Vm-dev March 2016

vm-dev@lists.squeakfoundation.org

33 participants
139 discussions

Re: [Vm-dev] VM Maker: VMMaker.oscog-cb.1617.mcz
by Clément Bera 03 Mar '16

03 Mar '16

Ok. According to my attempts, Stack and Cog Simulators are working fine (both Spur and SqueakV3). However the VM compiled to C crashes while trying to assign a block into an instance variable. So something is still wrong. 2016-01-06 14:46 GMT+01:00 <commits(a)source.squeak.org>: > > Eliot Miranda uploaded a new version of VMMaker to project VM Maker: > http://source.squeak.org/VMMaker/VMMaker.oscog-cb.1617.mcz > > ==================== Summary ==================== > > Name: VMMaker.oscog-cb.1617 > Author: cb > Time: 6 January 2016, 2:46:27.798 pm > UUID: 81fe5b89-69de-45b8-9b65-f2f4a8bd187d > Ancestors: VMMaker.oscog-rmacnak.1616 > > As Eliot ntoed, I made a dumb mistake breking instance variable stores in > the interpreter. > > This commit fixes the mistake. In addition, storing into an immutable > object in a primitive now signals a no modification error instead of > inappropriate error. > > I checked and the StackVMSimulator is working fine after this commit. > Unfortunately, the CogVMSimulator does not work in my machine, likely due > to other bugs. I am going to check that it works and that compilation to C > works right now. > > =============== Diff against VMMaker.oscog-rmacnak.1616 =============== > > Item was changed: > ----- Method: CoInterpreterPrimitives>>primitiveObjectAtPut (in category > 'object access primitives') ----- > primitiveObjectAtPut > "Store a literal into a CompiledMethod at the given index. Defined > for CompiledMethods only." > | thisReceiver rawHeader realHeader index newValue | > newValue := self stackValue: 0. > index := self stackValue: 1. > (objectMemory isNonIntegerObject: index) ifTrue: > [^self primitiveFailFor: PrimErrBadArgument]. > index := objectMemory integerValueOf: index. > thisReceiver := self stackValue: 2. > self cppIf: IMMUTABILITY > + ifTrue: [ (objectMemory isImmutable: thisReceiver) ifTrue: > [ ^self primitiveFailFor: PrimErrNoModification ] ]. > - ifTrue: [ (objectMemory isImmutable: thisReceiver) ifTrue: > [ ^self primitiveFailFor: PrimErrInappropriate ] ]. > rawHeader := self rawHeaderOf: thisReceiver. > realHeader := (self isCogMethodReference: rawHeader) > ifTrue: [(self cCoerceSimple: > rawHeader to: #'CogMethod *') methodHeader] > ifFalse: [rawHeader]. > (index > 0 > and: [index <= ((objectMemory literalCountOfMethodHeader: > realHeader) + LiteralStart)]) ifFalse: > [^self primitiveFailFor: PrimErrBadIndex]. > index = 1 > ifTrue: > [((objectMemory isNonIntegerObject: newValue) > or: [(objectMemory literalCountOfMethodHeader: > newValue) ~= (objectMemory literalCountOfMethodHeader: realHeader)]) ifTrue: > [^self primitiveFailFor: > PrimErrBadArgument]. > (self isCogMethodReference: rawHeader) > ifTrue: [(self cCoerceSimple: rawHeader > to: #'CogMethod *') methodHeader: newValue] > ifFalse: [objectMemory > storePointerUnchecked: 0 ofObject: thisReceiver withValue: newValue]] > ifFalse: > [objectMemory storePointer: index - 1 ofObject: > thisReceiver withValue: newValue]. > self pop: 3 thenPush: newValue! > > Item was changed: > ----- Method: InterpreterPrimitives>>primitiveFloatAtPut (in category > 'indexing primitives') ----- > primitiveFloatAtPut > "Provide platform-independent access to 32-bit words comprising > a Float. Map index 1 onto the most significant word and index 2 > onto the least significant word." > | rcvr index oopToStore valueToStore | > <var: #valueToStore type: #usqInt> > oopToStore := self stackTop. > valueToStore := self positive32BitValueOf: oopToStore. > self successful ifFalse: > [^self primitiveFailFor: PrimErrBadArgument]. > rcvr := self stackValue: 2. > index := self stackValue: 1. > (objectMemory isImmediateFloat: rcvr) ifTrue: > [^self primitiveFailFor: PrimErrBadReceiver]. > self cppIf: IMMUTABILITY > + ifTrue: [ (objectMemory isImmutable: rcvr) ifTrue: [^self > primitiveFailFor: PrimErrNoModification] ]. > - ifTrue: [ (objectMemory isImmutable: rcvr) ifTrue: [^self > primitiveFailFor: PrimErrBadReceiver] ]. > index = ConstOne ifTrue: > [objectMemory storeLong32: (VMBIGENDIAN ifTrue: [0] > ifFalse: [1]) > ofObject: rcvr > withValue: valueToStore. > ^self pop: 3 thenPush: oopToStore]. > index = ConstTwo ifTrue: > [objectMemory storeLong32: (VMBIGENDIAN ifTrue: [1] > ifFalse: [0]) > ofObject: rcvr > withValue: valueToStore. > ^self pop: 3 thenPush: oopToStore]. > self primitiveFailFor: ((objectMemory isIntegerObject: index) > ifTrue: > [PrimErrBadIndex] > ifFalse: > [PrimErrBadArgument])! > > Item was changed: > ----- Method: InterpreterPrimitives>>primitiveIntegerAtPut (in category > 'sound primitives') ----- > primitiveIntegerAtPut > "Return the 32bit signed integer contents of a words receiver" > | index rcvr sz addr value valueOop | > <var: 'value' type: 'int'> > valueOop := self stackValue: 0. > index := self stackIntegerValue: 1. > value := self signed32BitValueOf: valueOop. > self successful ifFalse: > [^self primitiveFailFor: PrimErrBadArgument]. > rcvr := self stackValue: 2. > (objectMemory isWords: rcvr) ifFalse: > [^self primitiveFailFor: PrimErrInappropriate]. > self cppIf: IMMUTABILITY "isWords: ensure non immediate" > + ifTrue: [ (objectMemory isImmutable: rcvr) ifTrue: [ ^self > primitiveFailFor: PrimErrNoModification ] ]. > - ifTrue: [ (objectMemory isImmutable: rcvr) ifTrue: [ ^self > primitiveFailFor: PrimErrInappropriate ] ]. > sz := objectMemory lengthOf: rcvr. "number of fields" > (index >= 1 and: [index <= sz]) ifFalse: > [^self primitiveFailFor: PrimErrBadIndex]. > "4 = 32 bits / 8" > addr := rcvr + objectMemory baseHeaderSize + (index - 1 * 4). "for > zero indexing" > value := objectMemory intAt: addr put: value. > self pop: 3 thenPush: valueOop "pop all; return value" > ! > > Item was changed: > ----- Method: InterpreterPrimitives>>primitiveObjectAtPut (in category > 'object access primitives') ----- > primitiveObjectAtPut > "Store a literal into a CompiledMethod at the given index. Defined > for CompiledMethods only." > | thisReceiver index newValue | > newValue := self stackValue: 0. > index := self stackValue: 1. > ((objectMemory