[Intel-wired-lan] [PATCH iwl-next v2 3/3] igb: allow configuring RSS key via ethtool set_rxfh

[Loktionov, Aleksandr]

> -----Original Message-----
> From: Intel-wired-lan <intel-wired-lan-bounces at osuosl.org> On Behalf
> Of Loktionov, Aleksandr
> Sent: Thursday, January 8, 2026 1:28 PM
> To: Kohei Enju <enjuk at amazon.com>
> Cc: andrew+netdev at lunn.ch; Nguyen, Anthony L
> <anthony.l.nguyen at intel.com>; davem at davemloft.net;
> edumazet at google.com; intel-wired-lan at lists.osuosl.org;
> kuba at kernel.org; netdev at vger.kernel.org; pabeni at redhat.com; Kitszel,
> Przemyslaw <przemyslaw.kitszel at intel.com>; takkozu at amazon.com
> Subject: Re: [Intel-wired-lan] [PATCH iwl-next v2 3/3] igb: allow
> configuring RSS key via ethtool set_rxfh
> 
> 
> 
> > -----Original Message-----
> > From: Kohei Enju <enjuk at amazon.com>
> > Sent: Thursday, January 8, 2026 1:04 PM
> > To: Loktionov, Aleksandr <aleksandr.loktionov at intel.com>
> > Cc: andrew+netdev at lunn.ch; Nguyen, Anthony L
> > <anthony.l.nguyen at intel.com>; davem at davemloft.net;
> > edumazet at google.com; enjuk at amazon.com; intel-wired-
> > lan at lists.osuosl.org; kuba at kernel.org; netdev at vger.kernel.org;
> > pabeni at redhat.com; Kitszel, Przemyslaw
> <przemyslaw.kitszel at intel.com>;
> > takkozu at amazon.com
> > Subject: Re: RE: [Intel-wired-lan] [PATCH iwl-next v2 3/3] igb:
> allow
> > configuring RSS key via ethtool set_rxfh
> >
> > On Thu, 8 Jan 2026 07:29:19 +0000, Loktionov, Aleksandr wrote:
> >
> > >>
> > >> -	igb_write_rss_indir_tbl(adapter);
> > >> +	if (rxfh->key) {
> > >> +		adapter->has_user_rss_key = true;
> > >> +		memcpy(adapter->rss_key, rxfh->key, sizeof(adapter-
> > >> >rss_key));
> > >> +		igb_write_rss_key(adapter);
> > >It leads to race between ethtool RSS update and concurrent resets.
> > >Because igb_setup_mrqc() (called during resets) also calls
> > igb_write_rss_key(adapter).
> > >Non-fatal but breaks RSS configuration guarantees.
> >
> > At my first glance, rtnl lock serializes those operation, so it
> > doesn't seem to be racy as long as they are under the rtnl lock.
> >
> > As far as I skimmed the codes, functions such as igb_open()/
> > igb_up()/igb_reset_task(), which finally call igb_write_rss_key()
> are
> > serialized by rtnl lock or serializes igb_write_rss_key() call by
> > locking rtnl.
> >
> > Please let me know if I'm missing something and it's truly racy.
> I think you're right, and I've missed that missing rtnl_lock was added
> in upstream.
> 
> Thank you for clarification
> Reviewed-by: Aleksandr Loktionov <aleksandr.loktionov at intel.com>
> 

Afterthought, I think it could be nice to place ASSERT_RTNL() to show it explicitly.
What do you think about this idea?