openwrt/staging/blogic.git
5 years agonet: hns3: fix keep_alive_timer not stop problem
Huazhong Tan [Sat, 6 Apr 2019 07:43:36 +0000 (15:43 +0800)]
net: hns3: fix keep_alive_timer not stop problem

When hclgevf_client_start() fails or VF driver unloaded, there is
nobody to disable keep_alive_timer.

So this patch fixes them.

Fixes: a6d818e31d08 ("net: hns3: Add vport alive state checking support")
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: Peng Li <lipeng321@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: hns3: remove resetting check in hclgevf_reset_task_schedule
Huazhong Tan [Sat, 6 Apr 2019 07:43:35 +0000 (15:43 +0800)]
net: hns3: remove resetting check in hclgevf_reset_task_schedule

The checking of HCLGEVF_STATE_RST_HANDLING flag in the
hclgevf_reset_task_schedule() will make some scheduling of
reset pending fail. This flag will be checked in the
hclgevf_reset_service_task(), it is unnecessary to check it
in the hclgevf_reset_task_schedule(). So this patch removes it.

Fixes: 35a1e50343bd ("net: hns3: Add VF Reset Service Task to support event handling")
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: hns3: add error handler for initializing command queue
Huazhong Tan [Sat, 6 Apr 2019 07:43:34 +0000 (15:43 +0800)]
net: hns3: add error handler for initializing command queue

This patch adds error handler for the failure of command queue
initialization both PF and VF.

Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: Peng Li <lipeng321@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: hns3: stop mailbox handling when command queue need re-init
Huazhong Tan [Sat, 6 Apr 2019 07:43:33 +0000 (15:43 +0800)]
net: hns3: stop mailbox handling when command queue need re-init

If the command queue needs re-initialization, the mailbox handling
task should do nothing, otherwise this task will just get some error
print.

Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: Peng Li <lipeng321@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: hns3: handle pending reset while reset fail
Huazhong Tan [Sat, 6 Apr 2019 07:43:32 +0000 (15:43 +0800)]
net: hns3: handle pending reset while reset fail

The ongoing lower-level reset will fail when there is a higher-level
reset occurs, so the error handler should deal with this situation.

Fixes: 6a5f6fa382f3 ("net: hns3: add error handler for hclgevf_reset()")
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: Peng Li <lipeng321@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: hns3: do not request reset when hardware resetting
Huazhong Tan [Sat, 6 Apr 2019 07:43:31 +0000 (15:43 +0800)]
net: hns3: do not request reset when hardware resetting

When hardware reset does not finish, the driver should not
request a new reset, otherwise the ongoing hardware reset
will get problem.

Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: hns3: ignore lower-level new coming reset
Huazhong Tan [Sat, 6 Apr 2019 07:43:30 +0000 (15:43 +0800)]
net: hns3: ignore lower-level new coming reset

It is unnecessary to deal with the new coming reset if
it is lower than the ongoing one.

Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: Peng Li <lipeng321@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: hns3: deactive the reset timer when reset successfully
Huazhong Tan [Sat, 6 Apr 2019 07:43:29 +0000 (15:43 +0800)]
net: hns3: deactive the reset timer when reset successfully

If the reset has been done successfully, the ongoing reset timer
is unnecessary.

Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: Peng Li <lipeng321@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: hns3: adjust the timing of hns3_client_stop when unloading
Huazhong Tan [Sat, 6 Apr 2019 07:43:28 +0000 (15:43 +0800)]
net: hns3: adjust the timing of hns3_client_stop when unloading

hns3_client_stop() should be called after unregister_netdev(),
otherwise the ongoing reset task may start the client just after it.

Fixes: a6d818e31d08 ("net: hns3: Add vport alive state checking support")
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: Peng Li <lipeng321@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: hns3: not reset vport who not alive when PF reset
Huazhong Tan [Sat, 6 Apr 2019 07:43:27 +0000 (15:43 +0800)]
net: hns3: not reset vport who not alive when PF reset

If a vport is not alive, it is unnecessary to notify it to reset
before PF asserting a reset. So before inform vport to reset,
we need to check its alive state firstly.

Fixes: aa5c4f175be6 ("net: hns3: add reset handling for VF when doing PF reset")
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: Peng Li <lipeng321@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: hns3: set up the vport alive state while reinitializing
Huazhong Tan [Sat, 6 Apr 2019 07:43:26 +0000 (15:43 +0800)]
net: hns3: set up the vport alive state while reinitializing

When reinitializing, the vport alive state needs to be set up.

Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: Peng Li <lipeng321@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: hns3: set vport alive state to default while resetting
Huazhong Tan [Sat, 6 Apr 2019 07:43:25 +0000 (15:43 +0800)]
net: hns3: set vport alive state to default while resetting

When resetting, the vport alive state should be set to default,
otherwise the alive state of the vport whose driver not running
is wrong before the timer to check it out.

Fixes: a6d818e31d08 ("net: hns3: Add vport alive state checking support")
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com>
Signed-off-by: Peng Li <lipeng321@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoMerge branch 'ipv4-Enable-support-for-IPv6-gateway-with-IPv4-routes'
David S. Miller [Mon, 8 Apr 2019 22:22:41 +0000 (15:22 -0700)]
Merge branch 'ipv4-Enable-support-for-IPv6-gateway-with-IPv4-routes'

David Ahern says:

====================
ipv4: Enable support for IPv6 gateway with IPv4 routes

Last set of three with the end goal of enabling IPv6 gateways with IPv4
routes.

This set adds fib6_nh_init and release to the IPv6 stubs, and adds neighbor
helpers that IPv4 code invokes to resolve an IPv6 address. When using
an IPv6 neighbor entry the hh_cache is bypassed as it contains the wrong
ethernet header for an IPv4 packet.

The nh_common nhc_has_gw was a temporary field used to convert existing
code from fib{6}_nh to fib_nh_common. That field is now converted to
nhc_gw_family to differentiate the address family of the gateway entry
as opposed to the address family of the container of fib_nh_common.

Existing code for rtable and fib_config is refactored to prepare
for a v6 address and then support is added. From there various
miscellaneous functions are updated to handle a v6 gateway - from
validating the v6 address to lookups in bpf code to verifying the
nexthop state.

Offload drivers - mlxsw and rocker - are modified to detect the v6
gateway and reject the route as 'unsupported'. e.g.,

    $ ip ro add 172.16.101.0/24 via inet6 fe80::202:ff:fe00:b dev swp1s0
    Error: mlxsw_spectrum: IPv6 gateway with IPv4 route is not supported.

This can be removed in time once support is added to each.

With the infrastructure changes in place, patch 17 enables it by adding
support for RTA_VIA to IPv4. RTA_VIA can be used for IPv4 addresses as
well. Only one of RTA_VIA and RTA_GATEWAY can be passed in a request.

Patch 18 adds a few test cases to fib_tests.sh.

v2
- comments from Ido - fixed typos as noted and updated messages
- add commit message to patch 1
- In patch 9, ipv4: Add fib_check_nh_v6_gw, moved the call to
  fib6_nh_release under the 'if (!err)' check as the intention is
  that release should not be called if init fails.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoselftests: fib_tests: Add tests for ipv6 gateway with ipv4 route
David Ahern [Fri, 5 Apr 2019 23:30:41 +0000 (16:30 -0700)]
selftests: fib_tests: Add tests for ipv6 gateway with ipv4 route

Add tests for ipv6 gateway with ipv4 route. Tests include basic
single path with ping to verify connectivity and multipath.

Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoipv4: Allow ipv6 gateway with ipv4 routes
David Ahern [Fri, 5 Apr 2019 23:30:40 +0000 (16:30 -0700)]
ipv4: Allow ipv6 gateway with ipv4 routes

Add support for RTA_VIA and allow an IPv6 nexthop for v4 routes:
   $ ip ro add 172.16.1.0/24 via inet6 2001:db8::1 dev eth0
   $ ip ro ls
   ...
   172.16.1.0/24 via inet6 2001:db8::1 dev eth0

For convenience and simplicity, userspace can use RTA_VIA to specify
AF_INET or AF_INET6 gateway.

The common fib_nexthop_info dump function compares the gateway address
family to the nh_common family to know if the gateway should be encoded
as RTA_VIA or RTA_GATEWAY.

Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoipv4: Flag fib_info with a fib_nh using IPv6 gateway
David Ahern [Fri, 5 Apr 2019 23:30:39 +0000 (16:30 -0700)]
ipv4: Flag fib_info with a fib_nh using IPv6 gateway

Until support is added to the offload drivers, they need to be able to
reject routes with an IPv6 gateway. To that end add a flag to fib_info
that indicates if any fib_nh has a v6 gateway. The flag allows the drivers
to efficiently know the use of a v6 gateway without walking all fib_nh
tied to a fib_info each time a route is added.

Update mlxsw and rocker to reject the routes with extack message as to why.

Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoipv4: Handle ipv6 gateway in fib_good_nh
David Ahern [Fri, 5 Apr 2019 23:30:38 +0000 (16:30 -0700)]
ipv4: Handle ipv6 gateway in fib_good_nh

Update fib_good_nh to handle an ipv6 gateway.

Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoipv4: Handle ipv6 gateway in fib_detect_death
David Ahern [Fri, 5 Apr 2019 23:30:37 +0000 (16:30 -0700)]
ipv4: Handle ipv6 gateway in fib_detect_death

Update fib_detect_death to handle an ipv6 gateway.

Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoipv4: Handle ipv6 gateway in ipv4_confirm_neigh
David Ahern [Fri, 5 Apr 2019 23:30:36 +0000 (16:30 -0700)]
ipv4: Handle ipv6 gateway in ipv4_confirm_neigh

Update ipv4_confirm_neigh to handle an ipv6 gateway.

Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agobpf: Handle ipv6 gateway in bpf_ipv4_fib_lookup
David Ahern [Fri, 5 Apr 2019 23:30:35 +0000 (16:30 -0700)]
bpf: Handle ipv6 gateway in bpf_ipv4_fib_lookup

Update bpf_ipv4_fib_lookup to handle an ipv6 gateway.

Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoipv4: Add helpers for neigh lookup for nexthop
David Ahern [Fri, 5 Apr 2019 23:30:34 +0000 (16:30 -0700)]
ipv4: Add helpers for neigh lookup for nexthop

A common theme in the output path is looking up a neigh entry for a
nexthop, either the gateway in an rtable or a fallback to the daddr
in the skb:

        nexthop = (__force u32)rt_nexthop(rt, ip_hdr(skb)->daddr);
        neigh = __ipv4_neigh_lookup_noref(dev, nexthop);
        if (unlikely(!neigh))
                neigh = __neigh_create(&arp_tbl, &nexthop, dev, false);

To allow the nexthop to be an IPv6 address we need to consider the
family of the nexthop and then call __ipv{4,6}_neigh_lookup_noref based
on it.

To make this simpler, add a ip_neigh_gw4 helper similar to ip_neigh_gw6
added in an earlier patch which handles:

        neigh = __ipv4_neigh_lookup_noref(dev, nexthop);
        if (unlikely(!neigh))
                neigh = __neigh_create(&arp_tbl, &nexthop, dev, false);

And then add a second one, ip_neigh_for_gw, that calls either
ip_neigh_gw4 or ip_neigh_gw6 based on the address family of the gateway.

Update the output paths in the VRF driver and core v4 code to use
ip_neigh_for_gw simplifying the family based lookup and making both
ready for a v6 nexthop.

ipv4_neigh_lookup has a different need - the potential to resolve a
passed in address in addition to any gateway in the rtable or skb. Since
this is a one-off, add ip_neigh_gw4 and ip_neigh_gw6 diectly. The
difference between __neigh_create used by the helpers and neigh_create
called by ipv4_neigh_lookup is taking a refcount, so add rcu_read_lock_bh
and bump the refcnt on the neigh entry.

Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoneighbor: Add skip_cache argument to neigh_output
David Ahern [Fri, 5 Apr 2019 23:30:33 +0000 (16:30 -0700)]
neighbor: Add skip_cache argument to neigh_output

A later patch allows an IPv6 gateway with an IPv4 route. The neighbor
entry will exist in the v6 ndisc table and the cached header will contain
the ipv6 protocol which is wrong for an IPv4 packet. For an IPv4 packet to
use the v6 neighbor entry, neigh_output needs to skip the cached header
and just use the output callback for the neigh entry.

A future patchset can look at expanding the hh_cache to handle 2
protocols. For now, IPv6 gateways with an IPv4 route will take the
extra overhead of generating the header.

Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoipv4: Add fib_check_nh_v6_gw
David Ahern [Fri, 5 Apr 2019 23:30:32 +0000 (16:30 -0700)]
ipv4: Add fib_check_nh_v6_gw

Add helper to use fib6_nh_init to validate a nexthop spec with an IPv6
gateway.

Signed-off-by: David Ahern <dsahern@gmail.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoipv4: Refactor fib_check_nh
David Ahern [Fri, 5 Apr 2019 23:30:31 +0000 (16:30 -0700)]
ipv4: Refactor fib_check_nh

fib_check_nh is currently huge covering multiple uses cases - device only,
device + gateway, and device + gateway with ONLINK. The next patch adds
validation checks for IPv6 which only further complicates it. So, break
fib_check_nh into 2 helpers - one for gateway validation and one for device
only.

Signed-off-by: David Ahern <dsahern@gmail.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoipv4: Add support to fib_config for IPv6 gateway
David Ahern [Fri, 5 Apr 2019 23:30:30 +0000 (16:30 -0700)]
ipv4: Add support to fib_config for IPv6 gateway

Add support for an IPv6 gateway to fib_config. Since a gateway is either
IPv4 or IPv6, make it a union with fc_gw4 where fc_gw_family decides
which address is in use. Update current checks on family and gw4 to
handle ipv6 as well.

Signed-off-by: David Ahern <dsahern@gmail.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoipv4: Add support to rtable for ipv6 gateway
David Ahern [Fri, 5 Apr 2019 23:30:29 +0000 (16:30 -0700)]
ipv4: Add support to rtable for ipv6 gateway

Add support for an IPv6 gateway to rtable. Since a gateway is either
IPv4 or IPv6, make it a union with rt_gw4 where rt_gw_family decides
which address is in use.

When dumping the route data, encode an ipv6 nexthop using RTA_VIA.

Signed-off-by: David Ahern <dsahern@gmail.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoipv4: Prepare fib_config for IPv6 gateway
David Ahern [Fri, 5 Apr 2019 23:30:28 +0000 (16:30 -0700)]
ipv4: Prepare fib_config for IPv6 gateway

Similar to rtable, fib_config needs to allow the gateway to be either an
IPv4 or an IPv6 address. To that end, rename fc_gw to fc_gw4 to mean an
IPv4 address and add fc_gw_family. Checks on 'is a gateway set' are changed
to see if fc_gw_family is set. In the process prepare the code for a
fc_gw_family == AF_INET6.

Signed-off-by: David Ahern <dsahern@gmail.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoipv4: Prepare rtable for IPv6 gateway
David Ahern [Fri, 5 Apr 2019 23:30:27 +0000 (16:30 -0700)]
ipv4: Prepare rtable for IPv6 gateway

To allow the gateway to be either an IPv4 or IPv6 address, remove
rt_uses_gateway from rtable and replace with rt_gw_family. If
rt_gw_family is set it implies rt_uses_gateway. Rename rt_gateway
to rt_gw4 to represent the IPv4 version.

Signed-off-by: David Ahern <dsahern@gmail.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: Replace nhc_has_gw with nhc_gw_family
David Ahern [Fri, 5 Apr 2019 23:30:26 +0000 (16:30 -0700)]
net: Replace nhc_has_gw with nhc_gw_family

Allow the gateway in a fib_nh_common to be from a different address
family than the outer fib{6}_nh. To that end, replace nhc_has_gw with
nhc_gw_family and update users of nhc_has_gw to check nhc_gw_family.
Now nhc_family is used to know if the nh_common is part of a fib_nh
or fib6_nh (used for container_of to get to route family specific data),
and nhc_gw_family represents the address family for the gateway.

Signed-off-by: David Ahern <dsahern@gmail.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoipv6: Add neighbor helpers that use the ipv6 stub
David Ahern [Fri, 5 Apr 2019 23:30:25 +0000 (16:30 -0700)]
ipv6: Add neighbor helpers that use the ipv6 stub

Add ipv6 helpers to handle ndisc references via the stub. Update
bpf_ipv6_fib_lookup to use __ipv6_neigh_lookup_noref_stub instead of
the open code ___neigh_lookup_noref with the stub.

Signed-off-by: David Ahern <dsahern@gmail.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoipv6: Add fib6_nh_init and release to stubs
David Ahern [Fri, 5 Apr 2019 23:30:24 +0000 (16:30 -0700)]
ipv6: Add fib6_nh_init and release to stubs

Add fib6_nh_init and fib6_nh_release to ipv6_stubs. If fib6_nh_init fails,
callers should not invoke fib6_nh_release, so there is no reason to have
a dummy stub for the IPv6 is not enabled case.

Signed-off-by: David Ahern <dsahern@gmail.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: phy: improve link partner capability detection
Heiner Kallweit [Fri, 5 Apr 2019 19:23:13 +0000 (21:23 +0200)]
net: phy: improve link partner capability detection

genphy_read_status() so far checks phydev->supported, not the actual
PHY capabilities. This can make a difference if the supported speeds
have been limited by of_set_phy_supported() or phy_set_max_speed().

It seems that this issue only affects the link partner advertisements
as displayed by ethtool. Also this patch wouldn't apply to older
kernels because linkmode bitmaps have been introduced recently.
Therefore net-next.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoMerge tag 'mlx5-updates-2019-04-02' of git://git.kernel.org/pub/scm/linux/kernel...
David S. Miller [Mon, 8 Apr 2019 21:31:25 +0000 (14:31 -0700)]
Merge tag 'mlx5-updates-2019-04-02' of git://git./linux/kernel/git/saeed/linux

Saeed Mamameed says:

====================
mlx5-updates-2019-04-02

This series provides misc updates to mlx5 driver

1) Aya Levin (1): Handle event of power detection in the PCIE slot

2) Eli Britstein (6):
  Some TC VLAN related updates and fixes to the previous VLAN modify action
  support patchset.
  Offload TC e-switch rules with egress/ingress VLAN devices

3) Max Gurtovoy (1): Fix double mutex initialization in esiwtch.c

4) Tariq Toukan (3): Misc small updates
  A write memory barrier is sufficient in EQ ci update
  Obsolete param field holding a constant value
  Unify logic of MTU boundaries

5) Tonghao Zhang (4): Misc updates to en_tc.c
  Make the log friendly when decapsulation offload not supported
  Remove 'parse_attr' argument in parse_tc_fdb_actions()
  Deletes unnecessary setting of esw_attr->parse_attr
  Return -EOPNOTSUPP when attempting to offload an unsupported action
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agocxgb4: Don't return EAGAIN when TCAM is full.
Vishal Kulkarni [Mon, 8 Apr 2019 12:33:49 +0000 (18:03 +0530)]
cxgb4: Don't return EAGAIN when TCAM is full.

