[PATCH] D103153: [InstCombine] Add fold for extracting known elements from a stepvector

[Caroline via Phabricator via llvm-commits]

CarolineConcatto created this revision.
Herald added a subscriber: hiraditya.
CarolineConcatto requested review of this revision.
Herald added a project: LLVM.
Herald added a subscriber: llvm-commits.

This patch allows folding stepvector + extract to the lane when the lane is
lower than the minimum size of the scalable vector. This fold is possible
because lane X of a stepvector is also X!
For instance, extracting element 3 of a <vscale x 4 x i64>stepvector is 3.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D103153

Files:
  llvm/lib/Transforms/InstCombine/InstCombineVectorOps.cpp
  llvm/test/Transforms/InstCombine/vscale_extractelement.ll


Index: llvm/test/Transforms/InstCombine/vscale_extractelement.ll
===================================================================
--- llvm/test/Transforms/InstCombine/vscale_extractelement.ll
+++ llvm/test/Transforms/InstCombine/vscale_extractelement.ll
@@ -183,3 +183,57 @@
   %E = extractelement <vscale x 4 x i32> %vec_int, i32 2147483647
   ret i32 %E
 }
+
+; Step vector optimization
+
+define i64 @ext_lane0_from_stepvec() {
+; CHECK-LABEL: @ext_lane0_from_stepvec(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    ret i64 0
+;
+entry:
+  %0 = call <vscale x 4 x i64> @llvm.experimental.stepvector.nxv4i64()
+  %1 = extractelement <vscale x 4 x i64> %0, i32 0
+  ret i64 %1
+}
+
+define i64 @ext_lane_out_of_range_from_stepvec() {
+; CHECK-LABEL: @ext_lane_out_of_range_from_stepvec(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.experimental.stepvector.nxv4i64()
+; CHECK-NEXT:    [[TMP1:%.*]] = extractelement <vscale x 4 x i64> [[TMP0]], i32 5
+; CHECK-NEXT:    ret i64 [[TMP1]]
+;
+entry:
+  %0 = call <vscale x 4 x i64> @llvm.experimental.stepvector.nxv4i64()
+  %1 = extractelement <vscale x 4 x i64> %0, i32 5
+  ret i64 %1
+}
+
+define i64 @ext_lane_invalid_from_stepvec() {
+; CHECK-LABEL: @ext_lane_invalid_from_stepvec(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.experimental.stepvector.nxv4i64()
+; CHECK-NEXT:    [[TMP1:%.*]] = extractelement <vscale x 4 x i64> [[TMP0]], i32 -1
+; CHECK-NEXT:    ret i64 [[TMP1]]
+;
+entry:
+  %0 = call <vscale x 4 x i64> @llvm.experimental.stepvector.nxv4i64()
+  %1 = extractelement <vscale x 4 x i64> %0, i32 -1
+  ret i64 %1
+}
+
+define i64 @ext_lane_unknown_from_stepvec(i32 %v) {
+; CHECK-LABEL: @ext_lane_unknown_from_stepvec(
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[TMP0:%.*]] = call <vscale x 4 x i64> @llvm.experimental.stepvector.nxv4i64()
+; CHECK-NEXT:    [[TMP1:%.*]] = extractelement <vscale x 4 x i64> [[TMP0]], i32 [[V:%.*]]
+; CHECK-NEXT:    ret i64 [[TMP1]]
+;
+entry:
+  %0 = call <vscale x 4 x i64> @llvm.experimental.stepvector.nxv4i64()
+  %1 = extractelement <vscale x 4 x i64> %0, i32 %v
+  ret i64 %1
+}
+
+declare <vscale x 4 x i64> @llvm.experimental.stepvector.nxv4i64();
Index: llvm/lib/Transforms/InstCombine/InstCombineVectorOps.cpp
===================================================================
--- llvm/lib/Transforms/InstCombine/InstCombineVectorOps.cpp
+++ llvm/lib/Transforms/InstCombine/InstCombineVectorOps.cpp
@@ -347,6 +347,22 @@
     ElementCount EC = EI.getVectorOperandType()->getElementCount();
     unsigned NumElts = EC.getKnownMinValue();
 
+    if (IntrinsicInst *II = dyn_cast<IntrinsicInst>(SrcVec)) {
+      if (Intrinsic::ID IID = II->getIntrinsicID()) {
+        switch (IID) {
+        case Intrinsic::experimental_stepvector:
+          if (IndexC->getZExtValue() <= NumElts) {
+            auto *Idx = ConstantInt::get(
+                cast<VectorType>(EI.getVectorOperandType())->getScalarType(),
+                IndexC->getZExtValue());
+            return replaceInstUsesWith(EI, Idx);
+          }
+          break;
+        default:
+          break;
+        }
+      }
+    }
     // InstSimplify should handle cases where the index is invalid.
     // For fixed-length vector, it's invalid to extract out-of-range element.
     if (!EC.isScalable() && IndexC->getValue().uge(NumElts))


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D103153.347913.patch
Type: text/x-patch
Size: 3393 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20210526/062ad725/attachment.bin>