From: Ajit Khaparde <ajit.khaparde@broadcom.com>
To: dev@dpdk.org
Cc: Damodharam Ammepalli <damodharam.ammepalli@broadcom.com>
Subject: [PATCH 01/18] net/bnxt: add support for UDP GSO
Date: Thu, 21 Dec 2023 10:05:12 -0800 [thread overview]
Message-ID: <20231221180529.18687-2-ajit.khaparde@broadcom.com> (raw)
In-Reply-To: <20231221180529.18687-1-ajit.khaparde@broadcom.com>
[-- Attachment #1: Type: text/plain, Size: 3238 bytes --]
UDP GSO aka UDP Fragmentation Offload allows an application or stack
to provide a data payload larger than the MTU. The application then
updates the mbuf ol_flags and sets the PKT_TX_UDP_SEG flag. Then based
on the tso_segs and tso_mss fields in the mbuf the PMD can indicate
the UDP GSO transmit request to the hardware.
This feature is supported on Thor2 and will be enabled when the firmware
sets the UDP GSO support via the HWRM_FUNC_QCAPS.
Signed-off-by: Ajit Khaparde <ajit.khaparde@broadcom.com>
Reviewed-by: Damodharam Ammepalli <damodharam.ammepalli@broadcom.com>
---
drivers/net/bnxt/bnxt.h | 1 +
drivers/net/bnxt/bnxt_hwrm.c | 2 ++
drivers/net/bnxt/bnxt_txq.c | 2 ++
drivers/net/bnxt/bnxt_txr.c | 7 ++++++-
4 files changed, 11 insertions(+), 1 deletion(-)
diff --git a/drivers/net/bnxt/bnxt.h b/drivers/net/bnxt/bnxt.h
index 7aed4c3da3..4b5c2c4b8f 100644
--- a/drivers/net/bnxt/bnxt.h
+++ b/drivers/net/bnxt/bnxt.h
@@ -801,6 +801,7 @@ struct bnxt {
(BNXT_CHIP_P5_P7((bp)) && \
(bp)->hwrm_spec_code >= HWRM_VERSION_1_9_2 && \
!BNXT_VF((bp)))
+#define BNXT_FW_CAP_UDP_GSO BIT(13)
#define BNXT_TRUFLOW_EN(bp) ((bp)->fw_cap & BNXT_FW_CAP_TRUFLOW_EN &&\
(bp)->app_id != 0xFF)
diff --git a/drivers/net/bnxt/bnxt_hwrm.c b/drivers/net/bnxt/bnxt_hwrm.c
index e56f7693af..37cf179938 100644
--- a/drivers/net/bnxt/bnxt_hwrm.c
+++ b/drivers/net/bnxt/bnxt_hwrm.c
@@ -950,6 +950,8 @@ static int __bnxt_hwrm_func_qcaps(struct bnxt *bp)
flags_ext2 = rte_le_to_cpu_32(resp->flags_ext2);
if (flags_ext2 & HWRM_FUNC_QCAPS_OUTPUT_FLAGS_EXT2_RX_ALL_PKTS_TIMESTAMPS_SUPPORTED)
bp->fw_cap |= BNXT_FW_CAP_RX_ALL_PKT_TS;
+ if (flags_ext2 & HWRM_FUNC_QCAPS_OUTPUT_FLAGS_EXT2_UDP_GSO_SUPPORTED)
+ bp->fw_cap |= BNXT_FW_CAP_UDP_GSO;
unlock:
HWRM_UNLOCK();
diff --git a/drivers/net/bnxt/bnxt_txq.c b/drivers/net/bnxt/bnxt_txq.c
index 4df4604975..f99ad211db 100644
--- a/drivers/net/bnxt/bnxt_txq.c
+++ b/drivers/net/bnxt/bnxt_txq.c
@@ -42,6 +42,8 @@ uint64_t bnxt_get_tx_port_offloads(struct bnxt *bp)
tx_offload_capa |= RTE_ETH_TX_OFFLOAD_GENEVE_TNL_TSO;
if (BNXT_TUNNELED_OFFLOADS_CAP_IPINIP_EN(bp))
tx_offload_capa |= RTE_ETH_TX_OFFLOAD_IPIP_TNL_TSO;
+ if (bp->fw_cap & BNXT_FW_CAP_UDP_GSO)
+ tx_offload_capa |= RTE_ETH_TX_OFFLOAD_UDP_TSO;
return tx_offload_capa;
}
diff --git a/drivers/net/bnxt/bnxt_txr.c b/drivers/net/bnxt/bnxt_txr.c
index 899986764f..38da2d2829 100644
--- a/drivers/net/bnxt/bnxt_txr.c
+++ b/drivers/net/bnxt/bnxt_txr.c
@@ -123,6 +123,10 @@ bnxt_xmit_need_long_bd(struct rte_mbuf *tx_pkt, struct bnxt_tx_queue *txq)
return false;
}
+/* Used for verifying TSO segments during TCP Segmentation Offload or
+ * UDP Fragmentation Offload. tx_pkt->tso_segsz stores the number of
+ * segments or fragments in those cases.
+ */
static bool
bnxt_zero_data_len_tso_segsz(struct rte_mbuf *tx_pkt, uint8_t data_len_chk)
{
@@ -308,7 +312,8 @@ static uint16_t bnxt_start_xmit(struct rte_mbuf *tx_pkt,
else
txbd1->cfa_action = txq->bp->tx_cfa_action;
- if (tx_pkt->ol_flags & RTE_MBUF_F_TX_TCP_SEG) {
+ if (tx_pkt->ol_flags & RTE_MBUF_F_TX_TCP_SEG ||
+ tx_pkt->ol_flags & RTE_MBUF_F_TX_UDP_SEG) {
uint16_t hdr_size;
/* TSO */
--
2.39.2 (Apple Git-143)
[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 4218 bytes --]
next prev parent reply other threads:[~2023-12-21 18:05 UTC|newest]
Thread overview: 19+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-12-21 18:05 [PATCH 00/18] bnxt patchset Ajit Khaparde
2023-12-21 18:05 ` Ajit Khaparde [this message]
2023-12-21 18:05 ` [PATCH 02/18] net/bnxt: add support for compressed Rx CQE Ajit Khaparde
2023-12-21 18:05 ` [PATCH 03/18] net/bnxt: fix a typo while parsing link speed Ajit Khaparde
2023-12-21 18:05 ` [PATCH 04/18] net/bnxt: fix setting 50G and 100G forced speed Ajit Khaparde
2023-12-21 18:05 ` [PATCH 05/18] net/bnxt: fix speed change from 200G to 25G on Thor Ajit Khaparde
2023-12-21 18:05 ` [PATCH 06/18] net/bnxt: support backward compatibility Ajit Khaparde
2023-12-21 18:05 ` [PATCH 07/18] net/bnxt: reattempt mbuf allocation for Rx and AGG rings Ajit Khaparde
2023-12-21 18:05 ` [PATCH 08/18] net/bnxt: refactor Rx doorbell during Rx flush Ajit Khaparde
2023-12-21 18:05 ` [PATCH 09/18] net/bnxt: extend RSS hash support for P7 devices Ajit Khaparde
2023-12-21 18:05 ` [PATCH 10/18] net/bnxt: add flow query callback Ajit Khaparde
2023-12-21 18:05 ` [PATCH 11/18] net/bnxt: add ESP and AH header based RSS support Ajit Khaparde
2023-12-21 18:05 ` [PATCH 12/18] net/bnxt: set allmulti mode if multicast filter fails Ajit Khaparde
2023-12-21 18:05 ` [PATCH 13/18] net/bnxt: add VF FLR async event handler Ajit Khaparde
2023-12-21 18:05 ` [PATCH 14/18] net/bnxt: add tunnel TPA support Ajit Khaparde
2023-12-21 18:05 ` [PATCH 15/18] net/bnxt: add 400G get support for P7 devices Ajit Khaparde
2023-12-21 18:05 ` [PATCH 16/18] net/bnxt: query extended stats from firmware Ajit Khaparde
2023-12-21 18:05 ` [PATCH 17/18] net/bnxt: add AVX2 support for compressed CQE Ajit Khaparde
2023-12-21 18:05 ` [PATCH 18/18] net/bnxt: enable SSE mode " Ajit Khaparde
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20231221180529.18687-2-ajit.khaparde@broadcom.com \
--to=ajit.khaparde@broadcom.com \
--cc=damodharam.ammepalli@broadcom.com \
--cc=dev@dpdk.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).