Processing any incoming packets with a with a napi budget of 0
is incorrect driver behavior.

This matters as netpoll will shortly call drivers with a budget of 0
to avoid receive packet processing happening in hard irq context.
Signed-off-by: N"Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Processing any incoming packets with a with a napi budget of 0
is incorrect driver behavior.

This matters as netpoll will shortly call drivers with a budget of 0
to avoid receive packet processing happening in hard irq context.
Signed-off-by: N"Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Processing any incoming packets with a with a napi budget of 0
is incorrect driver behavior.

This matters as netpoll will shortly call drivers with a budget of 0
to avoid receive packet processing happening in hard irq context.
Signed-off-by: N"Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Processing any incoming packets with a with a napi budget of 0
is incorrect driver behavior.

This matters as netpoll will shortly call drivers with a budget of 0
to avoid receive packet processing happening in hard irq context.
Signed-off-by: N"Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Hariprasad Shenai says:

====================
Doorbell drop Avoidance Bug fix for iw_cxgb4

This patch series provides fixes for Chelsio T4/T5 adapters
related to DB Drop avoidance and other small fix related to keepalive on
iw-cxgb4.

The patches series is created against David Miller's 'net-next' tree.
And includes patches on cxgb4 and iw_cxgb4 driver.

We would like to request this patch series to get merged via David Miller's
'net-next' tree.

We have included all the maintainers of respective drivers. Kindly review the
change and let us know in case of any review comments.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The current logic suffers from a slow response time to disable user DB
usage, and also fails to avoid DB FIFO drops under heavy load. This commit
fixes these deficiencies and makes the avoidance logic more optimal.
This is done by more efficiently notifying the ULDs of potential DB
problems, and implements a smoother flow control algorithm in iw_cxgb4,
which is the ULD that puts the most load on the DB fifo.

Design:

cxgb4:

Direct ULD callback from the DB FULL/DROP interrupt handler.  This allows
the ULD to stop doing user DB writes as quickly as possible.

While user DB usage is disabled, the LLD will accumulate DB write events
for its queues.  Then once DB usage is reenabled, a single DB write is
done for each queue with its accumulated write count.  This reduces the
load put on the DB fifo when reenabling.

iw_cxgb4:

Instead of marking each qp to indicate DB writes are disabled, we create
a device-global status page that each user process maps.  This allows
iw_cxgb4 to only set this single bit to disable all DB writes for all
user QPs vs traversing the idr of all the active QPs.  If the libcxgb4
doesn't support this, then we fall back to the old approach of marking
each QP.  Thus we allow the new driver to work with an older libcxgb4.

When the LLD upcalls iw_cxgb4 indicating DB FULL, we disable all DB writes
via the status page and transition the DB state to STOPPED.  As user
processes see that DB writes are disabled, they call into iw_cxgb4
to submit their DB write events.  Since the DB state is in STOPPED,
the QP trying to write gets enqueued on a new DB "flow control" list.
As subsequent DB writes are submitted for this flow controlled QP, the
amount of writes are accumulated for each QP on the flow control list.
So all the user QPs that are actively ringing the DB get put on this
list and the number of writes they request are accumulated.

When the LLD upcalls iw_cxgb4 indicating DB EMPTY, which is in a workq
context, we change the DB state to FLOW_CONTROL, and begin resuming all
the QPs that are on the flow control list.  This logic runs on until
the flow control list is empty or we exit FLOW_CONTROL mode (due to
a DB DROP upcall, for example).  QPs are removed from this list, and
their accumulated DB write counts written to the DB FIFO.  Sets of QPs,
called chunks in the code, are removed at one time. The chunk size is 64.
So 64 QPs are resumed at a time, and before the next chunk is resumed, the
logic waits (blocks) for the DB FIFO to drain.  This prevents resuming to
quickly and overflowing the FIFO.  Once the flow control list is empty,
the db state transitions back to NORMAL and user QPs are again allowed
to write directly to the user DB register.

The algorithm is designed such that if the DB write load is high enough,
then all the DB writes get submitted by the kernel using this flow
controlled approach to avoid DB drops.  As the load lightens though, we
resume to normal DB writes directly by user applications.
Signed-off-by: NSteve Wise <swise@opengridcomputing.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Based on original work by Anand Priyadarshee <anandp@chelsio.com>.
Signed-off-by: NSteve Wise <swise@opengridcomputing.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Replace the bh safe variant with the hard irq safe variant.

We need a hard irq safe variant to deal with netpoll transmitting
packets from hard irq context, and we need it in most if not all of
the places using the bh safe variant.

Except on 32bit uni-processor the code is exactly the same so don't
bother with a bh variant, just have a hard irq safe variant that
everyone can use.
Signed-off-by: N"Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Conflicts:
	drivers/net/usb/r8152.c
	drivers/net/xen-netback/netback.c

Both the r8152 and netback conflicts were simple overlapping
changes.
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Veaceslav Falico says:

====================
bonding: use correct ether type for alb

There have been reports that, while using the ETH_P_LOOP ether type
(0x0060), the ether type is treated as its packet length.

To avoid that and to not break already existing apps - add new ether type
ETH_P_LOOPBACK that contains the correct id - 0x9000.
====================
Signed-off-by: NVeaceslav Falico <vfalico@redhat.com>

Currently it's using the wrong ETH_P_LOOP type, which is sometimes treated
as packet length instead of ether type (because it's 0x0060).

Use the new ETH_P_LOOPBACK type.

CC: Jay Vosburgh <fubar@us.ibm.com>
CC: Andy Gospodarek <andy@greyhouse.net>
Signed-off-by: NVeaceslav Falico <vfalico@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Per IEEE 802.3*, the correct packet type for loopback 0x9000. There's
already one ETH_P_LOOP 0x0060, which has been there for ages, however it's
plainly wrong as anything that small is considered a length field.

We can't remove it because legacy, so add a new type which corresponds to
the correct id.

http://www.iana.org/assignments/ieee-802-numbers/ieee-802-numbers.xhtml

CC: "David S. Miller" <davem@davemloft.net>
CC: Stefan Richter <stefanr@s5r6.in-berlin.de>
CC: Simon Wunderlich <sw@simonwunderlich.de>
CC: Neil Jerram <Neil.Jerram@metaswitch.com>
CC: Simon Horman <horms@verge.net.au>
CC: Arvid Brodin <Arvid.Brodin@xdin.com>
Signed-off-by: NVeaceslav Falico <vfalico@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Jeff Kirsher says:

====================
Intel Wired LAN Driver Updates

This series contains updates to igb, i40e and i40evf.

I provide a code comment fix which David Miller noticed in the last
series of patches I submitted.

Shannon provides a patch to cleanup the NAPI structs when deleting the
netdev.

Anjali provides several patches for i40e, first fixes a bug in the update
filter logic which was causing a kernel panic.  Then provides a fix to
rename an error bit to correctly indicate the error.  Adds a definition
for a new state variable to keep track of features automatically disabled
due to hardware resource limitations versus user enforced feature disabled.
Anjali provides a patch to add code to handle when there is a filter
programming error due to a full table, which also resolves a previous
compile warning about an unused "*pf" variable introduced in the last i40e
series patch submission.

Jesse provides three i40e patches to cleanup strings to make more
consistent and to align with other Intel drivers.

Akeem cleans up a misleading function header comment for i40e.

Mitch provides a fix for i40e/i40evf to use the correctly reported number
of MSI-X vectors in the PF an VF.  Then provides a patch to use
dma_set_mask_and_coherent() which was introduced in v3.13 and simplifies
the DMA mapping code a bit.

v2:
- dropped the 2 ixgbe patches from Emil based on feedback from David Miller,
  where the 2 fixes should be handled in the net core to fix all drivers
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Phoebe Buckheister says:

====================
ieee802154: fix endianness and header handling

This patch set enforces network byte order on all internal operations and
fields of the 802.15.4 stack and adds a general representation of 802.15.4
headers with operations to create and parse those headers. This reduces code
duplication in the current stack and also allows for upper layers to read
headers of packets they have just received; it is also necessary for 802.15.4
link layer security, which requires header mangling.

Changes since v1:
 * fixed lowpan packet rx after reassembly. Control blocks were used to
   retrieve source/dest addresses, but the CB is clobbered by reassembly.
   Instead, parse the header anew in lowpan.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Have mac802154 header_ops.create fail with -EMSGSIZE if the length
passed will be too large to fit a frame. Since 6lowpan will ensure that
no packet payload will be too large, pass a length of 0 there. 802.15.4
dgram sockets will also return -EMSGSIZE on payloads larger than the
device MTU instead of -EINVAL.
Signed-off-by: NPhoebe Buckheister <phoebe.buckheister@itwm.fraunhofer.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Fragmentation and reassembly information for 6lowpan is independent from
the 802.15.4 stack and used only by the 6lowpan reassembly process. Move
the ieee802154_frag_info struct to a private are, it needn't be in the
802.15.4 skb control block.
Signed-off-by: NPhoebe Buckheister <phoebe.buckheister@itwm.fraunhofer.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Change all internal uses of ieee802154_addr_sa to ieee802154_addr,
except for those instances that communicate directly with userspace.
Signed-off-by: NPhoebe Buckheister <phoebe.buckheister@itwm.fraunhofer.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Use the operations on 802.15.4 header structs introduced in a previous
patch to create and parse all headers in the mac802154 stack. This patch
reduces code duplication between different parts of the mac802154 stack
that needed information from headers, and also fixes a few bugs that
seem to have gone unnoticed until now:

 * 802.15.4 dgram sockets would return a slightly incorrect value for
   the SIOCINQ ioctl
 * mac802154 would not drop frames with the "security enabled" bit set,
   even though it does not support security, in violation of the
   standard
Signed-off-by: NPhoebe Buckheister <phoebe.buckheister@itwm.fraunhofer.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch provides a set of structures to represent 802.15.4 MAC
headers, and a set of operations to push/pull/peek these structs from
skbs. We cannot simply pointer-cast the skb MAC header pointer to these
structs, because 802.15.4 headers are wildly variable - depending on the
first three bytes, virtually all other fields of the header may be
present or not, and be present with different lengths.

The new header creation/parsing routines also support 802.15.4 security
headers, which are currently not supported by the mac802154
implementation of the protocol.
Signed-off-by: NPhoebe Buckheister <phoebe.buckheister@itwm.fraunhofer.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Enable sparse warnings about endianness, replace the remaining fields
regarding network operations without explicit endianness annotations
with such that are annotated, and propagate this through the entire
stack.

Uses of ieee802154_addr_sa are not changed yet, this patch is only
concerned with all other fields (such as address filters, operation
parameters and the likes).
Signed-off-by: NPhoebe Buckheister <phoebe.buckheister@itwm.fraunhofer.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Add a replacement ieee802154_addr struct with proper endianness on
fields. Short address fields are stored as __le16 as on the network,
extended (EUI64) addresses are __le64 as opposed to the u8[8] format
used previously. This disconnect with the netdev address, which is
stored as big-endian u8[8], is intentional.
Signed-off-by: NPhoebe Buckheister <phoebe.buckheister@itwm.fraunhofer.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The struct as currently defined uses host byte order for some fields,
and most big endian/EUI display byte order for other fields. Inside the
stack, endianness should ideally match network byte order where possible
to minimize the number of byteswaps done in critical paths, but this
patch does not address this; it is only preparatory.
Signed-off-by: NPhoebe Buckheister <phoebe.buckheister@itwm.fraunhofer.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Pull x86 fixes from Peter Anvin:
 "Two x86 fixes: Suresh's eager FPU fix, and a fix to the NUMA quirk for
  AMD northbridges.

  This only includes Suresh's fix patch, not the "mostly a cleanup"
  patch which had __init issues"

* 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/amd/numa: Fix northbridge quirk to assign correct NUMA node
  x86, fpu: Check tsk_used_math() in kernel_fpu_end() for eager FPU

Pull ACPI and power management fixes from Rafael Wysocki:
 "Three of these are regression fixes, for two recent regressions and
  one introduced during the 3.13 cycle, and the fourth one is a working
  version of the fix that had to be reverted last time.

  Specifics:

   - A recent ACPI resources handling fix overlooked the fact that it
     had to update the ACPI PNP subsystem's resources parsing too and
     caused confusing warning messages to be printed during system
     intialization on some systems (with arguably buggy ACPI tables).
     Fix from Zhang Rui.

   - Moving the early ACPI initialization before timekeeping_init()
     earlier in this cycle broke fast TSC calibration on at least one
     system, so it needs to be done later, but still before
     efi_enter_virtual_mode() to allow the EFI initialization to refer
     to ACPI.

   - A change related to code duplication reduction in the cpufreq core
     inadvertently caused cpufreq intialization to fail for some CPUs
     handled by intel_pstate by adding checks that may fail for that
     driver, but aren't even necessary when it is used.  The issue is
     addressed by preventing those checks from run in the configurations
     in which they aren't needed.

   - If the Hardware Reduced ACPI flag is set in the ACPI tables, system
     suspend, hibernation and ACPI power off will only work when special
     sleep control and sleep status registeres are provided (their
     addresses in the ACPI tables are not zero).  If those registers are
     not available, the features in question have no chances to work, so
     they shouldn't even be regarded as supported.  That helps with
     power off in particular, because alternative power off methods may
     be used then and they may actually work"

* tag 'pm+acpi-3.14-rc7' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
  ACPI / sleep: Add extra checks for HW Reduced ACPI mode sleep states
  ACPI / init: Invoke early ACPI initialization later
  cpufreq: Skip current frequency initialization for ->setpolicy drivers
  PNP / ACPI: proper handling of ACPI IO/Memory resource parsing failures

Pull device-mapper fixes form Mike Snitzer:
 "Two small fixes for the DM cache target:

   - fix corruption with >2TB fast device due to truncation bug
   - fix access beyond end of origin device due to a partial block"

* tag 'dm-3.14-fixes-4' of git://git.kernel.org/pub/scm/linux/kernel/git/device-mapper/linux-dm:
  dm cache: fix access beyond end of origin device
  dm cache: fix truncation bug when copying a block to/from >2TB fast device

In Linux 3.13, dma_set_mask_and_coherent was introduced, and we have
been encouraged to use it. It simplifies the DMA mapping code a bit as
well.

Change-ID: I66e340245af7d0dedfa8b40fec1f5e352754432e
Signed-off-by: NMitch Williams <mitch.a.williams@intel.com>
Signed-off-by: NCatherine Sullivan  <catherine.sullivan@intel.com>
Tested-by: NKavindya Deegala <kavindya.s.deegala@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Now that the 2.4 firmware reports the correct number of MSI-X vectors,
use this value correctly when communicating with the VF, and when
setting up the interrupt linked list.

The PF has always reported the correct number of MSI-X vectors, so we
should never increment the value in the vf driver.

Change-ID: Ifeefc631c321390192219ce2af9ada6180c1492f
Signed-off-by: NMitch Williams <mitch.a.williams@intel.com>
Signed-off-by: NCatherine Sullivan <catherine.sullivan@intel.com>
Tested-by: NSibai Li <sibai.li@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

We have a separate handler for MDD events, a generic reset is not required.

Change-ID: I77858e2d479e4e65c52aede67109464649ea0253
Signed-off-by: NAnjali Singhai Jain <anjali.singhai@intel.com>
Tested-by: NKavindya Deegala <kavindya.s.deegala@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

The FDIR replay logic was being run a little too soon (before the
queues were enabled) and hence the tail bump was not effective till
a later transaction happened on the queue.

Change-ID: Icfd7cd2e79fc3cae3cbd3f703a2b3a148b4e7bf6
Signed-off-by: NAnjali Singhai Jain <anjali.singhai@intel.com>
Signed-off-by: NCatherine Sullivan <catherine.sullivan@intel.com>
Tested-by: NKavindya Deegala <kavindya.s.deegala@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Add code to enforce the following policy:
- If the HW reports filter programming error, we check if it's due to a
  full table.
- If so, we go ahead and turn off new rule addition for ATR and then SB
  in that order.
- We monitor the programmed filter count, if enough room is created due
  to filter deletion/reset, we then re-enable SB and ATR new rule addition.

Change-ID: I69d24b29e5c45bc4fa861258e11c2fa7b8868748
Signed-off-by: NAnjali Singhai Jain <anjali.singhai@intel.com>
Signed-off-by: NCatherine Sullivan <catherine.sullivan@intel.com>
Tested-by: NKavindya Deegala <kavindya.s.deegala@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This variable is a bit mask. It is needed to differentiate between
user enforced feature disables and auto disable of features due to
HW resource limitations.

Change-ID: Ib4b4f6ae1bb2668c12e482d2555100bc8ad713d5
Signed-off-by: NAnjali Singhai Jain <anjali.singhai@intel.com>
Signed-off-by: NCatherine Sullivan <catherine.sullivan@intel.com>
Tested-by: NKavindya Deegala <kavindya.s.deegala@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Correct misleading function comment.

Change-ID: I3f66cff5cc00250a285756b6500a58fad8eba4b5
Signed-off-by: NAkeem G Abodunrin <akeem.g.abodunrin@intel.com>
Signed-off-by: NCatherine Sullivan <catherine.sullivan@intel.com>
Tested-by: NKavindya Deegala <kavindya.s.deegala@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

In a similar way to how ixgbe works, print a short one-line string
showing what features and number of queues the driver and hardware has
enabled at probe time.

Example (wrapped for the commit message):
i40e 0000:06:00.1: Features: PF-id[1] VFs: 64 VSIs: 66 QP: 32 FDir RSS
ATR NTUPLE DCB

Change-ID: I177bf7f93d1c4c921529c92fdf66e614f6b4f755
Signed-off-by: NJesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: NCatherine Sullivan <catherine.sullivan@intel.com>
Tested-by: NKavindya Deegala <kavindya.s.deegala@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This patch cleans up the strings that the driver prints during normal
operation and moves many strings into dev_dbg.  It also cleans up
strings printed during reset.

Change-ID: I1835cc4e3c3b22596182b683284e6bb87eac61b2
Signed-off-by: NJesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: NCatherine Sullivan <catherine.sullivan@intel.com>
Tested-by: NKavindya Deegala <kavindya.s.deegala@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

This cleans up strings for consistency, q is replaced with queue.

Change-ID: Ia5f9dfae9af261f4c24485854264e02363729cf3
Signed-off-by: NJesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: NCatherine Sullivan <catherine.sullivan@intel.com>
Tested-by: NKavindya Deegala <kavindya.s.deegala@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

1) Fix a name of the error bit to correctly indicate the error.
2) Added a fd_id field in the 32 byte desc at the place(qw0) where it gets
reported in the programming error desc WB. In a normal data desc
the fd_id field is reported in qw3.

Change-ID: Ide9a24bff7273da5889c36635d629bc3b5212010
Signed-off-by: NAnjali Singhai Jain <anjali.singhai@intel.com>
Acked-by: NShannon Nelson <shannon.nelson@intel.com>
Signed-off-by: NKevin Scott <kevin.c.scott@intel.com>
Signed-off-by: NCatherine Sullivan <catherine.sullivan@intel.com>
Tested-by: NKavindya Deegala <kavindya.s.deegala@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

The update filter logic was causing a kernel panic in the original code.
We need to compare the input set to decide whether or not to delete a
filter since we do not have a hash stored. This new design helps fix the issue.

Change-ID: I2462b108e58ca4833312804cda730b4660cc18c9
Signed-off-by: NAnjali Singhai Jain <anjali.singhai@intel.com>
Signed-off-by: NCatherine Sullivan <catherine.sullivan@intel.com>
Tested-by: NKavindya Deegala <kavindya.s.deegala@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

We've been deleting the netdev before getting around to deleting the napi
structs.  Unfortunately, we then didn't delete the napi structs because we
have a check for netdev, thus we were leaving garbage around in the system.

Change-ID: Ife540176f6c9f801147495b3f2d2ac2e61ddcc58
Signed-off-by: NShannon Nelson <shannon.nelson@intel.com>
Signed-off-by: NCatherine Sullivan <catherine.sullivan@intel.com>
Tested-by: NKavindya Deegala <kavindya.s.deegala@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Recently added code comment was missing a space that is needed.
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

For systems with multiple servers and routed fabric, all
northbridges get assigned to the first server. Fix this by also
using the node reported from the PCI bus. For single-fabric
systems, the northbriges are on PCI bus 0 by definition, which
are on NUMA node 0 by definition, so this is invarient on most
systems.

Tested on fam10h and fam15h single and multi-fabric systems and
candidate for stable.
Signed-off-by: NDaniel J Blueman <daniel@numascale.com>
Acked-by: NSteffen Persvold <sp@numascale.com>
Acked-by: NBorislav Petkov <bp@suse.de>
Cc: <stable@vger.kernel.org>
Link: http://lkml.kernel.org/r/1394710981-3596-1-git-send-email-daniel@numascale.comSigned-off-by: NIngo Molnar <mingo@kernel.org>