netdev - Re: [PATCH net-next 06/11] netfilter: nf_flow_table: count and limit hw offloaded entries

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <YodG+REOiDa2PMUl@salvia>
Date:   Fri, 20 May 2022 09:44:57 +0200
From:   Pablo Neira Ayuso <pablo@...filter.org>
To:     Jakub Kicinski <kuba@...nel.org>
Cc:     netfilter-devel@...r.kernel.org, davem@...emloft.net,
        netdev@...r.kernel.org, pabeni@...hat.com,
        Felix Fietkau <nbd@....name>
Subject: Re: [PATCH net-next 06/11] netfilter: nf_flow_table: count and limit
 hw offloaded entries

On Thu, May 19, 2022 at 04:11:36PM -0700, Jakub Kicinski wrote:
> On Fri, 20 May 2022 00:02:01 +0200 Pablo Neira Ayuso wrote:
> > To improve hardware offload debuggability and scalability introduce
> > 'nf_flowtable_count_hw' and 'nf_flowtable_max_hw' sysctl entries in new
> > dedicated 'net/netfilter/ft' namespace. Add new pernet struct nf_ft_net in
> > order to store the counter and sysctl header of new sysctl table.
> > 
> > Count the offloaded flows in workqueue add task handler. Verify that
> > offloaded flow total is lower than allowed maximum before calling the
> > driver callbacks. To prevent spamming the 'add' workqueue with tasks when
> > flows can't be offloaded anymore also check that count is below limit
> > before queuing offload work. This doesn't prevent all redundant workqueue
> > task since counter can be taken by concurrent work handler after the check
> > had been performed but before the offload job is executed but it still
> > greatly reduces such occurrences. Note that flows that were not offloaded
> > due to counter being larger than the cap can still be offloaded via refresh
> > function.
> > 
> > Ensure that flows are accounted correctly by verifying IPS_HW_OFFLOAD_BIT
> > value before counting them. This ensures that add/refresh code path
> > increments the counter exactly once per flow when setting the bit and
> > decrements it only for accounted flows when deleting the flow with the bit
> > set.
> 
> Why a sysctl and not a netlink attr per table or per device?

Per-device is not an option, because the flowtable represents a
compound of devices.

Moreover, in tc ct act the flowtable is not bound to a device, while
in netfilter/nf_tables it is.

tc ct act does not expose flowtables to userspace in any way, they
internally allocate one flowtable per zone. I assume there os no good
netlink interface for them.

For netfilter/nftables, it should be possible to add per-flowtable
netlink attributes, my plan is to extend the flowtable netlink
attribute to add a flowtable maximum size.

This sysctl count and limit hw will just work as a global limit (which
is optional), my plan is that the upcoming per-flowtable limit will
just override this global limit.

I think it is a reasonable tradeoff for the different requirements of
the flowtable infrastructure users given there are two clients
currently for this code.