Skip to content
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

Jump to bottom

Add PspaMM support for Rome and HSW #59

Merged
merged 1 commit into from
Aug 24, 2023
Merged

Add PspaMM support for Rome and HSW #59

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
9 changes: 8 additions & 1 deletion yateto/codegen/gemm/gemmgen.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -233,6 +233,13 @@ def __call__(self, routineName, fileName):
cpu_arch = self._arch.host_name if self._arch.host_name else self._arch.name

if self._mode == 'pspamm':
pspamm_arch = cpu_arch
if cpu_arch == 'a64fx':
pspamm_arch = 'arm_sve'
elif cpu_arch in ['naples', 'rome', 'milan']:
# names are Zen1, Zen2, Zen3, respectively
# no explicit support for these archs yet, but they have the same instruction sets (AVX2+FMA3) that HSW also needs
pspamm_arch = 'hsw'
argList = [
self._cmd,
self._gemmDescr['M'],
Expand All @@ -244,7 +251,7 @@ def __call__(self, routineName, fileName):
self._gemmDescr['alpha'],
self._gemmDescr['beta'],
'--arch',
'arm_sve' if cpu_arch == 'a64fx' else cpu_arch,
pspamm_arch,
'--prefetching',
self._gemmDescr['prefetch'],
'--output_funcname',
Expand Down
Loading