1 Release 3.23.0 (?? Apr 2024)
2 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
4 This release supports X86/Linux, AMD64/Linux, ARM32/Linux, ARM64/Linux,
5 PPC32/Linux, PPC64BE/Linux, PPC64LE/Linux, S390X/Linux, MIPS32/Linux,
6 MIPS64/Linux, ARM/Android, ARM64/Android, MIPS32/Android, X86/Android,
7 X86/Solaris, AMD64/Solaris, AMD64/MacOSX 10.12, X86/FreeBSD and
8 AMD64/FreeBSD. There is also preliminary support for X86/macOS 10.13,
9 AMD64/macOS 10.13 and nanoMIPS/Linux.
11 * ==================== CORE CHANGES ===================
13 * ================== PLATFORM CHANGES =================
15 * ==================== TOOL CHANGES ===================
17 * ==================== FIXED BUGS ====================
19 The following bugs have been fixed or resolved. Note that "n-i-bz"
20 stands for "not in bugzilla" -- that is, a bug that was reported to us
21 but never got a bugzilla entry. We encourage you to file bugs in
22 bugzilla (https://bugs.kde.org/enter_bug.cgi?product=valgrind) rather
23 than mailing the developers (or mailing lists) directly -- bugs that
24 are not entered into bugzilla tend to get forgotten about or ignored.
27 401284 False positive "Source and destination overlap in strncat"
28 437790 valgrind reports "Conditional jump or move depends on uninitialised
29 value" in memchr of macOS 10.12-10.15
30 460616 disInstr(arm64): unhandled instruction 0x4E819402 (dotprod/ASIMDDP)
31 475498 Add reallocarray wrapper
32 476320 Build failure with GCC
33 476535 Difference in allocation size for massif/tests/overloaded-new between
34 clang++/libc++ and g++/libstdc++
35 476708 valgrind-monitor.py regular expressions should use raw strings
36 476780 Extend strlcat and strlcpy wrappers to GNU libc
37 476787 Build of Valgrind 3.21.0 fails when SOLARIS_PT_SUNDWTRACE_THRP is
39 476887 WARNING: unhandled amd64-freebsd syscall: 578
41 To see details of a given bug, visit
42 https://bugs.kde.org/show_bug.cgi?id=XXXXXX
43 where XXXXXX is the bug number as listed above.
45 (3.23.0.RC1: ?? Apr 2024)
47 Release 3.22.0 (31 Oct 2023)
48 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
50 This release supports X86/Linux, AMD64/Linux, ARM32/Linux, ARM64/Linux,
51 PPC32/Linux, PPC64BE/Linux, PPC64LE/Linux, S390X/Linux, MIPS32/Linux,
52 MIPS64/Linux, ARM/Android, ARM64/Android, MIPS32/Android, X86/Android,
53 X86/Solaris, AMD64/Solaris, AMD64/MacOSX 10.12, X86/FreeBSD and
54 AMD64/FreeBSD. There is also preliminary support for X86/macOS 10.13,
55 AMD64/macOS 10.13 and nanoMIPS/Linux.
57 * ==================== CORE CHANGES ===================
59 * A new configure option --with-gdbscripts-dir lets you install
60 the gdb valgrind python monitor scripts in a specific location.
61 For example a distro could use it to install the scripts in a
62 safe load location --with-gdbscripts-dir=%{_datadir}/gdb/auto-load
63 It is also possible to configure --without-gdb-scripts-dir so no
64 .debug_gdb_scripts section is added to the vgpreload library and
65 no valgrind-monitor python scripts are installed at all.
67 * ================== PLATFORM CHANGES =================
69 * Support has been added for FreeBSD 14 and FreeBSD 15.
70 * Add support for the folllowing FreeBSD system calls:
71 close_range, kqueuex, membarrier, timerfd_create,
72 timerfd_settime and timerfd_gettime (all added in FreeBSD 15).
