[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Xen-devel] [2/11] [NET] back: Make use of the simplicity of tasklets in net_rx_action


[NET] back: Make use of the simplicity of tasklets in net_rx_action

Tasklets have the property that each one is running on only one CPU at
any time.  This means that you don't have to worry about the tasklet
racing against itself.  Therefore any resources used by just a single
tasklet does not need to be guarded by locks.

Since net_rx_action is the only user of alloc_mfn, we can remove the
mfn_lock that guard it.

The notify_list array is huge by Linux standards so placing it on the
stack is unsafe.  Since net_rx_action is not re-entrant, we can simply
make it static.

Signed-off-by: Herbert Xu <herbert@xxxxxxxxxxxxxxxxxxx>

Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <herbert@xxxxxxxxxxxxxxxxxxx>
Home Page: http://gondor.apana.org.au/~herbert/
PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt
diff -r dd443604e775 -r e90f22eee9a1 
--- a/linux-2.6-xen-sparse/drivers/xen/netback/netback.c        Fri Jul 07 
23:35:42 2006 +1000
+++ b/linux-2.6-xen-sparse/drivers/xen/netback/netback.c        Fri Jul 07 
23:35:46 2006 +1000
@@ -99,24 +99,21 @@ static spinlock_t net_schedule_list_lock
 #define MAX_MFN_ALLOC 64
 static unsigned long mfn_list[MAX_MFN_ALLOC];
 static unsigned int alloc_index = 0;
-static DEFINE_SPINLOCK(mfn_lock);
 static unsigned long alloc_mfn(void)
-       unsigned long mfn = 0, flags;
+       unsigned long mfn = 0;
        struct xen_memory_reservation reservation = {
                .nr_extents   = MAX_MFN_ALLOC,
                .extent_order = 0,
                .domid        = DOMID_SELF
        set_xen_guest_handle(reservation.extent_start, mfn_list);
-       spin_lock_irqsave(&mfn_lock, flags);
        if ( unlikely(alloc_index == 0) )
                alloc_index = HYPERVISOR_memory_op(
                        XENMEM_increase_reservation, &reservation);
        if ( alloc_index != 0 )
                mfn = mfn_list[--alloc_index];
-       spin_unlock_irqrestore(&mfn_lock, flags);
        return mfn;
@@ -222,9 +219,13 @@ static void net_rx_action(unsigned long 
        unsigned long vdata, old_mfn, new_mfn;
        struct sk_buff_head rxq;
        struct sk_buff *skb;
-       u16 notify_list[NET_RX_RING_SIZE];
        int notify_nr = 0;
        int ret;
+       /*
+        * Putting hundreds of bytes on the stack is considered rude.
+        * Static works because a tasklet can only be on one CPU at any time.
+        */
+       static u16 notify_list[NET_RX_RING_SIZE];

Xen-devel mailing list



Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.