platform/kernel/linux-starfive.git
5 years agomlxsw: spectrum_router: Generalize mlxsw_sp_netdevice_router_port_event()
Petr Machata [Thu, 13 Dec 2018 11:54:48 +0000 (11:54 +0000)]
mlxsw: spectrum_router: Generalize mlxsw_sp_netdevice_router_port_event()

Prepare mlxsw_sp_netdevice_router_port_event() for handling of
NETDEV_PRE_CHANGEADDR. Split out the part that deals with the actual
changes and call it for the two events currently handled.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: bonding: Issue NETDEV_PRE_CHANGEADDR
Petr Machata [Thu, 13 Dec 2018 11:54:46 +0000 (11:54 +0000)]
net: bonding: Issue NETDEV_PRE_CHANGEADDR

Give interested parties an opportunity to veto an impending HW address
change.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: bonding: Give bond_set_dev_addr() a return value
Petr Machata [Thu, 13 Dec 2018 11:54:44 +0000 (11:54 +0000)]
net: bonding: Give bond_set_dev_addr() a return value

Before NETDEV_CHANGEADDR, bond driver should emit NETDEV_PRE_CHANGEADDR,
and allow consumers to veto the address change. To propagate further the
return code from NETDEV_PRE_CHANGEADDR, give the function that
implements address change a return value.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: ipvlan: Issue NETDEV_PRE_CHANGEADDR
Petr Machata [Thu, 13 Dec 2018 11:54:41 +0000 (11:54 +0000)]
net: ipvlan: Issue NETDEV_PRE_CHANGEADDR

A NETDEV_CHANGEADDR event implies a change of address of each of the
IPVLANs of this IPVLAN device. Therefore propagate NETDEV_PRE_CHANGEADDR
to all the IPVLANs.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: bridge: Handle NETDEV_PRE_CHANGEADDR from ports
Petr Machata [Thu, 13 Dec 2018 11:54:39 +0000 (11:54 +0000)]
net: bridge: Handle NETDEV_PRE_CHANGEADDR from ports

When a port device seeks approval of a potential new MAC address, make
sure that should the bridge device end up using this address, all
interested parties would agree with it.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: bridge: Issue NETDEV_PRE_CHANGEADDR
Petr Machata [Thu, 13 Dec 2018 11:54:37 +0000 (11:54 +0000)]
net: bridge: Issue NETDEV_PRE_CHANGEADDR

When a port is attached to a bridge, the address of the bridge in
question may change as well. Even if it would not change at this
point (because the current bridge address is lower), it might end up
changing later as a result of detach of another port, which can't be
vetoed.

Therefore issue NETDEV_PRE_CHANGEADDR regardless of whether the address
will be used at this point or not, and make sure all involved parties
would agree with the change.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: dev: Issue NETDEV_PRE_CHANGEADDR
Petr Machata [Thu, 13 Dec 2018 11:54:35 +0000 (11:54 +0000)]
net: dev: Issue NETDEV_PRE_CHANGEADDR

When a device address is about to be changed, or an address added to the
list of device HW addresses, it is necessary to ensure that all
interested parties can support the address. Therefore, send the
NETDEV_PRE_CHANGEADDR notification, and if anyone bails on it, do not
change the address.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: dev: Add NETDEV_PRE_CHANGEADDR
Petr Machata [Thu, 13 Dec 2018 11:54:33 +0000 (11:54 +0000)]
net: dev: Add NETDEV_PRE_CHANGEADDR

The NETDEV_CHANGEADDR notification is emitted after a device address
changes. Extending this message to allow vetoing is certainly possible,
but several other notification types have instead adopted a simple
two-stage approach: first a "pre" notification is sent to make sure all
interested parties are OK with a change that's about to be done. Then
the change is done, and afterwards a "post" notification is sent.

This dual approach is easier to use: when the change is vetoed, nothing
has changed yet, and it's therefore unnecessary to roll anything back.
Therefore adopt it for NETDEV_CHANGEADDR as well.

To that end, add NETDEV_PRE_CHANGEADDR and an info structure to go along
with it.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: dev: Add extack argument to dev_set_mac_address()
Petr Machata [Thu, 13 Dec 2018 11:54:30 +0000 (11:54 +0000)]
net: dev: Add extack argument to dev_set_mac_address()

A follow-up patch will add a notifier type NETDEV_PRE_CHANGEADDR, which
allows vetoing of MAC address changes. One prominent path to that
notification is through dev_set_mac_address(). Therefore give this
function an extack argument, so that it can be packed together with the
notification. Thus a textual reason for rejection (or a warning) can be
communicated back to the user.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoMerge tag 'mlx5e-updates-2018-12-11' of git://git.kernel.org/pub/scm/linux/kernel...
David S. Miller [Thu, 13 Dec 2018 06:22:53 +0000 (22:22 -0800)]
Merge tag 'mlx5e-updates-2018-12-11' of git://git./linux/kernel/git/saeed/linux

Saeed Mahameed says:

====================
mlx5e-updates-2018-12-11

From Eli Britstein,
Patches 1-10 adds remote mirroring support.
Patches 1-4 refactor encap related code as pre-steps for using per
destination encapsulation properties.
Patches 5-7 use extended destination feature for single/multi
destination scenarios that have a single encap destination.
Patches 8-10 enable multiple encap destinations for a TC flow.

From, Daniel Jurgens,
Patch 11, Use CQE padding for Ethernet CQs, PPC showed up to a 24%
improvement in small packet throughput

From Eyal Davidovich,
patches 12-14, FW monitor counter support
FW monitor counters feature came to solve the delayed reporting of
FW stats in the atomic get_stats64 ndo, since we can't access the
FW at that stage, this feature will enable immediate FW stats updates
in the driver via fw events on specific stats updates.

Patch 12, cleanup to avoid querying a FW counter when it is not
supported
Patch 13, Monitor counters FW commands support
Patch 14, Use monitor counters in ethernet netdevice to update FW
stats reported in the atomic get_stats64 ndo.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet-next: stmmac: dwmac-mediatek: add module license info
Biao Huang [Thu, 13 Dec 2018 02:41:37 +0000 (10:41 +0800)]
net-next: stmmac: dwmac-mediatek: add module license info

Add MODULE_LICENSE info to fix this:
WARNING: modpost: missing MODULE_LICENSE() in drivers/net/ethernet/stmicro/stmmac/dwmac-mediatek.o

Signed-off-by: Biao Huang <biao.huang@mediatek.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet/mlx5e: Remove set but not used variable 'upriv'
YueHaibing [Wed, 12 Dec 2018 08:33:53 +0000 (08:33 +0000)]
net/mlx5e: Remove set but not used variable 'upriv'

Fixes gcc '-Wunused-but-set-variable' warning:

drivers/net/ethernet/mellanox/mlx5/core/en_rep.c: In function 'mlx5e_vport_rep_load':
drivers/net/ethernet/mellanox/mlx5/core/en_rep.c:1490:21: warning:
 variable 'upriv' set but not used [-Wunused-but-set-variable]

drivers/net/ethernet/mellanox/mlx5/core/en_rep.c: In function 'mlx5e_vport_rep_unload':
drivers/net/ethernet/mellanox/mlx5/core/en_rep.c:1557:21: warning:
 variable 'upriv' set but not used [-Wunused-but-set-variable]

It not used any more since commit ef381359e3a8 ("net/mlx5e: Replace egdev with
indirect block notifications"). Also remove unused variable 'uplink_rpriv'
after this change.

Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Acked-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoMerge branch 'Pass-extack-to-SWITCHDEV_PORT_OBJ_ADD'
David S. Miller [Thu, 13 Dec 2018 00:34:22 +0000 (16:34 -0800)]
Merge branch 'Pass-extack-to-SWITCHDEV_PORT_OBJ_ADD'

Petr Machata says:

====================
Pass extack to SWITCHDEV_PORT_OBJ_ADD

Drivers may need to do validation as a result of port object addition.
An example is mlxsw, which needs to check the configuration of a VXLAN
device attached to an offloaded bridge. Without a mapped VLAN, the
invalidity of the device is not important, but as soon as a pvid,
untagged VLAN is configured for the device, it has to be validated and
offloaded. Should the validation fail, there's currently no way to
communicate details of the failure to the user, beyond an error number.

Because currently, extack is not available at all in that area of code,
this patch starts down at the RTNL level and progresses up towards the
driver(s).

In patch #1, ndo_bridge_setlink is updated to include extack, and
callbacks of all clients are updated as well (ignoring the argument).

In patch #2, the bridge driver is updated to propagate the extack
through to the switchdev border, br_switchdev_port_vlan_add().

Patches #3, #4 and #5 then gradually extend switchdev to pass the extack
argument through to the switchdev blocking notifier chain.

Patches #6 and #7 then update mlxsw to pass the extack argument from
VXLAN events resp. port events on to mlxsw_sp_bridge_8021q_vxlan_join().

Finally in patches #8 and #9, the code paths from the previous two
patches are verified to yield an error message.

v2:
- Patch #1:
    - In ndo_bridge_setlink(), keep the whole extack declaration on the
      same line.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoselftests: mlxsw: extack: Test VLAN add on a port device
Petr Machata [Wed, 12 Dec 2018 17:03:06 +0000 (17:03 +0000)]
selftests: mlxsw: extack: Test VLAN add on a port device

Test mapping a VLAN at a port device such that on the same VLAN, there
already is an unoffloadable VXLAN device.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoselftests: mlxsw: extack: Test VLAN add on a VXLAN device
Petr Machata [Wed, 12 Dec 2018 17:03:04 +0000 (17:03 +0000)]
selftests: mlxsw: extack: Test VLAN add on a VXLAN device

Test mapping a VLAN at a VXLAN device that can't be offloaded.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agomlxsw: spectrum_switchdev: Propagate extack on port VLAN events
Petr Machata [Wed, 12 Dec 2018 17:03:02 +0000 (17:03 +0000)]
mlxsw: spectrum_switchdev: Propagate extack on port VLAN events

After switchdev_handle_port_obj_add() was extended in a preceding patch,
mlxsw_sp_port_obj_add() now takes an extack argument. Propagate it
further by extending a callee chain from mlxsw_sp_port_vlans_add(), via
mlxsw_sp_bridge_port_vlan_add() via mlxsw_sp_port_vlan_bridge_join() via
mlxsw_sp_port_vlan_fid_join() to mlxsw_sp_bridge_ops.fid_get, adding an
extack argument for each of them.

This code path is used when a VLAN is added to a port netdevice if there
already is an unoffloadable VXLAN device with that VLAN mapped.

mlxsw_sp_bridge_8021d_port_join() is updated to obey the new interfaces
changed by the abovementioned code, propagating extack ultimately from
NETDEV_CHANGEUPPER events.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agomlxsw: spectrum_switchdev: Propagate extack on VXLAN VLAN events
Petr Machata [Wed, 12 Dec 2018 17:03:00 +0000 (17:03 +0000)]
mlxsw: spectrum_switchdev: Propagate extack on VXLAN VLAN events

Now that VLAN port object addition notifications carry an extack,
propagate it from mlxsw_sp_switchdev_vxlan_vlans_add() through
mlxsw_sp_switchdev_vxlan_vlan_add() to
mlxsw_sp_bridge_8021q_vxlan_join().

This code path is used when a VLAN is added to a VXLAN netdevice that
cannot be offloaded.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: switchdev: Add extack to switchdev_handle_port_obj_add() callback
Petr Machata [Wed, 12 Dec 2018 17:02:56 +0000 (17:02 +0000)]
net: switchdev: Add extack to switchdev_handle_port_obj_add() callback

Drivers use switchdev_handle_port_obj_add() to handle recursive descent
through lower devices. Change this function prototype to take add_cb
that itself takes an extack argument. Decode extack from
switchdev_notifier_port_obj_info and pass it to add_cb.

Update mlxsw and ocelot drivers which use this helper.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Acked-by: Ivan Vecera <ivecera@redhat.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: switchdev: Add extack to struct switchdev_notifier_info
Petr Machata [Wed, 12 Dec 2018 17:02:54 +0000 (17:02 +0000)]
net: switchdev: Add extack to struct switchdev_notifier_info

In order to pass extack to the drivers that need it, add an extack field
to struct switchdev_notifier_info, and an extack argument to the
function call_switchdev_blocking_notifiers(). Also add a helper function
switchdev_notifier_info_to_extack().

Signed-off-by: Petr Machata <petrm@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Acked-by: Ivan Vecera <ivecera@redhat.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: switchdev: Add extack argument to switchdev_port_obj_add()
Petr Machata [Wed, 12 Dec 2018 17:02:52 +0000 (17:02 +0000)]
net: switchdev: Add extack argument to switchdev_port_obj_add()

After the previous patch, bridge driver has extack argument available to
pass to switchdev. Therefore extend switchdev_port_obj_add() with this
argument, updating all callers, and passing the argument through to
switchdev_port_obj_notify().

Signed-off-by: Petr Machata <petrm@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Acked-by: Ivan Vecera <ivecera@redhat.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: bridge: Propagate extack to switchdev
Petr Machata [Wed, 12 Dec 2018 17:02:50 +0000 (17:02 +0000)]
net: bridge: Propagate extack to switchdev

ndo_bridge_setlink has been updated in the previous patch to have extack
available, and changelink RTNL op has had this argument since the time
extack was added. Propagate both through the bridge driver to eventually
reach br_switchdev_port_vlan_add(), where it will be used by subsequent
patches.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Acked-by: Nikolay Aleksandrov <nikolay@cumulusnetworks.com>
Acked-by: Ivan Vecera <ivecera@redhat.com>
Acked-by: Roopa Prabhu <roopa@cumulusnetworks.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: ndo_bridge_setlink: Add extack
Petr Machata [Wed, 12 Dec 2018 17:02:48 +0000 (17:02 +0000)]
net: ndo_bridge_setlink: Add extack

Drivers may not be able to implement a VLAN addition or reconfiguration.
In those cases it's desirable to explain to the user that it was
rejected (and why).

To that end, add extack argument to ndo_bridge_setlink. Adapt all users
to that change.

Following patches will use the new argument in the bridge driver.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Acked-by: Jiri Pirko <jiri@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agobnxt: remove printing of hwrm message
Jonathan Toppins [Wed, 12 Dec 2018 16:58:51 +0000 (11:58 -0500)]
bnxt: remove printing of hwrm message

bnxt_en 0000:19:00.0 (unregistered net_device) (uninitialized): hwrm
req_type 0x190 seq id 0x6 error 0xffff

The message above is commonly seen when a newer driver is used on
hardware with older firmware. The issue is this message means nothing to
anyone except Broadcom. Remove the message to not confuse users as this
message is really not very informative.

Signed-off-by: Jonathan Toppins <jtoppins@redhat.com>
Acked-by: Michael Chan <michael.chan@broadcom.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonetdevsim: convert to DEFINE_SHOW_ATTRIBUTE
Yangtao Li [Wed, 12 Dec 2018 16:40:07 +0000 (11:40 -0500)]
netdevsim: convert to DEFINE_SHOW_ATTRIBUTE

Use DEFINE_SHOW_ATTRIBUTE macro to simplify the code.

Signed-off-by: Yangtao Li <tiny.windzz@gmail.com>
Acked-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoMerge branch 'stmmac-mt2712-support'
David S. Miller [Wed, 12 Dec 2018 23:21:01 +0000 (15:21 -0800)]
Merge branch 'stmmac-mt2712-support'

Biao Huang says:

====================
add Ethernet driver support for mt2712

Changes in v6:
 modifications according to comments from Rob/Andrew/Sean:
 1. use delay_ps instead of delay stage.
 2. add comments in driver to avoid confusion.
 2. rewrite set_delay function.
 3. modify binding document for properties: tx-delay-ps/rx-delay-ps/pericfg etc.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agodt-binding: mediatek-dwmac: add binding document for MediaTek MT2712 DWMAC
Biao Huang [Wed, 12 Dec 2018 09:35:32 +0000 (17:35 +0800)]
dt-binding: mediatek-dwmac: add binding document for MediaTek MT2712 DWMAC

The commit adds the device tree binding documentation for the MediaTek DWMAC
found on MediaTek MT2712.

Signed-off-by: Biao Huang <biao.huang@mediatek.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agostmmac: dwmac-mediatek: add support for mt2712
Biao Huang [Wed, 12 Dec 2018 09:35:31 +0000 (17:35 +0800)]
stmmac: dwmac-mediatek: add support for mt2712

Add Ethernet support for MediaTek SoCs from the mt2712 family

Signed-off-by: Biao Huang <biao.huang@mediatek.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoMerge branch 'mlxsw-Add-Spectrum-2-multicast-routing-support'
David S. Miller [Wed, 12 Dec 2018 07:01:34 +0000 (23:01 -0800)]
Merge branch 'mlxsw-Add-Spectrum-2-multicast-routing-support'

Ido Schimmel says:

====================
mlxsw: Add Spectrum-2 multicast routing support

Nir says:

In Spectrum the firmware provided an abstraction for multicast routing
on top of the policy engine. In Spectrum-2 this is no longer the case
and the driver must interact directly with the policy engine in order to
program multicast routes. Every route is written as an ACL rule, its
priority set according to route type (*,G) or (S,G) and its action is an
appropriate multicast routing action. Multicast routes are written to a
specific ACL group which is bound to the appropriate IP protocol
IPv4/IPv6.

Patch #1 adds PEMRBT register needed to declare which ACL group is
dedicated for each IP protocol multicast routing function.

Patch #2 Changes initialization order and puts ACL before router as
multicast router now uses ACL module.

Patch #3 adds Spectrum-2 ACL keys needed for multicast route matching.

Patch #4 adds another ACL profile - in addition to existing flower
profile - which allows the multicast routing module to program rules
directly into the ACL block.

Patch #5 adds the ability to update ACL rules' action, since multicast
routes actions may be updated after being configured.

Patch #6 separates rule creation operation and rule action creation
operation as in multicast router the action is created before the route
is inserted.

Patch #7 sharpens priority handling in Spectrum-2, to ensure incorrect
values are not set to rule's priority.

Patch #8 adds the implementation of multicast routing for IPv4 and IPv6
over existing ACL rule programming

Finally, patch #9 adds a test for IPv4/IPv6 multicast routing.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoselftests: forwarding: Add multicast routing test
Nir Dotan [Mon, 10 Dec 2018 07:11:46 +0000 (07:11 +0000)]
selftests: forwarding: Add multicast routing test

Introduce basic testing for both IPv4 and IPv6 multicast. The test creates
an (S,G) type route, sends traffic and verifies traffic arrives when the
route is present and then verifies traffic does not arrive after deleting
the route.
This test requires smcroute - https://github.com/troglobit/smcroute which
is a tool that allows creation of static multicast routes.

Signed-off-by: Nir Dotan <nird@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agomlxsw: spectrum_router: Add Multicast routing support for Spectrum-2
Nir Dotan [Mon, 10 Dec 2018 07:11:45 +0000 (07:11 +0000)]
mlxsw: spectrum_router: Add Multicast routing support for Spectrum-2

Add implementation of Spectrum-2 multicast routes for both IPv4 and IPv6 by
using ACL module explicitly.
In Spectrum-2, multicast routes are set as ACL rules, so initialization
takes care of creating dedicated ACL groups and binding them to the
appropriate multicast routing protocol IPv4/IPv6, and afterwards routes
configuration translates to setting explicit ACL rules.

Signed-off-by: Nir Dotan <nird@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agomlxsw: spectrum_acl: Limit priority value
Nir Dotan [Mon, 10 Dec 2018 07:11:44 +0000 (07:11 +0000)]
mlxsw: spectrum_acl: Limit priority value

In Spectrum-2, higher priority value wins and priority valid values are in
the range of {1,cap_kvd_size-1}. mlxsw_sp_acl_tcam_priority_get converts
from lower-bound priorities alike tc flower to Spectrum-2 HW range. Up
until now tc flower did not provide priority 0 or reached the maximal
value, however multicast routing does provide priority 0.

Therefore, Change mlxsw_sp_acl_tcam_priority_get to verify priority is in
the correct range. Make sure priority is never set to zero and never
exceeds the maximal allowed value.

Signed-off-by: Nir Dotan <nird@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agomlxsw: spectrum_acl: Support rule creation without action creation
Nir Dotan [Mon, 10 Dec 2018 07:11:43 +0000 (07:11 +0000)]
mlxsw: spectrum_acl: Support rule creation without action creation

Up until now, when ACL rule was created its action was created with it.
It suits well for tc flower where ACL rule always needs an action, however
it does not suit multicast router, where the action is created prior to
setting a route, which in Spectrum-2 is actually an ACL rule.

Add support for rule creation without action creation. Do it by adding
afa_block argument to mlxsw_sp_acl_rule_create, which if NULL then an
action would be created, also add an indication within struct
mlxsw_sp_acl_rule_info that tells if the action should be destroyed when
the rule is destroyed.

Signed-off-by: Nir Dotan <nird@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agomlxsw: spectrum_acl: Add replace rule action operation
Nir Dotan [Mon, 10 Dec 2018 07:11:41 +0000 (07:11 +0000)]
mlxsw: spectrum_acl: Add replace rule action operation

Multicast routes actions may be updated after creation. An example for that
is an addition of an egress interface to an existing route.
So far, as tc flower API dictated, ACL rules were either created or
deleted. Since multicast routes in Spectrum-2 are written to ACL as any
rule, it is required to allow the update of a rule's action as it may
change.

Add methods and operations to support updating rule's action. This is
supported only for Spectrum-2.

Signed-off-by: Nir Dotan <nird@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agomlxsw: spectrum_acl: Add multicast router profile operations
Nir Dotan [Mon, 10 Dec 2018 07:11:40 +0000 (07:11 +0000)]
mlxsw: spectrum_acl: Add multicast router profile operations

Add specific ACL operations needed for programming multicast routing ACL
groups and routes.

Signed-off-by: Nir Dotan <nird@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agomlxsw: spectrum_acl: Add Spectrum-2 keys
Nir Dotan [Mon, 10 Dec 2018 07:11:39 +0000 (07:11 +0000)]
mlxsw: spectrum_acl: Add Spectrum-2 keys

Add virtual router ID fields to Spectrum-2 key blocks set, as the field is
required for multicast routing.

Signed-off-by: Nir Dotan <nird@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agomlxsw: spectrum: Change stage of ACL initialization
Nir Dotan [Mon, 10 Dec 2018 07:11:37 +0000 (07:11 +0000)]
mlxsw: spectrum: Change stage of ACL initialization

In Spectrum-2, MC routing is implemented using ACL block explicitly, so
router initialization should take place after ACL initialization.
Set the initialization of the ACL block before IP router initizalization
takes place, so multicast router will be able to allocate ACL data
structures and create its required chains.

Signed-off-by: Nir Dotan <nird@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agomlxsw: reg: Add Policy Engine Multicast Router Binding Table Register
Nir Dotan [Mon, 10 Dec 2018 07:11:33 +0000 (07:11 +0000)]
mlxsw: reg: Add Policy Engine Multicast Router Binding Table Register

In Spectrum-2, multicast routing is implemented explicitly using policy
engine (ACL) block. PEMRBT register is used to bind a dedicated ACL group
to a specific IP protocol.
Add the register to be later used in multicast router implementation.

Signed-off-by: Nir Dotan <nird@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agohinic: optmize rx refill buffer mechanism
Xue Chaojing [Sun, 9 Dec 2018 19:14:19 +0000 (19:14 +0000)]
hinic: optmize rx refill buffer mechanism

There is no need to schedule a different tasklet for refill,
This patch remove it.

Suggested-by: Neil Horman <nhorman@redhat.com>
Signed-off-by: Xue Chaojing <xuechaojing@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet/mlx5e: Use monitor counters for update stats
Eyal Davidovich [Sat, 20 Oct 2018 13:18:00 +0000 (16:18 +0300)]
net/mlx5e: Use monitor counters for update stats

- Adding new notifier block (struct mlx5_nb) monitor_counters_nb
  for handeling MONITOR_COUNTER new event type.
- Adding work queue element: monitor_counters_work for re-arm and
  update stats.
- We re-queue the update stat work, only when working over firmware
  that doesn't support the monitored counters.

Signed-off-by: Eyal Davidovich <eyald@mellanox.com>
Reviewed-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5e: Monitor counters commands support
Eyal Davidovich [Mon, 8 Oct 2018 09:16:01 +0000 (12:16 +0300)]
net/mlx5e: Monitor counters commands support

new file monitor_stats.c for the new API.
add arm_monitor_counter new command support.
add set_monitor_counter new command support.

The device can monitor specific counters and provide an event to notify
when these counters are changed.
The monitoring is done in best effort manner where the minimum
notification period is 200 ms, however when the device is loaded, the
notification might be delayed.
To configure the required counters to be monitored, the
SET_MONITOR_COUNTER command shall be used with a list of counters to be
monitored.
The device firmware can monitor up to HCA_CAP.max_num_of_monitor_counters.
The configuration is done based on counter type (such as ppcnt, q counter,
etc) and additional param according to the type of counter selected.
Upon monitor counter change, the device will generate
Monitor_Counter_Change event.
The device will not generate new events unless the driver re-arms the
monitoring functionality, using the ARM_MONITOR_COUNTER command.

Signed-off-by: Eyal Davidovich <eyald@mellanox.com>
Reviewed-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5e: Avoid query PPCNT register if not supported by the device
Eyal Davidovich [Sun, 7 Oct 2018 12:18:37 +0000 (15:18 +0300)]
net/mlx5e: Avoid query PPCNT register if not supported by the device

PPCNT is not supported if PCAM access reg is supported and ppcnt bit is clear.

Signed-off-by: Eyal Davidovich <eyald@mellanox.com>
Reviewed-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5e: Use CQE padding for Ethernet CQs
Daniel Jurgens [Mon, 5 Nov 2018 22:05:37 +0000 (16:05 -0600)]
net/mlx5e: Use CQE padding for Ethernet CQs

Writing 64B CQEs to 128B cache lines results in a RMW operation. Padding
the CQEs to 128B if possible improves performance on 128B cache line
systems like PPC.

Testing on PPC showed up to a 24% improvement in small packet throughput
vs the default behavior, depending on the workload and system topology.

Signed-off-by: Daniel Jurgens <danielj@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5e: Support multiple encapsulations for a TC flow
Eli Britstein [Sun, 18 Nov 2018 12:21:46 +0000 (14:21 +0200)]
net/mlx5e: Support multiple encapsulations for a TC flow

Currently a flow is associated with a single encap structure. The FW
extended destination features enables the driver to associate a flow
with multiple encap instances.

Change the encap id field from a flow scope to a per destination value
in the flow attributes struct. Use the encaps array to associate a flow
table entry with multiple encap entries.

Update the neigh logic to offload only if all encapsulations used in a
flow are connected, and un-offload upon the first one disconnected.

Note that the driver can now support up to two encap destinations.

Signed-off-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Oz Shlomo <ozsh@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5e: Allow association of a flow to multiple encaps
Eli Britstein [Sun, 2 Dec 2018 07:44:13 +0000 (09:44 +0200)]
net/mlx5e: Allow association of a flow to multiple encaps

Currently a flow can be associated with a single encap entry. The
extended destination feature enables the driver to configure multiple
encap entries per flow.

Change the encap flow association field to array as a pre-step towards
supporting multiple encap destinations. Use only the first array
element, with no functional change.

Signed-off-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Oz Shlomo <ozsh@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5e: Change parse attr struct to accommodate multiple tunnel infos
Eli Britstein [Sun, 18 Nov 2018 11:16:52 +0000 (13:16 +0200)]
net/mlx5e: Change parse attr struct to accommodate multiple tunnel infos

Currently the driver can support only a single TC tunnel_set action.
Change the tunnel info fields to arrays, as a pre-step to support
multiple encapsulations for a single flow, with no functional change.

Signed-off-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Oz Shlomo <ozsh@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5e: Support header rewrite actions with remote port mirroring
Eli Britstein [Mon, 26 Nov 2018 05:24:08 +0000 (07:24 +0200)]
net/mlx5e: Support header rewrite actions with remote port mirroring

A rule with the following actions is split to a two level FDB:
1. Forward to local mirror vport
2. Header rewrite
3. Forward to local vport
In the first level flow table, forward the packet to the local port and
forward the packet to the second level flow table for header rewrite and
local port forwarding. This configuration fails when mirroring to a
remote encapsulated destination because currently an FTE cannot support
encap and table destinations.

Use the extended destination capabilities to configure the first level
flow table with a multi-destination FTE to the uplink and second level
table and the second level flow table for the header rewrite and local
port forwarding.

Signed-off-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Oz Shlomo <ozsh@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5e: Replace the split logic with extended destination
Eli Britstein [Thu, 22 Nov 2018 12:04:33 +0000 (14:04 +0200)]
net/mlx5e: Replace the split logic with extended destination

Currently the FTE encap flag applies to all destinations.
To support mirroring encapsulated traffic to a local port the driver
split the two destinations to two flow table entries:
Table#0: - FWD to the local vport
         - Goto table#1
Table#1: - Encap and FWD to wire
The firmware extended destination capabilities enable the driver to set
an encapsulation flag per destination.

Remove the split logic and use the extended destination mechanism
instead.

Note that split technique is still required for pedit and VLAN push
scenarios.

Signed-off-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Oz Shlomo <ozsh@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5e: Annul encap action ordering requirement
Eli Britstein [Tue, 4 Dec 2018 06:06:41 +0000 (08:06 +0200)]
net/mlx5e: Annul encap action ordering requirement

Currently a FW syndrome is emitted if the driver configures a
multi-destination FTE where the first destination is a tunneled uplink
port and the second destination is a local vPort.

Support this scenario by creating a multi-destination FTE using the
firmware's extended destination capabilities.

Signed-off-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Oz Shlomo <ozsh@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5e: Move flow attr reformat action bit to per dest flags
Eli Britstein [Sat, 1 Dec 2018 07:40:43 +0000 (09:40 +0200)]
net/mlx5e: Move flow attr reformat action bit to per dest flags

Flow attr reformat action bit is moved from the global action bits to a
per destination flags field, as a pre-step for adding additional flags
to support encapsulation properties per destination, with no
functionality change.

Signed-off-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Oz Shlomo <ozsh@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5e: Refactor eswitch flow attr for destination specific properties
Eli Britstein [Sat, 1 Dec 2018 07:40:43 +0000 (09:40 +0200)]
net/mlx5e: Refactor eswitch flow attr for destination specific properties

Currently the eswitch flow attr structure stores each destination
specific property in its own specific array.
Group them in an array of destination structures as a pre-step towards
adding additional destination specific field properties.

Signed-off-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Oz Shlomo <ozsh@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5: E-Switch, Rename esw attr mirror count field
Eli Britstein [Fri, 23 Nov 2018 07:38:27 +0000 (09:38 +0200)]
net/mlx5: E-Switch, Rename esw attr mirror count field

The mirror count esw attributes field is used to determine if splitting
the rule to two FTEs is required while programming e-switch mirroring.
Rename it to split count, making it clearer with no functional change.

Signed-off-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Oz Shlomo <ozsh@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5: Consider encapsulation properties when comparing destinations
Eli Britstein [Sun, 18 Nov 2018 06:25:17 +0000 (08:25 +0200)]
net/mlx5: Consider encapsulation properties when comparing destinations

Currently the driver identifies identical vport destinations by
comparing the vport ID. The FW extended destination feature enables the
driver to forward the packet to the same vport with multiple
encapsulation properties.

Change the vport destination comparison logic to compare
the encapsulation properties in addition to the vport ID.

Signed-off-by: Eli Britstein <elibr@mellanox.com>
Acked-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Oz Shlomo <ozsh@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agobpf: fix up uapi helper description and sync bpf header with tools
Daniel Borkmann [Tue, 11 Dec 2018 09:26:33 +0000 (10:26 +0100)]
bpf: fix up uapi helper description and sync bpf header with tools

Minor markup fixup from bpf-next into net-next merge in the BPF helper
description of bpf_sk_lookup_tcp() and bpf_sk_lookup_udp(). Also sync
up the copy of bpf.h from tooling infrastructure.

Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Reviewed-by: Quentin Monnet <quentin.monnet@netronome.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next
David S. Miller [Tue, 11 Dec 2018 02:00:43 +0000 (18:00 -0800)]
Merge git://git./linux/kernel/git/bpf/bpf-next

Daniel Borkmann says:

====================
pull-request: bpf-next 2018-12-11

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

It has three minor merge conflicts, resolutions:

1) tools/testing/selftests/bpf/test_verifier.c

 Take first chunk with alignment_prevented_execution.

2) net/core/filter.c

  [...]
  case bpf_ctx_range_ptr(struct __sk_buff, flow_keys):
  case bpf_ctx_range(struct __sk_buff, wire_len):
        return false;
  [...]

3) include/uapi/linux/bpf.h

  Take the second chunk for the two cases each.

The main changes are:

1) Add support for BPF line info via BTF and extend libbpf as well
   as bpftool's program dump to annotate output with BPF C code to
   facilitate debugging and introspection, from Martin.

2) Add support for BPF_ALU | BPF_ARSH | BPF_{K,X} in interpreter
   and all JIT backends, from Jiong.

3) Improve BPF test coverage on archs with no efficient unaligned
   access by adding an "any alignment" flag to the BPF program load
   to forcefully disable verifier alignment checks, from David.

4) Add a new bpf_prog_test_run_xattr() API to libbpf which allows for
   proper use of BPF_PROG_TEST_RUN with data_out, from Lorenz.

5) Extend tc BPF programs to use a new __sk_buff field called wire_len
   for more accurate accounting of packets going to wire, from Petar.

6) Improve bpftool to allow dumping the trace pipe from it and add
   several improvements in bash completion and map/prog dump,
   from Quentin.

7) Optimize arm64 BPF JIT to always emit movn/movk/movk sequence for
   kernel addresses and add a dedicated BPF JIT backend allocator,
   from Ard.

8) Add a BPF helper function for IR remotes to report mouse movements,
   from Sean.

9) Various cleanups in BPF prog dump e.g. to make UAPI bpf_prog_info
   member naming consistent with existing conventions, from Yonghong
   and Song.

10) Misc cleanups and improvements in allowing to pass interface name
    via cmdline for xdp1 BPF example, from Matteo.

11) Fix a potential segfault in BPF sample loader's kprobes handling,
    from Daniel T.

12) Fix SPDX license in libbpf's README.rst, from Andrey.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoneighbor: gc_list changes should be protected by table lock
David Ahern [Mon, 10 Dec 2018 21:54:07 +0000 (13:54 -0800)]
neighbor: gc_list changes should be protected by table lock

Adding and removing neighbor entries to / from the gc_list need to be
done while holding the table lock; a couple of places were missed in the
original patch.

Move the list_add_tail in neigh_alloc to ___neigh_create where the lock
is already obtained. Since neighbor entries should rarely be moved
to/from PERMANENT state, add lock/unlock around the gc_list changes in
neigh_change_state rather than extending the lock hold around all
neighbor updates.

Fixes: 58956317c8de ("neighbor: Improve garbage collection")
Reported-by: Andrei Vagin <avagin@gmail.com>
Reported-by: syzbot+6cc2fd1d3bdd2e007363@syzkaller.appspotmail.com
Reported-by: syzbot+35e87b87c00f386b041f@syzkaller.appspotmail.com
Reported-by: syzbot+b354d1fb59091ea73c37@syzkaller.appspotmail.com
Reported-by: syzbot+3ddead5619658537909b@syzkaller.appspotmail.com
Reported-by: syzbot+424d47d5c456ce8b2bbe@syzkaller.appspotmail.com
Reported-by: syzbot+e4d42eb35f6a27b0a628@syzkaller.appspotmail.com
Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoMerge tag 'mlx5e-updates-2018-12-10' of git://git.kernel.org/pub/scm/linux/kernel...
David S. Miller [Tue, 11 Dec 2018 01:06:58 +0000 (17:06 -0800)]
Merge tag 'mlx5e-updates-2018-12-10' of git://git./linux/kernel/git/saeed/linux

Saeed Mahameed:

====================
mlx5e-updates-2018-12-10 (gre)

This patch set adds GRE offloading support to Mellanox ethernet driver.

Patches 1-5 replace the existing egdev mechanism with the new TC indirect
block binds mechanism that was introduced by Netronome:
7f76fa36754b ("net: sched: register callbacks for indirect tc block binds")

Patches 6-9 add GRE offloading support along with some required
refactoring work.

Patch 10, Add netif_is_gretap()/netif_is_ip6gretap()
 - Changed the is_gretap_dev and is_ip6gretap_dev logic from structure
   comparison to string comparison of the rtnl_link_ops kind field.

Patch 11, add GRE offloading support to mlx5.

Patch 12 removes the egdev mechanism from TC as it is no longer used by
any of the drivers.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet/sched: Remove egdev mechanism
Oz Shlomo [Tue, 6 Nov 2018 07:58:37 +0000 (09:58 +0200)]
net/sched: Remove egdev mechanism

The egdev mechanism was replaced by the TC indirect block notifications
platform.

Signed-off-by: Oz Shlomo <ozsh@mellanox.com>
Reviewed-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Jiri Pirko <jiri@mellanox.com>
Cc: John Hurley <john.hurley@netronome.com>
Cc: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5e: Add GRE protocol offloading
Oz Shlomo [Mon, 29 Oct 2018 06:54:42 +0000 (08:54 +0200)]
net/mlx5e: Add GRE protocol offloading

Add HW offloading support for TC flower filters configured on
gretap/ip6gretap net devices.

Signed-off-by: Oz Shlomo <ozsh@mellanox.com>
Reviewed-by: Eli Britstein <elibr@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet: Add netif_is_gretap()/netif_is_ip6gretap()
Oz Shlomo [Wed, 21 Nov 2018 10:15:34 +0000 (12:15 +0200)]
net: Add netif_is_gretap()/netif_is_ip6gretap()

Changed the is_gretap_dev and is_ip6gretap_dev logic from structure
comparison to string comparison of the rtnl_link_ops kind field.

This approach aligns with the current identification methods and function
names of vxlan and geneve network devices.

Convert mlxsw to use these helpers and use them in downstream mlx5 patch.

Signed-off-by: Oz Shlomo <ozsh@mellanox.com>
Reviewed-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5e: Move TC tunnel offloading code to separate source file
Oz Shlomo [Sun, 2 Dec 2018 12:43:27 +0000 (14:43 +0200)]
net/mlx5e: Move TC tunnel offloading code to separate source file

Move tunnel offloading related code to a separate source file for better
code maintainability.

Code refactoring with no functional change.

Signed-off-by: Oz Shlomo <ozsh@mellanox.com>
Reviewed-by: Eli Britstein <elibr@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5e: Branch according to classified tunnel type
Oz Shlomo [Wed, 14 Nov 2018 13:21:27 +0000 (15:21 +0200)]
net/mlx5e: Branch according to classified tunnel type

Currently the tunnel offloading encap/decap methods assumes that VXLAN
is the sole tunneling protocol. Lay the infrastructure for supporting
multiple tunneling protocols by branching according to the tunnel
net device kind.

Encap filters tunnel type is determined according to the egress/mirred
net device. Decap filters classify the tunnel type according to the
filter's ingress net device kind.

Distinguish between the tunnel type as defined by the SW model and
the FW reformat type that specifies the HW operation being made.

Signed-off-by: Oz Shlomo <ozsh@mellanox.com>
Reviewed-by: Eli Britstein <elibr@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5e: Refactor VXLAN tunnel decap offloading code
Oz Shlomo [Wed, 14 Nov 2018 13:41:50 +0000 (15:41 +0200)]
net/mlx5e: Refactor VXLAN tunnel decap offloading code

Separates the vxlan header match handling from the matching on the
general fields of ipv4/6 tunnels, thus allowing the common IP tunnel
match code to branch in down stream patch, to multiple IP tunnels.

This patch doesn't add any functionality.

Signed-off-by: Oz Shlomo <ozsh@mellanox.com>
Reviewed-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5e: Refactor VXLAN tunnel encap offloading code
Oz Shlomo [Wed, 14 Nov 2018 13:41:50 +0000 (15:41 +0200)]
net/mlx5e: Refactor VXLAN tunnel encap offloading code

Separates the vxlan header encap logic from the general ipv4/6
encapsulation methods, thus allowing the common IP encap/decap code to
branch in downstream patch to multiple IP tunnels.

Code refactoring with no functional change.

Signed-off-by: Oz Shlomo <ozsh@mellanox.com>
Reviewed-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5e: Replace egdev with indirect block notifications
Oz Shlomo [Sun, 28 Oct 2018 11:03:54 +0000 (13:03 +0200)]
net/mlx5e: Replace egdev with indirect block notifications

Use TC indirect block notifications to offload filters that
are configured on higher level device interfaces (e.g. tunnel
devices). This mechanism replaces the current egdev implementation.

Signed-off-by: Oz Shlomo <ozsh@mellanox.com>
Reviewed-by: Eli Britstein <elibr@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5e: Propagate the filter's net device to mlx5e structures
Oz Shlomo [Sun, 28 Oct 2018 08:46:34 +0000 (10:46 +0200)]
net/mlx5e: Propagate the filter's net device to mlx5e structures

Propagate the filter's net_device parameter to the tc flower parsed
attributes structure so that it can later be used in tunnel decap
offloading sequences.

Pre-step for replacing egdev logic with the indirect block
notification mechanism.

Signed-off-by: Oz Shlomo <ozsh@mellanox.com>
Reviewed-by: Eli Britstein <elibr@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5e: Provide the TC filter netdev as parameter to flower callbacks
Oz Shlomo [Sun, 28 Oct 2018 07:14:50 +0000 (09:14 +0200)]
net/mlx5e: Provide the TC filter netdev as parameter to flower callbacks

Currently the driver controls flower filters that are installed on its
devices. However, with the introduction of the indirect block
notifications platform the driver may receive control events for filters
that are installed on higher level net devices (e.g. tunnel devices).
Therefore, the driver filter control API will not be able to implicitly
assume the filter's net device.

Explicitly specify the filter's net device, no functional change

Signed-off-by: Oz Shlomo <ozsh@mellanox.com>
Reviewed-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5e: Support TC indirect block notifications for eswitch uplink reprs
Oz Shlomo [Sun, 28 Oct 2018 06:34:51 +0000 (08:34 +0200)]
net/mlx5e: Support TC indirect block notifications for eswitch uplink reprs

Towards using this mechanism as the means to offload tunnel decap rules
set on SW tunnel devices instead of egdev, add the supporting structures
and functions.

Signed-off-by: Oz Shlomo <ozsh@mellanox.com>
Reviewed-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5e: Store eswitch uplink representor state on a dedicated struct
Oz Shlomo [Thu, 25 Oct 2018 18:51:11 +0000 (21:51 +0300)]
net/mlx5e: Store eswitch uplink representor state on a dedicated struct

Currently only a single field in the representor private structure
is relevant for uplink representors.  As a pre-step to allow adding
additional uplink representor fields, introduce uplink representor
private structure.

This is prepration step towards replacing egdev logic with the
indirect block notification mechanism. This patch doesn't change
any functionality.

Signed-off-by: Oz Shlomo <ozsh@mellanox.com>
Reviewed-by: Eli Britstein <elibr@mellanox.com>
Acked-by: Or Gerlitz <ogerlitz@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 [Mon, 10 Dec 2018 23:43:47 +0000 (15:43 -0800)]
Merge branch 'mlx5-next' of git://git./linux/kernel/git/mellanox/linux

mlx5-next shared branch with rdma subtree to avoid mlx5 rdma v.s. netdev
conflicts.

Highlights:

1) RDMA ODP  (On Demand Paging) improvements and moving ODP logic to
mlx5 RDMA driver
2) Improved mlx5 core driver and device events handling and provided API
for upper layers to subscribe to device events.
3) RDMA only code cleanup from mlx5 core
4) Add helper to get CQE opcode
5) Rework handling of port module events
6) shared mlx5_ifc.h updates to avoid conflicts

Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agoMerge branch 'rename-info_cnt-to-nr_info'
Alexei Starovoitov [Mon, 10 Dec 2018 22:51:45 +0000 (14:51 -0800)]
Merge branch 'rename-info_cnt-to-nr_info'

Yonghong Song says:

====================
Before func_info and line_info are added to the kernel, there are several
fields in structure bpf_prog_info specifying the "count" of a user buffer, e.g.,
        __u32 nr_jited_ksyms;
        __u32 nr_jited_func_lens;
The naming convention has the prefix "nr_".

The func_info and line_info support added several fields
        __u32 func_info_cnt;
        __u32 line_info_cnt;
        __u32 jited_line_info_cnt;
to indicate the "count" of buffers func_info, line_info and jited_line_info.
The original intention is to keep the field names the same as those in
structure bpf_attr, so it will be clear that the "count" returned to user
space will be the same as the one passed to the kernel during prog load.

Unfortunately, the field names *_info_cnt are not consistent with
other existing fields in bpf_prog_info.
This patch set renamed the fields *_info_cnt to nr_*_info
to keep naming convention consistent.
====================

Signed-off-by: Alexei Starovoitov <ast@kernel.org>
5 years agotools/bpf: rename *_info_cnt to nr_*_info
Yonghong Song [Mon, 10 Dec 2018 22:14:10 +0000 (14:14 -0800)]
tools/bpf: rename *_info_cnt to nr_*_info

Rename all occurances of *_info_cnt field access
to nr_*_info in tools directory.

The local variables finfo_cnt, linfo_cnt and jited_linfo_cnt
in function do_dump() of tools/bpf/bpftool/prog.c are also
changed to nr_finfo, nr_linfo and nr_jited_linfo to
keep naming convention consistent.

Acked-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Yonghong Song <yhs@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
5 years agotools/bpf: sync kernel uapi bpf.h to tools directory
Yonghong Song [Mon, 10 Dec 2018 22:14:09 +0000 (14:14 -0800)]
tools/bpf: sync kernel uapi bpf.h to tools directory

Sync kernel uapi bpf.h "*_info_cnt => nr_*_info"
changes to tools directory.

Acked-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Yonghong Song <yhs@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
5 years agobpf: rename *_info_cnt to nr_*_info in bpf_prog_info
Yonghong Song [Mon, 10 Dec 2018 22:14:08 +0000 (14:14 -0800)]
bpf: rename *_info_cnt to nr_*_info in bpf_prog_info

In uapi bpf.h, currently we have the following fields in
the struct bpf_prog_info:
__u32 func_info_cnt;
__u32 line_info_cnt;
__u32 jited_line_info_cnt;
The above field names "func_info_cnt" and "line_info_cnt"
also appear in union bpf_attr for program loading.

The original intention is to keep the names the same
between bpf_prog_info and bpf_attr
so it will imply what we returned to user space will be
the same as what the user space passed to the kernel.

Such a naming convention in bpf_prog_info is not consistent
with other fields like:
        __u32 nr_jited_ksyms;
        __u32 nr_jited_func_lens;

This patch made this adjustment so in bpf_prog_info
newly introduced *_info_cnt becomes nr_*_info.

Acked-by: Song Liu <songliubraving@fb.com>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Yonghong Song <yhs@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
5 years agobpf: clean up bpf_prog_get_info_by_fd()
Song Liu [Mon, 10 Dec 2018 19:17:50 +0000 (11:17 -0800)]
bpf: clean up bpf_prog_get_info_by_fd()

info.nr_jited_ksyms and info.nr_jited_func_lens cannot be 0 in these two
statements, so we don't need to check them.

Signed-off-by: Song Liu <songliubraving@fb.com>
Acked-by: Martin KaFai Lau <kafai@fb.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
5 years agonet/mlx5: Remove the get protocol device interface entry
Or Gerlitz [Mon, 10 Dec 2018 21:15:17 +0000 (13:15 -0800)]
net/mlx5: Remove the get protocol device interface entry

This isn't used anywhere across the mlx5 driver stack,
remove it.

Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5: Support extended destination format in flow steering command
Eli Britstein [Mon, 10 Dec 2018 21:15:16 +0000 (13:15 -0800)]
net/mlx5: Support extended destination format in flow steering command

Update the flow steering command formatting according to the extended
destination API.
Note that the FW dictates that multi destination FTEs that involve at
least one encap must use the extended destination format, while single
destination ones must use the legacy format.
Using extended destination format requires FW support. Check for its
capabilities and return error if not supported.

Signed-off-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Oz Shlomo <ozsh@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5: E-Switch, Change vhca id valid bool field to bit flag
Eli Britstein [Mon, 10 Dec 2018 21:15:15 +0000 (13:15 -0800)]
net/mlx5: E-Switch, Change vhca id valid bool field to bit flag

Change the driver flow destination struct to use bit flags with the vhca
id valid being the 1st one. The flags field is more extendable and will
be used in downstream patch.

Signed-off-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Oz Shlomo <ozsh@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5: Introduce extended destination fields
Eli Britstein [Mon, 10 Dec 2018 21:15:14 +0000 (13:15 -0800)]
net/mlx5: Introduce extended destination fields

Extended destinations provide the ability to configure different
encapsulation properties per destination on a single FTE. This is
needed for use-cases such as remote mirroring over tunneled networks.

Signed-off-by: Eli Britstein <elibr@mellanox.com>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Reviewed-by: Oz Shlomo <ozsh@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5: Revise gre and nvgre key formats
Oz Shlomo [Mon, 10 Dec 2018 21:15:13 +0000 (13:15 -0800)]
net/mlx5: Revise gre and nvgre key formats

GRE RFC defines a 32 bit key field. NVGRE RFC splits the 32 bit
key field to 24 bit VSID (gre_key_h) and 8 bit flow entropy (gre_key_l).

Define the two key parsing alternatives in a union, thus enabling both
access methods.

Signed-off-by: Oz Shlomo <ozsh@mellanox.com>
Reviewed-by: Eli Britstein <elibr@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5: Add monitor commands layout and event data
Eyal Davidovich [Mon, 10 Dec 2018 21:15:12 +0000 (13:15 -0800)]
net/mlx5: Add monitor commands layout and event data

Will be used in downstream patch to monitor counter changes
by the HCA and report it to the driver by an event.
The driver will update its counters cached data accordingly.

Signed-off-by: Eyal Davidovich <eyald@mellanox.com>
Reviewed-by: Eran Ben Elisha <eranbe@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5: Add support for plugged-disabled cable status in PME
Mikhael Goikhman [Mon, 10 Dec 2018 21:15:11 +0000 (13:15 -0800)]
net/mlx5: Add support for plugged-disabled cable status in PME

Support a new hardware module status in port module events:
- module_status=0x4 (Cable plugged, but disabled)

Signed-off-by: Mikhael Goikhman <migo@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5: Add support for PCIe power slot exceeded error in PME
Mikhael Goikhman [Mon, 10 Dec 2018 21:15:10 +0000 (13:15 -0800)]
net/mlx5: Add support for PCIe power slot exceeded error in PME

Support a new hardware error type in port module events:
- error_type=0xc (PCIe system power slot exceeded)

Signed-off-by: Mikhael Goikhman <migo@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5: Rework handling of port module events
Mikhael Goikhman [Mon, 10 Dec 2018 21:15:09 +0000 (13:15 -0800)]
net/mlx5: Rework handling of port module events

Add explicit HW defined error values. For simplicity, keep counters for all
statuses starting from 0, although currently status=0 is not used.

Additionally, when HW signals an unexpected cable status, it is reported
now rather than ignored. And status counter is now updated on errors.

Signed-off-by: Mikhael Goikhman <migo@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agobpf: bpftool: Fix newline and p_err issue
Martin KaFai Lau [Mon, 10 Dec 2018 18:53:24 +0000 (10:53 -0800)]
bpf: bpftool: Fix newline and p_err issue

This patch fixes a few newline issues and also
replaces p_err with p_info in prog.c

Fixes: b053b439b72a ("bpf: libbpf: bpftool: Print bpf_line_info during prog dump")
Cc: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: Martin KaFai Lau <kafai@fb.com>
Acked-by: Jakub Kicinski <jakub.kicinski@netronome.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
5 years agotcp: handle EOR and FIN conditions the same in tcp_tso_should_defer()
Eric Dumazet [Mon, 10 Dec 2018 14:10:02 +0000 (06:10 -0800)]
tcp: handle EOR and FIN conditions the same in tcp_tso_should_defer()

In commit f9bfe4e6a9d0 ("tcp: lack of available data can also cause
TSO defer") we moved the test in tcp_tso_should_defer() for packets
with a FIN flag, and we mentioned that the same would be done
later for EOR flag.

Both flags should be handled at the same time, after all other
heuristics have been considered. They both mean that no more bytes
can be added to this skb by an application.

Signed-off-by: Eric Dumazet <edumazet@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoMerge branch 'dsa-ksz-Add-reset-GPIO-handling'
David S. Miller [Mon, 10 Dec 2018 20:07:56 +0000 (12:07 -0800)]
Merge branch 'dsa-ksz-Add-reset-GPIO-handling'

Marek Vasut says:

====================
net: dsa: ksz: Add reset GPIO handling

Add code to handle optional reset GPIO in the KSZ switch driver and a
matching DT property adjustments.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: dsa: ksz: Add reset GPIO handling
Marek Vasut [Mon, 10 Dec 2018 13:43:06 +0000 (14:43 +0100)]
net: dsa: ksz: Add reset GPIO handling

Add code to handle optional reset GPIO in the KSZ switch driver. The switch
has a reset GPIO line which can be controlled by the CPU, so make sure it is
configured correctly in such setups.

Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Vivien Didelot <vivien.didelot@savoirfairelinux.com>
Cc: Woojung Huh <woojung.huh@microchip.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: Tristram Ha <Tristram.Ha@microchip.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: dsa: ksz: Add optional reset GPIO to Microchip KSZ switch binding
Marek Vasut [Mon, 10 Dec 2018 13:43:05 +0000 (14:43 +0100)]
net: dsa: ksz: Add optional reset GPIO to Microchip KSZ switch binding

Add optional reset GPIO, as such a signal is available on the KSZ switches.

Signed-off-by: Marek Vasut <marex@denx.de>
Cc: Andrew Lunn <andrew@lunn.ch>
Cc: Florian Fainelli <f.fainelli@gmail.com>
Cc: Woojung Huh <Woojung.Huh@microchip.com>
Cc: David S. Miller <davem@davemloft.net>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agobonding: convert to DEFINE_SHOW_ATTRIBUTE
Yangtao Li [Mon, 10 Dec 2018 16:08:45 +0000 (11:08 -0500)]
bonding: convert to DEFINE_SHOW_ATTRIBUTE

Use DEFINE_SHOW_ATTRIBUTE macro to simplify the code.

Signed-off-by: Yangtao Li <tiny.windzz@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agofjes: convert to DEFINE_SHOW_ATTRIBUTE
Yangtao Li [Mon, 10 Dec 2018 15:57:38 +0000 (10:57 -0500)]
fjes: convert to DEFINE_SHOW_ATTRIBUTE

Use DEFINE_SHOW_ATTRIBUTE macro to simplify the code.

Signed-off-by: Yangtao Li <tiny.windzz@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: xenbus: convert to DEFINE_SHOW_ATTRIBUTE
Yangtao Li [Mon, 10 Dec 2018 15:53:29 +0000 (10:53 -0500)]
net: xenbus: convert to DEFINE_SHOW_ATTRIBUTE

Use DEFINE_SHOW_ATTRIBUTE macro to simplify the code.

Signed-off-by: Yangtao Li <tiny.windzz@gmail.com>
Acked-by: Wei Liu <wei.liu2@citrix.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoieee802154: at86rf230: convert to DEFINE_SHOW_ATTRIBUTE
Yangtao Li [Mon, 10 Dec 2018 15:46:03 +0000 (10:46 -0500)]
ieee802154: at86rf230: convert to DEFINE_SHOW_ATTRIBUTE

Use DEFINE_SHOW_ATTRIBUTE macro to simplify the code.

Signed-off-by: Yangtao Li <tiny.windzz@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agoipvlan: Remove a useless comparison
YueHaibing [Mon, 10 Dec 2018 11:25:38 +0000 (19:25 +0800)]
ipvlan: Remove a useless comparison

Fix following gcc warning:

drivers/net/ipvlan/ipvlan_main.c:543:12: warning:
 comparison is always false due to limited range of data type [-Wtype-limits]

'mode' is a u16 variable, IPVLAN_MODE_L2 is zero,
the comparison is always false

Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet: hns3: fix spelling mistake "offser" -> "offset"
Colin Ian King [Mon, 10 Dec 2018 09:29:48 +0000 (09:29 +0000)]
net: hns3: fix spelling mistake "offser" -> "offset"

There is a spelling mistake in a msg string, fix this.

Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agobpf: relax verifier restriction on BPF_MOV | BPF_ALU
Jiong Wang [Fri, 7 Dec 2018 17:16:18 +0000 (12:16 -0500)]
bpf: relax verifier restriction on BPF_MOV | BPF_ALU

Currently, the destination register is marked as unknown for 32-bit
sub-register move (BPF_MOV | BPF_ALU) whenever the source register type is
SCALAR_VALUE.

This is too conservative that some valid cases will be rejected.
Especially, this may turn a constant scalar value into unknown value that
could break some assumptions of verifier.

For example, test_l4lb_noinline.c has the following C code:

    struct real_definition *dst

1:  if (!get_packet_dst(&dst, &pckt, vip_info, is_ipv6))
2:    return TC_ACT_SHOT;
3:
4:  if (dst->flags & F_IPV6) {

get_packet_dst is responsible for initializing "dst" into valid pointer and
return true (1), otherwise return false (0). The compiled instruction
sequence using alu32 will be:

  412: (54) (u32) r7 &= (u32) 1
  413: (bc) (u32) r0 = (u32) r7
  414: (95) exit

insn 413, a BPF_MOV | BPF_ALU, however will turn r0 into unknown value even
r7 contains SCALAR_VALUE 1.

This causes trouble when verifier is walking the code path that hasn't
initialized "dst" inside get_packet_dst, for which case 0 is returned and
we would then expect verifier concluding line 1 in the above C code pass
the "if" check, therefore would skip fall through path starting at line 4.
Now, because r0 returned from callee has became unknown value, so verifier
won't skip analyzing path starting at line 4 and "dst->flags" requires
dereferencing the pointer "dst" which actually hasn't be initialized for
this path.

This patch relaxed the code marking sub-register move destination. For a
SCALAR_VALUE, it is safe to just copy the value from source then truncate
it into 32-bit.

A unit test also included to demonstrate this issue. This test will fail
before this patch.

This relaxation could let verifier skipping more paths for conditional
comparison against immediate. It also let verifier recording a more
accurate/strict value for one register at one state, if this state end up
with going through exit without rejection and it is used for state
comparison later, then it is possible an inaccurate/permissive value is
better. So the real impact on verifier processed insn number is complex.
But in all, without this fix, valid program could be rejected.

>From real benchmarking on kernel selftests and Cilium bpf tests, there is
no impact on processed instruction number when tests ares compiled with
default compilation options. There is slightly improvements when they are
compiled with -mattr=+alu32 after this patch.

Also, test_xdp_noinline/-mattr=+alu32 now passed verification. It is
rejected before this fix.

Insn processed before/after this patch:

                        default     -mattr=+alu32

Kernel selftest

===
test_xdp.o              371/371      369/369
test_l4lb.o             6345/6345    5623/5623
test_xdp_noinline.o     2971/2971    rejected/2727
test_tcp_estates.o      429/429      430/430

Cilium bpf
===
bpf_lb-DLB_L3.o:        2085/2085     1685/1687
bpf_lb-DLB_L4.o:        2287/2287     1986/1982
bpf_lb-DUNKNOWN.o:      690/690       622/622
bpf_lxc.o:              95033/95033   N/A
bpf_netdev.o:           7245/7245     N/A
bpf_overlay.o:          2898/2898     3085/2947

NOTE:
  - bpf_lxc.o and bpf_netdev.o compiled by -mattr=+alu32 are rejected by
    verifier due to another issue inside verifier on supporting alu32
    binary.
  - Each cilium bpf program could generate several processed insn number,
    above number is sum of them.

v1->v2:
 - Restrict the change on SCALAR_VALUE.
 - Update benchmark numbers on Cilium bpf tests.

Signed-off-by: Jiong Wang <jiong.wang@netronome.com>
Signed-off-by: Alexei Starovoitov <ast@kernel.org>
5 years agoMerge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net
David S. Miller [Mon, 10 Dec 2018 05:27:48 +0000 (21:27 -0800)]
Merge git://git./linux/kernel/git/davem/net

Several conflicts, seemingly all over the place.

I used Stephen Rothwell's sample resolutions for many of these, if not
just to double check my own work, so definitely the credit largely
goes to him.

The NFP conflict consisted of a bug fix (moving operations
past the rhashtable operation) while chaning the initial
argument in the function call in the moved code.

The net/dsa/master.c conflict had to do with a bug fix intermixing of
making dsa_master_set_mtu() static with the fixing of the tagging
attribute location.

cls_flower had a conflict because the dup reject fix from Or
overlapped with the addition of port range classifiction.

__set_phy_supported()'s conflict was relatively easy to resolve
because Andrew fixed it in both trees, so it was just a matter
of taking the net-next copy.  Or at least I think it was :-)

Joe Stringer's fix to the handling of netns id 0 in bpf_sk_lookup()
intermixed with changes on how the sdif and caller_net are calculated
in these code paths in net-next.

The remaining BPF conflicts were largely about the addition of the
__bpf_md_ptr stuff in 'net' overlapping with adjustments and additions
to the relevant data structure where the MD pointer macros are used.

Signed-off-by: David S. Miller <davem@davemloft.net>
5 years agonet/mlx5: Move flow counters data structures from flow steering header
Saeed Mahameed [Wed, 5 Dec 2018 02:03:03 +0000 (18:03 -0800)]
net/mlx5: Move flow counters data structures from flow steering header

After the following flow counters API refactoring:
("net/mlx5: Use flow counter IDs and not the wrapping cache object")
flow counters private data structures mlx5_fc_cache and mlx5_fc are
redundantly exposed in fs_core.h, they have nothing to do with flow
steering core and they are private to fs_counter.c, this patch moves them
to where they belong and reduces their exposure in the driver.

Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agoIB/mlx5: Use helper to get CQE opcode
Tariq Toukan [Wed, 5 Dec 2018 02:03:02 +0000 (18:03 -0800)]
IB/mlx5: Use helper to get CQE opcode

Use the new helper that extracts the opcode
from a CQE (completion queue entry) structure.

Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Acked-by: Leon Romanovsky <leonro@mellanox.com>
5 years agonet/mlx5: Use helper to get CQE opcode
Tariq Toukan [Wed, 5 Dec 2018 02:03:01 +0000 (18:03 -0800)]
net/mlx5: Use helper to get CQE opcode

Introduce and use a helper that extracts the opcode
from a CQE (completion queue entry) structure.

Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
5 years agonet/mlx5: When fetching CQEs return CQE instead of void pointer
Daniel Jurgens [Wed, 5 Dec 2018 02:03:00 +0000 (18:03 -0800)]
net/mlx5: When fetching CQEs return CQE instead of void pointer

The function is only used to retrieve CQEs, use the proper type as the
return value.

Signed-off-by: Daniel Jurgens <danielj@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>