The original crash I got was from Debian 5, 32bit x86, running in a vm. I was able to reproduce it after building from the source package in Debian (which builds with debug symbols). The stack trace from the crash was useless however.
What other info do you want? I used the same environment variable as I posted earlier to trigger. -- Silvio On Tue, Jun 7, 2011 at 11:31 PM, Bastien ROUCARIES < [email protected]> wrote: > Could you give me a test case ? > > On Tue, May 31, 2011 at 4:53 AM, Silvio Cesare <[email protected]> > wrote: > > Sorry again for the delays. > > > > It looks like an off-by-1 in the option parsing code for long options. My > > patch is is included. > > > > -- > > Silvio > > > > diff -ru imagemagick-6.3.7.9.dfsg2/magick/option.c > > imagemagick-6.3.7.9.dfsg2-fix/magick/option.c > > --- imagemagick-6.3.7.9.dfsg2/magick/option.c 2007-12-13 > > 07:41:12.000000000 +1100 > > +++ imagemagick-6.3.7.9.dfsg2-fix/magick/option.c 2011-05-31 > > 12:31:27.000000000 +1000 > > @@ -1866,7 +1866,7 @@ > > while (((isspace((int) ((unsigned char) *p)) == 0) && (*p != ',')) > && > > (*p != '\0')) > > { > > - if ((q-token) >= MaxTextExtent) > > + if ((q-token) >= (MaxTextExtent - 1)) > > break; > > *q++=(*p++); > > } > > > > > > On Mon, May 30, 2011 at 7:38 PM, Bastien ROUCARIES > > <[email protected]> wrote: > >> > >> Package: imagemagick > >> Version: 8:6.6.0.4-3 > >> > >> Could you send me the info asked ? > >> > >> Thanks > >> > >> Bastien > > > > >
