Re: bad htb_{en,re}queue return codes causing corrupt data in drivers

Re: bad htb_{en,re}queue return codes causing corrupt data in drivers

[Jan-Aage Frydenbø-Bruvoll]

> It hasn't, we're doing an audit of the entire packet scheduler layer
> to make sure we fix this bug everywhere it occurs.
>
> Thanks for pinging us.

Hi,

As I've just installed 2.6.28 (vanilla) and seen the same type of
crash that has been bugging me for the last couple of weeks, I was
just wondering how I can get at this fix for my boxes, or when the fix
could be expected to enter the mainstream tree?

Thanks
Jan

Re: bnx2_poll panicking kernel

[Josip Rodin]

On Mon, Jul 14, 2008 at 11:48:51PM +0200, Jarek Poplawski wrote:
> > OK, I'll give it a whirl.
> > 
> > (Should I update to .26 at the same time, now that it went live in the
> > meantime?)
> 
> My proposal is to try .25 first.

It looks like the patch fixed my issues - no relevant printk's appeared
since I've applied it, so apparently there is no more cloned skb splitting
or corrupted data getting to bnx2.

I used the latest variant, from Jarek's mail
<20080714212223.GA4849@ami.dom.local>, on the 2.6.25.6 kernel.

Thanks all!

(Now if only someone could make sure this propagates into the stable kernel
branches, since those were pretty nasty crashes...)

-- 
     2. That which causes joy or happiness.

Re: bnx2_poll panicking kernel

[David Miller]

From: Josip Rodin <joy@entuzijast.net>
Date: Thu, 17 Jul 2008 23:30:58 +0200

> It looks like the patch fixed my issues - no relevant printk's appeared
> since I've applied it, so apparently there is no more cloned skb splitting
> or corrupted data getting to bnx2.
> 
> I used the latest variant, from Jarek's mail
> <20080714212223.GA4849@ami.dom.local>, on the 2.6.25.6 kernel.
> 
> Thanks all!
> 
> (Now if only someone could make sure this propagates into the stable kernel
> branches, since those were pretty nasty crashes...)

We will make sure the fix propagates in some form, thanks for testing.

bad htb_{en,re}queue return codes causing corrupt data in drivers [was Re: bnx2_poll panicking kernel]

[Josip Rodin]

On Thu, Jul 17, 2008 at 02:44:07PM -0700, David Miller wrote:
> From: Josip Rodin <joy@entuzijast.net>
> Date: Thu, 17 Jul 2008 23:30:58 +0200
> 
> > It looks like the patch fixed my issues - no relevant printk's appeared
> > since I've applied it, so apparently there is no more cloned skb splitting
> > or corrupted data getting to bnx2.
> > 
> > I used the latest variant, from Jarek's mail
> > <20080714212223.GA4849@ami.dom.local>, on the 2.6.25.6 kernel.
> > 
> > Thanks all!
> > 
> > (Now if only someone could make sure this propagates into the stable kernel
> > branches, since those were pretty nasty crashes...)
> 
> We will make sure the fix propagates in some form, thanks for testing.

Hi again, sorry to bother you, but it's been a few weeks now and I can't
seem see that fix in any of stable/linux-2.6.2[56].y.git, davem/net-2.6.git,
sfr/linux-next.git, so I just wanted to check that it hasn't fallen through
the cracks :)

-- 
     2. That which causes joy or happiness.

Re: bad htb_{en,re}queue return codes causing corrupt data in drivers

[David Miller]

From: Josip Rodin <joy@entuzijast.net>
Date: Sat, 2 Aug 2008 14:28:22 +0200

> On Thu, Jul 17, 2008 at 02:44:07PM -0700, David Miller wrote:
> > From: Josip Rodin <joy@entuzijast.net>
> > Date: Thu, 17 Jul 2008 23:30:58 +0200
> > 
> > > It looks like the patch fixed my issues - no relevant printk's appeared
> > > since I've applied it, so apparently there is no more cloned skb splitting
> > > or corrupted data getting to bnx2.
> > > 
> > > I used the latest variant, from Jarek's mail
> > > <20080714212223.GA4849@ami.dom.local>, on the 2.6.25.6 kernel.
> > > 
> > > Thanks all!
> > > 
> > > (Now if only someone could make sure this propagates into the stable kernel
> > > branches, since those were pretty nasty crashes...)
> > 
> > We will make sure the fix propagates in some form, thanks for testing.
> 
> Hi again, sorry to bother you, but it's been a few weeks now and I can't
> seem see that fix in any of stable/linux-2.6.2[56].y.git, davem/net-2.6.git,
> sfr/linux-next.git, so I just wanted to check that it hasn't fallen through
> the cracks :)

It hasn't, we're doing an audit of the entire packet scheduler layer
to make sure we fix this bug everywhere it occurs.

Thanks for pinging us.