Vm-dev October 2010

vm-dev@lists.squeakfoundation.org

27 participants
60 discussions

VM Maker: VMMaker-dtl.193.mcz
by squeak-dev-noreply＠lists.squeakfoundation.org 15 Oct '10

15 Oct '10

Dave Lewis uploaded a new version of VMMaker to project VM Maker: http://www.squeaksource.com/VMMaker/VMMaker-dtl.193.mcz ==================== Summary ==================== Name: VMMaker-dtl.193 Author: dtl Time: 15 October 2010, 12:14:38 pm UUID: c9cc3efd-173a-41c0-acbb-d9ea99337eec Ancestors: VMMaker-dtl.192 VMMaker 4.3.7 Adopt various updates from oscog branch. Add TNotImplementedNode to allow Slangbrowser to fail less awkwardly. Add Object>>buildCodeGeneratorUpTo: and Object>>codeGeneratorClass to enable SlangBrowser translation of classes (e.g. MemoryAccess) not part of normal class translation hierarchy.

1 0

How to access Interpreter code from plugin?
by Mariano Martinez Peck 13 Oct '10

13 Oct '10

Hi. Sorry if the question is newbie. I am still reading the blue book, so in case this is explained there I would just appreciate a link to it. The problem is the following: I have implemented some functionality in Interpreter to trace objects usage. I have modified all bytecodes and different primitives and now I am almost sure I am covering all the "methods sends". Now, I have a problem with bitOr: and bitAnd: for example. If I do 9999999999999999 bitOr: 8888888888888888888 I would like to trace both intances of LargePositiveInteger as used. LargePositiveInteger >> bitOr: bitOr: anInteger "Primitive. Answer an Integer whose bits are the logical OR of the receiver's bits and those of the argument. Fail if the receiver or argument is greater than 32 bits. See Object documentation whatIsAPrimitive." <primitive: 15> ^ super bitOr: anInteger So...clearly, this will fail since both obects are more than 32 bits. So....we will do ^ super bitOr: anInteger bitOr: n "Answer an Integer whose bits are the logical OR of the receiver's bits and those of the argument, n." | norm | <primitive: 'primDigitBitOr' module:'LargeIntegers'> norm := n normalize. ^ self digitLogic: norm op: #bitOr: length: (self digitLength max: norm digitLength) So...it will call the primDigitBitOr in the plugin LargeIntegers In Interpreter I put an instVar for example to flag if I need to trace or not, and several helpers and primitives methods. The "main" method is: Interpreter >> traceObjectUsage: rcvr ((self isIntegerObject: rcvr) not and: [hasToTrace]) ifTrue: [ self internalTurnOnUsedBit: rcvr. ] Now, the problem is, how can I trace that 8888888888888888888 was used??? 9999999999999999 is used, because at least one it goes trough the #normalSend Two options: 1) In the fail of Interprter >> primitiveBitOr: I check if the class is LargePositiveInteger, I mark it. I don't like this solution because actually I "think" it will be used, but this is just because of the implemetnation on bitOr: in the image. Maybe it changes, I don't know. 2) Ttry to trace this in the LargeInteger plugin. I would like this idea, but I found several problems: 2.a) It is ok that I cannot access/send messages to Interpreter from plugins? It seems I am only allowed to do to InterpreterPlugin. But, I don't want to duplicate all my code from Interpreter to InterpreterProxy. 2.b) So my idea was to do something like this: InterpreterProxy >> traceObjectUsage: anObject anObject markAsUsed And Object >> markAsUsed will just be a primitive that actually calls #traceObjectUsage Trying to do that, I failed because sqVirtualMachine.h was not generated automatically, thus such struct didn't understand traceObjectUsage. For that, I added by hand to sqVirtualMachine.h something like this: #if VM_PROXY_MINOR > 8 sqInt (*traceObjectUsage)(sqInt oop); #endif With this it compiles, but when trying to run (actually in the start) the image I get a bad access error. 2.c) why is not sqVirtualMemory, or let's say, InterpreterProxy methods not automatically generated with VMMaker? I am doing something wrong ? 2.d) I saw there are more than one sqVirtualMemory. One in "/platforms/Cross/vm" and another one in "/platforms/Cross/plugins/IA32ABI/sqVirtualMachine.h" . After testin a little, it seems in my case it is using the second one. Why we have 2? and why it is using the second one ? Ok, thanks for any help you can give me and sorry for so many (and probably newbie) questions. Best regards, mariano

3 7

[commit] r2329 - iOS make private sqGNU.h a copy of the standard one. (delete at some point in the future after evaluating powerpc performance issues)
by commits＠squeakvm.org 13 Oct '10

13 Oct '10

Author: johnmci Date: 2010-10-13 11:39:14 -0700 (Wed, 13 Oct 2010) New Revision: 2329 Modified: trunk/platforms/iOS/vm/sqGnu.h Log: iOS make private sqGNU.h a copy of the standard one. (delete at some point in the future after evaluating powerpc performance issues) Modified: trunk/platforms/iOS/vm/sqGnu.h =================================================================== --- trunk/platforms/iOS/vm/sqGnu.h 2010-10-13 18:37:38 UTC (rev 2328) +++ trunk/platforms/iOS/vm/sqGnu.h 2010-10-13 18:39:14 UTC (rev 2329) @@ -45,17 +45,24 @@ #include "sq.h" -#define CASE(N) case N: _##N: +#define CASE(N) case N: _##N: + #if defined(__powerpc__) || defined(PPC) || defined(_POWER) || defined(_IBMR2) || defined(__ppc__) - #define BREAK goto *jumpTableR[currentBytecode] - #define PPC_REG_JUMP ; register void **jumpTableR JP_REG; jumpTableR = &jumpTable[0] +# define JUMP_TABLE_PTR ; register void **jumpTableP JP_REG; jumpTableP= &jumpTable[0] +# define BREAK goto *jumpTableP[currentBytecode] #else - #define BREAK goto *jumpTable[currentBytecode] - #define PPC_REG_JUMP +# define JUMP_TABLE_PTR +# define BREAK goto *jumpTable[currentBytecode] #endif -#define PRIM_DISPATCH goto *jumpTable[GIV(primitiveIndex)] + +#if defined(SQ_USE_GLOBAL_STRUCT) +# define PRIM_DISPATCH goto *jumpTable[GIV(primitiveIndex)] +#else +# define PRIM_DISPATCH goto *jumpTable[primitiveIndex] +#endif + #define JUMP_TABLE \ - void *jumpTable[256]= { \ + static void *jumpTable[256]= { \ &&_0, &&_1, &&_2, &&_3, &&_4, &&_5, &&_6, &&_7, &&_8, &&_9, \ &&_10, &&_11, &&_12, &&_13, &&_14, &&_15, &&_16, &&_17, &&_18, &&_19, \ &&_20, &&_21, &&_22, &&_23, &&_24, &&_25, &&_26, &&_27, &&_28, &&_29, \ @@ -82,7 +89,8 @@ &&_230, &&_231, &&_232, &&_233, &&_234, &&_235, &&_236, &&_237, &&_238, &&_239, \ &&_240, &&_241, &&_242, &&_243, &&_244, &&_245, &&_246, &&_247, &&_248, &&_249, \ &&_250, &&_251, &&_252, &&_253, &&_254, &&_255 \ - } PPC_REG_JUMP; + } JUMP_TABLE_PTR + #define PRIM_TABLE \ static void *jumpTable[700]= { \ &&_0, &&_1, &&_2, &&_3, &&_4, &&_5, &&_6, &&_7, &&_8, &&_9, \ @@ -157,50 +165,69 @@ &&_690, &&_691, &&_692, &&_693, &&_694, &&_695, &&_696, &&_697, &&_698, &&_699, \ } - /* - IP_REG, FP_REG, SP_REG, CB_REG + /* + IP_REG, SP_REG, CB_REG the machine registers in which to place localIP, localFP, localSP and currentBytecode. Wins big on register-deficient architectures -- especially Intel. */ #if defined(__mips__) -# define IP_REG asm("$16") -# define SP_REG asm("$17") -# define CB_REG asm("$18") +# define IP_REG __asm__("$16") +# define SP_REG __asm__("$17") +# define CB_REG __asm__("$18") #endif #if defined(__sparc__) -# define IP_REG asm("%l0") -# define SP_REG asm("%l1") -# define CB_REG asm("%l2") +# define IP_REG __asm__("%l0") +# define SP_REG __asm__("%l1") +# define CB_REG __asm__("%l2") #endif #if defined(__alpha__) -# define IP_REG asm("$9") -# define SP_REG asm("$10") -# define CB_REG asm("$11") +# define IP_REG __asm__("$9") +# define SP_REG __asm__("$10") +# define CB_REG __asm__("$11") #endif #if defined(__i386__) -# define IP_REG asm("%esi") -# define SP_REG asm("%edi") -# define CB_REG asm("%ebx") +# define IP_REG __asm__("%esi") +# define SP_REG __asm__("%edi") +# if (__GNUC__ > 2) || ((__GNUC__ == 2) && (__GNUC_MINOR__ >= 95)) +# define CB_REG __asm__("%ebx") +# else +# define CB_REG /* avoid undue register pressure */ +# endif #endif #if defined(__powerpc__) || defined(PPC) || defined(_POWER) || defined(_IBMR2) || defined(__ppc__) -# define FOO_REG asm("13") -# define JP_REG asm("14") -# define IP_REG asm("15") -# define SP_REG asm("16") -# define CB_REG asm("17") +# define GP_REG __asm__("24") +# define JP_REG __asm__("25") +# define IP_REG __asm__("26") +# define SP_REG __asm__("27") +# define CB_REG __asm__("28") #endif #if defined(__hppa__) -# define IP_REG asm("%r18") -# define SP_REG asm("%r17") -# define CB_REG asm("%r16") +# define IP_REG __asm__("%r18") +# define SP_REG __asm__("%r17") +# define CB_REG __asm__("%r16") #endif #if defined(__mc68000__) -# define IP_REG asm("a5") -# define SP_REG asm("a4") -# define CB_REG asm("d7") +# define IP_REG __asm__("a5") +# define SP_REG __asm__("a4") +# define CB_REG __asm__("d7") #endif +#if !defined(JP_REG) +# define JP_REG +#endif +#ifndef IP_REG +# define IP_REG +#endif +#if !defined(SP_REG) +# define SP_REG +#endif +#if !defined(CB_REG) +# define CB_REG +#endif +#if !defined(GP_REG) +# define GP_REG +#endif #if !defined(FOO_REG) # define FOO_REG /* nada */ #endif

