[llvm] r179957 - SimplifyCFG: If convert single conditional stores

[Hal Finkel]

----- Original Message -----
> From: "Arnold Schwaighofer" <aschwaighofer at apple.com>
> To: "Chris Lattner" <clattner at apple.com>, "Commit Messages and Patches for LLVM" <llvm-commits at cs.uiuc.edu>
> Cc: "Jakob Stoklund Olesen" <jolesen at apple.com>
> Sent: Wednesday, April 24, 2013 7:13:27 PM
> Subject: Re: [llvm] r179957 - SimplifyCFG: If convert single conditional stores
> 
> I have CC’ed all folks who voiced opinions on this in this email
> conversation. I am not going to move forward until I have received
> an okay from all of them.
> 
> 
> So far my understanding is that Andrew, Jakob, and Chris are okay
> with this (please correct me if I am wrong). May I ask a ‘yay’ or
> ‘nay’ from the rest?
> 
> My intend is to:
>  * Enable the SimplifyCFG optimization now (i.e resurrect 179957)
> and
> * Add the Early If-converter optimization of stores (this patch might
> need some polishing, in principle it works though) for when we
> remove selectification from simplifycfg.

FWIW, I'm also in favor of doing this now.

 -Hal

> 
> Best,
> Arnold
> 
> 
> On Apr 24, 2013, at 6:58 PM, Chris Lattner <clattner at apple.com>
> wrote:
> 
> > 
> > On Apr 24, 2013, at 12:17 PM, Arnold Schwaighofer
> > <aschwaighofer at apple.com> wrote:
> > 
> >> Given that we have the optimization implemented in both
> >> SimplifyCFG and the Early-If converter now, I believe we can take
> >> a pragmatic approach:
> >> 
> >> * Enable the SimplifyCFG optimization now. This will allow us not
> >> to look bad on Phoronix/hmmer and spec/hmmer in llvm 3.3 (20%!).
> >> And it seems people agree that this is generally a good thing in
> >> the current compilation pipeline.
> > 
> > I'm in favor of doing this, but please make sure folks who opposed
> > it are ok with it.
> > 
> > -Chris
> 
> 
> _______________________________________________
> llvm-commits mailing list
> llvm-commits at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
>