[LLVM][tests/CodeGen/AArch64] Convert instances of ConstantExpr based splats to use splat().

This is mostly NFC but some output does change due to consistently
inserting into poison rather than undef and using i64 as the index
type for inserts.

Author: paulwalker-arm

llvm/test/CodeGen/AArch64/complex-deinterleaving-splat-scalable.ll

@@ -40,11 +40,11 @@ entry:
   %7 = fmul fast <vscale x 2 x double> %2, %0
   %8 = fmul fast <vscale x 2 x double> %3, %1
   %9 = fsub fast <vscale x 2 x double> %7, %8
-  %10 = fmul fast <vscale x 2 x double> %9, shufflevector (<vscale x 2 x double> insertelement (<vscale x 2 x double> poison, double 3.000000e+00, i64 0), <vscale x 2 x double> poison, <vscale x 2 x i32> zeroinitializer)
-  %11 = fmul fast <vscale x 2 x double> %6, shufflevector (<vscale x 2 x double> insertelement (<vscale x 2 x double> poison, double 1.100000e+01, i64 0), <vscale x 2 x double> poison, <vscale x 2 x i32> zeroinitializer)
+  %10 = fmul fast <vscale x 2 x double> %9, splat (double 3.000000e+00)
+  %11 = fmul fast <vscale x 2 x double> %6, splat (double 1.100000e+01)
   %12 = fadd fast <vscale x 2 x double> %10, %11
-  %13 = fmul fast <vscale x 2 x double> %9, shufflevector (<vscale x 2 x double> insertelement (<vscale x 2 x double> poison, double 1.100000e+01, i64 0), <vscale x 2 x double> poison, <vscale x 2 x i32> zeroinitializer)
-  %14 = fmul fast <vscale x 2 x double> %6, shufflevector (<vscale x 2 x double> insertelement (<vscale x 2 x double> poison, double 3.000000e+00, i64 0), <vscale x 2 x double> poison, <vscale x 2 x i32> zeroinitializer)
+  %13 = fmul fast <vscale x 2 x double> %9, splat (double 1.100000e+01)
+  %14 = fmul fast <vscale x 2 x double> %6, splat (double 3.000000e+00)
   %15 = fsub fast <vscale x 2 x double> %13, %14
   %interleaved.vec = tail call <vscale x 4 x double> @llvm.experimental.vector.interleave2.nxv4f64(<vscale x 2 x double> %15, <vscale x 2 x double> %12)
   ret <vscale x 4 x double> %interleaved.vec

llvm/test/CodeGen/AArch64/dag-combine-concat-vectors.ll

@@ -63,7 +63,7 @@ define fastcc i8 @allocno_reload_assign() {
   br label %1
 
 1:                                                ; preds = %1, %0
-  call void @llvm.masked.scatter.nxv16i8.nxv16p0(<vscale x 16 x i8> zeroinitializer, <vscale x 16 x ptr> zeroinitializer, i32 0, <vscale x 16 x i1> xor (<vscale x 16 x i1> shufflevector (<vscale x 16 x i1> icmp eq (<vscale x 16 x ptr> insertelement (<vscale x 16 x ptr> poison, ptr null, i64 0), <vscale x 16 x ptr> zeroinitializer), <vscale x 16 x i1> poison, <vscale x 16 x i32> zeroinitializer), <vscale x 16 x i1> shufflevector (<vscale x 16 x i1> insertelement (<vscale x 16 x i1> poison, i1 true, i32 0), <vscale x 16 x i1> poison, <vscale x 16 x i32> zeroinitializer)))
+  call void @llvm.masked.scatter.nxv16i8.nxv16p0(<vscale x 16 x i8> zeroinitializer, <vscale x 16 x ptr> zeroinitializer, i32 0, <vscale x 16 x i1> xor (<vscale x 16 x i1> shufflevector (<vscale x 16 x i1> icmp eq (<vscale x 16 x ptr> insertelement (<vscale x 16 x ptr> poison, ptr null, i64 0), <vscale x 16 x ptr> zeroinitializer), <vscale x 16 x i1> poison, <vscale x 16 x i32> zeroinitializer), <vscale x 16 x i1> splat (i1 true)))
   br label %1
 }
 

llvm/test/CodeGen/AArch64/fold-int-pow2-with-fmul-or-fdiv.ll

@@ -611,9 +611,9 @@ define <vscale x 4 x float> @fdiv_pow2_nx4xfloat(<vscale x 4 x i32> %i) "target-
 ; CHECK-NEXT:    ucvtf z0.s, p0/m, z0.s
 ; CHECK-NEXT:    fdivr z0.s, p0/m, z0.s, z1.s
 ; CHECK-NEXT:    ret
-  %p2 = shl <vscale x 4 x i32> shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> poison, i32 1, i64 0), <vscale x 4 x i32> poison, <vscale x 4 x i32> zeroinitializer), %i
+  %p2 = shl <vscale x 4 x i32> splat (i32 1), %i
   %p2_f = uitofp <vscale x 4 x i32> %p2 to <vscale x 4 x float>
-  %r = fdiv <vscale x 4 x float> shufflevector (<vscale x 4 x float> insertelement (<vscale x 4 x float> poison, float 9.000000e+00, i64 0), <vscale x 4 x float> poison, <vscale x 4 x i32> zeroinitializer), %p2_f
+  %r = fdiv <vscale x 4 x float> splat (float 9.000000e+00), %p2_f
   ret <vscale x 4 x float> %r
 }
 
@@ -626,6 +626,6 @@ define <vscale x 2 x double> @scalable2(<vscale x 2 x i64> %0) "target-features"
 ; CHECK-NEXT:    fdivr z0.d, p0/m, z0.d, z1.d
 ; CHECK-NEXT:    ret
   %2 = uitofp <vscale x 2 x i64> %0 to <vscale x 2 x double>
-  %3 = fdiv <vscale x 2 x double> shufflevector (<vscale x 2 x double> insertelement (<vscale x 2 x double> poison, double 1.000000e+00, i64 0), <vscale x 2 x double> poison, <vscale x 2 x i32> zeroinitializer), %2
+  %3 = fdiv <vscale x 2 x double> splat (double 1.000000e+00), %2
   ret <vscale x 2 x double> %3
 }

llvm/test/CodeGen/AArch64/sve-calling-convention-mixed.ll

@@ -724,7 +724,7 @@ define void @verify_all_operands_are_initialised() {
 ; CHECK-NEXT:    add sp, sp, #16
 ; CHECK-NEXT:    ldp x29, x30, [sp], #16 // 16-byte Folded Reload
 ; CHECK-NEXT:    ret
-  call void @func_f8_and_v0_passed_via_memory(float 0.0, float 1.0, float 2.0, float 3.0, float 4.0, float 5.0, float 6.0, float 7.0, float 8.0, <vscale x 4 x float> shufflevector (<vscale x 4 x float> insertelement (<vscale x 4 x float> poison, float 9.000000e+00, i64 0), <vscale x 4 x float> poison, <vscale x 4 x i32> zeroinitializer))
+  call void @func_f8_and_v0_passed_via_memory(float 0.0, float 1.0, float 2.0, float 3.0, float 4.0, float 5.0, float 6.0, float 7.0, float 8.0, <vscale x 4 x float> splat (float 9.000000e+00))
   ret void
 }
 

llvm/test/CodeGen/AArch64/sve-expand-div.ll

@@ -16,7 +16,7 @@ define <vscale x 16 x i8> @sdiv_i8(<vscale x 16 x i8> %a) #0 {
 ; CHECK-NEXT:    lsr z1.b, z0.b, #7
 ; CHECK-NEXT:    add z0.b, z0.b, z1.b
 ; CHECK-NEXT:    ret
-  %div = sdiv <vscale x 16 x i8> %a, shufflevector (<vscale x 16 x i8> insertelement (<vscale x 16 x i8> undef, i8 3, i32 0), <vscale x 16 x i8> undef, <vscale x 16 x i32> zeroinitializer)
+  %div = sdiv <vscale x 16 x i8> %a, splat (i8 3)
   ret <vscale x 16 x i8> %div
 }
 
@@ -30,7 +30,7 @@ define <vscale x 8 x i16> @sdiv_i16(<vscale x 8 x i16> %a) #0 {
 ; CHECK-NEXT:    lsr z1.h, z0.h, #15
 ; CHECK-NEXT:    add z0.h, z0.h, z1.h
 ; CHECK-NEXT:    ret
-  %div = sdiv <vscale x 8 x i16> %a, shufflevector (<vscale x 8 x i16> insertelement (<vscale x 8 x i16> undef, i16 3, i32 0), <vscale x 8 x i16> undef, <vscale x 8 x i32> zeroinitializer)
+  %div = sdiv <vscale x 8 x i16> %a, splat (i16 3)
   ret <vscale x 8 x i16> %div
 }
 
@@ -45,7 +45,7 @@ define <vscale x 4 x i32> @sdiv_i32(<vscale x 4 x i32> %a) #0 {
 ; CHECK-NEXT:    lsr z1.s, z0.s, #31
 ; CHECK-NEXT:    add z0.s, z0.s, z1.s
 ; CHECK-NEXT:    ret
-  %div = sdiv <vscale x 4 x i32> %a, shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> undef, i32 3, i32 0), <vscale x 4 x i32> undef, <vscale x 4 x i32> zeroinitializer)
+  %div = sdiv <vscale x 4 x i32> %a, splat (i32 3)
   ret <vscale x 4 x i32> %div
 }
 
@@ -60,7 +60,7 @@ define <vscale x 2 x i64> @sdiv_i64(<vscale x 2 x i64> %a) #0 {
 ; CHECK-NEXT:    lsr z1.d, z0.d, #63
 ; CHECK-NEXT:    add z0.d, z0.d, z1.d
 ; CHECK-NEXT:    ret
-  %div = sdiv <vscale x 2 x i64> %a, shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> undef, i64 3, i32 0), <vscale x 2 x i64> undef, <vscale x 2 x i32> zeroinitializer)
+  %div = sdiv <vscale x 2 x i64> %a, splat (i64 3)
   ret <vscale x 2 x i64> %div
 }
 
@@ -76,7 +76,7 @@ define <vscale x 16 x i8> @udiv_i8(<vscale x 16 x i8> %a) #0 {
 ; CHECK-NEXT:    umulh z0.b, p0/m, z0.b, z1.b
 ; CHECK-NEXT:    lsr z0.b, z0.b, #1
 ; CHECK-NEXT:    ret
-  %div = udiv <vscale x 16 x i8> %a, shufflevector (<vscale x 16 x i8> insertelement (<vscale x 16 x i8> undef, i8 3, i32 0), <vscale x 16 x i8> undef, <vscale x 16 x i32> zeroinitializer)
+  %div = udiv <vscale x 16 x i8> %a, splat (i8 3)
   ret <vscale x 16 x i8> %div
 }
 
@@ -89,7 +89,7 @@ define <vscale x 8 x i16> @udiv_i16(<vscale x 8 x i16> %a) #0 {
 ; CHECK-NEXT:    umulh z0.h, p0/m, z0.h, z1.h
 ; CHECK-NEXT:    lsr z0.h, z0.h, #1
 ; CHECK-NEXT:    ret
-  %div = udiv <vscale x 8 x i16> %a, shufflevector (<vscale x 8 x i16> insertelement (<vscale x 8 x i16> undef, i16 3, i32 0), <vscale x 8 x i16> undef, <vscale x 8 x i32> zeroinitializer)
+  %div = udiv <vscale x 8 x i16> %a, splat (i16 3)
   ret <vscale x 8 x i16> %div
 }
 
@@ -103,7 +103,7 @@ define <vscale x 4 x i32> @udiv_i32(<vscale x 4 x i32> %a) #0 {
 ; CHECK-NEXT:    umulh z0.s, p0/m, z0.s, z1.s
 ; CHECK-NEXT:    lsr z0.s, z0.s, #1
 ; CHECK-NEXT:    ret
-  %div = udiv <vscale x 4 x i32> %a, shufflevector (<vscale x 4 x i32> insertelement (<vscale x 4 x i32> undef, i32 3, i32 0), <vscale x 4 x i32> undef, <vscale x 4 x i32> zeroinitializer)
+  %div = udiv <vscale x 4 x i32> %a, splat (i32 3)
   ret <vscale x 4 x i32> %div
 }
 
@@ -117,7 +117,7 @@ define <vscale x 2 x i64> @udiv_i64(<vscale x 2 x i64> %a) #0 {
 ; CHECK-NEXT:    umulh z0.d, p0/m, z0.d, z1.d
 ; CHECK-NEXT:    lsr z0.d, z0.d, #1
 ; CHECK-NEXT:    ret
-  %div = udiv <vscale x 2 x i64> %a, shufflevector (<vscale x 2 x i64> insertelement (<vscale x 2 x i64> undef, i64 3, i32 0), <vscale x 2 x i64> undef, <vscale x 2 x i32> zeroinitializer)
+  %div = udiv <vscale x 2 x i64> %a, splat (i64 3)
   ret <vscale x 2 x i64> %div
 }
 

