Nishanth' tech rambles

--NAND flash preps for 4 bits per cell-- Single-chip 16-Gbit MLC NAND flash devices from 50-nanometer-classprocesses were introduced in the first half of 2007. http://newsletter.embedded.com/cgi-bin4/DM/y/eBFuD0FyTZ10FrY0FhNF0Ef --NAND to scale for several years-- NAND flash is projected to scale at least for another five or soyears, pushing out the need for next-generation or universal memorytechnologies. http://newsletter.embedded.com/cgi-bin4/DM/y/eBFuD0FyTZ10FrY0FSzc0EX --16-Gbit MLC NAND flash weighs in-- To meet increasing demand for more digital storage, leading flashmemory manufacturers are touting their latest single-chip 16-Gbit MLCNAND flash devices. Earlier this year, Toshiba announced a 16-Gbit MLCNAND flash manufactured in the 56-nm process node. Samsung, meanwhile,introduced a 16-Gbit MLC NAND flash using a 51-nm process, a half aprocess node ahead of Toshiba's. http://newsletter.embedded.com/cgi-bin4/DM/y/eBFuD0FyTZ10FrY0FhNG0Eg

Sony goes ahead with a change in policy - quote: "The revised Terms of Service and User Agreement will be effective no sooner than October 30th, 2007".. Since the Mail is on Nov 9th, Weee that is "-9 " days of notice.. "oh by the way," 11. PUBLIC INFORMATION SCEA may provide all users with information relating to your game play. (ADDED FOLLOWING) You hereby authorize SCEA to use, distribute, copy, display, and publish information relating to your game play for gaming purposes including tournaments and ranking, without payment to you . In addition, you will have the option to post, stream or transmit pictures, photographs, game related materials, or other information through PSN to share with the PSN community ("User Material"), provided no rights of others are violated. You authorize and sublicense SCEA the right to authorize anyone to use, distribute, copy, modify, display, and publish your User Material in any manner on any medium without paym...

http://www.computerworld.com/action/article.do?command=viewArticleBasic&taxonomyId=154&articleId=9044179&intsrc=hm_topic

Here is what a friend of mine Jayan John has to say (reposted with permission) on a discussion on various processor variants today in Cellphone business. The relevant debate may not be “standalone baseband processors” vs”SOC", but whether or not SOC is the best option... Why ? Quote: “Convergence in the industry of computing capabilities and communication protocols has resulted in the development of devices with enhanced functionalities and capabilities. Providing a range of functions at a low cost while maintaining the reliability and compatibility of existing systems and devices is a major challenge today. As devices become more complex with numerous features, it is of prime importance to find out ways to extract more power from the batteries". With this in mind, choosing the right level of integration is a difficult decision. To make this interesting… Read Texas Instruments vs Philips back in 2005 ( http://www.eetimes.com/news/latest/showArticle.jhtml?articleID=159901628 ...

Here are some interesting news articles in recent days: Appilcation processors are dead ? http://www.linuxdevices.com/news/NS9250186181.html -> that'd be painful... ;) Broadcom : http://www.news.com/8301-10784_3-9797486-7.html?part=rss&subj=news&tag=2547-1_3-0-5 Cnet has some details though not radical http://www.linuxdevices.com/news/NS8838596703.html - that is some indepth info.. Should the world tremble? Colored printk : http://kerneltrap.org/Linux/Colored_printk_Output i like jazzy stuff.. mebbe i will steal this for uboot v2??...

So, I am on a slow roll on uboot2 changes for OMAP.. uboot 2 is lot less mature as it can be seen digging into the code, but it is most promising. The main reason I want to jump on uboot2 is that omap support is very nascent here, and lots can be done.. Some areas of concern: * i2c seems pretty unclear * I like the dev concept, but should modules of OMAP be also dev based? I dont think so.. generic modules - yes..i2c etc.. * on dev -> it looks very immature considering that include/driver.h has: 45 struct driver_d { 55 ssize_t (*erase) (struct device_d*, size_t count, unsigned long offset); 56 int (*protect)(struct device_d*, size_t count, unsigned long offset, int prot); 59 void (*info) (struct device_d *); Errrr... erase,protect,info ?? that is flash type device struct rt? it does not fit into a generic driver structure.. * I dont like the fact that there is still a config.h (include/configs/xyz.h SYMLINKed to include/config.h) which makes NO sense to me.. on...

Ok today I had a look of Makefile.. Denx supports a unified make file unlike the strategy in kernel. There are some interesting snippets of code in the same. In short, we could have something for configs as follows: _ _ _config Where, SiliconName = omap2420, omap3430,et.al. boardName = sdp,osk,apollon etc Special = Boot Mode -> e.g. nand/onenand or defined boot mode for that platform. Now, should i have es versions??? 3430 es1 and es2 are different enough to have two code definitions? One thing I don’t agree to : u-boot.arm/doc/README.ARM-SoC cpu/$(CPU)/$(SOC)/ does not facilitate re-use -> coz I might want to reuse across arm versions.. and this does not let me do it.. I am getting diverted again -> this time it is U-boot v2!!!! cg-clone http://www.denx.de/git/u-boot-v2.git/ u-boot.v2 let me see what sacha has got...

Where are the files? u-boot.arm is what i am basing myself on. Directory : u-boot.arm/lib_arm/ Description: ARM specific library function implementation Files: armlinux.c _ashrdi3.S cache.c _divsi3.S _modsi3.S _umodsi3.S _ashldi3.S board.c div0.c Makefile _udivsi3.S Contains: some important (if I may use such a word - all look important): bootm implementation, banner display cache functions start_armboot Directory :/u-boot.arm/cpu/arm1136 Description: ARM1136 CPU specific logic. This is kind of common for most part. I can see that ARM integrator and 2420 are the two folks using this so far at least.. Files: config.mk cpu.c interrupts.c Makefile start.S Contains: start.s is the basic starting point of the boot process.. __start is where the code starts the boot.. So... after almost an year and a half of CCS assembly, I need to brush up the GNU ARM Assembly.. i got it somewhere.. hmm.. Ref: GNU ARM Assembly Quick Reference Guide: http://www.microcross.com/GNU-ARM-Assy-Quick-Ref.pdf An I...

I am starting off with a fresh OMAP port of uboot . Nothing is proper unless i answer some of my questions.. Why Do this? * The main reason of doing this is coz I am sick and tired of crappy code and repatch folks have been doing.. and am hoping that I can do a clean reuse for OMAP platform * U-boot is largely ignored for maybe the lack of fame and fortune stuff, but I kinda like it at the moment, coz it reminds me the first line of code i wrote for 2430 port.. it was in uboot .. on virtual platform. kind of presilicon . pressures were high, could never do a clean job then.. here is a chance to get things right.. * I am kind of tired doing my "project" work * My wife is getting busy for her exams and other than an hour's walk, I wont get troubled much for a week at least (I hope ;) ). What is my plan? I am hoping to integrate a lot of modules and avoid repetitions where possible.. e.g.: why have multiple board file for each of omap 2420 platforms? GPMC as an example...

cordic algo seems pretty powerful: http://www.ddj.com/embedded/196700769 C Code for the same: http://my.execpc.com/~geezer/embed/cordic.c Now to see if VFP and Neon support can handle float ops.. never tried it so far... Hmm...

Damn... they killed my template!!!!!! by the way, USB 3.0 is out!!! here is more what my friend passed on: USB 3.0 ---------- Superspeed Personal USB interface 10 times faster than the present USB optimized for low power and improved protocol efficiency USB 3.0 ports and cabling will be designed to enable backward compatibility as well as future-proofing for optical capabilitie USB 3.0 specification is expected by the first half of 2008 Intel, HP, NEC, NXP, TI, Microsoft are the 3.0 promotors Ref: http://www.intel.com/pressroom/archive/releases/20070918comp.htm http://focus.ti.com/docs/pr/pressrelease.jhtml?prelId=sc07155

Brain dead Windows Vista says: C:\Users\XYZ>ipconfig /release The requested operation requires elevation. Zen and the art of higher conciousness in Vista??? or is it yogic powers?? XP used to be down to earth like us common folks.. ;)

It is interesting to see http://www.linuxdevices.com/news/NS2766418755.html .. if windriver manages to get taken over by IBM, do we see a better set of patches coming the opensource way? i'd say there'd be finally a good competition to Montavista in embedded support sphere Keep track of kernel developments... if u track kernel trap like me.. u'd want this too in ur list: http://www.linux-foundation.org/en/Linux_Weather_Forecast

With Denx opening the gates on Dat's Das Uboot, it is kind of interesting to see how things will flow on the "much belated" uboot push into the uboot official tree. For those interested: Uboot home page: http://www.denx.de/wiki/UBoot Custodians: http://www.denx.de/wiki/UBoot/Custodians ARM custodian is:Peter Pearse ARM Tree weblink: http://www.denx.de/cgi-bin/gitweb.cgi?p=u-boot/u-boot-arm.git;a=summary How to pull a uboot git: http://www.denx.de/wiki/UBoot/CustodianGitTrees Patch status: http://www.denx.de/wiki/UBoot/PatchStatus Compiler to use was already covered in my document: http://www.geocities.com/crecmca98/Linux/omap2430_From_almost_scratch.pdf Codegen is the best.. Things I am planning on doing: start with 2430, move to 3430 patches. I am thinking strongly of forcing muxing of individual drivers(not in uboot) to be done exclusively by kernel..

http://www.youtube.com/watch?v=Cy1uqv9nNGs The board looks good.. the next generation of boards much more "compressed" and deadlier!!! Love being on the bleeding edge of tech!!!

Searching for a ready made rpm is pain if yum does not pull it out automagically. instead download source, and run "rpmbuild -ta git-1.5.2.3.tar.gz"

Denx is out looking for a new u-boot maintanence mechanism similar to what Linux has. some new(?) guidelines on development: http://www.denx.de/wiki/UBoot/DesignPrinciples See this thread on the maintainer stuff: http://sourceforge.net/mailarchive/forum.php?thread_name=20070704002828.E7F34353B2C%40atlas.denx.de&forum_name=u-boot-users here is more on code refactor stuff: http://sourceforge.net/mailarchive/forum.php?thread_name=E1I5qQW-0007cq-3i%40jdl.com&forum_name=u-boot-users

Yesterday night, I was looking at my linux box, and thought, let me revamp my old kernel driver skills.. thought I'd write a small character LCD driver.. checked the datasheet, gathered that it was childish. Till i started looking at userspace functionality... Thats when I groaned.. Options: LCDProc: http://lcdproc.omnipotent.net/ LCD Module: http://lcd.sourceforge.net/ LCD mod: http://lcd-mod.sourceforge.net/ etc.... in the end.. hmm.. I kind of have decided on lcd proc.. dunno how difficult it is.. the worst part is that.. there are no standard interfaces in the kernel itself.. Will update this post as I develop on this..

http://news.com.com/2100-7344_3-6194139.html?part=rss&tag=2547-1_3-0-5&subj=news As of last, Linus was against pulling in GPL3 for kernel... what sayeth he now??? Gotta read it sometime later this weekend.. Update: http://linuxdevices.com/news/NS6228340866.html GPL V3: http://gplv3.fsf.org/ - there is GPL and LGPL - Good!! One feature I am worried upon is the "non tviozation" concept: Imagine.. one day Linus says ok to GPL v3 for linux, all apps are GPL V3. Now, using today's technology, the h/w vendor who releases a GPL V3 based software for a cellphone..needs to ensure that you dont have a malicious "cracked" version running on the board. How does they do it? by providing signature for images.. Does "non tvio" software mean the signatures are published? Errr.. beats the purpose right? Does this imply that the vendor should impart the tools to make the modified s/w run on thier boards -Errrr.... why should the vendor want to give the key to t...

Intrest Story: Amusia Musica La MainFrama : http://www.wired.com/culture/art/news/2007/07/IBM1401_Musical Ze Linux Killer?? Reiser SNAFU! For some one who had hacked around with File systems, name such as JFFS2,JFFS3 etc reminded strongly of reiserfs -> the filesystem that made ext2fs look crap.. Now.. this!! http://www.wired.com/techbiz/people/magazine/15-07/ff_hansreiser Voca Silencia : http://arstechnica.com/news.ars/post/20070626-internet-radio-day-of-silence-hushes-thousands-of-stations.html Signed-off-by: Nishanth Menon http://arstechnica.com/news.ars/post/20070628-rep-inslee-game-simply-not-over-for-net-radio-now-or-on-july-15.html Essentially.. get ur ipods and shared drives ready.. internet stations are dying!!! Power: Lol.. intel understands it today??? but then the media was always slow compared to research community.. http://arstechnica.com/articles/paedia/cpu/power-plays.ars Open Kerala Project: Will the Open Kerala become Open Karela (bittergord in Hindi)?? Knowing th...

http://www.arnnet.com.au/index.php/id;262730519;fp;4;fpid;319049444

Business : --Programmer shortage: A problem, but what kind? How bad? And how to solve it?-- Convinced that the software programmer crisis is due to fewer college students coming into technical fields, Jack Ganssle points to "stark numbers" that have convinced him of the urgency of the problem. http://newsletter.embedded.com/cgi-bin4/DM/y/eBAJ10FyTZ10FrY0FLqX0Ex --Five chipmakers control 32% of manufacturing, says analyst-- Samsung, TSMC, Intel, Toshiba, and UMC had the five largest shares of the world's chip manufacturing capacity at the end of 2006, with a combined capacity of just over 2.9 million 200-mm equivalent wafer per month. They accounted for 32 percent of total wafer capacity as of year-end 2006, according to market research company IC Insights. http://newsletter.eetimes.com/cgi-bin4/DM/y/eBALV0FyTZ10FrK0FLvK0EG --Big shakeup predicted in flash memory markets-- The market for trapped charge and phase change memories is set to outpace many other flash memory sec...

People who know me, know my love for UML. Ever since I used it for my Master's project, I have been a fan of UML. After years of experimentation between Visio and Rational Rose(depending on the resources available to the project I worked on), I have always felt that Rational Rose took the prize! In my opinion, opensource was always lagging behind. I played with ArgoUML , KUML , Dia and bunch of other projects, never got convinced.. Today, the search began again, since my Rational Rose 98 License expired and I was lazy to contact IT for getting a new one.. Then hit this deadly project: Star UML . Love its capability - a cross between Visio and Rational Rose. It can read my Rose MDL files, generate C++ code and overall pretty usefull.. Loves: Formatting Options -> For someone who plays with usecases, aligning them properly was a pain in Rational Rose, easy on Visio.. Ability to generate code easily (unlike Grr.. Rational rose which needs a deployment and component view to do anyt...

From u boot mailing list - one lil nugget of a link: http://download.micron.com/pdf/technotes/nand/tn2916.pdf At least the concepts remain same for all platforms... they might miss a little bit on certificates etc.. Oh by the way, Mentor Graphics is sponsoring how they "empowered" USB on TI's DaVinci processor, next month on embedded magazine!! grok brok pow mow...

For someone of physics background... this news is interesting: http://www.wired.com/science/discoveries/magazine/15-06/st_best Future vehicles: http://www.wired.com/cars/futuretransport/news/2007/06/future_engines

For those who have seen Tesla in the movie Prestige will enjoy this: http://newsletter.eetimes.com/cgi-bin4/DM/y/e8rv0FyTZ10FrX0FKEC0AM --Wibree becomes Ultra Low Power Bluetooth--The Bluetooth Special Interest Group and Wibree forum are merging in amove that makes the Nokia pioneered ultra low power wirelesstechnology part of the Bluetooth specification as an ultra low powerBluetooth technology. http://newsletter.eetimes.com/cgi-bin4/DM/y/e8rv0FyTZ10FrX0FKEA0AK --Intel details R&D on multi-core--In a handful of recent technical papers, Intel Corp. is disclosingmore details about its research into the future of multi-coreprocessors. http://newsletter.eetimes.com/cgi-bin4/DM/y/e8rv0FyTZ10FrX0FKD80AA

http://www.arm.linux.org.uk/mailinglists/etiquette.php Love his spelling... If you reply to a message, avoid top-posting like this: Thank you for your reply. Wouldn't it be a better idea to frobnicate foo? Bob -----Original Message----- From: Alice [ mailto:alice@example.com ] Sent: Tuesday, January 11, 2005 8:47 PM To: Bob Cc: linux-arm-kernel Subject: Re: What is foo supposed to do? It's to make sure bar does not eat the gnats.

For those folks interested in audio drivers... ALSA Vs OSS debate rolls on... Thread: http://linuxaudio.org/pipermail/linux-audio-dev/2007-June/019294.html

Micro AB receptacles to replace Mini AB receptacles http://www.usb.org/developers/Deprecation_Announcement_052507.pdf announces this. So, to see a such a connector or cable, a bit of google: From here I got the following info: Micro-A plug (white) Micro-AB receptacle (gray) Micro-B plug (black) Micro-B receptacle (black) hehe that is useful for color blind folks too :) Standard-A receptacle to Micro-A plug (no other adaptors allowed). Quote: Micro-USB plug is rated for 10,000 connect-disconnect cycles. It is about half the height of the mini-USB connector in widespread use today but features a similar width. I hear it is stainless steel also :).. Ok, now my Mot A1200 gets plugged in and out of USB cable at my desk around 5 times a day, and once in house on average, that gives me like 4 year or so life time for my cable.. Err.. where are those days where u buy a TV and expect it to last 20 years ;).. just kiding ofcourse.. (my dad has a 15 year old TV and loves it).. Tons of pages lat...

Khasim finally started posting OMAP 3430 patches to git.. will take me a while to get to looking at the patches.. but.. it is good to see them come finally!! interestingly some devices using various OMAPs: http://pdadb.net/index.php?m=pdamaster i suppose lots of similar sites exist.. Yesterday Steve K showed me a cool setup -> he got himself a USB->VGA convertor (some 70 bucks), plugged the 2430 to a USBhub, plugged in a USB keyboard, and vroom.. he is all set.. lovely.. all those stuff.. throw them out.. just provide tons of USB ports ;)

Played arnd with my Dell inspiron 9300 laptop.. I had been lazy to update linux to new WPA-PSK key at home.. just finished it..along with a bunch of other stuff.. (I got a debian etch now): WPA apt-get install wpa_supplicant then installed /etc/network/interfaces----> # The loopback network interface auto lo iface lo inet loopback address 127.0.0.1 netmask 255.0.0.0 # eth0-wired The primary network interface iface eth0 inet dhcp #Wireless - eth1 auto eth1 iface eth1 inet dhcp wpa-driver wext wpa-roam /etc/wpa_supplicant/wpa_supplicant.conf Minimal /etc/wpa_supplicant.conf to associate with open ctrl_interface=/var/run/wpa_supplicant ctrl_interface_group=0 eapol_version=1 ap_scan=1 fast_reauth=1 ### Home network network={ ssid="MyNetworkSSID" key_mgmt=WPA-PSK psk="MyNetworkPASSKEY" } Then ofcourse, i tried playing around with this: http://www.beryl-project.org/ installed and got it running.. but for some reasons.. it is to...

If you already know how to build a filesystem, good else for OMAP2430, here it is: http://www.geocities.com/crecmca98/Linux/omap2430_From_almost_scratch.pdf lmsensor for OMAP has been a slight pain so far.. lmsensor needs sysfsutils to build properly.. pick them up here: http://sourceforge.net/project/showfiles.php?group_id=44427 ./configure --target=arm-linux --build=i386-pc-linux-gnu --prefix=/home/fs/nmenon/arm/ --host=arm-linux CC=arm-none-linux-gnueabi-gcc mymake install lmsensor: Download it from here: http://www.lm-sensors.org/wiki/Download compiling this is a pain: first I modified Makefile, changed I2C_HEADERS := $(LINUX_HEADERS) WARN := 1 DEBUG := 1 CC := $(CROSS_COMPILE)gcc the last one was not sensible... but personally, I hate seeing CC=gcc in Makefiles... might have been good to have ETCDIR := $(PREFIX)/etc Next compilation started mymake LINUX=/home/nmenon/opensource/copy/linux-omap.git/ MACHINE=ARM CFLAGS="-I/home/fs/nmenon/arm/include/" V=1 PREFIX=/home/fs/...

some interesting questions put up by shardul in the ARM orkut forum: "what is 'it' that actually affects power consumed by a procesor...?the bus traffic?internal register altrations?memory reads / cache interaction?von-numann- / harvard- base?what is the factor that differentiatessome processor designs (like ARM etc.) to be power-efficientcompared to other processors (like x86) ?" What indeed.. I'd say CPU Speed is a factor.. but that just means all the gates in the processor run at a lesser speed.. hence the per gate consumption is lower. but really.. what is the architecture factor that gives the edge for ARM?? I am keeping a close watch on this discussion..

An out of context post, but sad day today, the manager I most respected in TI left..

--TI's CEO sees new chip drivers --The semiconductor market may be mature and feeling the brunt of aslowdown. But there are still many new and untapped markets in the ICarena, namely in medical, video, among others, said Richard Templeton,president and chief executive at Texas Instruments. http://newsletter.eetimes.com/cgi-bin4/DM/y/e5Mi0FyTZ10FrK0E67y0Ec --ARM tips plans for next-generation processor--ARM Holdings plc, the licensor of processors and related intellectualproperty, has tipped sparse details of plans for a next-generationprocessor a follow on to the three processor cores the comprise theCortex series. http://newsletter.eetimes.com/cgi-bin4/DM/y/e5MV0FyTZ10FrX0E64b0A1 --1.2 percent dip in chip sales 'reflect seasonal decline'--Sales of semiconductors in January were worth $21.47 billion inJanuary, up 9.2 percent on the corresponding month in 2006, accordingto the Semiconductor Industry Association. However, this represents aseasonal decline of 1.2 percent from ...

Object oriented programming speaks of Objects and classes. Classes have operations and attributes.. How can we exploit this for Embedded baseport environment? Operations: read/write/ioctls etc Attributes: variants accross boards and silicons: such as base address, Chip selects etc.. Values: the initializing value for the driver creating this.. benefits: portability.. linux already has something of this sorts: resource structure.. only, it can be extended to a larger extent. OMAP has IP Blocks.. such as i2c controllers, DMA4 controllers etc... we can use the above concept to create "in theory" a highly portable model.

(I am not in general a technology prediction guru.. but here is a interesting another orkut driven discussion: "do we use ARM processor in Embedded Systems? We have already 8085 & 8086 processors. Then why do we go for it?" Some answers: power. Multiprocessor (DSP+ARM), more address space, IO, ISA, power again.. here is one set: 1) speed/MIPS.2) co-processor support3) No. of I/O s4) the peripherals supported by both the processors5) amount of addressable memory6) security options now, my view:(note, I have been a ARM guy through out my professional life). Have ya all heard of Intel's SOC with Pentium? if u guys been following news of intel selling off its Xscale processor family to Marvin, u guys might be aware of this.. dont yet sign off Intel from the match.. with it comes a BIG set of apps... http://www.linuxdevices.com/news/NS2460545479.html and ofcourse.. two things to keep in mind: the next gen 35nm phreus or what ever new hafnium based chips will boost intel a...

was in orkut today and hit this neat document on ARM intro.. recomended for those starters: Ajay 11/20/06 Hi Pankaj, not sure why - something wrong with Rapidshare. I have uploaded it to another location as well: http://ajaydudani.googlepages.com/Programming_the_ARM_Microprocessor_for_Embedded_Systems.pdf Link to the community: http://www.orkut.com/CommMsgs.aspx?cmm=5818698&tid=2499023171415881585&start=1 From another post: http://www.heyrick.co.uk/assembler/

heh heh.. forget abt mot support.. there are "not so lazy" blokes arnd http://www.motorolafans.com/MotorolaFansPHPbb/viewforum.php?f=21 and orkut has a smattering of moto ming hackers(just search for A1200)... mebbe next post might be - I dommed my cell :(( heh heh... calvin, hobbes.. here I comeeeee...

Phew... moved most of the tech posts of old livejournal here.. hopefully will be useful for someone... sometime..wish it was easier though.... am back from Indian vacation and did post a SDP2430 bring up document here: http://www.geocities.com/crecmca98/Linux/omap2430_From_almost_scratch.pdf Also decided to move to orkut and linkedIn.. just to meetup with like minded folks :)