Pull networking updates from David Miller:
"Here is my initial pull request for the networking subsystem during
this merge window:
1) Support for ESN in AH (RFC 4302) from Fan Du.
2) Add full kernel doc for ethtool command structures, from Ben
Hutchings.
3) Add BCM7xxx PHY driver, from Florian Fainelli.
4) Export computed TCP rate information in netlink socket dumps, from
Eric Dumazet.
5) Allow IPSEC SA to be dumped partially using a filter, from Nicolas
Dichtel.
6) Convert many drivers to pci_enable_msix_range(), from Alexander
Gordeev.
7) Record SKB timestamps more efficiently, from Eric Dumazet.
8) Switch to microsecond resolution for TCP round trip times, also
from Eric Dumazet.
9) Clean up and fix 6lowpan fragmentation handling by making use of
the existing inet_frag api for it's implementation.
10) Add TX grant mapping to xen-netback driver, from Zoltan Kiss.
11) Auto size SKB lengths when composing netlink messages based upon
past message sizes used, from Eric Dumazet.
12) qdisc dumps can take a long time, add a cond_resched(), From Eric
Dumazet.
13) Sanitize netpoll core and drivers wrt. SKB handling semantics.
Get rid of never-used-in-tree netpoll RX handling. From Eric W
Biederman.
14) Support inter-address-family and namespace changing in VTI tunnel
driver(s). From Steffen Klassert.
15) Add Altera TSE driver, from Vince Bridgers.
16) Optimizing csum_replace2() so that it doesn't adjust the checksum
by checksumming the entire header, from Eric Dumazet.
17) Expand BPF internal implementation for faster interpreting, more
direct translations into JIT'd code, and much cleaner uses of BPF
filtering in non-socket ocntexts. From Daniel Borkmann and Alexei
Starovoitov"
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1976 commits)
netpoll: Use skb_irq_freeable to make zap_completion_queue safe.
net: Add a test to see if a skb is freeable in irq context
qlcnic: Fix build failure due to undefined reference to `vxlan_get_rx_port'
net: ptp: move PTP classifier in its own file
net: sxgbe: make "core_ops" static
net: sxgbe: fix logical vs bitwise operation
net: sxgbe: sxgbe_mdio_register() frees the bus
Call efx_set_channels() before efx->type->dimension_resources()
xen-netback: disable rogue vif in kthread context
net/mlx4: Set proper build dependancy with vxlan
be2net: fix build dependency on VxLAN
mac802154: make csma/cca parameters per-wpan
mac802154: allow only one WPAN to be up at any given time
net: filter: minor: fix kdoc in __sk_run_filter
netlink: don't compare the nul-termination in nla_strcmp
can: c_can: Avoid led toggling for every packet.
can: c_can: Simplify TX interrupt cleanup
can: c_can: Store dlc private
can: c_can: Reduce register access
can: c_can: Make the code readable
...
doubling of the default queue length though.
Cheers,
Rusty.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.14 (GNU/Linux)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=AoMS
-----END PGP SIGNATURE-----
Merge tag 'virtio-next-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux
Pull virtio updates from Rusty Russell:
"Nothing exciting: virtio-blk users might see a bit of a boost from the
doubling of the default queue length though"
* tag 'virtio-next-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux:
virtio-blk: base queue-depth on virtqueue ringsize or module param
Revert a02bbb1ccfe8: MAINTAINERS: add virtio-dev ML for virtio
virtio: fail adding buffer on broken queues.
virtio-rng: don't crash if virtqueue is broken.
virtio_balloon: don't crash if virtqueue is broken.
virtio_blk: don't crash, report error if virtqueue is broken.
virtio_net: don't crash if virtqueue is broken.
virtio_balloon: don't softlockup on huge balloon changes.
virtio: Use pci_enable_msix_exact() instead of pci_enable_msix()
MAINTAINERS: virtio-dev is subscribers only
tools/virtio: add a missing )
tools/virtio: fix missing kmemleak_ignore symbol
tools/virtio: update internal copies of headers
Conflicts:
drivers/net/ethernet/marvell/mvneta.c
The mvneta.c conflict is a case of overlapping changes,
a conversion to devm_ioremap_resource() vs. a conversion
to netdev_alloc_pcpu_stats.
Signed-off-by: David S. Miller <davem@davemloft.net>
Current error handling of virtqueue_kick() was wrong in two places:
- The skb were freed immediately when virtqueue_kick() fail during
xmit. This may lead double free since the skb was not detached from
the virtqueue.
- try_fill_recv() returns false when virtqueue_kick() fail. This will
lead unnecessary rescheduling of refill work.
Actually, it's safe to just ignore the kick failure in those two
places. So this patch fixes this by partially revert commit
6797590118.
Fixes 6797590118
(virtio_net: verify if virtqueue_kick() succeeded).
Cc: Heinz Graalfs <graalfs@linux.vnet.ibm.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Replace dev_kfree_skb with dev_kfree_skb_any in start_xmit which can
be called in hard irq and other contexts.
start_xmit only frees skbs that it is dropping.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Replace the bh safe variant with the hard irq safe variant.
We need a hard irq safe variant to deal with netpoll transmitting
packets from hard irq context, and we need it in most if not all of
the places using the bh safe variant.
Except on 32bit uni-processor the code is exactly the same so don't
bother with a bh variant, just have a hard irq safe variant that
everyone can use.
Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
A bad implementation of virtio might cause us to mark the virtqueue
broken: we'll dev_err() in that case, and the device is useless, but
let's not BUG_ON().
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We should alloc big buffers also when guest can receive UFO
packets to let the big packets fit into guest rx buffer.
Fixes 5c5167515d
(virtio-net: Allow UFO feature to be set and advertised.)
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: Sridhar Samudrala <sri@us.ibm.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Rusty Russell <rusty@rustcorp.com.au>
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>
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>
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>
Conflicts:
drivers/net/ethernet/qlogic/qlcnic/qlcnic_sriov_pf.c
net/ipv6/ip6_tunnel.c
net/ipv6/ip6_vti.c
ipv6 tunnel statistic bug fixes conflicting with consolidation into
generic sw per-cpu net stats.
qlogic conflict between queue counting bug fix and the addition
of multiple MAC address support.
Signed-off-by: David S. Miller <davem@davemloft.net>
During restoring, try_fill_recv() was called with neither napi lock nor napi
disabled. This can lead two try_fill_recv() was called in the same time. Fix
this by refilling before trying to enable napi.
Fixes 0741bcb558
(virtio: net: Add freeze, restore handlers to support S4).
Cc: Amit Shah <amit.shah@redhat.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
All the code passes NULL for the last sg list (in).
Simplify by just removing it.
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Merge 'net' into 'net-next' to get the AF_PACKET bug fix that
Daniel's direct transmit changes depend upon.
Signed-off-by: David S. Miller <davem@davemloft.net>
When a packet with invalid length arrives, ensure that the packet
is freed correctly if mergeable packet buffers and big packets
(GUEST_TSO4) are both enabled.
Signed-off-by: Michael Dalton <mwdalton@google.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Andrew Vagin <avagin@openvz.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Several files refer to an old address for the Free Software Foundation
in the file header comment. Resolve by replacing the address with
the URL <http://www.gnu.org/licenses/> so that we do not have to keep
updating the header comments anytime the address changes.
CC: Jay Vosburgh <fubar@us.ibm.com>
CC: Veaceslav Falico <vfalico@redhat.com>
CC: Andy Gospodarek <andy@greyhouse.net>
CC: Haiyang Zhang <haiyangz@microsoft.com>
CC: "K. Y. Srinivasan" <kys@microsoft.com>
CC: Paul Mackerras <paulus@samba.org>
CC: Ian Campbell <ian.campbell@citrix.com>
CC: Wei Liu <wei.liu2@citrix.com>
CC: Rusty Russell <rusty@rustcorp.com.au>
CC: "Michael S. Tsirkin" <mst@redhat.com>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Acked-by: Wei Liu <wei.liu2@citrix.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
receive mergeable now handles errors internally.
Do same for big and small packet paths, otherwise
the logic is too hard to follow.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Dumazet noticed that if we encounter an error
when processing a mergeable buffer, we don't
dequeue all of the buffers from this packet,
the result is almost sure to be loss of networking.
Jason Wang noticed that we also leak a page and that we don't decrement
the rq buf count, so we won't repost buffers (a resource leak).
Fix both issues.
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Michael Dalton <mwdalton@google.com>
Reported-by: Eric Dumazet <edumazet@google.com>
Reported-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
"MAC filter" sounds more reasonable than "MAC fitler".
Signed-off-by: Thomas Huth <thuth@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pull networking fixes from David Miller:
"Mostly these are fixes for fallout due to merge window changes, as
well as cures for problems that have been with us for a much longer
period of time"
1) Johannes Berg noticed two major deficiencies in our genetlink
registration. Some genetlink protocols we passing in constant
counts for their ops array rather than something like
ARRAY_SIZE(ops) or similar. Also, some genetlink protocols were
using fixed IDs for their multicast groups.
We have to retain these fixed IDs to keep existing userland tools
working, but reserve them so that other multicast groups used by
other protocols can not possibly conflict.
In dealing with these two problems, we actually now use less state
management for genetlink operations and multicast groups.
2) When configuring interface hardware timestamping, fix several
drivers that simply do not validate that the hwtstamp_config value
is one the driver actually supports. From Ben Hutchings.
3) Invalid memory references in mwifiex driver, from Amitkumar Karwar.
4) In dev_forward_skb(), set the skb->protocol in the right order
relative to skb_scrub_packet(). From Alexei Starovoitov.
5) Bridge erroneously fails to use the proper wrapper functions to make
calls to netdev_ops->ndo_vlan_rx_{add,kill}_vid. Fix from Toshiaki
Makita.
6) When detaching a bridge port, make sure to flush all VLAN IDs to
prevent them from leaking, also from Toshiaki Makita.
7) Put in a compromise for TCP Small Queues so that deep queued devices
that delay TX reclaim non-trivially don't have such a performance
decrease. One particularly problematic area is 802.11 AMPDU in
wireless. From Eric Dumazet.
8) Fix crashes in tcp_fastopen_cache_get(), we can see NULL socket dsts
here. Fix from Eric Dumzaet, reported by Dave Jones.
9) Fix use after free in ipv6 SIT driver, from Willem de Bruijn.
10) When computing mergeable buffer sizes, virtio-net fails to take the
virtio-net header into account. From Michael Dalton.
11) Fix seqlock deadlock in ip4_datagram_connect() wrt. statistic
bumping, this one has been with us for a while. From Eric Dumazet.
12) Fix NULL deref in the new TIPC fragmentation handling, from Erik
Hugne.
13) 6lowpan bit used for traffic classification was wrong, from Jukka
Rissanen.
14) macvlan has the same issue as normal vlans did wrt. propagating LRO
disabling down to the real device, fix it the same way. From Michal
Kubecek.
15) CPSW driver needs to soft reset all slaves during suspend, from
Daniel Mack.
16) Fix small frame pacing in FQ packet scheduler, from Eric Dumazet.
17) The xen-netfront RX buffer refill timer isn't properly scheduled on
partial RX allocation success, from Ma JieYue.
18) When ipv6 ping protocol support was added, the AF_INET6 protocol
initialization cleanup path on failure was borked a little. Fix
from Vlad Yasevich.
19) If a socket disconnects during a read/recvmsg/recvfrom/etc that
blocks we can do the wrong thing with the msg_name we write back to
userspace. From Hannes Frederic Sowa. There is another fix in the
works from Hannes which will prevent future problems of this nature.
20) Fix route leak in VTI tunnel transmit, from Fan Du.
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: (106 commits)
genetlink: make multicast groups const, prevent abuse
genetlink: pass family to functions using groups
genetlink: add and use genl_set_err()
genetlink: remove family pointer from genl_multicast_group
genetlink: remove genl_unregister_mc_group()
hsr: don't call genl_unregister_mc_group()
quota/genetlink: use proper genetlink multicast APIs
drop_monitor/genetlink: use proper genetlink multicast APIs
genetlink: only pass array to genl_register_family_with_ops()
tcp: don't update snd_nxt, when a socket is switched from repair mode
atm: idt77252: fix dev refcnt leak
xfrm: Release dst if this dst is improper for vti tunnel
netlink: fix documentation typo in netlink_set_err()
be2net: Delete secondary unicast MAC addresses during be_close
be2net: Fix unconditional enabling of Rx interface options
net, virtio_net: replace the magic value
ping: prevent NULL pointer dereference on write to msg_name
bnx2x: Prevent "timeout waiting for state X"
bnx2x: prevent CFC attention
bnx2x: Prevent panic during DMAE timeout
...
It is more appropriate to use # of queue pairs currently used by
the driver instead of a magic value.
Signed-off-by: Zhi Yong Wu <wuzhy@linux.vnet.ibm.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
some robustness fixes for broken virtio devices, plus minor tweaks.
[vs last pull request: added the virtio-scsi broken vq escape patch, which
I somehow lost.]
Cheers,
Rusty.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.14 (GNU/Linux)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=vY2L
-----END PGP SIGNATURE-----
Merge tag 'virtio-next-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux
Pull virtio updates from Rusty Russell:
"Nothing really exciting: some groundwork for changing virtio endian,
and some robustness fixes for broken virtio devices, plus minor
tweaks"
* tag 'virtio-next-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux:
virtio_scsi: verify if queue is broken after virtqueue_get_buf()
x86, asmlinkage, lguest: Pass in globals into assembler statement
virtio: mmio: fix signature checking for BE guests
virtio_ring: adapt to notify() returning bool
virtio_net: verify if queue is broken after virtqueue_get_buf()
virtio_console: verify if queue is broken after virtqueue_get_buf()
virtio_blk: verify if queue is broken after virtqueue_get_buf()
virtio_ring: add new function virtqueue_is_broken()
virtio_test: verify if virtqueue_kick() succeeded
virtio_net: verify if virtqueue_kick() succeeded
virtio_ring: let virtqueue_{kick()/notify()} return a bool
virtio_ring: change host notification API
virtio_config: remove virtio_config_val
virtio: use size-based config accessors.
virtio_config: introduce size-based accessors.
virtio_ring: plug kmemleak false positive.
virtio: pm: use CONFIG_PM_SLEEP instead of CONFIG_PM
Commit 2613af0ed1 ("virtio_net: migrate mergeable rx buffers to page
frag allocators") changed the mergeable receive buffer size from PAGE_SIZE
to MTU-size. However, the merge buffer size does not take into account the
size of the virtio-net header. Consequently, packets that are MTU-size
will take two buffers intead of one (to store the virtio-net header),
substantially decreasing the throughput of MTU-size traffic due to TCP
window / SKB truesize effects.
This commit changes the mergeable buffer size to include the virtio-net
header. The buffer size is cacheline-aligned because skb_page_frag_refill
will not automatically align the requested 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 and
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. Transmit offloads and mergeable receive
buffers are enabled, but guest_tso4 / guest_csum are explicitly disabled to
force MTU-sized packets on the receiver.
next-net trunk before 2613af0ed1 (PAGE_SIZE buf): 3861.08Gb/s
net-next trunk (MTU 1500- packet uses two buf due to size bug): 4076.62Gb/s
net-next trunk (MTU 1480- packet fits in one buf): 6301.34Gb/s
net-next trunk w/ size fix (MTU 1500 - packet fits in one buf): 6445.44Gb/s
Suggested-by: Eric Northup <digitaleric@google.com>
Signed-off-by: Michael Dalton <mwdalton@google.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pull core locking changes from Ingo Molnar:
"The biggest changes:
- add lockdep support for seqcount/seqlocks structures, this
unearthed both bugs and required extra annotation.
