Skip to content

[Torch] Disable building with MKL#5583

Merged
giordano merged 2 commits intoJuliaPackaging:masterfrom
stemann:stemann/torch_fixes
Dec 2, 2022
Merged

[Torch] Disable building with MKL#5583
giordano merged 2 commits intoJuliaPackaging:masterfrom
stemann:stemann/torch_fixes

Conversation

@stemann
Copy link
Contributor

@stemann stemann commented Sep 27, 2022

No description provided.

@stemann stemann force-pushed the stemann/torch_fixes branch 2 times, most recently from 0e79969 to b5bd876 Compare October 2, 2022 07:26
@stemann stemann force-pushed the stemann/torch_fixes branch 8 times, most recently from 0cb084d to 860bfa1 Compare October 3, 2022 11:09
@ViralBShah
Copy link
Member

What needs to be done here to get this merged?

@stemann stemann force-pushed the stemann/torch_fixes branch 2 times, most recently from 6e35c22 to 109d5cc Compare November 29, 2022 13:42
@stemann
Copy link
Contributor Author

stemann commented Nov 29, 2022

What needs to be done here to get this merged?

Fighting a bit with the build process to have it find LAPACK - and BLAS.

... and considering libblastrampoline though I guess it would constrain compatibility to Julia v1.7+ - would prefer something compatible with v1.6.

@ViralBShah
Copy link
Member

ViralBShah commented Nov 30, 2022

I suggest not worrying about 1.6 or even 1.7. These capabilities are fast moving, and we'll have a new LTS next year (I hope - not official statement). That will make it much easier to work with LBT.

@stemann stemann force-pushed the stemann/torch_fixes branch 2 times, most recently from da0852b to bef6194 Compare December 1, 2022 14:34
@stemann
Copy link
Contributor Author

stemann commented Dec 1, 2022

I have confirmed that the generated Torch_jll can be imported on x86_64-linux-gnu-cxx11.

@stemann stemann marked this pull request as ready for review December 1, 2022 14:37
Also,
* Changed recipe to use OpenBLAS (OpenBLAS32) and not BLIS. Building with BLIS cannot find LAPACK (in OpenBLAS) - and OpenBLAS takes precedence over BLIS.
* Added Declarations.yaml FileProduct - needed for generating Torch wrapper.
@stemann stemann force-pushed the stemann/torch_fixes branch from bef6194 to ffaf28c Compare December 1, 2022 14:37
@stemann stemann marked this pull request as draft December 1, 2022 18:32
@stemann stemann force-pushed the stemann/torch_fixes branch from bff3507 to e06e3c0 Compare December 2, 2022 09:22
@stemann stemann marked this pull request as ready for review December 2, 2022 09:59
@ViralBShah
Copy link
Member

This may need a @giordano review.

@giordano giordano merged commit f6fd019 into JuliaPackaging:master Dec 2, 2022
@giordano
Copy link
Member

giordano commented Dec 2, 2022

Thanks a lot @stemann!

Dependency("XNNPACK_jll"; compat = "0.0.20210622"),
Dependency(PackageSpec("protoc_jll", Base.UUID("c7845625-083e-5bbe-8504-b32d602b7110")); compat="~3.13.0"),
HostBuildDependency(PackageSpec("protoc_jll", Base.UUID("c7845625-083e-5bbe-8504-b32d602b7110"), v"3.13.0")),
RuntimeDependency("CUDA_Runtime_jll"),
Copy link
Member

Choose a reason for hiding this comment

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

Was this supposed to be a dependency for all platforms? Or only the CUDA ones?

Copy link
Contributor Author

@stemann stemann Dec 2, 2022

Choose a reason for hiding this comment

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

It’s required, I believe, by the CUDA augment block that I added - I guess that is necessary in order to be able to get hold of the CUDA artifacts.

@stemann stemann deleted the stemann/torch_fixes branch December 3, 2022 09:36
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.

3 participants