>> In general, I consider it discourteous to send patches for code
>> owned and maintained by someone else directly for inclusion in the
>> kernel rather than to the author, except perhaps for critical
>> bugs.
> I agree with your concerns Leonard. But I think it is necessary to
> come up with a solution for the case where someone makes a quite
> resonable change to an interface which is used by many files in the
> kernel, encompassing perhaps 40 or 50 maintainers who would need to
> be contacted.
> What do you suppose be done in such a case?
Personally, I'd see that falling into three subgroups which have quite
different solutions, namely:
1. There is a maintainer specifically for the interface in question,
to whom the maintainers of the myriad files ought to bow when
determining how the interface they're using works.
In this case, the obvious solution is to send the patch to the
maintainer of the interface since that's who is responsible for
the relevant area anyway. Problem solved.
2. The interface in question has no specific maintainer, and is used
my files maintained by not more than six different people.
In this case, it isn't exactly hard to just send a CC of the email
with the patch to each maintainer involved, and allow them all to
have a look at the parts of the patch that affect their files.
Again, problem solved.
3. The interface in question has no specific maintainer, but is used
by files maintained by more than six different people.
In this case, the only sensible solution is for somebody to
become the maintainer of the relevant interface.
I know that if I developed a patch effecting a reasonable change to an
interface that was used by even half a dozen files with different
people maintaining them, and the interface itself had no specific
maintrainer, I would be VERY tempted to declare myself to be the
maintainer of that interface simply so I didn't have to deal with this
very situation. My reasoning would also be quite simple: If that
interface is used by sufficient different maintainers to fall in this
category, then it's important enough to have a maintainer of its own.
OK, the boundary between categories (2) and (3) could be a bit fuzzier
than stated, since different people will be willing to type in
different quantities of email addresses, but the general principle
still applies.
Comments, anybody?
Best wishes from Riley.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu