Vm-dev May 2013

vm-dev@lists.squeakfoundation.org

20 participants
145 discussions

Corrupt BlockClosure
by Max Leske 23 May '13

23 May '13

Hi Marcus Roberto sent this mail to the Fuel-dev list. When we looked at the problem we noticed that serialization fails because of a corrupt BlockClosure. Since that's not really our territorry Mariano suggested to forward this to you. The image with the corrupt BlockClosure is available here: https://dl.dropboxusercontent.com/u/6281855/DFlow-SubscriptOutOfBounds.zip. To see the stack, right click on the only entry in the right window and click "export session". You'll find the corrupt BlockClosure at BlockClosure>>fuelAccept: @Roberto Is it possible to reproduce the stack with a new session? Or did this only happen once? Cheers, Max Begin forwarded message: > From: Mariano Martinez Peck <marianopeck(a)gmail.com> > Subject: Re: [Pharo-fuel] [Fuel] SubscriptOutOfBounds: 27 > Date: 23. Mai 2013 14:23:45 MESZ > To: The Fuel Project <pharo-fuel(a)lists.gforge.inria.fr> > Reply-To: The Fuel Project <pharo-fuel(a)lists.gforge.inria.fr> > > Indeed, it would be nice if you can known which CompiledMethod and blockclosure are having the problem. Not only the source code but the real bytecodes. > Also, can you isolate and just try to serialize them alone and reproduce the problem? In other words, a reproducible test case? :) > Thanks! > > > On Thu, May 23, 2013 at 8:41 AM, Max Leske <maxleske(a)gmail.com> wrote: > What's the compiled method? Is it a special one? Which selector in which class? > > Max > > On 23.05.2013, at 12:45, "roberto.minelli(a)usi.ch" <roberto.minelli(a)usi.ch> wrote: > > > Hi, > > > > While using Fuel in a 2.0 image to serialize some objects (including CompiledMethods) I'm getting this error: SubscriptOutOfBounds: 27. > > > > I spent hours in the debugger before writing this email, but I'm not able to figure out why this is happening. I attach the full stack of the error (STACK#1). > > > > I understood that it has something to do with a particular CompiledMethod in my image and/or a BlockClosure which for some reason cannot be printed (i.e., the printString returns <error in printString: evaluate "self printString" to debug>) and when I try to printString to debug I got the STACK#2 which in turn does not bring me to any possible solution. > > > > Hope some of you will have an intuition on that or at least tell me where to look or how to script a possible workaround. > > > > Thanks in advance, > > Roby > > > > ############################################################################################ > > ######################################### STACK#1 ########################################### > > ########################################################################################### > > CompiledMethod(Object)>>errorSubscriptBounds: > > CompiledMethod(Object)>>at: > > InstructionStream>>interpretNextInstructionFor: > > CompiledMethod>>abstractBytecodeMessageAt: in Block: [(InstructionStream new method: self pc: pc)... > > BlockClosure>>on:do: > > CompiledMethod>>abstractBytecodeMessageAt: > > BlockClosure>>blockCreationBytecodeMessage > > BlockClosure>>endPC > > BlockClosure>>abstractBytecodeMessagesDo: > > BlockClosure>>isClean > > BlockClosure>>shouldBeSubstitutedByCleanCopy > > BlockClosure>>fuelAccept: > > FLFullGeneralMapper(FLLightGeneralMapper)>>mapAndTrace: > > FLFullGlobalMapper>>mapAndTrace: in Block: [(anObject class == CompiledMethod... > > FLLargeIdentityDictionary>>at:ifAbsent: > > FLFullGlobalMapper>>mapAndTrace: > > FLPluggableSubstitutionMapper>>mapAndTrace: > > FLPluggableSubstitutionMapper>>mapAndTrace: > > FLAnalysis>>mapAndTrace: > > FLAnalysis>>run > > FLAnalyzer>>setDefaultAnalysis in Block: [:anObject | (FLAnalysis... > > FLAnalyzer>>analysisFor: > > FLSerialization>>analysisStep > > FLSerialization>>run > > DFSerializer(FLSerializer)>>setDefaultSerialization in Block: [:anObject :anEncoder | (FLSerialization... > > DFSerializer(FLSerializer)>>serialize:on: in Block: [:anEncoder | ... > > FLEncoder class>>on:globalEnvironment:do: in Block: [aBlock value: anEncoder] > > BlockClosure>>ensure: > > FLEncoder class>>on:globalEnvironment:do: > > DFSerializer(FLSerializer)>>serialize:on: > > > > > > ############################################################################################ > > ######################################### STACK#2 ########################################### > > ############################################################################################ > > Decompiler(Object)>>error: > > Decompiler>>decompileBlock: > > BlockClosure>>decompile > > BlockClosure>>printOn: > > BlockClosure(Object)>>printStringLimitedTo: in Block: [:s | self printOn: s] > > String class(SequenceableCollection class)>>streamContents:limitedTo: > > BlockClosure(Object)>>printStringLimitedTo: > > BlockClosure(Object)>>printString > > BlockClosure>>DoIt > > Compiler>>evaluate:in:to:notifying:ifFail:logged: > > SmalltalkEditor>>evaluateSelectionAndDo: in Block: [rcvr class evaluatorClass new... > > BlockClosure>>on:do: > > SmalltalkEditor>>evaluateSelectionAndDo: > > SmalltalkEditor>>evaluateSelection > > PluggableTextMorph>>doIt in Block: [textMorph editor evaluateSelection] > > PluggableTextMorph>>handleEdit: in Block: [result := editBlock value] > > TextMorphForEditView(TextMorph)>>handleEdit: > > PluggableTextMorph>>handleEdit: > > PluggableTextMorph>>doIt > > SmalltalkEditor class>>buildSmalltalkEditorKeymappingsOn: in Block: [:morph | morph doIt] > > BlockClosure>>cull: > > BlockClosure>>cull:cull: > > BlockClosure>>cull:cull:cull: > > KMCategoryTarget>>completeMatch:buffer: > > KMKeymap>>notifyCompleteMatchTo:buffer: in Block: [:l | l completeMatch: self buffer: aBuffer] > > Array(SequenceableCollection)>>do: > > KMKeymap>>notifyCompleteMatchTo:buffer: > > KMKeymap>>onMatchWith:notify:andDo: > > KMCategory>>onMatchWith:notify:andDo: in Block: [:entry | entry... > > Set>>do: > > > > > > > > _______________________________________________ > > Pharo-fuel mailing list > > Pharo-fuel(a)lists.gforge.inria.fr > > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-fuel > > _______________________________________________ > Pharo-fuel mailing list > Pharo-fuel(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-fuel > > > > -- > Mariano > http://marianopeck.wordpress.com > _______________________________________________ > Pharo-fuel mailing list > Pharo-fuel(a)lists.gforge.inria.fr > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-fuel

