From mboxrd@z Thu Jan 1 00:00:00 1970
Return-Path: <xiao.w.wang@intel.com>
Received: from mga11.intel.com (mga11.intel.com [192.55.52.93])
by dpdk.org (Postfix) with ESMTP id 5D13D91B7
for <dev@dpdk.org>; Wed, 21 Oct 2015 09:23:44 +0200 (CEST)
Received: from fmsmga001.fm.intel.com ([10.253.24.23])
by fmsmga102.fm.intel.com with ESMTP; 21 Oct 2015 00:23:36 -0700
X-ExtLoop1: 1
X-IronPort-AV: E=Sophos;i="5.17,709,1437462000"; d="scan'208";a="815909402"
Received: from fmsmsx107.amr.corp.intel.com ([10.18.124.205])
by fmsmga001.fm.intel.com with ESMTP; 21 Oct 2015 00:23:35 -0700
Received: from fmsmsx101.amr.corp.intel.com (10.18.124.199) by
fmsmsx107.amr.corp.intel.com (10.18.124.205) with Microsoft SMTP Server (TLS)
id 14.3.248.2; Wed, 21 Oct 2015 00:23:34 -0700
Received: from shsmsx102.ccr.corp.intel.com (10.239.4.154) by
fmsmsx101.amr.corp.intel.com (10.18.124.199) with Microsoft SMTP Server (TLS)
id 14.3.248.2; Wed, 21 Oct 2015 00:23:34 -0700
Received: from shsmsx101.ccr.corp.intel.com ([169.254.1.96]) by
shsmsx102.ccr.corp.intel.com ([169.254.2.253]) with mapi id 14.03.0248.002;
Wed, 21 Oct 2015 15:23:32 +0800
From: "Wang, Xiao W" <xiao.w.wang@intel.com>
To: "Lu, Wenzhuo" <wenzhuo.lu@intel.com>, "dev@dpdk.org" <dev@dpdk.org>
Thread-Topic: [PATCH v3 1/2] e1000: enable igb TSO support
Thread-Index: AQHRCymiTPCYrCYih0mcISY5ZXBWQJ50+n6AgACRNqA=
Date: Wed, 21 Oct 2015 07:23:31 +0000
Message-ID: <B7F2E978279D1D49A3034B7786DACF403C823393@SHSMSX101.ccr.corp.intel.com>
References: <1444472872-26866-2-git-send-email-xiao.w.wang@intel.com>
<1445340132-2630-1-git-send-email-xiao.w.wang@intel.com>
<1445340132-2630-2-git-send-email-xiao.w.wang@intel.com>
<6A0DE07E22DDAD4C9103DF62FEBC0909020A11B5@shsmsx102.ccr.corp.intel.com>
In-Reply-To: <6A0DE07E22DDAD4C9103DF62FEBC0909020A11B5@shsmsx102.ccr.corp.intel.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach:
X-MS-TNEF-Correlator:
x-originating-ip: [10.239.127.40]
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Subject: Re: [dpdk-dev] [PATCH v3 1/2] e1000: enable igb TSO support
X-BeenThere: dev@dpdk.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: patches and discussions about DPDK <dev.dpdk.org>
List-Unsubscribe: <http://dpdk.org/ml/options/dev>,
<mailto:dev-request@dpdk.org?subject=unsubscribe>
List-Archive: <http://dpdk.org/ml/archives/dev/>
List-Post: <mailto:dev@dpdk.org>
List-Help: <mailto:dev-request@dpdk.org?subject=help>
List-Subscribe: <http://dpdk.org/ml/listinfo/dev>,
<mailto:dev-request@dpdk.org?subject=subscribe>
X-List-Received-Date: Wed, 21 Oct 2015 07:23:45 -0000
-----Original Message-----
From: Lu, Wenzhuo=20
Sent: Wednesday, October 21, 2015 2:41 PM
To: Wang, Xiao W; dev@dpdk.org
Cc: Richardson, Bruce; Zhang, Helin
Subject: RE: [PATCH v3 1/2] e1000: enable igb TSO support
Hi Xiao,
> -----Original Message-----
> From: Wang, Xiao W
> Sent: Tuesday, October 20, 2015 7:22 PM
> To: dev@dpdk.org
> Cc: Richardson, Bruce; Zhang, Helin; Lu, Wenzhuo; Wang, Xiao W
> Subject: [PATCH v3 1/2] e1000: enable igb TSO support
>=20
> This patch enables igb TSO feature, the feature works on both PF and VF.
> The TCP segmentation offload needs to write the offload related=20
> information into the advanced context descriptors, which is similar to ch=
ecksum offload.
>=20
> Signed-off-by: Wang Xiao W <xiao.w.wang@intel.com>
> ---
> drivers/net/e1000/igb_ethdev.c | 6 +-
> drivers/net/e1000/igb_rxtx.c | 200 +++++++++++++++++++++++++----------=
----
> --
> 2 files changed, 127 insertions(+), 79 deletions(-)
>=20
> diff --git a/drivers/net/e1000/igb_ethdev.c=20
> b/drivers/net/e1000/igb_ethdev.c index 848ef6e..2e69394 100644
> --- a/drivers/net/e1000/igb_ethdev.c
> +++ b/drivers/net/e1000/igb_ethdev.c
> @@ -1497,7 +1497,8 @@ eth_igb_infos_get(struct rte_eth_dev *dev,=20
> struct rte_eth_dev_info *dev_info)
> DEV_TX_OFFLOAD_IPV4_CKSUM |
> DEV_TX_OFFLOAD_UDP_CKSUM |
> DEV_TX_OFFLOAD_TCP_CKSUM |
> - DEV_TX_OFFLOAD_SCTP_CKSUM;
> + DEV_TX_OFFLOAD_SCTP_CKSUM |
> + DEV_TX_OFFLOAD_TCP_TSO;
>=20
> switch (hw->mac.type) {
> case e1000_82575:
> @@ -1588,7 +1589,8 @@ eth_igbvf_infos_get(struct rte_eth_dev *dev,=20
> struct rte_eth_dev_info *dev_info)
> DEV_TX_OFFLOAD_IPV4_CKSUM |
> DEV_TX_OFFLOAD_UDP_CKSUM |
> DEV_TX_OFFLOAD_TCP_CKSUM |
> - DEV_TX_OFFLOAD_SCTP_CKSUM;
> + DEV_TX_OFFLOAD_SCTP_CKSUM |
> + DEV_TX_OFFLOAD_TCP_TSO;
> switch (hw->mac.type) {
> case e1000_vfadapt:
> dev_info->max_rx_queues =3D 2;
> diff --git a/drivers/net/e1000/igb_rxtx.c=20
> b/drivers/net/e1000/igb_rxtx.c index
> 19905fd..afb8478 100644
> --- a/drivers/net/e1000/igb_rxtx.c
> +++ b/drivers/net/e1000/igb_rxtx.c
> @@ -76,7 +76,8 @@
> #define IGB_TX_OFFLOAD_MASK ( \
> PKT_TX_VLAN_PKT | \
> PKT_TX_IP_CKSUM | \
> - PKT_TX_L4_MASK)
> + PKT_TX_L4_MASK | \
> + PKT_TX_TCP_SEG)
>=20
> static inline struct rte_mbuf *
> rte_rxmbuf_alloc(struct rte_mempool *mp) @@ -146,32 +147,40 @@ enum=20
> igb_advctx_num { };
>=20
> /** Offload features */
> -union igb_vlan_macip {
> - uint32_t data;
> +union igb_tx_offload {
> + uint64_t data;
> struct {
> - uint16_t l2_l3_len; /**< 7bit L2 and 9b L3 lengths combined
> */
> - uint16_t vlan_tci;
> - /**< VLAN Tag Control Identifier (CPU order). */
> - } f;
> + uint64_t l3_len:9; /**< L3 (IP) Header Length. */
> + uint64_t l2_len:7; /**< L2 (MAC) Header Length. */
> + uint64_t vlan_tci:16; /**< VLAN Tag Control Identifier(CPU
> order). */
> + uint64_t l4_len:8; /**< L4 (TCP/UDP) Header Length. */
> + uint64_t tso_segsz:16; /**< TCP TSO segment size. */
> +
> + /* uint64_t unused:8; */
> + };
> };
>=20
> /*
> - * Compare mask for vlan_macip_len.data,
> - * should be in sync with igb_vlan_macip.f layout.
> + * Compare mask for igb_tx_offload.data,
> + * should be in sync with igb_tx_offload layout.
> * */
> -#define TX_VLAN_CMP_MASK 0xFFFF0000 /**< VLAN length - 16-bits. =
*/
> -#define TX_MAC_LEN_CMP_MASK 0x0000FE00 /**< MAC length - 7-bits.
> */
> -#define TX_IP_LEN_CMP_MASK 0x000001FF /**< IP length - 9-bits. */
> -/** MAC+IP length. */
> -#define TX_MACIP_LEN_CMP_MASK (TX_MAC_LEN_CMP_MASK |
> TX_IP_LEN_CMP_MASK)
> +#define TX_MACIP_LEN_CMP_MASK 0x000000000000FFFF /**< L2L3
> header mask. */
> +#define TX_VLAN_CMP_MASK 0x00000000FFFF0000 /**< Vlan mask.
> */
I think it's too long. Should be 0x00000000FFFF0000ULL?
Yes, it's mask for uint64_t, adding ULL will be better.=20
> +#define TX_TCP_LEN_CMP_MASK 0x000000FF00000000 /**<
> TCP header mask. */
> +#define TX_TSO_MSS_CMP_MASK 0x00FFFF0000000000 /**<
> TSO segsz mask. */
> +/** Mac + IP + TCP + Mss mask. */
> +#define TX_TSO_CMP_MASK \
> + (TX_MACIP_LEN_CMP_MASK | TX_TCP_LEN_CMP_MASK |
> TX_TSO_MSS_CMP_MASK)
>=20
> /**
> * Strucutre to check if new context need be built
> */
> struct igb_advctx_info {
> uint64_t flags; /**< ol_flags related to context build. */
> - uint32_t cmp_mask; /**< compare mask for vlan_macip_lens */
> - union igb_vlan_macip vlan_macip_lens; /**< vlan, mac & ip length.
> */
> + /** tx offload: vlan, tso, l2-l3-l4 lengths. */
> + union igb_tx_offload tx_offload;
> + /** compare mask for tx offload. */
> + union igb_tx_offload tx_offload_mask;
> };
>=20
> /**
> @@ -221,6 +230,8 @@ struct igb_tx_queue {
> * Macro for VMDq feature for 1 GbE NIC.
> */
> #define E1000_VMOLR_SIZE (8)
> +#define IGB_TSO_MAX_HDRLEN (512)
> +#define IGB_TSO_MAX_MSS (9216)
>=20
>=20
> /***************************************************************
> ******
> *
> @@ -229,6 +240,23 @@ struct igb_tx_queue {
>=20
> ****************************************************************
> ******/
>=20
> /*
> + *There're some limitations in hardware for TCP segmentation offload.
> +We *should check whether the parameters are valid.
> + */
> +static inline uint64_t
> +check_tso_para(uint64_t ol_req, union igb_tx_offload ol_para) {
> + if (!(ol_req & PKT_TX_TCP_SEG))
> + return ol_req;
> + if ((ol_para.tso_segsz > IGB_TSO_MAX_MSS) || (ol_para.l2_len +
> + ol_para.l3_len + ol_para.l4_len >
> IGB_TSO_MAX_HDRLEN)) {
> + ol_req &=3D ~PKT_TX_TCP_SEG;
> + ol_req |=3D PKT_TX_TCP_CKSUM;
> + }
> + return ol_req;
> +}
> +
> +/*
> * Advanced context descriptor are almost same between igb/ixgbe
> * This is a separate function, looking for optimization opportunity her=
e
> * Rework required to go with the pre-defined values.
