search for: xen_netbk

..._down(). 3. Change the usage of kthread_should_stop(). 4. Use __get_free_pages() to replace kzalloc(). 5. Modify the changes to netif_be_dbg(). 6. Use MODPARM_netback_kthread to determine whether using tasklet or kernel thread. 7. Put small fields in the front, and large arrays in the end of struct xen_netbk. 8. Add more checks in netif_page_release(). Current netback uses one pair of tasklets for Tx/Rx data transaction. Netback tasklet could only run at one CPU at a time, and it is used to serve all the netfronts. Therefore it has become a performance bottle neck. This patch is to use multiple taskle...

...credit = min(max_credit, max_burst); } +static void tx_add_packets(struct xenvif *vif) +{ + vif->remaining_packets = vif->credit_packets; +} + static void tx_credit_callback(unsigned long data) { struct xenvif *vif = (struct xenvif *)data; tx_add_credit(vif); + tx_add_packets(vif); xen_netbk_check_rx_xenvif(vif); } @@ -1419,6 +1425,38 @@ static bool tx_credit_exceeded(struct xenvif *vif, unsigned size) return false; } +static bool tx_packets_exceeded(struct xenvif *vif) +{ + unsigned long now = jiffies; + unsigned long next_credit = + vif->credit_timeout.expires + + msecs_...

...t xenvif_interrupt(int irq, void *dev_id) static int xenvif_start_xmit(struct sk_buff *skb, struct net_device *dev) { struct xenvif *vif = netdev_priv(dev); + unsigned int ring_slots_required; BUG_ON(skb->dev != dev); if (vif->netbk == NULL) goto drop; + ring_slots_required = xen_netbk_count_skb_slots(vif, skb); + + /* Reserve ring slots for the worst-case number of fragments. */ + vif->rx_req_cons_peek += ring_slots_required; + /* Drop the packet if the target domain has no receive buffers. */ - if (!xenvif_rx_schedulable(vif)) + if (!xenvif_rx_schedulable(vif)) { + vif-&g...

...39;t an insane number of requests on the ring (i.e. more than would fit in the ring). If the ring contains garbage then previously is was possible to loop over this insane number, getting an error each time and therefore not generating any more pending requests and therefore not exiting the loop in xen_netbk_tx_build_gops for an externded period. Also turn various netdev_dbg calls which no precipitate a fatal error into netdev_err, they are rate limited because the device is shutdown afterwards. This fixes at least one known DoS/softlockup of the backend domain. Signed-off-by: Ian Campbell <ian.c...

...ers/net/xen-netback/interface.c b/drivers/net/xen-netback/interface.c index fa4d46d..c9ebe21 100644 --- a/drivers/net/xen-netback/interface.c +++ b/drivers/net/xen-netback/interface.c @@ -60,7 +60,8 @@ static int xenvif_rx_schedulable(struct xenvif *vif) return xenvif_schedulable(vif) && !xen_netbk_rx_ring_full(vif); } -static irqreturn_t xenvif_interrupt(int irq, void *dev_id) +/* Tx interrupt handler used when feature-split-event-channels == 1 */ +static irqreturn_t xenvif_tx_interrupt(int tx_irq, void *dev_id) { struct xenvif *vif = dev_id; @@ -69,12 +70,31 @@ static irqreturn_t xe...

...for required slot in netbk_gop_skb. In netbk_gop_skb, required slots are calculated based on offset and len in page of header data. It is possible that required slots here is larger than the one calculated in earlier netbk_count_requests. This inconsistency directly results in rx_req_cons_peek and xen_netbk_rx_ring_full judgement are wrong. Then it comes to situation the ring is actually full, but netback thinks it is not and continues to create responses. This results in response overlaps request in the ring, then grantcopy gets wrong grant reference and throws out error, for example "(XEN) gra...

Tune xen_netbk_count_requests to not touch working array beyond limit, so that we can make working array size constant. Signed-off-by: Wei Liu <wei.liu2@citrix.com> --- drivers/net/xen-netback/netback.c | 26 +++++++++++++++++++++----- 1 file changed, 21 insertions(+), 5 deletions(-) diff --git a/drive...