[PATCH] D63459: Loop Cache Analysis

[Florian Hahn via Phabricator via llvm-commits]

fhahn added a comment.

Can you re-run clang-format on the latest version of the patch? I think it would be good to get an in-tree user of this soon, to make sure the modeling works as expected on real hardware/benchmarks. Do you have a timeline to get this used? I think you mentioned LoopFusion as one of the first planned users?



================
Comment at: llvm/include/llvm/Analysis/LoopCacheAnalysis.h:198
+
+  const ArrayRef<LoopCacheCostTy> getLoopCosts() const {
+    return LoopCosts;
----------------
IIRC this returns the ordered loop costs, right? Please document.


================
Comment at: llvm/test/Analysis/LoopCacheAnalysis/PowerPC/matvecmul.ll:17
+
+; CHECK-DAG: Loop 'k_loop' has cost = 30000000000
+; CHECK-DAG: Loop 'j_loop' has cost = 30000000000
----------------
IIRC the costs should be ordered. If so, please use check next for entries with differing costs. (here and at other places)


Repository:
  rL LLVM

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D63459/new/

https://reviews.llvm.org/D63459