This change moves all of the ring flags into a single value.  The advantage
to this is that there is one central area for all of these flags and they
can all make use of the set/test bit operations.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NAaron Brown  <aaron.f.brown@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

There are a number of places where we have values that are stored as u16
but are being converted to int unnecessarily.  In order to avoid that we
should convert all variables that deal with the next_to_clean, next_to_use,
and count to u16 values.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NAaron Brown  <aaron.f.brown@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This change is meant to update the ring and vector allocations so that they
are per node instead of allocating everything on the node that
ifconfig/modprobe is called on.  By doing this we can cut down
significantly on cross node traffic.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NAaron Brown  <aaron.f.brown@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Instead of storing most of the data for the TX hot path in the stack until
we are ready to write the descriptor we can save ourselves some time and
effort by pushing the SKB, tx_flags, gso_size, bytecount, and protocol into
the first igb_tx_buffer since that is where we will end up putting it
anyway.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NAaron Brown  <aaron.f.brown@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This change will combine the writes of tx_buffer_info and the Tx data
descriptors into a single function. The advantage of this is that we can
avoid needless memory reads from the buffer info struct and speed things up
by keeping the accesses to the local registers.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NAaron Brown  <aaron.f.brown@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This change is meant to combine all of the TX flags fields into one u32
flags field so that it can be stored into the tx_buffer_info structure.
This includes the time stamp flag as well as mapped_as_page flag info.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NAaron Brown  <aaron.f.brown@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This change is meant to cleanup the protocol handling in the transmit path
so that it correctly offloads software VLAN tagged frames.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NAaron Brown  <aaron.f.brown@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This change is meant to improve the readability of the driver by separating
out the cmd_type configuration and the olinfo configuration into their own
functions.  By doing this it is much easier to determine which ingredients
go into setting up these to portions of the descriptor.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NAaron Brown  <aaron.f.brown@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This change converts two tx_buffer_info index values into pointers.  The
advantage to this is that we reduce unnecessary computations and in the case
of next_to_watch we get an added bonus of the value being able to provide
additional information as a NULL value indicates it is unset versus a 0 not
having any meaning for the index value.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NAaron Brown  <aaron.f.brown@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This patch is meant to simplify the transmit path by reducing the overhead
for creating a transmit context descriptor.  The current implementation is
split with igb_tso and igb_tx_csum doing two separate implementations on
how to setup the tx_buffer_info structure and the tx_desc.  By combining
them it is possible to reduce code and simplify things since now only one
function will create context descriptors.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NAaron Brown  <aaron.f.brown@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

In order to be able to improve the performance of the TX path it has been
necessary to add addition info to the tx_buffer_info structure.  However a
side effect is that the structure has gotten larger and this in turn has
also increased the size of the RX buffer info structure.  In order to avoid
this in the future I am splitting the single buffer_info structure into two
separate ones and instead I will join them by making the buffer_info
pointer in the ring a union of the two.
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Tested-by: NAaron Brown  <aaron.f.brown@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This change is to make the NAPI budget limits for transmit
adjustable.  Currently they are only set to 128, and when
the changes/improvements to NAPI occur to allow for adjustability,
it would be possible to tune the value for optimal
performance with applications such as routing.

v2: remove tie between NAPI and interrupt moderation
    fix work limit define name (s/IXGBE/IGB/)
    Update patch description to better reflect patch
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Tested-by: NAaron Brown <aaron.f.brown@intel.com>

When short packets are received with jumbos enabled on 82579, they can be
interpreted to have a receive address that does not match any configured
address.  This is due to a hardware bug that can be worked around by
reducing the number of IPG octets added when the packet is transferred from
the PHY to the MAC.
Signed-off-by: NBruce Allan <bruce.w.allan@intel.com>
Tested-by: NJeff Pieper <jeffrey.e.pieper@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

The e1000 driver when running with lockdep could run into
some possible deadlocks between the work items acquiring
rtnl and the rtnl lock being acquired before work items
were cancelled.

Use a private mutex to make sure lock ordering isn't violated.
The private mutex is only used to protect areas not generally
covered by the rtnl lock already.
Signed-off-by: NJesse Brandeburg <jesse.brandeburg@intel.com>
CC: Thomas Gleixner <tglx@linutronix.de>
CC: Tushar Dave <tushar.n.dave@intel.com>
Tested-by: NAaron Brown <aaron.f.brown@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

With the previous commit, there are several functions
that are only ever called from thread context, and are
able to sleep with msleep instead of mdelay.
Signed-off-by: NJesse Brandeburg <jesse.brandeburg@intel.com>
CC: Thomas Gleixner <tglx@linutronix.de>
CC: Tushar Dave <tushar.n.dave@intel.com>
Tested-by: NAaron Brown <aaron.f.brown@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Thomas Gleixner (tglx) reported that e1000 was delaying for many milliseconds
(using mdelay) from inside timer/interrupt context.  None of these paths are
performance critical and can be moved into threads/work items.  This patch
implements the work items and the next patch changes the mdelays to msleeps.
Signed-off-by: NJesse Brandeburg <jesse.brandeburg@intel.com>
CC: Thomas Gleixner <tglx@linutronix.de>
CC: Tushar Dave <tushar.n.dave@intel.com>
Tested-by: NAaron Brown <aaron.f.brown@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Bump the version string to better match pair up with the out of tree
driver that contains the same functionality.
Signed-off-by: NDon Skidmore <donald.c.skidmore@intel.com>
Tested-by: NPhil Schmitt <phillip.j.schmitt@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Only function 1 has support for Alternate MAC Address in the EEPROM before,
this update now allow function 2 and 3 to have support for Alternate MAC
Address in the EEPROM.
Signed-off-by: N"Akeem G. Abodunrin" <akeem.g.abodunrin@intel.com>
Tested-by: NAaron Brown  <aaron.f.brown@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This code check word 0x37 in the EEPROM, if it is 0xFFFF _or_ 0x0000, then
there is no Alternate MAC Address in the EEPROM.
Signed-off-by: N"Akeem G. Abodunrin" <akeem.g.abodunrin@intel.com>
Tested-by: NAaron Brown  <aaron.f.brown@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This patch fixes "overwrite" problem. without this fix, SFP I2C EEPROM
data, which is located at A0 can be overwritten by the phy write function.
Signed-off-by: N"Akeem G. Abodunrin" <akeem.g.abodunrin@intel.com>
Tested-by: NAaron Brown <aaron.f.brown@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Receiving PFC (priority flow control) frames while the feature
is off should not pause the traffic class. On the X540 devices
the traffic class react to frames if it was previously enabled
because the field is incorrectly cleared.
Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Tested-by: NRoss Brattain <ross.b.brattain@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

X540 devices can only support up to 4 traffic classes and
guarantee a "lossless" traffic class on some platforms.
This patch sets the X540 devices to initialize a max
traffic class value of 4 at probe time.
Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Tested-by: NRoss Brattain <ross.b.brattain@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This patch correctly configures DCB when less than 8 traffic classes
are available in hardware.
Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Tested-by: NRoss Brattain <ross.b.brattain@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Fix PFC mask generation to OR in only a single bit for each priority in
the PFC mask returned via netlink.
Signed-off-by: NMark Rustad <mark.d.rustad@intel.com>
Tested-by: NRoss Brattain <ross.b.brattain@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

PCI device ID 0x1501 has a hardware bug when the link downshifts for
whatever reason which requires a workaround.  The workaround already exists
for other similar devices but is not called for 0x1501 (it should be called
for any ICH8-based device that uses a GbE PHY).  There is also one other
instance when the workaround should be called - after disabling gigabit
speed when going to Sx.
Signed-off-by: NBruce Allan <bruce.w.allan@intel.com>
Tested-by: NJeff Pieper <jeffrey.e.pieper@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

During suspend, the PHY must be reset for workaround updates to take effect
without restarting auto-negotiation.  Also, set the disable GbE and enable
Low Power Link Up (LPLU) if the EEPROM is configured to do likewise in
either D0 or non-D0a instead of just the latter.
Signed-off-by: NBruce Allan <bruce.w.allan@intel.com>
Tested-by: NJeff Pieper <jeffrey.e.pieper@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

82598 and 82599 do not ship with this type of PHY
Signed-off-by: NEmil Tantilov <emil.s.tantilov@intel.com>
Tested-by: NStephen Ko <stephen.s.ko@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

There is a problem in the ixgbe driver with the reporting of the flow
control parameters. The autoneg parameter is shown to be of if
*either* it really is off, or current modes for both tx and rx are off.

The problem is seen when the parameters are read or set when the link
is down. In this case, the driver sees that tx and rx are currently off
and therefore autoneg parameter is incorrectly reported to be off too.
Also, the ethtool binary can not set the autoneg off since it sees that
it already is. When a link later comes up, the autonegotiation is
carried out normally and the driver later on reports the autoneg
parameter to be on (as it is) and then it can also be changed with
ethtool.

The patch is made against v3.0 kernel, but the problem seems to be there
since v2.6.30-rc1.

Reviewer comments: What we are trying to do is to disable flow control
while the cable is disconnected. Since ixgbe defaults to full flow
control, we call ethtool -A autoneg off rx off tx off while the cable
is disconnected. This doesn't work, because the driver sets
hw->fc.current_mode = ixgbe_fc_none if the cable is unplugged.
ixgbe_get_pauseparam() then reports to ethtool that nothing needs to be
done. The code fixes this, but it might have some unknown consequences.
Signed-off-by: NMika Lansirinne <mika.lansirinne@stonesoft.com>
Reviewed-by: NEsa-Pekka Pyokkimies <esa-pekka.pyokkimies@stonesoft.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Disabling flow control in ixgbe_check_mac_link() results in incorrect
reporting by ethtool when link goes down, so remove it.
Signed-off-by: NEmil Tantilov <emil.s.tantilov@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

MFLCN register is used to set Rx flow control on parts newer than 82598.

This patch sends the value of MFLCN to ethtool, so it can be used in a
register dump (ethtool -d).
Signed-off-by: NEmil Tantilov <emil.s.tantilov@intel.com>
Tested-by: NStephen Ko <stephen.s.ko@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This patch adds support for new device ID.
Signed-off-by: NEmil Tantilov <emil.s.tantilov@intel.com>
Tested-by: NPhil Schmitt <phillip.j.schmitt@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This patch fixes an issue with storing the driver version for the
firmware. If the os does not support the particular firmware
management tools, the firmware requires a driver version to be written
as 0xFFFFFFFF rather than the actual driver version.
Signed-off-by: NJacob Keller <jacob.e.keller@intel.com>
Tested-by: NStephen Ko <stephen.s.ko@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Since ixgbe_raise_i2c_clk() can never return anything else than 0
this patch removes it's return value and all checks for it.
Signed-off-by: NEmil Tantilov <emil.s.tantilov@intel.com>
Tested-by: NPhil Schmitt <phillip.j.schmitt@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Clear the data field in ixgbe_read_i2c_byte_generic so it does not
accumulate 1 bit using the same variable multiple times.
Signed-off-by: NEmil Tantilov <emil.s.tantilov@intel.com>
Tested-by: NPhil Schmitt <phillip.j.schmitt@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

It some situations the driver sets __IXGBE_RESETTING and then
__IXGBE_DOWN flags. It is possible a link check may sneak in
between.

This patch adds check for both flags.
The idea is to reduce register reads while the PHY is resetting.
Signed-off-by: NEmil Tantilov <emil.s.tantilov@intel.com>
Tested-by: NPhil Schmitt <phillip.j.schmitt@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

The initial function and setup tables can be marked as constant.
Reported-by: NStephen Hemminger <shemminger@vyatta.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Tested-by: NAaron Brown <aaron.f.brown@intel.com>

Noticed that the legacy Interrupt handler didn't have the same
ECC warning as did the MSI.  So this patch adds it.

Signed-off-by: Don Skidmore <donald.c.skidmore>
Tested-by: NPhil Schmitt <phillip.j.schmitt@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

The X540 thermal sensor interrupt isn't a General Purpose Interrupt
so doesn't need to be enabled in ixgbe_setup_gpie().  Likewise X540 doesn't
use the SDP0 for thermal sensor so it doesn't need to be enabled for any
device other than 82599.
Signed-off-by: NDon Skidmore <donald.c.skidmore@intel.com>
Tested-by: NPhil Schmitt <phillip.j.schmitt@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Add code to enable thermal sensors for the x540 hardware, as well as a
thermal interrupt check which will exit with a critical message of a
thermal overheat is detected. Intent of code allows other mac types to
be added with different configuration in the future.

Fixed in this version is the addition of setting the temp_sensor
capable flag which was previously only set for a specific mac.
Signed-off-by: NJacob Keller <jacob.e.keller@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Revise high and low threshold marks wrt flow control to account
for the X540 devices and latency introduced by the loopback
switch.

Without this it was in theory possible to drop frames on a
supposedly lossless link with X540 or SR-IOV enabled.

Previously we used a magic number in a define to calculate the
threshold values. This made it difficult to sort out exactly
which latencies were or were not being accounted for. Here
I was overly explicit and tried to used #define names that would
be recognizable after reading the IEEE 802.1Qbb specification.
Signed-off-by: NJohn Fastabend <john.r.fastabend@intel.com>
Tested-by: NRoss Brattain <ross.b.brattain@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>