* Worker threads in D state since c5a94a618e7ac86 (workqueue: Use TASK_IDLE)
@ 2017-09-10 7:36 Markus Trippelsdorf
2017-09-11 13:11 ` Tejun Heo
0 siblings, 1 reply; 11+ messages in thread
From: Markus Trippelsdorf @ 2017-09-10 7:36 UTC (permalink / raw)
To: Peter Zijlstra; +Cc: Tejun Heo, linux-kernel
[-- Attachment #1: Type: text/plain, Size: 429 bytes --]
Since:
commit c5a94a618e7ac86b20f53d947f68d7cee6a4c6bc
Author: Peter Zijlstra <peterz@infradead.org>
Date: Wed Aug 23 13:58:44 2017 +0200
workqueue: Use TASK_IDLE
all worker threads are in D state. They all show up when using "magic
SysRq w". In htop they all have big fat red 'D' in the state column.
Is this really desirable?
I have attached the output of "ps aux" after boot and the SysRq-w
output.
--
Markus
[-- Attachment #2: ps_aux --]
[-- Type: text/plain, Size: 13591 bytes --]
USER PID %CPU %MEM VSZ RSS TTY STAT START TIME COMMAND
root 1 7.5 0.0 180 36 ? S 09:21 0:02 /sbin/minit
root 2 0.0 0.0 0 0 ? S 09:21 0:00 [kthreadd]
root 3 0.2 0.0 0 0 ? D 09:21 0:00 [kworker/0:0]
root 4 0.0 0.0 0 0 ? D< 09:21 0:00 [kworker/0:0H]
root 5 1.0 0.0 0 0 ? D 09:21 0:00 [kworker/u8:0]
root 6 0.0 0.0 0 0 ? D< 09:21 0:00 [mm_percpu_wq]
root 7 0.0 0.0 0 0 ? S 09:21 0:00 [ksoftirqd/0]
root 8 0.0 0.0 0 0 ? S 09:21 0:00 [rcu_sched]
root 9 0.0 0.0 0 0 ? S 09:21 0:00 [rcu_bh]
root 10 0.0 0.0 0 0 ? S 09:21 0:00 [migration/0]
root 11 0.0 0.0 0 0 ? S 09:21 0:00 [cpuhp/0]
root 12 0.0 0.0 0 0 ? S 09:21 0:00 [cpuhp/1]
root 13 0.0 0.0 0 0 ? S 09:21 0:00 [migration/1]
root 14 0.0 0.0 0 0 ? S 09:21 0:00 [ksoftirqd/1]
root 15 0.0 0.0 0 0 ? D 09:21 0:00 [kworker/1:0]
root 16 0.0 0.0 0 0 ? D< 09:21 0:00 [kworker/1:0H]
root 17 0.0 0.0 0 0 ? S 09:21 0:00 [cpuhp/2]
root 18 0.0 0.0 0 0 ? S 09:21 0:00 [migration/2]
root 19 0.0 0.0 0 0 ? S 09:21 0:00 [ksoftirqd/2]
root 20 0.2 0.0 0 0 ? D 09:21 0:00 [kworker/2:0]
root 21 0.0 0.0 0 0 ? D< 09:21 0:00 [kworker/2:0H]
root 22 0.0 0.0 0 0 ? S 09:21 0:00 [cpuhp/3]
root 23 0.0 0.0 0 0 ? S 09:21 0:00 [migration/3]
root 24 0.0 0.0 0 0 ? S 09:21 0:00 [ksoftirqd/3]
root 25 0.0 0.0 0 0 ? D 09:21 0:00 [kworker/3:0]
root 26 0.0 0.0 0 0 ? D< 09:21 0:00 [kworker/3:0H]
root 27 0.0 0.0 0 0 ? S 09:21 0:00 [kdevtmpfs]
root 28 0.0 0.0 0 0 ? D< 09:21 0:00 [netns]
root 29 0.1 0.0 0 0 ? D 09:21 0:00 [kworker/0:1]
root 30 0.0 0.0 0 0 ? D 09:21 0:00 [kworker/1:1]
root 31 0.0 0.0 0 0 ? D 09:21 0:00 [kworker/2:1]
root 32 0.0 0.0 0 0 ? S 09:21 0:00 [oom_reaper]
root 33 0.0 0.0 0 0 ? D< 09:21 0:00 [writeback]
root 34 0.0 0.0 0 0 ? S 09:21 0:00 [kcompactd0]
root 35 0.0 0.0 0 0 ? D< 09:21 0:00 [kblockd]
root 36 0.0 0.0 0 0 ? D< 09:21 0:00 [edac-poller]
root 37 0.1 0.0 0 0 ? D 09:21 0:00 [kworker/3:1]
root 38 0.0 0.0 0 0 ? S 09:21 0:00 [kswapd0]
root 39 0.0 0.0 0 0 ? D< 09:21 0:00 [ttm_swap]
root 40 0.0 0.0 0 0 ? S 09:21 0:00 [scsi_eh_0]
root 41 0.0 0.0 0 0 ? D< 09:21 0:00 [scsi_tmf_0]
root 42 0.0 0.0 0 0 ? S 09:21 0:00 [scsi_eh_1]
root 43 0.0 0.0 0 0 ? D< 09:21 0:00 [scsi_tmf_1]
root 44 0.0 0.0 0 0 ? S 09:21 0:00 [scsi_eh_2]
root 45 0.0 0.0 0 0 ? D< 09:21 0:00 [scsi_tmf_2]
root 46 0.0 0.0 0 0 ? S 09:21 0:00 [scsi_eh_3]
root 47 0.0 0.0 0 0 ? D< 09:21 0:00 [scsi_tmf_3]
root 48 0.0 0.0 0 0 ? S 09:21 0:00 [scsi_eh_4]
root 49 0.0 0.0 0 0 ? D< 09:21 0:00 [scsi_tmf_4]
root 50 0.0 0.0 0 0 ? S 09:21 0:00 [scsi_eh_5]
root 51 0.0 0.0 0 0 ? D< 09:21 0:00 [scsi_tmf_5]
root 52 0.5 0.0 0 0 ? D 09:21 0:00 [kworker/u8:1]
root 53 0.5 0.0 0 0 ? D 09:21 0:00 [kworker/u8:2]
root 54 0.5 0.0 0 0 ? D 09:21 0:00 [kworker/u8:3]
root 55 0.9 0.0 0 0 ? D 09:21 0:00 [kworker/u8:4]
root 56 2.4 0.0 0 0 ? D 09:21 0:00 [kworker/u8:5]
root 57 0.3 0.0 0 0 ? D 09:21 0:00 [kworker/u8:6]
root 58 0.0 0.0 0 0 ? D 09:21 0:00 [kworker/2:2]
root 59 0.0 0.0 0 0 ? D 09:21 0:00 [kworker/2:3]
root 60 0.8 0.0 0 0 ? D 09:21 0:00 [kworker/u8:7]
root 61 0.0 0.0 0 0 ? D< 09:21 0:00 [kworker/0:1H]
root 62 0.0 0.0 0 0 ? D< 09:21 0:00 [kworker/3:1H]
root 63 0.0 0.0 0 0 ? D< 09:21 0:00 [kworker/2:1H]
root 64 0.0 0.0 0 0 ? D 09:21 0:00 [kworker/0:2]
root 65 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-worker]
root 66 0.0 0.0 0 0 ? D< 09:21 0:00 [kworker/u9:0]
root 67 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-worker-hi]
root 68 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-delalloc]
root 69 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-flush_del]
root 70 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-cache]
root 71 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-submit]
root 72 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-fixup]
root 73 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-endio]
root 74 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-endio-met]
root 75 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-endio-met]
root 76 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-endio-rai]
root 77 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-endio-rep]
root 78 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-rmw]
root 79 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-endio-wri]
root 80 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-freespace]
root 81 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-delayed-m]
root 82 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-readahead]
root 83 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-qgroup-re]
root 84 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-extent-re]
root 85 0.0 0.0 0 0 ? D< 09:21 0:00 [kworker/1:1H]
root 86 0.0 0.0 0 0 ? S 09:21 0:00 [btrfs-cleaner]
root 87 0.0 0.0 0 0 ? S 09:21 0:00 [btrfs-transacti]
root 94 0.0 0.0 0 0 ? S 09:21 0:00 [jbd2/sdc2-8]
root 95 0.0 0.0 0 0 ? D< 09:21 0:00 [ext4-rsv-conver]
root 96 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-worker]
root 97 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-worker-hi]
root 98 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-delalloc]
root 99 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-flush_del]
root 100 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-cache]
root 101 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-submit]
root 102 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-fixup]
root 103 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-endio]
root 104 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-endio-met]
root 105 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-endio-met]
root 106 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-endio-rai]
root 107 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-endio-rep]
root 108 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-rmw]
root 109 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-endio-wri]
root 110 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-freespace]
root 111 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-delayed-m]
root 112 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-readahead]
root 113 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-qgroup-re]
root 114 0.0 0.0 0 0 ? D< 09:21 0:00 [btrfs-extent-re]
root 115 0.0 0.0 0 0 ? D 09:21 0:00 [kworker/3:2]
root 117 0.0 0.0 0 0 ? D 09:21 0:00 [kworker/1:2]
root 118 0.0 0.0 0 0 ? S 09:21 0:00 [btrfs-cleaner]
root 119 0.0 0.0 0 0 ? S 09:21 0:00 [btrfs-transacti]
root 122 0.7 0.0 12268 3580 ? Ss 09:21 0:00 /sbin/udevd --daemon
root 148 0.5 0.0 0 0 ? D 09:21 0:00 [kworker/u8:8]
root 151 0.0 0.0 8948 6276 ? Ss 09:21 0:00 syslog-ng --foreground
mpd 157 1.0 0.2 409956 21728 ? Ssl 09:21 0:00 mpd --no-daemon
root 158 0.1 0.0 73352 7868 ? SLsl 09:21 0:00 ntpd -n
root 159 0.1 0.0 9724 5136 ? Ss 09:21 0:00 cupsd -f
root 161 0.0 0.0 2700 120 ? Ss 09:21 0:00 fcron -b
root 163 0.6 0.0 0 0 ? D 09:21 0:00 [kworker/u8:9]
root 164 1.1 0.0 0 0 ? D 09:21 0:00 [kworker/u8:10]
root 166 0.3 0.0 0 0 ? D 09:21 0:00 [kworker/u8:11]
root 167 1.6 0.0 0 0 ? D 09:21 0:00 [kworker/u8:12]
root 194 0.2 0.0 3780 2776 tty1 Ss 09:21 0:00 /bin/login -f
root 195 0.0 0.0 5536 1712 tty2 Ss+ 09:21 0:00 agetty 38400 tty2 linux
markus 198 0.0 0.0 8252 3952 tty1 S 09:21 0:00 -zsh
markus 199 0.0 0.0 7072 3160 tty1 S+ 09:21 0:00 /bin/sh /usr/bin/startx
markus 215 0.0 0.0 4172 2668 tty1 S+ 09:21 0:00 xinit /home/markus/.xinitrc -- /etc/X11/xinit/xserverrc :0 -auth /home/markus/.serverauth.199
root 216 3.0 0.8 275060 66748 tty3 Ssl+ 09:21 0:00 /usr/bin/X -nolisten tcp :0 -auth /home/markus/.serverauth.199
root 217 0.7 0.0 0 0 ? D 09:21 0:00 [kworker/u8:13]
markus 221 0.2 0.2 1073767784 16816 tty1 S 09:21 0:00 /home/markus/.xmonad/xmonad-x86_64-linux
root 225 0.0 0.0 0 0 ? D< 09:21 0:00 [kworker/1:2H]
markus 230 0.0 0.0 4188 2212 ? S 09:21 0:00 xbindkeys
markus 247 0.1 0.0 9196 5096 tty1 S 09:21 0:00 xscreensaver
markus 251 0.0 0.0 7072 1544 tty1 S 09:21 0:00 /bin/sh /home/markus/.xinitrc
markus 254 0.0 0.0 4428 2540 tty1 S 09:21 0:00 unclutter -noevents -idle 1
root 256 0.0 0.0 0 0 ? D< 09:21 0:00 [kworker/u9:1]
markus 257 0.2 0.2 1073770164 19108 ? Ss 09:21 0:00 xmobar
markus 265 3.9 0.9 490256 81336 tty1 Sl 09:21 0:00 konsole -e tmux attach-session
markus 271 0.0 0.0 4780 2272 tty1 S 09:21 0:00 dbus-launch --autolaunch 500fac476577f8a0a3d748f800000095 --binary-syntax --close-stderr
markus 272 0.0 0.0 3292 196 ? Ss 09:21 0:00 /usr/bin/dbus-daemon --fork --print-pid 5 --print-address 7 --session
markus 275 0.0 0.0 9796 4080 pts/0 Ss+ 09:21 0:00 /usr/bin/tmux attach-session
markus 278 0.7 0.0 10492 4632 ? Ss 09:21 0:00 /usr/bin/tmux attach-session
root 279 0.0 0.0 6688 3384 pts/1 Ss 09:21 0:00 sudo su
root 280 0.0 0.0 6688 3332 pts/2 Ss 09:21 0:00 sudo su
markus 281 0.0 0.0 7072 3148 pts/3 Ss+ 09:21 0:00 sh /home/markus/multitail
markus 282 0.3 0.3 62116 29028 pts/4 Ss+ 09:21 0:00 ncmpcpp -h 192.168.1.2 -p 55555
markus 283 1.7 0.4 43556 37856 pts/5 Ss+ 09:21 0:00 mutt
markus 284 1.2 0.2 20748 16656 pts/6 Ss 09:21 0:00 zsh
markus 285 1.0 0.1 20208 15956 pts/7 Ss+ 09:21 0:00 zsh
markus 286 1.0 0.1 20208 15996 pts/8 Ss+ 09:21 0:00 zsh
markus 287 1.0 0.1 20168 15920 pts/9 Ss+ 09:21 0:00 -zsh
markus 288 1.0 0.1 20168 16016 pts/10 Ss+ 09:21 0:00 -zsh
root 300 0.0 0.0 0 0 ? D< 09:21 0:00 [kworker/3:2H]
root 301 0.0 0.0 6312 2868 pts/1 S 09:21 0:00 su
root 302 0.0 0.0 6312 2592 pts/2 S 09:21 0:00 su
root 306 0.4 0.0 11776 7572 pts/2 S+ 09:21 0:00 zsh
root 347 0.5 0.0 11776 7580 pts/1 S+ 09:21 0:00 zsh
root 351 0.0 0.0 6688 3208 pts/3 S+ 09:21 0:00 sudo multitail --no-mark-change -M 0 -csn /var/log/messages -cS kernel /var/log/kern.log
root 352 0.2 0.1 14944 10760 pts/3 S+ 09:21 0:00 multitail --no-mark-change -M 0 -csn /var/log/messages -cS kernel /var/log/kern.log
root 353 0.0 0.0 5436 620 pts/3 S 09:21 0:00 tail --follow=name -n 172 /var/log/messages
root 354 0.0 0.0 5436 700 pts/3 S 09:21 0:00 tail --follow=name -n 50 /var/log/kern.log
root 355 0.0 0.0 0 0 ? D< 09:21 0:00 [kworker/u9:2]
markus 362 0.0 0.0 6328 2400 pts/6 R+ 09:21 0:00 ps aux
[-- Attachment #3: SysRq_w --]
[-- Type: text/plain, Size: 25515 bytes --]
sysrq: SysRq : Show Blocked State
task PC stack pid father
kworker/0:0 D 0 3 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/0:0H D 0 4 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/u8:0 D 0 5 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
mm_percpu_wq D 0 6 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/1:0 D 0 15 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/1:0H D 0 16 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/2:0 D 0 20 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/2:0H D 0 21 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/3:0 D 0 25 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/3:0H D 0 26 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
netns D 0 28 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/0:1 D 0 29 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/1:1 D 0 30 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/2:1 D 0 31 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? pm_runtime_work+0x79/0x80
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
writeback D 0 33 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kblockd D 0 35 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
edac-poller D 0 36 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/3:1 D 0 37 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
ttm_swap D 0 39 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
scsi_tmf_0 D 0 41 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
scsi_tmf_1 D 0 43 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
scsi_tmf_2 D 0 45 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
scsi_tmf_3 D 0 47 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
scsi_tmf_4 D 0 49 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
scsi_tmf_5 D 0 51 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/u8:1 D 0 52 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/u8:2 D 0 53 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? edac_pci_workq_function+0x54/0x70
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/u8:3 D 0 54 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/u8:4 D 0 55 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/u8:5 D 0 56 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/u8:6 D 0 57 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/2:2 D 0 58 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/2:3 D 0 59 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/u8:7 D 0 60 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/0:1H D 0 61 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/3:1H D 0 62 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/2:1H D 0 63 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/0:2 D 0 64 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-worker D 0 65 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/u9:0 D 0 66 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-worker-hi D 0 67 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-delalloc D 0 68 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-flush_del D 0 69 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-cache D 0 70 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-submit D 0 71 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-fixup D 0 72 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-endio D 0 73 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-endio-met D 0 74 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-endio-met D 0 75 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-endio-rai D 0 76 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-endio-rep D 0 77 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-rmw D 0 78 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-endio-wri D 0 79 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-freespace D 0 80 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-delayed-m D 0 81 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-readahead D 0 82 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-qgroup-re D 0 83 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-extent-re D 0 84 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/1:1H D 0 85 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
ext4-rsv-conver D 0 95 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-worker D 0 96 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-worker-hi D 0 97 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-delalloc D 0 98 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-flush_del D 0 99 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? do_group_exit+0x35/0xa0
? ret_from_fork+0x22/0x30
btrfs-cache D 0 100 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-submit D 0 101 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-fixup D 0 102 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-endio D 0 103 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-endio-met D 0 104 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-endio-met D 0 105 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-endio-rai D 0 106 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-endio-rep D 0 107 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? do_group_exit+0x35/0xa0
? ret_from_fork+0x22/0x30
btrfs-rmw D 0 108 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-endio-wri D 0 109 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? do_group_exit+0x35/0xa0
? ret_from_fork+0x22/0x30
btrfs-freespace D 0 110 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-delayed-m D 0 111 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-readahead D 0 112 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-qgroup-re D 0 113 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-extent-re D 0 114 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? rescuer_thread+0x2f1/0x340
? __cancel_work+0x70/0x70
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/3:2 D 0 115 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/1:2 D 0 117 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
btrfs-transacti D 0 119 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? schedule_timeout+0x1a9/0x240
? __schedule+0x177/0x5b0
? io_schedule_timeout+0x1e/0x40
? wait_for_completion_io+0x92/0xf0
? do_task_dead+0x40/0x40
? write_all_supers+0x9d5/0xac0
? btrfs_commit_transaction+0x685/0x860
? start_transaction+0x94/0x3a0
? transaction_kthread+0x185/0x1a0
? btrfs_cleanup_transaction+0x460/0x460
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/u8:8 D 0 148 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/u8:9 D 0 163 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/u8:10 D 0 164 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? do_group_exit+0x35/0xa0
? ret_from_fork+0x22/0x30
kworker/u8:11 D 0 166 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? do_group_exit+0x35/0xa0
? ret_from_fork+0x22/0x30
kworker/u8:12 D 0 167 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/u8:13 D 0 217 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? kmem_cache_free+0xdf/0x100
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/1:2H D 0 225 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/u9:1 D 0 256 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/3:2H D 0 300 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? do_group_exit+0x35/0xa0
? ret_from_fork+0x22/0x30
kworker/u9:2 D 0 355 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/u9:3 D 0 419 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? ret_from_fork+0x22/0x30
kworker/u9:4 D 0 420 2 0x00000000
Call Trace:
? __schedule+0x16f/0x5b0
? schedule+0x2d/0x80
? worker_thread+0xaf/0x3f0
? process_one_work+0x340/0x340
? kthread+0x106/0x120
? __kthread_create_on_node+0x170/0x170
? do_group_exit+0x35/0xa0
? ret_from_fork+0x22/0x30
^ permalink raw reply [flat|nested] 11+ messages in thread* Re: Worker threads in D state since c5a94a618e7ac86 (workqueue: Use TASK_IDLE) 2017-09-10 7:36 Worker threads in D state since c5a94a618e7ac86 (workqueue: Use TASK_IDLE) Markus Trippelsdorf @ 2017-09-11 13:11 ` Tejun Heo 2017-09-11 14:21 ` Markus Trippelsdorf 0 siblings, 1 reply; 11+ messages in thread From: Tejun Heo @ 2017-09-11 13:11 UTC (permalink / raw) To: Markus Trippelsdorf; +Cc: Peter Zijlstra, linux-kernel Hello, On Sun, Sep 10, 2017 at 09:36:53AM +0200, Markus Trippelsdorf wrote: > Since: > > commit c5a94a618e7ac86b20f53d947f68d7cee6a4c6bc > Author: Peter Zijlstra <peterz@infradead.org> > Date: Wed Aug 23 13:58:44 2017 +0200 > > workqueue: Use TASK_IDLE > > > all worker threads are in D state. They all show up when using "magic > SysRq w". In htop they all have big fat red 'D' in the state column. > Is this really desirable? > > I have attached the output of "ps aux" after boot and the SysRq-w > output. Hmm.... looks like we better revert until we figure out how this should get presented in debugging facilities / to userspace. Peter? Thanks. -- tejun ^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: Worker threads in D state since c5a94a618e7ac86 (workqueue: Use TASK_IDLE) 2017-09-11 13:11 ` Tejun Heo @ 2017-09-11 14:21 ` Markus Trippelsdorf 2017-09-21 11:08 ` Markus Trippelsdorf 0 siblings, 1 reply; 11+ messages in thread From: Markus Trippelsdorf @ 2017-09-11 14:21 UTC (permalink / raw) To: Tejun Heo Cc: Peter Zijlstra, linux-kernel, Luis R. Rodriguez, Eric W. Biederman, Paul E. McKenney On 2017.09.11 at 06:11 -0700, Tejun Heo wrote: > Hello, > > On Sun, Sep 10, 2017 at 09:36:53AM +0200, Markus Trippelsdorf wrote: > > Since: > > > > commit c5a94a618e7ac86b20f53d947f68d7cee6a4c6bc > > Author: Peter Zijlstra <peterz@infradead.org> > > Date: Wed Aug 23 13:58:44 2017 +0200 > > > > workqueue: Use TASK_IDLE > > > > > > all worker threads are in D state. They all show up when using "magic > > SysRq w". In htop they all have big fat red 'D' in the state column. > > Is this really desirable? > > > > I have attached the output of "ps aux" after boot and the SysRq-w > > output. > > Hmm.... looks like we better revert until we figure out how this > should get presented in debugging facilities / to userspace. Peter? BTW rcu recently introduced the same issue: commit d5374226c3e444239e063f005dfb59cae4390db4 Author: Luis R. Rodriguez <mcgrof@kernel.org> Date: Tue Jun 20 14:45:47 2017 -0700 rcu: Use idle versions of swait to make idle-hack clear -- Markus ^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: Worker threads in D state since c5a94a618e7ac86 (workqueue: Use TASK_IDLE) 2017-09-11 14:21 ` Markus Trippelsdorf @ 2017-09-21 11:08 ` Markus Trippelsdorf 2017-09-21 12:30 ` Peter Zijlstra 0 siblings, 1 reply; 11+ messages in thread From: Markus Trippelsdorf @ 2017-09-21 11:08 UTC (permalink / raw) To: Tejun Heo Cc: Peter Zijlstra, linux-kernel, Luis R. Rodriguez, Eric W. Biederman, Paul E. McKenney, Linus Torvalds On 2017.09.11 at 16:21 +0200, Markus Trippelsdorf wrote: > On 2017.09.11 at 06:11 -0700, Tejun Heo wrote: > > Hello, > > > > On Sun, Sep 10, 2017 at 09:36:53AM +0200, Markus Trippelsdorf wrote: > > > Since: > > > > > > commit c5a94a618e7ac86b20f53d947f68d7cee6a4c6bc > > > Author: Peter Zijlstra <peterz@infradead.org> > > > Date: Wed Aug 23 13:58:44 2017 +0200 > > > > > > workqueue: Use TASK_IDLE > > > > > > > > > all worker threads are in D state. They all show up when using "magic > > > SysRq w". In htop they all have big fat red 'D' in the state column. > > > Is this really desirable? > > > > > > I have attached the output of "ps aux" after boot and the SysRq-w > > > output. > > > > Hmm.... looks like we better revert until we figure out how this > > should get presented in debugging facilities / to userspace. Peter? > > BTW rcu recently introduced the same issue: > > commit d5374226c3e444239e063f005dfb59cae4390db4 > Author: Luis R. Rodriguez <mcgrof@kernel.org> > Date: Tue Jun 20 14:45:47 2017 -0700 > > rcu: Use idle versions of swait to make idle-hack clear Ping? You may call it a cosmetic issue, but still it makes debugging much harder. Finding "real" blocked tasks is now like finding a needle in a haystack. -- Markus ^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: Worker threads in D state since c5a94a618e7ac86 (workqueue: Use TASK_IDLE) 2017-09-21 11:08 ` Markus Trippelsdorf @ 2017-09-21 12:30 ` Peter Zijlstra 2017-09-21 14:41 ` Markus Trippelsdorf 0 siblings, 1 reply; 11+ messages in thread From: Peter Zijlstra @ 2017-09-21 12:30 UTC (permalink / raw) To: Markus Trippelsdorf Cc: Tejun Heo, linux-kernel, Luis R. Rodriguez, Eric W. Biederman, Paul E. McKenney, Linus Torvalds On Thu, Sep 21, 2017 at 01:08:42PM +0200, Markus Trippelsdorf wrote: > On 2017.09.11 at 16:21 +0200, Markus Trippelsdorf wrote: > > On 2017.09.11 at 06:11 -0700, Tejun Heo wrote: > > > Hello, > > > > > > On Sun, Sep 10, 2017 at 09:36:53AM +0200, Markus Trippelsdorf wrote: > > > > Since: > > > > > > > > commit c5a94a618e7ac86b20f53d947f68d7cee6a4c6bc > > > > Author: Peter Zijlstra <peterz@infradead.org> > > > > Date: Wed Aug 23 13:58:44 2017 +0200 > > > > > > > > workqueue: Use TASK_IDLE > > > > > > > > > > > > all worker threads are in D state. They all show up when using "magic > > > > SysRq w". In htop they all have big fat red 'D' in the state column. > > > > Is this really desirable? > > > > > > > > I have attached the output of "ps aux" after boot and the SysRq-w > > > > output. > > > > > > Hmm.... looks like we better revert until we figure out how this > > > should get presented in debugging facilities / to userspace. Peter? > > > > BTW rcu recently introduced the same issue: > > > > commit d5374226c3e444239e063f005dfb59cae4390db4 > > Author: Luis R. Rodriguez <mcgrof@kernel.org> > > Date: Tue Jun 20 14:45:47 2017 -0700 > > > > rcu: Use idle versions of swait to make idle-hack clear > > Ping? > You may call it a cosmetic issue, but still it makes debugging much > harder. Finding "real" blocked tasks is now like finding a needle in a > haystack. Sorry, was out traveling. We can easily fix sysrq-w, not sure we can do much about htop (I've never seen it). I suppose we can try and make the state character not be D, is that really worth the trouble, or would it simply break htop if we were to return a new character? ^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: Worker threads in D state since c5a94a618e7ac86 (workqueue: Use TASK_IDLE) 2017-09-21 12:30 ` Peter Zijlstra @ 2017-09-21 14:41 ` Markus Trippelsdorf 2017-09-22 9:35 ` Markus Trippelsdorf 0 siblings, 1 reply; 11+ messages in thread From: Markus Trippelsdorf @ 2017-09-21 14:41 UTC (permalink / raw) To: Peter Zijlstra Cc: Tejun Heo, linux-kernel, Luis R. Rodriguez, Eric W. Biederman, Paul E. McKenney, Linus Torvalds On 2017.09.21 at 14:30 +0200, Peter Zijlstra wrote: > On Thu, Sep 21, 2017 at 01:08:42PM +0200, Markus Trippelsdorf wrote: > > On 2017.09.11 at 16:21 +0200, Markus Trippelsdorf wrote: > > > On 2017.09.11 at 06:11 -0700, Tejun Heo wrote: > > > > Hello, > > > > > > > > On Sun, Sep 10, 2017 at 09:36:53AM +0200, Markus Trippelsdorf wrote: > > > > > Since: > > > > > > > > > > commit c5a94a618e7ac86b20f53d947f68d7cee6a4c6bc > > > > > Author: Peter Zijlstra <peterz@infradead.org> > > > > > Date: Wed Aug 23 13:58:44 2017 +0200 > > > > > > > > > > workqueue: Use TASK_IDLE > > > > > > > > > > > > > > > all worker threads are in D state. They all show up when using "magic > > > > > SysRq w". In htop they all have big fat red 'D' in the state column. > > > > > Is this really desirable? > > > > > > > > > > I have attached the output of "ps aux" after boot and the SysRq-w > > > > > output. > > > > > > > > Hmm.... looks like we better revert until we figure out how this > > > > should get presented in debugging facilities / to userspace. Peter? > > > > > > BTW rcu recently introduced the same issue: > > > > > > commit d5374226c3e444239e063f005dfb59cae4390db4 > > > Author: Luis R. Rodriguez <mcgrof@kernel.org> > > > Date: Tue Jun 20 14:45:47 2017 -0700 > > > > > > rcu: Use idle versions of swait to make idle-hack clear > > > > Ping? > > You may call it a cosmetic issue, but still it makes debugging much > > harder. Finding "real" blocked tasks is now like finding a needle in a > > haystack. > > Sorry, was out traveling. We can easily fix sysrq-w, not sure we can do > much about htop (I've never seen it). > > I suppose we can try and make the state character not be D, is that > really worth the trouble, or would it simply break htop if we were to > return a new character? It seems to work. Simply returning "I (idle)" from get_task_state() in fs/proc/array.c when the state is TASK_IDLE does the trick. I've tested top, htop and ps. -- Markus ^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: Worker threads in D state since c5a94a618e7ac86 (workqueue: Use TASK_IDLE) 2017-09-21 14:41 ` Markus Trippelsdorf @ 2017-09-22 9:35 ` Markus Trippelsdorf 2017-09-22 11:54 ` [RFC][PATCH] sched: Cleanup task->state printing Peter Zijlstra 0 siblings, 1 reply; 11+ messages in thread From: Markus Trippelsdorf @ 2017-09-22 9:35 UTC (permalink / raw) To: Peter Zijlstra Cc: Tejun Heo, linux-kernel, Luis R. Rodriguez, Eric W. Biederman, Paul E. McKenney, Linus Torvalds On 2017.09.21 at 16:41 +0200, Markus Trippelsdorf wrote: > On 2017.09.21 at 14:30 +0200, Peter Zijlstra wrote: > > On Thu, Sep 21, 2017 at 01:08:42PM +0200, Markus Trippelsdorf wrote: > > > On 2017.09.11 at 16:21 +0200, Markus Trippelsdorf wrote: > > > > On 2017.09.11 at 06:11 -0700, Tejun Heo wrote: > > > > > Hello, > > > > > > > > > > On Sun, Sep 10, 2017 at 09:36:53AM +0200, Markus Trippelsdorf wrote: > > > > > > Since: > > > > > > > > > > > > commit c5a94a618e7ac86b20f53d947f68d7cee6a4c6bc > > > > > > Author: Peter Zijlstra <peterz@infradead.org> > > > > > > Date: Wed Aug 23 13:58:44 2017 +0200 > > > > > > > > > > > > workqueue: Use TASK_IDLE > > > > > > > > > > > > > > > > > > all worker threads are in D state. They all show up when using "magic > > > > > > SysRq w". In htop they all have big fat red 'D' in the state column. > > > > > > Is this really desirable? > > > > > > > > > > > > I have attached the output of "ps aux" after boot and the SysRq-w > > > > > > output. > > > > > > > > > > Hmm.... looks like we better revert until we figure out how this > > > > > should get presented in debugging facilities / to userspace. Peter? > > > > > > > > BTW rcu recently introduced the same issue: > > > > > > > > commit d5374226c3e444239e063f005dfb59cae4390db4 > > > > Author: Luis R. Rodriguez <mcgrof@kernel.org> > > > > Date: Tue Jun 20 14:45:47 2017 -0700 > > > > > > > > rcu: Use idle versions of swait to make idle-hack clear > > > > > > Ping? > > > You may call it a cosmetic issue, but still it makes debugging much > > > harder. Finding "real" blocked tasks is now like finding a needle in a > > > haystack. > > > > Sorry, was out traveling. We can easily fix sysrq-w, not sure we can do > > much about htop (I've never seen it). > > > > I suppose we can try and make the state character not be D, is that > > really worth the trouble, or would it simply break htop if we were to > > return a new character? > > It seems to work. Simply returning "I (idle)" from get_task_state() in > fs/proc/array.c when the state is TASK_IDLE does the trick. > I've tested top, htop and ps. So perhaps something like this: diff --git a/fs/proc/array.c b/fs/proc/array.c index 525157ca25cb..741687be3b0d 100644 --- a/fs/proc/array.c +++ b/fs/proc/array.c @@ -142,6 +142,9 @@ static inline const char *get_task_state(struct task_struct *tsk) BUILD_BUG_ON(1 + ilog2(TASK_REPORT) != ARRAY_SIZE(task_state_array)-1); + if (tsk->state == TASK_IDLE) + return "I (idle)"; + return task_state_array[fls(state)]; } diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 18a6966567da..83681990d3f9 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -5188,7 +5188,8 @@ void show_state_filter(unsigned long state_filter) */ touch_nmi_watchdog(); touch_all_softlockup_watchdogs(); - if (!state_filter || (p->state & state_filter)) + if (!state_filter || + (!(p->state == TASK_IDLE) && p->state & state_filter)) sched_show_task(p); } -- Markus ^ permalink raw reply related [flat|nested] 11+ messages in thread
* [RFC][PATCH] sched: Cleanup task->state printing 2017-09-22 9:35 ` Markus Trippelsdorf @ 2017-09-22 11:54 ` Peter Zijlstra 2017-09-22 12:40 ` Markus Trippelsdorf 2017-09-22 14:12 ` Steven Rostedt 0 siblings, 2 replies; 11+ messages in thread From: Peter Zijlstra @ 2017-09-22 11:54 UTC (permalink / raw) To: Markus Trippelsdorf Cc: Tejun Heo, linux-kernel, Luis R. Rodriguez, Eric W. Biederman, Paul E. McKenney, Linus Torvalds, Steven Rostedt, Thomas Gleixner, Ingo Molnar On Fri, Sep 22, 2017 at 11:35:33AM +0200, Markus Trippelsdorf wrote: > > It seems to work. Simply returning "I (idle)" from get_task_state() in > > fs/proc/array.c when the state is TASK_IDLE does the trick. > > I've tested top, htop and ps. I ended up with the below; there was quite a lot of inconsistent state printing around it seems. I should probably split this thing into a bunch of patches :/ Alongside an explicit idle state, this also exposes TASK_PARKED, although arguably we could map that to idle too. Opinions? --- fs/proc/array.c | 35 ++++++++++++----------- include/linux/sched.h | 58 +++++++++++++++++++++++---------------- include/trace/events/sched.h | 24 +++++++++++----- kernel/sched/core.c | 22 ++++++++++++++- kernel/sched/debug.c | 2 -- kernel/trace/trace_output.c | 21 ++++---------- kernel/trace/trace_sched_wakeup.c | 12 ++++---- 7 files changed, 103 insertions(+), 71 deletions(-) diff --git a/fs/proc/array.c b/fs/proc/array.c index 88c355574aa0..5a076854857f 100644 --- a/fs/proc/array.c +++ b/fs/proc/array.c @@ -118,28 +118,31 @@ static inline void task_name(struct seq_file *m, struct task_struct *p) * simple bit tests. */ static const char * const task_state_array[] = { - "R (running)", /* 0 */ - "S (sleeping)", /* 1 */ - "D (disk sleep)", /* 2 */ - "T (stopped)", /* 4 */ - "t (tracing stop)", /* 8 */ - "X (dead)", /* 16 */ - "Z (zombie)", /* 32 */ + /* states inside TASK_REPORT */ + + "R (running)", /* 0x00 */ + "S (sleeping)", /* 0x01 */ + "D (disk sleep)", /* 0x02 */ + "T (stopped)", /* 0x04 */ + "t (tracing stop)", /* 0x08 */ + "X (dead)", /* 0x10 */ + "Z (zombie)", /* 0x20 */ + "P (parked)", /* 0x40 */ + + /* extra states, beyond TASK_REPORT */ + + "I (idle)", /* 0x80 */ }; static inline const char *get_task_state(struct task_struct *tsk) { - unsigned int state = (tsk->state | tsk->exit_state) & TASK_REPORT; + unsigned int tsk_state = READ_ONCE(tsk->state); + unsigned int state = (tsk_state | tsk->exit_state) & TASK_REPORT; - /* - * Parked tasks do not run; they sit in __kthread_parkme(). - * Without this check, we would report them as running, which is - * clearly wrong, so we report them as sleeping instead. - */ - if (tsk->state == TASK_PARKED) - state = TASK_INTERRUPTIBLE; + if (tsk_state == TASK_IDLE) + state = TASK_REPORT_MAX; - BUILD_BUG_ON(1 + ilog2(TASK_REPORT) != ARRAY_SIZE(task_state_array)-1); + BUILD_BUG_ON(1 + ilog2(TASK_REPORT) != ARRAY_SIZE(task_state_array)-2); return task_state_array[fls(state)]; } diff --git a/include/linux/sched.h b/include/linux/sched.h index 68b38335d33c..7ae81efb17bd 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -65,25 +65,27 @@ struct task_group; */ /* Used in tsk->state: */ -#define TASK_RUNNING 0 -#define TASK_INTERRUPTIBLE 1 -#define TASK_UNINTERRUPTIBLE 2 -#define __TASK_STOPPED 4 -#define __TASK_TRACED 8 +#define TASK_RUNNING 0x0000 +#define TASK_INTERRUPTIBLE 0x0001 +#define TASK_UNINTERRUPTIBLE 0x0002 +#define __TASK_STOPPED 0x0004 +#define __TASK_TRACED 0x0008 /* Used in tsk->exit_state: */ -#define EXIT_DEAD 16 -#define EXIT_ZOMBIE 32 +#define EXIT_DEAD 0x0010 +#define EXIT_ZOMBIE 0x0020 #define EXIT_TRACE (EXIT_ZOMBIE | EXIT_DEAD) /* Used in tsk->state again: */ -#define TASK_DEAD 64 -#define TASK_WAKEKILL 128 -#define TASK_WAKING 256 -#define TASK_PARKED 512 -#define TASK_NOLOAD 1024 -#define TASK_NEW 2048 -#define TASK_STATE_MAX 4096 +#define TASK_PARKED 0x0040 +#define TASK_REPORT_MAX 0x0080 -#define TASK_STATE_TO_CHAR_STR "RSDTtXZxKWPNn" +/* Not in TASK_REPORT: */ +#define TASK_DEAD 0x0080 +#define TASK_WAKEKILL 0x0100 +#define TASK_WAKING 0x0200 +#define TASK_NOLOAD 0x0400 +#define TASK_NEW 0x0800 + +#define TASK_STATE_MAX 0x1000 /* Convenience macros for the sake of set_current_state: */ #define TASK_KILLABLE (TASK_WAKEKILL | TASK_UNINTERRUPTIBLE) @@ -96,10 +98,11 @@ struct task_group; #define TASK_NORMAL (TASK_INTERRUPTIBLE | TASK_UNINTERRUPTIBLE) #define TASK_ALL (TASK_NORMAL | __TASK_STOPPED | __TASK_TRACED) -/* get_task_state(): */ +/* task_state_to_char(), get_task_state(), trace_sched_switch() */ #define TASK_REPORT (TASK_RUNNING | TASK_INTERRUPTIBLE | \ TASK_UNINTERRUPTIBLE | __TASK_STOPPED | \ - __TASK_TRACED | EXIT_ZOMBIE | EXIT_DEAD) + __TASK_TRACED | EXIT_DEAD | EXIT_ZOMBIE | \ + TASK_PARKED) #define task_is_traced(task) ((task->state & __TASK_TRACED) != 0) @@ -1244,17 +1247,24 @@ static inline pid_t task_pgrp_nr(struct task_struct *tsk) return task_pgrp_nr_ns(tsk, &init_pid_ns); } -static inline char task_state_to_char(struct task_struct *task) +static inline char __task_state_to_char(unsigned int state) { - const char stat_nam[] = TASK_STATE_TO_CHAR_STR; - unsigned long state = task->state; + static const char state_char[] = "RSDTtXZPI"; - state = state ? __ffs(state) + 1 : 0; + BUILD_BUG_ON(1 + ilog2(TASK_REPORT) != sizeof(state_char) - 3); + + return state_char[fls(state)]; +} + +static inline char task_state_to_char(struct task_struct *task) +{ + unsigned int tsk_state = READ_ONCE(task->state); + unsigned int state = (tsk_state | task->exit_state) & TASK_REPORT; - /* Make sure the string lines up properly with the number of task states: */ - BUILD_BUG_ON(sizeof(TASK_STATE_TO_CHAR_STR)-1 != ilog2(TASK_STATE_MAX)+1); + if (tsk_state == TASK_IDLE) + state = TASK_REPORT_MAX; - return state < sizeof(stat_nam) - 1 ? stat_nam[state] : '?'; + return __task_state_to_char(state); } /** diff --git a/include/trace/events/sched.h b/include/trace/events/sched.h index ae1409ffe99a..af1858a01335 100644 --- a/include/trace/events/sched.h +++ b/include/trace/events/sched.h @@ -106,6 +106,8 @@ DEFINE_EVENT(sched_wakeup_template, sched_wakeup_new, #ifdef CREATE_TRACE_POINTS static inline long __trace_sched_switch_state(bool preempt, struct task_struct *p) { + unsigned int state = READ_ONCE(p->state); + #ifdef CONFIG_SCHED_DEBUG BUG_ON(p != current); #endif /* CONFIG_SCHED_DEBUG */ @@ -114,10 +116,18 @@ static inline long __trace_sched_switch_state(bool preempt, struct task_struct * * Preemption ignores task state, therefore preempted tasks are always * RUNNING (we will not have dequeued if state != RUNNING). */ - return preempt ? TASK_RUNNING | TASK_STATE_MAX : p->state; + if (preempt) + return TASK_REPORT_MAX << 1; + + if (state == TASK_IDLE) + return TASK_REPORT_MAX; + + return (state | p->exit_state) & TASK_REPORT; } #endif /* CREATE_TRACE_POINTS */ +#define TRACE_REPORT_MASK ((TASK_REPORT_MAX << 1) - 1) + /* * Tracepoint for task switches, performed by the scheduler: */ @@ -152,13 +162,13 @@ TRACE_EVENT(sched_switch, TP_printk("prev_comm=%s prev_pid=%d prev_prio=%d prev_state=%s%s ==> next_comm=%s next_pid=%d next_prio=%d", __entry->prev_comm, __entry->prev_pid, __entry->prev_prio, - __entry->prev_state & (TASK_STATE_MAX-1) ? - __print_flags(__entry->prev_state & (TASK_STATE_MAX-1), "|", + + __entry->prev_state & TRACE_REPORT_MASK ? + __print_flags(__entry->prev_state & TRACE_REPORT_MASK, "|", { 1, "S"} , { 2, "D" }, { 4, "T" }, { 8, "t" }, - { 16, "Z" }, { 32, "X" }, { 64, "x" }, - { 128, "K" }, { 256, "W" }, { 512, "P" }, - { 1024, "N" }) : "R", - __entry->prev_state & TASK_STATE_MAX ? "+" : "", + { 16, "X" }, { 32, "Z" }, { 64, "P" }, + { 128, "I" }) : "R", + __entry->prev_state & (TASK_REPORT_MAX << 1) ? "+" : "", __entry->next_comm, __entry->next_pid, __entry->next_prio) ); diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 703f5831738e..431e2d6c709e 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -5164,6 +5164,26 @@ void sched_show_task(struct task_struct *p) put_task_stack(p); } +static inline bool state_filter_match(unsigned long state_filter, struct task_struct *p) +{ + /* no filter, everything matches */ + if (!state_filter) + return true; + + /* filter, but doesn't match */ + if (!(p->state & state_filter)) + return false; + + /* + * When looking for TASK_UNINTERRUPTIBLE, skip TASK_IDLE, but allow + * TASK_KILLABLE. + */ + if (state_filter == TASK_UNINTERRUPTIBLE && p->state == TASK_IDLE) + return false; + + return true; +} + void show_state_filter(unsigned long state_filter) { struct task_struct *g, *p; @@ -5186,7 +5206,7 @@ void show_state_filter(unsigned long state_filter) */ touch_nmi_watchdog(); touch_all_softlockup_watchdogs(); - if (!state_filter || (p->state & state_filter)) + if (state_filter_match(state_filter, p)) sched_show_task(p); } diff --git a/kernel/sched/debug.c b/kernel/sched/debug.c index 4a23bbc3111b..244619e402cc 100644 --- a/kernel/sched/debug.c +++ b/kernel/sched/debug.c @@ -461,8 +461,6 @@ static char *task_group_path(struct task_group *tg) } #endif -static const char stat_nam[] = TASK_STATE_TO_CHAR_STR; - static void print_task(struct seq_file *m, struct rq *rq, struct task_struct *p) { diff --git a/kernel/trace/trace_output.c b/kernel/trace/trace_output.c index bac629af2285..c738e764e2a5 100644 --- a/kernel/trace/trace_output.c +++ b/kernel/trace/trace_output.c @@ -656,15 +656,6 @@ int trace_print_lat_context(struct trace_iterator *iter) return !trace_seq_has_overflowed(s); } -static const char state_to_char[] = TASK_STATE_TO_CHAR_STR; - -static int task_state_char(unsigned long state) -{ - int bit = state ? __ffs(state) + 1 : 0; - - return bit < sizeof(state_to_char) - 1 ? state_to_char[bit] : '?'; -} - /** * ftrace_find_event - find a registered event * @type: the type of event to look for @@ -930,8 +921,8 @@ static enum print_line_t trace_ctxwake_print(struct trace_iterator *iter, trace_assign_type(field, iter->ent); - T = task_state_char(field->next_state); - S = task_state_char(field->prev_state); + T = __task_state_to_char(field->next_state); + S = __task_state_to_char(field->prev_state); trace_find_cmdline(field->next_pid, comm); trace_seq_printf(&iter->seq, " %5d:%3d:%c %s [%03d] %5d:%3d:%c %s\n", @@ -966,8 +957,8 @@ static int trace_ctxwake_raw(struct trace_iterator *iter, char S) trace_assign_type(field, iter->ent); if (!S) - S = task_state_char(field->prev_state); - T = task_state_char(field->next_state); + S = __task_state_to_char(field->prev_state); + T = __task_state_to_char(field->next_state); trace_seq_printf(&iter->seq, "%d %d %c %d %d %d %c\n", field->prev_pid, field->prev_prio, @@ -1002,8 +993,8 @@ static int trace_ctxwake_hex(struct trace_iterator *iter, char S) trace_assign_type(field, iter->ent); if (!S) - S = task_state_char(field->prev_state); - T = task_state_char(field->next_state); + S = __task_state_to_char(field->prev_state); + T = __task_state_to_char(field->next_state); SEQ_PUT_HEX_FIELD(s, field->prev_pid); SEQ_PUT_HEX_FIELD(s, field->prev_prio); diff --git a/kernel/trace/trace_sched_wakeup.c b/kernel/trace/trace_sched_wakeup.c index ddec53b67646..b14caa0afd35 100644 --- a/kernel/trace/trace_sched_wakeup.c +++ b/kernel/trace/trace_sched_wakeup.c @@ -380,7 +380,7 @@ probe_wakeup_migrate_task(void *ignore, struct task_struct *task, int cpu) } static void -tracing_sched_switch_trace(struct trace_array *tr, +tracing_sched_switch_trace(bool preempt, struct trace_array *tr, struct task_struct *prev, struct task_struct *next, unsigned long flags, int pc) @@ -397,10 +397,10 @@ tracing_sched_switch_trace(struct trace_array *tr, entry = ring_buffer_event_data(event); entry->prev_pid = prev->pid; entry->prev_prio = prev->prio; - entry->prev_state = prev->state; + entry->prev_state = __trace_sched_switch_state(preempt, prev); entry->next_pid = next->pid; entry->next_prio = next->prio; - entry->next_state = next->state; + entry->next_state = __trace_sched_switch_state(false, next); entry->next_cpu = task_cpu(next); if (!call_filter_check_discard(call, entry, buffer, event)) @@ -425,10 +425,10 @@ tracing_sched_wakeup_trace(struct trace_array *tr, entry = ring_buffer_event_data(event); entry->prev_pid = curr->pid; entry->prev_prio = curr->prio; - entry->prev_state = curr->state; + entry->prev_state = __trace_sched_switch_state(false, curr); entry->next_pid = wakee->pid; entry->next_prio = wakee->prio; - entry->next_state = wakee->state; + entry->next_state = __trace_sched_switch_state(false, wakee); entry->next_cpu = task_cpu(wakee); if (!call_filter_check_discard(call, entry, buffer, event)) @@ -482,7 +482,7 @@ probe_wakeup_sched_switch(void *ignore, bool preempt, data = per_cpu_ptr(wakeup_trace->trace_buffer.data, wakeup_cpu); __trace_function(wakeup_trace, CALLER_ADDR0, CALLER_ADDR1, flags, pc); - tracing_sched_switch_trace(wakeup_trace, prev, next, flags, pc); + tracing_sched_switch_trace(preempt, wakeup_trace, prev, next, flags, pc); T0 = data->preempt_timestamp; T1 = ftrace_now(cpu); ^ permalink raw reply related [flat|nested] 11+ messages in thread
* Re: [RFC][PATCH] sched: Cleanup task->state printing 2017-09-22 11:54 ` [RFC][PATCH] sched: Cleanup task->state printing Peter Zijlstra @ 2017-09-22 12:40 ` Markus Trippelsdorf 2017-09-22 14:12 ` Steven Rostedt 1 sibling, 0 replies; 11+ messages in thread From: Markus Trippelsdorf @ 2017-09-22 12:40 UTC (permalink / raw) To: Peter Zijlstra Cc: Tejun Heo, linux-kernel, Luis R. Rodriguez, Eric W. Biederman, Paul E. McKenney, Linus Torvalds, Steven Rostedt, Thomas Gleixner, Ingo Molnar On 2017.09.22 at 13:54 +0200, Peter Zijlstra wrote: > On Fri, Sep 22, 2017 at 11:35:33AM +0200, Markus Trippelsdorf wrote: > > > It seems to work. Simply returning "I (idle)" from get_task_state() in > > > fs/proc/array.c when the state is TASK_IDLE does the trick. > > > I've tested top, htop and ps. > > I ended up with the below; there was quite a lot of inconsistent state > printing around it seems. > > I should probably split this thing into a bunch of patches :/ > > Alongside an explicit idle state, this also exposes TASK_PARKED, > although arguably we could map that to idle too. Opinions? Looks good to me and works as expected. Many thanks. -- Markus ^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [RFC][PATCH] sched: Cleanup task->state printing 2017-09-22 11:54 ` [RFC][PATCH] sched: Cleanup task->state printing Peter Zijlstra 2017-09-22 12:40 ` Markus Trippelsdorf @ 2017-09-22 14:12 ` Steven Rostedt 2017-09-22 15:56 ` Peter Zijlstra 1 sibling, 1 reply; 11+ messages in thread From: Steven Rostedt @ 2017-09-22 14:12 UTC (permalink / raw) To: Peter Zijlstra Cc: Markus Trippelsdorf, Tejun Heo, linux-kernel, Luis R. Rodriguez, Eric W. Biederman, Paul E. McKenney, Linus Torvalds, Thomas Gleixner, Ingo Molnar On Fri, 22 Sep 2017 13:54:30 +0200 Peter Zijlstra <peterz@infradead.org> wrote: > I should probably split this thing into a bunch of patches :/ Yes please. Convert form dec to hex in one patch and one patch only. Because I'm not sure if you meant to change numbers or not. > /* Used in tsk->state again: */ > -#define TASK_DEAD 64 > -#define TASK_WAKEKILL 128 > -#define TASK_WAKING 256 > -#define TASK_PARKED 512 > -#define TASK_NOLOAD 1024 > -#define TASK_NEW 2048 > -#define TASK_STATE_MAX 4096 > +#define TASK_PARKED 0x0040 > +#define TASK_REPORT_MAX 0x0080 > > -#define TASK_STATE_TO_CHAR_STR "RSDTtXZxKWPNn" > +/* Not in TASK_REPORT: */ > +#define TASK_DEAD 0x0080 TASK_DEAD went from 64 to 128 (0x40 to 0x80) As well as all the defines below that. Was this on purpose? > +#define TASK_WAKEKILL 0x0100 > +#define TASK_WAKING 0x0200 > +#define TASK_NOLOAD 0x0400 > +#define TASK_NEW 0x0800 > + > +#define TASK_STATE_MAX 0x1000 -- Steve ^ permalink raw reply [flat|nested] 11+ messages in thread
* Re: [RFC][PATCH] sched: Cleanup task->state printing 2017-09-22 14:12 ` Steven Rostedt @ 2017-09-22 15:56 ` Peter Zijlstra 0 siblings, 0 replies; 11+ messages in thread From: Peter Zijlstra @ 2017-09-22 15:56 UTC (permalink / raw) To: Steven Rostedt Cc: Markus Trippelsdorf, Tejun Heo, linux-kernel, Luis R. Rodriguez, Eric W. Biederman, Paul E. McKenney, Linus Torvalds, Thomas Gleixner, Ingo Molnar On Fri, Sep 22, 2017 at 10:12:45AM -0400, Steven Rostedt wrote: > On Fri, 22 Sep 2017 13:54:30 +0200 > Peter Zijlstra <peterz@infradead.org> wrote: > > > I should probably split this thing into a bunch of patches :/ > > Yes please. Convert form dec to hex in one patch and one patch only. Yeah, was already on it, did more cleanups too. > Because I'm not sure if you meant to change numbers or not. > > > > /* Used in tsk->state again: */ > > -#define TASK_DEAD 64 > > -#define TASK_WAKEKILL 128 > > -#define TASK_WAKING 256 > > -#define TASK_PARKED 512 > > -#define TASK_NOLOAD 1024 > > -#define TASK_NEW 2048 > > -#define TASK_STATE_MAX 4096 > > +#define TASK_PARKED 0x0040 > > +#define TASK_REPORT_MAX 0x0080 > > > > -#define TASK_STATE_TO_CHAR_STR "RSDTtXZxKWPNn" > > +/* Not in TASK_REPORT: */ > > +#define TASK_DEAD 0x0080 > > TASK_DEAD went from 64 to 128 (0x40 to 0x80) > > As well as all the defines below that. Was this on purpose? Yes, was on purpose. I moved TASK_PARKED up, such that I could include it in the TASK_REPORT mask and keep that contiguous. ^ permalink raw reply [flat|nested] 11+ messages in thread
end of thread, other threads:[~2017-09-22 15:56 UTC | newest] Thread overview: 11+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2017-09-10 7:36 Worker threads in D state since c5a94a618e7ac86 (workqueue: Use TASK_IDLE) Markus Trippelsdorf 2017-09-11 13:11 ` Tejun Heo 2017-09-11 14:21 ` Markus Trippelsdorf 2017-09-21 11:08 ` Markus Trippelsdorf 2017-09-21 12:30 ` Peter Zijlstra 2017-09-21 14:41 ` Markus Trippelsdorf 2017-09-22 9:35 ` Markus Trippelsdorf 2017-09-22 11:54 ` [RFC][PATCH] sched: Cleanup task->state printing Peter Zijlstra 2017-09-22 12:40 ` Markus Trippelsdorf 2017-09-22 14:12 ` Steven Rostedt 2017-09-22 15:56 ` Peter Zijlstra
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).