isNonIntegerObject: index) > or: [index = ConstOne and: [(objectMemory isNonIntegerObject: > newValue)]]) ifTrue: > [^self primitiveFailFor: PrimErrBadArgument]. > index := objectMemory integerValueOf: index. > thisReceiver := self stackValue: 2. > self cppIf: IMMUTABILITY > + ifTrue: [ (objectMemory isImmutable: thisReceiver) ifTrue: > [ ^self primitiveFailFor: PrimErrNoModification ] ]. > - ifTrue: [ (objectMemory isImmutable: thisReceiver) ifTrue: > [ ^self primitiveFailFor: PrimErrInappropriate ] ]. > (index > 0 and: [index <= ((objectMemory literalCountOf: > thisReceiver) + LiteralStart)]) ifFalse: > [^self primitiveFailFor: PrimErrBadIndex]. > objectMemory storePointer: index - 1 ofObject: thisReceiver > withValue: newValue. > self pop: 3 thenPush: newValue! > > Item was changed: > ----- Method: InterpreterPrimitives>>primitiveShortAtPut (in category > 'sound primitives') ----- > primitiveShortAtPut > "Treat the receiver, which can be indexible by either bytes or > words, as an array > of signed 16-bit values. Set the contents of the given index to > the given value. > Note that the index specifies the i-th 16-bit entry, not the i-th > byte or word." > > | index rcvr value | > value := self stackTop. > index := self stackValue: 1. > ((objectMemory isIntegerObject: value) > and: [(objectMemory isIntegerObject: index) > and: [value := objectMemory integerValueOf: value. > (value >= -32768) and: [value <= 32767]]]) ifFalse: > [^self primitiveFailFor: PrimErrBadArgument]. > rcvr := self stackValue: 2. > (objectMemory isWordsOrBytes: rcvr) ifFalse: > [^self primitiveFailFor: PrimErrInappropriate]. > self cppIf: IMMUTABILITY "isWordsOrBytes ensure non immediate" > + ifTrue: [ (objectMemory isImmutable: rcvr) ifTrue: [ ^self > primitiveFailFor: PrimErrNoModification ] ]. > - ifTrue: [ (objectMemory isImmutable: rcvr) ifTrue: [ ^self > primitiveFailFor: PrimErrInappropriate ] ]. > index := objectMemory integerValueOf: index. > (index >= 1 and: [index <= (objectMemory num16BitUnitsOf: rcvr)]) > ifFalse: > [^self primitiveFailFor: PrimErrBadIndex]. > objectMemory storeShort16: index - 1 ofObject: rcvr withValue: > value. > self pop: 3 thenPush: (objectMemory integerObjectOf: value)! > > Item was changed: > ----- Method: SpurMemoryManager>>isOopValidBecome: (in category 'become > implementation') ----- > isOopValidBecome: oop > "Answers 0 if the oop can be become. > Answers an error code in the other case" > (self isImmediate: oop) ifTrue: [^PrimErrInappropriate]. > (self isPinned: oop) ifTrue: [^PrimErrObjectIsPinned]. > self > cppIf: IMMUTABILITY > + ifTrue: [ (self isImmutable: oop) ifTrue: > [^PrimErrNoModification] ]. > - ifTrue: [ (self isImmutable: oop) ifTrue: > [^PrimErrInappropriate] ]. > ^ 0! > > Item was changed: > ----- Method: StackInterpreter>>extendedStoreAndPopBytecode (in category > 'stack bytecodes') ----- > extendedStoreAndPopBytecode > <inline: true> > self extendedStoreBytecodePop: true > + "may not be reached (immutable receiver)"! > - ! > > Item was changed: > ----- Method: StackInterpreter>>extendedStoreBytecode (in category > 'stack bytecodes') ----- > extendedStoreBytecode > <inline: true> > + self extendedStoreBytecodePop: false > + "may not be reached (immutable receiver)"! > - self extendedStoreBytecodePop: false! > > Item was changed: > ----- Method: StackInterpreter>>storeAndPopReceiverVariableBytecode (in > category 'stack bytecodes') ----- > storeAndPopReceiverVariableBytecode > "Note: This code uses > storePointerUnchecked:ofObject:withValue: and does the > store check explicitely in order to help the translator > produce better code." > | rcvr top | > rcvr := self receiver. > top := self internalStackTop. > self internalPop: 1. > + self > + cCode: "Slang will inline currentBytecode to a constant so > this will work in C" > + [self fetchNextBytecode. > + objectMemory > + storePointerImmutabilityCheck: > (currentBytecode bitAnd: 7) > + ofObject: rcvr > + withValue: top] > + inSmalltalk: "But in Smalltalk we must use the > currentBytecode's value, not the next. > + We cant use the following code when generating C > code as slang > + won't inline currentBytecode correctly due to the > extra temp." > + [ | instVarIndex | > + instVarIndex := currentBytecode bitAnd: 7. > + self fetchNextBytecode. > + objectMemory > + storePointerImmutabilityCheck: instVarIndex > + ofObject: rcvr > + withValue: top]! > - self fetchNextBytecode. > - objectMemory storePointerImmutabilityCheck: (currentBytecode > bitAnd: 7) ofObject: rcvr withValue: top.! > > Item was changed: > ----- Method: StackInterpreterPrimitives>>primitiveInstVarAtPut (in > category 'object access primitives') ----- > primitiveInstVarAtPut > | newValue index rcvr hdr fmt totalLength fixedFields | > newValue := self stackTop. > index := self stackValue: 1. > rcvr := self stackValue: 2. > ((objectMemory isNonIntegerObject: index) > or: [argumentCount > 2 "e.g. object:instVarAt:put:" > and: [objectMemory isOopForwarded: rcvr]]) ifTrue: > [^self primitiveFailFor: PrimErrBadArgument]. > + (objectMemory isImmediate: rcvr) ifTrue: [ ^ self > primitiveFailFor: PrimErrInappropriate]. > + self > + cppIf: IMMUTABILITY > + ifTrue: [ (objectMemory isImmutable: rcvr) ifTrue: [ ^ > self primitiveFailFor: PrimErrNoModification] ]. > - self cppIf: IMMUTABILITY > - ifTrue: [ (objectMemory isOopImmutable: rcvr) ifTrue: > [^self primitiveFailFor: PrimErrInappropriate] ] > - ifFalse: [ (objectMemory isImmediate: rcvr) ifTrue: [^self > primitiveFailFor: PrimErrInappropriate] ]. > index := objectMemory integerValueOf: index. > hdr := objectMemory baseHeader: rcvr. > fmt := objectMemory formatOfHeader: hdr. > totalLength := objectMemory lengthOf: rcvr baseHeader: hdr format: > fmt. > fixedFields := objectMemory fixedFieldsOf: rcvr format: fmt > length: totalLength. > (index >= 1 and: [index <= fixedFields]) ifFalse: > [^self primitiveFailFor: PrimErrBadIndex]. > (fmt = objectMemory indexablePointersFormat > and: [objectMemory isContextHeader: hdr]) > ifTrue: [self externalInstVar: index - 1 ofContext: rcvr > put: newValue] > ifFalse: [self subscript: rcvr with: index storing: > newValue format: fmt]. > self pop: argumentCount + 1 thenPush: newValue! > >

