Upstream commit: 59a6d5e9477695c41d6feef7ef8636f8f744f3c5
- Add AArch64 HWCAP2 values from Linux 6.2 to bits/hwcap.h
- crypt: Remove invalid end of page test badsalttest
- S390: Fix _FPU_SETCW/GETCW when compiling with Clang [BZ #30130]
- s390x: Regenerate ULPs.
- Add Arm HWCAP values from Linux 6.2 to bits/hwcap.h
- htl: Add pthreadtypes-arch.h for x86_64
- hurd: Implement TLS for x86_64
- htl: Make pthread_mutex_t pointer-aligned
- x86_64: Update libm test ulps
- localedata: de_DE should not use Fräulein
- LoongArch: Add math-barriers.h
- cdefs.h: fix "__clang_major" typo
- hppa: Drop old parisc-specific MADV_* constants
- hurd: Generalize init-first.c to support x86_64
- hurd: Simplify init-first.c further
- hurd: Mark some audit tests as unsupported
- htl: Mark select loop test as unsupported
- hurd: Mark RLIMIT_AS tests as unsupported
- aarch64: update libm test ulps
- powerpc:Regenerate ulps for hypot
- Update syscall lists for Linux 6.2
- tunables.texi: Change \code{1} to @code{1}
- x86-64: Add glibc.cpu.prefer_map_32bit_exec [BZ #28656]
- gmon: fix memory corruption issues [BZ# 30101]
- gmon: improve mcount overflow handling [BZ# 27576]
- gmon: Fix allocated buffer overflow (bug 29444)
- malloc: remove redundant check of unsorted bin corruption
- Use Linux 6.2 in build-many-glibcs.py
- Ignore MAP_VARIABLE in tst-mman-consts.py
- AArch64: Fix HP_TIMING_DIFF computation [BZ# 29329]
Upstream commit: a704fd9a133bfb10510e18702f48a6a9c88dbbd5:
- Create ChangeLog.old/ChangeLog.26. (tag: glibc-2.37)
- Prepare for glibc 2.37 release.
- x86: Fix strncat-avx2.S reading past length [BZ #30065]
- Update install.texi, and regenerate INSTALL.
- Update manual/contrib.texi.
- Update NEWS file with bug fixes.
- Regenerate configure.
- Update all PO files in preparation for release.
- doc: correct _FORTIFY_SOURCE doc in features.h
- libio: Update number of written bytes in dprintf implementation
- Drop already included glibc-dprintf-length.patch patch.
- Apply glibc-printf-grouping-swbz30068.patch to fix swbz#30068.
Upstream commit: 2f39e44a8417b4186a7f15bfeac5d0b557e63e03
- Account for octal marker in %#o format (rhbz#2165869)
- Use binutils 2.40 branch in build-many-glibcs.py
- Use MPFR 4.2.0, MPC 1.3.1 in build-many-glibcs.py
- Apply glibc-dprintf-length.patch to fix dprintf return value
regression.
Upstream commit: 5f55b22f4b3ea14c777a60f239d25dc4555eb804
- Drop glibc-rh2155825.patch; fix applied upstream
- hurd getcwd: Fix memory leak on error
- hurd fcntl: Make LOCKED macro more robust
- hurd: Make dl-sysdep __sbrk check __vm_allocate call
- htl: Drop duplicate check in __pthread_stack_alloc
- hurd hurdstartup: Initialize remaining fields of hurd_startup_data
- hurd _S_msg_add_auth: Initialize new arrays to 0
- htl: Check error returned by __getrlimit
- getdelim: ensure error indicator is set on error (bug 29917)
- htl: Fix sem_wait race between read and gsync_wait
- Avoid use of atoi in malloc
- Linux: Pass size argument of epoll_create to the kernel
- Simplify scripts/cross-test-ssh.sh configuration.
- Define MADV_COLLAPSE from Linux 6.1
- powerpc64: Increase SIGSTKSZ and MINSIGSTKSZ
- Update all PO files in preparation for release.
- Update kernel version to 6.1 in header constant tests
- Update syscall lists for Linux 6.1
- libio: Convert __vswprintf_internal to buffers (bug 27857)
- libio: Convert __obstack_vprintf_internal to buffers (bug 27124)
- libio: Convert __vdprintf_internal to buffers
- libio: Convert __vasprintf_internal to buffers
- libio: Convert __vsprintf_internal to buffers
- stdio-common: Add lock optimization to vfprintf and vfwprintf
- stdio-common: Convert vfprintf and related functions to buffers
- stdio-common: Add __translated_number_width
- stdio-common: Add __printf_function_invoke
- stdio-common: Introduce buffers for implementing printf
- locale: Implement struct grouping_iterator
- Use Linux 6.1 in build-many-glibcs.py
- Avoid use of atoi in some places in libc
Upstream commit c1c0dea38833751f36a145c322ce53c9a08332e1
- Linux: Remove epoll_create, inotify_init from syscalls.list (#2154747)
- Linux: Reflow and sort some Makefile variables
- mach: Drop remnants of old_CFLAGS
- mach: Fix passing -ffreestanding when checking for gnumach headers
- Force use of -ffreestanding when checking for gnumach headers
- elf: Fix tst-relro-symbols.py argument passing
- x86: Prevent SIGSEGV in memcmp-sse2 when data is concurrently modified [BZ #29863]
- Allow _Qp_fgt in sparc64 localplt.data
Upstream commit: 5dcd2d0ad02ff12c76355ef4f40947c1857ac482
- stdlib: Move _IO_cleanup to call_function_static_weak
- elf: Do not assume symbol order on tst-audit25{a,b}
- time: Use 64 bit time on tzfile
- nscd: Use 64 bit time_t on libc nscd routines (BZ# 29402)
- nis: Build libnsl with 64 bit time_t
- realloc: Return unchanged if request is within usable size
- Linux: Consolidate typesizes.h
- Linux: Make generic fcntl.h the default one
- Linux: make generic xstatver.h the default one
- Linux: Remove generic sysdep
- Linux: Assume and consolidate shutdown wire-up syscall
- Linux: Assume and consolidate listen wire-up syscall
- Linux: Assume and consolidate socketpair wire-up syscall
- Linux: Assume and consolidate socket wire-up syscall
- Linux: Assume and consolidate bind wire-up syscall
- Linux: consolidate ____longjmp_chk
- Linux: consolidate sendfile implementation
- Linux: consolidate unlink implementation
- Linux: consolidate symlink implementation
- Linux: consolidate rmdir implementation
- Linux: consolidate readlink implementation
- Linux: consolidate mkdir implementation
- Linux: consolidate link implementation
- Linux: consolidate lchown implementation
- Linux: consolidate inotify_init implementation
- Lninux: consolidate epoll_create implementation
- Linux: consolidate dup2 implementation
- Linux: consolidate chown implementation
- Linux: consolidate chmod implementation
- linux: Consolidate dl-origin.c
- linux: Use long int for syscall return value
- LoongArch: Use medium cmodel build libc_nonshared.a.
- x86_64: State assembler is being tested on sysdeps/x86/configure
- configure: Remove AS check
- configure: Remove check if ld is GNU
- configure: Remove check if as is GNU
- configure: Move locale tools early
Upstream commit: 8fb923ddc38dd5f4bfac4869d70fd80483fdb87a
- hurd: Make getrandom cache the server port
- powerpc64: Remove old strncmp optimization
- x86-64 strncpy: Properly handle the length parameter [BZ# 29839]
- x86-64 strncat: Properly handle the length parameter [BZ# 24097]
- ARC: update definitions in elf/elf.h
- scripts: Add "|" operator support to glibcpp's parsing
- Apply asm redirections in syslog.h before first use [BZ #27087]
- LoongArch: Add support for ilogb[f]
- LoongArch: Add support for scalb[f]
- LoongArch: Add support for scalbn[f]
- LoongArch: Use __builtin_logb{,f} with GCC >= 13
- Use GCC builtins for logb functions if desired.
- LoongArch: Use __builtin_llrint{,f} with GCC >= 13
- Use GCC builtins for llrint functions if desired.
- LoongArch: Use __builtin_lrint{,f} with GCC >= 13
- Use GCC builtins for lrint functions if desired.
- LoongArch: Use __builtin_rint{,f} with GCC >= 13
Upstream commit: f704192911c6c7b65a54beab3ab369fca7609a5d
- x86/fpu: Factor out shared avx2/avx512 code in svml_{s|d}_wrapper_impl.h
- x86/fpu: Cleanup code in svml_{s|d}_wrapper_impl.h
- x86/fpu: Reformat svml_{s|d}_wrapper_impl.h
- x86/fpu: Fix misspelled evex512 section in variety of svml files
- x86/fpu: Add missing ISA sections to variety of svml files
- stdio-common: Add missing dependencies (bug 29780)
- i386: Avoid rely on linker optimization to avoid relocation
- elf: Fix rtld-audit trampoline for aarch64
- Define in_int32_t_range to check if the 64 bit time_t syscall should be used
Upstream commit: 94628de77888c3292fc103840731ff85f283368e
- elf/tst-tlsopt-powerpc fails when compiled with -mcpu=power10 (BZ# 29776)
- LoongArch: Hard Float Support for fmaximum_mag_num{f/ }, fminimum_mag_num{f/ }.
- LoongArch: Hard Float Support for fmaximum_mag{f/ }, fminimum_mag{f/ }.
- LoongArch: Hard Float Support for fmaxmag{f/ }, fminmag{f/ }.
- LoongArch: Hard Float Support for fmaximum_num{f/ }, fminimum_num{f/ }.
- LoongArch: Hard Float Support for fmaximum{f/ }, fminimum{f/ }.
- LoongArch: Hard Float Support for float-point classification functions.
- LoongArch: Use __builtin_{fma, fmaf} to implement function {fma, fmaf}.
Upstream commit: 22a46dee24351fd5f4f188ad80554cad79c82524
- Linux: Support __IPC_64 in sysvctl *ctl command arguments (bug 29771)
- riscv: Get level 3 cache's information
- debug: Fix typo in tests-unsupported rule
- iconvdata/tst-table-charmap.sh: remove handling of old, borrowed format
- Makerules: Generate shlib.lds with -fuse-ld=bfd
- x86: Add avx2 optimized functions for the wchar_t strcpy family
- x86: Add evex optimized functions for the wchar_t strcpy family
- x86: Optimize and shrink st{r|p}{n}{cat|cpy}-avx2 functions
- x86: Optimize and shrink st{r|p}{n}{cat|cpy}-evex functions
- benchtests: Make str{n}{cat|cpy} benchmarks output json
- x86: Use VMM API in memcmpeq-evex.S and minor changes
- x86: Use VMM API in memcmp-evex-movbe.S and minor changes
- string: Add len=0 to {w}memcmp{eq} tests and benchtests
- Linux: Add ppoll fortify symbol for 64 bit time_t (BZ# 29746)
- hurd: Add sigtimedwait and sigwaitinfo support
Upstream commit: 8d291eabd541029d7ac705cc1ea112c58dfbb05f
- Apply asm redirection in gmp.h before first use
- Rewrite find_cxx_header config configure.ac
- elf/tlsdeschtab.h: Add the Malloc return value check in _dl_make_tlsdesc_dynamic()
- elf: Disable some subtests of ifuncmain1, ifuncmain5 for !PIE
- posix: Make posix_spawn extensions available by default
- x86_64: Implement evex512 version of strrchr and wcsrchr
- elf: Introduce <dl-call_tls_init_tp.h> and call_tls_init_tp (bug 29249)
- LoongArch: Fix ABI related macros in elf.h to keep consistent with binutils[1].
- scripts/glibcelf.py: Properly report <elf.h> parsing failures
- elf: Rework exception handling in the dynamic loader [BZ #25486]
- linux: Drop useless include from fstatat.c
- Fix OOB read in stdlib thousand grouping parsing [BZ #29727]
- linux: Fix fstatat on MIPSn64 (BZ #29730)
- elf: Remove allocate use on _dl_debug_printf
- nptl: Fix pthread_create.c build with clang
- allocate_once: Apply asm redirection before first use
- alloc_buffer: Apply asm redirection before first use
- configure: Use -Wno-ignored-attributes if compiler warns about multiple aliases
- Disable use of -fsignaling-nans if compiler does not support it
- intl: Fix clang -Wunused-but-set-variable on plural.c
- Apply asm redirection in not-cancel before first use
- malloc: Use uintptr_t for pointer alignment
- Use uintptr_t in fts for pointer alignment
- Fix build with GCC 13 _FloatN, _FloatNx built-in functions
- elf: Build tst-relr-mod[34]a.so with $(LDFLAGS-rpath-ORIGIN)
- x86-64: Improve evex512 version of strlen functions
- Correctly determine libc.so 'OUTPUT_FORMAT' when cross-compiling.
- Remove unused scratch_buffer_dupfree
- Fix elf/tst-dlmopen-twice not to exhaust static TLS
- Use uintptr_t in string/tester for pointer alignment
- stdlib/strfrom: Add copysign to fix NAN issue on riscv (BZ #29501)
- Fix resource/bug-ulimit1 test
- Fix missing NUL terminator in stdio-common/scanf13 test
- Fix off-by-one OOB read in elf/tst-tls20
- elf: Fix alloca size in _dl_debug_vdprintf
- malloc: Use uintptr_t in alloc_buffer
- Fix invalid pointer dereference in wcpcpy_chk
- Fix invalid pointer dereference in wcscpy_chk
- aarch64: Fix the extension header write in getcontext and swapcontext
- aarch64: Don't build wordcopy
- scripts: Use bool in tunables initializer
- longlong.h: update from GCC for LoongArch clz/ctz support
Upstream commit: 6f360366f7f76b158a0f4bf20d42f2854ad56264
- elf: Introduce to _dl_call_fini
- ld.so: Export tls_init_tp_called as __rtld_tls_init_tp_called
- scripts/localplt.awk: Handle DT_JMPREL with empty PLT (for C-SKY)
- Remove lingering libSegfault Makefile entries
- aarch64: Use memcpy_simd as the default memcpy
- aarch64: Cleanup memset ifunc
- elf: Reinstate on DL_DEBUG_BINDINGS _dl_lookup_symbol_x
- x86_64: Implement evex512 version of strchrnul, strchr and wcschr
- linux: Fix generic struct_stat for 64 bit time (BZ# 29657)
- manual: Add missing % in int conversion list
- Avoid undefined behaviour in ibm128 implementation of llroundl (BZ #29488)
- Remove all assembly optimizations for htonl and htons
- Remove htonl.S for i386/x86_64
- Fix BZ #29463 in the ibm128 implementation of y1l too
- Add ADDRB from Linux 6.0 to bits/termios-c_cflag.h
- x86: Use `testb` for FSRM check in memmove-vec-unaligned-erms
- x86: Use `testb` for case-locale check in str{n}casecmp-sse42
- x86: Use `testb` for case-locale check in str{n}casecmp-sse2
- x86: Use `testb` for case-locale check in str{n}casecmp-avx2
- x86: Add support for VEC_SIZE == 64 in strcmp-evex.S impl
- x86: Remove AVX512-BVMI2 instruction from strrchr-evex.S
- sysdeps: arm: Fix preconfigure script for ARMv8/v9 targets [BZ #29698]
- nis: Fix nis_print_directory
- linux: Avoid shifting a negative signed on POSIX timer interface
- Bench: Improve benchtests for memchr, strchr, strnlen, strrchr
- x86: Optimize strrchr-evex.S and implement with VMM headers
- x86: Optimize memrchr-evex.S
- x86: Optimize strnlen-evex.S and implement with VMM headers
- x86: Shrink / minorly optimize strchr-evex and implement with VMM headers
- x86: Optimize memchr-evex.S and implement with VMM headers
- x86_64: Implement evex512 version of memchr, rawmemchr and wmemchr
- String: Improve test coverage for memchr
- Use PTR_MANGLE and PTR_DEMANGLE unconditionally in C sources
- Introduce <pointer_guard.h>, extracted from <sysdep.h>
- x86-64: Move LP_SIZE definition to its own header
- math: Fix asin and acos invalid exception with old gcc
- x86: Update strlen-evex-base to use new reg/vec macros.
- x86: Remove now unused vec header macros.
