20G is not supported by production hardware and only the 40GbaseCR4 standard
is supported.
Signed-off-by: NMichael Chan <mchan@broadcom.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Cleanup bnxt_probe_phy() to cleanly separate 2 code blocks for autoneg
on and off.  Autoneg flow control is possible only if autoneg is enabled.

In bnxt_get_settings(), Pause and Asym_Pause are always supported.
Only the advertisement bits change depending on the ethtool -A setting
in auto mode.
Signed-off-by: NMichael Chan <mchan@broadcom.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

1. Determine autoneg on|off setting from link_info->autoneg.  Using the
firmware returned setting can be misleading if autoneg is changed and
there hasn't been a phy update from the firmware.

2. If autoneg is disabled, link_info->autoneg should be set to 0 to
indicate both speed and flow control autoneg are disabled.

3. To enable autoneg flow control, speed autoneg must be enabled.
Signed-off-by: NMichael Chan <mchan@broadcom.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

A recent change to the mdb code confused the compiler to the point
where it did not realize that the port-group returned from
br_mdb_add_group() is always valid when the function returns a nonzero
return value, so we get a spurious warning:

net/bridge/br_mdb.c: In function 'br_mdb_add':
net/bridge/br_mdb.c:542:4: error: 'pg' may be used uninitialized in this function [-Werror=maybe-uninitialized]
    __br_mdb_notify(dev, entry, RTM_NEWMDB, pg);

Slightly rearranging the code in br_mdb_add_group() makes the problem
go away, as gcc is clever enough to see that both functions check
for 'ret != 0'.
Signed-off-by: NArnd Bergmann <arnd@arndb.de>
Fixes: 9e8430f8 ("bridge: mdb: Passing the port-group pointer to br_mdb module")
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NHariprasad Shenai <hariprasad@chelsio.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Alexander Kochetkov says:

====================
Fixes for rockchip EMAC

Here is a set of 3 patches what fix koops, memory leak and
rockchip EMAC hang. Tested on radxarock lite.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

EMAC could be disabled, while there is some sb_buff
in use. That buffers got lost for linux.

In order to reproduce run on device during active ethernet work:
    ifconfig eth0 down
Signed-off-by: NAlexander Kochetkov <al.kochet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

EMAC reset internal tx ring pointer to zero at statup.
txbd_curr and txbd_dirty can be different from zero.
That cause ethernet transfer hang (no packets transmitted).

In order to reproduce, run on device:
    ifconfig eth0 down
    ifconfig eth0 up
Signed-off-by: NAlexander Kochetkov <al.kochet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

There is a race between arc_emac_tx() and arc_emac_tx_clean().
sk_buff got freed by arc_emac_tx_clean() while arc_emac_tx()
submitting sk_buff.

In order to free sk_buff arc_emac_tx_clean() checks:
    if ((info & FOR_EMAC) || !txbd->data)
        break;
    ...
    dev_kfree_skb_irq(skb);

If condition false, arc_emac_tx_clean() free sk_buff.

In order to submit txbd, arc_emac_tx() do:
    priv->tx_buff[*txbd_curr].skb = skb;
    ...
    priv->txbd[*txbd_curr].data = cpu_to_le32(addr);
    ...
    ...  <== arc_emac_tx_clean() check condition here
    ...  <== (info & FOR_EMAC) is false
    ...  <== !txbd->data is false
    ...
    *info = cpu_to_le32(FOR_EMAC | FIRST_OR_LAST_MASK | len);

In order to reproduce the situation,
run device:
    # iperf -s
run on host:
    # iperf -t 600 -c <device-ip-addr>

[   28.396284] ------------[ cut here ]------------
[   28.400912] kernel BUG at .../net/core/skbuff.c:1355!
[   28.414019] Internal error: Oops - BUG: 0 [#1] SMP ARM
[   28.419150] Modules linked in:
[   28.422219] CPU: 0 PID: 0 Comm: swapper/0 Tainted: G    B           4.4.0+ #120
[   28.429516] Hardware name: Rockchip (Device Tree)
[   28.434216] task: c0665070 ti: c0660000 task.ti: c0660000
[   28.439622] PC is at skb_put+0x10/0x54
[   28.443381] LR is at arc_emac_poll+0x260/0x474
[   28.447821] pc : [<c03af580>]    lr : [<c028fec4>]    psr: a0070113
[   28.447821] sp : c0661e58  ip : eea68502  fp : ef377000
[   28.459280] r10: 0000012c  r9 : f08b2000  r8 : eeb57100
[   28.464498] r7 : 00000000  r6 : ef376594  r5 : 00000077  r4 : ef376000
[   28.471015] r3 : 0030488b  r2 : ef13e880  r1 : 000005ee  r0 : eeb57100
[   28.477534] Flags: NzCv  IRQs on  FIQs on  Mode SVC_32  ISA ARM  Segment none
[   28.484658] Control: 10c5387d  Table: 8eaf004a  DAC: 00000051
[   28.490396] Process swapper/0 (pid: 0, stack limit = 0xc0660210)
[   28.496393] Stack: (0xc0661e58 to 0xc0662000)
[   28.500745] 1e40:                                                       00000002 00000000
[   28.508913] 1e60: 00000000 ef376520 00000028 f08b23b8 00000000 ef376520 ef7b6900 c028fc64
[   28.517082] 1e80: 2f158000 c0661ea8 c0661eb0 0000012c c065e900 c03bdeac ffff95e9 c0662100
[   28.525250] 1ea0: c0663924 00000028 c0661ea8 c0661ea8 c0661eb0 c0661eb0 0000001e c0660000
[   28.533417] 1ec0: 40000003 00000008 c0695a00 0000000a c066208c 00000100 c0661ee0 c0027410
[   28.541584] 1ee0: ef0fb700 2f158000 00200000 ffff95e8 00000004 c0662100 c0662080 00000003
[   28.549751] 1f00: 00000000 00000000 00000000 c065b45c 0000001e ef005000 c0647a30 00000000
[   28.557919] 1f20: 00000000 c0027798 00000000 c005cf40 f0802100 c0662ffc c0661f60 f0803100
[   28.566088] 1f40: c0661fb8 c00093bc c000ffb4 60070013 ffffffff c0661f94 c0661fb8 c00137d4
[   28.574267] 1f60: 00000001 00000000 00000000 c001ffa0 00000000 c0660000 00000000 c065a364
[   28.582441] 1f80: c0661fb8 c0647a30 00000000 00000000 00000000 c0661fb0 c000ffb0 c000ffb4
[   28.590608] 1fa0: 60070013 ffffffff 00000051 00000000 00000000 c005496c c0662400 c061bc40
[   28.598776] 1fc0: ffffffff ffffffff 00000000 c061b680 00000000 c0647a30 00000000 c0695294
[   28.606943] 1fe0: c0662488 c0647a2c c066619c 6000406a 413fc090 6000807c 00000000 00000000
[   28.615127] [<c03af580>] (skb_put) from [<ef376520>] (0xef376520)
[   28.621218] Code: e5902054 e590c090 e3520000 0a000000 (e7f001f2)
[   28.627307] ---[ end trace 4824734e2243fdb6 ]---

