bogus patch in 3.8.y.z extended stable

Kamal Mostafa kamal at canonical.com
Wed Sep 18 18:01:45 UTC 2013


On Wed, 2013-09-18 at 18:16 +0200, Markus Kreidl wrote:
> Could you please explain the algorithm you use to pick up patches for
> the v3.8 stable series?

We have a script which selects and tries to apply each patch marked "Cc:
stable@" from mainline, and we also try to apply David Miller's batches
of stable networking patches.  For the most part, patches which apply
cleanly get queued up.  Patches that don't apply cleanly get inspected
for relevance and then either discarded or backported and queued up.

For each patch that gets queued up, the author (and all other SOB's) are
notified.  Before the stable series release is made the entire patch set
is posted for review to lkml@, stable@, etc.  Occasionally, an
irrelevant patch may still slip through unnoticed by those checkpoints.

>  I'm mostly interested how you ended up with
> 
> commit 98aab898043fdc0f63621dae282fb5448fc39f9a
> Author: Federico Vaga <federico.vaga at gmail.com>
> 
>     net/core/sock.c: add missing VSOCK string in af_family_*_key_strings
> 
>     [ Upstream commit 456db6a4d495f40777da6f1f32f62f13026f52db ]
> 
> in your tree.

It arrived in a batch of stable networking patches and applied cleanly.

>  It is not really valid.

I agree.  Happily, it also looks harmless.  I will revert it for the
next 3.8 release.  Thanks for point it out.

 -Kamal


>  If you have no idea why this is
> bogus since Greg applied it to v3.9 tree and nobody complained there I
> suggest a think tank. If it does not help here are two hints:
> 
> - VSOCK is not part of the v3.8 kernel
> - gcc throws out a few warnings after the patch has been applied.
> 
> And? Got it?
> 
> regards MK
> 

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: This is a digitally signed message part
URL: <https://lists.ubuntu.com/archives/kernel-team/attachments/20130918/24012b44/attachment.sig>


More information about the kernel-team mailing list