Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Original file line number Diff line number Diff line change
@@ -0,0 +1,313 @@
# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py UTC_ARGS: --version 5
# RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx942 -start-before=greedy,2 -stop-after=virtregrewriter,2 -o - %s | FileCheck %s

--- |

define amdgpu_kernel void @inflate_result_to_agpr__not_mfma() {
ret void
}

; FIXME: Need IR for amdgpu-waves-per-eu
define amdgpu_kernel void @av_split_is_phi_def_crash() #0 {
ret void
}

define amdgpu_kernel void @copy_physreg_to_av_crash() #0 {
ret void
}

attributes #0 = { "amdgpu-wave-limiter"="true" "amdgpu-waves-per-eu"="8,8" }
...

# Inflate pattern, except the defining instruction isn't an MFMA.
---
name: inflate_result_to_agpr__not_mfma
tracksRegLiveness: true
machineFunctionInfo:
isEntryFunction: true
stackPtrOffsetReg: '$sgpr32'
occupancy: 10
sgprForEXECCopy: '$sgpr100_sgpr101'
body: |
; CHECK-LABEL: name: inflate_result_to_agpr__not_mfma
; CHECK: bb.0:
; CHECK-NEXT: successors: %bb.1(0x80000000)
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: S_NOP 0, implicit-def $agpr0
; CHECK-NEXT: renamable $sgpr0 = S_MOV_B32 0
; CHECK-NEXT: renamable $vgpr8 = V_MOV_B32_e32 0, implicit $exec
; CHECK-NEXT: renamable $sgpr1 = COPY renamable $sgpr0
; CHECK-NEXT: renamable $vgpr2_vgpr3 = COPY killed renamable $sgpr0_sgpr1
; CHECK-NEXT: renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
; CHECK-NEXT: dead renamable $vgpr9 = COPY renamable $vgpr8
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: bb.1:
; CHECK-NEXT: successors: %bb.1(0x40000000), %bb.2(0x40000000)
; CHECK-NEXT: liveins: $vcc, $vgpr2_vgpr3
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: renamable $vgpr0_vgpr1 = GLOBAL_LOAD_DWORDX2 undef renamable $vgpr0_vgpr1, 0, 0, implicit $exec :: (load (s64), addrspace 1)
; CHECK-NEXT: S_NOP 0, implicit-def renamable $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15, implicit renamable $vgpr2_vgpr3, implicit renamable $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7_vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
; CHECK-NEXT: S_CBRANCH_VCCNZ %bb.1, implicit $vcc
; CHECK-NEXT: S_BRANCH %bb.2
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: bb.2:
; CHECK-NEXT: liveins: $agpr0_agpr1_agpr2_agpr3_agpr4_agpr5_agpr6_agpr7_agpr8_agpr9_agpr10_agpr11_agpr12_agpr13_agpr14_agpr15:0x00000000FFFFFFFF
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
; CHECK-NEXT: S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
; CHECK-NEXT: renamable $vgpr0 = V_MOV_B32_e32 0, implicit $exec
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr8_agpr9_agpr10_agpr11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr12_agpr13_agpr14_agpr15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR renamable $vgpr0, renamable $agpr0_agpr1_agpr2_agpr3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
; CHECK-NEXT: GLOBAL_STORE_DWORDX4_SADDR killed renamable $vgpr0, killed renamable $agpr4_agpr5_agpr6_agpr7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
; CHECK-NEXT: S_ENDPGM 0
bb.0:
S_NOP 0, implicit-def $agpr0
renamable $sgpr0 = S_MOV_B32 0
undef %0.sub8:vreg_512_align2 = V_MOV_B32_e32 0, implicit $exec
renamable $sgpr1 = COPY renamable $sgpr0
%1:vreg_64_align2 = COPY killed renamable $sgpr0_sgpr1
renamable $vcc = S_AND_B64 $exec, -1, implicit-def dead $scc
%0.sub9:vreg_512_align2 = COPY %0.sub8

bb.1:
liveins: $vcc

undef %4.sub0_sub1:vreg_512_align2 = GLOBAL_LOAD_DWORDX2 undef %3:vreg_64_align2, 0, 0, implicit $exec :: (load (s64), addrspace 1)
S_NOP 0, implicit-def %0:vreg_512_align2, implicit %1, implicit %4
S_CBRANCH_VCCNZ %bb.1, implicit $vcc
S_BRANCH %bb.2

bb.2:
; No VGPRs available for %0
S_NOP 0, implicit-def $vgpr0_vgpr1_vgpr2_vgpr3_vgpr4_vgpr5_vgpr6_vgpr7
S_NOP 0, implicit-def $vgpr8_vgpr9_vgpr10_vgpr11_vgpr12_vgpr13_vgpr14_vgpr15
S_NOP 0, implicit-def $vgpr16_vgpr17_vgpr18_vgpr19_vgpr20_vgpr21_vgpr22_vgpr23
S_NOP 0, implicit-def $vgpr24_vgpr25_vgpr26_vgpr27_vgpr28_vgpr29_vgpr30_vgpr31
S_NOP 0, implicit-def $vgpr32_vgpr33_vgpr34_vgpr35_vgpr36_vgpr37_vgpr38_vgpr39
S_NOP 0, implicit-def $vgpr40_vgpr41_vgpr42_vgpr43_vgpr44_vgpr45_vgpr46_vgpr47
S_NOP 0, implicit-def $vgpr48_vgpr49_vgpr50_vgpr51_vgpr52_vgpr53_vgpr54_vgpr55
S_NOP 0, implicit-def $vgpr56_vgpr57_vgpr58_vgpr59_vgpr60_vgpr61_vgpr62_vgpr63
%2:vgpr_32 = V_MOV_B32_e32 0, implicit $exec
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub8_sub9_sub10_sub11, undef $sgpr0_sgpr1, 32, 0, implicit $exec :: (store (s128), align 32, addrspace 1)
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub12_sub13_sub14_sub15, undef $sgpr0_sgpr1, 48, 0, implicit $exec :: (store (s128), addrspace 1)
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub0_sub1_sub2_sub3, undef $sgpr0_sgpr1, 0, 0, implicit $exec :: (store (s128), align 128, addrspace 1)
GLOBAL_STORE_DWORDX4_SADDR %2, %0.sub4_sub5_sub6_sub7, killed undef $sgpr0_sgpr1, 16, 0, implicit $exec :: (store (s128), addrspace 1)
S_ENDPGM 0

...

---
name: av_split_is_phi_def_crash
tracksRegLiveness: true
machineFunctionInfo:
isEntryFunction: true
waveLimiter: true
hasSpilledSGPRs: true
scratchRSrcReg: '$sgpr68_sgpr69_sgpr70_sgpr71'
stackPtrOffsetReg: '$sgpr32'
occupancy: 8
wwmReservedRegs:
- '$vgpr31'
sgprForEXECCopy: '$sgpr72_sgpr73'
body: |
; CHECK-LABEL: name: av_split_is_phi_def_crash
; CHECK: bb.0:
; CHECK-NEXT: successors: %bb.2(0x40000000), %bb.1(0x40000000)
; CHECK-NEXT: liveins: $vgpr0, $sgpr2_sgpr3, $sgpr8_sgpr9
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: renamable $vgpr8 = IMPLICIT_DEF
; CHECK-NEXT: renamable $vgpr2_vgpr3 = V_MOV_B64_PSEUDO 0, implicit $exec
; CHECK-NEXT: renamable $vgpr0 = V_MOV_B32_e32 0, implicit $exec
; CHECK-NEXT: S_CBRANCH_EXECZ %bb.2, implicit $exec
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: bb.1:
; CHECK-NEXT: successors: %bb.2(0x80000000)
; CHECK-NEXT: liveins: $vgpr8, $vgpr0_vgpr1:0x0000000000000003
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: renamable $vgpr2_vgpr3 = V_MOV_B64_PSEUDO 4607182418800017408, implicit $exec
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: bb.2:
; CHECK-NEXT: successors: %bb.3(0x80000000)
; CHECK-NEXT: liveins: $vgpr8, $vgpr31, $sgpr8_sgpr9, $vgpr0_vgpr1:0x0000000000000003, $vgpr2_vgpr3
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: renamable $vgpr1 = COPY renamable $vgpr8
; CHECK-NEXT: renamable $agpr0_agpr1 = COPY killed renamable $vgpr2_vgpr3
; CHECK-NEXT: renamable $vgpr2 = COPY renamable $vgpr1
; CHECK-NEXT: renamable $vgpr3 = COPY renamable $vgpr8
; CHECK-NEXT: renamable $vgpr6 = COPY renamable $vgpr1
; CHECK-NEXT: dead renamable $vgpr7 = COPY renamable $vgpr8
; CHECK-NEXT: renamable $vgpr6 = COPY renamable $vgpr1
; CHECK-NEXT: renamable $vgpr7 = COPY renamable $vgpr8
; CHECK-NEXT: S_NOP 0
; CHECK-NEXT: S_NOP 0
; CHECK-NEXT: S_NOP 0
; CHECK-NEXT: S_NOP 0
; CHECK-NEXT: S_NOP 0
; CHECK-NEXT: S_NOP 0
; CHECK-NEXT: S_NOP 0
; CHECK-NEXT: S_NOP 0
; CHECK-NEXT: S_NOP 0
; CHECK-NEXT: S_NOP 0
; CHECK-NEXT: S_NOP 0
; CHECK-NEXT: S_NOP 0
; CHECK-NEXT: S_NOP 0
; CHECK-NEXT: S_NOP 0
; CHECK-NEXT: S_NOP 0
; CHECK-NEXT: S_NOP 0
; CHECK-NEXT: S_NOP 0
; CHECK-NEXT: S_NOP 0
; CHECK-NEXT: S_NOP 0
; CHECK-NEXT: S_NOP 0
; CHECK-NEXT: S_NOP 0
; CHECK-NEXT: S_NOP 0
; CHECK-NEXT: S_NOP 0
; CHECK-NEXT: S_NOP 0
; CHECK-NEXT: S_NOP 0
; CHECK-NEXT: S_NOP 0
; CHECK-NEXT: S_NOP 0
; CHECK-NEXT: S_NOP 0
; CHECK-NEXT: renamable $vgpr10_vgpr11 = V_ADD_F64_e64 0, undef $sgpr6_sgpr7, 0, 0, 0, 0, implicit $mode, implicit $exec
; CHECK-NEXT: renamable $vgpr12_vgpr13 = V_ADD_F64_e64 0, undef $sgpr6_sgpr7, 0, 0, 0, 0, implicit $mode, implicit $exec
; CHECK-NEXT: renamable $vgpr14_vgpr15 = V_ADD_F64_e64 0, undef $sgpr6_sgpr7, 0, 0, 0, 0, implicit $mode, implicit $exec
; CHECK-NEXT: renamable $vgpr16_vgpr17 = V_ADD_F64_e64 0, undef $sgpr6_sgpr7, 0, 0, 0, 0, implicit $mode, implicit $exec
; CHECK-NEXT: renamable $vgpr18_vgpr19 = V_ADD_F64_e64 0, undef $sgpr2_sgpr3, 0, 0, 0, 0, implicit $mode, implicit $exec
; CHECK-NEXT: renamable $vgpr20_vgpr21 = V_ADD_F64_e64 0, undef $sgpr28_sgpr29, 0, 0, 0, 0, implicit $mode, implicit $exec
; CHECK-NEXT: renamable $vgpr22_vgpr23 = V_ADD_F64_e64 0, undef $sgpr24_sgpr25, 0, 0, 0, 0, implicit $mode, implicit $exec
; CHECK-NEXT: renamable $vgpr24_vgpr25 = V_ADD_F64_e64 0, undef $sgpr30_sgpr31, 0, 0, 0, 0, implicit $mode, implicit $exec
; CHECK-NEXT: renamable $vgpr26_vgpr27 = V_ADD_F64_e64 0, undef $sgpr26_sgpr27, 0, 0, 0, 0, implicit $mode, implicit $exec
; CHECK-NEXT: renamable $vgpr29 = COPY renamable $vgpr12
; CHECK-NEXT: renamable $vgpr28 = COPY renamable $vgpr12
; CHECK-NEXT: renamable $vgpr4_vgpr5 = V_MOV_B64_PSEUDO 0, implicit $exec
; CHECK-NEXT: renamable $vgpr4_vgpr5 = V_ADD_F64_e64 0, killed $vgpr28_vgpr29, 0, killed $vgpr4_vgpr5, 0, 0, implicit $mode, implicit $exec
; CHECK-NEXT: renamable $vgpr28_vgpr29 = COPY killed renamable $agpr0_agpr1
; CHECK-NEXT: renamable $vgpr4_vgpr5 = V_ADD_F64_e64 0, killed $vgpr4_vgpr5, 0, killed $vgpr28_vgpr29, 0, 0, implicit $mode, implicit $exec
; CHECK-NEXT: dead renamable $vgpr4 = COPY killed renamable $vgpr11
; CHECK-NEXT: dead renamable $vgpr4 = COPY killed renamable $vgpr14
; CHECK-NEXT: dead renamable $vgpr4 = COPY killed renamable $vgpr20
; CHECK-NEXT: dead renamable $vgpr4 = COPY killed renamable $vgpr24
; CHECK-NEXT: dead renamable $vgpr4 = COPY killed renamable $vgpr26
; CHECK-NEXT: renamable $vgpr4 = COPY killed renamable $vgpr6
; CHECK-NEXT: S_NOP 0
; CHECK-NEXT: dead renamable $vgpr4 = COPY killed renamable $vgpr16
; CHECK-NEXT: dead renamable $vgpr4 = COPY killed renamable $vgpr22
; CHECK-NEXT: dead renamable $vgpr4 = COPY killed renamable $vgpr18
; CHECK-NEXT: renamable $vgpr2 = COPY killed renamable $vgpr1
; CHECK-NEXT: dead renamable $vgpr3 = COPY killed renamable $vgpr8
; CHECK-NEXT: dead renamable $vgpr1 = COPY renamable $vgpr0
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: bb.3:
; CHECK-NEXT: S_ENDPGM 0
bb.0:
liveins: $vgpr0, $sgpr8_sgpr9, $sgpr2_sgpr3, $sgpr8_sgpr9

