Florian Westphal [Wed, 26 Feb 2020 09:14:49 +0000 (10:14 +0100)]
mptcp: add rmem queue accounting
If userspace never drains the receive buffers we must stop draining
the subflow socket(s) at some point.
This adds the needed rmem accouting for this.
If the threshold is reached, we stop draining the subflows.
Signed-off-by: Florian Westphal <fw@strlen.de>
Reviewed-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Florian Westphal [Wed, 26 Feb 2020 09:14:48 +0000 (10:14 +0100)]
mptcp: update mptcp ack sequence from work queue
If userspace is not reading data, all the mptcp-level acks contain the
ack_seq from the last time userspace read data rather than the most
recent in-sequence value.
This causes pointless retransmissions for data that is already queued.
The reason for this is that all the mptcp protocol level processing
happens at mptcp_recv time.
This adds work queue to move skbs from the subflow sockets receive
queue on the mptcp socket receive queue (which was not used so far).
This allows us to announce the correct mptcp ack sequence in a timely
fashion, even when the application does not call recv() on the mptcp socket
for some time.
We still wake userspace tasks waiting for POLLIN immediately:
If the mptcp level receive queue is empty (because the work queue is
still pending) it can be filled from in-sequence subflow sockets at
recv time without a need to wait for the worker.
The skb_orphan when moving skbs from subflow to mptcp level is needed,
because the destructor (sock_rfree) relies on skb->sk (ssk!) lock
being taken.
A followup patch will add needed rmem accouting for the moved skbs.
Other problem: In case application behaves as expected, and calls
recv() as soon as mptcp socket becomes readable, the work queue will
only waste cpu cycles. This will also be addressed in followup patches.
Signed-off-by: Florian Westphal <fw@strlen.de>
Reviewed-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Paolo Abeni [Wed, 26 Feb 2020 09:14:47 +0000 (10:14 +0100)]
mptcp: add work queue skeleton
Will be extended with functionality in followup patches.
Initial user is moving skbs from subflows receive queue to
the mptcp-level receive queue.
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
Reviewed-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Florian Westphal [Wed, 26 Feb 2020 09:14:46 +0000 (10:14 +0100)]
mptcp: add and use mptcp_data_ready helper
allows us to schedule the work queue to drain the ssk receive queue in
a followup patch.
This is needed to avoid sending all-to-pessimistic mptcp-level
acknowledgements. At this time, the ack_seq is what was last read by
userspace instead of the highest in-sequence number queued for reading.
Signed-off-by: Florian Westphal <fw@strlen.de>
Reviewed-by: Mat Martineau <mathew.j.martineau@linux.intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Thu, 27 Feb 2020 04:44:42 +0000 (20:44 -0800)]
Merge branch 'mlxsw-Small-driver-update'
Jiri Pirko says:
====================
mlxsw: Small driver update
This patchset contains couple of patches not related to each other. They
are small optimization and extension changes to the driver.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Petr Machata [Wed, 26 Feb 2020 08:39:20 +0000 (09:39 +0100)]
mlxsw: spectrum: Add mlxsw_sp_span_ops.buffsize_get for Spectrum-3
The buffer factor on Spectrum-3 is larger than on Spectrum-2. Add a new
callback and use it for mlxsw_sp->span_ops on Spectrum-3.
Signed-off-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Ido Schimmel [Wed, 26 Feb 2020 08:39:19 +0000 (09:39 +0100)]
mlxsw: spectrum: Initialize advertised speeds to supported speeds
During port initialization the driver instructs the device to only
advertise speeds that can be supported by the port's current width.
Since the device now returns the supported speeds based on the port's
current width, the driver no longer needs to compute the speeds that can
be advertised.
Simplify port initialization by setting the advertised speeds to the
queried supported speeds.
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Petr Machata [Wed, 26 Feb 2020 08:39:18 +0000 (09:39 +0100)]
mlxsw: spectrum: Move the ECN-marked packet counter to ethtool
Spectrum-1 and Spectrum-2 do not have a per-TC counter of number of packets
marked by ECN. The value reported currently is the total number of marked
packets. Showing this value at individual TC Qdiscs is misleading.
Move the counter to ethtool instead.
Signed-off-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jiri Pirko [Wed, 26 Feb 2020 08:39:17 +0000 (09:39 +0100)]
mlxsw: spectrum_switchdev: Optimize SFN records processing
Currently, only one SFN query is done from repetitive work at a time,
processing 64 entries. Another work iteration is scheduled in 100ms,
that means that the max rate of learned FDB entries is limited to 6400/s.
That is slow. Fix this by doing 2 optimizations:
1) Run 10 SFN queries at a time.
2) In case the SFN is not drained, schedule work with 0 delay to allow
to continue processing rest of the records.
On a testing setup with 500K entries the time to process decreased
from 870secs to 10secs.
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Tested-by: Alex Kushnarov <alexanderk@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Randy Dunlap [Wed, 26 Feb 2020 05:08:52 +0000 (21:08 -0800)]
af_llc: fix if-statement empty body warning
When debugging via dprintk() is not enabled, make the dprintk()
macro be an empty do-while loop, as is done in
<linux/sunrpc/debug.h>.
This fixes a gcc warning when -Wextra is set:
../net/llc/af_llc.c:974:51: warning: suggest braces around empty body in an ‘if’ statement [-Wempty-body]
I have verified that there is not object code change (with gcc 7.5.0).
Signed-off-by: Randy Dunlap <rdunlap@infradead.org>
Cc: netdev@vger.kernel.org
Cc: "David S. Miller" <davem@davemloft.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Thu, 27 Feb 2020 04:30:43 +0000 (20:30 -0800)]
Merge tag 'mlx5-updates-2020-02-25' of git://git./linux/kernel/git/saeed/linux
Saeed Mahameed says:
====================
mlx5-updates-2020-02-25
The following series provides some misc updates to mlx5 driver:
1) From Maxim, Refactoring for mlx5e netdev channels recreation flow.
- Add error handling
- Add context to the preactivate hook
- Use preactivate hook with context where it can be used
and subsequently unify channel recreation flow everywhere.
- Fix XPS cpumask to not reset upon channel recreation.
2) From Tariq:
- Use indirect calls wrapper on RX.
- Check LRO capability bit
3) Multiple small cleanups
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Thu, 27 Feb 2020 04:27:06 +0000 (20:27 -0800)]
Merge branch 'net-smc-improve-peer-ID-in-CLC-decline'
Hans Wippel says:
====================
net/smc: improve peer ID in CLC decline
The following two patches improve the peer ID in CLC decline messages if
RoCE devices are present in the host but no suitable device is found for
a connection. The first patch reworks the peer ID initialization. The
second patch contains the actual changes of the CLC decline messages.
Changes v1 -> v2:
* make smc_ib_is_valid_local_systemid() static in first patch
* changed if in smc_clc_send_decline() to remove curly braces
Changes RFC -> v1:
* split the patch into two parts
* removed zero assignment to global variable (thanks Leon)
Thanks to Leon Romanovsky and Karsten Graul for the feedback!
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Hans Wippel [Tue, 25 Feb 2020 21:41:22 +0000 (22:41 +0100)]
net/smc: improve peer ID in CLC decline for SMC-R
According to RFC 7609, all CLC messages contain a peer ID that consists
of a unique instance ID and the MAC address of one of the host's RoCE
devices. But if a SMC-R connection cannot be established, e.g., because
no matching pnet table entry is found, the current implementation uses a
zero value in the CLC decline message although the host's peer ID is set
to a proper value.
If no RoCE and no ISM device is usable for a connection, there is no LGR
and the LGR check in smc_clc_send_decline() prevents that the peer ID is
copied into the CLC decline message for both SMC-D and SMC-R. So, this
patch modifies the check to also accept the case of no LGR. Also, only a
valid peer ID is copied into the decline message.
Signed-off-by: Hans Wippel <ndev@hwipl.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Hans Wippel [Tue, 25 Feb 2020 21:41:21 +0000 (22:41 +0100)]
net/smc: rework peer ID handling
This patch initializes the peer ID to a random instance ID and a zero
MAC address. If a RoCE device is in the host, the MAC address part of
the peer ID is overwritten with the respective address. Also, a function
for checking if the peer ID is valid is added. A peer ID is considered
valid if the MAC address part contains a non-zero MAC address.
Signed-off-by: Hans Wippel <ndev@hwipl.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
Arjun Roy [Tue, 25 Feb 2020 20:38:54 +0000 (12:38 -0800)]
tcp-zerocopy: Update returned getsockopt() optlen.
TCP receive zerocopy currently does not update the returned optlen for
getsockopt() if the user passed in a larger than expected value.
Thus, userspace cannot properly determine if all the fields are set in
the passed-in struct. This patch sets the optlen for this case before
returning, in keeping with the expected operation of getsockopt().
Fixes:
c8856c051454 ("tcp-zerocopy: Return inq along with tcp receive zerocopy.")
Signed-off-by: Arjun Roy <arjunroy@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Soheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Thu, 27 Feb 2020 04:07:26 +0000 (20:07 -0800)]
Merge branch 'net-fix-sysfs-permssions-when-device-changes-network'
Christian Brauner says:
====================
net: fix sysfs permssions when device changes network
/* v7 */
This is v7 with a build warning fixup that slipped past me the last
time. It removes to unused variables in sysfs_group_change_owner(). I
observed no warning when building just now.
/* v6 */
This is v6 with two small fixups. I missed adapting the commit message
to reflect the renamed helper for changing the owner of sysfs files and
I also forgot to make the new dpm helper static inline.
/* v5 */
This is v5 with a small fixup requested by Rafael.
/* v4 */
This is v4 with more documentation and other fixes that Greg requested.
/* v3 */
This is v3 with explicit uid and gid parameters added to functions that
change sysfs object ownership as Greg requested.
(I've tagged this with net-next since it's triggered by a bug for
network device files but it also touches driver core aspects so it's
not clear-cut. I can of course split this series into separate
patchsets.)
We have been struggling with a bug surrounding the ownership of network
device sysfs files when moving network devices between network
namespaces owned by different user namespaces reported by multiple
users.
Currently, when moving network devices between network namespaces the
ownership of the corresponding sysfs entries is not changed. This leads
to problems when tools try to operate on the corresponding sysfs files.
I also causes a bug when creating a network device in a network
namespaces owned by a user namespace and moving that network device back
to the host network namespaces. Because when a network device is created
in a network namespaces it will be owned by the root user of the user
namespace and all its associated sysfs files will also be owned by the
root user of the corresponding user namespace.
If such a network device has to be moved back to the host network
namespace the permissions will still be set to the root user of the
owning user namespaces of the originating network namespace. This means
unprivileged users can e.g. re-trigger uevents for such incorrectly
owned devices on the host or in other network namespaces. They can also
modify the settings of the device itself through sysfs when they
wouldn't be able to do the same through netlink. Both of these things
are unwanted.
For example, quite a few workloads will create network devices in the
host network namespace. Other tools will then proceed to move such
devices between network namespaces owner by other user namespaces. While
the ownership of the device itself is updated in
net/core/net-sysfs.c:dev_change_net_namespace() the corresponding sysfs
entry for the device is not. Below you'll find that moving a network
device (here a veth device) from a network namespace into another
network namespaces owned by a different user namespace with a different
id mapping. As you can see the permissions are wrong even though it is
owned by the userns root user after it has been moved and can be
interacted with through netlink:
drwxr-xr-x 5 nobody nobody 0 Jan 25 18:08 .
drwxr-xr-x 9 nobody nobody 0 Jan 25 18:08 ..
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 addr_assign_type
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 addr_len
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 address
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 broadcast
-rw-r--r-- 1 nobody nobody 4096 Jan 25 18:09 carrier
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 carrier_changes
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 carrier_down_count
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 carrier_up_count
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 dev_id
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 dev_port
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 dormant
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 duplex
-rw-r--r-- 1 nobody nobody 4096 Jan 25 18:09 flags
-rw-r--r-- 1 nobody nobody 4096 Jan 25 18:09 gro_flush_timeout
-rw-r--r-- 1 nobody nobody 4096 Jan 25 18:09 ifalias
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 ifindex
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 iflink
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 link_mode
-rw-r--r-- 1 nobody nobody 4096 Jan 25 18:09 mtu
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 name_assign_type
-rw-r--r-- 1 nobody nobody 4096 Jan 25 18:09 netdev_group
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 operstate
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 phys_port_id
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 phys_port_name
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 phys_switch_id
drwxr-xr-x 2 nobody nobody 0 Jan 25 18:09 power
-rw-r--r-- 1 nobody nobody 4096 Jan 25 18:09 proto_down
drwxr-xr-x 4 nobody nobody 0 Jan 25 18:09 queues
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 speed
drwxr-xr-x 2 nobody nobody 0 Jan 25 18:09 statistics
lrwxrwxrwx 1 nobody nobody 0 Jan 25 18:08 subsystem -> ../../../../class/net
-rw-r--r-- 1 nobody nobody 4096 Jan 25 18:09 tx_queue_len
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 type
-rw-r--r-- 1 nobody nobody 4096 Jan 25 18:08 uevent
Constrast this with creating a device of the same type in the network
namespace directly. In this case the device's sysfs permissions will be
correctly updated.
(Please also note, that in a lot of workloads this strategy of creating
the network device directly in the network device to workaround this
issue can not be used. Either because the network device is dedicated
after it has been created or because it used by a process that is
heavily sandboxed and couldn't create network devices itself.):
drwxr-xr-x 5 root root 0 Jan 25 18:12 .
drwxr-xr-x 9 nobody nobody 0 Jan 25 18:08 ..
-r--r--r-- 1 root root 4096 Jan 25 18:12 addr_assign_type
-r--r--r-- 1 root root 4096 Jan 25 18:12 addr_len
-r--r--r-- 1 root root 4096 Jan 25 18:12 address
-r--r--r-- 1 root root 4096 Jan 25 18:12 broadcast
-rw-r--r-- 1 root root 4096 Jan 25 18:12 carrier
-r--r--r-- 1 root root 4096 Jan 25 18:12 carrier_changes
-r--r--r-- 1 root root 4096 Jan 25 18:12 carrier_down_count
-r--r--r-- 1 root root 4096 Jan 25 18:12 carrier_up_count
-r--r--r-- 1 root root 4096 Jan 25 18:12 dev_id
-r--r--r-- 1 root root 4096 Jan 25 18:12 dev_port
-r--r--r-- 1 root root 4096 Jan 25 18:12 dormant
-r--r--r-- 1 root root 4096 Jan 25 18:12 duplex
-rw-r--r-- 1 root root 4096 Jan 25 18:12 flags
-rw-r--r-- 1 root root 4096 Jan 25 18:12 gro_flush_timeout
-rw-r--r-- 1 root root 4096 Jan 25 18:12 ifalias
-r--r--r-- 1 root root 4096 Jan 25 18:12 ifindex
-r--r--r-- 1 root root 4096 Jan 25 18:12 iflink
-r--r--r-- 1 root root 4096 Jan 25 18:12 link_mode
-rw-r--r-- 1 root root 4096 Jan 25 18:12 mtu
-r--r--r-- 1 root root 4096 Jan 25 18:12 name_assign_type
-rw-r--r-- 1 root root 4096 Jan 25 18:12 netdev_group
-r--r--r-- 1 root root 4096 Jan 25 18:12 operstate
-r--r--r-- 1 root root 4096 Jan 25 18:12 phys_port_id
-r--r--r-- 1 root root 4096 Jan 25 18:12 phys_port_name
-r--r--r-- 1 root root 4096 Jan 25 18:12 phys_switch_id
drwxr-xr-x 2 root root 0 Jan 25 18:12 power
-rw-r--r-- 1 root root 4096 Jan 25 18:12 proto_down
drwxr-xr-x 4 root root 0 Jan 25 18:12 queues
-r--r--r-- 1 root root 4096 Jan 25 18:12 speed
drwxr-xr-x 2 root root 0 Jan 25 18:12 statistics
lrwxrwxrwx 1 nobody nobody 0 Jan 25 18:12 subsystem -> ../../../../class/net
-rw-r--r-- 1 root root 4096 Jan 25 18:12 tx_queue_len
-r--r--r-- 1 root root 4096 Jan 25 18:12 type
-rw-r--r-- 1 root root 4096 Jan 25 18:12 uevent
Now, when creating a network device in a network namespace owned by a
user namespace and moving it to the host the permissions will be set to
the id that the user namespace root user has been mapped to on the host
leading to all sorts of permission issues mentioned above:
458752
drwxr-xr-x 5 458752 458752 0 Jan 25 18:12 .
drwxr-xr-x 9 root root 0 Jan 25 18:08 ..
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 addr_assign_type
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 addr_len
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 address
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 broadcast
-rw-r--r-- 1 458752 458752 4096 Jan 25 18:12 carrier
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 carrier_changes
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 carrier_down_count
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 carrier_up_count
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 dev_id
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 dev_port
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 dormant
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 duplex
-rw-r--r-- 1 458752 458752 4096 Jan 25 18:12 flags
-rw-r--r-- 1 458752 458752 4096 Jan 25 18:12 gro_flush_timeout
-rw-r--r-- 1 458752 458752 4096 Jan 25 18:12 ifalias
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 ifindex
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 iflink
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 link_mode
-rw-r--r-- 1 458752 458752 4096 Jan 25 18:12 mtu
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 name_assign_type
-rw-r--r-- 1 458752 458752 4096 Jan 25 18:12 netdev_group
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 operstate
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 phys_port_id
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 phys_port_name
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 phys_switch_id
drwxr-xr-x 2 458752 458752 0 Jan 25 18:12 power
-rw-r--r-- 1 458752 458752 4096 Jan 25 18:12 proto_down
drwxr-xr-x 4 458752 458752 0 Jan 25 18:12 queues
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 speed
drwxr-xr-x 2 458752 458752 0 Jan 25 18:12 statistics
lrwxrwxrwx 1 root root 0 Jan 25 18:12 subsystem -> ../../../../class/net
-rw-r--r-- 1 458752 458752 4096 Jan 25 18:12 tx_queue_len
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 type
-rw-r--r-- 1 458752 458752 4096 Jan 25 18:12 uevent
Fix this by changing the basic sysfs files associated with network
devices when moving them between network namespaces. To this end we add
some infrastructure to sysfs.
The patchset takes care to only do this when the owning user namespaces
changes and the kids differ. So there's only a performance overhead,
when the owning user namespace of the network namespace is different
__and__ the kid mappings for the root user are different for the two
user namespaces:
Assume we have a netdev eth0 which we create in netns1 owned by userns1.
userns1 has an id mapping of 0 100000 100000. Now we move eth0 into
netns2 which is owned by userns2 which also defines an id mapping of 0
100000 100000. In this case sysfs doesn't need updating. The patch will
handle this case and not do any needless work. Now assume eth0 is moved
into netns3 which is owned by userns3 which defines an id mapping of 0
123456 65536. In this case the root user in each namespace corresponds
to different kid and sysfs needs updating.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Christian Brauner [Thu, 27 Feb 2020 03:37:19 +0000 (04:37 +0100)]
net: fix sysfs permssions when device changes network namespace
Now that we moved all the helpers in place and make use netdev_change_owner()
to fixup the permissions when moving network devices between network
namespaces.
Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Christian Brauner [Thu, 27 Feb 2020 03:37:18 +0000 (04:37 +0100)]
net-sysfs: add queue_change_owner()
Add a function to change the owner of the queue entries for a network device
when it is moved between network namespaces.
Currently, when moving network devices between network namespaces the
ownership of the corresponding queue sysfs entries are not changed. This leads
to problems when tools try to operate on the corresponding sysfs files. Fix
this.
Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Christian Brauner [Thu, 27 Feb 2020 03:37:17 +0000 (04:37 +0100)]
net-sysfs: add netdev_change_owner()
Add a function to change the owner of a network device when it is moved
between network namespaces.
