Inkscape SVN crashes on startup
Hi,
I build Inkscape from recent SVN and it crashes on startup. I have Ubuntu 9.04 and build Inkscape with:
make clean CFLAGS='-g -O0' CXXFLAGS='-g -O0' ./configure --prefix=/home/tobias/dev/bin/ make make install
This is the console output if I try to start Inkscape: http://paste.ubuntu.com/284467/
And this is what I get if I start it with gdb: http://paste.ubuntu.com/284470/
And this is the output from valgrind: http://paste.ubuntu.com/284474/
Regards, Tobias
Can you also provide the output of the 'backtrace' command after the glibc error? The glibc-generated backtrace doesn't have any symbolic information, while the GDB backtrace will have it.
Valgrind probably won't give you anything meaningful with Inkscape because we're currently using the Boehm garbage collector.
Regards, Krzysztof
2009/10/3 Tobias Jakobs <tobias.jakobs@...1439...>:
Hi,
I build Inkscape from recent SVN and it crashes on startup. I have Ubuntu 9.04 and build Inkscape with:
make clean CFLAGS='-g -O0' CXXFLAGS='-g -O0' ./configure --prefix=/home/tobias/dev/bin/ make make install
This is the console output if I try to start Inkscape: http://paste.ubuntu.com/284467/
And this is what I get if I start it with gdb: http://paste.ubuntu.com/284470/
And this is the output from valgrind: http://paste.ubuntu.com/284474/
Regards, Tobias
Come build with us! The BlackBerry® Developer Conference in SF, CA is the only developer event you need to attend this year. Jumpstart your developing skills, take BlackBerry mobile applications to market and stay ahead of the curve. Join us from November 9-12, 2009. Register now! http://p.sf.net/sfu/devconf _______________________________________________ Inkscape-devel mailing list Inkscape-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/inkscape-devel
2009/10/3 Krzysztof Kosiński <tweenk.pl@...400...>:
Can you also provide the output of the 'backtrace' command after the glibc error? The glibc-generated backtrace doesn't have any symbolic information, while the GDB backtrace will have it.
OK, here output from the backtrace: http://paste.ubuntu.com/284571/
I hope this helps.
Regards, Tobias
On Oct 3, 2009, at 5:17 AM, Tobias Jakobs wrote:
2009/10/3 Krzysztof Kosiński <tweenk.pl@...400...>:
Can you also provide the output of the 'backtrace' command after the glibc error? The glibc-generated backtrace doesn't have any symbolic information, while the GDB backtrace will have it.
OK, here output from the backtrace: http://paste.ubuntu.com/284571/
I hope this helps.
Thanks, that's a good start.
It did stop early, though
#15 0x082a0222 in Inkscape::SVG::PathString::_appendPoint (this=0xbfcd44e0, p= {_pt = {3.7118498975370057e-314, -5.6550316700445993e+284}}, sc=true) ---Type <return> to continue, or q <return> to quit---
If you can hit <return> while you still get that message it will then continue and finish showing us where the calls originated from. Off hand, it looks like those x and y coordinates for the point are garbage. They might be the cause of problems, or just might be a result of the true problem.
Thanks.
Hi!
On Sat, Oct 3, 2009 at 16:46, Jon A. Cruz <jon@...18...> wrote:
It did stop early, though
#15 0x082a0222 in Inkscape::SVG::PathString::_appendPoint (this=0xbfcd44e0, p= {_pt = {3.7118498975370057e-314, -5.6550316700445993e+284}}, sc=true) ---Type <return> to continue, or q <return> to quit---
If you can hit <return> while you still get that message it will then continue and finish showing us where the calls originated from. Off hand, it looks like those x and y coordinates for the point are garbage. They might be the cause of problems, or just might be a result of the true problem.
Here is the full backtrace: http://paste.ubuntu.com/284701/
Perhaps it's even interesting, that this is on an Notebook with a 1024x768px screen.
Regards, Tobias
On 3/10/09 18:02, Tobias Jakobs wrote:
Hi!
On Sat, Oct 3, 2009 at 16:46, Jon A. Cruz <jon@...18...> wrote:
It did stop early, though
#15 0x082a0222 in Inkscape::SVG::PathString::_appendPoint (this=0xbfcd44e0, p= {_pt = {3.7118498975370057e-314, -5.6550316700445993e+284}}, sc=true) ---Type <return> to continue, or q <return> to quit---
If you can hit <return> while you still get that message it will then continue and finish showing us where the calls originated from. Off hand, it looks like those x and y coordinates for the point are garbage. They might be the cause of problems, or just might be a result of the true problem.
Here is the full backtrace: http://paste.ubuntu.com/284701/
Perhaps it's even interesting, that this is on an Notebook with a 1024x768px screen.
from the backtrace:
precision=2
did you manually (or in an older version) change the SVG output precision in your preferences file? Which SVN revision are you compiling?
(see bug #399604 “'malloc: *** error' after changing numeric precision to 1 or 2 ”: https://bugs.launchpad.net/inkscape/+bug/399604)
Hello,
from the backtrace:
precision=2
did you manually (or in an older version) change the SVG output precision in your preferences file?
Not that I can remember. But now I renamed the .config/inkscape folder and Inkscape starts again. If you are interested in the old preferences, you can download my old inkscape config folder here: http://dl.getdropbox.com/u/1069285/inkscape_bak.tar.gz
Which SVN revision are you compiling?
It was SVN trunk from today ~10:00 MEZ.
Regards, Tobias
On Oct 3, 2009, at 10:34 AM, Tobias Jakobs wrote:
Not that I can remember. But now I renamed the .config/inkscape folder and Inkscape starts again. If you are interested in the old preferences, you can download my old inkscape config folder here: http://dl.getdropbox.com/u/1069285/inkscape_bak.tar.gz
Thanks.
That is exactly the kind of feedback and data that we need. It sounds like it will be straightforward to test, given that folder.
I'm at the SVG Open conference this weekend, so I can't actually put my hands on it until Tuesday. However I have a feeing someone else will at least take a look before then.
On 3/10/09 19:34, Tobias Jakobs wrote:
On 3/10/09 18:20, ~suv wrote:
from the backtrace:
precision=2
did you manually (or in an older version) change the SVG output precision in your preferences file?
Not that I can remember. But now I renamed the .config/inkscape folder and Inkscape starts again. If you are interested in the old preferences, you can download my old inkscape config folder here: http://dl.getdropbox.com/u/1069285/inkscape_bak.tar.gz
Which SVN revision are you compiling?
It was SVN trunk from today ~10:00 MEZ.
I tested your preferences.xml (which does indeed contain the setting 'numericprecision="2"') with Inkscape 0.46+devel r22379 on OS X 10.5.8.
Inkscape crashes as described in bug #399604, and the preferences file needs to be manually edited (reset precsision >= 3) to avoid the crash.
Console messages on OS 10.5.8:
inkscape-bin(6946,0xa036e720) malloc: *** error for object 0x5de2500: incorrect checksum for freed object - object was probably modified after being freed. *** set a breakpoint in malloc_error_break to debug
Emergency save activated! inkscape-bin(6946,0xa036e720) malloc: *** error for object 0x5ddfe90: incorrect checksum for freed object - object was probably modified after being freed. *** set a breakpoint in malloc_error_break to debug inkscape-bin(6946,0xa036e720) malloc: *** error for object 0x5ddfe90: incorrect checksum for freed object - object was probably modified after being freed. *** set a breakpoint in malloc_error_break to debug Emergency save completed. Inkscape will close now.
IMHO this confirms your crash which was caused by a currently incompatible preference setting from earlier versions.
~suv
participants (4)
-
Jon A. Cruz -
Krzysztof Kosiński -
Tobias Jakobs -
~suv