Add LLVM STATISTIC tracking for cached and recomputed values#2749
Open
brucechanglongxu wants to merge 1 commit intoEnzymeAD:mainfrom
Open
Add LLVM STATISTIC tracking for cached and recomputed values#2749brucechanglongxu wants to merge 1 commit intoEnzymeAD:mainfrom
brucechanglongxu wants to merge 1 commit intoEnzymeAD:mainfrom
Conversation
Track the number of values cached vs recomputed in the reverse pass using LLVM's STATISTIC macro. Statistics are incremented in computeMinCache() where the min-cut analysis decides which intermediate values must be cached and which can be safely recomputed. Addresses EnzymeAD#962.
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Adds LLVM
STATISTICmacros toGradientUtils.cppto track how many values get cached vs recomputed in the reverse pass. The counters are incremented incomputeMinCache()at both decision points — when a value must be cached due to single-use needs, and when the min-cut classifies intermediates.Viewable via
opt -stats.Partial fix for #962 (covers cached/recomputed tracking; GPU gradient accumulation, allocation merging, etc. could follow separately).