[PATCH] D112302: [TTI] `BasicTTIImplBase::getInterleavedMemoryOpCost()`: fix load discounting

[Roman Lebedev via Phabricator via llvm-commits]

lebedev.ri created this revision.
lebedev.ri added a reviewer: RKSimon.
lebedev.ri added a project: LLVM.
lebedev.ri requested review of this revision.

The math here is:
Cost of 1 load = cost of n loads / n
Cost of live loads = num live loads * Cost of 1 load
Cost of live loads = num live loads * (cost of n loads / n)
Cost of live loads = cost of n loads * (num live loads / n)

But, all the variables here are integers,
and integer division rounds down,
but this calculation clearly expects float semantics.

Instead multiply upfront, and then perform round-up-division.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D112302

Files:
  llvm/include/llvm/CodeGen/BasicTTIImpl.h
  llvm/test/Transforms/LoopVectorize/AArch64/interleaved_cost.ll


Index: llvm/test/Transforms/LoopVectorize/AArch64/interleaved_cost.ll
===================================================================
--- llvm/test/Transforms/LoopVectorize/AArch64/interleaved_cost.ll
+++ llvm/test/Transforms/LoopVectorize/AArch64/interleaved_cost.ll
@@ -168,7 +168,7 @@
 ; gaps.
 ;
 ; VF_2-LABEL: Checking a loop in "i64_factor_8"
-; VF_2:         Found an estimated cost of 6 for VF 2 For instruction: %tmp2 = load i64, i64* %tmp0, align 8
+; VF_2:         Found an estimated cost of 10 for VF 2 For instruction: %tmp2 = load i64, i64* %tmp0, align 8
 ; VF_2-NEXT:    Found an estimated cost of 0 for VF 2 For instruction: %tmp3 = load i64, i64* %tmp1, align 8
 ; VF_2-NEXT:    Found an estimated cost of 7 for VF 2 For instruction: store i64 0, i64* %tmp0, align 8
 ; VF_2-NEXT:    Found an estimated cost of 7 for VF 2 For instruction: store i64 0, i64* %tmp1, align 8
Index: llvm/include/llvm/CodeGen/BasicTTIImpl.h
===================================================================
--- llvm/include/llvm/CodeGen/BasicTTIImpl.h
+++ llvm/include/llvm/CodeGen/BasicTTIImpl.h
@@ -1214,7 +1214,7 @@
     //
     // TODO: Note that legalization can turn masked loads/stores into unmasked
     // (legalized) loads/stores. This can be reflected in the cost.
-    if (VecTySize > VecTyLTSize) {
+    if (Cost.isValid() && VecTySize > VecTyLTSize) {
       // The number of loads of a legal type it will take to represent a load
       // of the unlegalized vector type.
       unsigned NumLegalInsts = divideCeil(VecTySize, VecTyLTSize);
@@ -1231,7 +1231,8 @@
 
       // Scale the cost of the load by the fraction of legal instructions that
       // will be used.
-      Cost *= UsedInsts.count() / NumLegalInsts;
+      Cost = divideCeil(UsedInsts.count() * Cost.getValue().getValue(),
+                        NumLegalInsts);
     }
 
     // Then plus the cost of interleave operation.


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D112302.381493.patch
Type: text/x-patch
Size: 1911 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20211022/04573790/attachment.bin>