I don't see any particular reason why the implementations of the
frame_info_ptr object are in the header file. It only seems to add some
complexity. Since we can't include frame.h in frame-info.h, we have to
add declarations of functions defined in frame.c, in frame-info.h. By
moving the implementations to a new frame-info.c, we can avoid that.
Change-Id: I435c828f81b8a3392c43ef018af31effddf6be9c
Reviewed-By: Bruno Larsen <blarsen@redhat.com>
Reviewed-By: Tom Tromey <tom@tromey.com>
I noticed this crash:
$ ./gdb --data-directory=data-directory -nx -q \
testsuite/outputs/gdb.python/pretty-print-call-by-hand/pretty-print-call-by-hand \
-x testsuite/outputs/gdb.python/pretty-print-call-by-hand/pretty-print-call-by-hand.py \
-ex "b g" -ex r
(gdb) info frame
Stack level 0, frame at 0x7fffffffdd80:
rip = 0x555555555160 in g
(/home/simark/src/binutils-gdb/gdb/testsuite/gdb.python/pretty-print-call-by-hand.c:41); saved rip = 0x5555555551a3
called by frame at 0x7fffffffdda0
source language c.
Arglist at 0x7fffffffdd70, args: mt=mytype is 0x555555556004 "hello world",
depth=10
Fatal signal: Segmentation fault
This is another case of frame_info being invalidated under a function's
feet. The stack trace when the frame_info get invalidated looks like:
... many frames to pretty print the arg, that eventually invalidate the frame_infos ...
#35 0x00005568d0a8ab24 in print_frame_arg (fp_opts=..., arg=0x7ffc3216bcb0) at /home/simark/src/binutils-gdb/gdb/stack.c:489
#36 0x00005568d0a8cc75 in print_frame_args (fp_opts=..., func=0x621000233210, frame=..., num=-1, stream=0x60b000000300)
at /home/simark/src/binutils-gdb/gdb/stack.c:898
#37 0x00005568d0a9536d in info_frame_command_core (fi=..., selected_frame_p=true) at /home/simark/src/binutils-gdb/gdb/stack.c:1682
print_frame_args knows that print_frame_arg can invalidate frame_info
objects, and therefore calls prepare_reinflate/reinflate. However,
info_frame_command_core has a separate frame_info_ptr instance (it is
passed by value / copy). So info_frame_command_core needs to know that
print_frame_args can invalidate frame_info objects, and therefore needs
to prepare_reinflate/reinflate as well. Add those calls, and enhance
the gdb.python/pretty-print-call-by-hand.exp test to test that command.
Reviewed-By: Bruno Larsen <blarsen@redhat.com>
Change-Id: I9edaae06d62e97ffdb30938d364437737238a960
We do it in the move assignment operator, so I think it makes sense to
do it here too for consistency. I don't think it's absolutely necessary
to clear the other object's fields (in other words, copy constructor and
move constructor could be the same), as there is no exclusive resource
being transfered. The important thing is to leave the moved-from object
in an unknown, but valid state. But still, I think that clearing the
fields of the moved-from object is not a bad idea, it helps ensure we
don't rely on the moved-from object after.
Change-Id: Iee900ff9d25dad51d62765d694f2e01524351340
Reviewed-By: Bruno Larsen <blarsen@redhat.com>
When resolving overloaded functions, GDB relies on knowing relationships
between types, i.e. if a type inherits from another. However, some
compilers may not add complete information for given types as a way to
reduce unnecessary debug information. In these cases, GDB would just say
that it couldn't resolve the method or function, with no extra
information.
The problem is that sometimes the user may not know that the type
information is incomplete, and may just assume that there is a bug in
GDB. To improve the user experience, we attempt to detect if the
overload match failed because of an incomplete type, and warn the user
of this.
This commit also adds a testcase confirming that the message is only
triggered in the correct scenario. This test was not developed as an
expansion of gdb.cp/overload.cc because it needed the dwarf assembler,
and porting all of overload.cc seemed unnecessary.
Approved-By: Tom Tromey <tom@tromey.com>
When calling get_func_info inside a test case, it would cause failures
if the function was printed using a C++ style mangled name. The current
patch fixes this by allowing for mangled names along with the current
rules.
