Netdev Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Jamal Hadi Salim <jhs@mojatatu.com>
To: Vlad Buslov <vladbu@nvidia.com>,
	Simon Horman <simon.horman@corigine.com>
Cc: David Miller <davem@davemloft.net>,
	Jakub Kicinski <kuba@kernel.org>,
	Cong Wang <xiyou.wangcong@gmail.com>,
	Jiri Pirko <jiri@mellanox.com>,
	netdev@vger.kernel.org, oss-drivers@corigine.com,
	Baowen Zheng <baowen.zheng@corigine.com>,
	Louis Peens <louis.peens@corigine.com>
Subject: Re: [PATCH net-next 1/3] flow_offload: allow user to offload tc action to net device
Date: Tue, 27 Jul 2021 12:13:37 -0400	[thread overview]
Message-ID: <95d6873c-256c-0462-60f7-56dbffb8221b@mojatatu.com> (raw)
In-Reply-To: <ygnh7dhbrfd0.fsf@nvidia.com>

On 2021-07-27 10:38 a.m., Vlad Buslov wrote:
> 
> On Tue 27 Jul 2021 at 16:04, Simon Horman <simon.horman@corigine.com> wrote:

>>>
>>> Also showing a tc command line in the cover letter on how one would
>>> ask for a specific action to be offloaded.
>>
>> In practice actions are offloaded when a flow using them is offloaded.
>> So I think we need to consider what the meaning of IN_HW is.
>>
>> Is it that:
>>
>> * The driver (and potentially hardware, though not in our current
>>    implementation) has accepted the action for offload;
>> * That a classifier that uses the action has bee offloaded;
>> * Or something else?
> 
> I think we have the same issue with filters - they might not be in
> hardware after driver callback returned "success" (due to neigh state
> being invalid for tunnel_key encap, for example).
> 

Sounds like we need another state for this. Otherwise, how do you debug
that something is sitting in the driver and not in hardware after you
issued a command to offload it? How do i tell today?
Also knowing reason why something is sitting in the driver would be
helpful.

>> With regards to a counter, I'm not quite sure what this would be:
>>
>> * The number of devices where the action has been offloaded (which ties
>>    into the question of what we mean by IN_HW)
>> * The number of offloaded classifier instances using the action
>> * Something else
> 
> I would prefer to have semantics similar to filters:
> 
> 1. Count number of driver callbacks that returned "success".
> 
> 2. If count > 0, then set in_hw flag.
> 
> 3. Set in_hw_count to success count.
> 
> This would allow user to immediately determine whether action passed
> driver validation.
>

I didnt follow this:
Are we refering to the the "block" semantics (where a filter for
example applies to multiple devices)?

>>
>> Regarding a flag to control offload:
>>
>> * For classifiers (at least the flower classifier) there is the skip_sw and
>>    skip_hw flags, which allow control of placement of a classifier in SW and
>>    HW.
>> * We could add similar flags for actions, which at least in my
>>    world view would have the net-effect of controlling which classifiers can
>>    be added to sw and hw - f.e. a classifier that uses an action marked
>>    skip_hw could not be added to HW.

I guess it depends on the hardware implementation.
In S/W we have two modes:
Approach A: create an action and then 2) bind it to a filter.
Approach B: Create a filter and then bind it to an action.

And #2A can be repeated multiple times for the same action
(would require some index as a reference for the action)
To Simon's comment above that would mean allowing
"a classifier that uses an action marked skip_hw to be added to HW"
i.e
Some hardware is capable of doing both option #A and #B.

Todays offload assumes #B - in which both filter and action are assumed
offloaded.

I am hoping whatever approach we end up agreeing on doesnt limit
either mode.

>> * Doing so would add some extra complexity and its not immediately apparent
>>    to me what the use-case would be given that there are already flags for
>>    classifiers.
> Yeah, adding such flag for action offload seems to complicate things.
> Also, "skip_sw" flag doesn't even make much sense for actions. I thought
> that "skip_hw" flag would be nice to have for users that would like to
> avoid "spamming" their NIC drivers (potentially causing higher latency
> and resource consumption) for filters/actions they have no intention to
> offload to hardware, but I'm not sure how useful is that option really
> is.

Hold on Vlad.
So you are looking at this mostly as an optimization to speed up h/w
control updates? ;->

I was looking at it more as a (currently missing) feature improvement.
We already have a use case that is implemented by s/w today. The feature
mimics it in h/w.

