Vm-dev March 2012

vm-dev@lists.squeakfoundation.org

23 participants
41 discussions

ffi-test-main.c fails on FreeBSD
by Gary Dunn 12 Aug '16

12 Aug '16

In conjunction with getting Open Cobalt running on FreeBSD I have been looking into the problems with SqueakFFIPrims on FreeBSD and have reached a point that is beyond my understanding. I am not sure if the fix should come from the VM side or the FreeBSD port maintainer. I needed to get source code back into the ports tree so I did a make in /usr/ports/lang/squeak. I already had the 3.9 tarball, so that just repopulated the source and did the initial build. In the "files" folder I found patch files typical for a FreeBSD port. These patch source files from a generic tarball to work with FreeBSD. The one of interest was patch-platforms__unix__plugins__SqueakFFIPrims__ffi-config which contains ----x-----x----- --- platforms/unix/plugins/SqueakFFIPrims/ffi-config.org Wed Apr 26 20:27:53 2006 +++ platforms/unix/plugins/SqueakFFIPrims/ffi-config Wed Apr 26 20:29:00 2006 @@ -39,6 +39,7 @@ case ${abi} in linux) abi=sysv;; + freebsd*) abi=sysv;; darwin*) abi=darwin;; *) abi=libffi; lib="-lffi";; esac -----x-----x----- I then had the folder work/Squeak-3.9-7/platforms/unix/plugins/SqueakFFIPrims In it is a file 00README which explains how to test FFI. It says to run ffi-test-config but I have no such file. It says to type "make" to build a test suite but the build failed. It says to try with make CPU=any ABI=libffi LIB=-lffi which also failed. I traced the problem to the gcc search paths for includes and libraries -- turns out FreeBSD only searches /usr/include and /usr/lib when the required files are in /usr/local/include and /usr/local/lib. After some fiddling I finally had a "main" to test. It failed: ffi-test-main.c failed at line 361. Here is a bit from that file, ending with line 361: ffiInitialize(); for (ul= 0; ul < 15; ++ul) ffiPushSingleFloat(fa[ul]); GO(FFITypeSingleFloat, many); f= ffiReturnFloatValue(); ffiCleanup(); CHECK(f - ff < FLT_EPSILON); I worked out the CHECK macro but could not find what FLT_EPSILON is. At that point I decided to stop and come to you for help. Where do I go from here? -- Gary Dunn, Honolulu osp(a)aloha.com http://openslate.net/ http://e9erust.blogspot.com/ Sent from Slate001

5 8

Do we have the new primitive?? [WAS] Re: [Pharo-project] IdentitySet but using #hash rather than #identityHash ?
by Mariano Martinez Peck 16 Dec '15

