Squeak-dev January 2001

squeak-dev@lists.squeakfoundation.org

112 participants
318 discussions

Re: newbie. morphic is sluggish on my PC
by Lex Spoon 30 Jan '01

30 Jan '01

Doug Way <dway(a)riskmetrics.com> wrote: > As far as I know, MVC in Squeak has always done this, so it's not a new problem. (I remember up until 2.4 or so, Morphic did it too...) > No. For a bit over 2 years MVC hasn't hogged CPU in most of its idle states. See Controller>>interActivityPause and its senders. And on *my* nice slow machine, with Squeak 2.9a-3193, it in fact works fine -- MVC takes very low CPU when idle. It's probably the same bug, a #< where #<= would be appropriate: wait < MinActivityLapse "big waits happen after a snapshot" ifTrue: [DisplayScreen checkForNewScreenSize. What an idiot of a programmer; it's really amazing that this code ever woked. -Lex

1 0

Re: So where's the 2.8pre5 Unix sources?
by Lex Spoon 30 Jan '01

30 Jan '01

Ned Konz <ned(a)bike-nomad.com> wrote: > You have diffs against "Ian's Squeak 2.8 pre5" there, but I can't find it, > either at ftp://ftp.inria.fr/INRIA/Projects/SOR/users/piumarta/squeak > or at http://www-sor.inria.fr/~piumarta/squeak/unix/ > > Where can I find these sources? > > The latest one I can find (I think) is: > ftp://ftp.inria.fr/INRIA/Projects/SOR/users/piumarta/squeak/Squeak-2.9a.pre1-src.tar.bz2 > It's hard to find. (Digging through email...). Ah, it's at: http://www-sor.inria.fr/~piumarta/squeak/Jitter-3.1/ppc-linux-gnu/ I guess I'll link this from my page. -Lex

1 0

Re: [RFI] autogenerated type info in class comments
by Vassili Bykov 30 Jan '01

30 Jan '01

I uploaded the type collector on the Swiki in case someone wants to play with it separately from Whisker: http://minnow.cc.gatech.edu/squeak/uploads/TypeCollector.cs It has a few changes compared to the version I posted to this list about a year ago; the changes are essentially the same Doug made--limiting the number of instances to consider in case there are too many. As far as I remember, it modifies the comment template to include type information. My guess is it should be easy to combine this with Dan's inferencer, just make the inferencer produce one of TcTypes and then use it as a seed when collecting types. Doug Way wrote: > (I got this idea from a recent thread on comp.lang.smalltalk, in which > it was suggested that some sort of double-quoted warning that "THIS > CLASS IS NOT YET COMMENTED" should be added at the end of the class > definition pane in VisualWorks browsers, which is also an excellent idea > for Squeak. When all else fails, shame people into adding comments. ;-) ) I refined the idea a little since then, having implemented and used it for while. Here is what I found, in case someone wants to go in this direction. - It's really neat and more convenient than separate browser modes. Even though you only see the top of a comment this way, it's often enough to attract your attention and "seduce" into reading further. On a couple of occasions already, I've seen interesting things I would never see otherwise because I would never intentionally go read those classes' comments. - Another possible approach would be rolling the comment into the class definition message. I prefer to just have the browser combine them instead. This way it's easier to deal with formatted comment text. - For the same reason, I dropped the idea that the comment should be appended as a proper Smalltalk comment, in double quotes (and with double quotes inside doubled). Double double quotes look too weird, and as you edit the comment it's too easy to forget to double them. Instead, I made the definition view have two parts, upper for the definition, lower for the comment text. The comment text appears "as is", without any quotes or escapes. Two view parts are not in the sense of two separate widgets. It's a single scrollable text view, separated into two portions. When you accept, the contents of two portions are saved separately. In my experience, this is the most convenient arrangement. Yes, it does show "THIS CLASS HAS NOT BEEN COMMENTED" if the comment is missing, and yes, it is a strong shame factor. :) Cheers, --Vassili -- Vassili Bykov VisualWorks Development vassili(a)parcplace.com

