This patch adds support of XDP in native mode for hv_netvsc driver, and
transparently sets the XDP program on the associated VF NIC as well.
Setting / unsetting XDP program on synthetic NIC (netvsc) propagates to
VF NIC automatically. Setting / unsetting XDP program on VF NIC directly
is not recommended, also not propagated to synthetic NIC, and may be
overwritten by setting of synthetic NIC.
The Azure/Hyper-V synthetic NIC receive buffer doesn't provide headroom
for XDP. We thought about re-use the RNDIS header space, but it's too
small. So we decided to copy the packets to a page buffer for XDP. And,
most of our VMs on Azure have Accelerated Network (SRIOV) enabled, so
most of the packets run on VF NIC. The synthetic NIC is considered as a
fallback data-path. So the data copy on netvsc won't impact performance
significantly.
XDP program cannot run with LRO (RSC) enabled, so you need to disable LRO
before running XDP:
ethtool -K eth0 lro off
XDP actions not yet supported:
XDP_REDIRECT
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If negotiated NVSP version <= NVSP_PROTOCOL_VERSION_6, the offset may
be wrong (too small) due to a host bug. This can cause missing the
end of the send indirection table, and add multiple zero entries from
leading zeros before the data region. This bug adds extra burden on
channel 0.
So fix the offset by computing it from the data structure sizes. This
will ensure netvsc driver runs normally on unfixed hosts, and future
fixed hosts.
Fixes: 5b54dac856 ("hyperv: Add support for virtual Receive Side Scaling (vRSS)")
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
To reach the data region, the existing code adds offset in struct
nvsp_5_send_indirect_table on the beginning of this struct. But the
offset should be based on the beginning of its container,
struct nvsp_message. This bug causes the first table entry missing,
and adds an extra zero from the zero pad after the data region.
This can put extra burden on the channel 0.
So, correct the offset usage. Also add a boundary check to ensure
not reading beyond data region.
Fixes: 5b54dac856 ("hyperv: Add support for virtual Receive Side Scaling (vRSS)")
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Based on 1 normalized pattern(s):
this program is free software you can redistribute it and or modify
it under the terms and conditions of the gnu general public license
version 2 as published by the free software foundation this program
is distributed in the hope it will be useful but without any
warranty without even the implied warranty of merchantability or
fitness for a particular purpose see the gnu general public license
for more details you should have received a copy of the gnu general
public license along with this program if not see http www gnu org
licenses
extracted by the scancode license scanner the SPDX license identifier
GPL-2.0-only
has been chosen to replace the boilerplate/reference in 228 file(s).
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Allison Randal <allison@lohutok.net>
Reviewed-by: Steve Winslow <swinslow@gmail.com>
Reviewed-by: Richard Fontana <rfontana@redhat.com>
Reviewed-by: Alexios Zavras <alexios.zavras@intel.com>
Cc: linux-spdx@vger.kernel.org
Link: https://lkml.kernel.org/r/20190528171438.107155473@linutronix.de
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When the ring buffer is almost full due to RX completion messages, a
TX packet may reach the "low watermark" and cause the queue stopped.
If the TX completion arrives earlier than queue stopping, the wakeup
may be missed.
This patch moves the check for the last pending packet to cover both
EAGAIN and success cases, so the queue will be reliably waked up when
necessary.
Reported-and-tested-by: Stephan Klein <stephan.klein@wegfinder.at>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Minor comment merge conflict in mlx5.
Staging driver has a fixup due to the skb->xmit_more changes
in 'net-next', but was removed in 'net'.
Signed-off-by: David S. Miller <davem@davemloft.net>
There are two reasons for this.
First, the xmit_more flag conceptually doesn't fit into the skb, as
xmit_more is not a property related to the skb.
Its only a hint to the driver that the stack is about to transmit another
packet immediately.
Second, it was only done this way to not have to pass another argument
to ndo_start_xmit().
We can place xmit_more in the softnet data, next to the device recursion.
The recursion counter is already written to on each transmit. The "more"
indicator is placed right next to it.
Drivers can use the netdev_xmit_more() helper instead of skb->xmit_more
to check the "more packets coming" hint.
skb->xmit_more is retained (but always 0) to not cause build breakage.
This change takes care of the simple s/skb->xmit_more/netdev_xmit_more()/
conversions. Remaining drivers are converted in the next patches.
Suggested-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: Florian Westphal <fw@strlen.de>
Signed-off-by: David S. Miller <davem@davemloft.net>
After queue stopped, the wakeup mechanism may wake it up again
when ring buffer usage is lower than a threshold. This may cause
send path panic on NULL pointer when we stopped all tx queues in
netvsc_detach and start removing the netvsc device.
This patch fix it by adding a tx_disable flag to prevent unwanted
queue wakeup.
Fixes: 7b2ee50c0c ("hv_netvsc: common detach logic")
Reported-by: Mohammed Gamal <mgamal@redhat.com>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Changing mtu, channels, or buffer sizes ops call to netvsc_attach(),
rndis_set_subchannel(), which always reset the hash key to default
value. That will override hash key changed previously. This patch
fixes the problem by save the hash key, then restore it when we re-
add the netvsc device.
Fixes: ff4a441990 ("netvsc: allow get/set of RSS indirection table")
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Reviewed-by: Michael Kelley <mikelley@microsoft.com>
[sl: fix up subject line]
Signed-off-by: Sasha Levin <sashal@kernel.org>
LRO/RSC in the vSwitch is a feature available in Windows Server 2019
hosts and later. It reduces the per packet processing overhead by
coalescing multiple TCP segments when possible. This patch adds netvsc
driver support for this feature.
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Matching network device based on MAC address is problematic
since a non VF network device can be creted with a duplicate MAC
address causing confusion and problems. The VMBus API does provide
a serial number that is a better matching method.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If out ring is full temporarily and receive completion cannot go out,
we may still need to reschedule napi if certain conditions are met.
Otherwise the napi poll might be stopped forever, and cause network
disconnect.
Fixes: 7426b1a518 ("netvsc: optimize receive completions")
Signed-off-by: Stephen Hemminger <stephen@networkplumber.org>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When doing device hotplug the sub channel must be async to avoid
deadlock issues because device is discovered in softirq context.
When doing changes to MTU and number of channels, the setup
must be synchronous to avoid races such as when MTU and device
settings are done in a single ip command.
Reported-by: Thomas Walker <Thomas.Walker@twosigma.com>
Fixes: 8195b1396e ("hv_netvsc: fix deadlock on hotplug")
Fixes: 732e49850c ("netvsc: fix race on sub channel creation")
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This is mostly updates to the usual drivers: ufs, qedf, mpt3sas, lpfc,
xfcp, hisi_sas, cxlflash, qla2xxx. In the absence of Nic, we're also
taking target updates which are mostly minor except for the tcmu
refactor. The only real core change to worry about is the removal of
high page bouncing (in sas, storvsc and iscsi). This has been well
tested and no problems have shown up so far.
Signed-off-by: James E.J. Bottomley <jejb@linux.vnet.ibm.com>
-----BEGIN PGP SIGNATURE-----
iJwEABMIAEQWIQTnYEDbdso9F2cI+arnQslM7pishQUCWx1pbCYcamFtZXMuYm90
dG9tbGV5QGhhbnNlbnBhcnRuZXJzaGlwLmNvbQAKCRDnQslM7pishUucAP42pccS
ziKyiOizuxv9fZ4Q+nXd1A9zhI5tqqpkHjcQegEA40qiZSi3EKGKR8W0UpX7Ntmo
tqrZJGojx9lnrAM2RbQ=
=NMXg
-----END PGP SIGNATURE-----
Merge tag 'scsi-misc' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi
Pull SCSI updates from James Bottomley:
"This is mostly updates to the usual drivers: ufs, qedf, mpt3sas, lpfc,
xfcp, hisi_sas, cxlflash, qla2xxx.
In the absence of Nic, we're also taking target updates which are
mostly minor except for the tcmu refactor.
The only real core change to worry about is the removal of high page
bouncing (in sas, storvsc and iscsi). This has been well tested and no
problems have shown up so far"
* tag 'scsi-misc' of git://git.kernel.org/pub/scm/linux/kernel/git/jejb/scsi: (268 commits)
scsi: lpfc: update driver version to 12.0.0.4
scsi: lpfc: Fix port initialization failure.
scsi: lpfc: Fix 16gb hbas failing cq create.
scsi: lpfc: Fix crash in blk_mq layer when executing modprobe -r lpfc
scsi: lpfc: correct oversubscription of nvme io requests for an adapter
scsi: lpfc: Fix MDS diagnostics failure (Rx < Tx)
scsi: hisi_sas: Mark PHY as in reset for nexus reset
scsi: hisi_sas: Fix return value when get_free_slot() failed
scsi: hisi_sas: Terminate STP reject quickly for v2 hw
scsi: hisi_sas: Add v2 hw force PHY function for internal ATA command
scsi: hisi_sas: Include TMF elements in struct hisi_sas_slot
scsi: hisi_sas: Try wait commands before before controller reset
scsi: hisi_sas: Init disks after controller reset
scsi: hisi_sas: Create a scsi_host_template per HW module
scsi: hisi_sas: Reset disks when discovered
scsi: hisi_sas: Add LED feature for v3 hw
scsi: hisi_sas: Change common allocation mode of device id
scsi: hisi_sas: change slot index allocation mode
scsi: hisi_sas: Introduce hisi_sas_phy_set_linkrate()
scsi: hisi_sas: fix a typo in hisi_sas_task_prep()
...
