tech-kern archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]
Re: serious performance regression in .41
On 21.05.2014 20:08, Christos Zoulas wrote:
> Hi,
>
I don't understand what do you mean.
Maybe it is useful for statistics, when "tar xzf src.tgz" on 6.99.40.
lockstat sync
Elapsed time: 10.04 seconds.
-- Adaptive mutex spin
Total% Count Time/ms Lock Caller
------ ------- --------- ---------------------- ------------------------------
64.56 2501 0.15 vnode_free_list_lock <all>
32.41 1247 0.07 vnode_free_list_lock vremfree+2f
31.67 1241 0.07 vnode_free_list_lock vrelel+3f0
0.42 11 0.00 vnode_free_list_lock getnewvnode+44
0.03 1 0.00 vnode_free_list_lock vnfree+2d
0.02 1 0.00 vnode_free_list_lock vdrain_thread+45
24.24 21 0.06 fffffe822f737240 uvm_fault_internal+865
5.03 58 0.01 uvm_pageqlock <all>
2.38 14 0.01 uvm_pageqlock genfs_getpages+a09
0.83 10 0.00 uvm_pageqlock ubc_fault+258
0.60 8 0.00 uvm_pageqlock uvm_aio_aiodone_pages+65
0.58 15 0.00 uvm_pageqlock uvm_fault_internal+95b
0.38 2 0.00 uvm_pageqlock ufs_balloc_range+18d
0.11 4 0.00 uvm_pageqlock uvm_fault_upper_enter.isra.5+11
0.09 3 0.00 uvm_pageqlock uvm_fault_lower_enter+189
0.04 1 0.00 uvm_pageqlock ubc_release+1ab
0.03 1 0.00 uvm_pageqlock uvm_fault_internal+557
3.40 66 0.01 bufcache_lock <all>
2.12 49 0.00 bufcache_lock vflushbuf+55
0.82 10 0.00 bufcache_lock brelse+19
0.16 1 0.00 bufcache_lock getblk+2f
0.14 3 0.00 bufcache_lock bwrite+e5
0.09 2 0.00 bufcache_lock vinvalbuf+7f
0.07 1 0.00 bufcache_lock breadn+3f
1.37 27 0.00 mntvnode_lock <all>
1.30 24 0.00 mntvnode_lock vfs_insmntque+2a
0.07 3 0.00 mntvnode_lock vfs_vnode_iterator_next+38
0.93 2 0.00 fffffe822f737b00 <all>
0.89 1 0.00 fffffe822f737b00 ubc_release+a3
0.04 1 0.00 fffffe822f737b00 ubc_alloc+6a
0.18 3 0.00 fffffe820ddb4a40 <all>
0.09 1 0.00 fffffe820ddb4a40 pipe_write+2e2
0.06 1 0.00 fffffe820ddb4a40 pipe_read+27
0.03 1 0.00 fffffe820ddb4a40 pipe_write+28
0.13 2 0.00 fffffe822f7370c0 uvm_fault_internal+865
0.12 1 0.00 fffffe820ddb46c0 uvm_fault_internal+371
0.04 1 0.00 fffffe820ddb49c0 pipe_read+27
-- Spin mutex spin
Total% Count Time/ms Lock Caller
------ ------- --------- ---------------------- ------------------------------
55.35 43 0.03 uvm_fpageqlock <all>
40.41 23 0.02 uvm_fpageqlock uvm_pageout_done+13
9.32 16 0.00 uvm_fpageqlock uvm_pagealloc_strat+c2
5.61 4 0.00 uvm_fpageqlock uvm_pageidlezero+1f6
28.63 8 0.01 fffffe822f738f80 turnstile_lookup+22
13.70 15 0.01 fffffe822f737bc0 <all>
13.00 11 0.01 fffffe822f737bc0 mi_switch+bd
0.58 3 0.00 fffffe822f737bc0 sleepq_remove+103
0.13 1 0.00 fffffe822f737bc0 sched_idle+112
1.20 2 0.00 fffffe822f737d00 <all>
1.07 1 0.00 fffffe822f737d00 sched_idle+11b
0.13 1 0.00 fffffe822f737d00 sleepq_remove+103
0.85 1 0.00 fffffe822f736c40 cv_wakeup_all+6f
0.14 1 0.00 fffffe822f738680 wakeup+38
0.13 1 0.00 pv_hash_locks+500 pmap_remove_pv+b1
-- RW lock sleep (writer)
Total% Count Time/ms Lock Caller
------ ------- --------- ---------------------- ------------------------------
79.60 1 1.52 fffffe821d257f28 genfs_lock+7b
18.59 1 0.35 fffffe820ed544b8 genfs_lock+7b
1.81 5 0.03 fffffe822fbc9dc8 vm_map_lock+62
-- RW lock spin
Total% Count Time/ms Lock Caller
------ ------- --------- ---------------------- ------------------------------
100.00 4037 7.02 fffffe822fbc9dc8 vm_map_lock+62
-- Kernel lock spin
Total% Count Time/ms Lock Caller
------ ------- --------- ---------------------- ------------------------------
100.00 706 1.11 kernel_lock <all>
84.94 423 0.94 kernel_lock bdev_strategy+53
12.28 230 0.14 kernel_lock biodone2+6f
1.43 49 0.02 kernel_lock intr_biglock_wrapper+12
1.26 2 0.01 kernel_lock frag6_slowtimo+1b
0.09 2 0.00 kernel_lock frag6_fasttimo+1a
Home |
Main Index |
Thread Index |
Old Index