On Sat, Apr 11, 2015 at 09:40:36AM -0700, Matt Turner wrote: > On Sat, Apr 11, 2015 at 2:46 AM, Kenneth Graunke <kenn...@whitecape.org> > wrote: > > new_prim was declared as a stack variable within a nested scope; we > > tried to retain a pointer to that data beyond the scope, which is bogus. > > > > GCC with -O1 eliminated most of the code that set new_prim's fields. > > > > Move the declaration to fix the bug. > > > > Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=81025 > > Signed-off-by: Kenneth Graunke <kenn...@whitecape.org> > > Cc: Mark Janes <mark.a.ja...@intel.com> > > Cc: mesa-sta...@lists.freedesktop.org > > --- > > Wow, yeah, that's awful. > > Doesn't ib/new_ib suffer from the same problem? > > With ib/new_ib fixed as well,
Good catch, Matt. It looks like it is needed to me as well. (The author and reviewer of the original patch certainly make me feel better about the silly mistakes I make). Reviewed-by: Ben Widawsky <b...@bwidawsk.net> > > Reviewed-by: Matt Turner <matts...@gmail.com> > > FWIW, Coverity [0] has known about this since Dec 2013. I really > recommend people use this. It'll send you emails about new defects, > which is often nice because you just reviewed that code and missed > something. :) > > [0] https://scan.coverity.com/projects/139?tab=overview > _______________________________________________ > mesa-dev mailing list > mesa-dev@lists.freedesktop.org > http://lists.freedesktop.org/mailman/listinfo/mesa-dev _______________________________________________ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
