[Transforms] Add constant_tensors_folding pass #74
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
niuxiaog
force-pushed
the
xgniu/constant_weights_folding
branch
from
acf3ae8
to
94f2813
Compare
niuxiaog
force-pushed
the
xgniu/constant_weights_folding
branch
from
a0ddebe
to
d7663a5
Compare
This reverts commit 22c3d76.
niuxiaog
force-pushed
the
xgniu/constant_weights_folding
branch
from
387523a
to
d8d2d79
Compare
niuxiaog
requested review from
zhczhong,
Menooker,
AndreyPavlenko,
ciyongch and
ZhennanQin
and removed request for
zhczhong
niuxiaog
changed the title
ciyongch
reviewed
Sep 18, 2024
| void ConstantSubgraphAnalysis::runOnOperation() { | ||
| Operation *op = getOperation(); | ||
| auto &func = | ||
| op->getRegions().front().getBlocks().front().getOperations().front(); |
There was a problem hiding this comment.
Is there any shortcut for this kind of operation?
| @@ -53,6 +53,8 @@ void populateTensorPasses(mlir::OpPassManager &pm) { | |||
| // todo: padding propagation pass | |||
| // todo: layout propagation pass | |||
| // todo: tensor constant propagation pass | |||
| pm.addPass(createConstantSubgraphAnalysisPass()); | |||
| pm.addPass(createConstantTensorFoldingPass()); | |||
There was a problem hiding this comment.
Shall we combine these two passes into one, and provide an option to do the analysis only is needed?
There was a problem hiding this comment.
I feel the same. Maybe we should put the analysis into the pass, unless the const-subgraph-analysis is needed by more than one pass.
There was a problem hiding this comment.
OK, I will put them into to one.
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Outdated suggestions cannot be applied.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.
This PR implements constant_tensors_folding pass (RFC: PR #183) in issue #56 and issue #146:
entrywill be split into two functions:entry()andruntime_fold(). Theruntime_fold()function contains the constant operations whose input tensors and output tensors are all constant. The newentry()function contains other operations that depend on the variable values and folded constant tensors.compile_time_fold()can be enabled to fold constant tensors during compile time.