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 281059 Cannot connect to Oracle using valgrind
28 390269 unhandled amd64-darwin syscall: unix:464 (openat_nocancel)
29 401284 False positive "Source and destination overlap in strncat"
30 428364 Signals inside io_uring_enter not handled
31 437790 valgrind reports "Conditional jump or move depends on uninitialised
32 value" in memchr of macOS 10.12-10.15
33 460616 disInstr(arm64): unhandled instruction 0x4E819402 (dotprod/ASIMDDP)
34 466884 Missing writev uninit padding suppression for _XSend
35 475498 Add reallocarray wrapper
36 476320 Build failure with GCC
37 476535 Difference in allocation size for massif/tests/overloaded-new between
38 clang++/libc++ and g++/libstdc++
39 476548 valgrind 3.22.0 fails on assertion when loading debuginfo file produced by mold
40 476708 valgrind-monitor.py regular expressions should use raw strings
41 476780 Extend strlcat and strlcpy wrappers to GNU libc
42 476787 Build of Valgrind 3.21.0 fails when SOLARIS_PT_SUNDWTRACE_THRP is
44 476887 WARNING: unhandled amd64-freebsd syscall: 578
45 477198 Add fchmodat2 syscall on linux
46 477628 Add mremap support for Solaris
47 477630 Include ucontext.h rather than sys/ucontext.h in Solaris sources
48 477719 vgdb incorrectly replies to qRcmd packet
49 478211 Redundant code for vgdb.c and Valgrind core tools
50 478624 Valgrind incompatibility with binutils-2.42 on x86 with new nop patterns
51 (unhandled instruction bytes: 0x2E 0x8D 0xB4 0x26
52 478837 valgrind fails to read debug info for rust binaries
53 n-i-bz Add redirect for memccpy
55 To see details of a given bug, visit
56 https://bugs.kde.org/show_bug.cgi?id=XXXXXX
57 where XXXXXX is the bug number as listed above.
59 (3.23.0.RC1: ?? Apr 2024)
61 Release 3.22.0 (31 Oct 2023)
62 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
64 This release supports X86/Linux, AMD64/Linux, ARM32/Linux, ARM64/Linux,
65 PPC32/Linux, PPC64BE/Linux, PPC64LE/Linux, S390X/Linux, MIPS32/Linux,
66 MIPS64/Linux, ARM/Android, ARM64/Android, MIPS32/Android, X86/Android,
67 X86/Solaris, AMD64/Solaris, AMD64/MacOSX 10.12, X86/FreeBSD and
68 AMD64/FreeBSD. There is also preliminary support for X86/macOS 10.13,
69 AMD64/macOS 10.13 and nanoMIPS/Linux.
71 * ==================== CORE CHANGES ===================
73 * A new configure option --with-gdbscripts-dir lets you install
74 the gdb valgrind python monitor scripts in a specific location.
75 For example a distro could use it to install the scripts in a
76 safe load location --with-gdbscripts-dir=%{_datadir}/gdb/auto-load
77 It is also possible to configure --without-gdb-scripts-dir so no
78 .debug_gdb_scripts section is added to the vgpreload library and
79 no valgrind-monitor python scripts are installed at all.
81 * ================== PLATFORM CHANGES =================
83 * Support has been added for FreeBSD 14 and FreeBSD 15.
84 * Add support for the folllowing FreeBSD system calls:
85 close_range, kqueuex, membarrier, timerfd_create,
86 timerfd_settime and timerfd_gettime (all added in FreeBSD 15).
