On Mon, May 12, 2025 at 01:54:39PM +0100, Anatoly Burakov wrote:
> Currently, there are duplicate implementations of Tx mbuf recycle in some
> drivers, specifically ixgbe and i40e. Move them into a common header.
> 
> Signed-off-by: Anatoly Burakov <anatoly.bura...@intel.com>
> ---
>  drivers/net/intel/common/recycle_mbufs.h      | 98 +++++++++++++++++++
>  drivers/net/intel/common/tx.h                 |  1 +
>  .../i40e/i40e_recycle_mbufs_vec_common.c      | 88 +----------------
>  .../ixgbe/ixgbe_recycle_mbufs_vec_common.c    | 89 +----------------
>  4 files changed, 107 insertions(+), 169 deletions(-)
> 
> diff --git a/drivers/net/intel/common/recycle_mbufs.h 
> b/drivers/net/intel/common/recycle_mbufs.h
> index fd31c5c1ff..88779c5aa4 100644
> --- a/drivers/net/intel/common/recycle_mbufs.h
> +++ b/drivers/net/intel/common/recycle_mbufs.h
> @@ -64,4 +64,102 @@ ci_rx_recycle_mbufs(struct ci_rx_queue *rxq, const 
> uint16_t nb_mbufs,
>       rte_write32_wc_relaxed(rte_cpu_to_le_32(rx_id), rxq->qrx_tail);
>  }
>  
> +/**
> + * Recycle buffers on Tx. Note: the function must first perform a 
> driver-specific
> + * DD-bit-set check to ensure that the Tx descriptors are ready for 
> recycling.
> + *
> + * @param txq Tx queue pointer
> + * @param recycle_rxq_info recycling mbuf information
> + *
> + * @return how many buffers were recycled
> + */
> +static __rte_always_inline uint16_t
> +ci_tx_recycle_mbufs(struct ci_tx_queue *txq,
> +     struct rte_eth_recycle_rxq_info *recycle_rxq_info)
> +{
> +     struct ci_tx_entry *txep;
> +     struct rte_mbuf **rxep;
> +     int i, n;
> +     uint16_t nb_recycle_mbufs;
> +     uint16_t avail = 0;
> +     uint16_t mbuf_ring_size = recycle_rxq_info->mbuf_ring_size;
> +     uint16_t mask = recycle_rxq_info->mbuf_ring_size - 1;
> +     uint16_t refill_requirement = recycle_rxq_info->refill_requirement;
> +     uint16_t refill_head = *recycle_rxq_info->refill_head;
> +     uint16_t receive_tail = *recycle_rxq_info->receive_tail;
> +
> +     /* Get available recycling Rx buffers. */
> +     avail = (mbuf_ring_size - (refill_head - receive_tail)) & mask;
> +
> +     /* Check Tx free thresh and Rx available space. */
> +     if (txq->nb_tx_free > txq->tx_free_thresh || avail <= txq->tx_rs_thresh)
> +             return 0;
> +
> +     n = txq->tx_rs_thresh;
> +     nb_recycle_mbufs = n;
> +
> +     /* Mbufs recycle mode can only support no ring buffer wrapping around.
> +      * Two case for this:
> +      *
> +      * case 1: The refill head of Rx buffer ring needs to be aligned with
> +      * mbuf ring size. In this case, the number of Tx freeing buffers
> +      * should be equal to refill_requirement.
> +      *
> +      * case 2: The refill head of Rx ring buffer does not need to be aligned
> +      * with mbuf ring size. In this case, the update of refill head can not
> +      * exceed the Rx mbuf ring size.
> +      */
> +     if ((refill_requirement && refill_requirement != n) ||
> +             (!refill_requirement && (refill_head + n > mbuf_ring_size)))
> +             return 0;
> +
> +     /* First buffer to free from S/W ring is at index
> +      * tx_next_dd - (tx_rs_thresh-1).
> +      */
> +     txep = &txq->sw_ring[txq->tx_next_dd - (n - 1)];
> +     rxep = recycle_rxq_info->mbuf_ring;
> +     rxep += refill_head;
> +
> +     /* is fast-free enabled in offloads? */
> +     if (txq->offloads & RTE_ETH_TX_OFFLOAD_MBUF_FAST_FREE) {
> +             /* Avoid txq containing buffers from unexpected mempool. */
> +             if (unlikely(recycle_rxq_info->mp
> +                                     != txep[0].mbuf->pool))
> +                     return 0;
> +
> +             /* Directly put mbufs from Tx to Rx. */
> +             for (i = 0; i < n; i++)
> +                     rxep[i] = txep[i].mbuf;
> +     } else {
> +             for (i = 0; i < n; i++) {
> +                     rxep[i] = rte_pktmbuf_prefree_seg(txep[i].mbuf);
> +
> +                     /* If Tx buffers are not the last reference or from
> +                      * unexpected mempool, previous copied buffers are
> +                      * considered as invalid.
> +                      */
> +                     if (unlikely(rxep[i] == NULL ||
> +                             recycle_rxq_info->mp != txep[i].mbuf->pool))
> +                             nb_recycle_mbufs = 0;
> +             }
> +             /* If Tx buffers are not the last reference or
> +              * from unexpected mempool, all recycled buffers
> +              * are put into mempool.
> +              */
> +             if (nb_recycle_mbufs == 0)
> +                     for (i = 0; i < n; i++) {
> +                             if (rxep[i] != NULL)
> +                                     rte_mempool_put(rxep[i]->pool, rxep[i]);
> +                     }
> +     }
> +
> +     /* Update counters for Tx. */
> +     txq->nb_tx_free = (uint16_t)(txq->nb_tx_free + txq->tx_rs_thresh);
> +     txq->tx_next_dd = (uint16_t)(txq->tx_next_dd + txq->tx_rs_thresh);
> +     if (txq->tx_next_dd >= txq->nb_tx_desc)
> +             txq->tx_next_dd = (uint16_t)(txq->tx_rs_thresh - 1);
> +
> +     return nb_recycle_mbufs;
> +}
> +
>  #endif
> diff --git a/drivers/net/intel/common/tx.h b/drivers/net/intel/common/tx.h
> index c99bd5420f..cc70fa7db4 100644
> --- a/drivers/net/intel/common/tx.h
> +++ b/drivers/net/intel/common/tx.h
> @@ -37,6 +37,7 @@ struct ci_tx_queue {
>               volatile struct ice_tx_desc *ice_tx_ring;
>               volatile struct idpf_base_tx_desc *idpf_tx_ring;
>               volatile union ixgbe_adv_tx_desc *ixgbe_tx_ring;
> +             volatile void *tx_ring; /**< Generic. */
>       };
>       volatile uint8_t *qtx_tail;               /* register address of tail */
>       union {
> diff --git a/drivers/net/intel/i40e/i40e_recycle_mbufs_vec_common.c 
> b/drivers/net/intel/i40e/i40e_recycle_mbufs_vec_common.c
> index 073357bee2..19edee781d 100644
> --- a/drivers/net/intel/i40e/i40e_recycle_mbufs_vec_common.c
> +++ b/drivers/net/intel/i40e/i40e_recycle_mbufs_vec_common.c
> @@ -23,92 +23,12 @@ i40e_recycle_tx_mbufs_reuse_vec(void *tx_queue,
>       struct rte_eth_recycle_rxq_info *recycle_rxq_info)
>  {
>       struct ci_tx_queue *txq = tx_queue;
> -     struct ci_tx_entry *txep;
> -     struct rte_mbuf **rxep;
> -     int i, n;
> -     uint16_t nb_recycle_mbufs;
> -     uint16_t avail = 0;
> -     uint16_t mbuf_ring_size = recycle_rxq_info->mbuf_ring_size;
> -     uint16_t mask = recycle_rxq_info->mbuf_ring_size - 1;
> -     uint16_t refill_requirement = recycle_rxq_info->refill_requirement;
> -     uint16_t refill_head = *recycle_rxq_info->refill_head;
> -     uint16_t receive_tail = *recycle_rxq_info->receive_tail;
> +     const uint64_t ctob = 
> txq->i40e_tx_ring[txq->tx_next_dd].cmd_type_offset_bsz;
>  
> -     /* Get available recycling Rx buffers. */
> -     avail = (mbuf_ring_size - (refill_head - receive_tail)) & mask;
> -
> -     /* Check Tx free thresh and Rx available space. */
> -     if (txq->nb_tx_free > txq->tx_free_thresh || avail <= txq->tx_rs_thresh)
> -             return 0;
> -
> -     /* check DD bits on threshold descriptor */
> -     if ((txq->i40e_tx_ring[txq->tx_next_dd].cmd_type_offset_bsz &
> -                             rte_cpu_to_le_64(I40E_TXD_QW1_DTYPE_MASK)) !=
> +     /* are Tx descriptors ready for recycling? */
> +     if ((ctob & rte_cpu_to_le_64(I40E_TXD_QW1_DTYPE_MASK)) !=
>                       rte_cpu_to_le_64(I40E_TX_DESC_DTYPE_DESC_DONE))
>               return 0;

There is the function i40e_tx_desc_done (and similar functions for other
drivers) to do this check. In the tx cleanup code we pass that function in
as a callback - you could probably shorten things a little by doing so
here. Due to inlining, the indirect function call is eliminated by the
compiler (as constant propagation), leading to no perf penalty.

/Bruce

Reply via email to