Skip to content

Commit ef8e4f7

Browse files
vg0204vg0204
committed
[LIT] Updated the regressing LIT tests to accomodate patch changes.
1 parent e14f6a3 commit ef8e4f7

File tree

301 files changed

+11638
-14323
lines changed

Some content is hidden

Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.

301 files changed

+11638
-14323
lines changed

‎llvm/test/CodeGen/AMDGPU/GlobalISel/add.ll‎

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -438,9 +438,9 @@ define void @s_uaddo_uadde(i64 inreg %a, i64 inreg %b, ptr addrspace(1) %res, pt
438438
; GFX7-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
439439
; GFX7-NEXT: s_add_u32 s4, s16, s18
440440
; GFX7-NEXT: s_addc_u32 s5, s17, s19
441-
; GFX7-NEXT: v_mov_b32_e32 v4, s4
442441
; GFX7-NEXT: s_mov_b32 s6, 0
443442
; GFX7-NEXT: s_cselect_b32 s8, 1, 0
443+
; GFX7-NEXT: v_mov_b32_e32 v4, s4
444444
; GFX7-NEXT: v_mov_b32_e32 v5, s5
445445
; GFX7-NEXT: s_mov_b32 s7, 0xf000
446446
; GFX7-NEXT: s_mov_b64 s[4:5], 0
@@ -455,8 +455,8 @@ define void @s_uaddo_uadde(i64 inreg %a, i64 inreg %b, ptr addrspace(1) %res, pt
455455
; GFX9-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
456456
; GFX9-NEXT: s_add_u32 s4, s16, s18
457457
; GFX9-NEXT: s_addc_u32 s5, s17, s19
458-
; GFX9-NEXT: v_mov_b32_e32 v4, s4
459458
; GFX9-NEXT: s_cselect_b32 s6, 1, 0
459+
; GFX9-NEXT: v_mov_b32_e32 v4, s4
460460
; GFX9-NEXT: v_mov_b32_e32 v5, s5
461461
; GFX9-NEXT: global_store_dwordx2 v[0:1], v[4:5], off
462462
; GFX9-NEXT: v_mov_b32_e32 v0, s6
@@ -469,8 +469,8 @@ define void @s_uaddo_uadde(i64 inreg %a, i64 inreg %b, ptr addrspace(1) %res, pt
469469
; GFX8-NEXT: s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
470470
; GFX8-NEXT: s_add_u32 s4, s16, s18
471471
; GFX8-NEXT: s_addc_u32 s5, s17, s19
472-
; GFX8-NEXT: v_mov_b32_e32 v4, s4
473472
; GFX8-NEXT: s_cselect_b32 s6, 1, 0
473+
; GFX8-NEXT: v_mov_b32_e32 v4, s4
474474
; GFX8-NEXT: v_mov_b32_e32 v5, s5
475475
; GFX8-NEXT: flat_store_dwordx2 v[0:1], v[4:5]
476476
; GFX8-NEXT: v_mov_b32_e32 v0, s6

‎llvm/test/CodeGen/AMDGPU/GlobalISel/addo.ll‎

Lines changed: 3 additions & 3 deletions
Original file line numberDiff line numberDiff line change
@@ -676,8 +676,8 @@ define amdgpu_ps i64 @s_saddo_i64(i64 inreg %a, i64 inreg %b) {
676676
; GFX7-LABEL: s_saddo_i64:
677677
; GFX7: ; %bb.0:
678678
; GFX7-NEXT: s_add_u32 s4, s0, s2
679-
; GFX7-NEXT: v_mov_b32_e32 v0, s0
680679
; GFX7-NEXT: s_addc_u32 s5, s1, s3
680+
; GFX7-NEXT: v_mov_b32_e32 v0, s0
681681
; GFX7-NEXT: v_mov_b32_e32 v1, s1
682682
; GFX7-NEXT: v_cmp_lt_i64_e32 vcc, s[4:5], v[0:1]
683683
; GFX7-NEXT: v_cmp_lt_i64_e64 s[0:1], s[2:3], 0
@@ -693,8 +693,8 @@ define amdgpu_ps i64 @s_saddo_i64(i64 inreg %a, i64 inreg %b) {
693693
; GFX8-LABEL: s_saddo_i64:
694694
; GFX8: ; %bb.0:
695695
; GFX8-NEXT: s_add_u32 s4, s0, s2
696-
; GFX8-NEXT: v_mov_b32_e32 v0, s0
697696
; GFX8-NEXT: s_addc_u32 s5, s1, s3
697+
; GFX8-NEXT: v_mov_b32_e32 v0, s0
698698
; GFX8-NEXT: v_mov_b32_e32 v1, s1
699699
; GFX8-NEXT: v_cmp_lt_i64_e32 vcc, s[4:5], v[0:1]
700700
; GFX8-NEXT: v_cmp_lt_i64_e64 s[0:1], s[2:3], 0
@@ -710,8 +710,8 @@ define amdgpu_ps i64 @s_saddo_i64(i64 inreg %a, i64 inreg %b) {
710710
; GFX9-LABEL: s_saddo_i64:
711711
; GFX9: ; %bb.0:
712712
; GFX9-NEXT: s_add_u32 s4, s0, s2
713-
; GFX9-NEXT: v_mov_b32_e32 v0, s0
714713
; GFX9-NEXT: s_addc_u32 s5, s1, s3
714+
; GFX9-NEXT: v_mov_b32_e32 v0, s0
715715
; GFX9-NEXT: v_mov_b32_e32 v1, s1
716716
; GFX9-NEXT: v_cmp_lt_i64_e32 vcc, s[4:5], v[0:1]
717717
; GFX9-NEXT: v_cmp_lt_i64_e64 s[0:1], s[2:3], 0