74 * ==================== TOOL CHANGES ===================
76 * Memcheck now tests and warns about the values used for
77 alignment and size. These apply to various functions: memalign,
78 posix_memalign and aligned_alloc in C and various overloads
79 of operators new and delete in C++. The kinds of error that can
81 - invalid alignment, for instance the alignment is usually required
83 - mismatched alignment between aligned allocation and aligned
85 - mismatched size when sized delete is used
86 - bad size for functions that have implementation defined behaviour
87 when the requested size is zero
90 - You can now profile part of a program's execution using the new
91 `CACHEGRIND_START_INSTRUMENTATION` and `CACHEGRIND_STOP_INSTRUMENTATION`
92 client requests, along with the new `--instr-at-start` option. The
93 behaviour is the same as Callgrind's equivalent functionality.
95 * ==================== FIXED BUGS ====================
97 The following bugs have been fixed or resolved. Note that "n-i-bz"
98 stands for "not in bugzilla" -- that is, a bug that was reported to us
99 but never got a bugzilla entry. We encourage you to file bugs in
100 bugzilla (https://bugs.kde.org/enter_bug.cgi?product=valgrind) rather
101 than mailing the developers (or mailing lists) directly -- bugs that
102 are not entered into bugzilla tend to get forgotten about or ignored.
104 390871 ELF debug info reader confused with multiple .rodata* sections
105 417993 vbit-test fail on s390x with Iop_Add32: spurious dependency on uninit
106 426751 Valgrind reports "still reachable" memory using musl
107 (alpine running inside docker)
108 432801 Valgrind 3.16.1 reports a jump based on uninitialized memory somehow
109 related to clang and signals
110 433857 Add validation to C++17 aligned new/delete alignment size
111 433859 Add mismatched detection to C++ 17 aligned new/delete
112 460192 Add epoll_pwait2
113 461074 DWARF2 CFI reader: unhandled DW_OP_ 0x11 (consts) DW_OP_ 0x92 (bregx)
114 465782 s390x: Valgrind doesn't compile with Clang on s390x
115 466105 aligned_alloc problems, part 2
116 467441 Add mismatched detection to C++ 14 sized delete
117 469049 link failure on ppc64 (big endian) valgrind 3.20
118 469146 massif --ignore-fn does not ignore inlined functions
119 469768 Make it possible to install gdb scripts in a different location
120 470121 Can't run callgrind_control with valgrind 3.21.0 because of perl errors
121 470132 s390x: Assertion failure on VGM instruction
122 470520 Multiple realloc zero errors crash in MC_(eq_Error)
123 470713 Failure on the Yosys project: valgrind: m_libcfile.c:1802
124 (Bool vgPlain_realpath(const HChar *, HChar *)):
125 Assertion 'resolved' failed
126 470830 Don't print actions vgdb me ... continue for vgdb --multi mode
127 470978 s390x: Valgrind cannot start qemu-kvm when "sysctl vm.allocate_pgste=0"
128 471311 gdb --multi mode stdout redirecting to stderr
129 471807 Add support for lazy reading and downloading of DWARF debuginfo
130 472219 Syscall param ppoll(ufds.events) points to uninitialised byte(s)
131 472875 none/tests/s390x/dfp-1 failure
132 472963 Broken regular expression in configure.ac
133 473604 Fix bug472219.c compile failure with Clang 16
134 473677 make check compile failure with Clang 16 based on GCC 13.x
135 473745 must-be-redirected function - strlen
136 473870 FreeBSD 14 applications fail early at startup
137 473944 Handle mold linker split RW PT_LOAD segments correctly
138 474332 aligned_alloc under Valgrind returns nullptr when alignment is not a multiple of sizeof(void *)
139 475650 DRD does not work with C11 threads
140 475652 Missing suppression for __wcsncpy_avx2 (strncpy-avx2.S:308)?
