[llvm] e844f24 - AMDGPU: Use aligned vgprs/agprs in gfx90a mir tests
Matt Arsenault via llvm-commits
llvm-commits at lists.llvm.org
Tue Feb 23 13:54:49 PST 2021
Author: Matt Arsenault
Date: 2021-02-23T16:46:22-05:00
New Revision: e844f24a278bf4b3622fb969d06e7b771a6a2aca
URL: https://github.com/llvm/llvm-project/commit/e844f24a278bf4b3622fb969d06e7b771a6a2aca
DIFF: https://github.com/llvm/llvm-project/commit/e844f24a278bf4b3622fb969d06e7b771a6a2aca.diff
LOG: AMDGPU: Use aligned vgprs/agprs in gfx90a mir tests
These would fail a verifier check in a future change.
Added:
Modified:
llvm/test/CodeGen/AMDGPU/mai-hazards-gfx90a.mir
llvm/test/CodeGen/AMDGPU/waitcnt-agpr.mir
Removed:
################################################################################
diff --git a/llvm/test/CodeGen/AMDGPU/mai-hazards-gfx90a.mir b/llvm/test/CodeGen/AMDGPU/mai-hazards-gfx90a.mir
index 305d30fd1afe..435529d1d2cb 100644
--- a/llvm/test/CodeGen/AMDGPU/mai-hazards-gfx90a.mir
+++ b/llvm/test/CodeGen/AMDGPU/mai-hazards-gfx90a.mir
@@ -464,8 +464,8 @@ body: |
name: dmfma4x4_write_vgpr_flat_read_overlap
body: |
bb.0:
- $vgpr3_vgpr4 = V_MFMA_F64_4X4X4F64_vgprcd_e64 $vgpr0_vgpr1, $vgpr0_vgpr1, $vgpr0_vgpr1, 0, 0, 0, implicit $mode, implicit $exec
- FLAT_STORE_DWORD $vgpr0_vgpr1, $vgpr4, 0, 0, 0, 0, 0, implicit $mode, implicit $exec, implicit $flat_scr
+ $vgpr4_vgpr5 = V_MFMA_F64_4X4X4F64_vgprcd_e64 $vgpr0_vgpr1, $vgpr0_vgpr1, $vgpr0_vgpr1, 0, 0, 0, implicit $mode, implicit $exec
+ FLAT_STORE_DWORD $vgpr0_vgpr1, $vgpr5, 0, 0, 0, 0, 0, implicit $mode, implicit $exec, implicit $flat_scr
...
# GCN-LABEL: name: dmfma4x4_write_vgpr_flat_read_full
# GCN: V_MFMA
@@ -475,8 +475,8 @@ body: |
name: dmfma4x4_write_vgpr_flat_read_full
body: |
bb.0:
- $vgpr3_vgpr4 = V_MFMA_F64_4X4X4F64_vgprcd_e64 $vgpr0_vgpr1, $vgpr0_vgpr1, $vgpr0_vgpr1, 0, 0, 0, implicit $mode, implicit $exec
- FLAT_STORE_DWORDX2 $vgpr0_vgpr1, $vgpr3_vgpr4, 0, 0, 0, 0, 0, implicit $mode, implicit $exec, implicit $flat_scr
+ $vgpr4_vgpr5 = V_MFMA_F64_4X4X4F64_vgprcd_e64 $vgpr0_vgpr1, $vgpr0_vgpr1, $vgpr0_vgpr1, 0, 0, 0, implicit $mode, implicit $exec
+ FLAT_STORE_DWORDX2 $vgpr0_vgpr1, $vgpr4_vgpr5, 0, 0, 0, 0, 0, implicit $mode, implicit $exec, implicit $flat_scr
...
# GCN-LABEL: name: dmfma16x16_write_vgpr_flat_read
# GCN: V_MFMA
@@ -530,8 +530,8 @@ body: |
name: dmfma4x4_write_vgpr_valu_read
body: |
bb.0:
- $vgpr3_vgpr4 = V_MFMA_F64_4X4X4F64_vgprcd_e64 $vgpr0_vgpr1, $vgpr0_vgpr1, $vgpr0_vgpr1, 0, 0, 0, implicit $mode, implicit $exec
- $vgpr5 = V_MOV_B32_e32 $vgpr4, implicit $exec
+ $vgpr4_vgpr5 = V_MFMA_F64_4X4X4F64_vgprcd_e64 $vgpr0_vgpr1, $vgpr0_vgpr1, $vgpr0_vgpr1, 0, 0, 0, implicit $mode, implicit $exec
+ $vgpr6 = V_MOV_B32_e32 $vgpr5, implicit $exec
...
# GCN-LABEL: name: dmfma16x16_write_vgpr_valu_read
# GCN: V_MFMA
@@ -594,8 +594,8 @@ body: |
name: dmfma4x4_write_vgpr_dot_read
body: |
bb.0:
- $vgpr3_vgpr4 = V_MFMA_F64_4X4X4F64_vgprcd_e64 $vgpr0_vgpr1, $vgpr0_vgpr1, $vgpr0_vgpr1, 0, 0, 0, implicit $mode, implicit $exec
- $vgpr1 = V_DOT4C_I32_I8_e32 $vgpr0, $vgpr4, $vgpr1, implicit $exec
+ $vgpr4_vgpr5 = V_MFMA_F64_4X4X4F64_vgprcd_e64 $vgpr0_vgpr1, $vgpr0_vgpr1, $vgpr0_vgpr1, 0, 0, 0, implicit $mode, implicit $exec
+ $vgpr1 = V_DOT4C_I32_I8_e32 $vgpr0, $vgpr5, $vgpr1, implicit $exec
...
# GCN-LABEL: name: dmfma16x16_write_vgpr_dot_read
# GCN: V_MFMA
@@ -714,8 +714,8 @@ body: |
name: dmfma4x4_write_vgpr_valu_write
body: |
bb.0:
- $vgpr3_vgpr4 = V_MFMA_F64_4X4X4F64_vgprcd_e64 $vgpr0_vgpr1, $vgpr0_vgpr1, $vgpr0_vgpr1, 0, 0, 0, implicit $mode, implicit $exec
- $vgpr3 = V_MOV_B32_e32 0, implicit $mode, implicit $exec
+ $vgpr4_vgpr5 = V_MFMA_F64_4X4X4F64_vgprcd_e64 $vgpr0_vgpr1, $vgpr0_vgpr1, $vgpr0_vgpr1, 0, 0, 0, implicit $mode, implicit $exec
+ $vgpr4 = V_MOV_B32_e32 0, implicit $mode, implicit $exec
...
# GCN-LABEL: name: dmfma16x16_write_vgpr_valu_write
# GCN: V_MFMA
@@ -839,7 +839,7 @@ body: |
name: dmfma4x4_read_srcc_vgpr_valu_write
body: |
bb.0:
- $vgpr3_vgpr4 = V_MFMA_F64_4X4X4F64_vgprcd_e64 $vgpr0_vgpr1, $vgpr0_vgpr1, $vgpr0_vgpr1, 0, 0, 0, implicit $mode, implicit $exec
+ $vgpr4_vgpr5 = V_MFMA_F64_4X4X4F64_vgprcd_e64 $vgpr0_vgpr1, $vgpr0_vgpr1, $vgpr0_vgpr1, 0, 0, 0, implicit $mode, implicit $exec
$vgpr1 = V_MOV_B32_e32 0, implicit $mode, implicit $exec
...
# GCN-LABEL: name: dmfma16x16_read_srcc_vgpr_valu_write
@@ -1202,8 +1202,8 @@ body: |
name: dmfma4x4_write_agpr_flat_read_overlap
body: |
bb.0:
- $agpr3_agpr4 = V_MFMA_F64_4X4X4F64_e64 $vgpr0_vgpr1, $vgpr0_vgpr1, $agpr0_agpr1, 0, 0, 0, implicit $mode, implicit $exec
- FLAT_STORE_DWORD $vgpr0_vgpr1, $agpr4, 0, 0, 0, 0, 0, implicit $mode, implicit $exec, implicit $flat_scr
+ $agpr4_agpr5 = V_MFMA_F64_4X4X4F64_e64 $vgpr0_vgpr1, $vgpr0_vgpr1, $agpr0_agpr1, 0, 0, 0, implicit $mode, implicit $exec
+ FLAT_STORE_DWORD $vgpr0_vgpr1, $agpr5, 0, 0, 0, 0, 0, implicit $mode, implicit $exec, implicit $flat_scr
...
