[GRASS-dev] Re: [GRASS GIS] #565: r.colors: glibc double free or corruption with -ae

GRASS GIS trac at osgeo.org
Wed Apr 22 00:36:12 EDT 2009


#565: r.colors: glibc double free or corruption with -ae
---------------------+------------------------------------------------------
  Reporter:  hamish  |       Owner:  grass-dev at lists.osgeo.org
      Type:  defect  |      Status:  new                      
  Priority:  minor   |   Milestone:  6.5.0                    
 Component:  Raster  |     Version:  svn-develbranch6         
Resolution:          |    Keywords:  r.colors                 
  Platform:  Linux   |         Cpu:  x86-32                   
---------------------+------------------------------------------------------
Comment (by glynn):

 Replying to [comment:1 hamish]:

 > here we go, happened again..
 {{{
 G> r.univar -g map1
 ...
 min=-23.8941449919674
 max=1999.8932486971
 ...
 }}}
 >
 {{{
 ...
 #7  0x0804b386 in get_fp_stats (name=0x9b72640 "map1", mapset=0x9b72030
 "user7",
     statf=0xbfd05f94, min=3.2146326349358647, max=7.601348984183665,
 geometric=0,
     geom_abs=1) at stats.c:136
 }}}

 If any of the raster values lie outside of the range specified by the
 min/max parameters, memory corruption will occur due to modifying an out-
 of-range array element.

 FWIW, I have noticed a bug in get_fp_stats(...,geom_abs=1): if the range
 is signed, values which are closer to zero than both min and max will also
 modify an out-of-range array element.

 I'll fix that one shortly, but it won't help if the min/max parameters are
 bogus.

-- 
Ticket URL: <https://trac.osgeo.org/grass/ticket/565#comment:4>
GRASS GIS <http://grass.osgeo.org>


More information about the grass-dev mailing list