Comments for EEMBC & CoreMark Blog http://www.eembc.org/wordpress This Blog focuses on EEMBC & CoreMark Tue, 03 Aug 2010 15:43:32 +0000 http://wordpress.org/?v=2.9.2 hourly 1 Comment on EEMBC Director of Software Engineering takes on sumo wrestler with EEMBC power by David Harris http://www.eembc.org/wordpress/?p=125&cpage=1#comment-192 David Harris Tue, 03 Aug 2010 15:43:32 +0000 http://www.eembc.org/wordpress/?p=125#comment-192 i like to watch wrestling on TV and i have always been a fan of WWE-.. i like to watch wrestling on TV and i have always been a fan of WWE-..

]]> Comment on Using Coremark on Multiple Cores by Suresh Shukla http://www.eembc.org/wordpress/?p=48&cpage=1#comment-179 Suresh Shukla Thu, 01 Jul 2010 05:31:47 +0000 http://www.coremark.org/wordpress/?p=48#comment-179 That's very true. On dual-core system CoreMark linearly scaled to 2 times of single-core performance. That’s very true.
On dual-core system CoreMark linearly scaled to 2 times of single-core performance.

]]> Comment on EEMBC Director of Software Engineering takes on sumo wrestler with EEMBC power by Ellis Gibson http://www.eembc.org/wordpress/?p=125&cpage=1#comment-170 Ellis Gibson Mon, 24 May 2010 17:23:20 +0000 http://www.eembc.org/wordpress/?p=125#comment-170 Sumo wrestling is my all time favorite. i always watch it a lot in Japanese channels.*~: Sumo wrestling is my all time favorite. i always watch it a lot in Japanese channels.*~:

]]> Comment on Data types by Paul Kimelman http://www.eembc.org/wordpress/?p=94&cpage=1#comment-153 Paul Kimelman Mon, 08 Mar 2010 23:11:47 +0000 http://www.coremark.org/wordpress/?p=94#comment-153 The reverse is true. 32-bit processors do best with all int/unsigned local variables (globals and statics do not matter much). If you use char or short locals, you impact performance because they have to sign-extend or unsign-extend over and over. The reverse is true. 32-bit processors do best with all int/unsigned local variables (globals and statics do not matter much). If you use char or short locals, you impact performance because they have to sign-extend or unsign-extend over and over.

]]> Comment on EEMBC Director of Software Engineering takes on sumo wrestler with EEMBC power by Satoshi http://www.eembc.org/wordpress/?p=125&cpage=1#comment-147 Satoshi Sun, 07 Mar 2010 01:21:25 +0000 http://www.eembc.org/wordpress/?p=125#comment-147 It looked like a really rare chance to fight against a sumo restler. It looked like a really rare chance to fight against a sumo restler.

]]> Comment on Data types by Denis Cabrol http://www.eembc.org/wordpress/?p=94&cpage=1#comment-132 Denis Cabrol Wed, 17 Feb 2010 16:15:33 +0000 http://www.coremark.org/wordpress/?p=94#comment-132 Bruno is absolutely correct. 1- The bench is heavily weighted towards 32-bit processing. We measured that 65% of the instructions generated by the compilation of the Coremark bench were of the Long type (32-bit). ALL processing instructions (ADD, SUB, MULT) in the bench work on 32-bit data. 2- They are thousands of applications where you will never instantiate 32-bit data. Think about: - remote controls - most appliances - electronic light dimmers - Thermostats - seat positioning, window lift and mirror control in automotive. The Coremark bench, by implicitly assuming that processing is always performed on 32-bit data (and not being very upfront about it), favors 32-bit architectures over 8- and 16-bit ones. My recommendations are: - be more upfront about what Coremark actually benches - provide versions for 32-bit (which we have) and 8/16-bit In any case this is already much better than the old Drystone bench. Denis Bruno is absolutely correct.
1- The bench is heavily weighted towards 32-bit processing. We measured that 65% of the instructions generated by the compilation of the Coremark bench were of the Long type (32-bit). ALL processing instructions (ADD, SUB, MULT) in the bench work on 32-bit data.

2- They are thousands of applications where you will never instantiate 32-bit data. Think about:
- remote controls
- most appliances
- electronic light dimmers
- Thermostats
- seat positioning, window lift and mirror control in automotive.

The Coremark bench, by implicitly assuming that processing is always performed on 32-bit data (and not being very upfront about it), favors 32-bit architectures over 8- and 16-bit ones.

My recommendations are:
- be more upfront about what Coremark actually benches
- provide versions for 32-bit (which we have) and 8/16-bit

In any case this is already much better than the old Drystone bench.
Denis

]]> Comment on What do YOU use to edit your code? by Jonas http://www.eembc.org/wordpress/?p=113&cpage=1#comment-131 Jonas Wed, 17 Feb 2010 08:08:39 +0000 http://www.eembc.org/wordpress/?p=113#comment-131 I use: - Eclipse for "hard-core" sesseions of C++/Python programming, mostly for its excellent cross-referencing feature (just hit F3 and it'll jump to the definition/implementation of a function/class/...) - Kate (KDE/Linux) for quick hacking, when "booting" Eclipse is too much of a hassle - Kile (KDE) for LaTeX - Vim for console editing, e.g. quickly changing a config file Regards I use:

- Eclipse for “hard-core” sesseions of C++/Python programming, mostly for its excellent cross-referencing feature (just hit F3 and it’ll jump to the definition/implementation of a function/class/…)
- Kate (KDE/Linux) for quick hacking, when “booting” Eclipse is too much of a hassle
- Kile (KDE) for LaTeX
- Vim for console editing, e.g. quickly changing a config file

Regards

]]> Comment on Data types by Bruno Richard http://www.eembc.org/wordpress/?p=94&cpage=1#comment-130 Bruno Richard Wed, 17 Feb 2010 07:52:17 +0000 http://www.coremark.org/wordpress/?p=94#comment-130 Hi Shay, I think Coremark is a good effort toward "modern" benchmarking. However, concerning data types, it suffers from a great problem: The arithmetic used to measure the "performance" of a compiler/target is 32-bit! This means that performance (speed) figures obtained on 8 or 16-bit processors are meaningless: Adding 2 32-bit objetcts is one instruction on a 32-bit processor, but can be a dozen on a 8-bitter. I know, the benchmark tries to mimick some real-world applications, which require 32-bit arithmetics. However, lots of real-world applications do NOT require 32 and can live well with 16-bits arithmetic (think about all-software MP3 players). When using the CoreMark results, engineers may get fooled by results and pick a 32-bit processor where an 8-bit would have been enough. My proposal would be: - Rewrite (just a little bit) the Coremark benchmark so that it has a meaningful 16-bit version. - Separate the results of 8/16-bit processors from the 32-bit ones on the results page, so the figures are more meaningful. Thanks for your excellent work anyway, Bruno Hi Shay,

I think Coremark is a good effort toward “modern” benchmarking.
However, concerning data types, it suffers from a great problem: The arithmetic used to measure the “performance” of a compiler/target is 32-bit! This means that performance (speed) figures obtained on 8 or 16-bit processors are meaningless: Adding 2 32-bit objetcts is one instruction on a 32-bit processor, but can be a dozen on a 8-bitter.

I know, the benchmark tries to mimick some real-world applications, which require 32-bit arithmetics. However, lots of real-world applications do NOT require 32 and can live well with 16-bits arithmetic (think about all-software MP3 players).

When using the CoreMark results, engineers may get fooled by results and pick a 32-bit processor where an 8-bit would have been enough.

My proposal would be:
- Rewrite (just a little bit) the Coremark benchmark so that it has a meaningful 16-bit version.
- Separate the results of 8/16-bit processors from the 32-bit ones on the results page, so the figures are more meaningful.

Thanks for your excellent work anyway,
Bruno

]]> Comment on What do YOU use to edit your code? by Anthony Green http://www.eembc.org/wordpress/?p=113&cpage=1#comment-129 Anthony Green Tue, 16 Feb 2010 01:09:22 +0000 http://www.eembc.org/wordpress/?p=113#comment-129 Hi Shay - I'm in the "still use Emacs" camp, but you say it like it's a bad thing! Eclipse admittedly beats out Emacs for Java hacking, but for C/C++/asm/Verilog/Lisp/etc, Emacs still sets a high bar. Witness the tweakage... http://github.com/atgreen/emacs/blob/master/init.el I recommend revisiting Emacs. (but remember to run it in server mode and use emacsclient for fast startup times!) AG Hi Shay -

I’m in the “still use Emacs” camp, but you say it like it’s a bad thing! Eclipse admittedly beats out Emacs for Java hacking, but for C/C++/asm/Verilog/Lisp/etc, Emacs still sets a high bar. Witness the tweakage… http://github.com/atgreen/emacs/blob/master/init.el
I recommend revisiting Emacs. (but remember to run it in server mode and use emacsclient for fast startup times!)

AG

]]> Comment on What do YOU use to edit your code? by webmaster http://www.eembc.org/wordpress/?p=113&cpage=1#comment-126 webmaster Thu, 11 Feb 2010 16:16:32 +0000 http://www.eembc.org/wordpress/?p=113#comment-126 I too use Notepad++, great editor. For HTML, Javascript, PHP, etc, I use Adobe Dreamweaver, the best HTML editor in my opinion. I too use Notepad++, great editor. For HTML, Javascript, PHP, etc, I use Adobe Dreamweaver, the best HTML editor in my opinion.

]]>