Squeak-dev December 2008

squeak-dev@lists.squeakfoundation.org

91 participants
114 discussions

Cog VM status update
by Andreas Raab 21 Dec '08

21 Dec '08

Folks - I just read Eliot's most recent blog post about the Cog VM and it reminded me how difficult it must be for others to see where this project stands. So here is a bit of an update on the current status: As you may recall we started this project in spring this year by hiring Eliot for the express purpose of building us a new VM that would speed up execution of our products. We decided to structure the work into stages at the end of each there would be a tangible deliverable i.e., a new VM that could be run and benchmarked. The first stage in this process is what we call the "Closure VM". It is nothing more (and nothing less) than a Squeak VM with closures and the required support (compiler, decompiler, debugger etc). Given past experience, we had originally expected this stage to cost us some speed (up to 20% were estimated) since closure support has a cost which at that stage is hard to offset with other improvements. However, thanks to a truly ingenious bit of engineering done by Eliot in the design for the closure implementation the resulting speed difference was negligible. Since there was no speed penalty we decided to jump ship earlier than we originally anticipated and the Closure VM has been the regular shipping VM with Qwaq products since September this year. The second stage in the process is the "Stack VM". It is a Closure VM that executes on the native stack and transparently maps contexts from and to stack frames as required. The VM itself is still an interpreter so any speed improvements come purely from the more efficient organization of the stack layout (no allocations, overlapping frames etc). For those of you having been around for long enough it is equivalent to what Anthony Hanan did a few years ago, except that it hides the existence of the native stack entirely and gives the programmer the naive view of just dealing with linked frames (contexts). The original expectations for the resulting speedups by Eliot were a little higher than we've seen in practice, but are in line with the results that Anthony got: approx. 30% improvements across the board in macro benchmarks. The work on the Stack VM was completed last month, we are currently rolling it out internally and the next product release will ship with the Stack VM. The third stage which has just begun is what we call the "Simple JIT VM" (well, really it doesn't have a name yet, I just made it up ;-) Its focus is send performance (as we see send performance as the single biggest current bottleneck). It will sport a very simple JIT w/ inline caches with the idea being to bring up send performance to the point where it's no longer the single biggest bottleneck, then measure performance again and figure out what the next best target is. I am not going to speculate on performance (we have been wrong every single step of the way ;-) but both Eliot and I do think that we'll see some nice improvements in application performance here. The fourth stage is a bit more speculation at this point because the concrete direction depends on what the results of stage 3 really show the new bottleneck to be. We have various candidates lined up: Very high on the list is a delayed code generator which can dramatically improve the code quality. Next to it are changes in the object format moving to a unified 32/64bit header model which would dramatically simplify some tests for inline caching and primitives etc. However, since this work is driven by product performance, it is possible (albeit unlikely at this point) that the focus might shift towards FFI speed or float inlining. There is no shortage of possible directions, the main issue will be to figure out what the bottlenecks at that point are and how to address them most efficiently. Stage four won't be the end of it, but from where we are this is how far we've planned at this point. And if you want to know all the gory details about the stuff that Eliot's working on, please do check out his blog at: http://cogblog.mirandabanda.org/ Cheers, - Andreas

7 9

How to make a changeset to remove a method.
by Jerome Peace 20 Dec '08

20 Dec '08

How to make a changeset to remove a method. Hi Tim, This is to answer a question Tim Johnson posed in another thread. I think the thread was on squeak-dev so I post it here. Basicly,removing a method includes creating a changeset record. So all you have to do is fileout the current changeset in the project in which you removed the method. Below is a recipe for making a new changeset with just the removal in it. To make a changeset to remove a method: Open a fresh morphic project. Open a browser on the method you want to remove. Remove it. (the current change set now has the removal change record in it.) Open a dual change sorter. ( the dualness is not important here. I have just never has occasion to open a single change sorter.) Rename the changeset to something meaningful. TextAnchorFix-tjp for example. Write a preamble for the changeset explaining what, why and how. Save changeset. You can now fileIn or install the removal change. You can also use the dualChange sorter to combine the removal with other changeset changes. I kind of remember this not being obvious to me the first time I came across the need to report a removal as a fix. hth, Yours in curiosity and service, --Jerome Peace

1 0

Re: 3.11 artifacts observations and a little bit of curiosity.
by Jerome Peace 20 Dec '08

20 Dec '08

[squeak-dev] 3.11 artifacts observations and a little bit of curiosity. *** >Keith Hodges keith_hodges at yahoo.co.uk >Thu Dec 18 22:07:26 UTC 2008 > > > >Chris Cunningham wrote: >> So, why not call it 3.11bc or 3.11.0bc? You do definitely have the >> authority to use that name. >> >> -Chris >Because it is an image derived from 3.10 for building stuff that works >in 3.10. The packages definitions/universe etc that you would expect to >work in this image are the 3.10 ones, not the 3.11 ones. Um. Are you sure that's the right way to think about it? 3.11 the beginning does not have to resemble 3.11 the final product. Almost all previous releases started where their predecessors left off. The first 3.9 alpha was exactly 3.8 final with just the version number changed. The final 3.9 release was something quite different than its beginning. I wondered the same thing as Chris. Why not call what you are doing both 3.11 alpha (the beginning) and when permission comes 3.10.3 (the maintenence collection of fixes from the 3.11 change stream? Or in your case the installer scripts including the 3.11 fixes that are useful for maintaining 3.10.x AFAICT that is what versioning number distinctions are all about. Also if you don't have permission to name something a 3.10.3 then we should get "someone in authority" into the loop Either we get permission or we get another release team that can have permission to fix the bugs 3.10.2 left broken. In practice, I don't really care about the numbering. I care about bug reduction and I care about repairs getting in to a distributable image. I care that names should not conflict. 3.10.2 should be and only be associated with the official squeak release. An official distribution will gain a reputation. That reputation should not be diluted by any similarly sounding but not similarly acting alternative. Thank you for considering these concerns. Yours in curiosity and service, --Jerome Peace

1 0

Re: [squeak-dev] [ANN] Maui 1.1
by Edgar J. De Cleene 20 Dec '08

20 Dec '08

On 19/12/2008, at 21:35, Chris Muller wrote: I am pleased to announce the latest and greatest release of Maui, the "naked objects" Morphic-based UI builder for Squeak 3.9, and all-new *detailed documentation*; more than 50 pages worth with lots of screenshots.. http://wiki.squeak.org/squeak/3836 Enjoy! Chris Chris: I have Maui into SqueakLightII without major troubles. Still reading the extense docs, but this could revamp Morphic. I don't have enough English to praise this kind of work, so I say BRAVO !!! I try to give time to this for feedback . Edgar

1 0

Some thought on Pharo fork...
by Hilaire Fernandes 20 Dec '08

20 Dec '08

http://blog.ofset.org/hilaire/index.php?post/2008/12/12/Pharo-birth-of-a-fr…

1 0

Algernon + OB Package Browser
by Mariano Martinez Peck 19 Dec '08

19 Dec '08

Hi everybody! can someone tell how can I tell Algernon to use my default class browser (OB Package Browser) ? It is using system browser when I actually have OB Package Browser as default. Is this OK ? should Algernon use my default class browser ? Any help? thanks, Mariano

6 9

Mantis fixes
by Keith Hodges 19 Dec '08

19 Dec '08

Could we change the message on Mantis to request that fixes be prefixed with "M1234-" rather than suffixed? Keith

3 5

Re: [ANN] ConflictFinder (help needed!)
by Stephan Eggermont 19 Dec '08

19 Dec '08

>> >> From: Andreas Raab <andreas.raab(a)gmx.de> > > HTTPSocket httpFileIn: 'installer.pbwiki.com/f/LPF.st'. > Installer mantis bug: 6980 fix: > 'SystemDictionary-recreateSpecialObjectsArray-M6980.st'. > FileStream readOnlyFileNamed: 'ConflictFinder.st'. >>> In which images is this supposed to work? (doesn't work in the pharo one-click 10074, or the Seaside-2.8-573) Stephan

2 1

Squeak + SqueakDBX + FFI + valgrind + segmentation fault
by Mariano Martinez Peck 19 Dec '08

19 Dec '08

Hi people: I think I am having a bug in SqueakDBX with mysql backend. SqueakDBX uses FFI to talk to openDBX C library. Sometimes (totally random), after running several times (4 or 5) our 80 tests, I got a segmentation fault. Always in the same function but in different calls. I mean, the same function is called many times (at least one per test). I don't know what the problem is and it is very difficult for me to find the cause of the problem. Because of this, I am trying to use valgrind and I have never use it before. In all of the valgrind examples I see in internet it uses the C program directly, for example: valgrind --leak-check=yes myprog arg1 arg2 But I do it from Squeak using FFI, so, I have no idea where to put the valgrind :( Sorry but I don't know too much about C. Thanks a lot! Mariano

1 0

3.11 artifacts observations and a little bit of curiosity.
by Jerome Peace 18 Dec '08

18 Dec '08

3.11 artifacts observations and a little bit of curiosity. Hi Keith, Yea for the appearence of 081214 and 081216 build candidates. Thank you. I am still playing with it. There are some early observations and questions. Why is it a 3.10.2 bc and not a 3.10.3? Ken finalized the former and it should probably stay that way. So people using the new know its new. I will also note that the long redundant naming gives my poor OS9.1 mac the fits when I try to unzip it. A small but useful (to me) favor would be to pick a shorter way of naming. I vote +1 on keeping a date relevant name. yymmvv naming might serve well as it does for Damien's releases. My curiosity notices that these images weigh in at 13.7 and 14 MB down from the 7179 images 17M. and the zips are about 7M compared to 8.8M for 7179. Puck says: "Shazam| How'd he do that? Maybe we can get loading and unpacking squeak to actually give you back space on your hard drive." My curiosity wants to know too. Anything we should know about been left out? Anyway, I want to acknowledge your spirit and courage for responding to the challenge with such little delay. Puck again: "Shazam! How did he do that?" Curiosity: "I wonder if he got Bob the builder working?" Good work. Good work. Good work. Yours in curiosity and service, --Jerome Peace

3 3

← Newer
1
2
3
4
5
6
7
8
...
12
Older →