[llvm-commits] [llvm] r134907 - /llvm/trunk/utils/TableGen/

[Chris Lattner]

On Jul 13, 2011, at 9:32 AM, David A. Greene wrote:

> Chris Lattner <clattner at apple.com> writes:
> 
>>> Does this sound reasonable?
>> 
>> No, this doesn't.  Not at all.  "If there is no response forthcoming,
>> the submitter may commit the patch" is tantamount to ignoring process.
> 
> Here's what I find confusing.  On the one hand, everyone argues about
> "following process" which seems to implicitly mean sending patches for
> review ahead of committing them.
> 
> On the other hand, I have been told that if patches are "small enough"
> go ahead and commit them and we'll post-commit review them.
> 
> So I'm left in the same situation.  Do I commit this patch or not?  If
> not, how long do I wait on non-responses before I go ahead and commit?

The patches you're proposing are clearly not "small enough".  Are you actually asking what the dividing line is here?

> I'm really trying to work with the system.

I know, thank you!!

>  I've done a ton of work over
> the last couple of days to break up this patch.  Even learned some new
> git-fu which will come in handy down the road.  So the exercise has been
> a little helpful, at least.  But now what?  Do I go ahead and commit the
> pieces or not?

Please propose the pieces, the "::get()-ification" and "not mutating things" pieces are great steps.  Thanks!

-Chris