Re: [PATCH] xfs: introduce object readahead to log recovery

2013-07-28 Thread Zhi Yong Wu
On Mon, Jul 29, 2013 at 10:45 AM, Dave Chinner wrote: > On Mon, Jul 29, 2013 at 09:38:11AM +0800, Zhi Yong Wu wrote: >> By the way, how do you try the workload which has nothing but dirty >> dquote objects? > > Create quota limits for non-existent users. That will allocate the > dquots and dirty t

Re: [PATCH] xfs: introduce object readahead to log recovery

2013-07-28 Thread Dave Chinner
On Mon, Jul 29, 2013 at 09:38:11AM +0800, Zhi Yong Wu wrote: > By the way, how do you try the workload which has nothing but dirty > dquote objects? Create quota limits for non-existent users. That will allocate the dquots and dirty them. Or if you already have a few hundred thousand dquots, just

Re: [PATCH] xfs: introduce object readahead to log recovery

2013-07-28 Thread Zhi Yong Wu
On Fri, Jul 26, 2013 at 7:35 PM, Dave Chinner wrote: > On Fri, Jul 26, 2013 at 02:36:15PM +0800, Zhi Yong Wu wrote: >> Dave, >> >> All comments are good to me, and will be applied to next version, thanks a >> lot. >> >> On Fri, Jul 26, 2013 at 10:50 AM, Dave Chinner wrote: >> > On Thu, Jul 25, 2

Re: [PATCH] xfs: introduce object readahead to log recovery

2013-07-26 Thread Dave Chinner
On Fri, Jul 26, 2013 at 02:36:15PM +0800, Zhi Yong Wu wrote: > Dave, > > All comments are good to me, and will be applied to next version, thanks a > lot. > > On Fri, Jul 26, 2013 at 10:50 AM, Dave Chinner wrote: > > On Thu, Jul 25, 2013 at 04:23:39PM +0800, zwu.ker...@gmail.com wrote: > >> Fro

Re: [PATCH] xfs: introduce object readahead to log recovery

2013-07-25 Thread Zhi Yong Wu
Dave, All comments are good to me, and will be applied to next version, thanks a lot. On Fri, Jul 26, 2013 at 10:50 AM, Dave Chinner wrote: > On Thu, Jul 25, 2013 at 04:23:39PM +0800, zwu.ker...@gmail.com wrote: >> From: Zhi Yong Wu >> >> It can take a long time to run log recovery operation

Re: [PATCH] xfs: introduce object readahead to log recovery

2013-07-25 Thread Dave Chinner
On Thu, Jul 25, 2013 at 04:23:39PM +0800, zwu.ker...@gmail.com wrote: > From: Zhi Yong Wu > > It can take a long time to run log recovery operation because it is > single threaded and is bound by read latency. We can find that it took > most of the time to wait for the read IO to occur, so if o