Please use GitHub pull requests for new patches. Avoid migrating existing patches. Phabricator shutdown timeline
Differential D142192 Diff 491241 llvm/test/CodeGen/AMDGPU/GlobalISel/llvm.amdgcn.raw.tbuffer.store.f16.ll
Changeset View
Changeset View
Standalone View
Standalone View
llvm/test/CodeGen/AMDGPU/GlobalISel/llvm.amdgcn.raw.tbuffer.store.f16.ll
Show First 20 Lines • Show All 134 Lines • ▼ Show 20 Lines | define amdgpu_ps void @raw_tbuffer_store_f16__vgpr_rsrc__vgpr_voffset__sgpr_soffset(half %val, <4 x i32> %rsrc, i32 %voffset, i32 inreg %soffset) { | ||||
; UNPACKED-NEXT: [[REG_SEQUENCE:%[0-9]+]]:vreg_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3 | ; UNPACKED-NEXT: [[REG_SEQUENCE:%[0-9]+]]:vreg_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3 | ||||
; UNPACKED-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr5 | ; UNPACKED-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr5 | ||||
; UNPACKED-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr2 | ; UNPACKED-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr2 | ||||
; UNPACKED-NEXT: [[S_MOV_B64_:%[0-9]+]]:sreg_64_xexec = S_MOV_B64 $exec | ; UNPACKED-NEXT: [[S_MOV_B64_:%[0-9]+]]:sreg_64_xexec = S_MOV_B64 $exec | ||||
; UNPACKED-NEXT: {{ $}} | ; UNPACKED-NEXT: {{ $}} | ||||
; UNPACKED-NEXT: bb.2: | ; UNPACKED-NEXT: bb.2: | ||||
; UNPACKED-NEXT: successors: %bb.3(0x80000000) | ; UNPACKED-NEXT: successors: %bb.3(0x80000000) | ||||
; UNPACKED-NEXT: {{ $}} | ; UNPACKED-NEXT: {{ $}} | ||||
; UNPACKED-NEXT: [[COPY7:%[0-9]+]]:vgpr_32 = COPY [[REG_SEQUENCE]].sub0 | ; UNPACKED-NEXT: [[V_READFIRSTLANE_B32_:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY1]], implicit $exec | ||||
; UNPACKED-NEXT: [[COPY8:%[0-9]+]]:vgpr_32 = COPY [[REG_SEQUENCE]].sub1 | ; UNPACKED-NEXT: [[V_READFIRSTLANE_B32_1:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY2]], implicit $exec | ||||
; UNPACKED-NEXT: [[COPY9:%[0-9]+]]:vgpr_32 = COPY [[REG_SEQUENCE]].sub2 | ; UNPACKED-NEXT: [[V_READFIRSTLANE_B32_2:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY3]], implicit $exec | ||||
; UNPACKED-NEXT: [[COPY10:%[0-9]+]]:vgpr_32 = COPY [[REG_SEQUENCE]].sub3 | ; UNPACKED-NEXT: [[V_READFIRSTLANE_B32_3:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY4]], implicit $exec | ||||
; UNPACKED-NEXT: [[V_READFIRSTLANE_B32_:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY7]], implicit $exec | |||||
; UNPACKED-NEXT: [[V_READFIRSTLANE_B32_1:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY8]], implicit $exec | |||||
; UNPACKED-NEXT: [[V_READFIRSTLANE_B32_2:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY9]], implicit $exec | |||||
; UNPACKED-NEXT: [[V_READFIRSTLANE_B32_3:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY10]], implicit $exec | |||||
; UNPACKED-NEXT: [[REG_SEQUENCE1:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[V_READFIRSTLANE_B32_]], %subreg.sub0, [[V_READFIRSTLANE_B32_1]], %subreg.sub1, [[V_READFIRSTLANE_B32_2]], %subreg.sub2, [[V_READFIRSTLANE_B32_3]], %subreg.sub3 | ; UNPACKED-NEXT: [[REG_SEQUENCE1:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[V_READFIRSTLANE_B32_]], %subreg.sub0, [[V_READFIRSTLANE_B32_1]], %subreg.sub1, [[V_READFIRSTLANE_B32_2]], %subreg.sub2, [[V_READFIRSTLANE_B32_3]], %subreg.sub3 | ||||
; UNPACKED-NEXT: [[COPY11:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub0_sub1 | ; UNPACKED-NEXT: [[COPY7:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub0_sub1 | ||||
; UNPACKED-NEXT: [[COPY12:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub2_sub3 | ; UNPACKED-NEXT: [[COPY8:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub2_sub3 | ||||
; UNPACKED-NEXT: [[COPY13:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub0_sub1 | ; UNPACKED-NEXT: [[COPY9:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub0_sub1 | ||||
; UNPACKED-NEXT: [[COPY14:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub2_sub3 | ; UNPACKED-NEXT: [[COPY10:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub2_sub3 | ||||
; UNPACKED-NEXT: [[V_CMP_EQ_U64_e64_:%[0-9]+]]:sreg_64_xexec = V_CMP_EQ_U64_e64 [[COPY13]], [[COPY11]], implicit $exec | ; UNPACKED-NEXT: [[V_CMP_EQ_U64_e64_:%[0-9]+]]:sreg_64_xexec = V_CMP_EQ_U64_e64 [[COPY9]], [[COPY7]], implicit $exec | ||||
