c4a60c9d34
This is an attempt to reland D42600 and enabling this optimisation by default. This also resolves the issue pointed out in the context of PGO build. Differential Revision: https://reviews.llvm.org/D42600
242 lines
11 KiB
LLVM
242 lines
11 KiB
LLVM
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
|
|
; RUN: llc -mtriple=armv6t2-eabi %s -o - | FileCheck %s --check-prefix=CHECK
|
|
|
|
; Checks SSAT is still generated when loop unrolling is on
|
|
|
|
define void @ssat_unroll(ptr %pSrcA, ptr %pSrcB, ptr %pDst, i32 %blockSize) {
|
|
; CHECK-LABEL: ssat_unroll:
|
|
; CHECK: @ %bb.0: @ %entry
|
|
; CHECK-NEXT: cmp r3, #0
|
|
; CHECK-NEXT: bxeq lr
|
|
; CHECK-NEXT: .LBB0_1: @ %while.body.preheader
|
|
; CHECK-NEXT: .save {r11, lr}
|
|
; CHECK-NEXT: push {r11, lr}
|
|
; CHECK-NEXT: sub r12, r3, #1
|
|
; CHECK-NEXT: tst r3, #1
|
|
; CHECK-NEXT: beq .LBB0_3
|
|
; CHECK-NEXT: @ %bb.2: @ %while.body.prol.preheader
|
|
; CHECK-NEXT: ldrsh lr, [r0], #2
|
|
; CHECK-NEXT: ldrsh r3, [r1], #2
|
|
; CHECK-NEXT: smulbb r3, r3, lr
|
|
; CHECK-NEXT: ssat r3, #16, r3, asr #14
|
|
; CHECK-NEXT: strh r3, [r2], #2
|
|
; CHECK-NEXT: mov r3, r12
|
|
; CHECK-NEXT: .LBB0_3: @ %while.body.prol.loopexit
|
|
; CHECK-NEXT: cmp r12, #0
|
|
; CHECK-NEXT: beq .LBB0_5
|
|
; CHECK-NEXT: .LBB0_4: @ %while.body
|
|
; CHECK-NEXT: @ =>This Inner Loop Header: Depth=1
|
|
; CHECK-NEXT: ldrsh r12, [r0]
|
|
; CHECK-NEXT: subs r3, r3, #2
|
|
; CHECK-NEXT: ldrsh lr, [r1]
|
|
; CHECK-NEXT: smulbb r12, lr, r12
|
|
; CHECK-NEXT: ssat r12, #16, r12, asr #14
|
|
; CHECK-NEXT: strh r12, [r2]
|
|
; CHECK-NEXT: ldrsh r12, [r0, #2]
|
|
; CHECK-NEXT: add r0, r0, #4
|
|
; CHECK-NEXT: ldrsh lr, [r1, #2]
|
|
; CHECK-NEXT: add r1, r1, #4
|
|
; CHECK-NEXT: smulbb r12, lr, r12
|
|
; CHECK-NEXT: ssat r12, #16, r12, asr #14
|
|
; CHECK-NEXT: strh r12, [r2, #2]
|
|
; CHECK-NEXT: add r2, r2, #4
|
|
; CHECK-NEXT: bne .LBB0_4
|
|
; CHECK-NEXT: .LBB0_5:
|
|
; CHECK-NEXT: pop {r11, lr}
|
|
; CHECK-NEXT: bx lr
|
|
entry:
|
|
%cmp.not7 = icmp eq i32 %blockSize, 0
|
|
br i1 %cmp.not7, label %while.end, label %while.body.preheader
|
|
|
|
while.body.preheader: ; preds = %entry
|
|
%0 = add i32 %blockSize, -1
|
|
%xtraiter = and i32 %blockSize, 1
|
|
%lcmp.mod.not = icmp eq i32 %xtraiter, 0
|
|
br i1 %lcmp.mod.not, label %while.body.prol.loopexit, label %while.body.prol.preheader
|
|
|
|
while.body.prol.preheader: ; preds = %while.body.preheader
|
|
%incdec.ptr.prol = getelementptr inbounds i16, ptr %pSrcA, i32 1
|
|
%1 = load i16, ptr %pSrcA
|
|
%conv.prol = sext i16 %1 to i32
|
|
%incdec.ptr1.prol = getelementptr inbounds i16, ptr %pSrcB, i32 1
|
|
%2 = load i16, ptr %pSrcB
|
|
%conv2.prol = sext i16 %2 to i32
|
|
%mul.prol = mul nsw i32 %conv2.prol, %conv.prol
|
|
%shr.prol = ashr i32 %mul.prol, 14
|
|
%3 = icmp sgt i32 %shr.prol, -32768
|
|
%4 = select i1 %3, i32 %shr.prol, i32 -32768
|
|
%5 = icmp slt i32 %4, 32767
|
|
%spec.select.i.prol = select i1 %5, i32 %4, i32 32767
|
|
%conv3.prol = trunc i32 %spec.select.i.prol to i16
|
|
%incdec.ptr4.prol = getelementptr inbounds i16, ptr %pDst, i32 1
|
|
store i16 %conv3.prol, ptr %pDst
|
|
br label %while.body.prol.loopexit
|
|
|
|
while.body.prol.loopexit: ; preds = %while.body.prol.preheader, %while.body.preheader
|
|
%blkCnt.011.unr = phi i32 [ %blockSize, %while.body.preheader ], [ %0, %while.body.prol.preheader ]
|
|
%pSrcA.addr.010.unr = phi ptr [ %pSrcA, %while.body.preheader ], [ %incdec.ptr.prol, %while.body.prol.preheader ]
|
|
%pDst.addr.09.unr = phi ptr [ %pDst, %while.body.preheader ], [ %incdec.ptr4.prol, %while.body.prol.preheader ]
|
|
%pSrcB.addr.08.unr = phi ptr [ %pSrcB, %while.body.preheader ], [ %incdec.ptr1.prol, %while.body.prol.preheader ]
|
|
%6 = icmp eq i32 %0, 0
|
|
br i1 %6, label %while.end, label %while.body
|
|
|
|
while.body: ; preds = %while.body.prol.loopexit, %while.body
|
|
%blkCnt.011 = phi i32 [ %dec.1, %while.body ], [ %blkCnt.011.unr, %while.body.prol.loopexit ]
|
|
%pSrcA.addr.010 = phi ptr [ %incdec.ptr.1, %while.body ], [ %pSrcA.addr.010.unr, %while.body.prol.loopexit ]
|
|
%pDst.addr.09 = phi ptr [ %incdec.ptr4.1, %while.body ], [ %pDst.addr.09.unr, %while.body.prol.loopexit ]
|
|
%pSrcB.addr.08 = phi ptr [ %incdec.ptr1.1, %while.body ], [ %pSrcB.addr.08.unr, %while.body.prol.loopexit ]
|
|
%incdec.ptr = getelementptr inbounds i16, ptr %pSrcA.addr.010, i32 1
|
|
%7 = load i16, ptr %pSrcA.addr.010
|
|
%conv = sext i16 %7 to i32
|
|
%incdec.ptr1 = getelementptr inbounds i16, ptr %pSrcB.addr.08, i32 1
|
|
%8 = load i16, ptr %pSrcB.addr.08
|
|
%conv2 = sext i16 %8 to i32
|
|
%mul = mul nsw i32 %conv2, %conv
|
|
%shr = ashr i32 %mul, 14
|
|
%9 = icmp sgt i32 %shr, -32768
|
|
%10 = select i1 %9, i32 %shr, i32 -32768
|
|
%11 = icmp slt i32 %10, 32767
|
|
%spec.select.i = select i1 %11, i32 %10, i32 32767
|
|
%conv3 = trunc i32 %spec.select.i to i16
|
|
%incdec.ptr4 = getelementptr inbounds i16, ptr %pDst.addr.09, i32 1
|
|
store i16 %conv3, ptr %pDst.addr.09
|
|
%incdec.ptr.1 = getelementptr inbounds i16, ptr %pSrcA.addr.010, i32 2
|
|
%12 = load i16, ptr %incdec.ptr
|
|
%conv.1 = sext i16 %12 to i32
|
|
%incdec.ptr1.1 = getelementptr inbounds i16, ptr %pSrcB.addr.08, i32 2
|
|
%13 = load i16, ptr %incdec.ptr1
|
|
%conv2.1 = sext i16 %13 to i32
|
|
%mul.1 = mul nsw i32 %conv2.1, %conv.1
|
|
%shr.1 = ashr i32 %mul.1, 14
|
|
%14 = icmp sgt i32 %shr.1, -32768
|
|
%15 = select i1 %14, i32 %shr.1, i32 -32768
|
|
%16 = icmp slt i32 %15, 32767
|
|
%spec.select.i.1 = select i1 %16, i32 %15, i32 32767
|
|
%conv3.1 = trunc i32 %spec.select.i.1 to i16
|
|
%incdec.ptr4.1 = getelementptr inbounds i16, ptr %pDst.addr.09, i32 2
|
|
store i16 %conv3.1, ptr %incdec.ptr4
|
|
%dec.1 = add i32 %blkCnt.011, -2
|
|
%cmp.not.1 = icmp eq i32 %dec.1, 0
|
|
br i1 %cmp.not.1, label %while.end, label %while.body
|
|
|
|
while.end: ; preds = %while.body, %while.body.prol.loopexit, %entry
|
|
ret void
|
|
}
|
|
|
|
define void @ssat_unroll_minmax(ptr nocapture readonly %pSrcA, ptr nocapture readonly %pSrcB, ptr nocapture writeonly %pDst, i32 %blockSize) {
|
|
; CHECK-LABEL: ssat_unroll_minmax:
|
|
; CHECK: @ %bb.0: @ %entry
|
|
; CHECK-NEXT: cmp r3, #0
|
|
; CHECK-NEXT: bxeq lr
|
|
; CHECK-NEXT: .LBB1_1: @ %while.body.preheader
|
|
; CHECK-NEXT: .save {r11, lr}
|
|
; CHECK-NEXT: push {r11, lr}
|
|
; CHECK-NEXT: sub r12, r3, #1
|
|
; CHECK-NEXT: tst r3, #1
|
|
; CHECK-NEXT: beq .LBB1_3
|
|
; CHECK-NEXT: @ %bb.2: @ %while.body.prol.preheader
|
|
; CHECK-NEXT: ldrsh lr, [r0], #2
|
|
; CHECK-NEXT: ldrsh r3, [r1], #2
|
|
; CHECK-NEXT: smulbb r3, r3, lr
|
|
; CHECK-NEXT: ssat r3, #16, r3, asr #14
|
|
; CHECK-NEXT: strh r3, [r2], #2
|
|
; CHECK-NEXT: mov r3, r12
|
|
; CHECK-NEXT: .LBB1_3: @ %while.body.prol.loopexit
|
|
; CHECK-NEXT: cmp r12, #0
|
|
; CHECK-NEXT: beq .LBB1_5
|
|
; CHECK-NEXT: .LBB1_4: @ %while.body
|
|
; CHECK-NEXT: @ =>This Inner Loop Header: Depth=1
|
|
; CHECK-NEXT: ldrsh r12, [r0]
|
|
; CHECK-NEXT: subs r3, r3, #2
|
|
; CHECK-NEXT: ldrsh lr, [r1]
|
|
; CHECK-NEXT: smulbb r12, lr, r12
|
|
; CHECK-NEXT: ssat r12, #16, r12, asr #14
|
|
; CHECK-NEXT: strh r12, [r2]
|
|
; CHECK-NEXT: ldrsh r12, [r0, #2]
|
|
; CHECK-NEXT: add r0, r0, #4
|
|
; CHECK-NEXT: ldrsh lr, [r1, #2]
|
|
; CHECK-NEXT: add r1, r1, #4
|
|
; CHECK-NEXT: smulbb r12, lr, r12
|
|
; CHECK-NEXT: ssat r12, #16, r12, asr #14
|
|
; CHECK-NEXT: strh r12, [r2, #2]
|
|
; CHECK-NEXT: add r2, r2, #4
|
|
; CHECK-NEXT: bne .LBB1_4
|
|
; CHECK-NEXT: .LBB1_5:
|
|
; CHECK-NEXT: pop {r11, lr}
|
|
; CHECK-NEXT: bx lr
|
|
entry:
|
|
%cmp.not7 = icmp eq i32 %blockSize, 0
|
|
br i1 %cmp.not7, label %while.end, label %while.body.preheader
|
|
|
|
while.body.preheader: ; preds = %entry
|
|
%0 = add i32 %blockSize, -1
|
|
%xtraiter = and i32 %blockSize, 1
|
|
%lcmp.mod.not = icmp eq i32 %xtraiter, 0
|
|
br i1 %lcmp.mod.not, label %while.body.prol.loopexit, label %while.body.prol.preheader
|
|
|
|
while.body.prol.preheader: ; preds = %while.body.preheader
|
|
%incdec.ptr.prol = getelementptr inbounds i16, ptr %pSrcA, i64 1
|
|
%1 = load i16, ptr %pSrcA, align 2
|
|
%conv.prol = sext i16 %1 to i32
|
|
%incdec.ptr1.prol = getelementptr inbounds i16, ptr %pSrcB, i64 1
|
|
%2 = load i16, ptr %pSrcB, align 2
|
|
%conv2.prol = sext i16 %2 to i32
|
|
%mul.prol = mul nsw i32 %conv2.prol, %conv.prol
|
|
%shr.prol = ashr i32 %mul.prol, 14
|
|
%3 = call i32 @llvm.smax.i32(i32 %shr.prol, i32 -32768)
|
|
%4 = call i32 @llvm.smin.i32(i32 %3, i32 32767)
|
|
%conv3.prol = trunc i32 %4 to i16
|
|
%incdec.ptr4.prol = getelementptr inbounds i16, ptr %pDst, i64 1
