[Bug c/44384] builtin_object_size_ treatment of multidimensional arrays is unexpected

[Bug c/44384] builtin_object_size_ treatment of multidimensional arrays is unexpected

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=44384

Jakub Jelinek <jakub at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |jakub at gcc dot gnu.org

--- Comment #1 from Jakub Jelinek <jakub at gcc dot gnu.org> 2011-03-28 14:05:34 UTC ---
__builtin_object_size is a builtin for -D_FORTIFY_SOURCE checking, it is not a
generic object size computation function, and for strcpy IMHO we don't want to
treat each dimension as a field separator.

[Bug c/44384] builtin_object_size_ treatment of multidimensional arrays is unexpected

[meklund at cisco dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=44384

--- Comment #2 from Mark Eklund <meklund at cisco dot com> 2011-03-29 20:31:15 UTC ---
I use FORTIFY_SOURCE with a USE_FORTIFY_LEVEL of 2
and was surprised that the below code allowed
writing into fu[4][0].  I expected that at a
level of 1, but not 2.

#include <string.h>

int bar ()
{
    char fu[10][10];
    strcpy(fu[3], "ten counts");
    return 0;
}

[Bug c/44384] builtin_object_size_ treatment of multidimensional arrays is unexpected

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=44384

--- Comment #3 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-03-30 08:47:32 UTC ---
The FE is also not very helpful when decaying arrays to pointers or when
generating array accesses.

[Bug c/44384] builtin_object_size_ treatment of multidimensional arrays is unexpected

[meklund at cisco dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=44384

--- Comment #4 from Mark Eklund <meklund at cisco dot com> 2011-04-21 15:17:37 UTC ---
Although "subobject" isn't well defined in the documentation,
the below comments from tree-object-size.c revision r171388 
lines 63-68 would indicate that subobject is the innermost array.

   /* 
     *
     * object_sizes[0] is upper bound for number of bytes till the end of
     * the object.
     * object_sizes[1] is upper bound for number of bytes till the end of
     * the subobject (innermost array or field with address taken).
     * object_sizes[2] is lower bound for number of bytes till the end of
     * the object and object_sizes[3] lower bound for subobject. 
     */

[Bug middle-end/44384] builtin_object_size_ treatment of multidimensional arrays is unexpected

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=44384

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |siddhesh at gcc dot gnu.org

--- Comment #6 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
*** Bug 110373 has been marked as a duplicate of this bug. ***