netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Aaron Conole <aconole@redhat.com>
To: Nicholas Piggin <npiggin@gmail.com>
Cc: netdev@vger.kernel.org,  dev@openvswitch.org,
	 Pravin B Shelar <pshelar@ovn.org>,
	 "Eelco Chaudron" <echaudro@redhat.com>,
	 "Ilya Maximets" <imaximet@redhat.com>,
	 "Flavio Leitner" <fbl@redhat.com>,
	Paolo Abeni <pabeni@redhat.com>, Jakub Kicinski <kuba@kernel.org>,
	"David S. Miller" <davem@davemloft.net>,
	Eric Dumazet <edumazet@google.com>
Subject: Re: [PATCH 0/7] net: openvswitch: Reduce stack usage
Date: Wed, 11 Oct 2023 09:23:18 -0400	[thread overview]
Message-ID: <f7ta5spe1ix.fsf@redhat.com> (raw)
In-Reply-To: <20231011034344.104398-1-npiggin@gmail.com> (Nicholas Piggin's message of "Wed, 11 Oct 2023 13:43:37 +1000")

Nicholas Piggin <npiggin@gmail.com> writes:

> Hi,
>
> I'll post this out again to keep discussion going. Thanks all for the
> testing and comments so far.

Thanks for the update - did you mean for this to be tagged RFC as well?

I don't see any performance data with the deployments on x86_64 and
ppc64le that cause these stack overflows.  Are you able to provide the
impact on ppc64le and x86_64?

I guess the change probably should be tagged as -next since it doesn't
really have a specific set of commits it is "fixing."  It's really like
a major change and shouldn't really go through stable trees, but I'll
let the maintainers tell me off if I got it wrong.

> Changes since the RFC
> https://lore.kernel.org/netdev/20230927001308.749910-1-npiggin@gmail.com/
>
> - Replace slab allocations for flow keys with expanding the use
>   of the per-CPU key allocator to ovs_vport_receive.
>
> - Drop patch 1 with Ilya's since they did the same thing (that is
>   added at patch 3).
>
> - Change push_nsh stack reduction from slab allocation to per-cpu
>   buffer.
>
> - Drop the ovs_fragment stack usage reduction for now sinc it used
>   slab and was a bit more complicated.
>
> I posted an initial version of the per-cpu flow allocator patch in
> the RFC thread. Since then I cleaned up some debug code and increased
> the allocator size to accommodate the additional user of it.
>
> Thanks,
> Nick
>
> Ilya Maximets (1):
>   openvswitch: reduce stack usage in do_execute_actions
>
> Nicholas Piggin (6):
>   net: openvswitch: generalise the per-cpu flow key allocation stack
>   net: openvswitch: Use flow key allocator in ovs_vport_receive
>   net: openvswitch: Reduce push_nsh stack usage
>   net: openvswitch: uninline action execution
>   net: openvswitch: uninline ovs_fragment to control stack usage
>   net: openvswitch: Reduce stack usage in ovs_dp_process_packet
>
>  net/openvswitch/actions.c  | 208 +++++++++++++++++++++++--------------
>  net/openvswitch/datapath.c |  56 +++++-----
>  net/openvswitch/flow.h     |   3 +
>  net/openvswitch/vport.c    |  27 +++--
>  4 files changed, 185 insertions(+), 109 deletions(-)


  parent reply	other threads:[~2023-10-11 13:23 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-10-11  3:43 [PATCH 0/7] net: openvswitch: Reduce stack usage Nicholas Piggin
2023-10-11  3:43 ` [PATCH 1/7] net: openvswitch: generalise the per-cpu flow key allocation stack Nicholas Piggin
2023-10-11  3:43 ` [PATCH 2/7] net: openvswitch: Use flow key allocator in ovs_vport_receive Nicholas Piggin
2023-10-11  3:43 ` [PATCH 3/7] openvswitch: reduce stack usage in do_execute_actions Nicholas Piggin
2023-10-11  3:43 ` [PATCH 4/7] net: openvswitch: Reduce push_nsh stack usage Nicholas Piggin
2023-10-11  3:43 ` [PATCH 5/7] net: openvswitch: uninline action execution Nicholas Piggin
2023-10-11  3:43 ` [PATCH 6/7] net: openvswitch: uninline ovs_fragment to control stack usage Nicholas Piggin
2023-10-11  3:43 ` [PATCH 7/7] net: openvswitch: Reduce stack usage in ovs_dp_process_packet Nicholas Piggin
2023-10-11 12:22 ` [PATCH 0/7] net: openvswitch: Reduce stack usage Ilya Maximets
2023-10-12  0:08   ` Nicholas Piggin
2023-10-11 13:23 ` Aaron Conole [this message]
2023-10-12  1:19   ` Nicholas Piggin
2023-10-13  8:27     ` David Laight
2023-10-20 17:04     ` Aaron Conole
2023-10-25  4:06       ` Nicholas Piggin

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=f7ta5spe1ix.fsf@redhat.com \
    --to=aconole@redhat.com \
    --cc=davem@davemloft.net \
    --cc=dev@openvswitch.org \
    --cc=echaudro@redhat.com \
    --cc=edumazet@google.com \
    --cc=fbl@redhat.com \
    --cc=imaximet@redhat.com \
    --cc=kuba@kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=npiggin@gmail.com \
    --cc=pabeni@redhat.com \
    --cc=pshelar@ovn.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
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;
as well as URLs for NNTP newsgroup(s).