[BUG] e1000e, scheduling while atomic

[BUG] e1000e, scheduling while atomic

[Jérôme Carretero]

[-- Attachment #1: Type: text/plain, Size: 361 bytes --]

Hi,


I opened https://bugzilla.kernel.org/show_bug.cgi?id=218740 because I'm
not quite sure the culprit is e1000e or some timer stuff.

I just verified that this happens on the latest master.


Let me know if I can (quickly) help,


-- 
Jérôme

PS: Sorry if I'm sending this to too many people, I did a
get_maintainer timer and e1000e code.

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 854 bytes --]

[Intel-wired-lan] [BUG] e1000e, scheduling while atomic

[Jérôme Carretero]

[-- Attachment #1: Type: text/plain, Size: 361 bytes --]

Hi,


I opened https://bugzilla.kernel.org/show_bug.cgi?id=218740 because I'm
not quite sure the culprit is e1000e or some timer stuff.

I just verified that this happens on the latest master.


Let me know if I can (quickly) help,


-- 
Jérôme

PS: Sorry if I'm sending this to too many people, I did a
get_maintainer timer and e1000e code.

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 854 bytes --]

Re: [Intel-wired-lan] [BUG] e1000e, scheduling while atomic

[Sasha Neftin]

On 17/04/2024 21:46, Jérôme Carretero wrote:
> Hi,
> 
> 
> I opened https://bugzilla.kernel.org/show_bug.cgi?id=218740 because I'm
> not quite sure the culprit is e1000e or some timer stuff.

Hello Jérôme,

You hit on regression introduced by 6dbdd4de0362 (used the wrong timer 
during code running in atomic contexts)
Please, incorporate 
https://patchwork.ozlabs.org/project/intel-wired-lan/patch/20240417190320.3159360-1-vitaly.lifshits@intel.com/ 
in your latest kernel.

Thanks,
sasha

> 
> I just verified that this happens on the latest master.
> 
> 
> Let me know if I can (quickly) help,
> 
> 

Re: [Intel-wired-lan] [BUG] e1000e, scheduling while atomic

[Sasha Neftin]

On 17/04/2024 21:46, Jérôme Carretero wrote:
> Hi,
> 
> 
> I opened https://bugzilla.kernel.org/show_bug.cgi?id=218740 because I'm
> not quite sure the culprit is e1000e or some timer stuff.

Hello Jérôme,

You hit on regression introduced by 6dbdd4de0362 (used the wrong timer 
during code running in atomic contexts)
Please, incorporate 
https://patchwork.ozlabs.org/project/intel-wired-lan/patch/20240417190320.3159360-1-vitaly.lifshits@intel.com/ 
in your latest kernel.

Thanks,
sasha

> 
> I just verified that this happens on the latest master.
> 
> 
> Let me know if I can (quickly) help,
> 
> 

Re: [Intel-wired-lan] [BUG] e1000e, scheduling while atomic (stable)

[Jérôme Carretero]

[-- Attachment #1: Type: text/plain, Size: 1144 bytes --]

Hi Sasha,


Thank you, sorry for the delay but I coudln't reboot.

Adding Greg KH because I don't know if stable will receive my e-mail
(not subscribed) but the regression was integrated in stable:
 commit 0a4e3c2d976aa4dd38951afd6267f74ef3fade0e
so they should get the fix ASAP too.


Tested-by: Jérôme Carretero <cJ-ko@zougloub.eu>


Best regards,

-- 
Jérôme

On Thu, 2024-04-18 at 06:44 +0300, Sasha Neftin wrote:
> On 17/04/2024 21:46, Jérôme Carretero wrote:
> > Hi,
> > 
> > 
> > I opened https://bugzilla.kernel.org/show_bug.cgi?id=218740 because
> > I'm
> > not quite sure the culprit is e1000e or some timer stuff.
> 
> Hello Jérôme,
> 
> You hit on regression introduced by 6dbdd4de0362 (used the wrong
> timer 
> during code running in atomic contexts)
> Please, incorporate 
> https://patchwork.ozlabs.org/project/intel-wired-lan/patch/20240417190320.3159360-1-vitaly.lifshits@intel.com/
>  
> in your latest kernel.
> 
> Thanks,
> sasha
> 
> > 
> > I just verified that this happens on the latest master.
> > 
> > 
> > Let me know if I can (quickly) help,
> > 
> > 
> 


[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 854 bytes --]

Re: [Intel-wired-lan] [BUG] e1000e, scheduling while atomic (stable)

[Sasha Neftin]

