From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jakub Kicinski Subject: Re: [PATCH bpf-next 0/4] i40e AF_XDP zero-copy buffer leak fixes Date: Wed, 5 Sep 2018 19:14:37 +0200 Message-ID: <20180905191437.35f7d049@cakuba> References: <20180904181105.10983-1-bjorn.topel@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Cc: ast@kernel.org, daniel@iogearbox.net, netdev@vger.kernel.org, jeffrey.t.kirsher@intel.com, intel-wired-lan@lists.osuosl.org, =?UTF-8?B?QmrDtnJuIFTDtnBlbA==?= , magnus.karlsson@intel.com, magnus.karlsson@gmail.com To: =?UTF-8?B?QmrDtnJuIFTDtnBlbA==?= Return-path: Received: from mail-pg1-f196.google.com ([209.85.215.196]:32800 "EHLO mail-pg1-f196.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1726046AbeIEVpx (ORCPT ); Wed, 5 Sep 2018 17:45:53 -0400 Received: by mail-pg1-f196.google.com with SMTP id s7-v6so3771559pgc.0 for ; Wed, 05 Sep 2018 10:14:46 -0700 (PDT) In-Reply-To: <20180904181105.10983-1-bjorn.topel@gmail.com> Sender: netdev-owner@vger.kernel.org List-ID: On Tue, 4 Sep 2018 20:11:01 +0200, Bj=C3=B6rn T=C3=B6pel wrote: > From: Bj=C3=B6rn T=C3=B6pel >=20 > This series addresses an AF_XDP zero-copy issue that buffers passed > from userspace to the kernel was leaked when the hardware descriptor > ring was torn down. >=20 > The patches fixes the i40e AF_XDP zero-copy implementation. >=20 > Thanks to Jakub Kicinski for pointing this out! >=20 > Some background for folks that don't know the details: A zero-copy > capable driver picks buffers off the fill ring and places them on the > hardware Rx ring to be completed at a later point when DMA is > complete. Similar on the Tx side; The driver picks buffers off the Tx > ring and places them on the Tx hardware ring. >=20 > In the typical flow, the Rx buffer will be placed onto an Rx ring > (completed to the user), and the Tx buffer will be placed on the > completion ring to notify the user that the transfer is done. >=20 > However, if the driver needs to tear down the hardware rings for some > reason (interface goes down, reconfiguration and such), the userspace > buffers cannot be leaked. They have to be reused or completed back to > userspace. >=20 > The implementation does the following: >=20 > * Outstanding Tx descriptors will be passed to the completion > ring. The Tx code has back-pressure mechanism in place, so that > enough empty space in the completion ring is guaranteed. >=20 > * Outstanding Rx descriptors are temporarily stored on a stash/reuse > queue. The reuse queue is based on Jakub's RFC. When/if the HW rings > comes up again, entries from the stash are used to re-populate the > ring. >=20 > * When AF_XDP ZC is enabled, disallow changing the number of hardware > descriptors via ethtool. Otherwise, the size of the stash/reuse > queue can grow unbounded. >=20 > Going forward, introducing a "zero-copy allocator" analogous to Jesper > Brouer's page pool would be a more robust and reuseable solution. >=20 > Jakub: I've made a minor checkpatch-fix to your RFC, prior adding it > into this series. Thanks for the fix! :) Out of curiosity, did checking the reuse queue have a noticeable impact in your test (i.e. always using the _rq() helpers)? You seem to be adding an indirect call, would that not be way worse on a retpoline kernel?