llvm/test/CodeGen/AArch64/sve-fp-int-min-max.ll

@@ -33,11 +33,11 @@ define i64 @scalable_int_min_max(ptr %arg, ptr %arg1, <vscale x 2 x ptr> %i37, <
 entry:
   %i56 = getelementptr inbounds float, ptr %arg, i64 0
   %i57 = load <vscale x 2 x float>, ptr %i56, align 4
-  %i58 = fmul <vscale x 2 x float> %i57, shufflevector (<vscale x 2 x float> insertelement (<vscale x 2 x float> poison, float 0x401D41D420000000, i64 0), <vscale x 2 x float> poison, <vscale x 2 x i32> zeroinitializer)
-  %i59 = fadd <vscale x 2 x float> %i58, shufflevector (<vscale x 2 x float> insertelement (<vscale x 2 x float> poison, float 1.023500e+03, i64 0), <vscale x 2 x float> poison, <vscale x 2 x i32> zeroinitializer)
+  %i58 = fmul <vscale x 2 x float> %i57, splat (float 0x401D41D420000000)
+  %i59 = fadd <vscale x 2 x float> %i58, splat (float 1.023500e+03)
   %i60 = fptosi <vscale x 2 x float> %i59 to <vscale x 2 x i32>
   %i61 = tail call <vscale x 2 x i32> @llvm.smax.nxv2i32(<vscale x 2 x i32> %i60, <vscale x 2 x i32> zeroinitializer)
-  %i62 = tail call <vscale x 2 x i32> @llvm.smin.nxv2i32(<vscale x 2 x i32> %i61, <vscale x 2 x i32> shufflevector (<vscale x 2 x i32> insertelement (<vscale x 2 x i32> poison, i32 1023, i64 0), <vscale x 2 x i32> poison, <vscale x 2 x i32> zeroinitializer))
+  %i62 = tail call <vscale x 2 x i32> @llvm.smin.nxv2i32(<vscale x 2 x i32> %i61, <vscale x 2 x i32> splat (i32 1023))
   %i63 = icmp ne <vscale x 2 x i32> %i62, zeroinitializer
   %i64 = getelementptr float, ptr %arg1, i64 0
   %i65 = tail call <vscale x 2 x float> @llvm.masked.load.nxv2f32.p0(ptr %i64, i32 4, <vscale x 2 x i1> %i63, <vscale x 2 x float> poison)

llvm/test/CodeGen/AArch64/sve-gather-scatter-dag-combine.ll

@@ -95,7 +95,7 @@ define <vscale x 16 x i8> @narrow_i64_gather_index_i8_zext(ptr %out, ptr %in, <v
   %wide.load = load <vscale x 16 x i8>, ptr %2, align 1
   %3 = zext <vscale x 16 x i8> %wide.load to <vscale x 16 x i64>
   %4 = getelementptr inbounds i8, ptr %in, <vscale x 16 x i64> %3
-  %wide.masked.gather = call <vscale x 16 x i8> @llvm.masked.gather.nxv16i8.nxv16p0(<vscale x 16 x ptr> %4, i32 1, <vscale x 16 x i1> shufflevector (<vscale x 16 x i1> insertelement (<vscale x 16 x i1> poison, i1 true, i32 0), <vscale x 16 x i1> poison, <vscale x 16 x i32> zeroinitializer), <vscale x 16 x i8> undef)
+  %wide.masked.gather = call <vscale x 16 x i8> @llvm.masked.gather.nxv16i8.nxv16p0(<vscale x 16 x ptr> %4, i32 1, <vscale x 16 x i1> splat (i1 true), <vscale x 16 x i8> undef)
   ret <vscale x 16 x i8> %wide.masked.gather
 }
 
@@ -121,7 +121,7 @@ define <vscale x 16 x i8> @narrow_i64_gather_index_i8_sext(ptr %out, ptr %in, <v
   %wide.load = load <vscale x 16 x i8>, ptr %2, align 1
   %3 = sext <vscale x 16 x i8> %wide.load to <vscale x 16 x i64>
   %4 = getelementptr inbounds i8, ptr %in, <vscale x 16 x i64> %3
-  %wide.masked.gather = call <vscale x 16 x i8> @llvm.masked.gather.nxv16i8.nxv16p0(<vscale x 16 x ptr> %4, i32 1, <vscale x 16 x i1> shufflevector (<vscale x 16 x i1> insertelement (<vscale x 16 x i1> poison, i1 true, i32 0), <vscale x 16 x i1> poison, <vscale x 16 x i32> zeroinitializer), <vscale x 16 x i8> undef)
+  %wide.masked.gather = call <vscale x 16 x i8> @llvm.masked.gather.nxv16i8.nxv16p0(<vscale x 16 x ptr> %4, i32 1, <vscale x 16 x i1> splat (i1 true), <vscale x 16 x i8> undef)
   ret <vscale x 16 x i8> %wide.masked.gather
 }
 
@@ -141,7 +141,7 @@ define <vscale x 8 x i16> @narrow_i64_gather_index_i16_zext(ptr %out, ptr %in, <
   %wide.load = load <vscale x 8 x i16>, ptr %2, align 1
   %3 = zext <vscale x 8 x i16> %wide.load to <vscale x 8 x i64>
   %4 = getelementptr inbounds i16, ptr %in, <vscale x 8 x i64> %3
-  %wide.masked.gather = call <vscale x 8 x i16> @llvm.masked.gather.nxv8i16.nxv8p0(<vscale x 8 x ptr> %4, i32 1, <vscale x 8 x i1> shufflevector (<vscale x 8 x i1> insertelement (<vscale x 8 x i1> poison, i1 true, i32 0), <vscale x 8 x i1> poison, <vscale x 8 x i32> zeroinitializer), <vscale x 8 x i16> undef)
+  %wide.masked.gather = call <vscale x 8 x i16> @llvm.masked.gather.nxv8i16.nxv8p0(<vscale x 8 x ptr> %4, i32 1, <vscale x 8 x i1> splat (i1 true), <vscale x 8 x i16> undef)
   ret <vscale x 8 x i16> %wide.masked.gather
 }
 
@@ -161,7 +161,7 @@ define <vscale x 8 x i16> @narrow_i64_gather_index_i16_sext(ptr %out, ptr %in, <
   %wide.load = load <vscale x 8 x i16>, ptr %2, align 1
   %3 = sext <vscale x 8 x i16> %wide.load to <vscale x 8 x i64>
   %4 = getelementptr inbounds i16, ptr %in, <vscale x 8 x i64> %3
-  %wide.masked.gather = call <vscale x 8 x i16> @llvm.masked.gather.nxv8i16.nxv8p0(<vscale x 8 x ptr> %4, i32 1, <vscale x 8 x i1> shufflevector (<vscale x 8 x i1> insertelement (<vscale x 8 x i1> poison, i1 true, i32 0), <vscale x 8 x i1> poison, <vscale x 8 x i32> zeroinitializer), <vscale x 8 x i16> undef)
+  %wide.masked.gather = call <vscale x 8 x i16> @llvm.masked.gather.nxv8i16.nxv8p0(<vscale x 8 x ptr> %4, i32 1, <vscale x 8 x i1> splat (i1 true), <vscale x 8 x i16> undef)
   ret <vscale x 8 x i16> %wide.masked.gather
 }
 
@@ -177,7 +177,7 @@ define <vscale x 4 x i32> @no_narrow_i64_gather_index_i32(ptr %out, ptr %in, <vs
   %wide.load = load <vscale x 4 x i32>, ptr %2, align 1
   %3 = zext <vscale x 4 x i32> %wide.load to <vscale x 4 x i64>
   %4 = getelementptr inbounds i32, ptr %in, <vscale x 4 x i64> %3
-  %wide.masked.gather = call <vscale x 4 x i32> @llvm.masked.gather.nxv4i32.nxv4p0(<vscale x 4 x ptr> %4, i32 1, <vscale x 4 x i1> shufflevector (<vscale x 4 x i1> insertelement (<vscale x 4 x i1> poison, i1 true, i32 0), <vscale x 4 x i1> poison, <vscale x 4 x i32> zeroinitializer), <vscale x 4 x i32> undef)
+  %wide.masked.gather = call <vscale x 4 x i32> @llvm.masked.gather.nxv4i32.nxv4p0(<vscale x 4 x ptr> %4, i32 1, <vscale x 4 x i1> splat (i1 true), <vscale x 4 x i32> undef)
   ret <vscale x 4 x i32> %wide.masked.gather
 }
 
@@ -192,7 +192,7 @@ define <vscale x 2 x i64> @no_narrow_i64_gather_index_i64(ptr %out, ptr %in, <vs
   %2 = bitcast ptr %1 to ptr
   %wide.load = load <vscale x 2 x i64>, ptr %2, align 1
   %3 = getelementptr inbounds i64, ptr %in, <vscale x 2 x i64> %wide.load
-  %wide.masked.gather = call <vscale x 2 x i64> @llvm.masked.gather.nxv2i64.nxv2p0(<vscale x 2 x ptr> %3, i32 1, <vscale x 2 x i1> shufflevector (<vscale x 2 x i1> insertelement (<vscale x 2 x i1> poison, i1 true, i32 0), <vscale x 2 x i1> poison, <vscale x 2 x i32> zeroinitializer), <vscale x 2 x i64> undef)
+  %wide.masked.gather = call <vscale x 2 x i64> @llvm.masked.gather.nxv2i64.nxv2p0(<vscale x 2 x ptr> %3, i32 1, <vscale x 2 x i1> splat (i1 true), <vscale x 2 x i64> undef)
   ret <vscale x 2 x i64> %wide.masked.gather
 }