[RISCV] Add tests for gather scatter pass for non-recursive mul strides

Reviewed By: craig.topper

Differential Revision: https://reviews.llvm.org/D146982

diff --git a/llvm/test/CodeGen/RISCV/rvv/strided-load-store.ll b/llvm/test/CodeGen/RISCV/rvv/strided-load-store.ll
index be151e4..31fcf10 100644 (file)
--- a/llvm/test/CodeGen/RISCV/rvv/strided-load-store.ll
+++ b/llvm/test/CodeGen/RISCV/rvv/strided-load-store.ll
@@ -89,6 +89,54 @@ for.cond.cleanup:                                 ; preds = %vector.body
   ret void
 }
 
+define <vscale x 1 x i64> @gather_loopless(ptr %p, i64 %stride) {
+; CHECK-LABEL: @gather_loopless(
+; CHECK-NEXT:    [[STEP:%.*]] = call <vscale x 1 x i64> @llvm.experimental.stepvector.nxv1i64()
+; CHECK-NEXT:    [[SPLAT_INSERT:%.*]] = insertelement <vscale x 1 x i64> poison, i64 [[STRIDE:%.*]], i64 0
+; CHECK-NEXT:    [[SPLAT:%.*]] = shufflevector <vscale x 1 x i64> [[SPLAT_INSERT]], <vscale x 1 x i64> poison, <vscale x 1 x i32> zeroinitializer
+; CHECK-NEXT:    [[OFFSETS:%.*]] = mul <vscale x 1 x i64> [[STEP]], [[SPLAT]]
+; CHECK-NEXT:    [[PTRS:%.*]] = getelementptr i32, ptr [[P:%.*]], <vscale x 1 x i64> [[OFFSETS]]
+; CHECK-NEXT:    [[X:%.*]] = call <vscale x 1 x i64> @llvm.masked.gather.nxv1i64.nxv1p0(<vscale x 1 x ptr> [[PTRS]], i32 8, <vscale x 1 x i1> shufflevector (<vscale x 1 x i1> insertelement (<vscale x 1 x i1> poison, i1 true, i64 0), <vscale x 1 x i1> poison, <vscale x 1 x i32> zeroinitializer), <vscale x 1 x i64> poison)
+; CHECK-NEXT:    ret <vscale x 1 x i64> [[X]]
+;
+  %step = call <vscale x 1 x i64> @llvm.experimental.stepvector.nxv1i64()
+  %splat.insert = insertelement <vscale x 1 x i64> poison, i64 %stride, i64 0
+  %splat = shufflevector <vscale x 1 x i64> %splat.insert, <vscale x 1 x i64> poison, <vscale x 1 x i32> zeroinitializer
+  %offsets = mul <vscale x 1 x i64> %step, %splat
+  %ptrs = getelementptr i32, ptr %p, <vscale x 1 x i64> %offsets
+  %x = call <vscale x 1 x i64> @llvm.masked.gather.nxv1i64.nxv1p0(
+  <vscale x 1 x ptr> %ptrs,
+  i32 8,
+  <vscale x 1 x i1> shufflevector (<vscale x 1 x i1> insertelement (<vscale x 1 x i1> poison, i1 1, i64 0), <vscale x 1 x i1> poison, <vscale x 1 x i32> zeroinitializer),
+  <vscale x 1 x i64> poison
+  )
+  ret <vscale x 1 x i64> %x
+}
+
+define void @scatter_loopless(<vscale x 1 x i64> %x, ptr %p, i64 %stride) {
+; CHECK-LABEL: @scatter_loopless(
+; CHECK-NEXT:    [[STEP:%.*]] = call <vscale x 1 x i64> @llvm.experimental.stepvector.nxv1i64()
+; CHECK-NEXT:    [[SPLAT_INSERT:%.*]] = insertelement <vscale x 1 x i64> poison, i64 [[STRIDE:%.*]], i64 0
+; CHECK-NEXT:    [[SPLAT:%.*]] = shufflevector <vscale x 1 x i64> [[SPLAT_INSERT]], <vscale x 1 x i64> poison, <vscale x 1 x i32> zeroinitializer
+; CHECK-NEXT:    [[OFFSETS:%.*]] = mul <vscale x 1 x i64> [[STEP]], [[SPLAT]]
+; CHECK-NEXT:    [[PTRS:%.*]] = getelementptr i32, ptr [[P:%.*]], <vscale x 1 x i64> [[OFFSETS]]
+; CHECK-NEXT:    call void @llvm.masked.scatter.nxv1i64.nxv1p0(<vscale x 1 x i64> [[X:%.*]], <vscale x 1 x ptr> [[PTRS]], i32 8, <vscale x 1 x i1> shufflevector (<vscale x 1 x i1> insertelement (<vscale x 1 x i1> poison, i1 true, i64 0), <vscale x 1 x i1> poison, <vscale x 1 x i32> zeroinitializer))
+; CHECK-NEXT:    ret void
+;
+  %step = call <vscale x 1 x i64> @llvm.experimental.stepvector.nxv1i64()
+  %splat.insert = insertelement <vscale x 1 x i64> poison, i64 %stride, i64 0
+  %splat = shufflevector <vscale x 1 x i64> %splat.insert, <vscale x 1 x i64> poison, <vscale x 1 x i32> zeroinitializer
+  %offsets = mul <vscale x 1 x i64> %step, %splat
+  %ptrs = getelementptr i32, ptr %p, <vscale x 1 x i64> %offsets
+  call void @llvm.masked.scatter.nxv1i64.nxv1p0(
+  <vscale x 1 x i64> %x,
+  <vscale x 1 x ptr> %ptrs,
+  i32 8,
+  <vscale x 1 x i1> shufflevector (<vscale x 1 x i1> insertelement (<vscale x 1 x i1> poison, i1 1, i64 0), <vscale x 1 x i1> poison, <vscale x 1 x i32> zeroinitializer)
+  )
+  ret void
+}
+
 declare i64 @llvm.vscale.i64()
 declare void @llvm.masked.scatter.nxv1i64.nxv1p0(<vscale x 1 x i64>, <vscale x 1 x ptr>, i32, <vscale x 1 x i1>)
 declare <vscale x 1 x i64> @llvm.masked.gather.nxv1i64.nxv1p0(<vscale x 1 x ptr>, i32, <vscale x 1 x i1>, <vscale x 1 x i64>)