Current driver dynamically allocates an skb and maps it as DMA Rx
buffer. In order to prepare for upcoming XDP changes, let's introduce a
different allocation scheme.
Buffers are allocated dynamically and mapped into hardware.
During the Rx operation the driver uses build_skb() to produce the
necessary buffers for the network stack.
This change increases performance ~15% on 64b packets with smmu disabled
and ~5% with smmu enabled
Signed-off-by: Ilias Apalodimas <ilias.apalodimas@linaro.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Stefan Wahren [Thu, 8 Nov 2018 13:38:21 +0000 (14:38 +0100)]
net: qca_spi: Add available buffer space verification
Interferences on the SPI line could distort the response of
available buffer space. So at least we should check that the
response doesn't exceed the maximum available buffer space.
In error case increase a new error counter and retry it later.
This behavior avoids buffer errors in the QCA7000, which
results in an unnecessary chip reset including packet loss.
Signed-off-by: Stefan Wahren <stefan.wahren@i2se.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David Barmann [Thu, 8 Nov 2018 14:13:35 +0000 (08:13 -0600)]
sock: Reset dst when changing sk_mark via setsockopt
When setting the SO_MARK socket option, if the mark changes, the dst
needs to be reset so that a new route lookup is performed.
This fixes the case where an application wants to change routing by
setting a new sk_mark. If this is done after some packets have already
been sent, the dst is cached and has no effect.
Signed-off-by: David Barmann <david.barmann@stackpath.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
please apply the following qeth patches to net-next.
The first patch allows one more device type to query the FW for a MAC address,
the others are all basically just removal of duplicated or unused code.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Julian Wiedmann [Thu, 8 Nov 2018 14:06:22 +0000 (15:06 +0100)]
s390/qeth: don't process hsuid in qeth_l3_setup_netdev()
qeth_l3_setup_netdev() checks if the hsuid attribute is set on the qeth
device, and propagates it to the net_device. In the past this was needed
to pick up any hsuid that was set before allocation of the net_device.
With commit d3d1b205e89f ("s390/qeth: allocate netdevice early") this
is no longer necessary, qeth_l3_dev_hsuid_store() always stores the
hsuid straight into dev->perm_addr.
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Julian Wiedmann [Thu, 8 Nov 2018 14:06:21 +0000 (15:06 +0100)]
s390/qeth: remove unused fallback in Layer3's MAC code
If the CREATE ADDR sent by qeth_l3_iqd_read_initial_mac() fails, its
callback sets a random MAC address on the net_device. The error then
propagates back, and qeth_l3_setup_netdev() bails out without
registering the net_device.
Any subsequent call to qeth_l3_setup_netdev() will then attempt a fresh
CREATE ADDR which either 1) also fails, or 2) sets a proper MAC address
on the net_device. Consequently, the net_device will never be registered
with a random MAC and we can drop the fallback code.
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Julian Wiedmann [Thu, 8 Nov 2018 14:06:20 +0000 (15:06 +0100)]
s390/qeth: remove two IPA command helpers
qeth_l3_send_ipa_arp_cmd() is merely a wrapper around
qeth_send_control_data() now. So push the length adjustment into
QETH_SETASS_BASE_LEN, and remove the wrapper. While at it, also remove
some redundant 0-initializations.
qeth_send_setassparms() requires that callers prepare their command
parameters, so that they can be copied into the parameter area in one
go. Skip the indirection, and just let callers set up the command
themselves.
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Julian Wiedmann [Thu, 8 Nov 2018 14:06:19 +0000 (15:06 +0100)]
s390/qeth: replace open-coded cmd setup
Call qeth_prepare_ipa_cmd() during setup of a new IPA cmd buffer, so
that it is used for all commands. Thus ARP and SNMP requests don't have
to do their own initialization.
This will now also set the proper MPC protocol version for SNMP requests
on L2 devices.
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Julian Wiedmann [Thu, 8 Nov 2018 14:06:17 +0000 (15:06 +0100)]
s390/qeth: unify transmit code
Since commit 82bf5c0867f6 ("s390/qeth: add support for IPv6 TSO"),
qeth_xmit() also knows how to build TSO packets and is practically
identical to qeth_l3_xmit().
Convert qeth_l3_xmit() into a thin wrapper that merely strips the
L2 header off a packet, and calls qeth_xmit() for the actual
TX processing.
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Julian Wiedmann [Thu, 8 Nov 2018 14:06:16 +0000 (15:06 +0100)]
s390/qeth: handle af_iucv skbs in qeth_l3_fill_header()
Filling the HW header from one single function will make it easier to
rip out all the duplicated transmit code in qeth_l3_xmit(). On top, this
saves one conditional branch in the TSO path.
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Julian Wiedmann [Thu, 8 Nov 2018 14:06:15 +0000 (15:06 +0100)]
s390/qeth: utilize virtual MAC for Layer2 OSD devices
By default, READ MAC on a Layer2 OSD device returns the adapter's
burnt-in MAC address. Given the default scenario of many virtual devices
on the same adapter, qeth can't make any use of this address and
therefore skips the READ MAC call for this device type.
But in some configurations, the READ MAC command for a Layer2 OSD device
actually returns a pre-provisioned, virtual MAC address. So enable the
READ MAC code to detect this situation, and let the L2 subdriver
call READ MAC for OSD devices.
This also removes the QETH_LAYER2_MAC_READ flag, which protects L2
devices against calling READ MAC multiple times. Instead protect the
whole call to qeth_l2_request_initial_mac().
Signed-off-by: Julian Wiedmann <jwi@linux.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
====================
ICMP error handling for UDP tunnels
This series introduces ICMP error handling for UDP tunnels and
encapsulations and related selftests. We need to handle ICMP errors to
support PMTU discovery and route redirection -- this support is entirely
missing right now:
- patch 1/11 adds a socket lookup for UDP tunnels that use, by design,
the same destination port on both endpoints -- i.e. VXLAN and GENEVE
- patches 2/11 to 7/11 are specific to VxLAN and GENEVE
- patches 8/11 and 9/11 add infrastructure for lookup of encapsulations
where sent packets cannot be matched via receiving socket lookup, i.e.
FoU and GUE
- patches 10/11 and 11/11 are specific to FoU and GUE
v2: changes are listed in the single patches
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Stefano Brivio [Thu, 8 Nov 2018 11:19:24 +0000 (12:19 +0100)]
selftests: pmtu: Introduce FoU and GUE PMTU exceptions tests
Introduce eight tests, for FoU and GUE, with IPv4 and IPv6 payload,
on IPv4 and IPv6 transport, that check that PMTU exceptions are created
with the right value when exceeding the MTU on a link of the path.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com> Reviewed-by: Sabrina Dubroca <sd@queasysnail.net> Signed-off-by: David S. Miller <davem@davemloft.net>
Stefano Brivio [Thu, 8 Nov 2018 11:19:23 +0000 (12:19 +0100)]
fou, fou6: ICMP error handlers for FoU and GUE
As the destination port in FoU and GUE receiving sockets doesn't
necessarily match the remote destination port, we can't associate errors
to the encapsulating tunnels with a socket lookup -- we need to blindly
try them instead. This means we don't even know if we are handling errors
for FoU or GUE without digging into the packets.
Hence, implement a single handler for both, one for IPv4 and one for IPv6,
that will check whether the packet that generated the ICMP error used a
direct IP encapsulation or if it had a GUE header, and send the error to
the matching protocol handler, if any.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com> Reviewed-by: Sabrina Dubroca <sd@queasysnail.net> Signed-off-by: David S. Miller <davem@davemloft.net>
Stefano Brivio [Thu, 8 Nov 2018 11:19:22 +0000 (12:19 +0100)]
udp: Support for error handlers of tunnels with arbitrary destination port
ICMP error handling is currently not possible for UDP tunnels not
employing a receiving socket with local destination port matching the
remote one, because we have no way to look them up.
Add an err_handler tunnel encapsulation operation that can be exported by
tunnels in order to pass the error to the protocol implementing the
encapsulation. We can't easily use a lookup function as we did for VXLAN
and GENEVE, as protocol error handlers, which would be in turn called by
implementations of this new operation, handle the errors themselves,
together with the tunnel lookup.
Without a socket, we can't be sure which encapsulation error handler is
the appropriate one: encapsulation handlers (the ones for FoU and GUE
introduced in the next patch, e.g.) will need to check the new error codes
returned by protocol handlers to figure out if errors match the given
encapsulation, and, in turn, report this error back, so that we can try
all of them in __udp{4,6}_lib_err_encap_no_sk() until we have a match.
v2:
- Name all arguments in err_handler prototypes (David Miller)
Signed-off-by: Stefano Brivio <sbrivio@redhat.com> Reviewed-by: Sabrina Dubroca <sd@queasysnail.net> Signed-off-by: David S. Miller <davem@davemloft.net>
Stefano Brivio [Thu, 8 Nov 2018 11:19:21 +0000 (12:19 +0100)]
net: Convert protocol error handlers from void to int
We'll need this to handle ICMP errors for tunnels without a sending socket
(i.e. FoU and GUE). There, we might have to look up different types of IP
tunnels, registered as network protocols, before we get a match, so we
want this for the error handlers of IPPROTO_IPIP and IPPROTO_IPV6 in both
inet_protos and inet6_protos. These error codes will be used in the next
patch.
For consistency, return sensible error codes in protocol error handlers
whenever handlers can't handle errors because, even if valid, they don't
match a protocol or any of its states.
This has no effect on existing error handling paths.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com> Reviewed-by: Sabrina Dubroca <sd@queasysnail.net> Signed-off-by: David S. Miller <davem@davemloft.net>
Stefano Brivio [Thu, 8 Nov 2018 11:19:20 +0000 (12:19 +0100)]
selftests: pmtu: Introduce tests for IPv4/IPv6 over GENEVE over IPv4/IPv6
Use a router between endpoints, implemented via namespaces, set a low MTU
between router and destination endpoint, exceed it and check PMTU value in
route exceptions.
v2:
- Introduce IPv4 tests right away, if iproute2 doesn't support the 'df'
link option they will be skipped (David Ahern)
Signed-off-by: Stefano Brivio <sbrivio@redhat.com> Reviewed-by: Sabrina Dubroca <sd@queasysnail.net> Signed-off-by: David S. Miller <davem@davemloft.net>
Stefano Brivio [Thu, 8 Nov 2018 11:19:19 +0000 (12:19 +0100)]
geneve: Allow configuration of DF behaviour
draft-ietf-nvo3-geneve-08 says:
It is strongly RECOMMENDED that Path MTU Discovery ([RFC1191],
[RFC1981]) be used by setting the DF bit in the IP header when Geneve
packets are transmitted over IPv4 (this is the default with IPv6).
Now that ICMP error handling is working for GENEVE, we can comply with
this recommendation.
Make this configurable, though, to avoid breaking existing setups. By
default, DF won't be set. It can be set or inherited from inner IPv4
packets. If it's configured to be inherited and we are encapsulating IPv6,
it will be set.
This only applies to non-lwt tunnels: if an external control plane is
used, tunnel key will still control the DF flag.
v2:
- DF behaviour configuration only applies for non-lwt tunnels, apply DF
setting only if (!geneve->collect_md) in geneve_xmit_skb()
(Stephen Hemminger)
Signed-off-by: Stefano Brivio <sbrivio@redhat.com> Reviewed-by: Sabrina Dubroca <sd@queasysnail.net> Signed-off-by: David S. Miller <davem@davemloft.net>
Stefano Brivio [Thu, 8 Nov 2018 11:19:18 +0000 (12:19 +0100)]
geneve: ICMP error lookup handler
Export an encap_err_lookup() operation to match an ICMP error against a
valid VNI.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com> Reviewed-by: Sabrina Dubroca <sd@queasysnail.net> Signed-off-by: David S. Miller <davem@davemloft.net>
Stefano Brivio [Thu, 8 Nov 2018 11:19:17 +0000 (12:19 +0100)]
selftests: pmtu: Introduce tests for IPv4/IPv6 over VXLAN over IPv4/IPv6
Use a router between endpoints, implemented via namespaces, set a low MTU
between router and destination endpoint, exceed it and check PMTU value in
route exceptions.
v2:
- Change all occurrences of VxLAN to VXLAN (Jiri Benc)
- Introduce IPv4 tests right away, if iproute2 doesn't support the 'df'
link option they will be skipped (David Ahern)
Signed-off-by: Stefano Brivio <sbrivio@redhat.com> Reviewed-by: Sabrina Dubroca <sd@queasysnail.net> Signed-off-by: David S. Miller <davem@davemloft.net>
Stefano Brivio [Thu, 8 Nov 2018 11:19:16 +0000 (12:19 +0100)]
vxlan: Allow configuration of DF behaviour
Allow users to set the IPv4 DF bit in outgoing packets, or to inherit its
value from the IPv4 inner header. If the encapsulated protocol is IPv6 and
DF is configured to be inherited, always set it.
For IPv4, inheriting DF from the inner header was probably intended from
the very beginning judging by the comment to vxlan_xmit(), but it wasn't
actually implemented -- also because it would have done more harm than
good, without handling for ICMP Fragmentation Needed messages.
According to RFC 7348, "Path MTU discovery MAY be used". An expired RFC
draft, draft-saum-nvo3-pmtud-over-vxlan-05, whose purpose was to describe
PMTUD implementation, says that "is a MUST that Vxlan gateways [...]
SHOULD set the DF-bit [...]", whatever that means.
Given this background, the only sane option is probably to let the user
decide, and keep the current behaviour as default.
This only applies to non-lwt tunnels: if an external control plane is
used, tunnel key will still control the DF flag.
v2:
- DF behaviour configuration only applies for non-lwt tunnels, move DF
setting to if (!info) block in vxlan_xmit_one() (Stephen Hemminger)
Signed-off-by: Stefano Brivio <sbrivio@redhat.com> Reviewed-by: Sabrina Dubroca <sd@queasysnail.net> Signed-off-by: David S. Miller <davem@davemloft.net>
Stefano Brivio [Thu, 8 Nov 2018 11:19:15 +0000 (12:19 +0100)]
vxlan: ICMP error lookup handler
Export an encap_err_lookup() operation to match an ICMP error against a
valid VNI.
Signed-off-by: Stefano Brivio <sbrivio@redhat.com> Reviewed-by: Sabrina Dubroca <sd@queasysnail.net> Signed-off-by: David S. Miller <davem@davemloft.net>
Stefano Brivio [Thu, 8 Nov 2018 11:19:14 +0000 (12:19 +0100)]
udp: Handle ICMP errors for tunnels with same destination port on both endpoints
For both IPv4 and IPv6, if we can't match errors to a socket, try
tunnels before ignoring them. Look up a socket with the original source
and destination ports as found in the UDP packet inside the ICMP payload,
this will work for tunnels that force the same destination port for both
endpoints, i.e. VXLAN and GENEVE.
Actually, lwtunnels could break this assumption if they are configured by
an external control plane to have different destination ports on the
endpoints: in this case, we won't be able to trace ICMP messages back to
them.
For IPv6 redirect messages, call ip6_redirect() directly with the output
interface argument set to the interface we received the packet from (as
it's the very interface we should build the exception on), otherwise the
new nexthop will be rejected. There's no such need for IPv4.
Tunnels can now export an encap_err_lookup() operation that indicates a
match. Pass the packet to the lookup function, and if the tunnel driver
reports a matching association, continue with regular ICMP error handling.
v2:
- Added newline between network and transport header sets in
__udp{4,6}_lib_err_encap() (David Miller)
- Removed redundant skb_reset_network_header(skb); in
__udp4_lib_err_encap()
- Removed redundant reassignment of iph in __udp4_lib_err_encap()
(Sabrina Dubroca)
- Edited comment to __udp{4,6}_lib_err_encap() to reflect the fact this
won't work with lwtunnels configured to use asymmetric ports. By the way,
it's VXLAN, not VxLAN (Jiri Benc)
Signed-off-by: Stefano Brivio <sbrivio@redhat.com> Reviewed-by: Sabrina Dubroca <sd@queasysnail.net> Signed-off-by: David S. Miller <davem@davemloft.net>
YueHaibing [Thu, 8 Nov 2018 02:08:43 +0000 (02:08 +0000)]
net: bcmgenet: return correct value 'ret' from bcmgenet_power_down
Fixes gcc '-Wunused-but-set-variable' warning:
drivers/net/ethernet/broadcom/genet/bcmgenet.c: In function 'bcmgenet_power_down':
drivers/net/ethernet/broadcom/genet/bcmgenet.c:1136:6: warning:
variable 'ret' set but not used [-Wunused-but-set-variable]
bcmgenet_power_down should return 'ret' instead of 0.
Fixes: ca8cf341903f ("net: bcmgenet: propagate errors from bcmgenet_power_down") Signed-off-by: YueHaibing <yuehaibing@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
====================
net: sched: prepare for more Qdisc offloads
This series refactors the "switchdev" Qdisc offloads a little. We have
a few Qdiscs which can be fully offloaded today to the forwarding plane
of switching devices.
First patch adds a helper for handing statistic dumps, the code seems
to be copy pasted between PRIO and RED. Second patch removes unnecessary
parameter from RED offload function. Third patch makes the MQ offload
use the dump helper which helps it behave much like PRIO and RED when
it comes to the TCQ_F_OFFLOADED flag. Patch 4 adds a graft helper,
similar to the dump helper.
Patch 5 is unrelated to offloads, qdisc_graft() code seemed ripe for a
small refactor - no functional changes there.
Last two patches move the qdisc_put() call outside of the sch_tree_lock
section for RED and PRIO. The child Qdiscs will get removed from the
hierarchy under the lock, but having the put (and potentially destroy)
called outside of the lock helps offload which may choose to sleep,
and it should generally lower the Qdisc change impact.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Jakub Kicinski [Thu, 8 Nov 2018 01:33:40 +0000 (17:33 -0800)]
net: sched: prio: delay destroying child qdiscs on change
Move destroying of the old child qdiscs outside of the sch_tree_lock()
section. This should improve the software qdisc replace but is even
more important for offloads. Calling offloads under a spin lock is
best avoided, and child's destroy would be called under sch_tree_lock().
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: John Hurley <john.hurley@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jakub Kicinski [Thu, 8 Nov 2018 01:33:39 +0000 (17:33 -0800)]
net: sched: red: delay destroying child qdisc on replace
Move destroying of the old child qdisc outside of the sch_tree_lock()
section. This should improve the software qdisc replace but is even
more important for offloads. Firstly calling offloads under a spin
lock is best avoided. Secondly the destroy event of existing child
would have been sent to the offload device before the replace, causing
confusion.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: John Hurley <john.hurley@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jakub Kicinski [Thu, 8 Nov 2018 01:33:38 +0000 (17:33 -0800)]
net: sched: refactor grafting Qdiscs with a parent
The code for grafting Qdiscs when there is a parent has two needless
indentation levels, and breaks the "keep the success path unindented"
guideline. Refactor.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: John Hurley <john.hurley@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jakub Kicinski [Thu, 8 Nov 2018 01:33:37 +0000 (17:33 -0800)]
net: sched: add an offload graft helper
Qdisc graft operation of offload-capable qdiscs performs a few
extra steps which are identical among all the qdiscs. Add
a helper to share this code.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: John Hurley <john.hurley@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jakub Kicinski [Thu, 8 Nov 2018 01:33:36 +0000 (17:33 -0800)]
net: sched: set TCQ_F_OFFLOADED flag for MQ
PRIO and RED mark the qdisc with TCQ_F_OFFLOADED upon successful offload,
make MQ do the same. The consistency will help with consistent
graft callback behaviour.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: John Hurley <john.hurley@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Offload dump helper does not use opt parameter, remove it.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: John Hurley <john.hurley@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jakub Kicinski [Thu, 8 Nov 2018 01:33:34 +0000 (17:33 -0800)]
net: sched: add an offload dump helper
Qdisc dump operation of offload-capable qdiscs performs a few
extra steps which are identical among all the qdiscs. Add
a helper to share this code.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: John Hurley <john.hurley@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
====================
net: phy: improve and simplify phylib state machine
This patch series is based on two axioms:
- During autoneg a PHY always reports the link being down
- Info in clause 22/45 registers doesn't allow to differentiate between
these two states:
1. Link is physically down
2. A link partner is connected and PHY is autonegotiating
In both cases "link up" and "aneg finished" bits aren't set.
One consequence is that having separate states PHY_NOLINK and PHY_AN
isn't needed.
By using these two axioms the state machine can be significantly
simplified.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Heiner Kallweit [Wed, 7 Nov 2018 19:45:58 +0000 (20:45 +0100)]
net: phy: add phy_check_link_status
In few places in the state machine the state is set to PHY_RUNNING or
PHY_NOLINK after doing a phy_read_status(). So factor this out to
phy_check_link_status().
First use it in phy_start_aneg(): By setting the state to PHY_RUNNING
or PHY_NOLINK directly we can remove the code to handle the case that
we're using interrupts and aneg was finished already.
Definition of phy_link_up and phy_link_down needs to be moved because
they are called in the new function.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Heiner Kallweit [Wed, 7 Nov 2018 19:44:56 +0000 (20:44 +0100)]
net: phy: remove useless check in state machine case PHY_RESUMING
If aneg isn't finished yet then the PHY reports the link as down.
There's no benefit in setting the state to PHY_AN because the next
state machine run would set the status to PHY_NOLINK anyway (except
in the meantime aneg has been finished and link is up). Therefore
we can set the state to PHY_RUNNING or PHY_NOLINK directly.
In addition change the do_carrier parameter in phy_link_down() to true.
If carrier was marked as up before (what should never be the case because
PHY was in state PHY_HALTED before) then we should mark it as down now.
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Thu, 8 Nov 2018 07:07:04 +0000 (23:07 -0800)]
Merge branch '1GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/next-queue
Jeff Kirsher says:
====================
Intel Wired LAN Driver Updates 2018-11-07
This series contains updates to almost all of the Intel wired LAN
drivers.
Lance Roy replaces a spin lock with lockdep_assert_held() for igbvf
driver in move toward trying to remove spin_is_locked().
Colin Ian King fixes a potential null pointer dereference by adding a
check in ixgbe. Also fixed the igc driver by properly assigning the
return error code of a function call, so that we can properly check it.
Shannon Nelson updates the ixgbe driver to not block IPsec offload when
in VEPA mode, in VEB mode, IPsec offload is still blocked because the
device drops packets into a black hole.
Jake adds support for software timestamping for packets sent over
ixgbevf. Also modifies i40e, iavf, igb, igc, and ixgbe to delay calling
skb_tx_timestamp() to the latest point possible, which is just prior to
notifying the hardware of the new Tx packet.
Todd adds the new WoL filter flag so that we properly report that we do
not support this new feature.
YueHaibing from Huawei fixes the igc driver by cleaning up variables
that are not "really" used.
Dan Carpenter cleans up igc whitespace issues.
Miroslav Lichvar fixes e1000e for potential underflow issue in the
timecounter, so modify the driver to use timecounter_cyc2time() to allow
non-monotonic SYSTIM readings.
Sasha provides additional igc cleanups based on community feedback.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
====================
nfp: add and use tunnel netdev helpers
A recent patch introduced the function netif_is_vxlan() to verify the
tunnel type of a given netdev as vxlan.
Add a similar function to detect geneve netdevs and make use of this
function in the NFP driver. Also make use of the vxlan helper where
applicable.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
John Hurley [Wed, 7 Nov 2018 18:32:50 +0000 (18:32 +0000)]
nfp: flower: include geneve as supported offload tunnel type
Offload of geneve decap rules is supported in NFP. Include geneve in the
check for supported types.
Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
John Hurley [Wed, 7 Nov 2018 18:32:49 +0000 (18:32 +0000)]
nfp: flower: use geneve and vxlan helpers
Make use of the recently added VXLAN and geneve helper functions to
determine the type of the netdev from its rtnl_link_ops.
Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
John Hurley [Wed, 7 Nov 2018 18:32:48 +0000 (18:32 +0000)]
net: add netif_is_geneve()
Add a helper function to determine if the type of a netdev is geneve based
on its rtnl_link_ops. This allows drivers that may wish to offload tunnels
to check the underlying type of the device.
A recent patch added a similar helper to vxlan.h
Signed-off-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Edward Cree [Wed, 7 Nov 2018 18:12:42 +0000 (18:12 +0000)]
sfc: add missing NVRAM partition types for EF10
Expose the MUM/SUC Firmware, UEFI Expansion ROM and MC Status partitions
of the NIC's NVRAM as MTDs if found on the NIC. The first two are needed
in order to properly update them when performing firmware updates; the MC
Status partition is used to determine whether a signed firmware image was
accepted or rejected by a Secure NIC.
Signed-off-by: Edward Cree <ecree@solarflare.com> Signed-off-by: David S. Miller <davem@davemloft.net>
====================
net/vlan: prepare for removal of VLAN_TAG_PRESENT
This is a preparatory patchset before removing the use of VLAN_TAG_PRESENT
bit in skb->vlan_tci as indication of VLAN offload. This set includes
only cleanups that allow abstracting of code testing VLAN tag presence
in drivers and networking code.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Davide Caratti [Wed, 7 Nov 2018 10:28:18 +0000 (11:28 +0100)]
net: vlan: add support for tunnel offload
GSO tunneled packets are always segmented in software before they are
transmitted by a VLAN, even when the lower device can offload tunnel
encapsulation and VLAN together (i.e., some bits in NETIF_F_GSO_ENCAP_ALL
mask are set in the lower device 'vlan_features'). If we let VLANs have
the same tunnel offload capabilities as their lower device, throughput
can improve significantly when CPU is limited on the transmitter side.
