[ovs-dev] [PATCH v12 2/6] netdev-dpdk: Convert initialization from cmdline to db

gowrishankar gowrishankar.m at linux.vnet.ibm.com
Tue Jun 14 06:53:17 UTC 2016


Hi Daniele,
Is there a possibility that this patch would also go in branch-2.5 
(2.5.1) ?.

Regards,
Gowrishankar


On Friday 29 April 2016 07:06 PM, Aaron Conole wrote:
> Daniele Di Proietto <diproiettod at vmware.com> writes:
>
>> Hi Aaron,
>>
>> thanks (again!) for this patch. Few comments:
>>
>> * As I mentioned on my previous round of review, I don't think it's necessary to pass the whole Open_vSwitch table to dpdk_init(). I.e., instead of doing
>>
>>    dpdk_init(&cfg);
>>
>>    I'd prefer
>>
>>    if (cfg) {
>>        dpdk_init(&cfg->other_config);
>>    }
>>
>>   This way we don't have to include "vswitch-idl.h".
>>
>>
>> * I suggested 'dpdk-mem-channels', because I think people who want to use that will be comfortable passing '-n' in dpdk-extra.  What do you think? Is there any reason why you think it's worth keeping?
> D'oh! I had done both of these changes, but they were dropped during the
> rebase. I'll cook a fix asap.
>
>> * Sorry for not noticing this before: it seems that netdev_dpdk_register() now always registers the netdev classes, even though they cannot be created.  The registered classes end up in the database in the iface_type column of the Open_vSwitch table, so controllers might think that they're available.  I think we should register the classes only when DPDK is initialized.
> I had an issue doing this, back when the I had the lazy
> initialization. I don't remember the details, though. I'll try it again,
> and see what happens.
>
>> Two minor nits inline,
>>
>> Thanks
> Thanks so much for the review, Daniele!
>
> -Aaron
>
>> On 26/04/2016 12:42, "Aaron Conole" <aconole at redhat.com> wrote:
>>
>>> Existing DPDK integration is provided by use of command line options which
>>> must be split out and passed to librte in a special manner. However, this
>>> forces any configuration to be passed by way of a special DPDK flag, and
>>> interferes with ovs+dpdk packaging solutions.
>>>
>>> This commit delays dpdk initialization until after the OVS database
>>> connection is established, at which point ovs initializes librte. It
>>> pulls all of the config data from the OVS database, and assembles a
>>> new argv/argc pair to be passed along.
>>>
>>> Signed-off-by: Aaron Conole <aconole at redhat.com>
>>> Acked-by: Kevin Traynor <kevin.traynor at intel.com>
>>> ---




More information about the dev mailing list