1 0

Re: [Vm-dev] VM Maker: VMMaker.oscog-eem.276.mcz
by Frank Shearar 22 May '13

22 May '13

On 21 March 2013 23:44, <commits(a)source.squeak.org> wrote: > > Eliot Miranda uploaded a new version of VMMaker to project VM Maker: > http://source.squeak.org/VMMaker/VMMaker.oscog-eem.276.mcz > > ==================== Summary ==================== > > Name: VMMaker.oscog-eem.276 > Author: eem > Time: 21 March 2013, 4:43:39.056 pm > UUID: 52d03a8f-11ec-4454-9c4a-ea1c383f339a > Ancestors: VMMaker.oscog-eem.275 > > Cogit: > Stop reporting EncounteredUnknownBytecode with an error message. > > Fix slip in ceSICMiss: that didn't link new PIC if an MNU case. > > Better document the calling conventions. > > Add an assert to genGetImplicitReceiverFor:forPush: that checks > the appropriate use of the IsNSSend annotation. > > Fix simulation bug in CogMethodSurrogate>>nextOpenPIC. > > Slang: > Nuke unused InterpreterSimulationObject. > > =============== Diff against VMMaker.oscog-eem.275 =============== I realise I've just reported a Cog build failure, but when might we see the unusedBytecode stuff land in a released VM? I'd love to play with MethodMassage, without having to custom build a Cog. (Although I suspect I could do just that for the moment...) frank

2 3

CogVM build failure
by Frank Shearar 20 May '13

20 May '13

Hi, The CogVM build has been failing since this job: http://build.squeak.org/job/CogVM/193/console Note the error message: 2013-05-14T03:17:13.709+01:00: install VMMaker.oscog and related packages 2013-05-14T03:17:13.709000001+01:00: install SharedPool-Speech 2013-05-14T03:17:14.407+01:00: install FFI-Pools 2013-05-14T03:17:15.447+01:00: install Alien 2013-05-14T03:17:17.991+01:00: install Balloon-Engine-Pools 2013-05-14T03:17:19.749+01:00: install VMMaker.oscog last object overwritten <snip> C stack backtrace: /var/lib/jenkins/workspace/CogVM/tmp/bin/../lib/squeak/4.0-2636/squeak[0x805ee4f] /var/lib/jenkins/workspace/CogVM/tmp/bin/../lib/squeak/4.0-2636/squeak(error+0x1f)[0x805f10d] /var/lib/jenkins/workspace/CogVM/tmp/bin/../lib/squeak/4.0-2636/squeak(instantiateClassindexableSize+0xc2)[0x807cee3] /var/lib/jenkins/workspace/CogVM/tmp/bin/../lib/squeak/4.0-2636/squeak[0x807f4b7] [0x7754df21] /var/lib/jenkins/workspace/CogVM/tmp/bin/../lib/squeak/4.0-2636/squeak(interpret+0x1ae)[0x8080116] /var/lib/jenkins/workspace/CogVM/tmp/bin/../lib/squeak/4.0-2636/squeak(main+0x4a5)[0x805f5e5] /lib/i686/nosegneg/libc.so.6(__libc_start_main+0xe6)[0xb7646ca6] /var/lib/jenkins/workspace/CogVM/tmp/bin/../lib/squeak/4.0-2636/squeak[0x805c5a1] Smalltalk stack dump: 0xbf9eadf4 M Dictionary(HashedCollection)>growTo: 0x7fe63598: a(n) Dictionary 0xbf9eae10 M Dictionary(HashedCollection)>grow 0x7fe63598: a(n) Dictionary 0xbf9eae28 M Dictionary(HashedCollection)>atNewIndex:put: 0x7fe63598: a(n) Dictionary 0xbf9eae50 M Dictionary>at:put: 0x7fe63598: a(n) Dictionary 0xbf9eae70 M [] in EncoderForV3PlusClosures(Encoder)>init:notifying: 0x7fe6353c: a(n) EncoderForV3PlusClosures 0xbf9eae9c M StackInterpreter class(ClassDescription)>instVarNamesAndOffsetsDo: 0x79cc903c: a(n) StackInterpreter class 0xbf9e4d08 M StackInterpreterPrimitives class(ClassDescription)>instVarNamesAndOffsetsDo: 0x79cd1a7c: a(n) StackInterpreterPrimitives class 0xbf9e4d30 M CoInterpreter class(ClassDescription)>instVarNamesAndOffsetsDo: 0x79cd24c8: a(n) CoInterpreter class 0xbf9e4d58 M CoInterpreterPrimitives class(ClassDescription)>instVarNamesAndOffsetsDo: 0x79cd69d4: a(n) CoInterpreterPrimitives class 0xbf9e4d80 M CoInterpreterMT class(ClassDescription)>instVarNamesAndOffsetsDo: 0x79cdc048: a(n) CoInterpreterMT class 0xbf9e4da8 M CogVMSimulator class(ClassDescription)>instVarNamesAndOffsetsDo: 0x79cdd490: a(n) CogVMSimulator class 0xbf9e4dc4 M CogVMSimulator class(Behavior)>variablesAndOffsetsDo: 0x79cdd490: a(n) CogVMSimulator class 0xbf9e4de4 M EncoderForV3PlusClosures(Encoder)>init:notifying: 0x7fe6353c: a(n) EncoderForV3PlusClosures 0xbf9e4e10 M Parser>parse:cue:noPattern:ifFail: 0x7fe632ec: a(n) Parser 0xbf9e4e38 M Compiler>translate:noPattern:ifFail: 0x7fe632d8: a(n) Compiler 0xbf9e4e5c M Compiler>compile:in:classified:notifying:ifFail: 0x7fe632d8: a(n) Compiler 0xbf9e4e8c M CogVMSimulator class(Behavior)>compile:classified:notifying:trailer:ifFail: 0x79cdd490: a(n) CogVMSimulator class 0xbf9ebd20 M MethodAddition>createCompiledMethod 0x79cdd7d0: a(n) MethodAddition 0xbf9ebd38 M [] in MCPackageLoader>basicLoad 0x79bfe9d8: a(n) MCPackageLoader <snip> frank

