Tom de Vries [Wed, 29 Mar 2023 06:01:42 +0000 (08:01 +0200)]
[gdb/testsuite] Fix gdb.guile/scm-objfile-script.exp for remote host
Fix test-case gdb.guile/scm-objfile-script.exp using gdb_remote_download.
Tested on x86_64-linux.
Tom de Vries [Wed, 29 Mar 2023 06:01:42 +0000 (08:01 +0200)]
[gdb/testsuite] Fix gdb.guile/scm-objfile-script.exp for remote host
Fix test-case gdb.guile/scm-objfile-script.exp using host_standard_output_file.
Tested on x86_64-linux.
Tom de Vries [Wed, 29 Mar 2023 06:01:42 +0000 (08:01 +0200)]
[gdb/testsuite] Fix gdb.guile/scm-cmd.exp without readline
Fix test-case gdb.guile/scm-cmd.exp using readline_is_used.
Tested on x86_64-linux.
Tom de Vries [Wed, 29 Mar 2023 06:01:42 +0000 (08:01 +0200)]
[gdb/testsuite] Fix gdb.guile/guile.exp for remote host
Fix test-case gdb.guile/guile.exp for remote host using gdb_remote_download.
Tested on x86_64-linux.
Alan Modra [Wed, 29 Mar 2023 02:15:16 +0000 (12:45 +1030)]
Sanity check section size in bfd_init_section_compress_status
This function doesn't just initialise for compression, it actually
compresses. This patch sanity checks section size before allocating
buffers for the uncompressed contents.
* compress.c (bfd_init_section_compress_status): Sanity check
section size.
Alan Modra [Tue, 28 Mar 2023 22:43:17 +0000 (09:13 +1030)]
Re: Fix an aout memory leak
We have way too much duplicated code in bfd. Apply
dd3a3d0af9f6 and
920581c57e08 to pdp11.c.
* pdp11.c (bfd_free_cached_info): Free line_buf. Return true
if tdata.aout_data is NULL.
Alan Modra [Tue, 28 Mar 2023 10:32:36 +0000 (21:02 +1030)]
ld testsuite CFLAGS_FOR_TARGET
run_host_cmd adds $gcc_B_opt and $ld_L_opt to the command line if it
detects the program being run is a compiler. Since the program being
run in lto.exp linking pr28138 is "sh", we need to add these by hand.
This isn't exactly as run_host_cmd does, as it lacks reordering of
any user -B option in $CC_FOR_TARGET, but it's better than ignoring
gcc_B_opt. This fixes a mips64 testsuite fail.
ld_compile adds CFLAGS_FOR_TARGET and other flags as well, so there
is no need for the ld_compile command line to include
CFLAGS_FOR_TARGET. Fixing this is just a tidy.
* testsuite/ld-plugin/lto.exp: Add gcc_B_opt, CFLAGS_FOR_TARGET
and $ld_L_opt to pr28138 link line.
* testsuite/lib/ld-lib.exp (run_ld_link_tests): Don't pass
unnecessary flags to ld_compile.
(run_ld_link_exec_tests, run_cc_link_tests): Likewise.
GDB Administrator [Wed, 29 Mar 2023 00:00:11 +0000 (00:00 +0000)]
Automatic date update in version.in
Tom Tromey [Mon, 20 Mar 2023 22:03:05 +0000 (16:03 -0600)]
Rename "raw" to "unrelocated"
Per an earlier discussion, this patch renames the existing "raw" APIs
to use the word "unrelocated" instead.
Tom Tromey [Sun, 19 Mar 2023 16:24:27 +0000 (10:24 -0600)]
Use unrelocated_addr in minimal symbols
This changes minimal symbols to use unrelocated_addr. I believe this
detected a latent bug in add_pe_forwarded_sym.
Tom Tromey [Sun, 19 Mar 2023 15:51:32 +0000 (09:51 -0600)]
Use unrelocated_addr in psymbols
This changes psymbols themselves to use unrelocated_addr. This
transform is largely mechanical. I don't think it finds any bugs.
Tom Tromey [Fri, 17 Mar 2023 22:37:15 +0000 (16:37 -0600)]
Use unrelocated_addr in partial symbol tables
This changes partial symbol tables to use unrelocated_addr for the
text_high and text_low members. This revealed some latent bugs in
ctfread.c, which are fixed here.
Tom Tromey [Sun, 19 Mar 2023 15:57:48 +0000 (09:57 -0600)]
Move definition of unrelocated_addr earlier
This moves the definition of unrelocated_addr a bit earlier in
symtab.h, so that it can be used elsewhere in the file.
Tom Tromey [Sun, 19 Mar 2023 22:12:58 +0000 (16:12 -0600)]
Use function_view in gdb_bfd_lookup_symbol
This changes gdb_bfd_lookup_symbol to use a function_view. This
simplifies the code a little bit.
Tom de Vries [Tue, 28 Mar 2023 20:15:18 +0000 (22:15 +0200)]
[gdb/testsuite] Fix gdb.btrace/multi-inferior.exp for remote host
Fix test-case gdb.btrace/multi-inferior.exp for remote host using
gdb_remote_download.
Tested on x86_64-linux.
Tom de Vries [Tue, 28 Mar 2023 20:15:18 +0000 (22:15 +0200)]
[gdb/testsuite] Fix gdb.btrace/gcore.exp for remote host
Fix test-case gdb.btrace/gcore.exp for remote host using
host_standard_output.
Tested on x86_64-linux.
Tom de Vries [Tue, 28 Mar 2023 20:15:18 +0000 (22:15 +0200)]
[gdb/testsuite] Fix gdb.btrace/reconnect.exp for remote target
Fix test-case gdb.btrace/reconnect.exp for target board
remote-gdbserver-on-localhost using gdb_remote_download.
Tested on x86_64-linux.
Tom Tromey [Thu, 26 May 2022 14:33:05 +0000 (08:33 -0600)]
Put pretty-printers to_string output in varobj result
PR mi/11335 points out that an MI varobj will not display the result
of a pretty-printer's "to_string" method. Instead, it always shows
"{...}".
This does not seem very useful, and there have been multiple
complaints about it over the years. This patch changes varobj to emit
this string when possible, and updates the test suite.
Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=11335
Simon Marchi [Tue, 28 Mar 2023 12:23:05 +0000 (08:23 -0400)]
gdb/testsuite: allow "require" callbacks to provide a reason
When an allow_* proc returns false, it can be a bit difficult what check
failed exactly, if the procedure does multiple checks. To make
investigation easier, I propose to allow the "require" callbacks to be
able to return a list of two elements: the zero/non-zero value, and a
reason string.
Use the new feature in allow_hipcc_tests to demonstrate it (it's also
where I hit actually hit this inconvenience). On my computer (where GDB
is built with amd-dbgapi support but where I don't have a suitable GPU
target), I get:
UNSUPPORTED: gdb.rocm/simple.exp: require failed: allow_hipcc_tests (no suitable amdgpu targets found)
vs before:
UNSUPPORTED: gdb.rocm/simple.exp: require failed: allow_hipcc_tests
Change-Id: Id1966535b87acfcbe9eac99f49dc1196398c6578
Approved-By: Tom de Vries <tdevries@suse.de>
Tom de Vries [Tue, 28 Mar 2023 15:48:35 +0000 (17:48 +0200)]
[gdb/testsuite] Fix gdb.server/server-kill-python.exp for remote host
Fix test-case gdb.server/server-kill-python.exp for remote host using
gdb_remote_download.
Tested on x86_64-linux.
Tom de Vries [Tue, 28 Mar 2023 15:48:35 +0000 (17:48 +0200)]
[gdb/testsuite] Fix gdb.server/sysroot.exp for remote host
Fix test-case gdb.server/sysroot.exp for remote host, by:
- using gdb_remote_download, and
- disabling the "local" scenario for remote host/target, unless
remote host == remote target.
Tested on x86_64-linux.
Tom de Vries [Tue, 28 Mar 2023 15:48:35 +0000 (17:48 +0200)]
[gdb/testsuite] Require non-remote host for gdb.server/multi-ui-errors.exp
Require non-remote host for test-case gdb.server/multi-ui-errors.exp, because
it uses "spawn -pty", which creates a pty on build, which gdb cannot use on
remote host.
Tested on x86_64-linux.
Tom de Vries [Tue, 28 Mar 2023 15:48:35 +0000 (17:48 +0200)]
[gdb/testsuite] Fix gdb.server/solib-list.exp for remote host
Fix test-case gdb.server/solib-list.exp for remote host using
gdb_remote_download.
Likewise in another test-case.
Tested on x86_64-linux.
Tom de Vries [Tue, 28 Mar 2023 15:48:35 +0000 (17:48 +0200)]
[gdb/testsuite] Fix gdb.server/file-transfer.exp for remote host
Fix test-case gdb.server/file-transfer.exp for remote host using
gdb_remote_download and host_standard_output_file.
Tested on x86_64-linux.
Tom de Vries [Tue, 28 Mar 2023 15:48:34 +0000 (17:48 +0200)]
[gdb/testsuite] Fix local-remote-host-native.exp for gdb.server tests
When running test-case gdb.server/stop-reply-no-thread-multi.exp with
host+target board local-remote-host-native, I run into a time-out:
...
(gdb) PASS: gdb.server/stop-reply-no-thread-multi.exp: target-non-stop=off: \
to_disable=: disconnect
builtin_spawn /usr/bin/ssh -t -l vries 127.0.0.1 gdbserver --once \
localhost:2346 stop-reply-no-thread-multi^M
Process stop-reply-no-thread-multi created; pid = 32600^M
Listening on port 2346^M
set remote threads-packet off^M
FAIL: gdb.server/stop-reply-no-thread-multi.exp: target-non-stop=off: \
to_disable=: set remote threads-packet off (timeout)
...
This is due to this line in ${board}_spawn:
...
set board_info($board,fileid) $spawn_id
...
We have the following series of events:
- gdb is spawned, setting fileid
- a few gdb commands (set height etc) are send using fileid, arrive at gdb and
are successful
- gdbserver is spawned, overwriting fileid
- the next gdb command is sent using fileid, so it's send
to gdbserver instead of gdb, and we run into the timeout.
There is some notion of current gdb, tracked in both gdb_spawn_id and fileid
of the host board (see switch_gdb_spawn_id). And because the host and target
board are the same, spawning something on the target overwrites the fileid on
host, and consequently the current gdb.
Fix this by only setting fileid when spawning gdb.
Tested on x86_64-linux.
Now gdb.server/*.exp passes for host+target board local-remote-host-native,
except for file-transfer.exp.
Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=29734
Enze Li [Wed, 1 Feb 2023 13:26:11 +0000 (21:26 +0800)]
gdb: use dynamic year in update-freebsd.sh
When running update-freebsd.sh on FreeBSD, I see the following
modification in freebsd.xml,
-<!-- Copyright (C) 2009-2023 Free Software Foundation, Inc.
+<!-- Copyright (C) 2009-2020 Free Software Foundation, Inc.
It means that each time, when we running the update-freebsd.sh on
FreeBSD, we have to correct the year of copyright manually. So fix this
issue by using dynamic year.
Tested by regenerating freebsd.xml on FreeBSD/amd64.
Reviewed-By: John Baldwin <jhb@FreeBSD.org>
Approved-By: Simon Marchi <simon.marchi@efficios.com>
Tom de Vries [Tue, 28 Mar 2023 12:18:44 +0000 (14:18 +0200)]
[gdb/testsuite] Fix gdb.server/non-existing-program.exp with remote-gdbserver-on-localhost
With test-case gdb.server/non-existing-program.exp and native, I have reliably:
...
(gdb) builtin_spawn gdbserver stdio non-existing-program^M
stdin/stdout redirected^M
/bin/bash: line 0: exec: non-existing-program: not found^M
During startup program exited with code 127.^M
Exiting^M
PASS: gdb.server/non-existing-program.exp: gdbserver exits cleanly
...
But with target board remote-gdbserver-on-localhost I sometimes have:
...
(gdb) builtin_spawn /usr/bin/ssh -t -l remote-target localhost gdbserver \
stdio non-existing-program^M
stdin/stdout redirected^M
/bin/bash: line 0: exec: non-existing-program: not found^M
During startup program exited with code 127.^M
Exiting^M
Connection to localhost closed.^M^M
PASS: gdb.server/non-existing-program.exp: gdbserver exits cleanly
...
and sometimes the exact same output, but a FAIL instead.
Fix this by replacing "Exiting\r\n$" with "Exiting\r\n" in the regexps.
Tested on x86_64-linux.
Alan Modra [Tue, 28 Mar 2023 09:55:26 +0000 (20:25 +1030)]
Avoid undefined behaviour in m68hc11 md_begin
Given p = A where p is a pointer to some type and A is an array of
that type, then the expression p - 1 + 1 evokes undefined behaviour
according to the C standard.
gcc-13 -fsanitize=address,undefined complains about this, but not
where the undefined behaviour actually occurs at tc-m68hc11.c:646.
Instead you get an error: "tc-m68hc11.c:708:20: runtime error: store
to address 0x62600000016c with insufficient space for an object of
type 'int'". Which is a lie. There most definitely is space there.
Oh well, diagnostics are sometimes hard to get right. The UB is easy
to avoid.
PR 30279
* config/tc-m68hc11.c (md_begin): Avoid undefined pointer
decrement. Remove unnecessary cast.
Tom de Vries [Tue, 28 Mar 2023 08:22:48 +0000 (10:22 +0200)]
[gdb/testsuite] Allow gdb.rust/expr.exp without rust compiler
Proc allow_rust_tests returns 0 when there's no rust compiler, but that gives
the wrong answer for gdb.rust/expr.exp, which doesn't require it.
Fix this by using can_compile rust in the test-cases that need it, and just
returning 1 in allow_rust_tests.
Tested on x86_64-linux.
Tom de Vries [Tue, 28 Mar 2023 08:22:48 +0000 (10:22 +0200)]
[gdb/testsuite] Add can_compile rust
If I deinstall the rust compiler, I get:
...
gdb compile failed, default_target_compile: Can't find rustc --color never.
UNTESTED: gdb.rust/watch.exp: failed to prepare
...
Fix this by adding can_compile rust, and using it in allow_rust_tests, such
that we have instead:
...
UNSUPPORTED: gdb.rust/watch.exp: require failed: allow_rust_tests
...
Since the rest of the code in allow_rust_tests is also about availability of
the rust compiler, move it to can_compile.
Tested on x86_64-linux.
Tom de Vries [Tue, 28 Mar 2023 08:22:48 +0000 (10:22 +0200)]
[gdb/testsuite] Unsupport gdb.rust for remote host
With test-case gdb.rust/watch.exp and remote host I run into:
...
Executing on host: gcc watch.rs -g -lm -o watch (timeout = 300)
...
ld:watch.rs: file format not recognized; treating as linker script
ld:watch.rs:1: syntax error
...
UNTESTED: gdb.rust/watch.exp: failed to prepare
...
The problem is that find_rustc returns "" for remote host, so we fall back to gcc, which fails.
Fix this by returning 0 in allow_rust_tests for remote host.
Tested on x86_64-linux.
