Hal Finkel 66d1fa6f4b Add a llvm.copysign intrinsic
This adds a llvm.copysign intrinsic; We already have Libfunc recognition for
copysign (which is turned into the FCOPYSIGN SDAG node). In order to
autovectorize calls to copysign in the loop vectorizer, we need a corresponding
intrinsic as well.

In addition to the expected changes to the language reference, the loop
vectorizer, BasicTTI, and the SDAG builder (the intrinsic is transformed into
an FCOPYSIGN node, just like the function call), this also adds FCOPYSIGN to a
few lists in LegalizeVector{Ops,Types} so that vector copysigns can be
expanded.

In TargetLoweringBase::initActions, I've made the default action for FCOPYSIGN
be Expand for vector types. This seems correct for all in-tree targets, and I
think is the right thing to do because, previously, there was no way to generate
vector-values FCOPYSIGN nodes (and most targets don't specify an action for
vector-typed FCOPYSIGN).

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@188728 91177308-0d34-0410-b5e6-96231b3b80d8
2013-08-19 23:35:46 +00:00

1035 lines
40 KiB
LLVM

; RUN: opt < %s -loop-vectorize -force-vector-unroll=1 -force-vector-width=4 -dce -instcombine -S | FileCheck %s
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"
;CHECK-LABEL: @sqrt_f32(
;CHECK: llvm.sqrt.v4f32
;CHECK: ret void
define void @sqrt_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
entry:
%cmp6 = icmp sgt i32 %n, 0
br i1 %cmp6, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
%0 = load float* %arrayidx, align 4
%call = tail call float @llvm.sqrt.f32(float %0) nounwind readnone
%arrayidx2 = getelementptr inbounds float* %x, i64 %indvars.iv
store float %call, float* %arrayidx2, align 4
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare float @llvm.sqrt.f32(float) nounwind readnone
;CHECK-LABEL: @sqrt_f64(
;CHECK: llvm.sqrt.v4f64
;CHECK: ret void
define void @sqrt_f64(i32 %n, double* noalias %y, double* noalias %x) nounwind uwtable {
entry:
%cmp6 = icmp sgt i32 %n, 0
br i1 %cmp6, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
%0 = load double* %arrayidx, align 8
%call = tail call double @llvm.sqrt.f64(double %0) nounwind readnone
%arrayidx2 = getelementptr inbounds double* %x, i64 %indvars.iv
store double %call, double* %arrayidx2, align 8
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare double @llvm.sqrt.f64(double) nounwind readnone
;CHECK-LABEL: @sin_f32(
;CHECK: llvm.sin.v4f32
;CHECK: ret void
define void @sin_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
entry:
%cmp6 = icmp sgt i32 %n, 0
br i1 %cmp6, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
%0 = load float* %arrayidx, align 4
%call = tail call float @llvm.sin.f32(float %0) nounwind readnone
%arrayidx2 = getelementptr inbounds float* %x, i64 %indvars.iv
store float %call, float* %arrayidx2, align 4
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare float @llvm.sin.f32(float) nounwind readnone
;CHECK-LABEL: @sin_f64(
;CHECK: llvm.sin.v4f64
;CHECK: ret void
define void @sin_f64(i32 %n, double* noalias %y, double* noalias %x) nounwind uwtable {
entry:
%cmp6 = icmp sgt i32 %n, 0
br i1 %cmp6, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
%0 = load double* %arrayidx, align 8
%call = tail call double @llvm.sin.f64(double %0) nounwind readnone
%arrayidx2 = getelementptr inbounds double* %x, i64 %indvars.iv
store double %call, double* %arrayidx2, align 8
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare double @llvm.sin.f64(double) nounwind readnone
;CHECK-LABEL: @cos_f32(
;CHECK: llvm.cos.v4f32
;CHECK: ret void
define void @cos_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
entry:
%cmp6 = icmp sgt i32 %n, 0
br i1 %cmp6, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
%0 = load float* %arrayidx, align 4
%call = tail call float @llvm.cos.f32(float %0) nounwind readnone
%arrayidx2 = getelementptr inbounds float* %x, i64 %indvars.iv
store float %call, float* %arrayidx2, align 4
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare float @llvm.cos.f32(float) nounwind readnone
;CHECK-LABEL: @cos_f64(
;CHECK: llvm.cos.v4f64
;CHECK: ret void
define void @cos_f64(i32 %n, double* noalias %y, double* noalias %x) nounwind uwtable {
entry:
%cmp6 = icmp sgt i32 %n, 0
br i1 %cmp6, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
%0 = load double* %arrayidx, align 8
%call = tail call double @llvm.cos.f64(double %0) nounwind readnone
%arrayidx2 = getelementptr inbounds double* %x, i64 %indvars.iv
store double %call, double* %arrayidx2, align 8
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare double @llvm.cos.f64(double) nounwind readnone
;CHECK-LABEL: @exp_f32(
;CHECK: llvm.exp.v4f32
;CHECK: ret void
define void @exp_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
entry:
%cmp6 = icmp sgt i32 %n, 0
br i1 %cmp6, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
%0 = load float* %arrayidx, align 4
%call = tail call float @llvm.exp.f32(float %0) nounwind readnone
%arrayidx2 = getelementptr inbounds float* %x, i64 %indvars.iv
store float %call, float* %arrayidx2, align 4
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare float @llvm.exp.f32(float) nounwind readnone
;CHECK-LABEL: @exp_f64(
;CHECK: llvm.exp.v4f64
;CHECK: ret void
define void @exp_f64(i32 %n, double* noalias %y, double* noalias %x) nounwind uwtable {
entry:
%cmp6 = icmp sgt i32 %n, 0
br i1 %cmp6, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
%0 = load double* %arrayidx, align 8
%call = tail call double @llvm.exp.f64(double %0) nounwind readnone
%arrayidx2 = getelementptr inbounds double* %x, i64 %indvars.iv
store double %call, double* %arrayidx2, align 8
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare double @llvm.exp.f64(double) nounwind readnone
;CHECK-LABEL: @exp2_f32(
;CHECK: llvm.exp2.v4f32
;CHECK: ret void
define void @exp2_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
entry:
%cmp6 = icmp sgt i32 %n, 0
br i1 %cmp6, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
%0 = load float* %arrayidx, align 4
%call = tail call float @llvm.exp2.f32(float %0) nounwind readnone
%arrayidx2 = getelementptr inbounds float* %x, i64 %indvars.iv
store float %call, float* %arrayidx2, align 4
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare float @llvm.exp2.f32(float) nounwind readnone
;CHECK-LABEL: @exp2_f64(
;CHECK: llvm.exp2.v4f64
;CHECK: ret void
define void @exp2_f64(i32 %n, double* noalias %y, double* noalias %x) nounwind uwtable {
entry:
%cmp6 = icmp sgt i32 %n, 0
br i1 %cmp6, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
%0 = load double* %arrayidx, align 8
%call = tail call double @llvm.exp2.f64(double %0) nounwind readnone
%arrayidx2 = getelementptr inbounds double* %x, i64 %indvars.iv
store double %call, double* %arrayidx2, align 8
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare double @llvm.exp2.f64(double) nounwind readnone
;CHECK-LABEL: @log_f32(
;CHECK: llvm.log.v4f32
;CHECK: ret void
define void @log_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
entry:
%cmp6 = icmp sgt i32 %n, 0
br i1 %cmp6, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
%0 = load float* %arrayidx, align 4
%call = tail call float @llvm.log.f32(float %0) nounwind readnone
%arrayidx2 = getelementptr inbounds float* %x, i64 %indvars.iv
store float %call, float* %arrayidx2, align 4
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare float @llvm.log.f32(float) nounwind readnone
;CHECK-LABEL: @log_f64(
;CHECK: llvm.log.v4f64
;CHECK: ret void
define void @log_f64(i32 %n, double* noalias %y, double* noalias %x) nounwind uwtable {
entry:
%cmp6 = icmp sgt i32 %n, 0
br i1 %cmp6, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
%0 = load double* %arrayidx, align 8
%call = tail call double @llvm.log.f64(double %0) nounwind readnone
%arrayidx2 = getelementptr inbounds double* %x, i64 %indvars.iv
store double %call, double* %arrayidx2, align 8
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare double @llvm.log.f64(double) nounwind readnone
;CHECK-LABEL: @log10_f32(
;CHECK: llvm.log10.v4f32
;CHECK: ret void
define void @log10_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
entry:
%cmp6 = icmp sgt i32 %n, 0
br i1 %cmp6, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
%0 = load float* %arrayidx, align 4
%call = tail call float @llvm.log10.f32(float %0) nounwind readnone
%arrayidx2 = getelementptr inbounds float* %x, i64 %indvars.iv
store float %call, float* %arrayidx2, align 4
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare float @llvm.log10.f32(float) nounwind readnone
;CHECK-LABEL: @log10_f64(
;CHECK: llvm.log10.v4f64
;CHECK: ret void
define void @log10_f64(i32 %n, double* noalias %y, double* noalias %x) nounwind uwtable {
entry:
%cmp6 = icmp sgt i32 %n, 0
br i1 %cmp6, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
%0 = load double* %arrayidx, align 8
%call = tail call double @llvm.log10.f64(double %0) nounwind readnone
%arrayidx2 = getelementptr inbounds double* %x, i64 %indvars.iv
store double %call, double* %arrayidx2, align 8
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare double @llvm.log10.f64(double) nounwind readnone
;CHECK-LABEL: @log2_f32(
;CHECK: llvm.log2.v4f32
;CHECK: ret void
define void @log2_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
entry:
%cmp6 = icmp sgt i32 %n, 0
br i1 %cmp6, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
%0 = load float* %arrayidx, align 4
%call = tail call float @llvm.log2.f32(float %0) nounwind readnone
%arrayidx2 = getelementptr inbounds float* %x, i64 %indvars.iv
store float %call, float* %arrayidx2, align 4
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare float @llvm.log2.f32(float) nounwind readnone
;CHECK-LABEL: @log2_f64(
;CHECK: llvm.log2.v4f64
;CHECK: ret void
define void @log2_f64(i32 %n, double* noalias %y, double* noalias %x) nounwind uwtable {
entry:
%cmp6 = icmp sgt i32 %n, 0
br i1 %cmp6, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
%0 = load double* %arrayidx, align 8
%call = tail call double @llvm.log2.f64(double %0) nounwind readnone
%arrayidx2 = getelementptr inbounds double* %x, i64 %indvars.iv
store double %call, double* %arrayidx2, align 8
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare double @llvm.log2.f64(double) nounwind readnone
;CHECK-LABEL: @fabs_f32(
;CHECK: llvm.fabs.v4f32
;CHECK: ret void
define void @fabs_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
entry:
%cmp6 = icmp sgt i32 %n, 0
br i1 %cmp6, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
%0 = load float* %arrayidx, align 4
%call = tail call float @llvm.fabs.f32(float %0) nounwind readnone
%arrayidx2 = getelementptr inbounds float* %x, i64 %indvars.iv
store float %call, float* %arrayidx2, align 4
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare float @llvm.fabs.f32(float) nounwind readnone
define void @fabs_f64(i32 %n, double* noalias %y, double* noalias %x) nounwind uwtable {
entry:
%cmp6 = icmp sgt i32 %n, 0
br i1 %cmp6, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
%0 = load double* %arrayidx, align 8
%call = tail call double @llvm.fabs(double %0) nounwind readnone
%arrayidx2 = getelementptr inbounds double* %x, i64 %indvars.iv
store double %call, double* %arrayidx2, align 8
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare double @llvm.fabs(double) nounwind readnone
;CHECK-LABEL: @copysign_f32(
;CHECK: llvm.copysign.v4f32
;CHECK: ret void
define void @copysign_f32(i32 %n, float* noalias %y, float* noalias %x, float* noalias %z) nounwind uwtable {
entry:
%cmp6 = icmp sgt i32 %n, 0
br i1 %cmp6, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
%0 = load float* %arrayidx, align 4
%arrayidx1 = getelementptr inbounds float* %z, i64 %indvars.iv
%1 = load float* %arrayidx1, align 4
%call = tail call float @llvm.copysign.f32(float %0, float %1) nounwind readnone
%arrayidx2 = getelementptr inbounds float* %x, i64 %indvars.iv
store float %call, float* %arrayidx2, align 4
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare float @llvm.copysign.f32(float, float) nounwind readnone
define void @copysign_f64(i32 %n, double* noalias %y, double* noalias %x, double* noalias %z) nounwind uwtable {
entry:
%cmp6 = icmp sgt i32 %n, 0
br i1 %cmp6, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
%0 = load double* %arrayidx, align 8
%arrayidx1 = getelementptr inbounds double* %z, i64 %indvars.iv
%1 = load double* %arrayidx, align 8
%call = tail call double @llvm.copysign(double %0, double %1) nounwind readnone
%arrayidx2 = getelementptr inbounds double* %x, i64 %indvars.iv
store double %call, double* %arrayidx2, align 8
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare double @llvm.copysign(double, double) nounwind readnone
;CHECK-LABEL: @floor_f32(
;CHECK: llvm.floor.v4f32
;CHECK: ret void
define void @floor_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
entry:
%cmp6 = icmp sgt i32 %n, 0
br i1 %cmp6, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
%0 = load float* %arrayidx, align 4
%call = tail call float @llvm.floor.f32(float %0) nounwind readnone
%arrayidx2 = getelementptr inbounds float* %x, i64 %indvars.iv
store float %call, float* %arrayidx2, align 4
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare float @llvm.floor.f32(float) nounwind readnone
;CHECK-LABEL: @floor_f64(
;CHECK: llvm.floor.v4f64
;CHECK: ret void
define void @floor_f64(i32 %n, double* noalias %y, double* noalias %x) nounwind uwtable {
entry:
%cmp6 = icmp sgt i32 %n, 0
br i1 %cmp6, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
%0 = load double* %arrayidx, align 8
%call = tail call double @llvm.floor.f64(double %0) nounwind readnone
%arrayidx2 = getelementptr inbounds double* %x, i64 %indvars.iv
store double %call, double* %arrayidx2, align 8
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare double @llvm.floor.f64(double) nounwind readnone
;CHECK-LABEL: @ceil_f32(
;CHECK: llvm.ceil.v4f32
;CHECK: ret void
define void @ceil_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
entry:
%cmp6 = icmp sgt i32 %n, 0
br i1 %cmp6, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
%0 = load float* %arrayidx, align 4
%call = tail call float @llvm.ceil.f32(float %0) nounwind readnone
%arrayidx2 = getelementptr inbounds float* %x, i64 %indvars.iv
store float %call, float* %arrayidx2, align 4
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare float @llvm.ceil.f32(float) nounwind readnone
;CHECK-LABEL: @ceil_f64(
;CHECK: llvm.ceil.v4f64
;CHECK: ret void
define void @ceil_f64(i32 %n, double* noalias %y, double* noalias %x) nounwind uwtable {
entry:
%cmp6 = icmp sgt i32 %n, 0
br i1 %cmp6, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
%0 = load double* %arrayidx, align 8
%call = tail call double @llvm.ceil.f64(double %0) nounwind readnone
%arrayidx2 = getelementptr inbounds double* %x, i64 %indvars.iv
store double %call, double* %arrayidx2, align 8
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare double @llvm.ceil.f64(double) nounwind readnone
;CHECK-LABEL: @trunc_f32(
;CHECK: llvm.trunc.v4f32
;CHECK: ret void
define void @trunc_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
entry:
%cmp6 = icmp sgt i32 %n, 0
br i1 %cmp6, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
%0 = load float* %arrayidx, align 4
%call = tail call float @llvm.trunc.f32(float %0) nounwind readnone
%arrayidx2 = getelementptr inbounds float* %x, i64 %indvars.iv
store float %call, float* %arrayidx2, align 4
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare float @llvm.trunc.f32(float) nounwind readnone
;CHECK-LABEL: @trunc_f64(
;CHECK: llvm.trunc.v4f64
;CHECK: ret void
define void @trunc_f64(i32 %n, double* noalias %y, double* noalias %x) nounwind uwtable {
entry:
%cmp6 = icmp sgt i32 %n, 0
br i1 %cmp6, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
%0 = load double* %arrayidx, align 8
%call = tail call double @llvm.trunc.f64(double %0) nounwind readnone
%arrayidx2 = getelementptr inbounds double* %x, i64 %indvars.iv
store double %call, double* %arrayidx2, align 8
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare double @llvm.trunc.f64(double) nounwind readnone
;CHECK-LABEL: @rint_f32(
;CHECK: llvm.rint.v4f32
;CHECK: ret void
define void @rint_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
entry:
%cmp6 = icmp sgt i32 %n, 0
br i1 %cmp6, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
%0 = load float* %arrayidx, align 4
%call = tail call float @llvm.rint.f32(float %0) nounwind readnone
%arrayidx2 = getelementptr inbounds float* %x, i64 %indvars.iv
store float %call, float* %arrayidx2, align 4
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare float @llvm.rint.f32(float) nounwind readnone
;CHECK-LABEL: @rint_f64(
;CHECK: llvm.rint.v4f64
;CHECK: ret void
define void @rint_f64(i32 %n, double* noalias %y, double* noalias %x) nounwind uwtable {
entry:
%cmp6 = icmp sgt i32 %n, 0
br i1 %cmp6, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
%0 = load double* %arrayidx, align 8
%call = tail call double @llvm.rint.f64(double %0) nounwind readnone
%arrayidx2 = getelementptr inbounds double* %x, i64 %indvars.iv
store double %call, double* %arrayidx2, align 8
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare double @llvm.rint.f64(double) nounwind readnone
;CHECK-LABEL: @nearbyint_f32(
;CHECK: llvm.nearbyint.v4f32
;CHECK: ret void
define void @nearbyint_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
entry:
%cmp6 = icmp sgt i32 %n, 0
br i1 %cmp6, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
%0 = load float* %arrayidx, align 4
%call = tail call float @llvm.nearbyint.f32(float %0) nounwind readnone
%arrayidx2 = getelementptr inbounds float* %x, i64 %indvars.iv
store float %call, float* %arrayidx2, align 4
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare float @llvm.nearbyint.f32(float) nounwind readnone
;CHECK-LABEL: @nearbyint_f64(
;CHECK: llvm.nearbyint.v4f64
;CHECK: ret void
define void @nearbyint_f64(i32 %n, double* noalias %y, double* noalias %x) nounwind uwtable {
entry:
%cmp6 = icmp sgt i32 %n, 0
br i1 %cmp6, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
%0 = load double* %arrayidx, align 8
%call = tail call double @llvm.nearbyint.f64(double %0) nounwind readnone
%arrayidx2 = getelementptr inbounds double* %x, i64 %indvars.iv
store double %call, double* %arrayidx2, align 8
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare double @llvm.nearbyint.f64(double) nounwind readnone
;CHECK-LABEL: @round_f32(
;CHECK: llvm.round.v4f32
;CHECK: ret void
define void @round_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
entry:
%cmp6 = icmp sgt i32 %n, 0
br i1 %cmp6, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
%0 = load float* %arrayidx, align 4
%call = tail call float @llvm.round.f32(float %0) nounwind readnone
%arrayidx2 = getelementptr inbounds float* %x, i64 %indvars.iv
store float %call, float* %arrayidx2, align 4
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare float @llvm.round.f32(float) nounwind readnone
;CHECK-LABEL: @round_f64(
;CHECK: llvm.round.v4f64
;CHECK: ret void
define void @round_f64(i32 %n, double* noalias %y, double* noalias %x) nounwind uwtable {
entry:
%cmp6 = icmp sgt i32 %n, 0
br i1 %cmp6, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
%0 = load double* %arrayidx, align 8
%call = tail call double @llvm.round.f64(double %0) nounwind readnone
%arrayidx2 = getelementptr inbounds double* %x, i64 %indvars.iv
store double %call, double* %arrayidx2, align 8
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare double @llvm.round.f64(double) nounwind readnone
;CHECK-LABEL: @fma_f32(
;CHECK: llvm.fma.v4f32
;CHECK: ret void
define void @fma_f32(i32 %n, float* noalias %y, float* noalias %x, float* noalias %z, float* noalias %w) nounwind uwtable {
entry:
%cmp12 = icmp sgt i32 %n, 0
br i1 %cmp12, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
%0 = load float* %arrayidx, align 4
%arrayidx2 = getelementptr inbounds float* %w, i64 %indvars.iv
%1 = load float* %arrayidx2, align 4
%arrayidx4 = getelementptr inbounds float* %z, i64 %indvars.iv
%2 = load float* %arrayidx4, align 4
%3 = tail call float @llvm.fma.f32(float %0, float %2, float %1)
%arrayidx6 = getelementptr inbounds float* %x, i64 %indvars.iv
store float %3, float* %arrayidx6, align 4
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare float @llvm.fma.f32(float, float, float) nounwind readnone
;CHECK-LABEL: @fma_f64(
;CHECK: llvm.fma.v4f64
;CHECK: ret void
define void @fma_f64(i32 %n, double* noalias %y, double* noalias %x, double* noalias %z, double* noalias %w) nounwind uwtable {
entry:
%cmp12 = icmp sgt i32 %n, 0
br i1 %cmp12, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
%0 = load double* %arrayidx, align 8
%arrayidx2 = getelementptr inbounds double* %w, i64 %indvars.iv
%1 = load double* %arrayidx2, align 8
%arrayidx4 = getelementptr inbounds double* %z, i64 %indvars.iv
%2 = load double* %arrayidx4, align 8
%3 = tail call double @llvm.fma.f64(double %0, double %2, double %1)
%arrayidx6 = getelementptr inbounds double* %x, i64 %indvars.iv
store double %3, double* %arrayidx6, align 8
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare double @llvm.fma.f64(double, double, double) nounwind readnone
;CHECK-LABEL: @fmuladd_f32(
;CHECK: llvm.fmuladd.v4f32
;CHECK: ret void
define void @fmuladd_f32(i32 %n, float* noalias %y, float* noalias %x, float* noalias %z, float* noalias %w) nounwind uwtable {
entry:
%cmp12 = icmp sgt i32 %n, 0
br i1 %cmp12, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
%0 = load float* %arrayidx, align 4
%arrayidx2 = getelementptr inbounds float* %w, i64 %indvars.iv
%1 = load float* %arrayidx2, align 4
%arrayidx4 = getelementptr inbounds float* %z, i64 %indvars.iv
%2 = load float* %arrayidx4, align 4
%3 = tail call float @llvm.fmuladd.f32(float %0, float %2, float %1)
%arrayidx6 = getelementptr inbounds float* %x, i64 %indvars.iv
store float %3, float* %arrayidx6, align 4
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare float @llvm.fmuladd.f32(float, float, float) nounwind readnone
;CHECK-LABEL: @fmuladd_f64(
;CHECK: llvm.fmuladd.v4f64
;CHECK: ret void
define void @fmuladd_f64(i32 %n, double* noalias %y, double* noalias %x, double* noalias %z, double* noalias %w) nounwind uwtable {
entry:
%cmp12 = icmp sgt i32 %n, 0
br i1 %cmp12, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
%0 = load double* %arrayidx, align 8
%arrayidx2 = getelementptr inbounds double* %w, i64 %indvars.iv
%1 = load double* %arrayidx2, align 8
%arrayidx4 = getelementptr inbounds double* %z, i64 %indvars.iv
%2 = load double* %arrayidx4, align 8
%3 = tail call double @llvm.fmuladd.f64(double %0, double %2, double %1)
%arrayidx6 = getelementptr inbounds double* %x, i64 %indvars.iv
store double %3, double* %arrayidx6, align 8
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare double @llvm.fmuladd.f64(double, double, double) nounwind readnone
;CHECK-LABEL: @pow_f32(
;CHECK: llvm.pow.v4f32
;CHECK: ret void
define void @pow_f32(i32 %n, float* noalias %y, float* noalias %x, float* noalias %z) nounwind uwtable {
entry:
%cmp9 = icmp sgt i32 %n, 0
br i1 %cmp9, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
%0 = load float* %arrayidx, align 4
%arrayidx2 = getelementptr inbounds float* %z, i64 %indvars.iv
%1 = load float* %arrayidx2, align 4
%call = tail call float @llvm.pow.f32(float %0, float %1) nounwind readnone
%arrayidx4 = getelementptr inbounds float* %x, i64 %indvars.iv
store float %call, float* %arrayidx4, align 4
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
declare float @llvm.pow.f32(float, float) nounwind readnone
;CHECK-LABEL: @pow_f64(
;CHECK: llvm.pow.v4f64
;CHECK: ret void
define void @pow_f64(i32 %n, double* noalias %y, double* noalias %x, double* noalias %z) nounwind uwtable {
entry:
%cmp9 = icmp sgt i32 %n, 0
br i1 %cmp9, label %for.body, label %for.end
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
%arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
%0 = load double* %arrayidx, align 8
%arrayidx2 = getelementptr inbounds double* %z, i64 %indvars.iv
%1 = load double* %arrayidx2, align 8
%call = tail call double @llvm.pow.f64(double %0, double %1) nounwind readnone
%arrayidx4 = getelementptr inbounds double* %x, i64 %indvars.iv
store double %call, double* %arrayidx4, align 8
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, %n
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body, %entry
ret void
}
; CHECK: fabs_libm
; CHECK: call <4 x float> @llvm.fabs.v4f32
; CHECK: ret void
define void @fabs_libm(float* nocapture %x) nounwind {
entry:
br label %for.body
for.body: ; preds = %entry, %for.body
%indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
%arrayidx = getelementptr inbounds float* %x, i64 %indvars.iv
%0 = load float* %arrayidx, align 4
%call = tail call float @fabsf(float %0) nounwind readnone
store float %call, float* %arrayidx, align 4
%indvars.iv.next = add i64 %indvars.iv, 1
%lftr.wideiv = trunc i64 %indvars.iv.next to i32
%exitcond = icmp eq i32 %lftr.wideiv, 1024
br i1 %exitcond, label %for.end, label %for.body
for.end: ; preds = %for.body
ret void
}
declare float @fabsf(float) nounwind readnone
declare double @llvm.pow.f64(double, double) nounwind readnone