Squeak-dev June 2006

squeak-dev@lists.squeakfoundation.org

149 participants
250 discussions

BUG FFI/unix vm ? Hackers help requested !
by nicolas cellier 29 Jun '06

29 Jun '06

Hello gods of FFI and of the vm, I am blocked for a while in the development of Smallapack (Smalltalk interface to LAPACK) on squeak. I have a very strange behaviour isolated on a small test case (see below): I want to call DLANGE, a LAPACK FORTRAN routine to compute the norm of a matrix. And on the seventh call, i always get a stange result (result is 0.0 but will answer false to = 0.0). This fail on unix with image 3.9alpha7029, vm is: squeak -version 3.7-7 #1 Sat Mar 19 13:12:20 PST 2005 gcc 3.3.5 Squeak3.7 of '4 September 2004' [latest update: #5989] Linux squeak.hpl.hp.com 2.4.27-1-386 #1 Fri Sep 3 06:24:46 UTC 2004 i686 GNU/Linux default plugin location: /usr/local/lib/squeak/3.7-7/*.so Same code seem to work on windows (i have tried it today)... Beware, i presume this can corrupt your image. Maybe i am doing something wrong, could someone explain me please ? Nicolas ---------------------------------------------------------------------------------------------------- My definition of dlange2.c (translated with f2c then modified to simply answer 0.0) is: /* #include "f2c.h" */ typedef double doublereal; typedef long integer; typedef long ftnlen; /*< DOUBLE PRECISION FUNCTION DLANGE( NORM, M, N, A, LDA, WORK ) >*/ doublereal dlange2_(char *norm, integer *m, integer *n, doublereal *a, integer *lda, doublereal *work, ftnlen norm_len) { return 0.0; } /* dlange2_ */ you just compile with: gcc -c dlange2.c; ld -shared -o libdlange2.so dlange2.o and call from Squeak with: ExternalLibrary subclass: #DLANGE2Library instanceVariableNames: '' classVariableNames: '' poolDictionaries: '' category: 'Smallapack-Test-DLANGE' DLANGE2Library class>>moduleName ^'dlange2' DLANGE2Library>>dlange2Withnorm: norm m: m n: n a: a lda: lda work: work length: lengthOfnorm <cdecl: double 'dlange2_'( char * long * long * double * long * double * long )> ^self externalCallFailed DLANGE2Library class>>testDLANGE2 "DLANGE2Library testDLANGE2" | a m n lda norm cm cn clda work | m := lda := 3. n := 4. a := ExternalData fromHandle: (ByteArray new: m*n*8) type: ExternalType double. "AS FORTRAN IS PASSING POINTERS, DO ALLOCATE ExternalData" cm := ExternalData fromHandle: ((ByteArray new: 4) signedLongAt: 1 put: m; yourself) type: ExternalType long. cn := ExternalData fromHandle: ((ByteArray new: 4) signedLongAt: 1 put: n; yourself) type: ExternalType long. clda := ExternalData fromHandle: ((ByteArray new: 4) signedLongAt: 1 put: lda; yourself) type: ExternalType long. norm := 'M'. work := nil. ^(1 to: 10) collect: [:i | (self new dlange2Withnorm: norm m: cm n: cn a: a lda: clda work: work length: 1) = 0.0] "you always obtain false from the seventh entry on..."

3 5

Re: BUG FFI/unix vm ? Hackers help requested !
by ncellier＠ifrance.com 29 Jun '06

29 Jun '06

Hello vincent, yes, the FFI call is always returning 0.0, but then the vm fails somewhere after when testing = 0.0. If you execute step by step in the debugger, it does not fail... something weird. I guess we'd better quit without saving the image when such symptoms occurs... Nicolas bouchet vincent: > Hi nicolas, > > I just test this : > > ^(1 to: 10) collect: [:i | r := (self new dlange2Withnorm: norm m: cm n: cn a: a lda: clda work: work length: 1) = 0.0. > Transcript show: r asString. > r ]. > > > And its works...... but if I remove the "Transcrip show" I obtain : #(true true false false false false false false false false) > > vincent. > > 2006/6/28, nicolas cellier <ncellier(a)ifrance.com>:Nobody alive on the thread? > The bug is now at http://bugs.impara.de/view.php?id=3929 > > Nicolas > > > > ________________________________________________________________________ iFRANCE, exprimez-vous ! http://web.ifrance.com

1 0

Squeak Chat Monday July 3rd 00:00 GMT
by Ken Causey 29 Jun '06

29 Jun '06

Just a reminder that the next Squeak Chat will take place between 00:00 (Midnight) and 04:00 (4 AM) GMT Monday July 3rd, 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 Friday July 07, 2006 08:00 (08:00 AM) GMT Tuesday July 11, 2006 12:00 (12:00 PM) GMT Saturday July 15, 2006 16:00 (04:00 PM) GMT Wednesday July 19, 2006 20:00 (08:00 PM) GMT Sunday July 23, 2006 00:00 (12:00 AM) GMT Friday July 28, 2006 04:00 (04:00 AM) GMT Tuesday August 01, 2006

1 0

Looking for papers about 'Dealing with workflows in Smalltalk'
by Damien Cassou 28 Jun '06

28 Jun '06

Hi, Karsten Kusche and I are working for Georg Heeg eK to improve Smalltalk developers workflow while keeping flexibility. Ideas have to be grabbed to help average Smalltalk developers to make a good design and follow an established workflow. We would like to finish our work with a tool that everybody can use for development. It should be an extension to the existing browser that displays more information. Attention is being put in easing unit tests, making the comprehension of an existing system simpler, and giving lots of information about methods, classes and packages. If you have ideas, publications or things that we can test or read, it would be interesting to share with us. Thank you for your help, we hope we will be able to present something at ESUG 2006. Bye -- Damien Cassou

4 6

Indic font can be seen in squeak
by Prabhakar katlakunta 28 Jun '06

28 Jun '06

Hi All, We have worked on providing Indic support to squeak. And we are successfull in providing partial support to it. We have Utilized the support of "Pango" and "EasyBMP". the source file is at http://www.yousendit.com/transfer.php?action=download&ufid=829E9F8538F936A9 regards, CITE, IIIT Hyderabad, INDIA.

1 0

Workflow with state pattern
by Mathieu SUEN 28 Jun '06

28 Jun '06

Hi, I was wondering if it make sens to use state pattern for a worklow. I have to implement a system to let user to place an order. Also if you have some small stuff about it I will be happy to know them. :-) Thanks Cheers, Math

3 3

Mock Objects
by Torsten Bergmann 28 Jun '06

28 Jun '06

Hi Stef, the MockObject class from Chad is just a simplified version of the Teachable package I wrote about before. Teachable is able to handle more cases. see http://astares.blogspot.com/2005/04/teaching-behavior.html Since Teachable does not implement #initialize (like it is the case in Chads MockObject class) you are also able to teach the mock how to behave when an initialize method is sent to it. Just download the package using SM and have a look at the provided test case. Teachable is also more intention revealing. Instead of just working with blocks you can write Chads example with Teachable as: cart := Cart new. item := Teachable new. item whenSend: #total return: 2.99. cart add: item ... But as I warned in the previous email: creating dynamic mock objects (using either Chads code or the Teachable package) is sometimes dangerous. Think of testing persistence layer code in an application. If you want to get the tests running without having an actual database connection you can mock it. You can either either teach an object dynamically how to behave like the real DB connection class or implement the mock also as a class with the same signature. In such a case I would prefer to implement a real mock class since it can become a nightmare to debug the code with the #doesNotUnderstand: trick (especially when it uses blocks). It is also hard if you refactor the mocked code - this always means to refactor the mock object in a similar way. The more behavior/logic to mock the better it is to implement a special mock class with a similar interface. We just need better tools. Smalltalk with it's dynamic nature is very powerful - especially for testing. Think of a general mock framework where mocks can dynamically created from the "mocked object" or where one can use SUnit to even test the mock: "self assert: (aMockObject behavesLike: mockedObject)". Bye Torsten -- "Feel free" – 10 GB Mailbox, 100 FreeSMS/Monat ... Jetzt GMX TopMail testen: http://www.gmx.net/de/go/topmail

3 2

Mock Objects
by Chad Nantais 28 Jun '06

28 Jun '06

Hi. Could someone please direct me to a Squeak project (or some documentation) that uses mock objects in its tests, specifically dealing with testing network code like a web service. I'm looking for a way to speed up some unit tests that hit a remote service. Thanks in advance, Chad

4 6

Acquiring multiple locks
by Andreas Raab 28 Jun '06

28 Jun '06

Hi Folks - I have an interesting little problem that can be trivially solved with help from the VM but that I'm curious if there is a way of doing it from Squeak directly. Here is the problem: Given a set of N locks (semaphores), how can multiple processes acquire subsets of those locks without deadlocking? E.g., it's fairly obvious that if one process attempts to acquire lock A and B and another one lock B and A that this leads to deadlock in a hurry. What's in particular problematic for my use case is that I don't know which semaphores are part of said set of locks and that I can't establish a total order to make sure that lock A always gets acquired before lock B. What I've been contemplating is to add a primitive which does the following: It attempts to acquire all the locks given as argument and if one of them cannot be acquired it immediately releases all the locks acquired so far and leaves the process suspended on the "failing" lock. The return value will indicate whether all the locks were acquired and therefore one can write a loop like here to acquire all of the locks: [self primitiveAcquireAll: locks] whileFalse. In other words, if all the locks can be acquired the code just marches through. If it fails to acquire a lock the process will stay suspended until that particular lock is released and then retry to acquire the locks. Is it possible to do that without a VM change? Cheers, - Andreas

11 16

Indic Support in Squeak
by Prabhakar katlakunta 28 Jun '06

28 Jun '06

Hi All, my mail got bounced because i included the source.. this support is in Windows OS. i am providing the link from where the source file can be downloded... http://www.yousendit.com/transfer.php?action=download&ufid=829E9F8538F936A9 Our support is partial. please have a look at it ... we have utilized the support of "Pango" for rendering. and "EazyBMP" for building the bitmap. this bitmap is brought into squeak .... this is a monocrome bitmap. regards, CITE, IIIT Hyderabad, INDIA.

1 0

← Newer
1
2
3
4
5
6
7
...
25
Older →

Jump to page: