floraachy
/
OpenBLAS
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Use p2align instead of align for OSX compatibility 

fixes #1902
This commit is contained in:
Martin Kroeker 2018-12-03 13:06:43 +01:00 committed by GitHub
parent 721c56c224
commit 701ea88347
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
kernel/x86_64/dgemm_kernel_4x8_skylakex.c
View File

@ -869,7 +869,7 @@ CNAME(BLASLONG m, BLASLONG n, BLASLONG k, double alpha, double * __restrict__ A,
"vmovapd %%zmm1, %%zmm27\n"
"vmovapd %%zmm1, %%zmm28\n"
"jmp .label24\n"
".align 32\n"
".p2align 5\n"
/* Inner math loop */
".label24:\n"
"vmovupd -128(%[AO]),%%zmm0\n"
@ -1037,7 +1037,7 @@ CNAME(BLASLONG m, BLASLONG n, BLASLONG k, double alpha, double * __restrict__ A,
"vmovapd %%zmm1, %%zmm17\n"
"vmovapd %%zmm1, %%zmm18\n"
"jmp .label16\n"
".align 32\n"
".p2align 5\n"
/* Inner math loop */
".label16:\n"
"vmovupd -128(%[AO]),%%zmm0\n"
@ -1165,7 +1165,7 @@ CNAME(BLASLONG m, BLASLONG n, BLASLONG k, double alpha, double * __restrict__ A,
"vmovapd %%zmm1, %%zmm8\n"
"vbroadcastsd (%[alpha]), %%zmm9\n"
"jmp .label1\n"
".align 32\n"
".p2align 5\n"
/* Inner math loop */
".label1:\n"
"vmovupd -128(%[AO]),%%zmm0\n"