Re: [PATCH v2 1/7] vhost: refactor rte_vhost_dequeue_burst

[Yuanhan Liu <yuanhan.liu@linux.intel.com>]

On Mon, Mar 07, 2016 at 02:59:55AM +0000, Xie, Huawei wrote:
> On 3/7/2016 10:47 AM, Yuanhan Liu wrote:
> > On Mon, Mar 07, 2016 at 02:32:46AM +0000, Xie, Huawei wrote:
> >> On 3/4/2016 10:15 AM, Yuanhan Liu wrote:
> >>> On Thu, Mar 03, 2016 at 04:30:42PM +0000, Xie, Huawei wrote:
> >>>> On 2/18/2016 9:48 PM, Yuanhan Liu wrote:
> >>>> We could always allocate the head mbuf before the loop, then we save the
> >>>> following branch and make the code more streamlined.
> >>>> It reminds me that this change prevents the possibility of mbuf bulk
> >>>> allocation, one solution is we pass the head mbuf from an additional
> >>>> parameter.
> >>> Yep, that's also something I have thought of.
> >>>
> >>>> Btw, put unlikely before the check of mbuf_avail and checks elsewhere.
> >>> I don't think so. It would benifit for the small packets. What if,
> >>> however, when TSO or jumbo frame is enabled that we have big packets?
> >> Prefer to favor the path that packet could fit in one mbuf.
> > Hmmm, why? While I know that TSO and mergeable buf is disable by default
> > in our vhost example vhost-switch, they are enabled by default in real
> > life.
> 
> mergable is only meaningful in RX path.
> If you mean big packets in TX path,

Sorry, and yes, I meant that.

> i personally favor the path that
> packet fits in one mbuf.

Sorry, that will not convince me.

	--yliu