about summary refs log tree commit diff
path: root/sysdeps
Commit message (Collapse)AuthorAgeFilesLines
* x86: Remove .tfloat usageAdhemerval Zanella2022-10-039-26/+47
| | | | | Some compiler does not support it (such as clang integrated assembler) neither gcc emits it.
* hppa: Fix initialization of dp register [BZ 29635]John David Anglin2022-10-011-5/+19
| | | | | | | | | | | | | After upgrading glibc to Debian 2.35-1, gdb faulted on startup and dropped core in a function call in the main application. This was caused by not initializing the global dp register for the main application early enough. Restore the code to initialize dp in _dl_start_user. It was removed when code was added to initialize dp in elf_machine_runtime_setup. Signed-off-by: John David Anglin <dave.anglin@bell.net>
* malloc: Do not clobber errno on __getrandom_nocancel (BZ #29624)Adhemerval Zanella2022-09-302-3/+11
| | | | | | | | | | Use INTERNAL_SYSCALL_CALL instead of INLINE_SYSCALL_CALL. This requires emulate the semantic for hurd call (so __arc4random_buf uses the fallback). Checked on x86_64-linux-gnu. Reviewed-by: Wilco Dijkstra <Wilco.Dijkstra@arm.com>
* stdlib: Fix __getrandom_nocancel type and arc4random usage (BZ #29638)Adhemerval Zanella2022-09-301-1/+1
| | | | | | | | | Using an unsigned type prevents the fallback to be used if kernel does not support getrandom syscall. Checked on x86_64-linux-gnu. Reviewed-by: Wilco Dijkstra <Wilco.Dijkstra@arm.com>
* LoongArch: Add static PIE supportXi Ruoyao2022-09-303-3/+95
| | | | | | | If the compiler is new enough, enable static PIE support. In the static PIE version of _start (in rcrt1.o), use la.pcrel instead of la.got because in a static PIE we cannot use GOT entries until the dynamic relocations for GOT are resolved.
* x86: Fix wcsnlen-avx2 page cross length comparison [BZ #29591]Noah Goldstein2022-09-281-5/+2
| | | | | | | | | | | Previous implementation was adjusting length (rsi) to match bytes (eax), but since there is no bound to length this can cause overflow. Fix is to just convert the byte-count (eax) to length by dividing by sizeof (wchar_t) before the comparison. Full check passes on x86-64 and build succeeds w/ and w/o multiarch.
* Update _FloatN header support for C++ in GCC 13Joseph Myers2022-09-285-15/+15
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | GCC 13 adds support for _FloatN and _FloatNx types in C++, so breaking the installed glibc headers that assume such support is not present. GCC mostly works around this with fixincludes, but that doesn't help for building glibc and its tests (glibc doesn't itself contain C++ code, but there's C++ code built for tests). Update glibc's bits/floatn-common.h and bits/floatn.h headers to handle the GCC 13 support directly. In general the changes match those made by fixincludes, though I think the ones in sysdeps/powerpc/bits/floatn.h, where the header tests __LDBL_MANT_DIG__ == 113 or uses #elif, wouldn't match the existing fixincludes patterns. Some places involving special C++ handling in relation to _FloatN support are not changed. There's no need to change the __HAVE_FLOATN_NOT_TYPEDEF definition (also in a form that wouldn't be matched by the fixincludes fixes) because it's only used in relation to macro definitions using features not supported for C++ (__builtin_types_compatible_p and _Generic). And there's no need to change the inline function overloads for issignaling, iszero and iscanonical in C++ because cases where types have the same format but are no longer compatible types are handled automatically by the C++ overload resolution rules. This patch also does not change the overload handling for iseqsig, and there I think changes *are* needed, beyond those in this patch or made by fixincludes. The way that overload is defined, via a template parameter to a structure type, requires overloads whenever the types are incompatible, even if they have the same format. So I think we need to add overloads with GCC 13 for every supported _FloatN and _FloatNx type, rather than just having one for _Float128 when it has a different ABI to long double as at present (but for older GCC, such overloads must not be defined for types that end up defined as typedefs for another type). Tested with build-many-glibcs.py: compilers build for aarch64-linux-gnu ia64-linux-gnu mips64-linux-gnu powerpc-linux-gnu powerpc64le-linux-gnu x86_64-linux-gnu; glibcs build for aarch64-linux-gnu ia64-linux-gnu i686-linux-gnu mips-linux-gnu mips64-linux-gnu-n32 powerpc-linux-gnu powerpc64le-linux-gnu x86_64-linux-gnu.
* hurd: Fix typoSamuel Thibault2022-09-281-1/+1
|
* get_nscd_addresses: Fix subscript typos [BZ #29605]Jörg Sonnenberger2022-09-281-3/+3
| | | | | | | | | Fix the subscript on air->family, which was accidentally set to COUNT when it should have remained as I. Resolves: BZ #29605 Reviewed-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
* hurd: Increase SOMAXCONN to 4096Samuel Thibault2022-09-271-1/+1
| | | | Notably fakeroot-tcp may introduce a lot of parallel connections.
* Use atomic_exchange_release/acquireWilco Dijkstra2022-09-2623-28/+28
| | | | | | | Rename atomic_exchange_rel/acq to use atomic_exchange_release/acquire since these map to the standard C11 atomic builtins. Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
* Use C11 atomics instead of atomic_decrement_and_testWilco Dijkstra2022-09-231-1/+1
| | | | | | | | | | Replace atomic_decrement_and_test with atomic_fetch_add_relaxed. These are simple counters which do not protect any shared data from concurrent accesses. Also remove the unused file cond-perf.c. Passes regress on AArch64. Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
* Use C11 atomics instead of atomic_increment(_val)Wilco Dijkstra2022-09-231-3/+3
| | | | | | | | | | | Replace atomic_increment and atomic_increment_val with atomic_fetch_add_relaxed. One case in sem_post.c uses release semantics (see comment above it). The others are simple counters and do not protect any shared data from concurrent accesses. Passes regress on AArch64. Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
* riscv: Remove RV32 floating point functionsAlistair Francis2022-09-218-132/+40
| | | | | | | We don't need RV32 specific floating point functions, instead make them generic for RISC-V. Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
* riscv: Consolidate the libm-test-ulpsAlistair Francis2022-09-214-1406/+0
| | | | | | | Both RV32 and RV64 should have the same libm-test-ulps, so consolidate them into a single file. Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
* hurd: Fix SIOCADD/DELRT ioctlsSamuel Thibault2022-09-211-2/+2
| | | | The hurd network stack uses struct ifrtreq rather than ortentry.
* hurd: Drop struct rtentry and in6_rtmsgSamuel Thibault2022-09-211-38/+0
| | | | These were cargo-culted, they are not used at all in Hurd interfaces.
* hurd: Add _IOT_ifrtreq to <net/route.h>Damien Zammit2022-09-211-0/+1
| | | | So that we can use struct ifrtreq in ioctls.
* hurd: Use IF_NAMESIZE rather than IFNAMSIZSamuel Thibault2022-09-211-1/+1
| | | | The latter is not available without __USE_MISC.
* hurd: Add ifrtreq structure to net/route.hDamien Zammit2022-09-211-0/+17
| | | | As used by the hurdish route ioctls.
* hppa: undef __ASSUME_SET_ROBUST_LISTJohn David Anglin2022-09-201-0/+3
| | | | | | | QEMU does not support support set_robust_list. Thus, we need to enable detection of set_robust_list system call. Signed-off-by: John David Anglin <dave.anglin@bell.net>
* linux: Use same type for MMAP2_PAGE_UNITAdhemerval Zanella2022-09-204-4/+4
| | | | | | | It avoid a possible compiler warning where right size of operator is converted from a negative value to unsigned. Reviewed-by: Florian Weimer <fweimer@redhat.com>
* m68k: Enforce 4-byte alignment on internal locks (BZ #29537)Adhemerval Zanella2022-09-204-2/+59
| | | | | | | | | A new internal definition, __LIBC_LOCK_ALIGNMENT, is used to force the 4-byte alignment only for m68k, other architecture keep the natural alignment of the type used internally (and hppa does not require 16-byte alignment for kernel-assisted CAS). Reviewed-by: Florian Weimer <fweimer@redhat.com>
* Linux: Do not skip d_ino == 0 entries in readdir, readdir64 (bug 12165)Florian Weimer2022-09-192-108/+69
| | | | | | | | POSIX does not say this value is special. For example, old XFS file systems may still use inode number zero. Also update the comment regarding ENOENT. Linux may return ENOENT for some file systems.
* hurd: Factorize at/non-at functionsSamuel Thibault2022-09-1726-226/+116
| | | | | | | | | | | | | | | | | | Non-at functions can be implemented by just calling the corresponding at function with AT_FDCWD and zero at_flags. In the linkat case, the at behavior is different (O_NOLINK), so this introduces __linkat_common to pass O_NOLINK as appropriate. lstat functions can also be implemented with fstatat by adding __fstatat64_common which takes a flags parameter in addition to the at_flags parameter, In the end this factorizes chmod, chown, link, lstat64, mkdir, readlink, rename, stat64, symlink, unlink, utimes. This also makes __lstat, __lxstat64, __stat and __xstat64 directly use __fstatat64_common instead of __lstat64 or __stat64.
* RISC-V: Allow long jumps to __syscall_errorŁukasz Stelmach2022-09-166-6/+6
| | | | | | | | | | __syscall_error may end up farther than 1MiB away from a caller, especially when linking statically large binaries. tail allows for 4GiB jumps and is reduced to j when a linked symbol is within range. Fixes: 36960f0c76 ("RISC-V: Linux Syscall Interface") Fixes: 7f33b09c65 ("RISC-V: Linux ABI") Signed-off-by: Łukasz Stelmach <l.stelmach@samsung.com>
* hurd: Make readlink* just reopen the file used for statSamuel Thibault2022-09-152-24/+40
| | | | | | | 9e5c991106cb ("hurd: Fix readlink() hanging on fifo") separated opening the file for the stat call from opening the file for the read call. That however opened a small window for the file to change. Better make this atomic by reopening the file with O_READ.
* hurd: Fix readlink() hanging on fifoSamuel Thibault2022-09-142-2/+8
| | | | | | | | readlink() opens the target with O_READ to be able to read the symlink content. When the target is actually a fifo, that would hang waiting for a writer (caught in the coreutils testsuite). We thus have to first lookup the target without O_READ to perform io_stat and lookout for fifos, and only after checking the symlink type, we can re-lookup with O_READ.
* Use relaxed atomics since there is no MO dependenceWilco Dijkstra2022-09-131-7/+1
| | | | | | | | Replace the 3 uses of atomic_bit_set and atomic_bit_test_set with atomic_fetch_or_relaxed. Using relaxed MO is correct since the atomics are used to ensure memory is released only once. Reviewed-by: Florian Weimer <fweimer@redhat.com>
* Use C11 atomics instead atomic_add(_zero)Wilco Dijkstra2022-09-091-3/+3
| | | | | | Replace atomic_add and atomic_add_zero with atomic_fetch_add_relaxed. Reviewed-by: DJ Delorie <dj@redhat.com>
* errlist: add missing entry for EDEADLOCK (bug 29545)Andreas Schwab2022-09-081-0/+3
| | | | | | | Some architectures (mips, powerpc and sparc) define separate values for EDEADLOCK and EDEADLK. Readd the errlist entry for EDEADLOCK for those configurations. Also use the dependency files from generating the auxiliary errlist and siglist files.
* Do not define static_assert or thread_local in headers for C2xJoseph Myers2022-09-071-1/+3
| | | | | | | | | | | | | | | | | | | | | C2x makes static_assert and thread_local into keywords, removing the definitions as macros in assert.h and threads.h. Thus, disable those macros in those glibc headers for C2x. The disabling is done based on a combination of language version and __GNUC_PREREQ, *not* based on __GLIBC_USE (ISOC2X), on the principle that users of the header (when requesting C11 or later APIs - not assert.h for C99 and older API versions) should always have the names static_assert or thread_local available after inclusion of the header, whether as a keyword or as a macro. Thus, when using a compiler without the keywords (whether an older compiler, possibly in C2x mode, or _GNU_SOURCE with any compiler but in an older language mode, for example) the macros should be defined, even when C2x APIs have been requested. The __GNUC_PREREQ conditionals here may well need updating with the versions of other compilers that gained support for these keywords in C2x mode. Tested for x86_64.
* elf: Rename _dl_sort_maps parameter from skip to force_firstFlorian Weimer2022-09-061-2/+4
| | | | | | | The new implementation will not be able to skip an arbitrary number of objects. Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
* math: x86: Use prefix for FP_INIT_ROUNDMODEAdhemerval Zanella2022-09-051-1/+7
| | | | | | | Not all compilers support the inline asm prefix '%v' to emit the avx instruction if AVX is enable. Use a prefix instead. Checked on x86_64-linux-gnu and i686-linux-gnu.