The calls up from the napi poll reading the receive ring had many
places where an argument was being recreated. I.e the caller already
had the value and wasn't passing it, then the callee would use
known relationship to determine the same value. Simpler and faster
to just pass arguments needed.
Also, add const in a couple places where message is being only read.
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds the NetVSP v6 and 6.1 message structures, and includes
these versions into NetVSC/NetVSP version negotiation process.
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
In Vmbus, we have defined a function to calculate available ring buffer
percentage to write.
Use that function and remove netvsc's private version.
[mkp: typo]
Signed-off-by: Long Li <longli@microsoft.com>
Acked-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
The callers to netvsc_revoke_*_buf() and netvsc_teardown_*_gpadl()
already have their net_device instances. Pass them as a paramaeter to
the function instead of obtaining them from netvsc_device struct
everytime
Signed-off-by: Mohammed Gamal <mgamal@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Prior to commit 0cf737808a ("hv_netvsc: netvsc_teardown_gpadl() split")
the call sequence in netvsc_device_remove() was as follows (as
implemented in netvsc_destroy_buf()):
1- Send NVSP_MSG1_TYPE_REVOKE_RECV_BUF message
2- Teardown receive buffer GPADL
3- Send NVSP_MSG1_TYPE_REVOKE_SEND_BUF message
4- Teardown send buffer GPADL
5- Close vmbus
This didn't work for WS2016 hosts. Commit 0cf737808a
("hv_netvsc: netvsc_teardown_gpadl() split") rearranged the
teardown sequence as follows:
1- Send NVSP_MSG1_TYPE_REVOKE_RECV_BUF message
2- Send NVSP_MSG1_TYPE_REVOKE_SEND_BUF message
3- Close vmbus
4- Teardown receive buffer GPADL
5- Teardown send buffer GPADL
That worked well for WS2016 hosts, but it prevented guests on older hosts from
shutting down after changing network settings. Commit 0ef58b0a05
("hv_netvsc: change GPAD teardown order on older versions") ensured the
following message sequence for older hosts
1- Send NVSP_MSG1_TYPE_REVOKE_RECV_BUF message
2- Send NVSP_MSG1_TYPE_REVOKE_SEND_BUF message
3- Teardown receive buffer GPADL
4- Teardown send buffer GPADL
5- Close vmbus
However, with this sequence calling `ip link set eth0 mtu 1000` hangs and the
process becomes uninterruptible. On futher analysis it turns out that on tearing
down the receive buffer GPADL the kernel is waiting indefinitely
in vmbus_teardown_gpadl() for a completion to be signaled.
Here is a snippet of where this occurs:
int vmbus_teardown_gpadl(struct vmbus_channel *channel, u32 gpadl_handle)
{
struct vmbus_channel_gpadl_teardown *msg;
struct vmbus_channel_msginfo *info;
unsigned long flags;
int ret;
info = kmalloc(sizeof(*info) +
sizeof(struct vmbus_channel_gpadl_teardown), GFP_KERNEL);
if (!info)
return -ENOMEM;
init_completion(&info->waitevent);
info->waiting_channel = channel;
[....]
ret = vmbus_post_msg(msg, sizeof(struct vmbus_channel_gpadl_teardown),
true);
if (ret)
goto post_msg_err;
wait_for_completion(&info->waitevent);
[....]
}
The completion is signaled from vmbus_ongpadl_torndown(), which gets called when
the corresponding message is received from the host, which apparently never happens
in that case.
This patch works around the issue by restoring the first mentioned message sequence
for older hosts
Fixes: 0ef58b0a05 ("hv_netvsc: change GPAD teardown order on older versions")
Signed-off-by: Mohammed Gamal <mgamal@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Split each of the functions into two for each of send/recv buffers.
This will be needed in order to implement a fine-grained messaging
sequence to the host so that we accommodate the requirements of
different Windows versions
Fixes: 0ef58b0a05 ("hv_netvsc: change GPAD teardown order on older versions")
Signed-off-by: Mohammed Gamal <mgamal@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
When changing network interface settings, Windows guests
older than WS2016 can no longer shutdown. This was addressed
by commit 0ef58b0a05 ("hv_netvsc: change GPAD teardown order
on older versions"), however the issue also occurs on WS2012
guests that share NVSP protocol versions with WS2016 guests.
Hence we use Windows version directly to differentiate them.
Fixes: 0ef58b0a05 ("hv_netvsc: change GPAD teardown order on older versions")
Signed-off-by: Mohammed Gamal <mgamal@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This patch adds range checking for rx packet offset and length.
It may only happen if there is a host side bug.
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
As defined in hyperv_net.h, the NVSP_STAT_SUCCESS is one not zero.
Some functions returns 0 when it actually means NVSP_STAT_SUCCESS.
This patch fixes them.
In netvsc_receive(), it puts the last RNDIS packet's receive status
for all packets in a vmxferpage which may contain multiple RNDIS
packets.
This patch puts NVSP_STAT_FAIL in the receive completion if one of
the packets in a vmxferpage fails.
Signed-off-by: Haiyang Zhang <haiyangz@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Fun set of conflict resolutions here...
For the mac80211 stuff, these were fortunately just parallel
adds. Trivially resolved.
In drivers/net/phy/phy.c we had a bug fix in 'net' that moved the
function phy_disable_interrupts() earlier in the file, whilst in
'net-next' the phy_error() call from this function was removed.
In net/ipv4/xfrm4_policy.c, David Ahern's changes to remove the
'rt_table_id' member of rtable collided with a bug fix in 'net' that
added a new struct member "rt_mtu_locked" which needs to be copied
over here.
The mlxsw driver conflict consisted of net-next separating
the span code and definitions into separate files, whilst
a 'net' bug fix made some changes to that moved code.
The mlx5 infiniband conflict resolution was quite non-trivial,
the RDMA tree's merge commit was used as a guide here, and
here are their notes:
====================
Due to bug fixes found by the syzkaller bot and taken into the for-rc
branch after development for the 4.17 merge window had already started
being taken into the for-next branch, there were fairly non-trivial
merge issues that would need to be resolved between the for-rc branch
and the for-next branch. This merge resolves those conflicts and
provides a unified base upon which ongoing development for 4.17 can
be based.
Conflicts:
drivers/infiniband/hw/mlx5/main.c - Commit 42cea83f95
(IB/mlx5: Fix cleanup order on unload) added to for-rc and
commit b5ca15ad7e (IB/mlx5: Add proper representors support)
add as part of the devel cycle both needed to modify the
init/de-init functions used by mlx5. To support the new
representors, the new functions added by the cleanup patch
needed to be made non-static, and the init/de-init list
added by the representors patch needed to be modified to
match the init/de-init list changes made by the cleanup
patch.
Updates:
drivers/infiniband/hw/mlx5/mlx5_ib.h - Update function
prototypes added by representors patch to reflect new function
names as changed by cleanup patch
drivers/infiniband/hw/mlx5/ib_rep.c - Update init/de-init
stage list to match new order from cleanup patch
====================
Signed-off-by: David S. Miller <davem@davemloft.net>
Make common function for detaching internals of device
during changes to MTU and RSS. Make sure no more packets
are transmitted and all packets have been received before
doing device teardown.
Change the wait logic to be common and use usleep_range().
Changes transmit enabling logic so that transmit queues are disabled
during the period when lower device is being changed. And enabled
only after sub channels are setup. This avoids issue where it could
be that a packet was being sent while subchannel was not initialized.
Fixes: 8195b1396e ("hv_netvsc: fix deadlock on hotplug")
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
On older versions of Windows, the host ignores messages after
vmbus channel is closed.
Workaround this by doing what Windows does and send the teardown
before close on older versions of NVSP protocol.
Reported-by: Mohammed Gamal <mgamal@redhat.com>
Fixes: 0cf737808a ("hv_netvsc: netvsc_teardown_gpadl() split")
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
The receive processing may continue to happen while the
internal network device state is in RCU grace period.
The internal RNDIS structure is associated with the
internal netvsc_device structure; both have the same
RCU lifetime.
Defer freeing all associated parts until after grace
period.
Fixes: 0cf737808a ("hv_netvsc: netvsc_teardown_gpadl() split")
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This makes sure that no CPU is still process packets when
the channel is closed.
Fixes: 76bb5db5c7 ("netvsc: fix use after free on module removal")
Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
This adds tracepoints to the driver which has proved useful in
debugging startup and shutdown race conditions.
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>
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>
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>
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 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>
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 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>
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>
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>
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>
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>
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>
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>
