]> git.ipfire.org Git - thirdparty/valgrind.git/log
thirdparty/valgrind.git
22 hours agoFreeBSD cleanup: mostly using VKI prefix in vki-freebsd.h master
Paul Floyd [Wed, 1 May 2024 14:06:53 +0000 (16:06 +0200)] 
FreeBSD cleanup: mostly using VKI prefix in vki-freebsd.h

23 hours agoFreeBSD ioctl: CAMIOCOMMAND with subcommmand XPT_DEV_MATCH
Paul Floyd [Wed, 1 May 2024 13:17:50 +0000 (15:17 +0200)] 
FreeBSD ioctl: CAMIOCOMMAND with subcommmand XPT_DEV_MATCH

CAM is the FreeBSD disk I/O subsystem (Common Access Method).

I had picked up an implementation of this from bitbucket done
by a FreeBSD dev. It was done a long time ago, around 2013 on
Valgrind 3.8. Mercurial bitbucket is now kickbucket. There
were 2 big problems.

1.  Bit rot. The CAM ioctl is versioned and the version that I
copies was 0x15 whilst the current version is 0x2a, 6 versions
later. That meant that the ioctl wrapper wasn't even getting
called. In addition, as the versioning implies, the data structures
were also out of date.

2. Might also be bit rot but the data structures were copied with
QUEUE_MACRO_DEBUG_TRACE enabled. For that to work the whole
FreeBSD system needs to be built with it turned on. With that
macro turned on in Valgrind the data structures were offset
compared to the kernel, and nothing worked correctly.

There are still a huge number of CAM ioctls unimplemented.
"camcontrol devlist" now works wihtout generating false positives.

27 hours agoFreeBSD ioctl: add BIOCSETF
Paul Floyd [Wed, 1 May 2024 09:08:19 +0000 (11:08 +0200)] 
FreeBSD ioctl: add BIOCSETF

And a bit more header cleanup

2 days agoDRD regtest: try to make drd/tests/pth_mutex_signal less flaky
Paul Floyd [Tue, 30 Apr 2024 05:58:38 +0000 (07:58 +0200)] 
DRD regtest: try to make drd/tests/pth_mutex_signal less flaky

Use a loop of 1ms usleeps rather than a single 1s usleep.
This is usually more releable in ensuring that there's a context switch and that signals
get handled in the order that we want.

2 days agoDarwin regtest: don't visit dSYM directories with tests/vg_regtest
Paul Floyd [Mon, 29 Apr 2024 19:11:53 +0000 (21:11 +0200)] 
Darwin regtest: don't visit dSYM directories with tests/vg_regtest

Due to the extra directory depth of dSYM directories the following line
    my $dashes = "-" x (50 - length $full_dir);
generates a warning on Darwin because the length of $full_dir is more than 50
("Negative repeat count does nothing"). This change makes test_one_dir return
early if the directory contains "dSYM".

3 days agoFreeBSD ioctl: add SIOCGIFCONF
Paul Floyd [Sun, 28 Apr 2024 19:02:11 +0000 (21:02 +0200)] 
FreeBSD ioctl: add SIOCGIFCONF

Also remove a load of commented out Linux stuff from vki-freebsd.h

3 days agoFreeBSD ioctl: enable SIOCGIFMEDIA PCIOCGETCONF CAMIOCOMMAND ioctls and add SIOCGIFSTATUS
Paul Floyd [Sun, 28 Apr 2024 16:22:43 +0000 (18:22 +0200)] 
FreeBSD ioctl: enable SIOCGIFMEDIA PCIOCGETCONF CAMIOCOMMAND ioctls and add SIOCGIFSTATUS

3 days agofd leaks: add an assert for the error extra data
Paul Floyd [Sun, 28 Apr 2024 16:19:55 +0000 (18:19 +0200)] 
fd leaks: add an assert for the error extra data

Coverity claims that it could cause a NULL dereference. I don't
think it's possible. Adding an assert will at least make things
clearer if Coverity ever turns out to be right.

4 days agomc-manual.xml fix new/delete expresions -> expression
Mark Wielaard [Sat, 27 Apr 2024 20:19:54 +0000 (22:19 +0200)] 
mc-manual.xml fix new/delete expresions -> expression

Spotted by Philippe Waroquiers

4 days agoregtest warnings: use %z for printf size_t
Paul Floyd [Sat, 27 Apr 2024 15:41:19 +0000 (17:41 +0200)] 
regtest warnings: use %z for printf size_t

5 days agowarnings: arm64 unused function in test and missing return for FreeBSD getauxval
Paul Floyd [Sat, 27 Apr 2024 11:42:57 +0000 (13:42 +0200)] 
warnings: arm64 unused function in test and missing return for FreeBSD getauxval

5 days agowarnings: remove two unused variables
Paul Floyd [Sat, 27 Apr 2024 11:30:35 +0000 (13:30 +0200)] 
warnings: remove two unused variables

5 days agoREADMEs: add details on dependencies needed to build and run tests.
Paul Floyd [Sat, 27 Apr 2024 11:23:21 +0000 (13:23 +0200)] 
READMEs: add details on dependencies needed to build and run tests.

5 days ago486180 - [Valgrind][MIPS] 'VexGuestArchState' has no member named 'guest_IP_AT_SYSCALL'
Paul Floyd [Sat, 27 Apr 2024 11:07:07 +0000 (13:07 +0200)] 
486180  - [Valgrind][MIPS] 'VexGuestArchState' has no member named 'guest_IP_AT_SYSCALL'

5 days agoUpdate memcheck description of C++ new and delete
Mark Wielaard [Fri, 26 Apr 2024 20:07:46 +0000 (22:07 +0200)] 
Update memcheck description of C++ new and delete

Lifted from valgrind-htdocs commit 73eaff0ad.

Left out the paragraph on C23 free_sized and free_sized_aligned not
being supported (they are now).

5 days agoAdd some website spelling fixes to xml files
Mark Wielaard [Fri, 26 Apr 2024 19:21:04 +0000 (21:21 +0200)] 
Add some website spelling fixes to xml files

Pick up some commits from valgrind-htdocs:

commit d97a1a24e

    Fix description of memory xtree.

    Was referring to function g2 which is not present in the screenshot.

    Suggested by Fabien Launay on valgrind-developers.

commit bc2801d98

Fix a few "the the" usages.

    One reported by Fabien Launay on valgrind-developers. The other
    two found with grep. There's a fourth one, but it's in a patch
    file so I've left it for now.

commit acc56391e

    Consistent spelling of visualize(r)

    Suggested by Fabien Launay on valgrind-developers

    I checked my Shorter Oxford English Dictionary, and there it is
    with a z. Good enough for me.

5 days ago-> 3.23.0 final VALGRIND_3_23_0
Mark Wielaard [Fri, 26 Apr 2024 16:44:47 +0000 (18:44 +0200)] 
-> 3.23.0 final

5 days agoNEWS: Add new instruction support for arm64, amd64, s390x and x86
Mark Wielaard [Fri, 26 Apr 2024 15:03:49 +0000 (17:03 +0200)] 
NEWS: Add new instruction support for arm64, amd64, s390x and x86

5 days agoAdd various none/tests/ppc64 test binaries to .gitignore
Mark Wielaard [Fri, 26 Apr 2024 12:52:46 +0000 (14:52 +0200)] 
Add various none/tests/ppc64 test binaries to .gitignore

5 days agoAdd valgrind-*tar.gz and bz2 to .gitignore
Mark Wielaard [Fri, 26 Apr 2024 12:48:33 +0000 (14:48 +0200)] 
Add valgrind-*tar.gz and bz2 to .gitignore

They are generated by make distcheck.

6 days agoUse 32 bit (with space) in sized_aligned_new_delete_misaligned test supp
Mark Wielaard [Thu, 25 Apr 2024 20:11:29 +0000 (22:11 +0200)] 
Use 32 bit (with space) in sized_aligned_new_delete_misaligned test supp

To be consistent with the space used in the exp-32 test file and all
other supp names.

6 days agos390x: Add support for the `NNPA' instruction
Andreas Arnez [Thu, 25 Apr 2024 12:38:45 +0000 (14:38 +0200)] 
s390x: Add support for the `NNPA' instruction

Finalize the NNPA facility support by handling the NNPA instruction
itself, using the new "extension" approach.  Also reflect NNPA facility
support in HWCAP and in the STFLE bits.

7 days agoSet version to 3.23.0-RC2
Mark Wielaard [Wed, 24 Apr 2024 22:45:49 +0000 (00:45 +0200)] 
Set version to 3.23.0-RC2

7 days agoglibc-2.X-helgrind.supp.in: Also recognize variants on gethostbyname2_r
Mark Wielaard [Wed, 24 Apr 2024 20:11:11 +0000 (22:11 +0200)] 
glibc-2.X-helgrind.supp.in: Also recognize variants on gethostbyname2_r

7 days agoFreeBSD syscall: add wrapper for kcmp
Paul Floyd [Wed, 24 Apr 2024 19:34:03 +0000 (21:34 +0200)] 
FreeBSD syscall: add wrapper for kcmp

For FreeBSD 15.0-CURRENT. Will need to add for 14.1 when it is out.
Mostly the same as the Linux kcmp syscall (small differences in the
supported types).

7 days agoAdd prereqs for tests using python 3.9+
Mark Wielaard [Wed, 24 Apr 2024 18:50:43 +0000 (20:50 +0200)] 
Add prereqs for tests using python 3.9+

Adds a script tests/python_test.sh that checks whether there is a python3
binary and that it supports python version 3.9 or higher. Use this script
in the various cachegrind/tests annotate vgtests as prereq.

7 days agoFreeBSD syswrap: wrong length for __sysctlbyname(name)
Paul Floyd [Wed, 24 Apr 2024 18:14:40 +0000 (20:14 +0200)] 
FreeBSD syswrap: wrong length for __sysctlbyname(name)

Copied and pasted from syscall where name is a pointer to array of ints
so the size has a '*sizeof(int)'. -byname name is a char* so should not
have the *4 factor.

