From: John L. <le...@mo...> - 2005-05-25 03:44:16
|
On Wed, May 25, 2005 at 10:47:55AM +0800, Lv Zheng wrote: > 1. we still get kernel panic occasionally on ARM platform in > execution of spinlock rwsem's down_read, which seems caused I don't think anybody has the time to debug problems caused by the work you've done yourself... > 2. Even using 2.6 kernel, we could not associate samples sampled > after do_exit with the relevant process, that is, the samples for > do_exit is associated with "vmlinux", not relevant program with > "separate=kernel". Since handoff_task is implemented in 2.6, we > could not find any possible reasons that will cause this problem. How did you ascertain this? You're very short on information here. Whilst current still has the right information, we should be getting samples for the right pid. regards, john |