On Fri, Aug 24, 2018 at 05:46:42PM +0100, Ferruh Yigit wrote:
> On 8/24/2018 3:54 PM, Tomasz Duszynski wrote:
> > Fix used_bpools array initialization by using range initializer.
> > This way all necessary variables are properly initialized regardless
> > of PP2_NUM_PKT_PROC value.
> >
> > Fixes: 0ddc9b815b11 ("net/mrvl: add net PMD skeleton")
> > Cc: sta...@dpdk.org
> >
> > Signed-off-by: Tomasz Duszynski <t...@semihalf.com>
> > ---
> > drivers/net/mvpp2/mrvl_ethdev.c | 3 +--
> > 1 file changed, 1 insertion(+), 2 deletions(-)
> >
> > diff --git a/drivers/net/mvpp2/mrvl_ethdev.c
> > b/drivers/net/mvpp2/mrvl_ethdev.c
> > index 7ea8946..60ca6e0 100644
> > --- a/drivers/net/mvpp2/mrvl_ethdev.c
> > +++ b/drivers/net/mvpp2/mrvl_ethdev.c
> > @@ -84,8 +84,7 @@ static const char * const valid_args[] = {
> > static int used_hifs = MRVL_MUSDK_HIFS_RESERVED;
> > static struct pp2_hif *hifs[RTE_MAX_LCORE];
> > static int used_bpools[PP2_NUM_PKT_PROC] = {
> > - MRVL_MUSDK_BPOOLS_RESERVED,
> > - MRVL_MUSDK_BPOOLS_RESERVED
> > + [0 ... PP2_NUM_PKT_PROC - 1] = MRVL_MUSDK_BPOOLS_RESERVED
>
> range initializer is not part of standard, although I am sure this won't be
> first GCC extension we rely on, and I don't have any other option than memset
> for this.
>
> So just a reminder about extension usage, if you are happy with this
> implementation, lets keep it.
I would prefer to leave it as is. As you pointed out DPDK has
already started using it anyway.
--
- Tomasz Duszyński
