Vm-dev October 2008

vm-dev@lists.squeakfoundation.org

11 participants
15 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

Joys of Squeak-dev image, VM Compiling on MacIntel and Spoon remote OtherBrowsing
by Ken G. Brown 16 Nov '10

16 Nov '10

MacPro OS X 10.4.9 Xcode 2.4.1 Squeak-dev-114-2, VMMaker 2.8b6.1, Squeak 3.8.15beta6U.app, and version 1657 of svn MacOS VM sources, and Spoon v2a12. Downloaded http://www.squeakvm.org/cgi-bin/viewcvs.cgi/ trunk version 1657 for MacOS Ran squeak-dev-beta-114-2.image on Squeak 3.8.15beta6U.app Loaded VMMaker 3.8b6.1 from dev image Universes. Generated vm sources by following instructions in <http://www.squeakvm.org/cgi-bin/viewcvs.cgi/trunk/platforms/Mac%20OS/vm/Doc…> Fought with Xcode wanting to use MacOSX10.3.9.sdk for ppc which was giving an error which was known and documented in the Readme. Finally figured out what Xcode wanted so built universal app just using 10.4.9 sdk. Couple issues: Exupery references remain in sqnamedPrims.h and sqPlatformSpecific.h from svn version 1657, I just commented them out. Notes said: Info-NPSqueak__Upgraded_.plist was to be found in the Developer directory. I could not find it so scammed a info.plist out of John's Squeak 3.8.15beta6U.app vm. Not sure what impact this has. And much, much diddling about figuring out various things... Finally got a build. It's a lot smaller (3.9MB) than John McIntosh's Squeak 3.8.15beta6U.app (8.1MB), and it looks like it is a PowerPC one but at least it runs. I'll look into the Xcode build process some more. Then: Replaced interp.c with Craig's interp.c for Spoon, rebuilt and got a VM with Spoon client stuff. Downloaded Spoon release 2a12. Moved a copy of Flow plugin from within Craig's naiad.app package into a plugins subdirectory where the new spoon client compatible MacIntel universal VM was run from. Followed Craig's Spoon instructions for loading niaid.st and niaid.pr into the 3.10 image (Squeak-dev-114-2) running on the new VM. Did the run spoon.app thing from Spoon.dmg (version 2a12), needs to be done twice if Safari is not running. Got Spoon processor running. Followed instructions again in the Spoon client and lo! Got OtherBrowser browsing the Spoon memory running on processor. Yay Craig! Who do i get to sign my certificate? :) Ken

4 7

How to use Croquet VMMaker
by Matthew Fulmer 29 Apr '09

29 Apr '09

Hi. In anticipation of working on HydraVM, I am trying to test as close a configuration as possible to HydraVM without being hydra. That is, load the nearest ancestor VMMaker and C sources to hydra, and try to make a working VM. This should help ensure that there are no "normal" bugs once I switch to Hydra, only hydra bugs. So, I did that: Here is my setup: - Fedora core 7 on x86 processor - Release 3.9 image/VM from squeak.org - Load FFI from SqueakMap - Load VMMaker-ar.70 from http://jabberwocky.croquetproject.org:8889/Homebase/ - Checkout http://squeakvm.org/svn/squeak/trunk revision 1796 (The nearest ancestor of HydraVM) I pointed VMMaker at the platforms directory in the checkout, and told it to build source in platforms/unix/src. I selected all plugins to be external, but moved FilePlugin and SocketPlugin to be internal. Then I did mkdir unixbuild cd unixbuild ../platforms/unix/config/configure make make fails linking gnu-interp.o: /bin/sh /home/matthewf/squeak/vm/hydravm-trunk-base/unixbuild/libtool --mode=link gcc -g -O2 -fomit-frame-pointer -DLSB_FIRST=1 -export-dynamic -R/usr/local/lib -o squeak vm/vm.a FilePlugin/FilePlugin.a SocketPlugin/SocketPlugin.a disabledPlugins.o version.o -lutil -ldl -lm -lnsl vm/vm.a gcc -g -O2 -fomit-frame-pointer -DLSB_FIRST=1 -o squeak disabledPlugins.o version.o -Wl,--export-dynamic vm/vm.a FilePlugin/FilePlugin.a SocketPlugin/SocketPlugin.a -lutil -ldl -lm -lnsl vm/vm.a -Wl,--rpath -Wl,/usr/local/lib vm/vm.a(gnu-interp.o): In function `primitiveGetWindowLabel': /home/matthewf/squeak/vm/hydravm-trunk-base/unixbuild/vm/gnu-interp.c:15141: undefined reference to `ioGetWindowLabel' vm/vm.a(gnu-interp.o): In function `primitiveSetWindowLabel': /home/matthewf/squeak/vm/hydravm-trunk-base/unixbuild/vm/gnu-interp.c:18770: undefined reference to `ioSetWindowLabelOfSize' collect2: ld returned 1 exit status Building this VM worked when using the latest VMMaker from SqueakMap (VMMaker-tpr.58), everything else the same. Is there something else I need to do in order to build using the Croquet VMMaker? I want to use it specifically, since the Hydra VMMaker is derived from it. The dev notes on Hydra are at http://squeakvm.org/~sig/hydravm/devnotes.html -- Matthew Fulmer -- http://mtfulmer.wordpress.com/ Help improve Squeak Documentation: http://wiki.squeak.org/squeak/808