88 * ==================== TOOL CHANGES ===================
90 * Memcheck now tests and warns about the values used for
91 alignment and size. These apply to various functions: memalign,
92 posix_memalign and aligned_alloc in C and various overloads
93 of operators new and delete in C++. The kinds of error that can
95 - invalid alignment, for instance the alignment is usually required
97 - mismatched alignment between aligned allocation and aligned
99 - mismatched size when sized delete is used
100 - bad size for functions that have implementation defined behaviour
101 when the requested size is zero
104 - You can now profile part of a program's execution using the new
105 `CACHEGRIND_START_INSTRUMENTATION` and `CACHEGRIND_STOP_INSTRUMENTATION`
106 client requests, along with the new `--instr-at-start` option. The
107 behaviour is the same as Callgrind's equivalent functionality.
109 * ==================== FIXED BUGS ====================
111 The following bugs have been fixed or resolved. Note that "n-i-bz"
112 stands for "not in bugzilla" -- that is, a bug that was reported to us
113 but never got a bugzilla entry. We encourage you to file bugs in
114 bugzilla (https://bugs.kde.org/enter_bug.cgi?product=valgrind) rather
115 than mailing the developers (or mailing lists) directly -- bugs that
116 are not entered into bugzilla tend to get forgotten about or ignored.
118 390871 ELF debug info reader confused with multiple .rodata* sections
119 417993 vbit-test fail on s390x with Iop_Add32: spurious dependency on uninit
120 426751 Valgrind reports "still reachable" memory using musl
121 (alpine running inside docker)
122 432801 Valgrind 3.16.1 reports a jump based on uninitialized memory somehow
123 related to clang and signals
124 433857 Add validation to C++17 aligned new/delete alignment size
125 433859 Add mismatched detection to C++ 17 aligned new/delete
126 460192 Add epoll_pwait2
127 461074 DWARF2 CFI reader: unhandled DW_OP_ 0x11 (consts) DW_OP_ 0x92 (bregx)
128 465782 s390x: Valgrind doesn't compile with Clang on s390x
129 466105 aligned_alloc problems, part 2
130 467441 Add mismatched detection to C++ 14 sized delete
131 469049 link failure on ppc64 (big endian) valgrind 3.20
132 469146 massif --ignore-fn does not ignore inlined functions
133 469768 Make it possible to install gdb scripts in a different location
134 470121 Can't run callgrind_control with valgrind 3.21.0 because of perl errors
135 470132 s390x: Assertion failure on VGM instruction
136 470520 Multiple realloc zero errors crash in MC_(eq_Error)
137 470713 Failure on the Yosys project: valgrind: m_libcfile.c:1802
138 (Bool vgPlain_realpath(const HChar *, HChar *)):
139 Assertion 'resolved' failed
140 470830 Don't print actions vgdb me ... continue for vgdb --multi mode
141 470978 s390x: Valgrind cannot start qemu-kvm when "sysctl vm.allocate_pgste=0"
142 471311 gdb --multi mode stdout redirecting to stderr
143 471807 Add support for lazy reading and downloading of DWARF debuginfo
144 472219 Syscall param ppoll(ufds.events) points to uninitialised byte(s)
145 472875 none/tests/s390x/dfp-1 failure
146 472963 Broken regular expression in configure.ac
147 473604 Fix bug472219.c compile failure with Clang 16
148 473677 make check compile failure with Clang 16 based on GCC 13.x
149 473745 must-be-redirected function - strlen
150 473870 FreeBSD 14 applications fail early at startup
151 473944 Handle mold linker split RW PT_LOAD segments correctly
152 474332 aligned_alloc under Valgrind returns nullptr when alignment is not a multiple of sizeof(void *)
153 475650 DRD does not work with C11 threads
154 475652 Missing suppression for __wcsncpy_avx2 (strncpy-avx2.S:308)?
155 476108 vg_replace_malloc DELETE checks size
156 n-i-bz Allow arguments with spaces in .valgrindrc files
157 n-i-bz FreeBSD fixed reading of Valgrind tools own debuginfo
159 To see details of a given bug, visit
160 https://bugs.kde.org/show_bug.cgi?id=XXXXXX
161 where XXXXXX is the bug number as listed above.
163 (3.22.0.RC1: 17 Oct 2023)
164 (3.22.0.RC2: 26 Oct 2023)
166 Release 3.21.0 (28 Apr 2023)
167 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
169 This release supports X86/Linux, AMD64/Linux, ARM32/Linux, ARM64/Linux,
170 PPC32/Linux, PPC64BE/Linux, PPC64LE/Linux, S390X/Linux, MIPS32/Linux,
171 MIPS64/Linux, ARM/Android, ARM64/Android, MIPS32/Android, X86/Android,
172 X86/Solaris, AMD64/Solaris, AMD64/MacOSX 10.12, X86/FreeBSD and
173 AMD64/FreeBSD. There is also preliminary support for X86/macOS 10.13,
174 AMD64/macOS 10.13 and nanoMIPS/Linux.
176 * ==================== CORE CHANGES ===================
178 * When GDB is used to debug a program running under valgrind using
179 the valgrind gdbserver, GDB will automatically load some
180 python code provided in valgrind defining GDB front end commands
181 corresponding to the valgrind monitor commands.
182 These GDB front end commands accept the same format as
183 the monitor commands directly sent to the Valgrind gdbserver.
184 These GDB front end commands provide a better integration
185 in the GDB command line interface, so as to use for example
186 GDB auto-completion, command specific help, searching for
187 a command or command help matching a regexp, ...
188 For relevant monitor commands, GDB will evaluate arguments
189 to make the use of monitor commands easier.
190 For example, instead of having to print the address of a variable
191 to pass it to a subsequent monitor command, the GDB front end
192 command will evaluate the address argument. It is for example
194 (gdb) memcheck who_points_at &some_struct sizeof(some_struct)
197 $2 = (some_struct_type *) 0x1130a0 <some_struct>
198 (gdb) p sizeof(some_struct)
200 (gdb) monitor who_point_at 0x1130a0 40
202 * The vgdb utility now supports extended-remote protocol when
203 invoked with --multi. In this mode the GDB run command is
204 supported. Which means you don't need to run gdb and valgrind
205 from different terminals. So for example to start your program
206 in gdb and run it under valgrind you can do:
208 (gdb) set remote exec-file prog
210 (gdb) target extended-remote | vgdb --multi
213 * The behaviour of realloc with a size of zero can now
214 be changed for tools that intercept malloc. Those
215 tools are memcheck, helgrind, drd, massif and dhat.
216 Realloc implementations generally do one of two things
217 - free the memory like free() and return NULL
218 (GNU libc and ptmalloc).
219 - either free the memory and then allocate a
220 minimum sized block or just return the
221 original pointer. Return NULL if the
222 allocation of the minimum sized block fails
223 (jemalloc, musl, snmalloc, Solaris, macOS).
224 When Valgrind is configured and built it will
225 try to match the OS and libc behaviour. However
226 if you are using a non-default library to replace
227 malloc and family (e.g., musl on a glibc Linux or
228 tcmalloc on FreeBSD) then you can use a command line
229 option to change the behaviour of Valgrind:
230 --realloc-zero-bytes-frees=yes|no [yes on Linux glibc, no otherwise]
232 * ================== PLATFORM CHANGES =================
234 * Make the address space limit on FreeBSD amd64 128Gbytes
235 (the same as Linux and Solaris, it was 32Gbytes)
237 * ==================== TOOL CHANGES ===================
240 - When doing a delta leak_search, it is now possible to only
241 output the new loss records compared to the previous leak search.
242 This is available in the memcheck monitor command 'leak_search'
243 by specifying the "new" keyword or in your program by using
244 the client request VALGRIND_DO_NEW_LEAK_CHECK.
245 Whenever a "delta" leak search is done (i.e. when specifying
246 "new" or "increased" or "changed" in the monitor command),
247 the new loss records have a "new" marker.
248 - Valgrind now contains python code that defines GDB memcheck
249 front end monitor commands. See CORE CHANGES.
250 - Performs checks for the use of realloc with a size of zero.
251 This is non-portable and a source of errors. If memcheck
252 detects such a usage it will generate an error
253 realloc() with size 0
254 followed by the usual callstacks.
255 A switch has been added to allow this to be turned off:
256 --show-realloc-size-zero=yes|no [yes]
259 - The option ---history-backtrace-size=<number> allows to configure
260 the number of entries to record in the stack traces of "old"
261 accesses. Previously, this number was hardcoded to 8.
262 - Valgrind now contains python code that defines GDB helgrind
263 front end monitor commands. See CORE CHANGES.
266 - `--cache-sim=no` is now the default. The cache simulation is old and
267 unlikely to match any real modern machine. This means only the `Ir`
268 event are gathered by default, but that is by far the most useful
270 - `cg_annotate`, `cg_diff`, and `cg_merge` have been rewritten in
271 Python. As a result, they all have more flexible command line
272 argument handling, e.g. supporting `--show-percs` and
273 `--no-show-percs` forms as well as the existing `--show-percs=yes`
274 and `--show-percs=no`.
275 - `cg_annotate` has some functional changes.
276 - It's much faster, e.g. 3-4x on common cases.
277 - It now supports diffing (with `--diff`, `--mod-filename`, and
278 `--mod-funcname`) and merging (by passing multiple data files).
279 - It now provides more information at the file and function level.
280 There are now "File:function" and "Function:file" sections. These
281 are very useful for programs that use inlining a lot.
282 - Support for user-annotated files and the `-I`/`--include` option
283 has been removed, because it was of little use and blocked other
285 - The `--auto` option is renamed `--annotate`, though the old
286 `--auto=yes`/`--auto=no` forms are still supported.
287 - `cg_diff` and `cg_merge` are now deprecated, because `cg_annotate`
288 now does a better job of diffing and merging.
289 - The Cachegrind output file format has changed very slightly, but in
290 ways nobody is likely to notice.
293 - Valgrind now contains python code that defines GDB callgrind
294 front end monitor commands. See CORE CHANGES.
297 - Valgrind now contains python code that defines GDB massif
298 front end monitor commands. See CORE CHANGES.
301 - A new kind of user request has been added which allows you to
302 override the 1024 byte limit on access count histograms for blocks
303 of memory. The client request is DHAT_HISTOGRAM_MEMORY.
305 * ==================== FIXED BUGS ====================
307 The following bugs have been fixed or resolved. Note that "n-i-bz"
308 stands for "not in bugzilla" -- that is, a bug that was reported to us
309 but never got a bugzilla entry. We encourage you to file bugs in
310 bugzilla (https://bugs.kde.org/enter_bug.cgi?product=valgrind) rather
311 than mailing the developers (or mailing lists) directly -- bugs that
312 are not entered into bugzilla tend to get forgotten about or ignored.
314 170510 Don't warn about ioctl of size 0 without direction hint
315 241072 List tools in --help output
316 327548 false positive while destroying mutex
317 382034 Testcases build fixes for musl
318 351857 confusing error message about valid command line option
319 374596 inconsistent RDTSCP support on x86_64
320 392331 Spurious lock not held error from inside pthread_cond_timedwait
321 397083 Likely false positive "uninitialised value(s)" for __wmemchr_avx2 and __wmemcmp_avx2_movbe
322 400793 pthread_rwlock_timedwrlock false positive
323 419054 Unhandled syscall getcpu on arm32
324 433873 openat2 syscall unimplemented on Linux
325 434057 Add stdio mode to valgrind's gdbserver
326 435441 valgrind fails to interpose malloc on musl 1.2.2 due to weak symbol name and no libc soname
327 436413 Warn about realloc of size zero
328 439685 compiler warning in callgrind/main.c
329 444110 priv/guest_ppc_toIR.c:36198:31: warning: duplicated 'if' condition.
330 444487 hginfo test detects an extra lock inside data symbol "_rtld_local"
331 444488 Use glibc.pthread.stack_cache_size tunable
332 444568 drd/tests/pth_barrier_thr_cr fails on Fedora 38
333 445743 "The impossible happened: mutex is locked simultaneously by two threads"
334 while using mutexes with priority inheritance and signals
335 449309 Missing loopback device ioctl(s)
336 459476 vgdb: allow address reuse to avoid "address already in use" errorsuse" errors
337 460356 s390: Sqrt32Fx4 -- cannot reduce tree
338 462830 WARNING: unhandled amd64-freebsd syscall: 474
339 463027 broken check for MPX instruction support in assembler
340 464103 Enhancement: add a client request to DHAT to mark memory to be histogrammed
341 464476 Firefox fails to start under Valgrind
342 464609 Valgrind memcheck should support Linux pidfd_open
343 464680 Show issues caused by memory policies like selinux deny_execmem
344 464859 Build failures with GCC-13 (drd tsan_unittest)
345 464969 D language demangling
346 465435 m_libcfile.c:66 (vgPlain_safe_fd): Assertion 'newfd >= VG_(fd_hard_limit)' failed.
347 466104 aligned_alloc problems, part 1
348 467036 Add time cost statistics for Regtest
349 467482 Build failure on aarch64 Alpine
350 467714 fdleak_* and rlimit tests fail when parent process has more than
351 64 descriptors opened
352 467839 Gdbserver: Improve compatibility of library directory name
353 468401 [PATCH] Add a style file for clang-format
354 468556 Build failure for vgdb
355 468606 build: remove "Valgrind relies on GCC" check/output
356 469097 ppc64(be) doesn't support SCV syscall instruction
357 n-i-bz FreeBSD rfork syscall fail with EINVAL or ENOSYS rather than VG_(unimplemented)
359 To see details of a given bug, visit
360 https://bugs.kde.org/show_bug.cgi?id=XXXXXX
361 where XXXXXX is the bug number as listed above.
363 * ==================== KNOWN ISSUES ===================
365 * configure --enable-lto=yes is know to not work in all setups.
366 See bug 469049. Workaround: Build without LTO.
368 (3.21.0.RC1: 14 Apr 2023)
369 (3.21.0.RC2: 21 Apr 2023)
371 Release 3.20.0 (24 Oct 2022)
372 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
374 This release supports X86/Linux, AMD64/Linux, ARM32/Linux, ARM64/Linux,
375 PPC32/Linux, PPC64BE/Linux, PPC64LE/Linux, S390X/Linux, MIPS32/Linux,
376 MIPS64/Linux, ARM/Android, ARM64/Android, MIPS32/Android, X86/Android,
377 X86/Solaris, AMD64/Solaris, AMD64/MacOSX 10.12, X86/FreeBSD and
378 AMD64/FreeBSD. There is also preliminary support for X86/macOS 10.13,
379 AMD64/macOS 10.13 and nanoMIPS/Linux.
381 * ==================== CORE CHANGES ===================
383 * The option "--vgdb-stop-at=event1,event2,..." accepts the new value abexit.
384 This indicates to invoke gdbserver when your program exits abnormally
385 (i.e. with a non zero exit code).
386 * Fix Rust v0 name demangling.
387 * The Linux rseq syscall is now implemented as (silently) returning ENOSYS.
388 * Add FreeBSD syscall wrappers for __specialfd and __realpathat.
389 * Remove FreeBSD dependencies on COMPAT10, which fixes compatibility with
391 * The option --enable-debuginfod=<no|yes> [default: yes] has been added on
393 * More DWARF5 support as generated by clang14.
