tqchen opened a new pull request, #19770:
URL: https://github.com/apache/tvm/pull/19770

   ## Summary
   
   The CUDA runtime can rely on the shared tvm-ffi CUDA error helper instead of 
carrying a TVM-local common header. This keeps CUDA error handling aligned with 
the FFI CUDA support and removes the remaining reason for 
src/backend/cuda/runtime/cuda_common.h.
   
   Main changes:
   
   - Move CUDA workspace thread-local state into cuda_device_api.cc.
   - Replace CUDA_CALL use sites with TVM_FFI_CHECK_CUDA_ERROR from 
tvm/ffi/extra/cuda/base.h.
   - Inline the remaining CUDA driver checks at their call sites and delete 
cuda_common.h.


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to