In the tables which are used to parse the fields for the set_fields
command, there should never be a entry which has a size set to 8
bytes, and two pointers defined. Not only would it result in
undefined behavior in the compiled code, it doesn't make any sense and
is definitely a bug.
Reported-by: Xi Wang <xi.wang@gmail.com>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
n = num & mask;
switch (size) {
case 8:
- *u.ptr64 = n;
+ /* Should never get here */
+ fprintf(stderr, "64-bit field %s has a second 64-bit field\n"
+ "defined; BUG?!?\n", info->name);
+ *u.ptr64 = 0;
break;
case 4:
*u.ptr32 = n;