Vm-dev March 2021

vm-dev@lists.squeakfoundation.org

17 participants
57 discussions

[OpenSmalltalk/opensmalltalk-vm] sqUnixEvent.c missing from platforms/unix/vm (#555)
by alexshendi 24 Mar '21

24 Mar '21

Hello OpenSmalltalk, Apparently the file "sqUnixEvent.c" is missing from the directory platforms/unix/vm. It is still needed for the X11 plugin. My quick fix was to checkout tag "202003021730", and to copy the file to the folder. -- 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/555

5 6

Build failed: opensmalltalk-vm 1.0.2418
by AppVeyor 24 Mar '21

24 Mar '21

1 0

[OpenSmalltalk/opensmalltalk-vm] 4b5f05: Improve the comment in platforms/unix/vm/include_u...
by Eliot Miranda 24 Mar '21

24 Mar '21

Branch: refs/heads/Cog Home: https://github.com/OpenSmalltalk/opensmalltalk-vm Commit: 4b5f053169427eb4e17587d5fa2cbffdddbcf292 https://github.com/OpenSmalltalk/opensmalltalk-vm/commit/4b5f053169427eb4e1… Author: Eliot Miranda <eliot.miranda(a)gmail.com> Date: 2021-03-24 (Wed, 24 Mar 2021) Changed paths: M platforms/unix/vm/include_ucontext.h Log Message: ----------- Improve the comment in platforms/unix/vm/include_ucontext.h [ci skip] Commit: 4f6b75b56b314c20b1842078872bd845d61344a4 https://github.com/OpenSmalltalk/opensmalltalk-vm/commit/4f6b75b56b314c20b1… Author: Eliot Miranda <eliot.miranda(a)gmail.com> Date: 2021-03-24 (Wed, 24 Mar 2021) Changed paths: A platforms/unix/vm/sqUnixEvent.c Log Message: ----------- Merge branch 'Cog' of https://github.com/OpenSmalltalk/opensmalltalk-vm into Cog Compare: https://github.com/OpenSmalltalk/opensmalltalk-vm/compare/d494240736f7...4f…

1 0

Build failed: opensmalltalk-vm 1.0.2417
by AppVeyor 22 Mar '21

22 Mar '21

1 0

[OpenSmalltalk/opensmalltalk-vm] d49424: Restore sqUnixEvent.c which was mistakenly deleted...
by David T Lewis 22 Mar '21

22 Mar '21

Branch: refs/heads/Cog Home: https://github.com/OpenSmalltalk/opensmalltalk-vm Commit: d494240736f7c0309e3e819784feb1d53ed0985a https://github.com/OpenSmalltalk/opensmalltalk-vm/commit/d494240736f7c0309e… Author: David T. Lewis <lewis(a)mail.msen.com> Date: 2021-03-21 (Sun, 21 Mar 2021) Changed paths: A platforms/unix/vm/sqUnixEvent.c Log Message: ----------- Restore sqUnixEvent.c which was mistakenly deleted in an earlier commit

1 0

regression: TestObjectsAsMethods crash in 4.19.5 (not in 4.16.7)
by stes＠PANDORA.BE 21 Mar '21