3 4

Re: [squeak-dev] The Trunk: System-eem.803.mcz
by Tobias Pape 03 Mar '16

03 Mar '16

Hi all, Hi Eliot, Hi Clément Well, you know I'm the opponent here regarding the Immutability naming. I hate that I have to come back to the naming thing here once again. But it is rather important for me, since I work on immutability concepts that are unlike what immutability would mean here. I think it is also important to know that immutability in most other languages means "it won't change whatsoever" (including OCaml[1], Racket[2], Haskell[3], or object-oriented programming in general[4]) Now, I see the need, usefulness, and practical implementation for objects that cannot be written and hence throw an error, which _then_ can be circumvented/made be writable. I'm all for it and I like it. Except for the name. As Clément put it: "As argued on the virtual machine mailing list, we are talking about a write-barrier more than immutability itself." I would hence propose to name those objects locked objects with the accompanying Someone-tries-to-write-that-Error named ObjectIsLockedError and the state-change messages Object>>lock Object>>unlock and likewise the VM-information SmalltalkImage>>supportsLockedObjects I know, re-iterating this is tedious, but it will avoid unmet expectations by newcomers from other languages as well as name clashes with actual immutability implementations (how should we name those, then)? With apologies -Tobias [1]: http://typeocaml.com/2015/01/02/immutable/ [2]: a) http://docs.racket-lang.org/reference/pairs.html "Pairs are not mutable" b) http://docs.racket-lang.org/reference/strings.html [3]: https://wiki.haskell.org/A_brief_introduction_to_Haskell#Immutable_declarat… [4]: https://en.wikipedia.org/wiki/Immutable_object On 03.03.2016, at 03:19, commits(a)source.squeak.org wrote: > Eliot Miranda uploaded a new version of System to project The Trunk: > http://source.squeak.org/trunk/System-eem.803.mcz > > ==================== Summary ==================== > > Name: System-eem.803 > Author: eem > Time: 2 March 2016, 7:18:21.713694 pm > UUID: 934ce776-0717-469b-8818-300954393791 > Ancestors: System-bf.802 > > Add getters that answer whether the VM supports immutability or multiple bytecode sets. > > =============== Diff against System-bf.802 =============== > > Item was added: > + ----- Method: SmalltalkImage>>supportsImmutability (in category 'system attributes') ----- > + supportsImmutability > + "Answer whether the VM observes the per-object immutability flag and consequently > + aborts writes to inst vars of, and fails primitives that attempt to write, to immutable objects." > + "SmalltalkImage current supportsImmutability" > + > + ^(self vmParameterAt: 65) > + ifNil: [false] > + ifNotNil: > + [:param| "In older VMs this is a boolean reflecting MULTIPLE_BYTECODE_SETS" > + param isInteger "In newer VMs it is a set of integer flags, bit 1 of which is IMMUTABILITY" > + ifTrue: [param anyMask: 2] > + ifFalse: [false]]! > > Item was added: > + ----- Method: SmalltalkImage>>supportsMultipleBytecodeSets (in category 'system attributes') ----- > + supportsMultipleBytecodeSets > + "Answer whether the VM supports multiple bytecodeSets." > + "SmalltalkImage current supportsMultipleBytecodeSets" > + > + ^(self vmParameterAt: 65) > + ifNil: [false] > + ifNotNil: > + [:param| "In older VMs this is a boolean reflecting MULTIPLE_BYTECODE_SETS" > + param isInteger "In newer VMs it is a set of integer flags, bit 0 of which is MULTIPLE_BYTECODE_SETS" > + ifTrue: [param anyMask: 1] > + ifFalse: [param]]! > >

4 4

VM Recipe: How to build an interpreter VM on Linux
by David T. Lewis 03 Mar '16

03 Mar '16

I put a copy of the recipe on the wiki: http://wiki.squeak.org/squeak/6354 This is the quick-start for compiling a context interpreter VM suitable for running most images up to but not including Spur. For a VM capable of running earlier images, see the SqueakJS VM at try.squeak.org. Dave

1 0

VM Maker: VMMaker.oscog-eem.1705.mcz
by commits＠source.squeak.org 03 Mar '16

03 Mar '16

Eliot Miranda uploaded a new version of VMMaker to project VM Maker: http://source.squeak.org/VMMaker/VMMaker.oscog-eem.1705.mcz ==================== Summary ==================== Name: VMMaker.oscog-eem.1705 Author: eem Time: 2 March 2016, 7:08:07.484944 pm UUID: 47d289d4-7673-412f-95a2-c2633e2b0181 Ancestors: VMMaker.oscog-eem.1704 Fix immutability suport in the Cogit by making sure that the Slang output for the bytecode generatorTable marks inst var store bytecode generators as mapped if IMMUTABILITY is set. Change vm parameter 65 to be a set of flags, not merely the boolean for MULTIPLE_BYTECODE_SETS, and make the flags answer MULTIPLE_BYTECODE_SETS & IMMUTABILITY. Add a PC map printer to the Cogit (used to debug the above issue). Make the current image facade support more of the printing api so that the map printer can be tested in-image. Nuke an obsolete method. =============== Diff against VMMaker.oscog-eem.1704 =============== Item was changed: ----- Method: CogBytecodeDescriptor>>isMapped (in category 'accessing') ----- isMapped + "Answer the value of isMapped. Note we send value because this may be the binding for #IMMUTABILITY" + ^isMapped value! - "Answer the value of isMapped" - - ^isMapped! Item was changed: ----- Method: CogBytecodeDescriptor>>printCInitializerOn:in: (in category 'translation') ----- printCInitializerOn: aStream in: aCCodeGenerator <doNotGenerate> | first | first := true. aStream nextPut: ${; space. self class instVarNamesAndTypesForTranslationDo: [:ivn :type| | value | first ifTrue: [first := false] ifFalse: [aStream nextPut: $,; space]. value := self instVarNamed: ivn. + value isVariableBinding + ifTrue: [aStream nextPutAll: value key] + ifFalse: + [(#(#'unsigned char' #'signed char' #('unsigned' ' : 1')) includes: type) + ifTrue: [value isInteger + ifTrue: [ivn = 'opcode' + ifTrue: [aStream nextPutAll: (CogRTLOpcodes nameForOpcode: value)] + ifFalse: [aStream print: value]] + ifFalse: [aStream nextPut: ((value notNil and: [value]) ifTrue: [$1] ifFalse: [$0])]] + ifFalse: [(false and: [#('spanFunction' 'isBackwardBranchFunction') includes: ivn]) ifTrue: + [aStream nextPut: $(; nextPutAll: type first; nextPutAll: type last; nextPut: $)]. + aStream nextPutAll: (value + ifNotNil: [aCCodeGenerator cFunctionNameFor: value] + ifNil: ['0'])]]]. - (#(#'unsigned char' #'signed char' #('unsigned' ' : 1')) includes: type) - ifTrue: [value isInteger - ifTrue: [ivn = 'opcode' - ifTrue: [aStream nextPutAll: (CogRTLOpcodes nameForOpcode: value)] - ifFalse: [aStream print: value]] - ifFalse: [aStream nextPut: ((value notNil and: [value]) ifTrue: [$1] ifFalse: [$0])]] - ifFalse: [(false and: [#('spanFunction' 'isBackwardBranchFunction') includes: ivn]) ifTrue: - [aStream nextPut: $(; nextPutAll: type first; nextPutAll: type last; nextPut: $)]. - aStream nextPutAll: (value - ifNotNil: [aCCodeGenerator cFunctionNameFor: value] - ifNil: ['0'])]]. aStream space; nextPut: $}! Item was removed: - ----- Method: Cogit class>>bytecodeTable (in category 'translation') ----- - bytecodeTable - | selectors | - selectors := Set new. - generatorTable object do: - [:bytecodeDescriptor| - selectors add: bytecodeDescriptor generator. - bytecodeDescriptor spanFunction ifNotNil: - [selectors add: bytecodeDescriptor spanFunction]]. - ^selectors! Item was changed: ----- Method: Cogit class>>generatorTableFrom: (in category 'class initialization') ----- generatorTableFrom: anArray | blockCreationBytecodeSize | generatorTable := CArrayAccessor on: (Array new: 256). anArray do: [:tuple| | descriptor | (descriptor := CogBytecodeDescriptor new) numBytes: tuple first; generator: tuple fourth; isReturn: (tuple includes: #return); + isMapped: ((tuple includes: #isMappedIfImmutability) + ifTrue: [self bindingOf: #IMMUTABILITY] + ifFalse: [tuple includes: #isMapped]); - isMapped: ((tuple includes: #isMapped) or: [IMMUTABILITY and: [ tuple includes: #isMappedIfImmutability]]); isMappedInBlock: (tuple includes: #isMappedInBlock); isBlockCreation: (tuple includes: #block); spanFunction: (((tuple includes: #block) or: [(tuple includes: #branch)]) ifTrue: [tuple detect: [:thing| thing isSymbol and: [thing numArgs = 4]]]); isBranchTrue: (tuple includes: #isBranchTrue); isBranchFalse: (tuple includes: #isBranchFalse); isExtension: (tuple includes: #extension); isInstVarRef: (tuple includes: #isInstVarRef); "for Spur" hasIRC: (tuple includes: #hasIRC); "for Newspeak" yourself. "As a hack to cut down on descriptor flags, use opcode to tag unusedBytecode for scanning. Currently descriptors are exactly 16 bytes with all 8 flag bits used (in Newspeak at least 17 bytes, 9 flag bits). As another hack to eliminate a test in scanMethod mark unknowns as extensions." descriptor generator == #unknownBytecode ifTrue: [descriptor opcode: Nop; isExtension: true]. descriptor isBlockCreation ifTrue: [blockCreationBytecodeSize ifNil: [blockCreationBytecodeSize := descriptor numBytes] ifNotNil: [self assert: blockCreationBytecodeSize = descriptor numBytes]]. tuple do: [:thing| thing isSymbol ifTrue: [(thing beginsWith: #needsFrame) ifTrue: [descriptor needsFrameFunction: thing]. (CogRTLOpcodes classPool at: thing ifAbsent: []) ifNotNil: [:opcode| descriptor opcode: opcode]]]. tuple last isInteger ifTrue: [descriptor stackDelta: tuple last] ifFalse: [descriptor needsFrameFunction ifNotNil: [self error: 'frameless block bytecodes must specify a stack delta']]. tuple second to: tuple third do: [:index| generatorTable at: index put: descriptor]]. BlockCreationBytecodeSize := blockCreationBytecodeSize. ^generatorTable! Item was added: + ----- Method: Cogit>>printPCMapPairsFor: (in category 'method map') ----- + printPCMapPairsFor: cogMethod + <api> + <var: 'cogMethod' type: #'CogMethod *'> + | mcpc map mapByte annotation value | + mcpc := cogMethod asInteger + cmNoCheckEntryOffset. + map := self mapStartFor: cogMethod. + [(mapByte := objectMemory byteAt: map) ~= MapEnd] whileTrue: + [annotation := mapByte >> AnnotationShift. + annotation = IsAnnotationExtension + ifTrue: + [value := (mapByte bitAnd: DisplacementMask) + IsSendCall] + ifFalse: + [value := annotation. + mcpc := mcpc + (backEnd codeGranularity + * (annotation = IsDisplacementX2N + ifTrue: [mapByte - DisplacementX2N << AnnotationShift] + ifFalse: [mapByte bitAnd: DisplacementMask]))]. + coInterpreter + printHexnp: map; + print: ': '. + self + cCode: [self print: '%02x' f: mapByte] + inSmalltalk: + [mapByte < 16 ifTrue: + [coInterpreter putchar: $0]. + coInterpreter printHexnp: mapByte]. + coInterpreter + printChar: $ ; + printNum: annotation; + print: ' ('. + self cppIf: NewspeakVM + ifTrue: + [value + caseOf: { + [IsDisplacementX2N] -> [coInterpreter print: 'IsDisplacementX2N']. + [IsAnnotationExtension] -> [coInterpreter print: 'IsAnnotationExtension']. + [IsObjectReference] -> [coInterpreter print: 'IsObjectReference']. + [IsAbsPCReference] -> [coInterpreter print: 'IsAbsPCReference']. + [HasBytecodePC] -> [coInterpreter print: 'HasBytecodePC']. + [IsRelativeCall] -> [coInterpreter print: 'IsRelativeCall']. + [IsNSSendCall] -> [coInterpreter print: 'IsNSSendCall']. + [IsSendCall] -> [coInterpreter print: 'IsSendCall']. + [IsSuperSend] -> [coInterpreter print: 'IsSuperSend']. + [IsDirectedSuperSend] -> [coInterpreter print: 'IsDirectedSuperSend']. + [IsNSSelfSend] -> [coInterpreter print: 'IsNSSelfSend']. + [IsNSDynamicSuperSend] -> [coInterpreter print: 'IsNSDynamicSuperSend']. + [IsNSImplicitReceiverSend] -> [coInterpreter print: 'IsNSImplicitReceiverSend'] } + otherwise: [coInterpreter print: '??? '; printHexnp: value]] + ifFalse: + [value + caseOf: { + [IsDisplacementX2N] -> [coInterpreter print: 'IsDisplacementX2N']. + [IsAnnotationExtension] -> [coInterpreter print: 'IsAnnotationExtension']. + [IsObjectReference] -> [coInterpreter print: 'IsObjectReference']. + [IsAbsPCReference] -> [coInterpreter print: 'IsAbsPCReference']. + [HasBytecodePC] -> [coInterpreter print: 'HasBytecodePC']. + [IsRelativeCall] -> [coInterpreter print: 'IsRelativeCall']. + [IsSendCall] -> [coInterpreter print: 'IsSendCall']. + [IsSuperSend] -> [coInterpreter print: 'IsSuperSend']. + [IsDirectedSuperSend] -> [coInterpreter print: 'IsDirectedSuperSend'] } + otherwise: [coInterpreter print: '??? '; printHexnp: value]]. + coInterpreter + print: ') '; + printHexnp: (mapByte bitAnd: DisplacementMask); + printChar: $ ; + putchar: $@; + printHex: mcpc; + cr; + flush. + map := map - 1]! Item was changed: ----- Method: Cogit>>printPCMapPairsFor:on: (in category 'method map') ----- printPCMapPairsFor: cogMethod on: aStream <doNotGenerate> <inline: true> | mcpc map mapByte annotation | mcpc := cogMethod asInteger + cmNoCheckEntryOffset. map := self mapStartFor: cogMethod. [(mapByte := objectMemory byteAt: map) ~= MapEnd] whileTrue: [annotation := mapByte >> AnnotationShift. annotation ~= IsAnnotationExtension ifTrue: [mcpc := mcpc + (backEnd codeGranularity * (annotation = IsDisplacementX2N ifTrue: [mapByte - DisplacementX2N << AnnotationShift] ifFalse: [mapByte bitAnd: DisplacementMask]))]. aStream ensureCr. map printOn: aStream base: 16. aStream nextPutAll: ': '. + mapByte printOn: aStream base: 16 length: 2 padded: true. - mapByte printOn: aStream base: 16. aStream space. annotation printOn: aStream base: 16. aStream nextPutAll: ' ('; print: (AnnotationConstantNames at: annotation + 1); nextPutAll: ') '. (mapByte bitAnd: DisplacementMask) printOn: aStream base: 16. aStream space. aStream nextPut: $@. mcpc printOn: aStream base: 16. aStream flush. map := map - 1]! Item was added: + ----- Method: CurrentImageCoInterpreterFacade>>flush (in category 'printing') ----- + flush + coInterpreter transcript flush! Item was added: + ----- Method: CurrentImageCoInterpreterFacade>>printChar: (in category 'printing') ----- + printChar: aCharacter + coInterpreter transcript nextPut: aCharacter! Item was changed: ----- Method: CurrentImageCoInterpreterFacade>>printHex: (in category 'printing') ----- printHex: anInteger | it16 | it16 := anInteger radix: 16. coInterpreter transcript next: 8 - it16 size put: Character space; + nextPutAll: (anInteger printStringBase: 16)! - nextPutAll: (anInteger storeStringBase: 16)! Item was added: + ----- Method: CurrentImageCoInterpreterFacade>>printHexnp: (in category 'printing') ----- + printHexnp: anInteger + coInterpreter transcript nextPutAll: (anInteger printStringBase: 16)! Item was added: + ----- Method: CurrentImageCoInterpreterFacade>>printNum: (in category 'printing') ----- + printNum: anInteger + coInterpreter transcript print: anInteger! Item was added: + ----- Method: CurrentImageCoInterpreterFacade>>putchar: (in category 'printing') ----- + putchar: aCharacter + coInterpreter transcript nextPut: aCharacter! Item was changed: ----- Method: CurrentImageCoInterpreterFacade>>space (in category 'printing') ----- space + coInterpreter transcript space! - Transcript space! Item was changed: ----- Method: StackInterpreter>>flush (in category 'debug printing') ----- flush + <api> <cmacro: '() fflush(stdout)'>! Item was added: + ----- Method: StackInterpreter>>getCogVMFeatureFlags (in category 'internal interpreter access') ----- + getCogVMFeatureFlags + "Answer an array of flags indicating various optional features of the Cog VM. + Bit 0: supports two btecode sets (MULTIPLEBYTECODESETS) + Bit 1: supports immtablity (IMMUTABILITY)" + ^objectMemory integerObjectOf: (MULTIPLEBYTECODESETS ifTrue: [1] ifFalse: [0]) + + (IMMUTABILITY ifTrue: [2] ifFalse: [0])! Item was changed: ----- Method: StackInterpreter>>printHexnp: (in category 'debug printing') ----- printHexnp: n + <api> "Print n in hex, in the form '0x1234', unpadded" self print: '0x%lx' f: (self cCoerceSimple: n to: #'unsigned long')! Item was changed: ----- Method: StackInterpreterPrimitives>>primitiveVMParameter (in category 'system control primitives') ----- (excessive size, no diff calculated)

1 0

The four Andreas Tests for Interruptability
by David T. Lewis 02 Mar '16

02 Mar '16

This is a follow up from off line discussion with Eliot. In order to verify the operation of the low space interrupt handling mechanism, Andreas proposed the following four tests: * "[true] whileTrue" * "[[true] whileTrue] forkAt: Processor userSchedulingPriority + 1" * "Smalltalk createStackOverflow" * "[Smalltalk createStackOverflow] forkAt: Processor userSchedulingPriority + 1" Each of these should be interruptable, and the mechanism should work in both Morphic and MVC. When running on a virtual memory system with swapping (e.g. Linux), the VM should be started with a specific memory size, otherwise the system will bog down in swapping without ever encountering the low space limit. When we make a change to process scheduling, we would want to make sure that these four tests still work (which might in turn require changing the low space interrupt handling). Background is on Mantis 0001041: [VM][ENH] LowSpaceAndInterruptHandler-dtl at http://bugs.squeak.org/view.php?id=1041, and Andreas' original discussion would be on squeak-dev somewhere in the early 2005 time frame. Dave

1 0

Call for Papers: 11th Workshop on Implementation, Compilation, Optimization of OO Languages, Programs and Systems
by Stefan Marr 02 Mar '16

02 Mar '16

Call for Papers: ICOOOLPS’16 ============================ 11th Workshop on Implementation, Compilation, Optimization of OO Languages, Programs and Systems Co-located with ECOOP July 18, 2016, Rome, Italy URL: http://2016.ecoop.org/track/ICOOOLPS-2016 Twitter: @ICOOOLPS The ICOOOLPS workshop series brings together researchers and practitioners working in the field of language implementation and optimization. The goal of the workshop is to discuss emerging problems and research directions as well as new solutions to classic performance challenges. The topics of interest for the workshop include techniques for the implementation and optimization of a wide range of languages including but not limited to object-oriented ones. Furthermore, meta-compilation techniques or language-agnostic approaches are welcome, too. A non-exclusive list of topics follows: - implementation and optimization of fundamental languages features (from automatic memory management to zero-overhead metaprogramming) - runtime systems technology (libraries, virtual machines) - static, adaptive, and speculative optimizations and compiler techniques - meta-compilation techniques and language-agnostic approaches for the efficient implementation of languages - compilers (intermediate representations, offline and online optimizations,...) - empirical studies on language usage, benchmark design, and benchmarking methodology - resource-sensitive systems (real-time, low power, mobile, cloud) - studies on design choices and tradeoffs (dynamic vs. static compilation, heuristics vs. programmer input,...) - tooling support, debuggability and observability of languages as well as their implementations ### Workshop Format and Submissions This workshop welcomes the presentation and discussion of new ideas and emerging problems that give a chance for interaction and exchange. More mature work is welcome as part of a mini-conference format, too. We aim to interleave interactive brainstorming and demonstration sessions between the formal presentations to foster an active exchange of ideas. The workshop papers will be published either in the ACM DL or in the Dagstuhl LIPIcs ECOOP Workshop proceedings. Until further notice, please use the ACM SIGPLAN template with a 10pt font size: http://www.sigplan.org/Resources/Author/ - position and work-in-progress paper: 1-4 pages - technical paper: max. 10 pages - demos and posters: 1-page abstract For the submission, please use the HotCRP system: http://ssw.jku.at/icooolps/ ### Important Dates - abstract submission: April 11, 2016 - paper submission: April 15, 2016 - notification: May 13, 2016 - all deadlines: Anywhere on Earth (AoE), i.e., GMT/UTC−12:00 hour - workshop: July 18th, 2016 ### Program Committee Edd Barrett, King’s College London, UK Clement Bera, Inria Lille, France Maxime Chevalier-Boisvert, Université de Montréal, Canada Tim Felgentreff, Hasso Plattner Institute, Germany Roland Ducournau, LIRMM, Université de Montpellier, France Elisa Gonzalez Boix, Vrije Universiteit Brussel, Belgium David Gregg, Trinity College Dublin, Ireland Matthias Grimmer, Johannes Kepler University Linz, Austria Michael Haupt, Oracle, Germany Richard Jones, University of Kent, UK Tomas Kalibera, Northeastern University, USA Hidehiko Masuhara, Tokyo Institute of Technology, Japan Tiark Rompf, Purdue University, USA Jennifer B. Sartor, Ghent University, Belgium Sam Tobin-Hochstadt, Indiana University, USA ### Workshop Organizers Stefan Marr, Johannes Kepler University Linz, Austria Eric Jul, University of Oslo, Norway For questions or concerns, please mail to stefan.marr at jku.at or contact us via https://twitter.com/icooolps.

