Vm-dev July 2018

vm-dev@lists.squeakfoundation.org

21 participants
55 discussions

VM Maker: VMMaker.oscog-eem.2422.mcz
by commits＠source.squeak.org 26 Jul '18

26 Jul '18

Eliot Miranda uploaded a new version of VMMaker to project VM Maker: http://source.squeak.org/VMMaker/VMMaker.oscog-eem.2422.mcz ==================== Summary ==================== Name: VMMaker.oscog-eem.2422 Author: eem Time: 25 July 2018, 5:45:27.688681 pm UUID: c3a5106e-a29f-41cb-9df3-119b03c01fa5 Ancestors: VMMaker.oscog-eem.2421 Cogits. Revamp pc mapping tests so they can apply to methods in the simulation (used to debug the issue fixed in VMMaker.oscog-eem.2421). Fix some bogus receivers in bcpcsDescriptorsAndStartpcsFor:bsOffset:do:. Fix the range check in VMCompiledMethodProxy>>at:, and cache some values (so that pcPreviousTo: doesn't crawl). =============== Diff against VMMaker.oscog-eem.2421 =============== Item was changed: ----- Method: Cogit>>bcpcsDescriptorsAndStartpcsFor:bsOffset:do: (in category 'tests-method map') ----- bcpcsDescriptorsAndStartpcsFor: aMethod bsOffset: bsOffset do: quinternaryBlock "Evaluate quinternaryBlock with the pc, byte, descriptor and numExtensions for all the bytecodes in aMethod. Evaluate with byte, descriptor and numExtensions nil for the initialPC of the mehtod and any blocks within it." <doNotGenerate> | nExts byte descriptor endpc latestContinuation pc primIdx blockEndPCs startpcs | ((primIdx := coInterpreter primitiveIndexOf: aMethod) > 0 and: [coInterpreter isQuickPrimitiveIndex: primIdx]) ifTrue: [^self]. latestContinuation := pc := coInterpreter startPCOfMethod: aMethod. startpcs := OrderedCollection with: pc. + blockEndPCs := OrderedCollection with: (objectMemory numBytesOf: aMethod). - blockEndPCs := OrderedCollection with: (coInterpreter numBytesOf: aMethod). quinternaryBlock value: pc value: nil value: nil value: 0 value: pc. "stackCheck/entry pc" primIdx > 0 ifTrue: [pc := pc + (self deltaToSkipPrimAndErrorStoreIn: aMethod + header: (objectMemory methodHeaderOf: aMethod))]. - header: (coInterpreter methodHeaderOf: aMethod))]. nExts := 0. endpc := objectMemory numBytesOf: aMethod. [pc <= endpc] whileTrue: [byte := objectMemory fetchByte: pc ofObject: aMethod. descriptor := self generatorAt: byte + bsOffset. descriptor isExtension ifFalse: [quinternaryBlock value: pc value: byte value: descriptor value: nExts value: startpcs last]. descriptor isReturn ifTrue: [pc >= latestContinuation ifTrue: [endpc := pc]]. (descriptor isBranch or: [descriptor isBlockCreation]) ifTrue: [| targetPC | targetPC := self latestContinuationPCFor: descriptor at: pc exts: nExts in: aMethod. descriptor isBlockCreation ifTrue: [quinternaryBlock value: (startpcs addLast: pc + descriptor numBytes) value: nil value: nil value: 0 value: startpcs last. blockEndPCs addLast: targetPC]. "stackCheck/entry pc" self assert: targetPC < endpc. latestContinuation := latestContinuation max: targetPC]. descriptor isReturn ifTrue: [pc + descriptor numBytes >= blockEndPCs last ifTrue: [blockEndPCs removeLast. startpcs removeLast]]. pc := pc + descriptor numBytes. nExts := descriptor isExtension ifTrue: [nExts + 1] ifFalse: [0]]! Item was added: + ----- Method: Cogit>>resolveToOopOrNil: (in category 'tests-method map') ----- + resolveToOopOrNil: proxyOrSpecialSelector + "Answer the oop of a proxy or a special selector. When testing pc mapping on simulation + methods (as opposed to CurrentImageFacades) selectorForSendBefore:in: answers either + VMObjectProxy instances (normal sends) or symbols (special selector sends)." + <doNotGenerate> + | index | + proxyOrSpecialSelector isSymbol ifFalse: + [^proxyOrSpecialSelector oop]. + index := Smalltalk specialSelectors indexOf: proxyOrSpecialSelector. + ^index > 0 ifTrue: + [coInterpreter specialSelector: index - 1 // 2]! Item was changed: ----- Method: Cogit>>testMcToBcPcMappingForCompiledMethod:cogMethod: (in category 'tests-method map') ----- testMcToBcPcMappingForCompiledMethod: aCompiledMethod cogMethod: cogMethod <doNotGenerate> | bcMethod subMethods prevMcpc | "self disassembleMethod: cogMethod" "coInterpreter symbolicMethod: cogMethod methodObject" "coInterpreter printOop: cogMethod methodObject" "self printPCMapPairsFor: cogMethod on: Transcript" cogMethod stackCheckOffset = 0 ifTrue: "frameless" [^self]. bcMethod := coInterpreter isCurrentImageFacade ifTrue: [coInterpreter objectForOop: cogMethod methodObject] ifFalse: [VMCompiledMethodProxy new for: cogMethod methodObject coInterpreter: coInterpreter objectMemory: objectMemory]. subMethods := self subMethodsAsRangesFor: cogMethod. self mapFor: cogMethod do: [:annotation :mcpc| | subMethod subCogMethod bcpc mappedpc | (self isPCMappedAnnotation: annotation) ifTrue: [subMethod := subMethods detect: [:range| range includes: mcpc] ifNone: ["a trailing call ceNonLocalReturnTrampoline's following pc is the start of a following block or the end of the map" subMethods detect: [:range| range includes: mcpc - 1]]. mcpc > subMethod first ifTrue: [bcpc := self bytecodePCFor: mcpc startBcpc: subMethod startpc in: (subCogMethod := subMethod cogMethod). self assert: bcpc ~= 0. mappedpc := self mcPCFor: bcpc startBcpc: subMethod startpc in: subCogMethod. subCogMethod stackCheckOffset = 0 ifTrue: [self assert: mappedpc > (subCogMethod address + self noCheckEntryOffset)] ifFalse: [self assert: mappedpc >= (subCogMethod address + subCogMethod stackCheckOffset)]. "mcpc = mappedpc is obviously what we want and expect. prevMcpc = mappedpc hacks around frame building accessors where the first bytecode is mapped twice, once for the stack check and once for the context inst var access. The bytecode pc can only map back to a single mcpc, the first, so the second map entry will fail without this hack." self assert: (mcpc = mappedpc or: [prevMcpc = mappedpc]). (self isSendAnnotation: annotation) ifTrue: [| mcSelector bcSelector | mcSelector := self selectorForSendAt: mcpc annotation: annotation in: aCompiledMethod. "sends map to the following pc. need to find the selector for the previous pc" bcSelector := self selectorForSendBefore: bcpc in: bcMethod. + self assert: (mcSelector = bcSelector + or: [mcSelector = (self resolveToOopOrNil: bcSelector)])]]. - self assert: mcSelector = bcSelector]]. prevMcpc := mcpc]. false "keep scanning"]! Item was changed: ----- Method: Cogit>>testPCMappingForCompiledMethod:cogMethod: (in category 'tests-method map') ----- testPCMappingForCompiledMethod: aCompiledMethod cogMethod: cm <doNotGenerate> methodObj := methodHeader := nil. + cm isInteger ifTrue: "If testing in a simulation..." + [^self testPCMappingForCompiledMethod: aCompiledMethod cogMethod: (self cogMethodSurrogateAt: cm)]. self testMcToBcPcMappingForCompiledMethod: aCompiledMethod cogMethod: cm; testBcToMcPcMappingForCogMethod: cm! Item was changed: VMObjectProxy subclass: #VMCompiledMethodProxy + instanceVariableNames: 'size numLiterals' - instanceVariableNames: '' classVariableNames: '' poolDictionaries: '' category: 'VMMaker-Support'! !VMCompiledMethodProxy commentStamp: 'eem 8/6/2014 14:48' prior: 0! A VMCompiledMethodProxy is a wrapper for the oop of a CompiledMethod object in the simulator VM's heap that provides accessd to the oop as if it were a CompiledMethod object.! Item was changed: ----- Method: VMCompiledMethodProxy>>at: (in category 'accessing') ----- at: index + ^(index between: self numLiterals + 1 * objectMemory wordSize + 1 and: self size) - ^(index between: 1 and: (objectMemory numBytesOf: oop)) ifTrue: [objectMemory fetchByte: index - 1 ofObject: oop] ifFalse: [self errorSubscriptBounds: index]! Item was changed: ----- Method: VMCompiledMethodProxy>>numLiterals (in category 'accessing') ----- numLiterals "Answer the number of literals used by the receiver." + ^numLiterals ifNil: [numLiterals := objectMemory literalCountOf: oop]! - ^objectMemory literalCountOf: oop! Item was added: + ----- Method: VMCompiledMethodProxy>>pcPreviousTo: (in category 'scanning') ----- + pcPreviousTo: thePC + "Answer the pc of the bytecode before the bytecode at thePC." + | pc prevPc byte encoderClass | + thePC > self endPC ifTrue: [^self endPC]. + pc := self initialPC. + encoderClass := self encoderClass. + [pc < thePC] whileTrue: + [byte := self at: (prevPc := pc). + [pc := pc + (encoderClass bytecodeSize: byte). + encoderClass isExtension: byte] whileTrue: + [byte := self at: pc]]. + ^prevPc! Item was changed: ----- Method: VMCompiledMethodProxy>>size (in category 'accessing') ----- size + ^size ifNil: [size := objectMemory numBytesOf: oop]! - ^objectMemory numBytesOf: oop!

1 0

VM Maker: VMMaker.oscog-eem.2421.mcz
by commits＠source.squeak.org 26 Jul '18

26 Jul '18

Eliot Miranda uploaded a new version of VMMaker to project VM Maker: http://source.squeak.org/VMMaker/VMMaker.oscog-eem.2421.mcz ==================== Summary ==================== Name: VMMaker.oscog-eem.2421 Author: eem Time: 25 July 2018, 5:17:07.310927 pm UUID: f273bc45-113b-4685-822c-f0db0293c206 Ancestors: VMMaker.oscog-eem.2420 Cogits. Fix bad bug in pc mapping of methods containing primitives. methodUsesPrimitiveErrorCode:header: used the initialPC inst var, rather than deriving the initialPC of the method parameter. initialPC is stale/over ripe, being that of the last JITTED method. So many methods containing primitives get their PC mapping wrong. We haven't noticed this much because primiitves typically succeed. Nuke an obsolete version of the method. Have the Spur immutability store generators voidReceiverOptStatus, since with immutability, stores are suspension pioints (when an immutability check fails) and so receiverResultReg cannot be assumed to be live after a store. Hence also implement SimpleStackBasedCogit>>voidReceiverOptStatus for compatibility ()a noop). Cleanups: Extend the comment for genGenericStorePop:slotIndex:destReg:needsStoreCheck:needsRestoreRcvr:needsImmutabilityCheck: (and hence also (forgive me!) reformat to fix confusing indentation). Remove ssAllocateRequiredRegMask:upThrough:, inlining it into its callers, ssAllocateCall/RequiredReg:... Clean up the computation of the register mask in SistaCogit>>genForwardersInlinedIdenticalOrNotIf: Add a helpful comment to genCallMustBeBooleanFor: (I can never remember the selector the trampoline invokes). Eliminate cr & lf in literal printing when decorating disassembly. =============== Diff against VMMaker.oscog-eem.2420 =============== Item was changed: ----- Method: CogObjectRepresentationForSpur>>genStoreWithImmutabilityAndStoreCheckSourceReg:slotIndex:destReg:scratchReg:needRestoreRcvr: (in category 'compile abstract instructions') ----- genStoreWithImmutabilityAndStoreCheckSourceReg: sourceReg slotIndex: index destReg: destReg scratchReg: scratchReg needRestoreRcvr: needRestoreRcvr "Store check code is duplicated to use a single trampoline" <option: #IMMUTABILITY> | immutableJump jmpImmediate jmpDestYoung jmpSourceOld jmpAlreadyRemembered | <var: #immutableJump type: #'AbstractInstruction *'> <var: #jmpImmediate type: #'AbstractInstruction *'> <var: #jmpDestYoung type: #'AbstractInstruction *'> <var: #jmpSourceOld type: #'AbstractInstruction *'> <var: #jmpAlreadyRemembered type: #'AbstractInstruction *'> immutableJump := self genJumpImmutable: destReg scratchReg: scratchReg. cogit genTraceStores. "do the store" cogit MoveR: sourceReg Mw: index * objectMemory wordSize + objectMemory baseHeaderSize r: destReg. "store check" jmpImmediate := self genJumpImmediate: sourceReg. "Get the old/new boundary in scratchReg" cogit MoveCw: objectMemory storeCheckBoundary R: scratchReg. "Is target young? If so we're done" cogit CmpR: scratchReg R: destReg. "N.B. FLAGS := destReg - scratchReg" jmpDestYoung := cogit JumpBelow: 0. "Is value stored old? If so we're done." cogit CmpR: scratchReg R: sourceReg. "N.B. FLAGS := valueReg - scratchReg" jmpSourceOld := cogit JumpAboveOrEqual: 0. "value is young and target is old. Need to remember this only if the remembered bit is not already set." CheckRememberedInTrampoline ifFalse: [jmpAlreadyRemembered := self genCheckRememberedBitOf: destReg scratch: scratchReg]. "Set the inst var index for the benefit of the immutability check. The trampoline will repeat the check to choose between the immutbality violation and the store check." immutableJump jmpTarget: cogit Label. self genStoreTrampolineCall: index. + cogit voidReceiverOptStatus. + needRestoreRcvr ifTrue: + [cogit putSelfInReceiverResultReg]. - needRestoreRcvr ifTrue: [ cogit putSelfInReceiverResultReg ]. jmpImmediate jmpTarget: (jmpDestYoung jmpTarget: (jmpSourceOld jmpTarget: cogit Label)). CheckRememberedInTrampoline ifFalse: [jmpAlreadyRemembered jmpTarget: jmpSourceOld getJmpTarget]. ^ 0! Item was changed: ----- Method: CogObjectRepresentationForSpur>>genStoreWithImmutabilityButNoStoreCheckSourceReg:slotIndex:destReg:scratchReg:needRestoreRcvr: (in category 'compile abstract instructions') ----- genStoreWithImmutabilityButNoStoreCheckSourceReg: sourceReg slotIndex: index destReg: destReg scratchReg: scratchReg needRestoreRcvr: needRestoreRcvr "Gen an immutability check with no store check (e.g. assigning an immediate literal)" <option: #IMMUTABILITY> <var: #mutableJump type: #'AbstractInstruction *'> <var: #immutabilityFailure type: #'AbstractInstruction *'> | immutabilityFailure mutableJump | "imm check has its own trampoline" mutableJump := self genJumpMutable: destReg scratchReg: scratchReg. self genStoreTrampolineCall: index. + cogit voidReceiverOptStatus. + needRestoreRcvr ifTrue: + [cogit putSelfInReceiverResultReg]. - needRestoreRcvr ifTrue: [ cogit putSelfInReceiverResultReg ]. immutabilityFailure := cogit Jump: 0. mutableJump jmpTarget: cogit Label. cogit genTraceStores. "do the store" cogit MoveR: sourceReg Mw: index * objectMemory wordSize + objectMemory baseHeaderSize r: destReg. immutabilityFailure jmpTarget: cogit Label. ^ 0! Item was changed: ----- Method: Cogit>>genCallMustBeBooleanFor: (in category 'trampoline support') ----- genCallMustBeBooleanFor: boolean + "Call ceSendMustBeBooleanTo: via the relevant trampoline." ^self CallRT: (boolean = objectMemory falseObject ifTrue: [ceSendMustBeBooleanAddFalseTrampoline] ifFalse: [ceSendMustBeBooleanAddTrueTrampoline])! Item was changed: ----- Method: CurrentImageCoInterpreterFacade>>lookupAddress: (in category 'labels') ----- lookupAddress: address | thing | thing := objectMap keyAtValue: address ifAbsent: [variables keyAtValue: address ifAbsent: [^nil]]. + ^(((thing isLiteral - ^(thing isLiteral ifTrue: [thing storeString] + ifFalse: [thing asString]) contractTo: 64) + copyReplaceAll: String cr with: '\r') + copyReplaceAll: String lf with: '\n'! - ifFalse: [thing asString]) contractTo: 64! Item was removed: - ----- Method: SimpleStackBasedCogit>>methodUsesPrimitiveErrorCode (in category 'compile abstract instructions') ----- - methodUsesPrimitiveErrorCode - "Answer if methodObj contains a primitive and uses the primitive error code." - ^(coInterpreter primitiveIndexOfMethod: methodObj header: methodHeader) > 0 - and: [(coInterpreter longStoreBytecodeForHeader: methodHeader) - = (objectMemory - fetchByte: initialPC + (coInterpreter sizeOfCallPrimitiveBytecode: methodHeader) - ofObject: methodObj)]! Item was changed: ----- Method: SimpleStackBasedCogit>>methodUsesPrimitiveErrorCode:header: (in category 'compile abstract instructions') ----- methodUsesPrimitiveErrorCode: aMethodObj header: aMethodHeader "Answer if aMethodObj contains a primitive and uses the primitive error code." <inline: true> ^(coInterpreter primitiveIndexOfMethod: aMethodObj header: aMethodHeader) > 0 and: [(coInterpreter longStoreBytecodeForHeader: aMethodHeader) = (objectMemory + fetchByte: (coInterpreter startPCOfMethod: aMethodObj) + (coInterpreter sizeOfCallPrimitiveBytecode: aMethodHeader) - fetchByte: initialPC + (coInterpreter sizeOfCallPrimitiveBytecode: aMethodHeader) ofObject: aMethodObj)]! Item was added: + ----- Method: SimpleStackBasedCogit>>voidReceiverOptStatus (in category 'bytecode generator support') ----- + voidReceiverOptStatus + "No op in this cogit. Provided for compatibility with the cleverer subclasses." + <inline: true>! Item was changed: ----- Method: SistaCogit>>genForwardersInlinedIdenticalOrNotIf: (in category 'bytecode generators') ----- genForwardersInlinedIdenticalOrNotIf: orNot "Override to count inlined branches if followed by a conditional branch. We borrow the following conditional branch's counter and when about to inline the comparison we decrement the counter (without writing it back) and if it trips simply abort the inlining, falling back to the normal send which will then continue to the conditional branch which will trip and enter the abort." | nextPC postBranchPC targetBytecodePC branchDescriptor counterReg fixup jumpEqual jumpNotEqual counterAddress countTripped unforwardArg unforwardRcvr argReg rcvrReg regMask | <var: #fixup type: #'BytecodeFixup *'> <var: #countTripped type: #'AbstractInstruction *'> <var: #label type: #'AbstractInstruction *'> <var: #branchDescriptor type: #'BytecodeDescriptor *'> <var: #jumpEqual type: #'AbstractInstruction *'> <var: #jumpNotEqual type: #'AbstractInstruction *'> ((coInterpreter isOptimizedMethod: methodObj) or: [needsFrame not]) ifTrue: [^super genForwardersInlinedIdenticalOrNotIf: orNot]. regMask := 0. self extractMaybeBranchDescriptorInto: [ :descr :next :postBranch :target | branchDescriptor := descr. nextPC := next. postBranchPC := postBranch. targetBytecodePC := target ]. unforwardRcvr := (objectRepresentation isUnannotatableConstant: (self ssValue: 1)) not. unforwardArg := (objectRepresentation isUnannotatableConstant: self ssTop) not. "If an operand is an annotable constant, it may be forwarded, so we need to store it into a register so the forwarder check can jump back to the comparison after unforwarding the constant. However, if one of the operand is an unnanotable constant, does not allocate a register for it (machine code will use operations on constants)." - rcvrReg:= argReg := NoReg. self allocateEqualsEqualsRegistersArgNeedsReg: unforwardArg rcvrNeedsReg: unforwardRcvr into: [ :rcvr :arg | rcvrReg:= rcvr. argReg := arg ]. + - - argReg ~= NoReg ifTrue: [ regMask := self registerMaskFor: argReg ]. - rcvrReg ~= NoReg ifTrue: [ regMask := regMask bitOr: (self registerMaskFor: rcvrReg) ]. - "Only interested in inlining if followed by a conditional branch." (branchDescriptor isBranchTrue or: [branchDescriptor isBranchFalse]) ifFalse: [^ self genIdenticalNoBranchArgIsConstant: unforwardArg not rcvrIsConstant: unforwardRcvr not argReg: argReg rcvrReg: rcvrReg orNotIf: orNot]. "If branching the stack must be flushed for the merge" self ssFlushTo: simStackPtr - 2. unforwardArg ifTrue: [ objectRepresentation genEnsureOopInRegNotForwarded: argReg scratchReg: TempReg ]. unforwardRcvr ifTrue: [ objectRepresentation genEnsureOopInRegNotForwarded: rcvrReg scratchReg: TempReg ]. + regMask := argReg = NoReg + ifTrue: [self registerMaskFor: rcvrReg] + ifFalse: + [rcvrReg = NoReg + ifTrue: [self registerMaskFor: argReg] + ifFalse: [self registerMaskFor: rcvrReg and: argReg]]. counterReg := self allocateRegNotConflictingWith: regMask. self genExecutionCountLogicInto: [ :cAddress :countTripBranch | counterAddress := cAddress. countTripped := countTripBranch ] counterReg: counterReg. self assert: (unforwardArg or: [ unforwardRcvr ]). self genCmpArgIsConstant: unforwardArg not rcvrIsConstant: unforwardRcvr not argReg: argReg rcvrReg: rcvrReg. self ssPop: 2. orNot == branchDescriptor isBranchTrue "orNot is true for ~~" ifFalse: [ fixup := (self ensureNonMergeFixupAt: postBranchPC) asUnsignedInteger. self JumpZero: (self ensureNonMergeFixupAt: targetBytecodePC) asUnsignedInteger ] ifTrue: [ fixup := (self ensureNonMergeFixupAt: targetBytecodePC) asUnsignedInteger. self JumpZero: (self ensureNonMergeFixupAt: postBranchPC) asUnsignedInteger ]. self genFallsThroughCountLogicCounterReg: counterReg counterAddress: counterAddress. self Jump: fixup. countTripped jmpTarget: self Label. "inlined version of #== ignoring the branchDescriptor if the counter trips to have normal state for the optimizer" self ssPop: -2. self genCmpArgIsConstant: unforwardArg not rcvrIsConstant: unforwardRcvr not argReg: argReg rcvrReg: rcvrReg. self ssPop: 2. "This code necessarily directly falls through the jumpIf: code which pops the top of the stack into TempReg. We therefore directly assign the result to TempReg to save one move instruction" jumpEqual := orNot ifFalse: [self JumpZero: 0] ifTrue: [self JumpNonZero: 0]. self genMoveFalseR: TempReg. jumpNotEqual := self Jump: 0. jumpEqual jmpTarget: (self genMoveTrueR: TempReg). jumpNotEqual jmpTarget: self Label. self ssPushRegister: TempReg. (self fixupAt: nextPC) notAFixup ifTrue: [ branchReachedOnlyForCounterTrip := true ]. ^ 0! Item was changed: ----- Method: StackToRegisterMappingCogit>>genGenericStorePop:slotIndex:destReg:needsStoreCheck:needsRestoreRcvr:needsImmutabilityCheck: (in category 'bytecode generator stores') ----- genGenericStorePop: popBoolean slotIndex: slotIndex destReg: destReg needsStoreCheck: needsStoreCheck needsRestoreRcvr: needsRestoreReceiver needsImmutabilityCheck: needsImmCheck "Generates a store into an object that *cannot* be a context. This code is common between multiple stores (litVar, instVar, remoteInstVar, RemoteTemp) Multiple settings: - needsStoreCheck (young into old object check) - needRestoreRcvr (ensures the receiver is live across the store) - needsImmCheck (do the call-back if the receiver is immutable)" "We have two very different paths as only the immutability path requires a specific register + for the value on top of stack as well as the stack flush. + N.B. If IMMUTABILITY then ReceiverResultReg/destReg will be smashed if needsImmCheck. + If not IMMUTABILITY then ReceiverResultReg will be preserved by the ceStoreCheck trampoline." - for the value on top of stack as well as the stack flush" | topReg | <inline: true> + self cppIf: IMMUTABILITY ifTrue: + [needsImmCheck ifTrue: + [self ssAllocateRequiredReg: ClassReg upThrough: simStackPtr - 1. "If already classReg don't spill it" + "we replace the top value for the flush" + self ssStoreAndReplacePop: popBoolean toReg: ClassReg. + self ssFlushTo: simStackPtr. + ^objectRepresentation + genStoreWithImmutabilityCheckSourceReg: ClassReg + slotIndex: slotIndex + destReg: destReg + scratchReg: TempReg + needsStoreCheck: needsStoreCheck + needRestoreRcvr: needsRestoreReceiver]]. + topReg := self + allocateRegForStackEntryAt: 0 + notConflictingWith: (self registerMaskFor: destReg). + self ssStorePop: popBoolean toReg: topReg. + ^objectRepresentation + genStoreSourceReg: topReg + slotIndex: slotIndex + destReg: destReg + scratchReg: TempReg + inFrame: needsFrame + needsStoreCheck: needsStoreCheck! - self - cppIf: IMMUTABILITY - ifTrue: - [needsImmCheck - ifTrue: - [self ssAllocateRequiredReg: ClassReg upThrough: simStackPtr - 1. "If already classReg don't spill it" - "we replace the top value for the flush" - self ssStoreAndReplacePop: popBoolean toReg: ClassReg. - self ssFlushTo: simStackPtr. - ^objectRepresentation - genStoreWithImmutabilityCheckSourceReg: ClassReg - slotIndex: slotIndex - destReg: destReg - scratchReg: TempReg - needsStoreCheck: needsStoreCheck - needRestoreRcvr: needsRestoreReceiver]]. - topReg := self - allocateRegForStackEntryAt: 0 - notConflictingWith: (self registerMaskFor: destReg). - self ssStorePop: popBoolean toReg: topReg. - ^objectRepresentation - genStoreSourceReg: topReg - slotIndex: slotIndex - destReg: destReg - scratchReg: TempReg - inFrame: needsFrame - needsStoreCheck: needsStoreCheck - ! Item was changed: ----- Method: StackToRegisterMappingCogit>>ssAllocateCallReg: (in category 'simulation stack') ----- ssAllocateCallReg: requiredReg "Allocate a register needed in a run-time call (i.e. flush uses of the register to the real stack). Since the run-time can smash any and all caller-saved registers also flush all caller-saved registers." + <inline: true> self ssAllocateRequiredRegMask: (CallerSavedRegisterMask bitOr: (self registerMaskFor: requiredReg)) + upThrough: simStackPtr + upThroughNative: simNativeStackPtr! - upThrough: simStackPtr! Item was changed: ----- Method: StackToRegisterMappingCogit>>ssAllocateCallReg:and: (in category 'simulation stack') ----- ssAllocateCallReg: requiredReg1 and: requiredReg2 "Allocate registers needed in a run-time call (i.e. flush uses of the registers to the real stack). Since the run-time can smash any and all caller-saved registers also flush all caller-saved registers." + <inline: true> self ssAllocateRequiredRegMask: (CallerSavedRegisterMask bitOr: ((self registerMaskFor: requiredReg1) bitOr: (self registerMaskFor: requiredReg2))) + upThrough: simStackPtr + upThroughNative: simNativeStackPtr! - upThrough: simStackPtr! Item was changed: ----- Method: StackToRegisterMappingCogit>>ssAllocateCallReg:and:and: (in category 'simulation stack') ----- ssAllocateCallReg: requiredReg1 and: requiredReg2 and: requiredReg3 "Allocate registers needed in a run-time call (i.e. flush uses of the registers to the real stack). Since the run-time can smash any and all caller-saved registers also flush all caller-saved registers." + <inline: true> self ssAllocateRequiredRegMask: (CallerSavedRegisterMask bitOr: (self registerMaskFor: requiredReg1 and: requiredReg2 and: requiredReg3)) + upThrough: simStackPtr + upThroughNative: simNativeStackPtr! - upThrough: simStackPtr! Item was changed: ----- Method: StackToRegisterMappingCogit>>ssAllocateCallReg:and:and:and: (in category 'simulation stack') ----- ssAllocateCallReg: requiredReg1 and: requiredReg2 and: requiredReg3 and: requiredReg4 "Allocate registers needed in a run-time call (i.e. flush uses of the registers to the real stack). Since the run-time can smash any and all caller-saved registers also flush all caller-saved registers." + <inline: true> self ssAllocateRequiredRegMask: (CallerSavedRegisterMask bitOr: ((self registerMaskFor: requiredReg1) bitOr: ((self registerMaskFor: requiredReg2) bitOr: ((self registerMaskFor: requiredReg3) bitOr: (self registerMaskFor: requiredReg4))))) + upThrough: simStackPtr + upThroughNative: simNativeStackPtr! - upThrough: simStackPtr! Item was changed: ----- Method: StackToRegisterMappingCogit>>ssAllocateRequiredReg: (in category 'simulation stack') ----- ssAllocateRequiredReg: requiredReg + <inline: true> self ssAllocateRequiredRegMask: (self registerMaskFor: requiredReg) + upThrough: simStackPtr + upThroughNative: simNativeStackPtr! - upThrough: simStackPtr! Item was changed: ----- Method: StackToRegisterMappingCogit>>ssAllocateRequiredReg:and: (in category 'simulation stack') ----- ssAllocateRequiredReg: requiredReg1 and: requiredReg2 + <inline: true> self ssAllocateRequiredRegMask: ((self registerMaskFor: requiredReg1) bitOr: (self registerMaskFor: requiredReg2)) + upThrough: simStackPtr + upThroughNative: simNativeStackPtr! - upThrough: simStackPtr! Item was changed: ----- Method: StackToRegisterMappingCogit>>ssAllocateRequiredReg:upThrough: (in category 'simulation stack') ----- ssAllocateRequiredReg: requiredReg upThrough: stackPtr + <inline: true> self ssAllocateRequiredRegMask: (self registerMaskFor: requiredReg) + upThrough: stackPtr + upThroughNative: simNativeStackPtr! - upThrough: stackPtr! Item was removed: - ----- Method: StackToRegisterMappingCogit>>ssAllocateRequiredRegMask:upThrough: (in category 'simulation stack') ----- - ssAllocateRequiredRegMask: requiredRegsMask upThrough: stackPtr - <inline: true> - self ssAllocateRequiredRegMask: requiredRegsMask upThrough: stackPtr upThroughNative: simNativeStackPtr. - !

1 0

[OpenSmalltalk/opensmalltalk-vm] 201807240904
by Fabio Niephaus 25 Jul '18

25 Jul '18

-- You are receiving this because you are subscribed to this thread. View it on GitHub: https://github.com/OpenSmalltalk/opensmalltalk-vm/releases/tag/201807240904

3 4

[OpenSmalltalk/opensmalltalk-vm] a402a7: ARM VMs are always threaded
by GitHub 25 Jul '18

25 Jul '18

Branch: refs/heads/Cog Home: https://github.com/OpenSmalltalk/opensmalltalk-vm Commit: a402a7fdc7786631f0eb4820d01f1b579cc66e00 https://github.com/OpenSmalltalk/opensmalltalk-vm/commit/a402a7fdc7786631f0… Author: Fabio Niephaus <code(a)fniephaus.com> Date: 2018-07-25 (Wed, 25 Jul 2018) Changed paths: M deploy/pack-vm.sh Log Message: ----------- ARM VMs are always threaded **NOTE:** This service been marked for deprecation: https://developer.github.com/changes/2018-04-25-github-services-deprecation/ Functionality will be removed from GitHub.com on January 31st, 2019.

1 0

Build completed: opensmalltalk-vm 1.0.1351
by AppVeyor 25 Jul '18

25 Jul '18

1 0

Build failed: opensmalltalk-vm 1.0.1350
by AppVeyor 25 Jul '18

25 Jul '18

1 0

[OpenSmalltalk/opensmalltalk-vm]
by GitHub 25 Jul '18

25 Jul '18

Branch: refs/tags/201807240904 Home: https://github.com/OpenSmalltalk/opensmalltalk-vm **NOTE:** This service been marked for deprecation: https://developer.github.com/changes/2018-04-25-github-services-deprecation/ Functionality will be removed from GitHub.com on January 31st, 2019.

2 1

Re: [Vm-dev] [squeak-dev] Finishing 5.2 (was Re: Project for 'after 5.2 release')
by Tobias Pape 24 Jul '18

24 Jul '18

> On 24.07.2018, at 11:02, Levente Uzonyi <leves(a)caesar.elte.hu> wrote: > > On Tue, 24 Jul 2018, Fabio Niephaus wrote: > >> On Tue, Jul 24, 2018 at 10:02 AM Edgar J. De Cleene <edgardec2005(a)gmail.com> wrote: >> I think 5.2 is stable now. >> I using as base for things I working on. >> Could have a system on Linux box , use the share directory as link in Mac >> and open from Mac and save changes. >> So the .image works from Linux or from Mac remote on same network. >> I check all bugs reports of 5.0 to now in Mantis for if still relevant. >> Test on Linux gives only one red . >> Wish change status to beta, close any new thing and only acept fixes. >> And all Chris said and more could be 5.3 plan. >> >> Edgar >> @morplenauta >> Hi Edgar, >> Shall we update all VMs one more time before the release? I think I saw some SqueakSSL-related bug fixes we might want to ship with the new >> release. > > Definitely. But before that, I'd prefer to have this integrated: https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/273 Done -t > > Levente > >> On the other hand, we might want to test these VMs for a few days... >> Cheers, >> Fabio >> >> >> On 23/07/2018, 23:24, "David T. Lewis" <lewis(a)mail.msen.com> wrote: >> >> > On Mon, Jul 23, 2018 at 06:54:55PM -0700, Chris Cunningham wrote: >> > Just >> > noticing the number of things that at least one person has flagged for >> > >> > 'after 5.2': >> > >> > * ProvideAnswerNotification and UIManager ( >> > >> > http://forum.world.st/ProvideAnswerNotification-and-UIManager-td5081488.html >> > >> > ) >> > * Making ServerDirectory example work ( >> > >> > http://forum.world.st/ServerDirectory-example-login-is-dead-td5061704.html ) >> > >> > * Fixing DateAndTimeLeapTest>>#testAsSeconds ( >> > >> > http://forum.world.st/template/NamlServlet.jtp?macro=search_page&node=12947… >> > query=Some+failing+tests+in+Squeak5.2alpha&n=1294792 >> > ) >> > * Making EToys and >> > Morphic separation clearer ( >> > >> > http://forum.world.st/The-Inbox-Morphic-hjh-1453-mcz-td5079588.html#a5079826 >> > >> > ) >> > * Make EToys unloadable ( >> > >> > http://forum.world.st/The-Inbox-EToys-hjh-333-mcz-td5079273.html#a5079454 ) >> > >> > * EToysProject ( >> > >> > http://forum.world.st/MorphicProject-subclass-EtoysProject-td4974975i60.htm… >> > 5078814 >> > ) >> > * HelpBrowser fixes for Text ( >> > >> > http://forum.world.st/Improving-SqueakToolsHelp-class-gt-gt-fontSizeSummary… >> > veals-problems-in-Text-amp-styles-td5081938.html >> > ) >> > >> > Thanks, >> > cbc >> >> That >> > is a good thing :-) The items in this list are things that are likely >> to >> > introduce instability to trunk if we start working on them right now, so >> it is >> > better to wait for the release before pushing changes into the trunk >> for >> > these issues. >> >> Let's focus on getting the 5.2 release done, including giving >> > Edgar our full >> support to close any issues that are critical to the release. >> > Once that is >> done, we can address the issues that are flagged for follow >> > up. >> >> Dave

4 3

Re: [Vm-dev] [OpenSmalltalk/opensmalltalk-vm] Control keys broken on OS X (#43)
by Tobias Pape 24 Jul '18

24 Jul '18

No complaints in nearly a year. close. -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/OpenSmalltalk/opensmalltalk-vm/issues/43#issuecomment-40…

1 0

Re: [Vm-dev] [OpenSmalltalk/opensmalltalk-vm] Control keys broken on OS X (#43)
by Tobias Pape 24 Jul '18

24 Jul '18

Closed #43. -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/OpenSmalltalk/opensmalltalk-vm/issues/43#event-1749623909

1 0

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

Vm-dev July 2018