From 86b9a4f892b9fc6cf5929f33a529570acd7abc71 Mon Sep 17 00:00:00 2001 From: Alexey Bataev Date: Mon, 29 Apr 2024 06:42:35 -0700 Subject: [SLP][NFC]Add a test with the skipped gather node, which is same, as vectorized node. --- .../X86/gather-node-same-as-vect-but-order.ll | 144 +++++++++++++++++++++ 1 file changed, 144 insertions(+) create mode 100644 llvm/test/Transforms/SLPVectorizer/X86/gather-node-same-as-vect-but-order.ll diff --git a/llvm/test/Transforms/SLPVectorizer/X86/gather-node-same-as-vect-but-order.ll b/llvm/test/Transforms/SLPVectorizer/X86/gather-node-same-as-vect-but-order.ll new file mode 100644 index 0000000..e66d38b --- /dev/null +++ b/llvm/test/Transforms/SLPVectorizer/X86/gather-node-same-as-vect-but-order.ll @@ -0,0 +1,144 @@ +; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 4 +; RUN: opt -S --passes=slp-vectorizer -mtriple=x86_64-unknown-linux-gnu -mcpu=x86-64-v4 < %s | FileCheck %s + +%struct.rect = type { float, float, float, float } + +define void @foo(ptr %i7, i32 %0, i1 %tobool62.not) { +; CHECK-LABEL: define void @foo( +; CHECK-SAME: ptr [[I7:%.*]], i32 [[TMP0:%.*]], i1 [[TOBOOL62_NOT:%.*]]) #[[ATTR0:[0-9]+]] { +; CHECK-NEXT: entry: +; CHECK-NEXT: [[RC21:%.*]] = alloca [0 x [0 x %struct.rect]], i32 0, align 4 +; CHECK-NEXT: [[TMP1:%.*]] = load float, ptr [[RC21]], align 4 +; CHECK-NEXT: [[TMP2:%.*]] = insertelement <2 x i32> poison, i32 [[TMP0]], i32 0 +; CHECK-NEXT: [[TMP3:%.*]] = shufflevector <2 x i32> [[TMP2]], <2 x i32> poison, <2 x i32> zeroinitializer +; CHECK-NEXT: [[TMP4:%.*]] = sitofp <2 x i32> [[TMP3]] to <2 x float> +; CHECK-NEXT: [[TMP5:%.*]] = shufflevector <2 x float> [[TMP4]], <2 x float> poison, <4 x i32> +; CHECK-NEXT: [[X1:%.*]] = getelementptr i8, ptr [[RC21]], i64 4 +; CHECK-NEXT: [[TMP6:%.*]] = load <2 x float>, ptr [[X1]], align 4 +; CHECK-NEXT: [[TMP7:%.*]] = load float, ptr [[I7]], align 4 +; CHECK-NEXT: [[TMP8:%.*]] = insertelement <4 x float> poison, float [[TMP1]], i32 0 +; CHECK-NEXT: [[TMP9:%.*]] = shufflevector <2 x float> [[TMP6]], <2 x float> poison, <4 x i32> +; CHECK-NEXT: [[TMP10:%.*]] = shufflevector <4 x float> [[TMP8]], <4 x float> [[TMP9]], <4 x i32> +; CHECK-NEXT: [[TMP11:%.*]] = insertelement <4 x float> [[TMP10]], float [[TMP7]], i32 3 +; CHECK-NEXT: [[TMP12:%.*]] = fcmp olt <4 x float> [[TMP11]], zeroinitializer +; CHECK-NEXT: [[TMP13:%.*]] = load <4 x float>, ptr poison, align 16 +; CHECK-NEXT: [[TMP14:%.*]] = fcmp olt <4 x float> [[TMP13]], zeroinitializer +; CHECK-NEXT: [[TMP15:%.*]] = select <4 x i1> [[TMP14]], <4 x float> [[TMP5]], <4 x float> zeroinitializer +; CHECK-NEXT: [[TMP16:%.*]] = select <4 x i1> [[TMP12]], <4 x float> zeroinitializer, <4 x float> [[TMP15]] +; CHECK-NEXT: store <4 x float> [[TMP16]], ptr [[RC21]], align 4 +; CHECK-NEXT: br label [[IF_END:%.*]] +; CHECK: entry.if.end72_crit_edge: +; CHECK-NEXT: br label [[IF_END72:%.*]] +; CHECK: if.then63: +; CHECK-NEXT: br label [[IF_END]] +; CHECK: if.end: +; CHECK-NEXT: [[TMP17:%.*]] = phi <4 x float> [ poison, [[IF_THEN63:%.*]] ], [ [[TMP16]], [[ENTRY:%.*]] ] +; CHECK-NEXT: [[TMP18:%.*]] = call <4 x float> @llvm.round.v4f32(<4 x float> [[TMP17]]) +; CHECK-NEXT: [[TMP19:%.*]] = fptosi <4 x float> [[TMP18]] to <4 x i32> +; CHECK-NEXT: br label [[IF_END72]] +; CHECK: if.end72: +; CHECK-NEXT: [[TMP20:%.*]] = phi <4 x i32> [ poison, [[ENTRY_IF_END72_CRIT_EDGE:%.*]] ], [ [[TMP19]], [[IF_END]] ] +; CHECK-NEXT: [[TMP21:%.*]] = shufflevector <4 x i32> [[TMP20]], <4 x i32> poison, <4 x i32> +; CHECK-NEXT: br i1 [[TOBOOL62_NOT]], label [[IF_END75:%.*]], label [[IF_THEN74:%.*]] +; CHECK: if.then74: +; CHECK-NEXT: br label [[IF_END75]] +; CHECK: if.end75: +; CHECK-NEXT: [[TMP22:%.*]] = phi <4 x i32> [ [[TMP20]], [[IF_THEN74]] ], [ [[TMP21]], [[IF_END72]] ] +; CHECK-NEXT: [[TMP23:%.*]] = or <4 x i32> [[TMP22]], +; CHECK-NEXT: [[TMP24:%.*]] = shufflevector <2 x i32> [[TMP3]], <2 x i32> poison, <4 x i32> zeroinitializer +; CHECK-NEXT: [[TMP25:%.*]] = mul <4 x i32> [[TMP23]], [[TMP24]] +; CHECK-NEXT: [[TMP26:%.*]] = sitofp <4 x i32> [[TMP25]] to <4 x float> +; CHECK-NEXT: [[TMP27:%.*]] = shufflevector <4 x float> [[TMP26]], <4 x float> poison, <4 x i32> +; CHECK-NEXT: store <4 x float> [[TMP27]], ptr [[RC21]], align 4 +; CHECK-NEXT: ret void +; +entry: + %rc21 = alloca [0 x [0 x %struct.rect]], i32 0, align 4 + %1 = load float, ptr %rc21, align 4 + %cmp = fcmp olt float %1, 0.000000e+00 + %conv = sitofp i32 %0 to float + %cmp2 = fcmp olt float %conv, 0.000000e+00 + %cond = select i1 %cmp2, float %conv, float 0.000000e+00 + %cond9 = select i1 %cmp, float 0.000000e+00, float %cond + store float %cond9, ptr %rc21, align 4 + %x1 = getelementptr i8, ptr %rc21, i64 4 + %2 = load float, ptr %x1, align 4 + %cmp11 = fcmp olt float %2, 0.000000e+00 + %conv16 = sitofp i32 %0 to float + %cmp17 = fcmp olt float %conv16, 0.000000e+00 + %cond24 = select i1 %cmp17, float %conv16, float 0.000000e+00 + %cond26 = select i1 %cmp11, float 0.000000e+00, float %cond24 + store float %cond26, ptr %x1, align 4 + %y0 = getelementptr i8, ptr %rc21, i64 8 + %3 = load float, ptr %y0, align 4 + %cmp28 = fcmp olt float %3, 0.000000e+00 + %cmp34 = fcmp olt float %conv, 0.000000e+00 + %cond41 = select i1 %cmp34, float %conv, float 0.000000e+00 + %cond43 = select i1 %cmp28, float 0.000000e+00, float %cond41 + store float %cond43, ptr %y0, align 4 + %y11 = getelementptr i8, ptr %rc21, i64 12 + %4 = load float, ptr %i7, align 4 + %cmp45 = fcmp olt float %4, 0.000000e+00 + %cmp51 = fcmp olt float %conv16, 0.000000e+00 + %cond58 = select i1 %cmp51, float %conv16, float 0.000000e+00 + %cond60 = select i1 %cmp45, float 0.000000e+00, float %cond58 + store float %cond60, ptr %y11, align 4 + br label %if.end + +entry.if.end72_crit_edge: + br label %if.end72 + +if.then63: + br label %if.end + +if.end: + %5 = phi float [ 0.000000e+00, %if.then63 ], [ %cond60, %entry ] + %6 = phi float [ 0.000000e+00, %if.then63 ], [ %cond26, %entry ] + %7 = phi float [ 0.000000e+00, %if.then63 ], [ %cond43, %entry ] + %8 = phi float [ 0.000000e+00, %if.then63 ], [ %cond9, %entry ] + %9 = call float @llvm.round.f32(float %8) + %conv65 = fptosi float %9 to i32 + %10 = call float @llvm.round.f32(float %7) + %conv67 = fptosi float %10 to i32 + %11 = call float @llvm.round.f32(float %6) + %conv69 = fptosi float %11 to i32 + %12 = call float @llvm.round.f32(float %5) + %conv71 = fptosi float %12 to i32 + br label %if.end72 + +if.end72: + %.pre100 = phi i32 [ 0, %entry.if.end72_crit_edge ], [ %conv71, %if.end ] + %.pre99 = phi i32 [ 0, %entry.if.end72_crit_edge ], [ %conv67, %if.end ] + %.pre98 = phi i32 [ 0, %entry.if.end72_crit_edge ], [ %conv69, %if.end ] + %.pre97 = phi i32 [ 0, %entry.if.end72_crit_edge ], [ %conv65, %if.end ] + br i1 %tobool62.not, label %if.end75, label %if.then74 + +if.then74: + br label %if.end75 + +if.end75: + %13 = phi i32 [ %.pre99, %if.then74 ], [ %.pre100, %if.end72 ] + %14 = phi i32 [ %.pre100, %if.then74 ], [ %.pre99, %if.end72 ] + %15 = phi i32 [ %.pre97, %if.then74 ], [ %.pre98, %if.end72 ] + %16 = phi i32 [ %.pre98, %if.then74 ], [ %.pre97, %if.end72 ] + %sub = or i32 %16, 1 + %mul = mul i32 %sub, %0 + %conv77 = sitofp i32 %mul to float + store float %conv77, ptr %rc21, align 4 + %x178 = getelementptr i8, ptr %rc21, i64 4 + %sub79 = or i32 %15, 1 + %mul80 = mul i32 %sub79, %0 + %conv81 = sitofp i32 %mul80 to float + store float %conv81, ptr %x178, align 4 + %y082 = getelementptr i8, ptr %rc21, i64 8 + %sub83 = or i32 %14, 1 + %mul84 = mul i32 %sub83, %0 + %conv85 = sitofp i32 %mul84 to float + store float %conv85, ptr %y082, align 4 + %y186 = getelementptr i8, ptr %rc21, i64 12 + %sub87 = or i32 %13, 1 + %mul88 = mul i32 %sub87, %0 + %conv89 = sitofp i32 %mul88 to float + store float %conv89, ptr %y186, align 4 + ret void +} -- cgit v1.1