On 15/04/10 20:53, Marco d'Itri wrote:
On Apr 15, Meelis Roos<[email protected]> wrote:
Please find out exactly which operation is causing the traps.
Are there any news?
I just found a moment to look at this:
Core was generated by `depmod -a'.
Program terminated with signal 10, Bus error.
#0 0x000000012000e34c in native_endianness () at ../util.c:205
205 return (char) *((uint32_t*)("\1\0\0\2"));
(gdb) bt
#0 0x000000012000e34c in native_endianness () at ../util.c:205
#1 0x000000012000738c in main (argc=2, argv=0x11facdc98) at
../depmod.c:1312
(gdb) list
200 int __attribute__ ((pure)) native_endianness()
201 {
202 /* Encoding the endianness enums in a string and then reading
that
203 * string as a 32-bit int, returns the correct endianness
automagically.
204 */
205 return (char) *((uint32_t*)("\1\0\0\2"));
206 }
Yeah, nice programming guys. Allocate a string, i.e. an array of
char, which the compiler will merely byte align and then read it
back as a long word which must be long word aligned. And the
programmer thinks he is so smart that he descibes it
as "automagically"! Doh.
Cheers
Michael.
--
To UNSUBSCRIBE, email to [email protected]
with a subject of "unsubscribe". Trouble? Contact [email protected]
