Jan Beulich [Wed, 27 Sep 2023 12:16:09 +0000 (14:16 +0200)]
x86: fold FMA VEX and EVEX templates
Following the folding of some generic AVX/AVX2 templates with their
AVX512F counterpart ones, do this for FMA ones as well, requiring one
further adjustment to cpu_flags_match().
Jan Beulich [Wed, 27 Sep 2023 12:15:19 +0000 (14:15 +0200)]
x86: fold certain VEX and EVEX templates
In anticipation of APX introduce logic to reduce the number of templates
we have now, allowing to limit some the number of ones we then need to
gain.
The fundamental requirements are that
- attributes be compatible, which specifically means VexW needs to be
the same in the templates (which often isn't the case, for VEX
encodings having far more WIG tha, EVEX ones),
- the EVEX form being AVX512F (with or without AVX512VL), not any of its
extensions (the same will then be required for APX - it'll need to be
APX_F).
Note that in check_register() there's now a redundant zmm check. Since
this logic will need revisiting for APX anyway, I'd like to keep it that
way for now. (Similarly a couple of if()-s which could be folded are
kept separate, to reduce code churn when adding APX support.)
Jan Beulich [Wed, 27 Sep 2023 08:53:38 +0000 (10:53 +0200)]
x86-64: fix suffix-less PUSH of symbol address
PR gas/30856
In 5cc007751cdb ("x86: further adjust extend-to-32bit-address
conditions") I neglected the case of PUSH, which is the only insn
allowing (proper) symbol addresses to be used as immediates (not
displacements, like CALL/JMP) in the absence of any register operands.
Since it defaults to 64-bit operand size, guessing an L suffix is wrong
there.
Simon Marchi [Fri, 8 Sep 2023 01:53:55 +0000 (21:53 -0400)]
gdb/testsuite: add xfail for gdb/29965 in gdb.threads/process-exit-status-is-leader-exit-status.exp
Bug 29965 shows on a Linux kernel >= 6.1, that test fails consistently
with:
FAIL: gdb.threads/process-exit-status-is-leader-exit-status.exp: iteration=0: continue (the program exited)
...
FAIL: gdb.threads/process-exit-status-is-leader-exit-status.exp: iteration=9: continue (the program exited)
This is due to a change in Linux kernel behavior [1] that affects
exactly what this test tests. That is, if multiple threads (including
the leader) call SYS_exit, the exit status of the process should be the
exit status of the leader. After that change in the kernel, it is no
longer the case.
Add an xfail in the test, based on the Linux kernel version. The goal
is that if a regression is introduced in GDB regarding this feature, it
should be caught if running on an older kernel where the behavior was
consistent.
Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=29965
Change-Id: If6ab7171c92bfc1a3b961c7179e26611773969eb Approved-By: Tom de Vries <tdevries@suse.de>
Tom Tromey [Tue, 26 Sep 2023 15:51:56 +0000 (09:51 -0600)]
Remove some unnecessary qualification from printing.py
printing.py references "gdb.printing" in a few spots, but there's no
need for this. I think this is leftover from when this code was
(briefly) in some other module. This patch removes the unnecessary
qualifications. Tested on x86-64 Fedora 36.
Tom Tromey [Fri, 8 Sep 2023 19:18:17 +0000 (13:18 -0600)]
Add two new pretty-printer methods
This adds two new pretty-printer methods, to support random access to
children. The methods are implemented for the no-op array printer,
and DAP is updated to use this.
We've known about the need for printer extensibility for a while, but
have been hampered by backward-compatibilty concerns: gdb never
documented that printers might acquire new methods, and so existing
printers may have attribute name clashes.
To solve this problem, this patch adds a new pretty-printer tag class
that signals to gdb that the printer follows new extensibility rules.
Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=30816 Reviewed-By: Eli Zaretskii <eliz@gnu.org>
Tom de Vries [Tue, 26 Sep 2023 14:28:48 +0000 (16:28 +0200)]
[gdb/testsuite] Fix gdb.server/ext-run.exp in container
When running the gdb testsuite inside a container, I run into:
...
(gdb) gdb_expect_list pattern: /1 +root +[/a-z]*(init|systemd)/
FAIL: gdb.server/ext-run.exp: get process list (pattern 2)
...
because there's no process with pid 1 and cmd init or systemd.
In the host system (where the test passes), I have:
...
$ ps -f 1
UID PID PPID C STIME TTY STAT TIME CMD
root 1 0 0 Sep25 ? Ss 0:03 /usr/lib/systemd/systemd ...
...
but in the container instead:
...
UID PID PPID C STIME TTY STAT TIME CMD
root 1 0 0 11:45 pts/0 Ss 0:00 /bin/bash
...
Following on from the previous patch to make the feature macros take
a word number, this one increases the number of flag words from 1 to 2.
The patch uses some dummy features to push the number of features
over 64. The intention is that these should be reused by real
features rather than kept as-is.
The AArch64 feature-flag code is currently limited to a maximum
of 64 features. This patch reworks it so that the limit can be
increased more easily. The basic idea is:
(1) Turn the ARM_FEATURE_FOO macros into an enum, with the enum
counting bit positions.
(2) Make the feature-list macros take an array index argument
(currently always 0). The macros then return the
aarch64_feature_set contents for that array index.
An N-element array would then be initialised as:
{ MACRO (0), ..., MACRO (N - 1) }
(3) Provide convenience macros for initialising an
aarch64_feature_set for:
- a single feature
- a list of individual features
- an architecture version
- an architecture version + a list of additional features
(2) and (3) use the preprocessor to generate static initialisers.
The main restriction was that uses of the same preprocessor macro
cannot be nested. So if a macro wants to do something for N individual
arguments, it needs to use a chain of N macros to do it. There then
needs to be a way of deriving N, as a preprocessor token suitable for
pasting.
The easiest way of doing that was to precede each list of features
by the number of features in the list. So an aarch64_feature_set
initialiser for three features A, B and C would be written:
AARCH64_FEATURES (3, A, B, C)
This scheme makes it difficult to keep AARCH64_FEATURE_CRYPTO as a
synonym for SHA2+AES, so the patch expands the former to the latter.
Tom de Vries [Tue, 26 Sep 2023 13:51:27 +0000 (15:51 +0200)]
[gdb/dap] Fix dap for python < 3.8
With any gdb.dap test and python 3.6 I run into:
...
Error occurred in Python: 'code' object has no attribute 'co_posonlyargcount'
ERROR: eof reading json header
...
The attribute is not supported before python 3.8, which introduced the
"Positional−only Parameters" concept.
Fix this by using try/except AttributeError.
Tested on x86_64-linux:
- openSUSE Leap 15.4 with python 3.6, and
- openSUSE Tumbleweed with python 3.11.5.
Enze Li [Tue, 26 Sep 2023 12:55:36 +0000 (20:55 +0800)]
fbsd-nat: Fix build failure with GCC 12
A user pointed out that the build failed on FreeBSD/amd64 with my last
commit. The problem is that I'm not using the proper way to tell the
compiler that the variable has been "used". This patch fixes this issue
as suggested by John. Pushed as obvious.
Tested both on FreeBSD/amd64 and FreeBSD/aarch64 by rebuilding.
Fix to step instruction due to P10 prefix instruction.
In AIX, power 10 instructions like paddi occupy 8 bytes, while the other instructions
4 bytes of space. Due to this when we do a stepi on paddi instruction we get a SIGILL interrupt. Hence, we
need to check during stepi if we are able to step 8 bytes during this instruction execution and is the
breakpoint to this instruction set correctly in both 32- and 64-bit mode.
Tom Tromey [Thu, 21 Sep 2023 19:31:14 +0000 (13:31 -0600)]
Use string_file::release in some places
I found a few spots like:
string_file f;
std::string x = f.string ();
However, string_file::string returns a 'const std::string &'... so it
seems to me that this must be copying the string (? I find it hard to
reason about this in C++).
This patch changes these spots to use release() instead, which moves
the string.
Reviewed-by: Keith Seitz <keiths@redhat.com> Reviewed-by: Lancelot Six <lancelot.six@amd.com>
PR 30792
* dwarf.h (struct debug_info): Remove range_versions field.
* dwarf.c (fetch_indexed_offset): New function. (read_and_display_attr_value): Use it for DW_FORM_rnglistx. Remove code to initialise range_versions. (skip_attribute): New function. (read_bases): Read and reccord all range and address bases in a CU. (process_debug_info): Call read_bases. (display_debug_rnglists): Rename to display_debug_rnglists_unit_header and only display the range list header information. (display_debug_ranges): Adjust.
arc: Update bfd arc pattern file to allow enable-targets=all
The ARC backend uses a BFD pattern file to generate three ARC targets:
- an BFD ARC target for ARCv1 and ARCv2 CPU families. It also works
for big-endian variants.
- an BFD ARC64 target for ARCv3 64bit machines. It also allows working
with ARCv3 32bit machines.
- an BFD ARC32 target for ARCv4 32bit machines. It also allows working
with ARCv3 64bit machines.
When configuring with `--enable-targets=all` some patterns are defined
multiple times. Fix this issue.
Andreas Schwab [Thu, 21 Sep 2023 14:49:41 +0000 (16:49 +0200)]
RISC-V: Protect .got with relro
Move .got before .data so that it can be protected with -zrelro. Also
separate .got.plt from .got if -znow is not in effect; the first two words
of .got.plt are placed within the relro region.
arc: Update ARC's Gnu Assembler backend with ARCv3 ISA.
The new Synopsys ARCv3 ISA has a similar instruction format like
the old ARCv1 and ARCv2 ISA. Thus, the ARCv3 addition is using
whatever we have for old ARC processors plus some ARCv3 spcific mods.
To distinguish between various ARC variants, we introduced two new
configure defines named TARGET_ARCv3_32 and TARGET_ARCv3_64 which are
set when we choose either an ARC32 (ARCv3/32) ISA toolchain or an
ARC64 (ARCv3/64) ISA toolchain.
* gas/config/tc-arc.h: Selectively define default target macros.
* gas/configure.ac: Add ARC64 target.
* gas/configure.tgt: Likewise.
* gas/configure: Regenerate
* gas/config.in: Regenerate.
* gas/config/tc-arc.c (DEFAULT_ARCH): New macro.
(default_arch): New variable.
(md_pseudo_table): Add xword.
(md_shortopts): Only a few options are recognized by the new ARC64
assembler.
(md_longopts): Likewise.
(ARC_CPU_TYPE_A64x): New define.
(ARC_CPU_TYPE_A32x): Likewise.
(cpu_type): New arch field.
(selected_cpu): Update fields.
(arc_opcode_hash_entry_iterator_init): Formating.
(arc_opcode_hash_entry_iterator_next): Likewise.
(arc_select_cpu): Likewise.
(arc_option): Likewise.
(check_cpu_feature): Likewise.
(debug_exp): Recognize new expression operands.
(parse_reloc_symbol): Parse new signed/unsigend cases.
(parse_opcode_flags): Update for the case when the flags needs
insert/extract functions.
(find_opcode_match): Match new signed/unsigned 32-bit immediates.
(autodetect_attributes): PLT34 only available for ARC64.
(md_assemble): Extend match characters.
(declare_fp_set): New function.
(init_default_arch): Likewise.
(md_begin): Detect and initialize the correct CPU and coresponding
registers.
(md_pcrel_from_section): Add new relocs.
(arc_target_format): New function.
(md_apply_fix): Add new relocs.
(md_parse_option): Update options.
(arc_show_cpu_list): Update with ARC64 cpus.
(md_show_usage): Update messages.
(may_relax_expr): Add PLT34 case.
(assemble_insn): Update for ARC64.
(arc_make_nops): New function.
(arc_handle_align): Refurbish this function, use arc_make_nops.
(tc_arc_fix_adjustable): Update messages.
arc: Update opcode related include files for ARCv3.
Add new ARCv3 CPUs and required bits to decode/encode ARCv3 ISA
opcodes. Fix 32 bit relocations which were set as signed but should be
bitfield: ARC_32_ME, ARC_GLOB_DAT, ARC_JMP_SLOT, ARC_RELATIVE. Remove
non-ABI relocation ARC_32_ME_S.
include/
xxxx-xx-xx Claudiu Zissulescu <claziss@synopsys.com>
Cupertino Miranda <cupertinomiranda@gmail.com>
Bruno Mauricio <brunoasmauricio@gmail.com>
* include/elf/arc-cpu.def: Add new HS5x and HS6x CPUs.
* include/elf/arc-reloc.def: Add new ARC64 relocations.
* include/elf/arc.h (EF_ARC_CPU_ARC64): New define.
* include/opcode/arc-attrs.h (FEATURE_LIST_NAME): Update predicate.
* include/opcode/arc-func.h: Update formating.
(replace_disp8ls): New function.
(replace_disp9s): Likewise.
(replace_disp6s): Likewise.
(replace_disp7s): Likewise.
(replace_disp12s): Likewise.
* include/opcode/arc.h (ARC_OPCODE_ARC64): New define.
(ARC_OPCODE_ARC32): Likewise.
(ARC_OPERAND_FP): Likewise.
(HARD_FIELDF): Likewise.
(ARC_OPCODE_ARCVx): New macro.
(arc_flag_class): Update structure to hold new extract/insert
functions for flags.
(INSN3OP): Update macro.
(FP_SIZE, TPOF, DPOF, SOPF, COPF, CONVOPS): New enums.
The new Synopsys's ARCv3 ISA is capable to run either 64-bit or
32-bit ISA. The new 32-bit ISA is not compatible with the old
Synopsys ARCv1/ARCv2 ISA, however, it retains a lot of common
concepts. Thus, this patch is reusing the old ARC BFD backend and
adds the necessary bits for the new architecture in a similar way as
it is done for RISCV backend.
bfd/
xxxx-xx-xx Claudiu Zissulescu <claziss@synopsys.com>
Cupertino Miranda <cupertinomiranda@gmail.com>
* bfd/Makefile.am: Add ARC64 files.
* bfd/Makefile.in: Regerate.
* bfd/arc-got.h (TCB_SIZE): Depends on the target architecture.
(GOT_ENTRY_SIZE): New define.
(write_in_got): Likewise.
(read_from_got): Likewise.
(align_power): Likewise.
(arc_got_entry_type_for_reloc): Use RELA_SIZE and GOT_ENTRY_SIZE.
(arc_fill_got_info_for_reloc): Update formating.
(relocate_fix_got_relocs_for_got_info): Likewise.
(arc_static_sym_data): Deleted structure.
(get_static_sym_data): Deleted function.
(relocate_fix_got_relocs_for_got_info): Use symbol static data.
(create_got_dynrelocs_for_single_entry): Update formating.
(create_got_dynrelocs_for_got_info): Likewise.
* bfd/arc-plt.c: New file.
* bfd/arc-plt.def: Add ARC64 PLT entry.
* bfd/arc-plt.h: Clean it up, move functionality to arc-plt.c file.
* bfd/archures.c: Add ARC64 target.
* bfd/config.bfd: Likewise.
* bfd/configure.ac: Likewise.
* bfd/bfd-in2.h: Regenerate.
* bfd/configure: Likewise.
* bfd/libbfd.h: Likewise.
* bfd/cpu-arc.c: Clean it up.
* bfd/cpu-arc64.c: New file.
* bfd/elf32-arc.c: Renamed to elfnn-arc.c.
* bfd/elfnn-arc.c: New file.
* bfd/reloc.c: Add new ARC64 relocs.
* bfd/targets.c: Add ARC64 target.
Enze Li [Mon, 11 Sep 2023 14:40:43 +0000 (22:40 +0800)]
fbsd-nat: Pacify gcc with no functional changes
I see these errors on FreeBSD/aarch64 when using gcc 12 without passing
--disable-werror.
=====================================================================
CXX fbsd-nat.o
fbsd-nat.c: In member function 'void fbsd_nat_target::resume_one_process(ptid_t, int, gdb_signal)':
fbsd-nat.c:1208:11: error: unused variable 'request' [-Werror=unused-variable]
1208 | int request;
| ^~~~~~~
fbsd-nat.c: In member function 'virtual ptid_t fbsd_nat_target::wait(ptid_t, target_waitstatus*, target_wait_flags)':
fbsd-nat.c:1726:22: error: declaration of 'inf' shadows a previous local [-Werror=shadow=compatible-local]
1726 | for (inferior *inf : all_non_exited_inferiors (this))
| ^~~
fbsd-nat.c:1697:17: note: shadowed declaration is here
1697 | inferior *inf = find_inferior_ptid (this, wptid);
| ^~~
fbsd-nat.c: In member function 'virtual void fbsd_nat_target::detach(inferior*, int)':
fbsd-nat.c:2044:18: error: variable 'wptid' set but not used [-Werror=unused-but-set-variable]
2044 | ptid_t wptid = wait_1 (ptid, &ws, 0);
| ^~~~~
cc1plus: all warnings being treated as errors
=====================================================================
This patch includes the following non-functional changes,
1. Remove unused variable "request".
2. Rename inf to inf_p to avoid shadowed declaration warnings.
3. Mark wptid as used when USE_SIGTRAP_SIGINFO is defined.
Tom Tromey [Tue, 19 Sep 2023 17:02:27 +0000 (11:02 -0600)]
Remove keywords from target debug printer names
I recently checked in a patch that removed the use of the "struct"
keyword in some spots. Doing this pointed out that the target
delegate code preserves this keyword -- but, with C++, it does not
really need to. This patch changes make-target-delegates.py to remove
these keywords, and updates target-debug.h to follow. This pointed
out that there was already one redudancy: both
target_debug_print_struct_inferior_p and target_debug_print_inferior_p
existed.
gprofng: 30834 improve disassembly output for call and branch instructions
This patch makes the gprofng disassembler to emit, e.g.
call fprintf@plt [ 0x401060, .-0x49c]
instead of
call 0xfffffffffffffb64
I use bfd_get_synthetic_symtab() to get function names in the .plt section.
I have not yet modified Elf-reader in gprofng to remove parsing of .symtab or
.dynsym sections. But we plan to do it.
gprofng/ChangeLog
2023-09-20 Vladimir Mezentsev <vladimir.mezentsev@oracle.com>
PR gprofng/30834
* src/Disasm.cc: Show the function name in the call instruction
and the relative address in the branch instruction. Remove unused code.
* src/Disasm.h (map_PC_to_func, get_funcname_in_plt): New functions.
* src/Elf.cc: Get function names for the .plt section.
* src/Elf.h (get_funcname_in_plt, get_bfd_symbols): New functions.
* src/Stabs.cc: Add pltSym to SymLst. Remove the conversion to uint32_t.
ld: write resolved path to included file to dependency-file
In ldfile_open_command_file_1() name written to the dependency files is
the name as specified passed to the "INCLUDE" directive.
This is before include-path processing so the tracked dependency
location is most likely wrong.
Instead track the opened file at the point where the resolved path is
actually available, in try_open().
Signed-off-by: Thomas Weißschuh <thomas@t-8ch.de>
Tom Tromey [Wed, 20 Sep 2023 23:08:28 +0000 (17:08 -0600)]
Remove stray trailing "," from DAP breakpoint.py
The buildbot pointed out that the last DAP series I checked in had an
issue. Looking into it, it seems there is a stray trailing "," in
breakpoint.py. This patch removes it.
This seems to point out a test suite deficiency. I will look into
fixing that.
Tom Tromey [Wed, 20 Sep 2023 02:34:23 +0000 (20:34 -0600)]
Remove explanatory comments from includes
I noticed a comment by an include and remembered that I think these
don't really provide much value -- sometimes they are just editorial,
and sometimes they are obsolete. I think it's better to just remove
them. Tested by rebuilding.
Gregory Anders [Fri, 1 Sep 2023 21:02:20 +0000 (16:02 -0500)]
gdb/dap: only include sourceReference if file path does not exist
According to the DAP specification if the "sourceReference" field is
included in a Source object, then the DAP client _must_ make a "source"
request to the debugger to retrieve file contents, even if the Source
object also includes path information.
If the Source's path field is a valid path that the DAP client is able
to read from the filesystem, having to make another request to the
debugger to get the file contents is wasteful and leads to incorrect
results (DAP clients will try to get the contents from the server and
display those contents as a file with the name in "source.path", but
this will conflict with the _acutal_ existing file at "source.path").
Instead, only set "sourceReference" if the source file path does not
exist.
Gregory Anders [Fri, 1 Sep 2023 21:02:19 +0000 (16:02 -0500)]
gdb/dap: use breakpoint fullname to resolve source
If the breakpoint has a fullname, use that as the source path when
resolving the breakpoint source information. This is consistent with
other callers of make_source which also use "fullname" if it exists (see
e.g. DAPFrameDecorator which returns the symtab's fullname).
Gregory Anders [Fri, 1 Sep 2023 21:02:18 +0000 (16:02 -0500)]
gdb/dap: ignore unused keyword args in step_out
Some DAP clients may send additional parameters in the stepOut command
(e.g. "granularity") which are not used by GDB, but should nonetheless
be accepted without error.
Gregory Anders [Fri, 1 Sep 2023 21:02:17 +0000 (16:02 -0500)]
gdb/dap: check for breakpoint source before unpacking
Not all breakpoints have a source location. For example, a breakpoint
set on a raw address will have only the "address" field populated, but
"source" will be None, which leads to a RuntimeError when attempting to
unpack the filename and line number.
Before attempting to unpack the filename and line number from the
breakpoint, ensure that the source information is not None. Also
populate the source and line information separately from the
"instructionReference" field, so that breakpoints that include only an
address are still included.
gdb/tui: add 'set tui mouse-events off' to restore mouse selection
Rationale:
I use the mouse with my terminal to select and copy text. In gdb, I use
the mouse to select a function name to set a breakpoint, or a variable
name to print, for example.
When gdb is compiled with ncurses mouse support, gdb's TUI mode
intercepts mouse events. Left-clicking and dragging, which would
normally select text, seems to do nothing. This means I cannot select
text using my mouse anymore. This makes it harder to set breakpoints,
print variables, etc.
Solution:
I tried to fix this issue by editing the 'mousemask' call to only enable
buttons 4 and 5. However, this still caused my terminal (gnome-terminal)
to not allow text to be selected. The only way I could make it work is
by calling 'mousemask (0, NULL);'. But doing so disables the mouse code
entirely, which other people might want.
I therefore decided to make a setting in gdb called 'tui mouse-events'.
If enabled (the default), the behavior is as it is now: terminal mouse
events are given to gdb, disabling the terminal's default behavior.
If disabled (opt-in), the behavior is as it was before the year 2020:
terminal mouse events are not given to gdb, therefore the mouse can be
used to select and copy text.
Notes:
I am not attached to the setting name or its description. Feel free to
suggest better wording.
Testing:
I tested this change in gnome-terminal by performing the following steps
manually:
1. Run: gdb --args ./myprogram
2. Enable TUI: press ctrl-x ctrl-a
3. Click and drag text with the mouse. Observe no selection.
4. Input: set tui mouse-events off
5. Click and drag text with the mouse. Observe that selection works now.
6. Input: set tui mouse-events on.
7. Click and drag text with the mouse. Observe no selection.
Tom de Vries [Wed, 20 Sep 2023 14:05:55 +0000 (16:05 +0200)]
[gdb/symtab] Error out for .debug_types section in dwz file
There are two methods to factor out type information in a dwarf4 executable:
- use -fdebug-info-types to generate type units in a .debug_types section, and
- use dwz to create partial units.
The dwz method has an extra benefit: it also allows to factor out information
between executables into a newly created .dwz file, pointed to by a
.gnu_debugaltlink section.
There is nothing prohibiting a .gnu_debugaltlink file to contain a
.debug_types section.
It's just not generated by dwz or any other tool atm, and consequently gdb has
no support for it. Enhancement PR symtab/30838 is open about the lack of
support.
Make the current situation explicit by emitting a dwarf error:
...
(gdb) file struct-with-sig-2^M
Reading symbols from struct-with-sig-2...^M
Dwarf Error: .debug_types section not supported in dwz file^M
...
and add an assert in write_gdbindex:
...
+ /* See enhancement PR symtab/30838. */
+ gdb_assert (!(per_cu->is_dwz && per_cu->is_debug_types));
...
to clarify why we can use:
...
data_buf &cu_list = (per_cu->is_debug_types
? types_cu_list
: per_cu->is_dwz ? dwz_cu_list : objfile_cu_list);
...
The test-case is a modified copy from gdb.dwarf2/struct-with-sig.exp, so it
keeps the copyright years range.
Tom Tromey [Tue, 5 Sep 2023 15:13:14 +0000 (09:13 -0600)]
Handle pointers and references correctly in DAP
A user pointed out that the current DAP variable code does not let the
client deference a pointer. Oops!
Fixing this oversight is simple enough -- adding a new no-op
pretty-printer for pointers and references is quite simple.
However, doing this naive caused a regession in scopes.exp, which
expected there to be no children of a 'const char *' variable. This
problem was fixed by the preceding patches in the series, which ensure
that a C type of this kind is recognized as a string.
Tom Tromey [Tue, 5 Sep 2023 18:29:23 +0000 (12:29 -0600)]
Give a language to a type
This changes main_type to hold a language, and updates the debug
readers to set this field. This is done by adding the language to the
type-allocator object.
Note that the non-DWARF readers are changed on a "best effort" basis.
This patch also reimplements type::is_array_like to use the type's
language, and it adds a new type::is_string_like as well. This in
turn lets us change the Python implementation of these methods to
simply defer to the type.
Tom Tromey [Tue, 5 Sep 2023 19:08:29 +0000 (13:08 -0600)]
Add is_array_like and to_array to language_defn
This adds new is_array_like and to_array methods to language_defn.
This will be used in a subsequent patch that generalizes the new
Python array- and string-handling code.
Tom Tromey [Tue, 5 Sep 2023 18:54:40 +0000 (12:54 -0600)]
Regularize some DWARF type initialization
In one spot, it will be convenient for a subsequent patch if the CU is
passed to a type-creation helper function. In another spot, remove
the redundant 'objfile' parameter to another such function.
Tom Tromey [Tue, 5 Sep 2023 18:43:50 +0000 (12:43 -0600)]
Pass a type allocator to init_fixed_point_type
init_fixed_point_type currently takes an objfile and creates its own
type allocator. However, for a later patch it is more convenient if
this function accepts a type allocator. This patch makes this change.
Tom Tromey [Fri, 15 Sep 2023 18:05:57 +0000 (12:05 -0600)]
Use gdb::checked_static_cast for tracepoints
This replaces some casts to 'tracepoint *' with checked_static_cast.
Some functions are changed to accept a 'tracepoint *' now, for better
type safety.
Approved-By: Simon Marchi <simon.marchi@efficios.com>
gdb, breakpoint: add a destructor to the watchpoint struct
Make sure to unlink the related breakpoint when the watchpoint instance
is deleted. This prevents having a wp-related breakpoint that is
linked to a NULL watchpoint (e.g. the watchpoint instance is being
deleted when the 'watch' command fails). With the below scenario,
having such a left out breakpoint will lead to a GDB hang, and this
is due to an infinite loop when deleting all inferior breakpoints.
Scenario:
(gdb) set can-use-hw-watchpoints 0
(gdb) awatch <SCOPE VAR>
Can't set read/access watchpoint when hardware watchpoints are disabled.
(gdb) rwatch <SCOPE VAR>
Can't set read/access watchpoint when hardware watchpoints are disabled.
(gdb) <continue the program until the end>
>> HANG <<
Signed-off-by: Mohamed Bouhaouel <mohamed.bouhaouel@intel.com> Reviewed-by: Bruno Larsen <blarsen@redhat.com>
After the series that added this command was pushed, Pedro mentioned
that the news description could easily be misinterpreted, as well as
some code and test improvements that should be made.
While fixing the test, I realized that code repetition wasn't
happening as it should, so I took care of that too.
Approved-By: Andrew Burgess <aburgess@redhat.com> Reviewed-By: Eli Zaretskii <eliz@gnu.org>
Tom Tromey [Tue, 5 Sep 2023 18:30:08 +0000 (12:30 -0600)]
Move val_prettyformat to valprint.h
I stumbled across an ancient FIXME comment that was easy to fix --
val_prettyformat does not need to be in defs.h, and is easily moved to
valprint.h, where (despite what the comment says) it belongs.
Tom de Vries [Sat, 16 Sep 2023 02:10:53 +0000 (04:10 +0200)]
[gdb/symtab] Fix overly large gdb-index file check for 32-bit
Add a unit test which checks that write_gdb_index_1 will throw
an error when the size of the file would exceed the maximum value
capable of being represented by 'offset_type'.
The unit test fails on 32-bit systems due to wrapping overflow. Fix this by
changing the type of total_len in write_gdbindex_1 from size_t to uint64_t.
Tested on x86_64-linux.
Co-Authored-By: Kevin Buettner <kevinb@redhat.com> Approved-by: Kevin Buettner <kevinb@redhat.com>
Simon Marchi [Wed, 6 Sep 2023 13:41:45 +0000 (09:41 -0400)]
gdb/amdgpu: add precise-memory support
The amd-dbgapi library exposes a setting called "memory precision" for
AMD GPUs [1]. Here's a copy of the description of the setting:
The AMD GPU can overlap the execution of memory instructions with other
instructions. This can result in a wave stopping due to a memory violation
or hardware data watchpoint hit with a program counter beyond the
instruction that caused the wave to stop.
Some architectures allow the hardware to be configured to always wait for
memory operations to complete before continuing. This will result in the
wave stopping at the instruction immediately after the one that caused the
stop event. Enabling this mode can make execution of waves significantly
slower.
Expose this option through a new "amdgpu precise-memory" setting.
The precise memory setting is per inferior. The setting is transferred
from one inferior to another when using the clone-inferior command, or
when a new inferior is created following an exec or a fork.
It can be set before starting the inferior, in which case GDB will
attempt to apply what the user wants when attaching amd-dbgapi. If the
user has requested to enable precise memory, but it can't be enabled
(not all hardware supports it), GDB prints a warning.
If precise memory is disabled, GDB prints a warning when hitting a
memory exception (translated into GDB_SIGNAL_SEGV or GDB_SIGNAL_BUS),
saying that the stop location may not be precise.
Note that the precise memory setting also affects memory watchpoint
reporting, but the watchpoint support for AMD GPUs hasn't been
upstreamed to GDB yet. When we do upstream watchpoint support, GDB will
produce a similar warning message when stopping due to a watchpoint if
precise memory is disabled.
Add a handful of tests. Add a util proc
"hip_devices_support_precise_memory", which indicates if all devices
used for testing support that feature.
Simon Marchi [Thu, 13 Apr 2023 20:18:34 +0000 (16:18 -0400)]
gdb: add inferior_cloned observable
The following patch makes the amdgpu port transfer a property from the
original inferior to the new inferior when using the clone-inferior
command. Add the inferior_cloned observable to help with this.
Change-Id: Id845a799813ec49b1b7b2fcb97b07d0a1e5e2631 Approved-By: Tom Tromey <tom@tromey.com>
Tom Tromey [Fri, 15 Sep 2023 16:13:24 +0000 (10:13 -0600)]
Fix build failure with GCC 4.8
A user pointed out that the build failed with GCC 4.8. The problem
was that the form used by the std::hash specialization of ptid_t was
not accepted. This patch rewrites this code into a form that is
acceptable to the older compiler.
Approved-By: Simon Marchi <simon.marchi@efficios.com>
Simon Marchi [Fri, 15 Sep 2023 15:56:18 +0000 (11:56 -0400)]
gdb: add Lancelot Six as maintainer of the AMD GPU port
Lancelot has accepted to take the role of maintainer for the AMD GPU
port. The AMD GPU port (amdgpu as I've written in the MAINTAINERS file)
is an umbrella term for everything needed to make this work: the amdgcn
arch, the amd-dbgapi target, solib-rocm, etc.
Thanks for accepting the role, and congratulations!
projects / thirdparty / binutils-gdb.git / log
