On Wed, 7 Dec 2022 14:33:31 -0800
Tyler Retzlaff <roret...@linux.microsoft.com> wrote:
> On Wed, Dec 07, 2022 at 01:03:41PM -0800, Stephen Hemminger wrote:
> > On Wed, 7 Dec 2022 11:00:13 -0800
> > Tyler Retzlaff <roret...@linux.microsoft.com> wrote:
> >
> > > +static char lcore_names[RTE_MAX_LCORE][RTE_LCORE_NAME_MAX_LEN];
> >
> > This copy would redundant on Linux.
> >
> > > +
> > > +int
> > > +rte_lcore_set_name(unsigned int lcore_id, const char *name)
> > > +{
> > > + if (unlikely(lcore_id >= RTE_MAX_LCORE))
> > > + return -EINVAL;
> > > +
> > > + if (strlen(name) >= RTE_LCORE_NAME_MAX_LEN)
> > > + return -ERANGE;
> > > +
> > > + (void)strcpy(&lcore_names[lcore_id][0], name);
> >
> > Why the void cast?
>
> it's a common convention used in various open source projects indicating
> the that ignoring the return value is intentional as opposed to being
> sloppy or accidental.
>
> if it's a violation of dpdk style i'll remove it. but i have come across
> a lot of dpdk code where i honestly can't tell if it is on purpose or
> just sloppyness. (sticks out in code reviews too).
I think it is an old BSD lint ism.
Haven't seen it used in years.
