Skip to content

Commit 0fb74d0

Browse files
arsenmarsenm
committed
AMDGPU: Fix broken attribute usage in test
1 parent 75a3f22 commit 0fb74d0

File tree

1 file changed

+38
-52
lines changed

1 file changed

+38
-52
lines changed

‎llvm/test/CodeGen/AMDGPU/GlobalISel/insertelement-stack-lower.ll

Lines changed: 38 additions & 52 deletions
Original file line numberDiff line numberDiff line change
@@ -16,6 +16,7 @@ define amdgpu_kernel void @v_insert_v64i32_varidx(ptr addrspace(1) %out.ptr, ptr
1616
; GCN-NEXT: s_load_dwordx16 s[36:51], s[22:23], 0x0
1717
; GCN-NEXT: s_load_dwordx16 s[52:67], s[22:23], 0x40
1818
; GCN-NEXT: s_load_dwordx16 s[4:19], s[22:23], 0x80
19+
; GCN-NEXT: v_mov_b32_e32 v64, 0
1920
; GCN-NEXT: s_waitcnt lgkmcnt(0)
2021
; GCN-NEXT: v_mov_b32_e32 v0, s36
2122
; GCN-NEXT: v_mov_b32_e32 v1, s37
@@ -157,23 +158,10 @@ define amdgpu_kernel void @v_insert_v64i32_varidx(ptr addrspace(1) %out.ptr, ptr
157158
; GCN-NEXT: buffer_load_dword v1, off, s[0:3], 0 offset:260
158159
; GCN-NEXT: buffer_load_dword v2, off, s[0:3], 0 offset:264
159160
; GCN-NEXT: buffer_load_dword v3, off, s[0:3], 0 offset:268
160-
; GCN-NEXT: s_waitcnt vmcnt(0)
161-
; GCN-NEXT: buffer_store_dword v0, off, s[0:3], 0 offset:512 ; 4-byte Folded Spill
162-
; GCN-NEXT: s_waitcnt vmcnt(0)
163-
; GCN-NEXT: buffer_store_dword v1, off, s[0:3], 0 offset:516 ; 4-byte Folded Spill
164-
; GCN-NEXT: buffer_store_dword v2, off, s[0:3], 0 offset:520 ; 4-byte Folded Spill
165-
; GCN-NEXT: buffer_store_dword v3, off, s[0:3], 0 offset:524 ; 4-byte Folded Spill
166-
; GCN-NEXT: buffer_load_dword v0, off, s[0:3], 0 offset:272
167-
; GCN-NEXT: s_nop 0
168-
; GCN-NEXT: buffer_load_dword v1, off, s[0:3], 0 offset:276
169-
; GCN-NEXT: buffer_load_dword v2, off, s[0:3], 0 offset:280
170-
; GCN-NEXT: buffer_load_dword v3, off, s[0:3], 0 offset:284
171-
; GCN-NEXT: s_waitcnt vmcnt(0)
172-
; GCN-NEXT: buffer_store_dword v0, off, s[0:3], 0 offset:528 ; 4-byte Folded Spill
173-
; GCN-NEXT: s_waitcnt vmcnt(0)
174-
; GCN-NEXT: buffer_store_dword v1, off, s[0:3], 0 offset:532 ; 4-byte Folded Spill
175-
; GCN-NEXT: buffer_store_dword v2, off, s[0:3], 0 offset:536 ; 4-byte Folded Spill
176-
; GCN-NEXT: buffer_store_dword v3, off, s[0:3], 0 offset:540 ; 4-byte Folded Spill
161+
; GCN-NEXT: buffer_load_dword v4, off, s[0:3], 0 offset:272
162+
; GCN-NEXT: buffer_load_dword v5, off, s[0:3], 0 offset:276
163+
; GCN-NEXT: buffer_load_dword v6, off, s[0:3], 0 offset:280
164+
; GCN-NEXT: buffer_load_dword v7, off, s[0:3], 0 offset:284
177165
; GCN-NEXT: buffer_load_dword v8, off, s[0:3], 0 offset:288
178166
; GCN-NEXT: buffer_load_dword v9, off, s[0:3], 0 offset:292
179167
; GCN-NEXT: buffer_load_dword v10, off, s[0:3], 0 offset:296
@@ -230,45 +218,43 @@ define amdgpu_kernel void @v_insert_v64i32_varidx(ptr addrspace(1) %out.ptr, ptr
230218
; GCN-NEXT: buffer_load_dword v61, off, s[0:3], 0 offset:500
231219
; GCN-NEXT: buffer_load_dword v62, off, s[0:3], 0 offset:504
232220
; GCN-NEXT: buffer_load_dword v63, off, s[0:3], 0 offset:508
233-
; GCN-NEXT: s_nop 0
234-
; GCN-NEXT: buffer_load_dword v1, off, s[0:3], 0 offset:512 ; 4-byte Folded Reload
235-
; GCN-NEXT: s_nop 0
236-
; GCN-NEXT: buffer_load_dword v2, off, s[0:3], 0 offset:516 ; 4-byte Folded Reload
237-
; GCN-NEXT: s_nop 0
238-
; GCN-NEXT: buffer_load_dword v3, off, s[0:3], 0 offset:520 ; 4-byte Folded Reload
239-
; GCN-NEXT: s_nop 0
240-
; GCN-NEXT: buffer_load_dword v4, off, s[0:3], 0 offset:524 ; 4-byte Folded Reload
241-
; GCN-NEXT: v_mov_b32_e32 v0, 0
242-
; GCN-NEXT: s_waitcnt vmcnt(0)
243-
; GCN-NEXT: global_store_dwordx4 v0, v[1:4], s[20:21]
244-
; GCN-NEXT: buffer_load_dword v1, off, s[0:3], 0 offset:528 ; 4-byte Folded Reload
245-
; GCN-NEXT: s_nop 0
246-
; GCN-NEXT: buffer_load_dword v2, off, s[0:3], 0 offset:532 ; 4-byte Folded Reload
247-
; GCN-NEXT: s_nop 0
248-
; GCN-NEXT: buffer_load_dword v3, off, s[0:3], 0 offset:536 ; 4-byte Folded Reload
249-
; GCN-NEXT: s_nop 0
250-
; GCN-NEXT: buffer_load_dword v4, off, s[0:3], 0 offset:540 ; 4-byte Folded Reload
251-
; GCN-NEXT: s_waitcnt vmcnt(0)
252-
; GCN-NEXT: global_store_dwordx4 v0, v[1:4], s[20:21] offset:16
253-
; GCN-NEXT: global_store_dwordx4 v0, v[8:11], s[20:21] offset:32
254-
; GCN-NEXT: global_store_dwordx4 v0, v[12:15], s[20:21] offset:48
255-
; GCN-NEXT: global_store_dwordx4 v0, v[16:19], s[20:21] offset:64
256-
; GCN-NEXT: global_store_dwordx4 v0, v[20:23], s[20:21] offset:80
257-
; GCN-NEXT: global_store_dwordx4 v0, v[24:27], s[20:21] offset:96
258-
; GCN-NEXT: global_store_dwordx4 v0, v[28:31], s[20:21] offset:112
259-
; GCN-NEXT: global_store_dwordx4 v0, v[32:35], s[20:21] offset:128
260-
; GCN-NEXT: global_store_dwordx4 v0, v[36:39], s[20:21] offset:144
261-
; GCN-NEXT: global_store_dwordx4 v0, v[40:43], s[20:21] offset:160
262-
; GCN-NEXT: global_store_dwordx4 v0, v[44:47], s[20:21] offset:176
263-
