xen-netback: fix race between napi_complete() and interrupt handler

When the NAPI budget was not all used, xenvif_poll() would call
napi_complete() /after/ enabling the interrupt.  This resulted in a
race between the napi_complete() and the napi_schedule() in the
interrupt handler.  The use of local_irq_save/restore() avoided by
race iff the handler is running on the same CPU but not if it was
running on a different CPU.

Fix this properly by calling napi_complete() before reenabling
interrupts (in the xenvif_napi_schedule_or_enable_irq() call).

Signed-off-by: David Vrabel <david.vrabel@citrix.com>
Acked-by: Wei Liu <wei.liu2@citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

3 files changed, 6 insertions, 30 deletions

diff --git a/drivers/net/xen-netback/common.h b/drivers/net/xen-netback/common.h
index 630a3fcf65bc..0d4a285cbd7e 100644
--- a/drivers/net/xen-netback/common.h
+++ b/drivers/net/xen-netback/common.h
@@ -226,7 +226,7 @@ int xenvif_map_frontend_rings(struct xenvif *vif,
                               grant_ref_t rx_ring_ref);
 
 /* Check for SKBs from frontend and schedule backend processing */
-void xenvif_check_rx_xenvif(struct xenvif *vif);
+void xenvif_napi_schedule_or_enable_events(struct xenvif *vif);
 
 /* Prevent the device from generating any further traffic. */
 void xenvif_carrier_off(struct xenvif *vif);
diff --git a/drivers/net/xen-netback/interface.c b/drivers/net/xen-netback/interface.c
index ef05c5c49d41..20e9defa1060 100644
--- a/drivers/net/xen-netback/interface.c
+++ b/drivers/net/xen-netback/interface.c
@@ -75,32 +75,8 @@ static int xenvif_poll(struct napi_struct *napi, int budget)
         work_done = xenvif_tx_action(vif, budget);
 
         if (work_done < budget) {
-                int more_to_do = 0;
-                unsigned long flags;
-
-                /* It is necessary to disable IRQ before calling
-                 * RING_HAS_UNCONSUMED_REQUESTS. Otherwise we might
-                 * lose event from the frontend.
-                 *
-                 * Consider:
-                 *   RING_HAS_UNCONSUMED_REQUESTS
-                 *   <frontend generates event to trigger napi_schedule>
-                 *   __napi_complete
-                 *
-                 * This handler is still in scheduled state so the
-                 * event has no effect at all. After __napi_complete
-                 * this handler is descheduled and cannot get
-                 * scheduled again. We lose event in this case and the ring
-                 * will be completely stalled.
-                 */
-
-                local_irq_save(flags);
-
-                RING_FINAL_CHECK_FOR_REQUESTS(&vif->tx, more_to_do);
-                if (!more_to_do)
-                        __napi_complete(napi);
-
-                local_irq_restore(flags);
+                napi_complete(napi);
+                xenvif_napi_schedule_or_enable_events(vif);
         }
 
         return work_done;
@@ -194,7 +170,7 @@ static void xenvif_up(struct xenvif *vif)
         enable_irq(vif->tx_irq);
         if (vif->tx_irq != vif->rx_irq)
                 enable_irq(vif->rx_irq);
-        xenvif_check_rx_xenvif(vif);
+        xenvif_napi_schedule_or_enable_events(vif);
 }
 
 static void xenvif_down(struct xenvif *vif)
diff --git a/drivers/net/xen-netback/netback.c b/drivers/net/xen-netback/netback.c
index 64ab1d141f1c..7367208ee8cd 100644
--- a/drivers/net/xen-netback/netback.c
+++ b/drivers/net/xen-netback/netback.c
@@ -716,7 +716,7 @@ done:
                 notify_remote_via_irq(vif->rx_irq);
 }
 
-void xenvif_check_rx_xenvif(struct xenvif *vif)
+void xenvif_napi_schedule_or_enable_events(struct xenvif *vif)
 {
         int more_to_do;
 
@@ -750,7 +750,7 @@ static void tx_credit_callback(unsigned long data)
 {
         struct xenvif *vif = (struct xenvif *)data;
         tx_add_credit(vif);
-        xenvif_check_rx_xenvif(vif);
+        xenvif_napi_schedule_or_enable_events(vif);
 }
 
 static void xenvif_tx_err(struct xenvif *vif,