Status of v4.9

Status of v4.9

[Mason]

Hello Greg,

A few months ago, you stated that you were considering making v4.9
the latest LTS version.

http://kroah.com/log/blog/2016/09/06/4-dot-9-equals-equals-next-lts-kernel/

Neither https://www.kernel.org/ nor https://www.kernel.org/category/releases.html
list 4.9 as an LTS version yet.

Could you clear (some of) my confusion?

Regards.

Re: Status of v4.9

[Greg Kroah-Hartman]

On Mon, Jan 16, 2017 at 09:39:39AM +0100, Mason wrote:
> Hello Greg,
> 
> A few months ago, you stated that you were considering making v4.9
> the latest LTS version.
> 
> http://kroah.com/log/blog/2016/09/06/4-dot-9-equals-equals-next-lts-kernel/
> 
> Neither https://www.kernel.org/ nor https://www.kernel.org/category/releases.html
> list 4.9 as an LTS version yet.
> 
> Could you clear (some of) my confusion?

It's a bit hard for a kernel to be "LTS" when it hasn't even had the
chance to move out of the "normal" stable release process, right? :)

Is there anything in your testing of 4.9 that you feel needs to be
resolved before you would feel comfortable using it as a LTS kernel?
How has it worked out for your platform and workload?  Any warning flags
that you feel would keep it from being a good LTS kernel?

thanks,

greg k-h

Re: Status of v4.9

[Mason]

On 16/01/2017 11:34, Greg Kroah-Hartman wrote:

> On Mon, Jan 16, 2017 at 09:39:39AM +0100, Mason wrote:
>
>> Hello Greg,
>>
>> A few months ago, you stated that you were considering making v4.9
>> the latest LTS version.
>>
>> http://kroah.com/log/blog/2016/09/06/4-dot-9-equals-equals-next-lts-kernel/
>>
>> Neither https://www.kernel.org/ nor https://www.kernel.org/category/releases.html
>> list 4.9 as an LTS version yet.
>>
>> Could you clear (some of) my confusion?
> 
> It's a bit hard for a kernel to be "LTS" when it hasn't even had the
> chance to move out of the "normal" stable release process, right? :)

I think it might be worthwhile mentioning somewhere that 4.9 is LTS.
(My manager thought 4.4 was the latest.)

> Is there anything in your testing of 4.9 that you feel needs to be
> resolved before you would feel comfortable using it as a LTS kernel?
> How has it worked out for your platform and workload?  Any warning flags
> that you feel would keep it from being a good LTS kernel?

I see that you back-ported the fix for the HOTPLUG crash regression
that affected my platform, so thanks for that. Other than that, we're
getting ready to run a full regression test, so I guess I'll know more
in a few weeks :-)

But my options are limited anyway. I need a recent LTS, and 4.4 is
getting pretty old... so 4.9 it is. (We're upgrading from 3.4)

Regards.

Re: Status of v4.9

[Greg Kroah-Hartman]

On Mon, Jan 16, 2017 at 01:35:38PM +0100, Mason wrote:
> On 16/01/2017 11:34, Greg Kroah-Hartman wrote:
> 
> > On Mon, Jan 16, 2017 at 09:39:39AM +0100, Mason wrote:
> >
> >> Hello Greg,
> >>
> >> A few months ago, you stated that you were considering making v4.9
> >> the latest LTS version.
> >>
> >> http://kroah.com/log/blog/2016/09/06/4-dot-9-equals-equals-next-lts-kernel/
> >>
> >> Neither https://www.kernel.org/ nor https://www.kernel.org/category/releases.html
> >> list 4.9 as an LTS version yet.
> >>
> >> Could you clear (some of) my confusion?
> > 
> > It's a bit hard for a kernel to be "LTS" when it hasn't even had the
> > chance to move out of the "normal" stable release process, right? :)
> 
> I think it might be worthwhile mentioning somewhere that 4.9 is LTS.
> (My manager thought 4.4 was the latest.)

As you can see by my questions, I haven't come to that final conclusion
yet :)

> > Is there anything in your testing of 4.9 that you feel needs to be
> > resolved before you would feel comfortable using it as a LTS kernel?
> > How has it worked out for your platform and workload?  Any warning flags
> > that you feel would keep it from being a good LTS kernel?
> 
> I see that you back-ported the fix for the HOTPLUG crash regression
> that affected my platform, so thanks for that. Other than that, we're
> getting ready to run a full regression test, so I guess I'll know more
> in a few weeks :-)

Please let me know if you find any issues.

thanks,

greg k-h

Re: Status of v4.9

[Mason]

On 16/01/2017 14:02, Greg Kroah-Hartman wrote:

