zzh
/
fast-bwa
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Make _mm_load_si128() explicit 

The previous code implicitly caused a load; change it so the load
intrinsic is explicitly invoked, as the others are. (This in fact
makes no difference to the generated code.)
This commit is contained in:
John Marshall 2022-06-17 18:42:07 +01:00
parent 0028f086d8
commit b5f4bdae91
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
ksw.c
View File

@ -267,7 +267,7 @@ kswr_t ksw_i16(kswq_t *q, int tlen, const uint8_t *target, int _o_del, int _e_de
h = _mm_load_si128(H0 + slen - 1); // h={2,5,8,11,14,17,-1,-1} in the above example h = _mm_load_si128(H0 + slen - 1); // h={2,5,8,11,14,17,-1,-1} in the above example
h = _mm_slli_si128(h, 2); h = _mm_slli_si128(h, 2);
for (j = 0; LIKELY(j < slen); ++j) { for (j = 0; LIKELY(j < slen); ++j) {
h = _mm_adds_epi16(h, *S++); h = _mm_adds_epi16(h, _mm_load_si128(S++));
e = _mm_load_si128(E + j); e = _mm_load_si128(E + j);
h = _mm_max_epi16(h, e); h = _mm_max_epi16(h, e);
h = _mm_max_epi16(h, f); h = _mm_max_epi16(h, f);