On Wed, Jan 25, 2012 at 3:46 AM, Philip Prindeville
<[email protected]> wrote:
> On 1/24/12 6:06 AM, Jonathan McCrohan wrote:
>> I also see svn as part of the problem. I think a move towards the
>> linux-kernel development model would be a great benefit.
>>
>> Using git would allow users to make many small fixes in their own tree
>> and send single a pull request for fixes to x,y and z to a member of the
>> patch review team for ACK or NAK who can then commit to master.
>> Hopefully this will result in fewer stray patches.
>>
>> The original user will then show up in git blame and will make tracing
>> errors far easier. Currently, unless you have commit rights, everything
>> comes from one of the few core developers and you have to manually look
>> up the changeset to figure out who is responsible for it.
>
> For those of us that submit fixes that then languish for months before being 
> committed, wouldn't that mean having to constantly rebase them?

Certainly, I do not want to fork this conversation further, and I would
hope that if the time between submittal and application can be brought down
that this ongoing problem will be reduced. It IS a pita however, and
like many I'm willing to volunteer to help, and I've seen a few suggestions
(like using git pulls from temporary repos to preserve ownership) go by
that look like further improvements on the process.

However, I would also like to find ways to deal with my other comment
on this thread:

"A second, large problem (in my mind), is that I would like to find a
process for getting stuff upstream into the mainline kernel."

Returning to handling the patch backlog:

One way to perhaps help this is for the overall schedule to be more
widely publicised and to work on more of a time based manner.
In my case I've been trying to adhere to the kernel release schedule
+ .1 + X, where .1 is the first followup to the 'stable' kernel release,
where X is the delta that it seems to take between a release and
patches appearing.

The ar71xx 3.1 process had 3? 4? patch sets go by that never made it into
openwrt proper (and I rebased each time), and the 3.2 patch set landed
a few days ago without (as best as I recall)
public review. I'm delighted that it appears to work, but confess to having
been a bit surprised. (I have been out of the loop for several weeks however)

Would adding a 'Tested-by' email, to patch sets going by, be a helpful addition?
It seems that also finding a way to track coverage sanely would be good
addition, like:

Tested-by: [email protected]
Coverage: ag71xx

Or on a proposed patch from a developer unable to fully test:

Coverage-needed: mips, arm, x86
>
> -Philip
>
> _______________________________________________
> openwrt-devel mailing list
> [email protected]
> https://lists.openwrt.org/mailman/listinfo/openwrt-devel



-- 
Dave Täht
SKYPE: davetaht
US Tel: 1-239-829-5608
FR Tel: 0638645374
http://www.bufferbloat.net
_______________________________________________
openwrt-devel mailing list
[email protected]
https://lists.openwrt.org/mailman/listinfo/openwrt-devel

Reply via email to