Vm-dev January 2014

vm-dev@lists.squeakfoundation.org

28 participants
97 discussions

Re: I would not have thought I ever say that: lets have a new primitve (was: Re: [squeak-dev] The Trunk: System-cwp.660.mcz)
by Bert Freudenberg 15 Jan '14

15 Jan '14

(moving to vm-dev) On 14.01.2014, at 10:56, Stefan Marr <smalltalk(a)stefan-marr.de> wrote: > Hi Tobias: > > On 14 Jan 2014, at 10:41, Tobias Pape <Das.Linux(a)gmx.de> wrote: > >>> How do you implement that efficiently in RPython, or Java? >>> Keeping some kind of object table around? >> >> Well, I have no Idea currently, but i think some #primAllObjectsDo: >> (or #primAllObjects, as it seems now) has IMHO less severe constraints >> than the #someObject/#nextObject way. >> OT is one path. Manual heap scanning another. Collecting objects from >> the roots (special objects array) a third. All these do not require >> the #nextObject invariant. > > I am just trying to think of how to implement that in the SOM VMs. > Without GC support for tracing/enumerating the objects, it’s not going to be pretty. > > #someObject/#nextObject don’t make any sense for me in the long term. Some VMs support those easily and they're more efficient, so I would not abandon them. But, as I wrote before, I'd make them optional, if a VM instead provides allObjects. > I want to support one form or another of parallel execution. So, if I want to offer #allObjects at all, it needs to be an atomic snapshot. Most likely with stop-the-world semantics, which is a severe burden on performance. Sure, but then allObjects shouldn't be used in production code anyway. Users need to be aware that it's going to be slow. > And without support of the underlying platform, I think, #allObjects can’t be complete. Just thinking of how TruffleSOM works, traversing only from specialObjectsArray on, will only find part of the live heap. You need to traverse from specialObjectsArray and activeContext. This will find all objects. > There are way to many important references on the Java stack, that I can’t easily access. And making them accessible, is going to cost a lot of performance even when #allObjects is never used. I think. :-/ But if you support thisContext then you must have those mechanisms in place already. > In case you find a nice solution for RSqueak, I would be very interested. In SqueakJS I support nextObject for oldspace objects only (old space is a linked list). If you call nextObject on a new object, I do a full GC, which makes it old. - Bert -

4 4

Re: I would not have thought I ever say that: lets have a new primitve (was: Re: [squeak-dev] The Trunk: System-cwp.660.mcz)
by Eliot Miranda 15 Jan '14

15 Jan '14

Hi Tobias, On Tue, Jan 14, 2014 at 1:41 AM, Tobias Pape <Das.Linux(a)gmx.de> wrote: > > On 13.01.2014, at 21:56, Stefan Marr <smalltalk(a)stefan-marr.de> wrote: > > > Hi Tobias: > > > > On 12 Jan 2014, at 18:47, Tobias Pape <Das.Linux(a)gmx.de> wrote: > > > >> I really don’t like to say it, but it seems to me, that a > >> VM primitive (say #primAllObjectsDo:) would be apt here. > >> • is more easy implemented by cousin VMs (say, Potato, SqueakJS, > >> SPy/RSqueakVM, SqueakMaxine† > > > > How do you implement that efficiently in RPython, or Java? > > Keeping some kind of object table around? > > > > Well, I have no Idea currently, but i think some #primAllObjectsDo: > (or #primAllObjects, as it seems now) has IMHO less severe constraints > than the #someObject/#nextObject way. > I've always thought that reentering the VI from a VM primitive, and hence being able to return to the VM primitive from arbitrary code executed in the block, would be problematic. It's a lot of new machinery to add for one case. Further, a GC could easily happen while executing the block. Do you think it's easier than I'm thinking? I know for example that the VisualAge VM does something somewhat analogous for its unwind-protects (ensure: ifCurtailed:). But neither the Squeak nor the VisualWorks VMs have anything like this; primitives always run a single contiguous sequence of instructions before either returning to Smalltalk (e.g. basicNew) or resuming Smalltalk (e.g. perform:). > OT is one path. Manual heap scanning another. Collecting objects from > the roots (special objects array) a third. All these do not require > the #nextObject invariant. > > Best > -Tobias > -- best, Eliot

3 3

VM Maker: VMMaker.oscog-eem.585.mcz
by commits＠source.squeak.org 15 Jan '14

15 Jan '14

Eliot Miranda uploaded a new version of VMMaker to project VM Maker: http://source.squeak.org/VMMaker/VMMaker.oscog-eem.585.mcz ==================== Summary ==================== Name: VMMaker.oscog-eem.585 Author: eem Time: 14 January 2014, 3:37:02.93 pm UUID: ffc9c619-e2b4-4432-977e-aba381f975ff Ancestors: VMMaker.oscog-eem.584 Fix become on compiled methods in Spur. followForwardedMethods must use isForwarded:/followForwarded: not shouldRemapObj: because the latter will confuse methods in newSpace for forwarded methods. Rename the interpreter followForwardedMethods' to followForwardedMethodsInMethodZone. In General: Fix some comment typos. Better comment the closed PIC prototype. Add CogMethodZone>>printCogYoungReferrers. =============== Diff against VMMaker.oscog-eem.584 =============== Item was removed: - ----- Method: CoInterpreter>>followForwardedMethods (in category 'object memory support') ----- - followForwardedMethods - <doNotGenerate> - cogit followForwardedMethods! Item was added: + ----- Method: CoInterpreter>>followForwardedMethodsInMethodZone (in category 'object memory support') ----- + followForwardedMethodsInMethodZone + <inline: true> + cogit followForwardedMethods! Item was added: + ----- Method: CogMethodZone>>printCogYoungReferrers (in category 'printing') ----- + printCogYoungReferrers + <api> + <returnTypeC: #void> + | pointer cogMethod | + <var: #cogMethod type: #'CogMethod *'> + pointer := youngReferrers. + [pointer < limitAddress] whileTrue: + [cogMethod := coInterpreter cCoerceSimple: (objectMemory longAt: pointer) to: #'CogMethod *'. + cogMethod cmRefersToYoung ifFalse: + [coInterpreter print: '* ']. + coInterpreter printCogMethod: cogMethod. + pointer := pointer + BytesPerWord]! Item was changed: ----- Method: Cogit>>compileClosedPICPrototype (in category 'in-line cacheing') ----- compileClosedPICPrototype + "Compile the abstract instructions for a full closed PIC used to initialize closedPICSize. + The loads into SendNumArgsReg are those for optional method objects which may be + used in MNU cases." - "Compile the abstract instructions for a full closed PIC used to initialize closedPICSize" | numArgs jumpNext | <var: #jumpNext type: #'AbstractInstruction *'> numArgs := 0. self compilePICProlog: numArgs. jumpNext := self compileCPICEntry. self MoveCw: 16r5EAF00D R: SendNumArgsReg. self JumpLong: 16r5EEDCA5E. jumpNext jmpTarget: (endCPICCase0 := self Label). 1 to: numPICCases - 1 do: [:h| self CmpCw: 16rBABE1F15+h R: TempReg. self MoveCw: 16rBADA550 + h R: SendNumArgsReg. self JumpLongZero: 16rBEDCA5E0. h = 1 ifTrue: [endCPICCase1 := self Label]]. self MoveCw: 16rAB5CE55 R: ClassReg. self JumpLong: (self cPICMissTrampolineFor: numArgs). ^0! Item was changed: ----- Method: Cogit>>followForwardedMethods (in category 'garbage collection') ----- followForwardedMethods <api> <option: #SpurObjectMemory> <var: #cogMethod type: #'CogMethod *'> | cogMethod freedPIC | <var: #cogMethod type: #'CogMethod *'> freedPIC := false. cogMethod := self cCoerceSimple: methodZoneBase to: #'CogMethod *'. [cogMethod < methodZone limitZony] whileTrue: [cogMethod cmType = CMMethod ifTrue: + [(objectMemory isForwarded: cogMethod methodObject) ifTrue: + [cogMethod methodObject: (objectMemory followForwarded: cogMethod methodObject). - [(objectMemory shouldRemapOop: cogMethod methodObject) ifTrue: - [cogMethod methodObject: (objectMemory remapObj: cogMethod methodObject). (cogMethod cmRefersToYoung not and: [objectMemory isYoungObject: cogMethod methodObject]) ifTrue: [methodZone addToYoungReferrers: cogMethod]]]. cogMethod cmType = CMClosedPIC ifTrue: + [(self followMethodReferencesInClosedPIC: cogMethod) ifTrue: - [(self mapObjectReferencesInClosedPIC: cogMethod) ifTrue: [freedPIC := true. methodZone freeMethod: cogMethod]]. cogMethod := methodZone methodAfter: cogMethod]. freedPIC ifTrue: [self unlinkSendsToFree. processor flushICacheFrom: codeBase to: methodZone limitZony asInteger]! Item was added: + ----- Method: Cogit>>followMaybeObjRefAt: (in category 'garbage collection') ----- + followMaybeObjRefAt: mcpc + "Follow a potential object reference from a closed PIC. + This may be a method reference or null. + Answer if the followed literal is young." + | object subject | + object := backEnd literalBeforeFollowingAddress: mcpc. + (objectRepresentation couldBeObject: object) ifFalse: + [^false]. + (objectMemory isForwarded: object) ifFalse: + [^objectMemory isYoungObject: object]. + subject := objectMemory followForwarded: object. + backEnd storeLiteral: subject beforeFollowingAddress: mcpc. + codeModified := true. + ^objectMemory isYoungObject: subject! Item was added: + ----- Method: Cogit>>followMethodReferencesInClosedPIC: (in category 'garbage collection') ----- + followMethodReferencesInClosedPIC: cPIC + "Remap all object references in the closed PIC. Answer if any references are young. + Set codeModified if any modifications are made." + <var: #cPIC type: #'CogMethod *'> + | pc refersToYoung | + pc := cPIC asInteger + firstCPICCaseOffset. + refersToYoung := self followMaybeObjRefAt: pc - backEnd jumpLongByteSize. + pc := pc + cPICCaseSize. + 2 to: cPIC cPICNumCases do: + [:i| + (self followMaybeObjRefAt: pc - backEnd jumpLongConditionalByteSize) ifTrue: + [refersToYoung := true]. + pc := pc + cPICCaseSize]. + ^refersToYoung! Item was changed: ----- Method: Cogit>>updateMaybeObjRefAt: (in category 'garbage collection') ----- updateMaybeObjRefAt: mcpc + "Update a potential object reference from a closed PIC. - "Update a potential object reference form a closed PIC. This may be an object reference, an inline cache tag or null. Answer if the updated literal is young." | object subject | object := backEnd literalBeforeFollowingAddress: mcpc. (objectRepresentation couldBeObject: object) ifFalse: [^false]. subject := objectRepresentation remapOop: object. object ~= subject ifTrue: [backEnd storeLiteral: subject beforeFollowingAddress: mcpc. codeModified := true]. ^objectMemory isYoungObject: subject! Item was changed: ----- Method: SpurMemoryManager>>followForwarded: (in category 'forwarding') ----- followForwarded: objOop "Follow a forwarding pointer. Alas we cannot prevent forwarders to forwarders being created by lazy become. Consider the following example by Igor Stasenko: array := { a. b. c }. - array at: 1 points to &a. array at: 2 points to &b. array at: 3 points to &c Ó a becomeForward: b - array at: 1 still points to &a. array at: 2 still points to &b. array at: 3 still points to &c b becomeForward: c. - array at: 1 still points to &a. array at: 2 still points to &b. array at: 3 still points to &c - when accessing array first one has to follow a forwarding chain: &a -> &b -> c" + <api> | referent | self assert: (self isForwarded: objOop). referent := self fetchPointer: 0 ofMaybeForwardedObject: objOop. [(self isOopForwarded: referent)] whileTrue: [referent := self fetchPointer: 0 ofMaybeForwardedObject: referent]. ^referent! Item was removed: - ----- Method: StackInterpreter>>followForwardedMethods (in category 'object memory support') ----- - followForwardedMethods - "This is just a stub for the CoInterpreter"! Item was added: + ----- Method: StackInterpreter>>followForwardedMethodsInMethodZone (in category 'object memory support') ----- + followForwardedMethodsInMethodZone + "This is just a stub for the CoInterpreter"! Item was changed: ----- Method: StackInterpreter>>postBecomeAction: (in category 'object memory support') ----- postBecomeAction: theBecomeEffectsFlags theBecomeEffectsFlags ~= 0 ifTrue: [self followForwardingPointersInStackZone: theBecomeEffectsFlags. (theBecomeEffectsFlags anyMask: BecameCompiledMethodFlag) ifTrue: + [self followForwardedMethodsInMethodCache. + self followForwardedMethodsInMethodZone]. "for CoInterpreter" - [self followForwardedMethodsInMethodCache]. - self followForwardedMethods. "for CoInterpreter" self followForwardingPointersInScheduler]!

1 0

Re: allObjectsDo: (was Re: [squeak-dev] The Inbox: System-cwp.662.mcz)
by David T. Lewis 14 Jan '14

14 Jan '14

On Sun, Jan 12, 2014 at 09:53:27PM +0100, Bert Freudenberg wrote: > > On 12.01.2014, at 20:42, David T. Lewis <lewis(a)mail.msen.com> wrote: > > > It is worth noting that allObjectsDo: relies on assumptions about how > > the objects memory works internally. It requires that #someObject will > > always answer the object at the lowest address in the object memory, and > > also that a newly allocated object will always be placed at a higher > > address location than all other objects. Either of these assumptions is > > likely to be a problem as new and better object memories and garbage > > collectors are implemented. > > > > Dave > > Right (as Eliot's vm-dev post shows). > > So IMHO the only sensible semantics of allObjectsDo: is as in "allObjects do:" - > which might be implemented as a primitive in some VMs soonish. It *should not* > enumerate objects created after calling the method. > On Sun, Jan 12, 2014 at 12:01:00PM -0800, Eliot Miranda wrote: > > The bug is in implementing allObjects in terms of someObject and nextObject > in the first place. It's cheap and cheerful but horribly error-prone and > restrictive. It's cheap because the collection of objects doesn't have to > be created, and on the original 16-bit Smalltalk machines that was really > important. It's horribly restrictive because it assumes much about the > implementation. > > Before closures a sentinel wasn't even needed because enumerating the block > didn't create a new object (the block context was reused). So the code had > to be rewritten just to support closures. > > Spur has a generation scavenger operating in a distinct new space and that > doesn't jive well with a consistent ordering at all. So far the system is > limping along by tenuring all objects on someObject and someInstance (so > that newSpace is either empty, or doesn't contain any instances of a > specific class) and having nextObject enumerate only objects in oldSpace. > > But I think now we can afford a primitive that answers all the objects > (remember that average object size means that such a collection will be ~ > 10% of the heap, average object size in Squeak V3 is about 10.6 words). At > least that's what Spur will do, along with an allInstancesOf: primitive. > And then the become example won't cause any problems at all. Far more > reliable. I suppose there are circumstances when enumerating without a > container is the only feasible approach, but VisualWorks has got along with > only an allObjects primitive for a long time now. I suspect we can too. > Implementation attached. Works on interpreter VM, not yet tested on Cog but it should be ok there also. If no objections or better suggestions I will commit it to VMMaker tomorrow. InterpreterPrimitives>>primitiveAllObjects "Answer an array of all objects that exist when the primitive is called, excluding those that may be garbage collected as a side effect of allocating the result array. Multiple references to nil in the last slots of the array are an indication that garbage collection occured, such that some of the unreferenced objects that existed at the time of calling the primitive no longer exist. Sender is responsible for handling multiple references to nil in the result array." Dave

