Browse Source

Merge pull request #2049 from Celelibi/fix_crash_sgemm_sse_x64

Fix crash in sgemm SSE/nano kernel on x86_64
tags/v0.3.6^2
Martin Kroeker GitHub 7 years ago
parent
commit
8d3d29e4d7
No known key found for this signature in database GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 2 additions and 2 deletions
  1. +1
    -1
      kernel/x86_64/gemm_kernel_4x8_nano.S
  2. +1
    -1
      kernel/x86_64/gemm_kernel_8x4_sse.S

+ 1
- 1
kernel/x86_64/gemm_kernel_4x8_nano.S View File

@@ -135,7 +135,7 @@
#endif

movq %rsp, %rbx # save old stack
subq $128 + LOCAL_BUFFER_SIZE, %rsp
subq $256 + LOCAL_BUFFER_SIZE, %rsp
andq $-4096, %rsp # align stack

STACK_TOUCHING


+ 1
- 1
kernel/x86_64/gemm_kernel_8x4_sse.S View File

@@ -383,7 +383,7 @@
EMMS

movq %rsp, %rbx # save old stack
subq $128 + LOCAL_BUFFER_SIZE, %rsp
subq $256 + LOCAL_BUFFER_SIZE, %rsp
andq $-4096, %rsp # align stack

STACK_TOUCHING


Loading…
Cancel
Save