I did not look closely at what was happening, but I experienced a
similar slowdown on 2.4.9-ac11. It took a while in my case, (several
weeks) as I have 1gb memory and don't have any massive files on my
system. I have dual PIII 1GHz CPUs and was downloading a 300mb file when
I started xpdf on a document that has lots of complex drawaings. The
download rate was 200-250kbps prior to starting xpdf and dropped to
20-40. If I stopped the download and restarted, it ran at 200-250 until
I ran xpdf again (that it was running did not seem important, but when I
displayed a new page and it ate a CPU). I thought it might be something
with xpdf because I was able to reproduce an xpdf slowdown by running
three copies at the same time.
However, now I am running 2.4.12-ac6 booted last night and I have seen
the slowdown in xpdf running just one. Last night after I re-booted, it
was running fine. The strange thing is that when this happens, the
system acts like it has no spare CPU because even switching to another
window takes a long time for the focus to change.
top reports:
12:21pm up 11:36, 16 users, load average: 0.30, 0.13, 0.09
140 processes: 138 sleeping, 2 running, 0 zombie, 0 stopped
CPU0 states: 0.5% user, 0.4% system, 0.0% nice, 98.1% idle
CPU1 states: 99.3% user, 0.2% system, 0.0% nice, 0.0% idle
Mem: 1028936K av, 817768K used, 211168K free, 4152K shrd, 282336K
buff
Swap: 2000880K av, 0K used, 2000880K free 189408K
cached
PID USER PRI NI SIZE RSS SHARE STAT %CPU %MEM TIME COMMAND
1117 root 18 0 75864 34M 5972 R 99.9 3.3 342:36 X
5491 linush 14 0 1016 1016 768 R 1.1 0.0 0:09 top
4859 linush 9 0 3544 3544 1616 S 0.5 0.3 0:04 xpdf
1 root 8 0 528 528 460 S 0.0 0.0 0:01 init
2 root 9 0 0 0 0 SW 0.0 0.0 0:00 keventd
Let me know if I can be of any assistance.
Linus
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/