; UNPACKED-NEXT: [[V_CMP_EQ_U64_e64_1:%[0-9]+]]:sreg_64_xexec = V_CMP_EQ_U64_e64 [[COPY14]], [[COPY12]], implicit $exec | ; UNPACKED-NEXT: [[V_CMP_EQ_U64_e64_1:%[0-9]+]]:sreg_64_xexec = V_CMP_EQ_U64_e64 [[COPY10]], [[COPY8]], implicit $exec | ||||
; UNPACKED-NEXT: [[S_AND_B64_:%[0-9]+]]:sreg_64_xexec = S_AND_B64 [[V_CMP_EQ_U64_e64_]], [[V_CMP_EQ_U64_e64_1]], implicit-def dead $scc | ; UNPACKED-NEXT: [[S_AND_B64_:%[0-9]+]]:sreg_64_xexec = S_AND_B64 [[V_CMP_EQ_U64_e64_]], [[V_CMP_EQ_U64_e64_1]], implicit-def dead $scc | ||||
; UNPACKED-NEXT: [[S_AND_SAVEEXEC_B64_:%[0-9]+]]:sreg_64_xexec = S_AND_SAVEEXEC_B64 killed [[S_AND_B64_]], implicit-def $exec, implicit-def $scc, implicit $exec | ; UNPACKED-NEXT: [[S_AND_SAVEEXEC_B64_:%[0-9]+]]:sreg_64_xexec = S_AND_SAVEEXEC_B64 killed [[S_AND_B64_]], implicit-def $exec, implicit-def $scc, implicit $exec | ||||
; UNPACKED-NEXT: {{ $}} | ; UNPACKED-NEXT: {{ $}} | ||||
; UNPACKED-NEXT: bb.3: | ; UNPACKED-NEXT: bb.3: | ||||
; UNPACKED-NEXT: successors: %bb.4(0x40000000), %bb.2(0x40000000) | ; UNPACKED-NEXT: successors: %bb.4(0x40000000), %bb.2(0x40000000) | ||||
; UNPACKED-NEXT: {{ $}} | ; UNPACKED-NEXT: {{ $}} | ||||
; UNPACKED-NEXT: TBUFFER_STORE_FORMAT_D16_X_gfx80_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE1]], [[COPY6]], 0, 94, 0, 0, implicit $exec :: (dereferenceable store (s16), align 1, addrspace 7) | ; UNPACKED-NEXT: TBUFFER_STORE_FORMAT_D16_X_gfx80_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE1]], [[COPY6]], 0, 94, 0, 0, implicit $exec :: (dereferenceable store (s16), align 1, addrspace 7) | ||||
; UNPACKED-NEXT: $exec = S_XOR_B64_term $exec, [[S_AND_SAVEEXEC_B64_]], implicit-def $scc | ; UNPACKED-NEXT: $exec = S_XOR_B64_term $exec, [[S_AND_SAVEEXEC_B64_]], implicit-def $scc | ||||
Show All 19 Lines | define amdgpu_ps void @raw_tbuffer_store_f16__vgpr_rsrc__vgpr_voffset__sgpr_soffset(half %val, <4 x i32> %rsrc, i32 %voffset, i32 inreg %soffset) { | ||||
; PACKED-NEXT: [[REG_SEQUENCE:%[0-9]+]]:vreg_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3 | ; PACKED-NEXT: [[REG_SEQUENCE:%[0-9]+]]:vreg_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3 | ||||
; PACKED-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr5 | ; PACKED-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr5 | ||||
; PACKED-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr2 | ; PACKED-NEXT: [[COPY6:%[0-9]+]]:sreg_32 = COPY $sgpr2 | ||||
; PACKED-NEXT: [[S_MOV_B64_:%[0-9]+]]:sreg_64_xexec = S_MOV_B64 $exec | ; PACKED-NEXT: [[S_MOV_B64_:%[0-9]+]]:sreg_64_xexec = S_MOV_B64 $exec | ||||
; PACKED-NEXT: {{ $}} | ; PACKED-NEXT: {{ $}} | ||||
; PACKED-NEXT: bb.2: | ; PACKED-NEXT: bb.2: | ||||
; PACKED-NEXT: successors: %bb.3(0x80000000) | ; PACKED-NEXT: successors: %bb.3(0x80000000) | ||||
; PACKED-NEXT: {{ $}} | ; PACKED-NEXT: {{ $}} | ||||
; PACKED-NEXT: [[COPY7:%[0-9]+]]:vgpr_32 = COPY [[REG_SEQUENCE]].sub0 | ; PACKED-NEXT: [[V_READFIRSTLANE_B32_:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY1]], implicit $exec | ||||
; PACKED-NEXT: [[COPY8:%[0-9]+]]:vgpr_32 = COPY [[REG_SEQUENCE]].sub1 | ; PACKED-NEXT: [[V_READFIRSTLANE_B32_1:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY2]], implicit $exec | ||||
; PACKED-NEXT: [[COPY9:%[0-9]+]]:vgpr_32 = COPY [[REG_SEQUENCE]].sub2 | ; PACKED-NEXT: [[V_READFIRSTLANE_B32_2:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY3]], implicit $exec | ||||
; PACKED-NEXT: [[COPY10:%[0-9]+]]:vgpr_32 = COPY [[REG_SEQUENCE]].sub3 | ; PACKED-NEXT: [[V_READFIRSTLANE_B32_3:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY4]], implicit $exec | ||||
; PACKED-NEXT: [[V_READFIRSTLANE_B32_:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY7]], implicit $exec | |||||
; PACKED-NEXT: [[V_READFIRSTLANE_B32_1:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY8]], implicit $exec | |||||
; PACKED-NEXT: [[V_READFIRSTLANE_B32_2:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY9]], implicit $exec | |||||
; PACKED-NEXT: [[V_READFIRSTLANE_B32_3:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY10]], implicit $exec | |||||
; PACKED-NEXT: [[REG_SEQUENCE1:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[V_READFIRSTLANE_B32_]], %subreg.sub0, [[V_READFIRSTLANE_B32_1]], %subreg.sub1, [[V_READFIRSTLANE_B32_2]], %subreg.sub2, [[V_READFIRSTLANE_B32_3]], %subreg.sub3 | ; PACKED-NEXT: [[REG_SEQUENCE1:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[V_READFIRSTLANE_B32_]], %subreg.sub0, [[V_READFIRSTLANE_B32_1]], %subreg.sub1, [[V_READFIRSTLANE_B32_2]], %subreg.sub2, [[V_READFIRSTLANE_B32_3]], %subreg.sub3 | ||||
; PACKED-NEXT: [[COPY11:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub0_sub1 | ; PACKED-NEXT: [[COPY7:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub0_sub1 | ||||
; PACKED-NEXT: [[COPY12:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub2_sub3 | ; PACKED-NEXT: [[COPY8:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub2_sub3 | ||||
; PACKED-NEXT: [[COPY13:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub0_sub1 | ; PACKED-NEXT: [[COPY9:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub0_sub1 | ||||
; PACKED-NEXT: [[COPY14:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub2_sub3 | ; PACKED-NEXT: [[COPY10:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub2_sub3 | ||||
; PACKED-NEXT: [[V_CMP_EQ_U64_e64_:%[0-9]+]]:sreg_64_xexec = V_CMP_EQ_U64_e64 [[COPY13]], [[COPY11]], implicit $exec | ; PACKED-NEXT: [[V_CMP_EQ_U64_e64_:%[0-9]+]]:sreg_64_xexec = V_CMP_EQ_U64_e64 [[COPY9]], [[COPY7]], implicit $exec | ||||
; PACKED-NEXT: [[V_CMP_EQ_U64_e64_1:%[0-9]+]]:sreg_64_xexec = V_CMP_EQ_U64_e64 [[COPY14]], [[COPY12]], implicit $exec | ; PACKED-NEXT: [[V_CMP_EQ_U64_e64_1:%[0-9]+]]:sreg_64_xexec = V_CMP_EQ_U64_e64 [[COPY10]], [[COPY8]], implicit $exec | ||||
; PACKED-NEXT: [[S_AND_B64_:%[0-9]+]]:sreg_64_xexec = S_AND_B64 [[V_CMP_EQ_U64_e64_]], [[V_CMP_EQ_U64_e64_1]], implicit-def dead $scc | ; PACKED-NEXT: [[S_AND_B64_:%[0-9]+]]:sreg_64_xexec = S_AND_B64 [[V_CMP_EQ_U64_e64_]], [[V_CMP_EQ_U64_e64_1]], implicit-def dead $scc | ||||
; PACKED-NEXT: [[S_AND_SAVEEXEC_B64_:%[0-9]+]]:sreg_64_xexec = S_AND_SAVEEXEC_B64 killed [[S_AND_B64_]], implicit-def $exec, implicit-def $scc, implicit $exec | ; PACKED-NEXT: [[S_AND_SAVEEXEC_B64_:%[0-9]+]]:sreg_64_xexec = S_AND_SAVEEXEC_B64 killed [[S_AND_B64_]], implicit-def $exec, implicit-def $scc, implicit $exec | ||||
; PACKED-NEXT: {{ $}} | ; PACKED-NEXT: {{ $}} | ||||
; PACKED-NEXT: bb.3: | ; PACKED-NEXT: bb.3: | ||||
; PACKED-NEXT: successors: %bb.4(0x40000000), %bb.2(0x40000000) | ; PACKED-NEXT: successors: %bb.4(0x40000000), %bb.2(0x40000000) | ||||
; PACKED-NEXT: {{ $}} | ; PACKED-NEXT: {{ $}} | ||||
; PACKED-NEXT: TBUFFER_STORE_FORMAT_D16_X_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE1]], [[COPY6]], 0, 94, 0, 0, implicit $exec :: (dereferenceable store (s16), align 1, addrspace 7) | ; PACKED-NEXT: TBUFFER_STORE_FORMAT_D16_X_OFFEN_exact [[COPY]], [[COPY5]], [[REG_SEQUENCE1]], [[COPY6]], 0, 94, 0, 0, implicit $exec :: (dereferenceable store (s16), align 1, addrspace 7) | ||||
; PACKED-NEXT: $exec = S_XOR_B64_term $exec, [[S_AND_SAVEEXEC_B64_]], implicit-def $scc | ; PACKED-NEXT: $exec = S_XOR_B64_term $exec, [[S_AND_SAVEEXEC_B64_]], implicit-def $scc | ||||
Show All 25 Lines | define amdgpu_ps void @raw_tbuffer_store_f16__vgpr_rsrc__vgpr_voffset__vgpr_soffset(half %val, <4 x i32> %rsrc, i32 %voffset, i32 %soffset) { | ||||
; UNPACKED-NEXT: [[REG_SEQUENCE:%[0-9]+]]:vreg_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3 | ; UNPACKED-NEXT: [[REG_SEQUENCE:%[0-9]+]]:vreg_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3 | ||||
; UNPACKED-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr5 | ; UNPACKED-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr5 | ||||
; UNPACKED-NEXT: [[COPY6:%[0-9]+]]:vgpr_32 = COPY $vgpr6 | ; UNPACKED-NEXT: [[COPY6:%[0-9]+]]:vgpr_32 = COPY $vgpr6 | ||||
; UNPACKED-NEXT: [[S_MOV_B64_:%[0-9]+]]:sreg_64_xexec = S_MOV_B64 $exec | ; UNPACKED-NEXT: [[S_MOV_B64_:%[0-9]+]]:sreg_64_xexec = S_MOV_B64 $exec | ||||
; UNPACKED-NEXT: {{ $}} | ; UNPACKED-NEXT: {{ $}} | ||||
; UNPACKED-NEXT: bb.2: | ; UNPACKED-NEXT: bb.2: | ||||
; UNPACKED-NEXT: successors: %bb.3(0x80000000) | ; UNPACKED-NEXT: successors: %bb.3(0x80000000) | ||||
; UNPACKED-NEXT: {{ $}} | ; UNPACKED-NEXT: {{ $}} | ||||
; UNPACKED-NEXT: [[COPY7:%[0-9]+]]:vgpr_32 = COPY [[REG_SEQUENCE]].sub0 | ; UNPACKED-NEXT: [[V_READFIRSTLANE_B32_:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY1]], implicit $exec | ||||
; UNPACKED-NEXT: [[COPY8:%[0-9]+]]:vgpr_32 = COPY [[REG_SEQUENCE]].sub1 | ; UNPACKED-NEXT: [[V_READFIRSTLANE_B32_1:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY2]], implicit $exec | ||||
; UNPACKED-NEXT: [[COPY9:%[0-9]+]]:vgpr_32 = COPY [[REG_SEQUENCE]].sub2 | ; UNPACKED-NEXT: [[V_READFIRSTLANE_B32_2:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY3]], implicit $exec | ||||
; UNPACKED-NEXT: [[COPY10:%[0-9]+]]:vgpr_32 = COPY [[REG_SEQUENCE]].sub3 | ; UNPACKED-NEXT: [[V_READFIRSTLANE_B32_3:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY4]], implicit $exec | ||||
; UNPACKED-NEXT: [[V_READFIRSTLANE_B32_:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY7]], implicit $exec | |||||
; UNPACKED-NEXT: [[V_READFIRSTLANE_B32_1:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY8]], implicit $exec | |||||
; UNPACKED-NEXT: [[V_READFIRSTLANE_B32_2:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY9]], implicit $exec | |||||
; UNPACKED-NEXT: [[V_READFIRSTLANE_B32_3:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY10]], implicit $exec | |||||
; UNPACKED-NEXT: [[REG_SEQUENCE1:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[V_READFIRSTLANE_B32_]], %subreg.sub0, [[V_READFIRSTLANE_B32_1]], %subreg.sub1, [[V_READFIRSTLANE_B32_2]], %subreg.sub2, [[V_READFIRSTLANE_B32_3]], %subreg.sub3 | ; UNPACKED-NEXT: [[REG_SEQUENCE1:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[V_READFIRSTLANE_B32_]], %subreg.sub0, [[V_READFIRSTLANE_B32_1]], %subreg.sub1, [[V_READFIRSTLANE_B32_2]], %subreg.sub2, [[V_READFIRSTLANE_B32_3]], %subreg.sub3 | ||||
; UNPACKED-NEXT: [[COPY11:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub0_sub1 | ; UNPACKED-NEXT: [[COPY7:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub0_sub1 | ||||
; UNPACKED-NEXT: [[COPY12:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub2_sub3 | ; UNPACKED-NEXT: [[COPY8:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub2_sub3 | ||||
; UNPACKED-NEXT: [[COPY13:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub0_sub1 | ; UNPACKED-NEXT: [[COPY9:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub0_sub1 | ||||
; UNPACKED-NEXT: [[COPY14:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub2_sub3 | ; UNPACKED-NEXT: [[COPY10:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub2_sub3 | ||||
; UNPACKED-NEXT: [[V_CMP_EQ_U64_e64_:%[0-9]+]]:sreg_64_xexec = V_CMP_EQ_U64_e64 [[COPY13]], [[COPY11]], implicit $exec | ; UNPACKED-NEXT: [[V_CMP_EQ_U64_e64_:%[0-9]+]]:sreg_64_xexec = V_CMP_EQ_U64_e64 [[COPY9]], [[COPY7]], implicit $exec | ||||
; UNPACKED-NEXT: [[V_CMP_EQ_U64_e64_1:%[0-9]+]]:sreg_64_xexec = V_CMP_EQ_U64_e64 [[COPY14]], [[COPY12]], implicit $exec | ; UNPACKED-NEXT: [[V_CMP_EQ_U64_e64_1:%[0-9]+]]:sreg_64_xexec = V_CMP_EQ_U64_e64 [[COPY10]], [[COPY8]], implicit $exec | ||||
; UNPACKED-NEXT: [[S_AND_B64_:%[0-9]+]]:sreg_64_xexec = S_AND_B64 [[V_CMP_EQ_U64_e64_]], [[V_CMP_EQ_U64_e64_1]], implicit-def dead $scc | ; UNPACKED-NEXT: [[S_AND_B64_:%[0-9]+]]:sreg_64_xexec = S_AND_B64 [[V_CMP_EQ_U64_e64_]], [[V_CMP_EQ_U64_e64_1]], implicit-def dead $scc | ||||
; UNPACKED-NEXT: [[V_READFIRSTLANE_B32_4:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY6]], implicit $exec | ; UNPACKED-NEXT: [[V_READFIRSTLANE_B32_4:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY6]], implicit $exec | ||||
; UNPACKED-NEXT: [[V_CMP_EQ_U32_e64_:%[0-9]+]]:sreg_64_xexec = V_CMP_EQ_U32_e64 [[V_READFIRSTLANE_B32_4]], [[COPY6]], implicit $exec | ; UNPACKED-NEXT: [[V_CMP_EQ_U32_e64_:%[0-9]+]]:sreg_64_xexec = V_CMP_EQ_U32_e64 [[V_READFIRSTLANE_B32_4]], [[COPY6]], implicit $exec | ||||
; UNPACKED-NEXT: [[S_AND_B64_1:%[0-9]+]]:sreg_64_xexec = S_AND_B64 [[S_AND_B64_]], [[V_CMP_EQ_U32_e64_]], implicit-def dead $scc | ; UNPACKED-NEXT: [[S_AND_B64_1:%[0-9]+]]:sreg_64_xexec = S_AND_B64 [[S_AND_B64_]], [[V_CMP_EQ_U32_e64_]], implicit-def dead $scc | ||||
; UNPACKED-NEXT: [[S_AND_SAVEEXEC_B64_:%[0-9]+]]:sreg_64_xexec = S_AND_SAVEEXEC_B64 killed [[S_AND_B64_1]], implicit-def $exec, implicit-def $scc, implicit $exec | ; UNPACKED-NEXT: [[S_AND_SAVEEXEC_B64_:%[0-9]+]]:sreg_64_xexec = S_AND_SAVEEXEC_B64 killed [[S_AND_B64_1]], implicit-def $exec, implicit-def $scc, implicit $exec | ||||
; UNPACKED-NEXT: {{ $}} | ; UNPACKED-NEXT: {{ $}} | ||||
