Re: RCU_USER_QS traces.

2012-10-05 Thread Dave Jones
On Fri, Oct 05, 2012 at 02:36:19PM +0200, Frederic Weisbecker wrote: > On Thu, Oct 04, 2012 at 10:41:06AM -0400, Dave Jones wrote: > > > We have CONFIG_RCU_USER_QS that is a specific case. It's an > > intermediate state > > > before we implement a true CONFIG_NO_HZ_FULL. But the option is

Re: RCU_USER_QS traces.

2012-10-05 Thread Frederic Weisbecker
On Thu, Oct 04, 2012 at 10:41:06AM -0400, Dave Jones wrote: > > We have CONFIG_RCU_USER_QS that is a specific case. It's an intermediate > state > > before we implement a true CONFIG_NO_HZ_FULL. But the option is useless on > its > > own for users. Worse, it introduces a real overhead. OTOH

Re: RCU_USER_QS traces.

2012-10-05 Thread Frederic Weisbecker
On Thu, Oct 04, 2012 at 10:41:06AM -0400, Dave Jones wrote: We have CONFIG_RCU_USER_QS that is a specific case. It's an intermediate state before we implement a true CONFIG_NO_HZ_FULL. But the option is useless on its own for users. Worse, it introduces a real overhead. OTOH we want

Re: RCU_USER_QS traces.

2012-10-05 Thread Dave Jones
On Fri, Oct 05, 2012 at 02:36:19PM +0200, Frederic Weisbecker wrote: On Thu, Oct 04, 2012 at 10:41:06AM -0400, Dave Jones wrote: We have CONFIG_RCU_USER_QS that is a specific case. It's an intermediate state before we implement a true CONFIG_NO_HZ_FULL. But the option is useless

Re: RCU_USER_QS traces.

2012-10-04 Thread Paul E. McKenney
On Thu, Oct 04, 2012 at 10:41:06AM -0400, Dave Jones wrote: > > We have CONFIG_RCU_USER_QS that is a specific case. It's an intermediate > state > > before we implement a true CONFIG_NO_HZ_FULL. But the option is useless on > its > > own for users. Worse, it introduces a real overhead. OTOH

RCU_USER_QS traces.

2012-10-04 Thread Dave Jones
> We have CONFIG_RCU_USER_QS that is a specific case. It's an intermediate > state > before we implement a true CONFIG_NO_HZ_FULL. But the option is useless on > its > own for users. Worse, it introduces a real overhead. OTOH we want it to be > upstream > to make the development of full

RCU_USER_QS traces.

2012-10-04 Thread Dave Jones
We have CONFIG_RCU_USER_QS that is a specific case. It's an intermediate state before we implement a true CONFIG_NO_HZ_FULL. But the option is useless on its own for users. Worse, it introduces a real overhead. OTOH we want it to be upstream to make the development of full

Re: RCU_USER_QS traces.

2012-10-04 Thread Paul E. McKenney
On Thu, Oct 04, 2012 at 10:41:06AM -0400, Dave Jones wrote: We have CONFIG_RCU_USER_QS that is a specific case. It's an intermediate state before we implement a true CONFIG_NO_HZ_FULL. But the option is useless on its own for users. Worse, it introduces a real overhead. OTOH we want