mirror of
git://sourceware.org/git/glibc.git
synced 2024-12-21 04:31:04 +08:00
71ae86478e
This patch adds an optimized memset implementation for POWER8. For sizes from 0 to 255 bytes, a word/doubleword algorithm similar to POWER7 optimized one is used. For size higher than 255 two strategies are used: 1. If the constant is different than 0, the memory is written with altivec vector instruction; 2. If constant is 0, dbcz instructions are used. The loop is unrolled to clear 512 byte at time. Using vector instructions increases throughput considerable, with a double performance for sizes larger than 1024. The dcbz loops unrolls also shows performance improvement, by doubling throughput for sizes larger than 8192 bytes. |
||
|---|---|---|
| .. | ||
| 970 | ||
| a2 | ||
| bits | ||
| cell | ||
| fpu | ||
| multiarch | ||
| power4 | ||
| power5 | ||
| power5+ | ||
| power6 | ||
| power6x | ||
| power7 | ||
| power8 | ||
| __longjmp-common.S | ||
| __longjmp.S | ||
| addmul_1.S | ||
| backtrace.c | ||
| bsd-_setjmp.S | ||
| bsd-setjmp.S | ||
| bzero.S | ||
| configure | ||
| configure.ac | ||
| crti.S | ||
| crtn.S | ||
| dl-dtprocnum.h | ||
| dl-irel.h | ||
| dl-machine.c | ||
| dl-machine.h | ||
| dl-trampoline.S | ||
| entry.h | ||
| ffsll.c | ||
| hp-timing.h | ||
| Implies | ||
| lshift.S | ||
| Makefile | ||
| memcpy.S | ||
| memset.S | ||
| mul_1.S | ||
| ppc-mcount.S | ||
| register-dump.h | ||
| rtld-memset.c | ||
| setjmp-common.S | ||
| setjmp.S | ||
| stackguard-macros.h | ||
| start.S | ||
| stpcpy.S | ||
| strchr.S | ||
| strcmp.S | ||
| strcpy.S | ||
| strlen.S | ||
| strncmp.S | ||
| submul_1.S | ||
| sysdep.h | ||
| tls-macros.h | ||
| tst-audit.h | ||