Skip to content

L
linux
Switch branch/tag
  1. 22 Sep, 2014 31 commits
  3. 20 Sep, 2014 1 commit
    • Andy Zhou's avatar
      udp_tunnel: Only build ip6_udp_tunnel.c when IPV6 is selected · 6d967f87
      Andy Zhou authored 
      Functions supplied in ip6_udp_tunnel.c are only needed when IPV6 is
selected. When IPV6 is not selected, those functions are stubbed out
in udp_tunnel.h.

==================================================================
 net/ipv6/ip6_udp_tunnel.c:15:5: error: redefinition of 'udp_sock_create6'
     int udp_sock_create6(struct net *net, struct udp_port_cfg *cfg,
 In file included from net/ipv6/ip6_udp_tunnel.c:9:0:
      include/net/udp_tunnel.h:36:19: note: previous definition of 'udp_sock_create6' was here
       static inline int udp_sock_create6(struct net *net, struct udp_port_cfg *cfg,
==================================================================

Fixes:  fd384412 udp_tunnel: Seperate ipv6 functions into its own file
Reported-by: default avatarkbuild test robot <fengguang.wu@intel.com>
Signed-off-by: default avatarAndy Zhou <azhou@nicira.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      6d967f87
  5. 19 Sep, 2014 8 commits
    • David S. Miller's avatar
      Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/net-next · 6c62f606
      David S. Miller authored 
      Jeff Kirsher says:

====================
Intel Wired LAN Driver Updates 2014-09-18

This series contains updates to ixgbe and ixgbevf.

Ethan Zhao cleans up ixgbe and ixgbevf by removing bd_number from the
adapter struct because it is not longer useful.

Mark fixes ixgbe where if a hardware transmit timestamp is requested,
an uninitialized workqueue entry may be scheduled.  Added a check for
a PTP clock to avoid that.

Jacob provides a number of cleanups for ixgbe.  Since we may call
ixgbe_acquire_msix_vectors() prior to registering our netdevice, we
should not use the netdevice specific printk and use e_dev_warn()
instead.  Similar to how ixgbevf handles acquiring MSI-X vectors, we
can return an error code instead of relying on the flag being set.
This makes it more clear that we have failed to setup MSI-X mode and
will make it easier to consolidate MSI-X related code into a single
function.  In the case of disabling DCB, it is not an error since we
still can function, we just have to let the user know.  So use
e_dev_warn() instead of e_err().  Added warnings for other features
that are disabled when we are without MSI-X support.  Cleanup flags
that are no longer used or needed.
====================
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      6c62f606
    • David S. Miller's avatar
      Merge branch 'mlx4-next' · 58310b3f
      David S. Miller authored 
      Or Gerlitz says:

====================
mlx4: CQE/EQE stride support

This series from Ido Shamay is intended for archs having
cache line larger then 64 bytes.

Since our CQE/EQEs are generally 64B in those systems, HW will write
twice to the same cache line consecutively, causing pipe locks due to
he hazard prevention mechanism. For elements in a cyclic buffer, writes
are consecutive, so entries smaller than a cache line should be
avoided, especially if they are written at a high rate.

Reduce consecutive writes to same cache line in CQs/EQs, by allowing the
driver to increase the distance between entries so that each will reside
in a different cache line.
====================
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      58310b3f
    • Ido Shamay's avatar
      net/mlx4_en: Add mlx4_en_get_cqe helper · b1b6b4da
      Ido Shamay authored 
      This function derives the base address of the CQE from the CQE size,
and calculates the real CQE context segment in it from the factor
(this is like before). Before this change the code used the factor to
calculate the base address of the CQE as well.

The factor indicates in which segment of the cqe stride the cqe information
is located. For 32-byte strides, the segment is 0, and for 64 byte strides,
the segment is 1 (bytes 32..63). Using the factor was ok as long as we had
only 32 and 64 byte strides. However, with larger strides, the factor is zero,
and so cannot be used to calculate the base of the CQE.

The helper uses the same method of CQE buffer pulling made by all other
components that reads the CQE buffer (mlx4_ib driver and libmlx4).
Signed-off-by: default avatarIdo Shamay <idos@mellanox.com>
Signed-off-by: default avatarJack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: default avatarOr Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      b1b6b4da
    • Ido Shamay's avatar
      net/mlx4_core: Cache line EQE size support · 43c816c6
      Ido Shamay authored 
      Enable mlx4 interrupt handler to work with EQE stride feature,
The feature may be enabled when cache line is bigger than 64B.
The EQE size will then be the cache line size, and the context
segment resides in [0-31] offset.
Signed-off-by: default avatarIdo Shamay <idos@mellanox.com>
Signed-off-by: default avatarJack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: default avatarOr Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      43c816c6
    • Ido Shamay's avatar
      net/mlx4_core: Enable CQE/EQE stride support · 77507aa2
      Ido Shamay authored 
      This feature is intended for archs having cache line larger then 64B.

Since our CQE/EQEs are generally 64B in those systems, HW will write
twice to the same cache line consecutively, causing pipe locks due to
he hazard prevention mechanism. For elements in a cyclic buffer, writes
are consecutive, so entries smaller than a cache line should be
avoided, especially if they are written at a high rate.

Reduce consecutive writes to same cache line in CQs/EQs, by allowing the
driver to increase the distance between entries so that each will reside
in a different cache line. Until the introduction of this feature, there
were two types of CQE/EQE:

1. 32B stride and context in the [0-31] segment
2. 64B stride and context in the [32-63] segment

This feature introduces two additional types:

3. 128B stride and context in the [0-31] segment (128B cache line)
4. 256B stride and context in the [0-31] segment (256B cache line)

Modify the mlx4_core driver to query the device for the CQE/EQE cache
line stride capability and to enable that capability when the host
cache line size is larger than 64 bytes (supported cache lines are
128B and 256B).

The mlx4 IB driver and libmlx4 need not be aware of this change. The PF
context behaviour is changed to require this change in VF drivers
running on such archs.
Signed-off-by: default avatarIdo Shamay <idos@mellanox.com>
Signed-off-by: default avatarJack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: default avatarOr Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      77507aa2
    • Sabrina Dubroca's avatar
      net: fix sparse warnings in SNMP_UPD_PO_STATS(_BH) · 54003f11
      Sabrina Dubroca authored 
      ptr used to be a non __percpu pointer (result of a this_cpu_ptr
assignment, 7d720c3e ("percpu: add __percpu sparse annotations to
net")). Since d25398df ("net: avoid reloads in SNMP_UPD_PO_STATS"),
that's no longer the case, SNMP_UPD_PO_STATS uses this_cpu_add and ptr
is now __percpu.

Silence sparse warnings by preserving the original type and
annotation, and remove the out-of-date comment.

warning: incorrect type in initializer (different address spaces)
   expected unsigned long long *ptr
   got unsigned long long [noderef] <asn:3>*<noident>
warning: incorrect type in initializer (different address spaces)
   expected void const [noderef] <asn:3>*__vpp_verify
   got unsigned long long *<noident>
warning: incorrect type in initializer (different address spaces)
   expected void const [noderef] <asn:3>*__vpp_verify
   got unsigned long long *<noident>
Signed-off-by: default avatarSabrina Dubroca <sd@queasysnail.net>
Acked-by: default avatarHannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      54003f11
    • David S. Miller's avatar
      Merge branch 'fou-next' · fb5690d2
      David S. Miller authored 
      Tom Herbert says:

====================
net: foo-over-udp (fou)

This patch series implements foo-over-udp. The idea is that we can
encapsulate different IP protocols in UDP packets. The rationale for
this is that networking devices such as NICs and switches are usually
implemented with UDP (and TCP) specific mechanims for processing. For
instance, many switches and routers will implement a 5-tuple hash
for UDP packets to perform Equal Cost Multipath Routing (ECMP) or
RSS (on NICs). Many NICs also only provide rudimentary checksum
offload (basic TCP and UDP packet), with foo-over-udp we may be
able to leverage these NICs to offload checksums of tunneled packets
(using checksum unnecessary conversion and eventually remote checksum
offload)

An example encapsulation of IPIP over FOU is diagrammed below. As
illustrated, the packet overhead for FOU is the 8 byte UDP header.

+------------------+
|    IPv4 hdr      |
+------------------+
|     UDP hdr      |
+------------------+
|    IPv4 hdr      |
+------------------+
|     TCP hdr      |
+------------------+
|   TCP payload    |
+------------------+

Conceptually, FOU should be able to encapsulate any IP protocol.
The FOU header (UDP hdr.) is essentially an inserted header between the
IP header and transport, so in the case of TCP or UDP encapsulation
the pseudo header would be based on the outer IP header and its length
field must not include the UDP header.

* Receive

In this patch set the RX path for FOU is implemented in a new fou
module. To enable FOU for a particular protocol, a UDP-FOU socket is
opened to the port to receive FOU packets. The socket is mapped to the
IP protocol for the packets. The XFRM mechanism used to receive
encapsulated packets (udp_encap_rcv) for the port. Upon reception, the
UDP is removed and packet is reinjected in the stack for the
corresponding protocol associated with the socket (return -protocol
from udp_encap_rcv function).

GRO is provided with the appropriate fou_gro_receive and
fou_gro_complete. These routines need to know the encapsulation
protocol so we save that in udp_offloads structure with the port
and pass it in the napi_gro_cb structure.

* TX

This patch series implements FOU transmit encapsulation for IPIP, GRE, and
SIT. This done by some common infrastructure in ip_tunnel including an
ip_tunnel_encap to perform FOU encapsulation and common configuration
to enable FOU on IP tunnels. FOU is configured on existing tunnels and
does not create any new interfaces. The transmit and receive paths are
independent, so use of FOU may be assymetric between tunnel endpoints.

* Configuration

The fou module using netlink to configure FOU receive ports. The ip
command can be augmented with a fou subcommand to support this. e.g. to
configure FOU for IPIP on port 5555:

  ip fou add port 5555 ipproto 4

GRE, IPIP, and SIT have been modified with netlink commands to
configure use of FOU on transmit. The "ip link" command will be
augmented with an encap subcommand (for supporting various forms of
secondary encapsulation). For instance, to configure an ipip tunnel
with FOU on port 5555:

  ip link add name tun1 type ipip \
    remote 192.168.1.1 local 192.168.1.2 ttl 225 \
    encap fou encap-sport auto encap-dport 5555

* Notes
  - This patch set does not implement GSO for FOU. The UDP encapsulation
    code assumes TEB, so that will need to be reimplemented.
  - When a packet is received through FOU, the UDP header is not
    actually removed for the skbuf, pointers to transport header
    and length in the IP header are updated (like in ESP/UDP RX). A
    side effect is the IP header will now appear to have an incorrect
    checksum by an external observer (e.g. tcpdump), it will be off
    by sizeof UDP header. If necessary we could adjust the checksum
    to compensate.
  - Performance results are below. My expectation is that FOU should
    entail little overhead (clearly there is some work to do :-) ).
    Optimizing UDP socket lookup for encapsulation ports should help
    significantly.
  - I really don't expect/want devices to have special support for any
    of this. Generic checksum offload mechanisms (NETIF_HW_CSUM
    and use of CHECKSUM_COMPLETE) should be sufficient. RSS and flow
    steering is provided by commonly implemented UDP hashing. GRO/GSO
    seem fairly comparable with LRO/TSO already.

* Performance

Ran netperf TCP_RR and TCP_STREAM tests across various configurations.
This was performed on bnx2x and I disabled TSO/GSO on sender to get
fair comparison for FOU versus non-FOU. CPU utilization is reported
for receive in TCP_STREAM.

  GRE
    IPv4, FOU, UDP checksum enabled
      TCP_STREAM
        24.85% CPU utilization
        9310.6 Mbps
      TCP_RR
        94.2% CPU utilization
        155/249/460 90/95/99% latencies
        1.17018e+06 tps
    IPv4, FOU, UDP checksum disabled
      TCP_STREAM
        31.04% CPU utilization
        9302.22 Mbps
      TCP_RR
        94.13% CPU utilization
        154/239/419 90/95/99% latencies
        1.17555e+06 tps
    IPv4, no FOU
      TCP_STREAM
        23.13% CPU utilization
        9354.58 Mbps
      TCP_RR
        90.24% CPU utilization
        156/228/360 90/95/99% latencies
        1.18169e+06 tps

  IPIP
    FOU, UDP checksum enabled
      TCP_STREAM
        24.13% CPU utilization
        9328 Mbps
      TCP_RR
        94.23
        149/237/429 90/95/99% latencies
        1.19553e+06 tps
    FOU, UDP checksum disabled
      TCP_STREAM
        29.13% CPU utilization
        9370.25 Mbps
      TCP_RR
        94.13% CPU utilization
        149/232/398 90/95/99% latencies
        1.19225e+06 tps
    No FOU
      TCP_STREAM
        10.43% CPU utilization
        5302.03 Mbps
      TCP_RR
        51.53% CPU utilization
        215/324/475 90/95/99% latencies
        864998 tps

  SIT
    FOU, UDP checksum enabled
      TCP_STREAM
        30.38% CPU utilization
        9176.76 Mbps
      TCP_RR
        96.9% CPU utilization
        170/281/581 90/95/99% latencies
        1.03372e+06 tps
    FOU, UDP checksum disabled
      TCP_STREAM
        39.6% CPU utilization
        9176.57 Mbps
      TCP_RR
        97.14% CPU utilization
        167/272/548 90/95/99% latencies
        1.03203e+06 tps
    No FOU
      TCP_STREAM
        11.2% CPU utilization
        4636.05 Mbps
      TCP_RR
        59.51% CPU utilization
        232/346/489 90/95/99% latencies
        813199 tps

v2:
  - Removed encap IP tunnel ioctls, configuration is done by netlink
    only.
  - Don't export fou_create and fou_destroy, they are currently
    intended to be called within fou module only.
  - Filled on tunnel netlink structures and functions for new values.

v3:
  - Fixed change logs for some of the patches.
  - Remove inline from fou_gro_receive and fou_gro_complete, let
    compiler decide on these.

v4:
  - Don't need to cast void in fou_from_sock
  - Removed incorrest htons for port in fou_destroy
  - Some minor cleanup for readability
====================
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      fb5690d2
    • Tom Herbert's avatar
      gre: Setup and TX path for gre/UDP foo-over-udp encapsulation · 4565e991
      Tom Herbert authored 
      Added netlink attrs to configure FOU encapsulation for GRE, netlink
handling of these flags, and properly adjust MTU for encapsulation.
ip_tunnel_encap is called from ip_tunnel_xmit to actually perform FOU
encapsulation.
Signed-off-by: default avatarTom Herbert <therbert@google.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      4565e991
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7