On Wed, Oct 13, 2010 at 04:26:12PM +0800, Wu Fengguang wrote: > On Wed, Oct 13, 2010 at 11:07:33AM +0800, Dave Chinner wrote: > > On Tue, Oct 12, 2010 at 10:17:16AM -0400, Christoph Hellwig wrote: > > > Wu, what's the state of this series? It looks like we'll need it > > > rather sooner than later - try to get at least the preparations in > > > ASAP would be really helpful. > > > > Not ready in it's current form. This load (creating millions of 1 > > byte files in parallel): > > > > $ /usr/bin/time ./fs_mark -D 10000 -S0 -n 100000 -s 1 -L 63 \ > > > -d /mnt/scratch/0 -d /mnt/scratch/1 \ > > > -d /mnt/scratch/2 -d /mnt/scratch/3 \ > > > -d /mnt/scratch/4 -d /mnt/scratch/5 \ > > > -d /mnt/scratch/6 -d /mnt/scratch/7 > > > > Locks up all the fs_mark processes spinning in traces like the > > following and no further progress is made when the inode cache > > fills memory. > > I reproduced the problem on a 6G/8p 2-socket 11-disk box. > > The root cause is, pageout() is somehow called with low scan priority, > which deserves more investigation. > > The direct cause is, balance_dirty_pages() then keeps nr_dirty too low, > which can be improved easily by not pushing down the soft dirty limit > to less than 1-second worth of dirty pages. > > My test box has two nodes, and their memory usage are rather unbalanced: > (Dave, maybe you have NUMA setup too?) No, I'm running the test in a single node VM. FYI, I'm running the test on XFS (16TB 12 disk RAID0 stripe), using the mount options "inode64,nobarrier,logbsize=262144,delaylog". Cheers, Dave. -- Dave Chinner david@xxxxxxxxxxxxx -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxxx For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>