summaryrefslogtreecommitdiff
path: root/lib/Target/ARM/Thumb1FrameLowering.cpp
diff options
context:
space:
mode:
authorTim Northover <tnorthover@apple.com>2013-11-08 17:18:07 +0000
committerTim Northover <tnorthover@apple.com>2013-11-08 17:18:07 +0000
commit323ac85d6ad7ba5d9593d8e151d879bd91d82e08 (patch)
tree1c3b17bc84524be55da9f3ffaf112d3263a564f9 /lib/Target/ARM/Thumb1FrameLowering.cpp
parent2b01682aa7b9509e9fa1865ebed3d0a7928f5b7a (diff)
downloadllvm-323ac85d6ad7ba5d9593d8e151d879bd91d82e08.tar.gz
llvm-323ac85d6ad7ba5d9593d8e151d879bd91d82e08.tar.bz2
llvm-323ac85d6ad7ba5d9593d8e151d879bd91d82e08.tar.xz
ARM: fold prologue/epilogue sp updates into push/pop for code size
ARM prologues usually look like: push {r7, lr} sub sp, sp, #4 If code size is extremely important, this can be optimised to the single instruction: push {r6, r7, lr} where we don't actually care about the contents of r6, but pushing it subtracts 4 from sp as a side effect. This should implement such a conversion, predicated on the "minsize" function attribute (-Oz) since I've yet to find any code it actually makes faster. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@194264 91177308-0d34-0410-b5e6-96231b3b80d8
Diffstat (limited to 'lib/Target/ARM/Thumb1FrameLowering.cpp')
-rw-r--r--lib/Target/ARM/Thumb1FrameLowering.cpp15
1 files changed, 11 insertions, 4 deletions
diff --git a/lib/Target/ARM/Thumb1FrameLowering.cpp b/lib/Target/ARM/Thumb1FrameLowering.cpp
index cfaa792fc9..d921c82cfb 100644
--- a/lib/Target/ARM/Thumb1FrameLowering.cpp
+++ b/lib/Target/ARM/Thumb1FrameLowering.cpp
@@ -164,11 +164,17 @@ void Thumb1FrameLowering::emitPrologue(MachineFunction &MF) const {
AFI->setDPRCalleeSavedAreaOffset(DPRCSOffset);
NumBytes = DPRCSOffset;
+ int FramePtrOffsetInBlock = 0;
+ if (tryFoldSPUpdateIntoPushPop(MF, prior(MBBI), NumBytes)) {
+ FramePtrOffsetInBlock = NumBytes;
+ NumBytes = 0;
+ }
+
// Adjust FP so it point to the stack slot that contains the previous FP.
if (HasFP) {
- int FramePtrOffset = MFI->getObjectOffset(FramePtrSpillFI) + GPRCS1Size;
+ FramePtrOffsetInBlock += MFI->getObjectOffset(FramePtrSpillFI) + GPRCS1Size;
AddDefaultPred(BuildMI(MBB, MBBI, dl, TII.get(ARM::tADDrSPi), FramePtr)
- .addReg(ARM::SP).addImm(FramePtrOffset / 4)
+ .addReg(ARM::SP).addImm(FramePtrOffsetInBlock / 4)
.setMIFlags(MachineInstr::FrameSetup));
if (NumBytes > 508)
// If offset is > 508 then sp cannot be adjusted in a single instruction,
@@ -292,8 +298,9 @@ void Thumb1FrameLowering::emitEpilogue(MachineFunction &MF,
&MBB.front() != MBBI &&
prior(MBBI)->getOpcode() == ARM::tPOP) {
MachineBasicBlock::iterator PMBBI = prior(MBBI);
- emitSPUpdate(MBB, PMBBI, TII, dl, *RegInfo, NumBytes);
- } else
+ if (!tryFoldSPUpdateIntoPushPop(MF, PMBBI, NumBytes))
+ emitSPUpdate(MBB, PMBBI, TII, dl, *RegInfo, NumBytes);
+ } else if (!tryFoldSPUpdateIntoPushPop(MF, MBBI, NumBytes))
emitSPUpdate(MBB, MBBI, TII, dl, *RegInfo, NumBytes);
}
}