Skip to content

L
linux
Switch branch/tag
  1. 09 Jul, 2015 5 commits
    • Alexander Duyck's avatar
      net: skb_defer_rx_timestamp should check for phydev before setting up classify · 1007f59d
      Alexander Duyck authored 
      This change makes it so that the call skb_defer_rx_timestamp will first
check for a phydev before going in and manipulating the skb->data and
skb->len values.  By doing this we can avoid unnecessary work on network
devices that don't support phydev.  As a result we reduce the total
instruction count needed to process this on most devices.
Signed-off-by: default avatarAlexander Duyck <alexander.h.duyck@redhat.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      1007f59d
    • Jon Maxwell's avatar
      tcp: v1 always send a quick ack when quickacks are enabled · 2251ae46
      Jon Maxwell authored 
      V1 of this patch contains Eric Dumazet's suggestion to move the per
dst RTAX_QUICKACK check into tcp_in_quickack_mode(). Thanks Eric.

I ran some tests and after setting the "ip route change quickack 1"
knob there were still many delayed ACKs sent. This occured
because when icsk_ack.quick=0 the !icsk_ack.pingpong value is
subsequently ignored as tcp_in_quickack_mode() checks both these
values. The condition for a quick ack to trigger requires
that both icsk_ack.quick != 0 and icsk_ack.pingpong=0. Currently
only icsk_ack.pingpong is controlled by the knob. But the
icsk_ack.quick value changes dynamically depending on heuristics.
The crux of the matter is that delayed acks still cannot be entirely
disabled even with the RTAX_QUICKACK per dst knob enabled. This
patch ensures that a quick ack is always sent when the RTAX_QUICKACK
per dst knob is turned on.

The "ip route change quickack 1" knob was recently added to enable
quickacks. It was modeled around the TCP_QUICKACK setsockopt() option.
This issue is that even with "ip route change quickack 1" enabled
we still see delayed ACKs under some conditions. It would be nice
to be able to completely disable delayed ACKs.

Here is an example:

# netstat -s|grep dela
    3 delayed acks sent

For all routes enable the knob

# ip route change quickack 1

Generate some traffic across a slow link and we still see the delayed
acks.

# netstat -s|grep dela
    106 delayed acks sent
    1 delayed acks further delayed because of locked socket

The issue is that both the "ip route change quickack 1" knob and
the TCP_QUICKACK option set the icsk_ack.pingpong variable to 0.
However at the business end in the __tcp_ack_snd_check() routine,
tcp_in_quickack_mode() checks that both icsk_ack.quick != 0
and icsk_ack.pingpong=0 in order to trigger a quickack. As
icsk_ack.quick is determined by heuristics it can be 0. When
that occurs the icsk_ack.pingpong value is ignored and a delayed
ACK is sent regardless.

This patch moves the RTAX_QUICKACK per dst check into the
tcp_in_quickack_mode() routine which ensures that a quickack is
always sent when the quickack knob is enabled for that dst.
Signed-off-by: default avatarJon Maxwell <jmaxwell37@gmail.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      2251ae46
    • Scott Feldman's avatar
      rocker: add change MTU support · 77a58c74
      Scott Feldman authored 
      Implement ndo_change_mtu: on MTU change, reallocate Rx ring bufs and signal
HW of new port MTU value.
Signed-off-by: default avatarScott Feldman <sfeldma@gmail.com>
Reviewed-by: default avatarSimon Horman <simon.horman@netronome.com>
Tested-by: default avatarSimon Horman <simon.horman@netronome.com>
Acked-by: default avatarJiri Pirko <jiri@resnulli.us>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      77a58c74
    • Vaishali Thakkar's avatar
      neterion: s2io: Use module_pci_driver · 910be1ab
      Vaishali Thakkar authored 
      Use module_pci_driver for drivers whose init and exit functions
only register and unregister, respectively.

A simplified version of the Coccinelle semantic patch that performs
this transformation is as follows:

@A@
identifier f, x;
@@
-static f(...) { return pci_register_driver(&x); }

@b depends on a@
identifier e, a.x;
statement S;
@@
-static e(...) {
-pci_unregister_driver(&x);
-DBG_PRINT(INIT_DBG,"S");
- }

@c depends on a && b@
identifier a.f;
declarer name module_init;
@@
-module_init(f);

@d depends on a && b && c@
identifier b.e, a.x;
declarer name module_exit;
declarer name module_pci_driver;
@@
-module_exit(e);
+module_pci_driver(x);
Signed-off-by: default avatarVaishali Thakkar <vthakkar1994@gmail.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      910be1ab
    • Hariprasad Shenai's avatar
      cxgb4vf: Fix check to use new User Doorbell mechanism · 71d3c0b4
      Hariprasad Shenai authored 
      If we don't have access to the new User GTS (T5+), use the old doorbell
mechanism; otherwise use the new BAR2 mechanism.
Signed-off-by: default avatarHariprasad Shenai <hariprasad@chelsio.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      71d3c0b4
  3. 08 Jul, 2015 26 commits
  5. 03 Jul, 2015 5 commits
  7. 02 Jul, 2015 4 commits
    • Nikolay Aleksandrov's avatar
      bridge: vlan: fix usage of vlan 0 and 4095 again · 462e1ead
      Nikolay Aleksandrov authored 
      Vlan ids 0 and 4095 were disallowed by commit:
8adff41c ("bridge: Don't use VID 0 and 4095 in vlan filtering")
but then the check was removed when vlan ranges were introduced by:
bdced7ef ("bridge: support for multiple vlans and vlan ranges in setlink and dellink requests")
So reintroduce the vlan range check.
Before patch:
[root@testvm ~]# bridge vlan add vid 0 dev eth0 master
(succeeds)
After Patch:
[root@testvm ~]# bridge vlan add vid 0 dev eth0 master
RTNETLINK answers: Invalid argument
Signed-off-by: default avatarNikolay Aleksandrov <nikolay@cumulusnetworks.com>
Fixes: bdced7ef ("bridge: support for multiple vlans and vlan ranges in setlink and dellink requests")
Acked-by: default avatarToshiaki Makita <toshiaki.makita1@gmail.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      462e1ead
    • David S. Miller's avatar
      Merge branch 'for-upstream' of git://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth · c4555d16
      David S. Miller authored 
      Johan Hedberg says:

====================
pull request: bluetooth 2015-07-02

A couple of regressions crept in because of a patch to use proper list
APIs rather than manually reading & writing the next/prev pointers
(commit 835a6a2f). Turns out this was
masking a few bugs: a missing INIT_LIST_HEAD() call and incorrectly
using list_del() rather than list_del_init(). The two patches in this
set fix these, and it'd be nice they could still make it to 4.2-rc1 to
avoid new bug reports from users.

Please let me know if there are any issues pulling. Thanks.
====================
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      c4555d16
    • Govindarajulu Varadarajan's avatar
      enic: fix issues in enic_poll · 25c14ef8
      Govindarajulu Varadarajan authored 
      In enic_poll, we clean tx and rx queues, when low latency busy socket polling
is happening, enic_poll will only clean tx queue. After cleaning tx, it should
return total budget for re-poll.

There is a small window between vnic_intr_unmask() and enic_poll_unlock_napi().
In this window if an irq occurs and napi is scheduled on different cpu, it tries
to acquire enic_poll_lock_napi() and fails. Unlock napi_poll before unmasking
the interrupt.

v2:
Do not change tx wonk done behaviour. Consider only rx work done for completing
napi.
Signed-off-by: default avatarGovindarajulu Varadarajan <_govind@gmx.com>
Signed-off-by: default avatarDavid S. Miller <davem@davemloft.net>
      25c14ef8
    • Linus Torvalds's avatar
      Merge tag 'module-misc-v4.1-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/paulg/linux · a611fb75
      Linus Torvalds authored 
      Pull init.h/module.h fragility fixes from Paul Gortmaker:
 "Fixup various init.h misuses that are fragile wrt code moving to
  module.h

  What started as a removal of no longer required include <linux/init.h>
  due to the earlier __cpuinit and __devinit removal led to the
  observation that some module specfic support was living in init.h
  itself, thus preventing the full removal from introducing compile
  regressions.

  This series includes a few final fixups needed prior to the relocation
  of the modular init code from <init.h> to <module.h>.  These are
  things that weren't easily categorized into any of the other previous
  series categories already requested for pull.

  That said, each fixup branch (including this one) is independent and
  there are no ordering constraints.  Only the final code relocation
  (which is NOT in this pull) requires that all my cleanup branches be
  merged first"

* tag 'module-misc-v4.1-rc8' of git://git.kernel.org/pub/scm/linux/kernel/git/paulg/linux:
  tile: add init.h to usb.c to avoid compile failure
  arm: fix implicit #include <linux/init.h> in entry asm.
  x86: replace __init_or_module with __init in non-modular vsmp_64.c
      a611fb75
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7