Knuth's recent interview

Andrew Binstock recently interviewed Knuth. I would highly recommend reading the interview. It was really sad to find out that Knuth has been diagnosed with Prostate Cancer, but here are some my favourite quotations from the interview

"The success of open source code is perhaps the only thing in the computer field that hasn’t surprised me during the past several decades. But it still hasn’t reached its full potential; I believe that open-source programs will begin to be completely dominant as the economy moves more and more from products towards services, and as more and more volunteers arise to improve the code

For example, open-source code can produce thousands of binaries, tuned perfectly to the configurations of individual users, whereas commercial software usually will exist in only a few versions. A generic binary executable file must include things like inefficient "sync" instructions that are totally inappropriate for many installations; such wastage goes away when the source code is highly configurable. This should be a huge win for open source.

Yet I think that a few programs, such as Adobe Photoshop, will always be superior to competitors like the Gimp—for some reason, I really don’t know why! I’m quite willing to pay good money for really good software, if I believe that it has been produced by the best programmers."

"I might as well flame a bit about my personal unhappiness with the current trend toward multicore architecture. To me, it looks more or less like the hardware designers have run out of ideas, and that they’re trying to pass the blame for the future demise of Moore’s Law to the software writers by giving us machines that work faster only on a few key benchmarks! I won’t be surprised at all if the whole multithreading idea turns out to be a flop, worse than the "Titanium" approach that was supposed to be so terrific—until it turned out that the wished-for compilers were basically impossible to write."

"The machine I use today has dual processors. I get to use them both only when I’m running two independent jobs at the same time; that’s nice, but it happens only a few minutes every week. If I had four processors, or eight, or more, I still wouldn’t be any better off, considering the kind of work I do—even though I’m using my computer almost every day during most of the day.

So why should I be so happy about the future that hardware vendors promise? They think a magic bullet will come along to make multicores speed up my kind of work; I think it’s a pipe dream. (No—that’s the wrong metaphor! "Pipelines" actually work for me, but threads don’t. Maybe the word I want is "bubble.") From the opposite point of view, I do grant that web browsing probably will get better with multicores"

"On a positive note, I’ve been pleased to discover that the conventions of CWEB are already standard equipment within preinstalled software such as Makefiles, when I get off-the-shelf Linux these days."

"I currently use Ubuntu Linux, on a standalone laptop—it has no Internet connection. I occasionally carry flash memory drives between this machine and the Macs that I use for network surfing and graphics; but I trust my family jewels only to Linux. Incidentally, with Linux I much prefer the keyboard focus that I can get with classic FVWM to the GNOME and KDE environments that other people seem to like better. To each his own."

"let me just say that almost everything I’ve ever heard associated with the term "extreme programming" sounds like exactly the wrong way to go...with one exception. The exception is the idea of working in teams and reading each other’s code. That idea is crucial, and it might even mask out all the terrible aspects of extreme programming that alarm me."

"I also must confess to a strong bias against the fashion for reusable code. To me, "re-editable code" is much, much better than an untouchable black box or toolkit. I could go on and on about this. If you’re totally convinced that reusable code is wonderful, I probably won’t be able to sway you anyway, but you’ll never convince me that reusable code isn’t mostly a menace."

Atomic access

I was involved in a set of emails exchanged about per CPU data and atomic access. The argument was basically this

"If you access 64 bit data on a 32 bit machine, what part of the access is atomic?". One would assume that with per-CPU data, it is safe to read the data from a different CPU". The answer lies hidden in the processor architecture. Several RISC architectures require certain types of data to be aligned to specific boundaries (this in turn ensures that the access is atomic). Usually processor manuals suggest that data that is contained within a single cache line is likely to be fetched atomically. Ideally, it is recommended that the data be aligned to the boundary of it's size, so that access is atomic. To explain that further

16 bit data be aligned to a 16 bit boundary
32 bit data be aligned to a 32 bit boundary
64 bit data be aligned to a 64 bit boundary

So the next time you think of atomic access, data sizes, per CPU data, look at your processor manual and determine if the data needs to be protected when accessed from a different CPU.

RSS feed is fixed

The RSS feed has been changed to export full content instead of the earlier partial content. The blog should be easier and more fun to read with an RSS reader now.

Feeling left out of the software world

It seems to me that there is a hard barrier between the software development world and the systems programming world. I feel left out of the developments in software that have taken the world over. Agile/Lean programming, EJB, J2EE, .NET, .NET Remoting, JSP, JSTL, Web containers, EJB, MVC, Spring, OpenID, SSO, Indigo, WCF, WPF, Aspect Oriented Programming, Web services, CSS, XML, XSLT, XQuery, ADO, LINQ, AJAX, RSS, RDF, etc. are keywords I hear everyday from my other friends everyday and wonder where I've been while the world has been changing so dramatically?

In retrospect, I've been working on concepts that were quite well understood in the 60's and 70's - Virtualization, Resource Management, Processes, Threads, etc. Concurrency is very interesting, specially due to multi-core now being ubiquitous. My TODO list looks very dull - Python, Algorithms, Architecture of so and so processor, etc.

It seems like there is a hard barrier that prevents people stuck in systems from working in the layer above. I feel left and cut-out from the innovation in the application space. While, I am still finishing up on the basic design patterns, the world has already moved on to domain specific patterns. I do feel lost, but there is hope. Hope that someday, I'll take a look on the other side, where the grass seems greener; jump the fence; make sense of the new world; the world that seems to be moving at a rapid pace.

Doing my time outside of systems programming

I spend a lot of time doing systems programming. As humbly as possible I don't claim to be an expert at it. While there is a lot of learning to be done on the systems side, the applications side also drags me to it. I feel I should do my time with them, it's their fair due that I owe them; to understand them better and to appreciate them.



I've just started reading Web Services Essentials. I finished chapter 1, it seemed quite well written and was easily understandable. Unfortunately, the websites listed in the book are no longer accessible. That is quite understandable given that the book was written in 2002.



The other favourite book these days is "C# Precisely". It's a well written compact book. The book consists of 30 sections and you get the feeling that you are learning C# with the least possible time investment. I look forward to putting Mono through the C# precisely test

There are also some other things on my plate that I hope to finish up quickly. I suspect that going through this exercise will give me more debating power to argue for/against technology with my friends and colleagues and also help me appreciate what lies on the other side.

Ups and Downs from India's Australian Cricket Tour

An Indian Fan's view of the ups and downs

Up

1. India did well without their seniors in the ODIs
2. Tendulkar helped India win two finals
3. Dhoni led from the front
4. Harbhajan Singh managed to dodge charges leveled against him without evidence
5. India let Brag Hogg go, even though he used abusive words against the Indian captain
6. Bad umpiring was penalized. Steve Bucknor got his well deserved punishment for bad umpiring (long due, he seemed biased against India)
7. Hayden was reprimanded for denigrating an Indian player
8. Australian public criticized the spirit of the Australian Cricket team for unfair play

Down

1. Adam Gilchrist, a well respected cricketer retired
2. Michael Clarke got away, even though he tackled Tendulkar in the finals
3. Symonds seemed out of form
4. Ricky Ponting seemed out of form
5. Australian players got away by claiming catches that they really failed to hold on to
6. Australian media for looking foolish after every charge they leveled against the Indian team players were waved off

Linux 2.6.25-rc1 (new features)

I am extremely glad to see 2.6.25-rc1, specially due to some of the new features that I've added.

A list of features is mentioned in the LWN article about the merge window for 2.6.25-rc1. This includes the memory controller (now called memory resource controller). Check out http://lwn.net/Articles/268937/ as well. We also got fake NUMA nodes for PowerPC and many other small fixes.

I have a new favorite website for git changelogs http://tree.celinuxforum.org/gitstat/index.php

Check out some of the cool graphs from that website

Linux.conf.au (presentations available)

Check out http://linux.conf.au/programme/presentations

They also have the talks in .spx format, wonderful!

Bhaskara's proof of the pythagoras's theorem

Bhaskara's Proof

Bhaskara's proof (1114 AD) of the theorem follows from the figure




So simple, elegant and beautiful. I wish I was aware of this proof as a student

Favourite Quote for the Day

Loosely speaking we can think of automata, grammar and computability as the study of what can be done by computers in principle, while complexity addresses what can be done in practice.

Peter Linz

Crazy mathematics

As I was shifting my books from one location to another, I found a big heap of books, I had not even started reading or looked at in detail. Being me, I got distracted by them and decided to at-least give them a chance to fulfill their destiny (to be read cover to cover by their owners or the friends of the owners).

I started with Tom Apostol's Mathematical Analysis. It is a wonderfully terse book and I found myself struggling by the time I reached the topic on "Point Set Topology". I had been through a quick journey of the past - real and complex numbers, sets and cardinality, prime numbers and many more things.

I also ran into the famous Cauchy-Schwarz inequality. I remember having proved the inequality in the exercise of Don Knuth's Art of computer programming volume I, but by now I had forgotten it all -- the sum manipulation, the whole of chapter 1 as hiding somewhere in my memory and seeking it was a hard/impossible task. This time, I found myself taking the help of a symbolic computation package - maxima. My dialog was as follows

(%i2) expand((a1^2 + a2^2)*(b1^2 + b2^2));
(%o2) a2^2*b2^2+a1^2*b2^2+a2^2*b1^2+a1^2*b1^2
(%i4) expand((a1b1 + a2b2)^2);
(%o4) a2b2^2+2*a1b1*a2b2+a1b1^2

(%i31) (a1^2 + a2^2)*(b1^2 + b2^2) - ((a1b1 + a2b2)^2);
(%o31) (a2^2+a1^2)*(b2^2+b1^2)-(a2b2+a1b1)^2
(%i32) expand(%);
(%o32) a2^2*b2^2+a1^2*b2^2+a2^2*b1^2+a1^2*b1^2-a2b2^2-2*a1b1*a2b2-a1b1^2

To cut the long output short, I ended up with Lagrange's inequality.

This is just a small part of the mathematics I have to (re)learn, there is also graph theory, game theory, probability theory, Markov chains and theory of computation.

All the mathematics is very exciting, but time consuming. I have loads of other information to read and digest. Let's see where all this I need to learn this crazy math attitude takes me.

Have a fantastic new year 2008

It's that time again, a year has gone by, brought happiness and sad moments, but it's gone! Remember to change the year in your dates, make new year resolutions, start afresh, be more positive, thankful and happy that we have a new chance to do/make things right.

HAVE A GREAT NEW YEAR and make every day count

Rusting Knowledge

Over a period of time, it seems like a layer of dust covers your deep understanding and knowledge of a subject matter, where in once I was considered an expert. My knowledge is rusting and it's time to re-read some old classics and books I used to learn the subject matter from.

I've been trying to learn C#, Python, etc, but I realize that I need to revisit my C skills and my C99 knowledge base again. In summary, knowledge rusts over a period of time, specially the items not practiced very often.

Get ready to clear away the dust from your rusting areas as well, as I explore well known areas and newer areas with a "I used to know, but forgot" perspective.

