[Split from "correct capacity math on ring full" -- Rusty]
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Capacity math on ring full is wrong: we are
looking at num_sg but that might be optimistic
because of indirect buffer use.
The implementation also penalizes fast path
with extra memory accesses for the benefit of
ring full condition handling which is slow path.
It's easy to query ring capacity so let's do just that.
This change also makes it easier to move vnet header
for tx around as follow-up patch does.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Obviously it should check !vi->rq.
Reported-by: Fengguang Wu <fengguang.wu@intel.com>
Cc: Jason Wang <jasowang@redhat.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>
This patch implements the ethtool_{set|get}_channels method of virtio-net to
allow user to change the number of queues when the device is running on demand.
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds the multiqueue (VIRTIO_NET_F_MQ) support to virtio_net
driver. VIRTIO_NET_F_MQ capable device could allow the driver to do packet
transmission and reception through multiple queue pairs and does the packet
steering to get better performance. By default, one one queue pair is used, user
could change the number of queue pairs by ethtool in the next patch.
When multiple queue pairs is used and the number of queue pairs is equal to the
number of vcpus. Driver does the following optimizations to implement per-cpu
virt queue pairs:
- select the txq based on the smp processor id.
- smp affinity hint to the cpu that owns the queue pairs.
This could be used with the flow steering support of the device to guarantee the
packets of a single flow is handled by the same cpu.
Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
To support multiqueue transmitq/receiveq, the first step is to separate queue
related structure from virtnet_info. This patch introduce send_queue and
receive_queue structure and use the pointer to them as the parameter in
functions handling sending/receiving.
Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
CONFIG_HOTPLUG is going away as an option. As result the __dev*
markings will be going away.
Remove use of __devinit, __devexit_p, __devinitdata, __devinitconst,
and __devexit.
Signed-off-by: Bill Pemberton <wfp5p@virginia.edu>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: virtualization@lists.linux-foundation.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
These can be converted to net_*_ratelimited().
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: David Miller <davem@davemloft.net>
Signed-off-by: Cong Wang <amwang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Pull networking changes from David Miller:
1) GRE now works over ipv6, from Dmitry Kozlov.
2) Make SCTP more network namespace aware, from Eric Biederman.
3) TEAM driver now works with non-ethernet devices, from Jiri Pirko.
4) Make openvswitch network namespace aware, from Pravin B Shelar.
5) IPV6 NAT implementation, from Patrick McHardy.
6) Server side support for TCP Fast Open, from Jerry Chu and others.
7) Packet BPF filter supports MOD and XOR, from Eric Dumazet and Daniel
Borkmann.
8) Increate the loopback default MTU to 64K, from Eric Dumazet.
9) Use a per-task rather than per-socket page fragment allocator for
outgoing networking traffic. This benefits processes that have very
many mostly idle sockets, which is quite common.
From Eric Dumazet.
10) Use up to 32K for page fragment allocations, with fallbacks to
smaller sizes when higher order page allocations fail. Benefits are
a) less segments for driver to process b) less calls to page
allocator c) less waste of space.
From Eric Dumazet.
11) Allow GRO to be used on GRE tunnels, from Eric Dumazet.
12) VXLAN device driver, one way to handle VLAN issues such as the
limitation of 4096 VLAN IDs yet still have some level of isolation.
From Stephen Hemminger.
13) As usual there is a large boatload of driver changes, with the scale
perhaps tilted towards the wireless side this time around.
Fix up various fairly trivial conflicts, mostly caused by the user
namespace changes.
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1012 commits)
hyperv: Add buffer for extended info after the RNDIS response message.
hyperv: Report actual status in receive completion packet
hyperv: Remove extra allocated space for recv_pkt_list elements
hyperv: Fix page buffer handling in rndis_filter_send_request()
hyperv: Fix the missing return value in rndis_filter_set_packet_filter()
hyperv: Fix the max_xfer_size in RNDIS initialization
vxlan: put UDP socket in correct namespace
vxlan: Depend on CONFIG_INET
sfc: Fix the reported priorities of different filter types
sfc: Remove EFX_FILTER_FLAG_RX_OVERRIDE_IP
sfc: Fix loopback self-test with separate_tx_channels=1
sfc: Fix MCDI structure field lookup
sfc: Add parentheses around use of bitfield macro arguments
sfc: Fix null function pointer in efx_sriov_channel_type
vxlan: virtual extensible lan
igmp: export symbol ip_mc_leave_group
netlink: add attributes to fdb interface
tg3: unconditionally select HWMON support when tg3 is enabled.
Revert "net: ti cpsw ethernet: allow reading phy interface mode from DT"
gre: fix sparse warning
...
system_nrt[_freezable]_wq are now spurious. Mark them deprecated and
convert all users to system[_freezable]_wq.
If you're cc'd and wondering what's going on: Now all workqueues are
non-reentrant, so there's no reason to use system_nrt[_freezable]_wq.
Please use system[_freezable]_wq instead.
This patch doesn't make any functional difference.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-By: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: David Airlie <airlied@linux.ie>
Cc: Jiri Kosina <jkosina@suse.cz>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Cc: David Howells <dhowells@redhat.com>
I believe net/core/dev.c is a better place for netif_notify_peers(),
because other net event notify functions also stay in this file.
And rename it to netdev_notify_peers().
Cc: David S. Miller <davem@davemloft.net>
Cc: Ian Campbell <Ian.Campbell@citrix.com>
Signed-off-by: Cong Wang <amwang@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Fix race condition in several network drivers when reading stats on 32bit
UP architectures. These drivers update their stats in a BH context and
therefore should use u64_stats_fetch_begin_bh/u64_stats_fetch_retry_bh
instead of u64_stats_fetch_begin/u64_stats_fetch_retry when reading the
stats.
Signed-off-by: Kevin Groeneveld <kgroeneveld@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
commit 3fa2a1df90 (virtio-net: per cpu 64 bit stats (v2)) added a race
on 32bit arches.
We must use separate syncp for rx and tx path as they can be run at the
same time on different cpus. Thus one sequence increment can be lost and
readers spin forever.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Stephen Hemminger <shemminger@vyatta.com>
Cc: Michael S. Tsirkin <mst@redhat.com>
Cc: 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>
disable_cb is just an optimization: it
can not guarantee that there are no callbacks.
In particular it doesn't have any effect when
event index is on.
Instead, detach, napi disable and reset on freeze ensure we don't run
concurrently with a callback.
Remove the useless calls so we get same behaviour
with and without event index.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
__napi_schedule might raise softirq but nothing
causes do_softirq to trigger, so it does not in fact
run. As a result,
the error message "NOHZ: local_softirq_pending 08"
sometimes occurs during boot of a KVM guest when the network service is
started and we are oom:
...
Bringing up loopback interface: [ OK ]
Bringing up interface eth0:
Determining IP information for eth0...NOHZ: local_softirq_pending 08
done.
[ OK ]
...
Further, receive queue processing might get delayed
indefinitely until some interrupt triggers:
virtio_net expected napi to be run immediately.
One way to cause do_softirq to be executed is by
invoking local_bh_enable(). As __napi_schedule is
normally called from bh or irq context, this
seems to make sense: disable bh before __napi_schedule
and enable afterwards.
In fact it's a very complicated way of calling do_softirq(),
and works since this function is only used when we are not
in interrupt context. It's not hot at all, in any ideal scenario.
Reported-by: Ulrich Obergfell <uobergfe@redhat.com>
Tested-by: Ulrich Obergfell <uobergfe@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Rusty Russell <rusty@rustcorp.com.au>
As hypervior does not have the knowledge of guest network configuration, it's
better to ask guest to send gratuitous packets when needed.
This patch implements VIRTIO_NET_F_GUEST_ANNOUNCE feature: hypervisor would
notice the guest when it thinks it's time for guest to announce the link
presnece. Guest tests VIRTIO_NET_S_ANNOUNCE bit during config change interrupt
and woule send gratuitous packets through netif_notify_peers() and ack the
notification through ctrl vq.
We need to make sure the atomicy of read and ack in guest otherwise we may ack
more times than being notified. This is done through handling the whole config
change interrupt in an non-reentrant workqueue.
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>
Commit 2e57b79cce misplaced its
parenthesis and now tx_fifo_errors will only be incremented if an
ENOMEM error is not written to the syslog.
Correct the parenthesis and indentation to the original goal of
counting all non ENOMEM errors and ratelimiting only the messages.
Signed-of-by: Torsten Kaiser <just.for.lkml@googlemail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
While it is desirable to rate limit certain messages, it is not
desirable to rate limit the incrementing of counters associated
with those messages.
Signed-off-by: Rick Jones <rick.jones2@hp.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>
Replace usage of random_ether_addr() with eth_hw_addr_random()
to set addr_assign_type correctly to NET_ADDR_RANDOM.
Change the trivial cases.
v2: adapt to renamed eth_hw_addr_random()
Signed-off-by: Danny Kukawka <danny.kukawka@bisect.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
commit 3fa2a1df90 (virtio-net: per cpu 64 bit stats (v2)) added extra
__percpu qualifiers and sparse errors.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Cc: Stephen Hemminger <shemminger@vyatta.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Remove all the vqs, disable napi and detach from the netdev on
hibernation.
Re-create vqs after restoring from a hibernated image, re-enable napi
and re-attach the netdev. This keeps networking working across
hibernation.
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
The remove and PM freeze functions will share this code.
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
The probe and PM restore functions will share this code.
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Remove wrapper functions. This makes the allocation type explicit in
all callers; I used GPF_KERNEL where it seemed obvious, left it at
GFP_ATOMIC otherwise.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Currently, the refill path for RX buffers will always allocate the
buffers as GFP_ATOMIC, even if we are in process context. This will
fail to apply memory pressure as the worker thread will not contribute
to the freeing of memory.
Fix this by changing add_recvbuf_small to use the gfp variant allocator,
__netdev_alloc_skb_ip_align().
Signed-off-by: Mike Waychison <mikew@google.com>
Acked-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Michael S. Tsirkin also noticed that we could run the refill work
multiple CPUs: if we kick off a refill on one CPU and then on another,
they would both manipulate the queue at the same time (they use
napi_disable to avoid racing against the receive handler itself).
Tejun points out that this is what the WQ_NON_REENTRANT flag is for,
and that there is a convenient system kthread we can use.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Michael S. Tsirkin noticed that we could run the refill work after
ndo_close, which can re-enable napi - we don't disable it until
virtnet_remove. This is clearly wrong, so move the workqueue control
to ndo_open and ndo_stop (aka. virtnet_open and virtnet_close).
One subtle point: virtnet_probe() could simply fail if it couldn't
allocate a receive buffer, but that's less polite in virtnet_open() so
we schedule a refill as we do in the normal receive path if we run out
of memory.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
module_param(bool) used to counter-intuitively take an int. In
fddd5201 (mid-2009) we allowed bool or int/unsigned int using a messy
trick.
It's time to remove the int/unsigned int option. For this version
it'll simply give a warning, but it'll break next kernel version.
(Thanks to Joe Perches for suggesting coccinelle for 0/1 -> true/false).
Cc: "David S. Miller" <davem@davemloft.net>
Cc: netdev@vger.kernel.org
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Let caller know the result of adding/removing vlan id to/from vlan
filter.
In some drivers I make those functions to just return 0. But in those
where there is able to see if hw setup went correctly, return value is
set appropriately.
Signed-off-by: Jiri Pirko <jpirko@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Tx_fifo_errors are tracked in start_xmit_ for virtio_net, but not
reported in the tallies returned by virtnet_stats(). Return them
as the rx "sub-stats" rx_length_errors and rx_frame_errors are.
Signed-off-by: Rick Jones <rick.jones2@hp.com>
Acked-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add a new .bus_name to virtio_config_ops then modify virtio_net to
call through to it in an ethtool .get_drvinfo routine to report
bus_info in ethtool -i output which is consistent with other
emulated NICs and the output of lspci.
Signed-off-by: Rick Jones <rick.jones2@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch modifies virtio-net to use virtio_config_val() instead
of a 'if(virtio_has_feature()) vdev->config->get()' construct to retrieve
optional values from the config space.
Cc: Amit Shah <amit.shah@redhat.com>
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: virtualization@lists.linux-foundation.org
Signed-off-by: Sasha Levin <levinsasha928@gmail.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Add support for reporting ring sizes via ethtool -g to the virtio_net
driver.
Signed-off-by: Rick Jones <rick.jones2@hp.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>
We must account in skb->truesize, the size of the fragments, not the
used part of them.
Doing this work is important to avoid unexpected OOM situations.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
CC: Rusty Russell <rusty@rustcorp.com.au>
CC: "Michael S. Tsirkin" <mst@redhat.com>
CC: virtualization@lists.linux-foundation.org
CC: Krishna Kumar <krkumar2@in.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Remove manual initialization in set_skb_frag, and instead
use __skb_fill_page_desc() to do the same. Patch tested
on net-next.
Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com>
Acked-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
To ease skb->truesize sanitization, its better to be able to localize
all references to skb frags size.
Define accessors : skb_frag_size() to fetch frag size, and
skb_frag_size_{set|add|sub}() to manipulate it.
Signed-off-by: Eric Dumazet <eric.dumazet@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch verifies that the length of a buffer stored in a linked list
of pages is small enough to fit into a skb.
If the size is larger than a max size of a skb, it means that we shouldn't
go ahead building skbs anyway since we won't be able to send the buffer as
the user requested.
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: virtualization@lists.linux-foundation.org
Cc: netdev@vger.kernel.org
Cc: kvm@vger.kernel.org
Signed-off-by: Sasha Levin <levinsasha928@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: "Michael S. Tsirkin" <mst@redhat.com>
Cc: virtualization@lists.linux-foundation.org
Cc: netdev@vger.kernel.org
Signed-off-by: David S. Miller <davem@davemloft.net>
Use IFF_UNICAST_FTL to find out if driver handles unicast address
filtering. In case it does not, promisc mode is entered.
Patch also fixes following drivers:
stmmac, niu: support uc filtering and yet it propagated
ndo_set_multicast_list
bna, benet, pxa168_eth, ks8851, ks8851_mll, ksz884x : has set
ndo_set_rx_mode but do not support uc filtering
Signed-off-by: Jiri Pirko <jpirko@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Fix a panic in virtnet_remove. unregister_netdev has already
freed up the netdev (and virtnet_info) due to dev->destructor
being set, while virtnet_info is still required. Remove
virtnet_free altogether, and move the freeing of the per-cpu
statistics from virtnet_free to virtnet_remove.
Tested patch below.
Signed-off-by: Krishna Kumar <krkumar2@in.ibm.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use per-cpu variables to maintain 64 bit statistics.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@conan.davemloft.net>
There's no need for the guest to validate the checksum if it have been
validated by host nics. So this patch introduces a new flag -
VIRTIO_NET_HDR_F_DATA_VALID which is used to bypass the checksum
examing in guest. The backend (tap/macvtap) may set this flag when
met skbs with CHECKSUM_UNNECESSARY to save cpu utilization.
No feature negotiation is needed as old driver just ignore this flag.
Iperf shows 12%-30% performance improvement for UDP traffic. For TCP,
when gro is on no difference as it produces skb with partial
checksum. But when gro is disabled, 20% or even higher improvement
could be measured by netperf.
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>
Ask for delayed callbacks on TX ring full, to give the
other side more of a chance to make progress.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Under harsh testing conditions, including low memory, the guest would
stop receiving packets. With this patch applied we no longer see any
problems in the driver while performing these tests for extended periods
of time.
Make sure napi is scheduled subsequent to each napi_enable.
Signed-off-by: Bruce Rogers <brogers@novell.com>
Signed-off-by: Olaf Kirch <okir@suse.de>
Cc: stable@kernel.org
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Replace skb->csum_start - skb_headroom(skb) with skb_checksum_start_offset().
Note for usb/smsc95xx: skb->data - skb->head == skb_headroom(skb).
Signed-off-by: Michał Mirosław <mirq-linux@rere.qmqm.pl>
Signed-off-by: David S. Miller <davem@davemloft.net>
For device that supports VIRTIO_NET_F_STATUS, there's no need to
assume the link is up and we need to call nerif_carrier_off() before
querying device status, otherwise we may get wrong operstate after
diver was loaded because the link watch event was not fired as
expected.
For device that does not support VIRITO_NET_F_STATUS, we could not get
its status through virtnet_update_status() and what we can only do is
always assuming the link is up.
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Jason Wang <jasowang@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
The driver name and bus address for a net_device can normally be found
through the driver model now. Instead of requiring drivers to provide
this information redundantly through the ethtool_ops::get_drvinfo
operation, use the driver model to do so if the driver does not define
the operation. Since ETHTOOL_GDRVINFO no longer requires the driver
to implement any operations, do not require net_device::ethtool_ops to
be set either.
Remove implementations of get_drvinfo and ethtool_ops that provide
only this information.
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
I often use "ethtool -i" command to check what driver controls the
ehternet device. But because current virtio_net driver doesn't
support "ethtool -i", it becomes the following:
# ethtool -i eth3
Cannot get driver information: Operation not supported
This patch simply adds the "ethtool -i" support. The following is the
result when using the virtio_net driver with my patch applied to.
# ethtool -i eth3
driver: virtio_net
version: N/A
firmware-version: N/A
bus-info: virtio0
Personally, "-i" is one of the most frequently-used option, and most
network drivers support "ethtool -i", so I think virtio_net also
should do.
Signed-off-by: Taku Izumi <izumi.taku@jp.fujitsu.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> (use ARRAY_SIZE)
Signed-off-by: David S. Miller <davem@davemloft.net>
virtio net will never try to overflow the TX ring, so the only reason
add_buf may fail is out of memory. Thus, we can not stop the
device until some request completes - there's no guarantee anything
at all is outstanding.
Make the error message clearer as well: error here does not
indicate queue full.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> (...and avoid TX_BUSY)
Cc: stable@kernel.org # .34.x (s/virtqueue_/vi->svq->vq_ops->/)
Signed-off-by: David S. Miller <davem@davemloft.net>
We currently fill all of RX ring, then add_buf
returns ENOSPC, which gets mis-detected as an out of
memory condition and causes us to reschedule the work,
and so on forever. Fix this by oom = err == -ENOMEM;
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: stable@kernel.org # .34.x
Signed-off-by: David S. Miller <davem@davemloft.net>
virtio-net bounces buffer allocations off to
a thread if it can't allocate buffers from the atomic
pool. However, if posting buffers still requires atomic
buffers, this is unlikely to succeed.
Fix by passing in the proper gfp_t parameter.
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' of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-for-linus: (27 commits)
drivers/char: Eliminate use after free
virtio: console: Accept console size along with resize control message
virtio: console: Store each console's size in the console structure
virtio: console: Resize console port 0 on config intr only if multiport is off
virtio: console: Add support for nonblocking write()s
virtio: console: Rename wait_is_over() to will_read_block()
virtio: console: Don't always create a port 0 if using multiport
virtio: console: Use a control message to add ports
virtio: console: Move code around for future patches
virtio: console: Remove config work handler
virtio: console: Don't call hvc_remove() on unplugging console ports
virtio: console: Return -EPIPE to hvc_console if we lost the connection
virtio: console: Let host know of port or device add failures
virtio: console: Add a __send_control_msg() that can send messages without a valid port
virtio: Revert "virtio: disable multiport console support."
virtio: add_buf_gfp
trans_virtio: use virtqueue_xxx wrappers
virtio-rng: use virtqueue_xxx wrappers
virtio_ring: remove a level of indirection
virtio_net: use virtqueue_xxx wrappers
...
Fix up conflicts in drivers/net/virtio_net.c due to new virtqueue_xxx
wrappers changes conflicting with some other cleanups.
Switch virtio_net to new virtqueue_xxx wrappers.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Add missing sg_init_table for sg_set_buf in virtio_net which
induced in defer skb patch.
Reported-by: Thomas Müller <thomas@mathtm.de>
Tested-by: Thomas Müller <thomas@mathtm.de>
Signed-off-by: Shirley Ma <xma@us.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Move sg structure off stack and into virtnet_info structure.
This helps remove extra sg_init_table calls as well as reduce
stack usage.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Tested-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Converts the list and the core manipulating with it to be the same as uc_list.
+uses two functions for adding/removing mc address (normal and "global"
variant) instead of a function parameter.
+removes dev_mcast.c completely.
+exposes netdev_hw_addr_list_* macros along with __hw_addr_* functions for
manipulation with lists on a sandbox (used in bonding and 80211 drivers)
Signed-off-by: Jiri Pirko <jpirko@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add missing sg_init_table for sg_set_buf in virtio_net which
induced in defer skb patch.
Reported-by: Thomas Müller <thomas@mathtm.de>
Tested-by: Thomas Müller <thomas@mathtm.de>
Signed-off-by: Shirley Ma <xma@us.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
percpu.h is included by sched.h and module.h and thus ends up being
included when building most .c files. percpu.h includes slab.h which
in turn includes gfp.h making everything defined by the two files
universally available and complicating inclusion dependencies.
percpu.h -> slab.h dependency is about to be removed. Prepare for
this change by updating users of gfp and slab facilities include those
headers directly instead of assuming availability. As this conversion
needs to touch large number of source files, the following script is
used as the basis of conversion.
http://userweb.kernel.org/~tj/misc/slabh-sweep.py
The script does the followings.
* Scan files for gfp and slab usages and update includes such that
only the necessary includes are there. ie. if only gfp is used,
gfp.h, if slab is used, slab.h.
* When the script inserts a new include, it looks at the include
blocks and try to put the new include such that its order conforms
to its surrounding. It's put in the include block which contains
core kernel includes, in the same order that the rest are ordered -
alphabetical, Christmas tree, rev-Xmas-tree or at the end if there
doesn't seem to be any matching order.
* If the script can't find a place to put a new include (mostly
because the file doesn't have fitting include block), it prints out
an error message indicating which .h file needs to be added to the
file.
The conversion was done in the following steps.
1. The initial automatic conversion of all .c files updated slightly
over 4000 files, deleting around 700 includes and adding ~480 gfp.h
and ~3000 slab.h inclusions. The script emitted errors for ~400
files.
2. Each error was manually checked. Some didn't need the inclusion,
some needed manual addition while adding it to implementation .h or
embedding .c file was more appropriate for others. This step added
inclusions to around 150 files.
3. The script was run again and the output was compared to the edits
from #2 to make sure no file was left behind.
4. Several build tests were done and a couple of problems were fixed.
e.g. lib/decompress_*.c used malloc/free() wrappers around slab
APIs requiring slab.h to be added manually.
5. The script was run on all .h files but without automatically
editing them as sprinkling gfp.h and slab.h inclusions around .h
files could easily lead to inclusion dependency hell. Most gfp.h
inclusion directives were ignored as stuff from gfp.h was usually
wildly available and often used in preprocessor macros. Each
slab.h inclusion directive was examined and added manually as
necessary.
6. percpu.h was updated not to include slab.h.
7. Build test were done on the following configurations and failures
were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my
distributed build env didn't work with gcov compiles) and a few
more options had to be turned off depending on archs to make things
build (like ipr on powerpc/64 which failed due to missing writeq).
* x86 and x86_64 UP and SMP allmodconfig and a custom test config.
* powerpc and powerpc64 SMP allmodconfig
* sparc and sparc64 SMP allmodconfig
* ia64 SMP allmodconfig
* s390 SMP allmodconfig
* alpha SMP allmodconfig
* um on x86_64 SMP allmodconfig
8. percpu.h modifications were reverted so that it could be applied as
a separate patch and serve as bisection point.
Given the fact that I had only a couple of failures from tests on step
6, I'm fairly confident about the coverage of this conversion patch.
If there is a breakage, it's likely to be something in one of the arch
headers which should be easily discoverable easily on most builds of
the specific arch.
Signed-off-by: Tejun Heo <tj@kernel.org>
Guess-its-ok-by: Christoph Lameter <cl@linux-foundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Lee Schermerhorn <Lee.Schermerhorn@hp.com>
This is no longer needed. I missed to remove this in
567ec874d1 ("net: convert multiple
drivers to use netdev_for_each_mc_addr, part6")
Signed-off-by: Jiri Pirko <jpirko@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Now we have a virtio detach API (in commit
f9bfbebf34), we don't need to track xmit
skbs in the virio_net driver, which improves transmission performance.
Signed-off-by: Shirley Ma <xma@us.ibm.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 replaces dev->mc_count in all drivers (hopefully I didn't miss
anything). Used spatch and did small tweaks and conding style changes when
it was suitable.
Jirka
Signed-off-by: Jiri Pirko <jpirko@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
virtio_net receives packets from its pre-allocated vring buffers, then it
delivers these packets to upper layer protocols as skb buffs. So it's not
necessary to pre-allocate skb for each mergable buffer, then frees extra
skbs when buffers are merged into a large packet. This patch has deferred
skb allocation in receiving packets for both big packets and mergeable buffers
to reduce skb pre-allocations and skb frees. It frees unused buffers by calling
detach_unused_buf in vring, so recv skb queue is not needed.
Signed-off-by: Shirley Ma <xma@us.ibm.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
I have seen RX stalls on a machine that experienced a suspected
OOM. After the stall, the RX buffer is empty on the guest side
and there are exactly 16 entries available on the host side. As
the number of entries is less than that required by a maximal
skb, the host cannot proceed.
The guest did not have a refill job scheduled.
My diagnosis is that an OOM had occured, with the delayed refill
job scheduled. The job was able to allocate at least one skb, but
not enough to overcome the minimum required by the host to proceed.
As the refill job would only reschedule itself if it failed completely
to allocate any skbs, this would lead to an RX stall.
The following patch removes this stall possibility by always
rescheduling the refill job until the ring is totally refilled.
Testing has shown that the RX stall no longer occurs whereas
previously it would occur within a day.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Acked-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch introduces three macros to work with uc list from net drivers.
Signed-off-by: Jiri Pirko <jpirko@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Only files where David Miller is the primary git-signer.
wireless, wimax, ixgbe, etc are not modified.
Compile tested x86 allyesconfig only
Not all files compiled (not x86 compatible)
Added a few > 80 column lines, which I ignored.
Existing checkpatch complaints ignored.
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6: (34 commits)
net/fsl_pq_mdio: add module license GPL
can: fix WARN_ON dump in net/core/rtnetlink.c:rtmsg_ifinfo()
can: should not use __dev_get_by_index() without locks
hisax: remove bad udelay call to fix build error on ARM
ipip: Fix handling of DF packets when pmtudisc is OFF
qlge: Set PCIe reset type for EEH to fundamental.
qlge: Fix early exit from mbox cmd complete wait.
ixgbe: fix traffic hangs on Tx with ioatdma loaded
ixgbe: Fix checking TFCS register for TXOFF status when DCB is enabled
ixgbe: Fix gso_max_size for 82599 when DCB is enabled
macsonic: fix crash on PowerBook 520
NET: cassini, fix lock imbalance
ems_usb: Fix byte order issues on big endian machines
be2net: Bug fix to send config commands to hardware after netdev_register
be2net: fix to set proper flow control on resume
netfilter: xt_connlimit: fix regression caused by zero family value
rt2x00: Don't queue ieee80211 work after USB removal
Revert "ipw2200: fix oops on missing firmware"
decnet: netdevice refcount leak
netfilter: nf_nat: fix NAT issue in 2.6.30.4+
...
Conflicts:
drivers/net/usb/cdc_ether.c
All CDC ethernet devices of type USB_CLASS_COMM need to use
'&mbm_info'.
Signed-off-by: David S. Miller <davem@davemloft.net>
Commit
3d1285b (move virtnet_remove to .devexit.text)
introduced the first reference to __devexit in struct virtio_driver
virtio_net which upset modpost ("Section mismatch in reference from the
variable virtio_net to the function .devexit.text:virtnet_remove()").
Fix this by renaming virtio_net to virtio_net_driver.
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Reported-by: Michael S. Tsirkin <mst@redhat.com>
Blame-taken-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6: (43 commits)
net: Fix 'Re: PACKET_TX_RING: packet size is too long'
netdev: usb: dm9601.c can drive a device not supported yet, add support for it
qlge: Fix firmware mailbox command timeout.
qlge: Fix EEH handling.
AF_RAW: Augment raw_send_hdrinc to expand skb to fit iphdr->ihl (v2)
bonding: fix a race condition in calls to slave MII ioctls
virtio-net: fix data corruption with OOM
sfc: Set ip_summed correctly for page buffers passed to GRO
cnic: Fix L2CTX_STATUSB_NUM offset in context memory.
MAINTAINERS: rt2x00 list is moderated
airo: Reorder tests, check bounds before element
mac80211: fix for incorrect sequence number on hostapd injected frames
libertas spi: fix sparse errors
mac80211: trivial: fix spelling in mesh_hwmp
cfg80211: sme: deauthenticate on assoc failure
mac80211: keep auth state when assoc fails
mac80211: fix ibss joining
b43: add 'struct b43_wl' missing declaration
b43: Fix Bugzilla #14181 and the bug from the previous 'fix'
rt2x00: Fix crypto in TX frame for rt2800usb
...
virtio net used to unlink skbs from send queues on error,
but ever since 48925e372f
we do not do this. This causes guest data corruption and crashes
with vhost since net core can requeue the skb or free it without
it being taken off the list.
This patch fixes this by queueing the skb after successful
transmit.
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>
* git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-for-linus:
move virtrng_remove to .devexit.text
move virtballoon_remove to .devexit.text
virtio_blk: Revert serial number support
virtio: let header files include virtio_ids.h
virtio_blk: revert QUEUE_FLAG_VIRT addition
Rusty,
commit 3ca4f5ca73
virtio: add virtio IDs file
moved all device IDs into a single file. While the change itself is
a very good one, it can break userspace applications. For example
if a userspace tool wanted to get the ID of virtio_net it used to
include virtio_net.h. This does no longer work, since virtio_net.h
does not include virtio_ids.h.
This patch moves all "#include <linux/virtio_ids.h>" from the C
files into the header files, making the header files compatible with
the old ones.
In addition, this patch exports virtio_ids.h to userspace.
CC: Fernando Luis Vazquez Cao <fernando@oss.ntt.co.jp>
Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
The function virtnet_remove is used only wrapped by __devexit_p so define
it using __devexit.
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Acked-by: Sam Ravnborg <sam@ravnborg.org>
Cc: David S. Miller <davem@davemloft.net>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Alex Williamson <alex.williamson@hp.com>
Cc: Mark McLoughlin <markmc@redhat.com>
Cc: netdev@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: David S. Miller <davem@davemloft.net>
Saves us one cycle of alloc-add-free if the queue was full.
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> (modified)
Now we can tell the theoretical capacity remaining in the output
queue, virtio_net can waste entries by stopping the queue early.
It doesn't work in the case of indirect buffers and kmalloc failure,
but that's rare (we could drop the packet in that case, but other
drivers return TX_BUSY for similar reasons).
For the record, I think this patch reflects poorly on the linux
network API.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: Dinesh Subhraveti <dineshs@us.ibm.com>
We put the virtio_net_hdr into the skb's cb region; turn this into a
union to clean up the code slightly and allow future expansion.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: Mark McLoughlin <markmc@redhat.com>
Cc: Dinesh Subhraveti <dineshs@us.ibm.com>
The virtio_net driver is complicated by the two methods of freeing old
xmit buffers (in addition to freeing old ones at the start of the xmit
path).
The original code used a 1/10 second timer attached to xmit_free(),
reset on every xmit. Before we orphaned skbs on xmit, the
transmitting userspace could block with a full socket until the timer
fired, the skb destructor was called, and they were re-woken.
So we added the VIRTIO_F_NOTIFY_ON_EMPTY feature: supporting devices
send an interrupt (even if normally suppressed) on an empty xmit ring
which makes us schedule xmit_tasklet(). This was a benchmark win.
Unfortunately, VIRTIO_F_NOTIFY_ON_EMPTY makes quite a lot of work: a
host which is faster than the guest will fire the interrupt every xmit
packet (slowing the guest down further). Attempting mitigation in the
host adds overhead of userspace timers (possibly with the additional
pain of signals), and risks increasing latency anyway if you get it
wrong.
In practice, this effect was masked by benchmarks which take advantage
of GSO (with its inherent transmit batching), but it's still there.
Now we orphan xmitted skbs, the pressure is off: remove both paths and
no longer request VIRTIO_F_NOTIFY_ON_EMPTY. Note that the current
QEMU will notify us even if we don't negotiate this feature (legal,
but suboptimal); a patch is outstanding to improve that.
Move the skb_orphan/nf_reset to after we've done the send and notified
the other end, for a slight optimization.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: Mark McLoughlin <markmc@redhat.com>
This effectively reverts 99ffc696d1
"virtio: wean net driver off NETDEV_TX_BUSY".
The complexity of queuing an skb (setting a tasklet to re-xmit) is
questionable, especially once we get rid of the other reason for the
tasklet in the next patch.
If the skb won't fit in the tx queue, just return NETDEV_TX_BUSY.
