The netvsc driver can get repeated calls to netvsc_rx_mode during
network setup; each of these calls ends up scheduling the lower
layers to update tha packet filter. This update requires an
request/response to the host. So avoid doing this if we already
know that the correct packet filter value is set.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The recent change to not always enable all multicast and broadcast
was broken; meant to set filter, not change flags.
Fixes: 009f766ca2 ("hv_netvsc: filter multicast/broadcast")
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The netvsc device should propagate filters to the SR-IOV VF
device (if present). The flags also need to be propagated to the
VF device as well. This only really matters on local Hyper-V
since Azure does not support multiple addresses.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The netvsc driver was always enabling all multicast and broadcast
even if netdevice flag had not enabled it.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When VF is used for accelerated networking it will likely have
more queues (and different policy) than the synthetic NIC.
This patch defers the queue policy to the VF so that all the
queues can be used. This impacts workloads like local generate UDP.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since the netvsc_channel_cb is already called in interrupt
context from vmbus, there is no need to do irqsave/restore.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
There is a race between napi_reschedule and re-enabling interrupts
which could lead to missed host interrrupts. This occurs when
interrupts are re-enabled (hv_end_read) and vmbus irq callback
(netvsc_channel_cb) has already scheduled NAPI.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Block setup of multiple channels earlier in the teardown
process. This avoids possible races between halt and subchannel
initialization.
Suggested-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Need to delete NAPI association if vmbus_open fails.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Don't wake transmit queues if link is not up yet.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Change the initialization order so that the device is ready to transmit
(ie connect vsp is completed) before setting the internal reference
to the device with RCU.
This avoids any races on initialization and prevents retry issues
on shutdown.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since we no longer localize channel/CPU affiliation within one NUMA
node, num_online_cpus() is used as the number of channel cap, instead of
the number of processors in a NUMA node.
This patch allows a bigger range for tuning the number of channels.
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If the transmit queue is known full, then don't keep aggregating
data. And the cp_partial flag which indicates that the current
aggregation buffer is full can be folded in to avoid more
conditionals.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
There is only ever a single instance of network device object
referencing the internal rndis object. Therefore the open_cnt atomic
is not necessary.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The netvsc_receive_callback function was using RCU to find the
appropriate underlying netvsc_device. Since calling function already
had that pointer, this was unnecessary.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The caller (netvsc_receive) already has the net device pointer,
and should just pass that to functions rather than the hyperv device.
This eliminates several impossible error paths in the process.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When skb can not be allocated, update ethtool statisitics
rather than rx_dropped which is intended for netif_receive.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since only caller does not care about return value.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The values were not computed correctly. There are no significant
visible impact, though.
The intended size of RX buffer is 16 MB, and the default slot size is 1728.
So, NETVSC_DEFAULT_RX should be 16*1024*1024 / 1728 = 9709.
The intended size of TX buffer is 1 MB, and the slot size is 6144.
So, NETVSC_DEFAULT_TX should be 1024*1024 / 6144 = 170.
The patch puts the formula directly into the macro, and moves them to
hyperv_net.h, together with related macros.
Fixes: 5023a6db73 ("netvsc: increase default receive buffer size")
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The max should be 31 MB on host with NVSP version > 2.
On legacy hosts (NVSP version <=2) only 15 MB receive buffer is allowed,
otherwise the buffer request will be rejected by the host, resulting
vNIC not coming up.
The NVSP version is only available after negotiation. So, we add the
limit checking for legacy hosts in netvsc_init_buf().
Fixes: 5023a6db73 ("netvsc: increase default receive buffer size")
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The memset of the whole maximum possible RNDIS header is unnecessary.
For the main part of the header use a structure assignment.
No need to memset the whole per packet info. Instead rely on caller to
set what it wants. Also get rid of cast to void and signed/unsigned
conversion. Now return pointer to per packet data (rather than the
header) which simplifies use by code setting up the packet data.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Every packet sent checks the available ring space. The calculation
can be sped up by using reciprocal divide which is multiplication.
Since ring_size can only be configured by module parameter, so it doesn't
have to be passed around everywhere. Also it should be unsigned
since it is number of pages.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Packet alignment is always a power of 2 therefore modulus can
be replaced with a faster and operation
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since skb is always non-NULL in the copy portion of netvsc_send
do not need local variable.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
rndis_filter_device_add() is called both from netvsc_probe() when we
initially create the device and from set channels/mtu/ringparam
routines where we basically remove the device and add it back.
hw_features is reset in rndis_filter_device_add() and filled with
host data. However, we lose all additional flags which are set outside
of the driver, e.g. register_netdevice() adds NETIF_F_SOFT_FEATURES and
many others.
Unfortunately, calls to rndis_{query_hwcaps(), _set_offload_params()}
calls cannot be avoided on every RNDIS reset: host expects us to set
required features explicitly. Moreover, in theory hardware capabilities
can change and we need to reflect the change in hw_features.
Reset net->hw_features bits according to host data in
rndis_netdev_set_hwcaps(), clear corresponding feature bits
from net->features in case some features went missing (will never happen
in real life I guess but let's be consistent).
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Reviewed-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Hyper-V hosts are known to send RNDIS messages even after we halt the
device in rndis_filter_halt_device(). Remove user visible messages
as they are not really useful.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
It was found that in some cases host refuses to teardown GPADL for send/
receive buffers (probably when some work with these buffere is scheduled or
ongoing). Change the teardown logic to be:
1) Send NVSP_MSG1_TYPE_REVOKE_* messages
2) Close the channel
3) Teardown GPADLs.
This seems to work reliably.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In some cases, like internal vSwitch, the host doesn't provide
send indirection table updates. This patch sets the table to be
equal weight after subchannels are all open. Otherwise, all workload
will be on one TX channel.
As tested, this patch has largely increased the throughput over
internal vSwitch.
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
tx_table is part of the private data of kernel net_device. It is only
zero-ed out when allocating net_device.
We may recreate netvsc_device w/o recreating net_device, so the private
netdev data, including tx_table, are not zeroed. It may contain channel
numbers for the older netvsc_device.
This patch adds initialization of tx_table each time we recreate
netvsc_device.
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Rename this variable because it is the Receive indirection
table.
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The patch supports the options to switch TCP hash level between
L3 and L4 by ethtool command. TCP over IPv4 and v6 can be set
differently. The default hash level is L4. We currently only
allow switching TX hash level from within the guests.
For example, for TCP over IPv4 on eth0:
To include TCP port numbers in hashing:
ethtool -N eth0 rx-flow-hash tcp4 sdfn
To exclude TCP port numbers in hashing:
ethtool -N eth0 rx-flow-hash tcp4 sd
To show TCP hash level:
ethtool -n eth0 rx-flow-hash tcp4
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This simplifies the logic and make it easier to add more
options.
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add extack arg to netdev_upper_dev_link and netdev_master_upper_dev_link
Signed-off-by: David Ahern <dsahern@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Report the numbers of events for stop_queue and wake_queue in
ethtool stats.
Example:
ethtool -S eth0
NIC statistics:
...
stop_queue: 7
wake_queue: 7
...
Signed-off-by: Simon Xiao <sixiao@microsoft.com>
Reviewed-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
For older hosts without multi-channel (vRSS) support, and some error
cases, we still need to set the real number of queues to one.
This patch adds this missing setting.
Fixes: 8195b1396e ("hv_netvsc: fix deadlock on hotplug")
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Reviewed-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Don't populate const array ver_list on the stack, instead make it
static. Makes the object code smaller by over 400 bytes:
Before:
text data bss dec hex filename
18444 3168 320 21932 55ac drivers/net/hyperv/netvsc.o
After:
text data bss dec hex filename
17950 3224 320 21494 53f6 drivers/net/hyperv/netvsc.o
(gcc 6.3.0, x86-64)
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Reviewed-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If MTU is changed the host would reject the send buffer change.
This problem is result of recent change to allow changing send
buffer size.
Every time we change the MTU, we store the previous net_device section
count before destroying the buffer, but we don’t store the previous
section size. When we reinitialize the buffer, its size is calculated
by multiplying the previous count and previous size. Since we
continuously increase the MTU, the host returns us a decreasing count
value while the section size is reinitialized to 1728 bytes every
time.
This eventually leads to a condition where the calculated buf_size is
so small that the host rejects it.
Fixes: 8b5327975a ("netvsc: allow controlling send/recv buffer size")
Signed-off-by: Alex Ng <alexng@microsoft.com>
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The default receive buffer size was reduced by recent change
to a value which was appropriate for 10G and Windows Server 2016.
But the value is too small for full performance with 40G on Azure.
Increase the default back to maximum supported by host.
Fixes: 8b5327975a ("netvsc: allow controlling send/recv buffer size")
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Only need to wakeup the initiator after all sub-channels
are opened.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When a virtual device is added dynamically (via host console), then
the vmbus sends an offer message for the primary channel. The processing
of this message for networking causes the network device to then
initialize the sub channels.
The problem is that setting up the sub channels needs to wait until
the subsequent subchannel offers have been processed. These offers
come in on the same ring buffer and work queue as where the primary
offer is being processed; leading to a deadlock.
This did not happen in older kernels, because the sub channel waiting
logic was broken (it wasn't really waiting).
The solution is to do the sub channel setup in its own work queue
context that is scheduled by the primary channel setup; and then
happens later.
Fixes: 732e49850c ("netvsc: fix race on sub channel creation")
Reported-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The limit of setting receive indirection table value should be
the current number of channels, not the VRSS_CHANNEL_MAX.
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Because of the following code, net->num_tx_queues equals to
VRSS_CHANNEL_MAX, and max_chn is less than or equals to VRSS_CHANNEL_MAX.
netvsc_drv.c:
alloc_etherdev_mq(sizeof(struct net_device_context),
VRSS_CHANNEL_MAX);
rndis_filter.c:
net_device->max_chn = min_t(u32, VRSS_CHANNEL_MAX, num_possible_rss_qs);
So this patch removes the unnecessary limit check before comparing
with "max_chn".
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The minus one and assignment to a local variable is not necessary.
This patch simplifies it.
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch removes the parameter, num_queue in
rndis_filter_set_rss_param(), which is no longer in use.
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If VF is attached then can still allow netvsc driver module to
be removed. Just have to make sure and do the cleanup.
Also, avoid extra rtnl round trip when calling unregister.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use one routine for datapath up/down. Don't need to reopen
the rndis layer.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
There is a deadlock possible when canceling the link status
delayed work queue. The removal process is run with RTNL held,
and the link status callback is acquring RTNL.
Resolve the issue by using trylock and rescheduling.
If cancel is in process, that block it from happening.
Fixes: 122a5f6410 ("staging: hv: use delayed_work for netvsc_send_garp()")
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We now remove rndis filter before unregister_netdev(), which calls
device close. It involves closing rndis filter already removed.
This patch fixes this error.
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The patch add the functions to switch UDP hash level between
L3 and L4 by ethtool command. UDP over IPv4 and v6 can be set
differently. The default hash level is L4. We currently only
allow switching TX hash level from within the guests.
On Azure, fragmented UDP packets have high loss rate with L4
hashing. Using L3 hashing is recommended in this case.
For example, for UDP over IPv4 on eth0:
To include UDP port numbers in hasing:
ethtool -N eth0 rx-flow-hash udp4 sdfn
To exclude UDP port numbers in hasing:
ethtool -N eth0 rx-flow-hash udp4 sd
To show UDP hash level:
ethtool -n eth0 rx-flow-hash udp4
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The only usage of vmbus_sendpacket_ctl was by vmbus_sendpacket.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The function vmbus_sendpacket_pagebuffer_ctl was never used directly.
Just have vmbus_send_pagebuffer
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add ethtool statistics for case where send chimmeny buffer is
exhausted and driver has to fall back to doing scatter/gather
send. Also, add statistic for case where ring buffer is full and
receive completions are delayed.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Control the size of the buffer areas via ethtool ring settings.
They aren't really traditional hardware rings, but host API breaks
receive and send buffer into chunks. The final size of the chunks are
controlled by the host.
The default value of send and receive buffer area for host DMA
is much larger than it needs to be. Experimentation shows that
4M receive and 1M send is sufficient.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The function init_page_array is always called with a valid pointer
to RNDIS header. No check for NULL is needed.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Assignment to a typed pointer is sufficient in C.
No cast is needed.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The send and receive buffers are both per-device (not per-channel).
The associated NUMA node is a property of the CPU which is per-channel
therefore it makes no sense to force the receive/send buffer to be
allocated on a particular node (since it is a shared resource).
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If setting new values fails, and the attempt to restore original
settings fails. Then log an error and leave device down.
This should never happen, but if it does don't go down in flames.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If VF is slaved to synthetic device, then any change to netvsc
MAC address should be propagated to the slave device.
If slave device doesn't support MAC address change then it
should also be an error to attempt to change synthetic NIC MAC
address.
It also fixes the error unwind in the original code.
If give a bad address, the old code would change the device
MAC address anyway.
Reviewed-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When hv_pkt_iter_next() returns NULL, it has already called
hv_pkt_iter_close(). Calling it twice can lead to extra host signal.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When VF device is discovered, delay bring it automatically up in
order to allow userspace to some simple changes (like renaming).
Reported-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The UDP offload conflict is dealt with by simply taking what is
in net-next where we have removed all of the UFO handling code
entirely.
The TCP conflict was a case of local variables in a function
being removed from both net and net-next.
In netvsc we had an assignment right next to where a missing
set of u64 stats sync object inits were added.
Signed-off-by: David S. Miller <davem@davemloft.net>
Go back to switching datapath directly in the notifier callback.
Otherwise datapath might not get switched on unregister.
No need for calling the NOTIFY_PEERS notifier since that is only for
a gratitious ARP/ND packet; but that is not required with Hyper-V
because both VF and synthetic NIC have the same MAC address.
Reported-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Fixes: 0c195567a8 ("netvsc: transparent VF management")
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
With new transparent VF support, it is possible to get a deadlock
when some of the deferred work is running and the unregister_vf
is trying to cancel the work element. The solution is to use
trylock and reschedule (similar to bonding and team device).
Reported-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Fixes: 0c195567a8 ("netvsc: transparent VF management")
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The existing sub channel code did not wait for all the sub-channels
to completely initialize. This could lead to race causing crash
in napi_netif_del() from bad list. The existing code would send
an init message, then wait only for the initial response that
the init message was received. It thought it was waiting for
sub channels but really the init response did the wakeup.
The new code keeps track of the number of open channels and
waits until that many are open.
Other issues here were:
* host might return less sub-channels than was requested.
* the new init status is not valid until after init was completed.
Fixes: b3e6b82a00 ("hv_netvsc: Wait for sub-channels to be processed during probe")
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch implements transparent fail over from synthetic NIC to
SR-IOV virtual function NIC in Hyper-V environment. It is a better
alternative to using bonding as is done now. Instead, the receive and
transmit fail over is done internally inside the driver.
Using bonding driver has lots of issues because it depends on the
script being run early enough in the boot process and with sufficient
information to make the association. This patch moves all that
functionality into the kernel.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Repeated dereference of nvmsg.msg.v1_msg.send_rndis_pkt can be
shortened by using a temporary. Do so.
No change in object code.
Miscellanea:
o Use * const for rpkt and nvchan
Signed-off-by: Joe Perches <joe@perches.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
On 32-bit hosts and with CONFIG_DEBUG_LOCK_ALLOC we should be seeing a
lockdep splat indicating this seqcount is not correctly initialized, fix
that. In commit 6c80f3fc23 ("netvsc: report per-channel stats in
ethtool statistics") netdev_alloc_pcpu_stats() was removed in favor of
open-coding the 64-bits statistics, except that u64_stats_init() was
missed.
Fixes: 6c80f3fc23 ("netvsc: report per-channel stats in ethtool statistics")
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Two minor conflicts in virtio_net driver (bug fix overlapping addition
of a helper) and MAINTAINERS (new driver edit overlapping revamp of
PHY entry).
Signed-off-by: David S. Miller <davem@davemloft.net>
Latency improvement related to NAPI conversion.
If all packets are processed from receive ring then need
to signal host.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If setting receive buffer fails, the error unwind would cause
kernel panic because it was not correctly doing RCU and NAPI
unwind. RCU'd pointer needs to be reset to NULL, and NAPI needs
to be disabled not deleted.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Optimize how receive completion ring are managed.
* Allocate only as many slots as needed for all buffers from host
* Allocate before setting up sub channel for better error detection
* Don't need to keep copy of initial receive section message
* Precompute the watermark for when receive flushing is needed
* Replace division with conditional test
* Replace atomic per-device variable with per-channel check.
* Handle corner case where receive completion send
fails if ring buffer to host is full.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The internal API was passing struct hv_page_buffer **
when only simple struct hv_page_buffer * was necessary
for passing an array.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Using %p to print pointer to packet meta-data doesn't give any
good info, and exposes kernel memory offsets.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This includes a bunch of fixups for issues reported by
lockdep.
* ethtool routines can assume RTNL
* send is done with RCU lock (and BH disable)
* avoid refetching internal device struct (netvsc)
instead pass it as a parameter.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The logic for computing page buffer scatter does not take into
account the impact of compound pages. Therefore the optimization
to compute number of slots was incorrect and could cause stack
corruption a skb was sent with lots of fragments from huge pages.
This reverts commit 60b86665af.
Fixes: 60b86665af ("netvsc: optimize calculation of number of slots")
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This value has been calculated in rndis_device_attach since 4.11.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since these files use rtnl_derefernce make sure and include rtnetlink.h
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The number of channels returned by rndis_filter_device_add maybe
less than the number requested. Therefore set correct real
number of queues.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In interrupt handler, prefetch the first incoming ring element
so that it is in cache by the time NAPI poll gets to it.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This condition already uses an object of type ipv6hdr in the line above.
Use the object directly instead of calling ipv6_hdr
Signed-off-by: Mohammed Gamal <mgamal@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Remove accidental rtnl_unlock from earlier testing.
Fixes: 3962981f48 ("netvsc: add rtnl annotations in rndis")
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
drivers/net/hyperv/netvsc_drv.c:737:8-14: WARNING: PTR_ERR_OR_ZERO can be used
Use PTR_ERR_OR_ZERO rather than if(IS_ERR(...)) + PTR_ERR
Generated by: scripts/coccinelle/api/ptr_ret.cocci
Fixes: 9749fed5d4 ("netvsc: use ERR_PTR to avoid dereference issues")
CC: stephen hemminger <stephen@networkplumber.org>
Signed-off-by: Fengguang Wu <fengguang.wu@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The rndis functions are used when changing device state.
Therefore the references from network device to internal state
are protected by RTNL mutex.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Keep back pointer in the per-channel data structure to
avoid any possible RCU related issues when napi poll is
called but netvsc_device is in RCU limbo.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The netvsc_device structure should be accessed by rcu_dereference
in the send path. Change arguments to netvsc_send() to make
this easier to do correctly.
Remove no longer needed hv_device_to_netvsc_device.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The rndis_filter_device_add function is called both in
probe context and RTNL context,and creates the netvsc_device
inner structure. It is easier to get the RTNL lock annotation
correct if it returns the object directly, rather than implicitly
by updating network device private data.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use device detach/attach to ensure that no packets are handed
to device during state changes. Call rndis_filter_open/close
directly as part of later VF related changes.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This fixes the error unwind logic for incorrect number of queues.
If netif_set_real_num_XX_queues failed then rndis_filter_device_add
would have been called twice. Since input arguments are already
ranged checked this is a hypothetical only problem, not possible
in actual code.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In a couple places RTNL is held, and the netvsc_device pointer
is acquired without annotation.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If two MTU changes are in less than update interval (2 seconds),
then the netvsc network device may get stuck with no carrier.
The netvsc driver debounces link status events which is fine
for unsolicited updates, but blocks getting the update after
down/up from MTU reinitialization.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When the VF NIC is opened, the synthetic NIC's carrier state is set to
off. This tells the host to transitions data path to the VF device. But
if startup script or user manipulates the admin state of the netvsc
device directly for example:
# ifconfig eth0 down
# ifconfig eth0 up
Then the carrier state of the synthetic NIC would be on, even though the
data path was still over the VF NIC. This patch sets the carrier state
of synthetic NIC with consideration of the related VF state.
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Reviewed-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We simply use rndis_device->link_state in the netdev_dbg. The variable,
link_state from struct netvsc_device_info, is not used anywhere else.
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Reviewed-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This structure member is hidden behind CONFIG_SYSFS, and we
get a build error when that is disabled:
drivers/net/hyperv/netvsc_drv.c: In function 'netvsc_set_channels':
drivers/net/hyperv/netvsc_drv.c:754:49: error: 'struct net_device' has no member named 'num_rx_queues'; did you mean 'num_tx_queues'?
drivers/net/hyperv/netvsc_drv.c: In function 'netvsc_set_rxfh':
drivers/net/hyperv/netvsc_drv.c:1181:25: error: 'struct net_device' has no member named 'num_rx_queues'; did you mean 'num_tx_queues'?
As the value is only set once to the argument of alloc_netdev_mq(),
we can compare against that constant directly.
Fixes: ff4a441990 ("netvsc: allow get/set of RSS indirection table")
Fixes: 2b01888d1b ("netvsc: allow more flexible setting of number of channels")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
A common pattern with skb_put() is to just want to memcpy()
some data into the new space, introduce skb_put_data() for
this.
An spatch similar to the one for skb_put_zero() converts many
of the places using it:
@@
identifier p, p2;
expression len, skb, data;
type t, t2;
@@
(
-p = skb_put(skb, len);
+p = skb_put_data(skb, data, len);
|
-p = (t)skb_put(skb, len);
+p = skb_put_data(skb, data, len);
)
(
p2 = (t2)p;
-memcpy(p2, data, len);
|
-memcpy(p, data, len);
)
@@
type t, t2;
identifier p, p2;
expression skb, data;
@@
t *p;
...
(
-p = skb_put(skb, sizeof(t));
+p = skb_put_data(skb, data, sizeof(t));
|
-p = (t *)skb_put(skb, sizeof(t));
+p = skb_put_data(skb, data, sizeof(t));
)
(
p2 = (t2)p;
-memcpy(p2, data, sizeof(*p));
|
-memcpy(p, data, sizeof(*p));
)
@@
expression skb, len, data;
@@
-memcpy(skb_put(skb, len), data, len);
+skb_put_data(skb, data, len);
(again, manually post-processed to retain some comments)
Reviewed-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: Johannes Berg <johannes.berg@intel.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
No longer need common code to find get_outbound_net_device.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Don't need to find netvsc_device structure, caller already had it.
Also rearrange declarations.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Mark if() statements used for error handling only as unlikely()
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The element netvsc_device:extension is always a pointer to RNDIS
information.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Don't need need to look at write space in netvsc_close.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Speed up transmit check for fragmented packets by using existing
macros to compute number of pages, and eliminate loop since
skb fragments each take a page. Number of slots is also unsigned.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The work queue and handling of network filter parameters should
be in rndis_device. This gets rid of warning from RCU checks,
eliminates a race and cleans up code.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The ndo_poll_controller function needs to schedule NAPI to pick
up arriving packets and send completions. Otherwise no data
will ever be received. For simple case of netconsole, it also
will allow send completions to happen. Without this netpoll
will eventually get stuck.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The ethtool info command calls the netvsc get_sset_count with RTNL
but not with RCU. Which causes warning:
drivers/net/hyperv/netvsc_drv.c:1010 suspicious rcu_dereference_check() usage!
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add direct #include statements for declarations of csum_tcpudp_magic()
and csum_ipv6_magic(). While the needed #include's are picked up
indirectly for the x86 architecture, they aren't on other
architectures, resulting in compile errors.
Signed-off-by: Michael Kelley <mikelley@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This fixes a race where vmbus callback for new packet arriving
could occur before NAPI is initialized.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
My change (introduced in 4.11) to use find_first_clear_bit
incorrectly assumed that the size argument was words, not bits.
The effect was only a small limited number of the available send
sections were being actually used. This can cause performance loss
with some workloads.
Since map_words is now used only during initialization, it can
be on stack instead of in per-device data.
Fixes: b58a185801 ("netvsc: simplify get next send section")
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The NAPI data structure is embedded in the netvsc_device structure
and is freed when device is closed. There is still a reference
(in NAPI list) to this which causes a crash in netif_napi_del
when device is removed. Fix by managing NAPI instances correctly.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
We will not be able to send packets over a channel that has been
rescinded. Make necessary adjustments so we can properly cleanup
even when the channel is rescinded. This issue can be trigerred
in the NIC hot-remove path.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The netvsc device supports full duplex by default.
This warnings in log from bonding device which did not like
seeing UNKNOWN duplex.
Signed-off-by: Simon Xiao <sixiao@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The statistics functionis called with RTNL held during probe
but with RCU held during access from /proc and elsewhere.
This is safe so update the lockdep annotation.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Azure hosts are not supporting non-TCP port numbers in vRSS hashing for
now. For example, UDP packet loss rate will be high if port numbers are
also included in vRSS hash.
So, we created this patch to use only IP numbers for hashing in non-TCP
traffic.
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Reviewed-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If the outgoing skb has a RX queue mapping available, we use the queue
number directly, other than put it through Send Indirection Table.
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Reviewed-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This allows using deferred skb freeing and with NAPI. And get buffer
recycling.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Prior to opening the channel we should have all the state setup to handle
interrupts. The current code does not do that; fix the bug. This bug
can result in faults in the interrupt path.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
ndev is being checked to see if it is a null pointer however before
the null check ndev is being dereferenced; hence there is a potential
null pointer dereference bug that needs fixing. Fix this by only
dereferencing ndev after the null check.
Detected by CoverityScan, CID#1420760, CID#140761 ("Dereference
before null check")
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Reviewed-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
All netvsc channels are handled via NAPI. Setup the "read mode" correctly
for the netvsc sub-channels.
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Fix warning from unused set_complete variable. And rearrange code
to eliminate unnecessary goto's.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since there already is a special case goto for control messages (skb == NULL)
in netvsc_send, there is no need for later checks in same code path.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The channel inbound lock was not being used at all by the netvsc
device, but the spin_lock was helpful by providing necessary
barrier before waiting.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Rather than a lock and variable, use a refcount_t to keep track
of the number of sub channels. Don't need to wait for subchannels
on device removal since wait was already done in device_add.
Also fix the error handling; don't wait forever in case of
an error on request to create sub channels.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
It is cleaner to use RCU protected pointer (nvdev_ctx->nvdev)
to indicate device is in removed state, rather than having a separate
boolean flag. By using the pointer the context can be checked
by static checkers and dynamic lockdep.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The netvsc driver has an internal structure (netvsc_device) which
is created when device is opened and released when device is closed.
And also opened/released when MTU or number of channels change.
Since this is referenced in the receive and transmit path, it is
safer to use RCU to protect/prevent use after free problems.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The default number of maximum channels should be limited to the
number of cpus available on the numa node of the primary channel.
This also makes sure maximum channels <= num_online_cpus
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If device is not up, then changing MTU (or number of channels)
should not re-enable the device.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When using NAPI, the single stream performance declined signifcantly
because the poll routine was updating host after every burst
of packets. This excess signalling caused host throttling.
This fix restores the old behavior. Host is only signalled
after the ring has been emptied.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add some short description of how callback's and NAPI interoperate.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Change the argument to channel callback from the channel pointer
to the internal data structure containing per-channel info.
This avoids any possible races when callback happens during
initialization and makes IRQ code simpler.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When device is being setup on boot, there is a small race where
network device callback is registered, but the netvsc_device pointer
is not set yet. This can cause a NULL ptr dereference if packet
arrives during this window.
Fixes: 46b4f7f5d1 ("netvsc: eliminate per-device outstanding send counter")
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Conflicts:
drivers/net/ethernet/broadcom/genet/bcmgenet.c
net/core/sock.c
Conflicts were overlapping changes in bcmgenet and the
lockdep handling of sockets.
Signed-off-by: David S. Miller <davem@davemloft.net>
The code in netvsc_device_remove was incorrectly calling napi_disable
repeatedly on the same element. This would cause attempts
to remove netvsc module to hang.
Fixes: 2506b1dc4bbe ("netvsc: implement NAPI")
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Since rndis_halt_device waits until all outstanding sends and
receives are completed. Netvsc device needs to still schedule
NAPI to see those completions.
Fixes: 2506b1dc4bbe ("netvsc: implement NAPI")
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Move the send indirection table from the inner device (netvsc)
to the network device context.
It is possible that netvsc_device is not present (remove in progress).
This solves potential use after free issues when packet is being
created during MTU change, shutdown, or queue count changes.
Fixes: d8e18ee0fa ("netvsc: enhance transmit select_queue")
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The ethtool api {get|set}_settings is deprecated.
We move this driver to new api {get|set}_link_ksettings.
As I don't have the hardware, I'd be very pleased if
someone may test this patch.
Signed-off-by: Philippe Reynes <tremyfr@gmail.com>
Tested-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use NAPI (softirq), to handle receive packets and send completions.
Previously this was handled by tasklet.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This is mostly just a refactoring of previous functions
(get_pkt_next_raw, put_pkt_raw and commit_rd_index) to make it easier
to use for other drivers and NAPI.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
There are two variables named packet in the same function. One is the
metadata descriptor from host (vmpacket_descriptor) and the other is
the control block in the skb used to hold metadata from send.
Change name to avoid possible confusion and bugs.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
