x86: Fallback {str|wcs}cmp RTM in the ncmp overflow case [BZ #29127]

Re-cherry-pick commit c627209832 for strcmp-avx2.S change which was
omitted in intial cherry pick because at the time this bug was not
present on release branch.

Fixes BZ #29127.

In the overflow fallback strncmp-avx2-rtm and wcsncmp-avx2-rtm would
call strcmp-avx2 and wcscmp-avx2 respectively. This would have
not checks around vzeroupper and would trigger spurious
aborts. This commit fixes that.

test-strcmp, test-strncmp, test-wcscmp, and test-wcsncmp all pass on
AVX2 machines with and without RTM.

Co-authored-by: H.J. Lu <hjl.tools@gmail.com>
(cherry picked from commit c6272098323153db373f2986c67786ea8c85f1cf)
This commit is contained in:
Noah Goldstein 2022-02-15 08:18:15 -06:00 committed by Sunil K Pandey
parent 478cd506ea
commit ac87df8d75

View File

@ -345,10 +345,10 @@ L(one_or_less):
movq %LOCALE_REG, %rdx
# endif
jb L(ret_zero)
# ifdef USE_AS_WCSCMP
/* 'nbe' covers the case where length is negative (large
unsigned). */
jnbe __wcscmp_avx2
jnbe OVERFLOW_STRCMP
# ifdef USE_AS_WCSCMP
movl (%rdi), %edx
xorl %eax, %eax
cmpl (%rsi), %edx
@ -357,10 +357,6 @@ L(one_or_less):
negl %eax
orl $1, %eax
# else
/* 'nbe' covers the case where length is negative (large
unsigned). */
jnbe __strcmp_avx2
movzbl (%rdi), %eax
movzbl (%rsi), %ecx
TOLOWER_gpr (%rax, %eax)