i40e: xsk: use xdp_build_skb_from_zc() for XDP_PASS
Subsystems (3)
PATCH v2 1/1 i40e: xsk: use xdp_build_skb_from_zc() for XDP_PASS
From: Chenguang Zhao <[email protected]>
Replace the driver-local i40e_construct_skb_zc() with the common helper xdp_build_skb_from_zc(). On failure, free the xdp buff in the caller.
xdp_build_skb_from_zc() already calls skb_record_rx_queue() and eth_type_trans(), so pull the remaining descriptor field setup into __i40e_process_skb_fields() and use that on the XDP_PASS path. Briefly restore the Ethernet header around eth_skb_pad() so padding sees the full L2 frame length.
---
v2:
- Extract descriptor field setup into __i40e_process_skb_fields() and use it on the XDP_PASS path, so skb_record_rx_queue()/eth_type_trans() are not repeated after xdp_build_skb_from_zc().
- Keep a brief __skb_push()/__skb_pull() around eth_skb_pad() so padding still sees the full L2 frame length.
v1:
- https://lore.kernel.org/all/[email protected]/
--- drivers/net/ethernet/intel/i40e/i40e_txrx.c | 32 ++++++-- .../ethernet/intel/i40e/i40e_txrx_common.h | 2 + drivers/net/ethernet/intel/i40e/i40e_xsk.c | 76 +++---------------- 3 files changed, 36 insertions(+), 74 deletions(-)
drivers/net/ethernet/intel/i40e/i40e_txrx.c +24 -8
@@ -1833,17 +1833,17 @@ static inline void i40e_rx_hash(struct i40e_ring *ring, }/** -* i40e_process_skb_fields - Populate skb header fields from Rx descriptor +* __i40e_process_skb_fields - Populate skb fields from Rx descriptor * @rx_ring: rx descriptor ring packet is being transacted on * @rx_desc: pointer to the EOP Rx descriptor * @skb: pointer to current skb being populated * -* This function checks the ring, descriptor, and packet information in -* order to populate the hash, checksum, VLAN, protocol, and -* other fields within the skb. +* Populate hash, checksum, PTP timestamp and VLAN from @rx_desc. Does not +* call skb_record_rx_queue() or eth_type_trans(); callers that already got +* those from xdp_build_skb_from_zc() should use this helper. **/ -void i40e_process_skb_fields (struct i40e_ring *rx_ring, -union i40e_rx_desc *rx_desc,struct sk_buff *skb) +void __i40e_process_skb_fields (struct i40e_ring *rx_ring, +union i40e_rx_desc *rx_desc,struct sk_buff *skb) {u64 qword =le64_to_cpu (rx_desc->wb.qword1.status_error_len);u32 rx_status =FIELD_GET (I40E_RXD_QW1_STATUS_MASK , qword);
@@ -1858,14 +1858,30 @@ void i40e_process_skb_fields(struct i40e_ring *rx_ring,i40e_rx_checksum (rx_ring->vsi, skb, rx_desc); -skb_record_rx_queue (skb, rx_ring->queue_index); -if (qword &BIT (I40E_RX_DESC_STATUS_L2TAG1P_SHIFT )) {__le16 vlan_tag = rx_desc->wb.qword0.lo_dword.l2tag1;__vlan_hwaccel_put_tag (skb,htons (ETH_P_8021Q ),le16_to_cpu (vlan_tag)); } +} + +/** +* i40e_process_skb_fields - Populate skb header fields from Rx descriptor +* @rx_ring: rx descriptor ring packet is being transacted on +* @rx_desc: pointer to the EOP Rx descriptor +* @skb: pointer to current skb being populated +* +* This function checks the ring, descriptor, and packet information in +* order to populate the hash, checksum, VLAN, protocol, and +* other fields within the skb. +**/ +void i40e_process_skb_fields (struct i40e_ring *rx_ring, +union i40e_rx_desc *rx_desc,struct sk_buff *skb) +{ +__i40e_process_skb_fields (rx_ring, rx_desc, skb); + +skb_record_rx_queue (skb, rx_ring->queue_index);/* modifies the skb - consumes the enet header */ skb->protocol =eth_type_trans (skb, rx_ring->netdev);
drivers/net/ethernet/intel/i40e/i40e_txrx_common.h +2 -0
@@ -9,6 +9,8 @@int i40e_xmit_xdp_tx_ring (struct xdp_buff *xdp,struct i40e_ring *xdp_ring);void i40e_clean_programming_status (struct i40e_ring *rx_ring,u64 qword0_raw,u64 qword1); +void __i40e_process_skb_fields (struct i40e_ring *rx_ring, +union i40e_rx_desc *rx_desc,struct sk_buff *skb);void i40e_process_skb_fields (struct i40e_ring *rx_ring,union i40e_rx_desc *rx_desc,struct sk_buff *skb);void i40e_xdp_ring_update_tail (struct i40e_ring *xdp_ring);
drivers/net/ethernet/intel/i40e/i40e_xsk.c +10 -66
107 lines not on this page · raw message
Signature · 1 line
2.25.1
Quoted 29 lines · Chenguang Zhao
From: Chenguang Zhao <[email protected]>
Replace the driver-local
i40e_construct_skb_zc()with the common helperxdp_build_skb_from_zc(). On failure, free the xdp buff in the caller.
xdp_build_skb_from_zc()already callsskb_record_rx_queue()andeth_type_trans(), so pull the remaining descriptor field setup into__i40e_process_skb_fields()and use that on theXDP_PASSpath. Briefly restore the Ethernet header aroundeth_skb_pad()so padding sees the full L2 frame length.Signed-off-by Chenguang Zhao <[email protected]>---
v2:
- Extract descriptor field setup into__i40e_process_skb_fields()and use it on theXDP_PASSpath, so skb_record_rx_queue()/eth_type_trans() are not repeated afterxdp_build_skb_from_zc().
- Keep a brief __skb_push()/__skb_pull() aroundeth_skb_pad()so padding still sees the full L2 frame length.v1:
- https://lore.kernel.org/all/[email protected]/--- drivers/net/ethernet/intel/i40e/i40e_txrx.c | 32 ++++++-- .../ethernet/intel/i40e/i40e_txrx_common.h | 2 + drivers/net/ethernet/intel/i40e/i40e_xsk.c | 76 +++---------------- 3 files changed, 36 insertions(+), 74 deletions(-)
[...]
Quoted 18 lines
static void i40e_handle_xdp_result_zc(struct
i40e_ring*rx_ring,drivers/net/ethernet/intel/i40e/i40e_xsk.c +3
-1struct xdp_buff *xdp_buff, union i40e_rx_desc *rx_desc,@@ -372,21 +309,28 @@ static void i40e_handle_xdp_result_zc(struct i40e_ring *rx_ring,* BIT(I40E_RXD_QW1_ERROR_SHIFT). This is due to that * SBP is *not* set in PRT_SBPVSI (default not set). */ - skb =i40e_construct_skb_zc (rx_ring, xdp_buff); + skb =xdp_build_skb_from_zc (xdp_buff);if (!skb) { +xsk_buff_free (xdp_buff); rx_ring->rx_stats.alloc_buff_failed++; *rx_packets =0 ; *rx_bytes =0 ;return ; } +/* xdp_build_skb_from_zc() already ran eth_type_trans() and
drivers/net/ethernet/intel/i40e/i40e_xsk.c static void i40e_handle_xdp_result_zc(struct i40e_ring *rx_ring, · line 331
Quoted 10 lines
drivers/net/ethernet/intel/i40e/i40e_xsk.c +5
-0+* skb_record_rx_queue(). Restore the Ethernet header only for +* eth_skb_pad(), then pull it back. +*/ +__skb_push (skb,ETH_HLEN );if (eth_skb_pad (skb)) { *rx_packets =0 ; *rx_bytes =0 ;return ; } +__skb_pull (skb,ETH_HLEN );
I am pretty sure that eth_skb_pad() is not needed on Rx, network stack handles small skbs just fine, and no other vendor does this. We were trying to figure out the historical reasons for eth_skb_pad() being there in the intel drivers, but with no luck.
Quoted 10 lines
drivers/net/ethernet/intel/i40e/i40e_xsk.c +1
-1-i40e_process_skb_fields (rx_ring, rx_desc, skb); +__i40e_process_skb_fields (rx_ring, rx_desc, skb);napi_gro_receive (&rx_ring->q_vector->napi, skb);return ; }Signature · 1 line
2.25.1