Hi, guys:
I have a situation where an application makes small writes to a file.
In such a case, gam_server hogs the CPU. Here's a screen capture from
top(1):
top - 11:15:58 up 1:19, 4 users, load average: 1.88, 0.71, 0.34
Tasks: 82 total, 1 running, 81 sleeping, 0 stopped, 0 zombie
Cpu(s): 9.7% us, 78.0% sy, 0.0% ni, 6.0% id, 0.0% wa, 5.3% hi, 1.0% si
Mem: 515688k total, 457868k used, 57820k free, 36652k buffers
Swap: 554232k total, 0k used, 554232k free, 259560k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
3729 zaitcev 25 0 3412 1232 856 S 80.3 0.2 1:35.21 gam_server
3988 root 16 0 3576 404 348 S 4.3 0.1 0:06.84 cat
3313 root 15 0 63572 16m 6228 S 1.7 3.3 0:42.10 X
3450 zaitcev 16 0 38088 16m 10m S 1.7 3.3 0:10.03 nautilus
3454 zaitcev 15 0 36680 14m 8624 S 0.7 2.8 0:14.44 gnome-terminal
3444 zaitcev 16 0 13468 7328 6056 S 0.3 1.4 0:06.38 metacity
3497 zaitcev 16 0 21336 9940 7700 S 0.3 1.9 0:03.52 wnck-applet
1 root 16 0 1684 552 480 S 0.0 0.1 0:00.61 init
I run Fedora Core 3 with Nautilus enabled, and so I cannot just remove
Gamin and be happy. Does anyone have any ideas what do about this?
-- Pete
P.S. In case anyone is interested in details, this is what happened.
I was looking at some USB problems with usbmon and noticed that usbmon
drops about 0.7% packets. At first I thought it was something with SMM
BIOS or whatnot. But then I noticed that CPU is eaten by Gamin and
this apparently causes cat to starve sometimes. Usbmon produces a
stream of records which are captured with:
cat /sys/kernel/debug/usbmon/1t > x.dump
It is rather lightweight, without any massive copying of data by the
kernel. E.g. the USB traffic itself creates much more copying and
memory and cache usage.
