Openmoko GTA02 NAND performance improvements

On Sunday night, after returning from a weekend trip to Hamburg, I sat down and looked at the NAND and S3C2442B data sheet to figure out the actual timing performance. Interestingly, the NAND timings were much more verbose and detailed (and had different names) than the timings described in the NAND controller section of the S3C24xx manual - and both are from Samsung ;)

Anyway, it seems like the current timing settings for the various stages (reading u-boot by the stepping stone mechanism, reading the kernel by u-boot as well as actual mtd-based access inside the Linux kernels) were extremely suboptimal. They're basically standard timings designed to work with most NAND flashes out there, ignoring the fact that GTA02 uses one specific flash with very good (fast) timings, at least according to the data sheet. There should also be no PCB / routing related issues such as capacitive overload preventing higher speeds, since the NAND flash die is stacked onto the CPU die in one package, and the NAND controller signals are not routed on the PCB anywhere.

Some initial experiments based on the calculations show that the performance can be easily improved by 41% over the stock GTA02 NAND performance. However, the actual speed (6.59MBytes/sec) is still much lower than the theoretical maximum read performance of 15.64MBytes/sec. It seems there is more room for improvement inside the MTD layer of the Linux kernel.

It's again quite amazing how much room there is for improvement in GTA02 performance, both power consumption wise (see my recent post about framebuffer blanking), as well as actual data throughput. Those are really low-hanging fruits, and it's very surprising that nobody working for Openmoko or in the Openmoko community has been able to spend some time to look into those...

About the new format / structure of FOSS.in

There has been quite some discussion on various places on the net about the recently-announced change of the FOSS.in conference format. Instead of lots of talks/presentations, there is an emphasis on workshops and similar more interactive and collaborative types of events.

I have been speaking to a number of developers who have been to FOSS.in before and who have been putting in proposals for FOSS.in/2008, too. They all think it is a very courageous step: going from a successful, working 'traditional conference' scheme with presentations, sufficient sponsors to cover travel expenses of foreign speakers, etc. to a very different, much more developer-community oriented event.

I also think it is a courageous experiment. I have not yet heard of any event similar to this before. Sure, there are project days and developer meetings or miniconfs or whatever you might call them. But not to the extent as, at least to my perception, FOSS.in is planning right now.

In any case, it depends on what your target is. 'typical' Linux conferences are basically focussing on either one (or multiple) of the following:

• Spread the word about Linux/FOSS, to generate more adoption
• Provide updates on development progress to various people in the community as well both individual and professional users

However, if you emphasize on the actual FOSS development, then I think it is quite legitimate to go for a event format that FOSS.in is heading to right now.

It is exactly FOSS.in who can try such a change, since it is a true community event without any commercial interest and without affiliation to particular companies.

And after all, who wants to see the same kind of event happening each and every year, with the same kind of people talking? Wouldn't that be boring after some time? Especially if there are a number of other events doing more or less the same?

In any case, personally I'm planning to do a FOSS.in WorkOut on a USRP+gnuradio based GSM scanner project. India is the perfect place on earth to get this done, since the government mandates A5/0 (no encryption) and thus all the packets can be captured and each and every bit implemented as wireshark plugin.