Squeak-dev April 2020

squeak-dev@lists.squeakfoundation.org

39 participants
235 discussions

efficient DateAndTime
by Chris Muller 01 May '20

01 May '20

I have a bunch of two-element Array log entries, where the first element is the timestamp, the second is the object being logged. But these entries may never be consumed, so I wish to minimize their cost. So, instead of storing the timestamps as DateAndTimes, I wanted to keep them as SmallIntegers. I thought I could use Time utcMicroseconds, now a SmallInteger in 64-bit Squeak. However, when I went to instantiate the DateAndTime lazily from that number it's said it's year 2089... DateAndTime utcMicroseconds: Time utcMicrosecondClock offset: 0 " 2089-04-29T00:49:26.3326+00:00" I realized this is because a different epoch being used (1901 vs. 1971) between the two. This seems inconsistent and confusing, but that's less important to me at the moment than the efficiency I'm looking for. Is there another way? - Chris

9 13

Package Universes?
by Russell Allen 01 May '20

01 May '20

5 8

Hackathon team - squeak on hashgraph?
by keithy＠consultant.com 30 Apr '20

30 Apr '20

1 0

The Trunk: Morphic-fn.1654.mcz
by commits＠source.squeak.org 30 Apr '20

30 Apr '20

Fabio Niephaus uploaded a new version of Morphic to project The Trunk: http://source.squeak.org/trunk/Morphic-fn.1654.mcz ==================== Summary ==================== Name: Morphic-fn.1654 Author: fn Time: 30 April 2020, 3:27:32.337614 pm UUID: 7c5e8137-2bf8-4480-8713-905cd3640a4d Ancestors: Morphic-mt.1653 Inject `style` variable via #ifNotNil:. =============== Diff against Morphic-mt.1653 =============== Item was changed: ----- Method: ImageMorph>>drawOn: (in category 'drawing') ----- drawOn: aCanvas + self borderStyle ifNotNil:[:style | - | style | - (style := self borderStyle) ifNotNil:[ style frameRectangle: bounds on: aCanvas. ]. self isOpaque ifTrue:[aCanvas drawImage: image at: self innerBounds origin] ifFalse:[aCanvas translucentImage: image at: self innerBounds origin]!

1 0

AllocationTest >> #testOutOfMemorySignal
by Marcel Taeumel 30 Apr '20

30 Apr '20

Hi all! The VM 202004292351 (cog.spur 64-bit) triggers an interesting system crash on my Surface Pro 6 while having a Skype video call. At that time, the VM process uses about 11-12 GiB of memory. I have two questions: 1. Why does primitive 254 fail when setting VM parameter 67 in 32-bit to 1688449936? It works on 64-bit. 1.6 GiB memory is not that much. 2. Is it really necessary to stress the system that much in AllocationTest >> #testOutOfMemorySignal ? It tries to allocate 1 terabyte of memory. I know that the system crash is not the VM's fault. But it is mildly annoying. ;-) Best, Marcel P.S.: FYI, the system crashes with a blue screen because VIDEO_SCHEDULER_INTERNAL_ERROR

2 1

The Trunk: KernelTests-mt.380.mcz
by commits＠source.squeak.org 30 Apr '20

30 Apr '20

Marcel Taeumel uploaded a new version of KernelTests to project The Trunk: http://source.squeak.org/trunk/KernelTests-mt.380.mcz ==================== Summary ==================== Name: KernelTests-mt.380 Author: mt Time: 30 April 2020, 12:22:16.766652 pm UUID: c45fab34-ced0-9340-9f23-8f6af2c21f88 Ancestors: KernelTests-mt.379 Split up memory allocation tests. =============== Diff against KernelTests-mt.379 =============== Item was changed: ----- Method: AllocationTest>>testOutOfMemorySignal (in category 'tests') ----- testOutOfMemorySignal + "Ensure that OutOfMemory is signaled eventually. Restrain the available memory first to not stress the machine too much." + - "Ensure that OutOfMemory is signaled eventually" | sz | self setFreeSpaceLimitOf: 1024 * 1024 * 1024 * (Smalltalk wordSize = 8 ifTrue: [4] ifFalse: [1.5]) around: [sz := 512*1024*1024. "work around the 1GB alloc bug" + self should: [(1 to: 2000) collect: [:i| Array new: sz]] raise: OutOfMemory].! - self should: [(1 to: 2000) collect: [:i| Array new: sz]] raise: OutOfMemory]. - - "Call me when this test fails, I want your machine." - "Current (2017) Spur VMs fail new: & basicNew: with #'bad argument' if given other than a non-negative SmallInteger." - sz := 1024*1024*1024*1024. - self should: [Array new: sz] - raise: OutOfMemory, Error - withExceptionDo: - [:ex| - ex class == Error ifTrue: - [self assert: [ex messageText includesSubstring: 'basicNew: with invalid argument']]]! Item was added: + ----- Method: AllocationTest>>testOutOfMemorySignalExtreme (in category 'tests') ----- + testOutOfMemorySignalExtreme + "Try to allocate a ridiculous amount of memory and check whether the expected error is signaled. Call Eliot when this test fails, he want your machine. :-) + + Note that current (2017) Spur VMs fail in #new: and #basicNew: with #'bad argument' if given other than a non-negative SmallInteger. + + Also note that this test can be quite stressful to your machine depending on how your operating system allocates the required memory behind the curtains. Better not triggering some robot fetching a tape from somewhere..." + + | sz | + sz := 1024*1024*1024*1024. "= 1 TiB" + self should: [Array new: sz] + raise: OutOfMemory, Error + withExceptionDo: + [:ex| + ex class == Error ifTrue: + [self assert: [ex messageText includesSubstring: 'basicNew: with invalid argument']]]!

1 0

My social distancing project - MCConfiguration maps with versioning, comments, and author timestamps.
by David T. Lewis 29 Apr '20

29 Apr '20

An MC update stream, such as the Squeak trunk update stream, relies on a sequence of update maps stored in a repository. The update maps (MCM files) do not retain information about their authors or the time and reason for committing the maps, and they do not retain change information if an existing update map is modified, as occasionally may be required to correct a problem in the update stream. I have found a way to extend the MCM configurations such that they can be stored in a format that saves the additional version information, and that is still readable by any earlier Squeak image. Updated images can read and understand the version information, and will store new MCM updates with comments, author initials, time stamps, and version history. I have a working version of this in my own image. It seems to work quite well, although I think I need to spend some more time using it before suggesting it for general use. If there is in interest, I can put a copy in the inbox. Dave

4 11

Re: [squeak-dev] Multi-Image Sista VM?
by Robert 29 Apr '20

29 Apr '20

I was thinking they’d Run Concurrently, of course. #SVS!!! K, r Sent from ProtonMail Mobile On Wed, Apr 29, 2020 at 14:43, Robert <robert.withers(a)pm.me> wrote: > I have not read about this anywhere I can remember. Would it be possible to run two images on one VM? > > Preggo, > > Robertotini > > Sent from ProtonMail Mobile

1 0

The Trunk: PreferenceBrowser-mt.103.mcz
by commits＠source.squeak.org 29 Apr '20

29 Apr '20

Marcel Taeumel uploaded a new version of PreferenceBrowser to project The Trunk: http://source.squeak.org/trunk/PreferenceBrowser-mt.103.mcz ==================== Summary ==================== Name: PreferenceBrowser-mt.103 Author: mt Time: 29 April 2020, 8:45:37.770261 pm UUID: b939aeb0-0229-c14b-94fe-2fd07189ecb8 Ancestors: PreferenceBrowser-mt.102 Fixes a bug with some redundant tooltip ... because that quick-and-dirty way to find existing preferences did not work out. Note that the preference wizard can provide its own namespace for preferences with custom state/getter and setter messages. Luckily, much of the wizard's selection matches something from the system-wide Preferences. Look at the wizard as something that provides a nice first contact to set up important stuff in the environment. That "stuff" is not necessarily present in Preferences but maybe also in some specific class such as for pragma preferences. So the wizard acts as a GoF Facade. =============== Diff against PreferenceBrowser-mt.102 =============== Item was changed: ----- Method: PreferenceWizardMorph>>initializePage04InteractionMore (in category 'initialization - pages') ----- initializePage04InteractionMore | currentPage pane | currentPage := pages add: self createPage. pane := self createScrollPane. currentPage addMorphBack: (self createLabel: 'Choose more interaction settings' color: Color white). currentPage addMorphBack: pane. pane scroller firstSubmorph addAllMorphsBack: { self createCheckbox: 'Windows raise on click' translated for: #WindowsRaiseOnClick help: #(window raise). self createCheckbox: 'Windows always active' for: #WindowsAlwaysActive help: #(window content active). self createCheckbox: 'Window buttons always active' translated for: #WindowButtonsAlwaysActive help: #(window control active). self createVerticalSpace. self createCheckbox: 'Smart horizontal splitters' translated for: #SmartHorizontalSplitters help: #(horizontal splitter). self createCheckbox: 'Smart vertical splitters' translated for: #SmartVerticalSplitters help: #(vertical splitter). self createVerticalSpace. + self createCheckbox: 'Filterable lists and trees' translated for: #FilterableLists help: 'Type something to filter lists. Press [return] to clear the filter. Press [backspace] to clear and restore prior selection.' translated. - self createCheckbox: 'Filterable lists and trees' translated for: #FilterableLists help: #(filterable). self createCheckbox: 'Filters clear if unfocused' translated for: #ClearFilterAutomatically help: #(filter clear). self createVerticalSpace. self createCheckbox: 'Attach tools to mouse' translated for: #AttachToolsToMouse help: #(tools attach). }.!

1 0

Re: [squeak-dev] [Vm-dev] How to create headless VM without some primitives?
by Robert 29 Apr '20

29 Apr '20

It’s Right in the middle! Klaarrung? Sent from ProtonMail Mobile On Wed, Apr 29, 2020 at 14:36, Robert <robert.withers(a)pm.me> wrote: > Are you saying that I can can run my 64-bit Squeak.image running PTAPAS [1] on the Pharo headless VM?!?!?! Now that, my Ladies and Gentlemen, would constitute a profound shout-out to the Heritage [2]. Hey, goddammit, don’t you forget your Heritage!!! You gained a profound VM for Pharo [3]! Do you feel me? > > [1] PTAPAS - https://squeak.slack.com/archives/C02LC8AMS/p1587255601170900 > [2] Squeak - https://Squeak.org > [3] Cog Spur 64-bit Sista Multi, Malto, hi to hair, dost PI wisson calculiert, Schnell, Vielen Schnell!!! > > Vas ist deinem mackht Los Corison Stein, meinen Schatchzie? Do you feel me, now? > > ^,^ > > Sent from ProtonMail Mobile > > On Wed, Apr 29, 2020 at 13:57, Ben Coman <btc(a)openinworld.com> wrote: > >> On Tue, 14 Apr 2020 at 10:43, tim Rowledge <tim(a)rowledge.org> wrote: >> >>> >>> a) it's way past time we stopped opening a windows when it isn't needed. No window should be created until and unless the system actually writes to the display and calls for a display update. This would enable image-controlled headlessness. RISC OS was able to cope with it 25 years ago. >> >> Pharo is doing that, in their branch https://github.com/pharo-project/opensmalltalk-vm/tree/headless >> which IIUC is the branch being used to build all Pharo VM going forward. Once they've worked the bugs out of it, >> it might be worth pulling that back to evaluate for Squeak to use. >> >> cheers -ben

1 0

← Newer
1
...
4
5
6
7
8
9
10
...
24
Older →

Jump to page: