Vm-dev August 2011

vm-dev@lists.squeakfoundation.org

40 participants
111 discussions

Profiling and delay accuracy
by Andreas Raab 06 Aug '11

06 Aug '11

Hi Folks - For a long time now I had this feeling that MessageTally reports pretty bogus results in many situations without being able to precisely track down why this would be the case. Today I was looking at primitiveResponse() and, much to my surprise, noticed that the checkForInterrupts() that used to be there for years is gone (see Squeak3.6 for example which still has it). The sole reason for this check was to get both high delay accuracy and good profiling behavior. Consider this example: fc := FormCanvas on: Display. MessageTally spyOn:[ 1 to: 100 do:[:i| fc fillColor: Color white. fc printString. ]. ]. Using 3.6 this will be measured correctly as: **Tree** 87.5% {243ms} FormCanvas>>fillColor: |87.5% {243ms} GrafPort>>fillRect:offset: | 87.5% {243ms} GrafPort>>copyBits 12.5% {35ms} FormCanvas(Object)>>printString 12.5% {35ms} FormCanvas(Object)>>printStringLimitedTo: whereas the 3.10 reports it as **Tree** 99.6% {670ms} FormCanvas(Object)>>printString 94.9% {639ms} primitives 4.6% {31ms} FormCanvas(Object)>>printStringLimitedTo: I think you can see the problem and just imagine what this does to any cumulative results. I also wonder what that change does to platforms that don't have high-resolution asynchronous timers. This used to be the one place where we would catch timer changes at high resolution and unless the platform forces high resolution timer interrupts asynchronously (by resetting the interrupt counter) we must be loosing a lot in terms of delay accuracy. (it just so happens that the Windows VM *does* do asynchronous timer interrupts so I never noticed this but I'm very curious about the difference on Linux which typically still has pretty bad timer granularity). Any opinions? My feeling is that the timer check should go back in for all of the above reasons. At the very least a quickCheckForInterrupt() is required but I actually think putting the full check back in is worth it. Cheers, - Andreas

3 3

InterpreterProxy->getStackPointer is null
by Igor Stasenko 05 Aug '11

05 Aug '11

I just found that a pointer to this function is null. Means that if you will use interpreterProxy->getStackPointer() in plugin, you will crash VM. Could it be that it declared twice in sqVirtualMachine.c ? ... sqInt getStackPointer(void); /* Newsqueak FFI */ void *startOfAlienData(sqInt); usqInt sizeOfAlienData(sqInt); sqInt signalNoResume(sqInt); #if VM_PROXY_MINOR > 8 sqInt getStackPointer(void); /* Alien FFI */ ... (at least this code looks suspicious) and why at assignment point it does conversions? VM->getStackPointer = (sqInt *(*)(void))getStackPointer; while rest of functions don't require them? In header it declared as: sqInt *(*getStackPointer)(void); NBInterpreterProxy majorVersion 1 NBInterpreterProxy minorVersion 12 (means that it should be there :) -- Best regards, Igor Stasenko AKA sig.

2 3

VM Maker: CMakeVMMaker-IgorStasenko.132.mcz
by squeak-dev-noreply＠lists.squeakfoundation.org 05 Aug '11

05 Aug '11

Igor Stasenko uploaded a new version of CMakeVMMaker to project VM Maker: http://www.squeaksource.com/VMMaker/CMakeVMMaker-IgorStasenko.132.mcz ==================== Summary ==================== Name: CMakeVMMaker-IgorStasenko.132 Author: IgorStasenko Time: 5 August 2011, 4:31:56 pm UUID: 21e7df3e-9297-49c4-8e07-7ef29a9845d7 Ancestors: CMakeVMMaker-EstebanLorenzano.131 - included QuickTime framework by default (again)

1 0

[OT] Smalltalk on JVM
by Torsten Bergmann 04 Aug '11

04 Aug '11

At http://medianetwork.oracle.com/media/show/16803 scroll down and there you will find a new video "A Renaissance VM: One Platform, Many Languages" with a talk about the new "invokedynamic" (JSR292). John Rose also talks about all this in the context of Smalltalk. So if you are interested in ST and VM's this may be interesting too. There is also RTalk which maybe interesting in this context too, see http://astares.blogspot.com/2011/08/smalltalk-on-jvm.html Bye T. -- NEU: FreePhone - 0ct/min Handyspartarif mit Geld-zurück-Garantie! Jetzt informieren: http://www.gmx.net/de/go/freephone

2 1

Re: [Pharo-project] identityHash bits not incremented for new objects?
by Henrik Johansen 03 Aug '11

03 Aug '11

On Aug 2, 2011, at 5:09 51PM, David T. Lewis wrote: > On Tue, Aug 02, 2011 at 03:31:30PM +0200, Henrik Johansen wrote: >> >> On Aug 2, 2011, at 3:14 15PM, Igor Stasenko wrote: >> >>> On 2 August 2011 14:18, Henrik Johansen <henrik.s.johansen(a)veloxit.no> wrote: >>>> >>>> On Aug 1, 2011, at 8:43 53AM, Igor Stasenko wrote: >>>> >>>>> I found that #testBecomeIdentityHash sometimes failing, sometimes not. >>>>> >>>>> It seems like VM 'forgets' to produce different identityHash bits for >>>>> two consequently allocated objects, >>>>> while test assumes that they are always different. >>>>> >>>>> If i insert a statement (see the code), test no longer fails. >>>>> >>>>> testBecomeIdentityHash >>>>> "Note. The identity hash of both objects seems to change after the become:" >>>>> >>>>> | a b c d | >>>>> >>>>> a := 'ab' copy. >>>>> b := 'cd' copy. >>>>> >>>>>>>> [ b identityHash = a identityHash ] whileTrue: [ b := b copy ]. >>>>> >>>>> c := a. >>>>> d := b. >>>>> >>>>> a become: b. >>>>> >>>>> self >>>>> assert: a identityHash = c identityHash; >>>>> assert: b identityHash = d identityHash; >>>>> deny: a identityHash = b identityHash. >>>>> >>>>> A simple piece of code reveals the problem: >>>>> >>>>> (1 to: 20) collect: [:i | >>>>> 'ab' copy basicIdentityHash ] #(954 954 955 955 956 956 957 957 958 >>>>> 958 959 959 960 960 961 961 962 962 963 963) >>>>> >>>>> >>>>> -- >>>>> Best regards, >>>>> Igor Stasenko AKA sig. >>>> >>>> What is the point of this test in the first place? >>>> To me it seems to test whether a VM's implementation of become: adheres to some properties. >>>> At least, VM bugs in traversal of references during become is the only way I could see the two first asserts EVER being false. >>>> >>>> As for the last assert (well, deny actually), in the context of what it's supposed to test as established above, it seems to me to say a VM should never give the same identityHash to subsequently copied objects. Which means >>>> a) It has nothing to do with become and identityHash, thus is in the test. >>>> b) Is this a property any code actually rely on? >>>> >>>> If not, why would we assert such a restriction on the VM implementation, or is it harmless? >>>> >>>> Either way, modifying the test to iterate till they ARE different, would make it meaningless in either case (and possible infinite loop in case of a really bad VM). >>>> >>> >>> There is strong reason to test identityHashes, because VM should >>> preserve an identity hash for oops. >>> >>> But of course test is wrong , because to test that , it should be like >>> following: >>> >>> | a b hash1 hash2 | >>> >>> a := 'ab' copy. >>> b := 'cd' copy. >>> hash1 := a identityHash. >>> hash2 := b identityHash. >>> >>> a become: b. >>> >>> self >>> assert: a identityHash = hash1; >>> assert: b identityHash = hash2. >>> >>> >>> -- >>> Best regards, >>> Igor Stasenko AKA sig. >>> >> >> Please include a comment with that. :) >> Like this has shown, it's rather hard to know wtf to do when you can't unambiguously read what its actually intended to test from the code itself. > > Henry, FYI Igor has found a small but important bug in the VM. There is > some additional follow up discussion on the vm-dev list that you can read > starting here: > > <http://lists.squeakfoundation.org/pipermail/vm-dev/2011-August/009006.html> > > Dave I saw, and it does eliminate the common case where two small objects get the same identityHash. Though, it doesn't really _ensure_ that two subsequently allocated objects will have different identityHashes, does it? IE. if an object is allocated with size which increases (freeStart >> someInt) by 4096, won't you still see the next object getting the same identityHash? Not to mention power-of-two sized objects will have a _really_ low number of different hashes. (though none subsequently equal) Hence my question if this is really an important property to require of the VM. Considering the amount of times I've seen complaints of Set performance due to bad hashing under Cog (in other words, 0), I don't think so. Like Nicolas, I am rather skeptic to the comment saying the values are "well-distributed and fairly random" (which is both true and false, depending on how you interpret "distributed" and "random") though, and would like it if it additionally pointed out edge-cases like above. Cheers, Henry PS. What's the state of NewObjectMemory usage in VMMaker? When I loaded it, I couldn't find definition/initialization of HashMaskUnshifted, in VMMaker.oscog those were in VMSqueakV3ObjectRepresentationConstants/ObjectMemory respectively.

2 1

Object format(s) as a contract with VM?
by Igor Stasenko 03 Aug '11

03 Aug '11

Hello, i thought about object formats and thought how to make them more flexible and came to following idea. Today's squeak and cog vms using a 4 bits in object header which denoting an object format: So, there is 16 possible object formats supported by VM: formatOf: oop " 0 no fields 1 fixed fields only (all containing pointers) 2 indexable fields only (all containing pointers) 3 both fixed and indexable fields (all containing pointers) 4 both fixed and indexable weak fields (all containing pointers). 5 unused 6 indexable word fields only (no pointers) 7 indexable long (64-bit) fields (only in 64-bit images) 8-11 indexable byte fields only (no pointers) (low 2 bits are low 2 bits of size) 12-15 compiled methods: # of literal oops specified in method header, followed by indexable bytes (same interpretation of low 2 bits as above) " And virtual machine (mostly an ObjectMemory class) knows difference between those numbers and how to deal with objects depending what format they having. The bad side of it, that most semantics around formats is hardcoded and spreaded across various methods, like: lastPointerOf: oop "Return the byte offset of the last pointer field of the given object. Works with CompiledMethods, as well as ordinary objects. Can be used even when the type bits are not correct." | fmt sz methodHeader header contextSize | <inline: true> <asmLabel: false> header := self baseHeader: oop. fmt := self formatOfHeader: header. fmt <= 4 ifTrue: [(fmt = 3 and: [self isContextHeader: header]) ifTrue: ["contexts end at the stack pointer" contextSize := self fetchStackPointerOf: oop. ^ CtxtTempFrameStart + contextSize * BytesPerWord]. sz := self sizeBitsOfSafe: oop. ^ sz - BaseHeaderSize "all pointers"]. fmt < 12 ifTrue: [^ 0]. "no pointers" "CompiledMethod: contains both pointers and bytes:" methodHeader := self longAt: oop + BaseHeaderSize. ^ (methodHeader >> 10 bitAnd: 255) * BytesPerWord + BaseHeaderSize See how smelly this code is? What i was thinking about is , that what if number which denotes an object format instead of just being a number, point to a table of functions, which implement a behavior, specific for that format. Then the above method could be turned into something as short and nice as following: lastPointerOf: oop <inline: true> <asmLabel: false> | table | table := (self formatTableOf: oop). table performSomeOperation: oop param: param which , when translated to C, will look like: sqInt formatNumber; formatNumber = fetchFormatOf(oop); return ObjectFormats[formatNumber].performSomeOperation(oop, param). So, the table is simple list of function pointers (which very closely resembling a method dictionary in smalltalk), to provide an implementation of certain behavior for given object format: struct ObjectFormatFunctions { int (*)(sqInt oop); // firstFixedSlot int (*)(sqInt oop); // numFixedSlots: void* (*)(sqInt oop); // lastPointerOf: .. firstPointerOf: .. numIndexableSlots: .. indexableSlotSize: ... etc ) so, then instead of hardcoding and spreading the smelly code around the VM, we can write something like: (self formatTableOf: oop) performSomeOperation: oop params: params and code no longer needs to know if format = 0 or 4 or whatever, but just dispatching to function using formats table. Then all methods which dealing with object format could be placed in nicely organized class structure and can be translated to C in form of tables. The downside of it, of course, that additional level of indirection will cause a serious slowdown to interpreter. Because C compiler don't sees directly the implementation of operation for certain format and must call a function, pointer to which held in a corresponding format table. Moreover, you cannot inline those functions, since you don't know which one will be used at concrete place because of dynamic dispatch. How we can deal with it? Well, we still can hardcode the stuff (granting that we are not modifying object format table(s) at runtime), so we can write: fmt = 5 ifTrue: [ ^ self inlineFunctionOfFormat: 5 oop: oop params: params ] this requires some trickery in C code generator, but it is still doable. So, in overall i think that drawbacks could be cleverly mitigated if not completely avoided. So, lets talk what opportinities it could give us: VM could support a dynamically changeable object format(s), if we extend the number of possible formats to more than 16. So, for first 16 numbers things will be hardcoded as today, but for higher numbers, VM will always dispatch using format table. The cost of indirection can be amortized quite well especially in presence of JIT (it can simply inline the code for corresponding semantics to access various bits in object when compiling the methods of class with such format). Then we could dynamically change (or create new formats) at run time. But currently we have only 4 bits for object format. Where we can get some more? A solution is extremely simple! We have a 5 bits in object header for compact class index. If we merge them with object format bits 4+5 = 9 -- 512 possible object formats! The trick is that we still can pretty easily identify special objects by reserving a concrete format number for them (hey we have plenty of numbers) So, then format numbers ranged from 0..15 will identify hardcoded formats , known by VM from beginning. And formats from 16..48 is for special objects. and rest 512-48 is dynamically defined. Of course, we could change the object header to look differently for new VM, then there may be more (or less) than 9 bits for object format. Not really matters, because concept remains the same: VM could know that the only field which it can safely access by itself is an object header. For accessing other fields (if any) it should use functions provided in object format table. And the last thing: why i titled topic format as a contract? Because the next logical step of it is to imprint formats in a kind of form of manifest, and store them in image file. Then VM, when booting an image will read manifest and translate it into machine code before first attempt to access any object(s) in object memory. So, what you thinking about it? It could be an overkill , but i like the idea that with such approach we could change the object format(s) dynamically at run time, without the need of changing VM. It also structuring the code in VM quite nicely which will serve clarity. It also opens a wide field for experiments with different object formats. -- Best regards, Igor Stasenko AKA sig.

4 5

VM Maker: VMMaker-dtl.245.mcz
by squeak-dev-noreply＠lists.squeakfoundation.org 02 Aug '11

02 Aug '11

Dave Lewis uploaded a new version of VMMaker to project VM Maker: http://www.squeaksource.com/VMMaker/VMMaker-dtl.245.mcz ==================== Summary ==================== Name: VMMaker-dtl.245 Author: dtl Time: 2 August 2011, 7:11:17 am UUID: 0035fb0a-e4af-4bc3-b047-30b2c2f49cdd Ancestors: VMMaker-dtl.244 VMMaker 4.7.2 Reference <http://lists.squeakfoundation.org/pipermail/vm-dev/2011-August/009006.html> Fix NewObjectMemory>>newObjectHash to use shiftForWord rather than bytesPerWord (check to ensure corresponding fix in oscog). Reference Mantis 6987: signed32BitValueOf:, signed64BitValueOf: etc. broken Discussion at <http://lists.squeakfoundation.org/pipermail/vm-dev/2011-July/008905.html> Fix #signed32BitValueOf: and #signed64BitValueOf: which are reported broken for some optimizing C compilers. Cast to unsigned integer to prevent undefined behavior on arithmetic overflow. Add InterpreterPrimitivesTest to provide some coverage of #signed64BitValueOf: Follow up required - Test to verify fix on problematic compilers. Merge with oscog.

1 0

[commit] r2484 - Move dialect-independence macros to sq.h and use them on Mac OS.
by commits＠squeakvm.org 01 Aug '11

01 Aug '11

Author: eliot Date: 2011-08-01 14:36:19 -0700 (Mon, 01 Aug 2011) New Revision: 2484 Modified: branches/Cog/nscogbuild/unixbuild/HowToBuild branches/Cog/platforms/Cross/vm/sq.h branches/Cog/platforms/Cross/vm/sqSCCSVersion.h branches/Cog/platforms/Mac OS/vm/sqMacMain.c branches/Cog/platforms/Mac OS/vm/sqMacUnixCommandLineInterface.c branches/Cog/platforms/unix/vm/sqUnixExternalPrims.c branches/Cog/platforms/unix/vm/sqUnixMain.c branches/Cog/unixbuild/HowToBuild Log: Move dialect-independence macros to sq.h and use them on Mac OS. Format platforms/unix/vm/sqUnixExternalPrims.c so I can make sense of it and hence avoid calling tryLoading with a null plugin. Update unix HowToBuilds with info on avoiding gcc 3.4.x and -O2. Modified: branches/Cog/nscogbuild/unixbuild/HowToBuild =================================================================== --- branches/Cog/nscogbuild/unixbuild/HowToBuild 2011-07-31 22:48:45 UTC (rev 2483) +++ branches/Cog/nscogbuild/unixbuild/HowToBuild 2011-08-01 21:36:19 UTC (rev 2484) @@ -5,6 +5,8 @@ - How to build the Newspeak Cog VM on Unix - Testing an external plugin has completely linked - Compiling in 32-bit mode on a 64-bit linux + - Testing an external plugin has completely linked + - N.B. Please read: Optimization level and gcc version How to build the Newspeak VM on Unix ------------------------------- @@ -59,3 +61,13 @@ symbols, remake the plugin, capturing the link step and then supply -Wl,--warn-unresolved-symbols -Wl,--no-allow-shlib-undefined when manually repeating the link command + + +Optimization level and gcc version +---------------------------------- +There appear to be issues with 3.4.x gcc version on RedHat. In particular +compiling the Newspeak VM with either of + gcc version 3.4.6 20060404 (Red Hat 3.4.6-3) + gcc version 3.4.6 20060404 (Red Hat 3.4.6-10) +using -O2 results in a VM that segfaults early in startup. For these compilers +it is probably wise to use -O1, even though -O3 seems to work. Modified: branches/Cog/platforms/Cross/vm/sq.h =================================================================== --- branches/Cog/platforms/Cross/vm/sq.h 2011-07-31 22:48:45 UTC (rev 2483) +++ branches/Cog/platforms/Cross/vm/sq.h 2011-08-01 21:36:19 UTC (rev 2484) @@ -23,6 +23,18 @@ #define false 0 #define null 0 /* using "null" because nil is predefined in Think C */ +#if !defined(IMAGE_DIALECT_NAME) +# if NewspeakVM +# define IMAGE_DIALECT_NAME "Newspeak" +# define DEFAULT_IMAGE_NAME "newspeak.image" +# define IMAGE_ENV_NAME "NEWSPEAK_IMAGE" +# else +# define IMAGE_DIALECT_NAME "Squeak" +# define DEFAULT_IMAGE_NAME "squeak.image" +# define IMAGE_ENV_NAME "SQUEAK_IMAGE" +# endif +#endif + /* Pluggable primitives macros. */ /* Note: All pluggable primitives are defined as Property changes on: branches/Cog/platforms/Cross/vm/sqSCCSVersion.h ___________________________________________________________________ Modified: checkindate - Sun Jul 31 15:48:22 PDT 2011 + Mon Aug 1 14:33:26 PDT 2011 Modified: branches/Cog/platforms/Mac OS/vm/sqMacMain.c =================================================================== --- branches/Cog/platforms/Mac OS/vm/sqMacMain.c 2011-07-31 22:48:45 UTC (rev 2483) +++ branches/Cog/platforms/Mac OS/vm/sqMacMain.c 2011-08-01 21:36:19 UTC (rev 2484) @@ -130,7 +130,7 @@ gSqueakUIFlushSecondaryCleanupDelayMilliseconds=20, gSqueakUIFlushSecondaryCheckForPossibleNeedEveryNMilliseconds=16, gSqueakDebug=0; -char gSqueakImageName[PATH_MAX] = "Squeak.image"; +char gSqueakImageName[PATH_MAX] = DEFAULT_IMAGE_NAME; char gSqueakUntrustedDirectoryName[PATH_MAX] = "/foobar/tooBar/forSqueak/bogus/"; char gSqueakTrustedDirectoryName[PATH_MAX] = "/foobar/tooBar/forSqueak/bogus/"; CFStringRef gSqueakImageNameStringRef; Modified: branches/Cog/platforms/Mac OS/vm/sqMacUnixCommandLineInterface.c =================================================================== --- branches/Cog/platforms/Mac OS/vm/sqMacUnixCommandLineInterface.c 2011-07-31 22:48:45 UTC (rev 2483) +++ branches/Cog/platforms/Mac OS/vm/sqMacUnixCommandLineInterface.c 2011-08-01 21:36:19 UTC (rev 2484) @@ -250,7 +250,7 @@ printf(" %s [<option>...] -- [<argument>...]\n", argVec[0]); printUsage(); printf("\nNotes:\n"); - printf(" <imageName> defaults to `Squeak.image'.\n"); + printf(" <imageName> defaults to `" DEFAULT_IMAGE_NAME "'.\n"); printUsageNotes(); exit(1); } @@ -282,8 +282,8 @@ static void printUsageNotes(void) { printf(" If `-memory' is not specified then the heap will grow dynamically.\n"); - printf(" <argument>s are ignored, but are processed by the Squeak image.\n"); - printf(" The first <argument> normally names a Squeak `script' to execute.\n"); + printf(" <argument>s are ignored, but are processed by the " IMAGE_DIALECT_NAME " image.\n"); + printf(" The first <argument> normally names a " IMAGE_DIALECT_NAME " `script' to execute.\n"); printf(" Precede <arguments> by `--' to use default image.\n"); } @@ -311,10 +311,12 @@ static void parseEnvironment(void) { - char *ev= 0; + char *ev; - if ((ev= getenv("SQUEAK_IMAGE"))) - resolveWhatTheImageNameIs(ev); - if ((ev= getenv("SQUEAK_MEMORY"))) gMaxHeapSize= strtobkm(ev); - if ((ev= getenv("SQUEAK_PATHENC"))) setEncodingType(ev); + if ((ev= getenv(IMAGE_ENV_NAME))) + resolveWhatTheImageNameIs(ev); + if ((ev= getenv("SQUEAK_MEMORY"))) + gMaxHeapSize= strtobkm(ev); + if ((ev= getenv("SQUEAK_PATHENC"))) + setEncodingType(ev); } Modified: branches/Cog/platforms/unix/vm/sqUnixExternalPrims.c =================================================================== --- branches/Cog/platforms/unix/vm/sqUnixExternalPrims.c 2011-07-31 22:48:45 UTC (rev 2483) +++ branches/Cog/platforms/unix/vm/sqUnixExternalPrims.c 2011-08-01 21:36:19 UTC (rev 2484) @@ -35,6 +35,7 @@ #endif #include "sq.h" /* sqUnixConfig.h */ +#include "sqAssert.h" #if (DEBUG) # define DPRINTF(ARGS) fprintf ARGS @@ -216,7 +217,8 @@ * moduleName and suffix. Answer the new module entry, or 0 if the shared * library could not be loaded. */ -static void *tryLoading(char *dirName, char *moduleName) +static void * +tryLoading(char *dirName, char *moduleName) { static char *prefixes[]= { "", "lib", 0 }; static char *suffixes[]= { "", ".so", ".dylib", 0 }; @@ -224,33 +226,32 @@ char **prefix= 0, **suffix= 0; for (prefix= prefixes; *prefix; ++prefix) - for (suffix= suffixes; *suffix; ++suffix) - { - char libName[NAME_MAX + 32]; /* headroom for prefix/suffix */ - struct stat buf; - int err; - sprintf(libName, "%s%s%s%s", dirName, *prefix, moduleName, *suffix); - if ((!(err= stat(libName, &buf))) && S_ISDIR(buf.st_mode)) - DPRINTF((stderr, "ignoring directory: %s\n", libName)); - else - { - DPRINTF((stderr, "tryLoading %s\n", libName)); - handle= dlopen(libName, RTLD_NOW | RTLD_GLOBAL); - if (handle == 0) - { - if ((!err) && !(sqIgnorePluginErrors)) - fprintf(stderr, "ioLoadModule(%s):\n %s\n", libName, dlerror()); - } - else - { -# if DEBUG - printf("%s: loaded plugin `%s'\n", exeName, libName); -# endif - return handle; - } - } - } - return 0; + for (suffix= suffixes; *suffix; ++suffix) { + char libName[NAME_MAX + 32]; /* headroom for prefix/suffix */ + struct stat buf; + int n; + n = sprintf(libName,"%s%s%s%s",dirName,*prefix,moduleName,*suffix); + assert(n >= 0 && n < NAME_MAX + 32); + if (!stat(libName, &buf)) { + if (S_ISDIR(buf.st_mode)) + DPRINTF((stderr, "ignoring directory: %s\n", libName)); + else { + DPRINTF((stderr, "tryLoading %s\n", libName)); + handle= dlopen(libName, RTLD_NOW | RTLD_GLOBAL); + if (handle == 0) { + if (!sqIgnorePluginErrors) + fprintf(stderr, "ioLoadModule(%s):\n %s\n", libName, dlerror()); + } + else { +# if DEBUG + printf("%s: loaded plugin `%s'\n", exeName, libName); +# endif + return handle; + } + } + } + } + return 0; } @@ -283,49 +284,44 @@ /* Find and load the named module. Answer 0 if not found (do NOT fail * the primitive!). */ -void *ioLoadModule(char *pluginName) +void * +ioLoadModule(char *pluginName) { - void *handle= 0; + void *handle= 0; - if ((pluginName == 0) || (pluginName[0] == '\0')) - { - handle= dlopen(0, RTLD_NOW | RTLD_GLOBAL); - if (handle == 0) - fprintf(stderr, "ioLoadModule(<intrinsic>): %s\n", dlerror()); - else - { - DPRINTF((stderr, "loaded: <intrinsic>\n")); - return handle; + if (!pluginName || !*pluginName) { + if (!(handle= dlopen(0, RTLD_NOW | RTLD_GLOBAL))) { + fprintf(stderr, "ioLoadModule(<intrinsic>): %s\n", dlerror()); + return 0; + } + DPRINTF((stderr, "loaded: <intrinsic>\n")); + return handle; } - } - if (squeakPlugins) - { - char path[NAME_MAX]; - char c, *in= squeakPlugins, *out= path; - while ((c= *in++)) - { - if (c == '%' && ((*in == 'n') || (*in == 'N'))) - { - ++in; - strcpy(out, pluginName); - out+= strlen(pluginName); - } - else - *out++= c; - } - *out= '\0'; - DPRINTF((stderr, "ioLoadModule plugins = %s\n path = %s\n", - squeakPlugins, path)); - if ((handle= tryLoading("", path))) - return handle; - if (!(out > path && *(out - 1) == '/')) { - *out++= '/'; - *out= '\0'; + if (squeakPlugins) { + char path[NAME_MAX]; + char c, *in= squeakPlugins, *out= path; + while ((c= *in++)) { + if (c == '%' && ((*in == 'n') || (*in == 'N'))) { + ++in; + strcpy(out, pluginName); + out+= strlen(pluginName); + } + else + *out++= c; + } + *out= '\0'; + DPRINTF((stderr, "ioLoadModule plugins = %s\n path = %s\n", + squeakPlugins, path)); + if ((handle= tryLoading("", path))) + return handle; + if (!(out > path && *(out - 1) == '/')) { + *out++= '/'; + *out= '\0'; + } + if ((handle= tryLoading(path, pluginName))) + return handle; } - if ((handle= tryLoading(path, pluginName))) - return handle; - } if (( handle= tryLoading( "./", pluginName)) || (handle= tryLoadingPath("SQUEAK_PLUGIN_PATH", pluginName)) @@ -390,7 +386,6 @@ #endif return 0; } - #endif /* USE_SIMPLIFIED_PLUGIN_LOGIC */ /* Find a function in a loaded module. Answer 0 if not found (do NOT Modified: branches/Cog/platforms/unix/vm/sqUnixMain.c =================================================================== --- branches/Cog/platforms/unix/vm/sqUnixMain.c 2011-07-31 22:48:45 UTC (rev 2483) +++ branches/Cog/platforms/unix/vm/sqUnixMain.c 2011-08-01 21:36:19 UTC (rev 2484) @@ -1182,15 +1182,6 @@ if (documentName) strcpy(shortImageName, documentName); -#if NewspeakVM -# define IMAGE_ENV_NAME "NEWSPEAK_IMAGE" -# define IMAGE_DIALECT_NAME "Newspeak" -# define DEFAULT_IMAGE_NAME "newspeak.image" -#else -# define IMAGE_ENV_NAME "SQUEAK_IMAGE" -# define IMAGE_DIALECT_NAME "Squeak" -# define DEFAULT_IMAGE_NAME "squeak.image" -#endif else if ((ev= getenv(IMAGE_ENV_NAME))) strcpy(shortImageName, ev); else Modified: branches/Cog/unixbuild/HowToBuild =================================================================== --- branches/Cog/unixbuild/HowToBuild 2011-07-31 22:48:45 UTC (rev 2483) +++ branches/Cog/unixbuild/HowToBuild 2011-08-01 21:36:19 UTC (rev 2484) @@ -6,6 +6,8 @@ - How to build the Cog Croquet VM on Unix - Testing an external plugin has completely linked - Compiling in 32-bit mode on a 64-bit linux + - Testing an external plugin has completely linked + - N.B. Please read: Optimization level and gcc version Checking out sources @@ -99,3 +101,13 @@ symbols, remake the plugin, capturing the link step and then supply -Wl,--warn-unresolved-symbols -Wl,--no-allow-shlib-undefined when manually repeating the link command + + +Optimization level and gcc version +---------------------------------- +There appear to be issues with 3.4.x gcc version on RedHat. In particular +compiling the Newspeak VM with either of + gcc version 3.4.6 20060404 (Red Hat 3.4.6-3) + gcc version 3.4.6 20060404 (Red Hat 3.4.6-10) +using -O2 results in a VM that segfaults early in startup. For these compilers +it is probably wise to use -O1, even though -O3 seems to work.

1 0

SqueakVMPathAnswersResources preference
by Eliot Miranda 01 Aug '11

01 Aug '11

Hi All, I just added code to the Cog branch for the Mac VM so that whether primVmPath/primitiveVMPath answers the directory containing the VM.app or answers the Resources directory within the .app is controlled by the SqueakVMPathAnswersResources preference in the Info.plist file. primVmPath is used to determine where to derive the sources file from. By default primVmPath answers the directory containing the VM.app and therefore the sources file is at the same level, e.g. /Applications/Squeak.app & /Applications/SqueakV41.sources. But some installations (e.g. Newspeak and the EToys all-in-one) may find it more convenient to put the sources in VM.app/Contents/Resources. This means one file to install, not 2, etc. I think it useful that this behavior be in the standard VM. Esteban, if you agree, perhaps you could integrate into the trunk platforms/Mac OS? -- best, Eliot

3 7

A trick to speedup #become: (but not becomeForward:)
by Igor Stasenko 01 Aug '11

01 Aug '11

An interesting observation: if two oops which we need to swap: oop1 and oop2 taking exactly same space in memory, then all we need to do is to swap their memory contents, instead of scanning heap and updating pointers! This makes a become operation extremely cheap for such pairs! As for objects of different sizes, we could still use slow algorithm. But knowing that if two objects with same size will be swapped much faster, a developers could adapt their algorithms to exploit this feature (use fixed-size objects instead of variable-sized ones etc). Magma using proxies, which then #becomeForward: to real objects when reified. This is very costly operation and main reason why it so slow on loading objects from server. To speed thing up, a special trick can be used: - each time server sends an object ID (instead of real object) it could also send its size in bytes (or even more clever - you can encode object size in its ID ;) ). Then a client will create a proxy for given object and will try to match the size, if its possible. (nobody said that proxy can't be variable-sized, right?) Then once client requesting to reify given proxy with real object from database, VM will just overwrite a proxy's memory contents with real object. -- Best regards, Igor Stasenko AKA sig.

4 11

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

Vm-dev August 2011