[PATCH RFC 0/9] sk_buff: optimize layout for GRO

[Paul Moore]

On Sat, Jul 24, 2021 at 2:51 PM Florian Westphal <fw at strlen.de> wrote:
> Paul Moore <paul at paul-moore.com> wrote:
>  > Tow main drivers on my side:
> > > - there are use cases/deployments that do not use them.
> > > - moving them around was doable in term of required changes.
> > >
> > > There are no "slow-path" implications on my side. For example, vlan_*
> > > fields are very critical performance wise, if the traffic is tagged.
> > > But surely there are busy servers not using tagget traffic which will
> > > enjoy the reduced cachelines footprint, and this changeset will not
> > > impact negatively the first case.
> > >
> > > WRT to the vlan example, secmark and nfct require an extra conditional
> > > to fetch the data. My understanding is that such additional conditional
> > > is not measurable performance-wise when benchmarking the security
> > > modules (or conntrack) because they have to do much more intersting
> > > things after fetching a few bytes from an already hot cacheline.
> > >
> > > Not sure if the above somehow clarify my statements.
> > >
> > > As for expanding secmark to 64 bits, I guess that could be an
> > > interesting follow-up discussion :)
> >
> > The intersection between netdev and the LSM has a long and somewhat
> > tortured past with each party making sacrifices along the way to get
> > where we are at today.  It is far from perfect, at least from a LSM
> > perspective, but it is what we've got and since performance is usually
> > used as a club to beat back any changes proposed by the LSM side, I
> > would like to object to these changes that negatively impact the LSM
> > performance without some concession in return.  It has been a while
> > since Casey and I have spoken about this, but I think the prefered
> > option would be to exchange the current __u32 "sk_buff.secmark" field
> > with a void* "sk_buff.security" field, like so many other kernel level
> > objects.  Previous objections have eventually boiled down to the
> > additional space in the sk_buff for the extra bits (there is some
> > additional editorializing that could be done here, but I'll refrain),
> > but based on the comments thus far in this thread it sounds like
> > perhaps we can now make a deal here: move the LSM field down to a
> > "colder" cacheline in exchange for converting the LSM field to a
> > proper pointer.
> >
> > Thoughts?
>
> Is there a summary disucssion somewhere wrt. what exactly LSMs need?

My network access is limited for the next week so I don't have the
ability to dig through the list archives, but if you look through the
netdev/LSM/lists over the past decade (maybe go back ~15 years?) you
will see multiple instances where we/I've brought up different
solutions with the netdev folks only to hit a brick wall.  The LSM ask
for sk_buff is really the same as any other kernel object that we want
to control with LSM access controls, e.g. inodes; we basically want a
void* blob with the necessary hooks so that the opaque blob can be
managed through the skb's lifetime.

> There is the skb extension infra, does that work for you?

I was hopeful that when the skb_ext capability was introduced we might
be able to use it for the LSM(s), but when I asked netdev if they
would be willing to accept patches to leverage the skb_ext
infrastructure I was told "no".

-- 
paul moore
www.paul-moore.com