1 0

RE: Fwd: Verizon looking for Smalltalk programmers
by Norton, Chris 30 Jan '01

30 Jan '01

Hmm. I didn't get one. So maybe you all belong to another list? Cheers, ---==> Chris PS> Maybe they just don't want ME! :-) > -----Original Message----- > From: Florin X Mateoc [SMTP:mateoc_florin@jpmorgan.com] > Sent: Monday, January 29, 2001 7:56 PM > To: squeak(a)cs.uiuc.edu > Subject: Re: Fwd: Verizon looking for Smalltalk programmers > > This is a case of spamming, I also got the same message, on my work (this) > address. I never use this address on c.l.s, so I guess somebody grabbed > addresses from the squeak mailing list... > > Florin > > > > > > dway(a)riskmetrics.com on 01/29/2001 06:32:34 PM > > Please respond to squeak(a)cs.uiuc.edu > > To: squeak(a)cs.uiuc.edu > cc: (bcc: Florin X Mateoc) > Subject: Re: Fwd: Verizon looking for Smalltalk programmers > > > > > > Jeff Szuhay wrote: > > > > Don't know how I got on this list but here is a request > > for programmers in Smalltalk (how often does one see > > THAT nowadays?) > > Actually, they're not that uncommon if you look on comp.lang.smalltalk. > Usually > a few a day or so. So, we probably don't really need to post general > Smalltalk > jobs here, IMHO. (no offense taken at your posting, though. :) ) > > Squeak-specific jobs, on the other hand, by all means should be posted > here... > they're still somewhat few and far between at this point. Hopefully that > will > be less true in the future. :) > > - Doug Way > dway(a)riskmetrics.com > > > > > > > > > This communication is for informational purposes only. It is not intended > as > an offer or solicitation for the purchase or sale of any financial > instrument > or as an official confirmation of any transaction. All market prices, data > and other information are not warranted as to completeness or accuracy and > are subject to change without notice. Any comments or statements made > herein > do not necessarily reflect those of J.P. Morgan Chase & Co. Incorporated, > its > subsidiaries and affiliates. >

1 0

Re: newbie. morphic is sluggish on my PC
by Bob Arning 30 Jan '01

30 Jan '01

On Mon, 29 Jan 2001 20:09:43 -0500 Doug Way <dway(a)riskmetrics.com> wrote: >"Pennell, David" wrote: >> >> > Then I tried looking at the code in the MessageTallys that >> > were just posted. I tried changing WorldState>>initialize so >> > that the MinCycleLapse is 40 instead of 20 (and >> > re-initializing), ... >> >> Thanks, Bob Arning sent me a fix for this yesterday. I saw similar >> results - the CPU usage dropped to around 1%. Bob said he was going >> to drop it in the update stream. > >Cool beans. > >> FWIW - MVC still takes all of the machine. Doug, (and any others interested in cpu usage), I am attaching the change I gave David that dropped the cpu % quite dramatically. I found this on a Mac which doesn't have quite the same processor monitoring facilities and while I expected it to help, I was suprised at the amount. I don't know if this drop is an image thing, a VM thing or an OS thing - if you get any ideas after trying it, I would love to hear them. Cheers, Bob 'From Squeak2.9alpha of 17 July 2000 [latest update: #3316] on 28 January 2001 at 4:29:49 pm'! "Change Set: interCyclePause Date: 28 January 2001 Author: Bob Arning A couple of tweaks to the inter-cycle pause: - There was a test for the wait being less than the target. If the cpu was fast, the wait could be equal and would thus be skipped. Changed to less or equal. - The lastCycleTime was recorded as the time before the wait which meant that on the next cycle it might appear that no wait was necessary. Changed to record the time after the wait. The effect of these changes is to reduce the number of morphic cycles run per second to something closer to the number expected for the value of MinCycleLapse. This means lower cpu utilization by Squeak when nothing is going on (which is good). It also means a possibly lower frame rate (which may not be so good). If the latter is the case, then we may need to think about giving the user some control over the cycle rate."! !WorldState methodsFor: 'update cycle' stamp: 'RAA 1/28/2001 16:22'! interCyclePause: milliSecs "delay enough that the next interaction cycle won't happen too soon after the original; thus, if all the system is doing is polling for interaction, the overall CPU usage of Squeak will be low" | currentTime wait | currentTime _ Time millisecondClockValue. (lastCycleTime notNil and: [CanSurrenderToOS ~~ false]) ifTrue: [ wait _ lastCycleTime + milliSecs - currentTime. wait > 0 ifTrue: [ wait <= milliSecs "big waits happen after a snapshot" ifTrue: [DisplayScreen checkForNewScreenSize. (Delay forMilliseconds: wait) wait ]. ]. ]. "record the time AFTER the wait" lastCycleTime _ Time millisecondClockValue. CanSurrenderToOS _ true.! !

1 0

RE: newbie. morphic is sluggish on my PC
by Pennell, David 30 Jan '01

30 Jan '01

[snip] > Then I tried looking at the code in the MessageTallys that > were just posted. I tried changing WorldState>>initialize so > that the MinCycleLapse is 40 instead of 20 (and > re-initializing), expecting CPU usage to be cut in half if I > were lucky. But it dropped down to 1%! Squeak's > responsiveness still seemed fine from what I could tell, too, > although this is a pretty fast machine. Trying some > different numbers revealed that there's a big jump (at least > on my PC) from setting MinCycleLapse to 31 versus 33. This > must be some sort of magic threshold at which the VM starts > churning events, or something, I don't know. Thanks, Bob Arning sent me a fix for this yesterday. I saw similar results - the CPU usage dropped to around 1%. Bob said he was going to drop it in the update stream. FWIW - MVC still takes all of the machine. -david

2 1

Re: [ENH][VM][UNIX]generalized async IO
by Lex Spoon 30 Jan '01

30 Jan '01

Ned Konz <ned(a)bike-nomad.com> wrote: > On Sunday 28 January 2001 02:20, Lex Spoon wrote: > > > > > The code for all this is at the bottom of: > > > > http://www.cc.gatech.edu/~lex/squeak/ > > I didn't see it... is it hidden somewhere? > Forgot -- our web servers only update once a day around 6:00 am. It's up now! -Lex

