Benchmark Programs - Revision history

Peter Huewe at 11:05, 28 October 2011

2011-10-28T11:05:49Z

HectorOron: /* Unix Bench */

2010-11-03T10:03:31Z

‎Unix Bench

HectorOron: /* Unix Bench */ Add new URL

2010-11-03T10:03:14Z

‎Unix Bench: Add new URL

Js: lmbench: mention hostname requirement

2010-08-26T12:43:50Z

lmbench: mention hostname requirement

Js: updated lmbench instructions

2010-08-26T12:33:24Z

updated lmbench instructions

Glenn: +Category:Development Tools

2008-02-11T21:16:24Z

+Category:Development Tools

TimBird: remove link in section title

2007-06-07T18:35:14Z

remove link in section title

TimBird: offset benchmark warning, and fix wording

2007-06-07T18:34:05Z

offset benchmark warning, and fix wording

TimBird at 18:33, 7 June 2007

2007-06-07T18:33:11Z

RBot: Bot (Edward's framework)

2007-03-06T03:30:58Z

Bot (Edward's framework)

New page

Here are some different programs for performing benchmarking.

/\ ''Note: It is important to recognize that benchmarks between systems may be misleading.
Benchmark usage by the forum is primarily to determine differences in performance for different
software configurations on the '''same''' hardware system.''

== [[Unix Bench]] ==
FYI, the URL to the UnixBench is as follows;

http://www.tux.org/pub/tux/benchmarks/System/unixbench/

UnixBench contains 9 kinds of tests:
# Dhrystone 2 using register variables
# Double-Precision Whetstone
# Execl Throughput
# File Copy
# Pipe Throughput
# Pipe-based Context Switching
# Process Creation
# Shell Script
# System Call Overhead

UnixBench is preferred over lmbench since (according to reports) lmbench
cannot be easily cross-compiled.

I have cross-compiled lmbench - it's not too hard.
My target didn't have "make" or a compiler, but it did have bash.

Here is my recipe:
# Obtain lmbench. Make sure to get it from sourceforge (I used 3.0-a4), not bitmover, because the bitmover package is slightly tangled with a bitkeeper file "bk.ver". It's
# relatively easy to debug and disentangle it, but someone's already done that and put it up on sourceforge.
# Unpack lmbench in your build system.
# cd to lmbench-3.0-a4/src. Note that this is not the top level, which does have a Makefile.
# do "make OS=armv5EJl-linux-gnu CC=arm-linux-gcc" (or whatever your os and cross-compiler are)
# After everything is compiled, transport the whole directory tree to the target.
# cd to lmbench-3.0-a4/scripts
# Do ./config.run and answer some questions. This creates a config file that you can reuse.
# Do ./results
# Inside lmbench-3.0-a4/results will be a folder named armv5EJl-linux-gnu or something similar; inside that folder is a text file named imx21_199.0 or something similar.
# This is the raw lmbench results. Transport it back to civilization.
# When lmbench tries to save the results, it increments the last part ".0" until it finds an unused name. Therefore, you can rerun lmbench many times with a simple one-line
# bash script "for i in {1..100}; do ./results; done", and the result files will not overlap.
# You can get various kinds of summary postprocessing from lmbench. The "getsummary" script was sufficient for my purposes.
# To figure out which binary is generating which measurement, it may be useful to read the "lmbench" shell script in parallel with the raw results file.

== lmbench ==
The [[LMBench]] home page is at: http://www.bitmover.com/lmbench/
The sourceforge project page is at: http://sourceforge.net/projects/lmbench

← Older revision		Revision as of 11:05, 28 October 2011
Line 58:		Line 58:

	../scripts/getsummary ../results/arm-linux/*		../scripts/getsummary ../results/arm-linux/*
		+
		+	== Wishlist ==
		+
		+	A list of benchmark results would be useful:
		+	* Comparing performance of different FFT implementations on Beagleboard-XM: http://pmeerw.dyndns.org/blog/programming/arm_fft.html

	[[Category:Development Tools]]		[[Category:Development Tools]]

← Older revision		Revision as of 10:03, 3 November 2010
Line 10:		Line 10:

	OLD site: http://www.tux.org/pub/tux/benchmarks/System/unixbench/		OLD site: http://www.tux.org/pub/tux/benchmarks/System/unixbench/
		+
	NEW site: http://code.google.com/p/byte-unixbench/		NEW site: http://code.google.com/p/byte-unixbench/

@@ Line 45: / Line 45: @@
   cd lmbench-3.0-a9/src
   OS=arm-linux ../scripts/config-run
   OS=arm-linux ../scripts/results

← Older revision		Revision as of 21:16, 11 February 2008
Line 48:		Line 48:
	The [[LMBench]] home page is at: http://www.bitmover.com/lmbench/		The [[LMBench]] home page is at: http://www.bitmover.com/lmbench/
	The sourceforge project page is at: http://sourceforge.net/projects/lmbench		The sourceforge project page is at: http://sourceforge.net/projects/lmbench
		+
		+	[[Category:Development Tools]]

@@ Line 9: / Line 9: @@
 FYI, the URL to the UnixBench is as follows;
-http://www.tux.org/pub/tux/benchmarks/System/unixbench/
+OLD site: http://www.tux.org/pub/tux/benchmarks/System/unixbench/
 UnixBench contains 9 kinds of tests:

@@ Line 22: / Line 22: @@
 # System Call Overhead
-Unix Bench is preferred over lmbench since (according to reports) lmbench
+== lmbench ==
-cannot be easily cross-compiled.
+The LMBench home page is at: http://www.bitmover.com/lmbench/ and/or http://lmbench.sourceforge.net/<br>
-I have cross-compiled lmbench - it's not too hard.
+=== Instructions for lmbench-3.0-a9 ===
-My target didn't have "make" or a compiler, but it did have bash.
-Here is my recipe:
+(Adjust CC and OS according to your needs.)
-# Obtain lmbench. Make sure to get it from sourceforge (I used 3.0-a4), not bitmover, because the bitmover package is slightly tangled with a bitkeeper file "bk.ver". It's
-# relatively easy to debug and disentangle it, but someone's already done that and put it up on sourceforge.
-# Unpack lmbench in your build system.
-# cd to lmbench-3.0-a4/src. Note that this is not the top level, which does have a Makefile.
-# do "make OS=armv5EJl-linux-gnu CC=arm-linux-gcc" (or whatever your os and cross-compiler are)
-# After everything is compiled, transport the whole directory tree to the target.
-# cd to lmbench-3.0-a4/scripts
-# Do ./config.run and answer some questions. This creates a config file that you can reuse.
-# Do ./results
-# Inside lmbench-3.0-a4/results will be a folder named armv5EJl-linux-gnu or something similar; inside that folder is a text file named imx21_199.0 or something similar.
-# This is the raw lmbench results. Transport it back to civilization.
-# When lmbench tries to save the results, it increments the last part ".0" until it finds an unused name. Therefore, you can rerun lmbench many times with a simple one-line
-# bash script "for i in {1..100}; do ./results; done", and the result files will not overlap.
-# You can get various kinds of summary postprocessing from lmbench. The "getsummary" script was sufficient for my purposes.
-# To figure out which binary is generating which measurement, it may be useful to read the "lmbench" shell script in parallel with the raw results file.
-== lmbench ==
+ cd lmbench-3.0-a9/src
-The [[LMBench]] home page is at: http://www.bitmover.com/lmbench/
+ make CC=arm-linux-gcc OS=arm-linux TARGET=linux
-The sourceforge project page is at: http://sourceforge.net/projects/lmbench
 [[Category:Development Tools]]

@@ Line 3: / Line 3: @@
 <div style="margin:0; margin-top:10px; margin-right:10px; border:1px solid #dfdfdf; padding:0 1em 1em 1em; background-color:#ffffcc; align:right; ">
 ''Note: It is important to recognize that benchmarks between systems may be misleading.
-Benchmark usage by the forum is primarily to determine differences in performance for different
+Benchmarks should primarily be used to determine differences in performance for different
 software configurations on the '''same''' hardware system.''
 </div>