%0:vgpr_32 = IMPLICIT_DEF
%1:vreg_64_align2 = V_MOV_B64_PSEUDO 0, implicit $exec
undef %2.sub0:vreg_64_align2 = V_MOV_B32_e32 0, implicit $exec
S_CBRANCH_EXECZ %bb.2, implicit $exec

bb.1:
%1:vreg_64_align2 = V_MOV_B64_PSEUDO 4607182418800017408, implicit $exec

bb.2:
liveins: $vgpr31, $sgpr8_sgpr9

%3:vgpr_32 = COPY %0
%4:vgpr_32 = COPY %0
undef %5.sub0:vreg_64_align2 = COPY %3
%5.sub1:vreg_64_align2 = COPY %4
undef %6.sub0:vreg_64_align2 = COPY %3
dead %6.sub1:vreg_64_align2 = COPY %4
undef %7.sub0:vreg_64_align2 = COPY %3
%7.sub1:vreg_64_align2 = COPY %4
S_NOP 0
S_NOP 0
S_NOP 0
S_NOP 0
S_NOP 0
S_NOP 0
S_NOP 0
S_NOP 0
S_NOP 0
S_NOP 0
S_NOP 0
S_NOP 0
S_NOP 0
S_NOP 0
S_NOP 0
S_NOP 0
S_NOP 0
S_NOP 0
S_NOP 0
S_NOP 0
S_NOP 0
S_NOP 0
S_NOP 0
S_NOP 0
S_NOP 0
S_NOP 0
S_NOP 0
S_NOP 0
%8:vreg_64_align2 = V_ADD_F64_e64 0, undef $sgpr6_sgpr7, 0, 0, 0, 0, implicit $mode, implicit $exec
%9:vreg_64_align2 = V_ADD_F64_e64 0, undef $sgpr6_sgpr7, 0, 0, 0, 0, implicit $mode, implicit $exec
%10:vreg_64_align2 = V_ADD_F64_e64 0, undef $sgpr6_sgpr7, 0, 0, 0, 0, implicit $mode, implicit $exec
%11:vreg_64_align2 = V_ADD_F64_e64 0, undef $sgpr6_sgpr7, 0, 0, 0, 0, implicit $mode, implicit $exec
%12:vreg_64_align2 = V_ADD_F64_e64 0, undef $sgpr2_sgpr3, 0, 0, 0, 0, implicit $mode, implicit $exec
%13:vreg_64_align2 = V_ADD_F64_e64 0, undef $sgpr28_sgpr29, 0, 0, 0, 0, implicit $mode, implicit $exec
%14:vreg_64_align2 = V_ADD_F64_e64 0, undef $sgpr24_sgpr25, 0, 0, 0, 0, implicit $mode, implicit $exec
%15:vreg_64_align2 = V_ADD_F64_e64 0, undef $sgpr30_sgpr31, 0, 0, 0, 0, implicit $mode, implicit $exec
%16:vreg_64_align2 = V_ADD_F64_e64 0, undef $sgpr26_sgpr27, 0, 0, 0, 0, implicit $mode, implicit $exec
undef %17.sub1:vreg_64_align2 = COPY %9.sub0
%17.sub0:vreg_64_align2 = COPY %9.sub0
%18:vreg_64_align2 = V_MOV_B64_PSEUDO 0, implicit $exec
%19:vreg_64_align2 = V_ADD_F64_e64 0, %17, 0, %18, 0, 0, implicit $mode, implicit $exec
%20:vreg_64_align2 = V_ADD_F64_e64 0, %19, 0, %1, 0, 0, implicit $mode, implicit $exec
dead %21:vgpr_32 = COPY %20.sub0
dead %22:vgpr_32 = COPY %8.sub1
dead %23:vgpr_32 = COPY %10.sub0
dead %24:vgpr_32 = COPY %13.sub0
dead %25:vgpr_32 = COPY %15.sub0
dead %26:vgpr_32 = COPY %16.sub0
%27:vgpr_32 = COPY %7.sub0
%28:vgpr_32 = COPY %27
S_NOP 0
dead %29:vgpr_32 = COPY %11.sub0
dead %30:vgpr_32 = COPY %14.sub0
dead %31:vgpr_32 = COPY %12.sub0
%32:vreg_64_align2 = COPY %5
undef %33.sub0:vreg_64_align2 = COPY %3
dead %33.sub1:vreg_64_align2 = COPY %4
dead undef %34.sub1:vreg_64_align2 = COPY %9.sub1
dead %2.sub1:vreg_64_align2 = COPY %2.sub0

bb.3:
S_ENDPGM 0

...

---
name: copy_physreg_to_av_crash
tracksRegLiveness: true
registers:
- { id: 0, class: av_32, preferred-register: '$agpr0' }
machineFunctionInfo:
isEntryFunction: true
stackPtrOffsetReg: '$sgpr32'
occupancy: 10
body: |
bb.0:
liveins: $vgpr0
; CHECK-LABEL: name: copy_physreg_to_av_crash
; CHECK: liveins: $vgpr0
; CHECK-NEXT: {{ $}}
; CHECK-NEXT: renamable $agpr0 = COPY $vgpr0
%0:av_32 = COPY $vgpr0
$agpr0 = COPY %0

...
Loading
Loading