Hi, for my presentations, I have a graphical counter on every slide, so metapost has to write a graphic for every slide/page that ConTeXt produces. This works with mkii, but when I process my file in mkiv, I get this error after page 16: report >> error: 15 output files written: .. Transcript written on mpout.log. Memory size overflow! [15.15] report >> error: no-term no-error report >> error: no-term no-error report >> error: no-term no-error report >> error: no-term no-error report >> error: no-term no-error [16.16] report >> error: no-term no-error report >> error: no-term no-error report >> error: no-term no-error report >> error: no-term no-error report >> error: no-term no-error [17.17] Is there any way that I can influence the memory size? 15 doesn't seem like a whole lot... Best Thomas
Thomas A. Schmitz wrote:
Hi,
for my presentations, I have a graphical counter on every slide, so metapost has to write a graphic for every slide/page that ConTeXt produces. This works with mkii, but when I process my file in mkiv, I get this error after page 16:
report >> error: 15 output files written: .. Transcript written on mpout.log.
Memory size overflow! [15.15] report >> error: no-term no-error report >> error: no-term no-error report >> error: no-term no-error report >> error: no-term no-error report >> error: no-term no-error [16.16] report >> error: no-term no-error report >> error: no-term no-error report >> error: no-term no-error report >> error: no-term no-error report >> error: no-term no-error [17.17]
Is there any way that I can influence the memory size? 15 doesn't seem like a whole lot...
hm, maybe something messy with the graphic ... i allocate 2MB of memory for metafun so ... anyhow, \appendtoks\ctxlua{metapost.reset("metafun")}\to\everyshipout or so, may help, although it's better to see what's the real reason does texexec --pdftex --automp yourfile.tex run ok? ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | fax: 038 477 53 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------
On Apr 24, 2008, at 1:44 AM, Hans Hagen wrote:
hm, maybe something messy with the graphic ... i allocate 2MB of memory for metafun so ...
anyhow,
\appendtoks\ctxlua{metapost.reset("metafun")}\to\everyshipout
or so, may help, although it's better to see what's the real reason
does
texexec --pdftex --automp yourfile.tex
run ok?
Yes, that runs OK. If I insert the \appendtoks... command into the preamble, I get errors while processing, so there appears to be something wrong with my graphic. I'll see whether I can nail it down this afternoon. Thanks, and best Thomas
Hans Hagen wrote:
Thomas A. Schmitz wrote:
Hi,
for my presentations, I have a graphical counter on every slide, so metapost has to write a graphic for every slide/page that ConTeXt produces. This works with mkii, but when I process my file in mkiv, I get this error after page 16:
report >> error: 15 output files written: .. Transcript written on mpout.log.
Memory size overflow! [15.15]
That error indeed comes from mplib, but from the overflow test in realloc(), not the memory array "overflow" function. You get this error when you attempt to allocate more than 2^31 bytes. Since that clearly should not happen, there is a bug somewhere in MPlib. But I'll need your source to dig it out. (likely you do something we have never tested yet). Best wishes, Taco
On Thu, 24 Apr 2008 08:38:51 +0200
Taco Hoekwater
That error indeed comes from mplib, but from the overflow test in realloc(), not the memory array "overflow" function. You get this error when you attempt to allocate more than 2^31 bytes. Since that clearly should not happen, there is a bug somewhere in MPlib. But I'll need your source to dig it out. (likely you do something we have never tested yet).
Thanks Taco. I will try to produce a minimal example this afternoon/evening. The problem occurs with one of my presentation modes, but I have other documents with the same module which don't trigger the error, so I'll have to spend some time debugging. Thomas
On Apr 24, 2008, at 8:38 AM, Taco Hoekwater wrote:
That error indeed comes from mplib, but from the overflow test in realloc(), not the memory array "overflow" function. You get this error when you attempt to allocate more than 2^31 bytes. Since that clearly should not happen, there is a bug somewhere in MPlib. But I'll need your source to dig it out. (likely you do something we have never tested yet).
Just for the record: Taco found and squished this bug; the new luatex trunk doesn't have it anymore. Thanks for the swift action! Best Thomas
Thomas A. Schmitz wrote:
On Apr 24, 2008, at 8:38 AM, Taco Hoekwater wrote:
That error indeed comes from mplib, but from the overflow test in realloc(), not the memory array "overflow" function. You get this error when you attempt to allocate more than 2^31 bytes. Since that clearly should not happen, there is a bug somewhere in MPlib. But I'll need your source to dig it out. (likely you do something we have never tested yet).
Just for the record: Taco found and squished this bug; the new luatex trunk doesn't have it anymore. Thanks for the swift action!
the next mkiv upload will have a bit better reporter for such cases ----------------------------------------------------------------- Hans Hagen | PRAGMA ADE Ridderstraat 27 | 8061 GH Hasselt | The Netherlands tel: 038 477 53 69 | fax: 038 477 53 74 | www.pragma-ade.com | www.pragma-pod.nl -----------------------------------------------------------------
participants (4)
-
Hans Hagen -
Taco Hoekwater -
Thomas A. Schmitz -
Thomas Schmitz