I try to update a 3.7 alpha image (5423). While loading update 5501 the vm crashes. Is anybody seeing this too?
Alex
I try to update a 3.7 alpha image (5423). While loading update 5501 the vm crashes. Is anybody seeing this too?
I guess you need a 3.6vm for that... I actually didn't think about the possibility that this crashes older vm (as the resulting image runs fine on older vms.). That's bad... what do we do? We really need to rebuild the specialObjectsArray. maybe this cs should check the VM version?
Marcus
On Wed, Oct 29, 2003 at 01:36:50PM +0100, Marcus Denker wrote:
I try to update a 3.7 alpha image (5423). While loading update 5501 the vm crashes. Is anybody seeing this too?
I guess you need a 3.6vm for that... I actually didn't think about the possibility that this crashes older vm (as the resulting image runs fine on older vms.). That's bad... what do we do? We really need to rebuild the specialObjectsArray. maybe this cs should check the VM version?
Ups, totally wrong idea: Alex is using a current vm (self-compiled)
3.6g-2 #1 Mi Okt 29 12:49:15 CET 2003 gcc-Version (Debian) Squeak3.6beta of '4 July 2003' [latest update: #5411] Linux renaud 2.4.22 #1 Do Sep 25 17:09:04 CEST 2003 i686 GNU/Linux default plugin location: /usr/local/lib/squeak/3.6g-2/*.so
(log is attached to his mail)
Marcus
Am Mittwoch, 29. Oktober 2003 15:14 schrieb Marcus:
Ups, totally wrong idea: Alex is using a current vm (self-compiled)
The same happens when I use the precompiled binary.
3.6g-2 #2 Tue Sep 16 20:16:57 CEST 2003 gcc 3.3.1 Squeak3.6beta of '4 July 2003' [latest update: #5411] Linux xombul.inria.fr 2.4.19 #2 Mon Aug 5 12:44:11 CEST 2002 i686 GNU/Linux default plugin location: /usr/lib/squeak/3.6g-2/*.so
Alex
On Wed, Oct 29, 2003 at 04:33:14PM +0100, Alexander Lazarevi? wrote: Content-Description: signed data
Am Mittwoch, 29. Oktober 2003 15:14 schrieb Marcus:
Ups, totally wrong idea: Alex is using a current vm (self-compiled)
The same happens when I use the precompiled binary.
3.6g-2 #2 Tue Sep 16 20:16:57 CEST 2003 gcc 3.3.1 Squeak3.6beta of '4 July 2003' [latest update: #5411] Linux xombul.inria.fr 2.4.19 #2 Mon Aug 5 12:44:11 CEST 2002 i686 GNU/Linux default plugin location: /usr/lib/squeak/3.6g-2/*.so
Just a short note: Alex just mailed me that everything works fine if you build the VM using the latest VMMaker...
Marcus
I just noticed this, too. I have a self-compiled 3.6g VM, running on Linux on an Athlon. The update goes fine on Windows.
Joshua
On Wed, Oct 29, 2003 at 01:09:28PM +0100, Alexander Lazarevi? wrote: Content-Description: signed data Content-Description: body text
I try to update a 3.7 alpha image (5423). While loading update 5501 the vm crashes. Is anybody seeing this too?
Alex
3.6g-2 #1 Mi Okt 29 12:49:15 CET 2003 gcc-Version (Debian) Squeak3.6beta of '4 July 2003' [latest update: #5411] Linux renaud 2.4.22 #1 Do Sep 25 17:09:04 CEST 2003 i686 GNU/Linux default plugin location: /usr/local/lib/squeak/3.6g-2/*.so
Am Mittwoch, 29. Oktober 2003 17:07 schrieb Joshua 'Schwa' Gargus:
I just noticed this, too. I have a self-compiled 3.6g VM, running on Linux on an Athlon. The update goes fine on Windows.
I see the crash happening at the beginning of the gc mark phase when the ephemeron queue gets initialized. Shouldn't that code be the same on linux and windows?
Alex
/* Mark phase of the mark and sweep garbage collector. Set the mark bits of all reachable objects. Free chunks are untouched by this process. Assume: All non-free objects are initially unmarked. Root objects were unmarked when they were made roots. (Make sure this stays true!!). */
int markPhase(void) { register struct foo * foo = &fum; int oop; int i; int classFormat; int keyMarked; int anyNew; int oop1; int ephemeron; int ephemeronKey; int nilOop; int header;
foo->freeContexts = NilContext;
/* initialize the ephemeron queue */
foo->freeLargeContexts = NilContext; foo->lastEphemeron = foo->nilObj; foo->theEphemeronKeyOffset = 0; if ((lengthOf(foo->specialObjectsOop)) >= ClassEphemeron) { foo->theClassEphemeron = longAt(((((char *) foo->specialObjectsOop)) + BaseHeaderSize) + (ClassEphemeron << 2)); /**** BOOOM ****/
On Wed, Oct 29, 2003 at 05:56:09PM +0100, Alexander Lazarevi? wrote: Content-Description: signed data
Am Mittwoch, 29. Oktober 2003 17:07 schrieb Joshua 'Schwa' Gargus:
I just noticed this, too. I have a self-compiled 3.6g VM, running on Linux on an Athlon. The update goes fine on Windows.
I see the crash happening at the beginning of the gc mark phase when the ephemeron queue gets initialized. Shouldn't that code be the same on linux and windows?
I couldn't say. However, my Windows VM version is 3.5.1, from June 21, 2003. I don't think that Andreas had included the ephemeron support at that time (or does time fly that fast?)
Joshua
Alex
/* Mark phase of the mark and sweep garbage collector. Set the mark bits of all reachable objects. Free chunks are untouched by this process. Assume: All non-free objects are initially unmarked. Root objects were unmarked when they were made roots. (Make sure this stays true!!). */
int markPhase(void) { register struct foo * foo = &fum; int oop; int i; int classFormat; int keyMarked; int anyNew; int oop1; int ephemeron; int ephemeronKey; int nilOop; int header;
foo->freeContexts = NilContext;
/* initialize the ephemeron queue */
foo->freeLargeContexts = NilContext; foo->lastEphemeron = foo->nilObj; foo->theEphemeronKeyOffset = 0; if ((lengthOf(foo->specialObjectsOop)) >= ClassEphemeron) { foo->theClassEphemeron = longAt(((((char *) foo->specialObjectsOop)) + BaseHeaderSize) + (ClassEphemeron << 2)); /**** BOOOM ****/
I did a new compile run of the 3.6g-2 sources with a fresh VMM version and generated sources. The resulting VM does not crash during update.
Alex
Aargh. I'll have to do the same, I guess. What version of Squeak did you generate the sources from?
Joshua
On Fri, Oct 31, 2003 at 11:56:07AM +0100, Alexander Lazarevi? wrote: Content-Description: signed data
I did a new compile run of the 3.6g-2 sources with a fresh VMM version and generated sources. The resulting VM does not crash during update.
Alex
I used
http://www-sor.inria.fr/~piumarta/squeak/devel/dist/Squeak-3.6g-2.src.tar.gz http://www-sor.inria.fr/~piumarta/squeak/devel/dist/Squeak-3.6g-5420.image.t... http://sumeru.stanford.edu/tim/pooters/SqFiles/packages/VMMaker/VMMaker3-6g2...
Alex
Am Freitag, 31. Oktober 2003 16:14 schrieb Joshua 'Schwa' Gargus:
Aargh. I'll have to do the same, I guess. What version of Squeak did you generate the sources from?
Joshua
On Fri, Oct 31, 2003 at 11:56:07AM +0100, Alexander Lazarevi? wrote: Content-Description: signed data
I did a new compile run of the 3.6g-2 sources with a fresh VMM version and generated sources. The resulting VM does not crash during update.
Alex
squeak-dev@lists.squeakfoundation.org
-
Alexander Lazarević -
Joshua 'Schwa' Gargus -
Marcus -
Marcus Denker