2011-06-17 05:00:17

by Diego Viola

[permalink] [raw]
Subject: Please help us resolve the shutdown problem that is affecting our computers

Hello,

We are currently experiencing an issue that affects the shutdown process.

The problem occurs when we shutdown our machines, the system would
lock/hang during the shutdown process with a message similar to:
"Disabling IRQ #19.".

I send this email to the mailing list, hoping to get the attention of
a knowledgeable maintainer in order to fix this issue, which is
affecting not only me but many people. I already have started a thread
in bugzilla, but we haven't received many responses from developers.

The link to the bug report is:

https://bugzilla.kernel.org/show_bug.cgi?id=33872

We are able to reproduce the issue and the problem goes away on my
machine if I unload the 'e1000e' module before shutdown. This
workaround has worked for other people also, but I don't consider it
to be a real fix. There's also other people who have different
hardware and don't have this 'e1000e' module loaded at all and yet
they also experience shutdown hangs. How can this be explained?

Please help us to resolve this issue. We would happily assist in order
to test proposed patches, etc. We just want to get this problem fixed.

Thanks,
Diego Viola


2011-06-17 12:12:17

by Diego Viola

[permalink] [raw]
Subject: Re: Please help us resolve the shutdown problem that is affecting our computers

Any idea please?

On Fri, Jun 17, 2011 at 1:00 AM, Diego Viola <[email protected]> wrote:
> Hello,
>
> We are currently experiencing an issue that affects the shutdown process.
>
> The problem occurs when we shutdown our machines, the system would
> lock/hang during the shutdown process with a message similar to:
> "Disabling IRQ #19.".
>
> I send this email to the mailing list, hoping to get the attention of
> a knowledgeable maintainer in order to fix this issue, which is
> affecting not only me but many people. I already have started a thread
> in bugzilla, but we haven't received many responses from developers.
>
> The link to the bug report is:
>
> https://bugzilla.kernel.org/show_bug.cgi?id=33872
>
> We are able to reproduce the issue and the problem goes away on my
> machine if I unload the 'e1000e' module before shutdown. This
> workaround has worked for other people also, but I don't consider it
> to be a real fix. There's also other people who have different
> hardware and don't have this 'e1000e' module loaded at all and yet
> they also experience shutdown hangs. How can this be explained?
>
> Please help us to resolve this issue. We would happily assist in order
> to test proposed patches, etc. We just want to get this problem fixed.
>
> Thanks,
> Diego Viola
>

2011-06-17 13:05:27

by Alan

[permalink] [raw]
Subject: Re: Please help us resolve the shutdown problem that is affecting our computers

On Fri, 17 Jun 2011 08:12:12 -0400
Diego Viola <[email protected]> wrote:

> Any idea please?

I think you are under the mistaken impression this is a support list -
it's not. It's also not a list that deals with networking much but with
co-ordinating kernel patches and merges.

As it says in the driver source

Contacts:
e1000-devel Mailing List <[email protected]>

Although if isn't a recent kernel you may be better off going via your
vendor (particularly for things like SLES, Centos, RHEL etc which deviate
a lot from the mainstream)

Alan

2011-06-17 13:12:36

by Diego Viola

[permalink] [raw]
Subject: Re: Please help us resolve the shutdown problem that is affecting our computers

ok thanks, and sorry for the mistake.

On Fri, Jun 17, 2011 at 9:07 AM, Alan Cox <[email protected]> wrote:
> On Fri, 17 Jun 2011 08:12:12 -0400
> Diego Viola <[email protected]> wrote:
>
>> Any idea please?
>
> I think you are under the mistaken impression this is a support list -
> it's not. It's also not a list that deals with networking much but with
> co-ordinating kernel patches and merges.
>
> As it says in the driver source
>
> Contacts:
> e1000-devel Mailing List <[email protected]>
>
> Although if isn't a recent kernel you may be better off going via your
> vendor (particularly for things like SLES, Centos, RHEL etc which deviate
> a lot from the mainstream)
>
> Alan
>