On Mon, Mar 07, 2016 at 02:59:55AM +, Xie, Huawei wrote:
> On 3/7/2016 10:47 AM, Yuanhan Liu wrote:
> > On Mon, Mar 07, 2016 at 02:32:46AM +, Xie, Huawei wrote:
> >> On 3/4/2016 10:15 AM, Yuanhan Liu wrote:
> >>> On Thu, Mar 03, 2016 at 04:30:42PM +, Xie, Huawei wrote:
> On
On Mon, Mar 07, 2016 at 02:32:46AM +, Xie, Huawei wrote:
> On 3/4/2016 10:15 AM, Yuanhan Liu wrote:
> > On Thu, Mar 03, 2016 at 04:30:42PM +, Xie, Huawei wrote:
> >> On 2/18/2016 9:48 PM, Yuanhan Liu wrote:
> >>> + mbuf_avail = 0;
> >>> + mbuf_offset = 0;
> >>> + while (desc_avail ||
On Mon, Mar 07, 2016 at 02:19:54AM +, Xie, Huawei wrote:
> On 3/4/2016 10:19 AM, Yuanhan Liu wrote:
> > On Thu, Mar 03, 2016 at 04:21:19PM +, Xie, Huawei wrote:
> >> On 2/18/2016 9:48 PM, Yuanhan Liu wrote:
> >>> The current rte_vhost_dequeue_burst() implementation is a bit messy
> >>> and
On 2/18/2016 9:48 PM, Yuanhan Liu wrote:
> + mbuf_avail = 0;
> + mbuf_offset = 0;
one cs nit, put it at the definition.
On 3/4/2016 10:30 AM, Yuanhan Liu wrote:
> On Thu, Mar 03, 2016 at 05:40:14PM +, Xie, Huawei wrote:
>> On 2/18/2016 9:48 PM, Yuanhan Liu wrote:
>>> The current rte_vhost_dequeue_burst() implementation is a bit messy
>> [...]
>>> +
>>> uint16_t
>>> rte_vhost_dequeue_burst(struct virtio_net
On 3/7/2016 10:47 AM, Yuanhan Liu wrote:
> On Mon, Mar 07, 2016 at 02:32:46AM +, Xie, Huawei wrote:
>> On 3/4/2016 10:15 AM, Yuanhan Liu wrote:
>>> On Thu, Mar 03, 2016 at 04:30:42PM +, Xie, Huawei wrote:
On 2/18/2016 9:48 PM, Yuanhan Liu wrote:
> + mbuf_avail = 0;
> +
On 3/4/2016 10:10 AM, Yuanhan Liu wrote:
> On Thu, Mar 03, 2016 at 05:19:42PM +, Xie, Huawei wrote:
>> On 2/18/2016 9:48 PM, Yuanhan Liu wrote:
>>> [...]
>> CCed changchun, the author for the chained handling of desc and mbuf.
>> The change makes the code more readable, but i think the
On 3/4/2016 10:15 AM, Yuanhan Liu wrote:
> On Thu, Mar 03, 2016 at 04:30:42PM +, Xie, Huawei wrote:
>> On 2/18/2016 9:48 PM, Yuanhan Liu wrote:
>>> + mbuf_avail = 0;
>>> + mbuf_offset = 0;
>>> + while (desc_avail || (desc->flags & VRING_DESC_F_NEXT) != 0) {
>>> + /* This desc
On 3/4/2016 10:19 AM, Yuanhan Liu wrote:
> On Thu, Mar 03, 2016 at 04:21:19PM +, Xie, Huawei wrote:
>> On 2/18/2016 9:48 PM, Yuanhan Liu wrote:
>>> The current rte_vhost_dequeue_burst() implementation is a bit messy
>>> and logic twisted. And you could see repeat code here and there: it
>>>
On Thu, Mar 03, 2016 at 05:40:14PM +, Xie, Huawei wrote:
> On 2/18/2016 9:48 PM, Yuanhan Liu wrote:
> > The current rte_vhost_dequeue_burst() implementation is a bit messy
> [...]
> > +
> > uint16_t
> > rte_vhost_dequeue_burst(struct virtio_net *dev, uint16_t queue_id,
> > struct
On Thu, Mar 03, 2016 at 04:21:19PM +, Xie, Huawei wrote:
> On 2/18/2016 9:48 PM, Yuanhan Liu wrote:
> > The current rte_vhost_dequeue_burst() implementation is a bit messy
> > and logic twisted. And you could see repeat code here and there: it
> > invokes rte_pktmbuf_alloc() three times at
On Thu, Mar 03, 2016 at 04:30:42PM +, Xie, Huawei wrote:
> On 2/18/2016 9:48 PM, Yuanhan Liu wrote:
> > + mbuf_avail = 0;
> > + mbuf_offset = 0;
> > + while (desc_avail || (desc->flags & VRING_DESC_F_NEXT) != 0) {
> > + /* This desc reachs to its end, get the next one */
> > +
On Thu, Mar 03, 2016 at 05:19:42PM +, Xie, Huawei wrote:
> On 2/18/2016 9:48 PM, Yuanhan Liu wrote:
> > [...]
> CCed changchun, the author for the chained handling of desc and mbuf.
> The change makes the code more readable, but i think the following
> commit message is simple and enough.
On 2/18/2016 9:48 PM, Yuanhan Liu wrote:
> The current rte_vhost_dequeue_burst() implementation is a bit messy
[...]
> +
> uint16_t
> rte_vhost_dequeue_burst(struct virtio_net *dev, uint16_t queue_id,
> struct rte_mempool *mbuf_pool, struct rte_mbuf **pkts, uint16_t count)
> {
> -
On 2/18/2016 9:48 PM, Yuanhan Liu wrote:
> [...]
CCed changchun, the author for the chained handling of desc and mbuf.
The change makes the code more readable, but i think the following
commit message is simple and enough.
>
> while (this_desc_is_not_drained_totally || has_next_desc) {
>
On 2/18/2016 9:48 PM, Yuanhan Liu wrote:
> + mbuf_avail = 0;
> + mbuf_offset = 0;
> + while (desc_avail || (desc->flags & VRING_DESC_F_NEXT) != 0) {
> + /* This desc reachs to its end, get the next one */
> + if (desc_avail == 0) {
> + desc
On 2/18/2016 9:48 PM, Yuanhan Liu wrote:
> The current rte_vhost_dequeue_burst() implementation is a bit messy
> and logic twisted. And you could see repeat code here and there: it
> invokes rte_pktmbuf_alloc() three times at three different places!
>
> However, rte_vhost_dequeue_burst() acutally
The current rte_vhost_dequeue_burst() implementation is a bit messy
and logic twisted. And you could see repeat code here and there: it
invokes rte_pktmbuf_alloc() three times at three different places!
However, rte_vhost_dequeue_burst() acutally does a simple job: copy
the packet data from vring
18 matches
Mail list logo