# GCN-LABEL: name: dmfma4x4_write_agpr_flat_read_full
# GCN: V_MFMA
@@ -1213,8 +1213,8 @@ body: |
name: dmfma4x4_write_agpr_flat_read_full
body: |
bb.0:
- $agpr3_agpr4 = V_MFMA_F64_4X4X4F64_e64 $vgpr0_vgpr1, $vgpr0_vgpr1, $agpr0_agpr1, 0, 0, 0, implicit $mode, implicit $exec
- FLAT_STORE_DWORDX2 $vgpr0_vgpr1, $agpr3_agpr4, 0, 0, 0, 0, 0, implicit $mode, implicit $exec, implicit $flat_scr
+ $agpr4_agpr5 = V_MFMA_F64_4X4X4F64_e64 $vgpr0_vgpr1, $vgpr0_vgpr1, $agpr0_agpr1, 0, 0, 0, implicit $mode, implicit $exec
+ FLAT_STORE_DWORDX2 $vgpr0_vgpr1, $agpr4_agpr5, 0, 0, 0, 0, 0, implicit $mode, implicit $exec, implicit $flat_scr
...
# GCN-LABEL: name: dmfma16x16_write_agpr_flat_read
# GCN: V_MFMA
@@ -1235,8 +1235,8 @@ body: |
name: dmfma4x4_write_agpr_valu_read
body: |
bb.0:
- $agpr3_agpr4 = V_MFMA_F64_4X4X4F64_e64 $agpr0_agpr1, $vgpr0_vgpr1, $agpr0_agpr1, 0, 0, 0, implicit $mode, implicit $exec
- $vgpr5 = V_ACCVGPR_READ_B32_e64 $agpr4, implicit $exec
+ $agpr4_agpr5 = V_MFMA_F64_4X4X4F64_e64 $agpr0_agpr1, $vgpr0_vgpr1, $agpr0_agpr1, 0, 0, 0, implicit $mode, implicit $exec
+ $vgpr5 = V_ACCVGPR_READ_B32_e64 $agpr5, implicit $exec
...
# GCN-LABEL: name: dmfma16x16_write_agpr_valu_read
# GCN: V_MFMA
@@ -1256,8 +1256,8 @@ body: |
name: dmfma4x4_write_agpr_valu_write
body: |
bb.0:
- $agpr3_agpr4 = V_MFMA_F64_4X4X4F64_e64 $vgpr0_vgpr1, $vgpr0_vgpr1, $agpr0_agpr1, 0, 0, 0, implicit $mode, implicit $exec
- $agpr3 = V_ACCVGPR_WRITE_B32_e64 0, implicit $mode, implicit $exec
+ $agpr4_agpr5 = V_MFMA_F64_4X4X4F64_e64 $vgpr0_vgpr1, $vgpr0_vgpr1, $agpr0_agpr1, 0, 0, 0, implicit $mode, implicit $exec
+ $agpr4 = V_ACCVGPR_WRITE_B32_e64 0, implicit $mode, implicit $exec
...
# GCN-LABEL: name: dmfma16x16_write_agpr_valu_write
# GCN: V_MFMA
@@ -1276,7 +1276,7 @@ body: |
name: dmfma4x4_read_srcc_agpr_valu_write
body: |
bb.0:
- $agpr3_agpr4 = V_MFMA_F64_4X4X4F64_e64 $vgpr0_vgpr1, $vgpr0_vgpr1, $agpr0_agpr1, 0, 0, 0, implicit $mode, implicit $exec
+ $agpr4_agpr5 = V_MFMA_F64_4X4X4F64_e64 $vgpr0_vgpr1, $vgpr0_vgpr1, $agpr0_agpr1, 0, 0, 0, implicit $mode, implicit $exec
$agpr1 = V_ACCVGPR_WRITE_B32_e64 0, implicit $mode, implicit $exec
...
# GCN-LABEL: name: dmfma16x16_read_srcc_agpr_valu_write
diff --git a/llvm/test/CodeGen/AMDGPU/waitcnt-agpr.mir b/llvm/test/CodeGen/AMDGPU/waitcnt-agpr.mir
index 0b2f02e79f37..a5043423840d 100644
--- a/llvm/test/CodeGen/AMDGPU/waitcnt-agpr.mir
+++ b/llvm/test/CodeGen/AMDGPU/waitcnt-agpr.mir
@@ -59,42 +59,42 @@ body: |
; GCN: bb.0:
; GCN: successors: %bb.1(0x80000000)
; GCN: S_WAITCNT 0
- ; GCN: $agpr0 = FLAT_LOAD_DWORD $vgpr1_vgpr2, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr :: (load 4 from %ir.global4, addrspace 1)
- ; GCN: $agpr3_agpr4_agpr5_agpr6 = FLAT_LOAD_DWORDX4 $vgpr7_vgpr8, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr :: (load 16 from %ir.global16, addrspace 1)
+ ; GCN: $agpr0 = FLAT_LOAD_DWORD $vgpr2_vgpr3, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr :: (load 4 from %ir.global4, addrspace 1)
+ ; GCN: $agpr4_agpr5_agpr6_agpr7 = FLAT_LOAD_DWORDX4 $vgpr8_vgpr9, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr :: (load 16 from %ir.global16, addrspace 1)
; GCN: S_WAITCNT 3953
; GCN: $agpr0 = V_ACCVGPR_MOV_B32 $agpr1, implicit $exec
; GCN: S_BRANCH %bb.1
; GCN: bb.1:
; GCN: successors: %bb.2(0x80000000)
- ; GCN: $agpr0 = FLAT_LOAD_DWORD $vgpr1_vgpr2, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
+ ; GCN: $agpr0 = FLAT_LOAD_DWORD $vgpr2_vgpr3, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
; GCN: S_WAITCNT 3952
- ; GCN: $agpr3_agpr4_agpr5_agpr6 = FLAT_LOAD_DWORDX4 $vgpr7_vgpr8, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr :: (load 16 from %ir.global16, addrspace 1)
+ ; GCN: $agpr4_agpr5_agpr6_agpr7 = FLAT_LOAD_DWORDX4 $vgpr8_vgpr9, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr :: (load 16 from %ir.global16, addrspace 1)
; GCN: $vgpr0 = V_ACCVGPR_READ_B32_e64 $agpr1, implicit $exec
; GCN: S_BRANCH %bb.2
; GCN: bb.2:
; GCN: S_WAITCNT 49279
- ; GCN: $agpr0 = FLAT_LOAD_DWORD $vgpr1_vgpr2, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr :: (load 4 from %ir.flat4)
+ ; GCN: $agpr0 = FLAT_LOAD_DWORD $vgpr2_vgpr3, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr :: (load 4 from %ir.flat4)
; GCN: S_WAITCNT 3952
- ; GCN: $agpr3_agpr4_agpr5_agpr6 = FLAT_LOAD_DWORDX4 $vgpr7_vgpr8, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr :: (load 16 from %ir.flat16)
+ ; GCN: $agpr4_agpr5_agpr6_agpr7 = FLAT_LOAD_DWORDX4 $vgpr8_vgpr9, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr :: (load 16 from %ir.flat16)
; GCN: $vgpr0 = V_ACCVGPR_READ_B32_e64 $agpr1, implicit $exec
; GCN: S_ENDPGM 0
bb.0:
successors: %bb.1
- $agpr0 = FLAT_LOAD_DWORD $vgpr1_vgpr2, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr :: (load 4 from %ir.global4)
- $agpr3_agpr4_agpr5_agpr6 = FLAT_LOAD_DWORDX4 $vgpr7_vgpr8, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr :: (load 16 from %ir.global16)
+ $agpr0 = FLAT_LOAD_DWORD $vgpr2_vgpr3, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr :: (load 4 from %ir.global4)
+ $agpr4_agpr5_agpr6_agpr7 = FLAT_LOAD_DWORDX4 $vgpr8_vgpr9, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr :: (load 16 from %ir.global16)
$agpr0 = V_ACCVGPR_MOV_B32 $agpr1, implicit $exec
S_BRANCH %bb.1
bb.1:
successors: %bb.2
- $agpr0 = FLAT_LOAD_DWORD $vgpr1_vgpr2, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
- $agpr3_agpr4_agpr5_agpr6 = FLAT_LOAD_DWORDX4 $vgpr7_vgpr8, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr :: (load 16 from %ir.global16)
+ $agpr0 = FLAT_LOAD_DWORD $vgpr2_vgpr3, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
+ $agpr4_agpr5_agpr6_agpr7 = FLAT_LOAD_DWORDX4 $vgpr8_vgpr9, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr :: (load 16 from %ir.global16)
$vgpr0 = V_ACCVGPR_READ_B32_e64 $agpr1, implicit $exec
S_BRANCH %bb.2
bb.2:
- $agpr0 = FLAT_LOAD_DWORD $vgpr1_vgpr2, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr :: (load 4 from %ir.flat4)
- $agpr3_agpr4_agpr5_agpr6 = FLAT_LOAD_DWORDX4 $vgpr7_vgpr8, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr :: (load 16 from %ir.flat16)
+ $agpr0 = FLAT_LOAD_DWORD $vgpr2_vgpr3, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr :: (load 4 from %ir.flat4)
+ $agpr4_agpr5_agpr6_agpr7 = FLAT_LOAD_DWORDX4 $vgpr8_vgpr9, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr :: (load 16 from %ir.flat16)
$vgpr0 = V_ACCVGPR_READ_B32_e64 $agpr1, implicit $exec
S_ENDPGM 0
...
@@ -110,19 +110,19 @@ body: |
; GCN: bb.0:
; GCN: successors: %bb.1(0x80000000)
; GCN: S_WAITCNT 0
- ; GCN: $agpr0 = FLAT_LOAD_DWORD $vgpr1_vgpr2, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
+ ; GCN: $agpr0 = FLAT_LOAD_DWORD $vgpr2_vgpr3, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
; GCN: bb.1:
- ; GCN: $vgpr3_vgpr4 = V_LSHLREV_B64_e64 4, $vgpr7_vgpr8, implicit $exec
+ ; GCN: $vgpr4_vgpr5 = V_LSHLREV_B64_e64 4, $vgpr8_vgpr9, implicit $exec
; GCN: S_WAITCNT 112
- ; GCN: FLAT_STORE_DWORD $vgpr3_vgpr4, $agpr0, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
+ ; GCN: FLAT_STORE_DWORD $vgpr4_vgpr5, $agpr0, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
; GCN: S_ENDPGM 0
bb.0:
successors: %bb.1
- $agpr0 = FLAT_LOAD_DWORD $vgpr1_vgpr2, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
+ $agpr0 = FLAT_LOAD_DWORD $vgpr2_vgpr3, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
bb.1:
- $vgpr3_vgpr4 = V_LSHLREV_B64_e64 4, $vgpr7_vgpr8, implicit $exec
- FLAT_STORE_DWORD $vgpr3_vgpr4, $agpr0, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
+ $vgpr4_vgpr5 = V_LSHLREV_B64_e64 4, $vgpr8_vgpr9, implicit $exec
+ FLAT_STORE_DWORD $vgpr4_vgpr5, $agpr0, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
S_ENDPGM 0
...
---
@@ -139,19 +139,19 @@ body: |
; GCN: bb.0:
; GCN: successors: %bb.2(0x80000000)
; GCN: S_WAITCNT 0
- ; GCN: $agpr0 = FLAT_LOAD_DWORD $vgpr1_vgpr2, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
+ ; GCN: $agpr0 = FLAT_LOAD_DWORD $vgpr2_vgpr3, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
; GCN: S_BRANCH %bb.2
; GCN: bb.1:
; GCN: FLAT_STORE_DWORD $vgpr8_vgpr9, $agpr10, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
; GCN: S_ENDPGM 0
; GCN: bb.2:
- ; GCN: $vgpr3_vgpr4 = V_LSHLREV_B64_e64 4, $vgpr7_vgpr8, implicit $exec
+ ; GCN: $vgpr4_vgpr5 = V_LSHLREV_B64_e64 4, $vgpr8_vgpr9, implicit $exec
; GCN: S_WAITCNT 112
- ; GCN: FLAT_STORE_DWORD $vgpr3_vgpr4, $agpr0, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
+ ; GCN: FLAT_STORE_DWORD $vgpr4_vgpr5, $agpr0, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
; GCN: S_ENDPGM 0
bb.0:
successors: %bb.2
- $agpr0 = FLAT_LOAD_DWORD $vgpr1_vgpr2, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
+ $agpr0 = FLAT_LOAD_DWORD $vgpr2_vgpr3, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
S_BRANCH %bb.2
bb.1:
@@ -159,8 +159,8 @@ body: |
S_ENDPGM 0
bb.2:
- $vgpr3_vgpr4 = V_LSHLREV_B64_e64 4, $vgpr7_vgpr8, implicit $exec
- FLAT_STORE_DWORD $vgpr3_vgpr4, $agpr0, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
+ $vgpr4_vgpr5 = V_LSHLREV_B64_e64 4, $vgpr8_vgpr9, implicit $exec
+ FLAT_STORE_DWORD $vgpr4_vgpr5, $agpr0, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
S_ENDPGM 0
...
