caio/clang-p2996
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
Files
be1294de9dd93ae02002bd1aa4b4a8d62e562c21
clang-p2996/llvm/test/CodeGen/Thumb2/LowOverheadLoops/inloop-vpsel-2.mir
David Green 9cb8f4d1ad [ARM] Add a tail-predication loop predicate register 
The semantics of tail predication loops means that the value of LR as an
instruction is executed determines the predicate. In other words:

mov r3, #3
DLSTP lr, r3        // Start tail predication, lr==3
VADD.s32 q0, q1, q2 // Lanes 0,1 and 2 are updated in q0.
mov lr, #1
VADD.s32 q0, q1, q2 // Only first lane is updated.

This means that the value of lr cannot be spilled and re-used in tail
predication regions without potentially altering the behaviour of the
program. More lanes than required could be stored, for example, and in
the case of a gather those lanes might not have been setup, leading to
alignment exceptions.

This patch adds a new lr predicate operand to MVE instructions in order
to keep a reference to the lr that they use as a tail predicate. It will
usually hold the zeroreg meaning not predicated, being set to the LR phi
value in the MVETPAndVPTOptimisationsPass. This will prevent it from
being spilled anywhere that it needs to be used.

A lot of tests needed updating.

Differential Revision: https://reviews.llvm.org/D107638
2021-09-02 13:42:58 +01:00

240 lines
13 KiB
YAML
Raw Blame History

# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py
# RUN: llc -mtriple=thumbv8.1m.main -mattr=+mve -run-pass=arm-low-overhead-loops %s -o - --verify-machineinstrs | FileCheck %s
# Test that, even if vpsels are allowed, that they aren't inserted into a vpt
# block - which would cause an assert here because of the number of insts in
# the block.
--- |
define dso_local i32 @vpsel_after_vpt(i16* nocapture readonly %a, i16* nocapture readonly %b, i16* nocapture readonly %c, i16* nocapture readonly %d, i32 %N) local_unnamed_addr #0 {
entry:
%cmp9 = icmp eq i32 %N, 0
%tmp = add i32 %N, 3
%tmp1 = lshr i32 %tmp, 2
%tmp2 = shl nuw i32 %tmp1, 2
%tmp3 = add i32 %tmp2, -4
%tmp4 = lshr i32 %tmp3, 2
%tmp5 = add nuw nsw i32 %tmp4, 1
br i1 %cmp9, label %for.cond.cleanup, label %vector.ph
vector.ph: ; preds = %entry
%start = call i32 @llvm.start.loop.iterations.i32(i32 %tmp5)
br label %vector.body
vector.body: ; preds = %vector.body, %vector.ph
%lsr.iv1 = phi i32 [ %lsr.iv.next, %vector.body ], [ %start, %vector.ph ]
%lsr.iv.d = phi i16* [ %scevgep.d, %vector.body ], [ %d, %vector.ph ]
%lsr.iv.c = phi i16* [ %scevgep.c, %vector.body ], [ %c, %vector.ph ]
%lsr.iv18 = phi i16* [ %scevgep19, %vector.body ], [ %b, %vector.ph ]
%lsr.iv = phi i16* [ %scevgep, %vector.body ], [ %a, %vector.ph ]
%vec.phi = phi <4 x i32> [ zeroinitializer, %vector.ph ], [ %tmp14, %vector.body ]
%tmp7 = phi i32 [ %N, %vector.ph ], [ %tmp9, %vector.body ]
%lsr.iv17 = bitcast i16* %lsr.iv to <4 x i16>*
%lsr.iv1820 = bitcast i16* %lsr.iv18 to <4 x i16>*
%lsr.iv1820.c = bitcast i16* %lsr.iv.c to <4 x i16>*
%lsr.iv17.d = bitcast i16* %lsr.iv.d to <4 x i16>*
%tmp8 = call <4 x i1> @llvm.arm.mve.vctp32(i32 %tmp7)
%tmp9 = sub i32 %tmp7, 4
%wide.masked.load = call <4 x i16> @llvm.masked.load.v4i16.p0v4i16(<4 x i16>* %lsr.iv17, i32 2, <4 x i1> %tmp8, <4 x i16> undef)
%tmp10 = sext <4 x i16> %wide.masked.load to <4 x i32>
%wide.masked.load14 = call <4 x i16> @llvm.masked.load.v4i16.p0v4i16(<4 x i16>* %lsr.iv1820, i32 2, <4 x i1> %tmp8, <4 x i16> undef)
%tmp11 = sext <4 x i16> %wide.masked.load14 to <4 x i32>
%wide.masked.load.c = call <4 x i16> @llvm.masked.load.v4i16.p0v4i16(<4 x i16>* %lsr.iv1820.c, i32 2, <4 x i1> %tmp8, <4 x i16> undef)
%sext.load.c = sext <4 x i16> %wide.masked.load.c to <4 x i32>
%wide.masked.load.d = call <4 x i16> @llvm.masked.load.v4i16.p0v4i16(<4 x i16>* %lsr.iv17.d, i32 2, <4 x i1> %tmp8, <4 x i16> undef)
%sext.load.d = sext <4 x i16> %wide.masked.load.d to <4 x i32>
%tmp12 = mul nsw <4 x i32> %tmp11, %tmp10
%mul.2 = mul nsw <4 x i32> %sext.load.c, %sext.load.d
%tmp13 = add <4 x i32> %tmp12, %mul.2
%acc = add <4 x i32> %tmp13, %vec.phi
%tmp14 = select <4 x i1> %tmp8, <4 x i32> %acc, <4 x i32> %vec.phi
%scevgep = getelementptr i16, i16* %lsr.iv, i32 4
%scevgep19 = getelementptr i16, i16* %lsr.iv18, i32 4
%scevgep.c = getelementptr i16, i16* %lsr.iv.c, i32 4
%scevgep.d = getelementptr i16, i16* %lsr.iv.d, i32 4
%tmp15 = call i32 @llvm.loop.decrement.reg.i32.i32.i32(i32 %lsr.iv1, i32 1)
%tmp16 = icmp ne i32 %tmp15, 0
%lsr.iv.next = add nsw i32 %lsr.iv1, -1
br i1 %tmp16, label %vector.body, label %middle.block
middle.block: ; preds = %vector.body
%tmp17 = call i32 @llvm.vector.reduce.add.v4i32(<4 x i32> %tmp14)
br label %for.cond.cleanup
for.cond.cleanup: ; preds = %middle.block, %entry
%res.0.lcssa = phi i32 [ 0, %entry ], [ %tmp17, %middle.block ]
ret i32 %res.0.lcssa
}
declare <4 x i16> @llvm.masked.load.v4i16.p0v4i16(<4 x i16>*, i32 immarg, <4 x i1>, <4 x i16>) #1
declare i32 @llvm.vector.reduce.add.v4i32(<4 x i32>) #2
declare i32 @llvm.start.loop.iterations.i32(i32) #3
declare i32 @llvm.loop.decrement.reg.i32.i32.i32(i32, i32) #3
declare <4 x i1> @llvm.arm.mve.vctp32(i32) #4
...
---
name: vpsel_after_vpt
alignment: 2
exposesReturnsTwice: false
legalized: false
regBankSelected: false
selected: false
failedISel: false
tracksRegLiveness: true
hasWinCFI: false
registers: []
liveins:
- { reg: '$r0', virtual-reg: '' }
- { reg: '$r1', virtual-reg: '' }
- { reg: '$r2', virtual-reg: '' }
- { reg: '$r3', virtual-reg: '' }
frameInfo:
isFrameAddressTaken: false
isReturnAddressTaken: false
hasStackMap: false
hasPatchPoint: false
stackSize: 16
offsetAdjustment: 0
maxAlignment: 4
adjustsStack: false
hasCalls: false
stackProtector: ''
maxCallFrameSize: 0
cvBytesOfCalleeSavedRegisters: 0
hasOpaqueSPAdjustment: false
hasVAStart: false
hasMustTailInVarArgFunc: false
localFrameSize: 0
savePoint: ''
restorePoint: ''
fixedStack:
- { id: 0, type: default, offset: 0, size: 4, alignment: 8, stack-id: default,
isImmutable: true, isAliased: false, callee-saved-register: '', callee-saved-restored: true,
debug-info-variable: '', debug-info-expression: '', debug-info-location: '' }
stack:
- { id: 0, name: '', type: spill-slot, offset: -4, size: 4, alignment: 4,
stack-id: default, callee-saved-register: '$lr', callee-saved-restored: false,
debug-info-variable: '', debug-info-expression: '', debug-info-location: '' }
- { id: 1, name: '', type: spill-slot, offset: -8, size: 4, alignment: 4,
stack-id: default, callee-saved-register: '$r7', callee-saved-restored: true,
debug-info-variable: '', debug-info-expression: '', debug-info-location: '' }
- { id: 2, name: '', type: spill-slot, offset: -12, size: 4, alignment: 4,
stack-id: default, callee-saved-register: '$r5', callee-saved-restored: true,
debug-info-variable: '', debug-info-expression: '', debug-info-location: '' }
- { id: 3, name: '', type: spill-slot, offset: -16, size: 4, alignment: 4,
stack-id: default, callee-saved-register: '$r4', callee-saved-restored: true,
debug-info-variable: '', debug-info-expression: '', debug-info-location: '' }
callSites: []
constants: []
machineFunctionInfo: {}
body: |
; CHECK-LABEL: name: vpsel_after_vpt
; CHECK: bb.0.entry:
; CHECK: successors: %bb.4(0x30000000), %bb.1(0x50000000)
; CHECK: liveins: $lr, $r0, $r1, $r2, $r3, $r4, $r5, $r7
; CHECK: frame-setup tPUSH 14 /* CC::al */, $noreg, killed $r4, killed $r5, killed $r7, killed $lr, implicit-def $sp, implicit $sp
; CHECK: frame-setup CFI_INSTRUCTION def_cfa_offset 16
; CHECK: frame-setup CFI_INSTRUCTION offset $lr, -4
; CHECK: frame-setup CFI_INSTRUCTION offset $r7, -8
; CHECK: frame-setup CFI_INSTRUCTION offset $r5, -12
; CHECK: frame-setup CFI_INSTRUCTION offset $r4, -16
; CHECK: renamable $r12 = t2LDRi12 $sp, 16, 14 /* CC::al */, $noreg :: (load (s32) from %fixed-stack.0, align 8)
; CHECK: t2CMPri renamable $r12, 0, 14 /* CC::al */, $noreg, implicit-def $cpsr
; CHECK: tBcc %bb.4, 0 /* CC::eq */, killed $cpsr
; CHECK: bb.1.vector.ph:
; CHECK: successors: %bb.2(0x80000000)
; CHECK: liveins: $r0, $r1, $r2, $r3, $r12
; CHECK: renamable $lr = t2ADDri renamable $r12, 3, 14 /* CC::al */, $noreg, $noreg
; CHECK: renamable $r4, dead $cpsr = tMOVi8 1, 14 /* CC::al */, $noreg
; CHECK: renamable $lr = t2BICri killed renamable $lr, 3, 14 /* CC::al */, $noreg, $noreg
; CHECK: renamable $q0 = MVE_VMOVimmi32 0, 0, $noreg, $noreg, undef renamable $q0
; CHECK: renamable $lr = t2SUBri killed renamable $lr, 4, 14 /* CC::al */, $noreg, $noreg
; CHECK: renamable $r5 = nuw nsw t2ADDrs killed renamable $r4, killed renamable $lr, 19, 14 /* CC::al */, $noreg, $noreg
; CHECK: dead $lr = t2DLS renamable $r5
; CHECK: $r4 = tMOVr killed $r5, 14 /* CC::al */, $noreg
; CHECK: bb.2.vector.body:
; CHECK: successors: %bb.2(0x7c000000), %bb.3(0x04000000)
; CHECK: liveins: $q0, $r0, $r1, $r2, $r3, $r4, $r12
; CHECK: renamable $vpr = MVE_VCTP32 renamable $r12, 0, $noreg, $noreg
; CHECK: MVE_VPST 2, implicit $vpr
; CHECK: renamable $r3, renamable $q1 = MVE_VLDRHS32_post killed renamable $r3, 8, 1, renamable $vpr, $noreg :: (load (s64) from %ir.lsr.iv17.d, align 2)
; CHECK: renamable $r2, renamable $q2 = MVE_VLDRHS32_post killed renamable $r2, 8, 1, renamable $vpr, $noreg :: (load (s64) from %ir.lsr.iv1820.c, align 2)
; CHECK: renamable $r1, renamable $q3 = MVE_VLDRHS32_post killed renamable $r1, 8, 1, renamable $vpr, $noreg :: (load (s64) from %ir.lsr.iv1820, align 2)
; CHECK: renamable $r0, renamable $q4 = MVE_VLDRHS32_post killed renamable $r0, 8, 1, renamable $vpr, $noreg :: (load (s64) from %ir.lsr.iv17, align 2)
; CHECK: renamable $q1 = nsw MVE_VMULi32 killed renamable $q2, killed renamable $q1, 0, $noreg, $noreg, undef renamable $q1
; CHECK: renamable $q2 = nsw MVE_VMULi32 killed renamable $q3, killed renamable $q4, 0, $noreg, $noreg, undef renamable $q2
; CHECK: $lr = tMOVr $r4, 14 /* CC::al */, $noreg
; CHECK: renamable $q1 = MVE_VADDi32 killed renamable $q2, killed renamable $q1, 0, $noreg, $noreg, undef renamable $q1
; CHECK: renamable $r4, dead $cpsr = nsw tSUBi8 killed $r4, 1, 14 /* CC::al */, $noreg
; CHECK: renamable $q1 = MVE_VADDi32 killed renamable $q1, renamable $q0, 0, $noreg, $noreg, undef renamable $q1
; CHECK: renamable $r12 = t2SUBri killed renamable $r12, 4, 14 /* CC::al */, $noreg, $noreg
; CHECK: renamable $q0 = MVE_VPSEL killed renamable $q1, killed renamable $q0, 0, killed renamable $vpr, $noreg
; CHECK: dead $lr = t2LEUpdate killed renamable $lr, %bb.2
; CHECK: bb.3.middle.block:
; CHECK: liveins: $q0
; CHECK: renamable $r0 = MVE_VADDVu32no_acc killed renamable $q0, 0, $noreg, $noreg
; CHECK: tPOP_RET 14 /* CC::al */, $noreg, def $r4, def $r5, def $r7, def $pc, implicit killed $r0
; CHECK: bb.4:
; CHECK: renamable $r0, dead $cpsr = tMOVi8 0, 14 /* CC::al */, $noreg
; CHECK: tPOP_RET 14 /* CC::al */, $noreg, def $r4, def $r5, def $r7, def $pc, implicit killed $r0
bb.0.entry:
successors: %bb.4(0x30000000), %bb.1(0x50000000)
liveins: $r0, $r1, $r2, $r3, $r4, $r5, $r7, $lr
frame-setup tPUSH 14, $noreg, killed $r4, killed $r5, killed $r7, killed $lr, implicit-def $sp, implicit $sp
