-
Notifications
You must be signed in to change notification settings - Fork 233
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
Support for qr of strided inputs (non-contiguous views) #1764
base: master
Are you sure you want to change the base?
Conversation
Co-authored-by: Valentin Churavy <[email protected]>
Looks like some more functions will need their signatures widened (as CI shows fallback functionality being hit). |
These two https://github.com/JuliaGPU/GPUArrays.jl/blob/b2f396dabdab907866b885cd7abb647214cddace/src/host/linalg.jl#L33-L34 are implemented for dense Arrays, but not for strided arrays. So we end up https://github.com/JuliaLang/julia/blob/c0d2c574d44231dba39f90d6a6b27448f71763ca/stdlib/LinearAlgebra/src/transpose.jl#L76-L77 |
No reason those shouldn't be an |
Changed this in JuliaGPU/GPUArrays.jl#452 |
I've tagged GPUArrays 8.6.3 with this fix. |
Linking JuliaLang/LinearAlgebra.jl#986 for the dimension issues in the tests |
d1d5e3f
to
2bd209c
Compare
5d585c4
to
c850163
Compare
Adding view support for CUDA.jl QR and tests