Jens Remus [Fri, 29 Nov 2024 14:37:19 +0000 (15:37 +0100)]
s390: Treat addressing operand sequence as one in disassembler
Reuse logic introduced with the preceding commit in the assembler to
treat addressing operand sequences D(X,B), D(B), and D(L,B) as one
with regards to optional last operands (i.e. optparm and optparm2).
With this "nop" now disassembles into "nop" instead of "nop 0".
opcodes/
* s390-dis.c (operand_count): New helper to count the remaining
operands, treating D(X,B), D(B), and D(L,B) as one.
(skip_optargs_p): New helper to test whether remaining operands
are optional.
(skip_optargs_zero_p): New helper to test whether remaining
operands are optional and their values are zero.
(s390_print_insn_with_opcode): Use skip_optargs_zero_p to skip
optional last operands with a value of zero.
gas/testsuite/
* gas/s390/zarch-optargs.d (nop): Adjust test case accordingly.
Jens Remus [Fri, 29 Nov 2024 14:37:19 +0000 (15:37 +0100)]
s390: Treat addressing operand sequence as one in assembler
The assembler erroneously treated any number of operands as optional,
if the instruction was flagged to have one or two optional operands
(i.e. optparm or optparm2).
Only treat the exact specified number of operands as optional while
treating addressing operand sequences D(X,B), D(B), and D(L,B) as one
operand.
gas/
* config/tc-s390.c (operand_count): New helper to count the
remaining operands, treating D(X,B), D(B), and D(L,B) as one.
(skip_optargs_p): Use new helper operand_count to treat
D(X,B), D(B), and D(L,B) as one operand.
(md_gather_operands): Use skip_optargs_p to skip only the
optional last operands.
Jens Remus [Fri, 29 Nov 2024 14:37:19 +0000 (15:37 +0100)]
s390: Fix disassembly of optional addressing operands
"nop D1(B1)" erroneously disassembled into "nop D1(B1" (missing
closing parenthesis). "nop D1(X1,0)" and "nop D1(X1,)" erroneously
disassembled into "nop D1(X1)" (missing zero base register) instead
of "nop D1(X1,0)".
Do not skip disassembly of optional operands if they are index (X)
or base (B) registers or length (L) in an addressing operand sequence
"D(X,B)", "D(B)", or "D(L,B). Index and base register operand values
of zero are being handled separately, as they may not be omitted
unconditionally. For instance a base register value of zero must be
printed in above mentioned case, to distinguish the index from the
base register. This also ensures proper formatting of addressing
operand sequences.
While at it add further test cases for instructions with optional
operands.
opcodes/
* s390-dis.c (s390_print_insn_with_opcode): Do not
unconditionally skip disassembly of optional operands with a
value of zero, if within an addressing operand sequence.
gas/testsuite/
* gas/s390/zarch-optargs.d: Add further test cases for
instructions with optional operands.
* gas/s390/zarch-optargs.s: Likewise.
Jan Beulich [Fri, 29 Nov 2024 08:37:13 +0000 (09:37 +0100)]
x86/Solaris: support Sun form of CMOVcc
Sun specifies an alternative form for CMOVcc [1], which for some reason
we never cared to support, even if - as per gcc's configure checking for
it - it may have been the only permitted form at some point.
While documentation doesn't indicate FCMOVcc to have similar alternative
forms, gcc assumes so. Hence cover FCMOVcc as well.
Assorted targets do not check, as the ELF targets do, that the program
name in a core file is NUL terminated. Fix some of them. I haven't
attempted to fix all targets because editing host specific code can
easily result in build bugs, which aren't discovered until someone
build binutils for that host. (Of the files edited here, I can't
easily compile hpux-core.c and osf-core.c on a linux system.)
In commit 58776901074 ("[gdb/syscalls] Update to linux v6.11") I updated to
linux v6.11, but a recent submission for loongarch [1] used a current trunk
version, so it makes sense to do this as well elsewhere.
Using linux current trunk with update-linux-from-src.sh gets us 4 more
syscalls:
- setxattrat
- getxattrat
- listxattrat
- removexattrat
Guinevere Larsen [Mon, 25 Nov 2024 18:15:36 +0000 (15:15 -0300)]
gdb/testsuite: skip gdb.threads/omp-par-scope.exp with clang
Since 2020 it has been reported to clang[1] that the debug information
around OpenMP is insufficient. The OpenMP section is not declared
within the correct scope, and instead clang marks as if the section was
a function in the global scope. This causes several failures in the
test gdb.threads/omp-par-scope.exp when using clang to test GDB.
Since this isn't a true failure of GDB, and there is little expectation
that clang will be able to fix this soon, this commit disables the
aforementioned test when clang is being used.
Tom de Vries [Wed, 27 Nov 2024 17:48:43 +0000 (18:48 +0100)]
[gdb/symtab] Fix parent map dump
Before the fix for PR symtab/32225, the parent map dump showed a mapping from
section offsets to cooked index entries:
...
0x0000000000000035 0x3ba9560 (0x34: sp1::A)
...
but now that's no longer the case:
...
0x00000000406f5405 0x410a04d0 (0x34: sp1::A)
...
Fix this by extending the annotation somewhat, such that we get:
...
map start:
0x0000000012c52405 0x135fd550
(section: .debug_info, offset: 0x35) -> (0x34: sp1::A)
...
[gdb/symtab] Fix parent map when handling .debug_info and .debug_types
Consider test-case:
...
$ cat test.c
namespace sp1 {
class A {
int i;
const int f1 = 1;
...
const int f29 = 1;
};
}
sp1::A a;
void _start (void) {}
$ cat test2.c
namespace sp2 {
class B {
float f;
const float f1 = 1;
...
const float f29 = 1;
};
}
sp2::B b;
...
compiled like this:
...
$ g++ test.c -gdwarf-4 -c -g -fdebug-types-section
$ g++ test2.c -gdwarf-5 -c -g -fdebug-types-section
$ g++ -g test.o test2.o -nostdlib
...
Using:
...
$ gdb -q -batch -iex "maint set worker-threads 0" a.out -ex "maint print objfiles"
...
we get a cooked index entry with incorrect parent:
...
[29] ((cooked_index_entry *) 0x3c57d1a0)
name: B
canonical: B
qualified: sp1::A::B
DWARF tag: DW_TAG_class_type
flags: 0x0 []
DIE offset: 0x154
parent: ((cooked_index_entry *) 0x3c57d110) [A]
...
The problem is that the parent map assumes that all offsets are in the same
section.
Fix this by using dwarf2_section_info::buffer-relative addresses instead,
which get us instead:
...
[29] ((cooked_index_entry *) 0x3f0962b0)
name: B
canonical: B
qualified: sp2::B
DWARF tag: DW_TAG_class_type
flags: 0x0 []
DIE offset: 0x154
parent: ((cooked_index_entry *) 0x3f096280) [sp2]
...
Sandra Loosemore [Tue, 26 Nov 2024 19:13:07 +0000 (19:13 +0000)]
nios2: Remove binutils support for Nios II target.
The Nios II architecture has been EOL'ed by the vendor. This patch
removes all binutils, bfd, gas, binutils, and opcodes support for this
target with the exception of the readelf utility. (The ELF EM_*
number remains valid and the relocation definitions from the Nios II
ABI will never change in future, so retaining the readelf support
seems consistent with its purpose as a utility that tries to parse the
headers in any ELF file provided as an argument regardless of target.)
Sandra Loosemore [Tue, 26 Nov 2024 19:13:07 +0000 (19:13 +0000)]
nios2: Remove all GDB support for Nios II targets.
Intel has EOL'ed the Nios II architecture, and it's time to remove support
from all toolchain components before it gets any more bit-rotten from
lack of maintenance or regular testing.
Alan Modra [Mon, 25 Nov 2024 21:54:19 +0000 (08:24 +1030)]
PR32387 ppc64 TLS optimization bug with -fno-plt code
The inline plt code emitted by gcc is incompatible with the
linker/ld.so --tls-get-addr-optimize scheme. This is the runtime
optimisation where the first call to __tls_get_addr results in
__tls_get_addr updating the tls_index pair, then the special linker
stub using that to short-circuit second and subsequent calls for a
given tls symbol. Enabled by default when the linker sees
__tls_get_addr_opt is preseent, and enabled in ld.so when DT_PPC64_OPT
has PPC64_OPT_TLS set. Note that this is distinct from link-time tls
optimisation.
PR 32387
* elf64-ppc.c (ppc64_elf_check_relocs): Disable tls_get_addr_opt
on detecting inline plt calls to __tls_get_addr.
Tom de Vries [Tue, 26 Nov 2024 09:02:37 +0000 (10:02 +0100)]
[gdb/syscalls] Sync with strace v6.12
I ran gdb/syscalls/update-linux-defaults.sh with strace sources v6.12, and got
one difference in gdb/syscalls/linux-defaults.xml.in:
...
+ <syscall name="mseal" groups="memory"/>
...
Rerun make to propagate this change to the xml files.
Tom de Vries [Tue, 26 Nov 2024 08:49:29 +0000 (09:49 +0100)]
[gdb/syscalls] Use update-linux-from-src.sh for arm-linux
I tried to use arm-linux.py to regenerate arm-linux.xml.in, but it didn't work.
Fix this by:
- adding handling of arm-linux.xml.in in update-linux-from-src.sh,
- regenerating arm-linux.xml.in using update-linux-from-src.sh and linux 6.11
sources,
- regenerating arm-linux.xml using make, and
- removing arm-linux.py.
This changes the name "oldolduname" into "olduname".
Tom de Vries [Tue, 26 Nov 2024 08:49:29 +0000 (09:49 +0100)]
[gdb/syscalls] Improve update-linux-from-src.sh
Some improvements in gdb/syscalls/update-linux-from-src.sh:
- use bash instead of sh
- use local to distinguish between local and global vars
(which brings to light that pre uses the global rather than the local
start_date)
- factor out main and parse_args
- factor out regen
- iterate over *.xml.in instead of *.in
Tested on aarch64-linux. Verified with shellcheck.
Simon Marchi [Mon, 4 Nov 2024 18:27:51 +0000 (13:27 -0500)]
Convert all_bfds to new hash table
This converts gdb_bfd.c to use the new hash table for all_bfds.
This patch slightly changes the htab_t pretty-printer test, which was
relying on all_bfds. Note that with the new hash table, gdb-specific
printers aren't needed; the libstdc++ printers suffice -- in fact,
they are better, because the true types of the contents are available.
Change-Id: I48b7bd142085287b34bdef8b6db5587581f94280 Co-Authored-By: Tom Tromey <tom@tromey.com> Approved-By: Tom Tromey <tom@tromey.com>
Simon Marchi [Mon, 4 Nov 2024 18:27:50 +0000 (13:27 -0500)]
Convert typedef hash to new hash table
This converts the typedef hash to use the new hash table.
This patch found a latent bug in the typedef code. Previously, the
hash function looked at the type name, but the hash equality function
used types_equal -- but that strips typedefs, meaning that equality of
types did not imply equality of hashes. This patch fixes the problem
and updates the relevant test.
Change-Id: I0d10236b01e74bac79621244a1c0c56f90d65594 Co-Authored-By: Tom Tromey <tom@tromey.com> Approved-By: Tom Tromey <tom@tromey.com>
Simon Marchi [Mon, 4 Nov 2024 18:27:47 +0000 (13:27 -0500)]
Convert gnu-v3-abi.c to new hash table
This converts gnu-v3-abi.c to use the new hash table.
This change shows how a std::vector can easily be made directly from
the hash table, simplifying the earlier approach of constructing a
vector and a hash table at the same time.
Change-Id: Ia0c387a035a52300db6b6f5a3a2e5c69efa01155 Co-Authored-By: Tom Tromey <tom@tromey.com> Approved-By: Tom Tromey <tom@tromey.com>
Simon Marchi [Mon, 4 Nov 2024 18:27:38 +0000 (13:27 -0500)]
Convert linespec.c to new hash table
This converts linespec.c to use the new hash table.
Note that more simplification could perhaps be done. Currently, the
collectors in this code insert an element into a set and then, if the
element has not been seen before, append it to a vector. If we know
the order does not matter, or if the results can be sorted later, we
could dispense with the vector. This would simplify the code some
more. (This technique is used in the vtable patch, later in this
series.)
Change-Id: Ie6828b1520d918d189ab5140dc8094a609152cf2 Co-Authored-By: Tom Tromey <tom@tromey.com> Approved-By: Tom Tromey <tom@tromey.com>
Simon Marchi [Mon, 4 Nov 2024 18:27:36 +0000 (13:27 -0500)]
Convert compile-c-symbols.c to new hash table
This converts compile-c-symbols.c to use the new hash table.
I made it use a set of string_view instead of a set of `symbol *`, to
avoid calling `symbol::natural_name` over and over. This appears safe
to do, since I don't expect the storage behing the natural names to
change during the lifetime of the map.
Change-Id: Ie9f9334d4f03b9a8ae6886287f82cd435eee217c Co-Authored-By: Tom Tromey <tom@tromey.com> Approved-By: Tom Tromey <tom@tromey.com>
Simon Marchi [Mon, 4 Nov 2024 18:27:35 +0000 (13:27 -0500)]
gdbsupport: add unordered_dense.h 4.4.0
Add a copy of unordered_dense.h from [1]. This file implements an
efficient hash map and hash set with a nice C++ interface (a near
drop-in for std::unordered_map and std::unordered_set). This is
expected to be used to replace uses of `htab_t`, for improved code
readability and type safety. Performance-wise, it is preferred to the
std types (std::unordered_map and std::unordered_set) due to it using
open addressing vs closed addressing for the std types.
I chose this particular implementation because it is simple to use, it's
a standalone header and it typically performs well in benchmarks I have
seen (e.g. [2]). The license being MIT, my understanding is that it's
not a problem to use it and re-distribute it.
Add two additional files, gdbsupport/unordered_map.h and
gdbsupport/unordered_set.h, which make the map and set offered by
gdbsupport/unordered_dense.h available as gdb::unordered_map and
gdb::unordered_set.
Simon Marchi [Mon, 4 Nov 2024 18:27:32 +0000 (13:27 -0500)]
gdb: rename abbrev_cache to abbrev_table_cache
This cache holds `abbrev_table` objects, so I think it's clearer and
more consistent to name it `abbrev_table_cache`. Rename it and
everything that goes along with it.
Change-Id: I43448c0aa538dd2c3ae5efd2f7b3e7b827409d8c Approved-By: Tom Tromey <tom@tromey.com>
Andrew Burgess [Mon, 11 Nov 2024 21:45:17 +0000 (21:45 +0000)]
gdb: do better in breakpoint_free_objfile
The breakpoint_free_objfile function is called from the objfile
destructor, and has the job of removing references to the soon to be
deleted objfile from all breakpoint locations.
The current implementation of breakpoint_free_objfile seems to miss
lots of possible objfile references within bp_location. Currently we
only check if bp_location::symtab is associated with the objfile in
question, but there's bp_location::section and bp_location::probe,
both of which might reference the soon to be deleted objfile.
Additionally bp_location::symbol and bp_location::msymbol if set will
surely be related to the objfile and should also be cleaned up.
I'm not aware that this causes any problems, but it doesn't seem like
a good idea to retain pointers to deleted state, so I propose that we
improve breakpoint_free_objfile to set these pointers back to nullptr.
In the future I plan to investigate the possibility of merging the
functionality of breakpoint_free_objfile into
disable_breakpoints_in_freed_objfile which is called via the
gdb::observers::free_objfile event. However, I already have a patch series
in progress which touches this area of GDB, and I'd like to avoid
conflicting with that earlier series:
Once this patch, and that earlier series have landed then I'll see if
I can merge breakpoint_free_objfile, but I don't think that this needs
to block this patch.
There should be no user visible changes after this commit.
Andrew Burgess [Tue, 5 Nov 2024 13:42:57 +0000 (13:42 +0000)]
gdb: remove an unnecessary scope block in update_breakpoint_locations
In update_breakpoint_locations there's a scope block which I don't
think adds any value. There is one local defined within the scope,
the local is currently an 'int' but should be a 'bool', either way
there's no destructor being triggered when we exit the scope.
This commit changes the local to a 'bool', removes the unnecessary
scope, and re-indents the code.
Within the (now removed) scope was a `for' loop. Inside the loop I
have converted this:
for (....)
{
if (CONDITION)
{
/* Body */
}
}
to this:
for (....)
{
if (!CONDITION)
continue;
/* Body */
}
which means that the body doesn't need to be indented as much, making
things easier to read.
There should be no functional change after this commit.
Reviewed-By: Klaus Gerlicher <klaus.gerlicher@intel.com>
Recalculate the checksum and replace whatever is at the end
of the packet with the newly calculated checksum. Then
replay the packet with the checksum added.
The motivation for this change is that I'd like to add a TCL test
which starts a communication with gdbsever setting the remotelog file.
Then, it modifies the remotelog, injects an error message instead of
the expected replay to some packet in order to test GDB reacts to
the error response properly.
Andrew Burgess [Tue, 5 Nov 2024 17:26:13 +0000 (17:26 +0000)]
gdb/testsuite: force TERM setting for some filename completion tests
Some of the filename completion tests perform mid-line completion.
That is we enter a partial line, then move the cursor back to the
middle of the line and perform completion.
The problem is that, emitting characters into the middle of a terminal
line relies on first emitting some control characters. And which
control characters are emitted will depend on the current TERM
setting.
When I initially added the mid-line completion tests I setup two
regexp that covered two different terminal types, but PR gdb/32338
identifies additional terminal types that emit different sequences of
control characters.
Rather than trying to handle all possible terminal types, lets just
force the TERM variable to something simple (i.e. "dumb") and then
just support that one case. The thing being tested for here was that
GDB would complete a filename in the middle of a line, the specific
terminal type was not really important.
I've simplified the regexp used to match the completion in two places,
and I now force TERM to be "dumb" for the mid-line completion tests.
I've tested this by setting my global environment TERM to 'ansi',
'xterm', 'xterm-mono', and 'dumb', and I see no failures in any mode
now.
Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=32338 Tested-By: Tom de Vries <tdevries@suse.de>
Hui Li [Mon, 25 Nov 2024 07:50:42 +0000 (15:50 +0800)]
gdb: Add LoongArch process record/replay support in NEWS and doc
At present, process record/replay and reverse debugging has been
implemented on LoongArch. Update the NEWS and doc to record this
new change.
Signed-off-by: Hui Li <lihui@loongson.cn> Reviewed-By: Eli Zaretskii <eliz@gnu.org> Approved-By: Tom Tromey <tom@tromey.com> Signed-off-by: Tiezhu Yang <yangtiezhu@loongson.cn>
Hui Li [Mon, 25 Nov 2024 07:46:00 +0000 (15:46 +0800)]
gdb: LoongArch: Add system call support for process record/replay
The process record and replay function also need record Linux
system call instruction. This patch adds LoongArch system call
number definitions in gdb/arch/loongarch-syscall.h, and adds
loongarch_linux_syscall_record() in gdb/loongarch-linux-tdep.c
to record system call execute log. With this patch, the main
functions of process record/replay and reverse debugging are
implemented.
The LoongArch system call numbers definitions are obtained from Linux kernel.
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/include/uapi/asm-generic/unistd.h
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/arch/loongarch/include/asm/unistd.h
Signed-off-by: Hui Li <lihui@loongson.cn> Approved-By: Guinevere Larsen <guinevere@redhat.com> (record-full) Approved-By: Tom Tromey <tom@tromey.com> Signed-off-by: Tiezhu Yang <yangtiezhu@loongson.cn>
Hui Li [Mon, 25 Nov 2024 07:40:03 +0000 (15:40 +0800)]
gdb: LoongArch: Add basic process record/replay support
GDB provides a special process record and replay target that can
record a log of the process execution, and replay it later with
both forward and reverse execution commands. This patch adds the
basic support of process record and replay on LoongArch, it allows
users to debug basic LoongArch instructions and provides reverse
debugging support.
Here is a simple example on LoongArch:
$ cat test.c
int a = 0;
int main()
{
a = 1;
a = 2;
return 0;
}
$ gdb test
...
(gdb) start
...
Temporary breakpoint 1, main () at test.c:4
4 a = 1;
(gdb) record
(gdb) p a
$1 = 0
(gdb) n
5 a = 2;
(gdb) n
6 return 0;
(gdb) p a
$2 = 2
(gdb) rn
5 a = 2;
(gdb) rn
Reached end of recorded history; stopping.
Backward execution from here not possible.
main () at test.c:4
4 a = 1;
(gdb) p a
$3 = 0
(gdb) record stop
Process record is stopped and all execution logs are deleted.
(gdb) c
Continuing.
[Inferior 1 (process 129178) exited normally]
Signed-off-by: Hui Li <lihui@loongson.cn> Approved-By: Guinevere Larsen <guinevere@redhat.com> (record-full) Approved-By: Tom Tromey <tom@tromey.com> Signed-off-by: Tiezhu Yang <yangtiezhu@loongson.cn>
Hui Li [Mon, 25 Nov 2024 07:31:34 +0000 (15:31 +0800)]
gdb: LoongArch: Add instruction definition for process record
GDB provides a special process record function that can record a log
of the process execution. The core of this feature is need to record
the execution of all instructions. This patch adds opcode definitions
and judgments in gdb/arch/loongarch-insn.h. This is preparation for
later patch on LoongArch, there is no effect for the other archs with
this patch.
The LoongArch opcode and mask definitions are obtained from
https://sourceware.org/git/?p=binutils-gdb.git;a=blob;f=opcodes/loongarch-opc.c
LoongArch instruction description refer to the LoongArch Reference Manual:
https://loongson.github.io/LoongArch-Documentation/LoongArch-Vol1-EN.html
Signed-off-by: Hui Li <lihui@loongson.cn> Reviewed-By: Guinevere Larsen <guinevere@redhat.com> Approved-By: Tom Tromey <tom@tromey.com> Signed-off-by: Tiezhu Yang <yangtiezhu@loongson.cn>
Tom de Vries [Sun, 24 Nov 2024 08:21:28 +0000 (09:21 +0100)]
opcodes: fix Werror=format build breaker in opcodes/riscv-dis.c
I build gdb on arm-linux and ran into:
...
CC riscv-dis.lo
opcodes/riscv-dis.c: In function ‘print_insn_args’:
opcodes/riscv-dis.c:743:29: error: format ‘%lu’ expects argument of type \
‘long unsigned int’, but argument 4 has type ‘insn_t’ \
{aka ‘long long unsigned int’} [-Werror=format=]
743 | "%lu", EXTRACT_ZCMT_INDEX (l));
| ~~^
| |
| long unsigned int
| %llu
...
Fix this by printing the insn_t value, which is a uint64_t, using PRIu64.
Tom de Vries [Sat, 23 Nov 2024 12:07:38 +0000 (13:07 +0100)]
[gdb/contrib] Add two words to common-misspellings.txt
While reviewing changes generated by spellcheck.sh for directory sim, I
noticed two more misspellings:
...
arrithemetic->arithmetic
electricaly->electrically
...
Add them to common-misspellings.txt, and fix them in directory sim.
Tom de Vries [Sat, 23 Nov 2024 11:20:34 +0000 (12:20 +0100)]
[gdb/contrib] Add two rules in common-misspellings.txt
Eli mentioned [1] that given that we use US English spelling in our
documentation, we should use "behavior" instead of "behaviour".
In wikipedia-common-misspellings.txt there's a rule:
...
behavour->behavior, behaviour
...
which leaves this as a choice.
Add an overriding rule to hardcode the choice to common-misspellings.txt:
...
behavour->behavior
...
and add a rule to rewrite behaviour into behavior:
...
behaviour->behavior
...
and re-run spellcheck.sh on gdb*.
Guinevere Larsen [Thu, 14 Nov 2024 12:31:00 +0000 (09:31 -0300)]
gdb/record: introduce recoding support for vpor
This commit adds recording support for the AVX instruction vpor, and the
AVX2 extension. Since the encoding of vpor and vpxor are the same, and
their semantics are basically the same, modulo the mathematical
operation, they are handled by the same switch case block.
This also updates the vpxor function, to test vpor and vpxor, and
updates the name to vpor_xor_test to better reflect what it does.
Guinevere Larsen [Tue, 12 Nov 2024 20:45:05 +0000 (17:45 -0300)]
gdb/record: Add support for recording vpmovmskb
This commit adds support for recording the AVX instruction vpmovmskb,
and tests to the relevant file. The test didn't really support checking
general purpose registers, so this commit also adds a proc to
gdb.reverse/i386-avx-reverse.exp, which can be used to test them
Guinevere Larsen [Tue, 12 Nov 2024 19:37:32 +0000 (16:37 -0300)]
gdb/record: Add support for all vpcmpeq instructions
This commit adds support to recording instructions of the form
VPCMPEQ[B|W|D]. They are all encoded in the same way and only
differentiated by the opcode, so they are all processed together. This
commit also updates the test to (quite exhaustively) test the new
instruction.
This commit adds support for recording the instruction vpxor,
introduced in the AVX extension, and extended in AVX2 to use 256 bit
registers. The test gdb.reverse/i386-avx-reverse.exp has been extended
to test this instruction as well.
Guinevere Larsen [Thu, 20 Jun 2024 17:52:35 +0000 (14:52 -0300)]
gdb: Introduce RAII signal handler setter
This patch is motivated by the wait function for the record-full target,
that would install a custom signal handler for SIGINT, but could throw
an exception and never reset the SIGINT handler. This is clearly a bad
idea, so this patch introduces the class scoped_signal_handler in a new
.h file. The file is added to gdbsupport, even though only gdb code is
using it, because it feels like an addition that would be useful for
more than just directly gdb.
The implementation of the RAII class is based on the implementation
on gdb/utils.c. That is, it uses preprocessor ifdefs to probe for
sigaction support, and uses it if possible, defaulting to a raw call to
signal only if sigaction isn't supported. sigaction is preferred based
on the "portability" section of the manual page for the signal function.
There are 3 places where this class can just be dropped in,
gdb/record-full.c, gdb/utils.c and gdb/extension.c. This third place
already had a specialized RAII signal handler setter, but it is
substituted for the new general purpose one.
Hannes Domani [Fri, 22 Nov 2024 19:10:22 +0000 (20:10 +0100)]
Use appropriate context flags for Wow64 processes
When I implemented debugging of Wow64 processes, I missed that there are
extra ContextFlags defines for them.
It's a bit surprising that the wrong ones actually worked, except that
CONTEXT_EXTENDED_REGISTERS is not available for x86_64, and they are
needed for i686, since that's where the xmm registers are stored.
So this replaces the ContextFlags values with their WOW64_* equivalents.
On gdbserver this also duplicates the fallback logic if the
GetThreadContext call failed with CONTEXT_EXTENDED_REGISTERS.
