In addition to my previous report I would like to add some
more data-points:
As some people suggested, I let crashme run on systems with
an underclocked AMD CPU. Result: all systems locked hard,
with a time period scaling with the lower clock frequency!
Here are the results in detail:
10) MB ASUS P/I-P55T2P4, 64MB RAM, AMD K6-200 at 133 MHz, 2GB IDE HDD,
64MB Swap, Linux-2.0.34pre15, egcs-1.0.2
(same system as in test #3, but CPU underclocked at 133 MHz)
-> Crash after 1 hour 50 minutes (17158 Processes)
11) MB ASUS P/I-P55T2P4, 256MB RAM, AMD K6-233 at 133MHz, 2xAHA2940UW,
2GB SCSI + 16GB UW-SCSI RAID5 (5xIBM DCAS-34330), 120MB Swap,
Linux-2.0.34pre16, gcc-2.7.3.2
(same system as in test #6, but CPU underclocked at 133MHz)
-> Crash after about 30 minutes
It's true that all crashing systems have a ASUS P55T2P4 motherboard.
I don't have any other MB available yet, but the boards in question
all have revision 3.1, which, according to the manual, support the
K6 200 & 233 CPUs.
Some people also mentioned possible problems with AMD CPU's under
heavy load, if there's a lack of cooling.
As I said in my previous report, all systems under test were used
for stress-testing linux-2.0.34prexy before. For several days they
executed tests with heavy disk- and network-I/O and high CPU load
(system load sometimes went up to 60 and more), all without a single
problem!
This is the main reason I think the crashes I see must have
something to to with the illegal CPU opcodes "crashme" executes,
and not with general system instability.
- andreas
-- Andreas Haumer | email: andreas@xss.co.at | PGP key available *x Software + Systeme | phone: +43.1.6001508 | on request. Buchengasse 67/8 | +43.664.3004449 | A-1100 Vienna, Austria | fax: +43.1.6001507 |- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu