> -----Original Message-----
> From: Intel-wired-lan <intel-wired-lan-boun...@osuosl.org> On Behalf Of
> Alexander Lobakin
> Sent: Tuesday, August 26, 2025 9:25 PM
> To: intel-wired-...@lists.osuosl.org
> Cc: Lobakin, Aleksander <aleksander.loba...@intel.com>; Kubiak, Michal
> <michal.kub...@intel.com>; Fijalkowski, Maciej
> <maciej.fijalkow...@intel.com>; Nguyen, Anthony L
> <anthony.l.ngu...@intel.com>; Kitszel, Przemyslaw
> <przemyslaw.kits...@intel.com>; Andrew Lunn <andrew+net...@lunn.ch>;
> David S. Miller <da...@davemloft.net>; Eric Dumazet
> <eduma...@google.com>; Jakub Kicinski <k...@kernel.org>; Paolo Abeni
> <pab...@redhat.com>; Alexei Starovoitov <a...@kernel.org>; Daniel
> Borkmann <dan...@iogearbox.net>; Simon Horman <ho...@kernel.org>;
> NXNE CNSE OSDT ITP Upstreaming
> <nxne.cnse.osdt.itp.upstream...@intel.com>; b...@vger.kernel.org;
> net...@vger.kernel.org; linux-ker...@vger.kernel.org
> Subject: [Intel-wired-lan] [PATCH iwl-next v5 02/13] idpf: fix Rx descriptor
> ready check barrier in splitq
>
> No idea what the current barrier position was meant for. At that point,
> nothing is read from the descriptor, only the pointer to the actual one is
> fetched.
> The correct barrier usage here is after the generation check, so that only the
> first qword is read if the descriptor is not yet ready and we need to stop
> polling. Debatable on coherent DMA as the Rx descriptor size is <= cacheline
> size, but anyway, the current barrier position only makes the codegen worse.
>
> Fixes: 3a8845af66ed ("idpf: add RX splitq napi poll support")
> Reviewed-by: Maciej Fijalkowski <maciej.fijalkow...@intel.com>
> Signed-off-by: Alexander Lobakin <aleksander.loba...@intel.com>
> ---
> drivers/net/ethernet/intel/idpf/idpf_txrx.c | 8 ++------
> 1 file changed, 2 insertions(+), 6 deletions(-)
>
Tested-by: R,Ramu <ram...@intel.com>
