binutils-gdb/gdb/target-delegates.c

4463 lines
153 KiB
C
Raw Normal View History

/* THIS FILE IS GENERATED -*- buffer-read-only: t -*- */
/* vi:set ro: */
/* To regenerate this file, run:*/
/* make-target-delegates target.h > target-delegates.c */
static void
delegate_post_attach (struct target_ops *self, int arg1)
{
self = self->beneath;
self->to_post_attach (self, arg1);
}
static void
tdefault_post_attach (struct target_ops *self, int arg1)
{
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_post_attach (struct target_ops *self, int arg1)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_post_attach (...)\n", debug_target.to_shortname);
debug_target.to_post_attach (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_post_attach (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg1);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
delegate_detach (struct target_ops *self, const char *arg1, int arg2)
{
self = self->beneath;
self->to_detach (self, arg1, arg2);
}
static void
tdefault_detach (struct target_ops *self, const char *arg1, int arg2)
{
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_detach (struct target_ops *self, const char *arg1, int arg2)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_detach (...)\n", debug_target.to_shortname);
debug_target.to_detach (&debug_target, arg1, arg2);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_detach (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_const_char_p (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg2);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
delegate_disconnect (struct target_ops *self, const char *arg1, int arg2)
{
self = self->beneath;
self->to_disconnect (self, arg1, arg2);
}
static void
tdefault_disconnect (struct target_ops *self, const char *arg1, int arg2)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_disconnect (struct target_ops *self, const char *arg1, int arg2)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_disconnect (...)\n", debug_target.to_shortname);
debug_target.to_disconnect (&debug_target, arg1, arg2);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_disconnect (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_const_char_p (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg2);
fputs_unfiltered (")\n", gdb_stdlog);
}
add target method delegation This patch replaces some code in the record targets with target method delegation. record-full.c stores pointers to many target methods when the record target is pushed. Then it later delegates some calls via these. This is wrong because it violates the target stack contract. In particular it is ok to unpush a target at any stratum, but record-full does not keep track of this, so it could potentially call into an unpushed target. This patch fixes the problem by using the newly-introduced generic approach to target delegation for the methods in question. 2014-02-19 Tom Tromey <tromey@redhat.com> * record-full.c (record_full_beneath_to_resume_ops) (record_full_beneath_to_resume, record_full_beneath_to_wait_ops) (record_full_beneath_to_wait) (record_full_beneath_to_store_registers_ops) (record_full_beneath_to_store_registers) (record_full_beneath_to_xfer_partial_ops) (record_full_beneath_to_xfer_partial) (record_full_beneath_to_insert_breakpoint_ops) (record_full_beneath_to_insert_breakpoint) (record_full_beneath_to_remove_breakpoint_ops) (record_full_beneath_to_remove_breakpoint) (record_full_beneath_to_stopped_by_watchpoint) (record_full_beneath_to_stopped_data_address) (record_full_beneath_to_async, tmp_to_resume_ops, tmp_to_resume) (tmp_to_wait_ops, tmp_to_wait, tmp_to_store_registers_ops) (tmp_to_store_registers, tmp_to_xfer_partial_ops) (tmp_to_xfer_partial, tmp_to_instmp_to_insert_breakpoint_ops) (tmp_to_insert_breakpoint, tmp_to_remove_breakpoint_ops) (tmp_to_remove_breakpoint, tmp_to_stopped_by_watchpoint) (tmp_to_stopped_data_address, tmp_to_async): Remove. (record_full_open_1, record_full_open): Update. Use RECORD_IS_USED. (record_full_resume, record_full_wait_1) (record_full_stopped_by_watchpoint, record_full_stopped_data_address) (record_full_store_registers, record_full_xfer_partial) (record_full_insert_breakpoint, record_full_remove_breakpoint) (record_full_async, record_full_core_xfer_partial): Use target delegation. * target-delegates.c: Rebuild. * target.c (current_xfer_partial): Remove. (update_current_target): Do not INHERIT or de_fault to_insert_breakpoint, to_remove_breakpoint, to_stopped_data_address, to_stopped_by_watchpoint, to_can_async_p, to_is_async_p, to_async. Do not set to_xfer_partial field. (default_xfer_partial): Simplify. (current_xfer_partial): Remove. (target_wait, target_resume): Simplify. (find_default_can_async_p, find_default_is_async_p): Update. (init_dummy_target): Don't set to_can_async_p, to_is_async_p, to_xfer_partial, to_stopped_by_watchpoint, to_stopped_data_address. (target_store_registers): Simplify. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. (target_remove_breakpoint, target_insert_breakpoint) (debug_to_insert_breakpoint, debug_to_remove_breakpoint): Update. * target.h (struct target_ops) <to_resume, to_wait, to_store_registers, to_insert_breakpoint, to_remove_breakpoint, to_stopped_by_watchpoint, to_stopped_data_address, to_can_async_p, to_is_async_p, to_async, to_xfer_partial>: Add TARGET_DEFAULT markup. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. * record-btrace.c (record_btrace_remove_breakpoint): Delegate directly. (record_btrace_insert_breakpoint): Delegate directly.
2013-07-13 03:00:34 +08:00
static void
delegate_resume (struct target_ops *self, ptid_t arg1, int arg2, enum gdb_signal arg3)
{
self = self->beneath;
self->to_resume (self, arg1, arg2, arg3);
}
static void
tdefault_resume (struct target_ops *self, ptid_t arg1, int arg2, enum gdb_signal arg3)
{
noprocess ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_resume (struct target_ops *self, ptid_t arg1, int arg2, enum gdb_signal arg3)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_resume (...)\n", debug_target.to_shortname);
debug_target.to_resume (&debug_target, arg1, arg2, arg3);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_resume (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_ptid_t (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_step (arg2);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_enum_gdb_signal (arg3);
fputs_unfiltered (")\n", gdb_stdlog);
}
add target method delegation This patch replaces some code in the record targets with target method delegation. record-full.c stores pointers to many target methods when the record target is pushed. Then it later delegates some calls via these. This is wrong because it violates the target stack contract. In particular it is ok to unpush a target at any stratum, but record-full does not keep track of this, so it could potentially call into an unpushed target. This patch fixes the problem by using the newly-introduced generic approach to target delegation for the methods in question. 2014-02-19 Tom Tromey <tromey@redhat.com> * record-full.c (record_full_beneath_to_resume_ops) (record_full_beneath_to_resume, record_full_beneath_to_wait_ops) (record_full_beneath_to_wait) (record_full_beneath_to_store_registers_ops) (record_full_beneath_to_store_registers) (record_full_beneath_to_xfer_partial_ops) (record_full_beneath_to_xfer_partial) (record_full_beneath_to_insert_breakpoint_ops) (record_full_beneath_to_insert_breakpoint) (record_full_beneath_to_remove_breakpoint_ops) (record_full_beneath_to_remove_breakpoint) (record_full_beneath_to_stopped_by_watchpoint) (record_full_beneath_to_stopped_data_address) (record_full_beneath_to_async, tmp_to_resume_ops, tmp_to_resume) (tmp_to_wait_ops, tmp_to_wait, tmp_to_store_registers_ops) (tmp_to_store_registers, tmp_to_xfer_partial_ops) (tmp_to_xfer_partial, tmp_to_instmp_to_insert_breakpoint_ops) (tmp_to_insert_breakpoint, tmp_to_remove_breakpoint_ops) (tmp_to_remove_breakpoint, tmp_to_stopped_by_watchpoint) (tmp_to_stopped_data_address, tmp_to_async): Remove. (record_full_open_1, record_full_open): Update. Use RECORD_IS_USED. (record_full_resume, record_full_wait_1) (record_full_stopped_by_watchpoint, record_full_stopped_data_address) (record_full_store_registers, record_full_xfer_partial) (record_full_insert_breakpoint, record_full_remove_breakpoint) (record_full_async, record_full_core_xfer_partial): Use target delegation. * target-delegates.c: Rebuild. * target.c (current_xfer_partial): Remove. (update_current_target): Do not INHERIT or de_fault to_insert_breakpoint, to_remove_breakpoint, to_stopped_data_address, to_stopped_by_watchpoint, to_can_async_p, to_is_async_p, to_async. Do not set to_xfer_partial field. (default_xfer_partial): Simplify. (current_xfer_partial): Remove. (target_wait, target_resume): Simplify. (find_default_can_async_p, find_default_is_async_p): Update. (init_dummy_target): Don't set to_can_async_p, to_is_async_p, to_xfer_partial, to_stopped_by_watchpoint, to_stopped_data_address. (target_store_registers): Simplify. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. (target_remove_breakpoint, target_insert_breakpoint) (debug_to_insert_breakpoint, debug_to_remove_breakpoint): Update. * target.h (struct target_ops) <to_resume, to_wait, to_store_registers, to_insert_breakpoint, to_remove_breakpoint, to_stopped_by_watchpoint, to_stopped_data_address, to_can_async_p, to_is_async_p, to_async, to_xfer_partial>: Add TARGET_DEFAULT markup. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. * record-btrace.c (record_btrace_remove_breakpoint): Delegate directly. (record_btrace_insert_breakpoint): Delegate directly.
2013-07-13 03:00:34 +08:00
static ptid_t
delegate_wait (struct target_ops *self, ptid_t arg1, struct target_waitstatus *arg2, int arg3)
{
self = self->beneath;
return self->to_wait (self, arg1, arg2, arg3);
}
static ptid_t
tdefault_wait (struct target_ops *self, ptid_t arg1, struct target_waitstatus *arg2, int arg3)
{
noprocess ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static ptid_t
debug_wait (struct target_ops *self, ptid_t arg1, struct target_waitstatus *arg2, int arg3)
{
ptid_t result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_wait (...)\n", debug_target.to_shortname);
result = debug_target.to_wait (&debug_target, arg1, arg2, arg3);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_wait (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_ptid_t (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_struct_target_waitstatus_p (arg2);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_options (arg3);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_ptid_t (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static void
delegate_fetch_registers (struct target_ops *self, struct regcache *arg1, int arg2)
{
self = self->beneath;
self->to_fetch_registers (self, arg1, arg2);
}
static void
tdefault_fetch_registers (struct target_ops *self, struct regcache *arg1, int arg2)
{
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_fetch_registers (struct target_ops *self, struct regcache *arg1, int arg2)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_fetch_registers (...)\n", debug_target.to_shortname);
debug_target.to_fetch_registers (&debug_target, arg1, arg2);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_fetch_registers (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_struct_regcache_p (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg2);
fputs_unfiltered (")\n", gdb_stdlog);
}
add target method delegation This patch replaces some code in the record targets with target method delegation. record-full.c stores pointers to many target methods when the record target is pushed. Then it later delegates some calls via these. This is wrong because it violates the target stack contract. In particular it is ok to unpush a target at any stratum, but record-full does not keep track of this, so it could potentially call into an unpushed target. This patch fixes the problem by using the newly-introduced generic approach to target delegation for the methods in question. 2014-02-19 Tom Tromey <tromey@redhat.com> * record-full.c (record_full_beneath_to_resume_ops) (record_full_beneath_to_resume, record_full_beneath_to_wait_ops) (record_full_beneath_to_wait) (record_full_beneath_to_store_registers_ops) (record_full_beneath_to_store_registers) (record_full_beneath_to_xfer_partial_ops) (record_full_beneath_to_xfer_partial) (record_full_beneath_to_insert_breakpoint_ops) (record_full_beneath_to_insert_breakpoint) (record_full_beneath_to_remove_breakpoint_ops) (record_full_beneath_to_remove_breakpoint) (record_full_beneath_to_stopped_by_watchpoint) (record_full_beneath_to_stopped_data_address) (record_full_beneath_to_async, tmp_to_resume_ops, tmp_to_resume) (tmp_to_wait_ops, tmp_to_wait, tmp_to_store_registers_ops) (tmp_to_store_registers, tmp_to_xfer_partial_ops) (tmp_to_xfer_partial, tmp_to_instmp_to_insert_breakpoint_ops) (tmp_to_insert_breakpoint, tmp_to_remove_breakpoint_ops) (tmp_to_remove_breakpoint, tmp_to_stopped_by_watchpoint) (tmp_to_stopped_data_address, tmp_to_async): Remove. (record_full_open_1, record_full_open): Update. Use RECORD_IS_USED. (record_full_resume, record_full_wait_1) (record_full_stopped_by_watchpoint, record_full_stopped_data_address) (record_full_store_registers, record_full_xfer_partial) (record_full_insert_breakpoint, record_full_remove_breakpoint) (record_full_async, record_full_core_xfer_partial): Use target delegation. * target-delegates.c: Rebuild. * target.c (current_xfer_partial): Remove. (update_current_target): Do not INHERIT or de_fault to_insert_breakpoint, to_remove_breakpoint, to_stopped_data_address, to_stopped_by_watchpoint, to_can_async_p, to_is_async_p, to_async. Do not set to_xfer_partial field. (default_xfer_partial): Simplify. (current_xfer_partial): Remove. (target_wait, target_resume): Simplify. (find_default_can_async_p, find_default_is_async_p): Update. (init_dummy_target): Don't set to_can_async_p, to_is_async_p, to_xfer_partial, to_stopped_by_watchpoint, to_stopped_data_address. (target_store_registers): Simplify. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. (target_remove_breakpoint, target_insert_breakpoint) (debug_to_insert_breakpoint, debug_to_remove_breakpoint): Update. * target.h (struct target_ops) <to_resume, to_wait, to_store_registers, to_insert_breakpoint, to_remove_breakpoint, to_stopped_by_watchpoint, to_stopped_data_address, to_can_async_p, to_is_async_p, to_async, to_xfer_partial>: Add TARGET_DEFAULT markup. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. * record-btrace.c (record_btrace_remove_breakpoint): Delegate directly. (record_btrace_insert_breakpoint): Delegate directly.
2013-07-13 03:00:34 +08:00
static void
delegate_store_registers (struct target_ops *self, struct regcache *arg1, int arg2)
{
self = self->beneath;
self->to_store_registers (self, arg1, arg2);
}
static void
tdefault_store_registers (struct target_ops *self, struct regcache *arg1, int arg2)
{
noprocess ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_store_registers (struct target_ops *self, struct regcache *arg1, int arg2)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_store_registers (...)\n", debug_target.to_shortname);
debug_target.to_store_registers (&debug_target, arg1, arg2);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_store_registers (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_struct_regcache_p (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg2);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
delegate_prepare_to_store (struct target_ops *self, struct regcache *arg1)
{
self = self->beneath;
self->to_prepare_to_store (self, arg1);
}
static void
tdefault_prepare_to_store (struct target_ops *self, struct regcache *arg1)
{
noprocess ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_prepare_to_store (struct target_ops *self, struct regcache *arg1)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_prepare_to_store (...)\n", debug_target.to_shortname);
debug_target.to_prepare_to_store (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_prepare_to_store (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_struct_regcache_p (arg1);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
delegate_files_info (struct target_ops *self)
{
self = self->beneath;
self->to_files_info (self);
}
static void
tdefault_files_info (struct target_ops *self)
{
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_files_info (struct target_ops *self)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_files_info (...)\n", debug_target.to_shortname);
debug_target.to_files_info (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_files_info (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (")\n", gdb_stdlog);
}
add target method delegation This patch replaces some code in the record targets with target method delegation. record-full.c stores pointers to many target methods when the record target is pushed. Then it later delegates some calls via these. This is wrong because it violates the target stack contract. In particular it is ok to unpush a target at any stratum, but record-full does not keep track of this, so it could potentially call into an unpushed target. This patch fixes the problem by using the newly-introduced generic approach to target delegation for the methods in question. 2014-02-19 Tom Tromey <tromey@redhat.com> * record-full.c (record_full_beneath_to_resume_ops) (record_full_beneath_to_resume, record_full_beneath_to_wait_ops) (record_full_beneath_to_wait) (record_full_beneath_to_store_registers_ops) (record_full_beneath_to_store_registers) (record_full_beneath_to_xfer_partial_ops) (record_full_beneath_to_xfer_partial) (record_full_beneath_to_insert_breakpoint_ops) (record_full_beneath_to_insert_breakpoint) (record_full_beneath_to_remove_breakpoint_ops) (record_full_beneath_to_remove_breakpoint) (record_full_beneath_to_stopped_by_watchpoint) (record_full_beneath_to_stopped_data_address) (record_full_beneath_to_async, tmp_to_resume_ops, tmp_to_resume) (tmp_to_wait_ops, tmp_to_wait, tmp_to_store_registers_ops) (tmp_to_store_registers, tmp_to_xfer_partial_ops) (tmp_to_xfer_partial, tmp_to_instmp_to_insert_breakpoint_ops) (tmp_to_insert_breakpoint, tmp_to_remove_breakpoint_ops) (tmp_to_remove_breakpoint, tmp_to_stopped_by_watchpoint) (tmp_to_stopped_data_address, tmp_to_async): Remove. (record_full_open_1, record_full_open): Update. Use RECORD_IS_USED. (record_full_resume, record_full_wait_1) (record_full_stopped_by_watchpoint, record_full_stopped_data_address) (record_full_store_registers, record_full_xfer_partial) (record_full_insert_breakpoint, record_full_remove_breakpoint) (record_full_async, record_full_core_xfer_partial): Use target delegation. * target-delegates.c: Rebuild. * target.c (current_xfer_partial): Remove. (update_current_target): Do not INHERIT or de_fault to_insert_breakpoint, to_remove_breakpoint, to_stopped_data_address, to_stopped_by_watchpoint, to_can_async_p, to_is_async_p, to_async. Do not set to_xfer_partial field. (default_xfer_partial): Simplify. (current_xfer_partial): Remove. (target_wait, target_resume): Simplify. (find_default_can_async_p, find_default_is_async_p): Update. (init_dummy_target): Don't set to_can_async_p, to_is_async_p, to_xfer_partial, to_stopped_by_watchpoint, to_stopped_data_address. (target_store_registers): Simplify. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. (target_remove_breakpoint, target_insert_breakpoint) (debug_to_insert_breakpoint, debug_to_remove_breakpoint): Update. * target.h (struct target_ops) <to_resume, to_wait, to_store_registers, to_insert_breakpoint, to_remove_breakpoint, to_stopped_by_watchpoint, to_stopped_data_address, to_can_async_p, to_is_async_p, to_async, to_xfer_partial>: Add TARGET_DEFAULT markup. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. * record-btrace.c (record_btrace_remove_breakpoint): Delegate directly. (record_btrace_insert_breakpoint): Delegate directly.
2013-07-13 03:00:34 +08:00
static int
delegate_insert_breakpoint (struct target_ops *self, struct gdbarch *arg1, struct bp_target_info *arg2)
{
self = self->beneath;
return self->to_insert_breakpoint (self, arg1, arg2);
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_insert_breakpoint (struct target_ops *self, struct gdbarch *arg1, struct bp_target_info *arg2)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_insert_breakpoint (...)\n", debug_target.to_shortname);
result = debug_target.to_insert_breakpoint (&debug_target, arg1, arg2);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_insert_breakpoint (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_struct_gdbarch_p (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_struct_bp_target_info_p (arg2);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
add target method delegation This patch replaces some code in the record targets with target method delegation. record-full.c stores pointers to many target methods when the record target is pushed. Then it later delegates some calls via these. This is wrong because it violates the target stack contract. In particular it is ok to unpush a target at any stratum, but record-full does not keep track of this, so it could potentially call into an unpushed target. This patch fixes the problem by using the newly-introduced generic approach to target delegation for the methods in question. 2014-02-19 Tom Tromey <tromey@redhat.com> * record-full.c (record_full_beneath_to_resume_ops) (record_full_beneath_to_resume, record_full_beneath_to_wait_ops) (record_full_beneath_to_wait) (record_full_beneath_to_store_registers_ops) (record_full_beneath_to_store_registers) (record_full_beneath_to_xfer_partial_ops) (record_full_beneath_to_xfer_partial) (record_full_beneath_to_insert_breakpoint_ops) (record_full_beneath_to_insert_breakpoint) (record_full_beneath_to_remove_breakpoint_ops) (record_full_beneath_to_remove_breakpoint) (record_full_beneath_to_stopped_by_watchpoint) (record_full_beneath_to_stopped_data_address) (record_full_beneath_to_async, tmp_to_resume_ops, tmp_to_resume) (tmp_to_wait_ops, tmp_to_wait, tmp_to_store_registers_ops) (tmp_to_store_registers, tmp_to_xfer_partial_ops) (tmp_to_xfer_partial, tmp_to_instmp_to_insert_breakpoint_ops) (tmp_to_insert_breakpoint, tmp_to_remove_breakpoint_ops) (tmp_to_remove_breakpoint, tmp_to_stopped_by_watchpoint) (tmp_to_stopped_data_address, tmp_to_async): Remove. (record_full_open_1, record_full_open): Update. Use RECORD_IS_USED. (record_full_resume, record_full_wait_1) (record_full_stopped_by_watchpoint, record_full_stopped_data_address) (record_full_store_registers, record_full_xfer_partial) (record_full_insert_breakpoint, record_full_remove_breakpoint) (record_full_async, record_full_core_xfer_partial): Use target delegation. * target-delegates.c: Rebuild. * target.c (current_xfer_partial): Remove. (update_current_target): Do not INHERIT or de_fault to_insert_breakpoint, to_remove_breakpoint, to_stopped_data_address, to_stopped_by_watchpoint, to_can_async_p, to_is_async_p, to_async. Do not set to_xfer_partial field. (default_xfer_partial): Simplify. (current_xfer_partial): Remove. (target_wait, target_resume): Simplify. (find_default_can_async_p, find_default_is_async_p): Update. (init_dummy_target): Don't set to_can_async_p, to_is_async_p, to_xfer_partial, to_stopped_by_watchpoint, to_stopped_data_address. (target_store_registers): Simplify. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. (target_remove_breakpoint, target_insert_breakpoint) (debug_to_insert_breakpoint, debug_to_remove_breakpoint): Update. * target.h (struct target_ops) <to_resume, to_wait, to_store_registers, to_insert_breakpoint, to_remove_breakpoint, to_stopped_by_watchpoint, to_stopped_data_address, to_can_async_p, to_is_async_p, to_async, to_xfer_partial>: Add TARGET_DEFAULT markup. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. * record-btrace.c (record_btrace_remove_breakpoint): Delegate directly. (record_btrace_insert_breakpoint): Delegate directly.
2013-07-13 03:00:34 +08:00
static int
delegate_remove_breakpoint (struct target_ops *self, struct gdbarch *arg1, struct bp_target_info *arg2)
{
self = self->beneath;
return self->to_remove_breakpoint (self, arg1, arg2);
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_remove_breakpoint (struct target_ops *self, struct gdbarch *arg1, struct bp_target_info *arg2)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_remove_breakpoint (...)\n", debug_target.to_shortname);
result = debug_target.to_remove_breakpoint (&debug_target, arg1, arg2);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_remove_breakpoint (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_struct_gdbarch_p (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_struct_bp_target_info_p (arg2);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
Teach GDB about targets that can tell whether a trap is a breakpoint event The moribund locations heuristics are problematic. This patch teaches GDB about targets that can reliably tell whether a trap was caused by a software or hardware breakpoint, and thus don't need moribund locations, thus bypassing all the problems that mechanism has. The non-stop-fair-events.exp test is frequently failing currently. E.g., see https://sourceware.org/ml/gdb-testers/2015-q1/msg03148.html. The root cause is a fundamental problem with moribund locations. For example, the stepped_breakpoint logic added by af48d08f breaks in this case (which is what happens with that test): - Step thread A, no breakpoint is set at PC. - The kernel doesn't schedule thread A yet. - Insert breakpoint at A's PC, for some reason (e.g., a step-resume breakpoint for thread B). - Kernel finally schedules thread A. - thread A's stepped_breakpoint flag is not set, even though it now stepped a breakpoint instruction. - adjust_pc_after_break gets the PC wrong, because PC == PREV_PC, but stepped_breakpoint is not set. We needed the stepped_breakpoint logic to workaround moribund locations, because otherwise adjust_pc_after_break could apply an adjustment when it shouldn't just because there _used_ to be a breakpoint at PC (a moribund breakpoint location). For example, on x86, that's wrong if the thread really hasn't executed an int3, but instead executed some other 1-byte long instruction. Getting the PC adjustment wrong of course leads to the inferior executing the wrong instruction. Other problems with moribund locations are: - if a true SIGTRAP happens to be raised when the program is executing the PC that used to have a breakpoint, GDB will assume that is a trap for a breakpoint that has recently been removed, and thus we miss reporting the random signal to the user. - to minimize that, we get rid of moribund location after a while. That while is defined as just a certain number of events being processed. That number of events sometimes passes by before a delayed breakpoint is processed, and GDB confuses the trap for a random signal, thus reporting the random trap. Once the user resumes the thread, the program crashes because the PC was not adjusted... The fix for all this is to bite the bullet and get rid of heuristics and instead rely on the target knowing accurately what caused the SIGTRAP. The target/kernel/stub is in the best position to know what that, because it can e.g. consult priviledged CPU flags GDB has no access to, or by knowing which exception vector entry was called when the instruction trapped, etc. Most debug APIs I've seen to date report breakpoint hits as a distinct event in some fashion. For example, on the Linux kernel, whether a breakpoint was executed is exposed to userspace in the si_code field of the SIGTRAP's siginfo. On Windows, the debug API reports a EXCEPTION_BREAKPOINT exception code. We needed to keep around deleted breakpoints in an on-the-side list (the moribund locations) for two main reasons: - Know that a SIGTRAP actually is a delayed event for a hit of a breakpoint that was removed before the event was processed, and thus should not be reported as a random signal. - So we still do the decr_pc_after_break adjustment in that case, so that the thread is resumed at the correct address. In the new model, if GDB processes an event the target tells is a breakpoint trap, and GDB doesn't find the corresponding breakpoint in its breakpoint tables, it means that event is a delayed event for a breakpoint that has since been removed, and thus the event should be ignored. For the decr_pc_after_after issue, it ends up being much simpler that on targets that can reliably tell whether a breakpoint trapped, for the breakpoint trap to present the PC already adjusted. Proper multi-threading support already implies that targets needs to be doing decr_pc_after_break adjustment themselves, otherwise for example, in all-stop if two threads hit a breakpoint simultaneously, and the user does "info threads", he'll see the non-event thread that hit the breakpoint stopped at the wrong PC. This way (target adjusts) also ends up eliminating the need for some awkward re-incrementing of the PC in the record-full and Linux targets that we do today, and the need for the target_decr_pc_after_break hook. If the target always adjusts, then there's a case where GDB needs to re-increment the PC. Say, on x86, an "int3" instruction that was explicitly written in the program traps. In this case, GDB should report a random SIGTRAP signal to the user, with the PC pointing at the instruction past the int3, just like if GDB was not debugging the program. The user may well decide to pass the SIGTRAP to the program because the program being debugged has a SIGTRAP handler that handles its own breakpoints, and expects the PC to be unadjusted. Tested on x86-64 Fedora 20. gdb/ChangeLog: 2015-03-04 Pedro Alves <palves@redhat.com> * breakpoint.c (need_moribund_for_location_type): New function. (bpstat_stop_status): Don't skipping checking moribund locations of breakpoint types which the target tell caused a stop. (program_breakpoint_here_p): New function, factored out from ... (bp_loc_is_permanent): ... this. (update_global_location_list): Don't create a moribund location if the target supports reporting stops of the type of the removed breakpoint. * breakpoint.h (program_breakpoint_here_p): New declaration. * infrun.c (adjust_pc_after_break): Return early if the target has already adjusted the PC. Add comments. (handle_signal_stop): If nothing explains a signal, and the target tells us the stop was caused by a software breakpoint, check if there's a breakpoint instruction in the memory. If so, adjust the PC before presenting the stop to the user. Otherwise, ignore the trap. If nothing explains a signal, and the target tells us the stop was caused by a hardware breakpoint, ignore the trap. * target.h (struct target_ops) <to_stopped_by_sw_breakpoint, to_supports_stopped_by_sw_breakpoint, to_stopped_by_hw_breakpoint, to_supports_stopped_by_hw_breakpoint>: New fields. (target_stopped_by_sw_breakpoint) (target_supports_stopped_by_sw_breakpoint) (target_stopped_by_hw_breakpoint) (target_supports_stopped_by_hw_breakpoint): Define. * target-delegates.c: Regenerate.
2015-03-05 04:41:15 +08:00
static int
delegate_stopped_by_sw_breakpoint (struct target_ops *self)
{
self = self->beneath;
return self->to_stopped_by_sw_breakpoint (self);
}
static int
tdefault_stopped_by_sw_breakpoint (struct target_ops *self)
{
return 0;
}
static int
debug_stopped_by_sw_breakpoint (struct target_ops *self)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_stopped_by_sw_breakpoint (...)\n", debug_target.to_shortname);
result = debug_target.to_stopped_by_sw_breakpoint (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_stopped_by_sw_breakpoint (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_supports_stopped_by_sw_breakpoint (struct target_ops *self)
{
self = self->beneath;
return self->to_supports_stopped_by_sw_breakpoint (self);
}
static int
tdefault_supports_stopped_by_sw_breakpoint (struct target_ops *self)
{
return 0;
}
static int
debug_supports_stopped_by_sw_breakpoint (struct target_ops *self)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_supports_stopped_by_sw_breakpoint (...)\n", debug_target.to_shortname);
result = debug_target.to_supports_stopped_by_sw_breakpoint (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_supports_stopped_by_sw_breakpoint (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_stopped_by_hw_breakpoint (struct target_ops *self)
{
self = self->beneath;
return self->to_stopped_by_hw_breakpoint (self);
}
static int
tdefault_stopped_by_hw_breakpoint (struct target_ops *self)
{
return 0;
}
static int
debug_stopped_by_hw_breakpoint (struct target_ops *self)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_stopped_by_hw_breakpoint (...)\n", debug_target.to_shortname);
result = debug_target.to_stopped_by_hw_breakpoint (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_stopped_by_hw_breakpoint (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_supports_stopped_by_hw_breakpoint (struct target_ops *self)
{
self = self->beneath;
return self->to_supports_stopped_by_hw_breakpoint (self);
}
static int
tdefault_supports_stopped_by_hw_breakpoint (struct target_ops *self)
{
return 0;
}
static int
debug_supports_stopped_by_hw_breakpoint (struct target_ops *self)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_supports_stopped_by_hw_breakpoint (...)\n", debug_target.to_shortname);
result = debug_target.to_supports_stopped_by_hw_breakpoint (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_supports_stopped_by_hw_breakpoint (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_can_use_hw_breakpoint (struct target_ops *self, int arg1, int arg2, int arg3)
{
self = self->beneath;
return self->to_can_use_hw_breakpoint (self, arg1, arg2, arg3);
}
static int
tdefault_can_use_hw_breakpoint (struct target_ops *self, int arg1, int arg2, int arg3)
{
return 0;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_can_use_hw_breakpoint (struct target_ops *self, int arg1, int arg2, int arg3)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_can_use_hw_breakpoint (...)\n", debug_target.to_shortname);
result = debug_target.to_can_use_hw_breakpoint (&debug_target, arg1, arg2, arg3);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_can_use_hw_breakpoint (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg2);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg3);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_ranged_break_num_registers (struct target_ops *self)
{
self = self->beneath;
return self->to_ranged_break_num_registers (self);
}
static int
tdefault_ranged_break_num_registers (struct target_ops *self)
{
return -1;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_ranged_break_num_registers (struct target_ops *self)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_ranged_break_num_registers (...)\n", debug_target.to_shortname);
result = debug_target.to_ranged_break_num_registers (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_ranged_break_num_registers (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_insert_hw_breakpoint (struct target_ops *self, struct gdbarch *arg1, struct bp_target_info *arg2)
{
self = self->beneath;
return self->to_insert_hw_breakpoint (self, arg1, arg2);
}
static int
tdefault_insert_hw_breakpoint (struct target_ops *self, struct gdbarch *arg1, struct bp_target_info *arg2)
{
return -1;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_insert_hw_breakpoint (struct target_ops *self, struct gdbarch *arg1, struct bp_target_info *arg2)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_insert_hw_breakpoint (...)\n", debug_target.to_shortname);
result = debug_target.to_insert_hw_breakpoint (&debug_target, arg1, arg2);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_insert_hw_breakpoint (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_struct_gdbarch_p (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_struct_bp_target_info_p (arg2);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_remove_hw_breakpoint (struct target_ops *self, struct gdbarch *arg1, struct bp_target_info *arg2)
{
self = self->beneath;
return self->to_remove_hw_breakpoint (self, arg1, arg2);
}
static int
tdefault_remove_hw_breakpoint (struct target_ops *self, struct gdbarch *arg1, struct bp_target_info *arg2)
{
return -1;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_remove_hw_breakpoint (struct target_ops *self, struct gdbarch *arg1, struct bp_target_info *arg2)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_remove_hw_breakpoint (...)\n", debug_target.to_shortname);
result = debug_target.to_remove_hw_breakpoint (&debug_target, arg1, arg2);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_remove_hw_breakpoint (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_struct_gdbarch_p (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_struct_bp_target_info_p (arg2);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_remove_watchpoint (struct target_ops *self, CORE_ADDR arg1, int arg2, int arg3, struct expression *arg4)
{
self = self->beneath;
return self->to_remove_watchpoint (self, arg1, arg2, arg3, arg4);
}
static int
tdefault_remove_watchpoint (struct target_ops *self, CORE_ADDR arg1, int arg2, int arg3, struct expression *arg4)
{
return -1;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_remove_watchpoint (struct target_ops *self, CORE_ADDR arg1, int arg2, int arg3, struct expression *arg4)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_remove_watchpoint (...)\n", debug_target.to_shortname);
result = debug_target.to_remove_watchpoint (&debug_target, arg1, arg2, arg3, arg4);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_remove_watchpoint (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_CORE_ADDR (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg2);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg3);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_struct_expression_p (arg4);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_insert_watchpoint (struct target_ops *self, CORE_ADDR arg1, int arg2, int arg3, struct expression *arg4)
{
self = self->beneath;
return self->to_insert_watchpoint (self, arg1, arg2, arg3, arg4);
}
static int
tdefault_insert_watchpoint (struct target_ops *self, CORE_ADDR arg1, int arg2, int arg3, struct expression *arg4)
{
return -1;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_insert_watchpoint (struct target_ops *self, CORE_ADDR arg1, int arg2, int arg3, struct expression *arg4)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_insert_watchpoint (...)\n", debug_target.to_shortname);
result = debug_target.to_insert_watchpoint (&debug_target, arg1, arg2, arg3, arg4);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_insert_watchpoint (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_CORE_ADDR (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg2);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg3);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_struct_expression_p (arg4);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_insert_mask_watchpoint (struct target_ops *self, CORE_ADDR arg1, CORE_ADDR arg2, int arg3)
{
self = self->beneath;
return self->to_insert_mask_watchpoint (self, arg1, arg2, arg3);
}
static int
tdefault_insert_mask_watchpoint (struct target_ops *self, CORE_ADDR arg1, CORE_ADDR arg2, int arg3)
{
return 1;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_insert_mask_watchpoint (struct target_ops *self, CORE_ADDR arg1, CORE_ADDR arg2, int arg3)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_insert_mask_watchpoint (...)\n", debug_target.to_shortname);
result = debug_target.to_insert_mask_watchpoint (&debug_target, arg1, arg2, arg3);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_insert_mask_watchpoint (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_CORE_ADDR (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_CORE_ADDR (arg2);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg3);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_remove_mask_watchpoint (struct target_ops *self, CORE_ADDR arg1, CORE_ADDR arg2, int arg3)
{
self = self->beneath;
return self->to_remove_mask_watchpoint (self, arg1, arg2, arg3);
}
static int
tdefault_remove_mask_watchpoint (struct target_ops *self, CORE_ADDR arg1, CORE_ADDR arg2, int arg3)
{
return 1;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_remove_mask_watchpoint (struct target_ops *self, CORE_ADDR arg1, CORE_ADDR arg2, int arg3)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_remove_mask_watchpoint (...)\n", debug_target.to_shortname);
result = debug_target.to_remove_mask_watchpoint (&debug_target, arg1, arg2, arg3);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_remove_mask_watchpoint (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_CORE_ADDR (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_CORE_ADDR (arg2);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg3);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
add target method delegation This patch replaces some code in the record targets with target method delegation. record-full.c stores pointers to many target methods when the record target is pushed. Then it later delegates some calls via these. This is wrong because it violates the target stack contract. In particular it is ok to unpush a target at any stratum, but record-full does not keep track of this, so it could potentially call into an unpushed target. This patch fixes the problem by using the newly-introduced generic approach to target delegation for the methods in question. 2014-02-19 Tom Tromey <tromey@redhat.com> * record-full.c (record_full_beneath_to_resume_ops) (record_full_beneath_to_resume, record_full_beneath_to_wait_ops) (record_full_beneath_to_wait) (record_full_beneath_to_store_registers_ops) (record_full_beneath_to_store_registers) (record_full_beneath_to_xfer_partial_ops) (record_full_beneath_to_xfer_partial) (record_full_beneath_to_insert_breakpoint_ops) (record_full_beneath_to_insert_breakpoint) (record_full_beneath_to_remove_breakpoint_ops) (record_full_beneath_to_remove_breakpoint) (record_full_beneath_to_stopped_by_watchpoint) (record_full_beneath_to_stopped_data_address) (record_full_beneath_to_async, tmp_to_resume_ops, tmp_to_resume) (tmp_to_wait_ops, tmp_to_wait, tmp_to_store_registers_ops) (tmp_to_store_registers, tmp_to_xfer_partial_ops) (tmp_to_xfer_partial, tmp_to_instmp_to_insert_breakpoint_ops) (tmp_to_insert_breakpoint, tmp_to_remove_breakpoint_ops) (tmp_to_remove_breakpoint, tmp_to_stopped_by_watchpoint) (tmp_to_stopped_data_address, tmp_to_async): Remove. (record_full_open_1, record_full_open): Update. Use RECORD_IS_USED. (record_full_resume, record_full_wait_1) (record_full_stopped_by_watchpoint, record_full_stopped_data_address) (record_full_store_registers, record_full_xfer_partial) (record_full_insert_breakpoint, record_full_remove_breakpoint) (record_full_async, record_full_core_xfer_partial): Use target delegation. * target-delegates.c: Rebuild. * target.c (current_xfer_partial): Remove. (update_current_target): Do not INHERIT or de_fault to_insert_breakpoint, to_remove_breakpoint, to_stopped_data_address, to_stopped_by_watchpoint, to_can_async_p, to_is_async_p, to_async. Do not set to_xfer_partial field. (default_xfer_partial): Simplify. (current_xfer_partial): Remove. (target_wait, target_resume): Simplify. (find_default_can_async_p, find_default_is_async_p): Update. (init_dummy_target): Don't set to_can_async_p, to_is_async_p, to_xfer_partial, to_stopped_by_watchpoint, to_stopped_data_address. (target_store_registers): Simplify. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. (target_remove_breakpoint, target_insert_breakpoint) (debug_to_insert_breakpoint, debug_to_remove_breakpoint): Update. * target.h (struct target_ops) <to_resume, to_wait, to_store_registers, to_insert_breakpoint, to_remove_breakpoint, to_stopped_by_watchpoint, to_stopped_data_address, to_can_async_p, to_is_async_p, to_async, to_xfer_partial>: Add TARGET_DEFAULT markup. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. * record-btrace.c (record_btrace_remove_breakpoint): Delegate directly. (record_btrace_insert_breakpoint): Delegate directly.
2013-07-13 03:00:34 +08:00
static int
delegate_stopped_by_watchpoint (struct target_ops *self)
{
self = self->beneath;
return self->to_stopped_by_watchpoint (self);
}
static int
tdefault_stopped_by_watchpoint (struct target_ops *self)
{
return 0;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_stopped_by_watchpoint (struct target_ops *self)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_stopped_by_watchpoint (...)\n", debug_target.to_shortname);
result = debug_target.to_stopped_by_watchpoint (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_stopped_by_watchpoint (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
add target method delegation This patch replaces some code in the record targets with target method delegation. record-full.c stores pointers to many target methods when the record target is pushed. Then it later delegates some calls via these. This is wrong because it violates the target stack contract. In particular it is ok to unpush a target at any stratum, but record-full does not keep track of this, so it could potentially call into an unpushed target. This patch fixes the problem by using the newly-introduced generic approach to target delegation for the methods in question. 2014-02-19 Tom Tromey <tromey@redhat.com> * record-full.c (record_full_beneath_to_resume_ops) (record_full_beneath_to_resume, record_full_beneath_to_wait_ops) (record_full_beneath_to_wait) (record_full_beneath_to_store_registers_ops) (record_full_beneath_to_store_registers) (record_full_beneath_to_xfer_partial_ops) (record_full_beneath_to_xfer_partial) (record_full_beneath_to_insert_breakpoint_ops) (record_full_beneath_to_insert_breakpoint) (record_full_beneath_to_remove_breakpoint_ops) (record_full_beneath_to_remove_breakpoint) (record_full_beneath_to_stopped_by_watchpoint) (record_full_beneath_to_stopped_data_address) (record_full_beneath_to_async, tmp_to_resume_ops, tmp_to_resume) (tmp_to_wait_ops, tmp_to_wait, tmp_to_store_registers_ops) (tmp_to_store_registers, tmp_to_xfer_partial_ops) (tmp_to_xfer_partial, tmp_to_instmp_to_insert_breakpoint_ops) (tmp_to_insert_breakpoint, tmp_to_remove_breakpoint_ops) (tmp_to_remove_breakpoint, tmp_to_stopped_by_watchpoint) (tmp_to_stopped_data_address, tmp_to_async): Remove. (record_full_open_1, record_full_open): Update. Use RECORD_IS_USED. (record_full_resume, record_full_wait_1) (record_full_stopped_by_watchpoint, record_full_stopped_data_address) (record_full_store_registers, record_full_xfer_partial) (record_full_insert_breakpoint, record_full_remove_breakpoint) (record_full_async, record_full_core_xfer_partial): Use target delegation. * target-delegates.c: Rebuild. * target.c (current_xfer_partial): Remove. (update_current_target): Do not INHERIT or de_fault to_insert_breakpoint, to_remove_breakpoint, to_stopped_data_address, to_stopped_by_watchpoint, to_can_async_p, to_is_async_p, to_async. Do not set to_xfer_partial field. (default_xfer_partial): Simplify. (current_xfer_partial): Remove. (target_wait, target_resume): Simplify. (find_default_can_async_p, find_default_is_async_p): Update. (init_dummy_target): Don't set to_can_async_p, to_is_async_p, to_xfer_partial, to_stopped_by_watchpoint, to_stopped_data_address. (target_store_registers): Simplify. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. (target_remove_breakpoint, target_insert_breakpoint) (debug_to_insert_breakpoint, debug_to_remove_breakpoint): Update. * target.h (struct target_ops) <to_resume, to_wait, to_store_registers, to_insert_breakpoint, to_remove_breakpoint, to_stopped_by_watchpoint, to_stopped_data_address, to_can_async_p, to_is_async_p, to_async, to_xfer_partial>: Add TARGET_DEFAULT markup. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. * record-btrace.c (record_btrace_remove_breakpoint): Delegate directly. (record_btrace_insert_breakpoint): Delegate directly.
2013-07-13 03:00:34 +08:00
static int
delegate_stopped_data_address (struct target_ops *self, CORE_ADDR *arg1)
{
self = self->beneath;
return self->to_stopped_data_address (self, arg1);
}
static int
tdefault_stopped_data_address (struct target_ops *self, CORE_ADDR *arg1)
{
return 0;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_stopped_data_address (struct target_ops *self, CORE_ADDR *arg1)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_stopped_data_address (...)\n", debug_target.to_shortname);
result = debug_target.to_stopped_data_address (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_stopped_data_address (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_CORE_ADDR_p (arg1);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_watchpoint_addr_within_range (struct target_ops *self, CORE_ADDR arg1, CORE_ADDR arg2, int arg3)
{
self = self->beneath;
return self->to_watchpoint_addr_within_range (self, arg1, arg2, arg3);
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_watchpoint_addr_within_range (struct target_ops *self, CORE_ADDR arg1, CORE_ADDR arg2, int arg3)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_watchpoint_addr_within_range (...)\n", debug_target.to_shortname);
result = debug_target.to_watchpoint_addr_within_range (&debug_target, arg1, arg2, arg3);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_watchpoint_addr_within_range (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_CORE_ADDR (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_CORE_ADDR (arg2);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg3);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_region_ok_for_hw_watchpoint (struct target_ops *self, CORE_ADDR arg1, int arg2)
{
self = self->beneath;
return self->to_region_ok_for_hw_watchpoint (self, arg1, arg2);
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_region_ok_for_hw_watchpoint (struct target_ops *self, CORE_ADDR arg1, int arg2)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_region_ok_for_hw_watchpoint (...)\n", debug_target.to_shortname);
result = debug_target.to_region_ok_for_hw_watchpoint (&debug_target, arg1, arg2);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_region_ok_for_hw_watchpoint (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_CORE_ADDR (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg2);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_can_accel_watchpoint_condition (struct target_ops *self, CORE_ADDR arg1, int arg2, int arg3, struct expression *arg4)
{
self = self->beneath;
return self->to_can_accel_watchpoint_condition (self, arg1, arg2, arg3, arg4);
}
static int
tdefault_can_accel_watchpoint_condition (struct target_ops *self, CORE_ADDR arg1, int arg2, int arg3, struct expression *arg4)
{
return 0;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_can_accel_watchpoint_condition (struct target_ops *self, CORE_ADDR arg1, int arg2, int arg3, struct expression *arg4)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_can_accel_watchpoint_condition (...)\n", debug_target.to_shortname);
result = debug_target.to_can_accel_watchpoint_condition (&debug_target, arg1, arg2, arg3, arg4);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_can_accel_watchpoint_condition (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_CORE_ADDR (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg2);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg3);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_struct_expression_p (arg4);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_masked_watch_num_registers (struct target_ops *self, CORE_ADDR arg1, CORE_ADDR arg2)
{
self = self->beneath;
return self->to_masked_watch_num_registers (self, arg1, arg2);
}
static int
tdefault_masked_watch_num_registers (struct target_ops *self, CORE_ADDR arg1, CORE_ADDR arg2)
{
return -1;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_masked_watch_num_registers (struct target_ops *self, CORE_ADDR arg1, CORE_ADDR arg2)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_masked_watch_num_registers (...)\n", debug_target.to_shortname);
result = debug_target.to_masked_watch_num_registers (&debug_target, arg1, arg2);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_masked_watch_num_registers (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_CORE_ADDR (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_CORE_ADDR (arg2);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static void
delegate_terminal_init (struct target_ops *self)
{
self = self->beneath;
self->to_terminal_init (self);
}
static void
tdefault_terminal_init (struct target_ops *self)
{
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_terminal_init (struct target_ops *self)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_terminal_init (...)\n", debug_target.to_shortname);
debug_target.to_terminal_init (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_terminal_init (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
delegate_terminal_inferior (struct target_ops *self)
{
self = self->beneath;
self->to_terminal_inferior (self);
}
static void
tdefault_terminal_inferior (struct target_ops *self)
{
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_terminal_inferior (struct target_ops *self)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_terminal_inferior (...)\n", debug_target.to_shortname);
debug_target.to_terminal_inferior (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_terminal_inferior (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
delegate_terminal_ours_for_output (struct target_ops *self)
{
self = self->beneath;
self->to_terminal_ours_for_output (self);
}
static void
tdefault_terminal_ours_for_output (struct target_ops *self)
{
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_terminal_ours_for_output (struct target_ops *self)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_terminal_ours_for_output (...)\n", debug_target.to_shortname);
debug_target.to_terminal_ours_for_output (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_terminal_ours_for_output (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
delegate_terminal_ours (struct target_ops *self)
{
self = self->beneath;
self->to_terminal_ours (self);
}
static void
tdefault_terminal_ours (struct target_ops *self)
{
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_terminal_ours (struct target_ops *self)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_terminal_ours (...)\n", debug_target.to_shortname);
debug_target.to_terminal_ours (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_terminal_ours (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
delegate_terminal_info (struct target_ops *self, const char *arg1, int arg2)
{
self = self->beneath;
self->to_terminal_info (self, arg1, arg2);
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_terminal_info (struct target_ops *self, const char *arg1, int arg2)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_terminal_info (...)\n", debug_target.to_shortname);
debug_target.to_terminal_info (&debug_target, arg1, arg2);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_terminal_info (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_const_char_p (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg2);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
delegate_kill (struct target_ops *self)
{
self = self->beneath;
self->to_kill (self);
}
static void
tdefault_kill (struct target_ops *self)
{
noprocess ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_kill (struct target_ops *self)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_kill (...)\n", debug_target.to_shortname);
debug_target.to_kill (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_kill (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
delegate_load (struct target_ops *self, const char *arg1, int arg2)
{
self = self->beneath;
self->to_load (self, arg1, arg2);
}
static void
tdefault_load (struct target_ops *self, const char *arg1, int arg2)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_load (struct target_ops *self, const char *arg1, int arg2)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_load (...)\n", debug_target.to_shortname);
debug_target.to_load (&debug_target, arg1, arg2);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_load (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_const_char_p (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg2);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
delegate_post_startup_inferior (struct target_ops *self, ptid_t arg1)
{
self = self->beneath;
self->to_post_startup_inferior (self, arg1);
}
static void
tdefault_post_startup_inferior (struct target_ops *self, ptid_t arg1)
{
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_post_startup_inferior (struct target_ops *self, ptid_t arg1)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_post_startup_inferior (...)\n", debug_target.to_shortname);
debug_target.to_post_startup_inferior (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_post_startup_inferior (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_ptid_t (arg1);
fputs_unfiltered (")\n", gdb_stdlog);
}
static int
delegate_insert_fork_catchpoint (struct target_ops *self, int arg1)
{
self = self->beneath;
return self->to_insert_fork_catchpoint (self, arg1);
}
static int
tdefault_insert_fork_catchpoint (struct target_ops *self, int arg1)
{
return 1;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_insert_fork_catchpoint (struct target_ops *self, int arg1)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_insert_fork_catchpoint (...)\n", debug_target.to_shortname);
result = debug_target.to_insert_fork_catchpoint (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_insert_fork_catchpoint (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg1);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_remove_fork_catchpoint (struct target_ops *self, int arg1)
{
self = self->beneath;
return self->to_remove_fork_catchpoint (self, arg1);
}
static int
tdefault_remove_fork_catchpoint (struct target_ops *self, int arg1)
{
return 1;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_remove_fork_catchpoint (struct target_ops *self, int arg1)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_remove_fork_catchpoint (...)\n", debug_target.to_shortname);
result = debug_target.to_remove_fork_catchpoint (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_remove_fork_catchpoint (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg1);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_insert_vfork_catchpoint (struct target_ops *self, int arg1)
{
self = self->beneath;
return self->to_insert_vfork_catchpoint (self, arg1);
}
static int
tdefault_insert_vfork_catchpoint (struct target_ops *self, int arg1)
{
return 1;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_insert_vfork_catchpoint (struct target_ops *self, int arg1)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_insert_vfork_catchpoint (...)\n", debug_target.to_shortname);
result = debug_target.to_insert_vfork_catchpoint (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_insert_vfork_catchpoint (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg1);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_remove_vfork_catchpoint (struct target_ops *self, int arg1)
{
self = self->beneath;
return self->to_remove_vfork_catchpoint (self, arg1);
}
static int
tdefault_remove_vfork_catchpoint (struct target_ops *self, int arg1)
{
return 1;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_remove_vfork_catchpoint (struct target_ops *self, int arg1)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_remove_vfork_catchpoint (...)\n", debug_target.to_shortname);
result = debug_target.to_remove_vfork_catchpoint (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_remove_vfork_catchpoint (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg1);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_follow_fork (struct target_ops *self, int arg1, int arg2)
{
self = self->beneath;
return self->to_follow_fork (self, arg1, arg2);
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_follow_fork (struct target_ops *self, int arg1, int arg2)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_follow_fork (...)\n", debug_target.to_shortname);
result = debug_target.to_follow_fork (&debug_target, arg1, arg2);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_follow_fork (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg2);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_insert_exec_catchpoint (struct target_ops *self, int arg1)
{
self = self->beneath;
return self->to_insert_exec_catchpoint (self, arg1);
}
static int
tdefault_insert_exec_catchpoint (struct target_ops *self, int arg1)
{
return 1;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_insert_exec_catchpoint (struct target_ops *self, int arg1)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_insert_exec_catchpoint (...)\n", debug_target.to_shortname);
result = debug_target.to_insert_exec_catchpoint (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_insert_exec_catchpoint (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg1);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_remove_exec_catchpoint (struct target_ops *self, int arg1)
{
self = self->beneath;
return self->to_remove_exec_catchpoint (self, arg1);
}
static int
tdefault_remove_exec_catchpoint (struct target_ops *self, int arg1)
{
return 1;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_remove_exec_catchpoint (struct target_ops *self, int arg1)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_remove_exec_catchpoint (...)\n", debug_target.to_shortname);
result = debug_target.to_remove_exec_catchpoint (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_remove_exec_catchpoint (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg1);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_set_syscall_catchpoint (struct target_ops *self, int arg1, int arg2, int arg3, int arg4, int *arg5)
{
self = self->beneath;
return self->to_set_syscall_catchpoint (self, arg1, arg2, arg3, arg4, arg5);
}
static int
tdefault_set_syscall_catchpoint (struct target_ops *self, int arg1, int arg2, int arg3, int arg4, int *arg5)
{
return 1;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_set_syscall_catchpoint (struct target_ops *self, int arg1, int arg2, int arg3, int arg4, int *arg5)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_set_syscall_catchpoint (...)\n", debug_target.to_shortname);
result = debug_target.to_set_syscall_catchpoint (&debug_target, arg1, arg2, arg3, arg4, arg5);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_set_syscall_catchpoint (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg2);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg3);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg4);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int_p (arg5);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_has_exited (struct target_ops *self, int arg1, int arg2, int *arg3)
{
self = self->beneath;
return self->to_has_exited (self, arg1, arg2, arg3);
}
static int
tdefault_has_exited (struct target_ops *self, int arg1, int arg2, int *arg3)
{
return 0;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_has_exited (struct target_ops *self, int arg1, int arg2, int *arg3)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_has_exited (...)\n", debug_target.to_shortname);
result = debug_target.to_has_exited (&debug_target, arg1, arg2, arg3);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_has_exited (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg2);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int_p (arg3);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static void
delegate_mourn_inferior (struct target_ops *self)
{
self = self->beneath;
self->to_mourn_inferior (self);
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_mourn_inferior (struct target_ops *self)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_mourn_inferior (...)\n", debug_target.to_shortname);
debug_target.to_mourn_inferior (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_mourn_inferior (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (")\n", gdb_stdlog);
}
static int
delegate_can_run (struct target_ops *self)
{
self = self->beneath;
return self->to_can_run (self);
}
static int
tdefault_can_run (struct target_ops *self)
{
return 0;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_can_run (struct target_ops *self)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_can_run (...)\n", debug_target.to_shortname);
result = debug_target.to_can_run (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_can_run (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static void
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
delegate_pass_signals (struct target_ops *self, int arg1, unsigned char * arg2)
{
self = self->beneath;
self->to_pass_signals (self, arg1, arg2);
}
static void
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
tdefault_pass_signals (struct target_ops *self, int arg1, unsigned char * arg2)
{
}
static void
debug_pass_signals (struct target_ops *self, int arg1, unsigned char * arg2)
{
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
fprintf_unfiltered (gdb_stdlog, "-> %s->to_pass_signals (...)\n", debug_target.to_shortname);
debug_target.to_pass_signals (&debug_target, arg1, arg2);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_pass_signals (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_signals (arg2);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
delegate_program_signals (struct target_ops *self, int arg1, unsigned char * arg2)
{
self = self->beneath;
self->to_program_signals (self, arg1, arg2);
}
static void
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
tdefault_program_signals (struct target_ops *self, int arg1, unsigned char * arg2)
{
}
static void
debug_program_signals (struct target_ops *self, int arg1, unsigned char * arg2)
{
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
fprintf_unfiltered (gdb_stdlog, "-> %s->to_program_signals (...)\n", debug_target.to_shortname);
debug_target.to_program_signals (&debug_target, arg1, arg2);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_program_signals (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_signals (arg2);
fputs_unfiltered (")\n", gdb_stdlog);
}
static int
delegate_thread_alive (struct target_ops *self, ptid_t arg1)
{
self = self->beneath;
return self->to_thread_alive (self, arg1);
}
static int
tdefault_thread_alive (struct target_ops *self, ptid_t arg1)
{
return 0;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_thread_alive (struct target_ops *self, ptid_t arg1)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_thread_alive (...)\n", debug_target.to_shortname);
result = debug_target.to_thread_alive (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_thread_alive (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_ptid_t (arg1);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static void
Push pruning old threads down to the target When GDB wants to sync the thread list with the target's (e.g., due to "info threads"), it calls update_thread_list: update_thread_list (void) { prune_threads (); target_find_new_threads (); update_threads_executing (); } And then prune_threads does: prune_threads (void) { struct thread_info *tp, *next; for (tp = thread_list; tp; tp = next) { next = tp->next; if (!thread_alive (tp)) delete_thread (tp->ptid); } } Calling thread_live on each thread one by one is expensive. E.g., on Linux, it ends up doing kill(SIG0) once for each thread. Not a big deal, but still a bunch of syscalls... With the remote target, it's cumbersome. That thread_alive call ends up generating one T packet per thread: Sending packet: $Tp2141.2150#82...Packet received: OK Sending packet: $Tp2141.214f#b7...Packet received: OK Sending packet: $Tp2141.2141#82...Packet received: OK Sending packet: $qXfer:threads:read::0,fff#03...Packet received: l<threads>\n<thread id="p2141.2141" core="2"/>\n<thread id="p2141.214f" core="1"/>\n<thread id="p2141.2150" core="2"/>\n</threads>\n That seems a bit silly when target_find_new_threads method implementations will always fetch the whole current set of target threads, and then add those that are not in GDB's thread list, to GDB's thread list. This patch thus pushes down the responsibility of pruning dead threads to the target_find_new_threads method instead, so a target may implement pruning dead threads however it wants. Once we do that, target_find_new_threads becomes a misnomer, so the patch renames it to target_update_thread_list. The patch doesn't attempt to do any optimization to any target yet. It simply exports prune_threads, and makes all implementations of target_update_thread_list call that. It's meant to be a no-op. gdb/ 2014-10-15 Pedro Alves <palves@redhat.com> * ada-tasks.c (print_ada_task_info, task_command_1): Adjust. * bsd-uthread.c (bsd_uthread_find_new_threads): Rename to ... (bsd_uthread_update_thread_list): ... this. Call prune_threads. (bsd_uthread_target): Adjust. * corelow.c (core_open): Adjust. * dec-thread.c (dec_thread_find_new_threads): Update comment. (dec_thread_update_thread_list): New function. (init_dec_thread_ops): Adjust. * gdbthread.h (prune_threads): New declaration. * linux-thread-db.c (thread_db_find_new_threads): Rename to ... (thread_db_update_thread_list): ... this. Call prune_threads. (init_thread_db_ops): Adjust. * nto-procfs.c (procfs_find_new_threads): Rename to ... (procfs_update_thread_list): ... this. Call prune_threads. (procfs_attach, procfs_create_inferior, init_procfs_targets): Adjust. * obsd-nat.c (obsd_find_new_threads): Rename to ... (obsd_update_thread_list): ... this. Call prune_threads. (obsd_add_target): Adjust. * procfs.c (procfs_target): Adjust. (procfs_notice_thread): Update comment. (procfs_find_new_threads): Rename to ... (procfs_update_thread_list): ... this. Call prune_threads. * ravenscar-thread.c (ravenscar_update_inferior_ptid): Update comment. (ravenscar_wait): Adjust. (ravenscar_find_new_threads): Rename to ... (ravenscar_update_thread_list): ... this. Call prune_threads. (init_ravenscar_thread_ops): Adjust. * record-btrace.c (record_btrace_find_new_threads): Rename to ... (record_btrace_update_thread_list): ... this. Adjust comment. (init_record_btrace_ops): Adjust. * remote.c (remote_threads_info): Rename to ... (remote_update_thread_list): ... this. Call prune_threads. (remote_start_remote, extended_remote_attach_1, init_remote_ops): Adjust. * sol-thread.c (check_for_thread_db): Adjust. (sol_find_new_threads_callback): Rename to ... (sol_update_thread_list_callback): ... this. (sol_find_new_threads): Rename to ... (sol_update_thread_list): ... this. Call prune_threads. Adjust. (sol_get_ada_task_ptid, init_sol_thread_ops): Adjust. * target-delegates.c: Regenerate. * target.c (target_find_new_threads): Rename to ... (target_update_thread_list): ... this. * target.h (struct target_ops): Rename to_find_new_threads field to to_update_thread_list. (target_find_new_threads): Rename to ... (target_update_thread_list): ... this. * thread.c (prune_threads): Make extern. (update_thread_list): Adjust.
2014-10-16 05:44:00 +08:00
delegate_update_thread_list (struct target_ops *self)
{
self = self->beneath;
Push pruning old threads down to the target When GDB wants to sync the thread list with the target's (e.g., due to "info threads"), it calls update_thread_list: update_thread_list (void) { prune_threads (); target_find_new_threads (); update_threads_executing (); } And then prune_threads does: prune_threads (void) { struct thread_info *tp, *next; for (tp = thread_list; tp; tp = next) { next = tp->next; if (!thread_alive (tp)) delete_thread (tp->ptid); } } Calling thread_live on each thread one by one is expensive. E.g., on Linux, it ends up doing kill(SIG0) once for each thread. Not a big deal, but still a bunch of syscalls... With the remote target, it's cumbersome. That thread_alive call ends up generating one T packet per thread: Sending packet: $Tp2141.2150#82...Packet received: OK Sending packet: $Tp2141.214f#b7...Packet received: OK Sending packet: $Tp2141.2141#82...Packet received: OK Sending packet: $qXfer:threads:read::0,fff#03...Packet received: l<threads>\n<thread id="p2141.2141" core="2"/>\n<thread id="p2141.214f" core="1"/>\n<thread id="p2141.2150" core="2"/>\n</threads>\n That seems a bit silly when target_find_new_threads method implementations will always fetch the whole current set of target threads, and then add those that are not in GDB's thread list, to GDB's thread list. This patch thus pushes down the responsibility of pruning dead threads to the target_find_new_threads method instead, so a target may implement pruning dead threads however it wants. Once we do that, target_find_new_threads becomes a misnomer, so the patch renames it to target_update_thread_list. The patch doesn't attempt to do any optimization to any target yet. It simply exports prune_threads, and makes all implementations of target_update_thread_list call that. It's meant to be a no-op. gdb/ 2014-10-15 Pedro Alves <palves@redhat.com> * ada-tasks.c (print_ada_task_info, task_command_1): Adjust. * bsd-uthread.c (bsd_uthread_find_new_threads): Rename to ... (bsd_uthread_update_thread_list): ... this. Call prune_threads. (bsd_uthread_target): Adjust. * corelow.c (core_open): Adjust. * dec-thread.c (dec_thread_find_new_threads): Update comment. (dec_thread_update_thread_list): New function. (init_dec_thread_ops): Adjust. * gdbthread.h (prune_threads): New declaration. * linux-thread-db.c (thread_db_find_new_threads): Rename to ... (thread_db_update_thread_list): ... this. Call prune_threads. (init_thread_db_ops): Adjust. * nto-procfs.c (procfs_find_new_threads): Rename to ... (procfs_update_thread_list): ... this. Call prune_threads. (procfs_attach, procfs_create_inferior, init_procfs_targets): Adjust. * obsd-nat.c (obsd_find_new_threads): Rename to ... (obsd_update_thread_list): ... this. Call prune_threads. (obsd_add_target): Adjust. * procfs.c (procfs_target): Adjust. (procfs_notice_thread): Update comment. (procfs_find_new_threads): Rename to ... (procfs_update_thread_list): ... this. Call prune_threads. * ravenscar-thread.c (ravenscar_update_inferior_ptid): Update comment. (ravenscar_wait): Adjust. (ravenscar_find_new_threads): Rename to ... (ravenscar_update_thread_list): ... this. Call prune_threads. (init_ravenscar_thread_ops): Adjust. * record-btrace.c (record_btrace_find_new_threads): Rename to ... (record_btrace_update_thread_list): ... this. Adjust comment. (init_record_btrace_ops): Adjust. * remote.c (remote_threads_info): Rename to ... (remote_update_thread_list): ... this. Call prune_threads. (remote_start_remote, extended_remote_attach_1, init_remote_ops): Adjust. * sol-thread.c (check_for_thread_db): Adjust. (sol_find_new_threads_callback): Rename to ... (sol_update_thread_list_callback): ... this. (sol_find_new_threads): Rename to ... (sol_update_thread_list): ... this. Call prune_threads. Adjust. (sol_get_ada_task_ptid, init_sol_thread_ops): Adjust. * target-delegates.c: Regenerate. * target.c (target_find_new_threads): Rename to ... (target_update_thread_list): ... this. * target.h (struct target_ops): Rename to_find_new_threads field to to_update_thread_list. (target_find_new_threads): Rename to ... (target_update_thread_list): ... this. * thread.c (prune_threads): Make extern. (update_thread_list): Adjust.
2014-10-16 05:44:00 +08:00
self->to_update_thread_list (self);
}
static void
Push pruning old threads down to the target When GDB wants to sync the thread list with the target's (e.g., due to "info threads"), it calls update_thread_list: update_thread_list (void) { prune_threads (); target_find_new_threads (); update_threads_executing (); } And then prune_threads does: prune_threads (void) { struct thread_info *tp, *next; for (tp = thread_list; tp; tp = next) { next = tp->next; if (!thread_alive (tp)) delete_thread (tp->ptid); } } Calling thread_live on each thread one by one is expensive. E.g., on Linux, it ends up doing kill(SIG0) once for each thread. Not a big deal, but still a bunch of syscalls... With the remote target, it's cumbersome. That thread_alive call ends up generating one T packet per thread: Sending packet: $Tp2141.2150#82...Packet received: OK Sending packet: $Tp2141.214f#b7...Packet received: OK Sending packet: $Tp2141.2141#82...Packet received: OK Sending packet: $qXfer:threads:read::0,fff#03...Packet received: l<threads>\n<thread id="p2141.2141" core="2"/>\n<thread id="p2141.214f" core="1"/>\n<thread id="p2141.2150" core="2"/>\n</threads>\n That seems a bit silly when target_find_new_threads method implementations will always fetch the whole current set of target threads, and then add those that are not in GDB's thread list, to GDB's thread list. This patch thus pushes down the responsibility of pruning dead threads to the target_find_new_threads method instead, so a target may implement pruning dead threads however it wants. Once we do that, target_find_new_threads becomes a misnomer, so the patch renames it to target_update_thread_list. The patch doesn't attempt to do any optimization to any target yet. It simply exports prune_threads, and makes all implementations of target_update_thread_list call that. It's meant to be a no-op. gdb/ 2014-10-15 Pedro Alves <palves@redhat.com> * ada-tasks.c (print_ada_task_info, task_command_1): Adjust. * bsd-uthread.c (bsd_uthread_find_new_threads): Rename to ... (bsd_uthread_update_thread_list): ... this. Call prune_threads. (bsd_uthread_target): Adjust. * corelow.c (core_open): Adjust. * dec-thread.c (dec_thread_find_new_threads): Update comment. (dec_thread_update_thread_list): New function. (init_dec_thread_ops): Adjust. * gdbthread.h (prune_threads): New declaration. * linux-thread-db.c (thread_db_find_new_threads): Rename to ... (thread_db_update_thread_list): ... this. Call prune_threads. (init_thread_db_ops): Adjust. * nto-procfs.c (procfs_find_new_threads): Rename to ... (procfs_update_thread_list): ... this. Call prune_threads. (procfs_attach, procfs_create_inferior, init_procfs_targets): Adjust. * obsd-nat.c (obsd_find_new_threads): Rename to ... (obsd_update_thread_list): ... this. Call prune_threads. (obsd_add_target): Adjust. * procfs.c (procfs_target): Adjust. (procfs_notice_thread): Update comment. (procfs_find_new_threads): Rename to ... (procfs_update_thread_list): ... this. Call prune_threads. * ravenscar-thread.c (ravenscar_update_inferior_ptid): Update comment. (ravenscar_wait): Adjust. (ravenscar_find_new_threads): Rename to ... (ravenscar_update_thread_list): ... this. Call prune_threads. (init_ravenscar_thread_ops): Adjust. * record-btrace.c (record_btrace_find_new_threads): Rename to ... (record_btrace_update_thread_list): ... this. Adjust comment. (init_record_btrace_ops): Adjust. * remote.c (remote_threads_info): Rename to ... (remote_update_thread_list): ... this. Call prune_threads. (remote_start_remote, extended_remote_attach_1, init_remote_ops): Adjust. * sol-thread.c (check_for_thread_db): Adjust. (sol_find_new_threads_callback): Rename to ... (sol_update_thread_list_callback): ... this. (sol_find_new_threads): Rename to ... (sol_update_thread_list): ... this. Call prune_threads. Adjust. (sol_get_ada_task_ptid, init_sol_thread_ops): Adjust. * target-delegates.c: Regenerate. * target.c (target_find_new_threads): Rename to ... (target_update_thread_list): ... this. * target.h (struct target_ops): Rename to_find_new_threads field to to_update_thread_list. (target_find_new_threads): Rename to ... (target_update_thread_list): ... this. * thread.c (prune_threads): Make extern. (update_thread_list): Adjust.
2014-10-16 05:44:00 +08:00
tdefault_update_thread_list (struct target_ops *self)
{
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
Push pruning old threads down to the target When GDB wants to sync the thread list with the target's (e.g., due to "info threads"), it calls update_thread_list: update_thread_list (void) { prune_threads (); target_find_new_threads (); update_threads_executing (); } And then prune_threads does: prune_threads (void) { struct thread_info *tp, *next; for (tp = thread_list; tp; tp = next) { next = tp->next; if (!thread_alive (tp)) delete_thread (tp->ptid); } } Calling thread_live on each thread one by one is expensive. E.g., on Linux, it ends up doing kill(SIG0) once for each thread. Not a big deal, but still a bunch of syscalls... With the remote target, it's cumbersome. That thread_alive call ends up generating one T packet per thread: Sending packet: $Tp2141.2150#82...Packet received: OK Sending packet: $Tp2141.214f#b7...Packet received: OK Sending packet: $Tp2141.2141#82...Packet received: OK Sending packet: $qXfer:threads:read::0,fff#03...Packet received: l<threads>\n<thread id="p2141.2141" core="2"/>\n<thread id="p2141.214f" core="1"/>\n<thread id="p2141.2150" core="2"/>\n</threads>\n That seems a bit silly when target_find_new_threads method implementations will always fetch the whole current set of target threads, and then add those that are not in GDB's thread list, to GDB's thread list. This patch thus pushes down the responsibility of pruning dead threads to the target_find_new_threads method instead, so a target may implement pruning dead threads however it wants. Once we do that, target_find_new_threads becomes a misnomer, so the patch renames it to target_update_thread_list. The patch doesn't attempt to do any optimization to any target yet. It simply exports prune_threads, and makes all implementations of target_update_thread_list call that. It's meant to be a no-op. gdb/ 2014-10-15 Pedro Alves <palves@redhat.com> * ada-tasks.c (print_ada_task_info, task_command_1): Adjust. * bsd-uthread.c (bsd_uthread_find_new_threads): Rename to ... (bsd_uthread_update_thread_list): ... this. Call prune_threads. (bsd_uthread_target): Adjust. * corelow.c (core_open): Adjust. * dec-thread.c (dec_thread_find_new_threads): Update comment. (dec_thread_update_thread_list): New function. (init_dec_thread_ops): Adjust. * gdbthread.h (prune_threads): New declaration. * linux-thread-db.c (thread_db_find_new_threads): Rename to ... (thread_db_update_thread_list): ... this. Call prune_threads. (init_thread_db_ops): Adjust. * nto-procfs.c (procfs_find_new_threads): Rename to ... (procfs_update_thread_list): ... this. Call prune_threads. (procfs_attach, procfs_create_inferior, init_procfs_targets): Adjust. * obsd-nat.c (obsd_find_new_threads): Rename to ... (obsd_update_thread_list): ... this. Call prune_threads. (obsd_add_target): Adjust. * procfs.c (procfs_target): Adjust. (procfs_notice_thread): Update comment. (procfs_find_new_threads): Rename to ... (procfs_update_thread_list): ... this. Call prune_threads. * ravenscar-thread.c (ravenscar_update_inferior_ptid): Update comment. (ravenscar_wait): Adjust. (ravenscar_find_new_threads): Rename to ... (ravenscar_update_thread_list): ... this. Call prune_threads. (init_ravenscar_thread_ops): Adjust. * record-btrace.c (record_btrace_find_new_threads): Rename to ... (record_btrace_update_thread_list): ... this. Adjust comment. (init_record_btrace_ops): Adjust. * remote.c (remote_threads_info): Rename to ... (remote_update_thread_list): ... this. Call prune_threads. (remote_start_remote, extended_remote_attach_1, init_remote_ops): Adjust. * sol-thread.c (check_for_thread_db): Adjust. (sol_find_new_threads_callback): Rename to ... (sol_update_thread_list_callback): ... this. (sol_find_new_threads): Rename to ... (sol_update_thread_list): ... this. Call prune_threads. Adjust. (sol_get_ada_task_ptid, init_sol_thread_ops): Adjust. * target-delegates.c: Regenerate. * target.c (target_find_new_threads): Rename to ... (target_update_thread_list): ... this. * target.h (struct target_ops): Rename to_find_new_threads field to to_update_thread_list. (target_find_new_threads): Rename to ... (target_update_thread_list): ... this. * thread.c (prune_threads): Make extern. (update_thread_list): Adjust.
2014-10-16 05:44:00 +08:00
debug_update_thread_list (struct target_ops *self)
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
{
Push pruning old threads down to the target When GDB wants to sync the thread list with the target's (e.g., due to "info threads"), it calls update_thread_list: update_thread_list (void) { prune_threads (); target_find_new_threads (); update_threads_executing (); } And then prune_threads does: prune_threads (void) { struct thread_info *tp, *next; for (tp = thread_list; tp; tp = next) { next = tp->next; if (!thread_alive (tp)) delete_thread (tp->ptid); } } Calling thread_live on each thread one by one is expensive. E.g., on Linux, it ends up doing kill(SIG0) once for each thread. Not a big deal, but still a bunch of syscalls... With the remote target, it's cumbersome. That thread_alive call ends up generating one T packet per thread: Sending packet: $Tp2141.2150#82...Packet received: OK Sending packet: $Tp2141.214f#b7...Packet received: OK Sending packet: $Tp2141.2141#82...Packet received: OK Sending packet: $qXfer:threads:read::0,fff#03...Packet received: l<threads>\n<thread id="p2141.2141" core="2"/>\n<thread id="p2141.214f" core="1"/>\n<thread id="p2141.2150" core="2"/>\n</threads>\n That seems a bit silly when target_find_new_threads method implementations will always fetch the whole current set of target threads, and then add those that are not in GDB's thread list, to GDB's thread list. This patch thus pushes down the responsibility of pruning dead threads to the target_find_new_threads method instead, so a target may implement pruning dead threads however it wants. Once we do that, target_find_new_threads becomes a misnomer, so the patch renames it to target_update_thread_list. The patch doesn't attempt to do any optimization to any target yet. It simply exports prune_threads, and makes all implementations of target_update_thread_list call that. It's meant to be a no-op. gdb/ 2014-10-15 Pedro Alves <palves@redhat.com> * ada-tasks.c (print_ada_task_info, task_command_1): Adjust. * bsd-uthread.c (bsd_uthread_find_new_threads): Rename to ... (bsd_uthread_update_thread_list): ... this. Call prune_threads. (bsd_uthread_target): Adjust. * corelow.c (core_open): Adjust. * dec-thread.c (dec_thread_find_new_threads): Update comment. (dec_thread_update_thread_list): New function. (init_dec_thread_ops): Adjust. * gdbthread.h (prune_threads): New declaration. * linux-thread-db.c (thread_db_find_new_threads): Rename to ... (thread_db_update_thread_list): ... this. Call prune_threads. (init_thread_db_ops): Adjust. * nto-procfs.c (procfs_find_new_threads): Rename to ... (procfs_update_thread_list): ... this. Call prune_threads. (procfs_attach, procfs_create_inferior, init_procfs_targets): Adjust. * obsd-nat.c (obsd_find_new_threads): Rename to ... (obsd_update_thread_list): ... this. Call prune_threads. (obsd_add_target): Adjust. * procfs.c (procfs_target): Adjust. (procfs_notice_thread): Update comment. (procfs_find_new_threads): Rename to ... (procfs_update_thread_list): ... this. Call prune_threads. * ravenscar-thread.c (ravenscar_update_inferior_ptid): Update comment. (ravenscar_wait): Adjust. (ravenscar_find_new_threads): Rename to ... (ravenscar_update_thread_list): ... this. Call prune_threads. (init_ravenscar_thread_ops): Adjust. * record-btrace.c (record_btrace_find_new_threads): Rename to ... (record_btrace_update_thread_list): ... this. Adjust comment. (init_record_btrace_ops): Adjust. * remote.c (remote_threads_info): Rename to ... (remote_update_thread_list): ... this. Call prune_threads. (remote_start_remote, extended_remote_attach_1, init_remote_ops): Adjust. * sol-thread.c (check_for_thread_db): Adjust. (sol_find_new_threads_callback): Rename to ... (sol_update_thread_list_callback): ... this. (sol_find_new_threads): Rename to ... (sol_update_thread_list): ... this. Call prune_threads. Adjust. (sol_get_ada_task_ptid, init_sol_thread_ops): Adjust. * target-delegates.c: Regenerate. * target.c (target_find_new_threads): Rename to ... (target_update_thread_list): ... this. * target.h (struct target_ops): Rename to_find_new_threads field to to_update_thread_list. (target_find_new_threads): Rename to ... (target_update_thread_list): ... this. * thread.c (prune_threads): Make extern. (update_thread_list): Adjust.
2014-10-16 05:44:00 +08:00
fprintf_unfiltered (gdb_stdlog, "-> %s->to_update_thread_list (...)\n", debug_target.to_shortname);
debug_target.to_update_thread_list (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_update_thread_list (", debug_target.to_shortname);
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (")\n", gdb_stdlog);
}
static char *
delegate_pid_to_str (struct target_ops *self, ptid_t arg1)
{
self = self->beneath;
return self->to_pid_to_str (self, arg1);
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static char *
debug_pid_to_str (struct target_ops *self, ptid_t arg1)
{
char * result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_pid_to_str (...)\n", debug_target.to_shortname);
result = debug_target.to_pid_to_str (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_pid_to_str (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_ptid_t (arg1);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_char_p (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static char *
delegate_extra_thread_info (struct target_ops *self, struct thread_info *arg1)
{
self = self->beneath;
return self->to_extra_thread_info (self, arg1);
}
static char *
tdefault_extra_thread_info (struct target_ops *self, struct thread_info *arg1)
{
return NULL;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static char *
debug_extra_thread_info (struct target_ops *self, struct thread_info *arg1)
{
char * result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_extra_thread_info (...)\n", debug_target.to_shortname);
result = debug_target.to_extra_thread_info (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_extra_thread_info (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_struct_thread_info_p (arg1);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_char_p (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static char *
delegate_thread_name (struct target_ops *self, struct thread_info *arg1)
{
self = self->beneath;
return self->to_thread_name (self, arg1);
}
static char *
tdefault_thread_name (struct target_ops *self, struct thread_info *arg1)
{
return NULL;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static char *
debug_thread_name (struct target_ops *self, struct thread_info *arg1)
{
char * result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_thread_name (...)\n", debug_target.to_shortname);
result = debug_target.to_thread_name (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_thread_name (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_struct_thread_info_p (arg1);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_char_p (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static void
delegate_stop (struct target_ops *self, ptid_t arg1)
{
self = self->beneath;
self->to_stop (self, arg1);
}
static void
tdefault_stop (struct target_ops *self, ptid_t arg1)
{
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_stop (struct target_ops *self, ptid_t arg1)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_stop (...)\n", debug_target.to_shortname);
debug_target.to_stop (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_stop (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_ptid_t (arg1);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
delegate_rcmd (struct target_ops *self, const char *arg1, struct ui_file *arg2)
{
self = self->beneath;
self->to_rcmd (self, arg1, arg2);
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_rcmd (struct target_ops *self, const char *arg1, struct ui_file *arg2)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_rcmd (...)\n", debug_target.to_shortname);
debug_target.to_rcmd (&debug_target, arg1, arg2);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_rcmd (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_const_char_p (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_struct_ui_file_p (arg2);
fputs_unfiltered (")\n", gdb_stdlog);
}
static char *
delegate_pid_to_exec_file (struct target_ops *self, int arg1)
{
self = self->beneath;
return self->to_pid_to_exec_file (self, arg1);
}
static char *
tdefault_pid_to_exec_file (struct target_ops *self, int arg1)
{
return NULL;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static char *
debug_pid_to_exec_file (struct target_ops *self, int arg1)
{
char * result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_pid_to_exec_file (...)\n", debug_target.to_shortname);
result = debug_target.to_pid_to_exec_file (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_pid_to_exec_file (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg1);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_char_p (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static void
delegate_log_command (struct target_ops *self, const char *arg1)
{
self = self->beneath;
self->to_log_command (self, arg1);
}
static void
tdefault_log_command (struct target_ops *self, const char *arg1)
{
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_log_command (struct target_ops *self, const char *arg1)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_log_command (...)\n", debug_target.to_shortname);
debug_target.to_log_command (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_log_command (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_const_char_p (arg1);
fputs_unfiltered (")\n", gdb_stdlog);
}
static struct target_section_table *
delegate_get_section_table (struct target_ops *self)
{
self = self->beneath;
return self->to_get_section_table (self);
}
static struct target_section_table *
tdefault_get_section_table (struct target_ops *self)
{
return NULL;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static struct target_section_table *
debug_get_section_table (struct target_ops *self)
{
struct target_section_table * result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_get_section_table (...)\n", debug_target.to_shortname);
result = debug_target.to_get_section_table (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_get_section_table (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_struct_target_section_table_p (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
add target method delegation This patch replaces some code in the record targets with target method delegation. record-full.c stores pointers to many target methods when the record target is pushed. Then it later delegates some calls via these. This is wrong because it violates the target stack contract. In particular it is ok to unpush a target at any stratum, but record-full does not keep track of this, so it could potentially call into an unpushed target. This patch fixes the problem by using the newly-introduced generic approach to target delegation for the methods in question. 2014-02-19 Tom Tromey <tromey@redhat.com> * record-full.c (record_full_beneath_to_resume_ops) (record_full_beneath_to_resume, record_full_beneath_to_wait_ops) (record_full_beneath_to_wait) (record_full_beneath_to_store_registers_ops) (record_full_beneath_to_store_registers) (record_full_beneath_to_xfer_partial_ops) (record_full_beneath_to_xfer_partial) (record_full_beneath_to_insert_breakpoint_ops) (record_full_beneath_to_insert_breakpoint) (record_full_beneath_to_remove_breakpoint_ops) (record_full_beneath_to_remove_breakpoint) (record_full_beneath_to_stopped_by_watchpoint) (record_full_beneath_to_stopped_data_address) (record_full_beneath_to_async, tmp_to_resume_ops, tmp_to_resume) (tmp_to_wait_ops, tmp_to_wait, tmp_to_store_registers_ops) (tmp_to_store_registers, tmp_to_xfer_partial_ops) (tmp_to_xfer_partial, tmp_to_instmp_to_insert_breakpoint_ops) (tmp_to_insert_breakpoint, tmp_to_remove_breakpoint_ops) (tmp_to_remove_breakpoint, tmp_to_stopped_by_watchpoint) (tmp_to_stopped_data_address, tmp_to_async): Remove. (record_full_open_1, record_full_open): Update. Use RECORD_IS_USED. (record_full_resume, record_full_wait_1) (record_full_stopped_by_watchpoint, record_full_stopped_data_address) (record_full_store_registers, record_full_xfer_partial) (record_full_insert_breakpoint, record_full_remove_breakpoint) (record_full_async, record_full_core_xfer_partial): Use target delegation. * target-delegates.c: Rebuild. * target.c (current_xfer_partial): Remove. (update_current_target): Do not INHERIT or de_fault to_insert_breakpoint, to_remove_breakpoint, to_stopped_data_address, to_stopped_by_watchpoint, to_can_async_p, to_is_async_p, to_async. Do not set to_xfer_partial field. (default_xfer_partial): Simplify. (current_xfer_partial): Remove. (target_wait, target_resume): Simplify. (find_default_can_async_p, find_default_is_async_p): Update. (init_dummy_target): Don't set to_can_async_p, to_is_async_p, to_xfer_partial, to_stopped_by_watchpoint, to_stopped_data_address. (target_store_registers): Simplify. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. (target_remove_breakpoint, target_insert_breakpoint) (debug_to_insert_breakpoint, debug_to_remove_breakpoint): Update. * target.h (struct target_ops) <to_resume, to_wait, to_store_registers, to_insert_breakpoint, to_remove_breakpoint, to_stopped_by_watchpoint, to_stopped_data_address, to_can_async_p, to_is_async_p, to_async, to_xfer_partial>: Add TARGET_DEFAULT markup. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. * record-btrace.c (record_btrace_remove_breakpoint): Delegate directly. (record_btrace_insert_breakpoint): Delegate directly.
2013-07-13 03:00:34 +08:00
static int
delegate_can_async_p (struct target_ops *self)
{
self = self->beneath;
return self->to_can_async_p (self);
}
fix regressions with target-async A patch in the target cleanup series caused a regression when using record with target-async. Version 4 of the patch is here: https://sourceware.org/ml/gdb-patches/2014-03/msg00159.html The immediate problem is that record supplies to_can_async_p and to_is_async_p methods, but does not supply a to_async method. So, when target-async is set, record claims to support async -- but if the underlying target does not support async, then the to_async method call will end up in that method's default implementation, namely tcomplain. This worked previously because the record target used to provide a to_async method; one that (erroneously, only at push time) checked the other members of the target stack, and then simply dropped to_async calls in the "does not implement async" case. My first thought was to simply drop tcomplain as the default for to_async. This works, but Pedro pointed out that the only reason record has to supply to_can_async_p and to_is_async_p is that these default to using the find_default_run_target machinery -- and these defaults are only needed by "run" and "attach". So, a nicer solution presents itself: change run and attach to explicitly call into the default run target when needed; and change to_is_async_p and to_can_async_p to default to "return 0". This makes the target stack simpler to use and lets us remove the method implementations from record. This is also in harmony with other plans for the target stack; namely trying to reduce the impact of find_default_run_target. This approach makes it clear that find_default_is_async_p is not needed -- it is asking whether a target that may not even be pushed is actually async, which seems like a nonsensical question. While an improvement, this approach proved to introduce the same bug when using the core target. Looking a bit deeper, the issue is that code in "attach" and "run" may need to use either the current target stack or the default run target -- but different calls into the target API in those functions could wind up querying different targets. This new patch makes the target to use more explicit in "run" and "attach". Then these commands explicitly make the needed calls against that target. This ensures that a single target is used for all relevant operations. This lets us remove a couple find_default_* functions from various targets, including the dummy target. I think this is a decent understandability improvement. One issue I see with this patch is that the new calls in "run" and "attach" are not very much like the rest of the target API. I think fundamentally this is due to bad factoring in the target API, which may need to be fixed for multi-target. Tackling that seemed ambitious for a regression fix. While working on this I noticed that there don't seem to be any test cases that involve both target-async and record, so this patch changes break-precsave.exp to add some. It also changes corefile.exp to add some target-async tests; these pass with current trunk and with this patch applied, but fail with the v1 patch. This patch differs from v4 in that it moves initialization of to_can_async_p and to_supports_non_stop into inf-child, adds some assertions to complete_target_initialization, and adds some comments to target.h. Built and regtested on x86-64 Fedora 20. 2014-03-12 Tom Tromey <tromey@redhat.com> * inf-child.c (return_zero): New function. (inf_child_target): Set to_can_async_p, to_supports_non_stop. * aix-thread.c (aix_thread_inferior_created): New function. (aix_thread_attach): Remove. (init_aix_thread_ops): Don't set to_attach. (_initialize_aix_thread): Register inferior_created observer. * corelow.c (init_core_ops): Don't set to_attach or to_create_inferior. * exec.c (init_exec_ops): Don't set to_attach or to_create_inferior. * infcmd.c (run_command_1): Use find_run_target. Make direct target calls. (attach_command): Use find_attach_target. Make direct target calls. * record-btrace.c (init_record_btrace_ops): Don't set to_create_inferior. * record-full.c (record_full_can_async_p, record_full_is_async_p): Remove. (init_record_full_ops, init_record_full_core_ops): Update. Don't set to_create_inferior. * target.c (complete_target_initialization): Add assertion. (target_create_inferior): Remove. (find_default_attach, find_default_create_inferior): Remove. (find_attach_target, find_run_target): New functions. (find_default_is_async_p, find_default_can_async_p) (target_supports_non_stop, target_attach): Remove. (init_dummy_target): Don't set to_create_inferior or to_supports_non_stop. * target.h (struct target_ops) <to_attach>: Add comment. Remove TARGET_DEFAULT_FUNC. <to_create_inferior>: Add comment. <to_can_async_p, to_is_async_p, to_supports_non_stop>: Use TARGET_DEFAULT_RETURN. <to_can_async_p, to_supports_non_stop, to_can_run>: Add comments. (find_attach_target, find_run_target): Declare. (target_create_inferior): Remove. (target_has_execution_1): Update comment. (target_supports_non_stop): Remove. * target-delegates.c: Rebuild. 2014-03-12 Tom Tromey <tromey@redhat.com> * gdb.base/corefile.exp (corefile_test_run, corefile_test_attach): New procs. Add target-async tests. * gdb.reverse/break-precsave.exp (precsave_tests): New proc. Add target-async tests.
2014-03-01 00:47:34 +08:00
static int
tdefault_can_async_p (struct target_ops *self)
{
return 0;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_can_async_p (struct target_ops *self)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_can_async_p (...)\n", debug_target.to_shortname);
result = debug_target.to_can_async_p (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_can_async_p (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
add target method delegation This patch replaces some code in the record targets with target method delegation. record-full.c stores pointers to many target methods when the record target is pushed. Then it later delegates some calls via these. This is wrong because it violates the target stack contract. In particular it is ok to unpush a target at any stratum, but record-full does not keep track of this, so it could potentially call into an unpushed target. This patch fixes the problem by using the newly-introduced generic approach to target delegation for the methods in question. 2014-02-19 Tom Tromey <tromey@redhat.com> * record-full.c (record_full_beneath_to_resume_ops) (record_full_beneath_to_resume, record_full_beneath_to_wait_ops) (record_full_beneath_to_wait) (record_full_beneath_to_store_registers_ops) (record_full_beneath_to_store_registers) (record_full_beneath_to_xfer_partial_ops) (record_full_beneath_to_xfer_partial) (record_full_beneath_to_insert_breakpoint_ops) (record_full_beneath_to_insert_breakpoint) (record_full_beneath_to_remove_breakpoint_ops) (record_full_beneath_to_remove_breakpoint) (record_full_beneath_to_stopped_by_watchpoint) (record_full_beneath_to_stopped_data_address) (record_full_beneath_to_async, tmp_to_resume_ops, tmp_to_resume) (tmp_to_wait_ops, tmp_to_wait, tmp_to_store_registers_ops) (tmp_to_store_registers, tmp_to_xfer_partial_ops) (tmp_to_xfer_partial, tmp_to_instmp_to_insert_breakpoint_ops) (tmp_to_insert_breakpoint, tmp_to_remove_breakpoint_ops) (tmp_to_remove_breakpoint, tmp_to_stopped_by_watchpoint) (tmp_to_stopped_data_address, tmp_to_async): Remove. (record_full_open_1, record_full_open): Update. Use RECORD_IS_USED. (record_full_resume, record_full_wait_1) (record_full_stopped_by_watchpoint, record_full_stopped_data_address) (record_full_store_registers, record_full_xfer_partial) (record_full_insert_breakpoint, record_full_remove_breakpoint) (record_full_async, record_full_core_xfer_partial): Use target delegation. * target-delegates.c: Rebuild. * target.c (current_xfer_partial): Remove. (update_current_target): Do not INHERIT or de_fault to_insert_breakpoint, to_remove_breakpoint, to_stopped_data_address, to_stopped_by_watchpoint, to_can_async_p, to_is_async_p, to_async. Do not set to_xfer_partial field. (default_xfer_partial): Simplify. (current_xfer_partial): Remove. (target_wait, target_resume): Simplify. (find_default_can_async_p, find_default_is_async_p): Update. (init_dummy_target): Don't set to_can_async_p, to_is_async_p, to_xfer_partial, to_stopped_by_watchpoint, to_stopped_data_address. (target_store_registers): Simplify. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. (target_remove_breakpoint, target_insert_breakpoint) (debug_to_insert_breakpoint, debug_to_remove_breakpoint): Update. * target.h (struct target_ops) <to_resume, to_wait, to_store_registers, to_insert_breakpoint, to_remove_breakpoint, to_stopped_by_watchpoint, to_stopped_data_address, to_can_async_p, to_is_async_p, to_async, to_xfer_partial>: Add TARGET_DEFAULT markup. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. * record-btrace.c (record_btrace_remove_breakpoint): Delegate directly. (record_btrace_insert_breakpoint): Delegate directly.
2013-07-13 03:00:34 +08:00
static int
delegate_is_async_p (struct target_ops *self)
{
self = self->beneath;
return self->to_is_async_p (self);
}
fix regressions with target-async A patch in the target cleanup series caused a regression when using record with target-async. Version 4 of the patch is here: https://sourceware.org/ml/gdb-patches/2014-03/msg00159.html The immediate problem is that record supplies to_can_async_p and to_is_async_p methods, but does not supply a to_async method. So, when target-async is set, record claims to support async -- but if the underlying target does not support async, then the to_async method call will end up in that method's default implementation, namely tcomplain. This worked previously because the record target used to provide a to_async method; one that (erroneously, only at push time) checked the other members of the target stack, and then simply dropped to_async calls in the "does not implement async" case. My first thought was to simply drop tcomplain as the default for to_async. This works, but Pedro pointed out that the only reason record has to supply to_can_async_p and to_is_async_p is that these default to using the find_default_run_target machinery -- and these defaults are only needed by "run" and "attach". So, a nicer solution presents itself: change run and attach to explicitly call into the default run target when needed; and change to_is_async_p and to_can_async_p to default to "return 0". This makes the target stack simpler to use and lets us remove the method implementations from record. This is also in harmony with other plans for the target stack; namely trying to reduce the impact of find_default_run_target. This approach makes it clear that find_default_is_async_p is not needed -- it is asking whether a target that may not even be pushed is actually async, which seems like a nonsensical question. While an improvement, this approach proved to introduce the same bug when using the core target. Looking a bit deeper, the issue is that code in "attach" and "run" may need to use either the current target stack or the default run target -- but different calls into the target API in those functions could wind up querying different targets. This new patch makes the target to use more explicit in "run" and "attach". Then these commands explicitly make the needed calls against that target. This ensures that a single target is used for all relevant operations. This lets us remove a couple find_default_* functions from various targets, including the dummy target. I think this is a decent understandability improvement. One issue I see with this patch is that the new calls in "run" and "attach" are not very much like the rest of the target API. I think fundamentally this is due to bad factoring in the target API, which may need to be fixed for multi-target. Tackling that seemed ambitious for a regression fix. While working on this I noticed that there don't seem to be any test cases that involve both target-async and record, so this patch changes break-precsave.exp to add some. It also changes corefile.exp to add some target-async tests; these pass with current trunk and with this patch applied, but fail with the v1 patch. This patch differs from v4 in that it moves initialization of to_can_async_p and to_supports_non_stop into inf-child, adds some assertions to complete_target_initialization, and adds some comments to target.h. Built and regtested on x86-64 Fedora 20. 2014-03-12 Tom Tromey <tromey@redhat.com> * inf-child.c (return_zero): New function. (inf_child_target): Set to_can_async_p, to_supports_non_stop. * aix-thread.c (aix_thread_inferior_created): New function. (aix_thread_attach): Remove. (init_aix_thread_ops): Don't set to_attach. (_initialize_aix_thread): Register inferior_created observer. * corelow.c (init_core_ops): Don't set to_attach or to_create_inferior. * exec.c (init_exec_ops): Don't set to_attach or to_create_inferior. * infcmd.c (run_command_1): Use find_run_target. Make direct target calls. (attach_command): Use find_attach_target. Make direct target calls. * record-btrace.c (init_record_btrace_ops): Don't set to_create_inferior. * record-full.c (record_full_can_async_p, record_full_is_async_p): Remove. (init_record_full_ops, init_record_full_core_ops): Update. Don't set to_create_inferior. * target.c (complete_target_initialization): Add assertion. (target_create_inferior): Remove. (find_default_attach, find_default_create_inferior): Remove. (find_attach_target, find_run_target): New functions. (find_default_is_async_p, find_default_can_async_p) (target_supports_non_stop, target_attach): Remove. (init_dummy_target): Don't set to_create_inferior or to_supports_non_stop. * target.h (struct target_ops) <to_attach>: Add comment. Remove TARGET_DEFAULT_FUNC. <to_create_inferior>: Add comment. <to_can_async_p, to_is_async_p, to_supports_non_stop>: Use TARGET_DEFAULT_RETURN. <to_can_async_p, to_supports_non_stop, to_can_run>: Add comments. (find_attach_target, find_run_target): Declare. (target_create_inferior): Remove. (target_has_execution_1): Update comment. (target_supports_non_stop): Remove. * target-delegates.c: Rebuild. 2014-03-12 Tom Tromey <tromey@redhat.com> * gdb.base/corefile.exp (corefile_test_run, corefile_test_attach): New procs. Add target-async tests. * gdb.reverse/break-precsave.exp (precsave_tests): New proc. Add target-async tests.
2014-03-01 00:47:34 +08:00
static int
tdefault_is_async_p (struct target_ops *self)
{
return 0;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_is_async_p (struct target_ops *self)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_is_async_p (...)\n", debug_target.to_shortname);
result = debug_target.to_is_async_p (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_is_async_p (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
add target method delegation This patch replaces some code in the record targets with target method delegation. record-full.c stores pointers to many target methods when the record target is pushed. Then it later delegates some calls via these. This is wrong because it violates the target stack contract. In particular it is ok to unpush a target at any stratum, but record-full does not keep track of this, so it could potentially call into an unpushed target. This patch fixes the problem by using the newly-introduced generic approach to target delegation for the methods in question. 2014-02-19 Tom Tromey <tromey@redhat.com> * record-full.c (record_full_beneath_to_resume_ops) (record_full_beneath_to_resume, record_full_beneath_to_wait_ops) (record_full_beneath_to_wait) (record_full_beneath_to_store_registers_ops) (record_full_beneath_to_store_registers) (record_full_beneath_to_xfer_partial_ops) (record_full_beneath_to_xfer_partial) (record_full_beneath_to_insert_breakpoint_ops) (record_full_beneath_to_insert_breakpoint) (record_full_beneath_to_remove_breakpoint_ops) (record_full_beneath_to_remove_breakpoint) (record_full_beneath_to_stopped_by_watchpoint) (record_full_beneath_to_stopped_data_address) (record_full_beneath_to_async, tmp_to_resume_ops, tmp_to_resume) (tmp_to_wait_ops, tmp_to_wait, tmp_to_store_registers_ops) (tmp_to_store_registers, tmp_to_xfer_partial_ops) (tmp_to_xfer_partial, tmp_to_instmp_to_insert_breakpoint_ops) (tmp_to_insert_breakpoint, tmp_to_remove_breakpoint_ops) (tmp_to_remove_breakpoint, tmp_to_stopped_by_watchpoint) (tmp_to_stopped_data_address, tmp_to_async): Remove. (record_full_open_1, record_full_open): Update. Use RECORD_IS_USED. (record_full_resume, record_full_wait_1) (record_full_stopped_by_watchpoint, record_full_stopped_data_address) (record_full_store_registers, record_full_xfer_partial) (record_full_insert_breakpoint, record_full_remove_breakpoint) (record_full_async, record_full_core_xfer_partial): Use target delegation. * target-delegates.c: Rebuild. * target.c (current_xfer_partial): Remove. (update_current_target): Do not INHERIT or de_fault to_insert_breakpoint, to_remove_breakpoint, to_stopped_data_address, to_stopped_by_watchpoint, to_can_async_p, to_is_async_p, to_async. Do not set to_xfer_partial field. (default_xfer_partial): Simplify. (current_xfer_partial): Remove. (target_wait, target_resume): Simplify. (find_default_can_async_p, find_default_is_async_p): Update. (init_dummy_target): Don't set to_can_async_p, to_is_async_p, to_xfer_partial, to_stopped_by_watchpoint, to_stopped_data_address. (target_store_registers): Simplify. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. (target_remove_breakpoint, target_insert_breakpoint) (debug_to_insert_breakpoint, debug_to_remove_breakpoint): Update. * target.h (struct target_ops) <to_resume, to_wait, to_store_registers, to_insert_breakpoint, to_remove_breakpoint, to_stopped_by_watchpoint, to_stopped_data_address, to_can_async_p, to_is_async_p, to_async, to_xfer_partial>: Add TARGET_DEFAULT markup. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. * record-btrace.c (record_btrace_remove_breakpoint): Delegate directly. (record_btrace_insert_breakpoint): Delegate directly.
2013-07-13 03:00:34 +08:00
static void
Simplify target_async hook interface All callers of target_async pass it the same callback (inferior_event_handler). Since both common code and target backends need to be able to put the target in and out of target async mode at any given time, there's really no way that a different callback could be passed. This commit simplifies things, and removes the indirection altogether. Bonus: with this, gdb's target_async method ends up with the same signature as gdbserver's. Tested on x86_64 Fedora 20, native and gdbserver. gdb/ChangeLog: 2015-03-25 Pedro Alves <palves@redhat.com> * target.h <to_async>: Replace 'callback' and 'context' parameters with boolean 'enable' parameter. (target_async): Replace CALLBACK and CONTEXT parameters with boolean ENABLE parameter. * inf-loop.c (inferior_event_handler): Adjust. * linux-nat.c (linux_nat_attach, linux_nat_resume) (linux_nat_resume): Adjust. (async_client_callback, async_client_context): Delete. (handle_target_event): Call inferior_event_handler directly. (linux_nat_async): Replace 'callback' and 'context' parameters with boolean 'enable' parameter. Adjust. Remove references to async_client_callback and async_client_context. (linux_nat_close): Adjust. * record-btrace.c (record_btrace_async): Replace 'callback' and 'context' parameters with boolean 'enable' parameter. Adjust. (record_btrace_resume): Adjust. * record-full.c (record_full_async): Replace 'callback' and 'context' parameters with boolean 'enable' parameter. Adjust. (record_full_resume, record_full_core_resume): Adjust. * remote.c (struct remote_state) <async_client_callback, async_client_context>: Delete fields. (remote_start_remote, extended_remote_attach_1, remote_resume) (extended_remote_create_inferior): Adjust. (remote_async_serial_handler): Call inferior_event_handler directly. (remote_async): Replace 'callback' and 'context' parameters with boolean 'enable' parameter. Adjust. * top.c (gdb_readline_wrapper_cleanup, gdb_readline_wrapper): Adjust. * target-delegates.c: Regenerate.
2015-03-25 19:28:31 +08:00
delegate_async (struct target_ops *self, int arg1)
add target method delegation This patch replaces some code in the record targets with target method delegation. record-full.c stores pointers to many target methods when the record target is pushed. Then it later delegates some calls via these. This is wrong because it violates the target stack contract. In particular it is ok to unpush a target at any stratum, but record-full does not keep track of this, so it could potentially call into an unpushed target. This patch fixes the problem by using the newly-introduced generic approach to target delegation for the methods in question. 2014-02-19 Tom Tromey <tromey@redhat.com> * record-full.c (record_full_beneath_to_resume_ops) (record_full_beneath_to_resume, record_full_beneath_to_wait_ops) (record_full_beneath_to_wait) (record_full_beneath_to_store_registers_ops) (record_full_beneath_to_store_registers) (record_full_beneath_to_xfer_partial_ops) (record_full_beneath_to_xfer_partial) (record_full_beneath_to_insert_breakpoint_ops) (record_full_beneath_to_insert_breakpoint) (record_full_beneath_to_remove_breakpoint_ops) (record_full_beneath_to_remove_breakpoint) (record_full_beneath_to_stopped_by_watchpoint) (record_full_beneath_to_stopped_data_address) (record_full_beneath_to_async, tmp_to_resume_ops, tmp_to_resume) (tmp_to_wait_ops, tmp_to_wait, tmp_to_store_registers_ops) (tmp_to_store_registers, tmp_to_xfer_partial_ops) (tmp_to_xfer_partial, tmp_to_instmp_to_insert_breakpoint_ops) (tmp_to_insert_breakpoint, tmp_to_remove_breakpoint_ops) (tmp_to_remove_breakpoint, tmp_to_stopped_by_watchpoint) (tmp_to_stopped_data_address, tmp_to_async): Remove. (record_full_open_1, record_full_open): Update. Use RECORD_IS_USED. (record_full_resume, record_full_wait_1) (record_full_stopped_by_watchpoint, record_full_stopped_data_address) (record_full_store_registers, record_full_xfer_partial) (record_full_insert_breakpoint, record_full_remove_breakpoint) (record_full_async, record_full_core_xfer_partial): Use target delegation. * target-delegates.c: Rebuild. * target.c (current_xfer_partial): Remove. (update_current_target): Do not INHERIT or de_fault to_insert_breakpoint, to_remove_breakpoint, to_stopped_data_address, to_stopped_by_watchpoint, to_can_async_p, to_is_async_p, to_async. Do not set to_xfer_partial field. (default_xfer_partial): Simplify. (current_xfer_partial): Remove. (target_wait, target_resume): Simplify. (find_default_can_async_p, find_default_is_async_p): Update. (init_dummy_target): Don't set to_can_async_p, to_is_async_p, to_xfer_partial, to_stopped_by_watchpoint, to_stopped_data_address. (target_store_registers): Simplify. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. (target_remove_breakpoint, target_insert_breakpoint) (debug_to_insert_breakpoint, debug_to_remove_breakpoint): Update. * target.h (struct target_ops) <to_resume, to_wait, to_store_registers, to_insert_breakpoint, to_remove_breakpoint, to_stopped_by_watchpoint, to_stopped_data_address, to_can_async_p, to_is_async_p, to_async, to_xfer_partial>: Add TARGET_DEFAULT markup. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. * record-btrace.c (record_btrace_remove_breakpoint): Delegate directly. (record_btrace_insert_breakpoint): Delegate directly.
2013-07-13 03:00:34 +08:00
{
self = self->beneath;
Simplify target_async hook interface All callers of target_async pass it the same callback (inferior_event_handler). Since both common code and target backends need to be able to put the target in and out of target async mode at any given time, there's really no way that a different callback could be passed. This commit simplifies things, and removes the indirection altogether. Bonus: with this, gdb's target_async method ends up with the same signature as gdbserver's. Tested on x86_64 Fedora 20, native and gdbserver. gdb/ChangeLog: 2015-03-25 Pedro Alves <palves@redhat.com> * target.h <to_async>: Replace 'callback' and 'context' parameters with boolean 'enable' parameter. (target_async): Replace CALLBACK and CONTEXT parameters with boolean ENABLE parameter. * inf-loop.c (inferior_event_handler): Adjust. * linux-nat.c (linux_nat_attach, linux_nat_resume) (linux_nat_resume): Adjust. (async_client_callback, async_client_context): Delete. (handle_target_event): Call inferior_event_handler directly. (linux_nat_async): Replace 'callback' and 'context' parameters with boolean 'enable' parameter. Adjust. Remove references to async_client_callback and async_client_context. (linux_nat_close): Adjust. * record-btrace.c (record_btrace_async): Replace 'callback' and 'context' parameters with boolean 'enable' parameter. Adjust. (record_btrace_resume): Adjust. * record-full.c (record_full_async): Replace 'callback' and 'context' parameters with boolean 'enable' parameter. Adjust. (record_full_resume, record_full_core_resume): Adjust. * remote.c (struct remote_state) <async_client_callback, async_client_context>: Delete fields. (remote_start_remote, extended_remote_attach_1, remote_resume) (extended_remote_create_inferior): Adjust. (remote_async_serial_handler): Call inferior_event_handler directly. (remote_async): Replace 'callback' and 'context' parameters with boolean 'enable' parameter. Adjust. * top.c (gdb_readline_wrapper_cleanup, gdb_readline_wrapper): Adjust. * target-delegates.c: Regenerate.
2015-03-25 19:28:31 +08:00
self->to_async (self, arg1);
add target method delegation This patch replaces some code in the record targets with target method delegation. record-full.c stores pointers to many target methods when the record target is pushed. Then it later delegates some calls via these. This is wrong because it violates the target stack contract. In particular it is ok to unpush a target at any stratum, but record-full does not keep track of this, so it could potentially call into an unpushed target. This patch fixes the problem by using the newly-introduced generic approach to target delegation for the methods in question. 2014-02-19 Tom Tromey <tromey@redhat.com> * record-full.c (record_full_beneath_to_resume_ops) (record_full_beneath_to_resume, record_full_beneath_to_wait_ops) (record_full_beneath_to_wait) (record_full_beneath_to_store_registers_ops) (record_full_beneath_to_store_registers) (record_full_beneath_to_xfer_partial_ops) (record_full_beneath_to_xfer_partial) (record_full_beneath_to_insert_breakpoint_ops) (record_full_beneath_to_insert_breakpoint) (record_full_beneath_to_remove_breakpoint_ops) (record_full_beneath_to_remove_breakpoint) (record_full_beneath_to_stopped_by_watchpoint) (record_full_beneath_to_stopped_data_address) (record_full_beneath_to_async, tmp_to_resume_ops, tmp_to_resume) (tmp_to_wait_ops, tmp_to_wait, tmp_to_store_registers_ops) (tmp_to_store_registers, tmp_to_xfer_partial_ops) (tmp_to_xfer_partial, tmp_to_instmp_to_insert_breakpoint_ops) (tmp_to_insert_breakpoint, tmp_to_remove_breakpoint_ops) (tmp_to_remove_breakpoint, tmp_to_stopped_by_watchpoint) (tmp_to_stopped_data_address, tmp_to_async): Remove. (record_full_open_1, record_full_open): Update. Use RECORD_IS_USED. (record_full_resume, record_full_wait_1) (record_full_stopped_by_watchpoint, record_full_stopped_data_address) (record_full_store_registers, record_full_xfer_partial) (record_full_insert_breakpoint, record_full_remove_breakpoint) (record_full_async, record_full_core_xfer_partial): Use target delegation. * target-delegates.c: Rebuild. * target.c (current_xfer_partial): Remove. (update_current_target): Do not INHERIT or de_fault to_insert_breakpoint, to_remove_breakpoint, to_stopped_data_address, to_stopped_by_watchpoint, to_can_async_p, to_is_async_p, to_async. Do not set to_xfer_partial field. (default_xfer_partial): Simplify. (current_xfer_partial): Remove. (target_wait, target_resume): Simplify. (find_default_can_async_p, find_default_is_async_p): Update. (init_dummy_target): Don't set to_can_async_p, to_is_async_p, to_xfer_partial, to_stopped_by_watchpoint, to_stopped_data_address. (target_store_registers): Simplify. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. (target_remove_breakpoint, target_insert_breakpoint) (debug_to_insert_breakpoint, debug_to_remove_breakpoint): Update. * target.h (struct target_ops) <to_resume, to_wait, to_store_registers, to_insert_breakpoint, to_remove_breakpoint, to_stopped_by_watchpoint, to_stopped_data_address, to_can_async_p, to_is_async_p, to_async, to_xfer_partial>: Add TARGET_DEFAULT markup. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. * record-btrace.c (record_btrace_remove_breakpoint): Delegate directly. (record_btrace_insert_breakpoint): Delegate directly.
2013-07-13 03:00:34 +08:00
}
static void
Simplify target_async hook interface All callers of target_async pass it the same callback (inferior_event_handler). Since both common code and target backends need to be able to put the target in and out of target async mode at any given time, there's really no way that a different callback could be passed. This commit simplifies things, and removes the indirection altogether. Bonus: with this, gdb's target_async method ends up with the same signature as gdbserver's. Tested on x86_64 Fedora 20, native and gdbserver. gdb/ChangeLog: 2015-03-25 Pedro Alves <palves@redhat.com> * target.h <to_async>: Replace 'callback' and 'context' parameters with boolean 'enable' parameter. (target_async): Replace CALLBACK and CONTEXT parameters with boolean ENABLE parameter. * inf-loop.c (inferior_event_handler): Adjust. * linux-nat.c (linux_nat_attach, linux_nat_resume) (linux_nat_resume): Adjust. (async_client_callback, async_client_context): Delete. (handle_target_event): Call inferior_event_handler directly. (linux_nat_async): Replace 'callback' and 'context' parameters with boolean 'enable' parameter. Adjust. Remove references to async_client_callback and async_client_context. (linux_nat_close): Adjust. * record-btrace.c (record_btrace_async): Replace 'callback' and 'context' parameters with boolean 'enable' parameter. Adjust. (record_btrace_resume): Adjust. * record-full.c (record_full_async): Replace 'callback' and 'context' parameters with boolean 'enable' parameter. Adjust. (record_full_resume, record_full_core_resume): Adjust. * remote.c (struct remote_state) <async_client_callback, async_client_context>: Delete fields. (remote_start_remote, extended_remote_attach_1, remote_resume) (extended_remote_create_inferior): Adjust. (remote_async_serial_handler): Call inferior_event_handler directly. (remote_async): Replace 'callback' and 'context' parameters with boolean 'enable' parameter. Adjust. * top.c (gdb_readline_wrapper_cleanup, gdb_readline_wrapper): Adjust. * target-delegates.c: Regenerate.
2015-03-25 19:28:31 +08:00
tdefault_async (struct target_ops *self, int arg1)
add target method delegation This patch replaces some code in the record targets with target method delegation. record-full.c stores pointers to many target methods when the record target is pushed. Then it later delegates some calls via these. This is wrong because it violates the target stack contract. In particular it is ok to unpush a target at any stratum, but record-full does not keep track of this, so it could potentially call into an unpushed target. This patch fixes the problem by using the newly-introduced generic approach to target delegation for the methods in question. 2014-02-19 Tom Tromey <tromey@redhat.com> * record-full.c (record_full_beneath_to_resume_ops) (record_full_beneath_to_resume, record_full_beneath_to_wait_ops) (record_full_beneath_to_wait) (record_full_beneath_to_store_registers_ops) (record_full_beneath_to_store_registers) (record_full_beneath_to_xfer_partial_ops) (record_full_beneath_to_xfer_partial) (record_full_beneath_to_insert_breakpoint_ops) (record_full_beneath_to_insert_breakpoint) (record_full_beneath_to_remove_breakpoint_ops) (record_full_beneath_to_remove_breakpoint) (record_full_beneath_to_stopped_by_watchpoint) (record_full_beneath_to_stopped_data_address) (record_full_beneath_to_async, tmp_to_resume_ops, tmp_to_resume) (tmp_to_wait_ops, tmp_to_wait, tmp_to_store_registers_ops) (tmp_to_store_registers, tmp_to_xfer_partial_ops) (tmp_to_xfer_partial, tmp_to_instmp_to_insert_breakpoint_ops) (tmp_to_insert_breakpoint, tmp_to_remove_breakpoint_ops) (tmp_to_remove_breakpoint, tmp_to_stopped_by_watchpoint) (tmp_to_stopped_data_address, tmp_to_async): Remove. (record_full_open_1, record_full_open): Update. Use RECORD_IS_USED. (record_full_resume, record_full_wait_1) (record_full_stopped_by_watchpoint, record_full_stopped_data_address) (record_full_store_registers, record_full_xfer_partial) (record_full_insert_breakpoint, record_full_remove_breakpoint) (record_full_async, record_full_core_xfer_partial): Use target delegation. * target-delegates.c: Rebuild. * target.c (current_xfer_partial): Remove. (update_current_target): Do not INHERIT or de_fault to_insert_breakpoint, to_remove_breakpoint, to_stopped_data_address, to_stopped_by_watchpoint, to_can_async_p, to_is_async_p, to_async. Do not set to_xfer_partial field. (default_xfer_partial): Simplify. (current_xfer_partial): Remove. (target_wait, target_resume): Simplify. (find_default_can_async_p, find_default_is_async_p): Update. (init_dummy_target): Don't set to_can_async_p, to_is_async_p, to_xfer_partial, to_stopped_by_watchpoint, to_stopped_data_address. (target_store_registers): Simplify. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. (target_remove_breakpoint, target_insert_breakpoint) (debug_to_insert_breakpoint, debug_to_remove_breakpoint): Update. * target.h (struct target_ops) <to_resume, to_wait, to_store_registers, to_insert_breakpoint, to_remove_breakpoint, to_stopped_by_watchpoint, to_stopped_data_address, to_can_async_p, to_is_async_p, to_async, to_xfer_partial>: Add TARGET_DEFAULT markup. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. * record-btrace.c (record_btrace_remove_breakpoint): Delegate directly. (record_btrace_insert_breakpoint): Delegate directly.
2013-07-13 03:00:34 +08:00
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
Simplify target_async hook interface All callers of target_async pass it the same callback (inferior_event_handler). Since both common code and target backends need to be able to put the target in and out of target async mode at any given time, there's really no way that a different callback could be passed. This commit simplifies things, and removes the indirection altogether. Bonus: with this, gdb's target_async method ends up with the same signature as gdbserver's. Tested on x86_64 Fedora 20, native and gdbserver. gdb/ChangeLog: 2015-03-25 Pedro Alves <palves@redhat.com> * target.h <to_async>: Replace 'callback' and 'context' parameters with boolean 'enable' parameter. (target_async): Replace CALLBACK and CONTEXT parameters with boolean ENABLE parameter. * inf-loop.c (inferior_event_handler): Adjust. * linux-nat.c (linux_nat_attach, linux_nat_resume) (linux_nat_resume): Adjust. (async_client_callback, async_client_context): Delete. (handle_target_event): Call inferior_event_handler directly. (linux_nat_async): Replace 'callback' and 'context' parameters with boolean 'enable' parameter. Adjust. Remove references to async_client_callback and async_client_context. (linux_nat_close): Adjust. * record-btrace.c (record_btrace_async): Replace 'callback' and 'context' parameters with boolean 'enable' parameter. Adjust. (record_btrace_resume): Adjust. * record-full.c (record_full_async): Replace 'callback' and 'context' parameters with boolean 'enable' parameter. Adjust. (record_full_resume, record_full_core_resume): Adjust. * remote.c (struct remote_state) <async_client_callback, async_client_context>: Delete fields. (remote_start_remote, extended_remote_attach_1, remote_resume) (extended_remote_create_inferior): Adjust. (remote_async_serial_handler): Call inferior_event_handler directly. (remote_async): Replace 'callback' and 'context' parameters with boolean 'enable' parameter. Adjust. * top.c (gdb_readline_wrapper_cleanup, gdb_readline_wrapper): Adjust. * target-delegates.c: Regenerate.
2015-03-25 19:28:31 +08:00
debug_async (struct target_ops *self, int arg1)
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_async (...)\n", debug_target.to_shortname);
Simplify target_async hook interface All callers of target_async pass it the same callback (inferior_event_handler). Since both common code and target backends need to be able to put the target in and out of target async mode at any given time, there's really no way that a different callback could be passed. This commit simplifies things, and removes the indirection altogether. Bonus: with this, gdb's target_async method ends up with the same signature as gdbserver's. Tested on x86_64 Fedora 20, native and gdbserver. gdb/ChangeLog: 2015-03-25 Pedro Alves <palves@redhat.com> * target.h <to_async>: Replace 'callback' and 'context' parameters with boolean 'enable' parameter. (target_async): Replace CALLBACK and CONTEXT parameters with boolean ENABLE parameter. * inf-loop.c (inferior_event_handler): Adjust. * linux-nat.c (linux_nat_attach, linux_nat_resume) (linux_nat_resume): Adjust. (async_client_callback, async_client_context): Delete. (handle_target_event): Call inferior_event_handler directly. (linux_nat_async): Replace 'callback' and 'context' parameters with boolean 'enable' parameter. Adjust. Remove references to async_client_callback and async_client_context. (linux_nat_close): Adjust. * record-btrace.c (record_btrace_async): Replace 'callback' and 'context' parameters with boolean 'enable' parameter. Adjust. (record_btrace_resume): Adjust. * record-full.c (record_full_async): Replace 'callback' and 'context' parameters with boolean 'enable' parameter. Adjust. (record_full_resume, record_full_core_resume): Adjust. * remote.c (struct remote_state) <async_client_callback, async_client_context>: Delete fields. (remote_start_remote, extended_remote_attach_1, remote_resume) (extended_remote_create_inferior): Adjust. (remote_async_serial_handler): Call inferior_event_handler directly. (remote_async): Replace 'callback' and 'context' parameters with boolean 'enable' parameter. Adjust. * top.c (gdb_readline_wrapper_cleanup, gdb_readline_wrapper): Adjust. * target-delegates.c: Regenerate.
2015-03-25 19:28:31 +08:00
debug_target.to_async (&debug_target, arg1);
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
fprintf_unfiltered (gdb_stdlog, "<- %s->to_async (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
Simplify target_async hook interface All callers of target_async pass it the same callback (inferior_event_handler). Since both common code and target backends need to be able to put the target in and out of target async mode at any given time, there's really no way that a different callback could be passed. This commit simplifies things, and removes the indirection altogether. Bonus: with this, gdb's target_async method ends up with the same signature as gdbserver's. Tested on x86_64 Fedora 20, native and gdbserver. gdb/ChangeLog: 2015-03-25 Pedro Alves <palves@redhat.com> * target.h <to_async>: Replace 'callback' and 'context' parameters with boolean 'enable' parameter. (target_async): Replace CALLBACK and CONTEXT parameters with boolean ENABLE parameter. * inf-loop.c (inferior_event_handler): Adjust. * linux-nat.c (linux_nat_attach, linux_nat_resume) (linux_nat_resume): Adjust. (async_client_callback, async_client_context): Delete. (handle_target_event): Call inferior_event_handler directly. (linux_nat_async): Replace 'callback' and 'context' parameters with boolean 'enable' parameter. Adjust. Remove references to async_client_callback and async_client_context. (linux_nat_close): Adjust. * record-btrace.c (record_btrace_async): Replace 'callback' and 'context' parameters with boolean 'enable' parameter. Adjust. (record_btrace_resume): Adjust. * record-full.c (record_full_async): Replace 'callback' and 'context' parameters with boolean 'enable' parameter. Adjust. (record_full_resume, record_full_core_resume): Adjust. * remote.c (struct remote_state) <async_client_callback, async_client_context>: Delete fields. (remote_start_remote, extended_remote_attach_1, remote_resume) (extended_remote_create_inferior): Adjust. (remote_async_serial_handler): Call inferior_event_handler directly. (remote_async): Replace 'callback' and 'context' parameters with boolean 'enable' parameter. Adjust. * top.c (gdb_readline_wrapper_cleanup, gdb_readline_wrapper): Adjust. * target-delegates.c: Regenerate.
2015-03-25 19:28:31 +08:00
target_debug_print_int (arg1);
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
fputs_unfiltered (")\n", gdb_stdlog);
}
fix regressions with target-async A patch in the target cleanup series caused a regression when using record with target-async. Version 4 of the patch is here: https://sourceware.org/ml/gdb-patches/2014-03/msg00159.html The immediate problem is that record supplies to_can_async_p and to_is_async_p methods, but does not supply a to_async method. So, when target-async is set, record claims to support async -- but if the underlying target does not support async, then the to_async method call will end up in that method's default implementation, namely tcomplain. This worked previously because the record target used to provide a to_async method; one that (erroneously, only at push time) checked the other members of the target stack, and then simply dropped to_async calls in the "does not implement async" case. My first thought was to simply drop tcomplain as the default for to_async. This works, but Pedro pointed out that the only reason record has to supply to_can_async_p and to_is_async_p is that these default to using the find_default_run_target machinery -- and these defaults are only needed by "run" and "attach". So, a nicer solution presents itself: change run and attach to explicitly call into the default run target when needed; and change to_is_async_p and to_can_async_p to default to "return 0". This makes the target stack simpler to use and lets us remove the method implementations from record. This is also in harmony with other plans for the target stack; namely trying to reduce the impact of find_default_run_target. This approach makes it clear that find_default_is_async_p is not needed -- it is asking whether a target that may not even be pushed is actually async, which seems like a nonsensical question. While an improvement, this approach proved to introduce the same bug when using the core target. Looking a bit deeper, the issue is that code in "attach" and "run" may need to use either the current target stack or the default run target -- but different calls into the target API in those functions could wind up querying different targets. This new patch makes the target to use more explicit in "run" and "attach". Then these commands explicitly make the needed calls against that target. This ensures that a single target is used for all relevant operations. This lets us remove a couple find_default_* functions from various targets, including the dummy target. I think this is a decent understandability improvement. One issue I see with this patch is that the new calls in "run" and "attach" are not very much like the rest of the target API. I think fundamentally this is due to bad factoring in the target API, which may need to be fixed for multi-target. Tackling that seemed ambitious for a regression fix. While working on this I noticed that there don't seem to be any test cases that involve both target-async and record, so this patch changes break-precsave.exp to add some. It also changes corefile.exp to add some target-async tests; these pass with current trunk and with this patch applied, but fail with the v1 patch. This patch differs from v4 in that it moves initialization of to_can_async_p and to_supports_non_stop into inf-child, adds some assertions to complete_target_initialization, and adds some comments to target.h. Built and regtested on x86-64 Fedora 20. 2014-03-12 Tom Tromey <tromey@redhat.com> * inf-child.c (return_zero): New function. (inf_child_target): Set to_can_async_p, to_supports_non_stop. * aix-thread.c (aix_thread_inferior_created): New function. (aix_thread_attach): Remove. (init_aix_thread_ops): Don't set to_attach. (_initialize_aix_thread): Register inferior_created observer. * corelow.c (init_core_ops): Don't set to_attach or to_create_inferior. * exec.c (init_exec_ops): Don't set to_attach or to_create_inferior. * infcmd.c (run_command_1): Use find_run_target. Make direct target calls. (attach_command): Use find_attach_target. Make direct target calls. * record-btrace.c (init_record_btrace_ops): Don't set to_create_inferior. * record-full.c (record_full_can_async_p, record_full_is_async_p): Remove. (init_record_full_ops, init_record_full_core_ops): Update. Don't set to_create_inferior. * target.c (complete_target_initialization): Add assertion. (target_create_inferior): Remove. (find_default_attach, find_default_create_inferior): Remove. (find_attach_target, find_run_target): New functions. (find_default_is_async_p, find_default_can_async_p) (target_supports_non_stop, target_attach): Remove. (init_dummy_target): Don't set to_create_inferior or to_supports_non_stop. * target.h (struct target_ops) <to_attach>: Add comment. Remove TARGET_DEFAULT_FUNC. <to_create_inferior>: Add comment. <to_can_async_p, to_is_async_p, to_supports_non_stop>: Use TARGET_DEFAULT_RETURN. <to_can_async_p, to_supports_non_stop, to_can_run>: Add comments. (find_attach_target, find_run_target): Declare. (target_create_inferior): Remove. (target_has_execution_1): Update comment. (target_supports_non_stop): Remove. * target-delegates.c: Rebuild. 2014-03-12 Tom Tromey <tromey@redhat.com> * gdb.base/corefile.exp (corefile_test_run, corefile_test_attach): New procs. Add target-async tests. * gdb.reverse/break-precsave.exp (precsave_tests): New proc. Add target-async tests.
2014-03-01 00:47:34 +08:00
static int
delegate_supports_non_stop (struct target_ops *self)
{
self = self->beneath;
return self->to_supports_non_stop (self);
}
static int
tdefault_supports_non_stop (struct target_ops *self)
{
return 0;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_supports_non_stop (struct target_ops *self)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_supports_non_stop (...)\n", debug_target.to_shortname);
result = debug_target.to_supports_non_stop (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_supports_non_stop (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_find_memory_regions (struct target_ops *self, find_memory_region_ftype arg1, void *arg2)
{
self = self->beneath;
return self->to_find_memory_regions (self, arg1, arg2);
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_find_memory_regions (struct target_ops *self, find_memory_region_ftype arg1, void *arg2)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_find_memory_regions (...)\n", debug_target.to_shortname);
result = debug_target.to_find_memory_regions (&debug_target, arg1, arg2);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_find_memory_regions (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_find_memory_region_ftype (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_void_p (arg2);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static char *
delegate_make_corefile_notes (struct target_ops *self, bfd *arg1, int *arg2)
{
self = self->beneath;
return self->to_make_corefile_notes (self, arg1, arg2);
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static char *
debug_make_corefile_notes (struct target_ops *self, bfd *arg1, int *arg2)
{
char * result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_make_corefile_notes (...)\n", debug_target.to_shortname);
result = debug_target.to_make_corefile_notes (&debug_target, arg1, arg2);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_make_corefile_notes (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_bfd_p (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int_p (arg2);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_char_p (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static gdb_byte *
delegate_get_bookmark (struct target_ops *self, const char *arg1, int arg2)
{
self = self->beneath;
return self->to_get_bookmark (self, arg1, arg2);
}
static gdb_byte *
tdefault_get_bookmark (struct target_ops *self, const char *arg1, int arg2)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static gdb_byte *
debug_get_bookmark (struct target_ops *self, const char *arg1, int arg2)
{
gdb_byte * result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_get_bookmark (...)\n", debug_target.to_shortname);
result = debug_target.to_get_bookmark (&debug_target, arg1, arg2);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_get_bookmark (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_const_char_p (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg2);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_gdb_byte_p (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static void
delegate_goto_bookmark (struct target_ops *self, const gdb_byte *arg1, int arg2)
{
self = self->beneath;
self->to_goto_bookmark (self, arg1, arg2);
}
static void
tdefault_goto_bookmark (struct target_ops *self, const gdb_byte *arg1, int arg2)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_goto_bookmark (struct target_ops *self, const gdb_byte *arg1, int arg2)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_goto_bookmark (...)\n", debug_target.to_shortname);
debug_target.to_goto_bookmark (&debug_target, arg1, arg2);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_goto_bookmark (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_const_gdb_byte_p (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg2);
fputs_unfiltered (")\n", gdb_stdlog);
}
static CORE_ADDR
delegate_get_thread_local_address (struct target_ops *self, ptid_t arg1, CORE_ADDR arg2, CORE_ADDR arg3)
{
self = self->beneath;
return self->to_get_thread_local_address (self, arg1, arg2, arg3);
}
static CORE_ADDR
tdefault_get_thread_local_address (struct target_ops *self, ptid_t arg1, CORE_ADDR arg2, CORE_ADDR arg3)
{
generic_tls_error ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static CORE_ADDR
debug_get_thread_local_address (struct target_ops *self, ptid_t arg1, CORE_ADDR arg2, CORE_ADDR arg3)
{
CORE_ADDR result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_get_thread_local_address (...)\n", debug_target.to_shortname);
result = debug_target.to_get_thread_local_address (&debug_target, arg1, arg2, arg3);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_get_thread_local_address (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_ptid_t (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_CORE_ADDR (arg2);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_CORE_ADDR (arg3);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_CORE_ADDR (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static enum target_xfer_status
delegate_xfer_partial (struct target_ops *self, enum target_object arg1, const char *arg2, gdb_byte *arg3, const gdb_byte *arg4, ULONGEST arg5, ULONGEST arg6, ULONGEST *arg7)
add target method delegation This patch replaces some code in the record targets with target method delegation. record-full.c stores pointers to many target methods when the record target is pushed. Then it later delegates some calls via these. This is wrong because it violates the target stack contract. In particular it is ok to unpush a target at any stratum, but record-full does not keep track of this, so it could potentially call into an unpushed target. This patch fixes the problem by using the newly-introduced generic approach to target delegation for the methods in question. 2014-02-19 Tom Tromey <tromey@redhat.com> * record-full.c (record_full_beneath_to_resume_ops) (record_full_beneath_to_resume, record_full_beneath_to_wait_ops) (record_full_beneath_to_wait) (record_full_beneath_to_store_registers_ops) (record_full_beneath_to_store_registers) (record_full_beneath_to_xfer_partial_ops) (record_full_beneath_to_xfer_partial) (record_full_beneath_to_insert_breakpoint_ops) (record_full_beneath_to_insert_breakpoint) (record_full_beneath_to_remove_breakpoint_ops) (record_full_beneath_to_remove_breakpoint) (record_full_beneath_to_stopped_by_watchpoint) (record_full_beneath_to_stopped_data_address) (record_full_beneath_to_async, tmp_to_resume_ops, tmp_to_resume) (tmp_to_wait_ops, tmp_to_wait, tmp_to_store_registers_ops) (tmp_to_store_registers, tmp_to_xfer_partial_ops) (tmp_to_xfer_partial, tmp_to_instmp_to_insert_breakpoint_ops) (tmp_to_insert_breakpoint, tmp_to_remove_breakpoint_ops) (tmp_to_remove_breakpoint, tmp_to_stopped_by_watchpoint) (tmp_to_stopped_data_address, tmp_to_async): Remove. (record_full_open_1, record_full_open): Update. Use RECORD_IS_USED. (record_full_resume, record_full_wait_1) (record_full_stopped_by_watchpoint, record_full_stopped_data_address) (record_full_store_registers, record_full_xfer_partial) (record_full_insert_breakpoint, record_full_remove_breakpoint) (record_full_async, record_full_core_xfer_partial): Use target delegation. * target-delegates.c: Rebuild. * target.c (current_xfer_partial): Remove. (update_current_target): Do not INHERIT or de_fault to_insert_breakpoint, to_remove_breakpoint, to_stopped_data_address, to_stopped_by_watchpoint, to_can_async_p, to_is_async_p, to_async. Do not set to_xfer_partial field. (default_xfer_partial): Simplify. (current_xfer_partial): Remove. (target_wait, target_resume): Simplify. (find_default_can_async_p, find_default_is_async_p): Update. (init_dummy_target): Don't set to_can_async_p, to_is_async_p, to_xfer_partial, to_stopped_by_watchpoint, to_stopped_data_address. (target_store_registers): Simplify. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. (target_remove_breakpoint, target_insert_breakpoint) (debug_to_insert_breakpoint, debug_to_remove_breakpoint): Update. * target.h (struct target_ops) <to_resume, to_wait, to_store_registers, to_insert_breakpoint, to_remove_breakpoint, to_stopped_by_watchpoint, to_stopped_data_address, to_can_async_p, to_is_async_p, to_async, to_xfer_partial>: Add TARGET_DEFAULT markup. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. * record-btrace.c (record_btrace_remove_breakpoint): Delegate directly. (record_btrace_insert_breakpoint): Delegate directly.
2013-07-13 03:00:34 +08:00
{
self = self->beneath;
return self->to_xfer_partial (self, arg1, arg2, arg3, arg4, arg5, arg6, arg7);
}
static enum target_xfer_status
tdefault_xfer_partial (struct target_ops *self, enum target_object arg1, const char *arg2, gdb_byte *arg3, const gdb_byte *arg4, ULONGEST arg5, ULONGEST arg6, ULONGEST *arg7)
add target method delegation This patch replaces some code in the record targets with target method delegation. record-full.c stores pointers to many target methods when the record target is pushed. Then it later delegates some calls via these. This is wrong because it violates the target stack contract. In particular it is ok to unpush a target at any stratum, but record-full does not keep track of this, so it could potentially call into an unpushed target. This patch fixes the problem by using the newly-introduced generic approach to target delegation for the methods in question. 2014-02-19 Tom Tromey <tromey@redhat.com> * record-full.c (record_full_beneath_to_resume_ops) (record_full_beneath_to_resume, record_full_beneath_to_wait_ops) (record_full_beneath_to_wait) (record_full_beneath_to_store_registers_ops) (record_full_beneath_to_store_registers) (record_full_beneath_to_xfer_partial_ops) (record_full_beneath_to_xfer_partial) (record_full_beneath_to_insert_breakpoint_ops) (record_full_beneath_to_insert_breakpoint) (record_full_beneath_to_remove_breakpoint_ops) (record_full_beneath_to_remove_breakpoint) (record_full_beneath_to_stopped_by_watchpoint) (record_full_beneath_to_stopped_data_address) (record_full_beneath_to_async, tmp_to_resume_ops, tmp_to_resume) (tmp_to_wait_ops, tmp_to_wait, tmp_to_store_registers_ops) (tmp_to_store_registers, tmp_to_xfer_partial_ops) (tmp_to_xfer_partial, tmp_to_instmp_to_insert_breakpoint_ops) (tmp_to_insert_breakpoint, tmp_to_remove_breakpoint_ops) (tmp_to_remove_breakpoint, tmp_to_stopped_by_watchpoint) (tmp_to_stopped_data_address, tmp_to_async): Remove. (record_full_open_1, record_full_open): Update. Use RECORD_IS_USED. (record_full_resume, record_full_wait_1) (record_full_stopped_by_watchpoint, record_full_stopped_data_address) (record_full_store_registers, record_full_xfer_partial) (record_full_insert_breakpoint, record_full_remove_breakpoint) (record_full_async, record_full_core_xfer_partial): Use target delegation. * target-delegates.c: Rebuild. * target.c (current_xfer_partial): Remove. (update_current_target): Do not INHERIT or de_fault to_insert_breakpoint, to_remove_breakpoint, to_stopped_data_address, to_stopped_by_watchpoint, to_can_async_p, to_is_async_p, to_async. Do not set to_xfer_partial field. (default_xfer_partial): Simplify. (current_xfer_partial): Remove. (target_wait, target_resume): Simplify. (find_default_can_async_p, find_default_is_async_p): Update. (init_dummy_target): Don't set to_can_async_p, to_is_async_p, to_xfer_partial, to_stopped_by_watchpoint, to_stopped_data_address. (target_store_registers): Simplify. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. (target_remove_breakpoint, target_insert_breakpoint) (debug_to_insert_breakpoint, debug_to_remove_breakpoint): Update. * target.h (struct target_ops) <to_resume, to_wait, to_store_registers, to_insert_breakpoint, to_remove_breakpoint, to_stopped_by_watchpoint, to_stopped_data_address, to_can_async_p, to_is_async_p, to_async, to_xfer_partial>: Add TARGET_DEFAULT markup. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. * record-btrace.c (record_btrace_remove_breakpoint): Delegate directly. (record_btrace_insert_breakpoint): Delegate directly.
2013-07-13 03:00:34 +08:00
{
return TARGET_XFER_E_IO;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static enum target_xfer_status
debug_xfer_partial (struct target_ops *self, enum target_object arg1, const char *arg2, gdb_byte *arg3, const gdb_byte *arg4, ULONGEST arg5, ULONGEST arg6, ULONGEST *arg7)
{
enum target_xfer_status result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_xfer_partial (...)\n", debug_target.to_shortname);
result = debug_target.to_xfer_partial (&debug_target, arg1, arg2, arg3, arg4, arg5, arg6, arg7);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_xfer_partial (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_enum_target_object (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_const_char_p (arg2);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_gdb_byte_p (arg3);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_const_gdb_byte_p (arg4);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_ULONGEST (arg5);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_ULONGEST (arg6);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_ULONGEST_p (arg7);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_enum_target_xfer_status (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static VEC(mem_region_s) *
delegate_memory_map (struct target_ops *self)
{
self = self->beneath;
return self->to_memory_map (self);
}
static VEC(mem_region_s) *
tdefault_memory_map (struct target_ops *self)
{
return NULL;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static VEC(mem_region_s) *
debug_memory_map (struct target_ops *self)
{
VEC(mem_region_s) * result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_memory_map (...)\n", debug_target.to_shortname);
result = debug_target.to_memory_map (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_memory_map (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_VEC_mem_region_s__p (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static void
delegate_flash_erase (struct target_ops *self, ULONGEST arg1, LONGEST arg2)
{
self = self->beneath;
self->to_flash_erase (self, arg1, arg2);
}
static void
tdefault_flash_erase (struct target_ops *self, ULONGEST arg1, LONGEST arg2)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_flash_erase (struct target_ops *self, ULONGEST arg1, LONGEST arg2)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_flash_erase (...)\n", debug_target.to_shortname);
debug_target.to_flash_erase (&debug_target, arg1, arg2);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_flash_erase (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_ULONGEST (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_LONGEST (arg2);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
delegate_flash_done (struct target_ops *self)
{
self = self->beneath;
self->to_flash_done (self);
}
static void
tdefault_flash_done (struct target_ops *self)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_flash_done (struct target_ops *self)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_flash_done (...)\n", debug_target.to_shortname);
debug_target.to_flash_done (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_flash_done (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (")\n", gdb_stdlog);
}
static const struct target_desc *
delegate_read_description (struct target_ops *self)
{
self = self->beneath;
return self->to_read_description (self);
}
static const struct target_desc *
tdefault_read_description (struct target_ops *self)
{
return NULL;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static const struct target_desc *
debug_read_description (struct target_ops *self)
{
const struct target_desc * result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_read_description (...)\n", debug_target.to_shortname);
result = debug_target.to_read_description (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_read_description (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_const_struct_target_desc_p (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static ptid_t
delegate_get_ada_task_ptid (struct target_ops *self, long arg1, long arg2)
{
self = self->beneath;
return self->to_get_ada_task_ptid (self, arg1, arg2);
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static ptid_t
debug_get_ada_task_ptid (struct target_ops *self, long arg1, long arg2)
{
ptid_t result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_get_ada_task_ptid (...)\n", debug_target.to_shortname);
result = debug_target.to_get_ada_task_ptid (&debug_target, arg1, arg2);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_get_ada_task_ptid (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_long (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_long (arg2);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_ptid_t (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_auxv_parse (struct target_ops *self, gdb_byte **arg1, gdb_byte *arg2, CORE_ADDR *arg3, CORE_ADDR *arg4)
{
self = self->beneath;
return self->to_auxv_parse (self, arg1, arg2, arg3, arg4);
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_auxv_parse (struct target_ops *self, gdb_byte **arg1, gdb_byte *arg2, CORE_ADDR *arg3, CORE_ADDR *arg4)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_auxv_parse (...)\n", debug_target.to_shortname);
result = debug_target.to_auxv_parse (&debug_target, arg1, arg2, arg3, arg4);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_auxv_parse (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_gdb_byte_pp (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_gdb_byte_p (arg2);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_CORE_ADDR_p (arg3);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_CORE_ADDR_p (arg4);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_search_memory (struct target_ops *self, CORE_ADDR arg1, ULONGEST arg2, const gdb_byte *arg3, ULONGEST arg4, CORE_ADDR *arg5)
{
self = self->beneath;
return self->to_search_memory (self, arg1, arg2, arg3, arg4, arg5);
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_search_memory (struct target_ops *self, CORE_ADDR arg1, ULONGEST arg2, const gdb_byte *arg3, ULONGEST arg4, CORE_ADDR *arg5)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_search_memory (...)\n", debug_target.to_shortname);
result = debug_target.to_search_memory (&debug_target, arg1, arg2, arg3, arg4, arg5);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_search_memory (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_CORE_ADDR (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_ULONGEST (arg2);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_const_gdb_byte_p (arg3);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_ULONGEST (arg4);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_CORE_ADDR_p (arg5);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_can_execute_reverse (struct target_ops *self)
{
self = self->beneath;
return self->to_can_execute_reverse (self);
}
static int
tdefault_can_execute_reverse (struct target_ops *self)
{
return 0;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_can_execute_reverse (struct target_ops *self)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_can_execute_reverse (...)\n", debug_target.to_shortname);
result = debug_target.to_can_execute_reverse (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_can_execute_reverse (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static enum exec_direction_kind
delegate_execution_direction (struct target_ops *self)
{
self = self->beneath;
return self->to_execution_direction (self);
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static enum exec_direction_kind
debug_execution_direction (struct target_ops *self)
{
enum exec_direction_kind result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_execution_direction (...)\n", debug_target.to_shortname);
result = debug_target.to_execution_direction (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_execution_direction (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_enum_exec_direction_kind (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_supports_multi_process (struct target_ops *self)
{
self = self->beneath;
return self->to_supports_multi_process (self);
}
static int
tdefault_supports_multi_process (struct target_ops *self)
{
return 0;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_supports_multi_process (struct target_ops *self)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_supports_multi_process (...)\n", debug_target.to_shortname);
result = debug_target.to_supports_multi_process (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_supports_multi_process (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_supports_enable_disable_tracepoint (struct target_ops *self)
{
self = self->beneath;
return self->to_supports_enable_disable_tracepoint (self);
}
static int
tdefault_supports_enable_disable_tracepoint (struct target_ops *self)
{
return 0;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_supports_enable_disable_tracepoint (struct target_ops *self)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_supports_enable_disable_tracepoint (...)\n", debug_target.to_shortname);
result = debug_target.to_supports_enable_disable_tracepoint (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_supports_enable_disable_tracepoint (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_supports_string_tracing (struct target_ops *self)
{
self = self->beneath;
return self->to_supports_string_tracing (self);
}
static int
tdefault_supports_string_tracing (struct target_ops *self)
{
return 0;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_supports_string_tracing (struct target_ops *self)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_supports_string_tracing (...)\n", debug_target.to_shortname);
result = debug_target.to_supports_string_tracing (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_supports_string_tracing (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_supports_evaluation_of_breakpoint_conditions (struct target_ops *self)
{
self = self->beneath;
return self->to_supports_evaluation_of_breakpoint_conditions (self);
}
static int
tdefault_supports_evaluation_of_breakpoint_conditions (struct target_ops *self)
{
return 0;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_supports_evaluation_of_breakpoint_conditions (struct target_ops *self)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_supports_evaluation_of_breakpoint_conditions (...)\n", debug_target.to_shortname);
result = debug_target.to_supports_evaluation_of_breakpoint_conditions (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_supports_evaluation_of_breakpoint_conditions (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_can_run_breakpoint_commands (struct target_ops *self)
{
self = self->beneath;
return self->to_can_run_breakpoint_commands (self);
}
static int
tdefault_can_run_breakpoint_commands (struct target_ops *self)
{
return 0;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_can_run_breakpoint_commands (struct target_ops *self)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_can_run_breakpoint_commands (...)\n", debug_target.to_shortname);
result = debug_target.to_can_run_breakpoint_commands (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_can_run_breakpoint_commands (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static struct gdbarch *
delegate_thread_architecture (struct target_ops *self, ptid_t arg1)
{
self = self->beneath;
return self->to_thread_architecture (self, arg1);
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static struct gdbarch *
debug_thread_architecture (struct target_ops *self, ptid_t arg1)
{
struct gdbarch * result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_thread_architecture (...)\n", debug_target.to_shortname);
result = debug_target.to_thread_architecture (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_thread_architecture (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_ptid_t (arg1);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_struct_gdbarch_p (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static struct address_space *
delegate_thread_address_space (struct target_ops *self, ptid_t arg1)
{
self = self->beneath;
return self->to_thread_address_space (self, arg1);
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static struct address_space *
debug_thread_address_space (struct target_ops *self, ptid_t arg1)
{
struct address_space * result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_thread_address_space (...)\n", debug_target.to_shortname);
result = debug_target.to_thread_address_space (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_thread_address_space (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_ptid_t (arg1);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_struct_address_space_p (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_filesystem_is_local (struct target_ops *self)
{
self = self->beneath;
return self->to_filesystem_is_local (self);
}
static int
tdefault_filesystem_is_local (struct target_ops *self)
{
return 1;
}
static int
debug_filesystem_is_local (struct target_ops *self)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_filesystem_is_local (...)\n", debug_target.to_shortname);
result = debug_target.to_filesystem_is_local (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_filesystem_is_local (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static void
delegate_trace_init (struct target_ops *self)
{
self = self->beneath;
self->to_trace_init (self);
}
static void
tdefault_trace_init (struct target_ops *self)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_trace_init (struct target_ops *self)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_trace_init (...)\n", debug_target.to_shortname);
debug_target.to_trace_init (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_trace_init (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
delegate_download_tracepoint (struct target_ops *self, struct bp_location *arg1)
{
self = self->beneath;
self->to_download_tracepoint (self, arg1);
}
static void
tdefault_download_tracepoint (struct target_ops *self, struct bp_location *arg1)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_download_tracepoint (struct target_ops *self, struct bp_location *arg1)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_download_tracepoint (...)\n", debug_target.to_shortname);
debug_target.to_download_tracepoint (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_download_tracepoint (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_struct_bp_location_p (arg1);
fputs_unfiltered (")\n", gdb_stdlog);
}
static int
delegate_can_download_tracepoint (struct target_ops *self)
{
self = self->beneath;
return self->to_can_download_tracepoint (self);
}
static int
tdefault_can_download_tracepoint (struct target_ops *self)
{
return 0;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_can_download_tracepoint (struct target_ops *self)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_can_download_tracepoint (...)\n", debug_target.to_shortname);
result = debug_target.to_can_download_tracepoint (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_can_download_tracepoint (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static void
delegate_download_trace_state_variable (struct target_ops *self, struct trace_state_variable *arg1)
{
self = self->beneath;
self->to_download_trace_state_variable (self, arg1);
}
static void
tdefault_download_trace_state_variable (struct target_ops *self, struct trace_state_variable *arg1)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_download_trace_state_variable (struct target_ops *self, struct trace_state_variable *arg1)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_download_trace_state_variable (...)\n", debug_target.to_shortname);
debug_target.to_download_trace_state_variable (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_download_trace_state_variable (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_struct_trace_state_variable_p (arg1);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
delegate_enable_tracepoint (struct target_ops *self, struct bp_location *arg1)
{
self = self->beneath;
self->to_enable_tracepoint (self, arg1);
}
static void
tdefault_enable_tracepoint (struct target_ops *self, struct bp_location *arg1)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_enable_tracepoint (struct target_ops *self, struct bp_location *arg1)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_enable_tracepoint (...)\n", debug_target.to_shortname);
debug_target.to_enable_tracepoint (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_enable_tracepoint (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_struct_bp_location_p (arg1);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
delegate_disable_tracepoint (struct target_ops *self, struct bp_location *arg1)
{
self = self->beneath;
self->to_disable_tracepoint (self, arg1);
}
static void
tdefault_disable_tracepoint (struct target_ops *self, struct bp_location *arg1)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_disable_tracepoint (struct target_ops *self, struct bp_location *arg1)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_disable_tracepoint (...)\n", debug_target.to_shortname);
debug_target.to_disable_tracepoint (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_disable_tracepoint (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_struct_bp_location_p (arg1);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
delegate_trace_set_readonly_regions (struct target_ops *self)
{
self = self->beneath;
self->to_trace_set_readonly_regions (self);
}
static void
tdefault_trace_set_readonly_regions (struct target_ops *self)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_trace_set_readonly_regions (struct target_ops *self)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_trace_set_readonly_regions (...)\n", debug_target.to_shortname);
debug_target.to_trace_set_readonly_regions (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_trace_set_readonly_regions (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
delegate_trace_start (struct target_ops *self)
{
self = self->beneath;
self->to_trace_start (self);
}
static void
tdefault_trace_start (struct target_ops *self)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_trace_start (struct target_ops *self)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_trace_start (...)\n", debug_target.to_shortname);
debug_target.to_trace_start (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_trace_start (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (")\n", gdb_stdlog);
}
static int
delegate_get_trace_status (struct target_ops *self, struct trace_status *arg1)
{
self = self->beneath;
return self->to_get_trace_status (self, arg1);
}
static int
tdefault_get_trace_status (struct target_ops *self, struct trace_status *arg1)
{
return -1;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_get_trace_status (struct target_ops *self, struct trace_status *arg1)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_get_trace_status (...)\n", debug_target.to_shortname);
result = debug_target.to_get_trace_status (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_get_trace_status (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_struct_trace_status_p (arg1);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static void
delegate_get_tracepoint_status (struct target_ops *self, struct breakpoint *arg1, struct uploaded_tp *arg2)
{
self = self->beneath;
self->to_get_tracepoint_status (self, arg1, arg2);
}
static void
tdefault_get_tracepoint_status (struct target_ops *self, struct breakpoint *arg1, struct uploaded_tp *arg2)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_get_tracepoint_status (struct target_ops *self, struct breakpoint *arg1, struct uploaded_tp *arg2)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_get_tracepoint_status (...)\n", debug_target.to_shortname);
debug_target.to_get_tracepoint_status (&debug_target, arg1, arg2);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_get_tracepoint_status (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_struct_breakpoint_p (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_struct_uploaded_tp_p (arg2);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
delegate_trace_stop (struct target_ops *self)
{
self = self->beneath;
self->to_trace_stop (self);
}
static void
tdefault_trace_stop (struct target_ops *self)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_trace_stop (struct target_ops *self)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_trace_stop (...)\n", debug_target.to_shortname);
debug_target.to_trace_stop (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_trace_stop (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (")\n", gdb_stdlog);
}
static int
delegate_trace_find (struct target_ops *self, enum trace_find_type arg1, int arg2, CORE_ADDR arg3, CORE_ADDR arg4, int *arg5)
{
self = self->beneath;
return self->to_trace_find (self, arg1, arg2, arg3, arg4, arg5);
}
static int
tdefault_trace_find (struct target_ops *self, enum trace_find_type arg1, int arg2, CORE_ADDR arg3, CORE_ADDR arg4, int *arg5)
{
return -1;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_trace_find (struct target_ops *self, enum trace_find_type arg1, int arg2, CORE_ADDR arg3, CORE_ADDR arg4, int *arg5)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_trace_find (...)\n", debug_target.to_shortname);
result = debug_target.to_trace_find (&debug_target, arg1, arg2, arg3, arg4, arg5);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_trace_find (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_enum_trace_find_type (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg2);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_CORE_ADDR (arg3);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_CORE_ADDR (arg4);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int_p (arg5);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_get_trace_state_variable_value (struct target_ops *self, int arg1, LONGEST *arg2)
{
self = self->beneath;
return self->to_get_trace_state_variable_value (self, arg1, arg2);
}
static int
tdefault_get_trace_state_variable_value (struct target_ops *self, int arg1, LONGEST *arg2)
{
return 0;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_get_trace_state_variable_value (struct target_ops *self, int arg1, LONGEST *arg2)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_get_trace_state_variable_value (...)\n", debug_target.to_shortname);
result = debug_target.to_get_trace_state_variable_value (&debug_target, arg1, arg2);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_get_trace_state_variable_value (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_LONGEST_p (arg2);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_save_trace_data (struct target_ops *self, const char *arg1)
{
self = self->beneath;
return self->to_save_trace_data (self, arg1);
}
static int
tdefault_save_trace_data (struct target_ops *self, const char *arg1)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_save_trace_data (struct target_ops *self, const char *arg1)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_save_trace_data (...)\n", debug_target.to_shortname);
result = debug_target.to_save_trace_data (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_save_trace_data (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_const_char_p (arg1);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_upload_tracepoints (struct target_ops *self, struct uploaded_tp **arg1)
{
self = self->beneath;
return self->to_upload_tracepoints (self, arg1);
}
static int
tdefault_upload_tracepoints (struct target_ops *self, struct uploaded_tp **arg1)
{
return 0;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_upload_tracepoints (struct target_ops *self, struct uploaded_tp **arg1)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_upload_tracepoints (...)\n", debug_target.to_shortname);
result = debug_target.to_upload_tracepoints (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_upload_tracepoints (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_struct_uploaded_tp_pp (arg1);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_upload_trace_state_variables (struct target_ops *self, struct uploaded_tsv **arg1)
{
self = self->beneath;
return self->to_upload_trace_state_variables (self, arg1);
}
static int
tdefault_upload_trace_state_variables (struct target_ops *self, struct uploaded_tsv **arg1)
{
return 0;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_upload_trace_state_variables (struct target_ops *self, struct uploaded_tsv **arg1)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_upload_trace_state_variables (...)\n", debug_target.to_shortname);
result = debug_target.to_upload_trace_state_variables (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_upload_trace_state_variables (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_struct_uploaded_tsv_pp (arg1);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static LONGEST
delegate_get_raw_trace_data (struct target_ops *self, gdb_byte *arg1, ULONGEST arg2, LONGEST arg3)
{
self = self->beneath;
return self->to_get_raw_trace_data (self, arg1, arg2, arg3);
}
static LONGEST
tdefault_get_raw_trace_data (struct target_ops *self, gdb_byte *arg1, ULONGEST arg2, LONGEST arg3)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static LONGEST
debug_get_raw_trace_data (struct target_ops *self, gdb_byte *arg1, ULONGEST arg2, LONGEST arg3)
{
LONGEST result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_get_raw_trace_data (...)\n", debug_target.to_shortname);
result = debug_target.to_get_raw_trace_data (&debug_target, arg1, arg2, arg3);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_get_raw_trace_data (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_gdb_byte_p (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_ULONGEST (arg2);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_LONGEST (arg3);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_LONGEST (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_get_min_fast_tracepoint_insn_len (struct target_ops *self)
{
self = self->beneath;
return self->to_get_min_fast_tracepoint_insn_len (self);
}
static int
tdefault_get_min_fast_tracepoint_insn_len (struct target_ops *self)
{
return -1;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_get_min_fast_tracepoint_insn_len (struct target_ops *self)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_get_min_fast_tracepoint_insn_len (...)\n", debug_target.to_shortname);
result = debug_target.to_get_min_fast_tracepoint_insn_len (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_get_min_fast_tracepoint_insn_len (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static void
delegate_set_disconnected_tracing (struct target_ops *self, int arg1)
{
self = self->beneath;
self->to_set_disconnected_tracing (self, arg1);
}
static void
tdefault_set_disconnected_tracing (struct target_ops *self, int arg1)
{
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_set_disconnected_tracing (struct target_ops *self, int arg1)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_set_disconnected_tracing (...)\n", debug_target.to_shortname);
debug_target.to_set_disconnected_tracing (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_set_disconnected_tracing (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg1);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
delegate_set_circular_trace_buffer (struct target_ops *self, int arg1)
{
self = self->beneath;
self->to_set_circular_trace_buffer (self, arg1);
}
static void
tdefault_set_circular_trace_buffer (struct target_ops *self, int arg1)
{
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_set_circular_trace_buffer (struct target_ops *self, int arg1)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_set_circular_trace_buffer (...)\n", debug_target.to_shortname);
debug_target.to_set_circular_trace_buffer (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_set_circular_trace_buffer (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg1);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
delegate_set_trace_buffer_size (struct target_ops *self, LONGEST arg1)
{
self = self->beneath;
self->to_set_trace_buffer_size (self, arg1);
}
static void
tdefault_set_trace_buffer_size (struct target_ops *self, LONGEST arg1)
{
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_set_trace_buffer_size (struct target_ops *self, LONGEST arg1)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_set_trace_buffer_size (...)\n", debug_target.to_shortname);
debug_target.to_set_trace_buffer_size (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_set_trace_buffer_size (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_LONGEST (arg1);
fputs_unfiltered (")\n", gdb_stdlog);
}
static int
delegate_set_trace_notes (struct target_ops *self, const char *arg1, const char *arg2, const char *arg3)
{
self = self->beneath;
return self->to_set_trace_notes (self, arg1, arg2, arg3);
}
static int
tdefault_set_trace_notes (struct target_ops *self, const char *arg1, const char *arg2, const char *arg3)
{
return 0;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_set_trace_notes (struct target_ops *self, const char *arg1, const char *arg2, const char *arg3)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_set_trace_notes (...)\n", debug_target.to_shortname);
result = debug_target.to_set_trace_notes (&debug_target, arg1, arg2, arg3);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_set_trace_notes (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_const_char_p (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_const_char_p (arg2);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_const_char_p (arg3);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_core_of_thread (struct target_ops *self, ptid_t arg1)
{
self = self->beneath;
return self->to_core_of_thread (self, arg1);
}
static int
tdefault_core_of_thread (struct target_ops *self, ptid_t arg1)
{
return -1;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_core_of_thread (struct target_ops *self, ptid_t arg1)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_core_of_thread (...)\n", debug_target.to_shortname);
result = debug_target.to_core_of_thread (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_core_of_thread (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_ptid_t (arg1);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_verify_memory (struct target_ops *self, const gdb_byte *arg1, CORE_ADDR arg2, ULONGEST arg3)
{
self = self->beneath;
return self->to_verify_memory (self, arg1, arg2, arg3);
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_verify_memory (struct target_ops *self, const gdb_byte *arg1, CORE_ADDR arg2, ULONGEST arg3)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_verify_memory (...)\n", debug_target.to_shortname);
result = debug_target.to_verify_memory (&debug_target, arg1, arg2, arg3);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_verify_memory (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_const_gdb_byte_p (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_CORE_ADDR (arg2);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_ULONGEST (arg3);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_get_tib_address (struct target_ops *self, ptid_t arg1, CORE_ADDR *arg2)
{
self = self->beneath;
return self->to_get_tib_address (self, arg1, arg2);
}
static int
tdefault_get_tib_address (struct target_ops *self, ptid_t arg1, CORE_ADDR *arg2)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_get_tib_address (struct target_ops *self, ptid_t arg1, CORE_ADDR *arg2)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_get_tib_address (...)\n", debug_target.to_shortname);
result = debug_target.to_get_tib_address (&debug_target, arg1, arg2);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_get_tib_address (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_ptid_t (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_CORE_ADDR_p (arg2);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static void
delegate_set_permissions (struct target_ops *self)
{
self = self->beneath;
self->to_set_permissions (self);
}
static void
tdefault_set_permissions (struct target_ops *self)
{
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_set_permissions (struct target_ops *self)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_set_permissions (...)\n", debug_target.to_shortname);
debug_target.to_set_permissions (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_set_permissions (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (")\n", gdb_stdlog);
}
static int
delegate_static_tracepoint_marker_at (struct target_ops *self, CORE_ADDR arg1, struct static_tracepoint_marker *arg2)
{
self = self->beneath;
return self->to_static_tracepoint_marker_at (self, arg1, arg2);
}
static int
tdefault_static_tracepoint_marker_at (struct target_ops *self, CORE_ADDR arg1, struct static_tracepoint_marker *arg2)
{
return 0;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_static_tracepoint_marker_at (struct target_ops *self, CORE_ADDR arg1, struct static_tracepoint_marker *arg2)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_static_tracepoint_marker_at (...)\n", debug_target.to_shortname);
result = debug_target.to_static_tracepoint_marker_at (&debug_target, arg1, arg2);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_static_tracepoint_marker_at (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_CORE_ADDR (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_struct_static_tracepoint_marker_p (arg2);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static VEC(static_tracepoint_marker_p) *
delegate_static_tracepoint_markers_by_strid (struct target_ops *self, const char *arg1)
{
self = self->beneath;
return self->to_static_tracepoint_markers_by_strid (self, arg1);
}
static VEC(static_tracepoint_marker_p) *
tdefault_static_tracepoint_markers_by_strid (struct target_ops *self, const char *arg1)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static VEC(static_tracepoint_marker_p) *
debug_static_tracepoint_markers_by_strid (struct target_ops *self, const char *arg1)
{
VEC(static_tracepoint_marker_p) * result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_static_tracepoint_markers_by_strid (...)\n", debug_target.to_shortname);
result = debug_target.to_static_tracepoint_markers_by_strid (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_static_tracepoint_markers_by_strid (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_const_char_p (arg1);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_VEC_static_tracepoint_marker_p__p (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static struct traceframe_info *
delegate_traceframe_info (struct target_ops *self)
{
self = self->beneath;
return self->to_traceframe_info (self);
}
static struct traceframe_info *
tdefault_traceframe_info (struct target_ops *self)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static struct traceframe_info *
debug_traceframe_info (struct target_ops *self)
{
struct traceframe_info * result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_traceframe_info (...)\n", debug_target.to_shortname);
result = debug_target.to_traceframe_info (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_traceframe_info (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_struct_traceframe_info_p (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_use_agent (struct target_ops *self, int arg1)
{
self = self->beneath;
return self->to_use_agent (self, arg1);
}
static int
tdefault_use_agent (struct target_ops *self, int arg1)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_use_agent (struct target_ops *self, int arg1)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_use_agent (...)\n", debug_target.to_shortname);
result = debug_target.to_use_agent (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_use_agent (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg1);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
delegate_can_use_agent (struct target_ops *self)
{
self = self->beneath;
return self->to_can_use_agent (self);
}
static int
tdefault_can_use_agent (struct target_ops *self)
{
return 0;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_can_use_agent (struct target_ops *self)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_can_use_agent (...)\n", debug_target.to_shortname);
result = debug_target.to_can_use_agent (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_can_use_agent (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static int
btrace: add format argument to supports_btrace Add a format argument to the various supports_btrace functions to check for support of a specific btrace format. This is to prepare for a new format. Removed two redundant calls. The check will be made in the subsequent btrace_enable call. 2015-02-09 Markus Metzger <markus.t.metzger@intel.com> * btrace.c (btrace_enable): Pass BTRACE_FORMAT_BTS. * record-btrace.c (record_btrace_open): Remove call to target_supports_btrace. * remote.c (remote_supports_btrace): Update parameters. * target.c (target_supports_btrace): Update parameters. * target.h (to_supports_btrace, target_supports_btrace): Update parameters. * target-delegates.c: Regenerate. * target-debug.h (target_debug_print_enum_btrace_format): New. * nat/linux-btrace.c (kernel_supports_btrace): Rename into ... (kernel_supports_bts): ... this. Update users. Update warning text. (intel_supports_btrace): Rename into ... (intel_supports_bts): ... this. Update users. (cpu_supports_btrace): Rename into ... (cpu_supports_bts): ... this. Update users. (linux_supports_btrace): Update parameters. Split into this and ... (linux_supports_bts): ... this. * nat/linux-btrace.h (linux_supports_btrace): Update parameters. gdbserver/ * server.c (handle_btrace_general_set): Remove call to target_supports_btrace. (supported_btrace_packets): New. (handle_query): Call supported_btrace_packets. * target.h: include btrace-common.h. (btrace_target_info): Removed. (supports_btrace, target_supports_btrace): Update parameters.
2014-01-17 20:29:19 +08:00
delegate_supports_btrace (struct target_ops *self, enum btrace_format arg1)
{
self = self->beneath;
btrace: add format argument to supports_btrace Add a format argument to the various supports_btrace functions to check for support of a specific btrace format. This is to prepare for a new format. Removed two redundant calls. The check will be made in the subsequent btrace_enable call. 2015-02-09 Markus Metzger <markus.t.metzger@intel.com> * btrace.c (btrace_enable): Pass BTRACE_FORMAT_BTS. * record-btrace.c (record_btrace_open): Remove call to target_supports_btrace. * remote.c (remote_supports_btrace): Update parameters. * target.c (target_supports_btrace): Update parameters. * target.h (to_supports_btrace, target_supports_btrace): Update parameters. * target-delegates.c: Regenerate. * target-debug.h (target_debug_print_enum_btrace_format): New. * nat/linux-btrace.c (kernel_supports_btrace): Rename into ... (kernel_supports_bts): ... this. Update users. Update warning text. (intel_supports_btrace): Rename into ... (intel_supports_bts): ... this. Update users. (cpu_supports_btrace): Rename into ... (cpu_supports_bts): ... this. Update users. (linux_supports_btrace): Update parameters. Split into this and ... (linux_supports_bts): ... this. * nat/linux-btrace.h (linux_supports_btrace): Update parameters. gdbserver/ * server.c (handle_btrace_general_set): Remove call to target_supports_btrace. (supported_btrace_packets): New. (handle_query): Call supported_btrace_packets. * target.h: include btrace-common.h. (btrace_target_info): Removed. (supports_btrace, target_supports_btrace): Update parameters.
2014-01-17 20:29:19 +08:00
return self->to_supports_btrace (self, arg1);
}
static int
btrace: add format argument to supports_btrace Add a format argument to the various supports_btrace functions to check for support of a specific btrace format. This is to prepare for a new format. Removed two redundant calls. The check will be made in the subsequent btrace_enable call. 2015-02-09 Markus Metzger <markus.t.metzger@intel.com> * btrace.c (btrace_enable): Pass BTRACE_FORMAT_BTS. * record-btrace.c (record_btrace_open): Remove call to target_supports_btrace. * remote.c (remote_supports_btrace): Update parameters. * target.c (target_supports_btrace): Update parameters. * target.h (to_supports_btrace, target_supports_btrace): Update parameters. * target-delegates.c: Regenerate. * target-debug.h (target_debug_print_enum_btrace_format): New. * nat/linux-btrace.c (kernel_supports_btrace): Rename into ... (kernel_supports_bts): ... this. Update users. Update warning text. (intel_supports_btrace): Rename into ... (intel_supports_bts): ... this. Update users. (cpu_supports_btrace): Rename into ... (cpu_supports_bts): ... this. Update users. (linux_supports_btrace): Update parameters. Split into this and ... (linux_supports_bts): ... this. * nat/linux-btrace.h (linux_supports_btrace): Update parameters. gdbserver/ * server.c (handle_btrace_general_set): Remove call to target_supports_btrace. (supported_btrace_packets): New. (handle_query): Call supported_btrace_packets. * target.h: include btrace-common.h. (btrace_target_info): Removed. (supports_btrace, target_supports_btrace): Update parameters.
2014-01-17 20:29:19 +08:00
tdefault_supports_btrace (struct target_ops *self, enum btrace_format arg1)
{
return 0;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
btrace: add format argument to supports_btrace Add a format argument to the various supports_btrace functions to check for support of a specific btrace format. This is to prepare for a new format. Removed two redundant calls. The check will be made in the subsequent btrace_enable call. 2015-02-09 Markus Metzger <markus.t.metzger@intel.com> * btrace.c (btrace_enable): Pass BTRACE_FORMAT_BTS. * record-btrace.c (record_btrace_open): Remove call to target_supports_btrace. * remote.c (remote_supports_btrace): Update parameters. * target.c (target_supports_btrace): Update parameters. * target.h (to_supports_btrace, target_supports_btrace): Update parameters. * target-delegates.c: Regenerate. * target-debug.h (target_debug_print_enum_btrace_format): New. * nat/linux-btrace.c (kernel_supports_btrace): Rename into ... (kernel_supports_bts): ... this. Update users. Update warning text. (intel_supports_btrace): Rename into ... (intel_supports_bts): ... this. Update users. (cpu_supports_btrace): Rename into ... (cpu_supports_bts): ... this. Update users. (linux_supports_btrace): Update parameters. Split into this and ... (linux_supports_bts): ... this. * nat/linux-btrace.h (linux_supports_btrace): Update parameters. gdbserver/ * server.c (handle_btrace_general_set): Remove call to target_supports_btrace. (supported_btrace_packets): New. (handle_query): Call supported_btrace_packets. * target.h: include btrace-common.h. (btrace_target_info): Removed. (supports_btrace, target_supports_btrace): Update parameters.
2014-01-17 20:29:19 +08:00
debug_supports_btrace (struct target_ops *self, enum btrace_format arg1)
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_supports_btrace (...)\n", debug_target.to_shortname);
btrace: add format argument to supports_btrace Add a format argument to the various supports_btrace functions to check for support of a specific btrace format. This is to prepare for a new format. Removed two redundant calls. The check will be made in the subsequent btrace_enable call. 2015-02-09 Markus Metzger <markus.t.metzger@intel.com> * btrace.c (btrace_enable): Pass BTRACE_FORMAT_BTS. * record-btrace.c (record_btrace_open): Remove call to target_supports_btrace. * remote.c (remote_supports_btrace): Update parameters. * target.c (target_supports_btrace): Update parameters. * target.h (to_supports_btrace, target_supports_btrace): Update parameters. * target-delegates.c: Regenerate. * target-debug.h (target_debug_print_enum_btrace_format): New. * nat/linux-btrace.c (kernel_supports_btrace): Rename into ... (kernel_supports_bts): ... this. Update users. Update warning text. (intel_supports_btrace): Rename into ... (intel_supports_bts): ... this. Update users. (cpu_supports_btrace): Rename into ... (cpu_supports_bts): ... this. Update users. (linux_supports_btrace): Update parameters. Split into this and ... (linux_supports_bts): ... this. * nat/linux-btrace.h (linux_supports_btrace): Update parameters. gdbserver/ * server.c (handle_btrace_general_set): Remove call to target_supports_btrace. (supported_btrace_packets): New. (handle_query): Call supported_btrace_packets. * target.h: include btrace-common.h. (btrace_target_info): Removed. (supports_btrace, target_supports_btrace): Update parameters.
2014-01-17 20:29:19 +08:00
result = debug_target.to_supports_btrace (&debug_target, arg1);
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
fprintf_unfiltered (gdb_stdlog, "<- %s->to_supports_btrace (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
btrace: add format argument to supports_btrace Add a format argument to the various supports_btrace functions to check for support of a specific btrace format. This is to prepare for a new format. Removed two redundant calls. The check will be made in the subsequent btrace_enable call. 2015-02-09 Markus Metzger <markus.t.metzger@intel.com> * btrace.c (btrace_enable): Pass BTRACE_FORMAT_BTS. * record-btrace.c (record_btrace_open): Remove call to target_supports_btrace. * remote.c (remote_supports_btrace): Update parameters. * target.c (target_supports_btrace): Update parameters. * target.h (to_supports_btrace, target_supports_btrace): Update parameters. * target-delegates.c: Regenerate. * target-debug.h (target_debug_print_enum_btrace_format): New. * nat/linux-btrace.c (kernel_supports_btrace): Rename into ... (kernel_supports_bts): ... this. Update users. Update warning text. (intel_supports_btrace): Rename into ... (intel_supports_bts): ... this. Update users. (cpu_supports_btrace): Rename into ... (cpu_supports_bts): ... this. Update users. (linux_supports_btrace): Update parameters. Split into this and ... (linux_supports_bts): ... this. * nat/linux-btrace.h (linux_supports_btrace): Update parameters. gdbserver/ * server.c (handle_btrace_general_set): Remove call to target_supports_btrace. (supported_btrace_packets): New. (handle_query): Call supported_btrace_packets. * target.h: include btrace-common.h. (btrace_target_info): Removed. (supports_btrace, target_supports_btrace): Update parameters.
2014-01-17 20:29:19 +08:00
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_enum_btrace_format (arg1);
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static struct btrace_target_info *
record btrace: add configuration struct Add a struct to describe the branch trace configuration and use it for enabling branch tracing. The user will be able to set configuration fields for each tracing format to be used for new threads. The actual configuration that is active for a given thread will be shown in the "info record" command. At the moment, the configuration struct only contains a format field that is set to the only available format. The format is the only configuration option that can not be set via set commands. It is given as argument to the "record btrace" command when starting recording. 2015-02-09 Markus Metzger <markus.t.metzger@intel.com> * Makefile.in (XMLFILES): Add btrace-conf.dtd. * x86-linux-nat.c (x86_linux_enable_btrace): Update parameters. (x86_linux_btrace_conf): New. (x86_linux_create_target): Initialize to_btrace_conf. * nat/linux-btrace.c (linux_enable_btrace): Update parameters. Check format. Split into this and ... (linux_enable_bts): ... this. (linux_btrace_conf): New. (perf_event_skip_record): Renamed into ... (perf_event_skip_bts_record): ... this. Updated users. (linux_disable_btrace): Split into this and ... (linux_disable_bts): ... this. (linux_read_btrace): Check format. * nat/linux-btrace.h (linux_enable_btrace): Update parameters. (linux_btrace_conf): New. (btrace_target_info)<ptid>: Moved. (btrace_target_info)<conf>: New. (btrace_target_info): Split into this and ... (btrace_tinfo_bts): ... this. Updated users. * btrace.c (btrace_enable): Update parameters. (btrace_conf, parse_xml_btrace_conf_bts, parse_xml_btrace_conf) (btrace_conf_children, btrace_conf_attributes) (btrace_conf_elements): New. * btrace.h (btrace_enable): Update parameters. (btrace_conf, parse_xml_btrace_conf): New. * common/btrace-common.h (btrace_config): New. * feature/btrace-conf.dtd: New. * record-btrace.c (record_btrace_conf): New. (record_btrace_cmdlist): New. (record_btrace_enable_warn, record_btrace_open): Pass &record_btrace_conf. (record_btrace_info): Print recording format. (cmd_record_btrace_bts_start): New. (cmd_record_btrace_start): Call cmd_record_btrace_bts_start. (_initialize_record_btrace): Add "record btrace bts" subcommand. Add "record bts" alias command. * remote.c (remote_state)<btrace_config>: New. (remote_btrace_reset, PACKET_qXfer_btrace_conf): New. (remote_protocol_features): Add qXfer:btrace-conf:read. (remote_open_1): Call remote_btrace_reset. (remote_xfer_partial): Handle TARGET_OBJECT_BTRACE_CONF. (btrace_target_info)<conf>: New. (btrace_sync_conf, btrace_read_config): New. (remote_enable_btrace): Update parameters. Call btrace_sync_conf and btrace_read_conf. (remote_btrace_conf): New. (init_remote_ops): Initialize to_btrace_conf. (_initialize_remote): Add qXfer:btrace-conf packet. * target.c (target_enable_btrace): Update parameters. (target_btrace_conf): New. * target.h (target_enable_btrace): Update parameters. (target_btrace_conf): New. (target_object)<TARGET_OBJECT_BTRACE_CONF>: New. (target_ops)<to_enable_btrace>: Update parameters and comment. (target_ops)<to_btrace_conf>: New. * target-delegates: Regenerate. * target-debug.h (target_debug_print_const_struct_btrace_config_p) (target_debug_print_const_struct_btrace_target_info_p): New. NEWS: Announce new command and new packet. doc/ * gdb.texinfo (Process Record and Replay): Describe the "record btrace bts" command. (General Query Packets): Describe qXfer:btrace-conf:read packet. (Branch Trace Configuration Format): New. gdbserver/ * linux-low.c (linux_low_enable_btrace): Update parameters. (linux_low_btrace_conf): New. (linux_target_ops)<to_btrace_conf>: Initialize. * server.c (current_btrace_conf): New. (handle_btrace_enable): Rename to ... (handle_btrace_enable_bts): ... this. Pass &current_btrace_conf to target_enable_btrace. Update comment. Update users. (handle_qxfer_btrace_conf): New. (qxfer_packets): Add btrace-conf entry. (handle_query): Report qXfer:btrace-conf:read as supported packet. * target.h (target_ops)<enable_btrace>: Update parameters and comment. (target_ops)<read_btrace_conf>: New. (target_enable_btrace): Update parameters. (target_read_btrace_conf): New. testsuite/ * gdb.btrace/delta.exp: Update "info record" output. * gdb.btrace/enable.exp: Update "info record" output. * gdb.btrace/finish.exp: Update "info record" output. * gdb.btrace/instruction_history.exp: Update "info record" output. * gdb.btrace/next.exp: Update "info record" output. * gdb.btrace/nexti.exp: Update "info record" output. * gdb.btrace/step.exp: Update "info record" output. * gdb.btrace/stepi.exp: Update "info record" output. * gdb.btrace/nohist.exp: Update "info record" output.
2013-11-28 22:44:13 +08:00
delegate_enable_btrace (struct target_ops *self, ptid_t arg1, const struct btrace_config *arg2)
{
self = self->beneath;
record btrace: add configuration struct Add a struct to describe the branch trace configuration and use it for enabling branch tracing. The user will be able to set configuration fields for each tracing format to be used for new threads. The actual configuration that is active for a given thread will be shown in the "info record" command. At the moment, the configuration struct only contains a format field that is set to the only available format. The format is the only configuration option that can not be set via set commands. It is given as argument to the "record btrace" command when starting recording. 2015-02-09 Markus Metzger <markus.t.metzger@intel.com> * Makefile.in (XMLFILES): Add btrace-conf.dtd. * x86-linux-nat.c (x86_linux_enable_btrace): Update parameters. (x86_linux_btrace_conf): New. (x86_linux_create_target): Initialize to_btrace_conf. * nat/linux-btrace.c (linux_enable_btrace): Update parameters. Check format. Split into this and ... (linux_enable_bts): ... this. (linux_btrace_conf): New. (perf_event_skip_record): Renamed into ... (perf_event_skip_bts_record): ... this. Updated users. (linux_disable_btrace): Split into this and ... (linux_disable_bts): ... this. (linux_read_btrace): Check format. * nat/linux-btrace.h (linux_enable_btrace): Update parameters. (linux_btrace_conf): New. (btrace_target_info)<ptid>: Moved. (btrace_target_info)<conf>: New. (btrace_target_info): Split into this and ... (btrace_tinfo_bts): ... this. Updated users. * btrace.c (btrace_enable): Update parameters. (btrace_conf, parse_xml_btrace_conf_bts, parse_xml_btrace_conf) (btrace_conf_children, btrace_conf_attributes) (btrace_conf_elements): New. * btrace.h (btrace_enable): Update parameters. (btrace_conf, parse_xml_btrace_conf): New. * common/btrace-common.h (btrace_config): New. * feature/btrace-conf.dtd: New. * record-btrace.c (record_btrace_conf): New. (record_btrace_cmdlist): New. (record_btrace_enable_warn, record_btrace_open): Pass &record_btrace_conf. (record_btrace_info): Print recording format. (cmd_record_btrace_bts_start): New. (cmd_record_btrace_start): Call cmd_record_btrace_bts_start. (_initialize_record_btrace): Add "record btrace bts" subcommand. Add "record bts" alias command. * remote.c (remote_state)<btrace_config>: New. (remote_btrace_reset, PACKET_qXfer_btrace_conf): New. (remote_protocol_features): Add qXfer:btrace-conf:read. (remote_open_1): Call remote_btrace_reset. (remote_xfer_partial): Handle TARGET_OBJECT_BTRACE_CONF. (btrace_target_info)<conf>: New. (btrace_sync_conf, btrace_read_config): New. (remote_enable_btrace): Update parameters. Call btrace_sync_conf and btrace_read_conf. (remote_btrace_conf): New. (init_remote_ops): Initialize to_btrace_conf. (_initialize_remote): Add qXfer:btrace-conf packet. * target.c (target_enable_btrace): Update parameters. (target_btrace_conf): New. * target.h (target_enable_btrace): Update parameters. (target_btrace_conf): New. (target_object)<TARGET_OBJECT_BTRACE_CONF>: New. (target_ops)<to_enable_btrace>: Update parameters and comment. (target_ops)<to_btrace_conf>: New. * target-delegates: Regenerate. * target-debug.h (target_debug_print_const_struct_btrace_config_p) (target_debug_print_const_struct_btrace_target_info_p): New. NEWS: Announce new command and new packet. doc/ * gdb.texinfo (Process Record and Replay): Describe the "record btrace bts" command. (General Query Packets): Describe qXfer:btrace-conf:read packet. (Branch Trace Configuration Format): New. gdbserver/ * linux-low.c (linux_low_enable_btrace): Update parameters. (linux_low_btrace_conf): New. (linux_target_ops)<to_btrace_conf>: Initialize. * server.c (current_btrace_conf): New. (handle_btrace_enable): Rename to ... (handle_btrace_enable_bts): ... this. Pass &current_btrace_conf to target_enable_btrace. Update comment. Update users. (handle_qxfer_btrace_conf): New. (qxfer_packets): Add btrace-conf entry. (handle_query): Report qXfer:btrace-conf:read as supported packet. * target.h (target_ops)<enable_btrace>: Update parameters and comment. (target_ops)<read_btrace_conf>: New. (target_enable_btrace): Update parameters. (target_read_btrace_conf): New. testsuite/ * gdb.btrace/delta.exp: Update "info record" output. * gdb.btrace/enable.exp: Update "info record" output. * gdb.btrace/finish.exp: Update "info record" output. * gdb.btrace/instruction_history.exp: Update "info record" output. * gdb.btrace/next.exp: Update "info record" output. * gdb.btrace/nexti.exp: Update "info record" output. * gdb.btrace/step.exp: Update "info record" output. * gdb.btrace/stepi.exp: Update "info record" output. * gdb.btrace/nohist.exp: Update "info record" output.
2013-11-28 22:44:13 +08:00
return self->to_enable_btrace (self, arg1, arg2);
}
static struct btrace_target_info *
record btrace: add configuration struct Add a struct to describe the branch trace configuration and use it for enabling branch tracing. The user will be able to set configuration fields for each tracing format to be used for new threads. The actual configuration that is active for a given thread will be shown in the "info record" command. At the moment, the configuration struct only contains a format field that is set to the only available format. The format is the only configuration option that can not be set via set commands. It is given as argument to the "record btrace" command when starting recording. 2015-02-09 Markus Metzger <markus.t.metzger@intel.com> * Makefile.in (XMLFILES): Add btrace-conf.dtd. * x86-linux-nat.c (x86_linux_enable_btrace): Update parameters. (x86_linux_btrace_conf): New. (x86_linux_create_target): Initialize to_btrace_conf. * nat/linux-btrace.c (linux_enable_btrace): Update parameters. Check format. Split into this and ... (linux_enable_bts): ... this. (linux_btrace_conf): New. (perf_event_skip_record): Renamed into ... (perf_event_skip_bts_record): ... this. Updated users. (linux_disable_btrace): Split into this and ... (linux_disable_bts): ... this. (linux_read_btrace): Check format. * nat/linux-btrace.h (linux_enable_btrace): Update parameters. (linux_btrace_conf): New. (btrace_target_info)<ptid>: Moved. (btrace_target_info)<conf>: New. (btrace_target_info): Split into this and ... (btrace_tinfo_bts): ... this. Updated users. * btrace.c (btrace_enable): Update parameters. (btrace_conf, parse_xml_btrace_conf_bts, parse_xml_btrace_conf) (btrace_conf_children, btrace_conf_attributes) (btrace_conf_elements): New. * btrace.h (btrace_enable): Update parameters. (btrace_conf, parse_xml_btrace_conf): New. * common/btrace-common.h (btrace_config): New. * feature/btrace-conf.dtd: New. * record-btrace.c (record_btrace_conf): New. (record_btrace_cmdlist): New. (record_btrace_enable_warn, record_btrace_open): Pass &record_btrace_conf. (record_btrace_info): Print recording format. (cmd_record_btrace_bts_start): New. (cmd_record_btrace_start): Call cmd_record_btrace_bts_start. (_initialize_record_btrace): Add "record btrace bts" subcommand. Add "record bts" alias command. * remote.c (remote_state)<btrace_config>: New. (remote_btrace_reset, PACKET_qXfer_btrace_conf): New. (remote_protocol_features): Add qXfer:btrace-conf:read. (remote_open_1): Call remote_btrace_reset. (remote_xfer_partial): Handle TARGET_OBJECT_BTRACE_CONF. (btrace_target_info)<conf>: New. (btrace_sync_conf, btrace_read_config): New. (remote_enable_btrace): Update parameters. Call btrace_sync_conf and btrace_read_conf. (remote_btrace_conf): New. (init_remote_ops): Initialize to_btrace_conf. (_initialize_remote): Add qXfer:btrace-conf packet. * target.c (target_enable_btrace): Update parameters. (target_btrace_conf): New. * target.h (target_enable_btrace): Update parameters. (target_btrace_conf): New. (target_object)<TARGET_OBJECT_BTRACE_CONF>: New. (target_ops)<to_enable_btrace>: Update parameters and comment. (target_ops)<to_btrace_conf>: New. * target-delegates: Regenerate. * target-debug.h (target_debug_print_const_struct_btrace_config_p) (target_debug_print_const_struct_btrace_target_info_p): New. NEWS: Announce new command and new packet. doc/ * gdb.texinfo (Process Record and Replay): Describe the "record btrace bts" command. (General Query Packets): Describe qXfer:btrace-conf:read packet. (Branch Trace Configuration Format): New. gdbserver/ * linux-low.c (linux_low_enable_btrace): Update parameters. (linux_low_btrace_conf): New. (linux_target_ops)<to_btrace_conf>: Initialize. * server.c (current_btrace_conf): New. (handle_btrace_enable): Rename to ... (handle_btrace_enable_bts): ... this. Pass &current_btrace_conf to target_enable_btrace. Update comment. Update users. (handle_qxfer_btrace_conf): New. (qxfer_packets): Add btrace-conf entry. (handle_query): Report qXfer:btrace-conf:read as supported packet. * target.h (target_ops)<enable_btrace>: Update parameters and comment. (target_ops)<read_btrace_conf>: New. (target_enable_btrace): Update parameters. (target_read_btrace_conf): New. testsuite/ * gdb.btrace/delta.exp: Update "info record" output. * gdb.btrace/enable.exp: Update "info record" output. * gdb.btrace/finish.exp: Update "info record" output. * gdb.btrace/instruction_history.exp: Update "info record" output. * gdb.btrace/next.exp: Update "info record" output. * gdb.btrace/nexti.exp: Update "info record" output. * gdb.btrace/step.exp: Update "info record" output. * gdb.btrace/stepi.exp: Update "info record" output. * gdb.btrace/nohist.exp: Update "info record" output.
2013-11-28 22:44:13 +08:00
tdefault_enable_btrace (struct target_ops *self, ptid_t arg1, const struct btrace_config *arg2)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static struct btrace_target_info *
record btrace: add configuration struct Add a struct to describe the branch trace configuration and use it for enabling branch tracing. The user will be able to set configuration fields for each tracing format to be used for new threads. The actual configuration that is active for a given thread will be shown in the "info record" command. At the moment, the configuration struct only contains a format field that is set to the only available format. The format is the only configuration option that can not be set via set commands. It is given as argument to the "record btrace" command when starting recording. 2015-02-09 Markus Metzger <markus.t.metzger@intel.com> * Makefile.in (XMLFILES): Add btrace-conf.dtd. * x86-linux-nat.c (x86_linux_enable_btrace): Update parameters. (x86_linux_btrace_conf): New. (x86_linux_create_target): Initialize to_btrace_conf. * nat/linux-btrace.c (linux_enable_btrace): Update parameters. Check format. Split into this and ... (linux_enable_bts): ... this. (linux_btrace_conf): New. (perf_event_skip_record): Renamed into ... (perf_event_skip_bts_record): ... this. Updated users. (linux_disable_btrace): Split into this and ... (linux_disable_bts): ... this. (linux_read_btrace): Check format. * nat/linux-btrace.h (linux_enable_btrace): Update parameters. (linux_btrace_conf): New. (btrace_target_info)<ptid>: Moved. (btrace_target_info)<conf>: New. (btrace_target_info): Split into this and ... (btrace_tinfo_bts): ... this. Updated users. * btrace.c (btrace_enable): Update parameters. (btrace_conf, parse_xml_btrace_conf_bts, parse_xml_btrace_conf) (btrace_conf_children, btrace_conf_attributes) (btrace_conf_elements): New. * btrace.h (btrace_enable): Update parameters. (btrace_conf, parse_xml_btrace_conf): New. * common/btrace-common.h (btrace_config): New. * feature/btrace-conf.dtd: New. * record-btrace.c (record_btrace_conf): New. (record_btrace_cmdlist): New. (record_btrace_enable_warn, record_btrace_open): Pass &record_btrace_conf. (record_btrace_info): Print recording format. (cmd_record_btrace_bts_start): New. (cmd_record_btrace_start): Call cmd_record_btrace_bts_start. (_initialize_record_btrace): Add "record btrace bts" subcommand. Add "record bts" alias command. * remote.c (remote_state)<btrace_config>: New. (remote_btrace_reset, PACKET_qXfer_btrace_conf): New. (remote_protocol_features): Add qXfer:btrace-conf:read. (remote_open_1): Call remote_btrace_reset. (remote_xfer_partial): Handle TARGET_OBJECT_BTRACE_CONF. (btrace_target_info)<conf>: New. (btrace_sync_conf, btrace_read_config): New. (remote_enable_btrace): Update parameters. Call btrace_sync_conf and btrace_read_conf. (remote_btrace_conf): New. (init_remote_ops): Initialize to_btrace_conf. (_initialize_remote): Add qXfer:btrace-conf packet. * target.c (target_enable_btrace): Update parameters. (target_btrace_conf): New. * target.h (target_enable_btrace): Update parameters. (target_btrace_conf): New. (target_object)<TARGET_OBJECT_BTRACE_CONF>: New. (target_ops)<to_enable_btrace>: Update parameters and comment. (target_ops)<to_btrace_conf>: New. * target-delegates: Regenerate. * target-debug.h (target_debug_print_const_struct_btrace_config_p) (target_debug_print_const_struct_btrace_target_info_p): New. NEWS: Announce new command and new packet. doc/ * gdb.texinfo (Process Record and Replay): Describe the "record btrace bts" command. (General Query Packets): Describe qXfer:btrace-conf:read packet. (Branch Trace Configuration Format): New. gdbserver/ * linux-low.c (linux_low_enable_btrace): Update parameters. (linux_low_btrace_conf): New. (linux_target_ops)<to_btrace_conf>: Initialize. * server.c (current_btrace_conf): New. (handle_btrace_enable): Rename to ... (handle_btrace_enable_bts): ... this. Pass &current_btrace_conf to target_enable_btrace. Update comment. Update users. (handle_qxfer_btrace_conf): New. (qxfer_packets): Add btrace-conf entry. (handle_query): Report qXfer:btrace-conf:read as supported packet. * target.h (target_ops)<enable_btrace>: Update parameters and comment. (target_ops)<read_btrace_conf>: New. (target_enable_btrace): Update parameters. (target_read_btrace_conf): New. testsuite/ * gdb.btrace/delta.exp: Update "info record" output. * gdb.btrace/enable.exp: Update "info record" output. * gdb.btrace/finish.exp: Update "info record" output. * gdb.btrace/instruction_history.exp: Update "info record" output. * gdb.btrace/next.exp: Update "info record" output. * gdb.btrace/nexti.exp: Update "info record" output. * gdb.btrace/step.exp: Update "info record" output. * gdb.btrace/stepi.exp: Update "info record" output. * gdb.btrace/nohist.exp: Update "info record" output.
2013-11-28 22:44:13 +08:00
debug_enable_btrace (struct target_ops *self, ptid_t arg1, const struct btrace_config *arg2)
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
{
struct btrace_target_info * result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_enable_btrace (...)\n", debug_target.to_shortname);
record btrace: add configuration struct Add a struct to describe the branch trace configuration and use it for enabling branch tracing. The user will be able to set configuration fields for each tracing format to be used for new threads. The actual configuration that is active for a given thread will be shown in the "info record" command. At the moment, the configuration struct only contains a format field that is set to the only available format. The format is the only configuration option that can not be set via set commands. It is given as argument to the "record btrace" command when starting recording. 2015-02-09 Markus Metzger <markus.t.metzger@intel.com> * Makefile.in (XMLFILES): Add btrace-conf.dtd. * x86-linux-nat.c (x86_linux_enable_btrace): Update parameters. (x86_linux_btrace_conf): New. (x86_linux_create_target): Initialize to_btrace_conf. * nat/linux-btrace.c (linux_enable_btrace): Update parameters. Check format. Split into this and ... (linux_enable_bts): ... this. (linux_btrace_conf): New. (perf_event_skip_record): Renamed into ... (perf_event_skip_bts_record): ... this. Updated users. (linux_disable_btrace): Split into this and ... (linux_disable_bts): ... this. (linux_read_btrace): Check format. * nat/linux-btrace.h (linux_enable_btrace): Update parameters. (linux_btrace_conf): New. (btrace_target_info)<ptid>: Moved. (btrace_target_info)<conf>: New. (btrace_target_info): Split into this and ... (btrace_tinfo_bts): ... this. Updated users. * btrace.c (btrace_enable): Update parameters. (btrace_conf, parse_xml_btrace_conf_bts, parse_xml_btrace_conf) (btrace_conf_children, btrace_conf_attributes) (btrace_conf_elements): New. * btrace.h (btrace_enable): Update parameters. (btrace_conf, parse_xml_btrace_conf): New. * common/btrace-common.h (btrace_config): New. * feature/btrace-conf.dtd: New. * record-btrace.c (record_btrace_conf): New. (record_btrace_cmdlist): New. (record_btrace_enable_warn, record_btrace_open): Pass &record_btrace_conf. (record_btrace_info): Print recording format. (cmd_record_btrace_bts_start): New. (cmd_record_btrace_start): Call cmd_record_btrace_bts_start. (_initialize_record_btrace): Add "record btrace bts" subcommand. Add "record bts" alias command. * remote.c (remote_state)<btrace_config>: New. (remote_btrace_reset, PACKET_qXfer_btrace_conf): New. (remote_protocol_features): Add qXfer:btrace-conf:read. (remote_open_1): Call remote_btrace_reset. (remote_xfer_partial): Handle TARGET_OBJECT_BTRACE_CONF. (btrace_target_info)<conf>: New. (btrace_sync_conf, btrace_read_config): New. (remote_enable_btrace): Update parameters. Call btrace_sync_conf and btrace_read_conf. (remote_btrace_conf): New. (init_remote_ops): Initialize to_btrace_conf. (_initialize_remote): Add qXfer:btrace-conf packet. * target.c (target_enable_btrace): Update parameters. (target_btrace_conf): New. * target.h (target_enable_btrace): Update parameters. (target_btrace_conf): New. (target_object)<TARGET_OBJECT_BTRACE_CONF>: New. (target_ops)<to_enable_btrace>: Update parameters and comment. (target_ops)<to_btrace_conf>: New. * target-delegates: Regenerate. * target-debug.h (target_debug_print_const_struct_btrace_config_p) (target_debug_print_const_struct_btrace_target_info_p): New. NEWS: Announce new command and new packet. doc/ * gdb.texinfo (Process Record and Replay): Describe the "record btrace bts" command. (General Query Packets): Describe qXfer:btrace-conf:read packet. (Branch Trace Configuration Format): New. gdbserver/ * linux-low.c (linux_low_enable_btrace): Update parameters. (linux_low_btrace_conf): New. (linux_target_ops)<to_btrace_conf>: Initialize. * server.c (current_btrace_conf): New. (handle_btrace_enable): Rename to ... (handle_btrace_enable_bts): ... this. Pass &current_btrace_conf to target_enable_btrace. Update comment. Update users. (handle_qxfer_btrace_conf): New. (qxfer_packets): Add btrace-conf entry. (handle_query): Report qXfer:btrace-conf:read as supported packet. * target.h (target_ops)<enable_btrace>: Update parameters and comment. (target_ops)<read_btrace_conf>: New. (target_enable_btrace): Update parameters. (target_read_btrace_conf): New. testsuite/ * gdb.btrace/delta.exp: Update "info record" output. * gdb.btrace/enable.exp: Update "info record" output. * gdb.btrace/finish.exp: Update "info record" output. * gdb.btrace/instruction_history.exp: Update "info record" output. * gdb.btrace/next.exp: Update "info record" output. * gdb.btrace/nexti.exp: Update "info record" output. * gdb.btrace/step.exp: Update "info record" output. * gdb.btrace/stepi.exp: Update "info record" output. * gdb.btrace/nohist.exp: Update "info record" output.
2013-11-28 22:44:13 +08:00
result = debug_target.to_enable_btrace (&debug_target, arg1, arg2);
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
fprintf_unfiltered (gdb_stdlog, "<- %s->to_enable_btrace (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_ptid_t (arg1);
record btrace: add configuration struct Add a struct to describe the branch trace configuration and use it for enabling branch tracing. The user will be able to set configuration fields for each tracing format to be used for new threads. The actual configuration that is active for a given thread will be shown in the "info record" command. At the moment, the configuration struct only contains a format field that is set to the only available format. The format is the only configuration option that can not be set via set commands. It is given as argument to the "record btrace" command when starting recording. 2015-02-09 Markus Metzger <markus.t.metzger@intel.com> * Makefile.in (XMLFILES): Add btrace-conf.dtd. * x86-linux-nat.c (x86_linux_enable_btrace): Update parameters. (x86_linux_btrace_conf): New. (x86_linux_create_target): Initialize to_btrace_conf. * nat/linux-btrace.c (linux_enable_btrace): Update parameters. Check format. Split into this and ... (linux_enable_bts): ... this. (linux_btrace_conf): New. (perf_event_skip_record): Renamed into ... (perf_event_skip_bts_record): ... this. Updated users. (linux_disable_btrace): Split into this and ... (linux_disable_bts): ... this. (linux_read_btrace): Check format. * nat/linux-btrace.h (linux_enable_btrace): Update parameters. (linux_btrace_conf): New. (btrace_target_info)<ptid>: Moved. (btrace_target_info)<conf>: New. (btrace_target_info): Split into this and ... (btrace_tinfo_bts): ... this. Updated users. * btrace.c (btrace_enable): Update parameters. (btrace_conf, parse_xml_btrace_conf_bts, parse_xml_btrace_conf) (btrace_conf_children, btrace_conf_attributes) (btrace_conf_elements): New. * btrace.h (btrace_enable): Update parameters. (btrace_conf, parse_xml_btrace_conf): New. * common/btrace-common.h (btrace_config): New. * feature/btrace-conf.dtd: New. * record-btrace.c (record_btrace_conf): New. (record_btrace_cmdlist): New. (record_btrace_enable_warn, record_btrace_open): Pass &record_btrace_conf. (record_btrace_info): Print recording format. (cmd_record_btrace_bts_start): New. (cmd_record_btrace_start): Call cmd_record_btrace_bts_start. (_initialize_record_btrace): Add "record btrace bts" subcommand. Add "record bts" alias command. * remote.c (remote_state)<btrace_config>: New. (remote_btrace_reset, PACKET_qXfer_btrace_conf): New. (remote_protocol_features): Add qXfer:btrace-conf:read. (remote_open_1): Call remote_btrace_reset. (remote_xfer_partial): Handle TARGET_OBJECT_BTRACE_CONF. (btrace_target_info)<conf>: New. (btrace_sync_conf, btrace_read_config): New. (remote_enable_btrace): Update parameters. Call btrace_sync_conf and btrace_read_conf. (remote_btrace_conf): New. (init_remote_ops): Initialize to_btrace_conf. (_initialize_remote): Add qXfer:btrace-conf packet. * target.c (target_enable_btrace): Update parameters. (target_btrace_conf): New. * target.h (target_enable_btrace): Update parameters. (target_btrace_conf): New. (target_object)<TARGET_OBJECT_BTRACE_CONF>: New. (target_ops)<to_enable_btrace>: Update parameters and comment. (target_ops)<to_btrace_conf>: New. * target-delegates: Regenerate. * target-debug.h (target_debug_print_const_struct_btrace_config_p) (target_debug_print_const_struct_btrace_target_info_p): New. NEWS: Announce new command and new packet. doc/ * gdb.texinfo (Process Record and Replay): Describe the "record btrace bts" command. (General Query Packets): Describe qXfer:btrace-conf:read packet. (Branch Trace Configuration Format): New. gdbserver/ * linux-low.c (linux_low_enable_btrace): Update parameters. (linux_low_btrace_conf): New. (linux_target_ops)<to_btrace_conf>: Initialize. * server.c (current_btrace_conf): New. (handle_btrace_enable): Rename to ... (handle_btrace_enable_bts): ... this. Pass &current_btrace_conf to target_enable_btrace. Update comment. Update users. (handle_qxfer_btrace_conf): New. (qxfer_packets): Add btrace-conf entry. (handle_query): Report qXfer:btrace-conf:read as supported packet. * target.h (target_ops)<enable_btrace>: Update parameters and comment. (target_ops)<read_btrace_conf>: New. (target_enable_btrace): Update parameters. (target_read_btrace_conf): New. testsuite/ * gdb.btrace/delta.exp: Update "info record" output. * gdb.btrace/enable.exp: Update "info record" output. * gdb.btrace/finish.exp: Update "info record" output. * gdb.btrace/instruction_history.exp: Update "info record" output. * gdb.btrace/next.exp: Update "info record" output. * gdb.btrace/nexti.exp: Update "info record" output. * gdb.btrace/step.exp: Update "info record" output. * gdb.btrace/stepi.exp: Update "info record" output. * gdb.btrace/nohist.exp: Update "info record" output.
2013-11-28 22:44:13 +08:00
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_const_struct_btrace_config_p (arg2);
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_struct_btrace_target_info_p (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static void
delegate_disable_btrace (struct target_ops *self, struct btrace_target_info *arg1)
{
self = self->beneath;
self->to_disable_btrace (self, arg1);
}
static void
tdefault_disable_btrace (struct target_ops *self, struct btrace_target_info *arg1)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_disable_btrace (struct target_ops *self, struct btrace_target_info *arg1)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_disable_btrace (...)\n", debug_target.to_shortname);
debug_target.to_disable_btrace (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_disable_btrace (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_struct_btrace_target_info_p (arg1);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
delegate_teardown_btrace (struct target_ops *self, struct btrace_target_info *arg1)
{
self = self->beneath;
self->to_teardown_btrace (self, arg1);
}
static void
tdefault_teardown_btrace (struct target_ops *self, struct btrace_target_info *arg1)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_teardown_btrace (struct target_ops *self, struct btrace_target_info *arg1)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_teardown_btrace (...)\n", debug_target.to_shortname);
debug_target.to_teardown_btrace (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_teardown_btrace (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_struct_btrace_target_info_p (arg1);
fputs_unfiltered (")\n", gdb_stdlog);
}
static enum btrace_error
btrace: add struct btrace_data Add a structure to hold the branch trace data and an enum to describe the format of that data. So far, only BTS is supported. Also added a NONE format to indicate that no branch trace data is available. This will make it easier to support different branch trace formats in the future. 2015-02-09 Markus Metzger <markus.t.metzger@intel.com> * Makefile.in (SFILES): Add common/btrace-common.c. (COMMON_OBS): Add common/btrace-common.o. (btrace-common.o): Add build rules. * btrace.c (parse_xml_btrace): Update parameters. (parse_xml_btrace_block): Set format field. (btrace_add_pc, btrace_fetch): Use struct btrace_data. (do_btrace_data_cleanup, make_cleanup_btrace_data): New. (btrace_compute_ftrace): Split into this and... (btrace_compute_ftrace_bts): ...this. (btrace_stitch_trace): Split into this and... (btrace_stitch_bts): ...this. * btrace.h (parse_xml_btrace): Update parameters. (make_cleanup_btrace_data): New. * common/btrace-common.c: New. * common/btrace-common.h: Include common-defs.h. (btrace_block_s): Update comment. (btrace_format): New. (btrace_format_string): New. (btrace_data_bts): New. (btrace_data): New. (btrace_data_init, btrace_data_fini, btrace_data_empty): New. * remote.c (remote_read_btrace): Update parameters. * target.c (target_read_btrace): Update parameters. * target.h (target_read_btrace): Update parameters. (target_ops)<to_read_btrace>: Update parameters. * x86-linux-nat.c (x86_linux_read_btrace): Update parameters. * target-delegates.c: Regenerate. * target-debug (target_debug_print_struct_btrace_data_p): New. * nat/linux-btrace.c (linux_read_btrace): Split into this and... (linux_read_bts): ...this. * nat/linux-btrace.h (linux_read_btrace): Update parameters. gdbserver/ * Makefile.in (SFILES): Add common/btrace-common.c. (OBS): Add common/btrace-common.o. (btrace-common.o): Add build rules. * linux-low: Include btrace-common.h. (linux_low_read_btrace): Use struct btrace_data. Call btrace_data_init and btrace_data_fini.
2013-11-13 22:31:07 +08:00
delegate_read_btrace (struct target_ops *self, struct btrace_data *arg1, struct btrace_target_info *arg2, enum btrace_read_type arg3)
{
self = self->beneath;
return self->to_read_btrace (self, arg1, arg2, arg3);
}
static enum btrace_error
btrace: add struct btrace_data Add a structure to hold the branch trace data and an enum to describe the format of that data. So far, only BTS is supported. Also added a NONE format to indicate that no branch trace data is available. This will make it easier to support different branch trace formats in the future. 2015-02-09 Markus Metzger <markus.t.metzger@intel.com> * Makefile.in (SFILES): Add common/btrace-common.c. (COMMON_OBS): Add common/btrace-common.o. (btrace-common.o): Add build rules. * btrace.c (parse_xml_btrace): Update parameters. (parse_xml_btrace_block): Set format field. (btrace_add_pc, btrace_fetch): Use struct btrace_data. (do_btrace_data_cleanup, make_cleanup_btrace_data): New. (btrace_compute_ftrace): Split into this and... (btrace_compute_ftrace_bts): ...this. (btrace_stitch_trace): Split into this and... (btrace_stitch_bts): ...this. * btrace.h (parse_xml_btrace): Update parameters. (make_cleanup_btrace_data): New. * common/btrace-common.c: New. * common/btrace-common.h: Include common-defs.h. (btrace_block_s): Update comment. (btrace_format): New. (btrace_format_string): New. (btrace_data_bts): New. (btrace_data): New. (btrace_data_init, btrace_data_fini, btrace_data_empty): New. * remote.c (remote_read_btrace): Update parameters. * target.c (target_read_btrace): Update parameters. * target.h (target_read_btrace): Update parameters. (target_ops)<to_read_btrace>: Update parameters. * x86-linux-nat.c (x86_linux_read_btrace): Update parameters. * target-delegates.c: Regenerate. * target-debug (target_debug_print_struct_btrace_data_p): New. * nat/linux-btrace.c (linux_read_btrace): Split into this and... (linux_read_bts): ...this. * nat/linux-btrace.h (linux_read_btrace): Update parameters. gdbserver/ * Makefile.in (SFILES): Add common/btrace-common.c. (OBS): Add common/btrace-common.o. (btrace-common.o): Add build rules. * linux-low: Include btrace-common.h. (linux_low_read_btrace): Use struct btrace_data. Call btrace_data_init and btrace_data_fini.
2013-11-13 22:31:07 +08:00
tdefault_read_btrace (struct target_ops *self, struct btrace_data *arg1, struct btrace_target_info *arg2, enum btrace_read_type arg3)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static enum btrace_error
btrace: add struct btrace_data Add a structure to hold the branch trace data and an enum to describe the format of that data. So far, only BTS is supported. Also added a NONE format to indicate that no branch trace data is available. This will make it easier to support different branch trace formats in the future. 2015-02-09 Markus Metzger <markus.t.metzger@intel.com> * Makefile.in (SFILES): Add common/btrace-common.c. (COMMON_OBS): Add common/btrace-common.o. (btrace-common.o): Add build rules. * btrace.c (parse_xml_btrace): Update parameters. (parse_xml_btrace_block): Set format field. (btrace_add_pc, btrace_fetch): Use struct btrace_data. (do_btrace_data_cleanup, make_cleanup_btrace_data): New. (btrace_compute_ftrace): Split into this and... (btrace_compute_ftrace_bts): ...this. (btrace_stitch_trace): Split into this and... (btrace_stitch_bts): ...this. * btrace.h (parse_xml_btrace): Update parameters. (make_cleanup_btrace_data): New. * common/btrace-common.c: New. * common/btrace-common.h: Include common-defs.h. (btrace_block_s): Update comment. (btrace_format): New. (btrace_format_string): New. (btrace_data_bts): New. (btrace_data): New. (btrace_data_init, btrace_data_fini, btrace_data_empty): New. * remote.c (remote_read_btrace): Update parameters. * target.c (target_read_btrace): Update parameters. * target.h (target_read_btrace): Update parameters. (target_ops)<to_read_btrace>: Update parameters. * x86-linux-nat.c (x86_linux_read_btrace): Update parameters. * target-delegates.c: Regenerate. * target-debug (target_debug_print_struct_btrace_data_p): New. * nat/linux-btrace.c (linux_read_btrace): Split into this and... (linux_read_bts): ...this. * nat/linux-btrace.h (linux_read_btrace): Update parameters. gdbserver/ * Makefile.in (SFILES): Add common/btrace-common.c. (OBS): Add common/btrace-common.o. (btrace-common.o): Add build rules. * linux-low: Include btrace-common.h. (linux_low_read_btrace): Use struct btrace_data. Call btrace_data_init and btrace_data_fini.
2013-11-13 22:31:07 +08:00
debug_read_btrace (struct target_ops *self, struct btrace_data *arg1, struct btrace_target_info *arg2, enum btrace_read_type arg3)
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
{
enum btrace_error result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_read_btrace (...)\n", debug_target.to_shortname);
result = debug_target.to_read_btrace (&debug_target, arg1, arg2, arg3);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_read_btrace (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
btrace: add struct btrace_data Add a structure to hold the branch trace data and an enum to describe the format of that data. So far, only BTS is supported. Also added a NONE format to indicate that no branch trace data is available. This will make it easier to support different branch trace formats in the future. 2015-02-09 Markus Metzger <markus.t.metzger@intel.com> * Makefile.in (SFILES): Add common/btrace-common.c. (COMMON_OBS): Add common/btrace-common.o. (btrace-common.o): Add build rules. * btrace.c (parse_xml_btrace): Update parameters. (parse_xml_btrace_block): Set format field. (btrace_add_pc, btrace_fetch): Use struct btrace_data. (do_btrace_data_cleanup, make_cleanup_btrace_data): New. (btrace_compute_ftrace): Split into this and... (btrace_compute_ftrace_bts): ...this. (btrace_stitch_trace): Split into this and... (btrace_stitch_bts): ...this. * btrace.h (parse_xml_btrace): Update parameters. (make_cleanup_btrace_data): New. * common/btrace-common.c: New. * common/btrace-common.h: Include common-defs.h. (btrace_block_s): Update comment. (btrace_format): New. (btrace_format_string): New. (btrace_data_bts): New. (btrace_data): New. (btrace_data_init, btrace_data_fini, btrace_data_empty): New. * remote.c (remote_read_btrace): Update parameters. * target.c (target_read_btrace): Update parameters. * target.h (target_read_btrace): Update parameters. (target_ops)<to_read_btrace>: Update parameters. * x86-linux-nat.c (x86_linux_read_btrace): Update parameters. * target-delegates.c: Regenerate. * target-debug (target_debug_print_struct_btrace_data_p): New. * nat/linux-btrace.c (linux_read_btrace): Split into this and... (linux_read_bts): ...this. * nat/linux-btrace.h (linux_read_btrace): Update parameters. gdbserver/ * Makefile.in (SFILES): Add common/btrace-common.c. (OBS): Add common/btrace-common.o. (btrace-common.o): Add build rules. * linux-low: Include btrace-common.h. (linux_low_read_btrace): Use struct btrace_data. Call btrace_data_init and btrace_data_fini.
2013-11-13 22:31:07 +08:00
target_debug_print_struct_btrace_data_p (arg1);
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_struct_btrace_target_info_p (arg2);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_enum_btrace_read_type (arg3);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_enum_btrace_error (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
record btrace: add configuration struct Add a struct to describe the branch trace configuration and use it for enabling branch tracing. The user will be able to set configuration fields for each tracing format to be used for new threads. The actual configuration that is active for a given thread will be shown in the "info record" command. At the moment, the configuration struct only contains a format field that is set to the only available format. The format is the only configuration option that can not be set via set commands. It is given as argument to the "record btrace" command when starting recording. 2015-02-09 Markus Metzger <markus.t.metzger@intel.com> * Makefile.in (XMLFILES): Add btrace-conf.dtd. * x86-linux-nat.c (x86_linux_enable_btrace): Update parameters. (x86_linux_btrace_conf): New. (x86_linux_create_target): Initialize to_btrace_conf. * nat/linux-btrace.c (linux_enable_btrace): Update parameters. Check format. Split into this and ... (linux_enable_bts): ... this. (linux_btrace_conf): New. (perf_event_skip_record): Renamed into ... (perf_event_skip_bts_record): ... this. Updated users. (linux_disable_btrace): Split into this and ... (linux_disable_bts): ... this. (linux_read_btrace): Check format. * nat/linux-btrace.h (linux_enable_btrace): Update parameters. (linux_btrace_conf): New. (btrace_target_info)<ptid>: Moved. (btrace_target_info)<conf>: New. (btrace_target_info): Split into this and ... (btrace_tinfo_bts): ... this. Updated users. * btrace.c (btrace_enable): Update parameters. (btrace_conf, parse_xml_btrace_conf_bts, parse_xml_btrace_conf) (btrace_conf_children, btrace_conf_attributes) (btrace_conf_elements): New. * btrace.h (btrace_enable): Update parameters. (btrace_conf, parse_xml_btrace_conf): New. * common/btrace-common.h (btrace_config): New. * feature/btrace-conf.dtd: New. * record-btrace.c (record_btrace_conf): New. (record_btrace_cmdlist): New. (record_btrace_enable_warn, record_btrace_open): Pass &record_btrace_conf. (record_btrace_info): Print recording format. (cmd_record_btrace_bts_start): New. (cmd_record_btrace_start): Call cmd_record_btrace_bts_start. (_initialize_record_btrace): Add "record btrace bts" subcommand. Add "record bts" alias command. * remote.c (remote_state)<btrace_config>: New. (remote_btrace_reset, PACKET_qXfer_btrace_conf): New. (remote_protocol_features): Add qXfer:btrace-conf:read. (remote_open_1): Call remote_btrace_reset. (remote_xfer_partial): Handle TARGET_OBJECT_BTRACE_CONF. (btrace_target_info)<conf>: New. (btrace_sync_conf, btrace_read_config): New. (remote_enable_btrace): Update parameters. Call btrace_sync_conf and btrace_read_conf. (remote_btrace_conf): New. (init_remote_ops): Initialize to_btrace_conf. (_initialize_remote): Add qXfer:btrace-conf packet. * target.c (target_enable_btrace): Update parameters. (target_btrace_conf): New. * target.h (target_enable_btrace): Update parameters. (target_btrace_conf): New. (target_object)<TARGET_OBJECT_BTRACE_CONF>: New. (target_ops)<to_enable_btrace>: Update parameters and comment. (target_ops)<to_btrace_conf>: New. * target-delegates: Regenerate. * target-debug.h (target_debug_print_const_struct_btrace_config_p) (target_debug_print_const_struct_btrace_target_info_p): New. NEWS: Announce new command and new packet. doc/ * gdb.texinfo (Process Record and Replay): Describe the "record btrace bts" command. (General Query Packets): Describe qXfer:btrace-conf:read packet. (Branch Trace Configuration Format): New. gdbserver/ * linux-low.c (linux_low_enable_btrace): Update parameters. (linux_low_btrace_conf): New. (linux_target_ops)<to_btrace_conf>: Initialize. * server.c (current_btrace_conf): New. (handle_btrace_enable): Rename to ... (handle_btrace_enable_bts): ... this. Pass &current_btrace_conf to target_enable_btrace. Update comment. Update users. (handle_qxfer_btrace_conf): New. (qxfer_packets): Add btrace-conf entry. (handle_query): Report qXfer:btrace-conf:read as supported packet. * target.h (target_ops)<enable_btrace>: Update parameters and comment. (target_ops)<read_btrace_conf>: New. (target_enable_btrace): Update parameters. (target_read_btrace_conf): New. testsuite/ * gdb.btrace/delta.exp: Update "info record" output. * gdb.btrace/enable.exp: Update "info record" output. * gdb.btrace/finish.exp: Update "info record" output. * gdb.btrace/instruction_history.exp: Update "info record" output. * gdb.btrace/next.exp: Update "info record" output. * gdb.btrace/nexti.exp: Update "info record" output. * gdb.btrace/step.exp: Update "info record" output. * gdb.btrace/stepi.exp: Update "info record" output. * gdb.btrace/nohist.exp: Update "info record" output.
2013-11-28 22:44:13 +08:00
static const struct btrace_config *
delegate_btrace_conf (struct target_ops *self, const struct btrace_target_info *arg1)
{
self = self->beneath;
return self->to_btrace_conf (self, arg1);
}
static const struct btrace_config *
tdefault_btrace_conf (struct target_ops *self, const struct btrace_target_info *arg1)
{
return NULL;
}
static const struct btrace_config *
debug_btrace_conf (struct target_ops *self, const struct btrace_target_info *arg1)
{
const struct btrace_config * result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_btrace_conf (...)\n", debug_target.to_shortname);
result = debug_target.to_btrace_conf (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_btrace_conf (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_const_struct_btrace_target_info_p (arg1);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_const_struct_btrace_config_p (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static void
delegate_stop_recording (struct target_ops *self)
{
self = self->beneath;
self->to_stop_recording (self);
}
static void
tdefault_stop_recording (struct target_ops *self)
{
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_stop_recording (struct target_ops *self)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_stop_recording (...)\n", debug_target.to_shortname);
debug_target.to_stop_recording (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_stop_recording (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
delegate_info_record (struct target_ops *self)
{
self = self->beneath;
self->to_info_record (self);
}
static void
tdefault_info_record (struct target_ops *self)
{
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_info_record (struct target_ops *self)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_info_record (...)\n", debug_target.to_shortname);
debug_target.to_info_record (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_info_record (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
delegate_save_record (struct target_ops *self, const char *arg1)
{
self = self->beneath;
self->to_save_record (self, arg1);
}
static void
tdefault_save_record (struct target_ops *self, const char *arg1)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_save_record (struct target_ops *self, const char *arg1)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_save_record (...)\n", debug_target.to_shortname);
debug_target.to_save_record (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_save_record (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_const_char_p (arg1);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
delegate_delete_record (struct target_ops *self)
{
self = self->beneath;
self->to_delete_record (self);
}
static void
tdefault_delete_record (struct target_ops *self)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_delete_record (struct target_ops *self)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_delete_record (...)\n", debug_target.to_shortname);
debug_target.to_delete_record (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_delete_record (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (")\n", gdb_stdlog);
}
static int
delegate_record_is_replaying (struct target_ops *self)
{
self = self->beneath;
return self->to_record_is_replaying (self);
}
static int
tdefault_record_is_replaying (struct target_ops *self)
{
return 0;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_record_is_replaying (struct target_ops *self)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_record_is_replaying (...)\n", debug_target.to_shortname);
result = debug_target.to_record_is_replaying (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_record_is_replaying (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static void
delegate_goto_record_begin (struct target_ops *self)
{
self = self->beneath;
self->to_goto_record_begin (self);
}
static void
tdefault_goto_record_begin (struct target_ops *self)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_goto_record_begin (struct target_ops *self)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_goto_record_begin (...)\n", debug_target.to_shortname);
debug_target.to_goto_record_begin (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_goto_record_begin (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
delegate_goto_record_end (struct target_ops *self)
{
self = self->beneath;
self->to_goto_record_end (self);
}
static void
tdefault_goto_record_end (struct target_ops *self)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_goto_record_end (struct target_ops *self)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_goto_record_end (...)\n", debug_target.to_shortname);
debug_target.to_goto_record_end (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_goto_record_end (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
delegate_goto_record (struct target_ops *self, ULONGEST arg1)
{
self = self->beneath;
self->to_goto_record (self, arg1);
}
static void
tdefault_goto_record (struct target_ops *self, ULONGEST arg1)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_goto_record (struct target_ops *self, ULONGEST arg1)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_goto_record (...)\n", debug_target.to_shortname);
debug_target.to_goto_record (&debug_target, arg1);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_goto_record (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_ULONGEST (arg1);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
delegate_insn_history (struct target_ops *self, int arg1, int arg2)
{
self = self->beneath;
self->to_insn_history (self, arg1, arg2);
}
static void
tdefault_insn_history (struct target_ops *self, int arg1, int arg2)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_insn_history (struct target_ops *self, int arg1, int arg2)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_insn_history (...)\n", debug_target.to_shortname);
debug_target.to_insn_history (&debug_target, arg1, arg2);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_insn_history (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg2);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
delegate_insn_history_from (struct target_ops *self, ULONGEST arg1, int arg2, int arg3)
{
self = self->beneath;
self->to_insn_history_from (self, arg1, arg2, arg3);
}
static void
tdefault_insn_history_from (struct target_ops *self, ULONGEST arg1, int arg2, int arg3)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_insn_history_from (struct target_ops *self, ULONGEST arg1, int arg2, int arg3)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_insn_history_from (...)\n", debug_target.to_shortname);
debug_target.to_insn_history_from (&debug_target, arg1, arg2, arg3);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_insn_history_from (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_ULONGEST (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg2);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg3);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
delegate_insn_history_range (struct target_ops *self, ULONGEST arg1, ULONGEST arg2, int arg3)
{
self = self->beneath;
self->to_insn_history_range (self, arg1, arg2, arg3);
}
static void
tdefault_insn_history_range (struct target_ops *self, ULONGEST arg1, ULONGEST arg2, int arg3)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_insn_history_range (struct target_ops *self, ULONGEST arg1, ULONGEST arg2, int arg3)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_insn_history_range (...)\n", debug_target.to_shortname);
debug_target.to_insn_history_range (&debug_target, arg1, arg2, arg3);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_insn_history_range (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_ULONGEST (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_ULONGEST (arg2);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg3);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
delegate_call_history (struct target_ops *self, int arg1, int arg2)
{
self = self->beneath;
self->to_call_history (self, arg1, arg2);
}
static void
tdefault_call_history (struct target_ops *self, int arg1, int arg2)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_call_history (struct target_ops *self, int arg1, int arg2)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_call_history (...)\n", debug_target.to_shortname);
debug_target.to_call_history (&debug_target, arg1, arg2);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_call_history (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg2);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
delegate_call_history_from (struct target_ops *self, ULONGEST arg1, int arg2, int arg3)
{
self = self->beneath;
self->to_call_history_from (self, arg1, arg2, arg3);
}
static void
tdefault_call_history_from (struct target_ops *self, ULONGEST arg1, int arg2, int arg3)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_call_history_from (struct target_ops *self, ULONGEST arg1, int arg2, int arg3)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_call_history_from (...)\n", debug_target.to_shortname);
debug_target.to_call_history_from (&debug_target, arg1, arg2, arg3);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_call_history_from (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_ULONGEST (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg2);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg3);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
delegate_call_history_range (struct target_ops *self, ULONGEST arg1, ULONGEST arg2, int arg3)
{
self = self->beneath;
self->to_call_history_range (self, arg1, arg2, arg3);
}
static void
tdefault_call_history_range (struct target_ops *self, ULONGEST arg1, ULONGEST arg2, int arg3)
{
tcomplain ();
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_call_history_range (struct target_ops *self, ULONGEST arg1, ULONGEST arg2, int arg3)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_call_history_range (...)\n", debug_target.to_shortname);
debug_target.to_call_history_range (&debug_target, arg1, arg2, arg3);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_call_history_range (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_ULONGEST (arg1);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_ULONGEST (arg2);
fputs_unfiltered (", ", gdb_stdlog);
target_debug_print_int (arg3);
fputs_unfiltered (")\n", gdb_stdlog);
}
static int
delegate_augmented_libraries_svr4_read (struct target_ops *self)
{
self = self->beneath;
return self->to_augmented_libraries_svr4_read (self);
}
static int
tdefault_augmented_libraries_svr4_read (struct target_ops *self)
{
return 0;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static int
debug_augmented_libraries_svr4_read (struct target_ops *self)
{
int result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_augmented_libraries_svr4_read (...)\n", debug_target.to_shortname);
result = debug_target.to_augmented_libraries_svr4_read (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_augmented_libraries_svr4_read (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_int (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static const struct frame_unwind *
delegate_get_unwinder (struct target_ops *self)
{
self = self->beneath;
return self->to_get_unwinder (self);
}
static const struct frame_unwind *
tdefault_get_unwinder (struct target_ops *self)
{
return NULL;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static const struct frame_unwind *
debug_get_unwinder (struct target_ops *self)
{
const struct frame_unwind * result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_get_unwinder (...)\n", debug_target.to_shortname);
result = debug_target.to_get_unwinder (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_get_unwinder (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_const_struct_frame_unwind_p (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static const struct frame_unwind *
delegate_get_tailcall_unwinder (struct target_ops *self)
{
self = self->beneath;
return self->to_get_tailcall_unwinder (self);
}
static const struct frame_unwind *
tdefault_get_tailcall_unwinder (struct target_ops *self)
{
return NULL;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static const struct frame_unwind *
debug_get_tailcall_unwinder (struct target_ops *self)
{
const struct frame_unwind * result;
fprintf_unfiltered (gdb_stdlog, "-> %s->to_get_tailcall_unwinder (...)\n", debug_target.to_shortname);
result = debug_target.to_get_tailcall_unwinder (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_get_tailcall_unwinder (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (") = ", gdb_stdlog);
target_debug_print_const_struct_frame_unwind_p (result);
fputs_unfiltered ("\n", gdb_stdlog);
return result;
}
static void
delegate_prepare_to_generate_core (struct target_ops *self)
{
self = self->beneath;
self->to_prepare_to_generate_core (self);
}
static void
tdefault_prepare_to_generate_core (struct target_ops *self)
{
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_prepare_to_generate_core (struct target_ops *self)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_prepare_to_generate_core (...)\n", debug_target.to_shortname);
debug_target.to_prepare_to_generate_core (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_prepare_to_generate_core (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
delegate_done_generating_core (struct target_ops *self)
{
self = self->beneath;
self->to_done_generating_core (self);
}
static void
tdefault_done_generating_core (struct target_ops *self)
{
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
debug_done_generating_core (struct target_ops *self)
{
fprintf_unfiltered (gdb_stdlog, "-> %s->to_done_generating_core (...)\n", debug_target.to_shortname);
debug_target.to_done_generating_core (&debug_target);
fprintf_unfiltered (gdb_stdlog, "<- %s->to_done_generating_core (", debug_target.to_shortname);
target_debug_print_struct_target_ops_p (&debug_target);
fputs_unfiltered (")\n", gdb_stdlog);
}
static void
install_delegators (struct target_ops *ops)
{
if (ops->to_post_attach == NULL)
ops->to_post_attach = delegate_post_attach;
if (ops->to_detach == NULL)
ops->to_detach = delegate_detach;
if (ops->to_disconnect == NULL)
ops->to_disconnect = delegate_disconnect;
add target method delegation This patch replaces some code in the record targets with target method delegation. record-full.c stores pointers to many target methods when the record target is pushed. Then it later delegates some calls via these. This is wrong because it violates the target stack contract. In particular it is ok to unpush a target at any stratum, but record-full does not keep track of this, so it could potentially call into an unpushed target. This patch fixes the problem by using the newly-introduced generic approach to target delegation for the methods in question. 2014-02-19 Tom Tromey <tromey@redhat.com> * record-full.c (record_full_beneath_to_resume_ops) (record_full_beneath_to_resume, record_full_beneath_to_wait_ops) (record_full_beneath_to_wait) (record_full_beneath_to_store_registers_ops) (record_full_beneath_to_store_registers) (record_full_beneath_to_xfer_partial_ops) (record_full_beneath_to_xfer_partial) (record_full_beneath_to_insert_breakpoint_ops) (record_full_beneath_to_insert_breakpoint) (record_full_beneath_to_remove_breakpoint_ops) (record_full_beneath_to_remove_breakpoint) (record_full_beneath_to_stopped_by_watchpoint) (record_full_beneath_to_stopped_data_address) (record_full_beneath_to_async, tmp_to_resume_ops, tmp_to_resume) (tmp_to_wait_ops, tmp_to_wait, tmp_to_store_registers_ops) (tmp_to_store_registers, tmp_to_xfer_partial_ops) (tmp_to_xfer_partial, tmp_to_instmp_to_insert_breakpoint_ops) (tmp_to_insert_breakpoint, tmp_to_remove_breakpoint_ops) (tmp_to_remove_breakpoint, tmp_to_stopped_by_watchpoint) (tmp_to_stopped_data_address, tmp_to_async): Remove. (record_full_open_1, record_full_open): Update. Use RECORD_IS_USED. (record_full_resume, record_full_wait_1) (record_full_stopped_by_watchpoint, record_full_stopped_data_address) (record_full_store_registers, record_full_xfer_partial) (record_full_insert_breakpoint, record_full_remove_breakpoint) (record_full_async, record_full_core_xfer_partial): Use target delegation. * target-delegates.c: Rebuild. * target.c (current_xfer_partial): Remove. (update_current_target): Do not INHERIT or de_fault to_insert_breakpoint, to_remove_breakpoint, to_stopped_data_address, to_stopped_by_watchpoint, to_can_async_p, to_is_async_p, to_async. Do not set to_xfer_partial field. (default_xfer_partial): Simplify. (current_xfer_partial): Remove. (target_wait, target_resume): Simplify. (find_default_can_async_p, find_default_is_async_p): Update. (init_dummy_target): Don't set to_can_async_p, to_is_async_p, to_xfer_partial, to_stopped_by_watchpoint, to_stopped_data_address. (target_store_registers): Simplify. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. (target_remove_breakpoint, target_insert_breakpoint) (debug_to_insert_breakpoint, debug_to_remove_breakpoint): Update. * target.h (struct target_ops) <to_resume, to_wait, to_store_registers, to_insert_breakpoint, to_remove_breakpoint, to_stopped_by_watchpoint, to_stopped_data_address, to_can_async_p, to_is_async_p, to_async, to_xfer_partial>: Add TARGET_DEFAULT markup. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. * record-btrace.c (record_btrace_remove_breakpoint): Delegate directly. (record_btrace_insert_breakpoint): Delegate directly.
