Squeak-dev May 2004

squeak-dev@lists.squeakfoundation.org

152 participants
288 discussions

[BUG] Package Browser horizontal scrollbar
by Anthony Hannan 29 May '04

29 May '04

With pop-up scrollbars and scrollbars on left, the horizontal scrollbar in the package-pane browser pops-up but is inaccessible because it disappears when you try to point to it; it is also transparent. Peculiarly, it works fine when pop-up scrollbars are on right. Cheers, Anthony __________________________________ Do you Yahoo!? Friends. Fun. Try the all-new Yahoo! Messenger. http://messenger.yahoo.com/

1 0

Trouble installing Chuck
by Stéphane Rollandin 28 May '04

28 May '04

I tried to install Chuck-Standard-ls.11.mcz both from SqueakMap and SqueakSource and got the following error message: Could not find version 'Chuck Type Inference-ls.12' help would be appreciated :) Stef

3 3

[BUG][VM][FIX]file truncate primitive broken on *nix ( [closed] Fixed in SF cvs sources )
by tim＠sumeru.stanford.edu 28 May '04

28 May '04

In the latest VM source download from SF this is fixed.

1 0

3.7 release in "a very few weeks" ???
by Dean_Swan＠Mitel.COM 28 May '04

28 May '04

Hi Tim, I do hope that the "3.7 Full Assembler" package on SM is going to be cleaned up some before any official 3.7 release. It would be nice if it finished up leaving the default project in a state that looks like earlier releases (i.e. with the 'Welcom to Squeak' and 'Readme' windows, 'Worlds of Squeak' project, etc.). -Dean Tim Rowledge <tim(a)sumeru.stanford.edu> Sent by: squeak-dev-bounces(a)lists.squeakfoundation.org 05/27/04 11:21 PM Please respond to The general-purpose Squeak developers list To: squeak-dev(a)lists.squeakfoundation.org cc: Subject: Re: [BUG] Pathological LargePositiveInteger instance discovered in the wild ( fix/workaround provided ) Stephan Rudlof <sr(a)evolgo.de> wrote: > > Seconded - and the sooner the better since we're close to a release > > How close? Hopefully a very few weeks. > > You could help me somewhat: > I have just installed the newest VMMaker from SqueakMap: does the actual > version work well? > Is it best to use VMMaker with the sources from SF? I remember a trial > to come away from SF... > So: is SF the right place just now? Right now, yes - get the latest from SF, the latest VMMaker37b4.sar and in a 5905 (I haven't tested later yet) it should most definitely work. Thanks for looking into this for us. tim -- Tim Rowledge, tim(a)sumeru.stanford.edu, http://sumeru.stanford.edu/tim Strange OpCodes: SCS: Spurious Cold Start

1 0

Clicks in audio recordings of standard image
by Brad Fuller 28 May '04

28 May '04

Hi All, I'm starting to explore the sound classes in Squeak (latest beta). I've noticed clicks when playing back the audio that was recorded using the supplied class SoundRecorder. I have not saved this sound to a file to look at it (to make sure that the clicks are not introduced in playback) but I thought I'd ask some questions anyway. First, has anyone else noticed this? (Note that I know it's not my hardware as I've used this with many professional audio software apps with no problems like this.) Secondly, if indeed these are introduced during recording, could another process be running (like garbage collection) to interrupt the buffer fill? This is the most common problem with introducing clicks, but does anyone know of any other possibilities in Squeak to look at? Thanks, brad _______________________________ Hear us online: www.Sonaural.com Brad Fuller (408) 799-6124

1 0

[BUG] Pathological LargePositiveInteger instance discovered in the wild
by David T. Lewis 28 May '04

28 May '04

A certain non-normalized instance of PositiveLargeInteger in the image crashes the VM with a segmentation fault. The instance of PositiveLargeInteger that I found cannot be directly inspected (VM crashes when inspecting it), but it has the following characteristics: badLargeInteger - 0 ==> 0 badLargeInteger + 0 ==> 0 badLargeInteger + 1 ==> 1 badLargeInteger - 1 ==> 4294967295 badLargeInteger - 2 ==> 4294967294 badLargeInteger - 1 + 1 ==> 4294967296 (badLargeInteger - 1 + 1) storeStringBase: 16 ==> '16r100000000' (badLargeInteger - 1 + 1) = 16r100000000 ==> true (badLargeInteger - 1 + 1) = (2 raisedTo: 32) ==> true badLargeInteger normalize ==> 0 badLargeInteger == > KABOOM badLargeInteger printString ==> KABOOM Working from a debugger, the closest I can get the the crash is: badLargeInteger digitDiv: 10 neg: false ==> KABOOM This is calling Integer>>digitDiv:neg: which calls primDigitDivNegative in the LargeIntegers plugin. Disabling the plugin prevents the VM crash, but badLargeInteger still cannot be inspected because the #digitDev:neg method fails. >From the debugger (with the plugin disabled) I can see that the four byte slots in badInteger are all zero (despite the '16r100000000' interpretation above), so I am inclined to think that this is an "impossible" instance of LargePositiveInteger, and fixing the plugin and/or #digitDiv:neg: method would just be masking the real problem somewhere else. Having said that, I have no idea where the underlying problem might lie, so I'm just posting this as a BUG report with no fix proposed. The problem can be reproduced as follows: Test system: Smalltalk vmVersion ==> 'Squeak3.7beta of ''1 April 2004'' [latest update: #5868]' Smalltalk osVersion ==> 'linux-gnu' - Open a new process browser. - Select the low space watcher process. - Explore context (right pane option). - In the explorer, select the semaphore. - Find objects point to the semaphore. - You should see a three element array. - Save your image. - Select element 1, an Association (no problem). - Select element 3, a Process (no problem). - Select element 2, a Array ==> KABOOM. - Restart the image, and from the inspector on the three-element array, open an inspector on element 2 (inspect "self at: 2"). - Save your image. - Inspect the elements of the array until you find the one that crashes the VM. This will be a LargePositiveInteger. It was element 33 in the array in my test. Here is the console (stderr) output when the VM crashes: lewis@dtlewis:~/squeak/squeak3.7> squeak -memory 10m BFAV Segmentation fault 1106979208 Integer>quo: 1106979116 LargePositiveInteger>quo: 1106979008 Integer>printOn:base: 1106978916 Number>printOn: 1106979848 [] in Object>printStringLimitedTo: 1106978824 SequenceableCollection class>streamContents:limitedTo: 1106978732 Object>printStringLimitedTo: 1106979572 [] in MethodContext>printDetails: 1106978596 BlockContext>on:do: 1106978412 BlockContext>ifError: 1106975848 MethodContext>printDetails: 1106963652 ContextPart>errorReportOn: 1106958108 SystemDictionary>logError:inContext:to: 1106744736 [] in Debugger class>openOn:context:label:contents:fullView: 1106744644 BlockContext>on:do: 1106744268 [] in Debugger class>openOn:context:label:contents:fullView: 1106744520 [] in BlockContext>newProcess Aborted

4 7

[BUG][FIX] sendersOfKeyBugFix-wiz (sm)
by Peace Jerome 28 May '04

28 May '04

Bugfix. Method was sending browseAllCallsOn: to Smalltalk which as a SysDic didn't understand it. SysNav default is the proper receiver." sendersOfSelectedKey "Create a browser on all senders of the selected key" | aKey | self selectionIndex = 0 ifTrue: [^ self changed: #flash]. ((aKey := keyArray at: selectionIndex) isKindOf: Symbol) ifFalse: [^ self changed: #flash]. object class == MethodDictionary ifTrue: [^ self changed: #flash]. SystemNavigation default browseAllCallsOn: aKey attached is gzip version of the change set. Yours in Service, Jerome Peace (wiz) __________________________________ Do you Yahoo!? Yahoo! Domains � Claim yours for only $14.70/year http://smallbusiness.promotions.yahoo.com/offer

2 1

[BUG][FIX] Integer>>printOn:base: fix prevents VM crash
by David T. Lewis 28 May '04

28 May '04

Inspecting the special objects array (Smalltalk specialObjectsArray) has the annoying side effect of crashing the VM. The offending element of the special objects array is an uninitialized instance of LargePositiveInteger intended for use as a template for fast large integer creation (but not actually used in the present Squeak image or VM). This change set corrects the problem with a small change to Integer>>printOn:base: Test case methods are provided to confirm reasonableness of the new #printOn:base: method. Related note: The VM crash occurs in the LargeIntegers plugin. If the plugin is disabled, other errors occur but are contained to image side problems that throw up a debugger. This means that some additional bounds checking in the LargeInteger plugin would be a good idea. This change set does not address that issue, but it does eliminate the symptoms and considerably reduces the likelihood of someone accidentally causing a crash during normal Squeak usage. This problem was originally reported as: "[BUG] Pathological LargePositiveInteger instance discovered in the wild" Dave

2 2

Re:Trouble installing Chuck
by Martin Wirblat 28 May '04

28 May '04

If I download per web interface from http://kilana.unibe.ch:8888/Chuck/ Chuck-Standard-ls.11.mcz or 13 or 15 I get strange packages. The source.st has 0 bytes. regards, Martin Bryce Kampjes <bryce(a)kampjes.demon.co.uk> wrote on 27.05.2004 21:06:11: > >Bryce Kampjes writes: > > Stéphane Rollandin writes: > > > > > > I tried to install Chuck-Standard-ls.11.mcz both from SqueakMap > > > and SqueakSource and got the following error message: > > > > > > Could not find version 'Chuck Type Inference-ls.12' > > > > > > help would be appreciated :) > > > > I tried and had a similar problem. If you install Monticello then > > get Chuck directly from SqueakSource it seems to work. It works > > well enough to get the AutoChuck preference which I've just > > selected and am still waiting for it to install. > > > > >From Monticello create a project called Chuck then with that > > >selected > > add an HTTP repository. The SqueakSource details are: > > > > MCHttpRepository > > location: 'http://kilana.unibe.ch:8888/Chuck' > > user: '' > > password: '' > >Hmm that doesn't quite work. Look's like some packages are missing, >there are a few DNUs. Oh, well, back to waiting for information from >the informed. > >Uninformed >Bryce

2 1

Re: [FIX][BUG] ExceptionTester>>simpleOuterTest fails
by Anthony Hannan 27 May '04

27 May '04

I should elaborate on the reason for the unwind loop implementation described in my previous email and repeated below: Anthony Hannan <ajhannan(a)yahoo.com> wrote: > In senders of #terminateTo:, like #return: and #restart, unwinding is done > while terminating. They find the next ensure:/ifCurtailed: (unwind) sender, > terminate to it, execute its unwind block, then repeat until the target home > context is found. It is important to terminate (pop contexts) to the ensure:/ifCurtail: context before executing its unwind block so the unwind block is executed within the correct dynamic scope (ie. exception handlers). If the unwind block is executed before terminating (or even worse, in another process) handlers created during the execution of the original block will still be there and catch errors raised by the unwind block. The intension is to unwind the stack first to the point of the ensure:/ifCurtail: then execute its unwind block. This is why #return: terminates a single chunk at a time instead of all at once in the end. Cheers, Anthony __________________________________ Do you Yahoo!? Friends. Fun. Try the all-new Yahoo! Messenger. http://messenger.yahoo.com/

1 0

← Newer
1
...
4
5
6
7
8
9
10
...
29
Older →

Jump to page: