Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 24 May 2004 17:23:39 -0700 (PDT)
From:      Luke <luked@pobox.com>
To:        Jon Noack <noackjr@alumni.rice.edu>
Cc:        current@freebsd.org
Subject:   Re: USB patch for better bus_dma and detachment support
Message-ID:  <Pine.NEB.4.58.0405241716510.8962@otaku.freeshell.org>
In-Reply-To: <40B18C47.9080102@alumni.rice.edu>
References:  <200405232109.aa46210@salmon.maths.tcd.ie> <40B18C47.9080102@alumni.rice.edu>

next in thread | previous in thread | raw e-mail | index | archive | help


On Mon, 24 May 2004, Jon Noack wrote:

> On 05/23/04 22:31, Luke wrote:
> > On Sun, 23 May 2004, Ian Dowse wrote:
> >
> >> In case anybody is interested in testing or helping with this,
> >> there's a patch at
> >>
> >> http://people.freebsd.org/~iedowse/usb.diff
> >>
> >> that attempts to begin addressing a number of problems with our
> >> current USB code:
> >
> > I'd love to try this out.  I'm using a Netgear FA120 USB ethernet
> > adapter with the axe driver, and my kernel panics with a page fault
> > in usb_transfer_complete every time I attempt to bring the interface
> > down. This means I can't reboot without a panic. It looks like this
> > code might address my problem.
> >
> > I've never tested a patch in this manner before though.  I assume
> > there's some utility like mergemaster to merge this into my existing
> > code.  If somebody can tell me how to merge in these changes I'll try
> > them out this week and let you know how it goes.
>
> The utility 'patch' is what you need to use.  Check out the patch(1) man
> page.  Something like the following should work (I haven't tried this):
>
> cd /usr/src/sys/dev/usb
> patch -p0 < usb.diff
>
> After applying the diff just rebuild and test.
>
> NOTE: you will have to reapply the diff after every CVSup (and possibly
> other update schemes).
>
> Jon Noack

Thanks.
I applied the patch, rebuilt and reinstalled the kernel, and I can see no
difference.  My system still hangs with the same page fault in
usb_transfer_complete when I attempt to shut it down or bring down the
network interface on my USB device.  The disconnecting/reconnecting
behavior appears unchanged as well, though it seems to detect the
reconnection a little faster... that might just be my imagination though.
It was plenty fast to start with.
I'm using ohci.  I don't believe I mentioned that earlier.



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.NEB.4.58.0405241716510.8962>