linux-toradex.git/drivers/net/ethernet, branch v3.2.68

enic: fix rx skb checksum

2015-02-20T00:49:41+00:00

[ Upstream commit 17e96834fd35997ca7cdfbf15413bcd5a36ad448 ]

Hardware always provides compliment of IP pseudo checksum. Stack expects
whole packet checksum without pseudo checksum if CHECKSUM_COMPLETE is set.

This causes checksum error in nf & ovs.

kernel: qg-19546f09-f2: hw csum failure
kernel: CPU: 9 PID: 0 Comm: swapper/9 Tainted: GF          O--------------   3.10.0-123.8.1.el7.x86_64 #1
kernel: Hardware name: Cisco Systems Inc UCSB-B200-M3/UCSB-B200-M3, BIOS B200M3.2.2.3.0.080820141339 08/08/2014
kernel: ffff881218f40000 df68243feb35e3a8 ffff881237a43ab8 ffffffff815e237b
kernel: ffff881237a43ad0 ffffffff814cd4ca ffff8829ec71eb00 ffff881237a43af0
kernel: ffffffff814c6232 0000000000000286 ffff8829ec71eb00 ffff881237a43b00
kernel: Call Trace:
kernel:   [] dump_stack+0x19/0x1b
kernel: [] netdev_rx_csum_fault+0x3a/0x40
kernel: [] __skb_checksum_complete_head+0x62/0x70
kernel: [] __skb_checksum_complete+0x11/0x20
kernel: [] nf_ip_checksum+0xcc/0x100
kernel: [] icmp_error+0x1f7/0x35c [nf_conntrack_ipv4]
kernel: [] ? netif_rx+0xb9/0x1d0
kernel: [] ? internal_dev_recv+0xdb/0x130 [openvswitch]
kernel: [] nf_conntrack_in+0xf0/0xa80 [nf_conntrack]
kernel: [] ? inet_del_offload+0x40/0x40
kernel: [] ipv4_conntrack_in+0x22/0x30 [nf_conntrack_ipv4]
kernel: [] nf_iterate+0xaa/0xc0
kernel: [] ? inet_del_offload+0x40/0x40
kernel: [] nf_hook_slow+0x84/0x140
kernel: [] ? inet_del_offload+0x40/0x40
kernel: [] ip_rcv+0x344/0x380

Hardware verifies IP & tcp/udp header checksum but does not provide payload
checksum, use CHECKSUM_UNNECESSARY. Set it only if its valid IP tcp/udp packet.

Cc: Jiri Benc 
Cc: Stefan Assmann 
Reported-by: Sunil Choudhary 
Signed-off-by: Govindarajulu Varadarajan <_govind@gmx.com>
Reviewed-by: Jiri Benc 
Signed-off-by: David S. Miller 
Signed-off-by: Ben Hutchings

tg3: tg3_disable_ints using uninitialized mailbox value to disable interrupts

2015-02-20T00:49:41+00:00

[ Upstream commit 05b0aa579397b734f127af58e401a30784a1e315 ]

During driver load in tg3_init_one, if the driver detects DMA activity before
intializing the chip tg3_halt is called. As part of tg3_halt interrupts are
disabled using routine tg3_disable_ints. This routine was using mailbox value
which was not initialized (default value is 0). As a result driver was writing
0x00000001 to pci config space register 0, which is the vendor id / device id.

This driver bug was exposed because of the commit a7877b17a667 (PCI: Check only
the Vendor ID to identify Configuration Request Retry). Also this issue is only
seen in older generation chipsets like 5722 because config space write to offset
0 from driver is possible. The newer generation chips ignore writes to offset 0.
Also without commit a7877b17a667, for these older chips when a GRC reset is
issued the Bootcode would reprogram the vendor id/device id, which is the reason
this bug was masked earlier.

Fixed by initializing the interrupt mailbox registers before calling tg3_halt.

Please queue for -stable.

Reported-by: Nils Holland 
Reported-by: Marcelo Ricardo Leitner 
Signed-off-by: Prashant Sreedharan 
Signed-off-by: Michael Chan 
Signed-off-by: David S. Miller 
Signed-off-by: Ben Hutchings

sunvnet: clean up objects created in vnet_new() on vnet_exit()

2014-08-06T17:07:39+00:00

[ Upstream commit a4b70a07ed12a71131cab7adce2ce91c71b37060 ]

Nothing cleans up the objects created by
vnet_new(), they are completely leaked.

vnet_exit(), after doing the vio_unregister_driver() to clean
up ports, should call a helper function that iterates over vnet_list
and cleans up those objects. This includes unregister_netdevice()
as well as free_netdev().

Signed-off-by: Sowmini Varadhan 
Acked-by: Dave Kleikamp 
Reviewed-by: Karl Volz 
Signed-off-by: David S. Miller 
Signed-off-by: Ben Hutchings

igb: do a reset on SR-IOV re-init if device is down

2014-08-06T17:07:37+00:00

commit 76252723e88681628a3dbb9c09c963e095476f73 upstream.

To properly re-initialize SR-IOV it is necessary to reset the device
even if it is already down. Not doing this may result in Tx unit hangs.

Signed-off-by: Stefan Assmann 
Tested-by: Aaron Brown 
Signed-off-by: Jeff Kirsher 
Signed-off-by: David S. Miller 
Signed-off-by: Ben Hutchings

net/mlx4_core: Preserve pci_dev_data after __mlx4_remove_one()

2014-07-11T12:33:58+00:00

[ Upstream commit befdf8978accecac2e0739e6b5075afc62db37fe ]

This patch wrap up a helper function __mlx4_remove_one() which does the tear
down function but preserve the drv_data. Functions like
mlx4_pci_err_detected() and mlx4_restart_one() will call this one with out
releasing drvdata.

Signed-off-by: Wei Yang 
Signed-off-by: Ben Hutchings

mlx4_core: Stash PCI ID driver_data in mlx4_priv structure

2014-07-11T12:33:58+00:00

[ No upstream commit, this is a cherry picked backport enabler. ]

That way we can check flags later on, when we've finished with the
pci_device_id structure.

This is a backport.

Signed-off-by: Roland Dreier 
Signed-off-by: Ben Hutchings

tg3: update rx_jumbo_pending ring param only when jumbo frames are enabled

2014-06-09T12:28:57+00:00

commit ba67b510035141bd89b40bf65efa0a79834311ca upstream.

The patch fixes a problem with dropped jumbo frames after usage of
'ethtool -G ... rx'.

Scenario:
1. ip link set eth0 up
2. ethtool -G eth0 rx N # <- This zeroes rx-jumbo
3. ip link set mtu 9000 dev eth0

The ethtool command set rx_jumbo_pending to zero so any received jumbo
packets are dropped and you need to use 'ethtool -G eth0 rx-jumbo N'
to workaround the issue.
The patch changes the logic so rx_jumbo_pending value is changed only if
jumbo frames are enabled (MTU > 1500).

Signed-off-by: Ivan Vecera 
Acked-by: Michael Chan 
Signed-off-by: David S. Miller 
Signed-off-by: Ben Hutchings

mlx4_en: don't use napi_synchronize inside mlx4_en_netpoll

2014-05-18T13:58:03+00:00

commit c98235cb8584a72e95786e17d695a8e5fafcd766 upstream.

The mlx4 driver is triggering schedules while atomic inside
mlx4_en_netpoll:

	spin_lock_irqsave(&cq->lock, flags);
	napi_synchronize(&cq->napi);
		^^^^^ msleep here
	mlx4_en_process_rx_cq(dev, cq, 0);
	spin_unlock_irqrestore(&cq->lock, flags);

This was part of a patch by Alexander Guller from Mellanox in 2011,
but it still isn't upstream.

Signed-off-by: Chris Mason 
Acked-By: Amir Vadai 
Signed-off-by: David S. Miller 
[bwh: Backported to 3.2: adjust context]
Signed-off-by: Ben Hutchings

tg3: Don't check undefined error bits in RXBD

2014-04-01T23:59:02+00:00

[ Upstream commit d7b95315cc7f441418845a165ee56df723941487 ]

Redefine the RXD_ERR_MASK to include only relevant error bits. This fixes
a customer reported issue of randomly dropping packets on the 5719.

Signed-off-by: Michael Chan 
Signed-off-by: David S. Miller 
Signed-off-by: Ben Hutchings

tg3: Fix deadlock in tg3_change_mtu()

2014-04-01T23:59:00+00:00

[ Upstream commit c6993dfd7db9b0c6b7ca7503a56fda9236a4710f ]

Quoting David Vrabel -
"5780 cards cannot have jumbo frames and TSO enabled together.  When
jumbo frames are enabled by setting the MTU, the TSO feature must be
cleared.  This is done indirectly by calling netdev_update_features()
which will call tg3_fix_features() to actually clear the flags.

netdev_update_features() will also trigger a new netlink message for the
feature change event which will result in a call to tg3_get_stats64()
which deadlocks on the tg3 lock."

tg3_set_mtu() does not need to be under the tg3 lock since converting
the flags to use set_bit(). Move it out to after tg3_netif_stop().

Reported-by: David Vrabel 
Tested-by: David Vrabel 
Signed-off-by: Michael Chan 
Signed-off-by: Nithin Nayak Sujir 
Signed-off-by: David S. Miller 
Signed-off-by: Ben Hutchings