My experiments with Linux

Moving to wordpress

2011-08-29T21:34:00.000+05:30

I revamped my homepage by moving it to a hosting provider and migrating to joomla. Fortunetly or unfortunately, Joomla has very good plugins for wordpress and not for blogspot. These plugins allow me to post at my wordpress blog while updates will be made available instantaneously at my home page. I have pulled all the content from flyingtux.blogspot.com to yetanothertux.wordpress.com. Do continue to visit me at

http://yetanothertux.wordpress.com
http://sarin.net.in

QT keyboard events

2011-06-24T12:40:00.000+05:30

I am here to make a quick post. I searched for this at many places but could not get the exact solutions. Every message I could read sounded cryptic.

Scenario:
I want to generate fake keyboard events and send to QWebView class.

Solution:
Extend this class to have slots for signals. Connect it to a signal that can pass an ascii character.

Issue:
Keyboard does not have 'a' and 'A'. It only has an 'A'. That means, I cannot directly send any lower case characters using the event (Which is not correct)

Solution:
The solution is simple. The constructor of QKeyEvent can take QString as one of its argument. The signature is as follows:

QKeyEvent ( Type type, int key, Qt::KeyboardModifiers modifiers, const QString & text = QString(), bool autorep = false, ushort count = 1 )

Even if we don't pass the QString, it will work. However, by passing the QString, we can control exactly what will be displayed.
This is how I create my QKeyEvent:

void MyWebView::rcvKeys(int k)
{

char str[2];
str[0]=k;
str[1]='\0';

QKeyEvent *k=new QKeyEvent(QEvent::KeyPress,k,Qt::NoModifier,QtString(str),false,0)

//-----------Pass it to postevent etc
}
}

Google fined USD 5 million for using patented code in Linux kernel

2011-04-24T21:50:00.001+05:30

http://www.bbc.co.uk/news/technology-13168296

Very interesting news. I am still not really sure if this is part of mainline. However, considering the fact that the holder was able to analyse the code and ascertain that it was there in the Google's server's kernel indicates it was surely there on the public domain whether it was a custom piece of code written by Google engineers or not. This is surely something very interesting thing to happen and it would be very interesting to see if Google takes it beyond district court and what exactly happens to this case at the end

/. link: http://linux.slashdot.org/story/11/04/21/2140249/Google-Loses-Bedrock-Suit-All-Linux-May-Infringe

Build error on Android rowboat

2011-04-13T17:39:00.000+05:30

I was building the code for TI omap3evm and I found that on FC14, I was getting the error from the make file of kernel. It told "Mixed implicit and normal rules". On a bit of googling, I found http://www.mail-archive.com/bug-make@gnu.org/msg06220.html

The file "hardware/ti/sgx/GFX_Linux_KM/eurasiacon/build/linux/omap3630_android/kbuild/../../kbuild/Makefile.kbuild" need to be modified as follows:

Change "all %:" to
all:
(Build command)
%::
(Build command)

Fips opessl

2011-03-06T00:25:00.000+05:30

http://olex.openlogic.com/packages/opensslfips

I was cross compiling this. The strangest thing about this is that it cannot be cross compiled. Because, they generate programs, run them, get signature from the programs and use that signature to compile them again!

I did that by little bit effort. What I am now looking at is the possibility of having it compiled inside qemu. Anyway, this is what I did
1. Configure for linux-ppc with compiler, prefix and ranlib specified during this step
2. Find and replace ar using powerpc ar in all makefiles
3. Make and when it breaks during sha1 calculation, replace the binary that does sha1 caculation with an x86 equivalent and complete this step
4. Make again and when it tries to execute openssl/tests, replace the necessary statements in fipsld with read function and supply the signatures for each program after taking them to target and running them there once
Well, another issue that I noticed with this package is that it does not behave well with parallel builds. I will post here again if I can compile this in Qemu. I am planning to use FC12 ARM/PPC for this.

Middle men

2011-02-19T21:09:00.000+05:30

http://www.pcworld.com/article/219776/rim_nokia_docomo_dodge_googles_dumb_pipe_menace.html

OK, losers consortium has taken shape. But, what are they driving at? The whole discussion is centered around how to enable the service provider to be the middle man and extort money from poor end user. With what face is Nokia and RIM going to face end customers now? A smart phone today costs 20-30k (INR). After taking such huge amounts from end customer, they want to create means for the letting the service provider loot the customer! Shame on them.

May be your service provider is subsidizing your phone. However, that reminds me about the phones (LG-LSI110) that Reliance used to sell for 3K! I am pretty much sure that it didn't have electronics and plastics worth even Rs. 1000. However, the phone came with the note that its cost is 10K and it is owned by Reliance. This was about 7 years back. I doubt there will be any change today. Anyway, would you prefer to get a smart phone at a subsidized cost when you pretty well know that this amount will be recovered from you through monopolistic practices of the service provider?

Well, I recently used a Galaxy tab to download and test an application (ZXing). The ability to customize your smart phone by pushing an apk to it after enabling USB debugging was probably the only reason needed for paying 30K for that device. Well, Nokia, go back and make some good Android phones and I will buy them (Yes, I would prefer to buy android from Nokia over any other vendor) when they become available.

Mozilla Dash

2011-01-30T02:33:00.000+05:30

I am on FF 4 beta 10 and I installed this addon. It was highly hyped at many sites. However, it was a bit of disappointment.
Why?

It is not intuitive enough

Even after being a heavy internet user, I had to struggle to get back my tabs

It does not show the url

This might be a good thing that you can search and find links. But, how do I make sure that the search result is correct and I am on the expected link? This is a problem especially when I am on sites that do financial transactions.

The UI was not up to the mark. I could see that the fonts of menu items were cut at bottom half

Well, so, whats good about it? Its still experimental. We can hope that it will get improved :)

.PHONY

2011-01-24T11:23:00.000+05:30

For a long time, I used to look at .PHONY targets in makefiles and wonder how useful those targets were. Today, I found the usefulness while playing with prelink (more on prelink later).

I created the following files and folders while doing this:

[root@sarin prelink]#ls -l
total 20
drwxr-xr-x 2 root root 4096 2011-01-24 11:14 bin
drwxr-xr-x 2 root root 4096 2011-01-24 11:14 lib
-rw-r--r-- 1 root root   57 2011-01-24 10:41 libutil.c
-rw-r--r-- 1 root root 273 2011-01-24 11:14 Makefile
-rw-r--r-- 1 root root 343 2011-01-24 10:56 prelink.c
[root@sarin prelink]#cat Makefile
all: bin

lib: libutil.c
        gcc -shared -Wl,-soname,libutil.so -o lib/libutil.so libutil.c

bin: lib prelink.c
         gcc -L/root/c/prelink/lib prelink.c -Wl,-rpath,lib -lpthread -lutil -o bin/prog

clean:
        rm -f lib/libutil.so bin/prog

Now, I did a make clean and did a make. Nothing happened. I got the message that "make: Nothing to be done for `all'.". If you notice, in the make file, target all had a dependency on 'bin'. Target 'bin' was dependent on target 'lib'. So, I was expecting that lib will be built first and then bin. But, It did not even try to build bin (I verified this by putting an echo in bin). After a long struggle, I found the reason.

As you can see, 'bin' and 'lib' were the names of two folders inside the current directory. So, make assumed that targets 'bin' and 'lib' where those folders. As these sub-folders were already present in the current folder, it assumed that there is nothing to be done for target all.

This is the issue that .PHONY target fixes. It clearly tells make to avoid looking at files and folders for these names. I added the following line to the top of the makefile.

.PHONY: all bin lib

Thats it, I was able to compile the program using my makefile!

Kernel oops

2011-01-21T14:02:00.000+05:30

I had to look into a kernel oops recently. This was produced on a ARM machine. It was relatively easy to understand this message. These messages have become much more simple to analyse these days. I am making this quick post to remember one register: LR

LR: This register holds the address to which PC will be set to once you hit return. This simply means, this is the caller of the function where the trouble occurred.

The point where failure happened can be seen by looking at PC. Also, SP points at the current stack location. When I scrolled down, I could see some good back tracing information too.

plymouth

2011-01-13T17:30:00.002+05:30

I have been looking for the source of rhgb and I was not able to find this source. Today, I somehow wanted to find this out and spent some time with yum. Eventually I found that this comes with plymouth (for Fedora-12). I am yet to find out how these two are related. I could see this:

1. There is a /usr/share/plymouth folder
2. The themes are present in this folder
3. There is a default.plymouth file which tells which is the default theme
4. plymouth uses png files to create animation
5. There is a text theme that can be (seems to be) altered easily by changing the ply-text-display.c (Needs to verify this)

Well, more on this after I play with this a bit more. :)

Update-1: Plymouth starts from initrd itself. So, modifying the stuff on the root file system will not help much. It has to go to initrd. I was able to change default.plymouth link in initrd and change from charge theme to text theme without much issue.
Update-2: While working with initrd, I found that fedora is now using darcut instead of old initrd. Will write about it later when I get some time to study it.

Update:
I tried modifying and recompiling the plymouth (version 0.8.0) and tried it on Fedora 12. The file that was modified was ply-text-progress-bar.c. The file can be found in src/libplybootsplash folder inside the plymouth package. I changed two functions get_os_string and ply_text_progress_bar_draw. The first function is very custom made function to read /etc/system-release file and get the OS name and version (Fedora 12). So, though modifying /etc/system-release might work in general, it is simpler to modify the function itself. The second function decides the colors of the progress bar and also the background and foreground color of the OS string. I modified the colors there so that the effect can be seen easily.

The code was compiled and I found that the o/p is present in src/libplybootsplash/.libs/libplybootsplash.so.2.0.0. This was copied to /usr/lib of initrd and I recreated initrd using this command
find . | cpio -H newc -o > ../ply-test.img. Later I gziped it and moved it to /boot (gzip -9 ply-test.img && mv ply-test.img.gz /boot/ply-test.img). I also modified grub config to create a new entry with this initrd. Once I rebooted, I could see the changes in the plymouth screen.

I am currently thinking if I should use this or I should use some graphical theme. May be changing a few functions, I can change the logic of drawing this progress bar and draw some other pattern. But, for changing a graphical theme, I may only have to change the images and that might have a better effect too. Anyway, plymouth seems to be in control and I will now move to something else. I want to revisit the qemu as I found that it is taking about 10 times more resources compared to vmware. I will see if using the qemu from android project gives me any advantage or not.

Compiling qemu on windows

2011-01-07T17:32:00.001+05:30

I started doing this today by following this page. However, I was always getting this error that config-all-devices.mak not found. I googled and found some patches. However, something told me that this is not a patch issue. So, I decided to look again and then saw this funny message "-uThe system cannot find the file specified."

This was actually getting caused due to incorrect sort being used. I had the windows programs in my path and a 'which sort' clearly told me what was going wrong.

bash-3.2$ which sort
/c/WINDOWS/system32/sort

A change in path declaration fixed this issue

PATH=/home/lassauge/MyDocuments/Qemu/libusb-win32-device-bin-0.1.12.2/bin:$PATH CFLAGS="-O4 -march=i686" \

to

PATH=../libusb-win32-device-bin-0.1.12.2/bin:/usr/bin:$PATH CFLAGS="-O4 -march=i686" \

Anyway, I have not completed the build. Currently I am getting some usb library related errors. Will update this post once I completely compile this.

Edit:

Completed this build. Though the above fix removed the error, it did not create the proper config-all-devices.mak. So, I had to manually create this by running the appropriate command

cat i386-softmmu/config-devices.mak | grep =y | sort -u > config-all-devices.mak

Again, I got the error that isa vga device was not created. So, again, I manually entered the vga-isa.so as a dependency for target "all" in the Makefile.hw.
Also, it did not pickup the usb library include folders correctly. To fix this, I added the following to config-host.mak
CFLAGS=-O2 -g -O4 -march=i686 -I../libusb-win32-device-bin-0.1.12.2/include

Another issue was the fmod libraries needed absolute path. I changed that also in the configure script. However, usb libraries worked fine with relative path from qemu folder. I think, fmod would have got linked from a sub-folder inside qemu where as the usb lib would have got used from the qemu folder itself. Anyway, I am running the qemu compiled my me :)

CES 2011

2011-01-06T14:54:00.000+05:30

Smart phones -- read Androids :) --, tablets are ruling CES 2011.

http://twitter.com/search?q=%23ces

Not that easy to watch it, hundreds of updates are coming per minute.

fossdotin

2010-12-18T08:44:00.000+05:30

I was watching #fossdotin till late into the night yesterday. I usually don't like/do social networking much. But, I could not resist this one. The channel showed a mixture of energy, enthusiasm, nostalgia and anguish. Kudos to all those who have made this happen.

Though I left early yesterday, I don't feel bad about it. There was surely a certain pleasure about being able to create that electrified environment in my mind sitting all alone in my room with my eyes glued to monitor and fossdotin channel on twitter. Such thought streams are usually far from the imperfections of reality and hence stronger and sweeter.

I saw the picture of team FOSS.IN in one of the tweets. My immediate thought was "most of them are kids". But, what they have achieved is tremendous. Probably the reason why they could achieve all these was that they were "kids" - People with Free and Open minds. Hats off to the this team and their leader who is now really proven to be "Atul".

I saw a few tweets about rekindling FOSS.IN again next year. I think that should only be done with the full consent of Atul Chitnis. This person has become the hero of the event and hijacking the event from him (for whatever reason it may be) is simply going to be cruel. After all, you can start your own FOSS event. Isn't that what the "Free as in Freedom" signifies?

QEMU/ARM hands-on (FOSS.IN)

2010-12-17T18:40:00.002+05:30

https://github.com/tuxdna/arm-experiments (Saleem Ansari)

This was the most interesting and useful session that I had in entire FOSS.IN. Above site will give details. However, I guess, there are some issues with instructions. For example, the kernel link is broken. But, it will work if you remove the .af from the url. Also, the image did not load the rootfs. I tried ext2fs instead of cpio and gzip and that loaded the image. Also, rcS is not present at the site. I wrote that from memory. Again, I seem to have missed something there. It fails with the message that "could not open initial console". I guess /dev/ttyX stuff might be missing. Will try to fix that later and post back here once I do that. Anyway, this was a enlightening talk and it now gives me enough confidence to try my own rootfs on different architecture. Many thanks to the presenter.

Update:
I fixed this stuff. For those who are interested, I am listing the important changes
0. The correct link for kernel is: http://www.kernel.org/pub/linux/kernel/v2.6/testing/v2.6.24/linux-2.6.24-rc7.tar.gz
0.1 May be this was not needed. But, I did create a symlink from busybox to /init. Also, it only worked when I did a static build of busybox
1. Some device inodes were missing. Created /dev/{null,console,tty,tty0,tty1,tty2,tty3,tty4,tty5}
2. Created /etc/init.d/rcS and made it executable

#!/bin/sh
mount -t proc proc /proc
mount -t sysfs sys /sys
/sbin/mdev -s

3. Most important of all, EABI support needs to be turned on in kernel. Else, nothing will work. Turn it on before compiling the kernel
4. The filesystem cannot be cpio. May be it can still work with a kernel config change. However, I tried with ext2fs and it worked perfectly well
4.1 To create a ext2fs,

dd if=/dev/zero of=root.fs bs=1M count=4
(As a side note, if you refer to the commands given in the above link, you will see bs=1. That is very slow)

mkfs -t ext2 root.fs
(Press 'y' for the question about root.fs not being a real device)

mkdir rootfs
mount -o loop root.fs rootfs
(Mount your new filesystem somewhere so that you can copy the contents to it)

cd $ROOTFS_PATH
(go to where your rootfs is)

tar c * | tar x -C /path/to/rootfs
(copy the contents to new your new filesystem)

umount /path/to/rootfs
(Now you can start using your root.fs just like rootfs.cpio.gz as described in the link at the top of this post and continue from mkimage part)

Thats all! you will have a bootable setup. I am thinking of creating a small script that will automate all these. Will post it here if I ever do that.
Update: I did make a script. Unfortunately, I was in office and our security policy does not allow me to upload that. Its like magic, u run it, have a coffee and when you come back, you will see qemu running your own small Linux image :)

mediawiki (FOSS.IN miniconf)

2010-12-17T17:58:00.001+05:30

Today I attended a miniconf on mediawiki that took place on the sidelines of FOSS.IN. It was a very interesting talk by Erik Moeller. During the talk Erik kept making notes using etherpad (http://eiximenis.wikimedia.org/MediaWikiNotes). I have copied them below. It was really amazing to know about the kind of effort that goes into making these things happen (which sadly the user community, that includes me, takes for granted). It was also nice to know about a bunch of allied technologies and a number of new enhancements that are coming to mediawiki. I was really fascinated by the new enhanced editor and I downloaded and installed the svn trunk when the talk was on to get a feel of the interface. I will be soon upgrading the wiki at office.

-------------------------------------------------------------------------------

Mediawiki Workshop 2010

Erik Moeller

FOSS.IN 2010

[Miniconf]

Mailing list for developers: http://lists.wikimedia.org/mailman/listinfo/wikitech-l

Manual about how to write a MediaWiki extension: http://www.mediawiki.org/wiki/Manual:Extensions

Subversion check-out instructions: http://www.mediawiki.org/wiki/Subversion

Get commit access: http://www.mediawiki.org/wiki/Commit_accesss

Wikimedia configuration files for MediaWiki: http://noc.wikimedia.org/conf/

Wikipedia's configurations: http://en.wikipedia.org/wiki/Special:Version

IRC: irc.freenode.net #mediawiki #wikimedia-tech

http://dbpedia.org DBPedia - extract info from Wikipedia

SPARQL - RDF extraction for DBPedia

http://semantic-mediawiki.org Semantic Media Wiki - semantic storage for MediaWiki

example:bases wiki http://baseswiki.org

http://baseswiki.org Info about non-judicial grievance mechanisms

http://svn.wikimedia.org/ Wikimedia Subversion Repository

how to enable the new toolbar of wikipedia in my mediawiki? WikiEditor extension is the answer and it works if we change the user preference.

browser toolbar: Go to https://addons.mozilla.org/firefox/addon/6401 click add to Firefox and then after 5 secs.addon is installed.

http://en.wikipedia.org/wiki/Main_Page/?useskin=monobook : old theme of wikipedia (new version: Vector)

http://www.wikia.com -wysiwyg ediing

http://www.wikia.com/Special:LandingPage?uselang=

http://harrypotter.wikia.com/wiki/Main_Page

http://uncyclopedia.wikia.org : the sarcastic wikipedia

http://svn.wikia-code.com/

http://www.dbpedia.org/UseCases

http://atlas.gnowledge.org

View Your Mind - a better alternative for mind mapping than freemind perhaps

http://en.wikipedia.org/wiki/Wikipedia:Community_Portal/Opentask

http://en.wikipedia.org/wiki/Wikipedia:WikiProject

http://outreach.wikimedia.org/wiki/Bookshelf/

http://en.wikipedia.org/wiki/Wikipedia:WikiProject_Screencast

http://en.wikipedia.org/wiki/Wikipedia:Notability

http://en.wikipedia.org/wiki/Wikipedia:BLP -- Biographies Guidelines

http://en.wikipedia.org/wiki/Wikipedia:Tools

http://en.wikipedia.org/wiki/Wikipedia:Twinkle - tool for friendly reverts

http://grouplens.org/ conducts research on user interaction

http://en.m.wikipedia.org - Mobile version

http://commons.wikimedia.org/wiki/Special:UploadWizard - do login first!

http://strategy.wikimedia.org/wiki/Proposal:Media_review

http://wikibhasha.com/ a conversion utility where articles may be easily converted into indic languages

http://www.openstreetmap.org/ an open source map( edit and contribute :) )

http://liquidthreads.labs.wikimedia.org/wiki/Talk:LiquidThreads_testing

http://en.wikipedia.org/wiki/Wikipedia:Pending_changes

http://translatewiki.net - localizatioin translation communities, language communities, and free and open source projects.

http://stats.wikimedia.org --

http://download.wikipedia.org

http://www.openwebanalytics.com/ alternate to google analytics

Flash plugin on 64 bit Chome OS running on Fedora 13 64 bit

2010-12-11T20:50:00.000+05:30

I installed chrome and youtube.com told me that I don't have flash installed
On going to adobe flash page, they told me that it has to be there as I am using chrome and I just need to enable it
about:plugins on chrome told me that it is not there :(
Now, what do I do? The answer follows

1. Download the 64 bit flash plugin preview from adobe (See my previous post)
2. Copy libflashplayer.so to a preferred location (I copied it to /usr/lib64/mozilla/plugins)
3. Edit ~/.config/google-chrome/Default/Preferences file and add the section shown in reverse video

4. Restart chrome and don't mind about it complaining about issues with config getting corrupted. You will see new entry for flash when you do about:plugins

5. Just go to youtube and start watching your favourite videos :)

64 bit flash preview

2010-12-09T20:25:00.000+05:30

Adobe is back with 64 bit flash for Linux. This is a preview and grab it before they close it again. Last time when they had tried this, there were a number of security issues and preview was pulled down soon.

