Squeak-dev February 2008

squeak-dev@lists.squeakfoundation.org

159 participants
219 discussions

Why is Heap>>#species => Array?
by Klaus D. Witzel 26 Feb '08

26 Feb '08

Subject line says it all, check yourself, (Heap withAll: 'array') reject: [:x | x = $r] What's the rationale (there's no doc, no comment)? Archive shows that #species was changed to fix another (anonymous) bug but, this way the senders of #species can impossibly do what Smalltalk users expect from the collection hierarchy (and there is #asArray ...) - http://lists.squeakfoundation.org/pipermail/squeak-dev/2000-March/011043.ht… /Klaus P.S. Heap is used in not-so-uncritical parts, like #startTimerEventLoop and WorldState, so I don't want to "just" play with Heap's #species without any background info on what was fixed.

4 16

Re: A better looking text caret (IMO)
by Jerome Peace 26 Feb '08

26 Feb '08

[squeak-dev] A better looking text caret (IMO) *** >Igor Stasenko siguctua at gmail.com >Sun Feb 24 02:49:19 UTC 2008 > > >On 24/02/2008, Matthew Fulmer <tapplek at gmail.com> wrote: >> On Sun, Feb 24, 2008 at 04:38:32AM +0200, Igor Stasenko wrote: >> > Sometimes it's hard to notice a caret position in text, because it's >> > thin and my display is soo big :) >> > This patch draws caret as 5 pixel width vertical bar. It's much easier >> > for my eyes to track it's position in text. >> > >> > I'm also wanted to make it blinking.. but hmm.. i think it will >> > require adding too much for achieving that, because caret is not >> > considered as some kind of visual object in text morph, and drawn only >> > when selection is empty and only during text update :( >> >> >> You should submit this as a bug report, not an email >> > >It's not a bug :) >It's just another way to draw cursor caret , to anyone who might like >it more than current one. An enhancement is a perfectly good topic for a mantis issue. I would usually tag it with [ENH]. Mantis tracks issues. So putting a report on mantis is a perfectly good way to begin to change squeak. The attention to mail is ephemeral. Here today gone tommorrow. Mail is a good way to draw attention to an issue. Mantis is a good way to see an issue followed up. A good sequence is Mantis report any clearly describable complaint. Follow with mail pointing to the report. Add analysis and useful code to mantis (if no one else beats you to it.) Then call attention to mantis issues ready for harvest when a good* release team is in place. Yours in service and curiosity, --Jerome Peace * A good release team is one that respects you. And trusts your work iff you have earned their trust. ____________________________________________________________________________________ Be a better friend, newshound, and know-it-all with Yahoo! Mobile. Try it now. http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ

11 22

Re: [squeak-dev] Google Summer of Code?
by giovanni 26 Feb '08

26 Feb '08

--- Andreas Raab <andreas.raab(a)gmx.de> wrote: > Hi - > > Slashdot just reminded me that it's this time of the year again. > Apparently Google accepts new applications starting Monday, March 3rd > (see http://code.google.com/soc/2008). Anyone interested? > > Also, last year there were a couple of projects which are listed at > http://code.google.com/soc/2007/squeak/about.html and I realized that I > never found out about their status/completion. Is there a web page / > code / documents available on any of these? > Hi, I meant to forward the announcement yesterday, but I was too busy at home. Ken reminded me yesterday that I didn't send a final status report to the mailing list (sorry for that!); I will send a report, together with my observation/proposals for this year, either tonight or tomorrow. Giovanni

