April 2010 - Pharo-dev - lists.pharo.org

[Pharo-project] Fwd: [squeak-dev] OB performance (Re: Slime)
by Mariano Martinez Peck April 7, 2010

April 7, 2010

Thanks Andreas :) Lukas, do you think this is worth? ---------- Forwarded message ---------- From: Andreas Raab <andreas.raab(a)gmx.de> Date: Wed, Apr 7, 2010 at 5:25 AM Subject: [squeak-dev] OB performance (Re: Slime) To: The general-purpose Squeak developers list < squeak-dev(a)lists.squeakfoundation.org> On 4/3/2010 2:22 PM, Colin Putney wrote: > > On 2010-04-03, at 11:50 AM, Andreas Raab wrote: > > On 4/3/2010 11:27 AM, Levente Uzonyi wrote: >> >>> On Sat, 3 Apr 2010, Andreas Raab wrote: >>> >>>> Interesting. My encounters with OB mostly originate from Pharo where >>>> it's always been dog-slow so I was never interested in looked deeper. >>>> Any ideas how to install OB without that stuff? >>>> >>> >>> Try this: >>> >> >> Thanks. It's definitely better than I remembered. I'm still noticing a >> real delay in some areas though, for example when I click on a system >> category. One thing you can try to test this is to open a regular browser >> and OB and then just keep scrolling through the system categories with >> cursor up or cursor down. In the regular browser this entirely smooth; in OB >> it's "stuttering" and at least I notice the same delay when clicking on the >> list. >> > > Yeah, this one area where OB is certainly not as fast. That's because OB > sorts classes according to inheritance, rather than name. It's just more > work. I find the tradeoff worthwhile - it's not that slow, and the extra > information imparted by the sort order is useful - but reasonable people may > disagree. It wouldn't be hard to make this a preference. > Nope it's not that. A bit of profiling shows the particular problem is caused by some gross inefficiencies when updating. If you file in the attached tweaks, OB scrolls just as smoothly as the standard browser. I'm not sure if the tweaks have side effects though; you might want to verify that (in particular because of the comment in OBColumn>>switchFilter: which indicates some interference although I can't see how that could make a difference if the filter is identical). Cheers, - Andreas

2 1

Re: [Pharo-project] Redline Smalltalk
by Mariano Martinez Peck April 7, 2010

April 7, 2010

> > Are you aware of the work I did that enables class fault detection in > Squeak? > > If a class has a nil methodDictionary, it causes a fault (ie it sends > recoverFromMDFault to the class). That code then restores the MD in > addition to any anyalysis you want. I never cleaned it up for posterity, but > it enabled lots of interesting work... > > By nilling all but a few MDs, and then running some application, you > could > tell exactly what classes were needed and which were not. > See the method discoverActiveClasses. > A special feature even recorded the call stack that caused each > class to be > touched. > What I was doing was restructuring the entire system as a bunch of > image segments that could be swapped out and brought in > on demand -- a sort of user-programmable swapping memory. > This all worked at one point. Swapping of projects was a bit > flakier > because of our weak discipline on pointer scope, but it > even worked in many cases. > > Where I was going with the whole "reorganizeEverything" project was that > the SystemDictionary would be partitioned into many separate dictionaries in > each project, and that namespaces would be the same as the project swapping > unit, and the whole thing would be simple and fast. It actually worked once > (;-) also. > > Loom was cool, but it could never come close to the speed of imageSegments. > I played around with swapping out the entire VM Support category with all > classes and methods, and it would swap in the entirety in less than 200 > milliseconds, and this was a decade ago. > > Sorry for the long digression, but maybe some of this thinking or even the > actual mechanism can be of use to your experiments. > > Hi Dan. First let me say that it is really cool to have you in the list :) Yes, I know about that MD faults and I did some experiments with it. As part of my work, I remove that from core and I put it in a separate package. I also rewrite some things a little. I invite you, if you want/have time to look at it. It is in http://www.squeaksource.com/ClaseUseDiscovery You can see the class comments of CUDClassesUseDiscoverer and also the CUDClassesUseDiscovererTest This tool is working perfect. I sent an email some weeks ago: http://n4.nabble.com/ANN-ClaseUseDiscovery-tool-td1296859.html#a1296861 Now, my conclusions were that it is a great tool but it has the problem that when ever you do a Smalltalk allClassesDo or similar, it brings all the classes back. I saw there was the *gentlly* so that you could do allClassesDoGently etc and the gently checks for isInMemory, where in ImageSegmentRootStub answers false. The problem is that you have to put those gently everywhere and once you forget about that....you are dead. The other problem is the gradunalirity. I mean, even when sending a simple message to a single object of a class, the whole class is loaded. Even with all that, it was great to see and have in mind a percentage of the used classes against different scenarios. Cheers Mariano > - Dan > > > > >> James Ladd <james_ladd(a)hotmail.com> wrote... > >>>> Hi Pharo-ites, > >>>> > >>>> I'm working on a port of Pharo to the Java Virtual Machine called > Redline Smalltalk. > >>>> (read more here: http://jamesladdcode.com/?p=323) > >>>> > >>>> I'm wanting to port every single class over time, but initially enough > to get the > >>>> compiler working and have a minimum runtime. > >>>> > >>>> Of the 1600+/- classes which are those necessary to make a small > workable compiler > >>>> and runtime environment? > >> > >> Hi, James - > >> > >> Bert Freudenberg forwarded this along to me, as I'm not on a Pharo > mailing list. > >> > >> I think you know of my JSqueak project (a.k.a. Potato) -- a Squeak VM > written in Java. The image that I used with it, is the same that I always > use for this kind of experiment -- mini.image built in Squeak 2.2. > >> > >> The reason it is so cool is that it includes browser, editor, compiler, > inspector debugger, and files -- everything you need for self-support. Plus > it has the decompiler, and special temp-name hack, so you can actually > browse sources with temp names before any file system works. My favorite > thing about this image is that although it is only 600k, if you decompile > all its sources, it comprises over 850k of Smalltalk, so it's like a > compressed version of the sources with a full-function IDE thrown in for > free ;-). > >> > >> As I said, the image is about 600k, of which 240k is code -- 202 classes > (x2 if you count the metas) with 4590 methods. This should be a good guide > for you in choosing classes to start with. > >> > >> You can actually run this image live in JSqueak on the web if you have > Java installed and the planets are aligned correctly... > >> http://Weather-dimensions.com/Dan/JSqueak.jnlp > >> Another interesting statistic is that the Java .jar file is only 433k, > including not only the VM, but also the entire image! This is possible > because the image is so small that the pointers in it are mostly zeroes, so > it compresses nicely. I just checked that this runs, and was happy to see > that it putts along at 21 million bytecodes/sec on my laptop. > > > > >> I haven't looked recently, but I think you'll find Mini2.2.image or > something like it in the same place where all the historical releases of > Squeak are kept. Let me know if you can't find it. > >> > >> Have fun with your project! I hope this helps > >> > >> - Dan > >> _______________________________________________ > >> Pharo-project mailing list > >> Pharo-project(a)lists.gforge.inria.fr > >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > > > > >_______________________________________________ > >Pharo-project mailing list > >Pharo-project(a)lists.gforge.inria.fr > >http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >

1 0

[Pharo-project] FFI+Linux+double return value
by Schwab,Wilhelm K April 7, 2010

April 7, 2010

Hello all, I had some strange results with double return values and boiled it down to a C function Double AnyDouble() { return 1.0; } called via anyDouble < cdecl: double 'AnyDouble' () > and driven by | out | out := Array writeStream. 100 timesRepeat:[ out nextPut:Library default anyDouble ]. out contents. The results are not good. It works for the first three elements, after which it produces NaN. The real problem appears to be comparison of floats being broken after that point, as evidenced by a walkback that arises when trying to inspect the resulting collection - exploring it works. I need to try this on the latest image/vm, but wanted to start gathering opinions. Any ideas? Bill

3 8

[Pharo-project] Mac carbon VM goes to 4.2.4beta1U
by John M McIntosh April 7, 2010

April 7, 2010

In order to wrap up some VM fixes that should be pushed into the Squeak 4.x offering I've compiled up a 4.2.4beta1U VM This will be the last 4.x series of macintosh VMs as the 5.x series gains support. Someone should run the Sunit and smoke test to ensure the VM is sane. Follow the macintosh link from http://www.squeakvm.org/index.html 4.2.4b1 We update to VMMaker 168 Change #versionString from 3.x.x to 4.0.2. Fix loading of image segments that are from older, but compatible image versions. fix bounds bug in primitiveSoundRecordSamples. Add primitiveUtcWithOffset. Provide an explicit imageFormatBackwardCompatibilityVersion Merge some/*lots* of 64bit fixes -- =========================================================================== John M. McIntosh <johnmci(a)smalltalkconsulting.com> Twitter: squeaker68882 Corporate Smalltalk Consulting Ltd. http://www.smalltalkconsulting.com ===========================================================================

2 1

[Pharo-project] S2JTranslator ...DNU's
by James Ladd April 6, 2010

April 6, 2010

Hi Pharo-ites, I'm trying to run Alexandre Bergel's S2JTranslator on the latest Pharo image. I'm finding a lot of DoesNotUnderstand errors like RBMethodNode DNU #primitiveText Has anyone got the S2JTranslator working with latest Pharo? Rgs, James. _________________________________________________________________ New, Used, Demo, Dealer or Private? Find it at CarPoint.com.au http://clk.atdmt.com/NMN/go/206222968/direct/01/

1 0

Re: [Pharo-project] Redline Smalltalk
by James Ladd April 6, 2010

April 6, 2010

Hi Dan, We spoke (email) a while back when I was petitioning Sun (Guy Steele) to help with a Smalltalk for the JVM. I like your work a lot and I get inspiration from it. Thank you for taking the time to respond to this list and me. Bert Freudenberg - Thank you for forwarding to Dan. Rgs, James. > Bert Freudenberg forwarded this along to me, as I'm not on a Pharo > mailing list. > > Have fun with your project! I hope this helps > > - Dan _________________________________________________________________ View photos of singles in your area! Looking for a hot date? http://clk.atdmt.com/NMN/go/150855801/direct/01/

1 0

[Pharo-project] nebraska?
by Lawson English April 6, 2010

April 6, 2010

I was fooling around and noticed that Nebraska isn't in the Pharo RC image. Is it compatible at all with pharo, since it is based on Morphic? lawson

10 25

[Pharo-project] Making the operating system close window event prompt for squeak window closure.
by John M McIntosh April 6, 2010

April 6, 2010

Ok, before I do this I need to ask the Pharo folks if they handle the close window event. To test someone can alter the info.plist setting as describe below and see what happens. Begin forwarded message: > From: Bert Freudenberg <bert(a)freudenbergs.de> > Date: April 6, 2010 3:35:26 AM PDT > To: Squeak Virtual Machine Development Discussion <vm-dev(a)lists.squeakfoundation.org> > Subject: Re: [Vm-dev] New Windows VM: 4.0.1 > Reply-To: Squeak Virtual Machine Development Discussion <vm-dev(a)lists.squeakfoundation.org> > > > On 28.03.2010, at 22:59, Andreas Raab wrote: >> >> >> 1) The VM will now default to pass the window close event to the image instead of handling it directly. You can see that in a current trunk image by clicking on the close button (you'll now get a Squeak notifier instead of the previous VM notifier). The old behavior is still available if you SHIFT-Click on the close button. This is very useful if your image hangs. > > John, > > could you change the default for the Mac VM too? Both Linux and Windows VMs now deliver the window event rather than killing the image. > > For Etoys we use > > SqueakWindowAttribute <8200001f> > > in the VM's Info.plist. That at least enables the red window close button (though "Quit" from the dock menu does not work). > > - Bert - > > -- =========================================================================== John M. McIntosh <johnmci(a)smalltalkconsulting.com> Twitter: squeaker68882 Corporate Smalltalk Consulting Ltd. http://www.smalltalkconsulting.com ===========================================================================

2 1

[Pharo-project] Problem loading Exupery on PharoCore 1.0
by laurent laffont April 6, 2010

April 6, 2010

Hi, I want to try to build Exupery VM. On PharoCore 1.0 I do: Gofer new squeaksource: 'MetacelloRepository'; package: 'ConfigurationOfVMMaker'; load. (ConfigurationOfVMMaker project version: '1.2') load. Then Gofer new squeaksource: 'Exupery'; package: 'Exupery'; load. I have this error: This package depends on the following classes: MOAbstractLayout You must resolve these dependencies before you will be able to load these definitions: MOGraphVizLayout MOGraphVizLayout>>doExecute: MOGraphVizLayout>>findTopLeftCorner: I suppose I need GraphViz package Gofer new squeaksource: 'GraphViz'; package: 'GraphViz'; load. But when I load it I have a Syntax Error: runWithSystemFramework: string <> expected ->apicall: short 'system' (char*) module: 'System.framework'> ^ self externalCallFailed Laurent Laffont

6 11

[Pharo-project] Handle Close Event in Image - necessary for 1.0?
by Torsten Bergmann April 6, 2010

April 6, 2010

I think I've found something that should be fixed before releasing 1.0 since it is meanwhile a major problem on Windows. In "old" Squeak/Pharo you could close the native world window using the "X" or hit ALT+F4 on Windows and the system asked you if you really want to close the whole Smalltalk system using a native dialog which was provided by the VM. With the new VM's (4.0.1 and 4.0.2 on Windows, see [1]) the native "window close event" is now dispatched from the VM into the image so it could be handled by Smalltalk code. This is currently not done/handled in Pharo. Only in Squeak so far. The recent Squeak trunk/release candidate already handles the event within the image and now provides an in-image dialog to ask the user to quit. So: in Pharo now NOTHING happens when you try to close the window using "X", you can only quit using the world menu - which is IMHO really a show stopper since you cant close using "X" or ALT+F4 (the key combination to close any windows program). Two possible options: A) we should handle the event now in-image similar to Squeak by fixing and integration issue 2131 (see [2]), so not only for 1.1. but also for 1.0 B) release and distribute Pharo 1.0 with an old VM < 4.0 BTW: You can get the old behavior on Windows by additionally using SHIFT, so SHIFT + ALT + F4 allows to close. But this is not very convinient, doesnt follow what the typical Windows user expects and one has to know about it... Bye T. [1] http://lists.squeakfoundation.org/pipermail/vm-dev/2010-March/004142.html [2] http://code.google.com/p/pharo/issues/detail?id=2131 -- GRATIS fÃ¼r alle GMX-Mitglieder: Die maxdome Movie-FLAT! Jetzt freischalten unter http://portal.gmx.net/de/go/maxdome01

4 3