This new firmware for the qed* adapters fixes multiple issues:
- Prevent accidental drops in Tx-switching between VFs.
- Corrct VF cleanup for higher VFs.
- Better handling of RoCE error flows.
- MR registration/deregistration corrections.
- Prevent possible HW attention during offloaded TCP teardown.
- Corrections to iSCSI retransmit flows.
In addition, this firmware is a requirement for supporting the
QL41xxx series of adapters, as it contains the necessary init
and firmware logic for those.
Signed-off-by: Yuval Mintz <Yuval.Mintz@cavium.com>
Signed-off-by: Kyle McMartin <kyle@kernel.org>
This new firmware for the qed* adpaters fixes several issues:
- Better blocking of malicious VFs.
- After FLR, Tx-switching [internal routing] of packets might
be incorrect.
- Deletion of unicast MAC filters would sometime have side-effect
of corrupting the MAC filters configred for a device.
It also contains fixes for future qed* drivers that *hopefully* would be
sent for review in the near future.
In addition, it would allow driver some new functionality, including:
- Allowing PF/VF driver compaitibility with old drivers [running
pre-8.10.5.0 firmware].
- Better debug facilities.
Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: Kyle McMartin <kyle@kernel.org>
The new QED firmware contains several new additions;
As for bug-fixes, it fixes a couple of issues:
- Wrong classification of packets in 4-port devices.
- Anti-spoof interoperability with encapsulated packets.
- Tx-switching of encapsulated packets.
It also slightly improves Tx performance in several scenarios.
>From the driver's feature-set POV, this firmware would
allow us to add 2 major enhancements in the near future:
- It would allow us produce a register dump from the device,
as well as other debug-related information.
- It should be sufficient for our iscsi driver, which we [hopefully]
plan to push upstream soon.
Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: Kyle McMartin <kyle@kernel.org>
This patch pushes the next firmware for the QED module.
This fixes numerous bugs, most notably:
- FLR flow isn't properly working.
- Chip might become unrespronsive after closing a vport.
- Chip gets stuck when transmitting packet with length > mtu.
It would also allow us to push future features such as:
- Vlan filter offload.
- Encapsulation offload.
- Single Root IOV support.
- Timesync.
- Non-L2 protocols.
>From L2 perspecitve, it's also note-worthy to state that it:
- Changes fastpath HSI - fastpath will be moved to into using
Jumbo-over-BD.
- Improves throughput on single connection traffic.
Signed-off-by: Sudarsana Reddy Kalluru <Sudarsana.Kalluru@qlogic.com>
Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: Kyle McMartin <kyle@kernel.org>
Yuval Mintz