Squeak-dev July 2006

squeak-dev@lists.squeakfoundation.org

132 participants
241 discussions

Re: Help with FFI - another try
by nicolas cellier 24 Jul '06

24 Jul '06

I retry sending because it failed... ---------- Message transmis ---------- Subject: Re: Help with FFI - another try Date: Vendredi 21 Juillet 2006 23:13 From: nicolas cellier <ncellier(a)ifrance.com> To: Squeak-dev developers list general-purpose Squeak <squeak-dev(a)lists.squeakfoundation.org> Le Vendredi 21 Juillet 2006 22:49, nicolas cellier a écrit : > Le Vendredi 21 Juillet 2006 13:15, Eugene Beschastnov a écrit : > > I digged a little into C-code and found that hbeStream is pointer to > > some LAME data (some global flags are stored there). I suppose that > > root of problem is that this data is destroyed or moved between > > apicalls. Is there any way to avoid this? > > One solution is to allocate this data on external heap. > (See ExternalAddress class>>allocate:) instead of using a ByteArray in > Smalltalk memory. > > Of course, you'll have to free data yourself as in C. > Maybe the trick explained at http://bugs.impara.de/view.php?id=3692 that > should be in latest image can help... > > Nicolas My last post was too short: i remember it is not that obvious to guess the right object which should be passed as FFI call argument. It is an ExternalData, a combination of an ExternalAddress and an ExternalType. And one more thing: don't expect ExternalAddress to survive across image snapshots. If you want to survive, you must transfer data into a ByteArray before saving the image, and transfer to a newly allocated ExternalAddress before next use... But maybe this is not a problem. in Smallapack-Collection, i have a class ExternalArray doing this kind of tirck: ExternalArray>>resetExternalData "Reset the ExternalData object used as FFI argument. This is cached in an inst-var instead of being recreated at each call. However, this must be reset whenever handle is changed" | cType | cType := self class type. data := cType isStructureType ifTrue: [ "Arrays of structure cannot carry type checking in FFI" ExternalData fromHandle: handle type: ExternalType void] ifFalse: [ "Atomic types" ExternalData fromHandle: handle type: cType] where handle is sometimes a ByteArray in Smalltalk memory, sometimes an ExternalAddress according to my needs. ExternalArray>>copyInCSpace "Answer a copy of self with data in C space" ^self shallowCopy postCopyInCSpace ExternalArray>>postCopyInCSpace "Copy the data into a new array" | newHandle bs | handle isNil ifFalse: [bs := self byteSize. newHandle := ExternalAddress gcallocate: bs. self memcpyDest: newHandle fromSource: handle nBytes: bs. self setHandle: newHandle] Maybe you have enough elements or you can browse more at squeak source http://www.squeaksource.com/@ZGaVjYGiHOshVjSM/LGmmIzpi Nicolas -------------------------------------------------------

1 0

Re: Help with FFI - another try
by nicolas cellier 24 Jul '06

24 Jul '06

I retry sending since it failed... ---------- Message transmis ---------- Subject: Re: Help with FFI - another try Date: Vendredi 21 Juillet 2006 22:49 From: nicolas cellier <ncellier(a)ifrance.com> To: The general-purpose Squeak developers list <squeak-dev(a)lists.squeakfoundation.org> Le Vendredi 21 Juillet 2006 13:15, Eugene Beschastnov a écrit : > I digged a little into C-code and found that hbeStream is pointer to > some LAME data (some global flags are stored there). I suppose that > root of problem is that this data is destroyed or moved between > apicalls. Is there any way to avoid this? One solution is to allocate this data on external heap. (See ExternalAddress class>>allocate:) instead of using a ByteArray in Smalltalk memory. Of course, you'll have to free data yourself as in C. Maybe the trick explained at http://bugs.impara.de/view.php?id=3692 that should be in latest image can help... Nicolas -------------------------------------------------------

1 0

Squeak Chat Friday July 28th 00:00 GMT
by Ken Causey 24 Jul '06

24 Jul '06

Just a reminder that the next Squeak Chat will take place between 00:00 (Midnight) and 04:00 (4 AM) GMT Friday July 28th, 2006. As usual we will meet in #squeak on irc.freenode.net. More information about the #squeak IRC channel can be found at http://people.squeakfoundation.org/article/7.html This is a good time to contribute to the development of Squeak by participating in the Harvesting Process. One of the main ways to contribute is by Reviewing submitted fixes and enhancements. For information about the Harvesting Process see http://minnow.cc.gatech.edu/squeak/3152 If you have the time please attend and work toward the future of Squeak. The next seven chats are scheduled for 04:00 (04:00 AM) GMT Tuesday August 01, 2006 08:00 (08:00 AM) GMT Saturday August 05, 2006 12:00 (12:00 PM) GMT Wednesday August 09, 2006 16:00 (04:00 PM) GMT Sunday August 13, 2006 20:00 (08:00 PM) GMT Thursday August 17, 2006 00:00 (12:00 AM) GMT Tuesday August 22, 2006 04:00 (04:00 AM) GMT Saturday August 26, 2006

1 0

Re: DependentsArray size and subscript is out of bounds
by nicolas cellier 23 Jul '06

23 Jul '06

Le Jeudi 20 Juillet 2006 00:38, mathieu a écrit : > Hi, > > It is normal that: > > (DependentsArray new: 100) size > > return 0? Yes > If yes so why? DependentsArray count only non nil elements > And if I do: > | o | > > o := WriteStream on: (DependentsArray new: 100). > 1 to: 24 do: [:each | o nextPut: Object new]. "do it sevral time" > > it raise an error 'subscript is out of bounds: 22' from time to time > why? You cannot do that with DependentsArray. DependentsArray are weak array: that mean that storing an object into a DependentsArray does not prevent it from being garbageCollected. When an element is garbage collected, corresponding index in DependentsArray is replaced by nil. All is occuring in pastEndPut: Since initial size (non nil elements) is zero, then the array is grown to size: oldSize + ((oldSize max: 20) min: 1000000) When no garbageCollection occurs, that works fine (except a new DependentsArray is created each time you add an Object). When a garbage collect occurs, all unreferenced (Object new) are removed from the DependentsArray, except one (I do not know who is referencing it, maybe the BlockContext...). Thus you allocate a DependentsArray of size 21 and fail when storing element number 22... > Thx > Math What do you need DependentsArray for? Nicolas

6 12

Re: [ANN] small miracle :-) The Rest of Squeak for the kernel image
by Alan Kay 23 Jul '06

23 Jul '06

Great! Just curious ...how many lines of code is all this? Cheers, Alan ----------- At 01:29 AM 7/22/2006, Pavel Krivanek wrote: >This is the experimental package The Rest of Squeak for the kernel image: > >http://www.comtalk.net/Squeak/uploads/20/RoS.zip > >Use kernel image with condensed changes. The size of the resultant >image is about 9.9MB and it doesn't include FFI and Speach. > >The code of this package is generated automatically. Only the >initialization order and operations is modified manually. > >The result is working Morphic image. Of course, not everything works >well but at least it shows that removing and reloading of Morphic, MVC >etc. is possible (see attached screenshot). > >Enjoy... > >-- Pavel > > >Content-Type: image/jpeg; name=world.jpg >X-Attachment-Id: f_epxpijt6 >Content-Disposition: attachment; filename="world.jpg"

3 2

[ANN] small miracle :-) The Rest of Squeak for the kernel image
by Pavel Krivanek 23 Jul '06

23 Jul '06

This is the experimental package The Rest of Squeak for the kernel image: http://www.comtalk.net/Squeak/uploads/20/RoS.zip Use kernel image with condensed changes. The size of the resultant image is about 9.9MB and it doesn't include FFI and Speach. The code of this package is generated automatically. Only the initialization order and operations is modified manually. The result is working Morphic image. Of course, not everything works well but at least it shows that removing and reloading of Morphic, MVC etc. is possible (see attached screenshot). Enjoy... -- Pavel

3 2

[SqNOS] Release 16-Jul-2006: Basic networking is working!
by Gerardo Richarte 22 Jul '06

22 Jul '06

Javier Diaz-Reinoso dijo: > >/ I've tested now in a VirtualPC and worked ok! > / > Same here, however the mouse when move have "droppings" over the screen. > > This "dropping" are a problem of how the mouse cursor is drawn. When squeak runs on an operating system the cursor is normally handled by the system, however, here, we have to take care of that. And Squeak doesn't have nice hooks to know when it's going to draw to the screen, so hidding and showing the cursor again in the right moments is not trivial. However, this can (and must) be solved by touching a few methods in Squeak, so if anybody wants to give it a try and find a nice and clean solution, please, go ahead!!!! thanks for all the tests! I should start taking notes of all reports gera

1 0

REMINDER: Squeak Chat Sunday July 23rd 20:00 GMT
by Ken Causey 22 Jul '06

22 Jul '06

Just a reminder that the next Squeak Chat will take place between 20:00 (8 PM) GMT Sunday July 23rd and 00:00 (Midnight) GMT Monday July 24th, 2006. As usual we will meet in #squeak on irc.freenode.net. More information about the #squeak IRC channel can be found at http://people.squeakfoundation.org/article/7.html This is a good time to contribute to the development of Squeak by participating in the Harvesting Process. One of the main ways to contribute is by Reviewing submitted fixes and enhancements. For information about the Harvesting Process see http://minnow.cc.gatech.edu/squeak/3152 If you have the time please attend and work toward the future of Squeak. The next seven chats are scheduled for 00:00 (12:00 AM) GMT Friday July 28, 2006 04:00 (04:00 AM) GMT Tuesday August 01, 2006 08:00 (08:00 AM) GMT Saturday August 05, 2006 12:00 (12:00 PM) GMT Wednesday August 09, 2006 16:00 (04:00 PM) GMT Sunday August 13, 2006 20:00 (08:00 PM) GMT Thursday August 17, 2006 00:00 (12:00 AM) GMT Tuesday August 22, 2006

1 0

PSHORT (was RE: Help with FFI - another try)
by Peter Crowther 22 Jul '06

22 Jul '06

> From: Eugene Beschastnov > [PSHORT] is typedef for > 'short*', but it's standard typedef. Really? Which ANSI version introduced that, then? I'm aware of the grotesque Hungarian notation adopted by Microsoft for years, and hence foisted on the rest of the world; but even they have moved away from it for .Net. - Peter

1 0

[SqNOS] Release 16-Jul-2006: Basic networking is working!
by Gerardo Richarte 22 Jul '06

22 Jul '06

> Boot ok, but I don't get control about the image, the mouse nor > keyboard seems to respond. uhm... is it a ps/2 (or older) keyboard with a ps/2 mouse? if not, well... it won't work. Like USB mouse and keyboards are not supported yet :-( if not, try again, and don't touch anything until the Transcript stops moving (won't be too long). I'm not sure, but there may be some issues with keyboard events triggered before initialization is done. Let me know what happens, and thanks for testing! gera

4 4

← Newer
1
...
4
5
6
7
8
9
10
...
25
Older →

Jump to page: