On Fri, Jul 25, 2025 at 12:49:09PM +0000, Ciara Loftus wrote:
> In the interest of simplicity, let the primary process select the Rx
> path to be used by all processes using the given device.
>
> The many logs which report individual Rx path selections have been
> consolidated into one single log
>
> Signed-off-by: Ciara Loftus <ciara.lof...@intel.com>
> ---
> drivers/net/intel/i40e/i40e_ethdev.h | 20 +++-
> drivers/net/intel/i40e/i40e_rxtx.c | 168 ++++++++++++---------------
> 2 files changed, 93 insertions(+), 95 deletions(-)
>
> diff --git a/drivers/net/intel/i40e/i40e_ethdev.h
> b/drivers/net/intel/i40e/i40e_ethdev.h
> index 44864292d0..308039c363 100644
> --- a/drivers/net/intel/i40e/i40e_ethdev.h
> +++ b/drivers/net/intel/i40e/i40e_ethdev.h
> @@ -1226,6 +1226,22 @@ struct i40e_vsi_vlan_pvid_info {
> #define I40E_MBUF_CHECK_F_TX_SEGMENT (1ULL << 2)
> #define I40E_MBUF_CHECK_F_TX_OFFLOAD (1ULL << 3)
>
> +enum i40e_rx_func_type {
> + I40E_RX_DEFAULT,
> + I40E_RX_BULK_ALLOC,
> + I40E_RX_SCATTERED,
> + I40E_RX_SSE,
> + I40E_RX_AVX2,
> + I40E_RX_SSE_SCATTERED,
> + I40E_RX_AVX2_SCATTERED,
> + I40E_RX_AVX512,
> + I40E_RX_AVX512_SCATTERED,
> + I40E_RX_NEON,
> + I40E_RX_NEON_SCATTERED,
> + I40E_RX_ALTIVEC,
> + I40E_RX_ALTIVEC_SCATTERED,
> +};
> +
As with previous patch, I'd suggest working out what the most logical order
for these elements is, and using that. Having AVX2 and SSE interleaved
seems a bit strange.
Otherwise:
Acked-by: Bruce Richardson <bruce.richard...@intel.com>
