[PATCH] D75981: [LV] Allow large RT checks, if they are a fraction of the scalar cost (WIP)

[Roman Lebedev via Phabricator via llvm-commits]

lebedev.ri added inline comments.


================
Comment at: llvm/lib/Transforms/Vectorize/LoopVectorizationLegality.cpp:269-270
       NumRuntimePointerChecks > VectorizerParams::RuntimeMemoryCheckThreshold;
-  if ((ThresholdReached && !Hints.allowReordering()) ||
-      PragmaThresholdReached) {
     ORE.emit([&]() {
----------------
Hm, why does `PragmaThresholdReached` not check for `Hints.allowReordering()`?
If i really asked for loop to be vectorized, why are there further limits on the sanity checks?
Or, for that matter, doesn't forcing vectorization disable those checks in the first place?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D75981/new/

https://reviews.llvm.org/D75981