Message-ID: <e6babb600611270923h139a59ecr82af786f3aedf607@mail.gmail.com>
Date:	Mon, 27 Nov 2006 10:23:17 -0700
From:	"Robert Crocombe" <rcrocomb@...il.com>
To:	linux-kernel <linux-kernel@...r.kernel.org>
Subject: Re: [patch] x86: unify/rewrite SMP TSC sync code

The difference that Wink reports is tiny compared to that measured on
my Opteron machines:

dual (2.6.17):

rcrocomb@...fires-aptest:cyclecounter_test$ ./rdtsc-pref 1000000
rdtsc:   average ticks=  10
gtod:    average ticks=4296
gtod_us: average ticks=4328

quad (2.6.16-rt29):

rcrocomb@...nky:wink_saville_test$ ./rdtsc-pref 1000000
rdtsc:   average ticks=  10
gtod:    average ticks=5688
gtod_us: average ticks=5711

I have my own little test that I'll attach, but it gives a similar
result.  Here are the results from the 2x box:

rcrocomb@...fires-aptest:cyclecounter_test$ ./timing
Using the cycle counter
Calibrated timer as 2593081969.758825 Hz
4194304 iterations in 0.016 seconds is 0.004 useconds per iteration.

rcrocomb@...fires-aptest:cyclecounter_test$ ./timing_gettimeofday
Using gettimeofday
4194304 iterations in 6.793 seconds is 1.620 useconds per iteration.

I have used the pthread affinity and/or cpuset, etc. mechanisms to try
and inject some reliability into the measurement.

Using gtod() can amount to a substantial disturbance of the thing to
be measured.  Using rdtsc, things seem reliable so far, and we have an
FPGA (accessed through the PCI bus) that has been programmed to give
access to an 8MHz clock and we do some checks against that.

-- 
Robert Crocombe
rcrocomb@...il.com

View attachment "timing.cpp" of type "text/x-c++src" (3566 bytes)

Powered by blists - more mailing lists