Hello, ladies and gents and be welcome to this week’s edition of kwn!
-We kickoff the week with Stefan Richter’s git pull request for some fixes and updates
on the firewire tree, while David Miller has some networking updates, introduced by him
as follows : “Here are the bug fixes that have queued up since -rc1,
several “touches netdev queue before device registry”
cases as well as the “net dst” percpu fixup you want to see merged ASAP.
Please pull, thanks a lot!” . Other updates/pull requests for this first part of the week include
Marcelo Tosatti’s kvm-related pull request (-rc1), Paul Mundt sh and rmobile updates (-rc2),
Greg KH – tty updates (-rc1) and Theodore Ts’o with ext4 bugfixes for -rc2.
-Paul Mundt announced the birth of a new git tree, namely fbdev. Here’s the full text :
“As I find myself spending more time reviewing and merging fbdev patches
lately, and in an effort to offload some of the pressure on -mm, I’ve
opted to start a git tree to start collecting and merging fbdev changes.
I plan to get a patchwork queue established for the list as well so that
things don’t fall through the cracks (the link will be posted once it’s
been established). People will be able to see the status of their patches
this way, rather than wondering whether things are being picked up,
overlooked, or forced through an alternate merge strategy (We have had
reasonable success with this with the SH tree, so here’s hoping that the
same approach will work well here, too).
I’ve already made a note about this to akpm off-list, so fbdev patches
that pop up outside of the regular channels that I may be unaware of
should still have a reasonable chance of being intercepted and rerouted
(although the mmotm is rather devoid of drivers/video churn at the
I’ll be combing through the list archives to see if there is anything
glaringly outstanding (the HDMI/EDID work comes to mind) and working on
getting the oustanding issues there resolved. If anyone has any
outstanding patches that have fallen through the cracks, now would be a
good time to provide pointers!
Given the point that we’re at with .38 bits now and a cursory look at
-next shows that there’s already quite a bit of drivers/video churn
coming in from git trees. It will likely take a few cycles to get people
used to the idea of there being an active fbdev tree to merge through,
but with any luck it will be a fairly natural transition. I’ll be
contacting and working with git tree owners separately so there aren’t
The tree can be found at:
Stephen, if you could add this with the master branch to -next, it would
be appreciated. For the moment I expect most of the patches to follow a
fairly linear line of development, but there may be cause for topic
branches as new (or contested) patch sets arise.”
-Alex Elder issued an update for the XFS tree aimed @ -rc2 (or -rc3, if
it’s too late for -rc2) and John W. Linville announced a batch of updates
for the wireless tree (10.11.2010) :
“David [Miller, n.a.],
Here is a batch of fixes intended for 2.6.37. Also mixed-in are a few
device ID updates. I think the changelog entries are reasonably
documentary of the issues being fixed, so I won’t belabor them. ;-)
This also includes a round of Bluetooth fixes from Gustavo:
“The following batch contains some bugfixes for 2.6.37. A fix for unaligned
access in L2CAP, a Kconfig error, two fixes related to security of
the Bluetooth links, and support for the a MacBook Air Bluetooth device.
There is also a one line patch from Matthew Garret, that enables USB
autosuspend for btusb module, which shall be completely safe. Please
Please let me know if there are problems!”
-Ingo Molnar issued a series of fixes for the perf tree, as well as x86 fixes,
while Jens Axboe announced “block bits for for 2.6.37-rc1″ (as the subject line
These are the fixes that have been collected since the initial push.
It’s a bit larger than I would have liked, but the majority of the code
churn there is simply finishing the conversion away from the previous
REQ_HARDBARRIER. In summary:
– Two fixes for RCU locking around io priority support from Sergey and
– Four fixes from me to report issues around missing validation of parts
of the iov interface inside the block layer. All of these were
reported by Dan Rosenberg.
– Removal of copy_io_context(), it’s been unused since the AS io
scheduler was killed from me.
– Documentational pruning of the AS io scheduler from Randy Dunlap.
– An kernel stack leak to userspace has been fixed up, missing a memset,
– Some fixes for cciss, mostly from HP but a fixup from me as well.
– Correct use of i_size_read() in various places instead of direct read
of the inode size from Mike Snitzer.
– A drbd fix, and three patches to finish the transition to the new
– Killing of REQ_HARDBARRIER in the block core from Christoph.
-Jeremy Fitzhardinge has a couple of mails with pull requests regarding the Xen tree and
2.6.37-rc1, the input tree benefits from a series of updates and fixes thanks to Dmitry
Torokhov and Dave Airlie has “a bunch of drm fixes, it includes couple of regression fixers on
radeon that could cause oops/memory corruptions, along with few Intel
fixers. It also fixes the Kconfig for the poulsbo stub.”
-Last minute (literally!) updates comprise of Greg KH’s updates for the staging and USB
targetted at -rc1, as well as tty updates, also by Greg. That’s about it for me this week, see
you around and have a nice weekend!