e1000e NVM mutex contention warnings

e1000e NVM mutex contention warnings

[Ben Hutchings]

[-- Attachment #1: Type: text/plain, Size: 256 bytes --]

David,

What's your plan for collecting information about NVM mutex contention
and when do you intend to remove the warnings?  I have to say I find
this way of testing very user-unfriendly and would like to see it gone
as soon as possible.

Ben.


[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

RE: e1000e NVM mutex contention warnings

[Brandeburg, Jesse]

Ben Hutchings wrote:
> What's your plan for collecting information about NVM mutex contention
> and when do you intend to remove the warnings?  I have to say I find
> this way of testing very user-unfriendly and would like to see it gone
> as soon as possible.

Hi Ben,

checking linus' tree and davem's current net-2.6 tree shows that the
code was removed as of commit: 0a834a36ac92375cd82d9e4fe4f571e257997d6a

it was put in 2/14/2009, and was included in 2.6.30-rc1 by linus.

are you suggesting that we maybe try to push to -stable too?

Jesse

RE: e1000e NVM mutex contention warnings

[Ben Hutchings]

[-- Attachment #1: Type: text/plain, Size: 992 bytes --]

On Thu, 2009-04-23 at 15:15 -0700, Brandeburg, Jesse wrote:
> Ben Hutchings wrote:
> > What's your plan for collecting information about NVM mutex contention
> > and when do you intend to remove the warnings?  I have to say I find
> > this way of testing very user-unfriendly and would like to see it gone
> > as soon as possible.
> 
> Hi Ben,
> 
> checking linus' tree and davem's current net-2.6 tree shows that the
> code was removed as of commit: 0a834a36ac92375cd82d9e4fe4f571e257997d6a
> 
> it was put in 2/14/2009, and was included in 2.6.30-rc1 by linus.

Thanks.  I had a look but somehow missed that this was already on the
way out in 2.6.30.

> are you suggesting that we maybe try to push to -stable too?

If you now have the information you need about contention, then it seems
to me that it would be worthwhile to remove the unnecessary warnings.
If you don't want to submit a stable update then I can patch it out in
the Debian kernel package.

Ben.


[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: e1000e NVM mutex contention warnings

[Jeff Kirsher]

On Thu, Apr 23, 2009 at 4:14 PM, Ben Hutchings <ben@decadent.org.uk> wrote:
> On Thu, 2009-04-23 at 15:15 -0700, Brandeburg, Jesse wrote:
>> Ben Hutchings wrote:
>> > What's your plan for collecting information about NVM mutex contention
>> > and when do you intend to remove the warnings?  I have to say I find
>> > this way of testing very user-unfriendly and would like to see it gone
>> > as soon as possible.
>>
>> Hi Ben,
>>
>> checking linus' tree and davem's current net-2.6 tree shows that the
>> code was removed as of commit: 0a834a36ac92375cd82d9e4fe4f571e257997d6a
>>
>> it was put in 2/14/2009, and was included in 2.6.30-rc1 by linus.
>
> Thanks.  I had a look but somehow missed that this was already on the
> way out in 2.6.30.
>
>> are you suggesting that we maybe try to push to -stable too?
>
> If you now have the information you need about contention, then it seems
> to me that it would be worthwhile to remove the unnecessary warnings.
> If you don't want to submit a stable update then I can patch it out in
> the Debian kernel package.
>
> Ben.

I am fine with pushing this patch to -stable.

-- 
Cheers,
Jeff