[Polly][Refactor] IslAst [6/7]

[Johannes Doerfert]

Attached the old patch 6 now based on the old patch 7 and passing your new test.

--

Johannes Doerfert
Employee of Qualcomm Innovation Center, Inc.
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, hosted by The Linux Foundation


-----Original Message-----
From: llvm-commits-bounces at cs.uiuc.edu [mailto:llvm-commits-bounces at cs.uiuc.edu] On Behalf Of Johannes Doerfert
Sent: Wednesday, July 30, 2014 1:41 AM
To: Tobias Grosser
Cc: llvm-commits at cs.uiuc.edu
Subject: Re: [Polly][Refactor] IslAst [6/7]

On 07/30, Tobias Grosser wrote:
> On 30/07/2014 00:23, Johannes Doerfert wrote:
> >How do we proceed now? I need to get a lot more patches out (at least 
> >one the list) and I have little time left.
> 
> You can always post patches to the list. ;-)
The question for me is, if we can commit them at some point... 

> Specifically, the fact that you changed behavior could have found by 
> any other patch reviewer, with me only really being needed to explain 
> why the current behavior is as it is.
I'm not sure about that,.. I didn't intend the behavior change,.. the code generation we have didn't suffer any behaviour changes, only the printed AST did... which is not something we see in real benchmarks and does not even give "wrong" information to the user.

> >I either (a) commit this patch with the changed comments, or (b) 
> >rebase the next one to work without this one and forget about it for now.
> >
> >Just tell me if you want to stick with the old flags for now.
> 
> I like the current idea of having the isl ast show to the user what 
> will be code generated and would like to preserve this behavior if 
> there are no strong reasons against it.
I told you why this is not always the case at the moment and gave a couple of reasons why we should not focus on that idea for the future, but if you think that we should stick with the current implementation, fine with me.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0002-Refactor-Remove-unecessary-check-and-function.patch
Type: application/octet-stream
Size: 7549 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20140731/d2b64c2b/attachment.obj>