[   34.377068] Internal error: Oops: 17 [#1] SMP ARM
[   34.382854] Modules linked in:
[   34.385947] CPU: 0 PID: 3 Comm: ksoftirqd/0 Not tainted 4.4.0+ #120
[   34.392219] Hardware name: Rockchip (Device Tree)
[   34.396937] task: ef02d040 ti: ef05c000 task.ti: ef05c000
[   34.402376] PC is at __dev_kfree_skb_irq+0x4/0x80
[   34.407121] LR is at arc_emac_poll+0x130/0x474
[   34.411583] pc : [<c03bb640>]    lr : [<c028fd94>]    psr: 60030013
[   34.411583] sp : ef05de68  ip : 0008e83c  fp : ef377000
[   34.423062] r10: c001bec4  r9 : 00000000  r8 : f08b24c8
[   34.428296] r7 : f08b2400  r6 : 00000075  r5 : 00000019  r4 : ef376000
[   34.434827] r3 : 00060000  r2 : 00000042  r1 : 00000001  r0 : 00000000
[   34.441365] Flags: nZCv  IRQs on  FIQs on  Mode SVC_32  ISA ARM  Segment none
[   34.448507] Control: 10c5387d  Table: 8f25c04a  DAC: 00000051
[   34.454262] Process ksoftirqd/0 (pid: 3, stack limit = 0xef05c210)
[   34.460449] Stack: (0xef05de68 to 0xef05e000)
[   34.464827] de60:                   ef376000 c028fd94 00000000 c0669480 c0669480 ef376520
[   34.473022] de80: 00000028 00000001 00002ae4 ef376520 ef7b6900 c028fc64 2f158000 ef05dec0
[   34.481215] dea0: ef05dec8 0000012c c065e900 c03bdeac ffff983f c0662100 c0663924 00000028
[   34.489409] dec0: ef05dec0 ef05dec0 ef05dec8 ef05dec8 ef7b6000 ef05c000 40000003 00000008
[   34.497600] dee0: c0695a00 0000000a c066208c 00000100 ef05def8 c0027410 ef7b6000 40000000
[   34.505795] df00: 04208040 ffff983e 00000004 c0662100 c0662080 00000003 ef05c000 ef027340
[   34.513985] df20: ef05c000 c0666c2c 00000000 00000001 00000002 00000000 00000000 c0027568
[   34.522176] df40: ef027340 c003ef48 ef027300 00000000 ef027340 c003edd4 00000000 00000000
[   34.530367] df60: 00000000 c003c37c ffffff7f 00000001 00000000 ef027340 00000000 00030003
[   34.538559] df80: ef05df80 ef05df80 00000000 00000000 ef05df90 ef05df90 ef05dfac ef027300
[   34.546750] dfa0: c003c2a4 00000000 00000000 c000f578 00000000 00000000 00000000 00000000
[   34.554939] dfc0: 00000000 00000000 00000000 00000000 00000000 00000000 00000000 00000000
[   34.563129] dfe0: 00000000 00000000 00000000 00000000 00000013 00000000 ffffffff dfff7fff
[   34.571360] [<c03bb640>] (__dev_kfree_skb_irq) from [<c028fd94>] (arc_emac_poll+0x130/0x474)
[   34.579840] [<c028fd94>] (arc_emac_poll) from [<c03bdeac>] (net_rx_action+0xdc/0x28c)
[   34.587712] [<c03bdeac>] (net_rx_action) from [<c0027410>] (__do_softirq+0xcc/0x1f8)
[   34.595482] [<c0027410>] (__do_softirq) from [<c0027568>] (run_ksoftirqd+0x2c/0x50)
[   34.603168] [<c0027568>] (run_ksoftirqd) from [<c003ef48>] (smpboot_thread_fn+0x174/0x18c)
[   34.611466] [<c003ef48>] (smpboot_thread_fn) from [<c003c37c>] (kthread+0xd8/0xec)
[   34.619075] [<c003c37c>] (kthread) from [<c000f578>] (ret_from_fork+0x14/0x3c)
[   34.626317] Code: e8bd8010 e3a00000 e12fff1e e92d4010 (e59030a4)
[   34.632572] ---[ end trace cca5a3d86a82249a ]---
Signed-off-by: NAlexander Kochetkov <al.kochet@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

This patch corrects the unaligned accesses seen on GRE TEB tunnels when
generating hash keys.  Specifically what this patch does is make it so that
we force the use of skb_copy_bits when the GRE inner headers will be
unaligned due to NET_IP_ALIGNED being a non-zero value.
Signed-off-by: NAlexander Duyck <aduyck@mirantis.com>
Acked-by: NTom Herbert <tom@herbertland.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Saeed Mahameed says:

====================
mlx5 driver fixes for 4.5-rc2

We added here a patch from Matan and Alaa for addressing Linus comments on
the mess w.r.t reserved field names in the driver/firmware auto-generated file.

Once the patch hits linus tree, we'll ask Doug to rebase his tree on that
rc so both net-next and rdma-next development for 4.6 will be done under
the fixed robust form.

Also provided two patches that addresses the dynamic ndo initialization
issue of mlx5e netdevice.

Or and Saeed.

changes from V1: (Only first patch was changed)
In this V we fixed the issues addressed in Or's previous e-mail.
	1. Offsets took into account two dimensional u8 arrays
	2. Offsets took into account nesting unions and structs
	3. Offsets for unions
	4. Offsets for any reserved field
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently our netdevice ops is a one static global variable which
is referenced by all mlx5e netdevice instances. This can be
problematic when different driver instances do not share same
HW capabilities (e.g SRIOV PF and VFs probed to the host).

Now we have two constant global netdevice ops variables, one
for basic netdevice ops and the other with extended SRIOV ops,
on netdevice construction we choose the one suitable for
current device capabilities.

Fixes: 66e49ded ("net/mlx5e: Add support for SR-IOV ndos")
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Currently mlx5e_select_queue is redundant since num_tc is always 1.
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

mlx5_ifc.h is a header file representing the API and ABI between
the driver to the firmware and hardware. This file is used from
both the mlx5_ib and mlx5_core drivers.

Previously, this file used incrementing counter to indicate
reserved fields, for example:

struct mlx5_ifc_odp_per_transport_service_cap_bits {
        u8         send[0x1];
        u8         receive[0x1];
        u8         write[0x1];
        u8         read[0x1];
        u8         reserved_0[0x1];
        u8         srq_receive[0x1];
        u8         reserved_1[0x1a];
};

If one developer implements through net-next feature A that uses
reserved_0, they replace it with featureA and renames reserved_1 to
reserved_0. In the same kernel cycle, a 2nd developer could implement
feature B through the rdma tree, that uses reserved_1 and split it to
featureB and a smaller reserved_1 field. This will cause a conflict
when the two trees are merged.

The source of this conflict is that the 1st developer changed *all*
reserved fields.

As Linus suggested, we change the layout of structs to:

struct mlx5_ifc_odp_per_transport_service_cap_bits {
	u8         send[0x1];
	u8         receive[0x1];
	u8         write[0x1];
	u8         read[0x1];
	u8         reserved_at_4[0x1];
	u8         srq_receive[0x1];
	u8         reserved_at_6[0x1a];
};

This makes the conflicts much more rare and preserves the locality of
changes.
Signed-off-by: NMatan Barak <matanb@mellanox.com>
Signed-off-by: NAlaa Hleihel <alaa@mellanox.com>
Reported-by: NLinus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: NSaeed Mahameed <saeedm@mellanox.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

There is presently a race condition between the bonding periodic
link monitor and the updating of a slave's speed and duplex.  The former
occurs on a periodic basis, and the latter in response to a driver's
calling of netif_carrier_on.

	It is possible for the periodic monitor to run between the
driver call of netif_carrier_on and the receipt of the NETDEV_CHANGE
event that causes bonding to update the slave's speed and duplex.  This
manifests most notably as a report that a slave is up and "0 Mbps full
duplex" after enslavement, but in principle could report an incorrect
speed and duplex after any link up event if the device comes up with a
different speed or duplex.  This affects the 802.3ad aggregator
selection, as the speed and duplex are selection criteria.

	This is fixed by updating the speed and duplex in the periodic
monitor, prior to using that information.

	This was done historically in bonding, but the call to
bond_update_speed_duplex was removed in commit 876254ae ("bonding:
don't call update_speed_duplex() under spinlocks"), as it might sleep
under lock.  Later, the locking was changed to only hold RTNL, and so
after commit 876254ae ("bonding: don't call update_speed_duplex()
under spinlocks") this call is again safe.
Tested-by: N"Tantilov, Emil S" <emil.s.tantilov@intel.com>
Cc: Veaceslav Falico <vfalico@gmail.com>
Cc: dingtianhong <dingtianhong@huawei.com>
Fixes: 876254ae ("bonding: don't call update_speed_duplex() under spinlocks")
Signed-off-by: NJay Vosburgh <jay.vosburgh@canonical.com>
Acked-by: NDing Tianhong <dingtianhong@huawei.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The am79c961a.c driver fails to build with clang because of an
unusual inline assembly construct:

drivers/net/ethernet/amd/am79c961a.c:53:7: error: invalid % escape in inline assembly string
 "str%?h        %1, [%2]        @ NET_RAP\n\t"

The same change has been done a decade ago in arch/arm as of
6a39dd62 ("[ARM] 3759/2: Remove uses of %?"), but apparently
some drivers were missed.
Signed-off-by: NArnd Bergmann <arnd@arndb.de>
Acked-by: NRussell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The smc91x driver doesn't honor the probe deferral mechanism when the
interrupt source is not yet available, such as one provided by a gpio
controller not probed.

Fix this by propagating the platform_get_irq() error code as the probe
return value.
Signed-off-by: NRobert Jarzmik <robert.jarzmik@free.fr>
Cc: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Florian Fainelli says:

====================
Subject: [PATCH net v2 0/4] net: phy: bcm7xxx 40nm PHY fixes

Here is a collection of fixes for the 40nm Ethernet PHY supported
by the 7xxx PHY driver, please also queue these fixes for stable.

Changes in v2:

- dropped the cleanup patch, not appropriate
- added another patch removing bogus wildcard entries
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Remove the two wildcard entries, they serve no purpose and will match way too
many devices, some of them being covered by the driver in
drivers/net/phy/broadcom.c. Remove the now unused bcm7xxx_dummy_config_init()
function which would produce a warning.

Fixes: b560a58c ("net: phy: add Broadcom BCM7xxx internal PHY driver")
Signed-off-by: NFlorian Fainelli <f.fainelli@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Since we were wrongly advertising gigabit features for these 10/100 only
Ethernet PHYs, bcm7xxx_config_init() which is supposed to apply workaround
would have not run since the check would be true, now that we have fixed the
PHY features, remove that check since it has no reasoning to be there anymore.

Fixes: e18556ee ("net: phy: bcm7xxx: do not use PHY_BRCM_100MBPS_WAR")
Signed-off-by: NFlorian Fainelli <f.fainelli@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The PHY entries for BCM7425/29/35 declare the 40nm Ethernet PHY as being
10/100/1000 capable, while this is just a 10/100 capable PHY device, fix that.

Fixes: d068b02c ("net: phy: add BCM7425 and BCM7429 PHYs")
Fixes: 9458ceab ("net: phy: bcm7xxx: Add entry for BCM7435")
Signed-off-by: NFlorian Fainelli <f.fainelli@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The clear and set masks in the call to phy_set_clr_bits() called from
bcm7xxx_config_init() are inverted. We need to fix this by swapping the two
arguments, that is, set 0 bits, but clear the shade mode 2 enable bit.

Fixes: b560a58c ("net: phy: add Broadcom BCM7xxx internal PHY driver")
Signed-off-by: NFlorian Fainelli <f.fainelli@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Sergei Shtylyov says:

====================
ravb: fix the fallout of R-Car gen3 gPTP support

   Here's a set of 2 patches against DaveM's 'net.git' repo fixing up the
incomplete commit f5d7837f ("ravb: ptp: Add CONFIG mode support").
I'm proposing these as fixes but they can be merged as cleanups as well...

[1/2] ravb: kill duplicate setting of CCC.CSEL
[2/2] ravb: skip gPTP start/stop on R-Car gen3
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When adding support for the  R-Car gen3 gPTP active in configuration mode,
some call sites of ravb_ptp_{init|stop}() were missed due to an oversight.
Add  checks for the R-Car gen2 SoCs around these...

Fixes: f5d7837f ("ravb: ptp: Add CONFIG mode support")
Signed-off-by: NSergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When  adding support for the  R-Car gen3 gPTP active in configuration mode,
the code setting the CCC.CSEL field  was duplicated due to an oversight.
For R-Car gen 2 it's just redundant  and for R-Car gen3 the write at this
time is probably  ignored due to CCC.GAC bit being already  set...

Fixes: f5d7837f ("ravb: ptp: Add CONFIG mode support")
Signed-off-by: NSergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Pablo Neira Ayuso says:

====================
Netfilter fixes for net

The following patchset contain a rather large batch for your net that
includes accumulated bugfixes, they are:

1) Run conntrack cleanup from workqueue process context to avoid hitting
   soft lockup via watchdog for large tables. This is required by the
   IPv6 masquerading extension. From Florian Westphal.

2) Use original skbuff from nfnetlink batch when calling netlink_ack()
   on error since this needs to access the skb->sk pointer.

3) Incremental fix on top of recent Sasha Levin's lock fix for conntrack
   resizing.

4) Fix several problems in nfnetlink batch message header sanitization
   and error handling, from Phil Turnbull.

5) Select NF_DUP_IPV6 based on CONFIG_IPV6, from Arnd Bergmann.

6) Fix wrong signess in return values on nf_tables counter expression,
   from Anton Protopopov.

Due to the NetDev 1.1 organization burden, I had no chance to pass up
this to you any sooner in this release cycle, sorry about that.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The unix_dgram_sendmsg routine use the following test

if (unlikely(unix_peer(other) != sk && unix_recvq_full(other))) {

to determine if sk and other are in an n:1 association (either
established via connect or by using sendto to send messages to an
unrelated socket identified by address). This isn't correct as the
specified address could have been bound to the sending socket itself or
because this socket could have been connected to itself by the time of
the unix_peer_get but disconnected before the unix_state_lock(other). In
both cases, the if-block would be entered despite other == sk which
might either block the sender unintentionally or lead to trying to unlock
the same spin lock twice for a non-blocking send. Add a other != sk
check to guard against this.

Fixes: 7d267278 ("unix: avoid use-after-free in ep_remove_wait_queue")
Reported-By: NPhilipp Hahn <pmhahn@pmhahn.de>
Signed-off-by: NRainer Weikusat <rweikusat@mobileactivedefense.com>
Tested-by: NPhilipp Hahn <pmhahn@pmhahn.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The present unix_stream_read_generic contains various code sequences of
the form

err = -EDISASTER;
if (<test>)
	goto out;

This has the unfortunate side effect of possibly causing the error code
to bleed through to the final

out:
	return copied ? : err;

and then to be wrongly returned if no data was copied because the caller
didn't supply a data buffer, as demonstrated by the program available at

http://pad.lv/1540731

Change it such that err is only set if an error condition was detected.

