xen-netfront: Fix Rx stall during network stress and OOM
authorVineeth Remanan Pillai <vineethp@amazon.com>
Thu, 19 Jan 2017 16:35:39 +0000 (08:35 -0800)
committerGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Wed, 5 Jul 2017 12:37:18 +0000 (14:37 +0200)
commit230fe9c7d8141d846ee9d30fd67204a44fc53943
treeab87ce38cc2ba8efbfe212ef98038559fbad7702
parent8df98ff6c39491360c6bdad55729e237da9b1d94
xen-netfront: Fix Rx stall during network stress and OOM

[ Upstream commit 90c311b0eeead647b708a723dbdde1eda3dcad05 ]

During an OOM scenario, request slots could not be created as skb
allocation fails. So the netback cannot pass in packets and netfront
wrongly assumes that there is no more work to be done and it disables
polling. This causes Rx to stall.

The issue is with the retry logic which schedules the timer if the
created slots are less than NET_RX_SLOTS_MIN. The count of new request
slots to be pushed are calculated as a difference between new req_prod
and rsp_cons which could be more than the actual slots, if there are
unconsumed responses.

The fix is to calculate the count of newly created slots as the
difference between new req_prod and old req_prod.

Signed-off-by: Vineeth Remanan Pillai <vineethp@amazon.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <alexander.levin@verizon.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
drivers/net/xen-netfront.c