Skip to content

Conversation

@netrunnereve
Copy link
Collaborator

Instead of skipping the tests when we get a CI machine with no AVX-512 we can run them using the Intel SDE emulator. It's pretty slow but it'll still finish much faster than the CUDA build.

@ggerganov ggerganov merged commit a7fac01 into ggml-org:master Nov 5, 2023
@netrunnereve netrunnereve deleted the intel_sde branch November 5, 2023 16:44
olexiyb pushed a commit to Sanctum-AI/llama.cpp that referenced this pull request Nov 23, 2023
@Green-Sky
Copy link
Collaborator

Green-Sky commented Jan 20, 2024

the ${env:SDE_VERSION} in

curl.exe -o $env:RUNNER_TEMP/sde.tar.xz -L "https://downloadmirror.intel.com/777395/sde-external-${env:SDE_VERSION}-win.tar.xz"

does not seem to get expanded and ci is failing.

edit: someone removed SDE_VERSION: 9.21.1-2023-04-24 from the ci env

@cebtenzzre
Copy link
Collaborator

edit: someone removed SDE_VERSION: 9.21.1-2023-04-24 from the ci env

No, it's still there:

https://github.com/ggerganov/llama.cpp/blob/97c1549808d2742d37584a3c9df28154bdf34417/.github/workflows/build.yml#L298

@netrunnereve
Copy link
Collaborator Author

netrunnereve commented Jan 21, 2024

It looks like this is because Intel deleted our version off their site, the link is now https://downloadmirror.intel.com/813591/sde-external-9.33.0-2024-01-07-win.tar.xz for the latest version.

It might make sense to use the older version https://downloadmirror.intel.com/788820/sde-external-9.27.0-2023-09-13-win.tar.xz as Intel seems to be archiving this one.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants