Skip to content

[AMDGPU][NFC] Check more autogenerated llc tests for COV5 #75219

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Dec 15, 2023
Merged

[AMDGPU][NFC] Check more autogenerated llc tests for COV5 #75219

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
45 changes: 24 additions & 21 deletions llvm/test/CodeGen/AMDGPU/indirect-addressing-term.ll
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -11,10 +11,10 @@ define amdgpu_kernel void @extract_w_offset_vgpr(ptr addrspace(1) %out) {
; GCN-LABEL: name: extract_w_offset_vgpr
; GCN: bb.0.entry:
; GCN-NEXT: successors: %bb.1(0x80000000)
; GCN-NEXT: liveins: $vgpr0, $sgpr4_sgpr5
; GCN-NEXT: liveins: $vgpr0, $sgpr2_sgpr3
; GCN-NEXT: {{ $}}
; GCN-NEXT: [[COPY:%[0-9]+]]:vgpr_32(s32) = COPY killed $vgpr0
; GCN-NEXT: renamable $sgpr0_sgpr1 = S_LOAD_DWORDX2_IMM killed renamable $sgpr4_sgpr5, 36, 0 :: (dereferenceable invariant load (s64) from %ir.out.kernarg.offset, align 4, addrspace 4)
; GCN-NEXT: renamable $sgpr0_sgpr1 = S_LOAD_DWORDX2_IMM killed renamable $sgpr2_sgpr3, 36, 0 :: (dereferenceable invariant load (s64) from %ir.out.kernarg.offset, align 4, addrspace 4)
; GCN-NEXT: renamable $sgpr6 = COPY renamable $sgpr1
; GCN-NEXT: renamable $sgpr0 = COPY renamable $sgpr0, implicit killed $sgpr0_sgpr1
; GCN-NEXT: renamable $sgpr4 = S_MOV_B32 61440
Expand Down Expand Up @@ -56,22 +56,22 @@ define amdgpu_kernel void @extract_w_offset_vgpr(ptr addrspace(1) %out) {
; GCN-NEXT: [[COPY14:%[0-9]+]]:vgpr_32 = COPY killed renamable $sgpr2
; GCN-NEXT: [[COPY15:%[0-9]+]]:vgpr_32 = COPY killed renamable $sgpr1
; GCN-NEXT: [[COPY16:%[0-9]+]]:vgpr_32 = COPY killed renamable $sgpr0
; GCN-NEXT: undef %35.sub0:vreg_512 = COPY [[COPY1]]
; GCN-NEXT: %35.sub1:vreg_512 = COPY [[COPY2]]
; GCN-NEXT: %35.sub2:vreg_512 = COPY [[COPY3]]
; GCN-NEXT: %35.sub3:vreg_512 = COPY [[COPY4]]
; GCN-NEXT: %35.sub4:vreg_512 = COPY [[COPY5]]
; GCN-NEXT: %35.sub5:vreg_512 = COPY [[COPY6]]
; GCN-NEXT: %35.sub6:vreg_512 = COPY [[COPY7]]
; GCN-NEXT: %35.sub7:vreg_512 = COPY [[COPY8]]
; GCN-NEXT: %35.sub8:vreg_512 = COPY [[COPY9]]
; GCN-NEXT: %35.sub9:vreg_512 = COPY [[COPY10]]
; GCN-NEXT: %35.sub10:vreg_512 = COPY [[COPY11]]
; GCN-NEXT: %35.sub11:vreg_512 = COPY [[COPY12]]
; GCN-NEXT: %35.sub12:vreg_512 = COPY [[COPY13]]