> On Mon, Jan 16, 2017 at 01:35:38PM +0100, Mason wrote:
>
>> On 16/01/2017 11:34, Greg Kroah-Hartman wrote:
>>
>>> On Mon, Jan 16, 2017 at 09:39:39AM +0100, Mason wrote:
>>>
>>>> Hello Greg,
>>>>
>>>> A few months ago, you stated that you were considering making v4.9
>>>> the latest LTS version.
>>>>
>>>> http://kroah.com/log/blog/2016/09/06/4-dot-9-equals-equals-next-lts-kernel/
>>>>
>>>> Neither https://www.kernel.org/ nor https://www.kernel.org/category/releases.html
>>>> list 4.9 as an LTS version yet.
>>>>
>>>> Could you clear (some of) my confusion?
>>>
>>> It's a bit hard for a kernel to be "LTS" when it hasn't even had the
>>> chance to move out of the "normal" stable release process, right? :)
>>
>> I think it might be worthwhile mentioning somewhere that 4.9 is LTS.
>> (My manager thought 4.4 was the latest.)
> 
> As you can see by my questions, I haven't come to that final conclusion
> yet :)

Sorry, I must be missing something obvious :-)

What are the advantages of not listing 4.9 in the LTS list(*)
until 4.11 is released?

(*) https://www.kernel.org/category/releases.html

Regards.

Re: Status of v4.9

[Greg Kroah-Hartman]

On Mon, Jan 16, 2017 at 03:11:58PM +0100, Mason wrote:
> On 16/01/2017 14:02, Greg Kroah-Hartman wrote:
> 
> > On Mon, Jan 16, 2017 at 01:35:38PM +0100, Mason wrote:
> >
> >> On 16/01/2017 11:34, Greg Kroah-Hartman wrote:
> >>
> >>> On Mon, Jan 16, 2017 at 09:39:39AM +0100, Mason wrote:
> >>>
> >>>> Hello Greg,
> >>>>
> >>>> A few months ago, you stated that you were considering making v4.9
> >>>> the latest LTS version.
> >>>>
> >>>> http://kroah.com/log/blog/2016/09/06/4-dot-9-equals-equals-next-lts-kernel/
> >>>>
> >>>> Neither https://www.kernel.org/ nor https://www.kernel.org/category/releases.html
> >>>> list 4.9 as an LTS version yet.
> >>>>
> >>>> Could you clear (some of) my confusion?
> >>>
> >>> It's a bit hard for a kernel to be "LTS" when it hasn't even had the
> >>> chance to move out of the "normal" stable release process, right? :)
> >>
> >> I think it might be worthwhile mentioning somewhere that 4.9 is LTS.
> >> (My manager thought 4.4 was the latest.)
> > 
> > As you can see by my questions, I haven't come to that final conclusion
> > yet :)
> 
> Sorry, I must be missing something obvious :-)
> 
> What are the advantages of not listing 4.9 in the LTS list(*)
> until 4.11 is released?
> 
> (*) https://www.kernel.org/category/releases.html

It gives me the chance to not make 4.9 a LTS kernel if something shows
up that is a really big problem.  People have only been really testing
4.9 for a few weeks (really only 2 given the holidays.)  Let's see what
shakes out first, ok?

Has anyone run any benchmarks yet even?  Have you?

What is the advantage of me listing it now?  Is that going to cause
anyone to do any extra work today for helping to test this out that it
wouldn't have?

If you need to point a manager at something, point them at my blog post,
that is why I wrote it.  And test the heck out of it starting now!

thanks,

greg k-h

Re: Status of v4.9

[Willy Tarreau]

On Mon, Jan 16, 2017 at 03:29:24PM +0100, Greg Kroah-Hartman wrote:
> Has anyone run any benchmarks yet even?  Have you?

At least FWIW I've upgraded my laptop to it two weeks ago in the hope
to spot any issue, and my build farm is not running on it as well. No
perf nor stability issues to report yet, but it's early ;-)

Willy

Re: Status of v4.9

[Greg Kroah-Hartman]

On Mon, Jan 16, 2017 at 03:39:44PM +0100, Willy Tarreau wrote:
> On Mon, Jan 16, 2017 at 03:29:24PM +0100, Greg Kroah-Hartman wrote:
> > Has anyone run any benchmarks yet even?  Have you?
> 
> At least FWIW I've upgraded my laptop to it two weeks ago in the hope
> to spot any issue, and my build farm is not running on it as well. No
> perf nor stability issues to report yet, but it's early ;-)

Ah, that's good to hear, thanks for letting me know.

greg k-h

Re: Status of v4.9

[Willy Tarreau]

On Mon, Jan 16, 2017 at 04:55:24PM +0100, Greg Kroah-Hartman wrote:
> On Mon, Jan 16, 2017 at 03:39:44PM +0100, Willy Tarreau wrote:
> > On Mon, Jan 16, 2017 at 03:29:24PM +0100, Greg Kroah-Hartman wrote:
> > > Has anyone run any benchmarks yet even?  Have you?
> > 
> > At least FWIW I've upgraded my laptop to it two weeks ago in the hope
> > to spot any issue, and my build farm is not running on it as well. No
> > perf nor stability issues to report yet, but it's early ;-)
> 
> Ah, that's good to hear, thanks for letting me know.

Hmmm that was before I tried to use WiFi :-)

------------
iwlwifi 0000:03:00.0: L1 Enabled - LTR Disabled
iwlwifi 0000:03:00.0: L1 Enabled - LTR Disabled
iwlwifi 0000:03:00.0: Radio type=0x1-0x2-0x0
iwlwifi 0000:03:00.0: L1 Enabled - LTR Disabled
iwlwifi 0000:03:00.0: L1 Enabled - LTR Disabled
iwlwifi 0000:03:00.0: Radio type=0x1-0x2-0x0
wlan0: authenticate with c0:56:27:eb:e7:b2
wlan0: send auth to c0:56:27:eb:e7:b2 (try 1/3)
wlan0: authenticated
wlan0: waiting for beacon from c0:56:27:eb:e7:b2
wlan0: associate with c0:56:27:eb:e7:b2 (try 1/3)
wlan0: RX AssocResp from c0:56:27:eb:e7:b2 (capab=0x431 status=0 aid=4)
wlan0: associated
wlan0: deauthenticated from c0:56:27:eb:e7:b2 (Reason: 6=CLASS2_FRAME_FROM_NONAUTH_STA)
wlan0: authenticate with c0:56:27:eb:e7:b2
wlan0: send auth to c0:56:27:eb:e7:b2 (try 1/3)
wlan0: authenticated
wlan0: associate with c0:56:27:eb:e7:b2 (try 1/3)
wlan0: RX AssocResp from c0:56:27:eb:e7:b2 (capab=0x431 status=0 aid=4)
wlan0: associated
wlan0: deauthenticated from c0:56:27:eb:e7:b2 (Reason: 6=CLASS2_FRAME_FROM_NONAUTH_STA)
wlan0: authenticate with c0:56:27:eb:e7:b2
wlan0: send auth to c0:56:27:eb:e7:b2 (try 1/3)
wlan0: authenticated
wlan0: associate with c0:56:27:eb:e7:b2 (try 1/3)
wlan0: RX AssocResp from c0:56:27:eb:e7:b2 (capab=0x431 status=0 aid=4)
wlan0: associated
wlan0: deauthenticated from c0:56:27:eb:e7:b2 (Reason: 6=CLASS2_FRAME_FROM_NONAUTH_STA)
wlan0: authenticate with c0:56:27:eb:e7:b2
wlan0: send auth to c0:56:27:eb:e7:b2 (try 1/3)
wlan0: authenticated
wlan0: associate with c0:56:27:eb:e7:b2 (try 1/3)
wlan0: RX AssocResp from c0:56:27:eb:e7:b2 (capab=0x431 status=0 aid=4)
wlan0: associated
wlan0: deauthenticated from c0:56:27:eb:e7:b2 (Reason: 9=STA_REQ_ASSOC_WITHOUT_AUTH)
------------

I'll try to bisect, it was rock-solid in 4.4.38 (strange for WiFi I know)
and doesn't work at all in 4.9.1-4.9.4.

Willy

Re: Status of v4.9

[Willy Tarreau]

On Mon, Jan 16, 2017 at 06:42:30PM +0100, Willy Tarreau wrote:
> I'll try to bisect, it was rock-solid in 4.4.38 (strange for WiFi I know)
> and doesn't work at all in 4.9.1-4.9.4.

OK please disregard this, I should have shut up, it suddenly stopped
working in 4.4 as well, so it's just working as one could expect a wire
made of thin air to work. There's no regression here, I'll have to try
to debug this later if I need it.

Regards,
Willy

Re: Status of v4.9

[Greg Kroah-Hartman]

On Mon, Jan 16, 2017 at 07:34:03PM +0100, Willy Tarreau wrote:
> On Mon, Jan 16, 2017 at 06:42:30PM +0100, Willy Tarreau wrote:
> > I'll try to bisect, it was rock-solid in 4.4.38 (strange for WiFi I know)
> > and doesn't work at all in 4.9.1-4.9.4.
> 
> OK please disregard this, I should have shut up, it suddenly stopped
> working in 4.4 as well, so it's just working as one could expect a wire
> made of thin air to work. There's no regression here, I'll have to try
> to debug this later if I need it.

Is this a hardware issue?  What wireless driver/chip is it?

thanks,

greg k-h

Re: Status of v4.9

[Willy Tarreau]

On Tue, Jan 17, 2017 at 08:15:31AM +0100, Greg Kroah-Hartman wrote:
> On Mon, Jan 16, 2017 at 07:34:03PM +0100, Willy Tarreau wrote:
> > On Mon, Jan 16, 2017 at 06:42:30PM +0100, Willy Tarreau wrote:
> > > I'll try to bisect, it was rock-solid in 4.4.38 (strange for WiFi I know)
> > > and doesn't work at all in 4.9.1-4.9.4.
> > 
> > OK please disregard this, I should have shut up, it suddenly stopped
> > working in 4.4 as well, so it's just working as one could expect a wire
> > made of thin air to work. There's no regression here, I'll have to try
> > to debug this later if I need it.
> 
> Is this a hardware issue?  What wireless driver/chip is it?

Probably just an AP issue only, that's why I really don't care. I've just
verified that it works fine at home. I use it once every few months, if
it doesn't work anymore with my previous kernel I consider that it's not
a kernel regression :-)  It just happens that I discovered the issue for
the first time in 4.9 so I believed it was a regression. If it had been
broken since 4.4, many daily users would have reported it ;-)

The driver is iwlwifi BTW.

Cheers,
Willy

Re: Status of v4.9

[Christoph Biedl]

Greg Kroah-Hartman wrote...

> If you need to point a manager at something, point them at my blog post,
> that is why I wrote it.  And test the heck out of it starting now!

Don't worry, 4.9 is already running on more than the half of my boxes
now. So far, transition (from 4.4 mostly) went very smoothly, even on
embedded hardware where I'm always a bit more concerned about
regressions. While some computers gave the feeling of being a bit
slower, there is no data that substantiates it.

The only thing that requires checking is the X server in a KVM guest
that fails to start. But since I haven got around to take a closer look
yet, I preferred not the bother anybody for the time being.

    Christoph

Re: Status of v4.9

[Mason]

On 16/01/2017 14:02, Greg Kroah-Hartman wrote:

> On Mon, Jan 16, 2017 at 01:35:38PM +0100, Mason wrote:
>
>> On 16/01/2017 11:34, Greg Kroah-Hartman wrote:
>>
>>> On Mon, Jan 16, 2017 at 09:39:39AM +0100, Mason wrote:
>>>
>>>> Hello Greg,
>>>>
>>>> A few months ago, you stated that you were considering making v4.9
>>>> the latest LTS version.
>>>>
>>>> http://kroah.com/log/blog/2016/09/06/4-dot-9-equals-equals-next-lts-kernel/
>>>>
>>>> Neither https://www.kernel.org/ nor https://www.kernel.org/category/releases.html
>>>> list 4.9 as an LTS version yet.
>>>>
>>>> Could you clear (some of) my confusion?
>>>
>>> It's a bit hard for a kernel to be "LTS" when it hasn't even had the
>>> chance to move out of the "normal" stable release process, right? :)
>>
>> I think it might be worthwhile mentioning somewhere that 4.9 is LTS.
>> (My manager thought 4.4 was the latest.)
> 
> As you can see by my questions, I haven't come to that final conclusion
> yet :)

I had somehow managed to misinterpret this sentence.

I now understand the process better.

Regards.

Re: Status of v4.9

[Christoph Biedl]

[ That was in January ]

Christoph Biedl wrote...

> Greg Kroah-Hartman wrote...
>
> > If you need to point a manager at something, point them at my blog post,
> > that is why I wrote it.  And test the heck out of it starting now!
>
> The only thing that requires checking is the X server in a KVM guest
> that fails to start. But since I haven got around to take a closer look
> yet, I preferred not the bother anybody for the time being.

After a round tuit finally arrived: This is not a regression in the kernel,
just a stricter behaviour some userland cannot cope with (yet).

It was introduced in

    commit 21266be9ed542f13436bd9c75316d43e1e84f6ae
    Author: Dan Williams <dan.j.williams@intel.com>
    Date:   Thu Nov 19 18:19:29 2015 -0800

        arch: consolidate CONFIG_STRICT_DEVM in lib/Kconfig.debug

and manifests only if CONFIG_IO_STRICT_DEVMEM is set. Then some X
servers will report "Internal error: could not map framebuffer range
(22, Invalid argument)" or even happily segfault. It's beyond my
knowledge whether they can handle that situation. In my opinion, they
should, if ever possible.

    Christoph