2 1

Distributing network load using HydraVM
by Igor Stasenko 26 Feb '08

26 Feb '08

My head just visited another idea, and i'd like to share it :) There was a discussion on how to pass sockets between images safely. And due to reasons that it's not safe, not easy, and require a lot changes which may break compatibility, solution was postponed for a future. Now, i think i found the way how to pass the newly created socket to another image by using a special accept primitive. What this primitive should do: by taking parameters: interpreter handle, socket handle and channel name, it should 1. do validity checks (all handles should be valid e.t.c) 2. accept new incoming socket connection (socket handle passed to primitive should be in listen mode) 3. if system fails with accept(), primitive fails 4. create an event and put new socket handle , and channel name to it. 5. enqueue given event to target interpreter event queue. -- Event handling: - register socket handle in SocketPlugin internal structures. - then find channel with given name, and put SocketHandle bytearray as packet. voila.. Now at language side, you simply need a channel, which listening for packets, containing socket handles. Once packet received, you know, that it's newly accepted socket connection and can do anything you want with it. ............ by finishing writing this, i just thought that passing ownership of already open socket can be done in similar way: a primitive can remove given socket from SocketPlugin internal structures and pass it with event to another interpreter. -- Best regards, Igor Stasenko AKA sig.

2 2

Re: Squeak-dev Digest, Vol 62, Issue 68
by Giuseppe Luigi Punzi 26 Feb '08

26 Feb '08

Is normal thath Squeak-dev digest maiol is coming without date? On squeak-dev-request(a)lists.squeakfoundation.org wrote: > Send Squeak-dev mailing list submissions to > squeak-dev(a)lists.squeakfoundation.org > > To subscribe or unsubscribe via the World Wide Web, visit > http://lists.squeakfoundation.org/cgi-bin/mailman/listinfo/squeak-dev > or, via email, send a message with subject or body 'help' to > squeak-dev-request(a)lists.squeakfoundation.org > > You can reach the person managing the list at > squeak-dev-owner(a)lists.squeakfoundation.org > > When replying, please edit your Subject line so it is more specific > than "Re: Contents of Squeak-dev digest..." > > > Today's Topics: > > 1. Re: Hydra benchmarking (Igor Stasenko) > 2. Google Summer of Code? (Andreas Raab) > 3. Re: Hydra benchmarking (John M McIntosh) > 4. Re: FFI on 3.10 Dev image? (Andrew Tween) > 5. Re: Re: Why is Heap>>#species => Array? (Andrew P. Black) > 6. Re: Craig's answers to the 2008 candidate questions > (goran(a)krampe.se) > 7. Fwd: Web page with status of squeak license issue > resolution? (Damien Pollet) > 8. Re: Why is Heap>>#species => Array? (Paolo Bonzini) > > > ---------------------------------------------------------------------- > > Message: 1 > Date: Tue, 26 Feb 2008 09:21:20 +0200 > From: "Igor Stasenko" <siguctua(a)gmail.com> > Subject: Re: [squeak-dev] Hydra benchmarking > To: "The general-purpose Squeak developers list" > <squeak-dev(a)lists.squeakfoundation.org> > Message-ID: > <4a5f5f320802252321m77263d66l67ea1d78106a8e08(a)mail.gmail.com> > Content-Type: text/plain; charset=UTF-8 > > > Also, i'd do it like following: > > > > [ HydraVM doIt: 'Transcript cr; show: ''Background: '', [ 0 > > tinyBenchmarks ] timeToRun' > > at: 2. > > Transcript cr; show: 'Foreground: ', [0 tinyBenchmarks] timeToRun. > > ] timeToRun > > > > On single core, sum of local+remote doit, should be roughly equivalent > > to outer timeToRun. > > On dual core all #timeToRun should be roughly equivalent. > > > > oops, no it shouldn't work as i described :) > > > -- > > Best regards, > > Igor Stasenko AKA sig. > > > > > -- > Best regards, > Igor Stasenko AKA sig. > > > ------------------------------ > > Message: 2 > Date: Mon, 25 Feb 2008 23:24:05 -0800 > From: Andreas Raab <andreas.raab(a)gmx.de> > Subject: [squeak-dev] Google Summer of Code? > To: The general-purpose Squeak developers list > <squeak-dev(a)lists.squeakfoundation.org> > Message-ID: <47C3BE95.1080100(a)gmx.de> > Content-Type: text/plain; charset=ISO-8859-1; format=flowed > > Hi - > > Slashdot just reminded me that it's this time of the year again. > Apparently Google accepts new applications starting Monday, March 3rd > (see http://code.google.com/soc/2008). Anyone interested? > > Also, last year there were a couple of projects which are listed at > http://code.google.com/soc/2007/squeak/about.html and I realized that I > never found out about their status/completion. Is there a web page / > code / documents available on any of these? > > Cheers, > - Andreas > > > ------------------------------ > > Message: 3 > Date: Tue, 26 Feb 2008 00:03:41 -0800 > From: John M McIntosh <johnmci(a)smalltalkconsulting.com> > Subject: Re: [squeak-dev] Hydra benchmarking > To: The general-purpose Squeak developers list > <squeak-dev(a)lists.squeakfoundation.org> > Message-ID: > <E47A4C4B-8110-475F-9F51-35DC3EAB8BE3(a)smalltalkconsulting.com> > Content-Type: text/plain; charset=US-ASCII; format=flowed; delsp=yes > > > I think it's because main thread have slightly higher priority in > > windows. > > Ah, that reminds me, you cannot do UI work on anything other than the > main thread in carbon and I think > Cocoa applications in os-x. > > When we migrated from InputSensor to EventSensor I had changed th pre- > os-x (OS8.x) squeak VM to use cooperative threading which allowed us > to run the > VM on one thread, and the UI on another. This was changed when OS-X > came out to run the interpret() on a pthread with the main pthread > handing the UI. > The UI thread used some mutex semaphores to place data on the VM event > queue. All was happy, but Apple started enforcing the rules about UI > calls only on the main thread. > > Although I had a lockUI call to swap the call or FFI call to the main > thread which worked quite well, Sophie started getting serious about > FFI calls to quicktime we ran into major problems > with it's usage of Open/GL. > > This was fixed in 3.8.12b1 when I switch to dispatching the > interpreter loop via a carbon event and back to a form of co-operative > threading. > Mac Carbon 3.8.12b1 Alter VM so that the interpreter() loop is > dispatched on the UI thread and we poll for pending UI events. This is > to allow quicktime FFI calls to work on the main thread that contains > the UI, and to avoid cross thread OpenG/L calls which seem to crash > the VM > > I note this because someday someone might want to make a UI related > FFI call on a Hydra thread and then something unexpected could > happen. More research required to understand if this is an issue with > Cocoa applications. > -- > = > = > = > ======================================================================== > John M. McIntosh <johnmci(a)smalltalkconsulting.com> > Corporate Smalltalk Consulting Ltd. http://www.smalltalkconsulting.com > = > = > = > ======================================================================== > > > > > ------------------------------ > > Message: 4 > Date: Tue, 26 Feb 2008 08:05:36 -0000 > From: "Andrew Tween" <amtween(a)hotmail.com> > Subject: [squeak-dev] Re: FFI on 3.10 Dev image? > To: squeak-dev(a)lists.squeakfoundation.org > Message-ID: <fq0h9b$hh$1(a)ger.gmane.org> > Content-Type: text/plain; charset="iso-8859-1" > > Hi, > I noticed the same problem and tracked it down to this... > ExternalStructure class implements doneCompiling, and doneCompiling is sent > during the MC1.5 loading process. > (specifically - it is sent from > ClassBuilder>>name:inEnvironment:subclassOf:type:instanceVariableNames:classVariableNames:poolDictionaries:category:unsafe: > ) > > ExternalStructure class>>doneCompiling then calls ExternalType > class>>noticeModificationOf: , which then accesses its StructTypes class > var. > But ExternalType hasn't been sent #initialize yet, so its StructTypes class > var is nil. > > I presume that the old MC loader loads/initializes things in a different > order, which is why the problem doesn't occur in a standard 3.10 image. > > To workaround the problem, and get FFI loaded, I hacked > ClassBuilder>>name:inEnvironment:subclassOf:type:instanceVariableNames:classVariableNames:poolDictionaries:category:unsafe: > so that errors occuring during doneCompiling are ignored. (hacked method > attached). > FFI then loads ok from Universes, and seems to function correctly. > > I've also noticed that preambles and postscripts aren't being run by the > MC1.5 loader, which causes problems for other packages (such as FreeType > Plus). > > I have added both these issues to Mantis... > http://bugs.squeak.org/view.php?id=6952 > http://bugs.squeak.org/view.php?id=6953 > > Cheers, > Andy > > > > "Ken G. Brown" <kbrown(a)mac.com> wrote in message > news:p0624043cc3e90f0bbeb8@[169.254.1.2]... > > Is FFI supposed to load into 3.10? > > When I try loading FFI() from SqueakMap into a fresh sq3.10-7159dev08.02.1 > > I get the following failure: > > > > Ken G. Brown > > ----------------------------------------------------------------------------------- > > VM: Mac OS - a SmalltalkImage > > Image: Squeak3.10beta [latest update: #7159] > > > > SecurityManager state: > > Restricted: false > > FileAccess: true > > SocketAccess: true > > Working Dir /mySqueakStuff/Sqkb/Sqkb3.10/sq3.10-7159dev08.02.1 > > Trusted Dir /foobar/tooBar/forSqueak/bogus > > Untrusted Dir /Users/kbrownMPro/Library/Preferences/Squeak/Internet/My > > Squeak > > > > UndefinedObject(Object)>>doesNotUnderstand: #at:ifAbsent: > > Receiver: nil > > Arguments and temporary variables: > > aMessage: at: #ExternalStructure ifAbsent: [] in ExternalType > > class>>noticeModi...etc... > > Receiver's instance variables: > > nil > > > > [] in ExternalType class>>noticeModificationOf: {[:cls | type := > > StructTypes at: cls name ifAbsent: []. type ifNo...]