Re: [PATCH rdma-rc 4/7] IB/mlx4: Fix possible vl/sl field mismatch in LRH header in QP1 packets

On 9/18/2016 2:53 AM, Leon Romanovsky wrote:

On Fri, Sep 16, 2016 at 02:01:57PM -0400, Doug Ledford wrote:

quoted

On 9/12/2016 12:16 PM, Leon Romanovsky wrote:

quoted

From: Jack Morgenstein <redacted>

In MLX qp packets, the LRH (built by the driver) has both a VL field
and an SL field. When building a QP1 packet, the VL field should
reflect the SLtoVL mapping and not arbitrarily contain zero (as is
done now). This bug causes credit problems in IB switches at
high rates of QP1 packets.

The fix is to cache the SL to VL mapping in the driver, and look up
the VL mapped to the SL provided in the send request when sending
QP1 packets.

For FW versions which support generating a port_management_config_change
event with subtype sl-to-vl-table-change, the driver uses that event
to update its sl-to-vl mapping cache.  Otherwise, the driver snoops
incoming SMP mads to update the cache.

There remains the case where the FW is running in secure-host mode
(so no QP0 packets are delivered to the driver), and the FW does not
generate the sl2vl mapping change event. To support this case, the
driver updates (via querying the FW) its sl2vl mapping cache when
running in secure-host mode when it receives either a Port Up event
or a client-reregister event (where the port is still up, but there
may have been an opensm failover).
OpenSM modifies the sl2vl mapping before Port Up and Client-reregister
events occur, so if there is a mapping change the driver's cache will
be properly updated.

Fixes: 225c7b1feef1 ("IB/mlx4: Add a driver Mellanox ConnectX InfiniBand adapters")
Signed-off-by: Jack Morgenstein <redacted>
Signed-off-by: Leon Romanovsky <redacted>
---
 drivers/infiniband/hw/mlx4/mad.c        |  64 ++++++++++++++++++-
 drivers/infiniband/hw/mlx4/main.c       | 110 +++++++++++++++++++++++++++++++-
 drivers/infiniband/hw/mlx4/mlx4_ib.h    |   7 ++
 drivers/infiniband/hw/mlx4/qp.c         |  23 ++++++-
 drivers/net/ethernet/mellanox/mlx4/fw.c |  13 ++--
 include/linux/mlx4/device.h             |  13 +++-
 6 files changed, 220 insertions(+), 10 deletions(-)

That's a lot more code churn than I would like to see in a late RC.  I'm
going to drop this patch and move it to 4.9 instead.  If this fixed an
oops or something like that, I would be more open to taking it now, but
the problem being resolved is credits on a switch.  That isn't the sort
of showstopper issue that would justify this large of a patch this late
in the cycle.

No problem, Thanks.

This is now in my 4.9-misc branch.


-- 
Doug Ledford [off-list ref]
    GPG Key ID: 0E572FDD