Approved-By: Tom Tromey <tom@tromey.com>
The idea here is the stop tools from allocating up to 32G per section
for the arelent pointer array, only to find a little later that the
section reloc count was fuzzed. This usually doesn't hurt much (on
systems that allow malloc overcommit) except when compiled with asan.
We already do this for ELF targets, and while fixing the logic
recently I decided other targets ought to do the same.
* elf64-sparc.c (elf64_sparc_get_reloc_upper_bound): Sanity check
section reloc count against file size.
* mach-o.c (bfd_mach_o_get_reloc_upper_bound): Likewise.
* aoutx.h (get_reloc_upper_bound): Likewise, and don't duplicate
check done in bfd_get_reloc_upper_bound.
* pdp11.c (get_reloc_upper_bound): Likewise.
* coffgen.c (coff_get_reloc_upper_bound): Likewise.
The test case introduced in bafcc33526 (Fix stepping in rtld without
debug symbol) fails on some systems as reported by PR/29768. This can
be seen if the system does not have debug info for the libc:
(gdb) step^M
Single stepping until exit from function main,^M
which has no line number information.^M
hello world[Inferior 1 (process 48203) exited normally]^M
(gdb) PASS: gdb.base/rtld-step-nodebugsym.exp: step
continue^M
The program is not being run.^M
(gdb) FAIL: gdb.base/rtld-step-nodebugsym.exp: continue until exit (the program is no longer running)
Without glibc debug info, GDB steps until the program finishes, and
then "gdb_continue_to_end" fails.
As this test was designed to check that GDB does not crash in the "step"
command, the continue does not carry real meaning to the test.
Replace it by "print 0" so we still check that after the step command
GDB is still alive, which is what we care about.
Tested on Ubuntu-22.04 x86_64, with and without libc6-dbg.
Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=29768
Approved-By: Simon Marchi <simon.marchi@efficios.com>
Nothing passes this to dgen, and even if it did, nothing would happen
because the generated spreg.[ch] files don't include any references
back to the original data table. So drop it to simplify.
Since we know we'll return 0 by default, we don't have to output case
statements for readonly or length fields whose values are also zero.
This is the most common case by far and thus generates a much smaller
switch table in the end.
Instead of writing:
case 1:
return 1;
case 2:
return 1;
...etc...
Output a single return so we get:
case 1:
case 2:
case ...
return 1;
This saves ~100 lines of code. Hopefully the compiler was already
smart enough to optimize to the same code, but if not, this probably
helps there too :).
This saves a single line for the same result. By itself, it's not
interesting, but we can further optimize the generated output and
completely omit the switch table in some cases. Which we'll do in
follow up commits.
PR tdep/28796
As reported, we are using some memory read routines that don't handle read
errors gracefully. Convert those to use the safe_* versions if available.
This allows the code to handle those read errors in a more sensible way.
Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=28796
Commit be6276e0ae "Allow debugging of runtime loader / dynamic linker"
introduced a small regression when stepping into the runtime loader /
dynamic linker from function we do not have debug information for. This
is reported in PR/29747.
This can be shown by the following example (given by Simon Marchi in
buzilla bug report):
$ cat test.c
#include <stdio.h>
int main()
{
printf("Hi\n");
return 0;
}
$ gcc test.c -O0 -o test
$ ./gdb -q -nx --data-directory=data-directory test -ex start -ex s
Reading symbols from test...
(No debugging symbols found in test)
Temporary breakpoint 1 at 0x1151
Starting program: .../binutils-gdb/gdb/test
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
Temporary breakpoint 1, 0x0000555555555151 in main ()
Single stepping until exit from function main,
which has no line number information.
