Using napi_complete_done() instead of napi_complete() allows
us to use /sys/class/net/ethX/gro_flush_timeout

GRO layer can aggregate more packets if the flush is delayed a bit,
without having to set too big coalescing parameters that impact
latencies.

Tested:

lpx:~# echo 0 >/sys/class/net/eth1/gro_flush_timeout

lpx:~# sar -n DEV 1 10 | grep eth1
10:36:25 AM      eth1  81290.00  40617.00 120479.67   2777.01      0.00      0.00      0.00
10:36:26 AM      eth1  81283.00  40608.00 120481.81   2778.13      0.00      0.00      1.00
10:36:27 AM      eth1  81304.00  40639.00 120518.42   2778.28      0.00      0.00      0.00
10:36:28 AM      eth1  81255.00  40605.00 120437.34   2775.95      0.00      0.00      1.00
10:36:29 AM      eth1  81306.00  40630.00 120521.44   2777.70      0.00      0.00      0.00
10:36:30 AM      eth1  81286.00  40564.00 120480.20   2773.31      0.00      0.00      0.00
10:36:31 AM      eth1  81256.00  40599.00 120438.81   2776.27      0.00      0.00      0.00
10:36:32 AM      eth1  81287.00  40594.00 120480.69   2776.69      0.00      0.00      0.00
10:36:33 AM      eth1  81279.00  40601.00 120478.53   2775.84      0.00      0.00      0.00
10:36:34 AM      eth1  81277.00  40610.00 120476.94   2776.25      0.00      0.00      0.00
Average:         eth1  81282.30  40606.70 120479.39   2776.54      0.00      0.00      0.20

lpx:~# echo 13000 >/sys/class/net/eth1/gro_flush_timeout

lpx:~# sar -n DEV 1 10 | grep eth1
10:36:43 AM      eth1  81257.00   7747.00 120437.44    530.00      0.00      0.00      0.00
10:36:44 AM      eth1  81278.00   7748.00 120480.00    529.85      0.00      0.00      0.00
10:36:45 AM      eth1  81282.00   7752.00 120479.09    531.09      0.00      0.00      0.00
10:36:46 AM      eth1  81282.00   7751.00 120478.80    530.90      0.00      0.00      0.00
10:36:47 AM      eth1  81276.00   7745.00 120478.31    529.64      0.00      0.00      0.00
10:36:48 AM      eth1  81278.00   7747.00 120478.50    529.81      0.00      0.00      0.00
10:36:49 AM      eth1  81282.00   7749.00 120478.88    530.01      0.00      0.00      0.00
10:36:50 AM      eth1  81284.00   7751.00 120481.52    530.20      0.00      0.00      0.00
10:36:51 AM      eth1  81299.00   7769.00 120481.74    533.81      0.00      0.00      0.00
10:36:52 AM      eth1  81281.00   7748.00 120478.62    529.96      0.00      0.00      0.00
Average:         eth1  81279.90   7750.70 120475.29    530.53      0.00      0.00      0.00
Signed-off-by: NEric Dumazet <edumazet@google.com>
Acked-by: NMichael Chan <mchan@broadcom.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

As pointed out by Ben Hutchings, ioremap uses unsigned long as
its parameter type, so we should be using that instead of u32
or int.
Reported-by: NBen Hutchings <ben@decadent.org.uk>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

To allow for better scalability on systems with large core counts, we
will try and allocate enough RDMA Concentrator IQs and MSI/X vectors as
we have cores. If we cannot get enough MSI/X vectors, fall back to the
minimum required: 1 per adapter rx channel.

Also clean up cxgb_enable_msix() to make it readable and correct a bug
where the vectors are not correctly assigned if the driver doesn't get
the full amount requested.
Signed-off-by: NSteve Wise <swise@opengridcomputing.com>
Signed-off-by: NHariprasad Shenai <hariprasad@chelsio.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Adds a common function for all Rx queue allocation.
Signed-off-by: NHariprasad Shenai <hariprasad@chelsio.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Bump i40e to 1.2.11 and i40evf to 1.2.5

Change-ID: Ie13375941606b0a027e5b5dbc235f5f5f03b75c8
Signed-off-by: NSravanthi Tangeda <sravanthi.tangeda@intel.com>
Tested-by: NJim Young <james.m.young@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

The PF driver spams the system log with messages about VF VSI when VFs
are created, as well as each time they are reset. This is annoying, and
the information isn't even useful most of the time.

