Bug#313552: kernel-image-2.6.11-alpha: qla1280 driver doesn't work with ISP1020 (PCI ID: 1077:1020)
On Thu, Dec 04, 2008 at 03:02:29PM -0800, Steve Langasek wrote:
> On Thu, Dec 04, 2008 at 11:44:11PM +0100, Moritz Muehlenhoff wrote:
> > On Mon, Jan 02, 2006 at 09:06:12PM -0800, Steve Langasek wrote:
>
> > > On Mon, Jan 02, 2006 at 11:32:50PM +0100, David Schmitt wrote:
>
> > > > Could you please retest this with a current 2.6.14 or 2.6.15-rc (from
> > > > experimental) image?
>
> > > The bug still exists with 2.6.14. I haven't tested 2.6.15 yet, but I
> > > eventually worked around the bug by throwing a second SCSI controller in the
> > > box so I could make some headway on d-i for etch.
>
> > Does this error still occur with more recent kernel versions?
>
> Yes.
>
> [4194078.920013] qla1280: QLA1040 found on PCI bus 1, dev 1
> [4194078.920013] PCI: Setting latency timer of device 0000:01:01.0 to 64
> [4194078.920989] qla1280: Failed mbox check
> [4194079.693450] qla1280_mailbox_command: Command failed, mailbox0 = 0x0007,
> mailbox_out0 = 0x4003, istatus = 0x0000
> [4194079.693450] m0 4003, m1 aaaa, m2 0000, m3 aa55
> [4194079.693450] m4 55aa, m5 a5a5, m6 a5a5, m7 a5a5
> [4194079.693450] scsi(1): RISC checksum failed.
> [4194079.693450] scsi(1): initialize: pci probe failed!
> [4194079.693450] qla1x160: Failed to initialize adapter
> [4194079.694426] qla1280: QLA1040 found on PCI bus 1, dev 2
> [4194079.694426] PCI: Setting latency timer of device 0000:01:02.0 to 64
> [4194079.694426] qla1280: Failed mbox check
> [4194080.804777] qla1280_mailbox_command: Command failed, mailbox0 = 0x0007,
> mailbox_out0 = 0x4003, istatus = 0x0000
> [4194080.804777] m0 4003, m1 aaaa, m2 0000, m3 aa55
> [4194080.804777] m4 55aa, m5 a5a5, m6 a5a5, m7 a5a5
> [4194080.804777] scsi(2): RISC checksum failed.
> [4194080.804777] scsi(2): initialize: pci probe failed!
> [4194080.806731] qla1x160: Failed to initialize adapter
>
> $ uname -a
> Linux quetzlcoatl 2.6.26-1-alpha-generic #1 Sat Oct 18 14:48:43 UTC 2008 alpha GNU/Linux
> $
Since the bug is originally from three years ago, it's very likely still present
in mainline. Could you file this upstream?
Cheers,
Moritz
Reply to: