The Tx accounting when cleaning during NAPI was not completely properly.
We should use the work_limit to determine when to finish cleaning, and
use the same to return the cleaned status.  The impact of running like this
causes the NAPI clean for this Tx to get stuck in a scheduling loop, and
can result in Tx not getting cleaned, ending with a Tx hang and device
reset.
Signed-off-by: NJesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: NPeter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Occasionally if the driver was loaded in a system that
didn't support MSI-X or MSI and was on a shared interrupt,
the driver would then panic in NAPI on the first shared
interrupt because we hadn't called napi_add yet.

Solution: call napi_add before calling request_irq
Signed-off-by: NJesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: NPeter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The interrupt models using EITR have changed in 82599.  The way the register
is laid out, the change is transparent to some of the existing code.
However, some of it isn't.  This patch fixes all the cases where EITR
handling is different than 82598.
Signed-off-by: NJesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: NPeter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

82599 mistakenly enabled drop on Rx queues in the packet buffer.  The
default mode should be store-and-forward from the FIFO.
Signed-off-by: NPeter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Acked-by: NMallikarjuna R Chilakala <mallikarjuna.chilakala@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The rx_no_dma_resources counter reported by ethtool -S ethX is not
counting correctly.  In 82599, the queue mappings for the counters need
to be mapped properly, and accounted for properly.
Signed-off-by: NPeter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Acked-by: NMallikarjuna R Chilakala <mallikarjuna.chilakala@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

With the hardware-specific code in place, add all supported device id's,
along with base driver changes to enable 82599 devices.  The devices
being enabled are:

8086:10f7: 82599EB 10 Gigabit KX4 Network Connection
8086:10fb: 82599EB 10 Gigabit Network Connection

The device 8086:10fb is a fully-pluggable SFP+ NIC.
Signed-off-by: NPeter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

it was pointed out on the list that ixgbe was failing when using 64kB pages
and large 16kB MTU.

since with a 64kB PAGE_SIZE MAX_SKB_FRAGS = 3, the way the driver was
configuring page usage was assuming 2kB is half a page, and was only
ever dmaing that much data to a half page.

(16kB - header size) / 2048 = 7 or 8 pages, which would far exceed 3

adjust the driver to account for these large pages, the hardware can
support DMA to up to 16kB for each descriptor.
Signed-off-by: NJesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The failure happens when an interrupt occurs and the driver is reading
EICR.  This read will cause a clear-by-read which leads to two TLP
being inserted in the PCIe retry buffer leading to an overflow of the
buffer and corruption of TLPs.

The solution is different depending where the reading of EICR takes place.

For ixgbe_msix_lsc() since we are in MSIX mode and know OCD is enabled a
clear-by-write is done instead of the normal clear-by-read.

For ixgbe_intr() 0xffffffff is written to EIMC before the read, masking the
interrupts.
Signed-off-by: NDon Skidmore <donald.c.skidmore@intel.com>
Acked-by: NPeter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Removing the unused PAGE_USE_COUNT() macro. Also removing the unused
pages variable at ixgbe_configure_rx() function.
Signed-off-by: NBreno Leitão <leitao@linux.vnet.ibm.com>
Signed-off-by: NJay Vosburgh <fubar@us.ibm.com>
Acked-by: NPeter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Now that our set_num_queues() routines for each feature are re-entrant, and
can be called at any point, they shouldn't zero out the feature's indices
or mask bits.  Subsequent calls into those routines for those features can
result in zero Rx and Tx queues being assigned, causing a panic later in
driver reinitialization.
Signed-off-by: NPeter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Tx Head writeback is causing multi-microsecond stalls on PCIe chipsets, due
to partial cacheline writebacks.  Removing this feature removes these
issues.
Signed-off-by: NPeter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The flow control handling is overly complicated and difficult to maintain.
This patch cleans up the flow control handling and makes it much more
explicit.  It also adds 1G flow control autonegotiation, for 1G copper
links, 1G KX links, and 1G fiber links.
Signed-off-by: NPeter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently ixgbe does not display the EEPROM version in ethtool -i, where
other drivers do.  The EEPROM version is located at offset 0x29.  This
patch adds support to display it.
Signed-off-by: NPeter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The current code has some flaws in it when performing autonegotiation,
especially on KX/KX4 links.  This patch updates the code to better handle
the autonegotiation states on link setup.  The patch also removes a redundant
link configuration call on driver load, and moves link configuration to
the ->open() path.
Signed-off-by: NPeter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The current code to determine the number of queues the device will want
on driver initialization is ugly and difficult to maintain.  It also
doesn't allow for easy expansion for future features or future hardware.
This patch refactors these routines, and make them easier to deal with.
Signed-off-by: NPeter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

