Dropping i386 non-PAE as a supported kernel flavour in Precise Pangolin

classic Classic list List threaded Threaded
38 messages Options
12
Reply | Threaded
Open this post in threaded view
|

Dropping i386 non-PAE as a supported kernel flavour in Precise Pangolin

Tim Gardner-2
Per discussion at UDS the kernel team is proposing to drop the non-PAE
i386 flavour. The upgrade path for non-PAE users will be the PAE kernel.
Those CPUs that do not have i686 and PAE support will be orphaned. To
the best of my knowledge, these include Intel CPUs prior to Pentium II,
400Mhz Pentium M, VIA C3, and Geode LX. As far as I know, there are no
laptop or desktop class CPUs being produced that do not meet these
minimum requirements.

Before I do something that is difficult to revert, I would like to hear
from the development community why we should continue to maintain a
kernel flavour that is (in my opinion) getting increasingly low
utilization. It is my feeling that an extremely high percentage of users
of the non-PAE kernel have a CPU that is PAE capable.

If there is sufficient community demand (and support), I would be
willing to sponsor the first non-PAE kernel upload to Universe.

https://wiki.ubuntu.com/KernelTeam/Specs/PreciseKernelConfigReview

We'll be conducting a similar survey for powerpc.

rtg

P.S. For those of you that are totally confused by this email, PAE
(Physical Address Extension) was an addition to 32 bit x86 CPUs that
allowed them to address more then 4GB physical memory.

http://en.wikipedia.org/wiki/Physical_Address_Extension
--
Tim Gardner [hidden email]

--
ubuntu-devel mailing list
[hidden email]
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel
Reply | Threaded
Open this post in threaded view
|

Re: Dropping i386 non-PAE as a supported kernel flavour in Precise Pangolin

Amir Eldor-2
Hello Tim,

The only reason I see for keeping support of weird old CPUs is for people who have an old computer and they would like to linux-ize it. I'm not sure if the "old" of today are computers running with powerpc or non-PAE CPUs, so my original claim might be void.

I think these kind of computers will probably won't be installed with Ubuntu at all, but X/Lubuntu might be a choice for them. As I understand the kernel of Ubuntu is shared between the 'sister-releases', am I right?

Thanks,
Amir

On Wed, Nov 9, 2011 at 11:43 PM, Tim Gardner <[hidden email]> wrote:
Per discussion at UDS the kernel team is proposing to drop the non-PAE i386 flavour. The upgrade path for non-PAE users will be the PAE kernel. Those CPUs that do not have i686 and PAE support will be orphaned. To the best of my knowledge, these include Intel CPUs prior to Pentium II, 400Mhz Pentium M, VIA C3, and Geode LX. As far as I know, there are no laptop or desktop class CPUs being produced that do not meet these minimum requirements.

Before I do something that is difficult to revert, I would like to hear from the development community why we should continue to maintain a kernel flavour that is (in my opinion) getting increasingly low utilization. It is my feeling that an extremely high percentage of users of the non-PAE kernel have a CPU that is PAE capable.

If there is sufficient community demand (and support), I would be willing to sponsor the first non-PAE kernel upload to Universe.

https://wiki.ubuntu.com/KernelTeam/Specs/PreciseKernelConfigReview

We'll be conducting a similar survey for powerpc.

rtg

P.S. For those of you that are totally confused by this email, PAE (Physical Address Extension) was an addition to 32 bit x86 CPUs that allowed them to address more then 4GB physical memory.

http://en.wikipedia.org/wiki/Physical_Address_Extension
--
Tim Gardner [hidden email]

--
ubuntu-devel mailing list
[hidden email]
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel


--
ubuntu-devel mailing list
[hidden email]
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel
Reply | Threaded
Open this post in threaded view
|

Re: Dropping i386 non-PAE as a supported kernel flavour in Precise Pangolin

Stéphane Graber-2
In reply to this post by Tim Gardner-2
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA512

On 11/09/2011 04:43 PM, Tim Gardner wrote:

> Per discussion at UDS the kernel team is proposing to drop the
> non-PAE i386 flavour. The upgrade path for non-PAE users will be
> the PAE kernel. Those CPUs that do not have i686 and PAE support
> will be orphaned. To the best of my knowledge, these include Intel
> CPUs prior to Pentium II, 400Mhz Pentium M, VIA C3, and Geode LX.
> As far as I know, there are no laptop or desktop class CPUs being
> produced that do not meet these minimum requirements.
>
> Before I do something that is difficult to revert, I would like to
> hear from the development community why we should continue to
> maintain a kernel flavour that is (in my opinion) getting
> increasingly low utilization. It is my feeling that an extremely
> high percentage of users of the non-PAE kernel have a CPU that is
> PAE capable.