3 2

Re: Fwd: Verizon looking for Smalltalk programmers
by Florin X Mateoc 30 Jan '01

30 Jan '01

This is a case of spamming, I also got the same message, on my work (this) address. I never use this address on c.l.s, so I guess somebody grabbed addresses from the squeak mailing list... Florin dway(a)riskmetrics.com on 01/29/2001 06:32:34 PM Please respond to squeak(a)cs.uiuc.edu To: squeak(a)cs.uiuc.edu cc: (bcc: Florin X Mateoc) Subject: Re: Fwd: Verizon looking for Smalltalk programmers Jeff Szuhay wrote: > > Don't know how I got on this list but here is a request > for programmers in Smalltalk (how often does one see > THAT nowadays?) Actually, they're not that uncommon if you look on comp.lang.smalltalk. Usually a few a day or so. So, we probably don't really need to post general Smalltalk jobs here, IMHO. (no offense taken at your posting, though. :) ) Squeak-specific jobs, on the other hand, by all means should be posted here... they're still somewhat few and far between at this point. Hopefully that will be less true in the future. :) - Doug Way dway(a)riskmetrics.com This communication is for informational purposes only. It is not intended as an offer or solicitation for the purchase or sale of any financial instrument or as an official confirmation of any transaction. All market prices, data and other information are not warranted as to completeness or accuracy and are subject to change without notice. Any comments or statements made herein do not necessarily reflect those of J.P. Morgan Chase & Co. Incorporated, its subsidiaries and affiliates.

1 0

Color Map flashing on Unix VM
by William Barnett-Lewis 30 Jan '01

30 Jan '01

Just installed Squeak 2.7 on my Sun IPX under NetBSD 1.5. It has the standard 8bit GX framebuffer installed with a greyscale monitor attached. On startup and whenever I try to do anything, I get a very bad level of color map flashing. I notice that, unlike the Mac VM, there is no option for 1,2, or 4 bit color. Is this something that can be compiled back in? Or perhaps there is some other work around? I did check the swiki, but didn't see anything. Thanks, William -- You better watch out, what you wish for; It better be worth it, so much to die for. Courtney Love

2 5

Fwd: Verizon looking for Smalltalk programmers
by Jeff Szuhay 30 Jan '01

30 Jan '01

Don't know how I got on this list but here is a request for programmers in Smalltalk (how often does one see THAT nowadays?) Anyway, here it is. Please do not reply to me. <mailto:joel.poulter@verizon.com> Jeff 8-) >From: joel.poulter(a)verizon.com >Date: Mon, 29 Jan 2001 13:59:57 -0500 >Subject: Smalltalk > >Jeff, > >I represent Verizon Communications, a Fortune 10 company formed by the >merger of Bell Atlantic and GTE. Verizon is currently the largest >Telecommunications Company in the United States and one of the largest >in the world. We are building systems that represent the future of >telecommunications and are searching for top technical talent to help >us accomplish this mission. > >We are actively seeking individuals with Smalltalk OOA/D development >experience. These permanent positions are located in the Arlington, >Virginia area. The positions offer relocation, excellent benefits and >salary. Included in this email is additional details about the >project (Express Trak) and Arlington. > >If you or someone you know might be interested in learning more about >these opportunities, please let me know. Thank you for your time! > >Sincerely, >Joel Poulter >Verizon Recruitment > > >Project Info: >Express Trak is Verizon's implementation of a Strategic Billing System >targeted to replace the existing legacy Billing System across all of >its geographical regions. A multi-release approach has been developed >to accomplish this task. The main application is a client / server >solution developed with Smalltalk and Gemstone on the front end and >Cobol and DB2 on the backend. Each Express TRAK release will >incorporate more of the existing functionality. It is an ongoing >long-term project meaning stability, but still offers full life cycle >development projects. This project supports several products >including long distance, ATM, ADSL and Frame Relay. Verizon's Express >Trak is the largest smalltalk project ever implemented. >City Info: >Arlington is located in the country's most influential region and the >power capital of the globe -- the Washington D.C. metropolitan area. >The city is home to Washington National Airport making D.C. easily >accessible. All major roadways and the five major bridges into >Washington, D.C. from Virginia come through Arlington. Eleven of the >thirty-three stations for D.C.'s Metro Subway system are located in >Arlington. > >Arlington has been on the cutting edge of technology since the >beginning of the digital revolution. The Internet was born here, at >the Defense Advanced Research Project Agency and the National Science >Foundation. Today, Arlington is at the center of the region known as >the "Netplex," as Fortune magazine has labeled the nation's fastest >growing communications center. > >The city has a 4.5% Sales Tax and has a lower cost of living than most >major cities, including Boston, New York, San Francisco, and Toronto. -- ************************************************** Jeff Szuhay A randomly-directed www.szuhay.org chaotical wetware pattern jeff(a)szuhay.org recognizer/generator. <this space for rent>

2 1

Jump to page: