On Tue, Jun 21, 2016 at 9:58 PM, William Tu <u9012...@gmail.com> wrote:
> Hi Pravin,
>
> Thanks for the feedback.
>
>>> @@ -4084,10 +4089,13 @@ dp_execute_cb(void *aux_, struct dp_packet_batch
>>> *packets_,
>>> int i;
>>>
>>> if (!may_steal) {
>>> - dp_packet_batch_clone(&tnl_pkt, packets_);
>>> - packets_ = &tnl_pkt;
>>> + dp_packet_batch_clone(&tnl_pkt, packets_);
>>> + packets_ = &tnl_pkt;
>>> }
>>>
>>> + dp_packet_batch_cutlen(packets_);
>>> + dp_packet_batch_reset_cutlen(orig_packets_);
>>> +
>> is there any need to call dp_packet_batch_reset_cutlen() function
>> after dp_packet_batch_cutlen()?
>
> Yes, if may_steal == false, we need to clear the cutlen at the
> original packets so that it won't leak to other actions.
>
Can you move this to the block which test for may_steal?
>>> @@ -4107,22 +4115,41 @@ dp_execute_cb(void *aux_, struct dp_packet_batch
>>> *packets_,
>>>
>>> case OVS_ACTION_ATTR_USERSPACE:
>>> if (!fat_rwlock_tryrdlock(&dp->upcall_rwlock)) {
>>> + struct dp_packet_bat
>>> @@ -1585,6 +1585,12 @@ netdev_dpdk_send__(struct netdev_dpdk *dev, int qid,
>>>
>>> for (i = 0; i < cnt; i++) {
>>> int size = dp_packet_size(pkts[i]);
>>> + uint32_t cutlen = dp_packet_get_cutlen(pkts[i]);
>>> +
>>> + if (cutlen > 0) {
>>> + size -= cutlen;
>>> + dp_packet_set_size(pkts[i], size);
>>> + }
>> No need to check if cutlen is positive value, subtracting zero does
>> not matter anyways.
>
> Yes, no need to check.
> I did this check on purpose since for dpdk, dp_packet_set_size() does
> two assignment, set data_len and pkt_len. If cutlen is rarely used, I
> feel it's more efficient adding this check.
>
It can be optimized using batch truncate flag. But this can be
addressed later after the API changes.
_______________________________________________
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