1 0

[commit] r2328 - iOS take Cog soundplugin.h back to thrunk implementation
by commits＠squeakvm.org 13 Oct '10

13 Oct '10

Author: johnmci Date: 2010-10-13 11:37:38 -0700 (Wed, 13 Oct 2010) New Revision: 2328 Modified: branches/Cog/platforms/Cross/plugins/SoundPlugin/SoundPlugin.h Log: iOS take Cog soundplugin.h back to thrunk implementation Modified: branches/Cog/platforms/Cross/plugins/SoundPlugin/SoundPlugin.h =================================================================== --- branches/Cog/platforms/Cross/plugins/SoundPlugin/SoundPlugin.h 2010-10-11 04:29:11 UTC (rev 2327) +++ branches/Cog/platforms/Cross/plugins/SoundPlugin/SoundPlugin.h 2010-10-13 18:37:38 UTC (rev 2328) @@ -1,39 +1,28 @@ /* SoundPlugin.h - header for Sound Plugins :-) tim(a)sumeru.stanford.edu */ /* module initialization/shutdown */ -int soundInit(void); -int soundShutdown(void); +sqInt soundInit(void); +sqInt soundShutdown(void); /* sound output */ -int snd_AvailableSpace(void); -int snd_InsertSamplesFromLeadTime(int frameCount, int srcBufPtr, int samplesOfLeadTime); -int snd_PlaySamplesFromAtLength(int frameCount, int arrayIndex, int startIndex); -int snd_PlaySilence(void); -int snd_Start(int frameCount, int samplesPerSec, int stereo, int semaIndex); -int snd_Stop(void); +sqInt snd_AvailableSpace(void); +sqInt snd_InsertSamplesFromLeadTime(sqInt frameCount, void *srcBufPtr, sqInt samplesOfLeadTime); +sqInt snd_PlaySamplesFromAtLength(sqInt frameCount, void *srcBufPtr, sqInt startIndex); +sqInt snd_PlaySilence(void); +sqInt snd_Start(sqInt frameCount, sqInt samplesPerSec, sqInt stereo, sqInt semaIndex); +sqInt snd_Stop(void); /* sound input */ -int snd_SetRecordLevel(int level); -int snd_GetRecordLevel(void); // howard.stearns(a)qwaq.com October 24 2008 -int snd_StartRecording(int desiredSamplesPerSec, int stereo, int semaIndex); -int snd_StopRecording(void); +void snd_SetRecordLevel(sqInt level); +sqInt snd_StartRecording(sqInt desiredSamplesPerSec, sqInt stereo, sqInt semaIndex); +sqInt snd_StopRecording(void); double snd_GetRecordingSampleRate(void); -int snd_RecordSamplesIntoAtLength(int buf, int startSliceIndex, int bufferSizeInBytes); -void snd_Volume(double *left, double *right); //johnmci(a)smalltalkconsulting.com Nov 6th 2000 -void snd_SetVolume(double left, double right);//johnmci(a)smalltalkconsulting.com Nov 6th 2000 -int snd_EnableAEC(int trueOrFalse); //josh.gargus(a)teleplace.com March 23 2010 -int snd_SupportsAEC(); //josh.gargus(a)teleplace.com March 24 2010 +sqInt snd_RecordSamplesIntoAtLength(void *buf, sqInt startSliceIndex, sqInt bufferSizeInBytes); +void snd_Volume(double *left, double *right); /* johnmci(a)smalltalkconsulting.com Nov 6th 2000 */ +void snd_SetVolume(double left, double right); /* johnmci(a)smalltalkconsulting.com Nov 6th 2000 */ -int getNumberOfSoundPlayerDevices(void); // howard.stearns(a)qwaq.com October 20 2008 ... -int getNumberOfSoundRecorderDevices(void); -char * getDefaultSoundPlayer(void); -char * getDefaultSoundRecorder(void); -char * getSoundPlayerDeviceName(int i); -char * getSoundRecorderDeviceName(int i); -void setDefaultSoundPlayer(char *deviceName); -void setDefaultSoundRecorder(char *deviceName); - /* mixer interface */ -int snd_GetSwitch(int id, int captureFlag, int channel); -int snd_SetSwitch(int id, int captureFlag, int parameter); -int snd_SetDevice(int id, char *name); +sqInt snd_GetSwitch(sqInt identifier, sqInt captureFlag, sqInt channel); +sqInt snd_SetSwitch(sqInt identifier, sqInt captureFlag, sqInt parameter); +sqInt snd_SetDevice(sqInt identifier, char *name); +

1 0

FreeType fonts scan annoyance on windows
by Igor Stasenko 13 Oct '10

13 Oct '10

If you're on Windows, a freetype searching fonts everywhere on all drive letters. This causes opening annoying 'Insert disk in drive X:' dialog. guessWindowsFontFolderPath "Guess the location of the Windows font folder" | possibles | possibles := Set new. 'cdefghijklmnopqrstuvwxyz' do:[:drive | #('\windows\fonts' '\winnt\fonts') do:[:path | | d | (d := FileDirectory on: drive asString, ':', path) exists ifTrue:[possibles add: d]]]. possibles := possibles asSortedCollection: [:a :b | a directoryEntry creationTime >= b directoryEntry creationTime]. possibles ifNotEmpty:[^possibles first pathName]. ^nil Here what i found occasionally, when read ReadFile() function documentation in MSDN: If you are attempting to read from a floppy drive that does not have a floppy disk, the system displays a message box that prompts the user to retry the operation. To prevent the system from displaying this message box, call the SetErrorMode function with SEM_NOOPENFILEERRORBOX. But it will require changing VM to call SetErrorMode() during startup to prevent these promts from appearing. -- Best regards, Igor Stasenko AKA sig.

1 0

Building OSProcess plugin on win32 fails
by Igor Stasenko 13 Oct '10

13 Oct '10

I tried to build an OSProcess plugin on windoze, with VMMaker Name: VMMaker-dtl.186 Author: dtl Time: 19 September 2010, 11:03:20 am UUID: d2271d63-e922-47de-9bf7-161900f8e7d0 Ancestors: VMMaker-dtl.185 And it fails compiling, unable to reach FilePlugin and SocketPlugin headers.. make -C ./obj/Win32OSProcessPlugin -fMakefile.dll LIBNAME=Win32OSProcessPlugin \ PLATDIR=../../../platforms BASEDIR=../../. \ Win32OSProcessPlugin.dll make[1]: Entering directory `F:/projects/squeak/!trunk/src/winbuild/obj/Win32OSProcessPlugin' gcc -o ../.././obj/Win32OSProcessPlugin/Win32OSProcessPlugin.o -g -mpentium -mwindows -fomit-frame-pointer -funroll-loops -fschedule-insns2 -O2 -I. -Ic:/dx7sdk/include -I../.././src/vm -I../../../platforms/Win32/vm -I../../../platforms/Cross/vm -I../.././src/Win32OSProcessPlugin -I../../../platforms/Win32/plugins/Win32OSProcessPlugin -I../../../platforms/Cross/plugins/Win32OSProcessPlugin -DVM_VERSION=\""3.10.6 (release)"\" -DWIN32_FILE_SUPPORT -DNO_SERVICE -DNO_STD_FILE_SUPPORT -DNDEBUG -DLSB_FIRST -DVM_NAME=\"\" -DX86 -c ../.././src/Win32OSProcessPlugin/Win32OSProcessPlugin.c ../.././src/Win32OSProcessPlugin/Win32OSProcessPlugin.c:39: FilePlugin.h: No such file or directory ../.././src/Win32OSProcessPlugin/Win32OSProcessPlugin.c:40: SocketPlugin.h: No such file or directory make[1]: *** [Win32OSProcessPlugin.o] Error 1 make[1]: Leaving directory `F:/projects/squeak/!trunk/src/winbuild/obj/Win32OSProcessPlugin' make: *** [Win32OSProcessPlugin.dll] Error 2 looks like something wrong with headers search path, and i need to update/modify makefile(s). -- Best regards, Igor Stasenko AKA sig.

2 1

64-bit compilation beneath a 32-bit image...
by Eliot Miranda 13 Oct '10

13 Oct '10

Hi David, John, All, I'm working on integrating callbacks into the FFI and have the old version of callbacks working. In integrating with Cog I've realised some cleanliness can come from a slightly different approach, the details of which are irrelevant to my question, which is, when we're running a 32-bit image above a 32-bit VM compiled in 64-bit mode, is there any way from the image we can ask the VM if it is 64-bit or 32-bit? Am I to understand correctly that you, John, have arranged that when we're compiling the iOS VM on a 64-bit Mac OS X, sizeof(long) = 8, and everything still works? If not, the question is moot. If so, how from the image can I ask whether the VM is 64-bit, e.g. by asking what sizeof(long) or sizeof(void *) is? If there isn't a way, I propose we add vmParameterAt: 70 to answer the VM's underlying word size. Tis is different to vmParameterAt: 40, which is the image's underlying word size. I need the answer to this so that when a callback tries to answer e.g. an integer result it can tell whether it should use some Alien signedLongAt: N or Alien signedLongLongAt: N to answer the result, and what machine code (x86 vs x86-64) to use for the callback thunk. best Eliot

3 3

Mac VM: SecurityManager default untrustedUserDirectory
by Lukas Renggli 12 Oct '10

12 Oct '10

Depending on the VM version evaluating SecurityManager default untrustedUserDirectory gives inconsistent results: 4.2.5beta1U: '/Users/renggli/Library/Preferences/Squeak/Internet/My Squeak' (ok) 5.8b10: '/foobar/tooBar/forSqueak/bogus/' (strange) Looks like there is something missing in for 5.8b10? This makes certain tests of the Filesystem library fail. Lukas -- Lukas Renggli www.lukas-renggli.ch

4 5

Cog crash during GC compaction (5.8b9)
by Henrik Johansen 12 Oct '10

12 Oct '10

Just encountered my first Crash in awhile, attached is the error log. Not doing anything particularily interesting at that point, IIRC I was inspecting the WeakDependents, and had browsed a few Monticello repositories. Sorry I can't provide any more details/have time to dig deeper into details up front this time :( Cheers, Henry

1 0

[CfPart] BoF: Non-deterministic Programming for Manycore Systems at SPLASH'10
by Stefan Marr 12 Oct '10

12 Oct '10

============================================================================ Call for Participation Non-deterministic Programming for Manycore Systems Birds of a Feather Tuesday, October 19, starting 5:30pm Reno/Tahoe, Nevada, 2010. Co-located with SPLASH 2010 http://soft.vub.ac.be/~smarr/bof_non-determenistic_programming.html ============================================================================ Non-deterministic Programming: A Solution for the Manycore Challenge? Informal Workshop on the Opportunities of Embracing Non-determinism in a Highly Parallel World Goal ---- This BoF is meant to bring together people who are interested in solving the challenges that the fine-grained parallelism of manycore and exascale systems present. Today, non-determinism is usually considered a problem researchers have to solve when they confronted with parallel programming. However, non-deterministic programming and its potential has been the subject of research for decades past. With the increasing and unavoidable degree of parallelism, it may be the time to discuss it from a new perspective. Thus, the main question of the workshop is, what if the downsides of non-determinism are outweighed by its benefits? We like to discuss the question of how a programming model for non-deterministic parallel programming could be designed. Thus, we want to think about what challenges need to be overcome and what research questions need to be formulated. Of interest are also which classic problems, for instance sorting, could be solved in a non-deterministic manner, and what kind of computational problems could benefit from speedups provided by additional parallelism even when trading off precision for speed. Attendance ---------- This Birds of a Feather is supposed to propose and discus new ideas, without any restriction of the creativity of its attendees. Every attendee is welcome to give a brief presentation, show a demo, or just engage into a discussion with the audience. Renaissance Project ------------------- This BoF will also be used as a venue to present the Renaissance Virtual Machine (RVM) of IBM Research[3]. On this occasion, the RVM will be opensourced and attendees will have the opportunity to get introduced to the technical details of it. The RVM is meant to provide a common foundation for research regarding non-deterministic parallel programming for manycore systems. It is a Squeak/Pharo compatible Smalltalk VM supporting a classic shared-memory programming model. Thus, the programmer has the illusion of a single object heap on which Smalltalk Processes can work in parallel. On top of the RVM, Ly was developed to embrace non-determinism and to harness emergence. The first results will be reported in an Onward! presentation on Thursday afternoon [1]: Harnessing Emergence for Manycore Programming: Early Experience Integrating Ensembles, Adverbs, and Object-based Inheritance by David Ungar, Sam S. Adams Confirmed Participants ---------------------- David Ungar, Doug Kimelman, IBM Research Brief introduction into the Renaissance project and demo of the current status of their Ly/Sly language and manycore virtual machine. Max OrHai, Andrew Black, Portland State University Giving an overview of an example application to test the Renaissance ensemble code as well as an outlook to a scalable non-deterministic parallel sorting algorithm. Theo D'Hondt, Vrije Universiteit Brussel Lessons from the past: Recapitulating the relations of non-deterministic programming, AI, and highly parallel systems like Connection Machine. Organizers ---------- Theo D'Hondt, Software Languages Lab, Vrije Universiteit Brussel Stefan Marr, Software Languages Lab, Vrije Universiteit Brussel [1] http://splashcon.org/program/onward-short-papers/154-onward-short-papers-b [2] http://splashcon.org/schedule/tuesday-oct-19-/235 [3] https://researcher.ibm.com/researcher/view_project.php?id=245 -- Stefan Marr Software Languages Lab Vrije Universiteit Brussel Pleinlaan 2 / B-1050 Brussels / Belgium http://soft.vub.ac.be/~smarr Phone: +32 2 629 2974 Fax: +32 2 629 3525

1 0

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

Vm-dev October 2010