Skip to content
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

AMDGPU: Fix inst-selection of large scratch offsets with sgpr base #110256

Merged
merged 1 commit into from
Sep 30, 2024
+9 −9
Merged

AMDGPU: Fix inst-selection of large scratch offsets with sgpr base #110256

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
6 changes: 3 additions & 3 deletions llvm/lib/Target/AMDGPU/AMDGPUISelDAGToDAG.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1911,7 +1911,7 @@ bool AMDGPUDAGToDAGISel::SelectScratchSAddr(SDNode *Parent, SDValue Addr,
0);
}

Offset = CurDAG->getTargetConstant(COffsetVal, DL, MVT::i16);
Offset = CurDAG->getTargetConstant(COffsetVal, DL, MVT::i32);
petar-avramovic marked this conversation as resolved.
Show resolved Hide resolved

return true;
}
Expand Down Expand Up @@ -1966,7 +1966,7 @@ bool AMDGPUDAGToDAGISel::SelectScratchSVAddr(SDNode *N, SDValue Addr,
return false;
if (checkFlatScratchSVSSwizzleBug(VAddr, SAddr, SplitImmOffset))
return false;
Offset = CurDAG->getTargetConstant(SplitImmOffset, SDLoc(), MVT::i16);
Offset = CurDAG->getTargetConstant(SplitImmOffset, SDLoc(), MVT::i32);
return true;
}
}
Expand Down Expand Up @@ -1999,7 +1999,7 @@ bool AMDGPUDAGToDAGISel::SelectScratchSVAddr(SDNode *N, SDValue Addr,
if (checkFlatScratchSVSSwizzleBug(VAddr, SAddr, ImmOffset))
return false;
SAddr = SelectSAddrFI(CurDAG, SAddr);
Offset = CurDAG->getTargetConstant(ImmOffset, SDLoc(), MVT::i16);
Offset = CurDAG->getTargetConstant(ImmOffset, SDLoc(), MVT::i32);
return true;
}

Expand Down
12 changes: 6 additions & 6 deletions llvm/test/CodeGen/AMDGPU/flat-scratch.ll
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4926,7 +4926,7 @@ define amdgpu_gs void @sgpr_base_large_offset(ptr addrspace(1) %out, ptr addrspa
;
; GFX12-LABEL: sgpr_base_large_offset:
; GFX12: ; %bb.0: ; %entry
; GFX12-NEXT: scratch_load_b32 v2, off, s0 offset:-24
; GFX12-NEXT: scratch_load_b32 v2, off, s0 offset:65512
; GFX12-NEXT: s_wait_loadcnt 0x0
; GFX12-NEXT: global_store_b32 v[0:1], v2, off
; GFX12-NEXT: s_nop 0
Expand Down Expand Up @@ -4985,7 +4985,7 @@ define amdgpu_gs void @sgpr_base_large_offset(ptr addrspace(1) %out, ptr addrspa
;
; GFX12-PAL-LABEL: sgpr_base_large_offset:
; GFX12-PAL: ; %bb.0: ; %entry
; GFX12-PAL-NEXT: scratch_load_b32 v2, off, s0 offset:-24
; GFX12-PAL-NEXT: scratch_load_b32 v2, off, s0 offset:65512
; GFX12-PAL-NEXT: s_wait_loadcnt 0x0
; GFX12-PAL-NEXT: global_store_b32 v[0:1], v2, off
; GFX12-PAL-NEXT: s_nop 0
Expand Down Expand Up @@ -5038,7 +5038,7 @@ define amdgpu_gs void @sgpr_base_large_offset_split(ptr addrspace(1) %out, ptr a
; GFX12: ; %bb.0: ; %entry
; GFX12-NEXT: v_mov_b32_e32 v2, 0x1000000
; GFX12-NEXT: s_and_b32 s0, s0, -4
; GFX12-NEXT: scratch_load_b32 v2, v2, s0 offset:-24 scope:SCOPE_SYS
; GFX12-NEXT: scratch_load_b32 v2, v2, s0 offset:65512 scope:SCOPE_SYS
; GFX12-NEXT: s_wait_loadcnt 0x0
; GFX12-NEXT: global_store_b32 v[0:1], v2, off
; GFX12-NEXT: s_nop 0
Expand Down Expand Up @@ -5103,7 +5103,7 @@ define amdgpu_gs void @sgpr_base_large_offset_split(ptr addrspace(1) %out, ptr a
; GFX12-PAL: ; %bb.0: ; %entry
; GFX12-PAL-NEXT: v_mov_b32_e32 v2, 0x1000000
; GFX12-PAL-NEXT: s_and_b32 s0, s0, -4
; GFX12-PAL-NEXT: scratch_load_b32 v2, v2, s0 offset:-24 scope:SCOPE_SYS
; GFX12-PAL-NEXT: scratch_load_b32 v2, v2, s0 offset:65512 scope:SCOPE_SYS
; GFX12-PAL-NEXT: s_wait_loadcnt 0x0
; GFX12-PAL-NEXT: global_store_b32 v[0:1], v2, off
; GFX12-PAL-NEXT: s_nop 0
Expand Down Expand Up @@ -5159,7 +5159,7 @@ define amdgpu_gs void @sgpr_base_plus_sgpr_plus_vgpr_plus_large_imm_offset(ptr a
; GFX12: ; %bb.0: ; %bb
; GFX12-NEXT: v_mov_b32_e32 v1, 15
; GFX12-NEXT: s_add_co_i32 s0, s0, s1
; GFX12-NEXT: scratch_store_b32 v0, v1, s0 offset:-24 scope:SCOPE_SYS
; GFX12-NEXT: scratch_store_b32 v0, v1, s0 offset:65512 scope:SCOPE_SYS
; GFX12-NEXT: s_wait_storecnt 0x0
; GFX12-NEXT: s_endpgm
;
Expand Down Expand Up @@ -5221,7 +5221,7 @@ define amdgpu_gs void @sgpr_base_plus_sgpr_plus_vgpr_plus_large_imm_offset(ptr a
; GFX12-PAL: ; %bb.0: ; %bb
; GFX12-PAL-NEXT: v_mov_b32_e32 v1, 15
; GFX12-PAL-NEXT: s_add_co_i32 s0, s0, s1
; GFX12-PAL-NEXT: scratch_store_b32 v0, v1, s0 offset:-24 scope:SCOPE_SYS
; GFX12-PAL-NEXT: scratch_store_b32 v0, v1, s0 offset:65512 scope:SCOPE_SYS
; GFX12-PAL-NEXT: s_wait_storecnt 0x0
; GFX12-PAL-NEXT: s_endpgm
bb:
Expand Down
Loading