[UR][CUDA] Add opportunistic queue serialize prop, impl for cuda #18443
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Signed-off-by: JackAKirk <[email protected]>
Signed-off-by: JackAKirk <[email protected]>
Signed-off-by: JackAKirk <[email protected]>
|
Reasonable chance this will interact/conflict with #18385 |
|
yeah I'm going to need to rethink how devices reporting support for different properties looks I think |
jchlanda
approved these changes
May 14, 2025
| @@ -12320,6 +12323,9 @@ typedef union ur_exp_launch_property_value_t { | |||
| /// [in] non-zero value indicates the amount of work group memory to | |||
| /// allocate in bytes | |||
| size_t workgroup_mem_size; | |||
| /// [in] non-zero value indicates a opportunistic native queue serialized | |||
There was a problem hiding this comment.
Suggested change
| /// [in] non-zero value indicates a opportunistic native queue serialized | |
| /// [in] non-zero value indicates an opportunistic native queue serialized |
| @@ -56,6 +58,10 @@ members: | |||
| name: workgroup_mem_size | |||
| desc: "[in] non-zero value indicates the amount of work group memory to allocate in bytes" | |||
| tag: $X_EXP_LAUNCH_PROPERTY_ID_WORK_GROUP_MEMORY | |||
| - type: int | |||
| name: opportunistic_queue_serialize | |||
| desc: "[in] non-zero value indicates a opportunistic native queue serialized kernel" | |||
There was a problem hiding this comment.
Suggested change
| desc: "[in] non-zero value indicates a opportunistic native queue serialized kernel" | |
| desc: "[in] non-zero value indicates an opportunistic native queue serialized kernel" |
aarongreig
approved these changes
May 14, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Makes short kernels that don't need to see the same global memory (or user guarantees global memory writes are complete) launch faster. See https://docs.nvidia.com/cuda/cuda-c-programming-guide/index.html#programmatic-dependent-launch-and-synchronization
Makes lots of short kernels in cutlass great again. cc @FMarno who identified this performance gap.