Vm-dev April 2011

vm-dev@lists.squeakfoundation.org

36 participants
176 discussions

A couple of things about CMake
by Mariano Martinez Peck 07 Apr '11

07 Apr '11

Hi Igor. I was thinking a couple of things: 1) So...once I hace download anything from GIT, what I always need to do is: a) take a pharo core 1.3 b) open the LoadVMMaker.st so that I can see which versions of metacello and cmake I need. c) evaluate that over the pharo image d) move my image to /build so that it finds the /src and /platform by default So...I wonder what are the reason of not having such PharoCore image already commited in GIT under /image. You can even save it with a workspace opened with the code of LoadVMMaker.st is it just the size of it? If I want to do it for myself, what is the best strategy? clone to my own fork, add the image there and then just pull the changes? 2) #prepareVMMaker does this: prepareVMMaker | maker allPlugins | maker := VMMaker forPlatform: 'Cross'. maker sourceDirectoryName: (self srcDir assureExistence fullName). maker platformRootDirectoryName: (self topDir / 'platforms' ) fullName. And srcDir "Answer the value of srcDir" ^ srcDir ifNil: [ srcDir := self topDir / 'src' ] This is cool because I can script the cong to use another src. But you don't do the same for 'platform'. So I cannot do: CogDebugMacOSConfig new srcDir: 'myFolder'; platformsDir: 'myPlatofrm'; generateSources; generate. Since the platform directory will be overwritten. Cheers Mariano

2 4

CMake for Windows
by Mariano Martinez Peck 06 Apr '11

06 Apr '11

Hi Igor. I saw CogMsWindowsConfig but I have a couple of questions: 1) is it working ? 2) how do you use cmake in a windows box ? cygwin ? 3) how are you now compiling the windows vm (in case you are) ? cheers Mariano

2 7

VM Maker: CMakeVMMaker-MarianoMartinezPeck.69.mcz
by squeak-dev-noreply＠lists.squeakfoundation.org 06 Apr '11

06 Apr '11

Mariano Martínez Peck uploaded a new version of CMakeVMMaker to project VM Maker: http://www.squeaksource.com/VMMaker/CMakeVMMaker-MarianoMartinezPeck.69.mcz ==================== Summary ==================== Name: CMakeVMMaker-MarianoMartinezPeck.69 Author: MarianoMartinezPeck Time: 6 April 2011, 7:54:31 pm UUID: 3b75bf4c-cf11-448c-9aad-e858a324f9d6 Ancestors: CMakeVMMaker-MarianoMartinezPeck.68 Change the outputDir so that now all confs in all OS are uniform: all of them let the binary in the parent directoty under a directory /results. So.../results will be just next to /platforms, /src, etc. Putting /results inside /build was VERY difficult to find.

1 0

Event VM: current status?
by Dimitry Golubovsky 06 Apr '11

06 Apr '11

Hi, Andreas Raab wrote: ... > You participated in the discussion :-) > http://lists.squeakfoundation.org/pipermail/vm-dev/2009-November/003385.html > http://lists.squeakfoundation.org/pipermail/vm-dev/2009-November/003437.html Thanks Andreas for these links. I am reading through. At least I can learn some history ;) -- Dimitry Golubovsky Anywhere on the Web

1 0

Event VM: current status?
by Dimitry Golubovsky 06 Apr '11

06 Apr '11

Hi, It is my understanding that the version of Squeak VM ("Event VM") that was used in the original Android port (and the same I am using in my fork) differs from the "mainline" VM that interpret() is not an infinite loop but rather a handler of some events queue (a ring buffer as I conclude from the sources) which returns once the queue is empty. Please correct me if I am wrong. There is a zip-file (witn a Windows executable inside) for this VM dated 13-Nov-2009. Has there been any progress since? Can the event VM be recreated these days using VMMaker and related tools, or was it an ad-hoc modification of a VMMaker code output? Such VM looks very convenient for embedding (just like the Android port is an example of embedding into JVM), so Squeak might be used as a very powerful GUI frontend library with backend being anything else. Thanks for any ideas. -- Dimitry Golubovsky Anywhere on the Web

3 3

next steps for cocoa cog
by Esteban Lorenzano 06 Apr '11

06 Apr '11

Hi, as you can see... I'm finally building plugins for cocoa cog vms... but there are still some new actions I want to take. So, I'm informing my planned next steps: 1) add FT2Plugin and OSProcessPlugin to cocoa builds 2) collect (I mean "collect", not create... at least not in all cases) a suite of tests to ensure plugins are working at least "more or less", to add to quality production check 3 and 4) start migrating the carbon plugins to cocoa plugins (there are not so many, but some of them need really work) and ensure all the plugins work (using the tests for that) In parallel: I will build a CMakeConfig for StackVM in the iPhone Well... as you can see... I think my "VM Year" is really full of work... I will be working on this as much as my daily work allows me... so... this are my REQUESTS TO THE COMMUNITY: 1) If you think I'm missing something important, now is the time to say it :) 2) I want TESTS... as many tests as I can collect, related to plugins... or at least ideas of how to test things... 3) patience... I will be sending a lot of noise mails like this one ;) Cheers, Esteban

2 1

latest cog build plugins
by Esteban Lorenzano 06 Apr '11

06 Apr '11

Hi, Latest cocoa cog builds (from ci.pharo-project.org) are being built with this external plugins: ThreadedIA32FFIPlugin or ReentrantIA32FFIPlugin (if CogMT or Cog) UUIDPlugin FloatMathPlugin AsynchFilePlugin SerialPlugin Mpeg3Plugin CroquetPlugin JoystickTabletPlugin MIDIPlugin B3DAcceleratorPlugin LocalePlugin In addition to usual internal plugins: ADPCMCodecPlugin BMPReadWriterPlugin BalloonEnginePlugin BitBltSimulation DSAPlugin DeflatePlugin DropPlugin FFTPlugin FilePlugin FloatArrayPlugin GeniePlugin HostWindowPlugin JPEGReadWriter2Plugin JPEGReaderPlugin KlattSynthesizerPlugin LargeIntegersPlugin Matrix2x3Plugin MiscPrimitivePlugin RePlugin SecurityPlugin SocketPlugin SoundCodecPlugin SoundGenerationPlugin SoundPlugin StarSqueakPlugin SurfacePlugin I think they are now a good set of plugins... Maybe cobalt guys can now test it and provide some feedback (as I said, I'm compiling the plugins, but no idea if they are working... at least until I construct a test suite for them). Enjoy :) Esteban

2 1

[Pharo-project][Vm-dev] OSS sound plugin.
by Lemuus 06 Apr '11

06 Apr '11

Hi, is it possible to include the vm-sound-OSS sound plugin in the cog build on Hudson?. This is the only plugin that seems to work on linux. Thanks. Daniel Galdames G.

3 3

FreeBSD r2378 and status
by David Graham 06 Apr '11

06 Apr '11

I just uploaded a new binary for Cog FreeBSD. http://www.unthinkable.org/dl/cogvm-r2378-freebsd-i386.tgz It's built from the current squeakvm source configured with: ../../platforms/unix/config/configure CFLAGS="-g -O2 -msse2 -DNDEBUG -DNO_VM_PROFILE=1 -DCOGMTVM=0 -DNOEXECINFO -DDEBUGVM=0" LIBS=-pthread --without-npsqueak I don't currently have a way to test OpenGL, (I'm doing dev on a remote VM) but the plugin is there. Also, I don't have sound or BSD native UUID enabled (but am hoping to get UUID working soon). I haven't had a chance to work on the CMakeVMMaker code. Is there a way to disable the download link on Hudson until we can get a functional build? (The source compiles 100%, but doesn't link properly) If someone want's to take it up, there is probably some coordinated refactoring to do in the CUnix classes since half of my overrides were to remove linux specific compiler flags.

2 1

Problem with alien callbacks
by Javier Pimás 05 Apr '11

05 Apr '11

Hi, we are having a problem with callbacks in alien and we would like to see if we are doing something wrong or if it is a bug in the implementation (for the standard old vm). We are receiving the callback just in the middle of a primitiveExternalCall (actually to a function that will fail because the plugin is not present, but i don't think that's important). We pinned it to occur always in the same line, which is longAtput((lit + (BASE_HEADER_SIZE)) + (2 << (SHIFT_FOR_WORD)), ConstZero); of primitiveExternalCall. When the callback occurs, the thunkEntry is called, which if we understand correctly, saves the active context and runs the interpreter by calling sendInvokeCallbackStackRegistersJmpbuf. The problem is that things like messageSelector and lkupClass, which are global variables are not saved while saving the context, and when the callback returns, the last line of primitiveExternalCall, rewriteMethodCacheSelclassprimIndex(messageSelector, lkupClass, 0); puts a 0 in the wrong place. Also, probably as las message sent was primReturnFromContext:through: (because we just returned from the context), we get a primitiveFailed, but not for the original called function but for primReturnFromContext:through:. What do you think? are we missing something? Regards, Javier. -- Javier Pimás Ciudad de Buenos Aires

3 6

Jump to page:

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

Vm-dev April 2011