[AMDGPU] Update tests for last-use in global/scratch/flat/buffer load…

[mariusz-sikora-at-amd]

… instructions

[llvmbot]

@llvm/pr-subscribers-backend-amdgpu

Author: Mariusz Sikora (mariusz-sikora-at-amd)

Changes

… instructions

---

Full diff: https://github.com/llvm/llvm-project/pull/94975.diff

4 Files Affected:

• (added) llvm/test/CodeGen/AMDGPU/llvm.amdgcn.buffer.load-last-use.ll (+27)
• (modified) llvm/test/CodeGen/AMDGPU/memory-legalizer-flat-lastuse.ll (+2-5)
• (modified) llvm/test/CodeGen/AMDGPU/memory-legalizer-global-lastuse.ll (+2-5)
• (modified) llvm/test/CodeGen/AMDGPU/memory-legalizer-private-lastuse.ll (+2-5)

diff --git a/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.buffer.load-last-use.ll b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.buffer.load-last-use.ll
new file mode 100644
index 0000000000000..e19db7f7eb15f
--- /dev/null
+++ b/llvm/test/CodeGen/AMDGPU/llvm.amdgcn.buffer.load-last-use.ll
@@ -0,0 +1,27 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+;RUN: llc < %s -mtriple=amdgcn -mcpu=gfx1200 -amdgpu-enable-delay-alu=0 -verify-machineinstrs | FileCheck %s --check-prefix=GCN
+;RUN: llc < %s -global-isel -mtriple=amdgcn -mcpu=gfx1200 -amdgpu-enable-delay-alu=0 -verify-machineinstrs | FileCheck %s --check-prefix=GCN
+
+define amdgpu_ps float @raw_buffer_load(<4 x i32> inreg) {
+; GCN-LABEL: raw_buffer_load:
+; GCN:       ; %bb.0: ; %main_body
+; GCN-NEXT:    buffer_load_b32 v0, off, s[0:3], null th:TH_LOAD_LU
+; GCN-NEXT:    s_wait_loadcnt 0x0
+; GCN-NEXT:    ; return to shader part epilog
+main_body:
+  %data = call float @llvm.amdgcn.raw.buffer.load.f32(<4 x i32> %0, i32 0, i32 0, i32 3)
+  ret float %data
+}
+
+define amdgpu_ps float @struct_buffer_load(<4 x i32> inreg) {
+; GCN-LABEL: struct_buffer_load:
+; GCN:       ; %bb.0: ; %main_body
+; GCN-NEXT:    v_mov_b32_e32 v0, 0
+; GCN-NEXT:    buffer_load_b32 v0, v0, s[0:3], null idxen th:TH_LOAD_LU
+; GCN-NEXT:    s_wait_loadcnt 0x0
+; GCN-NEXT:    ; return to shader part epilog
+main_body:
+  %data = call float @llvm.amdgcn.struct.buffer.load.f32(<4 x i32> %0, i32 0, i32 0, i32 0, i32 3)
+  ret float %data
+}
+
diff --git a/llvm/test/CodeGen/AMDGPU/memory-legalizer-flat-lastuse.ll b/llvm/test/CodeGen/AMDGPU/memory-legalizer-flat-lastuse.ll
index e7c6044b3fb6b..fb40274cac1ba 100644
--- a/llvm/test/CodeGen/AMDGPU/memory-legalizer-flat-lastuse.ll
+++ b/llvm/test/CodeGen/AMDGPU/memory-legalizer-flat-lastuse.ll
@@ -1,6 +1,6 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; RUN: llc -mtriple=amdgcn-amd-amdhsa -O0 -mcpu=gfx1200 < %s | FileCheck --check-prefixes=GFX12,GFX12-WGP %s
-; RUN: llc -mtriple=amdgcn-amd-amdhsa -O0 -mcpu=gfx1200 -mattr=+cumode < %s | FileCheck --check-prefixes=GFX12,GFX12-CU %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -O0 -mcpu=gfx1200 < %s | FileCheck --check-prefix=GFX12 %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -O0 -mcpu=gfx1200 -mattr=+cumode < %s | FileCheck --check-prefix=GFX12 %s
 
 define amdgpu_kernel void @flat_last_use_load_0(ptr %in, ptr %out) {
 ; GFX12-LABEL: flat_last_use_load_0:
@@ -107,6 +107,3 @@ entry:
 
