After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 127202 - Recalc in the wrong order will overflow the stack
Recalc in the wrong order will overflow the stack
Status: RESOLVED DUPLICATE of bug 92131
Product: Gnumeric
Classification: Applications
Component: General
git master
Other Linux
: High major
: ---
Assigned To: Jody Goldberg
Jody Goldberg
Depends on:
Blocks:
 
 
Reported: 2003-11-17 15:57 UTC by Frederic Parrenin
Modified: 2004-12-22 21:47 UTC
See Also:
GNOME target: ---
GNOME version: ---


Description Frederic Parrenin 2003-11-17 15:57:29 UTC 
I have a .gnumeric file that crashes gnumeric.
This file is in fact a .xls file I have saved as .gnumeric file.

The original .xls file is to large to attached it in bugzilla.
You can find it at:
ftp://lgge.obs.ujf-grenoble.fr/pub/depot/parrenin/test.xls
Comment 1 Morten Welinder 2003-11-17 16:43:08 UTC 
This loads fine for me, but it's clear that gnumeric uses an insane
amount of memory for this.  How much do you have?

(With the experimental SAX loader, which requires much less memory,
it does crash:)

Program received signal SIGSEGV, Segmentation fault.
0xfb044 in workbook_sheet_by_name (wb=0x6d20bf, name=0x4ecb00
"Modèle") at workbook.c:903
903             g_return_val_if_fail (IS_WORKBOOK (wb), NULL);
Comment 2 Morten Welinder 2003-11-17 17:05:25 UTC 
xml_sax crash fixed in cvs.
Comment 3 Frederic Parrenin 2003-11-17 17:32:05 UTC 
Morten,
Thanks for your quick reply and the quick fix in the sax importer.

I have 512Mb of memory in the AthlonXP2000+ I am using now.
I am surprised that you did not experienced a crash when loading the
gnumeric file.
Have you really saved the file as .gnumeric and reopened it ?

I have putted the .gnumeric file that crashed my gnumeric-1.2.1 at the
same place:
ftp://lgge.obs.ujf-grenoble.fr/pub/depot/parrenin/test.gnumeric
Please give a look at it.

Anyway, this is a nice sheet to test performance issues :-)

   Frederic
Comment 4 Jody Goldberg 2003-11-17 22:35:47 UTC 
I can replicate.

The crash is a known problem which we can not solve for 1.2.x
In certain cases gnumeric can run out of stack space when calculating
poorly ordered expressions.

For 1.2.x we are tuned so that things work for descending A2=A1+foo,
not ascending which is what you have there.
Comment 5 Jody Goldberg 2004-02-18 18:40:59 UTC 

*** This bug has been marked as a duplicate of 92131 ***