Skip to content

Enable GPU offloading in Clang if CUDA is included as a dep#1716

Merged
boegel merged 9 commits intoeasybuilders:developfrom
ocaisa:clang_linker
May 9, 2019
Merged

Enable GPU offloading in Clang if CUDA is included as a dep#1716
boegel merged 9 commits intoeasybuilders:developfrom
ocaisa:clang_linker

Conversation

@ocaisa
Copy link
Copy Markdown
Member

@ocaisa ocaisa commented May 8, 2019

and support building lld

@ocaisa ocaisa changed the title Enable GPU offloading if CUDA is included as a dep Enable GPU offloading in Clang if CUDA is included as a dep May 8, 2019
Comment thread easybuild/easyblocks/c/clang.py Outdated
Comment thread easybuild/easyblocks/c/clang.py Outdated
Comment thread easybuild/easyblocks/c/clang.py Outdated
Copy link
Copy Markdown
Member

@boegel boegel left a comment

Choose a reason for hiding this comment

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

(see comments)

@boegel boegel added this to the 3.x milestone May 8, 2019
Alan O'Cais added 3 commits May 9, 2019 09:07
Required to correctly decide if we should include NVPTX
Also use self.config.update everywhere possible
@ocaisa
Copy link
Copy Markdown
Member Author

ocaisa commented May 9, 2019

I moved defining the build targets until the configure step so that we can automatically include architectures that require a dependency (CUDA) while still respecting if someone sets the architectures explicitly.

@boegel boegel modified the milestones: 3.x, next release (3.9.1) May 9, 2019
@boegel
Copy link
Copy Markdown
Member

boegel commented May 9, 2019

lgtm, tested with Clang 8.0.0 easyconfigs from #8254 and #8276, so going in, thanks @ocaisa!

@boegel boegel merged commit 0d52c8f into easybuilders:develop May 9, 2019
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.

2 participants