RPCSX/llvm
1
0
Fork 0
mirror of https://github.com/RPCSX/llvm.git synced 2025-02-05 03:46:27 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

AMDGPU: Run LoadStoreVectorizer pass by default

Browse Source 
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@281112 91177308-0d34-0410-b5e6-96231b3b80d8
This commit is contained in:
Matt Arsenault 2016-09-09 22:29:28 +00:00
parent 2c9e9a6439
commit d5a5e9043a
9 changed files with 25 additions and 24 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
lib/Target/AMDGPU/AMDGPUISelDAGToDAG.cpp
View File

@ -786,6 +786,9 @@ bool AMDGPUDAGToDAGISel::SelectDS64Bit4ByteAligned(SDValue Addr, SDValue &Base,
}
// default case
// FIXME: This is broken on SI where we still need to check if the base
// pointer is positive here.
Base = Addr;
Offset0 = CurDAG->getTargetConstant(0, DL, MVT::i8);
Offset1 = CurDAG->getTargetConstant(1, DL, MVT::i8);

2
lib/Target/AMDGPU/AMDGPUTargetMachine.cpp
View File

@ -58,7 +58,7 @@ static cl::opt<bool> EnableR600IfConvert(
static cl::opt<bool> EnableLoadStoreVectorizer(
"amdgpu-load-store-vectorizer",
cl::desc("Enable load store vectorizer"),
cl::init(false),
cl::init(true),
cl::Hidden);
extern "C" void LLVMInitializeAMDGPUTarget() {

12
test/CodeGen/AMDGPU/amdgpu.private-memory.ll
View File

@ -1,9 +1,9 @@
; RUN: llc -show-mc-encoding -mattr=+promote-alloca -verify-machineinstrs -march=amdgcn < %s | FileCheck %s -check-prefix=SI-PROMOTE -check-prefix=SI -check-prefix=FUNC
; RUN: llc -show-mc-encoding -mattr=+promote-alloca -verify-machineinstrs -mtriple=amdgcn--amdhsa -mcpu=kaveri -mattr=-unaligned-buffer-access < %s | FileCheck %s -check-prefix=SI-PROMOTE -check-prefix=SI -check-prefix=FUNC -check-prefix=HSA-PROMOTE
; RUN: llc -show-mc-encoding -mattr=-promote-alloca -verify-machineinstrs -march=amdgcn < %s | FileCheck %s -check-prefix=SI-ALLOCA -check-prefix=SI -check-prefix=FUNC
; RUN: llc -show-mc-encoding -mattr=-promote-alloca -verify-machineinstrs -mtriple=amdgcn-amdhsa -mcpu=kaveri -mattr=-unaligned-buffer-access < %s | FileCheck %s -check-prefix=SI-ALLOCA -check-prefix=SI -check-prefix=FUNC -check-prefix=HSA-ALLOCA
; RUN: llc -show-mc-encoding -mattr=+promote-alloca -verify-machineinstrs -mtriple=amdgcn-amdhsa -march=amdgcn -mcpu=tonga -mattr=-unaligned-buffer-access < %s | FileCheck %s -check-prefix=SI-PROMOTE -check-prefix=SI -check-prefix=FUNC
; RUN: llc -show-mc-encoding -mattr=-promote-alloca -verify-machineinstrs -mtriple=amdgcn-amdhsa -march=amdgcn -mcpu=tonga -mattr=-unaligned-buffer-access < %s | FileCheck %s -check-prefix=SI-ALLOCA -check-prefix=SI -check-prefix=FUNC
; RUN: llc -show-mc-encoding -mattr=+promote-alloca -amdgpu-load-store-vectorizer=0 -verify-machineinstrs -march=amdgcn < %s | FileCheck %s -check-prefix=SI-PROMOTE -check-prefix=SI -check-prefix=FUNC
; RUN: llc -show-mc-encoding -mattr=+promote-alloca -amdgpu-load-store-vectorizer=0 -verify-machineinstrs -mtriple=amdgcn--amdhsa -mcpu=kaveri -mattr=-unaligned-buffer-access < %s | FileCheck %s -check-prefix=SI-PROMOTE -check-prefix=SI -check-prefix=FUNC -check-prefix=HSA-PROMOTE
; RUN: llc -show-mc-encoding -mattr=-promote-alloca -amdgpu-load-store-vectorizer=0 -verify-machineinstrs -march=amdgcn < %s | FileCheck %s -check-prefix=SI-ALLOCA -check-prefix=SI -check-prefix=FUNC
; RUN: llc -show-mc-encoding -mattr=-promote-alloca -amdgpu-load-store-vectorizer=0 -verify-machineinstrs -mtriple=amdgcn-amdhsa -mcpu=kaveri -mattr=-unaligned-buffer-access < %s | FileCheck %s -check-prefix=SI-ALLOCA -check-prefix=SI -check-prefix=FUNC -check-prefix=HSA-ALLOCA
; RUN: llc -show-mc-encoding -mattr=+promote-alloca -amdgpu-load-store-vectorizer=0 -verify-machineinstrs -mtriple=amdgcn-amdhsa -march=amdgcn -mcpu=tonga -mattr=-unaligned-buffer-access < %s | FileCheck %s -check-prefix=SI-PROMOTE -check-prefix=SI -check-prefix=FUNC
; RUN: llc -show-mc-encoding -mattr=-promote-alloca -amdgpu-load-store-vectorizer=0 -verify-machineinstrs -mtriple=amdgcn-amdhsa -march=amdgcn -mcpu=tonga -mattr=-unaligned-buffer-access < %s | FileCheck %s -check-prefix=SI-ALLOCA -check-prefix=SI -check-prefix=FUNC
; RUN: opt -S -mtriple=amdgcn-unknown-amdhsa -mcpu=kaveri -amdgpu-promote-alloca < %s | FileCheck -check-prefix=HSAOPT -check-prefix=OPT %s
; RUN: opt -S -mtriple=amdgcn-unknown-unknown -mcpu=kaveri -amdgpu-promote-alloca < %s | FileCheck -check-prefix=NOHSAOPT -check-prefix=OPT %s

12
test/CodeGen/AMDGPU/ds_read2_offset_order.ll
View File

@ -1,17 +1,15 @@
; RUN: llc -march=amdgcn -mcpu=bonaire -verify-machineinstrs -mattr=+load-store-opt < %s | FileCheck -strict-whitespace -check-prefix=SI %s
; RUN: llc -march=amdgcn -mcpu=tonga -verify-machineinstrs -mattr=+load-store-opt < %s | FileCheck -strict-whitespace -check-prefix=SI %s
; RUN: llc -march=amdgcn -mcpu=bonaire -verify-machineinstrs < %s | FileCheck -strict-whitespace -check-prefix=SI %s
; RUN: llc -march=amdgcn -mcpu=tonga -verify-machineinstrs < %s | FileCheck -strict-whitespace -check-prefix=SI %s
@lds = addrspace(3) global [512 x float] undef, align 4
; offset0 is larger than offset1
; SI-LABEL: {{^}}offset_order:
; SI-DAG: ds_read2st64_b32 v[{{[0-9]+}}:{{[0-9]+}}], v{{[0-9]+}} offset1:4{{$}}
; SI-DAG: ds_read2_b32 v[{{[0-9]+}}:{{[0-9]+}}], v{{[0-9]+}} offset0:2 offset1:3
; SI-DAG: ds_read2_b32 v[{{[0-9]+}}:{{[0-9]+}}], v{{[0-9]+}} offset0:12 offset1:14
; SI-DAG: ds_read_b32 v{{[0-9]+}}, v{{[0-9]+}} offset:44
; SI-DAG: ds_read_b32 v{{[0-9]+}}, v{{[0-9]+}} offset:56
; SI-DAG: ds_read2_b32 v[{{[0-9]+}}:{{[0-9]+}}], v{{[0-9]+}} offset0:11 offset1:12
define void @offset_order(float addrspace(1)* %out) {
entry:
%ptr0 = getelementptr inbounds [512 x float], [512 x float] addrspace(3)* @lds, i32 0, i32 0

2
test/CodeGen/AMDGPU/invariant-load-no-alias-store.ll
View File

@ -1,4 +1,4 @@
; RUN: llc -march=amdgcn -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
; RUN: llc -march=amdgcn -amdgpu-load-store-vectorizer=0 -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
; GatherAllAliases gives up on trying to analyze cases where the
; pointer may have been loaded from an aliased store, so make sure

