Split the receive hot path code into two, one for packet split and one
for single buffer. This improves receive performance since we only need
to check if the ring is in packet split mode once per NAPI poll time,
not several times per packet. The single buffer code is further improved
by the removal of a bunch of code and several variables that are not
needed. On a receive-oriented test this can improve single-threaded
throughput.

Also refactor the packet split receive path to use a fixed buffer for
headers, like ixgbe does. This vastly reduces the number of DMA mappings
and unmappings we need to do, allowing for much better performance in
the presence of an IOMMU.

Lastly, correct packet split descriptor types now that we are actually
using them.

Change-ID: I3a194a93af3d2c31e77ff17644ac7376da6f3e4b
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>

Change debugfs command from "clear_stats pf" to "clear_stats port"
to be clearer what the action is.  Also, limit the action to the
base PF, not the NPAR partitions.

Change-ID: I22aa39c0962d83a83a985097b1000ed7f8c66f3f
Signed-off-by: NShannon Nelson <shannon.nelson@intel.com>
Tested-by: NJim Young <james.m.young@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This change is to refactor the read/write functions to support
future work.

Change-ID: I13150d5e3042f2c617362c0140dc7e6473ebcdee
Signed-off-by: NKamil Krawczyk <kamil.krawczyk@intel.com>
Signed-off-by: NShannon Nelson <shannon.nelson@intel.com>
Tested-by: NJim Young <james.m.young@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Fixed a few cases of when we used msleep() when we should have been
using usleep_range().  Also updated the usage of min/max() to use
min_t/max_t().
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Tested-by: NAaron Brown <aaron.f.brown@intel.com>

Fix the code comments to align with the drivers/net/ commenting style.
Also fix other checkpatch errors such as using tabs where possible and
properly wrap lines to conform to the 80 char limit (unless it is
a string).
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
Tested-by: NAaron Brown <aaron.f.brown@intel.com>

i219 is the next-generation LOM that will be available on systems with the
Sunrise Point Platform Controller Hub (PCH) chipset from Intel.  This patch
provides the initial support for the device.
Signed-off-by: NDave Ertman <david.m.ertman@intel.com>
Tested-by: NAaron Brown <aaron.f.brown@intel.com>
Tested-by: NCarmen Edwards <carmenx.edwards@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Every time be_setup() is called, the driver will have to wait for the
function/FW to be properly initialized. So, it make sense to move this call
inside be_setup().
Signed-off-by: NSathya Perla <sathya.perla@emulex.com>
Signed-off-by: NKalesh AP <kalesh.purayil@emulex.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Most of the resource cleanup sequences performed in be_suspend(),
be_eeh_err_detected() and be_err_detection_task() are same. Moved the
common code to a new routine be_cleanup() to avoid code duplication.
Signed-off-by: NKalesh AP <kalesh.purayil@emulex.com>
Signed-off-by: NSathya Perla <sathya.perla@emulex.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Most of the adapter initialisation sequences performed in be_resume(),
be_eeh_resume() and be_err_recover() are same. Renamed be_resume() to
be_pci_resume() and moved the common code to a new routine be_resume() to
avoid code duplication.
Signed-off-by: NKalesh AP <kalesh.purayil@emulex.com>
Signed-off-by: NSathya Perla <sathya.perla@emulex.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The steps needed for Lancer's reset/initialization sequence are:
	a) wait for SLIPORT_STAUS RDY bit to be set
	b) set the SLIPORT_CONTROL IP bit
	c) repeat step "a"

The code needed for this sequence is already covered by the be_func_init()
routine (with minor modifications.) So, get rid of the
lancer_test_and_set_rdy_state() and lancer_provisioning_error() routines
that unnecessarily duplicate this code. Also fixed the error recovery
function to take care of these changes
Signed-off-by: NSathya Perla <sathya.perla@emulex.com>
Signed-off-by: NKalesh AP <kalesh.purayil@emulex.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently when an error is detected, the queue resources are being cleaned up
in the recovery routine. The resources are better cleaned up in the error
detection routine itself (similar to EEH code.)
So, this patch re-factors error processing logic to follow the following sequence:
	- check if there is an error in adapter
	- if error,
		- cleanup resources
		- attempt recovery

The patch renames lancer_recover_func() to be_err_recover() as this
routine will be used in the future for error recovery on Skyhawk too.
Signed-off-by: NKalesh AP <kalesh.purayil@emulex.com>
Signed-off-by: NSathya Perla <sathya.perla@emulex.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Also rename func_recovery_work/task to err_detection_work/task as error
detection is the primary goal of this task while recovery is not guaranteed.
Signed-off-by: NSathya Perla <sathya.perla@emulex.com>
Signed-off-by: NKalesh AP <kalesh.purayil@emulex.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Function initialization sequence is executed in be_probe(), be_resume
and be_eeh_resume(). Move this code to a new routine called be_func_init()
to prevent code duplication.
Signed-off-by: NSathya Perla <sathya.perla@emulex.com>
Signed-off-by: NKalesh AP <kalesh.purayil@emulex.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Most of the code to fetch the adapter state is in be_setup()->be_get_config().
So, move the code from be_get_initial_config() to be_get_config().
Signed-off-by: NSathya Perla <sathya.perla@emulex.com>
Signed-off-by: NKalesh AP <kalesh.purayil@emulex.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The members of be_adapter struct were being allocated in two separate
routines -- be_ctrl_init() and be_stats_init(). Also, some other members
were allocated elsewhere. This patch moves the alloc/free code into
be_drv_init/cleanup() routines. The be_pci_map_bars() routine that was
called from be_ctrl_init() is now called directly from be_probe().
The new routine be_drv_init() will now be the place-holder for allocating
memory for any new be_adapter{} members in the future.
Some routines needed to be moved to provide forward definitions for their
calls.
Signed-off-by: NSathya Perla <sathya.perla@emulex.com>
Signed-off-by: NKalesh AP <kalesh.purayil@emulex.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch updates the BNA driver version to 3.2.25.1 and the firmware version
to 3.2.5.1
Signed-off-by: NRasesh Mody <rasesh.mody@qlogic.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Re-brand the BNA driver to QLogic.
Signed-off-by: NRasesh Mody <rasesh.mody@qlogic.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

An error return from __vnet_tx_trigger() sets the TX descriptor to
VIO_DESC_FREE while leaving port->tx_bufs[txi].skb set. This leads
to a BUG_ON() the next time this descriptor is used.

This patch frees the pending skb when getting a trigger error to
match the VIO_DESC_FREE state.
Signed-off-by: NDavid L Stevens <david.stevens@oracle.com>
Acked-by: NSowmini Varadhan <sowmini.varadhan@oracle.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch converts all __raw_readl and __raw_writel function calls
to their corresponding readl_relaxed and writel_relaxed variants.

It also tells the driver to set ahb_endian_swp_mgmt_en bit in dma_cfg
when the CPU is configured in big endian mode.
Signed-off-by: NArun Chandran <achandran@mvista.com>
Acked-by: NNicolas Ferre <nicolas.ferre@atmel.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch updates BNX2 driver version to 2.2.6 and CNIC driver version
to 2.5.21.
Signed-off-by: NRasesh Mody <rasesh.mody@qlogic.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Do not enable filter SORT MODE in chip init routine. This patch addresses an
issue where BCM5716 sporadically drops packets when changing multicast list.
Signed-off-by: NRasesh Mody <rasesh.mody@qlogic.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch provides additional changes as a part of BNX2 and CNIC driver
re-branding effort.
Signed-off-by: NRasesh Mody <rasesh.mody@qlogic.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

- Check qfull condition
- Check qwake condition
- Check pkts pending completion
Signed-off-by: NSriharsha Basavapatna <sriharsha.basavapatna@emulex.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

- To avoid multiple accesses to CQE, extract compl_status and end_idx from
  be_tx_compl_get().
Signed-off-by: NSriharsha Basavapatna <sriharsha.basavapatna@emulex.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

- Reduce code duplication by moving WRB-frags setup into a function.
- Do not setup WRB-header before frags are setup, which is unncessary if
  there's errors while setting up frags. We should only grab an entry for
  the header, setup the frags and if everything is fine setup the header.
- The error cleanup can be moved into a small function.
Signed-off-by: NSriharsha Basavapatna <sriharsha.basavapatna@emulex.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The WRB header is setup by wrb_fill_hdr() routine. This routine currently
gets some of the WRB params as args and figures out rest of the WRB params
by looking at various fields in skb (like gso, checksum, vlan-tag etc).
All these params could instead be retrieved from the skb into a structure
and passed to this routine. This separates wrb_fill_hdr() to only provide
chip-specific code to fill the WRB. This also makes it simple to support
chips with different WRB formats.
Signed-off-by: NSriharsha Basavapatna <sriharsha.basavapatna@emulex.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This debug output is not really an error message since mac reset can fail
if the phy clocks are gated, specifically when the phy has been placed in
a powered down or isolation mode. The netdev output masks were changed from
err to dbg, and comments added in the code.
Signed-off-by: NVince Bridgers <vbridger@opensource.altera.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Change use of Altera TSE's MDIO access from phy 0 registers to phy 1
registers. This allows support for GMII, MII, RGMII, and SGMII
designs where the external PHY is always accesible through
Altera TSE's MDIO phy 1 registers and Altera's PCS is accessible
through MDIO phy 0 registers for SGMII applications.
Signed-off-by: NVince Bridgers <vbridger@opensource.altera.com>
Tested-by: NKai Lin Ng <kailng@altera.com>
Tested-by: NDalon Westergreen <dwesterg@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Provide a file creation function that also takes an initial size so that the
caller doesn't have to set i_size, thus meaning that we don't have to call
deal with ->d_inode in the callers.
Signed-off-by: NDavid Howells <dhowells@redhat.com>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

The patch "module: fix types of device tables aliases" newly requires that
invocations of

MODULE_DEVICE_TABLE(type, name);

come *after* the definition of `name'.  That is reasonable, but some
drivers weren't doing this.  Fix them.

Cc: James Bottomley <James.Bottomley@HansenPartnership.com>
Cc: Andrey Ryabinin <a.ryabinin@samsung.com>
Cc: David Miller <davem@davemloft.net>
Cc: Hans Verkuil <hverkuil@xs4all.nl>
Acked-by: NMauro Carvalho Chehab <mchehab@osg.samsung.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

printk and friends can now format bitmaps using '%*pb[l]'.  cpumask
and nodemask also provide cpumask_pr_args() and nodemask_pr_args()
respectively which can be used to generate the two printf arguments
necessary to format the specified cpu/nodemask.
Signed-off-by: NTejun Heo <tj@kernel.org>
Cc: Chris Metcalf <cmetcalf@tilera.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

tg3_init_one() calls tg3_halt() without tp->lock despite its assumption
and causes deadlock.
If lockdep is enabled, a warning like this shows up before the stall:

  [ BUG: bad unlock balance detected! ]
  3.19.0test #3 Tainted: G            E
  -------------------------------------
  insmod/369 is trying to release lock (&(&tp->lock)->rlock) at:
  [<ffffffffa02d5a1d>] tg3_chip_reset+0x14d/0x780 [tg3]
  but there are no more locks to release!

tg3_init_one() doesn't call tg3_halt() under normal situation but
during kexec kdump I hit this problem.

Fixes: 932f19de ("tg3: Release tp->lock before invoking synchronize_irq()")
Signed-off-by: NJun'ichi Nomura <j-nomura@ce.jp.nec.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

On Northstar (Broadcom's ARM architecture) we need to manually enable
all cores. Code for that is already in place, but the condition for it
was wrong.
Signed-off-by: NRafał Miłecki <zajec5@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Driver keeps adding multicast addresses without deleting removed MACs and
worrying about adapters filter limit. This results into actual count of programmed
multicast addresses get accumulated over the time and overruns the adapter's
filter limit without putting device in ACCEPT_ALL_MULTI mode. This causes
newly added multicast traffic to fail after the sequence of addition - deletion
in certain pattern.

This issue is seen only when netdev's mcast list count is less than adapters
mcast filter limit.

e.g. If adapters multicast filter limit is 38 per function
     then following sequence would result in multicast traffic failure for
     newly added MACs.
     - add less than 38 multicast MACs
     - remove previously added multicast MACs
     - add new multicast MACs (less than 38)
Signed-off-by: NShahed Shaikh <shahed.shaikh@qlogic.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The current code failed to configure the page size for architectures with page
size different than 4K - PPC for example.
Signed-off-by: NCarol L Soto <clsoto@us.ibm.com>
Signed-off-by: NEli Cohen <eli@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch unclones an skb for the case where the sunvnet driver needs to
change the segmentation size so that it doesn't interfere with TCP SACK's
use of them.
Signed-off-by: NDavid L Stevens <david.stevens@oracle.com>
Acked-by: NEric Dumazet <edumazet@google.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch introduces the use of functions setup_timer
and mod_timer.

This is done using Coccinelle and semantic patch used
for this as follows:

// <smpl>
@@
expression x,y,z,a,b;
@@

-init_timer (&x);
+setup_timer (&x, y, z);
+mod_timer (&a, b);
-x.function = y;
-x.data = z;
-x.expires = b;
-add_timer(&a);

// </smpl>
Signed-off-by: NVaishali Thakkar <vthakkar1994@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

If CONFIG_NET_XGENE=y but CONFIG_OF=n:

drivers/net/ethernet/apm/xgene/xgene_enet_main.c:1033: warning: ‘xgene_enet_of_match’ defined but not used
Signed-off-by: NGeert Uytterhoeven <geert@linux-m68k.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This is only an API consolidation and should make things more readable.
Converting milliseconds to jiffies by "val * HZ / 1000" is technically
OK but msecs_to_jiffies(val) is the cleaner solution and handles all
corner cases correctly. This is a minor API cleanup only.
Signed-off-by: NNicholas Mc Guire <hofrat@osadl.org>
Acked-by: NMark Einon <mark.einon@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Use the normal {} instead of a macro to terminate an array.

Remove the macro too.
Signed-off-by: NJoe Perches <joe@perches.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>