XDP hardware hints discussion mail archive
 help / color / mirror / Atom feed
From: Song Yoong Siang <yoong.siang.song@intel.com>
To: "David S . Miller" <davem@davemloft.net>,
	Eric Dumazet <edumazet@google.com>,
	Jakub Kicinski <kuba@kernel.org>, Paolo Abeni <pabeni@redhat.com>,
	Jonathan Corbet <corbet@lwn.net>, Bjorn Topel <bjorn@kernel.org>,
	Magnus Karlsson <magnus.karlsson@intel.com>,
	Maciej Fijalkowski <maciej.fijalkowski@intel.com>,
	Jonathan Lemon <jonathan.lemon@gmail.com>,
	Alexei Starovoitov <ast@kernel.org>,
	Daniel Borkmann <daniel@iogearbox.net>,
	Jesper Dangaard Brouer <hawk@kernel.org>,
	John Fastabend <john.fastabend@gmail.com>,
	Stanislav Fomichev <sdf@google.com>,
	Lorenzo Bianconi <lorenzo@kernel.org>,
	Tariq Toukan <tariqt@nvidia.com>,
	Willem de Bruijn <willemb@google.com>,
	Maxime Coquelin <mcoquelin.stm32@gmail.com>,
	Andrii Nakryiko <andrii@kernel.org>,
	Mykola Lysenko <mykolal@fb.com>,
	Martin KaFai Lau <martin.lau@linux.dev>,
	Song Liu <song@kernel.org>,
	Yonghong Song <yonghong.song@linux.dev>,
	KP Singh <kpsingh@kernel.org>, Hao Luo <haoluo@google.com>,
	Jiri Olsa <jolsa@kernel.org>, Shuah Khan <shuah@kernel.org>,
	Alexandre Torgue <alexandre.torgue@foss.st.com>,
	Jose Abreu <joabreu@synopsys.com>
Cc: netdev@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-doc@vger.kernel.org, bpf@vger.kernel.org,
	xdp-hints@xdp-project.net,
	linux-stm32@st-md-mailman.stormreply.com,
	linux-arm-kernel@lists.infradead.org,
	linux-kselftest@vger.kernel.org,
	Song Yoong Siang <yoong.siang.song@intel.com>
Subject: [xdp-hints] [PATCH bpf-next v2 1/3] xsk: add ETF support to XDP Tx metadata
Date: Fri,  1 Dec 2023 14:24:19 +0800	[thread overview]
Message-ID: <20231201062421.1074768-2-yoong.siang.song@intel.com> (raw)
In-Reply-To: <20231201062421.1074768-1-yoong.siang.song@intel.com>

This patch extends the XDP Tx metadata framework to include Earliest
TxTime First (ETF) HW offload support where the NIC will schedule the
packet for transmission at a pre-determined time called txtime. The value
of txtime is communicated from user space to Ethernet driver via txtime
field of struct xsk_tx_metadata.

Suggested-by: Stanislav Fomichev <sdf@google.com>
Signed-off-by: Song Yoong Siang <yoong.siang.song@intel.com>
---
 Documentation/netlink/specs/netdev.yaml      | 4 ++++
 Documentation/networking/xsk-tx-metadata.rst | 5 +++++
 include/net/xdp_sock.h                       | 9 +++++++++
 include/net/xdp_sock_drv.h                   | 1 +
 include/uapi/linux/if_xdp.h                  | 9 +++++++++
 include/uapi/linux/netdev.h                  | 3 +++
 net/core/netdev-genl.c                       | 2 ++
 net/xdp/xsk.c                                | 3 +++
 tools/include/uapi/linux/if_xdp.h            | 9 +++++++++
 tools/include/uapi/linux/netdev.h            | 3 +++
 tools/net/ynl/generated/netdev-user.c        | 1 +
 11 files changed, 49 insertions(+)

diff --git a/Documentation/netlink/specs/netdev.yaml b/Documentation/netlink/specs/netdev.yaml
index 00439bcbd2e3..339cdcddbfc0 100644
--- a/Documentation/netlink/specs/netdev.yaml
+++ b/Documentation/netlink/specs/netdev.yaml
@@ -66,6 +66,10 @@ definitions:
         name: tx-checksum
         doc:
           L3 checksum HW offload is supported by the driver.
+      -
+        name: txtime
+        doc:
+          Earliest TxTime First (ETF) HW offload is supported by the driver.
 
 attribute-sets:
   -
diff --git a/Documentation/networking/xsk-tx-metadata.rst b/Documentation/networking/xsk-tx-metadata.rst
index 97ecfa480d00..654ff692062a 100644
--- a/Documentation/networking/xsk-tx-metadata.rst
+++ b/Documentation/networking/xsk-tx-metadata.rst
@@ -44,6 +44,10 @@ The flags field enables the particular offload:
   checksum. ``csum_start`` specifies byte offset of where the checksumming
   should start and ``csum_offset`` specifies byte offset where the
   device should store the computed checksum.
+- ``XDP_TXMD_FLAGS_TXTIME``: requests Earliest TxTime First (ETF) HW
+  offload to launch the packet at a pre-determined time. ``txtime``
+  indicates the time which the NIC should schedule the packet for
+  transmission.
 
 Besides the flags above, in order to trigger the offloads, the first
 packet's ``struct xdp_desc`` descriptor should set ``XDP_TX_METADATA``
@@ -68,6 +72,7 @@ Refer to ``xsk-flags`` features bitmask in
 
 - ``tx-timestamp``: device supports ``XDP_TXMD_FLAGS_TIMESTAMP``
 - ``tx-checksum``: device supports ``XDP_TXMD_FLAGS_CHECKSUM``
+- ``txtime``: device supports ``XDP_TXMD_FLAGS_TXTIME``
 
 See ``tools/net/ynl/samples/netdev.c`` on how to query this information.
 
diff --git a/include/net/xdp_sock.h b/include/net/xdp_sock.h
index 3cb4dc9bd70e..0651b5264b72 100644
--- a/include/net/xdp_sock.h
+++ b/include/net/xdp_sock.h
@@ -110,11 +110,16 @@ struct xdp_sock {
  *     indicates position where checksumming should start.
  *     csum_offset indicates position where checksum should be stored.
  *
+ * void (*tmo_request_txtime)(u64 txtime, void *priv)
+ *     Called when AF_XDP frame requested Earliest TxTime First (ETF) HW offload
+ *     support. txtime indicates the time which the NIC should schedule the
+ *     packet for transmission.
  */
 struct xsk_tx_metadata_ops {
 	void	(*tmo_request_timestamp)(void *priv);
 	u64	(*tmo_fill_timestamp)(void *priv);
 	void	(*tmo_request_checksum)(u16 csum_start, u16 csum_offset, void *priv);
+	void	(*tmo_request_txtime)(u64 txtime, void *priv);
 };
 
 #ifdef CONFIG_XDP_SOCKETS
@@ -170,6 +175,10 @@ static inline void xsk_tx_metadata_request(const struct xsk_tx_metadata *meta,
 		if (meta->flags & XDP_TXMD_FLAGS_CHECKSUM)
 			ops->tmo_request_checksum(meta->request.csum_start,
 						  meta->request.csum_offset, priv);
+
+	if (ops->tmo_request_txtime)
+		if (meta->flags & XDP_TXMD_FLAGS_TXTIME)
+			ops->tmo_request_txtime(meta->request.txtime, priv);
 }
 
 /**
diff --git a/include/net/xdp_sock_drv.h b/include/net/xdp_sock_drv.h
index 81e02de3f453..e66d597e7079 100644
--- a/include/net/xdp_sock_drv.h
+++ b/include/net/xdp_sock_drv.h
@@ -168,6 +168,7 @@ static inline void *xsk_buff_raw_get_data(struct xsk_buff_pool *pool, u64 addr)
 #define XDP_TXMD_FLAGS_VALID ( \
 		XDP_TXMD_FLAGS_TIMESTAMP | \
 		XDP_TXMD_FLAGS_CHECKSUM | \
+		XDP_TXMD_FLAGS_TXTIME | \
 	0)
 
 static inline bool xsk_buff_valid_tx_metadata(struct xsk_tx_metadata *meta)
diff --git a/include/uapi/linux/if_xdp.h b/include/uapi/linux/if_xdp.h
index d31698410410..24d123bce7f3 100644
--- a/include/uapi/linux/if_xdp.h
+++ b/include/uapi/linux/if_xdp.h
@@ -123,6 +123,12 @@ struct xdp_options {
  */
 #define XDP_TXMD_FLAGS_CHECKSUM			(1 << 1)
 
+/* Request Earliest TxTime First (ETF) HW offload to launch the packet at a
+ * pre-determined time. The time which the NIC should schedule the packet for
+ * transmission is communicated via txtime field of struct xsk_tx_metadata.
+ */
+#define XDP_TXMD_FLAGS_TXTIME			(1 << 2)
+
 /* AF_XDP offloads request. 'request' union member is consumed by the driver
  * when the packet is being transmitted. 'completion' union member is
  * filled by the driver when the transmit completion arrives.
@@ -138,6 +144,9 @@ struct xsk_tx_metadata {
 			__u16 csum_start;
 			/* Offset from csum_start where checksum should be stored. */
 			__u16 csum_offset;
+
+			/* XDP_TXMD_FLAGS_TXTIME */
+			__u64 txtime;
 		} request;
 
 		struct {
diff --git a/include/uapi/linux/netdev.h b/include/uapi/linux/netdev.h
index 48d5477a668c..03b913757e1c 100644
--- a/include/uapi/linux/netdev.h
+++ b/include/uapi/linux/netdev.h
@@ -59,10 +59,13 @@ enum netdev_xdp_rx_metadata {
  *   by the driver.
  * @NETDEV_XSK_FLAGS_TX_CHECKSUM: L3 checksum HW offload is supported by the
  *   driver.
+ * @NETDEV_XSK_FLAGS_TXTIME: Earliest TxTime First (ETF) HW offload is supported
+ *   by the driver.
  */
 enum netdev_xsk_flags {
 	NETDEV_XSK_FLAGS_TX_TIMESTAMP = 1,
 	NETDEV_XSK_FLAGS_TX_CHECKSUM = 2,
+	NETDEV_XSK_FLAGS_TXTIME = 3,
 
 	/* private: */
 	NETDEV_XSK_FLAGS_MASK = 3,
diff --git a/net/core/netdev-genl.c b/net/core/netdev-genl.c
index 10f2124e9e23..aecc6e26f839 100644
--- a/net/core/netdev-genl.c
+++ b/net/core/netdev-genl.c
@@ -33,6 +33,8 @@ XDP_METADATA_KFUNC_xxx
 			xsk_features |= NETDEV_XSK_FLAGS_TX_TIMESTAMP;
 		if (netdev->xsk_tx_metadata_ops->tmo_request_checksum)
 			xsk_features |= NETDEV_XSK_FLAGS_TX_CHECKSUM;
+		if (netdev->xsk_tx_metadata_ops->tmo_request_txtime)
+			xsk_features |= NETDEV_XSK_FLAGS_TXTIME;
 	}
 
 	if (nla_put_u32(rsp, NETDEV_A_DEV_IFINDEX, netdev->ifindex) ||
diff --git a/net/xdp/xsk.c b/net/xdp/xsk.c
index 281d49b4fca4..4d48fc6caf7c 100644
--- a/net/xdp/xsk.c
+++ b/net/xdp/xsk.c
@@ -751,6 +751,9 @@ static struct sk_buff *xsk_build_skb(struct xdp_sock *xs,
 						goto free_err;
 				}
 			}
+
+			if (meta->flags & XDP_TXMD_FLAGS_TXTIME)
+				skb->skb_mstamp_ns = meta->request.txtime;
 		}
 	}
 
diff --git a/tools/include/uapi/linux/if_xdp.h b/tools/include/uapi/linux/if_xdp.h
index 638c606dfa74..abcf79e8a879 100644
--- a/tools/include/uapi/linux/if_xdp.h
+++ b/tools/include/uapi/linux/if_xdp.h
@@ -123,6 +123,12 @@ struct xdp_options {
  */
 #define XDP_TXMD_FLAGS_CHECKSUM			(1 << 1)
 
+/* Request Earliest TxTime First (ETF) HW offload to launch the packet at a
+ * pre-determined time. The time which the NIC should schedule the packet for
+ * transmission is communicated via txtime field of struct xsk_tx_metadata.
+ */
+#define XDP_TXMD_FLAGS_TXTIME			(1 << 2)
+
 /* AF_XDP offloads request. 'request' union member is consumed by the driver
  * when the packet is being transmitted. 'completion' union member is
  * filled by the driver when the transmit completion arrives.
@@ -138,6 +144,9 @@ struct xsk_tx_metadata {
 			__u16 csum_start;
 			/* Offset from csum_start where checksum should be stored. */
 			__u16 csum_offset;
+
+			/* XDP_TXMD_FLAGS_TXTIME */
+			__u64 txtime;
 		} request;
 
 		struct {
diff --git a/tools/include/uapi/linux/netdev.h b/tools/include/uapi/linux/netdev.h
index 48d5477a668c..03b913757e1c 100644
--- a/tools/include/uapi/linux/netdev.h
+++ b/tools/include/uapi/linux/netdev.h
@@ -59,10 +59,13 @@ enum netdev_xdp_rx_metadata {
  *   by the driver.
  * @NETDEV_XSK_FLAGS_TX_CHECKSUM: L3 checksum HW offload is supported by the
  *   driver.
+ * @NETDEV_XSK_FLAGS_TXTIME: Earliest TxTime First (ETF) HW offload is supported
+ *   by the driver.
  */
 enum netdev_xsk_flags {
 	NETDEV_XSK_FLAGS_TX_TIMESTAMP = 1,
 	NETDEV_XSK_FLAGS_TX_CHECKSUM = 2,
+	NETDEV_XSK_FLAGS_TXTIME = 3,
 
