Leap 42.3 and Kernel

classic Classic list List threaded Threaded
9 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Leap 42.3 and Kernel

Bjoern Voigt
Is it already clear which kernel Leap 42.3 will get?

I tried one of the latest 42.3 snapshot releases (with kernel 4.4.x) on
a Lenovo Ideapad 500-15ACZ laptop with a AMD A10-8700P CPU. The laptop
is from 2016, so not brand-new. "dmesg" was showing a lot of warnings
because of unimplemented powersave features of the AMD CPU. After
installing kernel 4.11.x from Kernel:stable repository, the warnings
went disappeared. Currently I do not have access to the laptop, but I
can post the "dmesg" warnings on Tuesday or later.

Probably the brand-new AMD Ryzen CPUs will have more incompatibilities
with kernel 4.4.

4.4 is a longterm kernel. What about the latest longterm kernel series 4.9?

Greetings,
Björn
--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Leap 42.3 and Kernel

Carlos E. R.-2
On 2017-06-17 13:51, Bjoern Voigt wrote:
> Is it already clear which kernel Leap 42.3 will get?

More or less the same as you see in the beta.

> 4.4 is a longterm kernel. What about the latest longterm kernel series 4.9?

Don't expect it.
This is one of the effects of basing Leap on SLES.

--
Cheers / Saludos,

                Carlos E. R.
                (from 42.2 x86_64 "Malachite" at Telcontar)


signature.asc (188 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Leap 42.3 and Kernel

Karl Cheng
On 17 June 2017 at 22:39, Carlos E. R. <[hidden email]> wrote:
> Don't expect it.
> This is one of the effects of basing Leap on SLES.

Well that doesn't necessarily mean Leap *must* have the same kernel
(e.g. 42.1), but I would agree that it's unlikely we'd see a kernel
update at this point in time.

IMO (if there isn't one somewhere I'm not aware of) there should be
some sort of checklist for each release where decisions for important
updates are recorded down.

--
- Karl Cheng (Qantas94Heavy)
--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Leap 42.3 and Kernel

Carlos E. R.-2
On 2017-06-18 07:38, Karl Cheng wrote:
> On 17 June 2017 at 22:39, Carlos E. R. <[hidden email]> wrote:
>> Don't expect it.
>> This is one of the effects of basing Leap on SLES.
>
> Well that doesn't necessarily mean Leap *must* have the same kernel
> (e.g. 42.1), but I would agree that it's unlikely we'd see a kernel
> update at this point in time.

Well, basically it does :-)

It is not the exact same version, though.

> IMO (if there isn't one somewhere I'm not aware of) there should be
> some sort of checklist for each release where decisions for important
> updates are recorded down.

This was clear years ago, when Leap was first designed and made. No
surprise at all.

--
Cheers / Saludos,

                Carlos E. R.
                (from 42.2 x86_64 "Malachite" at Telcontar)


signature.asc (188 bytes) Download Attachment
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Leap 42.3 and Kernel

Larry Finger
On 06/18/2017 05:42 AM, Carlos E. R. wrote:

> On 2017-06-18 07:38, Karl Cheng wrote:
>> On 17 June 2017 at 22:39, Carlos E. R. <[hidden email]> wrote:
>>> Don't expect it.
>>> This is one of the effects of basing Leap on SLES.
>>
>> Well that doesn't necessarily mean Leap *must* have the same kernel
>> (e.g. 42.1), but I would agree that it's unlikely we'd see a kernel
>> update at this point in time.
>
> Well, basically it does :-)
>
> It is not the exact same version, though.

Although Leap42.3 is using a 4.4.X kernel, it does have some fixes ported from
later kernels so that it can support newer hardware. These backports are the
reason you see snippets like the following in spec files:

%if 0%{?sle_version} == 120300
%patch114 -p1
%endif

That one is needed for VirtualBox to build as the get_user_pages() API has been
changed to match the one updated in kernel 4.10.

Larry


--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Leap 42.3 and Kernel

Takashi Iwai
On Sun, 18 Jun 2017 17:20:57 +0200,
Larry Finger wrote:

>
> On 06/18/2017 05:42 AM, Carlos E. R. wrote:
> > On 2017-06-18 07:38, Karl Cheng wrote:
> >> On 17 June 2017 at 22:39, Carlos E. R. <[hidden email]> wrote:
> >>> Don't expect it.
> >>> This is one of the effects of basing Leap on SLES.
> >>
> >> Well that doesn't necessarily mean Leap *must* have the same kernel
> >> (e.g. 42.1), but I would agree that it's unlikely we'd see a kernel
> >> update at this point in time.
> >
> > Well, basically it does :-)
> >
> > It is not the exact same version, though.
>
> Although Leap42.3 is using a 4.4.X kernel, it does have some fixes
> ported from later kernels so that it can support newer hardware.

Right, there are tons of patches, namely Leap 42.3 has ca 16,000
patches on top of vanilla 4.4 kernel.

Also, the update of the whole KMS graphics stack is provided via
drm-kmp for Leap 42.3, instead of yet another a few thousands of
backport patches.  It'll be automatically drug onto system via rpm
supplements tag for the corresponding hardware.

Note that the package description of drm-kmp shows "the update for
kaby lake", but it's not true, it's the update for the whole
graphics.  The fix of the package description is already in queue.


Takashi
--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Leap 42.3 and Kernel

Andrei Borzenkov
18.06.2017 22:49, Takashi Iwai пишет:
>
> Also, the update of the whole KMS graphics stack is provided via
> drm-kmp for Leap 42.3, instead of yet another a few thousands of
> backport patches.  It'll be automatically drug onto system via rpm
> supplements tag for the corresponding hardware.
>

Can this be the reason for failed nvidia-drm loading? It sounds like drm
module is missing unless some extra RPM is installed?
--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Leap 42.3 and Kernel

Takashi Iwai
On Sun, 18 Jun 2017 22:13:51 +0200,
Andrei Borzenkov wrote:

>
> 18.06.2017 22:49, Takashi Iwai пишет:
> >
> > Also, the update of the whole KMS graphics stack is provided via
> > drm-kmp for Leap 42.3, instead of yet another a few thousands of
> > backport patches.  It'll be automatically drug onto system via rpm
> > supplements tag for the corresponding hardware.
> >
>
> Can this be the reason for failed nvidia-drm loading? It sounds like drm
> module is missing unless some extra RPM is installed?

Yes, it was my suspect and I asked for it on bugzilla, but the
reporter said it didn't help.


Takashi
--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Leap 42.3 and Kernel

Roman Bysh
On 18/06/17 04:35 PM, Takashi Iwai wrote:

> On Sun, 18 Jun 2017 22:13:51 +0200,
> Andrei Borzenkov wrote:
> >
> > 18.06.2017 22:49, Takashi Iwai пишет:
> >>
> >> Also, the update of the whole KMS graphics stack is provided via
> >> drm-kmp for Leap 42.3, instead of yet another a few thousands of
> >> backport patches.  It'll be automatically drug onto system via rpm
> >> supplements tag for the corresponding hardware.
> >>
> >
> > Can this be the reason for failed nvidia-drm loading? It sounds like drm
> > module is missing unless some extra RPM is installed?
>
> Yes, it was my suspect and I asked for it on bugzilla, but the
> reporter said it didn't help.
>
>
> Takashi
>
Thanks for your help. Nvidia driver now works.
--
To unsubscribe, e-mail: [hidden email]
To contact the owner, e-mail: [hidden email]

Loading...