diff options
author | Adhemerval Zanella <adhemerval.zanella@linaro.org> | 2019-02-05 18:55:59 -0200 |
---|---|---|
committer | Adhemerval Zanella <adhemerval.zanella@linaro.org> | 2019-02-27 10:00:37 -0300 |
commit | 457208b1e9498c1238b7af4387529388df0414c1 (patch) | |
tree | d7310c4d50b115ceef62d4c7a1de7fff9b2657e9 /math/libm-test-j1.inc | |
parent | 30a7e2081c690dbb22022e1a7d276341b7391434 (diff) | |
download | glibc-457208b1e9498c1238b7af4387529388df0414c1.zip glibc-457208b1e9498c1238b7af4387529388df0414c1.tar.gz glibc-457208b1e9498c1238b7af4387529388df0414c1.tar.bz2 |
wcsmbs: optimize wcsnlen
This patch rewrites wcsnlen using wmemchr. The generic wmemchr
already uses the strategy (loop unrolling and tail handling) and
by using it it allows architectures that have optimized wmemchr
(s390 and x86_64) to optimize wcsnlen as well.
Checked on x86_64-linux-gnu.
* wcsmbs/wcsnlen.c (__wcsnlen): Rewrite using wmemchr.
Diffstat (limited to 'math/libm-test-j1.inc')
0 files changed, 0 insertions, 0 deletions