9
test/CodeGen/AMDGPU/merge-stores.ll
View File

@ -1,9 +1,10 @@
; RUN: llc -march=amdgcn -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=GCN -check-prefix=GCN-NOAA %s
; RUN: llc -march=amdgcn -mcpu=bonaire -verify-machineinstrs < %s | FileCheck -check-prefix=SI -check-prefix=GCN -check-prefix=GCN-NOAA %s
; RUN: llc -march=amdgcn -verify-machineinstrs -amdgpu-load-store-vectorizer=0 < %s | FileCheck -check-prefix=SI -check-prefix=GCN -check-prefix=GCN-NOAA %s
; RUN: llc -march=amdgcn -mcpu=bonaire -verify-machineinstrs -amdgpu-load-store-vectorizer=0 < %s | FileCheck -check-prefix=SI -check-prefix=GCN -check-prefix=GCN-NOAA %s
; RUN: llc -march=amdgcn -verify-machineinstrs -combiner-alias-analysis < %s | FileCheck -check-prefix=SI -check-prefix=GCN -check-prefix=GCN-AA %s
; RUN: llc -march=amdgcn -mcpu=bonaire -verify-machineinstrs -combiner-alias-analysis < %s | FileCheck -check-prefix=SI -check-prefix=GCN -check-prefix=GCN-AA %s
; RUN: llc -march=amdgcn -verify-machineinstrs -combiner-alias-analysis -amdgpu-load-store-vectorizer=0 < %s | FileCheck -check-prefix=SI -check-prefix=GCN -check-prefix=GCN-AA %s
; RUN: llc -march=amdgcn -mcpu=bonaire -verify-machineinstrs -combiner-alias-analysis -amdgpu-load-store-vectorizer=0 < %s | FileCheck -check-prefix=SI -check-prefix=GCN -check-prefix=GCN-AA %s
; This test is mostly to test DAG store merging, so disable the vectorizer.
; Run with devices with different unaligned load restrictions.
; TODO: Vector element tests

4
test/CodeGen/AMDGPU/promote-alloca-stored-pointer-value.ll
View File

@ -1,5 +1,5 @@
; RUN: llc -march=amdgcn -mattr=+promote-alloca -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
; RUN: llc -march=amdgcn -mattr=-promote-alloca -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
; RUN: llc -march=amdgcn -mattr=+promote-alloca,+max-private-element-size-4 -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
; RUN: llc -march=amdgcn -mattr=-promote-alloca,+max-private-element-size-4 -verify-machineinstrs < %s | FileCheck -check-prefix=GCN %s
; Pointer value is stored in a candidate for LDS usage.

1
test/CodeGen/AMDGPU/schedule-global-loads.ll
View File

@ -1,6 +1,5 @@
; RUN: llc -march=amdgcn -verify-machineinstrs < %s | FileCheck -check-prefix=FUNC -check-prefix=SI %s
; FIXME: This currently doesn't do a great job of clustering the
; loads, which end up with extra moves between them. Right now, it
; seems the only things areLoadsFromSameBasePtr is accomplishing is

4
test/CodeGen/AMDGPU/scratch-buffer.ll
View File

@ -48,8 +48,8 @@ done:
}
; GCN-LABEL: {{^}}legal_offset_fi_offset:
; GCN: buffer_store_dword v{{[0-9]+}}, v{{[0-9]+}}, s[{{[0-9]+}}:{{[0-9]+}}], s{{[0-9]+}} offen{{$}}
; GCN: v_add_i32_e32 [[OFFSET:v[0-9]+]], vcc, 0x8000
; GCN-DAG: buffer_store_dword v{{[0-9]+}}, v{{[0-9]+}}, s[{{[0-9]+}}:{{[0-9]+}}], s{{[0-9]+}} offen{{$}}
; GCN-DAG: v_add_i32_e32 [[OFFSET:v[0-9]+]], vcc, 0x8000
; GCN: buffer_store_dword v{{[0-9]+}}, [[OFFSET]], s[{{[0-9]+}}:{{[0-9]+}}], s{{[0-9]+}} offen{{$}}
define void @legal_offset_fi_offset(i32 addrspace(1)* %out, i32 %cond, i32 addrspace(1)* %offsets, i32 %if_offset, i32 %else_offset) {