5 9

UUID new crashes 64bit vm
by Philippe Marschall 29 Jan '09

29 Jan '09

Hi I built a 64bit VM (for 32bit images) from SVN trunk and VMMaker-dtl.102. UUID new causes the VM to segmentation fault. The reason I did this is because I need to have a Unix VM with the cUrl plugin and my system is 64 bit. I need to have the cUrl plugin because I need an https client. If I'm doing it wrong what would be the right way of doing it? Or is there even a recent 64bit Unix VM available that was a cUrl plugin? Cheers Philippe

5 8

SmartSyntaxPluginTMethod>>nullReturnExpr, possible bug?
by Esteban Lorenzano 01 Dec '08

01 Dec '08

Hi, I had to change null output for squeak generated plugins (it was colliding with other libraries needed) and I found this: SmartSyntaxPluginTMethod>>nullReturnExpr ^ TReturnNode new setExpression: (TVariableNode new setName: 'null') this generates correct return lines, but just because "null" is defined previously. If I change null to, e.g. sqNil, the previous method still generates "return null" lines. So I changed the method as follows: SmartSyntaxPluginTMethod>>nullReturnExpr ^ TReturnNode new setExpression: (TVariableNode new setName: 'nil') just replaced "null" for "nil", and everything seems to be working fine. Is this hack correct? Cheers, Esteban

2 1

mmap, changes to co-ordinate mapping to particular address to avoid oops address remapping at startup time.
by John M McIntosh 26 Nov '08

26 Nov '08

In the past year or so there has been discussion of using mmap to set a start address for the offset of object memory so that memory address swizzling does not occur. Also various operating systems vendors say using mmap to page in large files is faster than using fread, this in fact is true and noticeable on the iPhone. This requires a change to VMMaker so that sqAllocateMemory can pass the file pointer and the header size, since in the sqAllocateMemory we need to figure out the file (image size) and also understand the headersize of the image file to return a proper address to work with. #define sqAllocateMemory(x,y,f,headersize) (a) map in the image file as MAP_PRIVATE (copy on write) at a set location, noting that on os-x mmap (file offset) does not allow for non- multiples of page size, so we return the offset into the file as the address. (b) map in the free space at the page boundary rounded up based on the file size, plus the set location. This should give two mmap memory regions side by side. Obvious issues, if any mmap fails we exit. We cannot be sure what will happen for memory mapping in the future, should we attempt recover at some other start location? Source code below for comment. ?bug memory refers to (startOfAnonymousMemory+headersize) yet calculation in sqGrowMemoryBy uses memory and gMaxHeapSize. lazy adding of extra page size to freeSpaceRoundedUpToPageSize is preventing disaster, proper fix is exercise for the reader... extern usqInt gMaxHeapSize; static usqInt gHeapSize; static void *startOfmmapForANONMemory,*startOfmmapForImageFile; static size_t fileRoundedUpToPageSize,freeSpaceRoundedUpToPageSize; /* compute the desired memory allocation */ extern unsigned char *memory; usqInt sqGetAvailableMemory() { return gMaxHeapSize; } usqInt sqAllocateMemoryMac(sqInt minHeapSize, sqInt *desiredHeapSize, FILE * f,usqInt headersize) { void *possibleLocation,*startOfAnonymousMemory; off_t fileSize; struct stat sb; size_t pageSize= getpagesize(); size_t pageMask= ~(pageSize - 1); #define valign(x) ((x) & pageMask) #pragma unused(minHeapSize,desiredHeapSize) possibleLocation = 500*1024*1024; gHeapSize = gMaxHeapSize; /* Lets see about mmap the image file into a chunk of memory at the 500MB boundary rounding up to the page size Then we on the next page anonymously allocate the required free space for young space*/ /* Thanks to David Pennell for suggesting this */ fstat(fileno((FILE *)f), &sb); fileSize = sb.st_size; fileRoundedUpToPageSize = valign(fileSize+pageSize-1); startOfAnonymousMemory = (void *) ((size_t) fileRoundedUpToPageSize + (size_t) possibleLocation); startOfmmapForImageFile = mmap(possibleLocation, fileRoundedUpToPageSize, PROT_READ|PROT_WRITE, MAP_FILE| MAP_PRIVATE,fileno((FILE *)f), 0); if (startOfmmapForImageFile != possibleLocation) { fprintf(stderr, "errno %d\n", errno); perror("startOfmmapForImageFile failed"); exit(42); } freeSpaceRoundedUpToPageSize = valign(gMaxHeapSize)- fileRoundedUpToPageSize+pageSize; startOfmmapForANONMemory = mmap(startOfAnonymousMemory, freeSpaceRoundedUpToPageSize, PROT_READ|PROT_WRITE, MAP_ANON| MAP_SHARED,0,0); if (startOfmmapForANONMemory != startOfAnonymousMemory) { fprintf(stderr, "errno %d\n", errno); perror("startOfmmapForANONMemory failed"); exit(42); } return (usqInt) startOfmmapForImageFile+headersize; } sqInt sqGrowMemoryBy(sqInt memoryLimit, sqInt delta) { if ((usqInt) memoryLimit + (usqInt) delta - (usqInt) memory > gMaxHeapSize) return memoryLimit; gHeapSize += delta; return memoryLimit + delta; } sqInt sqShrinkMemoryBy(sqInt memoryLimit, sqInt delta) { returnsqGrowMemoryBy(memoryLimit,0-delta); } sqInt sqMemoryExtraBytesLeft(sqInt flag) { return gMaxHeapSize - gHeapSize; } void sqMacMemoryFree() { if (memory == NULL) return; munmap(startOfmmapForImageFile,fileRoundedUpToPageSize); munmap(startOfmmapForANONMemory,freeSpaceRoundedUpToPageSize); memory = NULL; }