; UNPACKED-NEXT: bb.3: | ; UNPACKED-NEXT: bb.3: | ||||
; UNPACKED-NEXT: successors: %bb.4(0x40000000), %bb.2(0x40000000) | ; UNPACKED-NEXT: successors: %bb.4(0x40000000), %bb.2(0x40000000) | ||||
Show All 22 Lines | define amdgpu_ps void @raw_tbuffer_store_f16__vgpr_rsrc__vgpr_voffset__vgpr_soffset(half %val, <4 x i32> %rsrc, i32 %voffset, i32 %soffset) { | ||||
; PACKED-NEXT: [[REG_SEQUENCE:%[0-9]+]]:vreg_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3 | ; PACKED-NEXT: [[REG_SEQUENCE:%[0-9]+]]:vreg_128 = REG_SEQUENCE [[COPY1]], %subreg.sub0, [[COPY2]], %subreg.sub1, [[COPY3]], %subreg.sub2, [[COPY4]], %subreg.sub3 | ||||
; PACKED-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr5 | ; PACKED-NEXT: [[COPY5:%[0-9]+]]:vgpr_32 = COPY $vgpr5 | ||||
; PACKED-NEXT: [[COPY6:%[0-9]+]]:vgpr_32 = COPY $vgpr6 | ; PACKED-NEXT: [[COPY6:%[0-9]+]]:vgpr_32 = COPY $vgpr6 | ||||
; PACKED-NEXT: [[S_MOV_B64_:%[0-9]+]]:sreg_64_xexec = S_MOV_B64 $exec | ; PACKED-NEXT: [[S_MOV_B64_:%[0-9]+]]:sreg_64_xexec = S_MOV_B64 $exec | ||||
; PACKED-NEXT: {{ $}} | ; PACKED-NEXT: {{ $}} | ||||
; PACKED-NEXT: bb.2: | ; PACKED-NEXT: bb.2: | ||||
; PACKED-NEXT: successors: %bb.3(0x80000000) | ; PACKED-NEXT: successors: %bb.3(0x80000000) | ||||
; PACKED-NEXT: {{ $}} | ; PACKED-NEXT: {{ $}} | ||||
; PACKED-NEXT: [[COPY7:%[0-9]+]]:vgpr_32 = COPY [[REG_SEQUENCE]].sub0 | ; PACKED-NEXT: [[V_READFIRSTLANE_B32_:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY1]], implicit $exec | ||||
; PACKED-NEXT: [[COPY8:%[0-9]+]]:vgpr_32 = COPY [[REG_SEQUENCE]].sub1 | ; PACKED-NEXT: [[V_READFIRSTLANE_B32_1:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY2]], implicit $exec | ||||
; PACKED-NEXT: [[COPY9:%[0-9]+]]:vgpr_32 = COPY [[REG_SEQUENCE]].sub2 | ; PACKED-NEXT: [[V_READFIRSTLANE_B32_2:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY3]], implicit $exec | ||||
; PACKED-NEXT: [[COPY10:%[0-9]+]]:vgpr_32 = COPY [[REG_SEQUENCE]].sub3 | ; PACKED-NEXT: [[V_READFIRSTLANE_B32_3:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY4]], implicit $exec | ||||
; PACKED-NEXT: [[V_READFIRSTLANE_B32_:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY7]], implicit $exec | |||||
; PACKED-NEXT: [[V_READFIRSTLANE_B32_1:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY8]], implicit $exec | |||||
; PACKED-NEXT: [[V_READFIRSTLANE_B32_2:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY9]], implicit $exec | |||||
; PACKED-NEXT: [[V_READFIRSTLANE_B32_3:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY10]], implicit $exec | |||||
; PACKED-NEXT: [[REG_SEQUENCE1:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[V_READFIRSTLANE_B32_]], %subreg.sub0, [[V_READFIRSTLANE_B32_1]], %subreg.sub1, [[V_READFIRSTLANE_B32_2]], %subreg.sub2, [[V_READFIRSTLANE_B32_3]], %subreg.sub3 | ; PACKED-NEXT: [[REG_SEQUENCE1:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[V_READFIRSTLANE_B32_]], %subreg.sub0, [[V_READFIRSTLANE_B32_1]], %subreg.sub1, [[V_READFIRSTLANE_B32_2]], %subreg.sub2, [[V_READFIRSTLANE_B32_3]], %subreg.sub3 | ||||
; PACKED-NEXT: [[COPY11:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub0_sub1 | ; PACKED-NEXT: [[COPY7:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub0_sub1 | ||||
; PACKED-NEXT: [[COPY12:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub2_sub3 | ; PACKED-NEXT: [[COPY8:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub2_sub3 | ||||
; PACKED-NEXT: [[COPY13:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub0_sub1 | ; PACKED-NEXT: [[COPY9:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub0_sub1 | ||||
; PACKED-NEXT: [[COPY14:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub2_sub3 | ; PACKED-NEXT: [[COPY10:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub2_sub3 | ||||
; PACKED-NEXT: [[V_CMP_EQ_U64_e64_:%[0-9]+]]:sreg_64_xexec = V_CMP_EQ_U64_e64 [[COPY13]], [[COPY11]], implicit $exec | ; PACKED-NEXT: [[V_CMP_EQ_U64_e64_:%[0-9]+]]:sreg_64_xexec = V_CMP_EQ_U64_e64 [[COPY9]], [[COPY7]], implicit $exec | ||||
; PACKED-NEXT: [[V_CMP_EQ_U64_e64_1:%[0-9]+]]:sreg_64_xexec = V_CMP_EQ_U64_e64 [[COPY14]], [[COPY12]], implicit $exec | ; PACKED-NEXT: [[V_CMP_EQ_U64_e64_1:%[0-9]+]]:sreg_64_xexec = V_CMP_EQ_U64_e64 [[COPY10]], [[COPY8]], implicit $exec | ||||
