Uhm, this method (and many others in the extensions-category) have been present in their expected-by-Alien form in Pharo since at least 1.0 Probably a better idea to separate the Alien-"Core" into two packages, one which is actually needed, and one which you only need to load in really old Squeak/Cuis images in order to have it work. In which case you probably won't have method trailers either, and generate: #(0 0 0 0) is the proper extension. Come to think of it, you probably have to go back to pre-closure enabled for that to be the case, with the amount of platform support for alien vms offered back then, it might not be a bad idea to just ditch it entirely. Cheers, Henry On Sep 29, 2010, at 3:33 58PM, Fernando olivero wrote:
I fixed this bug,
MethodNode>>generate , was sending generate: #(0 0 0 0)
instead of sending self generate: CompiledMethodTrailer empty.
Name: Alien-Core-FernandoOlivero.64 Author: FernandoOlivero
On Sep 29, 2010, at 12:56 PM, Fernando olivero wrote:
This Error prevents from using Alien in the lastest Pharo image.
I will try to fix this today.
Also i will remove the ConfigurationOfAlien from the Alien repository, because the version in MetacelloRepository should be used instead. So there are no future misunderstandings
Thanks to Eliot, and using his precompiled Vms for all platforms Alien would be available now! (if the mentioned bug were fixed).
Fernando
On Sep 28, 2010, at 12:05 PM, Igor Stasenko wrote:
On 28 September 2010 11:51, Henrik Johansen <henrik.s.johansen@veloxit.no> wrote:
On Sep 28, 2010, at 10:34 16AM, Alberto Bacchelli wrote:
On 9/27/10 8:13 PM, Eliot Miranda wrote:
2010/9/27 Fernando olivero<fernando.olivero@usi.ch<mailto:fernando.olivero@usi.ch>> Does the prebuilt CogVM for linux include Alien support? http://www.mirandabanda.org/files/Cog/VM/VM.r2313/
Only of the data manipulation and callout parts, not the callback parts. I'm trying to find time to work on integrating the Alien callback machinery with the FFI.
best Eliot
We had a problem after installing Alien, when running the tests.
The following does not understand
MessageNotUnderstood: Array>>createMethod:header:methodClass:
sent by CompileMethod class>>newBytes:trailerBytes:nArgs:nTemps:nStack:nLits:primitive:
It appears to be cog related, is it?
Thank you, Alberto
No, it's Method-trailer related.
IIRC, the convention used to be to pass the empty trailer bytes #(0 0 0 0), find the offending sender and change it to pass CompiledMethodTrailer empty instead.
Right. A more compatible way is to use SomeClass defaultMethodTrailer Where SomeClass should be a class, where you installing this method.
Cheers, Henry
_______________________________________________ Pharo-project mailing list Pharo-project@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
-- Best regards, Igor Stasenko AKA sig.
_______________________________________________ Pharo-project mailing list Pharo-project@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
_______________________________________________ Pharo-project mailing list Pharo-project@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project
_______________________________________________ Pharo-project mailing list Pharo-project@lists.gforge.inria.fr http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project