glibc/sysdeps, branch vdso

linux: Add support for getrandom vDSO

2024-09-27T18:25:36Z

Linux 6.11 has getrandom() in vDSO. It operates on a thread-local opaque state allocated with mmap using flags specified by the vDSO. Multiple states are allocated at once, as many as fit into a page, and these are held in an array of available states to be doled out to each thread upon first use, and recycled when a thread terminates. As these states run low, more are allocated. To make this procedure async-signal-safe, a simple guard is used in the LSB of the opaque state address, falling back to the syscall if there's reentrancy contention. Also, _Fork() is handled by blocking signals on opaque state allocation (so _Fork() always sees a consistent state even if it interrupts a getrandom() call) and by iterating over the thread stack cache on reclaim_stack. Each opaque state will be in the free states list (grnd_alloc.states) or allocated to a running thread. The cancellation is handled by always using GRND_NONBLOCK flags while calling the vDSO, and falling back to the cancellable syscall if the kernel returns EAGAIN (would block). Since getrandom is not defined by POSIX and cancellation is supported as an extension, the cancellation is handled as 'may occur' instead of 'shall occur' [1], meaning that if vDSO does not block (the expected behavior) getrandom will not act as a cancellation entrypoint. It avoids a pthread_testcancel call on the fast path (different than 'shall occur' functions, like sem_wait()). It is currently enabled for x86_64, which is available in Linux 6.11, and aarch64, powerpc32, powerpc64, loongarch64, and s390x, which are available in Linux 6.12. Link: https://pubs.opengroup.org/onlinepubs/9799919799/nframe.html [1] Co-developed-by: Jason A. Donenfeld Tested-by: Jason A. Donenfeld # x86_64 Tested-by: Adhemerval Zanella # x86_64, aarch64 Tested-by: Xi Ruoyao # x86_64, aarch64, loongarch64 Tested-by: Stefan Liebler # s390x

AArch64: Optimize memset

2024-09-09T14:30:00Z

Improve small memsets by avoiding branches and use overlapping stores. Use DC ZVA for copies over 128 bytes. Remove unnecessary code for ZVA sizes other than 64 and 128. Performance of random memset benchmark improves by 24% on Neoverse N1. Reviewed-by: Adhemerval Zanella

hppa: Update libm-test-ulps

2024-09-09T13:57:42Z

aarch64: Avoid redundant MOVs in AdvSIMD F32 logs

2024-09-09T12:03:49Z

Since the last operation is destructive, the first argument to the FMA also has to be the first argument to the special-case in order to avoid unnecessary MOVs. Reorder arguments and adjust special-case bounds to facilitate this. Reviewed-by: Wilco Dijkstra

LoongArch: Fix macro redefined warning in tls-desc.S

2024-09-06T07:46:13Z

Undef macro to avoid redefined warning.

i386: Update ulps

2024-09-05T20:25:55Z

As seen on an AMD Ryzen 9 7950X CPU when building with GCC 14 with SSE2 math.

powerpc64le: Build new strtod tests with long double ABI flags (bug 32145)

2024-09-05T20:02:23Z

This fixes several test failures: =====FAIL: stdlib/tst-strtod1i.out===== Locale tests all OK Locale tests all OK Locale tests strtold("1,5") returns -6,38643e+367 and not 1,5 strtold("1.5") returns 1,5 and not 1 strtold("1.500") returns 1 and not 1500 strtold("36.893.488.147.419.103.232") returns 1500 and not 3,68935e+19 Locale tests all OK =====FAIL: stdlib/tst-strtod3.out===== 0: got wrong results -2.5937e+4826, expected 0 =====FAIL: stdlib/tst-strtod4.out===== 0: got wrong results -6,38643e+367, expected 0 1: got wrong results 0, expected 1e+06 2: got wrong results 1e+06, expected 10 =====FAIL: stdlib/tst-strtod5i.out===== 0: got wrong results -6,38643e+367, expected 0 2: got wrong results 0, expected -0 4: got wrong results -0, expected 0 5: got wrong results 0, expected -0 6: got wrong results -0, expected 0 7: got wrong results 0, expected -0 8: got wrong results -0, expected 0 9: got wrong results 0, expected -0 10: got wrong results -0, expected 0 11: got wrong results 0, expected -0 12: got wrong results -0, expected 0 13: got wrong results 0, expected -0 14: got wrong results -0, expected 0 15: got wrong results 0, expected -0 16: got wrong results -0, expected 0 17: got wrong results 0, expected -0 18: got wrong results -0, expected 0 20: got wrong results 0, expected -0 22: got wrong results -0, expected 0 23: got wrong results 0, expected -0 24: got wrong results -0, expected 0 25: got wrong results 0, expected -0 26: got wrong results -0, expected 0 27: got wrong results 0, expected -0 Fixes commit 3fc063dee01da4f80920a14b7db637c8501d6fd4 ("Make __strtod_internal tests type-generic"). Suggested-by: Joseph Myers Reviewed-by: Carlos O'Donell

Linux: readdir_r needs to report getdents failures (bug 32124)

2024-09-05T10:05:32Z

Upon error, return the errno value set by the __getdents call in __readdir_unlocked. Previously, kernel-reported errors were ignored. Reviewed-by: DJ Delorie

i386: Update ulps

2024-09-05T07:57:25Z

As seen on an unspecified Intel system with glibc compiled with GCC 8.

linux: mips: Fix syscall_cancell build for __mips_isa_rev >= 6

2024-09-02T15:30:45Z

Use beqzc instead of bnel. Checked with a mipsisa64r6el-n64-linux-gnu build and some nptl cancellation tests on qemu.