Skip to content

Fix and reinstate the Cooper Lake/Sapphire Rapids microkernel for non-transpose SBGEMV #5237

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
martin-frbg merged 2 commits into from
Apr 22, 2025
Merged

Fix and reinstate the Cooper Lake/Sapphire Rapids microkernel for non-transpose SBGEMV #5237

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
96d8080
Reinstate the CooperLake microkernel
martin-frbg Apr 21, 2025
99d9f1f
Fix conditional
martin-frbg Apr 21, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 3 additions & 3 deletions kernel/x86_64/sbgemv_n.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -28,9 +28,9 @@ USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

#include "common.h"

//#if defined (COOPERLAKE) || defined (SAPPHIRERAPIDS)
//#include "sbgemv_n_microk_cooperlake.c"
//#endif
#if defined (COOPERLAKE) || defined (SAPPHIRERAPIDS)
#include "sbgemv_n_microk_cooperlake.c"
#endif

#define ALIGN64_ALLOC(alloc_size, TYPE, ptr_align, ptr) \
ptr = (TYPE *) malloc(sizeof(TYPE)*alloc_size + 63); \
Expand Down
2 changes: 1 addition & 1 deletion kernel/x86_64/sbgemv_n_microk_cooperlake_template.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -231,7 +231,7 @@ static int sbgemv_kernel_32xN_lda_direct(BLASLONG m, BLASLONG n, float alpha, bf
accum512_8 = _mm512_permutex2var_ps(accum512_0, idx_base_0, accum512_1);
accum512_9 = _mm512_permutex2var_ps(accum512_0, idx_base_1, accum512_1);

if ((m-tag_m_32x) > 16) {
if ((m-tag_m_32x) >= 16) {
STORE16_COMPLETE_RESULT(accum512_8, y+tag_m_32x+0)
STORE16_MASK_COMPLETE_RESULT(accum512_9, y+tag_m_32x+16, store_tail_mask)
} else {
Expand Down
Loading