Release 3.23.0 (?? Apr 2024)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
This release supports X86/Linux, AMD64/Linux, ARM32/Linux, ARM64/Linux,
PPC32/Linux, PPC64BE/Linux, PPC64LE/Linux, S390X/Linux, MIPS32/Linux,
MIPS64/Linux, ARM/Android, ARM64/Android, MIPS32/Android, X86/Android,
X86/Solaris, AMD64/Solaris, AMD64/MacOSX 10.12, X86/FreeBSD and
AMD64/FreeBSD. There is also preliminary support for X86/macOS 10.13,
AMD64/macOS 10.13 and nanoMIPS/Linux.
* ==================== CORE CHANGES ===================
* ================== PLATFORM CHANGES =================
* ==================== TOOL CHANGES ===================
* ==================== FIXED BUGS ====================
The following bugs have been fixed or resolved. Note that "n-i-bz"
stands for "not in bugzilla" -- that is, a bug that was reported to us
but never got a bugzilla entry. We encourage you to file bugs in
bugzilla (https://bugs.kde.org/enter_bug.cgi?product=valgrind) rather
than mailing the developers (or mailing lists) directly -- bugs that
are not entered into bugzilla tend to get forgotten about or ignored.
475498 Add reallocarray wrapper
476320 Build failure with GCC
476535 Difference in allocation size for massif/tests/overloaded-new between clang++/libc++ and g++/libstdc++
To see details of a given bug, visit
https://bugs.kde.org/show_bug.cgi?id=XXXXXX
where XXXXXX is the bug number as listed above.
(3.23.0.RC1: ?? Apr 2024)
Release 3.22.0 (31 Oct 2023)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
This release supports X86/Linux, AMD64/Linux, ARM32/Linux, ARM64/Linux,
PPC32/Linux, PPC64BE/Linux, PPC64LE/Linux, S390X/Linux, MIPS32/Linux,
MIPS64/Linux, ARM/Android, ARM64/Android, MIPS32/Android, X86/Android,
X86/Solaris, AMD64/Solaris, AMD64/MacOSX 10.12, X86/FreeBSD and
AMD64/FreeBSD. There is also preliminary support for X86/macOS 10.13,
AMD64/macOS 10.13 and nanoMIPS/Linux.
* ==================== CORE CHANGES ===================
* A new configure option --with-gdbscripts-dir lets you install
the gdb valgrind python monitor scripts in a specific location.
For example a distro could use it to install the scripts in a
safe load location --with-gdbscripts-dir=%{_datadir}/gdb/auto-load
It is also possible to configure --without-gdb-scripts-dir so no
.debug_gdb_scripts section is added to the vgpreload library and
no valgrind-monitor python scripts are installed at all.
* ================== PLATFORM CHANGES =================
* Support has been added for FreeBSD 14 and FreeBSD 15.
* Add support for the folllowing FreeBSD system calls:
close_range, kqueuex, membarrier, timerfd_create,
timerfd_settime and timerfd_gettime (all added in FreeBSD 15).
* ==================== TOOL CHANGES ===================
* Memcheck now tests and warns about the values used for
alignment and size. These apply to various functions: memalign,
posix_memalign and aligned_alloc in C and various overloads
of operators new and delete in C++. The kinds of error that can
be detected are
- invalid alignment, for instance the alignment is usually required
to be a power of 2
- mismatched alignment between aligned allocation and aligned
deallocation
- mismatched size when sized delete is used
- bad size for functions that have implementation defined behaviour
when the requested size is zero
* Cachegrind:
- You can now profile part of a program's execution using the new
`CACHEGRIND_START_INSTRUMENTATION` and `CACHEGRIND_STOP_INSTRUMENTATION`
client requests, along with the new `--instr-at-start` option. The
behaviour is the same as Callgrind's equivalent functionality.
* ==================== FIXED BUGS ====================
The following bugs have been fixed or resolved. Note that "n-i-bz"
stands for "not in bugzilla" -- that is, a bug that was reported to us
but never got a bugzilla entry. We encourage you to file bugs in
bugzilla (https://bugs.kde.org/enter_bug.cgi?product=valgrind) rather
than mailing the developers (or mailing lists) directly -- bugs that
are not entered into bugzilla tend to get forgotten about or ignored.
390871 ELF debug info reader confused with multiple .rodata* sections
417993 vbit-test fail on s390x with Iop_Add32: spurious dependency on uninit
426751 Valgrind reports "still reachable" memory using musl
(alpine running inside docker)
432801 Valgrind 3.16.1 reports a jump based on uninitialized memory somehow
related to clang and signals
433857 Add validation to C++17 aligned new/delete alignment size
433859 Add mismatched detection to C++ 17 aligned new/delete
460192 Add epoll_pwait2
461074 DWARF2 CFI reader: unhandled DW_OP_ 0x11 (consts) DW_OP_ 0x92 (bregx)
465782 s390x: Valgrind doesn't compile with Clang on s390x
466105 aligned_alloc problems, part 2
467441 Add mismatched detection to C++ 14 sized delete
469049 link failure on ppc64 (big endian) valgrind 3.20
469146 massif --ignore-fn does not ignore inlined functions
469768 Make it possible to install gdb scripts in a different location
470121 Can't run callgrind_control with valgrind 3.21.0 because of perl errors
470132 s390x: Assertion failure on VGM instruction
470520 Multiple realloc zero errors crash in MC_(eq_Error)
470713 Failure on the Yosys project: valgrind: m_libcfile.c:1802
(Bool vgPlain_realpath(const HChar *, HChar *)):
Assertion 'resolved' failed
470830 Don't print actions vgdb me ... continue for vgdb --multi mode
470978 s390x: Valgrind cannot start qemu-kvm when "sysctl vm.allocate_pgste=0"
471311 gdb --multi mode stdout redirecting to stderr
471807 Add support for lazy reading and downloading of DWARF debuginfo
472219 Syscall param ppoll(ufds.events) points to uninitialised byte(s)
472875 none/tests/s390x/dfp-1 failure
472963 Broken regular expression in configure.ac
473604 Fix bug472219.c compile failure with Clang 16
473677 make check compile failure with Clang 16 based on GCC 13.x
473745 must-be-redirected function - strlen
473870 FreeBSD 14 applications fail early at startup
473944 Handle mold linker split RW PT_LOAD segments correctly
474332 aligned_alloc under Valgrind returns nullptr when alignment is not a multiple of sizeof(void *)
475650 DRD does not work with C11 threads
475652 Missing suppression for __wcsncpy_avx2 (strncpy-avx2.S:308)?
476108 vg_replace_malloc DELETE checks size
n-i-bz Allow arguments with spaces in .valgrindrc files
n-i-bz FreeBSD fixed reading of Valgrind tools own debuginfo
To see details of a given bug, visit
https://bugs.kde.org/show_bug.cgi?id=XXXXXX
where XXXXXX is the bug number as listed above.
(3.22.0.RC1: 17 Oct 2023)
(3.22.0.RC2: 26 Oct 2023)
Release 3.21.0 (28 Apr 2023)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
This release supports X86/Linux, AMD64/Linux, ARM32/Linux, ARM64/Linux,
PPC32/Linux, PPC64BE/Linux, PPC64LE/Linux, S390X/Linux, MIPS32/Linux,
MIPS64/Linux, ARM/Android, ARM64/Android, MIPS32/Android, X86/Android,
X86/Solaris, AMD64/Solaris, AMD64/MacOSX 10.12, X86/FreeBSD and
AMD64/FreeBSD. There is also preliminary support for X86/macOS 10.13,
AMD64/macOS 10.13 and nanoMIPS/Linux.
* ==================== CORE CHANGES ===================
* When GDB is used to debug a program running under valgrind using
the valgrind gdbserver, GDB will automatically load some
python code provided in valgrind defining GDB front end commands
corresponding to the valgrind monitor commands.
These GDB front end commands accept the same format as
the monitor commands directly sent to the Valgrind gdbserver.
These GDB front end commands provide a better integration
in the GDB command line interface, so as to use for example
GDB auto-completion, command specific help, searching for
a command or command help matching a regexp, ...
For relevant monitor commands, GDB will evaluate arguments
to make the use of monitor commands easier.
For example, instead of having to print the address of a variable
to pass it to a subsequent monitor command, the GDB front end
command will evaluate the address argument. It is for example
possible to do:
(gdb) memcheck who_points_at &some_struct sizeof(some_struct)
instead of:
(gdb) p &some_struct
$2 = (some_struct_type *) 0x1130a0 <some_struct>
(gdb) p sizeof(some_struct)
$3 = 40
(gdb) monitor who_point_at 0x1130a0 40
* The vgdb utility now supports extended-remote protocol when
invoked with --multi. In this mode the GDB run command is
supported. Which means you don't need to run gdb and valgrind
from different terminals. So for example to start your program
in gdb and run it under valgrind you can do:
$ gdb prog
(gdb) set remote exec-file prog
(gdb) set sysroot /
(gdb) target extended-remote | vgdb --multi
(gdb) start
* The behaviour of realloc with a size of zero can now
be changed for tools that intercept malloc. Those
tools are memcheck, helgrind, drd, massif and dhat.
Realloc implementations generally do one of two things
- free the memory like free() and return NULL
(GNU libc and ptmalloc).
- either free the memory and then allocate a
minimum sized block or just return the
original pointer. Return NULL if the
allocation of the minimum sized block fails
(jemalloc, musl, snmalloc, Solaris, macOS).
When Valgrind is configured and built it will
try to match the OS and libc behaviour. However
if you are using a non-default library to replace
malloc and family (e.g., musl on a glibc Linux or
tcmalloc on FreeBSD) then you can use a command line
option to change the behaviour of Valgrind:
--realloc-zero-bytes-frees=yes|no [yes on Linux glibc, no otherwise]
* ================== PLATFORM CHANGES =================
* Make the address space limit on FreeBSD amd64 128Gbytes
(the same as Linux and Solaris, it was 32Gbytes)
* ==================== TOOL CHANGES ===================
* Memcheck:
- When doing a delta leak_search, it is now possible to only
output the new loss records compared to the previous leak search.
This is available in the memcheck monitor command 'leak_search'
by specifying the "new" keyword or in your program by using
the client request VALGRIND_DO_NEW_LEAK_CHECK.
Whenever a "delta" leak search is done (i.e. when specifying
"new" or "increased" or "changed" in the monitor command),
the new loss records have a "new" marker.
- Valgrind now contains python code that defines GDB memcheck
front end monitor commands. See CORE CHANGES.
- Performs checks for the use of realloc with a size of zero.
This is non-portable and a source of errors. If memcheck
detects such a usage it will generate an error
realloc() with size 0
followed by the usual callstacks.
A switch has been added to allow this to be turned off:
--show-realloc-size-zero=yes|no [yes]
* Helgrind:
- The option ---history-backtrace-size=<number> allows to configure
the number of entries to record in the stack traces of "old"
accesses. Previously, this number was hardcoded to 8.
- Valgrind now contains python code that defines GDB helgrind
front end monitor commands. See CORE CHANGES.
* Cachegrind:
- `--cache-sim=no` is now the default. The cache simulation is old and
unlikely to match any real modern machine. This means only the `Ir`
event are gathered by default, but that is by far the most useful
event.
- `cg_annotate`, `cg_diff`, and `cg_merge` have been rewritten in
Python. As a result, they all have more flexible command line
argument handling, e.g. supporting `--show-percs` and
`--no-show-percs` forms as well as the existing `--show-percs=yes`
and `--show-percs=no`.
- `cg_annotate` has some functional changes.
- It's much faster, e.g. 3-4x on common cases.
- It now supports diffing (with `--diff`, `--mod-filename`, and
`--mod-funcname`) and merging (by passing multiple data files).
- It now provides more information at the file and function level.
There are now "File:function" and "Function:file" sections. These
are very useful for programs that use inlining a lot.
- Support for user-annotated files and the `-I`/`--include` option
has been removed, because it was of little use and blocked other
improvements.
- The `--auto` option is renamed `--annotate`, though the old
`--auto=yes`/`--auto=no` forms are still supported.
- `cg_diff` and `cg_merge` are now deprecated, because `cg_annotate`
now does a better job of diffing and merging.
- The Cachegrind output file format has changed very slightly, but in
ways nobody is likely to notice.
* Callgrind:
- Valgrind now contains python code that defines GDB callgrind
front end monitor commands. See CORE CHANGES.
* Massif:
- Valgrind now contains python code that defines GDB massif
front end monitor commands. See CORE CHANGES.
* DHAT:
- A new kind of user request has been added which allows you to
override the 1024 byte limit on access count histograms for blocks
of memory. The client request is DHAT_HISTOGRAM_MEMORY.
* ==================== FIXED BUGS ====================
The following bugs have been fixed or resolved. Note that "n-i-bz"
stands for "not in bugzilla" -- that is, a bug that was reported to us
but never got a bugzilla entry. We encourage you to file bugs in
bugzilla (https://bugs.kde.org/enter_bug.cgi?product=valgrind) rather
than mailing the developers (or mailing lists) directly -- bugs that
are not entered into bugzilla tend to get forgotten about or ignored.
170510 Don't warn about ioctl of size 0 without direction hint
241072 List tools in --help output
327548 false positive while destroying mutex
382034 Testcases build fixes for musl
351857 confusing error message about valid command line option
374596 inconsistent RDTSCP support on x86_64
392331 Spurious lock not held error from inside pthread_cond_timedwait
397083 Likely false positive "uninitialised value(s)" for __wmemchr_avx2 and __wmemcmp_avx2_movbe
400793 pthread_rwlock_timedwrlock false positive
419054 Unhandled syscall getcpu on arm32
433873 openat2 syscall unimplemented on Linux
434057 Add stdio mode to valgrind's gdbserver
435441 valgrind fails to interpose malloc on musl 1.2.2 due to weak symbol name and no libc soname
436413 Warn about realloc of size zero
439685 compiler warning in callgrind/main.c
444110 priv/guest_ppc_toIR.c:36198:31: warning: duplicated 'if' condition.
444487 hginfo test detects an extra lock inside data symbol "_rtld_local"
444488 Use glibc.pthread.stack_cache_size tunable
444568 drd/tests/pth_barrier_thr_cr fails on Fedora 38
445743 "The impossible happened: mutex is locked simultaneously by two threads"
while using mutexes with priority inheritance and signals
449309 Missing loopback device ioctl(s)
459476 vgdb: allow address reuse to avoid "address already in use" errorsuse" errors
460356 s390: Sqrt32Fx4 -- cannot reduce tree
462830 WARNING: unhandled amd64-freebsd syscall: 474
463027 broken check for MPX instruction support in assembler
464103 Enhancement: add a client request to DHAT to mark memory to be histogrammed
464476 Firefox fails to start under Valgrind
464609 Valgrind memcheck should support Linux pidfd_open
464680 Show issues caused by memory policies like selinux deny_execmem
464859 Build failures with GCC-13 (drd tsan_unittest)
464969 D language demangling
465435 m_libcfile.c:66 (vgPlain_safe_fd): Assertion 'newfd >= VG_(fd_hard_limit)' failed.
466104 aligned_alloc problems, part 1
467036 Add time cost statistics for Regtest
467482 Build failure on aarch64 Alpine
467714 fdleak_* and rlimit tests fail when parent process has more than
64 descriptors opened
467839 Gdbserver: Improve compatibility of library directory name
468401 [PATCH] Add a style file for clang-format
468556 Build failure for vgdb
468606 build: remove "Valgrind relies on GCC" check/output
469097 ppc64(be) doesn't support SCV syscall instruction
n-i-bz FreeBSD rfork syscall fail with EINVAL or ENOSYS rather than VG_(unimplemented)
To see details of a given bug, visit
https://bugs.kde.org/show_bug.cgi?id=XXXXXX
where XXXXXX is the bug number as listed above.
* ==================== KNOWN ISSUES ===================
* configure --enable-lto=yes is know to not work in all setups.
See bug 469049. Workaround: Build without LTO.
(3.21.0.RC1: 14 Apr 2023)
(3.21.0.RC2: 21 Apr 2023)
Release 3.20.0 (24 Oct 2022)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
This release supports X86/Linux, AMD64/Linux, ARM32/Linux, ARM64/Linux,
PPC32/Linux, PPC64BE/Linux, PPC64LE/Linux, S390X/Linux, MIPS32/Linux,
MIPS64/Linux, ARM/Android, ARM64/Android, MIPS32/Android, X86/Android,
X86/Solaris, AMD64/Solaris, AMD64/MacOSX 10.12, X86/FreeBSD and
AMD64/FreeBSD. There is also preliminary support for X86/macOS 10.13,
AMD64/macOS 10.13 and nanoMIPS/Linux.
* ==================== CORE CHANGES ===================
* The option "--vgdb-stop-at=event1,event2,..." accepts the new value abexit.
This indicates to invoke gdbserver when your program exits abnormally
(i.e. with a non zero exit code).
* Fix Rust v0 name demangling.
* The Linux rseq syscall is now implemented as (silently) returning ENOSYS.
* Add FreeBSD syscall wrappers for __specialfd and __realpathat.
* Remove FreeBSD dependencies on COMPAT10, which fixes compatibility with
HardenedBSD
* The option --enable-debuginfod=<no|yes> [default: yes] has been added on
Linux.
* More DWARF5 support as generated by clang14.
* ==================== FIXED BUGS ====================
The following bugs have been fixed or resolved. Note that "n-i-bz"
stands for "not in bugzilla" -- that is, a bug that was reported to us
but never got a bugzilla entry. We encourage you to file bugs in
bugzilla (https://bugs.kde.org/enter_bug.cgi?product=valgrind) rather
than mailing the developers (or mailing lists) directly -- bugs that
are not entered into bugzilla tend to get forgotten about or ignored.
131186 writev reports error in (vector[...])
434764 iconv_open causes ld.so v2.28+ to use optimised strncmp
446754 Improve error codes from alloc functions under memcheck
452274 memcheck crashes with Assertion 'sci->status.what == SsIdle' failed
452779 Valgrind fails to build on FreeBSD 13.0 with llvm-devel (15.0.0)
453055 shared_timed_mutex drd test fails with "Lock shared failed" message
453602 Missing command line option to enable/disable debuginfod
452802 Handle lld 9+ split RW PT_LOAD segments correctly
454040 s390x: False-positive memcheck:cond in memmem on arch13 systems
456171 [PATCH] FreeBSD: Don't record address errors when accessing the 'kern.ps_strings' sysctl struct
n-i-bz Implement vgdb invoker on FreeBSD
458845 PowerPC: The L field for the dcbf and sync instruction should be
3 bits in ISA 3.1.
458915 Remove register cache to fix 458915 gdbserver causes wrong syscall return
459031 Documentation on --error-exitcode incomplete
459477 XERROR messages lacks ending '\n' in vgdb
462007 Implicit int in none/tests/faultstatus.c
To see details of a given bug, visit
https://bugs.kde.org/show_bug.cgi?id=XXXXXX
where XXXXXX is the bug number as listed above.
(3.20.0.RC1: 20 Oct 2022)
Release 3.19.0 (11 Apr 2022)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
This release supports X86/Linux, AMD64/Linux, ARM32/Linux, ARM64/Linux,
PPC32/Linux, PPC64BE/Linux, PPC64LE/Linux, S390X/Linux, MIPS32/Linux,
MIPS64/Linux, ARM/Android, ARM64/Android, MIPS32/Android, X86/Android,
X86/Solaris, AMD64/Solaris, AMD64/MacOSX 10.12, X86/FreeBSD and
AMD64/FreeBSD. There is also preliminary support for X86/macOS 10.13,
AMD64/macOS 10.13 and nanoMIPS/Linux.
* ==================== CORE CHANGES ===================
* Fix Rust v0 name demangling.
* The Linux rseq syscall is now implemented as (silently) returning ENOSYS.
* Add FreeBSD syscall wrappers for __specialfd and __realpathat.
* Remove FreeBSD dependencies on COMPAT10, which fixes compatibility with HardenedBSD
* ================== PLATFORM CHANGES =================
* arm64:
- ignore the "v8.x" architecture levels, only look at actual CPU features
present. Fixes mismatch detected between RDMA and atomics features
preventing startup on some QEMU configurations.
- Implement LD{,A}XP and ST{,L}XP
- Fix incorrect code emitted for doubleword CAS.
* s390:
- Fix sys_ipc semtimedop syscall
- Fix VFLRX and WFLRX instructions
- Fix EXRL instruction with negative offset
* ppc64:
- Reimplement the vbpermq instruction support to generate less Iops and
avoid overflowing internal buffers.
- Fix checking for scv support to avoid "Facility 'SCV' unavailable (12),
exception" messages in dmsg.
- Fix setting condition code for Vector Compare quad word instructions.
- Fix fix lxsibzx, lxsihzx and lxsihzx instructions so they only load
their respective sized data.
- Fix the prefixed stq instruction in PC relative mode.
* ==================== TOOL CHANGES ===================
* Memcheck:
- Speed up --track-origins=yes for large (in the range of hundreds to
thousands of megabytes) mmap/munmaps.
* DRD/Helgrind:
- Several fixes for new versions of libstd++ using new posix try_lock
functions
* ==================== FIXED BUGS ====================
The following bugs have been fixed or resolved. Note that "n-i-bz"
stands for "not in bugzilla" -- that is, a bug that was reported to us
but never got a bugzilla entry. We encourage you to file bugs in
bugzilla (https://bugs.kde.org/enter_bug.cgi?product=valgrind) rather
than mailing the developers (or mailing lists) directly -- bugs that
are not entered into bugzilla tend to get forgotten about or ignored.
403802 leak_cpp_interior fails with some reachable blocks different than expected
435732 memcheck/tests/leak_cpp_interior fails with gcc11
444242 s390x: Valgrind crashes on EXRL with negative offset
444399 arm64: unhandled instruction 0xC87F2D89 (LD{,A}XP and ST{,L}XP).
== 434283
444481 gdb_server test failures on s390x
444495 dhat/tests/copy fails on s390x
444552 memcheck/tests/sem fails on s390x with glibc 2.34
444571 PPC, fix the lxsibzx and lxsihzx so they only load their respective
sized data.
444836 PPC, pstq instruction for R=1 is not storing to the correct address.
444925 fexecve syscall wrapper not properly implemented
445032 valgrind/memcheck crash with SIGSEGV when SIGVTALRM timer used and
libthr.so associated
445211 Fix out of tree builds
445300 [PATCH] Fix building tests with Musl
445011 SIGCHLD is sent when valgrind uses debuginfod-find
445354 arm64 backend: incorrect code emitted for doubleword CAS
445415 arm64 front end: alignment checks missing for atomic instructions
445504 Using C++ condition_variable results in bogus "mutex is locked simultaneously by two threads" warning
445607 Unhandled amd64-freebsd syscall: 247
445668 Inline stack frame generation is broken for Rust binaries
445916 Demangle Rust v0 symbols with .llvm suffix
446139 DRD/Helgrind with std::shared_timed_mutex::try_lock_until and try_lock_shared_until false positives
446138 DRD/Helgrind with std::timed_mutex::try_lock_until false positives
446281 Add a DRD suppression for fwrite
446103 Memcheck: `--track-origins=yes` causes extreme slowdowns for large mmap/munmap
446139 DRD/Helgrind with std::shared_timed_mutex::try_lock_until and try_lock_shared_until false
446251 TARGET_SIGNAL_THR added to enum target_signal
446823 FreeBSD - missing syscalls when using libzm4
447991 s390x: Valgrind indicates illegal instruction on wflrx
447995 Valgrind segfault on power10 due to hwcap checking code
449483 Powerpc: vcmpgtsq., vcmpgtuq,, vcmpequq. instructions not setting the
condition code correctly.
449672 ppc64 --track-origins=yes failures because of bad cmov addHRegUse
449838 sigsegv liburing the 'impossible' happened for io_uring_setup
450025 Powerc: ACC file not implemented as a logical overlay of the VSR
registers.
450437 Warn for execve syscall with argv or argv[0] being NULL
450536 Powerpc: valgrind throws 'facility scv unavailable exception'
451626 Syscall param bpf(attr->raw_tracepoint.name) points to unaddressable byte(s)
451827 [ppc64le] VEX temporary storage exhausted with several vbpermq instructions
451843 valgrind fails to start on a FreeBSD system which enforces W^X
To see details of a given bug, visit
https://bugs.kde.org/show_bug.cgi?id=XXXXXX
where XXXXXX is the bug number as listed above.
(3.19.0.RC1: 02 Apr 2022)
(3.19.0.RC2: 08 Apr 2022)
