- Pharo-users - lists.pharo.org

Re: [Pharo-users] VM crash on Ubuntu 14.04

by Johan Fabry

OK, back with access to the Ubuntu machine, I reproduced the crash when running under gdb (reproducing it is never an issue, finding the minimal case is the issue). See below for gdb output. But my vm has no debug information compiled, so it is not really useful. Can I download a vm with debug info compiled from somewhere? I would prefer to avoid having to download the sources and compile et cetera â¦ Once I have such a VM I will make some time to further investigate, and report results. Program received signal SIGSEGV, Segmentation fault. 0xb8c2bc31 in ?? () (gdb) bt #0 0xb8c2bc31 in ?? () #1 0x080fc9f2 in primitiveNativeCall () #2 0xb4fba770 in ?? () #3 0xb503b05e in ?? () #4 0xb503afcb in ?? () #5 0xb5011abd in ?? () #6 0xb4fba5c0 in ?? () (gdb) > On Oct 27, 2015, at 19:09, Nicolai Hess <nicolaihess(a)gmail.com> wrote: > > > > 2015-10-26 15:03 GMT+01:00 Johan Fabry <jfabry(a)dcc.uchile.cl <mailto:jfabry@dcc.uchile.cl>>: > Hi Nicolai, > > thanks for following up on this. > > I am in the process of determining what causes the crash. I fear that to reproduce the problem you will need to have ROS installed, I have not yet been able to reproduce it without ROS running. I do not want to inflict the installation of ROS on you without being 100% sure. > > I am using Roassal because this is for Live Robot Programming (http://pleiad.cl/LRP <http://pleiad.cl/LRP>) which uses a dynamic visualization of the running code. To reproduce the bug (until now) I need to develop code in LRP when using ROS, and then it happens randomly. I realize that this is too much to ask for an outsider to reproduce the bug (I did not expect anybody to download and install PhaROS to try to reproduce, thanks for that!!). > > Can you inform me how I can get you more useful debug info to you? I will keep trying to find minimal steps to reproduce in the mean time. > > No, the best we can hope for is a crash when it is running from debugger, you can start the vm with a debugger or > attach the debugger to a running process. If we are lucky, the debugger will show a stacktrace. > > >> On Oct 24, 2015, at 19:45, Nicolai Hess <nicolaihess(a)gmail.com <mailto:nicolaihess@gmail.com>> wrote: >> >> Can you provide the image file. >> What exactly runs in the image (does it depend on ROS or >> does it crash even without it). For what do you >> use Roassal in this environment. >> Can you reproduce the crash with a clean image (and maybe loading >> some package - without all of the ROS-stuff? >> (I just followed the installation guide for PhaROS, but I have no idea >> how to go from here to do anything to reproduce the crash). >> >> >> >> 2015-10-24 21:32 GMT+02:00 Johan Fabry <jfabry(a)dcc.uchile.cl <mailto:jfabry@dcc.uchile.cl>>: >> >> Ben, thanks for replying but since no VM guys are following up on this, apparently, it does not make much sense for me to go this way. >> >> Guys, am I to assume there is not help for me on this topic? >> >> > On Oct 23, 2015, at 11:03, Ben Coman <btc(a)openInWorld.com <mailto:btc@openInWorld.com>> wrote: >> > >> > On Thu, Oct 22, 2015 at 2:25 AM, Johan Fabry <jfabry(a)dcc.uchile.cl <mailto:jfabry@dcc.uchile.cl>> wrote: >> >> Hi all, >> >> >> >> I am having some unpleasant experiences on Ubuntu 14.04 (experimenting with Live Robot Programming on PhaROS). The VM crashes after a random amount of time (< 25 minutes), with no clear sequence of steps on how to reproduce the crash. I am using Pharo 4 (Pharo 40622 as installed by PhaROS) + Roassal which required an install of libcairo2:i386 . VM version details below. >> >> >> >> There is no printout on stdout, and the debug.log does not contain recent entries (timestamp is always some minutes before the actual crash happens). >> >> >> >> Is this a known problem? If so, how can I fix it, and if not how can I provide more debugging info? >> >> >> >> Thanks in advance! >> > >> > I'm not much help since I haven't done these myself, but just some >> > ideas (maybe someone can add more details) >> > * compile and run a debug vm >> > * trace all function calls, http://tinyurl.com/gdb-trace-all <http://tinyurl.com/gdb-trace-all> >> > >> >> >> >> ---> Save our in-boxes! http://emailcharter.org <http://emailcharter.org/> <--- >> >> Johan Fabry - http://pleiad.cl/~jfabry <http://pleiad.cl/~jfabry> >> PLEIAD and RyCh labs - Computer Science Department (DCC) - University of Chile >> >> >> > > > > ---> Save our in-boxes! http://emailcharter.org <http://emailcharter.org/> <--- > > Johan Fabry - http://pleiad.cl/~jfabry <http://pleiad.cl/~jfabry> > PLEIAD and RyCh labs - Computer Science Department (DCC) - University of Chile ---> Save our in-boxes! http://emailcharter.org <--- Johan Fabry - http://pleiad.cl/~jfabry PLEIAD and RyCh labs - Computer Science Department (DCC) - University of Chile

Oct. 29, 2015

Re: [Pharo-users] What is your preferred way for storing ByteArray's as text?

by Sven Van Caekenberghe

> On 29 Oct 2015, at 17:02, Esteban A. Maringolo <emaringolo(a)gmail.com> wrote: > > 2015-10-29 12:44 GMT-03:00 Sven Van Caekenberghe <sven(a)stfx.eu>: >> >> >>> On 29 Oct 2015, at 16:35, Esteban A. Maringolo <emaringolo(a)gmail.com> wrote: >>> >>> Hello all, >>> >>> As some already know I'm working in the new GLORP port from VW to Pharo. >>> >>> And as part of minor modifications I'm making to the code, I'd like to know what is your preferred way of storing ByteArrays as text representation/CLOB, etc... >>> >>> Option A) The printString of ByteArray, it is #[23 12 253 ...] >>> Option B) A Base64 encoded version of the ByteArray >>> Option C) Some other encoding >> >> Does that not dependent, at least in part, on the actual underlying SQL DB ? > > Yes it does, but certain backends (like SQLite) doesn't provide BLOB > storage, and that's why I'm talking about CLOB (character instead of > byte based) or text representations for it. I'd like to have a lowest > common denominator to use as a fallback. > >> I think HEX and Base64 are make the most sense. > > Is Base64 more compact than Hex? (I guess it is, given it's 64vs16 :D) (0 to: 255) asByteArray base64Encoded size. "348" (0 to: 255) asByteArray hex size. "512" Beware the newlines in Base64 (better disable them) > Regards! > > Esteban A. Maringolo

Oct. 29, 2015

Re: [Pharo-users] Multiple method deletion

by Ferlicot D. Cyril

Le 29/10/2015 16:33, stepharo a Ã©crit : > Excellent news! > >> Hi, >> >> I think I found how to reproduce it: >> >> https://pharo.fogbugz.com/f/cases/16891/FT-Multi-selection-can-happen-when-… >> >> >> I think this is a FastTable problem with Nautilus. >> > > The problem is that when we try to remove something, the MethodsWidget give the selected methods. These methods are store inside an identityDictonary with the CompiledMethod as key and a boolean as value. Usually that works but when we compile a method a new instance of CompiledMethod is store on the dictionary. So with the identity dictionary we now have two elements for the same method since the pointers are not the same. Does someone know if it is important that the selected methods of the MethodWidget are store on an IdentityDictionary and not in a Dictionary ? If this is not important we can change it, else at the compilation we will need to remove this method from the IdentityDictionary. -- Cyril Ferlicot http://www.synectique.eu 165 Avenue Bretagne Lille 59000 France

Oct. 29, 2015

Re: [Pharo-users] fast subclassing/class creation in tests

by Marcus Denker

Hi, yes, I already felt that test runs involving unlogged code generation is very slowâ¦ can you add an entry on the issue tracker? > On 26 Oct 2015, at 13:14, Peter UhnÃ¡k <i.uhnak(a)gmail.com> wrote: > > Thanks Nicolai, the suspension seems to have quite a significant performance impact. > > Out of curiosity I did some benchmarking. Although the numbers seems to vary a lot based on something (the size of the image jumped from 25 to 40MB after I was done... and there seems to be some correlation with the performance). Repeatedly running a benchmark seems also detrimental as the performance can drop by over 30%... so probably it would be best to run each benchmark in a completely new image... > However the fact that the performance decreases over time is problematic, because that means that running the test suite will get worse over time. > > I find it also peculiar to see that unlogged anonymous classes are faster when the announcer is NOT suspended. > > And while for classes the improvement was ~40x, for methods it was over 200x faster... so I guess the best way is to modify performTest behavior to something like > > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > MyTestClass>>performTest > SystemAnnouncer uniqueInstance suspendAllWhile: [ super performTest ] > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > > > Benchmarks (I've removed decimal parts and thousand separators because they are always confusing me): > > Class creation with announcer: > > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > [ |cls| > cls := Object subclass: #Something. > cls removeFromSystem. > ] bench. "'9 per second'" > > [ |cls| > cls := Object subclass: #Something. > cls removeFromSystemUnlogged. > ] bench. "'173 per second'" > > [ |cls| > cls := Object newAnonymousSubclass. > cls removeFromSystem. > ] bench. "'4 per second'" > > [ |cls| > cls := Object newAnonymousSubclass. > cls removeFromSystemUnlogged. > ] bench. "'512 per second'" > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > > Class creation without announcer: > > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > [ SystemAnnouncer uniqueInstance suspendAllWhile: [ |cls| > cls := Object subclass: #Something. > cls removeFromSystem. > ] ] bench. "'371 per second'" > > [ SystemAnnouncer uniqueInstance suspendAllWhile: [ |cls| > cls := Object subclass: #Something. > cls removeFromSystemUnlogged. > ] ] bench. "'351 per second'" > > [ SystemAnnouncer uniqueInstance suspendAllWhile: [ |cls| > cls := Object newAnonymousSubclass. > cls removeFromSystem. > ] ] bench. "'368 per second'" > > [ SystemAnnouncer uniqueInstance suspendAllWhile: [ |cls| > cls := Object newAnonymousSubclass. > cls removeFromSystemUnlogged. > ] ] bench. "'430 per second'" > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > > Methods with announcer: > > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > cls := Object subclass: #Something. > [ > cls compile: 'method ^ 1'. > ] bench. "'9 per second'" > > anon := Object newAnonymousSubclass. > [ > anon compile: 'method ^ 1'. > ] bench. "'9 per second'" > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > > Methods without announcer > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > cls := Object subclass: #Something. > [ SystemAnnouncer uniqueInstance suspendAllWhile: [ > cls compile: 'method ^ 1'. > ] ] bench. "'1814 per second'" > > anon := Object newAnonymousSubclass. > [ SystemAnnouncer uniqueInstance suspendAllWhile: [ > anon compile: 'method ^ 1'. > ] ] bench. "'2063 per second'" > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ > > Peter > > On Sun, Oct 25, 2015 at 12:29 AM, Robert Withers <robert.w.withers(a)gmail.com <mailto:robert.w.withers@gmail.com>> wrote: > Extend Pharo byte codes to support Newspeak namespaces. #justsayin > > --- > robert > > On Oct 24, 2015, at 6:22 PM, Nicolai Hess <nicolaihess(a)gmail.com <mailto:nicolaihess@gmail.com>> wrote: > >> You can wrap the subclassing and the \removeFromSystem with >> SystemAnnouncer uniqueInstance suspendAllWhile: [ ] >> >> 2015-10-24 20:43 GMT+02:00 Peter UhnÃ¡k <i.uhnak(a)gmail.com <mailto:i.uhnak@gmail.com>>: >> Hi, >> >> is it possible to do fast, non-system wide class subclassing in tests? >> >> Currently I would do something like >> >> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ >> MyTest>>testSomething >> cls := SomeParent subclass: #Something. >> "... do some tests ..." >> cls removeFromSystem >> ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ >> >> This is for example what ClassTest is doing. >> >> However this is very slow, and even for small tests suite --- just ten (so far) test methods it takes over ten seconds to test it all, which is quite bad for TDD. >> >> My guess is that since it has to write to class to the environment and disk and then remove it it takes a while: >> >> ~~~~~~~~~~~~~ >> [ Object subclass: #Something ] timeToRun. "0:00:00:00.021" >> [ #Something asClass removeFromSystem ] timeToRun. "0:00:00:00.526" >> ~~~~~~~~~~~~~ >> >> is it possible to speed it up? Create classes that are fast to remove? >> >> Thanks, >> Peter >> >

Oct. 29, 2015

Re: [Pharo-users] What is your preferred way for storing ByteArray's as text?

by Esteban A. Maringolo

2015-10-29 12:44 GMT-03:00 Sven Van Caekenberghe <sven(a)stfx.eu>: > > > > On 29 Oct 2015, at 16:35, Esteban A. Maringolo <emaringolo(a)gmail.com> wrote: > > > > Hello all, > > > > As some already know I'm working in the new GLORP port from VW to Pharo. > > > > And as part of minor modifications I'm making to the code, I'd like to know what is your preferred way of storing ByteArrays as text representation/CLOB, etc... > > > > Option A) The printString of ByteArray, it is #[23 12 253 ...] > > Option B) A Base64 encoded version of the ByteArray > > Option C) Some other encoding > > Does that not dependent, at least in part, on the actual underlying SQL DB ? Yes it does, but certain backends (like SQLite) doesn't provide BLOB storage, and that's why I'm talking about CLOB (character instead of byte based) or text representations for it. I'd like to have a lowest common denominator to use as a fallback. > I think HEX and Base64 are make the most sense. Is Base64 more compact than Hex? (I guess it is, given it's 64vs16 :D) Regards! Esteban A. Maringolo

Oct. 29, 2015

Re: [Pharo-users] What is your preferred way for storing ByteArray's as text?

by Sven Van Caekenberghe

> On 29 Oct 2015, at 16:35, Esteban A. Maringolo <emaringolo(a)gmail.com> wrote: > > Hello all, > > As some already know I'm working in the new GLORP port from VW to Pharo. > > And as part of minor modifications I'm making to the code, I'd like to know what is your preferred way of storing ByteArrays as text representation/CLOB, etc... > > Option A) The printString of ByteArray, it is #[23 12 253 ...] > Option B) A Base64 encoded version of the ByteArray > Option C) Some other encoding Does that not dependent, at least in part, on the actual underlying SQL DB ? I think HEX and Base64 are make the most sense. > I posted the same question in the #databases channel at Pharo's Slack team (https://pharoproject.slack.com/archives/databases/p1446130304000022) > > Regards! > > > Esteban A. Maringolo

Oct. 29, 2015

What is your preferred way for storing ByteArray's as text?

by Esteban A. Maringolo

Hello all, As some already know I'm working in the new GLORP port from VW to Pharo. And as part of minor modifications I'm making to the code, I'd like to know what is your preferred way of storing ByteArrays as text representation/CLOB, etc... Option A) The printString of ByteArray, it is #[23 12 253 ...] Option B) A Base64 encoded version of the ByteArray Option C) Some other encoding I posted the same question in the #databases channel at Pharo's Slack team ( https://pharoproject.slack.com/archives/databases/p1446130304000022) Regards! Esteban A. Maringolo

Oct. 29, 2015

Re: [Pharo-users] Multiple method deletion

by stepharo

Excellent news! Le 29/10/15 10:18, Ferlicot D. Cyril a Ã©crit : > Le 09/10/2015 16:44, Alexandre Bergel a Ã©crit : >> Hi! >> >> When I select several methods in Nautilus, and press cmd-m-x it seems to >> delete more method that I have selected. It makes me lose code :-( >> >> Cheers, >> Alexandre >> -- >> _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: >> Alexandre Bergel http://www.bergel.eu >> ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. >> >> >> > Hi, > > I think I found how to reproduce it: > > https://pharo.fogbugz.com/f/cases/16891/FT-Multi-selection-can-happen-when-… > > I think this is a FastTable problem with Nautilus. >

Oct. 29, 2015

Re: [Pharo-users] Multiple method deletion

by Ferlicot D. Cyril

Le 09/10/2015 16:44, Alexandre Bergel a Ã©crit : > Hi! > > When I select several methods in Nautilus, and press cmd-m-x it seems to > delete more method that I have selected. It makes me lose code :-( > > Cheers, > Alexandre > -- > _,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;: > Alexandre Bergel http://www.bergel.eu > ^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;._,.;:~^~:;. > > > Hi, I think I found how to reproduce it: https://pharo.fogbugz.com/f/cases/16891/FT-Multi-selection-can-happen-when-… I think this is a FastTable problem with Nautilus. -- Cyril Ferlicot http://www.synectique.eu 165 Avenue Bretagne Lille 59000 France

Oct. 29, 2015

Re: [Pharo-users] Fossil and Filetree

by stepharo

> Stef, > > There is no problem. > I just *lack time* to keep moving on DrGeo features and migrate to > Pharo4.0 at the same time. > > I quickly try Dr. Geo code on Pharo4 and realize it will need some work > and lot of user testing to get DrGeo as robust as it is now. We will > conduct end user experiment in primary school, so I don't think > migrating now will be a good idea. > > I will very likely skip directly to Pharo 5.0, as I did from 1.4 to > Pharo 3.0 directly. I understand :) > > Hilaire > > Le 20/10/2015 16:02, stepharo a Ã©crit : >> Hilaire >> >> what is the problem that you encounter to still be in Pharo3.0? >> >> Stef >> >> >> Le 20/10/15 12:13, Hilaire a Ã©crit : >>> Le 20/10/2015 11:20, Thierry Goubier a Ã©crit : >>>> So it looks very similar to a filetree + git workflow ? >>>> >>>> Any taker on adding a fossil interface to filetree with the same >>>> approach as GitFileTree ? >>>> >>>> Thierry >>> Which version of GitFileTree should I take for Pharo3? >>> >>> Hilaire >>> >> >> >

Oct. 29, 2015