While it doesn't affect me personally, I've noticed that quite a few
friends of mine still have Pentium M hardware that don't support PAE.

These are mostly older Thinkapds and DELL laptops made in the last
decade and that can still run Oneiric pretty decently with no possible
fallback once the PAE kernel is dropped.
The only other thing that's not working with stock Oneiric on these is
Unity 3D because of some missing OpenGL extensions, Unity 2D works
fine though, AFAIK everything else is currently working as expected.

I'm sure that maintaining a separate non-PAE kernel must cost you
quite a lot of time, but I think it'd be great if we could have 12.04
as the last release supporting these.
As it's an LTS, that's going to give these users an extra 5 years of
support for their hardware which should be plenty enough for that
relatively old hardware.

> If there is sufficient community demand (and support), I would be
> willing to sponsor the first non-PAE kernel upload to Universe.
>
> https://wiki.ubuntu.com/KernelTeam/Specs/PreciseKernelConfigReview
>
> We'll be conducting a similar survey for powerpc.
>
> rtg
>
> P.S. For those of you that are totally confused by this email, PAE
> (Physical Address Extension) was an addition to 32 bit x86 CPUs
> that allowed them to address more then 4GB physical memory.
>
> http://en.wikipedia.org/wiki/Physical_Address_Extension


- --
Stéphane Graber
Ubuntu developer
http://www.ubuntu.com
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
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=6umr
-----END PGP SIGNATURE-----

--
ubuntu-devel mailing list
[hidden email]
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel
Reply | Threaded
Open this post in threaded view
|

Re: Dropping i386 non-PAE as a supported kernel flavour in Precise Pangolin

Mackenzie Morgan-3
In reply to this post by Tim Gardner-2

Laptops were still being produced without pae support after pentium 4 came out. I have a laptop from the early 2000s (second hand, so don't know the exact year, but an old dorm mate had it as her laptop when she started uni in 2004, so I'd guess that's the year) that kernel panics on pae kernels (learned that when I put ubuntu server on in 2007). It's the edubuntu one for teaching little cousins to code now.

Maco


--
ubuntu-devel mailing list
[hidden email]
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel
Reply | Threaded
Open this post in threaded view
|

Re: Dropping i386 non-PAE as a supported kernel flavour in Precise Pangolin

Colin Watson
In reply to this post by Tim Gardner-2
On Wed, Nov 09, 2011 at 02:43:28PM -0700, Tim Gardner wrote:
> Per discussion at UDS the kernel team is proposing to drop the
> non-PAE i386 flavour. The upgrade path for non-PAE users will be the
> PAE kernel. Those CPUs that do not have i686 and PAE support will be
> orphaned. To the best of my knowledge, these include Intel CPUs
> prior to Pentium II, 400Mhz Pentium M, VIA C3, and Geode LX. As far
> as I know, there are no laptop or desktop class CPUs being produced
> that do not meet these minimum requirements.

Does KVM work properly with PAE kernels at the moment?  I've had trouble
with it within the last six months, and when running server
installations I've had to tweak them on the fly to install the generic
kernel in order that I could boot the installed system.

> Before I do something that is difficult to revert, I would like to
> hear from the development community why we should continue to
> maintain a kernel flavour that is (in my opinion) getting
> increasingly low utilization.

I'd have thought we needed data here?  I'm worried about dropping the
kernel that's been the default during the installer for some time, in
one step.  If we want to switch the installer to generic-pae and then
drop the non-PAE kernel in the next cycle if that works out well, I'd be
happier with that approach; that gives us a much more graceful fallback
plan in the event that our opinions are mistaken.

Cheers,

--
Colin Watson                                       [[hidden email]]

--
ubuntu-devel mailing list
[hidden email]
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel
Reply | Threaded
Open this post in threaded view
|

Re: Dropping i386 non-PAE as a supported kernel flavour in Precise Pangolin

Scott Kitterman-3
In reply to this post by Stéphane Graber-2
On 11/09/2011 05:12 PM, Stéphane Graber wrote:
...
> As it's an LTS, that's going to give these users an extra 5 years of
> support for their hardware which should be plenty enough for that
> relatively old hardware.
...

As the proud owner of hardware that will not run on any release newer
than Hardy, I can vouch for this being really helpful.  As it is, I'll
probably get 4 more years of service out of the box than I would have
due to it being supported in an LTS.

I think right after an LTS is the best time to drop hardware support.

Scott K

--
ubuntu-devel mailing list
[hidden email]
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel
Reply | Threaded
Open this post in threaded view
|

Re: Dropping i386 non-PAE as a supported kernel flavour in Precise Pangolin

Colin Watson
In reply to this post by Amir Eldor-2
On Wed, Nov 09, 2011 at 11:53:26PM +0200, Amir Eldor wrote:
> I think these kind of computers will probably won't be installed with
> Ubuntu at all, but X/Lubuntu might be a choice for them. As I understand
> the kernel of Ubuntu is shared between the 'sister-releases', am I right?

Yes, any change to the Ubuntu kernel would affect Xubuntu and Lubuntu
too.

--
Colin Watson                                       [[hidden email]]

--
ubuntu-devel mailing list
[hidden email]
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel
Reply | Threaded
Open this post in threaded view
|

Re: Dropping i386 non-PAE as a supported kernel flavour in Precise Pangolin

Scott Kitterman-3
On 11/10/2011 04:26 AM, Colin Watson wrote:
> On Wed, Nov 09, 2011 at 11:53:26PM +0200, Amir Eldor wrote:
>> I think these kind of computers will probably won't be installed with
>> Ubuntu at all, but X/Lubuntu might be a choice for them. As I understand
>> the kernel of Ubuntu is shared between the 'sister-releases', am I right?
>
> Yes, any change to the Ubuntu kernel would affect Xubuntu and Lubuntu
> too.
>
It also affects servers.  I tend to use old desktops for specific server
tasks that aren't particularly CPU intensive and so I think that's
another use case that would be affected by this.

Scott K

--
ubuntu-devel mailing list
[hidden email]
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel
Reply | Threaded
Open this post in threaded view
|

Re: Dropping i386 non-PAE as a supported kernel flavour in Precise Pangolin

Adilson Oliveira-2
In reply to this post by Amir Eldor-2
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Em 09-11-2011 16:53, Amir Eldor escreveu:

> Hello Tim,
>
> The only reason I see for keeping support of weird old CPUs is for
> people who have an old computer and they would like to linux-ize
> it. I'm not sure if the "old" of today are computers running with
> powerpc or non-PAE CPUs, so my original claim might be void.
>
> I think these kind of computers will probably won't be installed
> with Ubuntu at all, but X/Lubuntu might be a choice for them. As I
> understand the kernel of Ubuntu is shared between the
> 'sister-releases', am I right?

In poor countries, many places like schools use recicled computers
donated by enterprises, non-profits, etc. Dropping the non-PAE kernel
can be a problem in this context.

[]s

Adilson.

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAk67xs8ACgkQYaKG37RGLIqfdgCeOnISzDZJnEyia7PT52xELEli
rlUAn091zFC69/RRtHhcZMZ28lBIo7Z9
=CM9c
-----END PGP SIGNATURE-----

--
ubuntu-devel mailing list
[hidden email]
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel
Reply | Threaded
Open this post in threaded view
|

Re: Dropping i386 non-PAE as a supported kernel flavour in Precise Pangolin

Tim Gardner-2
In reply to this post by Colin Watson
On 11/09/2011 03:14 PM, Colin Watson wrote:

> On Wed, Nov 09, 2011 at 02:43:28PM -0700, Tim Gardner wrote:
>> Per discussion at UDS the kernel team is proposing to drop the
>> non-PAE i386 flavour. The upgrade path for non-PAE users will be the
>> PAE kernel. Those CPUs that do not have i686 and PAE support will be
>> orphaned. To the best of my knowledge, these include Intel CPUs
>> prior to Pentium II, 400Mhz Pentium M, VIA C3, and Geode LX. As far
>> as I know, there are no laptop or desktop class CPUs being produced
>> that do not meet these minimum requirements.
>
> Does KVM work properly with PAE kernels at the moment?  I've had trouble
> with it within the last six months, and when running server
> installations I've had to tweak them on the fly to install the generic
> kernel in order that I could boot the installed system.
>

This just seems like a bug. If we don't address it early in this cycle,
then what incentive would we have to address it during the 12.10 dev cycle?

>> Before I do something that is difficult to revert, I would like to
>> hear from the development community why we should continue to
>> maintain a kernel flavour that is (in my opinion) getting
>> increasingly low utilization.
>
> I'd have thought we needed data here?

As far as I can tell there hasn't been a mass produced non-PAE cpu in
over 5 years that we (as a distro) care about. The consumer grade
electronics lifecycle is _well_ below 5 years. Furthermore, the distro
focus has been desktop with high performance 3D graphics and servers.
Where do non-PAE CPUs fit in that world? There are better distro choices
to fill that niche.

> I'm worried about dropping the
> kernel that's been the default during the installer for some time, in
> one step.  If we want to switch the installer to generic-pae and then
> drop the non-PAE kernel in the next cycle if that works out well, I'd be
> happier with that approach; that gives us a much more graceful fallback
> plan in the event that our opinions are mistaken.
>

I want to drop the non-PAE kernel _before_ the LTS. Otherwise we have to
deal with the complexities of LTS backported kernels _not_ having the
same flavour set as the released LTS kernel (something I'd prefer not to
have to do).

What do you think about dropping x86 32 bit kernels altogether for 14.04
? By then we should have _really_ good multi-arch support, and the CPUs
that we care about will all be 64 bit capable.

rtg
--
Tim Gardner [hidden email]

--
ubuntu-devel mailing list
[hidden email]
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel
Reply | Threaded
Open this post in threaded view
|

Re: Dropping i386 non-PAE as a supported kernel flavour in Precise Pangolin

Dotan Cohen
On Thu, Nov 10, 2011 at 17:14, Tim Gardner <[hidden email]> wrote:
> As far as I can tell there hasn't been a mass produced non-PAE cpu in over 5
> years that we (as a distro) care about. The consumer grade electronics
> lifecycle is _well_ below 5 years.

Most people that I know are using 3-5 year old hardware, and I would
say that the 5-8 year hardware is more common than 0-3 year old
hardware. Maybe we have less of a wasteful culture than those who
replace their entire desktops every <5 years, but even in instances
where a user _does_ replace his hardware so often, what becomes of
that hardware? I'll tell you: it becomes some relative's computer. Or
some neighbour's.

In fact, I pretty much maintain the computers of most of the 28
households in my building, and more than half are running some sub
1gHz Pentium 3 or such, with 256 MB of RAM. These are not "computer
enthusiasts" but rather normal people who make do with Kubuntu or XP
and browse Facebook and Picasa online. These people do not feel that
they are missing anything and have no need to upgrade.

--
Dotan Cohen

http://gibberish.co.il
http://what-is-what.com

--
ubuntu-devel mailing list
[hidden email]
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel
Reply | Threaded
Open this post in threaded view
|

Re: Dropping i386 non-PAE as a supported kernel flavour in Precise Pangolin

Mackenzie Morgan-3
In reply to this post by Adilson Oliveira-2

On Nov 10, 2011 10:49 AM, "Adilson Oliveira" <[hidden email]> wrote:
> In poor countries, many places like schools use recicled computers
> donated by enterprises, non-profits, etc. Dropping the non-PAE kernel
> can be a problem in this context.

I've been wondering for a while if my impression that ubuntu will help the digital divide was misplaced. :-/ Lubuntu still could, but not if this kernel gets dropped.

Maco


--
ubuntu-devel mailing list
[hidden email]
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel
Reply | Threaded
Open this post in threaded view
|

Re: Dropping i386 non-PAE as a supported kernel flavour in Precise Pangolin

Mackenzie Morgan-3
In reply to this post by Tim Gardner-2

On Nov 10, 2011 10:15 AM, "Tim Gardner" <[hidden email]> wrote:

> As far as I can tell there hasn't been a mass produced non-PAE cpu in over 5 years that we (as a distro) care about. The consumer grade electronics lifecycle is _well_ below 5 years. Furthermore, the distro focus has been desktop with high performance 3D graphics and servers. Where do non-PAE CPUs fit in that world? There are better distro choices to fill that niche.

Among gamer nerds or among people whose needs ubuntu can actually fill: the ones who just need a web browser and a word processor?

My experience with the latter group is that until the hardware craps out, they have no intention of spending money on a new one. Dad got rid of the pentium 2 when the motherboard died less than a year ago. Mom still has a c 2002 system running Ubuntu, and I'm pretty sure the last time any of their friends asked me to fix a windows blue screen it was well over 5 years old too.

Computers are replaced as frequently as refrigerators by people who don't care how quickly it loads a page or makes ice: when it stops turning on.

Maco


--
ubuntu-devel mailing list
[hidden email]
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel
Reply | Threaded
Open this post in threaded view
|

Re: Dropping i386 non-PAE as a supported kernel flavour in Precise Pangolin

Tim Gardner-2
In reply to this post by Tim Gardner-2
On 11/10/2011 08:14 AM, Tim Gardner wrote:

> On 11/09/2011 03:14 PM, Colin Watson wrote:
>> Does KVM work properly with PAE kernels at the moment? I've had trouble
>> with it within the last six months, and when running server
>> installations I've had to tweak them on the fly to install the generic
>> kernel in order that I could boot the installed system.
>>
>
> This just seems like a bug. If we don't address it early in this cycle,
> then what incentive would we have to address it during the 12.10 dev cycle?
>

I tested this on Precise today using testdrive on a 32 bit PAE server
kernel to host a 32 bit Precise PAE guest kernel. Similarly, I also
tested using a 64 bit host and a 32 bit PAE guest kernel.

Are those combinations sufficient ?

rtg
--
Tim Gardner [hidden email]

--
ubuntu-devel mailing list
[hidden email]
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel
Reply | Threaded
Open this post in threaded view
|

Re: Dropping i386 non-PAE as a supported kernel flavour in Precise Pangolin

Ben Hutchings-3
In reply to this post by Tim Gardner-2
On Wed, 2011-11-09 at 14:43 -0700, Tim Gardner wrote:
> Per discussion at UDS the kernel team is proposing to drop the non-PAE
> i386 flavour. The upgrade path for non-PAE users will be the PAE kernel.
> Those CPUs that do not have i686 and PAE support will be orphaned. To
> the best of my knowledge, these include Intel CPUs prior to Pentium II,
> 400Mhz Pentium M, VIA C3, and Geode LX. As far as I know, there are no
> laptop or desktop class CPUs being produced that do not meet these
> minimum requirements.

Assuming that '400Mhz Pentium M' means the 'Banias' models with a
400 MHz FSB, this agrees with my understanding of PAE support.

> Before I do something that is difficult to revert, I would like to hear
> from the development community why we should continue to maintain a
> kernel flavour that is (in my opinion) getting increasingly low
> utilization. It is my feeling that an extremely high percentage of users
> of the non-PAE kernel have a CPU that is PAE capable.

I agree.  In Debian testing/unstable we replaced the '686' flavour with
'686-pae', with a check on installation to tell people if it won't work.
There's been very little complaint about this, though of course we do
maintain a non-PAE flavour.

> If there is sufficient community demand (and support), I would be
> willing to sponsor the first non-PAE kernel upload to Universe.
[...]

Coincidentally, none of those non-PAE processors support SMP (at least
not in the standard way Linux supports).  So if you have a specifically
non-PAE flavour it's a useful optimisation to configure it as !SMP.

Ben.

--
Ben Hutchings
The program is absolutely right; therefore, the computer must be wrong.

--
ubuntu-devel mailing list
[hidden email]
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel

signature.asc (845 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Dropping i386 non-PAE as a supported kernel flavour in Precise Pangolin

Julien Lavergne-3
In reply to this post by Tim Gardner-2
On Thu, 10 Nov 2011 08:14:14 -0700
Tim Gardner <[hidden email]> wrote:

> As far as I can tell there hasn't been a mass produced non-PAE cpu in
> over 5 years that we (as a distro) care about. The consumer grade
> electronics lifecycle is _well_ below 5 years. Furthermore, the distro
> focus has been desktop with high performance 3D graphics and servers.
> Where do non-PAE CPUs fit in that world? There are better distro choices
> to fill that niche.

This type of hardware is the main target for Lubuntu. After reading some replies, it seems that more hardwares are affected by this. Depending of this list of affected hardwares, it may just kill the idea of Lubuntu (keeping people in Ubuntu universe, even if they use an old computer).

I also don't think it's wise to do it just before an LTS. Last time hardware support was dropped, it was for 10.10, making possible to send people affected to 10.04 with its LTS support.

Regards,
Julien Lavergne

--
Julien Lavergne <[hidden email]>

--
ubuntu-devel mailing list
[hidden email]
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel
Reply | Threaded
Open this post in threaded view
|

Re: Dropping i386 non-PAE as a supported kernel flavour in Precise Pangolin

Martin Pool-3
In reply to this post by Mackenzie Morgan-3
On 11 November 2011 04:01, Mackenzie Morgan <[hidden email]> wrote:

> Among gamer nerds or among people whose needs ubuntu can actually fill: the
> ones who just need a web browser and a word processor?
>
> My experience with the latter group is that until the hardware craps out,
> they have no intention of spending money on a new one. Dad got rid of the
> pentium 2 when the motherboard died less than a year ago. Mom still has a c
> 2002 system running Ubuntu, and I'm pretty sure the last time any of their
> friends asked me to fix a windows blue screen it was well over 5 years old
> too.
>
> Computers are replaced as frequently as refrigerators by people who don't
> care how quickly it loads a page or makes ice: when it stops turning on.

Those people are probably not upgrading their refrigerator firmware
all that often either.  They may not want a major new OS release.
They might install an update/backport of a particular app.

There is a group of people who want the latest-and-greatest software
on old or small hardware, but they're necessarily the crowd you're
describing here.

m

--
ubuntu-devel mailing list
[hidden email]
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel
Reply | Threaded
Open this post in threaded view
|

Re: Dropping i386 non-PAE as a supported kernel flavour in Precise Pangolin

Barry Warsaw-2
In reply to this post by Dotan Cohen
On Nov 10, 2011, at 06:08 PM, Dotan Cohen wrote:

>Most people that I know are using 3-5 year old hardware, and I would
>say that the 5-8 year hardware is more common than 0-3 year old
>hardware. Maybe we have less of a wasteful culture than those who
>replace their entire desktops every <5 years, but even in instances
>where a user _does_ replace his hardware so often, what becomes of
>that hardware? I'll tell you: it becomes some relative's computer. Or
>some neighbour's.

I've used this fact very effectively to migrate people to Ubuntu.  Many times,
they have an older computer that just starts dogging under Windows, maybe
because of bloatware, malware, or whatever.  Instead of spending $$$ on a new
computer, I'll give them a CD and tell them to give Ubuntu a try.  I've had
more than one friend/relative become very happy Ubuntu users because it
breathed new life into old hardware.  How awesome is it that you just saved a
friend several hundred to a thousand dollars *and* stealthily gave them some
freedom too?

(The servers in my closet are both at least a decade old, and happily run
10.04 LTS.  I hope to be able to upgrade them to 12.04 LTS.)

If you do end up dropping non-PAE, I think you need to do two important
things:

* Provide a very easy way for folks to determine whether their hardware is
  affected or not.
* Absolutely, positively, refuse to begin to upgrade such machines.  There's
  nothing worse than bricking their working hardware.

I'll note with a little sadness my inability to upgrade my first gen MacBook
Pro 1,1 to Lion because it doesn't support the Core Duo chip in that old
machine.  Hey wait, I know a good free operating system that *does* run on it!