2 1

Incomprehendable behaviour of ContextPart class >>#contextOn:do:
by Lars Wassermann 19 May '13

19 May '13

Hello VM-dev, I habe problems understanding what exactly happens when I step through: (ContextPart contextOn: Error do: []) halt in a workspace. When I step over the parenthesis in the debugger, everything works kind of like expected (the context of contextOn:do: misses one temporary but the returned value is what the methods description promised). When I step into or through the code, an error is raised when executing bytecode 54 of #contextOn:do:, because the temps-array illegally has been poped (by bytecode 53). My question now is: Why does the normal vm not only break when runing this code, but also return the correct value (the first element of the last temporary). I appended a graphic trying to visualize what happens up to the miracle of three skipped bytecodes. I replicated the behavior in Squeak 4.0 and 4.4 images on interpreter VMs 3.7.7, 4.10, and a newer cog VM. A fix to the problem of stepping and comprehension as well as the missing temp in the frame would be changing the method to: contextOn: exceptionClass do: block "Create an #on:do: context that is ready to return from executing its receiver" | ctxt chain | ctxt := thisContext. [chain := thisContext sender cut: ctxt. ctxt push: nil. ctxt jump] on: exceptionClass do: block. "jump above will resume here without unwinding chain" ^ chain The difference is that after cutting off ctxt, we ensure that it is a top context by pushing a meaningless value. But this still doesn't help me understanding the original behaviour. Thank you and all the best Lars

