Markus reported that ASNS breaks target record-btrace. In particular,
the gdb.btrace/multi-thread-step.exp test fails (both with BTS and PT
tracing) with a crash in py-inferior.c:
Program received signal SIGSEGV, Segmentation fault.
0x00000000006aa40d in add_thread_object (tp=0x27d32d0)
at /users/mmetzger/team/gdb/git/gdb/python/py-inferior.c:337
337 entry->next = inf_obj->threads;
My machine doesn't support BTS nor PT, so I missed this...
Disabling ASNS temporarily on x86 until this is addressed.
Tested on x86_64 Fedora 20.
gdb/ChangeLog:
2015-08-18 Pedro Alves <palves@redhat.com>
* linux-nat.c (linux_nat_always_non_stop_p): If the linux_ops
target implements to_always_non_stop_p, call it.
* x86-linux-nat.c (x86_linux_always_non_stop_p): New function.
(x86_linux_create_target): Install it as to_always_non_stop_p
method.
bfd/
* elf64-ppc.c (ppc64_elf_func_desc_adjust): Don't redefine .TOC.
if already defined, and set linker_def.
(ppc64_elf_set_toc): Use .TOC. value if defined other than by
the backend.
ld/
* ldexp.c (exp_fold_tree_1): Clear linker_def on symbol assignment.
In D, all named enums are explicitly scoped (the C++ equivalent of enum class)
so they should be handled as such in the language-specific symbol lookup
routines. However so as to support D compilers that don't emit enums as
DW_AT_enum_class, need to make sure that appropriate checks for
TYPE_DECLARED_CLASS are done.
gdb/ChangeLog
* d-exp.y (type_aggregate_p): New function.
(PrimaryExpression : TypeExp '.' IdentifierExp): Use it.
(classify_inner_name): Likewise.
* d-namespace.c (d_lookup_nested_symbol): Handle TYPE_CODE_ENUM.
One of the build slaves shows this error running explicit.exp:
(gdb) strace -m gdbfoobarbaz
Remote failure reply: E.In-process agent library not loaded in process.
Fast and static tracepoints unavailable.
(gdb) FAIL: gdb.linespec/explicit.exp: strace -m gdbfoobarbaz
There are two big problems with this test:
1) The expected output is actually not what the test is meant to test for.
2) This test should really only run where it is supported.
This is most easily fixed by moving the test to gdb.trace/strace.exp.
gdb/testsuite/ChangeLog
* gdb.linespec/explicit.exp: Move strace test from here ...
* gdb.trace/strace.exp: ... to here.
HJ recently removed trailing space in binutils files, but unfortunately
they return when regenerating files in opcodes. This fixes the regen
process.
* cgen.sh: Trim trailing space from cgen output.
* ia64-gen.c (print_dependency_table): Don't generate trailing space.
(print_dis_table): Likewise.
* opc2c.c (dump_lines): Likewise.
(orig_filename): Warning fix.
* ia64-asmtab.c: Regenerate.
Invoking either of the above commands on an inferior that's not running
triggers the following assert failure:
.../binutils-gdb/gdb/thread.c:514: internal-error: any_thread_of_process: Assertion `pid != 0' failed.
The fix is straightforward. This patch also adds a test to check the
basic functionality of these commands, along with testing this fix in
particular. Tested on x86_64 Linux.
gdb/ChangeLog:
* inferior.c (detach_inferior_command): Don't call
any_thread_of_process when pid is 0.
(kill_inferior_command): Likewise.
gdb/testsuite/ChangeLog:
* gdb.base/kill-detach-inferiors-cmd.exp: New test file.
* gdb.base/kill-detach-inferiors-cmd.c: New test file.
The "source centric" /m option to the disassemble command is often
unhelpful, e.g., in the presence of optimized code.
This patch adds a /s modifier that is better.
