DPDK-dev Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Mark Blasko <blasko@google.com>
To: stephen@networkplumber.org
Cc: dev@dpdk.org, Mark Blasko <blasko@google.com>,
	 Joshua Washington <joshwash@google.com>,
	"Jasper Tran O'Leary" <jtranoleary@google.com>
Subject: [PATCH v2 5/6] net/gve: support read clock ethdev op
Date: Fri, 15 May 2026 23:19:34 +0000	[thread overview]
Message-ID: <20260515231936.3296603-6-blasko@google.com> (raw)
In-Reply-To: <20260515231936.3296603-1-blasko@google.com>

Implement the read_clock operation in eth_dev_ops. The function calls
the AdminQ command to fetch the current NIC timestamp synchronously,
updates the cached timestamp used for reconstruction, and returns the
full 64-bit value.

Signed-off-by: Mark Blasko <blasko@google.com>
Reviewed-by: Joshua Washington <joshwash@google.com>
Reviewed-by: Jasper Tran O'Leary <jtranoleary@google.com>
---
v2:
    - Scoped read_clock ethdev operation strictly to DQO queues.
---
 drivers/net/gve/gve_ethdev.c | 28 ++++++++++++++++++++++++++++
 1 file changed, 28 insertions(+)

diff --git a/drivers/net/gve/gve_ethdev.c b/drivers/net/gve/gve_ethdev.c
index ee21fb3ec4..c600062faf 100644
--- a/drivers/net/gve/gve_ethdev.c
+++ b/drivers/net/gve/gve_ethdev.c
@@ -1271,6 +1271,34 @@ gve_flow_ops_get(struct rte_eth_dev *dev, const struct rte_flow_ops **ops)
 	return 0;
 }
 
+static int
+gve_read_clock(struct rte_eth_dev *dev, uint64_t *clock)
+{
+	struct gve_priv *priv = dev->data->dev_private;
+	uint64_t ts;
+	int err;
+
+	if (!priv->nic_timestamp_supported)
+		return -EOPNOTSUPP;
+
+	if (!priv->nic_ts_report_mz)
+		return -EIO;
+
+	err = gve_adminq_report_nic_timestamp(priv, priv->nic_ts_report_mz->iova);
+	if (err != 0)
+		return err;
+
+	ts = be64_to_cpu(priv->nic_ts_report->nic_timestamp);
+	*clock = ts;
+
+	/* Update the cached value */
+	rte_atomic_store_explicit(&priv->last_read_nic_timestamp, ts, rte_memory_order_relaxed);
+	rte_atomic_store_explicit(&priv->nic_ts_read_fails, 0, rte_memory_order_relaxed);
+	rte_atomic_store_explicit(&priv->nic_ts_stale, 0, rte_memory_order_release);
+
+	return 0;
+}
+
 static const struct eth_dev_ops gve_eth_dev_ops = {
 	.dev_configure        = gve_dev_configure,
 	.dev_start            = gve_dev_start,
-- 
2.54.0.563.g4f69b47b94-goog


  parent reply	other threads:[~2026-05-15 23:20 UTC|newest]

Thread overview: 15+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-05-11 22:43 [PATCH 0/6] net/gve: add hardware timestamping support mark-blasko
2026-05-11 22:43 ` [PATCH 1/6] net/gve: add thread safety to admin queue mark-blasko
2026-05-11 22:43 ` [PATCH 2/6] net/gve: add device option support for HW timestamps mark-blasko
2026-05-11 22:43 ` [PATCH 3/6] net/gve: add AdminQ command for NIC timestamps mark-blasko
2026-05-11 22:43 ` [PATCH 4/6] net/gve: add periodic NIC clock synchronization mark-blasko
2026-05-11 22:43 ` [PATCH 5/6] net/gve: support read clock ethdev op mark-blasko
2026-05-11 22:43 ` [PATCH 6/6] net/gve: reconstruct HW timestamps from DQO mark-blasko
2026-05-12  7:14 ` [PATCH 0/6] net/gve: add hardware timestamping support Stephen Hemminger
2026-05-15 23:19 ` [PATCH v2 " Mark Blasko
2026-05-15 23:19   ` [PATCH v2 1/6] net/gve: add thread safety to admin queue Mark Blasko
2026-05-15 23:19   ` [PATCH v2 2/6] net/gve: add device option support for HW timestamps Mark Blasko
2026-05-15 23:19   ` [PATCH v2 3/6] net/gve: add AdminQ command for NIC timestamps Mark Blasko
2026-05-15 23:19   ` [PATCH v2 4/6] net/gve: add periodic NIC clock synchronization Mark Blasko
2026-05-15 23:19   ` Mark Blasko [this message]
2026-05-15 23:19   ` [PATCH v2 6/6] net/gve: reconstruct HW timestamps from DQO Mark Blasko

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=20260515231936.3296603-6-blasko@google.com \
    --to=blasko@google.com \
    --cc=dev@dpdk.org \
    --cc=joshwash@google.com \
    --cc=jtranoleary@google.com \
    --cc=stephen@networkplumber.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