Previously the driver used pcie_get_minimum_link() to warn when the NIC
is in a slot that can't supply as much bandwidth as the NIC could use.

pcie_get_minimum_link() can be misleading because it finds the slowest link
and the narrowest link (which may be different links) without considering
the total bandwidth of each link.  For a path with a 16 GT/s x1 link and a
2.5 GT/s x16 link, it returns 2.5 GT/s x1, which corresponds to 250 MB/s of
bandwidth, not the true available bandwidth of about 1969 MB/s for a
16 GT/s x1 link.

Use pcie_print_link_status() to report PCIe link speed and possible
limitations instead of implementing this in the driver itself.  This finds
the slowest link in the path to the device by computing the total bandwidth
of each link and compares that with the capabilities of the device.

The dmesg change is:

  - PCIe: Speed %s Width x%d
  + %u.%03u Gb/s available PCIe bandwidth (%s x%d link)
Signed-off-by: NBjorn Helgaas <bhelgaas@google.com>

Previously the driver used pcie_get_minimum_link() to warn when the NIC
is in a slot that can't supply as much bandwidth as the NIC could use.

pcie_get_minimum_link() can be misleading because it finds the slowest link
and the narrowest link (which may be different links) without considering
the total bandwidth of each link.  For a path with a 16 GT/s x1 link and a
2.5 GT/s x16 link, it returns 2.5 GT/s x1, which corresponds to 250 MB/s of
bandwidth, not the true available bandwidth of about 1969 MB/s for a
16 GT/s x1 link.

Use pcie_print_link_status() to report PCIe link speed and possible
limitations instead of implementing this in the driver itself.  This finds
the slowest link in the path to the device by computing the total bandwidth
of each link and compares that with the capabilities of the device.

The dmesg change is:

  - %s (%c%d) PCI-E x%d %s found at mem %lx, IRQ %d, node addr %pM
  + %s (%c%d) PCI-E found at mem %lx, IRQ %d, node addr %pM
  + %u.%03u Gb/s available PCIe bandwidth (%s x%d link)
Signed-off-by: NBjorn Helgaas <bhelgaas@google.com>

I added dumping of link information about tun devices over netlink in
commit 1ec010e7 ("tun: export flags, uid, gid, queue information
over netlink"), but didn't add the missing netlink notifications when
the device's exported properties change.

This patch adds notifications when owner/group or flags are modified,
when queues are attached/detached, and when a tun fd is closed.
Reported-by: NThomas Haller <thaller@redhat.com>
Fixes: 1ec010e7 ("tun: export flags, uid, gid, queue information over netlink")
Signed-off-by: NSabrina Dubroca <sd@queasysnail.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Otherwise, register_netdevice advertises the creation of the device with
the default flags, instead of what the user requested.
Reported-by: NThomas Haller <thaller@redhat.com>
Fixes: 1ec010e7 ("tun: export flags, uid, gid, queue information over netlink")
Signed-off-by: NSabrina Dubroca <sd@queasysnail.net>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Commit 92571a1a ("lan78xx: Connect phy early") moves the PHY
initialisation into lan78xx_probe, but lan78xx_open subsequently calls
lan78xx_reset. As well as forcing a second round of link negotiation,
this reset frequently prevents the phy interrupt from being generated
(even though the link is up), rendering the interface unusable.

Fix this issue by removing the lan78xx_reset call from lan78xx_open.

Fixes: 92571a1a ("lan78xx: Connect phy early")
Signed-off-by: NPhil Elwell <phil@raspberrypi.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When open fails during ethtool -L ring change, for example, the driver
may crash at bnxt_free_irq() because bp->bnapi is NULL.

If we fail to allocate all the new rings, bnxt_open_nic() will free
all the memory including bp->bnapi.  Subsequent call to bnxt_close_nic()
will try to dereference bp->bnapi in bnxt_free_irq().

Fix it by checking for !bp->bnapi in bnxt_free_irq().

Fixes: e5811b8c ("bnxt_en: Add IRQ remapping logic.")
Signed-off-by: NMichael Chan <michael.chan@broadcom.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

With recent changes to reserve both L2 and RDMA rings, we need to include
the RDMA rings in bnxt_check_rings().  Otherwise we will under-estimate
the rings we need during ethtool -L and may lead to failure.

Fixes: fbcfc8e4 ("bnxt_en: Reserve completion rings and MSIX for bnxt_re RDMA driver.")
Signed-off-by: NMichael Chan <michael.chan@broadcom.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

While a VF is configured with a bigger mtu (> 1500), any packets that
are punted to the VF-rep (slow-path) get dropped by OVS kernel-datapath
with the following message: "dropped over-mtu packet". Fix this by
returning the max-mtu value for a VF-rep derived from its corresponding VF.
VF-rep's mtu can be changed using 'ip' command as shown in this example:

	$ ip link set bnxt0_pf0vf0 mtu 9000
Signed-off-by: NSriharsha Basavapatna <sriharsha.basavapatna@broadcom.com>
Signed-off-by: NMichael Chan <michael.chan@broadcom.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The driver currently uses src port field (along with other fields) in the
decap tunnel key, while looking up and adding tunnel nodes. This leads to
redundant cfa_decap_filter_alloc() requests to the FW and flow-miss in the
flow engine. Fix this by ignoring the src port field in decap tunnel nodes.

Fixes: f484f678 ("bnxt_en: add hwrm FW cmds for cfa_encap_record and decap_filter")
Signed-off-by: NSriharsha Basavapatna <sriharsha.basavapatna@broadcom.com>
Signed-off-by: NMichael Chan <michael.chan@broadcom.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Before this patch the following commands would succeed as far as the
user was concerned:

$ tc qdisc add dev p1p1 ingress
$ tc filter add dev p1p1 parent ffff: protocol all \
	flower skip_sw action drop
$ tc filter add dev p1p1 parent ffff: protocol ipv4 \
	flower skip_sw src_mac 00:02:00:00:00:01/44 action drop

The current flow offload infrastructure used does not support wildcard
matching for ethernet headers, so do not allow the second or third
commands to succeed.  If a user wants to drop traffic on that interface
the protocol and MAC addresses need to be specified explicitly:

$ tc qdisc add dev p1p1 ingress
$ tc filter add dev p1p1 parent ffff: protocol arp \
	flower skip_sw action drop
$ tc filter add dev p1p1 parent ffff: protocol ipv4 \
	flower skip_sw action drop
...
$ tc filter add dev p1p1 parent ffff: protocol ipv4 \
	flower skip_sw src_mac 00:02:00:00:00:01 action drop
$ tc filter add dev p1p1 parent ffff: protocol ipv4 \
	flower skip_sw src_mac 00:02:00:00:00:02 action drop
...

There are also checks for VLAN parameters in this patch as other callers
may wildcard those parameters even if tc does not.  Using different
flow infrastructure could allow this to work in the future for L2 flows,
but for now it does not.

Fixes: 2ae7408f ("bnxt_en: bnxt: add TC flower filter offload support")
Signed-off-by: NAndy Gospodarek <gospo@broadcom.com>
Signed-off-by: NMichael Chan <michael.chan@broadcom.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Fix ethtool .get_rxfh() crash by checking for valid indirection table
address before copying the data.
Signed-off-by: NMichael Chan <michael.chan@broadcom.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This is preferred to opencoding an IDA_INIT.

Link: http://lkml.kernel.org/r/20180313132639.17387-2-willy@infradead.orgSigned-off-by: NMatthew Wilcox <mawilcox@microsoft.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

In case netdev is closed at the moment of pci shutdown, aq_nic_stop
gets called second time. napi_disable in that case hangs indefinitely.
In other case, if device was never opened at all, we get oops because
of null pointer access.

We should invoke aq_nic_stop conditionally, only if device is running
at the moment of shutdown.
Reported-by: NDavid Arcari <darcari@redhat.com>
Fixes: 90869ddf ("net: aquantia: Implement pci shutdown callback")
Signed-off-by: NIgor Russkikh <igor.russkikh@aquantia.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

On ASUS XG-C100C with 1.5.44 firmware a special mode called "dirty wake"
is active. With this mode when motherboard gets powered (but no poweron
happens yet), NIC automatically enables powersave link and watches
for WOL packet.
This normally allows to powerup the PC after AC power failures.

Not all motherboards or bios settings gives power to PCI slots,
so this mode is not enabled on all the hardware.

4.16 linux driver introduced full hardware reset sequence
This is required since before that we had no NIC hardware
reset implemented and there were side effects of "not clean start".

But this full reset is incompatible with "dirty wake" WOL feature
it keeps the PHY link in a special mode forever. As a consequence,
driver sees no link and no traffic.

To fix this we forcibly change FW state to idle state before doing
the full reset. This makes FW to restore link state.

Fixes: c8c82eb3 net: aquantia: Introduce global AQC hardware reset sequence
Signed-off-by: NIgor Russkikh <igor.russkikh@aquantia.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The Cinterion AHS8 is a 3G device with one embedded WWAN interface
using cdc_ether as a driver.

The modem is controlled via AT commands through the exposed TTYs.

AT+CGDCONT write command can be used to activate or deactivate a WWAN
connection for a PDP context defined with the same command. UE
supports one WWAN adapter.
Signed-off-by: NBassem Boubaker <bassem.boubaker@actia.fr>
Acked-by: NOliver Neukum <oneukum@suse.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

On receiving a packet the state index points to the rstate which must be
used to fill up IP and TCP headers. But if the state index points to a
rstate which is unitialized, i.e. filled with zeros, it gets stuck in an
infinite loop inside ip_fast_csum trying to compute the ip checsum of a
header with zero length.

89.666953:   <2> [<ffffff9dd3e94d38>] slhc_uncompress+0x464/0x468
89.666965:   <2> [<ffffff9dd3e87d88>] ppp_receive_nonmp_frame+0x3b4/0x65c
89.666978:   <2> [<ffffff9dd3e89dd4>] ppp_receive_frame+0x64/0x7e0
89.666991:   <2> [<ffffff9dd3e8a708>] ppp_input+0x104/0x198
89.667005:   <2> [<ffffff9dd3e93868>] pppopns_recv_core+0x238/0x370
89.667027:   <2> [<ffffff9dd4428fc8>] __sk_receive_skb+0xdc/0x250
89.667040:   <2> [<ffffff9dd3e939e4>] pppopns_recv+0x44/0x60
89.667053:   <2> [<ffffff9dd4426848>] __sock_queue_rcv_skb+0x16c/0x24c
89.667065:   <2> [<ffffff9dd4426954>] sock_queue_rcv_skb+0x2c/0x38
89.667085:   <2> [<ffffff9dd44f7358>] raw_rcv+0x124/0x154
89.667098:   <2> [<ffffff9dd44f7568>] raw_local_deliver+0x1e0/0x22c
89.667117:   <2> [<ffffff9dd44c8ba0>] ip_local_deliver_finish+0x70/0x24c
89.667131:   <2> [<ffffff9dd44c92f4>] ip_local_deliver+0x100/0x10c

./scripts/faddr2line vmlinux slhc_uncompress+0x464/0x468 output:
 ip_fast_csum at arch/arm64/include/asm/checksum.h:40
 (inlined by) slhc_uncompress at drivers/net/slip/slhc.c:615

Adding a variable to indicate if the current rstate is initialized. If
such a packet arrives, move to toss state.
Signed-off-by: NTejaswi Tanikella <tejaswit@codeaurora.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

lan78xx_defer_event generates an error message whenever the work item
is already scheduled. lan78xx_open defers three events -
EVENT_STAT_UPDATE, EVENT_DEV_OPEN and EVENT_LINK_RESET. Being aware
of the likelihood (or certainty) of an error message, the DEV_OPEN
event is added to the set of pending events directly, relying on
the subsequent deferral of the EVENT_LINK_RESET call to schedule the
work.  Take the same precaution with EVENT_STAT_UPDATE to avoid a
totally unnecessary error message.
Signed-off-by: NPhil Elwell <phil@raspberrypi.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

lan78xx_read_otp tries to return -EINVAL in the event of invalid OTP
content, but the value gets overwritten before it is returned and the
read goes ahead anyway. Make the read conditional as it should be
and preserve the error code.

Fixes: 55d7de9d ("Microchip's LAN7800 family USB 2/3 to 10/100/1000 Ethernet device driver")
Signed-off-by: NPhil Elwell <phil@raspberrypi.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

It is too expensive to pass u64 values via linked list, instead
allocate array for them by overall number of mac addresses from netdev.

This eventually removes multiple kmalloc() calls, aviod memory
fragmentation and allow to put single null check on kmalloc
return value in order to prevent a potential null pointer dereference.

Addresses-Coverity-ID: 1467429 ("Dereference null return value")
Fixes: 37c3347e ("net: thunderx: add ndo_set_rx_mode callback implementation for VF")
Reported-by: NDan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: NVadim Lomovtsev <Vadim.Lomovtsev@cavium.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The datasheet specifies a 3uS pause after performing a software
reset. The default implementation of genphy_soft_reset() does not
provide this, so implement soft_reset with the needed pause.
Signed-off-by: NEsben Haabendal <eha@deif.com>
Reviewed-by: NAndrew Lunn <andrew@lunn.ch>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This resolves race during initialization where the resources with
ops are registered before driver and the structures used by occ_get
op is initialized. So keep occ_get callbacks registered only when
all structs are initialized.

The example flows, as it is in mlxsw:
1) driver load/asic probe:
   mlxsw_core
      -> mlxsw_sp_resources_register
        -> mlxsw_sp_kvdl_resources_register
          -> devlink_resource_register IDX
   mlxsw_spectrum
      -> mlxsw_sp_kvdl_init
        -> mlxsw_sp_kvdl_parts_init
          -> mlxsw_sp_kvdl_part_init
            -> devlink_resource_size_get IDX (to get the current setup
                                              size from devlink)
        -> devlink_resource_occ_get_register IDX (register current
                                                  occupancy getter)
2) reload triggered by devlink command:
  -> mlxsw_devlink_core_bus_device_reload
    -> mlxsw_sp_fini
      -> mlxsw_sp_kvdl_fini
	-> devlink_resource_occ_get_unregister IDX
    (struct mlxsw_sp *mlxsw_sp is freed at this point, call to occ get
     which is using mlxsw_sp would cause use-after free)
    -> mlxsw_sp_init
      -> mlxsw_sp_kvdl_init
        -> mlxsw_sp_kvdl_parts_init
          -> mlxsw_sp_kvdl_part_init
            -> devlink_resource_size_get IDX (to get the current setup
                                              size from devlink)
        -> devlink_resource_occ_get_register IDX (register current
                                                  occupancy getter)

Fixes: d9f9b9a4 ("devlink: Add support for resource abstraction")
Signed-off-by: NJiri Pirko <jiri@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This introduces a simpler and generic method for for finding (and mapping)
the TBIPA register.

Instead of relying of complicated logic for finding the TBIPA register
address based on the MDIO or MII register block base
address, which even in some cases relies on undocumented shadow registers,
a second "reg" entry for the mdio bus devicetree node specifies the TBIPA
register.

Backwards compatibility is kept, as the existing logic is applied when
only a single "reg" mapping is specified.
Signed-off-by: NEsben Haabendal <eha@deif.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When resetting the ibmvnic driver after a partition migration occurs
there is no requirement to do a reset of the main CRQ. The current
driver code does the required re-enable of the main CRQ, then does
a reset of the main CRQ later.

What we should be doing for a driver reset after a migration is to
re-enable the main CRQ, release all the sub-CRQs, and then allocate
new sub-CRQs after capability negotiation.

This patch updates the handling of mobility resets to do the proper
work and not reset the main CRQ. To do this the initialization/reset
of the main CRQ had to be moved out of the ibmvnic_init routine
and in to the ibmvnic_probe and do_reset routines.
Signed-off-by: NNathan Fontenot <nfont@linux.vnet.ibm.com>
Signed-off-by: NThomas Falcon <tlfalcon@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

There is a failover case for a non-redundant pseries VNIC
configuration that was not being handled properly. The current
implementation assumes that the driver will always have a redandant
device to communicate with following a failover notification. There
are cases, however, when a non-redundant configuration can receive
a failover request. If that happens, the driver should wait until
it receives a signal that the device is ready for operation.

The driver is agnostic of its backing hardware configuration,
so this fix necessarily affects all device failover management.
The driver needs to wait until it receives a signal that the device
is ready for resetting. A flag is introduced to track this intermediary
state where the driver is waiting for an active device.
Signed-off-by: NThomas Falcon <tlfalcon@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In some cases, if the driver is waiting for a reset following
a device parameter change, failure to schedule a reset can result
in a hang since a completion signal is never sent.

If the device configuration is being altered by a tool such
as ethtool or ifconfig, it could cause the console to hang
if the reset request does not get scheduled. Add some additional
error handling code to exit the wait_for_completion if there is
one in progress.
Signed-off-by: NThomas Falcon <tlfalcon@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The counter that tracks used TX descriptors pending completion
needs to be zeroed as part of a device reset. This change fixes
a bug causing transmit queues to be stopped unnecessarily and in
some cases a transmit queue stall and timeout reset. If the counter
is not reset, the remaining descriptors will not be "removed",
effectively reducing queue capacity. If the queue is over half full,
it will cause the queue to stall if stopped.
Signed-off-by: NThomas Falcon <tlfalcon@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Fix some mistakes caught by the DMA debugger. The first change
fixes a unnecessary unmap that should have been removed in an
earlier update. The next hunk fixes another bad unmap by zeroing
the bit checked to determine that an unmap is needed. The final
change fixes some buffers that are unmapped with the wrong
direction specified.
Signed-off-by: NThomas Falcon <tlfalcon@linux.vnet.ibm.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Joe Perches noted that we have a few source files that for some
inexplicable reason (read: I'm too lazy to even go look at the history)
are marked executable:

  drivers/gpu/drm/amd/amdgpu/vce_v4_0.c
  drivers/net/ethernet/cadence/macb_ptp.c

A simple git command line to show executable C/asm/header files is this:

    git ls-files -s '*.[chsS]' | grep '^100755'

and then you can fix them up with scripting by just feeding that output
into:

    | cut -f2 | xargs chmod -x

and commit it.

Which is exactly what this commit does.
Reported-by: NJoe Perches <joe@perches.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The LED2[2]/INTn pin on Marvell 88E1318S as well as 88E1510/12/14/18 needs
to be configured to be usable as interrupt not only when WOL is enabled,
but whenever we rely on interrupts from the PHY.
Signed-off-by: NEsben Haabendal <eha@deif.com>
Cc: Rasmus Villemoes <rasmus.villemoes@prevas.dk>
Reviewed-by: NAndrew Lunn <andrew@lunn.ch>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

1) Return correct size from ice_get_regs_len.
2) Fix incorrect use of ARRAY_SIZE in ice_get_regs.

Fixes: fcea6f3d (ice: Add stats and ethtool support)
Signed-off-by: NAnirudh Venkataramanan <anirudh.venkataramanan@intel.com>
Tested-by: NTony Brelinski <tonyx.brelinski@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Fix to return error code ICE_ERR_NO_MEMORY from the alloc error
handling case instead of 0, as done elsewhere in this function.

Fixes: dc49c772 ("ice: Get MAC/PHY/link info and scheduler topology")
Signed-off-by: NWei Yongjun <weiyongjun1@huawei.com>
Acked-by: NAnirudh Venkataramanan <anirudh.venkataramanan@intel.com>
Tested-by: NTony Brelinski <tonyx.brelinski@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>

Since the subsequent changes will need a TICK_USEC definition
analogous to TICK_NSEC, rename the existing TICK_USEC as
USER_TICK_USEC, update its users and redefine TICK_USEC
accordingly.
Suggested-by: NPeter Zijlstra <peterz@infradead.org>
Signed-off-by: NRafael J. Wysocki <rafael.j.wysocki@intel.com>
Acked-by: NPeter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: NFrederic Weisbecker <frederic@kernel.org>

Currently <linux/slab.h> #includes <linux/kmemleak.h> for no obvious
reason.  It looks like it's only a convenience, so remove kmemleak.h
from slab.h and add <linux/kmemleak.h> to any users of kmemleak_* that
don't already #include it.  Also remove <linux/kmemleak.h> from source
files that do not use it.

This is tested on i386 allmodconfig and x86_64 allmodconfig.  It would
be good to run it through the 0day bot for other $ARCHes.  I have
neither the horsepower nor the storage space for the other $ARCHes.

Update: This patch has been extensively build-tested by both the 0day
bot & kisskb/ozlabs build farms.  Both of them reported 2 build failures
for which patches are included here (in v2).

[ slab.h is the second most used header file after module.h; kernel.h is
  right there with slab.h. There could be some minor error in the
  counting due to some #includes having comments after them and I didn't
  combine all of those. ]

[akpm@linux-foundation.org: security/keys/big_key.c needs vmalloc.h, per sfr]
Link: http://lkml.kernel.org/r/e4309f98-3749-93e1-4bb7-d9501a39d015@infradead.org
Link: http://kisskb.ellerman.id.au/kisskb/head/13396/Signed-off-by: NRandy Dunlap <rdunlap@infradead.org>
Reviewed-by: NIngo Molnar <mingo@kernel.org>
Reported-by: Michael Ellerman <mpe@ellerman.id.au>	[2 build failures]
Reported-by: Fengguang Wu <fengguang.wu@intel.com>	[2 build failures]
Reviewed-by: NAndrew Morton <akpm@linux-foundation.org>
Cc: Wei Yongjun <weiyongjun1@huawei.com>
Cc: Luis R. Rodriguez <mcgrof@kernel.org>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Mimi Zohar <zohar@linux.vnet.ibm.com>
Cc: John Johansen <john.johansen@canonical.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The callers to netvsc_revoke_*_buf() and netvsc_teardown_*_gpadl()
already have their net_device instances. Pass them as a paramaeter to
the function instead of obtaining them from netvsc_device struct
everytime
Signed-off-by: NMohammed Gamal <mgamal@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Prior to commit 0cf73780 ("hv_netvsc: netvsc_teardown_gpadl() split")
the call sequence in netvsc_device_remove() was as follows (as
implemented in netvsc_destroy_buf()):
1- Send NVSP_MSG1_TYPE_REVOKE_RECV_BUF message
2- Teardown receive buffer GPADL
3- Send NVSP_MSG1_TYPE_REVOKE_SEND_BUF message
4- Teardown send buffer GPADL
5- Close vmbus

This didn't work for WS2016 hosts. Commit 0cf73780
("hv_netvsc: netvsc_teardown_gpadl() split") rearranged the
teardown sequence as follows:
1- Send NVSP_MSG1_TYPE_REVOKE_RECV_BUF message
2- Send NVSP_MSG1_TYPE_REVOKE_SEND_BUF message
3- Close vmbus
4- Teardown receive buffer GPADL
5- Teardown send buffer GPADL

That worked well for WS2016 hosts, but it prevented guests on older hosts from
shutting down after changing network settings. Commit 0ef58b0a
("hv_netvsc: change GPAD teardown order on older versions") ensured the
following message sequence for older hosts
1- Send NVSP_MSG1_TYPE_REVOKE_RECV_BUF message
2- Send NVSP_MSG1_TYPE_REVOKE_SEND_BUF message
3- Teardown receive buffer GPADL
4- Teardown send buffer GPADL
5- Close vmbus

However, with this sequence calling `ip link set eth0 mtu 1000` hangs and the
process becomes uninterruptible. On futher analysis it turns out that on tearing
down the receive buffer GPADL the kernel is waiting indefinitely
in vmbus_teardown_gpadl() for a completion to be signaled.

Here is a snippet of where this occurs:
int vmbus_teardown_gpadl(struct vmbus_channel *channel, u32 gpadl_handle)
{
        struct vmbus_channel_gpadl_teardown *msg;
        struct vmbus_channel_msginfo *info;
        unsigned long flags;
        int ret;

        info = kmalloc(sizeof(*info) +
                       sizeof(struct vmbus_channel_gpadl_teardown), GFP_KERNEL);
        if (!info)
                return -ENOMEM;

        init_completion(&info->waitevent);
        info->waiting_channel = channel;
[....]
        ret = vmbus_post_msg(msg, sizeof(struct vmbus_channel_gpadl_teardown),
                             true);

        if (ret)
                goto post_msg_err;

        wait_for_completion(&info->waitevent);
[....]
}

The completion is signaled from vmbus_ongpadl_torndown(), which gets called when
the corresponding message is received from the host, which apparently never happens
in that case.
This patch works around the issue by restoring the first mentioned message sequence
for older hosts

Fixes: 0ef58b0a ("hv_netvsc: change GPAD teardown order on older versions")
Signed-off-by: NMohammed Gamal <mgamal@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Split each of the functions into two for each of send/recv buffers.
This will be needed in order to implement a fine-grained messaging
sequence to the host so that we accommodate the requirements of
different Windows versions

Fixes: 0ef58b0a ("hv_netvsc: change GPAD teardown order on older versions")
Signed-off-by: NMohammed Gamal <mgamal@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When changing network interface settings, Windows guests
older than WS2016 can no longer shutdown. This was addressed
by commit 0ef58b0a ("hv_netvsc: change GPAD teardown order
on older versions"), however the issue also occurs on WS2012
guests that share NVSP protocol versions with WS2016 guests.
Hence we use Windows version directly to differentiate them.

Fixes: 0ef58b0a ("hv_netvsc: change GPAD teardown order on older versions")
Signed-off-by: NMohammed Gamal <mgamal@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Boundary check in mvpp2_prs_init_from_hw must be done according to the
passed "tid" parameter, not the mvpp2_prs_entry index, which is not yet
initialized at the time of the check.

Fixes: 47e0e14e ("net: mvpp2: Make mvpp2_prs_hw_read a parser entry init function")
Signed-off-by: NMaxime Chevallier <maxime.chevallier@bootlin.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This change updates the mlx5 interface to create mkey
on the device.

The updates in the command mailbox include increasing the
access mode type field to 5 bits in order to support additional
types such as MLX5_MKC_ACCESS_MODE_MEMIC which represents device
memory access type and will be used when registering MR on allocated
device memory.

All the places that use the old access mode format are adjusted as
well.
Signed-off-by: NAriel Levkovich <lariel@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>

This patch adds querying of device memory capabilities by the mlx5_core
driver during initialization.

Device memory capabilities is a new capability type and structure
which contains the necessary data that is needed for future device
memory allocation.

The presence of this new capabilities struct is indicated in the
general capabilities struct which is queried first by the driver.
If the presence bit is set, the driver will also query the new
capabilities struct and save it in the device context.
Signed-off-by: NAriel Levkovich <lariel@mellanox.com>
Signed-off-by: NLeon Romanovsky <leonro@mellanox.com>
Signed-off-by: NJason Gunthorpe <jgg@mellanox.com>