Browse Source

Merge pull request #3104 from martin-frbg/issue3103

Enable optimized Haswell/AVX2 kernels for sasum/dasum and srot/drot on Ryzen
tags/v0.3.14^2
Martin Kroeker GitHub 5 years ago
parent
commit
074d9bff7f
No known key found for this signature in database GPG Key ID: 4AEE18F83AFDEB23
5 changed files with 6 additions and 4 deletions
  1. +2
    -0
      kernel/x86_64/KERNEL.ZEN
  2. +1
    -1
      kernel/x86_64/dasum.c
  3. +1
    -1
      kernel/x86_64/drot.c
  4. +1
    -1
      kernel/x86_64/sasum.c
  5. +1
    -1
      kernel/x86_64/srot.c

+ 2
- 0
kernel/x86_64/KERNEL.ZEN View File

@@ -97,3 +97,5 @@ ZTRSMKERNEL_RT = ../generic/trsm_kernel_RT.c
CGEMM3MKERNEL = cgemm3m_kernel_8x4_haswell.c
ZGEMM3MKERNEL = zgemm3m_kernel_4x4_haswell.c

SROTKERNEL = srot.c
DROTKERNEL = drot.c

+ 1
- 1
kernel/x86_64/dasum.c View File

@@ -6,7 +6,7 @@

#if defined(SKYLAKEX)
#include "dasum_microk_skylakex-2.c"
#elif defined(HASWELL)
#elif defined(HASWELL) || defined(ZEN)
#include "dasum_microk_haswell-2.c"
#endif



+ 1
- 1
kernel/x86_64/drot.c View File

@@ -2,7 +2,7 @@

#if defined(SKYLAKEX)
#include "drot_microk_skylakex-2.c"
#elif defined(HASWELL)
#elif defined(HASWELL) || defined(ZEN)
#include "drot_microk_haswell-2.c"
#endif



+ 1
- 1
kernel/x86_64/sasum.c View File

@@ -11,7 +11,7 @@

#if defined(SKYLAKEX)
#include "sasum_microk_skylakex-2.c"
#elif defined(HASWELL)
#elif defined(HASWELL) || defined(ZEN)
#include "sasum_microk_haswell-2.c"
#endif



+ 1
- 1
kernel/x86_64/srot.c View File

@@ -2,7 +2,7 @@

#if defined(SKYLAKEX)
#include "srot_microk_skylakex-2.c"
#elif defined(HASWELL)
#elif defined(HASWELL) || defined(ZEN)
#include "srot_microk_haswell-2.c"
#endif



Loading…
Cancel
Save