Squeak-dev October 2007

squeak-dev@lists.squeakfoundation.org

151 participants
195 discussions

a meta comment (Re: Multy-core CPUs)
by Yoshiki Ohshima 26 Oct '07

26 Oct '07

Hello list, but mostly Jason and Peter, This is a meta-comment, but could you guys please refrain from writing many emails in a row on the same topic, just to comment on different paragraphs? People who is participating to this discussion should have had enough time to gather certain ideas, so instead of replying to paragraph by paragraph (that in theory results in exponential increse of traffic), writing up a coherent story might help. -- Yoshiki

3 2

Re: Multy-core CPUs
by Rob Withers 26 Oct '07

26 Oct '07

Hi Peter, My response here is not formatted as I would wish, but is the imit of my current technology. I work on SqueakElib, and though I am certainly no expert on E, I am somewhat familiar with it's design and implementation. That said, a comment you made caught my eye... > ----- Original Message ---- > From: Peter William Lount peter(a)smalltalk.org > > If a number of messages are waiting in the input queue of a process > that can only process one message at a time since it's not multi-threaded > then those messages are BLOCKED while in the thread. Now imagine > another process with messages in it's queue that are also BLOCKED > since they are waiting in the queue and only one message can be > processed at a time. Now imagine that process A and process B each > have messages that the other needs before it can proceed but those > messages are BLOCKED waiting for processing in the queues. > > This is a real example of what can happen with message queues. The > order isn't guaranteed. Simple concurrency solutions often have deadlock > scenarios. This can occur when objects must synchronize events or > information. As soon as you have multiple threads of execution you've got > problems that need solving regardless of the concurrency model in place. Messages don't get blocked, when they are queued for the Vat (thread/process) to service them. They do get queued when they are sent to a Promise, but it does not affect the Vat. When the Promise resolves, the messages are forwarded to the Vat for processing, in order, so partial msg ordering is maintained. Msg ordering is partial because of this case: object a in Vat A, object b in Vat B, and object c in Vat C are intercommunicating. Both b and c have references to a. At time t1, b sends msg m1 to a, at time t2, c sends msg m2 to a, but there is higher network latency for b, so m2 gets to a at t3 and m1 gets to a at t4 and they are out of order from the remote invocation order. I hope this helps. Read more about Elib. cheers, Rob

6 6

Re: Does anybody use < to compare lookup keys?
by Jerome Peace 26 Oct '07

26 Oct '07

Does anybody use < to compare lookup keys? Hi Nicolas, Probably everybody. <list of associations> asSortedArray for example. The problem is at the String >>#< level both #a < 1 and #a < #1 cause errors. The argument needs to be treated more polymorphicly (adapt to string and send type stuff ) and this ought be a mantis item. Cheers. Hth, Yours in curiosity and service, -Jerome Peace *** >nicolas cellier ncellier at ifrance.com >Mon Oct 8 22:58:29 UTC 2007 > >LookupKey as a descendant of Magnitude does implement <, suggesting the >set of LookupKeys can be fully ordered. >(that's the logic behind Margnitude >=, <=, and >). > >Though, you cannot compare arbitrary keys. >(1->nil) < (#a->nil). "false" >(#a->nil) < (1->nil). "MNU isBytes" >(Object new->nil) < (Object new ->nil). "MNU <" > >So, is there any application of LookupKey < ? *** - __________________________________________________ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com

1 0

Commanche
by Jason Johnson 25 Oct '07

25 Oct '07

Hi, I just noticed from my RSS feed that Squeaksource got a set of changes to Commanche put in like a machine gun (every 20 seconds or so). Is that legit, or is this some kind of replay thing? If it's legit, what's going on? :) Are the longstanding bugs people talk about getting fixed?

5 14

[ANN] Tree library for Squeak
by Hernán Morales 25 Oct '07

25 Oct '07

Hello. I've ported the Bruce Samuelson TreeLW implementation from VisualWorks to Squeak. http://www.squeaksource.com/TreeLW.html Hope it helps somebody. Cheers Hernán

2 2

Re: Multy-core CPUs
by Rob Withers 25 Oct '07

25 Oct '07

Peter, Of all the links you have been given, put this one at the top of your list. Commit to spending an half hour reading, much less than this thread! :) > (concerning distributed programming, and event-loop concurrency) > http://www.erights.org/elib/index.html I also wasn't as clear as I could have been. > When a message is sitting in an inbound queue on a process waiting > for that process to do get to it it is in essence blocked until it gets > processed. If the process is only processing one inbound message at > once it is possible for a dead lock to occur when multiple processes > are in a similar situation with each other. Both processes waiting for > the message that would let them continue which is sitting blocked on > the other's queue. That's all. Classic deadlock case. Ok, with your terminology, and some of mine, a message in a queue feeding a Vat is blocked until the Vat gets to it. Ditto with Vat 2. The mistake in your classic deadlock case is to assume that the message could also get blocked because it is waiting on the msg in Vat 2. This does not happen. If there was an unresolved reference waiting on resolution from a remote computation, it is represented as a promise. So message 1 would have a promise for the outcome of message 2. It would NOT be on the queue for Vat 1. It would be sitting as the argument in a #whenMoreResolved msg to the promise. And when that promise resolves, it would then run. Now, you are further saying that promise will not resolve because both are mutually dependent on each others results. So you are saying that in a sense they are deadlocked. Again this cannot happen. There is no way for you to kick off computation 1, with the promise to computation 2, unless the computation 2 had already been kicked off. DItto for the other direction. So it is impossible to construct this promise deadlock. Cheers, Rob

1 0

Inst var mix-up after adding accessors with OB
by Zulq Alam 25 Oct '07

25 Oct '07

If you follow these steps using OB 0.382: 1) Define class Thing 2) Add inst var "foo" 3) Add accessors for "foo" using menu 4) Add inst var "bar" AFTER "foo" 5) Add accessors for "bar" using menu the instance variable name list will be sorted from: instanceVariableNames: 'foo bar' to: instanceVariableNames: 'bar foo' and the accesors for "foo" will get and set "bar" such that: thing := Thing new. thing foo: 'foo'. thing bar: 'bar'. thing foo -> thing bar "'bar'->'bar'" If you upgrade to 0.391 the instance variable name list is not sorted so effectivley the problem goes away. I'd like to understand how the OB (or refactoring engine) could modify the class description without triggering a recompile of the methods? I tried writing a test case but without knowing how it's side stepping the usual recompile I got stuck. Regards, Zulq.

2 4

Heap>>first broken?
by Andreas Raab 25 Oct '07

25 Oct '07

Hi - Just stumbled over this: self should:[Heap new first] raise: Error. Right? However, in Squeak 3.9 and later it doesn't. From the 3.9 change log it appears it got introduced here: ------ Change Set: decompileBlock-md Date: 24 November 2005 Author: Marcus Denker A simplification for BlockContext>>decompileBlock: The Method got the tempNames by compiling the source of the method. This is already implemented in methodNode tempNames... thus not needed here. ------ remove emptycheck for Heap>>fist and SequencableCollection #first, #last, #middle ------ but no comment as to why. Any insights into the reasons for this change anyone? Cheers, - Andreas

3 2

Re: [Newbies] Delete translations
by stephane ducasse 25 Oct '07

25 Oct '07

Bert How do you store the translations? Because it would be cool to get them per packages as ressources package. I know that OLPC does not have mc but it would be good for Squeak. I would like to see a merge or flow from OLPC new features to Squeak. What would be the process to make sure that nothing is lost for squeak? Stef > > Actually, we do not use NLT anymore for OLPC. We exported all > translations to po files: > > https://translations.launchpad.net/etoys/first-deployment/+pots/etoys > > This lets use use the "gettext" translation tools that are the > standard in Linux. > >> It is very easy to add languages, merge, etc, but I do not find >> how to delete one of them. >> >> Another question: >> Which is Babel function? > > Not sure what the question is. > > - Bert - > > > _______________________________________________ > Beginners mailing list > Beginners(a)lists.squeakfoundation.org > http://lists.squeakfoundation.org/mailman/listinfo/beginners >

3 7

OSProcess - return value from command.
by John Thornborrow 25 Oct '07

25 Oct '07

Hello All, I've been trying to use OSProcess to execute commands; which works - however I need to be able to retrieve the value returned to the terminal by the command. To be precise, I am executing: OSProcess command: 'svn list svn://server/repo' and would like to retrive the string that is returned to the terminal. Could anyone advise on how to achieve this? Many thanks, John www.pinesoft.co.uk Pinesoft Computers are registered in England, Registered number: 2914825. Registered office: 266-268 High Street, Waltham Cross, Herts, EN8 7EA This message has been scanned for viruses by BlackSpider MailControl - www.blackspider.com

4 4

← Newer
1
2
3
4
5
6
7
8
9
...
20
Older →

Jump to page: