[llvm] r228180 - Utils: Resolve cycles under distinct MDNodes

Duncan P. N. Exon Smith dexonsmith at apple.com
Wed Feb 4 13:44:40 PST 2015


> On 2015-Feb-04, at 13:30, Hans Wennborg <hans at chromium.org> wrote:
> 
> On Wed, Feb 4, 2015 at 1:16 PM, Duncan P. N. Exon Smith
> <dexonsmith at apple.com> wrote:
>> 
>>> On 2015-Feb-04, at 13:02, Duncan P. N. Exon Smith <dexonsmith at apple.com> wrote:
>>> 
>>> 
>>>> On 2015-Feb-04, at 13:00, Hans Wennborg <hans at chromium.org> wrote:
>>>> 
>>>> On Wed, Feb 4, 2015 at 11:44 AM, Duncan P. N. Exon Smith
>>>> <dexonsmith at apple.com> wrote:
>>>>> Author: dexonsmith
>>>>> Date: Wed Feb  4 13:44:34 2015
>>>>> New Revision: 228180
>>>>> 
>>>>> URL: http://llvm.org/viewvc/llvm-project?rev=228180&view=rev
>>>>> Log:
>>>>> Utils: Resolve cycles under distinct MDNodes
>>>>> 
>>>>> Track unresolved nodes under distinct `MDNode`s during `MapMetadata()`,
>>>>> and resolve them at the end.  Previously, these cycles wouldn't get
>>>>> resolved.
>>>> 
>>>> PR22456 reproduces on the 3.6 branch. Should we merge this fix?
>>> 
>>> Yes we should, but it doesn't merge cleanly since `MapMetadata()` has
>>> been essentially rewritten since then.
>>> 
>>> I'm just compiling a fix now.  I'll let you know when I'm done.
>> 
>> The attached passes `make check`.  I would have committed myself, but
>> I wasn't sure whether you wanted a particular format (or how to update
>> the svn mergeinfo given that I wrote the patch on a git branch).
> 
> Thanks! Feel free to commit this on the branch (or let me know, and
> I'll do it). Don't worry about the svn mergeinfo; the commit message
> explains what's going on, so that's good enough.
> 
> Thanks,
> Hans

r228199.



More information about the llvm-commits mailing list