; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -mcpu=pwr9 -O3 -verify-machineinstrs -ppc-vsr-nums-as-vr \
; RUN: -mtriple=powerpc64le-unknown-linux-gnu \
; RUN: < %s | FileCheck %s --check-prefixes=P9LE
; RUN: llc -mcpu=pwr9 -O3 -verify-machineinstrs -ppc-vsr-nums-as-vr \
; RUN: -mtriple=powerpc64-unknown-linux-gnu \
; RUN: < %s | FileCheck %s --check-prefixes=P9BE
; RUN: llc -mcpu=pwr9 -O3 -verify-machineinstrs -ppc-vsr-nums-as-vr \
; RUN: -mtriple=powerpc64-ibm-aix-xcoff -vec-extabi \
; RUN: < %s | FileCheck %s --check-prefixes=P9BE-AIX
; RUN: llc -mcpu=pwr9 -O3 -verify-machineinstrs -ppc-vsr-nums-as-vr \
; RUN: -mtriple=powerpc-ibm-aix-xcoff -vec-extabi \
; RUN: < %s | FileCheck %s --check-prefixes=P9BE-AIX32
define void @test64(ptr nocapture readonly %pix2, i32 signext %i_pix2) {
; P9LE-LABEL: test64:
; P9LE: # %bb.0: # %entry
; P9LE-NEXT: add 5, 3, 4
; P9LE-NEXT: lfdx 0, 3, 4
; P9LE-NEXT: addis 3, 2, .LCPI0_0@toc@ha
; P9LE-NEXT: xxlxor 2, 2, 2
; P9LE-NEXT: vspltisw 4, 8
; P9LE-NEXT: lxsd 3, 4(5)
; P9LE-NEXT: addi 3, 3, .LCPI0_0@toc@l
; P9LE-NEXT: vadduwm 4, 4, 4
; P9LE-NEXT: lxv 1, 0(3)
; P9LE-NEXT: addis 3, 2, .LCPI0_1@toc@ha
; P9LE-NEXT: addi 3, 3, .LCPI0_1@toc@l
; P9LE-NEXT: xxperm 2, 0, 1
; P9LE-NEXT: lxv 0, 0(3)
; P9LE-NEXT: xxperm 3, 3, 0
; P9LE-NEXT: vnegw 3, 3
; P9LE-NEXT: vslw 3, 3, 4
; P9LE-NEXT: vsubuwm 2, 3, 2
; P9LE-NEXT: xxswapd 0, 2
; P9LE-NEXT: stxv 0, 0(3)
; P9LE-NEXT: blr
;
; P9BE-LABEL: test64:
; P9BE: # %bb.0: # %entry
; P9BE-NEXT: add 5, 3, 4
; P9BE-NEXT: lxsdx 2, 3, 4
; P9BE-NEXT: addis 3, 2, .LCPI0_0@toc@ha
; P9BE-NEXT: xxlxor 1, 1, 1
; P9BE-NEXT: vspltisw 4, 8
; P9BE-NEXT: lxsd 3, 4(5)
; P9BE-NEXT: addi 3, 3, .LCPI0_0@toc@l
; P9BE-NEXT: vadduwm 4, 4, 4
; P9BE-NEXT: lxv 0, 0(3)
; P9BE-NEXT: addis 3, 2, .LCPI0_1@toc@ha
; P9BE-NEXT: addi 3, 3, .LCPI0_1@toc@l
; P9BE-NEXT: xxperm 2, 1, 0
; P9BE-NEXT: lxv 0, 0(3)
; P9BE-NEXT: xxperm 3, 3, 0
; P9BE-NEXT: vnegw 3, 3
; P9BE-NEXT: vslw 3, 3, 4
; P9BE-NEXT: vsubuwm 2, 3, 2
; P9BE-NEXT: xxswapd 0, 2
; P9BE-NEXT: stxv 0, 0(3)
; P9BE-NEXT: blr
;
; P9BE-AIX-LABEL: test64:
; P9BE-AIX: # %bb.0: # %entry
; P9BE-AIX-NEXT: add 5, 3, 4
; P9BE-AIX-NEXT: lxsdx 2, 3, 4
; P9BE-AIX-NEXT: ld 3, L..C0(2) # %const.0
; P9BE-AIX-NEXT: xxlxor 1, 1, 1
; P9BE-AIX-NEXT: vspltisw 4, 8
; P9BE-AIX-NEXT: lxsd 3, 4(5)
; P9BE-AIX-NEXT: lxv 0, 0(3)
; P9BE-AIX-NEXT: ld 3, L..C1(2) # %const.1
; P9BE-AIX-NEXT: vadduwm 4, 4, 4
; P9BE-AIX-NEXT: xxperm 2, 1, 0
; P9BE-AIX-NEXT: lxv 0, 0(3)
; P9BE-AIX-NEXT: xxperm 3, 3, 0
; P9BE-AIX-NEXT: vnegw 3, 3
; P9BE-AIX-NEXT: vslw 3, 3, 4
; P9BE-AIX-NEXT: vsubuwm 2, 3, 2
; P9BE-AIX-NEXT: xxswapd 0, 2
; P9BE-AIX-NEXT: stxv 0, 0(3)
; P9BE-AIX-NEXT: blr
;
; P9BE-AIX32-LABEL: test64:
; P9BE-AIX32: # %bb.0: # %entry
; P9BE-AIX32-NEXT: add 5, 3, 4
; P9BE-AIX32-NEXT: lxvwsx 0, 3, 4
; P9BE-AIX32-NEXT: li 3, 4
; P9BE-AIX32-NEXT: xxlxor 2, 2, 2
; P9BE-AIX32-NEXT: vspltisw 4, 8
; P9BE-AIX32-NEXT: lxvwsx 1, 5, 3
; P9BE-AIX32-NEXT: lwz 3, L..C0(2) # %const.0
; P9BE-AIX32-NEXT: vadduwm 4, 4, 4
; P9BE-AIX32-NEXT: xxmrghw 2, 0, 1
; P9BE-AIX32-NEXT: lxv 0, 0(3)
; P9BE-AIX32-NEXT: li 3, 8
; P9BE-AIX32-NEXT: xxperm 2, 2, 0
; P9BE-AIX32-NEXT: lxvwsx 0, 5, 3
; P9BE-AIX32-NEXT: lwz 3, L..C1(2) # %const.1
; P9BE-AIX32-NEXT: xxmrghw 3, 1, 0
; P9BE-AIX32-NEXT: lxv 0, 0(3)
; P9BE-AIX32-NEXT: xxperm 3, 3, 0
; P9BE-AIX32-NEXT: vnegw 3, 3
; P9BE-AIX32-NEXT: vslw 3, 3, 4
; P9BE-AIX32-NEXT: vsubuwm 2, 3, 2
; P9BE-AIX32-NEXT: xxswapd 0, 2
; P9BE-AIX32-NEXT: stxv 0, 0(3)
; P9BE-AIX32-NEXT: blr
entry:
%idx.ext63 = sext i32 %i_pix2 to i64
%add.ptr64 = getelementptr inbounds i8, ptr %pix2, i64 %idx.ext63
%arrayidx5.1 = getelementptr inbounds i8, ptr %add.ptr64, i64 4
%0 = load <4 x i16>, ptr %add.ptr64, align 1
%reorder_shuffle117 = shufflevector <4 x i16> %0, <4 x i16> undef, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
%1 = zext <4 x i16> %reorder_shuffle117 to <4 x i32>
%2 = sub nsw <4 x i32> zeroinitializer, %1
%3 = load <4 x i16>, ptr %arrayidx5.1, align 1
%reorder_shuffle115 = shufflevector <4 x i16> %3, <4 x i16> undef, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
%4 = zext <4 x i16> %reorder_shuffle115 to <4 x i32>
%5 = sub nsw <4 x i32> zeroinitializer, %4
%6 = shl nsw <4 x i32> %5, <i32 16, i32 16, i32 16, i32 16>
%7 = add nsw <4 x i32> %6, %2
%8 = sub nsw <4 x i32> %7, zeroinitializer
%9 = shufflevector <4 x i32> undef, <4 x i32> %8, <4 x i32> <i32 2, i32 7, i32 0, i32 5>
%10 = add nsw <4 x i32> zeroinitializer, %9
%11 = shufflevector <4 x i32> %10, <4 x i32> undef, <4 x i32> <i32 0, i32 1, i32 6, i32 7>
store <4 x i32> %11, ptr undef, align 16
ret void
}
define void @test32(ptr nocapture readonly %pix2, i32 signext %i_pix2) {
; P9LE-LABEL: test32:
; P9LE: # %bb.0: # %entry
; P9LE-NEXT: add 5, 3, 4
; P9LE-NEXT: lxsiwzx 2, 3, 4
; P9LE-NEXT: addis 3, 2, .LCPI1_0@toc@ha
; P9LE-NEXT: xxlxor 0, 0, 0
; P9LE-NEXT: vspltisw 4, 8
; P9LE-NEXT: addi 3, 3, .LCPI1_0@toc@l
; P9LE-NEXT: lxv 1, 0(3)
; P9LE-NEXT: li 3, 4
; P9LE-NEXT: vadduwm 4, 4, 4
; P9LE-NEXT: lxsiwzx 3, 5, 3
; P9LE-NEXT: xxperm 2, 0, 1
; P9LE-NEXT: xxperm 3, 0, 1
; P9LE-NEXT: vnegw 3, 3
; P9LE-NEXT: vslw 3, 3, 4
; P9LE-NEXT: vsubuwm 2, 3, 2
; P9LE-NEXT: xxswapd 0, 2
; P9LE-NEXT: stxv 0, 0(3)
; P9LE-NEXT: blr
;
; P9BE-LABEL: test32:
; P9BE: # %bb.0: # %entry
; P9BE-NEXT: add 5, 3, 4
; P9BE-NEXT: lxsiwzx 2, 3, 4
; P9BE-NEXT: addis 3, 2, .LCPI1_0@toc@ha
; P9BE-NEXT: xxlxor 0, 0, 0
; P9BE-NEXT: vspltisw 4, 8
; P9BE-NEXT: addi 3, 3, .LCPI1_0@toc@l
; P9BE-NEXT: lxv 1, 0(3)
; P9BE-NEXT: li 3, 4
; P9BE-NEXT: vadduwm 4, 4, 4
; P9BE-NEXT: lxsiwzx 3, 5, 3
; P9BE-NEXT: xxperm 2, 0, 1
; P9BE-NEXT: xxperm 3, 0, 1
; P9BE-NEXT: vnegw 3, 3
; P9BE-NEXT: vslw 3, 3, 4
; P9BE-NEXT: vsubuwm 2, 3, 2
; P9BE-NEXT: xxswapd 0, 2
; P9BE-NEXT: stxv 0, 0(3)
; P9BE-NEXT: blr
;
; P9BE-AIX-LABEL: test32:
; P9BE-AIX: # %bb.0: # %entry
; P9BE-AIX-NEXT: add 5, 3, 4
; P9BE-AIX-NEXT: lxsiwzx 2, 3, 4
; P9BE-AIX-NEXT: ld 3, L..C2(2) # %const.0
; P9BE-AIX-NEXT: xxlxor 0, 0, 0
; P9BE-AIX-NEXT: vspltisw 4, 8
; P9BE-AIX-NEXT: lxv 1, 0(3)
; P9BE-AIX-NEXT: li 3, 4
; P9BE-AIX-NEXT: vadduwm 4, 4, 4
; P9BE-AIX-NEXT: lxsiwzx 3, 5, 3
; P9BE-AIX-NEXT: xxperm 2, 0, 1
; P9BE-AIX-NEXT: xxperm 3, 0, 1
; P9BE-AIX-NEXT: vnegw 3, 3
; P9BE-AIX-NEXT: vslw 3, 3, 4
; P9BE-AIX-NEXT: vsubuwm 2, 3, 2
; P9BE-AIX-NEXT: xxswapd 0, 2
; P9BE-AIX-NEXT: stxv 0, 0(3)
; P9BE-AIX-NEXT: blr
;
; P9BE-AIX32-LABEL: test32:
; P9BE-AIX32: # %bb.0: # %entry
; P9BE-AIX32-NEXT: add 5, 3, 4
; P9BE-AIX32-NEXT: lxsiwzx 2, 3, 4
; P9BE-AIX32-NEXT: lwz 3, L..C2(2) # %const.0
; P9BE-AIX32-NEXT: xxlxor 0, 0, 0
; P9BE-AIX32-NEXT: vspltisw 4, 8
; P9BE-AIX32-NEXT: lxv 1, 0(3)
; P9BE-AIX32-NEXT: li 3, 4
; P9BE-AIX32-NEXT: vadduwm 4, 4, 4
; P9BE-AIX32-NEXT: lxsiwzx 3, 5, 3
; P9BE-AIX32-NEXT: xxperm 2, 0, 1
; P9BE-AIX32-NEXT: xxperm 3, 0, 1
; P9BE-AIX32-NEXT: vnegw 3, 3
; P9BE-AIX32-NEXT: vslw 3, 3, 4
; P9BE-AIX32-NEXT: vsubuwm 2, 3, 2
; P9BE-AIX32-NEXT: xxswapd 0, 2
; P9BE-AIX32-NEXT: stxv 0, 0(3)
; P9BE-AIX32-NEXT: blr
entry:
%idx.ext63 = sext i32 %i_pix2 to i64
%add.ptr64 = getelementptr inbounds i8, ptr %pix2, i64 %idx.ext63
%arrayidx5.1 = getelementptr inbounds i8, ptr %add.ptr64, i64 4
%0 = load <4 x i8>, ptr %add.ptr64, align 1
%reorder_shuffle117 = shufflevector <4 x i8> %0, <4 x i8> undef, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
%1 = zext <4 x i8> %reorder_shuffle117 to <4 x i32>
%2 = sub nsw <4 x i32> zeroinitializer, %1
%3 = load <4 x i8>, ptr %arrayidx5.1, align 1
%reorder_shuffle115 = shufflevector <4 x i8> %3, <4 x i8> undef, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
%4 = zext <4 x i8> %reorder_shuffle115 to <4 x i32>
%5 = sub nsw <4 x i32> zeroinitializer, %4
%6 = shl nsw <4 x i32> %5, <i32 16, i32 16, i32 16, i32 16>
%7 = add nsw <4 x i32> %6, %2
%8 = sub nsw <4 x i32> %7, zeroinitializer
%9 = shufflevector <4 x i32> undef, <4 x i32> %8, <4 x i32> <i32 2, i32 7, i32 0, i32 5>
%10 = add nsw <4 x i32> zeroinitializer, %9
%11 = shufflevector <4 x i32> %10, <4 x i32> undef, <4 x i32> <i32 0, i32 1, i32 6, i32 7>
store <4 x i32> %11, ptr undef, align 16
ret void
}
define void @test16(ptr nocapture readonly %sums, i32 signext %delta, i32 signext %thresh) {
; P9LE-LABEL: test16:
; P9LE: # %bb.0: # %entry
; P9LE-NEXT: sldi 4, 4, 1
; P9LE-NEXT: li 7, 16
; P9LE-NEXT: add 6, 3, 4
; P9LE-NEXT: lxsihzx 4, 3, 4
; P9LE-NEXT: addis 3, 2, .LCPI2_0@toc@ha
; P9LE-NEXT: lxsihzx 2, 6, 7
; P9LE-NEXT: li 6, 0
; P9LE-NEXT: addi 3, 3, .LCPI2_0@toc@l
; P9LE-NEXT: mtvsrd 3, 6
; P9LE-NEXT: lxv 0, 0(3)
; P9LE-NEXT: li 3, 0
; P9LE-NEXT: vmrghh 4, 3, 4
; P9LE-NEXT: vmrghh 2, 3, 2
; P9LE-NEXT: vsplth 3, 3, 3
; P9LE-NEXT: xxmrglw 3, 4, 3
; P9LE-NEXT: xxperm 3, 2, 0
; P9LE-NEXT: xxspltw 2, 3, 2
; P9LE-NEXT: vadduwm 2, 3, 2
; P9LE-NEXT: vextuwrx 3, 3, 2
; P9LE-NEXT: cmpw 3, 5
; P9LE-NEXT: bgelr+ 0
; P9LE-NEXT: # %bb.1: # %if.then
;
; P9BE-LABEL: test16:
; P9BE: # %bb.0: # %entry
; P9BE-NEXT: sldi 4, 4, 1
; P9BE-NEXT: li 7, 16
; P9BE-NEXT: add 6, 3, 4
; P9BE-NEXT: lxsihzx 1, 3, 4
; P9BE-NEXT: addis 3, 2, .LCPI2_1@toc@ha
; P9BE-NEXT: lxsihzx 2, 6, 7
; P9BE-NEXT: addis 6, 2, .LCPI2_0@toc@ha
; P9BE-NEXT: addi 3, 3, .LCPI2_1@toc@l
; P9BE-NEXT: addi 6, 6, .LCPI2_0@toc@l
; P9BE-NEXT: lxv 0, 0(6)
; P9BE-NEXT: li 6, 0
; P9BE-NEXT: mtvsrwz 3, 6
; P9BE-NEXT: xxperm 2, 3, 0
; P9BE-NEXT: xxperm 1, 3, 0
; P9BE-NEXT: vsplth 3, 3, 3
; P9BE-NEXT: lxv 0, 0(3)
; P9BE-NEXT: li 3, 0
; P9BE-NEXT: xxmrghw 3, 3, 1
; P9BE-NEXT: xxperm 2, 3, 0
; P9BE-NEXT: xxspltw 3, 2, 1
; P9BE-NEXT: vadduwm 2, 2, 3
; P9BE-NEXT: vextuwlx 3, 3, 2
; P9BE-NEXT: cmpw 3, 5
; P9BE-NEXT: bgelr+ 0
; P9BE-NEXT: # %bb.1: # %if.then
;
; P9BE-AIX-LABEL: test16:
; P9BE-AIX: # %bb.0: # %entry
; P9BE-AIX-NEXT: sldi 4, 4, 1
; P9BE-AIX-NEXT: li 7, 16
; P9BE-AIX-NEXT: add 6, 3, 4
; P9BE-AIX-NEXT: lxsihzx 1, 3, 4
; P9BE-AIX-NEXT: ld 3, L..C3(2) # %const.1
; P9BE-AIX-NEXT: lxsihzx 2, 6, 7
; P9BE-AIX-NEXT: ld 6, L..C4(2) # %const.0
; P9BE-AIX-NEXT: lxv 0, 0(6)
; P9BE-AIX-NEXT: li 6, 0
; P9BE-AIX-NEXT: mtvsrwz 3, 6
; P9BE-AIX-NEXT: xxperm 2, 3, 0
; P9BE-AIX-NEXT: xxperm 1, 3, 0
; P9BE-AIX-NEXT: vsplth 3, 3, 3
; P9BE-AIX-NEXT: lxv 0, 0(3)
; P9BE-AIX-NEXT: li 3, 0
; P9BE-AIX-NEXT: xxmrghw 3, 3, 1
; P9BE-AIX-NEXT: xxperm 2, 3, 0
; P9BE-AIX-NEXT: xxspltw 3, 2, 1
; P9BE-AIX-NEXT: vadduwm 2, 2, 3
; P9BE-AIX-NEXT: vextuwlx 3, 3, 2
; P9BE-AIX-NEXT: cmpw 3, 5
; P9BE-AIX-NEXT: bgelr+ 0
; P9BE-AIX-NEXT: # %bb.1: # %if.then
;
; P9BE-AIX32-LABEL: test16:
; P9BE-AIX32: # %bb.0: # %entry
; P9BE-AIX32-NEXT: slwi 4, 4, 1
; P9BE-AIX32-NEXT: li 6, 0
; P9BE-AIX32-NEXT: lhzux 4, 3, 4
; P9BE-AIX32-NEXT: lhz 3, 16(3)
; P9BE-AIX32-NEXT: sth 6, -64(1)
; P9BE-AIX32-NEXT: lxv 2, -64(1)
; P9BE-AIX32-NEXT: sth 4, -48(1)
; P9BE-AIX32-NEXT: lxv 4, -48(1)
; P9BE-AIX32-NEXT: sth 3, -32(1)
; P9BE-AIX32-NEXT: lwz 3, L..C3(2) # %const.0
; P9BE-AIX32-NEXT: lxv 3, -32(1)
; P9BE-AIX32-NEXT: vmrghh 4, 2, 4
; P9BE-AIX32-NEXT: lxv 0, 0(3)
; P9BE-AIX32-NEXT: vmrghh 3, 2, 3
; P9BE-AIX32-NEXT: vsplth 2, 2, 0
; P9BE-AIX32-NEXT: xxmrghw 2, 2, 4
; P9BE-AIX32-NEXT: xxperm 3, 2, 0
; P9BE-AIX32-NEXT: xxspltw 2, 3, 1
; P9BE-AIX32-NEXT: vadduwm 2, 3, 2
; P9BE-AIX32-NEXT: stxv 2, -16(1)
; P9BE-AIX32-NEXT: lwz 3, -16(1)
; P9BE-AIX32-NEXT: cmpw 3, 5
; P9BE-AIX32-NEXT: bgelr+ 0
; P9BE-AIX32-NEXT: # %bb.1: # %if.then
entry:
%idxprom = sext i32 %delta to i64
%add14 = add nsw i32 %delta, 8
%idxprom15 = sext i32 %add14 to i64
br label %for.body
for.body: ; preds = %entry
%arrayidx8 = getelementptr inbounds i16, ptr %sums, i64 %idxprom
%0 = load i16, ptr %arrayidx8, align 2
%arrayidx16 = getelementptr inbounds i16, ptr %sums, i64 %idxprom15
%1 = load i16, ptr %arrayidx16, align 2
%2 = insertelement <4 x i16> undef, i16 %0, i32 2
%3 = insertelement <4 x i16> %2, i16 %1, i32 3
%4 = zext <4 x i16> %3 to <4 x i32>
%5 = sub nsw <4 x i32> zeroinitializer, %4
%6 = sub nsw <4 x i32> zeroinitializer, %5
%7 = select <4 x i1> undef, <4 x i32> %6, <4 x i32> %5
%bin.rdx = add <4 x i32> %7, zeroinitializer
%rdx.shuf54 = shufflevector <4 x i32> %bin.rdx, <4 x i32> undef, <4 x i32> <i32 1, i32 undef, i32 undef, i32 undef>
%bin.rdx55 = add <4 x i32> %bin.rdx, %rdx.shuf54
%8 = extractelement <4 x i32> %bin.rdx55, i32 0
%op.extra = add nuw i32 %8, 0
%cmp25 = icmp slt i32 %op.extra, %thresh
br i1 %cmp25, label %if.then, label %if.end
if.then: ; preds = %for.body
unreachable
if.end: ; preds = %for.body
ret void
}
define void @test8(ptr nocapture readonly %sums, i32 signext %delta, i32 signext %thresh) {
; P9LE-LABEL: test8:
; P9LE: # %bb.0: # %entry
; P9LE-NEXT: add 6, 3, 4
; P9LE-NEXT: lxsibzx 2, 3, 4
; P9LE-NEXT: li 3, 0
; P9LE-NEXT: mtvsrd 3, 3
; P9LE-NEXT: li 3, 8
; P9LE-NEXT: lxsibzx 5, 6, 3
; P9LE-NEXT: vspltb 4, 3, 7
; P9LE-NEXT: addis 3, 2, .LCPI3_0@toc@ha
; P9LE-NEXT: vmrghb 2, 3, 2
; P9LE-NEXT: addi 3, 3, .LCPI3_0@toc@l
; P9LE-NEXT: vmrglh 2, 2, 4
; P9LE-NEXT: lxv 0, 0(3)
; P9LE-NEXT: li 3, 0
; P9LE-NEXT: vmrghb 3, 3, 5
; P9LE-NEXT: xxmrglw 2, 2, 4
; P9LE-NEXT: vmrglh 3, 3, 4
; P9LE-NEXT: xxmrglw 3, 4, 3
; P9LE-NEXT: xxperm 2, 3, 0
; P9LE-NEXT: xxspltw 3, 2, 2
; P9LE-NEXT: vadduwm 2, 2, 3
; P9LE-NEXT: vextuwrx 3, 3, 2
; P9LE-NEXT: cmpw 3, 5
; P9LE-NEXT: bgelr+ 0
; P9LE-NEXT: # %bb.1: # %if.then
;
; P9BE-LABEL: test8:
; P9BE: # %bb.0: # %entry
; P9BE-NEXT: add 6, 3, 4
; P9BE-NEXT: li 7, 8
; P9BE-NEXT: lxsibzx 3, 3, 4
; P9BE-NEXT: addis 3, 2, .LCPI3_1@toc@ha
; P9BE-NEXT: lxsibzx 0, 6, 7
; P9BE-NEXT: addis 6, 2, .LCPI3_0@toc@ha
; P9BE-NEXT: addi 3, 3, .LCPI3_1@toc@l
; P9BE-NEXT: addi 6, 6, .LCPI3_0@toc@l
; P9BE-NEXT: lxv 1, 0(6)
; P9BE-NEXT: li 6, 0
; P9BE-NEXT: mtvsrwz 2, 6
; P9BE-NEXT: xxperm 0, 2, 1
; P9BE-NEXT: xxperm 3, 2, 1
; P9BE-NEXT: vspltb 2, 2, 7
; P9BE-NEXT: vmrghh 3, 3, 2
; P9BE-NEXT: xxspltw 1, 2, 0
; P9BE-NEXT: xxmrghw 3, 3, 0
; P9BE-NEXT: lxv 0, 0(3)
; P9BE-NEXT: li 3, 0
; P9BE-NEXT: xxperm 3, 1, 0
; P9BE-NEXT: xxspltw 2, 3, 1
; P9BE-NEXT: vadduwm 2, 3, 2
; P9BE-NEXT: vextuwlx 3, 3, 2
; P9BE-NEXT: cmpw 3, 5
; P9BE-NEXT: bgelr+ 0
; P9BE-NEXT: # %bb.1: # %if.then
;
; P9BE-AIX-LABEL: test8:
; P9BE-AIX: # %bb.0: # %entry
; P9BE-AIX-NEXT: add 6, 3, 4
; P9BE-AIX-NEXT: li 7, 8
; P9BE-AIX-NEXT: lxsibzx 3, 3, 4
; P9BE-AIX-NEXT: ld 3, L..C5(2) # %const.1
; P9BE-AIX-NEXT: lxsibzx 0, 6, 7
; P9BE-AIX-NEXT: ld 6, L..C6(2) # %const.0
; P9BE-AIX-NEXT: lxv 1, 0(6)
; P9BE-AIX-NEXT: li 6, 0
; P9BE-AIX-NEXT: mtvsrwz 2, 6
; P9BE-AIX-NEXT: xxperm 0, 2, 1
; P9BE-AIX-NEXT: xxperm 3, 2, 1
; P9BE-AIX-NEXT: vspltb 2, 2, 7
; P9BE-AIX-NEXT: vmrghh 3, 3, 2
; P9BE-AIX-NEXT: xxspltw 1, 2, 0
; P9BE-AIX-NEXT: xxmrghw 3, 3, 0
; P9BE-AIX-NEXT: lxv 0, 0(3)
; P9BE-AIX-NEXT: li 3, 0
; P9BE-AIX-NEXT: xxperm 3, 1, 0
; P9BE-AIX-NEXT: xxspltw 2, 3, 1
; P9BE-AIX-NEXT: vadduwm 2, 3, 2
; P9BE-AIX-NEXT: vextuwlx 3, 3, 2
; P9BE-AIX-NEXT: cmpw 3, 5
; P9BE-AIX-NEXT: bgelr+ 0
; P9BE-AIX-NEXT: # %bb.1: # %if.then
;
; P9BE-AIX32-LABEL: test8:
; P9BE-AIX32: # %bb.0: # %entry
; P9BE-AIX32-NEXT: add 6, 3, 4
; P9BE-AIX32-NEXT: li 7, 8
; P9BE-AIX32-NEXT: lxsibzx 3, 3, 4
; P9BE-AIX32-NEXT: lwz 3, L..C4(2) # %const.1
; P9BE-AIX32-NEXT: lxsibzx 0, 6, 7
; P9BE-AIX32-NEXT: lwz 6, L..C5(2) # %const.0
; P9BE-AIX32-NEXT: lxv 1, 0(6)
; P9BE-AIX32-NEXT: li 6, 0
; P9BE-AIX32-NEXT: mtvsrwz 2, 6
; P9BE-AIX32-NEXT: xxperm 0, 2, 1
; P9BE-AIX32-NEXT: xxperm 3, 2, 1
; P9BE-AIX32-NEXT: vspltb 2, 2, 7
; P9BE-AIX32-NEXT: vmrghh 3, 3, 2
; P9BE-AIX32-NEXT: xxspltw 1, 2, 0
; P9BE-AIX32-NEXT: xxmrghw 3, 3, 0
; P9BE-AIX32-NEXT: lxv 0, 0(3)
; P9BE-AIX32-NEXT: xxperm 3, 1, 0
; P9BE-AIX32-NEXT: xxspltw 2, 3, 1
; P9BE-AIX32-NEXT: vadduwm 2, 3, 2
; P9BE-AIX32-NEXT: stxv 2, -16(1)
; P9BE-AIX32-NEXT: lwz 3, -16(1)
; P9BE-AIX32-NEXT: cmpw 3, 5
; P9BE-AIX32-NEXT: bgelr+ 0
; P9BE-AIX32-NEXT: # %bb.1: # %if.then
entry:
%idxprom = sext i32 %delta to i64
%add14 = add nsw i32 %delta, 8
%idxprom15 = sext i32 %add14 to i64
br label %for.body
for.body: ; preds = %entry
%arrayidx8 = getelementptr inbounds i8, ptr %sums, i64 %idxprom
%0 = load i8, ptr %arrayidx8, align 2
%arrayidx16 = getelementptr inbounds i8, ptr %sums, i64 %idxprom15
%1 = load i8, ptr %arrayidx16, align 2
%2 = insertelement <4 x i8> undef, i8 %0, i32 2
%3 = insertelement <4 x i8> %2, i8 %1, i32 3
%4 = zext <4 x i8> %3 to <4 x i32>
%5 = sub nsw <4 x i32> zeroinitializer, %4
%6 = sub nsw <4 x i32> zeroinitializer, %5
%7 = select <4 x i1> undef, <4 x i32> %6, <4 x i32> %5
%bin.rdx = add <4 x i32> %7, zeroinitializer
%rdx.shuf54 = shufflevector <4 x i32> %bin.rdx, <4 x i32> undef, <4 x i32> <i32 1, i32 undef, i32 undef, i32 undef>
%bin.rdx55 = add <4 x i32> %bin.rdx, %rdx.shuf54
%8 = extractelement <4 x i32> %bin.rdx55, i32 0
%op.extra = add nuw i32 %8, 0
%cmp25 = icmp slt i32 %op.extra, %thresh
br i1 %cmp25, label %if.then, label %if.end
if.then: ; preds = %for.body
unreachable
if.end: ; preds = %for.body
ret void
}
Codebase Browser
llvm