Mfma prune #1577
Open
Mfma prune #1577
+57
−28
Conversation
This file contains 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
would cause register spills. Current value is hardcoded to 256, but should be generalized.
krzysz00
reviewed
Jul 19, 2024
| int64_t numOutputVectorElements = accelParams.numOutputVectorElements(); | ||
|
|
||
| // would be best to have register count be a part of arch, is not necessarily totalVGPRPerEu | ||
| if(numOutputVectorElements > 256) { |
There was a problem hiding this comment.
Let's go ahead and make that an arch field, then
I am doing another tuning run this weekend and testing it further, atm cannot tell. I should have results by Monday or Tuesday for that question. |
|
Did we ever get the experiment run? |
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Outdated suggestions cannot be applied.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.
Try pruning configs that take too many registers as likely non-performant. This adds a tentative check in the XDLOPs
specificCouldBePerformant()function to see if the output size (numOutputVectorElements) is greater than the number of available registers.