Upstream commit: e9e7f24543e6d1b0a31641f144697e261df6ccd7
- configure: Replace obsolete AC_TRY_LINK with AC_LINK_IFELSE
- configure: Remove obsolete AC_CHECK_TOOL_PREFIX
- configure: Replace obsoleted AC_HELP_STRING with AS_HELP_STRING
- Update syscall lists for Linux 5.11.
- Correct hppa EFD_NONBLOCK, IN_NONBLOCK, SFD_NONBLOCK and TFD_NONBLOCK defines.
- string: Work around GCC PR 98512 in rawmemchr
- tst: time: Provide Y2038 tests for mktime (tst-mktime4.c)
- tst: Provide test for difftime
- tst: Provide test for ctime
- tst: Provide test for sched_rr_get_interval
- S390: Add new hwcap values.
- aarch64: Fix sys/ptrace.h if linux headers are included
- elf: Do not copy vDSO soname when setting up link map
- x86: Remove the extra space between "# endif"
- hurd: Fix fstatfs build failure
- linux: Remove stat-check.c
- linux: Remove overflow.h
- linux: Consolidate internal_statvfs
- linux: Consolidate statvfs implementations
- linux: Consolidate fstatvfs implementations
- linux: Consolidate statfs implementations
- linux: Consolidate fstatfs implementations
- linux: Set LFS statfs as default
- linux: Set default kernel_stat.h to LFS
- linux: Fix STATFS_IS_STATFS64 definition
Upstream commit: df359a25ba6f6bda06104229fbfe284c1fb30915
- Revert "Make libc symbols hidden in static PIE" [BZ #27237]
The previous auto-sync failed to build due to the above bug.
Upstream commit: 3725ee39dbd4d7c17fc276b3ea4248cc16e4b522
- benchtests: Do not build bench-timing-type with MODULE_NAME=libc
- aarch64: Fix the list of tested IFUNC variants [BZ #26818]
- Update INSTALL with package versions that are known to work
- aarch64: Move and update the definition of MTE_ENABLED
- Fix misplaced const
- Update C-SKY libm-test-ulps
- manual: Correct argument order in mount examples [BZ #27207]
- linux: mips: Fix getdents64 fallback on mips64-n32
- x86: Properly match CPU features in /proc/cpuinfo [BZ #27222]
- x86-64: Update tst-glibc-hwcaps-2.c for x86-64 baseline
- powerpc64: Select POWER9 machine for the scv instruction
- x86: Check ifunc resolver with CPU_FEATURE_USABLE [BZ #27072]
- Revert "linux: Move {f}xstat{at} to compat symbols" for static build
- aarch64: revert memcpy optimze for kunpeng to avoid performance degradation
- Make libc symbols hidden in static PIE
- csu: Move static pie self relocation later [BZ #27072]
- Use hidden visibility for early static PIE code
- csu: Avoid weak ref for __ehdr_start in static PIE
- configure: Check for static PIE support
- elf: Avoid RELATIVE relocs in __tunables_init
- elf: Make the tunable struct definition internal only
- <sys/platform/x86.h>: Remove the C preprocessor magic
- posix: Fix fnmatch.c on bootstrap
- stdlib: Add testcase for BZ #26241
- posix: Fix regex_internal.h on bootstrap
- Use <startup.h> in __libc_init_secure
- elf: Avoid RELATIVE relocation for _dl_sysinfo
- libmvec: Add extra-test-objs to test-extras
- Hurd: Add rtld-strncpy-c.c
- Update MIPS libm-test-ulps.
- Update arm libm-test-ulps.
- Update powerpc-nofpu libm-test-ulps.
- Update hppa libm-test-ulps
- ARC: nofpu: Regenerate ulps
- ld.so: Add --list-tunables to print tunable values
- math/test-tgmath2: Fix fabs failure when no long double
- x86: Move x86 processor cache info to cpu_features
- Fix x86 build with --enable-tunable=no
- ifuncmain6pie: Remove the circular IFUNC dependency [BZ #20019]
The info pages of The GNU C Library Reference Manual included in the
devel subpackage, were (slightly) increasing the installed size of
systems that required the devel object files but not the manual.
In addition to this, a number of unnecessary files (READMEs, NEWS,
INSTALL, and an example gai.conf) were included in either the main
package or the common subpackage.
This commit introduces a new glibc-doc.noarch subpackage and moves the
info pages from the devel subpackage and the debugger interface
documentation from the main package into glibc-doc. It also removes all
the unnecessary documentation-like files mentioned above.
nscd is planned for a phased removal, with deprecation in Fedora 34:
https://fedoraproject.org/wiki/Changes/DeprecateNSCD
... and removal in Fedora 35:
https://fedoraproject.org/wiki/Changes/RemoveNSCD
systemd-resolved is now enabled by default for DNS caching in Fedora,
and sssd is capable of caching the remaining named services that nscd
handles. It is therefore time to retire nscd in Fedora and move to more
modern named services caches.
Upstream commit: cf1290064598def8dfeddec3d86d98495aee1fba
- Use the right argument code in unnormal tests
- ldconfig/x86: Store ISA level in cache and aux cache
- elf: work around a gcc bug in elf_get_dynamic_info
- x86: Set header.feature_1 in TCB for always-on CET [BZ #27177]
- posix: consume less entropy on tempname
- Makerules: Do not require startup files for format.lds probe object
- install: Replace scripts/output-format.sed with objdump -f [BZ #26559]
- math: Add BZ#18980 fix back on dbl-64 cosh
- posix: Sync tempname with gnulib [BZ #26648]
- posix: Fix return value of system if shell can not be executed [BZ #27053]
- support: Add xchmod wrapper
- Update STATX_ATTR_DAX value from Linux 5.10.