; GCN-NEXT: global_store_dwordx4 v0, v[48:51], s[20:21] offset:192
264-
; GCN-NEXT: global_store_dwordx4 v0, v[52:55], s[20:21] offset:208
265-
; GCN-NEXT: global_store_dwordx4 v0, v[56:59], s[20:21] offset:224
266-
; GCN-NEXT: global_store_dwordx4 v0, v[60:63], s[20:21] offset:240
221+
; GCN-NEXT: s_waitcnt vmcnt(60)
222+
; GCN-NEXT: global_store_dwordx4 v64, v[0:3], s[20:21]
223+
; GCN-NEXT: s_waitcnt vmcnt(57)
224+
; GCN-NEXT: global_store_dwordx4 v64, v[4:7], s[20:21] offset:16
225+
; GCN-NEXT: s_waitcnt vmcnt(54)
226+
; GCN-NEXT: global_store_dwordx4 v64, v[8:11], s[20:21] offset:32
227+
; GCN-NEXT: s_waitcnt vmcnt(51)
228+
; GCN-NEXT: global_store_dwordx4 v64, v[12:15], s[20:21] offset:48
229+
; GCN-NEXT: s_waitcnt vmcnt(48)
230+
; GCN-NEXT: global_store_dwordx4 v64, v[16:19], s[20:21] offset:64
231+
; GCN-NEXT: s_waitcnt vmcnt(45)
232+
; GCN-NEXT: global_store_dwordx4 v64, v[20:23], s[20:21] offset:80
233+
; GCN-NEXT: s_waitcnt vmcnt(42)
234+
; GCN-NEXT: global_store_dwordx4 v64, v[24:27], s[20:21] offset:96
235+
; GCN-NEXT: s_waitcnt vmcnt(39)
236+
; GCN-NEXT: global_store_dwordx4 v64, v[28:31], s[20:21] offset:112
237+
; GCN-NEXT: s_waitcnt vmcnt(36)
238+
; GCN-NEXT: global_store_dwordx4 v64, v[32:35], s[20:21] offset:128
239+
; GCN-NEXT: s_waitcnt vmcnt(33)
240+
; GCN-NEXT: global_store_dwordx4 v64, v[36:39], s[20:21] offset:144
241+
; GCN-NEXT: s_waitcnt vmcnt(30)
242+
; GCN-NEXT: global_store_dwordx4 v64, v[40:43], s[20:21] offset:160
243+
; GCN-NEXT: s_waitcnt vmcnt(27)
244+
; GCN-NEXT: global_store_dwordx4 v64, v[44:47], s[20:21] offset:176
245+
; GCN-NEXT: s_waitcnt vmcnt(24)
246+
; GCN-NEXT: global_store_dwordx4 v64, v[48:51], s[20:21] offset:192
247+
; GCN-NEXT: s_waitcnt vmcnt(21)
248+
; GCN-NEXT: global_store_dwordx4 v64, v[52:55], s[20:21] offset:208
249+
; GCN-NEXT: s_waitcnt vmcnt(18)
250+
; GCN-NEXT: global_store_dwordx4 v64, v[56:59], s[20:21] offset:224
251+
; GCN-NEXT: s_waitcnt vmcnt(15)
252+
; GCN-NEXT: global_store_dwordx4 v64, v[60:63], s[20:21] offset:240
267253
; GCN-NEXT: s_endpgm
268254
%vec = load <64 x i32>, ptr addrspace(1) %ptr
269255
%insert = insertelement <64 x i32> %vec, i32 %val, i32 %idx
270256
store <64 x i32> %insert, ptr addrspace(1) %out.ptr
271257
ret void
272258
}
273259

274-
attributes #0 = { "amdgpu-flat-workgroup-size"="1,256" "amdgpu-waves-per-eu"="1,10" }
260+
attributes #0 = { "amdgpu-flat-work-group-size"="1,256" "amdgpu-waves-per-eu"="1,10" }

0 commit comments

Comments
 (0)