Vm-dev April 2013

vm-dev@lists.squeakfoundation.org

36 participants
107 discussions

[commit][2727] commit some Raspberry Pi/Raspbian changes; note that unix/ config/confgure[.ac] has -msse commented out and this needs to be done more neatly
by commits＠squeakvm.org 27 Apr '13

27 Apr '13

Revision: 2727 Author: rowledge Date: 2013-04-26 16:56:18 -0700 (Fri, 26 Apr 2013) Log Message: ----------- commit some Raspberry Pi/Raspbian changes; note that unix/config/confgure[.ac] has -msse commented out and this needs to be done more neatly Modified Paths: -------------- branches/Cog/platforms/Cross/vm/sqAtomicOps.h branches/Cog/platforms/Cross/vm/sqMemoryFence.h branches/Cog/platforms/unix/config/configure branches/Cog/platforms/unix/config/configure.ac branches/Cog/platforms/unix/vm/sqUnixITimerHeartbeat.c Modified: branches/Cog/platforms/Cross/vm/sqAtomicOps.h =================================================================== --- branches/Cog/platforms/Cross/vm/sqAtomicOps.h 2013-04-25 18:38:59 UTC (rev 2726) +++ branches/Cog/platforms/Cross/vm/sqAtomicOps.h 2013-04-26 23:56:18 UTC (rev 2727) @@ -1,11 +1,12 @@ /**************************************************************************** * PROJECT: Atomic operations for multi-threading. * Atomic reads and writes of 64-bit values (e.g. for 64-bit clock). -* Atomic 32-bit increment (e.g. for signalSemaphoreWithIndex:). * get64(sqLong variable) * set64(sqLong variable, sqLong value) +* Atomic 32-bit increment (e.g. for signalSemaphoreWithIndex:). * sqAtomicAddConst(var,n) * sqCompareAndSwap(var,old,new) +* sqCompareAndSwapRes(var,old,new,res) * FILE: sqAtomicOps.h * * AUTHOR: Eliot Miranda @@ -29,6 +30,8 @@ * we can make a good guess that __LONG_MAX__ implies 32-bits or 64-bits. */ +// tpr; Raspbian does not define ILP32 or LP32, therefore LP32 is #def'd below + #if defined(__LONG_MAX__) && !defined(LP32) && !defined(ILP32) \ && !defined(LP64) && !defined(ILP64) && !defined(LLP64) # if __LONG_MAX__ > 0xFFFFFFFF @@ -44,6 +47,8 @@ # define set64(variable,value) (variable = value) #elif LP32 || ILP32 + + # if TARGET_OS_IS_IPHONE static inline void AtomicSet(uint64_t *target, uint64_t new_value) @@ -131,13 +136,24 @@ : "memory", "eax", "ebx", "ecx", "edx", "cc") # endif /* __SSE2__ */ # else /* TARGET_OS_IS_IPHONE elif x86 variants etc */ + +#if defined(__arm__) && defined(__ARM_ARCH_6__) +/* tpr - this is code intended for the Raspberry Pi Raspbian OS + * We'll experimentally trust in our MMU to keep 64bit accesses atomic */ +#define get64(var) \ + (var) +#define set64(var,value) \ + (var) = (value) + +#else /* Dear implementor, you have choices. For example consider defining get64 & * set64 thusly - * #define get64(var) read64(&(var)) + * #define get64(var) read64(&(var)) * #define set64(var,val) write64(&(var),val) * and get the JIT to generate read64 & write64 above atomic 64-bit read/write. */ # error atomic access of 64-bit variables not yet defined for this platform +#endif # endif #else /* LP32 || ILP32 else LP64 || ILP64 || LLP64 */ @@ -159,13 +175,20 @@ # define sqAtomicAddConst(var,n) \ asm volatile ("lock addl %1, %0" : "=m" (var) : "i" (n), "m" (var)) #endif +#elif defined TARGET_OS_IS_IPHONE +#define sqAtomicAddConst(var,n) OSAtomicAdd32(n,&var) +#elif defined(__arm__) && defined(__ARM_ARCH_6__) +/* tpr - this is code intended for the Raspberry Pi Raspbian OS */ +/* We'll experimentally use the gcc inbuilt functions detailed in + * http://gcc.gnu.org/onlinedocs/gcc-4.1.2/gcc/Atomic-Builtins.html + */ +#define sqAtomicAddConst(var,n) \ + __sync_fetch_and_add((int *)&var, n) #else -#ifdef TARGET_OS_IS_IPHONE -#define sqAtomicAddConst(var,n) OSAtomicAdd32(n,&var) -#endif /* Dear implementor, you have choices. Google atomic increment and you will * find a number of implementations for other architectures. */ +# error atomic increment of 32-bit variables not yet defined for this platfom #endif @@ -197,19 +220,29 @@ : "g"(old), "r"(new), "m"(var)\ : "memory", "%eax") -# define sqCompareAndSwapRes(var,old,new,res) \ +# define sqCompareAndSwapRes(var,old,new,res) \/tmp/ccf1WMiY.s: asm volatile ("movl %2, %%eax; lock cmpxchg %3, %0; movl %%eax, %1" \ : "=m"(var), "=g"(res) \ : "g"(old), "r"(new), "m"(var) \ : "memory", "%eax") #endif -#else -#if TARGET_OS_IS_IPHONE +#elif defined TARGET_OS_IS_IPHONE # define sqCompareAndSwap(var,old,new) OSAtomicCompareAndSwap32(old, new, &var) # define sqCompareAndSwapRes(var,old,new,res) res = var; OSAtomicCompareAndSwap32(old, new, &var) -#endif +#elif defined(__arm__) && defined(__ARM_ARCH_6__) +/* tpr - this is code intended for the Raspberry Pi Raspbian OS */ +/* We'll experimentally use the gcc inbuilt functions detailed in + * http://gcc.gnu.org/onlinedocs/gcc-4.1.2/gcc/Atomic-Builtins.html */ +# define sqCompareAndSwap(var,old,new) \ + __sync_bool_compare_and_swap(&(var), (old), (new)) + +# define sqCompareAndSwapRes(var,old,new,res) \ + __sync_val_compare_and_swap(&(var), (old), (new)) + +#else /* Dear implementor, you have choices. Google atomic increment and you will * find a number of implementations for other architectures. */ +# error atomic compare/swap of 32-bit variables not yet defined for this platfom #endif Modified: branches/Cog/platforms/Cross/vm/sqMemoryFence.h =================================================================== --- branches/Cog/platforms/Cross/vm/sqMemoryFence.h 2013-04-25 18:38:59 UTC (rev 2726) +++ branches/Cog/platforms/Cross/vm/sqMemoryFence.h 2013-04-26 23:56:18 UTC (rev 2727) @@ -47,7 +47,7 @@ # define sqLowLevelMFence() asm volatile ("mfence") # endif #else -# if defined(TARGET_OS_IS_IPHONE) +# if defined(TARGET_OS_IS_IPHONE) || (defined(__arm__) && defined(__ARM_ARCH_6__)) # define sqLowLevelMFence() __sync_synchronize() # elif !defined(sqLowLevelMFence) extern void sqLowLevelMFence(void); Modified: branches/Cog/platforms/unix/config/configure =================================================================== --- branches/Cog/platforms/unix/config/configure 2013-04-25 18:38:59 UTC (rev 2726) +++ branches/Cog/platforms/unix/config/configure 2013-04-26 23:56:18 UTC (rev 2727) @@ -23833,7 +23833,8 @@ echo "$as_me:$LINENO: result: $CFLAGS_32" >&5 echo "${ECHO_T}$CFLAGS_32" >&6 -CFLAGS="$CFLAGS_32 $CFLAGS -msse" +# CFLAGS="$CFLAGS_32 $CFLAGS -msse" +CFLAGS="$CFLAGS_32 $CFLAGS" echo "$as_me:$LINENO: checking whether byte ordering is bigendian" >&5 echo $ECHO_N "checking whether byte ordering is bigendian... $ECHO_C" >&6 Modified: branches/Cog/platforms/unix/config/configure.ac =================================================================== --- branches/Cog/platforms/unix/config/configure.ac 2013-04-25 18:38:59 UTC (rev 2726) +++ branches/Cog/platforms/unix/config/configure.ac 2013-04-26 23:56:18 UTC (rev 2727) @@ -265,8 +265,10 @@ AC_MODULE_LIB_PREFIX AC_64BIT_ARCH -CFLAGS="$CFLAGS_32 $CFLAGS -msse" +# tpr - try to keep msse out of it for Pi CFLAGS="$CFLAGS_32 $CFLAGS -msse" +CFLAGS="$CFLAGS_32 $CFLAGS" + AC_C_BYTEORDER AC_C_DOUBLE_ALIGNMENT Modified: branches/Cog/platforms/unix/vm/sqUnixITimerHeartbeat.c =================================================================== --- branches/Cog/platforms/unix/vm/sqUnixITimerHeartbeat.c 2013-04-25 18:38:59 UTC (rev 2726) +++ branches/Cog/platforms/unix/vm/sqUnixITimerHeartbeat.c 2013-04-26 23:56:18 UTC (rev 2727) @@ -164,6 +164,8 @@ || defined(i486) || defined(__i486) || defined (__i486__) \ || defined(intel) || defined(x86) || defined(i86pc) ) __asm__ __volatile__ ("rdtsc" : "=A"(value)); +#elif defined(__arm__) && defined(__ARM_ARCH_6__) + /* tpr - do nothing for now; needs input from eliot to decide further */ #else # error "no high res clock defined" #endif

