Skip to content

cloud-kernel
cloud-kernel

openanolis / cloud-kernel
12 个月 前同步成功

通知 152
Star 36
Fork 7
cloud-kernel
cloud-kernel

体验新版 GitCode，发现更多精彩内容 >>

切换分支/标签
  1. 22 6月, 2017 24 次提交
  3. 21 6月, 2017 16 次提交
    • P
      sock: avoid dirtying incoming_cpu if not needed · 34cfb542
      由 Paolo Abeni 提交于 
      for connected socket, the incoming_cpu field in the sock struct
is not going to change frequently, but we are setting it
unconditionally for each packet.

Since sk_incoming_cpu and sk_flags share the same cacheline,
and the latter is access by udp_recvmsg(), this cause a cache
miss for each packet for UDP connected socket.

With this patch, we set the incoming cpu field only when the
ingress cpu really changes.

This gives a small but measurable performance improvement for
connected UDP socket.
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      34cfb542
    • D
      net: introduce SO_PEERGROUPS getsockopt · 28b5ba2a
      由 David Herrmann 提交于 
      This adds the new getsockopt(2) option SO_PEERGROUPS on SOL_SOCKET to
retrieve the auxiliary groups of the remote peer. It is designed to
naturally extend SO_PEERCRED. That is, the underlying data is from the
same credentials. Regarding its syntax, it is based on SO_PEERSEC. That
is, if the provided buffer is too small, ERANGE is returned and @optlen
is updated. Otherwise, the information is copied, @optlen is set to the
actual size, and 0 is returned.

While SO_PEERCRED (and thus `struct ucred') already returns the primary
group, it lacks the auxiliary group vector. However, nearly all access
controls (including kernel side VFS and SYSVIPC, but also user-space
polkit, DBus, ...) consider the entire set of groups, rather than just
the primary group. But this is currently not possible with pure
SO_PEERCRED. Instead, user-space has to work around this and query the
system database for the auxiliary groups of a UID retrieved via
SO_PEERCRED.

Unfortunately, there is no race-free way to query the auxiliary groups
of the PID/UID retrieved via SO_PEERCRED. Hence, the current user-space
solution is to use getgrouplist(3p), which itself falls back to NSS and
whatever is configured in nsswitch.conf(3). This effectively checks
which groups we *would* assign to the user if it logged in *now*. On
normal systems it is as easy as reading /etc/group, but with NSS it can
resort to quering network databases (eg., LDAP), using IPC or network
communication.

Long story short: Whenever we want to use auxiliary groups for access
checks on IPC, we need further IPC to talk to the user/group databases,
rather than just relying on SO_PEERCRED and the incoming socket. This
is unfortunate, and might even result in dead-locks if the database
query uses the same IPC as the original request.

So far, those recursions / dead-locks have been avoided by using
primitive IPC for all crucial NSS modules. However, we want to avoid
re-inventing the wheel for each NSS module that might be involved in
user/group queries. Hence, we would preferably make DBus (and other IPC
that supports access-management based on groups) work without resorting
to the user/group database. This new SO_PEERGROUPS ioctl would allow us
to make dbus-daemon work without ever calling into NSS.

Cc: Michal Sekletar <msekleta@redhat.com>
Cc: Simon McVittie <simon.mcvittie@collabora.co.uk>
Reviewed-by: NTom Gundersen <teg@jklm.no>
Signed-off-by: NDavid Herrmann <dh.herrmann@gmail.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      28b5ba2a
    • P
      udp: prefetch rmem_alloc in udp_queue_rcv_skb() · dd99e425
      由 Paolo Abeni 提交于 
      On UDP packets processing, if the BH is the bottle-neck, it
always sees a cache miss while updating rmem_alloc; try to
avoid it prefetching the value as soon as we have the socket
available.

Performances under flood with multiple NIC rx queues used are
unaffected, but when a single NIC rx queue is in use, this
gives ~10% performance improvement.
Signed-off-by: NPaolo Abeni <pabeni@redhat.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      dd99e425
    • C
      qede: Fix compilation without QED_RDMA · da2e9cf0
      由 Chad Dupuis 提交于 
      When CONFIG_QED_RDMA isn't defined, we'd hit the following:

 /include/linux/qed/qede_rdma.h:84:19:
 warning: ‘qede_rdma_dev_add’ used but never defined [enabled by default]
 static inline int qede_rdma_dev_add(struct qede_dev *dev);

Fixes: bbfcd1e8 ("qed*: Set rdma generic functions prefix")
Signed-off-by: NChad Dupuis <chad.dupuis@cavium.com>
Signed-off-by: NYuval Mintz <Yuval.Mintz@cavium.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      da2e9cf0
    • H
      r8152: correct the definition · b65c0c9b
      由 hayeswang 提交于 
      Replace VLAN_HLEN and CRC_SIZE with ETH_FCS_LEN.
Signed-off-by: NHayes Wang <hayeswang@realtek.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      b65c0c9b
    • D
      Merge branch '40GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/next-queue · adff41f9
      由 David S. Miller 提交于 
      Jeff Kirsher says:

====================
40GbE Intel Wired LAN Driver Updates 2017-06-20

This series contains updates to i40e and i40evf only.

Björn adds additional XDP support for i40e, by adding pass and drop actions
and XDP_TX action support.

Jake fixes a possible NULL pointer dereference in
i40evf_get_ethtool_stats() which could occur if the VF fails to recover
from a reset, and then a user requests statistics.  Changed the use of
dev_info() to dev_dbg() for vf_capability client routine so that the
standard log is not spammed with this information which "might" cause
administrators to worry.  Also added more code comments to help explain
why udp_port has be in host byte order and to avoid future changes which
may cause this to break.  Fixed the holding of the RTNL lock for the
entire reset routine, reduced the scope so that the reset function will
handle its own lock, so that we do not have to wrap every reference
to i40e_do_reset() with RTNL lock/unlock.

Alice updates flags related to firmware interactions for WoL and admin
queue command address with the correct value.

Sudheer makes a fix to ensure that the array is not accessed past the
size of the array.

Greg fixes the parsing of firmware 4.33 admin queue commmand "Get CEE
DCBX PER CFG" because the firmware now creates the oper_prio_tc nibbles
reversed from those in the CDD Priority Group sub-TLV.

Carolyn adds a check and message to let users know that when in MFP mode,
changing RSS hash input set is not supported.

Shannon makes the partition bandwidth control more generic since it is not
in just one form of multi-function partitioning (MFP).  Also fixes a bug
which was causing the firmware confusion in some reset sequences, when
we were disabling interrupts and we were clearing the whole register.
Instead we should only be clearing the CAUSE_ENA bit when disabling
interrupts.

Filip adds support for OEM firmware version, so that if a OEM specific
adapter is detected, ethtool reports the OEM product version in the
firmware version string instead of etrack id.

Alan fixes a bug where the driver was not correctly exiting overflow
promiscuous mode, which can happen if "too many" MAC filters are added,
putting the driver into overflow promiscuous mode, and the filters are
then removed.  The bug occurs because the conditional for toggling
promiscuous mode was only be executed when enabled and not when it was
disabled.
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      adff41f9
    • D
      Merge branch 'ipmr-ip6mr-add-Netlink-notifications-on-cache-reports' · 2f89a795
      由 David S. Miller 提交于 
      Julien Gomes says:

====================
ipmr/ip6mr: add Netlink notifications on cache reports

Currently, all ipmr/ip6mr cache reports are sent through the
mroute/mroute6 socket only.
This forces the use of a single socket for mroute programming, cache
reports and, regarding ipmr, IGMP messages without Router Alert option
reception.

The present patches are aiming to send Netlink notifications in addition
to the existing igmpmsg/mrt6msg to give user programs a way to handle
cache reports in parallel with multiple sockets other than the
mroute/mroute6 socket.

Changes in v2:
- Changed attributes naming from {IPMRA,IP6MRA}_CACHEREPORTA_* to
  {IPMRA,IP6MRA}_CREPORT_*
- Improved packet data copy to handle non-linear packets in
  ipmr/ip6mr cache report Netlink notification creation
- Added two rtnetlink groups with restricted-binding
- Changed cache report notified groups from RTNL_{IPV4,IPV6}_MROUTE to
  the new restricted groups in ipmr/ip6mr

Changes in v3:
- Put message size calculation for {igmp,mrt6}msg_netlink_event in separate
  functions
- Increased vif id attributes size from u8 to u32
====================
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      2f89a795
    • J
      ip6mr: add netlink notifications on mrt6msg cache reports · dd12d15c
      由 Julien Gomes 提交于 
      Add Netlink notifications on cache reports in ip6mr, in addition to the
existing mrt6msg sent to mroute6_sk.
Send RTM_NEWCACHEREPORT notifications to RTNLGRP_IPV6_MROUTE_R.

MSGTYPE, MIF_ID, SRC_ADDR and DST_ADDR Netlink attributes contain the
same data as their equivalent fields in the mrt6msg header.
PKT attribute is the packet sent to mroute6_sk, without the added
mrt6msg header.
Suggested-by: NRyan Halbrook <halbrook@arista.com>
Signed-off-by: NJulien Gomes <julien@arista.com>
Reviewed-by: NNikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      dd12d15c
    • J
      ipmr: add netlink notifications on igmpmsg cache reports · 5a645dd8
      由 Julien Gomes 提交于 
      Add Netlink notifications on cache reports in ipmr, in addition to the
existing igmpmsg sent to mroute_sk.
Send RTM_NEWCACHEREPORT notifications to RTNLGRP_IPV4_MROUTE_R.

MSGTYPE, VIF_ID, SRC_ADDR and DST_ADDR Netlink attributes contain the
same data as their equivalent fields in the igmpmsg header.
PKT attribute is the packet sent to mroute_sk, without the added igmpmsg
header.
Suggested-by: NRyan Halbrook <halbrook@arista.com>
Signed-off-by: NJulien Gomes <julien@arista.com>
Reviewed-by: NNikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      5a645dd8
    • J
      rtnetlink: add restricted rtnl groups for ipv4 and ipv6 mroute · 5f729eaa
      由 Julien Gomes 提交于 
      Add RTNLGRP_{IPV4,IPV6}_MROUTE_R as two new restricted groups for the
NETLINK_ROUTE family.
Binding to these groups specifically requires CAP_NET_ADMIN to allow
multicast of sensitive messages (e.g. mroute cache reports).
Suggested-by: NNikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: NJulien Gomes <julien@arista.com>
Signed-off-by: NNikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      5f729eaa
    • J
      rtnetlink: add NEWCACHEREPORT message type · 94df30a6
      由 Julien Gomes 提交于 
      New NEWCACHEREPORT message type to be used for cache reports sent
via Netlink, effectively allowing splitting cache report reception from
mroute programming.
Suggested-by: NRyan Halbrook <halbrook@arista.com>
Signed-off-by: NJulien Gomes <julien@arista.com>
Reviewed-by: NNikolay Aleksandrov <nikolay@cumulusnetworks.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      94df30a6
    • A
      tcp: md5: hide unused variable · 083a0326
      由 Arnd Bergmann 提交于 
      Changing from a memcpy to per-member comparison left the
size variable unused:

net/ipv4/tcp_ipv4.c: In function 'tcp_md5_do_lookup':
net/ipv4/tcp_ipv4.c:910:15: error: unused variable 'size' [-Werror=unused-variable]

This does not show up when CONFIG_IPV6 is enabled, but the
variable can be removed either way, along with the now unused
assignment.

Fixes: 6797318e ("tcp: md5: add an address prefix for key lookup")
Signed-off-by: NArnd Bergmann <arnd@arndb.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      083a0326
    • Y
      idsn: fix wrong skb_put() used · e1d20e22
      由 yuan linyu 提交于 
      in my commit b952f4df,
-	*(u8 *)skb_put(skb_out, 1) = (u8)(accm >> 24);	\
+	skb_put(skb_out, (u8)(accm >> 24));	\
it should skb_put_u8()

Fixes: b952f4df ("net: manual clean code which call skb_put_[data:zero])")
Signed-off-by: Nyuan linyu <Linyu.Yuan@alcatel-sbell.com.cn>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>
      e1d20e22
    • J
      i40e: don't hold RTNL lock for the entire reset · dfc4ff64
      由 Jacob Keller 提交于 
      We recently refactored i40e_do_reset() and its friends to be able to
hold the RTNL lock only for the portions that actually need to be
protected. However, a separate refactoring added several new callers of
these functions during the PCIe error recovery and suspend/resume
cycles.

When merging the changes together, it was not noticed that we could
reduce the RTNL scope by letting the reset function handle the lock
itself, as previously it was not possible.

Fix this by replacing these call sites to indicate that the reset
function should handle its own lock. This enables multiple PFs to reset
or resume simultaneously without serializing the resets via the RTNL
lock. The end result is that on systems with lots of PFs and VFs the
resets don't stall waiting for each other to finish.

It is probable that we can also do the same for i40e_do_reset_safe, but
this author did not research that change carefully enough to be
confident.
Signed-off-by: NJacob Keller <jacob.e.keller@intel.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
      dfc4ff64
    • C
      i40e: Handle PE_CRITERR properly with IWARP enabled · 7642984b
      由 Catherine Sullivan 提交于 
      When IWARP is enabled, we weren't clearing the PE_CRITERR, just logging
it and removing it from the mask. We need to do a corer to reset the
PE_CRITERR register, so set the bit for that as we handle the
interrupt.

We should also be checking for the error against the PFINT_ICR0 register,
and only need to clear it in the value getting written to
PFINT_ICR0_ENA.
Signed-off-by: NCatherine Sullivan <catherine.sullivan@intel.com>
Signed-off-by: NMitch Williams <mitch.a.williams@intel.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
      7642984b
    • S
      i40e: clear only cause_ena bit · 2e5c26ea
      由 Shannon Nelson 提交于 
      When disabling interrupts, we should only be clearing the CAUSE_ENA bit,
not clearing the whole register.  Clearing the whole register sets the
NEXTQ_IDX field to 0 instead of 0x7ff which can confuse the Firmware in
some reset sequences.
Signed-off-by: NShannon Nelson <shannon.nelson@intel.com>
Signed-off-by: NMitch Williams <mitch.a.williams@intel.com>
Tested-by: NAndrew Bowers <andrewx.bowers@intel.com>
Signed-off-by: NJeff Kirsher <jeffrey.t.kirsher@intel.com>
      2e5c26ea