Skip to content

L
linux
Switch branch/tag
  1. 09 Jan, 2014 11 commits
  3. 08 Jan, 2014 18 commits
  5. 07 Jan, 2014 11 commits
    • David S. Miller's avatar
      Merge branch 'tipc' · 8752b5ca
      David S. Miller authored 
      Jon Maloy says:

====================
tipc: link setup and failover improvements

This series consists of four unrelated commits with different purposes.

- Commit #1 is purely cosmetic and pedagogic, hopefully making the
  failover/tunneling logics slightly easier to understand.
- Commit #2 fixes a bug that has always been in the code, but was not
  discovered until very recently.
- Commit #3 fixes a non-fatal race issue in the neighbour discovery
  code.
- Commit #4 removes an unnecessary indirection step during link
  startup.
====================
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      8752b5ca
    • Jon Paul Maloy's avatar
      tipc: make link start event synchronous · 581465fa
      Jon Paul Maloy authored 
      When a link is created we delay the start event by launching it
to be executed later in a tasklet. As we hold all the
necessary locks at the moment of creation, and there is no risk
of deadlock or contention, this delay serves no purpose in the
current code.

We remove this obsolete indirection step, and the associated function
link_start(). At the same time, we rename the function tipc_link_stop()
to the more appropriate tipc_link_purge_queues().
Signed-off-by: default avatarJon Maloy <jon.maloy@ericsson.com>
Reviewed-by: default avatarPaul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      581465fa
    • Ying Xue's avatar
      tipc: introduce new spinlock to protect struct link_req · f9a2c80b
      Ying Xue authored 
      Currently, only 'bearer_lock' is used to protect struct link_req in
the function disc_timeout(). This is unsafe, since the member fields
'num_nodes' and 'timer_intv' might be accessed by below three different
threads simultaneously, none of them grabbing bearer_lock in the
critical region:

link_activate()
  tipc_bearer_add_dest()
    tipc_disc_add_dest()
      req->num_nodes++;

tipc_link_reset()
  tipc_bearer_remove_dest()
    tipc_disc_remove_dest()
      req->num_nodes--
      disc_update()
        read req->num_nodes
	write req->timer_intv

disc_timeout()
  read req->num_nodes
  read/write req->timer_intv

Without lock protection, the only symptom of a race is that discovery
messages occasionally may not be sent out. This is not fatal, since such
messages are best-effort anyway. On the other hand, since discovery
messages are not time critical, adding a protecting lock brings no
serious overhead either. So we add a new, dedicated spinlock in
order to guarantee absolute data consistency in link_req objects.
This also helps reduce the overall role of the bearer_lock, which
we want to remove completely in a later commit series.
Signed-off-by: default avatarYing Xue <ying.xue@windriver.com>
Reviewed-by: default avatarPaul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: default avatarJon Maloy <jon.maloy@ericsson.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      f9a2c80b
    • Jon Paul Maloy's avatar
      tipc: remove 'has_redundant_link' flag from STATE link protocol messages · b9d4c339
      Jon Paul Maloy authored 
      The flag 'has_redundant_link' is defined only in RESET and ACTIVATE
protocol messages. Due to an ambiguity in the protocol specification it
is currently also transferred in STATE messages. Its value is used to
initialize a link state variable, 'permit_changeover', which is used
to inhibit futile link failover attempts when it is known that the
peer node has no working links at the moment, although the local node
may still think it has one.

The fact that 'has_redundant_link' incorrectly is read from STATE
messages has the effect that 'permit_changeover' sometimes gets a wrong
value, and permanently blocks any links from being re-established. Such
failures can only occur in in dual-link systems, and are extremely rare.
This bug seems to have always been present in the code.

Furthermore, since commit b4b56102
("tipc: Ensure both nodes recognize loss of contact between them"),
the 'permit_changeover' field serves no purpose any more. The task of
enforcing 'lost contact' cycles at both peer endpoints is now taken
by a new mechanism, using the flags WAIT_NODE_DOWN and WAIT_PEER_DOWN
in struct tipc_node to abort unnecessary failover attempts.

We therefore remove the 'has_redundant_link' flag from STATE messages,
as well as the now redundant 'permit_changeover' variable.
Signed-off-by: default avatarJon Maloy <jon.maloy@ericsson.com>
Reviewed-by: default avatarYing Xue <ying.xue@windriver.com>
Reviewed-by: default avatarPaul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      b9d4c339
    • Jon Paul Maloy's avatar
      tipc: rename functions related to link failover and improve comments · 170b3927
      Jon Paul Maloy authored 
      The functionality related to link addition and failover is unnecessarily
hard to understand and maintain. We try to improve this by renaming
some of the functions, at the same time adding or improving the
explanatory comments around them. Names such as "tipc_rcv()" etc. also
align better with what is used in other networking components.