From
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=278566

7 days agoAdd new bar_bad exp files to drd/tests/Makefile.am (EXTRA_DIST).
Mark Wielaard [Wed, 24 Apr 2024 17:24:02 +0000 (19:24 +0200)] 
Add new bar_bad exp files to drd/tests/Makefile.am (EXTRA_DIST).

7 days agoAdd another drd/tests/bar_bad exp variant.
Mark Wielaard [Wed, 24 Apr 2024 16:42:42 +0000 (16:42 +0000)] 
Add another drd/tests/bar_bad exp variant.

An uninitialized barrier might not be recognized as a barrier at all.

7 days agos390x: Improve operand names in trackers for PRNO
Andreas Arnez [Tue, 23 Apr 2024 14:27:11 +0000 (16:27 +0200)] 
s390x: Improve operand names in trackers for PRNO

The operand names in the trackers called in the implementation of the PRNO
extension do not indicate the extension being handled.  Change this by
adding the name PRNO to all these names.

7 days agoAdd dl_new_hash suppression to drd/tests/std_thread2.supp
Mark Wielaard [Wed, 24 Apr 2024 14:00:44 +0000 (16:00 +0200)] 
Add dl_new_hash suppression to drd/tests/std_thread2.supp

_dl_lookup_symbol_x is already suppressed and normally _dl_new_hash
seems inlined, but on some systems it isn't.

8 days agoVG_(show_open_fds) Only emit empty line when not creating xml output
Mark Wielaard [Wed, 24 Apr 2024 11:20:34 +0000 (13:20 +0200)] 
VG_(show_open_fds) Only emit empty line when not creating xml output

The "empty" line containing '== pid == ' would corrupt the xml output
with --track-fds=yes.

9 days agoFilter away "main" differences in filter_fdleak
Mark Wielaard [Tue, 23 Apr 2024 12:14:33 +0000 (14:14 +0200)] 
Filter away "main" differences in filter_fdleak

Stack traces showing where fds were created show some differences in
the "main" function, different line numbers, or (in binary) or (below
main). Since the precise location of the original call in the main
function isn't the goal of these tests just filer them all out and
replace them with a simple "main".

11 days agofilter out in /absolute/path in drd/tests stderr filter
Mark Wielaard [Sat, 20 Apr 2024 20:56:05 +0000 (22:56 +0200)] 
filter out in /absolute/path in drd/tests stderr filter

This filters out a line like:
by 0x........: ??? (in /home/mjw/valgrind/helgrind/tests/tc04_free_lock)
and replaces it with
by 0x........: ???

11 days agoAdd new .exp-32 files to memcheck/tests/Makefile.am EXTRA_DIST
Mark Wielaard [Sat, 20 Apr 2024 19:58:09 +0000 (21:58 +0200)] 
Add new .exp-32 files to memcheck/tests/Makefile.am EXTRA_DIST

11 days ago32 bit new_delete_mismatch_size and sized_aligned_new_delete_misaligned .exp
Mark Wielaard [Sat, 20 Apr 2024 19:30:59 +0000 (21:30 +0200)] 
32 bit new_delete_mismatch_size and sized_aligned_new_delete_misaligned .exp

Add 32 bit variants of the memcheck/tests/new_delete_mismatch_size_supp,
memcheck/tests/sized_aligned_new_delete_misaligned{2,3}_supp .exp files.

11 days agoFreeBSD suppression: puts reachable
Paul Floyd [Sat, 20 Apr 2024 16:42:41 +0000 (18:42 +0200)] 
FreeBSD suppression: puts reachable

12 days agoFreeBSD suppressions: yet more reachables
Paul Floyd [Sat, 20 Apr 2024 06:29:51 +0000 (08:29 +0200)] 
FreeBSD suppressions: yet more reachables

12 days agoFreeBSD suppression: reachable for setlocale
Paul Floyd [Sat, 20 Apr 2024 05:54:05 +0000 (07:54 +0200)] 
FreeBSD suppression: reachable for setlocale

Normally I have LANG=en_GB.UTF-8 and LC_ALL=C but if I switch to
C.UTF-8 for everything (which is the default for a fresh install)
then setlocale comes into the picture with a reachable allocation
on exit.

12 days agoFreeBSD suppression: add a suppression for __swbuf seen on arm64
Paul Floyd [Sat, 20 Apr 2024 05:50:31 +0000 (07:50 +0200)] 
FreeBSD suppression: add a suppression for __swbuf seen on arm64

Still no libc freeres.

12 days agoSet version to 3.23.0-RC1
Mark Wielaard [Sat, 20 Apr 2024 01:40:29 +0000 (03:40 +0200)] 
Set version to 3.23.0-RC1

12 days agoUpdate emitted Copyright years to include 2024
Mark Wielaard [Sat, 20 Apr 2024 01:36:01 +0000 (03:36 +0200)] 
Update emitted Copyright years to include 2024

Also adjust test exp and filters.

12 days agocore errors suppression support
Mark Wielaard [Fri, 19 Apr 2024 20:46:11 +0000 (22:46 +0200)] 
core errors suppression support

Add two new functions core_get_extra_suppression_info and
core_get_error_name as alternatives for tool suppression callbacks.
These functions are used in gen_suppression.

Instead of a tool name, a core error component name is "CoreError".

Two new suppression kinds FdBadCloseSupp and FdNotClosedSupp
were added. Corresponding to the FdBadClose and FdNotClosed
error kinds.

core_error_matches_suppression matches these suppression kinds
with error kinds.

core_get_extra_suppression_info and core_print_extra_suppression_use
are noops for core errors.

is_suppressible_error, supp_matches_error, load_one_suppressions_file
and show_used_suppressions have been adjusted to work with core
error kinds.

A new function VG_(found_or_suppressed_errs) helps to not output
an empty error summary if only core errors are requested, but no
errors were detected.

VG_(clo_track_fds) has been moved from pub_core_options.h to
pub_tool_options.h. And VG_(needs_core_errors) now takes a Bool
that can be set to false in the tool post_clo_init handler. This
is used in the none tool to request core errors, but disable all
reporting if no option has been given that enables such errors.

This make sure only the none/tests/fdleak_ipv4.stderr.exp needs
adjustment. For all other tests the output is exactly as before.

https://bugs.kde.org/show_bug.cgi?id=485778

12 days agoFix const signedness issue in VG_(show_open_fds)
Mark Wielaard [Fri, 19 Apr 2024 14:59:25 +0000 (16:59 +0200)] 
Fix const signedness issue in VG_(show_open_fds)

The colon variable should be a const HChar.

13 days agoImprove file descriptor xml output, add fd and path elements
Alexandra Hájková [Thu, 18 Apr 2024 09:21:49 +0000 (05:21 -0400)] 
Improve file descriptor xml output, add fd and path elements

Add needs_xml_output to none tool so it can also output xml when
--track-fds enabled.

Use xml protocolversion 5 when clo_track_fds is enabled

Split OpenFD pathname and description. Add description when a file
descriptor is closed so it can be used in a future error. On error
print <fd> element and (if known) a <path> element.

Add docs/internals/xml-output-protocol5.txt.

https://bugs.kde.org/show_bug.cgi?id=328563

Co-Authored-By: Mark Wielaard <mark@klomp.org>
13 days agoAdd core errors and use them to implement file descriptor tracker
Mark Wielaard [Wed, 20 Mar 2024 05:44:23 +0000 (01:44 -0400)] 
Add core errors and use them to implement file descriptor tracker

All the tool error callbacks now have a core error equivalent.
core errors are negative (while tool errors are positive).
There are two new ones for tracking issues with file descriptors.
FdBadClose (-2) and FdNotClosed (-3).

Add following core error functions with delegates to file descriptor
specific functions (implemented in syswrap-generic):

- core_eq_Error (fd_eq_Error)
  Compares core errors to detect duplicates
- core_before_pp_Error (fd_before_pp_Error)
  Currently prints nothing for known core errors and
  exists with FATAL for unknown core errors
- core_pp_Error (fd_pp_Error)
  For FdBadClose prints the backtraces for the file descriptor was
  opened, where it was originally closed and where it was closed again.
  For FdNotClosed prints where the file descriptor was opened.
- core_update_extra (fd_update_extra)
  Returns the size of the BadCloseExtra or FdNotClosedExtra struct
  which data needs to be saved (the fd number, pathname/description
  and previous backtraces).

We now accept the error (ExeContext) where to be NULL.
This is necessary for reporting not closed file descriptors when
the descriptor is inherited from the parent (so wasn't actually
created and doesn't have a 'where' in the current process code).

All the testcases still pass since the (stderr) output is the
same. But now they count as "real" errors. And so --error-exitcode
does now also work for file descriptor errors or leaks.

https://bugs.kde.org/show_bug.cgi?id=362680

Co-authored-by: Alexandra Hájková <ahajkova@redhat.com>
13 days agoAdd FreeBSD arm64 to README list of platforms
Paul Floyd [Thu, 18 Apr 2024 18:53:58 +0000 (20:53 +0200)] 
Add FreeBSD arm64 to README list of platforms

13 days agoDon't include pub_tool_tooliface.h in priv_types_n_macros.h
Mark Wielaard [Thu, 18 Apr 2024 15:45:16 +0000 (17:45 +0200)] 
Don't include pub_tool_tooliface.h in priv_types_n_macros.h

post-regtest-checks complains:
*** File coregrind/m_extension/priv_types_n_macros.h must not include pub_tool_tooliface.h

Which isn't really necessary anyway since priv_types_n_macros.h
already includes pub_core_tooliface.h.

13 days agoAdd pub_core_extension.h to coregrind/Makefile.am noinst_HEADERS
Mark Wielaard [Thu, 18 Apr 2024 13:58:37 +0000 (15:58 +0200)] 
Add pub_core_extension.h to coregrind/Makefile.am noinst_HEADERS

2 weeks agoAdd extension concept and exploit it for s390x `PRNO'
Andreas Arnez [Thu, 18 Apr 2024 11:49:52 +0000 (13:49 +0200)] 
Add extension concept and exploit it for s390x `PRNO'

z/Architecture specifies various "complex" machine instructions whose
register and memory effects cannot be expressed with Valgrind's current
mechanisms.

One example is the PPNO instruction -- "perform pseudorandom number
operation", whose main purpose is to generate a chunk of pseudorandom
data.  When doing so, it reads and writes a so-called "parameter block"
while also writing the variable-sized pseudorandom data itself.  These
memory effects cannot be adequately represented by the IRDirty concept.

Instead of using dirty helpers, a different approach is to treat such
instructions similar to syscalls, adding a new IRJumpKind for this
purpose.

This patch provides an implementation of this approach and an illustration
of its use at the example of PPNO (or its newer instantiation "PRNO").

2 weeks agos390x: Re-implement `vpdi' without intermediate updates
Andreas Arnez [Tue, 16 Apr 2024 15:24:47 +0000 (17:24 +0200)] 
s390x: Re-implement `vpdi' without intermediate updates

The vpdi instruction is implemented incorrectly for the case where the
result vector coincides with one of the input operands, because it
performs an intermediate update.  An example where this produces wrong
results is for

vpdi    %v3,%v3,%v3,4

where the instruction is used to swap the two halves of %v3.

Fix this by avoiding the intermediate update and using Iop_64HLtoV128
instead.

2 weeks agoBug 481874 - Add arm64 support for FreeBSD
Paul Floyd [Tue, 16 Apr 2024 05:39:33 +0000 (07:39 +0200)] 
Bug 481874 - Add arm64 support for FreeBSD

2 weeks agonon-linux regtest: update cmdline expecteds
Paul Floyd [Mon, 15 Apr 2024 05:18:42 +0000 (07:18 +0200)] 
non-linux regtest: update cmdline expecteds

2 weeks agoAdd Bug 481676 to NEWS
Paul Floyd [Mon, 15 Apr 2024 05:03:23 +0000 (07:03 +0200)] 
Add Bug 481676 to NEWS

Forgot I'd opened a bugzi for the problem.

2 weeks agoAdd @configure_input@ to all filer files generated by configure
Mark Wielaard [Sun, 14 Apr 2024 21:56:31 +0000 (23:56 +0200)] 
Add @configure_input@ to all filer files generated by configure

This will add a comment at the top of each filter file saying:

 filter_file.  Generated from filter_file.in by configure.

2 weeks agoMissing edit of filter_gdb.in
Philippe Waroquiers [Sun, 14 Apr 2024 20:54:49 +0000 (22:54 +0200)] 
Missing edit of filter_gdb.in

.exp files were updated to correspond to GDB 15,
filter_gdb being responsible to transform old GDB output to newly
expected output.

Sadly, I edited filter_gdb instead of filter_gdb.in

Tests should succeed again with older GDB.

2 weeks agoadd_hardwired_spec for ld-linux-x86-64.so.2 strcmp
Florian Weimer [Sun, 14 Apr 2024 19:56:03 +0000 (21:56 +0200)] 
add_hardwired_spec for ld-linux-x86-64.so.2 strcmp

glibc built with -march=x86-64-v3 does not work due to ld.so strcmp

https://bugs.kde.org/show_bug.cgi?id=485487

2 weeks agonightly/bin/nightly: Use grep -E instead of egrep
Mark Wielaard [Sun, 14 Apr 2024 19:42:12 +0000 (21:42 +0200)] 
nightly/bin/nightly: Use grep -E instead of egrep

2 weeks agoEnsure gdbserver tests work with a recent git version of GDB
Philippe Waroquiers [Sun, 14 Apr 2024 19:58:11 +0000 (21:58 +0200)] 
Ensure gdbserver tests work with a recent git version of GDB

Tested with GDB 12, 13, 14 and 15.0.50.20240414

2 weeks agoensure error output of vgdb relay mode is shown to the GDB user
Philippe Waroquiers [Sun, 14 Apr 2024 19:15:24 +0000 (21:15 +0200)] 
ensure error output of vgdb relay mode is shown to the GDB user

With GDB14.1, when there is more than one valgrind process, 'target remote | vgdb' shows:
  (gdb) tar rem | vgdb
  Remote debugging using | vgdb
  no --pid= arg given and multiple valgrind pids found:
  use --pid=913621 for ./Inst/bin/valgrind --vgdb-stop-at=startup ./gdbserver_tests/sleepers
  use --pid=913622 for ./Inst/bin/valgrind --vgdb-stop-at=startup ./gdbserver_tests/sleepers
  Remote communication error.  Target disconnected: Connection reset by peer.
  (gdb)

With GDB 15.0.50.20240414-git, we obtain:
  (gdb) tar rem | vgdb
  Remote debugging using | vgdb
  Remote communication error.  Target disconnected: error while reading: Connection reset by peer.
  (gdb)

This looks like a race condition:

When vgdb exits due to several pid or due to any other error (e.g. an argument
error), GDB gets a SIGPIPE and closes the pipe to/from vgdb.  To avoid losing
the error messages In such cases, have vgdb wait for the first packet from GDB
before exiting.

