[llvm] [RISCV][TTI] Implement instruction cost for vp.splice. (PR #124221)

[via llvm-commits]

llvmbot wrote:


<!--LLVM PR SUMMARY COMMENT-->

@llvm/pr-subscribers-backend-risc-v

Author: Elvis Wang (ElvisWang123)

<details>
<summary>Changes</summary>

This patch implement the instruction cost for vp.splice intrinsic.

To support type-based query for LV, adding a constant index when quering
`getShuffleCost()`. We get the same cost no matter what 
 `index` is because it only change the cost from `vslide.vx` to `vslide.vi` and
 the cost of `vslide.vx` is same as `vslide.vi` in current implementation.

---
Full diff: https://github.com/llvm/llvm-project/pull/124221.diff


2 Files Affected:

- (modified) llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp (+8) 
- (modified) llvm/test/Analysis/CostModel/RISCV/vp-intrinsics.ll (+52) 


``````````diff
diff --git a/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp b/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
index add82dc80c4290..d63680fa97d2eb 100644
--- a/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
+++ b/llvm/lib/Target/RISCV/RISCVTargetTransformInfo.cpp
@@ -1235,6 +1235,14 @@ RISCVTTIImpl::getIntrinsicInstrCost(const IntrinsicCostAttributes &ICA,
                                                   : RISCV::VMV_V_X,
                                               LT.second, CostKind);
   }
+  case Intrinsic::experimental_vp_splice: {
+    // To support type-based query from vectorizer, set the index to 0.
+    // Note that index only change the cost from vslide.vx to vslide.vi and in
+    // current implementations they have same costs.
+    return getShuffleCost(TTI::SK_Splice,
+                          cast<VectorType>(ICA.getArgTypes()[0]), {}, CostKind,
+                          0, cast<VectorType>(ICA.getReturnType()));
+  }
   }
 
   if (ST->hasVInstructions() && RetTy->isVectorTy()) {
diff --git a/llvm/test/Analysis/CostModel/RISCV/vp-intrinsics.ll b/llvm/test/Analysis/CostModel/RISCV/vp-intrinsics.ll
index 5126a6a0a3cbcd..c9798ea03b0733 100644
--- a/llvm/test/Analysis/CostModel/RISCV/vp-intrinsics.ll
+++ b/llvm/test/Analysis/CostModel/RISCV/vp-intrinsics.ll
@@ -2351,6 +2351,58 @@ define void @splat() {
   ret void
 }
 
+define void @splice() {
+; CHECK-LABEL: 'splice'
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice_nxv16i8 = call <vscale x 16 x i8> @llvm.experimental.vp.splice.nxv16i8(<vscale x 16 x i8> zeroinitializer, <vscale x 16 x i8> zeroinitializer, i32 1, <vscale x 16 x i1> zeroinitializer, i32 undef, i32 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice_nxv32i8 = call <vscale x 32 x i8> @llvm.experimental.vp.splice.nxv32i8(<vscale x 32 x i8> zeroinitializer, <vscale x 32 x i8> zeroinitializer, i32 1, <vscale x 32 x i1> zeroinitializer, i32 undef, i32 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv2i16 = call <vscale x 2 x i16> @llvm.experimental.vp.splice.nxv2i16(<vscale x 2 x i16> zeroinitializer, <vscale x 2 x i16> zeroinitializer, i32 1, <vscale x 2 x i1> zeroinitializer, i32 undef, i32 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv4i16 = call <vscale x 4 x i16> @llvm.experimental.vp.splice.nxv4i16(<vscale x 4 x i16> zeroinitializer, <vscale x 4 x i16> zeroinitializer, i32 1, <vscale x 4 x i1> zeroinitializer, i32 undef, i32 