- riscv: Initialize $gp before resolving the IRELATIVE relocation
- riscv: support GNU indirect function
- posix: Correct attribute access mode on readlinkat [BZ #27024].
- Add xfchmod to libsupport
- Add xchdir to libsupport.
- POSIX locale: Fix typo in comment
- ARC: Regenerate ulps
- mntent: Use __putc_unlocked instead of fputc_unlocked
- aarch64: define PI_STATIC_AND_HIDDEN
- Update NEWS for CVE-2019-25013.
- x86: Support GNU_PROPERTY_X86_ISA_1_V[234] marker [BZ #26717]
- Remove dbl-64/wordsize-64 (part 2)
- Remove dbl-64/wordsize-64
- Add SEGV_MTEAERR and SEGV_MTESERR from Linux 5.10.
- support: Add support_small_thread_stack_size
- stdlib: Sync canonicalize with gnulib [BZ #10635] [BZ #26592] [BZ #26341] [BZ #24970]
- malloc: Add scratch_buffer_dupfree
- Import filename.h from gnulib
- Import idx.h from gnulib
- alpha: Provide wait4 for static library [BZ #27150]
- aarch64: push the set of rules before falling into slow path
- nptl: Fix comment typo in pthread_cond_wait.c
- x86: Check IFUNC definition in unrelocated executable [BZ #20019]
- hurd: Fix mmap(!MAP_FIXED) on bogus address
Upstream commit: 3ec5d83d2a237d39e7fd6ef7a0bc8ac4c171a4a5
- Drop glibc-fedora-__libc_multiple_libcs.patch. Replaced by upstream's
__libc_initial flag.
- Adjust glibc-rh819430.patch to upstream's gnulib merge of fnmatch.
- Disable -Werror again due to GCC PR98512.
- x86-64: Avoid rep movsb with short distance [BZ #27130]
- aarch64: fix stack missing after sp is updated
- nptl: Remove set*id, set*gid files which are not built
- Drop nan-pseudo-number.h usage from tests
- posix: Sync fnmatch with gnulib
- Sync flexmember.h with gnulib
- Sync intprops.h with gnulib
- posix: Sync glob code with gnulib
- posix: Sync regex code with gnulib
- Move generic nan-pseudo-number.h to ldbl-96
- Sync FDL from https://www.gnu.org/licenses/fdl-1.3.texi
- Sync move-if-change from Gnulib
- Update automatically-generated copyright dates
- Update copyright dates not handled by scripts/update-copyrights.
- Update copyright dates with scripts/update-copyrights
- aarch64: use PTR_ARG and SIZE_ARG instead of DELOUSE
- nonstring: Enable __FORTIFY_LEVEL=3
- string: Enable __FORTIFY_LEVEL=3
- Introduce _FORTIFY_SOURCE=3
- Warn on unsupported fortification levels
- powerpc: Use scv instruction on clone when available
- powerpc: Runtime selection between sc and scv for syscalls
- malloc: preserve errno on mcheck hooks [BZ #17924]
- x86 long double: Add tests for pseudo normal numbers
- x86 long double: Consider pseudo numbers as signaling
- io: Remove xmknod{at} implementations
- io: Remove xstat implementations
- free: preserve errno [BZ#17924]
- hurd: Accept including hurd/version.h
- hurd: Add WSTOPPED/WCONTINUED/WEXITED/WNOWAIT support [BZ #23091]
- support: Make support_process_state_wait wait less
- hurd: set sigaction for signal preemptors in arch-independent file
- hurd: Fix spawni SPAWN_XFLAGS_TRY_SHELL with empty argv
- hurd: Try shell in posix_spawn* only in compat mode
- Remove _ISOMAC check from <cpu-features.h>
- x86: Remove the duplicated CPU_FEATURE_CPU_P
- Partially revert 681900d29683722b1cb0a8e565a0585846ec5a61
- x86 long double: Support pseudo numbers in isnanl
- x86 long double: Support pseudo numbers in fpclassifyl
- MTE: Do not pad size in realloc_check
- tests-mcheck: New variable to run tests with MALLOC_CHECK_=3
- elf: Account for glibc-hwcaps/ prefix in _dl_important_hwcaps
- misc: Use __ferror_unlocked instead of ferror
- s390x: Regenerate ulps
- powerpc: Regenerate ulps
- addmntent: Remove unbounded alloca usage from getmntent [BZ#27083]
- <sys/platform/x86.h>: Add Intel LAM support
- i386: Regenerate ulps
- aarch64: update ulps.
- aarch64: Add aarch64-specific files for memory tagging support
- aarch64: Add sysv specific enabling code for memory tagging
- linux: Add compatibility definitions to sys/prctl.h for MTE
- malloc: Basic support for memory tagging in the malloc() family
- elf: Add a tunable to control use of tagged memory
- config: Allow memory tagging to be enabled when configuring glibc
- alpha: Remove anonymous union in struct stat [BZ #27042]
- add inputs to auto-libm-test-in yielding larger errors (binary64, x86_64)
- m68k: fix clobbering a5 in setjmp() [BZ #24202]
- iconv add iconv_close before the function returned with bad value.
- iconv: use iconv_close after iconv_open
- Fix buffer overrun in EUC-KR conversion module (bz #24973)
- hurd: Make trampoline fill siginfo ss_sp from sc_uesp
- Hurd: make sigstates hold a reference on thread ports
- profil-counter: Add missing SIGINFO case
- hurd: implement SA_SIGINFO signal handlers.
- hurd: Fix ELF_MACHINE_USER_ADDRESS_MASK value
- hurd: Note when the vm_map kernel bug was fixed
- hurd: Also turn KERN_INVALID_ADDRESS to EINVAL
- ieee754: Remove unused __sin32 and __cos32
- ieee754: Remove slow paths from asin and acos
- getenv: Move call to strlen to the branch it's used in.
- Update kernel version to 5.10 in tst-mman-consts.py.
- s390x: Require GCC 7.1 or later to build glibc.
- malloc: Use __libc_initial to detect an inner libc
- Replace __libc_multiple_libcs with __libc_initial flag
- {nptl,htl}/semaphoreP.h: clean up
- htl: Get sem_open/sem_close/sem_unlink support [BZ #25524]
- pthread: Move semaphore initialization for open to semaphoreP.h
- Mark __libc_freeres_fn as used [BZ #27002]
- Update syscall lists for Linux 5.10.
- htl: Add pshared semaphore support
- hurd: Add LLL_PRIVATE and LLL_SHARED
- hurd: Add __libc_open and __libc_close
- htl: Add futex-internal.h
- hurd: Add __lll_abstimed_wait_intr
- hurd: make lll_* take a variable instead of a ptr
- hurd: Rename LLL_INITIALIZER to LLL_LOCK_INITIALIZER
- Use Linux 5.10 in build-many-glibcs.py.
Note: valgrind requires either internal or external symbol tables so
it's skipped if we're disabling debug packages, as ld-linux.so is
normally stripped.
Upstream commit: 4d0985543f479a6f421d4d8a9e0d1dc71c9c2c53
- elf: Record libc.so link map when it is the main program (bug 20972)
- Use GMP 6.2.1 in build-many-glibcs.py.
- aarch64: remove the strlen_asimd symbol
- aarch64: fix static PIE start code for BTI [BZ #27068]
- elf: Fix failure handling in _dl_map_object_from_fd
- elf: inline lose for error handling
- Remove strtoimax, strtoumax, wcstoimax, wcstoumax inlines
- nsswitch: handle missing actions properly
- x86: Remove the default REP MOVSB threshold tunable value [BZ #27061]
- elf.h: Remove SHF_GNU_BUILD_NOTE.
- elf.h: fix spelling typos in comments
- Fix spelling and grammar in several comments
- malloc: Detect infinite-loop in _int_free when freeing tcache [BZ#27052]
- elf: Fix dl-load.c
- elf: Include libc.so.6 as main program in dependency sort (bug 20972)
- support: Add support_slibdir_prefix variable
- aarch64: Use mmap to add PROT_BTI instead of mprotect [BZ #26831]
- elf: Pass the fd to note processing
- elf: Move note processing after l_phdr is updated
- aarch64: align address for BTI protection [BZ #26988]
- aarch64: Fix missing BTI protection from dependencies [BZ #26926]
- Fix linknamespace errors in nss_database.c if build with -Os.
- treewide: fix incorrect spelling of indices in comments
- linux: Consolidate brk implementation
- elf: Include <sys/param.h> in cache.c
- s390x: Add glibc-hwcaps support
- elf: Fix run-time dependencies of tst-dlopen-fail-2
- Handle out-of-memory case in svc_tcp.c/svc_unix.c:rendezvous_request.
- elf: Fix incorrect comparison in sort_priorities_by_name
- S390: Derive float_t from FLT_EVAL_METHOD
- Fix parsing of /sys/devices/system/cpu/online (bug 25859)
- Make strtoimax, strtoumax, wcstoimax, wcstoumax into aliases
- Fixed typos in "NEWS for version 2.32"
- Add NEWS entry for CVE-2020-29562 (BZ #26923)
- iconv: Fix incorrect UCS4 inner loop bounds (BZ#26923)
- Drop glibc-rh1906066 and glibc-rh741105 patches fixed by sync.
Upstream commit: 088e9625378f25607acff3daf7a79cbdee497043
- x86: Rename readelflib.c
- nsswitch: use new internal API (callers)
- nsswitch: user new internal API (tests)
- nsswitch: use new internal API (core)
- nss: Implement <nss_database.h>
- <nss_action.h>: New abstraction for combining NSS modules and NSS actions
- nss: Introduce <nss_module.h>
- Add scripts/move-symbol-to-libc.py
Upstream commit: 4c38c1a229bc3628269ad98bd7e8d31d118d91f6
- Drop glibc-revert-fxstat-compat.patch; applied upstream.
- Drop glibc-revert-mknod-compat.patch; applied upstream.
- powerpc64le: Add glibc-hwcaps support
- x86: Adjust tst-cpu-features-supports.c for GCC 11
- x86: Set RDRAND usable if CPU supports RDRAND
- elf: Add missing <stddef.h> header to elf/dl-hwcaps.h
- lowlevellock-futex: Remove not used macros
- futex: Remove not used futex_reltimed_wait{_cancelable}
- y2038: Convert gai_suspend to support 64 bit time
- symbols: Add defines for libanl's libanl_hidden_{def|proto}
- x86_64: Add glibc-hwcaps support
- elf: Add glibc-hwcaps subdirectory support to ld.so cache processing
- elf: Process glibc-hwcaps subdirectories in ldconfig
- elf: Implement tail merging of strings in ldconfig
- elf: Implement a string table for ldconfig, with tail merging
- elf: Add extension mechanism to ld.so.cache
- elf: Add endianness markup to ld.so.cache (bug 27008)
- elf: Add glibc-hwcaps support for LD_LIBRARY_PATH
- elf: Synchronize <elf.h> section header flags with binutils
- x86: Fix THREAD_SELF definition to avoid ld.so crash (bug 27004)
- htl: Add hidden def for __pthread_create/detach
- manual: Clarify File Access Modes section and add O_PATH
- htl: Add missing symbols
- Revert "linux: Move xmknod{at} to compat symbols"
- Revert "linux: Move {f}xstat{at} to compat symbols"
- elf.h: Fix spelling of EM_TILE64 comment
- nptl: Fix __futex_clocklock64 return error check [BZ #26964]
- powerpc64le: ifunc select *f128 routines in multiarch mode
- y2038: Convert aio_suspend to support 64 bit time
- Fix typo in NEWS file
- nptl: Add EOVERFLOW checks for futex calls
- nptl: Fix PTHREAD_PRIO_PROTECT timed lock
- sh: Add sh4 fpu Implies folder
- io: nftw/ftw: Fix stack overflow with large nopenfd [BZ #26353]
- elf: Introduce enum opt_format in the ldconfig implementation
- support: Add support_copy_file
- NEWS entry for commit b4f020c9b408fb3d1d3d4901c4a71839145f8791
- timezone: Change zdump installation to bin directory
- nptl: Return EINVAL for invalid clock for pthread_clockjoin_np
- nptl: Return EINVAL for pthread_mutex_clocklock/PI with CLOCK_MONOTONIC [BZ #26801]
- nptl: Replace lll_futex_wake with futex-internal.h
- nptl: Replace lll_futex_supported_clockid with futex-internal.h
- nptl: Replace lll_futex_{timed_}wait by futex-internal.h
- nptl: Replace lll_timedwait with __futex_abstimed_wait64
- nptl: Replace __futex_clocklock_wait64 with __futex_abstimed_wait64
- nptl: Remove _futex_clock_wait_bitset64
- nptl: Consolidate __futex_abstimed_wait_{cancelable}64
- nptl: Extend __futex_abstimed_wait_cancelable64 comment
- nptl: Remove clockwait_tid
- nptl: Remove futex_wait_cancelable
- nptl: Remove unused internal futex functions
- Mark mtrace tests UNSUPPORTED if bug-ga2.mtrace or tst-leaks2.mtrace are missing
- elf: Fix uninitialized variable for _dl_write
- powerpc: Make PT_THREAD_POINTER available to assembly code
- Use libnss_files.so for tests posix/bug-ga2 and resolv/tst-leaks2 [BZ #26821]
- hurd report-wait: Fix stpcpy usage
- hurd S_msg_report_wait: Fix detecting fd ports
- hurd S_msg_report_wait: Fix reporting ports
- hurd: Fix strcpy calls
- hurd: Fix _S_msg_get/set_env_variable prototype
- hurd: Enable using ifunc
- Add {,sysdep-}ld-library-path make variable
- nptl: Move stack list variables into _rtld_global
- hurd: let _dl_argv and __libc_stack_end be relro
- hurd: Remove some remnants of cthreads
- nanosleep: Pass NULL when rem == NULL on ports with __TIMESIZE != 64
- y2038: Convert thrd_sleep to support 64 bit time
- y2038: Convert mtx_timedlock to support 64 bit time
- y2038: Convert cnd_timedwait to support 64 bit time
- hurd: Drop CLOCK_MONOTONIC change which slipped in
- hurd: make ptsname fail with ENOTTY on non-master-pty
- mach: Add missing assert.h include
- hurd: break relocation loop between libc.so and lib{mach,hurd}user.so
- Remove obsolete defines for HPUX support from fcntl.h and update O_NONBLOCK.
- Remove tls.h inclusion from internal errno.h
- nptl: Eliminate <smp.h> and __is_smp
- powerpc: Eliminate UP macro conditionals
- x86: Remove UP macro. Define LOCK_PREFIX unconditionally.
- alpha: Remove UP preprocessor conditionals
- hurd: Make sure signals get started
- hurd: initialize libpthread before starting the signal thread
- hurd: Make _hurd_libc_proc_init idempotent
- powerpc: Add optimized stpncpy for POWER9
- powerpc: Add optimized strncpy for POWER9
- Don't use nested function in test-ffs
- Use __builtin___stpncpy_chk when available
- tests: Remove NULL check for an array
- hurd: Move {,f,l}xstat{,at} and xmknod{at} to compat symbols
- hurd: Notify the proc server later during initialization
- htl: Initialize later
- htl: Keep thread signals blocked during its initialization
- htl: Fix spurious symbols in namespaces
- Use O_CLOEXEC in sysconf [BZ #26791]
- struct _Unwind_Exception alignment should not depend on compiler flags
- hurd: keep only required PLTs in ld.so
- hurd: Add missing startup calls
- riscv: Get cache information through sysconf
- RISC-V: Add _dl_start_user.
With the release of Fedora 33 the host systems affected by the
older systemd-nspawn have been updated and the workaround is no
longer required.
Resolve: #1869030
Upstream commit: e156dabc766d6f6f99ce9402999eae380a3ec1f2
- aarch64: Add variant PCS lazy binding test [BZ #26798]
- aarch64: Fix DT_AARCH64_VARIANT_PCS handling [BZ #26798]
- hurd: Correct 'ethenet' spelling
- Avoid -Wstringop-overflow warning in pthread_cleanup_push macros
- Disable spurious -Warray-bounds for ypclnt.c (bug 26687)
- Do not use array parameter to new_composite_name (bug 26726)
- Disable spurious -Wstringop-overflow for setjmp/longjmp (bug 26647)
- malloc debug: fix compile error when enable macro MALLOC_DEBUG > 1
- tst-tcfree2: adjust coding style.
- elf: In ldconfig, extract the new_sub_entry function from search_dir
- Use MPC 1.2.1 in build-many-glibcs.py.
- Argument Syntax: Use "option", @option, and @command.
- elf: Unify old and new format cache handling code in ld.so
- x86: Restore processing of cache size tunables in init_cacheinfo
- Make elf.h header self contained.
- x86: Optimizing memcpy for AMD Zen architecture.
- Hurd: Fix ftime build
- Add IP_RECVERR_RFC4884 and IPV6_RECVERR_RFC4884 from Linux 5.9.
- misc: Add internal __getauxval2 function
- Remove NEWS entry about ftime removal
- time: Add 64-bit time_t support for ftime
- Reinstate ftime and add deprecate message on ftime usage
- Update kernel version to 5.9 in tst-mman-consts.py.
- Amend grammar and add a description
- Fix typo in NEWS file
- Remove timing related checks of time/tst-cpuclock1
- Update syscall lists for Linux 5.9.
- Use Linux 5.9 in build-many-glibcs.py.
- Reword description of SXID_* tunable properties
- New benchtest: pthread locks
- y2038: nptl: Provide __futex_clock_wait_bitset64 to support 64 bit bitset
- C-SKY: Make dynamic linker's name compitable with the older gcc.
- Revert "C-SKY:Fix dynamic linker's name when mfloat-abi=softfp."
- Move vtimes to a compatibility symbol
- y2038: linux: Provide __time64 implementation
- rt: Fix typos in comments in <aio.h>
- C-SKY:Fix dynamic linker's name when mfloat-abi=softfp.
- Drop the glibc-revert-ftime-compat.patch.
malloc/mtrace.pl is a perl script which is used for testing for
leaks throughout the testsuite (tests ending in "-mem" typically).
The "Requires" entry is correctly autodetected.
Upstream commit: 0f09154c64005e78b61484ae87b5ea2028051ea0
- x86: Initialize CPU info via IFUNC relocation [BZ 26203]
- Add NEWS entry for ftime compatibility move
- support: Add create_temp_file_in_dir
- linux: Add __readdir_unlocked
- linux: Simplify opendir buffer allocation
- linux: Move posix dir implementations to Linux
- linux: Add 64-bit time_t support for wait3
- Move ftime to a compatibility symbol
- linux: Fix time64 support for futimesat
- linux: Use INTERNAL_SYSCALL on fstatat{64}
- shm tests: Append PID to names passed to shm_open [BZ #26737]
- sysvipc: Fix tst-sysvshm-linux on x32
- x86/CET: Update vfork to prevent child return
- resolv: Serialize processing in resolv/tst-resolv-txnid-collision
- statfs: add missing f_flags assignment
- y2038: Remove not used __fstatat_time64 define
- y2038: nptl: Convert pthread_mutex_{clock|timed}lock to support 64 bit
- sysvipc: Return EINVAL for invalid shmctl commands
- sysvipc: Fix IPC_INFO and SHM_INFO handling [BZ #26636]
- AArch64: Use __memcpy_simd on Neoverse N2/V1
- resolv: Handle transaction ID collisions in parallel queries (bug 26600)
- support: Provide a way to clear the RA bit in DNS server responses
- support: Provide a way to reorder responses within the DNS test server
- Add missing stat/mknod symbol on libc.abilist some ABIs
- manual: correct the spelling of "MALLOC_PERTURB_" [BZ #23015]
- manual: replace an obsolete collation example with a valid one
- rtld: fix typo in comment
- elf: Add missing <dl-procinfo.h> header to elf/dl-usage.c
- hurd: support clock_gettime(CLOCK_PROCESS/THREAD_CPUTIME_ID)
- linux: Move xmknod{at} to compat symbols
- linux: Add {f}stat{at} y2038 support
- linux: Move {f}xstat{at} to compat symbols
- linux: Disentangle fstatat from fxstatat
- linux: Implement {l}fstat{at} in terms of fstatat
- linux: Move the struct stat{64} to struct_stat.h
- Remove mknod wrapper functions, move them to symbols
- Remove stat wrapper functions, move them to exported symbols
- <sys/platform/x86.h>: Add FSRCS/FSRS/FZLRM support
- <sys/platform/x86.h>: Add Intel HRESET support
- <sys/platform/x86.h>: Add AVX-VNNI support
- <sys/platform/x86.h>: Add AVX512_FP16 support
- <sys/platform/x86.h>: Add Intel UINTR support
- elf: Do not pass GLRO(dl_platform), GLRO(dl_platformlen) to _dl_important_hwcaps
- elf: Enhance ld.so --help to print HWCAP subdirectories
- elf: Add library search path information to ld.so --help
- sunrpc: Adjust RPC function declarations to match Sun's (bug 26686]
- Avoid GCC 11 -Warray-parameter warnings [BZ #26686].