Currently, when moving network devices between network namespaces the
ownership of the corresponding sysfs entries is not changed. This leads
to problems when tools try to operate on the corresponding sysfs files.
This leads to a bug whereby a network device that is created in a
network namespaces owned by a user namespace will have its corresponding
sysfs entry owned by the root user of the corresponding user namespace.
If such a network device has to be moved back to the host network
namespace the permissions will still be set to the user namespaces. This
means unprivileged users can e.g. trigger uevents for such incorrectly
owned devices. They can also modify the settings of the device itself.
Both of these things are unwanted.
For example, workloads will create network devices in the host network
namespace. Other tools will then proceed to move such devices between
network namespaces owner by other user namespaces. While the ownership
of the device itself is updated in
net/core/net-sysfs.c:dev_change_net_namespace() the corresponding sysfs
entry for the device is not:
drwxr-xr-x 5 nobody nobody 0 Jan 25 18:08 .
drwxr-xr-x 9 nobody nobody 0 Jan 25 18:08 ..
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 addr_assign_type
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 addr_len
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 address
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 broadcast
-rw-r--r-- 1 nobody nobody 4096 Jan 25 18:09 carrier
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 carrier_changes
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 carrier_down_count
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 carrier_up_count
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 dev_id
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 dev_port
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 dormant
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 duplex
-rw-r--r-- 1 nobody nobody 4096 Jan 25 18:09 flags
-rw-r--r-- 1 nobody nobody 4096 Jan 25 18:09 gro_flush_timeout
-rw-r--r-- 1 nobody nobody 4096 Jan 25 18:09 ifalias
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 ifindex
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 iflink
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 link_mode
-rw-r--r-- 1 nobody nobody 4096 Jan 25 18:09 mtu
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 name_assign_type
-rw-r--r-- 1 nobody nobody 4096 Jan 25 18:09 netdev_group
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 operstate
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 phys_port_id
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 phys_port_name
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 phys_switch_id
drwxr-xr-x 2 nobody nobody 0 Jan 25 18:09 power
-rw-r--r-- 1 nobody nobody 4096 Jan 25 18:09 proto_down
drwxr-xr-x 4 nobody nobody 0 Jan 25 18:09 queues
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 speed
drwxr-xr-x 2 nobody nobody 0 Jan 25 18:09 statistics
lrwxrwxrwx 1 nobody nobody 0 Jan 25 18:08 subsystem -> ../../../../class/net
-rw-r--r-- 1 nobody nobody 4096 Jan 25 18:09 tx_queue_len
-r--r--r-- 1 nobody nobody 4096 Jan 25 18:09 type
-rw-r--r-- 1 nobody nobody 4096 Jan 25 18:08 uevent
However, if a device is created directly in the network namespace then
the device's sysfs permissions will be correctly updated:
drwxr-xr-x 5 root root 0 Jan 25 18:12 .
drwxr-xr-x 9 nobody nobody 0 Jan 25 18:08 ..
-r--r--r-- 1 root root 4096 Jan 25 18:12 addr_assign_type
-r--r--r-- 1 root root 4096 Jan 25 18:12 addr_len
-r--r--r-- 1 root root 4096 Jan 25 18:12 address
-r--r--r-- 1 root root 4096 Jan 25 18:12 broadcast
-rw-r--r-- 1 root root 4096 Jan 25 18:12 carrier
-r--r--r-- 1 root root 4096 Jan 25 18:12 carrier_changes
-r--r--r-- 1 root root 4096 Jan 25 18:12 carrier_down_count
-r--r--r-- 1 root root 4096 Jan 25 18:12 carrier_up_count
-r--r--r-- 1 root root 4096 Jan 25 18:12 dev_id
-r--r--r-- 1 root root 4096 Jan 25 18:12 dev_port
-r--r--r-- 1 root root 4096 Jan 25 18:12 dormant
-r--r--r-- 1 root root 4096 Jan 25 18:12 duplex
-rw-r--r-- 1 root root 4096 Jan 25 18:12 flags
-rw-r--r-- 1 root root 4096 Jan 25 18:12 gro_flush_timeout
-rw-r--r-- 1 root root 4096 Jan 25 18:12 ifalias
-r--r--r-- 1 root root 4096 Jan 25 18:12 ifindex
-r--r--r-- 1 root root 4096 Jan 25 18:12 iflink
-r--r--r-- 1 root root 4096 Jan 25 18:12 link_mode
-rw-r--r-- 1 root root 4096 Jan 25 18:12 mtu
-r--r--r-- 1 root root 4096 Jan 25 18:12 name_assign_type
-rw-r--r-- 1 root root 4096 Jan 25 18:12 netdev_group
-r--r--r-- 1 root root 4096 Jan 25 18:12 operstate
-r--r--r-- 1 root root 4096 Jan 25 18:12 phys_port_id
-r--r--r-- 1 root root 4096 Jan 25 18:12 phys_port_name
-r--r--r-- 1 root root 4096 Jan 25 18:12 phys_switch_id
drwxr-xr-x 2 root root 0 Jan 25 18:12 power
-rw-r--r-- 1 root root 4096 Jan 25 18:12 proto_down
drwxr-xr-x 4 root root 0 Jan 25 18:12 queues
-r--r--r-- 1 root root 4096 Jan 25 18:12 speed
drwxr-xr-x 2 root root 0 Jan 25 18:12 statistics
lrwxrwxrwx 1 nobody nobody 0 Jan 25 18:12 subsystem -> ../../../../class/net
-rw-r--r-- 1 root root 4096 Jan 25 18:12 tx_queue_len
-r--r--r-- 1 root root 4096 Jan 25 18:12 type
-rw-r--r-- 1 root root 4096 Jan 25 18:12 uevent
Now, when creating a network device in a network namespace owned by a
user namespace and moving it to the host the permissions will be set to
the id that the user namespace root user has been mapped to on the host
leading to all sorts of permission issues:
458752
drwxr-xr-x 5 458752 458752 0 Jan 25 18:12 .
drwxr-xr-x 9 root root 0 Jan 25 18:08 ..
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 addr_assign_type
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 addr_len
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 address
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 broadcast
-rw-r--r-- 1 458752 458752 4096 Jan 25 18:12 carrier
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 carrier_changes
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 carrier_down_count
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 carrier_up_count
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 dev_id
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 dev_port
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 dormant
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 duplex
-rw-r--r-- 1 458752 458752 4096 Jan 25 18:12 flags
-rw-r--r-- 1 458752 458752 4096 Jan 25 18:12 gro_flush_timeout
-rw-r--r-- 1 458752 458752 4096 Jan 25 18:12 ifalias
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 ifindex
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 iflink
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 link_mode
-rw-r--r-- 1 458752 458752 4096 Jan 25 18:12 mtu
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 name_assign_type
-rw-r--r-- 1 458752 458752 4096 Jan 25 18:12 netdev_group
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 operstate
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 phys_port_id
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 phys_port_name
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 phys_switch_id
drwxr-xr-x 2 458752 458752 0 Jan 25 18:12 power
-rw-r--r-- 1 458752 458752 4096 Jan 25 18:12 proto_down
drwxr-xr-x 4 458752 458752 0 Jan 25 18:12 queues
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 speed
drwxr-xr-x 2 458752 458752 0 Jan 25 18:12 statistics
lrwxrwxrwx 1 root root 0 Jan 25 18:12 subsystem -> ../../../../class/net
-rw-r--r-- 1 458752 458752 4096 Jan 25 18:12 tx_queue_len
-r--r--r-- 1 458752 458752 4096 Jan 25 18:12 type
-rw-r--r-- 1 458752 458752 4096 Jan 25 18:12 uevent
Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Christian Brauner [Thu, 27 Feb 2020 03:37:16 +0000 (04:37 +0100)]
drivers/base/power: add dpm_sysfs_change_owner()
Add a helper to change the owner of a device's power entries. This
needs to happen when the ownership of a device is changed, e.g. when
moving network devices between network namespaces.
This function will be used to correctly account for ownership changes,
e.g. when moving network devices between network namespaces.
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Reviewed-by: "Rafael J. Wysocki" <rafael@kernel.org>
Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Christian Brauner [Thu, 27 Feb 2020 03:37:15 +0000 (04:37 +0100)]
device: add device_change_owner()
Add a helper to change the owner of a device's sysfs entries. This
needs to happen when the ownership of a device is changed, e.g. when
moving network devices between network namespaces.
This function will be used to correctly account for ownership changes,
e.g. when moving network devices between network namespaces.
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Christian Brauner [Thu, 27 Feb 2020 03:37:14 +0000 (04:37 +0100)]
sysfs: add sysfs_change_owner()
Add a helper to change the owner of sysfs objects.
This function will be used to correctly account for kobject ownership
changes, e.g. when moving network devices between network namespaces.
This mirrors how a kobject is added through driver core which in its guts is
done via kobject_add_internal() which in summary creates the main directory via
create_dir(), populates that directory with the groups associated with the
ktype of the kobject (if any) and populates the directory with the basic
attributes associated with the ktype of the kobject (if any). These are the
basic steps that are associated with adding a kobject in sysfs.
Any additional properties are added by the specific subsystem itself (not by
driver core) after it has registered the device. So for the example of network
devices, a network device will e.g. register a queue subdirectory under the
basic sysfs directory for the network device and than further subdirectories
within that queues subdirectory. But that is all specific to network devices
and they call the corresponding sysfs functions to do that directly when they
create those queue objects. So anything that a subsystem adds outside of what
driver core does must also be changed by it (That's already true for removal of
files it created outside of driver core.) and it's the same for ownership
changes.
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Christian Brauner [Thu, 27 Feb 2020 03:37:13 +0000 (04:37 +0100)]
sysfs: add sysfs_group{s}_change_owner()
Add helpers to change the owner of sysfs groups.
This function will be used to correctly account for kobject ownership
changes, e.g. when moving network devices between network namespaces.
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Christian Brauner [Thu, 27 Feb 2020 03:37:12 +0000 (04:37 +0100)]
sysfs: add sysfs_link_change_owner()
Add a helper to change the owner of a sysfs link.
This function will be used to correctly account for kobject ownership
changes, e.g. when moving network devices between network namespaces.
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Christian Brauner [Thu, 27 Feb 2020 03:37:11 +0000 (04:37 +0100)]
sysfs: add sysfs_file_change_owner()
Add helpers to change the owner of a sysfs files.
This function will be used to correctly account for kobject ownership
changes, e.g. when moving network devices between network namespaces.
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: Christian Brauner <christian.brauner@ubuntu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Gustavo A. R. Silva [Tue, 25 Feb 2020 00:18:26 +0000 (18:18 -0600)]
net: cisco: Replace zero-length array with flexible-array member
The current codebase makes use of the zero-length array language
extension to the C90 standard, but the preferred mechanism to declare
variable-length types such as these ones is a flexible array member[1][2],
introduced in C99:
struct foo {
int stuff;
struct boo array[];
};
By making use of the mechanism above, we will get a compiler warning
in case the flexible array does not occur last in the structure, which
will help us prevent some kind of undefined behavior bugs from being
inadvertently introduced[3] to the codebase from now on.
Also, notice that, dynamic memory allocations won't be affected by
this change:
"Flexible array members have incomplete type, and so the sizeof operator
may not be applied. As a quirk of the original implementation of
zero-length arrays, sizeof evaluates to zero."[1]
Lastly, fix the following checkpatch warning:
CHECK: Prefer kernel type 'u32' over 'u_int32_t'
#61: FILE: drivers/net/ethernet/cisco/enic/vnic_devcmd.h:653:
+ u_int32_t val[];
This issue was found with the help of Coccinelle.
[1] https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html
[2] https://github.com/KSPP/linux/issues/21
[3] commit
76497732932f ("cxgb3/l2t: Fix undefined behaviour")
Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Gustavo A. R. Silva [Tue, 25 Feb 2020 00:12:13 +0000 (18:12 -0600)]
net: marvell: Replace zero-length array with flexible-array member
The current codebase makes use of the zero-length array language
extension to the C90 standard, but the preferred mechanism to declare
variable-length types such as these ones is a flexible array member[1][2],
introduced in C99:
struct foo {
int stuff;
struct boo array[];
};
By making use of the mechanism above, we will get a compiler warning
in case the flexible array does not occur last in the structure, which
will help us prevent some kind of undefined behavior bugs from being
inadvertently introduced[3] to the codebase from now on.
Also, notice that, dynamic memory allocations won't be affected by
this change:
"Flexible array members have incomplete type, and so the sizeof operator
may not be applied. As a quirk of the original implementation of
zero-length arrays, sizeof evaluates to zero."[1]
This issue was found with the help of Coccinelle.
[1] https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html
[2] https://github.com/KSPP/linux/issues/21
[3] commit
76497732932f ("cxgb3/l2t: Fix undefined behaviour")
Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Gustavo A. R. Silva [Tue, 25 Feb 2020 00:09:39 +0000 (18:09 -0600)]
net: hns: Replace zero-length array with flexible-array member
The current codebase makes use of the zero-length array language
extension to the C90 standard, but the preferred mechanism to declare
variable-length types such as these ones is a flexible array member[1][2],
introduced in C99:
struct foo {
int stuff;
struct boo array[];
};
By making use of the mechanism above, we will get a compiler warning
in case the flexible array does not occur last in the structure, which
will help us prevent some kind of undefined behavior bugs from being
inadvertently introduced[3] to the codebase from now on.
Also, notice that, dynamic memory allocations won't be affected by
this change:
"Flexible array members have incomplete type, and so the sizeof operator
may not be applied. As a quirk of the original implementation of
zero-length arrays, sizeof evaluates to zero."[1]
This issue was found with the help of Coccinelle.
[1] https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html
[2] https://github.com/KSPP/linux/issues/21
[3] commit
76497732932f ("cxgb3/l2t: Fix undefined behaviour")
Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Gustavo A. R. Silva [Tue, 25 Feb 2020 00:06:47 +0000 (18:06 -0600)]
sfc: Replace zero-length array with flexible-array member
The current codebase makes use of the zero-length array language
extension to the C90 standard, but the preferred mechanism to declare
variable-length types such as these ones is a flexible array member[1][2],
introduced in C99:
struct foo {
int stuff;
struct boo array[];
};
By making use of the mechanism above, we will get a compiler warning
in case the flexible array does not occur last in the structure, which
will help us prevent some kind of undefined behavior bugs from being
inadvertently introduced[3] to the codebase from now on.
Also, notice that, dynamic memory allocations won't be affected by
this change:
"Flexible array members have incomplete type, and so the sizeof operator
may not be applied. As a quirk of the original implementation of
zero-length arrays, sizeof evaluates to zero."[1]
This issue was found with the help of Coccinelle.
[1] https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html
[2] https://github.com/KSPP/linux/issues/21
[3] commit
76497732932f ("cxgb3/l2t: Fix undefined behaviour")
Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
Acked-by: Martin Habets <mhabets@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Gustavo A. R. Silva [Tue, 25 Feb 2020 00:03:55 +0000 (18:03 -0600)]
qlogic: Replace zero-length array with flexible-array member
The current codebase makes use of the zero-length array language
extension to the C90 standard, but the preferred mechanism to declare
variable-length types such as these ones is a flexible array member[1][2],
introduced in C99:
struct foo {
int stuff;
struct boo array[];
};
By making use of the mechanism above, we will get a compiler warning
in case the flexible array does not occur last in the structure, which
will help us prevent some kind of undefined behavior bugs from being
inadvertently introduced[3] to the codebase from now on.
Also, notice that, dynamic memory allocations won't be affected by
this change:
"Flexible array members have incomplete type, and so the sizeof operator
may not be applied. As a quirk of the original implementation of
zero-length arrays, sizeof evaluates to zero."[1]
This issue was detected with the help of Coccinelle.
[1] https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html
[2] https://github.com/KSPP/linux/issues/21
[3] commit
76497732932f ("cxgb3/l2t: Fix undefined behaviour")
Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Florian Fainelli [Mon, 24 Feb 2020 23:44:26 +0000 (15:44 -0800)]
Revert "net: dsa: bcm_sf2: Also configure Port 5 for 2Gb/sec on 7278"
This reverts commit
7458bd540fa0a90220b9e8c349d910d9dde9caf8 ("net: dsa:
bcm_sf2: Also configure Port 5 for 2Gb/sec on 7278") as it causes
advanced congestion buffering issues with 7278 switch devices when using
their internal Giabit PHY. While this is being debugged, continue with
conservative defaults that work and do not cause packet loss.
Fixes:
7458bd540fa0 ("net: dsa: bcm_sf2: Also configure Port 5 for 2Gb/sec on 7278")
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Vivien Didelot <vivien.didelot@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jiri Pirko [Wed, 26 Feb 2020 08:21:57 +0000 (09:21 +0100)]
iavf: use tc_cls_can_offload_and_chain0() instead of chain check
Looks like the iavf code actually experienced a race condition, when a
developer took code before the check for chain 0 was put to helper.
So use tc_cls_can_offload_and_chain0() helper instead of direct check and
move the check to _cb() so this is similar to i40e code.
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Saeed Mahameed [Wed, 22 Jan 2020 23:06:35 +0000 (15:06 -0800)]
net/mlx5: sparse: warning: Using plain integer as NULL pointer
Return NULL instead of 0.
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Reviewed-by: Moshe Shemesh <moshe@mellanox.com>
Saeed Mahameed [Wed, 22 Jan 2020 23:05:21 +0000 (15:05 -0800)]
net/mlx5: sparse: warning: incorrect type in assignment
drivers/net/ethernet/mellanox/mlx5/core/diag/fw_tracer.c:191:13:
sparse: warning: incorrect type in assignment (different base types)
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Reviewed-by: Moshe Shemesh <moshe@mellanox.com>
Nathan Chancellor [Fri, 21 Feb 2020 05:24:37 +0000 (22:24 -0700)]
net/mlx5: Fix header guard in rsc_dump.h
Clang warns:
In file included from
../drivers/net/ethernet/mellanox/mlx5/core/main.c:73:
../drivers/net/ethernet/mellanox/mlx5/core/diag/rsc_dump.h:4:9: warning:
'__MLX5_RSC_DUMP_H' is used as a header guard here, followed by #define
of a different macro [-Wheader-guard]
#ifndef __MLX5_RSC_DUMP_H
^~~~~~~~~~~~~~~~~
../drivers/net/ethernet/mellanox/mlx5/core/diag/rsc_dump.h:5:9: note:
'__MLX5_RSC_DUMP__H' is defined here; did you mean '__MLX5_RSC_DUMP_H'?
#define __MLX5_RSC_DUMP__H
^~~~~~~~~~~~~~~~~~
__MLX5_RSC_DUMP_H
1 warning generated.
Make them match to get the intended behavior and remove the warning.
Fixes:
12206b17235a ("net/mlx5: Add support for resource dump")
Link: https://github.com/ClangBuiltLinux/linux/issues/897
Signed-off-by: Nathan Chancellor <natechancellor@gmail.com>
Reviewed-by: Jason Gunthorpe <jgg@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Hans Wippel [Tue, 18 Feb 2020 19:52:59 +0000 (20:52 +0100)]
Documentation: fix vxlan typo in mlx5.rst
Fix a vxlan typo in the mlx5 driver documentation.
Signed-off-by: Hans Wippel <ndev@hwipl.net>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Tariq Toukan [Mon, 27 Jan 2020 11:34:31 +0000 (13:34 +0200)]
net/mlx5e: RX, Use indirect calls wrapper for handling compressed completions
We can avoid an indirect call per compressed completion wrapping the
completion handling call with the appropriate helper.
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Reviewed-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Tariq Toukan [Mon, 27 Jan 2020 11:28:42 +0000 (13:28 +0200)]
net/mlx5e: RX, Use indirect calls wrapper for posting descriptors
We can avoid an indirect call per NAPI cycle wrapping the RX descriptors
posting call with the appropriate helper.
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Reviewed-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Maxim Mikityanskiy [Thu, 14 Nov 2019 11:06:16 +0000 (13:06 +0200)]
net/mlx5e: Change inline mode correctly when changing trust state
The current steps that are performed when the trust state changes, if
the channels are active:
1. The trust state is changed in hardware.
2. The new inline mode is calculated.
3. If the new inline mode is different, the channels are recreated using
the new inline mode.
This approach has some issues:
1. There is a time gap between changing trust state in hardware and
starting sending enough inline headers (the latter happens after
recreation of channels). It leads to failed transmissions and error
CQEs.
2. If the new channels fail to open, we'll be left with the old ones,
but the hardware will be configured for the new trust state, so the
interval when we can see TX errors never ends.
This patch fixes the issues above by moving the trust state change into
the preactivate hook that runs during the recreation of the channels
when no channels are active, so it eliminates the gap of partially
applied configuration. If the inline mode doesn't change with the change
of the trust state, the channels won't be recreated, just like before
this patch.
Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Maxim Mikityanskiy [Mon, 2 Dec 2019 13:48:25 +0000 (15:48 +0200)]
net/mlx5e: Add context to the preactivate hook
Sometimes the preactivate hook of mlx5e_safe_switch_channels needs more
parameters than just struct mlx5e_priv *. For such cases, a new
parameter (void *context) is added to preactivate hooks.
Some of the existing normal functions are currently used as preactivate
callbacks. To avoid adding an extra unused parameter, they are wrapped
in an automatic way using the MLX5E_DEFINE_PREACTIVATE_WRAPPER_CTX
macro.
Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Maxim Mikityanskiy [Wed, 13 Nov 2019 16:07:29 +0000 (18:07 +0200)]
net/mlx5e: Allow mlx5e_switch_priv_channels to fail and recover
Currently mlx5e_switch_priv_channels expects that the preactivate hook
doesn't fail, however, it can fail, because it may set hardware
parameters. This commit addresses this issue and provides a way to
recover from failures of the preactivate hook: the old channels are not
closed until the point where nothing can fail anymore, so in case
preactivate fails, the driver can roll back the old channels and
activate them again.
Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Maxim Mikityanskiy [Mon, 25 Nov 2019 12:29:46 +0000 (14:29 +0200)]
net/mlx5e: Remove unneeded netif_set_real_num_tx_queues
The number of queues is now updated by mlx5e_update_netdev_queues in a
centralized way, when no channels are active. Remove an extra occurrence
of netif_set_real_num_tx_queues to prepare it for the next commit.
Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Maxim Mikityanskiy [Tue, 3 Sep 2019 14:55:45 +0000 (17:55 +0300)]
net/mlx5e: Fix configuration of XPS cpumasks and netdev queues in corner cases
Currently, mlx5e notifies the kernel about the number of queues and sets
the default XPS cpumasks when channels are activated. This
implementation has several corner cases, in which the kernel may not be
updated on time, or XPS cpumasks may be reset when not directly touched
by the user.
This commit fixes these corner cases to match the following expected
behavior:
1. The number of queues always corresponds to the number of channels
configured.
2. XPS cpumasks are set to driver's defaults on netdev attach.
3. XPS cpumasks set by user are not reset, unless the number of channels
changes. If the number of channels changes, they are reset to driver's
defaults. (In general case, when the number of channels increases or
decreases, it's not possible to guess how to convert the current XPS
cpumasks to work with the new number of channels, so we let the user
reconfigure it if they change the number of channels.)
XPS cpumasks are no longer stored per channel. Only one temporary
cpumask is used. The old stored cpumasks didn't reflect the user's
changes and were not used after applying them.
A scratchpad area is added to struct mlx5e_priv. As cpumask_var_t
requires allocation, and the preactivate hook can't fail, we need to
preallocate the temporary cpumask in advance. It's stored in the
scratchpad.
Fixes:
149e566fef81 ("net/mlx5e: Expand XPS cpumask to cover all online cpus")
Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Maxim Mikityanskiy [Mon, 4 Nov 2019 10:02:14 +0000 (12:02 +0200)]
net/mlx5e: Use preactivate hook to set the indirection table
mlx5e_ethtool_set_channels updates the indirection table before
switching to the new channels. If the switch fails, the indirection
table is new, but the channels are old, which is wrong. Fix it by using
the preactivate hook of mlx5e_safe_switch_channels to update the
indirection table at the stage when nothing can fail anymore.
As the code that updates the indirection table is now encapsulated into
a new function, use that function in the attach flow when the driver has
to reduce the number of channels, and prepare the code for the next
commit.
Fixes:
85082dba0a ("net/mlx5e: Correctly handle RSS indirection table when changing number of channels")
Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Maxim Mikityanskiy [Thu, 31 Oct 2019 07:39:34 +0000 (09:39 +0200)]
net/mlx5e: Rename hw_modify to preactivate
mlx5e_safe_switch_channels accepts a callback to be called before
activating new channels. It is intended to configure some hardware
parameters in cases where channels are recreated because some
configuration has changed.
Recently, this callback has started being used to update the driver's
internal MLX5E_STATE_XDP_OPEN flag, and the following patches also
intend to use this callback for software preparations. This patch
renames the hw_modify callback to preactivate, so that the name fits
better.
Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Maxim Mikityanskiy [Tue, 3 Sep 2019 14:38:43 +0000 (17:38 +0300)]
net/mlx5e: Encapsulate updating netdev queues into a function
As a preparation for one of the following commits, create a function to
encapsulate the code that notifies the kernel about the new amount of
RX and TX queues. The code will be called multiple times in the next
commit.
Signed-off-by: Maxim Mikityanskiy <maximmi@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Tariq Toukan [Thu, 2 Jan 2020 14:17:41 +0000 (16:17 +0200)]
net/mlx5e: Add missing LRO cap check
The LRO boolean state in params->lro_en must not be set in case
the NIC is not capable.
Enforce this check and remove the TODO comment.
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Eran Ben Elisha [Sun, 8 Dec 2019 12:29:45 +0000 (14:29 +0200)]
net/mlx5e: Define one flow for TXQ selection when TCs are configured
We shall always extract channel index out of the txq, regardless
of the relation between txq_ix and num channels. The extraction is
always valid, as if txq is smaller than number of channels,
txq_ix == priv->txq2sq[txq_ix]->ch_ix.
By doing so, we can remove an if clause from the select queue method,
and have one flow for all packets.
Signed-off-by: Eran Ben Elisha <eranbe@mellanox.com>
Reviewed-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
David S. Miller [Tue, 25 Feb 2020 19:05:55 +0000 (11:05 -0800)]
Merge branch 'mlxsw-Implement-ACL-dropped-packets-identification'
Jiri Pirko says:
====================
mlxsw: Implement ACL-dropped packets identification
mlxsw hardware allows to insert a ACL-drop action with a value defined
by user that would be later on passed with a dropped packet.
To implement this, use the existing TC action cookie and pass it to the
driver. As the cookie format coming down from TC and the mlxsw HW cookie
format is different, do the mapping of these two using idr and rhashtable.
The cookie is passed up from the HW through devlink_trap_report() to
drop_monitor code. A new metadata type is used for that.
Example:
$ tc qdisc add dev enp0s16np1 clsact
$ tc filter add dev enp0s16np1 ingress protocol ip pref 10 flower skip_sw dst_ip 192.168.1.2 action drop cookie
3b45fa38c8
^^^^^^^^^^
$ devlink trap set pci/0000:00:10.0 trap acl action trap
$ dropwatch
Initializing null lookup method
dropwatch> set hw true
setting hardware drops monitoring to 1
dropwatch> set alertmode packet
Setting alert mode
Alert mode successfully set
dropwatch> start
Enabling monitoring...
Kernel monitoring activated.
Issue Ctrl-C to stop monitoring
drop at: ingress_flow_action_drop (acl_drops)
origin: hardware
input port ifindex: 30
input port name: enp0s16np1
cookie:
3b45fa38c8 <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
timestamp: Fri Jan 24 17:10:53 2020
715387671 nsec
protocol: 0x800
length: 98
original length: 98
This way the user may insert multiple drop rules and monitor the dropped
packets with the information of which action caused the drop.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Jiri Pirko [Tue, 25 Feb 2020 10:45:27 +0000 (11:45 +0100)]
selftests: netdevsim: Extend devlink trap test to include flow action cookie
Extend existing devlink trap test to include metadata type for flow
action cookie.
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jiri Pirko [Tue, 25 Feb 2020 10:45:26 +0000 (11:45 +0100)]
netdevsim: add ACL trap reporting cookie as a metadata
Add new trap ACL which reports flow action cookie in a metadata. Allow
used to setup the cookie using debugfs file.
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jiri Pirko [Tue, 25 Feb 2020 10:45:25 +0000 (11:45 +0100)]
mlxsw: spectrum_trap: Lookup and pass cookie down to devlink_trap_report()
Use the cookie index received along with the packet to lookup original
flow_offload cookie binary and pass it down to devlink_trap_report().
Add "fa_cookie" metadata to the ACL trap.
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jiri Pirko [Tue, 25 Feb 2020 10:45:24 +0000 (11:45 +0100)]
mlxsw: pci: Extract cookie index for ACL discard trap packets
In case the received packet comes in due to one of ACL discard traps,
take the user_def_val_orig_pkt_len field from CQE and store it
in skb->cb as ACL cookie index.
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jiri Pirko [Tue, 25 Feb 2020 10:45:23 +0000 (11:45 +0100)]
mlxsw: core_acl_flex_actions: Implement flow_offload action cookie offload
Track cookies coming down to driver by flow_offload.
Assign a cookie_index to each unique cookie binary. Use previously
defined "Trap with userdef" flex action to ask HW to pass cookie_index
alongside with the dropped packets.
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jiri Pirko [Tue, 25 Feb 2020 10:45:22 +0000 (11:45 +0100)]
mlxsw: core_acl_flex_actions: Add trap with userdef action
Expose "Trap action with userdef". It is the same as already
defined "Trap action" with a difference that it would ask the policy
engine to pass arbitrary value (userdef) alongside with received packets.
This would be later on used to carry cookie index.
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jiri Pirko [Tue, 25 Feb 2020 10:45:21 +0000 (11:45 +0100)]
devlink: extend devlink_trap_report() to accept cookie and pass
Add cookie argument to devlink_trap_report() allowing driver to pass in
the user cookie. Pass on the cookie down to drop monitor code.
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jiri Pirko [Tue, 25 Feb 2020 10:45:20 +0000 (11:45 +0100)]
drop_monitor: extend by passing cookie from driver
If driver passed along the cookie, push it through Netlink.
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jiri Pirko [Tue, 25 Feb 2020 10:45:19 +0000 (11:45 +0100)]
devlink: add trap metadata type for cookie
Allow driver to indicate cookie metadata for registered traps.
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jiri Pirko [Tue, 25 Feb 2020 10:45:18 +0000 (11:45 +0100)]
flow_offload: pass action cookie through offload structures
Extend struct flow_action_entry in order to hold TC action cookie
specified by user inserting the action.
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: Ido Schimmel <idosch@mellanox.com>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Mon, 24 Feb 2020 23:41:54 +0000 (15:41 -0800)]
Merge tag 'mac80211-next-for-net-next-2020-02-24' of git://git./linux/kernel/git/jberg/mac80211-next
Johannes Berg says:
====================
A new set of changes:
* lots of small documentation fixes, from Jérôme Pouiller
* beacon protection (BIGTK) support from Jouni Malinen
* some initial code for TID configuration, from Tamizh chelvam
* I reverted some new API before it's actually used, because
it's wrong to mix controlled port and preauth
* a few other cleanups/fixes
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Colin Ian King [Mon, 24 Feb 2020 17:35:53 +0000 (17:35 +0000)]
net: qrtr: fix spelling mistake "serivce" -> "service"
There is a spelling mistake in a pr_err message. Fix it.
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Ahmad Fatoum [Mon, 24 Feb 2020 17:29:55 +0000 (18:29 +0100)]
net: ethernet: stmmac: don't warn about missing optional wakeup IRQ
The "stm32_pwr_wakeup" is optional per the binding and the driver
handles its absence gracefully. Request it with
platform_get_irq_byname_optional, so its absence doesn't needlessly
clutter the log.
Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
Ahmad Fatoum [Mon, 24 Feb 2020 17:29:54 +0000 (18:29 +0100)]
net: ethernet: stmmac: demote warnings about missing optional clocks
The specification of a "eth-ck" and a "ptp_ref" clock is optional per
the binding and the driver handles them gracefully.
Demote the output to an info message accordingly.
Signed-off-by: Ahmad Fatoum <a.fatoum@pengutronix.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Mon, 24 Feb 2020 23:31:23 +0000 (15:31 -0800)]
Merge branch 'Add-ACPI-bindings-to-the-genet'
Jeremy Linton says:
====================
Add ACPI bindings to the genet
This patch series allows the BCM GENET, as used on the RPi4,
to attach when booted in an ACPI environment. The DSDT entry to
trigger this is seen below. Of note, the first patch adds a
small extension to the mdio layer which allows drivers to find
the mii_bus without firmware assistance. The fifth patch in
the set retrieves the MAC address from the umac registers
rather than carrying it directly in the DSDT. This of course
requires the firmware to pre-program it, so we continue to fall
back on a random one if it appears to be garbage.
v1 -> v2:
fail on missing phy-mode property
replace phy-mode internal property read string with
device_get_phy_mode() equivalent
rework mac address detection logic so that it merges
the acpi/DT case into device_get_mac_address()
allowing _DSD mac address properties.
some commit messages justifying why phy_find_first()
isn't the worst choice for this driver.
+ Device (ETH0)
+ {
+ Name (_HID, "BCM6E4E")
+ Name (_UID, 0)
+ Name (_CCA, 0x0)
+ Method (_STA)
+ {
+ Return (0xf)
+ }
+ Method (_CRS, 0x0, Serialized)
+ {
+ Name (RBUF, ResourceTemplate ()
+ {
+ Memory32Fixed (ReadWrite, 0xFd580000, 0x10000, )
+ Interrupt (ResourceConsumer, Level, ActiveHigh, Exclusive) { 0xBD }
+ Interrupt (ResourceConsumer, Level, ActiveHigh, Exclusive) { 0xBE }
+ })
+ Return (RBUF)
+ }
+ Name (_DSD, Package () {
+ ToUUID("
daffd814-6eba-4d8c-8a91-
bc9bbf4aa301"),
+ Package () {
+ Package () { "phy-mode", "rgmii-rxid" },
+ }
+ })
+ }
====================
Tested-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jeremy Linton [Mon, 24 Feb 2020 22:54:03 +0000 (16:54 -0600)]
net: bcmgenet: reduce severity of missing clock warnings
If one types "failed to get enet clock" or similar into google
there are ~370k hits. The vast majority are people debugging
problems unrelated to this adapter, or bragging about their
rpi's. Further, the DT clock bindings here are optional.
Given that its not a fatal situation with common DT based
systems, lets reduce the severity so people aren't seeing failure
messages in everyday operation.
Signed-off-by: Jeremy Linton <jeremy.linton@arm.com>
Reviewed-by: Nicolas Saenz Julienne <nsaenzjulienne@suse.de>
Acked-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jeremy Linton [Mon, 24 Feb 2020 22:54:02 +0000 (16:54 -0600)]
net: bcmgenet: Fetch MAC address from the adapter
ARM/ACPI machines should utilize self describing hardware
when possible. The MAC address on the BCMGENET can be
read from the adapter if a full featured firmware has already
programmed it. Lets try using the address already programmed,
if it appears to be valid.
It should be noted that while we move the macaddr logic below
the clock and power logic in the driver, none of that code will
ever be active in an ACPI environment as the device will be
attached to the acpi power domain, and brought to full power
with all clocks enabled immediately before the device probe
routine is called.
One side effect of the above tweak is that while its now
possible to read the MAC address via _DSD properties, it should
be avoided.
Signed-off-by: Jeremy Linton <jeremy.linton@arm.com>
Acked-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jeremy Linton [Mon, 24 Feb 2020 22:54:01 +0000 (16:54 -0600)]
net: bcmgenet: Initial bcmgenet ACPI support
The rpi4 is capable of booting in ACPI mode with the latest
edk2-platform commits. As such it would be helpful if the genet
platform device were usable.
To achieve this we add a new MODULE_DEVICE_TABLE, and convert
a few dt specific methods to their generic device_ calls. Until
the next patch, ACPI based machines will fallback on random
mac addresses.
Signed-off-by: Jeremy Linton <jeremy.linton@arm.com>
Acked-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jeremy Linton [Mon, 24 Feb 2020 22:54:00 +0000 (16:54 -0600)]
net: bcmgenet: enable automatic phy discovery
The unimac mdio driver falls back to scanning the
entire bus if its given an appropriate mask. In ACPI
mode we expect that the system is well behaved and
conforms to recent versions of the specification.
We then utilize phy_find_first(), and
phy_connect_direct() to find and attach to the
discovered phy during net_device open. While its
apparently possible to build a genet based device
with multiple phys on a single mdio bus, this works
for current machines. Further, this driver makes
a number of assumptions about the platform device,
mac, mdio and phy all being 1:1. Lastly, It also
avoids having to create references across the ACPI
namespace hierarchy.
Signed-off-by: Jeremy Linton <jeremy.linton@arm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jeremy Linton [Mon, 24 Feb 2020 22:53:59 +0000 (16:53 -0600)]
net: bcmgenet: refactor phy mode configuration
The DT phy mode is similar to what we want for ACPI
lets factor it out of the of path, and change the
of_ call to device_.
Signed-off-by: Jeremy Linton <jeremy.linton@arm.com>
Acked-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jeremy Linton [Mon, 24 Feb 2020 22:53:58 +0000 (16:53 -0600)]
mdio_bus: Add generic mdio_find_bus()
It appears most ethernet drivers follow one of two main strategies
for mdio bus/phy management. A monolithic model where the net driver
itself creates, probes and uses the phy, and one where an external
mdio/phy driver instantiates the mdio bus/phy and the net driver
only attaches to a known phy. Usually in this latter model the phys
are discovered via DT relationships or simply phy name/address
hardcoding.
This is a shame because modern well behaved mdio buses are self
describing and can be probed. The mdio layer itself is fully capable
of this, yet there isn't a clean way for a standalone net driver
to attach and enumerate the discovered devices. This is because
outside of of_mdio_find_bus() there isn't a straightforward way
to acquire the mii_bus pointer.
So, lets add a mdio_find_bus which can return the mii_bus based
only on its name.
Signed-off-by: Jeremy Linton <jeremy.linton@arm.com>
Acked-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Gustavo A. R. Silva [Mon, 24 Feb 2020 16:43:46 +0000 (10:43 -0600)]
freescale: Replace zero-length array with flexible-array member
The current codebase makes use of the zero-length array language
extension to the C90 standard, but the preferred mechanism to declare
variable-length types such as these ones is a flexible array member[1][2],
introduced in C99:
struct foo {
int stuff;
struct boo array[];
};
By making use of the mechanism above, we will get a compiler warning
in case the flexible array does not occur last in the structure, which
will help us prevent some kind of undefined behavior bugs from being
inadvertently introduced[3] to the codebase from now on.
Also, notice that, dynamic memory allocations won't be affected by
this change:
"Flexible array members have incomplete type, and so the sizeof operator
may not be applied. As a quirk of the original implementation of
zero-length arrays, sizeof evaluates to zero."[1]
This issue was found with the help of Coccinelle.
[1] https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html
[2] https://github.com/KSPP/linux/issues/21
[3] commit
76497732932f ("cxgb3/l2t: Fix undefined behaviour")
Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Gustavo A. R. Silva [Mon, 24 Feb 2020 16:41:06 +0000 (10:41 -0600)]
intel: Replace zero-length array with flexible-array member
The current codebase makes use of the zero-length array language
extension to the C90 standard, but the preferred mechanism to declare
variable-length types such as these ones is a flexible array member[1][2],
introduced in C99:
struct foo {
int stuff;
struct boo array[];
};
By making use of the mechanism above, we will get a compiler warning
in case the flexible array does not occur last in the structure, which
will help us prevent some kind of undefined behavior bugs from being
inadvertently introduced[3] to the codebase from now on.
Also, notice that, dynamic memory allocations won't be affected by
this change:
"Flexible array members have incomplete type, and so the sizeof operator
may not be applied. As a quirk of the original implementation of
zero-length arrays, sizeof evaluates to zero."[1]
This issue was found with the help of Coccinelle.
[1] https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html
[2] https://github.com/KSPP/linux/issues/21
[3] commit
76497732932f ("cxgb3/l2t: Fix undefined behaviour")
Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Gustavo A. R. Silva [Mon, 24 Feb 2020 16:36:52 +0000 (10:36 -0600)]
netronome: Replace zero-length array with flexible-array member
The current codebase makes use of the zero-length array language
extension to the C90 standard, but the preferred mechanism to declare
variable-length types such as these ones is a flexible array member[1][2],
introduced in C99:
struct foo {
int stuff;
struct boo array[];
};
By making use of the mechanism above, we will get a compiler warning
in case the flexible array does not occur last in the structure, which
will help us prevent some kind of undefined behavior bugs from being
inadvertently introduced[3] to the codebase from now on.
Also, notice that, dynamic memory allocations won't be affected by
this change:
"Flexible array members have incomplete type, and so the sizeof operator
may not be applied. As a quirk of the original implementation of
zero-length arrays, sizeof evaluates to zero."[1]
This issue was found with the help of Coccinelle.
[1] https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html
[2] https://github.com/KSPP/linux/issues/21
[3] commit
76497732932f ("cxgb3/l2t: Fix undefined behaviour")
Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Gustavo A. R. Silva [Mon, 24 Feb 2020 16:32:52 +0000 (10:32 -0600)]
toshiba: Replace zero-length array with flexible-array member
The current codebase makes use of the zero-length array language
extension to the C90 standard, but the preferred mechanism to declare
variable-length types such as these ones is a flexible array member[1][2],
introduced in C99:
struct foo {
int stuff;
struct boo array[];
};
By making use of the mechanism above, we will get a compiler warning
in case the flexible array does not occur last in the structure, which
will help us prevent some kind of undefined behavior bugs from being
inadvertently introduced[3] to the codebase from now on.
Also, notice that, dynamic memory allocations won't be affected by
this change:
"Flexible array members have incomplete type, and so the sizeof operator
may not be applied. As a quirk of the original implementation of
zero-length arrays, sizeof evaluates to zero."[1]
This issue was found with the help of Coccinelle.
[1] https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html
[2] https://github.com/KSPP/linux/issues/21
[3] commit
76497732932f ("cxgb3/l2t: Fix undefined behaviour")
Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Gustavo A. R. Silva [Mon, 24 Feb 2020 16:30:24 +0000 (10:30 -0600)]
chelsio: Replace zero-length array with flexible-array member
The current codebase makes use of the zero-length array language
extension to the C90 standard, but the preferred mechanism to declare
variable-length types such as these ones is a flexible array member[1][2],
introduced in C99:
struct foo {
int stuff;
struct boo array[];
};
By making use of the mechanism above, we will get a compiler warning
in case the flexible array does not occur last in the structure, which
will help us prevent some kind of undefined behavior bugs from being
inadvertently introduced[3] to the codebase from now on.
Also, notice that, dynamic memory allocations won't be affected by
this change:
"Flexible array members have incomplete type, and so the sizeof operator
may not be applied. As a quirk of the original implementation of
zero-length arrays, sizeof evaluates to zero."[1]
This issue was found with the help of Coccinelle.
[1] https://gcc.gnu.org/onlinedocs/gcc/Zero-Length.html
[2] https://github.com/KSPP/linux/issues/21
[3] commit
76497732932f ("cxgb3/l2t: Fix undefined behaviour")
Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Mon, 24 Feb 2020 23:12:11 +0000 (15:12 -0800)]
Merge branch 'Remainder-for-DT-bindings-for-Felix-DSA-switch-on-LS1028A'
Vladimir Oltean says:
====================
Remainder for "DT bindings for Felix DSA switch on LS1028A"
This series is the remainder of patchset [0] which has been merged
through Shawn Guo's devicetree tree.
It contains changes to the PHY mode validation in the Felix driver
("gmii" to "internal") and the documentation for the DT bindings.
[0]: https://patchwork.ozlabs.org/cover/1242716/
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Vladimir Oltean [Mon, 24 Feb 2020 12:15:34 +0000 (14:15 +0200)]
dt-bindings: net: dsa: ocelot: document the vsc9959 core
This patch adds the required documentation for the embedded L2 switch
inside the NXP LS1028A chip.
I've submitted it in the legacy format instead of yaml schema, because
DSA itself has not yet been converted to yaml, and this driver defines
no custom bindings.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
Vladimir Oltean [Mon, 24 Feb 2020 12:15:33 +0000 (14:15 +0200)]
net: dsa: felix: Use PHY_INTERFACE_MODE_INTERNAL instead of GMII
phy-mode = "gmii" is confusing because it may mean that the port
supports the 8-bit-wide parallel data interface pinout, which it
doesn't.
It may also be confusing because one of the "gmii" internal ports is
actually overclocked to run at 2.5Gbps (even though, yes, as far as the
switch MAC is concerned, it still thinks it's gigabit).
So use the phy-mode = "internal" property to describe the internal ports
inside the NXP LS1028A chip (the ones facing the ENETC). The change
should be fine, because the device tree bindings document is yet to be
introduced, and there are no stable DT blobs in use.
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Tested-by: Michael Walle <michael@walle.cc>
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Mon, 24 Feb 2020 22:40:54 +0000 (14:40 -0800)]
bareudp: Fix uninitialized variable warnings.
drivers/net/bareudp.c: In function 'bareudp_xmit_skb':
drivers/net/bareudp.c:346:9: warning: 'err' may be used uninitialized in this function [-Wmaybe-uninitialized]
346 | return err;
| ^~~
drivers/net/bareudp.c: In function 'bareudp6_xmit_skb':
drivers/net/bareudp.c:407:9: warning: 'err' may be used uninitialized in this function [-Wmaybe-uninitialized]
407 | return err;
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Mon, 24 Feb 2020 21:31:42 +0000 (13:31 -0800)]
Merge branch 'Bare-UDP-L3-Encapsulation-Module'
Martin Varghese says:
====================
Bare UDP L3 Encapsulation Module
There are various L3 encapsulation standards using UDP being discussed to
leverage the UDP based load balancing capability of different networks.
MPLSoUDP (__ https://tools.ietf.org/html/rfc7510) is one among them.
The Bareudp tunnel module provides a generic L3 encapsulation tunnelling
support for tunnelling different L3 protocols like MPLS, IP, NSH etc. inside
a UDP tunnel.
Special Handling
----------------
The bareudp device supports special handling for MPLS & IP as they can have
multiple ethertypes.
MPLS procotcol can have ethertypes ETH_P_MPLS_UC (unicast) & ETH_P_MPLS_MC (multicast).
IP protocol can have ethertypes ETH_P_IP (v4) & ETH_P_IPV6 (v6).
This special handling can be enabled only for ethertypes ETH_P_IP & ETH_P_MPLS_UC
with a flag called multiproto mode.
Usage
------
1) Device creation & deletion
a) ip link add dev bareudp0 type bareudp dstport 6635 ethertype 0x8847.
This creates a bareudp tunnel device which tunnels L3 traffic with ethertype
0x8847 (MPLS traffic). The destination port of the UDP header will be set to
6635.The device will listen on UDP port 6635 to receive traffic.
b) ip link delete bareudp0
2) Device creation with multiple proto mode enabled
There are two ways to create a bareudp device for MPLS & IP with multiproto mode
enabled.
a) ip link add dev bareudp0 type bareudp dstport 6635 ethertype 0x8847 multiproto
b) ip link add dev bareudp0 type bareudp dstport 6635 ethertype mpls
3) Device Usage
The bareudp device could be used along with OVS or flower filter in TC.
The OVS or TC flower layer must set the tunnel information in SKB dst field before
sending packet buffer to the bareudp device for transmission. On reception the
bareudp device extracts and stores the tunnel information in SKB dst field before
passing the packet buffer to the network stack.
Why not FOU ?
------------
FOU by design does l4 encapsulation.It maps udp port to ipproto (IP protocol number for l4 protocol).
Bareudp acheives a generic l3 encapsulation.It maps udp port to l3 ethertype.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Martin Varghese [Mon, 24 Feb 2020 05:28:35 +0000 (10:58 +0530)]
net: Special handling for IP & MPLS.
Special handling is needed in bareudp module for IP & MPLS as they
support more than one ethertypes.
MPLS has 2 ethertypes. 0x8847 for MPLS unicast and 0x8848 for MPLS multicast.
While decapsulating MPLS packet from UDP packet the tunnel destination IP
address is checked to determine the ethertype. The ethertype of the packet
will be set to 0x8848 if the tunnel destination IP address is a multicast
IP address. The ethertype of the packet will be set to 0x8847 if the
tunnel destination IP address is a unicast IP address.
IP has 2 ethertypes.0x0800 for IPV4 and 0x86dd for IPv6. The version
field of the IP header tunnelled will be checked to determine the ethertype.
This special handling to tunnel additional ethertypes will be disabled
by default and can be enabled using a flag called multiproto. This flag can
be used only with ethertypes 0x8847 and 0x0800.
Signed-off-by: Martin Varghese <martin.varghese@nokia.com>
Acked-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Martin Varghese [Mon, 24 Feb 2020 05:27:50 +0000 (10:57 +0530)]
net: UDP tunnel encapsulation module for tunnelling different protocols like MPLS, IP, NSH etc.
The Bareudp tunnel module provides a generic L3 encapsulation
tunnelling module for tunnelling different protocols like MPLS,
IP,NSH etc inside a UDP tunnel.
Signed-off-by: Martin Varghese <martin.varghese@nokia.com>
Acked-by: Willem de Bruijn <willemb@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jules Irenge [Sun, 23 Feb 2020 23:16:56 +0000 (23:16 +0000)]
af_unix: Add missing annotation for unix_wait_for_peer()
Sparse reports a warning unix_wait_for_peer()
warning: context imbalance in unix_wait_for_peer() - unexpected unlock
The root cause is the missing annotation at unix_wait_for_peer()
Add the missing annotation __releases(&unix_sk(other)->lock)
Signed-off-by: Jules Irenge <jbi.octave@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jules Irenge [Sun, 23 Feb 2020 23:16:55 +0000 (23:16 +0000)]
dccp: Add missing annotation for dccp_child_process()
Sparse reports a warning at dccp_child_process()
warning: context imbalance in dccp_child_process() - unexpected unlock
The root cause is the missing annotation at dccp_child_process()
Add the missing __releases(child) annotation
Signed-off-by: Jules Irenge <jbi.octave@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jules Irenge [Sun, 23 Feb 2020 23:16:54 +0000 (23:16 +0000)]
net: netrom: Add missing annotation for nr_neigh_stop()
Sparse reports a warning at nr_neigh_stop()
warning: context imbalance in nr_neigh_stop() - unexpected unlock
The root cause is the missing annotation at nr_neigh_stop()
Add the missing __releases(&nr_neigh_list_lock) annotation
Signed-off-by: Jules Irenge <jbi.octave@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jules Irenge [Sun, 23 Feb 2020 23:16:53 +0000 (23:16 +0000)]
net: netrom: Add missing annotation for nr_neigh_start()
Sparse reports a warning at nr_neigh_start()
warning: context imbalance in nr_neigh_start() - wrong count at exit
The root cause is the missing annotation at nr_neigh_start()
Add the missing __acquires(&nr_neigh_list_lock) annotation
Signed-off-by: Jules Irenge <jbi.octave@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jules Irenge [Sun, 23 Feb 2020 23:16:52 +0000 (23:16 +0000)]
net: netrom: Add missing annotation for nr_node_stop()
Sparse reports a warning at nr_node_stop()
warning: context imbalance in nr_node_stop() - wrong count at exit
The root cause is the missing annotation at nr_node_stop()
Add the missing __releases(&nr_node_list_lock) annotation
Signed-off-by: Jules Irenge <jbi.octave@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jules Irenge [Sun, 23 Feb 2020 23:16:51 +0000 (23:16 +0000)]
net: netrom: Add missing annotation for nr_node_start()
Sparse reports a warning at nr_node_start()
warning: context imbalance in nr_node_start() - wrong count at exit
The root cause is the missing annotation at nr_node_start()
Add the missing __acquires(&nr_node_list_lock) annotation
Signed-off-by: Jules Irenge <jbi.octave@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jules Irenge [Sun, 23 Feb 2020 23:16:50 +0000 (23:16 +0000)]
netrom: Add missing annotation for nr_info_stop()
Sparse reports a warning at nr_info_stop()
warning: context imbalance in nr_info_stop() - unexpected unlock
The root cause is the missing annotation at nr_info_stop()
Add the missing __releases(&nr_list_lock)
Signed-off-by: Jules Irenge <jbi.octave@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jules Irenge [Sun, 23 Feb 2020 23:16:49 +0000 (23:16 +0000)]
netrom: Add missing annotation for nr_info_start()
Sparse reports a warning at nr_info_start()
warning: context imbalance in nr_info_start() - wrong count at exit
The root cause is the missing annotation at nr_info_start()
Add the missing __acquires(&nr_list_lock)
Signed-off-by: Jules Irenge <jbi.octave@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jules Irenge [Sun, 23 Feb 2020 23:16:48 +0000 (23:16 +0000)]
net: Add missing annotation for llc_seq_start()
Sparse reports a warning at llc_seq_start()
warning: context imbalance in llc_seq_start() - wrong count at exit
The root cause is the msiing annotation at llc_seq_start()
Add the missing __acquires(RCU) annotation
Signed-off-by: Jules Irenge <jbi.octave@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jules Irenge [Sun, 23 Feb 2020 23:16:47 +0000 (23:16 +0000)]
sctp: Add missing annotation for sctp_transport_walk_stop()
Sparse reports a warning at sctp_transport_walk_stop()
warning: context imbalance in sctp_transport_walk_stop
- wrong count at exit
The root cause is the missing annotation at sctp_transport_walk_stop()
Add the missing __releases(RCU) annotation
Signed-off-by: Jules Irenge <jbi.octave@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jules Irenge [Sun, 23 Feb 2020 23:16:46 +0000 (23:16 +0000)]
sctp: Add missing annotation for sctp_transport_walk_start()
Sparse reports a warning at sctp_transport_walk_start()
warning: context imbalance in sctp_transport_walk_start
- wrong count at exit
The root cause is the missing annotation at sctp_transport_walk_start()
Add the missing __acquires(RCU) annotation
Signed-off-by: Jules Irenge <jbi.octave@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jules Irenge [Sun, 23 Feb 2020 23:16:45 +0000 (23:16 +0000)]
sctp: Add missing annotation for sctp_err_finish()
Sparse reports a warning at sctp_err_finish()
warning: context imbalance in sctp_err_finish() - unexpected unlock
The root cause is a missing annotation at sctp_err_finish()
Add the missing __releases(&((__sk)->sk_lock.slock)) annotation
Signed-off-by: Jules Irenge <jbi.octave@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Amol Grover [Sat, 22 Feb 2020 16:57:27 +0000 (22:27 +0530)]
ip6mr: Fix RCU list debugging warning
ip6mr_for_each_table() macro uses list_for_each_entry_rcu()
for traversing outside an RCU read side critical section
but under the protection of rtnl_mutex. Hence add the
corresponding lockdep expression to silence the following
false-positive warnings:
[ 4.319479] =============================
[ 4.319480] WARNING: suspicious RCU usage
[ 4.319482] 5.5.4-stable #17 Tainted: G E
[ 4.319483] -----------------------------
[ 4.319485] net/ipv6/ip6mr.c:1243 RCU-list traversed in non-reader section!!
[ 4.456831] =============================
[ 4.456832] WARNING: suspicious RCU usage
[ 4.456834] 5.5.4-stable #17 Tainted: G E
[ 4.456835] -----------------------------
[ 4.456837] net/ipv6/ip6mr.c:1582 RCU-list traversed in non-reader section!!
Signed-off-by: Amol Grover <frextrite@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Min Li [Fri, 21 Feb 2020 20:03:51 +0000 (15:03 -0500)]
ptp: Add a ptp clock driver for IDT 82P33 SMU.
The IDT 82P33 Synchronization Management Unit (SMU) family provides
tools to manage timing references, clock sources and
timing paths for IEEE 1588 / Precision Time Protocol (PTP) and
Synchronous Ethernet (SyncE) based clocks. The device supports up
to three independent timing paths that control: PTP clock synthesis;
SyncE clock generation; and general purpose frequency translation.
The device supports physical layer timing with Digital PLLs (DPLLs)
and it supports packet based timing with Digitally Controlled
Oscillators (DCOs). This patch adds support for ptp clock based on
the device.
Changes since v1:
- As suggested by Richard Cochran:
1. Replace _mask_bit_count with the existing hweight8
2. Prefix all functions with idt82p33
3. Fix white space issues in Kconfig and Makefile
4. Remove forward declaration
5. Use adjfine instead of adjfreq for better resolution
- As suggested by David Miller:
1. Replace CHAN_INIT macro with a static function
idt82p33_channel_init
2. Employ reverse christmas tree ordering for local
variables
3. Fix indentation problem by appropriate number of
TAB then SPACE character
Signed-off-by: Min Li <min.li.xe@renesas.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Min Li [Fri, 21 Feb 2020 21:48:38 +0000 (16:48 -0500)]
dt-bindings: ptp: Add device tree binding for IDT 82P33 based PTP clock
Add device tree binding doc for the PTP clock based on IDT 82P33
Synchronization Management Unit (SMU).
Changes since v1:
- As suggested by Rob Herring:
1. Drop reg description for i2c
2. Replace i2c@1 with i2c
3. Add addtionalProperties: false
Signed-off-by: Min Li <min.li.xe@renesas.com>
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Amol Grover [Fri, 21 Feb 2020 17:57:14 +0000 (23:27 +0530)]
tcp: ipv4: Pass lockdep expression to RCU lists
md5sig->head maybe traversed using hlist_for_each_entry_rcu
outside an RCU read-side critical section but under the protection
of socket lock.
Hence, add corresponding lockdep expression to silence false-positive
warnings, and harden RCU lists.
Signed-off-by: Amol Grover <frextrite@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Madhuparna Bhowmik [Fri, 21 Feb 2020 16:19:47 +0000 (21:49 +0530)]
net: 802: psnap.c: Use built-in RCU list checking
list_for_each_entry_rcu() has built-in RCU and lock checking.
Pass cond argument to list_for_each_entry_rcu() to silence
false lockdep warning when CONFIG_PROVE_RCU_LIST is enabled
by default.
Signed-off-by: Madhuparna Bhowmik <madhuparnabhowmik10@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Amol Grover [Fri, 21 Feb 2020 15:15:38 +0000 (20:45 +0530)]
tcp, ulp: Pass lockdep expression to RCU lists
tcp_ulp_list is traversed using list_for_each_entry_rcu
outside an RCU read-side critical section but under the protection
of tcp_ulp_list_lock.
Hence, add corresponding lockdep expression to silence false-positive
warnings, and harden RCU lists.t
Signed-off-by: Amol Grover <frextrite@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>