-
Notifications
You must be signed in to change notification settings - Fork 117
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.
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
[HIP][CUDA] Change unions in ur_mem_handle_t_ to stdvariant #961
[HIP][CUDA] Change unions in ur_mem_handle_t_ to stdvariant #961
Conversation
intel/llvm testing: intel/llvm#11539 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM, sorry about the wait.
@kbenzie can we merge this please? |
We already have 2 UR changes that can't merge to intel/llvm because of test failures, I feel like merging this will only add to that problem. |
@kbenzie OK sure yes no rush |
2b2e234
to
86f96f0
Compare
This is required after oneapi-src#961
Update UR SHA to oneapi-src/unified-runtime#961 change unions to std::variant in the HIP and CUDA adapters. --------- Co-authored-by: Kenneth Benzie (Benie) <k.benzie83@gmail.com>
Update UR SHA to oneapi-src/unified-runtime#961 change unions to std::variant in the HIP and CUDA adapters. --------- Co-authored-by: Kenneth Benzie (Benie) <k.benzie83@gmail.com>
Std::variant is safer than unions in the
ur_mem_handle_t_
class and it will make it easier to incrementally move to multi device ctx, which requires addingstd::vector
s to the MemImpl, which is not possible in a union.