Skip to content

[AMGXWrapper][CUDA] Enable position independent code globally#19

Open
fmgjcoppens wants to merge 1 commit into
ledac/next_gpu_rebasedfrom
coppens/fix_build_amgxwrapper
Open

[AMGXWrapper][CUDA] Enable position independent code globally#19
fmgjcoppens wants to merge 1 commit into
ledac/next_gpu_rebasedfrom
coppens/fix_build_amgxwrapper

Conversation

@fmgjcoppens

@fmgjcoppens fmgjcoppens commented Jul 17, 2026

Copy link
Copy Markdown
Collaborator

This PR fixes the ICoCo build when CUDA is enabled. By activating position independent code at CMake level, -fPIC is automatically passed to all compilers that support it.

-DCMAKE_CXX_FLAGS=-fPIC is not propagated correctly to the host
compiler when compiling .cu-files. This causes the ICoCo build later to
fail in the linking stage.

@fmgjcoppens
fmgjcoppens requested a review from pledac July 17, 2026 20:17
By enabling position indendend code at CMake level -fPIC is
automatically passed to all compilers that support it.

`-DCMAKE_CXX_FLAGS=-fPIC` is not propagated correctly to the host
compiler when compiling .cu-files. This causes the ICoCo build later to
fail in the linking stage.
@fmgjcoppens
fmgjcoppens force-pushed the coppens/fix_build_amgxwrapper branch from 488429c to fc2451b Compare July 17, 2026 20:27
@fmgjcoppens fmgjcoppens changed the title [AMGXWrapper] Enable position independent code globally [AMGXWrapper][CUDA] Enable position independent code globally Jul 17, 2026
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.

1 participant