[dpdk-dev] [PATCH 1/3 v3] ethdev: add rx offload to drop error packets

[Nipun Gupta]

> -----Original Message-----
> From: Andrew Rybchenko <andrew.rybchenko at oktetlabs.ru>
> Sent: Tuesday, October 13, 2020 12:52 PM
> To: Nipun Gupta <nipun.gupta at nxp.com>; dev at dpdk.org
> Cc: thomas at monjalon.net; ferruh.yigit at intel.com; Hemant Agrawal
> <hemant.agrawal at nxp.com>; Sachin Saxena <sachin.saxena at nxp.com>; Rohit
> Raj <rohit.raj at nxp.com>; jerinjacobk at gmail.com;
> stephen at networkplumber.org; asafp at nvidia.com
> Subject: Re: [dpdk-dev] [PATCH 1/3 v3] ethdev: add rx offload to drop error
> packets
> 
> On 10/12/20 3:22 PM, Andrew Rybchenko wrote:
> > On 10/12/20 2:30 PM, Nipun Gupta wrote:
> >>> -----Original Message-----
> >>> From: Andrew Rybchenko <Andrew.Rybchenko at oktetlabs.ru>
> >>> Sent: Monday, October 12, 2020 1:32 PM
> >>> To: Nipun Gupta <nipun.gupta at nxp.com>; dev at dpdk.org
> >>> Cc: thomas at monjalon.net; ferruh.yigit at intel.com;
> arybchenko at solarflare.com;
> >>> Hemant Agrawal <hemant.agrawal at nxp.com>; Sachin Saxena
> >>> <sachin.saxena at nxp.com>; Rohit Raj <rohit.raj at nxp.com>;
> >>> jerinjacobk at gmail.com; stephen at networkplumber.org; asafp at nvidia.com
> >>> Subject: Re: [dpdk-dev] [PATCH 1/3 v3] ethdev: add rx offload to drop error
> >>> packets
> >>>
> >>> On 10/9/20 4:13 PM, nipun.gupta at nxp.com wrote:
> >>>> From: Nipun Gupta <nipun.gupta at nxp.com>
> >>>>
> >>>> This change adds a RX offload capability and configuration to
> >>>> enable hardware to drop the packets in case of any error in the
> >>>> packets such as L3 checksum error or L4 checksum.
> >>>>
> >>>> Signed-off-by: Nipun Gupta <nipun.gupta at nxp.com>
> >>>> Signed-off-by: Rohit Raj <rohit.raj at nxp.com>
> >>>> Reviewed-by: Asaf Penso <asafp at nvidia.com>
> 
> Thinking a bit more about it I agree with Thomas idea that
> it should be flow API based solution in fact.
> Drop is just a one of possible actions to be done with
> packets with bad checksum on one or another layer.
> Such packets could be redirected to a slow path
> (dedicated queue or port ID (PF, VF)).
> It is just a missing feature in various layer
> pattern match to say if we want to proceed with packets
> with only good or only bad chehcksum (or we don't care
> as we do right now). Exact match for checksums is hardly
> useful except UDP with zero checksum case.

I would think of it applicable to both, i.e. it could fit in config option
as well as in flow (e.g. RSS we also have as part of both config and flow).
Having this in flow would increase usage like redirecting as you mentioned,
and having in the config will increase utility when simple config like RSS is
used without flow API's.