Re: 2.6.24.4 Sonics Silicon Backplane problems in dmesg

["Mark Knecht" <markknecht@gmail.com>]

On Sun, May 4, 2008 at 4:35 PM, Larry Finger <Larry.Finger@lwfinger.net> wrote:
> Larry Finger wrote:
>
> > I'm attempting to bring up a Compaq F752US laptop running Gentoo. The
> > kernel.org 2.6.24.4, Gentoo Vanilla-sources and 2.6.24-gentoo-r4
> > kernels all fail to boot without selecting the 'noapic' option on the
> > kernel command line. If I add noapic irqpoll the machine does boot but
> > I see this problem for all 3 kernels:
> >
> > Is this possibly caused by having to choose the noapic irqpoll options?
> >
> > <SNIP>
> > WARNING: at drivers/ssb/main.c:883 ssb_tmslow_reject_bitmask()
> > Pid: 1, comm: swapper Not tainted 2.6.24-gentoo-r4 #4
> > [<c0375e3f>] ssb_tmslow_reject_bitmask+0x4b/0x56
> > [<c0376504>] ssb_device_is_enabled+0x9/0x2d
> >
>
>  The warning above is caused by the configuration of the device below. I
> found
>  and fixed this problem several months ago, but after the 2.6.24 window
> closed.
>

Thanks Larry. I downloaded 2.6.25 but hadn't had a chance to build it
yet. I'll get on that.

>
>
> > 03:00.0 Network controller: Broadcom Corporation BCM94311MCG wlan
> > mini-PCI (rev 02)
> >
>
>  The problem is fixed in the wireless-testing.git tree, and in mainline
> 2.6.25. It is unlikely that it is causing the boot problem; however that
> could be checked by trying 2.6.25.
>

Yeah, doesn't seem likely. The non-boot issue was earlier in the boot
process with messages like these:


Intel machine check architecture supported.
Intel machine check reporting enabled on CPU#1
CPU1: AMD Athlon(tm) 64 X2 Mobile Technology TK-53 stepping 01
Total of 2 processors activated (6834.25 BogoMIPS).
ENABLING IO-APIC IRQs
..TIMER: vector=0x31 apic1=0 pin1=2 apic2=-1 pin2=-1

and then it just hangs.

Since I wrote the message this morning I munged through the kernel
config and came up with a config file that doesn't hang without the
options. I'll send the two config files along to the list under a
different subject line that speaks directly to this problem.

Thanks,
Mark