How to find if BIOS has already enabled the device
Quoted 2 lines · Parag Warudkar
Quoted 1 line
This current problem of Hang-On-Boot if USB drive is attached does not
happen
Quoted 2 lines
Quoted 2 lines
with Windows - so it is some sort of additional (unnecessary?) thing which Linux does and the BIOS doesn't like. (Like re-enabling the controller
even
Quoted 6 lines
Quoted 3 lines
if BIOS has already enabled it or some such.)
Alan Cox wrote:
Provide dmesg output and we might be able to guess. The first obvious candidate would be the BIOS refusing to do a handover if it booted from USB disk.
Hi Alan
Sorry for digging this out so late - (Quick recap - My machine hangs for couple minutes on boot if USB storage disk is attached - hang occurs in pci_enable_device).
I have filed a bug to track this one -
http://bugme.osdl.org/show_bug.cgi?id=4711
Further analysis points towards wrong/differing IRQ assigments being the cause of this hang. I observed that when the machine hangs, the IRQ assignment looks like -
18: 379 IO-APIC-level eth0
19: 3 IO-APIC-level ohci1394 20: 1439 IO-APIC-level ohci_hcd, NVidia nForce3 Modem 21: 0 IO-APIC-level ohci_hcd, NVidia nForce3 22: 12884 IO-APIC-level ehci_hcd
And when it does NOT hang it looks like -
16: 3 IO-APIC-level ohci1394
18: 49277 IO-APIC-level nvidia 19: 1753 IO-APIC-level eth0 20: 6253 IO-APIC-level ohci_hcd 21: 646 IO-APIC-level ohci_hcd, NVidia nForce3 22: 2225 IO-APIC-level ehci_hcd, NVidia nForce3 Modem
Seems to me like the OHCI controller doesn't like to be assigned IRQ 19. Is this difference in IRQ assignment normal or is it a bug somewhere?
Parag