diff options
| author | Ulrich Weigand <ulrich.weigand@de.ibm.com> | 2014-06-18 18:33:36 +0000 |
|---|---|---|
| committer | Ulrich Weigand <ulrich.weigand@de.ibm.com> | 2014-06-18 18:33:36 +0000 |
| commit | ffbd906558a6bb927ba81570491472393ee7ebac (patch) | |
| tree | 11dfc15e21fd5f365d45c5d6e66889be7e79edad /lib/Target/PowerPC | |
| parent | 75b56dc6fe8c8e17ba37683c158dcb63abcd9056 (diff) | |
| download | llvm-ffbd906558a6bb927ba81570491472393ee7ebac.tar.gz llvm-ffbd906558a6bb927ba81570491472393ee7ebac.tar.bz2 llvm-ffbd906558a6bb927ba81570491472393ee7ebac.tar.xz | |
[PowerPC] Remove unnecessary load of r12 in indirect call
When looking at the 64-bit SVR4 indirect call sequence, I noticed
an unnecessary load of r12. And indeed the code says:
// R12 must contain the address of an indirect callee.
But this is not correct; in the 64-bit SVR4 (ELFv1) ABI, there is
no need to load r12 at this point. It seems this code and comment
is a remnant of code originally shared with the Darwin ABI ...
This patch simply removes the unnecessary load.
git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@211203 91177308-0d34-0410-b5e6-96231b3b80d8
Diffstat (limited to 'lib/Target/PowerPC')
| -rw-r--r-- | lib/Target/PowerPC/PPCISelLowering.cpp | 4 |
1 files changed, 0 insertions, 4 deletions
diff --git a/lib/Target/PowerPC/PPCISelLowering.cpp b/lib/Target/PowerPC/PPCISelLowering.cpp index 201d124199..f6884d5a27 100644 --- a/lib/Target/PowerPC/PPCISelLowering.cpp +++ b/lib/Target/PowerPC/PPCISelLowering.cpp @@ -4399,10 +4399,6 @@ PPCTargetLowering::LowerCall_64SVR4(SDValue Chain, SDValue Callee, SDValue AddPtr = DAG.getNode(ISD::ADD, dl, PtrVT, StackPtr, PtrOff); Chain = DAG.getStore(Val.getValue(1), dl, Val, AddPtr, MachinePointerInfo(), false, false, 0); - // R12 must contain the address of an indirect callee. This does not - // mean the MTCTR instruction must use R12; it's easier to model this - // as an extra parameter, so do that. - RegsToPass.push_back(std::make_pair((unsigned)PPC::X12, Callee)); } // Build a sequence of copy-to-reg nodes chained together with token chain |