diff options
| author | Dimitry Andric <dim@FreeBSD.org> | 2013-04-08 18:41:23 +0000 |
|---|---|---|
| committer | Dimitry Andric <dim@FreeBSD.org> | 2013-04-08 18:41:23 +0000 |
| commit | 4a16efa3e43e35f0cc9efe3a67f620f0017c3d36 (patch) | |
| tree | 06099edc18d30894081a822b756f117cbe0b8207 /test/Transforms/BBVectorize | |
| parent | 482e7bddf617ae804dc47133cb07eb4aa81e45de (diff) | |
Notes
Diffstat (limited to 'test/Transforms/BBVectorize')
| -rw-r--r-- | test/Transforms/BBVectorize/X86/pr15289.ll | 98 | ||||
| -rw-r--r-- | test/Transforms/BBVectorize/X86/simple-int.ll | 79 | ||||
| -rw-r--r-- | test/Transforms/BBVectorize/cycle.ll | 2 | ||||
| -rw-r--r-- | test/Transforms/BBVectorize/ld1.ll | 2 | ||||
| -rw-r--r-- | test/Transforms/BBVectorize/loop1.ll | 4 | ||||
| -rw-r--r-- | test/Transforms/BBVectorize/req-depth.ll | 4 | ||||
| -rw-r--r-- | test/Transforms/BBVectorize/search-limit.ll | 4 | ||||
| -rw-r--r-- | test/Transforms/BBVectorize/simple-int.ll | 38 | ||||
| -rw-r--r-- | test/Transforms/BBVectorize/simple-ldstr.ll | 4 | ||||
| -rw-r--r-- | test/Transforms/BBVectorize/simple-sel.ll | 4 | ||||
| -rw-r--r-- | test/Transforms/BBVectorize/simple.ll | 2 | ||||
| -rw-r--r-- | test/Transforms/BBVectorize/simple3.ll | 2 |
12 files changed, 225 insertions, 18 deletions
diff --git a/test/Transforms/BBVectorize/X86/pr15289.ll b/test/Transforms/BBVectorize/X86/pr15289.ll new file mode 100644 index 000000000000..07cc5d8b96b7 --- /dev/null +++ b/test/Transforms/BBVectorize/X86/pr15289.ll @@ -0,0 +1,98 @@ +; RUN: opt < %s -basicaa -bb-vectorize -disable-output +; This is a bugpoint-reduced test case. It did not always assert, but does reproduce the bug +; and running under valgrind (or some similar tool) will catch the error. + +target datalayout = "e-p:64:64:64-S128-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f16:16:16-f32:32:32-f64:64:64-f128:128:128-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64" +target triple = "x86_64-apple-darwin12.2.0" + +%0 = type { [10 x { float, float }], [10 x { float, float }], [10 x { float, float }], [10 x { float, float }], [10 x { float, float }] } +%1 = type { [10 x [8 x i8]] } +%2 = type { i64, i64 } +%3 = type { [10 x i64], i64, i64, i64, i64, i64 } +%4 = type { i64, i64, i64, i64, i64, i64 } +%5 = type { [10 x i64] } +%6 = type { [10 x float], [10 x float], [10 x float], [10 x float] } +%struct.__st_parameter_dt.1.3.5.7 = type { %struct.__st_parameter_common.0.2.4.6, i64, i64*, i64*, i8*, i8*, i32, i32, i8*, i8*, i32, i32, i8*, [256 x i8], i32*, i64, i8*, i32, i32, i8*, i8*, i32, i32, i8*, i8*, i32, i32, i8*, i8*, i32, [4 x i8] } +%struct.__st_parameter_common.0.2.4.6 = type { i32, i32, i8*, i32, i32, i8*, i32* } + +@cctenso_ = external unnamed_addr global %0, align 32 +@ctenso_ = external unnamed_addr global %1, align 32 +@i_dim_ = external unnamed_addr global %2, align 16 +@itenso1_ = external unnamed_addr global %3, align 32 +@itenso2_ = external unnamed_addr global %4, align 32 +@ltenso_ = external unnamed_addr global %5, align 32 +@rtenso_ = external unnamed_addr global %6, align 32 +@.cst = external unnamed_addr constant [8 x i8], align 8 +@.cst1 = external unnamed_addr constant [3 x i8], align 8 +@.cst2 = external unnamed_addr constant [29 x i8], align 8 +@.cst3 = external unnamed_addr constant [32 x i8], align 64 + +define void @cart_to_dc2y_(double* noalias nocapture %xx, double* noalias nocapture %yy, double* noalias nocapture %zz, [5 x { double, double }]* noalias nocapture %c2ten) nounwind uwtable { +entry: + %0 = fmul double undef, undef + %1 = fmul double undef, undef + %2 = fadd double undef, undef + %3 = fmul double undef, 0x3FE8B8B76E3E9919 + %4 = fsub double %0, %1 + %5 = fsub double -0.000000e+00, undef + %6 = fmul double undef, undef + %7 = fmul double %4, %6 + %8 = fmul double undef, 2.000000e+00 + %9 = fmul double %8, undef + %10 = fmul double undef, %9 + %11 = fmul double %10, undef + %12 = fsub double undef, %7 + %13 = fmul double %3, %12 + %14 = fmul double %3, undef + %15 = getelementptr inbounds [5 x { double, double }]* %c2ten, i64 0, i64 0, i32 0 + store double %13, double* %15, align 8, !tbaa !0 + %16 = getelementptr inbounds [5 x { double, double }]* %c2ten, i64 0, i64 0, i32 1 + %17 = fmul double undef, %8 + %18 = fmul double %17, undef + %19 = fmul double undef, %18 + %20 = fadd double undef, undef + %21 = fmul double %3, %19 + %22 = fsub double -0.000000e+00, %21 + %23 = getelementptr inbounds [5 x { double, double }]* %c2ten, i64 0, i64 1, i32 0 + store double %22, double* %23, align 8, !tbaa !0 + %24 = getelementptr inbounds [5 x { double, double }]* %c2ten, i64 0, i64 1, i32 1 + %25 = fmul double undef, 0x3FE42F601A8C6794 + %26 = fmul double undef, 2.000000e+00 + %27 = fsub double %26, %0 + %28 = fmul double %6, undef + %29 = fsub double undef, %28 + %30 = getelementptr inbounds [5 x { double, double }]* %c2ten, i64 0, i64 2, i32 0 + store double undef, double* %30, align 8, !tbaa !0 + %31 = getelementptr inbounds [5 x { double, double }]* %c2ten, i64 0, i64 2, i32 1 + %32 = fmul double undef, %17 + %33 = fmul double undef, %17 + %34 = fmul double undef, %32 + %35 = fmul double undef, %33 + %36 = fsub double undef, %35 + %37 = fmul double %3, %34 + %38 = getelementptr inbounds [5 x { double, double }]* %c2ten, i64 0, i64 3, i32 0 + store double %37, double* %38, align 8, !tbaa !0 + %39 = getelementptr inbounds [5 x { double, double }]* %c2ten, i64 0, i64 3, i32 1 + %40 = fmul double undef, %8 + %41 = fmul double undef, %40 + %42 = fmul double undef, %41 + %43 = fsub double undef, %42 + %44 = fmul double %3, %43 + %45 = getelementptr inbounds [5 x { double, double }]* %c2ten, i64 0, i64 4, i32 0 + store double %13, double* %45, align 8, !tbaa !0 + %46 = getelementptr inbounds [5 x { double, double }]* %c2ten, i64 0, i64 4, i32 1 + %47 = fsub double -0.000000e+00, %14 + store double %47, double* %16, align 8, !tbaa !0 + store double undef, double* %24, align 8, !tbaa !0 + store double -0.000000e+00, double* %31, align 8, !tbaa !0 + store double undef, double* %39, align 8, !tbaa !0 + store double undef, double* %46, align 8, !tbaa !0 + ret void +} + +attributes #0 = { nounwind uwtable } +attributes #1 = { nounwind readnone } +attributes #2 = { nounwind } + +!0 = metadata !{metadata !"alias set 17: real(kind=8)", metadata !1} +!1 = metadata !{metadata !1} diff --git a/test/Transforms/BBVectorize/X86/simple-int.ll b/test/Transforms/BBVectorize/X86/simple-int.ll new file mode 100644 index 000000000000..f5dbe46b1480 --- /dev/null +++ b/test/Transforms/BBVectorize/X86/simple-int.ll @@ -0,0 +1,79 @@ +target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128" +; RUN: opt < %s -mtriple=x86_64-unknown-linux-gnu -mcpu=corei7 -bb-vectorize -bb-vectorize-req-chain-depth=3 -instcombine -gvn -S | FileCheck %s + +declare double @llvm.fma.f64(double, double, double) +declare double @llvm.fmuladd.f64(double, double, double) +declare double @llvm.cos.f64(double) +declare double @llvm.powi.f64(double, i32) + +; Basic depth-3 chain with fma +define double @test1(double %A1, double %A2, double %B1, double %B2, double %C1, double %C2) { + %X1 = fsub double %A1, %B1 + %X2 = fsub double %A2, %B2 + %Y1 = call double @llvm.fma.f64(double %X1, double %A1, double %C1) + %Y2 = call double @llvm.fma.f64(double %X2, double %A2, double %C2) + %Z1 = fadd double %Y1, %B1 + %Z2 = fadd double %Y2, %B2 + %R = fmul double %Z1, %Z2 + ret double %R +; CHECK: @test1 +; CHECK: ret double %R +} + +; Basic depth-3 chain with fmuladd +define double @test1a(double %A1, double %A2, double %B1, double %B2, double %C1, double %C2) { + %X1 = fsub double %A1, %B1 + %X2 = fsub double %A2, %B2 + %Y1 = call double @llvm.fmuladd.f64(double %X1, double %A1, double %C1) + %Y2 = call double @llvm.fmuladd.f64(double %X2, double %A2, double %C2) + %Z1 = fadd double %Y1, %B1 + %Z2 = fadd double %Y2, %B2 + %R = fmul double %Z1, %Z2 + ret double %R +; CHECK: @test1a +; CHECK: ret double %R +} + +; Basic depth-3 chain with cos +define double @test2(double %A1, double %A2, double %B1, double %B2) { + %X1 = fsub double %A1, %B1 + %X2 = fsub double %A2, %B2 + %Y1 = call double @llvm.cos.f64(double %X1) + %Y2 = call double @llvm.cos.f64(double %X2) + %Z1 = fadd double %Y1, %B1 + %Z2 = fadd double %Y2, %B2 + %R = fmul double %Z1, %Z2 + ret double %R +; CHECK: @test2 +; CHECK: ret double %R +} + +; Basic depth-3 chain with powi +define double @test3(double %A1, double %A2, double %B1, double %B2, i32 %P) { + %X1 = fsub double %A1, %B1 + %X2 = fsub double %A2, %B2 + %Y1 = call double @llvm.powi.f64(double %X1, i32 %P) + %Y2 = call double @llvm.powi.f64(double %X2, i32 %P) + %Z1 = fadd double %Y1, %B1 + %Z2 = fadd double %Y2, %B2 + %R = fmul double %Z1, %Z2 + ret double %R +; CHECK: @test3 +; CHECK: ret double %R +} + +; Basic depth-3 chain with powi (different powers: should not vectorize) +define double @test4(double %A1, double %A2, double %B1, double %B2, i32 %P) { + %X1 = fsub double %A1, %B1 + %X2 = fsub double %A2, %B2 + %P2 = add i32 %P, 1 + %Y1 = call double @llvm.powi.f64(double %X1, i32 %P) + %Y2 = call double @llvm.powi.f64(double %X2, i32 %P2) + %Z1 = fadd double %Y1, %B1 + %Z2 = fadd double %Y2, %B2 + %R = fmul double %Z1, %Z2 + ret double %R +; CHECK: @test4 +; CHECK: ret double %R +} + diff --git a/test/Transforms/BBVectorize/cycle.ll b/test/Transforms/BBVectorize/cycle.ll index e8e82ce02479..bdcb30da887f 100644 --- a/test/Transforms/BBVectorize/cycle.ll +++ b/test/Transforms/BBVectorize/cycle.ll @@ -1,5 +1,5 @@ target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128" -; RUN: opt < %s -bb-vectorize -bb-vectorize-req-chain-depth=3 -instcombine -gvn -S | FileCheck %s +; RUN: opt < %s -bb-vectorize -bb-vectorize-req-chain-depth=3 -bb-vectorize-ignore-target-info -instcombine -gvn -S | FileCheck %s ; This test checks the non-trivial pairing-induced cycle avoidance. Without this cycle avoidance, the algorithm would otherwise ; want to select the pairs: diff --git a/test/Transforms/BBVectorize/ld1.ll b/test/Transforms/BBVectorize/ld1.ll index cea225d076e1..ea5cb5dd93f7 100644 --- a/test/Transforms/BBVectorize/ld1.ll +++ b/test/Transforms/BBVectorize/ld1.ll @@ -1,5 +1,5 @@ target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128" -; RUN: opt < %s -bb-vectorize -bb-vectorize-req-chain-depth=3 -instcombine -gvn -S | FileCheck %s +; RUN: opt < %s -bb-vectorize -bb-vectorize-req-chain-depth=3 -bb-vectorize-ignore-target-info -instcombine -gvn -S | FileCheck %s define double @test1(double* %a, double* %b, double* %c) nounwind uwtable readonly { entry: diff --git a/test/Transforms/BBVectorize/loop1.ll b/test/Transforms/BBVectorize/loop1.ll index c22ea5852a1b..e592edb44a02 100644 --- a/test/Transforms/BBVectorize/loop1.ll +++ b/test/Transforms/BBVectorize/loop1.ll @@ -1,7 +1,7 @@ target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128" target triple = "x86_64-unknown-linux-gnu" -; RUN: opt < %s -bb-vectorize -bb-vectorize-req-chain-depth=3 -instcombine -gvn -S | FileCheck %s -; RUN: opt < %s -basicaa -loop-unroll -unroll-threshold=45 -unroll-allow-partial -bb-vectorize -bb-vectorize-req-chain-depth=3 -instcombine -gvn -S | FileCheck %s -check-prefix=CHECK-UNRL +; RUN: opt < %s -bb-vectorize -bb-vectorize-req-chain-depth=3 -bb-vectorize-ignore-target-info -instcombine -gvn -S | FileCheck %s +; RUN: opt < %s -basicaa -loop-unroll -unroll-threshold=45 -unroll-allow-partial -bb-vectorize -bb-vectorize-req-chain-depth=3 -bb-vectorize-ignore-target-info -instcombine -gvn -S | FileCheck %s -check-prefix=CHECK-UNRL ; The second check covers the use of alias analysis (with loop unrolling). define void @test1(double* noalias %out, double* noalias %in1, double* noalias %in2) nounwind uwtable { diff --git a/test/Transforms/BBVectorize/req-depth.ll b/test/Transforms/BBVectorize/req-depth.ll index 8c9cc3c188e3..e0120059b952 100644 --- a/test/Transforms/BBVectorize/req-depth.ll +++ b/test/Transforms/BBVectorize/req-depth.ll @@ -1,6 +1,6 @@ target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128" -; RUN: opt < %s -bb-vectorize -bb-vectorize-req-chain-depth 3 -S | FileCheck %s -check-prefix=CHECK-RD3 -; RUN: opt < %s -bb-vectorize -bb-vectorize-req-chain-depth 2 -S | FileCheck %s -check-prefix=CHECK-RD2 +; RUN: opt < %s -bb-vectorize -bb-vectorize-req-chain-depth 3 -bb-vectorize-ignore-target-info -S | FileCheck %s -check-prefix=CHECK-RD3 +; RUN: opt < %s -bb-vectorize -bb-vectorize-req-chain-depth 2 -bb-vectorize-ignore-target-info -S | FileCheck %s -check-prefix=CHECK-RD2 define double @test1(double %A1, double %A2, double %B1, double %B2) { %X1 = fsub double %A1, %B1 diff --git a/test/Transforms/BBVectorize/search-limit.ll b/test/Transforms/BBVectorize/search-limit.ll index aeaf98865bc9..a694e45bc181 100644 --- a/test/Transforms/BBVectorize/search-limit.ll +++ b/test/Transforms/BBVectorize/search-limit.ll @@ -1,6 +1,6 @@ target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128" -; RUN: opt < %s -bb-vectorize -bb-vectorize-req-chain-depth=3 -instcombine -gvn -S | FileCheck %s -; RUN: opt < %s -bb-vectorize -bb-vectorize-req-chain-depth=3 -bb-vectorize-search-limit=4 -instcombine -gvn -S | FileCheck %s -check-prefix=CHECK-SL4 +; RUN: opt < %s -bb-vectorize -bb-vectorize-req-chain-depth=3 -bb-vectorize-ignore-target-info -instcombine -gvn -S | FileCheck %s +; RUN: opt < %s -bb-vectorize -bb-vectorize-req-chain-depth=3 -bb-vectorize-search-limit=4 -bb-vectorize-ignore-target-info -instcombine -gvn -S | FileCheck %s -check-prefix=CHECK-SL4 define double @test1(double %A1, double %A2, double %B1, double %B2) { ; CHECK: @test1 diff --git a/test/Transforms/BBVectorize/simple-int.ll b/test/Transforms/BBVectorize/simple-int.ll index ae1d63bfd852..e4d51526ca11 100644 --- a/test/Transforms/BBVectorize/simple-int.ll +++ b/test/Transforms/BBVectorize/simple-int.ll @@ -1,7 +1,8 @@ target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128" -; RUN: opt < %s -bb-vectorize -bb-vectorize-req-chain-depth=3 -instcombine -gvn -S | FileCheck %s +; RUN: opt < %s -bb-vectorize -bb-vectorize-req-chain-depth=3 -bb-vectorize-ignore-target-info -instcombine -gvn -S | FileCheck %s declare double @llvm.fma.f64(double, double, double) +declare double @llvm.fmuladd.f64(double, double, double) declare double @llvm.cos.f64(double) declare double @llvm.powi.f64(double, i32) @@ -31,6 +32,32 @@ define double @test1(double %A1, double %A2, double %B1, double %B2, double %C1, ; CHECK: ret double %R } +; Basic depth-3 chain with fmuladd +define double @test1a(double %A1, double %A2, double %B1, double %B2, double %C1, double %C2) { + %X1 = fsub double %A1, %B1 + %X2 = fsub double %A2, %B2 + %Y1 = call double @llvm.fmuladd.f64(double %X1, double %A1, double %C1) + %Y2 = call double @llvm.fmuladd.f64(double %X2, double %A2, double %C2) + %Z1 = fadd double %Y1, %B1 + %Z2 = fadd double %Y2, %B2 + %R = fmul double %Z1, %Z2 + ret double %R +; CHECK: @test1a +; CHECK: %X1.v.i1.1 = insertelement <2 x double> undef, double %B1, i32 0 +; CHECK: %X1.v.i1.2 = insertelement <2 x double> %X1.v.i1.1, double %B2, i32 1 +; CHECK: %X1.v.i0.1 = insertelement <2 x double> undef, double %A1, i32 0 +; CHECK: %X1.v.i0.2 = insertelement <2 x double> %X1.v.i0.1, double %A2, i32 1 +; CHECK: %X1 = fsub <2 x double> %X1.v.i0.2, %X1.v.i1.2 +; CHECK: %Y1.v.i2.1 = insertelement <2 x double> undef, double %C1, i32 0 +; CHECK: %Y1.v.i2.2 = insertelement <2 x double> %Y1.v.i2.1, double %C2, i32 1 +; CHECK: %Y1 = call <2 x double> @llvm.fmuladd.v2f64(<2 x double> %X1, <2 x double> %X1.v.i0.2, <2 x double> %Y1.v.i2.2) +; CHECK: %Z1 = fadd <2 x double> %Y1, %X1.v.i1.2 +; CHECK: %Z1.v.r1 = extractelement <2 x double> %Z1, i32 0 +; CHECK: %Z1.v.r2 = extractelement <2 x double> %Z1, i32 1 +; CHECK: %R = fmul double %Z1.v.r1, %Z1.v.r2 +; CHECK: ret double %R +} + ; Basic depth-3 chain with cos define double @test2(double %A1, double %A2, double %B1, double %B2) { %X1 = fsub double %A1, %B1 @@ -97,7 +124,10 @@ define double @test4(double %A1, double %A2, double %B1, double %B2, i32 %P) { ; CHECK: ret double %R } -; CHECK: declare <2 x double> @llvm.fma.v2f64(<2 x double>, <2 x double>, <2 x double>) nounwind readnone -; CHECK: declare <2 x double> @llvm.cos.v2f64(<2 x double>) nounwind readonly -; CHECK: declare <2 x double> @llvm.powi.v2f64(<2 x double>, i32) nounwind readonly +; CHECK: declare <2 x double> @llvm.fma.v2f64(<2 x double>, <2 x double>, <2 x double>) #0 +; CHECK: declare <2 x double> @llvm.fmuladd.v2f64(<2 x double>, <2 x double>, <2 x double>) #0 +; CHECK: declare <2 x double> @llvm.cos.v2f64(<2 x double>) #1 +; CHECK: declare <2 x double> @llvm.powi.v2f64(<2 x double>, i32) #1 +; CHECK: attributes #0 = { nounwind readnone } +; CHECK: attributes #1 = { nounwind readonly } diff --git a/test/Transforms/BBVectorize/simple-ldstr.ll b/test/Transforms/BBVectorize/simple-ldstr.ll index 7dd77c933f6d..8e51d297e8ec 100644 --- a/test/Transforms/BBVectorize/simple-ldstr.ll +++ b/test/Transforms/BBVectorize/simple-ldstr.ll @@ -1,6 +1,6 @@ target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128" -; RUN: opt < %s -bb-vectorize -bb-vectorize-req-chain-depth=3 -instcombine -gvn -S | FileCheck %s -; RUN: opt < %s -bb-vectorize -bb-vectorize-req-chain-depth=3 -bb-vectorize-aligned-only -instcombine -gvn -S | FileCheck %s -check-prefix=CHECK-AO +; RUN: opt < %s -bb-vectorize -bb-vectorize-req-chain-depth=3 -bb-vectorize-ignore-target-info -instcombine -gvn -S | FileCheck %s +; RUN: opt < %s -bb-vectorize -bb-vectorize-req-chain-depth=3 -bb-vectorize-aligned-only -bb-vectorize-ignore-target-info -instcombine -gvn -S | FileCheck %s -check-prefix=CHECK-AO ; Simple 3-pair chain with loads and stores define void @test1(double* %a, double* %b, double* %c) nounwind uwtable readonly { diff --git a/test/Transforms/BBVectorize/simple-sel.ll b/test/Transforms/BBVectorize/simple-sel.ll index 15ecb597025a..8caccfd32c34 100644 --- a/test/Transforms/BBVectorize/simple-sel.ll +++ b/test/Transforms/BBVectorize/simple-sel.ll @@ -1,6 +1,6 @@ target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128" -; RUN: opt < %s -bb-vectorize -bb-vectorize-req-chain-depth=3 -instcombine -gvn -S | FileCheck %s -; RUN: opt < %s -bb-vectorize -bb-vectorize-req-chain-depth=3 -bb-vectorize-no-bools -instcombine -gvn -S | FileCheck %s -check-prefix=CHECK-NB +; RUN: opt < %s -bb-vectorize -bb-vectorize-req-chain-depth=3 -bb-vectorize-ignore-target-info -instcombine -gvn -S | FileCheck %s +; RUN: opt < %s -bb-vectorize -bb-vectorize-req-chain-depth=3 -bb-vectorize-no-bools -bb-vectorize-ignore-target-info -instcombine -gvn -S | FileCheck %s -check-prefix=CHECK-NB ; Basic depth-3 chain with select define double @test1(double %A1, double %A2, double %B1, double %B2, i1 %C1, i1 %C2) { diff --git a/test/Transforms/BBVectorize/simple.ll b/test/Transforms/BBVectorize/simple.ll index 3527ae75b457..a447908d16cc 100644 --- a/test/Transforms/BBVectorize/simple.ll +++ b/test/Transforms/BBVectorize/simple.ll @@ -1,5 +1,5 @@ target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128" -; RUN: opt < %s -bb-vectorize -bb-vectorize-req-chain-depth=3 -instcombine -gvn -S | FileCheck %s +; RUN: opt < %s -bb-vectorize -bb-vectorize-req-chain-depth=3 -bb-vectorize-ignore-target-info -instcombine -gvn -S | FileCheck %s ; Basic depth-3 chain define double @test1(double %A1, double %A2, double %B1, double %B2) { diff --git a/test/Transforms/BBVectorize/simple3.ll b/test/Transforms/BBVectorize/simple3.ll index 153be73f832f..78bcc9f83080 100644 --- a/test/Transforms/BBVectorize/simple3.ll +++ b/test/Transforms/BBVectorize/simple3.ll @@ -1,5 +1,5 @@ target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128" -; RUN: opt < %s -bb-vectorize -bb-vectorize-req-chain-depth=3 -bb-vectorize-vector-bits=192 -instcombine -gvn -S | FileCheck %s +; RUN: opt < %s -bb-vectorize -bb-vectorize-req-chain-depth=3 -bb-vectorize-vector-bits=192 -bb-vectorize-ignore-target-info -instcombine -gvn -S | FileCheck %s ; Basic depth-3 chain define double @test1(double %A1, double %A2, double %A3, double %B1, double %B2, double %B3) { |