* LoongArch: Add soft float support.caiyinyu2022-09-0116-19/+164
|
* nptl: x86_64: Use same code for CURRENT_STACK_FRAME and stackinfo_get_spAdhemerval Zanella2022-08-312-2/+6
| | | | | | | | | | | | | It avoids the possible warning of uninitialized 'frame' variable when building with clang: ../sysdeps/nptl/jmp-unwind.c:27:42: error: variable 'frame' is uninitialized when used here [-Werror,-Wuninitialized] __pthread_cleanup_upto (env->__jmpbuf, CURRENT_STACK_FRAME); The resulting code is similar to CURRENT_STACK_FRAME. Checked on x86_64-linux-gnu.
* posix: Fix macro expansion producing 'defined' has undefined behaviorAdhemerval Zanella2022-08-301-3/+7
| | | | | | | | | | | | | | | | | The NEED_CHECK_SPEC is defined as: #define NEED_CHECK_SPEC \ (!defined _XBS5_ILP32_OFF32 || !defined _XBS5_ILP32_OFFBIG \ || !defined _XBS5_LP64_OFF64 || !defined _XBS5_LPBIG_OFFBIG \ || !defined _POSIX_V6_ILP32_OFF32 || !defined _POSIX_V6_ILP32_OFFBIG \ || !defined _POSIX_V6_LP64_OFF64 || !defined _POSIX_V6_LPBIG_OFFBIG \ || !defined _POSIX_V7_ILP32_OFF32 || !defined _POSIX_V7_ILP32_OFFBIG \ || !defined _POSIX_V7_LP64_OFF64 || !defined _POSIX_V7_LPBIG_OFFBIG) Which is undefined behavior accordingly to C Standard (Preprocessing directives, p4). Checked on x86_64-linux-gnu.
* S390: Always use svc 0Stefan Liebler2022-08-3013-66/+46
| | | | | | | | | | | | | | | | | | | On s390x syscalls are triggered by svc instruction. One can pass the syscall number encoded in the instruction "svc 123" or by storing it in r1: lghi r1,123 svc 0 If the syscall number is encoded in the instruction, this can cause broken syscall restarts. Therefore this patch is now just passing the syscall number in r1. See also kernel-commit: "s390/signal: switch to using vdso for sigreturn and syscall restart" https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/arch/s390/[%e2%80%a6]call.c?h=v6.0-rc1&id=df29a7440c4b5c65765c8f60396b3b13063e24e9 As information, the "svc 0" feature was introduced in kernel 2.5.62: commit b5aad611393ef2e132e3648fa4c6e56a9cfa8708
* LoongArch: Use __builtin_{fmax,fmaxf,fmin,fminf} with GCC >= 13Xi Ruoyao2022-08-302-0/+20
| | | | | | | | GCC 13 compiles these built-ins to {fmax,fmin}.{s/d} instruction, use them instead of the generic implementation. Link: https://gcc.gnu.org/r13-2085 Signed-off-by: Xi Ruoyao <xry111@xry111.site>
* LoongArch: Fix ptr mangling/demangling features.caiyinyu2022-08-303-19/+18
|
* Makeconfig: Set pie-ccflag to -fPIE by default [BZ# 29514]Richard Henderson2022-08-291-3/+0
| | | | | | | | | | | | We should default to the larger code model, in order to support larger applications built with -static -pie. This should be consistent with pic-ccflag, which defaults to -fPIC. Remove the now redundant override from sysdeps/sparc/Makefile. Note that -fno-pie and -fno-PIE have the same effect. Reviewed-by: Carlos O'Donell <carlos@redhat.com> Tested-by: Carlos O'Donell <carlos@redhat.com>
* hurd: Fix vm_size_t incoherenciesSamuel Thibault2022-08-296-7/+10
| | | | | | | In gnumach, 3e1702a65fb3 ("add rpc_versions for vm types") changed the type of vm_size_t, making it always a unsigned long. This made it incompatible on x86 with size_t. Even if we may want to revert it to unsigned int, it's better to fix the types of parameters according to the .defs files.
* mach: Make xpg_strerror_r set a message on errorSamuel Thibault2022-08-271-6/+10
| | | | | | | | | posix advises to have strerror_r fill a message even when we are returning an error. This makes mach's xpg_strerror_r do this, like the generic version does. Spotted by the libunistring testsuite test-strerror_r
* mach: Fix incoherency between perror and strerrorSamuel Thibault2022-08-271-1/+1
| | | | | | | | | | | | 08d2024b4167 ("string: Simplify strerror_r") inadvertently made __strerror_r print unknown error system in decimal while the original code was printing it in hexadecimal. perror was kept printing in hexadecimal in 725eeb4af14c ("string: Use tls-internal on strerror_l"), let us keep both coherent. This also fixes a duplicate ':' Spotted by the libunistring testsuite test-perror2
* csu: Change start code license to have link exceptionSzabolcs Nagy2022-08-264-10/+76
| | | | | | | | | | | | | | | The start code can get linked into dynamic linked executables where LGPL would require shipping the source or linkable binaries when the executable is distributed. On some targets the license exception was missing in start.S (which is compiled into crt1.o and Scrt1.o which may end up linked into PDE and PIE binaries). I did not review what other code may end up in executables, just fixed the start.S license inconsistency across targets. Reviewed-by: Carlos O'Donell <carlos@redhat.com>
* s390: Move hwcaps/platform names out of _rtld_global_roFlorian Weimer2022-08-255-64/+44
| | | | | | | | | | | Changes to these arrays are often backported to stable releases, but additions to these arrays shift the offsets of the following _rltd_global_ro members, thus breaking the GLIBC_PRIVATE ABI. Obviously, this change is itself an internal ABI break, but at least it will avoid further ABI breaks going forward. Reviewed-by: Carlos O'Donell <carlos@redhat.com>
* Revert "Detect ld.so and libc.so version inconsistency during startup"Florian Weimer2022-08-251-4/+0
| | | | | | | | | | | | | | | | This reverts commit 6f85dbf102ad7982409ba0fe96886caeb6389fef. Once this change hits the release branches, it will require relinking of all statically linked applications before static dlopen works again, for the majority of updates on release branches: The NEWS file is regularly updated with bug references, so the __libc_early_init suffix changes, and static dlopen cannot find the function anymore. While this ABI check is still technically correct (we do require rebuilding & relinking after glibc updates to keep static dlopen working), it is too drastic for stable release branches. Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
* Detect ld.so and libc.so version inconsistency during startupFlorian Weimer2022-08-241-0/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The files NEWS, include/link.h, and sysdeps/generic/ldsodefs.h contribute to the version fingerprint used for detection. The fingerprint can be further refined using the --with-extra-version-id configure argument. _dl_call_libc_early_init is replaced with _dl_lookup_libc_early_init. The new function is used store a pointer to libc.so's __libc_early_init function in the libc_map_early_init member of the ld.so namespace structure. This function pointer can then be called directly, so the separate invocation function is no longer needed. The versioned symbol lookup needs the symbol versioning data structures, so the initialization of libc_map and libc_map_early_init is now done from _dl_check_map_versions, after this information becomes available. (_dl_map_object_from_fd does not set this up in time, so the initialization code had to be moved from there.) This means that the separate initialization code can be removed from dl_main because _dl_check_map_versions covers all maps, including the initial executable loaded by the kernel. The lookup still happens before relocation and the invocation of IFUNC resolvers, so IFUNC resolvers are protected from ABI mismatch. The __libc_early_init function pointer is not protected because so little code runs between the pointer write and the invocation (only dynamic linker code and IFUNC resolvers). Reviewed-by: Carlos O'Donell <carlos@redhat.com>
* Merge _GL_UNUSED C23 patch from GnulibPaul Eggert2022-08-231-2/+2
| | | | | | | | | * posix/getopt.c (_getopt_initialize): * sysdeps/posix/tempname.c (try_dir, try_nocreate): Put _GL_UNUSED before args instead of after. This makes no difference for glibc. It is needed for Gnulib when being compiled on non-GCC C23 compilers.
* LoongArch: Fix dl-machine.h code formatting.Xi Ruoyao2022-08-241-8/+4
| | | | No functional change.
generated by cgit-pink 1.4.1 (git 2.36.1) at 2024-07-12 02:53:07 +0000