2 6

building on Darwin (Mac Unix)
by Craig Latta 14 Jan '14

14 Jan '14

Hi-- Having isolated the Spoon VM changes[1], I'm testing them with the latest VM sources (and Mac OS/devtools). As always, when I do this once a year, I get errors... Am I doing something wrong? - I go to http://squeakvm.org/unix and follow the instructions. - svn co http://squeakvm.org/svn/squeak/trunk squeak-svn - cd platforms/unix; make - I get the following: *** % make [ -d bld ] || mkdir bld [ -f bld/Makefile ] || ( cd bld; ../cmake/configure; ) -- Configuring squeak 4.10.2-2840 for i386-apple-darwin13.1.0 -- Using source directory /Users/craig/me/behavior/spoon/processor/kernel/mainstream/squeak-svn/platforms/unix/src -- Using CFLAGS -O3 -DNDEBUG -I/usr/local/include -L/usr/local/lib !! Cannot optimise interpreter performance for GCC !! FT2Plugin disabled -- SocketPlugin: libsocket not found -- SocketPlugin: libnsl not found -- CameraPlugin: linux/videodev2.h not found !! CameraPlugin disabled !! DBusPlugin disabled -- SqueakFFIPrims: any-libffi -- SqueakFFIPrims: /usr/lib/libffi.dylib -- Warning: Did not find file Compiler/Clang-ASM -- MIDIPlugin: alsa/asoundlib.h not found -- MIDIPlugin: libasound not found !! MIDIPlugin disabled -- RomePlugin: cairo-ft.h not found !! RomePlugin disabled -- UUIDPlugin: uuid.h not found -- UUIDPlugin: sys/uuid.h not found -- UUIDPlugin: /System/Library/Frameworks/Kernel.framework/Headers/uuid/uuid/uuid.h -- UUIDPlugin: libuuid not found !! UnicodePlugin disabled -- UnixOSProcessPlugin: /usr/lib/libpthread.dylib -- WeDoPlugin: linux/hiddev.h not found !! WeDoPlugin disabled -- vm-display-fbdev: linux/fb.h not found !! vm-display-fbdev disabled -- vm-sound-ALSA: alsa/asoundlib.h not found -- vm-sound-ALSA: libasound not found !! vm-sound-ALSA disabled -- vm-sound-NAS: audio/audio.h not found -- vm-sound-NAS: libaudio not found !! vm-sound-NAS disabled -- vm-sound-OSS: alsa/asoundlib.h not found -- vm-sound-OSS: libasound not found !! vm-sound-OSS disabled -- vm-sound-pulse: libpulse-simple not found !! vm-sound-pulse disabled -- vm-sound-Sun: sys/audioio.h not found -- vm-sound-Sun: sun/audioio.h not found -- vm-sound-Sun: stropts.h not found !! vm-sound-Sun disabled You have called ADD_LIBRARY for library ADPCMCodecPlugin without any source files. This typically indicates a problem with your CMakeLists.txt file You have called ADD_LIBRARY for library BMPReadWriterPlugin without any source files. This typically indicates a problem with your CMakeLists.txt file You have called ADD_LIBRARY for library B2DPlugin without any source files. This typically indicates a problem with your CMakeLists.txt file You have called ADD_LIBRARY for library DSAPrims without any source files. This typically indicates a problem with your CMakeLists.txt file You have called ADD_LIBRARY for library ZipPlugin without any source files. This typically indicates a problem with your CMakeLists.txt file You have called ADD_LIBRARY for library FFTPlugin without any source files. This typically indicates a problem with your CMakeLists.txt file You have called ADD_LIBRARY for library FloatArrayPlugin without any source files. This typically indicates a problem with your CMakeLists.txt file You have called ADD_LIBRARY for library GeniePlugin without any source files. This typically indicates a problem with your CMakeLists.txt file You have called ADD_LIBRARY for library JPEGReaderPlugin without any source files. This typically indicates a problem with your CMakeLists.txt file You have called ADD_LIBRARY for library Klatt without any source files. This typically indicates a problem with your CMakeLists.txt file You have called ADD_LIBRARY for library LargeIntegers without any source files. This typically indicates a problem with your CMakeLists.txt file You have called ADD_LIBRARY for library Matrix2x3Plugin without any source files. This typically indicates a problem with your CMakeLists.txt file You have called ADD_LIBRARY for library MiscPrimitivePlugin without any source files. This typically indicates a problem with your CMakeLists.txt file You have called ADD_LIBRARY for library SlangTestSupportPlugin without any source files. This typically indicates a problem with your CMakeLists.txt file You have called ADD_LIBRARY for library StarSqueakPlugin without any source files. This typically indicates a problem with your CMakeLists.txt file CMake Warning (dev) at CMakeLists.txt:162 (ADD_SUBDIRECTORY): The source directory /Users/craig/me/behavior/spoon/processor/kernel/mainstream/squeak-svn/platforms/unix/npsqueak does not contain a CMakeLists.txt file. CMake does not support this case but it used to work accidentally and is being allowed for compatibility. Policy CMP0014 is not set: Input directories must have CMakeLists.txt. Run "cmake --help-policy CMP0014" for policy details. Use the cmake_policy command to set the policy and suppress this warning. This warning is for project developers. Use -Wno-dev to suppress it. -- Configuring done CMake Error at bld/BitBltPlugin/CMakeLists.txt:14 (ADD_LIBRARY): Cannot find source file: /Users/craig/me/behavior/spoon/processor/kernel/mainstream/squeak-svn/platforms/unix/src/plugins/BitBltPlugin/BitBltPlugin.c Tried extensions .c .C .c++ .cc .cpp .cxx .m .M .mm .h .hh .h++ .hm .hpp .hxx .in .txx -- Build files have been written to: /Users/craig/me/behavior/spoon/processor/kernel/mainstream/squeak-svn/platforms/unix/bld make: *** [all] Error 1 % *** Is something wrong with the CMake stuff? Does no one actually use this, and I should be doing something completely different? Mac OS 10.9.2, XCode 5.2. thanks, -C [1] http://tinyurl.com/p999omz -- Craig Latta www.netjam.org/resume +1 510 984 8117 (Skype rings this until 31 January 2014)

7 15

[commit][2846] CogVM source as per VMMaker.oscog-eem.584
by commits＠squeakvm.org 14 Jan '14

14 Jan '14

Revision: 2846 Author: eliot Date: 2014-01-13 18:14:07 -0800 (Mon, 13 Jan 2014) Log Message: ----------- CogVM source as per VMMaker.oscog-eem.584 Make sure SIGALRM is unblocked if using the interval timer heartbeat on linux. Spur: Fix swizzling on startup. Enumerate up to freeOldSpaceStart, not endOfMemory. Fixes startup of e.g. bootstrapped Squeak4.3 image. Admin: Move SqueakSSLPlugin to VMMaker from SqueakSSL. Modified Paths: -------------- branches/Cog/nscogsrc/plugins/SqueakSSL/SqueakSSL.c branches/Cog/platforms/unix/vm/sqUnixITimerHeartbeat.c branches/Cog/platforms/unix/vm/sqUnixITimerTickerHeartbeat.c branches/Cog/platforms/unix/vm/sqUnixMain.c branches/Cog/spursrc/vm/cointerp.c branches/Cog/spursrc/vm/cointerp.h branches/Cog/spursrc/vm/gcc3x-cointerp.c branches/Cog/spursrc/vm/interp.h branches/Cog/spursrc/vm/vmCallback.h branches/Cog/spurstacksrc/vm/gcc3x-interp.c branches/Cog/spurstacksrc/vm/interp.c branches/Cog/spurstacksrc/vm/interp.h branches/Cog/spurstacksrc/vm/vmCallback.h branches/Cog/src/plugins/SqueakSSL/SqueakSSL.c Property Changed: ---------------- branches/Cog/platforms/Cross/vm/sqSCCSVersion.h branches/Cog/spurcogbuild/cygwinbuild/ branches/Cog/spurcogbuild/unixbuild/abldt/ branches/Cog/spurcogbuild/unixbuild/bldt/ branches/Cog/spurcogbuild/unixbuild/dbldt/ Modified: branches/Cog/nscogsrc/plugins/SqueakSSL/SqueakSSL.c =================================================================== --- branches/Cog/nscogsrc/plugins/SqueakSSL/SqueakSSL.c 2014-01-14 01:59:30 UTC (rev 2845) +++ branches/Cog/nscogsrc/plugins/SqueakSSL/SqueakSSL.c 2014-01-14 02:14:07 UTC (rev 2846) @@ -1,9 +1,9 @@ /* Automatically generated by - VMPluginCodeGenerator VMMaker.oscog-eem.580 uuid: 751b08d4-d92e-440a-b3f6-cb2c76f52514 + VMPluginCodeGenerator VMMaker.oscog-eem.584 uuid: f8e3994b-6462-4df0-bc8e-c71034adf6aa from - SqueakSSLPlugin SqueakSSL-Plugin-eem.4 uuid: 5e2ad007-e58e-4c86-9657-47e0cfe12918 + SqueakSSLPlugin VMMaker.oscog-eem.584 uuid: f8e3994b-6462-4df0-bc8e-c71034adf6aa */ -static char __buildInfo[] = "SqueakSSLPlugin SqueakSSL-Plugin-eem.4 uuid: 5e2ad007-e58e-4c86-9657-47e0cfe12918 " __DATE__ ; +static char __buildInfo[] = "SqueakSSLPlugin VMMaker.oscog-eem.584 uuid: f8e3994b-6462-4df0-bc8e-c71034adf6aa " __DATE__ ; @@ -101,9 +101,9 @@ struct VirtualMachine* interpreterProxy; static const char *moduleName = #ifdef SQUEAK_BUILTIN_PLUGIN - "SqueakSSL SqueakSSL-Plugin-eem.4 (i)" + "SqueakSSL VMMaker.oscog-eem.584 (i)" #else - "SqueakSSL SqueakSSL-Plugin-eem.4 (e)" + "SqueakSSL VMMaker.oscog-eem.584 (e)" #endif ; Property changes on: branches/Cog/platforms/Cross/vm/sqSCCSVersion.h ___________________________________________________________________ Modified: checkindate - Fri Jan 10 13:54:16 PST 2014 + Mon Jan 13 18:13:36 PST 2014 Modified: branches/Cog/platforms/unix/vm/sqUnixITimerHeartbeat.c =================================================================== --- branches/Cog/platforms/unix/vm/sqUnixITimerHeartbeat.c 2014-01-14 01:59:30 UTC (rev 2845) +++ branches/Cog/platforms/unix/vm/sqUnixITimerHeartbeat.c 2014-01-14 02:14:07 UTC (rev 2846) @@ -411,6 +411,7 @@ extern sqInt suppressHeartbeatFlag; int er; struct sigaction heartbeat_handler_action; + sigset_t ss; if (suppressHeartbeatFlag) return; @@ -446,6 +447,11 @@ exit(1); } + /* Make sure SIGALRM is unblocked. */ + sigemptyset(&ss); + sigaddset(&ss, ITIMER_SIGNAL); + sigprocmask(SIG_UNBLOCK, &ss, NULL); + setIntervalTimer(beatMilliseconds); } Modified: branches/Cog/platforms/unix/vm/sqUnixITimerTickerHeartbeat.c =================================================================== --- branches/Cog/platforms/unix/vm/sqUnixITimerTickerHeartbeat.c 2014-01-14 01:59:30 UTC (rev 2845) +++ branches/Cog/platforms/unix/vm/sqUnixITimerTickerHeartbeat.c 2014-01-14 02:14:07 UTC (rev 2846) @@ -540,6 +540,7 @@ int er; struct timespec halfAMo; struct sigaction heartbeat_handler_action, ticker_handler_action; + sigset_t ss; if (suppressHeartbeatFlag) return; @@ -604,6 +605,11 @@ exit(1); } + /* Make sure SIGALRM is unblocked. */ + sigemptyset(&ss); + sigaddset(&ss, ITIMER_SIGNAL); + sigprocmask(SIG_UNBLOCK, &ss, NULL); + setIntervalTimer(beatMilliseconds); } Modified: branches/Cog/platforms/unix/vm/sqUnixMain.c =================================================================== --- branches/Cog/platforms/unix/vm/sqUnixMain.c 2014-01-14 01:59:30 UTC (rev 2845) +++ branches/Cog/platforms/unix/vm/sqUnixMain.c 2014-01-14 02:14:07 UTC (rev 2846) @@ -1740,12 +1740,24 @@ if (extraMemory) useMmap= 0; else +#if SPURVM + { off_t size = (long)sb.st_size; + + size = 1 << highBit(size-1); + size = size + size / 4; +# ifdef DEBUG_IMAGE + printf("image size %ld + heap size %ld (useMmap = %d)\n", (long)sb.st_size, size, useMmap); +# endif + readImageFromFileHeapSizeStartingAt(f, size + size / 4, 0); + } +#else extraMemory= DefaultHeapSize * 1024 * 1024; # ifdef DEBUG_IMAGE printf("image size %ld + heap size %ld (useMmap = %d)\n", (long)sb.st_size, extraMemory, useMmap); # endif extraMemory += (long)sb.st_size; readImageFromFileHeapSizeStartingAt(f, extraMemory, 0); +#endif sqImageFileClose(f); break; } Property changes on: branches/Cog/spurcogbuild/cygwinbuild ___________________________________________________________________ Added: svn:ignore + build buildast builddbg eitheror.exe mkNamedPrims.exe sqNamedPrims.h Property changes on: branches/Cog/spurcogbuild/unixbuild/abldt ___________________________________________________________________ Added: svn:ignore + * Property changes on: branches/Cog/spurcogbuild/unixbuild/bldt ___________________________________________________________________ Added: svn:ignore + * Property changes on: branches/Cog/spurcogbuild/unixbuild/dbldt ___________________________________________________________________ Added: svn:ignore + * Modified: branches/Cog/spursrc/vm/cointerp.c =================================================================== --- branches/Cog/spursrc/vm/cointerp.c 2014-01-14 01:59:30 UTC (rev 2845) +++ branches/Cog/spursrc/vm/cointerp.c 2014-01-14 02:14:07 UTC (rev 2846) @@ -1,9 +1,9 @@ /* Automatically generated by - CCodeGeneratorGlobalStructure VMMaker.oscog-eem.581 uuid: 926531fd-20be-42b4-9332-015b90e6734f + CCodeGeneratorGlobalStructure VMMaker.oscog-eem.584 uuid: f8e3994b-6462-4df0-bc8e-c71034adf6aa from - CoInterpreter VMMaker.oscog-eem.581 uuid: 926531fd-20be-42b4-9332-015b90e6734f + CoInterpreter VMMaker.oscog-eem.584 uuid: f8e3994b-6462-4df0-bc8e-c71034adf6aa */ -static char __buildInfo[] = "CoInterpreter VMMaker.oscog-eem.581 uuid: 926531fd-20be-42b4-9332-015b90e6734f " __DATE__ ; +static char __buildInfo[] = "CoInterpreter VMMaker.oscog-eem.584 uuid: f8e3994b-6462-4df0-bc8e-c71034adf6aa " __DATE__ ; char *__interpBuildInfo = __buildInfo; @@ -1459,6 +1459,7 @@ _iss usqLong longRunningPrimitiveStopUsecs; _iss sqInt numSegInfos; _iss sqInt statCheckForEvents; +_iss usqInt freeOldSpaceStart; _iss sqInt gcMode; _iss sqInt lastCoggableInterpretedBlockMethod; _iss sqInt lastUncoggableInterpretedBlockMethod; @@ -1468,7 +1469,6 @@ _iss sqInt cogCodeSize; _iss sqInt externalPrimitiveTableFirstFreeIndex; _iss sqInt flagInterpretedMethods; -_iss usqInt freeOldSpaceStart; _iss sqInt metaclassNumSlots; _iss sqInt pendingFinalizationSignals; _iss usqInt scavengeThreshold; @@ -2164,7 +2164,7 @@ /* 575 */ (void (*)(void))0, 0 }; usqInt heapBase; -const char *interpreterVersion = "Croquet Closure Cog VM [CoInterpreter VMMaker.oscog-eem.581]"; +const char *interpreterVersion = "Croquet Closure Cog VM [CoInterpreter VMMaker.oscog-eem.584]"; sqInt minBackwardJumpCountForCompile = MinBackwardJumpCountForCompile /* 10 */; volatile int sendTrace; @@ -6374,6 +6374,7 @@ /* begin internalPush: */ longAtPointerput((localSP -= BytesPerOop), longAt((rcvr1 + (BaseHeaderSize)) + (byte3 << 2))); } + null; goto l4; } if (opType == 3) { @@ -6383,6 +6384,7 @@ assert(GIV(method) == (iframeMethod(localFP))); object = longAt((GIV(method) + (BaseHeaderSize)) + ((byte3 + LiteralStart) << 2)); longAtPointerput((localSP -= BytesPerOop), object); + null; goto l4; } if (opType == 4) { @@ -6410,6 +6412,7 @@ /* begin internalPush: */ longAtPointerput((localSP -= BytesPerOop), longAt((litVar1 + (BaseHeaderSize)) + (ValueIndex << 2))); + null; goto l4; } top = longAtPointer(localSP); @@ -14896,7 +14899,8 @@ : methodHeader1); primIndex = ((primBits = ((usqInt) methodHeader) >> 1), (primBits & 0x1FF) + ((((usqInt) primBits) >> 19) & 512)); - assert(primitiveFunctionPointer == (functionPointerForinClass(primIndex, nilObject()))); + assert((primIndex == 117) + || (primitiveFunctionPointer == (functionPointerForinClass(primIndex, nilObject())))); assert(GIV(argumentCount) == (argumentCountOf(GIV(newMethod)))); accessorDepth = ((primIndex == 117) && (primitiveFunctionPointer != primitiveExternalCall) @@ -25479,7 +25483,7 @@ obj = (numSlots == 0xFF ? GIV(oldSpaceStart) + (BaseHeaderSize) : GIV(oldSpaceStart)); - while ((((usqInt) obj)) < (((usqInt) GIV(endOfMemory)))) { + while ((((usqInt) obj)) < (((usqInt) GIV(freeOldSpaceStart)))) { if (((longAt(obj)) & 0x3FFFFF) == 0) { /* begin swizzleFieldsOfFreeChunk: */ field = longAt((obj + (BaseHeaderSize)) + (0 << 2)); @@ -26838,7 +26842,7 @@ static sqInt isValidSegmentBridge(sqInt objOop) { - return (addressCouldBeObj(objOop - (BaseHeaderSize))) + return (addressCouldBeObj(objOop)) && ((((longAt(objOop)) & 0x3FFFFF) == 3) && (((rawNumSlotsOf(objOop)) == 0xFF) || ((numSlotsOfAny(objOop)) == 0))); Modified: branches/Cog/spursrc/vm/cointerp.h =================================================================== --- branches/Cog/spursrc/vm/cointerp.h 2014-01-14 01:59:30 UTC (rev 2845) +++ branches/Cog/spursrc/vm/cointerp.h 2014-01-14 02:14:07 UTC (rev 2846) @@ -1,5 +1,5 @@ /* Automatically generated by - CCodeGeneratorGlobalStructure VMMaker.oscog-eem.581 uuid: 926531fd-20be-42b4-9332-015b90e6734f + CCodeGeneratorGlobalStructure VMMaker.oscog-eem.584 uuid: f8e3994b-6462-4df0-bc8e-c71034adf6aa */ Modified: branches/Cog/spursrc/vm/gcc3x-cointerp.c =================================================================== --- branches/Cog/spursrc/vm/gcc3x-cointerp.c 2014-01-14 01:59:30 UTC (rev 2845) +++ branches/Cog/spursrc/vm/gcc3x-cointerp.c 2014-01-14 02:14:07 UTC (rev 2846) @@ -2,11 +2,11 @@ /* Automatically generated by - CCodeGeneratorGlobalStructure VMMaker.oscog-eem.581 uuid: 926531fd-20be-42b4-9332-015b90e6734f + CCodeGeneratorGlobalStructure VMMaker.oscog-eem.584 uuid: f8e3994b-6462-4df0-bc8e-c71034adf6aa from - CoInterpreter VMMaker.oscog-eem.581 uuid: 926531fd-20be-42b4-9332-015b90e6734f + CoInterpreter VMMaker.oscog-eem.584 uuid: f8e3994b-6462-4df0-bc8e-c71034adf6aa */ -static char __buildInfo[] = "CoInterpreter VMMaker.oscog-eem.581 uuid: 926531fd-20be-42b4-9332-015b90e6734f " __DATE__ ; +static char __buildInfo[] = "CoInterpreter VMMaker.oscog-eem.584 uuid: f8e3994b-6462-4df0-bc8e-c71034adf6aa " __DATE__ ; char *__interpBuildInfo = __buildInfo; @@ -1462,6 +1462,7 @@ _iss usqLong longRunningPrimitiveStopUsecs; _iss sqInt numSegInfos; _iss sqInt statCheckForEvents; +_iss usqInt freeOldSpaceStart; _iss sqInt gcMode; _iss sqInt lastCoggableInterpretedBlockMethod; _iss sqInt lastUncoggableInterpretedBlockMethod; @@ -1471,7 +1472,6 @@ _iss sqInt cogCodeSize; _iss sqInt externalPrimitiveTableFirstFreeIndex; _iss sqInt flagInterpretedMethods; -_iss usqInt freeOldSpaceStart; _iss sqInt metaclassNumSlots; _iss sqInt pendingFinalizationSignals; _iss usqInt scavengeThreshold; @@ -2167,7 +2167,7 @@ /* 575 */ (void (*)(void))0, 0 }; usqInt heapBase; -const char *interpreterVersion = "Croquet Closure Cog VM [CoInterpreter VMMaker.oscog-eem.581]"; +const char *interpreterVersion = "Croquet Closure Cog VM [CoInterpreter VMMaker.oscog-eem.584]"; sqInt minBackwardJumpCountForCompile = MinBackwardJumpCountForCompile /* 10 */; volatile int sendTrace; @@ -6383,6 +6383,7 @@ /* begin internalPush: */ longAtPointerput((localSP -= BytesPerOop), longAt((rcvr1 + (BaseHeaderSize)) + (byte3 << 2))); } + null; goto l4; } if (opType == 3) { @@ -6392,6 +6393,7 @@ assert(GIV(method) == (iframeMethod(localFP))); object = longAt((GIV(method) + (BaseHeaderSize)) + ((byte3 + LiteralStart) << 2)); longAtPointerput((localSP -= BytesPerOop), object); + null; goto l4; } if (opType == 4) { @@ -6419,6 +6421,7 @@ /* begin internalPush: */ longAtPointerput((localSP -= BytesPerOop), longAt((litVar1 + (BaseHeaderSize)) + (ValueIndex << 2))); + null; goto l4; } top = longAtPointer(localSP); @@ -14905,7 +14908,8 @@ : methodHeader1); primIndex = ((primBits = ((usqInt) methodHeader) >> 1), (primBits & 0x1FF) + ((((usqInt) primBits) >> 19) & 512)); - assert(primitiveFunctionPointer == (functionPointerForinClass(primIndex, nilObject()))); + assert((primIndex == 117) + || (primitiveFunctionPointer == (functionPointerForinClass(primIndex, nilObject())))); assert(GIV(argumentCount) == (argumentCountOf(GIV(newMethod)))); accessorDepth = ((primIndex == 117) && (primitiveFunctionPointer != primitiveExternalCall) @@ -25488,7 +25492,7 @@ obj = (numSlots == 0xFF ? GIV(oldSpaceStart) + (BaseHeaderSize) : GIV(oldSpaceStart)); - while ((((usqInt) obj)) < (((usqInt) GIV(endOfMemory)))) { + while ((((usqInt) obj)) < (((usqInt) GIV(freeOldSpaceStart)))) { if (((longAt(obj)) & 0x3FFFFF) == 0) { /* begin swizzleFieldsOfFreeChunk: */ field = longAt((obj + (BaseHeaderSize)) + (0 << 2)); @@ -26847,7 +26851,7 @@ static sqInt isValidSegmentBridge(sqInt objOop) { - return (addressCouldBeObj(objOop - (BaseHeaderSize))) + return (addressCouldBeObj(objOop)) && ((((longAt(objOop)) & 0x3FFFFF) == 3) && (((rawNumSlotsOf(objOop)) == 0xFF) || ((numSlotsOfAny(objOop)) == 0))); Modified: branches/Cog/spursrc/vm/interp.h =================================================================== --- branches/Cog/spursrc/vm/interp.h 2014-01-14 01:59:30 UTC (rev 2845) +++ branches/Cog/spursrc/vm/interp.h 2014-01-14 02:14:07 UTC (rev 2846) @@ -1,5 +1,5 @@ /* Automatically generated by - CCodeGeneratorGlobalStructure VMMaker.oscog-eem.581 uuid: 926531fd-20be-42b4-9332-015b90e6734f + CCodeGeneratorGlobalStructure VMMaker.oscog-eem.584 uuid: f8e3994b-6462-4df0-bc8e-c71034adf6aa */ #define VM_PROXY_MAJOR 1 Modified: branches/Cog/spursrc/vm/vmCallback.h =================================================================== --- branches/Cog/spursrc/vm/vmCallback.h 2014-01-14 01:59:30 UTC (rev 2845) +++ branches/Cog/spursrc/vm/vmCallback.h 2014-01-14 02:14:07 UTC (rev 2846) @@ -1,5 +1,5 @@ /* Automatically generated by - CCodeGeneratorGlobalStructure VMMaker.oscog-eem.581 uuid: 926531fd-20be-42b4-9332-015b90e6734f + CCodeGeneratorGlobalStructure VMMaker.oscog-eem.584 uuid: f8e3994b-6462-4df0-bc8e-c71034adf6aa */ #define VM_CALLBACK_INC 1 Modified: branches/Cog/spurstacksrc/vm/gcc3x-interp.c =================================================================== --- branches/Cog/spurstacksrc/vm/gcc3x-interp.c 2014-01-14 01:59:30 UTC (rev 2845) +++ branches/Cog/spurstacksrc/vm/gcc3x-interp.c 2014-01-14 02:14:07 UTC (rev 2846) @@ -2,11 +2,11 @@ /* Automatically generated by - CCodeGeneratorGlobalStructure VMMaker.oscog-eem.581 uuid: 926531fd-20be-42b4-9332-015b90e6734f + CCodeGeneratorGlobalStructure VMMaker.oscog-eem.584 uuid: f8e3994b-6462-4df0-bc8e-c71034adf6aa from - StackInterpreter VMMaker.oscog-eem.581 uuid: 926531fd-20be-42b4-9332-015b90e6734f + StackInterpreter VMMaker.oscog-eem.584 uuid: f8e3994b-6462-4df0-bc8e-c71034adf6aa */ -static char __buildInfo[] = "StackInterpreter VMMaker.oscog-eem.581 uuid: 926531fd-20be-42b4-9332-015b90e6734f " __DATE__ ; +static char __buildInfo[] = "StackInterpreter VMMaker.oscog-eem.584 uuid: f8e3994b-6462-4df0-bc8e-c71034adf6aa " __DATE__ ; char *__interpBuildInfo = __buildInfo; @@ -1271,11 +1271,11 @@ _iss usqLong longRunningPrimitiveStopUsecs; _iss sqInt numSegInfos; _iss sqInt statCheckForEvents; +_iss usqInt freeOldSpaceStart; _iss usqInt lowSpaceThreshold; _iss usqLong statGCEndUsecs; _iss sqInt classTableIndex; _iss sqInt externalPrimitiveTableFirstFreeIndex; -_iss usqInt freeOldSpaceStart; _iss sqInt lkupClass; _iss sqInt metaclassNumSlots; _iss sqInt pendingFinalizationSignals; @@ -1957,7 +1957,7 @@ 0 }; char * breakSelector; sqInt breakSelectorLength = -1; -const char *interpreterVersion = "Croquet Closure Stack VM [StackInterpreter VMMaker.oscog-eem.581]"; +const char *interpreterVersion = "Croquet Closure Stack VM [StackInterpreter VMMaker.oscog-eem.584]"; volatile int sendTrace; sqInt suppressHeartbeatFlag; @@ -13726,7 +13726,8 @@ (primBits & 0x1FF) + ((((usqInt) primBits) >> 19) & 512)) # endif /* MULTIPLEBYTECODESETS */ ; - assert(primitiveFunctionPointer == (functionPointerForinClass(primIndex, nilObject()))); + assert((primIndex == 117) + || (primitiveFunctionPointer == (functionPointerForinClass(primIndex, nilObject())))); assert(GIV(argumentCount) == (argumentCountOf(GIV(newMethod)))); accessorDepth = ((primIndex == 117) && (primitiveFunctionPointer != primitiveExternalCall) @@ -22897,7 +22898,7 @@ obj = (numSlots == 0xFF ? GIV(oldSpaceStart) + (BaseHeaderSize) : GIV(oldSpaceStart)); - while ((((usqInt) obj)) < (((usqInt) GIV(endOfMemory)))) { + while ((((usqInt) obj)) < (((usqInt) GIV(freeOldSpaceStart)))) { if (((longAt(obj)) & 0x3FFFFF) == 0) { /* begin swizzleFieldsOfFreeChunk: */ field = longAt((obj + (BaseHeaderSize)) + (0 << 2)); @@ -24147,7 +24148,7 @@ static sqInt isValidSegmentBridge(sqInt objOop) { - return (addressCouldBeObj(objOop - (BaseHeaderSize))) + return (addressCouldBeObj(objOop)) && ((((longAt(objOop)) & 0x3FFFFF) == 3) && (((rawNumSlotsOf(objOop)) == 0xFF) || ((numSlotsOfAny(objOop)) == 0))); Modified: branches/Cog/spurstacksrc/vm/interp.c =================================================================== --- branches/Cog/spurstacksrc/vm/interp.c 2014-01-14 01:59:30 UTC (rev 2845) +++ branches/Cog/spurstacksrc/vm/interp.c 2014-01-14 02:14:07 UTC (rev 2846) @@ -1,9 +1,9 @@ /* Automatically generated by - CCodeGeneratorGlobalStructure VMMaker.oscog-eem.581 uuid: 926531fd-20be-42b4-9332-015b90e6734f + CCodeGeneratorGlobalStructure VMMaker.oscog-eem.584 uuid: f8e3994b-6462-4df0-bc8e-c71034adf6aa from - StackInterpreter VMMaker.oscog-eem.581 uuid: 926531fd-20be-42b4-9332-015b90e6734f + StackInterpreter VMMaker.oscog-eem.584 uuid: f8e3994b-6462-4df0-bc8e-c71034adf6aa */ -static char __buildInfo[] = "StackInterpreter VMMaker.oscog-eem.581 uuid: 926531fd-20be-42b4-9332-015b90e6734f " __DATE__ ; +static char __buildInfo[] = "StackInterpreter VMMaker.oscog-eem.584 uuid: f8e3994b-6462-4df0-bc8e-c71034adf6aa " __DATE__ ; char *__interpBuildInfo = __buildInfo; @@ -1268,11 +1268,11 @@ _iss usqLong longRunningPrimitiveStopUsecs; _iss sqInt numSegInfos; _iss sqInt statCheckForEvents; +_iss usqInt freeOldSpaceStart; _iss usqInt lowSpaceThreshold; _iss usqLong statGCEndUsecs; _iss sqInt classTableIndex; _iss sqInt externalPrimitiveTableFirstFreeIndex; -_iss usqInt freeOldSpaceStart; _iss sqInt lkupClass; _iss sqInt metaclassNumSlots; _iss sqInt pendingFinalizationSignals; @@ -1954,7 +1954,7 @@ 0 }; char * breakSelector; sqInt breakSelectorLength = -1; -const char *interpreterVersion = "Croquet Closure Stack VM [StackInterpreter VMMaker.oscog-eem.581]"; +const char *interpreterVersion = "Croquet Closure Stack VM [StackInterpreter VMMaker.oscog-eem.584]"; volatile int sendTrace; sqInt suppressHeartbeatFlag; @@ -13717,7 +13717,8 @@ (primBits & 0x1FF) + ((((usqInt) primBits) >> 19) & 512)) # endif /* MULTIPLEBYTECODESETS */ ; - assert(primitiveFunctionPointer == (functionPointerForinClass(primIndex, nilObject()))); + assert((primIndex == 117) + || (primitiveFunctionPointer == (functionPointerForinClass(primIndex, nilObject())))); assert(GIV(argumentCount) == (argumentCountOf(GIV(newMethod)))); accessorDepth = ((primIndex == 117) && (primitiveFunctionPointer != primitiveExternalCall) @@ -22888,7 +22889,7 @@ obj = (numSlots == 0xFF ? GIV(oldSpaceStart) + (BaseHeaderSize) : GIV(oldSpaceStart)); - while ((((usqInt) obj)) < (((usqInt) GIV(endOfMemory)))) { + while ((((usqInt) obj)) < (((usqInt) GIV(freeOldSpaceStart)))) { if (((longAt(obj)) & 0x3FFFFF) == 0) { /* begin swizzleFieldsOfFreeChunk: */ field = longAt((obj + (BaseHeaderSize)) + (0 << 2)); @@ -24138,7 +24139,7 @@ static sqInt isValidSegmentBridge(sqInt objOop) { - return (addressCouldBeObj(objOop - (BaseHeaderSize))) + return (addressCouldBeObj(objOop)) && ((((longAt(objOop)) & 0x3FFFFF) == 3) && (((rawNumSlotsOf(objOop)) == 0xFF) || ((numSlotsOfAny(objOop)) == 0))); Modified: branches/Cog/spurstacksrc/vm/interp.h =================================================================== --- branches/Cog/spurstacksrc/vm/interp.h 2014-01-14 01:59:30 UTC (rev 2845) +++ branches/Cog/spurstacksrc/vm/interp.h 2014-01-14 02:14:07 UTC (rev 2846) @@ -1,5 +1,5 @@ /* Automatically generated by - CCodeGeneratorGlobalStructure VMMaker.oscog-eem.581 uuid: 926531fd-20be-42b4-9332-015b90e6734f + CCodeGeneratorGlobalStructure VMMaker.oscog-eem.584 uuid: f8e3994b-6462-4df0-bc8e-c71034adf6aa */ #define VM_PROXY_MAJOR 1 Modified: branches/Cog/spurstacksrc/vm/vmCallback.h =================================================================== --- branches/Cog/spurstacksrc/vm/vmCallback.h 2014-01-14 01:59:30 UTC (rev 2845) +++ branches/Cog/spurstacksrc/vm/vmCallback.h 2014-01-14 02:14:07 UTC (rev 2846) @@ -1,5 +1,5 @@ /* Automatically generated by - CCodeGeneratorGlobalStructure VMMaker.oscog-eem.581 uuid: 926531fd-20be-42b4-9332-015b90e6734f + CCodeGeneratorGlobalStructure VMMaker.oscog-eem.584 uuid: f8e3994b-6462-4df0-bc8e-c71034adf6aa */ #define VM_CALLBACK_INC 1 Modified: branches/Cog/src/plugins/SqueakSSL/SqueakSSL.c =================================================================== --- branches/Cog/src/plugins/SqueakSSL/SqueakSSL.c 2014-01-14 01:59:30 UTC (rev 2845) +++ branches/Cog/src/plugins/SqueakSSL/SqueakSSL.c 2014-01-14 02:14:07 UTC (rev 2846) @@ -1,9 +1,9 @@ /* Automatically generated by - VMPluginCodeGenerator VMMaker.oscog-eem.580 uuid: 751b08d4-d92e-440a-b3f6-cb2c76f52514 + VMPluginCodeGenerator VMMaker.oscog-eem.584 uuid: f8e3994b-6462-4df0-bc8e-c71034adf6aa from - SqueakSSLPlugin SqueakSSL-Plugin-eem.4 uuid: 5e2ad007-e58e-4c86-9657-47e0cfe12918 + SqueakSSLPlugin VMMaker.oscog-eem.584 uuid: f8e3994b-6462-4df0-bc8e-c71034adf6aa */ -static char __buildInfo[] = "SqueakSSLPlugin SqueakSSL-Plugin-eem.4 uuid: 5e2ad007-e58e-4c86-9657-47e0cfe12918 " __DATE__ ; +static char __buildInfo[] = "SqueakSSLPlugin VMMaker.oscog-eem.584 uuid: f8e3994b-6462-4df0-bc8e-c71034adf6aa " __DATE__ ; @@ -101,9 +101,9 @@ struct VirtualMachine* interpreterProxy; static const char *moduleName = #ifdef SQUEAK_BUILTIN_PLUGIN - "SqueakSSL SqueakSSL-Plugin-eem.4 (i)" + "SqueakSSL VMMaker.oscog-eem.584 (i)" #else - "SqueakSSL SqueakSSL-Plugin-eem.4 (e)" + "SqueakSSL VMMaker.oscog-eem.584 (e)" #endif ;

