[PATCH] D29641: [SLP] Fix for PR31847: Assertion failed: (isLoopInvariant(Operands[i], L) && "SCEVAddRecExpr operand is not loop-invariant!")

Michael Kuperstein via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Tue Feb 7 13:24:10 PST 2017


mkuper added a comment.

Test case?

Also, this looks a bit weird. Could you explain what exactly is going wrong?
I'm asking because this somehow seems like it's the wrong granularity for this - after any change, you invalidate the disposition of the innermost loop that contains the basic block in which the root/seed lives. I'm not sure why this is the right thing to do.


https://reviews.llvm.org/D29641





More information about the llvm-commits mailing list