[LTP] [PATCH 1/4] syscalls/sync01: Remove it

[Petr Vorel <pvorel@suse.cz>]

Hi,

> On 11/7/20 12:47 AM, Cyril Hrubis wrote:
> > Hi!
> > > I have a doubt after reading Xu's patch[1] and Martin's patch[2]:

> > > 1) Xu removed sync01 because sync() always return 0.
> > Actually sync() is defined as void function, so the tests were bogusly
> > checking the TST_RET value which haven't been set at all.

> Hi Cyril,

> Oops, I gave a wrong example. :-(

> On error, I just wonder if we need to check all return value(i.e. negative
> value except -1).

> IOW, Is it possible for syscall to get a error value which is not -1?
There are probably other examples, but I've found only these:

man malloc_get_state(3)

	If the implementation detects that state does not point to a correctly
	formed data structure, malloc_set_state() returns -1.
	If the implementation detects that the version of the data structure
	referred to by state is a more recent version than this implementation knows
	about, malloc_set_state() returns -2.

man mmap(2)
	On error, the value MAP_FAILED (that is, (void *) -1) is returned.

> Best Regards,

> Xiao Yang


Kind regards,
Petr