floraachy
/
OpenBLAS
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Add shortcuts for (small) cases that do not need expensive buffer allocation

This commit is contained in:
Martin Kroeker 2021-05-29 22:28:00 +02:00 committed by GitHub
parent 734bd265a8
commit f84197c1a7
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 12 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
interface/trsv.c
View File

@ -188,6 +188,12 @@ void CNAME(enum CBLAS_ORDER order, enum CBLAS_UPLO Uplo,
if (n == 0) return; if (n == 0) return;
if (incx == 1 && trans == 0 && n < 50) {
buffer = NULL;
(trsv[(trans<<2) | (uplo<<1) | unit])(n, a, lda, x, incx, buffer);
return;
}
IDEBUG_START; IDEBUG_START;
FUNCTION_PROFILE_START(); FUNCTION_PROFILE_START();

6
interface/ztrsv.c
View File

@ -199,6 +199,12 @@ void CNAME(enum CBLAS_ORDER order, enum CBLAS_UPLO Uplo,
if (n == 0) return; if (n == 0) return;
if (incx == 1 && trans == 0 && n < 50) {
buffer = NULL;
(trsv[(trans<<2) | (uplo<<1) | unit])(n, a, lda, x, incx, buffer);
return;
}
IDEBUG_START; IDEBUG_START;
FUNCTION_PROFILE_START(); FUNCTION_PROFILE_START();