3 4

Fwd: [Pharo-fuel] Possible collections problem
by Max Leske 16 May '13

16 May '13

I forgot to mention that I did use the builtin collection instead of the special Fuel collections as suggested by Mariano. That didn't change anything either. > Hi > > I'm forwarding this because I'd like to rule out a VM problem. Short summary: > I fork a squeak image and then serialize objects with Fuel. In roughly 40% of the cases the fork suddenly locks and consumes 100% CPU. The trace I most often see with gdb in that case is the one with > "#0 0x08060453 in updatePointersInRootObjectsFromto ()" at the top. > > The object processed when the lockup occufrs is always of class Timestamp, although that doesn't necessarily mean anything. Maybe it's more about the number of objects. > > I'm working on Debian, 32-bit and I can reproduce the problem with SqueakVM 4.4.7-2364 and 4.0.3-2202 (the newer ones wont run because of glibc). I haven't tried Cog yet. > > I also just checked that the problem occurs even if I don't serialize any timestamps (nor Process, Delay, Monitor, Semaphore; just to be sure). > > So if anyone has a clue as to what might be going on I'd really appreciate the help. > > Cheers, > Max > > Begin forwarded message: > >> From: Mariano Martinez Peck <marianopeck(a)gmail.com> >> Subject: Re: [Pharo-fuel] Possible collections problem >> Date: 15. Mai 2013 16:53:10 MESZ >> To: The Fuel Project <pharo-fuel(a)lists.gforge.inria.fr> >> Reply-To: The Fuel Project <pharo-fuel(a)lists.gforge.inria.fr> >> >> I cannot see anything in particular. Too many GC stuff. >> So I wouldn't spend more time trying to debug. I would try the none large collections. Then I would try with latest Cog and latest StackVM. >> >> >> >> On Wed, May 15, 2013 at 11:47 AM, Max Leske <maxleske(a)gmail.com> wrote: >> I've had several forks hanging around just now. Apart from one all of these were locked. I attached gdb and generated the c stack for all of them. Not sure if there's anything really interesting in there although clearly a lot of time is spent in GC and with creation of objects. That doesn't have to mean anything though. >> >> I haven't yet tried your suggestion Mariano. >> >> Cheers, >> Max >> >> >> [Thread debugging using libthread_db enabled] >> 0x08060453 in updatePointersInRootObjectsFromto () >> >> #0 0x08060453 in updatePointersInRootObjectsFromto () >> #1 0x08060a77 in mapPointersInObjectsFromto () >> #2 0x08060bb0 in incCompBody () >> #3 0x08065fa7 in incrementalGC () >> #4 0x080661a4 in sufficientSpaceAfterGC () >> #5 0x08069420 in primitiveNew () >> #6 0x0806de15 in interpret () >> #7 0x08073dfe in main () >> >> >> >> [Thread debugging using libthread_db enabled] >> 0x08060453 in updatePointersInRootObjectsFromto () >> >> #0 0x08060453 in updatePointersInRootObjectsFromto () >> #1 0x08060a77 in mapPointersInObjectsFromto () >> #2 0x08060bb0 in incCompBody () >> #3 0x08065fa7 in incrementalGC () >> #4 0x080661a4 in sufficientSpaceAfterGC () >> #5 0x08069420 in primitiveNew () >> #6 0x0806de15 in interpret () >> #7 0x08073dfe in main () >> >> >> >> >> [Thread debugging using libthread_db enabled] >> 0x08060453 in updatePointersInRootObjectsFromto () >> >> #0 0x08060453 in updatePointersInRootObjectsFromto () >> #1 0x08060a77 in mapPointersInObjectsFromto () >> #2 0x08060bb0 in incCompBody () >> #3 0x08065fa7 in incrementalGC () >> #4 0x080661a4 in sufficientSpaceAfterGC () >> #5 0x0806fed2 in clone () >> #6 0x08070095 in primitiveClone () >> #7 0x0806de15 in interpret () >> #8 0x08073dfe in main () >> >> >> [Thread debugging using libthread_db enabled] >> 0x08060453 in updatePointersInRootObjectsFromto () >> >> #0 0x08060453 in updatePointersInRootObjectsFromto () >> #1 0x08060a77 in mapPointersInObjectsFromto () >> #2 0x08060bb0 in incCompBody () >> #3 0x08065fa7 in incrementalGC () >> #4 0x080661a4 in sufficientSpaceAfterGC () >> #5 0x08069270 in primitiveNewWithArg () >> #6 0x0806de15 in interpret () >> #7 0x08073dfe in main () >> >> >> [Thread debugging using libthread_db enabled] >> 0xb76f0f68 in select () from /lib/libc.so.6 >> >> #0 0xb76f0f68 in select () from /lib/libc.so.6 >> #1 0x08070880 in aioPoll () >> #2 0xb762419e in ?? () from /usr/lib/squeak/4.0.3-2202//so.vm-display-X11 >> #3 0x08073595 in ioRelinquishProcessorForMicroseconds () >> #4 0x08061f24 in primitiveRelinquishProcessor () >> #5 0x0806de15 in interpret () >> #6 0x08073dfe in main () >> >> >> [Thread debugging using libthread_db enabled] >> 0x08060453 in updatePointersInRootObjectsFromto () >> >> #0 0x08060453 in updatePointersInRootObjectsFromto () >> #1 0x08060a77 in mapPointersInObjectsFromto () >> #2 0x08060bb0 in incCompBody () >> #3 0x08065fa7 in incrementalGC () >> #4 0x080661a4 in sufficientSpaceAfterGC () >> #5 0x08069420 in primitiveNew () >> #6 0x0806de15 in interpret () >> #7 0x08073dfe in main () >> >> >> >> [Thread debugging using libthread_db enabled] >> 0x08064e7e in markAndTrace () >> >> #0 0x08064e7e in markAndTrace () >> #1 0x0806593a in markPhase () >> #2 0x08065f60 in incrementalGC () >> #3 0x080661a4 in sufficientSpaceAfterGC () >> #4 0x0806fed2 in clone () >> #5 0x08070095 in primitiveClone () >> #6 0x0806de15 in interpret () >> #7 0x08073dfe in main () >> >> >> >> >> On 15.05.2013, at 13:59, Mariano Martinez Peck <marianopeck(a)gmail.com> wrote: >> >>> Ok. So, first thing you should try, is to replace the uses of LargeIdentityDictionary with IdentityDictionary. And LargeIdentitySet with IdentitySet. >>> If the problem disappears, then yes, there is something wrong with LargeCollections. If there is a problem with them, try updating VM, since they use a particular primitive. >>> Let us know! >>> >>> >>> On Tue, May 14, 2013 at 9:29 AM, Max Leske <maxleske(a)gmail.com> wrote: >>> >>> On 14.05.2013, at 13:52, Mariano Martinez Peck <marianopeck(a)gmail.com> wrote: >>> >>>> Hi Max. Question, are you able to reproduce the problem? >>> >>> Yes, but not "on purpose". The situation usually happens once or twice a day and then with consistent log entries. That's why I want to use gdb the next time it happens. >>> >>>> >>>> >>>> On Tue, Apr 30, 2013 at 3:57 PM, Max Leske <maxleske(a)gmail.com> wrote: >>>> Hi guys >>>> >>>> I have a problem serializing a graph. Sometimes (not always) the image will consume +/- 100% CPU and stop responding. I was able to pin the problem down a bit: >>>> - fails always in FLIteratingCluster>>registerIndexesOn: when called from FLFixedObjectCluster with class TimeStamp (this might not actually be relevan but it's consistent) >>>> - the problem *might* be in FLLargeIdentityDictionary>>at:put: (or further up the stack) >>>> >>>> I've done excessive logging to a file but even with flushing after every write the results are not consistent. Sometimes the image locks after leaving #at:put: sometimes it does somewhere in the middle or in #registerIndexesOn: (but remember: the logging might not be precise). >>>> >>>> It's probably not the size of the objects in the cluster (the graph is big but not overly large), since there are other clusters with more objects. >>>> >>>> What I did find is that the #grow operation for HashedCollections can be *very* slow, up to 20 seconds or more, at other times the snapshot runs through within no time. >>>> >>>> So here's my theory: There migth be a VM problem with HashedCollections. >>>> Now, the VM is a rather old one and I haven't had the possibility to test this with a newer one (but I'll probably have to). The version is Squeak4.0.3-2202 running on 32-bit Debian Squeeze. >>>> >>>> I'll try some more but if anyone has any ideas I'd be very happy :) >>>> >>>> Cheers, >>>> Max >>>> _______________________________________________ >>>> Pharo-fuel mailing list >>>> Pharo-fuel(a)lists.gforge.inria.fr >>>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-fuel >>>> >>>> >>>> >>>> -- >>>> Mariano >>>> http://marianopeck.wordpress.com >>>> _______________________________________________ >>>> Pharo-fuel mailing list >>>> Pharo-fuel(a)lists.gforge.inria.fr >>>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-fuel >>> >>> >>> _______________________________________________ >>> Pharo-fuel mailing list >>> Pharo-fuel(a)lists.gforge.inria.fr >>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-fuel >>> >>> >>> >>> >>> -- >>> Mariano >>> http://marianopeck.wordpress.com >>> _______________________________________________ >>> Pharo-fuel mailing list >>> Pharo-fuel(a)lists.gforge.inria.fr >>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-fuel >> >> >> _______________________________________________ >> Pharo-fuel mailing list >> Pharo-fuel(a)lists.gforge.inria.fr >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-fuel >> >> >> >> >> -- >> Mariano >> http://marianopeck.wordpress.com >> _______________________________________________ >> Pharo-fuel mailing list >> Pharo-fuel(a)lists.gforge.inria.fr >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-fuel >

1 0

VM Maker: VMMaker.oscog-tpr.291.mcz
by commits＠source.squeak.org 16 May '13

16 May '13

tim Rowledge uploaded a new version of VMMaker to project VM Maker: http://source.squeak.org/VMMaker/VMMaker.oscog-tpr.291.mcz ==================== Summary ==================== Name: VMMaker.oscog-tpr.291 Author: tpr Time: 15 May 2013, 6:08:02.276 pm UUID: ec89405f-12f5-4a55-98f7-b246fcbec660 Ancestors: VMMaker.oscog-tpr.290 Fix typo in SQ_USE_GLOBAL_STRUCT_REG. Insert batshit insane C macro double expansion lunacy to make use of -DUSE_GLOBAL_STRUCT_REG=r6 convert to register struct foo * foo asm("r6"); in final code. This does now build a Raspbian vm with global reg var for the pointer to the global struct. Doesn't seem to be any faster though. :-( =============== Diff against VMMaker.oscog-tpr.290 =============== Item was changed: ----- Method: CCodeGeneratorGlobalStructure>>emitCVariablesOn: (in category 'C code generator') ----- emitCVariablesOn: aStream "Store the global variable declarations on the given stream. Break logic into vars for structure and vars for non-structure." | structure nonstruct | structure := WriteStream on: (String new: 32768). nonstruct := WriteStream on: (String new: 32768). aStream nextPutAll: '/*** Variables ***/'; cr. structure nextPutAll: '#if SQ_USE_GLOBAL_STRUCT'; cr; nextPutAll: '# define _iss /* define in-struct static as void */'; cr; nextPutAll: 'static struct foo {'; cr; nextPutAll: '#else'; cr; nextPutAll: '# define _iss static'; cr; nextPutAll: '#endif'; cr. self buildSortedVariablesCollection do: [ :var | | decl varString inStruct target | target := (inStruct := self placeInStructure: (varString := var asString)) ifTrue: [structure] ifFalse: [nonstruct]. decl := variableDeclarations at: varString ifAbsent: ['sqInt ' , varString]. decl first == $# "support cgen var: #bytecodeSetSelector declareC: '#define bytecodeSetSelector 0' hack" ifTrue: [target nextPutAll: decl; cr] ifFalse: [self isGeneratingPluginCode ifTrue: [varString = 'interpreterProxy' ifTrue: "quite special..." [self preDeclareInterpreterProxyOn: target] ifFalse: [target nextPutAll: 'static ']] ifFalse: [(vmClass mustBeGlobal: varString) ifFalse: [target nextPutAll: (inStruct ifTrue: ['_iss '] ifFalse: ['static '])]]. target nextPutAll: decl; nextPut: $;; cr]]. structure nextPutAll: '#undef _iss'; cr; nextPutAll: '#if SQ_USE_GLOBAL_STRUCT'; cr; nextPutAll: ' } fum;'; cr; + nextPutAll: ' #if SQ_USE_GLOBAL_STRUCT_REG';cr; - nextPutAll: ' #if SQ_USE_GLOBAL_STRUCT_REGISTER';cr; nextPutAll: '# define DECL_MAYBE_SQ_GLOBAL_STRUCT /* using a global reg pointer */'; cr; nextPutAll: '# define DECL_MAYBE_VOLATILE_SQ_GLOBAL_STRUCT /* using a global reg pointer */'; cr; nextPutAll:'#else';cr; nextPutAll: '# define DECL_MAYBE_SQ_GLOBAL_STRUCT register struct foo * foo = &fum;'; cr; nextPutAll: '# define DECL_MAYBE_VOLATILE_SQ_GLOBAL_STRUCT volatile register struct foo * foo = &fum;'; cr; nextPutAll: '#endif';cr; nextPutAll: '# define GIV(interpreterInstVar) (foo->interpreterInstVar)'; cr; nextPutAll: '#else'; cr; nextPutAll: '# define DECL_MAYBE_SQ_GLOBAL_STRUCT /* oh, no mr bill!! */'; cr; nextPutAll: '# define DECL_MAYBE_VOLATILE_SQ_GLOBAL_STRUCT /* oh no, mr bill!! */'; cr; nextPutAll: '# define GIV(interpreterInstVar) interpreterInstVar'; cr; nextPutAll: '#endif'; cr. + "if the machine needs the fum structure defining locally, do it now; global register users don't need that, but DO need some batshit insane C macro fudging in order to convert the define of USE_GLOBAL_STRUCT_REG into a simple string to use in the asm clause below. Sigh." - "if the machine needs the fum structure defining locally, do it now; global register users don't need this" structure nextPutAll: '#if SQ_USE_GLOBAL_STRUCT'; cr; + nextPutAll: '#if SQ_USE_GLOBAL_STRUCT_REG';cr; + nextPutAll: '#define fooxstr(s) foostr(s)'; cr; + nextPutAll: '#define foostr(s) #s'; cr; + nextPutAll: 'register struct foo * foo asm(fooxstr(USE_GLOBAL_STRUCT_REG));'; cr; - nextPutAll: '#if SQ_USE_GLOBAL_STRUCT_REGISTER';cr; - nextPutAll: 'register struct foo * foo asm("USE_GLOBAL_STRUCT_REG")'; cr; nextPutAll: '#else'; cr; nextPutAll: 'static struct foo * foo = &fum;'; cr; nextPutAll: '#endif'; cr; nextPutAll: '#endif'; cr. aStream nextPutAll: structure contents; nextPutAll: nonstruct contents; cr!

