March 2018 - Pharo-dev - lists.pharo.org

Re: [Pharo-dev] [Vm-dev] Image crashing on startup, apparently during GC

by Eliot Miranda

Hi Cyril, On Sat, Mar 24, 2018 at 1:50 AM, Cyril Ferlicot D. <cyril.ferlicot(a)gmail.com > wrote: > Le 23/03/2018 Ã 21:52, Eliot Miranda a Ã©crit : > > Hi Damien, > > > > Indeed the image is corrupt at start-up. See below. > > > > > > Right. This VM is prior to the bug fixes in VMMaker.oscog-eem.2320: > > > > Spur: > > Fix a bad bug in SpurPlnningCompactor. > > unmarkObjectsFromFirstFreeObject, used when the compactor requires more > > than one pass due to insufficient savedFirstFieldsSpace, expects the > > corpse of a moved object to be unmarked, but > > copyAndUnmarkObject:to:bytes:firstField: only unmarked the target. > > Unmarking the corpse before the copy unmarks both. This fixes a crash > > with ReleaseBuilder class>>saveAsNewRelease when non-use of cacheDuring: > > creates lots of files, enough to push the system into the multi-pass > regime. > > > > > > Pharo urgently needs to upgrade the VM to one more up to date than 2017 > > 08 27 (in fact more up-to-date than opensmalltalk/vm commit > > 0fe1e1ea108e53501a0e728736048062c83a66ce, Fri Jan 19 13:17:57 2018 > > -0800). The bug that VMMaker.oscog-eem.2320 fixes can result in image > > corruption in large images, and can occur (as it has here) at start-up, > > causing one's work to be irretrievably lost. > > > > Hi Eliot, > > I think that there is a lot of people who would like to get a newer > stable vm for Pharo 6.1 and 7. The problem is that it is hard to know > which VM are stable enough to be promoted as stable. > > Some weeks ago Esteban tried to promote a VM as stable and he had to > revert it the same day because a regression occurred in the VM. > > If you're able to tell us which vms are stable in those present at > http://files.pharo.org/vm/pharo-spur32/ and > http://files.pharo.org/vm/pharo-spur64/ it would be a great help. > > Even better would be for the pharo community to have a way to know which > vms are stable or not without having to ask you. > The way to do that is to run tests on the CI infrastructure and mark VMs that pass the tests as stable. That;s what happens on travis with the Cog VMs built there. > > Have a nice day. > > > > > -- > > _,,,^..^,,,_ > > best, Eliot > -- > Cyril Ferlicot > https://ferlicot.fr > > -- _,,,^..^,,,_ best, Eliot

March 25, 2018

Re: [Pharo-dev] Slot Questions

by Sean P. DeNigris

webwarrior wrote > That's because you need to redefine #= and #definitionString > (alternatively: > redefine #printOn:) in your slot class. Ah, okay. After this existing part of Slot's class comment: "For customizing a subclass can override the meta-object-protocol methods. See subclasses for examples." I submitted a PR with class/method comments clarifying that: https://github.com/pharo-project/pharo/pull/1150 webwarrior wrote > Also regarding tool support: there's bug tracker entry Great, thanks. ----- Cheers, Sean -- Sent from: http://forum.world.st/Pharo-Smalltalk-Developers-f1294837.html

March 25, 2018

Re: [Pharo-dev] [pharo-project/pharo] 21646 AbstractFileReference should use "utilities" protocol name (#1149)

by Stephane Ducasse

Thanks Guille for raising these points. I already raised the issues of producing many small actions that are damaging our infrastructure. I talked to christophe and he told me that it should not impact us but it does :). I changed the in-image categorizer to specify some given protocols (for example initialization and not initialize-release) and I mentioned several times that I did a small tools to analyze the use of a given selector and perform automatic protocol renaming in AutomatedMethodCategorizer on SmalltalkHub. May be I should use it massively I guess that we could clean a lot in one shot. Stef On Sun, Mar 25, 2018 at 9:06 PM, Guillermo Polito <guillermopolito(a)gmail.com > wrote: > Hi Torsten, > > I like all this cleanups and recategorizations. > > But I have two requests, tell me what you think: > > - first and more important. I'd like that we define a convention document > in a wiki page or pillar document for protocols. What are the correct > protocols to use and why? This is important for newcomers and for making > tools. I see that you're doing a lot of recategorizations but I'd like if > we set them up and discuss them in the mailing list. > > - second and more technological. Why are all categorizations splitted in > mini pull requests, scoped to one class? Can we group them at least by > package? otherwise this is killing our integration infrastructure (which is > modest but for free :)). Also separating all this changes means that we > cannot easily find them in the history later on, I'd prefer that we group > related changes in related commits if possible. Marcus was telling me this > may be because people have trouble reviewing them if they are too big? I > can review if you put me ask me as reviewer. > > Tx, > Guilel > > On Sun, Mar 25, 2018 at 8:54 PM, Astares <notifications(a)github.com> wrote: > >> https://pharo.fogbugz.com/f/cases/21646/AbstractFileReferenc >> e-should-use-utilities-protocol-name >> ------------------------------ >> You can view, comment on, or merge this pull request online at: >> >> https://github.com/pharo-project/pharo/pull/1149 >> Commit Summary >> >> - 21646 AbstractFileReference should use "utilities" protocol name >> >> File Changes >> >> - *M* src/FileSystem-Core/AbstractFileReference.class.st >> <https://github.com/pharo-project/pharo/pull/1149/files#diff-0> (2) >> >> Patch Links: >> >> - https://github.com/pharo-project/pharo/pull/1149.patch >> - https://github.com/pharo-project/pharo/pull/1149.diff >> >> â >> You are receiving this because you are subscribed to this thread. >> Reply to this email directly, view it on GitHub >> <https://github.com/pharo-project/pharo/pull/1149>, or mute the thread >> <https://github.com/notifications/unsubscribe-auth/AArO4g3P9uj62nQGbRG-T842H…> >> . >> > > > > -- > > > > Guille Polito > > Research Engineer > > Centre de Recherche en Informatique, Signal et Automatique de Lille > > CRIStAL - UMR 9189 > > French National Center for Scientific Research - *http://www.cnrs.fr > <http://www.cnrs.fr>* > > > *Web:* *http://guillep.github.io* <http://guillep.github.io> > > *Phone: *+33 06 52 70 66 13 <+33%206%2052%2070%2066%2013> >

March 25, 2018

Re: [Pharo-dev] [pharo-project/pharo] 21646 AbstractFileReference should use "utilities" protocol name (#1149)

by Guillermo Polito

Hi Torsten, I like all this cleanups and recategorizations. But I have two requests, tell me what you think: - first and more important. I'd like that we define a convention document in a wiki page or pillar document for protocols. What are the correct protocols to use and why? This is important for newcomers and for making tools. I see that you're doing a lot of recategorizations but I'd like if we set them up and discuss them in the mailing list. - second and more technological. Why are all categorizations splitted in mini pull requests, scoped to one class? Can we group them at least by package? otherwise this is killing our integration infrastructure (which is modest but for free :)). Also separating all this changes means that we cannot easily find them in the history later on, I'd prefer that we group related changes in related commits if possible. Marcus was telling me this may be because people have trouble reviewing them if they are too big? I can review if you put me ask me as reviewer. Tx, Guilel On Sun, Mar 25, 2018 at 8:54 PM, Astares <notifications(a)github.com> wrote: > https://pharo.fogbugz.com/f/cases/21646/AbstractFileReference-should- > use-utilities-protocol-name > ------------------------------ > You can view, comment on, or merge this pull request online at: > > https://github.com/pharo-project/pharo/pull/1149 > Commit Summary > > - 21646 AbstractFileReference should use "utilities" protocol name > > File Changes > > - *M* src/FileSystem-Core/AbstractFileReference.class.st > <https://github.com/pharo-project/pharo/pull/1149/files#diff-0> (2) > > Patch Links: > > - https://github.com/pharo-project/pharo/pull/1149.patch > - https://github.com/pharo-project/pharo/pull/1149.diff > > â > You are receiving this because you are subscribed to this thread. > Reply to this email directly, view it on GitHub > <https://github.com/pharo-project/pharo/pull/1149>, or mute the thread > <https://github.com/notifications/unsubscribe-auth/AArO4g3P9uj62nQGbRG-T842H…> > . > -- Guille Polito Research Engineer Centre de Recherche en Informatique, Signal et Automatique de Lille CRIStAL - UMR 9189 French National Center for Scientific Research - *http://www.cnrs.fr <http://www.cnrs.fr>* *Web:* *http://guillep.github.io* <http://guillep.github.io> *Phone: *+33 06 52 70 66 13

