Date: Fri, 1 May 2009 21:30:28 +0200 From: Magnus Kling <klingfon@gmail.com> To: freebsd-acpi@freebsd.org Subject: Re: Fwd: Kernel panic on 7.2-RC1 when booting with ACPI enabled kernel. Message-ID: <43b1bb350905011230p1372e1ffw5ab61985e7672e19@mail.gmail.com> In-Reply-To: <200905011450.13899.jhb@freebsd.org> References: <43b1bb350904230622u4b7790f0p9f665b649c97a3b@mail.gmail.com> <200904300836.34238.jhb@freebsd.org> <43b1bb350905011130q4f58c018g66a9e4624c364b65@mail.gmail.com> <200905011450.13899.jhb@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
2009/5/1 John Baldwin <jhb@freebsd.org>
> On Friday 01 May 2009 2:30:28 pm Magnus Kling wrote:
> > 2009/4/30 John Baldwin <jhb@freebsd.org>
> >
> > > On Saturday 25 April 2009 6:27:23 am Magnus Kling wrote:
> > > > 2009/4/24 John Baldwin <jhb@freebsd.org>
> > > > > Can you do 'frame 10' followed by 'p *(struct acpi_pci_devinfo
> > > > > *)child->ivars'
> > > > >
> > > > > --
> > > > > John Baldwin
> > > >
> > > >
> > > > Sure, no problem. This is a none critical server so I can do alot o=
f
> > > > debugging and testing if that is needed.
> > > >
> > > >
> > > > (kgdb) frame 10
> > > > #10 0xc0db4ca8 in acpi_pci_child_location_str_method
> (cbdev=3D0xc2212680,
> > > > child=3D0xc2243400, buf=3D0xc22c2400 "slot=3D0 function=3D0 han=
dle=3D",
> > > > buflen=3D1024)
> > > > at
> /usr/src/sys/modules/acpi/acpi/../../../dev/acpica/acpi_pci.c:150
> > > > 150 strlcat(buf, acpi_name(dinfo->ap_handle), buflen);
> > > >
> > > > (kgdb) p *(struct acpi_pci_devinfo *)child->ivars
> > > > $1 =3D {ap_dinfo =3D {pci_links =3D {stqe_next =3D 0xc0b00f8c}, res=
ources =3D {
> > > > stqh_first =3D 0xc0b00f8c, stqh_last =3D 0x1030000}, cfg =3D =
{dev =3D
> 0x0,
> > > > bar =3D {4, 0, 0, 3257136600, 0, 0}, bios =3D 0, subvendor =
=3D 0,
> > > > subdevice =3D 0, vendor =3D 0, device =3D 0, cmdreg =3D 0, st=
atreg =3D 0,
> > > > baseclass =3D 0 '\0', subclass =3D 0 '\0', progif =3D 0 '\0',=
revid =3D
> 0
> > >
> > > Hmm, this is all completely wrong and trashed. What if you do 'p
> *child'?
> > >
> > > --
> > > John Baldwin
> > >
> > (kgdb) p *child
> > $2 =3D {ops =3D 0xc2161800, link =3D {tqe_next =3D 0xc2243380, tqe_prev=
=3D
> > 0xc2243484}, devlink =3D {tqe_next =3D 0xc2243380, tqe_prev =3D 0xc2243=
48c},
> > parent =3D 0xc2212680, children =3D {tqh_first =3D 0xc2262880, tqh_la=
st =3D
> > 0xc2262704}, driver =3D 0xc0b7066c, devclass =3D 0xc211e240, unit =3D 0=
,
> > nameunit =3D 0xc2241640 "atapci0", desc =3D 0xc223f900 "Promise PDC20=
621
> > UDMA100 controller", busy =3D 0, state =3D DS_ATTACHED, devflags =3D 0,
> > flags =3D 13, order =3D 0 '\0', pad =3D 0 '\0', ivars =3D 0xc223f5c0,=
softc =3D
> > 0xc2244800, sysctl_ctx =3D {tqh_first =3D 0xc2264380, tqh_last =3D 0xc2=
241594},
> > sysctl_tree =3D 0xc223f840}
> > (kgdb)
>
> Maybe try adding KTR traces for all calls to device_set_ivars(). I wonde=
r
> if
> something is trashing this device's ivars.
>
> Oh, dear. The ata(4) driver overwrites the ivars of some PCI devices it
> attaches to. This is very, very wrong. Which ATA controller do you have=
?
>
> --
> John Baldwin
>
Aha, I=B4m using a Promise Fasttrack SX4000 for a RAID1 setup. And the one
included on the motherboard for the OS.
And yes, I can confirm that without the Fasttrack SX4000 the system boots u=
p
correctly. (Pulled out the card and edited fstab.)
So you are right regarding that the ata driver messes something up. Do you
contact someone that is responsible for ata driver?
Thank you for taking the time to "correct" this,
Magnus
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?43b1bb350905011230p1372e1ffw5ab61985e7672e19>