2013-07-13 03:00:34 +08:00
if (ops->to_resume == NULL)
ops->to_resume = delegate_resume;
if (ops->to_wait == NULL)
ops->to_wait = delegate_wait;
if (ops->to_fetch_registers == NULL)
ops->to_fetch_registers = delegate_fetch_registers;
add target method delegation This patch replaces some code in the record targets with target method delegation. record-full.c stores pointers to many target methods when the record target is pushed. Then it later delegates some calls via these. This is wrong because it violates the target stack contract. In particular it is ok to unpush a target at any stratum, but record-full does not keep track of this, so it could potentially call into an unpushed target. This patch fixes the problem by using the newly-introduced generic approach to target delegation for the methods in question. 2014-02-19 Tom Tromey <tromey@redhat.com> * record-full.c (record_full_beneath_to_resume_ops) (record_full_beneath_to_resume, record_full_beneath_to_wait_ops) (record_full_beneath_to_wait) (record_full_beneath_to_store_registers_ops) (record_full_beneath_to_store_registers) (record_full_beneath_to_xfer_partial_ops) (record_full_beneath_to_xfer_partial) (record_full_beneath_to_insert_breakpoint_ops) (record_full_beneath_to_insert_breakpoint) (record_full_beneath_to_remove_breakpoint_ops) (record_full_beneath_to_remove_breakpoint) (record_full_beneath_to_stopped_by_watchpoint) (record_full_beneath_to_stopped_data_address) (record_full_beneath_to_async, tmp_to_resume_ops, tmp_to_resume) (tmp_to_wait_ops, tmp_to_wait, tmp_to_store_registers_ops) (tmp_to_store_registers, tmp_to_xfer_partial_ops) (tmp_to_xfer_partial, tmp_to_instmp_to_insert_breakpoint_ops) (tmp_to_insert_breakpoint, tmp_to_remove_breakpoint_ops) (tmp_to_remove_breakpoint, tmp_to_stopped_by_watchpoint) (tmp_to_stopped_data_address, tmp_to_async): Remove. (record_full_open_1, record_full_open): Update. Use RECORD_IS_USED. (record_full_resume, record_full_wait_1) (record_full_stopped_by_watchpoint, record_full_stopped_data_address) (record_full_store_registers, record_full_xfer_partial) (record_full_insert_breakpoint, record_full_remove_breakpoint) (record_full_async, record_full_core_xfer_partial): Use target delegation. * target-delegates.c: Rebuild. * target.c (current_xfer_partial): Remove. (update_current_target): Do not INHERIT or de_fault to_insert_breakpoint, to_remove_breakpoint, to_stopped_data_address, to_stopped_by_watchpoint, to_can_async_p, to_is_async_p, to_async. Do not set to_xfer_partial field. (default_xfer_partial): Simplify. (current_xfer_partial): Remove. (target_wait, target_resume): Simplify. (find_default_can_async_p, find_default_is_async_p): Update. (init_dummy_target): Don't set to_can_async_p, to_is_async_p, to_xfer_partial, to_stopped_by_watchpoint, to_stopped_data_address. (target_store_registers): Simplify. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. (target_remove_breakpoint, target_insert_breakpoint) (debug_to_insert_breakpoint, debug_to_remove_breakpoint): Update. * target.h (struct target_ops) <to_resume, to_wait, to_store_registers, to_insert_breakpoint, to_remove_breakpoint, to_stopped_by_watchpoint, to_stopped_data_address, to_can_async_p, to_is_async_p, to_async, to_xfer_partial>: Add TARGET_DEFAULT markup. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. * record-btrace.c (record_btrace_remove_breakpoint): Delegate directly. (record_btrace_insert_breakpoint): Delegate directly.
2013-07-13 03:00:34 +08:00
if (ops->to_store_registers == NULL)
ops->to_store_registers = delegate_store_registers;
if (ops->to_prepare_to_store == NULL)
ops->to_prepare_to_store = delegate_prepare_to_store;
if (ops->to_files_info == NULL)
ops->to_files_info = delegate_files_info;
add target method delegation This patch replaces some code in the record targets with target method delegation. record-full.c stores pointers to many target methods when the record target is pushed. Then it later delegates some calls via these. This is wrong because it violates the target stack contract. In particular it is ok to unpush a target at any stratum, but record-full does not keep track of this, so it could potentially call into an unpushed target. This patch fixes the problem by using the newly-introduced generic approach to target delegation for the methods in question. 2014-02-19 Tom Tromey <tromey@redhat.com> * record-full.c (record_full_beneath_to_resume_ops) (record_full_beneath_to_resume, record_full_beneath_to_wait_ops) (record_full_beneath_to_wait) (record_full_beneath_to_store_registers_ops) (record_full_beneath_to_store_registers) (record_full_beneath_to_xfer_partial_ops) (record_full_beneath_to_xfer_partial) (record_full_beneath_to_insert_breakpoint_ops) (record_full_beneath_to_insert_breakpoint) (record_full_beneath_to_remove_breakpoint_ops) (record_full_beneath_to_remove_breakpoint) (record_full_beneath_to_stopped_by_watchpoint) (record_full_beneath_to_stopped_data_address) (record_full_beneath_to_async, tmp_to_resume_ops, tmp_to_resume) (tmp_to_wait_ops, tmp_to_wait, tmp_to_store_registers_ops) (tmp_to_store_registers, tmp_to_xfer_partial_ops) (tmp_to_xfer_partial, tmp_to_instmp_to_insert_breakpoint_ops) (tmp_to_insert_breakpoint, tmp_to_remove_breakpoint_ops) (tmp_to_remove_breakpoint, tmp_to_stopped_by_watchpoint) (tmp_to_stopped_data_address, tmp_to_async): Remove. (record_full_open_1, record_full_open): Update. Use RECORD_IS_USED. (record_full_resume, record_full_wait_1) (record_full_stopped_by_watchpoint, record_full_stopped_data_address) (record_full_store_registers, record_full_xfer_partial) (record_full_insert_breakpoint, record_full_remove_breakpoint) (record_full_async, record_full_core_xfer_partial): Use target delegation. * target-delegates.c: Rebuild. * target.c (current_xfer_partial): Remove. (update_current_target): Do not INHERIT or de_fault to_insert_breakpoint, to_remove_breakpoint, to_stopped_data_address, to_stopped_by_watchpoint, to_can_async_p, to_is_async_p, to_async. Do not set to_xfer_partial field. (default_xfer_partial): Simplify. (current_xfer_partial): Remove. (target_wait, target_resume): Simplify. (find_default_can_async_p, find_default_is_async_p): Update. (init_dummy_target): Don't set to_can_async_p, to_is_async_p, to_xfer_partial, to_stopped_by_watchpoint, to_stopped_data_address. (target_store_registers): Simplify. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. (target_remove_breakpoint, target_insert_breakpoint) (debug_to_insert_breakpoint, debug_to_remove_breakpoint): Update. * target.h (struct target_ops) <to_resume, to_wait, to_store_registers, to_insert_breakpoint, to_remove_breakpoint, to_stopped_by_watchpoint, to_stopped_data_address, to_can_async_p, to_is_async_p, to_async, to_xfer_partial>: Add TARGET_DEFAULT markup. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. * record-btrace.c (record_btrace_remove_breakpoint): Delegate directly. (record_btrace_insert_breakpoint): Delegate directly.
2013-07-13 03:00:34 +08:00
if (ops->to_insert_breakpoint == NULL)
ops->to_insert_breakpoint = delegate_insert_breakpoint;
if (ops->to_remove_breakpoint == NULL)
ops->to_remove_breakpoint = delegate_remove_breakpoint;
Teach GDB about targets that can tell whether a trap is a breakpoint event The moribund locations heuristics are problematic. This patch teaches GDB about targets that can reliably tell whether a trap was caused by a software or hardware breakpoint, and thus don't need moribund locations, thus bypassing all the problems that mechanism has. The non-stop-fair-events.exp test is frequently failing currently. E.g., see https://sourceware.org/ml/gdb-testers/2015-q1/msg03148.html. The root cause is a fundamental problem with moribund locations. For example, the stepped_breakpoint logic added by af48d08f breaks in this case (which is what happens with that test): - Step thread A, no breakpoint is set at PC. - The kernel doesn't schedule thread A yet. - Insert breakpoint at A's PC, for some reason (e.g., a step-resume breakpoint for thread B). - Kernel finally schedules thread A. - thread A's stepped_breakpoint flag is not set, even though it now stepped a breakpoint instruction. - adjust_pc_after_break gets the PC wrong, because PC == PREV_PC, but stepped_breakpoint is not set. We needed the stepped_breakpoint logic to workaround moribund locations, because otherwise adjust_pc_after_break could apply an adjustment when it shouldn't just because there _used_ to be a breakpoint at PC (a moribund breakpoint location). For example, on x86, that's wrong if the thread really hasn't executed an int3, but instead executed some other 1-byte long instruction. Getting the PC adjustment wrong of course leads to the inferior executing the wrong instruction. Other problems with moribund locations are: - if a true SIGTRAP happens to be raised when the program is executing the PC that used to have a breakpoint, GDB will assume that is a trap for a breakpoint that has recently been removed, and thus we miss reporting the random signal to the user. - to minimize that, we get rid of moribund location after a while. That while is defined as just a certain number of events being processed. That number of events sometimes passes by before a delayed breakpoint is processed, and GDB confuses the trap for a random signal, thus reporting the random trap. Once the user resumes the thread, the program crashes because the PC was not adjusted... The fix for all this is to bite the bullet and get rid of heuristics and instead rely on the target knowing accurately what caused the SIGTRAP. The target/kernel/stub is in the best position to know what that, because it can e.g. consult priviledged CPU flags GDB has no access to, or by knowing which exception vector entry was called when the instruction trapped, etc. Most debug APIs I've seen to date report breakpoint hits as a distinct event in some fashion. For example, on the Linux kernel, whether a breakpoint was executed is exposed to userspace in the si_code field of the SIGTRAP's siginfo. On Windows, the debug API reports a EXCEPTION_BREAKPOINT exception code. We needed to keep around deleted breakpoints in an on-the-side list (the moribund locations) for two main reasons: - Know that a SIGTRAP actually is a delayed event for a hit of a breakpoint that was removed before the event was processed, and thus should not be reported as a random signal. - So we still do the decr_pc_after_break adjustment in that case, so that the thread is resumed at the correct address. In the new model, if GDB processes an event the target tells is a breakpoint trap, and GDB doesn't find the corresponding breakpoint in its breakpoint tables, it means that event is a delayed event for a breakpoint that has since been removed, and thus the event should be ignored. For the decr_pc_after_after issue, it ends up being much simpler that on targets that can reliably tell whether a breakpoint trapped, for the breakpoint trap to present the PC already adjusted. Proper multi-threading support already implies that targets needs to be doing decr_pc_after_break adjustment themselves, otherwise for example, in all-stop if two threads hit a breakpoint simultaneously, and the user does "info threads", he'll see the non-event thread that hit the breakpoint stopped at the wrong PC. This way (target adjusts) also ends up eliminating the need for some awkward re-incrementing of the PC in the record-full and Linux targets that we do today, and the need for the target_decr_pc_after_break hook. If the target always adjusts, then there's a case where GDB needs to re-increment the PC. Say, on x86, an "int3" instruction that was explicitly written in the program traps. In this case, GDB should report a random SIGTRAP signal to the user, with the PC pointing at the instruction past the int3, just like if GDB was not debugging the program. The user may well decide to pass the SIGTRAP to the program because the program being debugged has a SIGTRAP handler that handles its own breakpoints, and expects the PC to be unadjusted. Tested on x86-64 Fedora 20. gdb/ChangeLog: 2015-03-04 Pedro Alves <palves@redhat.com> * breakpoint.c (need_moribund_for_location_type): New function. (bpstat_stop_status): Don't skipping checking moribund locations of breakpoint types which the target tell caused a stop. (program_breakpoint_here_p): New function, factored out from ... (bp_loc_is_permanent): ... this. (update_global_location_list): Don't create a moribund location if the target supports reporting stops of the type of the removed breakpoint. * breakpoint.h (program_breakpoint_here_p): New declaration. * infrun.c (adjust_pc_after_break): Return early if the target has already adjusted the PC. Add comments. (handle_signal_stop): If nothing explains a signal, and the target tells us the stop was caused by a software breakpoint, check if there's a breakpoint instruction in the memory. If so, adjust the PC before presenting the stop to the user. Otherwise, ignore the trap. If nothing explains a signal, and the target tells us the stop was caused by a hardware breakpoint, ignore the trap. * target.h (struct target_ops) <to_stopped_by_sw_breakpoint, to_supports_stopped_by_sw_breakpoint, to_stopped_by_hw_breakpoint, to_supports_stopped_by_hw_breakpoint>: New fields. (target_stopped_by_sw_breakpoint) (target_supports_stopped_by_sw_breakpoint) (target_stopped_by_hw_breakpoint) (target_supports_stopped_by_hw_breakpoint): Define. * target-delegates.c: Regenerate.
2015-03-05 04:41:15 +08:00
if (ops->to_stopped_by_sw_breakpoint == NULL)
ops->to_stopped_by_sw_breakpoint = delegate_stopped_by_sw_breakpoint;
if (ops->to_supports_stopped_by_sw_breakpoint == NULL)
ops->to_supports_stopped_by_sw_breakpoint = delegate_supports_stopped_by_sw_breakpoint;
if (ops->to_stopped_by_hw_breakpoint == NULL)
ops->to_stopped_by_hw_breakpoint = delegate_stopped_by_hw_breakpoint;
if (ops->to_supports_stopped_by_hw_breakpoint == NULL)
ops->to_supports_stopped_by_hw_breakpoint = delegate_supports_stopped_by_hw_breakpoint;
if (ops->to_can_use_hw_breakpoint == NULL)
ops->to_can_use_hw_breakpoint = delegate_can_use_hw_breakpoint;
if (ops->to_ranged_break_num_registers == NULL)
ops->to_ranged_break_num_registers = delegate_ranged_break_num_registers;
if (ops->to_insert_hw_breakpoint == NULL)
ops->to_insert_hw_breakpoint = delegate_insert_hw_breakpoint;
if (ops->to_remove_hw_breakpoint == NULL)
ops->to_remove_hw_breakpoint = delegate_remove_hw_breakpoint;
if (ops->to_remove_watchpoint == NULL)
ops->to_remove_watchpoint = delegate_remove_watchpoint;
if (ops->to_insert_watchpoint == NULL)
ops->to_insert_watchpoint = delegate_insert_watchpoint;
if (ops->to_insert_mask_watchpoint == NULL)
ops->to_insert_mask_watchpoint = delegate_insert_mask_watchpoint;
if (ops->to_remove_mask_watchpoint == NULL)
ops->to_remove_mask_watchpoint = delegate_remove_mask_watchpoint;
add target method delegation This patch replaces some code in the record targets with target method delegation. record-full.c stores pointers to many target methods when the record target is pushed. Then it later delegates some calls via these. This is wrong because it violates the target stack contract. In particular it is ok to unpush a target at any stratum, but record-full does not keep track of this, so it could potentially call into an unpushed target. This patch fixes the problem by using the newly-introduced generic approach to target delegation for the methods in question. 2014-02-19 Tom Tromey <tromey@redhat.com> * record-full.c (record_full_beneath_to_resume_ops) (record_full_beneath_to_resume, record_full_beneath_to_wait_ops) (record_full_beneath_to_wait) (record_full_beneath_to_store_registers_ops) (record_full_beneath_to_store_registers) (record_full_beneath_to_xfer_partial_ops) (record_full_beneath_to_xfer_partial) (record_full_beneath_to_insert_breakpoint_ops) (record_full_beneath_to_insert_breakpoint) (record_full_beneath_to_remove_breakpoint_ops) (record_full_beneath_to_remove_breakpoint) (record_full_beneath_to_stopped_by_watchpoint) (record_full_beneath_to_stopped_data_address) (record_full_beneath_to_async, tmp_to_resume_ops, tmp_to_resume) (tmp_to_wait_ops, tmp_to_wait, tmp_to_store_registers_ops) (tmp_to_store_registers, tmp_to_xfer_partial_ops) (tmp_to_xfer_partial, tmp_to_instmp_to_insert_breakpoint_ops) (tmp_to_insert_breakpoint, tmp_to_remove_breakpoint_ops) (tmp_to_remove_breakpoint, tmp_to_stopped_by_watchpoint) (tmp_to_stopped_data_address, tmp_to_async): Remove. (record_full_open_1, record_full_open): Update. Use RECORD_IS_USED. (record_full_resume, record_full_wait_1) (record_full_stopped_by_watchpoint, record_full_stopped_data_address) (record_full_store_registers, record_full_xfer_partial) (record_full_insert_breakpoint, record_full_remove_breakpoint) (record_full_async, record_full_core_xfer_partial): Use target delegation. * target-delegates.c: Rebuild. * target.c (current_xfer_partial): Remove. (update_current_target): Do not INHERIT or de_fault to_insert_breakpoint, to_remove_breakpoint, to_stopped_data_address, to_stopped_by_watchpoint, to_can_async_p, to_is_async_p, to_async. Do not set to_xfer_partial field. (default_xfer_partial): Simplify. (current_xfer_partial): Remove. (target_wait, target_resume): Simplify. (find_default_can_async_p, find_default_is_async_p): Update. (init_dummy_target): Don't set to_can_async_p, to_is_async_p, to_xfer_partial, to_stopped_by_watchpoint, to_stopped_data_address. (target_store_registers): Simplify. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. (target_remove_breakpoint, target_insert_breakpoint) (debug_to_insert_breakpoint, debug_to_remove_breakpoint): Update. * target.h (struct target_ops) <to_resume, to_wait, to_store_registers, to_insert_breakpoint, to_remove_breakpoint, to_stopped_by_watchpoint, to_stopped_data_address, to_can_async_p, to_is_async_p, to_async, to_xfer_partial>: Add TARGET_DEFAULT markup. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. * record-btrace.c (record_btrace_remove_breakpoint): Delegate directly. (record_btrace_insert_breakpoint): Delegate directly.
2013-07-13 03:00:34 +08:00
if (ops->to_stopped_by_watchpoint == NULL)
ops->to_stopped_by_watchpoint = delegate_stopped_by_watchpoint;
if (ops->to_stopped_data_address == NULL)
ops->to_stopped_data_address = delegate_stopped_data_address;
if (ops->to_watchpoint_addr_within_range == NULL)
ops->to_watchpoint_addr_within_range = delegate_watchpoint_addr_within_range;
if (ops->to_region_ok_for_hw_watchpoint == NULL)
ops->to_region_ok_for_hw_watchpoint = delegate_region_ok_for_hw_watchpoint;
if (ops->to_can_accel_watchpoint_condition == NULL)
ops->to_can_accel_watchpoint_condition = delegate_can_accel_watchpoint_condition;
if (ops->to_masked_watch_num_registers == NULL)
ops->to_masked_watch_num_registers = delegate_masked_watch_num_registers;
if (ops->to_terminal_init == NULL)
ops->to_terminal_init = delegate_terminal_init;
if (ops->to_terminal_inferior == NULL)
ops->to_terminal_inferior = delegate_terminal_inferior;
if (ops->to_terminal_ours_for_output == NULL)
ops->to_terminal_ours_for_output = delegate_terminal_ours_for_output;
if (ops->to_terminal_ours == NULL)
ops->to_terminal_ours = delegate_terminal_ours;
if (ops->to_terminal_info == NULL)
ops->to_terminal_info = delegate_terminal_info;
if (ops->to_kill == NULL)
ops->to_kill = delegate_kill;
if (ops->to_load == NULL)
ops->to_load = delegate_load;
if (ops->to_post_startup_inferior == NULL)
ops->to_post_startup_inferior = delegate_post_startup_inferior;
if (ops->to_insert_fork_catchpoint == NULL)
ops->to_insert_fork_catchpoint = delegate_insert_fork_catchpoint;
if (ops->to_remove_fork_catchpoint == NULL)
ops->to_remove_fork_catchpoint = delegate_remove_fork_catchpoint;
if (ops->to_insert_vfork_catchpoint == NULL)
ops->to_insert_vfork_catchpoint = delegate_insert_vfork_catchpoint;
if (ops->to_remove_vfork_catchpoint == NULL)
ops->to_remove_vfork_catchpoint = delegate_remove_vfork_catchpoint;
if (ops->to_follow_fork == NULL)
ops->to_follow_fork = delegate_follow_fork;
if (ops->to_insert_exec_catchpoint == NULL)
ops->to_insert_exec_catchpoint = delegate_insert_exec_catchpoint;
if (ops->to_remove_exec_catchpoint == NULL)
ops->to_remove_exec_catchpoint = delegate_remove_exec_catchpoint;
if (ops->to_set_syscall_catchpoint == NULL)
ops->to_set_syscall_catchpoint = delegate_set_syscall_catchpoint;
if (ops->to_has_exited == NULL)
ops->to_has_exited = delegate_has_exited;
if (ops->to_mourn_inferior == NULL)
ops->to_mourn_inferior = delegate_mourn_inferior;
if (ops->to_can_run == NULL)
ops->to_can_run = delegate_can_run;
if (ops->to_pass_signals == NULL)
ops->to_pass_signals = delegate_pass_signals;
if (ops->to_program_signals == NULL)
ops->to_program_signals = delegate_program_signals;
if (ops->to_thread_alive == NULL)
ops->to_thread_alive = delegate_thread_alive;
Push pruning old threads down to the target When GDB wants to sync the thread list with the target's (e.g., due to "info threads"), it calls update_thread_list: update_thread_list (void) { prune_threads (); target_find_new_threads (); update_threads_executing (); } And then prune_threads does: prune_threads (void) { struct thread_info *tp, *next; for (tp = thread_list; tp; tp = next) { next = tp->next; if (!thread_alive (tp)) delete_thread (tp->ptid); } } Calling thread_live on each thread one by one is expensive. E.g., on Linux, it ends up doing kill(SIG0) once for each thread. Not a big deal, but still a bunch of syscalls... With the remote target, it's cumbersome. That thread_alive call ends up generating one T packet per thread: Sending packet: $Tp2141.2150#82...Packet received: OK Sending packet: $Tp2141.214f#b7...Packet received: OK Sending packet: $Tp2141.2141#82...Packet received: OK Sending packet: $qXfer:threads:read::0,fff#03...Packet received: l<threads>\n<thread id="p2141.2141" core="2"/>\n<thread id="p2141.214f" core="1"/>\n<thread id="p2141.2150" core="2"/>\n</threads>\n That seems a bit silly when target_find_new_threads method implementations will always fetch the whole current set of target threads, and then add those that are not in GDB's thread list, to GDB's thread list. This patch thus pushes down the responsibility of pruning dead threads to the target_find_new_threads method instead, so a target may implement pruning dead threads however it wants. Once we do that, target_find_new_threads becomes a misnomer, so the patch renames it to target_update_thread_list. The patch doesn't attempt to do any optimization to any target yet. It simply exports prune_threads, and makes all implementations of target_update_thread_list call that. It's meant to be a no-op. gdb/ 2014-10-15 Pedro Alves <palves@redhat.com> * ada-tasks.c (print_ada_task_info, task_command_1): Adjust. * bsd-uthread.c (bsd_uthread_find_new_threads): Rename to ... (bsd_uthread_update_thread_list): ... this. Call prune_threads. (bsd_uthread_target): Adjust. * corelow.c (core_open): Adjust. * dec-thread.c (dec_thread_find_new_threads): Update comment. (dec_thread_update_thread_list): New function. (init_dec_thread_ops): Adjust. * gdbthread.h (prune_threads): New declaration. * linux-thread-db.c (thread_db_find_new_threads): Rename to ... (thread_db_update_thread_list): ... this. Call prune_threads. (init_thread_db_ops): Adjust. * nto-procfs.c (procfs_find_new_threads): Rename to ... (procfs_update_thread_list): ... this. Call prune_threads. (procfs_attach, procfs_create_inferior, init_procfs_targets): Adjust. * obsd-nat.c (obsd_find_new_threads): Rename to ... (obsd_update_thread_list): ... this. Call prune_threads. (obsd_add_target): Adjust. * procfs.c (procfs_target): Adjust. (procfs_notice_thread): Update comment. (procfs_find_new_threads): Rename to ... (procfs_update_thread_list): ... this. Call prune_threads. * ravenscar-thread.c (ravenscar_update_inferior_ptid): Update comment. (ravenscar_wait): Adjust. (ravenscar_find_new_threads): Rename to ... (ravenscar_update_thread_list): ... this. Call prune_threads. (init_ravenscar_thread_ops): Adjust. * record-btrace.c (record_btrace_find_new_threads): Rename to ... (record_btrace_update_thread_list): ... this. Adjust comment. (init_record_btrace_ops): Adjust. * remote.c (remote_threads_info): Rename to ... (remote_update_thread_list): ... this. Call prune_threads. (remote_start_remote, extended_remote_attach_1, init_remote_ops): Adjust. * sol-thread.c (check_for_thread_db): Adjust. (sol_find_new_threads_callback): Rename to ... (sol_update_thread_list_callback): ... this. (sol_find_new_threads): Rename to ... (sol_update_thread_list): ... this. Call prune_threads. Adjust. (sol_get_ada_task_ptid, init_sol_thread_ops): Adjust. * target-delegates.c: Regenerate. * target.c (target_find_new_threads): Rename to ... (target_update_thread_list): ... this. * target.h (struct target_ops): Rename to_find_new_threads field to to_update_thread_list. (target_find_new_threads): Rename to ... (target_update_thread_list): ... this. * thread.c (prune_threads): Make extern. (update_thread_list): Adjust.
2014-10-16 05:44:00 +08:00
if (ops->to_update_thread_list == NULL)
ops->to_update_thread_list = delegate_update_thread_list;
if (ops->to_pid_to_str == NULL)
ops->to_pid_to_str = delegate_pid_to_str;
if (ops->to_extra_thread_info == NULL)
ops->to_extra_thread_info = delegate_extra_thread_info;
if (ops->to_thread_name == NULL)
ops->to_thread_name = delegate_thread_name;
if (ops->to_stop == NULL)
ops->to_stop = delegate_stop;
if (ops->to_rcmd == NULL)
ops->to_rcmd = delegate_rcmd;
if (ops->to_pid_to_exec_file == NULL)
ops->to_pid_to_exec_file = delegate_pid_to_exec_file;
if (ops->to_log_command == NULL)
ops->to_log_command = delegate_log_command;
if (ops->to_get_section_table == NULL)
ops->to_get_section_table = delegate_get_section_table;
add target method delegation This patch replaces some code in the record targets with target method delegation. record-full.c stores pointers to many target methods when the record target is pushed. Then it later delegates some calls via these. This is wrong because it violates the target stack contract. In particular it is ok to unpush a target at any stratum, but record-full does not keep track of this, so it could potentially call into an unpushed target. This patch fixes the problem by using the newly-introduced generic approach to target delegation for the methods in question. 2014-02-19 Tom Tromey <tromey@redhat.com> * record-full.c (record_full_beneath_to_resume_ops) (record_full_beneath_to_resume, record_full_beneath_to_wait_ops) (record_full_beneath_to_wait) (record_full_beneath_to_store_registers_ops) (record_full_beneath_to_store_registers) (record_full_beneath_to_xfer_partial_ops) (record_full_beneath_to_xfer_partial) (record_full_beneath_to_insert_breakpoint_ops) (record_full_beneath_to_insert_breakpoint) (record_full_beneath_to_remove_breakpoint_ops) (record_full_beneath_to_remove_breakpoint) (record_full_beneath_to_stopped_by_watchpoint) (record_full_beneath_to_stopped_data_address) (record_full_beneath_to_async, tmp_to_resume_ops, tmp_to_resume) (tmp_to_wait_ops, tmp_to_wait, tmp_to_store_registers_ops) (tmp_to_store_registers, tmp_to_xfer_partial_ops) (tmp_to_xfer_partial, tmp_to_instmp_to_insert_breakpoint_ops) (tmp_to_insert_breakpoint, tmp_to_remove_breakpoint_ops) (tmp_to_remove_breakpoint, tmp_to_stopped_by_watchpoint) (tmp_to_stopped_data_address, tmp_to_async): Remove. (record_full_open_1, record_full_open): Update. Use RECORD_IS_USED. (record_full_resume, record_full_wait_1) (record_full_stopped_by_watchpoint, record_full_stopped_data_address) (record_full_store_registers, record_full_xfer_partial) (record_full_insert_breakpoint, record_full_remove_breakpoint) (record_full_async, record_full_core_xfer_partial): Use target delegation. * target-delegates.c: Rebuild. * target.c (current_xfer_partial): Remove. (update_current_target): Do not INHERIT or de_fault to_insert_breakpoint, to_remove_breakpoint, to_stopped_data_address, to_stopped_by_watchpoint, to_can_async_p, to_is_async_p, to_async. Do not set to_xfer_partial field. (default_xfer_partial): Simplify. (current_xfer_partial): Remove. (target_wait, target_resume): Simplify. (find_default_can_async_p, find_default_is_async_p): Update. (init_dummy_target): Don't set to_can_async_p, to_is_async_p, to_xfer_partial, to_stopped_by_watchpoint, to_stopped_data_address. (target_store_registers): Simplify. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. (target_remove_breakpoint, target_insert_breakpoint) (debug_to_insert_breakpoint, debug_to_remove_breakpoint): Update. * target.h (struct target_ops) <to_resume, to_wait, to_store_registers, to_insert_breakpoint, to_remove_breakpoint, to_stopped_by_watchpoint, to_stopped_data_address, to_can_async_p, to_is_async_p, to_async, to_xfer_partial>: Add TARGET_DEFAULT markup. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. * record-btrace.c (record_btrace_remove_breakpoint): Delegate directly. (record_btrace_insert_breakpoint): Delegate directly.
2013-07-13 03:00:34 +08:00
if (ops->to_can_async_p == NULL)
ops->to_can_async_p = delegate_can_async_p;
if (ops->to_is_async_p == NULL)
ops->to_is_async_p = delegate_is_async_p;
if (ops->to_async == NULL)
ops->to_async = delegate_async;
fix regressions with target-async A patch in the target cleanup series caused a regression when using record with target-async. Version 4 of the patch is here: https://sourceware.org/ml/gdb-patches/2014-03/msg00159.html The immediate problem is that record supplies to_can_async_p and to_is_async_p methods, but does not supply a to_async method. So, when target-async is set, record claims to support async -- but if the underlying target does not support async, then the to_async method call will end up in that method's default implementation, namely tcomplain. This worked previously because the record target used to provide a to_async method; one that (erroneously, only at push time) checked the other members of the target stack, and then simply dropped to_async calls in the "does not implement async" case. My first thought was to simply drop tcomplain as the default for to_async. This works, but Pedro pointed out that the only reason record has to supply to_can_async_p and to_is_async_p is that these default to using the find_default_run_target machinery -- and these defaults are only needed by "run" and "attach". So, a nicer solution presents itself: change run and attach to explicitly call into the default run target when needed; and change to_is_async_p and to_can_async_p to default to "return 0". This makes the target stack simpler to use and lets us remove the method implementations from record. This is also in harmony with other plans for the target stack; namely trying to reduce the impact of find_default_run_target. This approach makes it clear that find_default_is_async_p is not needed -- it is asking whether a target that may not even be pushed is actually async, which seems like a nonsensical question. While an improvement, this approach proved to introduce the same bug when using the core target. Looking a bit deeper, the issue is that code in "attach" and "run" may need to use either the current target stack or the default run target -- but different calls into the target API in those functions could wind up querying different targets. This new patch makes the target to use more explicit in "run" and "attach". Then these commands explicitly make the needed calls against that target. This ensures that a single target is used for all relevant operations. This lets us remove a couple find_default_* functions from various targets, including the dummy target. I think this is a decent understandability improvement. One issue I see with this patch is that the new calls in "run" and "attach" are not very much like the rest of the target API. I think fundamentally this is due to bad factoring in the target API, which may need to be fixed for multi-target. Tackling that seemed ambitious for a regression fix. While working on this I noticed that there don't seem to be any test cases that involve both target-async and record, so this patch changes break-precsave.exp to add some. It also changes corefile.exp to add some target-async tests; these pass with current trunk and with this patch applied, but fail with the v1 patch. This patch differs from v4 in that it moves initialization of to_can_async_p and to_supports_non_stop into inf-child, adds some assertions to complete_target_initialization, and adds some comments to target.h. Built and regtested on x86-64 Fedora 20. 2014-03-12 Tom Tromey <tromey@redhat.com> * inf-child.c (return_zero): New function. (inf_child_target): Set to_can_async_p, to_supports_non_stop. * aix-thread.c (aix_thread_inferior_created): New function. (aix_thread_attach): Remove. (init_aix_thread_ops): Don't set to_attach. (_initialize_aix_thread): Register inferior_created observer. * corelow.c (init_core_ops): Don't set to_attach or to_create_inferior. * exec.c (init_exec_ops): Don't set to_attach or to_create_inferior. * infcmd.c (run_command_1): Use find_run_target. Make direct target calls. (attach_command): Use find_attach_target. Make direct target calls. * record-btrace.c (init_record_btrace_ops): Don't set to_create_inferior. * record-full.c (record_full_can_async_p, record_full_is_async_p): Remove. (init_record_full_ops, init_record_full_core_ops): Update. Don't set to_create_inferior. * target.c (complete_target_initialization): Add assertion. (target_create_inferior): Remove. (find_default_attach, find_default_create_inferior): Remove. (find_attach_target, find_run_target): New functions. (find_default_is_async_p, find_default_can_async_p) (target_supports_non_stop, target_attach): Remove. (init_dummy_target): Don't set to_create_inferior or to_supports_non_stop. * target.h (struct target_ops) <to_attach>: Add comment. Remove TARGET_DEFAULT_FUNC. <to_create_inferior>: Add comment. <to_can_async_p, to_is_async_p, to_supports_non_stop>: Use TARGET_DEFAULT_RETURN. <to_can_async_p, to_supports_non_stop, to_can_run>: Add comments. (find_attach_target, find_run_target): Declare. (target_create_inferior): Remove. (target_has_execution_1): Update comment. (target_supports_non_stop): Remove. * target-delegates.c: Rebuild. 2014-03-12 Tom Tromey <tromey@redhat.com> * gdb.base/corefile.exp (corefile_test_run, corefile_test_attach): New procs. Add target-async tests. * gdb.reverse/break-precsave.exp (precsave_tests): New proc. Add target-async tests.
2014-03-01 00:47:34 +08:00
if (ops->to_supports_non_stop == NULL)
ops->to_supports_non_stop = delegate_supports_non_stop;
if (ops->to_find_memory_regions == NULL)
ops->to_find_memory_regions = delegate_find_memory_regions;
if (ops->to_make_corefile_notes == NULL)
ops->to_make_corefile_notes = delegate_make_corefile_notes;
if (ops->to_get_bookmark == NULL)
ops->to_get_bookmark = delegate_get_bookmark;
if (ops->to_goto_bookmark == NULL)
ops->to_goto_bookmark = delegate_goto_bookmark;
if (ops->to_get_thread_local_address == NULL)
ops->to_get_thread_local_address = delegate_get_thread_local_address;
add target method delegation This patch replaces some code in the record targets with target method delegation. record-full.c stores pointers to many target methods when the record target is pushed. Then it later delegates some calls via these. This is wrong because it violates the target stack contract. In particular it is ok to unpush a target at any stratum, but record-full does not keep track of this, so it could potentially call into an unpushed target. This patch fixes the problem by using the newly-introduced generic approach to target delegation for the methods in question. 2014-02-19 Tom Tromey <tromey@redhat.com> * record-full.c (record_full_beneath_to_resume_ops) (record_full_beneath_to_resume, record_full_beneath_to_wait_ops) (record_full_beneath_to_wait) (record_full_beneath_to_store_registers_ops) (record_full_beneath_to_store_registers) (record_full_beneath_to_xfer_partial_ops) (record_full_beneath_to_xfer_partial) (record_full_beneath_to_insert_breakpoint_ops) (record_full_beneath_to_insert_breakpoint) (record_full_beneath_to_remove_breakpoint_ops) (record_full_beneath_to_remove_breakpoint) (record_full_beneath_to_stopped_by_watchpoint) (record_full_beneath_to_stopped_data_address) (record_full_beneath_to_async, tmp_to_resume_ops, tmp_to_resume) (tmp_to_wait_ops, tmp_to_wait, tmp_to_store_registers_ops) (tmp_to_store_registers, tmp_to_xfer_partial_ops) (tmp_to_xfer_partial, tmp_to_instmp_to_insert_breakpoint_ops) (tmp_to_insert_breakpoint, tmp_to_remove_breakpoint_ops) (tmp_to_remove_breakpoint, tmp_to_stopped_by_watchpoint) (tmp_to_stopped_data_address, tmp_to_async): Remove. (record_full_open_1, record_full_open): Update. Use RECORD_IS_USED. (record_full_resume, record_full_wait_1) (record_full_stopped_by_watchpoint, record_full_stopped_data_address) (record_full_store_registers, record_full_xfer_partial) (record_full_insert_breakpoint, record_full_remove_breakpoint) (record_full_async, record_full_core_xfer_partial): Use target delegation. * target-delegates.c: Rebuild. * target.c (current_xfer_partial): Remove. (update_current_target): Do not INHERIT or de_fault to_insert_breakpoint, to_remove_breakpoint, to_stopped_data_address, to_stopped_by_watchpoint, to_can_async_p, to_is_async_p, to_async. Do not set to_xfer_partial field. (default_xfer_partial): Simplify. (current_xfer_partial): Remove. (target_wait, target_resume): Simplify. (find_default_can_async_p, find_default_is_async_p): Update. (init_dummy_target): Don't set to_can_async_p, to_is_async_p, to_xfer_partial, to_stopped_by_watchpoint, to_stopped_data_address. (target_store_registers): Simplify. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. (target_remove_breakpoint, target_insert_breakpoint) (debug_to_insert_breakpoint, debug_to_remove_breakpoint): Update. * target.h (struct target_ops) <to_resume, to_wait, to_store_registers, to_insert_breakpoint, to_remove_breakpoint, to_stopped_by_watchpoint, to_stopped_data_address, to_can_async_p, to_is_async_p, to_async, to_xfer_partial>: Add TARGET_DEFAULT markup. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. * record-btrace.c (record_btrace_remove_breakpoint): Delegate directly. (record_btrace_insert_breakpoint): Delegate directly.
2013-07-13 03:00:34 +08:00
if (ops->to_xfer_partial == NULL)
ops->to_xfer_partial = delegate_xfer_partial;
if (ops->to_memory_map == NULL)
ops->to_memory_map = delegate_memory_map;
if (ops->to_flash_erase == NULL)
ops->to_flash_erase = delegate_flash_erase;
if (ops->to_flash_done == NULL)
ops->to_flash_done = delegate_flash_done;
if (ops->to_read_description == NULL)
ops->to_read_description = delegate_read_description;
if (ops->to_get_ada_task_ptid == NULL)
ops->to_get_ada_task_ptid = delegate_get_ada_task_ptid;
if (ops->to_auxv_parse == NULL)
ops->to_auxv_parse = delegate_auxv_parse;
if (ops->to_search_memory == NULL)
ops->to_search_memory = delegate_search_memory;
if (ops->to_can_execute_reverse == NULL)
ops->to_can_execute_reverse = delegate_can_execute_reverse;
if (ops->to_execution_direction == NULL)
ops->to_execution_direction = delegate_execution_direction;
if (ops->to_supports_multi_process == NULL)
ops->to_supports_multi_process = delegate_supports_multi_process;
if (ops->to_supports_enable_disable_tracepoint == NULL)
ops->to_supports_enable_disable_tracepoint = delegate_supports_enable_disable_tracepoint;
if (ops->to_supports_string_tracing == NULL)
ops->to_supports_string_tracing = delegate_supports_string_tracing;
if (ops->to_supports_evaluation_of_breakpoint_conditions == NULL)
ops->to_supports_evaluation_of_breakpoint_conditions = delegate_supports_evaluation_of_breakpoint_conditions;
if (ops->to_can_run_breakpoint_commands == NULL)
ops->to_can_run_breakpoint_commands = delegate_can_run_breakpoint_commands;
if (ops->to_thread_architecture == NULL)
ops->to_thread_architecture = delegate_thread_architecture;
if (ops->to_thread_address_space == NULL)
ops->to_thread_address_space = delegate_thread_address_space;
if (ops->to_filesystem_is_local == NULL)
ops->to_filesystem_is_local = delegate_filesystem_is_local;
if (ops->to_trace_init == NULL)
ops->to_trace_init = delegate_trace_init;
if (ops->to_download_tracepoint == NULL)
ops->to_download_tracepoint = delegate_download_tracepoint;
if (ops->to_can_download_tracepoint == NULL)
ops->to_can_download_tracepoint = delegate_can_download_tracepoint;
if (ops->to_download_trace_state_variable == NULL)
ops->to_download_trace_state_variable = delegate_download_trace_state_variable;
if (ops->to_enable_tracepoint == NULL)
ops->to_enable_tracepoint = delegate_enable_tracepoint;
if (ops->to_disable_tracepoint == NULL)
ops->to_disable_tracepoint = delegate_disable_tracepoint;
if (ops->to_trace_set_readonly_regions == NULL)
ops->to_trace_set_readonly_regions = delegate_trace_set_readonly_regions;
if (ops->to_trace_start == NULL)
ops->to_trace_start = delegate_trace_start;
if (ops->to_get_trace_status == NULL)
ops->to_get_trace_status = delegate_get_trace_status;
if (ops->to_get_tracepoint_status == NULL)
ops->to_get_tracepoint_status = delegate_get_tracepoint_status;
if (ops->to_trace_stop == NULL)
ops->to_trace_stop = delegate_trace_stop;
if (ops->to_trace_find == NULL)
ops->to_trace_find = delegate_trace_find;
if (ops->to_get_trace_state_variable_value == NULL)
ops->to_get_trace_state_variable_value = delegate_get_trace_state_variable_value;
if (ops->to_save_trace_data == NULL)
ops->to_save_trace_data = delegate_save_trace_data;
if (ops->to_upload_tracepoints == NULL)
ops->to_upload_tracepoints = delegate_upload_tracepoints;
if (ops->to_upload_trace_state_variables == NULL)
ops->to_upload_trace_state_variables = delegate_upload_trace_state_variables;
if (ops->to_get_raw_trace_data == NULL)
ops->to_get_raw_trace_data = delegate_get_raw_trace_data;
if (ops->to_get_min_fast_tracepoint_insn_len == NULL)
ops->to_get_min_fast_tracepoint_insn_len = delegate_get_min_fast_tracepoint_insn_len;
if (ops->to_set_disconnected_tracing == NULL)
ops->to_set_disconnected_tracing = delegate_set_disconnected_tracing;
if (ops->to_set_circular_trace_buffer == NULL)
ops->to_set_circular_trace_buffer = delegate_set_circular_trace_buffer;
if (ops->to_set_trace_buffer_size == NULL)
ops->to_set_trace_buffer_size = delegate_set_trace_buffer_size;
if (ops->to_set_trace_notes == NULL)
ops->to_set_trace_notes = delegate_set_trace_notes;
if (ops->to_core_of_thread == NULL)
ops->to_core_of_thread = delegate_core_of_thread;
if (ops->to_verify_memory == NULL)
ops->to_verify_memory = delegate_verify_memory;
if (ops->to_get_tib_address == NULL)
ops->to_get_tib_address = delegate_get_tib_address;
if (ops->to_set_permissions == NULL)
ops->to_set_permissions = delegate_set_permissions;
if (ops->to_static_tracepoint_marker_at == NULL)
ops->to_static_tracepoint_marker_at = delegate_static_tracepoint_marker_at;
if (ops->to_static_tracepoint_markers_by_strid == NULL)
ops->to_static_tracepoint_markers_by_strid = delegate_static_tracepoint_markers_by_strid;
if (ops->to_traceframe_info == NULL)
ops->to_traceframe_info = delegate_traceframe_info;
if (ops->to_use_agent == NULL)
ops->to_use_agent = delegate_use_agent;
if (ops->to_can_use_agent == NULL)
ops->to_can_use_agent = delegate_can_use_agent;
if (ops->to_supports_btrace == NULL)
ops->to_supports_btrace = delegate_supports_btrace;
if (ops->to_enable_btrace == NULL)
ops->to_enable_btrace = delegate_enable_btrace;
if (ops->to_disable_btrace == NULL)
ops->to_disable_btrace = delegate_disable_btrace;
if (ops->to_teardown_btrace == NULL)
ops->to_teardown_btrace = delegate_teardown_btrace;
if (ops->to_read_btrace == NULL)
ops->to_read_btrace = delegate_read_btrace;
record btrace: add configuration struct Add a struct to describe the branch trace configuration and use it for enabling branch tracing. The user will be able to set configuration fields for each tracing format to be used for new threads. The actual configuration that is active for a given thread will be shown in the "info record" command. At the moment, the configuration struct only contains a format field that is set to the only available format. The format is the only configuration option that can not be set via set commands. It is given as argument to the "record btrace" command when starting recording. 2015-02-09 Markus Metzger <markus.t.metzger@intel.com> * Makefile.in (XMLFILES): Add btrace-conf.dtd. * x86-linux-nat.c (x86_linux_enable_btrace): Update parameters. (x86_linux_btrace_conf): New. (x86_linux_create_target): Initialize to_btrace_conf. * nat/linux-btrace.c (linux_enable_btrace): Update parameters. Check format. Split into this and ... (linux_enable_bts): ... this. (linux_btrace_conf): New. (perf_event_skip_record): Renamed into ... (perf_event_skip_bts_record): ... this. Updated users. (linux_disable_btrace): Split into this and ... (linux_disable_bts): ... this. (linux_read_btrace): Check format. * nat/linux-btrace.h (linux_enable_btrace): Update parameters. (linux_btrace_conf): New. (btrace_target_info)<ptid>: Moved. (btrace_target_info)<conf>: New. (btrace_target_info): Split into this and ... (btrace_tinfo_bts): ... this. Updated users. * btrace.c (btrace_enable): Update parameters. (btrace_conf, parse_xml_btrace_conf_bts, parse_xml_btrace_conf) (btrace_conf_children, btrace_conf_attributes) (btrace_conf_elements): New. * btrace.h (btrace_enable): Update parameters. (btrace_conf, parse_xml_btrace_conf): New. * common/btrace-common.h (btrace_config): New. * feature/btrace-conf.dtd: New. * record-btrace.c (record_btrace_conf): New. (record_btrace_cmdlist): New. (record_btrace_enable_warn, record_btrace_open): Pass &record_btrace_conf. (record_btrace_info): Print recording format. (cmd_record_btrace_bts_start): New. (cmd_record_btrace_start): Call cmd_record_btrace_bts_start. (_initialize_record_btrace): Add "record btrace bts" subcommand. Add "record bts" alias command. * remote.c (remote_state)<btrace_config>: New. (remote_btrace_reset, PACKET_qXfer_btrace_conf): New. (remote_protocol_features): Add qXfer:btrace-conf:read. (remote_open_1): Call remote_btrace_reset. (remote_xfer_partial): Handle TARGET_OBJECT_BTRACE_CONF. (btrace_target_info)<conf>: New. (btrace_sync_conf, btrace_read_config): New. (remote_enable_btrace): Update parameters. Call btrace_sync_conf and btrace_read_conf. (remote_btrace_conf): New. (init_remote_ops): Initialize to_btrace_conf. (_initialize_remote): Add qXfer:btrace-conf packet. * target.c (target_enable_btrace): Update parameters. (target_btrace_conf): New. * target.h (target_enable_btrace): Update parameters. (target_btrace_conf): New. (target_object)<TARGET_OBJECT_BTRACE_CONF>: New. (target_ops)<to_enable_btrace>: Update parameters and comment. (target_ops)<to_btrace_conf>: New. * target-delegates: Regenerate. * target-debug.h (target_debug_print_const_struct_btrace_config_p) (target_debug_print_const_struct_btrace_target_info_p): New. NEWS: Announce new command and new packet. doc/ * gdb.texinfo (Process Record and Replay): Describe the "record btrace bts" command. (General Query Packets): Describe qXfer:btrace-conf:read packet. (Branch Trace Configuration Format): New. gdbserver/ * linux-low.c (linux_low_enable_btrace): Update parameters. (linux_low_btrace_conf): New. (linux_target_ops)<to_btrace_conf>: Initialize. * server.c (current_btrace_conf): New. (handle_btrace_enable): Rename to ... (handle_btrace_enable_bts): ... this. Pass &current_btrace_conf to target_enable_btrace. Update comment. Update users. (handle_qxfer_btrace_conf): New. (qxfer_packets): Add btrace-conf entry. (handle_query): Report qXfer:btrace-conf:read as supported packet. * target.h (target_ops)<enable_btrace>: Update parameters and comment. (target_ops)<read_btrace_conf>: New. (target_enable_btrace): Update parameters. (target_read_btrace_conf): New. testsuite/ * gdb.btrace/delta.exp: Update "info record" output. * gdb.btrace/enable.exp: Update "info record" output. * gdb.btrace/finish.exp: Update "info record" output. * gdb.btrace/instruction_history.exp: Update "info record" output. * gdb.btrace/next.exp: Update "info record" output. * gdb.btrace/nexti.exp: Update "info record" output. * gdb.btrace/step.exp: Update "info record" output. * gdb.btrace/stepi.exp: Update "info record" output. * gdb.btrace/nohist.exp: Update "info record" output.
2013-11-28 22:44:13 +08:00
if (ops->to_btrace_conf == NULL)
ops->to_btrace_conf = delegate_btrace_conf;
if (ops->to_stop_recording == NULL)
ops->to_stop_recording = delegate_stop_recording;
if (ops->to_info_record == NULL)
ops->to_info_record = delegate_info_record;
if (ops->to_save_record == NULL)
ops->to_save_record = delegate_save_record;
if (ops->to_delete_record == NULL)
ops->to_delete_record = delegate_delete_record;
if (ops->to_record_is_replaying == NULL)
ops->to_record_is_replaying = delegate_record_is_replaying;
if (ops->to_goto_record_begin == NULL)
ops->to_goto_record_begin = delegate_goto_record_begin;
if (ops->to_goto_record_end == NULL)
ops->to_goto_record_end = delegate_goto_record_end;
if (ops->to_goto_record == NULL)
ops->to_goto_record = delegate_goto_record;
if (ops->to_insn_history == NULL)
ops->to_insn_history = delegate_insn_history;
if (ops->to_insn_history_from == NULL)
ops->to_insn_history_from = delegate_insn_history_from;
if (ops->to_insn_history_range == NULL)
ops->to_insn_history_range = delegate_insn_history_range;
if (ops->to_call_history == NULL)
ops->to_call_history = delegate_call_history;
if (ops->to_call_history_from == NULL)
ops->to_call_history_from = delegate_call_history_from;
if (ops->to_call_history_range == NULL)
ops->to_call_history_range = delegate_call_history_range;
if (ops->to_augmented_libraries_svr4_read == NULL)
ops->to_augmented_libraries_svr4_read = delegate_augmented_libraries_svr4_read;
if (ops->to_get_unwinder == NULL)
ops->to_get_unwinder = delegate_get_unwinder;
if (ops->to_get_tailcall_unwinder == NULL)
ops->to_get_tailcall_unwinder = delegate_get_tailcall_unwinder;
if (ops->to_prepare_to_generate_core == NULL)
ops->to_prepare_to_generate_core = delegate_prepare_to_generate_core;
if (ops->to_done_generating_core == NULL)
ops->to_done_generating_core = delegate_done_generating_core;
}
static void
install_dummy_methods (struct target_ops *ops)
{
ops->to_post_attach = tdefault_post_attach;
ops->to_detach = tdefault_detach;
ops->to_disconnect = tdefault_disconnect;
add target method delegation This patch replaces some code in the record targets with target method delegation. record-full.c stores pointers to many target methods when the record target is pushed. Then it later delegates some calls via these. This is wrong because it violates the target stack contract. In particular it is ok to unpush a target at any stratum, but record-full does not keep track of this, so it could potentially call into an unpushed target. This patch fixes the problem by using the newly-introduced generic approach to target delegation for the methods in question. 2014-02-19 Tom Tromey <tromey@redhat.com> * record-full.c (record_full_beneath_to_resume_ops) (record_full_beneath_to_resume, record_full_beneath_to_wait_ops) (record_full_beneath_to_wait) (record_full_beneath_to_store_registers_ops) (record_full_beneath_to_store_registers) (record_full_beneath_to_xfer_partial_ops) (record_full_beneath_to_xfer_partial) (record_full_beneath_to_insert_breakpoint_ops) (record_full_beneath_to_insert_breakpoint) (record_full_beneath_to_remove_breakpoint_ops) (record_full_beneath_to_remove_breakpoint) (record_full_beneath_to_stopped_by_watchpoint) (record_full_beneath_to_stopped_data_address) (record_full_beneath_to_async, tmp_to_resume_ops, tmp_to_resume) (tmp_to_wait_ops, tmp_to_wait, tmp_to_store_registers_ops) (tmp_to_store_registers, tmp_to_xfer_partial_ops) (tmp_to_xfer_partial, tmp_to_instmp_to_insert_breakpoint_ops) (tmp_to_insert_breakpoint, tmp_to_remove_breakpoint_ops) (tmp_to_remove_breakpoint, tmp_to_stopped_by_watchpoint) (tmp_to_stopped_data_address, tmp_to_async): Remove. (record_full_open_1, record_full_open): Update. Use RECORD_IS_USED. (record_full_resume, record_full_wait_1) (record_full_stopped_by_watchpoint, record_full_stopped_data_address) (record_full_store_registers, record_full_xfer_partial) (record_full_insert_breakpoint, record_full_remove_breakpoint) (record_full_async, record_full_core_xfer_partial): Use target delegation. * target-delegates.c: Rebuild. * target.c (current_xfer_partial): Remove. (update_current_target): Do not INHERIT or de_fault to_insert_breakpoint, to_remove_breakpoint, to_stopped_data_address, to_stopped_by_watchpoint, to_can_async_p, to_is_async_p, to_async. Do not set to_xfer_partial field. (default_xfer_partial): Simplify. (current_xfer_partial): Remove. (target_wait, target_resume): Simplify. (find_default_can_async_p, find_default_is_async_p): Update. (init_dummy_target): Don't set to_can_async_p, to_is_async_p, to_xfer_partial, to_stopped_by_watchpoint, to_stopped_data_address. (target_store_registers): Simplify. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. (target_remove_breakpoint, target_insert_breakpoint) (debug_to_insert_breakpoint, debug_to_remove_breakpoint): Update. * target.h (struct target_ops) <to_resume, to_wait, to_store_registers, to_insert_breakpoint, to_remove_breakpoint, to_stopped_by_watchpoint, to_stopped_data_address, to_can_async_p, to_is_async_p, to_async, to_xfer_partial>: Add TARGET_DEFAULT markup. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. * record-btrace.c (record_btrace_remove_breakpoint): Delegate directly. (record_btrace_insert_breakpoint): Delegate directly.
2013-07-13 03:00:34 +08:00
ops->to_resume = tdefault_resume;
ops->to_wait = tdefault_wait;
ops->to_fetch_registers = tdefault_fetch_registers;
add target method delegation This patch replaces some code in the record targets with target method delegation. record-full.c stores pointers to many target methods when the record target is pushed. Then it later delegates some calls via these. This is wrong because it violates the target stack contract. In particular it is ok to unpush a target at any stratum, but record-full does not keep track of this, so it could potentially call into an unpushed target. This patch fixes the problem by using the newly-introduced generic approach to target delegation for the methods in question. 2014-02-19 Tom Tromey <tromey@redhat.com> * record-full.c (record_full_beneath_to_resume_ops) (record_full_beneath_to_resume, record_full_beneath_to_wait_ops) (record_full_beneath_to_wait) (record_full_beneath_to_store_registers_ops) (record_full_beneath_to_store_registers) (record_full_beneath_to_xfer_partial_ops) (record_full_beneath_to_xfer_partial) (record_full_beneath_to_insert_breakpoint_ops) (record_full_beneath_to_insert_breakpoint) (record_full_beneath_to_remove_breakpoint_ops) (record_full_beneath_to_remove_breakpoint) (record_full_beneath_to_stopped_by_watchpoint) (record_full_beneath_to_stopped_data_address) (record_full_beneath_to_async, tmp_to_resume_ops, tmp_to_resume) (tmp_to_wait_ops, tmp_to_wait, tmp_to_store_registers_ops) (tmp_to_store_registers, tmp_to_xfer_partial_ops) (tmp_to_xfer_partial, tmp_to_instmp_to_insert_breakpoint_ops) (tmp_to_insert_breakpoint, tmp_to_remove_breakpoint_ops) (tmp_to_remove_breakpoint, tmp_to_stopped_by_watchpoint) (tmp_to_stopped_data_address, tmp_to_async): Remove. (record_full_open_1, record_full_open): Update. Use RECORD_IS_USED. (record_full_resume, record_full_wait_1) (record_full_stopped_by_watchpoint, record_full_stopped_data_address) (record_full_store_registers, record_full_xfer_partial) (record_full_insert_breakpoint, record_full_remove_breakpoint) (record_full_async, record_full_core_xfer_partial): Use target delegation. * target-delegates.c: Rebuild. * target.c (current_xfer_partial): Remove. (update_current_target): Do not INHERIT or de_fault to_insert_breakpoint, to_remove_breakpoint, to_stopped_data_address, to_stopped_by_watchpoint, to_can_async_p, to_is_async_p, to_async. Do not set to_xfer_partial field. (default_xfer_partial): Simplify. (current_xfer_partial): Remove. (target_wait, target_resume): Simplify. (find_default_can_async_p, find_default_is_async_p): Update. (init_dummy_target): Don't set to_can_async_p, to_is_async_p, to_xfer_partial, to_stopped_by_watchpoint, to_stopped_data_address. (target_store_registers): Simplify. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. (target_remove_breakpoint, target_insert_breakpoint) (debug_to_insert_breakpoint, debug_to_remove_breakpoint): Update. * target.h (struct target_ops) <to_resume, to_wait, to_store_registers, to_insert_breakpoint, to_remove_breakpoint, to_stopped_by_watchpoint, to_stopped_data_address, to_can_async_p, to_is_async_p, to_async, to_xfer_partial>: Add TARGET_DEFAULT markup. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. * record-btrace.c (record_btrace_remove_breakpoint): Delegate directly. (record_btrace_insert_breakpoint): Delegate directly.
2013-07-13 03:00:34 +08:00
ops->to_store_registers = tdefault_store_registers;
ops->to_prepare_to_store = tdefault_prepare_to_store;
ops->to_files_info = tdefault_files_info;
add target method delegation This patch replaces some code in the record targets with target method delegation. record-full.c stores pointers to many target methods when the record target is pushed. Then it later delegates some calls via these. This is wrong because it violates the target stack contract. In particular it is ok to unpush a target at any stratum, but record-full does not keep track of this, so it could potentially call into an unpushed target. This patch fixes the problem by using the newly-introduced generic approach to target delegation for the methods in question. 2014-02-19 Tom Tromey <tromey@redhat.com> * record-full.c (record_full_beneath_to_resume_ops) (record_full_beneath_to_resume, record_full_beneath_to_wait_ops) (record_full_beneath_to_wait) (record_full_beneath_to_store_registers_ops) (record_full_beneath_to_store_registers) (record_full_beneath_to_xfer_partial_ops) (record_full_beneath_to_xfer_partial) (record_full_beneath_to_insert_breakpoint_ops) (record_full_beneath_to_insert_breakpoint) (record_full_beneath_to_remove_breakpoint_ops) (record_full_beneath_to_remove_breakpoint) (record_full_beneath_to_stopped_by_watchpoint) (record_full_beneath_to_stopped_data_address) (record_full_beneath_to_async, tmp_to_resume_ops, tmp_to_resume) (tmp_to_wait_ops, tmp_to_wait, tmp_to_store_registers_ops) (tmp_to_store_registers, tmp_to_xfer_partial_ops) (tmp_to_xfer_partial, tmp_to_instmp_to_insert_breakpoint_ops) (tmp_to_insert_breakpoint, tmp_to_remove_breakpoint_ops) (tmp_to_remove_breakpoint, tmp_to_stopped_by_watchpoint) (tmp_to_stopped_data_address, tmp_to_async): Remove. (record_full_open_1, record_full_open): Update. Use RECORD_IS_USED. (record_full_resume, record_full_wait_1) (record_full_stopped_by_watchpoint, record_full_stopped_data_address) (record_full_store_registers, record_full_xfer_partial) (record_full_insert_breakpoint, record_full_remove_breakpoint) (record_full_async, record_full_core_xfer_partial): Use target delegation. * target-delegates.c: Rebuild. * target.c (current_xfer_partial): Remove. (update_current_target): Do not INHERIT or de_fault to_insert_breakpoint, to_remove_breakpoint, to_stopped_data_address, to_stopped_by_watchpoint, to_can_async_p, to_is_async_p, to_async. Do not set to_xfer_partial field. (default_xfer_partial): Simplify. (current_xfer_partial): Remove. (target_wait, target_resume): Simplify. (find_default_can_async_p, find_default_is_async_p): Update. (init_dummy_target): Don't set to_can_async_p, to_is_async_p, to_xfer_partial, to_stopped_by_watchpoint, to_stopped_data_address. (target_store_registers): Simplify. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. (target_remove_breakpoint, target_insert_breakpoint) (debug_to_insert_breakpoint, debug_to_remove_breakpoint): Update. * target.h (struct target_ops) <to_resume, to_wait, to_store_registers, to_insert_breakpoint, to_remove_breakpoint, to_stopped_by_watchpoint, to_stopped_data_address, to_can_async_p, to_is_async_p, to_async, to_xfer_partial>: Add TARGET_DEFAULT markup. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. * record-btrace.c (record_btrace_remove_breakpoint): Delegate directly. (record_btrace_insert_breakpoint): Delegate directly.
2013-07-13 03:00:34 +08:00
ops->to_insert_breakpoint = memory_insert_breakpoint;
ops->to_remove_breakpoint = memory_remove_breakpoint;
Teach GDB about targets that can tell whether a trap is a breakpoint event The moribund locations heuristics are problematic. This patch teaches GDB about targets that can reliably tell whether a trap was caused by a software or hardware breakpoint, and thus don't need moribund locations, thus bypassing all the problems that mechanism has. The non-stop-fair-events.exp test is frequently failing currently. E.g., see https://sourceware.org/ml/gdb-testers/2015-q1/msg03148.html. The root cause is a fundamental problem with moribund locations. For example, the stepped_breakpoint logic added by af48d08f breaks in this case (which is what happens with that test): - Step thread A, no breakpoint is set at PC. - The kernel doesn't schedule thread A yet. - Insert breakpoint at A's PC, for some reason (e.g., a step-resume breakpoint for thread B). - Kernel finally schedules thread A. - thread A's stepped_breakpoint flag is not set, even though it now stepped a breakpoint instruction. - adjust_pc_after_break gets the PC wrong, because PC == PREV_PC, but stepped_breakpoint is not set. We needed the stepped_breakpoint logic to workaround moribund locations, because otherwise adjust_pc_after_break could apply an adjustment when it shouldn't just because there _used_ to be a breakpoint at PC (a moribund breakpoint location). For example, on x86, that's wrong if the thread really hasn't executed an int3, but instead executed some other 1-byte long instruction. Getting the PC adjustment wrong of course leads to the inferior executing the wrong instruction. Other problems with moribund locations are: - if a true SIGTRAP happens to be raised when the program is executing the PC that used to have a breakpoint, GDB will assume that is a trap for a breakpoint that has recently been removed, and thus we miss reporting the random signal to the user. - to minimize that, we get rid of moribund location after a while. That while is defined as just a certain number of events being processed. That number of events sometimes passes by before a delayed breakpoint is processed, and GDB confuses the trap for a random signal, thus reporting the random trap. Once the user resumes the thread, the program crashes because the PC was not adjusted... The fix for all this is to bite the bullet and get rid of heuristics and instead rely on the target knowing accurately what caused the SIGTRAP. The target/kernel/stub is in the best position to know what that, because it can e.g. consult priviledged CPU flags GDB has no access to, or by knowing which exception vector entry was called when the instruction trapped, etc. Most debug APIs I've seen to date report breakpoint hits as a distinct event in some fashion. For example, on the Linux kernel, whether a breakpoint was executed is exposed to userspace in the si_code field of the SIGTRAP's siginfo. On Windows, the debug API reports a EXCEPTION_BREAKPOINT exception code. We needed to keep around deleted breakpoints in an on-the-side list (the moribund locations) for two main reasons: - Know that a SIGTRAP actually is a delayed event for a hit of a breakpoint that was removed before the event was processed, and thus should not be reported as a random signal. - So we still do the decr_pc_after_break adjustment in that case, so that the thread is resumed at the correct address. In the new model, if GDB processes an event the target tells is a breakpoint trap, and GDB doesn't find the corresponding breakpoint in its breakpoint tables, it means that event is a delayed event for a breakpoint that has since been removed, and thus the event should be ignored. For the decr_pc_after_after issue, it ends up being much simpler that on targets that can reliably tell whether a breakpoint trapped, for the breakpoint trap to present the PC already adjusted. Proper multi-threading support already implies that targets needs to be doing decr_pc_after_break adjustment themselves, otherwise for example, in all-stop if two threads hit a breakpoint simultaneously, and the user does "info threads", he'll see the non-event thread that hit the breakpoint stopped at the wrong PC. This way (target adjusts) also ends up eliminating the need for some awkward re-incrementing of the PC in the record-full and Linux targets that we do today, and the need for the target_decr_pc_after_break hook. If the target always adjusts, then there's a case where GDB needs to re-increment the PC. Say, on x86, an "int3" instruction that was explicitly written in the program traps. In this case, GDB should report a random SIGTRAP signal to the user, with the PC pointing at the instruction past the int3, just like if GDB was not debugging the program. The user may well decide to pass the SIGTRAP to the program because the program being debugged has a SIGTRAP handler that handles its own breakpoints, and expects the PC to be unadjusted. Tested on x86-64 Fedora 20. gdb/ChangeLog: 2015-03-04 Pedro Alves <palves@redhat.com> * breakpoint.c (need_moribund_for_location_type): New function. (bpstat_stop_status): Don't skipping checking moribund locations of breakpoint types which the target tell caused a stop. (program_breakpoint_here_p): New function, factored out from ... (bp_loc_is_permanent): ... this. (update_global_location_list): Don't create a moribund location if the target supports reporting stops of the type of the removed breakpoint. * breakpoint.h (program_breakpoint_here_p): New declaration. * infrun.c (adjust_pc_after_break): Return early if the target has already adjusted the PC. Add comments. (handle_signal_stop): If nothing explains a signal, and the target tells us the stop was caused by a software breakpoint, check if there's a breakpoint instruction in the memory. If so, adjust the PC before presenting the stop to the user. Otherwise, ignore the trap. If nothing explains a signal, and the target tells us the stop was caused by a hardware breakpoint, ignore the trap. * target.h (struct target_ops) <to_stopped_by_sw_breakpoint, to_supports_stopped_by_sw_breakpoint, to_stopped_by_hw_breakpoint, to_supports_stopped_by_hw_breakpoint>: New fields. (target_stopped_by_sw_breakpoint) (target_supports_stopped_by_sw_breakpoint) (target_stopped_by_hw_breakpoint) (target_supports_stopped_by_hw_breakpoint): Define. * target-delegates.c: Regenerate.
2015-03-05 04:41:15 +08:00
ops->to_stopped_by_sw_breakpoint = tdefault_stopped_by_sw_breakpoint;
ops->to_supports_stopped_by_sw_breakpoint = tdefault_supports_stopped_by_sw_breakpoint;
ops->to_stopped_by_hw_breakpoint = tdefault_stopped_by_hw_breakpoint;
ops->to_supports_stopped_by_hw_breakpoint = tdefault_supports_stopped_by_hw_breakpoint;
ops->to_can_use_hw_breakpoint = tdefault_can_use_hw_breakpoint;
ops->to_ranged_break_num_registers = tdefault_ranged_break_num_registers;
ops->to_insert_hw_breakpoint = tdefault_insert_hw_breakpoint;
ops->to_remove_hw_breakpoint = tdefault_remove_hw_breakpoint;
ops->to_remove_watchpoint = tdefault_remove_watchpoint;
ops->to_insert_watchpoint = tdefault_insert_watchpoint;
ops->to_insert_mask_watchpoint = tdefault_insert_mask_watchpoint;
ops->to_remove_mask_watchpoint = tdefault_remove_mask_watchpoint;
add target method delegation This patch replaces some code in the record targets with target method delegation. record-full.c stores pointers to many target methods when the record target is pushed. Then it later delegates some calls via these. This is wrong because it violates the target stack contract. In particular it is ok to unpush a target at any stratum, but record-full does not keep track of this, so it could potentially call into an unpushed target. This patch fixes the problem by using the newly-introduced generic approach to target delegation for the methods in question. 2014-02-19 Tom Tromey <tromey@redhat.com> * record-full.c (record_full_beneath_to_resume_ops) (record_full_beneath_to_resume, record_full_beneath_to_wait_ops) (record_full_beneath_to_wait) (record_full_beneath_to_store_registers_ops) (record_full_beneath_to_store_registers) (record_full_beneath_to_xfer_partial_ops) (record_full_beneath_to_xfer_partial) (record_full_beneath_to_insert_breakpoint_ops) (record_full_beneath_to_insert_breakpoint) (record_full_beneath_to_remove_breakpoint_ops) (record_full_beneath_to_remove_breakpoint) (record_full_beneath_to_stopped_by_watchpoint) (record_full_beneath_to_stopped_data_address) (record_full_beneath_to_async, tmp_to_resume_ops, tmp_to_resume) (tmp_to_wait_ops, tmp_to_wait, tmp_to_store_registers_ops) (tmp_to_store_registers, tmp_to_xfer_partial_ops) (tmp_to_xfer_partial, tmp_to_instmp_to_insert_breakpoint_ops) (tmp_to_insert_breakpoint, tmp_to_remove_breakpoint_ops) (tmp_to_remove_breakpoint, tmp_to_stopped_by_watchpoint) (tmp_to_stopped_data_address, tmp_to_async): Remove. (record_full_open_1, record_full_open): Update. Use RECORD_IS_USED. (record_full_resume, record_full_wait_1) (record_full_stopped_by_watchpoint, record_full_stopped_data_address) (record_full_store_registers, record_full_xfer_partial) (record_full_insert_breakpoint, record_full_remove_breakpoint) (record_full_async, record_full_core_xfer_partial): Use target delegation. * target-delegates.c: Rebuild. * target.c (current_xfer_partial): Remove. (update_current_target): Do not INHERIT or de_fault to_insert_breakpoint, to_remove_breakpoint, to_stopped_data_address, to_stopped_by_watchpoint, to_can_async_p, to_is_async_p, to_async. Do not set to_xfer_partial field. (default_xfer_partial): Simplify. (current_xfer_partial): Remove. (target_wait, target_resume): Simplify. (find_default_can_async_p, find_default_is_async_p): Update. (init_dummy_target): Don't set to_can_async_p, to_is_async_p, to_xfer_partial, to_stopped_by_watchpoint, to_stopped_data_address. (target_store_registers): Simplify. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. (target_remove_breakpoint, target_insert_breakpoint) (debug_to_insert_breakpoint, debug_to_remove_breakpoint): Update. * target.h (struct target_ops) <to_resume, to_wait, to_store_registers, to_insert_breakpoint, to_remove_breakpoint, to_stopped_by_watchpoint, to_stopped_data_address, to_can_async_p, to_is_async_p, to_async, to_xfer_partial>: Add TARGET_DEFAULT markup. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. * record-btrace.c (record_btrace_remove_breakpoint): Delegate directly. (record_btrace_insert_breakpoint): Delegate directly.
2013-07-13 03:00:34 +08:00
ops->to_stopped_by_watchpoint = tdefault_stopped_by_watchpoint;
ops->to_stopped_data_address = tdefault_stopped_data_address;
ops->to_watchpoint_addr_within_range = default_watchpoint_addr_within_range;
ops->to_region_ok_for_hw_watchpoint = default_region_ok_for_hw_watchpoint;
ops->to_can_accel_watchpoint_condition = tdefault_can_accel_watchpoint_condition;
ops->to_masked_watch_num_registers = tdefault_masked_watch_num_registers;
ops->to_terminal_init = tdefault_terminal_init;
ops->to_terminal_inferior = tdefault_terminal_inferior;
ops->to_terminal_ours_for_output = tdefault_terminal_ours_for_output;
ops->to_terminal_ours = tdefault_terminal_ours;
ops->to_terminal_info = default_terminal_info;
ops->to_kill = tdefault_kill;
ops->to_load = tdefault_load;
ops->to_post_startup_inferior = tdefault_post_startup_inferior;
ops->to_insert_fork_catchpoint = tdefault_insert_fork_catchpoint;
ops->to_remove_fork_catchpoint = tdefault_remove_fork_catchpoint;
ops->to_insert_vfork_catchpoint = tdefault_insert_vfork_catchpoint;
ops->to_remove_vfork_catchpoint = tdefault_remove_vfork_catchpoint;
ops->to_follow_fork = default_follow_fork;
ops->to_insert_exec_catchpoint = tdefault_insert_exec_catchpoint;
ops->to_remove_exec_catchpoint = tdefault_remove_exec_catchpoint;
ops->to_set_syscall_catchpoint = tdefault_set_syscall_catchpoint;
ops->to_has_exited = tdefault_has_exited;
ops->to_mourn_inferior = default_mourn_inferior;
ops->to_can_run = tdefault_can_run;
ops->to_pass_signals = tdefault_pass_signals;
ops->to_program_signals = tdefault_program_signals;
ops->to_thread_alive = tdefault_thread_alive;
Push pruning old threads down to the target When GDB wants to sync the thread list with the target's (e.g., due to "info threads"), it calls update_thread_list: update_thread_list (void) { prune_threads (); target_find_new_threads (); update_threads_executing (); } And then prune_threads does: prune_threads (void) { struct thread_info *tp, *next; for (tp = thread_list; tp; tp = next) { next = tp->next; if (!thread_alive (tp)) delete_thread (tp->ptid); } } Calling thread_live on each thread one by one is expensive. E.g., on Linux, it ends up doing kill(SIG0) once for each thread. Not a big deal, but still a bunch of syscalls... With the remote target, it's cumbersome. That thread_alive call ends up generating one T packet per thread: Sending packet: $Tp2141.2150#82...Packet received: OK Sending packet: $Tp2141.214f#b7...Packet received: OK Sending packet: $Tp2141.2141#82...Packet received: OK Sending packet: $qXfer:threads:read::0,fff#03...Packet received: l<threads>\n<thread id="p2141.2141" core="2"/>\n<thread id="p2141.214f" core="1"/>\n<thread id="p2141.2150" core="2"/>\n</threads>\n That seems a bit silly when target_find_new_threads method implementations will always fetch the whole current set of target threads, and then add those that are not in GDB's thread list, to GDB's thread list. This patch thus pushes down the responsibility of pruning dead threads to the target_find_new_threads method instead, so a target may implement pruning dead threads however it wants. Once we do that, target_find_new_threads becomes a misnomer, so the patch renames it to target_update_thread_list. The patch doesn't attempt to do any optimization to any target yet. It simply exports prune_threads, and makes all implementations of target_update_thread_list call that. It's meant to be a no-op. gdb/ 2014-10-15 Pedro Alves <palves@redhat.com> * ada-tasks.c (print_ada_task_info, task_command_1): Adjust. * bsd-uthread.c (bsd_uthread_find_new_threads): Rename to ... (bsd_uthread_update_thread_list): ... this. Call prune_threads. (bsd_uthread_target): Adjust. * corelow.c (core_open): Adjust. * dec-thread.c (dec_thread_find_new_threads): Update comment. (dec_thread_update_thread_list): New function. (init_dec_thread_ops): Adjust. * gdbthread.h (prune_threads): New declaration. * linux-thread-db.c (thread_db_find_new_threads): Rename to ... (thread_db_update_thread_list): ... this. Call prune_threads. (init_thread_db_ops): Adjust. * nto-procfs.c (procfs_find_new_threads): Rename to ... (procfs_update_thread_list): ... this. Call prune_threads. (procfs_attach, procfs_create_inferior, init_procfs_targets): Adjust. * obsd-nat.c (obsd_find_new_threads): Rename to ... (obsd_update_thread_list): ... this. Call prune_threads. (obsd_add_target): Adjust. * procfs.c (procfs_target): Adjust. (procfs_notice_thread): Update comment. (procfs_find_new_threads): Rename to ... (procfs_update_thread_list): ... this. Call prune_threads. * ravenscar-thread.c (ravenscar_update_inferior_ptid): Update comment. (ravenscar_wait): Adjust. (ravenscar_find_new_threads): Rename to ... (ravenscar_update_thread_list): ... this. Call prune_threads. (init_ravenscar_thread_ops): Adjust. * record-btrace.c (record_btrace_find_new_threads): Rename to ... (record_btrace_update_thread_list): ... this. Adjust comment. (init_record_btrace_ops): Adjust. * remote.c (remote_threads_info): Rename to ... (remote_update_thread_list): ... this. Call prune_threads. (remote_start_remote, extended_remote_attach_1, init_remote_ops): Adjust. * sol-thread.c (check_for_thread_db): Adjust. (sol_find_new_threads_callback): Rename to ... (sol_update_thread_list_callback): ... this. (sol_find_new_threads): Rename to ... (sol_update_thread_list): ... this. Call prune_threads. Adjust. (sol_get_ada_task_ptid, init_sol_thread_ops): Adjust. * target-delegates.c: Regenerate. * target.c (target_find_new_threads): Rename to ... (target_update_thread_list): ... this. * target.h (struct target_ops): Rename to_find_new_threads field to to_update_thread_list. (target_find_new_threads): Rename to ... (target_update_thread_list): ... this. * thread.c (prune_threads): Make extern. (update_thread_list): Adjust.
2014-10-16 05:44:00 +08:00
ops->to_update_thread_list = tdefault_update_thread_list;
ops->to_pid_to_str = default_pid_to_str;
ops->to_extra_thread_info = tdefault_extra_thread_info;
ops->to_thread_name = tdefault_thread_name;
ops->to_stop = tdefault_stop;
ops->to_rcmd = default_rcmd;
ops->to_pid_to_exec_file = tdefault_pid_to_exec_file;
ops->to_log_command = tdefault_log_command;
ops->to_get_section_table = tdefault_get_section_table;
fix regressions with target-async A patch in the target cleanup series caused a regression when using record with target-async. Version 4 of the patch is here: https://sourceware.org/ml/gdb-patches/2014-03/msg00159.html The immediate problem is that record supplies to_can_async_p and to_is_async_p methods, but does not supply a to_async method. So, when target-async is set, record claims to support async -- but if the underlying target does not support async, then the to_async method call will end up in that method's default implementation, namely tcomplain. This worked previously because the record target used to provide a to_async method; one that (erroneously, only at push time) checked the other members of the target stack, and then simply dropped to_async calls in the "does not implement async" case. My first thought was to simply drop tcomplain as the default for to_async. This works, but Pedro pointed out that the only reason record has to supply to_can_async_p and to_is_async_p is that these default to using the find_default_run_target machinery -- and these defaults are only needed by "run" and "attach". So, a nicer solution presents itself: change run and attach to explicitly call into the default run target when needed; and change to_is_async_p and to_can_async_p to default to "return 0". This makes the target stack simpler to use and lets us remove the method implementations from record. This is also in harmony with other plans for the target stack; namely trying to reduce the impact of find_default_run_target. This approach makes it clear that find_default_is_async_p is not needed -- it is asking whether a target that may not even be pushed is actually async, which seems like a nonsensical question. While an improvement, this approach proved to introduce the same bug when using the core target. Looking a bit deeper, the issue is that code in "attach" and "run" may need to use either the current target stack or the default run target -- but different calls into the target API in those functions could wind up querying different targets. This new patch makes the target to use more explicit in "run" and "attach". Then these commands explicitly make the needed calls against that target. This ensures that a single target is used for all relevant operations. This lets us remove a couple find_default_* functions from various targets, including the dummy target. I think this is a decent understandability improvement. One issue I see with this patch is that the new calls in "run" and "attach" are not very much like the rest of the target API. I think fundamentally this is due to bad factoring in the target API, which may need to be fixed for multi-target. Tackling that seemed ambitious for a regression fix. While working on this I noticed that there don't seem to be any test cases that involve both target-async and record, so this patch changes break-precsave.exp to add some. It also changes corefile.exp to add some target-async tests; these pass with current trunk and with this patch applied, but fail with the v1 patch. This patch differs from v4 in that it moves initialization of to_can_async_p and to_supports_non_stop into inf-child, adds some assertions to complete_target_initialization, and adds some comments to target.h. Built and regtested on x86-64 Fedora 20. 2014-03-12 Tom Tromey <tromey@redhat.com> * inf-child.c (return_zero): New function. (inf_child_target): Set to_can_async_p, to_supports_non_stop. * aix-thread.c (aix_thread_inferior_created): New function. (aix_thread_attach): Remove. (init_aix_thread_ops): Don't set to_attach. (_initialize_aix_thread): Register inferior_created observer. * corelow.c (init_core_ops): Don't set to_attach or to_create_inferior. * exec.c (init_exec_ops): Don't set to_attach or to_create_inferior. * infcmd.c (run_command_1): Use find_run_target. Make direct target calls. (attach_command): Use find_attach_target. Make direct target calls. * record-btrace.c (init_record_btrace_ops): Don't set to_create_inferior. * record-full.c (record_full_can_async_p, record_full_is_async_p): Remove. (init_record_full_ops, init_record_full_core_ops): Update. Don't set to_create_inferior. * target.c (complete_target_initialization): Add assertion. (target_create_inferior): Remove. (find_default_attach, find_default_create_inferior): Remove. (find_attach_target, find_run_target): New functions. (find_default_is_async_p, find_default_can_async_p) (target_supports_non_stop, target_attach): Remove. (init_dummy_target): Don't set to_create_inferior or to_supports_non_stop. * target.h (struct target_ops) <to_attach>: Add comment. Remove TARGET_DEFAULT_FUNC. <to_create_inferior>: Add comment. <to_can_async_p, to_is_async_p, to_supports_non_stop>: Use TARGET_DEFAULT_RETURN. <to_can_async_p, to_supports_non_stop, to_can_run>: Add comments. (find_attach_target, find_run_target): Declare. (target_create_inferior): Remove. (target_has_execution_1): Update comment. (target_supports_non_stop): Remove. * target-delegates.c: Rebuild. 2014-03-12 Tom Tromey <tromey@redhat.com> * gdb.base/corefile.exp (corefile_test_run, corefile_test_attach): New procs. Add target-async tests. * gdb.reverse/break-precsave.exp (precsave_tests): New proc. Add target-async tests.
2014-03-01 00:47:34 +08:00
ops->to_can_async_p = tdefault_can_async_p;
ops->to_is_async_p = tdefault_is_async_p;
add target method delegation This patch replaces some code in the record targets with target method delegation. record-full.c stores pointers to many target methods when the record target is pushed. Then it later delegates some calls via these. This is wrong because it violates the target stack contract. In particular it is ok to unpush a target at any stratum, but record-full does not keep track of this, so it could potentially call into an unpushed target. This patch fixes the problem by using the newly-introduced generic approach to target delegation for the methods in question. 2014-02-19 Tom Tromey <tromey@redhat.com> * record-full.c (record_full_beneath_to_resume_ops) (record_full_beneath_to_resume, record_full_beneath_to_wait_ops) (record_full_beneath_to_wait) (record_full_beneath_to_store_registers_ops) (record_full_beneath_to_store_registers) (record_full_beneath_to_xfer_partial_ops) (record_full_beneath_to_xfer_partial) (record_full_beneath_to_insert_breakpoint_ops) (record_full_beneath_to_insert_breakpoint) (record_full_beneath_to_remove_breakpoint_ops) (record_full_beneath_to_remove_breakpoint) (record_full_beneath_to_stopped_by_watchpoint) (record_full_beneath_to_stopped_data_address) (record_full_beneath_to_async, tmp_to_resume_ops, tmp_to_resume) (tmp_to_wait_ops, tmp_to_wait, tmp_to_store_registers_ops) (tmp_to_store_registers, tmp_to_xfer_partial_ops) (tmp_to_xfer_partial, tmp_to_instmp_to_insert_breakpoint_ops) (tmp_to_insert_breakpoint, tmp_to_remove_breakpoint_ops) (tmp_to_remove_breakpoint, tmp_to_stopped_by_watchpoint) (tmp_to_stopped_data_address, tmp_to_async): Remove. (record_full_open_1, record_full_open): Update. Use RECORD_IS_USED. (record_full_resume, record_full_wait_1) (record_full_stopped_by_watchpoint, record_full_stopped_data_address) (record_full_store_registers, record_full_xfer_partial) (record_full_insert_breakpoint, record_full_remove_breakpoint) (record_full_async, record_full_core_xfer_partial): Use target delegation. * target-delegates.c: Rebuild. * target.c (current_xfer_partial): Remove. (update_current_target): Do not INHERIT or de_fault to_insert_breakpoint, to_remove_breakpoint, to_stopped_data_address, to_stopped_by_watchpoint, to_can_async_p, to_is_async_p, to_async. Do not set to_xfer_partial field. (default_xfer_partial): Simplify. (current_xfer_partial): Remove. (target_wait, target_resume): Simplify. (find_default_can_async_p, find_default_is_async_p): Update. (init_dummy_target): Don't set to_can_async_p, to_is_async_p, to_xfer_partial, to_stopped_by_watchpoint, to_stopped_data_address. (target_store_registers): Simplify. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. (target_remove_breakpoint, target_insert_breakpoint) (debug_to_insert_breakpoint, debug_to_remove_breakpoint): Update. * target.h (struct target_ops) <to_resume, to_wait, to_store_registers, to_insert_breakpoint, to_remove_breakpoint, to_stopped_by_watchpoint, to_stopped_data_address, to_can_async_p, to_is_async_p, to_async, to_xfer_partial>: Add TARGET_DEFAULT markup. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. * record-btrace.c (record_btrace_remove_breakpoint): Delegate directly. (record_btrace_insert_breakpoint): Delegate directly.