Remove this message to reduce user annoyance.

Change-ID: I8de90d05380f54b038c9c8c3265150be87c9242c
Signed-off-by: NMitch Williams <mitch.a.williams@intel.com>
Tested-by: NJim Young <james.m.young@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Move the IRQ tracking setup and teardown into the same routines that
do the IRQ setup and teardown.  This keeps like activities together and
allows us to track exactly the number of vectors reserved from the OS,
which may be fewer than are available from the HW.

Change-ID: I6b2b1a955c5f0ac6b94c3084304ed0b2ea6777cf
Signed-off-by: NShannon Nelson <shannon.nelson@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

For future device support we do not want to map the whole CSR space since some
of it is mapped by other drivers with different mapping methods.

Note: As a side effect, the flash region (if exposed through the memory map)
gets unmapped too since it follows the future use region.

Change-ID: Ic729a2eacd692984220b1a415ff4fa0f98ea419a
Signed-off-by: NAnjali Singhai Jain <anjali.singhai@intel.com>
Signed-off-by: NJesse Brandeburg <jesse.brandeburg@intel.com>
Tested-by: NJim Young <james.m.young@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Fix some double blank lines and un-split a function declaration that all
fits on one line. Also make i40e_get_priv_flags static.

Change-ID: I11b5d25d1153a06b286d0d2f5d916d7727c58e4a
Signed-off-by: NJesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: NNeerav Parikh <neerav.parikh@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Add the 10G and 40G AOC PHY types to the case statement in get_media_type
and ethtool get_settings so that the correct information gets reported
back to the user.

Change-ID: I1b4849d22199a9acf7c8807166d0317c1faad375
Signed-off-by: NCatherine Sullivan <catherine.sullivan@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

If the system administrator is requesting an offline diagnostic test using
'ethtool -t' then we should, you know, actually take the device offline
before doing the testing.

Change-ID: I6afa1cbfcc821c9ab6e6f47ed4d8dc2d8dd20e82
Signed-off-by: NGreg Rose <gregory.v.rose@intel.com>
Tested-by: NJim Young <james.m.young@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Some FW versions are incorrectly reporting a breakout cable as PHY type
0x3 when it should be 0x16 (I40E_PHY_TYPE_10GBASE_SFPP_CU).
If we get this value back from FW and the version is < 4.40, reassign it
to I40E_PHY_TYPE_10GBASE_SFPP_CU.

Change-ID: Ibb41a0e3cd2c0753744e8553959240df6ed13ae8
Signed-off-by: NCatherine Sullivan <catherine.sullivan@intel.com>
Tested-by: NJim Young <james.m.young@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

During resets (possibly caused by a Tx hang) the driver would
accidentally clear the XPS mask for all queues back to 0.

This caused higher CPU utilization and had some other performance impacts
for transmit tests.

Change-ID: I95f112432c9e643a153eaa31cd28cdcbfdd01831
Signed-off-by: NJesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Use automatic sign extension by replacing 0xffff... constants
with ~(u64)0 or ~(u32)0.

Change-ID: I73cab4cd2611795bb12e00f0f24fafaaee07457c
Signed-off-by: NJesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: NKevin Scott <kevin.c.scott@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

0x2A is the NVM version so it has useful data but it is per image
version every image can have a different one. 0x18 is the dev starter
version which all the images for release will have the same version.
Of the two 0x18 is more useful and is what should be displayed.

Change-ID: Idf493da13a42ab211e2de0bef287f5de51033cca
Signed-off-by: NShannon Nelson <shannon.nelson@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

In CEE mode the firmware does not set the operational status bit of
the application TLV status as returned from the "Get CEE DCBX Oper Cfg"
AQ command. This occurs whenever a DCBX configuration is changed.

This is a workaround to remove the check for the operational and sync bits
of the application TLV status till a firmware fix is provided.

Change-ID: I1a31ff2fcadcb06feb5b55776a33593afc6ea176
Signed-off-by: NNeerav Parikh <neerav.parikh@intel.com>
Acked-by: NShannon Nelson <shannon.nelson@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Modify our get and set LED functions so they ignore activity LEDs,
as we are required to blink the link LEDs only.

Change-ID: I647ea67a6fc95cbbab6e3cd01d81ec9ae096a9ad
Signed-off-by: NMatt Jared <matthew.a.jared@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Recent changes to the driver initialization have caused the BW
configurations to not take effect.  We use a BW configuration read and
write back to "kick" the Tx scheduler into action.

Change-ID: I94ab377c58d3a3986e3de62b6c199be3fd2ee5e6
Signed-off-by: NGreg Rose <gregory.v.rose@intel.com>
Tested-by: NJim Young <james.m.young@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

1. Use c_index and ring->c_index to determine how many TxCBs/TxBDs are
   ready for cleanup
   - c_index = the current value of TDMA_CONS_INDEX
   - TDMA_CONS_INDEX is HW-incremented and auto-wraparound (0x0-0xFFFF)
   - ring->c_index = __bcmgenet_tx_reclaim() cleaned up to this point on
     the previous invocation

2. Add bcmgenet_tx_ring->clean_ptr
   - index of the next TxCB to be cleaned
   - incremented as TxCBs/TxBDs are processed
   - value always in range [ring->cb_ptr, ring->end_ptr]

3. Fix incrementing of dev->stats.tx_packets
   - should be incremented only when tx_cb_ptr->skb != NULL

These changes simplify __bcmgenet_tx_reclaim(). Furthermore, Tx ring size
can now be any value.

With the old code, Tx ring size had to be a power-of-2:
   num_tx_bds = ring->size;
   c_index &= (num_tx_bds - 1);
   last_c_index &= (num_tx_bds - 1);
Signed-off-by: NPetri Gynther <pgynther@google.com>
Reviewed-by: NFlorian Fainelli <f.fainelli@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch implements the .sriov_configure() PCI method to allow for
runtime enabling/disabling of VFs. The module param "num_vfs" is now
deprecated.
At the time of driver load the PF-pool resources are allocated to the PF.
When the user enables VFs, the resources are then re-distributed across
PFs and VFs based on the number of VFs enabled.
Signed-off-by: NVasundhara Volam <vasundhara.volam@emulex.com>
Signed-off-by: NSathya Perla <sathya.perla@emulex.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When SR-IOV is enabled in the adapter, the FW distributes resources
evenly across the PF and it's VFs. This is currently done only for some
resources.

This patch adds support for a new cmd that queries the FW for the list
of resources for which the distribution is allowed and distributes them
accordingly.
Signed-off-by: NVasundhara Volam <vasundhara.volam@emulex.com>
Signed-off-by: NSathya Perla <sathya.perla@emulex.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

On BE2, BE3 and Skhawk-R chips one non-RSS (called "default") RXQ was
needed to receive non-IP traffic. Some FW versions now export a
capability called IFACE_FLAGS_DEFQ_RSS where this requirement doesn't hold.
On such FWs the driver now does not create the non-RSS default queue.
This prevents wasting one RXQ per VF.
Signed-off-by: NVasundhara Volam <vasundhara.volam@emulex.com>
Signed-off-by: NSathya Perla <sathya.perla@emulex.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Remove Kconfig dependency and enable driver for
all ARCHs.
Signed-off-by: NMichal Simek <michal.simek@xilinx.com>
Acked-by: NSören Brinkmann <soren.brinkmann@xilinx.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The statistics registers have write-clear behaviour, which means we
will lose any increment between the read and write.  Mitigate this by
only clearing when we read a non-zero value, so we will never falsely
report a total of zero.  This also saves time as we only handle
error statistics here and they won't often be incremented.
Signed-off-by: NBen Hutchings <ben.hutchings@codethink.co.uk>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NBen Hutchings <ben.hutchings@codethink.co.uk>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

There are many different sets of registers implemented by the
different versions of this controller, and we can only expect this to
get more complicated in future.  Limit how much ethtool needs to know
by including an explicit bitmap of which registers are included in the
dump, allowing room for future growth in the number of possible
registers.

As I don't have datasheets for all of these, I've only included
registers that are:

- defined in all 5 register type arrays, or
- used by the driver, or
- documented in the datasheet I have

Add one new capability flag so we can tell whether the RTRATE
register is implemented.

Delete the TSU_ADRL0 and TSU_ADR{H,L}31 definitions, as they weren't
used and the address table is already assumed to be contiguous.
Signed-off-by: NBen Hutchings <ben.hutchings@codethink.co.uk>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently we may silently read/write a register at offset 0.  Change
this to WARN and then ignore the write or read-back all-ones.
Signed-off-by: NBen Hutchings <ben.hutchings@codethink.co.uk>
Acked-by: NSergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

At least on the R8A7790, RFS8 reflects the RINT8 (multicast) MAC
status flag.
Signed-off-by: NBen Hutchings <ben.hutchings@codethink.co.uk>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Use the built-in function instead of memset.
Signed-off-by: NJoe Perches <joe@perches.com>
Acked-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Module dependencies are broken in the case where CONFIG_I40E=y and
CONFIG_CONFIGFS_FS=m.  This fixes the broken dependency.
Signed-off-by: NGreg Rose <gregory.v.rose@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Commit 6ce29b0e ("gianfar: Avoid unnecessary reg accesses in adjust_link()")
eliminates unnecessary calls to adjust_link for phy devices which don't support
interrupts and need polling. As part of that work, the 'new_state' local flag,
which was used to reduce logging noise on the console, was eliminated.

Unfortunately, that means that a 'Link is Down' log message will now be
issued continuously if a link is configured as UP, the link state is down,
and the associated phy requires polling. This occurs because priv->oldduplex
is -1 in this case, which always differs from phydev->duplex. In addition,
phydev->speed may also differ from priv->oldspeed.  gfar_update_link_state()
is therefore called each time a phy is polled, even if the link state did not
change.

Cc: Claudiu Manoil <claudiu.manoil@freescale.com>
Signed-off-by: NGuenter Roeck <linux@roeck-us.net>
Reviewed-by: NClaudiu Manoil <claudiu.manoil@freescale.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add a function that will enable changing the MAC address
of an ibmveth interface while it is still running.
Signed-off-by: NThomas Falcon <tlfalcon@linux.vnet.ibm.com>
Reviewed-by: NJiri Pirko <jiri@resnulli.us>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Bump i40e to 1.2.10 and i40evf to 1.2.4

Change-ID: I48aa64df05fcc8356e7026f3a9e69ecf78d0c785
Signed-off-by: NSravanthi Tangeda <sravanthi.tangeda@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

In case of MFP mode the driver was returning incorrect number of TCs
for partitions that are not enabled for iSCSI. Though the driver does
not configure these TCs in the Tx scheduler for the NIC partitions;
it does use this map to setup the queue mappings.

This patch fixes this and keeps all the NIC partitions to the default
PF TC i.e. TC0.

Change-ID: Iede214c907e7bac1356e999049b9f642759512b3
Signed-off-by: NNeerav Parikh <neerav.parikh@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Allow DCBNL operations in MFP mode to allow query of port DCB settings
via all the PFs and register iSCSI APP on iSCSI enabled PF.

Change-ID: I34cc39b4665d0a631847d4079350d3814f02381e
Signed-off-by: NNeerav Parikh <neerav.parikh@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Add an extra check to make sure that the indirection table pointer is
valid before dereferencing it.

Change-ID: I698adbf3daff03081d01f489dc95a9f1ad8b12f1
Reported-by: NBen Hutchings <ben@decadent.org.uk>
Signed-off-by: NMitch Williams <mitch.a.williams@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

With changes to default number of queue pairs that the interface comes up with
from 1 per online CPU to 1 per lan_msix, we need to make sure we recalculate
rss_size. We will now recalculate rss_size based on number of queues enabled in
the VSI.

Without this fix if the max_lan_msix < num_online_cpu we will be coming up
with fewer queues but will be populating rss_size based on num_online_cpus.
This will result in packets getting silently dropped because RSS LUT has queues
that are not enabled.

Change-ID: Ifac8796ce1be1758bb0c34f38dbf4a3a76621e76
Signed-off-by: NAnjali Singhai Jain <anjali.singhai@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Since all VSIs don't have the same RSS table size,
have one for each VSI instead of having a single define
for RSS table size

Change-ID: Ic2c7c66e4a389d4b6c8841a707510a9735041f02
Signed-off-by: NAnjali Singhai Jain <anjali.singhai@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

We were missing a few packet types for VXLAN offload. This patch fixes
that.

Change-ID: I4b23aa0b08e40ed49d0df6c49a5ed9f2009b44ce
Signed-off-by: NAnjali Singhai Jain <anjali.singhai@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This patch fixes an issue where the driver is not flushing out the
DCBNL app table for applications that are not present in the local
DCBX application configuration TLVs.

Change-ID: I1f1ee04c81c145071b2ab15657546eb10b81fadb
Signed-off-by: NNeerav Parikh <neerav.parikh@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>