- Pharo-dev - lists.pharo.org

Re: [Pharo-dev] Memory leaks

by stepharong

Thanks John For the record we are investigating a leak on points. In pharo 60 we discovered that in certain images I could have 300 000 or more points. :) Now marcus was analysing the problems with pavel and we had the impression that some of the points may have been tenured too fast. Pavel shows that every 5 ms we get a new point. Stef On Tue, 31 Jan 2017 05:00:47 +0100, John Brant <brant(a)refactoryworkers.com> wrote: > I donât know if there is an existing object reference crawler, but I > ported mine from VW. It finds the shortest paths so it is good for > finding the reference that is causing the memory leak. > MCHttpRepository > location: > 'http://www.smalltalkhub.com/mc/JohnBrant/ReferenceFinder/main' > user: '' > password: ââ > Once loaded, you can evaluate something like "ReferenceFinder > findPathToInstanceOf: MyClassâ. That will return a reference path or > nil if no path exists from the Smalltalk global to an instance of > MyClass. If you want to find a path to a particular instance, you can > use âReferenceFinder findPathTo: myInstâ. -- Using Opera's mail client: http://www.opera.com/mail/

Jan. 31, 2017

[pharo-project/pharo-core] 605aea: 60366

by GitHub

Branch: refs/heads/6.0 Home: https://github.com/pharo-project/pharo-core Commit: 605aea3271c974e38784e70c39e4ae1009be9c4d https://github.com/pharo-project/pharo-core/commit/605aea3271c974e38784e70c… Author: Jenkins Build Server <board(a)pharo-project.org> Date: 2017-01-31 (Tue, 31 Jan 2017) Changed paths: M Collections-Tests.package/ArrayTest.class/instance/testing/testComplexIsSelfEvaluating.st M Collections-Tests.package/DictionaryTest.class/instance/requirements/otherDictionaryClasses.st A Collections-Tests.package/ManifestCollectionsTests.class/class/meta-data - dependency analyser/manuallyResolvedDependencies.st R Collections-Tests.package/SplitJoinTest.class/instance/running/testSplitStringOnRegex.st A Compression-Tests.package/ZipExtensionTests.class/README.md A Compression-Tests.package/ZipExtensionTests.class/definition.st A Compression-Tests.package/ZipExtensionTests.class/instance/tests/testZipped.st M ConfigurationOfDependencyAnalyser.package/ConfigurationOfDependencyAnalyser.class/instance/symbolic versions/stable_.st A ConfigurationOfDependencyAnalyser.package/ConfigurationOfDependencyAnalyser.class/instance/versions/version121_.st M Graphics-Display Objects.package/DisplayScreen.class/class/initialize-release/initialize.st M Graphics-Display Objects.package/DisplayScreen.class/instance/private/findAnyDisplayDepthIfNone_.st M Kernel-Tests.package/MethodContextTest.class/instance/tests/verifyJumpWithSelector_.st A Kernel.package/ModificationForbidden.class/README.md A Kernel.package/ModificationForbidden.class/class/instance creation/for_atIndex_with_.st A Kernel.package/ModificationForbidden.class/class/instance creation/for_atInstVar_with_.st A Kernel.package/ModificationForbidden.class/definition.st A Kernel.package/ModificationForbidden.class/instance/accessing/fieldIndex.st A Kernel.package/ModificationForbidden.class/instance/accessing/fieldIndex_.st A Kernel.package/ModificationForbidden.class/instance/accessing/newValue.st A Kernel.package/ModificationForbidden.class/instance/accessing/newValue_.st A Kernel.package/ModificationForbidden.class/instance/accessing/object.st A Kernel.package/ModificationForbidden.class/instance/accessing/object_.st A Kernel.package/ModificationForbidden.class/instance/accessing/retrySelector.st A Kernel.package/ModificationForbidden.class/instance/accessing/retrySelector_.st A Kernel.package/ModificationForbidden.class/instance/printing/messageText.st A Kernel.package/ModificationForbidden.class/instance/printing/printObject_on_.st A Kernel.package/ModificationForbidden.class/instance/printing/standardMessageText.st A Kernel.package/ModificationForbidden.class/instance/retrying/retryModification.st A Kernel.package/ModificationForbidden.class/instance/testing/isResumable.st