Hi Peter, On 2016/04/20 at 21:49, Xunlei Pang wrote: > On 2016/04/20 at 21:19, Peter Zijlstra wrote: >> On Thu, Apr 14, 2016 at 07:37:03PM +0800, Xunlei Pang wrote: >>> + /* Updated under pi_lock and rtmutex lock */ >>> struct rb_node *pi_waiters_leftmost; >>> + struct rb_node *pi_waiters_leftmost_copy; >>> struct task_struct *rt_mutex_get_top_task(struct task_struct *task) >>> { >>> + if (!task->pi_waiters_leftmost_copy) >>> return NULL; >>> >>> + return rb_entry(task->pi_waiters_leftmost_copy, >>> + struct rt_mutex_waiter, pi_tree_entry)->task; >>> } >> why ?! Why not keep a regular task_struct pointer and avoid this stuff? > I meant to make it semantically consistent, but I can change it since you > think task_struct is better.
What do you think this version of PATCH1 and PATCH2? If you are fine with it, I can sent it out as v4 separately, then we can focus on the issue in PATCH5. Thanks! > > Regards, > Xunlei