[rt2x00-users] [bugzilla-daemon at bugzilla.kernel.org: [Bug 42828] rt2800pci unstable - chokes after too much I/O]

Andreas Hartmann andihartmann at 01019freenet.de
Tue Oct 23 14:42:51 AEDT 2012


On Mon, Oct 22, 2012 22:30:17 +0200
Gertjan van Wingerde <gwingerde at gmail.com> wrote:

> Hi Andreas,
> 
> On Mon, Oct 22, 2012 at 9:43 PM, Andreas Hartmann
> <andihartmann at 01019freenet.de> wrote:
> > Gertjan van Wingerde wrote:
> >> On 22 okt. 2012, at 05:44, Andreas Hartmann
> >> <andihartmann at 01019freenet.de> wrote:
> >>
> >>> Gertjan van Wingerde wrote:
> >>>> On 10/21/12 20:53, Andreas Hartmann wrote:
> >>>>> Francisco Pina Martins wrote:
> >>>>>> On Sun, 2012-10-21 at 11:02 +0200, Andreas Hartmann wrote:
> >>>>>>
> >>>>>>> Francisco Pina Martins wrote:
> >>>>>>>> Here is my update on this: I cannot boot with the patched
> >>>>>>>> kernel - I get a kernel panic at about 9 secs during
> >>>>>>>> boot.
> >>>>>>>
> >>>>>>> Stanislaws patch hangs the machine for me, too. See my
> >>>>>>> other post.
> >>>>>>
> >>>>>> Guess it's a no-go then.
> >>>>>
> >>>>> It was only a test patch. He wanted to help me, but I did it
> >>>>> manually to get the information, too. Therefore: no problem.
> >>>>
> >>>> Yeah, the problem with Stanislaw's patch was that he was trying
> >>>> to access fields that were no longer valid at that point in the
> >>>> code.
> >>>>
> >>>> Could you check if the attached (also untested) patch works
> >>>> better?
> >>>
> >>> Well, the patch doesn't crash the machine, anyway it doesn't work
> >>> as expected: the AP is recognized as STA, too.
> >>>
> >>
> >> Hmmm, that's odd. That would mean mac80211 would not indicate that
> >> it is an AP interface, unless there is a bug in the detection of
> >> this, but I cannot see that. Unfortunately I cannot diagnose this
> >> further until the weekend, as I am away from home for the week.
> >
> >
> >
> > 1. I inserted a comment at this point:
> >
> > +       if (!(skbdesc->flags & SKBDESC_TX_IS_AP) && !success) {
> > +               INFO(rt2x00dev,"That's a STA\n");
> > +               tx_info->flags |= IEEE80211_TX_STAT_AMPDU_NO_BACK;
> > +       }
> >
> > -> I got "That's a STA" when running the AP (I'm running it w/o
> > monitor device). This shouldn't come up when running as STA.
> >
> >
> > 2. The behaviour is the same as without
> > "!(skbdesc->flags & SKBDESC_TX_IS_AP) &&" - and the same as it was
> > before my patch but with "mac80211: retry sending failed BAR frames
> > later instead of tearing down aggr" applied.
> >
> 
> Thanks for that Andreas. I have found where the problem with my patch
> is (it is similar yet different from the problem with Stanislaw's
> original patch).
> I also found a way to generate a new patch on my Windows laptop.

Cool! Thanks for your fast support!

> Find attached take 2 of this patch. This time it is uncompiled and
> untested, but I have good hopes for this version.

This time, the detection of the AP seems to work as expected (only
short test by now).
More this evening!


Thanks,
kind regards,
Andreas




More information about the users mailing list