With this change, the early errors of vgdb are shown to the user.

Tested on debian, with GDB 12, 13, 15 and 15.0.50.20240414-git.

2 weeks agoBug 485148 - vfmadd213ss instruction is instrumented incorrectly (the remaining part...
Paul Floyd [Sun, 14 Apr 2024 15:59:30 +0000 (17:59 +0200)] 
Bug 485148 - vfmadd213ss instruction is instrumented incorrectly (the remaining part of the register is cleared instead of kept unmodified)

Initial version contributed by Bruno Lathuilière <bruno.lathuiliere@edf.fr>
Initial test contributed by Petr <kobalicek.petr@gmail.com>

2 weeks agoFreeBSD syscall: aio_error fails with EINVAL if the aiocb pointer is not accessible.
Paul Floyd [Sun, 14 Apr 2024 14:54:43 +0000 (16:54 +0200)] 
FreeBSD syscall: aio_error fails with EINVAL if the aiocb pointer is not accessible.

2 weeks agoAvoid format '%u' expect unsigned int warning
Philippe Waroquiers [Sun, 14 Apr 2024 14:28:37 +0000 (16:28 +0200)] 
Avoid format '%u' expect unsigned int warning

2 weeks agoFreeBSD regtest: fix for scalar sigaltstack
Paul Floyd [Sun, 14 Apr 2024 13:32:41 +0000 (15:32 +0200)] 
FreeBSD regtest: fix for scalar sigaltstack

I was lazy and using pointers to the same struct for the new and old
data (which isn't allowed, the new is 'restrict'). The current gets
copied to the old first so under Valgrind it worked. In the kernel
there are separate structs (which get copied in and copied out) and
it doesn't work.

Maybe we should consider changing VG_(do_sys_sigaltstack) to have at
least one local copy in order to behave like the kernel.

2 weeks agoAllow to see more details about suppressed errors.
Philippe Waroquiers [Sun, 14 Apr 2024 13:35:16 +0000 (15:35 +0200)] 
Allow to see more details about suppressed errors.

Extend --show-error-list=no|yes with all to allow to look also the
list of suppressed errors and see which supp entry is suppressing it.
Add an option also_suppressed in the monitor command 'c.vinfo all_errors'
to similarly be able to show the suppressed errors.

Small fix in check_makefile_consistency: avoid it considers emacs ~ files.

2 weeks agoFreeBSD regtest: setcontext only reads the context
Paul Floyd [Sun, 14 Apr 2024 13:31:22 +0000 (15:31 +0200)] 
FreeBSD regtest: setcontext only reads the context

2 weeks agoIllumos: also has wcpncpy
Paul Floyd [Sat, 13 Apr 2024 19:35:49 +0000 (21:35 +0200)] 
Illumos: also has wcpncpy

2 weeks agoregtest arm64: add a test for frinta and frintn
Paul Floyd [Sat, 13 Apr 2024 16:19:25 +0000 (18:19 +0200)] 
regtest arm64: add a test for frinta and frintn

Round ties above 0 (frinta) and to even (frintn). Both only scalar.
Vector not yet implemented.

2 weeks agoUpdate NEWS for bug476025
Paul Floyd [Sat, 13 Apr 2024 14:39:24 +0000 (16:39 +0200)] 
Update NEWS for bug476025

2 weeks agoamd64: Implement VFMADD213 for Iop_MAddF32 and Iop_MAddF64
Mark Wielaard [Sat, 13 Apr 2024 12:33:19 +0000 (14:33 +0200)] 
amd64: Implement VFMADD213 for Iop_MAddF32 and Iop_MAddF64

Speed up F32 and F64 FMA on amd64. Add priv/host_amd64_maddf.c
implementing h_amd64_calc_MAddF32_fma4 and h_amd64_calc_MAddF64_fma4
to be used instead of the generic variants h_generic_calc_MAddF32
and h_generic_calc_MAddF64 when host has VEX_HWCAPS_AMD64_FMA4.
Add fma3 and fma4 detection m_machine.c (machine_get_hwcaps).

This patch also fixes the memcheck/tests/vcpu_fnfns and
none/tests/amd64/fma testcases when run on a x86-64-v3 system.

Patch contributed by Grazvydas Ignotas <notasas@gmail.com> and
Bruno Lathuilière <bruno.lathuiliere@edf.fr>

https://bugs.kde.org/show_bug.cgi?id=481127
https://bugs.kde.org/show_bug.cgi?id=463463
https://bugs.kde.org/show_bug.cgi?id=463458

2 weeks agoAdd gethostbyname2_r to glibc-2.X-helgrind.supp.in
Mark Wielaard [Sat, 13 Apr 2024 01:03:26 +0000 (03:03 +0200)] 
Add gethostbyname2_r to glibc-2.X-helgrind.supp.in

This fixes helgrind/tests/getaddrinfo with glibc-2.28 which might
use gethostbyname2_r.

