Hello,

I have been testing an older mainboard and cpu, an ASUS SP3G and an Intel
486DX4/100.  With an AMD DX4/100 I was having some problems with certain
PCI options in the BIOS.  With the Intel everything appears to be fine.
My method of testing is to run about 10 copies of "eatmem", a small
utility someone had posted way back that grabs memory and randomly
writes to pages.  This abuse would consistently panic my AMD cpu unless
I disabled several default options in the BIOS.  Anyway, my machine
has been running 1.4 with 13 copies of eatmem, each with 10MB allocated.
It has been running for about 24 hours now, and top shows me the 
following interesting information.  Note how 2372 and 2376 are gobbling
up the RAM/CPU, and the rest appear starved?  Is this expected behavior?
I would expect the scheduler provide more equal treatment of these processes.
(other than the 5 niced to 20)  All 13 processes were started up at
about the same time. 


load averages: 13.40, 13.11, 12.83                                     14:09:03
32 processes:  8 running, 24 sleeping
CPU states: 83.4% user,  0.0% nice, 16.6% system,  0.0% interrupt,  0.0% idle
Memory: 20M Act 10M Inact 216K Wired 1144K Free 110M Swap 18M Swap free

  PID USERNAME PRI NICE   SIZE   RES STATE   TIME   WCPU    CPU COMMAND
 2372 root      63    0    10M 9616K run   668:54 44.14% 44.14% eatmem
 2376 root      62    0    10M   10M run   663:02 43.75% 43.75% eatmem
 4037 root      49    0   276K  232K run     0:00  0.05%  0.05% top
 2449 root       2    0   276K  400K sleep  13:33  0.00%  0.00% top
 2369 root      -5    0    10M   64K sleep   4:12  0.00%  0.00% eatmem
 2367 root      -5    0    10M   64K sleep   4:11  0.00%  0.00% eatmem
 2373 root      -5    0    10M   80K sleep   4:00  0.00%  0.00% eatmem
 2368 root      -5    0    10M   84K sleep   3:56  0.00%  0.00% eatmem
 2375 root      -5    0    10M   68K sleep   3:54  0.00%  0.00% eatmem
 2374 root      -5    0    10M    4K sleep   3:50  0.00%  0.00% eatmem
 2228 root       2    0    72K   96K sleep   1:31  0.00%  0.00% ypbind
 2360 root      68   20    10M    4K run     1:22  0.00%  0.00% eatmem
 2361 root      68   20    10M    4K run     1:10  0.00%  0.00% eatmem
 2329 root      68   20    10M    4K run     0:19  0.00%  0.00% eatmem
 2267 root      18    0    16K    4K sleep   0:18  0.00%  0.00% update
 2326 root      68   20    10M    4K run     0:17  0.00%  0.00% eatmem
 2328 root      68   20    10M    4K run     0:16  0.00%  0.00% eatmem
 2269 root      -5    0   268K   28K sleep   0:13  0.00%  0.00% cron
 2217 root      -5    0    96K   36K sleep   0:12  0.00%  0.00% syslogd
 2278 root      18    0   364K    4K sleep   0:01  0.00%  0.00% <csh>
 4032 root      18    0   352K    4K sleep   0:00  0.00%  0.00% <csh>
 2232 root      10    0  8648K    4K sleep   0:00  0.00%  0.00% <mount_mfs>
    1 root      10    0   252K    4K sleep   0:00  0.00%  0.00% <init>
 2240 root      10    0    16K    4K sleep   0:00  0.00%  0.00% nfsiod


Thanks.

-Andrew
-- 
-----------------------------------------------------------------
Andrew Gillham                            | This space left blank
gillham@whirlpool.com                     | inadvertently.
I speak for myself, not for my employer.  | Contact the publisher.