 	/* private: */
 	NETDEV_XSK_FLAGS_MASK = 3,
diff --git a/tools/net/ynl/generated/netdev-user.c b/tools/net/ynl/generated/netdev-user.c
index 6283d87dad37..02f1bd4cd97b 100644
--- a/tools/net/ynl/generated/netdev-user.c
+++ b/tools/net/ynl/generated/netdev-user.c
@@ -61,6 +61,7 @@ const char *netdev_xdp_rx_metadata_str(enum netdev_xdp_rx_metadata value)
 static const char * const netdev_xsk_flags_strmap[] = {
 	[0] = "tx-timestamp",
 	[1] = "tx-checksum",
+	[2] = "txtime"
 };
 
 const char *netdev_xsk_flags_str(enum netdev_xsk_flags value)
-- 
2.34.1


  reply	other threads:[~2023-12-01  6:25 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-12-01  6:24 [xdp-hints] [PATCH bpf-next v2 0/3] xsk: TX metadata txtime support Song Yoong Siang
2023-12-01  6:24 ` Song Yoong Siang [this message]
2023-12-01  6:24 ` [xdp-hints] [PATCH bpf-next v2 2/3] net: stmmac: Add txtime support to XDP ZC Song Yoong Siang
2023-12-01 15:02   ` [xdp-hints] " Jesper Dangaard Brouer
2023-12-01 15:26     ` Willem de Bruijn
2023-12-03 10:11     ` Song, Yoong Siang
2023-12-04 14:57       ` Willem de Bruijn
2023-12-05 14:43         ` Song, Yoong Siang
2023-12-05 14:55           ` Willem de Bruijn
2023-12-05 15:28             ` Song, Yoong Siang
2023-12-01  6:24 ` [xdp-hints] [PATCH bpf-next v2 3/3] selftests/bpf: Add txtime to xdp_hw_metadata Song Yoong Siang
2023-12-01 10:46 ` [xdp-hints] Re: [PATCH bpf-next v2 0/3] xsk: TX metadata txtime support Jesper Dangaard Brouer
2023-12-01 13:43   ` Song, Yoong Siang
2023-12-01 15:09     ` Willem de Bruijn
2023-12-01 15:39       ` Jesper Dangaard Brouer
2023-12-02 14:15         ` Willem de Bruijn
2023-12-03  9:16           ` Song, Yoong Siang

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

  List information: https://lists.xdp-project.net/postorius/lists/xdp-hints.xdp-project.net/

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20231201062421.1074768-2-yoong.siang.song@intel.com \
    --to=yoong.siang.song@intel.com \
    --cc=alexandre.torgue@foss.st.com \
    --cc=andrii@kernel.org \
    --cc=ast@kernel.org \
    --cc=bjorn@kernel.org \
    --cc=bpf@vger.kernel.org \
    --cc=corbet@lwn.net \
    --cc=daniel@iogearbox.net \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=haoluo@google.com \
    --cc=hawk@kernel.org \
    --cc=joabreu@synopsys.com \
    --cc=john.fastabend@gmail.com \
    --cc=jolsa@kernel.org \
    --cc=jonathan.lemon@gmail.com \
    --cc=kpsingh@kernel.org \
    --cc=kuba@kernel.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-doc@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-kselftest@vger.kernel.org \
    --cc=linux-stm32@st-md-mailman.stormreply.com \
    --cc=lorenzo@kernel.org \
    --cc=maciej.fijalkowski@intel.com \
    --cc=magnus.karlsson@intel.com \
    --cc=martin.lau@linux.dev \
    --cc=mcoquelin.stm32@gmail.com \
    --cc=mykolal@fb.com \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=sdf@google.com \
    --cc=shuah@kernel.org \
    --cc=song@kernel.org \
    --cc=tariqt@nvidia.com \
    --cc=willemb@google.com \
    --cc=xdp-hints@xdp-project.net \
    --cc=yonghong.song@linux.dev \
    /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