The changes in this commit are purely cosmetic, no functional changes
are made.
Signed-off-by: default avatarJon Maloy <jon.maloy@ericsson.com>
Reviewed-by: default avatarYing Xue <ying.xue@windriver.com>
Reviewed-by: default avatarPaul Gortmaker <paul.gortmaker@windriver.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      170b3927
    • Daniel Borkmann's avatar
      net: skbuff: const-ify casts in skb_queue_* functions · fd44b93c
      Daniel Borkmann authored 
      We should const-ify comparisons on skb_queue_* inline helper
functions as their parameters are const as well, so lets not
drop that.
Suggested-by: default avatarBrad Spengler <spender@grsecurity.net>
Signed-off-by: default avatarDaniel Borkmann <dborkman@redhat.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      fd44b93c
    • Daniel Borkmann's avatar
      net: xfrm: xfrm_policy: fix inline not at beginning of declaration · be7928d2
      Daniel Borkmann authored 
      Fix three warnings related to:

  net/xfrm/xfrm_policy.c:1644:1: warning: 'inline' is not at beginning of declaration [-Wold-style-declaration]
  net/xfrm/xfrm_policy.c:1656:1: warning: 'inline' is not at beginning of declaration [-Wold-style-declaration]
  net/xfrm/xfrm_policy.c:1668:1: warning: 'inline' is not at beginning of declaration [-Wold-style-declaration]

Just removing the inline keyword is sufficient as the compiler will
decide on its own about inlining or not.
Signed-off-by: default avatarDaniel Borkmann <dborkman@redhat.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      be7928d2
    • Shawn Bohrer's avatar
      mlx4_en: Select PTP_1588_CLOCK · 74b9c3ea
      Shawn Bohrer authored 
      Now that mlx4_en includes a PHC driver it must select PTP_1588_CLOCK.

   drivers/built-in.o: In function `mlx4_en_get_ts_info':
>> en_ethtool.c:(.text+0x391a11): undefined reference to `ptp_clock_index'
   drivers/built-in.o: In function `mlx4_en_remove_timestamp':
>> (.text+0x397913): undefined reference to `ptp_clock_unregister'
   drivers/built-in.o: In function `mlx4_en_init_timestamp':
>> (.text+0x397b20): undefined reference to `ptp_clock_register'

Fixes: ad7d4eae ("mlx4_en: Add PTP hardware clock")
Signed-off-by: default avatarShawn Bohrer <sbohrer@rgmadvisors.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      74b9c3ea
    • Jerry Chu's avatar
      net-gre-gro: Add GRE support to the GRO stack · bf5a755f
      Jerry Chu authored 
      This patch built on top of Commit 299603e8
("net-gro: Prepare GRO stack for the upcoming tunneling support") to add
the support of the standard GRE (RFC1701/RFC2784/RFC2890) to the GRO
stack. It also serves as an example for supporting other encapsulation
protocols in the GRO stack in the future.

The patch supports version 0 and all the flags (key, csum, seq#) but
will flush any pkt with the S (seq#) flag. This is because the S flag
is not support by GSO, and a GRO pkt may end up in the forwarding path,
thus requiring GSO support to break it up correctly.

Currently the "packet_offload" structure only contains L3 (ETH_P_IP/
ETH_P_IPV6) GRO offload support so the encapped pkts are limited to
IP pkts (i.e., w/o L2 hdr). But support for other protocol type can
be easily added, so is the support for GRE variations like NVGRE.

The patch also support csum offload. Specifically if the csum flag is on
and the h/w is capable of checksumming the payload (CHECKSUM_COMPLETE),
the code will take advantage of the csum computed by the h/w when
validating the GRE csum.

Note that commit 60769a5d "ipv4: gre:
add GRO capability" already introduces GRO capability to IPv4 GRE
tunnels, using the gro_cells infrastructure. But GRO is done after
GRE hdr has been removed (i.e., decapped). The following patch applies
GRO when pkts first come in (before hitting the GRE tunnel code). There
is some performance advantage for applying GRO as early as possible.
Also this approach is transparent to other subsystem like Open vSwitch
where GRE decap is handled outside of the IP stack hence making it
harder for the gro_cells stuff to apply. On the other hand, some NICs
are still not capable of hashing on the inner hdr of a GRE pkt (RSS).
In that case the GRO processing of pkts from the same remote host will
all happen on the same CPU and the performance may be suboptimal.

I'm including some rough preliminary performance numbers below. Note
that the performance will be highly dependent on traffic load, mix as
usual. Moreover it also depends on NIC offload features hence the
following is by no means a comprehesive study. Local testing and tuning
will be needed to decide the best setting.

All tests spawned 50 copies of netperf TCP_STREAM and ran for 30 secs.
(super_netperf 50 -H 192.168.1.18 -l 30)

An IP GRE tunnel with only the key flag on (e.g., ip tunnel add gre1
mode gre local 10.246.17.18 remote 10.246.17.17 ttl 255 key 123)
is configured.

The GRO support for pkts AFTER decap are controlled through the device
feature of the GRE device (e.g., ethtool -K gre1 gro on/off).

1.1 ethtool -K gre1 gro off; ethtool -K eth0 gro off
thruput: 9.16Gbps
CPU utilization: 19%

1.2 ethtool -K gre1 gro on; ethtool -K eth0 gro off
thruput: 5.9Gbps
CPU utilization: 15%

1.3 ethtool -K gre1 gro off; ethtool -K eth0 gro on
thruput: 9.26Gbps
CPU utilization: 12-13%

1.4 ethtool -K gre1 gro on; ethtool -K eth0 gro on
thruput: 9.26Gbps
CPU utilization: 10%

The following tests were performed on a different NIC that is capable of
csum offload. I.e., the h/w is capable of computing IP payload csum
(CHECKSUM_COMPLETE).

2.1 ethtool -K gre1 gro on (hence will use gro_cells)

2.1.1 ethtool -K eth0 gro off; csum offload disabled
thruput: 8.53Gbps
CPU utilization: 9%

2.1.2 ethtool -K eth0 gro off; csum offload enabled
thruput: 8.97Gbps
CPU utilization: 7-8%

2.1.3 ethtool -K eth0 gro on; csum offload disabled
thruput: 8.83Gbps
CPU utilization: 5-6%

2.1.4 ethtool -K eth0 gro on; csum offload enabled
thruput: 8.98Gbps
CPU utilization: 5%

2.2 ethtool -K gre1 gro off

2.2.1 ethtool -K eth0 gro off; csum offload disabled
thruput: 5.93Gbps
CPU utilization: 9%

2.2.2 ethtool -K eth0 gro off; csum offload enabled
thruput: 5.62Gbps
CPU utilization: 8%

2.2.3 ethtool -K eth0 gro on; csum offload disabled
thruput: 7.69Gbps
CPU utilization: 8%

2.2.4 ethtool -K eth0 gro on; csum offload enabled
thruput: 8.96Gbps
CPU utilization: 5-6%
Signed-off-by: default avatarH.K. Jerry Chu <hkchu@google.com>
Reviewed-by: default avatarEric Dumazet <edumazet@google.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      bf5a755f
    • Benjamin Poirier's avatar
      net: Do not enable tx-nocache-copy by default · cdb3f4a3
      Benjamin Poirier authored 
      There are many cases where this feature does not improve performance or even
reduces it.

For example, here are the results from tests that I've run using 3.12.6 on one
Intel Xeon W3565 and one i7 920 connected by ixgbe adapters. The results are
from the Xeon, but they're similar on the i7. All numbers report the
mean±stddev over 10 runs of 10s.

1) latency tests similar to what is described in "c6e1a0d1 net: Allow no-cache
copy from user on transmit"
There is no statistically significant difference between tx-nocache-copy
on/off.
nic irqs spread out (one queue per cpu)

200x netperf -r 1400,1
tx-nocache-copy off
        692000±1000 tps
        50/90/95/99% latency (us): 275±2/643.8±0.4/799±1/2474.4±0.3
tx-nocache-copy on
        693000±1000 tps
        50/90/95/99% latency (us): 274±1/644.1±0.7/800±2/2474.5±0.7

200x netperf -r 14000,14000
tx-nocache-copy off
        86450±80 tps
        50/90/95/99% latency (us): 334.37±0.02/838±1/2100±20/3990±40
tx-nocache-copy on
        86110±60 tps
        50/90/95/99% latency (us): 334.28±0.01/837±2/2110±20/3990±20

2) single stream throughput tests
tx-nocache-copy leads to higher service demand

                        throughput  cpu0        cpu1        demand
                        (Gb/s)      (Gcycle)    (Gcycle)    (cycle/B)

nic irqs and netperf on cpu0 (1x netperf -T0,0 -t omni -- -d send)

tx-nocache-copy off     9402±5      9.4±0.2                 0.80±0.01
tx-nocache-copy on      9403±3      9.85±0.04               0.838±0.004

nic irqs on cpu0, netperf on cpu1 (1x netperf -T1,1 -t omni -- -d send)

tx-nocache-copy off     9401±5      5.83±0.03   5.0±0.1     0.923±0.007
tx-nocache-copy on      9404±2      5.74±0.03   5.523±0.009 0.958±0.002

As a second example, here are some results from Eric Dumazet with latest
net-next.
tx-nocache-copy also leads to higher service demand

(cpu is Intel(R) Xeon(R) CPU X5660  @ 2.80GHz)

lpq83:~# ./ethtool -K eth0 tx-nocache-copy on
lpq83:~# perf stat ./netperf -H lpq84 -c
MIGRATED TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to lpq84.prod.google.com () port 0 AF_INET
Recv   Send    Send                          Utilization       Service Demand
Socket Socket  Message  Elapsed              Send     Recv     Send    Recv
Size   Size    Size     Time     Throughput  local    remote   local   remote
bytes  bytes   bytes    secs.    10^6bits/s  % S      % U      us/KB   us/KB

 87380  16384  16384    10.00      9407.44   2.50     -1.00    0.522   -1.000

 Performance counter stats for './netperf -H lpq84 -c':

       4282.648396 task-clock                #    0.423 CPUs utilized
             9,348 context-switches          #    0.002 M/sec
                88 CPU-migrations            #    0.021 K/sec
               355 page-faults               #    0.083 K/sec
    11,812,797,651 cycles                    #    2.758 GHz                     [82.79%]
     9,020,522,817 stalled-cycles-frontend   #   76.36% frontend cycles idle    [82.54%]
     4,579,889,681 stalled-cycles-backend    #   38.77% backend  cycles idle    [67.33%]
     6,053,172,792 instructions              #    0.51  insns per cycle
                                             #    1.49  stalled cycles per insn [83.64%]
       597,275,583 branches                  #  139.464 M/sec                   [83.70%]
         8,960,541 branch-misses             #    1.50% of all branches         [83.65%]

      10.128990264 seconds time elapsed

lpq83:~# ./ethtool -K eth0 tx-nocache-copy off
lpq83:~# perf stat ./netperf -H lpq84 -c
MIGRATED TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to lpq84.prod.google.com () port 0 AF_INET
Recv   Send    Send                          Utilization       Service Demand
Socket Socket  Message  Elapsed              Send     Recv     Send    Recv
Size   Size    Size     Time     Throughput  local    remote   local   remote
bytes  bytes   bytes    secs.    10^6bits/s  % S      % U      us/KB   us/KB

 87380  16384  16384    10.00      9412.45   2.15     -1.00    0.449   -1.000

 Performance counter stats for './netperf -H lpq84 -c':

       2847.375441 task-clock                #    0.281 CPUs utilized
            11,632 context-switches          #    0.004 M/sec
                49 CPU-migrations            #    0.017 K/sec
               354 page-faults               #    0.124 K/sec
     7,646,889,749 cycles                    #    2.686 GHz                     [83.34%]
     6,115,050,032 stalled-cycles-frontend   #   79.97% frontend cycles idle    [83.31%]
     1,726,460,071 stalled-cycles-backend    #   22.58% backend  cycles idle    [66.55%]
     2,079,702,453 instructions              #    0.27  insns per cycle
                                             #    2.94  stalled cycles per insn [83.22%]
       363,773,213 branches                  #  127.757 M/sec                   [83.29%]
         4,242,732 branch-misses             #    1.17% of all branches         [83.51%]

      10.128449949 seconds time elapsed

CC: Tom Herbert <therbert@google.com>
Signed-off-by: default avatarBenjamin Poirier <bpoirier@suse.de>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      cdb3f4a3
    • Jiri Pirko's avatar
      ipv4: loopback device: ignore value changes after device is upped · dfd1582d
      Jiri Pirko authored 
      When lo is brought up, new ifa is created. Then, devconf and neigh values
bitfield should be set so later changes of default values would not
affect lo values.

Note that the same behaviour is in ipv6. Also note that this is likely
not an issue in many distros (for example Fedora 19) because userspace
sets address to lo manually before bringing it up.
Signed-off-by: default avatarJiri Pirko <jiri@resnulli.us>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      dfd1582d
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7