-Barry

--
ubuntu-devel mailing list
[hidden email]
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel
Reply | Threaded
Open this post in threaded view
|

Re: Dropping i386 non-PAE as a supported kernel flavour in Precise Pangolin

John Meinel
In reply to this post by Martin Pool-3
...

>> Computers are replaced as frequently as refrigerators by people who don't
>> care how quickly it loads a page or makes ice: when it stops turning on.
>
> Those people are probably not upgrading their refrigerator firmware
> all that often either.  They may not want a major new OS release.
> They might install an update/backport of a particular app.
>
> There is a group of people who want the latest-and-greatest software
> on old or small hardware, but they're necessarily the crowd you're
> describing here.
>
> m
>

I think you mean 'not necessarily'. I agree, though I know we dealt with
a lot of this in our 'bzr python-compatibility' discussions. In that
particular case it was "we don't want to upgrade the OS, or even the
system libraries/python version, but we do want to upgrade a given
application". Which is a different level than "we don't want to upgrade
our hardware, but we do want to upgrade all of the OS and applications."

Certainly it is a bit different when one upgrade is $$ and the other is
free.

Still, it seems an open question for how to handle users that want the
latest-and-greatest X, but don't want the latest-and-greatest Y, even
though X depends on Y.

John
=:->

--
ubuntu-devel mailing list
[hidden email]
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel
Reply | Threaded
Open this post in threaded view
|

Re: Dropping i386 non-PAE as a supported kernel flavour in Precise Pangolin

John Meinel
In reply to this post by Barry Warsaw-2
On 11/11/2011 1:15 PM, Barry Warsaw wrote:

> On Nov 10, 2011, at 06:08 PM, Dotan Cohen wrote:
>
>> Most people that I know are using 3-5 year old hardware, and I would
>> say that the 5-8 year hardware is more common than 0-3 year old
>> hardware. Maybe we have less of a wasteful culture than those who
>> replace their entire desktops every<5 years, but even in instances
>> where a user _does_ replace his hardware so often, what becomes of
>> that hardware? I'll tell you: it becomes some relative's computer. Or
>> some neighbour's.
>
> I've used this fact very effectively to migrate people to Ubuntu.  Many times,
> they have an older computer that just starts dogging under Windows, maybe
> because of bloatware, malware, or whatever.  Instead of spending $$$ on a new
> computer, I'll give them a CD and tell them to give Ubuntu a try.  I've had
> more than one friend/relative become very happy Ubuntu users because it
> breathed new life into old hardware.  How awesome is it that you just saved a
> friend several hundred to a thousand dollars *and* stealthily gave them some
> freedom too?
>
> (The servers in my closet are both at least a decade old, and happily run
> 10.04 LTS.  I hope to be able to upgrade them to 12.04 LTS.)
>

PAE was first added in Pentium Pro in 1995. So your servers have to be
~16 years old to not support it.

So yes "a decade old" is still new enough, 2 decades old is not. Looking
here:
   http://en.wikipedia.org/wiki/Physical_Address_Extension

I'm trying to correlate that with:
   http://www.intel.com/pressroom/kits/quickrefyr.htm

And it looks like there is a 400MHz processor introduced around 1999. So
you're still looking at 1 decade being new enough, but slightly more
than that not.

I certainly ran a 450MHz dual-celeron A server for many years before I
finally upgraded it to PIII 700 (fastest I could find that could fit in
the slot).

 From what I can tell, the PIII version should have PAE.

> If you do end up dropping non-PAE, I think you need to do two important
> things:
>
> * Provide a very easy way for folks to determine whether their hardware is
>    affected or not.
> * Absolutely, positively, refuse to begin to upgrade such machines.  There's
>    nothing worse than bricking their working hardware.
>
> I'll note with a little sadness my inability to upgrade my first gen MacBook
> Pro 1,1 to Lion because it doesn't support the Core Duo chip in that old
> machine.  Hey wait, I know a good free operating system that *does* run on it!
>
> -Barry
>

John
=:->

--
ubuntu-devel mailing list
[hidden email]
Modify settings or unsubscribe at: https://lists.ubuntu.com/mailman/listinfo/ubuntu-devel
12