From: Leon Romanovsky [mailto:l...@kernel.org]
Sent: Tuesday, June 27, 2017 8:37 AM
>
> On Mon, Jun 26, 2017 at 09:06:51PM +0300, Michal Kalderon wrote:
> > iWARP personality introduced the need for differentiating in several
> > places in the code whether we are RoCE, iWARP or either. This leads to
> > introducing new macros for querying the personality.
> >
> > Signed-off-by: Michal Kalderon
> > Signed-off-by: Yuval Mintz
> > Signed-off-by: Ariel Elior
> >
> > ---
> > drivers/net/ethernet/qlogic/qed/qed.h | 26 +++--
> -
> > drivers/net/ethernet/qlogic/qed/qed_cxt.c | 8
> > drivers/net/ethernet/qlogic/qed/qed_dev.c | 12 +---
> > drivers/net/ethernet/qlogic/qed/qed_l2.c | 3 +--
> > drivers/net/ethernet/qlogic/qed/qed_ll2.c | 2 +-
> > drivers/net/ethernet/qlogic/qed/qed_main.c | 17 -
> > include/linux/qed/common_hsi.h | 2 +-
> > 7 files changed, 43 insertions(+), 27 deletions(-)
>
> I see that these changes are in Ethernet part of your driver, but for RDMA
> part,
> there are already available inline functions:
> rdma_protocol_iwarp, rdma_protocol_roce.
>
> Please avoid introducing new IS_IWARP/IS_ROCE macros and the decision
> should be taken on port level and not on device, despite the fact that
> probably
> your ib_device has only one port.
>
> Thanks
We still need to differentiate between iWARP/RoCE before
rdma_protocol_iwarp/rdma_protocol_roce will be valid
(allocating resources, registering ib device, initializing ib_device, and
capabilities.
For the rest of the places, after the port capabilities are valid, we'll use
the inline functions.
