[gs-bugs] [Bug 690642] conversion to PDF aborts after 800 pages

bugs.ghostscript.com-bugzilla-daemon at ghostscript.com bugs.ghostscript.com-bugzilla-daemon at ghostscript.com
Tue Jul 21 08:59:11 PDT 2009


------- Additional Comments From ken.sharp at artifex.com  2009-07-21 08:59 -------
Well, my attempt crashed (no 'abort' but a Windows GP fault, Unix SEGV) on page
940. At this point GS was using > 1.5GB of memory (!)

It looks to me like this is probably a memory problem, I suspect that we have
exhausted memory, something is trying to allocate more and failing and most
likely not detecting the failure, resulting in an attempt to write to location 0
or similar. I'll have to debug it to confirm that, which will take quite a long
time, since it takes 10-15 minutes to get to this point.

The indeterminacy is 'probably' due to the amount of memory available to the
process, which will be different on all machines, and very possibly will vary
from moment to moment. I'm guessing that the crashes are not all in the same
place (they may be though, if its a big allocation), even if I fix the crash,
the file will still not process correctly since there is insufficient memory.

I'd suggest that possibly processing the file in 2 parts may help, run pages
0-500 then 501-1140, then merge the two. It mat well be faster too, GS is
running pretty slowly by the time it reaches page 800.

------- You are receiving this mail because: -------
You are the QA contact for the bug, or are watching the QA contact.

More information about the gs-bugs mailing list