October 2013 - Pharo-dev - lists.pharo.org

Re: [Pharo-dev] Please try to reproduce this bug

by Tudor Girba

I just tried on Windows with: http://files.pharo.org/platform/Pharo2.0-win.zip and it works fine. Doru On Mon, Oct 7, 2013 at 11:04 AM, Goubier Thierry <thierry.goubier(a)cea.fr>wrote: > In short: > > does not happens if you use the get.pharo.org vm. > > Happens if you use the ubuntu pharo ppa. > > (ubuntu 13.04 64bits) > > > Thierry > > Le 07/10/2013 10:12, Damien Cassou a Ã©crit : > >> Can you please try to reproduce this bug: >> >> >> 1- Open a fresh Pharo 2.0 >> 2- Open Monticello >> 3- Select the package AST-Core >> 4- Click browse >> >> you get a "PharoV20.sources does not exist." error message. You can't >> escape. >> >> Please write your findings in >> https://pharo.fogbugz.com/f/**cases/11803/PharoV20-sources-** >> does-not-exist-when-browsing-**with-Monticello<https://pharo.fogbugz.com/f/cases/11803/PharoV20-sources-does-not-exist-whe…> >> . >> >> For me, it looks like the problem is in MCPackage>>snapshot >> >> > -- > Thierry Goubier > CEA list > Laboratoire des Fondations des SystÃ¨mes Temps RÃ©el EmbarquÃ©s > 91191 Gif sur Yvette Cedex > France > Phone/Fax: +33 (0) 1 69 08 32 92 / 83 95 > > -- www.tudorgirba.com "Every thing has its own flow"

