From: Thomas Monjalon <thomas@monjalon.net>
To: Andrew Rybchenko <andrew.rybchenko@oktetlabs.ru>
Cc: Rongwei Liu <rongweil@nvidia.com>,
matan@nvidia.com, viacheslavo@nvidia.com, orika@nvidia.com,
Aman Singh <aman.deep.singh@intel.com>,
Yuying Zhang <yuying.zhang@intel.com>,
Ferruh Yigit <ferruh.yigit@amd.com>,
dev@dpdk.org, rasland@nvidia.com
Subject: Re: [PATCH v4] ethdev: add special flags when creating async transfer table
Date: Wed, 09 Nov 2022 11:50:48 +0100 [thread overview]
Message-ID: <6951385.dE46n4Xy2H@thomas> (raw)
In-Reply-To: <0e3fc6fc-40e0-bca3-0d18-c426160186e6@oktetlabs.ru>
09/11/2022 10:36, Andrew Rybchenko:
> On 11/9/22 12:03, Thomas Monjalon wrote:
> > 09/11/2022 09:53, Andrew Rybchenko:
> >> On 11/8/22 18:25, Thomas Monjalon wrote:
> >>> 08/11/2022 15:38, Andrew Rybchenko:
> >>>> On 11/8/22 16:29, Thomas Monjalon wrote:
> >>>>> 08/11/2022 12:47, Andrew Rybchenko:
> >>>>>> On 11/8/22 14:39, Andrew Rybchenko wrote:
> >>>>>>> On 11/4/22 13:44, Rongwei Liu wrote:
> >>>>>>>> diff --git a/lib/ethdev/rte_flow.h b/lib/ethdev/rte_flow.h
> >>>>>>>> index 8858b56428..1eab12796f 100644
> >>>>>>>> --- a/lib/ethdev/rte_flow.h
> >>>>>>>> +++ b/lib/ethdev/rte_flow.h
> >>>>>>>> @@ -5186,6 +5186,34 @@ rte_flow_actions_template_destroy(uint16_t
> >>>>>>>> port_id,
> >>>>>>>> */
> >>>>>>>> struct rte_flow_template_table;
> >>>>>>>> +/**
> >>>>>>>> + * @warning
> >>>>>>>> + * @b EXPERIMENTAL: this API may change without prior notice.
> >>>>>>>> + *
> >>>>>>>> + * Special optional flags for template table attribute.
> >>>>>>>> + * Each bit stands for a table specialization
> >>>>>>>> + * offering a potential optimization at PMD layer.
> >>>>>>>> + * PMD can ignore the unsupported bits silently.
> >>>>>>>> + */
> >>>>>>>> +enum rte_flow_template_table_specialize {
> >>>>>>>> + /**
> >>>>>>>> + * Specialize table for transfer flows which come only from wire.
> >>>>>>>> + * It allows PMD not to allocate resources for non-wire
> >>>>>>>> originated traffic.
> >>>>>>>> + * This bit is not a matching criteria, just an optimization hint.
> >>>>>>>> + * Flow rules which match non-wire originated traffic will be missed
> >>>>>>>> + * if the hint is supported.
> >>>>>>
> >>>>>> Sorry, but if so, the hint changes behavior.
> >>>>>
> >>>>> Yes the hint may change behaviour.
> >>>>>
> >>>>>> Let's consider a rule which matches both VF originating and
> >>>>>> wire originating traffic. Will the rule be missed (ignored)
> >>>>>> regardless if the hint is supported or not?
> >>>>>
> >>>>> If the hint RTE_FLOW_TRANSFER_WIRE_ORIG is used,
> >>>>> the PMD may assume the table won't be used for traffic
> >>>>> which is not coming from wire ports.
> >>>>> As a consequence, the table may be implemented on the path
> >>>>> of wire traffic only.
> >>>>> In this case, the traffic coming from virtual ports
> >>>>> won't be affected by this table.
> >>>>> To answer the question, a rule matching both virtual and wire traffic
> >>>>> will be applied in a table affecting only wire traffic,
> >>>>> so it will still apply (not completely ignored).
> >>>>
> >>>> If so, it is not a hint. It becomes matching criteria
> >>>> which should be in pattern as we discussed.
