On Thu, 22 Jan 2026 07:33:23 +0000, "Loktionov, Aleksandr" wrote:
> > -----Original Message----- > > From: Intel-wired-lan <[email protected]> On Behalf > > Of Kohei Enju > > Sent: Monday, January 19, 2026 6:58 PM > > To: [email protected]; [email protected] > > Cc: Nguyen, Anthony L <[email protected]>; Kitszel, > > Przemyslaw <[email protected]>; Andrew Lunn > > <[email protected]>; David S. Miller <[email protected]>; Eric > > Dumazet <[email protected]>; Jakub Kicinski <[email protected]>; Paolo > > Abeni <[email protected]>; [email protected]; Takashi Kozu > > <[email protected]>; Kohei Enju <[email protected]> > > Subject: [Intel-wired-lan] [PATCH v1 iwl-next] igb: set skb hash type > > from RSS_TYPE > >=20 > > igb always marks the RX hash as L3 regardless of RSS_TYPE in the > > advanced descriptor, which may indicate L4 (TCP/UDP) hash. This can > > trigger unnecessary SW hash recalculation and breaks toeplitz > > selftests. > >=20 > > Use RSS_TYPE from pkt_info to set the correct PKT_HASH_TYPE_* > >=20 > > Tested by toeplitz.py with the igb RSS key get/set patches applied as > > they are required for toeplitz.py (see Link below). > > # ethtool -N $DEV rx-flow-hash udp4 sdfn # ethtool -N $DEV rx-flow- > > hash udp6 sdfn # python toeplitz.py | grep -E "^# Totals" > >=20 > > Without patch: > > # Totals: pass:0 fail:12 xfail:0 xpass:0 skip:0 error:0 > >=20 > > With patch: > > # Totals: pass:12 fail:0 xfail:0 xpass:0 skip:0 error:0 > >=20 > > Link: https://lore.kernel.org/intel-wired-lan/20260119084511.95287-5- > > [email protected]/ > > Signed-off-by: Kohei Enju <[email protected]> > > --- > > If a Fixes tag is needed, it would be Fixes: 42bdf083fe70 ("net: igb > > calls skb_set_hash"). > >=20 > > I'm not sure this qualifies as a fix, since the RX hash marking has > > been wrong for a long time without reported issues. So I'm leaning > > toward omitting Fixes. > > --- > > drivers/net/ethernet/intel/igb/e1000_82575.h | 21 > > ++++++++++++++++++++ > > drivers/net/ethernet/intel/igb/igb_main.c | 18 +++++++++++++---- > > 2 files changed, 35 insertions(+), 4 deletions(-) > >=20 > > diff --git a/drivers/net/ethernet/intel/igb/e1000_82575.h > > b/drivers/net/ethernet/intel/igb/e1000_82575.h > > index 63ec253ac788..a855bc10f5d4 100644 > > --- a/drivers/net/ethernet/intel/igb/e1000_82575.h > > +++ b/drivers/net/ethernet/intel/igb/e1000_82575.h > > @@ -87,6 +87,27 @@ union e1000_adv_rx_desc { > > } wb; /* writeback */ > > }; > >=20 > > +#define E1000_RSS_TYPE_NO_HASH 0 > > +#define E1000_RSS_TYPE_HASH_TCP_IPV4 1 > > +#define E1000_RSS_TYPE_HASH_IPV4 2 > > +#define E1000_RSS_TYPE_HASH_TCP_IPV6 3 > > +#define E1000_RSS_TYPE_HASH_IPV6_EX 4 > > +#define E1000_RSS_TYPE_HASH_IPV6 5 > > +#define E1000_RSS_TYPE_HASH_TCP_IPV6_EX 6 > > +#define E1000_RSS_TYPE_HASH_UDP_IPV4 7 > > +#define E1000_RSS_TYPE_HASH_UDP_IPV6 8 > > +#define E1000_RSS_TYPE_HASH_UDP_IPV6_EX 9 > > + > > +#define E1000_RSS_TYPE_MASK GENMASK(3, 0) /* 4-bits (3:0) > > =3D mask 0x0F */ > 80columns violation, it's not critical, but I'd suggest just remove redunda= > nt comment > Otherwise, it smells like AI generated... > #define E1000_RSS_TYPE_MASK GENMASK(3, 0) Ugh, somehow I missed the warning in checkpatch.pl. Thank you for poiting out. TBH I copied that line from igc/igc.h with the comment, but the comment is not necessary :) I'll fix it in v2. > > Everything else looks fine. > Reviewed-by: Aleksandr Loktionov <[email protected]> >
