Bug 11690

Summary:	ohci-1394 makes the kernel hang while booting
Product:	Drivers	Reporter:	Rorschach (r0rschach)
Component:	IEEE1394	Assignee:	drivers_ieee1394
Status:	REJECTED INVALID
Severity:	blocking	CC:	bunk, r0rschach
Priority:	P1
Hardware:	All
OS:	Linux
Kernel Version:	2.6.26.5	Subsystem:
Regression:	Yes	Bisected commit-id:

Description Rorschach 2008-10-03 06:19:19 UTC

Latest known working kernel version: 2.6.24-19
Earliest known failing kernel version: 2.6.26.5
Distribution: Ubuntu Hardy Heron
Hardware Environment: Toshiba Satellite A100-773

Problem Description: I tried to compile the newest kernel. But the kernel hangs while booting with the following message:

[       9.248281] ohci 1394: fw-host0: OHCI-1394 1.1 (PCI): IRQ=[17] MMIO=[f8a06000-f8a067ff] Max Packet=[2048] IR/IT contexts=[4/8]


In the kernel config just "OHCI-1394 controllers" and "OHCI HCD support" are activated (as modules). The same configuration works with 2.6.24-19 kernel!

Comment 1 Stefan Richter 2008-10-03 07:16:53 UTC

Which FireWire controller is built into the A100-773 according to lspci?

Please test the following:
  - Make sure that ohci1394 is not auto-loaded during boot.  I don't know how
    this is done in Ubuntu, but maybe you need to remove ohci1394 from
    /etc/modules or need to add the line "blacklist ohci1394" to
    /etc/modprobe.conf.
  - Then boot, check that everything started up properly and that no error
    messages are in dmesg.
  - Switch away from X11 to a text console and load ohci1394 (modprobe
    ohci1394).  Watch the kernel log.

This should confirm that ohci1394 (and not something else) is what hangs the kernel, and possibly yield some more information about how the hang happens. Attach the complete dmesg from that session here.

Furthermore, since 2.6.24-19 is a distributor kernel (and you probably used it as compiled package) while you apparently built 2.6.26.5 from vanilla sources yourself, it would be interesting to have a self-built vanilla 2.6.24.7 tested too.

Thanks.

Comment 2 Rorschach 2008-10-09 07:54:20 UTC

Okay I marked this bug as invalid, because I can't reproduce it anymore. The kernel know boots fine.