Oct. 7, 2013

Re: [Pharo-dev] [update 3.0] #30448

by Pavel Krivanek

Hi, place two BookmarksPeriscopePlugin into Periscope class >> #defaultPlugins and then eval Periscope reset. You will see only one button bar but you should see two. The window you see is not empty, it contains list that is part of the BookmarksPeriscopePlugin but it should contain some buttons too. -- Pavel 2013/10/7 Benjamin <Benjamin.VanRyseghem.Pharo(a)gmail.com>: > I am trying your code right now, and everything I get so far is an empty > Periscope window. > > What should I do ? :) > > Ben > > On Oct 5, 2013, at 10:12 PM, Pavel Krivanek <pavel.krivanek(a)gmail.com> > wrote: > > Hi Ben, > > Gofer new > smalltalkhubUser: 'PavelKrivanek' project: 'Periscope'; > package: 'Periscope'; > load. > > Nautilus pluginClasses add: {(Smalltalk at: #NautilusPeriscopePlugin). > #none} . > > Very useful code: ;-) > > Periscope reset > > It still without experiments with own Morphic bindings, that will be > the next step :-) Feel free to commit any changes. > > Chees, > -- Pavel > > 2013/10/5 Benjamin <Benjamin.VanRyseghem.Pharo(a)gmail.com>: > > Can you point me where to load your code and how to end up in a situation > where I can play with the spec ? :P > > Ben > > On Oct 4, 2013, at 11:11 PM, Pavel Krivanek <pavel.krivanek(a)gmail.com> > wrote: > > Damn, the last row of the last plugins fills all the window... > > -- Pavel > > 2013/10/4 Pavel Krivanek <pavel.krivanek(a)gmail.com>: > > Thanks, this seems to work: > > defaultSpec > > | layout | > layout := SpecLayout composed. > self pluginClasses do: [ :c | > layout add: {#model. #class. #pluginOf:. c. #defaultSpec} ]. > > ^ layout. > > > plugin code on the instance side looks like this: > > defaultSpec > > ^ SpecLayout composed > newRow:[: r | > r add: #finderButtonModel; > add: #browserButtonModel] > height: 25. > > > -- Pavel > > 2013/10/4 Benjamin <Benjamin.VanRyseghem.Pharo(a)gmail.com>: > > Since you are doing some really edge use of Spec, > it's not covered by Spec easily. > > But if on you model you have a method returning your pluginClasses, you > could do maybe: > > | layout | > layout := SpecLayout composed. > self pluginClasses do: [ :c | > layout add: {#model. pluginOf:. c} ] > > I am not sure it works, but it worth a try :P > > Ben > > On Oct 4, 2013, at 9:28 PM, Pavel Krivanek <pavel.krivanek(a)gmail.com> wrote: > > This was the original code for the spec that embeds plugins. As you > can see, it had to use ugly addMorph:. I do not think that I need to > do something special. Has the current Spec a better way how to do > that? > > defaultSpec > > | spec | > spec := OrderedCollection with: #ContainerModel. > spec addAll: {#changeTableLayout. > #listDirection:. #topToBottom. > #hResizing:. #spaceFill. > #vResizing:. #spaceFill.}. > self pluginClasses do: [:c | > spec add: #addMorph:. > spec add: {(self pluginOf: c) defaultSpec}]. > ^ spec asArray. > > 2013/10/4 Pavel Krivanek <pavel.krivanek(a)gmail.com>: > > In other words... I have a model that has some plugins. I need to > generate a spec that embeds specs generated for each plugin. Models > for this embedded specs must be the plugins, not the original model. > > -- Pavel > > 2013/10/4 Pavel Krivanek <pavel.krivanek(a)gmail.com>: > > well, and the version with "SpecLayout composed" when I do not need > direct fractions? > > ^ SpecLayout composed > newRow:[: r | > r add: #finderButtonModel] > height: 25 > > How to send #finderButtonModel to different object than SpecInterpreter > model? > > -- Pavel > > 2013/10/4 Benjamin <Benjamin.VanRyseghem.Pharo(a)gmail.com>: > > ^{ #ContainerModel. > > #vShrinkWrap. > > #add:. {{self. #finderButtonModel}. > > #layout:. #(FrameLayout > > rightFraction: 0.125)}. > > #add:. {{self. #browserButtonModel}. > > #layout:. #(FrameLayout > > leftFraction: 0.125 > rightFraction: 0.25)}} > > This should work, but looks strange though :P > > Ben > > On Oct 4, 2013, at 8:13 PM, Pavel Krivanek <pavel.krivanek(a)gmail.com> wrote: > > Ben, > > how now should look code like this: > > defaultSpec > > ^{ #ComposableSpec. > #vResizing:. #shrinkWrap. > #add:. {{self. #finderButtonModel}. > #layout:. #(FrameLayout > rightFraction: 0.125)}. > #add:. {{self. #browserButtonModel}. > #layout:. #(FrameLayout > leftFraction: 0.125 > rightFraction: 0.25)}} > > -- Pavel > > 2013/10/4 Benjamin <Benjamin.VanRyseghem.Pharo(a)gmail.com>: > > For Spec users: > > The change is mostly backward compatible :) > 2 things really changed: > - direct access to a model's widget is not returning what it used to return. > This is done on purpose since > it's very often a bad idea to directly access the widget. > - some behaviour on trees are now only available after the tree build. This > is because some properties > have been moved on the adapter directly, which is reachable only after the > model build. > You can use `whenBuiltDo:` to add post-build actions. > > Enjoy this new Spec :P > > Ben > > On Oct 4, 2013, at 2:31 PM, Marcus Denker <marcus.denker(a)inria.fr> wrote: > > 30448 > ----- > > 11740 Remove Morphic dependency from Spec > https://pharo.fogbugz.com/f/cases/11740 > > A Huge change from Ben... > > Diff information: > http://smalltalkhub.com/mc/Pharo/Pharo30/main/Tools-MarcusDenker.1292.diff > http://smalltalkhub.com/mc/Pharo/Pharo30/main/Tabs-MarcusDenker.33.diff > http://smalltalkhub.com/mc/Pharo/Pharo30/main/Spec-Widgets-MarcusDenker.265… > http://smalltalkhub.com/mc/Pharo/Pharo30/main/Spec-Tools-MarcusDenker.150.d… > http://smalltalkhub.com/mc/Pharo/Pharo30/main/Spec-Tests-MarcusDenker.35.di… > http://smalltalkhub.com/mc/Pharo/Pharo30/main/Spec-MorphicAdapters-MarcusDe… > http://smalltalkhub.com/mc/Pharo/Pharo30/main/Spec-PolyWidgets-MarcusDenker… > http://smalltalkhub.com/mc/Pharo/Pharo30/main/Spec-Layout-MarcusDenker.58.d… > http://smalltalkhub.com/mc/Pharo/Pharo30/main/Spec-Inspector-MarcusDenker.1… > http://smalltalkhub.com/mc/Pharo/Pharo30/main/Spec-Examples-MarcusDenker.48… > http://smalltalkhub.com/mc/Pharo/Pharo30/main/Spec-Debugger-MarcusDenker.16… > http://smalltalkhub.com/mc/Pharo/Pharo30/main/Spec-Core-MarcusDenker.175.di… > http://smalltalkhub.com/mc/Pharo/Pharo30/main/Spec-Builder-MarcusDenker.43.… > http://smalltalkhub.com/mc/Pharo/Pharo30/main/Spec-Bindings-MarcusDenker.35… > http://smalltalkhub.com/mc/Pharo/Pharo30/main/NautilusRefactoring-MarcusDen… > http://smalltalkhub.com/mc/Pharo/Pharo30/main/Spec-Adapters-MarcusDenker.3.… > http://smalltalkhub.com/mc/Pharo/Pharo30/main/Manifest-CriticBrowser-Marcus… > http://smalltalkhub.com/mc/Pharo/Pharo30/main/Keymapping-Tools-Spec-MarcusD… > http://smalltalkhub.com/mc/Pharo/Pharo30/main/KeyChain-MarcusDenker.58.diff > http://smalltalkhub.com/mc/Pharo/Pharo30/main/HudsonBuildTools20-MarcusDenk… > > > > > > > > > > >