> @@ -237,64 +265,81 @@ struct igb_tx_queue { static inline void=20
> igbe_set_xmit_ctx(struct igb_tx_queue* txq,
> volatile struct e1000_adv_tx_context_desc *ctx_txd,
> - uint64_t ol_flags, uint32_t vlan_macip_lens)
> + uint64_t ol_flags, union igb_tx_offload tx_offload)
> {
> uint32_t type_tucmd_mlhl;
> uint32_t mss_l4len_idx;
> uint32_t ctx_idx, ctx_curr;
> - uint32_t cmp_mask;
> + uint32_t vlan_macip_lens;
> + union igb_tx_offload tx_offload_mask;
>=20
> ctx_curr =3D txq->ctx_curr;
> ctx_idx =3D ctx_curr + txq->ctx_start;
>=20
> - cmp_mask =3D 0;
> + tx_offload_mask.data =3D 0;
> type_tucmd_mlhl =3D 0;
>=20
> - if (ol_flags & PKT_TX_VLAN_PKT) {
> - cmp_mask |=3D TX_VLAN_CMP_MASK;
> - }
> -
> - if (ol_flags & PKT_TX_IP_CKSUM) {
> - type_tucmd_mlhl =3D E1000_ADVTXD_TUCMD_IPV4;
> - cmp_mask |=3D TX_MACIP_LEN_CMP_MASK;
> - }
> -
> /* Specify which HW CTX to upload. */
> mss_l4len_idx =3D (ctx_idx << E1000_ADVTXD_IDX_SHIFT);
> - switch (ol_flags & PKT_TX_L4_MASK) {
> - case PKT_TX_UDP_CKSUM:
> - type_tucmd_mlhl |=3D E1000_ADVTXD_TUCMD_L4T_UDP |
> +
> + if (ol_flags & PKT_TX_VLAN_PKT)
> + tx_offload_mask.data |=3D TX_VLAN_CMP_MASK;
> +
> + /* check if TCP segmentation required for this packet */
> + if (ol_flags & PKT_TX_TCP_SEG) {
> + /* implies IP cksum in IPv4 */
> + if (ol_flags & PKT_TX_IP_CKSUM)
> + type_tucmd_mlhl =3D E1000_ADVTXD_TUCMD_IPV4 |
> + E1000_ADVTXD_TUCMD_L4T_TCP |
> E1000_ADVTXD_DTYP_CTXT |
> E1000_ADVTXD_DCMD_DEXT;
> - mss_l4len_idx |=3D sizeof(struct udp_hdr) <<
> E1000_ADVTXD_L4LEN_SHIFT;
> - cmp_mask |=3D TX_MACIP_LEN_CMP_MASK;
> - break;
> - case PKT_TX_TCP_CKSUM:
> - type_tucmd_mlhl |=3D E1000_ADVTXD_TUCMD_L4T_TCP |
> + else
> + type_tucmd_mlhl =3D E1000_ADVTXD_TUCMD_IPV6 |
> + E1000_ADVTXD_TUCMD_L4T_TCP |
> E1000_ADVTXD_DTYP_CTXT |
> E1000_ADVTXD_DCMD_DEXT;
> - mss_l4len_idx |=3D sizeof(struct tcp_hdr) <<
> E1000_ADVTXD_L4LEN_SHIFT;
> - cmp_mask |=3D TX_MACIP_LEN_CMP_MASK;
> - break;
> - case PKT_TX_SCTP_CKSUM:
> - type_tucmd_mlhl |=3D E1000_ADVTXD_TUCMD_L4T_SCTP |
> +
> + tx_offload_mask.data |=3D TX_TSO_CMP_MASK;
> + mss_l4len_idx |=3D tx_offload.tso_segsz <<
> E1000_ADVTXD_MSS_SHIFT;
> + mss_l4len_idx |=3D tx_offload.l4_len <<
> E1000_ADVTXD_L4LEN_SHIFT;
> + } else { /* no TSO, check if hardware checksum is needed */
> + if (ol_flags & (PKT_TX_IP_CKSUM | PKT_TX_L4_MASK))
> + tx_offload_mask.data |=3D TX_MACIP_LEN_CMP_MASK;
> +
> + if (ol_flags & PKT_TX_IP_CKSUM)
> + type_tucmd_mlhl =3D E1000_ADVTXD_TUCMD_IPV4;
> +
> + switch (ol_flags & PKT_TX_L4_MASK) {
> + case PKT_TX_UDP_CKSUM:
> + type_tucmd_mlhl |=3D
> E1000_ADVTXD_TUCMD_L4T_UDP |
> E1000_ADVTXD_DTYP_CTXT |
> E1000_ADVTXD_DCMD_DEXT;
> - mss_l4len_idx |=3D sizeof(struct sctp_hdr) <<
> E1000_ADVTXD_L4LEN_SHIFT;
> - cmp_mask |=3D TX_MACIP_LEN_CMP_MASK;
> - break;
> - default:
> - type_tucmd_mlhl |=3D E1000_ADVTXD_TUCMD_L4T_RSV |
> + mss_l4len_idx |=3D sizeof(struct udp_hdr) <<
> E1000_ADVTXD_L4LEN_SHIFT;
> + break;
> + case PKT_TX_TCP_CKSUM:
> + type_tucmd_mlhl |=3D
> E1000_ADVTXD_TUCMD_L4T_TCP |
> E1000_ADVTXD_DTYP_CTXT |
> E1000_ADVTXD_DCMD_DEXT;
> - break;
> + mss_l4len_idx |=3D sizeof(struct tcp_hdr) <<
> E1000_ADVTXD_L4LEN_SHIFT;
> + break;
> + case PKT_TX_SCTP_CKSUM:
> + type_tucmd_mlhl |=3D
> E1000_ADVTXD_TUCMD_L4T_SCTP |
> + E1000_ADVTXD_DTYP_CTXT |
> E1000_ADVTXD_DCMD_DEXT;
> + mss_l4len_idx |=3D sizeof(struct sctp_hdr) <<
> E1000_ADVTXD_L4LEN_SHIFT;
> + break;
> + default:
> + type_tucmd_mlhl |=3D
> E1000_ADVTXD_TUCMD_L4T_RSV |
> + E1000_ADVTXD_DTYP_CTXT |
> E1000_ADVTXD_DCMD_DEXT;
> + break;
> + }
> }
>=20
> - txq->ctx_cache[ctx_curr].flags =3D ol_flags;
> - txq->ctx_cache[ctx_curr].cmp_mask =3D cmp_mask;
> - txq->ctx_cache[ctx_curr].vlan_macip_lens.data =3D
> - vlan_macip_lens & cmp_mask;
> + txq->ctx_cache[ctx_curr].flags =3D ol_flags;
> + txq->ctx_cache[ctx_idx].tx_offload.data =3D
> + tx_offload_mask.data & tx_offload.data;
> + txq->ctx_cache[ctx_idx].tx_offload_mask =3D tx_offload_mask;
>=20
> ctx_txd->type_tucmd_mlhl =3D rte_cpu_to_le_32(type_tucmd_mlhl);
> + vlan_macip_lens =3D (uint32_t)tx_offload.data;
> ctx_txd->vlan_macip_lens =3D rte_cpu_to_le_32(vlan_macip_lens);
> - ctx_txd->mss_l4len_idx =3D rte_cpu_to_le_32(mss_l4len_idx);
> - ctx_txd->seqnum_seed =3D 0;
> + ctx_txd->mss_l4len_idx =3D rte_cpu_to_le_32(mss_l4len_idx);
> + ctx_txd->seqnum_seed =3D 0;
> }
>=20
> /*
> @@ -303,20 +348,20 @@ igbe_set_xmit_ctx(struct igb_tx_queue* txq,
> */
> static inline uint32_t
> what_advctx_update(struct igb_tx_queue *txq, uint64_t flags,
> - uint32_t vlan_macip_lens)
