Re: [dpdk-dev] [PATCH 18/28] drivers/virtio_vhost_user: add virtio PCI framework

[Maxime Coquelin <maxime.coquelin@redhat.com>]

Hi Nikos,

On 9/9/19 10:42 AM, Nikos Dragazis wrote:
> On 5/9/19 7:34 μ.μ., Maxime Coquelin wrote:
>> Hi Nikos,
>>
>> On 6/19/19 5:14 PM, Nikos Dragazis wrote:
>>> The virtio-vhost-user transport requires a driver for the
>>> virtio-vhost-user PCI device, hence it needs a virtio-pci driver.  There
>>> is currently no librte_virtio API that we can use.
>>>
>>> This commit is a hack that duplicates the virtio pci code from
>>> drivers/net/ into drivers/virtio_vhost_user/.  A better solution would
>>> be to extract the code cleanly from drivers/net/ and share it.  Or
>>> perhaps we could backport SPDK's lib/virtio/.
>> I think it would make sense to have a Virtio library, that could be re-
>> used by net, crypto and virtio-vhost-user.
>>
>> I didn't know about SPDK's lib. Maybe it is better to start from virtio-
>> net PMD codebase and then convert crypto and SPDK to use it.
>>
>> What do you think?
>>
>>> drivers/virtio_vhost_user/ will host the virtio-vhost-user transport
>>> implementation in the upcoming patches.
>> Thanks,
>> Maxime
> 
> Hi Maxime,
> 
> thanks for your comments. I agree with you. There is no point in having
> duplicated code here and there. A standalone virtio library sounds like
> a better approach. I will come back with a proper patchset for this
> purpose.

Great, I really appreciate the effort.

As you suggest, I agree it is better to have it done in a dedicated
patch set, the virtio-vhost-user series being already quite big :)

Regards,
Maxime

> Best regards,
> Nikos
>