Re: Flows! Offload them.

On 02/27/2015 12:53 AM, Jiri Pirko wrote:

Thu, Feb 26, 2015 at 10:11:23PM CET, john.r.fastabend@intel.com wrote:

quoted

On 02/26/2015 12:16 PM, Neil Horman wrote:

quoted

On Thu, Feb 26, 2015 at 07:23:36AM -0800, John Fastabend wrote:

quoted

On 02/26/2015 05:33 AM, Thomas Graf wrote:

quoted

On 02/26/15 at 10:16am, Jiri Pirko wrote:

quoted

Well, on netdev01, I believe that a consensus was reached that for every
switch offloaded functionality there has to be an implementation in
kernel.

Agreed. This should not prevent the policy being driven from user
space though.

quoted

What John's Flow API originally did was to provide a way to
configure hardware independently of kernel. So the right way is to
configure kernel and, if hw allows it, to offload the configuration to hw.

In this case, seems to me logical to offload from one place, that being
TC. The reason is, as I stated above, the possible conversion from OVS
datapath to TC.

Offloading of TC definitely makes a lot of sense. I think that even in
that case you will already encounter independent configuration of
hardware and kernel. Example: The hardware provides a fixed, generic
function to push up to n bytes onto a packet. This hardware function
could be used to implement TC actions "push_vlan", "push_vxlan",
"push_mpls". You would you would likely agree that TC should make use
of such a function even if the hardware version is different from the
software version. So I don't think we'll have a 1:1 mapping for all
configurations, regardless of whether the how is decided in kernel or
user space.

Just to expand slightly on this. I don't think you can get to a 1:1
mapping here. One reason is hardware typically has a TCAM and limited
size. So you need a _policy_ to determine when to push rules into the
hardware. The kernel doesn't know when to do this and I don't believe
its the kernel's place to start enforcing policy like this. One thing I likely
need to do is get some more "worlds" in rocker so we aren't stuck only
thinking about the infinite size OF_DPA world. The OF_DPA world is only
one world and not a terribly flexible one at that when compared with the
NPU folk. So minimally you need a flag to indicate rules go into hardware
vs software.

That said I think the bigger mismatch between software and hardware is
you program it differently because the data structures are different. Maybe
a u32 example would help. For parsing with u32 you might build a parse
graph with a root and some leaf nodes. In hardware you want to collapse
this down onto the hardware. I argue this is not a kernel task because
there are lots of ways to do this and there are trade-offs made with
respect to space and performance and which table to use when it could be
handled by a set of tables. Another example is a virtual switch possibly
OVS but we have others. The software does some "unmasking" (there term)
before sending the rules into the software dataplane cache. Basically this
means we can ignore priority in the hash lookup. However this is not how you
would optimally use hardware. Maybe I should do another write up with
some more concrete examples.

There are also lots of use cases to _not_ have hardware and software in
sync. A flag allows this.

My only point is I think we need to allow users to optimally use there
hardware either via 'tc' or my previous 'flow' tool. Actually in my
opinion I still think its best to have both interfaces.

I'll go get some coffee now and hopefully that is somewhat clear.

I've been thinking about the policy apect of this, and the more I think about
it, the more I wonder if not allowing some sort of common policy in the kernel
is really the right thing to do here.  I know thats somewhat blasphemous, but
this isn't really administrative poilcy that we're talking about, at least not
100%.  Its more of a behavioral profile that we're trying to enforce.  That may
be splitting hairs, but I think theres precidence for the latter.  That is to
say, we configure qdiscs to limit traffic flow to certain rates, and configure
policies which drop traffic that violates it (which includes random discard,
which is the antithesis of deterministic policy).  I'm not sure I see this as
any different, espcially if we limit its scope.  That is to say, why couldn't we
allow the kernel to program a predetermined set of policies that the admin can
set (i.e. offload routing to a hardware cache of X size with an lru
victimization).  If other well defined policies make sense, we can add them and
exposes options via iproute2 or some such to set them.  For the use case where
such pre-packaged policies don't make sense, we have things like the flow api to
offer users who want to be able to control their hardware in a more fine grained
approach.

Neil

Hi Neil,

I actually like this idea a lot. I might tweak a bit in that we could have some
feature bits or something like feature bits that expose how to split up the
hardware cache and give sizes.

So the hypervisor (see I think of end hosts) or administrators could come in and
say I want a route table and a nft table. This creates a "flavor" over how the
hardware is going to be used. Another use case may not be doing routing at all
but have an application that wants to manage the hardware at a more fine grained
level with the exception of some nft commands so it could have a "nft"+"flow"
flavor. Insert your favorite use case here.

I'm not sure I understand. You said that admin could say: "I want a
route table and a nft table". But how does he say it? Isn't is enough
just to insert some rules into these 2 things and that would give hw a
clue what the admin is doing and what he wants? I believe that this
offload should happen transparently.

Of course, you may want to balance resources as you said the hw capacity
is limited. But I would leave that optional. API unknown so far...

You can leave it optional and queue off the rule inserts but its not
very robust if the hardware capacity is smaller than the software
pipeline. This is the case for some of the setups we would like to get
working on Linux. In this case you can't have the transparent offload
because 'tc' or whatever may load a large set of irrelevant rules into
the hardware pipeline consuming the resources for flows that are control
traffic or exception cases. Perhaps worse what gets put into the
hardware is not very predictable.

So I agree we can support both cases we just need the API/feature bits
to turn it on/off. As far as the API I was thinking a "create" table
API would be sufficient. The kernel could request a "tc" table of size
1k and a "route" table of 4k for example then user space applications
could manage any un-reserved tables. I'm experimenting a bit with how
to make this work now. Maybe "create" should be "reserve".

.John

--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

-- 
John Fastabend         Intel Corporation