[PATCH] D33946: [InlineCost] Find identical loads in the callee

Mehdi AMINI via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Mon Jun 26 23:05:41 PDT 2017


mehdi_amini added a comment.

The description says "The motivation example is a fully unrolled 3x3 matrix multiplication. It loads every data in matrix a and b three times because of the Stores between them", but from reading the code I don't see how this case is handled since it seems that you bail as soon as you see a store (or a call)? What did I miss? Is the description still up-to-date?



================
Comment at: lib/Analysis/InlineCost.cpp:265
   unsigned SROACostSavingsLost;
+  unsigned CSELoadCostSavings;
 
----------------
chandlerc wrote:
> Similarly, I would call this `LoadCSECostSavings`.
Why do we have both `LoadEliminationCostSavings` and `LoadEliminationCost` ?


Repository:
  rL LLVM

https://reviews.llvm.org/D33946





More information about the llvm-commits mailing list