Vm-dev September 2019

vm-dev@lists.squeakfoundation.org

21 participants
118 discussions

Re: [Vm-dev] [squeak-dev] Difficult to debug VM crash with full blocks and Sista V1
by Eliot Miranda 21 Sep '19

21 Sep '19

Hi All, two further reports. One is that attempting to simulate SocketTest suite run with a small code zone to provoke code comp-actions did not produce a crash. Sad, because I left it running over night. There's still more work to do with simulation here though because the simulation is far from perfect; only two out of 14 tests pass. At least they run ;-) 14 run, 2 passes, 0 expected failures, 6 failures, 6 errors, 0 unexpected passes squeak> The other report is that I have much more information on the nature of the crash and reason to suspect something to do with code zone management. The JITted version of the Delay class>>#startTimerEventLoop method is badly corrupted from its last field on: (lldb) call printCogMethodFor((void *)0x10cefe70a) 0x10cefe288 <-> 0x10cefe760: method: 0x1117a96a8 selector: 0x7fff58e9af6d " is no where obvious" *(CogMethod *)0x10cefe288 (CogMethod) $14 = { objectHeader = 0x008000000a000035 cmNumArgs = 0 cmType = 2 CMMethod cmRefersToYoung = false cpicHasMNUCaseOrCMIsFullBlock = false cmUsageCount = 1 cmUsesPenultimateLit = false cbUsesInstVars = false cmHasMovableLiteral = true cmUnusedFlag = 0x00000001 stackCheckOffset = 116/0x74 blockSize = 1240/0x4d8 blockEntryOffset = 1154/0x482 (1149/0x479 in simulator) methodObject = 0x1117a96a8 methodHeader = 0x00000000000000b1 selector = 0x00007fff58e9af6d } Everything is fine up to the selector field which is well out of range of the heap, an odd value ending in 16rD, and something followed by garbage instructions: (lldb) print/x (CogMethod *)0x10cefe288 + 1 (CogMethod *) $24 = 0x000000010cefe2b0 (lldb) x/10i 0x000000010cefe2b0 0x10cefe2b0: d0 0c bd ef fe 7f 00 rorb 0x7ffeef(,%rdi,4) 0x10cefe2b7: 00 48 89 addb %cl, -0x77(%rax) 0x10cefe2ba: d0 48 83 rorb -0x7d(%rax) 0x10cefe2bd: e0 07 loopne 0x10cefe2c6 0x10cefe2bf: 75 09 jne 0x10cefe2ca 0x10cefe2c1: 48 8b 02 movq (%rdx), %rax 0x10cefe2c4: 48 25 ff ff 3f 00 andq $0x3fffff, %rax ; imm = 0x3FFFFF 0x10cefe2ca: 48 39 c8 cmpq %rcx, %rax 0x10cefe2cd: 75 e4 jne 0x10cefe2b3 0x10cefe2cf: 4c 8b de movq %rsi, %r11 the crash occurs I believe when the sort block for SuspendedDelays created in Delay class>>#startTimerEventLoop is activated via the code in CoInterpreter>>#executeCogBlock:closure:mayContextSwitch: which jumps to the blockEntry code for the method (code that dispatches between the two blocks in startTimerEventLoop). This dispatch code is also corrupted: (lldb) print/x 0x10cefe288 + 0x482 (CogMethod for startTimerEventLoop + its blockEntryOffset) (long) $36 = 0x000000010cefe70a (lldb) x/10i 0x10cefe70a 0x10cefe70a: e7 ef outl %eax, $0xef 0x10cefe70c: 0c 01 orb $0x1, %al 0x10cefe70e: 00 00 addb %al, (%rax) 0x10cefe710: 00 00 addb %al, (%rax) 0x10cefe712: 00 00 addb %al, (%rax) whereas it should look like 00002c74: xorq %r9, %r9 : 4D 31 C9 00002c77: jmp .+0x3 (0x2c7c=startTimerEventLoop@47C) : EB 03 blockEntry: 00002c79: movq %rdx, %r9 : 49 89 D1 00002c7c: movq methodDict@16(%rdx), %rax : 48 8B 42 10 00002c80: cmpq $0x661, %rax : 48 3D 61 06 00 00 00002c86: jle .-0xE4 (0x2ba8=startTimerEventLoop@3A8) : 0F 8E 1C FF FF FF block startpc: CB/659 00002c8c: jmp .-0x66 (0x2c28=startTimerEventLoop@428) : EB 9A block startpc: DF/6F9 So my current best guess is that JITted version of the Delay class>>#startTimerEventLoop method is corrupted by some bug in code zone compaction. A second route is indicated by the prim trace log which strangely shows two stack overflow events, not one, very soon before the crash, which could indicate a bug in the interpreter/jit control flow on context switch: (lldb) call dumpPrimTraceLog() … forceDisplayUpdate utcMicrosecondClock utcMicrosecondClock findNextUnwindContextUpTo: tempAt: tempAt:put: tempAt: terminateTo: tempAt:put: findNextUnwindContextUpTo: terminateTo: basicNew: primSocketConnectionStatus: utcMicrosecondClock primSocketSendDone: primSocket:sendData:startIndex:count: utcMicrosecondClock primSocketSendDone: primSocketConnectionStatus: utcMicrosecondClock **StackOverflow** **StackOverflow** wait signal utcMicrosecondClock(newMethod -> 0x10cec3230 : 0x111f565a8) This log is showing non-jitted primitive invocations and context switch/code zone reclamation events outside of JITted code (for obvious performance reasons we don't record such events in JITted code). So, still more digging to narrow this down, and possibly fixing simulation of the suite is worth-while. But what I really need is a more reproducible case in the real VM. Sigh :-) _,,,^..^,,,_ best, Eliot

2 2

VM Maker: VMMaker.oscog-eem.2567.mcz
by commits＠source.squeak.org 21 Sep '19

21 Sep '19

Eliot Miranda uploaded a new version of VMMaker to project VM Maker: http://source.squeak.org/VMMaker/VMMaker.oscog-eem.2567.mcz ==================== Summary ==================== Name: VMMaker.oscog-eem.2567 Author: eem Time: 21 September 2019, 9:48:03.990619 am UUID: b3d92f49-e94c-48f3-99b0-1a644c1d37f9 Ancestors: VMMaker.oscog-eem.2566 Fix two slips, one in printing contexts, the other in simulating primitiveSocketListenWithOrWithoutBacklog. =============== Diff against VMMaker.oscog-eem.2566 =============== Item was changed: ----- Method: CoInterpreter>>printMethodFieldForPrintContext: (in category 'debug printing') ----- printMethodFieldForPrintContext: aContext <inline: true> | meth | meth := objectMemory fetchPointer: MethodIndex ofObject: aContext. (self isMarriedOrWidowedContext: aContext) ifFalse: + [self printOopShortInner: meth. + (self methodHasCogMethod: meth) ifTrue: + [self space; print: '('; printHexnp: (self cogMethodOf: meth); print: ')']. + self cr] - [self shortPrintOop: meth. - (self methodHasCogMethod: meth) ifTrue: - [self space; printHexnp: (self cogMethodOf: meth)]] ifTrue: [(self methodHasCogMethod: meth) ifTrue: [self printHexnp: (self cogMethodOf: meth); space]. self shortPrintOop: meth]! Item was changed: ----- Method: SocketPluginSimulator>>primitiveSocketListenWithOrWithoutBacklog (in category 'primitives') ----- primitiveSocketListenWithOrWithoutBacklog "This is a hack to get around limitations in the simulated dispatch of smart syntax primitives." interpreterProxy methodArgumentCount = 2 ifTrue: ["self primitive: 'primitiveSocketListenOnPort' parameters: #(#Oop #SmallInteger)" ^self primitiveSocket: (interpreterProxy stackValue: 1) listenOnPort: (interpreterProxy stackIntegerValue: 0)]. "self primitive: 'primitiveSocketListenOnPortBacklog' parameters: #(#Oop #SmallInteger #SmallInteger)" + ^self primitiveSocket: (interpreterProxy stackValue: 2) + listenOnPort: (interpreterProxy stackIntegerValue: 1) - ^self primitiveSocket: (interpreterProxy stackValue: 1) - listenOnPort: (interpreterProxy stackIntegerValue: 0) backlogSize: (interpreterProxy stackIntegerValue: 0)!

1 0

VM Maker: VMMaker.oscog-eem.2566.mcz
by commits＠source.squeak.org 21 Sep '19

21 Sep '19

Eliot Miranda uploaded a new version of VMMaker to project VM Maker: http://source.squeak.org/VMMaker/VMMaker.oscog-eem.2566.mcz ==================== Summary ==================== Name: VMMaker.oscog-eem.2566 Author: eem Time: 18 September 2019, 9:23:01.791049 pm UUID: 71a20c6c-db1d-45e9-b85d-54aa2fd0cf8e Ancestors: VMMaker.oscog-eem.2565 Simulator: Complete the simulation support for the SocketTest suite tests. Make sure that the processes spawned by the SocketPluginSimulator run at a higher priority than the VM simulator and that they do not preempt it until they are all spawned. Fix a slip in SocketPluginSimulator>>closeAndDestroy:. =============== Diff against VMMaker.oscog-eem.2565 =============== Item was changed: ----- Method: SocketPluginSimulator>>closeAndDestroy: (in category 'initialize-release') ----- closeAndDestroy: socketHandle "c.f. Socket closeAndDestroy: timeoutSeconds" | fakeSocket | fakeSocket := Socket basicNew. [(fakeSocket primSocketConnectionStatus: socketHandle) = (Socket classPool at: #Connected) ifTrue: [fakeSocket primSocketCloseConnection: socketHandle]. fakeSocket primSocketAbortConnection: socketHandle; primSocketDestroy: socketHandle] on: SocketPrimitiveFailed + do: [:ex| Transcript cr; show: ex messageText]! - do: [:ex| Transcript cr; show: ex message]! Item was changed: ----- Method: SocketPluginSimulator>>map:to:type:register:spawning:and:and: (in category 'simulation support') ----- map: hostSocketHandle to: simSockPtr type: socketType register: semaphores spawning: blockOne and: blockTwo and: blockThree + | simSocket ourPriority | - | simSocket | "SQSocket is typedef struct { int sessionID; int socketType; void *privateSocketPtr; } SQSocket" simSocket := ByteArray new: (self sizeof: #SQSocket). simSocket unsignedLongAt: 1 put: interpreterProxy getThisSessionID; unsignedLongAt: 5 put: socketType. simSocket size = 12 ifTrue: [simSocket unsignedLongAt: 9 put: (fakeAddressCounter := fakeAddressCounter + 64)] ifFalse: [simSocket unsignedLongLongAt: 9 put: (fakeAddressCounter := fakeAddressCounter + 80)]. self assert: ((interpreterProxy isBytes: simSockPtr cPtrAsOop) and: [(interpreterProxy numBytesOf: simSockPtr cPtrAsOop) = simSocket size]). 1 to: simSocket size do: [:i| simSockPtr at: i - 1 put: (simSocket at: i)]. self assert: (self simSocketHandleFrom: simSockPtr) = simSocket. openSocketHandles add: hostSocketHandle. hostSocketToSimSocketMap at: hostSocketHandle put: simSocket. simSocketToHostSocketMap at: simSocket put: hostSocketHandle. externalSemaphores addAll: semaphores. "N.B. These don't need registering. Eventually they will end up waiting on semaphores that have been unregistered, and hence + will get garbage collected, along with these processes. But a) run them at one higher than the current priority, and + b) ensure that they all effectively resume at the same time; i.e. ensure this code doesn;t get preempted by the first + process to start running." + ourPriority := Processor activePriority. + [({blockOne. blockTwo. blockThree} + collect: [:b| b newProcess priority: ourPriority + 1; yourself]) + do: [:each| each resume]] + valueAt: ourPriority + 2! - will get garbage collected, along with these processes." - blockOne fork. - blockTwo fork. - blockThree fork! Item was added: + ----- Method: SocketPluginSimulator>>primitiveSocketListenWithOrWithoutBacklog (in category 'primitives') ----- + primitiveSocketListenWithOrWithoutBacklog + "This is a hack to get around limitations in the simulated dispatch of smart syntax primitives." + interpreterProxy methodArgumentCount = 2 ifTrue: + ["self primitive: 'primitiveSocketListenOnPort' parameters: #(#Oop #SmallInteger)" + ^self primitiveSocket: (interpreterProxy stackValue: 1) + listenOnPort: (interpreterProxy stackIntegerValue: 0)]. + "self primitive: 'primitiveSocketListenOnPortBacklog' parameters: #(#Oop #SmallInteger #SmallInteger)" + ^self primitiveSocket: (interpreterProxy stackValue: 1) + listenOnPort: (interpreterProxy stackIntegerValue: 0) + backlogSize: (interpreterProxy stackIntegerValue: 0)! Item was added: + ----- Method: SocketPluginSimulator>>sqSocket:ListenOnPort: (in category 'simulation') ----- + sqSocket: socketHandle ListenOnPort: portNumber + ^[Socket basicNew + primSocket: ((self hostSocketHandleFromSimSocketHandle: socketHandle) ifNil: [^self]) + listenOn: portNumber] + on: SocketPrimitiveFailed + do: [:ex| + interpreterProxy primitiveFail. + 0]! Item was added: + ----- Method: SocketPluginSimulator>>sqSocketLocalPort: (in category 'simulation') ----- + sqSocketLocalPort: socketHandleCArray + ^[Socket basicNew + primSocketLocalPort: ((self hostSocketHandleFromSimSocketHandle: socketHandleCArray) ifNil: [^false])] + on: SocketPrimitiveFailed + do: [:ex| + interpreterProxy primitiveFail. + false]!

3 2

VM Maker: VMMaker.oscog-eem.2565.mcz
by commits＠source.squeak.org 19 Sep '19

19 Sep '19

Eliot Miranda uploaded a new version of VMMaker to project VM Maker: http://source.squeak.org/VMMaker/VMMaker.oscog-eem.2565.mcz ==================== Summary ==================== Name: VMMaker.oscog-eem.2565 Author: eem Time: 18 September 2019, 6:10:19.899413 pm UUID: 4878d905-e31e-4027-9660-593effe84ebc Ancestors: VMMaker.oscog-cb.2564 Cogit: Eliminate a bogus assert in compileFullBlockMethodFrameBuild: (and explain why). Use #= to compare integers in isNonForwarderReceiver: instead of #==. Simulator: Also add some Socket simulation (sorry Nicolas, we're working on trhe same thing. My version is correct, because I created the Socket simulation; forgive me for discarding your changes). These get as far as primitiveSocketListenWithOrWithoutBacklog which breaks the hack used to invoke the smart-syntax methods. To be clear we should discard VMMaker.oscog-nice.2565 and continue from these changes. Fix some errors in initializing in-image compilation and include full blocks in the pc mapping tests. Nuke the now obsolete Cogit>>sizeof:, and fix four instances of a comment typo. =============== Diff against VMMaker.oscog-cb.2564 =============== Item was changed: ----- Method: Cogit class>>initializedInstanceForTests: (in category 'in-image compilation support') ----- initializedInstanceForTests: optionsDictionaryOrArray "Answer an instance of a Cogit suitable for running tests that has initialized its method zone (generated trampolines etc)" | cogit coInterpreter | cogit := self instanceForTests: optionsDictionaryOrArray. coInterpreter := CurrentImageCoInterpreterFacade forCogit: cogit. [cogit setInterpreter: coInterpreter; singleStep: true; + initializeCodeZoneFrom: 1024 upTo: coInterpreter memory byteSize. - initializeCodeZoneFrom: 1024 upTo: coInterpreter memory size. 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]. ^cogit! 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 n | cogit := self initializedInstanceForTests: optionsDictionaryOrArray. coInterpreter := cogit coInterpreter. n := -1. SystemNavigation new allSelect: + [:m| | retrym cm | + (m isQuick not + and: [aBlock value: m]) ifTrue: - [:m| | cm | - (m isQuick not - and: [aBlock value: m]) ifTrue: [(n := n + 1) \\ 10 = 0 ifTrue: [Transcript nextPut: $.; flush]. + retrym := true. + [(cm := cogit - cm := cogit cog: (coInterpreter oopForObject: m) + selector: (coInterpreter oopForObject: m selector)) isNil and: [retrym]] whileTrue: + [retrym := false. + cogit methodZone clearCogCompiledCode. + coInterpreter initializeObjectMap]. + cm + ifNil: [Transcript halt show: 'After 1 Cog compiled code compaction, still not able to generate the cog method...'] + ifNotNil: [cogit testPCMappingForCompiledMethod: m cogMethod: cm]. + m encoderClass supportsFullBlocks ifTrue: + [m nestedBlockMethods do: + [:bm| | retryb cbm | + (n := n + 1) \\ 10 = 0 ifTrue: [Transcript nextPut: $,; flush]. + retryb := true. + [(cbm := cogit + cogFullBlockMethod: (coInterpreter oopForObject: bm) + numCopied: bm numCopiedValues) isNil and: [retryb]] whileTrue: + [retryb := false. + cogit methodZone clearCogCompiledCode. + coInterpreter initializeObjectMap]. + cbm + ifNil: [Transcript show: 'After 1 Cog compiled code compaction, still not able to generate the cog block...'] + ifNotNil: [cogit testPCMappingForCompiledMethod: bm cogMethod: cbm]]]]. - selector: (coInterpreter oopForObject: m selector). - cm ifNil: - [cogit methodZone clearCogCompiledCode. - coInterpreter initializeObjectMap. - cm := cogit - cog: (coInterpreter oopForObject: m) - selector: (coInterpreter oopForObject: m selector). - cm ifNil: [Transcript show: 'After 1 Cog compiled code compaction, still not able to generate the cog method...' ] ]. - cm ifNotNil: - [cogit testPCMappingForCompiledMethod: m cogMethod: cm]]. false]! Item was removed: - ----- Method: Cogit>>sizeof: (in category 'translation support') ----- - sizeof: aCType - <doNotGenerate> - | bfc | - aCType == #BytecodeFixup ifTrue: - [bfc := self class bytecodeFixupClass. - ^bfc alignedByteSizeOf: bfc forClient: self]. - ^super sizeof: aCType! Item was changed: ----- Method: SocketPlugin>>primitiveSocket:getOptions: (in category 'primitives') ----- primitiveSocket: socket getOptions: optionName - | s optionNameStart optionNameSize returnedValue errorCode results | <var: #s type: #SocketPtr> <var: #optionNameStart type: #'char *'> + self primitive: 'primitiveSocketGetOptions' parameters: #(Oop Oop). - self primitive: 'primitiveSocketGetOptions' - parameters: #(Oop Oop). s := self socketValueOf: socket. interpreterProxy success: (interpreterProxy isBytes: optionName). optionNameStart := self cCoerce: (interpreterProxy firstIndexableField: optionName) to: #'char *'. optionNameSize := interpreterProxy slotSizeOf: optionName. interpreterProxy failed ifTrue: [^nil]. returnedValue := 0. errorCode := self sqSocketGetOptions: s optionNameStart: optionNameStart optionNameSize: optionNameSize + returnedValue: (self addressOf: returnedValue put: [:val| returnedValue := val]). - returnedValue: (self addressOf: returnedValue). results := interpreterProxy instantiateClass: interpreterProxy classArray indexableSize: 2. + interpreterProxy + storePointer: 0 ofObject: results withValue: errorCode asSmallIntegerObj; + storePointer: 1 ofObject: results withValue: returnedValue asSmallIntegerObj. + ^results! - interpreterProxy storePointer: 0 ofObject: results withValue: errorCode asSmallIntegerObj. - interpreterProxy storePointer: 1 ofObject: results withValue: returnedValue asSmallIntegerObj. - ^ results! Item was changed: ----- Method: SocketPlugin>>primitiveSocket:setOptions:value: (in category 'primitives') ----- primitiveSocket: socket setOptions: optionName value: optionValue "THIS BADLY NEEDS TO BE REWRITTEN TO TAKE Booleans AND Integers AS WELL AS (OR INSTEAD OF) Strings. It is only used with booleans and integers and parsing these back out of strings in sqSocketSetOptions:optionNameStart:optionNameSize:optionValueStart:optionValueSize:returnedValue: is STUPID." | s optionNameStart optionNameSize optionValueStart optionValueSize returnedValue errorCode results | <var: #s type: #SocketPtr> <var: #optionNameStart type: #'char *'> <var: #optionValueStart type: #'char *'> + self primitive: 'primitiveSocketSetOptions' parameters: #(Oop Oop Oop). - self primitive: 'primitiveSocketSetOptions' - parameters: #(Oop Oop Oop). s := self socketValueOf: socket. interpreterProxy success: (interpreterProxy isBytes: optionName). optionNameStart := self cCoerce: (interpreterProxy firstIndexableField: optionName) to: #'char *'. optionNameSize := interpreterProxy slotSizeOf: optionName. interpreterProxy success: (interpreterProxy isBytes: optionValue). optionValueStart:= self cCoerce: (interpreterProxy firstIndexableField: optionValue) to: #'char *'. optionValueSize := interpreterProxy slotSizeOf: optionValue. interpreterProxy failed ifTrue: [^nil]. returnedValue := 0. errorCode := self sqSocketSetOptions: s optionNameStart: optionNameStart optionNameSize: optionNameSize optionValueStart: optionValueStart optionValueSize: optionValueSize + returnedValue: (self addressOf: returnedValue put: [:val| returnedValue := val]). - returnedValue: (self addressOf: returnedValue). results := interpreterProxy instantiateClass: interpreterProxy classArray indexableSize: 2. + interpreterProxy + storePointer: 0 ofObject: results withValue: errorCode asSmallIntegerObj; + storePointer: 1 ofObject: results withValue: returnedValue asSmallIntegerObj. + ^results! - interpreterProxy storePointer: 0 ofObject: results withValue: errorCode asSmallIntegerObj. - interpreterProxy storePointer: 1 ofObject: results withValue: returnedValue asSmallIntegerObj. - ^ results! Item was added: + ----- Method: SocketPluginSimulator>>sqSocket:AcceptFrom:RecvBytes:SendBytes:SemaID:ReadSemaID:WriteSemaID: (in category 'simulation') ----- + sqSocket: sockPtr AcceptFrom: serverSocket RecvBytes: recvBufSize SendBytes: sendBufSize SemaID: semaIndex ReadSemaID: readSemaIndex WriteSemaID: writeSemaIndex + "Simulate the sqSocketAcceptFromRecvBytesSendBytesSemaIDReadSemaIDWriteSemaID function. + We descend beneath the Socket abstraction to simulate as accurately as possible." + | semaphoresAndIndexes semaphores indexes socketHandle | + semaphoresAndIndexes := Smalltalk newExternalSemaphores: 3. + semaphores := semaphoresAndIndexes first. + indexes := semaphoresAndIndexes second. + socketHandle := [Socket basicNew + primAcceptFrom: serverSocket + receiveBufferSize: recvBufSize + sendBufSize: sendBufSize + semaIndex: indexes first + readSemaIndex: indexes second + writeSemaIndex: indexes third] + on: SocketPrimitiveFailed + do: [:ex| + #failed]. + socketHandle == #failed ifTrue: + [interpreterProxy primitiveFail. + Smalltalk unregisterExternalObjects: semaphores. + ^self]. + "N.B. There is now a Processor yield in doSignalExternalSemaphores: every 100 virtual microseconds. + This allows these to make progress. Their job is to map a host signal into a signal of the relevant index." + self map: socketHandle + to: sockPtr + type: nil + register: semaphores + spawning: [[semaphores first wait. interpreterProxy signalSemaphoreWithIndex: semaIndex] repeat] + and: [[semaphores second wait. interpreterProxy signalSemaphoreWithIndex: readSemaIndex] repeat] + and: [[semaphores third wait. interpreterProxy signalSemaphoreWithIndex: writeSemaIndex] repeat]! Item was added: + ----- Method: SocketPluginSimulator>>sqSocket:ListenOnPort:BacklogSize:Interface: (in category 'simulation') ----- + sqSocket: socketHandle ListenOnPort: portNumber BacklogSize: backlog Interface: addr + ^[Socket basicNew + primSocket: ((self hostSocketHandleFromSimSocketHandle: socketHandle) ifNil: [^self]) + listenOn: portNumber + backlogSize: backlog + interface: (self netAddressAsByteArrayFromInt: addr)] + on: SocketPrimitiveFailed + do: [:ex| + interpreterProxy primitiveFail. + 0]! Item was added: + ----- Method: SocketPluginSimulator>>sqSocketError: (in category 'simulation') ----- + sqSocketError: socketHandleCArray + ^[Socket basicNew + primSocketError: ((self hostSocketHandleFromSimSocketHandle: socketHandleCArray) ifNil: [^false])] + on: SocketPrimitiveFailed + do: [:ex| + interpreterProxy primitiveFail. + false]! Item was added: + ----- Method: SocketPluginSimulator>>sqSocketGetOptions:optionNameStart:optionNameSize:returnedValue: (in category 'simulation') ----- + sqSocketGetOptions: socketHandle + optionNameStart: optionNameStart optionNameSize: optionNameSize + returnedValue: valuePtrBlock + "Simulate sqSocketGetOptionsoptionNameStartoptionNameSizereturnedValue. + Answer the error code and assign the value through valuePtrBlock." + | optionName | + optionName := self asString: optionNameStart size: optionNameSize. + ^[| errAndValue | + errAndValue := Socket basicNew + primSocket: ((self hostSocketHandleFromSimSocketHandle: socketHandle) ifNil: [^self]) + getOption: optionName. + valuePtrBlock at: 0 put: errAndValue second. + errAndValue first] + on: SocketPrimitiveFailed + do: [:ex| + interpreterProxy primitiveFail. + 0]! Item was added: + ----- Method: SocketPluginSimulator>>sqSocketRemoteAddress: (in category 'simulation') ----- + sqSocketRemoteAddress: socketHandleCArray + ^[| remoteAddress | + remoteAddress := Socket basicNew + primSocketRemoteAddress: ((self hostSocketHandleFromSimSocketHandle: socketHandleCArray) ifNil: [^false]). + "Effectively netAddressToInt: remoteAddress" + ((remoteAddress at: 4)) + + ((remoteAddress at: 3) <<8) + + ((remoteAddress at: 2) <<16) + + ((remoteAddress at: 1) <<24)] + on: SocketPrimitiveFailed + do: [:ex| + interpreterProxy primitiveFail. + false]! Item was added: + ----- Method: SocketPluginSimulator>>sqSocketSetOptions:optionNameStart:optionNameSize:optionValueStart:optionValueSize:returnedValue: (in category 'simulation') ----- + sqSocketSetOptions: socketHandle + optionNameStart: optionNameStart optionNameSize: optionNameSize optionValueStart: optionValueStart optionValueSize: optionValueSize + returnedValue: valuePtrBlock + "Simulate sqSocketSetOptionsoptionNameStartoptionNameSizeoptionValueStartoptionValueSizereturnedValue. + Answer the error code and assign the value through valuePtrBlock." + | optionName optionValue | + optionName := self asString: optionNameStart size: optionNameSize. + optionValue := self asString: optionValueStart size: optionValueSize. + ^[| errAndValue | + errAndValue := Socket basicNew + primSocket: ((self hostSocketHandleFromSimSocketHandle: socketHandle) ifNil: [^self]) + setOption: optionName + value: optionValue. + valuePtrBlock at: 0 put: errAndValue second. + errAndValue first] + on: SocketPrimitiveFailed + do: [:ex| + interpreterProxy primitiveFail. + 0]! Item was changed: ----- Method: StackToRegisterMappingCogit>>compileFullBlockMethodFrameBuild: (in category 'compile abstract instructions') ----- compileFullBlockMethodFrameBuild: numCopied <option: #SistaV1BytecodeSet> useTwoPaths ifTrue: [ "method with only inst var store, we compile only slow path for now" useTwoPaths := false. self cppIf: IMMUTABILITY ifTrue: [ needsFrame := true ] ]. needsFrame ifFalse: + ["it is OK for numCopied to be non-zero provided that the block does not actually use the copied values. + There are some blocks like this, e.g. that simply reference copied values to mark them as used for Slang. + See e.g. CroquetPlugin>>#primitiveGatherEntropy which contains the block [bufPtr. bufSize. false], + which the bytecode compiler optimizes to [false]." + false ifTrue: [self assert: numCopied = 0]. - [self assert: numCopied = 0. self compileFullBlockFramelessEntry: numCopied. self initSimStackForFramelessBlock: initialPC. ^self]. super compileFullBlockMethodFrameBuild: numCopied. self initSimStackForFramefulMethod: initialPC! Item was changed: ----- Method: StackToRegisterMappingCogit>>isNonForwarderReceiver: (in category 'testing') ----- isNonForwarderReceiver: reg "Do not inline (inBlock access)" ^self receiverIsInReceiverResultReg + and: [inBlock = 0 "Method, rcvr may be forwarder in blocks without inst var access." + and: [reg = ReceiverResultReg]]! - and: [inBlock == 0 "Method, rcvr may be forwarder in blocks without inst var access." - and: [reg == ReceiverResultReg]]! Item was changed: ----- Method: VMClass>>asByteArray: (in category 'C library extensions') ----- asByteArray: aStringOrStringIndex "aStringOrStringIndex is either a string or an address in the heap. + Create a ByteArray of the requested length from the bytes in the - Create a ByteArray of the requested length form the bytes in the heap starting at stringIndex." <doNotGenerate> | sz | aStringOrStringIndex isString ifTrue: [^aStringOrStringIndex asByteArray]. sz := self strlen: aStringOrStringIndex. ^self strncpy: (ByteArray new: sz) _: aStringOrStringIndex _: sz! Item was changed: ----- Method: VMClass>>asByteArray:size: (in category 'C library extensions') ----- asByteArray: baIndex size: baSize "baIndex is an address in the heap. Create a ByteArray of the requested length + from the bytes in the heap starting at baIndex." - form the bytes in the heap starting at baIndex." <doNotGenerate> ^self strncpy: (ByteArray new: baSize) _: baIndex _: baSize! Item was changed: ----- Method: VMClass>>asString: (in category 'C library extensions') ----- asString: aStringOrStringIndex "aStringOrStringIndex is either a string or an address in the heap. + Create a String of the requested length from the bytes in the - Create a String of the requested length form the bytes in the heap starting at stringIndex." <doNotGenerate> | sz | aStringOrStringIndex isString ifTrue: [^aStringOrStringIndex]. sz := self strlen: aStringOrStringIndex. ^self strncpy: (ByteString new: sz) _: aStringOrStringIndex _: sz! Item was changed: ----- Method: VMClass>>asString:size: (in category 'C library extensions') ----- asString: stringIndex size: stringSize "stringIndex is an address in the heap. Create a String of the requested length + from the bytes in the heap starting at stringIndex." - form the bytes in the heap starting at stringIndex." <doNotGenerate> ^self strncpy: (ByteString new: stringSize) _: stringIndex _: stringSize!

1 0

VM Maker: VMMaker.oscog-nice.2565.mcz
by commits＠source.squeak.org 18 Sep '19

18 Sep '19

Nicolas Cellier uploaded a new version of VMMaker to project VM Maker: http://source.squeak.org/VMMaker/VMMaker.oscog-nice.2565.mcz ==================== Summary ==================== Name: VMMaker.oscog-nice.2565 Author: nice Time: 18 September 2019, 11:18:16.74572 pm UUID: e21f30a8-9e63-4922-a3ae-c8f85d56e28e Ancestors: VMMaker.oscog-cb.2564 Add some simulation support (Socket) =============== Diff against VMMaker.oscog-cb.2564 =============== Item was added: + ----- Method: SocketPluginSimulator>>primitiveSocket:setOptions:value: (in category 'simulation') ----- + primitiveSocket: socket setOptions: optionName value: optionValue + "THIS BADLY NEEDS TO BE REWRITTEN TO TAKE Booleans AND Integers AS WELL AS (OR INSTEAD OF) Strings. + It is only used with booleans and integers and parsing these back out of strings in + sqSocketSetOptions:optionNameStart:optionNameSize:optionValueStart:optionValueSize:returnedValue: + is STUPID." + | s optionNameStart optionNameSize optionValueStart optionValueSize returnedValue errorCode results | + <var: #s type: #SocketPtr> + <var: #optionNameStart type: #'char *'> + <var: #optionValueStart type: #'char *'> + self primitive: 'primitiveSocketSetOptions' + parameters: #(Oop Oop Oop). + + s := self socketValueOf: socket. + interpreterProxy success: (interpreterProxy isBytes: optionName). + optionNameStart := self cCoerce: (interpreterProxy firstIndexableField: optionName) to: #'char *'. + optionNameSize := interpreterProxy slotSizeOf: optionName. + interpreterProxy success: (interpreterProxy isBytes: optionValue). + optionValueStart:= self cCoerce: (interpreterProxy firstIndexableField: optionValue) to: #'char *'. + optionValueSize := interpreterProxy slotSizeOf: optionValue. + + interpreterProxy failed ifTrue: [^nil]. + errorCode := 0. "?" + returnedValue := self sqSocket: s setOption: optionName value: optionValue. + returnedValue isNil ifTrue: [returnedValue := 0. errorCode := 1 "?"]. + + results := interpreterProxy instantiateClass: interpreterProxy classArray indexableSize: 2. + interpreterProxy storePointer: 0 ofObject: results withValue: errorCode asSmallIntegerObj. + interpreterProxy storePointer: 1 ofObject: results withValue: returnedValue asSmallIntegerObj. + ^ results! Item was added: + ----- Method: SocketPluginSimulator>>primitiveSocketListenWithOrWithoutBacklog (in category 'simulation') ----- + primitiveSocketListenWithOrWithoutBacklog + interpreterProxy methodArgumentCount = 2 + ifTrue:[^simulator primitiveSocketListenOnPort] + ifFalse:[^simulator primitiveSocketListenOnPortBacklog]! Item was added: + ----- Method: SocketPluginSimulator>>sqSocket:ListenOnPort: (in category 'simulation') ----- + sqSocket: s ListenOnPort: port + ^[Socket basicNew + primSocket: s listenOn: port] + on: SocketPrimitiveFailed + do: [:ex| + interpreterProxy primitiveFail. + 0]! Item was added: + ----- Method: SocketPluginSimulator>>sqSocket:ListenOnPort:BacklogSize:Interface: (in category 'simulation') ----- + sqSocket: s ListenOnPort: port BacklogSize: backlog Interface: addr + ^[Socket basicNew + primSocket: s listenOn: port backlogSize: backlog interface: addr] + on: SocketPrimitiveFailed + do: [:ex| + interpreterProxy primitiveFail. + 0]! Item was added: + ----- Method: SocketPluginSimulator>>sqSocket:setOption:value: (in category 'simulation') ----- + sqSocket: s setOption: aString value: aStringValue + ^[Socket basicNew + primSocket: s setOption: aString value: aStringValue] + on: SocketPrimitiveFailed + do: [:ex| + interpreterProxy primitiveFail. + 0]! Item was added: + ----- Method: SocketPluginSimulator>>sqSocketError: (in category 'simulation') ----- + sqSocketError: s + ^[Socket basicNew + primSocketError: s] + on: SocketPrimitiveFailed + do: [:ex| + interpreterProxy primitiveFail. + -1]! Item was added: + ----- Method: SocketPluginSimulator>>sqSocketRemoteAddress: (in category 'simulation') ----- + sqSocketRemoteAddress: s + ^[Socket basicNew + primSocketRemoteAddress: s] + on: SocketPrimitiveFailed + do: [:ex| + interpreterProxy primitiveFail. + 0]! Item was changed: ----- Method: VMClass>>strncpy:_:_: (in category 'C library simulation') ----- strncpy: dest _: src _: n <doNotGenerate> "implementation of strncpy(3). See e.g. https://manpages.debian.org/stretch/manpages-dev/strncpy.3.en.html The C version always takes an address; the simulation allows a String, ByteArray, CArray or address within the simulation object memory (Positive Integer)" | getBlock setBlock count | count := n. "Determine the source and destination access blocks based on the parameter type" getBlock := src isCollection ifTrue: [count := count min: src size. src isString ifTrue: [[ :idx | src basicAt: idx]] "basicAt: answers integers" ifFalse: [src class == ByteArray ifTrue: [[ :idx | src at: idx]]]] ifFalse: [src isInteger ifTrue: [[ :idx | self byteAt: src + idx - 1]] ifFalse: [src isCArray ifTrue: [[ :idx | src at: idx - 1]]]]. getBlock ifNil: [self error: 'unhandled type of source string']. setBlock := dest isCollection ifTrue: [dest isString ifTrue: [[ :idx | dest basicAt: idx put: (getBlock value: idx)]] "basicAt:put: stores integers" ifFalse: [dest class == ByteArray ifTrue: [[ :idx | dest at: idx put: (getBlock value: idx)]]]] ifFalse: + [dest isInteger + ifTrue: [[ :idx | self byteAt: dest + idx - 1 put: (getBlock value: idx)]] + ifFalse: [dest isCArray ifTrue: + [[ :idx | dest at: idx - 1 put: (getBlock value: idx)]]]]. - [dest isInteger ifTrue: - [[ :idx | self byteAt: dest + idx - 1 put: (getBlock value: idx)]]]. setBlock ifNil: [self error: 'unhandled type of destination string']. 1 to: count do: setBlock. "SVr4, 4.3BSD, C89, C99 require the remainder of the buffer be filled with nulls" getBlock := [:idx| 0]. count + 1 to: n do: setBlock. ^dest!

1 0

[OpenSmalltalk/opensmalltalk-vm] Restore support for large cursor on linux X11 VM (#424)
by Nicolas Cellier 18 Sep '19

18 Sep '19

Large cursor support is present but optional on X11 It is triggered by compiler define HAVE_LIBXRENDER It depends on availability of Xrender library which is an optional X extension Try to test presence of header and library in configure macro Note that it is also necessary to install the proper package e.g. on debian flavours sudo apt-get install libxrender-dev You can view, comment on, or merge this pull request online at: https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/424 -- Commit Summary -- * Restore support for large cursor on linux X11 VM -- File Changes -- M platforms/unix/vm-display-X11/acinclude.m4 (5) M scripts/ci/travis_install.sh (4) -- Patch Links -- https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/424.patch https://github.com/OpenSmalltalk/opensmalltalk-vm/pull/424.diff -- 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/pull/424

3 8

Still Failing: OpenSmalltalk/opensmalltalk-vm#1810 (Cog - f5de9a4)
by Travis CI 15 Sep '19

15 Sep '19

Build Update for OpenSmalltalk/opensmalltalk-vm ------------------------------------- Build: #1810 Status: Still Failing Duration: 40 mins and 28 secs Commit: f5de9a4 (Cog) Author: Nicolas Cellier Message: Let clang compile linux VM `CC=clang ./mvm` barks: a void function cannot return a value View the changeset: https://github.com/OpenSmalltalk/opensmalltalk-vm/compare/218e97a9b120...f5… View the full build log and details: https://travis-ci.org/OpenSmalltalk/opensmalltalk-vm/builds/585058733?utm_m… -- You can unsubscribe from build emails from the OpenSmalltalk/opensmalltalk-vm repository going to https://travis-ci.org/account/preferences/unsubscribe?repository=8795279&ut…. Or unsubscribe from *all* email updating your settings at https://travis-ci.org/account/preferences/unsubscribe?utm_medium=notificati…. Or configure specific recipients for build notifications in your .travis.yml file. See https://docs.travis-ci.com/user/notifications.

1 0

[OpenSmalltalk/opensmalltalk-vm] f5de9a: Let clang compile linux VM
by Nicolas Cellier 14 Sep '19

14 Sep '19

Branch: refs/heads/Cog Home: https://github.com/OpenSmalltalk/opensmalltalk-vm Commit: f5de9a4dcad610df4798e62f1f917ffb09ff8132 https://github.com/OpenSmalltalk/opensmalltalk-vm/commit/f5de9a4dcad610df47… Author: Nicolas Cellier <nicolas.cellier.aka.nice(a)gmail.com> Date: 2019-09-14 (Sat, 14 Sep 2019) Changed paths: M platforms/unix/vm-sound-pulse/sqUnixSoundPulseAudio.c Log Message: ----------- Let clang compile linux VM `CC=clang ./mvm` barks: a void function cannot return a value

1 0

Re: [Vm-dev] [squeak-dev] Difficult to debug VM crash with full blocks and Sista V1
by Phil B 14 Sep '19

14 Sep '19

I'm sure you're right (assuming the GUI tool still exists... the top answer is quite old.) Mainly I linked to the answer since it seems to cover a spectrum of approaches which might Eliot might find useful. On Sat, Sep 14, 2019 at 12:52 AM Nicola Mingotti <nmingotti(a)gmail.com> wrote: > > With high probability that GUI-tool is a wrapper to 'dummynet'. Anyhow, > for sure it will be faster to deploy. > > n. > > > On 9/13/19 9:32 PM, Phil B wrote: > > It sounds like simulating a terrible (dropping packets) or painfully slow > (kbps speeds) network connection might be a more controlled way to > reproduce the issue: > https://apple.stackexchange.com/questions/24066/how-to-simulate-slow-intern… > > On Fri, Sep 13, 2019 at 11:16 PM Eliot Miranda <eliot.miranda(a)gmail.com> > wrote: > >> >> Hi All, >> >> there is a VM bug in 64-bit Spur with the Sista V1 bytecode set and >> full blocks. The symptom is that when waiting for a remote Monticello >> repository to update and/or deliver a package version the system crashes in >> JITTED code after what appears to be some kind of wait. >> >> This is a reliably occurring bug b ut maddeningly difficult to >> reproduce. The bug reliably occurs when interacting with a remote >> rep[ository (e.g. http://source.squeak.org/VMMaker) when the server is >> "cold", and hence makes the image wait. Every time I have tried to repeat >> the failing sequence the crash has not occurre3d, I think because the >> server is now "hot" and serves up the version quickly. Today I even tried >> shutting down my machine for over an hour and rebooting. But I could not >> get the crash to occur even though it seems to me that every time I try it >> the first time in the4 day it does crash. >> >> This is an important bug to fix. If it cannot be fixed then full blocks >> and Sista V1 are not ready for use in the upcoming Squeak release. I am >> looking for help in debugging this. >> >> - is anyone else uising the 64-bit VM with full blocks and Sista V1 who >> sees hard VM crashes? If so, under what circumstances? >> >> - is it possible to flush caches in the http://source.squeak.org/VMMaker >> server, or could people tolerate me rebooting the server? >> >> - is there a way of introducing network delays in Mac OS that might help >> me induce the bug? >> >> - can anyone think of any other strategies I might take to try and >> reproduce this? >> >> I may have to try and reproduce e the bug in the simulator to have a >> chance of identifying the bug. Does anyone have a good enough mental model >> of the Monticello server interaction and have energy to help me figure this >> one out? >> >> Here is some information from the last crash I did see in the debugger >> (alas it is incomplete; there are a number of additional pieces of info I >> could have collected). >> >> (lldb) thr b >> >> * thread #1, queue = 'com.apple.main-thread', stop reason = >> EXC_BAD_INSTRUCTION (code=EXC_I386_INVOP, subcode=0x0) >> >> * frame #0: 0x000000010de5700a >> >> frame #1: 0x000000010dd7b174 >> >> frame #2: 0x000000010dd45f1c >> >> frame #3: 0x000000010dd44534 >> >> frame #4: 0x000000010dd44c60 >> >> (lldb) x/10i 0x000000010de5700a >> >> >> (lldb) call printStackCallStackOf($rbp) >> >> 0x7ffeefbdfc30 M Heap>upHeap: 0x11273ca90: a(n) Heap >> >> 0x7ffeefbdfc68 M Heap>add: 0x11273ca90: a(n) Heap >> >> 0x7ffeefbdfca0 M Delay class>scheduleDelay:from: 0x1123ebfb8: a(n) >> Delay class >> >> 0x7ffeefbdfcf0 M Delay class>handleTimerEvent 0x1123ebfb8: a(n) Delay >> class >> >> 0x7ffeefbdfd20 M Delay class>runTimerEventLoop 0x1123ebfb8: a(n) >> Delay class >> >> >> (lldb) x/10i 0x000000010dd7b174 >> >> 0x10dd7b174: 48 8b 55 10 movq 0x10(%rbp), %rdx >> >> 0x10dd7b178: 48 89 ec movq %rbp, %rsp >> >> 0x10dd7b17b: 5d popq %rbp >> >> 0x10dd7b17c: c2 10 00 retq $0x10 >> >> 0x10dd7b17f: cc int3 >> >> 0x10dd7b180: cc int3 >> >> 0x10dd7b181: cc int3 >> >> 0x10dd7b182: cc int3 >> >> 0x10dd7b183: cc int3 >> >> 0x10dd7b184: cc int3 >> >> (lldb) print whereIs(0x000000010dd7b174) >> >> (char *) $0 = 0x00000001000f83ff " is in generated methods" >> >> (lldb) call printCogMethodFor((void *)0x000000010dd7b174) >> >> 0x10dd7afc0 <-> 0x10dd7b198: method: 0x112f23c10 >> selector: 0x112232c20 add: >> >> (lldb) print whereIs(0x000000010de5700a) >> >> (char *) $1 = 0x00000001000f83ff " is in generated methods" >> >> (lldb) call printCogMethodFor((void *)0x000000010de5700a) >> >> 0x10de56ba0 <-> 0x10de57078: method: 0x1126ec218 >> prim 23856 selector: 0x7ffeefbf3d20 >> >> this method ends up being the fitted version of Delay class>> >> startTimerEventLoop >> _,,,^..^,,,_ >> best, Eliot >> > > >

1 0

Still Failing: OpenSmalltalk/opensmalltalk-vm#1809 (Cog - 218e97a)
by Travis CI 13 Sep '19

13 Sep '19

Build Update for OpenSmalltalk/opensmalltalk-vm ------------------------------------- Build: #1809 Status: Still Failing Duration: 38 mins and 47 secs Commit: 218e97a (Cog) Author: Nicolas Cellier Message: Fix Issue #426 - a BitBlt buffer overrun View the changeset: https://github.com/OpenSmalltalk/opensmalltalk-vm/compare/a3b48bd211e9...21… View the full build log and details: https://travis-ci.org/OpenSmalltalk/opensmalltalk-vm/builds/584740970?utm_m… -- You can unsubscribe from build emails from the OpenSmalltalk/opensmalltalk-vm repository going to https://travis-ci.org/account/preferences/unsubscribe?repository=8795279&ut…. Or unsubscribe from *all* email updating your settings at https://travis-ci.org/account/preferences/unsubscribe?utm_medium=notificati…. Or configure specific recipients for build notifications in your .travis.yml file. See https://docs.travis-ci.com/user/notifications.

1 0

← Newer
1
2
3
4
5
6
7
8
9
...
12
Older →

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

Vm-dev September 2019