[XLA:GPU] Replace dynamic-slice with slice during autotuning #67160
+434
−46
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Outdated suggestions cannot be applied.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.
[XLA:GPU] Replace dynamic-slice with slice during autotuning
Dynamic slice requires specific values for the indices that it takes in as operands. However, under computation extraction in the autotuner, all operands are replaced with random parameter values (which then represent incorrect/invalid offsets.) In order to avoid this, we replace dynamic-slice instructions with static slice instructions indexed at 0 along all dimensions.
This CL is needed for a later CL which will add support for fusing dynamic-slice into triton fusions.
The original author of this CL is jvstokes and I rewrote it a bit.
FUTURE_COPYBARA_INTEGRATE_REVIEW=openxla/xla#12161 from jaro-sevcik:serialize-as-stream-executor-executable 818765f7b1f5f7282d3f2756b2a03ac556201a6c