Phil Dawes' Stuff - Latest Comments

Re: Phil Dawes Stuff >> Idea for a global interpreter lock optimized for low contention

Anonymous Coward — Fri, 06 Nov 2009 08:37:31 -0000

Spinlocks, http://repnop.org/analysis/spinlocks.html

Re: Phil Dawes Stuff >> Idea for a global interpreter lock optimized for low contention

Nathan — Tue, 03 Nov 2009 16:23:51 -0000

You should look at what the Java folks have done with thin locks, too. They noticed that a large number of locks are grabbed and uncontended, so they came up with a locking design that's optimized for that case.

Re: Phil Dawes Stuff >> Idea for a global interpreter lock optimized for low contention

xilun — Thu, 22 Oct 2009 13:29:19 -0000

You should look at Linux futexes.

Note that futexes are (very) tricky to implement correctly, so you should rather study them carefully. Google for "Futexes Are Tricky".

Given that pthread stuffs are implemented using futex under Linux, it would be an error to add another layer.

Re: Phil Dawes Stuff >> Hand-coding multi-platform assembler using Factor compiler intrinsics

Adam — Thu, 08 Oct 2009 15:12:19 -0000

After reading this, I really can't wait for your next post on INSN: and CODEGEN: words. I have some ideas that I want the compiler to emit (just learning experiments) but this really was a good tutorial on at least the intrinsics support.

Re: Phil Dawes Stuff >> Making a C codebase reentrant by turning it into a big C++ object

foobar — Mon, 21 Sep 2009 19:07:53 -0000

If you want to keep it C-style, you can encapsulate global state in a struct and pass it around.

int x;

int foo(){ return x; }

Becomes:

typedef struct vm_t{
int x;
} *VM;

void foo(VM vm){ return vm->x; }

Re: Phil Dawes Stuff >> Making a C codebase reentrant by turning it into a big C++ object

Brian — Mon, 21 Sep 2009 09:53:18 -0000

I thought of this idea for another project but haven't tried it yet. I'm glad to see the approach works and your 5 steps will save me a few false starts.

Re: Phil Dawes Stuff >> BTriples - a model for aggregating structured data

Name — Thu, 23 Jul 2009 08:40:05 -0000

Why don't you call it an RDF molecule?

Re: Phil Dawes Stuff >> BTriples - a model for aggregating structured data

dominicsayers — Mon, 20 Jul 2009 04:54:21 -0000

Tribbles, surely.

Re: The Factor Attraction

Roger Levy — Tue, 02 Jun 2009 22:05:31 -0000

Hi,

Do you feel that your agility in Factor has improved since this post?

Roger

Re: Disqus comments

phildawes — Mon, 11 May 2009 07:44:21 -0000

Thanks for the pointer - I've cleaned up the spam and regrettably added some moderation

Re: Disqus comments

Dominic Sayers — Mon, 11 May 2009 06:07:24 -0000

I'm loving the comments thread for this post. Can't decide whether to get my upholstery cleaned or do something about my fast food obesity.

Re: Factor makes you write better code

phildawes — Sun, 12 Apr 2009 14:57:30 -0000

Cool - thanks Eric

Re: Factor makes you write better code

Eric Mertens — Sun, 12 Apr 2009 03:25:21 -0000

I pasted some code that does the moving sum in factor.

http://paste.factorcode.org/paste?id=569#282

Re: Factor makes you write better code

phildawes — Tue, 31 Mar 2009 08:54:52 -0000

This comment on proggit was particularly good I thought

Re: Factor makes you write better code

phildawes — Tue, 31 Mar 2009 05:26:48 -0000

More discussion at proggit

Re: Factor makes you write better code

Alex13 — Tue, 31 Mar 2009 02:16:54 -0000

Here is moving average in J:

[arr =. i.10
0 1 2 3 4 5 6 7 8 9

ave =. +/%#

4 ave \ arr
1.5 2.5 3.5 4.5 5.5 6.5 7.5

Re: Factor makes you write better code

Sam Danielson — Mon, 30 Mar 2009 21:24:59 -0000

I think you can eliminate the inner loop by storing a "moving" sum. I'm sorry but I don't know how to put this into factor.

var sum = 0;
for (i=0; i<arr.length; i++) {
if (i >= period) sum -= arr[i - period];
sum += arr[i];
out.push (sum / period);
}

Re: And another new programming language

Jeff — Fri, 13 Mar 2009 21:00:20 -0000

I'm still looking for a minimal machine that supports true hardware parallelism. Any thoughts? Pointers?

Re: Spread Betting

Lorraine — Mon, 26 Jan 2009 16:51:30 -0000

How is it going, a few weeks later?

Re: Spread Betting

Andy — Tue, 16 Dec 2008 19:19:03 -0000

nice read i understand your pain about requiring compliance permission for each trade as i used to work for a brokerage company and had the same problem, the current system is pretty ridiculous - at the time i used to trade forex which didn't require disclosure ;)

Andy

http://www.financial-spread-betting.com

Re: Disqus comments

phildawes — Thu, 02 Oct 2008 10:19:14 -0000

Thanks for pointing me at your script - I'll defo check it out

Re: Really simple html templating in factor

Slava Pestov — Thu, 02 Oct 2008 05:24:14 -0000

Phil, by sheer coincidence, Doug Coleman (erg) implemented something similar in extra/interpolate. He's using it to construct SQL stored procedures, IIRC. Perhaps you could look at his code for ideas, or merge your efforts? On a related note, I'd love to have your work (possibly after refactoring it to use 'interpolate') in the repository under basis/html/templates/simple, or something like that. Having more options for templating is always good.

Re: Disqus comments

davemw — Wed, 01 Oct 2008 11:24:21 -0000

For the past 5 years my blog has consisted of a gently accumulating Python script. It is currently around 300 lines, up from 200 or so when I initially created it. Around 2006 I decided to write a script to import the (filesystem-based) articles into a WordPress install. Instant dissatisfaction. WordPress is huge, insecure, and a little overfeatured for the average Joe.

So I'm still here with my horrible little Python script, and quite happy to be using it over WordPress. I'm presently adding an archives facility, integration with Disqus, and my first attempt at a pretty theme.

Take a look, it does HTML and RSS 2.0 output, and supports either HTML or a horrible wiki-like formatting for posts:

http://pastebin.com/f1be9ea59

Re: Really simple html templating in factor

ryantm — Tue, 30 Sep 2008 17:25:02 -0000

Have you seen Haml? It's a templating language that makes some complex control stuff with HTML or XML look good. I'm not really sure how it would integrate with Factor evaluation methods though.

http://haml.hamptoncatlin.com/

Re: Searching arrays in X86 assembler with a bloom filter pt 3

Phil Dawes — Sat, 13 Sep 2008 06:30:13 -0000

Slava Pestov wrote:
> Hi Phil,
>
> I couldn't post a comment on your blog for some reason so I'm posting
> this to the list instead.
>
> The seq>hash word you write already exists in the sets vocabulary, its
> called unique.
>
> And (prepare-filter) looks nicer if you use fry:
>
> : (prepare-filter) ( filter seq -- )
> '[ 1048576 mod _ set-bit ] each ;
>
> Slava