; GCN-NEXT: %35.sub13:vreg_512 = COPY [[COPY14]]
; GCN-NEXT: %35.sub14:vreg_512 = COPY [[COPY15]]
; GCN-NEXT: %35.sub15:vreg_512 = COPY [[COPY16]]
; GCN-NEXT: undef [[COPY17:%[0-9]+]].sub0:vreg_512 = COPY [[COPY1]]
; GCN-NEXT: [[COPY17:%[0-9]+]].sub1:vreg_512 = COPY [[COPY2]]
; GCN-NEXT: [[COPY17:%[0-9]+]].sub2:vreg_512 = COPY [[COPY3]]
; GCN-NEXT: [[COPY17:%[0-9]+]].sub3:vreg_512 = COPY [[COPY4]]
; GCN-NEXT: [[COPY17:%[0-9]+]].sub4:vreg_512 = COPY [[COPY5]]
; GCN-NEXT: [[COPY17:%[0-9]+]].sub5:vreg_512 = COPY [[COPY6]]
; GCN-NEXT: [[COPY17:%[0-9]+]].sub6:vreg_512 = COPY [[COPY7]]
; GCN-NEXT: [[COPY17:%[0-9]+]].sub7:vreg_512 = COPY [[COPY8]]
; GCN-NEXT: [[COPY17:%[0-9]+]].sub8:vreg_512 = COPY [[COPY9]]
; GCN-NEXT: [[COPY17:%[0-9]+]].sub9:vreg_512 = COPY [[COPY10]]
; GCN-NEXT: [[COPY17:%[0-9]+]].sub10:vreg_512 = COPY [[COPY11]]
; GCN-NEXT: [[COPY17:%[0-9]+]].sub11:vreg_512 = COPY [[COPY12]]
; GCN-NEXT: [[COPY17:%[0-9]+]].sub12:vreg_512 = COPY [[COPY13]]
; GCN-NEXT: [[COPY17:%[0-9]+]].sub13:vreg_512 = COPY [[COPY14]]
; GCN-NEXT: [[COPY17:%[0-9]+]].sub14:vreg_512 = COPY [[COPY15]]
; GCN-NEXT: [[COPY17:%[0-9]+]].sub15:vreg_512 = COPY [[COPY16]]
; GCN-NEXT: renamable $sgpr0_sgpr1 = S_MOV_B64 $exec
; GCN-NEXT: SI_SPILL_S64_SAVE killed $sgpr0_sgpr1, %stack.0, implicit $exec, implicit $sgpr32 :: (store (s64) into %stack.0, align 4, addrspace 5)
; GCN-NEXT: [[DEF:%[0-9]+]]:vgpr_32 = IMPLICIT_DEF
Expand All @@ -81,12 +81,12 @@ define amdgpu_kernel void @extract_w_offset_vgpr(ptr addrspace(1) %out) {
; GCN-NEXT: successors: %bb.1(0x40000000), %bb.3(0x40000000)
; GCN-NEXT: {{ $}}
; GCN-NEXT: $sgpr0_sgpr1 = SI_SPILL_S64_RESTORE %stack.2, implicit $exec, implicit $sgpr32 :: (load (s64) from %stack.2, align 4, addrspace 5)
; GCN-NEXT: dead [[COPY17:%[0-9]+]]:vgpr_32 = COPY [[DEF]]
; GCN-NEXT: dead [[COPY18:%[0-9]+]]:vgpr_32 = COPY [[DEF]]
; GCN-NEXT: renamable $sgpr2 = V_READFIRSTLANE_B32 [[COPY]](s32), implicit $exec
; GCN-NEXT: renamable $sgpr0_sgpr1 = V_CMP_EQ_U32_e64 $sgpr2, [[COPY]](s32), implicit $exec
; GCN-NEXT: renamable $sgpr0_sgpr1 = S_AND_SAVEEXEC_B64 killed renamable $sgpr0_sgpr1, implicit-def $exec, implicit-def dead $scc, implicit $exec
; GCN-NEXT: [[V_INDIRECT_REG_READ_GPR_IDX_B32_V16_:%[0-9]+]]:vgpr_32 = V_INDIRECT_REG_READ_GPR_IDX_B32_V16 %35, killed $sgpr2, 11, implicit-def $m0, implicit $m0, implicit $exec
