Unfortunately, I also do not have any numbers at that granularity to support that fact. But, I thought this is a logical thing to do..May be not impacting that much since it is already sharded.. Thanks & Regards Somnath -----Original Message----- From: 池信泽 [mailto:xmdxcxz@xxxxxxxxx] Sent: Friday, October 30, 2015 8:38 AM To: Somnath Roy Cc: ceph-devel@xxxxxxxxxxxxxxx Subject: Re: why we should use two Mutex in OSD ShardData? I do not see any improvement by moving to single mutex. I just fell puzzle why we use two mutex. But I also do not see any improvement using two mutex in my environment. Thanks for your explanation. 2015-10-30 22:59 GMT+08:00 Somnath Roy <Somnath.Roy@xxxxxxxxxxx>: > > Hi xinze, > This is mainly for reducing lock contention on a single mutex. Conditional wakeup on a mutex is expensive and that's why we wanted to make it separate from the mutex protecting Sharddata priority queue and pg_for_processing map. > Are you seeing any improvement by moving to single mutex ? > > Thanks & Regards > Somnath > > -----Original Message----- > From: ceph-devel-owner@xxxxxxxxxxxxxxx [mailto:ceph-devel-owner@xxxxxxxxxxxxxxx] On Behalf Of ??? > Sent: Friday, October 30, 2015 6:35 AM > To: ceph-devel@xxxxxxxxxxxxxxx > Subject: why we should use two Mutex in OSD ShardData? > > hi, all: > > There are two Mutex in ShardData, one is sdata_lock and the other one is sdata_op_ordering_lock. > > I wonder could we replace sdata_lock with sdata_op_ordering_lock? > > -- > Regards, > xinze > -- > To unsubscribe from this list: send the line "unsubscribe ceph-devel" > in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo > info at http://vger.kernel.org/majordomo-info.html -- Regards, xinze ��.n��������+%������w��{.n����z��u���ܨ}���Ơz�j:+v�����w����ޙ��&�)ߡ�a����z�ޗ���ݢj��w�f
