Tachyon-list Janvier 2011

tachyon-list@iro.umontreal.ca

2 participants
8 discussions

stdlib/arrays.js commit breaks make test
by chevalma＠iro.umontreal.ca 31 Jan '11

31 Jan '11

stdlib/arrays.js:919: ReferenceError: error is not defined error("lastIndexOf, results are different"); ^ error is defined in utility/debug.js. Please make sure to always run "make test" before pushing your commits. - Maxime

2 5

Fwd: [racket] moby/wescheme & TCO
by Marc Feeley 31 Jan '11

31 Jan '11

It seems like tail-calls will become mandatory in the not too distant future. What support do we have for tail-calls? I remember discussing this early on in the project, but I'm not sure that at this point in time the intermediate representation supports it. Marc Begin forwarded message: > From: David Herman <dherman(a)ccs.neu.edu> > Date: January 30, 2011 12:04:29 PM EST > To: YC <yinso.chen(a)gmail.com> > Cc: Racket Users <users(a)racket-lang.org> > Subject: Re: [racket] moby/wescheme & TCO > > I don't have your answer, but an interesting anecdote: I learned recently why the setTimeout(k, 0) trick is so slow. Apparently too many web sites (including nytimes.com, I believe) rely on a minimum delay of somewhere between 4 and 10ms to do animations, so browsers have to throttle the events. As I understand it, Firefox uses a 10ms minimum delay, and Chrome uses 4ms. > > Since Chrome seems to have gotten away with dropping it to 4ms, we may be able to do the same -- though I haven't heard anyone specifically say they planned to. Maybe at some point we can eliminate this silliness altogether, and setTimeout could become a more viable tool for compiler-writers. But happily, the ECMAScript committee recently agreed to try to mandate proper tail calls in the next version of the standard. So with a little (okay, maybe more than a little) luck, compiler-writers may eventually be able to rely on having tail calls built in to the web. > > Dave > > On Jan 29, 2011, at 9:12 PM, YC wrote: > >> Hi all - >> >> I am wondering if there are documentations on how moby/wescheme solve the tail call optimization issue to compile down to javascript. My google'fu is failing me. >> >> If not - can someone shed some light on how it's done? >> >> Thanks, >> yc >> >> _________________________________________________ >> For list-related administrative tasks: >> http://lists.racket-lang.org/listinfo/users > > _________________________________________________ > For list-related administrative tasks: > http://lists.racket-lang.org/listinfo/users

2 8

make perf
by Marc Feeley 31 Jan '11

31 Jan '11

I got this output from a "make perf": Generating report Report file: benchreport.html uname: illegal option -- o usage: uname [-amnprsv] cat: /proc/cpuinfo: No such file or directory sh: free: command not found TypeError: Cannot read property '1' of undefined at Object.getSystemInfo (bench/reports.js:240:35) at Object.genReport (bench/reports.js:72:40) at main (bench/benchmain.js:120:15) at bench/benchmain.js:130:5 It would seem that it assumes a Linux system. Can this be modified to work on Mac OS X also? For uname, you could use "uname -a" which works on Mac OS X. /proc/cpuinfo does not exist on Mac OS X. On Mac OS X you can get a bunch of system parameters by executing the system program "system_profiler". You'll get this on stdout: % system_profiler Hardware: Hardware Overview: Model Name: MacBook Pro Model Identifier: MacBookPro5,3 Processor Name: Intel Core 2 Duo Processor Speed: 2.8 GHz Number Of Processors: 1 Total Number Of Cores: 2 L2 Cache: 6 MB Memory: 4 GB Bus Speed: 1.07 GHz Boot ROM Version: MBP53.00AC.B03 SMC Version (system): 1.48f2 Serial Number (system): W893814U64C Hardware UUID: 3BC167BA-45C4-529A-BA30-3658F1E9966F Sudden Motion Sensor: State: Enabled Network: Bluetooth: ... Marc

1 0

Array operations
by Marc Feeley 30 Jan '11

30 Jan '11

I've implemented a bunch of Array library functions in stdlib/arrays.js . The code includes unit tests. Specifically the following functions are implemented: Array.prototype.toString Array.prototype.concat Array.prototype.join Array.prototype.pop Array.prototype.push Array.prototype.reverse Array.prototype.shift Array.prototype.slice Array.prototype.sort Array.prototype.splice Array.prototype.unshift Array.prototype.indexOf Array.prototype.lastIndexOf Array.prototype.forEach Array.prototype.map Array.prototype.filter For the sort function I have used a non-recursive version of mergesort which I designed from scratch (I'd be interested to know if this algorithm already exists). To activate the functions, the assignments of the form Array.prototype.array_toString = array_toString; must be changed to Array.prototype.toString = array_toString; Marc

2 2

[Tachyon-list] http://www.cs.cmu.edu/~fp/courses/15213-s07/misc/gdb-commands.txt
by Marc Feeley 25 Jan '11

25 Jan '11

And this too. http://www.cs.cmu.edu/~fp/courses/15213-s07/misc/gdb-commands.txt Marc

1 0

[Tachyon-list] http://www.cs.cmu.edu/~fp/courses/15213-s07/misc/asm64-handout.pdf
by Marc Feeley 25 Jan '11

25 Jan '11

Here's a really useful reference for x86-64 assembly code programming. http://www.cs.cmu.edu/~fp/courses/15213-s07/misc/asm64-handout.pdf Marc

1 0

Computing address of a label
by Marc Feeley 22 Jan '11

22 Jan '11

Erick, you are modifying the "mov" instruction to allow moving the address of a label to a register. For the x86-32 the linker has to be involved because there is no "pc-relative" addressing (except for jumps). On the x86-64, pc-relative addressing is one of the addressing modes. So loading the address of label XXX in %rdi could be done with a "load effective address" instruction like this: leaq XXX(%rip), %rdi The advantage is that the linker is not involved (less work to do when loading the program). The instruction encoding may be a little shorter too, and maybe faster. So try to use this instruction when generating code for the x86-64. Marc

1 0

Fwd: getting/building spider monkey
by Marc Feeley 21 Jan '11

21 Jan '11

Here's some info sent to me by Andreas Gal on compiling spider monkey. Marc Begin forwarded message: > From: Andreas Gal <gal(a)mozilla.com> > Date: January 21, 2011 2:20:29 PM EST > To: Marc Feeley <feeley(a)iro.umontreal.ca> > Subject: Re: getting/building spider monkey > > Hi Marc, > > you can obtain spidermonkey from our repository: > > hg clone http://hg.mozilla.org/tracemonkey > > That clones the entire firefox tree. > > And then > > autoconf213 > > and > > autoconf213 in js/src > > then > > cd js/src > > mkdir debug-build > cd debug-build > ../configure --enable-debug --disable-optimize > make > > That builds "js", the JS engine shell. > > To get an optimzied build, just do > > mkdir opt-build > cd opt-build > ../configure --disable-debug --enable-optimize > make > > Build prerequisites are listed here: > > https://developer.mozilla.org/en/build_documentation > > Let me know if you have any problems building the engine. > > Andreas > > On Jan 21, 2011, at 7:22 AM, Marc Feeley wrote: > >> >> On 2010-12-21, at 9:21 AM, Marc Feeley wrote: >> >>> Hello Andreas. Can you describe the steps to get and build spider monkey? I've been Googling around and the latest version I can find (1.8.0-rc1) is more than a year old. >>> >>> Marc >> >> Andreas, I haven't heard from you in a while. I hope everything is going OK. >> >> Do you think you can help me with getting spider monkey to build on my Mac? >> >> Marc >>

1 0

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

Tachyon-list Janvier 2011