Installation:

Download the preview version from Adobe labs site
Untar the downloaded plugin
Copy the resulting plugin to your firefox plugin folder

I had to first create this: mkdir /usr/lib64/firefox-3.6/plugins
Copy libflashplayer.so to that folder

Just re-start firefox and you are ready to browse the flash sites :)

Download link
http://labs.adobe.com/downloads/flashplayer10_square.html

(Its not easy to find the link. It took me 10 minutes to find this. Though there is a get beta note on the download page, it does not take you directly here)

Silence the console

2010-11-20T07:23:00.000+05:30

Well, at times, its very much annoying to hear the console beep from a Linux terminal. Sometimes, it is very untimely. For example, you might be stuck at some command and might press a tab. If there is no possible completion, terminal beeps. Its more like your leg is stuck between tracks when the train is approaching and you are desperately trying to escape. Then the guard standing near you blows his whistle and asks you to get of from the track fast, without trying to help you. Anyway, Here is how you disable it

For text terminals

setterm -blength 0

For graphical terminals

xset b off

And u have peace....

Chrome with built-in support for pdf

2010-11-07T07:20:00.000+05:30

http://chrome.blogspot.com/2010/11/pdf-goodness-in-chrome.html
Well, I can see cloud printing really picking up. PDF is the only supported mime type for cloud printing via Google cloud. Google had earlier mentioned that it expects pdf to be De facto standard for printing documents over cloud. With their Chrome OS release expected only a few days from now, I could not see much movement in this front. But, I guess, Google's plans are very clear now. With this beta release of Chrome, others will have no way other than support pdf extensively on web. I like Google's way of doing things :)

Fedora 14 on its way!

2010-10-24T11:20:00.002+05:30

Has Linux lost the Desktop war?

2010-10-20T22:37:00.001+05:30

http://www.techeye.net/software/linuxs-chance-has-gone
http://www.pcworld.com/businesscenter/article/207999/desktop_linux_the_dream_is_dead.html?tk=hp_new

Those are the two interesting links that I came across recently. They make the statement (rather convincingly) that the Linux has lost in the Desktop front. I look at it from a different perspective. I would say that Desktop has lost its place as a PC (personal computer).

Today, we can clearly see that Desktops are rapidly getting replaced with smart phones and MIDs. And, what are the two most interesting players in that domain? Though a subject of argument, Android is surely one of the forerunners for the title. Next, look at the cloud computing. Though in RC stage, Google chrome OS has made generated enough buzz among the tech community. I am sure that with the weight of Google behind these two distributions, Linux will stay close to you though the Desktop might not!

http://www.documentfoundation.org

2010-10-07T23:19:00.000+05:30

http://www.documentfoundation.org

Its only today I saw this. I am currently downloading this. Good move considering the way Oracle is going now. I am just waiting to see bytecodelangfoundation.org :)

SCU on ARM11MPCore will not give you coherency

2010-09-26T19:24:00.000+05:30

Well, we thought otherwise. But, that was not correct. This was a very interesting experience. We started writing the communication module to make two different cores of ARM11MPCore communicate and found that things are not going fine as we were expecting them to. We write from one core and we cannot read it back from other. First suspect was the new virtual addresses that had happened in VxWorks. However, that was not the issue. After reading about it a bit, we were convinced that it must have been a linear mapping in our case. Now, what was the reason? The next suspect was the cache. However, the process was not simple. No one except me in the whole group believed that it can be caused by cache. Also, I myself was doubtful as I felt SCU should take care of it.

To clear our doubt, we talked to a manager in our company who was an expert in JTAG. He appeared completely assured of the fact that we are doing something dumb with JTAG as we cannot see this memory even in JTAG. When I asked about cache coherency issues, he brushed aside my doubts saying that it can never happen.

