C++: Fix two more dataflow-related joins #18307
Merged
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Outdated suggestions cannot be applied.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.
The join order fix in #18233 turned out to be slightly incomplete. This can actually be seen in the 'after' tuple counts, but I decided to ignore it since I couldn't find a repository where it was too bad 🙈
Well, jokes on me, because there was actually a repository where this happened (thanks for flagging this up to me, @jketema):
(I stopped the run before it was finished)
This PR fixes that join by ordering the
IRGuards::IRGuardCondition.controls
join before the second join withValueNumberingInternal::tvalueNumber
which is clearly better:Ideally, I would have liked to order the joins with
ExprNode.getConvertedExpr
before the join withNode.hasIndexInBlock
, but I couldn't quite make that work out. In any case, this join order is still much better than before.The same project also showed an unfortunate materialization of
ensuresEq
:Both of these have been removed in f351558.