|
|
store i16 %conv3.prol, ptr %pDst, align 2
|
|
br label %while.body.prol.loopexit
|
|
|
|
while.body.prol.loopexit: ; preds = %while.body.prol.preheader, %while.body.preheader
|
|
%blkCnt.011.unr = phi i32 [ %blockSize, %while.body.preheader ], [ %0, %while.body.prol.preheader ]
|
|
%pSrcA.addr.010.unr = phi ptr [ %pSrcA, %while.body.preheader ], [ %incdec.ptr.prol, %while.body.prol.preheader ]
|
|
%pDst.addr.09.unr = phi ptr [ %pDst, %while.body.preheader ], [ %incdec.ptr4.prol, %while.body.prol.preheader ]
|
|
%pSrcB.addr.08.unr = phi ptr [ %pSrcB, %while.body.preheader ], [ %incdec.ptr1.prol, %while.body.prol.preheader ]
|
|
%5 = icmp eq i32 %0, 0
|
|
br i1 %5, label %while.end, label %while.body
|
|
|
|
while.body: ; preds = %while.body.prol.loopexit, %while.body
|
|
%blkCnt.011 = phi i32 [ %dec.1, %while.body ], [ %blkCnt.011.unr, %while.body.prol.loopexit ]
|
|
%pSrcA.addr.010 = phi ptr [ %incdec.ptr.1, %while.body ], [ %pSrcA.addr.010.unr, %while.body.prol.loopexit ]
|
|
%pDst.addr.09 = phi ptr [ %incdec.ptr4.1, %while.body ], [ %pDst.addr.09.unr, %while.body.prol.loopexit ]
|
|
%pSrcB.addr.08 = phi ptr [ %incdec.ptr1.1, %while.body ], [ %pSrcB.addr.08.unr, %while.body.prol.loopexit ]
|
|
%incdec.ptr = getelementptr inbounds i16, ptr %pSrcA.addr.010, i64 1
|
|
%6 = load i16, ptr %pSrcA.addr.010, align 2
|
|
%conv = sext i16 %6 to i32
|
|
%incdec.ptr1 = getelementptr inbounds i16, ptr %pSrcB.addr.08, i64 1
|
|
%7 = load i16, ptr %pSrcB.addr.08, align 2
|
|
%conv2 = sext i16 %7 to i32
|
|
%mul = mul nsw i32 %conv2, %conv
|
|
%shr = ashr i32 %mul, 14
|
|
%8 = call i32 @llvm.smax.i32(i32 %shr, i32 -32768)
|
|
%9 = call i32 @llvm.smin.i32(i32 %8, i32 32767)
|
|
%conv3 = trunc i32 %9 to i16
|
|
%incdec.ptr4 = getelementptr inbounds i16, ptr %pDst.addr.09, i64 1
|
|
store i16 %conv3, ptr %pDst.addr.09, align 2
|
|
%incdec.ptr.1 = getelementptr inbounds i16, ptr %pSrcA.addr.010, i64 2
|
|
%10 = load i16, ptr %incdec.ptr, align 2
|
|
%conv.1 = sext i16 %10 to i32
|
|
%incdec.ptr1.1 = getelementptr inbounds i16, ptr %pSrcB.addr.08, i64 2
|
|
%11 = load i16, ptr %incdec.ptr1, align 2
|
|
%conv2.1 = sext i16 %11 to i32
|
|
%mul.1 = mul nsw i32 %conv2.1, %conv.1
|
|
%shr.1 = ashr i32 %mul.1, 14
|
|
%12 = call i32 @llvm.smax.i32(i32 %shr.1, i32 -32768)
|
|
%13 = call i32 @llvm.smin.i32(i32 %12, i32 32767)
|
|
%conv3.1 = trunc i32 %13 to i16
|
|
%incdec.ptr4.1 = getelementptr inbounds i16, ptr %pDst.addr.09, i64 2
|
|
store i16 %conv3.1, ptr %incdec.ptr4, align 2
|
|
%dec.1 = add i32 %blkCnt.011, -2
|
|
%cmp.not.1 = icmp eq i32 %dec.1, 0
|
|
br i1 %cmp.not.1, label %while.end, label %while.body
|
|
|
|
while.end: ; preds = %while.body, %while.body.prol.loopexit, %entry
|
|
ret void
|
|
}
|
|
|
|
declare i32 @llvm.smax.i32(i32, i32) #1
|
|
declare i32 @llvm.smin.i32(i32, i32) #1
|