Re: [dpdk-dev] [PATCH] ethdev: add flow tag

[Slava Ovsiienko <viacheslavo@mellanox.com>]

> -----Original Message-----
> From: Yigit, Ferruh <ferruh.yigit@linux.intel.com>
> Sent: Tuesday, October 8, 2019 15:57
> To: Yongseok Koh <yskoh@mellanox.com>; Adrien Mazarguil
> <adrien.mazarguil@6wind.com>
> Cc: Shahaf Shuler <shahafs@mellanox.com>; Thomas Monjalon
> <thomas@monjalon.net>; Ferruh Yigit <ferruh.yigit@intel.com>; Andrew
> Rybchenko <arybchenko@solarflare.com>; Olivier Matz
> <olivier.matz@6wind.com>; dev <dev@dpdk.org>; Slava Ovsiienko
> <viacheslavo@mellanox.com>
> Subject: Re: [dpdk-dev] [PATCH] ethdev: add flow tag
> 
> On 7/11/2019 2:59 AM, Yongseok Koh wrote:
> > On Tue, Jul 09, 2019 at 10:38:06AM +0200, Adrien Mazarguil wrote:
> >> On Fri, Jul 05, 2019 at 06:05:50PM +0000, Yongseok Koh wrote:
> >>>> On Jul 5, 2019, at 6:54 AM, Adrien Mazarguil
> <adrien.mazarguil@6wind.com> wrote:
> >>>>
> >>>> On Thu, Jul 04, 2019 at 04:23:02PM -0700, Yongseok Koh wrote:
> >>>>> A tag is a transient data which can be used during flow match.
> >>>>> This can be used to store match result from a previous table so
> >>>>> that the same pattern need not be matched again on the next table.
> >>>>> Even if outer header is decapsulated on the previous match, the match
> result can be kept.
> >>>>>
> >>>>> Some device expose internal registers of its flow processing
> >>>>> pipeline and those registers are quite useful for stateful
> >>>>> connection tracking as it keeps status of flow matching. Multiple
> >>>>> tags are supported by specifying index.
> >>>>>
> >>>>> Example testpmd commands are:
> >>>>>
> >>>>>  flow create 0 ingress pattern ... / end
> >>>>>    actions set_tag index 2 value 0xaa00bb mask 0xffff00ff /
> >>>>>            set_tag index 3 value 0x123456 mask 0xffffff /
> >>>>>            vxlan_decap / jump group 1 / end
> >>>>>
> >>>>>  flow create 0 ingress pattern ... / end
> >>>>>    actions set_tag index 2 value 0xcc00 mask 0xff00 /
> >>>>>            set_tag index 3 value 0x123456 mask 0xffffff /
> >>>>>            vxlan_decap / jump group 1 / end
> >>>>>
> >>>>>  flow create 0 ingress group 1
> >>>>>    pattern tag index is 2 value spec 0xaa00bb value mask 0xffff00ff /
> >>>>>            eth ... / end
> >>>>>    actions ... jump group 2 / end
> >>>>>
> >>>>>  flow create 0 ingress group 1
> >>>>>    pattern tag index is 2 value spec 0xcc00 value mask 0xff00 /
> >>>>>            tag index is 3 value spec 0x123456 value mask 0xffffff /
> >>>>>            eth ... / end
> >>>>>    actions ... / end
> >>>>>
> >>>>>  flow create 0 ingress group 2
> >>>>>    pattern tag index is 3 value spec 0x123456 value mask 0xffffff /
> >>>>>            eth ... / end
> >>>>>    actions ... / end
> >>>>>
> >>>>> Signed-off-by: Yongseok Koh <yskoh@mellanox.com>
> >>>>
> >>>> Hi Yongseok,
> >>>>
> >>>> Only high level questions for now, while it unquestionably looks
> >>>> useful, from a user standpoint exposing the separate index seems
> >>>> redundant and not necessarily convenient. Using the following example
> to illustrate:
> >>>>
> >>>> actions set_tag index 3 value 0x123456 mask 0xfffff
> >>>>
> >>>> pattern tag index is 3 value spec 0x123456 value mask 0xffffff
> >>>>
> >>>> I might be missing something, but why isn't this enough:
> >>>>
> >>>> pattern tag index is 3 # match whatever is stored at index 3
> >>>>
> >>>> Assuming it can work, then why bother with providing value
> >>>> spec/mask on set_tag? A flow rule pattern matches something, sets
> >>>> some arbitrary tag to be matched by a subsequent flow rule and
> >>>> that's it. It even seems like relying on the index only on both occasions
> is enough for identification.
> >>>>
> >>>> Same question for the opposite approach; relying on the value,
> >>>> never mentioning the index.
> >>>>
> >>>> I'm under the impression that the index is a hardware-specific
> >>>> constraint that shouldn't be exposed (especially since it's an
> >>>> 8-bit field). If so, a PMD could keep track of used indices without
> >>>> having them exposed through the public API.
> >>>
> >>>
> >>> Thank you for review, Adrien.
> >>> Hope you are doing well. It's been long since we talked each other.
> >>> :-)
> >>
> >> Yeah clearly! Hope you're doing well too. I'm somewhat busy hence
> >> slow to answer these days...
> >>
> >>  <dev@dpdk.org> hey!
> >>  <dev@dpdk.org> no private talks!
> >>
> >> Back to the topic:
> >>
> >>> Your approach will work too in general but we have a request from
> >>> customer that they want to partition this limited tag storage.
> >>> Assuming that HW exposes 32bit tags (those are 'registers' in HW
> >>> pipeline in mlx5 HW). Then, customers want to store multiple data
> >>> even in a 32-bit storage. For example, 16bit vlan tag, 8bit table id
> >>> and 8bit flow id. As they want to split one 32bit storage, I thought
> >>> it is better to provide mask when setting/matching the value. Even
