[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20260128060958.6333-1-kohei@enjuk.jp>
Date: Wed, 28 Jan 2026 06:09:44 +0000
From: Kohei Enju <kohei@...uk.jp>
To: anthony.l.nguyen@...el.com
Cc: aleksandr.loktionov@...el.com, andrew+netdev@...n.ch, davem@...emloft.net,
edumazet@...gle.com, enjuk@...zon.com,
intel-wired-lan@...ts.osuosl.org, kohei@...uk.jp, kuba@...nel.org,
netdev@...r.kernel.org, pabeni@...hat.com, piotr.kwapulinski@...el.com,
pmenzel@...gen.mpg.de, przemyslaw.kitszel@...el.com,
takkozu@...zon.com
Subject: Re: [PATCH iwl-next v4 3/3] igb: allow configuring RSS key via
On Tue, 27 Jan 2026 14:33:24 -0800, Tony Nguyen wrote:
> On 1/25/2026 5:12 AM, Kohei Enju wrote:
> > On Tue, 20 Jan 2026 18:34:40 +0900, Takashi Kozu wrote:
> >
> >> Change igc_set_rxfh() to accept and save a userspace-provided
> >> RSS key. When a key is provided, store it in the adapter and write the
> >> E1000 registers accordingly.
> >>
> >> This can be tested using `ethtool -X <dev> hkey <key>`.
> >>
> >> Signed-off-by: Takashi Kozu <takkozu@...zon.com>
> >> ---
> >> drivers/net/ethernet/intel/igb/igb.h | 1 +
> >> drivers/net/ethernet/intel/igb/igb_ethtool.c | 49 +++++++++++---------
> >> drivers/net/ethernet/intel/igb/igb_main.c | 3 +-
> >> 3 files changed, 30 insertions(+), 23 deletions(-)
> >>
> >> diff --git a/drivers/net/ethernet/intel/igb/igb.h b/drivers/net/ethernet/intel/igb/igb.h
> >> index 8c9b02058cec..2509ec30acf3 100644
> >> --- a/drivers/net/ethernet/intel/igb/igb.h
> >> +++ b/drivers/net/ethernet/intel/igb/igb.h
> >> @@ -657,6 +657,7 @@ struct igb_adapter {
> >> u32 rss_indir_tbl_init;
> >> u8 rss_indir_tbl[IGB_RETA_SIZE];
> >> u8 rss_key[IGB_RSS_KEY_SIZE];
> >> + bool has_user_rss_key;
> >
> > Hi Kozu-san.
> >
> > While preparing for testing, I noticed that now 'has_user_rss_key' is
> > not necessary.
> >
> > Since netdev_rss_key_fill() is called in igb_sw_init() and igb_sw_init()
> > is called only once for the adapter's lifetime, adapter->rss_key
> > wouldn't be changed except for user-intended change.
> >
> > I'd drop that flag and related code (see below)
>
> Hi Kohei,
>
> I believe this igb implementation was based on your igc implementation
> which also has the same. Would it be possible for you to update the igc
> to do this as well?
Yes, you're right. I'd love to do that and post as new version,
so could you drop the series[1] from tnguy/next-queue.git?
[1] https://lore.kernel.org/intel-wired-lan/20251025150136.47618-1-enjuk@amazon.com/
>
> Thanks,
> Tony
Powered by blists - more mailing lists