linux.git/drivers/net, branch v3.11-rc2

macvtap: do not zerocopy if iov needs more pages than MAX_SKB_FRAGS

2013-07-18T20:04:25+00:00

We try to linearize part of the skb when the number of iov is greater than
MAX_SKB_FRAGS. This is not enough since each single vector may occupy more than
one pages, so zerocopy_sg_fromiovec() may still fail and may break the guest
network.

Solve this problem by calculate the pages needed for iov before trying to do
zerocopy and switch to use copy instead of zerocopy if it needs more than
MAX_SKB_FRAGS.

This is done through introducing a new helper to count the pages for iov, and
call uarg->callback() manually when switching from zerocopy to copy to notify
vhost.

We can do further optimization on top.

This bug were introduced from b92946e2919134ebe2a4083e4302236295ea2a73
(macvtap: zerocopy: validate vectors before building skb).

Cc: Michael S. Tsirkin 
Signed-off-by: Jason Wang 
Signed-off-by: David S. Miller

tuntap: do not zerocopy if iov needs more pages than MAX_SKB_FRAGS

2013-07-18T20:04:25+00:00

We try to linearize part of the skb when the number of iov is greater than
MAX_SKB_FRAGS. This is not enough since each single vector may occupy more than
one pages, so zerocopy_sg_fromiovec() may still fail and may break the guest
network.

Solve this problem by calculate the pages needed for iov before trying to do
zerocopy and switch to use copy instead of zerocopy if it needs more than
MAX_SKB_FRAGS.

This is done through introducing a new helper to count the pages for iov, and
call uarg->callback() manually when switching from zerocopy to copy to notify
vhost.

We can do further optimization on top.

The bug were introduced from commit 0690899b4d4501b3505be069b9a687e68ccbe15b
(tun: experimental zero copy tx support)

Cc: Michael S. Tsirkin 
Signed-off-by: Jason Wang 
Signed-off-by: David S. Miller

xen-netfront: pull on receive skb may need to happen earlier

2013-07-17T19:51:19+00:00

Due to commit 3683243b ("xen-netfront: use __pskb_pull_tail to ensure
linear area is big enough on RX") xennet_fill_frags() may end up
filling MAX_SKB_FRAGS + 1 fragments in a receive skb, and only reduce
the fragment count subsequently via __pskb_pull_tail(). That's a
result of xennet_get_responses() allowing a maximum of one more slot to
be consumed (and intermediately transformed into a fragment) if the
head slot has a size less than or equal to RX_COPY_THRESHOLD.

Hence we need to adjust xennet_fill_frags() to pull earlier if we
reached the maximum fragment count - due to the described behavior of
xennet_get_responses() this guarantees that at least the first fragment
will get completely consumed, and hence the fragment count reduced.

In order to not needlessly call __pskb_pull_tail() twice, make the
original call conditional upon the pull target not having been reached
yet, and defer the newly added one as much as possible (an alternative
would have been to always call the function right before the call to
xennet_fill_frags(), but that would imply more frequent cases of
needing to call it twice).

Signed-off-by: Jan Beulich 
Acked-by: Wei Liu 
Cc: Ian Campbell 
Cc: stable@vger.kernel.org (3.6 onwards)
Acked-by: Ian Campbell 
Signed-off-by: David S. Miller

vxlan: add necessary locking on device removal

2013-07-17T19:51:19+00:00

The socket management is now done in workqueue (outside of RTNL)
and protected by vn->sock_lock. There were two possible bugs, first
the vxlan device was removed from the VNI hash table per socket without
holding lock. And there was a race when device is created and the workqueue
could run after deletion.

Signed-off-by: Stephen Hemminger 
Signed-off-by: David S. Miller

hyperv: Fix the NETIF_F_SG flag setting in netvsc

2013-07-17T06:02:24+00:00

SG mode is not currently supported by netvsc, so remove this flag for now.
Otherwise, it will be unconditionally enabled by commit ec5f0615642
    "Kill link between CSUM and SG features"
Previously, the SG feature is disabled because CSUM is not set here.

Signed-off-by: Haiyang Zhang 
Reviewed-by: K. Y. Srinivasan 
Signed-off-by: David S. Miller

be2net: Fix to avoid hardware workaround when not needed

2013-07-16T20:01:57+00:00

Hardware workaround requesting hardware to skip vlan insertion is necessary
only when umc or qnq is enabled. Enabling this workaround in other scenarios
could cause controller to stall.

Signed-off-by: Sarveshwar Bandi 
Signed-off-by: David S. Miller

macvtap: do not assume 802.1Q when send vlan packets

2013-07-16T20:01:57+00:00

The hard-coded 8021.q proto will break 802.1ad traffic. So switch to use
vlan->proto.

Cc: Basil Gor 
Signed-off-by: Jason Wang 
Signed-off-by: David S. Miller

macvtap: fix the missing ret value of TUNSETQUEUE

2013-07-16T20:01:57+00:00

Commit 441ac0fcaadc76ad09771812382345001dd2b813
(macvtap: Convert to using rtnl lock) forget to return what
macvtap_ioctl_set_queue() returns to its caller. This may break multiqueue API
by always falling through to TUNGETFEATURES.

Cc: Vlad Yasevich 
Signed-off-by: Jason Wang 
Signed-off-by: David S. Miller

mlx5 core: Fix __udivdi3 when compiling for 32 bit arches

2013-07-16T19:36:39+00:00

Cc: Eli Cohen 
Signed-off-by: Tim Gardner 
Acked-by: Randy Dunlap 
Reported-by: Randy Dunlap 
Signed-off-by: David S. Miller

bgmac: add dependency to phylib

2013-07-16T19:34:19+00:00

bgmac is using functions from phylib, add the dependency.

Signed-off-by: Hauke Mehrtens 
Signed-off-by: David S. Miller