21 Mar '21

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 As can be seen in the screenshot at: http://docs.openindiana.org/handbook/community/squeak/index.html I'm deselecting the Tests-ObjectsAsMethods test (1 test), because it causes (reproducible) SIGSEGV on Solaris cc/OpenIndiana gcc. I think the segmentation fault is new in recent 4.19, I think it didn't happen a while ago in 4.16. I can test this as follows: when I install an older version squeak -version 4.16.7-3775 then I go into test runner: Tests-ObjectsAsMethods and select TestObjectsAsMethods that works in 4.16.7 Test Runner ... TestObjectsAsMethods 3 run, 3 passes, 0 expected failures, 0 failures, 0 errors, 0 unexpected passes But it stopped working in 4.19.x which is from I believe from: ContextInterpreter VMMaker-dtl.422 uuid: e72b95a0-204e-45a1-a4e4-3ac3c9e7a51a the interp.c file is automatically generated from VMMaker-dtl.422. It's reproducible in the sense that if I deselect all tests, and just select that one single test, I can repeatedly and reproducible SIGSEGV the VM. When I run the VM under a debugger: dbx: warning: Bad transition in runtime linker interface. CONSISTENT->CONSISTENT t@1 (l@1) signal SEGV (no mapping at the fault address) in interpret at line 9120 in file "interp.c" 9120 foo->freeContexts = longAt((newContext + (BASE_HEADER_SIZE)) + (0 << (SHIFT_FOR_WORD))); (dbx) where current thread: t@1 =>[1] interpret(), line 9120 in "interp.c" [2] main(argc = 1, argv = 0xfeffe250, envp = 0xfeffe258), line 1484 in "sqUnixMain.c" The above is from Solaris with cc/dbx but the same thing appears to happen for me on OpenIndiana with gcc/gdb. Unfortunately because the code of interp.c is automatically generated, it looks complicated to me and I don't see what's wrong with those " freeContext" code. The crash appears to be in: /* begin internalActivateNewMethod */ methodHeader = longAt((foo->newMethod + (BASE_HEADER_SIZE)) + (HeaderIndex << (SHIFT_FOR_WORD))); needsLarge = methodHeader & LargeContextBit; if ((needsLarge == 0) && (foo->freeContexts != NilContext)) { newContext = foo->freeContexts; /* begin setFreeContextsAfter: */ foo->freeContexts = longAt((newContext + (BASE_HEADER_SIZE)) + (0 << (SHIFT_FOR_WORD))); } else { /* begin externalizeIPandSP */ Has anyone seen this ? Also what is the test TestObjectsAsMethods actually doing please ? what is it testing ? Regards, David Stes -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQEcBAEBCAAGBQJgMnV5AAoJEAwpOKXMq1MafEgH/3EWQxaSaVm2g4r/2p99Wc21 P+U+ijqKpVTDfJ1smwV/GsgF0V8ZrZky0k7BzRDAyq3Gi/HGVm0e2bqOAKa1fo2Y MUS9JHOW4Lys+9qWgT0aLiWypjYlzThtYS0/Lfh013tsF1bBv2eppTceUyq/Zitv 6J0IFvDOspMN/zHwBw/ux3H6uR049boZ3mvk23sp3KIHDc2Yw2kF4TAXBwjZXmVO UFlIAC4EAahrtNZyLZSIBDbsXOl+wJGmQTsOIBG81pfSFpP6RBrIARcu6enZC3Wc bwsvWYADs49SKgVq3NBovfyzkZBIW30V82xlVKpOnp6A4FnOYXxQiVm9sNaOVXc= =34TJ -----END PGP SIGNATURE-----

5 20

VM Maker: VMMaker-dtl.423.mcz
by commits＠source.squeak.org 20 Mar '21

20 Mar '21

David T. Lewis uploaded a new version of VMMaker to project VM Maker: http://source.squeak.org/VMMaker/VMMaker-dtl.423.mcz ==================== Summary ==================== Name: VMMaker-dtl.423 Author: dtl Time: 20 March 2021, 12:27:52.291 pm UUID: b744c4a9-0a90-4104-8680-6e815df0f064 Ancestors: VMMaker-dtl.422 VMMaker 4.19.6 restore support for objects as methods. Extend the internal size of primitiveTable beyond MaxPrimitiveIndex and reinstall primitiveInvokeObjectAsMethod in the protected region. Allows invocation from lookupMethodInDictionary without exposing the method to the image as a numbered primitive. Also clean up a duplicate currentByteCode in class Interpreter declaration, not sure how that happened. =============== Diff against VMMaker-dtl.422 =============== Item was changed: ----- Method: ContextInterpreter class>>declareCVarsIn: (in category 'translation') ----- declareCVarsIn: aCCodeGenerator aCCodeGenerator addHeaderFile: '<setjmp.h>'. aCCodeGenerator var: #statGCTime type: #sqLong. aCCodeGenerator var: #statFullGCMSecs type: #sqLong. aCCodeGenerator var: #statIGCDeltaTime type: #sqLong. aCCodeGenerator var: #statIncrGCMSecs type: #sqLong. aCCodeGenerator var: #compilerHooks declareC: 'sqInt (*compilerHooks[' , (CompilerHooksSize + 1) printString , '])()'. aCCodeGenerator var: #interpreterVersion declareC: 'const char *interpreterVersion = "' , SmalltalkImage current datedVersion , ' [' , SmalltalkImage current lastUpdateString , ']"'. self declareCAsOop: {#newMethod. #activeContext. #theHomeContext} in: aCCodeGenerator. "Reinitialized at interpreter entry by #initializeImageFormatVersion" aCCodeGenerator var: #imageFormatVersionNumber declareC: 'sqInt imageFormatVersionNumber = 0'. "Declared here to prevent inclusion in foo struct by CCodeGeneratorGlobalStructure" aCCodeGenerator var: #imageFormatInitialVersion declareC: 'sqInt imageFormatInitialVersion = 0'. + aCCodeGenerator var: #primitiveTable declareC: 'void (*primitiveTable[', (MaxPrimitiveIndex + ProtectedPrimitiveCount + 1) asString, '])(void)'. - aCCodeGenerator var: #primitiveTable declareC: 'void (*primitiveTable[', MaxPrimitiveIndex asString, '])(void)'. aCCodeGenerator var: #primitiveTableDefaults declareC: 'void (*primitiveTableDefaults[' , (MaxPrimitiveIndex + 2) printString , '] )(void)= ' , self primitiveTableString. ! Item was changed: ----- Method: ContextInterpreter>>installPrimitiveTable (in category 'primitive table') ----- installPrimitiveTable <inline: false> + 0 to: MaxPrimitiveIndex + ProtectedPrimitiveCount do: [ :i | - 0 to: MaxPrimitiveIndex do: [ :i | primitiveTable at: i put: (primitiveTableDefaults at: i) ]. self updatePrimitiveTable. ! Item was changed: ----- Method: ContextInterpreter>>lookupMethodInDictionary: (in category 'message sending') ----- lookupMethodInDictionary: dictionary "This method lookup tolerates integers as Dictionary keys to support execution of images in which Symbols have been compacted out" | length index mask wrapAround nextSelector methodArray | <inline: true> length := objectMemory fetchWordLengthOf: dictionary. mask := length - SelectorStart - 1. (objectMemory isIntegerObject: messageSelector) ifTrue: [index := (mask bitAnd: (objectMemory integerValueOf: messageSelector)) + SelectorStart] ifFalse: [index := (mask bitAnd: (objectMemory hashBitsOf: messageSelector)) + SelectorStart]. "It is assumed that there are some nils in this dictionary, and search will stop when one is encountered. However, if there are no nils, then wrapAround will be detected the second time the loop gets to the end of the table." wrapAround := false. [true] whileTrue: [nextSelector := objectMemory fetchPointer: index ofObject: dictionary. nextSelector = objectMemory getNilObj ifTrue: [^ false]. nextSelector = messageSelector ifTrue: [methodArray := objectMemory fetchPointer: MethodArrayIndex ofObject: dictionary. newMethod := objectMemory fetchPointer: index - SelectorStart ofObject: methodArray. "Check if newMethod is a CompiledMethod." (objectMemory isCompiledMethod: newMethod) ifTrue: [primitiveIndex := self primitiveIndexOf: newMethod. primitiveIndex > MaxPrimitiveIndex ifTrue: ["If primitiveIndex is out of range, set to zero before putting in cache. This is equiv to primFail, and avoids the need to check on every send." primitiveIndex := 0]] ifFalse: ["indicate that this is no compiled method - use primitiveInvokeObjectAsMethod" + self flag: #primitiveInvokeObjectAsMethod. + primitiveIndex := MaxPrimitiveIndex + 1]. - primitiveIndex := 248]. ^ true]. index := index + 1. index = length ifTrue: [wrapAround ifTrue: [^ false]. wrapAround := true. index := SelectorStart]]! Item was changed: InterpreterPrimitives subclass: #Interpreter + instanceVariableNames: 'currentBytecode semaphoresUseBufferA semaphoresToSignalA semaphoresToSignalCountA semaphoresToSignalB semaphoresToSignalCountB extraVMMemory globalSessionID method instructionPointer stackPointer localIP localSP localReturnValue messageSelector primitiveFunctionPointer methodCache atCache lkupClass interruptPending savedWindowSize deferDisplayUpdates pendingFinalizationSignals interpreterProxy showSurfaceFn externalPrimitiveTable jmpBuf jmpDepth suspendedCallbacks suspendedMethods' + classVariableNames: 'AtCacheEntries AtCacheFixedFields AtCacheFmt AtCacheMask AtCacheOop AtCacheSize AtCacheTotalSize AtPutBase MaxExternalPrimitiveTableSize MaxJumpBuf MaxPrimitiveIndex ProtectedPrimitiveCount SemaphoresToSignalSize' - instanceVariableNames: 'currentBytecode semaphoresUseBufferA semaphoresToSignalA semaphoresToSignalCountA semaphoresToSignalB semaphoresToSignalCountB extraVMMemory globalSessionID method instructionPointer stackPointer localIP localSP localReturnValue messageSelector currentByteCode primitiveFunctionPointer methodCache atCache lkupClass interruptPending savedWindowSize deferDisplayUpdates pendingFinalizationSignals interpreterProxy showSurfaceFn externalPrimitiveTable jmpBuf jmpDepth suspendedCallbacks suspendedMethods' - classVariableNames: 'AtCacheEntries AtCacheFixedFields AtCacheFmt AtCacheMask AtCacheOop AtCacheSize AtCacheTotalSize AtPutBase MaxExternalPrimitiveTableSize MaxJumpBuf MaxPrimitiveIndex SemaphoresToSignalSize' poolDictionaries: 'VMMethodCacheConstants' category: 'VMMaker-Interpreter'! !Interpreter commentStamp: 'dtl 4/22/2016 22:14' prior: 0! This class is a complete implementation of the Smalltalk-80 virtual machine, derived originally from the Blue Book specification but quite different in some areas. ContextInterpreter is the Squeak interpreter VM as originally implemented by Dan Ingalls. StackInterpreter is the stack mapped interpreter by Eliot Miranda, which provides the basis for later Cog and Spur VMs.! Item was changed: ----- Method: Interpreter class>>declareCVarsIn: (in category 'translation') ----- declareCVarsIn: aCCodeGenerator aCCodeGenerator var: #interpreterProxy type: #'struct VirtualMachine*'. aCCodeGenerator var: 'semaphoresToSignalA' declareC: 'sqInt semaphoresToSignalA[' , (SemaphoresToSignalSize + 1) printString , ']'. aCCodeGenerator var: 'semaphoresToSignalB' declareC: 'sqInt semaphoresToSignalB[' , (SemaphoresToSignalSize + 1) printString , ']'. + self declareC: #(stackPointer instructionPointer method) - self declareC: #(instructionPointer method) as: #usqInt in: aCCodeGenerator. + self declareC: #(localIP localSP) - self declareC: #(stackPointer localIP localSP) as: #'char *' in: aCCodeGenerator. aCCodeGenerator var: #primitiveFunctionPointer declareC: 'void (*primitiveFunctionPointer)()'. aCCodeGenerator var: #methodCache declareC: 'long methodCache[MethodCacheSize + 1 /* ', (MethodCacheSize + 1) printString, ' */]'. aCCodeGenerator var: #atCache declareC: 'sqInt atCache[AtCacheTotalSize + 1 /* ', (AtCacheTotalSize + 1) printString, ' */]'. aCCodeGenerator var: #showSurfaceFn type: #'void *'. aCCodeGenerator var: #externalPrimitiveTable declareC: 'void (*externalPrimitiveTable[' , (MaxExternalPrimitiveTableSize + 1) printString , '])(void)'. aCCodeGenerator var: #jmpBuf declareC: 'jmp_buf jmpBuf[MaxJumpBuf + 1 /* ', (MaxJumpBuf + 1) printString, ' */]'. aCCodeGenerator var: #suspendedCallbacks declareC: 'usqInt suspendedCallbacks[MaxJumpBuf + 1 /* ', (MaxJumpBuf + 1) printString, ' */]'. aCCodeGenerator var: #suspendedMethods declareC: 'usqInt suspendedMethods[MaxJumpBuf + 1 /* ', (MaxJumpBuf + 1) printString, ' */]'. ! Item was changed: ----- Method: Interpreter class>>initializePrimitiveTable (in category 'initialization') ----- (excessive size, no diff calculated) Item was changed: ----- Method: VMMaker class>>versionString (in category 'version testing') ----- versionString "VMMaker versionString" + ^'4.19.6'! - ^'4.19.5'!

1 0

patch for allocateBuffer in AsyncFilePlugin
by stes＠PANDORA.BE 20 Mar '21

20 Mar '21

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA256 Hi, Is it please possible to fix a small issue in Squeak 4.19 (Classical VM). On my system the compiler is complaining that a function has no return type: file: platforms/unix/plugins/AsynchFilePlugin/sqUnixAsynchFile.c "platforms/unix/plugins/AsynchFilePlugin/sqUnixAsynchFile.c", line 145: warning: no explicit type given INLINE static allocateBuffer(struct FileBuf *buf, int size) The function is used only once in that file, as in : if (!allocateBuffer(&fp->buf, count)) Please apply my patch (in attach) to the classical VM ... Regards, David Stes -----BEGIN PGP SIGNATURE----- Version: GnuPG v2 iQEcBAEBCAAGBQJgU1sLAAoJEAwpOKXMq1Ma4gwH/Rwte3vJraeRSh7kj/EkXBL+ +Ctus32dJQR5echOLrVwzouifIEFZj8lSn64DGQiQaE1ZNczizn6HDaCeGvSB7K2 GNbmg8Fxq1Ghts5WX9QaeeGEotptcHDGGkZf9oaQnr+qrx2CbLLz1MFk9Lg/B+In sV2qL99vhFnS6kOjbpA926qj7Lb+bdsVg8ap7NO//t8UH6Irm7b4UMoRebJ68QpB YpuqtvtFaWmNr03Hsh5pcgGzOAoC+MITIL5umaIcs6nla1so0J9GJ/gnxfPKEety uo2kat03Onvjjn9FrwPi+qmmpq9hCDYMKFUqiV+OsYHWdfwGoXC6lKtSBgMvmGQ= =Ukc3 -----END PGP SIGNATURE-----

2 2

[Re: Copy Down to avoid Looking Up]
by ken.dickey＠whidbey.com 20 Mar '21

20 Mar '21

Thanks all for contributing to my education. Much to think about. Even with memory "on chip" these days for SOC, it seems IC (MonoMorphic) + PIC (PolyMorphic) win over copy-down based on CPU cache locality. For MegaMorphic case, one _might_ profitably specialize such selectors to contain/point-to caching dict of class->method as the lookups fault in. Have to test to see if caching the larger number of lookup results to avoid chronic dynamic lookup pays off. Simpler cache strategy to manage than copy-down, but equivalent based on Vanessa's squint test. Probably data dependent.. Thanks again to all! -KenD

1 0

Copy Down to avoid Looking Up
by ken.dickey＠whidbey.com 19 Mar '21

19 Mar '21

You can help me out. I keep getting this wacky idea which does not seem to go away. You can tell me what is wrong with it. OK. Here it is: To get rid of PICs and dynamic method lookups, each Class knows selectors of methods it defines. Each successful method definition is added not only to the method dictionary of the class, but copied down transitively to all subclasses, and their sub-subclasses, who have not defined a method with the same selector. At #DeepRehash time, _each_ method dictionary knows all of its selectors and creates a "perfect hash" function, or something close to it. So lookup is just into the specific method dictionary of the instance's class which yields the proper method (backstop method is DNU). Never a miss. No class hierarchy search for inherited methods. Memory is cheap these days, but selectors and compiled methods are mostly shared and the memory cost of method dictionaries is probably balanced out by elimination of PICS at the call sites. The UI tools need to keep some simple rules (e,g, transitively deleting methods from subclass method dictionaries), but these should be simple and obvious. The above seems simple enough, but nobody seems to be doing it. Why not? Thanks much for insights! -KenD

5 6

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

Vm-dev March 2021