Re: [PATCH v3 2/2] net/af_xdp: Refactor af_xdp_tx_zc()

2025-01-30 Thread Ariel Otilibili
Hello Maryam, On Wed, Jan 29, 2025 at 6:58 PM Maryam Tahhan wrote: > > +static struct rte_mbuf * > > +maybe_kick_tx(struct pkt_tx_queue *txq, uint32_t *idx_tx, struct > rte_mbuf *mbuf) > > +{ > > + struct rte_mbuf *ret = mbuf; > > + > > + if (!xsk_ring_prod__reserve(&txq->tx, 1, idx_tx))

Re: [PATCH v3 2/2] net/af_xdp: Refactor af_xdp_tx_zc()

2025-01-29 Thread Maryam Tahhan
On 28/01/2025 23:11, Ariel Otilibili wrote: Both legs of the loop share the same logic: either to go out of the function, or to fall through to the next instructions. For that, maybe_cpy_pkt() is introduced. Bugzilla ID: 1440 Signed-off-by: Ariel Otilibili --- drivers/net/af_xdp/rte_eth_af

[PATCH v3 2/2] net/af_xdp: Refactor af_xdp_tx_zc()

2025-01-28 Thread Ariel Otilibili
Both legs of the loop share the same logic: either to go out of the function, or to fall through to the next instructions. For that, maybe_cpy_pkt() is introduced. Bugzilla ID: 1440 Signed-off-by: Ariel Otilibili --- drivers/net/af_xdp/rte_eth_af_xdp.c | 131 ++-- 1 file