New year, new copyright date ranges.  Also bump the driver version
number to reflect many of the recent changes.
Signed-off-by: NPeter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Our current MSI-X allocation mechanism does not support new hardware
at all.  It also isn't getting the actual number of supported MSI-X vectors
from the device.

This patch allows the number of MSI-X vectors to be specific to a device,
plus it gets the number of MSI-X vectors available from PCIe configuration
space.
Signed-off-by: NPeter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add the device ID for BX devices using the 82598 MAC.
Signed-off-by: NDon Skidmore <donald.c.skidmore@intel.com>
Signed-off-by: NPeter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Base versions handle constant folding now.
Signed-off-by: NHarvey Harrison <harvey.harrison@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NDavid S. Miller <davem@davemloft.net>

And support for the KX/KX4 mezzanine card.  Device id 0x10B6.
Signed-off-by: NDon Skidmore <donald.c.skidmore@intel.com>
Signed-off-by: NPeter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Load times for NICs that use i2c to communicate with the phy were taking
~4.5 sec per port.  This fix first checks to see if the link is already
up before calling get_link_capabilities, since if it is we don't need
query the phy for link state.
Signed-off-by: NDon Skidmore <donald.c.skidmore@intel.com>
Signed-off-by: NPeter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch makes ixgbe invoke the GRO hooks instead of LRO.  As
GRO has a compatible external interface to LRO this is a very
straightforward replacement.

As GRO uses the napi structure to track the held packets, I've
modified the code paths involved to pass that along.
Signed-off-by: NHerbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Following the removal of the unused struct net_device * parameter from
the NAPI functions named *netif_rx_* in commit 908a7a16, they are
exactly equivalent to the corresponding *napi_* functions and are
therefore redundant.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>
Acked-by: NNeil Horman <nhorman@tuxdriver.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

netif_carrier_off() is sufficient to stop Tx into the driver.  Stopping the Tx
queues is redundant and unnecessary.  By the same token, netif_carrier_on()
will be sufficient to re-enable Tx, so waking the queues is unnecessary.
Signed-off-by: NPeter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Register VLAN ID 0 so that frames with VLAN ID 0 are received and get
their tag stripped when ixgbe is not in DCB mode.  VLAN ID 0 means
that the frame is 'priority tagged' only - it is not a VLAN, but the
priority value is the tag in valid.  The functions
ixgbe_vlan_rx_register() and ixgbe_vlan_rx_kill_vid() were moved up a
couple functions to correct compiling issues with this change.
Signed-off-by: NDon Skidmore <donald.c.skidmore@intel.com>
Signed-off-by: NPeter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: NEric W Multanen <eric.w.multanen@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The is an issue where setting Relaxed Ordering (RO) bit
(in a PCI-E write transaction) on 82598 causing the chipset
to drop DCA hints.  This patch forces RO not to be set for
descriptors as well as payload.  This will only be in effect
while DCA is enabled and no performance difference was
noticed in testing.
Signed-off-by: NDon Skidmore <donald.c.skidmore@intel.com>
Signed-off-by: NPeter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This change allows DCB mode to change the number of queues, and presumably
the number of NAPI instances, safely.
Signed-off-by: NPeter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When the napi api was changed to separate its 1:1 binding to the net_device
struct, the netif_rx_[prep|schedule|complete] api failed to remove the now
vestigual net_device structure parameter.  This patch cleans up that api by
properly removing it..
Signed-off-by: NNeil Horman <nhorman@tuxdriver.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This issue was initially reported by Joerg Roedel <joerg.roedel@amd.com>
It appears that ixgbe has had a long standing bug where it was unmapping a different size than it had mapped.

ixgbe 0000:02:00.0: PCI-DMA: device driver frees DMA memory with different sizes than it mapped.
ixgbe 0000:02:00.0: PCI-DMA: device driver frees DMA memory with different size [device address=0x0000000003fed812] [map size=258 bytes] [unmap size=256 bytes]
Pid: 6178, comm: rmmod Not tainted 2.6.28-rc5 #4 Call Trace:
 [<ffffffff8022a2ae>] iommu_queue_inv_iommu_pages+0x5e/0x70
 [<ffffffff80225956>] check_unmap+0x1c6/0x240  [<ffffffff80225ff5>] debug_unmap_single+0xb5/0x110  [<ffffffffa0213997>] ixgbe_clean_rx_ring+0x147/0x220  [<ffffffffa0214d7d>] ixgbe_down+0x2fd/0x3d0 [ixgbe]  [<ffffffffa02150b3>] ixgbe_close+0x13/0xc0 [ixgbe]  [<ffffffff80431326>] dev_close+0x56/0xa0  [<ffffffff804313b3>] rollback_registered+0x43/0x220  [<ffffffff804315a5>] unregister_netdevice+0x15/0x60  [<ffffffff80431601>] unregister_netdev+0x11/0x20  [<ffffffffa021aef8>] ixgbe_remove+0x48/0x16e [ixgbe]  [<ffffffff80386ffc>] pci_device_remove+0x2c/0x60  [<ffffffff803ef929>] __device_release_driver+0x99/0x100
 [<ffffffff803efa48>] driver_detach+0xb8/0xc0  [<ffffffff803eea6e>] bus_remove_driver+0x8e/0xd0  [<ffffffff80387374>] pci_unregister_driver+0x34/0x90  [<ffffffff8026c6c7>] sys_delete_module+0x1c7/0x2a0  [<ffffffff802a9ce9>] do_munmap+0x349/0x390  [<ffffffff80374481>] __up_write+0x21/0x150  [<ffffffff8020c30b>] system_call_fastpath+0x16/0x1b
Signed-off-by: NJesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch implements the PCIe Advanced Error Reporting callbacks in
ixgbe.  The 82598 hardware supports AER, so we enable it.
Signed-off-by: NPeter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Seems the ixgbe's code was copied from e1000.
The comment talks about something not exist.
Signed-off-by: NWang Chen <wangchen@cn.fujitsu.com>
Acked-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

ixgbe_init_interrupt_scheme() is called from ixgbe_resume().  Build that
with CONFIG_PM and without CONFIG_HOTPLUG and you've got a problem.
Several helpers called by it also are misannotated __devinit.
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

ixgbe_init_interrupt_scheme() is called from ixgbe_resume().  Build that
with CONFIG_PM and without CONFIG_HOTPLUG and you've got a problem.
Several helpers called by it also are misannotated __devinit.
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The system log messages created on a link status change need to follow a
specific format to work with tools some customers use.  This also makes
the messages consistant with other Intel driver link messages.
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NRobert Olsson <robert.olsson@its.uu.se>
Acked-by: NPeter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Since the netlink option for DCB is necessary to actually be useful,
simplified the Kconfig option.  In addition, added useful help text for the
Kconfig option.
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

remove redundant argument comments in files of drivers/net/*
Signed-off-by: NQinghuang Feng <qhfeng.kernel@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch adds support for SFP+ PHY in the following device ID's (10DB,
10F1, 10E1).  These SFP+ PHY's are accessed via an I2C interface so the
patch also includes functions to support this.

Another feature of note is that the PHY is pluggable and some
rearchitecting was needed to support this.
Signed-off-by: NDonald Skidmore <donald.c.skidmore@intel.com>
Signed-off-by: NPeter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This adds support for Data Center Bridging (DCB) features in the ixgbe
driver and adds an rtnetlink interface for configuring DCB to the
kernel.  The DCB feature support included are Priority Grouping (PG) -
which allows bandwidth guarantees to be allocated to groups to traffic
based on the 802.1q priority, and Priority Based Flow Control (PFC) -
which introduces a new MAC control PAUSE frame which works at
granularity of the 802.1p priority instead of the link (IEEE 802.3x).
Signed-off-by: NAlexander Duyck <alexander.h.duyck@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: NPeter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>