; GCN-NEXT: [[COPY18:%[0-9]+]]:vgpr_32 = COPY [[V_INDIRECT_REG_READ_GPR_IDX_B32_V16_]]
; GCN-NEXT: [[V_INDIRECT_REG_READ_GPR_IDX_B32_V16_:%[0-9]+]]:vgpr_32 = V_INDIRECT_REG_READ_GPR_IDX_B32_V16 [[COPY17]], killed $sgpr2, 11, implicit-def $m0, implicit $m0, implicit $exec
; GCN-NEXT: [[DEF:%[0-9]+]]:vgpr_32 = COPY [[V_INDIRECT_REG_READ_GPR_IDX_B32_V16_]]
; GCN-NEXT: renamable $sgpr2_sgpr3 = COPY renamable $sgpr0_sgpr1
; GCN-NEXT: SI_SPILL_S64_SAVE killed $sgpr2_sgpr3, %stack.2, implicit $exec, implicit $sgpr32 :: (store (s64) into %stack.2, align 4, addrspace 5)
; GCN-NEXT: $exec = S_XOR_B64_term $exec, killed renamable $sgpr0_sgpr1, implicit-def dead $scc
Expand All @@ -109,3 +109,6 @@ entry:
store i32 %value, ptr addrspace(1) %out
ret void
}

!llvm.module.flags = !{!0}
!0 = !{i32 1, !"amdgpu_code_object_version", i32 500}
113 changes: 58 additions & 55 deletions llvm/test/CodeGen/AMDGPU/insert-delay-alu-bug.ll
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -52,44 +52,44 @@ define <2 x i64> @f1() #0 {
define amdgpu_kernel void @f2(i32 %arg, i32 %arg1, i32 %arg2, i1 %arg3, i32 %arg4, i1 %arg5, ptr %arg6, i32 %arg7, i32 %arg8, i32 %arg9, i32 %arg10, i1 %arg11) {
; GFX11-LABEL: f2:
; GFX11: ; %bb.0: ; %bb
; GFX11-NEXT: s_mov_b64 s[16:17], s[4:5]
; GFX11-NEXT: s_load_b32 s21, s[2:3], 0x24
; GFX11-NEXT: v_mov_b32_e32 v31, v0
; GFX11-NEXT: s_load_b32 s24, s[16:17], 0x24
; GFX11-NEXT: s_mov_b32 s12, s13
; GFX11-NEXT: s_mov_b64 s[10:11], s[6:7]
; GFX11-NEXT: s_mov_b64 s[6:7], s[2:3]
; GFX11-NEXT: v_and_b32_e32 v0, 0x3ff, v31
; GFX11-NEXT: s_mov_b64 s[10:11], s[4:5]
; GFX11-NEXT: s_mov_b64 s[4:5], s[0:1]
; GFX11-NEXT: s_mov_b32 s3, 0
; GFX11-NEXT: s_mov_b32 s6, 0
; GFX11-NEXT: v_and_b32_e32 v0, 0x3ff, v31
; GFX11-NEXT: s_mov_b32 s0, -1
; GFX11-NEXT: s_mov_b32 s18, exec_lo
; GFX11-NEXT: s_mov_b32 s20, exec_lo
; GFX11-NEXT: s_mov_b32 s32, 0
; GFX11-NEXT: s_waitcnt lgkmcnt(0)
; GFX11-NEXT: v_mul_lo_u32 v0, s24, v0
; GFX11-NEXT: v_mul_lo_u32 v0, s21, v0
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1)
; GFX11-NEXT: v_cmpx_eq_u32_e32 0, v0
; GFX11-NEXT: s_cbranch_execz .LBB2_13
; GFX11-NEXT: ; %bb.1: ; %bb14
; GFX11-NEXT: s_load_b128 s[20:23], s[16:17], 0x2c
; GFX11-NEXT: s_mov_b32 s19, 0
; GFX11-NEXT: s_load_b128 s[16:19], s[2:3], 0x2c
; GFX11-NEXT: s_waitcnt lgkmcnt(0)
; GFX11-NEXT: s_bitcmp1_b32 s21, 0
; GFX11-NEXT: s_cselect_b32 s25, -1, 0
; GFX11-NEXT: s_bitcmp0_b32 s21, 0
; GFX11-NEXT: s_bitcmp1_b32 s17, 0
; GFX11-NEXT: s_cselect_b32 s22, -1, 0
; GFX11-NEXT: s_bitcmp0_b32 s17, 0
; GFX11-NEXT: s_mov_b32 s17, 0
; GFX11-NEXT: s_cbranch_scc0 .LBB2_3
; GFX11-NEXT: ; %bb.2: ; %bb15
; GFX11-NEXT: s_add_u32 s8, s16, 0x58
; GFX11-NEXT: s_addc_u32 s9, s17, 0
; GFX11-NEXT: s_add_u32 s8, s2, 0x58
; GFX11-NEXT: s_addc_u32 s9, s3, 0
; GFX11-NEXT: s_getpc_b64 s[0:1]
; GFX11-NEXT: s_add_u32 s0, s0, f0@gotpcrel32@lo+4
; GFX11-NEXT: s_addc_u32 s1, s1, f0@gotpcrel32@hi+12
; GFX11-NEXT: s_mov_b32 s13, s14
; GFX11-NEXT: s_load_b64 s[0:1], s[0:1], 0x0
; GFX11-NEXT: s_mov_b32 s3, s14
; GFX11-NEXT: s_mov_b32 s23, s14
; GFX11-NEXT: s_mov_b32 s14, s15
; GFX11-NEXT: s_mov_b64 s[6:7], s[2:3]
; GFX11-NEXT: s_waitcnt lgkmcnt(0)
; GFX11-NEXT: s_swappc_b64 s[30:31], s[0:1]
; GFX11-NEXT: s_mov_b32 s14, s3
; GFX11-NEXT: s_mov_b32 s14, s23
; GFX11-NEXT: s_mov_b64 s[2:3], s[6:7]
; GFX11-NEXT: s_mov_b32 s1, -1
; GFX11-NEXT: s_cbranch_execz .LBB2_4
; GFX11-NEXT: s_branch .LBB2_12
Expand All @@ -98,66 +98,66 @@ define amdgpu_kernel void @f2(i32 %arg, i32 %arg1, i32 %arg2, i1 %arg3, i32 %arg
; GFX11-NEXT: s_and_not1_b32 vcc_lo, exec_lo, s0
; GFX11-NEXT: s_cbranch_vccnz .LBB2_12
; GFX11-NEXT: .LBB2_4: ; %bb16
; GFX11-NEXT: s_load_b32 s2, s[16:17], 0x54
; GFX11-NEXT: s_bitcmp1_b32 s23, 0
; GFX11-NEXT: s_load_b32 s6, s[2:3], 0x54
; GFX11-NEXT: s_bitcmp1_b32 s19, 0
; GFX11-NEXT: s_cselect_b32 s0, -1, 0
; GFX11-NEXT: s_and_b32 s3, s23, 1
; GFX11-NEXT: s_and_b32 s7, s19, 1
; GFX11-NEXT: s_waitcnt lgkmcnt(0)
; GFX11-NEXT: s_bitcmp1_b32 s2, 0
; GFX11-NEXT: s_mov_b32 s2, -1
; GFX11-NEXT: s_bitcmp1_b32 s6, 0
; GFX11-NEXT: s_mov_b32 s6, -1
; GFX11-NEXT: s_cselect_b32 s8, -1, 0
; GFX11-NEXT: s_cmp_eq_u32 s3, 0
; GFX11-NEXT: s_cmp_eq_u32 s7, 0
; GFX11-NEXT: s_cbranch_scc0 .LBB2_8
; GFX11-NEXT: ; %bb.5: ; %bb18.preheader
; GFX11-NEXT: s_load_b128 s[28:31], s[16:17], 0x44
; GFX11-NEXT: s_load_b128 s[24:27], s[2:3], 0x44
; GFX11-NEXT: v_mov_b32_e32 v2, 0
; GFX11-NEXT: s_waitcnt lgkmcnt(0)
; GFX11-NEXT: s_mul_hi_u32 s2, s29, s28
; GFX11-NEXT: s_mul_i32 s3, s29, s28
; GFX11-NEXT: s_mul_hi_u32 s6, s25, s24
; GFX11-NEXT: s_mul_i32 s7, s25, s24
; GFX11-NEXT: s_delay_alu instid0(SALU_CYCLE_1) | instskip(SKIP_1) | instid1(VALU_DEP_1)
; GFX11-NEXT: v_alignbit_b32 v0, s2, s3, 1
; GFX11-NEXT: s_mov_b32 s3, 0
; GFX11-NEXT: v_readfirstlane_b32 s2, v0
; GFX11-NEXT: v_cndmask_b32_e64 v0, 0, 1, s25
; GFX11-NEXT: v_alignbit_b32 v0, s6, s7, 1
; GFX11-NEXT: s_mov_b32 s7, 0
; GFX11-NEXT: v_readfirstlane_b32 s6, v0
; GFX11-NEXT: v_cndmask_b32_e64 v0, 0, 1, s22
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(SALU_CYCLE_1)
; GFX11-NEXT: s_or_b32 s2, s2, 1
; GFX11-NEXT: s_lshr_b32 s2, s2, s30
; GFX11-NEXT: s_or_b32 s6, s6, 1
; GFX11-NEXT: s_lshr_b32 s6, s6, s26
; GFX11-NEXT: s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(SALU_CYCLE_1)
; GFX11-NEXT: s_mul_i32 s2, s2, s22
; GFX11-NEXT: s_mul_i32 s2, s2, s20
; GFX11-NEXT: s_mul_i32 s6, s6, s18
; GFX11-NEXT: s_mul_i32 s6, s6, s16
; GFX11-NEXT: s_delay_alu instid0(SALU_CYCLE_1) | instskip(NEXT) | instid1(SALU_CYCLE_1)
; GFX11-NEXT: s_or_b32 s2, s24, s2
; GFX11-NEXT: s_lshl_b64 s[20:21], s[2:3], 1
; GFX11-NEXT: global_load_u16 v1, v2, s[20:21]
; GFX11-NEXT: s_or_b32 s6, s21, s6
; GFX11-NEXT: s_lshl_b64 s[18:19], s[6:7], 1
; GFX11-NEXT: global_load_u16 v1, v2, s[18:19]
; GFX11-NEXT: s_waitcnt vmcnt(0)
; GFX11-NEXT: v_cmp_eq_u16_e32 vcc_lo, 0, v1
; GFX11-NEXT: v_cndmask_b32_e64 v1, 0, 1, vcc_lo
; GFX11-NEXT: .p2align 6
; GFX11-NEXT: .LBB2_6: ; %bb18
; GFX11-NEXT: ; =>This Inner Loop Header: Depth=1
; GFX11-NEXT: v_cmp_ne_u16_e64 s2, s3, 0
; GFX11-NEXT: v_cmp_ne_u16_e64 s6, s7, 0
; GFX11-NEXT: v_cmp_ne_u16_e32 vcc_lo, 0, v2
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(SKIP_2) | instid1(VALU_DEP_2)
; GFX11-NEXT: v_cndmask_b32_e64 v3, 0, 1, s2
; GFX11-NEXT: v_cndmask_b32_e64 v3, 0, 1, s6
; GFX11-NEXT: v_cndmask_b32_e64 v2, 0, 1, vcc_lo
; GFX11-NEXT: s_and_b32 vcc_lo, s8, vcc_lo
; GFX11-NEXT: v_cndmask_b32_e64 v3, v1, v3, s0
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(SKIP_1) | instid1(VALU_DEP_2)
; GFX11-NEXT: v_cndmask_b32_e32 v2, v0, v2, vcc_lo
; GFX11-NEXT: s_mov_b32 vcc_lo, 0
; GFX11-NEXT: v_readfirstlane_b32 s2, v3
; GFX11-NEXT: v_readfirstlane_b32 s6, v3
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_2)
; GFX11-NEXT: v_and_b32_e32 v2, 1, v2
; GFX11-NEXT: s_bitcmp1_b32 s2, 0
