mirror of
https://github.com/c64scene-ar/llvm-6502.git
synced 2024-12-15 04:30:12 +00:00
8b2b8a1835
This update was done with the following bash script:
find test/CodeGen -name "*.ll" | \
while read NAME; do
echo "$NAME"
if ! grep -q "^; *RUN: *llc.*debug" $NAME; then
TEMP=`mktemp -t temp`
cp $NAME $TEMP
sed -n "s/^define [^@]*@\([A-Za-z0-9_]*\)(.*$/\1/p" < $NAME | \
while read FUNC; do
sed -i '' "s/;\(.*\)\([A-Za-z0-9_-]*\):\( *\)$FUNC: *\$/;\1\2-LABEL:\3$FUNC:/g" $TEMP
done
sed -i '' "s/;\(.*\)-LABEL-LABEL:/;\1-LABEL:/" $TEMP
sed -i '' "s/;\(.*\)-NEXT-LABEL:/;\1-NEXT:/" $TEMP
sed -i '' "s/;\(.*\)-NOT-LABEL:/;\1-NOT:/" $TEMP
sed -i '' "s/;\(.*\)-DAG-LABEL:/;\1-DAG:/" $TEMP
mv $TEMP $NAME
fi
done
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@186280 91177308-0d34-0410-b5e6-96231b3b80d8
199 lines
5.6 KiB
LLVM
199 lines
5.6 KiB
LLVM
; RUN: llc < %s -march=arm -mattr=+neon | FileCheck %s
|
|
; RUN: llc < %s -march=arm -mattr=+neon -regalloc=basic | FileCheck %s
|
|
|
|
define <8 x i8> @vcgts8(<8 x i8>* %A, <8 x i8>* %B) nounwind {
|
|
;CHECK-LABEL: vcgts8:
|
|
;CHECK: vcgt.s8
|
|
%tmp1 = load <8 x i8>* %A
|
|
%tmp2 = load <8 x i8>* %B
|
|
%tmp3 = icmp sgt <8 x i8> %tmp1, %tmp2
|
|
%tmp4 = sext <8 x i1> %tmp3 to <8 x i8>
|
|
ret <8 x i8> %tmp4
|
|
}
|
|
|
|
define <4 x i16> @vcgts16(<4 x i16>* %A, <4 x i16>* %B) nounwind {
|
|
;CHECK-LABEL: vcgts16:
|
|
;CHECK: vcgt.s16
|
|
%tmp1 = load <4 x i16>* %A
|
|
%tmp2 = load <4 x i16>* %B
|
|
%tmp3 = icmp sgt <4 x i16> %tmp1, %tmp2
|
|
%tmp4 = sext <4 x i1> %tmp3 to <4 x i16>
|
|
ret <4 x i16> %tmp4
|
|
}
|
|
|
|
define <2 x i32> @vcgts32(<2 x i32>* %A, <2 x i32>* %B) nounwind {
|
|
;CHECK-LABEL: vcgts32:
|
|
;CHECK: vcgt.s32
|
|
%tmp1 = load <2 x i32>* %A
|
|
%tmp2 = load <2 x i32>* %B
|
|
%tmp3 = icmp sgt <2 x i32> %tmp1, %tmp2
|
|
%tmp4 = sext <2 x i1> %tmp3 to <2 x i32>
|
|
ret <2 x i32> %tmp4
|
|
}
|
|
|
|
define <8 x i8> @vcgtu8(<8 x i8>* %A, <8 x i8>* %B) nounwind {
|
|
;CHECK-LABEL: vcgtu8:
|
|
;CHECK: vcgt.u8
|
|
%tmp1 = load <8 x i8>* %A
|
|
%tmp2 = load <8 x i8>* %B
|
|
%tmp3 = icmp ugt <8 x i8> %tmp1, %tmp2
|
|
%tmp4 = sext <8 x i1> %tmp3 to <8 x i8>
|
|
ret <8 x i8> %tmp4
|
|
}
|
|
|
|
define <4 x i16> @vcgtu16(<4 x i16>* %A, <4 x i16>* %B) nounwind {
|
|
;CHECK-LABEL: vcgtu16:
|
|
;CHECK: vcgt.u16
|
|
%tmp1 = load <4 x i16>* %A
|
|
%tmp2 = load <4 x i16>* %B
|
|
%tmp3 = icmp ugt <4 x i16> %tmp1, %tmp2
|
|
%tmp4 = sext <4 x i1> %tmp3 to <4 x i16>
|
|
ret <4 x i16> %tmp4
|
|
}
|
|
|
|
define <2 x i32> @vcgtu32(<2 x i32>* %A, <2 x i32>* %B) nounwind {
|
|
;CHECK-LABEL: vcgtu32:
|
|
;CHECK: vcgt.u32
|
|
%tmp1 = load <2 x i32>* %A
|
|
%tmp2 = load <2 x i32>* %B
|
|
%tmp3 = icmp ugt <2 x i32> %tmp1, %tmp2
|
|
%tmp4 = sext <2 x i1> %tmp3 to <2 x i32>
|
|
ret <2 x i32> %tmp4
|
|
}
|
|
|
|
define <2 x i32> @vcgtf32(<2 x float>* %A, <2 x float>* %B) nounwind {
|
|
;CHECK-LABEL: vcgtf32:
|
|
;CHECK: vcgt.f32
|
|
%tmp1 = load <2 x float>* %A
|
|
%tmp2 = load <2 x float>* %B
|
|
%tmp3 = fcmp ogt <2 x float> %tmp1, %tmp2
|
|
%tmp4 = sext <2 x i1> %tmp3 to <2 x i32>
|
|
ret <2 x i32> %tmp4
|
|
}
|
|
|
|
define <16 x i8> @vcgtQs8(<16 x i8>* %A, <16 x i8>* %B) nounwind {
|
|
;CHECK-LABEL: vcgtQs8:
|
|
;CHECK: vcgt.s8
|
|
%tmp1 = load <16 x i8>* %A
|
|
%tmp2 = load <16 x i8>* %B
|
|
%tmp3 = icmp sgt <16 x i8> %tmp1, %tmp2
|
|
%tmp4 = sext <16 x i1> %tmp3 to <16 x i8>
|
|
ret <16 x i8> %tmp4
|
|
}
|
|
|
|
define <8 x i16> @vcgtQs16(<8 x i16>* %A, <8 x i16>* %B) nounwind {
|
|
;CHECK-LABEL: vcgtQs16:
|
|
;CHECK: vcgt.s16
|
|
%tmp1 = load <8 x i16>* %A
|
|
%tmp2 = load <8 x i16>* %B
|
|
%tmp3 = icmp sgt <8 x i16> %tmp1, %tmp2
|
|
%tmp4 = sext <8 x i1> %tmp3 to <8 x i16>
|
|
ret <8 x i16> %tmp4
|
|
}
|
|
|
|
define <4 x i32> @vcgtQs32(<4 x i32>* %A, <4 x i32>* %B) nounwind {
|
|
;CHECK-LABEL: vcgtQs32:
|
|
;CHECK: vcgt.s32
|
|
%tmp1 = load <4 x i32>* %A
|
|
%tmp2 = load <4 x i32>* %B
|
|
%tmp3 = icmp sgt <4 x i32> %tmp1, %tmp2
|
|
%tmp4 = sext <4 x i1> %tmp3 to <4 x i32>
|
|
ret <4 x i32> %tmp4
|
|
}
|
|
|
|
define <16 x i8> @vcgtQu8(<16 x i8>* %A, <16 x i8>* %B) nounwind {
|
|
;CHECK-LABEL: vcgtQu8:
|
|
;CHECK: vcgt.u8
|
|
%tmp1 = load <16 x i8>* %A
|
|
%tmp2 = load <16 x i8>* %B
|
|
%tmp3 = icmp ugt <16 x i8> %tmp1, %tmp2
|
|
%tmp4 = sext <16 x i1> %tmp3 to <16 x i8>
|
|
ret <16 x i8> %tmp4
|
|
}
|
|
|
|
define <8 x i16> @vcgtQu16(<8 x i16>* %A, <8 x i16>* %B) nounwind {
|
|
;CHECK-LABEL: vcgtQu16:
|
|
;CHECK: vcgt.u16
|
|
%tmp1 = load <8 x i16>* %A
|
|
%tmp2 = load <8 x i16>* %B
|
|
%tmp3 = icmp ugt <8 x i16> %tmp1, %tmp2
|
|
%tmp4 = sext <8 x i1> %tmp3 to <8 x i16>
|
|
ret <8 x i16> %tmp4
|
|
}
|
|
|
|
define <4 x i32> @vcgtQu32(<4 x i32>* %A, <4 x i32>* %B) nounwind {
|
|
;CHECK-LABEL: vcgtQu32:
|
|
;CHECK: vcgt.u32
|
|
%tmp1 = load <4 x i32>* %A
|
|
