Squeak-dev February 2001

squeak-dev@lists.squeakfoundation.org

144 participants
725 discussions

Re: Profiling help needed
by C. Keith Ray 02 Feb '01

02 Feb '01

I'm new to Squeak and this mailing list (I have Squeak 2.8.2 and Mac G4 450 MHz), could someone explain what "dsa prims" are? > Interesting results > > 400 MHz G3 MacOS: #(56426 67413 67494) no dsa prims > > So I rummaged around, found the dsa prim, put it in the Squeak folder and got: > > 400 MHz G3 MacOS: #(17516 49232 49119) w/dsa prim > > Hmm... that still doesn't look right, so I quit Squeak and restarted: > > 400 MHz G3 MacOS: #(17256 638 672) w/dsa prim plus restart > > I'm not clear on why the some of the primitives were detected on the second > test and others not until the third. > > Cheers, > Bob > ---- C. Keith Ray <http://homepage.mac.com/keithray/resume.html>

1 0

building a pluggable primitive
by Elzbieta Chludzinska 02 Feb '01

02 Feb '01

I'm trying to build a sample trial pluggable primitive using Slang that return the cosine of theta: myCosine |theta result| self export: true. self var: 'theta' declareC: 'float theta'. self var: 'result' declareC: 'float result'. self returnTypeC: 'float'. theta _ interpreterProxy stackFloatValue: 0. interpreterProxy success: (interpreterProxy isFloatObject: theta). interpreterProxy failed ifTrue:[^ nil]. self cCode: 'printf("theta = %f\n, theta)'. self cCode: 'result = cos(theta)' inSmalltalk: [result_theta cos]. self cCode: 'printf("result = %f\n", result)'. interpreterProxy failed ifTrue:[^ nil]. interpreterProxy pop: 2. interpreterProxy push: result. After #translate'ing the clase, I was forced to add #define HAVE_CONFIG_H to the generated .c file to get it compiled. When building the *.so, I used the following commands: gcc -I "../Squeak-2.8/src/generated" -c Test.c ld -shared -o Test.so Test.o I placed Test.so in the same directory as the VM. I'm running Squeak2.9a. When I run it, I get either primitiveFailed or VM crashed. Any suggestions??? Elzbieta

1 0

Swiki Server problems
by Jochen F. Rick 02 Feb '01

02 Feb '01

Hello, We are getting a strange error repeatedly on one of our Squeak servers. I was hoping somebody knows what is happening. Setup: Squeak2.8 Comanche 4.7 Swiki Beta 12. Solaris 2.8 VM (compiled by Lex from sources + a few file patches) Problem: All of a sudden, the Swiki no longer works and there is a debug window in the Squeak window. It says 'computation has been terminated'. This originates from MethodContext>>cannotReturn:. In the debug stack, the previous message is ComancheServer>>value:. After abandoning the error, stopping the server, and restarting, the server works again. Further Information: This is not happening on any of our Mac servers with a nearly identical setup. Also, the 2.8VM for Solaris seems to be having problems. I cannot for instance send the Squeak XWindow to another Solaris box. I can send it to my Linux box (RH6.2), but keystrokes don't go through. Headless mode doesn't work at all. Okay. Hopefully somebody has an answer or intuition. Thanks, Je77 and Bolot

1 1

Re: Profiling help needed
by Bob Arning 02 Feb '01

02 Feb '01

Interesting results 400 MHz G3 MacOS: #(56426 67413 67494) no dsa prims So I rummaged around, found the dsa prim, put it in the Squeak folder and got: 400 MHz G3 MacOS: #(17516 49232 49119) w/dsa prim Hmm... that still doesn't look right, so I quit Squeak and restarted: 400 MHz G3 MacOS: #(17256 638 672) w/dsa prim plus restart I'm not clear on why the some of the primitives were detected on the second test and others not until the third. Cheers, Bob

1 0

RE: [ENH][VM][UNIX]generalized async IO
by Jay Carlson 02 Feb '01

02 Feb '01

sr(a)evolgo.de [mailto:sr@evolgo.de] writes: > Bert Freudenberg wrote: > > > > On Sun, 28 Jan 2001, David T. Lewis wrote: > > > > > I would love to see this included in the Unix VM distribution. Ian? > > > > Exactly. Where's Ian? > > You could mail to him personally at > Ian.Piumarta(a)inria.fr > . > Personally I have the impression that Ian doesn't read and > communicates with > this mailing list continuously. I don't continuously read this mailing list either, so if you're really asking about the Helio port or BDF fonts, you'd better make sure my name is in the Cc: or To: fields.... :-) > But he is responsible for the Unix VM distribution though. > > This seems to be a problem... > > > How about releasing 2.8? How about taking the stuff > > to CVS now? > > To sucessfully use CVS as a system for *collaborative* work regarding Unix > VM, > there has to be *motivation* of the mostly important person here (Ian) to > work *together*! Linus doesn't use CVS for the Linux kernel source. The Linux/MIPS project takes the source releases from kernel.org and imports them as a vendor branch into CVS, and then does their work on that CVS server. When a new upstream kernel is released, they repeat the vendor branch merge. Then, Linux VR CVS treats the Linux/MIPS source as a vendor branch.... I guess what I'm saying is: you don't need the cooperation of the primary distributor of a package in order to do useful work on it with CVS. You can just send cvs diffs for the changes you've made to the author, and then she can merge them. This doesn't help if the author isn't interested in merging (your) outside changes. But then you're faced with the normal choice of "fork or suck it up". Jay

1 0

Re: [QUESTION?] Siren on windows NT?
by Helge Horch 02 Feb '01

02 Feb '01

At 11:16 02.02.2001 +0100, Hand Baveco wrote: > I downloaded the Siren package (2.8 image, changes and sources) and >installed it on NT. > It looks great, but the MIDI driver apparantly doesnot work when >called from Siren methods Grab <http://home.munich.netsurf.de/Helge.Horch/squeak/SirenSqueak-built.zip> (197 KB) and run the Siren 2.8 image with the VM therein, then file in the Fronkenshteen changeset. >Am I supposed to install an additional plug-in? Yes, if it existed (for win32). What my patch above does is to emulate the SirenMIDIPlugin with the builtin MIDIPlugin primitives. > From the documentation I understood that it should work with the standard >windows VM & primitives (using the latest 24/5/2000 VM). I think I read that too, but I guess the doc is wrong. ;-) BTW, you might be interested in the separate "Squeak-and-Siren-and-Music" mailing list, see <http://www.create.ucsb.edu/mailman/listinfo/squeakaudio> where I described the hows and whys in greater detail. HTH, Helge

1 0

Re: [updates] 85 new ones / Fast-track beta / Please test
by Henrik Gedenryd 02 Feb '01

02 Feb '01

Dan Ingalls wrote: > With sincere appreciation of all the points raised regarding 2.9ff > possibilities, we are declaring this to be a fast-track beta for Release 3.0, > targeted to freeze (in aobut a week) and ship on the CDROM for Mark and Kim's > forthcoming book on Squeak (*). Make sure to put up a prominent window on the startup desktop that somehow points to a net resource for updates to the 3.0 version on the CD. This way CD owners will easily be able to find the omissions that we will by natural law discover a couple of days after putting it in the mail. I suggest just a brief "click here for bug fixes" or so blurb with a link to an "Updating the 3.0 CD image" page on the Swiki. (Possibly also the explicit URL for typing into Netscape et al. as a baseline provision.) Then the actual description there. Possibly also a button that directly triggers "load updates from server". It may not be obvious how to find that command in the help (!) menu of the system menu. Henrik

1 0

Re: [pws] Swiki Server problems
by Bob Arning 02 Feb '01

02 Feb '01

On Thu, 1 Feb 2001 18:24:18 -0500 "Jochen F. Rick" <nadja(a)cc.gatech.edu> wrote: >Problem: > >All of a sudden, the Swiki no longer works and there is a debug window in >the Squeak window. It says 'computation has been terminated'. This >originates from MethodContext>>cannotReturn:. In the debug stack, the >previous message is ComancheServer>>value:. After abandoning the error, >stopping the server, and restarting, the server works again. > >Further Information: > >This is not happening on any of our Mac servers with a nearly identical >setup. Also, the 2.8VM for Solaris seems to be having problems. I cannot >for instance send the Squeak XWindow to another Solaris box. I can send >it to my Linux box (RH6.2), but keystrokes don't go through. Headless >mode doesn't work at all. Je77, FWIW, the only way I can recall this happening is while tinkering with the VM and screwing up some contexts. Cheers, Bob

1 0

RE: Sleeping in Squeak?
by Jim Beyer 01 Feb '01

01 Feb '01

Are you anywhere near Chicago? I just had an inquiry from a head hunter and I think he is looking over my level. FWIW jimB

1 0

[FIX] LayoutFix
by Jesse Welton 01 Feb '01

01 Feb '01

Changes LayoutFrames so that all offsets are properly treated as such. Changes SystemWindows to lay out their panes in a frame enclosing just the pane region and not the label area, eliminating odd offsets and some attendant display errors. -Jesse

1 0

Jump to page: