about summary refs log tree commit diff
Commit message (Collapse)AuthorAgeFilesLines
* manual: fix description for preadv()Armin Brauns2021-07-061-1/+1
|
* elf: Call free from base namespace on error in dl-libc.c [BZ #27646]Florian Weimer2021-07-061-1/+1
| | | | | | | | | | | | In dlerror_run, free corresponds to the local malloc in the namespace, but GLRO (dl_catch_error) uses the malloc from the base namespace. elf/tst-dlmopen-gethostbyname triggers this mismatch, but it does not crash, presumably because of a fastbin deallocation. Fixes commit c2059edce20c124d1a99f1a94cc52e83b77a917a ("elf: Use _dl_catch_error from base namespace in dl-libc.c [BZ #27646]") and commit b2964eb1d9a6b8ab1250e8a881cf406182da5875 ("dlfcn: Failures after dlmopen should not terminate process [BZ #24772]").
* linux: Check for null value msghdr struct before useKhem Raj2021-07-052-2/+4
| | | | | | | | This avoids crashes in libc when cmsg is null and refrencing msg structure when it is null Signed-off-by: Khem Raj <raj.khem@gmail.com> Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
* hooks.c: Remove incorrect commentSiddhesh Poyarekar2021-07-041-4/+0
| | | | | The comment about different values of glibc.malloc.check is no longer valid.
* mtrace: Add attribute nocommon to mallwatchTulio Magno Quites Machado Filho2021-07-021-1/+1
| | | | | | | | | | | Avoid compilation errors GCC versions that do not default to -fno-common, e.g. GCC <= 9. Fixes commit 00d28960c5388a582a0485e07629b553c32dde49 ("mtrace: Deprecate mallwatch and tr_break"). Suggested-by: Adhemerval Zanella <adhemerval.zanella@linaro.org> Suggested-by: Florian Weimer <fweimer@redhat.com>
* Move glibc.malloc.check implementation into its own fileSiddhesh Poyarekar2021-07-032-370/+391
| | | | | | | Separate the malloc check implementation from the malloc hooks. They still use the hooks but are now maintained in a separate file. Reviewed-by: Carlos O'Donell <carlos@redhat.com>
* mtrace: Deprecate mallwatch and tr_breakSiddhesh Poyarekar2021-07-032-37/+24
| | | | | | | | | | | | | The variable and function pair appear to provide a way for users to set conditional breakpoints in mtrace when a specific address is returned by the allocator. This can be achieved by using conditional breakpoints in gdb so it is redundant. There is no documentation of this interface in the manual either, so it appears to have been a hack that got added to debug malloc. Deprecate these symbols and do not call tr_break anymore. Reviewed-by: DJ Delorie <dj@redhat.com> Reviewed-by: Carlos O'Donell <carlos@redhat.com>
* Drop source dependencies on hooks.c and arena.cSiddhesh Poyarekar2021-07-031-4/+0
| | | | | | | | Dependencies on hooks.c and arena.c get auto-computed when generating malloc.o{,s}.d so there is no need to add them manually. Reviewed-by: Carlos O'Donell <carlos@redhat.com> Reviewed-by: Andreas Schwab <schwab@linux-m68k.org>
* malloc: Initiate tcache shutdown even without allocations [BZ #28028]JeffyChen2021-07-021-1/+2
| | | | | | | | | | After commit 1e26d35193efbb29239c710a4c46a64708643320 ("malloc: Fix tcache leak after thread destruction [BZ #22111]"), tcache_shutting_down is still not early enough. When we detach a thread with no tcache allocated, tcache_shutting_down would still be false. Reviewed-by: DJ Delorie <dj@redhat.com>
* Add mcheck tests to mallocSiddhesh Poyarekar2021-07-022-1/+49
| | | | | | | | | | | | | | | | | | | | | | | | | | | Like malloc-check, add generic rules to run all tests in malloc by linking with libmcheck.a so as to provide coverage for mcheck(). Currently the following 12 tests fail: FAIL: malloc/tst-malloc-backtrace-mcheck FAIL: malloc/tst-malloc-fork-deadlock-mcheck FAIL: malloc/tst-malloc-stats-cancellation-mcheck FAIL: malloc/tst-malloc-tcache-leak-mcheck FAIL: malloc/tst-malloc-thread-exit-mcheck FAIL: malloc/tst-malloc-thread-fail-mcheck FAIL: malloc/tst-malloc-usable-static-mcheck FAIL: malloc/tst-malloc-usable-static-tunables-mcheck FAIL: malloc/tst-malloc-usable-tunables-mcheck FAIL: malloc/tst-malloc_info-mcheck FAIL: malloc/tst-memalign-mcheck FAIL: malloc/tst-posix_memalign-mcheck and they have been added to tests-exclude-mcheck for now to keep status quo. At least the last two can be attributed to bugs in mcheck() but I haven't fixed them here since they should be fixed by removing malloc hooks. Others need to be triaged to check if they're due to mcheck bugs or due to actual bugs. Reviewed-by: DJ Delorie <dj@redhat.com>
* iconvconfig: Use the public feof_unlockedSiddhesh Poyarekar2021-07-021-1/+2
| | | | | | | | | Build of iconvconfig failed with CFLAGS=-Os since __feof_unlocked is not a public symbol. Replace with feof_unlocked (defined to __feof_unlocked when IS_IN (libc)) to fix this. Reported-by: Szabolcs Nagy <szabolcs.nagy@arm.com> Reviewed-by: Szabolcs Nagy <szabolcs.nagy@arm.com>
* resolv: Move libanl into libc (if libpthread is in libc)Florian Weimer2021-07-0278-206/+498
| | | | | | | | | | The symbols gai_cancel, gai_error, gai_suspend, getaddrinfo_a, __gai_suspend_time64 were moved using scripts/move-symbol-to-libc.py. For Hurd (which remains !PTHREAD_IN_LIBC), a few #define redirects had to be added because several pthread functions are not available under __. (Linux uses __ prefixes for most hidden aliases, and has to in some cases to avoid linknamespace issues.)
* powerpc: optimize strcpy/stpcpy for POWER9/10Pedro Franco de Carvalho2021-07-011-71/+89
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch modifies the current POWER9 implementation of strcpy and stpcpy to optimize it for POWER9/10. Since no new POWER10 instructions are used, the original POWER9 strcpy is modified instead of creating a new implementation for POWER10. This implementation is based on both the original POWER9 implementation of strcpy and the preamble of the new POWER10 implementation of strlen. The changes also affect stpcpy, which uses the same implementation with some additional code before returning. On POWER9, averaging improvements across the benchmark inputs (length/source alignment/destination alignment), for an experiment that ran the benchmark five times, bench-strcpy showed an improvement of 5.23%, and bench-stpcpy showed an improvement of 6.59%. On POWER10, bench-strcpy showed 13.16%, and bench-stpcpy showed 13.59%. The changes are: 1. Removed the null string optimization. Although this results in a few extra cycles for the null string, in combination with the second change, this resulted in improvements for for other cases. 2. Adapted the preamble from strlen for POWER10. This is the part of the function that handles up to the first 16 bytes of the string. 3. Increased number of unrolled iterations in the main loop to 6. Reviewed-by: Matheus Castanho <msc@linux.ibm.com> Tested-by: Matheus Castanho <msc@linux.ibm.com>
* soft-fp: Add __extendhfxf2 and __truncxfhf2H.J. Lu2021-07-012-0/+105
| | | | | | | | | | | 1. Add __extendhfxf2 to return an IEEE half converted to IEEE extended. 2. Add __truncxfhf2 to truncate IEEE extended into IEEE half. These are needed by x86 _Float16: https://software.intel.com/content/www/us/en/develop/download/intel-avx512-fp16-architecture-specification.html support in GCC.
* x86: Check RTM_ALWAYS_ABORT for RTM [BZ #28033]H.J. Lu2021-07-016-6/+14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | From https://www.intel.com/content/www/us/en/support/articles/000059422/processors.html * Intel TSX will be disabled by default. * The processor will force abort all Restricted Transactional Memory (RTM) transactions by default. * A new CPUID bit CPUID.07H.0H.EDX[11](RTM_ALWAYS_ABORT) will be enumerated, which is set to indicate to updated software that the loaded microcode is forcing RTM abort. * On processors that enumerate support for RTM, the CPUID enumeration bits for Intel TSX (CPUID.07H.0H.EBX[11] and CPUID.07H.0H.EBX[4]) continue to be set by default after microcode update. * Workloads that were benefited from Intel TSX might experience a change in performance. * System software may use a new bit in Model-Specific Register (MSR) 0x10F TSX_FORCE_ABORT[TSX_CPUID_CLEAR] functionality to clear the Hardware Lock Elision (HLE) and RTM bits to indicate to software that Intel TSX is disabled. 1. Add RTM_ALWAYS_ABORT to CPUID features. 2. Set RTM usable only if RTM_ALWAYS_ABORT isn't set. This skips the string/tst-memchr-rtm etc. testcases on the affected processors, which always fail after a microcde update. 3. Check RTM feature, instead of usability, against /proc/cpuinfo. This fixes BZ #28033.
* Update syscall lists for Linux 5.13Joseph Myers2021-07-0126-2/+80
| | | | | | | | | Linux 5.13 has three new syscalls (landlock_create_ruleset, landlock_add_rule, landlock_restrict_self). Update syscall-names.list and regenerate the arch-syscall.h headers with build-many-glibcs.py update-syscalls. Tested with build-many-glibcs.py.
* s390: Fix MEMCHR_Z900_G5 ifunc-variant if n>=0x80000000 [BZ #28024]Stefan Liebler2021-07-011-0/+13
| | | | | | | | | | | | | | | On s390 (31bit), the pointer to the first byte after s always wraps around with n >= 0x80000000 and can lead to stop searching before end of s. Thus this patch just use NULL as byte after s in this case and the srst instruction stops searching with "not found" when wrapping around from top address to zero. This is observable with testcase string/test-memchr starting with commit "String: Add overflow tests for strnlen, memchr, and strncat [BZ #27974]" https://sourceware.org/git/?p=glibc.git;a=commit;h=da5a6fba0febbfc90896ce1b2eb75c6d8a88a72d
* Fix extra PLT reference in libc.so due to __glob64_time64 if build with gcc ↵Stefan Liebler2021-07-013-0/+3
| | | | | | | | | | | | | | | | 7.5 on 32bit. Starting with recent commit 84f7ce84474c1648ce96884f1c91ca7b97ca3fc2 "posix: Add glob64 with 64-bit time_t support", elf/check-localplt fails due to extra PLT reference __glob64_time64 in __glob64_time64 itself. This is observable with gcc 7.5 on x86_64 with -m32 or s390x with -m31. E.g. if build with gcc 10, gcc is generating a call to __glob64_time64.localalias. This patch is adding a hidden version of __glob64_time64 in the same way as for __globfree64_time64.
* AArch64: Add hp-timing.hWilco Dijkstra2021-07-011-0/+47
| | | | | | | | | | Add hp-timing.h using the cntvct_el0 counter. Return timing in nanoseconds so it is fully compatible with generic hp-timing. Don't set HP_TIMING_INLINE in the dynamic linker since it adds unnecessary overheads and some ancient kernels may not handle emulating cntcvt correctly. Currently cntvct_el0 is only used for timing in the benchtests. Reviewed-by: Szabolcs Nagy <szabolcs.nagy@arm.com>
* AArch64: Improve strnlen performanceWilco Dijkstra2021-07-011-181/+89
| | | | | | | | Optimize strnlen by avoiding UMINV which is slow on most cores. On Neoverse N1 large strings are 1.8x faster than the current version, and bench-strnlen is 50% faster overall. This version is MTE compatible. Reviewed-by: Szabolcs Nagy <szabolcs.nagy@arm.com>
* Linux: Avoid calling malloc indirectly from __get_nprocsFlorian Weimer2021-06-301-26/+57
| | | | | | | | malloc initialization depends on __get_nprocs, so using scratch buffers in __get_nprocs may result in infinite recursion. Reviewed-by: Carlos O'Donell <carlos@redhat.com> Reviewed-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
* Use Linux 5.13 in build-many-glibcs.pyJoseph Myers2021-06-301-1/+1
| | | | | | | This patch makes build-many-glibcs.py use Linux 5.13. Tested with build-many-glibcs.py (host-libraries, compilers and glibcs builds).
* login: Move libutil into libcFlorian Weimer2021-06-3079-269/+640
| | | | | | | | | | | | | | | | The symbols forkpty, login, login_tty, logout, logwtmp, openpty were moved using scripts/move-symbol-to-libc.py. This is a single commit because most of the symbols are tied together via forkpty, for example. Several changes to use hidden prototypes are needed. This commit also updates pseudoterminal terminology on modified lines. For 390 (31-bit), this commit follows the existing style for the compat symbol version creation. Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
* login: Rework hidden prototypes for __setutent, __utmpname, __endutentFlorian Weimer2021-06-303-3/+9
| | | | | | | Replace attribute_hidden with a regular combination of libc_hidden_proto and libc_hidden_def. Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
* login: Hidden prototypes for _getpt, __ptsname_r, grantpt, unlockptFlorian Weimer2021-06-3012-2/+19
| | | | Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
* nptl_db: Re-use the ELF-to-abilist converter for ABI checkingFlorian Weimer2021-06-293-13/+13
| | | | | | | | | | The previous approach uses readelf -DWs, which does not produce a stable output format (older binutils versions do not include symbol version information). This commit re-uses scripts/abilist.awk with a tweak to include GLIBC_PRIVATE symbols. This awk script is based on objdump -T output, which appears to be stable over time. Reviewed-by: Szabolcs Nagy <szabolcs.nagy@arm.com>
* Add RFC 8335 Definitions from Linux 5.13Andreas Roeseler2021-06-292-0/+20
| | | | | | | | | | | | | | | | | | | | RFC 8335 defines the network utility PROBE, which builds off of the capabilities of Ping to query more detailed interface information from networking nodes. The definitions included in this patchset have been accepted into the linux net-next branch and will be included in Linux 5.13. This patchset adds the same definitions to the glibc for use in the iputils package. The relevant commits for the Linux definitions can be found here: https://github.com/torvalds/linux/commit/e542d29ca81d005651680a0a697b72ca13ddc4cc https://github.com/torvalds/linux/commit/750f4fc2a12f6632b5aa04526bf57fa06bfe8467 These changes have been tested by running the glibc tests on x86_64 Signed-off-by: Andreas Roeseler <andreas.a.roeseler@gmail.com> Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
* nss: Fix NSS_DECLARE_MODULE_FUNCTIONS handling of _nss_*_endnetgrentFlorian Weimer2021-06-291-1/+1
| | | | | The old version had an additional underscore, making the declaration ineffective.
* s390x: Update math: redirect roundeven functionStefan Liebler2021-06-293-1/+3
| | | | | | | | | | | | | After recent commit 447954a206837b5f153869cfeeeab44631c3fac9 "math: redirect roundeven function", building on s390x fails with: Error: symbol `__roundevenl' is already defined Similar to aarch64/riscv fix, this patch redirects target specific functions for s390x: commit 3213ed770cbc5821920d16caa93c85e92dd7b9f6 "Update math: redirect roundeven function"
* posix: Add _Fork [BZ #4737]Adhemerval Zanella2021-06-2842-17/+451
| | | | | | | | | | | | | | | | | | | | | | | | | Austin Group issue 62 [1] dropped the async-signal-safe requirement for fork and provided a async-signal-safe _Fork replacement that does not run the atfork handlers. It will be included in the next POSIX standard. It allow to close a long standing issue to make fork AS-safe (BZ#4737). As indicated on the bug, besides the internal lock for the atfork handlers itself; there is no guarantee that the handlers itself will not introduce more AS-safe issues. The idea is synchronize fork with the required internal locks to allow children in multithread processes to use mostly of standard function (even though POSIX states only AS-safe function should be used). On signal handles, _Fork should be used intead and only AS-safe functions should be used. For testing, the new tst-_Fork only check basic usage. I also added a new tst-mallocfork3 which uses the same strategy to check for deadlock of tst-mallocfork2 but using threads instead of subprocesses (and it does deadlock if it replaces _Fork with fork). [1] https://austingroupbugs.net/view.php?id=62
* nptl: Add glibc.pthread.stack_cache_size tunableFlorian Weimer2021-06-286-4/+32
| | | | | | | The valgrind/helgrind test suite needs a way to make stack dealloction more prompt, and this feature seems to be generally useful. Reviewed-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
* nptl: Export libthread_db-used symbols under GLIBC_PRIVATEFlorian Weimer2021-06-284-8/+71
| | | | | | | This allows distributions to strip debugging information from libc.so.6 without impacting the debugging experience. Reviewed-by: Carlos O'Donell <carlos@redhat.com>
* nptl: Rename nptl_version to __nptl_versionFlorian Weimer2021-06-282-2/+2
| | | | | | | This prepares it for exporting as a dynamic symbol. Reviewed-by: Carlos O'Donell <carlos@redhat.com> Tested-by: Carlos O'Donell <carlos@redhat.com>
* nptl_db: Clean up main/rtld variable handlingFlorian Weimer2021-06-282-10/+5
| | | | | | | | | | | Most symbols are now in libc.so.6. The "main" (exempted from coverage checks) status is therefore not necessary. Use DB_MAIN_VARIABLE for the remaining separate symbol, __nptl_initial_report_events. DB_RTLD_VARIABLE is now unused, so remove it. Reviewed-by: Carlos O'Donell <carlos@redhat.com> Tested-by: Carlos O'Donell <carlos@redhat.com>
* arm: align stack in clone [BZ 28020]Szabolcs Nagy2021-06-281-0/+2
| | | | | | | The arm PCS requires 8 byte aligned stack at function entry. Previously unaligned stack could crash the clone child. Fixes bug 28020.
* Linux: Cleanups after librt moveFlorian Weimer2021-06-2820-150/+44
| | | | | | | | | | | | librt.so is no longer installed for PTHREAD_IN_LIBC, and tests are not linked against it. $(librt) is introduced globally for shared tests that need to be linked for both PTHREAD_IN_LIBC and !PTHREAD_IN_LIBC. GLIBC_PRIVATE symbols that were needed during the transition are removed again. Reviewed-by: Carlos O'Donell <carlos@redhat.com>
* Linux: Move timer_settime, __timer_settime64 from librt to libcFlorian Weimer2021-06-2884-212/+245
| | | | | | | | | | | | | | | The symbols were moved using scripts/move-symbol-to-libc.py. The way the ABI intransition is implemented is changed with this commit: the implementation is now consolidated in one file with a TIMER_T_WAS_INT_COMPAT check. The shared librt is now empty, so this commit adds a placeholder symbol at the base version, GLIBC_2.2, and potentially at the GLIBC_2.3.3 version as well (the leftover from the int/timer_t ABI transition). Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
* Linux: Move timer_gettime, __timer_gettime64 from librt to libcFlorian Weimer2021-06-2879-119/+153
| | | | | | | | | | The symbols were moved using scripts/move-symbol-to-libc.py. The way the ABI intransition is implemented is changed with this commit: the implementation is now consolidated in one file with a TIMER_T_WAS_INT_COMPAT check. Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
* Linux: Move timer_getoverrun from librt to libcFlorian Weimer2021-06-2879-95/+107
| | | | | | | | | | The symbol was moved using scripts/move-symbol-to-libc.py. The way the ABI intransition is implemented is changed with this commit: the implementation is now consolidated in one file with a TIMER_T_WAS_INT_COMPAT check. Reviewed-by: Adhemerva Zanella <adhemerval.zanella@linaro.org>
* Linux: Move timer_create, timer_delete from librt to libcFlorian Weimer2021-06-2890-242/+276
| | | | | | | | | | | The symbols were moved using scripts/move-symbol-to-libc.py. timer_create and timer_delete are tied together via the int/timer_t compatibility code. The way the ABI intransition is implemented is changed with this commit: the implementation is now consolidated in one file with a TIMER_T_WAS_INT_COMPAT check. Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
* Linux: Define TIMER_T_WAS_INT_COMPAT in kernel-posix-timers.hFlorian Weimer2021-06-288-0/+138
| | | | | | | | | This is almost equivalent to __WORDSIZE == 64 && OTHER_SHLIB_COMPAT (librt, GLIBC_2_1, GLIBC_2_3_3), except that this expression is true for mips64/n64 targets as well, even though those did not undergo the timer_t transition. Reviewed-by: Carlos O'Donell <carlos@redhat.com>
* Install shared objects under their ABI namesFlorian Weimer2021-06-284-54/+15
| | | | | | | | | | | | | Previously, the installed objects were named like libc-2.33.so, and the ABI soname libc.so.6 was just a symbolic link. The Makefile targets to install these symbolic links are no longer needed after this, so they are removed with this commit. The more general $(make-link) command (which invokes scripts/rellns-sh) is retained because other symbolic links are still needed. Reviewed-by: Carlos O'Donell <carlos@redhat.com> Tested-by: Carlos O'Donell <carlos@rehdat.com>
* elf: Generalize name-based DSO recognition in ldconfigFlorian Weimer2021-06-284-4/+72
| | | | | | | | | This introduces <dl-is_dso.h> and the _dl_is_dso function. A test ensures that the official names of libc.so, ld.so, and their versioned names are recognized. Reviewed-by: Carlos O'Donell <carlos@redhat.com> Tested-by: Carlos O'Donell <carlos@redhat.com>
* Makerules: Remove lib-version, $(subdir-version)Florian Weimer2021-06-281-22/+4
| | | | | | | | Also clarify that the "versioned" term refers to the soname, not the glibc version (which also ends up in the installed file name). Reviewed-by: Carlos O'Donell <carlos@redhat.com> Tested-by: Carlos O'Donell <carlos@redhat.com>
* nptl_db: Install libthread_db under a regular implementation nameFlorian Weimer2021-06-281-2/+0
| | | | | | | | | | | Currently, the name is always libthread_db-1.0.so. It does not change with the glibc version, like the other libraries. GDB hard-codes libthread_db.so.1 (the soname), so this change does not affect loading libthread_db. Reviewed-by: Carlos O'Donell <carlos@redhat.com> Tested-by: Carlos O'Donell <carlos@redhat.com>
* iconvconfig: Fix multiple issuesSiddhesh Poyarekar2021-06-282-16/+10
| | | | | | | | | | | | | | | | | | | | | It was noticed on big-endian systems that msgfmt would fail with the following error: msgfmt: gconv_builtin.c:70: __gconv_get_builtin_trans: Assertion `cnt < sizeof (map) / sizeof (map[0])' failed. Aborted (core dumped) This is only seen on installed systems because it was due to a corrupted gconv-modules.cache. iconvconfig had the following issues (it was specifically freeing fulldir that caused this issue, but other cleanups are also needed) that this patch fixes. - Add prefix only if dir starts with '/' - Use asprintf instead of mempcpy so that the directory string is NULL terminated - Make a copy of the directory reference in new_module so that fulldir can be freed within the same scope in handle_dir. Reviewed-by: Florian Weimer <fweimer@redhat.com>
* wordexp: handle overflow in positional parameter number (bug 28011)Andreas Schwab2021-06-272-1/+2
| | | | Use strtoul instead of atoi so that overflow can be detected.
* Update math: redirect roundeven functionH.J. Lu2021-06-275-2/+5
| | | | | Redirect target specific roundeven functions for aarch64, ldbl-128ibm and riscv.
* Use GCC builtins for roundeven functions if desired.Shen-Ta Hsieh2021-06-276-0/+22
| | | | | | | | | | | | | This patch is using the corresponding GCC builtin for roundevenf, roundeven and roundevenl if the USE_FUNCTION_BUILTIN macros are defined to one in math-use-builtins.h. These builtin functions is supported since GCC 10. The code of the generic implementation is not changed. Signed-off-by: Shen-Ta Hsieh <ibmibmibm.tw@gmail.com> Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
* x86_64: roundeven with sse4.1 supportShen-Ta Hsieh2021-06-277-2/+118
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch adds support for the sse4.1 hardware floating point roundeven. Here is some benchmark results on my systems: =AMD Ryzen 9 3900X 12-Core Processor= * benchmark result before this commit | | roundeven | roundevenf | |------------|--------------|--------------| | duration | 3.75587e+09 | 3.75114e+09 | | iterations | 3.93053e+08 | 4.35402e+08 | | max | 52.592 | 58.71 | | min | 7.98 | 7.22 | | mean | 9.55563 | 8.61535 | * benchmark result after this commit | | roundeven | roundevenf | |------------|---------------|--------------| | duration | 3.73815e+09 | 3.73738e+09 | | iterations | 5.82692e+08 | 5.91498e+08 | | max | 56.468 | 51.642 | | min | 6.27 | 6.156 | | mean | 6.41532 | 6.3185 | =Intel(R) Pentium(R) CPU D1508 @ 2.20GHz= * benchmark result before this commit | | roundeven | roundevenf | |------------|--------------|--------------| | duration | 2.18208e+09 | 2.18258e+09 | | iterations | 2.39932e+08 | 2.46924e+08 | | max | 96.378 | 98.035 | | min | 6.776 | 5.94 | | mean | 9.09456 | 8.83907 | * benchmark result after this commit | | roundeven | roundevenf | |------------|--------------|--------------| | duration | 2.17415e+09 | 2.17005e+09 | | iterations | 3.56193e+08 | 4.09824e+08 | | max | 51.693 | 97.192 | | min | 5.926 | 5.093 | | mean | 6.10385 | 5.29507 | Signed-off-by: Shen-Ta Hsieh <ibmibmibm.tw@gmail.com> Reviewed-by: H.J. Lu <hjl.tools@gmail.com>
generated by cgit-pink 1.4.1 (git 2.36.1) at 2024-11-09 08:15:42 +0000