2.3.12 slower than 2.3.6?
Bob_Tracy (rct@gherkin.sa.wlk.com)
Thu, 29 Jul 1999 07:12:50 -0500 (CDT)
I reported a few days ago that rc5des under 2.3.11 was able to process
only 107 kkeys/s on my machine after two days of uptime. 2.3.12 seems
to have fixed the fatal flaws in 2.3.11, but rc5des still doesn't run
as efficiently as it did with 2.3.6. With 2.3.12 I'm getting around
320 kkeys/s vs. 340 kkeys/s for 2.3.6 on my Cyrix 6x86L/P200+. Note
that 2.3.12 doesn't feel slow the way 2.3.11 did, and the only reasons
I'm mentioning anything now are (a) I noticed the slight performance
decrease in the logs, and (b) my threshold of pain got recalibrated to
a much lower level after the 2.3.11 experience :-).
If this is an expected result due to changes in the process scheduling
algorithm, that's cool, but I'd like to know that rather than assume it.
Otherwise, we have a problem... Anyone else seeing similar things with
other "benchmark" programs?
--
Bob Tracy | "They couldn't hit an elephant at this dist- "
Firewall Security Corp. | - Last words of Union General John Sedgwick,
rct@frus.com | Battle of Spotsylvania Court House, U.S. Civil War
-
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/