Skip to content

Add Thread Throttling Optimization for Power10 in GEMV #5507

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
wants to merge 1 commit into
base: develop
Choose a base branch
from
Draft

Add Thread Throttling Optimization for Power10 in GEMV #5507

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
23 changes: 22 additions & 1 deletion interface/gemv.c
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -100,14 +100,35 @@ static inline int get_gemv_optimal_nthreads_neoversev2(BLASLONG MN, int ncpu) {
}
#endif

#if defined(DYNAMIC_ARCH) || defined(POWER10)
static inline int get_gemv_optimal_nthreads_power10(BLASLONG MN, int ncpu){

#ifdef DOUBLE
return
MN < 459684L ? 1
: MN <4000000L ? MIN(ncpu, 32)
: MN < 10240000L ? MIN(ncpu, 128)
: ncpu;
#else
return
MN < 459684L ? 1
: MN < 2890000L ? MIN(ncpu, 32)
: MN < 10240000L ? MIN(ncpu, 128)
: ncpu;
#endif
}
#endif

static inline int get_gemv_optimal_nthreads(BLASLONG MN) {
int ncpu = num_cpu_avail(3);
#if defined(_WIN64) && defined(_M_ARM64)
if (MN > 100000000L)
return num_cpu_avail(4);
return 1;
#endif
#if defined(NEOVERSEV1) && !defined(COMPLEX) && !defined(BFLOAT16)
#if defined(POWER10) && !defined(COMPLEX) && !defined(BFLOAT16)
return get_gemv_optimal_nthreads_power10(MN, ncpu);
#elif defined(NEOVERSEV1) && !defined(COMPLEX) && !defined(BFLOAT16)
return get_gemv_optimal_nthreads_neoversev1(MN, ncpu);
#elif defined(NEOVERSEV2) && !defined(COMPLEX) && !defined(DOUBLE) && !defined(BFLOAT16)
return get_gemv_optimal_nthreads_neoversev2(MN, ncpu);
Expand Down
Loading