@@ -175,9 +175,9 @@ machineFunctionInfo:
body: |
bb.0:
liveins: $vgpr1_vgpr2
- $agpr0 = FLAT_LOAD_DWORD $vgpr1_vgpr2, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
+ $agpr0 = FLAT_LOAD_DWORD $vgpr2_vgpr3, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
S_WAITCNT 0
- FLAT_STORE_DWORD $vgpr1_vgpr2, $agpr0, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
+ FLAT_STORE_DWORD $vgpr2_vgpr3, $agpr0, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
...
@@ -192,19 +192,19 @@ body: |
liveins: $vgpr1_vgpr2
; GCN-LABEL: name: bundle_no_waitcnt
; GCN: liveins: $vgpr1_vgpr2
- ; GCN: $agpr0 = FLAT_LOAD_DWORD $vgpr1_vgpr2, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
+ ; GCN: $agpr0 = FLAT_LOAD_DWORD $vgpr2_vgpr3, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
; GCN: BUNDLE {
; GCN: S_NOP 0
; GCN: S_NOP 0
; GCN: }
; GCN: S_WAITCNT 112
- ; GCN: FLAT_STORE_DWORD $vgpr1_vgpr2, $agpr0, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
- $agpr0 = FLAT_LOAD_DWORD $vgpr1_vgpr2, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
+ ; GCN: FLAT_STORE_DWORD $vgpr2_vgpr3, $agpr0, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
+ $agpr0 = FLAT_LOAD_DWORD $vgpr2_vgpr3, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
BUNDLE {
S_NOP 0
S_NOP 0
}
- FLAT_STORE_DWORD $vgpr1_vgpr2, $agpr0, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
+ FLAT_STORE_DWORD $vgpr2_vgpr3, $agpr0, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
...
@@ -220,18 +220,18 @@ body: |
liveins: $vgpr1_vgpr2
; GCN-LABEL: name: preexisting_waitcnt_in_bundle
; GCN: liveins: $vgpr1_vgpr2
- ; GCN: $agpr0 = FLAT_LOAD_DWORD $vgpr1_vgpr2, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
+ ; GCN: $agpr0 = FLAT_LOAD_DWORD $vgpr2_vgpr3, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
; GCN: BUNDLE {
; GCN: S_NOP 0
; GCN: S_WAITCNT 0
; GCN: }
- ; GCN: FLAT_STORE_DWORD $vgpr1_vgpr2, $agpr0, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
- $agpr0 = FLAT_LOAD_DWORD $vgpr1_vgpr2, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
+ ; GCN: FLAT_STORE_DWORD $vgpr2_vgpr3, $agpr0, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
+ $agpr0 = FLAT_LOAD_DWORD $vgpr2_vgpr3, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
BUNDLE {
S_NOP 0
S_WAITCNT 0
}
- FLAT_STORE_DWORD $vgpr1_vgpr2, $agpr0, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
+ FLAT_STORE_DWORD $vgpr2_vgpr3, $agpr0, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
...