2013-07-13 03:00:34 +08:00
ops->to_async = tdefault_async;
fix regressions with target-async A patch in the target cleanup series caused a regression when using record with target-async. Version 4 of the patch is here: https://sourceware.org/ml/gdb-patches/2014-03/msg00159.html The immediate problem is that record supplies to_can_async_p and to_is_async_p methods, but does not supply a to_async method. So, when target-async is set, record claims to support async -- but if the underlying target does not support async, then the to_async method call will end up in that method's default implementation, namely tcomplain. This worked previously because the record target used to provide a to_async method; one that (erroneously, only at push time) checked the other members of the target stack, and then simply dropped to_async calls in the "does not implement async" case. My first thought was to simply drop tcomplain as the default for to_async. This works, but Pedro pointed out that the only reason record has to supply to_can_async_p and to_is_async_p is that these default to using the find_default_run_target machinery -- and these defaults are only needed by "run" and "attach". So, a nicer solution presents itself: change run and attach to explicitly call into the default run target when needed; and change to_is_async_p and to_can_async_p to default to "return 0". This makes the target stack simpler to use and lets us remove the method implementations from record. This is also in harmony with other plans for the target stack; namely trying to reduce the impact of find_default_run_target. This approach makes it clear that find_default_is_async_p is not needed -- it is asking whether a target that may not even be pushed is actually async, which seems like a nonsensical question. While an improvement, this approach proved to introduce the same bug when using the core target. Looking a bit deeper, the issue is that code in "attach" and "run" may need to use either the current target stack or the default run target -- but different calls into the target API in those functions could wind up querying different targets. This new patch makes the target to use more explicit in "run" and "attach". Then these commands explicitly make the needed calls against that target. This ensures that a single target is used for all relevant operations. This lets us remove a couple find_default_* functions from various targets, including the dummy target. I think this is a decent understandability improvement. One issue I see with this patch is that the new calls in "run" and "attach" are not very much like the rest of the target API. I think fundamentally this is due to bad factoring in the target API, which may need to be fixed for multi-target. Tackling that seemed ambitious for a regression fix. While working on this I noticed that there don't seem to be any test cases that involve both target-async and record, so this patch changes break-precsave.exp to add some. It also changes corefile.exp to add some target-async tests; these pass with current trunk and with this patch applied, but fail with the v1 patch. This patch differs from v4 in that it moves initialization of to_can_async_p and to_supports_non_stop into inf-child, adds some assertions to complete_target_initialization, and adds some comments to target.h. Built and regtested on x86-64 Fedora 20. 2014-03-12 Tom Tromey <tromey@redhat.com> * inf-child.c (return_zero): New function. (inf_child_target): Set to_can_async_p, to_supports_non_stop. * aix-thread.c (aix_thread_inferior_created): New function. (aix_thread_attach): Remove. (init_aix_thread_ops): Don't set to_attach. (_initialize_aix_thread): Register inferior_created observer. * corelow.c (init_core_ops): Don't set to_attach or to_create_inferior. * exec.c (init_exec_ops): Don't set to_attach or to_create_inferior. * infcmd.c (run_command_1): Use find_run_target. Make direct target calls. (attach_command): Use find_attach_target. Make direct target calls. * record-btrace.c (init_record_btrace_ops): Don't set to_create_inferior. * record-full.c (record_full_can_async_p, record_full_is_async_p): Remove. (init_record_full_ops, init_record_full_core_ops): Update. Don't set to_create_inferior. * target.c (complete_target_initialization): Add assertion. (target_create_inferior): Remove. (find_default_attach, find_default_create_inferior): Remove. (find_attach_target, find_run_target): New functions. (find_default_is_async_p, find_default_can_async_p) (target_supports_non_stop, target_attach): Remove. (init_dummy_target): Don't set to_create_inferior or to_supports_non_stop. * target.h (struct target_ops) <to_attach>: Add comment. Remove TARGET_DEFAULT_FUNC. <to_create_inferior>: Add comment. <to_can_async_p, to_is_async_p, to_supports_non_stop>: Use TARGET_DEFAULT_RETURN. <to_can_async_p, to_supports_non_stop, to_can_run>: Add comments. (find_attach_target, find_run_target): Declare. (target_create_inferior): Remove. (target_has_execution_1): Update comment. (target_supports_non_stop): Remove. * target-delegates.c: Rebuild. 2014-03-12 Tom Tromey <tromey@redhat.com> * gdb.base/corefile.exp (corefile_test_run, corefile_test_attach): New procs. Add target-async tests. * gdb.reverse/break-precsave.exp (precsave_tests): New proc. Add target-async tests.
2014-03-01 00:47:34 +08:00
ops->to_supports_non_stop = tdefault_supports_non_stop;
ops->to_find_memory_regions = dummy_find_memory_regions;
ops->to_make_corefile_notes = dummy_make_corefile_notes;
ops->to_get_bookmark = tdefault_get_bookmark;
ops->to_goto_bookmark = tdefault_goto_bookmark;
ops->to_get_thread_local_address = tdefault_get_thread_local_address;
add target method delegation This patch replaces some code in the record targets with target method delegation. record-full.c stores pointers to many target methods when the record target is pushed. Then it later delegates some calls via these. This is wrong because it violates the target stack contract. In particular it is ok to unpush a target at any stratum, but record-full does not keep track of this, so it could potentially call into an unpushed target. This patch fixes the problem by using the newly-introduced generic approach to target delegation for the methods in question. 2014-02-19 Tom Tromey <tromey@redhat.com> * record-full.c (record_full_beneath_to_resume_ops) (record_full_beneath_to_resume, record_full_beneath_to_wait_ops) (record_full_beneath_to_wait) (record_full_beneath_to_store_registers_ops) (record_full_beneath_to_store_registers) (record_full_beneath_to_xfer_partial_ops) (record_full_beneath_to_xfer_partial) (record_full_beneath_to_insert_breakpoint_ops) (record_full_beneath_to_insert_breakpoint) (record_full_beneath_to_remove_breakpoint_ops) (record_full_beneath_to_remove_breakpoint) (record_full_beneath_to_stopped_by_watchpoint) (record_full_beneath_to_stopped_data_address) (record_full_beneath_to_async, tmp_to_resume_ops, tmp_to_resume) (tmp_to_wait_ops, tmp_to_wait, tmp_to_store_registers_ops) (tmp_to_store_registers, tmp_to_xfer_partial_ops) (tmp_to_xfer_partial, tmp_to_instmp_to_insert_breakpoint_ops) (tmp_to_insert_breakpoint, tmp_to_remove_breakpoint_ops) (tmp_to_remove_breakpoint, tmp_to_stopped_by_watchpoint) (tmp_to_stopped_data_address, tmp_to_async): Remove. (record_full_open_1, record_full_open): Update. Use RECORD_IS_USED. (record_full_resume, record_full_wait_1) (record_full_stopped_by_watchpoint, record_full_stopped_data_address) (record_full_store_registers, record_full_xfer_partial) (record_full_insert_breakpoint, record_full_remove_breakpoint) (record_full_async, record_full_core_xfer_partial): Use target delegation. * target-delegates.c: Rebuild. * target.c (current_xfer_partial): Remove. (update_current_target): Do not INHERIT or de_fault to_insert_breakpoint, to_remove_breakpoint, to_stopped_data_address, to_stopped_by_watchpoint, to_can_async_p, to_is_async_p, to_async. Do not set to_xfer_partial field. (default_xfer_partial): Simplify. (current_xfer_partial): Remove. (target_wait, target_resume): Simplify. (find_default_can_async_p, find_default_is_async_p): Update. (init_dummy_target): Don't set to_can_async_p, to_is_async_p, to_xfer_partial, to_stopped_by_watchpoint, to_stopped_data_address. (target_store_registers): Simplify. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. (target_remove_breakpoint, target_insert_breakpoint) (debug_to_insert_breakpoint, debug_to_remove_breakpoint): Update. * target.h (struct target_ops) <to_resume, to_wait, to_store_registers, to_insert_breakpoint, to_remove_breakpoint, to_stopped_by_watchpoint, to_stopped_data_address, to_can_async_p, to_is_async_p, to_async, to_xfer_partial>: Add TARGET_DEFAULT markup. (forward_target_remove_breakpoint) (forward_target_insert_breakpoint): Remove. * record-btrace.c (record_btrace_remove_breakpoint): Delegate directly. (record_btrace_insert_breakpoint): Delegate directly.
2013-07-13 03:00:34 +08:00
ops->to_xfer_partial = tdefault_xfer_partial;
ops->to_memory_map = tdefault_memory_map;
ops->to_flash_erase = tdefault_flash_erase;
ops->to_flash_done = tdefault_flash_done;
ops->to_read_description = tdefault_read_description;
ops->to_get_ada_task_ptid = default_get_ada_task_ptid;
ops->to_auxv_parse = default_auxv_parse;
ops->to_search_memory = default_search_memory;
ops->to_can_execute_reverse = tdefault_can_execute_reverse;
ops->to_execution_direction = default_execution_direction;
ops->to_supports_multi_process = tdefault_supports_multi_process;
ops->to_supports_enable_disable_tracepoint = tdefault_supports_enable_disable_tracepoint;
ops->to_supports_string_tracing = tdefault_supports_string_tracing;
ops->to_supports_evaluation_of_breakpoint_conditions = tdefault_supports_evaluation_of_breakpoint_conditions;
ops->to_can_run_breakpoint_commands = tdefault_can_run_breakpoint_commands;
ops->to_thread_architecture = default_thread_architecture;
ops->to_thread_address_space = default_thread_address_space;
ops->to_filesystem_is_local = tdefault_filesystem_is_local;
ops->to_trace_init = tdefault_trace_init;
ops->to_download_tracepoint = tdefault_download_tracepoint;
ops->to_can_download_tracepoint = tdefault_can_download_tracepoint;
ops->to_download_trace_state_variable = tdefault_download_trace_state_variable;
ops->to_enable_tracepoint = tdefault_enable_tracepoint;
ops->to_disable_tracepoint = tdefault_disable_tracepoint;
ops->to_trace_set_readonly_regions = tdefault_trace_set_readonly_regions;
ops->to_trace_start = tdefault_trace_start;
ops->to_get_trace_status = tdefault_get_trace_status;
ops->to_get_tracepoint_status = tdefault_get_tracepoint_status;
ops->to_trace_stop = tdefault_trace_stop;
ops->to_trace_find = tdefault_trace_find;
ops->to_get_trace_state_variable_value = tdefault_get_trace_state_variable_value;
ops->to_save_trace_data = tdefault_save_trace_data;
ops->to_upload_tracepoints = tdefault_upload_tracepoints;
ops->to_upload_trace_state_variables = tdefault_upload_trace_state_variables;
ops->to_get_raw_trace_data = tdefault_get_raw_trace_data;
ops->to_get_min_fast_tracepoint_insn_len = tdefault_get_min_fast_tracepoint_insn_len;
ops->to_set_disconnected_tracing = tdefault_set_disconnected_tracing;
ops->to_set_circular_trace_buffer = tdefault_set_circular_trace_buffer;
ops->to_set_trace_buffer_size = tdefault_set_trace_buffer_size;
ops->to_set_trace_notes = tdefault_set_trace_notes;
ops->to_core_of_thread = tdefault_core_of_thread;
Make compare-sections work against all targets; add compare-sections [-r] tests. This does two things: 1. Adds a test. Recently compare-sections got a new "-r" switch, but given no test existed for compare-sections, the patch was allowed in with no testsuite addition. This now adds a test for both compare-sections and compare-sections -r. 2. Makes the compare-sections command work against all targets. Currently, compare-sections only works with remote targets, and only those that support the qCRC packet. The patch makes it so that if the target doesn't support accelerating memory verification, then GDB falls back to comparing memory itself. This is of course slower, but it's better than nothing, IMO. While testing against extended-remote GDBserver I noticed that we send the qCRC request to the target if we're connected, but not yet running a program. That can't work of course -- the patch fixes that. This all also goes in the direction of bridging the local/remote parity gap. I didn't decouple 1. from 2., because that would mean that the test would need to handle the case of the target not supporting the command. Tested on x86_64 Fedora 17, native, remote GDBserver, and extended-remote GDBserver. I also hack-disabled qCRC support to make sure the fallback paths in remote.c work. gdb/doc/ 2014-05-20 Pedro Alves <palves@redhat.com> * gdb.texinfo (Memory) <compare-sections>: Generalize comments to not be remote specific. Add cross reference to the qCRC packet. (Separate Debug Files): Update cross reference to the qCRC packet. (General Query Packets) <qCRC packet>: Add anchor. gdb/ 2014-05-20 Pedro Alves <palves@redhat.com> * NEWS: Mention that compare-sections now works with all targets. * remote.c (PACKET_qCRC): New enum value. (remote_verify_memory): Don't send qCRC if the target has no execution. Use packet_support/packet_ok. If the target doesn't support the qCRC packet, fallback to a deep memory copy. (compare_sections_command): Say "target image" instead of "remote executable". (_initialize_remote): Add PACKET_qCRC to the list of config packets that have no associated command. Extend comment. * target.c (simple_verify_memory, default_verify_memory): New function. * target.h (struct target_ops) <to_verify_memory>: Default to default_verify_memory. (simple_verify_memory): New declaration. * target-delegates.c: Regenerate. gdb/testsuite/ 2014-05-20 Pedro Alves <palves@redhat.com> * gdb.base/compare-sections.c: New file. * gdb.base/compare-sections.exp: New file.
2014-05-21 02:11:39 +08:00
ops->to_verify_memory = default_verify_memory;
ops->to_get_tib_address = tdefault_get_tib_address;
ops->to_set_permissions = tdefault_set_permissions;
ops->to_static_tracepoint_marker_at = tdefault_static_tracepoint_marker_at;
ops->to_static_tracepoint_markers_by_strid = tdefault_static_tracepoint_markers_by_strid;
ops->to_traceframe_info = tdefault_traceframe_info;
ops->to_use_agent = tdefault_use_agent;
ops->to_can_use_agent = tdefault_can_use_agent;
ops->to_supports_btrace = tdefault_supports_btrace;
ops->to_enable_btrace = tdefault_enable_btrace;
ops->to_disable_btrace = tdefault_disable_btrace;
ops->to_teardown_btrace = tdefault_teardown_btrace;
ops->to_read_btrace = tdefault_read_btrace;
record btrace: add configuration struct Add a struct to describe the branch trace configuration and use it for enabling branch tracing. The user will be able to set configuration fields for each tracing format to be used for new threads. The actual configuration that is active for a given thread will be shown in the "info record" command. At the moment, the configuration struct only contains a format field that is set to the only available format. The format is the only configuration option that can not be set via set commands. It is given as argument to the "record btrace" command when starting recording. 2015-02-09 Markus Metzger <markus.t.metzger@intel.com> * Makefile.in (XMLFILES): Add btrace-conf.dtd. * x86-linux-nat.c (x86_linux_enable_btrace): Update parameters. (x86_linux_btrace_conf): New. (x86_linux_create_target): Initialize to_btrace_conf. * nat/linux-btrace.c (linux_enable_btrace): Update parameters. Check format. Split into this and ... (linux_enable_bts): ... this. (linux_btrace_conf): New. (perf_event_skip_record): Renamed into ... (perf_event_skip_bts_record): ... this. Updated users. (linux_disable_btrace): Split into this and ... (linux_disable_bts): ... this. (linux_read_btrace): Check format. * nat/linux-btrace.h (linux_enable_btrace): Update parameters. (linux_btrace_conf): New. (btrace_target_info)<ptid>: Moved. (btrace_target_info)<conf>: New. (btrace_target_info): Split into this and ... (btrace_tinfo_bts): ... this. Updated users. * btrace.c (btrace_enable): Update parameters. (btrace_conf, parse_xml_btrace_conf_bts, parse_xml_btrace_conf) (btrace_conf_children, btrace_conf_attributes) (btrace_conf_elements): New. * btrace.h (btrace_enable): Update parameters. (btrace_conf, parse_xml_btrace_conf): New. * common/btrace-common.h (btrace_config): New. * feature/btrace-conf.dtd: New. * record-btrace.c (record_btrace_conf): New. (record_btrace_cmdlist): New. (record_btrace_enable_warn, record_btrace_open): Pass &record_btrace_conf. (record_btrace_info): Print recording format. (cmd_record_btrace_bts_start): New. (cmd_record_btrace_start): Call cmd_record_btrace_bts_start. (_initialize_record_btrace): Add "record btrace bts" subcommand. Add "record bts" alias command. * remote.c (remote_state)<btrace_config>: New. (remote_btrace_reset, PACKET_qXfer_btrace_conf): New. (remote_protocol_features): Add qXfer:btrace-conf:read. (remote_open_1): Call remote_btrace_reset. (remote_xfer_partial): Handle TARGET_OBJECT_BTRACE_CONF. (btrace_target_info)<conf>: New. (btrace_sync_conf, btrace_read_config): New. (remote_enable_btrace): Update parameters. Call btrace_sync_conf and btrace_read_conf. (remote_btrace_conf): New. (init_remote_ops): Initialize to_btrace_conf. (_initialize_remote): Add qXfer:btrace-conf packet. * target.c (target_enable_btrace): Update parameters. (target_btrace_conf): New. * target.h (target_enable_btrace): Update parameters. (target_btrace_conf): New. (target_object)<TARGET_OBJECT_BTRACE_CONF>: New. (target_ops)<to_enable_btrace>: Update parameters and comment. (target_ops)<to_btrace_conf>: New. * target-delegates: Regenerate. * target-debug.h (target_debug_print_const_struct_btrace_config_p) (target_debug_print_const_struct_btrace_target_info_p): New. NEWS: Announce new command and new packet. doc/ * gdb.texinfo (Process Record and Replay): Describe the "record btrace bts" command. (General Query Packets): Describe qXfer:btrace-conf:read packet. (Branch Trace Configuration Format): New. gdbserver/ * linux-low.c (linux_low_enable_btrace): Update parameters. (linux_low_btrace_conf): New. (linux_target_ops)<to_btrace_conf>: Initialize. * server.c (current_btrace_conf): New. (handle_btrace_enable): Rename to ... (handle_btrace_enable_bts): ... this. Pass &current_btrace_conf to target_enable_btrace. Update comment. Update users. (handle_qxfer_btrace_conf): New. (qxfer_packets): Add btrace-conf entry. (handle_query): Report qXfer:btrace-conf:read as supported packet. * target.h (target_ops)<enable_btrace>: Update parameters and comment. (target_ops)<read_btrace_conf>: New. (target_enable_btrace): Update parameters. (target_read_btrace_conf): New. testsuite/ * gdb.btrace/delta.exp: Update "info record" output. * gdb.btrace/enable.exp: Update "info record" output. * gdb.btrace/finish.exp: Update "info record" output. * gdb.btrace/instruction_history.exp: Update "info record" output. * gdb.btrace/next.exp: Update "info record" output. * gdb.btrace/nexti.exp: Update "info record" output. * gdb.btrace/step.exp: Update "info record" output. * gdb.btrace/stepi.exp: Update "info record" output. * gdb.btrace/nohist.exp: Update "info record" output.
2013-11-28 22:44:13 +08:00
ops->to_btrace_conf = tdefault_btrace_conf;
ops->to_stop_recording = tdefault_stop_recording;
ops->to_info_record = tdefault_info_record;
ops->to_save_record = tdefault_save_record;
ops->to_delete_record = tdefault_delete_record;
ops->to_record_is_replaying = tdefault_record_is_replaying;
ops->to_goto_record_begin = tdefault_goto_record_begin;
ops->to_goto_record_end = tdefault_goto_record_end;
ops->to_goto_record = tdefault_goto_record;
ops->to_insn_history = tdefault_insn_history;
ops->to_insn_history_from = tdefault_insn_history_from;
ops->to_insn_history_range = tdefault_insn_history_range;
ops->to_call_history = tdefault_call_history;
ops->to_call_history_from = tdefault_call_history_from;
ops->to_call_history_range = tdefault_call_history_range;
ops->to_augmented_libraries_svr4_read = tdefault_augmented_libraries_svr4_read;
ops->to_get_unwinder = tdefault_get_unwinder;
ops->to_get_tailcall_unwinder = tdefault_get_tailcall_unwinder;
ops->to_prepare_to_generate_core = tdefault_prepare_to_generate_core;
ops->to_done_generating_core = tdefault_done_generating_core;
}
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
static void
init_debug_target (struct target_ops *ops)
{
ops->to_post_attach = debug_post_attach;
ops->to_detach = debug_detach;
ops->to_disconnect = debug_disconnect;
ops->to_resume = debug_resume;
ops->to_wait = debug_wait;
ops->to_fetch_registers = debug_fetch_registers;
ops->to_store_registers = debug_store_registers;
ops->to_prepare_to_store = debug_prepare_to_store;
ops->to_files_info = debug_files_info;
ops->to_insert_breakpoint = debug_insert_breakpoint;
ops->to_remove_breakpoint = debug_remove_breakpoint;
Teach GDB about targets that can tell whether a trap is a breakpoint event The moribund locations heuristics are problematic. This patch teaches GDB about targets that can reliably tell whether a trap was caused by a software or hardware breakpoint, and thus don't need moribund locations, thus bypassing all the problems that mechanism has. The non-stop-fair-events.exp test is frequently failing currently. E.g., see https://sourceware.org/ml/gdb-testers/2015-q1/msg03148.html. The root cause is a fundamental problem with moribund locations. For example, the stepped_breakpoint logic added by af48d08f breaks in this case (which is what happens with that test): - Step thread A, no breakpoint is set at PC. - The kernel doesn't schedule thread A yet. - Insert breakpoint at A's PC, for some reason (e.g., a step-resume breakpoint for thread B). - Kernel finally schedules thread A. - thread A's stepped_breakpoint flag is not set, even though it now stepped a breakpoint instruction. - adjust_pc_after_break gets the PC wrong, because PC == PREV_PC, but stepped_breakpoint is not set. We needed the stepped_breakpoint logic to workaround moribund locations, because otherwise adjust_pc_after_break could apply an adjustment when it shouldn't just because there _used_ to be a breakpoint at PC (a moribund breakpoint location). For example, on x86, that's wrong if the thread really hasn't executed an int3, but instead executed some other 1-byte long instruction. Getting the PC adjustment wrong of course leads to the inferior executing the wrong instruction. Other problems with moribund locations are: - if a true SIGTRAP happens to be raised when the program is executing the PC that used to have a breakpoint, GDB will assume that is a trap for a breakpoint that has recently been removed, and thus we miss reporting the random signal to the user. - to minimize that, we get rid of moribund location after a while. That while is defined as just a certain number of events being processed. That number of events sometimes passes by before a delayed breakpoint is processed, and GDB confuses the trap for a random signal, thus reporting the random trap. Once the user resumes the thread, the program crashes because the PC was not adjusted... The fix for all this is to bite the bullet and get rid of heuristics and instead rely on the target knowing accurately what caused the SIGTRAP. The target/kernel/stub is in the best position to know what that, because it can e.g. consult priviledged CPU flags GDB has no access to, or by knowing which exception vector entry was called when the instruction trapped, etc. Most debug APIs I've seen to date report breakpoint hits as a distinct event in some fashion. For example, on the Linux kernel, whether a breakpoint was executed is exposed to userspace in the si_code field of the SIGTRAP's siginfo. On Windows, the debug API reports a EXCEPTION_BREAKPOINT exception code. We needed to keep around deleted breakpoints in an on-the-side list (the moribund locations) for two main reasons: - Know that a SIGTRAP actually is a delayed event for a hit of a breakpoint that was removed before the event was processed, and thus should not be reported as a random signal. - So we still do the decr_pc_after_break adjustment in that case, so that the thread is resumed at the correct address. In the new model, if GDB processes an event the target tells is a breakpoint trap, and GDB doesn't find the corresponding breakpoint in its breakpoint tables, it means that event is a delayed event for a breakpoint that has since been removed, and thus the event should be ignored. For the decr_pc_after_after issue, it ends up being much simpler that on targets that can reliably tell whether a breakpoint trapped, for the breakpoint trap to present the PC already adjusted. Proper multi-threading support already implies that targets needs to be doing decr_pc_after_break adjustment themselves, otherwise for example, in all-stop if two threads hit a breakpoint simultaneously, and the user does "info threads", he'll see the non-event thread that hit the breakpoint stopped at the wrong PC. This way (target adjusts) also ends up eliminating the need for some awkward re-incrementing of the PC in the record-full and Linux targets that we do today, and the need for the target_decr_pc_after_break hook. If the target always adjusts, then there's a case where GDB needs to re-increment the PC. Say, on x86, an "int3" instruction that was explicitly written in the program traps. In this case, GDB should report a random SIGTRAP signal to the user, with the PC pointing at the instruction past the int3, just like if GDB was not debugging the program. The user may well decide to pass the SIGTRAP to the program because the program being debugged has a SIGTRAP handler that handles its own breakpoints, and expects the PC to be unadjusted. Tested on x86-64 Fedora 20. gdb/ChangeLog: 2015-03-04 Pedro Alves <palves@redhat.com> * breakpoint.c (need_moribund_for_location_type): New function. (bpstat_stop_status): Don't skipping checking moribund locations of breakpoint types which the target tell caused a stop. (program_breakpoint_here_p): New function, factored out from ... (bp_loc_is_permanent): ... this. (update_global_location_list): Don't create a moribund location if the target supports reporting stops of the type of the removed breakpoint. * breakpoint.h (program_breakpoint_here_p): New declaration. * infrun.c (adjust_pc_after_break): Return early if the target has already adjusted the PC. Add comments. (handle_signal_stop): If nothing explains a signal, and the target tells us the stop was caused by a software breakpoint, check if there's a breakpoint instruction in the memory. If so, adjust the PC before presenting the stop to the user. Otherwise, ignore the trap. If nothing explains a signal, and the target tells us the stop was caused by a hardware breakpoint, ignore the trap. * target.h (struct target_ops) <to_stopped_by_sw_breakpoint, to_supports_stopped_by_sw_breakpoint, to_stopped_by_hw_breakpoint, to_supports_stopped_by_hw_breakpoint>: New fields. (target_stopped_by_sw_breakpoint) (target_supports_stopped_by_sw_breakpoint) (target_stopped_by_hw_breakpoint) (target_supports_stopped_by_hw_breakpoint): Define. * target-delegates.c: Regenerate.
2015-03-05 04:41:15 +08:00
ops->to_stopped_by_sw_breakpoint = debug_stopped_by_sw_breakpoint;
ops->to_supports_stopped_by_sw_breakpoint = debug_supports_stopped_by_sw_breakpoint;
ops->to_stopped_by_hw_breakpoint = debug_stopped_by_hw_breakpoint;
ops->to_supports_stopped_by_hw_breakpoint = debug_supports_stopped_by_hw_breakpoint;
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
ops->to_can_use_hw_breakpoint = debug_can_use_hw_breakpoint;
ops->to_ranged_break_num_registers = debug_ranged_break_num_registers;
ops->to_insert_hw_breakpoint = debug_insert_hw_breakpoint;
ops->to_remove_hw_breakpoint = debug_remove_hw_breakpoint;
ops->to_remove_watchpoint = debug_remove_watchpoint;
ops->to_insert_watchpoint = debug_insert_watchpoint;
ops->to_insert_mask_watchpoint = debug_insert_mask_watchpoint;
ops->to_remove_mask_watchpoint = debug_remove_mask_watchpoint;
ops->to_stopped_by_watchpoint = debug_stopped_by_watchpoint;
ops->to_stopped_data_address = debug_stopped_data_address;
ops->to_watchpoint_addr_within_range = debug_watchpoint_addr_within_range;
ops->to_region_ok_for_hw_watchpoint = debug_region_ok_for_hw_watchpoint;
ops->to_can_accel_watchpoint_condition = debug_can_accel_watchpoint_condition;
ops->to_masked_watch_num_registers = debug_masked_watch_num_registers;
ops->to_terminal_init = debug_terminal_init;
ops->to_terminal_inferior = debug_terminal_inferior;
ops->to_terminal_ours_for_output = debug_terminal_ours_for_output;
ops->to_terminal_ours = debug_terminal_ours;
ops->to_terminal_info = debug_terminal_info;
ops->to_kill = debug_kill;
ops->to_load = debug_load;
ops->to_post_startup_inferior = debug_post_startup_inferior;
ops->to_insert_fork_catchpoint = debug_insert_fork_catchpoint;
ops->to_remove_fork_catchpoint = debug_remove_fork_catchpoint;
ops->to_insert_vfork_catchpoint = debug_insert_vfork_catchpoint;
ops->to_remove_vfork_catchpoint = debug_remove_vfork_catchpoint;
ops->to_follow_fork = debug_follow_fork;
ops->to_insert_exec_catchpoint = debug_insert_exec_catchpoint;
ops->to_remove_exec_catchpoint = debug_remove_exec_catchpoint;
ops->to_set_syscall_catchpoint = debug_set_syscall_catchpoint;
ops->to_has_exited = debug_has_exited;
ops->to_mourn_inferior = debug_mourn_inferior;
ops->to_can_run = debug_can_run;
ops->to_pass_signals = debug_pass_signals;
ops->to_program_signals = debug_program_signals;
ops->to_thread_alive = debug_thread_alive;
Push pruning old threads down to the target When GDB wants to sync the thread list with the target's (e.g., due to "info threads"), it calls update_thread_list: update_thread_list (void) { prune_threads (); target_find_new_threads (); update_threads_executing (); } And then prune_threads does: prune_threads (void) { struct thread_info *tp, *next; for (tp = thread_list; tp; tp = next) { next = tp->next; if (!thread_alive (tp)) delete_thread (tp->ptid); } } Calling thread_live on each thread one by one is expensive. E.g., on Linux, it ends up doing kill(SIG0) once for each thread. Not a big deal, but still a bunch of syscalls... With the remote target, it's cumbersome. That thread_alive call ends up generating one T packet per thread: Sending packet: $Tp2141.2150#82...Packet received: OK Sending packet: $Tp2141.214f#b7...Packet received: OK Sending packet: $Tp2141.2141#82...Packet received: OK Sending packet: $qXfer:threads:read::0,fff#03...Packet received: l<threads>\n<thread id="p2141.2141" core="2"/>\n<thread id="p2141.214f" core="1"/>\n<thread id="p2141.2150" core="2"/>\n</threads>\n That seems a bit silly when target_find_new_threads method implementations will always fetch the whole current set of target threads, and then add those that are not in GDB's thread list, to GDB's thread list. This patch thus pushes down the responsibility of pruning dead threads to the target_find_new_threads method instead, so a target may implement pruning dead threads however it wants. Once we do that, target_find_new_threads becomes a misnomer, so the patch renames it to target_update_thread_list. The patch doesn't attempt to do any optimization to any target yet. It simply exports prune_threads, and makes all implementations of target_update_thread_list call that. It's meant to be a no-op. gdb/ 2014-10-15 Pedro Alves <palves@redhat.com> * ada-tasks.c (print_ada_task_info, task_command_1): Adjust. * bsd-uthread.c (bsd_uthread_find_new_threads): Rename to ... (bsd_uthread_update_thread_list): ... this. Call prune_threads. (bsd_uthread_target): Adjust. * corelow.c (core_open): Adjust. * dec-thread.c (dec_thread_find_new_threads): Update comment. (dec_thread_update_thread_list): New function. (init_dec_thread_ops): Adjust. * gdbthread.h (prune_threads): New declaration. * linux-thread-db.c (thread_db_find_new_threads): Rename to ... (thread_db_update_thread_list): ... this. Call prune_threads. (init_thread_db_ops): Adjust. * nto-procfs.c (procfs_find_new_threads): Rename to ... (procfs_update_thread_list): ... this. Call prune_threads. (procfs_attach, procfs_create_inferior, init_procfs_targets): Adjust. * obsd-nat.c (obsd_find_new_threads): Rename to ... (obsd_update_thread_list): ... this. Call prune_threads. (obsd_add_target): Adjust. * procfs.c (procfs_target): Adjust. (procfs_notice_thread): Update comment. (procfs_find_new_threads): Rename to ... (procfs_update_thread_list): ... this. Call prune_threads. * ravenscar-thread.c (ravenscar_update_inferior_ptid): Update comment. (ravenscar_wait): Adjust. (ravenscar_find_new_threads): Rename to ... (ravenscar_update_thread_list): ... this. Call prune_threads. (init_ravenscar_thread_ops): Adjust. * record-btrace.c (record_btrace_find_new_threads): Rename to ... (record_btrace_update_thread_list): ... this. Adjust comment. (init_record_btrace_ops): Adjust. * remote.c (remote_threads_info): Rename to ... (remote_update_thread_list): ... this. Call prune_threads. (remote_start_remote, extended_remote_attach_1, init_remote_ops): Adjust. * sol-thread.c (check_for_thread_db): Adjust. (sol_find_new_threads_callback): Rename to ... (sol_update_thread_list_callback): ... this. (sol_find_new_threads): Rename to ... (sol_update_thread_list): ... this. Call prune_threads. Adjust. (sol_get_ada_task_ptid, init_sol_thread_ops): Adjust. * target-delegates.c: Regenerate. * target.c (target_find_new_threads): Rename to ... (target_update_thread_list): ... this. * target.h (struct target_ops): Rename to_find_new_threads field to to_update_thread_list. (target_find_new_threads): Rename to ... (target_update_thread_list): ... this. * thread.c (prune_threads): Make extern. (update_thread_list): Adjust.
2014-10-16 05:44:00 +08:00
ops->to_update_thread_list = debug_update_thread_list;
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
ops->to_pid_to_str = debug_pid_to_str;
ops->to_extra_thread_info = debug_extra_thread_info;
ops->to_thread_name = debug_thread_name;
ops->to_stop = debug_stop;
ops->to_rcmd = debug_rcmd;
ops->to_pid_to_exec_file = debug_pid_to_exec_file;
ops->to_log_command = debug_log_command;
ops->to_get_section_table = debug_get_section_table;
ops->to_can_async_p = debug_can_async_p;
ops->to_is_async_p = debug_is_async_p;
ops->to_async = debug_async;
ops->to_supports_non_stop = debug_supports_non_stop;
ops->to_find_memory_regions = debug_find_memory_regions;
ops->to_make_corefile_notes = debug_make_corefile_notes;
ops->to_get_bookmark = debug_get_bookmark;
ops->to_goto_bookmark = debug_goto_bookmark;
ops->to_get_thread_local_address = debug_get_thread_local_address;
ops->to_xfer_partial = debug_xfer_partial;
ops->to_memory_map = debug_memory_map;
ops->to_flash_erase = debug_flash_erase;
ops->to_flash_done = debug_flash_done;
ops->to_read_description = debug_read_description;
ops->to_get_ada_task_ptid = debug_get_ada_task_ptid;
ops->to_auxv_parse = debug_auxv_parse;
ops->to_search_memory = debug_search_memory;
ops->to_can_execute_reverse = debug_can_execute_reverse;
ops->to_execution_direction = debug_execution_direction;
ops->to_supports_multi_process = debug_supports_multi_process;
ops->to_supports_enable_disable_tracepoint = debug_supports_enable_disable_tracepoint;
ops->to_supports_string_tracing = debug_supports_string_tracing;
ops->to_supports_evaluation_of_breakpoint_conditions = debug_supports_evaluation_of_breakpoint_conditions;
ops->to_can_run_breakpoint_commands = debug_can_run_breakpoint_commands;
ops->to_thread_architecture = debug_thread_architecture;
ops->to_thread_address_space = debug_thread_address_space;
ops->to_filesystem_is_local = debug_filesystem_is_local;
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
ops->to_trace_init = debug_trace_init;
ops->to_download_tracepoint = debug_download_tracepoint;
ops->to_can_download_tracepoint = debug_can_download_tracepoint;
ops->to_download_trace_state_variable = debug_download_trace_state_variable;
ops->to_enable_tracepoint = debug_enable_tracepoint;
ops->to_disable_tracepoint = debug_disable_tracepoint;
ops->to_trace_set_readonly_regions = debug_trace_set_readonly_regions;
ops->to_trace_start = debug_trace_start;
ops->to_get_trace_status = debug_get_trace_status;
ops->to_get_tracepoint_status = debug_get_tracepoint_status;
ops->to_trace_stop = debug_trace_stop;
ops->to_trace_find = debug_trace_find;
ops->to_get_trace_state_variable_value = debug_get_trace_state_variable_value;
ops->to_save_trace_data = debug_save_trace_data;
ops->to_upload_tracepoints = debug_upload_tracepoints;
ops->to_upload_trace_state_variables = debug_upload_trace_state_variables;
ops->to_get_raw_trace_data = debug_get_raw_trace_data;
ops->to_get_min_fast_tracepoint_insn_len = debug_get_min_fast_tracepoint_insn_len;
ops->to_set_disconnected_tracing = debug_set_disconnected_tracing;
ops->to_set_circular_trace_buffer = debug_set_circular_trace_buffer;
ops->to_set_trace_buffer_size = debug_set_trace_buffer_size;
ops->to_set_trace_notes = debug_set_trace_notes;
ops->to_core_of_thread = debug_core_of_thread;
ops->to_verify_memory = debug_verify_memory;
ops->to_get_tib_address = debug_get_tib_address;
ops->to_set_permissions = debug_set_permissions;
ops->to_static_tracepoint_marker_at = debug_static_tracepoint_marker_at;
ops->to_static_tracepoint_markers_by_strid = debug_static_tracepoint_markers_by_strid;
ops->to_traceframe_info = debug_traceframe_info;
ops->to_use_agent = debug_use_agent;
ops->to_can_use_agent = debug_can_use_agent;
ops->to_supports_btrace = debug_supports_btrace;
ops->to_enable_btrace = debug_enable_btrace;
ops->to_disable_btrace = debug_disable_btrace;
ops->to_teardown_btrace = debug_teardown_btrace;
ops->to_read_btrace = debug_read_btrace;
record btrace: add configuration struct Add a struct to describe the branch trace configuration and use it for enabling branch tracing. The user will be able to set configuration fields for each tracing format to be used for new threads. The actual configuration that is active for a given thread will be shown in the "info record" command. At the moment, the configuration struct only contains a format field that is set to the only available format. The format is the only configuration option that can not be set via set commands. It is given as argument to the "record btrace" command when starting recording. 2015-02-09 Markus Metzger <markus.t.metzger@intel.com> * Makefile.in (XMLFILES): Add btrace-conf.dtd. * x86-linux-nat.c (x86_linux_enable_btrace): Update parameters. (x86_linux_btrace_conf): New. (x86_linux_create_target): Initialize to_btrace_conf. * nat/linux-btrace.c (linux_enable_btrace): Update parameters. Check format. Split into this and ... (linux_enable_bts): ... this. (linux_btrace_conf): New. (perf_event_skip_record): Renamed into ... (perf_event_skip_bts_record): ... this. Updated users. (linux_disable_btrace): Split into this and ... (linux_disable_bts): ... this. (linux_read_btrace): Check format. * nat/linux-btrace.h (linux_enable_btrace): Update parameters. (linux_btrace_conf): New. (btrace_target_info)<ptid>: Moved. (btrace_target_info)<conf>: New. (btrace_target_info): Split into this and ... (btrace_tinfo_bts): ... this. Updated users. * btrace.c (btrace_enable): Update parameters. (btrace_conf, parse_xml_btrace_conf_bts, parse_xml_btrace_conf) (btrace_conf_children, btrace_conf_attributes) (btrace_conf_elements): New. * btrace.h (btrace_enable): Update parameters. (btrace_conf, parse_xml_btrace_conf): New. * common/btrace-common.h (btrace_config): New. * feature/btrace-conf.dtd: New. * record-btrace.c (record_btrace_conf): New. (record_btrace_cmdlist): New. (record_btrace_enable_warn, record_btrace_open): Pass &record_btrace_conf. (record_btrace_info): Print recording format. (cmd_record_btrace_bts_start): New. (cmd_record_btrace_start): Call cmd_record_btrace_bts_start. (_initialize_record_btrace): Add "record btrace bts" subcommand. Add "record bts" alias command. * remote.c (remote_state)<btrace_config>: New. (remote_btrace_reset, PACKET_qXfer_btrace_conf): New. (remote_protocol_features): Add qXfer:btrace-conf:read. (remote_open_1): Call remote_btrace_reset. (remote_xfer_partial): Handle TARGET_OBJECT_BTRACE_CONF. (btrace_target_info)<conf>: New. (btrace_sync_conf, btrace_read_config): New. (remote_enable_btrace): Update parameters. Call btrace_sync_conf and btrace_read_conf. (remote_btrace_conf): New. (init_remote_ops): Initialize to_btrace_conf. (_initialize_remote): Add qXfer:btrace-conf packet. * target.c (target_enable_btrace): Update parameters. (target_btrace_conf): New. * target.h (target_enable_btrace): Update parameters. (target_btrace_conf): New. (target_object)<TARGET_OBJECT_BTRACE_CONF>: New. (target_ops)<to_enable_btrace>: Update parameters and comment. (target_ops)<to_btrace_conf>: New. * target-delegates: Regenerate. * target-debug.h (target_debug_print_const_struct_btrace_config_p) (target_debug_print_const_struct_btrace_target_info_p): New. NEWS: Announce new command and new packet. doc/ * gdb.texinfo (Process Record and Replay): Describe the "record btrace bts" command. (General Query Packets): Describe qXfer:btrace-conf:read packet. (Branch Trace Configuration Format): New. gdbserver/ * linux-low.c (linux_low_enable_btrace): Update parameters. (linux_low_btrace_conf): New. (linux_target_ops)<to_btrace_conf>: Initialize. * server.c (current_btrace_conf): New. (handle_btrace_enable): Rename to ... (handle_btrace_enable_bts): ... this. Pass &current_btrace_conf to target_enable_btrace. Update comment. Update users. (handle_qxfer_btrace_conf): New. (qxfer_packets): Add btrace-conf entry. (handle_query): Report qXfer:btrace-conf:read as supported packet. * target.h (target_ops)<enable_btrace>: Update parameters and comment. (target_ops)<read_btrace_conf>: New. (target_enable_btrace): Update parameters. (target_read_btrace_conf): New. testsuite/ * gdb.btrace/delta.exp: Update "info record" output. * gdb.btrace/enable.exp: Update "info record" output. * gdb.btrace/finish.exp: Update "info record" output. * gdb.btrace/instruction_history.exp: Update "info record" output. * gdb.btrace/next.exp: Update "info record" output. * gdb.btrace/nexti.exp: Update "info record" output. * gdb.btrace/step.exp: Update "info record" output. * gdb.btrace/stepi.exp: Update "info record" output. * gdb.btrace/nohist.exp: Update "info record" output.
2013-11-28 22:44:13 +08:00
ops->to_btrace_conf = debug_btrace_conf;
auto-generate most target debug methods The target debug methods are inconsistently maintained. Most to_* methods have some kind of targetdebug awareness, but not all of them do. The ones that do vary in the quantity and quality of output they generate. This patch changes most of the target debug methods to be automatically generated. All the arguments are printed, and separate lines are printed for entering and existing the outermost call to the target stack. For example now you'd see: -> multi-thread->to_terminal_ours (...) -> multi-thread->to_is_async_p (...) <- multi-thread->to_is_async_p (0x1ebb580) = 1 <- multi-thread->to_terminal_ours (0x1ebb580) -> multi-thread->to_thread_address_space (...) <- multi-thread->to_thread_address_space (0x1ebb580, 26802) = 1 In this case you can see nested calls. The "multi-thread" on the left hand side is the topmost target's shortname. There are some oddities with this patch. I'm on the fence about it all, I really just wrote it on a whim. It's not simple to convert every possible method, since a few don't participate in target delegation. Printing is done by type, so I introduced some new debug-printing-specific typedefs to handle cases where it is nicer to do something else. On the plus side, this lays the groundwork for making targetdebug affect every layer of the target stack. The idea would be to wrap each target_ops in the stack with its own debug_target, and then you could see calls propagate down the stack and back up; I suppose with indentation to make it prettier. (That said there are some gotchas lurking in this idea due to target stack introspection.) Regtested on x86-64 Fedora 20. 2014-07-24 Tom Tromey <tromey@redhat.com> * make-target-delegates (munge_type, write_debugmethod): New functions. (debug_names): New global. ($TARGET_DEBUG_PRINTER): New global. (write_function_header): Strip TARGET_DEBUG_PRINTER from the type name. Write debug methods. Generate init_debug_target. * target-debug.h: New file. * target-delegates.c: Rebuild. * target.c: Include target-debug.h. (debug_target): Hoist definition. (target_kill, target_get_section_table, target_memory_map) (target_flash_erase, target_flash_done, target_detach) (target_disconnect, target_wait, target_resume) (target_pass_signals, target_program_signals, target_follow_fork) (target_mourn_inferior, target_search_memory) (target_thread_address_space, target_close) (target_find_new_threads, target_core_of_thread) (target_verify_memory, target_insert_mask_watchpoint) (target_remove_mask_watchpoint): Remove targetdebug code. (debug_to_post_attach, debug_to_prepare_to_store) (debug_to_files_info, debug_to_insert_breakpoint) (debug_to_remove_breakpoint, debug_to_can_use_hw_breakpoint) (debug_to_region_ok_for_hw_watchpoint) (debug_to_can_accel_watchpoint_condition) (debug_to_stopped_by_watchpoint, debug_to_stopped_data_address) (debug_to_watchpoint_addr_within_range) (debug_to_insert_hw_breakpoint, debug_to_remove_hw_breakpoint) (debug_to_insert_watchpoint, debug_to_remove_watchpoint) (debug_to_terminal_init, debug_to_terminal_inferior) (debug_to_terminal_ours_for_output, debug_to_terminal_ours) (debug_to_terminal_save_ours, debug_to_terminal_info) (debug_to_load, debug_to_post_startup_inferior) (debug_to_insert_fork_catchpoint) (debug_to_remove_fork_catchpoint) (debug_to_insert_vfork_catchpoint) (debug_to_remove_vfork_catchpoint) (debug_to_insert_exec_catchpoint) (debug_to_remove_exec_catchpoint, debug_to_has_exited) (debug_to_can_run, debug_to_thread_architecture, debug_to_stop) (debug_to_rcmd, debug_to_pid_to_exec_file): Remove. (setup_target_debug): Call init_debug_target. * target.h (TARGET_DEBUG_PRINTER): New macro. (struct target_ops) <to_resume, to_wait, to_pass_signals, to_program_signals>: Use TARGET_DEBUG_PRINTER.
2014-06-16 11:28:57 +08:00
ops->to_stop_recording = debug_stop_recording;
ops->to_info_record = debug_info_record;
ops->to_save_record = debug_save_record;
ops->to_delete_record = debug_delete_record;
ops->to_record_is_replaying = debug_record_is_replaying;
ops->to_goto_record_begin = debug_goto_record_begin;
ops->to_goto_record_end = debug_goto_record_end;
ops->to_goto_record = debug_goto_record;
ops->to_insn_history = debug_insn_history;
ops->to_insn_history_from = debug_insn_history_from;
ops->to_insn_history_range = debug_insn_history_range;
ops->to_call_history = debug_call_history;
ops->to_call_history_from = debug_call_history_from;
ops->to_call_history_range = debug_call_history_range;
ops->to_augmented_libraries_svr4_read = debug_augmented_libraries_svr4_read;
ops->to_get_unwinder = debug_get_unwinder;
ops->to_get_tailcall_unwinder = debug_get_tailcall_unwinder;
ops->to_prepare_to_generate_core = debug_prepare_to_generate_core;
ops->to_done_generating_core = debug_done_generating_core;
}