netdev - Re: [PATCH intel-net 1/2] ice: xsk: change batched Tx descriptor cleaning

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]

Date:   Tue, 30 Aug 2022 14:43:50 +0200
From:   Maciej Fijalkowski <maciej.fijalkowski@...el.com>
To:     <intel-wired-lan@...ts.osuosl.org>
CC:     <netdev@...r.kernel.org>, <bpf@...r.kernel.org>,
        <anthony.l.nguyen@...el.com>, <magnus.karlsson@...el.com>
Subject: Re: [PATCH intel-net 1/2] ice: xsk: change batched Tx descriptor
 cleaning

On Tue, Aug 30, 2022 at 02:38:02PM +0200, Maciej Fijalkowski wrote:
> AF_XDP Tx descriptor cleaning in ice driver currently works in a "lazy"
> way - descriptors are not cleaned immediately after send. We rather hold
> on with cleaning until we see that free space in ring drops below
> particular threshold. This was supposed to reduce the amount of
> unnecessary work related to cleaning and instead of keeping the ring
> empty, ring was rather saturated.
> 
> In AF_XDP realm cleaning Tx descriptors implies producing them to CQ.
> This is a way of letting know user space that particular descriptor has
> been sent, as John points out in [0].
> 
> We tried to implement serial descriptor cleaning which would be used in
> conjunction with batched cleaning but it made code base more convoluted
> and probably harder to maintain in future. Therefore we step away from
> batched cleaning in a current form in favor of an approach where we set
> RS bit on every last descriptor from a batch and clean always at the
> beginning of ice_xmit_zc().
> 
> This means that we give up a bit of Tx performance, but this doesn't
> hurt l2fwd scenario due to the fact that Tx side is much faster than Rx
> and Rx is the one that has to catch Tx up. txonly can be treaten as
> AF_XDP based packet generator.
> 
> FWIW Tx descriptors are still produced in a batched way.
> 
> Fixes: 126cdfe1007a ("ice: xsk: Improve AF_XDP ZC Tx and use batching API")
> [0]: https://lore.kernel.org/bpf/62b0a20232920_3573208ab@john.notmuch/
> 
> Fixes: 126cdfe1007a ("ice: xsk: Improve AF_XDP ZC Tx and use batching API")

Ugh too much of a fixes tag. I'll send a v2.

> Signed-off-by: Maciej Fijalkowski <maciej.fijalkowski@...el.com>
> ---
>  drivers/net/ethernet/intel/ice/ice_txrx.c |   2 +-
>  drivers/net/ethernet/intel/ice/ice_xsk.c  | 143 +++++++++-------------
>  drivers/net/ethernet/intel/ice/ice_xsk.h  |   7 +-
>  3 files changed, 64 insertions(+), 88 deletions(-)
>