; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -ppc-asm-full-reg-names -verify-machineinstrs -ppc-formprep-chain-commoning \
; RUN: -mtriple=powerpc64le-unknown-linux-gnu -mcpu=pwr9 < %s | FileCheck %s
; addresses:
; 1: base1 + offset
; 2: + offset
; 3: + offset
; 4: + offset
;
; chains:
; 1: base: base1 + offset, offsets: (0, offset)
; 2: base: base1 + 3*offset, offsets: (0, offset)
;
; long long two_chain_same_offset_succ(char *p, long long offset, long long base1, long long n) {
; long long o1 = base1 + offset;
; long long o2 = base1 + 2 * offset;
; long long o3 = base1 + 3 * offset;
; long long o4 = base1 + 4 * offset;
; char *p1 = p + o1;
; char *p2 = p + o2;
; char *p3 = p + o3;
; char *p4 = p + o4;
; long long sum = 0;
; for (long long i = 0; i < n; ++i) {
; unsigned long x1 = *(unsigned long *)(p1 + i);
; unsigned long x2 = *(unsigned long *)(p2 + i);
; unsigned long x3 = *(unsigned long *)(p3 + i);
; unsigned long x4 = *(unsigned long *)(p4 + i);
; sum += x1 * x2 * x3 * x4;
; }
; return sum;
; }
;
define i64 @two_chain_same_offset_succ(i8* %p, i64 %offset, i64 %base1, i64 %n) {
; CHECK-LABEL: two_chain_same_offset_succ:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: cmpdi r6, 1
; CHECK-NEXT: blt cr0, .LBB0_4
; CHECK-NEXT: # %bb.1: # %for.body.preheader
; CHECK-NEXT: sldi r7, r4, 1
; CHECK-NEXT: mtctr r6
; CHECK-NEXT: add r8, r4, r7
; CHECK-NEXT: add r7, r5, r4
; CHECK-NEXT: add r5, r5, r8
; CHECK-NEXT: add r7, r3, r7
; CHECK-NEXT: add r5, r3, r5
; CHECK-NEXT: li r3, 0
; CHECK-NEXT: .p2align 4
; CHECK-NEXT: .LBB0_2: # %for.body
; CHECK-NEXT: #
; CHECK-NEXT: ld r6, 0(r7)
; CHECK-NEXT: ldx r8, r7, r4
; CHECK-NEXT: ld r9, 0(r5)
; CHECK-NEXT: ldx r10, r5, r4
; CHECK-NEXT: addi r7, r7, 1
; CHECK-NEXT: addi r5, r5, 1
; CHECK-NEXT: mulld r6, r8, r6
; CHECK-NEXT: mulld r6, r6, r9
; CHECK-NEXT: maddld r3, r6, r10, r3
; CHECK-NEXT: bdnz .LBB0_2
; CHECK-NEXT: # %bb.3: # %for.cond.cleanup
; CHECK-NEXT: blr
; CHECK-NEXT: .LBB0_4:
; CHECK-NEXT: li r3, 0
; CHECK-NEXT: blr
entry:
%mul = shl nsw i64 %offset, 1
%mul2 = mul nsw i64 %offset, 3
%mul4 = shl nsw i64 %offset, 2
%cmp46 = icmp sgt i64 %n, 0
br i1 %cmp46, label %for.body, label %for.cond.cleanup
for.cond.cleanup: ; preds = %for.body, %entry
%sum.0.lcssa = phi i64 [ 0, %entry ], [ %add16, %for.body ]
ret i64 %sum.0.lcssa
for.body: ; preds = %entry, %for.body
%sum.048 = phi i64 [ %add16, %for.body ], [ 0, %entry ]
%i.047 = phi i64 [ %inc, %for.body ], [ 0, %entry ]
%add = add i64 %i.047, %base1
%add.ptr9.idx = add i64 %add, %offset
%add.ptr9 = getelementptr inbounds i8, i8* %p, i64 %add.ptr9.idx
%0 = bitcast i8* %add.ptr9 to i64*
%1 = load i64, i64* %0, align 8
%add.ptr10.idx = add i64 %add, %mul
%add.ptr10 = getelementptr inbounds i8, i8* %p, i64 %add.ptr10.idx
%2 = bitcast i8* %add.ptr10 to i64*
%3 = load i64, i64* %2, align 8
%add.ptr11.idx = add i64 %add, %mul2
%add.ptr11 = getelementptr inbounds i8, i8* %p, i64 %add.ptr11.idx
%4 = bitcast i8* %add.ptr11 to i64*
%5 = load i64, i64* %4, align 8
%add.ptr12.idx = add i64 %add, %mul4
%add.ptr12 = getelementptr inbounds i8, i8* %p, i64 %add.ptr12.idx
%6 = bitcast i8* %add.ptr12 to i64*
%7 = load i64, i64* %6, align 8
%mul13 = mul i64 %3, %1
%mul14 = mul i64 %mul13, %5
%mul15 = mul i64 %mul14, %7
%add16 = add i64 %mul15, %sum.048
%inc = add nuw nsw i64 %i.047, 1
%exitcond.not = icmp eq i64 %inc, %n
br i1 %exitcond.not, label %for.cond.cleanup, label %for.body
}
; addresses:
; 1: base1 + offset
; 2: + offset
; 3: + offset
; 4: + offset
; 5: + offset
;
; It can not be commoned to chains because we need a chain for a single address.
; It is not profitable to common chains if not all addresses are in chains.
;
; long long not_perfect_chain_all_same_offset_fail(char *p, long long offset, long long base1, long long n) {
; long long o1 = base1 + offset;
; long long o2 = base1 + 2 * offset;
; long long o3 = base1 + 3 * offset;
; long long o4 = base1 + 4 * offset;
; long long o5 = base1 + 5 * offset;
; char *p1 = p + o1;
; char *p2 = p + o2;
; char *p3 = p + o3;
; char *p4 = p + o4;
; char *p5 = p + o5;
; long long sum = 0;
; for (long long i = 0; i < n; ++i) {
; unsigned long x1 = *(unsigned long *)(p1 + i);
; unsigned long x2 = *(unsigned long *)(p2 + i);
; unsigned long x3 = *(unsigned long *)(p3 + i);
; unsigned long x4 = *(unsigned long *)(p4 + i);
; unsigned long x5 = *(unsigned long *)(p5 + i);
; sum += x1 * x2 * x3 * x4 * x5;
; }
; return sum;
; }
;
define i64 @not_perfect_chain_all_same_offset_fail(i8* %p, i64 %offset, i64 %base1, i64 %n) {
; CHECK-LABEL: not_perfect_chain_all_same_offset_fail:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: cmpdi r6, 1
; CHECK-NEXT: std r30, -16(r1) # 8-byte Folded Spill
; CHECK-NEXT: blt cr0, .LBB1_4
; CHECK-NEXT: # %bb.1: # %for.body.preheader
; CHECK-NEXT: sldi r7, r4, 1
; CHECK-NEXT: sldi r9, r4, 2
; CHECK-NEXT: add r5, r3, r5
; CHECK-NEXT: li r3, 0
; CHECK-NEXT: add r8, r4, r7
; CHECK-NEXT: mtctr r6
; CHECK-NEXT: add r10, r4, r9
; CHECK-NEXT: .p2align 4
; CHECK-NEXT: .LBB1_2: # %for.body
; CHECK-NEXT: #
; CHECK-NEXT: ldx r6, r5, r4
; CHECK-NEXT: ldx r11, r5, r7
; CHECK-NEXT: ldx r12, r5, r8
; CHECK-NEXT: ldx r0, r5, r9
; CHECK-NEXT: mulld r6, r11, r6
; CHECK-NEXT: ldx r30, r5, r10
; CHECK-NEXT: addi r5, r5, 1
; CHECK-NEXT: mulld r6, r6, r12
; CHECK-NEXT: mulld r6, r6, r0
; CHECK-NEXT: maddld r3, r6, r30, r3
; CHECK-NEXT: bdnz .LBB1_2
; CHECK-NEXT: # %bb.3: # %for.cond.cleanup
; CHECK-NEXT: ld r30, -16(r1) # 8-byte Folded Reload
; CHECK-NEXT: blr
; CHECK-NEXT: .LBB1_4:
; CHECK-NEXT: li r3, 0
; CHECK-NEXT: ld r30, -16(r1) # 8-byte Folded Reload
; CHECK-NEXT: blr
entry:
%mul = shl nsw i64 %offset, 1
%mul2 = mul nsw i64 %offset, 3
%mul4 = shl nsw i64 %offset, 2
%mul6 = mul nsw i64 %offset, 5
%cmp58 = icmp sgt i64 %n, 0
br i1 %cmp58, label %for.body, label %for.cond.cleanup
for.cond.cleanup: ; preds = %for.body, %entry
%sum.0.lcssa = phi i64 [ 0, %entry ], [ %add21, %for.body ]
ret i64 %sum.0.lcssa
for.body: ; preds = %entry, %for.body
%sum.060 = phi i64 [ %add21, %for.body ], [ 0, %entry ]
%i.059 = phi i64 [ %inc, %for.body ], [ 0, %entry ]
%add = add i64 %i.059, %base1
%add.ptr12.idx = add i64 %add, %offset
%add.ptr12 = getelementptr inbounds i8, i8* %p, i64 %add.ptr12.idx
%0 = bitcast i8* %add.ptr12 to i64*
%1 = load i64, i64* %0, align 8
%add.ptr13.idx = add i64 %add, %mul
%add.ptr13 = getelementptr inbounds i8, i8* %p, i64 %add.ptr13.idx
%2 = bitcast i8* %add.ptr13 to i64*
%3 = load i64, i64* %2, align 8
%add.ptr14.idx = add i64 %add, %mul2
%add.ptr14 = getelementptr inbounds i8, i8* %p, i64 %add.ptr14.idx
%4 = bitcast i8* %add.ptr14 to i64*
%5 = load i64, i64* %4, align 8
%add.ptr15.idx = add i64 %add, %mul4
%add.ptr15 = getelementptr inbounds i8, i8* %p, i64 %add.ptr15.idx
%6 = bitcast i8* %add.ptr15 to i64*
%7 = load i64, i64* %6, align 8
%add.ptr16.idx = add i64 %add, %mul6
%add.ptr16 = getelementptr inbounds i8, i8* %p, i64 %add.ptr16.idx
%8 = bitcast i8* %add.ptr16 to i64*
%9 = load i64, i64* %8, align 8
%mul17 = mul i64 %3, %1
%mul18 = mul i64 %mul17, %5
%mul19 = mul i64 %mul18, %7
%mul20 = mul i64 %mul19, %9
%add21 = add i64 %mul20, %sum.060
%inc = add nuw nsw i64 %i.059, 1
%exitcond.not = icmp eq i64 %inc, %n
br i1 %exitcond.not, label %for.cond.cleanup, label %for.body
}
; addresses:
; 1: base1
; 2: + 2*offset
; 3: + offset
;
; We need at least 4 addresses to common 2 chains to reuse at least 1 offset.
;
; long long no_enough_elements_fail(char *p, long long offset, long long base1, long long n) {
; long long o1 = base1;
; long long o2 = base1 + 2 * offset;
; long long o3 = base1 + 3 * offset;
; char *p1 = p + o1;
; char *p2 = p + o2;
; char *p3 = p + o3;
; long long sum = 0;
; for (long long i = 0; i < n; ++i) {
; unsigned long x1 = *(unsigned long *)(p1 + i);
; unsigned long x2 = *(unsigned long *)(p2 + i);
; unsigned long x3 = *(unsigned long *)(p3 + i);
; sum += x1 * x2 * x3;
; }
; return sum;
; }
;
define i64 @no_enough_elements_fail(i8* %p, i64 %offset, i64 %base1, i64 %n) {
; CHECK-LABEL: no_enough_elements_fail:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: cmpdi r6, 1
; CHECK-NEXT: blt cr0, .LBB2_4
; CHECK-NEXT: # %bb.1: # %for.body.preheader
; CHECK-NEXT: sldi r7, r4, 1
; CHECK-NEXT: mtctr r6
; CHECK-NEXT: add r5, r3, r5
; CHECK-NEXT: li r3, 0
; CHECK-NEXT: add r4, r4, r7
; CHECK-NEXT: .p2align 5
; CHECK-NEXT: .LBB2_2: # %for.body
; CHECK-NEXT: #
; CHECK-NEXT: ld r6, 0(r5)
; CHECK-NEXT: ldx r8, r5, r7
; CHECK-NEXT: ldx r9, r5, r4
; CHECK-NEXT: addi r5, r5, 1
; CHECK-NEXT: mulld r6, r8, r6
; CHECK-NEXT: maddld r3, r6, r9, r3
; CHECK-NEXT: bdnz .LBB2_2
; CHECK-NEXT: # %bb.3: # %for.cond.cleanup
; CHECK-NEXT: blr
; CHECK-NEXT: .LBB2_4:
; CHECK-NEXT: li r3, 0
; CHECK-NEXT: blr
entry:
%mul = shl nsw i64 %offset, 1
%mul1 = mul nsw i64 %offset, 3
%cmp32 = icmp sgt i64 %n, 0
br i1 %cmp32, label %for.body, label %for.cond.cleanup
for.cond.cleanup: ; preds = %for.body, %entry
%sum.0.lcssa = phi i64 [ 0, %entry ], [ %add10, %for.body ]
ret i64 %sum.0.lcssa
for.body: ; preds = %entry, %for.body
%sum.034 = phi i64 [ %add10, %for.body ], [ 0, %entry ]
%i.033 = phi i64 [ %inc, %for.body ], [ 0, %entry ]
%add.ptr5.idx = add i64 %i.033, %base1
%add.ptr5 = getelementptr inbounds i8, i8* %p, i64 %add.ptr5.idx
%0 = bitcast i8* %add.ptr5 to i64*
%1 = load i64, i64* %0, align 8
%add.ptr6.idx = add i64 %add.ptr5.idx, %mul
%add.ptr6 = getelementptr inbounds i8, i8* %p, i64 %add.ptr6.idx
%2 = bitcast i8* %add.ptr6 to i64*
%3 = load i64, i64* %2, align 8
%add.ptr7.idx = add i64 %add.ptr5.idx, %mul1
%add.ptr7 = getelementptr inbounds i8, i8* %p, i64 %add.ptr7.idx
%4 = bitcast i8* %add.ptr7 to i64*
%5 = load i64, i64* %4, align 8
%mul8 = mul i64 %3, %1
%mul9 = mul i64 %mul8, %5
%add10 = add i64 %mul9, %sum.034
%inc = add nuw nsw i64 %i.033, 1
%exitcond.not = icmp eq i64 %inc, %n
br i1 %exitcond.not, label %for.cond.cleanup, label %for.body
}
; addresses:
; 1: base1
; 2: + 2*offset
; 3: + 2*offset
; 4: + 3*offset
;
; The diff between address 2 and address 1 is 2*offset, and this offset is not reused among other chains,
; so we can not common any chains.