1 0

Re: [squeak-dev] Re: [Vm-dev] Bug 956376 ??? Can not edit class comments
by Eliot Miranda 26 Apr '13

26 Apr '13

On Fri, Apr 26, 2013 at 9:25 AM, Chris Muller <asqueaker(a)gmail.com> wrote: > Frank are you saying that, under Ubuntu 13.04, after writing to a > particular place in a file, if you subsequently go back and _read_ > from that place, even using the same ReadWriteStream instance, that it > will not read back in what was just written until a #flush occurs? > > Hopefully that is not the case! > If it is, it is an ubuntu bug and not one we should work around, no? > On Fri, Apr 26, 2013 at 5:17 AM, Frank Shearar <frank.shearar(a)gmail.com> > wrote: > > On 25 April 2013 15:35, Frank Shearar <frank.shearar(a)gmail.com> wrote: > >> On 25 April 2013 13:54, David T. Lewis <lewis(a)mail.msen.com> wrote: > >>> > >>> On Thu, Apr 25, 2013 at 02:18:22PM +0200, Bert Freudenberg wrote: > >>>> > >>>> This didn't sound like a VM problem at first, but according to the > discussion it is: > >>>> > >>>> https://bugzilla.redhat.com/show_bug.cgi?id=956376 > >>>> > >>>> Any idea? > >>>> > >>> > >>> On my 64-bit Linux with an interpreter VM compiled in 64-bit, and using > >>> Squeak - no problem. With a 32-bit interpreter - no problem. With Cog - > >>> again, no problem. > >>> > >>> I edited the class comment for class Array in each case. > >>> > >>> Dave > >> > >> "Error: RemoteString past end of file" can indicate a broken/damages > >> changes file. > >> > >> It does when I get the error, at least. Usually that happens when I > >> have to hard-kill an image that's stored state in the changes file. > >> Restarting the image means there's a mismatch somehow in the state > >> between the image and the changes file. Once you get one of these > >> you're nearly completely hosed: you can't file things out, for > >> instance. > > > > For those not following the Pharo list, it looks like the root cause > > is that the glibc in Ubuntu 13.04 doesn't immediately write new > > content out to the file. Thus you can "write" to the file, and then > > read from the file, only your RemoteString starts off the end of the > > file. The fix is to make WriteStream >> #nextPutChunk: call "self > > flush". I've pushed a 4.5 fix to trunk, but have not yet done the same > > for the 4.4 and 4.3 update streams. > > > > frank > > > > -- best, Eliot

