- Pharo-dev - lists.pharo.org

Re: [Pharo-dev] context mirror primitives

by Christophe Demarey

Le 6 janv. 2016 Ã 08:11, Marcus Denker a Ã©crit : > >> On 05 Jan 2016, at 21:04, stepharo <stepharo(a)free.fr> wrote: >> >> All the Inria integration got a problem and this is restarting to work. >> So we will get there. >> > > Something is still broken: > > https://ci.inria.fr/pharo/ > > all slaves are blocked with threads that are dead. It looks like Jenkins is mess up with a job number. I need to find the job causing that and I think I could fix that

Jan. 6, 2016

[ANN] The next Pharo Sprints (Lille)

by Marcus Denker

Hello, We looked at the calendar and fixed the dates for all public sprints that we will do this year here locally. These are all Fridays: 29 Januar 26 Feb 25 March 29 April 27 May They will be at the Inria Lille, Building B, third floor (RMoD offices). Remotely, you can join us on Slack or the IRC channel #pharo on irc.freenode.net server. During the sprint, we will try to synchronize local and remote Pharo sprinters. One nice idea is to organise local sprints yourself. E.g. over the last sprints we had one in Chile at the same time. Marcus

Jan. 6, 2016

Re: [Pharo-dev] context mirror primitives

by Marcus Denker

> On 05 Jan 2016, at 21:04, stepharo <stepharo(a)free.fr> wrote: > > All the Inria integration got a problem and this is restarting to work. > So we will get there. > Something is still broken: https://ci.inria.fr/pharo/ all slaves are blocked with threads that are dead. Marcus

Jan. 6, 2016

Re: [Pharo-dev] [Ann] ReadWriteLock

by Tudor Girba

Thanks! Doru > On Jan 6, 2016, at 12:56 AM, Ben Coman <btc(a)openInWorld.com> wrote: > > okay. I'll have a go. > cheers -ben > > On Wed, Jan 6, 2016 at 3:28 AM, stepharo <stepharo(a)free.fr> wrote: >> Ben >> >> would you not be interested to write a little chapter on concurrency because >> we would all benefit from it. >> I can help but I'm pretty bad with concurrency. >> I'm sure that doing that we will find a lot of things to improve in Pharo :) >> >> Stef >> >> Le 5/1/16 18:15, Ben Coman a Ã©crit : >> >>> On Tue, Jan 5, 2016 at 11:59 PM, Denis Kudriashov <dionisiydk(a)gmail.com> >>> wrote: >>>> >>>> 2016-01-05 16:06 GMT+01:00 Ben Coman <btc(a)openinworld.com>: >>>>> >>>>> btw I have this growing suspicion that we should eliminate >>>>> Semaphore>>critical: and Semaphore>>forMutualExclusion. Semaphores >>>>> on their own should *not* provide a facility for to protect critical >>>>> regions since they suffer from accidental release [1] due to >>>>> *lack*of*ownership*. >>>> >>>> >>>> I have same feeling when analyse problem of critical: implementation. But >>>> I >>>> not think that Semaphore>>forMutualExclusion is bad. >>> >>> But Semaphore>>forMutualExclusion *encourages* people to believe >>> semaphores can be used *on*their*own* for mutual exclusion, when they >>> *shouldn't*. You *must* add ownership. If the mutual exclusion >>> object doesnât have ownership then, irrelevant of what it is called, >>> it is not a mutex!!! [1]. >>> >>> "Strictly speaking [2]... >>> -- A mutex is *locking* mechanism used to synchronize access to a >>> resource. Only one task (can be a thread or process based on OS >>> abstraction) can acquire the mutex. It means there is *ownership* >>> associated with mutex, and only the owner can release the lock >>> (mutex). >>> -- A semaphore is *signaling* mechanism (âI am done, you can carry >>> onâ kind of signal). For example, if you are listening songs (assume >>> it as one task) on your mobile and at the same time your friend calls >>> you, an interrupt is triggered upon which an interrupt service routine >>> (ISR) signals the call processing task to wakeup >>> >>> >>>> It is common case that semaphore "free" by default. >>> >>> With an implementation like.. >>> Semaphore>>forMutualExclusion >>> ^self new signal >>> >>> we don't gain a lot, and its a misleading convenience method since its >>> doesn't provide the proper facility for mutual exclusion (and I do >>> conflate mutual exclusion == mutex). Indeed, "Semaphore >>> forMutualExclusion + roll your ownership management + problems since >>> you weren't aware you needed to roll your own ownership management" is >>> not more convenient than "Mutex new" >>> >>> If you want to keep a convenience method for a pre signalled >>> Semaphore, then we could provide something like "Semaphore >>> newSignalled" >>> >>>> And this name is explicitly said about this. >>> >>> Actually no. It explicitly says it provides mutual exclusion - which >>> without ownership, it does not. It only implicitly provides a >>> pre-signalled Semaphore - you have to *know* of look at the source. >>> >>> [1] >>> https://blog.feabhas.com/2009/09/mutex-vs-semaphores-%E2%80%93-part-2-the-m… >>> [2] http://www.geeksforgeeks.org/mutex-vs-semaphore/ >>> >>> cheers -ben >>> >>> >> >> > -- www.tudorgirba.com www.feenk.com "Reasonable is what we are accustomed with."

Jan. 6, 2016

Re: [Pharo-dev] small refactor on files.pharo.org

by Dimitris Chloupis

thank you :) On Tue, Jan 5, 2016 at 9:57 PM Esteban Lorenzano <estebanlm(a)gmail.com> wrote: > On 05 Jan 2016, at 20:06, Dimitris Chloupis <kilon.alios(a)gmail.com> wrote: > > but this still works as expected yes ? > > wget -O- get.pharo.org/alpha+vmLatest | bash > > > yes > > > > > On Tue, Jan 5, 2016 at 8:23 PM Esteban Lorenzano <estebanlm(a)gmail.com> > wrote: > >> On 05 Jan 2016, at 19:14, Andrei Chis <chisvasileandrei(a)gmail.com> wrote: >> >> What's the recommended way of loading the VM through zeroconf scripts? >> >> Just http://get.pharo.org/vm loads the non-spur vm and >> http://get.pharo.org/vm50 loads the spur one. Will >> http://get.pharo.org/vm be updated to the spur one in the future? >> >> >> Hi, >> >> no it will not be updated. >> In fact if you see is not listed anymore (since almost one year now). It >> is just there for backward compatibility. >> >> you need to use the VM corresponding with your image version (in case of >> Pharo 50, vm50). >> >> Esteban >> >> ps: this was described when I made the change like one year agoâ¦ but now >> I do not find the mail :P >> >> >> Cheers, >> Andrei >> >> On Tue, Jan 5, 2016 at 11:47 AM, Esteban Lorenzano <estebanlm(a)gmail.com> >> wrote: >> >>> I move old VMs from files.pharo.org/vm to files.pharo.org/vm/old >>> >>> because it was confusing and some students were lost (Stef too :P). >>> >>> cheers, >>> Esteban >>> >> >>

Jan. 5, 2016

Re: [Pharo-dev] [Ann] ReadWriteLock

by Ben Coman

okay. I'll have a go. cheers -ben On Wed, Jan 6, 2016 at 3:28 AM, stepharo <stepharo(a)free.fr> wrote: > Ben > > would you not be interested to write a little chapter on concurrency because > we would all benefit from it. > I can help but I'm pretty bad with concurrency. > I'm sure that doing that we will find a lot of things to improve in Pharo :) > > Stef > > Le 5/1/16 18:15, Ben Coman a Ã©crit : > >> On Tue, Jan 5, 2016 at 11:59 PM, Denis Kudriashov <dionisiydk(a)gmail.com> >> wrote: >>> >>> 2016-01-05 16:06 GMT+01:00 Ben Coman <btc(a)openinworld.com>: >>>> >>>> btw I have this growing suspicion that we should eliminate >>>> Semaphore>>critical: and Semaphore>>forMutualExclusion. Semaphores >>>> on their own should *not* provide a facility for to protect critical >>>> regions since they suffer from accidental release [1] due to >>>> *lack*of*ownership*. >>> >>> >>> I have same feeling when analyse problem of critical: implementation. But >>> I >>> not think that Semaphore>>forMutualExclusion is bad. >> >> But Semaphore>>forMutualExclusion *encourages* people to believe >> semaphores can be used *on*their*own* for mutual exclusion, when they >> *shouldn't*. You *must* add ownership. If the mutual exclusion >> object doesnât have ownership then, irrelevant of what it is called, >> it is not a mutex!!! [1]. >> >> "Strictly speaking [2]... >> -- A mutex is *locking* mechanism used to synchronize access to a >> resource. Only one task (can be a thread or process based on OS >> abstraction) can acquire the mutex. It means there is *ownership* >> associated with mutex, and only the owner can release the lock >> (mutex). >> -- A semaphore is *signaling* mechanism (âI am done, you can carry >> onâ kind of signal). For example, if you are listening songs (assume >> it as one task) on your mobile and at the same time your friend calls >> you, an interrupt is triggered upon which an interrupt service routine >> (ISR) signals the call processing task to wakeup >> >> >>> It is common case that semaphore "free" by default. >> >> With an implementation like.. >> Semaphore>>forMutualExclusion >> ^self new signal >> >> we don't gain a lot, and its a misleading convenience method since its >> doesn't provide the proper facility for mutual exclusion (and I do >> conflate mutual exclusion == mutex). Indeed, "Semaphore >> forMutualExclusion + roll your ownership management + problems since >> you weren't aware you needed to roll your own ownership management" is >> not more convenient than "Mutex new" >> >> If you want to keep a convenience method for a pre signalled >> Semaphore, then we could provide something like "Semaphore >> newSignalled" >> >>> And this name is explicitly said about this. >> >> Actually no. It explicitly says it provides mutual exclusion - which >> without ownership, it does not. It only implicitly provides a >> pre-signalled Semaphore - you have to *know* of look at the source. >> >> [1] >> https://blog.feabhas.com/2009/09/mutex-vs-semaphores-%E2%80%93-part-2-the-m… >> [2] http://www.geeksforgeeks.org/mutex-vs-semaphore/ >> >> cheers -ben >> >> > >