395 * ==================== FIXED BUGS ====================
397 The following bugs have been fixed or resolved. Note that "n-i-bz"
398 stands for "not in bugzilla" -- that is, a bug that was reported to us
399 but never got a bugzilla entry. We encourage you to file bugs in
400 bugzilla (https://bugs.kde.org/enter_bug.cgi?product=valgrind) rather
401 than mailing the developers (or mailing lists) directly -- bugs that
402 are not entered into bugzilla tend to get forgotten about or ignored.
404 131186 writev reports error in (vector[...])
405 434764 iconv_open causes ld.so v2.28+ to use optimised strncmp
406 446754 Improve error codes from alloc functions under memcheck
407 452274 memcheck crashes with Assertion 'sci->status.what == SsIdle' failed
408 452779 Valgrind fails to build on FreeBSD 13.0 with llvm-devel (15.0.0)
409 453055 shared_timed_mutex drd test fails with "Lock shared failed" message
410 453602 Missing command line option to enable/disable debuginfod
411 452802 Handle lld 9+ split RW PT_LOAD segments correctly
412 454040 s390x: False-positive memcheck:cond in memmem on arch13 systems
413 456171 [PATCH] FreeBSD: Don't record address errors when accessing the 'kern.ps_strings' sysctl struct
414 n-i-bz Implement vgdb invoker on FreeBSD
415 458845 PowerPC: The L field for the dcbf and sync instruction should be
417 458915 Remove register cache to fix 458915 gdbserver causes wrong syscall return
418 459031 Documentation on --error-exitcode incomplete
419 459477 XERROR messages lacks ending '\n' in vgdb
420 462007 Implicit int in none/tests/faultstatus.c
422 To see details of a given bug, visit
423 https://bugs.kde.org/show_bug.cgi?id=XXXXXX
424 where XXXXXX is the bug number as listed above.
426 (3.20.0.RC1: 20 Oct 2022)
429 Release 3.19.0 (11 Apr 2022)
430 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
432 This release supports X86/Linux, AMD64/Linux, ARM32/Linux, ARM64/Linux,
433 PPC32/Linux, PPC64BE/Linux, PPC64LE/Linux, S390X/Linux, MIPS32/Linux,
434 MIPS64/Linux, ARM/Android, ARM64/Android, MIPS32/Android, X86/Android,
435 X86/Solaris, AMD64/Solaris, AMD64/MacOSX 10.12, X86/FreeBSD and
436 AMD64/FreeBSD. There is also preliminary support for X86/macOS 10.13,
437 AMD64/macOS 10.13 and nanoMIPS/Linux.
439 * ==================== CORE CHANGES ===================
441 * Fix Rust v0 name demangling.
442 * The Linux rseq syscall is now implemented as (silently) returning ENOSYS.
443 * Add FreeBSD syscall wrappers for __specialfd and __realpathat.
444 * Remove FreeBSD dependencies on COMPAT10, which fixes compatibility with HardenedBSD
446 * ================== PLATFORM CHANGES =================
449 - ignore the "v8.x" architecture levels, only look at actual CPU features
450 present. Fixes mismatch detected between RDMA and atomics features
451 preventing startup on some QEMU configurations.
452 - Implement LD{,A}XP and ST{,L}XP
453 - Fix incorrect code emitted for doubleword CAS.
456 - Fix sys_ipc semtimedop syscall
457 - Fix VFLRX and WFLRX instructions
458 - Fix EXRL instruction with negative offset
461 - Reimplement the vbpermq instruction support to generate less Iops and
462 avoid overflowing internal buffers.
463 - Fix checking for scv support to avoid "Facility 'SCV' unavailable (12),
464 exception" messages in dmsg.
465 - Fix setting condition code for Vector Compare quad word instructions.
466 - Fix fix lxsibzx, lxsihzx and lxsihzx instructions so they only load
467 their respective sized data.
468 - Fix the prefixed stq instruction in PC relative mode.
470 * ==================== TOOL CHANGES ===================
473 - Speed up --track-origins=yes for large (in the range of hundreds to
474 thousands of megabytes) mmap/munmaps.