‎llvm/test/CodeGen/AMDGPU/GlobalISel/atomic_optimizations_mul_one.ll‎

Lines changed: 6 additions & 6 deletions
Original file line numberDiff line numberDiff line change
@@ -96,8 +96,8 @@ define amdgpu_cs void @atomic_add_and_format(<4 x i32> inreg %arg) {
9696
; GCN-NEXT: s_waitcnt vmcnt(0)
9797
; GCN-NEXT: v_readfirstlane_b32 s4, v1
9898
; GCN-NEXT: v_add_i32_e32 v4, vcc, s4, v0
99-
; GCN-NEXT: s_waitcnt expcnt(0)
10099
; GCN-NEXT: v_mov_b32_e32 v0, s0
100+
; GCN-NEXT: s_waitcnt expcnt(0)
101101
; GCN-NEXT: v_mov_b32_e32 v1, s1
102102
; GCN-NEXT: v_mov_b32_e32 v2, s2
103103
; GCN-NEXT: v_mov_b32_e32 v3, s3
@@ -192,8 +192,8 @@ define amdgpu_cs void @atomic_sub_and_format(<4 x i32> inreg %arg) {
192192
; GCN-NEXT: s_waitcnt vmcnt(0)
193193
; GCN-NEXT: v_readfirstlane_b32 s4, v1
194194
; GCN-NEXT: v_sub_i32_e32 v4, vcc, s4, v0
195-
; GCN-NEXT: s_waitcnt expcnt(0)
196195
; GCN-NEXT: v_mov_b32_e32 v0, s0
196+
; GCN-NEXT: s_waitcnt expcnt(0)
197197
; GCN-NEXT: v_mov_b32_e32 v1, s1
198198
; GCN-NEXT: v_mov_b32_e32 v2, s2
199199
; GCN-NEXT: v_mov_b32_e32 v3, s3
@@ -294,8 +294,8 @@ define amdgpu_cs void @atomic_xor_and_format(<4 x i32> inreg %arg) {
294294
; GCN-NEXT: v_readfirstlane_b32 s4, v1
295295
; GCN-NEXT: v_and_b32_e32 v0, 1, v0
296296
; GCN-NEXT: v_xor_b32_e32 v4, s4, v0
297-
; GCN-NEXT: s_waitcnt expcnt(0)
298297
; GCN-NEXT: v_mov_b32_e32 v0, s0
298+
; GCN-NEXT: s_waitcnt expcnt(0)
299299
; GCN-NEXT: v_mov_b32_e32 v1, s1
300300
; GCN-NEXT: v_mov_b32_e32 v2, s2
301301
; GCN-NEXT: v_mov_b32_e32 v3, s3
@@ -392,8 +392,8 @@ define amdgpu_cs void @atomic_ptr_add_and_format(ptr addrspace(8) inreg %arg) {
392392
; GCN-NEXT: s_waitcnt vmcnt(0)
393393
; GCN-NEXT: v_readfirstlane_b32 s4, v1
394394
; GCN-NEXT: v_add_i32_e32 v4, vcc, s4, v0
395-
; GCN-NEXT: s_waitcnt expcnt(0)
396395
; GCN-NEXT: v_mov_b32_e32 v0, s0
396+
; GCN-NEXT: s_waitcnt expcnt(0)
397397
; GCN-NEXT: v_mov_b32_e32 v1, s1
398398
; GCN-NEXT: v_mov_b32_e32 v2, s2
399399
; GCN-NEXT: v_mov_b32_e32 v3, s3
@@ -492,8 +492,8 @@ define amdgpu_cs void @atomic_ptr_sub_and_format(ptr addrspace(8) inreg %arg) {
492492
; GCN-NEXT: s_waitcnt vmcnt(0)
493493
; GCN-NEXT: v_readfirstlane_b32 s4, v1
494494
; GCN-NEXT: v_sub_i32_e32 v4, vcc, s4, v0
495-
; GCN-NEXT: s_waitcnt expcnt(0)
496495
; GCN-NEXT: v_mov_b32_e32 v0, s0
496+
; GCN-NEXT: s_waitcnt expcnt(0)
497497
; GCN-NEXT: v_mov_b32_e32 v1, s1
498498
; GCN-NEXT: v_mov_b32_e32 v2, s2
499499
; GCN-NEXT: v_mov_b32_e32 v3, s3
@@ -598,8 +598,8 @@ define amdgpu_cs void @atomic_ptr_xor_and_format(ptr addrspace(8) inreg %arg) {
598598
; GCN-NEXT: v_readfirstlane_b32 s4, v1
599599
; GCN-NEXT: v_and_b32_e32 v0, 1, v0
600600
; GCN-NEXT: v_xor_b32_e32 v4, s4, v0
601-
; GCN-NEXT: s_waitcnt expcnt(0)
602601
; GCN-NEXT: v_mov_b32_e32 v0, s0
602+
; GCN-NEXT: s_waitcnt expcnt(0)
603603
; GCN-NEXT: v_mov_b32_e32 v1, s1
604604
; GCN-NEXT: v_mov_b32_e32 v2, s2
605605
; GCN-NEXT: v_mov_b32_e32 v3, s3

‎llvm/test/CodeGen/AMDGPU/GlobalISel/atomicrmw_fmax.ll‎

Lines changed: 11 additions & 14 deletions
Original file line numberDiff line numberDiff line change
@@ -8,6 +8,7 @@
88
; RUN: llc -global-isel -mtriple=amdgcn-amd-amdpal -mcpu=tonga < %s | FileCheck -check-prefix=GFX8 %s
99
; RUN: llc -global-isel -mtriple=amdgcn-amd-amdpal -mcpu=hawaii < %s | FileCheck -check-prefix=GFX7 %s
1010

11+
1112
; TODO: Delete this and add run lines to use *-atomicrmw-fmax.ll tests
1213

1314
define float @local_atomic_fmax_ret_f32(ptr addrspace(3) %ptr, float %val) {
@@ -1823,10 +1824,8 @@ define double @buffer_fat_ptr_agent_atomic_fmax_ret_f64__amdgpu_no_fine_grained_
18231824
; GFX12-NEXT: s_wait_storecnt 0x0
18241825
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
18251826
; GFX12-NEXT: v_max_num_f64_e32 v[2:3], v[0:1], v[6:7]
1826-
; GFX12-NEXT: v_mov_b32_e32 v0, v2
1827-
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_2)
1828-
; GFX12-NEXT: v_dual_mov_b32 v1, v3 :: v_dual_mov_b32 v2, v4
1829-
; GFX12-NEXT: v_mov_b32_e32 v3, v5
1827+
; GFX12-NEXT: v_dual_mov_b32 v0, v2 :: v_dual_mov_b32 v1, v3
1828+
; GFX12-NEXT: v_dual_mov_b32 v2, v4 :: v_dual_mov_b32 v3, v5
18301829
; GFX12-NEXT: buffer_atomic_cmpswap_b64 v[0:3], v8, s[0:3], null offen th:TH_ATOMIC_RETURN
18311830
; GFX12-NEXT: s_wait_loadcnt 0x0
18321831
; GFX12-NEXT: global_inv scope:SCOPE_DEV
@@ -1865,10 +1864,8 @@ define double @buffer_fat_ptr_agent_atomic_fmax_ret_f64__amdgpu_no_fine_grained_
18651864
; GFX11-NEXT: s_waitcnt_vscnt null, 0x0
18661865
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
18671866
; GFX11-NEXT: v_max_f64 v[2:3], v[0:1], v[6:7]
1868-
; GFX11-NEXT: v_mov_b32_e32 v0, v2
1869-
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_2)
1870-
; GFX11-NEXT: v_dual_mov_b32 v1, v3 :: v_dual_mov_b32 v2, v4
1871-
; GFX11-NEXT: v_mov_b32_e32 v3, v5
1867+
; GFX11-NEXT: v_dual_mov_b32 v0, v2 :: v_dual_mov_b32 v1, v3
1868+
; GFX11-NEXT: v_dual_mov_b32 v2, v4 :: v_dual_mov_b32 v3, v5
18721869
; GFX11-NEXT: buffer_atomic_cmpswap_b64 v[0:3], v8, s[0:3], 0 offen glc
18731870
; GFX11-NEXT: s_waitcnt vmcnt(0)
18741871
; GFX11-NEXT: buffer_gl1_inv
@@ -1989,10 +1986,10 @@ define void @buffer_fat_ptr_agent_atomic_fmax_noret_f64__amdgpu_no_fine_grained_
19891986
; GFX12-NEXT: ; =>This Inner Loop Header: Depth=1
19901987
; GFX12-NEXT: s_wait_loadcnt 0x0
19911988
; GFX12-NEXT: v_max_num_f64_e32 v[0:1], v[2:3], v[2:3]
1989+
; GFX12-NEXT: v_dual_mov_b32 v10, v3 :: v_dual_mov_b32 v9, v2
19921990
; GFX12-NEXT: s_wait_storecnt 0x0
1993-
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2)
1991+
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_1)
19941992
; GFX12-NEXT: v_max_num_f64_e32 v[0:1], v[0:1], v[4:5]
1995-
; GFX12-NEXT: v_dual_mov_b32 v10, v3 :: v_dual_mov_b32 v9, v2
19961993
; GFX12-NEXT: v_dual_mov_b32 v8, v1 :: v_dual_mov_b32 v7, v0
19971994
; GFX12-NEXT: buffer_atomic_cmpswap_b64 v[7:10], v6, s[0:3], null offen th:TH_ATOMIC_RETURN
19981995
; GFX12-NEXT: s_wait_loadcnt 0x0
@@ -2029,10 +2026,10 @@ define void @buffer_fat_ptr_agent_atomic_fmax_noret_f64__amdgpu_no_fine_grained_
20292026
; GFX11-NEXT: ; =>This Inner Loop Header: Depth=1
20302027
; GFX11-NEXT: s_waitcnt vmcnt(0)
20312028
; GFX11-NEXT: v_max_f64 v[0:1], v[2:3], v[2:3]
2029+
; GFX11-NEXT: v_dual_mov_b32 v10, v3 :: v_dual_mov_b32 v9, v2
20322030
; GFX11-NEXT: s_waitcnt_vscnt null, 0x0
2033-
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2)
2031+
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_1)
20342032
; GFX11-NEXT: v_max_f64 v[0:1], v[0:1], v[4:5]
2035-
; GFX11-NEXT: v_dual_mov_b32 v10, v3 :: v_dual_mov_b32 v9, v2
20362033
; GFX11-NEXT: v_dual_mov_b32 v8, v1 :: v_dual_mov_b32 v7, v0
20372034
; GFX11-NEXT: buffer_atomic_cmpswap_b64 v[7:10], v6, s[0:3], 0 offen glc
20382035
; GFX11-NEXT: s_waitcnt vmcnt(0)
@@ -2079,9 +2076,9 @@ define void @buffer_fat_ptr_agent_atomic_fmax_noret_f64__amdgpu_no_fine_grained_
20792076
; GFX908-NEXT: ; =>This Inner Loop Header: Depth=1
20802077
; GFX908-NEXT: s_waitcnt vmcnt(0)
20812078
; GFX908-NEXT: v_max_f64 v[0:1], v[2:3], v[2:3]
2082-
; GFX908-NEXT: v_max_f64 v[0:1], v[0:1], v[4:5]
20832079
; GFX908-NEXT: v_mov_b32_e32 v10, v3
20842080
; GFX908-NEXT: v_mov_b32_e32 v9, v2
2081+
; GFX908-NEXT: v_max_f64 v[0:1], v[0:1], v[4:5]
20852082
; GFX908-NEXT: v_mov_b32_e32 v8, v1
20862083
; GFX908-NEXT: v_mov_b32_e32 v7, v0
20872084
; GFX908-NEXT: buffer_atomic_cmpswap_x2 v[7:10], v6, s[16:19], 0 offen glc
@@ -2108,9 +2105,9 @@ define void @buffer_fat_ptr_agent_atomic_fmax_noret_f64__amdgpu_no_fine_grained_
21082105
; GFX8-NEXT: ; =>This Inner Loop Header: Depth=1
21092106
; GFX8-NEXT: s_waitcnt vmcnt(0)
21102107
; GFX8-NEXT: v_max_f64 v[0:1], v[2:3], v[2:3]
2111-
; GFX8-NEXT: v_max_f64 v[0:1], v[0:1], v[4:5]
21122108
; GFX8-NEXT: v_mov_b32_e32 v10, v3
21132109
; GFX8-NEXT: v_mov_b32_e32 v9, v2
2110+
; GFX8-NEXT: v_max_f64 v[0:1], v[0:1], v[4:5]
21142111
; GFX8-NEXT: v_mov_b32_e32 v8, v1
21152112
; GFX8-NEXT: v_mov_b32_e32 v7, v0
21162113
; GFX8-NEXT: buffer_atomic_cmpswap_x2 v[7:10], v6, s[16:19], 0 offen glc

‎llvm/test/CodeGen/AMDGPU/GlobalISel/atomicrmw_fmin.ll‎

Lines changed: 11 additions & 14 deletions
Original file line numberDiff line numberDiff line change
@@ -8,6 +8,7 @@
88
; RUN: llc -global-isel -mtriple=amdgcn-amd-amdpal -mcpu=tonga < %s | FileCheck -check-prefix=GFX8 %s
99
; RUN: llc -global-isel -mtriple=amdgcn-amd-amdpal -mcpu=hawaii < %s | FileCheck -check-prefix=GFX7 %s
1010

11+
1112
; TODO: Delete this and add run lines to use *-atomicrmw-fmin.ll tests
1213

1314
define float @local_atomic_fmin_ret_f32(ptr addrspace(3) %ptr, float %val) {
@@ -1823,10 +1824,8 @@ define double @buffer_fat_ptr_agent_atomic_fmin_ret_f64__amdgpu_no_fine_grained_
18231824
; GFX12-NEXT: s_wait_storecnt 0x0
18241825
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
18251826
; GFX12-NEXT: v_min_num_f64_e32 v[2:3], v[0:1], v[6:7]
1826-
; GFX12-NEXT: v_mov_b32_e32 v0, v2
1827-
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_2)
1828-
; GFX12-NEXT: v_dual_mov_b32 v1, v3 :: v_dual_mov_b32 v2, v4
1829-
; GFX12-NEXT: v_mov_b32_e32 v3, v5
1827+
; GFX12-NEXT: v_dual_mov_b32 v0, v2 :: v_dual_mov_b32 v1, v3
1828+
; GFX12-NEXT: v_dual_mov_b32 v2, v4 :: v_dual_mov_b32 v3, v5
18301829
; GFX12-NEXT: buffer_atomic_cmpswap_b64 v[0:3], v8, s[0:3], null offen th:TH_ATOMIC_RETURN
18311830
; GFX12-NEXT: s_wait_loadcnt 0x0
18321831
; GFX12-NEXT: global_inv scope:SCOPE_DEV
@@ -1865,10 +1864,8 @@ define double @buffer_fat_ptr_agent_atomic_fmin_ret_f64__amdgpu_no_fine_grained_
18651864
; GFX11-NEXT: s_waitcnt_vscnt null, 0x0
18661865
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(NEXT) | instid1(VALU_DEP_1)
18671866
; GFX11-NEXT: v_min_f64 v[2:3], v[0:1], v[6:7]
1868-
; GFX11-NEXT: v_mov_b32_e32 v0, v2
1869-
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_2)
1870-
; GFX11-NEXT: v_dual_mov_b32 v1, v3 :: v_dual_mov_b32 v2, v4
1871-
; GFX11-NEXT: v_mov_b32_e32 v3, v5
1867+
; GFX11-NEXT: v_dual_mov_b32 v0, v2 :: v_dual_mov_b32 v1, v3
1868+
; GFX11-NEXT: v_dual_mov_b32 v2, v4 :: v_dual_mov_b32 v3, v5
18721869
; GFX11-NEXT: buffer_atomic_cmpswap_b64 v[0:3], v8, s[0:3], 0 offen glc
18731870
; GFX11-NEXT: s_waitcnt vmcnt(0)
18741871
; GFX11-NEXT: buffer_gl1_inv
@@ -1989,10 +1986,10 @@ define void @buffer_fat_ptr_agent_atomic_fmin_noret_f64__amdgpu_no_fine_grained_
19891986
; GFX12-NEXT: ; =>This Inner Loop Header: Depth=1
19901987
; GFX12-NEXT: s_wait_loadcnt 0x0
19911988
; GFX12-NEXT: v_max_num_f64_e32 v[0:1], v[2:3], v[2:3]
1989+
; GFX12-NEXT: v_dual_mov_b32 v10, v3 :: v_dual_mov_b32 v9, v2
19921990
; GFX12-NEXT: s_wait_storecnt 0x0
1993-
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2)
1991+
; GFX12-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_1)
19941992
; GFX12-NEXT: v_min_num_f64_e32 v[0:1], v[0:1], v[4:5]
1995-
; GFX12-NEXT: v_dual_mov_b32 v10, v3 :: v_dual_mov_b32 v9, v2
19961993
; GFX12-NEXT: v_dual_mov_b32 v8, v1 :: v_dual_mov_b32 v7, v0
19971994
; GFX12-NEXT: buffer_atomic_cmpswap_b64 v[7:10], v6, s[0:3], null offen th:TH_ATOMIC_RETURN
19981995
; GFX12-NEXT: s_wait_loadcnt 0x0
@@ -2029,10 +2026,10 @@ define void @buffer_fat_ptr_agent_atomic_fmin_noret_f64__amdgpu_no_fine_grained_
20292026
; GFX11-NEXT: ; =>This Inner Loop Header: Depth=1
20302027
; GFX11-NEXT: s_waitcnt vmcnt(0)
20312028
; GFX11-NEXT: v_max_f64 v[0:1], v[2:3], v[2:3]
2029+
; GFX11-NEXT: v_dual_mov_b32 v10, v3 :: v_dual_mov_b32 v9, v2
20322030
; GFX11-NEXT: s_waitcnt_vscnt null, 0x0
2033-
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_1) | instskip(SKIP_1) | instid1(VALU_DEP_2)
2031+
; GFX11-NEXT: s_delay_alu instid0(VALU_DEP_2) | instskip(NEXT) | instid1(VALU_DEP_1)
20342032
; GFX11-NEXT: v_min_f64 v[0:1], v[0:1], v[4:5]
2035-
; GFX11-NEXT: v_dual_mov_b32 v10, v3 :: v_dual_mov_b32 v9, v2
20362033
; GFX11-NEXT: v_dual_mov_b32 v8, v1 :: v_dual_mov_b32 v7, v0
20372034
; GFX11-NEXT: buffer_atomic_cmpswap_b64 v[7:10], v6, s[0:3], 0 offen glc
20382035
; GFX11-NEXT: s_waitcnt vmcnt(0)
@@ -2079,9 +2076,9 @@ define void @buffer_fat_ptr_agent_atomic_fmin_noret_f64__amdgpu_no_fine_grained_
20792076
; GFX908-NEXT: ; =>This Inner Loop Header: Depth=1
20802077
; GFX908-NEXT: s_waitcnt vmcnt(0)
20812078
; GFX908-NEXT: v_max_f64 v[0:1], v[2:3], v[2:3]
2082-
; GFX908-NEXT: v_min_f64 v[0:1], v[0:1], v[4:5]
20832079
; GFX908-NEXT: v_mov_b32_e32 v10, v3
20842080
; GFX908-NEXT: v_mov_b32_e32 v9, v2
2081+
; GFX908-NEXT: v_min_f64 v[0:1], v[0:1], v[4:5]
20852082
; GFX908-NEXT: v_mov_b32_e32 v8, v1
20862083
; GFX908-NEXT: v_mov_b32_e32 v7, v0
20872084
; GFX908-NEXT: buffer_atomic_cmpswap_x2 v[7:10], v6, s[16:19], 0 offen glc
@@ -2108,9 +2105,9 @@ define void @buffer_fat_ptr_agent_atomic_fmin_noret_f64__amdgpu_no_fine_grained_
21082105
; GFX8-NEXT: ; =>This Inner Loop Header: Depth=1
21092106
; GFX8-NEXT: s_waitcnt vmcnt(0)
21102107
; GFX8-NEXT: v_max_f64 v[0:1], v[2:3], v[2:3]
2111-
; GFX8-NEXT: v_min_f64 v[0:1], v[0:1], v[4:5]
21122108
; GFX8-NEXT: v_mov_b32_e32 v10, v3
21132109
; GFX8-NEXT: v_mov_b32_e32 v9, v2
2110+
; GFX8-NEXT: v_min_f64 v[0:1], v[0:1], v[4:5]
21142111
; GFX8-NEXT: v_mov_b32_e32 v8, v1
21152112
; GFX8-NEXT: v_mov_b32_e32 v7, v0
21162113
; GFX8-NEXT: buffer_atomic_cmpswap_x2 v[7:10], v6, s[16:19], 0 offen glc

0 commit comments

Comments
 (0)