On 2023-05-20 03:19, Xi Ruoyao via Libc-alpha wrote:
> If a LTP test relies on a EFAULT here, then LTP has a bug.  You should
> tell LTP to fix it then.

LTP fstat03 allows either EFAULT or SIGSEGV, but no other behavior, when 
'stat' is passed a null pointer.

Therefore that test is too strict for POSIX, as the POSIX behavior is 
undefined.

It's also too strict for GNU/Linux, because it insists on SIGSEGV 
whereas any terminating signal will do (e.g., SIGABRT).

Proposed LTP patch attached.