Skip to content

Llvm kernel languages: Fix nvidia fatbin loading#22

Merged
kevinsala merged 4 commits into
jdoerfert:llvm_kernel_languagesfrom
cadivus:llvm_kernel_languages__fix_nvidia_fatbin_loading
Aug 7, 2025
Merged

Llvm kernel languages: Fix nvidia fatbin loading#22
kevinsala merged 4 commits into
jdoerfert:llvm_kernel_languagesfrom
cadivus:llvm_kernel_languages__fix_nvidia_fatbin_loading

Conversation

@cadivus

@cadivus cadivus commented Aug 5, 2025

Copy link
Copy Markdown

Makes it compatible with NVIDIA GPUs. Also chooses the right bin if it got compiled for multiple GPU archs.

@cadivus cadivus changed the title Llvm kernel languages fix nvidia fatbin loading Llvm kernel languages: Fix nvidia fatbin loading Aug 5, 2025
@cadivus cadivus force-pushed the llvm_kernel_languages__fix_nvidia_fatbin_loading branch from 44b7855 to f84ce60 Compare August 5, 2025 02:24
@kevinsala kevinsala self-requested a review August 5, 2025 18:40
Comment thread offload/languages/include/cuda/cuda_runtime.h Outdated
Comment thread offload/languages/kernel/src/LanguageRegistration.cpp Outdated
Comment thread offload/liboffload/include/generated/OffloadAPI.h Outdated
@cadivus cadivus marked this pull request as draft August 6, 2025 23:51
@cadivus cadivus force-pushed the llvm_kernel_languages__fix_nvidia_fatbin_loading branch from 93c6710 to 4fa8c5e Compare August 7, 2025 00:00
@cadivus cadivus marked this pull request as ready for review August 7, 2025 00:01
@cadivus cadivus force-pushed the llvm_kernel_languages__fix_nvidia_fatbin_loading branch from 4fa8c5e to b24c955 Compare August 7, 2025 00:08

@kevinsala kevinsala left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

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

LGTM

@kevinsala kevinsala merged commit 413f660 into jdoerfert:llvm_kernel_languages Aug 7, 2025
5 checks passed
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.

2 participants