wazevo: fuzz, fix regalloc ensure phi spills are consistent across siblings. #1845
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
Signed-off-by: Edoardo Vacchi <[email protected]>
Signed-off-by: Edoardo Vacchi <[email protected]>
Signed-off-by: Edoardo Vacchi <[email protected]>
Signed-off-by: Edoardo Vacchi <[email protected]>
Signed-off-by: Edoardo Vacchi <[email protected]>
Signed-off-by: Edoardo Vacchi <[email protected]>
evacchi
changed the title
evacchi
commented
Nov 23, 2023
There was a problem hiding this comment.
Huh I guess this was committed by mistake
evacchi
commented
Nov 23, 2023
| @@ -719,7 +720,7 @@ func Test1825(t *testing.T) { | |||
| }) | |||
| } | |||
|
|
|||
| // Test1825 tests that lowerFcopysignImpl allocates correctly the temporary registers. | |||
| // Test1826 tests that lowerFcopysignImpl allocates correctly the temporary registers. | |||
evacchi
commented
Nov 23, 2023
| @@ -734,3 +735,20 @@ func Test1826(t *testing.T) { | |||
| require.Equal(t, uint64(0), m.Globals[0].ValHi) | |||
| }) | |||
| } | |||
|
|
|||
| // Test1845 tests regalloc ensures that when one predecessor Y1 for a block X | |||
There was a problem hiding this comment.
Suggested change
| // Test1845 tests regalloc ensures that when one predecessor Y1 for a block X | |
| // Test1845 tests regalloc and ensures that when one predecessor Y1 for a block X |
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Outdated suggestions cannot be applied.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.
Fuzz case. Consider:
graph TD blk1 --> blk3 blk2 --> blk3where blk1 spills a given vReg, say
v183like in this case, and blk2 does not. Then if blk3 allocates its registers starting from blk1, then it will expect v183 to be on the stack.We need to ensure that
blk2correctly spillsv183even if locally it wouldn't be necessary, so that it can be unspilled byblk3regardless if the flow originates inblk1orblk2.This is an attempt to fix this, and it works in the given fuzz case without breaking any other tests, but it could be horribly misguided.
Feel free to close the PR if the solution is horrifying. I suspect I am missing some bits.