%tmp2 = load <4 x i32>* %B
|
|
%tmp3 = icmp ugt <4 x i32> %tmp1, %tmp2
|
|
%tmp4 = sext <4 x i1> %tmp3 to <4 x i32>
|
|
ret <4 x i32> %tmp4
|
|
}
|
|
|
|
define <4 x i32> @vcgtQf32(<4 x float>* %A, <4 x float>* %B) nounwind {
|
|
;CHECK-LABEL: vcgtQf32:
|
|
;CHECK: vcgt.f32
|
|
%tmp1 = load <4 x float>* %A
|
|
%tmp2 = load <4 x float>* %B
|
|
%tmp3 = fcmp ogt <4 x float> %tmp1, %tmp2
|
|
%tmp4 = sext <4 x i1> %tmp3 to <4 x i32>
|
|
ret <4 x i32> %tmp4
|
|
}
|
|
|
|
define <2 x i32> @vacgtf32(<2 x float>* %A, <2 x float>* %B) nounwind {
|
|
;CHECK-LABEL: vacgtf32:
|
|
;CHECK: vacgt.f32
|
|
%tmp1 = load <2 x float>* %A
|
|
%tmp2 = load <2 x float>* %B
|
|
%tmp3 = call <2 x i32> @llvm.arm.neon.vacgtd(<2 x float> %tmp1, <2 x float> %tmp2)
|
|
ret <2 x i32> %tmp3
|
|
}
|
|
|
|
define <4 x i32> @vacgtQf32(<4 x float>* %A, <4 x float>* %B) nounwind {
|
|
;CHECK-LABEL: vacgtQf32:
|
|
;CHECK: vacgt.f32
|
|
%tmp1 = load <4 x float>* %A
|
|
%tmp2 = load <4 x float>* %B
|
|
%tmp3 = call <4 x i32> @llvm.arm.neon.vacgtq(<4 x float> %tmp1, <4 x float> %tmp2)
|
|
ret <4 x i32> %tmp3
|
|
}
|
|
|
|
; rdar://7923010
|
|
define <4 x i32> @vcgt_zext(<4 x float>* %A, <4 x float>* %B) nounwind {
|
|
;CHECK-LABEL: vcgt_zext:
|
|
;CHECK: vmov.i32 [[Q0:q[0-9]+]], #0x1
|
|
;CHECK: vcgt.f32 [[Q1:q[0-9]+]]
|
|
;CHECK: vand [[Q2:q[0-9]+]], [[Q1]], [[Q0]]
|
|
%tmp1 = load <4 x float>* %A
|
|
%tmp2 = load <4 x float>* %B
|
|
%tmp3 = fcmp ogt <4 x float> %tmp1, %tmp2
|
|
%tmp4 = zext <4 x i1> %tmp3 to <4 x i32>
|
|
ret <4 x i32> %tmp4
|
|
}
|
|
|
|
declare <2 x i32> @llvm.arm.neon.vacgtd(<2 x float>, <2 x float>) nounwind readnone
|
|
declare <4 x i32> @llvm.arm.neon.vacgtq(<4 x float>, <4 x float>) nounwind readnone
|
|
|
|
define <8 x i8> @vcgti8Z(<8 x i8>* %A) nounwind {
|
|
;CHECK-LABEL: vcgti8Z:
|
|
;CHECK-NOT: vmov
|
|
;CHECK-NOT: vmvn
|
|
;CHECK: vcgt.s8
|
|
%tmp1 = load <8 x i8>* %A
|
|
%tmp3 = icmp sgt <8 x i8> %tmp1, <i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0>
|
|
%tmp4 = sext <8 x i1> %tmp3 to <8 x i8>
|
|
ret <8 x i8> %tmp4
|
|
}
|
|
|
|
define <8 x i8> @vclti8Z(<8 x i8>* %A) nounwind {
|
|
;CHECK-LABEL: vclti8Z:
|
|
;CHECK-NOT: vmov
|
|
;CHECK-NOT: vmvn
|
|
;CHECK: vclt.s8
|
|
%tmp1 = load <8 x i8>* %A
|
|
%tmp3 = icmp slt <8 x i8> %tmp1, <i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0>
|
|
%tmp4 = sext <8 x i1> %tmp3 to <8 x i8>
|
|
ret <8 x i8> %tmp4
|
|
}
|