1 0

VM Maker: VMMaker.oscog-eem.1704.mcz
by commits＠source.squeak.org 02 Mar '16

02 Mar '16

Eliot Miranda uploaded a new version of VMMaker to project VM Maker: http://source.squeak.org/VMMaker/VMMaker.oscog-eem.1704.mcz ==================== Summary ==================== Name: VMMaker.oscog-eem.1704 Author: eem Time: 1 March 2016, 6:04:07.799253 pm UUID: 21fb8ca6-b62c-4c06-8144-bd8be0832a8d Ancestors: VMMaker.oscog-eem.1703 Update primitiveFileRead for Spur. It doesn't need the PrimErrObjectMayMove handling sicne Spur has pinning. Allow the CurrentImageCoInterpreterFacade to print Cog methods (useful for looking at the methods generated during the pc mapping test). Modify the facade's management of the headerToMethodMap so that methods are not flushed on each jit compile and asserts don't fail (compare method headers not methods). =============== Diff against VMMaker.oscog-eem.1703 =============== Item was added: + ----- Method: CoInterpreter>>setUpForUseByFacade: (in category 'debug support') ----- + setUpForUseByFacade: aCurrentImageCoInterpreterFacade + "Set up variables with default values so that other initializations work. + numStackPages needs to be initialized so that interpreterAllocationReserveBytes + can be computed." + numStackPages := 0! Item was changed: ----- Method: Cogit class>>cog:selector:options: (in category 'in-image compilation') ----- cog: aCompiledMethod selector: aSelector options: optionsDictionaryOrArray "StackToRegisterMappingCogit cog: (Integer >> #benchFib) selector: #benchFib options: #(COGMTVM false)" + | cogit coInterpreter | + cogit := self instanceForTests: optionsDictionaryOrArray. + coInterpreter := CurrentImageCoInterpreterFacade forCogit: cogit. - | initOptions coInterpreter cogit | - initOptions := self asOptionsDictionary: optionsDictionaryOrArray. - CoInterpreter initializeWithOptions: initOptions. - CoInterpreter objectMemoryClass initializeWithOptions: initOptions. - self initializeWithOptions: initOptions. - coInterpreter := CurrentImageCoInterpreterFacade forCogit: (cogit := self new). [cogit setInterpreter: coInterpreter; singleStep: true; initializeCodeZoneFrom: 1024 upTo: coInterpreter memory size / 2. "leave space for rump C stack" cogit methodZone freeStart: (cogit methodZone freeStart roundUpTo: 1024)] on: Notification do: [:ex| (ex messageText beginsWith: 'cannot find receiver for') ifTrue: [ex resume: coInterpreter]. ex pass]. ^{ coInterpreter. cogit. cogit cog: (coInterpreter oopForObject: aCompiledMethod) selector: (coInterpreter oopForObject: aSelector) }! Item was changed: ----- Method: Cogit class>>genAndDisPICoptions: (in category 'in-image compilation') ----- genAndDisPICoptions: optionsDictionaryOrArray "StackToRegisterMappingCogit genAndDisPICoptions: #(ISA ARMv5 CogCompilerClass CogInLineLiteralsARMCompiler)" + | cogit coInterpreter | + cogit := self instanceForTests: optionsDictionaryOrArray. + coInterpreter := CurrentImageCoInterpreterFacade forCogit: cogit. - | coInterpreter cogit | - self initializeWithOptions: (self asOptionsDictionary: optionsDictionaryOrArray). - CoInterpreter initializeWithOptions: initializationOptions. - CoInterpreter objectMemoryClass initializeWithOptions: initializationOptions. - coInterpreter := CurrentImageCoInterpreterFacade forCogit: (cogit := self new). [cogit setInterpreter: coInterpreter; singleStep: true; initializeCodeZoneFrom: 1024 upTo: coInterpreter memory size / 2. "leave space for rump C stack" cogit methodZone freeStart: (cogit methodZone freeStart roundUpTo: 1024)] on: Notification do: [:ex| (ex messageText beginsWith: 'cannot find receiver for') ifTrue: [ex resume: coInterpreter]]. cogit disassembleFrom: cogit cPICPrototype + (cogit sizeof: CogMethod) to: cogit cPICPrototype + cogit closedPICSize! Item was added: + ----- Method: Cogit class>>instanceForTests: (in category 'in-image compilation') ----- + instanceForTests: optionsDictionaryOrArray + "Initialize all the relevant classes from the options and answer a new instance of me." + | initOptions | + initOptions := self asOptionsDictionary: optionsDictionaryOrArray. + CoInterpreter initializeWithOptions: initOptions. + CoInterpreter objectMemoryClass initializeWithOptions: initOptions. + self initializeWithOptions: initOptions. + ^self new! Item was changed: ----- Method: Cogit class>>testPCMappingSelect:options: (in category 'tests') ----- testPCMappingSelect: aBlock options: optionsDictionaryOrArray "Test pc mapping both ways using a selection of the methods in the current image." | cogit coInterpreter | + cogit := self instanceForTests: optionsDictionaryOrArray. - self initializeWithOptions: (self asOptionsDictionary: optionsDictionaryOrArray). - cogit := self new. coInterpreter := CurrentImageCoInterpreterFacade forCogit: cogit. [cogit setInterpreter: coInterpreter; singleStep: true; initializeCodeZoneFrom: 1024 upTo: coInterpreter memory size] on: Notification do: [:ex| (ex messageText beginsWith: 'cannot find receiver for') ifTrue: [ex resume: coInterpreter]]. SystemNavigation new allSelect: [:m| | cm | (m isQuick not and: [aBlock value: m]) ifTrue: + [Transcript nextPut: $.; flush. - [coInterpreter voidHeaderToMethodMap. - Transcript nextPut: $.; flush. [coInterpreter. cm := cogit cog: (coInterpreter oopForObject: m) selector: (coInterpreter oopForObject: m selector). cm isNil and: [coInterpreter isCogCompiledCodeCompactionCalledFor]] whileTrue: [cogit methodZone clearCogCompiledCode. coInterpreter clearCogCompiledCodeCompactionCalledFor. coInterpreter initializeObjectMap]. cogit testPCMappingForCompiledMethod: m cogMethod: cm]. false]! Item was changed: CogClass subclass: #CurrentImageCoInterpreterFacade instanceVariableNames: 'memory cogit coInterpreter objectMemory objectMap headerToMethodMap cachedObject cachedOop variables' classVariableNames: '' + poolDictionaries: 'CogMethodConstants VMBasicConstants VMObjectIndices VMSqueakClassIndices' - poolDictionaries: 'VMBasicConstants VMObjectIndices VMSqueakClassIndices' category: 'VMMaker-Support'! !CurrentImageCoInterpreterFacade commentStamp: 'eem 8/6/2014 14:59' prior: 0! A CurrentImageCoInterpreterFacade is a stand-in for an object memory (ObjectMemory, SpurMemoryManager, etc) that allows the Cogits to access image objects as if they were in the simulator VM's heap. hence it allows the Cogits to generate code for methdos in the current image, for testing, etc. Instance Variables cachedObject: <Object> cachedOop: <Integer> coInterpreter: <CoInterpreter> cogit: <Cogit> headerToMethodMap: <Dictionary> memory: <ByteArray> objectMap: <IdentityDictionary> objectMemory: <NewObjectMemory|SpurMemoryManager> variables: <Dictionary> cachedObject - the object matching cachedOop, to speed-up oop to obejct mapping cachedOop - the last used oop coInterpreter - the CoInterpreter simulator used by the cogit. cogit - the code egnerator in use headerToMethodMap - a map from header to CompiledMethod memory - a rump memory for holding various interpreter variables (e.g. stackLimit) that are accessed as memory locations by generated code objectMap - map from objects to their oops objectMemory - the object memory used to encode various values, answer queries, etc variables - a map from the names of variables to their addresses in memory ! Item was changed: ----- Method: CurrentImageCoInterpreterFacade>>cogit: (in category 'initialize-release') ----- cogit: aCogit cogit := aCogit. cogit objectMemory ifNil: [cogit instVarNamed: 'objectMemory' put: objectMemory]. coInterpreter cogit: aCogit. (objectMemory respondsTo: #cogit:) ifTrue: [objectMemory cogit: aCogit]. (objectMemory respondsTo: #coInterpreter:) ifTrue: [objectMemory coInterpreter: coInterpreter]. + coInterpreter setUpForUseByFacade: self. + objectMemory setUpForUseByFacade: self. #('stackLimit') do: [:l| self addressForLabel: l]. self initializeObjectMap! Item was removed: - ----- Method: CurrentImageCoInterpreterFacade>>initialPCForHeader:method: (in category 'accessing') ----- - initialPCForHeader: methodHeaderOop method: aMethodOop - self assert: (self objectForOop: aMethodOop) == (headerToMethodMap at: methodHeaderOop). - ^{ self objectForOop: aMethodOop. (self objectForOop: aMethodOop) initialPC }! Item was added: + ----- Method: CurrentImageCoInterpreterFacade>>printCogMethod: (in category 'printing') ----- + printCogMethod: cogMethod + | address primitive | + address := cogMethod asInteger. + self printHex: address; + print: ' <-> '; + printHex: address + cogMethod blockSize. + cogMethod cmType = CMMethod ifTrue: + [self print: ': method: '; + printHex: cogMethod methodObject. + primitive := self primitiveIndexOfMethod: cogMethod methodObject + header: cogMethod methodHeader. + primitive ~= 0 ifTrue: + [self print: ' prim '; printNum: primitive]]. + cogMethod cmType = CMBlock ifTrue: + [self print: ': block home: '; + printHex: (self cCoerceSimple: cogMethod to: #'CogBlockMethod *') cmHomeMethod asUnsignedInteger]. + cogMethod cmType = CMClosedPIC ifTrue: + [self print: ': Closed PIC N: '; + printHex: cogMethod cPICNumCases]. + cogMethod cmType = CMOpenPIC ifTrue: + [self print: ': Open PIC ']. + self print: ' selector: '; printHex: cogMethod selector. + cogMethod selector = objectMemory nilObject + ifTrue: [self print: ' (nil)'] + ifFalse: [self space; printStringOf: cogMethod selector]. + self cr! Item was added: + ----- Method: CurrentImageCoInterpreterFacade>>printStringOf: (in category 'printing') ----- + printStringOf: anOop + Transcript nextPutAll: (self objectForOop: anOop)! Item was changed: ----- Method: CurrentImageCoInterpreterFacade>>rawHeaderOf: (in category 'accessing') ----- rawHeaderOf: aMethodOop | method headerOop | method := self objectForOop: aMethodOop. headerOop := objectMemory integerObjectOf: (self objectForOop: aMethodOop) header. + self assert: method header = (headerToMethodMap at: headerOop ifAbsentPut: [method]) header. - self assert: method = (headerToMethodMap at: headerOop ifAbsentPut: [method]). ^headerOop! Item was added: + ----- Method: CurrentImageCoInterpreterFacade>>space (in category 'printing') ----- + space + Transcript space! Item was removed: - ----- Method: CurrentImageCoInterpreterFacade>>voidHeaderToMethodMap (in category 'accessing') ----- - voidHeaderToMethodMap - headerToMethodMap := Dictionary new! Item was changed: ----- Method: FilePlugin>>primitiveFileRead (in category 'file primitives') ----- primitiveFileRead <export: true> + self cppIf: SPURVM + ifTrue: [self primitiveFileReadWithPinning] + ifFalse: [self primitiveFileReadWithoutPinning]! - | retryCount count startIndex array file elementSize bytesRead | - <var: 'file' type: #'SQFile *'> - <var: 'count' type: #'size_t'> - <var: 'startIndex' type: #'size_t'> - <var: 'elementSize' type: #'size_t'> - - retryCount := 0. - count := interpreterProxy positive32BitValueOf: (interpreterProxy stackValue: 0). - startIndex := interpreterProxy positive32BitValueOf: (interpreterProxy stackValue: 1). - - [array := interpreterProxy stackValue: 2. - file := self fileValueOf: (interpreterProxy stackValue: 3). - - (interpreterProxy failed - "buffer can be any indexable words or bytes object except CompiledMethod" - or: [(interpreterProxy isWordsOrBytes: array) not]) ifTrue: - [^interpreterProxy primitiveFailFor: PrimErrBadArgument]. - - elementSize := (interpreterProxy isWords: array) ifTrue: [4] ifFalse: [1]. - (startIndex >= 1 - and: [(startIndex + count - 1) <= (interpreterProxy slotSizeOf: array)]) ifFalse: - [^interpreterProxy primitiveFailFor: PrimErrBadIndex]. - - "Note: adjust startIndex for zero-origin indexing" - bytesRead := self - sqFile: file - Read: count * elementSize - Into: (interpreterProxy cCoerce: (interpreterProxy firstIndexableField: array) to: #'char *') - At: (startIndex - 1) * elementSize. - interpreterProxy primitiveFailureCode = PrimErrObjectMayMove - and: [(retryCount := retryCount + 1) <= 2] "Two objects, the file and the array can move"] whileTrue: - [interpreterProxy - tenuringIncrementalGC; - primitiveFailFor: PrimNoErr]. - interpreterProxy failed ifFalse: - [interpreterProxy - pop: 5 "pop rcvr, file, array, startIndex, count" - thenPush:(interpreterProxy integerObjectOf: bytesRead // elementSize) "push # of elements read"]! Item was added: + ----- Method: FilePlugin>>primitiveFileReadWithPinning (in category 'file primitives') ----- + primitiveFileReadWithPinning + "This version of primitiveFileRead is for garbage collectors that support pinning." + | count startIndex array file elementSize bytesRead | + <inline: true> + <var: 'file' type: #'SQFile *'> + <var: 'count' type: #'size_t'> + <var: 'startIndex' type: #'size_t'> + <var: 'elementSize' type: #'size_t'> + count := interpreterProxy positive32BitValueOf: (interpreterProxy stackValue: 0). + startIndex := interpreterProxy positive32BitValueOf: (interpreterProxy stackValue: 1). + array := interpreterProxy stackValue: 2. + file := self fileValueOf: (interpreterProxy stackValue: 3). + + (interpreterProxy failed + "buffer can be any indexable words or bytes object except CompiledMethod" + or: [(interpreterProxy isWordsOrBytes: array) not]) ifTrue: + [^interpreterProxy primitiveFailFor: PrimErrBadArgument]. + + elementSize := (interpreterProxy isWords: array) ifTrue: [4] ifFalse: [1]. + (startIndex >= 1 + and: [(startIndex + count - 1) <= (interpreterProxy slotSizeOf: array)]) ifFalse: + [^interpreterProxy primitiveFailFor: PrimErrBadIndex]. + + "Note: adjust startIndex for zero-origin indexing" + bytesRead := self + sqFile: file + Read: count * elementSize + Into: (interpreterProxy cCoerce: (interpreterProxy firstIndexableField: array) to: #'char *') + At: (startIndex - 1) * elementSize. + interpreterProxy failed ifFalse: + [interpreterProxy + pop: 5 "pop rcvr, file, array, startIndex, count" + thenPush:(interpreterProxy integerObjectOf: bytesRead // elementSize) "push # of elements read"]! Item was added: + ----- Method: FilePlugin>>primitiveFileReadWithoutPinning (in category 'file primitives') ----- + primitiveFileReadWithoutPinning + "This version of primitiveFileRead is for garbage collectors without support for pinning." + | retryCount count startIndex array file elementSize bytesRead | + <inline: true> + <var: 'file' type: #'SQFile *'> + <var: 'count' type: #'size_t'> + <var: 'startIndex' type: #'size_t'> + <var: 'elementSize' type: #'size_t'> + retryCount := 0. + count := interpreterProxy positive32BitValueOf: (interpreterProxy stackValue: 0). + startIndex := interpreterProxy positive32BitValueOf: (interpreterProxy stackValue: 1). + + [array := interpreterProxy stackValue: 2. + file := self fileValueOf: (interpreterProxy stackValue: 3). + + (interpreterProxy failed + "buffer can be any indexable words or bytes object except CompiledMethod" + or: [(interpreterProxy isWordsOrBytes: array) not]) ifTrue: + [^interpreterProxy primitiveFailFor: PrimErrBadArgument]. + + elementSize := (interpreterProxy isWords: array) ifTrue: [4] ifFalse: [1]. + (startIndex >= 1 + and: [(startIndex + count - 1) <= (interpreterProxy slotSizeOf: array)]) ifFalse: + [^interpreterProxy primitiveFailFor: PrimErrBadIndex]. + + "Note: adjust startIndex for zero-origin indexing" + bytesRead := self + sqFile: file + Read: count * elementSize + Into: (interpreterProxy cCoerce: (interpreterProxy firstIndexableField: array) to: #'char *') + At: (startIndex - 1) * elementSize. + interpreterProxy primitiveFailureCode = PrimErrObjectMayMove + and: [(retryCount := retryCount + 1) <= 2] "Two objects, the file and the array can move"] whileTrue: + [interpreterProxy + tenuringIncrementalGC; + primitiveFailFor: PrimNoErr]. + interpreterProxy failed ifFalse: + [interpreterProxy + pop: 5 "pop rcvr, file, array, startIndex, count" + thenPush:(interpreterProxy integerObjectOf: bytesRead // elementSize) "push # of elements read"]! Item was added: + ----- Method: NewCoObjectMemorySimulator>>setUpForUseByFacade: (in category 'debug support') ----- + setUpForUseByFacade: aCurrentImageCoInterpreterFacade + "This is a noop"! Item was added: + ----- Method: Spur32BitMMLECoSimulator>>setUpForUseByFacade: (in category 'debug support') ----- + setUpForUseByFacade: aCurrentImageCoInterpreterFacade + "Make sure that eden etc are initialized, so that methods can be printed. + This is really to make addressCouldBeObj: et al work." + self edenBytes: 0. + self setHeapBase: self freeStart + memoryLimit: self endOfMemory + endOfMemory: self endOfMemory!

1 0

Socket>>#sendData:count: does no error checking and hence locks up the system
by Eliot Miranda 01 Mar '16

01 Mar '16

Hi Levente, Hi All, I'm trying to investigate the socket issues in aio.c but have found a much moire basic issue. With my recent changes to Network that more carefully checked for errors the SocketTest>>testSocketReuse test appears to lock up. In fact, the VM is fine, happily doing what it's being told by Socket>>#sendData:count: Socket>>sendData: buffer count: n "Send the amount of data from the given buffer" | sent | sent := 0. [sent < n] whileTrue:[ sent := sent + (self sendSomeData: buffer startIndex: sent+1 count: (n-sent))]. The VM keeps trying to send data on a socket that is being reused and gets an error from sendto, answers 0 as the number of bytes sent, as required, but Socket>>#sendData:count: pays no heed and spins hard. Here's the traces: The test is SocketTest>>testSocketReuse which spawns two processes, one to send and one to receive data. Here are the processes: Process 0x48641f8 priority 40 0xbfec0498 M Socket>sendSomeData:startIndex:count:for: 0x4864d18: a(n) Socket 0xbfec04c0 M Socket>sendSomeData:startIndex:count: 0x4864d18: a(n) Socket 0xbfec04ec M Socket>sendData:count: 0x4864d18: a(n) Socket 0xbfec0520 I [] in SocketTest>testSocketReuse 0x4864dd0: a(n) SocketTest 0xbfec0540 I [] in BlockClosure>newProcess 0x4864df0: a(n) BlockClosure Process 0x6543178 priority 40 0xbfec22c8 I [] in Delay>wait 0x4864ea0: a(n) Delay 0xbfec22f0 I BlockClosure>ifCurtailed: 0x4864eb8: a(n) BlockClosure 0xbfec2314 I Delay>wait 0x4864ea0: a(n) Delay 0xbfec2340 I [] in SocketTest>testSocketReuse 0x4864dd0: a(n) SocketTest 0xbfec2360 M BlockClosure>ensure: 0x4864fa8: a(n) BlockClosure 0xbfec2390 I SocketTest>testSocketReuse 0x4864dd0: a(n) SocketTest Process 0x4864168 priority 40 0xbfec3438 I [] in DelayWaitTimeout>wait 0x48652f8: a(n) DelayWaitTimeout 0xbfec3458 M BlockClosure>ensure: 0x4865378: a(n) BlockClosure 0xbfec347c I DelayWaitTimeout>wait 0x48652f8: a(n) DelayWaitTimeout 0xbfec34a0 I Semaphore>waitTimeoutMSecs: 0x48652e0: a(n) Semaphore 0xbfec34c4 I Socket>waitForDataIfClosed: 0x4865408: a(n) Socket 0xbfec34f0 I Socket>receiveDataInto:startingAt: 0x4865408: a(n) Socket 0xbfec3520 I [] in SocketTest>testSocketReuse 0x4864dd0: a(n) SocketTest 0xbfec3540 I [] in BlockClosure>newProcess 0x48654c0: a(n) BlockClosure And here's the VM spinning: 15726 0 sqUnixSocket.c:1128 UDP sendData(11, 16) 15726 0 sqUnixSocket.c:1134 UDP send failed 56 Socket is already connected 15726 0 sqUnixSocket.c:1128 UDP sendData(11, 16) 15726 0 sqUnixSocket.c:1134 UDP send failed 56 Socket is already connected 15726 0 sqUnixSocket.c:1128 UDP sendData(11, 16) 15726 0 sqUnixSocket.c:1134 UDP send failed 56 Socket is already connected ...etc... Ah!! Of course. Because I have changed the default scheduling semantics in Squeak 5 to make preemption not a yield point, Socket>>#sendData:count: never yields to the other processes. Previously when the Delay process woke up this would implicitly yield the process spinning in Socket>>#sendData:count:. So Socket>>#sendData:count: needs to do a yield if no data is sent. However, shouldn't but also check for errors if no data is sent and do something like return an error if it discovers, via Socket>>primSocketError:, that the socket is not happy? _,,,^..^,,,_ best, Eliot

2 3

SystemTracing: SystemTracing-tfel.42.mcz
by squeak-dev-noreply＠lists.squeakfoundation.org 01 Mar '16

01 Mar '16

Tim Felgentreff uploaded a new version of SystemTracing to project SystemTracing: http://www.squeaksource.com/SystemTracing/SystemTracing-tfel.42.mcz ==================== Summary ==================== Name: SystemTracing-tfel.42 Author: tfel Time: 1 March 2016, 4:20:04.019236 pm UUID: 7122c1cf-f4a9-4963-a8a7-ef6fac27a687 Ancestors: SystemTracing-tfel.41 make header putting also work with smallintegers

1 0

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

Vm-dev March 2016