AMDGPU: Merge dag and gisel tests for non-entry LDS handling

diff --git a/llvm/test/CodeGen/AMDGPU/GlobalISel/lds-global-non-entry-func.ll b/llvm/test/CodeGen/AMDGPU/GlobalISel/lds-global-non-entry-func.ll
deleted file mode 100644 (file)
index ea0376a..0000000
--- a/llvm/test/CodeGen/AMDGPU/GlobalISel/lds-global-non-entry-func.ll
+++ /dev/null
@@ -1,55 +0,0 @@
-; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; RUN: llc -global-isel -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -amdgpu-enable-lower-module-lds=false -o - %s 2> %t | FileCheck --check-prefix=GFX8 %s
-; RUN: FileCheck -check-prefix=ERR %s < %t
-
-; RUN: llc -global-isel -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -amdgpu-enable-lower-module-lds=false -o - %s 2> %t | FileCheck --check-prefix=GFX9 %s
-; RUN: FileCheck -check-prefix=ERR %s < %t
-
-@lds = internal addrspace(3) global float undef, align 4
-
-; ERR: warning: <unknown>:0:0: in function func_use_lds_global void (): local memory global used by non-kernel function
-define void @func_use_lds_global() {
-; GFX8-LABEL: func_use_lds_global:
-; GFX8:       ; %bb.0:
-; GFX8-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX8-NEXT:    v_mov_b32_e32 v0, 0
-; GFX8-NEXT:    s_mov_b32 m0, -1
-; GFX8-NEXT:    s_mov_b64 s[0:1], s[6:7]
-; GFX8-NEXT:    s_trap 2
-; GFX8-NEXT:    ds_write_b32 v0, v0
-; GFX8-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX8-NEXT:    s_setpc_b64 s[30:31]
-;
-; GFX9-LABEL: func_use_lds_global:
-; GFX9:       ; %bb.0:
-; GFX9-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX9-NEXT:    v_mov_b32_e32 v0, 0
-; GFX9-NEXT:    s_trap 2
-; GFX9-NEXT:    ds_write_b32 v0, v0
-; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_setpc_b64 s[30:31]
-  store float 0.0, ptr addrspace(3) @lds, align 4
-  ret void
-}
-
-; ERR: warning: <unknown>:0:0: in function func_use_lds_global_constexpr_cast void (): local memory global used by non-kernel function
-define void @func_use_lds_global_constexpr_cast() {
-; GFX8-LABEL: func_use_lds_global_constexpr_cast:
-; GFX8:       ; %bb.0:
-; GFX8-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX8-NEXT:    s_mov_b64 s[0:1], s[6:7]
-; GFX8-NEXT:    s_trap 2
-; GFX8-NEXT:    flat_store_dword v[0:1], v0
-; GFX8-NEXT:    s_waitcnt vmcnt(0)
-; GFX8-NEXT:    s_setpc_b64 s[30:31]
-;
-; GFX9-LABEL: func_use_lds_global_constexpr_cast:
-; GFX9:       ; %bb.0:
-; GFX9-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX9-NEXT:    s_trap 2
-; GFX9-NEXT:    global_store_dword v[0:1], v0, off
-; GFX9-NEXT:    s_waitcnt vmcnt(0)
-; GFX9-NEXT:    s_setpc_b64 s[30:31]
-  store i32 ptrtoint (ptr addrspace(3) @lds to i32), ptr addrspace(1) undef, align 4
-  ret void
-}

diff --git a/llvm/test/CodeGen/AMDGPU/lds-global-non-entry-func.ll b/llvm/test/CodeGen/AMDGPU/lds-global-non-entry-func.ll
index 2e3aefa..d7f82ce 100644 (file)
--- a/llvm/test/CodeGen/AMDGPU/lds-global-non-entry-func.ll
+++ b/llvm/test/CodeGen/AMDGPU/lds-global-non-entry-func.ll
@@ -1,51 +1,99 @@
 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
-; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -o - -amdgpu-enable-lower-module-lds=false %s 2> %t | FileCheck -check-prefixes=GFX8 %s
+; RUN: llc -global-isel=0 -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -o - -amdgpu-enable-lower-module-lds=false %s 2> %t | FileCheck -check-prefixes=GFX8,GFX8-SDAG %s
 ; RUN: FileCheck -check-prefix=ERR %s < %t
 
