kernel weekly news – 23.04.2011

Posted: April 23, 2011 in kernel

Hello everyone! …and welcome to this week’s kwn!

-Greg Kroah Hartman announces the release of 2.6.32.37/38,
2.6.33.10/11 and some usb tree fixes for 2.6.39, Eric van Hensbergen
announces 9p updates for -rc3, Ingo Molnar has futex, perf,
scheduler, timer and x86 fixes and Paul Gortmaker announces
the release of the 2.6.34.9 kernel , which is a long term
support release and can be found here :

http://git.kernel.org/?p=linux/kernel/git/longterm/linux-2.6.34.y.git

or here :
ftp://ftp.kernel.org/pub/linux/kernel/v2.6/longterm/v2.6.34/

-Benjamin Herrenschmidt has announced some powerpc updates,
Chris Mason has updates to the btrfs tree –

 Hi everyone,

The master branch of the btrfs unstable repo:

git://git.kernel.org/pub/scm/linux/kernel/git/mason/btrfs-unstable.git master

Has a nice pile of btrfs fixes.  Josef tackled a variety of DIO problems
including latencies and crc errors.  He also fixed a crash in the
mount -o free_space_cache mode, and dramatically reduced contention on
the journal mutex.  Less contention means we finish up endio processing
faster and we're able to process delalloc much faster as space gets
tight.

Our ENOSPC fixes have returned.  The cluster allocation code was
sometimes getting out of hand and leaving us with a filesystem full of
metadata chunks.

Fujitsu continues a long series of error handling improvements and
fixes, and Sergei Trofimovich fixed a memcpy vs memmove bug.

I've removed a duplicate commit from the shortlog below.  Josef and I
both managed to take his iov_base fix.  Git did the right thing and the
resulting code is correct.

-, Dmitry Torokhov has input
updates for -rc3 and Jens Axboe comes with block fixes , also
for -rc3 :

 With this in, I believe the only regression there's left is the one in
IDE which seems to be due to endless media events being generated. In
this round:

- Removal of the ->unplugged_fn(), as MD could not really use it. This
  adds a callback list instead, which MD can use to get notification on
  the level it requires.

- Neils MD fixes to fix with write intent bitmap hanging.

- Small fixup to use kblockd workqueue for the delay work, that was an
  oversight (it used the system workqueue instead).

- Small improvement to drop the queue lock before doing the async
  kblockd wakeup.

Please pull. 

-Linus Torvalds announces kernel 2.6.39-rc4, with the message :

 So things have sadly not continued to calm down even further. We had
more commits in -rc4 than we had in -rc3, and I sincerely hope that
upward trend doesn't continue.

That said, so far the only thing that has really caused problems this
release cycle has been the block layer plugging changes, and as of
-rc4 the issues we had with MD should hopefully now be behind us. So
we're making progress on that front too.

The plugging code still seems to trigger some issue with what looks
like an infinite stream of disk-change notifications on CD-ROMs - but
Jens is hopefully going to squish that problem soon. In the meantime,
you can avoid the problem by either running SMP or having preemption
enabled.

Other than that? We may have a bit more commits than in -rc3, but it
hasn't been _too_ bad. There's certainly nothing overly exciting:
aside from the block/MD fixups, we've got some filesystem updates
(btrfs, cifs and ubifs) and some driver updates (the largest chunk of
which is actually a duplicate driver removal). USB, some KMS, nothing
really earthshaking.

Shortlog appended for the curious.

                 Linus 

Since the shortlog ain’t up to its’ name, meaning it’s quite a beast,
I recommend you fire up your browser and find the log
@ kernel.org. :-)

-Ingo Molnar updates the perf and timer trees, Greg
Kroah Hartman starts the review for 2.6.38.4-rc1-stable,
David Miller has some networking updates -

 It's the all new Linux networking, now with less OOPS:

1) inetpeer tree traversal uses an on-stack stack, which is fine, but
   when we miss and end up creating a new entry we can trigger into
   the tree cleanup code which will allocate yet another one of these
   stacks on the stack.

   Several reported crashes are believed to be caused by this.

   Fix by sharing the stack state between these two code paths, from
   Eric Dumazet.

2) igmp_max_memberships has real meaning even when CONFIG_IP_MULTICAST
   is not set, so do not use that config knob to control whether the
   sysctl for it is available or not.  Fix from Joakim Tjernlund.

3) Fix double skb free in connector, from Patrick McHardy.

4) sis900 and natsemi forget to store permanent mac address, fix from
   Otavio Salvador.

5) irda_sendmsg() has locking imbalance, fix from Dave Jones.

6) ieee802154 has bogus cflags overrides in it's makefile, causes build
   issues of various sorts (reported by Dave Jones).  Just kill them.

7) We properly disable TSO when scatter-gather is disabled for a device,
   but we forget to handle TSO6 too.

   Similarly, when both TSO features are turned off it makes no sense
   to have TSO_ECN turned on.

   Fixes from Ben Hutchings.

8) SCTP oops fixes wrt retransmits and transport references to retrans
   paths, from Wei Yongjun.

9) ipset bug fixes from Jozsef Kadlecsik.  Dumping order was wrong, and
   entry reclaim was buggy.

10) OF layer conversion of mpc5xxx_can driver broke the build, fix
    from Anatolij Gustschin.

11) qlcnic and netxen have fragment handling issues causing deadlocks,
    fixes from Amit Kumar Salecha.

12) CAIF needs to use list_for_each_entry_safe() in cfmuxl_ctrlcmd() since
    the loop removes nodes.  Fix from Sjur Brændeland.

13) Length check is busted, and assumes linear SKBs, in llc_fixup_skb().
    Fixes bugzilla #32872

14) sfc driver fixes from Ben Hutchings, loopback self-tests leave
    TX queues active, and online self-test had unintended side effects.

15) Comment and log message typo fixes from Weiping Pan.

16) Fix some more cases of IP option parsing when coming out of a
    bridge, from Eric Dumazet.  Also, make ip_options_compile()
    resilient to packets with no attached route.

17) FLAG_POINTTOPOINT and FLAG_MULTI_PACKET overlap in usbnet layer,
    oops.  Fix from Huajun Li.

18) RX path memory leak in ban driver, from Debashis Dutt.

19) Allowing VLAN over loopback causes crashes, particularly in ipv6
    routing, simply disallow this.  From Krishna Kumar.

20) Revert TCP bind() change that causes regressions for various
    programs, such as haproxy and amavisd.  Fixes kernel bugzilla 32832.

Please pull, thanks a lot!

– and we have another review
from Greg, this time for 2.6.33.12-longterm .

-Dave Airlie announces drm fixes for radeon and nouveau, another review
awaits us for 2.6.32.39-longterm, Steven Rostedt has updates for
the trace subtree, Greg Kroah Hartman announces 2.6.38.4 and
since the weekend’s here, I’ll be off and see you next time!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s