Re: memcmp-tests module

[Ben Pfaff]

Eric Blake <address@hidden> writes:

> I think that it is a bug to attach the nonnull attribute to
> memcmp parameters (for the very reason that a compare of 0
> bytes should not dereference the pointer, so passing NULL is
> not necessarily a bug).

I am not sure that passing a null pointer to memcmp is valid even
if the number of bytes to compare is zero.

C99 7.21.1 "String function conventions" says, in part:

     Where an argument declared as size_t n specifies the length
     of the array for a function, n can have the value zero on a
     call to that function.  Unless explicitly stated otherwise
     in the description of a particular function in this
     subclause, pointer arguments on such a call shall still have
     valid values, as described in 7.1.4.

C99 7.1.4 says explicitly that a null pointer is an invalid
value:

     Each of the following statements applies unless explicitly
     stated otherwise in the detailed descriptions that follow:
     If an argument to a function has an invalid value (such as a
     value outside the domain of the function, or a pointer
     outside the address space of the program, or a null pointer,
     or a pointer to non-modifiable storage when the
     corresponding parameter is not const-qualified) or a type
     (after promotion) not expected by a function with variable
     number of arguments, the behavior is undefined.

The description of memcmp doesn't have any clarification.  I
don't see anything more specific in SUSv3 either.
-- 
Ben Pfaff 
http://benpfaff.org