Hi,
I have been using the Cog VM exclusively for several months now, and from time to time it crashes on me without leaving any trace in debug log. One frequent cause for the crash that I noticed is dealing with Monticello (both reading and writing). Does anyone experience similar things? Am I doing something wrong? The image I got is based on the sources from 27.06.2010. Perhaps a more general idea: given that this VM brings an important improvement, do we want to have a more concerted effort to beta-test it? Cheers, Doru -- www.tudorgirba.com "It's not how it is, it is how we see it." _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
Do you ordinarily get output in the log? I have found the Linux vm to be very stingy with such information, but something like
( /fullPathTo/squeakvm \ /anotherFullPath/Pharo-1.0.image # ) 1>pharo_out.txt 2>pharo_error.txt seemed to work. ________________________________________ From: [hidden email] [[hidden email]] On Behalf Of Tudor Girba [[hidden email]] Sent: Sunday, July 25, 2010 6:47 AM To: Pharo Development Subject: [Pharo-project] cog-based image crashes on mac Hi, I have been using the Cog VM exclusively for several months now, and from time to time it crashes on me without leaving any trace in debug log. One frequent cause for the crash that I noticed is dealing with Monticello (both reading and writing). Does anyone experience similar things? Am I doing something wrong? The image I got is based on the sources from 27.06.2010. Perhaps a more general idea: given that this VM brings an important improvement, do we want to have a more concerted effort to beta-test it? Cheers, Doru -- www.tudorgirba.com "It's not how it is, it is how we see it." _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
In reply to this post by Tudor Girba
On Sun, 25 Jul 2010, Tudor Girba wrote:
> Hi, > > I have been using the Cog VM exclusively for several months now, and from Did you get a VM from Teleplace before it was open sourced? > time to time it crashes on me without leaving any trace in debug log. > > One frequent cause for the crash that I noticed is dealing with Monticello > (both reading and writing). Does anyone experience similar things? Am I doing > something wrong? IIRC this issue was fixed in VMMaker-oscog.17. > > The image I got is based on the sources from 27.06.2010. You may be missing some image-side changes. The CogVM was open sourced on 20.06.2010. Levente > > Perhaps a more general idea: given that this VM brings an important > improvement, do we want to have a more concerted effort to beta-test it? > > Cheers, > Doru > > > -- > www.tudorgirba.com > > "It's not how it is, it is how we see it." > > > _______________________________________________ > Pharo-project mailing list > [hidden email] > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
Hi,
On 25 Jul 2010, at 17:32, Levente Uzonyi wrote: > On Sun, 25 Jul 2010, Tudor Girba wrote: > >> Hi, >> >> I have been using the Cog VM exclusively for several months now, >> and from > > Did you get a VM from Teleplace before it was open sourced? No. I directly got the sources from SVN and compile them for Mac. Is there an official Cog VM compiled for Mac? >> time to time it crashes on me without leaving any trace in debug log. >> >> One frequent cause for the crash that I noticed is dealing with >> Monticello (both reading and writing). Does anyone experience >> similar things? Am I doing something wrong? > > IIRC this issue was fixed in VMMaker-oscog.17. I am not sure what this means :). >> The image I got is based on the sources from 27.06.2010. > > You may be missing some image-side changes. The CogVM was open > sourced on 20.06.2010. Actually, where should I get the complete image-side changes from? I have those prepared by Lukas a while ago. Cheers, Doru > Levente > >> >> Perhaps a more general idea: given that this VM brings an important >> improvement, do we want to have a more concerted effort to beta- >> test it? >> >> Cheers, >> Doru >> >> >> -- >> www.tudorgirba.com >> >> "It's not how it is, it is how we see it." >> >> >> _______________________________________________ >> Pharo-project mailing list >> [hidden email] >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > _______________________________________________ > Pharo-project mailing list > [hidden email] > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project -- www.tudorgirba.com "Not knowing how to do something is not an argument for how it cannot be done." _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
>>> The image I got is based on the sources from 27.06.2010.
>> >> You may be missing some image-side changes. The CogVM was open sourced on >> 20.06.2010. > > Actually, where should I get the complete image-side changes from? I have > those prepared by Lukas a while ago. Yeah, these are the complete ones from the repository adapted to Pharo. Maybe these same socket related crashes that randomly appear with Seaside images? nanosleep: Invalid argument Exited with exit code: 1 Lukas -- Lukas Renggli www.lukas-renggli.ch _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
There's an update in svn that could fix crashes. I had some logging code that looks like
static unsigned long long useclog[LOGSIZE]; static int ulogidx = (unsigned int)-1; # define logusecs(usecs) do { sqLowLevelMFence(); \
if (logClock) useclog[++ulogidx % LOGSIZE] = (usecs); \ } while (0) which blows up when ulogidx wraps around. It the sign of -ve % +ve is undefined in C.
The fix is simple; make the index unsigned: static unsigned long long useclog[LOGSIZE]; static unsigned int ulogidx = (unsigned int)-1; # define logusecs(usecs) do { sqLowLevelMFence(); \
if (logClock) useclog[++ulogidx % LOGSIZE] = (usecs); \ } while (0) HTH Eliot On Mon, Jul 26, 2010 at 4:32 AM, Lukas Renggli <[hidden email]> wrote:
_______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
2010/7/26 Eliot Miranda <[hidden email]>:
> There's an update in svn that could fix crashes. No sorry, still the same (if I remember correctly I've never seen the bunch of error messages before the last two though) with the latest svn code: 26.07.10 20:39:33 [0x0-0x1d3dd3c].com.teleplace.Teleplace[4514] mprotect(x,y,PROT_READ | PROT_WRITE): Cannot allocate memory 26.07.10 20:39:33 [0x0-0x1d3dd3c].com.teleplace.Teleplace[4514] mprotect(x,y,PROT_READ | PROT_WRITE): Invalid argument 26.07.10 20:39:33 [0x0-0x1d3dd3c].com.teleplace.Teleplace[4514] squeak: could not load plugin `MiscPrimitivePlugin' 26.07.10 20:39:33 [0x0-0x1d3dd3c].com.teleplace.Teleplace[4514] squeak: could not load plugin `DropPlugin' 26.07.10 20:39:33 [0x0-0x1d3dd3c].com.teleplace.Teleplace[4514] squeak: could not load plugin `FilePlugin' 26.07.10 20:39:33 [0x0-0x1d3dd3c].com.teleplace.Teleplace[4514] squeak: could not load plugin `SecurityPlugin' 26.07.10 20:39:33 [0x0-0x1d3dd3c].com.teleplace.Teleplace[4514] squeak: could not load plugin `BitBltPlugin' 26.07.10 20:39:33 [0x0-0x1d3dd3c].com.teleplace.Teleplace[4514] squeak: could not load plugin `LargeIntegers' 26.07.10 20:39:33 [0x0-0x1d3dd3c].com.teleplace.Teleplace[4514] squeak: could not load plugin `B2DPlugin' 26.07.10 20:39:33 [0x0-0x1d3dd3c].com.teleplace.Teleplace[4514] squeak: could not load plugin `InternetConfigPlugin' 26.07.10 20:39:33 [0x0-0x1d3dd3c].com.teleplace.Teleplace[4514] squeak: could not load plugin `SocketPlugin' 26.07.10 20:40:03 [0x0-0x1d3dd3c].com.teleplace.Teleplace[4514] squeak: could not load plugin `DSAPrims' 26.07.10 20:40:06 [0x0-0x1d3dd3c].com.teleplace.Teleplace[4514] squeak: could not load plugin `JPEGReadWriter2Plugin' 26.07.10 20:41:33 [0x0-0x1d3dd3c].com.teleplace.Teleplace[4514] nanosleep: Invalid argument 26.07.10 20:41:33 com.apple.launchd.peruser.501[200] ([0x0-0x1d3dd3c].com.teleplace.Teleplace[4514]) Exited with exit code: 1 Lukas -- Lukas Renggli www.lukas-renggli.ch _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
I have a similar problem, with the latest svn version, 2249.
(localPrimIndex > 255) && (localPrimIndex < 520) 15376 squeak: could not load plugin `SocketPlugin' [Switching to process 26735] nanosleep: Invalid argument But it is not random. It happens everytime I try to run PharoCore 12069 tests. Cheers MAriano On Mon, Jul 26, 2010 at 8:43 PM, Lukas Renggli <[hidden email]> wrote: 2010/7/26 Eliot Miranda <[hidden email]>: _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
In reply to this post by Lukas Renggli
lukas
could you have a look at http://code.google.com/p/pharo/issues/detail?id=2579 because henrik mentioned that we missed something or could do it better. Stef On Jul 26, 2010, at 1:32 PM, Lukas Renggli wrote: >>>> The image I got is based on the sources from 27.06.2010. >>> >>> You may be missing some image-side changes. The CogVM was open sourced on >>> 20.06.2010. >> >> Actually, where should I get the complete image-side changes from? I have >> those prepared by Lukas a while ago. > > Yeah, these are the complete ones from the repository adapted to Pharo. > > Maybe these same socket related crashes that randomly appear with > Seaside images? > > nanosleep: Invalid argument > Exited with exit code: 1 > > Lukas > > -- > Lukas Renggli > www.lukas-renggli.ch > > _______________________________________________ > Pharo-project mailing list > [hidden email] > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
The changes I submitted are the ones of Eliot adapted to work in
Pharo. I don't understand what Henrik writes, if he has something better he should attach the change. Lukas On Sunday, August 1, 2010, Stéphane Ducasse <[hidden email]> wrote: > lukas > > could you have a look at > http://code.google.com/p/pharo/issues/detail?id=2579 > because henrik mentioned that we missed something or could do it better. > > Stef > > > On Jul 26, 2010, at 1:32 PM, Lukas Renggli wrote: > >>>>> The image I got is based on the sources from 27.06.2010. >>>> >>>> You may be missing some image-side changes. The CogVM was open sourced on >>>> 20.06.2010. >>> >>> Actually, where should I get the complete image-side changes from? I have >>> those prepared by Lukas a while ago. >> >> Yeah, these are the complete ones from the repository adapted to Pharo. >> >> Maybe these same socket related crashes that randomly appear with >> Seaside images? >> >> nanosleep: Invalid argument >> Exited with exit code: 1 >> >> Lukas >> >> -- >> Lukas Renggli >> www.lukas-renggli.ch >> >> _______________________________________________ >> Pharo-project mailing list >> [hidden email] >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > > _______________________________________________ > Pharo-project mailing list > [hidden email] > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > -- Lukas Renggli www.lukas-renggli.ch _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
Tx!
Stef On Aug 1, 2010, at 10:40 AM, Lukas Renggli wrote: > The changes I submitted are the ones of Eliot adapted to work in > Pharo. I don't understand what Henrik writes, if he has something > better he should attach the change. > > Lukas > > On Sunday, August 1, 2010, Stéphane Ducasse <[hidden email]> wrote: >> lukas >> >> could you have a look at >> http://code.google.com/p/pharo/issues/detail?id=2579 >> because henrik mentioned that we missed something or could do it better. >> >> Stef >> >> >> On Jul 26, 2010, at 1:32 PM, Lukas Renggli wrote: >> >>>>>> The image I got is based on the sources from 27.06.2010. >>>>> >>>>> You may be missing some image-side changes. The CogVM was open sourced on >>>>> 20.06.2010. >>>> >>>> Actually, where should I get the complete image-side changes from? I have >>>> those prepared by Lukas a while ago. >>> >>> Yeah, these are the complete ones from the repository adapted to Pharo. >>> >>> Maybe these same socket related crashes that randomly appear with >>> Seaside images? >>> >>> nanosleep: Invalid argument >>> Exited with exit code: 1 >>> >>> Lukas >>> >>> -- >>> Lukas Renggli >>> www.lukas-renggli.ch >>> >>> _______________________________________________ >>> Pharo-project mailing list >>> [hidden email] >>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >> >> >> _______________________________________________ >> Pharo-project mailing list >> [hidden email] >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >> > > -- > Lukas Renggli > www.lukas-renggli.ch > > _______________________________________________ > Pharo-project mailing list > [hidden email] > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
In reply to this post by Lukas Renggli
On Aug 1, 2010, at 10:40 53AM, Lukas Renggli wrote: > The changes I submitted are the ones of Eliot adapted to work in > Pharo. I don't understand what Henrik writes, if he has something > better he should attach the change. > > Lukas Not better, just simpler code. If I'm not mistaken, if primitive 38 fails in Cog, so will primitive 60. Primitive 38 will always fail in a non-cog vm. The failure handling code after the ec == nil check is the same in Float>>basicAt: and Object basicAt: Thus, writing Float>>basicAt: index <primitive:38> ^super basicAt: index would yield the exact same result, with the difference that you will have an extra primtive 60 failure in Cog vm's, an ec = nil check less in non-cog VM's, and no duplication of the fallback-handling code. The alternate code was attached, in issue 2581. Cheers, Henry > > On Sunday, August 1, 2010, Stéphane Ducasse <[hidden email]> wrote: >> lukas >> >> could you have a look at >> http://code.google.com/p/pharo/issues/detail?id=2579 >> because henrik mentioned that we missed something or could do it better. >> >> Stef >> >> >> On Jul 26, 2010, at 1:32 PM, Lukas Renggli wrote: >> >>>>>> The image I got is based on the sources from 27.06.2010. >>>>> >>>>> You may be missing some image-side changes. The CogVM was open sourced on >>>>> 20.06.2010. >>>> >>>> Actually, where should I get the complete image-side changes from? I have >>>> those prepared by Lukas a while ago. >>> >>> Yeah, these are the complete ones from the repository adapted to Pharo. >>> >>> Maybe these same socket related crashes that randomly appear with >>> Seaside images? >>> >>> nanosleep: Invalid argument >>> Exited with exit code: 1 >>> >>> Lukas >>> >>> -- >>> Lukas Renggli >>> www.lukas-renggli.ch >>> >>> _______________________________________________ >>> Pharo-project mailing list >>> [hidden email] >>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >> >> >> _______________________________________________ >> Pharo-project mailing list >> [hidden email] >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >> > > -- > Lukas Renggli > www.lukas-renggli.ch > > _______________________________________________ > Pharo-project mailing list > [hidden email] > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
thanks for the explanation and code.
On Aug 2, 2010, at 3:43 AM, Henrik Johansen wrote: > > On Aug 1, 2010, at 10:40 53AM, Lukas Renggli wrote: > >> The changes I submitted are the ones of Eliot adapted to work in >> Pharo. I don't understand what Henrik writes, if he has something >> better he should attach the change. >> >> Lukas > > Not better, just simpler code. > > If I'm not mistaken, if primitive 38 fails in Cog, so will primitive 60. > Primitive 38 will always fail in a non-cog vm. > The failure handling code after the ec == nil check is the same in Float>>basicAt: and Object basicAt: > > Thus, writing > Float>>basicAt: index > <primitive:38> > ^super basicAt: index > > would yield the exact same result, with the difference that you will have an extra primtive 60 failure in Cog vm's, an ec = nil check less in non-cog VM's, and no duplication of the fallback-handling code. > > The alternate code was attached, in issue 2581. > > Cheers, > Henry > >> >> On Sunday, August 1, 2010, Stéphane Ducasse <[hidden email]> wrote: >>> lukas >>> >>> could you have a look at >>> http://code.google.com/p/pharo/issues/detail?id=2579 >>> because henrik mentioned that we missed something or could do it better. >>> >>> Stef >>> >>> >>> On Jul 26, 2010, at 1:32 PM, Lukas Renggli wrote: >>> >>>>>>> The image I got is based on the sources from 27.06.2010. >>>>>> >>>>>> You may be missing some image-side changes. The CogVM was open sourced on >>>>>> 20.06.2010. >>>>> >>>>> Actually, where should I get the complete image-side changes from? I have >>>>> those prepared by Lukas a while ago. >>>> >>>> Yeah, these are the complete ones from the repository adapted to Pharo. >>>> >>>> Maybe these same socket related crashes that randomly appear with >>>> Seaside images? >>>> >>>> nanosleep: Invalid argument >>>> Exited with exit code: 1 >>>> >>>> Lukas >>>> >>>> -- >>>> Lukas Renggli >>>> www.lukas-renggli.ch >>>> >>>> _______________________________________________ >>>> Pharo-project mailing list >>>> [hidden email] >>>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >>> >>> >>> _______________________________________________ >>> Pharo-project mailing list >>> [hidden email] >>> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project >>> >> >> -- >> Lukas Renggli >> www.lukas-renggli.ch >> >> _______________________________________________ >> Pharo-project mailing list >> [hidden email] >> http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project > > > _______________________________________________ > Pharo-project mailing list > [hidden email] > http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project _______________________________________________ Pharo-project mailing list [hidden email] http://lists.gforge.inria.fr/cgi-bin/mailman/listinfo/pharo-project |
Free forum by Nabble | Edit this page |