 !0 = !{i32 1}
 declare i32 @llvm.amdgcn.workitem.id.x()
-;; NOTE: These prefixes are unused and the list is autogenerated. Do not add tests below this line:
-; GFX12-CU: {{.*}}
-; GFX12-WGP: {{.*}}
diff --git a/llvm/test/CodeGen/AMDGPU/memory-legalizer-global-lastuse.ll b/llvm/test/CodeGen/AMDGPU/memory-legalizer-global-lastuse.ll
index c889c67a5ca37..7a9cb992a0cd1 100644
--- a/llvm/test/CodeGen/AMDGPU/memory-legalizer-global-lastuse.ll
+++ b/llvm/test/CodeGen/AMDGPU/memory-legalizer-global-lastuse.ll
@@ -1,6 +1,6 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; RUN: llc -mtriple=amdgcn-amd-amdhsa -O0 -mcpu=gfx1200 < %s | FileCheck --check-prefixes=GFX12,GFX12-WGP %s
-; RUN: llc -mtriple=amdgcn-amd-amdhsa -O0 -mcpu=gfx1200 -mattr=+cumode < %s | FileCheck --check-prefixes=GFX12,GFX12-CU %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -O0 -mcpu=gfx1200 < %s | FileCheck --check-prefix=GFX12 %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -O0 -mcpu=gfx1200 -mattr=+cumode < %s | FileCheck --check-prefix=GFX12 %s
 
 define amdgpu_kernel void @global_last_use_load_0(ptr addrspace(1) %in, ptr addrspace(1) %out) {
 ; GFX12-LABEL: global_last_use_load_0:
@@ -92,6 +92,3 @@ entry:
 }
 !0 = !{i32 1}
 declare i32 @llvm.amdgcn.workitem.id.x()
-;; NOTE: These prefixes are unused and the list is autogenerated. Do not add tests below this line:
-; GFX12-CU: {{.*}}
-; GFX12-WGP: {{.*}}
diff --git a/llvm/test/CodeGen/AMDGPU/memory-legalizer-private-lastuse.ll b/llvm/test/CodeGen/AMDGPU/memory-legalizer-private-lastuse.ll
index 1f835349b12b0..61cec731feb56 100644
--- a/llvm/test/CodeGen/AMDGPU/memory-legalizer-private-lastuse.ll
+++ b/llvm/test/CodeGen/AMDGPU/memory-legalizer-private-lastuse.ll
@@ -1,6 +1,6 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; RUN: llc -mtriple=amdgcn-amd-amdhsa -O0 -mcpu=gfx1200 < %s | FileCheck --check-prefixes=GFX12,GFX12-WGP %s
-; RUN: llc -mtriple=amdgcn-amd-amdhsa -O0 -mcpu=gfx1200 -mattr=+cumode < %s | FileCheck --check-prefixes=GFX12,GFX12-CU %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -O0 -mcpu=gfx1200 < %s | FileCheck --check-prefix=GFX12 %s
+; RUN: llc -mtriple=amdgcn-amd-amdhsa -O0 -mcpu=gfx1200 -mattr=+cumode < %s | FileCheck --check-prefix=GFX12 %s
 
 define amdgpu_kernel void @private_last_use_load_0(ptr addrspace(5) %in, ptr addrspace(1) %out) {
 ; GFX12-LABEL: private_last_use_load_0:
@@ -85,6 +85,3 @@ entry:
 
 !0 = !{i32 1}
 declare i32 @llvm.amdgcn.workitem.id.x()
-;; NOTE: These prefixes are unused and the list is autogenerated. Do not add tests below this line:
-; GFX12-CU: {{.*}}
-; GFX12-WGP: {{.*}}