[Xen-devel] Re: [7/11] [NET] back: Added tx queue

[Keir Fraser <Keir.Fraser@xxxxxxxxxxxx>]

On 27 Jul 2006, at 14:23, Herbert Xu wrote:

> > Also, what does NETIF_F_LLTX have to do with whether we have a queue 
> > or
> > not -- couldn't we set that all the time? And you set queue length to 
> > 1
> Sorry, should've add a comment about that.  NETIF_F_LLTX is just an 
> unused
> bit from the features set to indicate the fact that queueing is 
> supported.
> I suppose I could add a new flag too.
That's pretty gross. NETIF_F_LLTX is used isn't it? To indicate whether 
an interface does its own locking in start_xmit()?
Why not have a private feature flag in an appropriate netback-private 
structure?
> > in the other case -- what's wrong with zero? Seems a saner value when
> > there is no queue: or do we need to commit to having *some* queue at
> > register_netdevice() time, and there's no way to go back from that by
> > the time we find out if netfront supports rx refill notifications (so
> > at that point we cannot change queue len to zero)?
> Exactly.  Once we've committed to having a queue, setting the queue
> length to zero will cause all packets to be dropped.  I even added
> a comment about that :)
Ok, sounds reasonable then.

But I would change queue length to 100 in your next spin of this patch. 
1000 is scary big.
 -- Keir


_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-devel