Adrian Bunk is now taking over the 2.6.16-stable branch
This is just a notice to everyone that Adrian is going to now be taking over the 2.6.16-stable kernel branch, for him to maintain for as long as he wants to.
He will still be following the same -stable rules that are documented in the Documentation/stable_kernel_rules.txt file, but just doing this for the 2.6.16 kernel tree for a much longer time than the current stable team is willing to do (we have moved on to the 2.6.17 kernel now.)
So, if you have any patches that meet the -stable requirements for the 2.6.16 kernel, please send them to him, and not the [email protected] account, as we will just delete them.
And I'd like to offer my best wishes to Adrian for doing this work. Personally I don't think it can be done for all that long of an amount of time, and I will be very happy to see him prove me wrong :)
thanks,
greg k-h
Quoted 3 lines · Greg KH
This is just a notice to everyone that Adrian is going to now be taking over the 2.6.16-stable kernel branch, for him to maintain for as long as he wants to.
Adrian, could you provide a bit of rationale as to why you want to do this? I'm just curious.
josh
Quoted 7 lines
Quoted 4 lines
This is just a notice to everyone that Adrian is going to now be taking over the 2.6.16-stable kernel branch, for him to maintain for as long as he wants to.
Adrian, could you provide a bit of rationale as to why you want to do this? I'm just curious.
Competition against the "LTS" of Ubuntu, perhaps? :-)
Jan Engelhardt
Quoted 7 lines · Josh Boyer
Quoted 3 lines · Greg KH
This is just a notice to everyone that Adrian is going to now be taking over the 2.6.16-stable kernel branch, for him to maintain for as long as he wants to.
Adrian, could you provide a bit of rationale as to why you want to do this? I'm just curious.
A long-term maintained stable series was missing in the current development model.
The 2.6 series itself is theoretically a stable series, but the amount of regressions is too big for some users.
Quoted 1 line
josh
cu
Adrian
Signature · 6 lines
Gentoo kernels are 42 times more popular than SUSE kernels among
KLive users (a service by SUSE contractor Andrea Arcangeli that
gathers data about kernels from many users worldwide).
There are three kinds of lies: Lies, Damn Lies, and Statistics.
Benjamin DisraeliQuoted 6 lines · Adrian Bunk
Quoted 5 lines · Josh Boyer
Quoted 3 lines · Greg KH
This is just a notice to everyone that Adrian is going to now be taking over the 2.6.16-stable kernel branch, for him to maintain for as long as he wants to.
Quoted 8 lines
Quoted 2 lines
Adrian, could you provide a bit of rationale as to why you want to do this? I'm just curious.
A long-term maintained stable series was missing in the current development model.
The 2.6 series itself is theoretically a stable series, but the amount of regressions is too big for some users.
Greg is electing new official maintainers, but Greg is doing other weird things as well:
http://www.cpushare.com/blog/andrea/article/42/
I'm certainly not happy with you as an official maintainer of a kernel that could potentially be used by lot of people (I say potentially because I doubt many would use it) given that recently you claimed to be happy with ext2, and you discredited the great work of many people that are testing new features and reporting feedback. I also don't trust you to act in the interest of all parties involved in the kernel community (given that most of them are funded by companies filing lots of US software patents). I don't need to comment on your signature since that's explicit enough without me needing to add anything.
Quoted 10 lines · Andrea Arcangeli
Quoted 9 lines · Adrian Bunk
Quoted 7 lines · Josh Boyer
Quoted 3 lines · Greg KH
This is just a notice to everyone that Adrian is going to now be taking over the 2.6.16-stable kernel branch, for him to maintain for as long as he wants to.
Adrian, could you provide a bit of rationale as to why you want to do this? I'm just curious.
Quoted 7 lines
Quoted 5 lines
A long-term maintained stable series was missing in the current development model.
The 2.6 series itself is theoretically a stable series, but the amount of regressions is too big for some users.
Greg is electing new official maintainers
Greg didn't "elect" anyone, Adrian volunteered to maintain something that had been dropped by the -stable developers and no one else was going to maintain.
Quoted 4 lines
, but Greg is doing other
weird things as well:http://www.cpushare.com/blog/andrea/article/42/
I'm only repeating what a whole lot of lawyers told me, nothing else.
thanks,
greg k-h
Quoted 3 lines · Greg KH
Greg didn't "elect" anyone, Adrian volunteered to maintain something that had been dropped by the -stable developers and no one else was going to maintain.
Did you ever call for a maintainer list of volunteers?
To me an official 2.6.16-stable in the hands of the only guy who proposed himself as maintainer, sounds worse than no stable tree at all. People won't know anymore if to run Greg's 2.6.18-stable or 2.6.16-stable.
If a 2.6-real-stable tree has to happen because 2.6-stable is not really stable/trustable enough, then give it up with your 2.6.18-stable and start doing 2.7 and leave 2.6 in the hands of somebody else.
An official kernel needs a critical mass to have a value, it's simply a wasted effort to open yet another official tree that will actually fragment the "production" userbase even more.
If 2.6.18-stable is sustainable with the current model, with the distro folks being the only ones forking off a real-stable tree, then you should drop 2.6.16-stable. If instead it's 2.6.18-stable that is not good enough for production usage and people really needs 2.6.16-stable, you should open 2.7, and not fragment the userbase like this.
I think it would be great to have the users choosing their preferred maintainer to end the era of maintainers being decided by other maintainers like you actually did. A simple website on kernel.org can achieve it, where users can registers for voting and the maintainers willing to maintain 2.6-stable can registers themself too. That's at least less random than the current status if what you said above is true and if 2.6.16-stable is meant to reach any critical mass.
Ar Llu, 2006-08-21 am 00:30 +0200, ysgrifennodd Andrea Arcangeli:
Quoted 6 lines · Greg KH
Quoted 3 lines
Greg didn't "elect" anyone, Adrian volunteered to maintain something that had been dropped by the -stable developers and no one else was going to maintain.
Did you ever call for a maintainer list of volunteers?
That isn't the normal way of doing things around here. Linus has always taken the approach of picking who he trusts to hand stuff onto, as have others when handing on further.
Quoted 3 lines
To me an official 2.6.16-stable in the hands of the only guy who proposed himself as maintainer, sounds worse than no stable tree at all.
2.6.16-stable is in the hands of someone that Greg (2.6.17-stable, ex 2.6.16-stable, 2.6 head temporary chief maintainer) thinks is a good person to do the job. That sounds to me quite sensible selection criteria, and Adrian is certainly up to the job.
Alan
Quoted 3 lines · Alan Cox
That isn't the normal way of doing things around here. Linus has always taken the approach of picking who he trusts to hand stuff onto, as have others when handing on further.
That doesn't make it right or better than what I proposed. Furthermore if Greg is so sure that nobody else will be willing to maintain 2.6.16-stable what I proposed will lead to the very same result of the current model, and then obviously there would be nothing left to argue about.
From: Alan Cox <[email protected]>
Date: Mon, 21 Aug 2006 00:05:02 +0100
Quoted 9 lines
Ar Llu, 2006-08-21 am 00:30 +0200, ysgrifennodd Andrea Arcangeli:
Quoted 3 lines
To me an official 2.6.16-stable in the hands of the only guy who proposed himself as maintainer, sounds worse than no stable tree at all.
2.6.16-stable is in the hands of someone that Greg (2.6.17-stable, ex 2.6.16-stable, 2.6 head temporary chief maintainer) thinks is a good person to do the job. That sounds to me quite sensible selection criteria, and Adrian is certainly up to the job.
Totally agreed.
This voting idea is absolute nonsense.
There is a tree of trust, and responsibilities like this are simply handed down.
Quoted 2 lines · David S. Miller
There is a tree of trust, and responsibilities like this are simply handed down.
That works as long as you trust and you are ok with every leaf of the tree. I'm not, both a developer and more important as an user.
I beg to remind you why Marcelo was picked, not only for his skills, but also because he was so universally friendly, and easy to talk with. I couldn't imagine Marcelo wanting to hurt anyone. To me that's important, I could never disagree with the Marcelo pick. But this is basically the opposite in my view. A maintainer needs qualities that go beyond technical skills. I think it's Greg's fault of course, I'm not blaming anybody else, but this clearly shows that the tree of trust has its limits, hence I'd like a different system to elect the new maintainers that has a better chance to work.
Andrea Arcangeli schrieb am 2006-08-21:
Quoted 9 lines
I beg to remind you why Marcelo was picked, not only for his skills, but also because he was so universally friendly, and easy to talk with. I couldn't imagine Marcelo wanting to hurt anyone. To me that's important, I could never disagree with the Marcelo pick. But this is basically the opposite in my view. A maintainer needs qualities that go beyond technical skills. I think it's Greg's fault of course, I'm not blaming anybody else, but this clearly shows that the tree of trust has its limits, hence I'd like a different system to elect the new maintainers that has a better chance to work.
I still liked your semi-technical reasons better than the personal.
Signature · 1 line
Matthias Andree
Quoted 6 lines · Andrea Arcangeli
Quoted 2 lines · David S. Miller
There is a tree of trust, and responsibilities like this are simply handed down.
That works as long as you trust and you are ok with every leaf of the tree. I'm not, both a developer and more important as an user.
As an user, that's not your problem.
As a developer, if you don't like Adrian doing that, perhaps you should have volunteered to maintain it yourself. Or you can work with Adrian to make sure he does nothing strange.
But I think Adrian is very good person to do this, and with strong review process in place we currently have.... I don't see what the problem is.
Pavel
Signature · 2 lines
(english) http://www.livejournal.com/~pavelmachek (cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
Quoted 7 lines · On Mon, 21 Aug 2006 00:30:46 +0200 Andrea Arcangeli <[email protected]> wrote:
I think it would be great to have the users choosing their preferred maintainer to end the era of maintainers being decided by other maintainers like you actually did. A simple website on kernel.org can achieve it, where users can registers for voting and the maintainers willing to maintain 2.6-stable can registers themself too. That's at least less random than the current status if what you said above is true and if 2.6.16-stable is meant to reach any critical mass.
There's no need for a vote. Users already vote for a maintainer when they decide to use a paticular kernel tree.
No user is forced to follow a particular maintainer. And anyone can step up and declare that they are also offering a maintained tree.
And this situation is already self correcting; if no users follow, it's unlikely that a maintainer will continue doing the required work. And if a maintainer doesn't do a satisfactory job, it's very unlikely many people will choose to use that tree.
Sean
A message that is not in the archive (sent off-list or not yet received): [email protected]
Quoted 5 lines · Sean
There's no need for a vote. Users already vote for a maintainer when they decide to use a paticular kernel tree.
No user is forced to follow a particular maintainer. And anyone can step up and declare that they are also offering a maintained tree.
I never said that 2.6.16-stable is going to succeed, all I'm saying is that all testing done on it will be a wasted effort, that's why there are no infinite competing trees. So one would hope that an official maintainer isn't choosed by random.
Quoted 1 line
And this situation is already self correcting; if no users follow, it's
The real ironic thing, is that the only feedback he has to know if users follow or not, is KLive. Which I'm going to shutdown anyway if nothing changes w.r.t. 2.6.16-stable since I'm not out to fight with anyone, I don't even have a degree in statistics, so it's up to the thousands getting a degree in statistics every year to argue with the -stable maintainers, certainly not me.
Quoted 7 lines · Andrea Arcangeli
On Sun, Aug 20, 2006 at 06:51:23PM -0400, Sean wrote:
...Quoted 1 line
And this situation is already self correcting; if no users follow, it's
The real ironic thing, is that the only feedback he has to know if users follow or not, is KLive.
...
If I was interested in statistics, I could also use [1] that shows statistics about running kernels automatically submitted my nearly ten times as many machines than in KLive.
2.6.16.27 is the latest 2.6.16 kernel, available for more than one month. KLive says no KLive user is using 2.6.16.27.
[1] says 0.2% of the machines reporting there are using 2.6.16.27.
If I was believing in such statistics, there was no 2.6.16 branch.
The important feedback are emails from people reporting a bug or saying "thank you".
This branch is an offer.
If it runs on millions of computers that's OK.
If it's used only by few people that's OK.
cu
Adrian
[1] http://counter.li.org/reports/systemstats.php
Signature · 4 lines
"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 SeedQuoted 10 lines · Andrea Arcangeli
Quoted 5 lines · Sean
There's no need for a vote. Users already vote for a maintainer when they decide to use a paticular kernel tree.
No user is forced to follow a particular maintainer. And anyone can step up and declare that they are also offering a maintained tree.
I never said that 2.6.16-stable is going to succeed, all I'm saying is that all testing done on it will be a wasted effort, that's why there are no infinite competing trees.
Andrea, I don't agree with you on this.
If 2.6.16 succeeds, some people who currently cannot rely on 2.6 due to its code changing too fast will be able to make the move. Also, those who need a good reliability will be able to check in one year if they consider it reliable enough for their use, based on other user's feedback.
Once those users depend on this kernel, they will probably send fixes back when they'll find a bug. Right now, many people using 2.6 just run it off the CD of their pet distro, and when something goes wrong, they decide the distro is broken and they change to anything else (which might have a different kernel version). Do not believe that everyone has enough knowledge to send valuable bug reports and/or fix bugs. Basically, you have them all reading this list on a somewhat regular basis.
Quoted 11 lines
So one would hope that an official
maintainer isn't choosed by random.Quoted 1 line
And this situation is already self correcting; if no users follow, it's
The real ironic thing, is that the only feedback he has to know if users follow or not, is KLive. Which I'm going to shutdown anyway if nothing changes w.r.t. 2.6.16-stable since I'm not out to fight with anyone, I don't even have a degree in statistics, so it's up to the thousands getting a degree in statistics every year to argue with the -stable maintainers, certainly not me.
Well, you've been fighting over opinions. Both of you have been a bit rude with the other one. Adrian has finally removed his mocking (and IMHO childish) signature. Why don't you consider the problem solved and keep your project online ?
Willy
Quoted 3 lines · Willy Tarreau
[..] Adrian has finally removed his mocking (and
IMHO childish) signature. Why don't you consider the problem solved and keep your project online ?
http://marc.theaimsgroup.com/?l=linux-kernel&m=115611817104179&w=2
http://marc.theaimsgroup.com/?l=linux-kernel&m=115592383000732&w=2
Here also a link sent to me from a KLive supporter:
http://en.wikipedia.org/wiki/W._Edwards_Deming
If I was to follow my emotions and my financial interest, KLive would be down already.
Hi Adrian,
Quoted 6 lines · Adrian Bunk
Quoted 5 lines · Josh Boyer
Quoted 3 lines · Greg KH
This is just a notice to everyone that Adrian is going to now be taking over the 2.6.16-stable kernel branch, for him to maintain for as long as he wants to.
Quoted 8 lines
Quoted 2 lines
Adrian, could you provide a bit of rationale as to why you want to do this? I'm just curious.
A long-term maintained stable series was missing in the current development model.
The 2.6 series itself is theoretically a stable series, but the amount of regressions is too big for some users.
Well, I really wish you success on this project. I completely agree on the need of a 2.4-like model to stabilize one branch of 2.6, and you probably remember that we've been talking about this 1 or 2 years ago, when I found it too hard to be started alone. It takes a very long time but apparently succeeds in the long term. I think it will quickly become a hard work because patches will get harder and harder to apply, and sometimes you'll have to adapt them a lot. But there's nothing impossible, I've been backporting fixes from 2.6 to 2.4 for a long time, so 2.6 to 2.6 should be feasible.
However, I hope that you realize that (if you succeed), your work might become a basis for some distros, as well as for some admins who will try to switch from 2.4 to 2.6. I mean, people will be *relying* on you to get fixes.
But for this, you will have to be seen as a serious person, and avoid childish fighting with other kernel developpers, such as this :
Quoted 3 lines
Gentoo kernels are 42 times more popular than SUSE kernels among KLive users (a service by SUSE contractor Andrea Arcangeli that gathers data about kernels from many users worldwide).
People who know the history will take this a as the teenager trying to take revenge on the other guy who stole his girlfriend, and people not aware of the history will take this for Gentoo advertisement based on the work of your good friend Andrea who is so much honnest that he doesn't mind publishing such results. In both cases, I think this is not what you're looking for, and it does not make you look like the serious guy on whom we can rely to get a stable kernel. At best, it will be used to show that Andrea is unbiased, which is exactly what is expected for the role you're taking.
And BTW, you'll look ridiculous when you'll post changelogs which will start with fixes from Andrea and will end with this signature ! Or maybe you'll decide to refuse his patches, which is a dangerous game when the original goal is to stabilize some code. I hope that everyone will grow up in order not to tarnish linux's image. There are already a few debian extremists to make us look like immature geeks in enterprise contexts, I think we don't need more of this.
Best wishes with 2.6.16,
Willy
Hi!
Thanks for doing this.
I believe I had 'fix pdflush after suspend' queued in Greg's tree. Is it still queued or should I resend?
Pavel
Signature · 1 line
Thanks for all the (sleeping) penguins.
Quoted 6 lines · Pavel Machek
Hi!
Thanks for doing this.
I believe I had 'fix pdflush after suspend' queued in Greg's tree. Is it still queued or should I resend?
Nothing is queued in the 2.6.16-stable tree right now, Adrian is starting with a blank list.
I'd suggest resending it to him.
thanks,
greg k-h
Quoted 6 lines · Pavel Machek
Hi!
Thanks for doing this.
I believe I had 'fix pdflush after suspend' queued in Greg's tree. Is it still queued or should I resend?
Is this "pdflush: handle resume wakeups"?
Quoted 1 line
Pavel
cu
Adrian
Signature · 6 lines
Gentoo kernels are 42 times more popular than SUSE kernels among
KLive users (a service by SUSE contractor Andrea Arcangeli that
gathers data about kernels from many users worldwide).
There are three kinds of lies: Lies, Damn Lies, and Statistics.
Benjamin DisraeliQuoted 10 lines · Adrian Bunk
Quoted 6 lines · Pavel Machek
Hi!
Thanks for doing this.
I believe I had 'fix pdflush after suspend' queued in Greg's tree. Is it still queued or should I resend?
Is this "pdflush: handle resume wakeups"?
Yes. Do you have it somewhere or should I dig it up?
Pavel
Signature · 2 lines
(english) http://www.livejournal.com/~pavelmachek (cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html
Quoted 5 lines · Pavel Machek
Quoted 4 lines · Adrian Bunk
Quoted 2 lines · Pavel Machek
Hi!
Quoted 8 lines
Quoted 6 lines
Quoted 4 lines
Thanks for doing this.
I believe I had 'fix pdflush after suspend' queued in Greg's tree. Is it still queued or should I resend?
Is this "pdflush: handle resume wakeups"?
Yes. Do you have it somewhere or should I dig it up?
I've applied it.
Quoted 1 line
Pavel
cu
Adrian
Signature · 4 lines
"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