1 0

[commit][2845] Fix paths in BocksIA32Plugin makefile
by commits＠squeakvm.org 14 Jan '14

14 Jan '14

Revision: 2845 Author: eliot Date: 2014-01-13 17:59:30 -0800 (Mon, 13 Jan 2014) Log Message: ----------- Fix paths in BocksIA32Plugin makefile Modified Paths: -------------- trunk/platforms/win32/plugins/BochsIA32Plugin/Makefile Modified: trunk/platforms/win32/plugins/BochsIA32Plugin/Makefile =================================================================== --- trunk/platforms/win32/plugins/BochsIA32Plugin/Makefile 2014-01-10 23:00:31 UTC (rev 2844) +++ trunk/platforms/win32/plugins/BochsIA32Plugin/Makefile 2014-01-14 01:59:30 UTC (rev 2845) @@ -4,9 +4,9 @@ DLLWRAPEXTRA := $(DLLTOOLEXTRA) LINK_WITH_GPP := yes -EXTRALIBS:= -L../../../processors/IA32/winbochs/cpu \ - -L../../../processors/IA32/winbochs/fpu \ - -L../../../processors/IA32/winbochs/disasm \ +EXTRALIBS:= -L$(PLATDIR)/../processors/IA32/winbochs/cpu \ + -L$(PLATDIR)/../processors/IA32/winbochs/fpu \ + -L$(PLATDIR)/../processors/IA32/winbochs/disasm \ -lcpu -lfpu -ldisasm include ../../Makefile.plugin

1 0

VM Maker: VMMaker.oscog-eem.584.mcz
by commits＠source.squeak.org 14 Jan '14

14 Jan '14

Eliot Miranda uploaded a new version of VMMaker to project VM Maker: http://source.squeak.org/VMMaker/VMMaker.oscog-eem.584.mcz ==================== Summary ==================== Name: VMMaker.oscog-eem.584 Author: eem Time: 13 January 2014, 5:36:41.137 pm UUID: f8e3994b-6462-4df0-bc8e-c71034adf6aa Ancestors: VMMaker.oscog-eem.583 Spur: Fix swizzling on startup. Enumerate up to freeOldSpaceStart, not endOfMemory. Fixes startup of e.g. bootstrapped Squeak4.3 image. Beef up asserts in initializeFromFreeChunks: (for bootstrap). Fix isValidSegmentBridge: to not subtract header size from bridge. Fix assert in checkForAndFollowForwardedPrimitiveState. =============== Diff against VMMaker.oscog-eem.583 =============== Item was changed: ----- Method: SpurMemoryManager>>adjustAllOopsBy: (in category 'snapshot') ----- adjustAllOopsBy: bytesToShift "Adjust all oop references by the given number of bytes. This is done just after reading in an image when the new base address of the object heap is different from the base address in the image, or when loading multiple segments that have been coalesced. Also set bits in the classTableBitmap corresponding to used classes." | obj | self countNumClassPagesPreSwizzle: bytesToShift; ensureAdequateClassTableBitmap. (bytesToShift ~= 0 or: [segmentManager numSegments > 1]) ifTrue: [self assert: self newSpaceIsEmpty. obj := self objectStartingAt: oldSpaceStart. + [self oop: obj isLessThan: freeOldSpaceStart] whileTrue: - [self oop: obj isLessThan: endOfMemory] whileTrue: [(self isFreeObject: obj) ifTrue: [self swizzleFieldsOfFreeChunk: obj] ifFalse: [self inClassTableBitmapSet: (self classIndexOf: obj). self swizzleFieldsOfObject: obj]. obj := self objectAfter: obj]] ifFalse: [self assert: self newSpaceIsEmpty. obj := self objectStartingAt: oldSpaceStart. [self oop: obj isLessThan: endOfMemory] whileTrue: [(self isFreeObject: obj) ifFalse: [self inClassTableBitmapSet: (self classIndexOf: obj)]. obj := self objectAfter: obj]]! Item was changed: ----- Method: SpurSegmentManager>>initializeFromFreeChunks: (in category 'simulation only') ----- initializeFromFreeChunks: freeChunks "For testing, create a set of segments using the freeChunks as bridges." + self assert: (freeChunks allSatisfy: [:f| manager hasOverflowHeader: f]). numSegments := freeChunks size. freeChunks do: [:f| manager initSegmentBridgeWithBytes: (manager bytesInObject: f) at: (manager startOfObject: f). self assert: (manager isSegmentBridge: f)]. segments := (1 to: numSegments) collect: [:i| | bridge start size | bridge := freeChunks at: i. start := i = 1 ifTrue: [manager newSpaceLimit] ifFalse: [manager addressAfter: (freeChunks at: i - 1)]. size := bridge + manager baseHeaderSize - start. SpurSegmentInfo new segStart: start; segSize: size; yourself]. manager setEndOfMemory: segments last segLimit. segments := CArrayAccessor on: segments. + freeChunks with: segments object do: + [:bridge :segment| + self assert: (self isValidSegmentBridge: bridge). + self assert: bridge = (self bridgeFor: segment)]! - freeChunks do: - [:bridge| self assert: (self isValidSegmentBridge: bridge)]! Item was changed: ----- Method: SpurSegmentManager>>isValidSegmentBridge: (in category 'testing') ----- isValidSegmentBridge: objOop "bridges bridge the gaps between segments. They are the last object in each segment." + ^(manager addressCouldBeObj: objOop) + and: [(manager isSegmentBridge: objOop) + and: [(manager hasOverflowHeader: objOop) - ^(manager addressCouldBeObj: objOop - manager baseHeaderSize) - and: [(manager isSegmentBridge: objOop) - and: [(manager hasOverflowHeader: objOop) or: [(manager numSlotsOfAny: objOop) = 0]]]! Item was changed: ----- Method: StackInterpreter>>checkForAndFollowForwardedPrimitiveState (in category 'primitive support') ----- checkForAndFollowForwardedPrimitiveState "In Spur a primitive may fail due to encountering a forwarder. On failure check the accessorDepth for the primitive and if non-negative scan the args to the depth, following any forwarders. Answer if any are found so the prim can be retried." <option: #SpurObjectMemory> | primIndex accessorDepth found | self assert: self successful not. found := false. primIndex := self primitiveIndexOf: newMethod. + self assert: (primIndex = 117 + or: [primitiveFunctionPointer = (self functionPointerFor: primIndex inClass: objectMemory nilObject)]). - self assert: (self - cCode: - [primitiveFunctionPointer = (self functionPointerFor: primIndex inClass: objectMemory nilObject)] - inSmalltalk: - [((primitiveFunctionPointer isInteger and: [primitiveFunctionPointer >= 1000]) - ifTrue: [#primitiveExternalCall] - ifFalse: [primitiveFunctionPointer]) = (self functionPointerFor: primIndex inClass: objectMemory nilObject)]). self assert: argumentCount = (self argumentCountOf: newMethod). accessorDepth := (primIndex = 117 and: [primitiveFunctionPointer ~~ #primitiveExternalCall]) ifTrue: [self primitiveAccessorDepthForExternalPrimitiveMethod: newMethod] ifFalse: [primitiveAccessorDepthTable at: primIndex]. accessorDepth < 0 ifTrue: [^false]. 0 to: argumentCount do: [:index| | oop | oop := self stackValue: index. (objectMemory isNonImmediate: oop) ifTrue: [(objectMemory isForwarded: oop) ifTrue: [self assert: index < argumentCount. "receiver should have been caught at send time." found := true. oop := objectMemory followForwarded: oop. self stackValue: index put: oop]. ((objectMemory hasPointerFields: oop) and: [objectMemory followForwardedObjectFields: oop toDepth: accessorDepth]) ifTrue: [found := true]]]. ^found!

1 0

Re: allObjectsDo: (was Re: [squeak-dev] The Inbox: System-cwp.662.mcz)
by David T. Lewis 14 Jan '14

14 Jan '14

On Mon, Jan 13, 2014 at 09:26:31AM -0800, Eliot Miranda wrote: > Hi David, > > On Sun, Jan 12, 2014 at 5:13 PM, David T. Lewis <lewis(a)mail.msen.com> wrote: > > > On Sun, Jan 12, 2014 at 09:53:27PM +0100, Bert Freudenberg wrote: > > > > > > On 12.01.2014, at 20:42, David T. Lewis <lewis(a)mail.msen.com> wrote: > > > > > > > It is worth noting that allObjectsDo: relies on assumptions about how > > > > the objects memory works internally. It requires that #someObject will > > > > always answer the object at the lowest address in the object memory, > > and > > > > also that a newly allocated object will always be placed at a higher > > > > address location than all other objects. Either of these assumptions is > > > > likely to be a problem as new and better object memories and garbage > > > > collectors are implemented. > > > > > > > > Dave > > > > > > Right (as Eliot's vm-dev post shows). > > > > > > So IMHO the only sensible semantics of allObjectsDo: is as in > > "allObjects do:" - > > > which might be implemented as a primitive in some VMs soonish. It > > *should not* > > > enumerate objects created after calling the method. > > > > > > > > > On Sun, Jan 12, 2014 at 12:01:00PM -0800, Eliot Miranda wrote: > > > > > > The bug is in implementing allObjects in terms of someObject and > > nextObject > > > in the first place. It's cheap and cheerful but horribly error-prone and > > > restrictive. It's cheap because the collection of objects doesn't have > > to > > > be created, and on the original 16-bit Smalltalk machines that was really > > > important. It's horribly restrictive because it assumes much about the > > > implementation. > > > > > > Before closures a sentinel wasn't even needed because enumerating the > > block > > > didn't create a new object (the block context was reused). So the code > > had > > > to be rewritten just to support closures. > > > > > > Spur has a generation scavenger operating in a distinct new space and > > that > > > doesn't jive well with a consistent ordering at all. So far the system > > is > > > limping along by tenuring all objects on someObject and someInstance (so > > > that newSpace is either empty, or doesn't contain any instances of a > > > specific class) and having nextObject enumerate only objects in oldSpace. > > > > > > But I think now we can afford a primitive that answers all the objects > > > (remember that average object size means that such a collection will be ~ > > > 10% of the heap, average object size in Squeak V3 is about 10.6 words). > > At > > > least that's what Spur will do, along with an allInstancesOf: primitive. > > > And then the become example won't cause any problems at all. Far more > > > reliable. I suppose there are circumstances when enumerating without a > > > container is the only feasible approach, but VisualWorks has got along > > with > > > only an allObjects primitive for a long time now. I suspect we can too. > > > > > > > Implementation attached. Works on interpreter VM, not yet tested on Cog but > > it should be ok there also. If no objections or better suggestions I will > > commit it to VMMaker tomorrow. > > > > InterpreterPrimitives>>primitiveAllObjects > > "Answer an array of all objects that exist when the primitive is > > called, excluding those > > that may be garbage collected as a side effect of allocating the > > result array. Multiple > > references to nil in the last slots of the array are an indication > > that garbage collection > > occured, such that some of the unreferenced objects that existed > > at the time of calling > > the primitive no longer exist. Sender is responsible for handling > > multiple references to > > nil in the result array." > > > > Instead of filling the unused slots with nil or 0, I think you should > shorten the object so that it contains each object only once, and contains > only the objects. Cog contains some code for shortening. See > [New]ObjectMemory>>shorten:toIndexableSize:. > That would be a better solution. However, I cannot offer an implementation in the near term because of: shorten: obj toIndexableSize: nSlots "Currently this works for pointer objects only, and is almost certainly wrong for 64 bits." Given that this is currently intended for pointer objects, it is probably fairly straightforward to get it working on the 64-bit object memory. In fact, it might already work as written. But I think that it will take some time to test so it's not going to happen tonight. We could consider a variation on Bert's suggestion, in which the result array might have trailing zeros if garbage collection has occurred. Later the primitive can be improved with shorten:toIndexableSize: after which the trailing zeros will never occur in practice. That would still put the burden on the image to ignore the trailing junk, so I don't know if it would be worth doing. Dave