This is frowned upon, so a followup patch uses a more complex solution.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: Herbert Xu <herbert@gondor.apana.org.au>
The complex transmit free logic was introduced to avoid hangs on
removing the ip_conntrack module and also because drivers aren't
generally supposed to keep stale skbs for unbounded times.
After some debate, it was decided that while doing skb_orphan()
generally is a rat's nest, we can do it in this driver. Following
patches take advantage of this.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Virtio IDs are spread all over the tree which makes assigning new IDs
bothersome. Putting them together should make the process less error-prone.
Signed-off-by: Fernando Luis Vazquez Cao <fernando@oss.ntt.co.jp>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
This API change means that virtio_net can tell how much capacity
remains for buffers. It's necessarily fuzzy, since
VIRTIO_RING_F_INDIRECT_DESC means we can fit any number of descriptors
in one, *if* we can kmalloc.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Cc: Dinesh Subhraveti <dineshs@us.ibm.com>
No need to put ethtool_ops in data, they should be const.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
These are all drivers that don't touch real hardware.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If we run out of memory, use keventd to fill the buffer. There's a
report of this happening: "Page allocation failures in guest",
Message-ID: <20090713115158.0a4892b0@mjolnir.ossman.eu>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
- Allow setting UFO on virtio-net and advertise to host.
Signed-off-by: Sridhar Samudrala <sri@us.ibm.com>
Acked-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch is inspired by patch recently posted by Johannes Berg. Basically what
my patch does is to group list and a count of addresses into newly introduced
structure netdev_hw_addr_list. This brings us two benefits:
1) struct net_device becames a bit nicer.
2) in the future there will be a possibility to operate with lists independently
on netdevices (with exporting right functions).
I wanted to introduce this patch before I'll post a multicast lists conversion.
Signed-off-by: Jiri Pirko <jpirko@redhat.com>
drivers/net/bnx2.c | 4 +-
drivers/net/e1000/e1000_main.c | 4 +-
drivers/net/ixgbe/ixgbe_main.c | 6 +-
drivers/net/mv643xx_eth.c | 2 +-
drivers/net/niu.c | 4 +-
drivers/net/virtio_net.c | 10 ++--
drivers/s390/net/qeth_l2_main.c | 2 +-
include/linux/netdevice.h | 17 +++--
net/core/dev.c | 130 ++++++++++++++++++--------------------
9 files changed, 89 insertions(+), 90 deletions(-)
Signed-off-by: David S. Miller <davem@davemloft.net>
This replaces find_vq/del_vq with find_vqs/del_vqs virtio operations,
and updates all drivers. This is needed for MSI support, because MSI
needs to know the total number of vectors upfront.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> (+ lguest/9p compile fixes)
Add a linked list of all virtqueues for a virtio device: this helps for
debugging and is also needed for upcoming interface change.
Also, add a "name" field for clearer debug messages.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We need to enforce the IP alignment on the non-mergeable RX path just
like the other RX path. Not doing so results in misaligned IP
headers.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Through a bug in the tun driver, I noticed that virtio_net is
producing bogus hdr_len values. In particular, it only includes
the IP header in the linear area, and excludes the entire TCP
header. This causes the TCP header to be copied twice for each
packet. (The bug omitted the second copy :)
This patch corrects this.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch converts unicast address list to standard list_head using
previously introduced struct netdev_hw_addr. It also relaxes the
locking. Original spinlock (still used for multicast addresses) is not
needed and is no longer used for a protection of this list. All
reading and writing takes place under rtnl (with no changes).
I also removed a possibility to specify the length of the address
while adding or deleting unicast address. It's always dev->addr_len.
The convertion touched especially e1000 and ixgbe codes when the
change is not so trivial.
Signed-off-by: Jiri Pirko <jpirko@redhat.com>
drivers/net/bnx2.c | 13 +--
drivers/net/e1000/e1000_main.c | 24 +++--
drivers/net/ixgbe/ixgbe_common.c | 14 ++--
drivers/net/ixgbe/ixgbe_common.h | 4 +-
drivers/net/ixgbe/ixgbe_main.c | 6 +-
drivers/net/ixgbe/ixgbe_type.h | 4 +-
drivers/net/macvlan.c | 11 +-
drivers/net/mv643xx_eth.c | 11 +-
drivers/net/niu.c | 7 +-
drivers/net/virtio_net.c | 7 +-
drivers/s390/net/qeth_l2_main.c | 6 +-
drivers/scsi/fcoe/fcoe.c | 16 ++--
include/linux/netdevice.h | 18 ++--
net/8021q/vlan.c | 4 +-
net/8021q/vlan_dev.c | 10 +-
net/core/dev.c | 195 +++++++++++++++++++++++++++-----------
net/dsa/slave.c | 10 +-
net/packet/af_packet.c | 4 +-
18 files changed, 227 insertions(+), 137 deletions(-)
Signed-off-by: David S. Miller <davem@davemloft.net>
We were avoiding calling sg_init* on scatterlists passed
into virtnet_send_command to prevent extraneous end markers.
This caused build warnings for uninitialized variables.
Cleanup the code to create proper scatterlists.
Signed-off-by: Alex Williamson <alex.williamson@hp.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
VIRTIO_NET_F_MAC indicates the presence of the mac field in config
space, not the validity of the value it contains. Allow the mac to be
changed at runtime, but only push the change into config space with the
VIRTIO_NET_F_MAC feature present.
Signed-off-by: Alex Williamson <alex.williamson@hp.com>
Acked-by: Christian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Impact: Make NetworkManager work with virtio_net
For now the semantics are simple: There is always carrier.
This allows a seamless experience with e.g., qemu/kvm
where NetworkManager just configures and sets up
everything automagically.
If/when a generally agreed-upon way to control
carrier on/off in the emulator/hypervisor level
emerges, it will be trivial to extend the driver
to support that too, but for now even this 2-liner
makes user experience that much better.
Signed-off-by: Pantelis Koukousoulas <pktoss@gmail.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Many physical NICs let the OS re-program the "hardware" MAC
address. Virtual NICs should allow this too.
Signed-off-by: Alex Williamson <alex.williamson@hp.com>
Acked-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
VLAN filtering allows the hypervisor to drop packets from VLANs
that we're not a part of, further reducing the number of extraneous
packets recieved. This makes use of the VLAN virtqueue command class.
The CTRL_VLAN feature bit tells us whether the backend supports VLAN
filtering.
Signed-off-by: Alex Williamson <alex.williamson@hp.com>
Acked-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Make use of the MAC control virtqueue class to support a MAC
filter table. The filter table is managed by the hypervisor.
We consider the table to be available if the CTRL_RX feature
bit is set. We leave it to the hypervisor to manage the table
and enable promiscuous or all-multi mode as necessary depending
on the resources available to it.
Signed-off-by: Alex Williamson <alex.williamson@hp.com>
Acked-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Make use of the RX_MODE control virtqueue class to enable the
set_rx_mode netdev interface. This allows us to selectively
enable/disable promiscuous and allmulti mode so we don't see
packets we don't want. For now, we automatically enable these
as needed if additional unicast or multicast addresses are
requested.
Signed-off-by: Alex Williamson <alex.williamson@hp.com>
Acked-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
This will be used for RX mode, MAC filter table, VLAN filtering, etc...
The control transaction consists of one or more "out" sg entries and
one or more "in" sg entries. The first out entry contains a header
defining the class and command. Additional out entries may provide
data for the command. The last in entry provides a status response
back from the command.
Virtqueues typically run asynchronous, running a callback function
when there's data in the channel. We can't readily make use of this
in the command paths where we need to use this. Instead, we kick
the virtqueue and spin. The kick causes an I/O write, triggering an
immediate trap into the hypervisor.
Signed-off-by: Alex Williamson <alex.williamson@hp.com>
Acked-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Without this fix, virtio_net makes incorrect usage of scatterlists. It sets
the end of the scatterlist chain after the first element, despite the fact
that more entries come after it.
If you try to run dma_map_sg() on one of the scatterlists given to you by
add_buf(), you will get a null pointer oops.
Signed-off-by: Ira W. Snyder <iws@ovro.caltech.edu>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
802.1Q expanded the maximum ethernet frame size by 4 bytes for the
VLAN tag. We're not taking this into account in virtio_net, which
means the buffers we provide to the backend in the virtqueue RX ring
aren't big enough to hold a full MTU VLAN packet. For QEMU/KVM,
this results in the backend exiting with a packet truncation error.
Signed-off-by: Alex Williamson <alex.williamson@hp.com>
Acked-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Allow the host to inform us that the link is down by adding
a VIRTIO_NET_F_STATUS which indicates that device status is
available in virtio_net config.
This is currently useful for simulating link down conditions
(e.g. using proposed qemu 'set_link' monitor command) but
would also be needed if we were to support device assignment
via virtio.
Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> (added future masking)
Signed-off-by: David S. Miller <davem@davemloft.net>
Following the removal of the unused struct net_device * parameter from
the NAPI functions named *netif_rx_* in commit 908a7a1, they are
exactly equivalent to the corresponding *napi_* functions and are
therefore redundant.
Signed-off-by: Ben Hutchings <bhutchings@solarflare.com>
Acked-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Acked-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When the napi api was changed to separate its 1:1 binding to the net_device
struct, the netif_rx_[prep|schedule|complete] api failed to remove the now
vestigual net_device structure parameter. This patch cleans up that api by
properly removing it..
Signed-off-by: Neil Horman <nhorman@tuxdriver.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We don't really have a max tx packet size limit, so allow configuring
the device with up to 64k tx MTU.
Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
If segmentation offload is enabled by the host, we currently allocate
maximum sized packet buffers and pass them to the host. This uses up
20 ring entries, allowing us to supply only 20 packet buffers to the
host with a 256 entry ring. This is a huge overhead when receiving
small packets, and is most keenly felt when receiving MTU sized
packets from off-host.
The VIRTIO_NET_F_MRG_RXBUF feature flag is set by hosts which support
using receive buffers which are smaller than the maximum packet size.
In order to transfer large packets to the guest, the host merges
together multiple receive buffers to form a larger logical buffer.
The number of merged buffers is returned to the guest via a field in
the virtio_net_hdr.
Make use of this support by supplying single page receive buffers to
the host. On receive, we extract the virtio_net_hdr, copy 128 bytes of
the payload to the skb's linear data buffer and adjust the fragment
offset to point to the remaining data. This ensures proper alignment
and allows us to not use any paged data for small packets. If the
payload occupies multiple pages, we simply append those pages as
fragments and free the associated skbs.
This scheme allows us to be efficient in our use of ring entries
while still supporting large packets. Benchmarking using netperf from
an external machine to a guest over a 10Gb/s network shows a 100%
improvement from ~1Gb/s to ~2Gb/s. With a local host->guest benchmark
with GSO disabled on the host side, throughput was seen to increase
from 700Mb/s to 1.7Gb/s.
Based on a patch from Herbert Xu.
Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> (use netdev_priv)
Signed-off-by: David S. Miller <davem@davemloft.net>
Seems like an oversight that we have set-tx-csum and set-sg hooked
up, but not set-tso.
Also leads to the strange situation that if you e.g. disable tx-csum,
then tso doesn't get disabled.
Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: David S. Miller <davem@davemloft.net>
Each time we re-fill the recv queue with buffers, we allocate
one too many skbs and free it again when adding fails. We should
recycle the pages allocated in this case.
A previous version of this patch made trim_pages() trim trailing
unused pages from skbs with some paged data, but this actually
caused a barely measurable slowdown.
Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> (use netdev_priv)
Signed-off-by: David S. Miller <davem@davemloft.net>
We have some reasons to kill netdev->priv:
1. netdev->priv is equal to netdev_priv().
2. netdev_priv() wraps the calculation of netdev->priv's offset, obviously
netdev_priv() is more flexible than netdev->priv.
But we cann't kill netdev->priv, because so many drivers reference to it
directly.
This patch is a safe convert for netdev->priv to netdev_priv(netdev).
Since all of the netdev->priv is only for read.
But it is too big to be sent in one mail.
I split it to 4 parts and make every part smaller than 100,000 bytes,
which is max size allowed by vger.
Signed-off-by: Wang Chen <wangchen@cn.fujitsu.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This converts pretty much everything to print_mac. There were
a few things that had conflicts which I have just dropped for
now, no harm done.
I've built an allyesconfig with this and looked at the files
that weren't built very carefully, but it's a huge patch.
Signed-off-by: Johannes Berg <johannes@sipsolutions.net>
Signed-off-by: David S. Miller <davem@davemloft.net>
If we hack the virtio_net driver to always allocate full-sized (64k+)
skbuffs, the driver slows down (lguest numbers):
Time to receive 1GB (small buffers): 10.85 seconds
Time to receive 1GB (64k+ buffers): 24.75 seconds
Of course, large buffers use up more space in the ring, so we increase
that from 128 to 2048:
Time to receive 1GB (64k+ buffers, 2k ring): 16.61 seconds
If we recycle pages rather than using alloc_page/free_page:
Time to receive 1GB (64k+ buffers, 2k ring, recycle pages): 10.81 seconds
This demonstrates that with efficient allocation, we don't need to
have a separate "small buffer" queue.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Finally this patch lets virtio_net receive GSO packets in addition
to sending them. This can definitely be optimised for the non-GSO
case. For comparison the Xen approach stores one page in each skb
and uses subsequent skb's pages to construct an SG skb instead of
preallocating the maximum amount of pages per skb.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> (added feature bits)
This patch adds some basic ethtool operations to virtio_net so
I could test SG without GSO (which was really useful because TSO
turned out to be buggy :)
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au> (remove MTU setting)
On Mon, 2008-05-26 at 17:42 +1000, Rusty Russell wrote:
> If we fail to transmit a packet, we assume the queue is full and put
> the skb into last_xmit_skb. However, if more space frees up before we
> xmit it, we loop, and the result can be transmitting the same skb twice.
>
> Fix is simple: set skb to NULL if we've used it in some way, and check
> before sending.
...
> diff -r 564237b31993 drivers/net/virtio_net.c
> --- a/drivers/net/virtio_net.c Mon May 19 12:22:00 2008 +1000
> +++ b/drivers/net/virtio_net.c Mon May 19 12:24:58 2008 +1000
> @@ -287,21 +287,25 @@ again:
> free_old_xmit_skbs(vi);
>
> /* If we has a buffer left over from last time, send it now. */
> - if (vi->last_xmit_skb) {
> + if (unlikely(vi->last_xmit_skb)) {
> if (xmit_skb(vi, vi->last_xmit_skb) != 0) {
> /* Drop this skb: we only queue one. */
> vi->dev->stats.tx_dropped++;
> kfree_skb(skb);
> + skb = NULL;
> goto stop_queue;
> }
> vi->last_xmit_skb = NULL;
With this, may drop an skb and then later in the function discover that
we could have sent it after all. Poor wee skb :)
How about the incremental patch below?
Cheers,
Mark.
Subject: [PATCH] virtio_net: Delay dropping tx skbs
Currently we drop the skb in start_xmit() if we have a
queued buffer and fail to transmit it.
However, if we delay dropping it until we've stopped the
queue and enabled the tx notification callback, then there
is a chance space might become available for it.
Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
We can handle receiving partial csums, so set the
appropriate feature bit.
Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
virtio_net uses a timer to free old transmitted packets, rather than
leaving callbacks enabled all the time. If the host promises to
always notify us when the transmit ring is empty, we can free packets
at that point and avoid the timer.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
virtio_net currently only frees old transmit skbs just
before queueing new ones. If the queue is full, it then
enables interrupts and waits for notification that more
work has been performed.
However, a side-effect of this scheme is that there are
always xmit skbs left dangling when no new packets are
sent, against the Documentation/networking/driver.txt
guideline:
"... it is not allowed for your TX mitigation scheme
to let TX packets "hang out" in the TX ring unreclaimed
forever if no new TX packets are sent."
Add a timer to ensure that any time we queue new TX
skbs, we will shortly free them again.
This fixes an easily reproduced hang at shutdown where
iptables attempts to unload nf_conntrack and nf_conntrack
waits for an skb it is tracking to be freed, but virtio_net
never frees it.
Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
hdr->csum_start is the offset from the start of the ethernet
header to the transport layer checksum field. skb->csum_start
is the offset from skb->head.
skb_partial_csum_set() assumes that skb->data points to the
ethernet header - i.e. it computes skb->csum_start by adding
the headroom to hdr->csum_start.
Since eth_type_trans() skb_pull()s the ethernet header,
skb_partial_csum_set() should be called before
eth_type_trans().
(Without this patch, GSO packets from a guest to the world outside the
host are corrupted).
Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Acked-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
Because we cache the last failed-to-xmit packet, if there are no
packets queued behind that one we may never send it (reproduced here
as TCP stalls, "cured" by an outgoing ping).
Cc: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
If we fail to transmit a packet, we assume the queue is full and put
the skb into last_xmit_skb. However, if more space frees up before we
xmit it, we loop, and the result can be transmitting the same skb twice.
Fix is simple: set skb to NULL if we've used it in some way, and check
before sending.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Jeff Garzik <jgarzik@redhat.com>
A recent proposed feature addition to the virtio block driver revealed
some flaws in the API: in particular, we assume that feature
negotiation is complete once a driver's probe function returns.
There is nothing in the API to require this, however, and even I
didn't notice when it was violated.
So instead, we require the driver to specify what features it supports
in a table, we can then move the feature negotiation into the virtio
core. The intersection of device and driver features are presented in
a new 'features' bitmap in the struct virtio_device.
Note that this highlights the difference between Linux unsigned-long
bitmaps where each unsigned long is in native endian, and a
straight-forward little-endian array of bytes.
Drivers can still remove feature bits in their probe routine if they
really have to.
API changes:
- dev->config->feature() no longer gets and acks a feature.
- drivers should advertise their features in the 'feature_table' field
- use virtio_has_feature() for extra sanity when checking feature bits
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
So, we previously had a 'VIRTIO_NET_F_GSO' bit which meant that 'the
host can handle csum offload, and any TSO (v4&v6 incl ECN) or UFO
packets you might want to send. I thought this was good enough for
Linux, but it actually isn't, since we don't do UFO in software.
So, add separate feature bits for what the host can handle. Add
equivalent ones for the guest to say what it can handle, because LRO
is coming too (thanks Herbert!).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Herbert tells me that returning NETDEV_TX_BUSY from hard_start_xmit is
seen as a poor thing to do; we should cache the packet and stop the queue.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Acked-by: Herbert Xu <herbert@gondor.apana.org.au>
Herbert Xu points out (within another patch) that my scatterlists are
too short: one entry for the gso header, one for the skb->data, and
MAX_SKB_FRAGS for all the fragments.
Fix both xmit and recv sides (recv currently unused, coming in later
patch).
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
get_buf() gives the length written by the other side, which will be
zero. We want to add the skb length.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6:
[NETNS][IPV6] tcp - assign the netns for timewait sockets
[IPV4]: Fix byte value boundary check in do_ip_getsockopt().
BNX2X: Correct bringing chip out of reset
[NETFILTER]: nf_nat: autoload IPv4 connection tracking
[NETFILTER]: xt_hashlimit: fix mask calculation
[XFRM]: xfrm_user: fix selector family initialization
rt61pci: rt61pci_beacon_update do not free skb twice
ssb-mipscore: Fix interrupt vectors
ssb-pcicore: Fix IRQ TPS flag handling
mac80211: use short_preamble mode from capability if ERP IE not present
[NET]: Undo code bloat in hot paths due to print_mac().
[TCP]: Don't allow FRTO to take place while MTU is being probed
[TCP]: tcp_simple_retransmit can cause S+L
[TCP]: Fix NewReno's fast rexmit/recovery problems with GSOed skb
[TCP]: Restore 2.6.24 mark_head_lost behavior for newreno/fack
nl80211: fix STA AID bug
b43legacy: fix bcm4303 crash
iwlwifi: fix n-band association problem
ipw2200: set MAC address on radiotap interface
libertas: fix mode initialization problem
Michael S. Tsirkin