- elf: Make __rtld_env_path_list and __rtld_search_dirs global variables
- elf: Print the full name of the dynamic loader in the ld.so help message
- elf: Use the term "program interpreter" in the ld.so help message
- scripts/update-copyrights: Update csu/version.c, elf/dl-usage.c
- elf: Implement ld.so --version
- nptl: Add missing cancellation flags on lockf
- Update mips64 libm-test-ulps
- Update alpha libm-test-ulps
- elf: Implement ld.so --help
- elf: Record whether paths come from LD_LIBRARY_PATH or --library-path
- elf: Move ld.so error/help output to _dl_usage
- elf: Extract command-line/environment variables state from rtld.c
Store the locales list as a Lua table in the spec file. Add Lua
code and a new Python script, parse-SUPPORTED.py, to compute
a common representation from it.
Upstream commit: 72d36ffd7db55ae599f4c77feb0eae25a0f3714e
- Drop glibc-fix-float128-benchtests.patch; applied upstream.
- elf: Implement __rtld_malloc_is_complete
- __vfscanf_internal: fix aliasing violation (bug 26690)
- Revert "Fix missing redirects in testsuite targets"
- nptl: Add missing cancellation flags on futex_internal and pselect32
- elf: Implement _dl_write
- elf: Do not search HWCAP subdirectories in statically linked binaries
- Linux: Require properly configured /dev/pts for PTYs
- Linux: unlockpt needs to fail with EINVAL, not ENOTTY (bug 26053)
- login/tst-grantpt: Convert to support framework, more error checking
- posix: Fix -Warray-bounds instances building timer_create [BZ #26687]
- Replace Minumum/minumum with Minimum/minimum
- Optimize scripts/merge-test-results.sh
- Fix GCC 11 -Warray-parameter warning for __sigsetjmp (bug 26647)
- manual: Fix typo
- y2038: nptl: Convert pthread_rwlock_{clock|timed}{rd|wr}lock to support
64 bit time
- Y2038: nptl: Provide futex_abstimed_wait64 supporting 64 bit time
- sysvipc: Return EINVAL for invalid msgctl commands
- sysvipc: Fix IPC_INFO and MSG_INFO handling [BZ #26639]
- sysvipc: Return EINVAL for invalid semctl commands
- sysvipc: Fix SEM_STAT_ANY kernel argument pass [BZ #26637]
- aarch64: enforce >=64K guard size [BZ #26691]
- sysvipc: Fix semtimedop for Linux < 5.1 for 64-bit ABI
- nptl: futex: Move __NR_futex_time64 alias to beginning of futex-internal.h
- nptl: Provide proper spelling for 32 bit version of futex_abstimed_wait
- string: Fix strerrorname_np return value [BZ #26555]
- Set tunable value as well as min/max values
- ld.so: add an --argv0 option [BZ #16124]
- Reversing calculation of __x86_shared_non_temporal_threshold
- linux: Add time64 recvmmsg support
- linux: Add time64 support for nanosleep
- linux: Consolidate utimes
- linux: Use 64-bit time_t syscall on clock_getcputclockid
- linux: Add time64 sigtimedwait support
- linux: Add time64 select support
- nptl: Fix __futex_abstimed_wait_cancellable32
- sysvipc: Fix semtimeop for !__ASSUME_DIRECT_SYSVIPC_SYSCALLS
- hurd: add ST_RELATIME
- intl: Handle translation output codesets with suffixes [BZ #26383]
- bench-strcmp.c: Add workloads on page boundary
- bench-strncmp.c: Add workloads on page boundary
- strcmp: Add a testcase for page boundary
- strncmp: Add a testcase for page boundary [BZ #25933]
- Set locale related environment variables in debugglibc.sh
- benchtests: Run _Float128 tests only on architectures that support it
- powerpc: Protect dl_powerpc_cpu_features on INIT_ARCH() [BZ #26615]
- x86: Harden printf against non-normal long double values (bug 26649)
- x86: Use one ldbl2mpn.c file for both i386 and x86_64
- Define __THROW to noexcept for C++11 and later
Upstream commit: 86a912c8634f581ea42ec6973553dde7f058cfbf
- Update i686 ulps.
- Use LFS readdir in generic POSIX getcwd [BZ# 22899]
- linux: Remove __ASSUME_ATFCTS
- Sync getcwd with gnulib
- x86-64: Fix FMA4 detection in ifunc [BZ #26534]
- y2038: nptl: Convert pthread_cond_{clock|timed}wait to support 64 bit time
- malloc: Fix mallinfo deprecation declaration
- x32: Add <fixup-asm-unistd.h> and regenerate arch-syscall.h
- Add mallinfo2 function that support sizes >= 4GB.