March 25, 2018

Re: [Pharo-dev] Slot Questions

by webwarrior

Sean P. DeNigris wrote > Sean P. DeNigris wrote >> Object subclass: #MyClass >> slots: { #var1 => MySlot adjustment: 10 } >> classVariables: { } >> category: '' >> â¦ >> but when I open a new browser on the class definition of MyClass, the >> slots >> are truncated to: >> slots: { #var1 => MySlot } > > It turns out that the slot is still "working" e.g. #adjustment is still > having its effect e.g. > read: anObject > ^ (super read: anObject) ifNotNil: [ :val | val + self adjustment ] > It's just that the class template has lost information. > > Another issue: > Changing the slots in the template from "{ #var1 => MySlot adjustment: 10 > }" > to "{ #var1 => MySlot adjustment: 100 }" has no effect unless I accept an > intermediate "{ #var1 }" > > > > ----- > Cheers, > Sean > -- > Sent from: http://forum.world.st/Pharo-Smalltalk-Developers-f1294837.html That's because you need to redefine #= and #definitionString (alternatively: redefine #printOn:) in your slot class. Also regarding tool support: there's bug tracker entry https://pharo.fogbugz.com/f/cases/19541/References-to-class-should-also-be-… <https://pharo.fogbugz.com/f/cases/19541/References-to-class-should-also-be-…> -- Sent from: http://forum.world.st/Pharo-Smalltalk-Developers-f1294837.html

March 25, 2018

Re: [Pharo-dev] Some segfault crashes and freezes when trying to re-run saved pharo 7 images

by Holger Freyther

> On 24. Mar 2018, at 19:22, Eliot Miranda <eliot.miranda(a)gmail.com> wrote: > Dear Eliot, >> d.) Re-write FreeType with Alien and just use the Plugin to conveniently link/load to freetype.. > > I wonder if there is sense in trying to come up with a general memory handle object that includes a session identifier, so that attempts to free stake memory always fail. I think it would help with the FT2Handle (to invalidate it on a new session) but it will not solve other image resume/FFI issues. Let's assume we have multiple calls into a C library (get a handle, call 1st method, call 2nd method, have a loop with backwards jumps). Something like: | memory | memory := GetSomeMemorySomewhere. 1 to: 5 do: [ self doSomeCStuffWithMemory: memory pointerAndCheckStillValid. ... more C stuff ]. We could be interrupted at any point and when the execution is resumed the handles might be invalid. In Python terminology something like a context manager could help: # Image saving is delayed/inhibited until after the callout chain CCallOut with: [ 1 to: 5 do: [ self doSomeCStuffWithMemory: memory pointerAndCheckStillValid. ... more C stuff ]. ] What do you think? holger

