Talk:NVIDIA: Difference between revisions

imported>SomeoneSerge
cudnn question
imported>SomeoneSerge
overriding cudnn: add link to an example thread
Line 16: Line 16:


Another aspect of the problem is in overriding downstream derivations, that take as inputs both cudnn and cudatoolkit.
Another aspect of the problem is in overriding downstream derivations, that take as inputs both cudnn and cudatoolkit.
For example, (speaking of nixpkgs-21.11) when a user overrides pytorch to use cudatoolkit_11, the build fails because pytorch gets two conflicting versions of cuda: cudatoolkit_11 from the user, and cudatoolkit_10 through the dependency on cudnn. Until we've fixed these issues on nixpkgs, we should at least make them and the workarounds visible to the users to reduce frustration
For example, (speaking of nixpkgs-21.11) when a user overrides pytorch to use cudatoolkit_11, the build fails because pytorch gets two conflicting versions of cuda: cudatoolkit_11 from the user, and cudatoolkit_10 through the dependency on cudnn. Until we've fixed these issues on nixpkgs, we should at least make them and the workarounds visible to the users to reduce frustration. See e.g. https://discourse.nixos.org/t/cant-get-pytorch-to-recognise-override/12082


--[[User:SomeoneSerge|SomeoneSerge]] ([[User talk:SomeoneSerge|talk]]) 21:58, 17 March 2022 (UTC)
--[[User:SomeoneSerge|SomeoneSerge]] ([[User talk:SomeoneSerge|talk]]) 21:58, 17 March 2022 (UTC)
Return to "NVIDIA" page.