[llvm-commits] [llvm] r63593 - in /llvm/trunk/lib/CodeGen/SelectionDAG: SelectionDAGBuild.cpp SelectionDAGISel.cpp

Evan Cheng evan.cheng at apple.com
Mon Feb 2 18:15:50 PST 2009


We can refine these later. It's easy to grep for getUnknonwLoc. :-)

Evan

On Feb 2, 2009, at 5:59 PM, Bill Wendling wrote:

> On Mon, Feb 2, 2009 at 5:49 PM, Evan Cheng <evan.cheng at apple.com>  
> wrote:
>>
>> On Feb 2, 2009, at 5:33 PM, Bill Wendling wrote:
>>
>>> +++ llvm/trunk/lib/CodeGen/SelectionDAG/SelectionDAGBuild.cpp Mon
>>> Feb  2 19:33:28 2009
>>> @@ -333,7 +333,8 @@
>>>        unsigned NumRegisters = TLI.getNumRegisters(VT);
>>>        const TargetInstrInfo *TII = MF->getTarget().getInstrInfo();
>>>        for (unsigned i = 0; i != NumRegisters; ++i)
>>> -          BuildMI(MBB, TII->get(TargetInstrInfo::PHI), PHIReg+i);
>>> +          BuildMI(MBB, DebugLoc::getUnknownLoc(),
>>> +                  TII->get(TargetInstrInfo::PHI), PHIReg + i);
>>>        PHIReg += NumRegisters;
>>
>> Perhaps these should have the same location as the first instruction
>> in the basic block? If there are no instructions, it has to traverse
>> up the predecessor chain.
>>
> At this point, there aren't any instructions in the BB (they're all
> being created, and only PHI nodes will be in the blocks after it's
> done). The only way I can think of to get information is to look for
> the "llvm.dbg.stoppoint" calls and get the info that way. But that
> involves having DAG around, which I don't think is available at this
> point...
>
> -bw
> _______________________________________________
> llvm-commits mailing list
> llvm-commits at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits




More information about the llvm-commits mailing list