> + union igb_tx_offload tx_offload)
> {
> /* If match with the current context */
> if (likely((txq->ctx_cache[txq->ctx_curr].flags =3D=3D flags) &&
> - (txq->ctx_cache[txq->ctx_curr].vlan_macip_lens.data =3D=3D
> - (txq->ctx_cache[txq->ctx_curr].cmp_mask &
> vlan_macip_lens)))) {
> + (txq->ctx_cache[txq->ctx_curr].tx_offload.data =3D=3D
> + (txq->ctx_cache[txq->ctx_curr].tx_offload_mask.data &
> +tx_offload.data)))) {
> return txq->ctx_curr;
> }
>=20
> /* If match with the second context */
> txq->ctx_curr ^=3D 1;
> if (likely((txq->ctx_cache[txq->ctx_curr].flags =3D=3D flags) &&
> - (txq->ctx_cache[txq->ctx_curr].vlan_macip_lens.data =3D=3D
> - (txq->ctx_cache[txq->ctx_curr].cmp_mask &
> vlan_macip_lens)))) {
> + (txq->ctx_cache[txq->ctx_curr].tx_offload.data =3D=3D
> + (txq->ctx_cache[txq->ctx_curr].tx_offload_mask.data &
> +tx_offload.data)))) {
> return txq->ctx_curr;
> }
>=20
> @@ -333,14 +378,19 @@ tx_desc_cksum_flags_to_olinfo(uint64_t ol_flags)
>=20
> tmp =3D l4_olinfo[(ol_flags & PKT_TX_L4_MASK) !=3D=20
> PKT_TX_L4_NO_CKSUM];
> tmp |=3D l3_olinfo[(ol_flags & PKT_TX_IP_CKSUM) !=3D 0];
> + tmp |=3D l4_olinfo[(ol_flags & PKT_TX_TCP_SEG) !=3D 0];
> return tmp;
> }
>=20
> static inline uint32_t
> tx_desc_vlan_flags_to_cmdtype(uint64_t ol_flags) {
> + uint32_t cmdtype;
> static uint32_t vlan_cmd[2] =3D {0, E1000_ADVTXD_DCMD_VLE};
> - return vlan_cmd[(ol_flags & PKT_TX_VLAN_PKT) !=3D 0];
> + static uint32_t tso_cmd[2] =3D {0, E1000_ADVTXD_DCMD_TSE};
> + cmdtype =3D vlan_cmd[(ol_flags & PKT_TX_VLAN_PKT) !=3D 0];
> + cmdtype |=3D tso_cmd[(ol_flags & PKT_TX_TCP_SEG) !=3D 0];
> + return cmdtype;
> }
>=20
> uint16_t
> @@ -354,14 +404,6 @@ eth_igb_xmit_pkts(void *tx_queue, struct rte_mbuf=20
> **tx_pkts,
> volatile union e1000_adv_tx_desc *txd;
> struct rte_mbuf *tx_pkt;
> struct rte_mbuf *m_seg;
> - union igb_vlan_macip vlan_macip_lens;
> - union {
> - uint16_t u16;
> - struct {
> - uint16_t l3_len:9;
> - uint16_t l2_len:7;
> - };
> - } l2_l3_len;
> uint64_t buf_dma_addr;
> uint32_t olinfo_status;
> uint32_t cmd_type_len;
> @@ -375,6 +417,7 @@ eth_igb_xmit_pkts(void *tx_queue, struct rte_mbuf=20
> **tx_pkts,
> uint64_t tx_ol_req;
> uint32_t new_ctx =3D 0;
> uint32_t ctx =3D 0;
> + union igb_tx_offload tx_offload =3D {0};
>=20
> txq =3D tx_queue;
> sw_ring =3D txq->sw_ring;
> @@ -399,16 +442,18 @@ eth_igb_xmit_pkts(void *tx_queue, struct=20
> rte_mbuf **tx_pkts,
> tx_last =3D (uint16_t) (tx_id + tx_pkt->nb_segs - 1);
>=20
> ol_flags =3D tx_pkt->ol_flags;
> - l2_l3_len.l2_len =3D tx_pkt->l2_len;
> - l2_l3_len.l3_len =3D tx_pkt->l3_len;
> - vlan_macip_lens.f.vlan_tci =3D tx_pkt->vlan_tci;
> - vlan_macip_lens.f.l2_l3_len =3D l2_l3_len.u16;
> tx_ol_req =3D ol_flags & IGB_TX_OFFLOAD_MASK;
>=20
> /* If a Context Descriptor need be built . */
> if (tx_ol_req) {
> - ctx =3D what_advctx_update(txq, tx_ol_req,
> - vlan_macip_lens.data);
> + tx_offload.l2_len =3D tx_pkt->l2_len;
> + tx_offload.l3_len =3D tx_pkt->l3_len;
> + tx_offload.l4_len =3D tx_pkt->l4_len;
> + tx_offload.vlan_tci =3D tx_pkt->vlan_tci;
> + tx_offload.tso_segsz =3D tx_pkt->tso_segsz;
> + tx_ol_req =3D check_tso_para(tx_ol_req, tx_offload);
> +
> + ctx =3D what_advctx_update(txq, tx_ol_req, tx_offload);
> /* Only allocate context descriptor if required*/
> new_ctx =3D (ctx =3D=3D IGB_CTX_NUM);
> ctx =3D txq->ctx_curr;
> @@ -500,6 +545,8 @@ eth_igb_xmit_pkts(void *tx_queue, struct rte_mbuf=20
> **tx_pkts,
> */
> cmd_type_len =3D txq->txd_type |
> E1000_ADVTXD_DCMD_IFCS |
> E1000_ADVTXD_DCMD_DEXT;
> + if (tx_ol_req & PKT_TX_TCP_SEG)
> + pkt_len -=3D (tx_pkt->l2_len + tx_pkt->l3_len + tx_pkt-
> >l4_len);
> olinfo_status =3D (pkt_len << E1000_ADVTXD_PAYLEN_SHIFT); #if=20
> defined(RTE_LIBRTE_IEEE1588)
> if (ol_flags & PKT_TX_IEEE1588_TMST) @@ -523,8 +570,7 @@=20
> eth_igb_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts,
> txe->mbuf =3D NULL;
> }
>=20
> - igbe_set_xmit_ctx(txq, ctx_txd, tx_ol_req,
> - vlan_macip_lens.data);
> + igbe_set_xmit_ctx(txq, ctx_txd, tx_ol_req,
> tx_offload);
>=20
> txe->last_id =3D tx_last;
> tx_id =3D txe->next_id;
> @@ -532,8 +578,8 @@ eth_igb_xmit_pkts(void *tx_queue, struct rte_mbuf=20
> **tx_pkts,
> }
>=20
> /* Setup the TX Advanced Data Descriptor */
> - cmd_type_len |=3D
> tx_desc_vlan_flags_to_cmdtype(ol_flags);
> - olinfo_status |=3D
> tx_desc_cksum_flags_to_olinfo(ol_flags);
> + cmd_type_len |=3D
> tx_desc_vlan_flags_to_cmdtype(tx_ol_req);
> + olinfo_status |=3D
> tx_desc_cksum_flags_to_olinfo(tx_ol_req);
> olinfo_status |=3D (ctx << E1000_ADVTXD_IDX_SHIFT);
> }
>=20
> --
> 1.9.3