On Sun, 24 Feb 2013 22:21:45 +0300 Dan Carpenter <dan.carpenter@xxxxxxxxxx> wrote: > Yeah. Sorry, if you can't reproduce the lockups on a stock kernel > then there isn't anything we can do. As I think you have suggested, I have enabled all kernel options seemingly related to lockdep. The boot messages seem to indicate that everything is OK ("all 218 testcases passed"), but the final result doesn't change. It's a hard lockup, the busy led on the card is on and the ethernet led indicates activity, although it doesn't respond anymore. This happens always while using lseek and write on a previously configured master window with a base address that actually exists. The test program (attached previous message) uses lots of lseeks and writes, so the context is the same, but the freezing point varies. Is there something else I can do? Thanks for any hint, -- Christoph _______________________________________________ devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxx http://driverdev.linuxdriverproject.org/mailman/listinfo/devel