[ovs-dev] [PATCH v2 3/3] dpdk: Use ovs-numa provided functions to manage thread affinity.

[David Marchand]

On Mon, Sep 2, 2019 at 3:38 PM Ilya Maximets <i.maximets at samsung.com> wrote:
>
> On 02.09.2019 16:17, David Marchand wrote:
> > On Mon, Sep 2, 2019 at 3:15 PM Ilya Maximets <i.maximets at samsung.com> wrote:
> >>
> >> On 02.09.2019 16:09, David Marchand wrote:
> >>> On Mon, Sep 2, 2019 at 1:27 PM Ilya Maximets <i.maximets at samsung.com> wrote:
> >>>>
> >>>> This allows to decrease code duplication and avoid using Linux-specific
> >>>> functions (this might be useful in the future if we'll try to allow
> >>>> running OvS+DPDK on FreeBSD).
> >>>>
> >>>> Signed-off-by: Ilya Maximets <i.maximets at samsung.com>
> >>>> Acked-by: William Tu <u9012063 at gmail.com>
> >>>> ---
> >>>>  lib/dpdk.c | 27 ++++++++++++---------------
> >>>>  1 file changed, 12 insertions(+), 15 deletions(-)
> >>>>
> >>>> diff --git a/lib/dpdk.c b/lib/dpdk.c
> >>>> index fc58de55a..6f297d918 100644
> >>>> --- a/lib/dpdk.c
> >>>> +++ b/lib/dpdk.c
> >>>> @@ -275,7 +275,7 @@ dpdk_init__(const struct smap *ovs_other_config)
> >>>>      int result;
> >>>>      bool auto_determine = true;
> >>>>      int err = 0;
> >>>
> >>> Nit: err can be removed.
> >>
> >> It's used 2 times in this function.
> >> We could only avoid initializing it here.
> >
> > Yes, but we do not care about err itself, we are just checking
> > function return values.
>
> I see. It seems for me more like a separate change. This could be
> done, for example, along with improving stat/pdump error messages.

This patch removed the only reason to keep this variable from my pov.
But yes, it can go in a cleanup later.


-- 
David Marchand