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

Work around Clang failing to disambiguate SVE intrinsics and add AppleClang crossbuild to MacOS/arm64 DYNAMIC_ARCH in AzureCI (#4140)

Browse Source 
* Add AppleClang crossbuild to MacOS/arm64 DYNAMIC_ARCH

* add casts to disambiguate svwhilelt for clang
This commit is contained in:
Martin Kroeker
2023-07-14 11:06:48 +02:00
committed by GitHub
parent affeef0b9c
commit 3d31191b0f
9 changed files with 37 additions and 24 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
kernel/arm64/dot_kernel_sve.c
View File

@@ -50,8 +50,8 @@ static FLOAT dot_kernel_sve(BLASLONG n, FLOAT *x, FLOAT *y) {
BLASLONG sve_width = SVE_WIDTH;
for (BLASLONG i = 0; i < n; i += sve_width * 2) {
svbool_t pg_a = SVE_WHILELT(i, n);
svbool_t pg_b = SVE_WHILELT(i + sve_width, n);
svbool_t pg_a = SVE_WHILELT((uint64_t)i, (uint64_t)n);
svbool_t pg_b = SVE_WHILELT((uint64_t)(i + sve_width), (uint64_t)n);
SVE_TYPE x_vec_a = svld1(pg_a, &x[i]);
SVE_TYPE y_vec_a = svld1(pg_a, &y[i]);