141 476108 vg_replace_malloc DELETE checks size
142 n-i-bz Allow arguments with spaces in .valgrindrc files
143 n-i-bz FreeBSD fixed reading of Valgrind tools own debuginfo
145 To see details of a given bug, visit
146 https://bugs.kde.org/show_bug.cgi?id=XXXXXX
147 where XXXXXX is the bug number as listed above.
149 (3.22.0.RC1: 17 Oct 2023)
150 (3.22.0.RC2: 26 Oct 2023)
152 Release 3.21.0 (28 Apr 2023)
153 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
155 This release supports X86/Linux, AMD64/Linux, ARM32/Linux, ARM64/Linux,
156 PPC32/Linux, PPC64BE/Linux, PPC64LE/Linux, S390X/Linux, MIPS32/Linux,
157 MIPS64/Linux, ARM/Android, ARM64/Android, MIPS32/Android, X86/Android,
158 X86/Solaris, AMD64/Solaris, AMD64/MacOSX 10.12, X86/FreeBSD and
159 AMD64/FreeBSD. There is also preliminary support for X86/macOS 10.13,
160 AMD64/macOS 10.13 and nanoMIPS/Linux.
162 * ==================== CORE CHANGES ===================
164 * When GDB is used to debug a program running under valgrind using
165 the valgrind gdbserver, GDB will automatically load some
166 python code provided in valgrind defining GDB front end commands
167 corresponding to the valgrind monitor commands.
168 These GDB front end commands accept the same format as
169 the monitor commands directly sent to the Valgrind gdbserver.
170 These GDB front end commands provide a better integration
171 in the GDB command line interface, so as to use for example
172 GDB auto-completion, command specific help, searching for
173 a command or command help matching a regexp, ...
174 For relevant monitor commands, GDB will evaluate arguments
175 to make the use of monitor commands easier.
176 For example, instead of having to print the address of a variable
177 to pass it to a subsequent monitor command, the GDB front end
178 command will evaluate the address argument. It is for example
180 (gdb) memcheck who_points_at &some_struct sizeof(some_struct)
183 $2 = (some_struct_type *) 0x1130a0 <some_struct>
184 (gdb) p sizeof(some_struct)
186 (gdb) monitor who_point_at 0x1130a0 40
188 * The vgdb utility now supports extended-remote protocol when
189 invoked with --multi. In this mode the GDB run command is
190 supported. Which means you don't need to run gdb and valgrind
191 from different terminals. So for example to start your program
192 in gdb and run it under valgrind you can do:
194 (gdb) set remote exec-file prog
196 (gdb) target extended-remote | vgdb --multi
199 * The behaviour of realloc with a size of zero can now
200 be changed for tools that intercept malloc. Those
201 tools are memcheck, helgrind, drd, massif and dhat.
202 Realloc implementations generally do one of two things
203 - free the memory like free() and return NULL
204 (GNU libc and ptmalloc).
205 - either free the memory and then allocate a
206 minimum sized block or just return the
207 original pointer. Return NULL if the
208 allocation of the minimum sized block fails
209 (jemalloc, musl, snmalloc, Solaris, macOS).
210 When Valgrind is configured and built it will
211 try to match the OS and libc behaviour. However
212 if you are using a non-default library to replace
213 malloc and family (e.g., musl on a glibc Linux or
214 tcmalloc on FreeBSD) then you can use a command line
215 option to change the behaviour of Valgrind:
216 --realloc-zero-bytes-frees=yes|no [yes on Linux glibc, no otherwise]
218 * ================== PLATFORM CHANGES =================
220 * Make the address space limit on FreeBSD amd64 128Gbytes
221 (the same as Linux and Solaris, it was 32Gbytes)
223 * ==================== TOOL CHANGES ===================
226 - When doing a delta leak_search, it is now possible to only
227 output the new loss records compared to the previous leak search.
228 This is available in the memcheck monitor command 'leak_search'
229 by specifying the "new" keyword or in your program by using
230 the client request VALGRIND_DO_NEW_LEAK_CHECK.
231 Whenever a "delta" leak search is done (i.e. when specifying
232 "new" or "increased" or "changed" in the monitor command),
233 the new loss records have a "new" marker.
234 - Valgrind now contains python code that defines GDB memcheck
235 front end monitor commands. See CORE CHANGES.
236 - Performs checks for the use of realloc with a size of zero.
237 This is non-portable and a source of errors. If memcheck
238 detects such a usage it will generate an error
239 realloc() with size 0
240 followed by the usual callstacks.
241 A switch has been added to allow this to be turned off:
242 --show-realloc-size-zero=yes|no [yes]
245 - The option ---history-backtrace-size=<number> allows to configure
246 the number of entries to record in the stack traces of "old"
247 accesses. Previously, this number was hardcoded to 8.
248 - Valgrind now contains python code that defines GDB helgrind
249 front end monitor commands. See CORE CHANGES.
252 - `--cache-sim=no` is now the default. The cache simulation is old and
253 unlikely to match any real modern machine. This means only the `Ir`
254 event are gathered by default, but that is by far the most useful
256 - `cg_annotate`, `cg_diff`, and `cg_merge` have been rewritten in
257 Python. As a result, they all have more flexible command line
258 argument handling, e.g. supporting `--show-percs` and
259 `--no-show-percs` forms as well as the existing `--show-percs=yes`
260 and `--show-percs=no`.
261 - `cg_annotate` has some functional changes.
262 - It's much faster, e.g. 3-4x on common cases.
263 - It now supports diffing (with `--diff`, `--mod-filename`, and
264 `--mod-funcname`) and merging (by passing multiple data files).
265 - It now provides more information at the file and function level.
266 There are now "File:function" and "Function:file" sections. These
267 are very useful for programs that use inlining a lot.
268 - Support for user-annotated files and the `-I`/`--include` option
269 has been removed, because it was of little use and blocked other
271 - The `--auto` option is renamed `--annotate`, though the old
272 `--auto=yes`/`--auto=no` forms are still supported.
273 - `cg_diff` and `cg_merge` are now deprecated, because `cg_annotate`
274 now does a better job of diffing and merging.
275 - The Cachegrind output file format has changed very slightly, but in
276 ways nobody is likely to notice.
279 - Valgrind now contains python code that defines GDB callgrind
280 front end monitor commands. See CORE CHANGES.
283 - Valgrind now contains python code that defines GDB massif
284 front end monitor commands. See CORE CHANGES.
287 - A new kind of user request has been added which allows you to
288 override the 1024 byte limit on access count histograms for blocks
289 of memory. The client request is DHAT_HISTOGRAM_MEMORY.
291 * ==================== FIXED BUGS ====================
293 The following bugs have been fixed or resolved. Note that "n-i-bz"
294 stands for "not in bugzilla" -- that is, a bug that was reported to us
295 but never got a bugzilla entry. We encourage you to file bugs in
296 bugzilla (https://bugs.kde.org/enter_bug.cgi?product=valgrind) rather
297 than mailing the developers (or mailing lists) directly -- bugs that
298 are not entered into bugzilla tend to get forgotten about or ignored.
300 170510 Don't warn about ioctl of size 0 without direction hint
301 241072 List tools in --help output
302 327548 false positive while destroying mutex
303 382034 Testcases build fixes for musl
304 351857 confusing error message about valid command line option
305 374596 inconsistent RDTSCP support on x86_64
306 392331 Spurious lock not held error from inside pthread_cond_timedwait
307 397083 Likely false positive "uninitialised value(s)" for __wmemchr_avx2 and __wmemcmp_avx2_movbe
308 400793 pthread_rwlock_timedwrlock false positive
309 419054 Unhandled syscall getcpu on arm32
310 433873 openat2 syscall unimplemented on Linux
311 434057 Add stdio mode to valgrind's gdbserver
312 435441 valgrind fails to interpose malloc on musl 1.2.2 due to weak symbol name and no libc soname
313 436413 Warn about realloc of size zero
314 439685 compiler warning in callgrind/main.c
315 444110 priv/guest_ppc_toIR.c:36198:31: warning: duplicated 'if' condition.
316 444487 hginfo test detects an extra lock inside data symbol "_rtld_local"
317 444488 Use glibc.pthread.stack_cache_size tunable
318 444568 drd/tests/pth_barrier_thr_cr fails on Fedora 38
319 445743 "The impossible happened: mutex is locked simultaneously by two threads"
320 while using mutexes with priority inheritance and signals
321 449309 Missing loopback device ioctl(s)
322 459476 vgdb: allow address reuse to avoid "address already in use" errorsuse" errors
323 460356 s390: Sqrt32Fx4 -- cannot reduce tree
324 462830 WARNING: unhandled amd64-freebsd syscall: 474
325 463027 broken check for MPX instruction support in assembler
326 464103 Enhancement: add a client request to DHAT to mark memory to be histogrammed
327 464476 Firefox fails to start under Valgrind
328 464609 Valgrind memcheck should support Linux pidfd_open
329 464680 Show issues caused by memory policies like selinux deny_execmem
330 464859 Build failures with GCC-13 (drd tsan_unittest)
331 464969 D language demangling
332 465435 m_libcfile.c:66 (vgPlain_safe_fd): Assertion 'newfd >= VG_(fd_hard_limit)' failed.
333 466104 aligned_alloc problems, part 1
334 467036 Add time cost statistics for Regtest
335 467482 Build failure on aarch64 Alpine
336 467714 fdleak_* and rlimit tests fail when parent process has more than
337 64 descriptors opened
338 467839 Gdbserver: Improve compatibility of library directory name
339 468401 [PATCH] Add a style file for clang-format
340 468556 Build failure for vgdb
341 468606 build: remove "Valgrind relies on GCC" check/output
342 469097 ppc64(be) doesn't support SCV syscall instruction
343 n-i-bz FreeBSD rfork syscall fail with EINVAL or ENOSYS rather than VG_(unimplemented)
345 To see details of a given bug, visit
346 https://bugs.kde.org/show_bug.cgi?id=XXXXXX
347 where XXXXXX is the bug number as listed above.
349 * ==================== KNOWN ISSUES ===================
351 * configure --enable-lto=yes is know to not work in all setups.
352 See bug 469049. Workaround: Build without LTO.
354 (3.21.0.RC1: 14 Apr 2023)
355 (3.21.0.RC2: 21 Apr 2023)
357 Release 3.20.0 (24 Oct 2022)
358 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
360 This release supports X86/Linux, AMD64/Linux, ARM32/Linux, ARM64/Linux,
361 PPC32/Linux, PPC64BE/Linux, PPC64LE/Linux, S390X/Linux, MIPS32/Linux,
362 MIPS64/Linux, ARM/Android, ARM64/Android, MIPS32/Android, X86/Android,
363 X86/Solaris, AMD64/Solaris, AMD64/MacOSX 10.12, X86/FreeBSD and
364 AMD64/FreeBSD. There is also preliminary support for X86/macOS 10.13,
365 AMD64/macOS 10.13 and nanoMIPS/Linux.
367 * ==================== CORE CHANGES ===================
369 * The option "--vgdb-stop-at=event1,event2,..." accepts the new value abexit.
370 This indicates to invoke gdbserver when your program exits abnormally
371 (i.e. with a non zero exit code).
372 * Fix Rust v0 name demangling.
373 * The Linux rseq syscall is now implemented as (silently) returning ENOSYS.
374 * Add FreeBSD syscall wrappers for __specialfd and __realpathat.
375 * Remove FreeBSD dependencies on COMPAT10, which fixes compatibility with
377 * The option --enable-debuginfod=<no|yes> [default: yes] has been added on
379 * More DWARF5 support as generated by clang14.