On 19/04/2024 19:20, Jérôme Carretero wrote:
> Hi Sasha,
> 
> 
> Thank you, sorry for the delay but I coudln't reboot.
> 
> Adding Greg KH because I don't know if stable will receive my e-mail
> (not subscribed) but the regression was integrated in stable:
>   commit 0a4e3c2d976aa4dd38951afd6267f74ef3fade0e
> so they should get the fix ASAP too.
> 
> 
> Tested-by: Jérôme Carretero <cJ-ko@zougloub.eu>
Hello Jérôme,

Can you put your "Tested-by:" tag under 
https://patchwork.ozlabs.org/project/intel-wired-lan/patch/20240417190320.3159360-1-vitaly.lifshits@intel.com/ 
?
(then we will forward it via fixes, hopefully v6.9 candidate will pick 
up it)

Thanks,
Sasha
> 
> 
> Best regards,
> 

Re: [Intel-wired-lan] [BUG] e1000e, scheduling while atomic (stable)

[Sasha Neftin]

On 19/04/2024 19:20, Jérôme Carretero wrote:
> Hi Sasha,
> 
> 
> Thank you, sorry for the delay but I coudln't reboot.
> 
> Adding Greg KH because I don't know if stable will receive my e-mail
> (not subscribed) but the regression was integrated in stable:
>   commit 0a4e3c2d976aa4dd38951afd6267f74ef3fade0e
> so they should get the fix ASAP too.
> 
> 
> Tested-by: Jérôme Carretero <cJ-ko@zougloub.eu>
Hello Jérôme,

Can you put your "Tested-by:" tag under 
https://patchwork.ozlabs.org/project/intel-wired-lan/patch/20240417190320.3159360-1-vitaly.lifshits@intel.com/ 
?
(then we will forward it via fixes, hopefully v6.9 candidate will pick 
up it)

Thanks,
Sasha
> 
> 
> Best regards,
> 

Re: [Intel-wired-lan] [BUG] e1000e, scheduling while atomic (stable)

[Greg Kroah-Hartman]

On Fri, Apr 19, 2024 at 12:20:05PM -0400, Jérôme Carretero wrote:
> Hi Sasha,
> 
> 
> Thank you, sorry for the delay but I coudln't reboot.
> 
> Adding Greg KH because I don't know if stable will receive my e-mail
> (not subscribed) but the regression was integrated in stable:
>  commit 0a4e3c2d976aa4dd38951afd6267f74ef3fade0e
> so they should get the fix ASAP too.

The fix needs to show up in Linus's tree, is it on its way there?

thanks,

greg k-h

Re: [Intel-wired-lan] [BUG] e1000e, scheduling while atomic (stable)

[Greg Kroah-Hartman]

On Fri, Apr 19, 2024 at 12:20:05PM -0400, Jérôme Carretero wrote:
> Hi Sasha,
> 
> 
> Thank you, sorry for the delay but I coudln't reboot.
> 
> Adding Greg KH because I don't know if stable will receive my e-mail
> (not subscribed) but the regression was integrated in stable:
>  commit 0a4e3c2d976aa4dd38951afd6267f74ef3fade0e
> so they should get the fix ASAP too.

The fix needs to show up in Linus's tree, is it on its way there?

thanks,

greg k-h

Re: [Intel-wired-lan] [BUG] e1000e, scheduling while atomic (stable)

[Sasha Neftin]

On 23/04/2024 19:23, Greg Kroah-Hartman wrote:
> On Fri, Apr 19, 2024 at 12:20:05PM -0400, Jérôme Carretero wrote:
>> Hi Sasha,
>>
>>
>> Thank you, sorry for the delay but I coudln't reboot.
>>
>> Adding Greg KH because I don't know if stable will receive my e-mail
>> (not subscribed) but the regression was integrated in stable:
>>   commit 0a4e3c2d976aa4dd38951afd6267f74ef3fade0e
>> so they should get the fix ASAP too.
> 
> The fix needs to show up in Linus's tree, is it on its way there?

yes. Routed via intel-wired-lan: 
https://patchwork.ozlabs.org/project/intel-wired-lan/patch/20240417190320.3159360-1-vitaly.lifshits@intel.com/

> 
> thanks,
> 
> greg k-h

Re: [Intel-wired-lan] [BUG] e1000e, scheduling while atomic (stable)

[Sasha Neftin]

On 23/04/2024 19:23, Greg Kroah-Hartman wrote:
> On Fri, Apr 19, 2024 at 12:20:05PM -0400, Jérôme Carretero wrote:
>> Hi Sasha,
>>
>>
>> Thank you, sorry for the delay but I coudln't reboot.
>>
>> Adding Greg KH because I don't know if stable will receive my e-mail
>> (not subscribed) but the regression was integrated in stable:
>>   commit 0a4e3c2d976aa4dd38951afd6267f74ef3fade0e
>> so they should get the fix ASAP too.
> 
> The fix needs to show up in Linus's tree, is it on its way there?

yes. Routed via intel-wired-lan: 
https://patchwork.ozlabs.org/project/intel-wired-lan/patch/20240417190320.3159360-1-vitaly.lifshits@intel.com/

> 
> thanks,
> 
> greg k-h

Re: [Intel-wired-lan] [BUG] e1000e, scheduling while atomic (stable)

[Artem S. Tashkinov]

Hello,

This fix is still not queued in 6.8 stable:

https://git.kernel.org/pub/scm/linux/kernel/git/stable/stable-queue.git/tree/queue-6.8

Why?

Regards,
Artem

Re: [Intel-wired-lan] [BUG] e1000e, scheduling while atomic (stable)

[Greg KH]

On Sat, Apr 27, 2024 at 10:54:23AM +0000, Artem S. Tashkinov wrote:
> Hello,
> 
> This fix is still not queued in 6.8 stable:

What fix?

> https://git.kernel.org/pub/scm/linux/kernel/git/stable/stable-queue.git/tree/queue-6.8
> 
> Why?

What is the git id of the change in Linus's tree?

thanks,

greg k-h

Re: [Intel-wired-lan] [BUG] e1000e, scheduling while atomic (stable)

[Linux regression tracking (Thorsten Leemhuis)]

On 27.04.24 13:14, Greg KH wrote:
> On Sat, Apr 27, 2024 at 10:54:23AM +0000, Artem S. Tashkinov wrote:
>> This fix is still not queued in 6.8 stable:
> What fix?

FWIW, this one:
https://lore.kernel.org/all/20240417190320.3159360-1-vitaly.lifshits@intel.com/

But sadly the review is still pending (and hence it's not yet eligible
for inclusion in stable afaics):
https://lore.kernel.org/lkml/ded3e7ae-6a7d-48b2-8acc-c125874ee09f@leemhuis.info/

HTH, Ciao, Thorsten

Re: [Intel-wired-lan] [BUG] e1000e, scheduling while atomic (stable)

[Tony Nguyen]

On 4/28/2024 1:01 AM, Linux regression tracking (Thorsten Leemhuis) wrote:
> On 27.04.24 13:14, Greg KH wrote:
>> On Sat, Apr 27, 2024 at 10:54:23AM +0000, Artem S. Tashkinov wrote:
>>> This fix is still not queued in 6.8 stable:
>> What fix?
> 
> FWIW, this one:
> https://lore.kernel.org/all/20240417190320.3159360-1-vitaly.lifshits@intel.com/
> 
> But sadly the review is still pending (and hence it's not yet eligible
> for inclusion in stable afaics):
> https://lore.kernel.org/lkml/ded3e7ae-6a7d-48b2-8acc-c125874ee09f@leemhuis.info/

The team that deals with this driver had holidays last week so that 
delayed things a bit. I was able to touch base with them today and the 
patch has been sent to netdev.

https://lore.kernel.org/netdev/20240429171040.1152516-1-anthony.l.nguyen@intel.com/

Thanks,
Tony

> HTH, Ciao, Thorsten

Re: [Intel-wired-lan] [BUG] e1000e, scheduling while atomic (stable)

[Tony Nguyen]

On 4/28/2024 1:01 AM, Linux regression tracking (Thorsten Leemhuis) wrote:
> On 27.04.24 13:14, Greg KH wrote:
>> On Sat, Apr 27, 2024 at 10:54:23AM +0000, Artem S. Tashkinov wrote:
>>> This fix is still not queued in 6.8 stable:
>> What fix?
> 
> FWIW, this one:
> https://lore.kernel.org/all/20240417190320.3159360-1-vitaly.lifshits@intel.com/
> 
> But sadly the review is still pending (and hence it's not yet eligible
> for inclusion in stable afaics):
> https://lore.kernel.org/lkml/ded3e7ae-6a7d-48b2-8acc-c125874ee09f@leemhuis.info/

The team that deals with this driver had holidays last week so that 
delayed things a bit. I was able to touch base with them today and the 
patch has been sent to netdev.

https://lore.kernel.org/netdev/20240429171040.1152516-1-anthony.l.nguyen@intel.com/

Thanks,
Tony

> HTH, Ciao, Thorsten