1 0

VM Maker: VMMaker.oscog-tpr.290.mcz
by commits＠source.squeak.org 15 May '13

15 May '13

tim Rowledge uploaded a new version of VMMaker to project VM Maker: http://source.squeak.org/VMMaker/VMMaker.oscog-tpr.290.mcz ==================== Summary ==================== Name: VMMaker.oscog-tpr.290 Author: tpr Time: 15 May 2013, 1:05:58.213 pm UUID: 31707f84-deff-4c57-8d55-df4e7407682f Ancestors: VMMaker.oscog-tpr.289 Clean up some global struct code; - remove need to build separate sources to accomodate use of register global ptr to struct - clean up recording/testing for use of global struct in methods - define USE_GLOBAL_STRUCT_REG to use the global struct reg; right now machine factors require a different define for each machine/OS. Maybe this can be improved =============== Diff against VMMaker.oscog-tpr.289 =============== Item was changed: ----- Method: CCodeGenerator>>checkForGlobalUsage:in: (in category 'utilities') ----- checkForGlobalUsage: vars in: aTMethod vars do: [:var | (variables includes: var) ifTrue: "find the set of method names using this global var" [(globalVariableUsage at: var ifAbsentPut: [Set new]) add: aTMethod selector]]. + aTMethod clearReferencesToGlobalStruct. - aTMethod referencesGlobalStructMakeZero. (aTMethod locals select: [:l| self reservedWords includes: l]) do: [:l| | em | em := aTMethod definingClass name, '>>', aTMethod selector, ' has variable that is a C reserved word: ', l. self error: em. self logger cr; nextPutAll: em; cr; flush]! Item was changed: ----- Method: CCodeGeneratorGlobalStructure>>checkForGlobalUsage:in: (in category 'utilities') ----- checkForGlobalUsage: vars in: aTMethod "override to handle global struct needs" super checkForGlobalUsage: vars in: aTMethod. + - "if localStructDef is false, we don't ever need to include a reference to it in a function" - localStructDef ifFalse:[^self]. vars asSet do: [:var | "if any var is global and in the global var struct + tell the TMethod it may be refering to the struct, depending upon the #defines" - tell the TMethod it will be refering to the struct" ((variables includes: var) and: [self placeInStructure: var]) ifTrue: + [aTMethod referencesGlobalStruct]]! - [aTMethod referencesGlobalStructIncrementBy: (vars occurrencesOf: var)]]! Item was changed: ----- Method: CCodeGeneratorGlobalStructure>>emitCCodeOn:doInlining:doAssertions: (in category 'C code generator') ----- emitCCodeOn: aStream doInlining: inlineFlag doAssertions: assertionFlag super emitCCodeOn: aStream doInlining: inlineFlag doAssertions: assertionFlag. + "we add an initialiser for the pointer to the global struct; " + aStream + cr; + nextPutAll: 'void initGlobalStructure(void) {';cr; + nextPutAll: '#if SQ_USE_GLOBAL_STRUCT_REG';cr; + nextPutAll: 'foo = &fum;' ; cr; + nextPutAll: '#endif'; cr; + nextPutAll:'}'; + cr! - "if the machine needs the globals structure defined locally in the interp.c file, don't add the folowing function" - localStructDef ifFalse:[self emitStructureInitFunctionOn: aStream]! Item was changed: ----- Method: CCodeGeneratorGlobalStructure>>emitCVariablesOn: (in category 'C code generator') ----- emitCVariablesOn: aStream "Store the global variable declarations on the given stream. Break logic into vars for structure and vars for non-structure." | structure nonstruct | structure := WriteStream on: (String new: 32768). nonstruct := WriteStream on: (String new: 32768). aStream nextPutAll: '/*** Variables ***/'; cr. structure nextPutAll: '#if SQ_USE_GLOBAL_STRUCT'; cr; nextPutAll: '# define _iss /* define in-struct static as void */'; cr; nextPutAll: 'static struct foo {'; cr; nextPutAll: '#else'; cr; nextPutAll: '# define _iss static'; cr; nextPutAll: '#endif'; cr. self buildSortedVariablesCollection do: [ :var | | decl varString inStruct target | target := (inStruct := self placeInStructure: (varString := var asString)) ifTrue: [structure] ifFalse: [nonstruct]. decl := variableDeclarations at: varString ifAbsent: ['sqInt ' , varString]. decl first == $# "support cgen var: #bytecodeSetSelector declareC: '#define bytecodeSetSelector 0' hack" ifTrue: [target nextPutAll: decl; cr] ifFalse: [self isGeneratingPluginCode ifTrue: [varString = 'interpreterProxy' ifTrue: "quite special..." [self preDeclareInterpreterProxyOn: target] ifFalse: [target nextPutAll: 'static ']] ifFalse: [(vmClass mustBeGlobal: varString) ifFalse: [target nextPutAll: (inStruct ifTrue: ['_iss '] ifFalse: ['static '])]]. target nextPutAll: decl; nextPut: $;; cr]]. structure nextPutAll: '#undef _iss'; cr; nextPutAll: '#if SQ_USE_GLOBAL_STRUCT'; cr; nextPutAll: ' } fum;'; cr; + nextPutAll: ' #if SQ_USE_GLOBAL_STRUCT_REGISTER';cr; + nextPutAll: '# define DECL_MAYBE_SQ_GLOBAL_STRUCT /* using a global reg pointer */'; cr; + nextPutAll: '# define DECL_MAYBE_VOLATILE_SQ_GLOBAL_STRUCT /* using a global reg pointer */'; cr; + nextPutAll:'#else';cr; nextPutAll: '# define DECL_MAYBE_SQ_GLOBAL_STRUCT register struct foo * foo = &fum;'; cr; nextPutAll: '# define DECL_MAYBE_VOLATILE_SQ_GLOBAL_STRUCT volatile register struct foo * foo = &fum;'; cr; + nextPutAll: '#endif';cr; nextPutAll: '# define GIV(interpreterInstVar) (foo->interpreterInstVar)'; cr; nextPutAll: '#else'; cr; nextPutAll: '# define DECL_MAYBE_SQ_GLOBAL_STRUCT /* oh, no mr bill!! */'; cr; nextPutAll: '# define DECL_MAYBE_VOLATILE_SQ_GLOBAL_STRUCT /* oh no, mr bill!! */'; cr; nextPutAll: '# define GIV(interpreterInstVar) interpreterInstVar'; cr; nextPutAll: '#endif'; cr. + "if the machine needs the fum structure defining locally, do it now; global register users don't need this" + structure + nextPutAll: '#if SQ_USE_GLOBAL_STRUCT'; cr; + nextPutAll: '#if SQ_USE_GLOBAL_STRUCT_REGISTER';cr; + nextPutAll: 'register struct foo * foo asm("USE_GLOBAL_STRUCT_REG")'; cr; + nextPutAll: '#else'; cr; + nextPutAll: 'static struct foo * foo = &fum;'; cr; + nextPutAll: '#endif'; cr; + nextPutAll: '#endif'; cr. - "if the machine needs the fum structure defining locally, do it now" - localStructDef ifTrue: - [structure - nextPutAll: '#if SQ_USE_GLOBAL_STRUCT'; cr; - nextPutAll: 'static struct foo * foo = &fum;'; cr; - nextPutAll: '#endif'; cr]. aStream nextPutAll: structure contents; nextPutAll: nonstruct contents; cr! Item was changed: ----- Method: CCodeGeneratorGlobalStructure>>emitGlobalStructFlagOn: (in category 'C code generator') ----- emitGlobalStructFlagOn: aStream + "Depending upon the value of structDefDefine (See also #structDefDefine:), define SQ_USE_GLOBAL_STRUCT before including the header. Also derive the flag for using the global register; define USE_GLOBAL_STRUCT_REG to do so" - "Define SQ_USE_GLOBAL_STRUCT before including the header." aStream nextPutAll: '#if '; nextPutAll: structDefDefine; cr; nextPutAll: '# define SQ_USE_GLOBAL_STRUCT 1'; cr; nextPutAll: '#else'; cr; nextPutAll: '# define SQ_USE_GLOBAL_STRUCT 0'; cr; nextPutAll: '#endif'; cr; + nextPutAll: '#if USE_GLOBAL_STRUCT_REG '; cr; + nextPutAll: '# define SQ_USE_GLOBAL_STRUCT_REG 1'; cr; + nextPutAll: '#else'; cr; + nextPutAll: '# define SQ_USE_GLOBAL_STRUCT_REG 0'; cr; + nextPutAll: '#endif'; cr; cr! Item was removed: - ----- Method: CCodeGeneratorGlobalStructure>>emitStructureInitFunctionOn: (in category 'C code generator') ----- - emitStructureInitFunctionOn: aStream - "For the VM using a global struct for most of the global vars (useful for ARM and PPC so far), append the initGlobalStructure() function" - aStream - cr; - nextPutAll: 'void initGlobalStructure(void) {foo = &fum;}'; - cr! Item was removed: - ----- Method: CCodeGeneratorGlobalStructure>>globalStructDefined: (in category 'C code generator') ----- - globalStructDefined: aBool - localStructDef := aBool! Item was changed: ----- Method: CCodeGeneratorGlobalStructure>>initialize (in category 'C code generator') ----- initialize super initialize. + localStructDef := nil. "ignored ivar - no longer used" - localStructDef := false. structDefDefine := '1'! Item was changed: ----- Method: CCodeGeneratorGlobalStructure>>structDefDefine: (in category 'initialize-release') ----- structDefDefine: aString + "set the string that will appear in the C file to define whether or not to use the global struct; reasonable values would be: + 'USE_GLOBAL_STRUCT' - which would be defined in a header or makefile + '0' - which would mean never do it + '1' - which would mean always do it" structDefDefine := aString! Item was changed: ----- Method: CrossPlatformVMMaker>>createCodeGenerator (in category 'initialize') ----- createCodeGenerator "Set up a CCodeGenerator for this VMMaker - A cross platform tree leaves it up to the makefiles to decide whether to use the global struct or not." ^CCodeGeneratorGlobalStructure new initialize; - globalStructDefined: true; structDefDefine: 'USE_GLOBAL_STRUCT'; logger: logger; options: optionsDictionary; yourself! Item was changed: ----- Method: MacOSPowerPCOS9VMMaker>>createCodeGenerator (in category 'initialize') ----- createCodeGenerator "Set up a CCodeGenerator for this VMMaker - Mac OS uses the global struct and local def of the structure. The global struct/loca def regime appears to be about 10% faster than the default regime for Smalltalk-intensive macro benchmarks for both the Intel and gcc 4.0 compiler on x86. eem 12/10/2008 14:34 2.16 GHz Intel Core Duo MacBook Pro Mac OS X 10.4.11" ^CCodeGeneratorGlobalStructure new initialize; - globalStructDefined: true; structDefDefine: '1'; "structDefDefine: 'defined(PPC) || defined(_POWER) || defined(__powerpc__) || defined(__ppc__)';" logger: logger; yourself! Item was changed: ----- Method: RiscOSVMMaker>>createCodeGenerator (in category 'initialize') ----- createCodeGenerator "set up a CCodeGenerator for this VMMaker - RiscOS uses the global struct and no local def of the structure because of the global register trickery" ^CCodeGeneratorGlobalStructure new initialize; - globalStructDefined: false; logger: logger; yourself! Item was added: + ----- Method: TMethod>>clearReferencesToGlobalStruct (in category 'accessing') ----- + clearReferencesToGlobalStruct + globalStructureBuildMethodHasFoo := false! Item was changed: ----- Method: TMethod>>emitCLocalsOn:generator: (in category 'C code generation') ----- emitCLocalsOn: aStream generator: aCodeGen "Emit a C function header for this method onto the given stream." | volatileVariables | volatileVariables := properties includesKey: #volatile. + self refersToGlobalStruct ifTrue: - self globalStructureBuildMethodHasFoo > 1 ifTrue: [aStream next: 3 put: Character space; "there's already an opening ${ on this line; see sender" nextPutAll: (volatileVariables ifTrue: ['DECL_MAYBE_VOLATILE_SQ_GLOBAL_STRUCT'] ifFalse: ['DECL_MAYBE_SQ_GLOBAL_STRUCT'])]. aStream cr. locals isEmpty ifFalse: [(aCodeGen sortStrings: locals) do: [ :var | aStream next: 4 put: Character space. volatileVariables ifTrue: [aStream nextPutAll: #volatile; space]. aStream nextPutAll: (self declarationAt: var); nextPut: $;; cr]. aStream cr]! Item was removed: - ----- Method: TMethod>>globalStructureBuildMethodHasFoo (in category 'accessing') ----- - globalStructureBuildMethodHasFoo - ^globalStructureBuildMethodHasFoo! Item was added: + ----- Method: TMethod>>referencesGlobalStruct (in category 'accessing') ----- + referencesGlobalStruct + globalStructureBuildMethodHasFoo := true! Item was removed: - ----- Method: TMethod>>referencesGlobalStructIncrementBy: (in category 'accessing') ----- - referencesGlobalStructIncrementBy: value - globalStructureBuildMethodHasFoo := globalStructureBuildMethodHasFoo + value.! Item was removed: - ----- Method: TMethod>>referencesGlobalStructMakeZero (in category 'accessing') ----- - referencesGlobalStructMakeZero - globalStructureBuildMethodHasFoo := 0! Item was added: + ----- Method: TMethod>>refersToGlobalStruct (in category 'accessing') ----- + refersToGlobalStruct + ^globalStructureBuildMethodHasFoo! Item was changed: ----- Method: TMethod>>setSelector:definingClass:args:locals:block:primitive:properties:comment: (in category 'initialization') ----- setSelector: sel definingClass: class args: argList locals: localList block: aBlockNode primitive: aNumber properties: methodProperties comment: aComment "Initialize this method using the given information." selector := sel. definingClass := class. returnType := #sqInt. "assume return type is long for now" args := argList asOrderedCollection collect: [:arg | arg key]. locals := (localList collect: [:arg | arg key]) asSet. declarations := Dictionary new. self addTypeForSelf. primitive := aNumber. properties := methodProperties. comment := aComment. parseTree := aBlockNode. "hack; allows nodes to find their parent, etc" parseTree := aBlockNode asTranslatorNodeIn: self. labels := OrderedCollection new. complete := false. "set to true when all possible inlining has been done" export := self extractExportDirective. static := self extractStaticDirective. canAsmLabel := self extractLabelDirective. self extractSharedCase. self removeFinalSelfReturn. "must preceed recordDeclarations because this may set returnType" self recordDeclarations. + globalStructureBuildMethodHasFoo := false! - globalStructureBuildMethodHasFoo := 0! Item was changed: ----- Method: UnixVMMaker>>createCodeGenerator (in category 'initialisation') ----- createCodeGenerator ^CCodeGeneratorGlobalStructure new initialize; - globalStructDefined: true; logger: logger; yourself! Item was changed: ----- Method: Win32VMMaker>>createCodeGenerator (in category 'initialize') ----- createCodeGenerator "Set up a CCodeGenerator for this VMMaker - On Windows we use the gcc 2.95.x compiler which does better without the global struct." ^CCodeGeneratorGlobalStructure new initialize; - globalStructDefined: true; structDefDefine: '0'; "structDefDefine: 'defined(PPC) || defined(_POWER) || defined(__powerpc__) || defined(__ppc__)';" logger: logger; yourself!

