macvlan: Disable LRO on lowerdev; warn if it's turned back on
Subsystems (2)
PATCH 1/1 macvlan: Disable LRO on lowerdev; warn if it's turned back on
Large Receive Offload (LRO) is only appropriate for packets that are destined for the host, and should be disabled if received packets may be forwarded.
Further, macvtap_skb_to_vnet_hdr() will BUG() on a packet received with LRO (but not GRO).
---
I'm not really familiar with macvlan so I'm not certain that this should be applied to all modes.
Ben.
drivers/net/macvlan.c | 7 +++++++ 1 files changed, 7 insertions(+), 0 deletions(-)
drivers/net/macvlan.c +7 -0
@@ -168,6 +168,11 @@ static rx_handler_result_t macvlan_handle_frame(struct sk_buff **pskb)unsigned int len =0 ;int ret =NET_RX_DROP ; +if (skb_warn_if_lro (skb)) { +kfree_skb (skb); +return RX_HANDLER_CONSUMED ; + } + port =macvlan_port_get_rcu (skb->dev);if (is_multicast_ether_addr (eth->h_dest)) { skb =ip_check_defrag (skb,IP_DEFRAG_MACVLAN );
@@ -717,6 +722,8 @@ int macvlan_common_newlink(struct net *src_net, struct net_device *dev,memcpy (dev->dev_addr, lowerdev->dev_addr,ETH_ALEN ); } +dev_disable_lro (lowerdev); + port->count +=1 ; err =register_netdevice (dev);if (err <0 )
Signature · 7 lines
1.7.7.6 -- Ben Hutchings, Staff Engineer, Solarflare Not speaking for my employer; that's the marketing department's job. They asked us to note that Solarflare product names are trademarked.
Quoted 8 lines · Ben Hutchings
Large Receive Offload (LRO) is only appropriate for packets that are destined for the host, and should be disabled if received packets may be forwarded.
Further,
macvtap_skb_to_vnet_hdr()willBUG()on a packet received with LRO (but not GRO).Signed-off-by Ben Hutchings <[email protected]>
Quoted 9 lines
---
I'm not really familiar with macvlan so I'm not certain that this should be applied to all modes.Ben.
drivers/net/macvlan.c | 7 +++++++ 1 files changed, 7 insertions(+), 0 deletions(-)
But this patch assumes the macvlan is forwarding traffic to a guest via macvtap. Which is an assumption that may not be true.
It seems more appropriate for the macvtap driver to do these checks after all its the driver that may BUG() with LRO.
Thanks,
John
Quoted 19 lines · John Fastabend
Quoted 8 lines · Ben Hutchings
Large Receive Offload (LRO) is only appropriate for packets that are destined for the host, and should be disabled if received packets may be forwarded.
Further,
macvtap_skb_to_vnet_hdr()willBUG()on a packet received with LRO (but not GRO).Signed-off-by Ben Hutchings <[email protected]>Quoted 9 lines
---
I'm not really familiar with macvlan so I'm not certain that this should be applied to all modes.Ben.
drivers/net/macvlan.c | 7 +++++++ 1 files changed, 7 insertions(+), 0 deletions(-)
Quoted 5 lines
But this patch assumes the macvlan is forwarding traffic to a guest via macvtap. Which is an assumption that may not be true.
It seems more appropriate for the macvtap driver to do these checks after all its the driver that may
BUG()with LRO.
That's what I thought at first, but then I looked through what macvlan was doing and it certainly appears to re-transmit skbs in all modes. That's not valid when gso_size != 0 and gso_type == 0.
Ben.
Signature · 3 lines
Ben Hutchings, Staff Engineer, Solarflare Not speaking for my employer; that's the marketing department's job. They asked us to note that Solarflare product names are trademarked.
Quoted 23 lines · Ben Hutchings
Quoted 22 lines · John Fastabend
Quoted 8 lines · Ben Hutchings
Large Receive Offload (LRO) is only appropriate for packets that are destined for the host, and should be disabled if received packets may be forwarded.
Further,
macvtap_skb_to_vnet_hdr()willBUG()on a packet received with LRO (but not GRO).Signed-off-by Ben Hutchings<[email protected]>Quoted 9 lines
---
I'm not really familiar with macvlan so I'm not certain that this should be applied to all modes.Ben.
drivers/net/macvlan.c | 7 +++++++ 1 files changed, 7 insertions(+), 0 deletions(-)But this patch assumes the macvlan is forwarding traffic to a guest via macvtap. Which is an assumption that may not be true.
Quoted 6 lines
Quoted 2 lines
It seems more appropriate for the macvtap driver to do these checks after all its the driver that may
BUG()with LRO.That's what I thought at first, but then I looked through what macvlan was doing and it certainly appears to re-transmit skbs in all modes. That's not valid when
gso_size!= 0 andgso_type== 0.
You can put an IP on a mac-vlan and receive packets on it like normal Ethernet interfaces. They wouldn't be re-transmitted in that case, would they?
Quoted 3 lines
Ben.
Signature · 2 lines
Ben Greear <[email protected]> Candela Technologies Inc http://www.candelatech.com
Quoted 27 lines · Ben Greear
Quoted 26 lines · Ben Hutchings
Quoted 24 lines · John Fastabend
Quoted 8 lines · Ben Hutchings
Large Receive Offload (LRO) is only appropriate for packets that are destined for the host, and should be disabled if received packets may be forwarded.
Further,
macvtap_skb_to_vnet_hdr()willBUG()on a packet received with LRO (but not GRO).Signed-off-by Ben Hutchings<[email protected]>Quoted 9 lines
---
I'm not really familiar with macvlan so I'm not certain that this should be applied to all modes.Ben.
drivers/net/macvlan.c | 7 +++++++ 1 files changed, 7 insertions(+), 0 deletions(-)But this patch assumes the macvlan is forwarding traffic to a guest via macvtap. Which is an assumption that may not be true.
It seems more appropriate for the macvtap driver to do these checks after all its the driver that may
BUG()with LRO.
Quoted 7 lines
Quoted 3 lines
That's what I thought at first, but then I looked through what macvlan was doing and it certainly appears to re-transmit skbs in all modes. That's not valid when
gso_size!= 0 andgso_type== 0.You can put an IP on a mac-vlan and receive packets on it like normal Ethernet interfaces. They wouldn't be re-transmitted in that case, would they?
Long story short.
With the macvlan driver the normal case is not to turn packets around but to go directly to the nic or to come directly from the nic.
In the case where packets are turned around and packets go from one software interface the design is that the drivers are supposed to do software emulation of hardware features like gro so that we don't have to take a performance hit when it happens.
If we are not doing proper software emulation of features like gro in macvtap that is a problem.
I am fuzzy about all of the details but last I looked we were doing proper software emulation of the features when just macvlan was involved. I also remember that the macvtap driver really wanted gro so that it could work efficiently with emulated hardware.
So I don't know why people are having a problem, but the correct solution is not to give up but to fix the silly software side of the drivers to actually handle things properly.
Disabling LRO on the lowerdev just so we can avoid writing the support in macvtap just sounds sad.
Eric
Quoted 11 lines · Eric W. Biederman
Quoted 9 lines · Ben Greear
Quoted 8 lines · Ben Hutchings
Quoted 7 lines · John Fastabend
Quoted 6 lines · Ben Hutchings
Large Receive Offload (LRO) is only appropriate for packets that are destined for the host, and should be disabled if received packets may be forwarded.
Further,
macvtap_skb_to_vnet_hdr()willBUG()on a packet received with LRO (but not GRO).
[...]
Quoted 10 lines
Long story short.
With the macvlan driver the normal case is not to turn packets around but to go directly to the nic or to come directly from the nic.
In the case where packets are turned around and packets go from one software interface the design is that the drivers are supposed to do software emulation of hardware features like gro so that we don't have to take a performance hit when it happens.
Quoted 7 lines
If we are not doing proper software emulation of features like gro in macvtap that is a problem.
I am fuzzy about all of the details but last I looked we were doing proper software emulation of the features when just macvlan was involved. I also remember that the macvtap driver really wanted gro so that it could work efficiently with emulated hardware.
This doesn't disable GRO; you can still benefit from that.
Quoted 3 lines
So I don't know why people are having a problem, but the correct solution is not to give up but to fix the silly software side of the drivers to actually handle things properly.
This is not a deficiency in macvtap. Any skb received with LRO cannot be re-sent through dev_queue_xmit() or dev_hard_start_xmit().
Quoted 2 lines
Disabling LRO on the lowerdev just so we can avoid writing the support in macvtap just sounds sad.
The regular bridge driver and the IPv4 and IPv6 protocols already disable LRO when enabling forwarding on a device. This is 'normal'.
Now we could decide that the possible loss of some header information through LRO is acceptable and we should just require drivers using LRO to label their skbs with the proper gso_type for re-sending. Indeed, although GRO preserves packet boundaries they are actually lost in re-transmission as GSO/TSO re-segments at regular intervals. So that's already happening to an extent.
Ben.
Signature · 3 lines
Ben Hutchings, Staff Engineer, Solarflare Not speaking for my employer; that's the marketing department's job. They asked us to note that Solarflare product names are trademarked.