[BENCHMARK] ctxbench and SYSENTER discussion

From: Bill Davidsen (davidsen@tmr.com)
Date: Wed Jan 01 2003 - 11:49:19 EST


I changed the report generator for ctxbench slightly after the recent
discussions of system call methods. I added a column of context loop rate
divided by MHz of the processor, and ran on the same or similar 2.4
kernels (this is a hardware thing, version matters only a little).

What it showed was a piss-poor value for the P4, best for my old P-II
Celeron, and Athlon vs. P-III closely swapping third place. In fact the P4
had a value about 3x slower than the P-II.

If someone wants to run the benchmark with the old and new system call
setup it might be useful to quantify the gain.

Benchmark is at www.unyuug.org/benchmarks, and if someone tests this I'd
like a copy of the raw results for my collection as well as the extract
posted to lkml.

-- 
bill davidsen, CTO TMR Associates, Inc <davidsen@tmr.com>
  Having the feature freeze for Linux 2.5 on Hallow'een is appropriate,
since using 2.5 kernels includes a lot of things jumping out of dark
corners to scare you.

- 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/



This archive was generated by hypermail 2b29 : Tue Jan 07 2003 - 22:00:15 EST