xen: Avoid allocations causing swap activity on the resume path
Avoid allocations causing swap activity on the resume path by
preventing the allocations from doing IO and allowing them
to access the emergency pools.
These paths are used when a frontend device is trying to connect
to its backend driver over Xenbus. These reconnections are triggered
on demand by IO, so by definition there is already IO underway,
and further IO would naturally deadlock. On resume, this path
is triggered when the running system tries to continue using its
devices. If it cannot then the resume will fail; to try to avoid this
we let it dip into the emergency pools.
[ linux-2.6.18-xen changesets e8b49cfbdac, fdb998e79aba ]
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: Jens Axboe <jens.axboe@oracle.com>
diff --git a/drivers/net/xen-netfront.c b/drivers/net/xen-netfront.c
index d26f69b..ef671d1 100644
--- a/drivers/net/xen-netfront.c
+++ b/drivers/net/xen-netfront.c
@@ -1324,7 +1324,7 @@
goto fail;
}
- txs = (struct xen_netif_tx_sring *)get_zeroed_page(GFP_KERNEL);
+ txs = (struct xen_netif_tx_sring *)get_zeroed_page(GFP_NOIO | __GFP_HIGH);
if (!txs) {
err = -ENOMEM;
xenbus_dev_fatal(dev, err, "allocating tx ring page");
@@ -1340,7 +1340,7 @@
}
info->tx_ring_ref = err;
- rxs = (struct xen_netif_rx_sring *)get_zeroed_page(GFP_KERNEL);
+ rxs = (struct xen_netif_rx_sring *)get_zeroed_page(GFP_NOIO | __GFP_HIGH);
if (!rxs) {
err = -ENOMEM;
xenbus_dev_fatal(dev, err, "allocating rx ring page");