artin/llama2.c
1
0
Fork 0
mirror of https://github.com/trholding/llama2.c.git synced 2026-02-06 11:26:53 +00:00
Code Issues Packages Projects Releases Wiki Activity

runq.c - Disabled cblas matmul

Browse Source 
May need invasive rewrite for 8bit quant. Won't fix.
This commit is contained in:
Vulcan 2024-03-20 17:32:16 +05:30
parent 103d3b147a
commit d62525d980
1 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
runq.c
View File

@ -509,9 +509,9 @@ void matmul(float* xout, QuantizedTensor *x, QuantizedTensor *w, int n, int d) {
int i;
// L2E Addition
#ifdef BLAS // TODO: FIX INTQ8
/* #ifdef BLAS // TODO: FIX INTQ8
cblas_sgemv(CblasRowMajor, CblasNoTrans, d, n, 1.0f, w, n, x, 1, 0.0f, xout, 1);
#else
#else */
#ifdef ACCEL
ACCEL(i) // OMP/OACC Macro
#endif
@ -533,7 +533,7 @@ void matmul(float* xout, QuantizedTensor *x, QuantizedTensor *w, int n, int d) {
xout[i] = val;
}
// L2E Addition
#endif
/* #endif */
// END L2E Addition
}