March 2011 - Pharo-dev - lists.pharo.org

[Pharo-project] why RefactoringEnvironment requires AST-Core

by stephane ducasse

Hi lukas we would like to reuse the BrowserEnvironment for an experiment and I saw package: 'Refactoring-Environment' with: [ spec requires: 'AST-Core' ]; package: 'Refactoring-Tests-Environment' with: [ spec requires: 'Refactoring-Environment' ]; Do you know if this is correct? And why? Stef

March 17, 2011

Re: [Pharo-project] Pharo 1.2. red again

by Igor Stasenko

On 17 March 2011 15:02, Torsten Bergmann <astares(a)gmx.de> wrote: > Hudson Console says: > Â "The spelling dictionary cannot be downloaded from <http://www.lukas-r..." > > Looks like a temporarily network outage. > https://pharo-ic.lille.inria.fr/hudson/view/Pharo/job/Pharo%201.2/182/conso… > > Bye > T. > -- > Empfehlen Sie GMX DSL Ihren Freunden und Bekannten und wir > belohnen Sie mit bis zu 50,- Euro! https://freundschaftswerbung.gmx.de > > -- Best regards, Igor Stasenko AKA sig.

March 17, 2011

[Pharo-project] Pharo 1.2. red again

by Torsten Bergmann

Hudson Console says: "The spelling dictionary cannot be downloaded from <http://www.lukas-r..." https://pharo-ic.lille.inria.fr/hudson/view/Pharo/job/Pharo%201.2/182/conso… Bye T. -- Empfehlen Sie GMX DSL Ihren Freunden und Bekannten und wir belohnen Sie mit bis zu 50,- Euro! https://freundschaftswerbung.gmx.de

March 17, 2011

Re: [Pharo-project] Squeaksource.com down

by Miguel CobÃ¡

Thanks :) El jue, 17-03-2011 a las 10:32 +0100, Fabrizio Perin escribiÃ³: > up again > > 2011/3/17 Miguel CobÃ¡ <miguel.coba(a)gmail.com> > > > SqueakSource Down > > SqueakSource is currently down. We are sorry for the > inconvenience and > working on the problem. In the meantime you can still commit > to a > directory repository on your disk and later copy the versions > to > SqueakSource. > > At > > -- > Miguel CobÃ¡ > http://twitter.com/MiguelCobaMtz > http://miguel.leugim.com.mx > > > > > -- Miguel CobÃ¡ http://twitter.com/MiguelCobaMtz http://miguel.leugim.com.mx

March 17, 2011

Re: [Pharo-project] Squeaksource.com down

by Fabrizio Perin

up again 2011/3/17 Miguel CobÃ¡ <miguel.coba(a)gmail.com> > > > SqueakSource Down > > SqueakSource is currently down. We are sorry for the inconvenience and > working on the problem. In the meantime you can still commit to a > directory repository on your disk and later copy the versions to > SqueakSource. > > At > > -- > Miguel CobÃ¡ > http://twitter.com/MiguelCobaMtz > http://miguel.leugim.com.mx > > > > >

March 17, 2011

Re: [Pharo-project] [Vm-dev] Socket Plugin primitives

by StÃ©phane Ducasse

Jannik I do not know you but even today on my mac I turned off the IPV6 supports because it was strange (not talking about Smalltalk). Now what you can probably do is to check the C files of the plugins in Cog and latest Squeak. Stef On Mar 17, 2011, at 5:28 AM, jannik.laval wrote: > > On Mar 16, 2011, at 21:52 , Levente Uzonyi wrote: > >> On Wed, 16 Mar 2011, jannik.laval wrote: >> >>> >>> Hi all, >>> >>> I am analyzing SocketPluggin for Ocean development. And I have some remarks/questions. >>> >>> First, to be sure, can you confirm my vision: >>> - the vm 4.x is the old one. Is it maintained ? >>> - vm 5.7.x is the normal vm (without JIT) >>> - vm 5.8.x is the Cog VM with normal pluggin >>> - vm Cog is Cog VM with less primitives than vm 5.8 (modified pluggins). >>> >>> Based on that, we can see that Cog has only 38 primitives related to SocketPluggin where normal pluggin provides 61 primitives. >>> Could anyone explain to me why some primitives disappear ? >> >> Cog was forked from an old version of the VM sources that didn't have the IPv6 primitives, so those are missing. > > Ok, now I understand. > >> There may be other differences, so someone has to merge the code of the two (Cog's and the interpreter VM's) plugins. > > Which kind of difference ?You want to say in the source code ? > > Jannik > >> >> >> Levente >> >>> I do not yet analyze which primitives disappear but maybe someone could help me to understand this difference between the two pluggins. >>> >>> Thanks in advance for your help, >>> --- >>> Jannik Laval >>> >>> >

March 17, 2011

[Pharo-project] Squeaksource.com down

by Miguel CobÃ¡

SqueakSource Down SqueakSource is currently down. We are sorry for the inconvenience and working on the problem. In the meantime you can still commit to a directory repository on your disk and later copy the versions to SqueakSource. At -- Miguel CobÃ¡ http://twitter.com/MiguelCobaMtz http://miguel.leugim.com.mx

March 17, 2011

Re: [Pharo-project] [Vm-dev] Socket Plugin primitives

by jannik.laval

On Mar 16, 2011, at 21:52 , Levente Uzonyi wrote: > On Wed, 16 Mar 2011, jannik.laval wrote: > >> >> Hi all, >> >> I am analyzing SocketPluggin for Ocean development. And I have some remarks/questions. >> >> First, to be sure, can you confirm my vision: >> - the vm 4.x is the old one. Is it maintained ? >> - vm 5.7.x is the normal vm (without JIT) >> - vm 5.8.x is the Cog VM with normal pluggin >> - vm Cog is Cog VM with less primitives than vm 5.8 (modified pluggins). >> >> Based on that, we can see that Cog has only 38 primitives related to SocketPluggin where normal pluggin provides 61 primitives. >> Could anyone explain to me why some primitives disappear ? > > Cog was forked from an old version of the VM sources that didn't have the IPv6 primitives, so those are missing. Ok, now I understand. > There may be other differences, so someone has to merge the code of the two (Cog's and the interpreter VM's) plugins. Which kind of difference ?You want to say in the source code ? Jannik > > > Levente > >> I do not yet analyze which primitives disappear but maybe someone could help me to understand this difference between the two pluggins. >> >> Thanks in advance for your help, >> --- >> Jannik Laval >> >>

March 17, 2011

Re: [Pharo-project] Building NativeBoost plugin (on linux)

by Javier PimÃ¡s

Done, now it's dying while running the tests. I'll try to make it work again tomorrow night (only if we finally get to release a new version of SqueakNOS, I promise). Here is the stack trace in case it's a known problem: Smalltalk stack dump: 0xbf78b038 M NBFFICallout class(NBNativeCodeGen class)>generateCode:andRetry: 2032660172: a(n) NBFFICallout class 0xbf78b9a8 I NativeBoostLinux32>getGlobalSymbolPointer: 2039879072: a(n) NativeBoostLinux32 0xbf78b9d0 I [] in NBUnixExternalHeapManager>mmapLength:prot:flags:into: 2041154832: a(n) NBUnixExternalHeapManager 0xbf78b9f4 I NBFFICallout>generate: 2041155916: a(n) NBFFICallout 0xbf78ba28 I [] in NBFFICallout class>cdecl:emitCall:options: 2032660172: a(n) NBFFICallout class 0xbf78ba64 I NBFFICallout class(NBNativeCodeGen class)>generateCode:andRetry: 2032660172: a(n) NBFFICallout class 0xbf78ba90 I NBFFICallout class(NBNativeCodeGen class)>handleFailureIn:nativeCode: 2032660172: a(n) NBFFICallout class 0xbf78babc I NBFFICallout class>cdecl:emitCall:options: 2032660172: a(n) NBFFICallout class 0xbf78bae8 I NBUnixExternalHeapManager>mmapLength:prot:flags:into: 2041154832: a(n) NBUnixExternalHeapManager 0xbf78bb20 I NBUnixExternalHeapManager>primAllocatePage: 2041154832: a(n) NBUnixExternalHeapManager 0xbf78bb50 I NBUnixExternalHeapManager(NBExternalHeapManager)>allocatePage: 2041154832: a(n) NBUnixExternalHeapManager 0xbf78bb74 M [] in NBUnixExternalHeapManager(NBExternalHeapManager)>allocate: 2041154832: a(n) NBUnixExternalHeapManager 0xbf78bb8c M IdentitySet(Collection)>detect:ifNone: 2041154916: a(n) IdentitySet 0xbf78bbc0 I [] in NBUnixExternalHeapManager(NBExternalHeapManager)>allocate: 2041154832: a(n) NBUnixExternalHeapManager 0xbf78bbe0 M [] in Semaphore>critical: 2041154856: a(n) Semaphore 0xbf78bc00 M BlockClosure>ensure: 2041155208: a(n) BlockClosure 0xbf78bc20 M Semaphore>critical: 2041154856: a(n) Semaphore 0xbf78bc44 I NBUnixExternalHeapManager(NBExternalHeapManager)>allocate: 2041154832: a(n) NBUnixExternalHeapManager 0xbf78bc6c I NativeBoostLinux32>basicAllocate: 2039879072: a(n) NativeBoostLinux32 0xbf78bc94 I NativeBoostLinux32(NativeBoost)>allocate: 2039879072: a(n) NativeBoostLinux32 0xbf78bcb8 I NativeBoostLinux32(NativeBoost)>initialize 2039879072: a(n) NativeBoostLinux32 0xbf78bcd0 M NativeBoostLinux32 class(Behavior)>new 2032913612: a(n) NativeBoostLinux32 class 0xbf78bcf4 I [] in NativeBoost class>discoverPlatformId 2032664204: a(n) NativeBoost class 0xbf78bd1c I [] in NativeBoost class(Behavior)>allSubclassesDo: 2032664204: a(n) NativeBoost class 0xbf78bd40 M Array(SequenceableCollection)>do: 2032915888: a(n) Array 0xbf78bd64 I NativeBoost class(Class)>subclassesDo: 2032664204: a(n) NativeBoost class 0xbf78bd88 I NativeBoost class(Behavior)>allSubclassesDo: 2032664204: a(n) NativeBoost class 0xbf78bdb0 I NativeBoost class>discoverPlatformId 2032664204: a(n) NativeBoost class 0xbf78bdd0 I NativeBoost class>forCurrentPlatform 2032664204: a(n) NativeBoost class 0xbf78bdf0 I NativeBoost class>allocate: 2032664204: a(n) NativeBoost class 0xbf78be10 M NBExternalAddressTests>testMemoryAccess 2039875280: a(n) NBExternalAddressTests 0xbf78be30 I NBExternalAddressTests(TestCase)>performTest 2039875280: a(n) NBExternalAddressTests 0xbf78be50 I [] in NBExternalAddressTests(TestCase)>runCase 2039875280: a(n) NBExternalAddressTests 0xbf78be70 M BlockClosure>ensure: 2039878812: a(n) BlockClosure 0xbf78be94 I NBExternalAddressTests(TestCase)>runCase 2039875280: a(n) NBExternalAddressTests On Wed, Mar 16, 2011 at 11:27 PM, Igor Stasenko <siguctua(a)gmail.com> wrote: > On 17 March 2011 03:20, Javier PimÃ¡s <elpochodelagente(a)gmail.com> wrote: > > I'm also seeing that NBCoInterpreter and NBCoObjectMemory only have small > > modifications compared to the non NB ones. Maybe it is possible to merge > > them directly into cog with an enable/disable flag, so that building NB > is > > the same as building Cog with NB enabled. > > > > Yes, the changes are very small, but i wanted to keep them separate, > so its easy to see them. > > So if you can't wait, you can just patch the original classes > and then generate VM sources using usual command. > But don't forget to include NativeBoostPlugin as internal one. > > If you using CMakeVMMaker, you do it like following: > > | config | > config := CogUnixConfig new. > config internalPlugins: (cog internalPlugins copyWith: #NativeBoostPlugin). > > config generateSourceFiles. > config generate. > > > > -- > Best regards, > Igor Stasenko AKA sig. > > -- Javier PimÃ¡s Ciudad de Buenos Aires

March 17, 2011

Re: [Pharo-project] Building NativeBoost plugin (on linux)

by Igor Stasenko

On 17 March 2011 03:27, Javier PimÃ¡s <elpochodelagente(a)gmail.com> wrote: > Maybe it is too much to ask, but having aÂ NBCogUnixDebugConfig would be > pretty useful in this context! > Not much.. just creating extra class + adding 1 extra method :) -- Best regards, Igor Stasenko AKA sig.

March 17, 2011