I can confirm similar behavior on my system as well. I'm running
2.3.99-pre7. I was doing a diff of two kernel trees and kswapd started
eating CPU like no tomorrow.
FYI, I have a AMD K6-II 400Mhz single processor system with one 6Gig
UDMA/33 hard drive on a Tyan 1598S motherboard, with 256Megs of ram.
System is a RedHat 6.1 installation with the new kernel.
Also, top and free both show that I'm not consuming any of my swap. That
was the part I found really odd. Also, after doing the diff once, I did
it again, and the behavior was exactly the same even though much of the
information should have been cached.
Matt Dharm
On Sat, 29 Apr 2000, Richard Torkar wrote:
>
> In short:
> Whenever I use the program vcdgear (which extracts mpg files from a
> bin/cue file) my kswapd goes sky high... See below. It stays between 60 -
> 80% CPU. Directly when vcdgear stops kswapd stops as well.
>
> There is extensive hd i/o.
> The hd reads and writes to the same partition for several minutes when
> using vcdgear, and that is of course normal. But is the kswapd usuage
> normal?
>
> Kernel 2.3.99-pre6
> Hardware: BP6, dual celeron 500, 128MB RAM
> I have the HD on a HPT366 interface (IBM-DJNA-371800, 17GB)
>
> ***
> Output of ver_linux:
> Linux ***.***.***.se 2.3.99-pre6 #3 SMP Fri Apr 28 22:06:17 CEST 2000 i686 unknown
> Kernel modules 2.3.10-pre1
> Gnu C egcs-2.91.66
> Binutils 2.9.5.0.22
> Linux C Library 2.1.3
> Dynamic linker ldd (GNU libc) 2.1.3
> Procps 2.0.6
> Mount 2.10f
> Net-tools 1.54
> Console-tools 0.3.3
> Sh-utils 2.0
> Modules Loaded 3c509 opl3 sb sb_lib uart401 sound soundcore
> ***
>
> 3:28pm up 17:08, 3 users, load average: 2.57, 2.18, 1.81
> 47 processes: 44 sleeping, 2 running, 1 zombie, 0 stopped
> CPU states: 1.4% user, 6.0% system, 0.0% nice, 22.8% idle
> Mem: 126548K av, 124260K used, 2288K free, 0K shrd, 700K buff
> Swap: 136512K av, 4092K used, 132420K free 94584K cached
>
> PID USER PRI NI SIZE RSS SHARE STAT LIB %CPU %MEM TIME COMMAND
> 2 root 18 0 0 0 0 SW 0 69.8 0.0 66:37 kswapd
> 513 root 12 0 111M 110M 1036 R 0 53.2 89.3 24:43 X
> 3441 root 15 0 512 512 408 D 0 53.2 0.4 1:49 vcdgear
> 3517 toor 6 0 840 840 652 R 0 9.9 0.6 0:00 top
> 617 toor 2 0 324 220 160 S 0 3.3 0.1 8:22 wmifs
> [...........................................................................]
>
>
>
> I have not tried any other kernel. If you want me to try diffrent kernels
> please tell me so.
>
> This might be a well-known 'feature' I have no idea.
>
>
>
> Richard
>
> ---
> Toor: Linux version 2.3.99-pre6 #1 SMP i686 (2244.61 BogoMips)
> Linux _is_ user-friendly. It's just selective about its friends.
> *** PGP4PINE 3.3 (20000103) by Marcin Marszalek <azzie@staszic.waw.pl>
> *** Decryption log - Sat Apr 29 18:52:12 PDT 2000
>
> Pretty Good Privacy(tm) Version 6.5.2
> (c) 1999 Network Associates Inc.
> Uses the RSAREF(tm) Toolkit, which is copyright RSA Data Security, Inc.
> Export of this software may be restricted by the U.S. government.
>
> File is signed. signature not checked.
> Signature made 2000/04/29 13:42 GMT
> key does not meet validity threshold.
>
> WARNING: Because this public key is not certified with a trusted
> signature, it is not known with high confidence that this public key
> actually belongs to: "(KeyID: 0x8A36DD1B)".
>
> *** End of decryption log
>
-- Matthew Dharm Home: mdharm@one-eyed-alien.net Senior Engineer, QCP Inc. Work: mdharm@qualcomm.comG: Let me guess, you started on the 'net with AOL, right? C: WOW! d00d! U r leet! -- Greg and Customer User Friendly, 2/12/1999
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sun Apr 30 2000 - 21:00:17 EST