saymrwulf/onnxruntime
1
0
Fork 0
mirror of https://github.com/saymrwulf/onnxruntime.git synced 2026-07-30 20:18:08 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

MLAS: clang u8u8 GEMM fix

Browse source
This commit is contained in:
Tracy Sharpe 2019-09-06 09:11:10 -07:00 committed by GitHub
parent b2a2326a45
commit a324ad7b96
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
2 changed files with 2 additions and 2 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
onnxruntime/core/mlas/lib/amd64/QgemmU8U8KernelAvx2.asm
View file

@ -215,7 +215,7 @@ ProcessNextRowM4:
mov rdx,rsi
mov rcx,rdi
lea rsi,[rsi+r8*4] ; advance next matrix A by 4 rows
lea rdi,[rdi+r11*(2*4)] ; advance next matrix D by 4 rows
lea rdi,[rdi+r11*8] ; advance next matrix D by 4 rows
mov rbx,r10 ; reload columns remaining
sub rbx,16
jb ProcessRemainingColumnsM4

2
onnxruntime/core/mlas/lib/x86_64/QgemmU8U8KernelAvx2.S
View file

@ -158,7 +158,7 @@ C_UNDERSCORE(MlasGemmU8U8CopyPackAAvx2):
mov rdx,rsi
mov rcx,rdi
lea rsi,[rsi+r10*4] # advance next matrix A by 4 rows
lea rdi,[rdi+r12*(2*4)] # advance next matrix D by 4 rows
lea rdi,[rdi+r12*8] # advance next matrix D by 4 rows
mov rbx,r8 # reload columns remaining
sub rbx,16
jb .LCopyPackA.ProcessRemainingColumnsM4