Well, after giving the advice, he walked away without turning back. But, I was the tech lead for this project and I had to get this working somehow. And, something kept telling me that this is a cache issue. So, I decided to keep hammering at that point. Now, we were writing from VxWorks and reading from Linux. I decided to look for code which will flush cache. But, being a noob in VxWorks, I could not find this code. We tried some silly tricks in the beginning like increasing the memory size in the hope that cache will get flushed auto-magically. Fortunately or unfortunately, these tricks did not work. However, one guy in our team was able to dig out VxWorks function to flush cache. It was a simple function that took the cache type, address and length as the arguments and it did its job! (Yes that was simple compared to what I found for Linux. We had to get physical address of the memory, get page structure and then flush the page from cache!) Yes, after flushing the cache, we were able to see what was being written by the second core into the memory location from the code that was running in first core. Later, another member in my team found that SCU will not do this in case of ARM11MPCore. It does this in case of some other newer chips (Cortex-A9). We also read that allocating DMA memory might do the flushing directly. We are yet to try this.

Linux IPI handling for ARM11MPCore (ASMP)

2010-09-26T18:49:00.000+05:30

We have been struggling for a while with this. Last Friday, I nailed him down.Ok, the background story...

We knew that IPI (Inter Processor Interrupt) can be used to communicate with other processor. Linux has support for this in SMP mode. However, we are doing ASMP. We wanted to use it slightly differently to device a custom communication module with another core running VxWorks. Now, the first hurdle came in the form of not being able to write to some registers. That was a dumb mistake of not ioremapping the register addresses. After that we faced another problem of not being able to register an interrupt. After a bit of debugging, we found that the reason is that there is no chip descriptor registered for interrupts 1-16 (They are the interrupt numbers for IPI). Well, the init code for interrupt was doing it for interrupts >= 29. But why?

It is this question that forced us to spend a week to resolve. There was one guy working on it continuously and I will break my head whenever I get time. Now, we tried the most obvious. We changed 29 to 1 in init code for interrupts. This enabled us to register the interrupt. However, this did not cause our interrupt function to be called when we wrote to the register. Last Friday, I decided to look at realview code. And it paid off. I could see that, they are also registering the interrupt from 29 onwards in the initialization code. Now, what was the problem? To know that, I had to delve much deeper. The answer came from the low level vector interface routines. The do_IRQ was not responsible for handling IPI. Instead, there was a do_IPI function. We immediately reverted the changes done in the irq initialization code. Then, we put some code into do_IPI

Well, thats it! our interrupts started working. We did face some issues while writing to the shared memory. All about that in another post.

$35 tablet

2010-09-12T00:46:00.000+05:30

Well, today I saw this news about $35 tablet

http://androidos.in/2010/09/the-truth-about-35-android-tablet-from-indian-government/

As per that page, governments claim of developing a tablet here in India is false. The blogger claims that the tablet is imported from China. But, the government claims that it is developed by a student in Vellore. Now, have a closer look at the Chinese one

If you see that closely, this device is based on ARM11. It has 256MB RAM. Well, I have been looking for the specs of the Indian tablet. Guess what I found:

http://armdevices.net/2010/07/23/indias-35-tablet-is-based-on-my-video/

The device seen there is very much speculated to be the Indian device that is priced at $35. However, that seem to be something made by a private company and not by universities or government. More details of this can be found here: http://www.allgosystems.com/html/android_stamp_i.mx233.htm

If that were the device which Mr. Sibal was talking about, then it is pretty clear that there is a difference in the processor itself (this one uses ARM9, i.MX233) and they are indeed different apart from the plastic out side. Sadly, I could not find more details on the stuff Mr. Sibal was talking about. The best I could get was this: http://www.engadget.com/2010/08/11/35-tablet-makes-an-appearance-on-indian-tv-video/

Well, now I am confused. What is the spec of the stuff that government is going to bring out? Who is this AllGo guy? Did the student after making it at campus go and join this AllGo? Was this a reference design that Freescale or someone like that floated and everyone copied? Did the student who designed it, copy the stuff from somewhere else?

The government tablet seems to have 3G support and 2G RAM. The AllGo one does not have 3G and the RAM is only 128MB. Well, I really doubt 2G RAM is a mistake. Else that itself will cost > $35. May be its 2GB flash.Well, I think I will wait for a while till more information is available.