frame-setup CFI_INSTRUCTION def_cfa_offset 16
frame-setup CFI_INSTRUCTION offset $lr, -4
frame-setup CFI_INSTRUCTION offset $r7, -8
frame-setup CFI_INSTRUCTION offset $r5, -12
frame-setup CFI_INSTRUCTION offset $r4, -16
renamable $r12 = t2LDRi12 $sp, 16, 14, $noreg :: (load (s32) from %fixed-stack.0, align 8)
t2CMPri renamable $r12, 0, 14, $noreg, implicit-def $cpsr
tBcc %bb.4, 0, killed $cpsr
bb.1.vector.ph:
successors: %bb.2(0x80000000)
liveins: $r0, $r1, $r2, $r3, $r12
renamable $lr = t2ADDri renamable $r12, 3, 14, $noreg, $noreg
renamable $r4, dead $cpsr = tMOVi8 1, 14, $noreg
renamable $lr = t2BICri killed renamable $lr, 3, 14, $noreg, $noreg
renamable $q0 = MVE_VMOVimmi32 0, 0, $noreg, $noreg, undef renamable $q0
renamable $lr = t2SUBri killed renamable $lr, 4, 14, $noreg, $noreg
renamable $r5 = nuw nsw t2ADDrs killed renamable $r4, killed renamable $lr, 19, 14, $noreg, $noreg
$lr = t2DoLoopStart renamable $r5
$r4 = tMOVr killed $r5, 14, $noreg
bb.2.vector.body:
successors: %bb.2(0x7c000000), %bb.3(0x04000000)
liveins: $q0, $r0, $r1, $r2, $r3, $r4, $r12
renamable $vpr = MVE_VCTP32 renamable $r12, 0, $noreg, $noreg
MVE_VPST 2, implicit $vpr
renamable $r3, renamable $q1 = MVE_VLDRHS32_post killed renamable $r3, 8, 1, renamable $vpr, $noreg :: (load (s64) from %ir.lsr.iv17.d, align 2)
renamable $r2, renamable $q2 = MVE_VLDRHS32_post killed renamable $r2, 8, 1, renamable $vpr, $noreg :: (load (s64) from %ir.lsr.iv1820.c, align 2)
renamable $r1, renamable $q3 = MVE_VLDRHS32_post killed renamable $r1, 8, 1, renamable $vpr, $noreg :: (load (s64) from %ir.lsr.iv1820, align 2)
renamable $r0, renamable $q4 = MVE_VLDRHS32_post killed renamable $r0, 8, 1, renamable $vpr, $noreg :: (load (s64) from %ir.lsr.iv17, align 2)
renamable $q1 = nsw MVE_VMULi32 killed renamable $q2, killed renamable $q1, 0, $noreg, $noreg, undef renamable $q1
renamable $q2 = nsw MVE_VMULi32 killed renamable $q3, killed renamable $q4, 0, $noreg, $noreg, undef renamable $q2
$lr = tMOVr $r4, 14, $noreg
renamable $q1 = MVE_VADDi32 killed renamable $q2, killed renamable $q1, 0, $noreg, $noreg, undef renamable $q1
renamable $r4, dead $cpsr = nsw tSUBi8 killed $r4, 1, 14, $noreg
renamable $q1 = MVE_VADDi32 killed renamable $q1, renamable $q0, 0, $noreg, $noreg, undef renamable $q1
renamable $r12 = t2SUBri killed renamable $r12, 4, 14, $noreg, $noreg
renamable $q0 = MVE_VPSEL killed renamable $q1, killed renamable $q0, 0, killed renamable $vpr, $noreg
renamable $lr = t2LoopDec killed renamable $lr, 1
t2LoopEnd killed renamable $lr, %bb.2, implicit-def dead $cpsr
tB %bb.3, 14, $noreg
bb.3.middle.block:
liveins: $q0
renamable $r0 = MVE_VADDVu32no_acc killed renamable $q0, 0, $noreg, $noreg
tPOP_RET 14, $noreg, def $r4, def $r5, def $r7, def $pc, implicit killed $r0
bb.4:
renamable $r0, dead $cpsr = tMOVi8 0, 14, $noreg
tPOP_RET 14, $noreg, def $r4, def $r5, def $r7, def $pc, implicit killed $r0
...
Reference in New Issue View Git Blame Copy Permalink