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice_nxv8i16 = call <vscale x 8 x i16> @llvm.experimental.vp.splice.nxv8i16(<vscale x 8 x i16> zeroinitializer, <vscale x 8 x i16> zeroinitializer, i32 1, <vscale x 8 x i1> zeroinitializer, i32 undef, i32 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice_nxv16i16 = call <vscale x 16 x i16> @llvm.experimental.vp.splice.nxv16i16(<vscale x 16 x i16> zeroinitializer, <vscale x 16 x i16> zeroinitializer, i32 1, <vscale x 16 x i1> zeroinitializer, i32 undef, i32 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice_nxv4i32 = call <vscale x 4 x i32> @llvm.experimental.vp.splice.nxv4i32(<vscale x 4 x i32> zeroinitializer, <vscale x 4 x i32> zeroinitializer, i32 1, <vscale x 4 x i1> zeroinitializer, i32 undef, i32 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice_nxv8i32 = call <vscale x 8 x i32> @llvm.experimental.vp.splice.nxv8i32(<vscale x 8 x i32> zeroinitializer, <vscale x 8 x i32> zeroinitializer, i32 1, <vscale x 8 x i1> zeroinitializer, i32 undef, i32 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice_nxv2i64 = call <vscale x 2 x i64> @llvm.experimental.vp.splice.nxv2i64(<vscale x 2 x i64> zeroinitializer, <vscale x 2 x i64> zeroinitializer, i32 1, <vscale x 2 x i1> zeroinitializer, i32 undef, i32 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice_nxv4i64 = call <vscale x 4 x i64> @llvm.experimental.vp.splice.nxv4i64(<vscale x 4 x i64> zeroinitializer, <vscale x 4 x i64> zeroinitializer, i32 1, <vscale x 4 x i1> zeroinitializer, i32 undef, i32 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice_nxv16i1 = call <vscale x 16 x i1> @llvm.experimental.vp.splice.nxv16i1(<vscale x 16 x i1> zeroinitializer, <vscale x 16 x i1> zeroinitializer, i32 1, <vscale x 16 x i1> zeroinitializer, i32 undef, i32 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv8i1 = call <vscale x 8 x i1> @llvm.experimental.vp.splice.nxv8i1(<vscale x 8 x i1> zeroinitializer, <vscale x 8 x i1> zeroinitializer, i32 1, <vscale x 8 x i1> zeroinitializer, i32 undef, i32 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv4i1 = call <vscale x 4 x i1> @llvm.experimental.vp.splice.nxv4i1(<vscale x 4 x i1> zeroinitializer, <vscale x 4 x i1> zeroinitializer, i32 1, <vscale x 4 x i1> zeroinitializer, i32 undef, i32 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv2i1 = call <vscale x 2 x i1> @llvm.experimental.vp.splice.nxv2i1(<vscale x 2 x i1> zeroinitializer, <vscale x 2 x i1> zeroinitializer, i32 1, <vscale x 2 x i1> zeroinitializer, i32 undef, i32 undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
+;
+; TYPEBASED-LABEL: 'splice'
+; TYPEBASED-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice_nxv16i8 = call <vscale x 16 x i8> @llvm.experimental.vp.splice.nxv16i8(<vscale x 16 x i8> zeroinitializer, <vscale x 16 x i8> zeroinitializer, i32 1, <vscale x 16 x i1> zeroinitializer, i32 undef, i32 undef)
+; TYPEBASED-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice_nxv32i8 = call <vscale x 32 x i8> @llvm.experimental.vp.splice.nxv32i8(<vscale x 32 x i8> zeroinitializer, <vscale x 32 x i8> zeroinitializer, i32 1, <vscale x 32 x i1> zeroinitializer, i32 undef, i32 undef)
