Bug 6400 - IEEE1394 related freezes on ppc64
Summary: IEEE1394 related freezes on ppc64
Status: REJECTED INSUFFICIENT_DATA
Alias: None
Product: Drivers
Classification: Unclassified
Component: IEEE1394 (show other bugs)
Hardware: i386 Linux
: P2 normal
Assignee: drivers_ieee1394
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2006-04-17 11:58 UTC by Eric Van Hensbergen
Modified: 2006-07-05 02:45 UTC (History)
1 user (show)

See Also:
Kernel Version: 2.6.12
Subsystem:
Regression: ---
Bisected commit-id:


Attachments

Description Eric Van Hensbergen 2006-04-17 11:58:30 UTC
Most recent kernel where this bug did not occur:
Distribution: Ubuntu
Hardware Environment: ppc64 (Apple G5)
Software Environment: 
Problem Description:

I have experienced chronic freezes when using IEEE1394 on PPC64 kernels -- I
don't seem to see these problems on 32-bit x86.  When the problem occurs the
entire system freezes and I don't see anything in debug-output or syslog to clue
me into the problem.

The problem is most likely to occur when using an IEEE1394 camera (such as an
Apple iSight).  Problem seems to occur rather quickly (within 5-10 minutes of
activating the camera via GnomeMeeting or other application).

I also have witnessed the problems with IEEE1394 connected storage (such as an
iPod) -- although the problem typically takes longer to happen (3-4 hours).

Perhaps some sort of a race condition complicated by large amounts of data transfer?

I don't believe this is the same problem as Bug#6393 as I don't have the
configuration it details.

Steps to reproduce:

Connect an IEEE1394 camera, access using Gnome Meeting, wait for system to go
unresponsive (sorry I don't have a better reproduction procedure..I've had this
problem for 2 years, and held off trying to find a better way to describe or
fault isolate the problem).
Comment 1 Adrian Bunk 2006-04-17 15:37:40 UTC
Is this issue still present in kernel 2.6.16.6?
Comment 2 Adrian Bunk 2006-07-04 14:57:14 UTC
I'm assuming this issue is already fixed.

Please reopen this bug if it's still present in kernel 2.6.17.
Comment 3 Stefan Richter 2006-07-05 02:45:46 UTC
Better yet, *if* you know how to patch and build an own kernel, try up-to-date
FireWire drivers from http://me.in-berlin.de/~s5r6/linux1394/updates/.

If the system totally freezes, one would usually need a remote debugger to get
any clue of what went wrong. However it is sometimes possible to get some last
messages right before the freeze this way:
 - Switch to a text console (i.e. away from X11), login as root.
 - Restart the kernel log daemon to run in this console:
   # killall klogd
   # klogd
   Klogd will go into background, so you can continue to work in this console.
 - Start whatever may cause the freeze but stay in the console.
 - Take a photo of the screen if messages appear when the system freezes.

Even if you do not know how to update the kernel or drivers it would
nevertheless be interesting to know if the klogd experiment gives any related
kernel messages.

Note You need to log in before you can comment on or make changes to this bug.