;
; long long no_reuseable_offset_fail(char *p, long long offset, long long base1, long long n) {
; long long o1 = base1;
; long long o2 = base1 + 2 * offset;
; long long o3 = base1 + 4 * offset;
; long long o4 = base1 + 7 * offset;
; char *p1 = p + o1;
; char *p2 = p + o2;
; char *p3 = p + o3;
; char *p4 = p + o4;
; long long sum = 0;
; for (long long i = 0; i < n; ++i) {
; unsigned long x1 = *(unsigned long *)(p1 + i);
; unsigned long x2 = *(unsigned long *)(p2 + i);
; unsigned long x3 = *(unsigned long *)(p3 + i);
; unsigned long x4 = *(unsigned long *)(p4 + i);
; sum += x1 * x2 * x3 * x4;
; }
; return sum;
; }
;
define i64 @no_reuseable_offset_fail(i8* %p, i64 %offset, i64 %base1, i64 %n) {
; CHECK-LABEL: no_reuseable_offset_fail:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: cmpdi r6, 1
; CHECK-NEXT: blt cr0, .LBB3_4
; CHECK-NEXT: # %bb.1: # %for.body.preheader
; CHECK-NEXT: sldi r9, r4, 3
; CHECK-NEXT: mtctr r6
; CHECK-NEXT: add r5, r3, r5
; CHECK-NEXT: li r3, 0
; CHECK-NEXT: sldi r7, r4, 1
; CHECK-NEXT: sldi r8, r4, 2
; CHECK-NEXT: sub r4, r9, r4
; CHECK-NEXT: .p2align 4
; CHECK-NEXT: .LBB3_2: # %for.body
; CHECK-NEXT: #
; CHECK-NEXT: ld r6, 0(r5)
; CHECK-NEXT: ldx r9, r5, r7
; CHECK-NEXT: ldx r10, r5, r8
; CHECK-NEXT: ldx r11, r5, r4
; CHECK-NEXT: addi r5, r5, 1
; CHECK-NEXT: mulld r6, r9, r6
; CHECK-NEXT: mulld r6, r6, r10
; CHECK-NEXT: maddld r3, r6, r11, r3
; CHECK-NEXT: bdnz .LBB3_2
; CHECK-NEXT: # %bb.3: # %for.cond.cleanup
; CHECK-NEXT: blr
; CHECK-NEXT: .LBB3_4:
; CHECK-NEXT: li r3, 0
; CHECK-NEXT: blr
entry:
%mul = shl nsw i64 %offset, 1
%mul1 = shl nsw i64 %offset, 2
%mul3 = mul nsw i64 %offset, 7
%cmp44 = icmp sgt i64 %n, 0
br i1 %cmp44, label %for.body, label %for.cond.cleanup
for.cond.cleanup: ; preds = %for.body, %entry
%sum.0.lcssa = phi i64 [ 0, %entry ], [ %add15, %for.body ]
ret i64 %sum.0.lcssa
for.body: ; preds = %entry, %for.body
%sum.046 = phi i64 [ %add15, %for.body ], [ 0, %entry ]
%i.045 = phi i64 [ %inc, %for.body ], [ 0, %entry ]
%add.ptr8.idx = add i64 %i.045, %base1
%add.ptr8 = getelementptr inbounds i8, i8* %p, i64 %add.ptr8.idx
%0 = bitcast i8* %add.ptr8 to i64*
%1 = load i64, i64* %0, align 8
%add.ptr9.idx = add i64 %add.ptr8.idx, %mul
%add.ptr9 = getelementptr inbounds i8, i8* %p, i64 %add.ptr9.idx
%2 = bitcast i8* %add.ptr9 to i64*
%3 = load i64, i64* %2, align 8
%add.ptr10.idx = add i64 %add.ptr8.idx, %mul1
%add.ptr10 = getelementptr inbounds i8, i8* %p, i64 %add.ptr10.idx
%4 = bitcast i8* %add.ptr10 to i64*
%5 = load i64, i64* %4, align 8
%add.ptr11.idx = add i64 %add.ptr8.idx, %mul3
%add.ptr11 = getelementptr inbounds i8, i8* %p, i64 %add.ptr11.idx
%6 = bitcast i8* %add.ptr11 to i64*
%7 = load i64, i64* %6, align 8
%mul12 = mul i64 %3, %1
%mul13 = mul i64 %mul12, %5
%mul14 = mul i64 %mul13, %7
%add15 = add i64 %mul14, %sum.046
%inc = add nuw nsw i64 %i.045, 1
%exitcond.not = icmp eq i64 %inc, %n
br i1 %exitcond.not, label %for.cond.cleanup, label %for.body
}
; addresses:
; 1: base1 + offset
; 2: + offset
; 3: + 3*offset
; 4: + 2*offset
; 5: + 1*offset
; 6: + 2*offset
;
; The diff between address 2 and address 1 is 1*offset, and this offset is reused between address 4 and address 5.
; but the diff between address 3 and address 2 (3*offset) is not the same with the diff between address 6
; and address 5(2*offset), so we can not common chains for these addresses.
;
; long long not_same_offset_fail(char *p, long long offset, long long base1, long long n) {
; long long o1 = base1 + offset;
; long long o2 = base1 + 2 * offset;
; long long o3 = base1 + 5 * offset;
; long long o4 = base1 + 7 * offset;
; long long o5 = base1 + 8 * offset;
; long long o6 = base1 + 10 * offset;
; char *p1 = p + o1;
; char *p2 = p + o2;
; char *p3 = p + o3;
; char *p4 = p + o4;
; char *p5 = p + o5;
; char *p6 = p + o6;
; long long sum = 0;
; for (long long i = 0; i < n; ++i) {
; unsigned long x1 = *(unsigned long *)(p1 + i);
; unsigned long x2 = *(unsigned long *)(p2 + i);
; unsigned long x3 = *(unsigned long *)(p3 + i);
; unsigned long x4 = *(unsigned long *)(p4 + i);
; unsigned long x5 = *(unsigned long *)(p5 + i);
; unsigned long x6 = *(unsigned long *)(p6 + i);
; sum += x1 * x2 * x3 * x4 * x5 * x6;
; }
; return sum;
; }
;
define i64 @not_same_offset_fail(i8* %p, i64 %offset, i64 %base1, i64 %n) {
; CHECK-LABEL: not_same_offset_fail:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: cmpdi r6, 1
; CHECK-NEXT: std r28, -32(r1) # 8-byte Folded Spill
; CHECK-NEXT: std r29, -24(r1) # 8-byte Folded Spill
; CHECK-NEXT: std r30, -16(r1) # 8-byte Folded Spill
; CHECK-NEXT: blt cr0, .LBB4_3
; CHECK-NEXT: # %bb.1: # %for.body.preheader
; CHECK-NEXT: mulli r11, r4, 10
; CHECK-NEXT: sldi r8, r4, 2
; CHECK-NEXT: add r5, r3, r5
; CHECK-NEXT: li r3, 0
; CHECK-NEXT: add r8, r4, r8
; CHECK-NEXT: sldi r9, r4, 3
; CHECK-NEXT: mtctr r6
; CHECK-NEXT: sldi r7, r4, 1
; CHECK-NEXT: sub r10, r9, r4
; CHECK-NEXT: .p2align 4
; CHECK-NEXT: .LBB4_2: # %for.body
; CHECK-NEXT: #
; CHECK-NEXT: ldx r6, r5, r4
; CHECK-NEXT: ldx r12, r5, r7
; CHECK-NEXT: ldx r0, r5, r8
; CHECK-NEXT: ldx r30, r5, r10
; CHECK-NEXT: mulld r6, r12, r6
; CHECK-NEXT: ldx r29, r5, r9
; CHECK-NEXT: ldx r28, r5, r11
; CHECK-NEXT: addi r5, r5, 1
; CHECK-NEXT: mulld r6, r6, r0
; CHECK-NEXT: mulld r6, r6, r30
; CHECK-NEXT: mulld r6, r6, r29
; CHECK-NEXT: maddld r3, r6, r28, r3
; CHECK-NEXT: bdnz .LBB4_2
; CHECK-NEXT: b .LBB4_4
; CHECK-NEXT: .LBB4_3:
; CHECK-NEXT: li r3, 0
; CHECK-NEXT: .LBB4_4: # %for.cond.cleanup
; CHECK-NEXT: ld r30, -16(r1) # 8-byte Folded Reload
; CHECK-NEXT: ld r29, -24(r1) # 8-byte Folded Reload
; CHECK-NEXT: ld r28, -32(r1) # 8-byte Folded Reload
; CHECK-NEXT: blr
entry:
%mul = shl nsw i64 %offset, 1
%mul2 = mul nsw i64 %offset, 5