2 5

Interpreter>>signed32BitValueOf: and signed64BitValueOf: broken
by Andreas Raab 26 Nov '08

26 Nov '08

Hi - I just noticed hat Interpreter>>signed32BitValueOf: and signed64BitValueOf: are broken for edge cases. The following example will illustrate the problem: array := IntegerArray new: 1. array at: 1 put: 16rFFFFFFFF. "should fail but doesn't" array at: 1. "answers -1 incorrectly" array := IntegerArray new: 1. array at: 1 put: -16rFFFFFFFF. "should fail but doesn't" array at: 1. "answers 1 incorrectly" The problem is that both signed32BitValueOf: as well as signed64BitValueOf: do not test whether the high bit of the magnitude is set (which it mustn't to fit into a signed integer). The fix is trivial in both cases - basically all that's needed at the end of both functions is this: "Filter out values out of range for the signed interpretation such as 16rFFFFFFFF (positive w/ bit 32 set) and -16rFFFFFFFF (negative w/ bit 32 set). Since the sign is implicit in the class we require that the high bit of the magnitude is not set which is a simple test here" value < 0 ifTrue:[^self primitiveFail]. negative ifTrue:[^0 - value] ifFalse:[^value] Cheers, - Andreas

4 12

Ubuntu 8 etoys/firefox crash
by Michael Rueger 29 Oct '08

29 Oct '08

Hi all, there was a report about a problem with etoys in Firefox on Ubuntu 8. I was able to easily reproduce it: - open firefox - make sure to have tabs open - load any etoys project from the squeakland web site - close the etoys project tab -> XIO: fatal IO error 11 (Resource temporarily unavailable) on X server ":0.0" after 204 requests (204 known processed) with 0 events remaining. Segmentation fault This doesn't seem to happen on Ubuntu 7. Any ideas? Michael P.S. Can' try on Ubuntu 6 as there is a problem with missing GLIBC_2.4 and I probably need to build a VM with -fno-stack-protector. see http://gbgames.com/blog/2008/04/linux-game-development-glibc_24-errors-solv…

1 0

Re: Bug with valueWithReceiver:arguments:
by Andreas Raab 26 Oct '08

26 Oct '08

Hi Frederic - Congrats! You found the first real VM bug in five years or so ;-) I'm not completely sure what the problem is but there is obviously something wrong here. It's easiest to recreate the problem by copying the #/ method and simply do something like: SmallInteger>>foo: aNumber "(SmallInteger>>#foo:) valueWithReceiver: 11 arguments: {2}" <primitive: 10> self halt. aNumber isZero ifTrue: [^(ZeroDivide dividend: self) signal]. (aNumber isMemberOf: SmallInteger) ifTrue: [^(Fraction numerator: self denominator: aNumber) reduced] ifFalse: [^super / aNumber] When you run this, you'll see that the debugger shows *two* frames with SmallInteger>>foo: on them. Inspecting the "parent" frame will eventually crash the system since it has a completely bogus stack pointer (-1). It seems that in a failing primitive method we end up with a bogus activation record in primitiveExecMethodWithArgs (removing the primitive from the above makes it work fine). I'm somewhat at a loss here as to what the problem might be given that the implementations of primPerformWithArgs and primExecWithArgs are so similar but interestingly, the equivalent: 11 perform: #foo: withArguments:{2}. works fine with or without the primitive. Somewhere there must be a difference ... Cheers, - Andreas Frederic Pluquet wrote: > Hello, > > I found a fondamental bug in Squeak and Pharo. The next code > > 11 / 2 > > gives the fraction (11/2). It's correct. But the next code > > (SmallInteger>>#/) valueWithReceiver: 11 arguments: {2} > > gives 1 ! > > The problem is the method valueWithReceiver:arguements: is used hugely > with method wrappers... > > After long time of debugging, I found a point to debug: this method > don't have the good behavior with compiled methods having a primitive > that fails and executes some code after (as in SmallInteger>>#/ method > when the division don't give a whole integer). In fact, when I send this > message, the vm executes normally the compiled method but, in place of > returns simply the good result, seems to rerun the the compiled method > with other arguments (completly wrong) and returns so a wrong result. > > For example, > (SmallInteger>>#/) valueWithReceiver: 11 arguments: {2} > > has the following execution trace : > > 2 isZero > | 2 = 0 > | returns: false > returns: false > 2 isMemberOf: SmallInteger > | 2 class > | returns: SmallInteger > | SmallInteger == SmallInteger > | returns: true > returns: true > Fraction numerator: 101 denominator: 2 > | Fraction new > | | Fraction basicNew > | | returns: a Fraction instance > | | (a Fraction instance) initialize > | | returns: a Fraction instance > | returns: a Fraction instance > | a Fraction instance setNumerator: 101 denominator: 2 > | | 2 = 0 > | | returns: false > | | 101 asInteger > | | returns: 101 > | | 2 asInteger > | | returns: 2 > | | 2 abs > | | | 2 < 0 > | | | returns: false > | | returns: 2 > | | 2 < 0 > | | returns: false > | returns: (101/2) > returns: (101/2) > (101/2) reduced > | 101 = 0 > | returns: false > | 101 gcd: 2 > | | 101 = 0 > | | returns: false > | | 2 \\ 101 > | | returns: 2 > | | 2 = 0 > | | returns: false > | | 101 \\ 2 > | | returns: 1 > | | 1 = 0 > | | returns: false > | | 2 \\ 1 > | | returns: 0 > | | 0 = 0 > | | returns: true > | | 1 abs > | | | 1 < 0 > | | | returns: false > | | returns: 1 > | returns: 1 > | 101 // 1 > | returns: 101 > | 2 // 1 > | returns: 2 > | 2 = 1 > | returns: false > | Fraction numerator: 101 denominator: 2 > | | Fraction new > | | | Fraction basicNew > | | | returns: a Fraction instance > | | | (a Fraction instance) initialize > | | | returns: a Fraction instance > | | returns: a Fraction instance > | | (a Fraction instance) setNumerator: 101 denominator: 2 > | | | 2 = 0 > | | | returns: false > | | | 101 asInteger > | | | returns: 101 > | | | 2 asInteger > | | | returns: 2 > | | | 2 abs > | | | | 2 < 0 > | | | | returns: false > | | | returns: 2 > | | | 2 < 0 > | | | returns: false > | | returns: (101/2) > | returns: (101/2) > returns: (101/2) > 2 isZero > | 2 = 0 > | returns: false > returns: false > false isMemberOf: SmallInteger > | false class > | returns: False > | False == SmallInteger > | returns: false > returns: false > > > Please help me to fix this bug. I really need it works fine ! > > Fréd > -- > Frédéric Pluquet > Université Libre de Bruxelles (ULB) > Assistant > http://www.ulb.ac.be/di/fpluquet > > > ------------------------------------------------------------------------ > >

2 1

Where do I found the Qwaq VM
by Mathieu Suen 25 Oct '08

25 Oct '08

Hi, Will trying to load Eliot changes it fail on the VM I use. So were do we find the Qwaq VM? Thanks Mth

2 2

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

Vm-dev October 2008