Oct. 7, 2013

Re: [Pharo-dev] [update 3.0] #30448

by Benjamin

I am trying your code right now, and everything I get so far is an empty Periscope window. What should I do ? :) Ben On Oct 5, 2013, at 10:12 PM, Pavel Krivanek <pavel.krivanek(a)gmail.com> wrote: > Hi Ben, > > Gofer new > smalltalkhubUser: 'PavelKrivanek' project: 'Periscope'; > package: 'Periscope'; > load. > > Nautilus pluginClasses add: {(Smalltalk at: #NautilusPeriscopePlugin). #none} . > > Very useful code: ;-) > > Periscope reset > > It still without experiments with own Morphic bindings, that will be > the next step :-) Feel free to commit any changes. > > Chees, > -- Pavel > > 2013/10/5 Benjamin <Benjamin.VanRyseghem.Pharo(a)gmail.com>: >> Can you point me where to load your code and how to end up in a situation >> where I can play with the spec ? :P >> >> Ben >> >> On Oct 4, 2013, at 11:11 PM, Pavel Krivanek <pavel.krivanek(a)gmail.com> >> wrote: >> >> Damn, the last row of the last plugins fills all the window... >> >> -- Pavel >> >> 2013/10/4 Pavel Krivanek <pavel.krivanek(a)gmail.com>: >> >> Thanks, this seems to work: >> >> defaultSpec >> >> | layout | >> layout := SpecLayout composed. >> self pluginClasses do: [ :c | >> layout add: {#model. #class. #pluginOf:. c. #defaultSpec} ]. >> >> ^ layout. >> >> >> plugin code on the instance side looks like this: >> >> defaultSpec >> >> ^ SpecLayout composed >> newRow:[: r | >> r add: #finderButtonModel; >> add: #browserButtonModel] >> height: 25. >> >> >> -- Pavel >> >> 2013/10/4 Benjamin <Benjamin.VanRyseghem.Pharo(a)gmail.com>: >> >> Since you are doing some really edge use of Spec, >> it's not covered by Spec easily. >> >> But if on you model you have a method returning your pluginClasses, you >> could do maybe: >> >> | layout | >> layout := SpecLayout composed. >> self pluginClasses do: [ :c | >> layout add: {#model. pluginOf:. c} ] >> >> I am not sure it works, but it worth a try :P >> >> Ben >> >> On Oct 4, 2013, at 9:28 PM, Pavel Krivanek <pavel.krivanek(a)gmail.com> wrote: >> >> This was the original code for the spec that embeds plugins. As you >> can see, it had to use ugly addMorph:. I do not think that I need to >> do something special. Has the current Spec a better way how to do >> that? >> >> defaultSpec >> >> | spec | >> spec := OrderedCollection with: #ContainerModel. >> spec addAll: {#changeTableLayout. >> #listDirection:. #topToBottom. >> #hResizing:. #spaceFill. >> #vResizing:. #spaceFill.}. >> self pluginClasses do: [:c | >> spec add: #addMorph:. >> spec add: {(self pluginOf: c) defaultSpec}]. >> ^ spec asArray. >> >> 2013/10/4 Pavel Krivanek <pavel.krivanek(a)gmail.com>: >> >> In other words... I have a model that has some plugins. I need to >> generate a spec that embeds specs generated for each plugin. Models >> for this embedded specs must be the plugins, not the original model. >> >> -- Pavel >> >> 2013/10/4 Pavel Krivanek <pavel.krivanek(a)gmail.com>: >> >> well, and the version with "SpecLayout composed" when I do not need >> direct fractions? >> >> ^ SpecLayout composed >> newRow:[: r | >> r add: #finderButtonModel] >> height: 25 >> >> How to send #finderButtonModel to different object than SpecInterpreter >> model? >> >> -- Pavel >> >> 2013/10/4 Benjamin <Benjamin.VanRyseghem.Pharo(a)gmail.com>: >> >> ^{ #ContainerModel. >> >> #vShrinkWrap. >> >> #add:. {{self. #finderButtonModel}. >> >> #layout:. #(FrameLayout >> >> rightFraction: 0.125)}. >> >> #add:. {{self. #browserButtonModel}. >> >> #layout:. #(FrameLayout >> >> leftFraction: 0.125 >> rightFraction: 0.25)}} >> >> This should work, but looks strange though :P >> >> Ben >> >> On Oct 4, 2013, at 8:13 PM, Pavel Krivanek <pavel.krivanek(a)gmail.com> wrote: >> >> Ben, >> >> how now should look code like this: >> >> defaultSpec >> >> ^{ #ComposableSpec. >> #vResizing:. #shrinkWrap. >> #add:. {{self. #finderButtonModel}. >> #layout:. #(FrameLayout >> rightFraction: 0.125)}. >> #add:. {{self. #browserButtonModel}. >> #layout:. #(FrameLayout >> leftFraction: 0.125 >> rightFraction: 0.25)}} >> >> -- Pavel >> >> 2013/10/4 Benjamin <Benjamin.VanRyseghem.Pharo(a)gmail.com>: >> >> For Spec users: >> >> The change is mostly backward compatible :) >> 2 things really changed: >> - direct access to a model's widget is not returning what it used to return. >> This is done on purpose since >> it's very often a bad idea to directly access the widget. >> - some behaviour on trees are now only available after the tree build. This >> is because some properties >> have been moved on the adapter directly, which is reachable only after the >> model build. >> You can use `whenBuiltDo:` to add post-build actions. >> >> Enjoy this new Spec :P >> >> Ben >> >> On Oct 4, 2013, at 2:31 PM, Marcus Denker <marcus.denker(a)inria.fr> wrote: >> >> 30448 >> ----- >> >> 11740 Remove Morphic dependency from Spec >> https://pharo.fogbugz.com/f/cases/11740 >> >> A Huge change from Ben... >> >> Diff information: >> http://smalltalkhub.com/mc/Pharo/Pharo30/main/Tools-MarcusDenker.1292.diff >> http://smalltalkhub.com/mc/Pharo/Pharo30/main/Tabs-MarcusDenker.33.diff >> http://smalltalkhub.com/mc/Pharo/Pharo30/main/Spec-Widgets-MarcusDenker.265… >> http://smalltalkhub.com/mc/Pharo/Pharo30/main/Spec-Tools-MarcusDenker.150.d… >> http://smalltalkhub.com/mc/Pharo/Pharo30/main/Spec-Tests-MarcusDenker.35.di… >> http://smalltalkhub.com/mc/Pharo/Pharo30/main/Spec-MorphicAdapters-MarcusDe… >> http://smalltalkhub.com/mc/Pharo/Pharo30/main/Spec-PolyWidgets-MarcusDenker… >> http://smalltalkhub.com/mc/Pharo/Pharo30/main/Spec-Layout-MarcusDenker.58.d… >> http://smalltalkhub.com/mc/Pharo/Pharo30/main/Spec-Inspector-MarcusDenker.1… >> http://smalltalkhub.com/mc/Pharo/Pharo30/main/Spec-Examples-MarcusDenker.48… >> http://smalltalkhub.com/mc/Pharo/Pharo30/main/Spec-Debugger-MarcusDenker.16… >> http://smalltalkhub.com/mc/Pharo/Pharo30/main/Spec-Core-MarcusDenker.175.di… >> http://smalltalkhub.com/mc/Pharo/Pharo30/main/Spec-Builder-MarcusDenker.43.… >> http://smalltalkhub.com/mc/Pharo/Pharo30/main/Spec-Bindings-MarcusDenker.35… >> http://smalltalkhub.com/mc/Pharo/Pharo30/main/NautilusRefactoring-MarcusDen… >> http://smalltalkhub.com/mc/Pharo/Pharo30/main/Spec-Adapters-MarcusDenker.3.… >> http://smalltalkhub.com/mc/Pharo/Pharo30/main/Manifest-CriticBrowser-Marcus… >> http://smalltalkhub.com/mc/Pharo/Pharo30/main/Keymapping-Tools-Spec-MarcusD… >> http://smalltalkhub.com/mc/Pharo/Pharo30/main/KeyChain-MarcusDenker.58.diff >> http://smalltalkhub.com/mc/Pharo/Pharo30/main/HudsonBuildTools20-MarcusDenk… >> >> >> >> >> >> >> >> >> >

Oct. 7, 2013

Re: [Pharo-dev] external semaphoresâ¦again

by Marcus Denker

On Oct 7, 2013, at 11:28 AM, Henrik Johansen <henrik.s.johansen(a)veloxit.no> wrote: > > On Oct 7, 2013, at 11:16 , Norbert Hartl <norbert(a)hartl.name> wrote: >> >> I have reported this once but got no feedback so I like to have a few opinions. >> >> The report is here: https://pharo.fogbugz.com/f/cases/10839/ >> >> Norbert > > The rationale is that inProduction would be some global setting, not yet in place when the code was writtenâ¦ > Excessive simultaneous Semaphore usage is something that should be caught during development, in which case it's better to get an active notification, than having it logged somewhere. > If I've understood correctly, it's moot on newer Pharo VM's, where there's no limit on the semtable size, but for legacy code a startup item setting size using maxExternalObjectsSilently: (as suggested in the Warning text), is still a more proper fix than setting inProduction to true and crossing your fingers hoping no signals will be lost during table growth. > The problem with this "set this secret flag in deployment" is that nobody knows it. *And* I am convinced that *everything* that is different in development to deployment will come back and hurt you *hart* when you least expect it. Marcus

Oct. 7, 2013

Re: [Pharo-dev] 100% CPU usage on image resume

by Camillo Bruni

which pharo version? which vm version? which OS? On 2013-10-07, at 11:19, Holger Hans Peter Freyther <holger(a)freyther.de> wrote: > Hi, > > I saved an image and today I wanted to resume it. Everything comes up, > for a second I can interact with it (e.g. move a window) but then the > pharo VM goes to 100% CPU usage. I let it run while writing this email > and it eventually aborted: > > C stack backtrace: > ./pharo[0x80a0c0c] > ./pharo(error+0x19)[0x80a0f69] > ./pharo[0x80785d3] > ./pharo[0x80786e1] > ./pharo[0x8078f10] > ./pharo[0x8090d97] > ./pharo(ceStackOverflow+0x6a)[0x80946ea] > [0x76ef122e] > ./pharo(interpret+0x7a6)[0x8094f36] > ./pharo(main+0x2b3)[0x80a18b3] > /lib/i386-linux-gnu/i686/cmov/libc.so.6(__libc_start_main+0xf5)[0xb74bb8c5] > > > Smalltalk stack dump: > 0xbfd5c610 M SmallInteger(Number)>negative 0x3841=7200 > 0xbfd5c630 M SmallInteger(Integer)>* 0x3841=7200 > 0xbfd5c64c M SmallInteger>* 0x3841=7200 > 0xbfd5c668 M Duration>asNanoSeconds 0x790ddef4: a(n) Duration > 0xbfd5c684 M Duration>= 0x7fe004a0: a(n) Duration > 0xbfd5c6a0 M DateAndTime>asLocal 0x7fe00488: a(n) DateAndTime > 0xbfd5c6c0 M DateAndTime>- 0x7fe0040c: a(n) DateAndTime > 0xbfd5c6dc M DateAndTime>asSeconds 0x7fe0040c: a(n) DateAndTime > 0xbfd5c700 M DateAndTime>< 0x78d2a5f8: a(n) DateAndTime > 0xbfd5c71c M DateAndTime(Magnitude)>> 0x7fe0040c: a(n) DateAndTime > 0xbfd5c73c M TimerScheduler>fireTimers: 0x77ef49ac: a(n) TimerScheduler > 0xbfd5c764 I [] in TimerScheduler>runTimers 0x77ef49ac: a(n) TimerScheduler > 0xbfd5c780 M BlockClosure>on:do: 0x78d2a60c: a(n) BlockClosure > 0xbfd5c7a4 M [] in Dispatcher>dispatch 0x77ef4764: a(n) Dispatcher > 0xbfd5c7c4 M BlockClosure>ensure: 0x78d2a798: a(n) BlockClosure > 0xbfd5c7f0 I [] in Dispatcher>dispatch 0x77ef4764: a(n) Dispatcher > 0xbfd5c810 I [] in BlockClosure>newProcess 0x78d2a634: a(n) BlockClosure > > Most recent primitives > > stack page bytes 4096 available headroom 3300 minimum unused headroom 3508 > > (last object overwritten) > Aborted > > > TimerScheduler/Dispatcher is code of mine but I wouldn't understand why this > code is getting blocked and would eventually cause a stack overflow? > > holger >

Oct. 7, 2013

Re: [Pharo-dev] external semaphoresâ¦again

by Henrik Johansen

On Oct 7, 2013, at 11:16 , Norbert Hartl <norbert(a)hartl.name> wrote: > As I need an image that runs longer than 24 hours I'm looking at some stuff and wonder. Can anybody explain me the rationale for a code like this > > maxExternalSemaphores: aSize > "This method should never be called as result of normal program > execution. If it is however, handle it differently: > - In development, signal an error to promt user to set a bigger size > at startup immediately. > - In production, accept the cost of potentially unhandled interrupts, > but log the action for later review. > > See comment in maxExternalObjectsSilently: why this behaviour is > desirable, " > "Can't find a place where development/production is decided. > Suggest Smalltalk image inProduction, but use an overridable temp > meanwhile. " > | inProduction | > self maxExternalSemaphores > ifNil: [^ 0]. > inProduction := false. > ^ inProduction > ifTrue: [self maxExternalSemaphoresSilently: aSize. > self crTrace: 'WARNING: Had to increase size of semaphore signal handling table due to many external objects concurrently in use'; > crTrace: 'You should increase this size at startup using #maxExternalObjectsSilently:'; > crTrace: 'Current table size: ' , self maxExternalSemaphores printString] > ifFalse: ["Smalltalk image" > self error: 'Not enough space for external objects, set a larger size at startup!' > "Smalltalk image"] > > I have reported this once but got no feedback so I like to have a few opinions. > > The report is here: https://pharo.fogbugz.com/f/cases/10839/ > > Norbert The rationale is that inProduction would be some global setting, not yet in place when the code was writtenâ¦ Excessive simultaneous Semaphore usage is something that should be caught during development, in which case it's better to get an active notification, than having it logged somewhere. If I've understood correctly, it's moot on newer Pharo VM's, where there's no limit on the semtable size, but for legacy code a startup item setting size using maxExternalObjectsSilently: (as suggested in the Warning text), is still a more proper fix than setting inProduction to true and crossing your fingers hoping no signals will be lost during table growth. Cheers. Henrik

Oct. 7, 2013

Re: [Pharo-dev] Please try to reproduce this bug

by Norbert Hartl

I have a lot of issues that this or like this. I get the sources problem as well as images I work for some time tell me that my working directory doesn't exist. I need to save and reopen the image to be able to do anything. Norbert Am 07.10.2013 um 10:12 schrieb Damien Cassou <damien.cassou(a)gmail.com>: > Can you please try to reproduce this bug: > > 1- Open a fresh Pharo 2.0 > 2- Open Monticello > 3- Select the package AST-Core > 4- Click browse > > you get a "PharoV20.sources does not exist." error message. You can't escape. > > Please write your findings in > https://pharo.fogbugz.com/f/cases/11803/PharoV20-sources-does-not-exist-whe…. > > For me, it looks like the problem is in MCPackage>>snapshot > > -- > Damien Cassou > http://damiencassou.seasidehosting.st > > "Success is the ability to go from one failure to another without > losing enthusiasm." > Winston Churchill >

