[LLVMdev] [cfe-dev] !!! 3.2 Release branch patching and the Code Owners

[Duncan Sands]

Hi Pawel,

>> Can you provide some examples of the problems you are seeing?
>
> Here is what happens.
>
> I get a message "could you please include/add/merge this r16xxxx into
> 3.2?". And my immediate reaction is sure, no problem this fixes
> PR/issue/crash so it is important. But are you the code owner
> and do you approve? So I have to go and start checking because
> that is the process. In the past few days CODE_OWNERS.TXT
> on the trunk has been changing while 3.2 has been stable,
> I work on 3.2 branch so I have sent couple of e-mails
> to wrong people.
>
> Anyway, it was not my intention to cause message storm and this is
> taking way too much bandwidth on the list. As always, change is
> causing breakages until we all learn how to do it efficiently.
> I have now a way to identify the code owners.
>
>
> All I am asking is for the code owners to state clearly that
> the change they ask for is approved. Couple of examples
> from real e-mails:
>
>
> approved APPROVED *approved*

what if there is no code owner?  I sent you a reassociate patch.  I may well be
the person who knows reassociate best these days, but I'm not the code owner
(there isn't one).  Should Chris be asked to confirm the patch is OK?

I likewise sent you an instruction combine patch.  I wrote the original wrong
code that the patch fixed, so I know it rather well.  As far as I know there is
no instruction combine code owner.  So again should Chris confirm?

Ciao, Duncan.

PS: The fixes in question:

http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20121112/155994.html
http://lists.cs.uiuc.edu/pipermail/llvm-commits/Week-of-Mon-20121112/156206.html

>
>
> Pawel
>
> P.S.
> I'll even take
> "I am the code owner and I approve this message!"
>
> _______________________________________________
> LLVM Developers mailing list
> LLVMdev at cs.uiuc.edu         http://llvm.cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
>