Hi all,
I have recently upgraded a Linux machine to Fedora 9 (it was Fedora 8) and the VW engine now crashes on startup (even before the splash screen) with: VisualWorks(R) 7.5 Non-Commercial Mar 3, 2008 src/mman/mmScavenge.c 2375 Out of memory. I have tried all the LD_ASSUME_KERNEL, setarch and font path tricks, including increasing the headroom with the -h engine option to no avail. The same happens with any version of VW. Anyone know what is going on? Thanks, Nick P.S. Machine details: Linux 2.6.25.3-18.fc9.i686 #1 SMP Tue May 13 05:38:53 EDT 2008 i686 athlon i386 GNU/Linux _______________________________________________ vwnc mailing list [hidden email] http://lists.cs.uiuc.edu/mailman/listinfo/vwnc |
Nick Yannakoyorgos wrote:
> Hi all, > > I have recently upgraded a Linux machine to Fedora 9 (it was Fedora 8) > and the VW engine now crashes on startup (even before the splash screen) > with: > > VisualWorks(R) 7.5 Non-Commercial Mar 3, 2008 > src/mman/mmScavenge.c 2375 > Out of memory. > > > I have tried all the LD_ASSUME_KERNEL, setarch and font path tricks, > including increasing the headroom with the -h engine option to no avail. > The same happens with any version of VW. Anyone know what is going on? > try starting the engine with -xq, like visual -xq visual.im This should give you a file stack.txt with stacks of all smalltalk processes. I expect that one of the processes has a recursion of errors. I hope you can find and correct the first error with that. Ralf -- Ralf Propach, [hidden email] Tel: +49 231 975 99 38 Fax: +49 231 975 99 20 Georg Heeg eK (Dortmund) Handelsregister: Amtsgericht Dortmund A 12812 _______________________________________________ vwnc mailing list [hidden email] http://lists.cs.uiuc.edu/mailman/listinfo/vwnc |
Ralf Propach wrote:
> Nick Yannakoyorgos wrote: > >> Hi all, >> >> I have recently upgraded a Linux machine to Fedora 9 (it was Fedora 8) >> and the VW engine now crashes on startup (even before the splash screen) >> with: >> >> VisualWorks(R) 7.5 Non-Commercial Mar 3, 2008 >> src/mman/mmScavenge.c 2375 >> Out of memory. >> >> >> I have tried all the LD_ASSUME_KERNEL, setarch and font path tricks, >> including increasing the headroom with the -h engine option to no avail. >> The same happens with any version of VW. Anyone know what is going on? >> >> > Hi Nick, > > try starting the engine with -xq, like > visual -xq visual.im > This should give you a file stack.txt with stacks of all smalltalk processes. > I expect that one of the processes has a recursion of errors. I hope > you can find and correct the first error with that. > > Ralf > > > > Thanks for the reply: yes it does and it seems that the fonts on the system are (again) at fault: context 0x0a5c92ec DebuggerService class(Object)>>doesNotUnderstand: context 0x0a5c92c8 DebuggerService class(Object)>>doesNotUnderstand: context 0x0a5c92a4 DebuggerService class(Object)>>doesNotUnderstand: context 0x0a5c9280 DebuggerService class(Object)>>doesNotUnderstand: context 0x0a5c925c DebuggerService class(Object)>>doesNotUnderstand: context 0x0a5c9238 DebuggerService class(Object)>>doesNotUnderstand: context 0x0a5c9214 DebuggerService class(Object)>>doesNotUnderstand: context 0x0a5c91f0 LargeArray class(ArrayedCollection class)>>withAll: context 0x0a5c91d8 XFont(ScreenFont)>>initializeFrom: context 0x0a5c91c0 Screen(HostGraphicsDevice)>>fontNamed:fromClass:encoding: context 0x0a5c9190 XFont class>>createFont:toResemble:on: context 0x0a5c9184 FontPolicy>>findFont: context 0x0a5c9178 FontPolicy>>initializeDefault context 0x0a5c916c Screen(GraphicsDevice)>>installFontPolicy context 0x0a5c9160 Screen(HostGraphicsDevice)>>installFontPolicy context 0x0a5c9154 Screen(GraphicsDevice)>>initialize context 0x0a5c9148 Screen(HostGraphicsDevice)>>initialize context 0x0a5c913c Screen>>initialize Any ideas about how to identify the offending font? Thanks, Nick _______________________________________________ vwnc mailing list [hidden email] http://lists.cs.uiuc.edu/mailman/listinfo/vwnc |
Got it: the offending package is:
xorg-x11-fonts-Type1 Removing this VW starts up fine. Thanks, Nick Nick Yannakoyorgos wrote: > Ralf Propach wrote: >> Nick Yannakoyorgos wrote: >> >>> Hi all, >>> >>> I have recently upgraded a Linux machine to Fedora 9 (it was Fedora >>> 8) and the VW engine now crashes on startup (even before the splash >>> screen) with: >>> >>> VisualWorks(R) 7.5 Non-Commercial Mar 3, 2008 >>> src/mman/mmScavenge.c 2375 >>> Out of memory. >>> >>> >>> I have tried all the LD_ASSUME_KERNEL, setarch and font path tricks, >>> including increasing the headroom with the -h engine option to no >>> avail. The same happens with any version of VW. Anyone know what is >>> going on? >>> >>> >> Hi Nick, >> >> try starting the engine with -xq, like >> visual -xq visual.im >> This should give you a file stack.txt with stacks of all smalltalk >> processes. >> I expect that one of the processes has a recursion of errors. I hope >> you can find and correct the first error with that. >> >> Ralf >> >> >> >> > Ralf, > > Thanks for the reply: yes it does and it seems that the fonts on the > system are (again) at fault: > > context 0x0a5c92ec DebuggerService class(Object)>>doesNotUnderstand: > context 0x0a5c92c8 DebuggerService class(Object)>>doesNotUnderstand: > context 0x0a5c92a4 DebuggerService class(Object)>>doesNotUnderstand: > context 0x0a5c9280 DebuggerService class(Object)>>doesNotUnderstand: > context 0x0a5c925c DebuggerService class(Object)>>doesNotUnderstand: > context 0x0a5c9238 DebuggerService class(Object)>>doesNotUnderstand: > context 0x0a5c9214 DebuggerService class(Object)>>doesNotUnderstand: > context 0x0a5c91f0 LargeArray class(ArrayedCollection class)>>withAll: > context 0x0a5c91d8 XFont(ScreenFont)>>initializeFrom: > context 0x0a5c91c0 > Screen(HostGraphicsDevice)>>fontNamed:fromClass:encoding: > context 0x0a5c9190 XFont class>>createFont:toResemble:on: > context 0x0a5c9184 FontPolicy>>findFont: > context 0x0a5c9178 FontPolicy>>initializeDefault > context 0x0a5c916c Screen(GraphicsDevice)>>installFontPolicy > context 0x0a5c9160 Screen(HostGraphicsDevice)>>installFontPolicy > context 0x0a5c9154 Screen(GraphicsDevice)>>initialize > context 0x0a5c9148 Screen(HostGraphicsDevice)>>initialize > context 0x0a5c913c Screen>>initialize > > Any ideas about how to identify the offending font? > > Thanks, > > Nick > _______________________________________________ vwnc mailing list [hidden email] http://lists.cs.uiuc.edu/mailman/listinfo/vwnc |
It would appear to me that there should be some exception
handling code that bypasses font exceptions during system startup. It should also record the offending font so it can be reported to the user after startup. Terry =========================================================== Terry Raymond Crafted Smalltalk 80 Lazywood Ln. Tiverton, RI 02878 (401) 624-4517 [hidden email] <http://www.craftedsmalltalk.com> =========================================================== > -----Original Message----- > From: [hidden email] [mailto:[hidden email]] On Behalf > Of Nick Yannakoyorgos > Sent: Wednesday, May 28, 2008 7:53 AM > To: [hidden email] > Subject: Re: [vwnc] Startup crash on Fedora 9 > > Got it: the offending package is: > > xorg-x11-fonts-Type1 > > Removing this VW starts up fine. > > Thanks, > > Nick > > Nick Yannakoyorgos wrote: > > Ralf Propach wrote: > >> Nick Yannakoyorgos wrote: > >> > >>> Hi all, > >>> > >>> I have recently upgraded a Linux machine to Fedora 9 (it was Fedora > >>> 8) and the VW engine now crashes on startup (even before the splash > >>> screen) with: > >>> > >>> VisualWorks(R) 7.5 Non-Commercial Mar 3, 2008 > >>> src/mman/mmScavenge.c 2375 > >>> Out of memory. > >>> > >>> > >>> I have tried all the LD_ASSUME_KERNEL, setarch and font path tricks, > >>> including increasing the headroom with the -h engine option to no > >>> avail. The same happens with any version of VW. Anyone know what is > >>> going on? > >>> > >>> > >> Hi Nick, > >> > >> try starting the engine with -xq, like > >> visual -xq visual.im > >> This should give you a file stack.txt with stacks of all smalltalk > >> processes. > >> I expect that one of the processes has a recursion of errors. I hope > >> you can find and correct the first error with that. > >> > >> Ralf > >> > >> > >> > >> > > Ralf, > > > > Thanks for the reply: yes it does and it seems that the fonts on the > > system are (again) at fault: > > > > context 0x0a5c92ec DebuggerService class(Object)>>doesNotUnderstand: > > context 0x0a5c92c8 DebuggerService class(Object)>>doesNotUnderstand: > > context 0x0a5c92a4 DebuggerService class(Object)>>doesNotUnderstand: > > context 0x0a5c9280 DebuggerService class(Object)>>doesNotUnderstand: > > context 0x0a5c925c DebuggerService class(Object)>>doesNotUnderstand: > > context 0x0a5c9238 DebuggerService class(Object)>>doesNotUnderstand: > > context 0x0a5c9214 DebuggerService class(Object)>>doesNotUnderstand: > > context 0x0a5c91f0 LargeArray class(ArrayedCollection class)>>withAll: > > context 0x0a5c91d8 XFont(ScreenFont)>>initializeFrom: > > context 0x0a5c91c0 > > Screen(HostGraphicsDevice)>>fontNamed:fromClass:encoding: > > context 0x0a5c9190 XFont class>>createFont:toResemble:on: > > context 0x0a5c9184 FontPolicy>>findFont: > > context 0x0a5c9178 FontPolicy>>initializeDefault > > context 0x0a5c916c Screen(GraphicsDevice)>>installFontPolicy > > context 0x0a5c9160 Screen(HostGraphicsDevice)>>installFontPolicy > > context 0x0a5c9154 Screen(GraphicsDevice)>>initialize > > context 0x0a5c9148 Screen(HostGraphicsDevice)>>initialize > > context 0x0a5c913c Screen>>initialize > > > > Any ideas about how to identify the offending font? > > > > Thanks, > > > > Nick > > > > _______________________________________________ > vwnc mailing list > [hidden email] > http://lists.cs.uiuc.edu/mailman/listinfo/vwnc _______________________________________________ vwnc mailing list [hidden email] http://lists.cs.uiuc.edu/mailman/listinfo/vwnc |
In reply to this post by Nick Yannakoyorgos-2
Nick Yannakoyorgos escreveu:
> Got it: the offending package is: > > xorg-x11-fonts-Type1 I find this as a /diagnostic/ > > Removing this VW starts up fine. > and this as a workaround. . . -- Cesar Rabak GNU/Linux User 52247. Get counted: http://counter.li.org/ _______________________________________________ vwnc mailing list [hidden email] http://lists.cs.uiuc.edu/mailman/listinfo/vwnc |
In reply to this post by Terry Raymond
HI Terry:
It sounds like you are suggesting a "least common denominator" approach to font handling at VW startup. If that is true, I applaud the suggestion. VW seems to be bitten by this sort of thing on a *far too regular* basis. On May 28, 2008, at 5:02 AM, Terry Raymond wrote: > It would appear to me that there should be some exception > handling code that bypasses font exceptions during system > startup. It should also record the offending font so it > can be reported to the user after startup. > > Terry > > =========================================================== > Terry Raymond > Crafted Smalltalk > 80 Lazywood Ln. > Tiverton, RI 02878 > (401) 624-4517 [hidden email] > <http://www.craftedsmalltalk.com> > =========================================================== > >> -----Original Message----- >> From: [hidden email] [mailto:[hidden email]] On >> Behalf >> Of Nick Yannakoyorgos >> Sent: Wednesday, May 28, 2008 7:53 AM >> To: [hidden email] >> Subject: Re: [vwnc] Startup crash on Fedora 9 >> >> Got it: the offending package is: >> >> xorg-x11-fonts-Type1 >> >> Removing this VW starts up fine. >> >> Thanks, >> >> Nick >> >> Nick Yannakoyorgos wrote: >>> Ralf Propach wrote: >>>> Nick Yannakoyorgos wrote: >>>> >>>>> Hi all, >>>>> >>>>> I have recently upgraded a Linux machine to Fedora 9 (it was >>>>> Fedora >>>>> 8) and the VW engine now crashes on startup (even before the >>>>> splash >>>>> screen) with: >>>>> >>>>> VisualWorks(R) 7.5 Non-Commercial Mar 3, 2008 >>>>> src/mman/mmScavenge.c 2375 >>>>> Out of memory. >>>>> >>>>> >>>>> I have tried all the LD_ASSUME_KERNEL, setarch and font path >>>>> tricks, >>>>> including increasing the headroom with the -h engine option to no >>>>> avail. The same happens with any version of VW. Anyone know what >>>>> is >>>>> going on? >>>>> >>>>> >>>> Hi Nick, >>>> >>>> try starting the engine with -xq, like >>>> visual -xq visual.im >>>> This should give you a file stack.txt with stacks of all smalltalk >>>> processes. >>>> I expect that one of the processes has a recursion of errors. I >>>> hope >>>> you can find and correct the first error with that. >>>> >>>> Ralf >>>> >>>> >>>> >>>> >>> Ralf, >>> >>> Thanks for the reply: yes it does and it seems that the fonts on the >>> system are (again) at fault: >>> >>> context 0x0a5c92ec DebuggerService class(Object)>>doesNotUnderstand: >>> context 0x0a5c92c8 DebuggerService class(Object)>>doesNotUnderstand: >>> context 0x0a5c92a4 DebuggerService class(Object)>>doesNotUnderstand: >>> context 0x0a5c9280 DebuggerService class(Object)>>doesNotUnderstand: >>> context 0x0a5c925c DebuggerService class(Object)>>doesNotUnderstand: >>> context 0x0a5c9238 DebuggerService class(Object)>>doesNotUnderstand: >>> context 0x0a5c9214 DebuggerService class(Object)>>doesNotUnderstand: >>> context 0x0a5c91f0 LargeArray class(ArrayedCollection >>> class)>>withAll: >>> context 0x0a5c91d8 XFont(ScreenFont)>>initializeFrom: >>> context 0x0a5c91c0 >>> Screen(HostGraphicsDevice)>>fontNamed:fromClass:encoding: >>> context 0x0a5c9190 XFont class>>createFont:toResemble:on: >>> context 0x0a5c9184 FontPolicy>>findFont: >>> context 0x0a5c9178 FontPolicy>>initializeDefault >>> context 0x0a5c916c Screen(GraphicsDevice)>>installFontPolicy >>> context 0x0a5c9160 Screen(HostGraphicsDevice)>>installFontPolicy >>> context 0x0a5c9154 Screen(GraphicsDevice)>>initialize >>> context 0x0a5c9148 Screen(HostGraphicsDevice)>>initialize >>> context 0x0a5c913c Screen>>initialize >>> >>> Any ideas about how to identify the offending font? >>> >>> Thanks, >>> >>> Nick >>> >> >> _______________________________________________ >> vwnc mailing list >> [hidden email] >> http://lists.cs.uiuc.edu/mailman/listinfo/vwnc > > _______________________________________________ > vwnc mailing list > [hidden email] > http://lists.cs.uiuc.edu/mailman/listinfo/vwnc Thanks!! Joseph Bacanskas [|] --- I use Smalltalk. My amp goes to eleven. _______________________________________________ vwnc mailing list [hidden email] http://lists.cs.uiuc.edu/mailman/listinfo/vwnc |
In reply to this post by Nick Yannakoyorgos-2
From: Joseph Bacanskas on 30.5.2008 7:36
> It sounds like you are suggesting a "least common denominator" > approach to font handling at VW startup. If that is true, I applaud > the suggestion. VW seems to be bitten by this sort of thing on a *far > too regular* basis. +1 Steve > On May 28, 2008, at 5:02 AM, Terry Raymond wrote: > > > It would appear to me that there should be some exception > > handling code that bypasses font exceptions during system > > startup. It should also record the offending font so it > > can be reported to the user after startup. > > > > Terry > > > > =========================================================== > > Terry Raymond > > Crafted Smalltalk > > 80 Lazywood Ln. > > Tiverton, RI 02878 > > (401) 624-4517 [hidden email] > > <http://www.craftedsmalltalk.com> > > =========================================================== > > > >> -----Original Message----- > >> From: [hidden email] [mailto:[hidden email]] On > >> Behalf > >> Of Nick Yannakoyorgos > >> Sent: Wednesday, May 28, 2008 7:53 AM > >> To: [hidden email] > >> Subject: Re: [vwnc] Startup crash on Fedora 9 > >> > >> Got it: the offending package is: > >> > >> xorg-x11-fonts-Type1 > >> > >> Removing this VW starts up fine. > >> > >> Thanks, > >> > >> Nick > >> > >> Nick Yannakoyorgos wrote: > >>> Ralf Propach wrote: > >>>> Nick Yannakoyorgos wrote: > >>>> > >>>>> Hi all, > >>>>> > >>>>> I have recently upgraded a Linux machine to Fedora 9 (it was > >>>>> Fedora > >>>>> 8) and the VW engine now crashes on startup (even before the > >>>>> splash > >>>>> screen) with: > >>>>> > >>>>> VisualWorks(R) 7.5 Non-Commercial Mar 3, 2008 > >>>>> src/mman/mmScavenge.c 2375 > >>>>> Out of memory. > >>>>> > >>>>> > >>>>> I have tried all the LD_ASSUME_KERNEL, setarch and font path > >>>>> tricks, > >>>>> including increasing the headroom with the -h engine option to > >>>>> avail. The same happens with any version of VW. Anyone know what > >>>>> is > >>>>> going on? > >>>>> > >>>>> > >>>> Hi Nick, > >>>> > >>>> try starting the engine with -xq, like > >>>> visual -xq visual.im > >>>> This should give you a file stack.txt with stacks of all > >>>> processes. > >>>> I expect that one of the processes has a recursion of errors. I > >>>> hope > >>>> you can find and correct the first error with that. > >>>> > >>>> Ralf > >>>> > >>>> > >>>> > >>>> > >>> Ralf, > >>> > >>> Thanks for the reply: yes it does and it seems that the fonts on > >>> system are (again) at fault: > >>> > >>> context 0x0a5c92ec DebuggerService class(Object)>>doesNotUnderstand: > >>> context 0x0a5c92c8 DebuggerService class(Object)>>doesNotUnderstand: > >>> context 0x0a5c92a4 DebuggerService class(Object)>>doesNotUnderstand: > >>> context 0x0a5c9280 DebuggerService class(Object)>>doesNotUnderstand: > >>> context 0x0a5c925c DebuggerService class(Object)>>doesNotUnderstand: > >>> context 0x0a5c9238 DebuggerService class(Object)>>doesNotUnderstand: > >>> context 0x0a5c9214 DebuggerService class(Object)>>doesNotUnderstand: > >>> context 0x0a5c91f0 LargeArray class(ArrayedCollection > >>> class)>>withAll: > >>> context 0x0a5c91d8 XFont(ScreenFont)>>initializeFrom: > >>> context 0x0a5c91c0 > >>> Screen(HostGraphicsDevice)>>fontNamed:fromClass:encoding: > >>> context 0x0a5c9190 XFont class>>createFont:toResemble:on: > >>> context 0x0a5c9184 FontPolicy>>findFont: > >>> context 0x0a5c9178 FontPolicy>>initializeDefault > >>> context 0x0a5c916c Screen(GraphicsDevice)>>installFontPolicy > >>> context 0x0a5c9160 Screen(HostGraphicsDevice)>>installFontPolicy > >>> context 0x0a5c9154 Screen(GraphicsDevice)>>initialize > >>> context 0x0a5c9148 Screen(HostGraphicsDevice)>>initialize > >>> context 0x0a5c913c Screen>>initialize > >>> > >>> Any ideas about how to identify the offending font? > >>> > >>> Thanks, > >>> > >>> Nick > >>> > >> > >> _______________________________________________ > >> vwnc mailing list > >> [hidden email] > >> http://lists.cs.uiuc.edu/mailman/listinfo/vwnc > > > > _______________________________________________ > > vwnc mailing list > > [hidden email] > > http://lists.cs.uiuc.edu/mailman/listinfo/vwnc > > Thanks!! > Joseph Bacanskas [|] > --- I use Smalltalk. My amp goes to eleven. > > _______________________________________________ > vwnc mailing list > [hidden email] > http://lists.cs.uiuc.edu/mailman/listinfo/vwnc _______________________________________________ vwnc mailing list [hidden email] http://lists.cs.uiuc.edu/mailman/listinfo/vwnc |
In reply to this post by Terry Raymond
Am 28.05.08 14:02 schrieb "Terry Raymond" unter
<[hidden email]>: > It would appear to me that there should be some exception > handling code that bypasses font exceptions during system > startup. It should also record the offending font so it > can be reported to the user after startup. It would appear to me that there should be some exception handling code in font lookup not only during system startup. ;-) We had to patch a method somewhere below findFont: because it used to crash our application when automatic software rollouts for Lotus Notes and MS Office installed fonts while the app was running. Apparently, Windows got confused about the font handles when that happened, or the VM didn't react appropriately. Fixing it was easy enough: When a font can't be allocated, answer the default font instead of raising a primitive failure which doesn't even tell you which primitive has failed and why. This may not be the right solution in all cases. Raising a Notification instead of an Error might be a better way to cope with this situation. The Notification could answer the default font in its defaultAction, such that if you don't handle it, the software can proceed, but you could handle it and do something else instead. Joachim Geidel _______________________________________________ vwnc mailing list [hidden email] http://lists.cs.uiuc.edu/mailman/listinfo/vwnc |
Free forum by Nabble | Edit this page |