Re: [PATCH net-next 3/3] net/mlx5: Support getcyclesx and getcrosscycles

From: Carolina Jubran
Date: Thu Jul 17 2025 - 12:02:48 EST




On 17/07/2025 13:55, Paolo Abeni wrote:
On 7/15/25 7:15 AM, Tariq Toukan wrote:
From: Carolina Jubran <cjubran@xxxxxxxxxx>

Implement the getcyclesx64 and getcrosscycles callbacks in ptp_info to
expose the device’s raw free-running counter.

Signed-off-by: Carolina Jubran <cjubran@xxxxxxxxxx>
Reviewed-by: Dragos Tatulea <dtatulea@xxxxxxxxxx>
Signed-off-by: Tariq Toukan <tariqt@xxxxxxxxxx>
---
.../ethernet/mellanox/mlx5/core/lib/clock.c | 74 ++++++++++++++++++-
1 file changed, 73 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c b/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c
index b1e2deeefc0c..2f75726674a9 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/lib/clock.c
@@ -306,6 +306,23 @@ static int mlx5_mtctr_syncdevicetime(ktime_t *device_time,
return 0;
}
+static int
+mlx5_mtctr_syncdevicecyclestime(ktime_t *device_time,
+ struct system_counterval_t *sys_counterval,
+ void *ctx)
+{
+ struct mlx5_core_dev *mdev = ctx;
+ u64 device;
+ int err;
+
+ err = mlx5_mtctr_read(mdev, false, sys_counterval, &device);
+ if (err)
+ return err;
+ *device_time = ns_to_ktime(device);

If the goal is providing a raw cycle counter, why still using a timespec
to the user space? A plain u64 would possibly be less ambiguous.

/P


getcycles64 and getcrosscycles already return the cycle counter in a timespec64/ktime format, so I kept the new ioctls consistent with them.