March 25, 2018

Re: [Pharo-dev] Some segfault crashes and freezes when trying to re-run saved pharo 7 images

by Holger Freyther

> On 24. Mar 2018, at 17:03, Holger Freyther <holger(a)freyther.de> wrote: > > > > FreeTypeExternalMemory>validate just checks if the handle isValid but remember 1st from my previous mail. We have not _yet_ cleared the FT2Handle SubInstances.. So the memory is all good. The snippet below is my work-around. I think the proper fix is to get FT2Handle to invalidate handles a lot earlier (Gui->System startup?). Author useAuthor: 'Hacks' during: [ FreeTypeExternalMemory addInstVarNamed: 'session'. FreeTypeExternalMemory compile: 'validate (session == Smalltalk session and: [ self isValid ]) ifTrue: [ ^self ]. bytes ifNil: [ ^ self ]. session := Smalltalk session. self primCopyToExternalMemory: bytes. self isValid ifTrue: [ self class register: self ]' ]. Smalltalk snapshot: true andQuit: true.

March 25, 2018

Re: [Pharo-dev] Slot Questions

by Sean P. DeNigris

Sean P. DeNigris wrote > Object subclass: #MyClass > slots: { #var1 => MySlot adjustment: 10 } > classVariables: { } > category: '' > â¦ > but when I open a new browser on the class definition of MyClass, the > slots > are truncated to: > slots: { #var1 => MySlot } It turns out that the slot is still "working" e.g. #adjustment is still having its effect e.g. read: anObject ^ (super read: anObject) ifNotNil: [ :val | val + self adjustment ] It's just that the class template has lost information. Another issue: Changing the slots in the template from "{ #var1 => MySlot adjustment: 10 }" to "{ #var1 => MySlot adjustment: 100 }" has no effect unless I accept an intermediate "{ #var1 }" ----- Cheers, Sean -- Sent from: http://forum.world.st/Pharo-Smalltalk-Developers-f1294837.html