Alan Modra [Tue, 28 Mar 2023 01:30:02 +0000 (12:00 +1030)]
ubsan: elfnn-aarch64.c:4595:19: runtime error: load of value 190
which is not a valid value for type '_Bool'
* elfnn-aarch64.c (stub_hash_newfunc): Clear all fields past root.
GDB Administrator [Tue, 28 Mar 2023 00:00:13 +0000 (00:00 +0000)]
Automatic date update in version.in
Tom de Vries [Mon, 27 Mar 2023 21:11:41 +0000 (23:11 +0200)]
[gdb/testsuite] Fix gnat_runtime_has_debug_info for remote host
Fix gnat_runtime_has_debug_info for remote host by checking for
allow_ada_tests.
This fixes an error for test-case gdb.testsuite/gdb-caching-proc.exp and
remote host.
Tested on x86_64-linux.
John Baldwin [Mon, 27 Mar 2023 19:51:02 +0000 (12:51 -0700)]
fbsd-nat: Use correct constant for target_waitstatus::sig.
Use GDB_SIGNAL_TRAP instead of SIGTRAP. This is a no-op since the
value of SIGTRAP on FreeBSD matches the value of GDB_SIGNAL_TRAP, but
it is more correct.
Approved-By: Simon Marchi <simon.marchi@efficios.com>
John Baldwin [Mon, 27 Mar 2023 19:51:02 +0000 (12:51 -0700)]
fbsd-nat: Avoid a direct write to target_waitstatus::kind.
This is in #ifdef'd code for a workaround for FreeBSD versions older
than 11.1 which is why it wasn't caught earlier.
Approved-By: Simon Marchi <simon.marchi@efficios.com>
John Baldwin [Mon, 27 Mar 2023 19:51:02 +0000 (12:51 -0700)]
fbsd-nat: Add missing spaces.
No functional change, just style fixes.
Approved-By: Simon Marchi <simon.marchi@efficios.com>
Vladimir Mezentsev [Thu, 23 Mar 2023 18:46:08 +0000 (11:46 -0700)]
gprofng: 30089 [display text] Invalid number of threads
The real problem is that libcollector doesn't interpose thread_create@GLIBC_2.34
We interpose a lot of libC functions (dlopen, fork, pthread_create, etc.).
Some of these functions have versions. For example, dlopen@GLIBC_2.34,
dlopen@GLIBC_2.17, dlopen@GLIBC_2.2.5, etc.
We have to interpose each of the functions because we don't know
which version of libC will be used during profiling.
Historically, we have used three versions of scripts (mapfile.aarch64-Linux,
mapfile.amd64-Linux, mapfile.intel-Linux).
Three are not needed. One is enough
The fixes below include:
- merged all version symbols into one version script.
- added new version symbols which are defined in latest versions of libC.
- removed unused defines and duplicated code.
- added the DCL_FUNC_VER macro to define the version symbols.
Tested on x86_64 and aarch64 (OL8/OL9). No regression.
gprofng/ChangeLog
2023-03-23 Vladimir Mezentsev <vladimir.mezentsev@oracle.com>
PR gprofng/30089
* libcollector/Makefile.am: Use libgprofng.ver instead of mapfile.*
* libcollector/configure.ac: Delete GPROFNG_VARIANT.
* src/collector_module.h: Move the SYMVER_ATTRIBUTE macro to collector.h
* libcollector/collector.h: Add macros (SYMVER_ATTRIBUTE, DCL_FUNC_VER).
Remove unused defines.
* libcollector/dispatcher.c: Interpose functions from libC.
Clean up the old code.
* libcollector/iotrace.c: Likewise.
* libcollector/libcol_util.c: Likewise.
* libcollector/linetrace.c: Likewise.
* libcollector/mmaptrace.c: Likewise.
* libcollector/synctrace.c: Likewise.
* libcollector/libgprofng.ver: New file.
* libcollector/Makefile.in: Rebuild.
* libcollector/configure: Rebuild.
* libcollector/mapfile.aarch64-Linux: Removed.
* libcollector/mapfile.amd64-Linux: Removed.
* libcollector/mapfile.intel-Linux: Removed.
* libcollector/mapfile.sparc-Linux: Removed.
* libcollector/mapfile.sparcv9-Linux: Removed.
Pedro Alves [Fri, 12 Nov 2021 20:50:29 +0000 (20:50 +0000)]
linux-nat: introduce pending_status_str
I noticed that some debug log output printing an lwp's pending status
wasn't considering lp->waitstatus. This fixes it, by introducing a
new pending_status_str function.
Also fix the comment in gdb/linux-nat.h describing
lwp_info::waitstatus and details the description of lwp_info::status
while at it.
Change-Id: I66e5c7a363d30a925b093b195d72925ce5b6b980
Approved-By: Andrew Burgess <aburgess@redhat.com>
Pedro Alves [Tue, 22 Jun 2021 14:42:51 +0000 (15:42 +0100)]
displaced step: pass down target_waitstatus instead of gdb_signal
This commit tweaks displaced_step_finish & friends to pass down a
target_waitstatus instead of a gdb_signal. This is needed because a
patch later in the step-over-{thread-exit,clone] series will want to
make displaced_step_buffers::finish handle
TARGET_WAITKIND_THREAD_EXITED. It also helps with the
TARGET_WAITKIND_THREAD_CLONED patch later in that same series.
It's also a bit more logical this way, as we don't have to pass down
signals when the thread didn't actually stop for a signal. So we can
also think of it as a clean up.
Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=27338
Change-Id: I4c5d338647b028071bc498c4e47063795a2db4c0
Approved-By: Andrew Burgess <aburgess@redhat.com>
Tom de Vries [Mon, 27 Mar 2023 16:06:19 +0000 (18:06 +0200)]
[gdb/testsuite] Fix gdb.stabs/exclfwd.exp for remote host
Fix test-case gdb.stabs/exclfwd.exp for remote host using include_file.
Tested on x86_64-linux.
Tom de Vries [Mon, 27 Mar 2023 16:06:19 +0000 (18:06 +0200)]
[gdb/testsuite] Fix gdb.stabs/weird.exp for remote host
Fix test-case gdb.stabs/weird.exp for remote host by not using an absolute
destfile argument to gdb_remote_download, which doesn't work well with remotedir.
Tom de Vries [Mon, 27 Mar 2023 15:40:06 +0000 (17:40 +0200)]
[gdb/testsuite] Fix gdb.gdb/unittest.exp for remote host
Fix test-case gdb.gdb/unittest.exp for remote host, by:
- disabling the completion tests if readline is not used, and
- not using with_gdb_cwd $dir for remote host (because it does
not support changing to ".").
Tested on x86_64-linux.
Tom de Vries [Mon, 27 Mar 2023 15:40:06 +0000 (17:40 +0200)]
[gdb/testsuite] Skip do_self_tests on remote host
In do_self_tests we try to find out the location of the gdb to debug, which
will then be copied and renamed to xgdb.
In principle, the host board specifies the location of GDB, on host.
With remote host, we could upload that gdb from host to build/target, but we
would miss the data directory (which is listed as the reason to skip
do_self_tests for remote target).
We could fix that by instead taking the gdb from build instead, but that
wouldn't work with installed testing.
It seems easier to just skip this on remote host.
It could be made to work for the "[is_remote host] && [is_remote target]
&& host == target" scenario (see board local-remote-host-native.exp), but
that doesn't seem worth the effort.
Tested on x86_64-linux.
Tom Tromey [Wed, 22 Mar 2023 19:43:03 +0000 (13:43 -0600)]
Change symbol::line to unsigned int
A user here at AdaCore noticed that, when debugging a certain program,
a stack frame reported line 34358, where it should have been line
99894.
After debugging a bit, I discovered:
(top) p (99894 & ~65536)
$60 = 34358
That line, symbol::line is too narrow.
This patch widens the member and changes all the uses that currently
use the narrower type.
Approved-By: Simon Marchi <simon.marchi@efficios.com>
Tom Tromey [Wed, 1 Mar 2023 22:13:39 +0000 (15:13 -0700)]
Fix 128-bit integer bug in Ada
While working on 128-bit integer support, I found one spot in Ada that
needed a fix as well.
Tom Tromey [Wed, 1 Mar 2023 22:13:21 +0000 (15:13 -0700)]
Use gdb_gmp for scalar arithmetic
This changes gdb to use scalar arithmetic for expression evaluation.
I suspect this patch is not truly complete, as there may be code paths
that still don't correctly handle 128-bit integers. However, many
things do work now.
Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=30190
Tom Tromey [Wed, 1 Mar 2023 19:23:43 +0000 (12:23 -0700)]
Use value_true in value_equal and value_less
Both value_equal and value_less use value_as_long to check a
presumably boolean result of calling value_binop. However,
value_binop in this case actually returns an int as wide as its
arguments, and this approach can then fail for integers wider than
LONGEST. Instead, rewrite this in a form that works for any size
integer.
Tom Tromey [Tue, 28 Feb 2023 14:03:55 +0000 (07:03 -0700)]
Simplify binop_promote
binop_promote currently only handles integer sizes up to
builtin_long_long. However, this may not handle 128-bit types.
Simplify this code, unify the C and non-C (but not OpenCL, as I don't
know how to test this) cases, and handle 128-bit integers as well.
This still doesn't exactly follow C or C++ rules. This could be
implemented, but if so, I think it makes more sense as a C-specific
expression node.
Tom Tromey [Mon, 27 Feb 2023 18:57:24 +0000 (11:57 -0700)]
Add value_as_mpz and value_from_mpz
This adds the two new functions, value_as_mpz and value_from_mpz,
useful for manipulation values via gdb_mpz.
Tom Tromey [Wed, 1 Mar 2023 19:58:11 +0000 (12:58 -0700)]
Add truncation mode to gdb_mpz
This renames gdb_mpz::safe_export to export_bits, and adds a new flag
to export a truncated value. This is needed by value arithmetic.
Tom Tromey [Wed, 1 Mar 2023 19:33:27 +0000 (12:33 -0700)]
Avoid a copy in gdb_mpz::safe_export
Currently, gdb_mpz::safe_export will always make a copy of *this.
However, this copy isn't always needed. This patch makes this code
slightly more efficient, by avoiding the copy when possible.
Tom Tromey [Mon, 27 Feb 2023 18:43:09 +0000 (11:43 -0700)]
Add many operators to gdb_mpz
This adds many operator overloads and other useful methods to gdb_mpz.
This is preparation for using this class for scalar arithmetic in gdb
expression evaluation.
Tom Tromey [Fri, 10 Feb 2023 14:18:15 +0000 (07:18 -0700)]
Populate seen_names hash in cooked_index_shard::do_finalize
Hannes pointed out that cooked_index_shard::do_finalize never
populates the seen_names hash table. This patch adds the necessary
store. This reduces memory use a little for "gdb gdb":
(before) Space used:
28909568 (+0 for this command)
(after) Space used:
28884992 (+0 for this command)
What this means, btw, is that in gdb there are not many symbols that
are both mentioned in many CUs and that also require name
canonicalization. It's possible this would differ in other programs.
Tom de Vries [Mon, 27 Mar 2023 12:33:15 +0000 (14:33 +0200)]
[gdb/testsuite] Fix gdb.asm/asm-source.exp for remote host
Fix test-case gdb.asm/asm-source.exp for remote host using
host_standard_output_file and gdb_remote_download.
Tested on x86_64-linux.
Tom de Vries [Mon, 27 Mar 2023 11:58:10 +0000 (13:58 +0200)]
[gdb/testsuite] Fix gdb.dwarf2/imported-unit-bp-c.exp for remote host
Fix test-case gdb.dwarf2/imported-unit-bp-c.exp on remote by removing a
downloaded source file.
Tested on x86_64-linux.
Tom de Vries [Mon, 27 Mar 2023 11:58:10 +0000 (13:58 +0200)]
[gdb/testsuite] Unsupport gdb.dwarf2/gdb-add-index-symlink.exp for remote host
Declare test-case gdb.dwarf2/gdb-add-index-symlink.exp unsupported for remote
host, because the current implementation of gdb_ensure_index doesn't support it.
Tested on x86_64-linux.
Tom de Vries [Mon, 27 Mar 2023 11:58:10 +0000 (13:58 +0200)]
[gdb/testsuite] Fix gdb.dwarf2/gdb-index-cxx.exp for remote host
Fix test-case gdb.dwarf2/gdb-index-cxx.exp for remote host using
host_standard_output_file.
Tested on x86_64-linux.
Tom de Vries [Mon, 27 Mar 2023 11:58:10 +0000 (13:58 +0200)]
[gdb/testsuite] Fix gdb.dwarf2/enqueued-cu-base-addr.exp for remote host
Fix test-case gdb.dwarf2/enqueued-cu-base-addr.exp for remote host by using
$testfile instead $binfile.
Tested on x86_64-linux.
Tom de Vries [Mon, 27 Mar 2023 11:58:10 +0000 (13:58 +0200)]
[gdb/testsuite] Fix gdb.dwarf2/gdb-index.exp on remote host
Fix test-case gdb.dwarf2/gdb-index.exp on remote host using
gdb_remote_download and host_standard_output_file.
Also declare the test-case unsupported with readnow.
Tested on x86_64-linux.
Tom de Vries [Mon, 27 Mar 2023 11:58:10 +0000 (13:58 +0200)]
[gdb/testsuite] Fix gdb.dwarf2/per-bfd-sharing.exp for remote host
Fix test-case gdb.dwarf2/per-bfd-sharing.exp for remote host using
gdb_remote_download.
Likewise in a few other test-cases.
Tested on x86_64-linux.
Tom de Vries [Mon, 27 Mar 2023 11:58:09 +0000 (13:58 +0200)]
[gdb/testsuite] Fix quoting issue in gdb.base/index-cache.exp
For test-case gdb.base/index-cache.exp and remote host, this:
...
lassign [remote_exec host sh "-c \"rm $cache_dir/*.gdb-index\""] ret
...
gives us:
...
Executing on host: sh -c rm /tmp/tmp.m3L7m2AVkL/*.gdb-index (timeout = 300)
builtin_spawn -ignore SIGHUP sh -c rm /tmp/tmp.m3L7m2AVkL/*.gdb-index^M
rm: missing operand^M
Try 'rm --help' for more information.^M
FAIL: gdb.dwarf2/per-bfd-sharing.exp: couldn't remove files in temporary cache dir
...
Fix this using quote_for_host. Likewise in gdb.dwarf2/per-bfd-sharing.exp.
Tested on x86_64-linux.
Tom de Vries [Mon, 27 Mar 2023 11:58:09 +0000 (13:58 +0200)]
[gdb/testsuite] Fix quoting issues in gdb.dwarf2 for remote host
A few test-cases in gdb.dwarf2 use something like:
...
additional_flags=\"-DFOO=BAR + 10\"
...
which doesn't work on remote host.
Fix this by introducing a new proc quote_for_host that also works for remote
host, such that we have:
...
additional_flags=[quote_for_host -DFOO=BAR + 10]
...
Tested on x86_64-linux.
Tom de Vries [Mon, 27 Mar 2023 11:58:09 +0000 (13:58 +0200)]
[gdb/testsuite] Fix have_index for remote host
Proc have_index is mostly used with $binfile, which gives problems
for remote host.
Fix this by using "file tail" on the proc argument.
Tested on x86_64-linux.
Tom de Vries [Mon, 27 Mar 2023 11:58:09 +0000 (13:58 +0200)]
[gdb/testsuite] Add missing include_file in gdb.dwarf/*.exp
Tom de Vries [Mon, 27 Mar 2023 11:47:51 +0000 (13:47 +0200)]
[gdb/testsuite] Add test-case gdb.dlang/dlang-start-2.exp
For test-case gdb.dlang/dlang-start.exp, I run into:
...
UNSUPPORTED: gdb.dlang/dlang-start.exp: require failed: can_compile d
...
My distro has no support for gdc, but I'd like to have the test-case
running and passing, so let's rewrite the test-case using dwarf assembly
and add it alongside (rather than replacing it, because it's good to use
actual compiler output if we have it available).
My distro does have a package providing dmd, so let's mimic that debug info in
the dwarf assembly. This gives us:
...
(gdb) start ^M
Temporary breakpoint 1 at 0x4004ab^M
Starting program: dlang-start-2 ^M
^M
Temporary breakpoint 1, 0x00000000004004ab in _Dmain ()^M
...
The "_Dmain ()" should probably be "D main", I've filed PR30276 about that.
Also add a "show language" to check that we automatically set the language
correctly to D.
Note that the dwarf assembly also describes main, otherwise the test-case
doesn't function as regression test for commit
47fe57c9281 ("Fix "start" for
D, Rust, etc").
Tested on x86_64-linux.
Alan Modra [Mon, 27 Mar 2023 06:32:15 +0000 (17:02 +1030)]
Tidy tc-ppc.c XCOFF auxent access
It's better not to drill down into u.auxent but instead use a pointer
to the combined_entry_type. That way the fix_scnlen field is
available, and no one looking at the codes needs to wonder whether
coffsymbol (symbol_get_bfdsym (sym))->native[i + 1] is the same
auxent.
* config/tc-ppc.c (ppc_frob_symbol): Tidy XCOFF auxent access.
(ppc_adjust_symtab): Likewise.
Alan Modra [Mon, 27 Mar 2023 06:15:57 +0000 (16:45 +1030)]
Remove coff_pointerize_aux table_end param
I'm fairly certain the table_end checks are redundant now. This
patch reverts commit
334d4ced42d3.
* coffgen.c (coff_pointerize_aux): Remove table_end parameter.
(coff_get_normalized_symtab): Adjust to suit.
Alan Modra [Sat, 25 Mar 2023 10:45:46 +0000 (21:15 +1030)]
Use stdint types in coff internal_auxent
long is a poor choice of type to store 32-bit values read from
objects files by H_GET_32. H_GET_32 doesn't sign extend so tests like
that in gdb/coffread.c for "negative" values won't work if long is
larger than 32 bits. If long is 32-bit then code needs to be careful
to not accidentally index negative array elements. (I'd rather see a
segfault on an unmapped 4G array index than silently reading bogus
data.) long is also a poor choice for x_sect.s_scnlen, which might
have 64-bit values. It's better to use unsigned exact width types to
avoid surprises.
I decided to change the field names too, which makes most of this
patch simply renaming. Besides that there are a few places where
casts are no longer needed, and where printf format strings or tests
need adjusting.
include/
* coff/internal.h (union internal_auxent): Use unsigned stdint
types. Rename l fields to u32 and u64 as appropriate.
bfd/
* coff-bfd.c,
* coff-rs6000.c,
* coff64-rs6000.c,
* coffcode.h,
* coffgen.c,
* cofflink.c,
* coffswap.h,
* peXXigen.c,
* xcofflink.c: Adjust to suit internal_auxent changes.
binutils/
* rdcoff.c: Adjust to suit internal_auxent changes.
gas/
* config/obj-coff.h,
* config/tc-ppc.c: Adjust to suit internal_auxent changes.
gdb/
* coffread.c,
* xcoffread.c: Adjust to suit internal_auxent changes.
ld/
* pe-dll.c: Adjust to suit internal_auxent changes.
Alan Modra [Mon, 27 Mar 2023 05:51:05 +0000 (16:21 +1030)]
Set proper union selector tag
* coff-bfd.c (bfd_coff_get_auxent): After converting sym pointer
to an index, reset the union tag.
Alan Modra [Sun, 26 Mar 2023 08:56:46 +0000 (19:26 +1030)]
coffgrok access of u.auxent.x_sym.x_tagndx.p
u.auxent.x_sym.x_tagndx is a union. The p field is only valid when
fix_tag is set. This patch fixes code in coffgrok.c that accessed the
field without first checking fix_tag, and removes a whole lot of code
validating bogus pointers to prevent segfaults (which no longer
happen, I checked the referenced PR 17512 testcases). The patch also
documents this in the fix_tag comment, makes is_sym a bitfield, and
sorts the selecter fields a little.
bfd/
* coffcode.h (combined_entry_type): Make is_sym a bitfield.
Sort and comment on union selectors.
* libcoff.h: Regenerate.
binutils/
* coffgrok.c (do_type): Make aux a combined_entry_type. Test
fix_tag before accessing u.auxent.x_sym.x_tagndx.p. Remove
now unnecessary pointer bounds checking.
Alan Modra [Sun, 26 Mar 2023 08:19:13 +0000 (18:49 +1030)]
Duplicate DW_AT_call_file leak
When given two or more DW_AT_call_file for a given function we
currently leak the concat memory.
* dwarf2.c (scan_unit_for_symbols): Don't leak on duplicate
DW_AT_call_file.
Alan Modra [Fri, 24 Mar 2023 23:13:40 +0000 (09:43 +1030)]
XCOFF sanity check
* coffcode.h (coff_pointerize_aux_hook): Sanity check
x_csect.x_scnlen against raw_syment_count.
Nick Clifton [Mon, 27 Mar 2023 10:10:10 +0000 (11:10 +0100)]
Add an option to the gold linker to put its version string into the .comment section.
PR 30187
* options.h (class General_options): Add enable-linker-version.
* layout.cc (Layout::create_gold_note): If linker-version is enabled put the version string into the .comment section.
Tom de Vries [Mon, 27 Mar 2023 09:35:26 +0000 (11:35 +0200)]
[gdb/testsuite] Handle missing gdc in gdb.dlang/dlang-start.exp
On openSUSE Leap 15.4, I get:
...
Running gdb.dlang/dlang-start.exp ...
gdb compile failed, default_target_compile: Can't find gdc.
UNTESTED: gdb.dlang/dlang-start.exp: failed to prepare
...
Fix this by:
- introducing a new proc can_compile, and
- requiring "can_compile d" in the test-case,
such that I have instead:
...
Running gdb.dlang/dlang-start.exp ...
UNSUPPORTED: gdb.dlang/dlang-start.exp: require failed: can_compile d
...
Tested on x86_64-linux, on openSUSE Leap 15.4 and Fedora 37.
Tom de Vries [Mon, 27 Mar 2023 09:35:26 +0000 (11:35 +0200)]
[gdb/testsuite] Remove superfluous pid in temp files
While trying to use gdb_can_simple_compile with a d program, I ran into:
...
/data/vries/gdb/f37/build/gdb/testsuite/temp/105856/can_compile_d-105856.d: \
error: module 'can_compile_d-105856' has non-identifier characters in \
filename, use module declaration instead
...
The d compiler has a problem with the filename can_compile_d-105856.d, which
contains the pid. The pid is added by gdb_simple_compile:
...
set obj [standard_temp_file $name-[pid].$postfix]
...
but it's unnecessary because standard_temp_file already uses the pid.
Fix this by removing "[pid]" in all calls to standard_temp_file.
Tested on x86_64-linux.
GDB Administrator [Mon, 27 Mar 2023 00:00:07 +0000 (00:00 +0000)]
Automatic date update in version.in
Tom de Vries [Sun, 26 Mar 2023 11:44:58 +0000 (13:44 +0200)]
[gdb/testsuite] Introduce allow_dap_tests
Simon pointed out that with gdb.dap/*.exp and target board
native-gdbserver, we run into problems.
I see for each test-case:
...
+++ run
Traceback (most recent call last):
File "startup.py", line 146, in exec_and_log
output = gdb.execute(cmd, from_tty=True, to_string=True)
gdb.error: Don't know how to run. Try "help target".
...
Likewise with target board native-extended-gdbserver.
Fix this by:
- adding a new proc allow_dap_tests,
- using it in all the gdb.dap tests, and
- bailing out if GDBFLAGS/INTERNAL_GDBFLAGS contains
"set auto-connect-native-target off".
Tested on x86_64-linux.
Reported-By: Simon Marchi <simon.marchi@efficios.com>
Approved-By: Tom Tromey <tom@tromey.com>
GDB Administrator [Sun, 26 Mar 2023 00:00:07 +0000 (00:00 +0000)]
Automatic date update in version.in
GDB Administrator [Sat, 25 Mar 2023 00:00:08 +0000 (00:00 +0000)]
Automatic date update in version.in
Tom Tromey [Mon, 20 Mar 2023 16:14:51 +0000 (10:14 -0600)]
Preserve name of range types
The type-allocation patches introduced a small regression that was
picked up by the AdaCore internal test suite. Previously, the name of
a range type was preserved by resolve_dynamic_range, but now it is
not. This patch changes this code to preserve the name.
Reviewed-By: Bruno Larsen <blarsen@redhat.com>
Tom Tromey [Thu, 16 Feb 2023 14:49:33 +0000 (07:49 -0700)]
Implement repl evaluation for DAP
The evaluate command supports a "context" parameter which tells the
adapter the context in which an evaluation occurs. One of the
supported values is "repl", which we took to mean evaluation of a gdb
command. That is what this patch implements.
Note that some gdb commands probably will not work correctly with the
rest of the protocol. For example if the user types "continue",
confusion may result.
This patch requires the earlier patch to fix up scopes in DAP.
Tom de Vries [Fri, 24 Mar 2023 14:45:56 +0000 (15:45 +0100)]
[gdb/symtab] Fix line number of static const class member
Since commit
6d263fe46e0 ("Avoid bad breakpoints with --gc-sections"), there
was a silent regression on openSUSE Leap 15.4 for test-case
gdb.cp/m-static.exp, from:
...
(gdb) info variable everywhere^M
All variables matching regular expression "everywhere":^M
^M
File /home/vries/tmp.local-remote-host-native/m-static.h:^M
8: const int gnu_obj_4::everywhere;^M
(gdb)
...
to:
...
(gdb) info variable everywhere^M
All variables matching regular expression "everywhere":^M
^M
File /data/vries/gdb/src/gdb/testsuite/gdb.cp/m-static.h:^M
8: const int gnu_obj_4::everywhere;^M
^M
File /data/vries/gdb/src/gdb/testsuite/gdb.cp/m-static1.cc:^M
8: const int gnu_obj_4::everywhere;^M
(gdb)
...
Another regression was found due to that commit, and it was fixed in commit
99d679e7b30 ("[gdb/symtab] Fix "file index out of range" complaint") by
limiting the scope of the fix in the original commit.
Fix this regression by yet further limiting the scope of that fix, making sure
that this bit in dwarf_decode_lines is executed again for m-static1.cc:
...
/* Make sure a symtab is created for every file, even files
which contain only variables (i.e. no code with associated
line numbers). */
...
Tested on x86_64-linux.
PR symtab/30265
Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=30265
Nick Alcock [Thu, 23 Mar 2023 00:15:17 +0000 (00:15 +0000)]
libctf: get the offsets of fields of unnamed structs/unions right
We were failing to add the offsets of the containing struct/union
in this case, leading to all offsets being relative to the unnamed
struct/union itself.
libctf/
PR libctf/30264
* ctf-types.c (ctf_member_info): Add the offset of the unnamed
member of the current struct as necessary.
* testsuite/libctf-lookup/unnamed-field-info*: New test.
Nick Alcock [Mon, 13 Mar 2023 17:36:33 +0000 (17:36 +0000)]
libctf: fix a comment typo
ctf_dedup's intern() function does not return a dynamically allocated
string, so I just spent ten minutes auditing for obvious memory leaks
that couldn't actually happen. Update the comment to note what it
actually returns (a pointer into an atoms table: i.e. possibly not
a new string, and not so easily leakable).
libctf/
* ctf-dedup.c (intern): Update comment.
Nick Alcock [Mon, 13 Mar 2023 17:32:53 +0000 (17:32 +0000)]
libctf: work around an uninitialized variable warning
GCC 11+ complains that sym is uninitialized in ctf_symbol_next. It
isn't, but it's not quite smart enough to figure that out (it requires
domain-specific knowledge of the state of the ctf_next_t iterator
over multiple calls).
libctf/
* ctf-lookup.c (ctf_symbol_next): Initialize sym to a suitable
value for returning if never reset during the function.
Nick Alcock [Mon, 23 Jan 2023 13:11:32 +0000 (13:11 +0000)]
libctf: fix assertion failure with no system qsort_r
If no suitable qsort_r is found in libc, we fall back to an
implementation in ctf-qsort.c. But this implementation routinely calls
the comparison function with two identical arguments. The comparison
function that ensures that the order of output types is stable is not
ready for this, misinterprets it as a type appearing more that once (a
can-never-happen condition) and fails with an assertion failure.
Fixed, audited for further instances of the same failure (none found)
and added a no-qsort test to my regular testsuite run.
libctf/:
PR libctf/30013
* ctf-dedup.c (sort_output_mapping): Inputs are always equal to
themselves.
Tom Tromey [Tue, 21 Mar 2023 14:35:09 +0000 (08:35 -0600)]
Fix race in DAP startup
Internal AdaCore DAP testing on Windows has had occasional failures
that show:
assert threading.current_thread() is _dap_thread
I think this is a race in DAP startup: the _dap_thread global is only
set on return from start_thread, but it seems possible that the thread
itself could already run and encounter a @in_dap_thread decorator.
This patch fixes the problem by setting the global before running any
of the code in the new thread. This also lets us remove a FIXME.
Luis Machado [Tue, 14 Mar 2023 10:30:53 +0000 (10:30 +0000)]
aarch64: Check for valid inferior thread/regcache before reading pauth registers
There were reports of gdb throwing internal errors when calling
inferior_thread ()/get_current_regcache () on a system with
Pointer Authentication enabled.
In such cases, gdb produces the following backtrace:
../../../repos/binutils-gdb/gdb/thread.c:86: internal-error: inferior_thread: Assertion `current_thread_ != nullptr' failed.
A problem internal to GDB has been detected,
further debugging may prove unreliable.
----- Backtrace -----
0xaaaae04a571f gdb_internal_backtrace_1
../../../repos/binutils-gdb/gdb/bt-utils.c:122
0xaaaae04a57f3 _Z22gdb_internal_backtracev
../../../repos/binutils-gdb/gdb/bt-utils.c:168
0xaaaae0b52ccf internal_vproblem
../../../repos/binutils-gdb/gdb/utils.c:401
0xaaaae0b5310b _Z15internal_verrorPKciS0_St9__va_list
../../../repos/binutils-gdb/gdb/utils.c:481
0xaaaae0e24b8f _Z18internal_error_locPKciS0_z
../../../repos/binutils-gdb/gdbsupport/errors.cc:58
0xaaaae0a88983 _Z15inferior_threadv
../../../repos/binutils-gdb/gdb/thread.c:86
0xaaaae0956c87 _Z20get_current_regcachev
../../../repos/binutils-gdb/gdb/regcache.c:428
0xaaaae035223f aarch64_remove_non_address_bits
../../../repos/binutils-gdb/gdb/aarch64-tdep.c:3572
0xaaaae03e8abb _Z31gdbarch_remove_non_address_bitsP7gdbarchm
../../../repos/binutils-gdb/gdb/gdbarch.c:3109
0xaaaae0a692d7 memory_xfer_partial
../../../repos/binutils-gdb/gdb/target.c:1620
0xaaaae0a695e3 _Z19target_xfer_partialP10target_ops13target_objectPKcPhPKhmmPm
../../../repos/binutils-gdb/gdb/target.c:1684
0xaaaae0a69e9f target_read_partial
../../../repos/binutils-gdb/gdb/target.c:1937
0xaaaae0a69fdf _Z11target_readP10target_ops13target_objectPKcPhml
../../../repos/binutils-gdb/gdb/target.c:1977
0xaaaae0a69937 _Z18target_read_memorymPhl
../../../repos/binutils-gdb/gdb/target.c:1773
0xaaaae08be523 ps_xfer_memory
../../../repos/binutils-gdb/gdb/proc-service.c:90
0xaaaae08be6db ps_pdread
../../../repos/binutils-gdb/gdb/proc-service.c:124
0x40001ed7c3b3 _td_fetch_value
/build/glibc-RIFKjK/glibc-2.31/nptl_db/fetch-value.c:115
0x40001ed791ef td_ta_map_lwp2thr
/build/glibc-RIFKjK/glibc-2.31/nptl_db/td_ta_map_lwp2thr.c:194
0xaaaae07f4473 thread_from_lwp
../../../repos/binutils-gdb/gdb/linux-thread-db.c:413
0xaaaae07f6d6f _ZN16thread_db_target4waitE6ptid_tP17target_waitstatus10enum_flagsI16target_wait_flagE
../../../repos/binutils-gdb/gdb/linux-thread-db.c:1420
0xaaaae0a6b33b _Z11target_wait6ptid_tP17target_waitstatus10enum_flagsI16target_wait_flagE
../../../repos/binutils-gdb/gdb/target.c:2586
0xaaaae0789cf7 do_target_wait_1
../../../repos/binutils-gdb/gdb/infrun.c:3825
0xaaaae0789e6f operator()
../../../repos/binutils-gdb/gdb/infrun.c:3884
0xaaaae078a167 do_target_wait
../../../repos/binutils-gdb/gdb/infrun.c:3903
0xaaaae078b0af _Z20fetch_inferior_eventv
../../../repos/binutils-gdb/gdb/infrun.c:4314
0xaaaae076652f _Z22inferior_event_handler19inferior_event_type
../../../repos/binutils-gdb/gdb/inf-loop.c:41
0xaaaae07dc68b handle_target_event
../../../repos/binutils-gdb/gdb/linux-nat.c:4206
0xaaaae0e25fbb handle_file_event
../../../repos/binutils-gdb/gdbsupport/event-loop.cc:573
0xaaaae0e264f3 gdb_wait_for_event
../../../repos/binutils-gdb/gdbsupport/event-loop.cc:694
0xaaaae0e24f9b _Z16gdb_do_one_eventi
../../../repos/binutils-gdb/gdbsupport/event-loop.cc:217
0xaaaae080f033 start_event_loop
../../../repos/binutils-gdb/gdb/main.c:411
0xaaaae080f1b7 captured_command_loop
../../../repos/binutils-gdb/gdb/main.c:475
0xaaaae0810b97 captured_main
../../../repos/binutils-gdb/gdb/main.c:1318
0xaaaae0810c1b _Z8gdb_mainP18captured_main_args
../../../repos/binutils-gdb/gdb/main.c:1337
0xaaaae0338453 main
../../../repos/binutils-gdb/gdb/gdb.c:32
---------------------
../../../repos/binutils-gdb/gdb/thread.c:86: internal-error: inferior_thread: Assertion `current_thread_ != nullptr' failed.
A problem internal to GDB has been detected,
further debugging may prove unreliable.
Quit this debugging session? (y or n)
We also see failures across the testsuite if the tests get executed on a target
that has native support for the pointer authentication feature. But
gdb.base/break.exp and gdb.base/access-mem-running.exp are two examples of
tests that run into errors and internal errors.
This issue started after commit
d88cb738e6a7a7179dfaff8af78d69250c852af1, which
enabled more broad use of pointer authentication masks to remove non-address
bits of pointers, but wasn't immediately detected because systems with native
support for pointer authentication are not that common yet.
The above crash happens because gdb is in the middle of handling an event,
and do_target_wait_1 calls switch_to_inferior_no_thread, nullifying the
current thread. This means a call to inferior_thread () will assert, and
attempting to call get_current_regcache () will also call inferior_thread (),
resulting in an assertion as well.
target_has_registers was one function that seemed useful for detecting these
types of situation where we don't have a register cache. The problem with that
is the inconsistent state of inferior_ptid, which is used by
target_has_registers.
Despite the call to switch_to_no_thread in switch_to_inferior_no_thread from
do_target_wait_1 in the backtrace above clearing inferior_ptid, the call to
ps_xfer_memory sets inferior_ptid momentarily before reading memory:
static ps_err_e
ps_xfer_memory (const struct ps_prochandle *ph, psaddr_t addr,
gdb_byte *buf, size_t len, int write)
{
scoped_restore_current_inferior restore_inferior;
set_current_inferior (ph->thread->inf);
scoped_restore_current_program_space restore_current_progspace;
set_current_program_space (ph->thread->inf->pspace);
scoped_restore save_inferior_ptid = make_scoped_restore (&inferior_ptid);
inferior_ptid = ph->thread->ptid;
CORE_ADDR core_addr = ps_addr_to_core_addr (addr);
int ret;
if (write)
ret = target_write_memory (core_addr, buf, len);
else
ret = target_read_memory (core_addr, buf, len);
return (ret == 0 ? PS_OK : PS_ERR);
}
Maybe this shouldn't happen, or maybe it is just an unfortunate state to be
in. But this prevents the use of target_has_registers to guard against the
lack of registers, since, although current_thread_ is still nullptr,
inferior_ptid is valid and is not null_ptid.
There is another crash scenario after we kill a previously active inferior, in
which case the gdbarch will still say we support pointer authentication but we
will also have no current thread (inferior_thread () will assert etc).
If the target has support for pointer authentication, gdb needs to use
a couple (or 4, for bare-metal) mask registers to mask off some bits of
pointers, and for that it needs to access the registers.
At some points, like the one from the backtrace above, there is no active
thread/current regcache because gdb is in the middle of doing event handling
and switching between threads.
Simon suggested the use of inferior_ptid to fetch the register cache, as
opposed to relying on the current register cache. Though we need to make sure
inferior_ptid is valid (not null_ptid), I think this works nicely.
With inferior_ptid, we can do safety checks along the way, making sure we have
a thread to fetch a register cache from and checking if the thread is actually
stopped or running.
The following patch implements this idea with safety checks to make sure we
don't run into assertions or errors. If any of the checks fail, we fallback to
using a default mask to remove non-address bits of a pointer.
I discussed with Pedro the possibility of caching the mask register values
(which are per-process and can change mid-execution), but there isn't a good
spot to cache those values. Besides, the mask registers can change constantly
for bare-metal debugging when switching between exception levels.
In some cases, it is just not possible to get access to these mask registers,
like the case where threads are running. In those cases, using a default mask
to remove the non-address bits should be enough.
This can happen when we let threads run in the background and then we attempt
to access a memory address (now that gdb is capable of reading memory even
with threads running). Thus gdb will attempt to remove non-address bits
of that memory access, will attempt to access registers, running into errors.
Regression-tested on aarch64-linux Ubuntu 20.04.
Alan Modra [Fri, 24 Mar 2023 06:00:48 +0000 (16:30 +1030)]
Tidy string_ptr increment
* peicode.h (pe_ILF_make_a_symbol): Use sprintf output to
increment string_ptr to end of new string.
Alan Modra [Fri, 24 Mar 2023 05:58:08 +0000 (16:28 +1030)]
Tidy dwarf1 cached section contents
* dwarf1.c (_bfd_dwarf1_cleanup_debug_info): New function.
* libbfd-in.h (_bfd_dwarf1_cleanup_debug_info): Declare.
* elf.c (_bfd_elf_close_and_cleanup): Call it.
* elf-bfd.h (struct elf_obj_tdata): Make dwarf1_find_line_info
a void*.
* libbfd.h: Regenerate.
Tom de Vries [Fri, 24 Mar 2023 09:45:37 +0000 (10:45 +0100)]
[gdb/testsuite] Fix unbalanced quotes in mi_expect_stop argument
In proc mi_expect_stop there's a proc argument reason that's handled like so:
...
set r "reason=\"$reason\","
...
That's fine for say:
...
set reason "foo"
...
for which this evaluates to:
...
set r "reason=\"foo\","
...
But there are more complex uses, for instance:
...
set reason "breakpoint-hit\",disp=\"keep\",bkptno=\"$decimal"
...
which evaluates to:
...
set r "\"breakpoint-hit\",disp=\"keep\",bkptno=\"$decimal\""
...
Note how in this reason argument, the first two '\"' seems to form a pair
surrounding ',disp=', which is not the case, which is confusing.
Fix this by only adding the quotes in mi_expect_stop if the string doesn't
already contain quotes, such that we have the more readable:
...
set reason "\"breakpoint-hit\",disp=\"keep\",bkptno=\"$decimal\""
...
Tested on x86_64-linux.
Tom de Vries [Fri, 24 Mar 2023 08:18:07 +0000 (09:18 +0100)]
[gdb/testsuite] Fix gdb.cp/m-static.exp regression on Ubuntu 20.04
In commit
722c4596034 ("[gdb/testsuite] Fix gdb.cp/*.exp for remote host"), I
needed to change ".*/" into "(.*/)?" in:
...
gdb_test "info variable everywhere" \
"File .*/m-static\[.\]h.*const int gnu_obj_4::everywhere;"
...
However, due to the fact that I got this output:
...
(gdb) info variable everywhere^M
All variables matching regular expression "everywhere":^M
^M
File /data/vries/gdb/src/gdb/testsuite/gdb.cp/m-static.h:^M
8: const int gnu_obj_4::everywhere;^M
^M
File /data/vries/gdb/src/gdb/testsuite/gdb.cp/m-static1.cc:^M
8: const int gnu_obj_4::everywhere;^M
...
I decided to make the matching somewhat stricter, to make sure that the two
matched lines were subsequent.
The commit turned out to be more strict than intended, and caused a regression
on Ubuntu 20.04, where the output was instead:
...
(gdb) info variable everywhere^M
All variables matching regular expression "everywhere":^M
^M
File /data/vries/gdb/src/gdb/testsuite/gdb.cp/m-static.h:^M
8: const int gnu_obj_4::everywhere;^M
...
At that point I realized I'm looking at a bug (filed as PR symtab/30265),
which manifests on openSUSE Leap 15.4 for native and readnow, and on Ubuntu
20.04 for readnow, but not for native.
Before my commit, the test-case passed whether the bug manifested or not.
After my commit, the test-case only passed when the bug manifested.
Fix the test-case regression by reverting to the situation before the commit:
pass whether the bug manifests or not. We could add an xfail for the PR, but
I'm expecting a fix soon, so that doesn't look worth the effort.
Tested on x86_64-linux, both on openSUSE Leap 15.4 and Ubuntu 20.04, both with
native and readnow.
Reported-By: Simon Marchi <simon.marchi@efficios.com>
Tom de Vries [Fri, 24 Mar 2023 08:08:10 +0000 (09:08 +0100)]
[gdb/dap] Add logging of ignored lines
This input sequence is accepted by DAP:
...
{"seq": 4, "type": "request", "command": "configurationDone"}Content-Length: 84
...
This input sequence has the same effect:
...
{"seq": 4, "type": "request", "command": "configurationDone"}ignorethis
Content-Length: 84
...
but the 'ignorethis' part is silently ignored.
Log the ignored bit, such that we have:
...
READ: <<<{"seq": 4, "type": "request", "command": "configurationDone"}>>>
WROTE: <<<{"request_seq": 4, "type": "response", "command": "configurationDone"
, "success": true}>>>
+++ run
IGNORED: <<<b'ignorethis'>>>
...
GDB Administrator [Fri, 24 Mar 2023 00:00:41 +0000 (00:00 +0000)]
Automatic date update in version.in
Tom Tromey [Thu, 23 Mar 2023 16:46:31 +0000 (10:46 -0600)]
Fix minor grammar issue in python.texi
I noticed a minor grammar problem in the 'GDB/MI Commands In Python'
node of the manual. I'm checking in this patch to correct it.
Frederic Cambus [Thu, 23 Mar 2023 15:19:38 +0000 (15:19 +0000)]
Add support to readelf for the PT_OPENBSD_MUTABLE segment type.
binutils * readelf.c (get_segment_type): Handle PT_OPENBSD_MUTABLE segment type.
include * elf/common.h (PT_OPENBSD_MUTABLE): Define.
Tom de Vries [Thu, 23 Mar 2023 13:54:28 +0000 (14:54 +0100)]
[gdb/testsuite] Use gdb_remote_download in allow_opencl_tests
Simon reported that doing:
...
$ while make check-parallel TESTS='gdb.opencl/*.exp' -j 100; do true; done
...
could run into:
...
ERROR: remote_download to target of \
/data/vries/gdb/src/gdb/testsuite/lib/opencl_kernel.cl to opencl_kernel.cl: \
cp: cannot create regular file 'opencl_kernel.cl': File exists
...
Fix this by using gdb_remote_download (instead of plain remote_download) in
allow_opencl_test, which takes care of:
- downloading to a location which is safe for parallel testing, by
using standard_output_file, and
- cleaning up the downloaded file, meaning we can remove the corresponding
"remote_file target delete ${clprogram}" lines in allow_opencl_test.
Tested on x86_64-linux.
Reported-by: Simon Marchi <simon.marchi@efficios.com>
Szabolcs Nagy [Fri, 3 Feb 2023 17:09:58 +0000 (17:09 +0000)]
bfd: aarch64: Optimize BTI stubs PR30076
Don't insert a second stub if the target is already compatible with
an indirect branch.
Szabolcs Nagy [Wed, 18 Jan 2023 12:56:46 +0000 (12:56 +0000)]
bfd: aarch64: Fix stubs that may break BTI PR30076
Insert two stubs in a BTI enabled binary when fixing long calls: The
first is near the call site and uses an indirect jump like before,
but it targets the second stub that is near the call target site and
uses a direct jump.
This is needed when a single stub breaks BTI compatibility.
The stub layout is kept fixed between sizing and building the stubs,
so the location of the second stub is known at build time, this may
introduce padding between stubs when those are relaxed. Stub layout
with BTI disabled is unchanged.