Oct. 7, 2013

100% CPU usage on image resume

by Holger Hans Peter Freyther

Hi, I saved an image and today I wanted to resume it. Everything comes up, for a second I can interact with it (e.g. move a window) but then the pharo VM goes to 100% CPU usage. I let it run while writing this email and it eventually aborted: C stack backtrace: ./pharo[0x80a0c0c] ./pharo(error+0x19)[0x80a0f69] ./pharo[0x80785d3] ./pharo[0x80786e1] ./pharo[0x8078f10] ./pharo[0x8090d97] ./pharo(ceStackOverflow+0x6a)[0x80946ea] [0x76ef122e] ./pharo(interpret+0x7a6)[0x8094f36] ./pharo(main+0x2b3)[0x80a18b3] /lib/i386-linux-gnu/i686/cmov/libc.so.6(__libc_start_main+0xf5)[0xb74bb8c5] Smalltalk stack dump: 0xbfd5c610 M SmallInteger(Number)>negative 0x3841=7200 0xbfd5c630 M SmallInteger(Integer)>* 0x3841=7200 0xbfd5c64c M SmallInteger>* 0x3841=7200 0xbfd5c668 M Duration>asNanoSeconds 0x790ddef4: a(n) Duration 0xbfd5c684 M Duration>= 0x7fe004a0: a(n) Duration 0xbfd5c6a0 M DateAndTime>asLocal 0x7fe00488: a(n) DateAndTime 0xbfd5c6c0 M DateAndTime>- 0x7fe0040c: a(n) DateAndTime 0xbfd5c6dc M DateAndTime>asSeconds 0x7fe0040c: a(n) DateAndTime 0xbfd5c700 M DateAndTime>< 0x78d2a5f8: a(n) DateAndTime 0xbfd5c71c M DateAndTime(Magnitude)>> 0x7fe0040c: a(n) DateAndTime 0xbfd5c73c M TimerScheduler>fireTimers: 0x77ef49ac: a(n) TimerScheduler 0xbfd5c764 I [] in TimerScheduler>runTimers 0x77ef49ac: a(n) TimerScheduler 0xbfd5c780 M BlockClosure>on:do: 0x78d2a60c: a(n) BlockClosure 0xbfd5c7a4 M [] in Dispatcher>dispatch 0x77ef4764: a(n) Dispatcher 0xbfd5c7c4 M BlockClosure>ensure: 0x78d2a798: a(n) BlockClosure 0xbfd5c7f0 I [] in Dispatcher>dispatch 0x77ef4764: a(n) Dispatcher 0xbfd5c810 I [] in BlockClosure>newProcess 0x78d2a634: a(n) BlockClosure Most recent primitives stack page bytes 4096 available headroom 3300 minimum unused headroom 3508 (last object overwritten) Aborted TimerScheduler/Dispatcher is code of mine but I wouldn't understand why this code is getting blocked and would eventually cause a stack overflow? holger

Oct. 7, 2013

external semaphoresâ¦again

by Norbert Hartl

As I need an image that runs longer than 24 hours I'm looking at some stuff and wonder. Can anybody explain me the rationale for a code like this maxExternalSemaphores: aSize "This method should never be called as result of normal program execution. If it is however, handle it differently: - In development, signal an error to promt user to set a bigger size at startup immediately. - In production, accept the cost of potentially unhandled interrupts, but log the action for later review. See comment in maxExternalObjectsSilently: why this behaviour is desirable, " "Can't find a place where development/production is decided. Suggest Smalltalk image inProduction, but use an overridable temp meanwhile. " | inProduction | self maxExternalSemaphores ifNil: [^ 0]. inProduction := false. ^ inProduction ifTrue: [self maxExternalSemaphoresSilently: aSize. self crTrace: 'WARNING: Had to increase size of semaphore signal handling table due to many external objects concurrently in use'; crTrace: 'You should increase this size at startup using #maxExternalObjectsSilently:'; crTrace: 'Current table size: ' , self maxExternalSemaphores printString] ifFalse: ["Smalltalk image" self error: 'Not enough space for external objects, set a larger size at startup!' "Smalltalk image"] I have reported this once but got no feedback so I like to have a few opinions. The report is here: https://pharo.fogbugz.com/f/cases/10839/ Norbert

Oct. 7, 2013

Tracking smalltalk database driver implementations

by Guillermo Polito

Hi guys! I'm tracking the smalltalk implementations of relational database drivers that are floating around :). So far I have: * http://smalltalkhub.com/mc/PharoExtras/PostgresV2/main* Open source - last commit 2013 - This is the only one we use in DBXTalk with glorp. *http://www.squeaksource.com/MySQL.html* Does not specify licence - last commit around 2012 *http://www.squeaksource.com/PostgresV3.html* MIT licence - last commit around 2010 *http://www.squeaksource.com/ODBC.html* MIT licence - last commit around 2011 http://www.squeaksource.com/SQLite.html * http://ss3.gemstone.com/ss/SQLite.html/Overview?_s=kR2mf_G3WE3A6p9g&_k=I7aw… * MIT licence Now, except for postgresv2, all the others are in squeaksource. I'd like to move them to smalltalkhub under some "RelationalDatabase" group, maybe DBXTalk. But at the same time I imagine they are not pharo specific, ideas?

Oct. 7, 2013