feat[venom]: add codesize optimization pass #4333
Open
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
harkal
reviewed
Oct 29, 2024
Comment on lines
+12
to
+15
| i = 0 | ||
| while i < len(bb.instructions): | ||
| inst = bb.instructions[i] | ||
| i += 1 |
There was a problem hiding this comment.
Suggested change
| i = 0 | |
| while i < len(bb.instructions): | |
| inst = bb.instructions[i] | |
| i += 1 | |
| for inst in bb.instructions: |
harkal
reviewed
Oct 29, 2024
|
|
||
| val = op.value | ||
|
|
||
| if val == (2**256 - 1): |
There was a problem hiding this comment.
Make use of constants that we have for numbers like these
harkal
reviewed
Oct 29, 2024
|
|
||
| # transform things like 0xffff...01 to (not 0xfe) | ||
| binz = bin(val)[2:] | ||
| if (ix := binz.find("0")) > 8: # `not` is 1 byte |
There was a problem hiding this comment.
Maybe, extract 8 and 24 to constants like BITS_FOR_SHL and BITS_FOR_NOT?
harkal
reviewed
Oct 29, 2024
Comment on lines
+41
to
+46
| inst.opcode = "shl" | ||
| # sanity check | ||
| assert (val >> ix) << ix == val, val | ||
| assert (val >> ix) & 1 == 1 | ||
|
|
||
| inst.operands = [IRLiteral(val >> ix), IRLiteral(ix)] |
There was a problem hiding this comment.
Suggested change
| inst.opcode = "shl" | |
| # sanity check | |
| assert (val >> ix) << ix == val, val | |
| assert (val >> ix) & 1 == 1 | |
| inst.operands = [IRLiteral(val >> ix), IRLiteral(ix)] | |
| # sanity check | |
| assert (val >> ix) << ix == val, val | |
| assert (val >> ix) & 1 == 1 | |
| inst.opcode = "shl" | |
| inst.operands = [IRLiteral(val >> ix), IRLiteral(ix)] |
harkal
reviewed
Oct 29, 2024
| for bb in self.function.get_basic_blocks(): | ||
| self._process_bb(bb) | ||
|
|
||
| def _process_bb(self, bb): |
There was a problem hiding this comment.
For readability, it might be a good thing to add comments to "separate" the three rules the processing applies, or make them separate methods?
harkal
requested changes
Oct 29, 2024
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.
What I did
basic codesize optimization which strengthens large literals to evm computations which are shorter
How I did it
How to verify it
saves 450 bytes (2.5%) on https://gist.github.com/charles-cooper/eed70340aee2a47478ca0fc2ea6d5140
Commit message
Commit message for the final, squashed PR. (Optional, but reviewers will appreciate it! Please see our commit message style guide for what we would ideally like to see in a commit message.)
Description for the changelog
Cute Animal Picture