For one, /m only prints instructions from the originating source file,
leaving out instructions from e.g., inlined functions from other files.
gdb/ChangeLog:
PR gdb/11833
* NEWS: Document new /s modifier for the disassemble command.
* cli/cli-cmds.c (disassemble_command): Add support for /s.
(_initialize_cli_cmds): Update online docs of disassemble command.
* disasm.c: #include "source.h".
(struct deprecated_dis_line_entry): Renamed from dis_line_entry.
All uses updated.
(dis_line_entry): New struct.
(hash_dis_line_entry, eq_dis_line_entry): New functions.
(allocate_dis_line_table): New functions.
(maybe_add_dis_line_entry, line_has_code_p): New functions.
(dump_insns): New arg end_pc. All callers updated.
(do_mixed_source_and_assembly_deprecated): Renamed from
do_mixed_source_and_assembly. All callers updated.
(do_mixed_source_and_assembly): New function.
(gdb_disassembly): Handle /s (DISASSEMBLY_SOURCE).
* disasm.h (DISASSEMBLY_SOURCE_DEPRECATED): Renamed from
DISASSEMBLY_SOURCE. All uses updated.
(DISASSEMBLY_SOURCE): New macro.
* mi/mi-cmd-disas.c (mi_cmd_disassemble): New modes 4,5.
gdb/doc/ChangeLog:
* gdb.texinfo (Machine Code): Update docs for mixed source/assembly
disassembly.
(GDB/MI Data Manipulation): Update docs for new disassembly modes.
gdb/testsuite/ChangeLog:
* gdb.mi/mi-disassemble.exp: Update.
* gdb.base/disasm-optim.S: New file.
* gdb.base/disasm-optim.c: New file.
* gdb.base/disasm-optim.h: New file.
* gdb.base/disasm-optim.exp: New file.
A recent patch introduced a variable named `typename' into d-exp.y,
and one of the --enable-with-cxx build slaves consequently failed to compile
this. This patch simply adds an underscore into the name to avoid the
reserved word.
gdb/ChangeLog
* d-exp.y (PrimaryExpression : TypeExp '.' IdentifierExp): Rename
`typename' to `type_name' to avoid C++ reserved word.
The locations patch I recently committed contains macro definitions
such as:
This causes an ARI error to be emitted by the server ("Do not use PTR, ISO C
90 implies `void *'"). While this ARI error is bogus in this context,
it is just easiest to squash the error completely by renaming the macro
parameters.
gdb/ChangeLog
* location.c (EL_TYPE, EL_LINESPEC, EL_PROBE, EL_ADDRESS)
(EL_EXPLICIT, EL_STRING): Change macro parameter to "P" to
silence ARI errors.
For some time now, GDB has permitted target-side evaluation of
breakpoint conditions. On targets that support this feature, GDB
may output an "evaluated-by" field into the breakpoint reply.
This patch adds handling for this option, and outputs a default
pattern to optionally recognize (and ignore) this pattern in the
reply.
gdb/testsuite/ChangeLog
* lib/mi-support.exp (mi_make_breakpoint): Add option/handling for
"evaluated-by".
This fixes four ARI warnings found in d-exp.y.
This is comprised of three uses of the && or || at the end of a line, and one
use of sprintf.
gdb/ChangeLog
* d-exp.y (PrimaryExpression : TypeExp '.' IdentifierExp): Use
xstrprintf instead of malloc and sprintf.
(PrimaryExpression : IdentifierExp): Avoid operator at end of line.
(lex_one_token): Likewise.
This tag allows debugging of MIPS position independent executables
and provides access to shared library information.
gdb/gdbserver/
* linux-low.c (get_r_debug): Handle DT_MIPS_RLD_MAP_REL.
gdb/
* solib-svr4.c (read_program_header): Add base_addr argument to
report the runtime address of the segment.
(find_program_interpreter): Update read_program_header call to pass
a NULL pointer for the new argument.
(scan_dyntag): Add ptr_addr argument to report the runtime address
of the tag payload.
(scan_dyntag_auxv): Likewise and use thew new base_addr argument of
read_program_header to get the base address of the dynamic segment.
(elf_locate_base): Update uses of scan_dyntag, scan_dyntag_auxv and
read_program_header.
(elf_locate_base): Scan for and handle DT_MIPS_RLD_MAP_REL.
This makes it so that alternating '.' and identifier tokens are resolved to
symbols as early as possible, which should all the addition of D properties -
such as EXP.sizeof and EXP.typeof - without the shift/reduce conflicts that
would occur in the current parsing strategy.
gdb/ChangeLog
* d-exp.y (%union): Add voidval.
(%token): Add UNKNOWN_NAME as a token to represent an unclassified
name in the lexing stage.
(PostfixExpression): Move symbol completion handling in grammar here
from PrimaryExpression.
(PrimaryExpression): Move routines to handle resolving identifier
tokens in the grammar here from push_expression_name.
(IdentifierExp): Remove the handling of alternating '.' and identifier
tokens.
(TypeExp): Allow TypeExp to be wrapped in parenthesis in the grammar.
(BasicType): Remove C-style typename rules.
(d_type_from_name, d_module_from_name, push_variable)
(push_fieldnames, push_type_name, push_module_name)
(push_expression_name): Remove.
(lex_one_token): Rename from yylex. Replace pstate with par_state.
(token_and_value): New type.
(token_fifo, popping, name_obstack): New globals.
(classify_name): New function.
(classify_inner_name): Likewise.
(yylex): Likewise.
(d_parse): Initialize token_fifo, popping and name_obstack.
In D, there is the notion of modules, and importing from one to the other,
whether it is a basic, selective or renamed import declaration.
module A;
import X;
void foo() {
import Y : bar;
}
If the compiler emits DW_TAG_imported_declaration at the appropriate locations,
then we can make use of what gdb stores in using_direct when performing
nonlocal symbol lookups.
gdb/ChangeLog
* Makefile.in (SFILES): Add d-namespace.c.
(COMMON_OBS): Add d-namespace.o.
* d-lang.c (d_language_defn): Use d_lookup_symbol_nonlocal as the
la_lookup_symbol_nonlocal callback function pointer.
* d-lang.h (d_lookup_symbol_nonlocal): New declaration.
(d_lookup_nested_symbol): New declaration.
* d-namespace.c: New file.
Valgrind shows:
==17026== Invalid write of size 8
==17026== at 0x54AA80: pending_frame_invalidate (py-unwind.c:477)
==17026== by 0x5AB934: do_my_cleanups (cleanups.c:155)
==17026== by 0x5AB9AF: do_cleanups (cleanups.c:177)
==17026== by 0x54B009: pyuw_sniffer (py-unwind.c:606)
==17026== by 0x755DAC: frame_unwind_try_unwinder (frame-unwind.c:105)
==17026== by 0x755EEE: frame_unwind_find_by_frame (frame-unwind.c:160)
==17026== by 0x750FFA: compute_frame_id (frame.c:454)
==17026== by 0x753BD6: get_prev_frame_if_no_cycle (frame.c:1781)
==17026== by 0x754292: get_prev_frame_always_1 (frame.c:1955)
==17026== by 0x7542DA: get_prev_frame_always (frame.c:1971)
==17026== by 0x7547BE: get_prev_frame (frame.c:2213)
==17026== by 0x7532BD: unwind_to_current_frame (frame.c:1450)
==17026== Address 0xd27b570 is 16 bytes inside a block of size 32 free'd
==17026== at 0x4A07577: free (in /usr/lib64/valgrind/vgpreload_memcheck-amd64-linux.so)
==17026== by 0x54B276: gdb_Py_DECREF (python-internal.h:185)
==17026== by 0x54B298: py_decref (py-utils.c:34)
==17026== by 0x5AB934: do_my_cleanups (cleanups.c:155)
==17026== by 0x5AB9AF: do_cleanups (cleanups.c:177)
==17026== by 0x54B009: pyuw_sniffer (py-unwind.c:606)
==17026== by 0x755DAC: frame_unwind_try_unwinder (frame-unwind.c:105)
==17026== by 0x755EEE: frame_unwind_find_by_frame (frame-unwind.c:160)
==17026== by 0x750FFA: compute_frame_id (frame.c:454)
==17026== by 0x753BD6: get_prev_frame_if_no_cycle (frame.c:1781)
==17026== by 0x754292: get_prev_frame_always_1 (frame.c:1955)
==17026== by 0x7542DA: get_prev_frame_always (frame.c:1971)
==17026==
Simply invalidate the object before releasing it.
gdb/ChangeLog:
2015-08-13 Pedro Alves <palves@redhat.com>
* python/py-unwind.c (pyuw_sniffer): Install the invalidate
cleanup after the decref cleanup, not before.
BuildBot reminded me that "explicit" is a reserved keyword in C++.
This patch simply renames all the (illegal) uses of "explicit". This should
fix the build errors with --enable-build-with-cxx bots.
gdb/ChangeLog
* break-catch-throw.c (re_set_exception_catchpoint) Rename
reserved C++ keyword "explicit" to "explicit_loc".
* breakpoint.c (create_overlay_event_breakpoint)
(create_longjmp_master_breakpoint)
(create_std_terminate_master_breakpoint)
(create_exception_master_breakpoint, update_static_tracepoint):
Rename reserved C++ keyword "explicit" to "explicit_loc".
* completer.c (collect_explicit_location_matches)
(explicit_location_completer): Rename reserved C++ keyword
"explicit" to "explicit_loc".
* linespec.c (struct linespec) <explicit>: Rename to "explicit_loc".
(canonicalize_linespec, create_sals_line_offset)
(convert_linespec_to_sals, convert_explicit_location_to_sals)
(event_location_to_sals, decode_objc): Rename reserved C++ keyword
"explicit" to "explicit_loc".
* location.c (struct event_location) <explicit>: Rename to
"explicit_loc".
(initialize_explicit_location, new_explicit_location)
(explicit_location_to_string_internal, explicit_location_to_linespec):
Rename reserved C++ keyword "explicit" to "explicit_loc".
* location.h (explicit_location_to_string)
(explicit_location_to_linespec, initialize_explicit_location)
(new_explicit_location): Rename reserved C++ keyword "explicit"
to "explicit_loc".
* mi/mi-cmd-break.c (mi_cmd_break_insert_1): Rename reserved C++
keyword "explicit" to "explicit_loc".
To load an ELF binary with DT_TEXTREL tag, the dynamic linker calls
__mprotect on the read-only segment with PROT_READ|PROT_WRITE before
applying dynamic relocation. It leads to segfault when performing
IFUNC relocations since the read-only segment has no execute permission.
This patch changes x86 linker to issue an error for read-only segment
with dynamic IFUNC relocations. Other backends with IFUNC support
may need a similar change.
bfd/
PR ld/18801
* elf32-i386.c (elf_i386_size_dynamic_sections): Issue an error
for read-only segment with dynamic IFUNC relocations.
* elf64-x86-64.c (elf_x86_64_size_dynamic_sections): Likewise.
ld/testsuite/
PR ld/18801
* ld-i386/i386.exp: Run pr18801.
* ld-x86-64/x86-64.exp: Likewise.
* ld-i386/pr18801.d: New file.
* ld-i386/pr18801.s: Likewise.
* ld-x86-64/pr18801.d: Likewise.
* ld-x86-64/pr18801.s: Likewise.
opcodes * arm-dis.c (print_insn_arm): Disassembling for all targets V6
and higher with ARM instruction set will now mark the 26-bit
versions of teq,tst,cmn and cmp as UNPREDICTABLE.
(arm_opcodes): Fix for unpredictable nop being recognized as a teq.
test * gas/arm/nops.d: New.
* gas/arm/nops.s: New.
* gas/arm/inst.d: Changed expectation file for 26-bit teq,
tst, cmn and cmp.
Consider the following declaration:
function Foo (I : Integer) return Integer renames Pack.Bar;
As Foo is not materialized as a routine whose name is derived from Foo,
GDB currently cannot use it:
(gdb) print foo(0)
No definition of "foo" in current context.
However, compilers can emit DW_TAG_imported_declaration in order to
materialize the fact that Foo is actually another name for Pack.Bar.
This commit enhances the DWARF reader to record global renamings (it
used to put global ones in a static block) and enhances the Ada engine
to leverage this information during symbol lookup.
gdb/ChangeLog:
* ada-lang.c: Include namespace.h
(aux_add_nonlocal_symbols): Fix a function name in comment.
(ada_add_block_renamings): New.
(add_nonlocal_symbols): Add global renamings handling.
(ada_lookup_symbol_list_worker): Move the symbol lookup part
to...
(ada_add_all_symbols): ... this new function.
(ada_add_block_symbols): Try to match the input name against the
"using directives list", perform a recursive symbol lookup on
the matched declarations.
* block.h (struct block): Move the_namespace to top-level as
namespace_info. Remove the language_specific field.
(BLOCK_NAMESPACE): Update access to the namespace_info field.
* buildsym.h (using_directives): Rename into...
(local_using_directives): ... this.
(global_using_directives): New.
(struct context_stack): Rename the using_directives field into
local_using_directives.
* buildsym.c (finish_block_internal): Deal with the proper
using directives repository (local or global).
(prepare_for_building): Reset local_using_directives. Assert
that there is no pending global using directive.
(reset_symtab_globals): Reset global_using_directives and
local_using_directives.
(end_symtab_get_static_block): Don't ignore symtabs that have
only using directives.
(push_context): Update references to local_using_directives.
(buildsym_init): Do not reset using_directives.
* cp-support.c: Include namespace.h.
* cp-support.h (struct using_direct): Move to namespace.h.
(cp_add_using_directives): Move to namespace.h.
* cp-namespace.c: Include namespace.h
(cp_add_using_directive): Move to namespace.c, rename it to
add_using_directive, add a "using_directives" argument and use
it as the pending using directives repository. All callers
updated.
* dwarf2read.c (using_directives): New.
(read_import_statement): Call using_directives.
(read_func_scope): Update references to local_using_directives.
(read_lexical_block_scope): Likewise.
(read_namespace): Update the heading comment, call
using_directives.
* namespace.h: New file.
* namespace.c: New file.
* Makefile.in (SFILES): Add namespace.c.
(COMMON_OBS): Add namespace.o
gdb/testsuite/ChangeLog:
* gdb.ada/fun_renaming.exp: New testcase.
* gdb.ada/fun_renaming/fun_renaming.adb: New file.
* gdb.ada/fun_renaming/pack.adb: New file.
* gdb.ada/fun_renaming/pack.ads: New file.
Tested on x86_64-linux. Support for this in GCC is in the pipeline: see
<https://gcc.gnu.org/ml/gcc-patches/2015-07/msg02166.html>.
_start:
.byte 0f-_start
0:
Fixes
..:2: Error: floating point number invalid
..:2: Error: junk at end of line, first unrecognized character is `_'
* expr.c (operand): Rewrite handling of operands starting with "0f".
If atof_generic only parses "-" or "+", treat as expression.
* expr.c (integer_constant): Return O_absent expression if eol.
(operand): For targets with both LOCAL_LABELS_FB and
NUMBERS_WITH_SUFFIX set, treat "0b" not followed by binary
digits as a local label reference. Correct handling of 0b prefix.
If a suffix is not allowed, error on 0B.
Keith reported that gdb.base/dso2dso.exp is broken, with the following
error:
| $ make check RUNTESTFLAGS=dso2dso.exp
| [snip]
| Running ../../../src/gdb/testsuite/gdb.base/dso2dso.exp ...
| ERROR: tcl error sourcing ../../../src/gdb/testsuite/gdb.base/dso2dso.exp.
| ERROR: couldn't open
| "../../../src/gdb/testsuite/gdb.base/../../../src/gdb/testsuite/gdb.base/dso2dso-dso1.c":
| no such file or directory
| while executing
| "error "$message""
| (procedure "gdb_get_line_number" line 14)
| invoked from within
| "gdb_get_line_number "STOP HERE" $srcfile_libdso1"
| (file "../../../src/gdb/testsuite/gdb.base/dso2dso.exp" line 60)
| invoked from within
| "source ../../../src/gdb/testsuite/gdb.base/dso2dso.exp"
| ("uplevel" body line 1)
| invoked from within
| "uplevel #0 source ../../../src/gdb/testsuite/gdb.base/dso2dso.exp"
| invoked from within
| "catch "uplevel #0 source $test_file_name""
This happens because gdb_get_line_number will prepend $srcdir/$subdir
if the given filename does not start with "/", and this happens when
GDB was configured using a relative path to the configure script.
When using an absolute path like I do, we avoid the pre-pending that
Keith is seeing.
gdb/testsuite/ChangeLog:
Keith Seitz <keiths@redhat.com>:
* gdb.base/dso2dso.exp: Pass basename of source file in call
to gdb_get_line_number.
Tested on x86_64-linux with both scenarios.
Since the backend elf_add_symbol_hook isn't called on local symbols,
the EI_OSABI field isn't to ELFOSABI_GNU where are local IFUNC symbols.
This patch changes the x86 backends to set has_gnu_symbols if there are
relocations against IFUNC symbols. Other backends with IFUNC support
may need a similar change.
This patch also changes the type of has_gnu_symbols from bfd_boolean to
enum elf_gnu_symbols.
bfd/
PR ld/18815
* elf-bfd.h (elf_gnu_symbols): New enum.
(elf_obj_tdata): Use elf_gnu_symbols on has_gnu_symbols.
* elf-s390-common.c (elf_s390_add_symbol_hook): Set
has_gnu_symbols to elf_gnu_symbol_any.
* elf32-arm.c (elf32_arm_add_symbol_hook): Likewise.
* elf32-m68k.c (elf_m68k_add_symbol_hook): Likewise.
* elf32-ppc.c (ppc_elf_add_symbol_hook): Likewise.
* elf32-sparc.c (elf32_sparc_add_symbol_hook): Likewise.
* elf64-ppc.c (ppc64_elf_add_symbol_hook): Likewise.
* elf64-sparc.c (elf64_sparc_add_symbol_hook): Likewise.
* lfxx-aarch64.c (_bfd_aarch64_elf_add_symbol_hook): Likewise.
* elf32-i386.c (elf_i386_check_relocs): Update has_gnu_symbols
if there are relocations against IFUNC symbols.
(elf_i386_add_symbol_hook): Don't check STT_GNU_IFUNC here.
* elf64-x86-64. (elf_x86_64_check_relocs): Update has_gnu_symbols
if there are relocations against IFUNC symbols.
(elf_x86_64_add_symbol_hook): Don't check STT_GNU_IFUNC here.
ld/testsuite/
PR ld/18815
* ld-i386/i386.exp: Run pr18815.
* ld-x86-64/x86-64.exp: Likewise.
* ld-i386/pr18815.d: New file.
* ld-i386/pr18815.s: Likewise.
* ld-x86-64/pr18815.d: Likewise.
* ld-x86-64/pr18815.s: Likewise.
Making all-stop run on top of non-stop caused a small regression
in behavior. This was observed on x86_64-linux. The attached testcase
is in C whereas the investigation was done with an Ada program,
but it's the same scenario, and using a C testcase allows wider testing.
Basically: I am debugging a single-threaded program, and currently
stopped inside a function provided by a shared-library, at a line
calling a subprogram provided by a second shared library, and trying
to "next" over that function call.
Before we changed the default all-stop behavior, we had:
7 Impl_Initialize; -- Stop here and try "next" over this line
(gdb) n
8 return 5; <<-- OK
But now, "next" just stops much earlier:
(gdb) n
0x00007ffff7bd8560 in impl.initialize@plt () from /[...]/lib/libpck.so
What happens is that next stops at a call instruction, which calls
the function's PLT, and GDB fails to notice that the inferior stepped
into a subroutine, and so decides that we're done. We can see another
symptom of the same issue by looking at the backtrace at the point
GDB stopped:
(gdb) bt
#0 0x00007ffff7bd8560 in impl.initialize@plt ()
from /[...]/lib/libpck.so
#1 0x00000000f7bd86f9 in ?? ()
#2 0x00007fffffffdf50 in ?? ()
#3 0x0000000000401893 in a () at /[...]/a.adb:7
Backtrace stopped: frame did not save the PC
With a functioning GDB, the backtrace looks like the following instead:
#0 0x00007ffff7bd8560 in impl.initialize@plt ()
from /[...]/lib/libpck.so
#1 0x00007ffff7bd86f9 in sub () at /[...]/pck.adb:7
#2 0x0000000000401893 in a () at /[...]/a.adb:7
Note how, for frame #1, the address looks quite similar, except
for the high-order bits not being set:
#1 0x00007ffff7bd86f9 in sub () at /[...]/pck.adb:7 <<<-- OK
#1 0x00000000f7bd86f9 in ?? () <<<-- WRONG
^^^^
||||
Wrong
Investigating this further led me to displaced stepping.
As we are "next"-ing from a location where a breakpoint is inserted,
we need to step out of it, and since we're on non-stop mode, we need
to do it using displaced stepping. And looking at
amd64-tdep.c:amd64_displaced_step_fixup, I found the code that handles
the return address:
regcache_cooked_read_unsigned (regs, AMD64_RSP_REGNUM, &rsp);
retaddr = read_memory_unsigned_integer (rsp, retaddr_len, byte_order);
retaddr = (retaddr - insn_offset) & 0xffffffffUL;
The mask used to compute retaddr looks wrong to me, keeping only
4 bytes instead of 8, and explains why the high order bits of
the backtrace are unset. What happens is that, after the displaced
stepping has completed, GDB restores that return address at the location
where the program expects it. But because the top half bits of
the address have been masked out, the return address is now invalid.
The incorrect behavior of the "next" command and the backtrace at
that location are the first symptoms of that. Another symptom is
that this actually alters the behavior of the program, where a "cont"
from there soon leads to a SEGV when the inferior tries to jump back
to that incorrect return address:
(gdb) c
Continuing.
Program received signal SIGSEGV, Segmentation fault.
0x00000000f7bd86f9 in ?? ()
^^^^^^^^^^^^^^^^^^
This patch fixes the issue by using a mask that seems more appropriate
for this architecture.
gdb/ChangeLog:
* amd64-tdep.c (amd64_displaced_step_fixup): Fix the mask used to
compute RETADDR.
gdb/testsuite/ChangeLog:
* gdb.base/dso2dso-dso2.c, gdb.base/dso2dso-dso2.h,
gdb.base/dso2dso-dso1.c, gdb.base/dso2dso-dso1.h, gdb.base/dso2dso.c,
gdb.base/dso2dso.exp: New files.
Tested on x86_64-linux, no regression.
Auto-litpools is the automated version of text-section-literals: literal
pool candidate frags are planted every N frags and during relaxation
they are turned into actual literal pools where literals are moved to
become reachable for their first reference by L32R instruction.
2015-08-12 David Weatherford <weath@cadence.com>
gas/
* config/tc-xtensa.c (struct litpool_frag, struct litpool_seg):
New structures.
(xtensa_maybe_create_literal_pool_frag): New function.
(litpool_seg_list, auto_litpools, auto_litpool_limit)
(litpool_buf, litpool_slotbuf): New static variables.
(option_auto_litpools, option_no_auto_litpools)
(option_auto_litpool_limit): New enum identifiers.
(md_longopts): Add entries for auto-litpools, no-auto-litpools
and auto-litpool-limit.
(md_parse_option): Handle option_auto_litpools,
option_no_auto_litpools and option_auto_litpool_limit.
(md_show_usage): Add help for --[no-]auto-litpools and
--auto-litpool-limit.
(xtensa_mark_literal_pool_location): Record a place for literal
pool with a call to xtensa_maybe_create_literal_pool_frag.
(get_literal_pool_location): Find highest priority literal pool
or convert candidate to literal pool when auto-litpools are used.
(xg_assemble_vliw_tokens): Create literal pool after jump
instruction.
(xtensa_check_frag_count): Create candidate literal pool every
auto_litpool_limit frags.
(xtensa_relax_frag): Add jump around literals to non-empty
literal pool.
(xtensa_move_literals): Estimate literal pool addresses and move
unreachable literals closer to their users, converting candidate
to literal pool if needed.
(xtensa_switch_to_non_abs_literal_fragment): Only emit error
about missing .literal_position in case auto-litpools are not
used.
* config/tc-xtensa.h (xtensa_relax_statesE): New relaxation
state: RELAX_LITERAL_POOL_CANDIDATE_BEGIN.
* doc/as.texinfo (Xtensa options): Document --auto-litpools and
--no-auto-litpools options.
* doc/c-xtensa.texi (Xtensa options): Likewise.
2015-08-12 Max Filippov <jcmvbkbc@gmail.com>
gas/testsuite/
* gas/xtensa/all.exp: Add auto-litpools to the list of xtensa
tests.
* gas/xtensa/auto-litpools.s: New file: auto-litpools test.
* gas/xtensa/auto-litpools.s: New file: auto-litpools test
result pattern.
Keith found out that several tests were failing when testing the
native-gdbserver board on Fedora (x86_64). Strangely, these failures
had not been reported by our BuildBot. Later, he found that the reason
for this was because the failures only happened when running the
testsuite without FORCE_PARALLEL (i.e., on serial mode; maybe it would
be worth having a builder testing things on serial...). Then, he
decided to start bisecting the changes to see which one introduced the
failure (it was not trivial to know this only by looking at gdb.log).
After a lot of time, he found that Pedro's commit
e1316e60d4 was the culprit. There was
nothing wrong in the code, but the new gdb.base/checkpoint-ns.exp
testcase did something that left the GDBFLAGS variable in an
inconsistent state. This test works by modifying this variable to set
non-stop on, sourcing gdb.base/checkpoint.exp (which does the hard
work), and then restoring the old value on GDBFLAGS. However, this was
not working because gdb.base/checkpoint.exp bails out if it is being
tested on gdbserver, and when it calls "continue" the control goes back
to the function calling the tests, and not to
gdb.base/checkpoint-ns.exp.
The fix is simple: just wrap the "source" call, and make
gdb.base/checkpoint-ns.exp aware of the "continue"/"return" calls made
by gdb.base/checkpoint.exp.
gdb/testsuite/ChangeLog:
2015-08-12 Sergio Durigan Junior <sergiodj@redhat.com>
Pedro Alves <palves@redhat.com>
Keith Seitz <keiths@redhat.com>
* gdb.base/checkpoint-ns.exp: Use save_vars to save and restore
GDBFLAGS.