- Remove obsolete default/nss code
- AArch64: Improve backwards memmove performance
- Add RISC-V 32-bit target to build-many-glibcs.py
- Documentation for the RISC-V 32-bit port
- RISC-V: Build infrastructure for 32-bit port
- RISC-V: Add rv32 path to RTLDLIST in ldd
- riscv32: Specify the arch_minimum_kernel as 5.4
- RISC-V: Fix llrint and llround missing exceptions on RV32
- RISC-V: Add the RV32 libm-test-ulps
- RISC-V: Add 32-bit ABI lists
- RISC-V: Add hard float support for 32-bit CPUs
- RISC-V: Support the 32-bit ABI implementation
- RISC-V: Add arch-syscall.h for RV32
- RISC-V: Add path of library directories for the 32-bit
- RISC-V: Support dynamic loader for the 32-bit
- RISC-V: Add support for 32-bit vDSO calls
- RISC-V: Use 64-bit-time syscall numbers with the 32-bit port
- RISC-V: Cleanup some of the sysdep.h code
- RISC-V: Use 64-bit time_t and off_t for RV32 and RV64
- io/lockf: Include bits/types.h before __OFF_T_MATCHES_OFF64_T check
- elf/tst-libc_dlvsym: Add a TEST_COMPAT around some symbol tests
- hurd: define BSD 4.3 ioctls only under __USE_MISC
- string: test strncasecmp and strncpy near page boundaries
- linux: Simplify utimensat
- linux: Simplify timerfd_settime
- linux: Simplify timer_gettime
- linux: Simplify sched_rr_get_interval
- linux: Simplify ppoll
- linux: Simplify mq_timedsend
- linux: Simplify mq_timedreceive
- linux: Simplify clock_settime
- linux: Simplify clock_nanosleep
- linux: Simplify clock_gettime
- linux: Simplify clock_adjtime
- linux: Add helper function to optimize 64-bit time_t fallback support
- S390: Sync HWCAP names with kernel by adding aliases [BZ #25971]
- [vcstocl] Import ProjectQuirks from its own file
- build-many-glibcs.py: Add a s390x -O3 glibc variant.
- Fix namespace violation in stdio.h and sys/stat.h if build with optimization. [BZ #26376]
- Add C2x BOOL_MAX and BOOL_WIDTH to limits.h.
- Use MPC 1.2.0 in build-many-glibcs.py.
- Add new STATX_* constants from Linux 5.8 to bits/statx-generic.h.
- Correct locking and cancellation cleanup in syslog functions (bug 26100)
In a discussion with Stephen Gallagher <sgallagh@redhat.com> we
identified that it has been 4 years since we transitioned to
language packs and that at this point we could switch the suggestion
from glibc-all-langpacks to glibc-minimal-langpack which would
prevent an installation from ever accidentally having the larger
installation with all language packs.
This does not yet fix bug 1380069 which is about the removal of
glibc language packs and the way in which the virtual provide is
used.
Upstream commit: cb7e7a5ca1d6d25d59bc038bdc09630e507c41e5
- nptl: Handle NULL abstime [BZ #26394]
- Update build-many-glibcs.py for binutils ia64 obsoletion.
- Update kernel version to 5.8 in tst-mman-consts.py.
- y2038: nptl: Convert pthread_{clock|timed}join_np to support 64 bit time
- aarch64: update ulps.
Upstream commit: 0be0845b7a674dbfb996f66cd03d675f0f6028dc
- S390: Regenerate ULPs.
- manual: Fix sigdescr_np and sigabbrev_np return type (BZ #26343)
- math: Update x86_64 ulps
- math: Regenerate auto-libm-test-out-j0
- manual: Put the istrerrorname_np and strerrordesc_np return type in braces
- Linux: Use faccessat2 to implement faccessat (bug 18683)
- manual: Fix strerrorname_np and strerrordesc_np return type (BZ #26343)
- math: Fix inaccuracy of j0f for x >= 2^127 when sin(x)+cos(x) is tiny
- Update syscall lists for Linux 5.8.
- Use Linux 5.8 in build-many-glibcs.py.
- htl: Enable tst-cancelx?[45]
- tst-cancel4: Make blocking on write more portable
- hurd: Add missing hidden def
- hurd: Rework sbrk
- hurd: Implement basic sched_get/setscheduler
- x86: Rename Intel CPU feature names
- manual: Fix some @code/@var formatting glitches chapter Date And Time
- Copy regex_internal.h from Gnulib
- Copy regex BITSET_WORD_BITS porting from Gnulib
- Sync regex.h from Gnulib
- Sync mktime.c from Gnulib
- Sync intprops.h from Gnulib
- Open master branch for glibc 2.33 development.
- Add glibc-deprecated-selinux-makedb.patch and
glibc-deprecated-selinux-nscd.patch to work around libselinux API
deprecations.
- Drop glibc-rseq-disable.patch; rseq support removed upstream.
Upstream commit: ec2f1fddf29053957d061dfe310f106388472a4f
- libio: Remove __libc_readline_unlocked
- shadow: Implement fgetspent_r using __nss_fgetent_r
- pwd: Implement fgetpwent_r using __nss_fgetent_r
- gshadow: Implement fgetsgent_r using __nss_fgetent_r (bug 20338)
- grp: Implement fgetgrent_r using __nss_fgetent_r
- nss: Add __nss_fgetent_r
- libio: Add fseterr_unlocked for internal use
- nss_files: Use generic result pointer in parse_line
- nss_files: Consolidate line parse declarations in <nss_files.h>
- nss_compat: Do not use mmap to read database files (bug 26258)
- nss_files: Consolidate file opening in __nss_files_fopen
- Update powerpc-nofpu libm-test-ulps.
- Use MPFR 4.1.0 in build-many-glibcs.py.
- elf: Change TLS static surplus default back to 1664
- hurd: Fix longjmp check for sigstate
- hurd: Fix longjmp early in initialization
- manual: New signal and errno string functions are AS-safe
- AArch64: Improve strlen_asimd performance (bug 25824)
- Move <rpc/netdb.h> from sunrpc to inet
- en_US: Minimize changes to date_fmt (Bug 25923)
- Linux: Remove rseq support
- manual: Use Unicode instead HTML entities for characters (bug 19737)
- Add NEWS entry for CVE-2020-6096 (bug 25620)
- arm: remove string/tst-memmove-overflow XFAIL
- AArch64: Rename IS_ARES to IS_NEOVERSE_N1
- AArch64: Add optimized Q-register memcpy
- AArch64: Align ENTRY to a cacheline
- Correct timespec implementation [BZ #26232]
- Remove --enable-obsolete-rpc configure flag
- hurd: Fix build-many-glibcs.py
- x86: Support usable check for all CPU features
- string: Make tst-strerror/tst-strsignal unsupported if msgfmt is not installed
- malloc: Deprecate more hook-related functionality
- elf: Support at least 32-byte alignment in static dlopen
- x86: Remove __ASSEMBLER__ check in init-arch.h
- x86: Remove the unused __x86_prefetchw
- Documentation for ARC port
- build-many-glibcs.py: Enable ARC builds
- ARC: Build Infrastructure
- ARC: ABI lists
- ARC: Linux Startup and Dynamic Loading
- ARC: Linux ABI
- ARC: Linux Syscall Interface
- ARC: hardware floating point support
- ARC: math soft float support
- ARC: Atomics and Locking primitives
- ARC: Thread Local Storage support
- ARC: startup and dynamic linking code
- ARC: ABI Implementation
- Fix time/tst-cpuclock1 intermitent failures
- powerpc64: Fix calls when r2 is not used [BZ #26173]
- Add NEWS entry for Update to Unicode 13.0.0 [BZ #25819]
- Update i686 libm-test-ulps
- Fix memory leak in __printf_fp_l (bug 26215).
- Fix double free in __printf_fp_l (bug 26214).
- linux: Fix syscall list generation instructions
- sysv: linux: Add 64-bit time_t variant for shmctl
- sysvipc: Remove the linux shm-pad.h file
- sysvipc: Split out linux struct shmid_ds
- sysv: linux: Add 64-bit time_t variant for msgctl
- sysvipc: Remove the linux msq-pad.h file
- sysvipc: Split out linux struct semid_ds
- sysv: linux: Add 64-bit time_t variant for semctl
Upstream commit: ffb17e7ba3a5ba9632cee97330b325072fbe41dd
- rtld: Avoid using up static TLS surplus for optimizations [BZ #25051]
- rtld: Account static TLS surplus for audit modules
- rtld: Add rtld.nns tunable for the number of supported namespaces
- Remove --enable-obsolete-nsl configure flag
- Move non-deprecated RPC-related functions from sunrpc to inet
- aarch64: add NEWS entry about branch protection support
- aarch64: redefine RETURN_ADDRESS to strip PAC
- aarch64: fix pac-ret support in _mcount
- aarch64: Add pac-ret support to assembly files
- aarch64: configure check for pac-ret code generation
- aarch64: ensure objects are BTI compatible
- aarch64: enable BTI at runtime
- aarch64: fix RTLD_START for BTI
- aarch64: fix swapcontext for BTI
- aarch64: Add BTI support to assembly files
- aarch64: Rename place holder .S files to .c
- aarch64: configure test for BTI support
- Rewrite abi-note.S in C.
- rtld: Clean up PT_NOTE and add PT_GNU_PROPERTY handling
- string: Move tst-strsignal tst-strerror to tests-container
- string: Fix prototype mismatch in sigabbrev_np, __sigdescr_np
- arm: CVE-2020-6096: Fix multiarch memcpy for negative length (#1820332)
- arm: CVE-2020-6096: fix memcpy and memmove for negative length (#1820332)
- sunrpc: Remove hidden aliases for global data symbols (bug 26210)
- hurd: Fix strerror not setting errno
- tst-strsignal: fix checking for RT signals support
- hurd: Evaluate fd before entering the critical section
- CVE-2016-10228: Rewrite iconv option parsing (#1428292)
- nss: Remove cryptographic key support from nss_files, nss_nis, nss_nisplus
- sunrpc: Do not export getrpcport by default
- sunrpc: Do not export key handling hooks by default
- sunrpc: Turn clnt_sperrno into a libc_hidden_nolink_sunrpc symbol
- string: Add strerrorname_np and strerrordesc_np
- string: Add sigabbrev_np and sigdescr_np
- string: Add strerror_l on test-strerror-errno
- string: Add strerror, strerror_r, and strerror_l test
- string: Add strsignal test
- string: Simplify strerror_r
- string: Use tls-internal on strerror_l
- string: Implement strerror in terms of strerror_l
- string: Remove old TLS usage on strsignal
- linux: Fix __NSIG_WORDS and add __NSIG_BYTES
- signal: Move sys_errlist to a compat symbol
- signal: Move sys_siglist to a compat symbol
- signal: Add signum-{generic,arch}.h
- Remove most vfprintf width/precision-dependent allocations (bug 14231, bug 26211).
- elf: Do not signal LA_ACT_CONSISTENT for an empty namespace [BZ #26076]
- Fix stringop-overflow errors from gcc 10 in iconv.
- x86: Add thresholds for "rep movsb/stosb" to tunables
- Use C2x return value from getpayload of non-NaN (bug 26073).
- x86: Detect Extended Feature Disable (XFD)
- x86: Correct bit_cpu_CLFSH [BZ #26208]
- manual: Document __libc_single_threaded
- Add the __libc_single_threaded variable
- Linux: rseq registration tests
- Linux: Use rseq in sched_getcpu if available
- Linux: Perform rseq registration at C startup and thread creation
- tst-cancel4: deal with ENOSYS errors
- manual: Show copyright information not just in the printed manual
