From: roopa <roopa@cumulusnetworks•com>
To: John Fastabend <john.fastabend@gmail•com>
Cc: Scott Feldman <sfeldma@gmail•com>, Jiri Pirko <jiri@resnulli•us>,
David Miller <davem@davemloft•net>,
Netdev <netdev@vger•kernel.org>,
Andy Gospodarek <andy@greyhouse•net>
Subject: Re: [PATCH net v2] switchdev: don't abort hardware ipv4 fib offload on failure to program fib entry in hardware
Date: Thu, 28 May 2015 22:37:14 -0700 [thread overview]
Message-ID: <5567FB0A.1060900@cumulusnetworks.com> (raw)
In-Reply-To: <55673DF5.7060401@gmail.com>
On 5/28/15, 9:10 AM, John Fastabend wrote:
> On 05/28/2015 08:40 AM, Scott Feldman wrote:
>> On Thu, May 28, 2015 at 2:42 AM, Jiri Pirko <jiri@resnulli•us> wrote:
>>> Mon, May 18, 2015 at 10:19:16PM CEST, davem@davemloft•net wrote:
>>>> From: Roopa Prabhu <roopa@cumulusnetworks•com>
>>>> Date: Sun, 17 May 2015 16:42:05 -0700
>>>>
>>>>> On most systems where you can offload routes to hardware,
>>>>> doing routing in software is not an option (the cpu limitations
>>>>> make routing impossible in software).
>>>>
>>>> You absolutely do not get to determine this policy, none of us
>>>> do.
>>>>
>>>> What matters is that by default the damn switch device being there
>>>> is %100 transparent to the user.
>>>>
>>>> And the way to achieve that default is to do software routes as
>>>> a fallback.
>>>>
>>>> I am not going to entertain changes of this nature which fail
>>>> route loading by default just because we've exceeded a device's
>>>> HW capacity to offload.
>>>>
>>>> I thought I was _really_ clear about this at netdev 0.1
>>>
>>> I certainly agree that by default, transparency 1:1 sw:hw mapping is
>>> what we need for fib. The current code is a good start!
>>>
>>> I see couple of issues regarding switchdev_fib_ipv4_abort:
>>> 1) If user adds and entry, switchdev_fib_ipv4_add fails, abort is
>>> executed -> and, error returned. I would expect that route entry
>>> should
>>> be added in this case. The next attempt of adding the same entry
>>> will
>>> be successful.
>>> The current behaviour breaks the transparency you are reffering to.
>>> 2) When switchdev_fib_ipv4_abort happens to be executed, the offload is
>>> disabled for good (until reboot). That is certainly not nice,
>>> alhough
>>> I understand that is the easiest solution for now.
>>>
>>> I believe that we all agree that the 1:1 transparency, although it is a
>>> default, may not be optimal for real-life usage. HW resources are
>>> limited and user does not know them. The danger of hitting _abort and
>>> screwing-up the whole system is huge, unacceptable.
>>>
>>> So here, there are couple of more or less simple things that I
>>> suggest to
>>> do in order to move a little bit forward:
>>> 1) Introduce system-wide option to switch _abort to just plain fail.
>>> When HW does not have capacity, do not flush and fallback to sw,
>>> but
>>> rather just fail to add the entry. This would not break anything.
>>> Userspace has to be prepared that entry add could fail.
>>> 2) Introduce a way to propagate resources to userspace. Driver knows
>>> about
>>> resources used/available/potentially_available. Switchdev infra
>>> could
>>> be extended in order to propagate the info to the user.
>>> 3) Introduce couple of flags for entry add that would alter the default
>>> behaviour. Something like:
>>> NLM_F_SKIP_KERNEL
>>> NLM_F_SKIP_OFFLOAD
>>> Again, this does not break the current users. On the other hand,
>>> this
>>> gives new users a leverage to instruct kernel where the entry
>>> should
>>> be added to (or not added to).
>>>
>>> Any thoughts? Objections?
>>
>> I don't like these. Breaks transparency and forces the user in a
>> position of having to know hardware failures modes (unique to each
>> hardware device). I presented an option d) which avoids this issues;
>> was it not understood?
>>
>
> Hi Scott,
>
> I understood your proposal. One caveat I had is in response to this,
>
> "Actually, now that I think of it, the device/driver could decide which
> related-prefix to evict from HW, if driver/device wanted to have a
> sense of which routes are more important to offload than other"
>
> hardware/driver/device shouldn't have a sense of which routes are more
> important than others.
correct. The routing daemons know this best.
> I think this is where the NLM_F_* flags come in.
> If userspace _wants_ to push policy into the kernel about what is
> important it can. If it doesn't we get a sensible heuristic that does
> a reasonable job offloading rules transparently. This is how we did
> L2 and I think that seems to work fairly well. At least for me but,
> always interested to hear other use cases though.
agree.
>
> Also I guess I'm not seeing the multitude of hardware failure modes. I
> see two either the hardware doesn't support the operation or it is out
> of resources. Both can be learned if the hardware exports a model of its
> capabilities and resources.
agree, A switchdev api to query hardware resource and capability is due.
We can start with rocker. It gives an app the choice to control the policy.
But, for our usecase/deployments today, i am more interested in a
system wide policy because
it is easier on my apps today.
thanks.
next prev parent reply other threads:[~2015-05-29 5:37 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-05-17 23:42 [PATCH net v2] switchdev: don't abort hardware ipv4 fib offload on failure to program fib entry in hardware Roopa Prabhu
2015-05-18 5:11 ` Scott Feldman
2015-05-18 20:19 ` David Miller
2015-05-19 0:21 ` John Fastabend
2015-05-19 3:48 ` David Miller
2015-05-19 5:58 ` roopa
2015-05-19 16:34 ` David Miller
2015-05-19 17:01 ` Jiri Pirko
2015-05-19 19:47 ` Andy Gospodarek
2015-05-19 20:28 ` David Miller
2015-05-20 14:37 ` Andy Gospodarek
2015-05-21 5:46 ` Scott Feldman
2015-05-21 15:37 ` roopa
2015-05-29 7:50 ` Jiri Pirko
2015-05-29 15:39 ` Scott Feldman
2015-05-30 9:00 ` Jiri Pirko
2015-05-31 4:19 ` John Fastabend
2015-05-31 6:34 ` Scott Feldman
2015-05-31 7:34 ` Scott Feldman
2015-05-19 5:57 ` roopa
2015-05-28 9:42 ` Jiri Pirko
2015-05-28 15:35 ` John Fastabend
2015-05-29 7:42 ` Jiri Pirko
2015-05-28 15:40 ` Scott Feldman
2015-05-28 16:10 ` John Fastabend
2015-05-29 5:37 ` roopa [this message]
2015-05-28 22:35 ` Andy Gospodarek
2015-05-29 5:51 ` roopa
2015-05-29 7:50 ` Jiri Pirko
2015-05-29 5:31 ` roopa
2015-05-29 15:12 ` Scott Feldman
2015-05-29 15:37 ` Jiri Pirko
-- strict thread matches above, loose matches on Subject: below --
2015-05-17 3:46 Roopa Prabhu
2015-05-17 23:41 ` roopa
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=5567FB0A.1060900@cumulusnetworks.com \
--to=roopa@cumulusnetworks$(echo .)com \
--cc=andy@greyhouse$(echo .)net \
--cc=davem@davemloft$(echo .)net \
--cc=jiri@resnulli$(echo .)us \
--cc=john.fastabend@gmail$(echo .)com \
--cc=netdev@vger$(echo .)kernel.org \
--cc=sfeldma@gmail$(echo .)com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
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