Fixes these fails:
FAIL: gdb.arch/i386-sse.exp: check float contents of %xmm0
FAIL: gdb.arch/i386-sse.exp: check int8 contents of %xmm0
FAIL: gdb.arch/i386-sse.exp: check float contents of %xmm1
FAIL: gdb.arch/i386-sse.exp: check int8 contents of %xmm1
FAIL: gdb.arch/i386-sse.exp: check float contents of %xmm2
FAIL: gdb.arch/i386-sse.exp: check int8 contents of %xmm2
FAIL: gdb.arch/i386-sse.exp: check float contents of %xmm3
FAIL: gdb.arch/i386-sse.exp: check int8 contents of %xmm3
FAIL: gdb.arch/i386-sse.exp: check float contents of %xmm4
FAIL: gdb.arch/i386-sse.exp: check int8 contents of %xmm4
FAIL: gdb.arch/i386-sse.exp: check float contents of %xmm5
FAIL: gdb.arch/i386-sse.exp: check int8 contents of %xmm5
FAIL: gdb.arch/i386-sse.exp: check float contents of %xmm6
FAIL: gdb.arch/i386-sse.exp: check int8 contents of %xmm6
FAIL: gdb.arch/i386-sse.exp: check float contents of %xmm7
FAIL: gdb.arch/i386-sse.exp: check int8 contents of %xmm7
FAIL: gdb.arch/i386-sse.exp: check contents of data[0]
FAIL: gdb.arch/i386-sse.exp: check contents of data[1]
FAIL: gdb.arch/i386-sse.exp: check contents of data[2]
FAIL: gdb.arch/i386-sse.exp: check contents of data[3]
FAIL: gdb.arch/i386-sse.exp: check contents of data[4]
FAIL: gdb.arch/i386-sse.exp: check contents of data[5]
FAIL: gdb.arch/i386-sse.exp: check contents of data[6]
FAIL: gdb.arch/i386-sse.exp: check contents of data[7]
Tom de Vries [Fri, 22 Nov 2024 18:34:24 +0000 (19:34 +0100)]
[gdb/python] Handle failure to initialize without exiting
I tried out making python initialization fail by passing an incorrect
PYTHONHOME, and got:
...
$ PYTHONHOME=foo ./gdb.sh -q
Python path configuration:
PYTHONHOME = 'foo'
...
Python initialization failed: \
failed to get the Python codec of the filesystem encoding
Python not initialized
$
...
The relevant part of the code is:
...
static void
gdbpy_initialize (const struct extension_language_defn *extlang)
{
if (!do_start_initialization () && py_isinitialized && PyErr_Occurred ())
gdbpy_print_stack ();
gdbpy_enter enter_py;
...
What happens is:
- gdbpy_enter::gdbpy_enter () is called, where we run into:
'if (!gdb_python_initialized) error (_("Python not initialized"));'
- the error propagates to gdb's toplevel
- gdb print the error and exits.
It seems unnecesssary that we exit gdb. We could continue the
session without python support.
Fix this by:
- bailing out of gdbpy_initialize if !do_start_initialization
- bailing out of finalize_python if !gdb_python_initialized
This gets us instead:
...
$ PYTHONHOME=foo gdb -q
Python path configuration:
PYTHONHOME = 'foo'
...
Python initialization failed: \
failed to get the Python codec of the filesystem encoding
(gdb) python print (1)
Python not initialized
(gdb)
...
Tom de Vries [Fri, 22 Nov 2024 18:34:24 +0000 (19:34 +0100)]
[gdb/python] Fix abort on Py_Initialize
I tried out making python initialization fail by passing an incorrect
PYTHONHOME with python 3.6, and got:
...
$ PYTHONHOME=foo gdb -q
Fatal Python error: Py_Initialize: Unable to get the locale encoding
ModuleNotFoundError: No module named 'encodings'
Current thread 0x0000ffff89269c80 (most recent call first):
This is as per spec: when Py_Initialize () fails, a fatal error is raised
using Py_FatalError.
This can be worked around using:
...
$ PYTHONHOME=foo gdb -q -eiex "set python ignore-environment on"
(gdb)
...
but it would be better if gdb didn't abort.
I found an article [1] describing two solutions:
- try out Py_Initialize in a separate process, and
- catch the abort using a signal handler.
This patch implements the latter solution.
Obviously we cannot call into python anymore after the abort, so we avoid
calling Py_IsInitialized (), and instead use a new variable py_isinitialized.
This gets us instead:
...
$ PYTHONHOME=foo gdb -q
Fatal Python error: Py_Initialize: Unable to get the locale encoding
ModuleNotFoundError: No module named 'encodings'
Current thread 0x0000fffecfd49c80 (most recent call first):
Python not initialized
$
...
Tested on aarch64-linux.
Approved-By: Tom Tromey <tom@tromey.com>
PR python/32379
Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=32379
Tom de Vries [Fri, 22 Nov 2024 18:34:24 +0000 (19:34 +0100)]
[gdb/python] Handle !Py_IsInitialized () in gdbpy_initialize
I tried out making python initialization fail by passing an incorrect
PYTHONHOME, and got:
...
$ PYTHONHOME=foo gdb -q
Python path configuration:
PYTHONHOME = 'foo'
...
Python Exception <class 'ModuleNotFoundError'>: No module named 'encodings'
Python not initialized
$
...
The relevant part of the code is:
...
static void
gdbpy_initialize (const struct extension_language_defn *extlang)
{
if (!do_start_initialization () && PyErr_Occurred ())
gdbpy_print_stack ();
gdbpy_enter enter_py;
...
What happens is that:
- do_start_initialization returns false because Py_InitializeFromConfig fails,
leaving us in the !Py_IsInitialized () state
- PyErr_Occurred () returns true
- gdbpy_print_stack is called, which prints
"Python Exception <class 'ModuleNotFoundError'>: No module named 'encodings"
The problem is that in the Py_IsInitialized () == false state, very few
functions can be called, and PyErr_Occurred is not one of them [1], and
likewise functions in gdbpy_print_stack.
Fix this by:
- guarding the PyErr_Occurred / gdbpy_print_stack part with Py_IsInitialized ().
- handling the !Py_IsInitialized () case by printing the failure PyStatus
in do_start_initialization
This gets us instead:
...
$ PYTHONHOME=foo ./gdb.sh -q
Python path configuration:
PYTHONHOME = 'foo'
...
Python initialization failed: failed to get the Python codec of the filesystem encoding
Python not initialized
$
...
Tested on aarch64-linux.
Approved-By: Tom Tromey <tom@tromey.com>
[1] https://docs.python.org/3/c-api/init.html#before-python-initialization
Tom de Vries [Fri, 22 Nov 2024 16:44:29 +0000 (17:44 +0100)]
[gdbsupport] Add gdb::{waitpid,read,write,close}
We have gdb::handle_eintr, which allows us to rewrite:
...
ssize_t ret;
do
{
errno = 0;
ret = ::write (pipe[1], "+", 1);
}
while (ret == -1 && errno == EINTR);
...
into:
...
ssize_t ret = gdb::handle_eintr (-1, ::write, pipe[1], "+", 1);
...
However, the call to write got a bit mangled, requiring effort to decode it
back to its original form.
Instead, add a new function gdb::write that allows us to write:
...
ssize_t ret = gdb::write (pipe[1], "+", 1);
...
Andrew Burgess [Mon, 18 Nov 2024 10:58:26 +0000 (10:58 +0000)]
gdb/disasm: fix demangling when disassembling the current function
When disassembling function symbols in C++ code, if GDB is asked to
disassemble a function by name then the function name in the header
line can be demangled by turning on `set print asm-demangle on`, e.g.:
(gdb) disassemble foo_type::some_function
Dump of assembler code for function _ZN8foo_type13some_functionE7my_type:
0x0000000000401142 <+0>: push %rbp
... etc ...
End of assembler dump.
(gdb) set print asm-demangle on
(gdb) disassemble foo_type::some_function
Dump of assembler code for function foo_type::some_function(my_type):
0x0000000000401142 <+0>: push %rbp
... etc ... │
End of assembler dump. │
However, if GDB is disassembling the current function, then this
demangling doesn't work, e.g.:
(gdb) break foo_type::some_function
Breakpoint 1 at 0x401152: file mangle.cc, line 16.
(gdb) run
Starting program: /tmp/mangle
Breakpoint 1, foo_type::some_function (this=0x7fffffffa597, obj=...) at mangle.cc:16
16 obj.update ();
(gdb) disassemble
Dump of assembler code for function _ZN8foo_type13some_functionE7my_type:
0x0000000000401142 <+0>: push %rbp
... etc ...
End of assembler dump.
(gdb) set print asm-demangle on
(gdb) disassemble
Dump of assembler code for function _ZN8foo_type13some_functionE7my_type:
0x0000000000401142 <+0>: push %rbp
... etc ...
End of assembler dump.
This commit fixes this issue, and extends gdb.cp/disasm-func-name.exp,
which was already testing the first case (disassemble by name) to also
cover disassembling the current function.
Tom de Vries [Fri, 22 Nov 2024 16:34:50 +0000 (17:34 +0100)]
[gdb/python] Ensure locale is restored in do_start_initialization
I noticed in do_start_initialization:
...
std::string oldloc = setlocale (LC_ALL, NULL);
setlocale (LC_ALL, "");
...
if (count == (size_t) -1)
{
fprintf (stderr, "Could not convert python path to string\n");
return false;
}
setlocale (LC_ALL, oldloc.c_str ());
...
that the old locale is not restored if the "return false" is triggered.
Sam James [Mon, 18 Nov 2024 05:53:54 +0000 (05:53 +0000)]
libiberty: sync with gcc
This imports the following commits from GCC as of r15-5375-gbeec291225be9b: 94bea5dd6c9a libiberity: ANSIfy test-demangle.c aa84020b2edb libiberty: Fix comment typos c1b2100e736c libiberty: Restore build with CP_DEMANGLE_DEBUG defined bb8dd0980b39 libiberty: Fix up > 64K section handling in simple_object_elf_copy_lto_debug_section [PR116614]
projects / thirdparty / binutils-gdb.git / log