16 Dec '15

> All I can say is that I am impressed by the numbers it is really much >> faster. >> I still don't understand why I send this email with a subject say >> IdentitySet because what I really need is a fast/large IdentityDictionary >> :( Anyway, there's a place where we can use this LargeIdentitySet in Fuel >> I think). >> >> So Levente, you say this is not possible to adapt this for dictionary? >> can >> we contact Eliot to provide such a primitive? >> > > As promised, I uploaded my LargeIdentityDictionary implementation to > http://leves.web.elte.hu/**squeak/**LargeIdentityDictionary.st<http://leves.web.elte.hu/squeak/LargeIdentityDictionary.st>. > The numbers will be a bit worse compared to LargeIdentitySet, because of > the lack of the primitive, but it's still 2-3x faster than other solutions > (IdentityDictionary, PluggableIdentityDictionary, subclassing, etc). I'm > about to propose this primitive with other improvements on the vm-dev list. > > Hi Eliot/Levente. What is the status of this? Do we have already the new primitive? If true, how can we adapt LargeIdentitySet to use such new primitive? Thanks! > > Levente > > >> thanks >> >> On Fri, Dec 16, 2011 at 3:28 PM, Levente Uzonyi <leves(a)elte.hu> wrote: >> >> On Fri, 16 Dec 2011, Henrik Sperre Johansen wrote: >>> >>> On 16.12.2011 03:26, Levente Uzonyi wrote: >>> >>>> >>>> >>>>> How about my numbers? :) >>>>> >>>>> "Preallocate objects, so we won't count gc time." >>>>> n := 1000000. >>>>> objects := Array new: n streamContents: [ :stream | >>>>> n timesRepeat: [ stream nextPut: Object new ] ]. >>>>> >>>>> set := IdentitySet new: n. >>>>> Smalltalk garbageCollect. >>>>> [1 to: n do: [ :i | set add: (objects at: i) ] ] timeToRun. "4949" >>>>> >>>>> set := LargeIdentitySet new. >>>>> Smalltalk garbageCollect. >>>>> [1 to: n do: [ :i | set add: (objects at: i) ] ] timeToRun. "331" >>>>> >>>>> set := (PluggableSet new: n) >>>>> hashBlock: [ :object | object identityHash * 4096 + object class >>>>> identityHash * 64 ]; "Change this to #basicIdentityHash in Pharo" >>>>> equalBlock: [ :a :b | a == b ]; >>>>> yourself. >>>>> Smalltalk garbageCollect. >>>>> [1 to: n do: [ :i | set add: (objects at: i) ] ] timeToRun. "5511" >>>>> >>>>> >>>>> I also have a LargeIdentityDictionary, which is relatively fast, but >>>>> not >>>>> as fast as LargeIdentitySet, because (for some unknown reason) we don't >>>>> have a primitive that could support it. If we had a primitive like >>>>> primitive 132 which would return the index of the element if found or >>>>> 0 if >>>>> not, then we could have a really fast LargeIdentityDictionary. >>>>> >>>>> >>>>> Levente >>>>> >>>>> Hehe yes, if writing a version fully exploiting the limited range, >>>> that's >>>> probably the approach I would go for as well. >>>> (IAssuming it's the version at http://leves.web.elte.hu/** >>>> squeak/LargeIdentitySet.st<htt**p://leves.web.elte.hu/squeak/** >>>> LargeIdentitySet.st<http://leves.web.elte.hu/squeak/LargeIdentitySet.st> >>>> > >>>> ) >>>> >>>> Mariano commented in the version at http://www.squeaksource.com/** >>>> FuelExperiments <http://www.squeaksource.com/**FuelExperiments<http://www.squeaksource.com/FuelExperiments>> >>>> that it's >>>> >>>> slow for them, which I guess is due to not adopting #identityHash calls >>>> to >>>> #basicIdentityHash calls for Pharo: >>>> ((0 to: 4095) collect: [:each | each << 22 \\ 4096 ]) asSet size -> 1 >>>> So it basically uses 1 bucket instead of 4096... Whoops. :) >>>> >>>> Uploaded a new version to the MC repository which is adapted for Pharo, >>>> on the same machine my numbers were taken from, it does the same test >>>> as I >>>> used above in 871 ms. (181 with preallocation). >>>> >>>> >>> Cool. One more thing: in Squeak the method using primitive 132 directly >>> was renamed to #instVarsInclude:, so now #pointsTo: works as expected. If >>> this was also added to Pharo, then the #pointsTo: sends should be changed >>> to #instVarsInclude:, otherwise Array can be reported as included even if >>> it wasn't added. >>> I'll upload my LargeIdentityDictionary implementation to the same place >>> this evening, since it's still 2-3 factor faster than other solutionts >>> and >>> there seem to be demand for it. >>> >>> >>> Levente >>> >>> >>> Cheers, >>>> Henry >>>> >>>> >>>> >>>> >>> >> >> -- >> Mariano >> http://marianopeck.wordpress.**com <http://marianopeck.wordpress.com> >> >> > -- Mariano http://marianopeck.wordpress.com

4 12

[CogVM-blessed] -headless stopped working
by Stefan Marr 02 May '12

02 May '12

Hi Esteban: I just downloaded a Cog-Mac-Cocoa-blessed from: https://ci.lille.inria.fr/pharo/job/Cog-Mac-Cocoa-blessed/ It tells me, it is currently 19days old. And it ignores the -headless parameter. The old VM I used so far is from Jul 6th and does execute headless without problems. I did not find the mail on the list that was naming the primitives to get more version infos out of the VMs, sorry. The latest StackVM also got that problem, but it works with the version from July. Is there actually something like -version on the command line to ask for the exact version infos? I checked -help, which also offers a lot less info then it used to. It still mentions -headless, but misses the options below. Best regards Stefan Missing help for commandline options: -breaksel selector set breakpoint on send of selector -eden <size>[mk] set eden memory to bytes -leakcheck num check for leaks in the heap -stackpages num use n stack pages -codesize <size>[mk] set machine code memory to bytes -sendtrace[=num] enable send tracing (optionally to a specific value) -tracestores enable store tracing (assert check stores) -cogmaxlits <n> set max number of literals for methods compiled to machine code -cogminjumps <n> set min number of backward jumps for interpreted methods to be considered for compilation to machine code -pathenc <enc> set encoding for pathnames (default: macintosh) -- Stefan Marr Software Languages Lab Vrije Universiteit Brussel Pleinlaan 2 / B-1050 Brussels / Belgium http://soft.vub.ac.be/~smarr Phone: +32 2 629 2974 Fax: +32 2 629 3525

4 5

another GCC issue?
by Igor Stasenko 03 Apr '12

03 Apr '12

or perhaps not.. i am trying to build the cog cocoa VMs using configs made by Esteban.. CogCocoaIOSConfig new " generateForDebug;" generateForRelease; addExternalPlugins: #( FT2Plugin ); addInternalPlugins: #( UnixOSProcessPlugin ); generateSources; generate. it is not a functional bug.. because if i use #generateForDebug, everything works fine (but slow ;) but if i do #genearateForRelease, VM crashes with following: Program received signal EXC_BAD_ACCESS, Could not access memory. Reason: KERN_INVALID_ADDRESS at address: 0x5eedca7a 0x0000dfb6 in concretizeAt () (gdb) bt #0 0x0000dfb6 in concretizeAt () #1 0x00014996 in generateInstructionsAt () #2 0x000000c9 in ?? () i am not sure how to squeeze more info about this crash point.. because stack trace is cut by gdb, which /.another rant here./ stupidly stops scanning stack frames once it discovers a frame which code is outside of memory covered by debug information.. Ah.. yes.. and addition info.. the difference between generateForDebug and generateForRelease compilerFlagsRelease ^#( "'-fobjc-direct-dispatch'" '-msse3' "'-msse4.1'" "'-msse4.2'" "'-mdynamic-no-pic'" "'-fwritable-strings'" '-Os' '-fvisibility=hidden' '-funroll-loops' "'-fno-asm'" '-fasm-blocks' '-finline-functions' '-mfpmath=sse' '-fomit-frame-pointer' '-march=pentium-m' '-mtune=prescott' '-falign-functions=16' '-fno-gcse' '-fno-cse-follow-jumps' '-std=gnu99' '-DBUILD_FOR_OSX' '-DUSE_INLINE_MEMORY_ACCESSORS' '-DLSB_FIRST' '-DUSE_INLINE_MEMORY_ACCESSORS' '-DHAVE_SYS_TIME_H' '-DHAVE_NANOSLEEP' '-DICC_DEBUG=0' '-DICC_OPTLEVEL="speedHLO"' '-DICC_OPT_IPO_FOR_SINGLE_FILE_COMPILATION=1' '-DICC_OPT_PARALLEL=0' '-DICC_OPT_PREFETCH_INSERTION=1' '-DICC_OPT_PROVIDE_FRAME_PTR=0' '-DICC_OPT_USE_ARCH_IA32="SSE42"') compilerFlagsDebug ^#( '-g3' '-O0' '-DDEBUGVM=1') -- Best regards, Igor Stasenko.

4 8

[Vm-dev] OSProcessPlugin: VMConstruction-Plugins-OSProcessPlugin.oscog-eem.34.mcz
by squeak-dev-noreply＠lists.squeakfoundation.org 31 Mar '12

31 Mar '12

Eliot Miranda uploaded a new version of VMConstruction-Plugins-OSProcessPlugin to project OSProcessPlugin: http://www.squeaksource.com/OSProcessPlugin/VMConstruction-Plugins-OSProces… ==================== Summary ==================== Name: VMConstruction-Plugins-OSProcessPlugin.oscog-eem.34 Author: eem Time: 31 March 2012, 1:56:42 am UUID: 90d36152-4f89-4764-b476-2cc44c195139 Ancestors: VMConstruction-Plugins-OSProcessPlugin.oscog-eem.33 Merge VMConstruction-Plugins-OSProcessPlugin-dtl.33's changes

1 0

VM Maker: VMMaker.oscog-eem.153.mcz
by commits＠source.squeak.org 31 Mar '12

31 Mar '12

Eliot Miranda uploaded a new version of VMMaker to project VM Maker: http://source.squeak.org/VMMaker/VMMaker.oscog-eem.153.mcz ==================== Summary ==================== Name: VMMaker.oscog-eem.153 Author: eem Time: 31 March 2012, 1:36:50.547 pm UUID: 8ace2909-a8c0-4bcf-8274-55850a71487d Ancestors: VMMaker.oscog-eem.152 Fix header file generation for bytecodeSetSelector. Make sure BytesPerOop is initialized in initializeWithOptions:. Add longPrintReferencesTo:. Search for configurations in package extensions to VMMaker. Comment fixes. =============== Diff against VMMaker.oscog-eem.152 =============== Item was changed: ----- Method: CCodeGenerator>>emitGlobalCVariablesOn: (in category 'C code generator') ----- emitGlobalCVariablesOn: aStream "Store the global variable declarations on the given stream." aStream cr; nextPutAll: '/*** Global Variables ***/'; cr. (self sortStrings: (variables select: [:v| vmClass mustBeGlobal: v])) do: [:var | | varString decl | varString := var asString. decl := variableDeclarations at: varString ifAbsent: ['sqInt ' , varString]. + decl first == $# "support cgen var: #bytecodeSetSelector declareC: '#define bytecodeSetSelector 0' hack" + ifTrue: + [aStream nextPutAll: decl; cr] + ifFalse: + [((decl includesSubString: ' private ') + or: [decl beginsWith: 'static']) ifFalse: "work-around hack to prevent localization of variables only referenced once." + [(decl includes: $=) ifTrue: + [decl := decl copyFrom: 1 to: (decl indexOf: $=) - 1]. + aStream + nextPutAll: decl; + nextPut: $;; + cr]]]. - ((decl includesSubString: ' private ') - or: [decl beginsWith: 'static']) ifFalse: "work-around hack to prevent localization of variables only referenced once." - [(decl includes: $=) ifTrue: - [decl := decl copyFrom: 1 to: (decl indexOf: $=) - 1]. - aStream - nextPutAll: decl; - nextPut: $;; - cr]]. aStream cr! Item was changed: ----- Method: Interpreter class>>initializeWithOptions: (in category 'initialization') ----- initializeWithOptions: optionsDictionary "Interpreter initializeWithOptions: Dictionary new" super initializeWithOptions: optionsDictionary. "initialize ObjectMemory constants" + self initializeMiscConstantsWith: optionsDictionary. "must preceed other initialization." - self initializeMiscConstants. "must preceed other initialization." self initializeAssociationIndex. self initializeBytecodeTable. self initializeCaches. self initializeCharacterIndex. self initializeCharacterScannerIndices. self initializeClassIndices. self initializeCompilerHooks. self initializeContextIndices. self initializeDirectoryLookupResultCodes. self initializeMessageIndices. self initializeMethodIndices. self initializePointIndices. self initializePrimitiveTable. self initializeSchedulerIndices. self initializeSmallIntegers. self initializeStreamIndices! Item was added: + ----- Method: NewObjectMemory>>longPrintReferencesTo: (in category 'debug printing') ----- + longPrintReferencesTo: anOop + "Scan the heap long printing the oops of any and all objects that refer to anOop" + | oop i prntObj | + <api> + prntObj := false. + oop := self firstAccessibleObject. + [oop = nil] whileFalse: + [((self isPointers: oop) or: [self isCompiledMethod: oop]) ifTrue: + [(self isCompiledMethod: oop) + ifTrue: + [i := (self literalCountOf: oop) - 1] + ifFalse: + [(self isContext: oop) + ifTrue: [i := CtxtTempFrameStart + (self fetchStackPointerOf: oop) - 1] + ifFalse: [i := (self lengthOf: oop) - 1]]. + [i >= 0] whileTrue: + [anOop = (self fetchPointer: i ofObject: oop) ifTrue: + [self printHex: oop; print: ' @ '; printNum: i; cr. + prntObj := true. + i := 0]. + i := i - 1]. + prntObj ifTrue: + [prntObj := false. + coInterpreter longPrintOop: oop]]. + oop := self accessibleObjectAfter: oop]! Item was changed: ----- Method: ObjectMemory class>>initializeWithOptions: (in category 'initialization') ----- initializeWithOptions: optionsDictionary "ObjectMemory initializeWithOptions: Dictionary new" + self initBytesPerWord: (optionsDictionary at: #BytesPerWord ifAbsent: [4]). + BytesPerOop := optionsDictionary at: #BytesPerOop ifAbsent: [BytesPerWord]. - self initBytesPerWord: (optionsDictionary at: #BytesPerWord ifAbsent: [4]).. "Translation flags (booleans that control code generation via conditional translation):" "generate assertion checks" DoAssertionChecks := optionsDictionary at: #DoAssertionChecks ifAbsent: [false]. DoExpensiveAssertionChecks := optionsDictionary at: #DoExpensiveAssertionChecks ifAbsent: [false]. self initializeSpecialObjectIndices. self initializeCompactClassIndices. self initializePrimitiveErrorCodes. self initializeObjectHeaderConstants. NilContext := 1. "the oop for the integer 0; used to mark the end of context lists" RemapBufferSize := 25. RootTableSize := 2500. "number of root table entries (4 bytes/entry)" RootTableRedZone := RootTableSize - 100. "red zone of root table - when reached we force IGC" WeakRootTableSize := RootTableSize + RemapBufferSize + 100. "tracer actions" StartField := 1. StartObj := 2. Upward := 3. Done := 4. ExtraRootSize := 2048. "max. # of external roots"! Item was changed: + ----- Method: ObjectMemory>>printInstancesOf: (in category 'debug printing') ----- - ----- Method: ObjectMemory>>printInstancesOf: (in category 'memory access') ----- printInstancesOf: aClassOop "Scan the heap printing the oops of any and all objects that are instances of aClassOop" | oop | <api> oop := self firstAccessibleObject. [oop = nil] whileFalse: [(self fetchClassOfNonInt: oop) = aClassOop ifTrue: [self printHex: oop; cr]. oop := self accessibleObjectAfter: oop]! Item was changed: + ----- Method: ObjectMemory>>printMethodReferencesTo: (in category 'debug printing') ----- - ----- Method: ObjectMemory>>printMethodReferencesTo: (in category 'memory access') ----- printMethodReferencesTo: anOop "Scan the heap printing the oops of any and all objects that refer to anOop" | oop i | <api> oop := self firstAccessibleObject. [oop = nil] whileFalse: [(self isCompiledMethod: oop) ifTrue: [i := (self literalCountOf: oop) - 1. [i >= 0] whileTrue: [anOop = (self fetchPointer: i ofObject: oop) ifTrue: [self printHex: oop; print: ' @ '; printNum: i; cr. i := 0]. i := i - 1]]. oop := self accessibleObjectAfter: oop]! Item was changed: + ----- Method: ObjectMemory>>printReferencesTo: (in category 'debug printing') ----- - ----- Method: ObjectMemory>>printReferencesTo: (in category 'memory access') ----- printReferencesTo: anOop "Scan the heap printing the oops of any and all objects that refer to anOop" | oop i | <api> oop := self firstAccessibleObject. [oop = nil] whileFalse: [((self isPointers: oop) or: [self isCompiledMethod: oop]) ifTrue: [(self isCompiledMethod: oop) ifTrue: [i := (self literalCountOf: oop) - 1] ifFalse: [(self isContext: oop) ifTrue: [i := CtxtTempFrameStart + (self fetchStackPointerOf: oop) - 1] ifFalse: [i := (self lengthOf: oop) - 1]]. [i >= 0] whileTrue: [anOop = (self fetchPointer: i ofObject: oop) ifTrue: [self printHex: oop; print: ' @ '; printNum: i; cr. i := 0]. i := i - 1]]. oop := self accessibleObjectAfter: oop]! Item was changed: ----- Method: SistaStackToRegisterMappingCogit>>genMustBeBooleanTrampolineFor:called: (in category 'initialization') ----- genMustBeBooleanTrampolineFor: boolean called: trampolineName + "This can be entered in one of two states, depending on SendNumArgsReg. See + e.g. genJumpIf:to:. If SendNumArgsReg is non-zero then this has been entered via + the initial test of the counter in the jump executed count (i.e. the counter has + tripped). In this case TempReg contains the boolean to be tested and should not + be offset, and ceCounterTripped should be invoked with the unoffset TempReg. + If SendNumArgsReg is zero then this has been entered for must-be-boolean + processing. TempReg has been offset by boolean and must be corrected and + ceSendMustBeBoolean: invoked with the corrected value." - "THis can be entered in one of two states, depending on SendNumArgsReg. - If SendNumArgsReg is non-zero then this has been entered via the initial test of - the counted in the jump executed count. In this case TempReg contains the - boolean to be tested and should not be offset, and ceCounterTripped should be - invoked with the unoffset TempReg. - If SendNumArgsReg is zero then this has been entered for must-be-boolean processing. - TempReg has been offset by boolean and must be corrected and ceSendMustBeBoolean: - invoked with the offset value." <var: #trampolineName type: #'char *'> | jumpMBB | <var: #jumpMBB type: #'AbstractInstruction *'> <inline: false> opcodeIndex := 0. self CmpCq: 0 R: SendNumArgsReg. jumpMBB := self JumpZero: 0. self compileTrampolineFor: #ceCounterTripped: asSymbol callJumpBar: true numArgs: 1 arg: TempReg arg: nil arg: nil arg: nil saveRegs: false resultReg: nil. "If the objectRepresentation does want true & false to be mobile then we need to record these addresses." self assert: (objectRepresentation shouldAnnotateObjectReference: boolean) not. jumpMBB jmpTarget: (self AddCq: boolean R: TempReg). ^self genTrampolineFor: #ceSendMustBeBoolean: asSymbol called: trampolineName callJumpBar: true numArgs: 1 arg: TempReg arg: nil arg: nil arg: nil saveRegs: false resultReg: nil appendOpcodes: true! Item was changed: ----- Method: StackInterpreter>>markContextAsDead: (in category 'frame access') ----- markContextAsDead: oop "Mark the argument, which must be a context, married, widowed or single, as dead. For married or widowed contexts this breaks any link to the spouse and makes the context single. + For all contexts, marks the context as inactive/having been returned from." - For all contexts, marks the comtext as inactive/having been returned from." <inline: true> self assert: (self isContext: oop). objectMemory storePointerUnchecked: SenderIndex ofObject: oop withValue: objectMemory nilObject; storePointerUnchecked: InstructionPointerIndex ofObject: oop withValue: objectMemory nilObject! Item was changed: ----- Method: VMMaker class>>generateConfiguration (in category 'configurations') ----- generateConfiguration "VMMaker generateConfiguration" + | configCategoryName selectors | + configCategoryName := self class whichCategoryIncludesSelector: thisContext selector. + selectors := Set new. + self class organization categories do: + [:cat| + (cat endsWith: configCategoryName) ifTrue: + [selectors addAll: (self class organization listAtCategoryNamed: cat)]]. + selectors remove: thisContext selector. + selectors := selectors asArray sort. - | selectors | - selectors := ((self class organization listAtCategoryNamed: (self class whichCategoryIncludesSelector: thisContext selector)) copyWithout: thisContext selector) sort. (UIManager default chooseFrom: (selectors collect: [:sel| (sel piecesCutWhere: [:a :b| a isLowercase and: [b isUppercase]]) allButFirst fold: [:a :b| a, ' ', b]]) values: selectors) ifNotNil: [:choice| self perform: choice]!

1 0

Re: [Pharo-project] direct object manipulation
by Eliot Miranda 30 Mar '12

30 Mar '12

On Fri, Mar 30, 2012 at 1:20 PM, Stéphane Ducasse <stephane.ducasse(a)inria.fr > wrote: > > On Mar 30, 2012, at 9:50 PM, phil(a)highoctane.be wrote: > > > This is a chicken and egg problem. > > And a popularity contest game as well. > > > > Without one click, you'll piss people off fast and they'll go look > elsewhere. Then you loose traction and that's bad. > > > > If I didn't had one click, I wouldn't have been learning as fast as I've > been. > > I agree with you. > > > Now I do have all pieces from different locations (even wanting to build > the VM on my own). > > > > For contrast, look at another community I am part of: Tiki Wiki CMS > Groupware. The system is topping SourceForge number of downloads, > committers, and is a general success for intranets in companies. > > > > Of course, the paradigm isn't pure at all, but easy enough to have a lot > of people contributing good stuff. > > > > We are at version 9 and I don't know any other package that is as > feature rich when it comes to the web. > > We are talking more than a million lines of code here. > > Now, we are moving further than one click: hosting providers provide a > full VM with everything installed for the stack. > > > > As of software developers, I do view 4 categories: > > > > - constructivists using established abstractions and not requiring > further exposure to inner workings of the system. They deliver business > value for common problems. > > - bridges understanding how the system works inside and are able to > convey usage patterns to the constructivists. They also feed back real > world usage problems to the next kind so that the system stays useful and > not pie in the skyish. > > - abstractionists: able to come up with a powerful set of abstractions > to solve a given problem in a general manner. Framework makers are here > > - technology makers: create core tech. Like a Cog-VM thing, CUDA, .. > > > > Try to pair types by jumping one strata and you are asking for trouble. > > > > Except for rare seasoned developers that are master of their craft and > are able to display behavioral flexibility. > > But these are a rare breed and you won't get them for a long time anyway > as these are seekers. > > I want a good installer for each platform :) > For newspeak there is a native installer for Mac OS and Windows. See http://www.squeakvm.org/svn/squeak/branches/Cog/nscogbuild/cygwinbuild/inst… http://www.squeakvm.org/svn/squeak/branches/Cog/nscogbuild/macbuild/install…. These allow one to install whatever one specifies using a native installer. The WIndows code is a little hairy and the Mac OS code is very simple. So if you consider a native installer a good introductory choice (it doesn't have to be the only one) then take a look. I can help with anyone trying to tailor to the WIndows installer. > And I want to let a chance to people to decide when they have to deal with > low level. > Pushing shit on their faces as a welcome message is not what we want. > > Now alex shipping cairo lib may be not be the best to do. It tooks some > time to compile on my machine > and it means some work to build the right libr for each sub platforms: 32, > 64….. > > Now you can also use VW. :) For me this is simply no way. ;) > > Stef > > > > -- best, Eliot

1 0

Re: [Pharo-project] Enhancing SmalltalkImage>>recreateSpecialObjectsArray to lower coupling
by Eliot Miranda 29 Mar '12

29 Mar '12

Hi Guillermo, On Sun, Mar 18, 2012 at 5:46 PM, Guillermo Polito <guillermopolito(a)gmail.com > wrote: > What about changing #recreateSpecialObjectsArray so having in mind some > classes like Bitmap or Display may not be in the kernel? > This only changes some lines like: > > ... > newArray at: 5 put: (self at: #Bitmap ifAbsent: [ nil ]). > > ... > > newArray at: 13 put: (self at: #Point ifAbsent: [ nil ]). > > ... > > newArray at: 15 put: (self at: #Display ifAbsent: [ nil ]). > > ... > > newArray at: 34 put: (self at: #Point ifPresent: [ :cls | 0@0] > ifAbsent: [ nil ]). > The last one isn't used in Cog (in fact indices 32, 33 & 34 are unused) and probably unused in the Interpreter. David? You may be able to get away with making Bitmap and DIsplay optional. But Point is dug pretty deep into the VM. Point is used for the special selector send bytecodes for #x, #y & #@. For #x & #y there's a check that the class of the receiver is Point, and if so, evaluate in-line. So that is fine; if Point is nil then #x will get sent, rather than evaluated inline. But if the system sends #@ it'll likely crash, trying to create an object whose class is nil. To do what you're proposing properly the VM would need to be modified to be safe. We should perhaps discuss what the right approach is. e.g. if Point is in fact nil then #x #y and #@ would all slow down slightly in normal use checking for the possibility of Point being nil. This isn't an issue in Cog since these messages are compiled to regular sends; but t=in the regular interpreter and the StackInterpreter (used on e.g. Android) it may be. A different approach might be to move these classes to Kernel, or to provide rump classes that raise errors when used, e.g. newArray at: 13 put: (self at: #Point ifAbsent: [ RumpPoint ]). (the rump of something being what's left when you take everything away, rump = behind, bottom etc) > > Does anyone know if this should this break something? > > Guille > -- best, Eliot

