Re: Fw: 2.5.61 oops running SDET

[Anton Blanchard]

> OK, I did the following, which is what I think you wanted, plus Zwane's
> observation that task_state acquires the task_struct lock (we're the only 
> caller, so I just removed it), but I still get the same panic and this time
> the box hung. No doubt I've just done something stupid in the patch ...
> (task_state takes the tasklist_lock ... is that safe to do inside task_lock?)

It looks like you now try and take the tasklist_lock inside a task_lock()
region, but task_lock says no-can-do:

/* Protects ->fs, ->files, ->mm, and synchronises with wait4().  Nests
 * inside tasklist_lock */
static inline void task_lock(struct task_struct *p)
{
...

Anton