You are not logged in.
Pages: 1
Topic closed
kernel: 4.11.2-1-ARCH
After bootup, iotop reports:
Total DISK READ : 0.00 B/s | Total DISK WRITE : 0.00 B/s
Actual DISK READ: 0.00 B/s | Actual DISK WRITE: 0.00 B/s
TID PRIO USER DISK READ DISK WRITE SWAPIN IO> COMMAND
7 be/4 root 0.00 B 0.00 B 0.00 % 99.99 % [ksoftirqd/0]
17 be/4 root 0.00 B 0.00 B 0.00 % 99.99 % [ksoftirqd/1]
After starting Xfce4, I normally get something similar to this:
Total DISK READ : 0.00 B/s | Total DISK WRITE : 0.00 B/s
Actual DISK READ: 0.00 B/s | Actual DISK WRITE: 0.00 B/s
TID PRIO USER DISK READ DISK WRITE SWAPIN IO> COMMAND
7 be/4 root 0.00 B 0.00 B 0.00 % 99.99 % [ksoftirqd/0]
17 be/4 root 0.00 B 0.00 B 0.00 % 99.99 % [ksoftirqd/1]
12836 be/4 tpfkanep 0.00 B 0.00 B 0.00 % 99.99 % opera --start-maxi~d [Chrome_IOThread]
12991 be/4 root 0.00 B 0.00 B 0.00 % 99.99 % python /usr/bin/iotop -ao
12935 be/4 tpfkanep 0.00 B 0.00 B 0.00 % 99.99 % opera --type=rende~1 [Chrome_ChildIOT]
11587 be/4 tpfkanep 0.00 B 0.00 B 0.00 % 99.99 % at-spi2-registryd --use-gnome-session
12807 be/4 tpfkanep 0.00 B 40.00 K 0.00 % 99.99 % firefox
12841 be/4 tpfkanep 0.00 B 0.00 B 0.00 % 99.99 % firefox [Timer]
12868 be/4 tpfkanep 0.00 B 8.00 K 0.00 % 99.99 % opera --start-maxi~d [BrowserBlocking]
12822 be/4 tpfkanep 0.00 B 0.00 B 0.00 % 99.99 % opera --start-maxi~ed [inotify_reader]
I use the deadline scheduler:
CONFIG_IOSCHED_DEADLINE=y
CONFIG_DEFAULT_DEADLINE=y
CONFIG_DEFAULT_IOSCHED="deadline"
CONFIG_MQ_IOSCHED_DEADLINE=y
noop and cfq also shows the same iotop results. The same behaviour is seen in zen and ck kernels.
top & htop does not show any unusual behaviour (ksoftirqd = 0% CPU) and no noticeable slowdowns in UI/commandline
4.10.13-1 does not have this behavior. What could be causing this?
Offline
Look at the numbers - there's no disk activity but apparently every process is on 99.9% I/O - looks like a misinterpretation/calculation by that script => are actually *all* processes on 99.9%
Online
Yeah it looks like all processes are reaching 99.99% over a period of time.
It normally starts with [ksoftirqd/0] and [ksoftirqd/1] at 99.9%. Then stuff like wvdial, xfwm, opera, [jfscommit], etc. all gradually go up to 99.9%. I first thought it was my custom kernels, but I just reinstalled the extra/linux-zen kernel from the repositories, but it exhibits the same behaviour as my custom zen kernel.
Is anyone else experiencing this...?
Offline
Been a week waiting on an email response from the developer of iotop...
I have resorted to running iotop without the --accumulated option, instead using iotop -o. It does not solve the reporting of 99.99% IO for the threads/processes (ksoftirqd, etc.) that does IO.
Offline
There is some minor bugs with Linux Kernel 4.11 and onwards.
The CPU frequency control does not work either.
Offline
That doesn't really add much to the thread and it is unclear if what you are saying is related to the thread or a general observation that might or might not be true. If you have an actual issue you'd like to discuss, please open a new thread about it.
https://wiki.archlinux.org/index.php/Co … bumping.22
Closing.
Offline
Pages: 1
Topic closed