Re: [patch net-next v2 00/20] net: sched: convert cls ndo_setup_tc offload calls to per-block callbacks

From: "Nambiar, Amritha" <amritha.nambiar@intel.com>
To: Jiri Pirko <jiri@resnulli.us>,
	Alexander Duyck <alexander.duyck@gmail.com>
Cc: David Miller <davem@davemloft.net>,
	Netdev <netdev@vger.kernel.org>,
	Jamal Hadi Salim <jhs@mojatatu.com>,
	Cong Wang <xiyou.wangcong@gmail.com>,
	mlxsw@mellanox.com, Andrew Lunn <andrew@lunn.ch>,
	Vivien Didelot <vivien.didelot@savoirfairelinux.com>,
	Florian Fainelli <f.fainelli@gmail.com>,
	Michael Chan <michael.chan@broadcom.com>,
	Ganesh Goudar <ganeshgr@chelsio.com>,
	Jeff Kirsher <jeffrey.t.kirsher@intel.com>,
	Saeed Mahameed <saeedm@mellanox.com>,
	Matan Barak <matanb@mellanox.com>,
	Leon Romanovsky <leonro@mellanox.com>,
	idosch@mellanox.com,
	Jakub Kicinski <jakub.kicinski@netronome.com>,
	Alexei Starovoitov <ast@kernel.org>,
	Daniel Borkmann <daniel@iogearbox.net>,
	Simon Horman <simon.horman@netronome.com>,
	pieter.jansenvanvuuren@netronome.com, john.hurl
Subject: Re: [patch net-next v2 00/20] net: sched: convert cls ndo_setup_tc offload calls to per-block callbacks
Date: Thu, 26 Oct 2017 13:24:31 -0700	[thread overview]
Message-ID: <2e4baa66-35e3-0259-6228-ab05ec826345@intel.com> (raw)
In-Reply-To: <20171025121503.GA1910@nanopsycho>

On 10/25/2017 5:15 AM, Jiri Pirko wrote:
> Tue, Oct 24, 2017 at 06:01:34PM CEST, alexander.duyck@gmail.com wrote:
> 
> [...]
> 
>> 1.       To offload filter into HW, the hw-tc-offload feature flag has
>> to be turned on before creating the ingress qdisc.
>>
>> Previously, this could also be turned on after the qdisc was created
>> and the filters could still be offloaded. Looks like this is because,
>> previously the offload flag was checked as a part of filter
>> integration in the classifier, and now it is checked as part of qdisc
>> creation (ingress_init). So, if no offload capability is advertised at
>> ingress qdisc creation time then hardware will not be asked to offload
>> filters later if the flag is enabled.
> 
> I have patchset that fixes this in my queue now. Will do some smoke
> testing and send later today.
> 
> 
>>
>> 2.       Deleting the ingress qdisc fails to remove filters added in
>> HW. Filters in SW gets deleted.
>>
>> We haven’t exactly root-caused this, the changes being extensive, but
>> our guess is again something wrong with the offload check or similar
>> while unregistering the block callback (tcf_block_cb_unregister) and
>> further to the classifier (CLS_U32/CLS_FLOWER etc.) with the
>> DESTROY/REMOVE command.
> 
> Hmm. How does this worked previously. I mean, do you see change of
> behaviour? I'm asking because I don't see how rules added only to HW
> could be removed, driver should care of it. Or are you talking about
> rules added to both SW and HW?

These are rules added to both SW and HW. Previously all cls_* had
ndo_setup_tc calls based on the offload capability.

commit 8d26d5636d "net: sched: avoid ndo_setup_tc calls for
TC_SETUP_CLS*" removed this bit to work with the new block callback. Is
there something similar in the block callback flow while acting on the
tcf_proto destroy call initiated when the qdisc is cleared?

> 
> Thanks!
> 