amd/blis
1
0
Fork 0
mirror of https://github.com/amd/blis.git synced 2026-05-11 17:50:00 +00:00
Code Issues Packages Projects Releases Wiki Activity

CPUPL-1074:

Browse Source 
- Bug fix in sgemmsup 1x16 Kernel for Beta Zero and with C col storage
       rcx register incrementing was missing because of this 4 values
       in output are overwritten

Change-Id: Ia3028040dce3e615f1db5a331498d86faadcf916
This commit is contained in:
bhaskarn
2020-08-10 17:45:23 +05:30
committed by Nallani Bhaskar
parent 7bbcae5a18
commit d186cfdf2e
1 changed files with 1 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

1
kernels/zen/3/sup/bli_gemmsup_rv_zen_asm_s6x16.c
View File

@@ -2479,6 +2479,7 @@ void bli_sgemmsup_rv_zen_asm_1x16
vmovss(xmm1, mem(rcx, rsi, 1))
vmovss(xmm2, mem(rcx, rsi, 2))
vmovss(xmm14, mem(rcx, rax, 1))
lea(mem(rcx, rsi, 4), rcx) // rcx += cs_c
vextractf128(imm(0x0), ymm5, xmm0)//c0-c3
vshufps(imm(0x01), xmm0, xmm0,xmm1)
vshufps(imm(0x02), xmm0, xmm0,xmm2)