At minimal all existing NICs should be able to support the counters
as mapped to simple actions like drop. I understand for example if some
cant support adding separately offloading of tunnels for example.
So the syntax is something along the lines of:

tc actions add action drop index 15 skip_sw
tc filter add dev ...parent ... protocol ip prio X ..\
u32/flower skip_sw match ... flowid 1:10 action gact index 15

You get an error if counter index 15 is not offloaded or
if skip_sw was left out..

And then later on, if you support sharing of actions:
tc filter add dev ...parent ... protocol ip prio X2 ..\
u32/flower skip_sw match ... flowid 1:10 action gact index 15

cheers,
jamal



  reply	other threads:[~2021-07-27 16:13 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-07-22  9:19 [PATCH net-next 0/3] flow_offload: hardware offload of TC actions Simon Horman
2021-07-22  9:19 ` [PATCH net-next 1/3] flow_offload: allow user to offload tc action to net device Simon Horman
2021-07-22 12:24   ` Roi Dayan
2021-07-22 13:19     ` Simon Horman
2021-07-22 13:29   ` Vlad Buslov
2021-07-22 13:33     ` Jamal Hadi Salim
2021-07-27 13:04       ` Simon Horman
2021-07-27 14:38         ` Vlad Buslov
2021-07-27 16:13           ` Jamal Hadi Salim [this message]
2021-07-27 16:47             ` Vlad Buslov
2021-07-28  7:46               ` Simon Horman
2021-07-28  8:05                 ` Vlad Buslov
2021-07-28 13:51                 ` Jamal Hadi Salim
2021-07-28 14:46                   ` Simon Horman
2021-07-30 10:17                     ` Jamal Hadi Salim
2021-07-30 11:40                       ` Vlad Buslov
2021-08-03  9:57                         ` Jamal Hadi Salim
2021-08-03 12:02                           ` tc offload debug-ability Jamal Hadi Salim
2021-08-03 12:14                             ` Vlad Buslov
2021-08-03 12:50                               ` Jamal Hadi Salim
2021-08-03 13:34                                 ` Ido Schimmel
2021-07-30 13:20                       ` [PATCH net-next 1/3] flow_offload: allow user to offload tc action to net device Simon Horman
2021-08-03 10:14                         ` Jamal Hadi Salim
2021-08-03 11:36                           ` Simon Horman
2021-08-03 11:45                             ` Jamal Hadi Salim
2021-08-03 12:31                               ` Simon Horman
2021-08-03 13:01                                 ` Jamal Hadi Salim
2021-08-03 14:46                                   ` Simon Horman
2021-07-22 13:57   ` kernel test robot
2021-07-22 15:31   ` kernel test robot
2021-08-03 10:50   ` Jamal Hadi Salim
2021-08-03 11:05   ` Jamal Hadi Salim
2021-08-03 11:31     ` Simon Horman
2021-07-22  9:19 ` [PATCH net-next 2/3] flow_offload: add process to delete offloaded actions from " Simon Horman
2021-07-22 14:25   ` Vlad Buslov
2021-07-22 14:50   ` kernel test robot
2021-07-22 17:07   ` kernel test robot
2021-08-03 10:59   ` Jamal Hadi Salim
2021-07-22  9:19 ` [PATCH net-next 3/3] flow_offload: add process to update action stats from hardware Simon Horman
2021-07-22 14:55   ` Vlad Buslov
2021-08-03 11:24   ` Jamal Hadi Salim
2021-08-03 11:35     ` Simon Horman

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=95d6873c-256c-0462-60f7-56dbffb8221b@mojatatu.com \
    --to=jhs@mojatatu.com \
    --cc=baowen.zheng@corigine.com \
    --cc=davem@davemloft.net \
    --cc=jiri@mellanox.com \
    --cc=kuba@kernel.org \
    --cc=louis.peens@corigine.com \
    --cc=netdev@vger.kernel.org \
    --cc=oss-drivers@corigine.com \
    --cc=simon.horman@corigine.com \
    --cc=vladbu@nvidia.com \
    --cc=xiyou.wangcong@gmail.com \
    --subject='Re: [PATCH net-next 1/3] flow_offload: allow user to offload tc action to net device' \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).