Linux 2.6.27

Linux 2.6.27

[Linus Torvalds]

So not a lot happened after -rc9, which is mostly good. I certainly don't 
want to have any upheavals at the end of an -rc series, although I 
wouldn't have minded more people looking at the regression list. That 
said, I think we're in pretty good shape, and it's not like I can hold 
back releases indefinitely anyway.

The shortlog from -rc9 really is small, and I doubt many people will 
notice the changes. And as usual, if you want to know about the big
changes since 2.6.26, the full ChangeLog is available, but people are 
much better off with the human-readable explanations at

	http://kernelnewbies.org/LinuxChanges

but most important is obviously to just test it. Especially if you've been 
a bad boy (or girl) and haven't been testing the -rc's. 

		Linus

---
Ali Saidi (1):
      tcp: Fix possible double-ack w/ user dma

Alistair John Strachan (2):
      hwmon: (abituguru3) Enable reading from AUX3 fan on Abit AT8 32X
      hwmon: (abituguru3) Enable DMI probing feature on Abit AT8 32X

Corentin Chary (1):
      eeepc-laptop: Fix hwmon interface

Daniele Lacamera (1):
      tcp: Fix tcp_hybla zero congestion window growth with small rho and large cwnd.

Darrick J. Wong (2):
      hwmon: Define sysfs interface for energy consumption register
      hwmon: (adt7473) Fix some bogosity in documentation file

David S. Miller (2):
      Revert "ax25: Fix std timer socket destroy handling."
      ax25: Quick fix for making sure unaccepted sockets get destroyed.

Herbert Xu (1):
      net: Fix netdev_run_todo dead-lock

Jarek Poplawski (1):
      netrom: Fix sock_orphan() use in nr_release

Jean Delvare (1):
      hwmon: (it87) Prevent power-off on Shuttle SN68PT

Linus Torvalds (3):
      Revert "V4L/DVB (8904): cx88: add missing unlock_kernel"
      Don't allow splice() to files opened with O_APPEND
      Linux 2.6.27

Marcel Holtmann (3):
      [Bluetooth] Fix double frees on error paths of btusb and bpa10x drivers
      [Bluetooth] Add reset quirk for new Targus and Belkin dongles
      [Bluetooth] Add reset quirk for A-Link BlueUSB21 dongle

Matt Mackall (2):
      SLOB: fix bogus ksize calculation
      SLOB: fix bogus ksize calculation fix

Németh Márton (1):
      [CPUFREQ] correct broken links and email addresses

Patrick McHardy (1):
      net: only invoke dev->change_rx_flags when device is UP

Ralf Baechle (1):
      [MIPS] Sibyte: Register PIO PATA device only for Swarm and Litte Sur

Linux 2.6.27 will be a longtime supported kernel

[Adrian Bunk]

2.6.16 has become a bit dated, and I'll maintain 2.6.27 for a few years 
as a replacement.

As with 2.6.16, I'll pickup maintainance when the normal -stable 
maintainance ends (at some point after 2.6.28 gets released in January).

It is intended to fill the niche for users who are not using 
distribution kernels but want to use a regression-free kernel for a 
longer time. It might be a small part of the userbase, but after the 
experiences with 2.6.16 I can say that there are quite a few users
who appreciate such an offering.

As people might have noted I wasn't able to do much kernel stuff 
recently, and 2.6.16 also suffered from this. I'll bring 2.6.16 back 
into shape and keep it maintained until at least mid-2009.

cu
Adrian

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed

Re: [stable] Linux 2.6.27 will be a longtime supported kernel

[Greg KH]

On Sun, Oct 12, 2008 at 12:03:40AM +0300, Adrian Bunk wrote:
> 2.6.16 has become a bit dated, and I'll maintain 2.6.27 for a few years 
> as a replacement.

Hm, you're not the only one going to be doing this, there are at least
three distros basing their latest releases on this kernel :)

> As with 2.6.16, I'll pickup maintainance when the normal -stable 
> maintainance ends (at some point after 2.6.28 gets released in January).

I was considering hanging on to .27 for a while longer than the .28
release, much like .25 has been working out.  But probably some time
after .29 I'll be glad to hand it off to you for you to maintain like
.16 was.  Is that ok?

> It is intended to fill the niche for users who are not using 
> distribution kernels but want to use a regression-free kernel for a 
> longer time. It might be a small part of the userbase, but after the 
> experiences with 2.6.16 I can say that there are quite a few users
> who appreciate such an offering.

Hey, I personally liked it as it made maintaining SLES a lot easier,
thanks a lot for your work there!

thanks,

greg k-h

Re: [stable] Linux 2.6.27 will be a longtime supported kernel

[Adrian Bunk]

On Sat, Oct 11, 2008 at 09:02:23PM -0700, Greg KH wrote:
> On Sun, Oct 12, 2008 at 12:03:40AM +0300, Adrian Bunk wrote:
>...
> > As with 2.6.16, I'll pickup maintainance when the normal -stable 
> > maintainance ends (at some point after 2.6.28 gets released in January).
> 
> I was considering hanging on to .27 for a while longer than the .28
> release, much like .25 has been working out.  But probably some time
> after .29 I'll be glad to hand it off to you for you to maintain like
> .16 was.  Is that ok?
>...

Sure - I don't want to disturb the normal -stable maintainance, and 
whenever you stop maintaining 2.6.27 I'll pick it up.

> thanks,
> 
> greg k-h

cu
Adrian

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed

Re: [stable] Linux 2.6.27 will be a longtime supported kernel

[Willy Tarreau]

On Sat, Oct 11, 2008 at 09:02:23PM -0700, Greg KH wrote:
> On Sun, Oct 12, 2008 at 12:03:40AM +0300, Adrian Bunk wrote:
> > 2.6.16 has become a bit dated, and I'll maintain 2.6.27 for a few years 
> > as a replacement.
> 
> Hm, you're not the only one going to be doing this, there are at least
> three distros basing their latest releases on this kernel :)

Oh that's good news as it will give it more exposure and bugs will be
found and fixed faster. Having 2.6.25 in opensuse may have contributed
to its stability.

> > As with 2.6.16, I'll pickup maintainance when the normal -stable 
> > maintainance ends (at some point after 2.6.28 gets released in January).
> 
> I was considering hanging on to .27 for a while longer than the .28
> release, much like .25 has been working out.

BTW, how long to you think you'll keep pushing fixes for .25 ? It's been
working really well and might have been the most reliable 2.6 I've seen,
2.6.16 aside. I'll keep using it waiting for 2.6.27 to settle down a bit.

> Hey, I personally liked it as it made maintaining SLES a lot easier,
> thanks a lot for your work there!

Not only there, but everywhere you need to embed a 2.6 without too many
features and without an easy ability to upgrade it, 2.6.16 has been a
good choice.

Thanks to you guys for the time you spend maintaining stable releases!
I really hope that 2.6.27 will be a viable replacement for 2.4 in the
long term.

Willy

Re: Linux 2.6.27 will be a longtime supported kernel

[Américo Wang]

On Sun, Oct 12, 2008 at 12:03:40AM +0300, Adrian Bunk wrote:
>2.6.16 has become a bit dated, and I'll maintain 2.6.27 for a few years 
>as a replacement.
>
>As with 2.6.16, I'll pickup maintainance when the normal -stable 
>maintainance ends (at some point after 2.6.28 gets released in January).
>
>It is intended to fill the niche for users who are not using 
>distribution kernels but want to use a regression-free kernel for a 
>longer time. It might be a small part of the userbase, but after the 
>experiences with 2.6.16 I can say that there are quite a few users
>who appreciate such an offering.
>
>As people might have noted I wasn't able to do much kernel stuff 
>recently, and 2.6.16 also suffered from this. I'll bring 2.6.16 back 
>into shape and keep it maintained until at least mid-2009.

Thanks for your work.

I am just wondering why you only selected .16 and .27, not others?
Are there some specical reasons why you did this?

Thank you.


-- 
"Sometimes the only way to stay sane is to go a little crazy."

Re: Linux 2.6.27 will be a longtime supported kernel

[Adrian Bunk]

On Mon, Oct 13, 2008 at 12:27:01PM +0100, Américo Wang wrote:
> On Sun, Oct 12, 2008 at 12:03:40AM +0300, Adrian Bunk wrote:
> >2.6.16 has become a bit dated, and I'll maintain 2.6.27 for a few years 
> >as a replacement.
> >
> >As with 2.6.16, I'll pickup maintainance when the normal -stable 
> >maintainance ends (at some point after 2.6.28 gets released in January).
> >
> >It is intended to fill the niche for users who are not using 
> >distribution kernels but want to use a regression-free kernel for a 
> >longer time. It might be a small part of the userbase, but after the 
> >experiences with 2.6.16 I can say that there are quite a few users
> >who appreciate such an offering.
> >
> >As people might have noted I wasn't able to do much kernel stuff 
> >recently, and 2.6.16 also suffered from this. I'll bring 2.6.16 back 
> >into shape and keep it maintained until at least mid-2009.
> 
> Thanks for your work.
> 
> I am just wondering why you only selected .16 and .27, not others?
> Are there some specical reasons why you did this?


If the question is why it isn't .26 or .28:

I'm checking whether my computer works fine with a kernel, and also 
very slightly what distributions might use a kernel, but in the end
it's pretty random.


If the question is about the big gap between them:

I don't want to maintain 10 kernels at once, and also for getting fixes 
from other kernel developers it's better when I don't do this for too 
many kernels.


> Thank you.

cu
Adrian

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed

Re: Linux 2.6.27 will be a longtime supported kernel

[Lennart Sorensen]

On Mon, Oct 13, 2008 at 03:37:05PM +0300, Adrian Bunk wrote:
> If the question is why it isn't .26 or .28:
> 
> I'm checking whether my computer works fine with a kernel, and also 
> very slightly what distributions might use a kernel, but in the end
> it's pretty random.

So any distributions using .27 by the looks of it?  It looks like Debian
is going with .26 for the 5.0 release, so if nothing else they will have
to be maintaining that one for quite a while.  I suppose having someone
maintain .27 makes sense in that case.

-- 
Len Sorensen

Re: [stable] Linux 2.6.27 will be a longtime supported kernel

[Greg KH]

On Tue, Oct 14, 2008 at 04:10:54PM -0400, Lennart Sorensen wrote:
> On Mon, Oct 13, 2008 at 03:37:05PM +0300, Adrian Bunk wrote:
> > If the question is why it isn't .26 or .28:
> > 
> > I'm checking whether my computer works fine with a kernel, and also 
> > very slightly what distributions might use a kernel, but in the end
> > it's pretty random.
> 
> So any distributions using .27 by the looks of it?  It looks like Debian
> is going with .26 for the 5.0 release, so if nothing else they will have
> to be maintaining that one for quite a while.  I suppose having someone
> maintain .27 makes sense in that case.

Fedora, Ubuntu, and openSUSE are going to be based on .27 from what I
can tell.

thanks,

greg k-h

Re: [stable] Linux 2.6.27 will be a longtime supported kernel

[Yinghai Lu]

On Tue, Oct 14, 2008 at 2:04 PM, Greg KH <greg@kroah.com> wrote:

> Fedora, Ubuntu, and openSUSE are going to be based on .27 from what I
> can tell.

opensuse 10.3 is 2.6.22
opensuse 11 is 2.6.25
so opensuse 12 will be 2.6.27? it is supposed to use 2.6.28, right?

YH

Re: [stable] Linux 2.6.27 will be a longtime supported kernel

[Jiri Kosina]

On Tue, 14 Oct 2008, Yinghai Lu wrote:

> > Fedora, Ubuntu, and openSUSE are going to be based on .27 from what I
> > can tell.
> opensuse 10.3 is 2.6.22
> opensuse 11 is 2.6.25
> so opensuse 12 will be 2.6.27? it is supposed to use 2.6.28, right?

FWIW -- opensuse 12 is out of question right now. 11.1 is the upcoming 
release (based on 2.6.27).

-- 
Jiri Kosina

Re: [stable] Linux 2.6.27 will be a longtime supported kernel

[Greg KH]

On Tue, Oct 14, 2008 at 03:43:26PM -0700, Yinghai Lu wrote:
> On Tue, Oct 14, 2008 at 2:04 PM, Greg KH <greg@kroah.com> wrote:
> 
> > Fedora, Ubuntu, and openSUSE are going to be based on .27 from what I
> > can tell.
> 
> opensuse 10.3 is 2.6.22
> opensuse 11 is 2.6.25
> so opensuse 12 will be 2.6.27? it is supposed to use 2.6.28, right?

openSUSE 11.1 will be 2.6.27.  As it is to be released pretty soon now,
it would be a bit hard to base it off of .28 which isn't even out yet :)

thanks,

greg k-h

Re: [stable] Linux 2.6.27 will be a longtime supported kernel

[Chris Friesen]

Greg KH wrote:
> On Tue, Oct 14, 2008 at 04:10:54PM -0400, Lennart Sorensen wrote:
> 
>>On Mon, Oct 13, 2008 at 03:37:05PM +0300, Adrian Bunk wrote:
>>
>>>If the question is why it isn't .26 or .28:
>>>
>>>I'm checking whether my computer works fine with a kernel, and also 
>>>very slightly what distributions might use a kernel, but in the end
>>>it's pretty random.
>>
>>So any distributions using .27 by the looks of it?  It looks like Debian
>>is going with .26 for the 5.0 release, so if nothing else they will have
>>to be maintaining that one for quite a while.  I suppose having someone
>>maintain .27 makes sense in that case.
> 
> 
> Fedora, Ubuntu, and openSUSE are going to be based on .27 from what I
> can tell.

According to distrowatch, Mandriva and Gentoo as well.

Chris

Re: [stable] Linux 2.6.27 will be a longtime supported kernel

[Greg KH]

On Tue, Oct 14, 2008 at 05:03:00PM -0600, Chris Friesen wrote:
> Greg KH wrote:
>> On Tue, Oct 14, 2008 at 04:10:54PM -0400, Lennart Sorensen wrote:
>>> On Mon, Oct 13, 2008 at 03:37:05PM +0300, Adrian Bunk wrote:
>>>
>>>> If the question is why it isn't .26 or .28:
>>>>
>>>> I'm checking whether my computer works fine with a kernel, and also very 
>>>> slightly what distributions might use a kernel, but in the end
>>>> it's pretty random.
>>>
>>> So any distributions using .27 by the looks of it?  It looks like Debian
>>> is going with .26 for the 5.0 release, so if nothing else they will have
>>> to be maintaining that one for quite a while.  I suppose having someone
>>> maintain .27 makes sense in that case.
>> Fedora, Ubuntu, and openSUSE are going to be based on .27 from what I
>> can tell.
>
> According to distrowatch, Mandriva and Gentoo as well.

Gentoo always uses whatever is released, once .28 is out, it will switch
to that after a short while :)

thanks,

greg k-h

Re: Linux 2.6.27 will be a longtime supported kernel

[Alex Howells]

Adrian Bunk wrote:
> It is intended to fill the niche for users who are not using 
> distribution kernels but want to use a regression-free kernel for a 
> longer time. It might be a small part of the userbase, but after the 
> experiences with 2.6.16 I can say that there are quite a few users
> who appreciate such an offering.

Your work is very much appreciated, we use 2.6.16 extensively and are
moving to base things off 2.6.27 for the future.  It's nice to know
it'll be supported for the foreseeable future.

++

Re: Linux 2.6.27 will be a longtime supported kernel

[Nick Piggin]

On Monday 13 October 2008 23:42, Alex Howells wrote:
> Adrian Bunk wrote:
> > It is intended to fill the niche for users who are not using
> > distribution kernels but want to use a regression-free kernel for a
> > longer time. It might be a small part of the userbase, but after the
> > experiences with 2.6.16 I can say that there are quite a few users
> > who appreciate such an offering.
>
> Your work is very much appreciated, we use 2.6.16 extensively and are
> moving to base things off 2.6.27 for the future.  It's nice to know
> it'll be supported for the foreseeable future.

Agreed, I think it is very useful work for many consumers of the
kernel.

Re: Linux 2.6.27 will be a longtime supported kernel

[Chris Friesen]

Adrian Bunk wrote:
> 2.6.16 has become a bit dated, and I'll maintain 2.6.27 for a few years 
> as a replacement.

Consider this another "thank you" for the work that you and also the 
-stable team are doing.

For various reasons (long release cycles, long-term support contracts, 
etc.) the release-early/often model can be difficult.  These periodic 
longer-term kernels are very useful.

Chris

Re: Linux 2.6.27

[Bill Davidsen]

Linus Torvalds wrote:
> So not a lot happened after -rc9, which is mostly good. I certainly don't 
> want to have any upheavals at the end of an -rc series, although I 
> wouldn't have minded more people looking at the regression list. That 
> said, I think we're in pretty good shape, and it's not like I can hold 
> back releases indefinitely anyway.
> 
> The shortlog from -rc9 really is small, and I doubt many people will 
> notice the changes. And as usual, if you want to know about the big
> changes since 2.6.26, the full ChangeLog is available, but people are 
> much better off with the human-readable explanations at
> 
> 	http://kernelnewbies.org/LinuxChanges
> 
It seems redundant to have the table of contents include links which point back
to the line containing the link. Ex: Clicking on the "3 network" link in "8
drivers" takes you back to the link with that link, not to the description which
you find by scrolling down to the entry you want. Perhaps not worth doing at all.

The description is fine, just unintuitive to use, most documents have a contents
click take you to the material you wnat to read...

And the athl1e driver seems to work really well, I've been running the -rc7 with
that network, PAE, and heavily using KVM without any problems to report. Another
good job.

-- 
Bill Davidsen <davidsen@tmr.com>
   "We have more to fear from the bungling of the incompetent than from
the machinations of the wicked."  - from Slashdot

Re: Linux 2.6.27

[Stefan Richter]

Bill Davidsen wrote:
>> 	http://kernelnewbies.org/LinuxChanges
>> 
> It seems redundant to have the table of contents include links which point back
> to the line containing the link. Ex: Clicking on the "3 network" link in "8
> drivers" takes you back to the link with that link, not to the description which
> you find by scrolling down to the entry you want.

The anchors at the LinuxChanges page are wrong.
Try http://kernelnewbies.org/Linux_2_6_27 .
-- 
Stefan Richter
-=====-==--- =-== -===-
http://arcgraph.de/sr/

Re: Linux 2.6.27

[Bill Davidsen]

Stefan Richter wrote:
> Bill Davidsen wrote:
>   
>>> 	http://kernelnewbies.org/LinuxChanges
>>>
>>>       
>> It seems redundant to have the table of contents include links which point back
>> to the line containing the link. Ex: Clicking on the "3 network" link in "8
>> drivers" takes you back to the link with that link, not to the description which
>> you find by scrolling down to the entry you want.
>>     
>
> The anchors at the LinuxChanges page are wrong.
> Try http://kernelnewbies.org/Linux_2_6_27 .
>   
Thank you, those are indeed correct.

-- 
Bill Davidsen <davidsen@tmr.com>
  "Woe unto the statesman who makes war without a reason that will still
  be valid when the war is over..." Otto von Bismark