1 0

VM Maker: VMMaker.oscog-tpr.289.mcz
by commits＠source.squeak.org 15 May '13

15 May '13

tim Rowledge uploaded a new version of VMMaker to project VM Maker: http://source.squeak.org/VMMaker/VMMaker.oscog-tpr.289.mcz ==================== Summary ==================== Name: VMMaker.oscog-tpr.289 Author: tpr Time: 14 May 2013, 3:19:03.426 pm UUID: 522ae12f-76a5-465d-9b22-0a418d901ad2 Ancestors: VMMaker.oscog-eem.288 USe asmLabel: false pragma to stop asm lables being put into floatValueOf and safeObejctAfter so that gcc -o3 doesn't fuck up; it tends to put the label in twice for some reason and then complains about that. Typical... =============== Diff against VMMaker.oscog-eem.288 =============== Item was changed: ----- Method: NewObjectMemory>>safeObjectAfter: (in category 'object enumeration') ----- safeObjectAfter: oop "Return the object or start of free space immediately following the given object or free chunk in memory. Return freeStart when enumeration is complete. This is for assertion checking only." | sz | + <asmLabel: false> (self isFreeObject: oop) ifTrue: [sz := self sizeOfFree: oop] ifFalse: [sz := self sizeBitsOf: oop]. ^oop + sz >= freeStart ifTrue: [freeStart] ifFalse: [self oopFromChunk: oop + sz]! Item was changed: ----- Method: StackInterpreter>>floatValueOf: (in category 'utilities') ----- floatValueOf: oop "Answer the C double precision floating point value of the argument, or fail if it is not a Float, and answer 0. Note: May be called by translated primitive code." | isFloat result | + <asmLabel: false> <returnTypeC: #double> <var: #result type: #double> isFloat := self isInstanceOfClassFloat: oop. isFloat ifTrue: [self cCode: '' inSmalltalk: [result := Float new: 2]. objectMemory fetchFloatAt: oop + BaseHeaderSize into: result. ^result]. self primitiveFail. ^0.0!

1 0

VM Maker: CMakeVMMaker-EstebanLorenzano.204.mcz
by commits＠source.squeak.org 14 May '13

14 May '13

Esteban Lorenzano uploaded a new version of CMakeVMMaker to project VM Maker: http://source.squeak.org/VMMaker/CMakeVMMaker-EstebanLorenzano.204.mcz ==================== Summary ==================== Name: CMakeVMMaker-EstebanLorenzano.204 Author: EstebanLorenzano Time: 14 May 2013, 7:51:34.962 pm UUID: 91ae4338-29e3-42a8-a944-7d703e6b9871 Ancestors: CMakeVMMaker-EstebanLorenzano.203 - missing one version change =============== Diff against CMakeVMMaker-EstebanLorenzano.203 =============== Item was changed: ----- Method: CMWin32ZLib>>unpackedDirName (in category 'as yet unclassified') ----- unpackedDirName + ^ 'zlib-1.2.8'! - ^ 'zlib-1.2.7'!

1 0

Re: [Vm-dev] VM Maker: VMMaker-oscog-EstebanLorenzano.238.mcz
by Eliot Miranda 14 May '13

14 May '13

I disagree. This is the tail wagging the dog. The warning is in this case erroneous. The old version is better in both time and space. I will not integrate this change. It's the wrong aesthetic ;) On Tue, May 14, 2013 at 8:04 AM, <commits(a)source.squeak.org> wrote: > > Esteban Lorenzano uploaded a new version of VMMaker to project VM Maker: > http://source.squeak.org/VMMaker/VMMaker-oscog-EstebanLorenzano.238.mcz > > ==================== Summary ==================== > > Name: VMMaker-oscog-EstebanLorenzano.238 > Author: EstebanLorenzano > Time: 14 May 2013, 5:03:09.451 pm > UUID: 5b4d9796-0c4c-4496-9513-42c47f6c0aa8 > Ancestors: VMMaker-oscog-EstebanLorenzano.237 > > - print: macro now resolves to printf("%s", ...) so debian guys do not > complain. > > =============== Diff against VMMaker-oscog-EstebanLorenzano.237 > =============== > > Item was changed: > ----- Method: Cogit>>print: (in category 'printing') ----- > print: aString > + <cmacro: '(aString) printf("%s", aString)'> > - <cmacro: '(aString) printf(aString)'> > coInterpreter transcript print: aString! > > -- best, Eliot

3 2

← Newer
1
2
3
4
5
6
7
8
9
...
15
Older →

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

Vm-dev May 2013