[PATCH] D102023: [SLP]Do not count perfect diamond matches for gathers several times.

[Alexey Bataev via Phabricator via llvm-commits]

ABataev created this revision.
ABataev added reviewers: spatel, RKSimon, dtemirbulatov, anton-afanasyev.
Herald added a subscriber: hiraditya.
ABataev requested review of this revision.
Herald added a project: LLVM.

Need to remove the old code for avoiding double counting of the gather
nodes with perfect diamond matches within the tree after we started
detecting perfect/shuffled matching in one of the previous patches. We
may skip the cost for such nodes completely.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D102023

Files:
  llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp
  llvm/test/Transforms/SLPVectorizer/AArch64/gather-cost.ll


Index: llvm/test/Transforms/SLPVectorizer/AArch64/gather-cost.ll
===================================================================
--- llvm/test/Transforms/SLPVectorizer/AArch64/gather-cost.ll
+++ llvm/test/Transforms/SLPVectorizer/AArch64/gather-cost.ll
@@ -10,7 +10,7 @@
 ; REMARK-LABEL: Function: gather_multiple_use
 ; REMARK:       Args:
 ; REMARK-NEXT:    - String: 'Vectorized horizontal reduction with cost '
-; REMARK-NEXT:    - Cost: '-16'
+; REMARK-NEXT:    - Cost: '-7'
 ;
 ; REMARK-NOT: Function: gather_load
 
Index: llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp
===================================================================
--- llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp
+++ llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp
@@ -4233,27 +4233,6 @@
   for (unsigned I = 0, E = VectorizableTree.size(); I < E; ++I) {
     TreeEntry &TE = *VectorizableTree[I].get();
 
-    // We create duplicate tree entries for gather sequences that have multiple
-    // uses. However, we should not compute the cost of duplicate sequences.
-    // For example, if we have a build vector (i.e., insertelement sequence)
-    // that is used by more than one vector instruction, we only need to
-    // compute the cost of the insertelement instructions once. The redundant
-    // instructions will be eliminated by CSE.
-    //
-    // We should consider not creating duplicate tree entries for gather
-    // sequences, and instead add additional edges to the tree representing
-    // their uses. Since such an approach results in fewer total entries,
-    // existing heuristics based on tree size may yield different results.
-    //
-    if (TE.State == TreeEntry::NeedToGather &&
-        std::any_of(std::next(VectorizableTree.begin(), I + 1),
-                    VectorizableTree.end(),
-                    [TE](const std::unique_ptr<TreeEntry> &EntryPtr) {
-                      return EntryPtr->State == TreeEntry::NeedToGather &&
-                             EntryPtr->isSame(TE.Scalars);
-                    }))
-      continue;
-
     InstructionCost C = getEntryCost(&TE);
     Cost += C;
     LLVM_DEBUG(dbgs() << "SLP: Adding cost " << C


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D102023.343494.patch
Type: text/x-patch
Size: 2171 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20210506/66fb2fee/attachment.bin>