svn co http://www.squeakvm.org/svn/squeak/branches/Cog says: svn: Can't find a temporary directory: Internal error tried on multiple different machines, different OSes.. it seems to be a problem on a server side. (Most probably space issues) -- Best regards, Igor Stasenko. |
On Tue, Apr 17, 2012 at 04:13:48PM +0200, Igor Stasenko wrote: > > svn co http://www.squeakvm.org/svn/squeak/branches/Cog > says: > svn: Can't find a temporary directory: Internal error > > tried on multiple different machines, different OSes.. > > it seems to be a problem on a server side. > (Most probably space issues) > > -- > Best regards, > Igor Stasenko. The /tmp directory is getting filled up with www junk. Ian runs a cron job to keep it cleaned up, but this happened once before and he needed to manually clean it up. I have some large files on the system (the 64-bit images etc) and I'll offload them when I get home about 8 hours from now to free up some space. I think that some web crawler has started browsing through our Subversion web interface, and the resulting tmp files are killing the system. I'm pretty sure that no human could generate this volume of temporary files on the system. I'll also cc Ian in case he does not see this on the list. Dave |
I'll be needing to run a checkout on that as well... But I'll be using the svn interface to pull something recent... Maybe from that I can find what is needed for Polymorph to bootstrap a more complete Amiga Smalltalk-VM as one of its modules On Wed, Apr 18, 2012 at 3:38 AM, David T. Lewis <[hidden email]> wrote: > > On Tue, Apr 17, 2012 at 04:13:48PM +0200, Igor Stasenko wrote: >> >> svn co http://www.squeakvm.org/svn/squeak/branches/Cog >> says: >> svn: Can't find a temporary directory: Internal error >> >> tried on multiple different machines, different OSes.. >> >> it seems to be a problem on a server side. >> (Most probably space issues) >> >> -- >> Best regards, >> Igor Stasenko. > > The /tmp directory is getting filled up with www junk. Ian runs a cron > job to keep it cleaned up, but this happened once before and he needed > to manually clean it up. I have some large files on the system (the > 64-bit images etc) and I'll offload them when I get home about 8 hours > from now to free up some space. > > I think that some web crawler has started browsing through our Subversion > web interface, and the resulting tmp files are killing the system. I'm > pretty sure that no human could generate this volume of temporary files > on the system. > > I'll also cc Ian in case he does not see this on the list. > > Dave > |
Hi Jeremy: On 17 Apr 2012, at 17:47, Jeremy Kajikawa wrote: > > I'll be needing to run a checkout on that as well... > > But I'll be using the svn interface to pull something recent... > > Maybe from that I can find what is needed for Polymorph to bootstrap a > more complete Amiga Smalltalk-VM as one of its modules If you are more into C perhaps you should have a look at the interpreter loop of the RoarVM, which is a Squeak-compatible VM written in some C++ subset. https://github.com/smarr/RoarVM/blob/master/vm/src/interpreter/interpreter_bytecodes.cpp https://github.com/smarr/RoarVM/blob/master/vm/src/interpreter/squeak_interpreter.cpp#L398 The RoarVM code is a lot more readable than the C code optimized and generated from VMMaker. (interp.c in the SVN) Best regards Stefan > > On Wed, Apr 18, 2012 at 3:38 AM, David T. Lewis <[hidden email]> wrote: >> >> On Tue, Apr 17, 2012 at 04:13:48PM +0200, Igor Stasenko wrote: >>> >>> svn co http://www.squeakvm.org/svn/squeak/branches/Cog >>> says: >>> svn: Can't find a temporary directory: Internal error >>> >>> tried on multiple different machines, different OSes.. >>> >>> it seems to be a problem on a server side. >>> (Most probably space issues) >>> >>> -- >>> Best regards, >>> Igor Stasenko. >> >> The /tmp directory is getting filled up with www junk. Ian runs a cron >> job to keep it cleaned up, but this happened once before and he needed >> to manually clean it up. I have some large files on the system (the >> 64-bit images etc) and I'll offload them when I get home about 8 hours >> from now to free up some space. >> >> I think that some web crawler has started browsing through our Subversion >> web interface, and the resulting tmp files are killing the system. I'm >> pretty sure that no human could generate this volume of temporary files >> on the system. >> >> I'll also cc Ian in case he does not see this on the list. >> >> Dave >> -- Stefan Marr Software Languages Lab Vrije Universiteit Brussel Pleinlaan 2 / B-1050 Brussels / Belgium http://soft.vub.ac.be/~smarr Phone: +32 2 629 2974 Fax: +32 2 629 3525 |
Hello, and online I use Belxjander ... Thanks for the reference abour RoarVM being in C/C++ I'll certainly take a peek at that since the VM system I am writing is internally seperated between the processing and opcode functions. It won't make much sense if I try to explain it here other than it is a working construct at least for classical CPUs being Emulated. and I don't have to care which CPU is actually Emulated... they *all* work the same way for the most part. On Wed, Apr 18, 2012 at 4:08 AM, Stefan Marr <[hidden email]> wrote: > > Hi Jeremy: > > On 17 Apr 2012, at 17:47, Jeremy Kajikawa wrote: > >> >> I'll be needing to run a checkout on that as well... >> >> But I'll be using the svn interface to pull something recent... >> >> Maybe from that I can find what is needed for Polymorph to bootstrap a >> more complete Amiga Smalltalk-VM as one of its modules > > If you are more into C perhaps you should have a look at the interpreter loop of the RoarVM, which is a Squeak-compatible VM written in some C++ subset. > > https://github.com/smarr/RoarVM/blob/master/vm/src/interpreter/interpreter_bytecodes.cpp > > https://github.com/smarr/RoarVM/blob/master/vm/src/interpreter/squeak_interpreter.cpp#L398 > > > The RoarVM code is a lot more readable than the C code optimized and generated from VMMaker. (interp.c in the SVN) > > Best regards > Stefan > > >> >> On Wed, Apr 18, 2012 at 3:38 AM, David T. Lewis <[hidden email]> wrote: >>> >>> On Tue, Apr 17, 2012 at 04:13:48PM +0200, Igor Stasenko wrote: >>>> >>>> svn co http://www.squeakvm.org/svn/squeak/branches/Cog >>>> says: >>>> svn: Can't find a temporary directory: Internal error >>>> >>>> tried on multiple different machines, different OSes.. >>>> >>>> it seems to be a problem on a server side. >>>> (Most probably space issues) >>>> >>>> -- >>>> Best regards, >>>> Igor Stasenko. >>> >>> The /tmp directory is getting filled up with www junk. Ian runs a cron >>> job to keep it cleaned up, but this happened once before and he needed >>> to manually clean it up. I have some large files on the system (the >>> 64-bit images etc) and I'll offload them when I get home about 8 hours >>> from now to free up some space. >>> >>> I think that some web crawler has started browsing through our Subversion >>> web interface, and the resulting tmp files are killing the system. I'm >>> pretty sure that no human could generate this volume of temporary files >>> on the system. >>> >>> I'll also cc Ian in case he does not see this on the list. >>> >>> Dave >>> > > -- > Stefan Marr > Software Languages Lab > Vrije Universiteit Brussel > Pleinlaan 2 / B-1050 Brussels / Belgium > http://soft.vub.ac.be/~smarr > Phone: +32 2 629 2974 > Fax: +32 2 629 3525 > |
Jeremy Kajikawa wrote: > Thanks for the reference abour RoarVM being in C/C++ If you are interested in Squeak VMs written in languages other than Squeak itself (the subset of it we call "Slang", actually) then you might want to look at Spy, which is written in RPython: https://bitbucket.org/pypy/lang-smalltalk/ http://codespeak.net/pypy/extradoc/paper/spy-s3.pdf And Dan Ingalls implemented a Squeak VM in Java, called "JSqueak", at Sun: http://labs.oracle.com/projects/JSqueak/ The source seems to be missing, but the project was further developed as "Potato": http://www.hpi.uni-potsdam.de/hirschfeld/projects/potato/index.html -- Jecel |
In reply to this post by Jeremy Kajikawa
Not so worried about anything more than a cSmalltalkVM implementation. I won't be able to permit the Smalltalk implimentation to Edit the VM shell module itself. I'm writing this for Amiga OS 4.x and have to do so blind about the Smalltalk language itself. You have seen me on the OpenCobalt discussion group as ジェレミー on occasions when staying with BrickViking. Inside the Smalltalk VM is still a black box and without being able to relate to the terms I already know the whole "Blue Book" reference may as well be written in Chinese and being presented as Japanese for readability of the English content as read in English by me. It is about that helpful in my attempts at reading it as my style of reading is more hunting for direct references I can use. I've also explained my never using printf or the standard C library despite actually learning the C language first by teaching myself. I'm using the same techniques where I managed to get a benchmark result between 16MHz and 20MHz when Interpreting an 80486DX processor on a 68040 processor based Amiga A4000 in writing this VM implementation. yes I managed to get a 32bit 68040 to Emulate a 32bit 80486 with FPU operations at a reasonable speed. I would like to do the same with Smalltalk operations if at all possible. Starting with the opcode values themselves, then how any data is formatted around them. followed up by what each opcode will modify. if possible to have the above described in some kind of classical sense similar to C or C++ |
In reply to this post by David T. Lewis
Hi Dave, I have removed the /tmp pollution and modified the cron job to delete files more than one day old every night. If that doesn't work we can make it files older than one hour every hour instead. Regards, Ian On Apr 17, 2012, at 08:38 , David T. Lewis wrote: > The /tmp directory is getting filled up with www junk. > I think that some web crawler has started browsing through our Subversion > web interface, and the resulting tmp files are killing the system. I'm > pretty sure that no human could generate this volume of temporary files > on the system. |
Free forum by Nabble | Edit this page |