+; TYPEBASED-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv2i16 = call <vscale x 2 x i16> @llvm.experimental.vp.splice.nxv2i16(<vscale x 2 x i16> zeroinitializer, <vscale x 2 x i16> zeroinitializer, i32 1, <vscale x 2 x i1> zeroinitializer, i32 undef, i32 undef)
+; TYPEBASED-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv4i16 = call <vscale x 4 x i16> @llvm.experimental.vp.splice.nxv4i16(<vscale x 4 x i16> zeroinitializer, <vscale x 4 x i16> zeroinitializer, i32 1, <vscale x 4 x i1> zeroinitializer, i32 undef, i32 undef)
+; TYPEBASED-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice_nxv8i16 = call <vscale x 8 x i16> @llvm.experimental.vp.splice.nxv8i16(<vscale x 8 x i16> zeroinitializer, <vscale x 8 x i16> zeroinitializer, i32 1, <vscale x 8 x i1> zeroinitializer, i32 undef, i32 undef)
+; TYPEBASED-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice_nxv16i16 = call <vscale x 16 x i16> @llvm.experimental.vp.splice.nxv16i16(<vscale x 16 x i16> zeroinitializer, <vscale x 16 x i16> zeroinitializer, i32 1, <vscale x 16 x i1> zeroinitializer, i32 undef, i32 undef)
+; TYPEBASED-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice_nxv4i32 = call <vscale x 4 x i32> @llvm.experimental.vp.splice.nxv4i32(<vscale x 4 x i32> zeroinitializer, <vscale x 4 x i32> zeroinitializer, i32 1, <vscale x 4 x i1> zeroinitializer, i32 undef, i32 undef)
+; TYPEBASED-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice_nxv8i32 = call <vscale x 8 x i32> @llvm.experimental.vp.splice.nxv8i32(<vscale x 8 x i32> zeroinitializer, <vscale x 8 x i32> zeroinitializer, i32 1, <vscale x 8 x i1> zeroinitializer, i32 undef, i32 undef)
+; TYPEBASED-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice_nxv2i64 = call <vscale x 2 x i64> @llvm.experimental.vp.splice.nxv2i64(<vscale x 2 x i64> zeroinitializer, <vscale x 2 x i64> zeroinitializer, i32 1, <vscale x 2 x i1> zeroinitializer, i32 undef, i32 undef)
+; TYPEBASED-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %splice_nxv4i64 = call <vscale x 4 x i64> @llvm.experimental.vp.splice.nxv4i64(<vscale x 4 x i64> zeroinitializer, <vscale x 4 x i64> zeroinitializer, i32 1, <vscale x 4 x i1> zeroinitializer, i32 undef, i32 undef)
+; TYPEBASED-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %splice_nxv16i1 = call <vscale x 16 x i1> @llvm.experimental.vp.splice.nxv16i1(<vscale x 16 x i1> zeroinitializer, <vscale x 16 x i1> zeroinitializer, i32 1, <vscale x 16 x i1> zeroinitializer, i32 undef, i32 undef)
+; TYPEBASED-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv8i1 = call <vscale x 8 x i1> @llvm.experimental.vp.splice.nxv8i1(<vscale x 8 x i1> zeroinitializer, <vscale x 8 x i1> zeroinitializer, i32 1, <vscale x 8 x i1> zeroinitializer, i32 undef, i32 undef)
+; TYPEBASED-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv4i1 = call <vscale x 4 x i1> @llvm.experimental.vp.splice.nxv4i1(<vscale x 4 x i1> zeroinitializer, <vscale x 4 x i1> zeroinitializer, i32 1, <vscale x 4 x i1> zeroinitializer, i32 undef, i32 undef)
+; TYPEBASED-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %splice_nxv2i1 = call <vscale x 2 x i1> @llvm.experimental.vp.splice.nxv2i1(<vscale x 2 x i1> zeroinitializer, <vscale x 2 x i1> zeroinitializer, i32 1, <vscale x 2 x i1> zeroinitializer, i32 undef, i32 undef)
+; TYPEBASED-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: ret void
+;
+  %splice_nxv16i8 = call <vscale x 16 x i8> @llvm.experimental.vp.splice.nxv16i8(<vscale x 16 x i8> zeroinitializer, <vscale x 16 x i8> zeroinitializer, i32 1, <vscale x 16 x i1> zeroinitializer, i32 undef, i32 undef)
+  %splice_nxv32i8 = call <vscale x 32 x i8> @llvm.experimental.vp.splice.nxv32i8(<vscale x 32 x i8> zeroinitializer, <vscale x 32 x i8> zeroinitializer, i32 1, <vscale x 32 x i1> zeroinitializer, i32 undef, i32 undef)
+  %splice_nxv2i16 = call <vscale x 2 x i16> @llvm.experimental.vp.splice.nxv2i16(<vscale x 2 x i16> zeroinitializer, <vscale x 2 x i16> zeroinitializer, i32 1, <vscale x 2 x i1> zeroinitializer, i32 undef, i32 undef)
+  %splice_nxv4i16 = call <vscale x 4 x i16> @llvm.experimental.vp.splice.nxv4i16(<vscale x 4 x i16> zeroinitializer, <vscale x 4 x i16> zeroinitializer, i32 1, <vscale x 4 x i1> zeroinitializer, i32 undef, i32 undef)
+  %splice_nxv8i16 = call <vscale x 8 x i16> @llvm.experimental.vp.splice.nxv8i16(<vscale x 8 x i16> zeroinitializer, <vscale x 8 x i16> zeroinitializer, i32 1, <vscale x 8 x i1> zeroinitializer, i32 undef, i32 undef)
+  %splice_nxv16i16 = call <vscale x 16 x i16> @llvm.experimental.vp.splice.nxv16i16(<vscale x 16 x i16> zeroinitializer, <vscale x 16 x i16> zeroinitializer, i32 1, <vscale x 16 x i1> zeroinitializer, i32 undef, i32 undef)
+  %splice_nxv4i32 = call <vscale x 4 x i32> @llvm.experimental.vp.splice.nxv4i32(<vscale x 4 x i32> zeroinitializer, <vscale x 4 x i32> zeroinitializer, i32 1, <vscale x 4 x i1> zeroinitializer, i32 undef, i32 undef)
+  %splice_nxv8i32 = call <vscale x 8 x i32> @llvm.experimental.vp.splice.nxv8i32(<vscale x 8 x i32> zeroinitializer, <vscale x 8 x i32> zeroinitializer, i32 1, <vscale x 8 x i1> zeroinitializer, i32 undef, i32 undef)
+  %splice_nxv2i64 = call <vscale x 2 x i64> @llvm.experimental.vp.splice.nxv2i64(<vscale x 2 x i64> zeroinitializer, <vscale x 2 x i64> zeroinitializer, i32 1, <vscale x 2 x i1> zeroinitializer, i32 undef, i32 undef)
+  %splice_nxv4i64 = call <vscale x 4 x i64> @llvm.experimental.vp.splice.nxv4i64(<vscale x 4 x i64> zeroinitializer, <vscale x 4 x i64> zeroinitializer, i32 1, <vscale x 4 x i1> zeroinitializer, i32 undef, i32 undef)
+  %splice_nxv16i1 = call <vscale x 16 x i1> @llvm.experimental.vp.splice.nxv16i1(<vscale x 16 x i1> zeroinitializer, <vscale x 16 x i1> zeroinitializer, i32 1, <vscale x 16 x i1> zeroinitializer, i32 undef, i32 undef)
+  %splice_nxv8i1 =  call <vscale x 8 x i1> @llvm.experimental.vp.splice.nxv8i1(<vscale x 8 x i1> zeroinitializer, <vscale x 8 x i1> zeroinitializer, i32 1, <vscale x 8 x i1> zeroinitializer, i32 undef, i32 undef)
+  %splice_nxv4i1 = call <vscale x 4 x i1> @llvm.experimental.vp.splice.nxv4i1(<vscale x 4 x i1> zeroinitializer, <vscale x 4 x i1> zeroinitializer, i32 1, <vscale x 4 x i1> zeroinitializer, i32 undef, i32 undef)
+  %splice_nxv2i1 = call <vscale x 2 x i1> @llvm.experimental.vp.splice.nxv2i1(<vscale x 2 x i1> zeroinitializer, <vscale x 2 x i1> zeroinitializer, i32 1, <vscale x 2 x i1> zeroinitializer, i32 undef, i32 undef)
+  ret void
+}
+
 declare <2 x i8> @llvm.vp.add.v2i8(<2 x i8>, <2 x i8>, <2 x i1>, i32)
 declare <4 x i8> @llvm.vp.add.v4i8(<4 x i8>, <4 x i8>, <4 x i1>, i32)
 declare <8 x i8> @llvm.vp.add.v8i8(<8 x i8>, <8 x i8>, <8 x i1>, i32)

``````````

</details>


https://github.com/llvm/llvm-project/pull/124221