[PATCH] D125754: [LoopUnroll] Avoid branch on poison for runtime unroll with multiple exits

[Nikita Popov via Phabricator via llvm-commits]

nikic added inline comments.


================
Comment at: llvm/lib/Transforms/Utils/LoopUnrollRuntime.cpp:755
   BasicBlock *UnrollingLoop = UseEpilogRemainder ? NewPreHeader : PrologExit;
+  // Freeze the condition if there are other exits before the latch that may
+  // cause the latch exit branch to never be executed.
----------------
reames wrote:
> Its not clear to me that you need this.  The condition being branched on should always be derived from the SCEV computed trip count.  After D124910, shouldn't we be assured that is non-poisonous?  Or at least, only poisonous if the first input is poison?  
> 
> Your code here is completely reasonable, except that I don't quite understand the case in which is matters.  :)
Runtime loop unrolling doesn't use the backedge taken count (which is what D124910 is about), it works only on the latch exit count. So if you have two exits at `n` and `m` (with `m` being the latch exit), runtime unroll will unroll using `m`, not `umin(n, m)`. That's why we need to handle this explicitly, it wouldn't be going through the umin_seq code.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D125754/new/

https://reviews.llvm.org/D125754