Skip to contentSkip to conversation

bonding: fix incorrect software timestamping report

PATCH Applied Hangbin Liu 3 participants 3 messages Jun 20, 2024 – Jun 21, 2024

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 bonding: fix incorrect software timestamping report

The __ethtool_get_ts_info function returns directly if the device has a get_ts_info() method. For bonding with an active slave, this works correctly as we simply return the real device's timestamping information. However, when there is no active slave, we only check the slave's TX software timestamp information. We still need to set the phc index and RX timestamp information manually. Otherwise, the result will be look like:

Time stamping parameters for bond0:
Capabilities:
software-transmit
PTP Hardware Clock: 0
Hardware Transmit Timestamp Modes: none
Hardware Receive Filter Modes: none

This issue does not affect VLAN or MACVLAN devices, as they only have one downlink and can directly use the downlink's timestamping information.

Fixes b8768dc40777 ("net: ethtool: Refactor identical get_ts_info implementations.")
---
 drivers/net/bonding/bond_main.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/net/bonding/bond_main.c +3 -0
index 3c3fcce4acd4..d19aabf5d4fb 100644
@@ -5773,6 +5773,9 @@ static int bond_ethtool_get_ts_info(struct net_device *bond_dev,
 	if (real_dev) {
 		ret = ethtool_get_ts_info_by_layer(real_dev, info);
 	} else {
+		info->phc_index = -1;
+		info->so_timestamping = SOF_TIMESTAMPING_RX_SOFTWARE |
+					SOF_TIMESTAMPING_SOFTWARE;
 		/* Check if all slaves support software tx timestamping */
 		rcu_read_lock();
 		bond_for_each_slave_rcu(bond, slave, iter) {
Signature · 1 line
2.45.0
Acked-by on v1 1/1
#1 lore raw
Quoted 14 lines · On Thu, 20 Jun 2024 16:56:26 +0800 Hangbin Liu <[email protected]> wrote:

The __ethtool_get_ts_info function returns directly if the device has a get_ts_info() method. For bonding with an active slave, this works correctly as we simply return the real device's timestamping information. However, when there is no active slave, we only check the slave's TX software timestamp information. We still need to set the phc index and RX timestamp information manually. Otherwise, the result will be look like:

Time stamping parameters for bond0:
Capabilities:
software-transmit
PTP Hardware Clock: 0
Hardware Transmit Timestamp Modes: none
Hardware Receive Filter Modes: none

Quoted 2 lines

This issue does not affect VLAN or MACVLAN devices, as they only have one downlink and can directly use the downlink's timestamping information.

--
Köry Maincent, Bootlin
Embedded Linux and kernel engineering
https://bootlin.com

#2 lore raw
In reply to Hangbin Liu
This patch was applied to netdev/net.git (main) by David S. Miller <[email protected]>:

Hello:

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

Quoted 8 lines · On Thu, 20 Jun 2024 16:56:26 +0800 you wrote:

The __ethtool_get_ts_info function returns directly if the device has a get_ts_info() method. For bonding with an active slave, this works correctly as we simply return the real device's timestamping information. However, when there is no active slave, we only check the slave's TX software timestamp information. We still need to set the phc index and RX timestamp information manually. Otherwise, the result will be look like:

[...]

Here is the summary with links:
- [net] bonding: fix incorrect software timestamping report

    https://git.kernel.org/netdev/net/c/a95b031c6796

You are awesome, thank you!

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