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.
26 To see details of a given bug, visit
27 https://bugs.kde.org/show_bug.cgi?id=XXXXXX
28 where XXXXXX is the bug number as listed above.
30 (3.23.0.RC1: ?? Apr 2024)
32 Release 3.22.0 (31 Oct 2023)
33 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
35 This release supports X86/Linux, AMD64/Linux, ARM32/Linux, ARM64/Linux,
36 PPC32/Linux, PPC64BE/Linux, PPC64LE/Linux, S390X/Linux, MIPS32/Linux,
37 MIPS64/Linux, ARM/Android, ARM64/Android, MIPS32/Android, X86/Android,
38 X86/Solaris, AMD64/Solaris, AMD64/MacOSX 10.12, X86/FreeBSD and
39 AMD64/FreeBSD. There is also preliminary support for X86/macOS 10.13,
40 AMD64/macOS 10.13 and nanoMIPS/Linux.
42 * ==================== CORE CHANGES ===================
44 * A new configure option --with-gdbscripts-dir lets you install
45 the gdb valgrind python monitor scripts in a specific location.
46 For example a distro could use it to install the scripts in a
47 safe load location --with-gdbscripts-dir=%{_datadir}/gdb/auto-load
48 It is also possible to configure --without-gdb-scripts-dir so no
49 .debug_gdb_scripts section is added to the vgpreload library and
50 no valgrind-monitor python scripts are installed at all.
52 * ================== PLATFORM CHANGES =================
54 * Support has been added for FreeBSD 14 and FreeBSD 15.
55 * Add support for the folllowing FreeBSD system calls:
56 close_range, kqueuex, membarrier, timerfd_create,
57 timerfd_settime and timerfd_gettime (all added in FreeBSD 15).
59 * ==================== TOOL CHANGES ===================
61 * Memcheck now tests and warns about the values used for
62 alignment and size. These apply to various functions: memalign,
63 posix_memalign and aligned_alloc in C and various overloads
64 of operators new and delete in C++. The kinds of error that can
66 - invalid alignment, for instance the alignment is usually required
68 - mismatched alignment between aligned allocation and aligned
70 - mismatched size when sized delete is used
71 - bad size for functions that have implementation defined behaviour
72 when the requested size is zero
75 - You can now profile part of a program's execution using the new
76 `CACHEGRIND_START_INSTRUMENTATION` and `CACHEGRIND_STOP_INSTRUMENTATION`
77 client requests, along with the new `--instr-at-start` option. The
78 behaviour is the same as Callgrind's equivalent functionality.
80 * ==================== FIXED BUGS ====================
82 The following bugs have been fixed or resolved. Note that "n-i-bz"
83 stands for "not in bugzilla" -- that is, a bug that was reported to us
84 but never got a bugzilla entry. We encourage you to file bugs in
85 bugzilla (https://bugs.kde.org/enter_bug.cgi?product=valgrind) rather
86 than mailing the developers (or mailing lists) directly -- bugs that
87 are not entered into bugzilla tend to get forgotten about or ignored.
