Re: glibc misc/sys/cdefs.h nonull - typo in comment

[Jonny Grant <jg@jguk.org>]

On 30/10/2023 10:10, Xi Ruoyao wrote:
> On Mon, 2023-10-30 at 10:04 +0100, Andreas Schwab wrote:
>> On Okt 29 2023, Jonny Grant wrote:
>>
>>> glibc does go beyond POSIX and set errno to EFAULT if a null pointer
>>> constant is passed.
>>> https://man7.org/linux/man-pages/man2/olduname.2.html
>>>
>>> Although I looked at glibc/posix/uname.c and it has EINVAL there,
>>> couldn't spot where the EFAULT comes from, probably there is another
>>> file.
>>
>> glibc never generates EFAULT on its own, it always comes from the
>> kernel.
> 
> And POSIX is clear that people should not rely on EFAULT, at all:
> 
> [EFAULT]
> Bad address. The system detected an invalid address in attempting to use
> an argument of a call. The reliable detection of this error cannot be
> guaranteed, and when not detected may result in the generation of a
> signal, indicating an address violation, which is sent to the process.
> 

Fair enough, as you and Andreas have said, the EFAULT is returned due to the kernel and POSIX doesn't state many functions that actually specify supporting a NULL pointer argument [a few eg. free(), realloc(), time()].

I usually just wrap calls to add extra checks when I'm not in control of how safe the rest of the user space software is, it is more tricky if 3rd party libraries are using symbols from libc directly.

Kind regards, Jonny