- Pharo-dev - lists.pharo.org

[Pharo-project] Cog VM doesn't start to listen on socket

by Lukas Renggli

Hi Eliot, I noticed that on the Cog VM the Kom Seaside server does not automatically start to listen, even if there is a process waiting at the semaphore of the socket. Singaling that semaphore from the process browser fixes the problem and the Cog image is fully working until the next time it is opened. The exact same image opened with an old VM does not show the problem and immediately starts to listen by itself. Did anybody else notice the same problem? Anybody found a solution? Lukas -- Lukas Renggli www.lukas-renggli.ch

Oct. 26, 2010

Re: [Pharo-project] Learning on Sockets

by GÃ¶ran Krampe

On 10/25/2010 09:21 AM, laurent laffont wrote: > Thank you Noury. > > Indeed for those interested in, I've just remembered that PBE is on > github. So we can read it here > http://github.com/SquareBracketAssociates/PharoByExample-english/blob/maste… > > Cheers, > > Laurent. Nice! Although I *do* think you should tell the reader that for most uses you would not need to use Socket directly - SocketStream is much more comfortable. I notice you have a planned section on SocketStream, but please let the reader know earlier in the text. regards, GÃ¶ran

Oct. 26, 2010

Re: [Pharo-project] [squeak-dev] How weak finalization affects the performance

by Igor Stasenko

On 26 October 2010 10:30, Andres Valloud <avalloud(a)smalltalk.comcastbiz.net> wrote: > Why does finalization do any work when running a block which I think just > creates garbage? Â Is finalization on a per class basis (i.e.: the VM > notifies the image that some objects might be notified, and the image just > enumerates through them), or on a per object basis (i.e.: the VM maintains a > finalization queue as in VisualWorks)? > There is no per-object or per-class finalization in Squeak. One must either register an object using WeakRegistry, or add own object to weak dependents (it should answer to #finalizeValues). Squeak VM signals the semaphore each time GC happen, and actually, a finalization process could do something arbitrary to react on such event. Putting delay there ensuring that even if you put a very ineffective finalization into weak dependents, it won't affect the performance too much. > On 10/26/10 0:25 , Igor Stasenko wrote: >> >> Here the idea, which came to Chris, >> put a delay into #finalizationProcess loop: >> >> finalizationProcess >> Â Â Â Â [true] whileTrue: >> Â Â Â Â Â Â Â Â [ WeakFinalizationList initTestPair. >> Â Â Â Â Â Â Â Â FinalizationSemaphore wait. >> Â Â Â Â Â Â Â Â FinalizationLock critical: >> Â Â Â Â Â Â Â Â Â Â Â Â [ >> Â Â Â Â Â Â Â Â Â Â Â Â WeakFinalizationList checkTestPair. >> Â Â Â Â Â Â Â Â Â Â Â Â FinalizationDependents do: >> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â [:weakDependent | >> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â weakDependent ifNotNil: >> Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â Â [weakDependent finalizeValues]]] >> Â Â Â Â Â Â Â Â Â Â Â Â ifError: >> Â Â Â Â Â Â Â Â Â Â Â Â [:msg :rcvr | rcvr error: msg]. >> Â Â Â Â Â Â Â Â 5 seconds asDelay wait. >> Â Â Â Â Â Â Â Â ]. >> >> >> And here a simple benchmark, which triggers GC often: >> >> [ Array new: 100 ] bench >> >> without delay: >> >> '2,450,000 per second.' >> '2,490,000 per second.' >> '2,490,000 per second.' >> '2,480,000 per second.' >> '2,530,000 per second.' >> >> with delay: >> >> '2,670,000 per second.' >> '2,680,000 per second.' >> '2,690,000 per second.' >> '2,730,000 per second.' >> >> roughly about ~8% faster :) >> >> But now lets put something big into weak array: >> >> | dict b | >> dict := WeakKeyDictionary new addAll: (( 1 to: 1000 ) collect: [:i | >> i->i] ); yourself. >> WeakArray addWeakDependent: dict. >> b := [ Array new: 100 ] bench. >> WeakArray removeWeakDependent: dict. >> b >> >> without delay: >> >> '1,840,000 per second.' >> '2,060,000 per second.' >> '2,130,000 per second.' >> >> with delay: >> >> '3,030,000 per second.' >> '2,880,000 per second.' >> '2,890,000 per second.' >> >> Do not forget to do: >> WeakArray restartFinalizationProcess >> >> when you changing the #finalizationProcess method, >> otherwise you won't see real numbers. >> >> So, i like the idea of putting delay there. >> Finalization is eventual, and there is no hard guarantees that it will >> happen in micro-second just after some object become garbage. >> So, be it 5 seconds or 1000 seconds not really matters. >> What is matters that with delay we win much more, by avoiding wasting >> time in finalization process too often. >> >> > > -- Best regards, Igor Stasenko AKA sig.

Oct. 26, 2010

[Pharo-project] Fwd: Re: [squeak-dev] How weak finalization affects the performance

by Andres Valloud

-------- Original Message -------- Subject: Re: [squeak-dev] How weak finalization affects the performance Date: Tue, 26 Oct 2010 03:30:13 -0400 From: Andres Valloud <avalloud(a)smalltalk.comcastbiz.net> Reply-To: andres.valloud(a)gmail.com <andres.valloud(a)gmail.com>, The general-purpose Squeak developers list <squeak-dev(a)lists.squeakfoundation.org> To: The general-purpose Squeak developers list <squeak-dev(a)lists.squeakfoundation.org> Why does finalization do any work when running a block which I think just creates garbage? Is finalization on a per class basis (i.e.: the VM notifies the image that some objects might be notified, and the image just enumerates through them), or on a per object basis (i.e.: the VM maintains a finalization queue as in VisualWorks)? On 10/26/10 0:25 , Igor Stasenko wrote: > Here the idea, which came to Chris, > put a delay into #finalizationProcess loop: > > finalizationProcess > [true] whileTrue: > [ WeakFinalizationList initTestPair. > FinalizationSemaphore wait. > FinalizationLock critical: > [ > WeakFinalizationList checkTestPair. > FinalizationDependents do: > [:weakDependent | > weakDependent ifNotNil: > [weakDependent finalizeValues]]] > ifError: > [:msg :rcvr | rcvr error: msg]. > 5 seconds asDelay wait. > ]. > > > And here a simple benchmark, which triggers GC often: > > [ Array new: 100 ] bench > > without delay: > > '2,450,000 per second.' > '2,490,000 per second.' > '2,490,000 per second.' > '2,480,000 per second.' > '2,530,000 per second.' > > with delay: > > '2,670,000 per second.' > '2,680,000 per second.' > '2,690,000 per second.' > '2,730,000 per second.' > > roughly about ~8% faster :) > > But now lets put something big into weak array: > > | dict b | > dict := WeakKeyDictionary new addAll: (( 1 to: 1000 ) collect: [:i | > i->i] ); yourself. > WeakArray addWeakDependent: dict. > b := [ Array new: 100 ] bench. > WeakArray removeWeakDependent: dict. > b > > without delay: > > '1,840,000 per second.' > '2,060,000 per second.' > '2,130,000 per second.' > > with delay: > > '3,030,000 per second.' > '2,880,000 per second.' > '2,890,000 per second.' > > Do not forget to do: > WeakArray restartFinalizationProcess > > when you changing the #finalizationProcess method, > otherwise you won't see real numbers. > > So, i like the idea of putting delay there. > Finalization is eventual, and there is no hard guarantees that it will > happen in micro-second just after some object become garbage. > So, be it 5 seconds or 1000 seconds not really matters. > What is matters that with delay we win much more, by avoiding wasting > time in finalization process too often. > >

Oct. 26, 2010

[Pharo-project] How weak finalization affects the performance

by Igor Stasenko

Here the idea, which came to Chris, put a delay into #finalizationProcess loop: finalizationProcess [true] whileTrue: [ WeakFinalizationList initTestPair. FinalizationSemaphore wait. FinalizationLock critical: [ WeakFinalizationList checkTestPair. FinalizationDependents do: [:weakDependent | weakDependent ifNotNil: [weakDependent finalizeValues]]] ifError: [:msg :rcvr | rcvr error: msg]. 5 seconds asDelay wait. ]. And here a simple benchmark, which triggers GC often: [ Array new: 100 ] bench without delay: '2,450,000 per second.' '2,490,000 per second.' '2,490,000 per second.' '2,480,000 per second.' '2,530,000 per second.' with delay: '2,670,000 per second.' '2,680,000 per second.' '2,690,000 per second.' '2,730,000 per second.' roughly about ~8% faster :) But now lets put something big into weak array: | dict b | dict := WeakKeyDictionary new addAll: (( 1 to: 1000 ) collect: [:i | i->i] ); yourself. WeakArray addWeakDependent: dict. b := [ Array new: 100 ] bench. WeakArray removeWeakDependent: dict. b without delay: '1,840,000 per second.' '2,060,000 per second.' '2,130,000 per second.' with delay: '3,030,000 per second.' '2,880,000 per second.' '2,890,000 per second.' Do not forget to do: WeakArray restartFinalizationProcess when you changing the #finalizationProcess method, otherwise you won't see real numbers. So, i like the idea of putting delay there. Finalization is eventual, and there is no hard guarantees that it will happen in micro-second just after some object become garbage. So, be it 5 seconds or 1000 seconds not really matters. What is matters that with delay we win much more, by avoiding wasting time in finalization process too often. -- Best regards, Igor Stasenko AKA sig.

Oct. 26, 2010

[Pharo-project] is there a way to gofer a configurationOf?

by stephane ducasse

Hi I would like in particular to do Gofer new squeaksource: 'Citezen'; package: 'ConfigurationOfCitezen; push and that all the dependent packages get pushed stef

Oct. 26, 2010

Re: [Pharo-project] About Adding/modifying in SUnit or not?

by Niko Schwarz

I completely agree. If we take charge of SUnit, it ought to stay backwards compatible. The new things we can put into extensions. On Mon, Oct 25, 2010 at 4:07 PM, Lukas Renggli <renggli(a)gmail.com> wrote: > I am all in for improving SUnit, but there must be some compatibility > mode that makes it easy not only to run, but also to write tests on > the official SUnit. This is crucial for projects like Seaside, > Magritte, Pier, PetitParser, ... that are supposed to work on various > other Smalltalk platforms. > > Lukas > > On 25 October 2010 15:51, Niko Schwarz <niko.schwarz(a)googlemail.com> wrote: >> Yea, well, it depends on whether we want to be compatible with camp >> smalltalk's SUnit, as maintained by Niall Ross: >> http://sunit.sourceforge.net/devel.htm. Pharo already is using >> extensions, such as the possibility to mark tests as expected failures >> at all (that is impossible in standard SUnit). >> >> My humble opinion is that we must control our testing environment, and >> we already have an excellent cleanup of SUnit, namely akuhn/SUnit, >> http://www.squeaksource.com/akuhn.html. And we have progress in the >> form of Phexample, which builds ontop of SUnit. >> http://smalltalkthoughts.blogspot.com/2009/11/phexample-because-examples-ex… >> . I'd be more than happy if Pharo took a new SUnit as its standard >> testing framework, and I'd even be thrilled if enough people found >> Phexample useful enough, so that it could become standard over time. >> >> Niko >> >> 2010/10/25 Luc Fabresse <luc.fabresse(a)gmail.com>: >>> Hi all, >>> Â I am working on pushing environments in SystemNavigation. >>> Â SystemNavigationTest previously used ClassFactoryForTestCase from SUnit for >>> generating Â classes. >>> Â Now I need an environment-awareÂ ClassFactoryForTestCase. >>> Â Should I directly modifyÂ ClassFactoryForTestCase since a class generator >>> should now be environment aware? >>> Â Should I subclassÂ ClassFactoryForTestCase and put it in SUnit? >>> Â Should I never changed SUnit and put it elsewhere? ;-) >>> #Luc >>> _______________________________________________ >>> Pharo-project mailing list >>> Pharo-project(a)lists.gforge.inria.fr >>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >>> >> >> >> >> -- >> http://scg.unibe.ch/staff/Schwarz >> twitter.com/nes1983 >> Tel: +41 076 235 8683 >> >> _______________________________________________ >> Pharo-project mailing list >> Pharo-project(a)lists.gforge.inria.fr >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >> > > > > -- > Lukas Renggli > www.lukas-renggli.ch > > _______________________________________________ > Pharo-project mailing list > Pharo-project(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > -- http://scg.unibe.ch/staff/Schwarz twitter.com/nes1983 Tel: +41 076 235 8683

Oct. 26, 2010

Re: [Pharo-project] [squeak-dev] private instance variables NOT objects? (Wikipedia)

by Bert Freudenberg

On 25.10.2010, at 23:10, Eliot Miranda wrote: > > > On Mon, Oct 25, 2010 at 1:59 PM, Lawson English <lenglish5(a)cox.net> wrote: > In the Wikipedia entry on Smalltalk, there is this line in the history section, 3rd paragraph, added roughly a year ago: "...Smalltalk-80 added [[metaclass]]es, to help maintain the "everything is an object" (except private instance variables) paradigm by associating properties and behavior with individual classes, ..." > > > Is this one of those esoteric details that mere mortals are not meant to understand, or is this an error? > > "(except private instance variables)" looks like a dig or a jibe. I would have put it > > "...Smalltalk-80 added [[metaclass]]es, to help maintain the "everything is an object" paradigm by allowing classes to have their own specific state and behavior, ..." > > Perhaps they were trying to say that because Smalltalk lacks private instance variables Smalltalk objects are not true objects, which is I suppose arguable. But it makes poor sense to state that private instance variables aren't objects; public inst vars aren't objects either. It's plain nonsense. Here's the edit that added the "private instance" modifier: http://en.wikipedia.org/w/index.php?title=Smalltalk&diff=prev&oldid=2800204… Someone should just correct it. - Bert -

Oct. 25, 2010

Re: [Pharo-project] private instance variables NOT objects? (Wikipedia)

by Schwab,Wilhelm K

Right or wrong (not sure which), I have always thought of meta classes as breaking a recursion that arises when classes aspire to be objects that have all have a class. Is there such a thing as private instance variables? Could that mean the header, special behavior masks (a Dolphinism), etc.? Independent of the truth, this **IS** something that mere mortals are not meant to understand :) ________________________________________ From: pharo-project-bounces(a)lists.gforge.inria.fr [pharo-project-bounces(a)lists.gforge.inria.fr] On Behalf Of Lawson English [lenglish5(a)cox.net] Sent: Monday, October 25, 2010 4:59 PM To: The general-purpose Squeak developers list; Pharo >> "Pharo-project(a)lists.gforge.inria.fr" Subject: [Pharo-project] private instance variables NOT objects? (Wikipedia) In the Wikipedia entry on Smalltalk, there is this line in the history section, 3rd paragraph, added roughly a year ago: "...Smalltalk-80 added [[metaclass]]es, to help maintain the "everything is an object" (except private instance variables) paradigm by associating properties and behavior with individual classes, ..." Is this one of those esoteric details that mere mortals are not meant to understand, or is this an error? Lawson _______________________________________________ Pharo-project mailing list Pharo-project(a)lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project

Oct. 25, 2010

Re: [Pharo-project] [squeak-dev] private instance variables NOT objects? (Wikipedia)

by Eliot Miranda

On Mon, Oct 25, 2010 at 1:59 PM, Lawson English <lenglish5(a)cox.net> wrote: > In the Wikipedia entry on Smalltalk, there is this line in the history > section, 3rd paragraph, added roughly a year ago: "...Smalltalk-80 added > [[metaclass]]es, to help maintain the "everything is an object" (except > private instance variables) paradigm by associating properties and behavior > with individual classes, ..." > > > Is this one of those esoteric details that mere mortals are not meant to > understand, or is this an error? > "(except private instance variables)" looks like a dig or a jibe. I would have put it "...Smalltalk-80 added [[metaclass]]es, to help maintain the "everything is an object" paradigm by allowing classes to have their own specific state and behavior, ..." Perhaps they were trying to say that because Smalltalk lacks private instance variables Smalltalk objects are not true objects, which is I suppose arguable. But it makes poor sense to state that private instance variables aren't objects; public inst vars aren't objects either. best, Eliot > > Lawson > > > > >

Oct. 25, 2010