Squeak-dev August 2000

squeak-dev@lists.squeakfoundation.org

130 participants
418 discussions

Re: [Q] Also looking for a TimeSlicer class (Squeak)
by Lex Spoon 01 Aug '00

01 Aug '00

Please note that we really don't want *all* processes to get preempted. It is convenient most of the time that you don't get preempted by processes at the same priority, because it usually allows you to do without locks. Processes should have to register if they want to be shuffled by the shuffler. By the way, what priority should the processes get toggled to? Perhaps the register method should have a parameter for the desired lower priority. One final note: multiple schedulars running in the same image is bound to cause trouble, and yet, but is there any one schedular that will make everyone happy? Why do people want preemption at all, by the way? -Lex Tim Rowledge <tim(a)sumeru.stanford.edu> wrote: > The canonical example of this is simply a high priority loop waiting on > a Delay. Each time the Delay fires it will stir up the process list at > the priority of the suspended process (and by extension any other > priority level that might get involved) assuming of course that the > suspend is implemented to add the suspended process at the back of the > list. Not all systems actually do that - VW used to put it at the front > of thelist and so the same process would get resumed. Might still be > like that for all I know. > > I think there is all the VM support you need already there, what with > the Delay etc. Note that checkForInterrupts is already involved since it > is the mechanism by which the Delay termination is noticed! > > I imagine that it might be useful to add some VM support to directly > deal with a special timer for process stirring and thereby reduce by a > small amount the allocating and setting up of actual Delay objects, but > it would be a pretty minimal improvement I suspect. > tim > > -- > Tim Rowledge, tim(a)sumeru.stanford.edu, http://sumeru.stanford.edu/tim > Never write software that anthropomorphizes the machine.

2 1

Re: Compiler, was Re: Block closures
by Roel Wuyts 01 Aug '00

01 Aug '00

[snip] >> If you have comments at this eary stage, welcome. My hope is that it will >> make it a bit "simplified and clarified" and more flexible. > > Speaking of "comments", the handling of comments in code is another thing > that needs to get fixed in the present comiler. It would greatly expand > the set of interesting parsing, viewing, and translating projects if the > underlying parse tree retained comments properly. > > There isn't a clear rule about where in the parse tree a given comment > should live. While pretty printing frequently makes the code look better, > it almost always undermines the effectiveness of the comments. Indeed, the 'comment-handling' is currently really bad. I would like to easily rewrite a parsetree (say, optimize it, refactor it, do Java-like before/after stuff), keeping the comments. It is currently possible, but not clean at all. Why not have a 'CommentNode', that keeps the position of the comment in the source. That way we can 'pull up' the comments from where they are living now. Also note that, depending on the builder you use when parsing, comments could be enabled or not. -- Roel Wuyts Programming Technology Lab rwuyts(a)vub.ac.be Vrije Universiteit Brussel http://prog.vub.ac.be/~rwuyts Webmaster of European Smalltalk User Group: www.esug.org

1 0

[BUG][FIX] (partial fix) Pane/Window resizing doesn't work right with touch screens
by Ned Konz 01 Aug '00

01 Aug '00

I've been trying to make Squeak work better on my touch-screen (pen) based WinCE device. A particularly annoying problem is that it's almost impossible to get out of window/pane resizing mode once you get in. The reason is that no mouse events occur between mouse up and mouse down. I fixed the pane resizing problems in MVC with the enclosed method change. But the Morphic resizing problem is trickier. The NewHandleMorph that is created to do the resizing does, in fact, delete itself when the mouse (pen) up occurs. However, the SystemWindow then notices that the mouse is still over the frame border and creates another NewHandleMorph. Now, since the pen is up and there aren't any mouse events while the pen is up, the NewHandleMorph stays around waiting for a click. When you finally put the pen down, it starts tracking the mouse movements. The whole process repeats. The only way you can get out of this mode is to move the pen into another window and tap inside it. The same thing happens, I think, with window border resizing (but I'm using the WindowFrame stuff so I can't tell easily). -- Ned Konz currently: Stanwood, WA email: ned(a)bike-nomad.com homepage: http://bike-nomad.com

1 0

Re: [ENH] Alternative EventSensor
by Lex Spoon 01 Aug '00

01 Aug '00

Tim Rowledge <tim(a)sumeru.stanford.edu> wrote: > > I like the concept of time stamps. > So do I, unless it costs a lot - although keystroke and mouse button > events are usually relatively slow (ie maybe a couple of hundred aminute > at worst) some events (people have been considering serial or > socket events for example) might be much higher bandwidth and could > cause a problem if getting the time is costly. Something to keep an eye > on; some OSs provide timestamps on event, some don't. Some have > expensive clock() calls, some are cheap. We have to check and make sure > we don't make life good on one platform by ruining it on another, just > like always. > > Hmm, I just put these two ideas together. On Unix, the timestamps for X events are going to be based on a completely separate clock from the ones for sockets. I guess the VM can just stamp the times itself, and ignore whatever X says, but something is definately lost: the input events will be less precise! The alternatives on Unix seem to be: 1. Only use *input events*, and allow for precise timings. 2. Mix all events into the same queue, but use imprecise timings. -Lex

3 2

[BUG] Symbol>>hash
by Andres Valloud 01 Aug '00

01 Aug '00

Hi. Now, 'To: ' = #'To: ', but their hashes don't match. Removing Symbol>>hash and rehashing all sets seems to work, but... is that ok? Symbol>>hash involved a primitive, and I think that if the primitive was there, it was because speed was important. Or maybe it was a leftover? Andres.

2 1

Automatic checkpointing, OOZE (Was Re: final(?) version of Object Engine Tour chapter online)
by Mark A. Schwenk 01 Aug '00

01 Aug '00

Alan Kay wrote: > > Tim -- > > Squeak Listers will perhaps be interested to know that the demo image > -- "The Flagship" -- is about 160MB and is indeed well over a 100 > projects. This runs in a much smaller footprint because I have been > using the segment swapping for the last 8 months or so. This works > perfectly and both the inswap and outswaps take place in the > transition from one project to another. No artifacts of the swapping > have ever been noticed by me. > > (A delicious thought for the future: If we had a very small image, > distributed code better, etc., then it is really possible to run with > just one or two projects in memory at a time with a footprint that is > just a little larger than the largest project. The small "extra > image" would allow us to do automatic checkpointing periodically with > little penalty, and the result would be a very crash proof system > with similar characteristics to the famous OOZE swapper done by Ted > and Dan at PARC (except it wouldn't ooze)). > > Cheers, > > Alan > I'm very interested in understanding how we can get automatic checkpointing in Squeak as well as learning more about the famous OOZE swapper. Do you have any references where I could learn more? -Mark Schwenk WellThot Inc.

1 0

new WinCE VM (2.8 or 2.9alpha) available?
by Ned Konz 01 Aug '00

01 Aug '00

I have a 2.7 WinCE VM. Is there a newer one anywhere? Is anyone working on this? -- Ned Konz currently: Stanwood, WA email: ned(a)bike-nomad.com homepage: http://bike-nomad.com

1 0

Re: [ENH] Alternative EventSensor
by Richard A. O'Keefe 01 Aug '00

01 Aug '00

Lex Spoon wrote: Amen. On my computer, and most of the ones I've seen, that key is labelled "Backspace". Perhaps sun keyboards do it differently? The Sun model 3 keyboard (nicest keyboard I've ever used; I've never forgiven Sun for dropping it) has a fat "Delete" key immediately above the "Return" key. Above that is the back-quote/tilde key, and at the very top above that is a "Backspace" key the same size as the character keys. From bottom to top, the right edge of the main keybank is [Alternate,Linefeed,Return,Delete,`~,Backspace]. The Sun model 4 keyboard looks pretty much like a Mac keyboard these days, except that the Mac has a fat key labelled Delete and the Sun 4 has a fat key labelled Backspace. From bottom to top, the right edge is [Alt Graph,Shift,Enter,\|,Backspace] on the Sun 4, [Control,Shift,Return,\|,Delete] on the Mac.

1 0

Jump to page: