There are two test cases:
1. Remove the 4 modules:hns_enet_drv/hns_dsaf/hnae/hns_mdio,
   and install them again, must use "ifconfig down/ifconfig up"
   command pair to bring port to work.

   This patch calls phy_stop function when init phy to fix this bug.

2. Remove the 2 modules:hns_enet_drv/hns_dsaf, and install them again,
   all ports can not use anymore, because of the phy devices register
   failed(phy devices already exists).

   Phy devices are registered when hns_dsaf installed, this patch
   removes them when hns_dsaf removed.

The two cases are sometimes related, fixing the second case also requires
fixing the first case, so fix them together.
Signed-off-by: Yonglong Liu <liuyonglong@huawei.com>
Signed-off-by: Peng Li <lipeng321@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

According to the hip06 Datasheet:
1. The offset of INGRESS_SW_VLAN_TAG_DISC should be 0x1A00+4*all_chn_num
2. The offset of INGRESS_IN_DATA_STP_DISC should be 0x1A50+4*all_chn_num
Signed-off-by: Yonglong Liu <liuyonglong@huawei.com>
Signed-off-by: Peng Li <lipeng321@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Sergey reported that forwarding was no longer working
if fq packet scheduler was used.

This is caused by the recent switch to EDT model, since incoming
packets might have been timestamped by __net_timestamp()

__net_timestamp() uses ktime_get_real(), while fq expects packets
using CLOCK_MONOTONIC base.

The fix is to clear skb->tstamp in forwarding paths.

Fixes: 80b14dee ("net: Add a new socket option for a future transmit time.")
Fixes: fb420d5d ("tcp/fq: move back to CLOCK_MONOTONIC")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Sergey Matyukevich <geomatsi@gmail.com>
Tested-by: Sergey Matyukevich <geomatsi@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Signed-off-by: Robert P. J. Day <rpjday@crashcourse.ca>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Since commit 7969e5c4 ("ip: discard IPv4 datagrams with overlapping
segments.") IPv4 reassembly code drops the whole queue whenever an
overlapping fragment is received. However, the test is written in a way
which detects duplicate fragments as overlapping so that in environments
with many duplicate packets, fragmented packets may be undeliverable.

Add an extra test and for (potentially) duplicate fragment, only drop the
new fragment rather than the whole queue. Only starting offset and length
are checked, not the contents of the fragments as that would be too
expensive. For similar reason, linear list ("run") of a rbtree node is not
iterated, we only check if the new fragment is a subset of the interval
covered by existing consecutive fragments.

v2: instead of an exact check iterating through linear list of an rbtree
node, only check if the new fragment is subset of the "run" (suggested
by Eric Dumazet)

Fixes: 7969e5c4 ("ip: discard IPv4 datagrams with overlapping segments.")
Signed-off-by: Michal Kubecek <mkubecek@suse.cz>
Signed-off-by: David S. Miller <davem@davemloft.net>

Added support for the Telit LN940 series cellular modules QMI interface.
QMI_QUIRK_SET_DTR quirk requied for Qualcomm MDM9x40 chipset.
Signed-off-by: Jörgen Storvist <jorgen.storvist@gmail.com>
Acked-by: Bjørn Mork <bjorn@mork.no>
Signed-off-by: David S. Miller <davem@davemloft.net>

Added support for Fibocom NL668 series QMI interface.
Using QMI_QUIRK_SET_DTR required for Qualcomm MDM9x07 chipsets.
Signed-off-by: Jörgen Storvist <jorgen.storvist@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Alexei Starovoitov says:

====================
pull-request: bpf 2018-12-15

The following pull-request contains BPF updates for your *net* tree.

The main changes are:

1) fix liveness propagation of callee saved registers, from Jakub.

2) fix overflow in bpf_jit_limit knob, from Daniel.

3) bpf_flow_dissector api fix, from Stanislav.

4) bpf_perf_event api fix on powerpc, from Sandipan.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

tipc_wait_for_cond() drops socket lock before going to sleep,
but tsk->group could be freed right after that release_sock().
So we have to re-check and reload tsk->group after it wakes up.

After this patch, tipc_wait_for_cond() returns -ERESTARTSYS when
tsk->group is NULL, instead of continuing with the assumption of
a non-NULL tsk->group.

(It looks like 'dsts' should be re-checked and reloaded too, but
it is a different bug.)

Similar for tipc_send_group_unicast() and tipc_send_group_anycast().

Reported-by: syzbot+10a9db47c3a0e13eb31c@syzkaller.appspotmail.com
Fixes: b7d42635 ("tipc: introduce flow control for group broadcast messages")
Fixes: ee106d7f ("tipc: introduce group anycast messaging")
Fixes: 27bd9ec0 ("tipc: introduce group unicast messaging")
Cc: Ying Xue <ying.xue@windriver.com>
Cc: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Acked-by: Ying Xue <ying.xue@windriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

While most distributions long ago switched to the iproute2 suite
of utilities, which allow class-e (240.0.0.0/4) address assignment,
distributions relying on busybox, toybox and other forms of
ifconfig cannot assign class-e addresses without this kernel patch.

While CIDR has been obsolete for 2 decades, and a survey of all the
open source code in the world shows the IN_whatever macros are also
obsolete... rather than obsolete CIDR from this ioctl entirely, this
patch merely enables class-e assignment, sanely.
Signed-off-by: Dave Taht <dave.taht@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

vr.mifi is indirectly controlled by user-space, hence leading to
a potential exploitation of the Spectre variant 1 vulnerability.

This issue was detected with the help of Smatch:

net/ipv6/ip6mr.c:1845 ip6mr_ioctl() warn: potential spectre issue 'mrt->vif_table' [r] (local cap)
net/ipv6/ip6mr.c:1919 ip6mr_compat_ioctl() warn: potential spectre issue 'mrt->vif_table' [r] (local cap)

Fix this by sanitizing vr.mifi before using it to index mrt->vif_table'

Notice that given that speculation windows are large, the policy is
to kill the speculation on the first load and not worry if it can be
completed with a dependent load/store [1].

[1] https://marc.info/?l=linux-kernel&m=152449131114778&w=2Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

The get_mac_address() function is normally inline, but when it is
not, we get a warning that this configuration is broken:

WARNING: vmlinux.o(.text+0x4aff00): Section mismatch in reference from the function w90p910_ether_setup() to the function .init.text:get_mac_address()
The function w90p910_ether_setup() references
the function __init get_mac_address().
This is often because w90p910_ether_setup lacks a __init

Remove the __init to make it always do the right thing.
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: David S. Miller <davem@davemloft.net>

The old code always starts from fixed port for VMADDR_PORT_ANY. Sometimes
when VMM crashed, there is still orphaned vsock which is waiting for
close timer, then it could cause connection time out for new started VM
if they are trying to connect to same port with same guest cid since the
new packets could hit that orphaned vsock. We could also fix this by doing
more in vhost_vsock_reset_orphans, but any way, it should be better to start
from a random local port instead of a fixed one.
Signed-off-by: Lepton Wu <ytht.net@gmail.com>
Reviewed-by: Jorgen Hansen <jhansen@vmware.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

All previous docks and dongles that have supported this feature use
the RTL8153-AD chip.

RTL8153-BND is a new chip that will be used in upcoming Dell type-C docks.
It should be added to the whitelist of devices to activate MAC address
pass through.

Per confirming with Realtek all devices containing RTL8153-BND should
activate MAC pass through and there won't use pass through bit on efuse
like in RTL8153-AD.
Signed-off-by: Mario Limonciello <mario.limonciello@dell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

recalculated send and receive window using linkspeed.
Determine correct value of eck_ok from SYN received and
option configured on local system.
Signed-off-by: Atul Gupta <atul.gupta@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

corrected macro used in tx path. removed redundant hdrlen
and check for !page in chtls_sendmsg
Signed-off-by: Atul Gupta <atul.gupta@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

listen fails when more than one tls capable device is
registered. tls_hw_hash is called for each dev which loops
again for each cdev_list causing listen failure. Hence
call chtls_listen_start/stop for specific device than loop over all
devices.
Signed-off-by: Atul Gupta <atul.gupta@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

HW unhash within mutex for registered tls devices cause sleep
when called from tcp_set_state for TCP_CLOSE. Release lock and
re-acquire after function call with ref count incr/dec.
defined kref and fp release for tls_device to ensure device
is not released outside lock.

BUG: sleeping function called from invalid context at
kernel/locking/mutex.c:748
in_atomic(): 1, irqs_disabled(): 0, pid: 0, name: swapper/7
INFO: lockdep is turned off.
CPU: 7 PID: 0 Comm: swapper/7 Tainted: G        W  O
Call Trace:
 <IRQ>
 dump_stack+0x5e/0x8b
 ___might_sleep+0x222/0x260
 __mutex_lock+0x5c/0xa50
 ? vprintk_emit+0x1f3/0x440
 ? kmem_cache_free+0x22d/0x2a0
 ? tls_hw_unhash+0x2f/0x80
 ? printk+0x52/0x6e
 ? tls_hw_unhash+0x2f/0x80
 tls_hw_unhash+0x2f/0x80
 tcp_set_state+0x5f/0x180
 tcp_done+0x2e/0xe0
 tcp_rcv_state_process+0x92c/0xdd3
 ? lock_acquire+0xf5/0x1f0
 ? tcp_v4_rcv+0xa7c/0xbe0
 ? tcp_v4_do_rcv+0x70/0x1e0
Signed-off-by: Atul Gupta <atul.gupta@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

create_ctx is called from tls_init and tls_hw_prot
hence initialize function pointers in common routine.
Signed-off-by: Atul Gupta <atul.gupta@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

Clang warns:

drivers/net/ethernet/apm/xgene/xgene_enet_main.c:33:36: warning:
tentative array definition assumed to have one element
static const struct acpi_device_id xgene_enet_acpi_match[];
                                   ^
1 warning generated.

Both xgene_enet_acpi_match and xgene_enet_of_match are defined before
their uses at the bottom of the file so this is unnecessary. When
CONFIG_ACPI is disabled, ACPI_PTR becomes NULL so xgene_enet_acpi_match
doesn't need to be defined.
Signed-off-by: Nathan Chancellor <natechancellor@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

When TIPC_NLA_UDP_REMOTE is an IPv6 mcast address but
TIPC_NLA_UDP_LOCAL is an IPv4 address, a NULL-ptr deref is triggered
as the UDP tunnel sock is initialized to IPv4 or IPv6 sock merely
based on the protocol in local address.

We should just error out when the remote address and local address
have different protocols.

Reported-by: syzbot+eb4da3a20fad2e52555d@syzkaller.appspotmail.com
Cc: Ying Xue <ying.xue@windriver.com>
Cc: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Acked-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

tipc_udp_xmit() drops the packet on error, there is no
need to drop it again.

Fixes: ef20cd4d ("tipc: introduce UDP replicast")
Reported-and-tested-by: syzbot+eae585ba2cc2752d3704@syzkaller.appspotmail.com
Cc: Ying Xue <ying.xue@windriver.com>
Cc: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

lock_sock() must be used in process context to be race-free with
other lock_sock() callers, for example, tipc_release(). Otherwise
using the spinlock directly can't serialize a parallel tipc_release().

As it is blocking, we have to hold the sock refcnt before
rhashtable_walk_stop() and release it after rhashtable_walk_start().

Fixes: 07f6c4bc ("tipc: convert tipc reference table to use generic rhashtable")
Reported-by: Dmitry Vyukov <dvyukov@google.com>
Cc: Ying Xue <ying.xue@windriver.com>
Cc: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

NETLINK_DUMP_STRICT_CHK can be used for all GET requests,
dumps as well as doit handlers.  Replace the DUMP in the
name with GET make that clearer.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

The value for OEM_CFG_UPDATE command differs between driver and the
Management firmware (mfw). Fix this gap with adding a reserved field.

Fixes: cac6f691 ("qed: Add support for Unified Fabric Port.")
Signed-off-by: Sudarsana Reddy Kalluru <Sudarsana.Kalluru@cavium.com>
Signed-off-by: Ariel Elior <ariel.elior@cavium.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

mlx5-fixes-2018-12-13

Subject: [pull request][net 0/9] Mellanox, mlx5 fixes 2018-12-13
Saeed Mahameed says:

====================
This series introduces some fixes to the mlx5 core and mlx5e netdevice
driver.

=======
Conflict with net-next: When merged with net-next this series will
cause a moderate conflict:

1) in drivers/net/ethernet/mellanox/mlx5/core/en_tc.c (2 hunks)
Take hunks from net only and just replace *attr->mirror_count to *attr->split_count
1.1) there is one more instance of slow_attr->mirror_count to be replaced
with slow_attr->split_count, it doesn't appear in the conflict, it will
cause a compilation error if left out.
2) in mlx5_ifc.h, take hunks only from net.

Example for the merge resolution can be found at:
https://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux.git/commit/?h=merge/mlx5-fixes&id=48830adf29804d85d77ed8a251d625db0eb5b8a8
branch merge/mlx5-fixes of git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux
(I simply merged this pull request tag into net-next and resolved the conflict)

I don't know if it's ok with you, but to save your time, you can just:
git pull git://git.kernel.org/pub/scm/linux/kernel/git/saeed/linux merge/mlx5-fixes
Into net-next, before your next net merge, and you will have a clean
merge of net into net-next (at least for mlx5 files).
======

Please pull and let me know if there's any problem.