@@ -248,14 +248,14 @@ body: |
liveins: $vgpr1_vgpr2
; GCN-LABEL: name: insert_in_bundle
; GCN: liveins: $vgpr1_vgpr2
- ; GCN: BUNDLE implicit-def $agpr0, implicit $vgpr1_vgpr2 {
- ; GCN: $agpr0 = FLAT_LOAD_DWORD $vgpr1_vgpr2, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
+ ; GCN: BUNDLE implicit-def $agpr0, implicit $vgpr2_vgpr3 {
+ ; GCN: $agpr0 = FLAT_LOAD_DWORD $vgpr2_vgpr3, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
; GCN: S_WAITCNT 112
- ; GCN: FLAT_STORE_DWORD $vgpr1_vgpr2, internal $agpr0, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
+ ; GCN: FLAT_STORE_DWORD $vgpr2_vgpr3, internal $agpr0, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
; GCN: }
- BUNDLE implicit-def $agpr0, implicit $vgpr1_vgpr2 {
- $agpr0 = FLAT_LOAD_DWORD $vgpr1_vgpr2, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
- FLAT_STORE_DWORD $vgpr1_vgpr2, internal $agpr0, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
+ BUNDLE implicit-def $agpr0, implicit $vgpr2_vgpr3 {
+ $agpr0 = FLAT_LOAD_DWORD $vgpr2_vgpr3, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
+ FLAT_STORE_DWORD $vgpr2_vgpr3, internal $agpr0, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
}
...
@@ -273,16 +273,16 @@ body: |
liveins: $vgpr1_vgpr2
; GCN-LABEL: name: exit_bundle
; GCN: liveins: $vgpr1_vgpr2
- ; GCN: BUNDLE implicit-def $agpr0, implicit $vgpr1_vgpr2 {
- ; GCN: $agpr0 = FLAT_LOAD_DWORD $vgpr1_vgpr2, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
+ ; GCN: BUNDLE implicit-def $agpr0, implicit $vgpr2_vgpr3 {
+ ; GCN: $agpr0 = FLAT_LOAD_DWORD $vgpr2_vgpr3, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
; GCN: }
; GCN: S_WAITCNT 112
- ; GCN: FLAT_STORE_DWORD $vgpr1_vgpr2, $agpr0, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
- BUNDLE implicit-def $agpr0, implicit $vgpr1_vgpr2 {
- $agpr0 = FLAT_LOAD_DWORD $vgpr1_vgpr2, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
+ ; GCN: FLAT_STORE_DWORD $vgpr2_vgpr3, $agpr0, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
+ BUNDLE implicit-def $agpr0, implicit $vgpr2_vgpr3 {
+ $agpr0 = FLAT_LOAD_DWORD $vgpr2_vgpr3, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
}
- FLAT_STORE_DWORD $vgpr1_vgpr2, $agpr0, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
+ FLAT_STORE_DWORD $vgpr2_vgpr3, $agpr0, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
...
@@ -300,17 +300,17 @@ body: |
liveins: $vgpr1_vgpr2
; GCN-LABEL: name: cross_bundle
; GCN: liveins: $vgpr1_vgpr2
- ; GCN: BUNDLE implicit-def $agpr0, implicit $vgpr1_vgpr2 {
- ; GCN: $agpr0 = FLAT_LOAD_DWORD $vgpr1_vgpr2, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
+ ; GCN: BUNDLE implicit-def $agpr0, implicit $vgpr2_vgpr3 {
+ ; GCN: $agpr0 = FLAT_LOAD_DWORD $vgpr2_vgpr3, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
; GCN: }
; GCN: S_WAITCNT 112
- ; GCN: BUNDLE implicit $agpr0, implicit $vgpr1_vgpr2 {
- ; GCN: FLAT_STORE_DWORD $vgpr1_vgpr2, $agpr0, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
+ ; GCN: BUNDLE implicit $agpr0, implicit $vgpr2_vgpr3 {
+ ; GCN: FLAT_STORE_DWORD $vgpr2_vgpr3, $agpr0, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
; GCN: }
- BUNDLE implicit-def $agpr0, implicit $vgpr1_vgpr2 {
- $agpr0 = FLAT_LOAD_DWORD $vgpr1_vgpr2, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
+ BUNDLE implicit-def $agpr0, implicit $vgpr2_vgpr3 {
+ $agpr0 = FLAT_LOAD_DWORD $vgpr2_vgpr3, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
}
- BUNDLE implicit $agpr0, implicit $vgpr1_vgpr2 {
- FLAT_STORE_DWORD $vgpr1_vgpr2, $agpr0, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
+ BUNDLE implicit $agpr0, implicit $vgpr2_vgpr3 {
+ FLAT_STORE_DWORD $vgpr2_vgpr3, $agpr0, 0, 0, 0, 0, 0, implicit $exec, implicit $flat_scr
}
...
More information about the llvm-commits
mailing list