R Kernel.package/NoModificationError.class/README.md R Kernel.package/NoModificationError.class/definition.st R Kernel.package/NoModificationError.class/instance/accessing/index_.st R Kernel.package/NoModificationError.class/instance/accessing/object_.st R Kernel.package/NoModificationError.class/instance/accessing/selector_.st R Kernel.package/NoModificationError.class/instance/accessing/value_.st R Kernel.package/NoModificationError.class/instance/printing/messageText.st R Kernel.package/NoModificationError.class/instance/printing/printObject_on_.st R Kernel.package/NoModificationError.class/instance/printing/standardMessageText.st R Kernel.package/NoModificationError.class/instance/retrying/retryModification.st R Kernel.package/NoModificationError.class/instance/testing/isResumable.st M Kernel.package/Object.class/instance/write barrier/attemptToAssign_withIndex_.st A Moose-Algos-Graph.package/MalGraphEdge.class/instance/accessing/printOn_.st M Moose-Algos-Graph.package/MalGraphReducer.class/instance/merging/mergeNodesInCircuit_.st M Moose-Algos-Graph.package/MalLongestPath.class/definition.st M Moose-Algos-Graph.package/MalLongestPath.class/instance/computing/computeStep.st M Moose-Algos-Graph.package/MalLongestPath.class/instance/running/run.st M Moose-Algos-Graph.package/MalLongestPath.class/instance/testing/isCompleted.st A Regex-Tests-Core.package/RxExtensionsTest.class/README.md A Regex-Tests-Core.package/RxExtensionsTest.class/definition.st A Regex-Tests-Core.package/RxExtensionsTest.class/instance/tests/testSplitJoinBoundaryCases.st A Regex-Tests-Core.package/RxExtensionsTest.class/instance/tests/testSplitStringOnRegex.st M Rubric.package/RubEditingMode.class/instance/accessing/interactive.st R ScriptLoader60.package/ScriptLoader.class/instance/pharo - scripts/script60365.st A ScriptLoader60.package/ScriptLoader.class/instance/pharo - scripts/script60366.st R ScriptLoader60.package/ScriptLoader.class/instance/pharo - updates/update60365.st A ScriptLoader60.package/ScriptLoader.class/instance/pharo - updates/update60366.st M ScriptLoader60.package/ScriptLoader.class/instance/public/commentForCurrentUpdate.st M Tool-DependencyAnalyser-Test.package/DAPackageTest.class/instance/tests/testDependenciesFrom.st M Tool-DependencyAnalyser-Test.package/DAPackageTest.class/instance/tests/testReferenceDependenciesFrom.st M Tool-DependencyAnalyser-UI.package/DAPackageAnalyzerTreeModel.class/instance/initialization/initializeButtons.st M Tool-DependencyAnalyser-UI.package/DAPackageDependenciesWelcome.class/class/world menu/menuCommandOn_.st M Tool-DependencyAnalyser-UI.package/DATraitDependencyNode.class/instance/accessing/icon.st M WriteBarrierTests.package/WriteBarrierTest.class/instance/setup/expectedFailures.st M WriteBarrierTests.package/WriteBarrierTest.class/instance/tests - object/testMutateIVObject.st A WriteBarrierTests.package/WriteBarrierTest.class/instance/tests - object/testRetryingInstVarModification.st Log Message: ----------- 60366 17352 DisplayScreen should have a class initialization method https://pharo.fogbugz.com/f/cases/17352 19544 Better method comment for RubEditingMode>>interactive https://pharo.fogbugz.com/f/cases/19544 19614 Better names for NoModificationError and friends https://pharo.fogbugz.com/f/cases/19614 18767 Remove external dependencies from Collection-Tests https://pharo.fogbugz.com/f/cases/18767 19325 Warning: dependency declared not detected as a dependency https://pharo.fogbugz.com/f/cases/19325 http://files.pharo.org/image/60/60366.zip

Jan. 31, 2017

[pharo-project/pharo-core]

by GitHub

Branch: refs/tags/60366 Home: https://github.com/pharo-project/pharo-core

Jan. 31, 2017

Re: [Pharo-dev] Two listener sockets on same port

by Denis Kudriashov

2017-01-31 11:56 GMT+01:00 Henrik Johansen <henrik.s.johansen(a)veloxit.no>: > IIRC, behaviour is different on all three platforms: > - Linux raises error > I got same behaviour on Debian as on Mac. It assigns different port