381 * ==================== FIXED BUGS ====================
383 The following bugs have been fixed or resolved. Note that "n-i-bz"
384 stands for "not in bugzilla" -- that is, a bug that was reported to us
385 but never got a bugzilla entry. We encourage you to file bugs in
386 bugzilla (https://bugs.kde.org/enter_bug.cgi?product=valgrind) rather
387 than mailing the developers (or mailing lists) directly -- bugs that
388 are not entered into bugzilla tend to get forgotten about or ignored.
390 131186 writev reports error in (vector[...])
391 434764 iconv_open causes ld.so v2.28+ to use optimised strncmp
392 446754 Improve error codes from alloc functions under memcheck
393 452274 memcheck crashes with Assertion 'sci->status.what == SsIdle' failed
394 452779 Valgrind fails to build on FreeBSD 13.0 with llvm-devel (15.0.0)
395 453055 shared_timed_mutex drd test fails with "Lock shared failed" message
396 453602 Missing command line option to enable/disable debuginfod
397 452802 Handle lld 9+ split RW PT_LOAD segments correctly
398 454040 s390x: False-positive memcheck:cond in memmem on arch13 systems
399 456171 [PATCH] FreeBSD: Don't record address errors when accessing the 'kern.ps_strings' sysctl struct
400 n-i-bz Implement vgdb invoker on FreeBSD
401 458845 PowerPC: The L field for the dcbf and sync instruction should be
403 458915 Remove register cache to fix 458915 gdbserver causes wrong syscall return
404 459031 Documentation on --error-exitcode incomplete
405 459477 XERROR messages lacks ending '\n' in vgdb
406 462007 Implicit int in none/tests/faultstatus.c
408 To see details of a given bug, visit
409 https://bugs.kde.org/show_bug.cgi?id=XXXXXX
410 where XXXXXX is the bug number as listed above.
412 (3.20.0.RC1: 20 Oct 2022)
415 Release 3.19.0 (11 Apr 2022)
416 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
418 This release supports X86/Linux, AMD64/Linux, ARM32/Linux, ARM64/Linux,
419 PPC32/Linux, PPC64BE/Linux, PPC64LE/Linux, S390X/Linux, MIPS32/Linux,
420 MIPS64/Linux, ARM/Android, ARM64/Android, MIPS32/Android, X86/Android,
421 X86/Solaris, AMD64/Solaris, AMD64/MacOSX 10.12, X86/FreeBSD and
422 AMD64/FreeBSD. There is also preliminary support for X86/macOS 10.13,
423 AMD64/macOS 10.13 and nanoMIPS/Linux.
425 * ==================== CORE CHANGES ===================
427 * Fix Rust v0 name demangling.
428 * The Linux rseq syscall is now implemented as (silently) returning ENOSYS.
429 * Add FreeBSD syscall wrappers for __specialfd and __realpathat.
430 * Remove FreeBSD dependencies on COMPAT10, which fixes compatibility with HardenedBSD
432 * ================== PLATFORM CHANGES =================
435 - ignore the "v8.x" architecture levels, only look at actual CPU features
436 present. Fixes mismatch detected between RDMA and atomics features
437 preventing startup on some QEMU configurations.
438 - Implement LD{,A}XP and ST{,L}XP
439 - Fix incorrect code emitted for doubleword CAS.
442 - Fix sys_ipc semtimedop syscall
443 - Fix VFLRX and WFLRX instructions
444 - Fix EXRL instruction with negative offset
447 - Reimplement the vbpermq instruction support to generate less Iops and
448 avoid overflowing internal buffers.
449 - Fix checking for scv support to avoid "Facility 'SCV' unavailable (12),
450 exception" messages in dmsg.
451 - Fix setting condition code for Vector Compare quad word instructions.
452 - Fix fix lxsibzx, lxsihzx and lxsihzx instructions so they only load
453 their respective sized data.
454 - Fix the prefixed stq instruction in PC relative mode.
456 * ==================== TOOL CHANGES ===================
459 - Speed up --track-origins=yes for large (in the range of hundreds to
460 thousands of megabytes) mmap/munmaps.
462 - Several fixes for new versions of libstd++ using new posix try_lock
465 * ==================== FIXED BUGS ====================
467 The following bugs have been fixed or resolved. Note that "n-i-bz"
468 stands for "not in bugzilla" -- that is, a bug that was reported to us
469 but never got a bugzilla entry. We encourage you to file bugs in
470 bugzilla (https://bugs.kde.org/enter_bug.cgi?product=valgrind) rather
471 than mailing the developers (or mailing lists) directly -- bugs that
472 are not entered into bugzilla tend to get forgotten about or ignored.
474 403802 leak_cpp_interior fails with some reachable blocks different than expected
475 435732 memcheck/tests/leak_cpp_interior fails with gcc11
476 444242 s390x: Valgrind crashes on EXRL with negative offset
477 444399 arm64: unhandled instruction 0xC87F2D89 (LD{,A}XP and ST{,L}XP).
479 444481 gdb_server test failures on s390x
480 444495 dhat/tests/copy fails on s390x
481 444552 memcheck/tests/sem fails on s390x with glibc 2.34
482 444571 PPC, fix the lxsibzx and lxsihzx so they only load their respective
484 444836 PPC, pstq instruction for R=1 is not storing to the correct address.
485 444925 fexecve syscall wrapper not properly implemented
486 445032 valgrind/memcheck crash with SIGSEGV when SIGVTALRM timer used and
488 445211 Fix out of tree builds
489 445300 [PATCH] Fix building tests with Musl
490 445011 SIGCHLD is sent when valgrind uses debuginfod-find
491 445354 arm64 backend: incorrect code emitted for doubleword CAS
492 445415 arm64 front end: alignment checks missing for atomic instructions
493 445504 Using C++ condition_variable results in bogus "mutex is locked simultaneously by two threads" warning
494 445607 Unhandled amd64-freebsd syscall: 247
495 445668 Inline stack frame generation is broken for Rust binaries
496 445916 Demangle Rust v0 symbols with .llvm suffix
497 446139 DRD/Helgrind with std::shared_timed_mutex::try_lock_until and try_lock_shared_until false positives
498 446138 DRD/Helgrind with std::timed_mutex::try_lock_until false positives
499 446281 Add a DRD suppression for fwrite
500 446103 Memcheck: `--track-origins=yes` causes extreme slowdowns for large mmap/munmap
501 446139 DRD/Helgrind with std::shared_timed_mutex::try_lock_until and try_lock_shared_until false
502 446251 TARGET_SIGNAL_THR added to enum target_signal
503 446823 FreeBSD - missing syscalls when using libzm4
504 447991 s390x: Valgrind indicates illegal instruction on wflrx
505 447995 Valgrind segfault on power10 due to hwcap checking code
506 449483 Powerpc: vcmpgtsq., vcmpgtuq,, vcmpequq. instructions not setting the
507 condition code correctly.
508 449672 ppc64 --track-origins=yes failures because of bad cmov addHRegUse
509 449838 sigsegv liburing the 'impossible' happened for io_uring_setup
510 450025 Powerc: ACC file not implemented as a logical overlay of the VSR
512 450437 Warn for execve syscall with argv or argv[0] being NULL
513 450536 Powerpc: valgrind throws 'facility scv unavailable exception'
514 451626 Syscall param bpf(attr->raw_tracepoint.name) points to unaddressable byte(s)
515 451827 [ppc64le] VEX temporary storage exhausted with several vbpermq instructions
516 451843 valgrind fails to start on a FreeBSD system which enforces W^X
518 To see details of a given bug, visit
519 https://bugs.kde.org/show_bug.cgi?id=XXXXXX
520 where XXXXXX is the bug number as listed above.
522 (3.19.0.RC1: 02 Apr 2022)
523 (3.19.0.RC2: 08 Apr 2022)