On 14/04/2016 05:37, "Kavanagh, Mark B" <mark.b.kavan...@intel.com> wrote:
>Hi Daniele,
>
>One comment inline.
>
>Thanks,
>Mark
>
>>
>>This introduces in dpif-netdev and netdev-dpdk the first use for the
>>newly introduce reconfigure netdev call.
>>
>>When a request to change the number of queues comes, netdev-dpdk will
>>remember this and notify the upper layer via
>>netdev_request_reconfigure().
>>
>>The datapath, instead of periodically calling netdev_set_multiq(), can
>>detect this and call reconfigure().
>>
>>This mechanism can also be used to:
>>* Automatically match the number of rxq with the one provided by qemu
>> via the new_device callback.
>>* Provide a way to change the MTU of dpdk devices at runtime.
>>* Move a DPDK vhost device to the proper NUMA socket.
>>
>>Signed-off-by: Daniele Di Proietto <diproiet...@vmware.com>
>>---
>> lib/dpif-netdev.c | 69 +++++++++---------
>> lib/netdev-dpdk.c | 195
>>++++++++++++++++++++++++++------------------------
>> lib/netdev-provider.h | 23 +++---
>> lib/netdev.c | 34 +++------
>> lib/netdev.h | 3 +-
>> 5 files changed, 155 insertions(+), 169 deletions(-)
[...]
>>@@ -312,12 +305,12 @@ struct netdev_class {
>> * making sure that these concurrent calls do not create a race
>>condition
>> * by using multiple hw queues or locking.
>> *
>>- * On error, the tx queue and rx queue configuration is
>>indeterminant.
>>- * Caller should make decision on whether to restore the previous or
>>- * the default configuration. Also, caller must make sure there is
>>no
>>- * other thread accessing the queues at the same time. */
>>- int (*set_multiq)(struct netdev *netdev, unsigned int n_txq,
>>- unsigned int n_rxq);
>>+ * The caller will call netdev_reconfigure() (if necessary) before
>>using
>>+ * netdev_send() on any of the newly configured queues, giving the
>>provider
>>+ * a chance to adjust its settings.
>>+ *
>>+ * On error, the tx queue configuration is unchanged. */
>>+ int (*set_multiq)(struct netdev *netdev, unsigned int n_txq);
>
>Since this function now deals only with TX queues, an identifier along
>the lines of 'set_tx_multiq' might more accurately describe its
>functionality. Specific netdev classes would need to modify the names of
>their own specific 'set_multiq' functions accordingly.
You're right, 'set_tx_multiq()' is definitely a better name for it. I
updated it.
Thanks for all your feedback
_______________________________________________
dev mailing list
dev@openvswitch.org
http://openvswitch.org/mailman/listinfo/dev
