[lldb-dev] large patches vs incremental changes

[Chandler Carruth]

On Tue, Jul 1, 2014 at 12:01 PM, Zachary Turner <zturner at google.com> wrote:

> Additionally, I frequently see patches going in without review.  What is
> LLDB's policy on this?


FWIW, I would assume (from the discussions with Chris when LLDB first
joined the LLVM projects) that it has the same policy as LLVM here. Chris,
I don't actually see any clarification about this on the LLDB website,
could we get that in there? I'm somewhat concerned that there is
essentially *no* developer policy posted for LLDB.

However, I want to point out that in LLVM you might see many patches going
in "without review" because the code review takes place *post-commit*. I
suspect that LLDB is much the same here, and a substantial place for code
review to take place is post-commit rather than pre-commit. Using
post-commit review to maintain a reasonable development velocity for
long-standing contributors is (IMO) an important and good part of the LLVM
development process.


>  Are we're ok with having broken code upstreamed for a short time on the
> condition that the committer is acting in good faith to fix it as soon as
> possible?


I think this is a separate question. Build bots should *never* be left
broken, etc.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/lldb-dev/attachments/20140701/adba0d4d/attachment.html>