} > > Arguments and temporary variables: > > aClass: ExternalStructure > > type: nil > > cls: ExternalStructure > > > > ExternalStructure class(Behavior)>>withAllSubclassesDo: > > Receiver: ExternalStructure > > Arguments and temporary variables: > > aBlock: [] in ExternalType class>>noticeModificationOf: {[:cls | type := > > Struc...etc... > > Receiver's instance variables: > > superclass: ExternalObject > > methodDict: a MethodDictionary(#free->a CompiledMethod (1328) > > #longPrintOn:->a ...etc... > > format: 132 > > instanceVariables: nil > > organization: ('as yet unclassified') > > > > subclasses: {ExternalData} > > name: #ExternalStructure > > classPool: nil > > sharedPools: an OrderedCollection(FFIConstants) > > environment: a SystemDictionary(lots of globals) > > category: #'FFI-Kernel' > > traitComposition: {} > > localSelectors: nil > > compiledSpec: nil > > > > ExternalType class>>noticeModificationOf: > > Receiver: ExternalType > > Arguments and temporary variables: > > aClass: ExternalStructure > > type: nil > > cls: ExternalStructure > > Receiver's instance variables: > > superclass: Object > > methodDict: a MethodDictionary(#asNonPointerType->a CompiledMethod (3691) > > #asPo...etc... > > format: 136 > > instanceVariables: #('compiledSpec' 'referentClass' 'referencedType') > > organization: ('as yet unclassified') > > > > subclasses: nil > > name: #ExternalType > > classPool: a Dictionary(#AtomicSelectors->nil #AtomicTypeNames->nil > > #AtomicType...etc... > > sharedPools: an OrderedCollection(FFIConstants) > > environment: a SystemDictionary(lots of globals) > > category: #'FFI-Kernel' > > traitComposition: {} > > localSelectors: nil > > > > > > --- The full stack --- > > UndefinedObject(Object)>>doesNotUnderstand: #at:ifAbsent: > > [] in ExternalType class>>noticeModificationOf: {[:cls | type := > > StructTypes at: cls name ifAbsent: []. type ifNo...]} > > ExternalStructure class(Behavior)>>withAllSubclassesDo: > > ExternalType class>>noticeModificationOf: > > - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - > > ExternalStructure class>>doneCompiling > > [] in > > ClassBuilder>>name:inEnvironment:subclassOf:type:instanceVariableNames:classVariableNames:poolDictionaries:category:unsafe: > > {[unsafe ifFalse: [(self validateSuperclass: newSuper forSubclass: > > oldClass...]} > > BlockContext>>ensure: > > ClassBuilder>>name:inEnvironment:subclassOf:type:instanceVariableNames:classVariableNames:poolDictionaries:category:unsafe: > > ClassBuilder>>name:inEnvironment:subclassOf:type:instanceVariableNames:classVariableNames:poolDictionaries:category: > > MCClassDefinition>>createUnionClassWith: > > MCClassDefinition>>postinstall > > [] in MCPackageLoader1b>>basicLoad {[:ea | ea postinstall]} > > OrderedCollection>>do: > > MCPackageLoader1b>>basicLoad > > [] in MCPackageLoader1b>>protectedLoad {[self basicLoad]} > > BlockContext>>on:do: > > [] in MCPackageLoader1b>>protectedLoad {[[self basicLoad] on: > > InMidstOfFileinNotification do: [:n | n resume: tr...]} > > BlockContext>>ensure: > > [] in MCPackageLoader1b>>protectedLoad {[[[self basicLoad] on: > > InMidstOfFileinNotification do: [:n | n resume: t...]} > > BlockContext>>on:do: > > MCPackageLoader1b>>protectedLoad > > [] in MCPackageLoader1b>>loadWithNameLike: {[self protectedLoad]} > > [] in MCPackageLoader1b>>useChangeSetNamed:during: {[aBlock value]} > > BlockContext>>ensure: > > MCPackageLoader1b>>useChangeSetNamed:during: > > MCPackageLoader1b>>useNewChangeSetNamedLike:during: > > MCPackageLoader1b>>loadWithNameLike: > > MCVersionLoader>>loadWithNameLike: > > MCVersionLoader>>load > > MCVersionLoader class>>loadVersion: > > MCVersion>>load > > [] in UndefinedObject>>DoIt {[:t2 :t3 | t1 value: t2 * 2 - 1. reader := > > repo versionReaderForFileNamed:...]} > > Array(SequenceableCollection)>>keysAndValuesDo: > > [] in UndefinedObject>>DoIt {[:t1 | #('FFI-Kernel-ar.8.mcz' > > 'FFI-Tests-ar.2.mcz' 'FFI-Examples-ar.1.mcz' ...]} > > ...etc... > > > > >

1 0

Google Summer of Code?
by Andreas Raab 26 Feb '08

26 Feb '08

Hi - Slashdot just reminded me that it's this time of the year again. Apparently Google accepts new applications starting Monday, March 3rd (see http://code.google.com/soc/2008). Anyone interested? Also, last year there were a couple of projects which are listed at http://code.google.com/soc/2007/squeak/about.html and I realized that I never found out about their status/completion. Is there a web page / code / documents available on any of these? Cheers, - Andreas

2 1

[ANN] Hydra VM: A multi-core capable Croquet VM
by Andreas Raab 26 Feb '08

26 Feb '08

Folks - In its ongoing commitment to Open Source, Qwaq has decided to release the first multi-core capable Croquet VM under the MIT license. Hydra VM is a virtual machine capable of running multiple Croquet images side-by-side, therefore being able to effectively utilize multi-core CPUs. Hydra VM is still in an early phase of development, but given the (very successful) early results, Qwaq has decided to make it available to general public in the hope that other people will help Igor to make further progress in developing Hydra VM. The offical release is available here: http://www.qwaq.com/HydraVM and contains the following files: * HydraVM-bin.zip - A prebuilt Windows version of Hydra VM. * HydraVM-src.zip - The (C) source tree for building Hydra VM. * HydraVM-pkg.zip - The (Squeak) packages needed to build Hydra VM. Enjoy! Cheers, - Andreas (on behalf of Qwaq)

10 40

FFI on 3.10 Dev image?
by Ken G. Brown 26 Feb '08

26 Feb '08

Is FFI supposed to load into 3.10? When I try loading FFI() from SqueakMap into a fresh sq3.10-7159dev08.02.1 I get the following failure: Ken G. Brown ----------------------------------------------------------------------------------- VM: Mac OS - a SmalltalkImage Image: Squeak3.10beta [latest update: #7159] SecurityManager state: Restricted: false FileAccess: true SocketAccess: true Working Dir /mySqueakStuff/Sqkb/Sqkb3.10/sq3.10-7159dev08.02.1 Trusted Dir /foobar/tooBar/forSqueak/bogus Untrusted Dir /Users/kbrownMPro/Library/Preferences/Squeak/Internet/My Squeak UndefinedObject(Object)>>doesNotUnderstand: #at:ifAbsent: Receiver: nil Arguments and temporary variables: aMessage: at: #ExternalStructure ifAbsent: [] in ExternalType class>>noticeModi...etc... Receiver's instance variables: nil [] in ExternalType class>>noticeModificationOf: {[:cls | type := StructTypes at: cls name ifAbsent: []. type ifNo...]} Arguments and temporary variables: aClass: ExternalStructure type: nil cls: ExternalStructure ExternalStructure class(Behavior)>>withAllSubclassesDo: Receiver: ExternalStructure Arguments and temporary variables: aBlock: [] in ExternalType class>>noticeModificationOf: {[:cls | type := Struc...etc... Receiver's instance variables: superclass: ExternalObject methodDict: a MethodDictionary(#free->a CompiledMethod (1328) #longPrintOn:->a ...etc... format: 132 instanceVariables: nil organization: ('as yet unclassified') subclasses: {ExternalData} name: #ExternalStructure classPool: nil sharedPools: an OrderedCollection(FFIConstants) environment: a SystemDictionary(lots of globals) category: #'FFI-Kernel' traitComposition: {} localSelectors: nil compiledSpec: nil ExternalType class>>noticeModificationOf: Receiver: ExternalType Arguments and temporary variables: aClass: ExternalStructure type: nil cls: ExternalStructure Receiver's instance variables: superclass: Object methodDict: a MethodDictionary(#asNonPointerType->a CompiledMethod (3691) #asPo...etc... format: 136 instanceVariables: #('compiledSpec' 'referentClass' 'referencedType') organization: ('as yet unclassified') subclasses: nil name: #ExternalType classPool: a Dictionary(#AtomicSelectors->nil #AtomicTypeNames->nil #AtomicType...etc... sharedPools: an OrderedCollection(FFIConstants) environment: a SystemDictionary(lots of globals) category: #'FFI-Kernel' traitComposition: {} localSelectors: nil --- The full stack --- UndefinedObject(Object)>>doesNotUnderstand: #at:ifAbsent: [] in ExternalType class>>noticeModificationOf: {[:cls | type := StructTypes at: cls name ifAbsent: []. type ifNo...]} ExternalStructure class(Behavior)>>withAllSubclassesDo: ExternalType class>>noticeModificationOf: - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ExternalStructure class>>doneCompiling [] in ClassBuilder>>name:inEnvironment:subclassOf:type:instanceVariableNames:classVariableNames:poolDictionaries:category:unsafe: {[unsafe ifFalse: [(self validateSuperclass: newSuper forSubclass: oldClass...]} BlockContext>>ensure: ClassBuilder>>name:inEnvironment:subclassOf:type:instanceVariableNames:classVariableNames:poolDictionaries:category:unsafe: ClassBuilder>>name:inEnvironment:subclassOf:type:instanceVariableNames:classVariableNames:poolDictionaries:category: MCClassDefinition>>createUnionClassWith: MCClassDefinition>>postinstall [] in MCPackageLoader1b>>basicLoad {[:ea | ea postinstall]} OrderedCollection>>do: MCPackageLoader1b>>basicLoad [] in MCPackageLoader1b>>protectedLoad {[self basicLoad]} BlockContext>>on:do: [] in MCPackageLoader1b>>protectedLoad {[[self basicLoad] on: InMidstOfFileinNotification do: [:n | n resume: tr...]} BlockContext>>ensure: [] in MCPackageLoader1b>>protectedLoad {[[[self basicLoad] on: InMidstOfFileinNotification do: [:n | n resume: t...]} BlockContext>>on:do: MCPackageLoader1b>>protectedLoad [] in MCPackageLoader1b>>loadWithNameLike: {[self protectedLoad]} [] in MCPackageLoader1b>>useChangeSetNamed:during: {[aBlock value]} BlockContext>>ensure: MCPackageLoader1b>>useChangeSetNamed:during: MCPackageLoader1b>>useNewChangeSetNamedLike:during: MCPackageLoader1b>>loadWithNameLike: MCVersionLoader>>loadWithNameLike: MCVersionLoader>>load MCVersionLoader class>>loadVersion: MCVersion>>load [] in UndefinedObject>>DoIt {[:t2 :t3 | t1 value: t2 * 2 - 1. reader := repo versionReaderForFileNamed:...]} Array(SequenceableCollection)>>keysAndValuesDo: [] in UndefinedObject>>DoIt {[:t1 | #('FFI-Kernel-ar.8.mcz' 'FFI-Tests-ar.2.mcz' 'FFI-Examples-ar.1.mcz' ...]} ...etc...

3 2

Hydra benchmarking
by Andreas Raab 26 Feb '08

26 Feb '08

Hi - I was demoing Hydra at VPRI today and Yoshiki came up with a nice little idea for a Hydra benchmark. It basically goes like this: HydraVM doIt: 'Transcript cr; show: ''Background: '', 0 tinyBenchmarks' at: 2. Transcript cr; show: 'Foreground: ', 0 tinyBenchmarks. The interesting part is that on a multi-core box it "does the right thing" which means it basically prints the same results in roughly the same time. What is weird is that on a single-core box, I get the "same results" but for the wrong reasons and apparently offset by a couple of seconds. It behaves as if first the foreground process runs and then the background process. Anyone having an idea how to fix the benchmark so that it comes up with the correct numbers on a single core box? [BTW, if you have dual-core I'm pretty sure that extending the above by a couple more images will show similarly skewed results] Cheers, - Andreas

3 3

Socket>>#waitForDataFor: and #primSocketReceiveDataAvailable
by Klaus D. Witzel 26 Feb '08

26 Feb '08

I'm trying to provoke the Exceptions which Socket>>#waitForDataFor: would signal on failure (and test that it does so), to no avail (no, the .image isn't locked and the process doesn't hang). The below runs in both the Squeak 3.9,3.8 final .images with both the 3.10.2,3.10.6 VMs on win32; the worker process always terminates without an exception. The implementation of #waitForDataFor: (#waitForDataFor:ifClosed:ifTimedOut:) looks like #dataAvailable returns true but there cannot be any data since nothing was ever sent. *but* after closing the client side, the other side's #dataAvailable (#primSocketReceiveDataAvailable) returns true ... Any hint(s) on what's going on (or wrong) will be recognized with gratitude. /Klaus | localhost serverSocket sessionSocket clientSocket worker | localhost := NetNameResolver localHostAddress. "warning:" Socket allInstancesDo: [:socket | socket destroy]. serverSocket := Socket newTCP listenOn: 7680 backlogSize: 1. clientSocket := Socket newTCP connectNonBlockingTo: localhost port: 7680. sessionSocket := serverSocket waitForAcceptFor: 1. worker := [sessionSocket waitForDataFor: 1] fork. Processor yield. Transcript cr; show: serverSocket statusString, ' [server]'; cr; show: clientSocket statusString, ' [client]'; cr; show: sessionSocket statusString, ' [session], dataAvailable: ' , sessionSocket dataAvailable storeString; cr; show: 'worker: ', worker browserPrintString. [clientSocket close] on: Exception do: [:ex|]. Processor yield. Transcript cr; show: serverSocket statusString, ' [server]'; cr; show: clientSocket statusString, ' [client]'; cr; show: sessionSocket statusString, ' [session], dataAvailable: ' , sessionSocket dataAvailable storeString; cr; show: 'worker: ', worker browserPrintString; cr. [sessionSocket destroy] on: Exception do: [:ex|]. [clientSocket destroy] on: Exception do: [:ex|]. [serverSocket destroy] on: Exception do: [:ex|]

3 5

← Newer
1
2
3
4
5
6
7
8
9
...
22
Older →

Jump to page: