2010/10/15 Arnd Bergmann <arnd@xxxxxxxx>: > On Friday 15 October 2010, Akinobu Mita wrote: >> +#define __set_le_bit(nr, addr) \ >> + __set_bit((nr)^(__BITOPS_WORDSIZE - 8), (addr)) >> +#define __clear_le_bit(nr, addr) \ >> + __clear_bit((nr)^(__BITOPS_WORDSIZE - 8), (addr)) >> +#define __test_and_set_le_bit(nr, addr) \ >> + __test_and_set_bit((nr)^(__BITOPS_WORDSIZE - 8), (addr)) >> +#define test_and_set_le_bit(lock, nr, addr) \ >> + test_and_set_bit((nr)^(__BITOPS_WORDSIZE - 8), (addr)) >> +#define __test_and_clear_le_bit(nr, addr) \ >> + __test_and_clear_bit((nr)^(__BITOPS_WORDSIZE - 8), (addr)) >> +#define test_and_clear_le_bit(lock, nr, addr) \ >> + test_and_clear_bit((nr)^(__BITOPS_WORDSIZE - 8), (addr)) >> +#define test_le_bit(nr, addr) \ >> + test_bit((nr)^(__BITOPS_WORDSIZE - 8), (addr)) > > The test_and_clear_le_bit and test_and_set_le_bit calling conventions > are a little surprising here. I would not pass the lock argument > here but leave that to the ext2 wrapper: > > #define test_and_set_le_bit(nr, addr) \ > test_and_set_bit((nr)^(__BITOPS_WORDSIZE - 8), (addr)) > #define test_and_clear_le_bit(nr, addr) \ > test_and_clear_bit((nr)^(__BITOPS_WORDSIZE - 8), (addr)) > > #define ext2_set_bit_atomic(lock, nr, addr) \ > test_and_set_le_bit((nr), (unsigned long *)(addr)) > #define ext2_clear_bit_atomic(lock, nr, addr) \ > test_and_clear_le_bit((nr), (unsigned long *)(addr)) > > Same thing on arm and m68k. Oops, test_and_{set,clear}_le_bit() is only used by ext2_{set,clear}_bit_atomic and I had one additional patch that cleans up these definisions across the architectures. So I didn't realize that mistake when I did cross compile. -- To unsubscribe from this list: send the line "unsubscribe linux-arch" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html