2 weeks agoregtest arm64: fixes for arm64 clang, part 5
Paul Floyd [Fri, 12 Apr 2024 18:50:39 +0000 (20:50 +0200)] 
regtest arm64: fixes for arm64 clang, part 5

last one for now
clang assembler not accepting sqrdmlah v0.4h, v1.4h, v2.4h[0]
It wants sqrdmlah v0.4h, v1.4h, v2.h[0]

2 weeks agoregtest arm64: fixes for arm64 clang, part 4
Paul Floyd [Fri, 12 Apr 2024 18:43:18 +0000 (20:43 +0200)] 
regtest arm64: fixes for arm64 clang, part 4

clang assembler being picky about unneeded annotations such as
LSL #0 but then insisting on a .0 that gas infers.

2 weeks agoregtest arm64: fixes for arm64 clang, part 3
Paul Floyd [Fri, 12 Apr 2024 18:28:05 +0000 (20:28 +0200)] 
regtest arm64: fixes for arm64 clang, part 3

clang assembler refuses to accept scvtf with q regs where v regs
should be used (again the arm manual agrees).

2 weeks agoregtest arm64: fixes for arm64 clang, part 2
Paul Floyd [Fri, 12 Apr 2024 18:14:18 +0000 (20:14 +0200)] 
regtest arm64: fixes for arm64 clang, part 2

More clang moans about using int where long long is expected
Also clang assembler refuses to accept opcodes like sub x21,x22,x23,sxtb
According to the arm manual it is right and it should be
sub x21,x22,w23,sxtb.
I asked on the LLVM discourse and they said they preferred to be strict
and unambiguous.

2 weeks agoregtest arm64: fixes for arm64 clang, part 1
Paul Floyd [Fri, 12 Apr 2024 17:59:35 +0000 (19:59 +0200)] 
regtest arm64: fixes for arm64 clang, part 1

clang warns about using integers where long longs are expected.
Secondly a few of the tests read the executable text. clang uses
different registers than GCC which causes a diff. Hence another
expected. The tests are potentially flaky so if we see instability
we need to modify the test.

2 weeks agolinux arm64: fix build with clang
Paul Floyd [Fri, 12 Apr 2024 10:50:07 +0000 (12:50 +0200)] 
linux arm64: fix build with clang

Same issue as https://bugs.kde.org/show_bug.cgi?id=421321
but with clang.

2 weeks agoAdd new helgrind race suppression for pth_mempcpy_false_races
Mark Wielaard [Fri, 12 Apr 2024 13:18:51 +0000 (15:18 +0200)] 
Add new helgrind race suppression for pth_mempcpy_false_races

In glibc 2.39+ the internal __printf_buffer_write might use memmove
instead of a memcpy variant. Add an extra suppression variant using
memmove.

This makes pth_mempcpy_false_races pass again.

See also https://bugs.kde.org/show_bug.cgi?id=450962

2 weeks agoAdd support for expensive cmpgt into vbits
Eyal Soha [Wed, 25 Oct 2023 03:52:56 +0000 (21:52 -0600)] 
Add support for expensive cmpgt into vbits

2 weeks agoAdd __set_vma_name suppression to drd/tests/std_thread2.supp
Mark Wielaard [Thu, 11 Apr 2024 17:48:22 +0000 (19:48 +0200)] 
Add __set_vma_name suppression to drd/tests/std_thread2.supp

glibc uses a atomic var to set whether the kernel supports
PR_SET_VMA_ANON_NAME. This looks like a conflicting access to drd.
Suppress it for this testcase.

2 weeks agoAdd memcheck/tests/linux/capget.stderr.exp4 variant
Mark Wielaard [Thu, 11 Apr 2024 16:29:01 +0000 (18:29 +0200)] 
Add memcheck/tests/linux/capget.stderr.exp4 variant

Add an expect file for capget where the inheritable bits are set,
when run as root.

3 weeks agoFreeBSD Helgrind: suppression for libstdc++ std::mutex::lock()
Paul Floyd [Mon, 8 Apr 2024 18:43:10 +0000 (20:43 +0200)] 
FreeBSD Helgrind: suppression for libstdc++ std::mutex::lock()

3 weeks agoDon't call vg_assert inside sync_signalhandler_from_kernel
Mark Wielaard [Fri, 5 Apr 2024 13:23:32 +0000 (15:23 +0200)] 
Don't call vg_assert inside sync_signalhandler_from_kernel

Calling vg_assert inside a core signal handler will mess up the core
backtrace. Making it even harder to find bugs in the core.

4 weeks agoUpdate libiberty demangler
Mark Wielaard [Wed, 3 Apr 2024 21:13:02 +0000 (23:13 +0200)] 
Update libiberty demangler

Update the libiberty demangler using the auxprogs/update-demangler
script to gcc git commit ca2f7c84927f85b95f0f48f82b93f1460c372db4.