> >>> some customer wants to store multiple flags bit by bit like ol_flags. They
> do want to alter only partial bits.
> >>>
> >>> And for the index, it is to reference an entry of tags array as HW
> >>> can provide larger registers than 32-bit. For example, mlx5 HW would
> >>> provide 4 of 32b storage which users can use for their own sake.
> >>> 	tag[0], tag[1], tag[2], tag[3]
> >>
> >> OK, looks like I missed the point then. I initially took it for a
> >> funky alternative to RTE_FLOW_ITEM_TYPE_META &
> >> RTE_FLOW_ACTION_TYPE_SET_META (ingress extended [1]) but while it
> >> could be used like that, it's more of a way to temporarily store and
> retrieve a small amount of data, correct?
> >
> > Correct.
> >
> >> Out of curiosity, are these registers independent from META and other
> >> items/actions in mlx5, otherwise what happens if they are combined?
> >
> > I thought about combining it but I chose this way. Because it is
> > transient. META can be set by packet descriptor on Tx and can be
> > delivered to host via mbuf on Rx, but this TAG item can't. If I
> > combine it, users have to query this capability for each 32b storage.
> > And also, there should be a way to request data from such storages
> > (i.e. new action , e.g. copy_meta). Let's say there are 4x32b storages
> > - meta[4]. If user wants to get one 32b data (meta[i]) out of them to
> > mbuf->metadata, it should be something like,
> > 	ingress / pattern .. /
> > 	actions ... set_meta index i data x / copy_meta_to_rx index i And if
> > user wants to set meta[i] via mbuf on Tx,
> > 	egress / pattern meta index is i data is x ... /
> > 	actions ... copy_meta_to_tx index i
> >
> > For sure, user is also responsible for querying these capabilities per
> > each meta[] storage.
> >
> > As copy_meta_to_tx/rx isn't a real action, this example would confuse
> user.
> > 	egress / pattern meta index is i data is x ... /
> > 	actions ... copy_meta_to_tx index i
> >
> > User might misunderstand the order of two things - item meta and
> > copy_meta action. I also thought about having capability bits per each
> > meta[] storage but it also looked complex.
> >
> > I do think rte_flow item/action is better to be simple, atomic and intuitive.
> > That's why I made this choice.
> >
> >> Are there other uses for these registers? Say, referencing their
> >> contents from other places in a flow rule so they don't have to be hard-
> coded?
> >
> > Possible.
> > Actually, this feature is needed by connection tracking of OVS-DPDK.
> >
> >> Right now I'm still uncomfortable with such a feature in the public
> >> API because compared to META [1], this approach looks very
> >> hardware-specific and seemingly difficult to map on different HW
> architectures.
> >
> > I wouldn't say it is HW-specific. Like I explained above, I just
> > define this new item/action to make things easy-to-use and intuitive.
> >
> >> However, the main problem is that as described, its end purpose seems
> >> redundant with META, which I think can cover the use cases you gave.
> >> So what can an application do with this that couldn't be done in a
> >> more generic fashion through META?
> >>
> >> I may still be missing something and I'm open to ideas, but assuming
> >> it doesn't make it into the public rte_flow API, it remains an
> >> interesting feature on its own merit which could be added to DPDK as
> >> PMD-specific pattern items/actions [2]. mlx5 doesn't have any yet,
> >> but it's pretty common for PMDs to expose a public header that
> >> dedicated applications can include to use this kind of features (look for
> rte_pmd_*.h, e.g. rte_pmd_ixgbe.h).
> >> No problem with that.
> >
> > That's good info. Thanks. But still considering
> > connection-tracking-like use-cases, this transient storage on multi-table
> flow pipeline is quite useful.
> >
> >
> > thanks,
> > Yongseok
> >
> >> [1] "[PATCH] ethdev: extend flow metadata"
> >>
> >>
> https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fmai
> >> ls.dpdk.org%2Farchives%2Fdev%2F2019-
> July%2F137305.html&amp;data=02%7C
> >>
> 01%7Cviacheslavo%40mellanox.com%7Cc0402133b8b2422fc23308d74bef1
> 4fd%7C
> >>
> a652971c7d2e4d9ba6a4d149256f461b%7C0%7C0%7C637061362537116332
> &amp;sda
> >>
> ta=I%2B%2BERHK8FXzLxXkbbjGTmNDf2e%2FsVRvQ%2FIJW4ZmaYrk%3D&a
> mp;reserve
> >> d=0
> >>
> >> [2] "Negative types"
> >>
> >>
> https://eur03.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdoc
> >> .dpdk.org%2Fguides%2Fprog_guide%2Frte_flow.html%23negative-
> types&amp;
> >>
> data=02%7C01%7Cviacheslavo%40mellanox.com%7Cc0402133b8b2422fc23
> 308d74
> >>
> bef14fd%7Ca652971c7d2e4d9ba6a4d149256f461b%7C0%7C0%7C63706136
> 25371163
> >>
> 32&amp;sdata=o6hcNuwWnv9fADGxNcy6S9B0xwCNdlNhbloIKRiMiNo%3D&
> amp;reser
> >> ved=0
> 
> Is this RFC still valid, will there be any follow up?
> If not am marking it as rejected in next a few days.

Yes, RFC is valid, v2 and support in mlx5 Is coming.

WBR, Slava