%mul4 = mul nsw i64 %offset, 7
%mul6 = shl nsw i64 %offset, 3
%mul8 = mul nsw i64 %offset, 10
%cmp70 = icmp sgt i64 %n, 0
br i1 %cmp70, label %for.body, label %for.cond.cleanup
for.cond.cleanup: ; preds = %for.body, %entry
%sum.0.lcssa = phi i64 [ 0, %entry ], [ %add26, %for.body ]
ret i64 %sum.0.lcssa
for.body: ; preds = %entry, %for.body
%sum.072 = phi i64 [ %add26, %for.body ], [ 0, %entry ]
%i.071 = phi i64 [ %inc, %for.body ], [ 0, %entry ]
%add = add i64 %i.071, %base1
%add.ptr15.idx = add i64 %add, %offset
%add.ptr15 = getelementptr inbounds i8, i8* %p, i64 %add.ptr15.idx
%0 = bitcast i8* %add.ptr15 to i64*
%1 = load i64, i64* %0, align 8
%add.ptr16.idx = add i64 %add, %mul
%add.ptr16 = getelementptr inbounds i8, i8* %p, i64 %add.ptr16.idx
%2 = bitcast i8* %add.ptr16 to i64*
%3 = load i64, i64* %2, align 8
%add.ptr17.idx = add i64 %add, %mul2
%add.ptr17 = getelementptr inbounds i8, i8* %p, i64 %add.ptr17.idx
%4 = bitcast i8* %add.ptr17 to i64*
%5 = load i64, i64* %4, align 8
%add.ptr18.idx = add i64 %add, %mul4
%add.ptr18 = getelementptr inbounds i8, i8* %p, i64 %add.ptr18.idx
%6 = bitcast i8* %add.ptr18 to i64*
%7 = load i64, i64* %6, align 8
%add.ptr19.idx = add i64 %add, %mul6
%add.ptr19 = getelementptr inbounds i8, i8* %p, i64 %add.ptr19.idx
%8 = bitcast i8* %add.ptr19 to i64*
%9 = load i64, i64* %8, align 8
%add.ptr20.idx = add i64 %add, %mul8
%add.ptr20 = getelementptr inbounds i8, i8* %p, i64 %add.ptr20.idx
%10 = bitcast i8* %add.ptr20 to i64*
%11 = load i64, i64* %10, align 8
%mul21 = mul i64 %3, %1
%mul22 = mul i64 %mul21, %5
%mul23 = mul i64 %mul22, %7
%mul24 = mul i64 %mul23, %9
%mul25 = mul i64 %mul24, %11
%add26 = add i64 %mul25, %sum.072
%inc = add nuw nsw i64 %i.071, 1
%exitcond.not = icmp eq i64 %inc, %n
br i1 %exitcond.not, label %for.cond.cleanup, label %for.body
}
; addresses:
; 1: base1 + offset
; 2: + offset
; 3: + 3*offset
; 4: + 2*offset
;
; chains:
; 1: base1 + offset, offsets: (0, 2*offset)
; 2: base1 + 4*offset, offsets: (0, 2*offset)
;
; long long two_chain_different_offsets_succ(char *p, long long offset, long long base1, long long n) {
; long long o1 = base1 + offset;
; long long o2 = base1 + 3 * offset;
; long long o3 = base1 + 4 * offset;
; long long o4 = base1 + 6 * offset;
; char *p1 = p + o1;
; char *p2 = p + o2;
; char *p3 = p + o3;
; char *p4 = p + o4;
; long long sum = 0;
; for (long long i = 0; i < n; ++i) {
; unsigned long x1 = *(unsigned long *)(p1 + i);
; unsigned long x2 = *(unsigned long *)(p2 + i);
; unsigned long x3 = *(unsigned long *)(p3 + i);
; unsigned long x4 = *(unsigned long *)(p4 + i);
; sum += x1 * x2 * x3 * x4;
; }
; return sum;
; }
;
define i64 @two_chain_different_offsets_succ(i8* %p, i64 %offset, i64 %base1, i64 %n) {
; CHECK-LABEL: two_chain_different_offsets_succ:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: cmpdi r6, 1
; CHECK-NEXT: blt cr0, .LBB5_4
; CHECK-NEXT: # %bb.1: # %for.body.preheader
; CHECK-NEXT: sldi r8, r4, 2
; CHECK-NEXT: add r7, r5, r4
; CHECK-NEXT: mtctr r6
; CHECK-NEXT: add r5, r5, r8
; CHECK-NEXT: add r7, r3, r7
; CHECK-NEXT: sldi r4, r4, 1
; CHECK-NEXT: add r5, r3, r5
; CHECK-NEXT: li r3, 0
; CHECK-NEXT: .p2align 4
; CHECK-NEXT: .LBB5_2: # %for.body
; CHECK-NEXT: #
; CHECK-NEXT: ld r6, 0(r7)
; CHECK-NEXT: ldx r8, r7, r4
; CHECK-NEXT: ld r9, 0(r5)
; CHECK-NEXT: ldx r10, r5, r4
; CHECK-NEXT: addi r7, r7, 1
; CHECK-NEXT: addi r5, r5, 1
; CHECK-NEXT: mulld r6, r8, r6
; CHECK-NEXT: mulld r6, r6, r9
; CHECK-NEXT: maddld r3, r6, r10, r3
; CHECK-NEXT: bdnz .LBB5_2
; CHECK-NEXT: # %bb.3: # %for.cond.cleanup
; CHECK-NEXT: blr
; CHECK-NEXT: .LBB5_4:
; CHECK-NEXT: li r3, 0
; CHECK-NEXT: blr
entry:
%mul = mul nsw i64 %offset, 3
%mul2 = shl nsw i64 %offset, 2
%mul4 = mul nsw i64 %offset, 6
%cmp46 = icmp sgt i64 %n, 0
br i1 %cmp46, label %for.body, label %for.cond.cleanup
for.cond.cleanup: ; preds = %for.body, %entry
%sum.0.lcssa = phi i64 [ 0, %entry ], [ %add16, %for.body ]
ret i64 %sum.0.lcssa
for.body: ; preds = %entry, %for.body
%sum.048 = phi i64 [ %add16, %for.body ], [ 0, %entry ]
%i.047 = phi i64 [ %inc, %for.body ], [ 0, %entry ]
%add = add i64 %i.047, %base1
%add.ptr9.idx = add i64 %add, %offset
%add.ptr9 = getelementptr inbounds i8, i8* %p, i64 %add.ptr9.idx
%0 = bitcast i8* %add.ptr9 to i64*
%1 = load i64, i64* %0, align 8
%add.ptr10.idx = add i64 %add, %mul
%add.ptr10 = getelementptr inbounds i8, i8* %p, i64 %add.ptr10.idx
%2 = bitcast i8* %add.ptr10 to i64*
%3 = load i64, i64* %2, align 8
%add.ptr11.idx = add i64 %add, %mul2
%add.ptr11 = getelementptr inbounds i8, i8* %p, i64 %add.ptr11.idx
%4 = bitcast i8* %add.ptr11 to i64*
%5 = load i64, i64* %4, align 8
%add.ptr12.idx = add i64 %add, %mul4
%add.ptr12 = getelementptr inbounds i8, i8* %p, i64 %add.ptr12.idx
%6 = bitcast i8* %add.ptr12 to i64*
%7 = load i64, i64* %6, align 8
%mul13 = mul i64 %3, %1
%mul14 = mul i64 %mul13, %5
%mul15 = mul i64 %mul14, %7
%add16 = add i64 %mul15, %sum.048
%inc = add nuw nsw i64 %i.047, 1
%exitcond.not = icmp eq i64 %inc, %n
br i1 %exitcond.not, label %for.cond.cleanup, label %for.body
}
; addresses:
; 1: base1 + offset
; 2: + 2*offset
; 3: + base2 - base1 - 2*offset
; 4: + 2*offset
;
; chains:
; 1: base1 + offset, offsets: (0, 2*offset)
; 2: base2 + offset, offsets: (0, 2*offset)
;
; long long two_chain_two_bases_succ(char *p, long long offset, long long base1, long long base2, long long n) {
; long long o1 = base1 + offset;
; long long o2 = base1 + 3 * offset;
; long long o3 = base2 + offset;
; long long o4 = base2 + 3 * offset;
; char *p1 = p + o1;
; char *p2 = p + o2;
; char *p3 = p + o3;
; char *p4 = p + o4;
; long long sum = 0;
; for (long long i = 0; i < n; ++i) {
; unsigned long x1 = *(unsigned long *)(p1 + i);
; unsigned long x2 = *(unsigned long *)(p2 + i);
; unsigned long x3 = *(unsigned long *)(p3 + i);
; unsigned long x4 = *(unsigned long *)(p4 + i);
; sum += x1 * x2 * x3 * x4;
; }
; return sum;
; }
;
define i64 @two_chain_two_bases_succ(i8* %p, i64 %offset, i64 %base1, i64 %base2, i64 %n) {
; CHECK-LABEL: two_chain_two_bases_succ:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: cmpdi r7, 1
; CHECK-NEXT: blt cr0, .LBB6_4
; CHECK-NEXT: # %bb.1: # %for.body.preheader
; CHECK-NEXT: add r6, r6, r4
; CHECK-NEXT: add r5, r5, r4
; CHECK-NEXT: mtctr r7
; CHECK-NEXT: sldi r4, r4, 1
; CHECK-NEXT: add r5, r3, r5
; CHECK-NEXT: add r6, r3, r6
; CHECK-NEXT: li r3, 0
; CHECK-NEXT: .p2align 4
; CHECK-NEXT: .LBB6_2: # %for.body
; CHECK-NEXT: #
; CHECK-NEXT: ld r7, 0(r5)
; CHECK-NEXT: ldx r8, r5, r4
; CHECK-NEXT: ld r9, 0(r6)
; CHECK-NEXT: ldx r10, r6, r4
; CHECK-NEXT: addi r5, r5, 1
; CHECK-NEXT: addi r6, r6, 1
; CHECK-NEXT: mulld r7, r8, r7
; CHECK-NEXT: mulld r7, r7, r9
; CHECK-NEXT: maddld r3, r7, r10, r3
; CHECK-NEXT: bdnz .LBB6_2
; CHECK-NEXT: # %bb.3: # %for.cond.cleanup
; CHECK-NEXT: blr
; CHECK-NEXT: .LBB6_4:
; CHECK-NEXT: li r3, 0
; CHECK-NEXT: blr
entry:
%mul = mul nsw i64 %offset, 3
%cmp44 = icmp sgt i64 %n, 0
br i1 %cmp44, label %for.body, label %for.cond.cleanup
for.cond.cleanup: ; preds = %for.body, %entry
%sum.0.lcssa = phi i64 [ 0, %entry ], [ %add15, %for.body ]
ret i64 %sum.0.lcssa
for.body: ; preds = %entry, %for.body
%sum.046 = phi i64 [ %add15, %for.body ], [ 0, %entry ]
%i.045 = phi i64 [ %inc, %for.body ], [ 0, %entry ]
%add = add i64 %i.045, %base1
%add.ptr8.idx = add i64 %add, %offset
%add.ptr8 = getelementptr inbounds i8, i8* %p, i64 %add.ptr8.idx
%0 = bitcast i8* %add.ptr8 to i64*
%1 = load i64, i64* %0, align 8
%add1 = add i64 %i.045, %mul
%add.ptr9.idx = add i64 %add1, %base1
%add.ptr9 = getelementptr inbounds i8, i8* %p, i64 %add.ptr9.idx
%2 = bitcast i8* %add.ptr9 to i64*
%3 = load i64, i64* %2, align 8
%add2 = add i64 %i.045, %base2
%add.ptr10.idx = add i64 %add2, %offset
%add.ptr10 = getelementptr inbounds i8, i8* %p, i64 %add.ptr10.idx
%4 = bitcast i8* %add.ptr10 to i64*
%5 = load i64, i64* %4, align 8
%add.ptr11.idx = add i64 %add2, %mul
%add.ptr11 = getelementptr inbounds i8, i8* %p, i64 %add.ptr11.idx
%6 = bitcast i8* %add.ptr11 to i64*
%7 = load i64, i64* %6, align 8
%mul12 = mul i64 %3, %1
%mul13 = mul i64 %mul12, %5
%mul14 = mul i64 %mul13, %7
%add15 = add i64 %mul14, %sum.046
%inc = add nuw nsw i64 %i.045, 1
%exitcond.not = icmp eq i64 %inc, %n
br i1 %exitcond.not, label %for.cond.cleanup, label %for.body
}
;
; Check chain commoning can reduce register pressure to save register spill/reload.
;
; int spill_reduce_succ(double *input1, double *input2, double *output, long long m, long long inc1, long long inc2, long long inc3, long long inc4, long long inc) {
; inc = inc4;
; #pragma unroll 4
; for (long long i = 0; i < 4 * m; i++) {
; output[inc + inc1] += input1[inc + inc1] * input2[inc + inc1];
; output[inc + inc2] += input1[inc + inc2] * input2[inc + inc2];
; output[inc + inc3] += input1[inc + inc3] * input2[inc + inc3];
; inc = inc + inc4;
; }
; return 0;
; }
;
define signext i32 @spill_reduce_succ(double* %input1, double* %input2, double* %output, i64 %m, i64 %inc1, i64 %inc2, i64 %inc3, i64 %inc4, i64 %inc) {
; CHECK-LABEL: spill_reduce_succ:
; CHECK: # %bb.0: # %entry
; CHECK-NEXT: cmpdi r6, 1
; CHECK-NEXT: std r14, -144(r1) # 8-byte Folded Spill
; CHECK-NEXT: std r15, -136(r1) # 8-byte Folded Spill
; CHECK-NEXT: std r16, -128(r1) # 8-byte Folded Spill
; CHECK-NEXT: std r17, -120(r1) # 8-byte Folded Spill
; CHECK-NEXT: std r18, -112(r1) # 8-byte Folded Spill
; CHECK-NEXT: std r19, -104(r1) # 8-byte Folded Spill
; CHECK-NEXT: std r20, -96(r1) # 8-byte Folded Spill
; CHECK-NEXT: std r21, -88(r1) # 8-byte Folded Spill
; CHECK-NEXT: std r22, -80(r1) # 8-byte Folded Spill
; CHECK-NEXT: std r23, -72(r1) # 8-byte Folded Spill
; CHECK-NEXT: std r24, -64(r1) # 8-byte Folded Spill
; CHECK-NEXT: std r25, -56(r1) # 8-byte Folded Spill
; CHECK-NEXT: std r26, -48(r1) # 8-byte Folded Spill
; CHECK-NEXT: std r27, -40(r1) # 8-byte Folded Spill
; CHECK-NEXT: std r28, -32(r1) # 8-byte Folded Spill
; CHECK-NEXT: std r29, -24(r1) # 8-byte Folded Spill
; CHECK-NEXT: std r30, -16(r1) # 8-byte Folded Spill
; CHECK-NEXT: std r31, -8(r1) # 8-byte Folded Spill
; CHECK-NEXT: std r2, -152(r1) # 8-byte Folded Spill
; CHECK-NEXT: std r9, -160(r1) # 8-byte Folded Spill
; CHECK-NEXT: std r8, -176(r1) # 8-byte Folded Spill
; CHECK-NEXT: std r7, -168(r1) # 8-byte Folded Spill
; CHECK-NEXT: blt cr0, .LBB7_7
; CHECK-NEXT: # %bb.1: # %for.body.preheader
; CHECK-NEXT: sldi r6, r6, 2
; CHECK-NEXT: li r7, 1
; CHECK-NEXT: mr r12, r10
; CHECK-NEXT: cmpdi r6, 1
; CHECK-NEXT: iselgt r7, r6, r7
; CHECK-NEXT: addi r8, r7, -1
; CHECK-NEXT: clrldi r6, r7, 63
; CHECK-NEXT: cmpldi r8, 3
; CHECK-NEXT: blt cr0, .LBB7_4
; CHECK-NEXT: # %bb.2: # %for.body.preheader.new
; CHECK-NEXT: rldicl r7, r7, 62, 2
; CHECK-NEXT: sldi r10, r12, 2
; CHECK-NEXT: ld r2, -168(r1) # 8-byte Folded Reload
; CHECK-NEXT: rldicl r7, r7, 2, 1
; CHECK-NEXT: std r7, -184(r1) # 8-byte Folded Spill
; CHECK-NEXT: ld r7, -160(r1) # 8-byte Folded Reload
; CHECK-NEXT: add r8, r7, r10
; CHECK-NEXT: mr r22, r7
; CHECK-NEXT: mr r7, r4
; CHECK-NEXT: mr r4, r3
; CHECK-NEXT: ld r3, -176(r1) # 8-byte Folded Reload
; CHECK-NEXT: sldi r8, r8, 3
; CHECK-NEXT: add r9, r5, r8
; CHECK-NEXT: add r8, r3, r10
; CHECK-NEXT: add r10, r2, r10
; CHECK-NEXT: sldi r10, r10, 3
; CHECK-NEXT: sldi r8, r8, 3
; CHECK-NEXT: add r30, r5, r10
; CHECK-NEXT: add r29, r7, r10
; CHECK-NEXT: add r28, r4, r10
; CHECK-NEXT: sldi r10, r12, 1
; CHECK-NEXT: add r8, r5, r8
; CHECK-NEXT: add r11, r12, r10
; CHECK-NEXT: add r0, r22, r11
; CHECK-NEXT: sldi r0, r0, 3
; CHECK-NEXT: add r27, r5, r0
; CHECK-NEXT: add r0, r3, r11
; CHECK-NEXT: add r11, r2, r11
; CHECK-NEXT: sldi r11, r11, 3
; CHECK-NEXT: sldi r0, r0, 3
; CHECK-NEXT: add r25, r5, r11
; CHECK-NEXT: add r24, r7, r11
; CHECK-NEXT: add r23, r4, r11
; CHECK-NEXT: add r11, r22, r10
; CHECK-NEXT: add r26, r5, r0
; CHECK-NEXT: mr r0, r22
; CHECK-NEXT: sldi r11, r11, 3
; CHECK-NEXT: add r22, r5, r11
; CHECK-NEXT: add r11, r3, r10
; CHECK-NEXT: add r10, r2, r10
; CHECK-NEXT: sldi r10, r10, 3
; CHECK-NEXT: sldi r11, r11, 3
; CHECK-NEXT: add r20, r5, r10
; CHECK-NEXT: add r19, r7, r10
; CHECK-NEXT: add r18, r4, r10
; CHECK-NEXT: add r10, r12, r0
; CHECK-NEXT: add r21, r5, r11
; CHECK-NEXT: sldi r11, r2, 3
; CHECK-NEXT: sldi r10, r10, 3
; CHECK-NEXT: add r17, r5, r10
; CHECK-NEXT: add r10, r12, r3
; CHECK-NEXT: sldi r10, r10, 3
; CHECK-NEXT: add r16, r5, r10
; CHECK-NEXT: add r10, r12, r2
; CHECK-NEXT: sldi r10, r10, 3
; CHECK-NEXT: add r15, r5, r10
; CHECK-NEXT: add r14, r7, r10
; CHECK-NEXT: add r31, r4, r10
; CHECK-NEXT: sldi r10, r3, 3
; CHECK-NEXT: mr r3, r4
; CHECK-NEXT: mr r4, r7
; CHECK-NEXT: ld r7, -160(r1) # 8-byte Folded Reload
; CHECK-NEXT: sub r0, r10, r11
; CHECK-NEXT: sldi r10, r7, 3
; CHECK-NEXT: ld r7, -184(r1) # 8-byte Folded Reload
; CHECK-NEXT: sub r2, r10, r11
; CHECK-NEXT: li r11, 0
; CHECK-NEXT: mr r10, r12
; CHECK-NEXT: addi r7, r7, -4
; CHECK-NEXT: rldicl r7, r7, 62, 2
; CHECK-NEXT: addi r7, r7, 1
; CHECK-NEXT: mtctr r7
; CHECK-NEXT: sldi r7, r12, 5
; CHECK-NEXT: .p2align 4
; CHECK-NEXT: .LBB7_3: # %for.body
; CHECK-NEXT: #
; CHECK-NEXT: lfd f0, 0(r31)
; CHECK-NEXT: lfd f1, 0(r14)
; CHECK-NEXT: add r10, r10, r12
; CHECK-NEXT: add r10, r10, r12
; CHECK-NEXT: xsmuldp f0, f0, f1
; CHECK-NEXT: lfd f1, 0(r15)
; CHECK-NEXT: add r10, r10, r12
; CHECK-NEXT: add r10, r10, r12
; CHECK-NEXT: xsadddp f0, f1, f0
; CHECK-NEXT: stfd f0, 0(r15)
; CHECK-NEXT: add r15, r15, r7
; CHECK-NEXT: lfdx f0, r31, r0
; CHECK-NEXT: lfdx f1, r14, r0
; CHECK-NEXT: xsmuldp f0, f0, f1
; CHECK-NEXT: lfdx f1, r16, r11
; CHECK-NEXT: xsadddp f0, f1, f0
; CHECK-NEXT: stfdx f0, r16, r11
; CHECK-NEXT: lfdx f0, r31, r2
; CHECK-NEXT: lfdx f1, r14, r2
; CHECK-NEXT: add r31, r31, r7
; CHECK-NEXT: add r14, r14, r7
; CHECK-NEXT: xsmuldp f0, f0, f1
; CHECK-NEXT: lfdx f1, r17, r11
; CHECK-NEXT: xsadddp f0, f1, f0
; CHECK-NEXT: stfdx f0, r17, r11
; CHECK-NEXT: lfd f0, 0(r18)
; CHECK-NEXT: lfd f1, 0(r19)
; CHECK-NEXT: xsmuldp f0, f0, f1
; CHECK-NEXT: lfdx f1, r20, r11
; CHECK-NEXT: xsadddp f0, f1, f0
; CHECK-NEXT: stfdx f0, r20, r11
; CHECK-NEXT: lfdx f0, r18, r0
; CHECK-NEXT: lfdx f1, r19, r0
; CHECK-NEXT: xsmuldp f0, f0, f1
; CHECK-NEXT: lfdx f1, r21, r11
; CHECK-NEXT: xsadddp f0, f1, f0
; CHECK-NEXT: stfdx f0, r21, r11
; CHECK-NEXT: lfdx f0, r18, r2
; CHECK-NEXT: lfdx f1, r19, r2
; CHECK-NEXT: add r18, r18, r7
; CHECK-NEXT: add r19, r19, r7
; CHECK-NEXT: xsmuldp f0, f0, f1
; CHECK-NEXT: lfdx f1, r22, r11
; CHECK-NEXT: xsadddp f0, f1, f0
; CHECK-NEXT: stfdx f0, r22, r11
; CHECK-NEXT: lfd f0, 0(r23)
; CHECK-NEXT: lfd f1, 0(r24)
; CHECK-NEXT: xsmuldp f0, f0, f1
; CHECK-NEXT: lfdx f1, r25, r11
; CHECK-NEXT: xsadddp f0, f1, f0
; CHECK-NEXT: stfdx f0, r25, r11
; CHECK-NEXT: lfdx f0, r23, r0
; CHECK-NEXT: lfdx f1, r24, r0
; CHECK-NEXT: xsmuldp f0, f0, f1
; CHECK-NEXT: lfdx f1, r26, r11
; CHECK-NEXT: xsadddp f0, f1, f0
; CHECK-NEXT: stfdx f0, r26, r11
; CHECK-NEXT: lfdx f0, r23, r2
; CHECK-NEXT: lfdx f1, r24, r2
; CHECK-NEXT: add r23, r23, r7
; CHECK-NEXT: add r24, r24, r7
; CHECK-NEXT: xsmuldp f0, f0, f1
; CHECK-NEXT: lfdx f1, r27, r11
; CHECK-NEXT: xsadddp f0, f1, f0
; CHECK-NEXT: stfdx f0, r27, r11
; CHECK-NEXT: lfd f0, 0(r28)
; CHECK-NEXT: lfd f1, 0(r29)
; CHECK-NEXT: xsmuldp f0, f0, f1
; CHECK-NEXT: lfdx f1, r30, r11
; CHECK-NEXT: xsadddp f0, f1, f0
; CHECK-NEXT: stfdx f0, r30, r11
; CHECK-NEXT: lfdx f0, r28, r0
; CHECK-NEXT: lfdx f1, r29, r0
; CHECK-NEXT: xsmuldp f0, f0, f1
; CHECK-NEXT: lfdx f1, r8, r11
; CHECK-NEXT: xsadddp f0, f1, f0
; CHECK-NEXT: stfdx f0, r8, r11
; CHECK-NEXT: lfdx f0, r28, r2
; CHECK-NEXT: lfdx f1, r29, r2
; CHECK-NEXT: add r28, r28, r7
; CHECK-NEXT: add r29, r29, r7
; CHECK-NEXT: xsmuldp f0, f0, f1
; CHECK-NEXT: lfdx f1, r9, r11
; CHECK-NEXT: xsadddp f0, f1, f0
; CHECK-NEXT: stfdx f0, r9, r11
; CHECK-NEXT: add r11, r11, r7
; CHECK-NEXT: bdnz .LBB7_3
; CHECK-NEXT: .LBB7_4: # %for.cond.cleanup.loopexit.unr-lcssa
; CHECK-NEXT: cmpldi r6, 0
; CHECK-NEXT: beq cr0, .LBB7_7
; CHECK-NEXT: # %bb.5: # %for.body.epil.preheader
; CHECK-NEXT: sldi r8, r12, 3
; CHECK-NEXT: ld r12, -176(r1) # 8-byte Folded Reload
; CHECK-NEXT: ld r7, -160(r1) # 8-byte Folded Reload
; CHECK-NEXT: add r12, r10, r12
; CHECK-NEXT: add r7, r10, r7
; CHECK-NEXT: sldi r0, r12, 3
; CHECK-NEXT: sldi r11, r7, 3
; CHECK-NEXT: add r12, r5, r0
; CHECK-NEXT: add r30, r4, r0
; CHECK-NEXT: add r29, r3, r0
; CHECK-NEXT: ld r0, -168(r1) # 8-byte Folded Reload
; CHECK-NEXT: add r7, r5, r11
; CHECK-NEXT: add r9, r4, r11
; CHECK-NEXT: add r11, r3, r11
; CHECK-NEXT: add r10, r10, r0
; CHECK-NEXT: sldi r10, r10, 3
; CHECK-NEXT: add r5, r5, r10
; CHECK-NEXT: add r4, r4, r10
; CHECK-NEXT: add r3, r3, r10
; CHECK-NEXT: li r10, 0
; CHECK-NEXT: .p2align 4
; CHECK-NEXT: .LBB7_6: # %for.body.epil
; CHECK-NEXT: #
; CHECK-NEXT: lfdx f0, r3, r10
; CHECK-NEXT: lfdx f1, r4, r10
; CHECK-NEXT: addi r6, r6, -1
; CHECK-NEXT: cmpldi r6, 0
; CHECK-NEXT: xsmuldp f0, f0, f1
; CHECK-NEXT: lfd f1, 0(r5)
; CHECK-NEXT: xsadddp f0, f1, f0
; CHECK-NEXT: stfd f0, 0(r5)
; CHECK-NEXT: add r5, r5, r8
; CHECK-NEXT: lfdx f0, r29, r10
; CHECK-NEXT: lfdx f1, r30, r10
; CHECK-NEXT: xsmuldp f0, f0, f1
; CHECK-NEXT: lfdx f1, r12, r10
; CHECK-NEXT: xsadddp f0, f1, f0
; CHECK-NEXT: stfdx f0, r12, r10
; CHECK-NEXT: lfdx f0, r11, r10
; CHECK-NEXT: lfdx f1, r9, r10
; CHECK-NEXT: xsmuldp f0, f0, f1
; CHECK-NEXT: lfdx f1, r7, r10
; CHECK-NEXT: xsadddp f0, f1, f0
; CHECK-NEXT: stfdx f0, r7, r10
; CHECK-NEXT: add r10, r10, r8
; CHECK-NEXT: bne cr0, .LBB7_6
; CHECK-NEXT: .LBB7_7: # %for.cond.cleanup
; CHECK-NEXT: ld r2, -152(r1) # 8-byte Folded Reload
; CHECK-NEXT: ld r31, -8(r1) # 8-byte Folded Reload
; CHECK-NEXT: ld r30, -16(r1) # 8-byte Folded Reload
; CHECK-NEXT: ld r29, -24(r1) # 8-byte Folded Reload
; CHECK-NEXT: li r3, 0
; CHECK-NEXT: ld r28, -32(r1) # 8-byte Folded Reload
; CHECK-NEXT: ld r27, -40(r1) # 8-byte Folded Reload
; CHECK-NEXT: ld r26, -48(r1) # 8-byte Folded Reload
; CHECK-NEXT: ld r25, -56(r1) # 8-byte Folded Reload
; CHECK-NEXT: ld r24, -64(r1) # 8-byte Folded Reload
; CHECK-NEXT: ld r23, -72(r1) # 8-byte Folded Reload
; CHECK-NEXT: ld r22, -80(r1) # 8-byte Folded Reload
; CHECK-NEXT: ld r21, -88(r1) # 8-byte Folded Reload
; CHECK-NEXT: ld r20, -96(r1) # 8-byte Folded Reload
; CHECK-NEXT: ld r19, -104(r1) # 8-byte Folded Reload
; CHECK-NEXT: ld r18, -112(r1) # 8-byte Folded Reload
; CHECK-NEXT: ld r17, -120(r1) # 8-byte Folded Reload
; CHECK-NEXT: ld r16, -128(r1) # 8-byte Folded Reload
; CHECK-NEXT: ld r15, -136(r1) # 8-byte Folded Reload
; CHECK-NEXT: ld r14, -144(r1) # 8-byte Folded Reload
; CHECK-NEXT: blr
entry:
%cmp49 = icmp sgt i64 %m, 0
br i1 %cmp49, label %for.body.preheader, label %for.cond.cleanup
for.body.preheader: ; preds = %entry
%0 = shl i64 %m, 2
%smax52 = call i64 @llvm.smax.i64(i64 %0, i64 1)
%1 = add nsw i64 %smax52, -1
%xtraiter = and i64 %smax52, 1
%2 = icmp ult i64 %1, 3
br i1 %2, label %for.cond.cleanup.loopexit.unr-lcssa, label %for.body.preheader.new
for.body.preheader.new: ; preds = %for.body.preheader
%unroll_iter = and i64 %smax52, 9223372036854775804
br label %for.body
for.cond.cleanup.loopexit.unr-lcssa: ; preds = %for.body, %for.body.preheader
%inc.addr.050.unr = phi i64 [ %inc4, %for.body.preheader ], [ %add23.3, %for.body ]
%lcmp.mod.not = icmp eq i64 %xtraiter, 0
br i1 %lcmp.mod.not, label %for.cond.cleanup, label %for.body.epil
for.body.epil: ; preds = %for.cond.cleanup.loopexit.unr-lcssa, %for.body.epil
%inc.addr.050.epil = phi i64 [ %add23.epil, %for.body.epil ], [ %inc.addr.050.unr, %for.cond.cleanup.loopexit.unr-lcssa ]
%epil.iter = phi i64 [ %epil.iter.sub, %for.body.epil ], [ %xtraiter, %for.cond.cleanup.loopexit.unr-lcssa ]
%add.epil = add nsw i64 %inc.addr.050.epil, %inc1
%arrayidx.epil = getelementptr inbounds double, double* %input1, i64 %add.epil
%3 = load double, double* %arrayidx.epil, align 8
%arrayidx2.epil = getelementptr inbounds double, double* %input2, i64 %add.epil
%4 = load double, double* %arrayidx2.epil, align 8
%mul3.epil = fmul double %3, %4
%arrayidx5.epil = getelementptr inbounds double, double* %output, i64 %add.epil
%5 = load double, double* %arrayidx5.epil, align 8
%add6.epil = fadd double %5, %mul3.epil
store double %add6.epil, double* %arrayidx5.epil, align 8
%add7.epil = add nsw i64 %inc.addr.050.epil, %inc2
%arrayidx8.epil = getelementptr inbounds double, double* %input1, i64 %add7.epil
%6 = load double, double* %arrayidx8.epil, align 8
%arrayidx10.epil = getelementptr inbounds double, double* %input2, i64 %add7.epil
%7 = load double, double* %arrayidx10.epil, align 8
%mul11.epil = fmul double %6, %7
%arrayidx13.epil = getelementptr inbounds double, double* %output, i64 %add7.epil
%8 = load double, double* %arrayidx13.epil, align 8
%add14.epil = fadd double %8, %mul11.epil
store double %add14.epil, double* %arrayidx13.epil, align 8
%add15.epil = add nsw i64 %inc.addr.050.epil, %inc3
%arrayidx16.epil = getelementptr inbounds double, double* %input1, i64 %add15.epil
%9 = load double, double* %arrayidx16.epil, align 8
%arrayidx18.epil = getelementptr inbounds double, double* %input2, i64 %add15.epil
%10 = load double, double* %arrayidx18.epil, align 8
%mul19.epil = fmul double %9, %10
%arrayidx21.epil = getelementptr inbounds double, double* %output, i64 %add15.epil
%11 = load double, double* %arrayidx21.epil, align 8
%add22.epil = fadd double %11, %mul19.epil
store double %add22.epil, double* %arrayidx21.epil, align 8
%add23.epil = add nsw i64 %inc.addr.050.epil, %inc4
%epil.iter.sub = add nsw i64 %epil.iter, -1
%epil.iter.cmp.not = icmp eq i64 %epil.iter.sub, 0
br i1 %epil.iter.cmp.not, label %for.cond.cleanup, label %for.body.epil
for.cond.cleanup: ; preds = %for.body.epil, %for.cond.cleanup.loopexit.unr-lcssa, %entry
ret i32 0
for.body: ; preds = %for.body, %for.body.preheader.new
%inc.addr.050 = phi i64 [ %inc4, %for.body.preheader.new ], [ %add23.3, %for.body ]
%niter = phi i64 [ %unroll_iter, %for.body.preheader.new ], [ %niter.nsub.3, %for.body ]
%add = add nsw i64 %inc.addr.050, %inc1
%arrayidx = getelementptr inbounds double, double* %input1, i64 %add
%12 = load double, double* %arrayidx, align 8
%arrayidx2 = getelementptr inbounds double, double* %input2, i64 %add
%13 = load double, double* %arrayidx2, align 8
%mul3 = fmul double %12, %13
%arrayidx5 = getelementptr inbounds double, double* %output, i64 %add
%14 = load double, double* %arrayidx5, align 8
%add6 = fadd double %14, %mul3
store double %add6, double* %arrayidx5, align 8
%add7 = add nsw i64 %inc.addr.050, %inc2
%arrayidx8 = getelementptr inbounds double, double* %input1, i64 %add7
%15 = load double, double* %arrayidx8, align 8
%arrayidx10 = getelementptr inbounds double, double* %input2, i64 %add7
%16 = load double, double* %arrayidx10, align 8
%mul11 = fmul double %15, %16
%arrayidx13 = getelementptr inbounds double, double* %output, i64 %add7
%17 = load double, double* %arrayidx13, align 8
%add14 = fadd double %17, %mul11
