Re: [PATCH 1/2] tuntap: reduce memory using of queues
From: "Michael S. Tsirkin" <mst@redhat.com>
Date: 2013-01-23 15:12:31
Also in:
lkml
On Wed, Jan 23, 2013 at 09:59:12PM +0800, Jason Wang wrote:
A MAX_TAP_QUEUES(1024) queues of tuntap device is always allocated unconditionally even userspace only requires a single queue device. This is unnecessary and will lead a very high order of page allocation when has a high possibility to fail. Solving this by creating a one queue net device when userspace only use one queue and also reduce MAX_TAP_QUEUES to DEFAULT_MAX_NUM_RSS_QUEUES which can guarantee the success of the allocation. Reported-by: Dirk Hohndel <redacted> Cc: Eric Dumazet <redacted> Cc: David Woodhouse <dwmw2@infradead.org> Cc: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Jason Wang <redacted>
Note: this is a 3.8 patch, it fixes a regression. Acked-by: Michael S. Tsirkin <mst@redhat.com>
quoted hunk ↗ jump to hunk
--- drivers/net/tun.c | 15 ++++++++------- 1 files changed, 8 insertions(+), 7 deletions(-)diff --git a/drivers/net/tun.c b/drivers/net/tun.c index c81680d..8939d21 100644 --- a/drivers/net/tun.c +++ b/drivers/net/tun.c@@ -109,11 +109,10 @@ struct tap_filter { unsigned char addr[FLT_EXACT_COUNT][ETH_ALEN]; }; -/* 1024 is probably a high enough limit: modern hypervisors seem to support on - * the order of 100-200 CPUs so this leaves us some breathing space if we want - * to match a queue per guest CPU. - */ -#define MAX_TAP_QUEUES 1024 +/* DEFAULT_MAX_NUM_RSS_QUEUES were choosed to let the rx/tx queues allocated for + * the netdevice to be fit in one page. So we can make sure the success of + * memory allocation. TODO: increase the limit. */ +#define MAX_TAP_QUEUES DEFAULT_MAX_NUM_RSS_QUEUES #define TUN_FLOW_EXPIRE (3 * HZ)@@ -1583,6 +1582,8 @@ static int tun_set_iff(struct net *net, struct file *file, struct ifreq *ifr) else { char *name; unsigned long flags = 0; + int queues = ifr->ifr_flags & IFF_MULTI_QUEUE ? + MAX_TAP_QUEUES : 1; if (!ns_capable(net->user_ns, CAP_NET_ADMIN)) return -EPERM;@@ -1606,8 +1607,8 @@ static int tun_set_iff(struct net *net, struct file *file, struct ifreq *ifr) name = ifr->ifr_name; dev = alloc_netdev_mqs(sizeof(struct tun_struct), name, - tun_setup, - MAX_TAP_QUEUES, MAX_TAP_QUEUES); + tun_setup, queues, queues); + if (!dev) return -ENOMEM;-- 1.7.1