2f885cbe90
We were executing this in a waterfall loop as a placeholder, but this should really be converted to a MUBUF load. Also execute in a waterfall loop if the resource isn't an SGPR. This is a case where the DAG handling was wrong because doing the right thing was too hard. Currently, this will mishandle 96-bit loads. There's currently no way to track the original memory size with an MMO, so these loads will be widened andd the resulting memory size will be 128-bits.
49 lines
2.6 KiB
YAML
49 lines
2.6 KiB
YAML
# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py
|
|
# RUN: llc -mtriple=amdgcn-mesa-mesa3d -mcpu=tahiti -run-pass=regbankselect -regbankselect-fast -o - %s | FileCheck -check-prefix=FAST %s
|
|
# RUN: llc -mtriple=amdgcn-mesa-mesa3d -mcpu=tahiti -run-pass=regbankselect -regbankselect-greedy -o - %s | FileCheck -check-prefix=GREEDY %s
|
|
|
|
# We see the offset is a VGPR, but this is due to a constant for some
|
|
# reason ending up in a VGPR. This shouldn't really ever happen, but
|
|
# make sure this doesn't break when looking through copies for the add
|
|
# operands.
|
|
|
|
---
|
|
name: s_buffer_load_f32_vgpr_offset_cross_bank_copy_add_offset
|
|
legalized: true
|
|
tracksRegLiveness: true
|
|
body: |
|
|
bb.0:
|
|
liveins: $sgpr0_sgpr1_sgpr2_sgpr3, $sgpr0
|
|
|
|
; FAST-LABEL: name: s_buffer_load_f32_vgpr_offset_cross_bank_copy_add_offset
|
|
; FAST: liveins: $sgpr0_sgpr1_sgpr2_sgpr3, $sgpr0
|
|
; FAST: [[COPY:%[0-9]+]]:sgpr(<4 x s32>) = COPY $sgpr0_sgpr1_sgpr2_sgpr3
|
|
; FAST: [[COPY1:%[0-9]+]]:sgpr(s32) = COPY $sgpr0
|
|
; FAST: [[C:%[0-9]+]]:sgpr(s32) = G_CONSTANT i32 256
|
|
; FAST: [[COPY2:%[0-9]+]]:vgpr(s32) = COPY [[C]](s32)
|
|
; FAST: [[COPY3:%[0-9]+]]:vgpr(s32) = COPY [[COPY1]](s32)
|
|
; FAST: [[ADD:%[0-9]+]]:vgpr(s32) = G_ADD [[COPY3]], [[COPY2]]
|
|
; FAST: [[C1:%[0-9]+]]:sgpr(s32) = G_CONSTANT i32 0
|
|
; FAST: [[C2:%[0-9]+]]:vgpr(s32) = G_CONSTANT i32 0
|
|
; FAST: [[AMDGPU_BUFFER_LOAD:%[0-9]+]]:vgpr(s32) = G_AMDGPU_BUFFER_LOAD [[COPY]](<4 x s32>), [[C2]](s32), [[COPY3]], [[C1]], 256, 0, 0 :: (dereferenceable invariant load 4)
|
|
; FAST: S_ENDPGM 0, implicit [[AMDGPU_BUFFER_LOAD]](s32)
|
|
; GREEDY-LABEL: name: s_buffer_load_f32_vgpr_offset_cross_bank_copy_add_offset
|
|
; GREEDY: liveins: $sgpr0_sgpr1_sgpr2_sgpr3, $sgpr0
|
|
; GREEDY: [[COPY:%[0-9]+]]:sgpr(<4 x s32>) = COPY $sgpr0_sgpr1_sgpr2_sgpr3
|
|
; GREEDY: [[COPY1:%[0-9]+]]:sgpr(s32) = COPY $sgpr0
|
|
; GREEDY: [[C:%[0-9]+]]:vgpr(s32) = G_CONSTANT i32 256
|
|
; GREEDY: [[COPY2:%[0-9]+]]:vgpr(s32) = COPY [[COPY1]](s32)
|
|
; GREEDY: [[ADD:%[0-9]+]]:vgpr(s32) = G_ADD [[COPY2]], [[C]]
|
|
; GREEDY: [[C1:%[0-9]+]]:sgpr(s32) = G_CONSTANT i32 0
|
|
; GREEDY: [[C2:%[0-9]+]]:vgpr(s32) = G_CONSTANT i32 0
|
|
; GREEDY: [[AMDGPU_BUFFER_LOAD:%[0-9]+]]:vgpr(s32) = G_AMDGPU_BUFFER_LOAD [[COPY]](<4 x s32>), [[C2]](s32), [[COPY2]], [[C1]], 256, 0, 0 :: (dereferenceable invariant load 4)
|
|
; GREEDY: S_ENDPGM 0, implicit [[AMDGPU_BUFFER_LOAD]](s32)
|
|
%0:_(<4 x s32>) = COPY $sgpr0_sgpr1_sgpr2_sgpr3
|
|
%1:_(s32) = COPY $sgpr0
|
|
%2:vgpr(s32) = G_CONSTANT i32 256
|
|
%3:_(s32) = G_ADD %1, %2
|
|
%4:_(s32) = G_AMDGPU_S_BUFFER_LOAD %0, %3, 0
|
|
S_ENDPGM 0, implicit %4
|
|
|
|
...
|