During hash filter programming, driver needs to return ENOSPC error
intead of EAGAIN when TCAM is full.

Signed-off-by: Vishal Kulkarni <vishal@chelsio.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: xilinx: emaclite: add minimal ndo_do_ioctl hook
Alexandru Ardelean [Mon, 8 Apr 2019 09:01:57 +0000 (12:01 +0300)]
net: xilinx: emaclite: add minimal ndo_do_ioctl hook

This hook only implements a minimal set of ioctl hooks to be able to access
MII regs by using phytool.
When using this simple MAC controller, it's pretty difficult to do
debugging of the PHY chip without checking MII regs.

Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Reviewed-by: Radhey Shyam Pandey <radhey.shyam.pandey@xilinx.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: xilinx: emaclite: add minimal ethtool ops
Alexandru Ardelean [Mon, 8 Apr 2019 09:01:36 +0000 (12:01 +0300)]
net: xilinx: emaclite: add minimal ethtool ops

This set adds a minimal set of ethtool hooks to the driver, which provide a
decent amount of link information via ethtool.
With this change, running `ethtool ethX` in user-space provides all the
neatly-formatted information about the link (what was negotiated, what is
advertised, etc).

Signed-off-by: Alexandru Ardelean <alexandru.ardelean@analog.com>
Reviewed-by: Radhey Shyam Pandey <radhey.shyam.pandey@xilinx.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agodatagram: remove rendundant 'peeked' argument
Paolo Abeni [Mon, 8 Apr 2019 08:15:59 +0000 (10:15 +0200)]
datagram: remove rendundant 'peeked' argument

After commit a297569fe00a ("net/udp: do not touch skb->peeked unless
really needed") the 'peeked' argument of __skb_try_recv_datagram()
and friends is always equal to !!'flags & MSG_PEEK'.

Since such argument is really a boolean info, and the callers have
already 'flags & MSG_PEEK' handy, we can remove it and clean-up the
code a bit.

Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Acked-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: sched: flower: insert filter to ht before offloading it to hw
Vlad Buslov [Fri, 5 Apr 2019 17:56:26 +0000 (20:56 +0300)]
net: sched: flower: insert filter to ht before offloading it to hw

John reports:

Recent refactoring of fl_change aims to use the classifier spinlock to
avoid the need for rtnl lock. In doing so, the fl_hw_replace_filer()
function was moved to before the lock is taken. This can create problems
for drivers if duplicate filters are created (commmon in ovs tc offload
due to filters being triggered by user-space matches).

Drivers registered for such filters will now receive multiple copies of
the same rule, each with a different cookie value. This means that the
drivers would need to do a full match field lookup to determine
duplicates, repeating work that will happen in flower __fl_lookup().
Currently, drivers do not expect to receive duplicate filters.

To fix this, verify that filter with same key is not present in flower
classifier hash table and insert the new filter to the flower hash table
before offloading it to hardware. Implement helper function
fl_ht_insert_unique() to atomically verify/insert a filter.

This change makes filter visible to fast path at the beginning of
fl_change() function, which means it can no longer be freed directly in
case of error. Refactor fl_change() error handling code to deallocate the
filter with rcu timeout.

Fixes: 620da4860827 ("net: sched: flower: refactor fl_change")
Reported-by: John Hurley <john.hurley@netronome.com>
Signed-off-by: Vlad Buslov <vladbu@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoMerge branch 'rhashtable-bitlocks'
David S. Miller [Mon, 8 Apr 2019 02:12:12 +0000 (19:12 -0700)]
Merge branch 'rhashtable-bitlocks'

NeilBrown says:

====================
Convert rhashtable to use bitlocks

This series converts rhashtable to use a per-bucket bitlock
rather than a separate array of spinlocks.
This:
  reduces memory usage
  results in slightly fewer memory accesses
  slightly improves parallelism
  makes a configuration option unnecessary

The main change from previous version is to use a distinct type for
the pointer in the bucket which has a bit-lock in it.  This
helped find two places where rht_ptr() was missed, one
in  rhashtable_free_and_destroy() in print_ht in the test code.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agorhashtable: add lockdep tracking to bucket bit-spin-locks.
NeilBrown [Mon, 1 Apr 2019 23:07:45 +0000 (10:07 +1100)]
rhashtable: add lockdep tracking to bucket bit-spin-locks.

Native bit_spin_locks are not tracked by lockdep.

The bit_spin_locks used for rhashtable buckets are local
to the rhashtable implementation, so there is little opportunity
for the sort of misuse that lockdep might detect.
However locks are held while a hash function or compare
function is called, and if one of these took a lock,
a misbehaviour is possible.

As it is quite easy to add lockdep support this unlikely
possibility seems to be enough justification.

So create a lockdep class for bucket bit_spin_lock and attach
through a lockdep_map in each bucket_table.

Without the 'nested' annotation in rhashtable_rehash_one(), lockdep
correctly reports a possible problem as this lock is taken
while another bucket lock (in another table) is held.  This
confirms that the added support works.
With the correct nested annotation in place, lockdep reports
no problems.

Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agorhashtable: use bit_spin_locks to protect hash bucket.
NeilBrown [Mon, 1 Apr 2019 23:07:45 +0000 (10:07 +1100)]
rhashtable: use bit_spin_locks to protect hash bucket.

This patch changes rhashtables to use a bit_spin_lock on BIT(1) of the
bucket pointer to lock the hash chain for that bucket.

The benefits of a bit spin_lock are:
 - no need to allocate a separate array of locks.
 - no need to have a configuration option to guide the
   choice of the size of this array
 - locking cost is often a single test-and-set in a cache line
   that will have to be loaded anyway.  When inserting at, or removing
   from, the head of the chain, the unlock is free - writing the new
   address in the bucket head implicitly clears the lock bit.
   For __rhashtable_insert_fast() we ensure this always happens
   when adding a new key.
 - even when lockings costs 2 updates (lock and unlock), they are
   in a cacheline that needs to be read anyway.

The cost of using a bit spin_lock is a little bit of code complexity,
which I think is quite manageable.

Bit spin_locks are sometimes inappropriate because they are not fair -
if multiple CPUs repeatedly contend of the same lock, one CPU can
easily be starved.  This is not a credible situation with rhashtable.
Multiple CPUs may want to repeatedly add or remove objects, but they
will typically do so at different buckets, so they will attempt to
acquire different locks.

As we have more bit-locks than we previously had spinlocks (by at
least a factor of two) we can expect slightly less contention to
go with the slightly better cache behavior and reduced memory
consumption.

To enhance type checking, a new struct is introduced to represent the
  pointer plus lock-bit
that is stored in the bucket-table.  This is "struct rhash_lock_head"
and is empty.  A pointer to this needs to be cast to either an
unsigned lock, or a "struct rhash_head *" to be useful.
Variables of this type are most often called "bkt".

Previously "pprev" would sometimes point to a bucket, and sometimes a
->next pointer in an rhash_head.  As these are now different types,
pprev is NULL when it would have pointed to the bucket. In that case,
'blk' is used, together with correct locking protocol.

Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agorhashtable: allow rht_bucket_var to return NULL.
NeilBrown [Mon, 1 Apr 2019 23:07:45 +0000 (10:07 +1100)]
rhashtable: allow rht_bucket_var to return NULL.

Rather than returning a pointer to a static nulls, rht_bucket_var()
now returns NULL if the bucket doesn't exist.
This will make the next patch, which stores a bitlock in the
bucket pointer, somewhat cleaner.

This change involves introducing __rht_bucket_nested() which is
like rht_bucket_nested(), but doesn't provide the static nulls,
and changing rht_bucket_nested() to call this and possible
provide a static nulls - as is still needed for the non-var case.

Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agorhashtable: use cmpxchg() in nested_table_alloc()
NeilBrown [Mon, 1 Apr 2019 23:07:45 +0000 (10:07 +1100)]
rhashtable: use cmpxchg() in nested_table_alloc()

nested_table_alloc() relies on the fact that there is
at most one spinlock allocated for every slot in the top
level nested table, so it is not possible for two threads
to try to allocate the same table at the same time.

This assumption is a little fragile (it is not explicit) and is
unnecessary as cmpxchg() can be used instead.

A future patch will replace the spinlocks by per-bucket bitlocks,
and then we won't be able to protect the slot pointer with a spinlock.

So replace rcu_assign_pointer() with cmpxchg() - which has equivalent
barrier properties.
If it the cmp fails, free the table that was just allocated.

Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoMerge branch 'net-hsr-improvements-and-bug-fixes'
David S. Miller [Sun, 7 Apr 2019 01:32:21 +0000 (18:32 -0700)]
Merge branch 'net-hsr-improvements-and-bug-fixes'

Murali Karicheri says:

====================
net: hsr: improvements and bug fixes

This series has some coding style fixes and other bug fixes.
Patch 12/14, I have also done SPDX conversion. Not sure if
that is the only thing needed and is correct. So please pay
close attention to this patch before merge as I would like to
avoid any issue related to licensing applicable for this code.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: hsr: Fix node prune function for forget time expiry
Aaron Kramer [Fri, 5 Apr 2019 17:31:36 +0000 (13:31 -0400)]
net: hsr: Fix node prune function for forget time expiry

HSR should forget nodes after configured node forget time expiry based
on HSR_NODE_FORGET_TIME. As part of hsr_prune_nodes(), code checks to
see if entries are to be flushed out if not heard for longer than forget
time. But currently hsr_prune_nodes() is called only once during device
creation. Restart the timer at the end of hsr_prune_nodes() so that
hsr_prune_nodes() gets called periodically and forgotten entries are
removed from node table.

Signed-off-by: Aaron Kramer <a-kramer@ti.com>
Signed-off-by: Murali Karicheri <m-karicheri2@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: hsr: add debugfs support for display node list
Murali Karicheri [Fri, 5 Apr 2019 17:31:35 +0000 (13:31 -0400)]
net: hsr: add debugfs support for display node list

This adds a debugfs interface to allow display the nodes learned
by the hsr master.

Signed-off-by: Murali Karicheri <m-karicheri2@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: hsr: convert to SPDX identifier
Murali Karicheri [Fri, 5 Apr 2019 17:31:34 +0000 (13:31 -0400)]
net: hsr: convert to SPDX identifier

Use SPDX-License-Identifier instead of a verbose license text.

Signed-off-by: Murali Karicheri <m-karicheri2@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: hsr: add blank line after function declaration
Murali Karicheri [Fri, 5 Apr 2019 17:31:33 +0000 (13:31 -0400)]
net: hsr: add blank line after function declaration

Add a blank line after function declaration as suggested by
checkpatch.pl -f

Signed-off-by: Murali Karicheri <m-karicheri2@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: hsr: remove camel case usage in the code
Murali Karicheri [Fri, 5 Apr 2019 17:31:32 +0000 (13:31 -0400)]
net: hsr: remove camel case usage in the code

Current driver code uses camel case in many places. This is
seen when ran checkpatch.pl -f on files under net/hsr. This
patch fixes the code to remove camel case usage.

Signed-off-by: Murali Karicheri <m-karicheri2@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: hsr: add missing space around operator in code
Murali Karicheri [Fri, 5 Apr 2019 17:31:31 +0000 (13:31 -0400)]
net: hsr: add missing space around operator in code

This patch add missing space around operator in code. This is
seen when ran checkpatch.pl -f on files under net/hsr.

Signed-off-by: Murali Karicheri <m-karicheri2@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: hsr: fix placement of logical operator in a multi-line statement
Murali Karicheri [Fri, 5 Apr 2019 17:31:30 +0000 (13:31 -0400)]
net: hsr: fix placement of logical operator in a multi-line statement

In a multi-line statement exceeding 80 characters, logical operator
should be at the end of a line instead of being at the start. This
is seen when ran checkpatch.pl -f on files under net/hsr. The change
is per suggestion from checkpatch.

Signed-off-by: Murali Karicheri <m-karicheri2@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: hsr: remove unnecessary space after a cast
Murali Karicheri [Fri, 5 Apr 2019 17:31:29 +0000 (13:31 -0400)]
net: hsr: remove unnecessary space after a cast

This patch removes unnecessary space after a cast. This is seen
when ran checkpatch.pl -f on files under net/hsr.

Signed-off-by: Murali Karicheri <m-karicheri2@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: hsr: fix NULL checks in the code
Murali Karicheri [Fri, 5 Apr 2019 17:31:28 +0000 (13:31 -0400)]
net: hsr: fix NULL checks in the code

This patch replaces all instance of NULL checks such as
    if (foo == NULL) with if (!foo)
Also
    if (foo != NULL) with if (foo)

This is seen when ran checkpatch.pl -f on files under net/hsr
and suggestion is to replace as above.

Signed-off-by: Murali Karicheri <m-karicheri2@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: hsr: fix lines that ends with a '('
Murali Karicheri [Fri, 5 Apr 2019 17:31:27 +0000 (13:31 -0400)]
net: hsr: fix lines that ends with a '('

This patch fixes function calls that ends with '(' in a line.
This is seen when ran checkpatch.pl -f option on files under
net/hsr.

Signed-off-by: Murali Karicheri <m-karicheri2@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: hsr: fix alignment issues in the code for functions
Murali Karicheri [Fri, 5 Apr 2019 17:31:26 +0000 (13:31 -0400)]
net: hsr: fix alignment issues in the code for functions

This patch fixes alignment issues in code for functions. This is
seen when ran checkpatch.pl -f option on files under net/hsr.

Signed-off-by: Murali Karicheri <m-karicheri2@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: hsr: remove unnecessary paranthesis from the code
Murali Karicheri [Fri, 5 Apr 2019 17:31:25 +0000 (13:31 -0400)]
net: hsr: remove unnecessary paranthesis from the code

This patch fixes unnecessary paranthesis from the code. This is
seen when ran checkpatch.pl -f option on files under net/hsr.

Signed-off-by: Murali Karicheri <m-karicheri2@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: hsr: fix multiple blank lines in the code
Murali Karicheri [Fri, 5 Apr 2019 17:31:24 +0000 (13:31 -0400)]
net: hsr: fix multiple blank lines in the code

This patch fixes multiple blank lines in the code. This is seen
when ran checkpatch.pl -f option for files under net/hsr

Signed-off-by: Murali Karicheri <m-karicheri2@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: hsr: fix lines exceeding 80 characters
Murali Karicheri [Fri, 5 Apr 2019 17:31:23 +0000 (13:31 -0400)]
net: hsr: fix lines exceeding 80 characters

This patch fixes lines exceeding 80 characters. This is seen
when ran checkpatch.pl with -f option for files under
net/hsr.

Signed-off-by: Murali Karicheri <m-karicheri2@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoselftests: forwarding: test for bridge mcast traffic after report and leave
Nikolay Aleksandrov [Fri, 5 Apr 2019 15:40:47 +0000 (18:40 +0300)]
selftests: forwarding: test for bridge mcast traffic after report and leave

This test is split in two, the first part checks if a report creates a
corresponding mdb entry and if traffic is properly forwarded to it, and
the second part checks if the mdb entry is deleted after a leave and
if traffic is *not* forwarded to it. Since the mcast querier is enabled
we should see standard mcast snooping bridge behaviour.

Signed-off-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: hns: fix unsigned comparison to less than zero
Colin Ian King [Fri, 5 Apr 2019 13:59:16 +0000 (14:59 +0100)]
net: hns: fix unsigned comparison to less than zero

Currently mskid is unsigned and hence comparisons with negative
error return values are always false. Fix this by making mskid an
int.

Fixes: f058e46855dc ("net: hns: fix ICMP6 neighbor solicitation messages discard problem")
Addresses-Coverity: ("Operands don't affect result")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Reviewed-by: Mukesh Ojha <mojha@codeaurora.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoMerge branch 'r8152-runtime-mac-changes'
David S. Miller [Sun, 7 Apr 2019 01:21:46 +0000 (18:21 -0700)]
Merge branch 'r8152-runtime-mac-changes'

Mario Limonciello says:

====================
r8152: Support runtime changes of vendor mac passthu policy

On some platforms ACPI method `\\_SB.AMAC` is dynamic and changes to it can
influence changing the behavior of MAC pass through and what MAC address is used.

When running USB reset, re-read the MAC address to use to support tools that
change the policy.

This is quite similar to using `SIOCSIFHWADDR` except that the actual MAC to use
comes from ASL rather than from userspace.

Changes from v1:
 * Remove an extra unneeded `ether_addr_copy` call
 * Use `dev_set_mac_address` to ensure all notifiers are called
 * Shuffle functions to allow code re-use.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agor8152: Refresh MAC address during USBDEVFS_RESET
Mario Limonciello [Thu, 4 Apr 2019 18:46:53 +0000 (13:46 -0500)]
r8152: Refresh MAC address during USBDEVFS_RESET

On some platforms it is possible to dynamically change the policy
of what MAC address is selected from the ASL at runtime.

These tools will reset the USB device and expect the change to be
made immediately.

Signed-off-by: Mario Limonciello <mario.limonciello@dell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agor8152: remove extra action copying ethernet address
Mario Limonciello [Thu, 4 Apr 2019 18:46:52 +0000 (13:46 -0500)]
r8152: remove extra action copying ethernet address

This already happens later on in `rtl8152_set_mac_address`

Signed-off-by: Mario Limonciello <mario.limonciello@dell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agotcp: remove redundant check on tskb
Colin Ian King [Thu, 4 Apr 2019 14:46:03 +0000 (15:46 +0100)]
tcp: remove redundant check on tskb

The non-null check on tskb is always false because it is in an else
path of a check on tskb and hence tskb is null in this code block.
This is check is therefore redundant and can be removed as well
as the label coalesc.

if (tsbk) {
        ...
} else {
        ...
        if (unlikely(!skb)) {
                if (tskb)       /* can never be true, redundant code */
                        goto coalesc;
                return;
        }
}

Addresses-Coverity: ("Logically dead code")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Reviewed-by: Mukesh Ojha <mojha@codeaurora.org>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoMerge branch 'net-phy-add-Amlogic-g12a-support'
David S. Miller [Sun, 7 Apr 2019 01:16:58 +0000 (18:16 -0700)]
Merge branch 'net-phy-add-Amlogic-g12a-support'

Jerome Brunet says:

====================
net: phy: add Amlogic g12a support

This patchset adds the necessary bits to support network on the Amlogic
g12a SoC family.

Only the internal PHY and related MDIO mux needed to be addressed.
The GMAC remains compatible with axg SoC family

This series has been tested on the u200 (S905D2) with both the internal
and external (Realtek) PHYs.

Change since v2 [1]:
* Change 'clk part' Reviewed-by as suggested
* Remove default callback from phy drivers
* Use exact match PHY macros
* Default MDIO g12a as module if ARCH_MESON is enabled
* Don't print error on probe defer in the g12a mdio mux

Change since v1 [0]:
* drop '_' from function name unrelated to locking
* fix peripheral clock disable on error
* fix variable declaration reverse Xmas trees
* fix Kconfig dependency on CCF
  (Actually needed for 'struct clk_hw', Thx Andrew !)
* Minor fix in the DT exemple as reported by Rob

[0] https://lkml.kernel.org/r/20190314140135.19184-1-jbrunet@baylibre.com
[1] https://lkml.kernel.org/r/20190329141512.29867-1-jbrunet@baylibre.com
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: phy: meson-gxl: clean-up gxl variant driver
Jerome Brunet [Thu, 4 Apr 2019 13:11:47 +0000 (15:11 +0200)]
net: phy: meson-gxl: clean-up gxl variant driver

The purpose of this change is to align the gxl and g12a driver
declaration.

Like on the g12a variant, remove genphy_aneg_done() from the driver
declaration as the net phy framework will default to it anyway.

Also, the gxl phy id should be an exact match as well, so let's change
this and use the macro provided.

Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: phy: meson-gxl: add g12a support
Jerome Brunet [Thu, 4 Apr 2019 13:11:46 +0000 (15:11 +0200)]
net: phy: meson-gxl: add g12a support

The g12a SoC family uses the type of internal PHY that was used on the
gxl family. The quirks of gxl family, like the LPA register corruption,
appear to have been resolved on this new SoC generation.

Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: phy: add amlogic g12a mdio mux support
Jerome Brunet [Thu, 4 Apr 2019 13:11:45 +0000 (15:11 +0200)]
net: phy: add amlogic g12a mdio mux support

Add support for the mdio mux and internal phy glue of the g12a SoC family

Reviewed-by: Neil Armstrong <narmstrong@baylibre.com> # clk parts
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agodt-bindings: net: phy: add g12a mdio mux documentation
Jerome Brunet [Thu, 4 Apr 2019 13:11:44 +0000 (15:11 +0200)]
dt-bindings: net: phy: add g12a mdio mux documentation

Add documentation for the device tree bindings of the MDIO mux of Amlogic
g12a SoC family

Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Jerome Brunet <jbrunet@baylibre.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agovirtio-net: Fix some minor formatting errors
Yuval Shaia [Wed, 3 Apr 2019 09:10:13 +0000 (12:10 +0300)]
virtio-net: Fix some minor formatting errors

Signed-off-by: Yuval Shaia <yuval.shaia@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agovirtio-net: Remove inclusion of pci.h
Yuval Shaia [Wed, 3 Apr 2019 08:20:45 +0000 (11:20 +0300)]
virtio-net: Remove inclusion of pci.h

This header is not in use - remove it.

Signed-off-by: Yuval Shaia <yuval.shaia@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agor8169: disable tx interrupt coalescing on RTL8168
Heiner Kallweit [Sun, 31 Mar 2019 13:43:59 +0000 (15:43 +0200)]
r8169: disable tx interrupt coalescing on RTL8168

In contrast to switching rx irq coalescing off what fixed an issue,
switching tx irq coalescing off is merely a latency optimization,
therefore net-next. As part of this change:

- Remove INTT_0 .. INTT_3 constants, they aren't used.

- Remove comment in rtl_hw_start_8169(), we now have a detailed
  description by the code in rtl_set_coalesce().

- Due to switching irq coalescing off per default we don't need the
  initialization in rtl_hw_start_8168(). If ethtool is used to switch
  on coalescing then rtl_set_coalesce() will configure this register.

For the sake of completeness: This patch just changes the default.
Users still have the option to configure irq coalescing via ethtool.

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
David S. Miller [Fri, 5 Apr 2019 21:14:19 +0000 (14:14 -0700)]
Merge git://git./linux/kernel/git/davem/net

Minor comment merge conflict in mlx5.

Staging driver has a fixup due to the skb->xmit_more changes
in 'net-next', but was removed in 'net'.

Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet/mlx5e: Unify logic of MTU boundaries
Tariq Toukan [Tue, 22 Jan 2019 11:42:10 +0000 (13:42 +0200)]
net/mlx5e: Unify logic of MTU boundaries

Expose a new helper that wraps the logic for setting the
netdevice's MTU boundaries.
Use it for the different components (Eth, rep, IPoIB).

Set the netdevice min MTU to ETH_MIN_MTU, and the max according
to both the FW capability and the kernel definition.

Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5e: Obsolete param field holding a constant value
Tariq Toukan [Thu, 17 Jan 2019 13:58:09 +0000 (15:58 +0200)]
net/mlx5e: Obsolete param field holding a constant value

The LRO WQE size is a constant, obsolete the parameter field
that holds it.

Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Reviewed-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5: A write memory barrier is sufficient in EQ ci update
Tariq Toukan [Sun, 16 Dec 2018 15:20:31 +0000 (17:20 +0200)]
net/mlx5: A write memory barrier is sufficient in EQ ci update

Soften the memory barrier call of mb() by a sufficient wmb() in the
consumer index update of the event queues.

Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5e: Do not rewrite fields with the same match
Eli Britstein [Sun, 17 Mar 2019 18:01:48 +0000 (18:01 +0000)]
net/mlx5e: Do not rewrite fields with the same match

If we have a match for the same value of a rewrite field, there is no
point for the rewrite. In order to save rewrite actions, and avoid
entirely rewrite actions (if all rewrites are the same), ignore such
rewrite fields.

Signed-off-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5e: Offload TC e-switch rules with ingress VLAN device
Eli Britstein [Thu, 14 Mar 2019 07:50:12 +0000 (07:50 +0000)]
net/mlx5e: Offload TC e-switch rules with ingress VLAN device

Offload TC rule on a VLAN device by matching the VLAN properties
of the VLAN device and emulating vlan pop actions.

Signed-off-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5e: Offload TC e-switch rules with egress VLAN device
Eli Britstein [Sun, 16 Dec 2018 14:57:44 +0000 (16:57 +0200)]
net/mlx5e: Offload TC e-switch rules with egress VLAN device

Upon redirection to an uplink VLAN device, emulate vlan push actions
according to the VLAN properties of the VLAN device and redirect to
the uplink.

Signed-off-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5e: Allow VLAN rewrite of prio field with the same match
Eli Britstein [Thu, 14 Mar 2019 14:55:09 +0000 (14:55 +0000)]
net/mlx5e: Allow VLAN rewrite of prio field with the same match

Changing the prio field of the VLAN is not supported. With
commit 37410902874c ("net/mlx5e: Support VLAN modify action") zero
value indicated "no-change". Allow the vid rewrite if the prio match
is the same as the prio set value.

Fixes: 37410902874c ("net/mlx5e: Support VLAN modify action")
Signed-off-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5e: Deny VLAN rewrite if there is no VLAN header match
Eli Britstein [Thu, 14 Mar 2019 14:41:40 +0000 (14:41 +0000)]
net/mlx5e: Deny VLAN rewrite if there is no VLAN header match

Rewrite of the packet in the VLAN offset may corrupt the packet if it's
not VLAN tagged. Deny the rewrite in this case.

Fixes: 37410902874c ("net/mlx5e: Support VLAN modify action")
Signed-off-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5e: Use helpers to get headers criteria and value pointers
Eli Britstein [Tue, 19 Mar 2019 07:04:56 +0000 (07:04 +0000)]
net/mlx5e: Use helpers to get headers criteria and value pointers

The headers criteria and value pointers may be either of the inner
packet, if a tunnel exists, or of the outer. Simplify the code by using
helper functions to retrieve them.

Signed-off-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5e: Return -EOPNOTSUPP when attempting to offload an unsupported action
Tonghao Zhang [Wed, 27 Feb 2019 15:31:19 +0000 (07:31 -0800)]
net/mlx5e: Return -EOPNOTSUPP when attempting to offload an unsupported action

* Now the encapsulation is not supported for mlx5 VFs. When we try to
offload that action, the -EINVAL is returned, but not -EOPNOTSUPP.
This patch changes the returned value and ignore to confuse user.
The command is shown as below [1].

* When max modify header action is zero, we return -EOPNOTSUPP
directly. In this way, we can ignore wrong message info (e.g.
"mlx5: parsed 0 pedit actions, can't do more"). This happens when
offloading pedit actions on mlx(cx4) VFs. The command is shown as below [2].

For example: (p2p1_0 is VF net device)
[1]
$ tc filter add dev p2p1_0 protocol ip  parent ffff: prio 1 flower skip_sw \
    src_mac e4:11:22:33:44:01    \
    action tunnel_key set        \
    src_ip 1.1.1.100        \
    dst_ip 1.1.1.200        \
    dst_port 4789 id 100        \
    action mirred egress redirect dev vxlan0

[2]
$ tc filter add dev p2p1_0 parent ffff: protocol ip prio 1 \
    flower skip_sw dst_mac 00:10:56:fb:64:e8 \
    dst_ip 1.1.1.100 src_ip 1.1.1.200 \
    action pedit ex munge eth src set 00:10:56:b4:5d:20

Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5e: Deletes unnecessary setting of esw_attr->parse_attr
Tonghao Zhang [Wed, 27 Feb 2019 15:31:18 +0000 (07:31 -0800)]
net/mlx5e: Deletes unnecessary setting of esw_attr->parse_attr

This patch deletes unnecessary setting of the esw_attr->parse_attr
to parse_attr in parse_tc_fdb_actions() because it is already done
by the mlx5e_flow_esw_attr_init() function.

Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5e: Remove 'parse_attr' argument in parse_tc_fdb_actions()
Tonghao Zhang [Wed, 27 Feb 2019 15:31:17 +0000 (07:31 -0800)]
net/mlx5e: Remove 'parse_attr' argument in parse_tc_fdb_actions()

This patch is a little improvement. Simplify the parse_tc_fdb_actions().

Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5e: Make the log friendly when decapsulation offload not supported
Tonghao Zhang [Wed, 27 Feb 2019 15:31:16 +0000 (07:31 -0800)]
net/mlx5e: Make the log friendly when decapsulation offload not supported

If we try to offload decapsulation actions to VFs hw, we get the log [1].
It's not friendly, because the kind of net device is null, and we don't
know what '0' means.

[1] "mlx5_core 0000:05:01.2 vf_0: decapsulation offload is not supported for  net device (0)"

Signed-off-by: Tonghao Zhang <xiangxia.m.yue@gmail.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5: E-Switch, Fix double mutex initialization
Max Gurtovoy [Wed, 27 Feb 2019 14:10:16 +0000 (16:10 +0200)]
net/mlx5: E-Switch, Fix double mutex initialization

Delete mutex_init call of a lock that's initialized in inner function.

Fixes: eca8cc389535 ("net/mlx5: E-Switch, Refactor offloads flow steering init/cleanup")
Signed-off-by: Max Gurtovoy <maxg@mellanox.com>
Reviewed-by: Roi Dayan <roid@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5: Handle event of power detection in the PCIE slot
Aya Levin [Sun, 24 Mar 2019 07:21:40 +0000 (09:21 +0200)]
net/mlx5: Handle event of power detection in the PCIE slot

Handle event of power state change in the PCIE slot. When the event
occurs, check if query power state and PCI power fields is supported. If
so, read these fields from MPEIN (management PCIE info) register and
issue a corresponding message.

Signed-off-by: Aya Levin <ayal@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agoMerge branch 'mlx5-next' of git://git.kernel.org/pub/scm/linux/kernel/git/mellanox...
Saeed Mahameed [Tue, 2 Apr 2019 22:43:45 +0000 (15:43 -0700)]
Merge branch 'mlx5-next' of git://git./linux/kernel/git/mellanox/linux

This merge commit includes some misc shared code updates from mlx5-next branch needed
for net-next.

1) From Maxim, Remove un-used macros and spinlock from mlx5 code.

2) From Aya, Expose Management PCIE info register layout and add rate limit
print macros.

3) From Tariq, Compilation warning fix in fs_core.c

4) From Vu, Huy and Saeed, Improve mlx5 initialization flow:
The goal is to provide a better logical separation of mlx5 core
device initialization flow and will help to seamlessly support
creating different mlx5 device types such as PF, VF and SF
mlx5 sub-function virtual devices.

Mlx5_core driver needs to separate HCA resources from pci resources.
Its initialize/load/unload will be broken into stages:
1. Initialize common data structures
2. Setup function which initializes pci resources (for PF/VF)
   or some other specific resources for virtual device
3. Initialize software objects according to hardware capabilities
4. Load all mlx5_core components

It is also necessary to detach mlx5_core mdev name/message from pci
device mdev->pdev name/message for a clearer report/debug of
different mlx5 device types.

Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agoMerge tag 'mm-compaction-5.1-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git...
Linus Torvalds [Fri, 5 Apr 2019 16:09:53 +0000 (06:09 -1000)]
Merge tag 'mm-compaction-5.1-rc4' of git://git./linux/kernel/git/mel/linux

Pull mm/compaction fixes from Mel Gorman:
 "The merge window for 5.1 introduced a number of compaction-related
  patches. with intermittent reports of corruption and functional
  issues. The bugs are due to sloopy checking of zone boundaries and a
  corner case where invalid indexes are used to access the free lists.

  Reports are not common but at least two users and 0-day have tripped
  over them. There is a chance that one of the syzbot reports are
  related but it has not been confirmed properly.

  The normal submission path is with Andrew but there have been some
  delays and I consider them urgent enough that they should be picked up
  before RC4 to avoid duplicate reports.

  All of these have been successfully tested on older RC windows. This
  will make this branch look like a rebase but in fact, they've simply
  been lifted again from Andrew's tree and placed on a fresh branch.
  I've no reason to believe that this has invalidated the testing given
  the lack of change in compaction and the nature of the fixes"

* tag 'mm-compaction-5.1-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/mel/linux:
  mm/compaction.c: abort search if isolation fails
  mm/compaction.c: correct zone boundary handling when resetting pageblock skip hints

5 years agotty: mark Siemens R3964 line discipline as BROKEN
Greg Kroah-Hartman [Fri, 5 Apr 2019 13:39:26 +0000 (15:39 +0200)]
tty: mark Siemens R3964 line discipline as BROKEN

The n_r3964 line discipline driver was written in a different time, when
SMP machines were rare, and users were trusted to do the right thing.
Since then, the world has moved on but not this code, it has stayed
rooted in the past with its lovely hand-crafted list structures and
loads of "interesting" race conditions all over the place.

After attempting to clean up most of the issues, I just gave up and am
now marking the driver as BROKEN so that hopefully someone who has this
hardware will show up out of the woodwork (I know you are out there!)
and will help with debugging a raft of changes that I had laying around
for the code, but was too afraid to commit as odds are they would break
things.

Many thanks to Jann and Linus for pointing out the initial problems in
this codebase, as well as many reviews of my attempts to fix the issues.
It was a case of whack-a-mole, and as you can see, the mole won.

Reported-by: Jann Horn <jannh@google.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
5 years agoMerge tag 'drm-fixes-2019-04-05' of git://anongit.freedesktop.org/drm/drm
Linus Torvalds [Fri, 5 Apr 2019 04:22:55 +0000 (18:22 -1000)]
Merge tag 'drm-fixes-2019-04-05' of git://anongit.freedesktop.org/drm/drm

Pull drm fixes from Dave Airlie:
 "Pretty quiet week, just some amdgpu and i915 fixes.

  i915:
   - deadlock fix
   - gvt fixes

  amdgpu:
   - PCIE dpm feature fix
   - Powerplay fixes"

* tag 'drm-fixes-2019-04-05' of git://anongit.freedesktop.org/drm/drm:
  drm/i915/gvt: Fix kerneldoc typo for intel_vgpu_emulate_hotplug
  drm/i915/gvt: Correct the calculation of plane size
  drm/amdgpu: remove unnecessary rlc reset function on gfx9
  drm/i915: Always backoff after a drm_modeset_lock() deadlock
  drm/i915/gvt: do not let pin count of shadow mm go negative
  drm/i915/gvt: do not deliver a workload if its creation fails
  drm/amd/display: VBIOS can't be light up HDMI when restart system
  drm/amd/powerplay: fix possible hang with 3+ 4K monitors
  drm/amd/powerplay: correct data type to avoid overflow
  drm/amd/powerplay: add ECC feature bit
  drm/amd/amdgpu: fix PCIe dpm feature issue (v3)

5 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
Linus Torvalds [Fri, 5 Apr 2019 04:07:12 +0000 (18:07 -1000)]
Merge git://git./linux/kernel/git/davem/net

Pull networking fixes from David Miller:

 1) Several hash table refcount fixes in batman-adv, from Sven
    Eckelmann.

 2) Use after free in bpf_evict_inode(), from Daniel Borkmann.

 3) Fix mdio bus registration in ixgbe, from Ivan Vecera.

 4) Unbounded loop in __skb_try_recv_datagram(), from Paolo Abeni.

 5) ila rhashtable corruption fix from Herbert Xu.

 6) Don't allow upper-devices to be added to vrf devices, from Sabrina
    Dubroca.

 7) Add qmi_wwan device ID for Olicard 600, from Bjørn Mork.

 8) Don't leave skb->next poisoned in __netif_receive_skb_list_ptype,
    from Alexander Lobakin.

 9) Missing IDR checks in mlx5 driver, from Aditya Pakki.

10) Fix false connection termination in ktls, from Jakub Kicinski.

11) Work around some ASPM issues with r8169 by disabling rx interrupt
    coalescing on certain chips. From Heiner Kallweit.

12) Properly use per-cpu qstat values on NOLOCK qdiscs, from Paolo
    Abeni.

13) Fully initialize sockaddr_in structures in SCTP, from Xin Long.

14) Various BPF flow dissector fixes from Stanislav Fomichev.

15) Divide by zero in act_sample, from Davide Caratti.

16) Fix bridging multicast regression introduced by rhashtable
    conversion, from Nikolay Aleksandrov.

* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (106 commits)
  ibmvnic: Fix completion structure initialization
  ipv6: sit: reset ip header pointer in ipip6_rcv
  net: bridge: always clear mcast matching struct on reports and leaves
  libcxgb: fix incorrect ppmax calculation
  vlan: conditional inclusion of FCoE hooks to match netdevice.h and bnx2x
  sch_cake: Make sure we can write the IP header before changing DSCP bits
  sch_cake: Use tc_skb_protocol() helper for getting packet protocol
  tcp: Ensure DCTCP reacts to losses
  net/sched: act_sample: fix divide by zero in the traffic path
  net: thunderx: fix NULL pointer dereference in nicvf_open/nicvf_stop
  net: hns: Fix sparse: some warnings in HNS drivers
  net: hns: Fix WARNING when remove HNS driver with SMMU enabled
  net: hns: fix ICMP6 neighbor solicitation messages discard problem
  net: hns: Fix probabilistic memory overwrite when HNS driver initialized
  net: hns: Use NAPI_POLL_WEIGHT for hns driver
  net: hns: fix KASAN: use-after-free in hns_nic_net_xmit_hw()
  flow_dissector: rst'ify documentation
  ipv6: Fix dangling pointer when ipv6 fragment
  net-gro: Fix GRO flush when receiving a GSO packet.
  flow_dissector: document BPF flow dissector environment
  ...

5 years agoibmvnic: Fix completion structure initialization
Thomas Falcon [Thu, 4 Apr 2019 23:58:26 +0000 (18:58 -0500)]
ibmvnic: Fix completion structure initialization

Fix device initialization completion handling for vNIC adapters.
Initialize the completion structure on probe and reinitialize when needed.
This also fixes a race condition during kdump where the driver can attempt
to access the completion struct before it is initialized:

Unable to handle kernel paging request for data at address 0x00000000
Faulting instruction address: 0xc0000000081acbe0
Oops: Kernel access of bad area, sig: 11 [#1]
LE SMP NR_CPUS=2048 NUMA pSeries
Modules linked in: ibmvnic(+) ibmveth sunrpc overlay squashfs loop
CPU: 19 PID: 301 Comm: systemd-udevd Not tainted 4.18.0-64.el8.ppc64le #1
NIP:  c0000000081acbe0 LR: c0000000081ad964 CTR: c0000000081ad900
REGS: c000000027f3f990 TRAP: 0300   Not tainted  (4.18.0-64.el8.ppc64le)
MSR:  800000010280b033 <SF,VEC,VSX,EE,FP,ME,IR,DR,RI,LE,TM[E]> CR: 28228288  XER: 00000006
CFAR: c000000008008934 DAR: 0000000000000000 DSISR: 40000000 IRQMASK: 1
GPR00: c0000000081ad964 c000000027f3fc10 c0000000095b5800 c0000000221b4e58
GPR04: 0000000000000003 0000000000000001 000049a086918581 00000000000000d4
GPR08: 0000000000000007 0000000000000000 ffffffffffffffe8 d0000000014dde28
GPR12: c0000000081ad900 c000000009a00c00 0000000000000001 0000000000000100
GPR16: 0000000000000038 0000000000000007 c0000000095e2230 0000000000000006
GPR20: 0000000000400140 0000000000000001 c00000000910c880 0000000000000000
GPR24: 0000000000000000 0000000000000006 0000000000000000 0000000000000003
GPR28: 0000000000000001 0000000000000001 c0000000221b4e60 c0000000221b4e58
NIP [c0000000081acbe0] __wake_up_locked+0x50/0x100
LR [c0000000081ad964] complete+0x64/0xa0
Call Trace:
[c000000027f3fc10] [c000000027f3fc60] 0xc000000027f3fc60 (unreliable)
[c000000027f3fc60] [c0000000081ad964] complete+0x64/0xa0
[c000000027f3fca0] [d0000000014dad58] ibmvnic_handle_crq+0xce0/0x1160 [ibmvnic]
[c000000027f3fd50] [d0000000014db270] ibmvnic_tasklet+0x98/0x130 [ibmvnic]
[c000000027f3fda0] [c00000000813f334] tasklet_action_common.isra.3+0xc4/0x1a0
[c000000027f3fe00] [c000000008cd13f4] __do_softirq+0x164/0x400
[c000000027f3fef0] [c00000000813ed64] irq_exit+0x184/0x1c0
[c000000027f3ff20] [c0000000080188e8] __do_irq+0xb8/0x210
[c000000027f3ff90] [c00000000802d0a4] call_do_irq+0x14/0x24
[c000000026a5b010] [c000000008018adc] do_IRQ+0x9c/0x130
[c000000026a5b060] [c000000008008ce4] hardware_interrupt_common+0x114/0x120

Signed-off-by: Thomas Falcon <tlfalcon@linux.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoipv6: sit: reset ip header pointer in ipip6_rcv
Lorenzo Bianconi [Thu, 4 Apr 2019 14:37:53 +0000 (16:37 +0200)]
ipv6: sit: reset ip header pointer in ipip6_rcv

ipip6 tunnels run iptunnel_pull_header on received skbs. This can
determine the following use-after-free accessing iph pointer since
the packet will be 'uncloned' running pskb_expand_head if it is a
cloned gso skb (e.g if the packet has been sent though a veth device)

[  706.369655] BUG: KASAN: use-after-free in ipip6_rcv+0x1678/0x16e0 [sit]
[  706.449056] Read of size 1 at addr ffffe01b6bd855f5 by task ksoftirqd/1/=
[  706.669494] Hardware name: HPE ProLiant m400 Server/ProLiant m400 Server, BIOS U02 08/19/2016
[  706.771839] Call trace:
[  706.801159]  dump_backtrace+0x0/0x2f8
[  706.845079]  show_stack+0x24/0x30
[  706.884833]  dump_stack+0xe0/0x11c
[  706.925629]  print_address_description+0x68/0x260
[  706.982070]  kasan_report+0x178/0x340
[  707.025995]  __asan_report_load1_noabort+0x30/0x40
[  707.083481]  ipip6_rcv+0x1678/0x16e0 [sit]
[  707.132623]  tunnel64_rcv+0xd4/0x200 [tunnel4]
[  707.185940]  ip_local_deliver_finish+0x3b8/0x988
[  707.241338]  ip_local_deliver+0x144/0x470
[  707.289436]  ip_rcv_finish+0x43c/0x14b0
[  707.335447]  ip_rcv+0x628/0x1138
[  707.374151]  __netif_receive_skb_core+0x1670/0x2600
[  707.432680]  __netif_receive_skb+0x28/0x190
[  707.482859]  process_backlog+0x1d0/0x610
[  707.529913]  net_rx_action+0x37c/0xf68
[  707.574882]  __do_softirq+0x288/0x1018
[  707.619852]  run_ksoftirqd+0x70/0xa8
[  707.662734]  smpboot_thread_fn+0x3a4/0x9e8
[  707.711875]  kthread+0x2c8/0x350
[  707.750583]  ret_from_fork+0x10/0x18

[  707.811302] Allocated by task 16982:
[  707.854182]  kasan_kmalloc.part.1+0x40/0x108
[  707.905405]  kasan_kmalloc+0xb4/0xc8
[  707.948291]  kasan_slab_alloc+0x14/0x20
[  707.994309]  __kmalloc_node_track_caller+0x158/0x5e0
[  708.053902]  __kmalloc_reserve.isra.8+0x54/0xe0
[  708.108280]  __alloc_skb+0xd8/0x400
[  708.150139]  sk_stream_alloc_skb+0xa4/0x638
[  708.200346]  tcp_sendmsg_locked+0x818/0x2b90
[  708.251581]  tcp_sendmsg+0x40/0x60
[  708.292376]  inet_sendmsg+0xf0/0x520
[  708.335259]  sock_sendmsg+0xac/0xf8
[  708.377096]  sock_write_iter+0x1c0/0x2c0
[  708.424154]  new_sync_write+0x358/0x4a8
[  708.470162]  __vfs_write+0xc4/0xf8
[  708.510950]  vfs_write+0x12c/0x3d0
[  708.551739]  ksys_write+0xcc/0x178
[  708.592533]  __arm64_sys_write+0x70/0xa0
[  708.639593]  el0_svc_handler+0x13c/0x298
[  708.686646]  el0_svc+0x8/0xc

[  708.739019] Freed by task 17:
[  708.774597]  __kasan_slab_free+0x114/0x228
[  708.823736]  kasan_slab_free+0x10/0x18
[  708.868703]  kfree+0x100/0x3d8
[  708.905320]  skb_free_head+0x7c/0x98
[  708.948204]  skb_release_data+0x320/0x490
[  708.996301]  pskb_expand_head+0x60c/0x970
[  709.044399]  __iptunnel_pull_header+0x3b8/0x5d0
[  709.098770]  ipip6_rcv+0x41c/0x16e0 [sit]
[  709.146873]  tunnel64_rcv+0xd4/0x200 [tunnel4]
[  709.200195]  ip_local_deliver_finish+0x3b8/0x988
[  709.255596]  ip_local_deliver+0x144/0x470
[  709.303692]  ip_rcv_finish+0x43c/0x14b0
[  709.349705]  ip_rcv+0x628/0x1138
[  709.388413]  __netif_receive_skb_core+0x1670/0x2600
[  709.446943]  __netif_receive_skb+0x28/0x190
[  709.497120]  process_backlog+0x1d0/0x610
[  709.544169]  net_rx_action+0x37c/0xf68
[  709.589131]  __do_softirq+0x288/0x1018

[  709.651938] The buggy address belongs to the object at ffffe01b6bd85580
                which belongs to the cache kmalloc-1024 of size 1024
[  709.804356] The buggy address is located 117 bytes inside of
                1024-byte region [ffffe01b6bd85580ffffe01b6bd85980)
[  709.946340] The buggy address belongs to the page:
[  710.003824] page:ffff7ff806daf600 count:1 mapcount:0 mapping:ffffe01c4001f600 index:0x0
[  710.099914] flags: 0xfffff8000000100(slab)
[  710.149059] raw: 0fffff8000000100 dead000000000100 dead000000000200 ffffe01c4001f600
[  710.242011] raw: 0000000000000000 0000000000380038 00000001ffffffff 0000000000000000
[  710.334966] page dumped because: kasan: bad access detected

Fix it resetting iph pointer after iptunnel_pull_header

Fixes: a09a4c8dd1ec ("tunnels: Remove encapsulation offloads on decap")
Tested-by: Jianlin Shi <jishi@redhat.com>
Signed-off-by: Lorenzo Bianconi <lorenzo.bianconi@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoMerge branch 'tipc-improve-TIPC-unicast-link-throughput'
David S. Miller [Fri, 5 Apr 2019 01:29:25 +0000 (18:29 -0700)]
Merge branch 'tipc-improve-TIPC-unicast-link-throughput'

Tuong Lien says:

====================
tipc: improve TIPC unicast link throughput

The series introduces an algorithm to improve TIPC throughput especially
in terms of packet loss, also tries to reduce packet duplication due to
overactive NACK sending mechanism.

The link failover situation is also covered by the patches.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agotipc: adapt link failover for new Gap-ACK algorithm
Tuong Lien [Thu, 4 Apr 2019 04:09:53 +0000 (11:09 +0700)]
tipc: adapt link failover for new Gap-ACK algorithm

In commit 0ae955e2656d ("tipc: improve TIPC throughput by Gap ACK
blocks"), we enhance the link transmq by releasing as many packets as
possible with the multi-ACKs from peer node. This also means the queue
is now non-linear and the peer link deferdq becomes vital.

Whereas, in the case of link failover, all messages in the link transmq
need to be transmitted as tunnel messages in such a way that message
sequentiality and cardinality per sender is preserved. This requires us
to maintain the link deferdq somehow, so that when the tunnel messages
arrive, the inner user messages along with the ones in the deferdq will
be delivered to upper layer correctly.

The commit accomplishes this by defining a new queue in the TIPC link
structure to hold the old link deferdq when link failover happens and
process it upon receipt of tunnel messages.

Also, in the case of link syncing, the link deferdq will not be purged
to avoid unnecessary retransmissions that in the worst case will fail
because the packets might have been freed on the sending side.

Acked-by: Ying Xue <ying.xue@windriver.com>
Acked-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Tuong Lien <tuong.t.lien@dektech.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agotipc: reduce duplicate packets for unicast traffic
Tuong Lien [Thu, 4 Apr 2019 04:09:52 +0000 (11:09 +0700)]
tipc: reduce duplicate packets for unicast traffic

For unicast transmission, the current NACK sending althorithm is over-
active that forces the sending side to retransmit a packet that is not
really lost but just arrived at the receiving side with some delay, or
even retransmit same packets that have already been retransmitted
before. As a result, many duplicates are observed also under normal
condition, ie. without packet loss.

One example case is: node1 transmits 1 2 3 4 10 5 6 7 8 9, when node2
receives packet #10, it puts into the deferdq. When the packet #5 comes
it sends NACK with gap [6 - 9]. However, shortly after that, when
packet #6 arrives, it pulls out packet #10 from the deferfq, but it is
still out of order, so it makes another NACK with gap [7 - 9] and so on
... Finally, node1 has to retransmit the packets 5 6 7 8 9 a number of
times, but in fact all the packets are not lost at all, so duplicates!

This commit reduces duplicates by changing the condition to send NACK,
also restricting the retransmissions on individual packets via a timer
of about 1ms. However, it also needs to say that too tricky condition
for NACKs or too long timeout value for retransmissions will result in
performance reducing! The criterias in this commit are found to be
effective for both the requirements to reduce duplicates but not affect
performance.

The tipc_link_rcv() is also improved to only dequeue skb from the link
deferdq if it is expected (ie. its seqno <= rcv_nxt).

Acked-by: Ying Xue <ying.xue@windriver.com>
Acked-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Tuong Lien <tuong.t.lien@dektech.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agotipc: improve TIPC throughput by Gap ACK blocks
Tuong Lien [Thu, 4 Apr 2019 04:09:51 +0000 (11:09 +0700)]
tipc: improve TIPC throughput by Gap ACK blocks

During unicast link transmission, it's observed very often that because
of one or a few lost/dis-ordered packets, the sending side will fastly
reach the send window limit and must wait for the packets to be arrived
at the receiving side or in the worst case, a retransmission must be
done first. The sending side cannot release a lot of subsequent packets
in its transmq even though all of them might have already been received
by the receiving side.
That is, one or two packets dis-ordered/lost and dozens of packets have
to wait, this obviously reduces the overall throughput!

This commit introduces an algorithm to overcome this by using "Gap ACK
blocks". Basically, a Gap ACK block will consist of <ack, gap> numbers
that describes the link deferdq where packets have been got by the
receiving side but with gaps, for example:

      link deferdq: [1 2 3 4      10 11      13 14 15       20]
--> Gap ACK blocks:       <4, 5>,   <11, 1>,      <15, 4>, <20, 0>

The Gap ACK blocks will be sent to the sending side along with the
traditional ACK or NACK message. Immediately when receiving the message
the sending side will now not only release from its transmq the packets
ack-ed by the ACK but also by the Gap ACK blocks! So, more packets can
be enqueued and transmitted.
In addition, the sending side can now do "multi-retransmissions"
according to the Gaps reported in the Gap ACK blocks.

The new algorithm as verified helps greatly improve the TIPC throughput
especially under packet loss condition.

So far, a maximum of 32 blocks is quite enough without any "Too few Gap
ACK blocks" reports with a 5.0% packet loss rate, however this number
can be increased in the furture if needed.

Also, the patch is backward compatible.

Acked-by: Ying Xue <ying.xue@windriver.com>
Acked-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: Tuong Lien <tuong.t.lien@dektech.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoMerge tag 'riscv-for-linus-5.1-rc4' of git://git.kernel.org/pub/scm/linux/kernel...
Linus Torvalds [Fri, 5 Apr 2019 01:04:00 +0000 (15:04 -1000)]
Merge tag 'riscv-for-linus-5.1-rc4' of git://git./linux/kernel/git/palmer/riscv-linux

Pull RISC-V fixes from Palmer Dabbelt:
 "I dropped the ball a bit here: these patches should all probably have
  been part of rc2, but I wanted to get around to properly testing them
  in the various configurations (qemu32, qeum64, unleashed) first.

  Unfortunately I've been traveling and didn't have time to actually do
  that, but since these fix concrete bugs and pass my old set of tests I
  don't want to delay the fixes any longer.

  There are four independent fixes here:

   - A fix for the rv32 port that corrects the 64-bit user accesor's
     fixup label address.

   - A fix for a regression introduced during the merge window that
     broke medlow configurations at run time. This patch also includes a
     fix that disables ftrace for the same set of functions, which was
     found by inspection at the same time.

   - A modification of the memory map to avoid overlapping the FIXMAP
     and VMALLOC regions on systems with small memory maps.

   - A fix to the module handling code to use the correct syntax for
     probing Kconfig entries.

  These have passed my standard test flow, but I didn't have time to
  expand that testing like I said I would"

* tag 'riscv-for-linus-5.1-rc4' of git://git.kernel.org/pub/scm/linux/kernel/git/palmer/riscv-linux:
  RISC-V: Use IS_ENABLED(CONFIG_CMODEL_MEDLOW)
  RISC-V: Fix FIXMAP_TOP to avoid overlap with VMALLOC area
  RISC-V: Always compile mm/init.c with cmodel=medany and notrace
  riscv: fix accessing 8-byte variable from RV32