Jan. 31, 2017

Re: [Pharo-dev] Two listener sockets on same port

by Sven Van Caekenberghe

> On 31 Jan 2017, at 11:56, Henrik Johansen <henrik.s.johansen(a)veloxit.no> wrote: > > >> On 30 Jan 2017, at 21:13 , Denis Kudriashov <dionisiydk(a)gmail.com> wrote: >> >> >> 2017-01-30 20:25 GMT+01:00 Sven Van Caekenberghe <sven(a)stfx.eu>: >> Well, you should do your utmost best to prevent the situation in the first place (use startup/shutdown behaviour correctly). >> >> I remember I had those problems a lot in the beginning of Zn, especially when running tests that sometimes failed. >> >> Have you seen Zn's #isListening ? I think that was written to catch the case you described, but I am not sure it is 100% rock solid. >> >> Yes. This is exactly how I fix my case. >> But this post is about general problem which I think is very critical. >> >> In #isListening code I see extra condition "self serverSocket localPort = self port". Could you explain what case it is covered? >> > > Doesn't work on Windows; > https://pharo.fogbugz.com/f/cases/16939/Intermittent-monkey-failures-due-to… > > IIRC, behaviour is different on all three platforms: > - Linux raises error > - Mac assigns different port than requested > - Win assigns requested port, but receives no traffic. This is the curse of being x-platform, we don't see them often (luckily) but there are important (semantic) differences that sometimes make life (very) difficult. > How to detect the last one from image is a very good question (tm) indeed... > > Cheers, > Henry

Jan. 31, 2017

Re: [Pharo-dev] Two listener sockets on same port

by Henrik Johansen

> On 30 Jan 2017, at 21:13 , Denis Kudriashov <dionisiydk(a)gmail.com> wrote: > > > 2017-01-30 20:25 GMT+01:00 Sven Van Caekenberghe <sven(a)stfx.eu <mailto:sven@stfx.eu>>: > Well, you should do your utmost best to prevent the situation in the first place (use startup/shutdown behaviour correctly). > > I remember I had those problems a lot in the beginning of Zn, especially when running tests that sometimes failed. > > Have you seen Zn's #isListening ? I think that was written to catch the case you described, but I am not sure it is 100% rock solid. > > Yes. This is exactly how I fix my case. > But this post is about general problem which I think is very critical. > > In #isListening code I see extra condition "self serverSocket localPort = self port". Could you explain what case it is covered? > Doesn't work on Windows; https://pharo.fogbugz.com/f/cases/16939/Intermittent-monkey-failures-due-to… <https://pharo.fogbugz.com/f/cases/16939/Intermittent-monkey-failures-due-to…> IIRC, behaviour is different on all three platforms: - Linux raises error - Mac assigns different port than requested - Win assigns requested port, but receives no traffic. How to detect the last one from image is a very good question (tm) indeed... Cheers, Henry

Jan. 31, 2017

Re: [Pharo-dev] Form/Spec based UI's in GT inspector extensions

by Tudor Girba

Hi, Indeed, this is a nice observation. We currently have basic support for handling forms: - GLMSpecPresentation offers a basic integration of Spec widgets inside Glamour. This ships with the Pharo image. - GLMMagrittePresentation allows us to take a Magritte-described object and render it through a form. But, it would definitely be worthwhile to have more elaborate possibilities. Cheers, Doru > On Jan 31, 2017, at 10:16 AM, Torsten Bergmann <astares(a)gmx.de> wrote: > > Hi, > > there are a lot of examples for GT inspector extensions in the image. I've seen > #tree(s), #list(s) or #table like UI's are used. Even a #text or an "image" displaying > a morph (using #form keyword sent to the given composite). > > It is very easy to provide extension methods to adopt the inspector to own needs and I > wish people would use this more in our environment and according projects. > > But what I still miss is an easy way to extend the inspector with REAL FORMS (which are dialog like) > to better display and manipulate attributes/ivars of an object. > > To give a few examples: > ======================= > When I inspect a color instance I would like to see (and change) the RGB > values as well as the HEX representation in a form in an own tab. Similar as > if I would be changing a color in a graphics program. > I would like to lively manipulate the RGB values easily with sliders instead > of sending messages to self in the "first" inspector code pane/tab. > > Or think of a Person class where you can provide an own custom inspector pane to > display/change the name and first name in an edit field or select the date of birth in > a datepicker together with a button to apply the new values to the object. > > Think of a database connection object where I can manipulate the ivar storing the > connection string in an edit field but also provide a "Reconnect" in the button > after I changed it. > > Currently we by default we have the "Raw" tab and need to manipulate the object state > with expressions in the code pane. For me this is not interactive enough and Pharo > has much potential to be a lively objects environment here. > > > Long story short: > ================= > 1. Is there any support for such kind of dialog based UI's in the > inspector tabs/composites. > > 2. Is there any support to use Spec based UIs to extend the GT inspector? > > 3. Is there support to build simple UI forms for the inspector with labels and > edit fields, combo boxes, radio buttons, date fields, field set, buttons, ... like in [2]. > > 4. If we dont have such a beast yet I guess we should work into this direction because > it would make Pharo not only a lively (for message sending) but very useful > environment for direct object interaction/manipulation of states. > > Are there any plans/work on this already? If not what would be needed to achieve that? > Maybe a GSOC project? > > Thanks > T. > > [1] http://101.wacom.com/UserHelp/en/Images/Dialog_OSDSetColor.png > [1] https://i.stack.imgur.com/X91VL.png > -- www.tudorgirba.com www.feenk.com "We cannot reach the flow of things unless we let go."

Jan. 31, 2017

Re: [Pharo-dev] Memory leaks

by Pavel Krivanek

Very nice tool. I think it cannot find objects pointed only by a stack of some processes so it may be handy to put them to the global space. Smalltalk globals at: #Stacks put: ((Process allInstances collect: [:p | (p suspendedContext) ifNotNil: #stack ]) reject: #isNil) asOrderedCollection. (Smalltalk globals at: #Stacks) add: thisContext stack. -- Pavel 2017-01-31 5:00 GMT+01:00 John Brant <brant(a)refactoryworkers.com>: > > On Jan 30, 2017, at 3:57 PM, Alexandre Bergel <alexandre.bergel(a)me.com> > wrote: > > > > > > The problem is still present. > > Looking at the pointers > > > > There was an object reference crawler no? I remember someone worked on > this? I tried sending #pointersTo but I without much success on identifying > the cause of this leak. > > > > Any idea? > > > > I donât know if there is an existing object reference crawler, but I > ported mine from VW. It finds the shortest paths so it is good for finding > the reference that is causing the memory leak. > > MCHttpRepository > location: 'http://www.smalltalkhub.com/ > mc/JohnBrant/ReferenceFinder/main' > user: '' > password: ââ > > Once loaded, you can evaluate something like "ReferenceFinder > findPathToInstanceOf: MyClassâ. That will return a reference path or nil > if no path exists from the Smalltalk global to an instance of MyClass. If > you want to find a path to a particular instance, you can use > âReferenceFinder findPathTo: myInstâ. > > In the past, Iâve found that NECController and some compiler > infrastructure (I forget the class) have been culprits for memory leaks > with their caches. > > > John Brant >

Jan. 31, 2017

[pharo-project/pharo-core]

by GitHub

Branch: refs/tags/60365 Home: https://github.com/pharo-project/pharo-core

Jan. 31, 2017

[pharo-project/pharo-core] a1d73b: 60365

by GitHub

Branch: refs/heads/6.0 Home: https://github.com/pharo-project/pharo-core Commit: a1d73b1704d27971ef636f12e05f999c6ef7eb78 https://github.com/pharo-project/pharo-core/commit/a1d73b1704d27971ef636f12… Author: Jenkins Build Server <board(a)pharo-project.org> Date: 2017-01-31 (Tue, 31 Jan 2017) Changed paths: M Deprecated60.package/extension/Collection/instance/ifEmpty_ifNotEmptyDo_.st M Kernel-Tests.package/ClassHierarchyTest.class/instance/tests/testSubclasses.st R ScriptLoader60.package/ScriptLoader.class/instance/pharo - scripts/script60364.st A ScriptLoader60.package/ScriptLoader.class/instance/pharo - scripts/script60365.st R ScriptLoader60.package/ScriptLoader.class/instance/pharo - updates/update60364.st A ScriptLoader60.package/ScriptLoader.class/instance/pharo - updates/update60365.st M ScriptLoader60.package/ScriptLoader.class/instance/public/commentForCurrentUpdate.st M Tool-Catalog.package/CatalogBrowser.class/class/utilities/iconFor_.st Log Message: ----------- 60365 19624 Wrong deprecated message in ifEmpty:ifNotEmptyDo: https://pharo.fogbugz.com/f/cases/19624 19625 Catalog: loaded config should not be displayed with white icon https://pharo.fogbugz.com/f/cases/19625 19129 Deprecation: The method Object>>name called from ClassHierarchyTest>>#testSubclasses has been deprecated. https://pharo.fogbugz.com/f/cases/19129 http://files.pharo.org/image/60/60365.zip

Jan. 31, 2017