- x86: Update memset to use new VEC macros
- x86: Update memmove to use new VEC macros
- x86: Update memrchr to use new VEC macros
- x86: Update VEC macros to complete API for evex/evex512 impls
- elf: Do not completely clear reused namespace in dlmopen (bug 29600)
- malloc: Switch global_max_fast to uint8_t
- Add NT_S390_PV_CPU_DATA from Linux 6.0 to elf.h
- Add AArch64 HWCAP2_EBF16 from Linux 6.0 to bits/hwcap.h
- String: Improve test coverage for memchr
- elf: Remove -fno-tree-loop-distribute-patterns usage on dl-support
- socket: Use offsetof in SUN_LEN (bug 29578)
- Expose all MAP_ constants in <sys/mman.h> unconditionally (bug 29375)
- LoongArch: Fix the condition to use PC-relative addressing in start.S
- arm: Enable USE_ATOMIC_COMPILER_BUILTINS (BZ #24774)
- csu: Disable stack protector for static-reloc for static-pie
- NEWS: Fix grammar
- elf: Simplify output of hwcap subdirectories in ld.so help
- elf: Remove _dl_string_hwcap
- Add NEWS entry for legacy hwcaps removal
- elf: Remove hwcap and bits_hwcap fields from struct cache_entry
- elf: Remove hwcap parameter from add_to_cache signature
- elf: Remove legacy hwcaps support from ldconfig
- elf: Remove legacy hwcaps support from the dynamic loader
- x86_64: Remove platform directory library loading test
- Update to Unicode 15.0.0 [BZ #29604]
- Update kernel version to 6.0 in header constant tests
- x86: Fix -Os build (BZ #29576)
- sunrpc: Suppress GCC -Os warning on user2netname
- rt: Initialize mq_send input on tst-mqueue{5,6}
- posix: Suppress -Os may be used uninitialized warnings on regexec
- posix: Suppress -Os warnings on fnmatch
- locale: prevent maybe-uninitialized errors with -Os [BZ #19444]
- Regenerate sysdeps/mach/hurd/bits/errno.h
- Update syscall lists for Linux 6.0
- nscd: Drop local address tuple variable [BZ #29607]
- Use Linux 6.0 in build-many-glibcs.py
- x86-64: Require BMI1/BMI2 for AVX2 strrchr and wcsrchr implementations
- x86-64: Require BMI2 and LZCNT for AVX2 memrchr implementation
- x86-64: Require BMI2 for AVX2 (raw|w)memchr implementations
- x86-64: Require BMI2 for AVX2 wcs(n)cmp implementations
- x86-64: Require BMI2 for AVX2 strncmp implementation
- x86-64: Require BMI2 for AVX2 strcmp implementation
- x86-64: Require BMI2 for AVX2 str(n)casecmp implementations
- x86: include BMI1 and BMI2 in x86-64-v3 level
- x86: Cleanup pthread_spin_{try}lock.S
- Benchtests: Add bench for pthread_spin_{try}lock and mutex_trylock
Upstream commit: c02e29a0ba47d636281e1a026444a1a0a254aa12
- nss: Use shared prefix in IPv4 address in tst-reload1
- nss: Enhance tst-reload1 coverage and logging
- Use C11 atomics instead of atomic_decrement_and_test
- Use C11 atomics instead of atomic_increment(_val)
- Use C11 atomics instead of atomic_and/or
- malloc: Print error when oldsize is not equal to the current size.
- Use '%z' instead of '%Z' on printf functions
- elf: Extract glibcelf constants from <elf.h>
- scripts: Enhance glibcpp to do basic macro processing
- scripts: Extract glibcpp.py from check-obsolete-constructs.py
- riscv: Remove RV32 floating point functions
- riscv: Consolidate the libm-test-ulps
- hurd: Fix SIOCADD/DELRT ioctls
- hurd: Drop struct rtentry and in6_rtmsg
- hurd: Add _IOT_ifrtreq to <net/route.h>
- elf: Use C11 atomics on _dl_mcount
- hurd: Use IF_NAMESIZE rather than IFNAMSIZ
- hurd: Add ifrtreq structure to net/route.h
- hppa: undef __ASSUME_SET_ROBUST_LIST
- linux: Use same type for MMAP2_PAGE_UNIT
- m68k: Enforce 4-byte alignment on internal locks (BZ #29537)
- nss: Fix tst-nss-files-hosts-long on single-stack hosts (bug 24816)
- nss: Implement --no-addrconfig option for getent
- gconv: Use 64-bit interfaces in gconv_parseconfdir (bug 29583)
- elf: Implement force_first handling in _dl_sort_maps_dfs (bug 28937)
- Linux: Do not skip d_ino == 0 entries in readdir, readdir64 (bug 12165)
- hurd: Factorize at/non-at functions
- tst-sprintf-errno: Update Hurd message length
- RISC-V: Allow long jumps to __syscall_error
- hurd: Make readlink* just reopen the file used for stat
- hurd: Fix readlink() hanging on fifo
- Fix BRE typos in check-safety.sh
- Makerules: fix MAKEFLAGS assignment for upcoming make-4.4 [BZ# 29564]
- Use relaxed atomics since there is no MO dependence
Upstream commit: f278835f594740f5913001430641cf1da4878670
- Drop glibc-deprecated-selinux-makedb.patch. Upstream has been ported
to new libselinux.
- makedb: fix build with libselinux >= 3.1 (Bug 26233)
- tst-sprintf-errno: Update Hurd message output
- Use C11 atomics instead of atomic_decrement(_val)
- Use C11 atomics instead atomic_add(_zero)
- mktime: improve heuristic for ca-1986 Indiana DST
- Assume HAVE_TZSET in time/mktime.c
- elf: Fix hwcaps string size overestimation
- errlist: add missing entry for EDEADLOCK (bug 29545)
- Do not define static_assert or thread_local in headers for C2x
- malloc: Use C11 atomics rather than atomic_exchange_and_add
- Add NEWS entry for CVE-2022-39046
- elf: Rename _dl_sort_maps parameter from skip to force_first
- scripts/dso-ordering-test.py: Generate program run-time dependencies
- math: x86: Use prefix for FP_INIT_ROUNDMODE
- scripts/build-many-glibcs.py: Use https:// for sourceware.org Git clones
- debug: test for more required cacellation points (BZ# 29274)
- elf.h: Remove duplicate definition of VER_FLG_WEAK
- syslog: Remove extra whitespace between timestamp and message (BZ#29544)
- LoongArch: Add soft float support.
- elf: Restore how vDSO dependency is printed with LD_TRACE_LOADED_OBJECTS (BZ #29539)
- nptl: x86_64: Use same code for CURRENT_STACK_FRAME and stackinfo_get_sp
This commit fixes wrap-find-debuginfo.sh so that:
* the .gnu_debuglink section of libc.so.6 is retained, and
* the redundant debuginfo file generated for ld.so is removed
Co-Authored-By: Benjamin Herrenschmidt <benh@amazon.com>
Reviewed-by: Florian Weimer <fweimer@redhat.com>
Upstream commit: c7509d49c4e8fa494120c5ead21338559dad16f5
- Apply asm redirections in wchar.h before first use
- resolv: Fix building tst-resolv-invalid-cname for earlier C standards
- syslog: Fix large messages (BZ#29536)
- posix: Fix macro expansion producing 'defined' has undefined behavior
- stdlib: Fix macro expansion producing 'defined' has undefined behavior
- S390: Always use svc 0
- nss_dns: Rewrite _nss_dns_gethostbyname4_r using current interfaces
- resolv: Add new tst-resolv-invalid-cname
- nss_dns: In gaih_getanswer_slice, skip strange aliases (bug 12154)
- nss_dns: Rewrite getanswer_r to match getanswer_ptr (bug 12154, bug 29305)
- nss_dns: Remove remnants of IPv6 address mapping
- nss_dns: Rewrite _nss_dns_gethostbyaddr2_r and getanswer_ptr
- nss_dns: Split getanswer_ptr from getanswer_r
- resolv: Add DNS packet parsing helpers geared towards wire format
- resolv: Add internal __ns_name_length_uncompressed function
- resolv: Add the __ns_samebinaryname function
- resolv: Add internal __res_binary_hnok function
- resolv: Add tst-resolv-aliases
- resolv: Add tst-resolv-byaddr for testing reverse lookup
- LoongArch: Use __builtin_{fmax,fmaxf,fmin,fminf} with GCC >= 13
- LoongArch: Fix ptr mangling/demangling features.
- nscd: Fix netlink cache invalidation if epoll is used [BZ #29415]
- Add test for bug 29530
- Makeconfig: Set pie-ccflag to -fPIE by default [BZ# 29514]
- hurd: Fix vm_size_t incoherencies
- mach: Make xpg_strerror_r set a message on error
- mach: Fix incoherency between perror and strerror
- elf: Call __libc_early_init for reused namespaces (bug 29528)
- csu: Change start code license to have link exception
- s390: Move hwcaps/platform names out of _rtld_global_ro
- Revert "Detect ld.so and libc.so version inconsistency during startup"
- Add NT_LOONGARCH_* from Linux 5.19 to elf.h
- Detect ld.so and libc.so version inconsistency during startup
- Merge getopt patch from Gnulib
- Merge _GL_UNUSED C23 patch from Gnulib
- LoongArch: Fix dl-machine.h code formatting.
- scripts/glibcelf.py: Add hashing support
- hurd: Fix starting static binaries with stack protection enabled
- htl: Make pthread*_cond_timedwait register wref before releasing mutex
- htl: make __pthread_hurd_cond_timedwait_internal check mutex is held
- Add AArch64 HWCAP2_* constants from Linux 5.19
Upstream commit: a727220b37efc9d4d558a77c5fc57f3af99a4829
- Add AGROUP from Linux 5.19 to sys/acct.h, remove Alpha version (bug 29502)
- alpha: Fix generic brk system call emulation in __brk_call (bug 29490)
- hurd: Assume non-suid during bootstrap
Upstream commit: 9125e43daf92e3d4e69044a54b9fe9ed88c861ad
- Use binutils 2.39 branch in build-many-glibcs.py
- S390: Fix werror=unused-variable in ifunc-impl-list.c.
- Ensure calculations happen with desired rounding mode in y1lf128
- localedata: Convert French language locales (fr_*) to UTF-8
- Linux: Fix enum fsconfig_command detection in <sys/mount.h>
- elf: Run tst-audit-tlsdesc, tst-audit-tlsdesc-dlopen everywhere
- Move ip_mreqn structure from Linux to generic
- malloc: Do not use MAP_NORESERVE to allocate heap segments
- Linux: Terminate subprocess on late failure in tst-pidfd (bug 29485)
- non-linux: bits/in.h: Add more RFC options
Upstream commit: 453b88efe6fa79f5c7c6fccc3a520c75fdd43074
- arm: Remove nested functionf rom relocate_pc24
- linux: Fix sys/mount.h usage with kernel headers
- linux: Use compile_c_snippet to check linux/mount.h availability
- linux: Mimic kernel defition for BLOCK_SIZE
- linux: Use compile_c_snippet to check linux/pidfd.h availability
- glibcextract.py: Add compile_c_snippet
- LoongArch: Add pointer mangling support.
- AArch64: Fix typo in sve configure check (BZ# 29394)
- libio: Improve performance of IO locks
- tst-process_madvise: Check process_madvise-syscall support.
- elf.h: Add ELFCOMPRESS_ZSTD
- inet: Turn __ivaliduser into a compatibility symbol
- x86: Fix `#define STRCPY` guard in strcpy-sse2.S
- elf: Replace `strcpy` call with `memcpy` [BZ #29454]
- soft-fp: Add fixhf[uns][di|si] and float[uns][di|si]hf
- i386: Use cmpl instead of cmp
- i386: Use fldt instead of fld on e_logl.S
- i386: Replace movzx with movzbl
- dlfcn: Pass caller pointer to static dlopen implementation (bug 29446)
- malloc: Correct the documentation of the top_pad default
- i386: Remove RELA support
- arm: Remove RELA support
- Remove ldd libc4 support
- Assume only FLAG_ELF_LIBC6 suport
- Remove left over LD_LIBRARY_VERSION usages
- Linux: Remove exit system call from _exit
- LoongArch: Add vdso support for gettimeofday.
- Update kernel version to 5.19 in header constant tests
- assert: Do not use stderr in libc-internal assert
- nptl: Remove uses of assert_perror
- stdio: Clean up __libc_message after unconditional abort
- Update syscall lists for Linux 5.19
- Use Linux 5.19 in build-many-glibcs.py
- socket: Check lengths before advancing pointer in CMSG_NXTHDR
- Don't use unsupported format string in ld.so (bug 29427)
- htl: Let pthread_self and cancellability called early
- stdlib: Simplify arc4random_uniform
- malloc: Use __getrandom_nocancel during tcache initiailization
- Remove spurious references to _dl_open_hook
- Open master branch for glibc 2.37 development
Upstream commit: 33f1b4c1452b33991e670f636ebe98b90a405e10
- wcsmbs: Add missing test-c8rtomb/test-mbrtoc8 dependency
- stdlib: Suppress gcc diagnostic that char8_t is a keyword in C++20 in uchar.h.
- Create ChangeLog.old/ChangeLog.25. (tag: glibc-2.36)
- Prepare for glibc 2.36 release.
- Update install.texi, and regenerate INSTALL.
- Update NEWS bug list.
- Update libc.pot for 2.36 release.
- tst-pidfd.c: UNSUPPORTED if we get EPERM on valid pidfd_getfd call
- stdlib: Tuned down tst-arc4random-thread internal parameters
- LoongArch: Add greg_t and gregset_t.
- LoongArch: Fix VDSO_HASH and VDSO_NAME.
- riscv: Update rv64 libm test ulps
- riscv: Update nofpu libm test ulps
No regressions seen in mass-prebuild of critical-path-base packages.
Upstream commit: 49889fb256a7f9b894b2d16fea23de1ac25b65e2
- x86: Add support to build st{p|r}{n}{cpy|cat} with explicit ISA level
- x86: Add support to build wcscpy with explicit ISA level
- x86: Add support to build strcmp/strlen/strchr with explicit ISA level
- elf: Fix wrong fscanf usage on tst-pldd
- Apply asm redirections in stdio.h before first use [BZ #27087]
- S390: Define SINGLE_THREAD_BY_GLOBAL only on s390x
- x86: Add missing rtm tests for strcmp family
- x86: Remove unneeded rtld-wmemcmp
- x86: Move wcslen SSE2 implementation to multiarch/wcslen-sse2.S
- x86: Move wcschr SSE2 implementation to multiarch/wcschr-sse2.S
- x86: Move strcat SSE2 implementation to multiarch/strcat-sse2.S
- x86: Move strchr SSE2 implementation to multiarch/strchr-sse2.S
- x86: Move strrchr SSE2 implementation to multiarch/strrchr-sse2.S
- x86: Move memrchr SSE2 implementation to multiarch/memrchr-sse2.S
- x86: Move strcpy SSE2 implementation to multiarch/strcpy-sse2.S
- x86: Move strlen SSE2 implementation to multiarch/strlen-sse2.S
- x86: Move strcmp SSE42 implementation to multiarch/strcmp-sse4_2.S
- x86: Move wcscmp SSE2 implementation to multiarch/wcscmp-sse2.S
- x86: Move strcmp SSE2 implementation to multiarch/strcmp-sse2.S
- x86: Rename STRCASECMP_NONASCII macro to STRCASECMP_L_NONASCII
- nptl: Fix ___pthread_unregister_cancel_restore asynchronous restore
- x86: Remove __mmask intrinsics in strstr-avx512.c
- x86: Remove generic strncat, strncpy, and stpncpy implementations
- i386: Remove -Wa,-mtune=i686
- x86-64: Remove redundant strcspn-generic/strpbrk-generic/strspn-generic
- elf: Rename tst-audit26 to tst-audit28
- x86-64: Don't mark symbols as hidden in strcmp-XXX.S
- stdlib: Tests for mbrtoc8, c8rtomb, and the char8_t typedef.
- stdlib: Implement mbrtoc8, c8rtomb, and the char8_t typedef.
- gconv: Correct Big5-HKSCS conversion to preserve all state bits. [BZ #25744]
- aarch64: Optimize string functions with shrn instruction
- test-container: return UNSUPPORTED for ENOSPC on clone()
- x86: Add support for building {w}memcmp{eq} with explicit ISA level
- x86: Add support for building {w}memset{_chk} with explicit ISA level
- x86: Add support for building {w}memmove{_chk} with explicit ISA level
- x86: Add support for building str{c|p}{brk|spn} with explicit ISA level
- x86: Add comment explaining no Slow_SSE4_2 check in ifunc-sse4_2
- Replace __libc_multiple_threads with __libc_single_threaded
- linux: Add mount_setattr
- linux: Add tst-mount to check for Linux new mount API
- linux: Add open_tree
- linux: Add fspick
- linux: Add fsconfig
- AArch64: Reset HWCAP2_AFP bits in FPCR for default fenv
- elf: Fix direction of NODELETE log messages during symbol lookup
We'll use baserelease here for two reasons:
- It is known to rpmdev-bumpspec, so it will be properly handled for mass-
rebuilds
- It allows using the Release number without the %%dist tag in the dependency
generator to make the generated requires interchangeable between Rawhide
and ELN (.elnYY < .fcXX).
Signed-off-by: Stephen Gallagher <sgallagh@redhat.com>
Upstream commit: 7519dee356a0ab21c8990e59ed05dd48a4e573a0
- malloc: Simplify checked_request2size interface
- stdlib: Simplify buffer management in canonicalize
- localedef: Support building for older C standards
- de_DE: Convert to UTF-8
- locale: localdef input files are now encoded in UTF-8
- locale: Introduce translate_unicode_codepoint into linereader.c
- locale: Fix signed char bug in lr_getc
- locale: Turn ADDC and ADDS into functions in linereader.c
- libc-symbols.h: remove unused macros
Upstream commit: 8ee2c043cfb35c48b45c7c5aed4022a8a7352bdc
- Fix hurd namespace issues for internal signal functions
- argp: Remove old includes in !_LIBC case
- Use GCC 12 branch in build-many-glibcs.py
- Refactor internal-signals.h
- riscv: Use memcpy to handle unaligned access when fixing R_RISCV_RELATIVE
- AArch64: Add asymmetric faulting mode for tag violations in mem.tagging tunable
- linux: Fix mq_timereceive check for 32 bit fallback code (BZ 29304)
- x86: Add missing IS_IN (libc) check to strncmp-sse4_2.S
- x86: Add missing IS_IN (libc) check to strcspn-sse4.c
- x86: Add missing IS_IN (libc) check to memmove-ssse3.S
- x86-64: Properly indent X86_IFUNC_IMPL_ADD_VN arguments
- x86-64: Small improvements to dl-trampoline.S
- x86: Move mem{p}{mov|cpy}_{chk_}erms to its own file
- x86: Move and slightly improve memset_erms
- x86: Add definition for __wmemset_chk AVX2 RTM in ifunc impl list
- linux: Remove unnecessary nice.c and signal.c
- nptl: Remove unused members from struct pthread
- Linux: Forward declaration of struct iovec for process_madvise
- x86: Add more feature definitions to isa-level.h
Upstream commit: a3563f3f369878467dd74aeb360448119a7a4b41
- elf: Fix -DNDEBUG warning in _dl_start_args_adjust
- elf: Fix compile error with -Werror and -DNDEBUG
- x86-64: Only define used SSE/AVX/AVX512 run-time resolvers
- x86: Move CPU_FEATURE{S}_{USABLE|ARCH}_P to isa-level.h
- x86: Fix backwards Prefer_No_VZEROUPPER check in ifunc-evex.h
- x86: Rename strstr_sse2 to strstr_generic as it uses string/strstr.c
- x86: Remove unused file wmemcmp-sse4
- x86: Put wcs{n}len-sse4.1 in the sse4.1 text section
- x86: Align entry for memrchr to 64-bytes.
- Makerules: Remove no-op -Wl,-d when linking libc_pic.os
- m68k: optimize RTLD_START
- misc: Optimize internal usage of __libc_single_threaded
- linux: Add move_mount
- linux: Add fsmount
- linux: Add fsopen
- resolv/tst-resolv-noaaaa: Support building for older C standards
- resolv: Implement no-aaaa stub resolver option
- support: Change non-address output format of support_format_dns_packet
- riscv: Use elf_machine_rela_relative to handle R_RISCV_RELATIVE
- x86: Remove faulty sanity tests for RTLD build with no multiarch
- stdlib: Fixup mbstowcs NULL __dst handling. [BZ #29279]
- x86: Replace all sse instructions with vex equivilent in avx+ files
- x86: Add support for compiling {raw|w}memchr with high ISA level
- x86: Add defines / utilities for making ISA specific x86 builds
- stdlib: Remove attr_write from mbstows if dst is NULL [BZ: 29265]
- stdlib: Remove trailing whitespace from Makefile
- debug: make __read_chk a cancellation point (bug 29274)
- s390: use LC_ALL=C for readelf call
- s390: use $READELF
Upstream commit: 999835533bc60fbd0b0b65d2412a6742e5a54b9d
- socket: Fix mistyped define statement in socket/sys/socket.h (BZ #29225)
- Declare timegm for ISO C2X
- Add PT_AARCH64_MEMTAG_MTE from Linux 5.18 to elf.h
- grep: egrep -> grep -E, fgrep -> grep -F
- string.h: Fix boolean spelling in comments
- elf: Add #include <errno.h> for use of E* constants.
- elf: Add #include <sys/param.h> for MAX usage.
- linux: Add process_mrelease
- linux: Add process_madvise
- linux: Set tst-pidfd-consts unsupported for kernels headers older than 5.10
- testrun.sh: Support passing strace and valgrind arguments
- Linux: Adjust struct rseq definition to current kernel version
- iconv: Use 64 bit stat for gconv_parseconfdir (BZ# 29213)
- catgets: Use 64 bit stat for __open_catalog (BZ# 29211)
- inet: Use 64 bit stat for ruserpass (BZ# 29210)
- socket: Use 64 bit stat for isfdtype (BZ# 29209)
- posix: Use 64 bit stat for fpathconf (_PC_ASYNC_IO) (BZ# 29208)
- posix: Use 64 bit stat for posix_fallocate fallback (BZ# 29207)
- misc: Use 64 bit stat for getusershell (BZ# 29203)
- misc: Use 64 bit stat for daemon (BZ# 29203)
- linux: use statx for fstat if neither newfstatat nor fstatat64 is present
- Add MADV_DONTNEED_LOCKED from Linux 5.18 to bits/mman-linux.h
- Add HWCAP2_MTE3 from Linux 5.18 to AArch64 bits/hwcap.h
- i686: Use generic sincosf implementation for SSE2 version
- benchtests: Add workload name for sincosf
- i686: Use generic sinf implementation for SSE2 version
- i686: Use generic cosf implementation for SSE2 version
- benchtests: Add workload name for cosf
- x86_64: Optimize sincos where sin/cos is optimized (bug 29193)
- manual: fix reference to source file
- Add SOL_SMC from Linux 5.18 to bits/socket.h
- elf: Remove _dl_skip_args
- x86_64: Remove _dl_skip_args usage
- sparc: Remove _dl_skip_args usage
- sh: Remove _dl_skip_args usage
- s390: Remove _dl_skip_args usage
- riscv: Remove _dl_skip_args usage
- nios2: Remove _dl_skip_args usage (BZ# 29187)
- mips: Remove _dl_skip_args usage
- microblaze: Remove _dl_skip_args usage
- m68k: Remove _dl_skip_args usage
- ia64: Remove _dl_skip_args usage
- i686: Remove _dl_skip_args usage
- hppa: Remove _dl_skip_args usage (BZ# 29165)
- csky: Remove _dl_skip_args usage
- arc: Remove _dl_skip_args usage
- arm: Remove _dl_skip_args usage
- alpha: Remove _dl_skip_args usage
- benchtests: Improve benchtests for strstr, memmem, and memchr
- dlsym: Make RTLD_NEXT prefer default version definition [BZ #14932]
- x86-64: Ignore r_addend for R_X86_64_GLOB_DAT/R_X86_64_JUMP_SLOT
- x86_64: Implement evex512 version of strlen, strnlen, wcslen and wcsnlen
- Update kernel version to 5.18 in header constant tests
- String: Improve overflow test coverage for strnlen
Upstream commit: 3d9926663cba19f40d26d8a8ab3b2a7cc09ffb13
- Update syscall-names.list for Linux 5.18
- Fix deadlock when pthread_atfork handler calls pthread_atfork or dlclose
- Use Linux 5.18 in build-many-glibcs.py
- stdio-common: Simplify printf_unknown interface in vfprintf-internal.c
- stdio-common: Move union printf_arg int <printf.h>
- stdio-common: Add printf specifier registry to <printf.h>
- elf/dl-reloc.c: Copyright The GNU Toolchain Authors
- benchtests: Improve bench-strnlen.c
- math: Add math-use-builtins-fabs (BZ#29027)
- linux: Add CLONE_NEWTIME from Linux 5.6 to bits/sched.h
- Revert "[ARM][BZ #17711] Fix extern protected data handling"
- Revert "[AArch64][BZ #17711] Fix extern protected data handling"
- elf: Rewrite long RESOLVE_MAP macro to an always_inline static function
Upstream commit 748df8126ac69e68e0b94e236ea3c2e11b1176cb
- dlfcn: Move RTLD_DEFAULT/RTLD_NEXT outside __USE_GNU
- elf: Optimize _dl_new_hash in dl-new-hash.h
- nss: Optimize nss_hash in nss_hash.c
- benchtests: Add benchtests for dl_elf_hash, dl_new_hash and nss_hash
- nss: Add tests for the nss_hash in nss_hash.h
- elf: Add tests for the dl hash funcs (_dl_new_hash and _dl_elf_hash)
- elf: Refactor dl_new_hash so it can be tested / benchmarked
- locale: Add more cached data to LC_CTYPE
- locale: Remove private union from struct __locale_data
- locale: Remove cleanup function pointer from struct __localedata
- locale: Call _nl_unload_locale from _nl_archive_subfreeres
- stdio-common: Add tst-memstream-string for open_memstream overflow
- __printf_fphex always uses LC_NUMERIC
- vfprintf: Consolidate some multibyte/wide character processing
- vfprintf: Move argument processing into vfprintf-process-arg.c
- stdio-common: Add tst-vfprintf-width-i18n to cover numeric field width
- string.h: fix __fortified_attr_access macro call [BZ #29162]
- Enable DT_RELR in glibc shared libraries and PIEs automatically
- S390: Enable static PIE
- linux: Add tst-pidfd.c
- linux: Add P_PIDFD
- linux: Add pidfd_send_signal
- linux: Add pidfd_getfd
- linux: Add pidfd_open
- aarch64: Move ld.so _start to separate file and drop _dl_skip_args
- linux: Add a getauxval test [BZ #23293]
- rtld: Remove DL_ARGV_NOT_RELRO and make _dl_skip_args const
- rtld: Use generic argv adjustment in ld.so [BZ #23293]
- scripts/glibcelf.py: Add *T_RISCV_* constants
- Remove dl-librecon.h header.
- elf: Remove ldconfig kernel version check
- Remove kernel version check
- linux: Use /sys/devices/system/cpu on __get_nprocs_conf (BZ#28991)
- csu: Implement and use _dl_early_allocate during static startup
- Linux: Introduce __brk_call for invoking the brk system call
- sys/cdefs.h: Do not require C++ compilers to define __STDC__
- fortify: Ensure that __glibc_fortify condition is a constant [BZ #29141]
- Update RISC-V specific ELF definitions
Upstream commit: 9403b71ae97e3f1a91c796ddcbb4e6f044434734
- x86_64: Remove bzero optimization
- RISC-V: Use an autoconf template to produce `preconfigure'
- MIPS: Use an autoconf template to produce `preconfigure'
- m68k: Use an autoconf template to produce `preconfigure'
- C-SKY: Use an autoconf template to produce `preconfigure'
- Remove configure fno_unit_at_a_time
- stdio: Remove the usage of $(fno-unit-at-a-time) for siglist.c
- stdio: Remove the usage of $(fno-unit-at-a-time) for errlist.c
- Add declare_object_symbol_alias for assembly codes (BZ #28128)
- wcrtomb: Make behavior POSIX compliant
Upstream commit: 8162147872491bb5b48e91543b19c49a29ae6b6d
- nptl: Add backoff mechanism to spinlock loop
- Linux: Implement a useful version of _startup_fatal
- ia64: Always define IA64_USE_NEW_STUB as a flag macro
- linux: Fix posix_spawn return code if clone fails (BZ#29109)
- benchtests: Add wcrtomb microbenchmark
- clock_settime/clock_gettime: Use __nonnull to avoid null pointer
- clock_adjtime: Use __nonnull to avoid null pointer
- ntp_xxxtimex: Use __nonnull to avoid null pointer
- adjtimex/adjtimex64: Use __nonnull to avoid null pointer
- hurd spawni: Fix reauthenticating closed fds
- Linux: Define MMAP_CALL_INTERNAL
- i386: Honor I386_USE_SYSENTER for 6-argument Linux system calls
- i386: Remove OPTIMIZE_FOR_GCC_5 from Linux libc-do-syscall.S
- manual: Clarify that abbreviations of long options are allowed