89 390871 ELF debug info reader confused with multiple .rodata* sections
90 417993 vbit-test fail on s390x with Iop_Add32: spurious dependency on uninit
91 426751 Valgrind reports "still reachable" memory using musl
92 (alpine running inside docker)
93 432801 Valgrind 3.16.1 reports a jump based on uninitialized memory somehow
94 related to clang and signals
95 433857 Add validation to C++17 aligned new/delete alignment size
96 433859 Add mismatched detection to C++ 17 aligned new/delete
97 460192 Add epoll_pwait2
98 461074 DWARF2 CFI reader: unhandled DW_OP_ 0x11 (consts) DW_OP_ 0x92 (bregx)
99 465782 s390x: Valgrind doesn't compile with Clang on s390x
100 466105 aligned_alloc problems, part 2
101 467441 Add mismatched detection to C++ 14 sized delete
102 469049 link failure on ppc64 (big endian) valgrind 3.20
103 469146 massif --ignore-fn does not ignore inlined functions
104 469768 Make it possible to install gdb scripts in a different location
105 470121 Can't run callgrind_control with valgrind 3.21.0 because of perl errors
106 470132 s390x: Assertion failure on VGM instruction
107 470520 Multiple realloc zero errors crash in MC_(eq_Error)
108 470713 Failure on the Yosys project: valgrind: m_libcfile.c:1802
109 (Bool vgPlain_realpath(const HChar *, HChar *)):
110 Assertion 'resolved' failed
111 470830 Don't print actions vgdb me ... continue for vgdb --multi mode
112 470978 s390x: Valgrind cannot start qemu-kvm when "sysctl vm.allocate_pgste=0"
113 471311 gdb --multi mode stdout redirecting to stderr
114 471807 Add support for lazy reading and downloading of DWARF debuginfo
115 472219 Syscall param ppoll(ufds.events) points to uninitialised byte(s)
116 472875 none/tests/s390x/dfp-1 failure
117 472963 Broken regular expression in configure.ac
118 473604 Fix bug472219.c compile failure with Clang 16
119 473677 make check compile failure with Clang 16 based on GCC 13.x
120 473745 must-be-redirected function - strlen
121 473870 FreeBSD 14 applications fail early at startup
122 473944 Handle mold linker split RW PT_LOAD segments correctly
123 474332 aligned_alloc under Valgrind returns nullptr when alignment is not a multiple of sizeof(void *)
124 475650 DRD does not work with C11 threads
125 475652 Missing suppression for __wcsncpy_avx2 (strncpy-avx2.S:308)?
126 476108 vg_replace_malloc DELETE checks size
127 n-i-bz Allow arguments with spaces in .valgrindrc files
128 n-i-bz FreeBSD fixed reading of Valgrind tools own debuginfo
130 To see details of a given bug, visit
131 https://bugs.kde.org/show_bug.cgi?id=XXXXXX
132 where XXXXXX is the bug number as listed above.
134 (3.22.0.RC1: 17 Oct 2023)
135 (3.22.0.RC2: 26 Oct 2023)
137 Release 3.21.0 (28 Apr 2023)
138 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
140 This release supports X86/Linux, AMD64/Linux, ARM32/Linux, ARM64/Linux,
141 PPC32/Linux, PPC64BE/Linux, PPC64LE/Linux, S390X/Linux, MIPS32/Linux,
142 MIPS64/Linux, ARM/Android, ARM64/Android, MIPS32/Android, X86/Android,
143 X86/Solaris, AMD64/Solaris, AMD64/MacOSX 10.12, X86/FreeBSD and
144 AMD64/FreeBSD. There is also preliminary support for X86/macOS 10.13,
145 AMD64/macOS 10.13 and nanoMIPS/Linux.
147 * ==================== CORE CHANGES ===================
149 * When GDB is used to debug a program running under valgrind using
150 the valgrind gdbserver, GDB will automatically load some
151 python code provided in valgrind defining GDB front end commands
152 corresponding to the valgrind monitor commands.
153 These GDB front end commands accept the same format as
154 the monitor commands directly sent to the Valgrind gdbserver.
155 These GDB front end commands provide a better integration
156 in the GDB command line interface, so as to use for example
157 GDB auto-completion, command specific help, searching for
158 a command or command help matching a regexp, ...
159 For relevant monitor commands, GDB will evaluate arguments
160 to make the use of monitor commands easier.
161 For example, instead of having to print the address of a variable
162 to pass it to a subsequent monitor command, the GDB front end
163 command will evaluate the address argument. It is for example
165 (gdb) memcheck who_points_at &some_struct sizeof(some_struct)
168 $2 = (some_struct_type *) 0x1130a0 <some_struct>
169 (gdb) p sizeof(some_struct)
171 (gdb) monitor who_point_at 0x1130a0 40
173 * The vgdb utility now supports extended-remote protocol when
174 invoked with --multi. In this mode the GDB run command is
175 supported. Which means you don't need to run gdb and valgrind
176 from different terminals. So for example to start your program
177 in gdb and run it under valgrind you can do:
179 (gdb) set remote exec-file prog
181 (gdb) target extended-remote | vgdb --multi
184 * The behaviour of realloc with a size of zero can now
185 be changed for tools that intercept malloc. Those
186 tools are memcheck, helgrind, drd, massif and dhat.
187 Realloc implementations generally do one of two things
188 - free the memory like free() and return NULL
189 (GNU libc and ptmalloc).
190 - either free the memory and then allocate a
191 minimum sized block or just return the
192 original pointer. Return NULL if the
193 allocation of the minimum sized block fails
194 (jemalloc, musl, snmalloc, Solaris, macOS).
195 When Valgrind is configured and built it will
196 try to match the OS and libc behaviour. However
197 if you are using a non-default library to replace
198 malloc and family (e.g., musl on a glibc Linux or
199 tcmalloc on FreeBSD) then you can use a command line
200 option to change the behaviour of Valgrind:
201 --realloc-zero-bytes-frees=yes|no [yes on Linux glibc, no otherwise]
203 * ================== PLATFORM CHANGES =================
205 * Make the address space limit on FreeBSD amd64 128Gbytes
206 (the same as Linux and Solaris, it was 32Gbytes)
208 * ==================== TOOL CHANGES ===================
211 - When doing a delta leak_search, it is now possible to only
212 output the new loss records compared to the previous leak search.
213 This is available in the memcheck monitor command 'leak_search'
214 by specifying the "new" keyword or in your program by using
215 the client request VALGRIND_DO_NEW_LEAK_CHECK.
216 Whenever a "delta" leak search is done (i.e. when specifying
217 "new" or "increased" or "changed" in the monitor command),
218 the new loss records have a "new" marker.
219 - Valgrind now contains python code that defines GDB memcheck
220 front end monitor commands. See CORE CHANGES.
221 - Performs checks for the use of realloc with a size of zero.
222 This is non-portable and a source of errors. If memcheck
223 detects such a usage it will generate an error
224 realloc() with size 0
225 followed by the usual callstacks.
226 A switch has been added to allow this to be turned off:
227 --show-realloc-size-zero=yes|no [yes]
230 - The option ---history-backtrace-size=<number> allows to configure
231 the number of entries to record in the stack traces of "old"
232 accesses. Previously, this number was hardcoded to 8.
233 - Valgrind now contains python code that defines GDB helgrind
234 front end monitor commands. See CORE CHANGES.
237 - `--cache-sim=no` is now the default. The cache simulation is old and
238 unlikely to match any real modern machine. This means only the `Ir`
239 event are gathered by default, but that is by far the most useful
241 - `cg_annotate`, `cg_diff`, and `cg_merge` have been rewritten in
242 Python. As a result, they all have more flexible command line
243 argument handling, e.g. supporting `--show-percs` and
244 `--no-show-percs` forms as well as the existing `--show-percs=yes`
245 and `--show-percs=no`.
246 - `cg_annotate` has some functional changes.
247 - It's much faster, e.g. 3-4x on common cases.
248 - It now supports diffing (with `--diff`, `--mod-filename`, and
249 `--mod-funcname`) and merging (by passing multiple data files).
250 - It now provides more information at the file and function level.
251 There are now "File:function" and "Function:file" sections. These
252 are very useful for programs that use inlining a lot.
253 - Support for user-annotated files and the `-I`/`--include` option
254 has been removed, because it was of little use and blocked other
256 - The `--auto` option is renamed `--annotate`, though the old
257 `--auto=yes`/`--auto=no` forms are still supported.
258 - `cg_diff` and `cg_merge` are now deprecated, because `cg_annotate`
259 now does a better job of diffing and merging.
260 - The Cachegrind output file format has changed very slightly, but in
261 ways nobody is likely to notice.
264 - Valgrind now contains python code that defines GDB callgrind
265 front end monitor commands. See CORE CHANGES.
268 - Valgrind now contains python code that defines GDB massif
269 front end monitor commands. See CORE CHANGES.
272 - A new kind of user request has been added which allows you to
273 override the 1024 byte limit on access count histograms for blocks
274 of memory. The client request is DHAT_HISTOGRAM_MEMORY.
276 * ==================== FIXED BUGS ====================
278 The following bugs have been fixed or resolved. Note that "n-i-bz"
279 stands for "not in bugzilla" -- that is, a bug that was reported to us
280 but never got a bugzilla entry. We encourage you to file bugs in
281 bugzilla (https://bugs.kde.org/enter_bug.cgi?product=valgrind) rather
282 than mailing the developers (or mailing lists) directly -- bugs that
283 are not entered into bugzilla tend to get forgotten about or ignored.
285 170510 Don't warn about ioctl of size 0 without direction hint
286 241072 List tools in --help output
287 327548 false positive while destroying mutex
288 382034 Testcases build fixes for musl
289 351857 confusing error message about valid command line option
290 374596 inconsistent RDTSCP support on x86_64
291 392331 Spurious lock not held error from inside pthread_cond_timedwait
292 397083 Likely false positive "uninitialised value(s)" for __wmemchr_avx2 and __wmemcmp_avx2_movbe
293 400793 pthread_rwlock_timedwrlock false positive
294 419054 Unhandled syscall getcpu on arm32
295 433873 openat2 syscall unimplemented on Linux
296 434057 Add stdio mode to valgrind's gdbserver
297 435441 valgrind fails to interpose malloc on musl 1.2.2 due to weak symbol name and no libc soname
298 436413 Warn about realloc of size zero
299 439685 compiler warning in callgrind/main.c
300 444110 priv/guest_ppc_toIR.c:36198:31: warning: duplicated 'if' condition.
301 444487 hginfo test detects an extra lock inside data symbol "_rtld_local"
302 444488 Use glibc.pthread.stack_cache_size tunable
303 444568 drd/tests/pth_barrier_thr_cr fails on Fedora 38
304 445743 "The impossible happened: mutex is locked simultaneously by two threads"
305 while using mutexes with priority inheritance and signals
306 449309 Missing loopback device ioctl(s)
307 459476 vgdb: allow address reuse to avoid "address already in use" errorsuse" errors
308 460356 s390: Sqrt32Fx4 -- cannot reduce tree
309 462830 WARNING: unhandled amd64-freebsd syscall: 474
310 463027 broken check for MPX instruction support in assembler
311 464103 Enhancement: add a client request to DHAT to mark memory to be histogrammed
312 464476 Firefox fails to start under Valgrind
313 464609 Valgrind memcheck should support Linux pidfd_open
314 464680 Show issues caused by memory policies like selinux deny_execmem
315 464859 Build failures with GCC-13 (drd tsan_unittest)
316 464969 D language demangling
317 465435 m_libcfile.c:66 (vgPlain_safe_fd): Assertion 'newfd >= VG_(fd_hard_limit)' failed.
318 466104 aligned_alloc problems, part 1
319 467036 Add time cost statistics for Regtest
320 467482 Build failure on aarch64 Alpine
321 467714 fdleak_* and rlimit tests fail when parent process has more than
322 64 descriptors opened
323 467839 Gdbserver: Improve compatibility of library directory name
324 468401 [PATCH] Add a style file for clang-format
325 468556 Build failure for vgdb
326 468606 build: remove "Valgrind relies on GCC" check/output
327 469097 ppc64(be) doesn't support SCV syscall instruction
328 n-i-bz FreeBSD rfork syscall fail with EINVAL or ENOSYS rather than VG_(unimplemented)
330 To see details of a given bug, visit
331 https://bugs.kde.org/show_bug.cgi?id=XXXXXX
332 where XXXXXX is the bug number as listed above.
334 * ==================== KNOWN ISSUES ===================
336 * configure --enable-lto=yes is know to not work in all setups.
337 See bug 469049. Workaround: Build without LTO.
339 (3.21.0.RC1: 14 Apr 2023)
340 (3.21.0.RC2: 21 Apr 2023)
342 Release 3.20.0 (24 Oct 2022)
343 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
345 This release supports X86/Linux, AMD64/Linux, ARM32/Linux, ARM64/Linux,
346 PPC32/Linux, PPC64BE/Linux, PPC64LE/Linux, S390X/Linux, MIPS32/Linux,
347 MIPS64/Linux, ARM/Android, ARM64/Android, MIPS32/Android, X86/Android,
348 X86/Solaris, AMD64/Solaris, AMD64/MacOSX 10.12, X86/FreeBSD and
349 AMD64/FreeBSD. There is also preliminary support for X86/macOS 10.13,
350 AMD64/macOS 10.13 and nanoMIPS/Linux.
352 * ==================== CORE CHANGES ===================
354 * The option "--vgdb-stop-at=event1,event2,..." accepts the new value abexit.
355 This indicates to invoke gdbserver when your program exits abnormally
356 (i.e. with a non zero exit code).
357 * Fix Rust v0 name demangling.
358 * The Linux rseq syscall is now implemented as (silently) returning ENOSYS.
359 * Add FreeBSD syscall wrappers for __specialfd and __realpathat.
360 * Remove FreeBSD dependencies on COMPAT10, which fixes compatibility with
362 * The option --enable-debuginfod=<no|yes> [default: yes] has been added on
364 * More DWARF5 support as generated by clang14.
366 * ==================== FIXED BUGS ====================
368 The following bugs have been fixed or resolved. Note that "n-i-bz"
369 stands for "not in bugzilla" -- that is, a bug that was reported to us
370 but never got a bugzilla entry. We encourage you to file bugs in
371 bugzilla (https://bugs.kde.org/enter_bug.cgi?product=valgrind) rather
372 than mailing the developers (or mailing lists) directly -- bugs that
373 are not entered into bugzilla tend to get forgotten about or ignored.
375 131186 writev reports error in (vector[...])
376 434764 iconv_open causes ld.so v2.28+ to use optimised strncmp
377 446754 Improve error codes from alloc functions under memcheck
378 452274 memcheck crashes with Assertion 'sci->status.what == SsIdle' failed
379 452779 Valgrind fails to build on FreeBSD 13.0 with llvm-devel (15.0.0)
380 453055 shared_timed_mutex drd test fails with "Lock shared failed" message
381 453602 Missing command line option to enable/disable debuginfod
382 452802 Handle lld 9+ split RW PT_LOAD segments correctly
383 454040 s390x: False-positive memcheck:cond in memmem on arch13 systems
384 456171 [PATCH] FreeBSD: Don't record address errors when accessing the 'kern.ps_strings' sysctl struct
385 n-i-bz Implement vgdb invoker on FreeBSD
386 458845 PowerPC: The L field for the dcbf and sync instruction should be
388 458915 Remove register cache to fix 458915 gdbserver causes wrong syscall return
389 459031 Documentation on --error-exitcode incomplete
390 459477 XERROR messages lacks ending '\n' in vgdb
391 462007 Implicit int in none/tests/faultstatus.c
393 To see details of a given bug, visit
394 https://bugs.kde.org/show_bug.cgi?id=XXXXXX
395 where XXXXXX is the bug number as listed above.
397 (3.20.0.RC1: 20 Oct 2022)
400 Release 3.19.0 (11 Apr 2022)
401 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
403 This release supports X86/Linux, AMD64/Linux, ARM32/Linux, ARM64/Linux,
404 PPC32/Linux, PPC64BE/Linux, PPC64LE/Linux, S390X/Linux, MIPS32/Linux,
405 MIPS64/Linux, ARM/Android, ARM64/Android, MIPS32/Android, X86/Android,
406 X86/Solaris, AMD64/Solaris, AMD64/MacOSX 10.12, X86/FreeBSD and
407 AMD64/FreeBSD. There is also preliminary support for X86/macOS 10.13,
408 AMD64/macOS 10.13 and nanoMIPS/Linux.
410 * ==================== CORE CHANGES ===================
412 * Fix Rust v0 name demangling.
413 * The Linux rseq syscall is now implemented as (silently) returning ENOSYS.
414 * Add FreeBSD syscall wrappers for __specialfd and __realpathat.
415 * Remove FreeBSD dependencies on COMPAT10, which fixes compatibility with HardenedBSD
417 * ================== PLATFORM CHANGES =================
420 - ignore the "v8.x" architecture levels, only look at actual CPU features
421 present. Fixes mismatch detected between RDMA and atomics features
422 preventing startup on some QEMU configurations.
423 - Implement LD{,A}XP and ST{,L}XP
424 - Fix incorrect code emitted for doubleword CAS.
427 - Fix sys_ipc semtimedop syscall
428 - Fix VFLRX and WFLRX instructions
429 - Fix EXRL instruction with negative offset
432 - Reimplement the vbpermq instruction support to generate less Iops and
433 avoid overflowing internal buffers.
434 - Fix checking for scv support to avoid "Facility 'SCV' unavailable (12),
435 exception" messages in dmsg.
436 - Fix setting condition code for Vector Compare quad word instructions.
437 - Fix fix lxsibzx, lxsihzx and lxsihzx instructions so they only load
438 their respective sized data.
439 - Fix the prefixed stq instruction in PC relative mode.
441 * ==================== TOOL CHANGES ===================
444 - Speed up --track-origins=yes for large (in the range of hundreds to
445 thousands of megabytes) mmap/munmaps.
447 - Several fixes for new versions of libstd++ using new posix try_lock
450 * ==================== FIXED BUGS ====================
452 The following bugs have been fixed or resolved. Note that "n-i-bz"
453 stands for "not in bugzilla" -- that is, a bug that was reported to us
454 but never got a bugzilla entry. We encourage you to file bugs in
455 bugzilla (https://bugs.kde.org/enter_bug.cgi?product=valgrind) rather
456 than mailing the developers (or mailing lists) directly -- bugs that
457 are not entered into bugzilla tend to get forgotten about or ignored.
459 403802 leak_cpp_interior fails with some reachable blocks different than expected
460 435732 memcheck/tests/leak_cpp_interior fails with gcc11
461 444242 s390x: Valgrind crashes on EXRL with negative offset
462 444399 arm64: unhandled instruction 0xC87F2D89 (LD{,A}XP and ST{,L}XP).
464 444481 gdb_server test failures on s390x
465 444495 dhat/tests/copy fails on s390x
466 444552 memcheck/tests/sem fails on s390x with glibc 2.34
467 444571 PPC, fix the lxsibzx and lxsihzx so they only load their respective
469 444836 PPC, pstq instruction for R=1 is not storing to the correct address.
470 444925 fexecve syscall wrapper not properly implemented
471 445032 valgrind/memcheck crash with SIGSEGV when SIGVTALRM timer used and
473 445211 Fix out of tree builds
474 445300 [PATCH] Fix building tests with Musl
475 445011 SIGCHLD is sent when valgrind uses debuginfod-find
476 445354 arm64 backend: incorrect code emitted for doubleword CAS
477 445415 arm64 front end: alignment checks missing for atomic instructions
478 445504 Using C++ condition_variable results in bogus "mutex is locked simultaneously by two threads" warning
479 445607 Unhandled amd64-freebsd syscall: 247
480 445668 Inline stack frame generation is broken for Rust binaries
481 445916 Demangle Rust v0 symbols with .llvm suffix
482 446139 DRD/Helgrind with std::shared_timed_mutex::try_lock_until and try_lock_shared_until false positives
483 446138 DRD/Helgrind with std::timed_mutex::try_lock_until false positives
484 446281 Add a DRD suppression for fwrite
485 446103 Memcheck: `--track-origins=yes` causes extreme slowdowns for large mmap/munmap
486 446139 DRD/Helgrind with std::shared_timed_mutex::try_lock_until and try_lock_shared_until false
487 446251 TARGET_SIGNAL_THR added to enum target_signal
488 446823 FreeBSD - missing syscalls when using libzm4
489 447991 s390x: Valgrind indicates illegal instruction on wflrx
490 447995 Valgrind segfault on power10 due to hwcap checking code
491 449483 Powerpc: vcmpgtsq., vcmpgtuq,, vcmpequq. instructions not setting the
492 condition code correctly.
493 449672 ppc64 --track-origins=yes failures because of bad cmov addHRegUse
494 449838 sigsegv liburing the 'impossible' happened for io_uring_setup
495 450025 Powerc: ACC file not implemented as a logical overlay of the VSR
497 450437 Warn for execve syscall with argv or argv[0] being NULL
498 450536 Powerpc: valgrind throws 'facility scv unavailable exception'
499 451626 Syscall param bpf(attr->raw_tracepoint.name) points to unaddressable byte(s)
500 451827 [ppc64le] VEX temporary storage exhausted with several vbpermq instructions
501 451843 valgrind fails to start on a FreeBSD system which enforces W^X
503 To see details of a given bug, visit
504 https://bugs.kde.org/show_bug.cgi?id=XXXXXX
505 where XXXXXX is the bug number as listed above.
507 (3.19.0.RC1: 02 Apr 2022)
508 (3.19.0.RC2: 08 Apr 2022)