2 1

[Vm-dev] OSProcessPlugin: VMConstruction-Plugins-OSProcessPlugin-dtl.33.mcz
by squeak-dev-noreply＠lists.squeakfoundation.org 28 Mar '12

28 Mar '12

Dave Lewis uploaded a new version of VMConstruction-Plugins-OSProcessPlugin to project OSProcessPlugin: http://www.squeaksource.com/OSProcessPlugin/VMConstruction-Plugins-OSProces… ==================== Summary ==================== Name: VMConstruction-Plugins-OSProcessPlugin-dtl.33 Author: dtl Time: 28 March 2012, 12:15:26 pm UUID: 458ec77f-53e7-42c8-bde1-2aa21992afe6 Ancestors: VMConstruction-Plugins-OSProcessPlugin-dtl.32 OSProcessPlugin 4.4.9 Fix missing type declaration in #newPthreadTypeByteArray: (thanks Camillo Bruni), and get rid of a #cCode: while we are at it. Clean up #printAllStacksFunctionAddress to eliminate compiler warnings.

1 0

Re: [Pharo-project] Bug in WAMemoryItem>>sizeOfObject
by Eliot Miranda 27 Mar '12

27 Mar '12

Hi Niko, the header size calculation is also wrong. If the byte size if the object body is > 255 bytes there needs to be an additional header size word. So something like byteSizeOfBody := anObject class isBytes ifTrue: [....] ifFalse: [.....]. headerSize := byteSizeOfBody > 255 ifTrue: [12] ifFalse: [anObject class indexIfCompact > 0 ifTrue: [4] ifFalse: [8]]. On Wed, Mar 14, 2012 at 1:05 PM, <niko.schwarz(a)googlemail.com> wrote: > Hi guys, > > I haven't checked if this bug exists already, so please ignore this if > it's been reported before. > > Bug in WAMemoryItem: > > sizeOfObject: anObject > | headerSize instanceSize variableSize | > headerSize := anObject class indexIfCompact > 0 > ifTrue: [ 4 ] > ifFalse: [ 8 ]. > instanceSize := anObject class instSize. > variableSize := anObject class isBytes > ifTrue: [ anObject basicSize ] > ifFalse: [ > anObject class isWords > ifTrue: [ Smalltalk wordSize * anObject > basicSize ] > ifFalse: [ Smalltalk wordSize * anObject > basicSize // 2 ] ]. > ^ headerSize + instanceSize + variableSize > > > In this snippet, the units that are added together in the last line are of > different units. > > instanceSize is in unit "number of machine words", whereas variableSize > and headerSize are in unit "bytes". > > > The method is fixed here: > > sizeOfObject: anObject > | headerSize instanceSize variableSize | > headerSize := anObject class indexIfCompact > 0 > ifTrue: [ 4 ] > ifFalse: [ 8 ]. > instanceSize := anObject class instSize * Smalltalk wordSize. > variableSize := anObject class isBytes > ifTrue: [ anObject basicSize ] > ifFalse: [ > anObject class isWords > ifTrue: [ Smalltalk wordSize * anObject > basicSize ] > ifFalse: [ Smalltalk wordSize * anObject > basicSize // 2 ] ]. > ^ headerSize + instanceSize + variableSize > > > > Example: > > |m| > dict := Dictionary new. > m := WAMemory new. dict traverseWithMemory: m seen: IdentitySet new. > m totalSize > > This will incorrectly spit out 10, but should return 16, on a 32 bit VM. > > Niko > > -- > http://scg.unibe.ch/staff/Schwarz > twitter.com/nes1983 > Tel: +41786126354 > > > -- best, Eliot

2 1

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

Vm-dev March 2012