> >>>
> >>> It is not a strict matching because the PMD is free to support it or not.
> >>
> >> It cannot be optional matching criteria. Matching criteria must
> >> be always mandatory. Otherwise application does not know what
> >> to expect and behaviour may legitimately vary on different
> >> vendors.
> >
> > I think you take it in the wrong direction.
> > The idea is not to have it as a criteria.
> > Let me explain again:
> >
> > If an application is using a flow table to manage flows
> > which *always* come from the same type of port (wire or virtual),
>
> What does guarantee it? Is it used a jump-table and jump rule
> must guarantee it? Or has pattern corresponding unit?
>
> It is very thin ice and I'm ready to bet money that finally
> it will be used as a matching criteria intentionally or not
> intentionally. Simply because it works as matching criteria
> on, for example, Mellanox. I.e. if rules from table with
> corresponding hint are programmed to HW which applies these
> rules on traffic from wire only - effectively it is a matching
> criteria. And it will be used this way. And it will be not
> portable to other HW which does not support the hint.
> So, we're making an API which is very easy to misuse if not
> to say more.
I completely understand your concern (I have same).
In other words, if the application misuse the hint,
it will become not portable.
That's why I made sure to highlight such misue consequence
in the API comments.
> You know better if it is OK or not to rely on liable users
> in the case of DPDK.
I do not rely on users, and I don't want to block innovation.
That's why I want to make sure all is explained and clear,
so freedom comes with responsibility.
> It would be much safer if we do not rely on application in this
> case, introduce a new pattern item to specify origin and
> require PMD to check that pattern has either a new pattern item
> or corresponding REPRESENTED_PORT/PORT_REPRESENTOR pattern
> item.
Safer is not often compatible with fastest :)
> I realize that my concerns could be not valid and it is just
> a paranoia. Just add your ack and let's move forward.
Let's wait for other opinions.
> > then the application can give this information to the driver.
> > With this assumption coming from the application,
> > the driver may do some optimizations.
> >
> > Now about what is explained above:
> > If the application gives such a hint
> > but does not respect its own assumption,
> > then confusion happens.
next prev parent reply other threads:[~2022-11-09 10:50 UTC|newest]
Thread overview: 96+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-09-07 2:40 [PATCH v1] ethdev: add direction info when creating the " Rongwei Liu
2022-09-11 8:22 ` Ori Kam
2022-09-12 16:57 ` Ivan Malov
2022-09-13 13:46 ` Rongwei Liu
2022-09-13 14:33 ` Ivan Malov
2022-09-14 5:16 ` Rongwei Liu
2022-09-14 7:32 ` Ivan Malov
2022-09-14 10:17 ` Rongwei Liu
2022-09-14 15:18 ` Ivan Malov
2022-09-14 21:02 ` Thomas Monjalon
2022-09-15 0:58 ` Rongwei Liu
2022-09-15 7:47 ` Ivan Malov
2022-09-15 8:18 ` Thomas Monjalon
2022-09-15 9:42 ` Ivan Malov
2022-09-15 8:48 ` Rongwei Liu
2022-09-15 10:59 ` Ivan Malov
2022-09-15 11:16 ` Thomas Monjalon
2022-09-20 9:41 ` Ori Kam
2022-09-20 12:45 ` Ivan Malov
2022-09-20 13:59 ` Ori Kam
2022-09-20 15:28 ` Ivan Malov
2022-09-21 7:34 ` Ori Kam
2022-09-21 8:39 ` Andrew Rybchenko
2022-09-21 9:04 ` Ivan Malov
2022-09-21 9:40 ` Thomas Monjalon
2022-09-21 10:04 ` Andrew Rybchenko
2022-09-21 12:41 ` Ori Kam
2022-09-21 12:51 ` Morten Brørup
2022-09-22 7:39 ` Andrew Rybchenko
2022-09-22 10:06 ` Ori Kam
2022-09-22 10:31 ` Andrew Rybchenko
2022-09-22 13:00 ` Ori Kam
2022-09-23 7:25 ` Andrew Rybchenko
2022-09-23 16:11 ` Ori Kam
2022-09-22 12:43 ` Ivan Malov
2022-09-22 14:46 ` Ori Kam
2022-09-28 9:24 ` [PATCH v3] ethdev: add hint when creating async " Rongwei Liu
2022-10-04 8:31 ` Andrew Rybchenko
2022-11-04 10:42 ` [PATCH v4] ethdev: add special flags " Rongwei Liu
2022-11-04 10:44 ` Rongwei Liu
2022-11-08 11:39 ` Andrew Rybchenko
2022-11-08 11:47 ` Andrew Rybchenko
2022-11-08 13:29 ` Thomas Monjalon
2022-11-08 14:38 ` Andrew Rybchenko
2022-11-08 15:25 ` Thomas Monjalon
2022-11-09 8:53 ` Andrew Rybchenko
2022-11-09 9:03 ` Thomas Monjalon
2022-11-09 9:36 ` Andrew Rybchenko
2022-11-09 10:50 ` Thomas Monjalon [this message]
2022-11-06 10:02 ` [PATCH v3] ethdev: add hint " Andrew Rybchenko
2022-11-07 1:58 ` Rongwei Liu
2022-11-08 9:19 ` Thomas Monjalon
2022-11-08 9:35 ` Andrew Rybchenko
2022-11-08 11:18 ` Thomas Monjalon
2022-11-08 11:48 ` Andrew Rybchenko
2022-11-14 8:47 ` [PATCH v6] ethdev: add special flags " Rongwei Liu
2022-11-14 11:59 ` [PATCH v7] " Rongwei Liu
2023-01-17 15:13 ` Ferruh Yigit
2023-01-17 17:01 ` Ferruh Yigit
2023-01-18 2:50 ` Rongwei Liu
2023-01-18 7:30 ` Andrew Rybchenko
2023-01-18 7:28 ` Andrew Rybchenko
2023-01-18 16:18 ` Thomas Monjalon
2023-02-01 10:17 ` Andrew Rybchenko
2023-02-01 10:58 ` Thomas Monjalon
2023-02-01 11:10 ` Andrew Rybchenko
2023-02-01 11:18 ` Thomas Monjalon
2023-02-01 11:38 ` Andrew Rybchenko
2023-02-01 13:48 ` Thomas Monjalon
2023-02-02 9:21 ` Andrew Rybchenko
2023-02-02 11:29 ` Thomas Monjalon
2023-02-02 12:24 ` Andrew Rybchenko
2023-02-01 11:22 ` Ori Kam
2023-02-01 11:29 ` Andrew Rybchenko
2023-02-01 11:12 ` Ori Kam
2023-02-01 11:20 ` Thomas Monjalon
2023-01-30 0:00 ` Ivan Malov
2023-01-30 2:34 ` Rongwei Liu
2023-01-30 7:40 ` Ivan Malov
2023-01-30 14:49 ` Rongwei Liu
2023-01-30 23:00 ` Ivan Malov
2023-01-31 3:06 ` Rongwei Liu
2023-01-31 5:30 ` Ivan Malov
2023-01-31 6:14 ` Rongwei Liu
2023-02-01 10:12 ` Thomas Monjalon
2023-02-01 11:50 ` Ivan Malov
2023-02-01 13:37 ` Thomas Monjalon
2023-02-01 14:04 ` Ivan Malov
2023-02-01 14:23 ` Thomas Monjalon
2023-02-01 14:29 ` Ori Kam
2023-02-02 11:19 ` [PATCH v8] ethdev: add optimization hints in flow template table Rongwei Liu
2023-02-02 11:33 ` Thomas Monjalon
2023-02-08 23:19 ` Ferruh Yigit
2022-11-09 8:11 ` [PATCH v5] ethdev: add special flags when creating async transfer table Rongwei Liu
2022-11-09 8:13 ` Rongwei Liu
2022-11-09 8:31 ` Thomas Monjalon
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=6951385.dE46n4Xy2H@thomas \
--to=thomas@monjalon.net \
--cc=aman.deep.singh@intel.com \
--cc=andrew.rybchenko@oktetlabs.ru \
--cc=dev@dpdk.org \
--cc=ferruh.yigit@amd.com \
--cc=matan@nvidia.com \
--cc=orika@nvidia.com \
--cc=rasland@nvidia.com \
--cc=rongweil@nvidia.com \
--cc=viacheslavo@nvidia.com \
--cc=yuying.zhang@intel.com \
/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
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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).