Hi.
Today again the same problem - however not that severe so I'm not sure if
below lockstat is representative.
# lockstat -kgIW sleep 5 | less
Profiling interrupt: 1968 events in 5.072 seconds (388 events/sec)
Count genr cuml rcnt nsec Hottest CPU+PIL Caller
-------------------------------------------------------------------------------
1560 79% ---- 0.00 2520 cpu[3] thread_start
1018 52% ---- 0.00 2069 cpu[2]+11 idle
851 43% ---- 0.00 2093 cpu[2]+11 disp_getwork
591 30% ---- 0.00 3647 cpu[1] metaslab_alloc
591 30% ---- 0.00 3647 cpu[1] metaslab_alloc_dva
591 30% ---- 0.00 3647 cpu[1] metaslab_group_alloc
404 21% ---- 0.00 3488 cpu[1] space_map_load
404 21% ---- 0.00 3488 cpu[1] metaslab_activate
382 19% ---- 0.00 3564 cpu[1] zio_dva_allocate
382 19% ---- 0.00 3564 cpu[1] zio_write_compress
382 19% ---- 0.00 3564 cpu[1] txg_sync_thread
382 19% ---- 0.00 3564 cpu[1] spa_sync
382 19% ---- 0.00 3564 cpu[1] dsl_pool_sync
382 19% ---- 0.00 3564 cpu[1] dsl_dataset_sync
382 19% ---- 0.00 3564 cpu[1] dnode_sync
382 19% ---- 0.00 3564 cpu[1] dmu_objset_sync
382 19% ---- 0.00 3564 cpu[1] dmu_objset_sync_dnodes
382 19% ---- 0.00 3564 cpu[1] dbuf_sync
382 19% ---- 0.00 3564 cpu[1] arc_write
363 18% ---- 0.00 3510 cpu[1] syscall_trap32
361 18% ---- 0.00 3514 cpu[1] nfssys
361 18% ---- 0.00 3514 cpu[1] svc_run
350 18% ---- 0.00 3532 cpu[1] svc_getreq
349 18% ---- 0.00 3533 cpu[1] common_dispatch
251 13% ---- 0.00 3813 cpu[1] avl_find
238 12% ---- 0.00 3320 cpu[1] space_map_remove
211 11% ---- 0.00 3798 cpu[1] zil_commit
209 11% ---- 0.00 3800 cpu[1] zio_alloc_blk
209 11% ---- 0.00 3800 cpu[1] zil_lwb_write_start
198 10% ---- 0.00 4256 cpu[1] metaslab_ff_alloc
196 10% ---- 0.00 3867 cpu[1] zfs_putpage
[...]
This message posted from opensolaris.org
_______________________________________________
zfs-discuss mailing list
zfs-discuss@opensolaris.org
http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
