llvm dev - Apr 2014 - [LLVMdev] PR17975 and trunk

Hi,

PR17975 was caused by r191059 which was reverted on the 3.4 branch in
r196521.  However, the problem still occurs with trunk (confirmed as
of r206186).
>From a thread on cfe-commits I see that Kai Nacke (the author of
r191059) was working on a patch to fix PR17975, but the conversation
ends:

http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20131202/197968.html

So my question is should we revert r191059 on trunk?  From what I can
see, the code has already been refactored once in r198768 (retaining
identical functionality).  Further work may make it even more
difficult to unpick.

Thanks,
Rob.

P.S. CC-ing Kai Nacke in case he still intends to follow up on his
original patch.

--
Robert Lougher
SN Systems - Sony Computer Entertainment Group

On Apr 14, 2014, at 1:14 PM, Robert Lougher <rob.lougher at gmail.com>
wrote:
> Hi,
> 
> PR17975 was caused by r191059 which was reverted on the 3.4 branch in
> r196521.  However, the problem still occurs with trunk (confirmed as
> of r206186).
> 
> From a thread on cfe-commits I see that Kai Nacke (the author of
> r191059) was working on a patch to fix PR17975, but the conversation
> ends:
> 
>
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20131202/197968.html
> 
> So my question is should we revert r191059 on trunk?  From what I can
> see, the code has already been refactored once in r198768 (retaining
> identical functionality).  Further work may make it even more
> difficult to unpick.
> 
> Thanks,
> Rob.
> 
> P.S. CC-ing Kai Nacke in case he still intends to follow up on his
> original patch.
This is the patch in question?
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20130916/188476.html

If so, yes, absolutely revert this.

-Chris

On Mon, Apr 14, 2014 at 5:21 PM, Chris Lattner <clattner at apple.com>
wrote:
> On Apr 14, 2014, at 1:14 PM, Robert Lougher <rob.lougher at
gmail.com> wrote:
>
> > Hi,
> >
> > PR17975 was caused by r191059 which was reverted on the 3.4 branch in
> > r196521.  However, the problem still occurs with trunk (confirmed as
> > of r206186).
> >
> > From a thread on cfe-commits I see that Kai Nacke (the author of
> > r191059) was working on a patch to fix PR17975, but the conversation
> > ends:
> >
> >
>
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20131202/197968.html
> >
> > So my question is should we revert r191059 on trunk?  From what I can
> > see, the code has already been refactored once in r198768 (retaining
> > identical functionality).  Further work may make it even more
> > difficult to unpick.
> >
> > Thanks,
> > Rob.
> >
> > P.S. CC-ing Kai Nacke in case he still intends to follow up on his
> > original patch.
>
> This is the patch in question?
>
>
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20130916/188476.html
>
> If so, yes, absolutely revert this.

No, its
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20130916/188462.html

And I think it should be reverted, yes.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20140414/babef6a9/attachment.html>