[PATCH] D32086: [DAG] Improve store merge candidate pruning.

[Nirav Dave via Phabricator via llvm-commits]

niravd updated this revision to Diff 95564.
niravd added a comment.

Update with Sanjay's suggestions and rebased with to show additional testcase


https://reviews.llvm.org/D32086

Files:
  lib/CodeGen/SelectionDAG/DAGCombiner.cpp
  test/CodeGen/AArch64/arm64-abi.ll
  test/CodeGen/X86/MergeConsecutiveStores.ll


Index: test/CodeGen/X86/MergeConsecutiveStores.ll
===================================================================
--- test/CodeGen/X86/MergeConsecutiveStores.ll
+++ test/CodeGen/X86/MergeConsecutiveStores.ll
@@ -596,14 +596,8 @@
   store i8 3, i8* %p3
   ret void
 ; CHECK-LABEL: almost_consecutive_stores
-; CHECK-DAG: movb	$0, (%rdi)
-; CHECK-DAG: movb	$1, 42(%rdi)
-; CHECK-DAG: movb	$2, 2(%rdi)
-; CHECK-DAG: movb	$3, 3(%rdi)
+; CHECK-DAG: movb $0, (%rdi)
+; CHECK-DAG: movb $1, 42(%rdi)
+; CHECK-DAG: movw $770, 2(%rdi)
 ; CHECK: retq
-
-; We should able to merge the final two stores into a 16-bit store
-; FIXMECHECK-DAG: movw $770, 2(%rdi)
-
-
 }
Index: test/CodeGen/AArch64/arm64-abi.ll
===================================================================
--- test/CodeGen/AArch64/arm64-abi.ll
+++ test/CodeGen/AArch64/arm64-abi.ll
@@ -43,9 +43,7 @@
 ; CHECK-LABEL: i8i16caller
 ; The 8th, 9th, 10th and 11th arguments are passed at sp, sp+2, sp+4, sp+5.
 ; They are i8, i16, i8 and i8.
-; CHECK-DAG: strb {{w[0-9]+}}, [sp, #5]
-; CHECK-DAG: strb {{w[0-9]+}}, [sp, #4]
-; CHECK-DAG: strh {{w[0-9]+}}, [sp, #2]
+; CHECK-DAG: stur {{w[0-9]+}}, [sp, #2]
 ; CHECK-DAG: strb {{w[0-9]+}}, [sp]
 ; CHECK: bl
 ; FAST-LABEL: i8i16caller
Index: lib/CodeGen/SelectionDAG/DAGCombiner.cpp
===================================================================
--- lib/CodeGen/SelectionDAG/DAGCombiner.cpp
+++ lib/CodeGen/SelectionDAG/DAGCombiner.cpp
@@ -12375,6 +12375,27 @@
               return LHS.OffsetFromBase < RHS.OffsetFromBase;
             });
 
+  // Store Merge attempts to merge the lowest stores. This generally
+  // works out as if successful, as the remaining stores are checked
+  // after the first collection of stores is merged. However in the
+  // case that a non-mergeable store is found first (e.g., p[-2],
+  // p[0],p[1],p[2],p[3]} we would fail and miss the subsequent
+  // mergable cases. To prevent this, we prune such stores from the
+  // front of StoreNodes here.
+
+  unsigned StartIdx = 0;
+  while ((StartIdx + 1 < StoreNodes.size()) &&
+         StoreNodes[StartIdx].OffsetFromBase + ElementSizeBytes !=
+             StoreNodes[StartIdx + 1].OffsetFromBase)
+    ++StartIdx;
+
+  // Bail if we've don't have enough Candidates to do a merge.
+  if (StartIdx + 1 >= StoreNodes.size())
+    return false;
+
+  if (StartIdx)
+    StoreNodes.erase(StoreNodes.begin(), StoreNodes.begin() + StartIdx);
+
   // Scan the memory operations on the chain and find the first non-consecutive
   // store memory address.
   unsigned NumConsecutiveStores = 0;


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D32086.95564.patch
Type: text/x-patch
Size: 2580 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20170418/ec0d745b/attachment.bin>