476 - Several fixes for new versions of libstd++ using new posix try_lock
479 * ==================== FIXED BUGS ====================
481 The following bugs have been fixed or resolved. Note that "n-i-bz"
482 stands for "not in bugzilla" -- that is, a bug that was reported to us
483 but never got a bugzilla entry. We encourage you to file bugs in
484 bugzilla (https://bugs.kde.org/enter_bug.cgi?product=valgrind) rather
485 than mailing the developers (or mailing lists) directly -- bugs that
486 are not entered into bugzilla tend to get forgotten about or ignored.
488 403802 leak_cpp_interior fails with some reachable blocks different than expected
489 435732 memcheck/tests/leak_cpp_interior fails with gcc11
490 444242 s390x: Valgrind crashes on EXRL with negative offset
491 444399 arm64: unhandled instruction 0xC87F2D89 (LD{,A}XP and ST{,L}XP).
493 444481 gdb_server test failures on s390x
494 444495 dhat/tests/copy fails on s390x
495 444552 memcheck/tests/sem fails on s390x with glibc 2.34
496 444571 PPC, fix the lxsibzx and lxsihzx so they only load their respective
498 444836 PPC, pstq instruction for R=1 is not storing to the correct address.
499 444925 fexecve syscall wrapper not properly implemented
500 445032 valgrind/memcheck crash with SIGSEGV when SIGVTALRM timer used and
502 445211 Fix out of tree builds
503 445300 [PATCH] Fix building tests with Musl
504 445011 SIGCHLD is sent when valgrind uses debuginfod-find
505 445354 arm64 backend: incorrect code emitted for doubleword CAS
506 445415 arm64 front end: alignment checks missing for atomic instructions
507 445504 Using C++ condition_variable results in bogus "mutex is locked simultaneously by two threads" warning
508 445607 Unhandled amd64-freebsd syscall: 247
509 445668 Inline stack frame generation is broken for Rust binaries
510 445916 Demangle Rust v0 symbols with .llvm suffix
511 446139 DRD/Helgrind with std::shared_timed_mutex::try_lock_until and try_lock_shared_until false positives
512 446138 DRD/Helgrind with std::timed_mutex::try_lock_until false positives
513 446281 Add a DRD suppression for fwrite
514 446103 Memcheck: `--track-origins=yes` causes extreme slowdowns for large mmap/munmap
515 446139 DRD/Helgrind with std::shared_timed_mutex::try_lock_until and try_lock_shared_until false
516 446251 TARGET_SIGNAL_THR added to enum target_signal
517 446823 FreeBSD - missing syscalls when using libzm4
518 447991 s390x: Valgrind indicates illegal instruction on wflrx
519 447995 Valgrind segfault on power10 due to hwcap checking code
520 449483 Powerpc: vcmpgtsq., vcmpgtuq,, vcmpequq. instructions not setting the
521 condition code correctly.
522 449672 ppc64 --track-origins=yes failures because of bad cmov addHRegUse
523 449838 sigsegv liburing the 'impossible' happened for io_uring_setup
524 450025 Powerc: ACC file not implemented as a logical overlay of the VSR
526 450437 Warn for execve syscall with argv or argv[0] being NULL
527 450536 Powerpc: valgrind throws 'facility scv unavailable exception'
528 451626 Syscall param bpf(attr->raw_tracepoint.name) points to unaddressable byte(s)
529 451827 [ppc64le] VEX temporary storage exhausted with several vbpermq instructions
530 451843 valgrind fails to start on a FreeBSD system which enforces W^X
532 To see details of a given bug, visit
533 https://bugs.kde.org/show_bug.cgi?id=XXXXXX
534 where XXXXXX is the bug number as listed above.
536 (3.19.0.RC1: 02 Apr 2022)
537 (3.19.0.RC2: 08 Apr 2022)