[openib-general] [PATCH/RFC] IB/ipoib: add selective tx signaling

[Michael S. Tsirkin]

> Subject: Re: [PATCH/RFC] IB/ipoib: add selective tx signaling
> 
>  > netif_queue_stopped is an atomic, I think that would be sufficient, no?
> 
> netif_queue_stopped does test_bit(), which is an implicit memory
> barrier.  But I think it's pretty subtle -- you have to look at every
> place tx_head and tx_tail are updated and decide if there needs to be
> a corresponding barrier.

I certainly agree it is subtle.

> For example, in:
> 
> +	if (netif_queue_stopped(dev) &&
> +	    priv->tx_head - priv->tx_tail < ipoib_sendq_size) {
> 
> there is a barrier in netif_queue_stopped() but no control in what
> order tx_head and tx_tail are read...

In this case, I think its fine since tx_tail is only updated in the
completion handler.

-- 
MST