On 8/2/21 6:49 PM, Ilya Maximets wrote:
On 8/1/21 12:22 PM, Andrew Rybchenko wrote:
By its very name, action PORT_ID means that packets hit an ethdev with the
given DPDK port ID. At least the current comments don't state the opposite.
That said, since port representors had been adopted, applications like OvS
have been misusing the action. They misread its purpose as sending packets
to the opposite end of the "wire" plugged to the given ethdev, for example,
redirecting packets to the VF itself rather than to its representor ethdev.
Another example: OvS relies on this action with the admin PF's ethdev port
ID specified in it in order to send offloaded packets to the physical port.
Hi, Andrew. The deprecation notice itself looks OK to me. But I'd suggest
to avoid words like "misuse" and "misread" in the commit message, because I
don't think that it's correct. Since documentation is ambiguous, different
people might interpret it differently. And also, implementation in DPDK
matches with the way how OVS uses the API, otherwise offloading in OVS would
just not work. So, OVS uses this API in a way as it is implemented in DPDK.
If the definition of a DPDK API allows interpretations that doesn't match with
the implementation inside the DPDK itself, that's not a fault of the external
application. And this can not be labeled as "misuse"/"misread". Let's not
create a precedent.
Yes, I agree. I'll send v2 tomorrow.
Thanks for feedback,
Andrew.
