[v5,8/9] ethdev: add capability of sub-function representor

  Old DPDK version or some drivers didn't support SubFunction representor.
For application to adapt different DPDK version automatically, or to be
used for different NICs, this patch introduces new eth device capability
of supporting SubFunction representor device.

Signed-off-by: Xueming Li <xuemingl@nvidia.com>
Acked-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>
Acked-by: Thomas Monjalon <thomas@monjalon.net>
---
 lib/librte_ethdev/rte_ethdev.h | 2 ++
 1 file changed, 2 insertions(+)
  

On 1/19/21 10:15 AM, Xueming Li wrote:
> Old DPDK version or some drivers didn't support SubFunction representor.
> For application to adapt different DPDK version automatically, or to be
> used for different NICs, this patch introduces new eth device capability
> of supporting SubFunction representor device.

Sorry, it does not sound sufficient motivation to introduce the
capability. I simply need real life example why application
need to know it.

> Signed-off-by: Xueming Li <xuemingl@nvidia.com>
> Acked-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>
> Acked-by: Thomas Monjalon <thomas@monjalon.net>
> ---
>  lib/librte_ethdev/rte_ethdev.h | 2 ++
>  1 file changed, 2 insertions(+)
>
> diff --git a/lib/librte_ethdev/rte_ethdev.h b/lib/librte_ethdev/rte_ethdev.h
> index 2cbce958cf..da65adf1ab 100644
> --- a/lib/librte_ethdev/rte_ethdev.h
> +++ b/lib/librte_ethdev/rte_ethdev.h
> @@ -1433,6 +1433,8 @@ struct rte_eth_conf {
>  #define RTE_ETH_DEV_CAPA_RUNTIME_RX_QUEUE_SETUP 0x00000001
>  /** Device supports Tx queue setup after device started. */
>  #define RTE_ETH_DEV_CAPA_RUNTIME_TX_QUEUE_SETUP 0x00000002
> +/** Device supports SubFunction representor. */
> +#define RTE_ETH_DEV_CAPA_REPRESENTOR_SF         0x00000004

Will we have a PMD which supports it in the release cycle?

>  /**@}*/
>  
>  /*
  

>-----Original Message-----
>From: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
>Sent: Tuesday, January 19, 2021 4:06 PM
>To: Xueming(Steven) Li <xuemingl@nvidia.com>
>Cc: dev@dpdk.org; Slava Ovsiienko <viacheslavo@nvidia.com>; Asaf Penso
><asafp@nvidia.com>; NBU-Contact-Thomas Monjalon
><thomas@monjalon.net>; Ferruh Yigit <ferruh.yigit@intel.com>
>Subject: Re: [PATCH v5 8/9] ethdev: add capability of sub-function representor
>
>On 1/19/21 10:15 AM, Xueming Li wrote:
>> Old DPDK version or some drivers didn't support SubFunction representor.
>> For application to adapt different DPDK version automatically, or to
>> be used for different NICs, this patch introduces new eth device
>> capability of supporting SubFunction representor device.
>
>Sorry, it does not sound sufficient motivation to introduce the capability. I
>simply need real life example why application need to know it.

I had same internal discussion on this as well :)
A simple example, for customer running DPDK based app with NICs from different vendors,
app need a flag to know whether the device support SF representor, hotplug SF if the
capability shows "support". This also happens with different model/fw even from same vendor.
PMD report device+driver capability that whether SF supported.

>
>> Signed-off-by: Xueming Li <xuemingl@nvidia.com>
>> Acked-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>
>> Acked-by: Thomas Monjalon <thomas@monjalon.net>
>> ---
>>  lib/librte_ethdev/rte_ethdev.h | 2 ++
>>  1 file changed, 2 insertions(+)
>>
>> diff --git a/lib/librte_ethdev/rte_ethdev.h
>> b/lib/librte_ethdev/rte_ethdev.h index 2cbce958cf..da65adf1ab 100644
>> --- a/lib/librte_ethdev/rte_ethdev.h
>> +++ b/lib/librte_ethdev/rte_ethdev.h
>> @@ -1433,6 +1433,8 @@ struct rte_eth_conf {  #define
>> RTE_ETH_DEV_CAPA_RUNTIME_RX_QUEUE_SETUP 0x00000001
>>  /** Device supports Tx queue setup after device started. */  #define
>> RTE_ETH_DEV_CAPA_RUNTIME_TX_QUEUE_SETUP 0x00000002
>> +/** Device supports SubFunction representor. */
>> +#define RTE_ETH_DEV_CAPA_REPRESENTOR_SF         0x00000004
>
>Will we have a PMD which supports it in the release cycle?

Please see reply above, it also reflect device hw capability.
>
>>  /**@}*/
>>
>>  /*
  

On 1/19/21 2:19 PM, Xueming(Steven) Li wrote:
> 
>> -----Original Message-----
>> From: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
>> Sent: Tuesday, January 19, 2021 4:06 PM
>> To: Xueming(Steven) Li <xuemingl@nvidia.com>
>> Cc: dev@dpdk.org; Slava Ovsiienko <viacheslavo@nvidia.com>; Asaf Penso
>> <asafp@nvidia.com>; NBU-Contact-Thomas Monjalon
>> <thomas@monjalon.net>; Ferruh Yigit <ferruh.yigit@intel.com>
>> Subject: Re: [PATCH v5 8/9] ethdev: add capability of sub-function representor
>>
>> On 1/19/21 10:15 AM, Xueming Li wrote:
>>> Old DPDK version or some drivers didn't support SubFunction representor.
>>> For application to adapt different DPDK version automatically, or to
>>> be used for different NICs, this patch introduces new eth device
>>> capability of supporting SubFunction representor device.
>>
>> Sorry, it does not sound sufficient motivation to introduce the capability. I
>> simply need real life example why application need to know it.
> 
> I had same internal discussion on this as well :)
> A simple example, for customer running DPDK based app with NICs from different vendors,
> app need a flag to know whether the device support SF representor, hotplug SF if the
> capability shows "support". This also happens with different model/fw even from same vendor.
> PMD report device+driver capability that whether SF supported.

Single feature bit is insufficient. Application needs to know
how many SF may be used on which PF.

>>> Signed-off-by: Xueming Li <xuemingl@nvidia.com>
>>> Acked-by: Viacheslav Ovsiienko <viacheslavo@nvidia.com>
>>> Acked-by: Thomas Monjalon <thomas@monjalon.net>

[snip]