This website works better with JavaScript.
Home
Issues
Pull Requests
Milestones
AI流水线
Repositories
Datasets
Forum
实训
竞赛
大数据
AI开发
Register
Sign In
OSchip
/
OpenBLAS
Not watched
Unwatch
Watch all
Watch but not notify
1
Star
0
Fork
0
Code
Releases
66
Wiki
evaluate
Activity
Issues
0
Pull Requests
0
Datasets
Model
Cloudbrain
HPC
Browse Source
VULCAN: Add optimized DGEMM implementation
tags/v0.2.20^2
Ashwin Sekhar T K
9 years ago
parent
4713e7c47f
commit
0b8e876d89
3 changed files
with
1829 additions
and
0 deletions
Split View
Diff Options
Show Stats
Download Patch File
Download Diff File
+10
-0
driver/others/parameter.c
+2
-0
kernel/arm64/KERNEL.VULCAN
+1817
-0
kernel/arm64/dgemm_kernel_8x4_vulcan.S
+ 10
- 0
driver/others/parameter.c
View File
@@ -730,12 +730,22 @@ void blas_set_parameter(void){
#if defined(ARCH_ARM64)
#if defined(VULCAN)
unsigned long vulcan_pre_a;
unsigned long vulcan_pre_b;
unsigned long vulcan_pre_c;
#endif
void blas_set_parameter(void)
{
#if defined(VULCAN)
dgemm_p = 160;
dgemm_q = 128;
dgemm_r = 4096;
vulcan_pre_a = 3584;
vulcan_pre_b = 512;
vulcan_pre_c = 128;
#endif
}
+ 2
- 0
kernel/arm64/KERNEL.VULCAN
View File
@@ -1,2 +1,4 @@
include $(KERNELDIR)/KERNEL.CORTEXA57
DGEMMKERNEL = dgemm_kernel_$(DGEMM_UNROLL_M)x$(DGEMM_UNROLL_N)_vulcan.S
+ 1817
- 0
kernel/arm64/dgemm_kernel_8x4_vulcan.S
File diff suppressed because it is too large
View File
Write
Preview
Loading…
Cancel
Save