On 01/15, Srivatsa Vaddagiri wrote: > > On Mon, Jan 15, 2007 at 03:54:01PM +0300, Oleg Nesterov wrote: > > > - singlethread_cpu needs to be hotplug safe (broken currently) > > > > Why? Could you explain? > > What if 'singlethread_cpu' dies?
Still can't understand you. Probably you missed what singlethread_cpu is. singlethread_cpu is just a "random" bit from cpu_possible_map. Single threaded workqueue is not bound to any cpu. We only need it to be sure that percpu_data.ptrs[singlethread_cpu] is populated by __percpu_alloc_mask(). > > > - Any reason why cpu_populated_map is not modified on CPU_DEAD? > > > > Because CPU_DEAD/CPU_UP_CANCELED doesn't wait for cwq->thread to exit. > > cpu_populated_map never shrinks, it only grows on CPU_UP_PREPARE. > > > > We can change this, but it needs some more code, and I am not sure > > we need it. Note that a "false" bit in cpu_populated_map only means > > that flush_work/flush_workqueue/destroy_workqueu will do lock/unlock > > of cwq->lock, nothing more. > > What abt __create_workqueue/schedule_on_each_cpu? As I said already __create_workqueue() needs a fix, schedule_on_each_cpu() is already broken, and should be fixed as well. > > > - I feel more comfortable if workqueue_cpu_callback were to take > > > workqueue_mutex in LOCK_ACQ and release it in LOCK_RELEASE > > > notifications. > > > > The whole purpose of this change to avoid this! > > I guess it depends on how __create_workqueue/schedule_on_each_cpu is > modified (whether we take/release lock upon LOCK_ACQ/LOCK_RELEASE) Sorry, can't understand this... > > > Finally, I wonder if these changes will be unnecessary if we move to > > > process freezer based hotplug locking ... > > > > This change ir not strictly necessary but imho make the code better and > > shrinks .text by 379 bytes. > > > > But I believe that freezer will change nothing for workqueue. We still > > need take_over_work(), and hacks like migrate_sequence. And no, CPU_DEAD > > can't just thaw cwq->thread which was bound to the dead CPU to complete > > kthread_stop(), we should thaw all processes. > > What abt stopping that thread in CPU_DOWN_PREPARE (before freezing > processes)? I understand that it may add to the latency, but compared to > the overall latency of process freezer, I suspect it may not be much. Srivatsa, why do you think this would be better? It add to the complexity! What do you mean by "stopping that thread" ? Kill it? - this is wrong. Make it TASK_STOPPED? - very untrivial and I can't see how this helps. Oleg. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