/home/smarchi/src/binutils-gdb/gdb/infrun.c:6960:64: runtime error: member call on null pointer of type 'struct symbol'
The crash happens here:
#0 __sanitizer::Die () at ../../../../src/libsanitizer/sanitizer_common/sanitizer_termination.cpp:50
#1 0x00007ffff5dd7128 in __ubsan::__ubsan_handle_type_mismatch_v1_abort (Data=<optimized out>, Pointer=<optimized out>) at ../../../../src/libsanitizer/ubsan/ubsan_handlers.cpp:148
#2 0x000055556183e1a7 in process_event_stop_test (ecs=0x7fffffffccd0) at .../binutils-gdb/gdb/infrun.c:6960
#3 0x0000555561838ea4 in handle_signal_stop (ecs=0x7fffffffccd0) at .../binutils-gdb/gdb/infrun.c:6615
#4 0x000055556182f77b in handle_inferior_event (ecs=0x7fffffffccd0) at .../binutils-gdb/gdb/infrun.c:5866
When evaluating:
6956 if (execution_direction != EXEC_REVERSE
6957 && ecs->event_thread->control.step_over_calls == STEP_OVER_UNDEBUGGABLE
6958 && in_solib_dynsym_resolve_code (ecs->event_thread->stop_pc ())
6959 && !in_solib_dynsym_resolve_code (
6961 ecs->event_thread->control.step_start_function->value_block ()
6962 ->entry_pc ()))
we dereference, ecs->event_thread->control.step_start_function which is
nullptr.
This patch changes this condition so it evaluates to true if
ecs->event_thread->control.step_start_function is nullptr since this
matches the behaviour before be6276e0ae.
Tested on ubuntu-22.04 x86_64.
Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=29747
Reviewed-By: Bruno Larsen <blarsen@redhat.com>
Approved-By: Kevin Buettner <kevinb@redhat.com>
When merging ppc configure checks into the top-level, these 2 funcs
were accidentally dropped (probably due to incorrect resolution of
conflicts). Restore them since the ppc code utilizes them both.
This controls only one thing: how to call mkdir(). The gnulib code
already has a mkdir module that provides this exact logic for us, so
punt the code entirely.
Gdbserver unconditionally reports support for btrace packets. Do not
report the support, if the underlying target does not say it supports
it. Otherwise GDB would query the server with btrace-related packets
unnecessarily.
PR exp/28359 points out that 'ptype/o' does not work when the current
language is "asm".
I tracked this down to a hard-coded list of languages in typeprint.c.
This patch replaces this list with a method on 'language_defn'
instead. If all languages are ever updated to have this feature, the
method could be removed; but in the meantime this lets each language
control what happens.
I looked at having each print_type method simply modify the flags
itself, but this doesn't work very well with the feature that disables
method-printing by default (but allows it via a flag).
Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=28359
Approved-By: Andrew Burgess <aburgess@redhat.com>
Approved-By: Keith Seitz <keiths@redhat.com>
This macro expansion was missing a set of outer-most parenthesis which
some compilers would complain about depending on how the macro is used.
This is just standard good macro hygiene too.
We've never run these helper programs directly. The igen program
includes the relevant source files directly and runs the code that
way. So stop wasting developer CPU time linking programs that are
never run. We leave the rules in place for people who need to test
and debug the specific bits of code every now & then.
Operand swapping was mistakenly suppressed when the first two operands
were immediate ones, not taking into account overall operand count. This
way EXTRQ / INSERTQ would have been accepted also with kind-of-AT&T
operand order.
For the testcase being extended, in order to not move around "GAS
LISTING" expectations, suppress pagination.
Like commit ffbe89531c this avoids more silliness writing output
that is going to be deleted. bfd_close and bfd_close_all_done differ
in that only the former calls _bfd_write_contents.
* objcopy.c (copy_archive): Don't call bfd_close for elements
that are going to be deleted, call bfd_close_all_done instead.
Do the same for the archive itself.
Although the encoding for scalar and fp registers is identical,
we should follow common pratice and use fp register names
when referencing fp registers.
The xtheadmemidx extension consists of indirect load/store instructions
which all load to or store from fp registers.
Let's use fp register names in this case and adjust the test cases
accordingly.
gas/
* testsuite/gas/riscv/x-thead-fmemidx-fail.l: Updated since rd need to
be float register.
* testsuite/gas/riscv/x-thead-fmemidx-fail.s: Likewise.
* testsuite/gas/riscv/x-thead-fmemidx.d: Likewise.
