I've got a Lenovo T530. Booting a 3.8 based Fedora kernel my machine
works just fine. Booting a 3.9-rc1 based Fedora kernel my USB 2.0 root
hub doesn't seem to notice when devices are plugged in. (The USB 3.0
root hub has no such problem) I'm assuming (maybe stupidly) it is
associated with:
Mar 06 09:52:35 localhost kernel: pci 0000:00:1c.0: can't find IRQ for PCI INT A; please try using pci=biosirq
Mar 06 09:52:35 localhost kernel: pci 0000:00:1c.1: can't find IRQ for PCI INT B; please try using pci=biosirq
Mar 06 09:52:35 localhost kernel: pci 0000:00:1c.2: can't find IRQ for PCI INT C; please try using pci=biosirq
(also, funny thing, pci=biosirq is not a valid command line option, glad
it told me to try that)
I also notice that the 3.9 kernels cannot reboot. The kernel output the
machine restart message and then just hangs forever. Again, this works
just fine on the 3.8 kernels.
Attached you will find my 3.8 kernel and boot log along with my 3.9
config and boot logs. Please tell me who to contact and what to collect
to try/debug
On Wed, Mar 6, 2013 at 10:59 AM, Eric Paris <[email protected]> wrote:
> I've got a Lenovo T530. Booting a 3.8 based Fedora kernel my machine
> works just fine. Booting a 3.9-rc1 based Fedora kernel my USB 2.0 root
> hub doesn't seem to notice when devices are plugged in. (The USB 3.0
> root hub has no such problem) I'm assuming (maybe stupidly) it is
> associated with:
>
> Mar 06 09:52:35 localhost kernel: pci 0000:00:1c.0: can't find IRQ for PCI INT A; please try using pci=biosirq
> Mar 06 09:52:35 localhost kernel: pci 0000:00:1c.1: can't find IRQ for PCI INT B; please try using pci=biosirq
> Mar 06 09:52:35 localhost kernel: pci 0000:00:1c.2: can't find IRQ for PCI INT C; please try using pci=biosirq
>
> (also, funny thing, pci=biosirq is not a valid command line option, glad
> it told me to try that)
>
> I also notice that the 3.9 kernels cannot reboot. The kernel output the
> machine restart message and then just hangs forever. Again, this works
> just fine on the 3.8 kernels.
You're booting via EFI, right? I'm debugging an identical issue on an
EFI machine here. Basically, it looks like the efi_info variable in the
bootparams thing is cleared so EFI/ACPI is all broken.
There's a suspect commit but I'm still bisecting so I won't be blaming
anything yet. If you'd like to follow along in Fedora land, I'm working
with bug 918408.
josh