| # NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py |
| # RUN: llc -mtriple=amdgcn-mesa-mesa3d -mcpu=hawaii -run-pass=amdgpu-regbankselect %s -o - | FileCheck -check-prefix=GFX7 %s |
| # RUN: llc -mtriple=amdgcn-mesa-mesa3d -mcpu=gfx1200 -run-pass=amdgpu-regbankselect %s -o - | FileCheck -check-prefix=GFX12 %s |
| |
| --- | |
| |
| define amdgpu_ps i96 @split_smrd_load_range(ptr addrspace(4) %ptr) { |
| %load = load i96, ptr addrspace(4) %ptr, !range !0 |
| ret i96 %load |
| } |
| |
| define amdgpu_ps <3 x i32> @split_smrd_load_tbaa(ptr addrspace(4) %ptr) { |
| %load = load <3 x i32>, ptr addrspace(4) %ptr, !tbaa !1 |
| ret <3 x i32> %load |
| } |
| |
| !0 = !{i96 0, i96 9223372036854775808} |
| !1 = !{!"omnipotent char", !2} |
| !2 = !{!"Simple C/C++ TBAA"} |
| ... |
| |
| # Make sure range metadata is not preserved when widening loads, but |
| # tbaa is. |
| |
| --- |
| name: split_smrd_load_range |
| legalized: true |
| body: | |
| bb.0: |
| liveins: $sgpr0_sgpr1 |
| |
| ; GFX7-LABEL: name: split_smrd_load_range |
| ; GFX7: liveins: $sgpr0_sgpr1 |
| ; GFX7-NEXT: {{ $}} |
| ; GFX7-NEXT: [[COPY:%[0-9]+]]:sgpr(p4) = COPY $sgpr0_sgpr1 |
| ; GFX7-NEXT: [[LOAD:%[0-9]+]]:sgpr(<2 x s32>) = G_LOAD [[COPY]](p4) :: (load (<2 x s32>), addrspace 4) |
| ; GFX7-NEXT: [[C:%[0-9]+]]:sgpr(s64) = G_CONSTANT i64 8 |
| ; GFX7-NEXT: [[PTR_ADD:%[0-9]+]]:sgpr(p4) = G_PTR_ADD [[COPY]], [[C]](s64) |
| ; GFX7-NEXT: [[LOAD1:%[0-9]+]]:sgpr(s32) = G_LOAD [[PTR_ADD]](p4) :: (load (s32) from unknown-address + 8, align 8, addrspace 4) |
| ; GFX7-NEXT: [[UV:%[0-9]+]]:sgpr(s32), [[UV1:%[0-9]+]]:sgpr(s32) = G_UNMERGE_VALUES [[LOAD]](<2 x s32>) |
| ; GFX7-NEXT: [[BUILD_VECTOR:%[0-9]+]]:sgpr(<3 x s32>) = G_BUILD_VECTOR [[UV]](s32), [[UV1]](s32), [[LOAD1]](s32) |
| ; GFX7-NEXT: $sgpr0_sgpr1_sgpr2 = COPY [[BUILD_VECTOR]](<3 x s32>) |
| ; GFX12-LABEL: name: split_smrd_load_range |
| ; GFX12: liveins: $sgpr0_sgpr1 |
| ; GFX12-NEXT: {{ $}} |
| ; GFX12-NEXT: [[COPY:%[0-9]+]]:sgpr(p4) = COPY $sgpr0_sgpr1 |
| ; GFX12-NEXT: [[LOAD:%[0-9]+]]:sgpr(<3 x s32>) = G_LOAD [[COPY]](p4) :: (load (<3 x s32>), align 8, !range !0, addrspace 4) |
| ; GFX12-NEXT: $sgpr0_sgpr1_sgpr2 = COPY [[LOAD]](<3 x s32>) |
| %0:_(p4) = COPY $sgpr0_sgpr1 |
| %1:_(<3 x s32>) = G_LOAD %0 :: (load (<3 x s32>), align 8, addrspace 4, !range !0) |
| $sgpr0_sgpr1_sgpr2 = COPY %1 |
| |
| ... |
| |
| --- |
| name: split_smrd_load_tbaa |
| legalized: true |
| body: | |
| bb.0: |
| liveins: $sgpr0_sgpr1 |
| |
| ; GFX7-LABEL: name: split_smrd_load_tbaa |
| ; GFX7: liveins: $sgpr0_sgpr1 |
| ; GFX7-NEXT: {{ $}} |
| ; GFX7-NEXT: [[COPY:%[0-9]+]]:sgpr(p4) = COPY $sgpr0_sgpr1 |
| ; GFX7-NEXT: [[LOAD:%[0-9]+]]:sgpr(<2 x s32>) = G_LOAD [[COPY]](p4) :: (load (<2 x s32>), !tbaa !2, addrspace 4) |
| ; GFX7-NEXT: [[C:%[0-9]+]]:sgpr(s64) = G_CONSTANT i64 8 |
| ; GFX7-NEXT: [[PTR_ADD:%[0-9]+]]:sgpr(p4) = G_PTR_ADD [[COPY]], [[C]](s64) |
| ; GFX7-NEXT: [[LOAD1:%[0-9]+]]:sgpr(s32) = G_LOAD [[PTR_ADD]](p4) :: (load (s32) from unknown-address + 8, align 8, !tbaa !2, addrspace 4) |
| ; GFX7-NEXT: [[UV:%[0-9]+]]:sgpr(s32), [[UV1:%[0-9]+]]:sgpr(s32) = G_UNMERGE_VALUES [[LOAD]](<2 x s32>) |
| ; GFX7-NEXT: [[BUILD_VECTOR:%[0-9]+]]:sgpr(<3 x s32>) = G_BUILD_VECTOR [[UV]](s32), [[UV1]](s32), [[LOAD1]](s32) |
| ; GFX7-NEXT: $sgpr0_sgpr1_sgpr2 = COPY [[BUILD_VECTOR]](<3 x s32>) |
| ; GFX12-LABEL: name: split_smrd_load_tbaa |
| ; GFX12: liveins: $sgpr0_sgpr1 |
| ; GFX12-NEXT: {{ $}} |
| ; GFX12-NEXT: [[COPY:%[0-9]+]]:sgpr(p4) = COPY $sgpr0_sgpr1 |
| ; GFX12-NEXT: [[LOAD:%[0-9]+]]:sgpr(<3 x s32>) = G_LOAD [[COPY]](p4) :: (load (<3 x s32>), align 8, !tbaa !2, addrspace 4) |
| ; GFX12-NEXT: $sgpr0_sgpr1_sgpr2 = COPY [[LOAD]](<3 x s32>) |
| %0:_(p4) = COPY $sgpr0_sgpr1 |
| %1:_(<3 x s32>) = G_LOAD %0 :: (load (<3 x s32>), align 8, addrspace 4, !tbaa !1) |
| $sgpr0_sgpr1_sgpr2 = COPY %1 |
| |
| ... |