3 2

VM Maker: Cog-eem.128.mcz
by commits＠source.squeak.org 13 Jan '14

13 Jan '14

Eliot Miranda uploaded a new version of Cog to project VM Maker: http://source.squeak.org/VMMaker/Cog-eem.128.mcz ==================== Summary ==================== Name: Cog-eem.128 Author: eem Time: 13 January 2014, 1:10:54.717 pm UUID: c941142d-2ead-413e-b128-c9c5a2777b4a Ancestors: Cog-eem.127 Spur bootstrap: Rescue the launchSaved short-cut by hacking newSpace initialization. Add report of average object size to the image bootstrap code. Provide a bigger eden so that the launchSaved route can accomodate the interpreter reserve. =============== Diff against Cog-eem.127 =============== Item was changed: Object subclass: #SpurBootstrap + instanceVariableNames: 'oldHeap newHeap oldHeapSize newHeapSize oldHeapNumObjs newHeapNumObjs map reverseMap classToIndex oldInterpreter lastClassTablePage literalMap methodClasses installedPrototypes sizeSym rehashSym classMetaclass' - instanceVariableNames: 'oldHeap newHeap oldHeapSize newHeapSize map reverseMap classToIndex oldInterpreter lastClassTablePage literalMap methodClasses installedPrototypes sizeSym rehashSym classMetaclass' classVariableNames: 'ImageHeaderFlags ImageName ImageScreenSize TransformedImage' poolDictionaries: 'VMObjectIndices' category: 'Cog-Bootstrapping'! !SpurBootstrap commentStamp: 'eem 9/11/2013 05:45' prior: 0! SpurBootstrap bootstraps an image in SpurMemoryManager format from a Squeak V3 + closures format. e.g. (SpurBootstrap32 new on: '/Users/eliot/Cog/startreader.image') transform; launch Bootstrap issues: - should it implement a deterministic Symbol identityHash? This means set a Symbol's identityHash at instance creation time based on its string hash so that e.g. MethodDIctionary instances have a deterministic order and don't need to be rehashed on load. - should it collapse ContextPart and MethodContext down onto Context (and perhaps eliminate BlockContext)? Instance Variables classToIndex: <Dictionary> lastClassTablePage: <Integer> map: <Dictionary> methodClasses: <Set> newHeap: <SpurMemoryManager> oldHeap: <NewObjectMemory> oldInterpreter: <StackInterpreterSimulator> reverseMap: <Dictionary> symbolMap: <Dictionary> classToIndex - oldClass to new classIndex map lastClassTablePage - oop in newHeap of last classTable page. U<sed in validation to filter-out class table. methodClasses - cache of methodClassAssociations for classes in which modified methods are installed map - oldObject to newObject map newHeap - the output, bootstrapped image oldHeap - the input, image oldInterpreter - the interpreter associated with oldHeap, needed for a hack to grab WeakArray reverseMap - newObject to oldObject map symbolMap - symbol toi symbol oop in oldHeap, used to map prototype methdos to methods in oldHeap! Item was changed: ----- Method: SpurBootstrap>>bootstrapImage (in category 'bootstrap image') ----- bootstrapImage oldHeap fullGC. + self measureOldHeapPostGC. - oldHeapSize := oldHeap freeStart. self initMaps. self findRequiredGlobals. Transcript cr; nextPutAll: 'transforming image...'; flush. self cloneNilTrueAndFalse. self allocateFreeLists. self buildClassMap. self allocateClassTable. self cloneObjects. self fillInObjects. self fillInClassTable. newHeapSize := newHeap freeStart. + newHeap initializePostBootstrap. + self measureNewHeapPostInitPostBootstrap! - newHeap initializePostBootstrap! Item was changed: ----- Method: SpurBootstrap>>launch:simulatorClass:headerFlags: (in category 'testing') ----- launch: heap simulatorClass: simulatorClass headerFlags: headerFlags | sim methodCacheSize | sim := simulatorClass onObjectMemory: heap. heap coInterpreter: sim. (sim class allInstVarNames includes: 'cogCodeSize') ifTrue: [sim initializeInterpreter: 0. methodCacheSize := sim methodCache size * heap wordSize. sim instVarNamed: 'heapBase' put: heap startOfMemory; instVarNamed: 'numStackPages' put: 8; instVarNamed: 'cogCodeSize' put: 1024*1024; moveMethodCacheToMemoryAt: sim cogCodeSize + sim computeStackZoneSize; movePrimTraceLogToMemoryAt: sim cogCodeSize + sim computeStackZoneSize + methodCacheSize; "sendTrace: 1+ 2 + 8 + 16;" initializeCodeGenerator] ifFalse: [sim initializeInterpreter: 0]. + heap + initializeNewSpaceVariables; + bootstrapping: false; + assimilateNewSegment: (heap segmentManager segments at: 0). sim setImageHeaderFlagsFrom: headerFlags; imageName: ImageName; flushExternalPrimitives; openAsMorph; transcript: Transcript. "deep copy copies this" "sim instVarNamed: 'printSends' put: true; instVarNamed: 'printReturns' put: true; instVarNamed: 'methodDictLinearSearchLimit' put: SmallInteger maxVal." "for now" heap setCheckForLeaks: 0; runLeakCheckerForFullGC: true. sim halt; run! Item was added: + ----- Method: SpurBootstrap>>measureNewHeapPostInitPostBootstrap (in category 'stats') ----- + measureNewHeapPostInitPostBootstrap + | savedEndOfMemory | + "need to hack around the fact that newHeap isn't all there yet. + In particular, it has no freeList so can't free space from + freeOldSpaceStart to endOfMemory to make oldSpace enumerable." + newHeapNumObjs := 0. + savedEndOfMemory := newHeap endOfMemory. + newHeap setEndOfMemory: newHeap freeOldSpaceStart. + newHeap allObjectsDo: [:o| newHeapNumObjs := newHeapNumObjs + 1]. + newHeap setEndOfMemory: savedEndOfMemory! Item was added: + ----- Method: SpurBootstrap>>measureOldHeapPostGC (in category 'stats') ----- + measureOldHeapPostGC + oldHeapSize := oldHeap freeStart. + oldHeapNumObjs := 0. + oldHeap allObjectsDo: [:o| oldHeapNumObjs := oldHeapNumObjs + 1]! Item was changed: ----- Method: SpurBootstrap>>on: (in category 'initialize-release') ----- on: imageName StackInterpreter initializeWithOptions: Dictionary new. oldInterpreter := StackInterpreterSimulator new. oldInterpreter openOn: imageName extraMemory: 0. oldHeap := oldInterpreter objectMemory. newHeap := Spur32BitMMLESimulator new. newHeap allocateMemoryOfSize: (oldHeap youngStart * 3 / 2 roundUpTo: 1024 * 1024) + newSpaceSize: 2 * 1024 * 1024 - newSpaceSize: 1024 * 1024 stackSize: 1024 * 1024 codeSize: 0. newHeap setCheckForLeaks: 15 - 6. "don't check become; or newSpace; soooo many rehashes in bootstrap" newHeap bootstrapping: true. self initMaps! Item was added: + ----- Method: SpurBootstrap>>reportSizes (in category 'bootstrap image') ----- + reportSizes + | change oldAvgBytes newAvgBytes | + change := newHeapSize - oldHeapSize / oldHeapSize. + oldAvgBytes := oldHeapSize asFloat / oldHeapNumObjs. + Transcript + nextPutAll: 'done.'; cr; + nextPutAll: 'old heap size: '; print: oldHeapSize; tab; + nextPutAll: ' (avg obj bytes '; print: (oldAvgBytes roundTo: 0.01); nextPutAll: ' words '; print: (oldAvgBytes / self wordSize roundTo: 0.01); nextPut: $); cr; + nextPutAll: 'initial new heap size: '; print: newHeapSize; cr; + nextPutAll: 'change: '; print: (change * 100.0 roundTo: 0.01); nextPut: $%; cr; + flush. + newHeapSize := newHeap endOfMemory + - newHeap scavenger eden limit + - newHeap totalFreeListBytes. + change := newHeapSize - oldHeapSize / oldHeapSize. + newAvgBytes := newHeapSize asFloat / newHeapNumObjs. + Transcript + nextPutAll: 'final new heap size: '; print: newHeapSize; tab; + nextPutAll: ' (avg obj bytes '; print: (newAvgBytes roundTo: 0.01); nextPutAll: ' words '; print: (newAvgBytes / self wordSize roundTo: 0.01); nextPut: $); cr; + nextPutAll: 'change: '; print: (change * 100.0 roundTo: 0.01); nextPut: $%; cr; + flush! Item was changed: ----- Method: SpurBootstrap>>transform (in category 'bootstrap image') ----- transform - | change | self rememberRehashSymbol. self findRequiredGlobals. self installModifiedMethods. self bootstrapImage. self validate. self rememberRehashSymbol. self rehashImage. self followForwardingPointers. self scavengeImage. self freeForwarders. self compactImage. + self reportSizes! - change := newHeapSize - oldHeapSize / oldHeapSize. - Transcript - nextPutAll: 'done.'; cr; - nextPutAll: 'old heap size: '; print: oldHeapSize; cr; - nextPutAll: 'initial new heap size: '; print: newHeapSize; cr; - nextPutAll: 'change: '; print: (change * 100.0 roundTo: 0.01); nextPut: $%; cr; - flush. - newHeapSize := newHeap endOfMemory - - newHeap scavenger eden limit - - newHeap totalFreeListBytes. - change := newHeapSize - oldHeapSize / oldHeapSize. - Transcript - nextPutAll: 'final new heap size: '; print: newHeapSize; cr; - nextPutAll: 'change: '; print: (change * 100.0 roundTo: 0.01); nextPut: $%; cr; - flush.!

1 0

← Newer
1
2
3
4
5
6
7
8
9
10
Older →

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

Vm-dev January 2014