; GFX11-NEXT: s_cselect_b32 s2, 0x100, 0
; GFX11-NEXT: s_bitcmp1_b32 s6, 0
; GFX11-NEXT: s_cselect_b32 s6, 0x100, 0
; GFX11-NEXT: s_delay_alu instid0(SALU_CYCLE_1)
; GFX11-NEXT: s_or_b32 s3, s2, s3
; GFX11-NEXT: s_or_b32 s7, s6, s7
; GFX11-NEXT: s_cbranch_vccz .LBB2_6
; GFX11-NEXT: ; %bb.7: ; %Flow
; GFX11-NEXT: s_mov_b32 s2, 0
; GFX11-NEXT: s_mov_b32 s6, 0
; GFX11-NEXT: .LBB2_8: ; %Flow12
; GFX11-NEXT: s_delay_alu instid0(SALU_CYCLE_1)
; GFX11-NEXT: s_and_b32 vcc_lo, exec_lo, s2
; GFX11-NEXT: s_and_b32 vcc_lo, exec_lo, s6
; GFX11-NEXT: s_cbranch_vccz .LBB2_12
; GFX11-NEXT: ; %bb.9:
; GFX11-NEXT: s_xor_b32 s0, s8, -1
Expand All @@ -167,17 +167,17 @@ define amdgpu_kernel void @f2(i32 %arg, i32 %arg1, i32 %arg2, i1 %arg3, i32 %arg
; GFX11-NEXT: s_and_b32 vcc_lo, exec_lo, s0
; GFX11-NEXT: s_cbranch_vccz .LBB2_10
; GFX11-NEXT: ; %bb.11: ; %Flow6
; GFX11-NEXT: s_mov_b32 s19, -1
; GFX11-NEXT: s_mov_b32 s17, -1
; GFX11-NEXT: .LBB2_12: ; %Flow11
; GFX11-NEXT: s_and_b32 s3, s1, exec_lo
; GFX11-NEXT: s_or_not1_b32 s0, s19, exec_lo
; GFX11-NEXT: s_and_b32 s6, s1, exec_lo
; GFX11-NEXT: s_or_not1_b32 s0, s17, exec_lo
; GFX11-NEXT: .LBB2_13: ; %Flow9
; GFX11-NEXT: s_or_b32 exec_lo, exec_lo, s18
; GFX11-NEXT: s_and_saveexec_b32 s18, s0
; GFX11-NEXT: s_or_b32 exec_lo, exec_lo, s20
; GFX11-NEXT: s_and_saveexec_b32 s7, s0
; GFX11-NEXT: s_cbranch_execz .LBB2_15
; GFX11-NEXT: ; %bb.14: ; %bb43
; GFX11-NEXT: s_add_u32 s8, s16, 0x58
; GFX11-NEXT: s_addc_u32 s9, s17, 0
; GFX11-NEXT: s_add_u32 s8, s2, 0x58
; GFX11-NEXT: s_addc_u32 s9, s3, 0
; GFX11-NEXT: s_getpc_b64 s[0:1]
; GFX11-NEXT: s_add_u32 s0, s0, f0@gotpcrel32@lo+4
; GFX11-NEXT: s_addc_u32 s1, s1, f0@gotpcrel32@hi+12
Expand All @@ -186,10 +186,10 @@ define amdgpu_kernel void @f2(i32 %arg, i32 %arg1, i32 %arg2, i1 %arg3, i32 %arg
; GFX11-NEXT: s_mov_b32 s14, s15
; GFX11-NEXT: s_waitcnt lgkmcnt(0)
; GFX11-NEXT: s_swappc_b64 s[30:31], s[0:1]
; GFX11-NEXT: s_or_b32 s3, s3, exec_lo
; GFX11-NEXT: s_or_b32 s6, s6, exec_lo
; GFX11-NEXT: .LBB2_15: ; %Flow14
; GFX11-NEXT: s_or_b32 exec_lo, exec_lo, s18
; GFX11-NEXT: s_and_saveexec_b32 s0, s3
; GFX11-NEXT: s_or_b32 exec_lo, exec_lo, s7
; GFX11-NEXT: s_and_saveexec_b32 s0, s6
; GFX11-NEXT: ; %bb.16: ; %UnifiedUnreachableBlock
; GFX11-NEXT: ; divergent unreachable
; GFX11-NEXT: ; %bb.17: ; %UnifiedReturnBlock
Expand Down Expand Up @@ -246,3 +246,6 @@ bb43:
}

attributes #0 = { noinline optnone }

!llvm.module.flags = !{!0}
!0 = !{i32 1, !"amdgpu_code_object_version", i32 500}
60 changes: 28 additions & 32 deletions llvm/test/CodeGen/AMDGPU/kernel-vgpr-spill-mubuf-with-voffset.ll
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,30 +9,22 @@ define amdgpu_kernel void @test_kernel(i32 %val) #0 {
; CHECK: ; %bb.0:
; CHECK-NEXT: s_mov_b32 s32, 0x180000
; CHECK-NEXT: s_mov_b32 s33, 0
; CHECK-NEXT: s_add_u32 flat_scratch_lo, s12, s17
; CHECK-NEXT: s_addc_u32 flat_scratch_hi, s13, 0
; CHECK-NEXT: s_add_u32 s0, s0, s17
; CHECK-NEXT: s_add_u32 flat_scratch_lo, s10, s15
; CHECK-NEXT: s_addc_u32 flat_scratch_hi, s11, 0
; CHECK-NEXT: s_add_u32 s0, s0, s15
; CHECK-NEXT: s_addc_u32 s1, s1, 0
; CHECK-NEXT: ; implicit-def: $vgpr3 : SGPR spill to VGPR lane
; CHECK-NEXT: v_writelane_b32 v3, s16, 0
; CHECK-NEXT: s_or_saveexec_b64 s[34:35], -1
; CHECK-NEXT: s_add_i32 s12, s33, 0x100200
; CHECK-NEXT: buffer_store_dword v3, off, s[0:3], s12 ; 4-byte Folded Spill
; CHECK-NEXT: s_mov_b64 exec, s[34:35]
; CHECK-NEXT: s_mov_b32 s13, s15
; CHECK-NEXT: s_mov_b32 s12, s14
; CHECK-NEXT: v_readlane_b32 s14, v3, 0
; CHECK-NEXT: s_mov_b64 s[16:17], s[8:9]
; CHECK-NEXT: s_mov_b64 s[10:11], s[8:9]
; CHECK-NEXT: v_mov_b32_e32 v3, v2
; CHECK-NEXT: v_mov_b32_e32 v2, v1
; CHECK-NEXT: v_mov_b32_e32 v1, v0
; CHECK-NEXT: s_or_saveexec_b64 s[34:35], -1
; CHECK-NEXT: s_add_i32 s8, s33, 0x100200
; CHECK-NEXT: buffer_load_dword v0, off, s[0:3], s8 ; 4-byte Folded Reload
; CHECK-NEXT: s_mov_b64 exec, s[34:35]
; CHECK-NEXT: s_load_dword s8, s[16:17], 0x0
; CHECK-NEXT: s_load_dword s8, s[6:7], 0x0
; CHECK-NEXT: s_waitcnt vmcnt(0) lgkmcnt(0)
; CHECK-NEXT: v_writelane_b32 v0, s8, 1
; CHECK-NEXT: v_writelane_b32 v0, s8, 0
; CHECK-NEXT: s_or_saveexec_b64 s[34:35], -1
; CHECK-NEXT: s_add_i32 s8, s33, 0x100200
; CHECK-NEXT: buffer_store_dword v0, off, s[0:3], s8 ; 4-byte Folded Spill
Expand All @@ -42,28 +34,29 @@ define amdgpu_kernel void @test_kernel(i32 %val) #0 {
; CHECK-NEXT: ;;#ASMEND
; CHECK-NEXT: s_add_i32 s8, s33, 0x100100
; CHECK-NEXT: buffer_store_dword v10, off, s[0:3], s8 ; 4-byte Folded Spill
; CHECK-NEXT: s_mov_b64 s[18:19], 8
; CHECK-NEXT: s_mov_b32 s8, s16
; CHECK-NEXT: s_mov_b32 s9, s17
; CHECK-NEXT: s_mov_b32 s16, s18
; CHECK-NEXT: s_mov_b32 s15, s19
; CHECK-NEXT: s_add_u32 s8, s8, s16
; CHECK-NEXT: s_addc_u32 s15, s9, s15
; CHECK-NEXT: s_mov_b64 s[16:17], 8
; CHECK-NEXT: s_mov_b32 s8, s6
; CHECK-NEXT: s_mov_b32 s6, s7
; CHECK-NEXT: s_mov_b32 s9, s16
; CHECK-NEXT: s_mov_b32 s7, s17
; CHECK-NEXT: s_add_u32 s8, s8, s9
; CHECK-NEXT: s_addc_u32 s6, s6, s7
; CHECK-NEXT: ; kill: def $sgpr8 killed $sgpr8 def $sgpr8_sgpr9
; CHECK-NEXT: s_mov_b32 s9, s15
; CHECK-NEXT: s_mov_b32 s9, s6
; CHECK-NEXT: v_mov_b32_e32 v0, 0x2000
; CHECK-NEXT: ; implicit-def: $sgpr15
; CHECK-NEXT: s_getpc_b64 s[16:17]
; CHECK-NEXT: s_add_u32 s16, s16, device_func@gotpcrel32@lo+4
; CHECK-NEXT: s_addc_u32 s17, s17, device_func@gotpcrel32@hi+12
; CHECK-NEXT: s_load_dwordx2 s[16:17], s[16:17], 0x0
; CHECK-NEXT: ; implicit-def: $sgpr6
; CHECK-NEXT: s_getpc_b64 s[6:7]
; CHECK-NEXT: s_add_u32 s6, s6, device_func@gotpcrel32@lo+4
; CHECK-NEXT: s_addc_u32 s7, s7, device_func@gotpcrel32@hi+12
; CHECK-NEXT: s_load_dwordx2 s[16:17], s[6:7], 0x0
; CHECK-NEXT: s_mov_b64 s[22:23], s[2:3]
; CHECK-NEXT: s_mov_b64 s[20:21], s[0:1]
; CHECK-NEXT: s_mov_b32 s15, 20
; CHECK-NEXT: v_lshlrev_b32_e64 v3, s15, v3
; CHECK-NEXT: s_mov_b32 s15, 10
; CHECK-NEXT: v_lshlrev_b32_e64 v2, s15, v2
; CHECK-NEXT: s_mov_b32 s6, 20
; CHECK-NEXT: v_lshlrev_b32_e64 v3, s6, v3
; CHECK-NEXT: s_mov_b32 s6, 10
; CHECK-NEXT: v_lshlrev_b32_e64 v2, s6, v2
; CHECK-NEXT: v_or3_b32 v31, v1, v2, v3
; CHECK-NEXT: ; implicit-def: $sgpr6_sgpr7
; CHECK-NEXT: ; implicit-def: $sgpr15
; CHECK-NEXT: s_mov_b64 s[0:1], s[20:21]
; CHECK-NEXT: s_mov_b64 s[2:3], s[22:23]
Expand All @@ -76,7 +69,7 @@ define amdgpu_kernel void @test_kernel(i32 %val) #0 {
; CHECK-NEXT: s_add_i32 s4, s33, 0x100100
; CHECK-NEXT: buffer_load_dword v10, off, s[0:3], s4 ; 4-byte Folded Reload
; CHECK-NEXT: s_waitcnt vmcnt(1)
; CHECK-NEXT: v_readlane_b32 s4, v0, 1
; CHECK-NEXT: v_readlane_b32 s4, v0, 0
; CHECK-NEXT: s_mov_b32 s5, 0
; CHECK-NEXT: s_cmp_eq_u32 s4, s5
; CHECK-NEXT: v_mov_b32_e32 v0, 0x4000
Expand Down Expand Up @@ -120,3 +113,6 @@ end:
declare void @device_func(ptr addrspace(5))

attributes #0 = { nounwind "frame-pointer"="all" }

!llvm.module.flags = !{!0}
!0 = !{i32 1, !"amdgpu_code_object_version", i32 500}
Loading