mirror of
https://github.com/amd/blis.git
synced 2026-05-21 00:48:56 +00:00
Arm SVE Typo Fix ZGEMM/CGEMM C Prefetch Reg
This commit is contained in:
@@ -116,23 +116,23 @@ GEMM_ACOLCMPLX_CONTIGUOUS_LOAD_FWD(z28,z29,p0,%0,x2)
|
||||
" b.ne END_CCOL_PRFM \n\t" // Do not prefetch for generic C storage.
|
||||
" mov x16, %2 \n\t"
|
||||
" prfm PLDL1KEEP, [x16] \n\t"
|
||||
" add x16, x16, %5 \n\t"
|
||||
" add x16, x16, %4 \n\t"
|
||||
" prfm PLDL1KEEP, [x16] \n\t"
|
||||
" add x16, x16, %5 \n\t"
|
||||
" add x16, x16, %4 \n\t"
|
||||
" prfm PLDL1KEEP, [x16] \n\t"
|
||||
" add x16, x16, %5 \n\t"
|
||||
" add x16, x16, %4 \n\t"
|
||||
" prfm PLDL1KEEP, [x16] \n\t"
|
||||
" add x16, x16, %5 \n\t"
|
||||
" add x16, x16, %4 \n\t"
|
||||
" prfm PLDL1KEEP, [x16] \n\t"
|
||||
" add x16, x16, %5 \n\t"
|
||||
" add x16, x16, %4 \n\t"
|
||||
" prfm PLDL1KEEP, [x16] \n\t"
|
||||
" add x16, x16, %5 \n\t"
|
||||
" add x16, x16, %4 \n\t"
|
||||
" prfm PLDL1KEEP, [x16] \n\t"
|
||||
" add x16, x16, %5 \n\t"
|
||||
" add x16, x16, %4 \n\t"
|
||||
" prfm PLDL1KEEP, [x16] \n\t"
|
||||
" add x16, x16, %5 \n\t"
|
||||
" add x16, x16, %4 \n\t"
|
||||
" prfm PLDL1KEEP, [x16] \n\t"
|
||||
" add x16, x16, %5 \n\t"
|
||||
" add x16, x16, %4 \n\t"
|
||||
" prfm PLDL1KEEP, [x16] \n\t"
|
||||
" END_CCOL_PRFM: \n\t"
|
||||
" \n\t"
|
||||
|
||||
@@ -116,23 +116,23 @@ GEMM_ACOLCMPLX_CONTIGUOUS_LOAD_FWD(z28,z29,p0,%0,x2)
|
||||
" b.ne END_CCOL_PRFM \n\t" // Do not prefetch for generic C storage.
|
||||
" mov x16, %2 \n\t"
|
||||
" prfm PLDL1KEEP, [x16] \n\t"
|
||||
" add x16, x16, %5 \n\t"
|
||||
" add x16, x16, %4 \n\t"
|
||||
" prfm PLDL1KEEP, [x16] \n\t"
|
||||
" add x16, x16, %5 \n\t"
|
||||
" add x16, x16, %4 \n\t"
|
||||
" prfm PLDL1KEEP, [x16] \n\t"
|
||||
" add x16, x16, %5 \n\t"
|
||||
" add x16, x16, %4 \n\t"
|
||||
" prfm PLDL1KEEP, [x16] \n\t"
|
||||
" add x16, x16, %5 \n\t"
|
||||
" add x16, x16, %4 \n\t"
|
||||
" prfm PLDL1KEEP, [x16] \n\t"
|
||||
" add x16, x16, %5 \n\t"
|
||||
" add x16, x16, %4 \n\t"
|
||||
" prfm PLDL1KEEP, [x16] \n\t"
|
||||
" add x16, x16, %5 \n\t"
|
||||
" add x16, x16, %4 \n\t"
|
||||
" prfm PLDL1KEEP, [x16] \n\t"
|
||||
" add x16, x16, %5 \n\t"
|
||||
" add x16, x16, %4 \n\t"
|
||||
" prfm PLDL1KEEP, [x16] \n\t"
|
||||
" add x16, x16, %5 \n\t"
|
||||
" add x16, x16, %4 \n\t"
|
||||
" prfm PLDL1KEEP, [x16] \n\t"
|
||||
" add x16, x16, %5 \n\t"
|
||||
" add x16, x16, %4 \n\t"
|
||||
" prfm PLDL1KEEP, [x16] \n\t"
|
||||
" END_CCOL_PRFM: \n\t"
|
||||
" \n\t"
|
||||
|
||||
@@ -123,17 +123,17 @@ GEMM_ACOLCMPLX_CONTIGUOUS_LOAD_FWD(z28,z29,p0,%0,x2)
|
||||
" b.ne END_CCOL_PRFM \n\t" // Do not prefetch for generic C storage.
|
||||
" mov x16, %2 \n\t"
|
||||
" prfm PLDL1KEEP, [x16] \n\t"
|
||||
" add x16, x16, %5 \n\t"
|
||||
" add x16, x16, %4 \n\t"
|
||||
" prfm PLDL1KEEP, [x16] \n\t"
|
||||
" add x16, x16, %5 \n\t"
|
||||
" add x16, x16, %4 \n\t"
|
||||
" prfm PLDL1KEEP, [x16] \n\t"
|
||||
" add x16, x16, %5 \n\t"
|
||||
" add x16, x16, %4 \n\t"
|
||||
" prfm PLDL1KEEP, [x16] \n\t"
|
||||
" add x16, x16, %5 \n\t"
|
||||
" add x16, x16, %4 \n\t"
|
||||
" prfm PLDL1KEEP, [x16] \n\t"
|
||||
" add x16, x16, %5 \n\t"
|
||||
" add x16, x16, %4 \n\t"
|
||||
" prfm PLDL1KEEP, [x16] \n\t"
|
||||
" add x16, x16, %5 \n\t"
|
||||
" add x16, x16, %4 \n\t"
|
||||
" prfm PLDL1KEEP, [x16] \n\t"
|
||||
" END_CCOL_PRFM: \n\t"
|
||||
" \n\t"
|
||||
|
||||
@@ -120,19 +120,19 @@ GEMM_ACOLCMPLX_CONTIGUOUS_LOAD_FWD(z16,z17,p0,%0,x2)
|
||||
" b.ne END_CCOL_PRFM \n\t" // Do not prefetch for generic C storage.
|
||||
" mov x16, %2 \n\t"
|
||||
" prfm PLDL1KEEP, [x16] \n\t"
|
||||
" add x16, x16, %5 \n\t"
|
||||
" add x16, x16, %4 \n\t"
|
||||
" prfm PLDL1KEEP, [x16] \n\t"
|
||||
" add x16, x16, %5 \n\t"
|
||||
" add x16, x16, %4 \n\t"
|
||||
" prfm PLDL1KEEP, [x16] \n\t"
|
||||
" add x16, x16, %5 \n\t"
|
||||
" add x16, x16, %4 \n\t"
|
||||
" prfm PLDL1KEEP, [x16] \n\t"
|
||||
" add x16, x16, %5 \n\t"
|
||||
" add x16, x16, %4 \n\t"
|
||||
" prfm PLDL1KEEP, [x16] \n\t"
|
||||
" add x16, x16, %5 \n\t"
|
||||
" add x16, x16, %4 \n\t"
|
||||
" prfm PLDL1KEEP, [x16] \n\t"
|
||||
" add x16, x16, %5 \n\t"
|
||||
" add x16, x16, %4 \n\t"
|
||||
" prfm PLDL1KEEP, [x16] \n\t"
|
||||
" add x16, x16, %5 \n\t"
|
||||
" add x16, x16, %4 \n\t"
|
||||
" prfm PLDL1KEEP, [x16] \n\t"
|
||||
" END_CCOL_PRFM: \n\t"
|
||||
" \n\t"
|
||||
|
||||
Reference in New Issue
Block a user