Improving the handling of PR:s
Jeremie Le Hen
jeremie at le-hen.org
Sun Jan 13 04:56:37 PST 2008
On Sat, Jan 12, 2008 at 10:16:22PM -0600, Mark Linimon wrote:
> On Sat, Jan 12, 2008 at 10:56:24PM +1100, Peter Jeremy wrote:
> > This PR brings up an issue that has been mentioned elsewhere in this
> > thread: Even where a PR contains a patch, a committer may be unwilling
> > to commit the change because they are unable to verify the patch
> > themselves.
> We need to generate some kind of way for people to volunteer to test
> some of these patches. e.g. "I am willing to test patches affecting
> the USB subsystem, and I have the following list of parts: ..."
I've just read the very interesting sub-thread started by Peter Schuller
two days ago.
As you guys have just said, it seems a major problem with PR relating to
drivers is that committers don't own the exact same hardware so are not
able to verify the patch correctness. On the other hand, I'm pretty
confident there is a non-marginal number of people (read,
non-committers) are willing to help handling PR: I'm not talking of any
official hat here, but those people could contribute to a database with
the hardware they own (think, pciconf(8)) so that they can be contacted
back if a patch affects the hardware they are using. I think that a
fair amount of folks running -CURRENT can and will help for this, why
would they be running it otherwise?
This practice is a step toward Peter Schuller's proposition of having a
group of self-proclaimed non-committers giving feed back to the PR.
Jeremie Le Hen
< jeremie at le-hen dot org >< ttz at chchile dot org >
More information about the freebsd-current