Fixes: 3822b5c2 ("af_unix: Revert 'lock_interruptible' in stream receive code")
Reported-by: NJoseph Salisbury <joseph.salisbury@canonical.com>
Signed-off-by: NRainer Weikusat <rweikusat@mobileactivedefense.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

My analysis in the below mail applies, although the second part is
unnecessary because i isn't used in arithmetic operations here:

https://marc.info/?l=openbsd-tech&m=145377854103866&w=2

Thanks for your time.
Signed-off-by: NMichael McConville <mmcco@mykolab.com>
Acked-by: NFrancois Romieu <romieu@fr.zoreil.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

BRIDGE_VLAN_FILTERING automatically adds a newly bridged port to the
VLAN with the bridge's default_pvid.

The mv88e6xxx driver currently reserves VLANs 4000+ for unbridged ports
isolation. When a port joins a bridge, it leaves its reserved VLAN. When
a port leaves a bridge, it joins again its reserved VLAN.

But if the VLAN filtering is disabled, or if this hardware VLAN is
already in use, the bridged port ends up with no default VLAN, and the
communication with the CPU is thus broken.

To fix this, make a port join its reserved VLAN once on setup, never
leave it, and restore its PVID after another one was eventually used.
Signed-off-by: NVivien Didelot <vivien.didelot@savoirfairelinux.com>
Tested-by: NAndrew Lunn <andrew@lunn.ch>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

The current bridge code calls switchdev_port_obj_del on a VLAN port even
if the corresponding switchdev_port_obj_add call returned -EOPNOTSUPP.

If the DSA driver doesn't return -EOPNOTSUPP for a software port VLAN in
its port_vlan_del function, the VLAN is not deleted. Unbridging the port
also generates a stack trace for the same reason.

This can be quickly tested on a VLAN filtering enabled system with:

    # brctl addbr br0
    # brctl addif br0 lan0
    # brctl addbr br1
    # brctl addif br1 lan1
    # brctl delif br1 lan1

Both bridges have a default default_pvid set to 1. lan0 uses the
hardware VLAN 1 while lan1 falls back to the software VLAN 1.

Unbridging lan1 does not delete its software VLAN, and thus generates
the following stack trace:

    [ 2991.681705] device lan1 left promiscuous mode
    [ 2991.686237] br1: port 1(lan1) entered disabled state
    [ 2991.725094] ------------[ cut here ]------------
    [ 2991.729761] WARNING: CPU: 0 PID: 869 at net/bridge/br_vlan.c:314 __vlan_group_free+0x4c/0x50()
    [ 2991.738437] Modules linked in:
    [ 2991.741546] CPU: 0 PID: 869 Comm: ip Not tainted 4.4.0 #16
    [ 2991.747039] Hardware name: Freescale Vybrid VF5xx/VF6xx (Device Tree)
    [ 2991.753511] Backtrace:
    [ 2991.756008] [<80014450>] (dump_backtrace) from [<8001469c>] (show_stack+0x20/0x24)
    [ 2991.763604]  r6:80512644 r5:00000009 r4:00000000 r3:00000000
    [ 2991.769343] [<8001467c>] (show_stack) from [<80268e44>] (dump_stack+0x24/0x28)
    [ 2991.776618] [<80268e20>] (dump_stack) from [<80025568>] (warn_slowpath_common+0x98/0xc4)
    [ 2991.784750] [<800254d0>] (warn_slowpath_common) from [<80025650>] (warn_slowpath_null+0x2c/0x34)
    [ 2991.793557]  r8:00000000 r7:9f786a8c r6:9f76c440 r5:9f786a00 r4:9f68ac00
    [ 2991.800366] [<80025624>] (warn_slowpath_null) from [<80512644>] (__vlan_group_free+0x4c/0x50)
    [ 2991.808946] [<805125f8>] (__vlan_group_free) from [<80514488>] (nbp_vlan_flush+0x44/0x68)
    [ 2991.817147]  r4:9f68ac00 r3:9ec70000
    [ 2991.820772] [<80514444>] (nbp_vlan_flush) from [<80506f08>] (del_nbp+0xac/0x130)
    [ 2991.828201]  r5:9f56f800 r4:9f786a00
    [ 2991.831841] [<80506e5c>] (del_nbp) from [<8050774c>] (br_del_if+0x40/0xbc)
    [ 2991.838724]  r7:80590f68 r6:00000000 r5:9ec71c38 r4:9f76c440
    [ 2991.844475] [<8050770c>] (br_del_if) from [<80503dc0>] (br_del_slave+0x1c/0x20)
    [ 2991.851802]  r5:9ec71c38 r4:9f56f800
    [ 2991.855428] [<80503da4>] (br_del_slave) from [<80484a34>] (do_setlink+0x324/0x7b8)
    [ 2991.863043] [<80484710>] (do_setlink) from [<80485e90>] (rtnl_newlink+0x508/0x6f4)
    [ 2991.870616]  r10:00000000 r9:9ec71ba8 r8:00000000 r7:00000000 r6:9f6b0400 r5:9f56f800
    [ 2991.878548]  r4:8076278c
    [ 2991.881110] [<80485988>] (rtnl_newlink) from [<80484048>] (rtnetlink_rcv_msg+0x18c/0x22c)
    [ 2991.889315]  r10:9f7d4e40 r9:00000000 r8:00000000 r7:00000000 r6:9f7d4e40 r5:9f6b0400
    [ 2991.897250]  r4:00000000
    [ 2991.899814] [<80483ebc>] (rtnetlink_rcv_msg) from [<80497c74>] (netlink_rcv_skb+0xb0/0xcc)
    [ 2991.908104]  r8:00000000 r7:9f7d4e40 r6:9f7d4e40 r5:80483ebc r4:9f6b0400
    [ 2991.914928] [<80497bc4>] (netlink_rcv_skb) from [<80483eb4>] (rtnetlink_rcv+0x34/0x3c)
    [ 2991.922874]  r6:9f5ea000 r5:00000028 r4:9f7d4e40 r3:80483e80
    [ 2991.928622] [<80483e80>] (rtnetlink_rcv) from [<80497604>] (netlink_unicast+0x180/0x200)
    [ 2991.936742]  r4:9f4edc00 r3:80483e80
    [ 2991.940362] [<80497484>] (netlink_unicast) from [<80497a88>] (netlink_sendmsg+0x33c/0x350)
    [ 2991.948648]  r8:00000000 r7:00000028 r6:00000000 r5:9f5ea000 r4:9ec71f4c
    [ 2991.955481] [<8049774c>] (netlink_sendmsg) from [<80457ff0>] (sock_sendmsg+0x24/0x34)
    [ 2991.963342]  r10:00000000 r9:9ec71e28 r8:00000000 r7:9f1e2140 r6:00000000 r5:00000000
    [ 2991.971276]  r4:9ec71f4c
    [ 2991.973849] [<80457fcc>] (sock_sendmsg) from [<80458af0>] (___sys_sendmsg+0x1fc/0x204)
    [ 2991.981809] [<804588f4>] (___sys_sendmsg) from [<804598d0>] (__sys_sendmsg+0x4c/0x7c)
    [ 2991.989640]  r10:00000000 r9:9ec70000 r8:80010824 r7:00000128 r6:7ee946c4 r5:00000000
    [ 2991.997572]  r4:9f1e2140
    [ 2992.000128] [<80459884>] (__sys_sendmsg) from [<80459918>] (SyS_sendmsg+0x18/0x1c)
    [ 2992.007725]  r6:00000000 r5:7ee9c7b8 r4:7ee946e0
    [ 2992.012430] [<80459900>] (SyS_sendmsg) from [<80010660>] (ret_fast_syscall+0x0/0x3c)
    [ 2992.020182] ---[ end trace 5d4bc29f4da04280 ]---

To fix this, return -EOPNOTSUPP in _mv88e6xxx_port_vlan_del instead of
-ENOENT if the hardware VLAN doesn't exist or the port is not a member.
Signed-off-by: NVivien Didelot <vivien.didelot@savoirfairelinux.com>
Tested-by: NAndrew Lunn <andrew@lunn.ch>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

smatch detected a suspicious looking bitop condition:

drivers/net/ethernet/cavium/liquidio/lio_main.c:2529
  handle_timestamp() warn: suspicious bitop condition

(skb_shinfo(skb)->tx_flags | SKBTX_IN_PROGRESS is always non-zero,
so the logic is definitely not correct.  Use & to mask the correct
bit.
Signed-off-by: NColin Ian King <colin.king@canonical.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Commit c0eb4540 ("hv_netvsc: Don't ask for additional head room in the
skb") got rid of needed_headroom setting for the driver. With the change I
hit the following issue trying to use ptkgen module:

[   57.522021] kernel BUG at net/core/skbuff.c:1128!
[   57.522021] invalid opcode: 0000 [#1] SMP DEBUG_PAGEALLOC
...
[   58.721068] Call Trace:
[   58.721068]  [<ffffffffa0144e86>] netvsc_start_xmit+0x4c6/0x8e0 [hv_netvsc]
...
[   58.721068]  [<ffffffffa02f87fc>] ? pktgen_finalize_skb+0x25c/0x2a0 [pktgen]
[   58.721068]  [<ffffffff814f5760>] ? __netdev_alloc_skb+0xc0/0x100
[   58.721068]  [<ffffffffa02f9907>] pktgen_thread_worker+0x257/0x1920 [pktgen]

Basically, we're calling skb_cow_head(skb, RNDIS_AND_PPI_SIZE) and crash on
    if (skb_shared(skb))
        BUG();

We probably need to restore needed_headroom setting (but shrunk to
RNDIS_AND_PPI_SIZE as we don't need more) to request the required headroom
space. In theory, it should not give us performance penalty.
Signed-off-by: NVitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Gregory CLEMENT says:

====================
mvneta fixes for SMP

Following this bug report:
http://thread.gmane.org/gmane.linux.ports.arm.kernel/468173 and the
suggestions from Russell King, I reviewed all the code involving
multi-CPU. It ended with this series of patches which should improve
the stability of the driver.

During my test I found another bug which is fixed by new patch (the
second one of this new version of the series)

The two first patches fix real bugs, the others fix potential issues
in the driver.

Changelog:

v1 -> v2
Fix spinlock comment. Pointed by David Miller

v2 -> v3
 - Fix typos and mistake in the comments. Pointed by Sergei Shtylyov
 - Add a new patch fixing the CPU choice in mvneta_percpu_elect
 - Use lock in last patch to prevent remaining race condition. Pointed
   by Jisheng
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When stopping the port, the CPU notifier are still there whereas the
mvneta_stop_dev function calls mvneta_percpu_disable() on each CPUs.
It was possible to have a new CPU coming at this point which could be
racy.

This patch adds a flag preventing executing the code notifier for a new
CPU when the port is stopping. It also uses the spinlock introduces
previously. To avoid the deadlock, the lock has been moved outside the
mvneta_percpu_elect function.
Signed-off-by: NGregory CLEMENT <gregory.clement@free-electrons.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Electing a CPU must be done in an atomic way: it should be done after or
before the removal/insertion of a CPU and this function is not reentrant.

During the loop of mvneta_percpu_elect we associates the queues to the
CPUs, if there is a topology change during this loop, then the mapping
between the CPUs and the queues could be wrong. During this loop the
interrupt mask is also updating for each CPUs, It should not be changed
in the same time by other part of the driver.

This patch adds spinlock to create the needed critical sections.
Signed-off-by: NGregory CLEMENT <gregory.clement@free-electrons.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

In the MVNETA_INTR_* registers, the queues related fields are per cpu,
according to the datasheet (comment in [] are added by me):
"In a multi-CPU system, bits of RX[or TX] queues for which the access by
the reading[or writing] CPU is disabled are read as 0, and cannot be
cleared[or written]."

That means that each time we want to manipulate these bits we had to do
it on each cpu and not only on the current cpu.
Signed-off-by: NGregory CLEMENT <gregory.clement@free-electrons.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Since the commit 2dcf75e2 ("net: mvneta: Associate RX queues with
each CPU") all the percpu irq are used and disabled at initialization, so
there is no point to disable them first.
Signed-off-by: NGregory CLEMENT <gregory.clement@free-electrons.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Instead of using a for_each_* loop in which we just call the
smp_call_function_single macro, it is more simple to directly use the
on_each_cpu macro. Moreover, this macro ensures that the calls will be
done all at once.
Suggested-by: NRussell King <rmk+kernel@arm.linux.org.uk>
Signed-off-by: NGregory CLEMENT <gregory.clement@free-electrons.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

When passing to the management of multiple RX queue, the
mvneta_percpu_elect function was broken. The use of the modulo can lead
to elect the wrong cpu. For example with rxq_def=2, if the CPU 2 goes
offline and then online, we ended with the third RX queue activated in
the same time on CPU 0 and CPU2, which lead to a kernel crash.

With this fix, we don't try to get "the closer" CPU if the default CPU is
gone, now we just use CPU 0 which always be there. Thanks to this, the
code becomes more readable, easier to maintain and more predicable.

Cc: stable@vger.kernel.org
Fixes: 2dcf75e2 ("net: mvneta: Associate RX queues with each CPU")
Signed-off-by: NGregory CLEMENT <gregory.clement@free-electrons.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>