Thumb1 load/store optimizer: Improve code to materialize new base register.

There are two add-immediate instructions in Thumb1: tADDi8 and tADDi3. Only
the latter supports using different source and destination registers, so
whenever we materialize a new base register (at a certain offset) we'd do
so by moving the base register value to the new register and then adding in
place. This patch changes the code to use a single tADDi3 if the offset is
small enough to fit in 3 bits.

Differential Revision: http://reviews.llvm.org/D5006

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@216193 91177308-0d34-0410-b5e6-96231b3b80d8
This commit is contained in:
Moritz Roth 2014-08-21 17:11:03 +00:00
parent eeb828f029
commit a6afad8b33
4 changed files with 44 additions and 7 deletions

View File

@ -360,13 +360,15 @@ ARMLoadStoreOpt::MergeOps(MachineBasicBlock &MBB,
int BaseOpc =
isThumb2 ? ARM::t2ADDri :
(isThumb1 && Offset < 8) ? ARM::tADDi3 :
isThumb1 ? ARM::tADDi8 : ARM::ADDri;
if (Offset < 0) {
Offset = - Offset;
BaseOpc =
isThumb2 ? ARM::t2SUBri :
(isThumb1 && Offset < 8) ? ARM::tSUBi3 :
isThumb1 ? ARM::tSUBi8 : ARM::SUBri;
Offset = - Offset;
}
if (!TL->isLegalAddImmediate(Offset))
@ -374,22 +376,28 @@ ARMLoadStoreOpt::MergeOps(MachineBasicBlock &MBB,
return false; // Probably not worth it then.
if (isThumb1) {
if (Base != NewBase) {
// Thumb1: depending on immediate size, use either
// ADD NewBase, Base, #imm3
// or
// MOV NewBase, Base
// ADD NewBase, #imm8.
if (Base != NewBase && Offset >= 8) {
// Need to insert a MOV to the new base first.
// FIXME: If the immediate fits in 3 bits, use ADD instead.
BuildMI(MBB, MBBI, dl, TII->get(ARM::tMOVr), NewBase)
.addReg(Base, getKillRegState(BaseKill))
.addImm(Pred).addReg(PredReg);
// Set up BaseKill and Base correctly to insert the ADDS/SUBS below.
Base = NewBase;
BaseKill = false;
}
AddDefaultT1CC(BuildMI(MBB, MBBI, dl, TII->get(BaseOpc), NewBase))
.addReg(NewBase, getKillRegState(true)).addImm(Offset)
.addReg(Base, getKillRegState(BaseKill)).addImm(Offset)
.addImm(Pred).addReg(PredReg);
} else {
BuildMI(MBB, MBBI, dl, TII->get(BaseOpc), NewBase)
.addReg(Base, getKillRegState(BaseKill)).addImm(Offset)
.addImm(Pred).addReg(PredReg).addReg(0);
}
Base = NewBase;
BaseKill = true; // New base is always killed straight away.
}

View File

@ -0,0 +1,29 @@
; RUN: llc < %s -mtriple=thumbv6m-eabi -verify-machineinstrs -o - | FileCheck %s
target datalayout = "e-m:e-p:32:32-i1:8:32-i8:8:32-i16:16:32-i64:64-v128:64:128-a:0:32-n32-S64"
target triple = "thumbv6m-none--eabi"
@a = external global i32*
@b = external global i32*
; Function Attrs: nounwind
define void @foo() #0 {
entry:
; CHECK-LABEL: foo:
; CHECK: ldr r[[SB:[0-9]]], .LCPI
; CHECK: ldr r[[LB:[0-9]]], .LCPI
; CHECK: adds r[[NLB:[0-9]]], r[[LB]], #4
; CHECK-NEXT: ldm r[[NLB]],
; CHECK: adds r[[NSB:[0-9]]], r[[SB]], #4
; CHECK-NEXT: stm r[[NSB]]
%0 = load i32** @a, align 4
%arrayidx = getelementptr inbounds i32* %0, i32 1
%1 = bitcast i32* %arrayidx to i8*
%2 = load i32** @b, align 4
%arrayidx1 = getelementptr inbounds i32* %2, i32 1
%3 = bitcast i32* %arrayidx1 to i8*
tail call void @llvm.memcpy.p0i8.p0i8.i32(i8* %1, i8* %3, i32 24, i32 4, i1 false)
ret void
}
; Function Attrs: nounwind
declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture readonly, i32, i32, i1) #1

View File

@ -1,4 +1,4 @@
; RUN: llc < %s -mtriple=thumbv6m-eabi -o - | FileCheck %s
; RUN: llc < %s -mtriple=thumbv6m-eabi -verify-machineinstrs -o - | FileCheck %s
@X = external global [0 x i32] ; <[0 x i32]*> [#uses=5]

View File

@ -1,4 +1,4 @@
; RUN: llc -mtriple=thumbv6m-eabi %s -o - | FileCheck %s
; RUN: llc -mtriple=thumbv6m-eabi -verify-machineinstrs %s -o - | FileCheck %s
@d = external global [64 x i32]
@s = external global [64 x i32]