Skip to content

Force-define _CCCL_ATOMIC_UNSAFE_AUTOMATIC_STORAGE for CUDA <13.1 #816

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.

Sign up for GitHub

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

Open
PointKernel wants to merge 1 commit into
base: dev
Choose a base branch
from
+9 −0
Open

Force-define _CCCL_ATOMIC_UNSAFE_AUTOMATIC_STORAGE for CUDA <13.1 #816

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
9 changes: 9 additions & 0 deletions include/cuco/detail/__config
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -34,6 +34,15 @@
#error "CCCL version 3.0.0 or later is required"
#endif

// WAR for cuCollections/804: work around the nvcc __cuda_is_local / isspacep.local
// miscompile on CUDA < 13.1. CCCL's auto-define of this macro only fires when
// NDEBUG is not set, but cuco builds Release, so define it unconditionally for
// affected toolkit versions.
#if !defined(_CCCL_ATOMIC_UNSAFE_AUTOMATIC_STORAGE) \
&& ((__CUDACC_VER_MAJOR__ < 13) || (__CUDACC_VER_MAJOR__ == 13 && __CUDACC_VER_MINOR__ < 1))
#define _CCCL_ATOMIC_UNSAFE_AUTOMATIC_STORAGE
#endif

// WAR for libcudacxx/296
#define CUCO_CUDA_MINIMUM_ARCH _NV_FIRST_ARG(__CUDA_ARCH_LIST__)

Expand Down
Loading