This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: Reduce Dwarf Debug Size
Mike Stump wrote:
On Mar 1, 2007, at 3:03 PM, Ian Lance Taylor wrote:
Can you describe post-checkin review?
Beyond:
http://llvm.org/docs/DeveloperPolicy.html#reviews
No, I'd not claim I have any experience with it in llvm, but, I do
sit next to people that do. :-)
FWIW, this does not seem to be materially different from GCC's policies,
although it may be different in how it's applied in practice. I note
the following text, from the "obtaining commit access" section:
"You are allowed to commit patches without approval to those portions of
LLVM that you have contributed or maintain (i.e., have been assigned
responsibility for), with the proviso that such commits must not break
the build. This is a "trust but verify" policy and commits of this
nature are reviewed after they are committed."
When that rule doesn't apply -- i.e., if one only has commit-after-
approval status, or the code in question is not something that one is a
maintainer of -- then the standard rules of requiring pre-approval
apply, just as in GCC.
And the only time those rules do apply is when one is a maintainer of
the code in question. According to the GCC Write Access Policies, GCC
maintainers likewise have permission to commit code in their areas of
maintenance without getting pre-approval.
Now, it may be that culturally, llvm maintainers tend to do a lot more
commits without asking for pre-review, and it may also be that llvm has
a higher percentage of maintainers among its contributors; I would be
interesting to know if those are the case or not. But the rules
themselves seem essentially the same.
- Brooks