[LLVMdev] 3.6.1 -rc1 has been tagged. Testing begins.
Daniel Sanders
Daniel.Sanders at imgtec.com
Fri May 15 06:32:17 PDT 2015
> -----Original Message-----
> From: Renato Golin [mailto:renato.golin at linaro.org]
> Sent: 15 May 2015 14:15
> To: Daniel Sanders
> Cc: Tom Stellard; llvmdev at cs.uiuc.edu; cfe-dev at cs.uiuc.edu
> Subject: Re: [LLVMdev] 3.6.1 -rc1 has been tagged. Testing begins.
>
> On 15 May 2015 at 13:34, Daniel Sanders <Daniel.Sanders at imgtec.com>
> wrote:
> > Not yet, these two failures were the first time we had seen the problem.
> We're still looking into fixing it properly but I'm currently thinking that the
> correct fix is to add
> > if (Subtarget.isGP64bit())
> > setOperationAction(ISD::SETCC, MVT::i32, Promote);
> > to MipsISelLowering.cpp and sort out the consequences of this on the
> patterns for all the comparison instructions. This is likely to be a fairly big
> change to our target.
>
> Right, so the best thing is probably reverting this one for 3.6.1.
>
> cheers,
> --renato
I agree. I reverted it this morning (in r237432) after confirming that doing so fixed the regressions. I've also kicked off the cross-compilation test-runs again just to be sure removing it doesn't break anything else.
More information about the llvm-dev
mailing list