Commit Graph

737196 Commits

Author SHA1 Message Date
Brandon Streiff
0336369d3a net: dsa: forward hardware timestamping ioctls to switch driver
This patch adds support to the dsa slave network device so that
switch drivers can implement the SIOC[GS]HWTSTAMP ioctls and the
ethtool timestamp-info interface.

Signed-off-by: Brandon Streiff <brandon.streiff@ni.com>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-14 14:33:37 -05:00
Brandon Streiff
4eb3be2937 net: dsa: mv88e6xxx: add support for event capture
This patch adds support for configuring mv88e6xxx GPIO lines as PTP
pins, so that they may be used for time stamping external events or for
periodic output.

Checkpatch and reverse Christmas tree fixes by Andrew Lunn

Periodic output removed by Richard Cochran, until a better abstraction
of a VCO is added to Linux in general.

Signed-off-by: Brandon Streiff <brandon.streiff@ni.com>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-14 14:33:37 -05:00
Brandon Streiff
a73ccd6106 net: dsa: mv88e6xxx: add support for GPIO configuration
MV88E6352 and later switches support GPIO control through the "Scratch
& Misc" global2 register. (Older switches do too, though with a slightly
different register interface. Only the 6352-style is implemented here.)

Add a new file, global2_scratch.c, for operations in the Scratch & Misc
space. Additionally, add a GPIO operations structure to present an
abstract view over GPIO manipulation.

Reverse Christmas tree and unsigned has been replaced with unsigned
int by Andrew Lunn.

Signed-off-by: Brandon Streiff <brandon.streiff@ni.com>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-14 14:33:36 -05:00
Brandon Streiff
2fa8d3af4b net: dsa: mv88e6xxx: expose switch time as a PTP hardware clock
This patch adds basic support for exposing the 32-bit timestamp counter
inside the mv88e6xxx switch as a ptp_clock.

Adjfine implemented by Richard Cochran.
Andrew Lunn: fix return value of PTP stub function.

Signed-off-by: Brandon Streiff <brandon.streiff@ni.com>
Signed-off-by: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-14 14:33:36 -05:00
Brandon Streiff
0d632c3d6f net: dsa: mv88e6xxx: add accessors for PTP/TAI registers
This patch implements support for accessing the Precision Time Protocol
and Time Application Interface registers via the AVB register interface
in the Global 2 register.

The register interface differs slightly between different models; older
models use a 3-bit operations field, while newer models use a 2-bit
field. The operations values and the special "global port" values are
different between the two. This is a similar split to the differences
in the "Ingress Rate" register between models, so, like in that case,
we call the two variants "6352" and "6390" and create an ops structure
to abstract between the two.

checkpatch fixups by Andrew Lunn

Signed-off-by: Brandon Streiff <brandon.streiff@ni.com>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-14 14:33:36 -05:00
Brandon Streiff
b000be95e5 net: dsa: mv88e6xxx: export g2 register accessors
Let the mv88e6xxx_g2_* register accessor functions be accessible
outside of global2.c.

Signed-off-by: Brandon Streiff <brandon.streiff@ni.com>
Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-14 14:33:36 -05:00
Andrew Lunn
052fddfb3c net: ptp: Add stub for ptp_classify_raw()
When NET_PTP_CLASSIFY is disabled, a stub function is required in
order that the drivers compile.

Signed-off-by: Andrew Lunn <andrew@lunn.ch>
Acked-by: Richard Cochran <richardcochran@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-14 14:33:36 -05:00
Eric Dumazet
e0f9759f53 tcp: try to keep packet if SYN_RCV race is lost
배석진 reported that in some situations, packets for a given 5-tuple
end up being processed by different CPUS.

This involves RPS, and fragmentation.

배석진 is seeing packet drops when a SYN_RECV request socket is
moved into ESTABLISH state. Other states are protected by socket lock.

This is caused by a CPU losing the race, and simply not caring enough.

Since this seems to occur frequently, we can do better and perform
a second lookup.

Note that all needed memory barriers are already in the existing code,
thanks to the spin_lock()/spin_unlock() pair in inet_ehash_insert()
and reqsk_put(). The second lookup must find the new socket,
unless it has already been accepted and closed by another cpu.

Note that the fragmentation could be avoided in the first place by
use of a correct TCP MSS option in the SYN{ACK} packet, but this
does not mean we can not be more robust.

Many thanks to 배석진 for a very detailed analysis.

Reported-by: 배석진 <soukjin.bae@samsung.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-14 14:21:45 -05:00
David S. Miller
1d631583ae Merge branch 'net-dev-Make-protocol-ptr-dependent-on-CONFIG'
David Ahern says:

====================
net: dev: Make protocol ptr dependent on CONFIG

Found these in a branch from 3-years ago. Still relevant today.
Make decnet, ax25, and atalk ptrs in net_device based on their
respective CONFIG.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-14 11:55:39 -05:00
David Ahern
e92bad5034 net: Remove atalk header from socket.c
Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-14 11:55:33 -05:00
David Ahern
89e58148fb net: Make atalk_ptr depend on ATALK or IRDA
Signed-off-by: David Ahern <dsa@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-14 11:55:33 -05:00
David Ahern
19ff13f2a4 net: Make ax25_ptr depend on CONFIG_AX25
Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-14 11:55:33 -05:00
David Ahern
330c7272c4 net: Make dn_ptr depend on CONFIG_DECNET
Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-14 11:55:33 -05:00
David S. Miller
7707968361 Merge branch '40GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/next-queue
Jeff Kirsher says:

====================
40GbE Intel Wired LAN Driver Updates 2018-02-13

This series contains updates to i40e and i40evf.

Wei Yongjun fixes a function that needed to be "static".  Also fixes the
use of GFP_KERNEL to GFP_ATOMIC when we have taken a spinlock.

Mitch cleans up several info messages to not include the memory
addresses being used on the off chance this information could be used
maliciously.

Alan provides several fixes to the broadcast filters starting with the
triggering of overflow promiscuous in circumstances where we run out of
space for broadcast filters to prevent traffic from being unexpectedly
dropped.  Refactored the code to improve the readability and
maintainability when we are concerned about when and how overflow
promiscuous is changed.

Harshitha cleans up a message to make it more clear on what is being
reset, so users are not confused and think the PF is resetting.

Dave fixes an issue where the MAC, firmware version and NPAR checks used
to determine if shutting off the firmware LLDP engine is supported or
not, instead set a hardware flag which ethtool can use.

Jake updates the VF driver to use __dev_uc_sync and __dev_mc_sync, like
the PF driver.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-13 22:31:23 -05:00
David Ahern
153e1b84f4 selftests: Add FIB onlink tests
Add test cases verifying FIB onlink commands work as expected in
various conditions - IPv4, IPv6, main table, and VRF.

Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-13 14:47:33 -05:00
Harshitha Ramamurthy
693acdd0f1 i40evf: Make VF reset warning message more clear
When the PF resets the VF, the VF puts out a warning message
indicating that the VF received a reset message from the PF.
Make this message more clear so that we do not mistakenly
think that the PF is undergoing a reset.

Signed-off-by: Harshitha Ramamurthy <harshitha.ramamurthy@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-02-13 11:40:11 -08:00
Jacob Keller
8946b56354 i40evf: use __dev_[um]c_sync routines in .set_rx_mode
Similar to changes done to the PF driver in commit 6622f5cdba ("i40e:
make use of __dev_uc_sync and __dev_mc_sync"), replace our
home-rolled method for updating the internal status of MAC filters with
__dev_uc_sync and __dev_mc_sync.

These new functions use internal state within the netdev struct in order
to efficiently break the question of "which filters in this list need to
be added or removed" into singular "add this filter" and "delete this
filter" requests.

This vastly improves our handling of .set_rx_mode especially with large
number of MAC filters being added to the device, and even results in
a simpler .set_rx_mode handler.

Under some circumstances, such as when attached to a bridge, we may
receive a request to delete our own permanent address. Prevent deletion
of this address during i40evf_addr_unsync so that we don't accidentally
stop receiving traffic.

Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-02-13 11:40:10 -08:00
Dave Ertman
7b63435a50 i40e: i40e: Change ethtool check from MAC to HW flag
The MAC, FW Version and NPAR check used to determine
if shutting off the FW LLDP engine is supported is not
using the usual feature check mechanism.

This patch fixes the problem by moving the feature check
to i40e_sw_init in order to set a flag in pf->hw_features
that ethtool will use for priv_flags disable operation.

Signed-off-by: Dave Ertman <david.m.ertman@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-02-13 11:40:10 -08:00
Alan Brady
7363115efb i40e: do not force filter failure in overflow promiscuous
Broadcast filters can now cause overflow promiscuous to trigger when
adding "too many" VLANs to all the ports of a device and the driver
needs a way to exit overflow promiscuous once triggered.

Currently the driver looks to see if there are "too many" filters and/or
we have any failed filters to determine when it is safe to exit overflow
promiscuous.  If we trigger overflow promiscuous with broadcast filters,
any new filters added will be "auto-failed" until we exit overflow
promiscuous.  Since the user can't manually remove the failed broadcast
filters for VLANs (nor should we expect the user to do such), there is
no way to exit overflow promiscuous without reloading the driver.

The easiest way to do this is to remove the shortcut to "auto-fail"
filters in overflow promiscuous.  If the user removes the VLANs, the
failed filters will be removed and since we're no longer "auto-failing"
new filters, we'll eventually get a good set of filters and exit
overflow promiscuous.

This has the side benefit of making filter state more explicit in that
if a filter says it's failed we know for a fact it failed and not just
assuming it will if we're in overflow promiscuous.  This is nice because
if the user removes some filters and then adds some, even if we're in
overflow promiscuous, the filter might succeed; we were just assuming it
won't because the user hasn't rectified other existing failed filters.

Signed-off-by: Alan Brady <alan.brady@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-02-13 11:40:10 -08:00
Alan Brady
cc6a96a419 i40e: refactor promisc_changed in i40e_sync_vsi_filters
This code here is quite complex and easy to screw up.  Let's see if we
can't improve the readability and maintainability a bit.  This refactors
out promisc_changed into two variables 'old_overflow' and 'new_overflow'
which makes it a bit clearer when we're concerned about when and how
overflow promiscuous is changed.  This also makes so that we no longer
need to pass a boolean pointer to i40e_aqc_add_filters.  Instead we can
simply check if we changed the overflow promiscuous flag since the
function start.

Signed-off-by: Alan Brady <alan.brady@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-02-13 11:40:10 -08:00
Harshitha Ramamurthy
fbd5eb54c2 i40evf: Use an iterator of the same type as the list
When iterating through the linked list of VLAN filters, make the
iterator the same type as that of the linked list.

Signed-off-by: Harshitha Ramamurthy <harshitha.ramamurthy@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-02-13 11:40:10 -08:00
Alan Brady
a48350c29b i40e: broadcast filters can trigger overflow promiscuous
When adding a bunch of VLANs to all the ports on a device, it's possible
to run out of space for broadcast filters.  The driver should trigger
overflow promiscuous in this circumstance to prevent traffic from being
unexpectedly dropped.

Signed-off-by: Alan Brady <alan.brady@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-02-13 11:40:10 -08:00
Mitch Williams
7be78aa444 i40e: don't leak memory addresses
Could a Bad Person do Bad Things to a server if they found these
addresses printed in the log? Who knows? But let's not take that risk.

Remove pointers from a bunch of printks. In some cases, I was able to
adjust the message to indicate whether or not the value was null. In
others, I just removed the entire message as there was really no hope of
saving it.

Signed-off-by: Mitch Williams <mitch.a.williams@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-02-13 11:40:10 -08:00
Wei Yongjun
03f431b33a i40evf: use GFP_ATOMIC under spin lock
A spin lock is taken here so we should use GFP_ATOMIC.

Fixes: 504398f0a7 ("i40evf: use spinlock to protect (mac|vlan)_filter_list")
Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com>
Acked-by: Jacob Keller <jacob.e.keller@intel.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-02-13 11:40:10 -08:00
Wei Yongjun
3758d2c74d i40e: Make local function i40e_get_link_speed static
Fixes the following sparse warning:

drivers/net/ethernet/intel/i40e/i40e_main.c:5440:5: warning:
 symbol 'i40e_get_link_speed' was not declared. Should it be static?

Signed-off-by: Wei Yongjun <weiyongjun1@huawei.com>
Tested-by: Andrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
2018-02-13 11:40:10 -08:00
David S. Miller
b95367e3bd Merge branch 'selftests-fib_tests-simplifications-verbosity-and-a-race'
David Ahern says:

====================
selftests: fib_tests: simplifications, verbosity and a race

Improve efficiency of fib_tests.sh and make the test result more verbose,
from this summary:
$ fib_tests.sh is failing in a VM:
    $ fib_tests.sh
    Running netdev unregister tests
    PASS: unicast route test
    PASS: multipath route test
    Running netdev down tests
    PASS: unicast route test
    PASS: multipath route test
    Running netdev carrier change tests
    PASS: local route carrier test
    FAIL: unicast route carrier test

where a single entry actually corresponds to many checks to a much more
verbse output that clarifies test cases:
$fib_tests.sh
Single path route carrier test
    ....
    Carrier down
        IPv4 fibmatch                                         [ OK ]
        IPv6 fibmatch                                         [ OK ]
        IPv4 linkdown flag set                                [FAIL]
        IPv6 linkdown flag set                                [FAIL]
    Second address added with carrier down
        IPv4 fibmatch                                         [ OK ]
        IPv6 fibmatch                                         [ OK ]
        IPv4 linkdown flag set                                [FAIL]
        IPv6 linkdown flag set                                [ OK ]

And then fix the race in changing carrier down on dummy device to checking
the corresponding routes.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-13 14:34:29 -05:00
David Ahern
e2ba732a16 selftests: fib_tests: sleep after changing carrier
sleep for a second after setting carrier down to allow linkwatch
to propagate the change to the routing stack via netdev_state_change.
As it stands there is a race setting carrier down on the dummy
device and then checking the linkdown flag in the routes.

Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-13 14:34:28 -05:00
David Ahern
ee395a5e72 selftests: fib_tests: Move admin of dummy0 to helpers
Move setup and teardown of testns and dummy0 to helpers.

Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-13 14:34:28 -05:00
David Ahern
1056691b26 selftests: fib_tests: Make test results more verbose
fib_tests.sh is failing in a VM:
    $ fib_tests.sh
    Running netdev unregister tests
    PASS: unicast route test
    PASS: multipath route test
    Running netdev down tests
    PASS: unicast route test
    PASS: multipath route test
    Running netdev carrier change tests
    PASS: local route carrier test
    FAIL: unicast route carrier test

The last test corresponds to fib_carrier_unicast_test which 12 places
that could be failing. Be more verbose in the output so a failure is
easier to track down and separate test setup failures with set -e and
set +e pairs.

With the verbose logging it is easier to see which checks are failing:
    $fib_tests.sh
    Single path route carrier test
        ....
        Carrier down
            IPv4 fibmatch                                         [ OK ]
            IPv6 fibmatch                                         [ OK ]
            IPv4 linkdown flag set                                [FAIL]
            IPv6 linkdown flag set                                [FAIL]
        Second address added with carrier down
            IPv4 fibmatch                                         [ OK ]
            IPv6 fibmatch                                         [ OK ]
            IPv4 linkdown flag set                                [FAIL]
            IPv6 linkdown flag set                                [ OK ]

Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-13 14:34:28 -05:00
David Ahern
07f4e62f1c selftests: fib_tests: simplify ip commands in a namespace
'ip netns exec testns ip' is more efficiently handled using 'ip -netns';
runs the ip command after switching the namespace and avoids an exec.

Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-13 14:34:28 -05:00
David Ahern
8c2ceabe99 net/ipv4: Unexport fib_multipath_hash and fib_select_path
Do not export fib_multipath_hash or fib_select_path; both are only used
by core ipv4 code.

Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-13 14:00:57 -05:00
David Ahern
0d876f2c6d net/ipv4: Simplify fib_select_path
If flow oif is set and it is not an l3mdev, then fib_select_path
can jump to the source address check.

Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-13 14:00:57 -05:00
David S. Miller
3174a94476 Merge branch 'sctp-rename-sctp-diag-file-and-add-file-comments-for-it'
Xin Long says:

====================
sctp: rename sctp diag file and add file comments for it

This patchset is to remove the sctp_ prefix for sctp diag file,
and also to add the missing file comments for it.

v1->v2:
  split them into two patches as Marcelo suggested.
====================

Acked-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-13 13:56:32 -05:00
Xin Long
6f68dc993a sctp: add file comments in diag.c
This patch is to add the missing file comments for sctp diag file.

Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-13 13:56:31 -05:00
Xin Long
ff22b5bf78 sctp: rename sctp_diag.c as diag.c
Remove 'sctp_' prefix for diag file, to keep consistent with other
files' names.

Signed-off-by: Xin Long <lucien.xin@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-13 13:56:31 -05:00
Arkadi Sharshevsky
6c677750f2 mlxsw: spectrum: Use NL_SET_ERR_MSG_MOD
Use NL_SET_ERR_MSG_MOD helper which adds the module name instead
of specifying the prefix each time.

Signed-off-by: Arkadi Sharshevsky <arkadis@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-13 12:27:20 -05:00
David S. Miller
d520d58124 Merge branch 'mlxsw-SPAN-cleanups'
Jiri Pirko says:

====================
mlxsw: SPAN cleanups

In patch one of this short series, a misplaced pointer star is moved to
the correct place.

In the second patch, we observe that if SPAN entries carry their
reference count anyway, it's redundant to also carry a "used" flag.

In the third patch, SPAN support code is moved to a separate module.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-13 12:26:26 -05:00
Petr Machata
a629ef210d mlxsw: spectrum: Move SPAN code to separate module
For the upcoming work on SPAN, it makes sense to move the current code
to a module of its own. It already has a well-defined API boundary to
the mirror management (which is used from matchall and ACL code). A
couple more functions need to be exported for the functions that
spectrum.c needs to use for MTU handling and subsystem init/fini.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-13 12:26:25 -05:00
Petr Machata
ce470b44e2 mlxsw: spectrum: Drop struct span_entry.used
The member ref_count already determines whether a given SPAN entry is
used, and is as easy to use as a dedicated boolean.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-13 12:26:25 -05:00
Petr Machata
306a934e5b mlxsw: spectrum: Fix a coding style nit
Signed-off-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-13 12:26:25 -05:00
David S. Miller
439ccd8ce9 Merge branch 'mlxsw-IPIP-cleanups'
Jiri Pirko says:

====================
mlxsw: IPIP cleanups

In the first patch, a forgotten #include is added. Even though the code
compiles as-is, the include is necessary for modules that should include
spectrum_ipip.h.

The second patch corrects an assumption that IPv6 tunnels use struct
ip_tunnel_parm to store tunnel parameters.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-13 12:24:28 -05:00
Petr Machata
e437f3b62d mlxsw: spectrum: Distinguish between IPv4/6 tunnels
struct ip_tunnel_parm, where GRE and several other tunnel types hold
information, is IPv4-specific. The current router / ipip code in mlxsw
however uses it as if it were generic.

Make it clear that it's not. Rename many functions from _params_ to
_params4_. mlxsw_sp_ipip_parms_saddr() and _daddr() take a proto
argument to dispatch on it. Move the dispatch logic to
mlxsw_sp_ipip_netdev_saddr() and _daddr(), and replace with
single-protocol functions.

In struct mlxsw_sp_ipip_entry, move the "parms" field to a (for the time
being, singleton) union. Update users throughout.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-13 12:24:28 -05:00
Petr Machata
fe735a3d2c mlxsw: spectrum_ipip: Add a forgotten include
struct ip_tunnel_parm, which is used in spectrum_ipip.h, is defined in
if_tunnel.h. However, the former neglects to include the latter.

Signed-off-by: Petr Machata <petrm@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-13 12:24:28 -05:00
Jake Moroni
d75de7b6e7 dpaa_eth: fix incorrect comment
The comment stated that a thread was started, but
that is not the case.

Signed-off-by: Jake Moroni <mail@jakemoroni.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-13 12:06:11 -05:00
David S. Miller
885842d89a Merge branch 'Replacing-net_mutex-with-rw_semaphore'
Kirill Tkhai says:

====================
Replacing net_mutex with rw_semaphore

this is the third version of the patchset introducing net_sem
instead of net_mutex. The patchset adds net_sem in addition
to net_mutex and allows pernet_operations to be "async". This
flag means, the pernet_operations methods are safe to be
executed with any other pernet_operations (un)initializing
another net.

If there are only async pernet_operations in the system,
net_mutex is not used either for setup_net() or for cleanup_net().

The pernet_operations converted in this patchset allow
to create minimal .config to have network working, and
the changes improve the performance like you may see
below:

    %for i in {1..10000}; do unshare -n bash -c exit; done

    *before*
    real 1m40,377s
    user 0m9,672s
    sys 0m19,928s

    *after*
    real 0m17,007s
    user 0m5,311s
    sys 0m11,779

    (5.8 times faster)

In the future, when all pernet_operations become async,
we'll just remove this "async" field tree-wide.

All the new logic is concentrated in patches [1-5/32].
The rest of patches converts specific operations:
review, rationale of they can be converted, and setting
of async flag.

Kirill

v3: Improved patches descriptions. Added comment into [5/32].
Added [32/32] converting netlink_tap_net_ops (new pernet operations
introduced in 2018).

v2: Single patch -> patchset with rationale of every conversion
====================

Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-13 10:36:10 -05:00
Kirill Tkhai
b86b47a395 net: Convert netlink_tap_net_ops
These pernet_operations init just allocated net memory,
and they obviously can be executed in parallel in any
others.

v3: New

Signed-off-by: Kirill Tkhai <ktkhai@virtuozzo.com>
Acked-by: Andrei Vagin <avagin@virtuozzo.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-13 10:36:09 -05:00
Kirill Tkhai
59a513587a net: Convert diag_net_ops
These pernet operations just create and destroy netlink
socket. The socket is pernet and else operations don't
touch it.

Signed-off-by: Kirill Tkhai <ktkhai@virtuozzo.com>
Acked-by: Andrei Vagin <avagin@virtuozzo.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-13 10:36:09 -05:00
Kirill Tkhai
2608e6b7ad net: Convert default_device_ops
These pernet operations consist of exit() and exit_batch() methods.

default_device_exit() moves not-local and virtual devices to init_net.
There is nothing exciting, because this may happen in any time
on a working system, and rtnl_lock() and synchronize_net() protect
us from all cases of external dereference.

The same for default_device_exit_batch(). Similar unregisteration
may happen in any time on a system. Here several lists (like todo_list),
which are accessed under rtnl_lock(). After rtnl_unlock() and
netdev_run_todo() all the devices are flushed.

Signed-off-by: Kirill Tkhai <ktkhai@virtuozzo.com>
Acked-by: Andrei Vagin <avagin@virtuozzo.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-13 10:36:09 -05:00
Kirill Tkhai
9a4d105de7 net: Convert loopback_net_ops
These pernet_operations have only init() method. It allocates
memory for net_device, calls register_netdev() and assigns
net::loopback_dev.

register_netdev() is allowed be used without additional locks,
as it's synchronized on rtnl_lock(). There are many examples
of using this functon directly from ioctl().

The only difference, compared to ioctl(), is that net is not
completely alive at this moment. But it looks like, there is
no way for parallel pernet_operations to dereference
the net_device, as the most of struct net_device lists,
where it's linked, are related to net, and the net is not liked.

The exceptions are net_device::unreg_list, close_list, todo_list,
used for unregistration, and ::link_watch_list, where net_device
may be linked to global lists.

Unregistration of loopback_dev obviously can't happen, when
loopback_net_init() is executing, as the net as alive. It occurs
in default_device_ops, which currently requires net_mutex,
and it behaves as a barrier at the moment. It will be considered
in next patch.

Speaking about link_watch_list, it seems, there is no way
for loopback_dev at time of registration to be linked in lweventlist
and be available for another pernet_operations.

Signed-off-by: Kirill Tkhai <ktkhai@virtuozzo.com>
Acked-by: Andrei Vagin <avagin@virtuozzo.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-13 10:36:09 -05:00
Kirill Tkhai
0bc9be6718 net: Convert addrconf_ops
These pernet_operations (un)register sysctl, which
are not touched by anybody else.

So, it's safe to make them async.

Signed-off-by: Kirill Tkhai <ktkhai@virtuozzo.com>
Acked-by: Andrei Vagin <avagin@virtuozzo.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
2018-02-13 10:36:09 -05:00