Skip to content

Add SME vector length detect #287

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.

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

Merged
merged 3 commits into from
May 22, 2025
Merged

Add SME vector length detect #287

merged 3 commits into from
May 22, 2025

Conversation

fbarchard
Copy link
Contributor

@fbarchard fbarchard commented Mar 28, 2025

Add cpuinfo_get_max_arm_sme_length() to return svl vector length in bits

Display length of SME vectors in isa-tool

SME may be enabled on cpus that do not have SVE

Tested isa-tool on qemu

SIMD extensions:
	ARM SVE: yes
	ARM SVE 2: yes
	ARM SME: yes
	ARM SME 2: no
ARM SVE Capabilities:
	SVE max length: 512
	SME max length: 256

- cpuinfo_get_max_arm_sme_length() returns svl vector length in bits
- Display length of SME vectors in isa-tool

SME may be enabled on cpus that do not have SVE
- cpuinfo_get_max_arm_sme_length() returns svl vector length in bits
- Display length of SME vectors in isa-tool
- Upgrade cmake-linux-riscv64 ubuntu-22.04 runners to ubuntu-24.04

SME may be enabled on cpus that do not have SVE
Copy link
Contributor Author

@fbarchard fbarchard left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

includes workflow fix for riscv to use ubuntu 24.04

@digantdesai
Copy link
Contributor

Unable to find image 'riscv64/ubuntu:24.04' locally

@digantdesai digantdesai merged commit b728a23 into pytorch:main May 22, 2025
9 of 12 checks passed
@malfet
Copy link
Contributor

malfet commented May 22, 2025

@digantdesai why did you merge this PR when it introduced formatting rule violation? Please stop doing that!!!

@malfet malfet mentioned this pull request May 22, 2025
malfet added a commit that referenced this pull request May 22, 2025
Regressions introduced by #275 and #287
@digantdesai
Copy link
Contributor

@digantdesai why did you merge this PR when it introduced formatting rule violation? Please stop doing that!!!

Yeah apologies about aarch64-isa.c. But this PR doesn't touch cpuinfo.c, and that made me believe it was unrelated. https://github.com/pytorch/cpuinfo/actions/runs/14137114188/job/39743929244?pr=287

Now that the CI is green. It should be easy to not get false positives.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

4 participants