Skip to contentSkip to conversation

gve: fix probe failure if clock read fails

PATCH Applied Harshitha Ramamurthy 2 participants 2 messages Jan 27 – Jan 28

Subsystems (4)
lore
Share & tools
lore

Harshitha Ramamurthy

author maintainer
#0 lore raw

PATCH 1/1 gve: fix probe failure if clock read fails

From: Jordan Rhee <[email protected]>

If timestamping is supported, GVE reads the clock during probe, which can fail for various reasons. Previously, this failure would abort the driver probe, rendering the device unusable. This behavior has been observed on production GCP VMs, causing driver initialization to fail completely.

This patch allows the driver to degrade gracefully. If gve_init_clock() fails, it logs a warning and continues loading the driver without PTP support.

Fixes a479a27f4da4 ("gve: Move gve_init_clock to after AQ CONFIGURE_DEVICE_RESOURCES call")
---
 drivers/net/ethernet/google/gve/gve.h         |  5 +++++
 drivers/net/ethernet/google/gve/gve_ethtool.c |  2 +-
 drivers/net/ethernet/google/gve/gve_main.c    | 12 +++++++-----
 drivers/net/ethernet/google/gve/gve_ptp.c     |  8 --------
 drivers/net/ethernet/google/gve/gve_rx_dqo.c  |  2 +-
 5 files changed, 14 insertions(+), 15 deletions(-)
drivers/net/ethernet/google/gve/gve.h +5 -0
index 970d5ca8..cbdf3a84 100644
@@ -1206,6 +1206,11 @@ static inline bool gve_supports_xdp_xmit(struct gve_priv *priv)
 	}
 }
 
+static inline bool gve_is_clock_enabled(struct gve_priv *priv)
+{
+	return priv->nic_ts_report;
+}
+
 /* gqi napi handler defined in gve_main.c */
 int gve_napi_poll(struct napi_struct *napi, int budget);
 
drivers/net/ethernet/google/gve/gve_ethtool.c +1 -1
index f7864ae7..137dd728 100644
@@ -952,7 +952,7 @@ static int gve_get_ts_info(struct net_device *netdev,
 
 	ethtool_op_get_ts_info(netdev, info);
 
-	if (priv->nic_timestamp_supported) {
+	if (gve_is_clock_enabled(priv)) {
 		info->so_timestamping |= SOF_TIMESTAMPING_RX_HARDWARE |
 					 SOF_TIMESTAMPING_RAW_HARDWARE;
 
drivers/net/ethernet/google/gve/gve_main.c +7 -5
index 96adbbe1..dbc84de3 100644
@@ -680,10 +680,12 @@ static int gve_setup_device_resources(struct gve_priv *priv)
 		}
 	}
 
-	err = gve_init_clock(priv);
-	if (err) {
-		dev_err(&priv->pdev->dev, "Failed to init clock");
-		goto abort_with_ptype_lut;
+	if (priv->nic_timestamp_supported) {
+		err = gve_init_clock(priv);
+		if (err) {
+			dev_warn(&priv->pdev->dev, "Failed to init clock, continuing without PTP support");
+			err = 0;
+		}
 	}
 
 	err = gve_init_rss_config(priv, priv->rx_cfg.num_queues);
@@ -2183,7 +2185,7 @@ static int gve_set_ts_config(struct net_device *dev,
 	}
 
 	if (kernel_config->rx_filter != HWTSTAMP_FILTER_NONE) {
-		if (!priv->nic_ts_report) {
+		if (!gve_is_clock_enabled(priv)) {
 			NL_SET_ERR_MSG_MOD(extack,
 					   "RX timestamping is not supported");
 			kernel_config->rx_filter = HWTSTAMP_FILTER_NONE;
drivers/net/ethernet/google/gve/gve_ptp.c +0 -8
index 073677d8..de42fc2c 100644
@@ -70,11 +70,6 @@ static int gve_ptp_init(struct gve_priv *priv)
 	struct gve_ptp *ptp;
 	int err;
 
-	if (!priv->nic_timestamp_supported) {
-		dev_dbg(&priv->pdev->dev, "Device does not support PTP\n");
-		return -EOPNOTSUPP;
-	}
-
 	priv->ptp = kzalloc(sizeof(*priv->ptp), GFP_KERNEL);
 	if (!priv->ptp)
 		return -ENOMEM;
@@ -116,9 +111,6 @@ int gve_init_clock(struct gve_priv *priv)
 {
 	int err;
 
-	if (!priv->nic_timestamp_supported)
-		return 0;
-
 	err = gve_ptp_init(priv);
 	if (err)
 		return err;
drivers/net/ethernet/google/gve/gve_rx_dqo.c +1 -1
index f1bd8f5d..63a96106 100644
@@ -484,7 +484,7 @@ int gve_xdp_rx_timestamp(const struct xdp_md *_ctx, u64 *timestamp)
 {
 	const struct gve_xdp_buff *ctx = (void *)_ctx;
 
-	if (!ctx->gve->nic_ts_report)
+	if (!gve_is_clock_enabled(ctx->gve))
 		return -ENODATA;
 
 	if (!(ctx->compl_desc->ts_sub_nsecs_low & GVE_DQO_RX_HWTSTAMP_VALID))
Signature · 1 line
2.52.0.457.g6b5491de43-goog
#1 lore raw
This patch was applied to netdev/net.git (main) by Jakub Kicinski <[email protected]>:

Hello:

This patch was applied to netdev/net.git (main)
by Jakub Kicinski <[email protected]>:

Quoted 9 lines · On Tue, 27 Jan 2026 01:02:10 +0000 you wrote:

From: Jordan Rhee <[email protected]>

If timestamping is supported, GVE reads the clock during probe, which can fail for various reasons. Previously, this failure would abort the driver probe, rendering the device unusable. This behavior has been observed on production GCP VMs, causing driver initialization to fail completely.

[...]

Here is the summary with links:
- [net] gve: fix probe failure if clock read fails

    https://git.kernel.org/netdev/net/c/a040afa3bca4

You are awesome, thank you!

Signature · 2 lines
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html