* testsuite/gas/riscv/x-thead-fmemidx.s: Likewise.
opcodes/
* riscv-opc.c (riscv_opcodes): Updated since rd need to be float register.
Signed-off-by: Christoph Müllner <christoph.muellner@vrull.eu>
I get this test failure on my CI;
FAIL: gdb.base/info-os.exp: get process list
The particularity of this setup is that builds are done in containers
who are allocated 4 CPUs on a machine that has 40. The code in
nat/linux-osdata.c fails to properly fetch the core number for each
task.
linux_xfer_osdata_processes uses `sysconf (_SC_NPROCESSORS_ONLN)`, which
returns 4, so it allocates an array of 4 integers. However, the core
numbers read from /proc/pid/task/tid/stat, by function
linux_common_core_of_thread, returns a value anywhere between 0 and 39.
The core numbers above 3 are therefore ignored, many processes end up
with no core value, and the regexp in the test doesn't match (it
requires an integer as the core field).
The way this the CPUs are exposed to the container is that the container
sees 40 CPUs "present" and "possible", but only 4 arbitrary CPUs
actually online:
root@ci-node-jammy-amd64-04-08:~# cat /sys/devices/system/cpu/present
0-39
root@ci-node-jammy-amd64-04-08:~# cat /sys/devices/system/cpu/online
5,11,24,31
root@ci-node-jammy-amd64-04-08:~# cat /sys/devices/system/cpu/possible
0-39
The solution proposed in this patch is to find out the number of
possible CPUs using /sys/devices/system/cpu/possible. In practice, this
will probably always contain `0-N`, where N is the number of CPUs, minus
one. But the documentation [1] doesn't such guarantee, so I'll assume
that it can contain a more complex range list such as `2,4-31,32-63`,
like the other files in that directory can have. The solution is to
iterate over these numbers to find the highest possible CPU id, and
use that that value plus one as the size of the array to allocate.
[1] https://www.kernel.org/doc/Documentation/admin-guide/cputopology.rst
Change-Id: I7abce2e43b000c1327fa94cd7b99d46e49d7ccf3
I would like to add more code to nat/linux-osdata.c that reads an entire
file from /proc or /sys and processes it as a string afterwards. I
would like to avoid duplicating the somewhat error-prone code that reads
an entire file to a buffer. I think we should have a utility function
that does that.
Add read_file_to_string to gdbsupport/filestuff.{c,h}, and make
linux_common_core_of_thread use it. I want to make the new function
return an std::string, and because strtok doesn't play well with
std::string (it requires a `char *`, std::string::c_str returns a `const
char *`), change linux_common_core_of_thread to use std::string methods
instead.
Approved-By: Tom Tromey <tom@tromey.com>
Change-Id: I1793fda72a82969c28b944a84acb953f74c9230a
Consider a hello world a.out, started using gdbserver:
...
$ gdbserver --once 127.0.0.1:2345 ./a.out
Process ./a.out created; pid = 15743
Listening on port 2345
...
that we can connect to using gdb:
...
$ gdb -ex "target remote 127.0.0.1:2345"
Remote debugging using 127.0.0.1:2345
Reading /home/vries/a.out from remote target...
...
0x00007ffff7dd4550 in _start () from target:/lib64/ld-linux-x86-64.so.2
(gdb)
...
After that, we can for instance quit with confirmation:
...
(gdb) quit
A debugging session is active.
Inferior 1 [process 16691] will be killed.
Quit anyway? (y or n) y
$
...
Or, kill with confirmation and quit:
...
(gdb) kill
Kill the program being debugged? (y or n) y
[Inferior 1 (process 16829) killed]
(gdb) quit
$
...
Or, monitor exit, kill with confirmation, and quit:
...
(gdb) monitor exit
(gdb) kill
Kill the program being debugged? (y or n) y
Remote connection closed
(gdb) quit
$
...
But when doing monitor exit followed by quit with confirmation, we get the gdb
prompt back, requiring us to do quit once more:
...
(gdb) monitor exit
(gdb) quit
A debugging session is active.
Inferior 1 [process 16944] will be killed.
Quit anyway? (y or n) y
Remote connection closed
(gdb) quit
$
...
So, the first quit didn't quit. This happens as follows:
- quit_command calls query_if_trace_running
- a TARGET_CLOSE_ERROR is thrown
- it's caught in remote_target::get_trace_status, but then
rethrown because it's TARGET_CLOSE_ERROR
- catch_command_errors catches the error, at which point the quit command
has been aborted.
The TARGET_CLOSE_ERROR is defined as:
...
/* Target throwing an error has been closed. Current command should be
aborted as the inferior state is no longer valid. */
TARGET_CLOSE_ERROR,
...
so in a way this is expected behaviour. But aborting quit because the inferior
state (which we've already confirmed we're not interested in) is no longer
valid, and having to type quit again seems pointless.
Furthermore, the purpose of not catching errors thrown by
query_if_trace_running as per commit 2f9d54cfce ("make -gdb-exit call
disconnect_tracing too, and don't lose history if the target errors on
"quit""), was to make sure that error (_("Not confirmed.") had effect.
Fix this in quit_command by catching only the TARGET_CLOSE_ERROR exception
during query_if_trace_running and reporting it:
...
(gdb) monitor exit
(gdb) quit
A debugging session is active.
Inferior 1 [process 19219] will be killed.
Quit anyway? (y or n) y
Remote connection closed
$
...
Tested on x86_64-linux.
PR server/15746
Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=15746
Approved-By: Tom Tromey <tom@tromey.com>
Remove test-cases from test-names, such that we don't have the redundant:
...
PASS: gdb.base/corefile.exp: backtrace in corefile.exp
...
but simply:
...
PASS: gdb.base/corefile.exp: backtrace
...
Fixed all instances found using:
...
$ grep ":.*:.*\.exp" gdb.sum
...
Tested on x86_64-linux.
With test-case gdb.base/bigcore.exp I run into:
...
(gdb) PASS: gdb.base/bigcore.exp: get inferior pid
signal SIGABRT^M
Continuing with signal SIGABRT.^M
^M
Program terminated with signal SIGABRT, Aborted.^M
The program no longer exists.^M
(gdb) PASS: gdb.base/bigcore.exp: signal SIGABRT
UNTESTED: gdb.base/bigcore.exp: can't generate a core file
...
due to find_core_file returning "".
There is a core file name core:
...
$ ls ./outputs/gdb.base/bigcore
bigcore bigcore.corefile core gdb.cmd.1 gdb.in.1 gdbserver.cmd.1
...
but it's not found.
The problem is this statement:
...
lappend files [list ${::testfile}.core core]
...
which adds a single list item "${::testfile}.core core".
Fix this in the most readable way:
...
lappend files ${::testfile}.core
lappend files core
...
Tested on x86_64-linux.
The intention of this code seems to be to indicate that this insn
should not be used and produces undefined behavior, so instead of
setting registers to bogus values, call Unpredictable. This fixes
build warnings due to 32-bit/64-bit type conversions, and outputs
a log message for users at runtime instead of silent corruption.
Bug: https://sourceware.org/PR29276
This hasn't been used by gdb in decades, and doesn't make sense with
a standalone sim program/library where the ABI is fixed. So punt it
to simplify the code.
Hi all,
This wrong comment was introduced by previous AVX-VNNI-INT8 commit.
Committed as obvious fix.
BRs,
Haochen
opcodes/ChangeLog:
* i386-dis.c (VEX_W_0F3851): Corrected from
VEX_W_0F3851_P_0.
The switch to linking with libtool now shows a very long link line
even when V=0. This patch arranges to silence libtool in this
situation.
Approved-By: Simon Marchi <simon.marchi@efficios.com>
We've been using this only to set the default word size to 32. We
can easily move this into the makefile via a -D compiler flag and
clean up the build logic quite a bit.
We've been using this only to set the default word size to 32. We
can easily move this into the makefile via a -D compiler flag and
clean up the build logic quite a bit.
We've been using this only to set the default word size to 32. We
can easily move this into the makefile via a -D compiler flag and
clean up the build logic quite a bit.
We've been using this only to set the default word size to 64. We
can easily move this into the makefile via a -D compiler flag and
clean up the build logic quite a bit.