This update includes:

* c++: mangle noexcept-expr [PR70790]
* c++: Fix templated convertion operator demangling
* c++: constrained hidden friends [PR109751]
* c++: mangle function template constraints
* Update copyright years.
* c++, demangle: Implement
  https://github.com/itanium-cxx-abi/cxx-abi/issues/148
  non-proposal
* libiberty: Invoke D demangler when --format=auto

4 weeks agoregtest: Add arm64 dir to DIST_SUBDIR
Paul Floyd [Wed, 3 Apr 2024 18:03:29 +0000 (20:03 +0200)] 
regtest: Add arm64 dir to DIST_SUBDIR

New directory was missing in memcheck/tests/Makefile.am

4 weeks agoBug 484935 - [patch] Valgrind reports false "Conditional jump or move depends on...
Paul Floyd [Wed, 3 Apr 2024 06:32:12 +0000 (08:32 +0200)] 
Bug 484935 - [patch] Valgrind reports false "Conditional jump or move depends on uninitialised value" errors for aarch64 signal handlers

Patch contributed by Peter Edwards <peadar@arista.com>

Testcase modified and added to a new directory, memcheck/tests/arm64

4 weeks agolinux regtest: merge the two prereq lines in memcheck sys-preadv2_pwritev2.vgtest
Paul Floyd [Tue, 2 Apr 2024 17:15:47 +0000 (19:15 +0200)] 
linux regtest: merge the two prereq lines in memcheck sys-preadv2_pwritev2.vgtest

I think only the last one gets used.

4 weeks agoregtest: fixes for old GCC versions
Paul Floyd [Tue, 2 Apr 2024 11:36:31 +0000 (13:36 +0200)] 
regtest: fixes for old GCC versions

Seen on arm64 with GCC 4.8.2

4 weeks agoregtest: one last change to none fdleak filter for arm64 linux
Paul Floyd [Mon, 1 Apr 2024 18:06:32 +0000 (20:06 +0200)] 
regtest: one last change to none fdleak filter for arm64 linux

4 weeks agoregtest: another fdleak filter in none/tests
Paul Floyd [Mon, 1 Apr 2024 17:59:20 +0000 (19:59 +0200)] 
regtest: another fdleak filter in none/tests

4 weeks agoFreeBSD DRD and Helgrind: add sem_clockwait_np wrappers
Paul Floyd [Sun, 31 Mar 2024 16:04:04 +0000 (18:04 +0200)] 
FreeBSD DRD and Helgrind: add sem_clockwait_np wrappers

Helgrind wrapper was complicated because sem_wait calls
sem_clockwait_np.

4 weeks agoDarwin: fix warning about unused Helgrind sem_timedwait wrapper
Paul Floyd [Sun, 31 Mar 2024 09:23:49 +0000 (11:23 +0200)] 
Darwin: fix warning about unused Helgrind sem_timedwait wrapper

4 weeks agoBug 484480 - False positives when using sem_trywait
Paul Floyd [Sat, 30 Mar 2024 15:31:12 +0000 (16:31 +0100)] 
Bug 484480 - False positives when using sem_trywait

4 weeks agoSolaris regtest: fix build of none socket_close
Paul Floyd [Sat, 30 Mar 2024 14:16:55 +0000 (15:16 +0100)] 
Solaris regtest: fix build of none socket_close

Also add to filter_fdleak

4 weeks agoSolaris: remove syscall close POST
Paul Floyd [Sat, 30 Mar 2024 12:25:13 +0000 (13:25 +0100)] 
Solaris: remove syscall close POST

And move the extra stuff that it did into PRE

4 weeks agoDarwin: add missing necp pieces
Paul Floyd [Fri, 29 Mar 2024 15:18:41 +0000 (16:18 +0100)] 
Darwin: add missing necp pieces

4 weeks agoDarwin regtest: add prereq for helgrind/tests/bug322621
Paul Floyd [Fri, 29 Mar 2024 14:50:03 +0000 (15:50 +0100)] 
Darwin regtest: add prereq for helgrind/tests/bug322621

It needs pthread_barrier which Darwin doesn't have

4 weeks agoDarwin: Helgrind suppressions for macOS 10.13
Paul Floyd [Fri, 29 Mar 2024 13:16:06 +0000 (14:16 +0100)] 
Darwin: Helgrind suppressions for macOS 10.13

4 weeks agoFreeBSD regtest: timerfd_settime has 4 args not 3
Paul Floyd [Fri, 29 Mar 2024 07:56:04 +0000 (08:56 +0100)] 
FreeBSD regtest: timerfd_settime has 4 args not 3

4 weeks agoregtest: rename none/tests/arm64/memory
Paul Floyd [Fri, 29 Mar 2024 06:33:29 +0000 (07:33 +0100)] 
regtest: rename none/tests/arm64/memory

This directory now contains a C++ test. 'memory' is also the name
of a C++ header. The include paths include ".". All that means that
if 'memory' builds before 'bug484426' then bug484426.cpp will try to
include the binary executable file 'memory' from the same directory
and not the header somewhere under /usr/include.