glibc/sysdeps/powerpc
Pedro Franco de Carvalho 813c6ec808 powerpc: optimize strcpy/stpcpy for POWER9/10
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>
2021-07-01 17:58:53 -03:00
..
bits Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
fpu powerpc: Update libm test ulps 2021-04-09 17:41:22 -03:00
nofpu Update powerpc-nofpu libm-test-ulps. 2021-01-18 20:21:07 +00:00
nptl nptl: Move pthread_spin_trylock into libc 2021-04-23 17:06:48 +02:00
power4 Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
power6 Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
powerpc32 Implement <unwind-link.h> for dynamically loading the libgcc_s unwinder 2021-03-01 15:58:01 +01:00
powerpc64 powerpc: optimize strcpy/stpcpy for POWER9/10 2021-07-01 17:58:53 -03:00
sys/platform Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
abort-instr.h
atomic-machine.h Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
cpu-features.c Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
cpu-features.h Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
dl-procinfo.c Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
dl-procinfo.h Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
dl-tls.c Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
dl-tls.h Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
dl-tunables.list Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
elf-initfini.h Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
ffs.c Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
fpu_control.h Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
gccframe.h Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
hwcapinfo.c Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
hwcapinfo.h Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
ifunc-sel.h Fix powerpc ifunc-sel.h build for -Os. 2018-03-02 22:27:56 +00:00
jmpbuf-offsets.h Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
jmpbuf-unwind.h Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
ldsodefs.h Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
libc-tls.c Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
locale-defines.sym powerpc: strcasestr optmization for power8 2016-04-22 19:23:13 +05:30
longjmp.c nptl: Move __pthread_unwind_next into libc 2021-04-21 19:49:50 +02:00
machine-gmon.h Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
Makefile Remove stale references to libdl.a 2021-06-09 19:14:02 +02:00
math-tests-snan-cast.h Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
memusage.h Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
mod-cache-ppc.c Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
mod-tlsopt-powerpc.c tst-tlsopt-powerpc as a shared lib 2017-08-03 15:39:21 +09:30
mp_clz_tab.c
novmx-longjmp.c Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
novmx-sigjmp.c Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
novmxsetjmp.h Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
preconfigure powerpc: Add support for POWER10 2020-06-29 10:08:38 -03:00
preconfigure.ac powerpc: Add support for POWER10 2020-06-29 10:08:38 -03:00
rtld-global-offsets.sym powerpc: Move cache line size to rtld_global_ro 2020-01-17 09:05:03 -03:00
sigjmp.c Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
sotruss-lib.c Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
stackinfo.h Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
sysdep.h Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
test-arith.c Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
test-arithf.c
test-get_hwcap-static.c Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
test-get_hwcap.c Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
test-gettimebase.c Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
tls-macros.h
tst-cache-ppc-static-dlopen.c Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
tst-cache-ppc-static.c Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
tst-cache-ppc.c Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
tst-set_ppr.c Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
tst-stack-align.h Properly check stack alignment [BZ #27901] 2021-05-24 07:42:12 -07:00
tst-tlsifunc-static.c Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
tst-tlsifunc.c Update copyright dates with scripts/update-copyrights 2021-01-02 12:17:34 -08:00
tst-tlsopt-powerpc.c tst-tlsopt-powerpc as a shared lib 2017-08-03 15:39:21 +09:30
Versions nptl: Move __pthread_unwind_next into libc 2021-04-21 19:49:50 +02:00