This is the mail archive of the libc-alpha@sourceware.org mailing list for the glibc project.

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH] Suppress sign-conversion warning from FD_SET

On 05/29/2012 03:14 AM, Paul Pluzhnikov wrote:

>> > -  ({ unsigned long int __d = (d);                                          \
>> > +  ({ long int __d = (d);                                                   \
>> >      (__builtin_constant_p (__d)                                           \
>> >       ? (__d >= __FD_SETSIZE                                               \
              ^^^^^^^^^^^^^^^^^^^
>> >  	 ? __fdelt_warn (__d) : (__d / __NFDBITS))			    \


>> >
>> > This should check for __d < 0 as well.
>> > Or, to make it clearer, you might write it this way:
>> >
>> >    (0 <= d && d < __FD_SETSIZE

                   ^^^^^^^^^^^^^^^^

>> >     ? __d / __NFDBITS
>> >     : __fdelt_warn (__d))


This:

> +      ? (0 <= __d && __FD_SETSIZE < __d					    \

                        ^^^^^^^^^^^^^^^^^^
> +	 ? (__d / __NFDBITS)						    \
> +	 : __fdelt_warn (__d))						    \

looks like swapped by mistake.
You really want it as Paul Eggert wrote:

> +      ? (0 <= __d && __d < __FD_SETSIZE 					    \

IOW, same as mathematical "0 <= __d < __FD_SETSIZE".

-- 
Pedro Alves
Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]