Digging out of history (longjmp is bad)

I had this program written out a long time back, it shows why some constructs are valid, but do not apply well today. In the evolution of C++, we've left behind some holes


 1 #include <stdio.h>
 2 #include <setjmp.h>
 3 #include <string.h>
 4
 5 class C {
 6 private:
 7         char *ptr;
 8 public:
 9         C (const char *name) {
10                 ptr = new char[strlen(name)+1];
11                 printf("constructed C\n");
12         }
13
14         ~C () {
15                 delete [] ptr;
16                 printf("destroyed C\n");
17         }
18 };
19
20 void funfunction(jmp_buf env)
21 {
22         C c("Hello");
23         /*
24          * Do something, at the end expect destructor to be
25          * called
26          */
27         longjmp(env, 1);
28 }
29
30 int main(void)
31 {
32         jmp_buf env;
33         int ret;
34
35         ret = setjmp(env);
36         if (ret) {
37                 printf("Looks like we jumped a long way\n");
38                 return ret;
39         }
40
41         funfunction(env);
42         return 0;
43 }


See if you can find the obvious problem with this code

Sounds in my life

The gentle and shrill sounds in my everyday life, call my attention to them. In earlier days, the only sound that one would hear is of people talking to them, children playing on the street or a baby calling for Mama's attention.

Compare and contrast that with what we hear today; the sound of the TV, the microwave, radio, music player, sound of vehicles zooming on the street, vehicles blowing horns, the telephone, the cell phone trying to tell you someone wants to talk to you (even when you are in the restroom), this someone could be a pre-recorded sales call; the computer buzzing to tell you that someone is trying to reach you through instant chat, the courier person ringing the doorbell, the alarm-clock ringing on my cell phone

How am I supposed to listen to myself and others around me; with these distracting sounds calling for my attention. We have moved into a new era, where we need more sounds to tell us, that we are needed elsewhere. I need to filter these sounds, so that I can hear my heart-beat, the pin-drop and my those around me giggle and laugh.

Guess what OS Don Knuth runs?

Check out http://www-cs-faculty.stanford.edu/~knuth/news.html

In the section "Wanted: A Name For High-Tech Grief" Knuth states and I quote

"I myself have often cried out for help to colleagues who have generously made house calls, in order to unwedge my highly customized Linux system"

We've known it for a while, looking at the customized fvwm2 configuration on Knuth's website, but it's good to see it in writing.

Some of you must be thinking, he has problems with Linux? We'll look at the keywords -- highly customized. Can't do that with other operating systems without hitting yourself in the head a few times or hitting your head against something. You don't hit yourself because your stupid, you do, because you brought the product :-)

Hmmm.... Now, if we can get everyone who has directly or in-directly been benefited from Knuth's work to run Linux, that would be a wonderful starting place. Of-course regular users are more than welcome to adopt and customize Linux.

Linux is a Trademark of Linus Torvalds

FOSS.IN delegate registration is open

Register at http://foss.in/2007/register/delegates/

The fee structure has changed this year. Do read the details before registering. This year we have project days and the main conference

Talk at VTU

I gave a talk on Introduction to Operating Systems via EDUSAT. The lecture was broadcast to several universities via satellite. I had no live audience, but it was fun to speak into a camera. The video has been stored/archived for students to see later.
Here's a link to the slides

New Code in the Linux Kernel

I got some new code included into the mainline linux 2.6.24-rc1 kernel

cgroupstats - Is an infrastructure to allow sending control group statistics to user space using taskstats/genetlink
cpu_acct - CPU accounting subsystem for control groups

Worked on some other miscellaneous stuff as well. Overall, 2.6.24 was an interesting merge cycle, lots of new stuff went it. Check it out! Please do report regressions

A Quick Estimate

I was trying to estimate how long it would take for a timer, measured in nano-seconds using 64 bits would take to roll over. Here's my quick estimate

It would take 2^64/10^9 seconds = 2^64/2^9*5^9

Approximating on all calculations

= 2^55/5^9

5^3 = 125 ~= 2^7
5^9 = 2^21

= 2^55/2^21

= 2^34 seconds

2^34 seconds = 2^34/86500 days ~= 2^34/80*2^10 = 2^24/2^6 days = 2^18 days

1 year = 365 days, 3 years ~= 2^10 days

That leaves us with 2^8 years * 3 years

= 256*3 years before the timer overflows

With a machine, usually one would expect at-least one reboot in 800 years, if so, we would be fine for a long time to come.

See I need no calculator :-)