File: llvm.amdgcn.lds.direct.load.ll

package info (click to toggle)
llvm-toolchain-20 1%3A20.1.6-1~exp1
  • links: PTS, VCS
  • area: main
  • in suites: experimental
  • size: 2,111,304 kB
  • sloc: cpp: 7,438,677; ansic: 1,393,822; asm: 1,012,926; python: 241,650; f90: 86,635; objc: 75,479; lisp: 42,144; pascal: 17,286; sh: 10,027; ml: 5,082; perl: 4,730; awk: 3,523; makefile: 3,349; javascript: 2,251; xml: 892; fortran: 672
file content (52 lines) | stat: -rw-r--r-- 2,518 bytes parent folder | download | duplicates (9)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
; RUN: llc -mtriple=amdgcn -mcpu=gfx1100 -verify-machineinstrs < %s | FileCheck -check-prefixes=GCN,GFX11 %s
; RUN: llc -global-isel -mtriple=amdgcn -mcpu=gfx1100 -verify-machineinstrs < %s | FileCheck -check-prefixes=GCN,GFX11 %s
; RUN: llc -mtriple=amdgcn -mcpu=gfx1200 -verify-machineinstrs < %s | FileCheck -check-prefixes=GCN,GFX12 %s
; RUN: llc -global-isel -mtriple=amdgcn -mcpu=gfx1200 -verify-machineinstrs < %s | FileCheck -check-prefixes=GCN,GFX12 %s

; GCN-LABEL: {{^}}lds_direct_load:
; GCN: s_mov_b32 m0
; GFX11: lds_direct_load v{{[0-9]+}}
; GFX12: ds_direct_load v{{[0-9]+}}
; GCN: s_mov_b32 m0
; GFX11: lds_direct_load v{{[0-9]+}}
; GFX12: ds_direct_load v{{[0-9]+}}
; GCN: s_mov_b32 m0
; GFX11: lds_direct_load v{{[0-9]+}}
; GFX11: s_waitcnt expcnt(2)
; GFX12: ds_direct_load v{{[0-9]+}}
; GFX12: s_wait_expcnt 0x2
; GCN: v_add_f32
; GCN: buffer_store_b32
; GFX11: s_waitcnt expcnt(1)
; GFX12: s_wait_expcnt 0x1
; GCN: buffer_store_b32
; GFX11: s_waitcnt expcnt(0)
; GFX12: s_wait_expcnt 0x0
; GCN: buffer_store_b32
; GCN: buffer_store_b32
; GCN: buffer_store_b32
; GCN: buffer_store_b32
define amdgpu_ps void @lds_direct_load(ptr addrspace(8) inreg %buf, i32 inreg %arg0,
                                       i32 inreg %arg1, i32 inreg %arg2) #0 {
main_body:
  %p0 = call float @llvm.amdgcn.lds.direct.load(i32 %arg0)
  ; Ensure memory clustering is occuring for lds_direct_load
  %p5 = fadd float %p0, 1.0
  %p1 = call float @llvm.amdgcn.lds.direct.load(i32 %arg1)
  %p2 = call float @llvm.amdgcn.lds.direct.load(i32 %arg2)
  %p3 = call float @llvm.amdgcn.lds.direct.load(i32 %arg1)
  %p4 = call float @llvm.amdgcn.lds.direct.load(i32 %arg2)
  call void @llvm.amdgcn.raw.ptr.buffer.store.f32(float %p5, ptr addrspace(8) %buf, i32 4, i32 0, i32 0)
  call void @llvm.amdgcn.raw.ptr.buffer.store.f32(float %p1, ptr addrspace(8) %buf, i32 4, i32 1, i32 0)
  call void @llvm.amdgcn.raw.ptr.buffer.store.f32(float %p2, ptr addrspace(8) %buf, i32 4, i32 2, i32 0)
  call void @llvm.amdgcn.raw.ptr.buffer.store.f32(float %p3, ptr addrspace(8) %buf, i32 4, i32 3, i32 0)
  call void @llvm.amdgcn.raw.ptr.buffer.store.f32(float %p4, ptr addrspace(8) %buf, i32 4, i32 4, i32 0)
  call void @llvm.amdgcn.raw.ptr.buffer.store.f32(float %p0, ptr addrspace(8) %buf, i32 4, i32 5, i32 0)
  ret void
}

declare float @llvm.amdgcn.lds.direct.load(i32) #1
declare void @llvm.amdgcn.raw.ptr.buffer.store.f32(float, ptr addrspace(8), i32, i32, i32)

attributes #0 = { nounwind }
attributes #1 = { nounwind readonly }