For -stable v4.18
338d615be484 ('net/mlx5e: Cancel DIM work on close SQ')
91f40f9904ad ('net/mlx5e: RX, Verify MPWQE stride size is in range')

For -stable v4.19
c5c7e1c41bbe ('net/mlx5e: Remove unused UDP GSO remaining counter')
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

Currently for liveness and state pruning the register parentage
chains don't include states of the callee.  This makes some sense
as the callee can't access those registers.  However, this means
that READs done after the callee returns will not propagate into
the states of the callee.  Callee will then perform pruning
disregarding differences in caller state.

Example:

   0: (85) call bpf_user_rnd_u32
   1: (b7) r8 = 0
   2: (55) if r0 != 0x0 goto pc+1
   3: (b7) r8 = 1
   4: (bf) r1 = r8
   5: (85) call pc+4
   6: (15) if r8 == 0x1 goto pc+1
   7: (05) *(u64 *)(r9 - 8) = r3
   8: (b7) r0 = 0
   9: (95) exit

   10: (15) if r1 == 0x0 goto pc+0
   11: (95) exit

Here we acquire unknown state with call to get_random() [1].  Then
we store this random state in r8 (either 0 or 1) [1 - 3], and make
a call on line 5.  Callee does nothing but a trivial conditional
jump (to create a pruning point).  Upon return caller checks the
state of r8 and either performs an unsafe read or not.

Verifier will first explore the path with r8 == 1, creating a pruning
point at [11].  The parentage chain for r8 will include only callers
states so once verifier reaches [6] it will mark liveness only on states
in the caller, and not [11].  Now when verifier walks the paths with
r8 == 0 it will reach [11] and since REG_LIVE_READ on r8 was not
propagated there it will prune the walk entirely (stop walking
the entire program, not just the callee).  Since [6] was never walked
with r8 == 0, [7] will be considered dead and replaced with "goto -1"
causing hang at runtime.

This patch weaves the callee's explored states onto the callers
parentage chain.  Rough parentage for r8 would have looked like this
before:

[0] [1] [2] [3] [4] [5]   [10]      [11]      [6]      [7]
     |           |      ,---|----.    |        |        |
  sl0:         sl0:    / sl0:     \ sl0:      sl0:     sl0:
  fr0: r8 <-- fr0: r8<+--fr0: r8   `fr0: r8  ,fr0: r8<-fr0: r8
                       \ fr1: r8 <- fr1: r8 /
                        \__________________/

after:

[0] [1] [2] [3] [4] [5]   [10]      [11]      [6]      [7]
     |           |          |         |        |        |
   sl0:         sl0:      sl0:       sl0:      sl0:     sl0:
   fr0: r8 <-- fr0: r8 <- fr0: r8 <- fr0: r8 <-fr0: r8<-fr0: r8
                          fr1: r8 <- fr1: r8

Now the mark from instruction 6 will travel through callees states.

Note that we don't have to connect r0 because its overwritten by
callees state on return and r1 - r5 because those are not alive
any more once a call is made.

v2:
 - don't connect the callees registers twice (Alexei: suggestion & code)
 - add more details to the comment (Ed & Alexei)
v1: don't unnecessarily link caller saved regs (Jiong)

Fixes: f4d7e40a ("bpf: introduce function calls (verification)")
Reported-by: David Beckett <david.beckett@netronome.com>
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Reviewed-by: Jiong Wang <jiong.wang@netronome.com>
Reviewed-by: Edward Cree <ecree@solarflare.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>

TXQ SQ closure is followed by closing the corresponding CQ. A pending
DIM work would try to modify the now non-existing CQ.
This would trigger an error:
[85535.835926] mlx5_core 0000:af:00.0: mlx5_cmd_check:769:(pid 124399):
MODIFY_CQ(0x403) op_mod(0x0) failed, status bad resource state(0x9), syndrome (0x1d7771)

Fix by making sure to cancel any pending DIM work before destroying the SQ.

Fixes: cbce4f44 ("net/mlx5e: Enable adaptive-TX moderation")
Signed-off-by: Tal Gilboa <talgi@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

Remove tx_udp_seg_rem counter from ethtool output, as it is no longer
being updated in the driver's data flow.

Fixes: 3f44899e ("net/mlx5e: Use PARTIAL_GSO for UDP segmentation")
Signed-off-by: Mikhael Goikhman <migo@mellanox.com>
Reviewed-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

Currently, we are deleting offloaded encap flows in case the relevant neigh
becomes unconnected while the encap is valid (a sign that it used to be
connected), or if the curr neigh mac is different from the cached mac
(a sign that the remote side changed their mac).

The 2nd check also applies when the neigh becomes connected on the 1st
time (we start with zero mac). Before the offending commit, the deleting
handler was practically no op, as no flows were offloaded. But since
that commit, we offload neigh-less encap flows to slow path.

Under mirroring scheme, we go into the delete handler, attempt to unoffload a
mirror rule which was never set (as we were offloading to slow path) and crash.

Fix that by calling the delete handler only when the encap is valid,
which covers both cases mentioned above.

Fixes: 5dbe906f ('net/mlx5e: Use a slow path rule instead if vxlan neighbour isn't available')
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Paul Blakey <paulb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

When a neighbour is resolved, we delete the goto slow path rule from HW.

The eswitch flow attributes where not properly initialized on that case,
hence we mess up the eswitch refcounts for chain zero (the default one).

Fix that along with making sure to use semicolons and not commas on that code;

Fixes: 5dbe906f ('net/mlx5e: Use a slow path rule instead if vxlan neighbour isn't available')
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Paul Blakey <paulb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

Just a leftover which was wrongly left there, remove it while spawning
a message to suggest firmware upgrade.

Fixes: bf07aa73 ('net/mlx5e: Support offloading tc priorities and chains for eswitch flows')
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Paul Blakey <paulb@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

Currently we are not supporting this and not err-ing on that either.

For now, just err if asked to do that.

Fixes: bf07aa73 ('net/mlx5e: Support offloading tc priorities and chains for eswitch flows')
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Reported-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Eli Britstein <elibr@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

Add check of MPWQE stride size is within range supported by HW. In case
calculated MPWQE stride size exceed range, linear SKB can't be used and
we should use non linear MPWQE instead.

Fixes: 619a8f2a ("net/mlx5e: Use linear SKB in Striding RQ")
Signed-off-by: Moshe Shemesh <moshe@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

The default amount of channels a representor opens was erroneously
changed from one to the maximum amount of channels, restore to its
intended value.

Fixes: 779d986d ("net/mlx5e: Do not ignore netdevice TX/RX queues number")
Signed-off-by: Gavi Teitz <gavi@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

The cap bits locations for the fdb caps of multi path to table (used for
local mirroring) and multi encap (used for prio/chains) were wrongly used
in swapped locations. This went unnoted so far b/c we tested the offending
patch with CX5 FW that supports both of them. On different environments where
not both caps are supported, we will be messed up, fix that.

Fixes: b9aa0ba1 ('net/mlx5: Add cap bits for multi fdb encap')
Signed-off-by: Vu Pham <vu@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Tested-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>

Jason Wang says:

====================
Fix various issue of vhost

This series tries to fix various issues of vhost:

- Patch 1 adds a missing write barrier between used idx updating and
  logging.
- Patch 2-3 brings back the protection of device IOTLB through vq
  mutex, this fixes possible use after free in device IOTLB entries.

Please consider them for -stable.

Changes from V2:
- drop dirty page fix and make it for net-next
Changes from V1:
- silent compiler warning for 32bit.
- use mutex_trylock() on slowpath instead of mutex_lock() even on fast
  path.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>

This reverts commit 78139c94. We don't
protect device IOTLB with vq mutex, which will lead e.g use after free
for device IOTLB entries. And since we've switched to use
mutex_trylock() in previous patch, it's safe to revert it without
having deadlock.

Fixes: commit 78139c94 ("net: vhost: lock the vqs one by one")
Cc: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

We used to hold the mutex of paired virtqueue in
vhost_net_busy_poll(). But this will results an inconsistent lock
order which may cause deadlock if we try to bring back the protection
of device IOTLB with vq mutex that requires to hold mutex of all
virtqueues at the same time.

Fix this simply by switching to use mutex_trylock(), when fail just
skip the busy polling. This can happen when device IOTLB is under
updating which should be rare.

Fixes: commit 78139c94 ("net: vhost: lock the vqs one by one")
Cc: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

We miss a write barrier that guarantees used idx is updated and seen
before log. This will let userspace sync and copy used ring before
used idx is update. Fix this by adding a barrier before log_write().

Fixes: 8dd014ad ("vhost-net: mergeable buffers support")
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>