; PACKED-NEXT: [[S_AND_B64_:%[0-9]+]]:sreg_64_xexec = S_AND_B64 [[V_CMP_EQ_U64_e64_]], [[V_CMP_EQ_U64_e64_1]], implicit-def dead $scc | ; PACKED-NEXT: [[S_AND_B64_:%[0-9]+]]:sreg_64_xexec = S_AND_B64 [[V_CMP_EQ_U64_e64_]], [[V_CMP_EQ_U64_e64_1]], implicit-def dead $scc | ||||
; PACKED-NEXT: [[V_READFIRSTLANE_B32_4:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY6]], implicit $exec | ; PACKED-NEXT: [[V_READFIRSTLANE_B32_4:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY6]], implicit $exec | ||||
; PACKED-NEXT: [[V_CMP_EQ_U32_e64_:%[0-9]+]]:sreg_64_xexec = V_CMP_EQ_U32_e64 [[V_READFIRSTLANE_B32_4]], [[COPY6]], implicit $exec | ; PACKED-NEXT: [[V_CMP_EQ_U32_e64_:%[0-9]+]]:sreg_64_xexec = V_CMP_EQ_U32_e64 [[V_READFIRSTLANE_B32_4]], [[COPY6]], implicit $exec | ||||
; PACKED-NEXT: [[S_AND_B64_1:%[0-9]+]]:sreg_64_xexec = S_AND_B64 [[S_AND_B64_]], [[V_CMP_EQ_U32_e64_]], implicit-def dead $scc | ; PACKED-NEXT: [[S_AND_B64_1:%[0-9]+]]:sreg_64_xexec = S_AND_B64 [[S_AND_B64_]], [[V_CMP_EQ_U32_e64_]], implicit-def dead $scc | ||||
; PACKED-NEXT: [[S_AND_SAVEEXEC_B64_:%[0-9]+]]:sreg_64_xexec = S_AND_SAVEEXEC_B64 killed [[S_AND_B64_1]], implicit-def $exec, implicit-def $scc, implicit $exec | ; PACKED-NEXT: [[S_AND_SAVEEXEC_B64_:%[0-9]+]]:sreg_64_xexec = S_AND_SAVEEXEC_B64 killed [[S_AND_B64_1]], implicit-def $exec, implicit-def $scc, implicit $exec | ||||
; PACKED-NEXT: {{ $}} | ; PACKED-NEXT: {{ $}} | ||||
; PACKED-NEXT: bb.3: | ; PACKED-NEXT: bb.3: | ||||
; PACKED-NEXT: successors: %bb.4(0x40000000), %bb.2(0x40000000) | ; PACKED-NEXT: successors: %bb.4(0x40000000), %bb.2(0x40000000) | ||||
Show All 29 Lines | define amdgpu_ps void @raw_tbuffer_store_f16__vgpr_rsrc__sgpr_voffset__vgpr_soffset(half %val, <4 x i32> %rsrc, i32 inreg %voffset, i32 %soffset) { | ||||
; UNPACKED-NEXT: [[COPY5:%[0-9]+]]:sreg_32 = COPY $sgpr2 | ; UNPACKED-NEXT: [[COPY5:%[0-9]+]]:sreg_32 = COPY $sgpr2 | ||||
; UNPACKED-NEXT: [[COPY6:%[0-9]+]]:vgpr_32 = COPY $vgpr5 | ; UNPACKED-NEXT: [[COPY6:%[0-9]+]]:vgpr_32 = COPY $vgpr5 | ||||
; UNPACKED-NEXT: [[COPY7:%[0-9]+]]:vgpr_32 = COPY [[COPY5]] | ; UNPACKED-NEXT: [[COPY7:%[0-9]+]]:vgpr_32 = COPY [[COPY5]] | ||||
; UNPACKED-NEXT: [[S_MOV_B64_:%[0-9]+]]:sreg_64_xexec = S_MOV_B64 $exec | ; UNPACKED-NEXT: [[S_MOV_B64_:%[0-9]+]]:sreg_64_xexec = S_MOV_B64 $exec | ||||
; UNPACKED-NEXT: {{ $}} | ; UNPACKED-NEXT: {{ $}} | ||||
; UNPACKED-NEXT: bb.2: | ; UNPACKED-NEXT: bb.2: | ||||
; UNPACKED-NEXT: successors: %bb.3(0x80000000) | ; UNPACKED-NEXT: successors: %bb.3(0x80000000) | ||||
; UNPACKED-NEXT: {{ $}} | ; UNPACKED-NEXT: {{ $}} | ||||
; UNPACKED-NEXT: [[COPY8:%[0-9]+]]:vgpr_32 = COPY [[REG_SEQUENCE]].sub0 | ; UNPACKED-NEXT: [[V_READFIRSTLANE_B32_:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY1]], implicit $exec | ||||
; UNPACKED-NEXT: [[COPY9:%[0-9]+]]:vgpr_32 = COPY [[REG_SEQUENCE]].sub1 | ; UNPACKED-NEXT: [[V_READFIRSTLANE_B32_1:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY2]], implicit $exec | ||||
; UNPACKED-NEXT: [[COPY10:%[0-9]+]]:vgpr_32 = COPY [[REG_SEQUENCE]].sub2 | ; UNPACKED-NEXT: [[V_READFIRSTLANE_B32_2:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY3]], implicit $exec | ||||
; UNPACKED-NEXT: [[COPY11:%[0-9]+]]:vgpr_32 = COPY [[REG_SEQUENCE]].sub3 | ; UNPACKED-NEXT: [[V_READFIRSTLANE_B32_3:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY4]], implicit $exec | ||||
; UNPACKED-NEXT: [[V_READFIRSTLANE_B32_:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY8]], implicit $exec | |||||
; UNPACKED-NEXT: [[V_READFIRSTLANE_B32_1:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY9]], implicit $exec | |||||
; UNPACKED-NEXT: [[V_READFIRSTLANE_B32_2:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY10]], implicit $exec | |||||
; UNPACKED-NEXT: [[V_READFIRSTLANE_B32_3:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY11]], implicit $exec | |||||
; UNPACKED-NEXT: [[REG_SEQUENCE1:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[V_READFIRSTLANE_B32_]], %subreg.sub0, [[V_READFIRSTLANE_B32_1]], %subreg.sub1, [[V_READFIRSTLANE_B32_2]], %subreg.sub2, [[V_READFIRSTLANE_B32_3]], %subreg.sub3 | ; UNPACKED-NEXT: [[REG_SEQUENCE1:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[V_READFIRSTLANE_B32_]], %subreg.sub0, [[V_READFIRSTLANE_B32_1]], %subreg.sub1, [[V_READFIRSTLANE_B32_2]], %subreg.sub2, [[V_READFIRSTLANE_B32_3]], %subreg.sub3 | ||||
; UNPACKED-NEXT: [[COPY12:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub0_sub1 | ; UNPACKED-NEXT: [[COPY8:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub0_sub1 | ||||
; UNPACKED-NEXT: [[COPY13:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub2_sub3 | ; UNPACKED-NEXT: [[COPY9:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub2_sub3 | ||||
; UNPACKED-NEXT: [[COPY14:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub0_sub1 | ; UNPACKED-NEXT: [[COPY10:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub0_sub1 | ||||
; UNPACKED-NEXT: [[COPY15:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub2_sub3 | ; UNPACKED-NEXT: [[COPY11:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub2_sub3 | ||||
; UNPACKED-NEXT: [[V_CMP_EQ_U64_e64_:%[0-9]+]]:sreg_64_xexec = V_CMP_EQ_U64_e64 [[COPY14]], [[COPY12]], implicit $exec | ; UNPACKED-NEXT: [[V_CMP_EQ_U64_e64_:%[0-9]+]]:sreg_64_xexec = V_CMP_EQ_U64_e64 [[COPY10]], [[COPY8]], implicit $exec | ||||
; UNPACKED-NEXT: [[V_CMP_EQ_U64_e64_1:%[0-9]+]]:sreg_64_xexec = V_CMP_EQ_U64_e64 [[COPY15]], [[COPY13]], implicit $exec | ; UNPACKED-NEXT: [[V_CMP_EQ_U64_e64_1:%[0-9]+]]:sreg_64_xexec = V_CMP_EQ_U64_e64 [[COPY11]], [[COPY9]], implicit $exec | ||||
; UNPACKED-NEXT: [[S_AND_B64_:%[0-9]+]]:sreg_64_xexec = S_AND_B64 [[V_CMP_EQ_U64_e64_]], [[V_CMP_EQ_U64_e64_1]], implicit-def dead $scc | ; UNPACKED-NEXT: [[S_AND_B64_:%[0-9]+]]:sreg_64_xexec = S_AND_B64 [[V_CMP_EQ_U64_e64_]], [[V_CMP_EQ_U64_e64_1]], implicit-def dead $scc | ||||
; UNPACKED-NEXT: [[V_READFIRSTLANE_B32_4:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY6]], implicit $exec | ; UNPACKED-NEXT: [[V_READFIRSTLANE_B32_4:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY6]], implicit $exec | ||||
; UNPACKED-NEXT: [[V_CMP_EQ_U32_e64_:%[0-9]+]]:sreg_64_xexec = V_CMP_EQ_U32_e64 [[V_READFIRSTLANE_B32_4]], [[COPY6]], implicit $exec | ; UNPACKED-NEXT: [[V_CMP_EQ_U32_e64_:%[0-9]+]]:sreg_64_xexec = V_CMP_EQ_U32_e64 [[V_READFIRSTLANE_B32_4]], [[COPY6]], implicit $exec | ||||
; UNPACKED-NEXT: [[S_AND_B64_1:%[0-9]+]]:sreg_64_xexec = S_AND_B64 [[S_AND_B64_]], [[V_CMP_EQ_U32_e64_]], implicit-def dead $scc | ; UNPACKED-NEXT: [[S_AND_B64_1:%[0-9]+]]:sreg_64_xexec = S_AND_B64 [[S_AND_B64_]], [[V_CMP_EQ_U32_e64_]], implicit-def dead $scc | ||||
; UNPACKED-NEXT: [[S_AND_SAVEEXEC_B64_:%[0-9]+]]:sreg_64_xexec = S_AND_SAVEEXEC_B64 killed [[S_AND_B64_1]], implicit-def $exec, implicit-def $scc, implicit $exec | ; UNPACKED-NEXT: [[S_AND_SAVEEXEC_B64_:%[0-9]+]]:sreg_64_xexec = S_AND_SAVEEXEC_B64 killed [[S_AND_B64_1]], implicit-def $exec, implicit-def $scc, implicit $exec | ||||
; UNPACKED-NEXT: {{ $}} | ; UNPACKED-NEXT: {{ $}} | ||||
; UNPACKED-NEXT: bb.3: | ; UNPACKED-NEXT: bb.3: | ||||
; UNPACKED-NEXT: successors: %bb.4(0x40000000), %bb.2(0x40000000) | ; UNPACKED-NEXT: successors: %bb.4(0x40000000), %bb.2(0x40000000) | ||||
Show All 23 Lines | define amdgpu_ps void @raw_tbuffer_store_f16__vgpr_rsrc__sgpr_voffset__vgpr_soffset(half %val, <4 x i32> %rsrc, i32 inreg %voffset, i32 %soffset) { | ||||
; PACKED-NEXT: [[COPY5:%[0-9]+]]:sreg_32 = COPY $sgpr2 | ; PACKED-NEXT: [[COPY5:%[0-9]+]]:sreg_32 = COPY $sgpr2 | ||||
; PACKED-NEXT: [[COPY6:%[0-9]+]]:vgpr_32 = COPY $vgpr5 | ; PACKED-NEXT: [[COPY6:%[0-9]+]]:vgpr_32 = COPY $vgpr5 | ||||
; PACKED-NEXT: [[COPY7:%[0-9]+]]:vgpr_32 = COPY [[COPY5]] | ; PACKED-NEXT: [[COPY7:%[0-9]+]]:vgpr_32 = COPY [[COPY5]] | ||||
; PACKED-NEXT: [[S_MOV_B64_:%[0-9]+]]:sreg_64_xexec = S_MOV_B64 $exec | ; PACKED-NEXT: [[S_MOV_B64_:%[0-9]+]]:sreg_64_xexec = S_MOV_B64 $exec | ||||
; PACKED-NEXT: {{ $}} | ; PACKED-NEXT: {{ $}} | ||||
; PACKED-NEXT: bb.2: | ; PACKED-NEXT: bb.2: | ||||
; PACKED-NEXT: successors: %bb.3(0x80000000) | ; PACKED-NEXT: successors: %bb.3(0x80000000) | ||||
; PACKED-NEXT: {{ $}} | ; PACKED-NEXT: {{ $}} | ||||
; PACKED-NEXT: [[COPY8:%[0-9]+]]:vgpr_32 = COPY [[REG_SEQUENCE]].sub0 | ; PACKED-NEXT: [[V_READFIRSTLANE_B32_:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY1]], implicit $exec | ||||
; PACKED-NEXT: [[COPY9:%[0-9]+]]:vgpr_32 = COPY [[REG_SEQUENCE]].sub1 | ; PACKED-NEXT: [[V_READFIRSTLANE_B32_1:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY2]], implicit $exec | ||||
; PACKED-NEXT: [[COPY10:%[0-9]+]]:vgpr_32 = COPY [[REG_SEQUENCE]].sub2 | ; PACKED-NEXT: [[V_READFIRSTLANE_B32_2:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY3]], implicit $exec | ||||
; PACKED-NEXT: [[COPY11:%[0-9]+]]:vgpr_32 = COPY [[REG_SEQUENCE]].sub3 | ; PACKED-NEXT: [[V_READFIRSTLANE_B32_3:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY4]], implicit $exec | ||||
; PACKED-NEXT: [[V_READFIRSTLANE_B32_:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY8]], implicit $exec | |||||
; PACKED-NEXT: [[V_READFIRSTLANE_B32_1:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY9]], implicit $exec | |||||
; PACKED-NEXT: [[V_READFIRSTLANE_B32_2:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY10]], implicit $exec | |||||
; PACKED-NEXT: [[V_READFIRSTLANE_B32_3:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY11]], implicit $exec | |||||
; PACKED-NEXT: [[REG_SEQUENCE1:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[V_READFIRSTLANE_B32_]], %subreg.sub0, [[V_READFIRSTLANE_B32_1]], %subreg.sub1, [[V_READFIRSTLANE_B32_2]], %subreg.sub2, [[V_READFIRSTLANE_B32_3]], %subreg.sub3 | ; PACKED-NEXT: [[REG_SEQUENCE1:%[0-9]+]]:sgpr_128 = REG_SEQUENCE [[V_READFIRSTLANE_B32_]], %subreg.sub0, [[V_READFIRSTLANE_B32_1]], %subreg.sub1, [[V_READFIRSTLANE_B32_2]], %subreg.sub2, [[V_READFIRSTLANE_B32_3]], %subreg.sub3 | ||||
; PACKED-NEXT: [[COPY12:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub0_sub1 | ; PACKED-NEXT: [[COPY8:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub0_sub1 | ||||
; PACKED-NEXT: [[COPY13:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub2_sub3 | ; PACKED-NEXT: [[COPY9:%[0-9]+]]:vreg_64 = COPY [[REG_SEQUENCE]].sub2_sub3 | ||||
; PACKED-NEXT: [[COPY14:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub0_sub1 | ; PACKED-NEXT: [[COPY10:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub0_sub1 | ||||
; PACKED-NEXT: [[COPY15:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub2_sub3 | ; PACKED-NEXT: [[COPY11:%[0-9]+]]:sreg_64 = COPY [[REG_SEQUENCE1]].sub2_sub3 | ||||
; PACKED-NEXT: [[V_CMP_EQ_U64_e64_:%[0-9]+]]:sreg_64_xexec = V_CMP_EQ_U64_e64 [[COPY14]], [[COPY12]], implicit $exec | ; PACKED-NEXT: [[V_CMP_EQ_U64_e64_:%[0-9]+]]:sreg_64_xexec = V_CMP_EQ_U64_e64 [[COPY10]], [[COPY8]], implicit $exec | ||||
; PACKED-NEXT: [[V_CMP_EQ_U64_e64_1:%[0-9]+]]:sreg_64_xexec = V_CMP_EQ_U64_e64 [[COPY15]], [[COPY13]], implicit $exec | ; PACKED-NEXT: [[V_CMP_EQ_U64_e64_1:%[0-9]+]]:sreg_64_xexec = V_CMP_EQ_U64_e64 [[COPY11]], [[COPY9]], implicit $exec | ||||
; PACKED-NEXT: [[S_AND_B64_:%[0-9]+]]:sreg_64_xexec = S_AND_B64 [[V_CMP_EQ_U64_e64_]], [[V_CMP_EQ_U64_e64_1]], implicit-def dead $scc | ; PACKED-NEXT: [[S_AND_B64_:%[0-9]+]]:sreg_64_xexec = S_AND_B64 [[V_CMP_EQ_U64_e64_]], [[V_CMP_EQ_U64_e64_1]], implicit-def dead $scc | ||||
; PACKED-NEXT: [[V_READFIRSTLANE_B32_4:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY6]], implicit $exec | ; PACKED-NEXT: [[V_READFIRSTLANE_B32_4:%[0-9]+]]:sreg_32 = V_READFIRSTLANE_B32 [[COPY6]], implicit $exec | ||||
; PACKED-NEXT: [[V_CMP_EQ_U32_e64_:%[0-9]+]]:sreg_64_xexec = V_CMP_EQ_U32_e64 [[V_READFIRSTLANE_B32_4]], [[COPY6]], implicit $exec | ; PACKED-NEXT: [[V_CMP_EQ_U32_e64_:%[0-9]+]]:sreg_64_xexec = V_CMP_EQ_U32_e64 [[V_READFIRSTLANE_B32_4]], [[COPY6]], implicit $exec | ||||
; PACKED-NEXT: [[S_AND_B64_1:%[0-9]+]]:sreg_64_xexec = S_AND_B64 [[S_AND_B64_]], [[V_CMP_EQ_U32_e64_]], implicit-def dead $scc | ; PACKED-NEXT: [[S_AND_B64_1:%[0-9]+]]:sreg_64_xexec = S_AND_B64 [[S_AND_B64_]], [[V_CMP_EQ_U32_e64_]], implicit-def dead $scc | ||||
; PACKED-NEXT: [[S_AND_SAVEEXEC_B64_:%[0-9]+]]:sreg_64_xexec = S_AND_SAVEEXEC_B64 killed [[S_AND_B64_1]], implicit-def $exec, implicit-def $scc, implicit $exec | ; PACKED-NEXT: [[S_AND_SAVEEXEC_B64_:%[0-9]+]]:sreg_64_xexec = S_AND_SAVEEXEC_B64 killed [[S_AND_B64_1]], implicit-def $exec, implicit-def $scc, implicit $exec | ||||
; PACKED-NEXT: {{ $}} | ; PACKED-NEXT: {{ $}} | ||||
; PACKED-NEXT: bb.3: | ; PACKED-NEXT: bb.3: | ||||
; PACKED-NEXT: successors: %bb.4(0x40000000), %bb.2(0x40000000) | ; PACKED-NEXT: successors: %bb.4(0x40000000), %bb.2(0x40000000) | ||||
▲ Show 20 Lines • Show All 152 Lines • Show Last 20 Lines |