Re: [PATCH] net/mlx5e: Move priv variable into case statement in mlx5e_setup_tc

From: Nathan Chancellor
Date: Thu Jul 11 2019 - 02:09:17 EST


On Wed, Jul 10, 2019 at 11:02:00PM -0700, Saeed Mahameed wrote:
> On Wed, Jul 10, 2019 at 12:05 PM Nathan Chancellor
> <natechancellor@xxxxxxxxx> wrote:
> >
> > There is an unused variable warning on arm64 defconfig when
> > CONFIG_MLX5_ESWITCH is unset:
> >
> > drivers/net/ethernet/mellanox/mlx5/core/en_main.c:3467:21: warning:
> > unused variable 'priv' [-Wunused-variable]
> > struct mlx5e_priv *priv = netdev_priv(dev);
> > ^
> > 1 warning generated.
> >
> > Move it down into the case statement where it is used.
> >
> > Fixes: 4e95bc268b91 ("net: flow_offload: add flow_block_cb_setup_simple()")
> > Link: https://github.com/ClangBuiltLinux/linux/issues/597
> > Signed-off-by: Nathan Chancellor <natechancellor@xxxxxxxxx>
> > ---
> > drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 7 ++++---
> > 1 file changed, 4 insertions(+), 3 deletions(-)
> >
> > diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
> > index 6d0ae87c8ded..651eb714eb5b 100644
> > --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
> > +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
> > @@ -3464,15 +3464,16 @@ static LIST_HEAD(mlx5e_block_cb_list);
> > static int mlx5e_setup_tc(struct net_device *dev, enum tc_setup_type type,
> > void *type_data)
> > {
> > - struct mlx5e_priv *priv = netdev_priv(dev);
> > -
> > switch (type) {
> > #ifdef CONFIG_MLX5_ESWITCH
> > - case TC_SETUP_BLOCK:
> > + case TC_SETUP_BLOCK: {
> > + struct mlx5e_priv *priv = netdev_priv(dev);
> > +
> > return flow_block_cb_setup_simple(type_data,
> > &mlx5e_block_cb_list,
> > mlx5e_setup_tc_block_cb,
> > priv, priv, true);
> > + }
>
> Hi Nathan,
>
> We have another patch internally that fixes this, and it is already
> queued up in my queue.
> it works differently as we want to pass priv instead of netdev to
> mlx5e_setup_tc_mqprio below,
> which will also solve warning ..
>
> So i would like to submit that patch if it is ok with you ?

Hi Saeed,

Whatever works best for you, I just care that the warning gets fixed,
not how it is done :) I wouldn't mind being put on CC so I can pick it
up for my local tests.

Thanks for the follow up!
Nathan