2015-12-03 18:30 GMT+03:00 Sasha Levin <[email protected]>: > Passing 0 to roundup_pow_of_two would lead to wrapping around and trying to > find the last set bit on (unsigned long)(-1), which is obviously wrong. > > Instead, deal with this case by rounding it up to the closest power of two > (2 ** 0). > > Signed-off-by: Sasha Levin <[email protected]> > --- > include/linux/log2.h | 3 +++ > 1 file changed, 3 insertions(+) > > diff --git a/include/linux/log2.h b/include/linux/log2.h > index fd7ff3d..b6bdf0c 100644 > --- a/include/linux/log2.h > +++ b/include/linux/log2.h > @@ -60,6 +60,9 @@ bool is_power_of_2(unsigned long n) > static inline __attribute__((const)) > unsigned long __roundup_pow_of_two(unsigned long n) > { > + if (n == 0) > + return 1UL << 0; > +
Perhaps we should fix callers instead? Comment near roundup_pow_of_two() says that result is undefined when n == 0: /** * roundup_pow_of_two - round the given value up to nearest power of two * @n - parameter * * round the given value up to the nearest power of two * - the result is undefined when n == 0 -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
