On Thu, Aug 25, 2011 at 11:58:22AM +0200, Josip Rodin wrote: > > It looks as if something is generally screwing with the PCI detection logic > from the get-go. > > Do you happen to have a working kernel to fall back to, and then do a > git bisect? Yes I agree that there's something screwy in the detection logic. I get a "console [tty0] enabled, bootconsole disabled" when I don't use the keep_bootcon kernel parameter, even though I have the input-device=keyboard and output-device=screen openprom variables set. So add bootconsole detection to the list of things that are screwy. The most recent working kernel I have is 2.6.20 from the gentoo 2007.0 livecd. I'm pretty sure I tried a 2.6.26 boot CD that failed, so I assume that the problems begin sometime between those versions. I'm happy to try different kernels from 2.6.20 onwards until I hit the bug, thought it will take me some time to work through them. Also unfortunately the keep_bootcon patch (http://patchwork.ozlabs.org/patch/78102/) doesn't apply cleanly to 2.6.20, which I would need to see the kernel panics. I'll get started and let you know when I track down where the problems begin. -- To unsubscribe from this list: send the line "unsubscribe sparclinux" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