-; RUN: llc -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -o - -amdgpu-enable-lower-module-lds=false %s 2> %t | FileCheck -check-prefixes=GFX9 %s
+; RUN: llc -global-isel=1 -mtriple=amdgcn-amd-amdhsa -mcpu=fiji -o - -amdgpu-enable-lower-module-lds=false %s 2> %t | FileCheck -check-prefixes=GFX8,GFX8-GISEL %s
 ; RUN: FileCheck -check-prefix=ERR %s < %t
 
-@lds = internal addrspace(3) global float undef, align 4
+; RUN: llc -global-isel=0 -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -o - -amdgpu-enable-lower-module-lds=false %s 2> %t | FileCheck -check-prefixes=GFX9,GFX9-SDAG %s
+; RUN: FileCheck -check-prefix=ERR %s < %t
+
+; RUN: llc -global-isel=1 -mtriple=amdgcn-amd-amdhsa -mcpu=gfx900 -o - -amdgpu-enable-lower-module-lds=false %s 2> %t | FileCheck -check-prefixes=GFX9,GFX9-GISEL %s
+; RUN: FileCheck -check-prefix=ERR %s < %t
+
+@lds = internal addrspace(3) global float poison, align 4
+
+; FIXME: The DAG should probably move the trap before the access.
 
 ; ERR: warning: <unknown>:0:0: in function func_use_lds_global void (): local memory global used by non-kernel function
 define void @func_use_lds_global() {
-; GFX8-LABEL: func_use_lds_global:
-; GFX8:       ; %bb.0:
-; GFX8-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX8-NEXT:    v_mov_b32_e32 v0, 0
-; GFX8-NEXT:    s_mov_b32 m0, -1
-; GFX8-NEXT:    ds_write_b32 v0, v0
-; GFX8-NEXT:    s_mov_b64 s[0:1], s[6:7]
-; GFX8-NEXT:    s_trap 2
-; GFX8-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX8-NEXT:    s_setpc_b64 s[30:31]
+; GFX8-SDAG-LABEL: func_use_lds_global:
+; GFX8-SDAG:       ; %bb.0:
+; GFX8-SDAG-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX8-SDAG-NEXT:    v_mov_b32_e32 v0, 0
+; GFX8-SDAG-NEXT:    s_mov_b32 m0, -1
+; GFX8-SDAG-NEXT:    ds_write_b32 v0, v0
+; GFX8-SDAG-NEXT:    s_mov_b64 s[0:1], s[6:7]
+; GFX8-SDAG-NEXT:    s_trap 2
+; GFX8-SDAG-NEXT:    s_waitcnt lgkmcnt(0)
+; GFX8-SDAG-NEXT:    s_setpc_b64 s[30:31]
 ;
-; GFX9-LABEL: func_use_lds_global:
-; GFX9:       ; %bb.0:
-; GFX9-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX9-NEXT:    v_mov_b32_e32 v0, 0
-; GFX9-NEXT:    ds_write_b32 v0, v0
-; GFX9-NEXT:    s_trap 2
-; GFX9-NEXT:    s_waitcnt lgkmcnt(0)
-; GFX9-NEXT:    s_setpc_b64 s[30:31]
-  store float 0.0, ptr addrspace(3) @lds, align 4
+; GFX8-GISEL-LABEL: func_use_lds_global:
+; GFX8-GISEL:       ; %bb.0:
+; GFX8-GISEL-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX8-GISEL-NEXT:    v_mov_b32_e32 v0, 0
+; GFX8-GISEL-NEXT:    s_mov_b32 m0, -1
+; GFX8-GISEL-NEXT:    s_mov_b64 s[0:1], s[6:7]
+; GFX8-GISEL-NEXT:    s_trap 2
+; GFX8-GISEL-NEXT:    ds_write_b32 v0, v0
+; GFX8-GISEL-NEXT:    s_waitcnt lgkmcnt(0)
+; GFX8-GISEL-NEXT:    s_setpc_b64 s[30:31]
+;
+; GFX9-SDAG-LABEL: func_use_lds_global:
+; GFX9-SDAG:       ; %bb.0:
+; GFX9-SDAG-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX9-SDAG-NEXT:    v_mov_b32_e32 v0, 0
+; GFX9-SDAG-NEXT:    ds_write_b32 v0, v0
+; GFX9-SDAG-NEXT:    s_trap 2
+; GFX9-SDAG-NEXT:    s_waitcnt lgkmcnt(0)
+; GFX9-SDAG-NEXT:    s_setpc_b64 s[30:31]
+;
+; GFX9-GISEL-LABEL: func_use_lds_global:
+; GFX9-GISEL:       ; %bb.0:
+; GFX9-GISEL-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX9-GISEL-NEXT:    v_mov_b32_e32 v0, 0
+; GFX9-GISEL-NEXT:    s_trap 2
+; GFX9-GISEL-NEXT:    ds_write_b32 v0, v0
+; GFX9-GISEL-NEXT:    s_waitcnt lgkmcnt(0)
+; GFX9-GISEL-NEXT:    s_setpc_b64 s[30:31]
+  store volatile float 0.0, ptr addrspace(3) @lds, align 4
   ret void
 }
 
 ; ERR: warning: <unknown>:0:0: in function func_use_lds_global_constexpr_cast void (): local memory global used by non-kernel function
 define void @func_use_lds_global_constexpr_cast() {
-; GFX8-LABEL: func_use_lds_global_constexpr_cast:
-; GFX8:       ; %bb.0:
-; GFX8-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX8-NEXT:    s_mov_b64 s[0:1], s[6:7]
-; GFX8-NEXT:    s_trap 2
-; GFX8-NEXT:    s_setpc_b64 s[30:31]
+; GFX8-SDAG-LABEL: func_use_lds_global_constexpr_cast:
+; GFX8-SDAG:       ; %bb.0:
+; GFX8-SDAG-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX8-SDAG-NEXT:    s_mov_b64 s[0:1], s[6:7]
+; GFX8-SDAG-NEXT:    s_trap 2
+; GFX8-SDAG-NEXT:    s_setpc_b64 s[30:31]
+;
+; GFX8-GISEL-LABEL: func_use_lds_global_constexpr_cast:
+; GFX8-GISEL:       ; %bb.0:
+; GFX8-GISEL-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX8-GISEL-NEXT:    s_mov_b64 s[0:1], s[6:7]
+; GFX8-GISEL-NEXT:    s_trap 2
+; GFX8-GISEL-NEXT:    flat_store_dword v[0:1], v0
+; GFX8-GISEL-NEXT:    s_waitcnt vmcnt(0)
+; GFX8-GISEL-NEXT:    s_setpc_b64 s[30:31]
+;
+; GFX9-SDAG-LABEL: func_use_lds_global_constexpr_cast:
+; GFX9-SDAG:       ; %bb.0:
+; GFX9-SDAG-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX9-SDAG-NEXT:    s_trap 2
+; GFX9-SDAG-NEXT:    s_setpc_b64 s[30:31]
 ;
-; GFX9-LABEL: func_use_lds_global_constexpr_cast:
-; GFX9:       ; %bb.0:
-; GFX9-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
-; GFX9-NEXT:    s_trap 2
-; GFX9-NEXT:    s_setpc_b64 s[30:31]
-  store i32 ptrtoint (ptr addrspace(3) @lds to i32), ptr addrspace(1) undef, align 4
+; GFX9-GISEL-LABEL: func_use_lds_global_constexpr_cast:
+; GFX9-GISEL:       ; %bb.0:
+; GFX9-GISEL-NEXT:    s_waitcnt vmcnt(0) expcnt(0) lgkmcnt(0)
+; GFX9-GISEL-NEXT:    s_trap 2
+; GFX9-GISEL-NEXT:    global_store_dword v[0:1], v0, off
+; GFX9-GISEL-NEXT:    s_waitcnt vmcnt(0)
+; GFX9-GISEL-NEXT:    s_setpc_b64 s[30:31]
+  store volatile i32 ptrtoint (ptr addrspace(3) @lds to i32), ptr addrspace(1) poison, align 4
   ret void
 }
+;; NOTE: These prefixes are unused and the list is autogenerated. Do not add tests below this line:
+; GFX8: {{.*}}
+; GFX9: {{.*}}