Commit ab7db91705 ("virtio-net: auto-tune mergeable rx buffer size for
improved performance") introduced a virtio-net dependency on EWMA.
The inclusion of EWMA is controlled by CONFIG_AVERAGE. Fix build error
when CONFIG_AVERAGE is not enabled by adding select AVERAGE to
virtio-net's Kconfig entry.
Build failure reported using config make ARCH=s390 defconfig.
Signed-off-by: Michael Dalton <mwdalton@google.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Aaron Brown says:
====================
Intel Wired LAN Driver Updates
This series contains an updates to ixgbe and ixgbevf.
Jacob add braces around some ixgbe_qv_lock_* calls lto better adhere
to Kernel style guidelines. Don bumps the versions on ixgbe and
ixgbevf to match internal driver functionality better.
====================
Reviewed-by: Ding Tianhong <dingtianhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Bump the version number to better match functionality provided with out of
tree driver of the same version.
Signed-off-by: Don Skidmore <donald.c.skidmore@intel.com>
Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com>
Signed-off-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Bump the version number to better match functionality provided with out
of tree driver of the same version.
Signed-off-by: Don Skidmore <donald.c.skidmore@intel.com>
Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com>
Signed-off-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds braces around the ixgbe_qv_lock_* calls which previously only
had braces around the if portion. Kernel style guidelines for this require
parenthesis around all conditions if they are required around one. In addition
the comment while not illegal C syntax makes the code look wrong at a cursory
glance. This patch corrects the style and adds braces so that the full if-else
block is uniform.
Signed-off-by: Jacob Keller <jacob.e.keller@intel.com>
Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com>
Signed-off-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The s390 bpf jit compiler emits the signed divide instructions "dr" and "d"
for unsigned divisions.
This can cause problems: the dividend will be zero extended to a 64 bit value
and the divisor is the 32 bit signed value as specified A or X accumulator,
even though A and X are supposed to be treated as unsigned values.
The divide instrunctions will generate an exception if the result cannot be
expressed with a 32 bit signed value.
This is the case if e.g. the dividend is 0xffffffff and the divisor either 1
or also 0xffffffff (signed: -1).
To avoid all these issues simply use unsigned divide instructions.
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In order to get correct drop monitor notifications for dropped
packets, we should call kfree_skb() instead of dev_kfree_skb()
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Scott Feldman says:
====================
bonding: add slave netlink and sysfs support
v2:
- Address review comment from Ding (and Veacesiav): handle kobj cleanup
if sysfs_create_file() fails when adding slave attribute nodes.
v1:
The following series adds bonding slave netlink and sysfs interfaces.
Slave interfaces get a new IFLA_SLAVE set of netlink attributes, along
with RTM_NEWLINK notification when slave's active status changes. The
sysfs interface adds read-only nodes for slave attributes under a /slave
dir, simliar to how bond interfaces get a /bonding dir for bonding
attributes.
====================
Reviewed-by: Ding Tianhong <dingtianhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If link is IFF_SLAVE, extend link dev netlink attributes to include
slave attributes with new IFLA_SLAVE nest. Add netlink notification
(RTM_NEWLINK) when slave status changes from backup to active, or
visa-versa.
Adds new ndo_get_slave op to net_device_ops to fill skb with IFLA_SLAVE
attributes. Currently only used by bonding driver, but could be
used by other aggregating devices with slaves.
Signed-off-by: Scott Feldman <sfeldma@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add sub-directory under /sys/class/net/<interface>/slave with
read-only attributes for slave. Directory only appears when
<interface> is a slave.
$ tree /sys/class/net/eth2/slave/
/sys/class/net/eth2/slave/
├── ad_aggregator_id
├── link_failure_count
├── mii_status
├── perm_hwaddr
├── queue_id
└── state
$ cat /sys/class/net/eth2/slave/*
2
0
up
40:02:10:ef:06:01
0
active
Signed-off-by: Scott Feldman <sfeldma@cumulusnetworks.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Jesse Brandeburg reported that commit acaf4e7099 caused a panic
when adding a network namespace while vxlan module was present in
the system:
[<ffffffff814d0865>] vxlan_lowerdev_event+0xf5/0x100
[<ffffffff816e9e5d>] notifier_call_chain+0x4d/0x70
[<ffffffff810912be>] __raw_notifier_call_chain+0xe/0x10
[<ffffffff810912d6>] raw_notifier_call_chain+0x16/0x20
[<ffffffff815d9610>] call_netdevice_notifiers_info+0x40/0x70
[<ffffffff815d9656>] call_netdevice_notifiers+0x16/0x20
[<ffffffff815e1bce>] register_netdevice+0x1be/0x3a0
[<ffffffff815e1dce>] register_netdev+0x1e/0x30
[<ffffffff814cb94a>] loopback_net_init+0x4a/0xb0
[<ffffffffa016ed6e>] ? lockd_init_net+0x6e/0xb0 [lockd]
[<ffffffff815d6bac>] ops_init+0x4c/0x150
[<ffffffff815d6d23>] setup_net+0x73/0x110
[<ffffffff815d725b>] copy_net_ns+0x7b/0x100
[<ffffffff81090e11>] create_new_namespaces+0x101/0x1b0
[<ffffffff81090f45>] copy_namespaces+0x85/0xb0
[<ffffffff810693d5>] copy_process.part.26+0x935/0x1500
[<ffffffff811d5186>] ? mntput+0x26/0x40
[<ffffffff8106a15c>] do_fork+0xbc/0x2e0
[<ffffffff811b7f2e>] ? ____fput+0xe/0x10
[<ffffffff81089c5c>] ? task_work_run+0xac/0xe0
[<ffffffff8106a406>] SyS_clone+0x16/0x20
[<ffffffff816ee689>] stub_clone+0x69/0x90
[<ffffffff816ee329>] ? system_call_fastpath+0x16/0x1b
Apparently loopback device is being registered first and thus we
receive an event notification when vxlan_net is not ready. Hence,
when we call net_generic() and request vxlan_net_id, we seem to
access garbage at that point in time. In setup_net() where we set
up a newly allocated network namespace, we traverse the list of
pernet ops ...
list_for_each_entry(ops, &pernet_list, list) {
error = ops_init(ops, net);
if (error < 0)
goto out_undo;
}
... and loopback_net_init() is invoked first here, so in the middle
of setup_net() we get this notification in vxlan. As currently we
only care about devices that unregister, move access through
net_generic() there. Fix is based on Cong Wang's proposal, but
only changes what is needed here. It sucks a bit as we only work
around the actual cure: right now it seems the only way to check if
a netns actually finished traversing all init ops would be to check
if it's part of net_namespace_list. But that I find quite expensive
each time we go through a notifier callback. Anyway, did a couple
of tests and it seems good for now.
Fixes: acaf4e7099 ("net: vxlan: when lower dev unregisters remove vxlan dev as well")
Reported-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: Daniel Borkmann <dborkman@redhat.com>
Tested-by: Jesse Brandeburg <jesse.brandeburg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Aaron Brown says:
====================
Intel Wired LAN Driver Updates
This series contains updates to ixgbe Ethan Zhao. The first one replaces
the magic number "63" with a macro, IXGBE_MAX_VFS_DRV_LIMIT, the second
moves the call to set driver_max_VFS to before SRIOV is enabled.
The code of these patches match the v3 (1/2) and v2 (2/2) versions sent
to the e1000-devel and netdev mailing lists. The intermediate versions
(v4, v5) are from sorting out style issues, mostly tabs to spaces and
split lines probably introduced via mailer.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
commit 43dc4e01 Limit number of reported VFs to device
specific value It doesn't work and always returns -EBUSY because VFs are
already enabled.
ixgbe_enable_sriov()
pci_enable_sriov()
sriov_enable()
{
... ..
iov->ctrl |= PCI_SRIOV_CTRL_VFE | PCI_SRIOV_CTRL_MSE;
pci_cfg_access_lock(dev);
... ...
}
pci_sriov_set_totalvfs()
{
... ...
if (dev->sriov->ctrl & PCI_SRIOV_CTRL_VFE)
return -EBUSY;
...
}
So should set driver_max_VFs with pci_sriov_set_totalvfs() before
enable VFs with ixgbe_enable_sriov().
V2: revised for net-next tree.
Signed-off-by: Ethan Zhao <ethan.kernel@gmail.com>
Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com>
Signed-off-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Because ixgbe driver limit the max number of VF
functions could be enabled to 63, so define one macro IXGBE_MAX_VFS_DRV_LIMIT
and cleanup the const 63 in code.
v3: revised for net-next tree.
Signed-off-by: Ethan Zhao <ethan.kernel@gmail.com>
Tested-by: Phil Schmitt <phillip.j.schmitt@intel.com>
Signed-off-by: Aaron Brown <aaron.f.brown@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch :
1) Remove a dst leak if DST_NOCACHE was set on dst
Fix this by holding a reference only if dst really cached.
2) Remove a lockdep warning in __tunnel_dst_set()
This was reported by Cong Wang.
3) Remove usage of a spinlock where xchg() is enough
4) Remove some spurious inline keywords.
Let compiler decide for us.
Fixes: 7d442fab0a ("ipv4: Cache dst in tunnels")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Cong Wang <cwang@twopensource.com>
Cc: Tom Herbert <therbert@google.com>
Cc: Maciej Żenczykowski <maze@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The r7s72100 SoC includes a fast ethernet controller.
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
Acked-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Return a boolean from sh_eth_is_gether() and refactor it as a one-liner.
Signed-off-by: Simon Horman <horms+renesas@verge.net.au>
Acked-by: Sergei Shtylyov <sergei.shtylyov@cogentembedded.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The RFC 3810 defines two type of messages for multicast
listeners. The "Current State Report" message, as the name
implies, refreshes the *current* state to the querier.
Since the querier sends Query messages periodically, there
is no need to retransmit the report.
On the other hand, any change should be reported immediately
using "State Change Report" messages. Since it's an event
triggered by a change and that it can be affected by packet
loss, the rfc states it should be retransmitted [RobVar] times
to make sure routers will receive timely.
Currently, we are sending "Current State Reports" after
DAD is completed. Before that, we send messages using
unspecified address (::) which should be silently discarded
by routers.
This patch changes to send "State Change Report" messages
after DAD is completed fixing the behavior to be RFC compliant
and also to pass TAHI IPv6 testsuite.
Signed-off-by: Flavio Leitner <fbl@redhat.com>
Acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
SO_MAX_PACING_RATE definition on parisc got a typo.
Its not too late to fix it, before 3.13 is official.
Fixes: 62748f32d5 ("net: introduce SO_MAX_PACING_RATE")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In commit 1ec047eb47 ("ipv6: introduce per-interface counter for
dad-completed ipv6 addresses") I build the detection of the first
operational link-local address much to complex. Additionally this code
now has a race condition.
Replace it with a much simpler variant, which just scans the address
list when duplicate address detection completes, to check if this is
the first valid link local address and send RS and MLD reports then.
Fixes: 1ec047eb47 ("ipv6: introduce per-interface counter for dad-completed ipv6 addresses")
Reported-by: Jiri Pirko <jiri@resnulli.us>
Cc: Flavio Leitner <fbl@redhat.com>
Signed-off-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Acked-by: Flavio Leitner <fbl@redhat.com>
Acked-by: Jiri Pirko <jiri@resnulli.us>
Signed-off-by: David S. Miller <davem@davemloft.net>
Because the tcp-metrics is an RCU-list, it may be that two
soft-interrupts are inside __tcp_get_metrics() for the same
destination-IP at the same time. If this destination-IP is not yet part of
the tcp-metrics, both soft-interrupts will end up in tcpm_new and create
a new entry for this IP.
So, we will have two tcp-metrics with the same destination-IP in the list.
This patch checks twice __tcp_get_metrics(). First without holding the
lock, then while holding the lock. The second one is there to confirm
that the entry has not been added by another soft-irq while waiting for
the spin-lock.
Fixes: 51c5d0c4b1 (tcp: Maintain dynamic metrics in local cache.)
Signed-off-by: Christoph Paasch <christoph.paasch@uclouvain.be>
Reviewed-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Remove function qlcnic_enable_eswitch which was defined
but never used in current code.
Compile tested only.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Functions only used in one file should be static.
Found by running make namespacecheck
Compile tested only.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch is following the commit b903d324be (ipv6: tcp: fix TCLASS
value in ACK messages sent from TIME_WAIT).
For the same reason than tclass, we have to store the flow label in the
inet_timewait_sock to provide consistency of flow label on the last ACK.
Signed-off-by: Florent Fourcot <florent.fourcot@enst-bretagne.fr>
Signed-off-by: David S. Miller <davem@davemloft.net>
commit ae4b46e9d "net: rds: use this_cpu_* per-cpu helper" broke per-cpu
handling for rds. chpfirst is the result of __this_cpu_read(), so it is
an absolute pointer and not __percpu. Therefore, __this_cpu_write()
should not operate on chpfirst, but rather on cache->percpu->first, just
like __this_cpu_read() did before.
Cc: <stable@vger.kernel.org> # 3.8+
Signed-off-byd Gerald Schaefer <gerald.schaefer@de.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
John W. Linville says:
====================
Please pull this batch of updates for the 3.14 stream!
For the mac80211 bits, Johannes says:
"This time I have uAPSD fixes since I was working on that, hwsim
improvements to make dynamic radios possible for the test suite, the
evidently long-overdue channel_change_time removal and a few other small
collected fix and improvements."
For the iwlwifi bits, Emmanuel says:
"Besides a few trivial patches, I have an important workaround for a HW
issue that has kept me busy for a long time. Along with it, a fix that
prevents an error from being printed.
Eyal fixes our behavior against SISO APs and Ilan fixes an issue with
multiple interface scenarios.
Eliad fixes an error path in our init flow.
We also have a few 'static analyzers' fix."
For the NFC bits, Samuel says:
"It includes:
* A new NFC driver for Marvell's 8897, and a few NCI fixes and
improvements needed to support this chipset.
* An LLCP fix for how we were setting the default MIU on a p2p link. If
there is no explicit MIU extension announced at connection time, we
must use the default one and not the one announced at LLCP link
establishement time.
* A pn544 EEPROM config update. Some of the currently EEPROM configured
values are overwriting the firmware ones while other should not be set
by the driver itself.
* Some NFC digital stack fixes and improvements. Asynchronous functions
are better documented, RF technologies and CRC functions are set upon
PSL_REQ reception, and a few minor bugs are fixed.
* Minor and miscelaneous pn533, mei_phy and port100 fixes."
For the ath bits, Kalle says:
"Janusz added Kconfig option for DFS. The DFS code was there already, but
after fixes to mac80211 we can now enable it.
Bartosz added a runtime firmware feature flag to disable P2P. Our 10.1
firmware branch doesn't support P2P and ath10k can now disable that. He
also added a limit for how many clients can connect to ath10k AP.
Michal fixed WEP shared authentication, in case someone still uses it.
And I added firmware debug log to help the firmware engineers."
Along with that is a small batch of ath9k updates and a few other bits
here and there.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Pull namespace fixes from Eric Biederman:
"This is a set of 3 regression fixes.
This fixes /proc/mounts when using "ip netns add <netns>" to display
the actual mount point.
This fixes a regression in clone that broke lxc-attach.
This fixes a regression in the permission checks for mounting /proc
that made proc unmountable if binfmt_misc was in use. Oops.
My apologies for sending this pull request so late. Al Viro gave
interesting review comments about the d_path fix that I wanted to
address in detail before I sent this pull request. Unfortunately a
bad round of colds kept from addressing that in detail until today.
The executive summary of the review was:
Al: Is patching d_path really sufficient?
The prepend_path, d_path, d_absolute_path, and __d_path family of
functions is a really mess.
Me: Yes, patching d_path is really sufficient. Yes, the code is mess.
No it is not appropriate to rewrite all of d_path for a regression
that has existed for entirely too long already, when a two line
change will do"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace:
vfs: Fix a regression in mounting proc
fork: Allow CLONE_PARENT after setns(CLONE_NEWPID)
vfs: In d_path don't call d_dname on a mount point
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.14 (GNU/Linux)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=ifJc
-----END PGP SIGNATURE-----
Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm
Pull KVM fix from Paolo Bonzini:
"Fix for a brown paper bag bug. Thanks to Drew Jones for noticing"
* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
kvm: x86: fix apic_base enable check
Michael Dalton says:
====================
virtio-net: mergeable rx buffer size auto-tuning
The virtio-net device currently uses aligned MTU-sized mergeable receive
packet buffers. Network throughput for workloads with large average
packet size can be improved by posting larger receive packet buffers.
However, due to SKB truesize effects, posting large (e.g, PAGE_SIZE)
buffers reduces the throughput of workloads that do not benefit from GRO
and have no large inbound packets.
This patchset introduces virtio-net mergeable buffer size auto-tuning,
with buffer sizes ranging from aligned MTU-size to PAGE_SIZE. Packet
buffer size is chosen based on a per-receive queue EWMA of incoming
packet size.
To unify mergeable receive buffer memory allocation and improve
SKB frag coalescing, all mergeable buffer memory allocation is
migrated to per-receive queue page frag allocators.
The per-receive queue mergeable packet buffer size is exported via
sysfs, and the network device sysfs layer has been extended to add
support for device-specific per-receive queue sysfs attribute groups.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Add initial support for per-rx queue sysfs attributes to virtio-net. If
mergeable packet buffers are enabled, adds a read-only mergeable packet
buffer size sysfs attribute for each RX queue.
Suggested-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael Dalton <mwdalton@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
To ensure ewma_read() without a lock returns a valid but possibly
out of date average, modify ewma_add() by using ACCESS_ONCE to prevent
intermediate wrong values from being written to avg->internal.
Suggested-by: Eric Dumazet <eric.dumazet@gmail.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Michael Dalton <mwdalton@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Extend existing support for netdevice receive queue sysfs attributes to
permit a device-specific attribute group. Initial use case for this
support will be to allow the virtio-net device to export per-receive
queue mergeable receive buffer size.
Signed-off-by: Michael Dalton <mwdalton@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit 2613af0ed1 ("virtio_net: migrate mergeable rx buffers to page frag
allocators") changed the mergeable receive buffer size from PAGE_SIZE to
MTU-size, introducing a single-stream regression for benchmarks with large
average packet size. There is no single optimal buffer size for all
workloads. For workloads with packet size <= MTU bytes, MTU + virtio-net
header-sized buffers are preferred as larger buffers reduce the TCP window
due to SKB truesize. However, single-stream workloads with large average
packet sizes have higher throughput if larger (e.g., PAGE_SIZE) buffers
are used.
This commit auto-tunes the mergeable receiver buffer packet size by
choosing the packet buffer size based on an EWMA of the recent packet
sizes for the receive queue. Packet buffer sizes range from MTU_SIZE +
virtio-net header len to PAGE_SIZE. This improves throughput for
large packet workloads, as any workload with average packet size >=
PAGE_SIZE will use PAGE_SIZE buffers.
These optimizations interact positively with recent commit
ba27524103 ("virtio-net: coalesce rx frags when possible during rx"),
which coalesces adjacent RX SKB fragments in virtio_net. The coalescing
optimizations benefit buffers of any size.
Benchmarks taken from an average of 5 netperf 30-second TCP_STREAM runs
between two QEMU VMs on a single physical machine. Each VM has two VCPUs
with all offloads & vhost enabled. All VMs and vhost threads run in a
single 4 CPU cgroup cpuset, using cgroups to ensure that other processes
in the system will not be scheduled on the benchmark CPUs. Trunk includes
SKB rx frag coalescing.
net-next w/ virtio_net before 2613af0ed1 (PAGE_SIZE bufs): 14642.85Gb/s
net-next (MTU-size bufs): 13170.01Gb/s
net-next + auto-tune: 14555.94Gb/s
Jason Wang also reported a throughput increase on mlx4 from 22Gb/s
using MTU-sized buffers to about 26Gb/s using auto-tuning.
Signed-off-by: Michael Dalton <mwdalton@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The virtio-net driver currently uses netdev_alloc_frag() for GFP_ATOMIC
mergeable rx buffer allocations. This commit migrates virtio-net to use
per-receive queue page frags for GFP_ATOMIC allocation. This change unifies
mergeable rx buffer memory allocation, which now will use skb_refill_frag()
for both atomic and GFP-WAIT buffer allocations.
To address fragmentation concerns, if after buffer allocation there
is too little space left in the page frag to allocate a subsequent
buffer, the remaining space is added to the current allocated buffer
so that the remaining space can be used to store packet data.
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael Dalton <mwdalton@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
skb_page_frag_refill currently permits only order-0 page allocs
unless GFP_WAIT is used. Change skb_page_frag_refill to attempt
higher-order page allocations whether or not GFP_WAIT is used. If
memory cannot be allocated, the allocator will fall back to
successively smaller page allocs (down to order-0 page allocs).
This change brings skb_page_frag_refill in line with the existing
page allocation strategy employed by netdev_alloc_frag, which attempts
higher-order page allocations whether or not GFP_WAIT is set, falling
back to successively lower-order page allocations on failure. Part
of migration of virtio-net to per-receive queue page frag allocators.
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Michael Dalton <mwdalton@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The error code was not set if change indev fail, so the error
condition wasn't reflected in the return value. Fix to return a
negative error code from this error handling case instead of 0.
Fixes: 2519a602c2 ('net_sched: optimize tcf_match_indev()')
Signed-off-by: Wei Yongjun <yongjun_wei@trendmicro.com.cn>
Signed-off-by: David S. Miller <davem@davemloft.net>
Ying Xue says:
====================
tipc: align TIPC behaviours of waiting for events with other stacks
Comparing the current implementations of waiting for events in TIPC
socket layer with other stacks, TIPC's behaviour is very different
because wait_event_interruptible_timeout()/wait_event_interruptible()
are always used by TIPC to wait for events while relevant socket or
port variables are fed to them as their arguments. As socket lock has
to be released temporarily before the two routines of waiting for
events are called, their arguments associated with socket or port
structures are out of socket lock protection. This might cause
serious issues where the process of calling socket syscall such as
sendsmg(), connect(), accept(), and recvmsg(), cannot be waken up
at all even if proper event arrives or improperly be woken up
although the condition of waking up the process is not satisfied
in practice.
Therefore, aligning its behaviours with similar functions implemented
in other stacks, for instance, sk_stream_wait_connect() and
inet_csk_wait_for_connect() etc, can avoid above risks for us.
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Standardize the behaviour of waiting for events in TIPC recvmsg()
so that all variables of socket or port structures are protected
within socket lock, allowing the process of calling recvmsg() to
be woken up at appropriate time.
Signed-off-by: Ying Xue <ying.xue@windriver.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Standardize the behaviour of waiting for events in TIPC send_packet()
so that all variables of socket or port structures are protected within
socket lock, allowing the process of calling sendmsg() to be woken up
at appropriate time.
Signed-off-by: Ying Xue <ying.xue@windriver.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Comparing the behaviour of how to wait for events in TIPC sendmsg()
with other stacks, the TIPC implementation might be perceived as
different, and sometimes even incorrect. For instance, sk_sleep()
and tport->congested variables associated with socket are exposed
without socket lock protection while wait_event_interruptible_timeout()
accesses them. So standardizing it with similar implementation
in other stacks can help us correct these errors which the process
of calling sendmsg() cannot be woken up event if an expected event
arrive at socket or improperly woken up although the wake condition
doesn't match.
Signed-off-by: Ying Xue <ying.xue@windriver.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Comparing the behaviour of how to wait for events in TIPC accept()
with other stacks, the TIPC implementation might be perceived as
different, and sometimes even incorrect. As sk_sleep() and
sk->sk_receive_queue variables associated with socket are not
protected by socket lock, the process of calling accept() may be
woken up improperly or sometimes cannot be woken up at all. After
standardizing it with inet_csk_wait_for_connect routine, we can
get benefits including: avoiding 'thundering herd' phenomenon,
adding a timeout mechanism for accept(), coping with a pending
signal, and having sk_sleep() and sk->sk_receive_queue being
always protected within socket lock scope and so on.
Signed-off-by: Ying Xue <ying.xue@windriver.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Comparing the behaviour of how to wait for events in TIPC connect()
with other stacks, the TIPC implementation might be perceived as
different, and sometimes even incorrect. For instance, as both
sock->state and sk_sleep() are directly fed to
wait_event_interruptible_timeout() as its arguments, and socket lock
has to be released before we call wait_event_interruptible_timeout(),
the two variables associated with socket are exposed out of socket
lock protection, thereby probably getting stale values so that the
process of calling connect() cannot be woken up exactly even if
correct event arrives or it is woken up improperly even if the wake
condition is not satisfied in practice. Therefore, standardizing its
behaviour with sk_stream_wait_connect routine can avoid these risks.
Additionally the implementation of connect routine is simplified as a
whole, allowing it to return correct values in all different cases.
Signed-off-by: Ying Xue <ying.xue@windriver.com>
Reviewed-by: Jon Maloy <jon.maloy@ericsson.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When go the right path, the status is 0, no need to assign it again.
So just remove the assignment.
Signed-off-by: Wang Weidong <wangweidong1@huawei.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
It looks like there's no need for those two fields:
- Unless there's a failure for the first refill try, rq->max should be always
equal to the vring size.
- rq->num is only used to determine the condition that we need to do the refill,
we could check vq->num_free instead.
- rq->num was required to be increased or decreased explicitly after each
get/put which results a bad API.
So this patch removes them both to make the code simpler.
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Acked-by: Rusty Russell <rusty@rustcorp.com.au>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch fixes following sparse warning
davinci_mdio.c:85:27: warning: symbol 'default_pdata' was not declared. Should it be static?
Also makes the default_pdata as a constant.
Signed-off-by: Lad, Prabhakar <prabhakar.csengg@gmail.com>
Acked-by: Mugunthan V N <mugunthanvnm@ti.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently, if a slave's name change, we just pass it by. However, if the
slave is a current primary_slave, then we end up with using a slave, whose
name != params.primary, for primary_slave. And vice-versa, if we don't have
a primary_slave but have params.primary set - we will not detected a new
primary_slave.
Fix this by catching the NETDEV_CHANGENAME event and setting primary_slave
accordingly. Also, if the primary_slave was changed, issue a reselection of
the active slave, cause the priorities have changed.
Reported-by: Ding Tianhong <dingtianhong@huawei.com>
CC: Ding Tianhong <dingtianhong@huawei.com>
CC: Jay Vosburgh <fubar@us.ibm.com>
CC: Andy Gospodarek <andy@greyhouse.net>
Signed-off-by: Veaceslav Falico <vfalico@redhat.com>
Acked-by: Ding Tianhong <dingtianhong@huawei.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In tcf_register_action() we check either ->type or ->kind to see if
there is an existing action registered, but ipt action registers two
actions with same type but different kinds. They should have different
types too.
Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Signed-off-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Cc: Jamal Hadi Salim <jhs@mojatatu.com>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
Acked-by: Jamal Hadi Salim <jhs@mojatatu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
