Skip to content

AMDGPU: VIMAGE/VSAMPLE are bundle candidates #140854

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 2 commits into from
May 28, 2025
Merged

AMDGPU: VIMAGE/VSAMPLE are bundle candidates #140854

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
95388de
AMDGPU: VIMAGE/VSAMPLE are bundle candidates
ruiling May 21, 2025
cae9aea
AMDGPU: add a test
ruiling May 27, 2025
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
3 changes: 2 additions & 1 deletion llvm/lib/Target/AMDGPU/SIPostRABundler.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -64,7 +64,8 @@ class SIPostRABundler {

constexpr uint64_t MemFlags = SIInstrFlags::MTBUF | SIInstrFlags::MUBUF |
SIInstrFlags::SMRD | SIInstrFlags::DS |
SIInstrFlags::FLAT | SIInstrFlags::MIMG;
SIInstrFlags::FLAT | SIInstrFlags::MIMG |
SIInstrFlags::VIMAGE | SIInstrFlags::VSAMPLE;

} // End anonymous namespace.

Expand Down
34 changes: 34 additions & 0 deletions llvm/test/CodeGen/AMDGPU/postra-bundle-vimage-vsample-gfx12.mir
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,34 @@
# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py
# RUN: llc -mtriple=amdgcn -mcpu=gfx1200 -passes=si-post-ra-bundler %s -o - | FileCheck -check-prefix=GFX12 %s

---
name: post_bundle_vimage
body: |
bb.0:
liveins: $vgpr1, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7
; GFX12-LABEL: name: post_bundle_vimage
; GFX12: liveins: $vgpr1, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: BUNDLE implicit-def $vgpr5, implicit-def $vgpr5_lo16, implicit-def $vgpr5_hi16, implicit-def $vgpr4, implicit-def $vgpr4_lo16, implicit-def $vgpr4_hi16, implicit killed $vgpr1, implicit killed $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, implicit $exec {
; GFX12-NEXT: $vgpr5 = IMAGE_LOAD_V1_V1_gfx12 $vgpr1, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, 2, 1, 0, 0, -1, 0, 0, implicit $exec :: (dereferenceable invariant load (s32), addrspace 8)
; GFX12-NEXT: $vgpr4 = IMAGE_LOAD_V1_V1_gfx12 killed $vgpr1, killed $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, 1, 1, 0, 0, -1, 0, 0, implicit $exec :: (dereferenceable invariant load (s32), addrspace 8)
; GFX12-NEXT: }
$vgpr5 = IMAGE_LOAD_V1_V1_gfx12 $vgpr1, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, 2, 1, 0, 0, -1, 0, 0, implicit $exec :: (dereferenceable invariant load (s32), addrspace 8)
$vgpr4 = IMAGE_LOAD_V1_V1_gfx12 killed $vgpr1, killed $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, 1, 1, 0, 0, -1, 0, 0, implicit $exec :: (dereferenceable invariant load (s32), addrspace 8)
...

---
name: post_bundle_vsample
body: |
bb.0:
liveins: $vgpr0, $vgpr1, $vgpr2, $vgpr3, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11
; GFX12-LABEL: name: post_bundle_vsample
; GFX12: liveins: $vgpr0, $vgpr1, $vgpr2, $vgpr3, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11
; GFX12-NEXT: {{ $}}
; GFX12-NEXT: BUNDLE implicit-def $vgpr6_vgpr7_vgpr8_vgpr9, implicit-def $vgpr6, implicit-def $vgpr6_lo16, implicit-def $vgpr6_hi16, implicit-def $vgpr7, implicit-def $vgpr7_lo16, implicit-def $vgpr7_hi16, implicit-def $vgpr8, implicit-def $vgpr8_lo16, implicit-def $vgpr8_hi16, implicit-def $vgpr9, implicit-def $vgpr9_lo16, implicit-def $vgpr9_hi16, implicit-def $vgpr6_vgpr7, implicit-def $vgpr6_vgpr7_vgpr8, implicit-def $vgpr7_vgpr8, implicit-def $vgpr7_vgpr8_vgpr9, implicit-def $vgpr8_vgpr9, implicit-def $vgpr10_vgpr11_vgpr12_vgpr13, implicit-def $vgpr10, implicit-def $vgpr10_lo16, implicit-def $vgpr10_hi16, implicit-def $vgpr11, implicit-def $vgpr11_lo16, implicit-def $vgpr11_hi16, implicit-def $vgpr12, implicit-def $vgpr12_lo16, implicit-def $vgpr12_hi16, implicit-def $vgpr13, implicit-def $vgpr13_lo16, implicit-def $vgpr13_hi16, implicit-def $vgpr10_vgpr11, implicit-def $vgpr10_vgpr11_vgpr12, implicit-def $vgpr11_vgpr12, implicit-def $vgpr11_vgpr12_vgpr13, implicit-def $vgpr12_vgpr13, implicit killed $vgpr0, implicit killed $vgpr1, implicit $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, implicit $sgpr8_sgpr9_sgpr10_sgpr11, implicit $exec, implicit killed $vgpr2, implicit killed $vgpr3 {
; GFX12-NEXT: $vgpr6_vgpr7_vgpr8_vgpr9 = IMAGE_SAMPLE_V4_V2_gfx12 killed $vgpr0, killed $vgpr1, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 15, 1, 0, 0, 0, 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s128), addrspace 8)
; GFX12-NEXT: $vgpr10_vgpr11_vgpr12_vgpr13 = IMAGE_SAMPLE_V4_V2_gfx12 killed $vgpr2, killed $vgpr3, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 15, 1, 0, 0, 0, 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s128), addrspace 8)
; GFX12-NEXT: }
$vgpr6_vgpr7_vgpr8_vgpr9 = IMAGE_SAMPLE_V4_V2_gfx12 killed $vgpr0, killed $vgpr1, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 15, 1, 0, 0, 0, 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s128), addrspace 8)
$vgpr10_vgpr11_vgpr12_vgpr13 = IMAGE_SAMPLE_V4_V2_gfx12 killed $vgpr2, killed $vgpr3, $sgpr0_sgpr1_sgpr2_sgpr3_sgpr4_sgpr5_sgpr6_sgpr7, $sgpr8_sgpr9_sgpr10_sgpr11, 15, 1, 0, 0, 0, 0, 0, 0, 0, implicit $exec :: (dereferenceable load (s128), addrspace 8)
...