[dpdk-dev] [PATCH v3 01/39] examples/l2fwd: convert to new ethdev offloads API

Ferruh Yigit ferruh.yigit at intel.com
Mon Jan 15 11:20:32 CET 2018


On 1/14/2018 10:37 AM, Shahaf Shuler wrote:
> Friday, January 12, 2018 3:31 PM, Ferruh Yigit:
>> On 12/26/2017 9:23 AM, Shahaf Shuler wrote:
>>> Ethdev offloads API has changed since:
>>>
>>> commit ce17eddefc20 ("ethdev: introduce Rx queue offloads API") commit
>>> cba7f53b717d ("ethdev: introduce Tx queue offloads API")
>>>
>>> This commit support the new API.
>>
>> This patch does three things:
>> 1- Convert bit-field Rx Offload information to new bitwise "offloads" variable.
>> 2- Use new queue specific offload configuration for Rx/Tx
>> 3- Enable new mbuf fast free Tx offload
>>
>>
>> 1 and 2 can be classified as "convert to new ethdev offloads", but I am not
>> sure about 3.
> 
> I think all of the 3 should be in a single patch.
> The reason is that the convert patch should maintain the same offloads configuration needed for the application.

Perhaps I am missing some details about "mbuf fast free" offload, can you please
give more details about it, what does having or not having it mean?
Currently no PMD seems implemented it.

> Before the convert patch the examples were using the default configuration set by the PMD. In there the txq flags were set to ignore ref count and to declare all mbufs are from the same pool. 
> The fast free Tx offload was added in order to keep this old offloads configuration.
> 
>>
>> Wouldn't be better to enable new offloadings in a separate patch, other than
>> convert one? And I don't know if we want to enable that specific offload for
>> all samples.
> 
> As you can see, not all the examples has the FAST_FREE offloads, only the entitled ones (i.e. single mempool and no ref count).
> For example, ipv4_multicast doesn't set this offload flag.
> 



More information about the dev mailing list