about summary refs log tree commit diff
Commit message (Collapse)AuthorAgeFilesLines
* nsswitch: use new internal API (callers)DJ Delorie2020-12-0426-262/+88
| | | | | | Stitch new ABI and types throughout all NSS callers. Reviewed-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
* nsswitch: user new internal API (tests)DJ Delorie2020-12-047-1/+524
| | | | | | Testsuite support and new test for new API. Reviewed-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
* nsswitch: use new internal API (core)DJ Delorie2020-12-048-845/+168
| | | | | | Core changes to switch the NSS internals to use the new API. Reviewed-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
* nss: Implement <nss_database.h>Florian Weimer2020-12-045-1/+539
| | | | | | | | | | | This code manages the mappings of the available databases in NSS (i.e. passwd, hosts, netgroup, etc) with the actions that should be taken to do a query on those databases. This is the main API between query functions scattered throughout glibc and the underlying code (actions, modules, etc). Reviewed-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
* <nss_action.h>: New abstraction for combining NSS modules and NSS actionsFlorian Weimer2020-12-044-1/+423
| | | | | | | | | | nss_action manages a set of lists of actions; these are the portions of the lines in nsswitch.conf to the right of the colons, like "dns [!UNAVAIL=return] files". Each permutation of actions and conditionals is cached for reuse, which limits memory growth, and refers to the static list of modules managed by nss_modules. Reviewed-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
* nss: Introduce <nss_module.h>Florian Weimer2020-12-045-61/+467
| | | | | | | | | | | | | | | | This provides the struct nss_module type, which combines the old struct service_library type with the known_function tree, by statically allocating space for all function pointers. struct nss_module is fairly large (536 bytes), but it will be shared across NSS databases. The old known_function handling had non-some per-function overhead (at least 32 bytes per looked-up function, but more for long function anmes), so overall, this is not too bad. Resolving all functions at load time simplifies locking, and the repeated lookups should be fast because the caches are hot at this point. Reviewed-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
* Add scripts/move-symbol-to-libc.pyFlorian Weimer2020-12-041-0/+217
| | | | | This helper script can be used to move symbols to libc.abilist across all architectures, while preserving their symbol version.
* powerpc64le: Add glibc-hwcaps supportFlorian Weimer2020-12-045-1/+135
| | | | | The "power10" and "power9" subdirectories are selected in a way that matches the -mcpu=power10 and -mcpu=power9 options of GCC.
* x86: Adjust tst-cpu-features-supports.c for GCC 11H.J. Lu2020-12-041-5/+10
| | | | | | Check HAS_CPU_FEATURE instead of CPU_FEATURE_USABLE for FSGSBASE, IBT, LM, SHSTK and XSAVES since FSGSBASE requires kernel support, IBT/SHSTK/LM require OS support and XSAVES is supervisor-mode only.
* x86: Set RDRAND usable if CPU supports RDRANDH.J. Lu2020-12-041-0/+1
| | | | Set RDRAND usable if CPU supports RDRAND.
* elf: Add missing <stddef.h> header to elf/dl-hwcaps.hMatheus Castanho2020-12-041-0/+1
| | | | | The lack of this header makes size_t unavailable on builds configured with --disable-tunables, causing compilation errors.
* lowlevellock-futex: Remove not used macrosLukasz Majewski2020-12-041-43/+0
| | | | | | | | | | | | Following macros: lll_futex_timed_lock_pi, lll_futex_clock_wait_bitset, lll_futex_wait_requeue_pi, lll_futex_timed_wait_requeue_pi are not used anymore so are eligible for removal. Build tests: ./src/scripts/build-many-glibcs.py glibcs Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org> Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
* futex: Remove not used futex_reltimed_wait{_cancelable}Lukasz Majewski2020-12-041-61/+0
| | | | | | | | | | | | | After gai_suspend and aio_suspend conversion to support 64 bit time and hence rewriting the code to use only absolute variants of futex wait functions (i.e. __futex_abstimed_wait64 and __futex_abstimed_wait_cancelable64) futex_reltimed_wait{_cancelable} are not needed anymore and can be removed. Build tests: ./src/scripts/build-many-glibcs.py glibcs Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org> Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
* y2038: Convert gai_suspend to support 64 bit timeLukasz Majewski2020-12-043-29/+50
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | This change uses (in gai_misc.h): - __futex_abstimed_wait64 (instead of futex_reltimed_wait) - __futex_abstimed_wait_cancellable64 (instead of futex_reltimed_wait_cancellable) from ./sysdeps/nptl/futex-helpers.h The gai_suspend() accepts relative timeout, which then is converted to absolute one. The i686-gnu port (HURD) do not define DONT_NEED_GAI_MISC_COND and as it doesn't (yet) support 64 bit time it uses not converted pthread_cond_timedwait(). The __gai_suspend() is supposed to be run on ports with __TIMESIZE !=64 and __WORDSIZE==32. It internally utilizes __gai_suspend_time64() and hence the conversion from 32 bit struct timespec to 64 bit one is required. For ports supporting 64 bit time the __gai_suspend_time64() will be used either via alias (to __gai_suspend when __TIMESIZE==64) or redirection (when -D_TIME_BITS=64 is passed). Build tests: ./src/scripts/build-many-glibcs.py glibcs Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org> Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
* symbols: Add defines for libanl's libanl_hidden_{def|proto}Lukasz Majewski2020-12-041-0/+8
| | | | | | | | | The __gai_suspend_time64, which supports 64 bit time on ports with __WORDSIZE == 32 && __TIMESIZE != 64, shall be exported from libanl (the same library from which original gai_suspend is exported). Up till now there were no defines for this library. This commit adds them.
* x86_64: Add glibc-hwcaps supportFlorian Weimer2020-12-045-1/+192
| | | | | | | | The subdirectories match those in the x86-64 psABI: https://gitlab.com/x86-psABIs/x86-64-ABI/-/commit/77566eb03bc6a326811cb7e9a6b9396884b67c7c Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
* elf: Add glibc-hwcaps subdirectory support to ld.so cache processingFlorian Weimer2020-12-0410-3/+508
| | | | | | | | | | | | | | This recognizes the DL_CACHE_HWCAP_EXTENSION flag in cache entries, and picks the supported cache entry with the highest priority. The elf/tst-glibc-hwcaps-prepend-cache test documents a non-desired aspect of the current cache implementation: If the cache selects a DSO that does not exist on disk, _dl_map_object falls back to open_path, which may or may not find an alternative implementation. This is an existing limitation that also applies to the legacy hwcaps processing for ld.so.cache. Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
* elf: Process glibc-hwcaps subdirectories in ldconfigFlorian Weimer2020-12-044-53/+427
| | | | | | | | Libraries from these subdirectories are added to the cache with a special hwcap bit DL_CACHE_HWCAP_EXTENSION, so that they are ignored by older dynamic loaders. Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
* elf: Implement tail merging of strings in ldconfigFlorian Weimer2020-12-042-39/+40
| | | | | | | | | | | | | This simplifies the string table construction in elf/cache.c because there is no more need to keep track of offsets explicitly; the string table implementation does this internally. This change slightly reduces the size of the cache on disk. The file format does not change as a result. The strings are null-terminated, without explicit length, so tail merging is transparent to readers. Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
* elf: Implement a string table for ldconfig, with tail mergingFlorian Weimer2020-12-045-1/+488
| | | | | | | | | | | | | | | | This will be used in ldconfig to reduce the ld.so.cache size slightly. Tail merging is an optimization where a pointer points into another string if the first string is a suffix of the second string. The hash function FNV-1a was chosen because it is simple and achieves good dispersion even for short strings (so that the hash table bucket count can be a power of two). It is clearly superior to the hsearch hash and the ELF hash in this regard. The hash table uses chaining for collision resolution. Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
* elf: Add extension mechanism to ld.so.cacheFlorian Weimer2020-12-042-1/+211
| | | | | | | | | | A previously unused new-format header field is used to record the address of an extension directory. This change adds a demo extension which records the version of ldconfig which builds a file. Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
* elf: Add endianness markup to ld.so.cache (bug 27008)Florian Weimer2020-12-043-2/+93
| | | | | | | | | Use a reserved byte in the new format cache header to indicate whether the file is in little endian or big endian format. Eventually, this information could be used to provide a unified cache for qemu-user and similiar scenarios. Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
* elf: Add glibc-hwcaps support for LD_LIBRARY_PATHFlorian Weimer2020-12-0416-33/+784
| | | | | | | | | | | | | | | | | | | | This hacks non-power-set processing into _dl_important_hwcaps. Once the legacy hwcaps handling goes away, the subdirectory handling needs to be reworked, but it is premature to do this while both approaches are still supported. ld.so supports two new arguments, --glibc-hwcaps-prepend and --glibc-hwcaps-mask. Each accepts a colon-separated list of glibc-hwcaps subdirectory names. The prepend option adds additional subdirectories that are searched first, in the specified order. The mask option restricts the automatically selected subdirectories to those listed in the option argument. For example, on systems where /usr/lib64 is on the library search path, --glibc-hwcaps-prepend=valgrind:debug causes the dynamic loader to search the directories /usr/lib64/glibc-hwcaps/valgrind and /usr/lib64/glibc-hwcaps/debug just before /usr/lib64 is searched. Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
* elf: Synchronize <elf.h> section header flags with binutilsFlorian Weimer2020-12-031-0/+2
| | | | | binutils 2.36 will add SHF_GNU_RETAIN support. SHF_GNU_BUILD_NOTE was also missing from the glibc header.
* x86: Fix THREAD_SELF definition to avoid ld.so crash (bug 27004)Jakub Jelinek2020-12-032-2/+12
| | | | | | | | | | | The previous definition of THREAD_SELF did not tell the compiler that %fs (or %gs) usage is invalid for the !DL_LOOKUP_GSCOPE_LOCK case in _dl_lookup_symbol_x. As a result, ld.so could try to use the TCB before it was initialized. As the comment in tls.h explains, asm volatile is undesirable here. Using the __seg_fs (or __seg_gs) namespace does not interfere with optimization, and expresses that THREAD_SELF is potentially trapping.
* htl: Add hidden def for __pthread_create/detachSamuel Thibault2020-12-033-0/+4
| | | | to avoid a PLT.
* manual: Clarify File Access Modes section and add O_PATHFlorian Weimer2020-12-031-28/+44
| | | | | | Kees Cook reported that the current text is misleading: <https://lore.kernel.org/lkml/202005150847.2B1ED8F81@keescook/>
* htl: Add missing symbolsSamuel Thibault2020-12-032-1/+3
| | | | | hurd/hurdsig.c needs to detect whether __pthread_detach and __pthread_create are available, so they need to be exposed.
* Revert "linux: Move xmknod{at} to compat symbols"Adhemerval Zanella2020-12-022-5/+0
| | | | | | | | | | | | | This reverts commit 81b83ff61f95f30ad53d6075247af0ea61a0b16e to move __xmknod{at} back to default symbols. ABIs with default symbol version of 2.33 or newer (such as riscv32) continue to just provide the mknod* symbols. The idea is to not force static libraries built against old glibc to update against new glibcs (since they reference the the xmknod{at} symbols). Checked on x86_64-linux-gnu and i686-linux-gnu.
* Revert "linux: Move {f}xstat{at} to compat symbols"Adhemerval Zanella2020-12-0220-133/+11
| | | | | | | | | | | | | This reverts commit 20b39d59467b0c1d858e89ded8b0cebe55e22f60 to move {f}xstat{at} back to default symbols. ABIs with default symbol version of 2.33 or newer (such as riscv32) continue to just provide the stat symbols. The idea is to not force static libraries built against old glibc to update against new glibcs (since they reference the old {f}xstat{at} symbols). Checked on x86_64-linux-gnu and i686-linux-gnu.
* elf.h: Fix spelling of EM_TILE64 commentFlorian Weimer2020-12-021-1/+1
| | | | Reported-By: Jakub Jelinek <jakub@redhat.com>
* nptl: Fix __futex_clocklock64 return error check [BZ #26964]Lucas A. M. Magalhaes2020-12-013-5/+142
| | | | | | | | | | | | | | | | | The earlier implementation of this, __lll_clocklock, calls lll_clockwait that doesn't return the futex syscall error codes. It always tries again if that fails. However in the current implementation, when the futex returns EAGAIN, __futex_clocklock64 will also return EGAIN, even if the futex is taken. This patch fixes the EAGAIN issue and also adds a check for EINTR. As futex syscall can return EINTR if the thread is interrupted by a signal. In this case I'm assuming the function should continue trying to lock as there is no mention to about it on POSIX. Also add a test for both scenarios. Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
* powerpc64le: ifunc select *f128 routines in multiarch modePaul E. Murphy2020-11-3016-197/+817
| | | | | | | | | | | | | | | | | | | | | | | | | Programatically generate simple wrappers for interesting libm *f128 objects. Selected functions are transcendental functions or those with trivial compiler builtins. This can result in a 2-3x speedup (e.g logf128 and expf128). A second set of implementation files are generated which include the first implementation encountered along the search path. This usually works, except when a wrapper is overriden and makefile search order slightly diverges from include order. Likewise, wrapper object files are created for each generated file. These hold the ifunc selection routines which export ABI. Next, several shared headers are intercepted to control renaming of asm function redirects are used first, and sometimes macro renames if the former is impractical. Notably, if the request machine supports hardware IEEE128 (i.e POWER9 and newer) this ifunc machinery is disabled. Likewise existing ifunc support for float128 is consolidated into this (e.g sqrtf128 and fmaf128). Reviewed-by: Tulio Magno Quites Machado Filho <tuliom@linux.ibm.com>
* y2038: Convert aio_suspend to support 64 bit timeLukasz Majewski2020-11-306-31/+57
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The aio_suspend function has been converted to support 64 bit time. This change uses (in aio_misc.h): - __futex_abstimed_wait64 (instead of futex_reltimed_wait) - __futex_abstimed_wait_cancellable64 (instead of futex_reltimed_wait_cancellable) from ./sysdeps/nptl/futex-helpers.h The aio_suspend() accepts relative timeout, which then is converted to absolute one. The i686-gnu port (HURD) do not define DONT_NEED_AIO_MISC_COND and as it doesn't (yet) support 64 bit time it uses not converted pthread_cond_timedwait(). The __aio_suspend() is supposed to be run on ports with __TIMESIZE !=64 and __WORDSIZE==32. It internally utilizes __aio_suspend_time64() and hence the conversion from 32 bit struct timespec to 64 bit one is required. For ports supporting 64 bit time the __aio_suspend_time64() will be used either via alias (to __aio_suspend when __TIMESIZE==64) or redirection (when -D_TIME_BITS=64 is passed). Build tests: ./src/scripts/build-many-glibcs.py glibcs Reviewed-by: Alistair Francis <alistair.francis@wdc.com> Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
* Fix typo in NEWS fileShuo Wang2020-11-301-3/+3
|
* nptl: Add EOVERFLOW checks for futex callsAdhemerval Zanella2020-11-274-13/+13
| | | | | | | | | | | | | | | | | | | | | Some futex-internal calls require additional check for EOVERFLOW (as indicated by [1] [2] [3]). For both mutex and rwlock code, EOVERFLOW is handle as ETIMEDOUT; since it indicate to the caller that the blocking operation could not be issued. For mutex it avoids a possible issue where PTHREAD_MUTEX_ROBUST_* might assume EOVERFLOW indicate futex has succeed, and for PTHREAD_MUTEX_PP_* it avoid a potential busy infinite loop. For rwlock and semaphores, it also avoids potential busy infinite loops. Checked on x86_64-linux-gnu and i686-linux-gnu, although EOVERFLOW won't be possible with current usage (since all timeouts on 32-bit architectures with 32-bit time_t support will be in the range of 32-bit time_t). [1] https://sourceware.org/pipermail/libc-alpha/2020-November/120079.html [2] https://sourceware.org/pipermail/libc-alpha/2020-November/120080.html [3] https://sourceware.org/pipermail/libc-alpha/2020-November/120127.html
* nptl: Fix PTHREAD_PRIO_PROTECT timed lockAdhemerval Zanella2020-11-276-27/+34
| | | | | | | | | | | | | | | | | The 878fe624d4 changed lll_futex_timed_wait, which expects a relative timeout, with a __futex_abstimed_wait64, which expects an absolute timeout. However the code still passes a relative timeout. Also, the PTHREAD_PRIO_PROTECT support for clocks different than CLOCK_REALTIME was broken since the inclusion of pthread_mutex_clocklock (9d20e22e46) since lll_futex_timed_wait always use CLOCK_REALTIME. This patch fixes by removing the relative time calculation. It also adds some xtests that tests both thread and inter-process usage. Checked on x86_64-linux-gnu.
* sh: Add sh4 fpu Implies folderAdhemerval Zanella2020-11-274-0/+4
| | | | | | | | | | | The commit 605f38177db (sh: Split BE/LE abilist) did not take in consideration the SH4 fpu support. Checked with a build for sh4-linux-gnu and manually checked that the implementations at sysdeps/sh/sh4/fpu/ are selected. John Paul Adrian Glaubitz also confirmed it fixes the build issues he encontered.
* io: nftw/ftw: Fix stack overflow with large nopenfd [BZ #26353]Xiaoming Ni2020-11-263-8/+81
| | | | | | | | | | | The nopenfd value is used as argument for the internal buffer on ftw_statup, which is allocated with alloca and might trigger a stack overflow for large values. This patch replaces the memory allocation to use malloc instead. Checked on x86_64-linux-gnu. Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
* elf: Introduce enum opt_format in the ldconfig implementationFlorian Weimer2020-11-263-18/+24
| | | | Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
* support: Add support_copy_fileFlorian Weimer2020-11-263-0/+49
| | | | Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
* NEWS entry for commit b4f020c9b408fb3d1d3d4901c4a71839145f8791Florian Weimer2020-11-251-0/+4
|
* timezone: Change zdump installation to bin directoryPrasanth R2020-11-251-1/+2
| | | | | | Commit 3bc4186c7a9, tzdata changed the zdump installation to bin directory, which creates duplicates of zdump in the Root filesystem.
* nptl: Return EINVAL for invalid clock for pthread_clockjoin_npAdhemerval Zanella2020-11-254-1/+92
| | | | | | | | | The align the GNU extension with the others one that accept specify which clock to wait for (such as pthread_mutex_clocklock). Check on x86_64-linux-gnu. Reviewed-by: Lukasz Majewski <lukma@denx.de>
* nptl: Return EINVAL for pthread_mutex_clocklock/PI with CLOCK_MONOTONIC [BZ ↵Adhemerval Zanella2020-11-255-1/+80
| | | | | | | | | | | | | | | | #26801] Linux futex FUTEX_LOCK_PI operation only supports CLOCK_REALTIME, so pthread_mutex_clocklock operation with priority aware mutexes may fail depending of the input timeout. Also, it is not possible to convert a CLOCK_MONOTONIC to a CLOCK_REALTIME due the possible wall clock time change which might invalid the requested timeout. Checked on x86_64-linux-gnu and i686-linux-gnu. Reviewed-by: Lukasz Majewski <lukma@denx.de>
* nptl: Replace lll_futex_wake with futex-internal.hAdhemerval Zanella2020-11-253-12/+7
| | | | | | | | | The idea is to make NPTL implementation to use on the functions provided by futex-internal.h. Checked on x86_64-linux-gnu and i686-linux-gnu. Reviewed-by: Lukasz Majewski <lukma@denx.de>
* nptl: Replace lll_futex_supported_clockid with futex-internal.hAdhemerval Zanella2020-11-251-1/+1
| | | | | | | | | The idea is to make NPTL implementation to use on the functions provided by futex-internal.h. Checked on x86_64-linux-gnu and i686-linux-gnu. Reviewed-by: Lukasz Majewski <lukma@denx.de>
* nptl: Replace lll_futex_{timed_}wait by futex-internal.hAdhemerval Zanella2020-11-254-12/+14
| | | | | | | | | The idea is to make NPTL implementation to use on the functions provided by futex-internal.h. Checked on x86_64-linux-gnu and i686-linux-gnu. Reviewed-by: Lukasz Majewski <lukma@denx.de>
* nptl: Replace lll_timedwait with __futex_abstimed_wait64Adhemerval Zanella2020-11-254-97/+2
| | | | | | Checked with x86_64-linux-gnu and i686-linux-gnu. Reviewed-by: Lukasz Majewski <lukma@denx.de>
* nptl: Replace __futex_clocklock_wait64 with __futex_abstimed_wait64Adhemerval Zanella2020-11-253-63/+5
| | | | | | | | | | | | | | | | | For non null timeouts, the __futex_clocklock_wait64 creates an a relative timeout by subtracting the current time from the input argument. The same behavior can be obtained with FUTEX_WAIT_BITSET without the need to calculate the relative timeout. Besides consolidate the code it also avoid the possible relative timeout issues [1]. The __futex_abstimed_wait64 needs also to return EINVAL syscall errors. Checked on x86_64-linux-gnu and i686-linux-gnu. [1] https://sourceware.org/pipermail/libc-alpha/2020-November/119881.html Reviewed-by: Lukasz Majewski <lukma@denx.de>