1 0

Call for ESUG 2013 presentations
by DeNigris Sean 26 Apr '13

26 Apr '13

Please distribute as widely as possible :) 21st International Smalltalk Joint Conference - Call for Contributions Annecy, France from 9 to 13 September 2013 http://www.esug.org/conferences/2013/ This call includes: Developer Forum Smalltalk Technology Award International Workshop http://www.esug.org/Conferences/2013/InternationalWorkshop Student Volunteer http://www.esug.org/Conferences/2013/StudentVolunteers Camp Smalltalk 8 September 2013 ---------------------------------------------------------------------- For the past 21 years, the European Smalltalk User Group (ESUG) has organised the International Smalltalk Conference, a lively forum on cutting edge software technologies that attract people from both academia and industry for a whole week. The attendees are both engineers using Smalltalk in business and students and teachers using Smalltalk both for research and didactic purposes. As every year, this year's edition of the largest European Smalltalk event will include the regular Smalltalk developers conference with renowned invited speakers, a Smalltalk camp that proves fruitful for interactions and discussions. This year will be held the 9th edition of the Innovation Technology Awards where prizes will be awarded to authors of best pieces of Smalltalk-related projects and an international workshop on Smalltalk and dynamic languages (http://www.esug.org/Conferences/2013/InternationalWorkshop) You can support the ESUG conference in many different ways: * Sponsor the conference. New sponsoring packages are described at http://www.esug.org/supportesug/becomeasponsor/ * Submit a talk, a software or a paper to one of the events. See below. * Attend the conference. We'd like to beat the previous record of attendance (170 people at Amsterdam 2008)! * Students can get free registration and hosting if they enrol into the the Student Volunteers program. See below. Developers Forum: International Smalltalk Developers Conference ------------------------------------------------------------------------ We are looking for YOUR experience on using Smalltalk. You will have 30 min for presentations and 45 min for hand-ons tutorial. The list of topics for the normal talks and tutorials includes, but is not limited to the following: * XP practices, Development tools, Experience reports * Model driven development, Web development, Team management * Meta-Modeling, Security, New libraries & frameworks * Educational material, Embedded systems and robotics * SOA and Web services, Interaction with other programming languages Teaching Pearls --------------- New this year!!! We want two news types of sessions in addition to default talk and show us your projects sessions. - Show your business 10 min session (Get prepared!!) - Teaching pearls : we want some session on how to teach some design aspects. We want your tip and tricks to teach Smalltalk or OOP. We expect to have several 10 to 15 min sessions aggregated. Open Forum: starting with Smalltalk ------------------------------------------------------------------------ New this year!! The event will be held at IAE Savoie Mont-Blanc - University of Savoie. This is an opportunity for students and local professionals to discover Smalltalk. A room will be freely accessible and we would like to propose some talks and workshop for Smalltalk beginners. The list of topics includes, but is not limited to the following: - Why Smalltalk ? - Panorama of Smalltalk solutions - Getting started with Smalltalk - Success stories How to submit? -------------- Submissions due on 15st June 2013 Notification of acceptance will be on done on the fly. More information at http://www.esug.org/conferences/2013 Pay attention: the places are limited so do not wait till the last minute to apply. Prospective presenters should submit a request to [hidden email] AND USE THE following header [ESUG 2013 Developers]. Please follow the template below the email will be automatically processed! Subject: [ESUG 2013 Developers] + your name First Name: Last Name: Email where you can always be reached: Title: Type: Tutorial/Talk/Teaching Pearl/OpenForum Abstract: Bio: Any presentation not respecting this form will be discarded automatically International Workshop on Smalltalk Technologies ------------------------------------------------------------------------ http://www.esug.org/Conferences/2013/InternationalWorkshop Smalltalk is considered as a design pearl and as a beacon in the realm of programming languages and programming environments. We are proud to invite submissions to the International Workshop on Smalltalk Technologies which is held as part of the ESUG 2013 joint event. The goals of the workshop is to create a forum around advances or experience in Smalltalk. We welcome contributions on all aspects, theoretical as well as practical, of Smalltalk related topics such as - aspect-oriented programming, meta-programming, - frameworks, interaction with other languages, - implementation, - new dialects or languages implemented in Smalltalk, tools, - meta-modeling, design patterns, experience reports Technology Award Competition ------------------------------------------------------------------------ The top 3 teams with the most innovative software will receive, respectively, 500 Euros, 300 Euros and 200 Euros during an awards ceremony at the conference. Developers of any Smalltalk-based software are welcome to compete. More information at http://www.esug.org/conferences/2013/ Student Volunteer Program ------------------------------------------------------------------------ If you are a student wanting to attend ESUG, have you considered being a student volunteer? Student volunteers help keep the conference running smoothly; in return, they have free accommodations, while still having most of the time to enjoy the conference. More information at http://www.esug.org/conferences/2013 http://www.esug.org/Conferences/2013/StudentVolunteers We hope to see you there and have fun together.

1 0

Re: [Pharo-project] RE : cannot modify class comments on Linux
by David T. Lewis 26 Apr '13

26 Apr '13

On Fri, Apr 26, 2013 at 12:44:49PM +0200, Igor Stasenko wrote: > On 26 April 2013 10:27, Pavel Krivanek <pavel.krivanek(a)gmail.com> wrote: > > The reason is in the optimalizations in the latest glibc. It really does not > > write the new content to the changes file immediately. We need to call flush > > at the end of nextChunkPut: > > > This is nasty one. > What i worry that it affects much more than code working with .changes file. > Indeed, I expect that a bug like this would effect *anything* that involves writing to a positionable file. It's not clear to me from the discussion if this is a bug in the runtime C library, or in the gcc optimizer (sic). Is it the case that a pre-compiled VM that works properly on one Linux platform will have this bug if run on another Linux platform that has a different runtime clib? Or is it the case that a VM compiled from source on one platform behaves differently than the same VM compiled from source on a different Linux platform? Dave > > > Cheers, > > -- Pavel > > > > > > On Fri, Apr 26, 2013 at 9:57 AM, GOUBIER Thierry <thierry.goubier(a)cea.fr> > > wrote: > >> > >> Reproduced on Ubuntu 13.04. > >> > >> Class organization remoteString filePositionHi is past the end of the > >> changes file (i.e. exactly equal to the file size if the comment is the last > >> operation). > >> > >> Thierry > >> ________________________________ > >> De : pharo-project-bounces(a)lists.gforge.inria.fr > >> [pharo-project-bounces(a)lists.gforge.inria.fr] de la part de Pavel Krivanek > >> [pavel.krivanek(a)gmail.com] > >> Date d'envoi : jeudi 25 avril 2013 21:32 > >> ?? : Pharo-project(a)lists.gforge.inria.fr > >> Objet : Re: [Pharo-project] cannot modify class comments on Linux > >> > >> Jaroslav wrote this: > >> This does work on F17, but it doesn't work on F18, F19. Maybe glibc issue? > >> I guess the reproducer needs glibc-2.16 and up, IMHO 12.04 has 2.15 same > >> as F17, so I guess this would be also reproducible on Ubuntu 13.04. > >> > >> -- Pavel > >> > >> On Thu, Apr 25, 2013 at 11:16 AM, Gisela Decuzzi <giseladecuzzi(a)gmail.com> > >> wrote: > >>> > >>> I'm on 64-bit Ubuntu 12.04 > >>> > >>> > >>> 2013/4/25 Pavel Krivanek <pavel.krivanek(a)gmail.com> > >>>> > >>>> Hi, > >>>> > >>>> what architecture and distro do you use because I tried it on a 32-bit > >>>> Debian and it worked. I'm on 64-bit Fedora 18. That may be the reason why > >>>> the tests are not failing on CI. This error appears on all old system > >>>> classes and on newely created classes too. > >>>> Some next info: https://bugzilla.redhat.com/show_bug.cgi?id=956376 > >>>> > >>>> -- Pavel > >>>> > >>>> On Thu, Apr 25, 2013 at 10:58 AM, Gisela Decuzzi > >>>> <giseladecuzzi(a)gmail.com> wrote: > >>>>> > >>>>> Hi Pavel, I tried in Pharo 2.0 and 3.0 and it work well for me. (just > >>>>> try in a random class comment) > >>>>> Are you trying to modify a particular class comment? > >>>>> > >>>>> > >>>>> 2013/4/25 Pavel Krivanek <pavel.krivanek(a)gmail.com> > >>>>>> > >>>>>> Hi, > >>>>>> > >>>>>> if you will try to modify a class comment on Linux, the error > >>>>>> "RemoteString past end of file" will appear. On windows VM it seems to work > >>>>>> well. Can someone confirm this behavior on Linux. Squeak has the same > >>>>>> problem. > >>>>>> > >>>>>> Cheers, > >>>>>> -- Pavel > >>>>> > >>>>> > >>>> > >>> > >> > > > > > > -- > Best regards, > Igor Stasenko.

1 0

Bug 956376 – Can not edit class comments
by Bert Freudenberg 26 Apr '13

26 Apr '13

This didn't sound like a VM problem at first, but according to the discussion it is: https://bugzilla.redhat.com/show_bug.cgi?id=956376 Any idea? - Bert -

3 3

[commit][2726] Ugh. Make the !defined(SA_NODEFER) compare-and-swap code comprehensible (i hope).
by commits＠squeakvm.org 25 Apr '13

25 Apr '13

Revision: 2726 Author: eliot Date: 2013-04-25 11:38:59 -0700 (Thu, 25 Apr 2013) Log Message: ----------- Ugh. Make the !defined(SA_NODEFER) compare-and-swap code comprehensible (i hope). Modified Paths: -------------- branches/Cog/platforms/unix/vm/sqUnixITimerHeartbeat.c branches/Cog/platforms/unix/vm/sqUnixITimerTickerHeartbeat.c Modified: branches/Cog/platforms/unix/vm/sqUnixITimerHeartbeat.c =================================================================== --- branches/Cog/platforms/unix/vm/sqUnixITimerHeartbeat.c 2013-04-25 18:34:40 UTC (rev 2725) +++ branches/Cog/platforms/unix/vm/sqUnixITimerHeartbeat.c 2013-04-25 18:38:59 UTC (rev 2726) @@ -350,10 +350,10 @@ heartbeat_handler(int sig, struct siginfo *sig_info, void *context) { #if !defined(SA_NODEFER) - { int zeroAndPreviousHandlingHeartbeat = 0; - sqCompareAndSwapRes(handling_heartbeat,zeroAndPreviousHandlingHeartbeat, - 1,zeroAndPreviousHandlingHeartbeat); - if (zeroAndPreviousHandlingHeartbeat) + { int zero = 0; + int previouslyHandlingHeartbeat; + sqCompareAndSwapRes(handling_heartbeat,zero,1,previouslyHandlingHeartbeat); + if (previouslyHandlingHeartbeat) return; } Modified: branches/Cog/platforms/unix/vm/sqUnixITimerTickerHeartbeat.c =================================================================== --- branches/Cog/platforms/unix/vm/sqUnixITimerTickerHeartbeat.c 2013-04-25 18:34:40 UTC (rev 2725) +++ branches/Cog/platforms/unix/vm/sqUnixITimerTickerHeartbeat.c 2013-04-25 18:38:59 UTC (rev 2726) @@ -480,10 +480,10 @@ } #if !defined(SA_NODEFER) - { int zeroAndPreviousHandlingHeartbeat = 0; - sqCompareAndSwapRes(handling_heartbeat,zeroAndPreviousHandlingHeartbeat, - 1,zeroAndPreviousHandlingHeartbeat); - if (zeroAndPreviousHandlingHeartbeat) + { int zero = 0; + int previouslyHandlingHeartbeat; + sqCompareAndSwapRes(handling_heartbeat,zero,1,previouslyHandlingHeartbeat); + if (previouslyHandlingHeartbeat) return; }

1 0

[commit][2725] Fix typos in -help on unix.
by commits＠squeakvm.org 25 Apr '13

25 Apr '13

Revision: 2725 Author: eliot Date: 2013-04-25 11:34:40 -0700 (Thu, 25 Apr 2013) Log Message: ----------- Fix typos in -help on unix. Fix use of sqCompareAndSwap in !defined(SA_NODEFER) regime in ticker heartbeats. Comment and better define iPhone imp of sqCompareAndSwapRes. Modified Paths: -------------- branches/Cog/platforms/Cross/vm/sqAtomicOps.h branches/Cog/platforms/unix/vm/sqUnixITimerHeartbeat.c branches/Cog/platforms/unix/vm/sqUnixITimerTickerHeartbeat.c branches/Cog/platforms/unix/vm/sqUnixMain.c Modified: branches/Cog/platforms/Cross/vm/sqAtomicOps.h =================================================================== --- branches/Cog/platforms/Cross/vm/sqAtomicOps.h 2013-04-19 21:24:17 UTC (rev 2724) +++ branches/Cog/platforms/Cross/vm/sqAtomicOps.h 2013-04-25 18:34:40 UTC (rev 2725) @@ -188,7 +188,8 @@ #ifdef TARGET_OS_IS_IPHONE # define sqCompareAndSwap(var,old,new) OSAtomicCompareAndSwap32(old, new, &var) -# define sqCompareAndSwapRes(var,old,new,res) res = var; OSAtomicCompareAndSwap32(old, new, &var) +/* N.B. This is not atomic in fetching var's old value :( */ +# define sqCompareAndSwapRes(var,old,new,res) do { res = var; if (OSAtomicCompareAndSwap32(old, new, &var)) res = new; } while (0) #else # define sqCompareAndSwap(var,old,new) \ asm volatile ("movl %1, %%eax; lock cmpxchg %2, %0" \ Modified: branches/Cog/platforms/unix/vm/sqUnixITimerHeartbeat.c =================================================================== --- branches/Cog/platforms/unix/vm/sqUnixITimerHeartbeat.c 2013-04-19 21:24:17 UTC (rev 2724) +++ branches/Cog/platforms/unix/vm/sqUnixITimerHeartbeat.c 2013-04-25 18:34:40 UTC (rev 2725) @@ -351,7 +351,8 @@ { #if !defined(SA_NODEFER) { int zeroAndPreviousHandlingHeartbeat = 0; - sqCompareAndSwap(handling_heartbeat,zeroAndPreviousHandlingHeartbeat,1); + sqCompareAndSwapRes(handling_heartbeat,zeroAndPreviousHandlingHeartbeat, + 1,zeroAndPreviousHandlingHeartbeat); if (zeroAndPreviousHandlingHeartbeat) return; } Modified: branches/Cog/platforms/unix/vm/sqUnixITimerTickerHeartbeat.c =================================================================== --- branches/Cog/platforms/unix/vm/sqUnixITimerTickerHeartbeat.c 2013-04-19 21:24:17 UTC (rev 2724) +++ branches/Cog/platforms/unix/vm/sqUnixITimerTickerHeartbeat.c 2013-04-25 18:34:40 UTC (rev 2725) @@ -481,7 +481,8 @@ #if !defined(SA_NODEFER) { int zeroAndPreviousHandlingHeartbeat = 0; - sqCompareAndSwap(handling_heartbeat,zeroAndPreviousHandlingHeartbeat,1); + sqCompareAndSwapRes(handling_heartbeat,zeroAndPreviousHandlingHeartbeat, + 1,zeroAndPreviousHandlingHeartbeat); if (zeroAndPreviousHandlingHeartbeat) return; } Modified: branches/Cog/platforms/unix/vm/sqUnixMain.c =================================================================== --- branches/Cog/platforms/unix/vm/sqUnixMain.c 2013-04-19 21:24:17 UTC (rev 2724) +++ branches/Cog/platforms/unix/vm/sqUnixMain.c 2013-04-25 18:34:40 UTC (rev 2725) @@ -1477,12 +1477,12 @@ printf(" -jit enable the dynamic compiler (if available)\n"); # endif printf(" -notimer disable interval timer for low-res clock\n"); - printf(" -display <dpy> quivalent to '-vm-display-X11 -display <dpy>'\n"); - printf(" -headless quivalent to '-vm-display-X11 -headless'\n"); - printf(" -nodisplay quivalent to '-vm-display-null'\n"); + printf(" -display <dpy> equivalent to '-vm-display-X11 -display <dpy>'\n"); + printf(" -headless equivalent to '-vm-display-X11 -headless'\n"); + printf(" -nodisplay equivalent to '-vm-display-null'\n"); printf(" -nomixer disable modification of mixer settings\n"); - printf(" -nosound quivalent to '-vm-sound-null'\n"); - printf(" -quartz quivalent to '-vm-display-Quartz'\n"); + printf(" -nosound equivalent to '-vm-sound-null'\n"); + printf(" -quartz equivalent to '-vm-display-Quartz'\n"); #endif }

1 0

Heartbeat & Linux
by Casey Ransberger 24 Apr '13

24 Apr '13

I'm assuming the problem is in the kernel, but making assumptions is usually a bad plan, so I'm asking. When I look, I find a rather confusing jumble: pthreads exists in different implementations under different operating systems. I'd kind of like the heartbeat to work right under Raspbian, because kids are going to use it, and the goal of producing better adults is pretty close to my heart. See also making Squeak faster. (Stack VM.) I want to arm myself with as much knowledge about this as possible, because I'm considering just bloody fixing it. If the (GNU/Linux) community has objections, and I'm able to pull off the fix, I can make a patch available to people who want it. I'm not confident that I can fix the bug with the knowledge I have now, but I'm confident that I can fix it eventually, if someone else doesn't beat me to the punch. Outside of the wiki page, are there any documents/people/mailing lists I should look at to be able to get a good handle on where in the kernel (or elsewhere) I should be looking for the code which isn't meeting our Cog-nitive requirements? TIA, C

4 11

Re: [squeak-dev] The Trunk: Graphics-fbs.209.mcz
by Eliot Miranda 24 Apr '13

24 Apr '13

Hi Frank, I need to be able to reproduce this. The error message means what it says; some code in the VM (primitives usually) has written beyond the end of an object, which, if it weren't happening to the last object in memory would smash the following object, crashing the GC. So when this happens please attempt to create an image that reproduces the bug and then I can track down the offending overwrite. The VM fills the nursery zone with a specific bit pattern (each 4-byte word contains its address) and checks on allocation that the allocation pointer is pointing at a word with the right contents. If not, it aborts. In this case the effects are benign (because the overwrite is writing beyond the allocation pointer where there are no objects). But if whatever code overwrote did so to an object in the middle of the heap a less easily diagnosed crash would occur. On Tue, Apr 23, 2013 at 2:34 AM, Frank Shearar <frank.shearar(a)gmail.com>wrote: > On 23 April 2013 09:50, <commits(a)source.squeak.org> wrote: > > Frank Shearar uploaded a new version of Graphics to project The Trunk: > > http://source.squeak.org/trunk/Graphics-fbs.209.mcz > > > > ==================== Summary ==================== > > > > Name: Graphics-fbs.209 > > Author: fbs > > Time: 23 April 2013, 9:50:00.413 am > > UUID: 4dd577f2-2a70-4e21-8efa-5847b3ad804c > > Ancestors: Graphics-tpr.208 > > > > #asGrafPort belongs in the Morphic package. > > > > =============== Diff against Graphics-tpr.208 =============== > > > > Item was removed: > > - ----- Method: BitBlt class>>asGrafPort (in category 'instance > creation') ----- > > - asGrafPort > > - "Return the GrafPort associated with the receiver" > > - ^GrafPort! > > I don't see why, but this and its pair Morphic-fbs-648 seem to have > triggered a nasty bug in Cog r.2714. > > To trigger it, > * clone https://github.com/frankshearar/squeak-ci/ > * open the Squeak4.5.image therein, with a (Linux) Cog r.2714 > * update it. > > Everything proceeds fine until installing this update. On my weak > little laptop things pause, and then I see the dreaded "(last object > overwritten)" message. On my much more powerful work machine (also a > Linux box) it just pauses. Squeak's burning a hole in the CPU, and > updating is uninterruptable. > > On my work machine, upgrading using an Interpreter VM just leaves me > hanging. > > <cog's output follows:> > last object overwritten > > Squeak VM version: 4.0-2714 #1 Thu Apr 4 11:52:39 PDT 2013 gcc 4.1.2 > Built from: CoInterpreter VMMaker.oscog-eem.282 uuid: > 6c2f2048-555b-4d37-a9eb-4d9252acf3b9 Apr 4 2013 > With: StackToRegisterMappingCogit VMMaker.oscog-eem.282 uuid: > 6c2f2048-555b-4d37-a9eb-4d9252acf3b9 Apr 4 2013 > Revision: VM: r2714 http://www.squeakvm.org/svn/squeak/branches/Cog > Plugins: r2545 > http://squeakvm.org/svn/squeak/trunk/platforms/Cross/plugins > Build host: Linux mcqfes 2.6.18-128.el5 #1 SMP Wed Jan 21 10:44:23 EST > 2009 i686 i686 i386 GNU/Linux > plugin path: target/cog.r2714/coglinux/bin/../lib/squeak/4.0-2714 > [default: > /home/frank/Documents/squeak-ci/target/cog.r2714/coglinux/lib/squeak/4.0-2714/] > > > C stack backtrace: > target/cog.r2714/coglinux/bin/../lib/squeak/4.0-2714/squeak[0x805cca1] > > target/cog.r2714/coglinux/bin/../lib/squeak/4.0-2714/squeak(error+0x19)[0x805cf89] > > target/cog.r2714/coglinux/bin/../lib/squeak/4.0-2714/squeak(instantiateClassindexableSize+0xa9)[0x807a159] > target/cog.r2714/coglinux/bin/../lib/squeak/4.0-2714/squeak[0x807b21b] > [0x77745451] > > target/cog.r2714/coglinux/bin/../lib/squeak/4.0-2714/squeak(interpret+0x1eb)[0x808080b] > > target/cog.r2714/coglinux/bin/../lib/squeak/4.0-2714/squeak(main+0x397)[0x805d367] > /lib/tls/i686/cmov/libc.so.6(__libc_start_main+0xe6)[0x12abd6] > target/cog.r2714/coglinux/bin/../lib/squeak/4.0-2714/squeak[0x805ad61] > > > Smalltalk stack dump: > 0xbfe2ad38 M Bitmap class(Behavior)>new: 0x77b25270: a(n) Bitmap class > 0xbfe2ad54 M Form>setExtent:depth: 0x7ff6e7a0: a(n) Form > 0xbfe2ad74 M Form class>extent:depth: 0x77b27480: a(n) Form class > 0xbfe2ada0 M DisplayText>composeForm 0x7ff6d8fc: a(n) DisplayText > 0xbfe2adb8 M DisplayText>form 0x7ff6d8fc: a(n) DisplayText > 0xbfe2add8 M DisplayText>displayOn:at:clippingBox:rule:fillColor: > 0x7ff6d8fc: a(n) DisplayText > 0xbfe2ae04 M DisplayText(DisplayObject)>displayOn:at: 0x7ff6d8fc: a(n) > DisplayText > 0xbfe2ae24 M ByteString(String)>displayOn:at:textColor: 0x7ff6d450: > a(n) ByteString > 0xbfe81ca0 M ByteString(String)>displayOn:at: 0x7ff6d450: a(n) ByteString > 0xbfe81cc0 M ByteString(String)>displayAt: 0x7ff6d450: a(n) ByteString > 0xbfe81cdc M Debugger class(Object)>primitiveError: 0x77dc864c: a(n) > Debugger class > 0xbfe81cfc M [] in Debugger > class>morphicOpenOn:context:label:contents:fullView: 0x77dc864c: a(n) > Debugger class > 0xbfe81d18 M BlockClosure>cull: 0x7ff577ac: a(n) BlockClosure > 0xbfe81d38 M [] in MethodContext(ContextPart)>handleSignal: > 0x7ff6cc68: a(n) MethodContext > 0xbfe81d58 M BlockClosure>ensure: 0x7ff6cd24: a(n) BlockClosure > 0xbfe81d78 M MethodContext(ContextPart)>handleSignal: 0x7ff6cc68: a(n) > MethodContext > 0xbfe81d94 M MessageNotUnderstood(Exception)>signal 0x7ff6cad4: a(n) > MessageNotUnderstood > 0xbfe81db4 M BitBlt class(Object)>doesNotUnderstand: asGrafPort > 0x77b22b48: a(n) BitBlt class > 0xbfe81dd0 M FormCanvas>portClass 0x7ff5ca70: a(n) FormCanvas > 0xbfe81de8 M FormCanvas>setForm: 0x7ff5ca70: a(n) FormCanvas > 0xbfe81e04 M FormCanvas class>extent:depth: 0x77b2d750: a(n) FormCanvas > class > 0xbfe81e28 M RectangleMorph(Morph)>imageForm:forRectangle: 0x7ff5c550: > a(n) RectangleMorph > 0xbfe39d14 M RectangleMorph(Morph)>imageFormForRectangle: 0x7ff5c550: > a(n) RectangleMorph > 0xbfe39d30 M RectangleMorph(Morph)>imageForm 0x7ff5c550: a(n) > RectangleMorph > 0xbfe39d5c M ScrollBar class>createArrowOfDirection:size:color: > 0x77b29268: a(n) ScrollBar class > 0xbfe39d80 M [] in ScrollBar class>createArrowImagesCache 0x77b29268: > a(n) ScrollBar class > 0xbfe39db0 M LRUCache>at: 0x785bcddc: a(n) LRUCache > 0xbfe39dcc M ScrollBar class>arrowOfDirection:size:color: 0x77b29268: > a(n) ScrollBar class > 0xbfe39df0 M ScrollBar>upImage 0x7ff5b408: a(n) ScrollBar > 0xbfe39e10 M ScrollBar>updateUpButtonImage 0x7ff5b408: a(n) ScrollBar > 0xbfe39e2c M ScrollBar>sliderColor: 0x7ff5b408: a(n) ScrollBar > 0xbfe44cf8 I ScrollBar>adoptPaneColor: 0x7ff5b408: a(n) ScrollBar > 0xbfe44d18 M [] in PluggableListMorphPlus(Morph)>adoptPaneColor: > 0x7ff5ab30: a(n) PluggableListMorphPlus > 0xbfe44d3c M Array(SequenceableCollection)>do: 0x7ff5bea4: a(n) Array > 0xbfe44d58 M PluggableListMorphPlus(Morph)>submorphsDo: 0x7ff5ab30: > a(n) PluggableListMorphPlus > 0xbfe44d74 M PluggableListMorphPlus(Morph)>adoptPaneColor: 0x7ff5ab30: > a(n) PluggableListMorphPlus > 0xbfe44d90 M PluggableListMorphPlus(ScrollPane)>adoptPaneColor: > 0x7ff5ab30: a(n) PluggableListMorphPlus > 0xbfe44dc0 M PluggableSystemWindow(SystemWindow)>addMorph:fullFrame: > 0x7ff5803c: a(n) PluggableSystemWindow > 0xbfe44de0 M PluggableSystemWindow>addPaneMorph: 0x7ff5803c: a(n) > PluggableSystemWindow > 0xbfe44dfc M MorphicToolBuilder>add:to: 0x7ff57880: a(n) MorphicToolBuilder > 0xbfe44e2c M MorphicToolBuilder>buildPluggableList: 0x7ff57880: a(n) > MorphicToolBuilder > 0xbfe6ccec M PluggableListSpec>buildWith: 0x7ff57b74: a(n) > PluggableListSpec > 0xbfe6cd08 M [] in MorphicToolBuilder(ToolBuilder)>buildAll:in: > 0x7ff57880: a(n) > MorphicToolBuilder > 0xbfe6cd28 M OrderedCollection>do: 0x7ff578dc: a(n) OrderedCollection > 0xbfe6cd48 M MorphicToolBuilder(ToolBuilder)>buildAll:in: 0x7ff57880: > a(n) MorphicToolBuilder > 0xbfe6cd74 M MorphicToolBuilder>buildPluggableWindow: 0x7ff57880: a(n) > MorphicToolBuilder > 0xbfe6cd90 M PluggableWindowSpec>buildWith: 0x7ff57898: a(n) > PluggableWindowSpec > 0xbfe6cdb4 I MorphicToolBuilder(ToolBuilder)>build: 0x7ff57880: a(n) > MorphicToolBuilder > 0xbfe6cddc I MorphicToolBuilder>open: 0x7ff57880: a(n) MorphicToolBuilder > 0xbfe6ce00 I MorphicToolBuilder(ToolBuilder)>openDebugger: 0x7ff57880: > a(n) MorphicToolBuilder > 0xbfe6ce28 M Debugger>openNotifierContents:label: 0x7ff577c4: a(n) Debugger > 0xbfea4d08 M [] in Debugger > class>morphicOpenOn:context:label:contents:fullView: 0x77dc864c: a(n) > Debugger class > 0xbfea4d24 M BlockClosure>on:do: 0x7ff5777c: a(n) BlockClosure > 0xbfea4d60 M [] in Debugger > class>morphicOpenOn:context:label:contents:fullView: 0x77dc864c: a(n) > Debugger class > 0xbfea4d88 M WorldState>runStepMethodsIn: 0x77834af8: a(n) WorldState > 0xbfea4da4 M PasteUpMorph>runStepMethods 0x77c6a5ec: a(n) PasteUpMorph > 0xbfea4dc0 M WorldState>doOneCycleNowFor: 0x77834af8: a(n) WorldState > 0xbfea4ddc M WorldState>doOneCycleFor: 0x77834af8: a(n) WorldState > 0xbfea4df8 M PasteUpMorph>doOneCycle 0x77c6a5ec: a(n) PasteUpMorph > 0xbfea4e10 M [] in Project class>spawnNewProcess 0x77c54960: a(n) Project > class > 0xbfea4e30 I [] in BlockClosure>newProcess 0x7ff557f4: a(n) BlockClosure > > Most recent primitives > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > perform: > at:put: > @ > @ > basicNew > new: > at:put: > basicNew > shallowCopy > shallowCopy > at:put: > basicScanCharactersFrom:to:in:rightX:stopConditions:kern: > new: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > perform: > basicScanCharactersFrom:to:in:rightX:stopConditions:kern: > new: > at:put: > at:put: > at:put: > at:put: > perform: > basicScanCharactersFrom:to:in:rightX:stopConditions:kern: > new: > at:put: > at:put: > at:put: > at:put: > perform: > basicScanCharactersFrom:to:in:rightX:stopConditions:kern: > new: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > perform: > basicScanCharactersFrom:to:in:rightX:stopConditions:kern: > new: > at:put: > at:put: > at:put: > at:put: > perform: > basicScanCharactersFrom:to:in:rightX:stopConditions:kern: > new: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > perform: > basicScanCharactersFrom:to:in:rightX:stopConditions:kern: > new: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > perform: > at:put: > @ > @ > basicNew > new: > at:put: > basicNew > shallowCopy > shallowCopy > at:put: > basicScanCharactersFrom:to:in:rightX:stopConditions:kern: > new: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > perform: > basicScanCharactersFrom:to:in:rightX:stopConditions:kern: > new: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > perform: > basicScanCharactersFrom:to:in:rightX:stopConditions:kern: > new: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > perform: > basicScanCharactersFrom:to:in:rightX:stopConditions:kern: > new: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > perform: > basicScanCharactersFrom:to:in:rightX:stopConditions:kern: > new: > at:put: > at:put: > at:put: > at:put: > perform: > basicScanCharactersFrom:to:in:rightX:stopConditions:kern: > new: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > at:put: > perform: > at:put: > species > new: > replaceFrom:to:with:startingAt: > @ > @ > basicNew > @ > @ > @ > @ > @ > @ > perform:with: > @ > @ > species > species > at:put: > @ > basicNew > basicNew: > > stack page bytes 4096 available headroom 3300 minimum unused headroom 2228 > > (last object overwritten) > > frank > > -- best, Eliot

2 1

[ANN] Pharo VM for Ubuntu moved!
by Damien Cassou 24 Apr '13

24 Apr '13

After a request from Esteban, I changed the Pharo VM PPA location from ppa:cassou/pharo to ppa:pharo/stable. Please issue the following commands to update your system: $ sudo rm -f /etc/apt/sources.list.d/cassou-emacs-quantal.list $ sudo add-apt-repository ppa:pharo/stable $ sudo apt-get update Sorry for the additional effort -- Damien Cassou http://damiencassou.seasidehosting.st "Success is the ability to go from one failure to another without losing enthusiasm." Winston Churchill

1 0

← Newer
1
2
3
4
5
6
7
8
...
11
Older →

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

Vm-dev April 2013