Bug#838486: inkscape: Segmentation fault in 0-48.5 src/display/nr-arena-image.cpp

[Alessandro Vesely]

Hi Mattia,

On Wed 21/Sep/2016 16:54:02 +0200 Mattia Rizzolo wrote:
> On Wed, Sep 21, 2016 at 02:13:24PM +0200, Alessandro Vesely wrote:
>> $ gdb -q --args /usr/bin/inkscape test-pdf.svg
>> Reading symbols from /usr/bin/inkscape...done.
>> (gdb) run
>> Starting program: /usr/bin/inkscape test-pdf.svg
>> [Thread debugging using libthread_db enabled]
>> Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
>> [New Thread 0x7fffe66dd700 (LWP 14025)]
>> [New Thread 0x7fff5442f700 (LWP 14030)]
>> [New Thread 0x7fff53bce700 (LWP 14033)]
>>
>> Program received signal SIGSEGV, Segmentation fault.
>> nr_arena_image_pick (item=0x29f5e00, p=..., delta=<optimized out>) at display
>> /nr-arena-image.cpp:318
>> 318             return (pix_ptr[3] > 0) ? item : NULL;
>
> nasty crash.
>
> Now, that's the stable release, though.  And most of the development
> efforts are concentrated in unstable.
> Can you please check whether the crash happens with 0.91?  you can just
> use what you find in jessie-backports for that.

I got a crash at a different point:

$ gdb -q --args /usr/bin/inkscape test-pdf.svg
Reading symbols from /usr/bin/inkscape...done.
(gdb) run
Starting program: /usr/bin/inkscape test-pdf.svg
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
[New Thread 0x7fffe3026700 (LWP 21884)]
[New Thread 0x7fff4c92b700 (LWP 21885)]
[New Thread 0x7fff47f9f700 (LWP 21886)]

Program received signal SIGSEGV, Segmentation fault.
bits_image_fetch_separable_convolution_affine (repeat_mode=PIXMAN_REPEAT_NONE, 
format=PIXMAN_a8r8g8b8, convert_pixel=<optimized out>, mask=0x0, 
buffer=0x7fffffff4090,
     width=<optimized out>, line=<optimized out>, offset=<optimized out>, 
image=0x555561e82e80) at ../../pixman/pixman-fast-path.c:2813
2813	../../pixman/pixman-fast-path.c: No such file or directory.
(gdb)


This looks similar to an older bug I reported in July:
https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=832579

I'll try and reapply the latter patch tomorrow, and see how it goes.

Ale