[PATCH] Fix optnone on Windows

[Reid Kleckner]

lgtm

Yep, fastisel can definitely bail out. It happens a lot on Windows for
unlowerable constructs.

On Mon, Nov 3, 2014 at 9:37 AM, Robinson, Paul <
Paul_Robinson at playstation.sony.com> wrote:

> ping.
>
> > -----Original Message-----
> > From: Robinson, Paul
> > Sent: Sunday, October 26, 2014 9:23 PM
> > To: Robinson, Paul; llvm-commits at cs.uiuc.edu
> > Subject: RE: [PATCH] Fix optnone on Windows
> >
> > Ping.
> >
> > > -----Original Message-----
> > > From: llvm-commits-bounces at cs.uiuc.edu [mailto:llvm-commits-
> > > bounces at cs.uiuc.edu] On Behalf Of Robinson, Paul
> > > Sent: Monday, October 20, 2014 3:55 PM
> > > To: llvm-commits at cs.uiuc.edu
> > > Subject: [PATCH] Fix optnone on Windows
> > >
> > > [PATCH] Fix optnone on Windows
> > >
> > > Normally an 'optnone' function goes through fast-isel, which does not
> > > call DAGCombiner.  But we ran into a case (on Windows) where the
> > > calling convention causes argument lowering to bail out of fast-isel,
> > > and we end up in CodeGenAndEmitDAG() which does run DAGCombiner.
> > > So, we need to make DAGCombiner check for 'optnone' after all.
> > >
> > > Along the way we discovered we hadn't committed all the tests we had
> > > written for 'optnone' so here are two more. 'fastmath-optnone.ll' is
> > > the test that caught this, 'optnone-simple.ll' is one we missed from
> > > our original implementation.
> > > Thanks,
> > > --paulr
>
>
> _______________________________________________
> llvm-commits mailing list
> llvm-commits at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20141103/a9b48c0b/attachment.html>