March 24, 2018

Re: [Pharo-dev] pharo64+load code=broken image

by Javier PimÃ¡s

On Fri, Mar 23, 2018 at 11:17 PM, Eliot Miranda <eliot.miranda(a)gmail.com> wrote: > > > On Fri, Mar 23, 2018 at 7:54 AM, Javier PimÃ¡s <elpochodelagente(a)gmail.com> > wrote: > >> Hi Eliot, the crash.dmp shows what I put in the first mail, gdb shows >> some more info: >> >> #0 __GI___libc_free (mem=0x7f15fc1d15d0) at malloc.c:2965 >> #1 0x00000000004c745b in primitiveFFIFree () at >> /home/travis/build/OpenSmalltalk/opensmalltalk-vm/src/ >> plugins/SqueakFFIPrims/X64SysVFFIPlugin.c:6352 >> #2 0x0000000000457f64 in primitiveExternalCall () at >> /home/travis/build/OpenSmalltalk/opensmalltalk-vm/spur64src/ >> vm/gcc3x-cointerp.c:75751 >> #3 0x00000000004595cb in interpretMethodFromMachineCode () at >> /home/travis/build/OpenSmalltalk/opensmalltalk-vm/spur64src/ >> vm/gcc3x-cointerp.c:18524 >> #4 0x000000000045bc69 in ceSendsupertonumArgs (selector=<optimized out>, >> superNormalBar=0, rcvr=<optimized out>, numArgs=0) >> at /home/travis/build/OpenSmalltalk/opensmalltalk-vm/spur64src/ >> vm/gcc3x-cointerp.c:16602 >> #5 0x00000000009000bb in ?? () >> #6 0x00007fffffffdbc0 in ?? () >> #7 0x000000000045ce56 in interpret () at /home/travis/build/OpenSmallta >> lk/opensmalltalk-vm/spur64src/vm/gcc3x-cointerp.c:2731 >> #8 0x00000000009ccf7f in ?? () >> #9 0x00000000045e39c8 in ?? () >> #10 0x00000000045e39c8 in ?? () >> ... >> >> As it is just loading, I think that it is trying to free some memory that >> was allocated before the image was saved. Looks like there is an external >> address that should have been niled but for some reason wasn't. >> > > I can take a look at the image in the simulator and/or debugger and see if > any stale pointers exist. Put the image and changes somewhere where I can > download them. > > Thanks! I'll try to do something better instead: do it by myself asking questions if needed, it might take some time before I start though. Cheers, Pocho >> On Fri, Mar 23, 2018 at 11:43 AM, Eliot Miranda <eliot.miranda(a)gmail.com> >> wrote: >> >>> Hi Javier, >>> >>> what does the crash.dmp file say? Or what does a gdb stacktrace say? >>> >>> _,,,^..^,,,_ (phone) >>> >>> On Mar 23, 2018, at 7:18 AM, Javier PimÃ¡s <elpochodelagente(a)gmail.com> >>> wrote: >>> >>> no luck with this one :( >>> >>> On Fri, Mar 23, 2018 at 4:48 AM, Alistair Grant <akgrant0710(a)gmail.com> >>> wrote: >>> >>>> Hi Javier, >>>> >>>> On 23 March 2018 at 04:54, Javier PimÃ¡s <elpochodelagente(a)gmail.com> >>>> wrote: >>>> > Hi folks, I take a fresh Pharo6.1 (spur64), load a git repository and >>>> save >>>> > the image. When reopening I get a segmentation fault. OS is ubuntu >>>> 17.04/64 >>>> > bits. I just cannot make it work, tried with all the vms I could >>>> (vmI61, >>>> > vmT61, vmLatest70, vmTLatest70). I suspect it is related to using git >>>> but >>>> > I'm not even sure. >>>> > >>>> > Below I put the involved code, crash dump at the end (seems to be the >>>> same >>>> > with any vm). btw, the vmILatest70 script does not seem to work (curl >>>> > https://get.pharo.org/64/vmILatest70 | bash). >>>> > >>>> > Any ideas? >>>> >>>> I'm using the following VM on Ubuntu 16.04 without any issues: >>>> >>>> http://files.pharo.org/vm/pharo-spur64/linux/pharo-linux-x86 >>>> _64threaded-201803160215-43a2f5c.zip >>>> >>>> >>>> HTH, >>>> Alistair >>>> >>>> >>> >>> >>> -- >>> Javier PimÃ¡s >>> Ciudad de Buenos Aires >>> >>> >> >> >> -- >> Javier PimÃ¡s >> Ciudad de Buenos Aires >> > > > > -- > _,,,^..^,,,_ > best, Eliot > -- Javier PimÃ¡s Ciudad de Buenos Aires

March 24, 2018

Re: [Pharo-dev] Slot Questions

by Sean P. DeNigris

Question #3: Shouldn't the following (from the paper): slots: { #staff => ManySlot opposite: #boss class: Clerk. } Be: slots: { #staff => (ManySlot opposite: #boss class: Clerk). } Also, what is that code actually doing? Is #opposite:class: a class-side method of ManySlot? If so, what does it return? Ah, I see now - #=>> does some magic to accept either a Slot class or instance. Question #4: Given: Object subclass: #MyClass slots: { #var1 => MySlot } classVariables: { } category: '' and MyClass >>#var1: anObject var1 := anObject MySlot>#write:to: doesn't seem to be sent when I `myClassInstance var1: 2`. I tried to insert a halt, 1 inspect, etc. I am able to intercept #read: on the other hand Question #5: I tried: Object subclass: #MyClass slots: { #var1 => MySlot adjustment: 10 } "with and without () around `MySlot adjustment: 10`" classVariables: { } category: '' where: InstanceVariableSlot subclass: #MySlot slots: { #adjustment } and: adjustment ^ adjustment ifNil: [ 0 ] but when I open a new browser on the class definition of MyClass, the slots are truncated to: slots: { #var1 => MySlot } ----- Cheers, Sean -- Sent from: http://forum.world.st/Pharo-Smalltalk-Developers-f1294837.html

March 24, 2018