[cip-dev] Next LTS will be 4.19

[cip-dev] Next LTS will be 4.19

[Jan Kiszka]

Hi all,

just a heads-up in case you didn't see read this yet, the next LTS is 
coming earlier: https://www.kernel.org/category/releases.html

I suspect that this increases the chances of Debian choosing the same 
version for buster because it gives more time to validate/stabilize it. 
And that should also help us picking it as next SLTS.

Jan

PS: 4.9 is now receiving 6 years maintenance from Greg as well, again 
due to Android needs.

-- 
Siemens AG, Corporate Technology, CT RDA IOT SES-DE
Corporate Competence Center Embedded Linux

[cip-dev] Next LTS will be 4.19

[Zoran S]

Hello Jan,

I see that there is correction to CIP initial plan/strategy. I should
say... Sounds much more realistic. :-)

Well... This is the Right Direction. There is no way to support old LTS
kernels for 20+ years (who, the heck, will support any RT Linux kernel for
the once released equipment for 20+ years, only critical bugs), Rather for
2 to 4. Then, the final freeze should come. IMHO.

As Fedora does (as example), correct (even for 14 months per release, at
most)? ;-)

Best Regards,
Zoran
_______

On Mon, Aug 27, 2018 at 8:32 AM Jan Kiszka <jan.kiszka@siemens.com> wrote:

> Hi all,
>
> just a heads-up in case you didn't see read this yet, the next LTS is
> coming earlier: https://www.kernel.org/category/releases.html
>
> I suspect that this increases the chances of Debian choosing the same
> version for buster because it gives more time to validate/stabilize it.
> And that should also help us picking it as next SLTS.
>
> Jan
>
> PS: 4.9 is now receiving 6 years maintenance from Greg as well, again
> due to Android needs.
>
> --
> Siemens AG, Corporate Technology, CT RDA IOT SES-DE
> Corporate Competence Center Embedded Linux
> _______________________________________________
> cip-dev mailing list
> cip-dev at lists.cip-project.org
> https://lists.cip-project.org/mailman/listinfo/cip-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.cip-project.org/pipermail/cip-dev/attachments/20180828/3edc9f0e/attachment.html>

[cip-dev] Next LTS will be 4.19

[Jan Kiszka]

Hi Zoran,

On 2018-08-28 19:14, Zoran S wrote:
> Hello Jan,
> 
> I see that there is correction to CIP initial plan/strategy. I should 
> say... Sounds much more realistic. :-)
> > Well... This is the Right Direction. There is no way to support old LTS
> kernels for 20+ years (who, the heck, will support any RT Linux kernel 
> for the once released equipment for 20+ years, only critical bugs), 
> Rather for 2 to 4. Then, the final freeze should come. IMHO.
> 
> As Fedora does (as example), correct (even for 14 months per release, at 
> most)? ;-)

There is no change to the CIP strategy. We are committed to 10 years 
maintenance for each kernel we pick up. The release of the next CIP 
kernel will not end the maintenance of the current 4.4 one. We will 
"only" stop taking feature backports at that point, because new 
developments should start on the new version.

Jan

> 
> Best Regards,
> Zoran
> _______
> 
> On Mon, Aug 27, 2018 at 8:32 AM Jan Kiszka <jan.kiszka@siemens.com 
> <mailto:jan.kiszka@siemens.com>> wrote:
> 
>     Hi all,
> 
>     just a heads-up in case you didn't see read this yet, the next LTS is
>     coming earlier: https://www.kernel.org/category/releases.html
> 
>     I suspect that this increases the chances of Debian choosing the same
>     version for buster because it gives more time to validate/stabilize it.
>     And that should also help us picking it as next SLTS.
> 
>     Jan
> 
>     PS: 4.9 is now receiving 6 years maintenance from Greg as well, again
>     due to Android needs.
> 
>     -- 
>     Siemens AG, Corporate Technology, CT RDA IOT SES-DE
>     Corporate Competence Center Embedded Linux
>     _______________________________________________
>     cip-dev mailing list
>     cip-dev at lists.cip-project.org <mailto:cip-dev@lists.cip-project.org>
>     https://lists.cip-project.org/mailman/listinfo/cip-dev
> 

-- 
Siemens AG, Corporate Technology, CT RDA IOT SES-DE
Corporate Competence Center Embedded Linux

[cip-dev] Next LTS will be 4.19

[Zoran S]

As I recall Daniel (Wagner), in early stages said that CIP has
committed to 15+ (20+, I should say, let me be very defensive on what
I heard about strategy) years of support.

But, anyway (discarded above statement as unreliable) even 10 years is
way too long. Half of it/committed time is more appropriate (if ?).
IMHO. You do (you/Siemens AG pay(s), CIP pays) as you intend. :-)

I am, after all, Nobody. But anyway, Nobody's two cent opinion.

Thank you,
Zoran
_______

On Wed, Aug 29, 2018 at 8:26 AM Jan Kiszka <jan.kiszka@siemens.com> wrote:
>
> Hi Zoran,
>
> On 2018-08-28 19:14, Zoran S wrote:
> > Hello Jan,
> >
> > I see that there is correction to CIP initial plan/strategy. I should
> > say... Sounds much more realistic. :-)
> > > Well... This is the Right Direction. There is no way to support old LTS
> > kernels for 20+ years (who, the heck, will support any RT Linux kernel
> > for the once released equipment for 20+ years, only critical bugs),
> > Rather for 2 to 4. Then, the final freeze should come. IMHO.
> >
> > As Fedora does (as example), correct (even for 14 months per release, at
> > most)? ;-)
>
> There is no change to the CIP strategy. We are committed to 10 years
> maintenance for each kernel we pick up. The release of the next CIP
> kernel will not end the maintenance of the current 4.4 one. We will
> "only" stop taking feature backports at that point, because new
> developments should start on the new version.
>
> Jan
>
> >
> > Best Regards,
> > Zoran
> > _______
> >
> > On Mon, Aug 27, 2018 at 8:32 AM Jan Kiszka <jan.kiszka@siemens.com
> > <mailto:jan.kiszka@siemens.com>> wrote:
> >
> >     Hi all,
> >
> >     just a heads-up in case you didn't see read this yet, the next LTS is
> >     coming earlier: https://www.kernel.org/category/releases.html
> >
> >     I suspect that this increases the chances of Debian choosing the same
> >     version for buster because it gives more time to validate/stabilize it.
> >     And that should also help us picking it as next SLTS.
> >
> >     Jan
> >
> >     PS: 4.9 is now receiving 6 years maintenance from Greg as well, again
> >     due to Android needs.
> >
> >     --
> >     Siemens AG, Corporate Technology, CT RDA IOT SES-DE
> >     Corporate Competence Center Embedded Linux
> >     _______________________________________________
> >     cip-dev mailing list
> >     cip-dev at lists.cip-project.org <mailto:cip-dev@lists.cip-project.org>
> >     https://lists.cip-project.org/mailman/listinfo/cip-dev
> >
>
> --
> Siemens AG, Corporate Technology, CT RDA IOT SES-DE
> Corporate Competence Center Embedded Linux

[cip-dev] Next LTS will be 4.19

[Chris Paterson]

Hello Zoran,

> From: cip-dev-bounces at lists.cip-project.org <cip-dev-bounces@lists.cip-
> project.org> On Behalf Of Zoran S
> Sent: 29 August 2018 09:27
> 
> As I recall Daniel (Wagner), in early stages said that CIP has committed to 15+
> (20+, I should say, let me be very defensive on what I heard about strategy)
> years of support.
> 
> But, anyway (discarded above statement as unreliable) even 10 years is way too
> long. Half of it/committed time is more appropriate (if ?).
> IMHO. You do (you/Siemens AG pay(s), CIP pays) as you intend. :-)

One of the main points of CIP is to provide a maintained Kernel for 10+ years to be used in industries with very long product lifecycles.

If we only planned to support Kernels for 5 years, there would be little need for CIP as Greg is already supporting select LTS Kernels for 6 years! 

Kind regards, Chris


> 
> I am, after all, Nobody. But anyway, Nobody's two cent opinion.
> 
> Thank you,
> Zoran
> _______
> 
> On Wed, Aug 29, 2018 at 8:26 AM Jan Kiszka <jan.kiszka@siemens.com> wrote:
> >
> > Hi Zoran,
> >
> > On 2018-08-28 19:14, Zoran S wrote:
> > > Hello Jan,
> > >
> > > I see that there is correction to CIP initial plan/strategy. I
> > > should say... Sounds much more realistic. :-)
> > > > Well... This is the Right Direction. There is no way to support
> > > > old LTS
> > > kernels for 20+ years (who, the heck, will support any RT Linux
> > > kernel for the once released equipment for 20+ years, only critical
> > > bugs), Rather for 2 to 4. Then, the final freeze should come. IMHO.
> > >
> > > As Fedora does (as example), correct (even for 14 months per
> > > release, at most)? ;-)
> >
> > There is no change to the CIP strategy. We are committed to 10 years
> > maintenance for each kernel we pick up. The release of the next CIP
> > kernel will not end the maintenance of the current 4.4 one. We will
> > "only" stop taking feature backports at that point, because new
> > developments should start on the new version.
> >
> > Jan
> >
> > >
> > > Best Regards,
> > > Zoran
> > > _______
> > >
> > > On Mon, Aug 27, 2018 at 8:32 AM Jan Kiszka <jan.kiszka@siemens.com
> > > <mailto:jan.kiszka@siemens.com>> wrote:
> > >
> > >     Hi all,
> > >
> > >     just a heads-up in case you didn't see read this yet, the next LTS is
> > >     coming earlier: https://www.kernel.org/category/releases.html
> > >
> > >     I suspect that this increases the chances of Debian choosing the same
> > >     version for buster because it gives more time to validate/stabilize it.
> > >     And that should also help us picking it as next SLTS.
> > >
> > >     Jan
> > >
> > >     PS: 4.9 is now receiving 6 years maintenance from Greg as well, again
> > >     due to Android needs.
> > >
> > >     --
> > >     Siemens AG, Corporate Technology, CT RDA IOT SES-DE
> > >     Corporate Competence Center Embedded Linux
> > >     _______________________________________________
> > >     cip-dev mailing list
> > >     cip-dev at lists.cip-project.org <mailto:cip-dev@lists.cip-project.org>
> > >     https://lists.cip-project.org/mailman/listinfo/cip-dev
> > >
> >
> > --
> > Siemens AG, Corporate Technology, CT RDA IOT SES-DE Corporate
> > Competence Center Embedded Linux
> _______________________________________________
> cip-dev mailing list
> cip-dev at lists.cip-project.org
> https://lists.cip-project.org/mailman/listinfo/cip-dev

[cip-dev] Next LTS will be 4.19

[Zoran S]

> If we only planned to support Kernels for 5 years, there would be little need for CIP as Greg is already supporting select LTS Kernels for 6 years!

5 years are silicon/Linux technology (endless time) ages, almost
unlimited... But, as INTEL failed repeatedly on 10nm UV silicon
technology (and Brian Krzanich was expelled as being INTEL CEO "INTEL
Inside" as indecent CoC performance (CoC = Code of Conduct))...

I asked Brian F2F some very important questions in Y2015, now 3+years
passed.... Brian did NOT have the answers. Because of that, maybe I
allow in my mind that you are very correct. If?

It is, after all, in this stage of the game, up to the top noch
silicon tech companies:
GlobalFoundries;
TSMC;
Samsung;
INTEL.

And this slowdown... Maybe it has some (serious) influence on CIP
strategy? No clear vision from my side (could not tell). :-(

Time Will Tell,
Zoran
_______
On Wed, Aug 29, 2018 at 10:31 AM Chris Paterson
<Chris.Paterson2@renesas.com> wrote:
>
> Hello Zoran,
>
> > From: cip-dev-bounces at lists.cip-project.org <cip-dev-bounces@lists.cip-
> > project.org> On Behalf Of Zoran S
> > Sent: 29 August 2018 09:27
> >
> > As I recall Daniel (Wagner), in early stages said that CIP has committed to 15+
> > (20+, I should say, let me be very defensive on what I heard about strategy)
> > years of support.
> >
> > But, anyway (discarded above statement as unreliable) even 10 years is way too
> > long. Half of it/committed time is more appropriate (if ?).
> > IMHO. You do (you/Siemens AG pay(s), CIP pays) as you intend. :-)
>
> One of the main points of CIP is to provide a maintained Kernel for 10+ years to be used in industries with very long product lifecycles.
>
> If we only planned to support Kernels for 5 years, there would be little need for CIP as Greg is already supporting select LTS Kernels for 6 years!
>
> Kind regards, Chris
>
>
> >
> > I am, after all, Nobody. But anyway, Nobody's two cent opinion.
> >
> > Thank you,
> > Zoran
> > _______
> >
> > On Wed, Aug 29, 2018 at 8:26 AM Jan Kiszka <jan.kiszka@siemens.com> wrote:
> > >
> > > Hi Zoran,
> > >
> > > On 2018-08-28 19:14, Zoran S wrote:
> > > > Hello Jan,
> > > >
> > > > I see that there is correction to CIP initial plan/strategy. I
> > > > should say... Sounds much more realistic. :-)
> > > > > Well... This is the Right Direction. There is no way to support
> > > > > old LTS
> > > > kernels for 20+ years (who, the heck, will support any RT Linux
> > > > kernel for the once released equipment for 20+ years, only critical
> > > > bugs), Rather for 2 to 4. Then, the final freeze should come. IMHO.
> > > >
> > > > As Fedora does (as example), correct (even for 14 months per
> > > > release, at most)? ;-)
> > >
> > > There is no change to the CIP strategy. We are committed to 10 years
> > > maintenance for each kernel we pick up. The release of the next CIP
> > > kernel will not end the maintenance of the current 4.4 one. We will
> > > "only" stop taking feature backports at that point, because new
> > > developments should start on the new version.
> > >
> > > Jan
> > >
> > > >
> > > > Best Regards,
> > > > Zoran
> > > > _______
> > > >
> > > > On Mon, Aug 27, 2018 at 8:32 AM Jan Kiszka <jan.kiszka@siemens.com
> > > > <mailto:jan.kiszka@siemens.com>> wrote:
> > > >
> > > >     Hi all,
> > > >
> > > >     just a heads-up in case you didn't see read this yet, the next LTS is
> > > >     coming earlier: https://www.kernel.org/category/releases.html
> > > >
> > > >     I suspect that this increases the chances of Debian choosing the same
> > > >     version for buster because it gives more time to validate/stabilize it.
> > > >     And that should also help us picking it as next SLTS.
> > > >
> > > >     Jan
> > > >
> > > >     PS: 4.9 is now receiving 6 years maintenance from Greg as well, again
> > > >     due to Android needs.
> > > >
> > > >     --
> > > >     Siemens AG, Corporate Technology, CT RDA IOT SES-DE
> > > >     Corporate Competence Center Embedded Linux
> > > >     _______________________________________________
> > > >     cip-dev mailing list
> > > >     cip-dev at lists.cip-project.org <mailto:cip-dev@lists.cip-project.org>
> > > >     https://lists.cip-project.org/mailman/listinfo/cip-dev
> > > >
> > >
> > > --
> > > Siemens AG, Corporate Technology, CT RDA IOT SES-DE Corporate
> > > Competence Center Embedded Linux
> > _______________________________________________
> > cip-dev mailing list
> > cip-dev at lists.cip-project.org
> > https://lists.cip-project.org/mailman/listinfo/cip-dev

[cip-dev] Next LTS will be 4.19

[Jan Kiszka]

On 2018-08-29 10:26, Zoran S wrote:
> As I recall Daniel (Wagner), in early stages said that CIP has
> committed to 15+ (20+, I should say, let me be very defensive on what
> I heard about strategy) years of support.

That's not originating from Daniel, it was rather some possibly unclear 
statements on early CIP publications.

At this point, we as CIP commit to 10 years, but we cannot promise 
anything beyond. That does not mean we categorically exclude it, but we 
first of all need to collect enough experience both with the long-term 
maintenance in the embedded context and with the value a support time 
frame of >10 years will have - in some 7 years from now.

I assume everyone would be happy if the whole software stack could have 
shorter major update cycles than a decade. Things like improving test 
coverage and automation is an important building block not only for 
long-term support but also for safer earlier updates. But there will be 
more needed, and that not only in the OSS part of the stack. Time will 
tell how industry will evolve and what will be the optimal balance 
between stability and maintainability (and "freshness") in the future.

HTH,
Jan

-- 
Siemens AG, Corporate Technology, CT RDA IOT SES-DE
Corporate Competence Center Embedded Linux

[cip-dev] Next LTS will be 4.19

[Zoran S]

> At this point, we as CIP commit to 10 years, but we cannot promise
> anything beyond. That does not mean we categorically exclude it, but we
> first of all need to collect enough experience both with the long-term
> maintenance in the embedded context and with the value a support time
> frame of >10 years will have - in some 7 years from now.

Fair enough. The only addendum to your last @.... For 7+ years old
support, you/CIP have to have rock solid history and documentation
written down. ;-)

Please, please... Do NOT forget this rule.

Thank you,
Zoran
_______
On Wed, Aug 29, 2018 at 12:38 PM Jan Kiszka <jan.kiszka@siemens.com> wrote:
>
> On 2018-08-29 10:26, Zoran S wrote:
> > As I recall Daniel (Wagner), in early stages said that CIP has
> > committed to 15+ (20+, I should say, let me be very defensive on what
> > I heard about strategy) years of support.
>
> That's not originating from Daniel, it was rather some possibly unclear
> statements on early CIP publications.
>
> At this point, we as CIP commit to 10 years, but we cannot promise
> anything beyond. That does not mean we categorically exclude it, but we
> first of all need to collect enough experience both with the long-term
> maintenance in the embedded context and with the value a support time
> frame of >10 years will have - in some 7 years from now.
>
> I assume everyone would be happy if the whole software stack could have
> shorter major update cycles than a decade. Things like improving test
> coverage and automation is an important building block not only for
> long-term support but also for safer earlier updates. But there will be
> more needed, and that not only in the OSS part of the stack. Time will
> tell how industry will evolve and what will be the optimal balance
> between stability and maintainability (and "freshness") in the future.
>
> HTH,
> Jan
>
> --
> Siemens AG, Corporate Technology, CT RDA IOT SES-DE
> Corporate Competence Center Embedded Linux