--c7rykKtsZvepKFKR
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

On Wed, Apr 02, 2008 at 10:50:39AM +0100, Vince Hoffman wrote:
> Kostik Belousov wrote:
> > On Tue, Apr 01, 2008 at 09:55:19AM +0100, Vince Hoffman wrote:
> >> Kostik Belousov wrote:
> >>> On Mon, Mar 31, 2008 at 12:13:31PM +0100, Vince Hoffman wrote:
> >>>> Kostik Belousov wrote:
> >>> Compile the ddb into the kernel (instructions are provided at the same
> >>> web page) and post the output of the console on the panic
> >> I tried this but other than the fact I can drop into ddb and step thou=

gh
> >> stuff I'm not sure what I can do with it. Since I cant (obviously) get=

a
> >> serial console on my laptop I have done it the other way and taken a
> >> picture of the console on panic. When I have more time I'll carry on
> >> reading up on ddb but I think I have a steep learning curve since my
> >> previous looks at programing stopped at 'hello world'
> >> Panic console at:
> >> http://unsane.co.uk/~jhary/freebsd/IMAGE_119.jpg

> >=20
> > Thanks, this is enough. Two chunks were missed from the rev. 1.210.
> > Please, try the patch below.
> >=20
> > diff --git a/sys/kern/kern_conf.c b/sys/kern/kern_conf.c
> > index 843498e..7e6e048 100644
> > --- a/sys/kern/kern_conf.c
> > +++ b/sys/kern/kern_conf.c
> > @@ -402,8 +402,7 @@ giant_ioctl(struct cdev *dev, u_long cmd, caddr_t d=

ata, int fflag, struct thread
> > if (dsw =3D=3D NULL)
> > return (ENXIO);
> > mtx_lock(&Giant);
> > - retval =3D dev->si_devsw->d_gianttrick->
> > - d_ioctl(dev, cmd, data, fflag, td);
> > + retval =3D dsw->d_gianttrick->d_ioctl(dev, cmd, data, fflag, td);
> > mtx_unlock(&Giant);
> > dev_relthread(dev);
> > return (retval);
> > @@ -419,8 +418,7 @@ giant_read(struct cdev *dev, struct uio *uio, int i=

oflag)
> > if (dsw =3D=3D NULL)
> > return (ENXIO);
> > mtx_lock(&Giant);
> > - retval =3D dev->si_devsw->d_gianttrick->
> > - d_read(dev, uio, ioflag);
> > + retval =3D dsw->d_gianttrick->d_read(dev, uio, ioflag);
> > mtx_unlock(&Giant);
> > dev_relthread(dev);
> > return (retval);

>=20
> Hi Kostik,
> Yes this stops the panic. I do still get the console spammed with
> informational messages
>=20
> Apr 2 10:36:30 prawn kernel: ucom0: > 0/0, rev 1.10/4.00, addr 3> on uhub2
> Apr 2 10:36:33 prawn login: ROOT LOGIN (toor) ON ttyv0
> Apr 2 10:36:56 prawn kernel: ucom0: ucomreadcb: IOERROR
> Apr 2 10:36:56 prawn kernel: ucom0: at uhub2 port 1 (addr 3) disconnected
> Apr 2 10:36:57 prawn kernel: Still 4294967295 threads in ttyU0
> Apr 2 10:37:16 prawn last message repeated 188 times
> Apr 2 10:37:16 prawn login: ROOT LOGIN (toor) ON ttyv1
> Apr 2 10:37:16 prawn kernel: Still 4294967295 threads in ttyU0
> Apr 2 10:37:47 prawn last message repeated 303 times
> Apr 2 10:37:55 prawn last message repeated 85 times
>=20
>=20
> but I can live with that.
>=20
> I had opened PR usb/122287 for this, and I have updated this with the
> relevent information from this thread.


I committed the fix to the HEAD. I tested the patch with the uark(4)
cable, and simply running the tip over the ttyU0 did not produced the
"Still -1 threads in ttyU0" message. Are you running tip, or something
else over the port ?


--c7rykKtsZvepKFKR
Content-Type: application/pgp-signature
Content-Disposition: inline

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.8 (FreeBSD)

iEYEARECAAYFAkfzaxwACgkQC3+MBN1Mb4hi7wCfevuibsMCrg Wt8C+e4Xmd7SIb
v1wAoPbaEIc4Ts34yHrnac/vbIaknIgb
=Mhhg
-----END PGP SIGNATURE-----

--c7rykKtsZvepKFKR--