Jan. 5, 2016

Re: [Pharo-dev] context mirror primitives

by Tudor Girba

Ok. No problem. Please let me know if I can help. Doru > On Jan 5, 2016, at 10:04 PM, stepharo <stepharo(a)free.fr> wrote: > > All the Inria integration got a problem and this is restarting to work. > So we will get there. > > Stef > > Le 5/1/16 20:38, Tudor Girba a Ã©crit : >> Hi, >> >> In summary, it seems we should integrate this one. >> >> I resolved the issue to "Resolved (Fix Review Needed)â, but it seems that the Monkey is still on holidays :). Does anyone know if there is a problem with it? >> >> Cheers, >> Doru >> >> >>> On Jan 4, 2016, at 5:42 PM, Tudor Girba <tudor(a)tudorgirba.com> wrote: >>> >>> Hi, >>> >>> The VM (both Cog and Spur) supports basic manipulation of objects without relying on sending messages to those objects. These primitives are meant to be handled by methods in Context. Eliot provided a changeset with these methods, and they are not yet in Pharo: >>> >>> object: anObject basicAt: index >>> object: anObject basicAt: index put: value >>> object: anObject eqeq: anOtherObject >>> object: anObject instVarAt: anIndex >>> object: anObject instVarAt: anIndex put: aValue >>> object: anObject perform: selector withArguments: argArray inClass: lookupClass >>> objectClass: anObject >>> objectSize: anObject >>> >>> I added an issue and the slice: >>> https://pharo.fogbugz.com/f/cases/17313/Missing-primitive-methods-in-Context >>> >>> As I note in the issue, I think it is certainly interesting to have a more powerful mirrors mechanism, but until then we should still be able to use what the VM already provides. >>> >>> I would like to use these primitives now to make the GT Inspector and Debugger be able to work with ProtoObject. I thought these methods were supposed to be in the Spur version of Pharo, but it seems they are not. So, I am wondering if maybe I am not missing something. Am I missing something or should we integrate these methods? >>> >>> Cheers, >>> Doru >>> >>> -- >>> www.tudorgirba.com >>> www.feenk.com >>> >>> "Problem solving efficiency grows with the abstractness level of problem understanding." >>> >>> >>> >>> >> -- >> www.tudorgirba.com >> www.feenk.com >> >> "We are all great at making mistakes." >> >> >> >> >> >> >> >> >> >> > > -- www.tudorgirba.com www.feenk.com "When people care, great things can happen."

Jan. 5, 2016

Re: [Pharo-dev] context mirror primitives

by stepharo

All the Inria integration got a problem and this is restarting to work. So we will get there. Stef Le 5/1/16 20:38, Tudor Girba a Ã©crit : > Hi, > > In summary, it seems we should integrate this one. > > I resolved the issue to "Resolved (Fix Review Needed)â, but it seems that the Monkey is still on holidays :). Does anyone know if there is a problem with it? > > Cheers, > Doru > > >> On Jan 4, 2016, at 5:42 PM, Tudor Girba <tudor(a)tudorgirba.com> wrote: >> >> Hi, >> >> The VM (both Cog and Spur) supports basic manipulation of objects without relying on sending messages to those objects. These primitives are meant to be handled by methods in Context. Eliot provided a changeset with these methods, and they are not yet in Pharo: >> >> object: anObject basicAt: index >> object: anObject basicAt: index put: value >> object: anObject eqeq: anOtherObject >> object: anObject instVarAt: anIndex >> object: anObject instVarAt: anIndex put: aValue >> object: anObject perform: selector withArguments: argArray inClass: lookupClass >> objectClass: anObject >> objectSize: anObject >> >> I added an issue and the slice: >> https://pharo.fogbugz.com/f/cases/17313/Missing-primitive-methods-in-Context >> >> As I note in the issue, I think it is certainly interesting to have a more powerful mirrors mechanism, but until then we should still be able to use what the VM already provides. >> >> I would like to use these primitives now to make the GT Inspector and Debugger be able to work with ProtoObject. I thought these methods were supposed to be in the Spur version of Pharo, but it seems they are not. So, I am wondering if maybe I am not missing something. Am I missing something or should we integrate these methods? >> >> Cheers, >> Doru >> >> -- >> www.tudorgirba.com >> www.feenk.com >> >> "Problem solving efficiency grows with the abstractness level of problem understanding." >> >> >> >> > -- > www.tudorgirba.com > www.feenk.com > > "We are all great at making mistakes." > > > > > > > > > >

Jan. 5, 2016

Re: [Pharo-dev] small refactor on files.pharo.org

by Esteban Lorenzano

> On 05 Jan 2016, at 20:06, Dimitris Chloupis <kilon.alios(a)gmail.com> wrote: > > but this still works as expected yes ? > > wget -O- get.pharo.org/alpha+vmLatest <http://get.pharo.org/alpha+vmLatest> | bash yes > > > > On Tue, Jan 5, 2016 at 8:23 PM Esteban Lorenzano <estebanlm(a)gmail.com <mailto:estebanlm@gmail.com>> wrote: >> On 05 Jan 2016, at 19:14, Andrei Chis <chisvasileandrei(a)gmail.com <mailto:chisvasileandrei@gmail.com>> wrote: >> >> What's the recommended way of loading the VM through zeroconf scripts? >> >> Just http://get.pharo.org/vm <http://get.pharo.org/vm> loads the non-spur vm and http://get.pharo.org/vm50 <http://get.pharo.org/vm50> loads the spur one. Will http://get.pharo.org/vm <http://get.pharo.org/vm> be updated to the spur one in the future? > > Hi, > > no it will not be updated. > In fact if you see is not listed anymore (since almost one year now). It is just there for backward compatibility. > > you need to use the VM corresponding with your image version (in case of Pharo 50, vm50). > > Esteban > > ps: this was described when I made the change like one year agoâ¦ but now I do not find the mail :P > >> >> Cheers, >> Andrei >> >> On Tue, Jan 5, 2016 at 11:47 AM, Esteban Lorenzano <estebanlm(a)gmail.com <mailto:estebanlm@gmail.com>> wrote: >> I move old VMs from files.pharo.org/vm <http://files.pharo.org/vm> to files.pharo.org/vm/old <http://files.pharo.org/vm/old> >> >> because it was confusing and some students were lost (Stef too :P). >> >> cheers, >> Esteban >>

Jan. 5, 2016

Process/Delay waiting on multiple Semaphores at once?

by Holger Freyther

Hi, I look into a way to implement something like Erlang's selective process by having Delay or something like it wait on multiple Semaphores (and maybe blocks in the future but that leads the question of side-effects if the event is not used) and fire on the first one ready without having to deal with many Processes. The usecases I have in mind right now are: * Wait for a delivery time or if any new work has come up * When connecting a call (e.g. two SIP calls) I want to manage the state from a central place (e.g. like an event inbox) and dispatch events from the same context and I do not know which side the event occurs. SelectiveReceive new check: database readableSemaphore do: [ev := self nextDatabaseEvent]; check: file readableSemaphore do: []; timeoutDo: []; runFor: 30 seconds. A prototype to get a feel for the API could be built by spawning n-processes that each wait on their own semaphore for the timeout and then create another semaphore that is signalled by these processes in case of a wakeup signal the new Semaphore and then in the runFor: selector kill all remaining processes.. What do you think of this pattern instead of having to deal with concurrency with more locking, more fine grained locking? What would it take to get this deeper into Pharo? I think fundamentally a Process would need to be able to sleep on a list of semaphores? Is that feasible? kind regards holger

Jan. 5, 2016