- move the various kernel locking primitives to the new
kernel/locking/ directory"
* 'core-locking-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (21 commits)
block: Use u64_stats_init() to initialize seqcounts
locking/lockdep: Mark __lockdep_count_forward_deps() as static
lockdep/proc: Fix lock-time avg computation
locking/doc: Update references to kernel/mutex.c
ipv6: Fix possible ipv6 seqlock deadlock
cpuset: Fix potential deadlock w/ set_mems_allowed
seqcount: Add lockdep functionality to seqcount/seqlock structures
net: Explicitly initialize u64_stats_sync structures for lockdep
locking: Move the percpu-rwsem code to kernel/locking/
locking: Move the lglocks code to kernel/locking/
locking: Move the rwsem code to kernel/locking/
locking: Move the rtmutex code to kernel/locking/
locking: Move the semaphore core to kernel/locking/
locking: Move the spinlock code to kernel/locking/
locking: Move the lockdep code to kernel/locking/
locking: Move the mutex code to kernel/locking/
hung_task debugging: Add tracepoint to report the hang
x86/locking/kconfig: Update paravirt spinlock Kconfig description
lockstat: Report avg wait and hold times
lockdep, x86/alternatives: Drop ancient lockdep fixup message
...
In order to enable lockdep on seqcount/seqlock structures, we
must explicitly initialize any locks.
The u64_stats_sync structure, uses a seqcount, and thus we need
to introduce a u64_stats_init() function and use it to initialize
the structure.
This unfortunately adds a lot of fairly trivial initialization code
to a number of drivers. But the benefit of ensuring correctness makes
this worth while.
Because these changes are required for lockdep to be enabled, and the
changes are quite trivial, I've not yet split this patch out into 30-some
separate patches, as I figured it would be better to get the various
maintainers thoughts on how to best merge this change along with
the seqcount lockdep enablement.
Feedback would be appreciated!
Signed-off-by: John Stultz <john.stultz@linaro.org>
Acked-by: Julian Anastasov <ja@ssi.bg>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org>
Cc: James Morris <jmorris@namei.org>
Cc: Jesse Gross <jesse@nicira.com>
Cc: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Mirko Lindner <mlindner@marvell.com>
Cc: Patrick McHardy <kaber@trash.net>
Cc: Roger Luethi <rl@hellgate.ch>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Simon Horman <horms@verge.net.au>
Cc: Stephen Hemminger <stephen@networkplumber.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Cc: Wensong Zhang <wensong@linux-vs.org>
Cc: netdev@vger.kernel.org
Link: http://lkml.kernel.org/r/1381186321-4906-2-git-send-email-john.stultz@linaro.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
We used to use a percpu structure vq_index to record the cpu to queue
mapping, this is suboptimal since it duplicates the work of XPS and
loses all other XPS functionality such as allowing user to configure
their own transmission steering strategy.
So this patch switches to use XPS and suggest a default mapping when
the number of cpus is equal to the number of queues. With XPS support,
there's no need for keeping per-cpu vq_index and .ndo_select_queue(),
so they were removed also.
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Rusty Russell <rusty@rustcorp.com.au>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit 2613af0ed1 (virtio_net: migrate mergeable
rx buffers to page frag allocators) try to increase the payload/truesize for
MTU-sized traffic. But this will introduce the extra overhead for GSO packets
received because of the frag list. This commit tries to reduce this issue by
coalesce the possible rx frags when possible during rx. Test result shows the
about 15% improvement on full size GSO packet receiving (and even better than
before commit 2613af0ed1).
Before this commit:
./netperf -H 192.168.100.4
MIGRATED TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.100.4
() port 0 AF_INET : demo
Recv Send Send
Socket Socket Message Elapsed
Size Size Size Time Throughput
bytes bytes bytes secs. 10^6bits/sec
87380 16384 16384 10.00 20303.87
After this commit:
./netperf -H 192.168.100.4
MIGRATED TCP STREAM TEST from 0.0.0.0 (0.0.0.0) port 0 AF_INET to 192.168.100.4
() port 0 AF_INET : demo
Recv Send Send
Socket Socket Message Elapsed
Size Size Size Time Throughput
bytes bytes bytes secs. 10^6bits/sec
87380 16384 16384 10.00 23841.26
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: Michael Dalton <mwdalton@google.com>
Cc: Eric Dumazet <edumazet@google.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Acked-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Conflicts:
drivers/net/ethernet/emulex/benet/be.h
drivers/net/netconsole.c
net/bridge/br_private.h
Three mostly trivial conflicts.
The net/bridge/br_private.h conflict was a function signature (argument
addition) change overlapping with the extern removals from Joe Perches.
In drivers/net/netconsole.c we had one change adjusting a printk message
whilst another changed "printk(KERN_INFO" into "pr_info(".
Lastly, the emulex change was a new inline function addition overlapping
with Joe Perches's extern removals.
Signed-off-by: David S. Miller <davem@davemloft.net>
The virtio_net driver's mergeable receive buffer allocator
uses 4KB packet buffers. For MTU-sized traffic, SKB truesize
is > 4KB but only ~1500 bytes of the buffer is used to store
packet data, reducing the effective TCP window size
substantially. This patch addresses the performance concerns
with mergeable receive buffers by allocating MTU-sized packet
buffers using page frag allocators. If more than MAX_SKB_FRAGS
buffers are needed, the SKB frag_list is used.
Signed-off-by: Michael Dalton <mwdalton@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If a virtqueue_get_buf() call returns a NULL pointer a possibly endless while
loop should be avoided by checking for a broken virtqueue.
Signed-off-by: Heinz Graalfs <graalfs@linux.vnet.ibm.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Verify if a host kick succeeded by checking return value of virtqueue_kick().
Signed-off-by: Heinz Graalfs <graalfs@linux.vnet.ibm.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We used to schedule the refill work unconditionally after changing the
number of queues. This may lead an issue if the device is not
up. Since we only try to cancel the work in ndo_stop(), this may cause
the refill work still work after removing the device. Fix this by only
schedule the work when device is up.
The bug were introduce by commit 9b9cd8024a.
(virtio-net: fix the race between channels setting and refill)
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We're trying to re-configure the affinity unconditionally in cpu hotplug
callback. This may lead the issue during resuming from s3/s4 since
- virt queues haven't been allocated at that time.
- it's unnecessary since thaw method will re-configure the affinity.
Fix this issue by checking the config_enable and do nothing is we're not ready.
The bug were introduced by commit 8de4b2f3ae
(virtio-net: reset virtqueue affinity when doing cpu hotplug).
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: Wanlong Gao <gaowanlong@cn.fujitsu.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This lets the transport do endian conversion if necessary, and insulates
the drivers from the difference.
Most drivers can use the simple helpers virtio_cread() and virtio_cwrite().
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
The freeze and restore functions defined in virtio drivers are used
for suspend and hibernate, so CONFIG_PM_SLEEP is more appropriate than
CONFIG_PM. This patch replace all CONFIG_PM with CONFIG_PM_SLEEP for
virtio drivers that implement freeze and restore callbacks.
Signed-off-by: Aaron Lu <aaron.lu@intel.com>
Reviewed-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
If the VIRTIO_NET_F_GUEST_CSUM virtio feature is available, the guest
does not have to calculate the checksums on all received packets. This
is pretty much the same feature as RX checksum offloading on real
network cards, so the virtio-net driver should report this by setting
the NETIF_F_RXCSUM flag. When the user now runs "ethtool -k", he or she
can see whether the virtio-net interface has to calculate RX checksums
or not.
Signed-off-by: Thomas Huth <thuth@linux.vnet.ibm.com>
Acked-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
For small packets we can simplify xmit processing
by linearizing buffers with the header:
most packets seem to have enough head room
we can use for this purpose.
Since existing hypervisors require that header
is the first s/g element, we need a feature bit
for this.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
virtio net called virtqueue_enable_cq on RX path after napi_complete, so
with NAPI_STATE_SCHED clear - outside the implicit napi lock.
This violates the requirement to synchronize virtqueue_enable_cq wrt
virtqueue_add_buf. In particular, used event can move backwards,
causing us to lose interrupts.
In a debug build, this can trigger panic within START_USE.
Jason Wang reports that he can trigger the races artificially,
by adding udelay() in virtqueue_enable_cb() after virtio_mb().
However, we must call napi_complete to clear NAPI_STATE_SCHED before
polling the virtqueue for used buffers, otherwise napi_schedule_prep in
a callback will fail, causing us to lose RX events.
To fix, call virtqueue_enable_cb_prepare with NAPI_STATE_SCHED
set (under napi lock), later call virtqueue_poll with
NAPI_STATE_SCHED clear (outside the lock).
Reported-by: Jason Wang <jasowang@redhat.com>
Tested-by: Jason Wang <jasowang@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit 55257d72bd (virtio-net: fill only rx queues
which are being used) tries to refill on demand when changing the number of
channels by call try_refill_recv() directly, this may race:
- the refill work who may do the refill in the same time
- the try_refill_recv() called in bh since napi was not disabled
Which may led guest complain during setting channels:
virtio_net virtio0: input.1:id 0 is not a head!
Solve this issue by scheduling a refill work which can guarantee the
serialization of refill.
Cc: Sasha Levin <sasha.levin@oracle.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Commit 55257d72bd (virtio-net: fill only rx
queues which are being used) only does the napi enabling during open for
curr_queue_pairs. This will break multiqueue receiving since napi of new queues
were still disabled after changing the number of queues.
This patch fixes this by enabling napi for all possible queues during open.
Cc: Sasha Levin <sasha.levin@oracle.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>
Since commit 82dc3c63c6 ("net: introduce NAPI_POLL_WEIGHT")
we warn drivers when they use napi weight higher than NAPI_POLL_WEIGHT,
but virtio_net still uses 128 by default. This patch makes its default
value to NAPI_POLL_WEIGHT.
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Eric Dumazet <eric.dumazet@gmail.com>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: Cong Wang <amwang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
I dived into lguest again, reworking the pagetable code so we can move
the switcher page: our fixmaps sometimes take more than 2MB now...
Cheers,
Rusty.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.12 (GNU/Linux)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=1tYL
-----END PGP SIGNATURE-----
Merge tag 'virtio-next-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux
Pull virtio & lguest updates from Rusty Russell:
"Lots of virtio work which wasn't quite ready for last merge window.
Plus I dived into lguest again, reworking the pagetable code so we can
move the switcher page: our fixmaps sometimes take more than 2MB now..."
Ugh. Annoying conflicts with the tcm_vhost -> vhost_scsi rename.
Hopefully correctly resolved.
* tag 'virtio-next-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux: (57 commits)
caif_virtio: Remove bouncing email addresses
lguest: improve code readability in lg_cpu_start.
virtio-net: fill only rx queues which are being used
lguest: map Switcher below fixmap.
lguest: cache last cpu we ran on.
lguest: map Switcher text whenever we allocate a new pagetable.
lguest: don't share Switcher PTE pages between guests.
lguest: expost switcher_pages array (as lg_switcher_pages).
lguest: extract shadow PTE walking / allocating.
lguest: make check_gpte et. al return bool.
lguest: assume Switcher text is a single page.
lguest: rename switcher_page to switcher_pages.
lguest: remove RESERVE_MEM constant.
lguest: check vaddr not pgd for Switcher protection.
lguest: prepare to make SWITCHER_ADDR a variable.
virtio: console: replace EMFILE with EBUSY for already-open port
virtio-scsi: reset virtqueue affinity when doing cpu hotplug
virtio-scsi: introduce multiqueue support
virtio-scsi: push vq lock/unlock into virtscsi_vq_done
virtio-scsi: pass struct virtio_scsi to virtqueue completion function
...
Due to MQ support we may allocate a whole bunch of rx queues but
never use them. With this patch we'll safe the space used by
the receive buffers until they are actually in use:
sh-4.2# free -h
total used free shared buffers cached
Mem: 490M 35M 455M 0B 0B 4.1M
-/+ buffers/cache: 31M 459M
Swap: 0B 0B 0B
sh-4.2# ethtool -L eth0 combined 8
sh-4.2# free -h
total used free shared buffers cached
Mem: 490M 162M 327M 0B 0B 4.1M
-/+ buffers/cache: 158M 331M
Swap: 0B 0B 0B
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Linus Torvalds