- set NETIF_F_GSO_ENCAP_ALL bits in the VLAN 'hw_features', to ensure
that 'features' will have those bits zeroed only when the lower device
has no hardware support for tunnel encapsulation.
- for the same reason, copy GSO-related bits of 'hw_enc_features' from
lower device to VLAN, and ensure to update that value when the lower
device changes its features.
- set NETIF_F_HW_CSUM bit in the VLAN 'hw_enc_features' if 'real_dev'
is able to compute checksums at least for a kind of packets, like done
with commit 8403debeead8 ("vlan: Keep NETIF_F_HW_CSUM similar to other
software devices"). This avoids software segmentation due to mismatching
checksum capabilities between VLAN's 'features' and 'hw_enc_features'.
Reported-by: Flavio Leitner <fbl@redhat.com> Signed-off-by: Davide Caratti <dcaratti@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Paolo Abeni [Wed, 7 Nov 2018 09:34:36 +0000 (10:34 +0100)]
tun: compute the RFS hash only if needed.
The tun XDP sendmsg code path, unconditionally computes the symmetric
hash of each packet for RFS's sake, even when we could skip it. e.g.
when the device has a single queue.
This change adds the check already in-place for the skb sendmsg path
to avoid unneeded hashing.
The above gives small, but measurable, performance gain for VM xmit
path when zerocopy is not enabled.
Signed-off-by: Paolo Abeni <pabeni@redhat.com> Acked-by: Jason Wang <jasowang@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Heiner Kallweit [Wed, 7 Nov 2018 07:52:46 +0000 (08:52 +0100)]
net: phy: realtek: load driver for all PHYs with a Realtek OUI
Instead of listing every single PHYID, load the driver for every PHYID
with a Realtek OUI, independent of model number and revision.
This patch also improves two further aspects:
- constify realtek_tbl[]
- the mask should have been 0xffffffff instead of 0x001fffff so far,
by masking out some bits a PHY from another vendor could have been
matched
Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Fixes: ae7a5aff783c ("net: dsa: bcm_sf2: Keep copy of inserted rules") CC: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: kbuild test robot <fengguang.wu@intel.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Justin Chen [Wed, 7 Nov 2018 00:37:44 +0000 (16:37 -0800)]
net: phy: bcm7xxx: Add entry for BCM7255
Add support for BCM7255 EPHY.
Signed-off-by: Justin Chen <justinpopo6@gmail.com> Signed-off-by: Florian Fainelli <f.fainelli@gmail.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Thu, 8 Nov 2018 00:23:05 +0000 (16:23 -0800)]
Merge branch 'udp-gro'
Paolo Abeni says:
====================
udp: implement GRO support
This series implements GRO support for UDP sockets, as the RX counterpart
of commit bec1f6f69736 ("udp: generate gso with UDP_SEGMENT").
The core functionality is implemented by the second patch, introducing a new
sockopt to enable UDP_GRO, while patch 3 implements support for passing the
segment size to the user space via a new cmsg.
UDP GRO performs a socket lookup for each ingress packets and aggregate datagram
directed to UDP GRO enabled sockets with constant l4 tuple.
UDP GRO packets can land on non GRO-enabled sockets, e.g. due to iptables NAT
rules, and that could potentially confuse existing applications.
The solution adopted here is to de-segment the GRO packet before enqueuing
as needed. Since we must cope with packet reinsertion after de-segmentation,
the relevant code is factored-out in ipv4 and ipv6 specific helpers and exposed
to UDP usage.
While the current code can probably be improved, this safeguard ,implemented in
the patches 4-7, allows future enachements to enable UDP GSO offload on more
virtual devices eventually even on forwarded packets.
The last 4 for patches implement some performance and functional self-tests,
re-using the existing udpgso infrastructure. The problematic scenario described
above is explicitly tested.
This revision of the series try to address the feedback provided by Willem and
Subash on previous iteration.
====================
Acked-by: Willem de Bruijn <willemb@google.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Paolo Abeni [Wed, 7 Nov 2018 11:38:37 +0000 (12:38 +0100)]
selftests: add functionals test for UDP GRO
Extends the existing udp programs to allow checking for proper
GRO aggregation/GSO size, and run the tests via a shell script, using
a veth pair with XDP program attached to trigger the GRO code path.
rfc v3 -> v1:
- use ip route to attach the xdp helper to the veth
rfc v2 -> rfc v3:
- add missing test program options documentation
- fix sporatic test failures (receiver faster than sender)
Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Paolo Abeni [Wed, 7 Nov 2018 11:38:35 +0000 (12:38 +0100)]
selftests: add dummy xdp test helper
This trivial XDP program does nothing, but will be used by the
next patch to test the GRO path in a net namespace, leveraging
the veth XDP implementation.
It's added here, despite its 'net' usage, to avoid the duplication
of the llc-related makefile boilerplate.
rfc v3 -> v1:
- move the helper implementation into the bpf directory, don't
touch udpgso_bench_rx
rfc v2 -> rfc v3:
- move 'x' option handling here
Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Paolo Abeni [Wed, 7 Nov 2018 11:38:33 +0000 (12:38 +0100)]
udp: cope with UDP GRO packet misdirection
In some scenarios, the GRO engine can assemble an UDP GRO packet
that ultimately lands on a non GRO-enabled socket.
This patch tries to address the issue explicitly checking for the UDP
socket features before enqueuing the packet, and eventually segmenting
the unexpected GRO packet, as needed.
We must also cope with re-insertion requests: after segmentation the
UDP code calls the helper introduced by the previous patches, as needed.
Segmentation is performed by a common helper, which takes care of
updating socket and protocol stats is case of failure.
rfc v3 -> v1
- fix compile issues with rxrpc
- when gso_segment returns NULL, treat is as an error
- added 'ipv4' argument to udp_rcv_segment()
rfc v2 -> rfc v3
- moved udp_rcv_segment() into net/udp.h, account errors to socket
and ns, always return NULL or segs list
Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Paolo Abeni [Wed, 7 Nov 2018 11:38:29 +0000 (12:38 +0100)]
udp: implement GRO for plain UDP sockets.
This is the RX counterpart of commit bec1f6f69736 ("udp: generate gso
with UDP_SEGMENT"). When UDP_GRO is enabled, such socket is also
eligible for GRO in the rx path: UDP segments directed to such socket
are assembled into a larger GSO_UDP_L4 packet.
The core UDP GRO support is enabled with setsockopt(UDP_GRO).
Initial benchmark numbers:
Before:
udp rx: 1079 MB/s 769065 calls/s
After:
udp rx: 1466 MB/s 24877 calls/s
This change introduces a side effect in respect to UDP tunnels:
after a UDP tunnel creation, now the kernel performs a lookup per ingress
UDP packet, while before such lookup happened only if the ingress packet
carried a valid internal header csum.
rfc v2 -> rfc v3:
- fixed typos in macro name and comments
- really enforce UDP_GRO_CNT_MAX, instead of UDP_GRO_CNT_MAX + 1
- acquire socket lock in UDP_GRO setsockopt
rfc v1 -> rfc v2:
- use a new option to enable UDP GRO
- use static keys to protect the UDP GRO socket lookup
Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Paolo Abeni [Wed, 7 Nov 2018 11:38:28 +0000 (12:38 +0100)]
udp: implement complete book-keeping for encap_needed
The *encap_needed static keys are enabled by UDP tunnels
and several UDP encapsulations type, but they are never
turned off. This can cause unneeded overall performance
degradation for systems where such features are used
transiently.
This patch introduces complete book-keeping for such keys,
decreasing the usage at socket destruction time, if needed,
and avoiding that the same socket could increase the key
usage multiple times.
rfc v3 -> v1:
- add socket lock around udp_tunnel_encap_enable()
rfc v2 -> rfc v3:
- use udp_tunnel_encap_enable() in setsockopt()
Signed-off-by: Paolo Abeni <pabeni@redhat.com> Signed-off-by: David S. Miller <davem@davemloft.net>
====================
vrf: allow simultaneous service instances in default and other VRFs
Services currently have to be VRF-aware if they are using an unbound
socket. One cannot have multiple service instances running in the
default and other VRFs for services that are not VRF-aware and listen
on an unbound socket. This is because there is no easy way of isolating
packets received in the default VRF from those arriving in other VRFs.
This series provides this isolation for stream sockets subject to the
existing kernel parameter net.ipv4.tcp_l3mdev_accept not being set,
given that this is documented as allowing a single service instance to
work across all VRF domains. Similarly, net.ipv4.udp_l3mdev_accept is
checked for datagram sockets, and net.ipv4.raw_l3mdev_accept is
introduced for raw sockets. The functionality applies to UDP & TCP
services as well as those using raw sockets, and is for IPv4 and IPv6.
Example of running ssh instances in default and blue VRF:
v1:
- Address Paolo Abeni's comments (patch 4/5)
- Fix build when CONFIG_NET_L3_MASTER_DEV not defined (patch 1/5)
v2:
- Address David Aherns' comments (patches 4/5 and 5/5)
- Remove patches 3/5 and 5/5 from series for individual submissions
- Include a sysctl for raw sockets as recommended by David Ahern
- Expand series into 10 patches and provide improved descriptions
v3:
- Update description for patch 1/10 and remove patch 6/10
v4:
- Set default to enabled for raw socket sysctl as recommended by David Ahern
v5:
- Address review comments from David Ahern in patches 2-5
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Dewi Morgan [Wed, 7 Nov 2018 15:36:10 +0000 (15:36 +0000)]
ipv6: do not drop vrf udp multicast packets
For bound udp sockets in a vrf, also check the sdif to get the index
for ingress devices enslaved to an l3mdev.
Signed-off-by: Dewi Morgan <morgand@vyatta.att-mail.com> Signed-off-by: Mike Manning <mmanning@vyatta.att-mail.com> Reviewed-by: David Ahern <dsahern@gmail.com> Tested-by: David Ahern <dsahern@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Mike Manning [Wed, 7 Nov 2018 15:36:09 +0000 (15:36 +0000)]
ipv6: handling of multicast packets received in VRF
If the skb for multicast packets marked as enslaved to a VRF are
received, then the secondary device index should be used to obtain
the real device. And verify the multicast address against the
enslaved rather than the l3mdev device.
Signed-off-by: Dewi Morgan <morgand@vyatta.att-mail.com> Signed-off-by: Mike Manning <mmanning@vyatta.att-mail.com> Reviewed-by: David Ahern <dsahern@gmail.com> Tested-by: David Ahern <dsahern@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Mike Manning [Wed, 7 Nov 2018 15:36:08 +0000 (15:36 +0000)]
ipv6: allow ping to link-local address in VRF
If link-local packets are marked as enslaved to a VRF, then to allow
ping to the link-local from a vrf, the error handling for IPV6_PKTINFO
needs to be relaxed to also allow the pkt ipi6_ifindex to be that of a
slave device to the vrf.
Note that the real device also needs to be retrieved in icmp6_iif()
to set the ipv6 flow oif to this for icmp echo reply handling. The
recent commit 24b711edfc34 ("net/ipv6: Fix linklocal to global address
with VRF") takes care of this, so the sdif does not need checking here.
This fix makes ping to link-local consistent with that to global
addresses, in that this can now be done from within the same VRF that
the address is in.
Signed-off-by: Mike Manning <mmanning@vyatta.att-mail.com> Reviewed-by: David Ahern <dsahern@gmail.com> Tested-by: David Ahern <dsahern@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Mike Manning [Wed, 7 Nov 2018 15:36:07 +0000 (15:36 +0000)]
vrf: mark skb for multicast or link-local as enslaved to VRF
The skb for packets that are multicast or to a link-local address are
not marked as being enslaved to a VRF, if they are received on a socket
bound to the VRF. This is needed for ND and it is preferable for the
kernel not to have to deal with the additional use-cases if ll or mcast
packets are handled as enslaved. However, this does not allow service
instances listening on unbound and bound to VRF sockets to distinguish
the VRF used, if packets are sent as multicast or to a link-local
address. The fix is for the VRF driver to also mark these skb as being
enslaved to the VRF.
Signed-off-by: Mike Manning <mmanning@vyatta.att-mail.com> Reviewed-by: David Ahern <dsahern@gmail.com> Tested-by: David Ahern <dsahern@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Duncan Eastoe [Wed, 7 Nov 2018 15:36:06 +0000 (15:36 +0000)]
net: fix raw socket lookup device bind matching with VRFs
When there exist a pair of raw sockets one unbound and one bound
to a VRF but equal in all other respects, when a packet is received
in the VRF context, __raw_v4_lookup() matches on both sockets.
This results in the packet being delivered over both sockets,
instead of only the raw socket bound to the VRF. The bound device
checks in __raw_v4_lookup() are replaced with a call to
raw_sk_bound_dev_eq() which correctly handles whether the packet
should be delivered over the unbound socket in such cases.
In __raw_v6_lookup() the match on the device binding of the socket is
similarly updated to use raw_sk_bound_dev_eq() which matches the
handling in __raw_v4_lookup().
Importantly raw_sk_bound_dev_eq() takes the raw_l3mdev_accept sysctl
into account.
Signed-off-by: Duncan Eastoe <deastoe@vyatta.att-mail.com> Signed-off-by: Mike Manning <mmanning@vyatta.att-mail.com> Reviewed-by: David Ahern <dsahern@gmail.com> Tested-by: David Ahern <dsahern@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Mike Manning [Wed, 7 Nov 2018 15:36:05 +0000 (15:36 +0000)]
net: provide a sysctl raw_l3mdev_accept for raw socket lookup with VRFs
Add a sysctl raw_l3mdev_accept to control raw socket lookup in a manner
similar to use of tcp_l3mdev_accept for stream and of udp_l3mdev_accept
for datagram sockets. Have this default to enabled for reasons of
backwards compatibility. This is so as to specify the output device
with cmsg and IP_PKTINFO, but using a socket not bound to the
corresponding VRF. This allows e.g. older ping implementations to be
run with specifying the device but without executing it in the VRF.
If the option is disabled, packets received in a VRF context are only
handled by a raw socket bound to the VRF, and correspondingly packets
in the default VRF are only handled by a socket not bound to any VRF.
Signed-off-by: Mike Manning <mmanning@vyatta.att-mail.com> Reviewed-by: David Ahern <dsahern@gmail.com> Tested-by: David Ahern <dsahern@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Mike Manning [Wed, 7 Nov 2018 15:36:04 +0000 (15:36 +0000)]
net: ensure unbound datagram socket to be chosen when not in a VRF
Ensure an unbound datagram skt is chosen when not in a VRF. The check
for a device match in compute_score() for UDP must be performed when
there is no device match. For this, a failure is returned when there is
no device match. This ensures that bound sockets are never selected,
even if there is no unbound socket.
Allow IPv6 packets to be sent over a datagram skt bound to a VRF. These
packets are currently blocked, as flowi6_oif was set to that of the
master vrf device, and the ipi6_ifindex is that of the slave device.
Allow these packets to be sent by checking the device with ipi6_ifindex
has the same L3 scope as that of the bound device of the skt, which is
the master vrf device. Note that this check always succeeds if the skt
is unbound.
Even though the right datagram skt is now selected by compute_score(),
a different skt is being returned that is bound to the wrong vrf. The
difference between these and stream sockets is the handling of the skt
option for SO_REUSEPORT. While the handling when adding a skt for reuse
correctly checks that the bound device of the skt is a match, the skts
in the hashslot are already incorrect. So for the same hash, a skt for
the wrong vrf may be selected for the required port. The root cause is
that the skt is immediately placed into a slot when it is created,
but when the skt is then bound using SO_BINDTODEVICE, it remains in the
same slot. The solution is to move the skt to the correct slot by
forcing a rehash.
Signed-off-by: Mike Manning <mmanning@vyatta.att-mail.com> Reviewed-by: David Ahern <dsahern@gmail.com> Tested-by: David Ahern <dsahern@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Mike Manning [Wed, 7 Nov 2018 15:36:03 +0000 (15:36 +0000)]
net: ensure unbound stream socket to be chosen when not in a VRF
The commit a04a480d4392 ("net: Require exact match for TCP socket
lookups if dif is l3mdev") only ensures that the correct socket is
selected for packets in a VRF. However, there is no guarantee that
the unbound socket will be selected for packets when not in a VRF.
By checking for a device match in compute_score() also for the case
when there is no bound device and attaching a score to this, the
unbound socket is selected. And if a failure is returned when there
is no device match, this ensures that bound sockets are never selected,
even if there is no unbound socket.
Signed-off-by: Mike Manning <mmanning@vyatta.att-mail.com> Reviewed-by: David Ahern <dsahern@gmail.com> Tested-by: David Ahern <dsahern@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Robert Shearman [Wed, 7 Nov 2018 15:36:02 +0000 (15:36 +0000)]
net: allow binding socket in a VRF when there's an unbound socket
Change the inet socket lookup to avoid packets arriving on a device
enslaved to an l3mdev from matching unbound sockets by removing the
wildcard for non sk_bound_dev_if and instead relying on check against
the secondary device index, which will be 0 when the input device is
not enslaved to an l3mdev and so match against an unbound socket and
not match when the input device is enslaved.
Change the socket binding to take the l3mdev into account to allow an
unbound socket to not conflict sockets bound to an l3mdev given the
datapath isolation now guaranteed.
Signed-off-by: Robert Shearman <rshearma@vyatta.att-mail.com> Signed-off-by: Mike Manning <mmanning@vyatta.att-mail.com> Reviewed-by: David Ahern <dsahern@gmail.com> Tested-by: David Ahern <dsahern@gmail.com> Signed-off-by: David S. Miller <davem@davemloft.net>
YueHaibing [Wed, 7 Nov 2018 02:33:53 +0000 (02:33 +0000)]
net: hns3: Remove set but not used variable 'reset_level'
Fixes gcc '-Wunused-but-set-variable' warning:
drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.c: In function 'hclge_log_and_clear_ppp_error':
drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_err.c:821:24: warning:
variable 'reset_level' set but not used [-Wunused-but-set-variable]
enum hnae3_reset_type reset_level = HNAE3_NONE_RESET;
It never used since introduction in commit 01865a50d78f ("net: hns3: Add enable and process hw errors of TM scheduler")
Signed-off-by: YueHaibing <yuehaibing@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
====================
nfp: more set actions and notifier refactor
This series brings updates to flower offload code. First Pieter adds
support for setting TTL, ToS, Flow Label and Hop Limit fields in IPv4
and IPv6 headers.
Remaining 5 patches deal with factoring out netdev notifiers from flower
code. We already have two instances, and more is coming, so it's time
to move to one central notifier which then feeds individual feature
handlers.
I start that part by cleaning up the existing notifiers. Next a central
notifier is added, and used by flower offloads.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Jakub Kicinski [Wed, 7 Nov 2018 01:07:34 +0000 (17:07 -0800)]
nfp: flower: use the common netdev notifier
Use driver's common notifier for LAG and tunnel configuration.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: John Hurley <john.hurley@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jakub Kicinski [Wed, 7 Nov 2018 01:07:33 +0000 (17:07 -0800)]
nfp: register a notifier handler in a central location for the device
Code interested in networking events registers its own notifier
handlers. Create one device-wide notifier instance.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: John Hurley <john.hurley@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jakub Kicinski [Wed, 7 Nov 2018 01:07:32 +0000 (17:07 -0800)]
nfp: flower: make nfp_fl_lag_changels_event() void
nfp_fl_lag_changels_event() never fails, and therefore we would
never return NOTIFY_BAD for NETDEV_CHANGELOWERSTATE. Make this
clearer by changing nfp_fl_lag_changels_event()'s return type
to void.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: John Hurley <john.hurley@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jakub Kicinski [Wed, 7 Nov 2018 01:07:31 +0000 (17:07 -0800)]
nfp: flower: don't try to nack device unregister events
Returning an error from a notifier means we want to veto the change.
We shouldn't veto NETDEV_UNREGISTER just because we couldn't find
the tracking info for given master.
I can't seem to find a way to trigger this unless we have some
other bug, so it's probably not fix-worthy.
While at it move the checking if the netdev really is of interest
into the handling functions, like we do for other events.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: John Hurley <john.hurley@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Jakub Kicinski [Wed, 7 Nov 2018 01:07:30 +0000 (17:07 -0800)]
nfp: flower: remove unnecessary iteration over devices
For flower tunnel offloads FW has to be informed about MAC addresses
of tunnel devices. We use a netdev notifier to keep track of these
addresses.
Remove unnecessary loop over netdevices after notifier is registered.
The intention of the loop was to catch devices which already existed
on the system before nfp driver got loaded, but netdev notifier will
replay NETDEV_REGISTER events.
Signed-off-by: Jakub Kicinski <jakub.kicinski@netronome.com> Reviewed-by: John Hurley <john.hurley@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
nfp: flower: add ipv6 set flow label and hop limit offload
Add ipv6 set flow label and hop limit action offload. Since pedit sets
headers per 4 byte word, we need to ensure that setting either version,
priority, payload_len or nexthdr does not get offloaded.
Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Add ipv4 set ttl and tos action offload. Since pedit sets headers per 4
byte word, we need to ensure that setting either version, ihl, protocol,
total length or checksum does not get offloaded.
Signed-off-by: Pieter Jansen van Vuuren <pieter.jansenvanvuuren@netronome.com> Reviewed-by: John Hurley <john.hurley@netronome.com> Reviewed-by: Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Huazhong Tan [Wed, 7 Nov 2018 04:06:18 +0000 (12:06 +0800)]
net: hns3: fix for cmd queue memory not freed problem during reset
It is not necessary to reallocate the descriptor and remap the
descriptor memory in reset process, otherwise it may cause memory
not freed problem.
Also, this patch initializes the cmd queue's spinlocks in
hclgevf_alloc_cmd_queue, and take the spinlocks when reinitializing
cmd queue' registers.
Fixes: fedd0c15d288 ("net: hns3: Add HNS3 VF IMP(Integrated Management Proc) cmd interface") Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com> Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Huazhong Tan [Wed, 7 Nov 2018 04:06:17 +0000 (12:06 +0800)]
net: hns3: add error handler for hclge_reset()
When hclge_reset() is called, it may fail for several reasons.
For example, an higher-level reset event occurs, memory allocation
failure, hardware reset timeout, etc. Therefore, it is necessary
to add corresponding error handling for these situations.
1. A high-level reset is required due to a high-level reset failure.
2. For memory allocation failure, a high-level reset is initiated by
the timer to recover. The reason for using the timer is to prevent this
new high-level reset to interrupt the reset process of other pf/vf;
3. For the case of hardware reset timeout, reschedule the reset task
to wait for the hardware to complete the reset.
For memory allocation failure and reset timeouts, in order to prevent
an infinite number of scheduled reset tasks, the number of error
recovery needs to be limited.
This patch also add some reset related debug log printing.
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Huazhong Tan [Wed, 7 Nov 2018 04:06:16 +0000 (12:06 +0800)]
net: hns3: call roce's reset notify callback when resetting
While doing resetting, roce should do its uninitailization part
before nic's, and do its initialization part after nic's.
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com> Signed-off-by: Peng Li <lipeng321@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Huazhong Tan [Wed, 7 Nov 2018 04:06:15 +0000 (12:06 +0800)]
net: hns3: adjust the process of PF reset
When doing PF reset, the driver needs to do some preparatory work
before asserting PF reset. Since when hardware is resetting, it
is necessary to stop tx/rx queue, clear hardware table, etc,
otherwise hardware may run into unrecoverable state if there is
still IO running when the hardware is resetting.
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Huazhong Tan [Wed, 7 Nov 2018 04:06:14 +0000 (12:06 +0800)]
net: hns3: move some reset information from hnae3_handle into hclge_dev/hclgevf_dev
Saving reset related information in the hclge_dev/hclgevf_dev
structure is more suitable than the hnae3_handle, since hardware
related information is kept in these two structure.
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Huazhong Tan [Wed, 7 Nov 2018 04:06:13 +0000 (12:06 +0800)]
net: hns3: ignore new coming low-level reset while doing high-level reset
When processing a higher level reset, the pending lower level reset
does not have to be processed anymore, because the higher level
reset is the superset of the lower level reset.
Therefore, when processing an higher level reset, the request of
lower level reset needs to be cleared.
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Huazhong Tan [Wed, 7 Nov 2018 04:06:12 +0000 (12:06 +0800)]
net: hns3: use HNS3_NIC_STATE_RESETTING to indicate resetting
While hclge is going to reset, it will notify its client with
HNAE3_DOWN_CLIENT, so this client should get into a resetting
status from this moment, other operations from the stack need to
be blocked as well. And when the reset is finished, the client
will be notified with HNAE3_UP_CLIENT, so this is the end of
the resetting status.
This patch uses HNS3_NIC_STATE_RESETTING flag to implement that,
and adds hns3_nic_resetting() to indicate which operation is not
allowed.
Signed-off-by: Huazhong Tan <tanhuazhong@huawei.com> Signed-off-by: David S. Miller <davem@davemloft.net>
projects / mirror_ubuntu-disco-kernel.git / log