store double %add14, double* %arrayidx13, align 8
%add15 = add nsw i64 %inc.addr.050, %inc3
%arrayidx16 = getelementptr inbounds double, double* %input1, i64 %add15
%18 = load double, double* %arrayidx16, align 8
%arrayidx18 = getelementptr inbounds double, double* %input2, i64 %add15
%19 = load double, double* %arrayidx18, align 8
%mul19 = fmul double %18, %19
%arrayidx21 = getelementptr inbounds double, double* %output, i64 %add15
%20 = load double, double* %arrayidx21, align 8
%add22 = fadd double %20, %mul19
store double %add22, double* %arrayidx21, align 8
%add23 = add nsw i64 %inc.addr.050, %inc4
%add.1 = add nsw i64 %add23, %inc1
%arrayidx.1 = getelementptr inbounds double, double* %input1, i64 %add.1
%21 = load double, double* %arrayidx.1, align 8
%arrayidx2.1 = getelementptr inbounds double, double* %input2, i64 %add.1
%22 = load double, double* %arrayidx2.1, align 8
%mul3.1 = fmul double %21, %22
%arrayidx5.1 = getelementptr inbounds double, double* %output, i64 %add.1
%23 = load double, double* %arrayidx5.1, align 8
%add6.1 = fadd double %23, %mul3.1
store double %add6.1, double* %arrayidx5.1, align 8
%add7.1 = add nsw i64 %add23, %inc2
%arrayidx8.1 = getelementptr inbounds double, double* %input1, i64 %add7.1
%24 = load double, double* %arrayidx8.1, align 8
%arrayidx10.1 = getelementptr inbounds double, double* %input2, i64 %add7.1
%25 = load double, double* %arrayidx10.1, align 8
%mul11.1 = fmul double %24, %25
%arrayidx13.1 = getelementptr inbounds double, double* %output, i64 %add7.1
%26 = load double, double* %arrayidx13.1, align 8
%add14.1 = fadd double %26, %mul11.1
store double %add14.1, double* %arrayidx13.1, align 8
%add15.1 = add nsw i64 %add23, %inc3
%arrayidx16.1 = getelementptr inbounds double, double* %input1, i64 %add15.1
%27 = load double, double* %arrayidx16.1, align 8
%arrayidx18.1 = getelementptr inbounds double, double* %input2, i64 %add15.1
%28 = load double, double* %arrayidx18.1, align 8
%mul19.1 = fmul double %27, %28
%arrayidx21.1 = getelementptr inbounds double, double* %output, i64 %add15.1
%29 = load double, double* %arrayidx21.1, align 8
%add22.1 = fadd double %29, %mul19.1
store double %add22.1, double* %arrayidx21.1, align 8
%add23.1 = add nsw i64 %add23, %inc4
%add.2 = add nsw i64 %add23.1, %inc1
%arrayidx.2 = getelementptr inbounds double, double* %input1, i64 %add.2
%30 = load double, double* %arrayidx.2, align 8
%arrayidx2.2 = getelementptr inbounds double, double* %input2, i64 %add.2
%31 = load double, double* %arrayidx2.2, align 8
%mul3.2 = fmul double %30, %31
%arrayidx5.2 = getelementptr inbounds double, double* %output, i64 %add.2
%32 = load double, double* %arrayidx5.2, align 8
%add6.2 = fadd double %32, %mul3.2
store double %add6.2, double* %arrayidx5.2, align 8
%add7.2 = add nsw i64 %add23.1, %inc2
%arrayidx8.2 = getelementptr inbounds double, double* %input1, i64 %add7.2
%33 = load double, double* %arrayidx8.2, align 8
%arrayidx10.2 = getelementptr inbounds double, double* %input2, i64 %add7.2
%34 = load double, double* %arrayidx10.2, align 8
%mul11.2 = fmul double %33, %34
%arrayidx13.2 = getelementptr inbounds double, double* %output, i64 %add7.2
%35 = load double, double* %arrayidx13.2, align 8
%add14.2 = fadd double %35, %mul11.2
store double %add14.2, double* %arrayidx13.2, align 8
%add15.2 = add nsw i64 %add23.1, %inc3
%arrayidx16.2 = getelementptr inbounds double, double* %input1, i64 %add15.2
%36 = load double, double* %arrayidx16.2, align 8
%arrayidx18.2 = getelementptr inbounds double, double* %input2, i64 %add15.2
%37 = load double, double* %arrayidx18.2, align 8
%mul19.2 = fmul double %36, %37
%arrayidx21.2 = getelementptr inbounds double, double* %output, i64 %add15.2
%38 = load double, double* %arrayidx21.2, align 8
%add22.2 = fadd double %38, %mul19.2
store double %add22.2, double* %arrayidx21.2, align 8
%add23.2 = add nsw i64 %add23.1, %inc4
%add.3 = add nsw i64 %add23.2, %inc1
%arrayidx.3 = getelementptr inbounds double, double* %input1, i64 %add.3
%39 = load double, double* %arrayidx.3, align 8
%arrayidx2.3 = getelementptr inbounds double, double* %input2, i64 %add.3
%40 = load double, double* %arrayidx2.3, align 8
%mul3.3 = fmul double %39, %40
%arrayidx5.3 = getelementptr inbounds double, double* %output, i64 %add.3
%41 = load double, double* %arrayidx5.3, align 8
%add6.3 = fadd double %41, %mul3.3
store double %add6.3, double* %arrayidx5.3, align 8
%add7.3 = add nsw i64 %add23.2, %inc2
%arrayidx8.3 = getelementptr inbounds double, double* %input1, i64 %add7.3
%42 = load double, double* %arrayidx8.3, align 8
%arrayidx10.3 = getelementptr inbounds double, double* %input2, i64 %add7.3
%43 = load double, double* %arrayidx10.3, align 8
%mul11.3 = fmul double %42, %43
%arrayidx13.3 = getelementptr inbounds double, double* %output, i64 %add7.3
%44 = load double, double* %arrayidx13.3, align 8
%add14.3 = fadd double %44, %mul11.3
store double %add14.3, double* %arrayidx13.3, align 8
%add15.3 = add nsw i64 %add23.2, %inc3
%arrayidx16.3 = getelementptr inbounds double, double* %input1, i64 %add15.3
%45 = load double, double* %arrayidx16.3, align 8
%arrayidx18.3 = getelementptr inbounds double, double* %input2, i64 %add15.3
%46 = load double, double* %arrayidx18.3, align 8
%mul19.3 = fmul double %45, %46
%arrayidx21.3 = getelementptr inbounds double, double* %output, i64 %add15.3
%47 = load double, double* %arrayidx21.3, align 8
%add22.3 = fadd double %47, %mul19.3
store double %add22.3, double* %arrayidx21.3, align 8
%add23.3 = add nsw i64 %add23.2, %inc4
%niter.nsub.3 = add i64 %niter, -4
%niter.ncmp.3.not = icmp eq i64 %niter.nsub.3, 0
br i1 %niter.ncmp.3.not, label %for.cond.cleanup.loopexit.unr-lcssa, label %for.body
}
declare i64 @llvm.smax.i64(i64, i64)