mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2024-10-19 02:52:53 +02:00
[x86] Propagate memory operands during call frame optimization
Summary: Propagate memory operands when folding load instructions into instructions that directly operate on memory. The original revision has been split. See D80140 for the other part of the changes. Reviewers: craig.topper, rnk, lebedev.ri, efriedma Reviewed By: craig.topper Subscribers: lebedev.ri, hiraditya, llvm-commits Tags: #llvm Differential Revision: https://reviews.llvm.org/D80062
This commit is contained in:
parent
e803204f39
commit
d3e7b3c0e9
@ -531,6 +531,7 @@ void X86CallFrameOptimization::adjustCallSequence(MachineFunction &MF,
|
||||
PushOpcode = Is64Bit ? X86::PUSH64i8 : X86::PUSH32i8;
|
||||
}
|
||||
Push = BuildMI(MBB, Context.Call, DL, TII->get(PushOpcode)).add(PushOp);
|
||||
Push->cloneMemRefs(MF, *Store);
|
||||
break;
|
||||
case X86::MOV32mr:
|
||||
case X86::MOV64mr: {
|
||||
@ -562,6 +563,7 @@ void X86CallFrameOptimization::adjustCallSequence(MachineFunction &MF,
|
||||
unsigned NumOps = DefMov->getDesc().getNumOperands();
|
||||
for (unsigned i = NumOps - X86::AddrNumOperands; i != NumOps; ++i)
|
||||
Push->addOperand(DefMov->getOperand(i));
|
||||
Push->cloneMergedMemRefs(MF, {&*DefMov, &*Store});
|
||||
|
||||
DefMov->eraseFromParent();
|
||||
} else {
|
||||
@ -569,6 +571,7 @@ void X86CallFrameOptimization::adjustCallSequence(MachineFunction &MF,
|
||||
Push = BuildMI(MBB, Context.Call, DL, TII->get(PushOpcode))
|
||||
.addReg(Reg)
|
||||
.getInstr();
|
||||
Push->cloneMemRefs(MF, *Store);
|
||||
}
|
||||
break;
|
||||
}
|
||||
|
109
test/CodeGen/X86/cf-opt-memops.mir
Normal file
109
test/CodeGen/X86/cf-opt-memops.mir
Normal file
@ -0,0 +1,109 @@
|
||||
# RUN: llc -o - -mtriple=x86_64-- -run-pass=x86-cf-opt %s | FileCheck %s
|
||||
|
||||
--- |
|
||||
; ModuleID = 'test.ll'
|
||||
source_filename = "code_io.c"
|
||||
target datalayout = "e-m:e-p270:32:32-p271:32:32-p272:64:64-i64:64-f80:128-n8:16:32:64-S128"
|
||||
|
||||
@.str.8 = private unnamed_addr constant [34 x i8] c"%10s%10s%10s%10s%10s%10s%10s%10s\0A\00", align 1
|
||||
@.str.9 = private unnamed_addr constant [6 x i8] c"nbody\00", align 1
|
||||
@.str.10 = private unnamed_addr constant [6 x i8] c"dtime\00", align 1
|
||||
@.str.11 = private unnamed_addr constant [4 x i8] c"eps\00", align 1
|
||||
@.str.12 = private unnamed_addr constant [4 x i8] c"tol\00", align 1
|
||||
@.str.13 = private unnamed_addr constant [6 x i8] c"dtout\00", align 1
|
||||
@.str.14 = private unnamed_addr constant [6 x i8] c"tstop\00", align 1
|
||||
@.str.15 = private unnamed_addr constant [7 x i8] c"fcells\00", align 1
|
||||
@.str.16 = private unnamed_addr constant [6 x i8] c"NPROC\00", align 1
|
||||
|
||||
define dso_local void @initoutput() local_unnamed_addr {
|
||||
entry:
|
||||
%call1 = tail call i32 (i8*, ...) @printf(i8* nonnull dereferenceable(1) getelementptr inbounds ([34 x i8], [34 x i8]* @.str.8, i64 0, i64 0), i8* getelementptr inbounds ([6 x i8], [6 x i8]* @.str.9, i64 0, i64 0), i8* getelementptr inbounds ([6 x i8], [6 x i8]* @.str.10, i64 0, i64 0), i8* getelementptr inbounds ([4 x i8], [4 x i8]* @.str.11, i64 0, i64 0), i8* getelementptr inbounds ([4 x i8], [4 x i8]* @.str.12, i64 0, i64 0), i8* getelementptr inbounds ([6 x i8], [6 x i8]* @.str.13, i64 0, i64 0), i8* getelementptr inbounds ([6 x i8], [6 x i8]* @.str.14, i64 0, i64 0), i8* getelementptr inbounds ([7 x i8], [7 x i8]* @.str.15, i64 0, i64 0), i8* getelementptr inbounds ([6 x i8], [6 x i8]* @.str.16, i64 0, i64 0))
|
||||
ret void
|
||||
}
|
||||
|
||||
declare dso_local i32 @printf(i8* nocapture readonly, ...) local_unnamed_addr
|
||||
|
||||
; Function Attrs: nounwind
|
||||
declare void @llvm.stackprotector(i8*, i8**) #0
|
||||
|
||||
attributes #0 = { nounwind }
|
||||
|
||||
...
|
||||
---
|
||||
name: initoutput
|
||||
alignment: 16
|
||||
exposesReturnsTwice: false
|
||||
legalized: false
|
||||
regBankSelected: false
|
||||
selected: false
|
||||
failedISel: false
|
||||
tracksRegLiveness: true
|
||||
hasWinCFI: false
|
||||
registers:
|
||||
- { id: 0, class: gr64, preferred-register: '' }
|
||||
- { id: 1, class: gr64, preferred-register: '' }
|
||||
- { id: 2, class: gr64, preferred-register: '' }
|
||||
- { id: 3, class: gr64, preferred-register: '' }
|
||||
- { id: 4, class: gr64, preferred-register: '' }
|
||||
- { id: 5, class: gr64, preferred-register: '' }
|
||||
- { id: 6, class: gr64, preferred-register: '' }
|
||||
- { id: 7, class: gr32, preferred-register: '' }
|
||||
- { id: 8, class: gr8, preferred-register: '' }
|
||||
- { id: 9, class: gr32, preferred-register: '' }
|
||||
liveins: []
|
||||
frameInfo:
|
||||
isFrameAddressTaken: false
|
||||
isReturnAddressTaken: false
|
||||
hasStackMap: false
|
||||
hasPatchPoint: false
|
||||
stackSize: 0
|
||||
offsetAdjustment: 0
|
||||
maxAlignment: 8
|
||||
adjustsStack: false
|
||||
hasCalls: true
|
||||
stackProtector: ''
|
||||
maxCallFrameSize: 4294967295
|
||||
cvBytesOfCalleeSavedRegisters: 0
|
||||
hasOpaqueSPAdjustment: false
|
||||
hasVAStart: false
|
||||
hasMustTailInVarArgFunc: false
|
||||
localFrameSize: 0
|
||||
savePoint: ''
|
||||
restorePoint: ''
|
||||
fixedStack: []
|
||||
stack: []
|
||||
callSites: []
|
||||
constants: []
|
||||
machineFunctionInfo: {}
|
||||
body: |
|
||||
bb.0.entry:
|
||||
ADJCALLSTACKDOWN64 24, 0, 0, implicit-def dead $rsp, implicit-def dead $eflags, implicit-def dead $ssp, implicit $rsp, implicit $ssp
|
||||
%0:gr64 = COPY $rsp
|
||||
MOV64mi32 %0, 1, $noreg, 16, $noreg, @.str.16 :: (store 8 into stack + 16)
|
||||
MOV64mi32 %0, 1, $noreg, 8, $noreg, @.str.15 :: (store 8 into stack + 8)
|
||||
MOV64mi32 %0, 1, $noreg, 0, $noreg, @.str.14 :: (store 8 into stack)
|
||||
%1:gr64 = MOV32ri64 @.str.8
|
||||
%2:gr64 = MOV32ri64 @.str.9
|
||||
%3:gr64 = MOV32ri64 @.str.10
|
||||
%4:gr64 = MOV32ri64 @.str.11
|
||||
%5:gr64 = MOV32ri64 @.str.12
|
||||
%6:gr64 = MOV32ri64 @.str.13
|
||||
%7:gr32 = MOV32r0 implicit-def dead $eflags
|
||||
%8:gr8 = COPY %7.sub_8bit
|
||||
$rdi = COPY %1
|
||||
$rsi = COPY %2
|
||||
$rdx = COPY %3
|
||||
$rcx = COPY %4
|
||||
$r8 = COPY %5
|
||||
$r9 = COPY %6
|
||||
$al = COPY %8
|
||||
CALL64pcrel32 @printf, csr_64, implicit $rsp, implicit $ssp, implicit $rdi, implicit $rsi, implicit $rdx, implicit $rcx, implicit $r8, implicit $r9, implicit $al, implicit-def $rsp, implicit-def $ssp, implicit-def $eax
|
||||
ADJCALLSTACKUP64 24, 0, implicit-def dead $rsp, implicit-def dead $eflags, implicit-def dead $ssp, implicit $rsp, implicit $ssp
|
||||
RET 0
|
||||
|
||||
# Call frame optimization should propagate memory operands
|
||||
# CHECK: PUSH64i32 @{{.*}} :: (store 8 into stack + 16)
|
||||
# CHECK: PUSH64i32 @{{.*}} :: (store 8 into stack + 8)
|
||||
# CHECK: PUSH64i32 @{{.*}} :: (store 8 into stack)
|
||||
|
||||
...
|
Loading…
Reference in New Issue
Block a user