We move register_netdev() to the end of dwceqos_probe() to close any
races where the netdev callbacks are called before the initialization
has finished.
Reported-by: Pavel Andrianov <andrianov@ispras.ru>
Signed-off-by: Lars Persson <larper@axis.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
During port init, we currently set the port's type to Ethernet after
setting its MAC address. However, the hardware documentation states this
should be the other way around.
Align the driver with the hardware documentation and set the port's MAC
address after setting its type.
Fixes: 56ade8fe3f ("mlxsw: spectrum: Add initial support for Spectrum ASIC")
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When neigh_init fails, we have to do proper cleanup including
router_fini call.
Fixes: 6cf3c971dc ("mlxsw: spectrum_router: Add private neigh table")
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Acked-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
There is no need to pad frames to ETH_ZLEN on RX.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Reviewed-by: Dinan Gunawardena <dinan.gunawardena@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Be more strict about FW versions. Drop support for old
transitional revisions which were never used in production.
Dropping support for FW ABI version 0.0.0.0 is particularly
useful because 0 could just be uninitialized memory.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Dinan Gunawardena <dinan.gunawardena@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Remove unnecessary version.h includes.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Simon Horman <simon.horman@netronome.com>
Reviewed-by: Dinan Gunawardena <dinan.gunawardena@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently vlan tagged packets were not parsed correctly
and assumed to be regular IPv4/IPv6 packets.
We should check for 802.1Q/802.1ad tags and update the lro header
accordingly.
This fixes the use case where LRO is on and rxvlan is off
(vlan stripping is off).
Fixes: e586b3b0ba ('net/mlx5: Ethernet Datapath files')
Signed-off-by: Gal Pressman <galp@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently when reading global PFC statistics we left the counter
iterator out of the equation and we ended up reading the same counter
over and over again.
Instead of reading the counter at index 0 on every iteration we now read
the counter at index (i).
Fixes: e989d5a532 ('net/mlx5e: Expose flow control counters to ethtool')
Signed-off-by: Gal Pressman <galp@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
On 64 bits architectures unsigned long is longer than u32,
casting to unsigned long will result in overflow.
We need to first allocate an unsigned long variable, then assign the
wanted value.
Fixes: 665bc53969 ('net/mlx5e: Use new ethtool get/set link ksettings API')
Signed-off-by: Gal Pressman <galp@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Update the xmit_more counter before notifying the HW,
to prevent a possible use-after-free of the skb.
Fixes: c8cf78fe10 ("net/mlx5e: Add ethtool counter for TX xmit_more")
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
There is a code path where we are calling __iowrite64_copy() on
an address that is not 64-bit aligned. This causes an exception on
some architectures such as arm64. Fix that code path by using
__iowrite32_copy().
Reported-by: JD Zheng <jiandong.zheng@broadcom.com>
Signed-off-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The build of m32r was giving warning:
In file included from drivers/net/ethernet/smsc/smc91x.c:92:0:
drivers/net/ethernet/smsc/smc91x.h:448:0: warning: "SMC_inb" redefined
#define SMC_inb(ioaddr, reg) ({ BUG(); 0; })
drivers/net/ethernet/smsc/smc91x.h:106:0:
note: this is the location of the previous definition
#define SMC_inb(a, r) inb(((u32)a) + (r))
drivers/net/ethernet/smsc/smc91x.h:449:0: warning: "SMC_outb" redefined
#define SMC_outb(x, ioaddr, reg) BUG()
drivers/net/ethernet/smsc/smc91x.h:108:0:
note: this is the location of the previous definition
#define SMC_outb(v, a, r) outb(v, ((u32)a) + (r))
Signed-off-by: Sudip Mukherjee <sudip.mukherjee@codethink.co.uk>
Signed-off-by: David S. Miller <davem@davemloft.net>
I'm still struggling to get this fix right..
Changes since v2:
- do not blindly modify SKB contents according to Dave's legitimate
objection
Changes since v1:
- dropped disabling HW checksum offload for Zynq
- initialize checksum similar to net/ethernet/freescale/fec_main.c
-- >8 --
MACB/GEM needs the checksum field initialized to 0 to get correct
results on transmit in all cases, e.g. on Zynq, UDP packets with
payload <= 2 otherwise contain a wrong checksums.
Signed-off-by: Helmut Buchsbaum <helmut.buchsbaum@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The /proc/irq/xx information is incorrect for smsc911x because
the request_irq is happening before the register_netdev has the
proper device name. Moving it to the open also fixes the case
of when the device is renamed.
Reported-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Jeremy Linton <jeremy.linton@arm.com>
Tested-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In preparation for the allocating/enabling interrupts
in the ndo_open routine move the irq handler before it.
Signed-off-by: Jeremy Linton <jeremy.linton@arm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Move phy startup/shutdown into the smsc911x_open/stop routines. This
allows the module to be unloaded because phy_connect_direct is no longer
always holding the module use count. This one change also resolves a
number of other problems.
The link status of a downed interface no longer reflects a stale state.
Errors caused by the net device being opened before the mdio/phy was
configured. There is also a potential power savings as the phy's don't
remain powered when the interface isn't running.
Signed-off-by: Jeremy Linton <jeremy.linton@arm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Rework the error handling in smsc911x open in preparation
for the mdio startup being moved here.
Signed-off-by: Jeremy Linton <jeremy.linton@arm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When PCI error is detected, in some architectures (like PowerPC) a slot
reset is performed - the driver's error handlers are in charge of "disable"
device before the reset, and re-enable it after a successful slot reset.
There are two cases though that another path is taken on the code: if the
slot reset is not successful or if too many errors already happened in the
specific adapter (meaning that possibly the device is experiencing a HW
failure that slot reset is not able to solve), the core PCI error mechanism
(called EEH in PowerPC) will remove the adapter from the system, since it
will consider this as a permanent failure on device. In this case, a path
is taken that leads to bnx2x_chip_cleanup() calling bnx2x_reset_hw(), which
then tries to perform a HW reset on chip. This reset won't succeed since
the HW is in a fault state, which can be seen by multiple messages on
kernel log like below:
bnx2x: [bnx2x_issue_dmae_with_comp:552(eth1)]DMAE timeout!
bnx2x: [bnx2x_write_dmae:600(eth1)]DMAE returned failure -1
After some time, the PCI error mechanism gives up on waiting the driver's
correct removal procedure and forcibly remove the adapter from the system.
We can see soft lockup while core PCI error mechanism is waiting for driver
to accomplish the right removal process.
This patch adds a verification to avoid a chip reset whenever the function
is in PCI error state - since this case is only reached when we have a
device being removed because of a permanent failure, the HW chip reset is
not expected to work fine neither is necessary.
Also, as a minor improvement in error path, we avoid the MCP information dump
in case of non-recoverable PCI error (when adapter is about to be removed),
since it will certainly fail.
Reported-by: Harsha Thyagaraja <hathyaga@in.ibm.com>
Signed-off-by: Guilherme G. Piccoli <gpiccoli@linux.vnet.ibm.com>
Acked-By: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
On ThunderX 88xx pass 2.x chips when TSO is offloaded to HW,
HW posts a CQE for every TSO segment transmitted. Current code
does handles this, but is prone to issues when segment sizes are
small resulting in SW processing too many CQEs and also at times
frees a SKB which is not yet transmitted.
This patch handles the errata in a different way and eliminates issues
with earlier approach, TSO packet is submitted to HW with post_cqe=0,
so that no CQE is posted upon completion of transmission of TSO packet
but a additional HDR + IMMEDIATE descriptors are added to SQ due to
which a CQE is posted and will have required info to be used while
cleanup in napi. This way only one CQE is posted for a TSO packet.
Signed-off-by: Sunil Goutham <sgoutham@cavium.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
There is a issue in HW where-in while sending GSO sized pkts
as part of TSO, if pkt len falls below configured min packet
size i.e 60, NIC will zero PAD packet and also updates IP total length.
Hence set this value to lessthan min pkt size of MAC + IP + TCP
headers, BGX will anyway do the padding to transmit 64 byte pkt
including FCS.
Signed-off-by: Sunil Goutham <sgoutham@cavium.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If DCB is configured on the link partner switch with an
unsupported traffic class configuration (e.g. non-contiguous TCs),
the driver is flagging DCB as disabled. But, for future DCB
LLDPDUs, the driver was checking if the interface was DCB capable
instead of enabled. This was causing a kernel panic when LLDP
was enabled/disabled on the link partner switch.
This patch corrects the situation by having the LLDP event handler
check the correct flag in the pf structure. It also cleans up the
setting and clearing of the enabled flag for other checks.
Signed-off-by: Dave Ertman <david.m.ertman@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The recent commit 087d7a8c91 "tg3: Fix for diasllow rx coalescing
time to be 0" disallow to set Rx coalescing time to be 0 as this stops
generating interrupts for the incoming packets. I found the zero
Tx coalescing time stops generating interrupts for outgoing packets
as well and fires Tx watchdog later. To avoid this, don't allow to set
Tx coalescing time to 0 and also remove subsequent checks that become
senseless.
Cc: satish.baddipadige@broadcom.com
Cc: siva.kallam@broadcom.com
Cc: michael.chan@broadcom.com
Signed-off-by: Ivan Vecera <ivecera@redhat.com>
Acked-by: Siva Reddy Kallam <siva.kallam@broadcom.com>
Acked-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
drivers/net/ethernet/qlogic/qed/qed_dcbx.c:1230:13-20: WARNING: kzalloc should be used for dcbx_info, instead of kmalloc/memset
drivers/net/ethernet/qlogic/qed/qed_dcbx.c:1192:13-20: WARNING: kzalloc should be used for dcbx_info, instead of kmalloc/memset
Use kzalloc rather than kmalloc followed by memset with 0
This considers some simple cases that are common and easy to validate
Note in particular that there are no ...s in the rule, so all of the
matched code has to be contiguous
Generated by: scripts/coccinelle/api/alloc/kzalloc-simple.cocci
CC: Sudarsana Reddy Kalluru <sudarsana.kalluru@qlogic.com>
Signed-off-by: Fengguang Wu <fengguang.wu@intel.com>
Acked-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When a VLAN is added on a bridge port we should use the existing unicast
flood configuration of the port instead of assuming it's enabled.
Fixes: 0293038e0c ("mlxsw: spectrum: Add support for flood control")
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In commit 14d39461b3 ("mlxsw: spectrum: Use per-FID struct for the
VLAN-aware bridge") I added a per-FID struct, which member ports can
take a reference on upon VLAN membership configuration.
However, sometimes only the VLAN flags (e.g. egress untagged) are
toggled without changing the VLAN membership. In these cases we
shouldn't take another reference on the FID.
Fixes: 14d39461b3 ("mlxsw: spectrum: Use per-FID struct for the VLAN-aware bridge")
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently the notifier is registered for every asic instance, however the
same block. Fix this by moving the registration to module init.
Fixes: c723c735fa ("mlxsw: spectrum_router: Periodically update the kernel's neigh table")
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Originally, I expected that there would be needed to call update
operation in case RALUE record action is changed. However, that is not
needed since write operation takes care of that nicely. Remove prepared
construct and always call the write operation.
Fixes: 61c503f976 ("mlxsw: spectrum_router: Implement fib4 add/del switchdev obj ops")
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In mlxsw we squash tables 254 and 255 together into HW. Kernel adds/dels
/32 ip to/from both 254 and 255. On del path, that causes the same
prefix being removed twice. Fix this by introducing reference counting
for private mlxsw fib entries. That required a bit of code reshuffle.
Also put dev into fib entry key so the same prefix could be represented
once per every router interface.
Fixes: 61c503f976 ("mlxsw: spectrum_router: Implement fib4 add/del switchdev obj ops")
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch takes care of clearing the uninitialized buffer before using it.
1. pfc pri-enable bitmap need to be cleared before setting the requested
enable bits. Without this, the un-touched values will be merged with
requested values and sent to MFW.
2. The data in app-entry field need to be cleared before using it.
3. Clear the output data buffer used in qed_dcbx_query_params().
Signed-off-by: Sudarsana Reddy Kalluru <sudarsana.kalluru@qlogic.com>
Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Management firmware requires the selection-field (SF) to be set for
configuring the application/protocol entry in IEEE mode. Without this
setting, the app entry will be configured incorrectly in MFW.
Signed-off-by: Sudarsana Reddy Kalluru <sudarsana.kalluru@qlogic.com>
Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Dcbx configuration is not supported for VF interfaces. Hence don't populate
the callbacks for VFs and also fail the dcbx-query for VFs.
Signed-off-by: Sudarsana Reddy Kalluru <sudarsana.kalluru@qlogic.com>
Signed-off-by: Yuval Mintz <Yuval.Mintz@qlogic.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Return -ENODEV if the MDIO bus is disabled in the device tree.
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Acked-by: John Crispin <john@phrozen.org>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
a lot of parts in the driver uses devm_* APIs to gain benefits from the
device resource management, so devm_mdiobus_alloc is also used instead
of mdiobus_alloc to have more elegant code flow.
Using common code provided by the devm_* helps to
1) have simplified the code flow as [1] says
2) decrease the risk of incorrect error handling by human
3) only a few drivers used it since it was proposed on linux 3.16,
so just hope to promote for this.
Ref:
[1] https://patchwork.ozlabs.org/patch/344093/
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds the missing of_node_put() after finishing the usage
of of_get_child_by_name.
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Acked-by: John Crispin <john@phrozen.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
mtk_stop() must be called to stop for freeing DMA
resources acquired and restoring state changed by mtk_open()
firstly when module removal.
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
original mdio_cleanup is not in the symmetric place against where
mdio_init is, so relocate mdio_cleanup to the right one.
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Acked-by: John Crispin <john@phrozen.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
these irqs are not used for shared irq and disabled during ethernet stops.
irq requested by devm_request_irq is safe to be freed automatically on
driver detach.
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Acked-by: John Crispin <john@phrozen.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
use skb_free_frag() instead of legacy put_page()
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Acked-by: John Crispin <john@phrozen.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
1) If the return value of devm_clk_get is EPROBE_DEFER, we should
defer probing the driver. The change is verified and works based
on 4.8-rc1 staying with the latest clk-next code for MT7623.
2) Changing with the usage of loops to work out if all clocks
required are fine
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
which net device the SKB is complete for depends on the forward port
on txd4 on the corresponding TX descriptor, but the information isn't
set up well in case of SKB fragments that would lead to watchdog timeout
from the upper layer, so fix it up.
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Acked-by: John Crispin <john@phrozen.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit b70661c708 ("net: smc91x: use run-time configuration on all ARM
machines") broke some ARM platforms through several mistakes. Firstly,
the access size must correspond to the following rule:
(a) at least one of 16-bit or 8-bit access size must be supported
(b) 32-bit accesses are optional, and may be enabled in addition to
the above.
Secondly, it provides no emulation of 16-bit accesses, instead blindly
making 16-bit accesses even when the platform specifies that only 8-bit
is supported.
Reorganise smc91x.h so we can make use of the existing 16-bit access
emulation already provided - if 16-bit accesses are supported, use
16-bit accesses directly, otherwise if 8-bit accesses are supported,
use the provided 16-bit access emulation. If neither, BUG(). This
exactly reflects the driver behaviour prior to the commit being fixed.
Since the conversion incorrectly cut down the available access sizes on
several platforms, we also need to go through every platform and fix up
the overly-restrictive access size: Arnd assumed that if a platform can
perform 32-bit, 16-bit and 8-bit accesses, then only a 32-bit access
size needed to be specified - not so, all available access sizes must
be specified.
This likely fixes some performance regressions in doing this: if a
platform does not support 8-bit accesses, 8-bit accesses have been
emulated by performing a 16-bit read-modify-write access.
Tested on the Intel Assabet/Neponset platform, which supports only 8-bit
accesses, which was broken by the original commit.
Fixes: b70661c708 ("net: smc91x: use run-time configuration on all ARM machines")
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>
Tested-by: Robert Jarzmik <robert.jarzmik@free.fr>
Signed-off-by: David S. Miller <davem@davemloft.net>
Ethtool has 11 flow tables, each flow table has its own priority.
Increase the number of priorities to be aligned with the number of flow
tables.
Fixes: 1174fce8d1 ('net/mlx5e: Support l3/l4 flow type specs in ethtool flow steering')
Signed-off-by: Maor Gottlieb <maorg@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Upon set ETS failure due to user invalid input, add error prints to
specify the exact error to the user.
Fixes: cdcf11212b ('net/mlx5e: Validate BW weight values of ETS')
Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Free 'in' command object also when mlx5_core_modify_tir fails.
Fixes: 724b2aa151 ("net/mlx5e: TIRs management refactoring")
Signed-off-by: Kamal Heib <kamalh@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add a counter in ethtool for the number of times that
TX xmit_more was used.
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The driver RQ has two possible configurations: striding RQ and
non-striding RQ. Until this patch, the driver always reported the
number of hardware WQEs (ring descriptors). For non striding RQ
configuration, this was OK since we have one WQE per pending packet
For striding RQ, multiple packets can fit into one WQE. For better
user experience we normalize the rx_pending parameter (size of wqe/mtu)
as the average ring size in case of striding RQ.
Fixes: 461017cb00 ('net/mlx5e: Support RX multi-packet WQE ...')
Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Instead of asking the firmware to flush the SQ (Send Queue) via
asynchronous completions when moved to error, we handle SQ flush
manually (mlx5e_free_tx_descs) same as we did when SQ flush got
timed out or on tx_timeout.
This will reduce SQs flush time and speedup interface down procedure.
Moved mlx5e_free_tx_descs to the end of en_tx.c for tx
critical code locality.
Fixes: 29429f3300 ('net/mlx5e: Timeout if SQ doesn't flush during close')
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
ICO (Internal control operations) SQ (Send Queue) is closed/disabled
after RQ (Receive Queue). After RQ is closed an ICO SQ completion
might post a fragmented MPWQE (Multi Packet Work Queue Element) into
that RQ.
As on regular RQ post, check if we are allowed to post to that
RQ (RQ is enabled). Cleanup in-progress UMR MPWQE on mlx5e_free_rx_descs
if needed.
Fixes: bc77b240b3 ('net/mlx5e: Add fragmented memory support for RX multi packet WQE')
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>