* [PATCH net-next v17 00/14] net: Make timestamping selectable
@ 2024-07-09 13:53 Kory Maincent
2024-07-09 13:53 ` [PATCH net-next v17 01/14] net_tstamp: Add TIMESTAMPING SOFTWARE and HARDWARE mask Kory Maincent
` (14 more replies)
0 siblings, 15 replies; 39+ messages in thread
From: Kory Maincent @ 2024-07-09 13:53 UTC (permalink / raw)
To: Florian Fainelli, Broadcom internal kernel review list,
Andrew Lunn, Heiner Kallweit, Russell King, David S. Miller,
Eric Dumazet, Jakub Kicinski, Paolo Abeni, Richard Cochran,
Radu Pirea, Jay Vosburgh, Andy Gospodarek, Nicolas Ferre,
Claudiu Beznea, Willem de Bruijn, Jonathan Corbet, Horatiu Vultur,
UNGLinuxDriver, Simon Horman, Vladimir Oltean, donald.hunter,
danieller, ecree.xilinx
Cc: Thomas Petazzoni, linux-kernel, netdev, linux-doc,
Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter, Kory Maincent
Up until now, there was no way to let the user select the hardware
PTP provider at which time stamping occurs. The stack assumed that PHY time
stamping is always preferred, but some MAC/PHY combinations were buggy.
This series updates the default MAC/PHY default timestamping and aims to
allow the user to select the desired hwtstamp provider administratively.
Changes in v17:
- Fix a documentation nit.
- Add a missing kernel_ethtool_tsinfo update from a new MAC driver.
- Link to v16: https://lore.kernel.org/r/20240705-feature_ptp_netnext-v16-0-5d7153914052@bootlin.com
Changes in v16:
- Add a new patch to separate tsinfo into a new tsconfig command to get
and set the hwtstamp config.
- Used call_rcu() instead of synchronize_rcu() to free the hwtstamp_provider
- Moved net core changes of patch 12 directly to patch 8.
- Link to v15: https://lore.kernel.org/r/20240612-feature_ptp_netnext-v15-0-b2a086257b63@bootlin.com
Changes in v15:
- Fix uninitialized ethtool_ts_info structure.
- Link to v14: https://lore.kernel.org/r/20240604-feature_ptp_netnext-v14-0-77b6f6efea40@bootlin.com
Changes in v14:
- Add back an EXPORT_SYMBOL() missing.
- Link to v13: https://lore.kernel.org/r/20240529-feature_ptp_netnext-v13-0-6eda4d40fa4f@bootlin.com
Changes in v13:
- Add PTP builtin code to fix build errors when building PTP as a module.
- Fix error spotted by smatch and sparse.
- Link to v12: https://lore.kernel.org/r/20240430-feature_ptp_netnext-v12-0-2c5f24b6a914@bootlin.com
Changes in v12:
- Add missing return description in the kdoc.
- Fix few nit.
- Link to v11: https://lore.kernel.org/r/20240422-feature_ptp_netnext-v11-0-f14441f2a1d8@bootlin.com
Changes in v11:
- Add netlink examples.
- Remove a change of my out of tree marvell_ptp patch in the patch series.
- Remove useless extern.
- Link to v10: https://lore.kernel.org/r/20240409-feature_ptp_netnext-v10-0-0fa2ea5c89a9@bootlin.com
Changes in v10:
- Move declarations to net/core/dev.h instead of netdevice.h
- Add netlink documentation.
- Add ETHTOOL_A_TSINFO_GHWTSTAMP netlink attributes instead of a bit in
ETHTOOL_A_TSINFO_TIMESTAMPING bitset.
- Send "Move from simple ida to xarray" patch standalone.
- Add tsinfo ntf command.
- Add rcu_lock protection mechanism to avoid memory leak.
- Fixed doc and kdoc issue.
- Link to v9: https://lore.kernel.org/r/20240226-feature_ptp_netnext-v9-0-455611549f21@bootlin.com
Changes in v9:
- Remove the RFC prefix.
- Correct few NIT fixes.
- Link to v8: https://lore.kernel.org/r/20240216-feature_ptp_netnext-v8-0-510f42f444fb@bootlin.com
Changes in v8:
- Drop the 6 first patch as they are now merged.
- Change the full implementation to not be based on the hwtstamp layer
(MAC/PHY) but on the hwtstamp provider which mean a ptp clock and a
phc qualifier.
- Made some patch to prepare the new implementation.
- Expand netlink tsinfo instead of a new ts command for new hwtstamp
configuration uAPI and for dumping tsinfo of specific hwtstamp provider.
- Link to v7: https://lore.kernel.org/r/20231114-feature_ptp_netnext-v7-0-472e77951e40@bootlin.com
Changes in v7:
- Fix a temporary build error.
- Link to v6: https://lore.kernel.org/r/20231019-feature_ptp_netnext-v6-0-71affc27b0e5@bootlin.com
Changes in v6:
- Few fixes from the reviews.
- Replace the allowlist to default_timestamp flag to know which phy is
using old API behavior.
- Rename the timestamping layer enum values.
- Move to a simple enum instead of the mix between enum and bitfield.
- Update ts_info and ts-set in software timestamping case.
Changes in v5:
- Update to ndo_hwstamp_get/set. This bring several new patches.
- Add few patches to make the glue.
- Convert macb to ndo_hwstamp_get/set.
- Add netlink specs description of new ethtool commands.
- Removed netdev notifier.
- Split the patches that expose the timestamping to userspace to separate
the core and ethtool development.
- Add description of software timestamping.
- Convert PHYs hwtstamp callback to use kernel_hwtstamp_config.
Changes in v4:
- Move on to ethtool netlink instead of ioctl.
- Add a netdev notifier to allow packet trapping by the MAC in case of PHY
time stamping.
- Add a PHY whitelist to not break the old PHY default time-stamping
preference API.
Changes in v3:
- Expose the PTP choice to ethtool instead of sysfs.
You can test it with the ethtool source on branch feature_ptp of:
https://github.com/kmaincent/ethtool
- Added a devicetree binding to select the preferred timestamp.
Changes in v2:
- Move selected_timestamping_layer variable of the concerned patch.
- Use sysfs_streq instead of strmcmp.
- Use the PHY timestamp only if available.
Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
---
Kory Maincent (14):
net_tstamp: Add TIMESTAMPING SOFTWARE and HARDWARE mask
net: Make dev_get_hwtstamp_phylib accessible
net: Make net_hwtstamp_validate accessible
net: Change the API of PHY default timestamp to MAC
net: net_tstamp: Add unspec field to hwtstamp_source enumeration
net: Add struct kernel_ethtool_ts_info
ptp: Add phc source and helpers to register specific PTP clock or get information
net: Add the possibility to support a selected hwtstamp in netdevice
net: netdevsim: ptp_mock: Convert to netdev_ptp_clock_register
net: macb: Convert to netdev_ptp_clock_register
net: ptp: Move ptp_clock_index() to builtin symbol
net: ethtool: tsinfo: Add support for reading tsinfo for a specific hwtstamp provider
net: ethtool: Add support for tsconfig command to get/set hwtstamp config
netlink: specs: Enhance tsinfo netlink attributes and add a tsconfig set command
Documentation/netlink/specs/ethtool.yaml | 73 +++++
Documentation/networking/ethtool-netlink.rst | 7 +-
Documentation/networking/timestamping.rst | 33 +-
drivers/net/bonding/bond_main.c | 4 +-
drivers/net/can/dev/dev.c | 2 +-
drivers/net/can/peak_canfd/peak_canfd.c | 2 +-
drivers/net/can/usb/gs_usb.c | 2 +-
drivers/net/can/usb/peak_usb/pcan_usb_core.c | 2 +-
drivers/net/can/usb/peak_usb/pcan_usb_core.h | 2 +-
drivers/net/dsa/hirschmann/hellcreek_hwtstamp.c | 2 +-
drivers/net/dsa/hirschmann/hellcreek_hwtstamp.h | 2 +-
drivers/net/dsa/microchip/ksz_ptp.c | 2 +-
drivers/net/dsa/microchip/ksz_ptp.h | 2 +-
drivers/net/dsa/mv88e6xxx/hwtstamp.c | 2 +-
drivers/net/dsa/mv88e6xxx/hwtstamp.h | 4 +-
drivers/net/dsa/ocelot/felix.c | 2 +-
drivers/net/dsa/sja1105/sja1105_ptp.c | 2 +-
drivers/net/dsa/sja1105/sja1105_ptp.h | 2 +-
drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c | 2 +-
.../net/ethernet/aquantia/atlantic/aq_ethtool.c | 2 +-
.../net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c | 2 +-
drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 2 +-
drivers/net/ethernet/broadcom/tg3.c | 2 +-
drivers/net/ethernet/cadence/macb.h | 2 +-
drivers/net/ethernet/cadence/macb_main.c | 4 +-
drivers/net/ethernet/cadence/macb_ptp.c | 2 +-
drivers/net/ethernet/cavium/liquidio/lio_ethtool.c | 2 +-
.../net/ethernet/cavium/thunder/nicvf_ethtool.c | 2 +-
drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c | 2 +-
drivers/net/ethernet/cisco/enic/enic_ethtool.c | 2 +-
drivers/net/ethernet/engleder/tsnep_ethtool.c | 2 +-
drivers/net/ethernet/freescale/dpaa/dpaa_ethtool.c | 2 +-
.../net/ethernet/freescale/dpaa2/dpaa2-ethtool.c | 2 +-
.../net/ethernet/freescale/enetc/enetc_ethtool.c | 2 +-
drivers/net/ethernet/freescale/fec_main.c | 2 +-
drivers/net/ethernet/freescale/gianfar_ethtool.c | 2 +-
.../net/ethernet/fungible/funeth/funeth_ethtool.c | 2 +-
drivers/net/ethernet/hisilicon/hns3/hnae3.h | 2 +-
drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c | 2 +-
.../net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.c | 2 +-
.../net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.h | 2 +-
drivers/net/ethernet/intel/e1000e/ethtool.c | 2 +-
drivers/net/ethernet/intel/i40e/i40e_ethtool.c | 2 +-
drivers/net/ethernet/intel/ice/ice_ethtool.c | 2 +-
drivers/net/ethernet/intel/igb/igb_ethtool.c | 2 +-
drivers/net/ethernet/intel/igc/igc_ethtool.c | 2 +-
drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c | 2 +-
drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c | 2 +-
.../ethernet/marvell/octeontx2/nic/otx2_ethtool.c | 2 +-
drivers/net/ethernet/mellanox/mlx4/en_ethtool.c | 2 +-
drivers/net/ethernet/mellanox/mlx5/core/en.h | 2 +-
.../net/ethernet/mellanox/mlx5/core/en_ethtool.c | 4 +-
.../ethernet/mellanox/mlx5/core/ipoib/ethtool.c | 2 +-
drivers/net/ethernet/mellanox/mlxsw/spectrum.h | 2 +-
.../net/ethernet/mellanox/mlxsw/spectrum_ethtool.c | 2 +-
drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.c | 4 +-
drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.h | 10 +-
drivers/net/ethernet/microchip/lan743x_ethtool.c | 2 +-
.../ethernet/microchip/lan966x/lan966x_ethtool.c | 2 +-
.../net/ethernet/microchip/sparx5/sparx5_ethtool.c | 2 +-
drivers/net/ethernet/mscc/ocelot_net.c | 2 +-
drivers/net/ethernet/mscc/ocelot_ptp.c | 2 +-
.../net/ethernet/pensando/ionic/ionic_ethtool.c | 2 +-
drivers/net/ethernet/qlogic/qede/qede_ethtool.c | 2 +-
drivers/net/ethernet/qlogic/qede/qede_ptp.c | 2 +-
drivers/net/ethernet/qlogic/qede/qede_ptp.h | 2 +-
drivers/net/ethernet/renesas/ravb_main.c | 2 +-
drivers/net/ethernet/renesas/rswitch.c | 2 +-
drivers/net/ethernet/renesas/rtsn.c | 2 +-
drivers/net/ethernet/sfc/ethtool.c | 2 +-
drivers/net/ethernet/sfc/falcon/nic.h | 2 +-
drivers/net/ethernet/sfc/ptp.c | 2 +-
drivers/net/ethernet/sfc/ptp.h | 5 +-
drivers/net/ethernet/sfc/siena/ethtool.c | 2 +-
drivers/net/ethernet/sfc/siena/ptp.c | 2 +-
drivers/net/ethernet/sfc/siena/ptp.h | 4 +-
.../net/ethernet/stmicro/stmmac/stmmac_ethtool.c | 2 +-
drivers/net/ethernet/ti/am65-cpsw-ethtool.c | 2 +-
drivers/net/ethernet/ti/cpsw_ethtool.c | 4 +-
drivers/net/ethernet/ti/cpsw_priv.h | 2 +-
drivers/net/ethernet/ti/icssg/icssg_ethtool.c | 2 +-
drivers/net/ethernet/ti/netcp_ethss.c | 4 +-
drivers/net/ethernet/xscale/ixp4xx_eth.c | 2 +-
drivers/net/macvlan.c | 2 +-
drivers/net/netdevsim/ethtool.c | 2 +-
drivers/net/netdevsim/netdev.c | 19 +-
drivers/net/phy/bcm-phy-ptp.c | 5 +-
drivers/net/phy/dp83640.c | 4 +-
drivers/net/phy/micrel.c | 10 +-
drivers/net/phy/mscc/mscc_ptp.c | 5 +-
drivers/net/phy/nxp-c45-tja11xx.c | 5 +-
drivers/net/phy/phy_device.c | 11 +
drivers/ptp/Makefile | 5 +
drivers/ptp/ptp_clock.c | 39 ++-
drivers/ptp/ptp_clock_consumer.c | 182 +++++++++++
drivers/ptp/ptp_ines.c | 2 +-
drivers/ptp/ptp_mock.c | 4 +-
drivers/ptp/ptp_private.h | 7 +
drivers/s390/net/qeth_ethtool.c | 2 +-
include/linux/can/dev.h | 2 +-
include/linux/ethtool.h | 29 +-
include/linux/mii_timestamper.h | 2 +-
include/linux/net_tstamp.h | 27 ++
include/linux/netdevice.h | 5 +
include/linux/phy.h | 21 +-
include/linux/ptp_clock_kernel.h | 188 +++++++++++
include/linux/ptp_mock.h | 4 +-
include/net/dsa.h | 2 +-
include/soc/mscc/ocelot.h | 2 +-
include/uapi/linux/ethtool_netlink.h | 29 +-
include/uapi/linux/net_tstamp.h | 11 +
net/8021q/vlan_dev.c | 2 +-
net/core/dev.h | 3 +
net/core/dev_ioctl.c | 55 +++-
net/core/timestamping.c | 49 ++-
net/dsa/user.c | 2 +-
net/ethtool/Makefile | 3 +-
net/ethtool/common.c | 40 ++-
net/ethtool/common.h | 5 +-
net/ethtool/ioctl.c | 14 +-
net/ethtool/netlink.c | 26 +-
net/ethtool/netlink.h | 8 +-
net/ethtool/ts.h | 52 +++
net/ethtool/tsconfig.c | 347 +++++++++++++++++++++
net/ethtool/tsinfo.c | 245 ++++++++++++++-
net/sched/sch_taprio.c | 2 +-
126 files changed, 1595 insertions(+), 184 deletions(-)
---
base-commit: ca778dde1ebe2816020d6fc53a9a715e4f9db417
change-id: 20231011-feature_ptp_netnext-3f278578e84b
Best regards,
--
Köry Maincent, Bootlin
Embedded Linux and kernel engineering
https://bootlin.com
^ permalink raw reply [flat|nested] 39+ messages in thread
* [PATCH net-next v17 01/14] net_tstamp: Add TIMESTAMPING SOFTWARE and HARDWARE mask
2024-07-09 13:53 [PATCH net-next v17 00/14] net: Make timestamping selectable Kory Maincent
@ 2024-07-09 13:53 ` Kory Maincent
2024-07-15 23:28 ` Jacob Keller
2024-07-09 13:53 ` [PATCH net-next v17 02/14] net: Make dev_get_hwtstamp_phylib accessible Kory Maincent
` (13 subsequent siblings)
14 siblings, 1 reply; 39+ messages in thread
From: Kory Maincent @ 2024-07-09 13:53 UTC (permalink / raw)
To: Florian Fainelli, Broadcom internal kernel review list,
Andrew Lunn, Heiner Kallweit, Russell King, David S. Miller,
Eric Dumazet, Jakub Kicinski, Paolo Abeni, Richard Cochran,
Radu Pirea, Jay Vosburgh, Andy Gospodarek, Nicolas Ferre,
Claudiu Beznea, Willem de Bruijn, Jonathan Corbet, Horatiu Vultur,
UNGLinuxDriver, Simon Horman, Vladimir Oltean, donald.hunter,
danieller, ecree.xilinx
Cc: Thomas Petazzoni, linux-kernel, netdev, linux-doc,
Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter, Kory Maincent
Timestamping software or hardware flags are often used as a group,
therefore adding these masks will ease future use.
I did not use SOF_TIMESTAMPING_SYS_HARDWARE flag as it is deprecated and
not used at all.
Reviewed-by: Willem de Bruijn <willemb@google.com>
Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
---
Changes in v7:
- Move the masks out of uapi to include/linux/net_tstamp.h
Changes in v9:
- Fix commit message typos
---
include/linux/net_tstamp.h | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/include/linux/net_tstamp.h b/include/linux/net_tstamp.h
index eb01c37e71e0..3799c79b6c83 100644
--- a/include/linux/net_tstamp.h
+++ b/include/linux/net_tstamp.h
@@ -5,6 +5,14 @@
#include <uapi/linux/net_tstamp.h>
+#define SOF_TIMESTAMPING_SOFTWARE_MASK (SOF_TIMESTAMPING_RX_SOFTWARE | \
+ SOF_TIMESTAMPING_TX_SOFTWARE | \
+ SOF_TIMESTAMPING_SOFTWARE)
+
+#define SOF_TIMESTAMPING_HARDWARE_MASK (SOF_TIMESTAMPING_RX_HARDWARE | \
+ SOF_TIMESTAMPING_TX_HARDWARE | \
+ SOF_TIMESTAMPING_RAW_HARDWARE)
+
enum hwtstamp_source {
HWTSTAMP_SOURCE_NETDEV,
HWTSTAMP_SOURCE_PHYLIB,
--
2.34.1
^ permalink raw reply related [flat|nested] 39+ messages in thread
* [PATCH net-next v17 02/14] net: Make dev_get_hwtstamp_phylib accessible
2024-07-09 13:53 [PATCH net-next v17 00/14] net: Make timestamping selectable Kory Maincent
2024-07-09 13:53 ` [PATCH net-next v17 01/14] net_tstamp: Add TIMESTAMPING SOFTWARE and HARDWARE mask Kory Maincent
@ 2024-07-09 13:53 ` Kory Maincent
2024-07-15 23:29 ` Jacob Keller
2024-07-09 13:53 ` [PATCH net-next v17 03/14] net: Make net_hwtstamp_validate accessible Kory Maincent
` (12 subsequent siblings)
14 siblings, 1 reply; 39+ messages in thread
From: Kory Maincent @ 2024-07-09 13:53 UTC (permalink / raw)
To: Florian Fainelli, Broadcom internal kernel review list,
Andrew Lunn, Heiner Kallweit, Russell King, David S. Miller,
Eric Dumazet, Jakub Kicinski, Paolo Abeni, Richard Cochran,
Radu Pirea, Jay Vosburgh, Andy Gospodarek, Nicolas Ferre,
Claudiu Beznea, Willem de Bruijn, Jonathan Corbet, Horatiu Vultur,
UNGLinuxDriver, Simon Horman, Vladimir Oltean, donald.hunter,
danieller, ecree.xilinx
Cc: Thomas Petazzoni, linux-kernel, netdev, linux-doc,
Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter, Kory Maincent
Make the dev_get_hwtstamp_phylib function accessible in prevision to use
it from ethtool to read the hwtstamp current configuration.
Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
---
Change in v8:
- New patch
Change in v10:
- Remove export symbol as ethtool can't be built as a module.
- Move the declaration to net/core/dev.h instead of netdevice.h
---
net/core/dev.h | 2 ++
net/core/dev_ioctl.c | 4 ++--
2 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/net/core/dev.h b/net/core/dev.h
index 5654325c5b71..9d4ceaf9bdc0 100644
--- a/net/core/dev.h
+++ b/net/core/dev.h
@@ -187,5 +187,7 @@ static inline void dev_xmit_recursion_dec(void)
int dev_set_hwtstamp_phylib(struct net_device *dev,
struct kernel_hwtstamp_config *cfg,
struct netlink_ext_ack *extack);
+int dev_get_hwtstamp_phylib(struct net_device *dev,
+ struct kernel_hwtstamp_config *cfg);
#endif
diff --git a/net/core/dev_ioctl.c b/net/core/dev_ioctl.c
index b9719ed3c3fd..b8cf8c55fa2d 100644
--- a/net/core/dev_ioctl.c
+++ b/net/core/dev_ioctl.c
@@ -268,8 +268,8 @@ static int dev_eth_ioctl(struct net_device *dev,
* -EOPNOTSUPP for phylib for now, which is still more accurate than letting
* the netdev handle the GET request.
*/
-static int dev_get_hwtstamp_phylib(struct net_device *dev,
- struct kernel_hwtstamp_config *cfg)
+int dev_get_hwtstamp_phylib(struct net_device *dev,
+ struct kernel_hwtstamp_config *cfg)
{
if (phy_has_hwtstamp(dev->phydev))
return phy_hwtstamp_get(dev->phydev, cfg);
--
2.34.1
^ permalink raw reply related [flat|nested] 39+ messages in thread
* [PATCH net-next v17 03/14] net: Make net_hwtstamp_validate accessible
2024-07-09 13:53 [PATCH net-next v17 00/14] net: Make timestamping selectable Kory Maincent
2024-07-09 13:53 ` [PATCH net-next v17 01/14] net_tstamp: Add TIMESTAMPING SOFTWARE and HARDWARE mask Kory Maincent
2024-07-09 13:53 ` [PATCH net-next v17 02/14] net: Make dev_get_hwtstamp_phylib accessible Kory Maincent
@ 2024-07-09 13:53 ` Kory Maincent
2024-07-15 23:30 ` Jacob Keller
2024-07-09 13:53 ` [PATCH net-next v17 04/14] net: Change the API of PHY default timestamp to MAC Kory Maincent
` (11 subsequent siblings)
14 siblings, 1 reply; 39+ messages in thread
From: Kory Maincent @ 2024-07-09 13:53 UTC (permalink / raw)
To: Florian Fainelli, Broadcom internal kernel review list,
Andrew Lunn, Heiner Kallweit, Russell King, David S. Miller,
Eric Dumazet, Jakub Kicinski, Paolo Abeni, Richard Cochran,
Radu Pirea, Jay Vosburgh, Andy Gospodarek, Nicolas Ferre,
Claudiu Beznea, Willem de Bruijn, Jonathan Corbet, Horatiu Vultur,
UNGLinuxDriver, Simon Horman, Vladimir Oltean, donald.hunter,
danieller, ecree.xilinx
Cc: Thomas Petazzoni, linux-kernel, netdev, linux-doc,
Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter, Kory Maincent
Make the net_hwtstamp_validate function accessible in prevision to use
it from ethtool to validate the hwtstamp configuration before setting it.
Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
---
Change in v8:
- New patch
Change in v10:
- Remove export symbol as ethtool can't be built as a module.
- Move the declaration to net/core/dev.h instead of netdevice.h
---
net/core/dev.h | 1 +
net/core/dev_ioctl.c | 2 +-
2 files changed, 2 insertions(+), 1 deletion(-)
diff --git a/net/core/dev.h b/net/core/dev.h
index 9d4ceaf9bdc0..df085c3e510c 100644
--- a/net/core/dev.h
+++ b/net/core/dev.h
@@ -189,5 +189,6 @@ int dev_set_hwtstamp_phylib(struct net_device *dev,
struct netlink_ext_ack *extack);
int dev_get_hwtstamp_phylib(struct net_device *dev,
struct kernel_hwtstamp_config *cfg);
+int net_hwtstamp_validate(const struct kernel_hwtstamp_config *cfg);
#endif
diff --git a/net/core/dev_ioctl.c b/net/core/dev_ioctl.c
index b8cf8c55fa2d..6aaa8326bf8f 100644
--- a/net/core/dev_ioctl.c
+++ b/net/core/dev_ioctl.c
@@ -184,7 +184,7 @@ static int dev_ifsioc_locked(struct net *net, struct ifreq *ifr, unsigned int cm
return err;
}
-static int net_hwtstamp_validate(const struct kernel_hwtstamp_config *cfg)
+int net_hwtstamp_validate(const struct kernel_hwtstamp_config *cfg)
{
enum hwtstamp_tx_types tx_type;
enum hwtstamp_rx_filters rx_filter;
--
2.34.1
^ permalink raw reply related [flat|nested] 39+ messages in thread
* [PATCH net-next v17 04/14] net: Change the API of PHY default timestamp to MAC
2024-07-09 13:53 [PATCH net-next v17 00/14] net: Make timestamping selectable Kory Maincent
` (2 preceding siblings ...)
2024-07-09 13:53 ` [PATCH net-next v17 03/14] net: Make net_hwtstamp_validate accessible Kory Maincent
@ 2024-07-09 13:53 ` Kory Maincent
2024-07-15 23:37 ` Jacob Keller
2024-07-09 13:53 ` [PATCH net-next v17 05/14] net: net_tstamp: Add unspec field to hwtstamp_source enumeration Kory Maincent
` (10 subsequent siblings)
14 siblings, 1 reply; 39+ messages in thread
From: Kory Maincent @ 2024-07-09 13:53 UTC (permalink / raw)
To: Florian Fainelli, Broadcom internal kernel review list,
Andrew Lunn, Heiner Kallweit, Russell King, David S. Miller,
Eric Dumazet, Jakub Kicinski, Paolo Abeni, Richard Cochran,
Radu Pirea, Jay Vosburgh, Andy Gospodarek, Nicolas Ferre,
Claudiu Beznea, Willem de Bruijn, Jonathan Corbet, Horatiu Vultur,
UNGLinuxDriver, Simon Horman, Vladimir Oltean, donald.hunter,
danieller, ecree.xilinx
Cc: Thomas Petazzoni, linux-kernel, netdev, linux-doc,
Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter, Kory Maincent
Change the API to select MAC default time stamping instead of the PHY.
Indeed the PHY is closer to the wire therefore theoretically it has less
delay than the MAC timestamping but the reality is different. Due to lower
time stamping clock frequency, latency in the MDIO bus and no PHC hardware
synchronization between different PHY, the PHY PTP is often less precise
than the MAC. The exception is for PHY designed specially for PTP case but
these devices are not very widespread. For not breaking the compatibility
default_timestamp flag has been introduced in phy_device that is set by
the phy driver to know we are using the old API behavior.
Reviewed-by: Rahul Rameshbabu <rrameshbabu@nvidia.com>
Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
---
Changes in v5:
- Extract the API change in this patch.
- Rename whitelist to allowlist.
- Set NETDEV_TIMESTAMPING in register_netdevice function.
- Add software timestamping case description in ts_info.
Change in v6:
- Replace the allowlist phy with a default_timestamp flag to know which
phy is using old API behavior.
- Fix dereferenced of a possible null pointer.
- Follow timestamping layer naming update.
- Update timestamp default set between MAC and software.
- Update ts_info returned in case of software timestamping.
Change in v8:
- Reform the implementation to use a simple phy_is_default_hwtstamp helper
instead of saving the hwtstamp in the net_device struct.
Change in v9:
- Update few nit following the review.
Change in v12:
- Add missing return description in the kdoc.
---
drivers/net/phy/bcm-phy-ptp.c | 3 +++
drivers/net/phy/dp83640.c | 2 ++
drivers/net/phy/micrel.c | 6 ++++++
drivers/net/phy/mscc/mscc_ptp.c | 3 +++
drivers/net/phy/nxp-c45-tja11xx.c | 3 +++
include/linux/phy.h | 19 +++++++++++++++++++
net/core/dev_ioctl.c | 8 +++-----
net/core/timestamping.c | 5 +++--
net/ethtool/common.c | 2 +-
9 files changed, 43 insertions(+), 8 deletions(-)
diff --git a/drivers/net/phy/bcm-phy-ptp.c b/drivers/net/phy/bcm-phy-ptp.c
index 617d384d4551..d3e825c951ee 100644
--- a/drivers/net/phy/bcm-phy-ptp.c
+++ b/drivers/net/phy/bcm-phy-ptp.c
@@ -931,6 +931,9 @@ struct bcm_ptp_private *bcm_ptp_probe(struct phy_device *phydev)
return ERR_CAST(clock);
priv->ptp_clock = clock;
+ /* Timestamp selected by default to keep legacy API */
+ phydev->default_timestamp = true;
+
priv->phydev = phydev;
bcm_ptp_init(priv);
diff --git a/drivers/net/phy/dp83640.c b/drivers/net/phy/dp83640.c
index 5c42c47dc564..d3e72d5c1472 100644
--- a/drivers/net/phy/dp83640.c
+++ b/drivers/net/phy/dp83640.c
@@ -1447,6 +1447,8 @@ static int dp83640_probe(struct phy_device *phydev)
for (i = 0; i < MAX_RXTS; i++)
list_add(&dp83640->rx_pool_data[i].list, &dp83640->rxpool);
+ /* Timestamp selected by default to keep legacy API */
+ phydev->default_timestamp = true;
phydev->mii_ts = &dp83640->mii_ts;
phydev->priv = dp83640;
diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c
index ebafedde0ab7..8d57225d8575 100644
--- a/drivers/net/phy/micrel.c
+++ b/drivers/net/phy/micrel.c
@@ -3781,6 +3781,9 @@ static void lan8814_ptp_init(struct phy_device *phydev)
ptp_priv->mii_ts.ts_info = lan8814_ts_info;
phydev->mii_ts = &ptp_priv->mii_ts;
+
+ /* Timestamp selected by default to keep legacy API */
+ phydev->default_timestamp = true;
}
static int lan8814_ptp_probe_once(struct phy_device *phydev)
@@ -5279,6 +5282,9 @@ static int lan8841_probe(struct phy_device *phydev)
phydev->mii_ts = &ptp_priv->mii_ts;
+ /* Timestamp selected by default to keep legacy API */
+ phydev->default_timestamp = true;
+
return 0;
}
diff --git a/drivers/net/phy/mscc/mscc_ptp.c b/drivers/net/phy/mscc/mscc_ptp.c
index eb0b032cb613..e66d20eff7c4 100644
--- a/drivers/net/phy/mscc/mscc_ptp.c
+++ b/drivers/net/phy/mscc/mscc_ptp.c
@@ -1570,6 +1570,9 @@ int vsc8584_ptp_probe(struct phy_device *phydev)
return PTR_ERR(vsc8531->load_save);
}
+ /* Timestamp selected by default to keep legacy API */
+ phydev->default_timestamp = true;
+
vsc8531->ptp->phydev = phydev;
return 0;
diff --git a/drivers/net/phy/nxp-c45-tja11xx.c b/drivers/net/phy/nxp-c45-tja11xx.c
index 3cf614b4cd52..d18c133e6013 100644
--- a/drivers/net/phy/nxp-c45-tja11xx.c
+++ b/drivers/net/phy/nxp-c45-tja11xx.c
@@ -1660,6 +1660,9 @@ static int nxp_c45_probe(struct phy_device *phydev)
priv->mii_ts.ts_info = nxp_c45_ts_info;
phydev->mii_ts = &priv->mii_ts;
ret = nxp_c45_init_ptp_clock(priv);
+
+ /* Timestamp selected by default to keep legacy API */
+ phydev->default_timestamp = true;
} else {
phydev_dbg(phydev, "PTP support not enabled even if the phy supports it");
}
diff --git a/include/linux/phy.h b/include/linux/phy.h
index bd68f9d8e74f..e7a38137211c 100644
--- a/include/linux/phy.h
+++ b/include/linux/phy.h
@@ -616,6 +616,8 @@ struct macsec_ops;
* handling shall be postponed until PHY has resumed
* @irq_rerun: Flag indicating interrupts occurred while PHY was suspended,
* requiring a rerun of the interrupt handler after resume
+ * @default_timestamp: Flag indicating whether we are using the phy
+ * timestamp as the default one
* @interface: enum phy_interface_t value
* @possible_interfaces: bitmap if interface modes that the attached PHY
* will switch between depending on media speed.
@@ -681,6 +683,8 @@ struct phy_device {
unsigned irq_suspended:1;
unsigned irq_rerun:1;
+ unsigned default_timestamp:1;
+
int rate_matching;
enum phy_state state;
@@ -1625,6 +1629,21 @@ static inline void phy_txtstamp(struct phy_device *phydev, struct sk_buff *skb,
phydev->mii_ts->txtstamp(phydev->mii_ts, skb, type);
}
+/**
+ * phy_is_default_hwtstamp - Is the PHY hwtstamp the default timestamp
+ * @phydev: Pointer to phy_device
+ *
+ * This is used to get default timestamping device taking into account
+ * the new API choice, which is selecting the timestamping from MAC by
+ * default if the phydev does not have default_timestamp flag enabled.
+ *
+ * Return: True if phy is the default hw timestamp, false otherwise.
+ */
+static inline bool phy_is_default_hwtstamp(struct phy_device *phydev)
+{
+ return phy_has_hwtstamp(phydev) && phydev->default_timestamp;
+}
+
/**
* phy_is_internal - Convenience function for testing if a PHY is internal
* @phydev: the phy_device struct
diff --git a/net/core/dev_ioctl.c b/net/core/dev_ioctl.c
index 6aaa8326bf8f..36cea843381f 100644
--- a/net/core/dev_ioctl.c
+++ b/net/core/dev_ioctl.c
@@ -259,9 +259,7 @@ static int dev_eth_ioctl(struct net_device *dev,
* @dev: Network device
* @cfg: Timestamping configuration structure
*
- * Helper for enforcing a common policy that phylib timestamping, if available,
- * should take precedence in front of hardware timestamping provided by the
- * netdev.
+ * Helper for calling the default hardware provider timestamping.
*
* Note: phy_mii_ioctl() only handles SIOCSHWTSTAMP (not SIOCGHWTSTAMP), and
* there only exists a phydev->mii_ts->hwtstamp() method. So this will return
@@ -271,7 +269,7 @@ static int dev_eth_ioctl(struct net_device *dev,
int dev_get_hwtstamp_phylib(struct net_device *dev,
struct kernel_hwtstamp_config *cfg)
{
- if (phy_has_hwtstamp(dev->phydev))
+ if (phy_is_default_hwtstamp(dev->phydev))
return phy_hwtstamp_get(dev->phydev, cfg);
return dev->netdev_ops->ndo_hwtstamp_get(dev, cfg);
@@ -327,7 +325,7 @@ int dev_set_hwtstamp_phylib(struct net_device *dev,
struct netlink_ext_ack *extack)
{
const struct net_device_ops *ops = dev->netdev_ops;
- bool phy_ts = phy_has_hwtstamp(dev->phydev);
+ bool phy_ts = phy_is_default_hwtstamp(dev->phydev);
struct kernel_hwtstamp_config old_cfg = {};
bool changed = false;
int err;
diff --git a/net/core/timestamping.c b/net/core/timestamping.c
index 04840697fe79..3717fb152ecc 100644
--- a/net/core/timestamping.c
+++ b/net/core/timestamping.c
@@ -25,7 +25,8 @@ void skb_clone_tx_timestamp(struct sk_buff *skb)
struct sk_buff *clone;
unsigned int type;
- if (!skb->sk)
+ if (!skb->sk || !skb->dev ||
+ !phy_is_default_hwtstamp(skb->dev->phydev))
return;
type = classify(skb);
@@ -47,7 +48,7 @@ bool skb_defer_rx_timestamp(struct sk_buff *skb)
struct mii_timestamper *mii_ts;
unsigned int type;
- if (!skb->dev || !skb->dev->phydev || !skb->dev->phydev->mii_ts)
+ if (!skb->dev || !phy_is_default_hwtstamp(skb->dev->phydev))
return false;
if (skb_headroom(skb) < ETH_HLEN)
diff --git a/net/ethtool/common.c b/net/ethtool/common.c
index 6b2a360dcdf0..01b7550f12c6 100644
--- a/net/ethtool/common.c
+++ b/net/ethtool/common.c
@@ -637,7 +637,7 @@ int __ethtool_get_ts_info(struct net_device *dev, struct ethtool_ts_info *info)
memset(info, 0, sizeof(*info));
info->cmd = ETHTOOL_GET_TS_INFO;
- if (phy_has_tsinfo(phydev))
+ if (phy_is_default_hwtstamp(phydev) && phy_has_tsinfo(phydev))
return phy_ts_info(phydev, info);
if (ops->get_ts_info)
return ops->get_ts_info(dev, info);
--
2.34.1
^ permalink raw reply related [flat|nested] 39+ messages in thread
* [PATCH net-next v17 05/14] net: net_tstamp: Add unspec field to hwtstamp_source enumeration
2024-07-09 13:53 [PATCH net-next v17 00/14] net: Make timestamping selectable Kory Maincent
` (3 preceding siblings ...)
2024-07-09 13:53 ` [PATCH net-next v17 04/14] net: Change the API of PHY default timestamp to MAC Kory Maincent
@ 2024-07-09 13:53 ` Kory Maincent
2024-07-15 23:37 ` Jacob Keller
2024-07-09 13:53 ` [PATCH net-next v17 06/14] net: Add struct kernel_ethtool_ts_info Kory Maincent
` (9 subsequent siblings)
14 siblings, 1 reply; 39+ messages in thread
From: Kory Maincent @ 2024-07-09 13:53 UTC (permalink / raw)
To: Florian Fainelli, Broadcom internal kernel review list,
Andrew Lunn, Heiner Kallweit, Russell King, David S. Miller,
Eric Dumazet, Jakub Kicinski, Paolo Abeni, Richard Cochran,
Radu Pirea, Jay Vosburgh, Andy Gospodarek, Nicolas Ferre,
Claudiu Beznea, Willem de Bruijn, Jonathan Corbet, Horatiu Vultur,
UNGLinuxDriver, Simon Horman, Vladimir Oltean, donald.hunter,
danieller, ecree.xilinx
Cc: Thomas Petazzoni, linux-kernel, netdev, linux-doc,
Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter, Kory Maincent
Prepare for future support of saving hwtstamp source in PTP xarray by
introducing HWTSTAMP_SOURCE_UNSPEC to hwtstamp_source enum, setting it
to 0 to match old behavior of no source defined.
Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
---
Change in v8:
- New patch
---
include/linux/net_tstamp.h | 1 +
1 file changed, 1 insertion(+)
diff --git a/include/linux/net_tstamp.h b/include/linux/net_tstamp.h
index 3799c79b6c83..662074b08c94 100644
--- a/include/linux/net_tstamp.h
+++ b/include/linux/net_tstamp.h
@@ -14,6 +14,7 @@
SOF_TIMESTAMPING_RAW_HARDWARE)
enum hwtstamp_source {
+ HWTSTAMP_SOURCE_UNSPEC,
HWTSTAMP_SOURCE_NETDEV,
HWTSTAMP_SOURCE_PHYLIB,
};
--
2.34.1
^ permalink raw reply related [flat|nested] 39+ messages in thread
* [PATCH net-next v17 06/14] net: Add struct kernel_ethtool_ts_info
2024-07-09 13:53 [PATCH net-next v17 00/14] net: Make timestamping selectable Kory Maincent
` (4 preceding siblings ...)
2024-07-09 13:53 ` [PATCH net-next v17 05/14] net: net_tstamp: Add unspec field to hwtstamp_source enumeration Kory Maincent
@ 2024-07-09 13:53 ` Kory Maincent
2024-07-15 23:41 ` Jacob Keller
2024-07-09 13:53 ` [PATCH net-next v17 07/14] ptp: Add phc source and helpers to register specific PTP clock or get information Kory Maincent
` (8 subsequent siblings)
14 siblings, 1 reply; 39+ messages in thread
From: Kory Maincent @ 2024-07-09 13:53 UTC (permalink / raw)
To: Florian Fainelli, Broadcom internal kernel review list,
Andrew Lunn, Heiner Kallweit, Russell King, David S. Miller,
Eric Dumazet, Jakub Kicinski, Paolo Abeni, Richard Cochran,
Radu Pirea, Jay Vosburgh, Andy Gospodarek, Nicolas Ferre,
Claudiu Beznea, Willem de Bruijn, Jonathan Corbet, Horatiu Vultur,
UNGLinuxDriver, Simon Horman, Vladimir Oltean, donald.hunter,
danieller, ecree.xilinx
Cc: Thomas Petazzoni, linux-kernel, netdev, linux-doc,
Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter, Kory Maincent
In prevision to add new UAPI for hwtstamp we will be limited to the struct
ethtool_ts_info that is currently passed in fixed binary format through the
ETHTOOL_GET_TS_INFO ethtool ioctl. It would be good if new kernel code
already started operating on an extensible kernel variant of that
structure, similar in concept to struct kernel_hwtstamp_config vs struct
hwtstamp_config.
Since struct ethtool_ts_info is in include/uapi/linux/ethtool.h, here
we introduce the kernel-only structure in include/linux/ethtool.h.
The manual copy is then made in the function called by ETHTOOL_GET_TS_INFO.
Acked-by: Shannon Nelson <shannon.nelson@amd.com>
Acked-by: Alexandra Winter <wintera@linux.ibm.com>
Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
---
Change in v8:
- New patch
Change in v11:
- Remove a change of my out of tree patch.
Change in v15:
- Fix uninitialized ethtool_ts_info structure.
Change in v17:
- Add a missing kernel_ethtool_tsinfo from a new MAC driver.
---
drivers/net/bonding/bond_main.c | 4 ++--
drivers/net/can/dev/dev.c | 2 +-
drivers/net/can/peak_canfd/peak_canfd.c | 2 +-
drivers/net/can/usb/gs_usb.c | 2 +-
drivers/net/can/usb/peak_usb/pcan_usb_core.c | 2 +-
drivers/net/can/usb/peak_usb/pcan_usb_core.h | 2 +-
drivers/net/dsa/hirschmann/hellcreek_hwtstamp.c | 2 +-
drivers/net/dsa/hirschmann/hellcreek_hwtstamp.h | 2 +-
drivers/net/dsa/microchip/ksz_ptp.c | 2 +-
drivers/net/dsa/microchip/ksz_ptp.h | 2 +-
drivers/net/dsa/mv88e6xxx/hwtstamp.c | 2 +-
drivers/net/dsa/mv88e6xxx/hwtstamp.h | 4 ++--
drivers/net/dsa/ocelot/felix.c | 2 +-
drivers/net/dsa/sja1105/sja1105_ptp.c | 2 +-
drivers/net/dsa/sja1105/sja1105_ptp.h | 2 +-
drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c | 2 +-
.../net/ethernet/aquantia/atlantic/aq_ethtool.c | 2 +-
.../net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c | 2 +-
drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c | 2 +-
drivers/net/ethernet/broadcom/tg3.c | 2 +-
drivers/net/ethernet/cadence/macb.h | 2 +-
drivers/net/ethernet/cadence/macb_main.c | 4 ++--
drivers/net/ethernet/cavium/liquidio/lio_ethtool.c | 2 +-
.../net/ethernet/cavium/thunder/nicvf_ethtool.c | 2 +-
drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c | 2 +-
drivers/net/ethernet/cisco/enic/enic_ethtool.c | 2 +-
drivers/net/ethernet/engleder/tsnep_ethtool.c | 2 +-
drivers/net/ethernet/freescale/dpaa/dpaa_ethtool.c | 2 +-
.../net/ethernet/freescale/dpaa2/dpaa2-ethtool.c | 2 +-
.../net/ethernet/freescale/enetc/enetc_ethtool.c | 2 +-
drivers/net/ethernet/freescale/fec_main.c | 2 +-
drivers/net/ethernet/freescale/gianfar_ethtool.c | 2 +-
.../net/ethernet/fungible/funeth/funeth_ethtool.c | 2 +-
drivers/net/ethernet/hisilicon/hns3/hnae3.h | 2 +-
drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c | 2 +-
.../net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.c | 2 +-
.../net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.h | 2 +-
drivers/net/ethernet/intel/e1000e/ethtool.c | 2 +-
drivers/net/ethernet/intel/i40e/i40e_ethtool.c | 2 +-
drivers/net/ethernet/intel/ice/ice_ethtool.c | 2 +-
drivers/net/ethernet/intel/igb/igb_ethtool.c | 2 +-
drivers/net/ethernet/intel/igc/igc_ethtool.c | 2 +-
drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c | 2 +-
drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c | 2 +-
.../ethernet/marvell/octeontx2/nic/otx2_ethtool.c | 2 +-
drivers/net/ethernet/mellanox/mlx4/en_ethtool.c | 2 +-
drivers/net/ethernet/mellanox/mlx5/core/en.h | 2 +-
.../net/ethernet/mellanox/mlx5/core/en_ethtool.c | 4 ++--
.../ethernet/mellanox/mlx5/core/ipoib/ethtool.c | 2 +-
drivers/net/ethernet/mellanox/mlxsw/spectrum.h | 2 +-
.../net/ethernet/mellanox/mlxsw/spectrum_ethtool.c | 2 +-
drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.c | 4 ++--
drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.h | 10 ++++-----
drivers/net/ethernet/microchip/lan743x_ethtool.c | 2 +-
.../ethernet/microchip/lan966x/lan966x_ethtool.c | 2 +-
.../net/ethernet/microchip/sparx5/sparx5_ethtool.c | 2 +-
drivers/net/ethernet/mscc/ocelot_net.c | 2 +-
drivers/net/ethernet/mscc/ocelot_ptp.c | 2 +-
.../net/ethernet/pensando/ionic/ionic_ethtool.c | 2 +-
drivers/net/ethernet/qlogic/qede/qede_ethtool.c | 2 +-
drivers/net/ethernet/qlogic/qede/qede_ptp.c | 2 +-
drivers/net/ethernet/qlogic/qede/qede_ptp.h | 2 +-
drivers/net/ethernet/renesas/ravb_main.c | 2 +-
drivers/net/ethernet/renesas/rswitch.c | 2 +-
drivers/net/ethernet/renesas/rtsn.c | 2 +-
drivers/net/ethernet/sfc/ethtool.c | 2 +-
drivers/net/ethernet/sfc/falcon/nic.h | 2 +-
drivers/net/ethernet/sfc/ptp.c | 2 +-
drivers/net/ethernet/sfc/ptp.h | 5 +++--
drivers/net/ethernet/sfc/siena/ethtool.c | 2 +-
drivers/net/ethernet/sfc/siena/ptp.c | 2 +-
drivers/net/ethernet/sfc/siena/ptp.h | 4 ++--
.../net/ethernet/stmicro/stmmac/stmmac_ethtool.c | 2 +-
drivers/net/ethernet/ti/am65-cpsw-ethtool.c | 2 +-
drivers/net/ethernet/ti/cpsw_ethtool.c | 4 ++--
drivers/net/ethernet/ti/cpsw_priv.h | 2 +-
drivers/net/ethernet/ti/icssg/icssg_ethtool.c | 2 +-
drivers/net/ethernet/ti/netcp_ethss.c | 4 ++--
drivers/net/ethernet/xscale/ixp4xx_eth.c | 2 +-
drivers/net/macvlan.c | 2 +-
drivers/net/netdevsim/ethtool.c | 2 +-
drivers/net/phy/bcm-phy-ptp.c | 2 +-
drivers/net/phy/dp83640.c | 2 +-
drivers/net/phy/micrel.c | 4 ++--
drivers/net/phy/mscc/mscc_ptp.c | 2 +-
drivers/net/phy/nxp-c45-tja11xx.c | 2 +-
drivers/ptp/ptp_ines.c | 2 +-
drivers/s390/net/qeth_ethtool.c | 2 +-
include/linux/can/dev.h | 2 +-
include/linux/ethtool.h | 25 +++++++++++++++++++---
include/linux/mii_timestamper.h | 2 +-
include/linux/phy.h | 2 +-
include/net/dsa.h | 2 +-
include/soc/mscc/ocelot.h | 2 +-
net/8021q/vlan_dev.c | 2 +-
net/dsa/user.c | 2 +-
net/ethtool/common.c | 6 +++---
net/ethtool/common.h | 2 +-
net/ethtool/ioctl.c | 14 +++++++++---
net/ethtool/tsinfo.c | 6 +++---
net/sched/sch_taprio.c | 2 +-
101 files changed, 151 insertions(+), 123 deletions(-)
diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index d19aabf5d4fb..af9ddd3902cc 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -5755,10 +5755,10 @@ static void bond_ethtool_get_drvinfo(struct net_device *bond_dev,
}
static int bond_ethtool_get_ts_info(struct net_device *bond_dev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct bonding *bond = netdev_priv(bond_dev);
- struct ethtool_ts_info ts_info;
+ struct kernel_ethtool_ts_info ts_info;
struct net_device *real_dev;
bool sw_tx_support = false;
struct list_head *iter;
diff --git a/drivers/net/can/dev/dev.c b/drivers/net/can/dev/dev.c
index 83e724e0ab87..87828f953073 100644
--- a/drivers/net/can/dev/dev.c
+++ b/drivers/net/can/dev/dev.c
@@ -376,7 +376,7 @@ EXPORT_SYMBOL(can_eth_ioctl_hwts);
* supporting hardware timestamps
*/
int can_ethtool_op_get_ts_info_hwts(struct net_device *dev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
info->so_timestamping =
SOF_TIMESTAMPING_TX_SOFTWARE |
diff --git a/drivers/net/can/peak_canfd/peak_canfd.c b/drivers/net/can/peak_canfd/peak_canfd.c
index 31c9c127e24b..b50005397463 100644
--- a/drivers/net/can/peak_canfd/peak_canfd.c
+++ b/drivers/net/can/peak_canfd/peak_canfd.c
@@ -777,7 +777,7 @@ static const struct net_device_ops peak_canfd_netdev_ops = {
};
static int peak_get_ts_info(struct net_device *dev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
info->so_timestamping =
SOF_TIMESTAMPING_TX_SOFTWARE |
diff --git a/drivers/net/can/usb/gs_usb.c b/drivers/net/can/usb/gs_usb.c
index 340297e3bec7..bc86e9b329fd 100644
--- a/drivers/net/can/usb/gs_usb.c
+++ b/drivers/net/can/usb/gs_usb.c
@@ -1148,7 +1148,7 @@ static int gs_usb_set_phys_id(struct net_device *netdev,
}
static int gs_usb_get_ts_info(struct net_device *netdev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct gs_can *dev = netdev_priv(netdev);
diff --git a/drivers/net/can/usb/peak_usb/pcan_usb_core.c b/drivers/net/can/usb/peak_usb/pcan_usb_core.c
index 1efa39e134f4..3d68fef46ded 100644
--- a/drivers/net/can/usb/peak_usb/pcan_usb_core.c
+++ b/drivers/net/can/usb/peak_usb/pcan_usb_core.c
@@ -897,7 +897,7 @@ int peak_usb_set_eeprom(struct net_device *netdev,
return 0;
}
-int pcan_get_ts_info(struct net_device *dev, struct ethtool_ts_info *info)
+int pcan_get_ts_info(struct net_device *dev, struct kernel_ethtool_ts_info *info)
{
info->so_timestamping =
SOF_TIMESTAMPING_TX_SOFTWARE |
diff --git a/drivers/net/can/usb/peak_usb/pcan_usb_core.h b/drivers/net/can/usb/peak_usb/pcan_usb_core.h
index f6cf84bb718f..abab00930b9d 100644
--- a/drivers/net/can/usb/peak_usb/pcan_usb_core.h
+++ b/drivers/net/can/usb/peak_usb/pcan_usb_core.h
@@ -145,7 +145,7 @@ void peak_usb_get_ts_time(struct peak_time_ref *time_ref, u32 ts, ktime_t *tv);
int peak_usb_netif_rx_64(struct sk_buff *skb, u32 ts_low, u32 ts_high);
void peak_usb_async_complete(struct urb *urb);
void peak_usb_restart_complete(struct peak_usb_device *dev);
-int pcan_get_ts_info(struct net_device *dev, struct ethtool_ts_info *info);
+int pcan_get_ts_info(struct net_device *dev, struct kernel_ethtool_ts_info *info);
/* common 32-bit CAN channel ID ethtool management */
int peak_usb_get_eeprom_len(struct net_device *netdev);
diff --git a/drivers/net/dsa/hirschmann/hellcreek_hwtstamp.c b/drivers/net/dsa/hirschmann/hellcreek_hwtstamp.c
index bd7aacc71a63..ca2500aba96f 100644
--- a/drivers/net/dsa/hirschmann/hellcreek_hwtstamp.c
+++ b/drivers/net/dsa/hirschmann/hellcreek_hwtstamp.c
@@ -16,7 +16,7 @@
#include "hellcreek_ptp.h"
int hellcreek_get_ts_info(struct dsa_switch *ds, int port,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct hellcreek *hellcreek = ds->priv;
diff --git a/drivers/net/dsa/hirschmann/hellcreek_hwtstamp.h b/drivers/net/dsa/hirschmann/hellcreek_hwtstamp.h
index 71af77efb28b..7d88da2134f2 100644
--- a/drivers/net/dsa/hirschmann/hellcreek_hwtstamp.h
+++ b/drivers/net/dsa/hirschmann/hellcreek_hwtstamp.h
@@ -48,7 +48,7 @@ void hellcreek_port_txtstamp(struct dsa_switch *ds, int port,
struct sk_buff *skb);
int hellcreek_get_ts_info(struct dsa_switch *ds, int port,
- struct ethtool_ts_info *info);
+ struct kernel_ethtool_ts_info *info);
long hellcreek_hwtstamp_work(struct ptp_clock_info *ptp);
diff --git a/drivers/net/dsa/microchip/ksz_ptp.c b/drivers/net/dsa/microchip/ksz_ptp.c
index 1fe105913c75..f0bd46e5d4ec 100644
--- a/drivers/net/dsa/microchip/ksz_ptp.c
+++ b/drivers/net/dsa/microchip/ksz_ptp.c
@@ -293,7 +293,7 @@ static int ksz_ptp_enable_mode(struct ksz_device *dev)
/* The function is return back the capability of timestamping feature when
* requested through ethtool -T <interface> utility
*/
-int ksz_get_ts_info(struct dsa_switch *ds, int port, struct ethtool_ts_info *ts)
+int ksz_get_ts_info(struct dsa_switch *ds, int port, struct kernel_ethtool_ts_info *ts)
{
struct ksz_device *dev = ds->priv;
struct ksz_ptp_data *ptp_data;
diff --git a/drivers/net/dsa/microchip/ksz_ptp.h b/drivers/net/dsa/microchip/ksz_ptp.h
index 0ca8ca4f804e..2f1783c0d723 100644
--- a/drivers/net/dsa/microchip/ksz_ptp.h
+++ b/drivers/net/dsa/microchip/ksz_ptp.h
@@ -38,7 +38,7 @@ int ksz_ptp_clock_register(struct dsa_switch *ds);
void ksz_ptp_clock_unregister(struct dsa_switch *ds);
int ksz_get_ts_info(struct dsa_switch *ds, int port,
- struct ethtool_ts_info *ts);
+ struct kernel_ethtool_ts_info *ts);
int ksz_hwtstamp_get(struct dsa_switch *ds, int port, struct ifreq *ifr);
int ksz_hwtstamp_set(struct dsa_switch *ds, int port, struct ifreq *ifr);
void ksz_port_txtstamp(struct dsa_switch *ds, int port, struct sk_buff *skb);
diff --git a/drivers/net/dsa/mv88e6xxx/hwtstamp.c b/drivers/net/dsa/mv88e6xxx/hwtstamp.c
index 331b4ca089ff..49e6e1355142 100644
--- a/drivers/net/dsa/mv88e6xxx/hwtstamp.c
+++ b/drivers/net/dsa/mv88e6xxx/hwtstamp.c
@@ -64,7 +64,7 @@ static int mv88e6xxx_ptp_read(struct mv88e6xxx_chip *chip, int addr,
#define TX_TSTAMP_TIMEOUT msecs_to_jiffies(40)
int mv88e6xxx_get_ts_info(struct dsa_switch *ds, int port,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
const struct mv88e6xxx_ptp_ops *ptp_ops;
struct mv88e6xxx_chip *chip;
diff --git a/drivers/net/dsa/mv88e6xxx/hwtstamp.h b/drivers/net/dsa/mv88e6xxx/hwtstamp.h
index cf7fb6d660b1..85acc758e3eb 100644
--- a/drivers/net/dsa/mv88e6xxx/hwtstamp.h
+++ b/drivers/net/dsa/mv88e6xxx/hwtstamp.h
@@ -121,7 +121,7 @@ void mv88e6xxx_port_txtstamp(struct dsa_switch *ds, int port,
struct sk_buff *skb);
int mv88e6xxx_get_ts_info(struct dsa_switch *ds, int port,
- struct ethtool_ts_info *info);
+ struct kernel_ethtool_ts_info *info);
int mv88e6xxx_hwtstamp_setup(struct mv88e6xxx_chip *chip);
void mv88e6xxx_hwtstamp_free(struct mv88e6xxx_chip *chip);
@@ -157,7 +157,7 @@ static inline void mv88e6xxx_port_txtstamp(struct dsa_switch *ds, int port,
}
static inline int mv88e6xxx_get_ts_info(struct dsa_switch *ds, int port,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
return -EOPNOTSUPP;
}
diff --git a/drivers/net/dsa/ocelot/felix.c b/drivers/net/dsa/ocelot/felix.c
index d12c4e85baa7..e554699f06d4 100644
--- a/drivers/net/dsa/ocelot/felix.c
+++ b/drivers/net/dsa/ocelot/felix.c
@@ -1234,7 +1234,7 @@ static int felix_get_sset_count(struct dsa_switch *ds, int port, int sset)
}
static int felix_get_ts_info(struct dsa_switch *ds, int port,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct ocelot *ocelot = ds->priv;
diff --git a/drivers/net/dsa/sja1105/sja1105_ptp.c b/drivers/net/dsa/sja1105/sja1105_ptp.c
index a7d41e781398..a1f4ca6ad888 100644
--- a/drivers/net/dsa/sja1105/sja1105_ptp.c
+++ b/drivers/net/dsa/sja1105/sja1105_ptp.c
@@ -111,7 +111,7 @@ int sja1105_hwtstamp_get(struct dsa_switch *ds, int port, struct ifreq *ifr)
}
int sja1105_get_ts_info(struct dsa_switch *ds, int port,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct sja1105_private *priv = ds->priv;
struct sja1105_ptp_data *ptp_data = &priv->ptp_data;
diff --git a/drivers/net/dsa/sja1105/sja1105_ptp.h b/drivers/net/dsa/sja1105/sja1105_ptp.h
index 416461ee95d2..8add2bd5f728 100644
--- a/drivers/net/dsa/sja1105/sja1105_ptp.h
+++ b/drivers/net/dsa/sja1105/sja1105_ptp.h
@@ -101,7 +101,7 @@ void sja1105pqrs_ptp_cmd_packing(u8 *buf, struct sja1105_ptp_cmd *cmd,
enum packing_op op);
int sja1105_get_ts_info(struct dsa_switch *ds, int port,
- struct ethtool_ts_info *ts);
+ struct kernel_ethtool_ts_info *ts);
void sja1105_ptp_txtstamp_skb(struct dsa_switch *ds, int slot,
struct sk_buff *clone);
diff --git a/drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c b/drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c
index 58e7e88aae5b..21407a26f806 100644
--- a/drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c
+++ b/drivers/net/ethernet/amd/xgbe/xgbe-ethtool.c
@@ -577,7 +577,7 @@ static int xgbe_set_rxfh(struct net_device *netdev,
}
static int xgbe_get_ts_info(struct net_device *netdev,
- struct ethtool_ts_info *ts_info)
+ struct kernel_ethtool_ts_info *ts_info)
{
struct xgbe_prv_data *pdata = netdev_priv(netdev);
diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c b/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c
index a2606ee3b0a5..d0aecd1d7357 100644
--- a/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c
+++ b/drivers/net/ethernet/aquantia/atlantic/aq_ethtool.c
@@ -652,7 +652,7 @@ static int aq_ethtool_set_wol(struct net_device *ndev,
}
static int aq_ethtool_get_ts_info(struct net_device *ndev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct aq_nic_s *aq_nic = netdev_priv(ndev);
diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c
index 58956ed8f531..c7b56a5e5425 100644
--- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_ethtool.c
@@ -3634,7 +3634,7 @@ static int bnx2x_set_channels(struct net_device *dev,
}
static int bnx2x_get_ts_info(struct net_device *dev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct bnx2x *bp = netdev_priv(dev);
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
index bf157f6cc042..fcbf38f79c23 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt_ethtool.c
@@ -5013,7 +5013,7 @@ static int bnxt_get_dump_data(struct net_device *dev, struct ethtool_dump *dump,
}
static int bnxt_get_ts_info(struct net_device *dev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct bnxt *bp = netdev_priv(dev);
struct bnxt_ptp_cfg *ptp;
diff --git a/drivers/net/ethernet/broadcom/tg3.c b/drivers/net/ethernet/broadcom/tg3.c
index 1589a49b876c..0ec5f01551f9 100644
--- a/drivers/net/ethernet/broadcom/tg3.c
+++ b/drivers/net/ethernet/broadcom/tg3.c
@@ -6141,7 +6141,7 @@ static void tg3_refclk_write(struct tg3 *tp, u64 newval)
static inline void tg3_full_lock(struct tg3 *tp, int irq_sync);
static inline void tg3_full_unlock(struct tg3 *tp);
-static int tg3_get_ts_info(struct net_device *dev, struct ethtool_ts_info *info)
+static int tg3_get_ts_info(struct net_device *dev, struct kernel_ethtool_ts_info *info)
{
struct tg3 *tp = netdev_priv(dev);
diff --git a/drivers/net/ethernet/cadence/macb.h b/drivers/net/ethernet/cadence/macb.h
index 122663ff7834..ea71612f6b36 100644
--- a/drivers/net/ethernet/cadence/macb.h
+++ b/drivers/net/ethernet/cadence/macb.h
@@ -1168,7 +1168,7 @@ struct macb_ptp_info {
s32 (*get_ptp_max_adj)(void);
unsigned int (*get_tsu_rate)(struct macb *bp);
int (*get_ts_info)(struct net_device *dev,
- struct ethtool_ts_info *info);
+ struct kernel_ethtool_ts_info *info);
int (*get_hwtst)(struct net_device *netdev,
struct kernel_hwtstamp_config *tstamp_config);
int (*set_hwtst)(struct net_device *netdev,
diff --git a/drivers/net/ethernet/cadence/macb_main.c b/drivers/net/ethernet/cadence/macb_main.c
index cecc3d6e630f..11665be3a22c 100644
--- a/drivers/net/ethernet/cadence/macb_main.c
+++ b/drivers/net/ethernet/cadence/macb_main.c
@@ -3399,7 +3399,7 @@ static s32 gem_get_ptp_max_adj(void)
}
static int gem_get_ts_info(struct net_device *dev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct macb *bp = netdev_priv(dev);
@@ -3440,7 +3440,7 @@ static struct macb_ptp_info gem_ptp_info = {
#endif
static int macb_get_ts_info(struct net_device *netdev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct macb *bp = netdev_priv(netdev);
diff --git a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c
index d3e07b6ed5e1..5835965dbc32 100644
--- a/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c
+++ b/drivers/net/ethernet/cavium/liquidio/lio_ethtool.c
@@ -2497,7 +2497,7 @@ static int lio_set_intr_coalesce(struct net_device *netdev,
}
static int lio_get_ts_info(struct net_device *netdev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct lio *lio = GET_LIO(netdev);
diff --git a/drivers/net/ethernet/cavium/thunder/nicvf_ethtool.c b/drivers/net/ethernet/cavium/thunder/nicvf_ethtool.c
index 34125b8cd935..6a04d2530176 100644
--- a/drivers/net/ethernet/cavium/thunder/nicvf_ethtool.c
+++ b/drivers/net/ethernet/cavium/thunder/nicvf_ethtool.c
@@ -836,7 +836,7 @@ static int nicvf_set_pauseparam(struct net_device *dev,
}
static int nicvf_get_ts_info(struct net_device *netdev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct nicvf *nic = netdev_priv(netdev);
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c
index 47eecde36285..3d091947ae00 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_ethtool.c
@@ -1550,7 +1550,7 @@ static int set_flash(struct net_device *netdev, struct ethtool_flash *ef)
return ret;
}
-static int get_ts_info(struct net_device *dev, struct ethtool_ts_info *ts_info)
+static int get_ts_info(struct net_device *dev, struct kernel_ethtool_ts_info *ts_info)
{
struct port_info *pi = netdev_priv(dev);
struct adapter *adapter = pi->adapter;
diff --git a/drivers/net/ethernet/cisco/enic/enic_ethtool.c b/drivers/net/ethernet/cisco/enic/enic_ethtool.c
index a42f3f280f3e..f2f1055880b2 100644
--- a/drivers/net/ethernet/cisco/enic/enic_ethtool.c
+++ b/drivers/net/ethernet/cisco/enic/enic_ethtool.c
@@ -599,7 +599,7 @@ static int enic_set_rxfh(struct net_device *netdev,
}
static int enic_get_ts_info(struct net_device *netdev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
info->so_timestamping = SOF_TIMESTAMPING_TX_SOFTWARE |
SOF_TIMESTAMPING_RX_SOFTWARE |
diff --git a/drivers/net/ethernet/engleder/tsnep_ethtool.c b/drivers/net/ethernet/engleder/tsnep_ethtool.c
index 65ec1abc9442..9aa286ba1f00 100644
--- a/drivers/net/ethernet/engleder/tsnep_ethtool.c
+++ b/drivers/net/ethernet/engleder/tsnep_ethtool.c
@@ -305,7 +305,7 @@ static void tsnep_ethtool_get_channels(struct net_device *netdev,
}
static int tsnep_ethtool_get_ts_info(struct net_device *netdev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct tsnep_adapter *adapter = netdev_priv(netdev);
diff --git a/drivers/net/ethernet/freescale/dpaa/dpaa_ethtool.c b/drivers/net/ethernet/freescale/dpaa/dpaa_ethtool.c
index 5bd0b36d1feb..0e923d805732 100644
--- a/drivers/net/ethernet/freescale/dpaa/dpaa_ethtool.c
+++ b/drivers/net/ethernet/freescale/dpaa/dpaa_ethtool.c
@@ -394,7 +394,7 @@ static int dpaa_set_rxnfc(struct net_device *dev, struct ethtool_rxnfc *cmd)
}
static int dpaa_get_ts_info(struct net_device *net_dev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct device *dev = net_dev->dev.parent;
struct device_node *mac_node = dev->of_node;
diff --git a/drivers/net/ethernet/freescale/dpaa2/dpaa2-ethtool.c b/drivers/net/ethernet/freescale/dpaa2/dpaa2-ethtool.c
index e80e9388c71f..7f476519b7ad 100644
--- a/drivers/net/ethernet/freescale/dpaa2/dpaa2-ethtool.c
+++ b/drivers/net/ethernet/freescale/dpaa2/dpaa2-ethtool.c
@@ -794,7 +794,7 @@ int dpaa2_phc_index = -1;
EXPORT_SYMBOL(dpaa2_phc_index);
static int dpaa2_eth_get_ts_info(struct net_device *dev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
if (!dpaa2_ptp)
return ethtool_op_get_ts_info(dev, info);
diff --git a/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c b/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c
index f7753ea5b57e..5e684b23c5f5 100644
--- a/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c
+++ b/drivers/net/ethernet/freescale/enetc/enetc_ethtool.c
@@ -841,7 +841,7 @@ static int enetc_set_coalesce(struct net_device *ndev,
}
static int enetc_get_ts_info(struct net_device *ndev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
int *phc_idx;
diff --git a/drivers/net/ethernet/freescale/fec_main.c b/drivers/net/ethernet/freescale/fec_main.c
index fb19295529a2..a923cb95cdc6 100644
--- a/drivers/net/ethernet/freescale/fec_main.c
+++ b/drivers/net/ethernet/freescale/fec_main.c
@@ -2768,7 +2768,7 @@ static void fec_enet_get_regs(struct net_device *ndev,
}
static int fec_enet_get_ts_info(struct net_device *ndev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct fec_enet_private *fep = netdev_priv(ndev);
diff --git a/drivers/net/ethernet/freescale/gianfar_ethtool.c b/drivers/net/ethernet/freescale/gianfar_ethtool.c
index 7a15b9245698..f581402ad740 100644
--- a/drivers/net/ethernet/freescale/gianfar_ethtool.c
+++ b/drivers/net/ethernet/freescale/gianfar_ethtool.c
@@ -1448,7 +1448,7 @@ static int gfar_get_nfc(struct net_device *dev, struct ethtool_rxnfc *cmd,
}
static int gfar_get_ts_info(struct net_device *dev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct gfar_private *priv = netdev_priv(dev);
struct platform_device *ptp_dev;
diff --git a/drivers/net/ethernet/fungible/funeth/funeth_ethtool.c b/drivers/net/ethernet/fungible/funeth/funeth_ethtool.c
index 4edd0adfc6c7..7f081e6e8c87 100644
--- a/drivers/net/ethernet/fungible/funeth/funeth_ethtool.c
+++ b/drivers/net/ethernet/fungible/funeth/funeth_ethtool.c
@@ -1040,7 +1040,7 @@ static int fun_set_rxfh(struct net_device *netdev,
}
static int fun_get_ts_info(struct net_device *netdev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
info->so_timestamping = SOF_TIMESTAMPING_RX_SOFTWARE |
SOF_TIMESTAMPING_RX_HARDWARE |
diff --git a/drivers/net/ethernet/hisilicon/hns3/hnae3.h b/drivers/net/ethernet/hisilicon/hns3/hnae3.h
index 7cebb08bd320..27dbe367f3d3 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hnae3.h
+++ b/drivers/net/ethernet/hisilicon/hns3/hnae3.h
@@ -786,7 +786,7 @@ struct hnae3_ae_ops {
void (*get_rx_hwts)(struct hnae3_handle *handle, struct sk_buff *skb,
u32 nsec, u32 sec);
int (*get_ts_info)(struct hnae3_handle *handle,
- struct ethtool_ts_info *info);
+ struct kernel_ethtool_ts_info *info);
int (*get_link_diagnosis_info)(struct hnae3_handle *handle,
u32 *status_code);
void (*clean_vf_config)(struct hnae3_ae_dev *ae_dev, int num_vfs);
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c b/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c
index 941cb529d671..b1e988347347 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3_ethtool.c
@@ -2009,7 +2009,7 @@ static int hns3_set_tunable(struct net_device *netdev,
ETHTOOL_RING_USE_TX_PUSH)
static int hns3_get_ts_info(struct net_device *netdev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct hnae3_handle *handle = hns3_get_handle(netdev);
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.c b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.c
index 507d7ce26d83..5fff8ed388f8 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.c
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.c
@@ -378,7 +378,7 @@ int hclge_ptp_set_cfg(struct hclge_dev *hdev, struct ifreq *ifr)
}
int hclge_ptp_get_ts_info(struct hnae3_handle *handle,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct hclge_vport *vport = hclge_get_vport(handle);
struct hclge_dev *hdev = vport->back;
diff --git a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.h b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.h
index bbee74cd8404..63483636c074 100644
--- a/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.h
+++ b/drivers/net/ethernet/hisilicon/hns3/hns3pf/hclge_ptp.h
@@ -138,6 +138,6 @@ int hclge_ptp_set_cfg(struct hclge_dev *hdev, struct ifreq *ifr);
int hclge_ptp_init(struct hclge_dev *hdev);
void hclge_ptp_uninit(struct hclge_dev *hdev);
int hclge_ptp_get_ts_info(struct hnae3_handle *handle,
- struct ethtool_ts_info *info);
+ struct kernel_ethtool_ts_info *info);
int hclge_ptp_cfg_qry(struct hclge_dev *hdev, u32 *cfg);
#endif
diff --git a/drivers/net/ethernet/intel/e1000e/ethtool.c b/drivers/net/ethernet/intel/e1000e/ethtool.c
index 85da20778e0f..9364bc2b4eb1 100644
--- a/drivers/net/ethernet/intel/e1000e/ethtool.c
+++ b/drivers/net/ethernet/intel/e1000e/ethtool.c
@@ -2263,7 +2263,7 @@ static int e1000e_set_eee(struct net_device *netdev, struct ethtool_keee *edata)
}
static int e1000e_get_ts_info(struct net_device *netdev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct e1000_adapter *adapter = netdev_priv(netdev);
diff --git a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
index 4e28785c9fb2..1d0d2e526adb 100644
--- a/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
+++ b/drivers/net/ethernet/intel/i40e/i40e_ethtool.c
@@ -2546,7 +2546,7 @@ static void i40e_get_strings(struct net_device *netdev, u32 stringset,
}
static int i40e_get_ts_info(struct net_device *dev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct i40e_pf *pf = i40e_netdev_to_pf(dev);
diff --git a/drivers/net/ethernet/intel/ice/ice_ethtool.c b/drivers/net/ethernet/intel/ice/ice_ethtool.c
index 62c8205fceba..b8f142f04330 100644
--- a/drivers/net/ethernet/intel/ice/ice_ethtool.c
+++ b/drivers/net/ethernet/intel/ice/ice_ethtool.c
@@ -3434,7 +3434,7 @@ ice_set_rxfh(struct net_device *netdev, struct ethtool_rxfh_param *rxfh,
}
static int
-ice_get_ts_info(struct net_device *dev, struct ethtool_ts_info *info)
+ice_get_ts_info(struct net_device *dev, struct kernel_ethtool_ts_info *info)
{
struct ice_pf *pf = ice_netdev_to_pf(dev);
diff --git a/drivers/net/ethernet/intel/igb/igb_ethtool.c b/drivers/net/ethernet/intel/igb/igb_ethtool.c
index 61d72250c0ed..06b9970dffad 100644
--- a/drivers/net/ethernet/intel/igb/igb_ethtool.c
+++ b/drivers/net/ethernet/intel/igb/igb_ethtool.c
@@ -2381,7 +2381,7 @@ static void igb_get_strings(struct net_device *netdev, u32 stringset, u8 *data)
}
static int igb_get_ts_info(struct net_device *dev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct igb_adapter *adapter = netdev_priv(dev);
diff --git a/drivers/net/ethernet/intel/igc/igc_ethtool.c b/drivers/net/ethernet/intel/igc/igc_ethtool.c
index 0cd2bd695db1..9c3adb18f922 100644
--- a/drivers/net/ethernet/intel/igc/igc_ethtool.c
+++ b/drivers/net/ethernet/intel/igc/igc_ethtool.c
@@ -1559,7 +1559,7 @@ static int igc_ethtool_set_channels(struct net_device *netdev,
}
static int igc_ethtool_get_ts_info(struct net_device *dev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct igc_adapter *adapter = netdev_priv(dev);
diff --git a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c
index 6e6e6f1847b6..4cac76254966 100644
--- a/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c
+++ b/drivers/net/ethernet/intel/ixgbe/ixgbe_ethtool.c
@@ -3170,7 +3170,7 @@ static int ixgbe_set_rxfh(struct net_device *netdev,
}
static int ixgbe_get_ts_info(struct net_device *dev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct ixgbe_adapter *adapter = netdev_priv(dev);
diff --git a/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c b/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c
index 9adf4301c9b1..8c45ad983abc 100644
--- a/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c
+++ b/drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c
@@ -5259,7 +5259,7 @@ static int mvpp2_get_ts_config(struct mvpp2_port *port, struct ifreq *ifr)
}
static int mvpp2_ethtool_get_ts_info(struct net_device *dev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct mvpp2_port *port = netdev_priv(dev);
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ethtool.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ethtool.c
index 7f786de61014..0db62eb0dab3 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ethtool.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ethtool.c
@@ -954,7 +954,7 @@ static u32 otx2_get_link(struct net_device *netdev)
}
static int otx2_get_ts_info(struct net_device *netdev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct otx2_nic *pfvf = netdev_priv(netdev);
diff --git a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
index 0606f18e5bbe..943d6918c2ec 100644
--- a/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx4/en_ethtool.c
@@ -1884,7 +1884,7 @@ static int mlx4_en_set_channels(struct net_device *dev,
}
static int mlx4_en_get_ts_info(struct net_device *dev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct mlx4_en_priv *priv = netdev_priv(dev);
struct mlx4_en_dev *mdev = priv->mdev;
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en.h b/drivers/net/ethernet/mellanox/mlx5/core/en.h
index 6a343a8f162f..5fd82c67b6ab 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en.h
@@ -1191,7 +1191,7 @@ int mlx5e_set_per_queue_coalesce(struct net_device *dev, u32 queue,
u32 mlx5e_ethtool_get_rxfh_key_size(struct mlx5e_priv *priv);
u32 mlx5e_ethtool_get_rxfh_indir_size(struct mlx5e_priv *priv);
int mlx5e_ethtool_get_ts_info(struct mlx5e_priv *priv,
- struct ethtool_ts_info *info);
+ struct kernel_ethtool_ts_info *info);
int mlx5e_ethtool_flash_device(struct mlx5e_priv *priv,
struct ethtool_flash *flash);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
index 3320f12ba2db..cbb3945529d7 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c
@@ -1658,7 +1658,7 @@ static int mlx5e_set_pauseparam(struct net_device *netdev,
}
int mlx5e_ethtool_get_ts_info(struct mlx5e_priv *priv,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct mlx5_core_dev *mdev = priv->mdev;
@@ -1682,7 +1682,7 @@ int mlx5e_ethtool_get_ts_info(struct mlx5e_priv *priv,
}
static int mlx5e_get_ts_info(struct net_device *dev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct mlx5e_priv *priv = netdev_priv(dev);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ethtool.c b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ethtool.c
index 779d92b762d3..905bdbaffb9a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/ipoib/ethtool.c
@@ -136,7 +136,7 @@ static int mlx5i_get_coalesce(struct net_device *netdev,
}
static int mlx5i_get_ts_info(struct net_device *netdev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct mlx5e_priv *priv = mlx5i_epriv(netdev);
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
index bb0586b45c8d..8d3c61287696 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum.h
@@ -238,7 +238,7 @@ struct mlxsw_sp_ptp_ops {
struct hwtstamp_config *config);
void (*shaper_work)(struct work_struct *work);
int (*get_ts_info)(struct mlxsw_sp *mlxsw_sp,
- struct ethtool_ts_info *info);
+ struct kernel_ethtool_ts_info *info);
int (*get_stats_count)(void);
void (*get_stats_strings)(u8 **p);
void (*get_stats)(struct mlxsw_sp_port *mlxsw_sp_port,
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ethtool.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ethtool.c
index c79da1411d33..2bed8c86b7cf 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ethtool.c
@@ -1082,7 +1082,7 @@ mlxsw_sp_set_module_eeprom_by_page(struct net_device *dev,
}
static int
-mlxsw_sp_get_ts_info(struct net_device *netdev, struct ethtool_ts_info *info)
+mlxsw_sp_get_ts_info(struct net_device *netdev, struct kernel_ethtool_ts_info *info)
{
struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(netdev);
struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.c
index cbb6c75a6620..5b174cb95eb8 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.c
@@ -1276,7 +1276,7 @@ int mlxsw_sp1_ptp_hwtstamp_set(struct mlxsw_sp_port *mlxsw_sp_port,
}
int mlxsw_sp1_ptp_get_ts_info(struct mlxsw_sp *mlxsw_sp,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
info->phc_index = ptp_clock_index(mlxsw_sp->clock->ptp);
@@ -1661,7 +1661,7 @@ int mlxsw_sp2_ptp_hwtstamp_set(struct mlxsw_sp_port *mlxsw_sp_port,
}
int mlxsw_sp2_ptp_get_ts_info(struct mlxsw_sp *mlxsw_sp,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
info->phc_index = ptp_clock_index(mlxsw_sp->clock->ptp);
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.h b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.h
index a8b88230959a..769095d4932d 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ptp.h
@@ -11,7 +11,7 @@ struct mlxsw_sp;
struct mlxsw_sp_port;
struct mlxsw_sp_ptp_clock;
-static inline int mlxsw_sp_ptp_get_ts_info_noptp(struct ethtool_ts_info *info)
+static inline int mlxsw_sp_ptp_get_ts_info_noptp(struct kernel_ethtool_ts_info *info)
{
info->so_timestamping = SOF_TIMESTAMPING_RX_SOFTWARE |
SOF_TIMESTAMPING_SOFTWARE;
@@ -50,7 +50,7 @@ int mlxsw_sp1_ptp_hwtstamp_set(struct mlxsw_sp_port *mlxsw_sp_port,
void mlxsw_sp1_ptp_shaper_work(struct work_struct *work);
int mlxsw_sp1_ptp_get_ts_info(struct mlxsw_sp *mlxsw_sp,
- struct ethtool_ts_info *info);
+ struct kernel_ethtool_ts_info *info);
int mlxsw_sp1_get_stats_count(void);
void mlxsw_sp1_get_stats_strings(u8 **p);
@@ -84,7 +84,7 @@ int mlxsw_sp2_ptp_hwtstamp_set(struct mlxsw_sp_port *mlxsw_sp_port,
struct hwtstamp_config *config);
int mlxsw_sp2_ptp_get_ts_info(struct mlxsw_sp *mlxsw_sp,
- struct ethtool_ts_info *info);
+ struct kernel_ethtool_ts_info *info);
int mlxsw_sp2_ptp_txhdr_construct(struct mlxsw_core *mlxsw_core,
struct mlxsw_sp_port *mlxsw_sp_port,
@@ -152,7 +152,7 @@ static inline void mlxsw_sp1_ptp_shaper_work(struct work_struct *work)
}
static inline int mlxsw_sp1_ptp_get_ts_info(struct mlxsw_sp *mlxsw_sp,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
return mlxsw_sp_ptp_get_ts_info_noptp(info);
}
@@ -227,7 +227,7 @@ mlxsw_sp2_ptp_hwtstamp_set(struct mlxsw_sp_port *mlxsw_sp_port,
}
static inline int mlxsw_sp2_ptp_get_ts_info(struct mlxsw_sp *mlxsw_sp,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
return mlxsw_sp_ptp_get_ts_info_noptp(info);
}
diff --git a/drivers/net/ethernet/microchip/lan743x_ethtool.c b/drivers/net/ethernet/microchip/lan743x_ethtool.c
index 0d1740d64676..3a63ec091413 100644
--- a/drivers/net/ethernet/microchip/lan743x_ethtool.c
+++ b/drivers/net/ethernet/microchip/lan743x_ethtool.c
@@ -1029,7 +1029,7 @@ static int lan743x_ethtool_set_rxfh(struct net_device *netdev,
}
static int lan743x_ethtool_get_ts_info(struct net_device *netdev,
- struct ethtool_ts_info *ts_info)
+ struct kernel_ethtool_ts_info *ts_info)
{
struct lan743x_adapter *adapter = netdev_priv(netdev);
diff --git a/drivers/net/ethernet/microchip/lan966x/lan966x_ethtool.c b/drivers/net/ethernet/microchip/lan966x/lan966x_ethtool.c
index c0fc85ac5db3..aec7066d83b3 100644
--- a/drivers/net/ethernet/microchip/lan966x/lan966x_ethtool.c
+++ b/drivers/net/ethernet/microchip/lan966x/lan966x_ethtool.c
@@ -538,7 +538,7 @@ static int lan966x_set_pauseparam(struct net_device *dev,
}
static int lan966x_get_ts_info(struct net_device *dev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct lan966x_port *port = netdev_priv(dev);
struct lan966x *lan966x = port->lan966x;
diff --git a/drivers/net/ethernet/microchip/sparx5/sparx5_ethtool.c b/drivers/net/ethernet/microchip/sparx5/sparx5_ethtool.c
index a06dc5a9b355..4f800c1a435d 100644
--- a/drivers/net/ethernet/microchip/sparx5/sparx5_ethtool.c
+++ b/drivers/net/ethernet/microchip/sparx5/sparx5_ethtool.c
@@ -1183,7 +1183,7 @@ static void sparx5_config_port_stats(struct sparx5 *sparx5, int portno)
}
static int sparx5_get_ts_info(struct net_device *dev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct sparx5_port *port = netdev_priv(dev);
struct sparx5 *sparx5 = port->sparx5;
diff --git a/drivers/net/ethernet/mscc/ocelot_net.c b/drivers/net/ethernet/mscc/ocelot_net.c
index 21a87a3fc556..7c9540a71725 100644
--- a/drivers/net/ethernet/mscc/ocelot_net.c
+++ b/drivers/net/ethernet/mscc/ocelot_net.c
@@ -980,7 +980,7 @@ static int ocelot_port_get_sset_count(struct net_device *dev, int sset)
}
static int ocelot_port_get_ts_info(struct net_device *dev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct ocelot_port_private *priv = netdev_priv(dev);
struct ocelot *ocelot = priv->port.ocelot;
diff --git a/drivers/net/ethernet/mscc/ocelot_ptp.c b/drivers/net/ethernet/mscc/ocelot_ptp.c
index cb32234a5bf1..b3c28260adf8 100644
--- a/drivers/net/ethernet/mscc/ocelot_ptp.c
+++ b/drivers/net/ethernet/mscc/ocelot_ptp.c
@@ -580,7 +580,7 @@ int ocelot_hwstamp_set(struct ocelot *ocelot, int port, struct ifreq *ifr)
EXPORT_SYMBOL(ocelot_hwstamp_set);
int ocelot_get_ts_info(struct ocelot *ocelot, int port,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
info->phc_index = ocelot->ptp_clock ?
ptp_clock_index(ocelot->ptp_clock) : -1;
diff --git a/drivers/net/ethernet/pensando/ionic/ionic_ethtool.c b/drivers/net/ethernet/pensando/ionic/ionic_ethtool.c
index 185a03514ae3..4619fd74f3e3 100644
--- a/drivers/net/ethernet/pensando/ionic/ionic_ethtool.c
+++ b/drivers/net/ethernet/pensando/ionic/ionic_ethtool.c
@@ -977,7 +977,7 @@ static int ionic_get_module_eeprom(struct net_device *netdev,
}
static int ionic_get_ts_info(struct net_device *netdev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct ionic_lif *lif = netdev_priv(netdev);
struct ionic *ionic = lif->ionic;
diff --git a/drivers/net/ethernet/qlogic/qede/qede_ethtool.c b/drivers/net/ethernet/qlogic/qede/qede_ethtool.c
index f497f6ca1018..97b059be1041 100644
--- a/drivers/net/ethernet/qlogic/qede/qede_ethtool.c
+++ b/drivers/net/ethernet/qlogic/qede/qede_ethtool.c
@@ -1137,7 +1137,7 @@ static int qede_set_channels(struct net_device *dev,
}
static int qede_get_ts_info(struct net_device *dev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct qede_dev *edev = netdev_priv(dev);
diff --git a/drivers/net/ethernet/qlogic/qede/qede_ptp.c b/drivers/net/ethernet/qlogic/qede/qede_ptp.c
index 747cc5e2bb78..63e3dac4d5f7 100644
--- a/drivers/net/ethernet/qlogic/qede/qede_ptp.c
+++ b/drivers/net/ethernet/qlogic/qede/qede_ptp.c
@@ -321,7 +321,7 @@ int qede_ptp_hw_ts(struct qede_dev *edev, struct ifreq *ifr)
sizeof(config)) ? -EFAULT : 0;
}
-int qede_ptp_get_ts_info(struct qede_dev *edev, struct ethtool_ts_info *info)
+int qede_ptp_get_ts_info(struct qede_dev *edev, struct kernel_ethtool_ts_info *info)
{
struct qede_ptp *ptp = edev->ptp;
diff --git a/drivers/net/ethernet/qlogic/qede/qede_ptp.h b/drivers/net/ethernet/qlogic/qede/qede_ptp.h
index 1db0f021c645..adafc894797e 100644
--- a/drivers/net/ethernet/qlogic/qede/qede_ptp.h
+++ b/drivers/net/ethernet/qlogic/qede/qede_ptp.h
@@ -17,7 +17,7 @@ void qede_ptp_tx_ts(struct qede_dev *edev, struct sk_buff *skb);
int qede_ptp_hw_ts(struct qede_dev *edev, struct ifreq *req);
void qede_ptp_disable(struct qede_dev *edev);
int qede_ptp_enable(struct qede_dev *edev);
-int qede_ptp_get_ts_info(struct qede_dev *edev, struct ethtool_ts_info *ts);
+int qede_ptp_get_ts_info(struct qede_dev *edev, struct kernel_ethtool_ts_info *ts);
static inline void qede_ptp_record_rx_ts(struct qede_dev *edev,
union eth_rx_cqe *cqe,
diff --git a/drivers/net/ethernet/renesas/ravb_main.c b/drivers/net/ethernet/renesas/ravb_main.c
index 6605e4f4af53..c02fb296bf7d 100644
--- a/drivers/net/ethernet/renesas/ravb_main.c
+++ b/drivers/net/ethernet/renesas/ravb_main.c
@@ -1737,7 +1737,7 @@ static int ravb_set_ringparam(struct net_device *ndev,
}
static int ravb_get_ts_info(struct net_device *ndev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct ravb_private *priv = netdev_priv(ndev);
const struct ravb_hw_info *hw_info = priv->info;
diff --git a/drivers/net/ethernet/renesas/rswitch.c b/drivers/net/ethernet/renesas/rswitch.c
index 24c90d8f5a44..ff50e20856ec 100644
--- a/drivers/net/ethernet/renesas/rswitch.c
+++ b/drivers/net/ethernet/renesas/rswitch.c
@@ -1809,7 +1809,7 @@ static const struct net_device_ops rswitch_netdev_ops = {
.ndo_set_mac_address = eth_mac_addr,
};
-static int rswitch_get_ts_info(struct net_device *ndev, struct ethtool_ts_info *info)
+static int rswitch_get_ts_info(struct net_device *ndev, struct kernel_ethtool_ts_info *info)
{
struct rswitch_device *rdev = netdev_priv(ndev);
diff --git a/drivers/net/ethernet/renesas/rtsn.c b/drivers/net/ethernet/renesas/rtsn.c
index ad69d47463cb..577227c007ab 100644
--- a/drivers/net/ethernet/renesas/rtsn.c
+++ b/drivers/net/ethernet/renesas/rtsn.c
@@ -1213,7 +1213,7 @@ static const struct net_device_ops rtsn_netdev_ops = {
};
static int rtsn_get_ts_info(struct net_device *ndev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct rtsn_private *priv = netdev_priv(ndev);
diff --git a/drivers/net/ethernet/sfc/ethtool.c b/drivers/net/ethernet/sfc/ethtool.c
index 0f5c68b8bab7..7c887160e2ef 100644
--- a/drivers/net/ethernet/sfc/ethtool.c
+++ b/drivers/net/ethernet/sfc/ethtool.c
@@ -226,7 +226,7 @@ static void efx_ethtool_get_fec_stats(struct net_device *net_dev,
}
static int efx_ethtool_get_ts_info(struct net_device *net_dev,
- struct ethtool_ts_info *ts_info)
+ struct kernel_ethtool_ts_info *ts_info)
{
struct efx_nic *efx = efx_netdev_priv(net_dev);
diff --git a/drivers/net/ethernet/sfc/falcon/nic.h b/drivers/net/ethernet/sfc/falcon/nic.h
index 9f413474bd9f..ada6e036fd97 100644
--- a/drivers/net/ethernet/sfc/falcon/nic.h
+++ b/drivers/net/ethernet/sfc/falcon/nic.h
@@ -297,7 +297,7 @@ static inline struct falcon_board *falcon_board(struct ef4_nic *efx)
return &data->board;
}
-struct ethtool_ts_info;
+struct kernel_ethtool_ts_info;
extern const struct ef4_nic_type falcon_a1_nic_type;
extern const struct ef4_nic_type falcon_b0_nic_type;
diff --git a/drivers/net/ethernet/sfc/ptp.c b/drivers/net/ethernet/sfc/ptp.c
index c3bffbf0ba2b..6fd2fdbaa418 100644
--- a/drivers/net/ethernet/sfc/ptp.c
+++ b/drivers/net/ethernet/sfc/ptp.c
@@ -1864,7 +1864,7 @@ static int efx_ptp_ts_init(struct efx_nic *efx, struct kernel_hwtstamp_config *i
return 0;
}
-void efx_ptp_get_ts_info(struct efx_nic *efx, struct ethtool_ts_info *ts_info)
+void efx_ptp_get_ts_info(struct efx_nic *efx, struct kernel_ethtool_ts_info *ts_info)
{
struct efx_ptp_data *ptp = efx->ptp_data;
struct efx_nic *primary = efx->primary;
diff --git a/drivers/net/ethernet/sfc/ptp.h b/drivers/net/ethernet/sfc/ptp.h
index 2f30dbb490d2..6946203499ef 100644
--- a/drivers/net/ethernet/sfc/ptp.h
+++ b/drivers/net/ethernet/sfc/ptp.h
@@ -12,7 +12,7 @@
#include <linux/net_tstamp.h>
#include "net_driver.h"
-struct ethtool_ts_info;
+struct kernel_ethtool_ts_info;
int efx_ptp_probe(struct efx_nic *efx, struct efx_channel *channel);
void efx_ptp_defer_probe_with_channel(struct efx_nic *efx);
struct efx_channel *efx_ptp_channel(struct efx_nic *efx);
@@ -23,7 +23,8 @@ int efx_ptp_set_ts_config(struct efx_nic *efx,
struct netlink_ext_ack *extack);
int efx_ptp_get_ts_config(struct efx_nic *efx,
struct kernel_hwtstamp_config *config);
-void efx_ptp_get_ts_info(struct efx_nic *efx, struct ethtool_ts_info *ts_info);
+void efx_ptp_get_ts_info(struct efx_nic *efx,
+ struct kernel_ethtool_ts_info *ts_info);
bool efx_ptp_is_ptp_tx(struct efx_nic *efx, struct sk_buff *skb);
int efx_ptp_get_mode(struct efx_nic *efx);
int efx_ptp_change_mode(struct efx_nic *efx, bool enable_wanted,
diff --git a/drivers/net/ethernet/sfc/siena/ethtool.c b/drivers/net/ethernet/sfc/siena/ethtool.c
index 14dd3893bdef..4c182d4edfc2 100644
--- a/drivers/net/ethernet/sfc/siena/ethtool.c
+++ b/drivers/net/ethernet/sfc/siena/ethtool.c
@@ -226,7 +226,7 @@ static void efx_ethtool_get_fec_stats(struct net_device *net_dev,
}
static int efx_ethtool_get_ts_info(struct net_device *net_dev,
- struct ethtool_ts_info *ts_info)
+ struct kernel_ethtool_ts_info *ts_info)
{
struct efx_nic *efx = netdev_priv(net_dev);
diff --git a/drivers/net/ethernet/sfc/siena/ptp.c b/drivers/net/ethernet/sfc/siena/ptp.c
index 4b5e2f0ba350..c473a4b6dd44 100644
--- a/drivers/net/ethernet/sfc/siena/ptp.c
+++ b/drivers/net/ethernet/sfc/siena/ptp.c
@@ -1780,7 +1780,7 @@ static int efx_ptp_ts_init(struct efx_nic *efx,
}
void efx_siena_ptp_get_ts_info(struct efx_nic *efx,
- struct ethtool_ts_info *ts_info)
+ struct kernel_ethtool_ts_info *ts_info)
{
struct efx_ptp_data *ptp = efx->ptp_data;
struct efx_nic *primary = efx->primary;
diff --git a/drivers/net/ethernet/sfc/siena/ptp.h b/drivers/net/ethernet/sfc/siena/ptp.h
index 6352f84424f6..b6133e7c5608 100644
--- a/drivers/net/ethernet/sfc/siena/ptp.h
+++ b/drivers/net/ethernet/sfc/siena/ptp.h
@@ -12,7 +12,7 @@
#include <linux/net_tstamp.h>
#include "net_driver.h"
-struct ethtool_ts_info;
+struct kernel_ethtool_ts_info;
void efx_siena_ptp_defer_probe_with_channel(struct efx_nic *efx);
struct efx_channel *efx_siena_ptp_channel(struct efx_nic *efx);
int efx_siena_ptp_set_ts_config(struct efx_nic *efx,
@@ -21,7 +21,7 @@ int efx_siena_ptp_set_ts_config(struct efx_nic *efx,
int efx_siena_ptp_get_ts_config(struct efx_nic *efx,
struct kernel_hwtstamp_config *config);
void efx_siena_ptp_get_ts_info(struct efx_nic *efx,
- struct ethtool_ts_info *ts_info);
+ struct kernel_ethtool_ts_info *ts_info);
bool efx_siena_ptp_is_ptp_tx(struct efx_nic *efx, struct sk_buff *skb);
int efx_siena_ptp_get_mode(struct efx_nic *efx);
int efx_siena_ptp_change_mode(struct efx_nic *efx, bool enable_wanted,
diff --git a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c
index 18468c0228f0..7008219fd88d 100644
--- a/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c
+++ b/drivers/net/ethernet/stmicro/stmmac/stmmac_ethtool.c
@@ -1199,7 +1199,7 @@ static int stmmac_set_channels(struct net_device *dev,
}
static int stmmac_get_ts_info(struct net_device *dev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct stmmac_priv *priv = netdev_priv(dev);
diff --git a/drivers/net/ethernet/ti/am65-cpsw-ethtool.c b/drivers/net/ethernet/ti/am65-cpsw-ethtool.c
index a1d0935d1ebe..b60976947da5 100644
--- a/drivers/net/ethernet/ti/am65-cpsw-ethtool.c
+++ b/drivers/net/ethernet/ti/am65-cpsw-ethtool.c
@@ -692,7 +692,7 @@ static void am65_cpsw_get_eth_mac_stats(struct net_device *ndev,
};
static int am65_cpsw_get_ethtool_ts_info(struct net_device *ndev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct am65_cpsw_common *common = am65_ndev_to_common(ndev);
unsigned int ptp_v2_filter;
diff --git a/drivers/net/ethernet/ti/cpsw_ethtool.c b/drivers/net/ethernet/ti/cpsw_ethtool.c
index f7b283353ba2..53ed23d68722 100644
--- a/drivers/net/ethernet/ti/cpsw_ethtool.c
+++ b/drivers/net/ethernet/ti/cpsw_ethtool.c
@@ -717,7 +717,7 @@ int cpsw_set_ringparam(struct net_device *ndev,
}
#if IS_ENABLED(CONFIG_TI_CPTS)
-int cpsw_get_ts_info(struct net_device *ndev, struct ethtool_ts_info *info)
+int cpsw_get_ts_info(struct net_device *ndev, struct kernel_ethtool_ts_info *info)
{
struct cpsw_common *cpsw = ndev_to_cpsw(ndev);
@@ -738,7 +738,7 @@ int cpsw_get_ts_info(struct net_device *ndev, struct ethtool_ts_info *info)
return 0;
}
#else
-int cpsw_get_ts_info(struct net_device *ndev, struct ethtool_ts_info *info)
+int cpsw_get_ts_info(struct net_device *ndev, struct kernel_ethtool_ts_info *info)
{
info->so_timestamping =
SOF_TIMESTAMPING_TX_SOFTWARE |
diff --git a/drivers/net/ethernet/ti/cpsw_priv.h b/drivers/net/ethernet/ti/cpsw_priv.h
index 7efa72502c86..1f448290b9f4 100644
--- a/drivers/net/ethernet/ti/cpsw_priv.h
+++ b/drivers/net/ethernet/ti/cpsw_priv.h
@@ -510,6 +510,6 @@ int cpsw_set_ringparam(struct net_device *ndev,
int cpsw_set_channels_common(struct net_device *ndev,
struct ethtool_channels *chs,
cpdma_handler_fn rx_handler);
-int cpsw_get_ts_info(struct net_device *ndev, struct ethtool_ts_info *info);
+int cpsw_get_ts_info(struct net_device *ndev, struct kernel_ethtool_ts_info *info);
#endif /* DRIVERS_NET_ETHERNET_TI_CPSW_PRIV_H_ */
diff --git a/drivers/net/ethernet/ti/icssg/icssg_ethtool.c b/drivers/net/ethernet/ti/icssg/icssg_ethtool.c
index c8d0f45cc5b1..524723ced9ed 100644
--- a/drivers/net/ethernet/ti/icssg/icssg_ethtool.c
+++ b/drivers/net/ethernet/ti/icssg/icssg_ethtool.c
@@ -110,7 +110,7 @@ static void emac_get_ethtool_stats(struct net_device *ndev,
}
static int emac_get_ts_info(struct net_device *ndev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct prueth_emac *emac = netdev_priv(ndev);
diff --git a/drivers/net/ethernet/ti/netcp_ethss.c b/drivers/net/ethernet/ti/netcp_ethss.c
index 02cb6474f6dc..d286709ca3b9 100644
--- a/drivers/net/ethernet/ti/netcp_ethss.c
+++ b/drivers/net/ethernet/ti/netcp_ethss.c
@@ -1999,7 +1999,7 @@ static int keystone_set_link_ksettings(struct net_device *ndev,
#if IS_ENABLED(CONFIG_TI_CPTS)
static int keystone_get_ts_info(struct net_device *ndev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct netcp_intf *netcp = netdev_priv(ndev);
struct gbe_intf *gbe_intf;
@@ -2027,7 +2027,7 @@ static int keystone_get_ts_info(struct net_device *ndev,
}
#else
static int keystone_get_ts_info(struct net_device *ndev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
info->so_timestamping =
SOF_TIMESTAMPING_TX_SOFTWARE |
diff --git a/drivers/net/ethernet/xscale/ixp4xx_eth.c b/drivers/net/ethernet/xscale/ixp4xx_eth.c
index 8aff6a73ca0a..56df37f8d50a 100644
--- a/drivers/net/ethernet/xscale/ixp4xx_eth.c
+++ b/drivers/net/ethernet/xscale/ixp4xx_eth.c
@@ -1015,7 +1015,7 @@ static void ixp4xx_get_drvinfo(struct net_device *dev,
}
static int ixp4xx_get_ts_info(struct net_device *dev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct port *port = netdev_priv(dev);
diff --git a/drivers/net/macvlan.c b/drivers/net/macvlan.c
index 67b7ef2d463f..24298a33e0e9 100644
--- a/drivers/net/macvlan.c
+++ b/drivers/net/macvlan.c
@@ -1084,7 +1084,7 @@ static int macvlan_ethtool_get_link_ksettings(struct net_device *dev,
}
static int macvlan_ethtool_get_ts_info(struct net_device *dev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct net_device *real_dev = macvlan_dev_real_dev(dev);
diff --git a/drivers/net/netdevsim/ethtool.c b/drivers/net/netdevsim/ethtool.c
index 3f9c9327f149..1436905bc106 100644
--- a/drivers/net/netdevsim/ethtool.c
+++ b/drivers/net/netdevsim/ethtool.c
@@ -148,7 +148,7 @@ nsim_get_fec_stats(struct net_device *dev, struct ethtool_fec_stats *fec_stats)
}
static int nsim_get_ts_info(struct net_device *dev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct netdevsim *ns = netdev_priv(dev);
diff --git a/drivers/net/phy/bcm-phy-ptp.c b/drivers/net/phy/bcm-phy-ptp.c
index d3e825c951ee..874a1b64b115 100644
--- a/drivers/net/phy/bcm-phy-ptp.c
+++ b/drivers/net/phy/bcm-phy-ptp.c
@@ -841,7 +841,7 @@ static int bcm_ptp_hwtstamp(struct mii_timestamper *mii_ts,
}
static int bcm_ptp_ts_info(struct mii_timestamper *mii_ts,
- struct ethtool_ts_info *ts_info)
+ struct kernel_ethtool_ts_info *ts_info)
{
struct bcm_ptp_private *priv = mii2priv(mii_ts);
diff --git a/drivers/net/phy/dp83640.c b/drivers/net/phy/dp83640.c
index d3e72d5c1472..075d2beea716 100644
--- a/drivers/net/phy/dp83640.c
+++ b/drivers/net/phy/dp83640.c
@@ -1395,7 +1395,7 @@ static void dp83640_txtstamp(struct mii_timestamper *mii_ts,
}
static int dp83640_ts_info(struct mii_timestamper *mii_ts,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct dp83640_private *dp83640 =
container_of(mii_ts, struct dp83640_private, mii_ts);
diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c
index 8d57225d8575..dd519805deee 100644
--- a/drivers/net/phy/micrel.c
+++ b/drivers/net/phy/micrel.c
@@ -2552,7 +2552,7 @@ static void lan8814_ptp_tx_ts_get(struct phy_device *phydev,
*seq_id = lanphy_read_page_reg(phydev, 5, PTP_TX_MSG_HEADER2);
}
-static int lan8814_ts_info(struct mii_timestamper *mii_ts, struct ethtool_ts_info *info)
+static int lan8814_ts_info(struct mii_timestamper *mii_ts, struct kernel_ethtool_ts_info *info)
{
struct kszphy_ptp_priv *ptp_priv = container_of(mii_ts, struct kszphy_ptp_priv, mii_ts);
struct phy_device *phydev = ptp_priv->phydev;
@@ -4317,7 +4317,7 @@ static irqreturn_t lan8841_handle_interrupt(struct phy_device *phydev)
}
static int lan8841_ts_info(struct mii_timestamper *mii_ts,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct kszphy_ptp_priv *ptp_priv;
diff --git a/drivers/net/phy/mscc/mscc_ptp.c b/drivers/net/phy/mscc/mscc_ptp.c
index e66d20eff7c4..c1ddae36a2ae 100644
--- a/drivers/net/phy/mscc/mscc_ptp.c
+++ b/drivers/net/phy/mscc/mscc_ptp.c
@@ -1134,7 +1134,7 @@ static int vsc85xx_hwtstamp(struct mii_timestamper *mii_ts,
}
static int vsc85xx_ts_info(struct mii_timestamper *mii_ts,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct vsc8531_private *vsc8531 =
container_of(mii_ts, struct vsc8531_private, mii_ts);
diff --git a/drivers/net/phy/nxp-c45-tja11xx.c b/drivers/net/phy/nxp-c45-tja11xx.c
index d18c133e6013..5af5ade4fc64 100644
--- a/drivers/net/phy/nxp-c45-tja11xx.c
+++ b/drivers/net/phy/nxp-c45-tja11xx.c
@@ -1058,7 +1058,7 @@ static int nxp_c45_hwtstamp(struct mii_timestamper *mii_ts,
}
static int nxp_c45_ts_info(struct mii_timestamper *mii_ts,
- struct ethtool_ts_info *ts_info)
+ struct kernel_ethtool_ts_info *ts_info)
{
struct nxp_c45_phy *priv = container_of(mii_ts, struct nxp_c45_phy,
mii_ts);
diff --git a/drivers/ptp/ptp_ines.c b/drivers/ptp/ptp_ines.c
index 385643f3f8fe..e6f7d2bf8dde 100644
--- a/drivers/ptp/ptp_ines.c
+++ b/drivers/ptp/ptp_ines.c
@@ -556,7 +556,7 @@ static bool ines_timestamp_expired(struct ines_timestamp *ts)
}
static int ines_ts_info(struct mii_timestamper *mii_ts,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
info->so_timestamping =
SOF_TIMESTAMPING_TX_HARDWARE |
diff --git a/drivers/s390/net/qeth_ethtool.c b/drivers/s390/net/qeth_ethtool.c
index c1caf7734c3e..f184c58ecf24 100644
--- a/drivers/s390/net/qeth_ethtool.c
+++ b/drivers/s390/net/qeth_ethtool.c
@@ -247,7 +247,7 @@ static int qeth_set_channels(struct net_device *dev,
}
static int qeth_get_ts_info(struct net_device *dev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
struct qeth_card *card = dev->ml_priv;
diff --git a/include/linux/can/dev.h b/include/linux/can/dev.h
index 1b92aed49363..23492213ea35 100644
--- a/include/linux/can/dev.h
+++ b/include/linux/can/dev.h
@@ -186,7 +186,7 @@ void close_candev(struct net_device *dev);
int can_change_mtu(struct net_device *dev, int new_mtu);
int can_eth_ioctl_hwts(struct net_device *netdev, struct ifreq *ifr, int cmd);
int can_ethtool_op_get_ts_info_hwts(struct net_device *dev,
- struct ethtool_ts_info *info);
+ struct kernel_ethtool_ts_info *info);
int register_candev(struct net_device *dev);
void unregister_candev(struct net_device *dev);
diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index e213b5508da6..6b38bbda5790 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -18,6 +18,7 @@
#include <linux/if_ether.h>
#include <linux/netlink.h>
#include <uapi/linux/ethtool.h>
+#include <uapi/linux/net_tstamp.h>
struct compat_ethtool_rx_flow_spec {
u32 flow_type;
@@ -713,6 +714,22 @@ struct ethtool_rxfh_param {
u8 input_xfrm;
};
+/**
+ * struct kernel_ethtool_ts_info - kernel copy of struct ethtool_ts_info
+ * @cmd: command number = %ETHTOOL_GET_TS_INFO
+ * @so_timestamping: bit mask of the sum of the supported SO_TIMESTAMPING flags
+ * @phc_index: device index of the associated PHC, or -1 if there is none
+ * @tx_types: bit mask of the supported hwtstamp_tx_types enumeration values
+ * @rx_filters: bit mask of the supported hwtstamp_rx_filters enumeration values
+ */
+struct kernel_ethtool_ts_info {
+ u32 cmd;
+ u32 so_timestamping;
+ int phc_index;
+ enum hwtstamp_tx_types tx_types;
+ enum hwtstamp_rx_filters rx_filters;
+};
+
/**
* struct ethtool_ops - optional netdev operations
* @cap_link_lanes_supported: indicates if the driver supports lanes
@@ -1020,7 +1037,7 @@ struct ethtool_ops {
int (*get_dump_data)(struct net_device *,
struct ethtool_dump *, void *);
int (*set_dump)(struct net_device *, struct ethtool_dump *);
- int (*get_ts_info)(struct net_device *, struct ethtool_ts_info *);
+ int (*get_ts_info)(struct net_device *, struct kernel_ethtool_ts_info *);
void (*get_ts_stats)(struct net_device *dev,
struct ethtool_ts_stats *ts_stats);
int (*get_module_info)(struct net_device *,
@@ -1181,7 +1198,8 @@ int ethtool_get_phc_vclocks(struct net_device *dev, int **vclock_index);
/* Some generic methods drivers may use in their ethtool_ops */
u32 ethtool_op_get_link(struct net_device *dev);
-int ethtool_op_get_ts_info(struct net_device *dev, struct ethtool_ts_info *eti);
+int ethtool_op_get_ts_info(struct net_device *dev,
+ struct kernel_ethtool_ts_info *eti);
/**
* ethtool_mm_frag_size_add_to_min - Translate (standard) additional fragment
@@ -1230,7 +1248,8 @@ static inline int ethtool_mm_frag_size_min_to_add(u32 val_min, u32 *val_add,
* @info: buffer to hold the result
* Returns zero on success, non-zero otherwise.
*/
-int ethtool_get_ts_info_by_layer(struct net_device *dev, struct ethtool_ts_info *info);
+int ethtool_get_ts_info_by_layer(struct net_device *dev,
+ struct kernel_ethtool_ts_info *info);
/**
* ethtool_sprintf - Write formatted string to ethtool string data
diff --git a/include/linux/mii_timestamper.h b/include/linux/mii_timestamper.h
index 26b04f73f214..995db62570f9 100644
--- a/include/linux/mii_timestamper.h
+++ b/include/linux/mii_timestamper.h
@@ -59,7 +59,7 @@ struct mii_timestamper {
struct phy_device *phydev);
int (*ts_info)(struct mii_timestamper *mii_ts,
- struct ethtool_ts_info *ts_info);
+ struct kernel_ethtool_ts_info *ts_info);
struct device *device;
};
diff --git a/include/linux/phy.h b/include/linux/phy.h
index e7a38137211c..04ae5c811cfb 100644
--- a/include/linux/phy.h
+++ b/include/linux/phy.h
@@ -1618,7 +1618,7 @@ static inline bool phy_rxtstamp(struct phy_device *phydev, struct sk_buff *skb,
}
static inline int phy_ts_info(struct phy_device *phydev,
- struct ethtool_ts_info *tsinfo)
+ struct kernel_ethtool_ts_info *tsinfo)
{
return phydev->mii_ts->ts_info(phydev->mii_ts, tsinfo);
}
diff --git a/include/net/dsa.h b/include/net/dsa.h
index f9ae3ca66b6f..96efdd9f90c9 100644
--- a/include/net/dsa.h
+++ b/include/net/dsa.h
@@ -934,7 +934,7 @@ struct dsa_switch_ops {
* ethtool timestamp info
*/
int (*get_ts_info)(struct dsa_switch *ds, int port,
- struct ethtool_ts_info *ts);
+ struct kernel_ethtool_ts_info *ts);
/*
* ethtool MAC merge layer
diff --git a/include/soc/mscc/ocelot.h b/include/soc/mscc/ocelot.h
index 1e1b40f4e664..6a37b29f4b4c 100644
--- a/include/soc/mscc/ocelot.h
+++ b/include/soc/mscc/ocelot.h
@@ -1016,7 +1016,7 @@ void ocelot_port_get_eth_mac_stats(struct ocelot *ocelot, int port,
void ocelot_port_get_eth_phy_stats(struct ocelot *ocelot, int port,
struct ethtool_eth_phy_stats *phy_stats);
int ocelot_get_ts_info(struct ocelot *ocelot, int port,
- struct ethtool_ts_info *info);
+ struct kernel_ethtool_ts_info *info);
void ocelot_set_ageing_time(struct ocelot *ocelot, unsigned int msecs);
int ocelot_port_vlan_filtering(struct ocelot *ocelot, int port, bool enabled,
struct netlink_ext_ack *extack);
diff --git a/net/8021q/vlan_dev.c b/net/8021q/vlan_dev.c
index 3efba4f857ac..217be32426b5 100644
--- a/net/8021q/vlan_dev.c
+++ b/net/8021q/vlan_dev.c
@@ -677,7 +677,7 @@ static void vlan_ethtool_get_drvinfo(struct net_device *dev,
}
static int vlan_ethtool_get_ts_info(struct net_device *dev,
- struct ethtool_ts_info *info)
+ struct kernel_ethtool_ts_info *info)
{
const struct vlan_dev_priv *vlan = vlan_dev_priv(dev);
return ethtool_get_ts_info_by_layer(vlan->real_dev, info);
diff --git a/net/dsa/user.c b/net/dsa/user.c
index e8f56a40b614..f5adfa1d978a 100644
--- a/net/dsa/user.c
+++ b/net/dsa/user.c
@@ -1729,7 +1729,7 @@ static int dsa_user_set_rxnfc(struct net_device *dev,
}
static int dsa_user_get_ts_info(struct net_device *dev,
- struct ethtool_ts_info *ts)
+ struct kernel_ethtool_ts_info *ts)
{
struct dsa_user_priv *p = netdev_priv(dev);
struct dsa_switch *ds = p->dp->ds;
diff --git a/net/ethtool/common.c b/net/ethtool/common.c
index 01b7550f12c6..461017a37955 100644
--- a/net/ethtool/common.c
+++ b/net/ethtool/common.c
@@ -629,7 +629,7 @@ int ethtool_check_ops(const struct ethtool_ops *ops)
return 0;
}
-int __ethtool_get_ts_info(struct net_device *dev, struct ethtool_ts_info *info)
+int __ethtool_get_ts_info(struct net_device *dev, struct kernel_ethtool_ts_info *info)
{
const struct ethtool_ops *ops = dev->ethtool_ops;
struct phy_device *phydev = dev->phydev;
@@ -651,7 +651,7 @@ int __ethtool_get_ts_info(struct net_device *dev, struct ethtool_ts_info *info)
int ethtool_get_phc_vclocks(struct net_device *dev, int **vclock_index)
{
- struct ethtool_ts_info info = { };
+ struct kernel_ethtool_ts_info info = { };
int num = 0;
if (!__ethtool_get_ts_info(dev, &info))
@@ -661,7 +661,7 @@ int ethtool_get_phc_vclocks(struct net_device *dev, int **vclock_index)
}
EXPORT_SYMBOL(ethtool_get_phc_vclocks);
-int ethtool_get_ts_info_by_layer(struct net_device *dev, struct ethtool_ts_info *info)
+int ethtool_get_ts_info_by_layer(struct net_device *dev, struct kernel_ethtool_ts_info *info)
{
return __ethtool_get_ts_info(dev, info);
}
diff --git a/net/ethtool/common.h b/net/ethtool/common.h
index 28b8aaaf9bcb..b9daeecbd84d 100644
--- a/net/ethtool/common.h
+++ b/net/ethtool/common.h
@@ -44,7 +44,7 @@ bool convert_legacy_settings_to_link_ksettings(
const struct ethtool_cmd *legacy_settings);
int ethtool_get_max_rxfh_channel(struct net_device *dev, u32 *max);
int ethtool_get_max_rxnfc_channel(struct net_device *dev, u64 *max);
-int __ethtool_get_ts_info(struct net_device *dev, struct ethtool_ts_info *info);
+int __ethtool_get_ts_info(struct net_device *dev, struct kernel_ethtool_ts_info *info);
extern const struct ethtool_phy_ops *ethtool_phy_ops;
extern const struct ethtool_pse_ops *ethtool_pse_ops;
diff --git a/net/ethtool/ioctl.c b/net/ethtool/ioctl.c
index d72b0fec89af..bc8988ddf51c 100644
--- a/net/ethtool/ioctl.c
+++ b/net/ethtool/ioctl.c
@@ -65,7 +65,8 @@ u32 ethtool_op_get_link(struct net_device *dev)
}
EXPORT_SYMBOL(ethtool_op_get_link);
-int ethtool_op_get_ts_info(struct net_device *dev, struct ethtool_ts_info *info)
+int ethtool_op_get_ts_info(struct net_device *dev,
+ struct kernel_ethtool_ts_info *info)
{
info->so_timestamping =
SOF_TIMESTAMPING_TX_SOFTWARE |
@@ -2569,13 +2570,20 @@ static int ethtool_get_dump_data(struct net_device *dev,
static int ethtool_get_ts_info(struct net_device *dev, void __user *useraddr)
{
- struct ethtool_ts_info info;
+ struct kernel_ethtool_ts_info kernel_info;
+ struct ethtool_ts_info info = {};
int err;
- err = __ethtool_get_ts_info(dev, &info);
+ err = __ethtool_get_ts_info(dev, &kernel_info);
if (err)
return err;
+ info.cmd = kernel_info.cmd;
+ info.so_timestamping = kernel_info.so_timestamping;
+ info.phc_index = kernel_info.phc_index;
+ info.tx_types = kernel_info.tx_types;
+ info.rx_filters = kernel_info.rx_filters;
+
if (copy_to_user(useraddr, &info, sizeof(info)))
return -EFAULT;
diff --git a/net/ethtool/tsinfo.c b/net/ethtool/tsinfo.c
index 57d496287e52..03d12d6f79ca 100644
--- a/net/ethtool/tsinfo.c
+++ b/net/ethtool/tsinfo.c
@@ -12,7 +12,7 @@ struct tsinfo_req_info {
struct tsinfo_reply_data {
struct ethnl_reply_data base;
- struct ethtool_ts_info ts_info;
+ struct kernel_ethtool_ts_info ts_info;
struct ethtool_ts_stats stats;
};
@@ -55,7 +55,7 @@ static int tsinfo_reply_size(const struct ethnl_req_info *req_base,
{
const struct tsinfo_reply_data *data = TSINFO_REPDATA(reply_base);
bool compact = req_base->flags & ETHTOOL_FLAG_COMPACT_BITSETS;
- const struct ethtool_ts_info *ts_info = &data->ts_info;
+ const struct kernel_ethtool_ts_info *ts_info = &data->ts_info;
int len = 0;
int ret;
@@ -136,7 +136,7 @@ static int tsinfo_fill_reply(struct sk_buff *skb,
{
const struct tsinfo_reply_data *data = TSINFO_REPDATA(reply_base);
bool compact = req_base->flags & ETHTOOL_FLAG_COMPACT_BITSETS;
- const struct ethtool_ts_info *ts_info = &data->ts_info;
+ const struct kernel_ethtool_ts_info *ts_info = &data->ts_info;
int ret;
if (ts_info->so_timestamping) {
diff --git a/net/sched/sch_taprio.c b/net/sched/sch_taprio.c
index b284a06b5a75..cc2df9f8c14a 100644
--- a/net/sched/sch_taprio.c
+++ b/net/sched/sch_taprio.c
@@ -1610,7 +1610,7 @@ static int taprio_parse_clockid(struct Qdisc *sch, struct nlattr **tb,
if (FULL_OFFLOAD_IS_ENABLED(q->flags)) {
const struct ethtool_ops *ops = dev->ethtool_ops;
- struct ethtool_ts_info info = {
+ struct kernel_ethtool_ts_info info = {
.cmd = ETHTOOL_GET_TS_INFO,
.phc_index = -1,
};
--
2.34.1
^ permalink raw reply related [flat|nested] 39+ messages in thread
* [PATCH net-next v17 07/14] ptp: Add phc source and helpers to register specific PTP clock or get information
2024-07-09 13:53 [PATCH net-next v17 00/14] net: Make timestamping selectable Kory Maincent
` (5 preceding siblings ...)
2024-07-09 13:53 ` [PATCH net-next v17 06/14] net: Add struct kernel_ethtool_ts_info Kory Maincent
@ 2024-07-09 13:53 ` Kory Maincent
2024-07-15 23:44 ` Jacob Keller
2024-07-09 13:53 ` [PATCH net-next v17 08/14] net: Add the possibility to support a selected hwtstamp in netdevice Kory Maincent
` (7 subsequent siblings)
14 siblings, 1 reply; 39+ messages in thread
From: Kory Maincent @ 2024-07-09 13:53 UTC (permalink / raw)
To: Florian Fainelli, Broadcom internal kernel review list,
Andrew Lunn, Heiner Kallweit, Russell King, David S. Miller,
Eric Dumazet, Jakub Kicinski, Paolo Abeni, Richard Cochran,
Radu Pirea, Jay Vosburgh, Andy Gospodarek, Nicolas Ferre,
Claudiu Beznea, Willem de Bruijn, Jonathan Corbet, Horatiu Vultur,
UNGLinuxDriver, Simon Horman, Vladimir Oltean, donald.hunter,
danieller, ecree.xilinx
Cc: Thomas Petazzoni, linux-kernel, netdev, linux-doc,
Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter, Kory Maincent
Prepare for future hardware timestamp selection by adding source and
corresponding pointers to ptp_clock structure.
Additionally, introduce helpers for registering specific phydev or netdev
PTP clocks, retrieving PTP clock information such as hwtstamp source or
phydev/netdev pointers, and obtaining the ptp_clock structure from the
phc index.
These helpers are added to a new ptp_clock_consumer.c file, built as
builtin. This is necessary because these helpers will be called by
ethtool or net timestamping, which are builtin code.
Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
---
Change in v8:
- New patch.
Change in v10:
- Add get and put function to avoid unregistering a ptp clock object used.
- Fix kdoc issues.
Change in v11:
- Remove useless extern.
Change in v12:
- Add missing return description in the kdoc.
Change in v13:
- Remove a semicolon which bring errors while not building PTP driver.
- Remove few useless EXPORT_SYMBOL().
- Separate PTP consumer symbole which are builtin from PTP provider.
Change in v14:
- Add back missing EXPORT_SYMBOL().
---
drivers/ptp/Makefile | 5 ++
drivers/ptp/ptp_clock.c | 33 ++++++++++-
drivers/ptp/ptp_clock_consumer.c | 100 +++++++++++++++++++++++++++++++
drivers/ptp/ptp_private.h | 7 +++
include/linux/ptp_clock_kernel.h | 125 +++++++++++++++++++++++++++++++++++++++
5 files changed, 269 insertions(+), 1 deletion(-)
diff --git a/drivers/ptp/Makefile b/drivers/ptp/Makefile
index 68bf02078053..9d7226b995cf 100644
--- a/drivers/ptp/Makefile
+++ b/drivers/ptp/Makefile
@@ -3,6 +3,11 @@
# Makefile for PTP 1588 clock support.
#
+ifdef CONFIG_PTP_1588_CLOCK
+# The ptp_clock consumer is built-in whenever PTP_1588_CLOCK is built-in
+# or module
+obj-y := ptp_clock_consumer.o
+endif
ptp-y := ptp_clock.o ptp_chardev.o ptp_sysfs.o ptp_vclock.o
ptp_kvm-$(CONFIG_X86) := ptp_kvm_x86.o ptp_kvm_common.o
ptp_kvm-$(CONFIG_HAVE_ARM_SMCCC) := ptp_kvm_arm.o ptp_kvm_common.o
diff --git a/drivers/ptp/ptp_clock.c b/drivers/ptp/ptp_clock.c
index c56cd0f63909..593b5c906314 100644
--- a/drivers/ptp/ptp_clock.c
+++ b/drivers/ptp/ptp_clock.c
@@ -34,7 +34,6 @@ const struct class ptp_class = {
static dev_t ptp_devt;
-static DEFINE_XARRAY_ALLOC(ptp_clocks_map);
/* time stamp event queue operations */
@@ -512,6 +511,38 @@ void ptp_cancel_worker_sync(struct ptp_clock *ptp)
}
EXPORT_SYMBOL(ptp_cancel_worker_sync);
+struct ptp_clock *netdev_ptp_clock_register(struct ptp_clock_info *info,
+ struct net_device *dev)
+{
+ struct ptp_clock *ptp;
+
+ ptp = ptp_clock_register(info, &dev->dev);
+ if (IS_ERR(ptp))
+ return ptp;
+
+ ptp->phc_source = HWTSTAMP_SOURCE_NETDEV;
+ ptp->netdev = dev;
+
+ return ptp;
+}
+EXPORT_SYMBOL(netdev_ptp_clock_register);
+
+struct ptp_clock *phydev_ptp_clock_register(struct ptp_clock_info *info,
+ struct phy_device *phydev)
+{
+ struct ptp_clock *ptp;
+
+ ptp = ptp_clock_register(info, &phydev->mdio.dev);
+ if (IS_ERR(ptp))
+ return ptp;
+
+ ptp->phc_source = HWTSTAMP_SOURCE_PHYLIB;
+ ptp->phydev = phydev;
+
+ return ptp;
+}
+EXPORT_SYMBOL(phydev_ptp_clock_register);
+
/* module operations */
static void __exit ptp_exit(void)
diff --git a/drivers/ptp/ptp_clock_consumer.c b/drivers/ptp/ptp_clock_consumer.c
new file mode 100644
index 000000000000..58b0c8948fc8
--- /dev/null
+++ b/drivers/ptp/ptp_clock_consumer.c
@@ -0,0 +1,100 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * PTP 1588 clock support
+ *
+ * Copyright (C) 2010 OMICRON electronics GmbH
+ */
+#include <linux/device.h>
+#include <linux/err.h>
+#include <linux/init.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/posix-clock.h>
+#include <linux/pps_kernel.h>
+#include <linux/slab.h>
+#include <linux/syscalls.h>
+#include <linux/uaccess.h>
+#include <linux/debugfs.h>
+#include <linux/xarray.h>
+#include <uapi/linux/sched/types.h>
+
+#include "ptp_private.h"
+
+DEFINE_XARRAY_ALLOC(ptp_clocks_map);
+EXPORT_SYMBOL(ptp_clocks_map);
+
+bool ptp_clock_from_phylib(struct ptp_clock *ptp)
+{
+ return ptp->phc_source == HWTSTAMP_SOURCE_PHYLIB;
+}
+
+bool ptp_clock_from_netdev(struct ptp_clock *ptp)
+{
+ return ptp->phc_source == HWTSTAMP_SOURCE_NETDEV;
+}
+
+struct net_device *ptp_clock_netdev(struct ptp_clock *ptp)
+{
+ if (ptp->phc_source != HWTSTAMP_SOURCE_NETDEV)
+ return NULL;
+
+ return ptp->netdev;
+}
+
+struct phy_device *ptp_clock_phydev(struct ptp_clock *ptp)
+{
+ if (ptp->phc_source != HWTSTAMP_SOURCE_PHYLIB)
+ return NULL;
+
+ return ptp->phydev;
+}
+EXPORT_SYMBOL(ptp_clock_phydev);
+
+int ptp_clock_get(struct device *dev, struct ptp_clock *ptp)
+{
+ struct device_link *link;
+
+ if (!ptp)
+ return 0;
+
+ if (!try_module_get(ptp->info->owner))
+ return -EPROBE_DEFER;
+
+ get_device(&ptp->dev);
+
+ link = device_link_add(dev, &ptp->dev, DL_FLAG_STATELESS);
+ if (!link)
+ dev_warn(dev, "failed to create device link to %s\n",
+ dev_name(&ptp->dev));
+
+ return 0;
+}
+
+struct ptp_clock *ptp_clock_get_by_index(struct device *dev, int index)
+{
+ struct ptp_clock *ptp;
+ int ret;
+
+ if (index < 0)
+ return NULL;
+
+ ptp = xa_load(&ptp_clocks_map, (unsigned long)index);
+ if (IS_ERR_OR_NULL(ptp))
+ return ptp;
+
+ ret = ptp_clock_get(dev, ptp);
+ if (ret)
+ return ERR_PTR(ret);
+
+ return ptp;
+}
+
+void ptp_clock_put(struct device *dev, struct ptp_clock *ptp)
+{
+ if (!ptp)
+ return;
+
+ device_link_remove(dev, &ptp->dev);
+ put_device(&ptp->dev);
+ module_put(ptp->info->owner);
+}
diff --git a/drivers/ptp/ptp_private.h b/drivers/ptp/ptp_private.h
index 18934e28469e..6a306e6e34c2 100644
--- a/drivers/ptp/ptp_private.h
+++ b/drivers/ptp/ptp_private.h
@@ -24,6 +24,8 @@
#define PTP_DEFAULT_MAX_VCLOCKS 20
#define PTP_MAX_CHANNELS 2048
+extern struct xarray ptp_clocks_map;
+
struct timestamp_event_queue {
struct ptp_extts_event buf[PTP_MAX_TIMESTAMPS];
int head;
@@ -41,6 +43,11 @@ struct ptp_clock {
struct ptp_clock_info *info;
dev_t devid;
int index; /* index into clocks.map */
+ enum hwtstamp_source phc_source;
+ union { /* Pointer of the phc_source device */
+ struct net_device *netdev;
+ struct phy_device *phydev;
+ };
struct pps_device *pps_source;
long dialed_frequency; /* remembers the frequency adjustment */
struct list_head tsevqs; /* timestamp fifo list */
diff --git a/include/linux/ptp_clock_kernel.h b/include/linux/ptp_clock_kernel.h
index 6e4b8206c7d0..860c1e293a62 100644
--- a/include/linux/ptp_clock_kernel.h
+++ b/include/linux/ptp_clock_kernel.h
@@ -9,7 +9,9 @@
#define _PTP_CLOCK_KERNEL_H_
#include <linux/device.h>
+#include <linux/netdevice.h>
#include <linux/pps_kernel.h>
+#include <linux/phy.h>
#include <linux/ptp_clock.h>
#include <linux/timecounter.h>
#include <linux/skbuff.h>
@@ -340,6 +342,106 @@ extern void ptp_clock_event(struct ptp_clock *ptp,
extern int ptp_clock_index(struct ptp_clock *ptp);
+/**
+ * netdev_ptp_clock_register() - Register a PTP hardware clock driver for
+ * a net device
+ *
+ * @info: Structure describing the new clock.
+ * @dev: Pointer of the net device.
+ *
+ * Return: Pointer of the PTP clock, error pointer otherwise.
+ */
+
+struct ptp_clock *
+netdev_ptp_clock_register(struct ptp_clock_info *info,
+ struct net_device *dev);
+
+/**
+ * phydev_ptp_clock_register() - Register a PTP hardware clock driver for
+ * a phy device
+ *
+ * @info: Structure describing the new clock.
+ * @phydev: Pointer of the phy device.
+ *
+ * Return: Pointer of the PTP clock, error pointer otherwise.
+ */
+
+struct ptp_clock *
+phydev_ptp_clock_register(struct ptp_clock_info *info,
+ struct phy_device *phydev);
+
+/**
+ * ptp_clock_from_phylib() - Does the PTP clock comes from phylib
+ *
+ * @ptp: The clock obtained from net/phy_ptp_clock_register().
+ *
+ * Return: True if the PTP clock comes from phylib, false otherwise.
+ */
+
+bool ptp_clock_from_phylib(struct ptp_clock *ptp);
+
+/**
+ * ptp_clock_from_netdev() - Does the PTP clock comes from netdev
+ *
+ * @ptp: The clock obtained from net/phy_ptp_clock_register().
+ *
+ * Return: True if the PTP clock comes from netdev, false otherwise.
+ */
+
+bool ptp_clock_from_netdev(struct ptp_clock *ptp);
+
+/**
+ * ptp_clock_netdev() - Obtain the net_device reference of PTP clock
+ *
+ * @ptp: The clock obtained from netdev_ptp_clock_register().
+ *
+ * Return: Pointer of the net device, NULL otherwise.
+ */
+
+struct net_device *ptp_clock_netdev(struct ptp_clock *ptp);
+
+/**
+ * ptp_clock_phydev() - Obtain the phy_device reference of a PTP clock
+ *
+ * @ptp: The clock obtained from phydev_ptp_clock_register().
+ *
+ * Return: Pointer of the phy device, NULL otherwise.
+ */
+
+struct phy_device *ptp_clock_phydev(struct ptp_clock *ptp);
+
+/**
+ * ptp_clock_get() - Increment refcount of the PTP clock
+ *
+ * @dev: The device which get the PTP clock.
+ * @ptp: Pointer of a PTP clock.
+ *
+ * Return: 0 in case of success, error otherwise.
+ */
+
+int ptp_clock_get(struct device *dev, struct ptp_clock *ptp);
+
+/**
+ * ptp_clock_get_by_index() - Obtain the PTP clock reference from a given
+ * PHC index
+ *
+ * @dev: The device which get the PTP clock.
+ * @index: The device index of a PTP clock.
+ *
+ * Return: Pointer of the PTP clock, error pointer otherwise.
+ */
+
+struct ptp_clock *ptp_clock_get_by_index(struct device *dev, int index);
+
+/**
+ * ptp_clock_put() - decrement refcount of the PTP clock
+ *
+ * @dev: The device which get the PTP clock.
+ * @ptp: Pointer of a PTP clock.
+ */
+
+void ptp_clock_put(struct device *dev, struct ptp_clock *ptp);
+
/**
* ptp_find_pin() - obtain the pin index of a given auxiliary function
*
@@ -405,6 +507,29 @@ static inline void ptp_clock_event(struct ptp_clock *ptp,
{ }
static inline int ptp_clock_index(struct ptp_clock *ptp)
{ return -1; }
+static inline struct ptp_clock *
+netdev_ptp_clock_register(struct ptp_clock_info *info,
+ struct net_device *dev)
+{ return NULL; }
+static inline struct ptp_clock *
+phydev_ptp_clock_register(struct ptp_clock_info *info,
+ struct phy_device *phydev)
+{ return NULL; }
+static inline bool ptp_clock_from_phylib(struct ptp_clock *ptp)
+{ return false; }
+static inline bool ptp_clock_from_netdev(struct ptp_clock *ptp)
+{ return false; }
+static inline struct net_device *ptp_clock_netdev(struct ptp_clock *ptp)
+{ return NULL; }
+static inline struct phy_device *ptp_clock_phydev(struct ptp_clock *ptp)
+{ return NULL; }
+static inline int ptp_clock_get(struct device *dev, struct ptp_clock *ptp)
+{ return -ENODEV; }
+static inline void ptp_clock_put(struct device *dev, struct ptp_clock *ptp)
+{ }
+static inline struct ptp_clock *ptp_clock_get_by_index(struct device *dev,
+ int index)
+{ return NULL; }
static inline int ptp_find_pin(struct ptp_clock *ptp,
enum ptp_pin_function func, unsigned int chan)
{ return -1; }
--
2.34.1
^ permalink raw reply related [flat|nested] 39+ messages in thread
* [PATCH net-next v17 08/14] net: Add the possibility to support a selected hwtstamp in netdevice
2024-07-09 13:53 [PATCH net-next v17 00/14] net: Make timestamping selectable Kory Maincent
` (6 preceding siblings ...)
2024-07-09 13:53 ` [PATCH net-next v17 07/14] ptp: Add phc source and helpers to register specific PTP clock or get information Kory Maincent
@ 2024-07-09 13:53 ` Kory Maincent
2024-07-17 17:22 ` Jacob Keller
2024-07-09 13:53 ` [PATCH net-next v17 09/14] net: netdevsim: ptp_mock: Convert to netdev_ptp_clock_register Kory Maincent
` (6 subsequent siblings)
14 siblings, 1 reply; 39+ messages in thread
From: Kory Maincent @ 2024-07-09 13:53 UTC (permalink / raw)
To: Florian Fainelli, Broadcom internal kernel review list,
Andrew Lunn, Heiner Kallweit, Russell King, David S. Miller,
Eric Dumazet, Jakub Kicinski, Paolo Abeni, Richard Cochran,
Radu Pirea, Jay Vosburgh, Andy Gospodarek, Nicolas Ferre,
Claudiu Beznea, Willem de Bruijn, Jonathan Corbet, Horatiu Vultur,
UNGLinuxDriver, Simon Horman, Vladimir Oltean, donald.hunter,
danieller, ecree.xilinx
Cc: Thomas Petazzoni, linux-kernel, netdev, linux-doc,
Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter, Kory Maincent
Introduce the description of a hwtstamp provider, mainly defined with a
ptp_clock pointer and a qualifier value.
Add a hwtstamp provider description within the netdev structure to
allow saving the hwtstamp we want to use. This prepares for future
support of an ethtool netlink command to select the desired hwtstamp
provider. By default, the old API that does not support hwtstamp
selectability is used, meaning the hwtstamp ptp_clock pointer is unset.
Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
---
Change in v8:
- New patch
Change in v10:
- Set hwtstamp in netdevice as a pointer for future use of rcu lock.
- Fix a nit in teh order of setting phydev pointer.
- Add a missing kdoc description.
Change in v13:
- Remove an include from netdevices.h.
Change in v16:
- Import the part of the patch 12 which belong to the hwtstamp provider
selectability of net core.
---
drivers/net/phy/phy_device.c | 11 +++++++++
drivers/ptp/ptp_clock_consumer.c | 10 ++++++++
include/linux/net_tstamp.h | 18 +++++++++++++++
include/linux/netdevice.h | 5 ++++
include/linux/ptp_clock_kernel.h | 10 ++++++++
include/uapi/linux/net_tstamp.h | 11 +++++++++
net/core/dev_ioctl.c | 43 ++++++++++++++++++++++++++++++++--
net/core/timestamping.c | 50 ++++++++++++++++++++++++++++++++++++----
8 files changed, 151 insertions(+), 7 deletions(-)
diff --git a/drivers/net/phy/phy_device.c b/drivers/net/phy/phy_device.c
index 70b07e621fb2..57a9943bb1c1 100644
--- a/drivers/net/phy/phy_device.c
+++ b/drivers/net/phy/phy_device.c
@@ -31,6 +31,7 @@
#include <linux/phy_led_triggers.h>
#include <linux/pse-pd/pse.h>
#include <linux/property.h>
+#include <linux/ptp_clock_kernel.h>
#include <linux/rtnetlink.h>
#include <linux/sfp.h>
#include <linux/skbuff.h>
@@ -1936,6 +1937,16 @@ void phy_detach(struct phy_device *phydev)
phy_suspend(phydev);
if (dev) {
+ struct hwtstamp_provider *hwtstamp;
+
+ hwtstamp = rtnl_dereference(dev->hwtstamp);
+ /* Disable timestamp if selected */
+ if (hwtstamp && ptp_clock_phydev(hwtstamp->ptp) == phydev) {
+ rcu_assign_pointer(dev->hwtstamp, NULL);
+ call_rcu(&hwtstamp->rcu_head,
+ remove_hwtstamp_provider);
+ }
+
phydev->attached_dev->phydev = NULL;
phydev->attached_dev = NULL;
}
diff --git a/drivers/ptp/ptp_clock_consumer.c b/drivers/ptp/ptp_clock_consumer.c
index 58b0c8948fc8..f5fab1c14b47 100644
--- a/drivers/ptp/ptp_clock_consumer.c
+++ b/drivers/ptp/ptp_clock_consumer.c
@@ -98,3 +98,13 @@ void ptp_clock_put(struct device *dev, struct ptp_clock *ptp)
put_device(&ptp->dev);
module_put(ptp->info->owner);
}
+
+void remove_hwtstamp_provider(struct rcu_head *rcu_head)
+{
+ struct hwtstamp_provider *hwtstamp;
+
+ hwtstamp = container_of(rcu_head, struct hwtstamp_provider, rcu_head);
+ ptp_clock_put(hwtstamp->dev, hwtstamp->ptp);
+ kfree(hwtstamp);
+}
+EXPORT_SYMBOL(remove_hwtstamp_provider);
diff --git a/include/linux/net_tstamp.h b/include/linux/net_tstamp.h
index 662074b08c94..077037f9ad0f 100644
--- a/include/linux/net_tstamp.h
+++ b/include/linux/net_tstamp.h
@@ -19,6 +19,22 @@ enum hwtstamp_source {
HWTSTAMP_SOURCE_PHYLIB,
};
+/**
+ * struct hwtstamp_provider - Description of a hwtstamp provider object
+ *
+ * @rcu_head: RCU callback used to free the struct.
+ * @dev: device attached to the hwtstamp provider used to put the ptp clock.
+ * @ptp: PTP clock pointer of the hwtstamp provider.
+ * @qualifier: hwtstamp provider qualifier.
+ */
+
+struct hwtstamp_provider {
+ struct rcu_head rcu_head;
+ struct device *dev;
+ struct ptp_clock *ptp;
+ enum hwtstamp_provider_qualifier qualifier;
+};
+
/**
* struct kernel_hwtstamp_config - Kernel copy of struct hwtstamp_config
*
@@ -31,6 +47,7 @@ enum hwtstamp_source {
* copied the ioctl request back to user space
* @source: indication whether timestamps should come from the netdev or from
* an attached phylib PHY
+ * @qualifier: qualifier of the hwtstamp provider
*
* Prefer using this structure for in-kernel processing of hardware
* timestamping configuration, over the inextensible struct hwtstamp_config
@@ -43,6 +60,7 @@ struct kernel_hwtstamp_config {
struct ifreq *ifr;
bool copied_to_user;
enum hwtstamp_source source;
+ enum hwtstamp_provider_qualifier qualifier;
};
static inline void hwtstamp_config_to_kernel(struct kernel_hwtstamp_config *kernel_cfg,
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 93558645c6d0..c18761c36455 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -81,6 +81,7 @@ struct xdp_frame;
struct xdp_metadata_ops;
struct xdp_md;
struct ethtool_netdev_state;
+struct hwtstamp_provider;
typedef u32 xdp_features_t;
@@ -2027,6 +2028,8 @@ enum netdev_reg_state {
* @dpll_pin: Pointer to the SyncE source pin of a DPLL subsystem,
* where the clock is recovered.
*
+ * @hwtstamp: Tracks which PTP performs hardware packet time stamping.
+ *
* FIXME: cleanup struct net_device such that network protocol info
* moves out.
*/
@@ -2406,6 +2409,8 @@ struct net_device {
/** @irq_moder: dim parameters used if IS_ENABLED(CONFIG_DIMLIB). */
struct dim_irq_moder *irq_moder;
+
+ struct hwtstamp_provider __rcu *hwtstamp;
};
#define to_net_dev(d) container_of(d, struct net_device, dev)
diff --git a/include/linux/ptp_clock_kernel.h b/include/linux/ptp_clock_kernel.h
index 860c1e293a62..6d9e6cd5c832 100644
--- a/include/linux/ptp_clock_kernel.h
+++ b/include/linux/ptp_clock_kernel.h
@@ -442,6 +442,14 @@ struct ptp_clock *ptp_clock_get_by_index(struct device *dev, int index);
void ptp_clock_put(struct device *dev, struct ptp_clock *ptp);
+/**
+ * remove_hwtstamp_provider() - Put and free the hwtstamp provider
+ *
+ * @rcu_head: RCU callback head.
+ */
+
+void remove_hwtstamp_provider(struct rcu_head *rcu_head);
+
/**
* ptp_find_pin() - obtain the pin index of a given auxiliary function
*
@@ -530,6 +538,8 @@ static inline void ptp_clock_put(struct device *dev, struct ptp_clock *ptp)
static inline struct ptp_clock *ptp_clock_get_by_index(struct device *dev,
int index)
{ return NULL; }
+static inline void remove_hwtstamp_provider(struct rcu_head *rcu_head)
+{ return; }
static inline int ptp_find_pin(struct ptp_clock *ptp,
enum ptp_pin_function func, unsigned int chan)
{ return -1; }
diff --git a/include/uapi/linux/net_tstamp.h b/include/uapi/linux/net_tstamp.h
index a2c66b3d7f0f..a9ed48ee8fc7 100644
--- a/include/uapi/linux/net_tstamp.h
+++ b/include/uapi/linux/net_tstamp.h
@@ -13,6 +13,17 @@
#include <linux/types.h>
#include <linux/socket.h> /* for SO_TIMESTAMPING */
+/*
+ * Possible type of htstamp provider. Mainly "precise" the default one
+ * is for IEEE 1588 quality and "approx" is for NICs DMA point.
+ */
+enum hwtstamp_provider_qualifier {
+ HWTSTAMP_PROVIDER_QUALIFIER_PRECISE,
+ HWTSTAMP_PROVIDER_QUALIFIER_APPROX,
+
+ HWTSTAMP_PROVIDER_QUALIFIER_CNT,
+};
+
/* SO_TIMESTAMPING flags */
enum {
SOF_TIMESTAMPING_TX_HARDWARE = (1<<0),
diff --git a/net/core/dev_ioctl.c b/net/core/dev_ioctl.c
index 36cea843381f..350d123c09c9 100644
--- a/net/core/dev_ioctl.c
+++ b/net/core/dev_ioctl.c
@@ -6,6 +6,7 @@
#include <linux/rtnetlink.h>
#include <linux/net_tstamp.h>
#include <linux/phylib_stubs.h>
+#include <linux/ptp_clock_kernel.h>
#include <linux/wireless.h>
#include <linux/if_bridge.h>
#include <net/dsa_stubs.h>
@@ -269,6 +270,22 @@ static int dev_eth_ioctl(struct net_device *dev,
int dev_get_hwtstamp_phylib(struct net_device *dev,
struct kernel_hwtstamp_config *cfg)
{
+ struct hwtstamp_provider *hwtstamp;
+
+ hwtstamp = rtnl_dereference(dev->hwtstamp);
+ if (hwtstamp) {
+ struct ptp_clock *ptp = hwtstamp->ptp;
+
+ cfg->qualifier = hwtstamp->qualifier;
+ if (ptp_clock_from_phylib(ptp))
+ return phy_hwtstamp_get(ptp_clock_phydev(ptp), cfg);
+
+ if (ptp_clock_from_netdev(ptp))
+ return dev->netdev_ops->ndo_hwtstamp_get(dev, cfg);
+
+ return -EOPNOTSUPP;
+ }
+
if (phy_is_default_hwtstamp(dev->phydev))
return phy_hwtstamp_get(dev->phydev, cfg);
@@ -325,11 +342,33 @@ int dev_set_hwtstamp_phylib(struct net_device *dev,
struct netlink_ext_ack *extack)
{
const struct net_device_ops *ops = dev->netdev_ops;
- bool phy_ts = phy_is_default_hwtstamp(dev->phydev);
struct kernel_hwtstamp_config old_cfg = {};
+ struct hwtstamp_provider *hwtstamp;
+ struct phy_device *phydev;
bool changed = false;
+ bool phy_ts;
int err;
+ hwtstamp = rtnl_dereference(dev->hwtstamp);
+ if (hwtstamp) {
+ struct ptp_clock *ptp = hwtstamp->ptp;
+
+ if (ptp_clock_from_phylib(ptp)) {
+ phy_ts = true;
+ phydev = ptp_clock_phydev(ptp);
+ } else if (ptp_clock_from_netdev(ptp)) {
+ phy_ts = false;
+ } else {
+ return -EOPNOTSUPP;
+ }
+
+ cfg->qualifier = hwtstamp->qualifier;
+ } else {
+ phy_ts = phy_is_default_hwtstamp(dev->phydev);
+ if (phy_ts)
+ phydev = dev->phydev;
+ }
+
cfg->source = phy_ts ? HWTSTAMP_SOURCE_PHYLIB : HWTSTAMP_SOURCE_NETDEV;
if (phy_ts && (dev->priv_flags & IFF_SEE_ALL_HWTSTAMP_REQUESTS)) {
@@ -351,7 +390,7 @@ int dev_set_hwtstamp_phylib(struct net_device *dev,
changed = kernel_hwtstamp_config_changed(&old_cfg, cfg);
if (phy_ts) {
- err = phy_hwtstamp_set(dev->phydev, cfg, extack);
+ err = phy_hwtstamp_set(phydev, cfg, extack);
if (err) {
if (changed)
ops->ndo_hwtstamp_set(dev, &old_cfg, NULL);
diff --git a/net/core/timestamping.c b/net/core/timestamping.c
index 3717fb152ecc..f8522fefd468 100644
--- a/net/core/timestamping.c
+++ b/net/core/timestamping.c
@@ -9,6 +9,7 @@
#include <linux/ptp_classify.h>
#include <linux/skbuff.h>
#include <linux/export.h>
+#include <linux/ptp_clock_kernel.h>
static unsigned int classify(const struct sk_buff *skb)
{
@@ -21,19 +22,38 @@ static unsigned int classify(const struct sk_buff *skb)
void skb_clone_tx_timestamp(struct sk_buff *skb)
{
+ struct hwtstamp_provider *hwtstamp;
struct mii_timestamper *mii_ts;
+ struct phy_device *phydev;
struct sk_buff *clone;
unsigned int type;
- if (!skb->sk || !skb->dev ||
- !phy_is_default_hwtstamp(skb->dev->phydev))
+ if (!skb->sk || !skb->dev)
return;
+ rcu_read_lock();
+ hwtstamp = rcu_dereference(skb->dev->hwtstamp);
+ if (hwtstamp) {
+ if (!ptp_clock_from_phylib(hwtstamp->ptp)) {
+ rcu_read_unlock();
+ return;
+ }
+
+ phydev = ptp_clock_phydev(hwtstamp->ptp);
+ } else {
+ phydev = skb->dev->phydev;
+ if (!phy_is_default_hwtstamp(phydev)) {
+ rcu_read_unlock();
+ return;
+ }
+ }
+ rcu_read_unlock();
+
type = classify(skb);
if (type == PTP_CLASS_NONE)
return;
- mii_ts = skb->dev->phydev->mii_ts;
+ mii_ts = phydev->mii_ts;
if (likely(mii_ts->txtstamp)) {
clone = skb_clone_sk(skb);
if (!clone)
@@ -45,12 +65,32 @@ EXPORT_SYMBOL_GPL(skb_clone_tx_timestamp);
bool skb_defer_rx_timestamp(struct sk_buff *skb)
{
+ struct hwtstamp_provider *hwtstamp;
struct mii_timestamper *mii_ts;
+ struct phy_device *phydev;
unsigned int type;
- if (!skb->dev || !phy_is_default_hwtstamp(skb->dev->phydev))
+ if (!skb->dev)
return false;
+ rcu_read_lock();
+ hwtstamp = rcu_dereference(skb->dev->hwtstamp);
+ if (hwtstamp) {
+ if (!ptp_clock_from_phylib(hwtstamp->ptp)) {
+ rcu_read_unlock();
+ return false;
+ }
+
+ phydev = ptp_clock_phydev(hwtstamp->ptp);
+ } else {
+ phydev = skb->dev->phydev;
+ if (!phy_is_default_hwtstamp(phydev)) {
+ rcu_read_unlock();
+ return false;
+ }
+ }
+ rcu_read_unlock();
+
if (skb_headroom(skb) < ETH_HLEN)
return false;
@@ -63,7 +103,7 @@ bool skb_defer_rx_timestamp(struct sk_buff *skb)
if (type == PTP_CLASS_NONE)
return false;
- mii_ts = skb->dev->phydev->mii_ts;
+ mii_ts = phydev->mii_ts;
if (likely(mii_ts->rxtstamp))
return mii_ts->rxtstamp(mii_ts, skb, type);
--
2.34.1
^ permalink raw reply related [flat|nested] 39+ messages in thread
* [PATCH net-next v17 09/14] net: netdevsim: ptp_mock: Convert to netdev_ptp_clock_register
2024-07-09 13:53 [PATCH net-next v17 00/14] net: Make timestamping selectable Kory Maincent
` (7 preceding siblings ...)
2024-07-09 13:53 ` [PATCH net-next v17 08/14] net: Add the possibility to support a selected hwtstamp in netdevice Kory Maincent
@ 2024-07-09 13:53 ` Kory Maincent
2024-07-17 17:23 ` Jacob Keller
2024-07-09 13:53 ` [PATCH net-next v17 10/14] net: macb: " Kory Maincent
` (5 subsequent siblings)
14 siblings, 1 reply; 39+ messages in thread
From: Kory Maincent @ 2024-07-09 13:53 UTC (permalink / raw)
To: Florian Fainelli, Broadcom internal kernel review list,
Andrew Lunn, Heiner Kallweit, Russell King, David S. Miller,
Eric Dumazet, Jakub Kicinski, Paolo Abeni, Richard Cochran,
Radu Pirea, Jay Vosburgh, Andy Gospodarek, Nicolas Ferre,
Claudiu Beznea, Willem de Bruijn, Jonathan Corbet, Horatiu Vultur,
UNGLinuxDriver, Simon Horman, Vladimir Oltean, donald.hunter,
danieller, ecree.xilinx
Cc: Thomas Petazzoni, linux-kernel, netdev, linux-doc,
Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter, Kory Maincent
The hardware registration clock for net device is now using
netdev_ptp_clock_register to save the net_device pointer within the PTP
clock xarray. netdevsim is registering its ptp through the mock driver.
It is the only driver using the mock driver to register a ptp clock.
Convert the mock driver to the new API.
Reviewed-by: Rahul Rameshbabu <rrameshbabu@nvidia.com>
Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
---
Changes in v8:
- New patch
---
drivers/net/netdevsim/netdev.c | 19 +++++++++++--------
drivers/ptp/ptp_mock.c | 4 ++--
include/linux/ptp_mock.h | 4 ++--
3 files changed, 15 insertions(+), 12 deletions(-)
diff --git a/drivers/net/netdevsim/netdev.c b/drivers/net/netdevsim/netdev.c
index 017a6102be0a..7806dea5d77f 100644
--- a/drivers/net/netdevsim/netdev.c
+++ b/drivers/net/netdevsim/netdev.c
@@ -663,17 +663,12 @@ static int nsim_init_netdevsim(struct netdevsim *ns)
struct mock_phc *phc;
int err;
- phc = mock_phc_create(&ns->nsim_bus_dev->dev);
- if (IS_ERR(phc))
- return PTR_ERR(phc);
-
- ns->phc = phc;
ns->netdev->netdev_ops = &nsim_netdev_ops;
ns->netdev->stat_ops = &nsim_stat_ops;
err = nsim_udp_tunnels_info_create(ns->nsim_dev, ns->netdev);
if (err)
- goto err_phc_destroy;
+ return err;
rtnl_lock();
err = nsim_queue_init(ns);
@@ -691,8 +686,18 @@ static int nsim_init_netdevsim(struct netdevsim *ns)
if (err)
goto err_ipsec_teardown;
rtnl_unlock();
+
+ phc = mock_phc_create(ns->netdev);
+ if (IS_ERR(phc)) {
+ err = PTR_ERR(phc);
+ goto err_register_netdevice;
+ }
+
+ ns->phc = phc;
return 0;
+err_register_netdevice:
+ unregister_netdevice(ns->netdev);
err_ipsec_teardown:
nsim_ipsec_teardown(ns);
nsim_macsec_teardown(ns);
@@ -702,8 +707,6 @@ static int nsim_init_netdevsim(struct netdevsim *ns)
err_utn_destroy:
rtnl_unlock();
nsim_udp_tunnels_info_destroy(ns->netdev);
-err_phc_destroy:
- mock_phc_destroy(ns->phc);
return err;
}
diff --git a/drivers/ptp/ptp_mock.c b/drivers/ptp/ptp_mock.c
index e7b459c846a2..1dcbe7426746 100644
--- a/drivers/ptp/ptp_mock.c
+++ b/drivers/ptp/ptp_mock.c
@@ -115,7 +115,7 @@ int mock_phc_index(struct mock_phc *phc)
}
EXPORT_SYMBOL_GPL(mock_phc_index);
-struct mock_phc *mock_phc_create(struct device *dev)
+struct mock_phc *mock_phc_create(struct net_device *dev)
{
struct mock_phc *phc;
int err;
@@ -147,7 +147,7 @@ struct mock_phc *mock_phc_create(struct device *dev)
spin_lock_init(&phc->lock);
timecounter_init(&phc->tc, &phc->cc, 0);
- phc->clock = ptp_clock_register(&phc->info, dev);
+ phc->clock = netdev_ptp_clock_register(&phc->info, dev);
if (IS_ERR(phc->clock)) {
err = PTR_ERR(phc->clock);
goto out_free_phc;
diff --git a/include/linux/ptp_mock.h b/include/linux/ptp_mock.h
index 72eb401034d9..e226011071f8 100644
--- a/include/linux/ptp_mock.h
+++ b/include/linux/ptp_mock.h
@@ -13,13 +13,13 @@ struct mock_phc;
#if IS_ENABLED(CONFIG_PTP_1588_CLOCK_MOCK)
-struct mock_phc *mock_phc_create(struct device *dev);
+struct mock_phc *mock_phc_create(struct net_device *dev);
void mock_phc_destroy(struct mock_phc *phc);
int mock_phc_index(struct mock_phc *phc);
#else
-static inline struct mock_phc *mock_phc_create(struct device *dev)
+static inline struct mock_phc *mock_phc_create(struct net_device *dev)
{
return NULL;
}
--
2.34.1
^ permalink raw reply related [flat|nested] 39+ messages in thread
* [PATCH net-next v17 10/14] net: macb: Convert to netdev_ptp_clock_register
2024-07-09 13:53 [PATCH net-next v17 00/14] net: Make timestamping selectable Kory Maincent
` (8 preceding siblings ...)
2024-07-09 13:53 ` [PATCH net-next v17 09/14] net: netdevsim: ptp_mock: Convert to netdev_ptp_clock_register Kory Maincent
@ 2024-07-09 13:53 ` Kory Maincent
2024-07-17 17:24 ` Jacob Keller
2024-07-09 13:53 ` [PATCH net-next v17 11/14] net: ptp: Move ptp_clock_index() to builtin symbol Kory Maincent
` (4 subsequent siblings)
14 siblings, 1 reply; 39+ messages in thread
From: Kory Maincent @ 2024-07-09 13:53 UTC (permalink / raw)
To: Florian Fainelli, Broadcom internal kernel review list,
Andrew Lunn, Heiner Kallweit, Russell King, David S. Miller,
Eric Dumazet, Jakub Kicinski, Paolo Abeni, Richard Cochran,
Radu Pirea, Jay Vosburgh, Andy Gospodarek, Nicolas Ferre,
Claudiu Beznea, Willem de Bruijn, Jonathan Corbet, Horatiu Vultur,
UNGLinuxDriver, Simon Horman, Vladimir Oltean, donald.hunter,
danieller, ecree.xilinx
Cc: Thomas Petazzoni, linux-kernel, netdev, linux-doc,
Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter, Kory Maincent
The hardware registration clock for net device is now using
netdev_ptp_clock_register to save the net_device pointer within the ptp
clock xarray. Convert the macb driver to the new API.
Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
---
Change in v8:
- New patch
---
drivers/net/ethernet/cadence/macb_ptp.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/cadence/macb_ptp.c b/drivers/net/ethernet/cadence/macb_ptp.c
index a63bf29c4fa8..50fa62a0ddc5 100644
--- a/drivers/net/ethernet/cadence/macb_ptp.c
+++ b/drivers/net/ethernet/cadence/macb_ptp.c
@@ -332,7 +332,7 @@ void gem_ptp_init(struct net_device *dev)
bp->tsu_rate = bp->ptp_info->get_tsu_rate(bp);
bp->ptp_clock_info.max_adj = bp->ptp_info->get_ptp_max_adj();
gem_ptp_init_timer(bp);
- bp->ptp_clock = ptp_clock_register(&bp->ptp_clock_info, &dev->dev);
+ bp->ptp_clock = netdev_ptp_clock_register(&bp->ptp_clock_info, dev);
if (IS_ERR(bp->ptp_clock)) {
pr_err("ptp clock register failed: %ld\n",
PTR_ERR(bp->ptp_clock));
--
2.34.1
^ permalink raw reply related [flat|nested] 39+ messages in thread
* [PATCH net-next v17 11/14] net: ptp: Move ptp_clock_index() to builtin symbol
2024-07-09 13:53 [PATCH net-next v17 00/14] net: Make timestamping selectable Kory Maincent
` (9 preceding siblings ...)
2024-07-09 13:53 ` [PATCH net-next v17 10/14] net: macb: " Kory Maincent
@ 2024-07-09 13:53 ` Kory Maincent
2024-07-17 17:24 ` Jacob Keller
2024-07-09 13:53 ` [PATCH net-next v17 12/14] net: ethtool: tsinfo: Add support for reading tsinfo for a specific hwtstamp provider Kory Maincent
` (3 subsequent siblings)
14 siblings, 1 reply; 39+ messages in thread
From: Kory Maincent @ 2024-07-09 13:53 UTC (permalink / raw)
To: Florian Fainelli, Broadcom internal kernel review list,
Andrew Lunn, Heiner Kallweit, Russell King, David S. Miller,
Eric Dumazet, Jakub Kicinski, Paolo Abeni, Richard Cochran,
Radu Pirea, Jay Vosburgh, Andy Gospodarek, Nicolas Ferre,
Claudiu Beznea, Willem de Bruijn, Jonathan Corbet, Horatiu Vultur,
UNGLinuxDriver, Simon Horman, Vladimir Oltean, donald.hunter,
danieller, ecree.xilinx
Cc: Thomas Petazzoni, linux-kernel, netdev, linux-doc,
Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter, Kory Maincent
Move ptp_clock_index() to builtin symbols to prepare for supporting get
and set hardware timestamps from ethtool, which is builtin.
Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
---
Change in v13:
- New patch
---
drivers/ptp/ptp_clock.c | 6 ------
drivers/ptp/ptp_clock_consumer.c | 6 ++++++
2 files changed, 6 insertions(+), 6 deletions(-)
diff --git a/drivers/ptp/ptp_clock.c b/drivers/ptp/ptp_clock.c
index 593b5c906314..fc4b266abe1d 100644
--- a/drivers/ptp/ptp_clock.c
+++ b/drivers/ptp/ptp_clock.c
@@ -460,12 +460,6 @@ void ptp_clock_event(struct ptp_clock *ptp, struct ptp_clock_event *event)
}
EXPORT_SYMBOL(ptp_clock_event);
-int ptp_clock_index(struct ptp_clock *ptp)
-{
- return ptp->index;
-}
-EXPORT_SYMBOL(ptp_clock_index);
-
int ptp_find_pin(struct ptp_clock *ptp,
enum ptp_pin_function func, unsigned int chan)
{
diff --git a/drivers/ptp/ptp_clock_consumer.c b/drivers/ptp/ptp_clock_consumer.c
index f5fab1c14b47..f521b07da231 100644
--- a/drivers/ptp/ptp_clock_consumer.c
+++ b/drivers/ptp/ptp_clock_consumer.c
@@ -108,3 +108,9 @@ void remove_hwtstamp_provider(struct rcu_head *rcu_head)
kfree(hwtstamp);
}
EXPORT_SYMBOL(remove_hwtstamp_provider);
+
+int ptp_clock_index(struct ptp_clock *ptp)
+{
+ return ptp->index;
+}
+EXPORT_SYMBOL(ptp_clock_index);
--
2.34.1
^ permalink raw reply related [flat|nested] 39+ messages in thread
* [PATCH net-next v17 12/14] net: ethtool: tsinfo: Add support for reading tsinfo for a specific hwtstamp provider
2024-07-09 13:53 [PATCH net-next v17 00/14] net: Make timestamping selectable Kory Maincent
` (10 preceding siblings ...)
2024-07-09 13:53 ` [PATCH net-next v17 11/14] net: ptp: Move ptp_clock_index() to builtin symbol Kory Maincent
@ 2024-07-09 13:53 ` Kory Maincent
2024-07-17 17:35 ` Jacob Keller
2024-07-09 13:53 ` [PATCH net-next v17 13/14] net: ethtool: Add support for tsconfig command to get/set hwtstamp config Kory Maincent
` (2 subsequent siblings)
14 siblings, 1 reply; 39+ messages in thread
From: Kory Maincent @ 2024-07-09 13:53 UTC (permalink / raw)
To: Florian Fainelli, Broadcom internal kernel review list,
Andrew Lunn, Heiner Kallweit, Russell King, David S. Miller,
Eric Dumazet, Jakub Kicinski, Paolo Abeni, Richard Cochran,
Radu Pirea, Jay Vosburgh, Andy Gospodarek, Nicolas Ferre,
Claudiu Beznea, Willem de Bruijn, Jonathan Corbet, Horatiu Vultur,
UNGLinuxDriver, Simon Horman, Vladimir Oltean, donald.hunter,
danieller, ecree.xilinx
Cc: Thomas Petazzoni, linux-kernel, netdev, linux-doc,
Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter, Kory Maincent
Either the MAC or the PHY can provide hwtstamp, so we should be able to
read the tsinfo for any hwtstamp provider.
Enhance 'get' command to retrieve tsinfo of hwtstamp providers within a
network topology.
Add support for a specific dump command to retrieve all hwtstamp
providers within the network topology, with added functionality for
filtered dump to target a single interface.
Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
---
Pointer attached_dev is used to know if the phy is on the net topology.
This might not be enough and might need Maxime Chevallier link topology
patch series:
https://lore.kernel.org/netdev/20240213150431.1796171-1-maxime.chevallier@bootlin.com/
Change in v8:
- New patch
Change in v9:
- Fix nit in done callback.
Change in v10:
- Add documentation.
- Add extack error messages.
- Add NLA_POLICY_* checks.
- Fix few nits.
- Add ETHTOOL_A_TSINFO_GHWTSTAMP netlink attributes instead of a bit in
ETHTOOL_A_TSINFO_TIMESTAMPING bitset.
- Add tsinfo_parse_hwtstamp_provider function for more readability.
- Move netdev_support_hwtstamp_qualifier function in core ptp instead of
core net.
- Add refcount put to release the ptp object.
- Use rcu lock to avoid memory leak.
Change in v12:
- Add missing return description in the kdoc.
- Fix possible leak due to uninitialised variable.
- Add a missing static.
Change in v13:
- Remove useless EXPORT_SYMBOL().
- Fix issues reported by sparse and smatch.
- Fix issues when building PTP as a module.
- Rename HWTSTAMP_PROVIDER_NEST to HWTSTAMP_PROVIDER.
Change in v16:
- Used call_rcu() instead of synchronize_rcu() to free the hwtstamp_provider
struct.
- Fix documentation typo.
- Fix few nit.
- Fix possible issue regardings stats.
- Remove hwtstamp config get and set from tsinfo.
---
Documentation/networking/ethtool-netlink.rst | 7 +-
drivers/ptp/ptp_clock_consumer.c | 66 ++++++++
include/linux/ethtool.h | 4 +
include/linux/ptp_clock_kernel.h | 53 ++++++
include/uapi/linux/ethtool_netlink.h | 11 +-
net/ethtool/common.c | 32 ++++
net/ethtool/common.h | 3 +
net/ethtool/netlink.c | 6 +-
net/ethtool/netlink.h | 5 +-
net/ethtool/ts.h | 52 ++++++
net/ethtool/tsinfo.c | 239 ++++++++++++++++++++++++++-
11 files changed, 464 insertions(+), 14 deletions(-)
diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst
index 3ab423b80e91..95a1d3576b37 100644
--- a/Documentation/networking/ethtool-netlink.rst
+++ b/Documentation/networking/ethtool-netlink.rst
@@ -1235,9 +1235,10 @@ Gets timestamping information like ``ETHTOOL_GET_TS_INFO`` ioctl request.
Request contents:
- ===================================== ====== ==========================
- ``ETHTOOL_A_TSINFO_HEADER`` nested request header
- ===================================== ====== ==========================
+ ======================================== ====== ============================
+ ``ETHTOOL_A_TSINFO_HEADER`` nested request header
+ ``ETHTOOL_A_TSINFO_HWTSTAMP_PROVIDER`` nested PTP hw clock provider
+ ======================================== ====== ============================
Kernel response contents:
diff --git a/drivers/ptp/ptp_clock_consumer.c b/drivers/ptp/ptp_clock_consumer.c
index f521b07da231..0eec0e910d1a 100644
--- a/drivers/ptp/ptp_clock_consumer.c
+++ b/drivers/ptp/ptp_clock_consumer.c
@@ -114,3 +114,69 @@ int ptp_clock_index(struct ptp_clock *ptp)
return ptp->index;
}
EXPORT_SYMBOL(ptp_clock_index);
+
+struct ptp_clock *netdev_ptp_clock_find(struct net_device *dev,
+ unsigned long *indexp)
+{
+ struct ptp_clock *ptp;
+ unsigned long index;
+
+ xa_for_each_start(&ptp_clocks_map, index, ptp, *indexp) {
+ if ((ptp->phc_source == HWTSTAMP_SOURCE_NETDEV &&
+ ptp->netdev == dev) ||
+ (ptp->phc_source == HWTSTAMP_SOURCE_PHYLIB &&
+ ptp->phydev->attached_dev == dev)) {
+ *indexp = index;
+ return ptp;
+ }
+ }
+
+ return NULL;
+};
+
+bool
+netdev_support_hwtstamp_qualifier(struct net_device *dev,
+ enum hwtstamp_provider_qualifier qualifier)
+{
+ const struct ethtool_ops *ops = dev->ethtool_ops;
+
+ if (!ops)
+ return false;
+
+ /* Return true with precise qualifier and with NIC without
+ * qualifier description to not break the old behavior.
+ */
+ if (!ops->supported_hwtstamp_qualifiers &&
+ qualifier == HWTSTAMP_PROVIDER_QUALIFIER_PRECISE)
+ return true;
+
+ if (ops->supported_hwtstamp_qualifiers & BIT(qualifier))
+ return true;
+
+ return false;
+}
+
+bool netdev_support_hwtstamp(struct net_device *dev,
+ struct hwtstamp_provider *hwtstamp)
+{
+ struct ptp_clock *tmp_ptp;
+ unsigned long index = 0;
+
+ netdev_for_each_ptp_clock_start(dev, index, tmp_ptp, 0) {
+ if (tmp_ptp != hwtstamp->ptp)
+ continue;
+
+ if (ptp_clock_from_phylib(hwtstamp->ptp) &&
+ hwtstamp->qualifier == HWTSTAMP_PROVIDER_QUALIFIER_PRECISE)
+ return true;
+
+ if (ptp_clock_from_netdev(hwtstamp->ptp) &&
+ netdev_support_hwtstamp_qualifier(dev,
+ hwtstamp->qualifier))
+ return true;
+
+ return false;
+ }
+
+ return false;
+}
diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index 6b38bbda5790..f21e54187185 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -719,6 +719,7 @@ struct ethtool_rxfh_param {
* @cmd: command number = %ETHTOOL_GET_TS_INFO
* @so_timestamping: bit mask of the sum of the supported SO_TIMESTAMPING flags
* @phc_index: device index of the associated PHC, or -1 if there is none
+ * @phc_qualifier: qualifier of the associated PHC
* @tx_types: bit mask of the supported hwtstamp_tx_types enumeration values
* @rx_filters: bit mask of the supported hwtstamp_rx_filters enumeration values
*/
@@ -726,6 +727,7 @@ struct kernel_ethtool_ts_info {
u32 cmd;
u32 so_timestamping;
int phc_index;
+ enum hwtstamp_provider_qualifier phc_qualifier;
enum hwtstamp_tx_types tx_types;
enum hwtstamp_rx_filters rx_filters;
};
@@ -746,6 +748,7 @@ struct kernel_ethtool_ts_info {
* argument to @create_rxfh_context and friends.
* @supported_coalesce_params: supported types of interrupt coalescing.
* @supported_ring_params: supported ring params.
+ * @supported_hwtstamp_qualifiers: bitfield of supported hwtstamp qualifier.
* @get_drvinfo: Report driver/device information. Modern drivers no
* longer have to implement this callback. Most fields are
* correctly filled in by the core using system information, or
@@ -959,6 +962,7 @@ struct ethtool_ops {
u32 rxfh_max_context_id;
u32 supported_coalesce_params;
u32 supported_ring_params;
+ u32 supported_hwtstamp_qualifiers;
void (*get_drvinfo)(struct net_device *, struct ethtool_drvinfo *);
int (*get_regs_len)(struct net_device *);
void (*get_regs)(struct net_device *, struct ethtool_regs *, void *);
diff --git a/include/linux/ptp_clock_kernel.h b/include/linux/ptp_clock_kernel.h
index 6d9e6cd5c832..a8babc15f5b3 100644
--- a/include/linux/ptp_clock_kernel.h
+++ b/include/linux/ptp_clock_kernel.h
@@ -450,6 +450,45 @@ void ptp_clock_put(struct device *dev, struct ptp_clock *ptp);
void remove_hwtstamp_provider(struct rcu_head *rcu_head);
+/**
+ * netdev_ptp_clock_find() - obtain the next PTP clock in the netdev
+ * topology
+ *
+ * @dev: Pointer of the net device.
+ * @indexp: Pointer of ptp clock index start point.
+ *
+ * Return: Pointer of the PTP clock found, NULL otherwise.
+ */
+
+struct ptp_clock *netdev_ptp_clock_find(struct net_device *dev,
+ unsigned long *indexp);
+
+/**
+ * netdev_support_hwtstamp_qualifier() - Verify if the net device support the
+ * hwtstamp qualifier
+ *
+ * @dev: Pointer of the net device.
+ * @qualifier: hwtstamp provider qualifier.
+ *
+ * Return: True if the net device support the htstamp qualifier false otherwise.
+ */
+
+bool netdev_support_hwtstamp_qualifier(struct net_device *dev,
+ enum hwtstamp_provider_qualifier qualifier);
+
+/**
+ * netdev_support_hwtstamp() - Verify if the hwtstamp belong to the netdev
+ * topology
+ *
+ * @dev: Pointer of the net device
+ * @hwtstamp: Pointer of the hwtstamp provider
+ *
+ * Return: True if the hwtstamp belong to the netdev topology false otherwise.
+ */
+
+bool netdev_support_hwtstamp(struct net_device *dev,
+ struct hwtstamp_provider *hwtstamp);
+
/**
* ptp_find_pin() - obtain the pin index of a given auxiliary function
*
@@ -538,6 +577,16 @@ static inline void ptp_clock_put(struct device *dev, struct ptp_clock *ptp)
static inline struct ptp_clock *ptp_clock_get_by_index(struct device *dev,
int index)
{ return NULL; }
+static inline struct ptp_clock *netdev_ptp_clock_find(struct net_device *dev,
+ unsigned long *indexp)
+{ return NULL; }
+static inline bool
+netdev_support_hwtstamp_qualifier(struct net_device *dev,
+ enum hwtstamp_provider_qualifier qualifier)
+{ return false; }
+static inline bool netdev_support_hwtstamp(struct net_device *dev,
+ struct hwtstamp_provider *hwtst)
+{ return false; }
static inline void remove_hwtstamp_provider(struct rcu_head *rcu_head)
{ return; }
static inline int ptp_find_pin(struct ptp_clock *ptp,
@@ -554,6 +603,10 @@ static inline void ptp_cancel_worker_sync(struct ptp_clock *ptp)
{ }
#endif
+#define netdev_for_each_ptp_clock_start(dev, index, entry, start) \
+ for (index = start, entry = netdev_ptp_clock_find(dev, &index); \
+ entry; index++, entry = netdev_ptp_clock_find(dev, &index))
+
#if IS_BUILTIN(CONFIG_PTP_1588_CLOCK)
/*
* These are called by the network core, and don't work if PTP is in
diff --git a/include/uapi/linux/ethtool_netlink.h b/include/uapi/linux/ethtool_netlink.h
index 6d5bdcc67631..0660e5a58b34 100644
--- a/include/uapi/linux/ethtool_netlink.h
+++ b/include/uapi/linux/ethtool_netlink.h
@@ -492,8 +492,16 @@ enum {
ETHTOOL_A_EEE_MAX = (__ETHTOOL_A_EEE_CNT - 1)
};
-/* TSINFO */
+enum {
+ ETHTOOL_A_TS_HWTSTAMP_PROVIDER_UNSPEC,
+ ETHTOOL_A_TS_HWTSTAMP_PROVIDER_INDEX, /* u32 */
+ ETHTOOL_A_TS_HWTSTAMP_PROVIDER_QUALIFIER, /* u32 */
+ __ETHTOOL_A_TS_HWTSTAMP_PROVIDER_CNT,
+ ETHTOOL_A_TS_HWTSTAMP_PROVIDER_MAX = (__ETHTOOL_A_TS_HWTSTAMP_PROVIDER_CNT - 1)
+};
+
+/* TSINFO */
enum {
ETHTOOL_A_TSINFO_UNSPEC,
ETHTOOL_A_TSINFO_HEADER, /* nest - _A_HEADER_* */
@@ -502,6 +510,7 @@ enum {
ETHTOOL_A_TSINFO_RX_FILTERS, /* bitset */
ETHTOOL_A_TSINFO_PHC_INDEX, /* u32 */
ETHTOOL_A_TSINFO_STATS, /* nest - _A_TSINFO_STAT */
+ ETHTOOL_A_TSINFO_HWTSTAMP_PROVIDER, /* nest - *_TS_HWTSTAMP_PROVIDER_* */
/* add new constants above here */
__ETHTOOL_A_TSINFO_CNT,
diff --git a/net/ethtool/common.c b/net/ethtool/common.c
index 461017a37955..532332e708fb 100644
--- a/net/ethtool/common.c
+++ b/net/ethtool/common.c
@@ -629,10 +629,42 @@ int ethtool_check_ops(const struct ethtool_ops *ops)
return 0;
}
+int ethtool_get_ts_info_by_phc(struct net_device *dev,
+ struct kernel_ethtool_ts_info *info,
+ struct hwtstamp_provider *hwtstamp)
+{
+ const struct ethtool_ops *ops = dev->ethtool_ops;
+
+ memset(info, 0, sizeof(*info));
+ info->cmd = ETHTOOL_GET_TS_INFO;
+ info->phc_qualifier = hwtstamp->qualifier;
+
+ if (!netdev_support_hwtstamp(dev, hwtstamp))
+ return -ENODEV;
+
+ if (ptp_clock_from_phylib(hwtstamp->ptp) &&
+ phy_has_tsinfo(ptp_clock_phydev(hwtstamp->ptp)))
+ return phy_ts_info(ptp_clock_phydev(hwtstamp->ptp), info);
+
+ if (ptp_clock_from_netdev(hwtstamp->ptp) && ops->get_ts_info)
+ return ops->get_ts_info(dev, info);
+
+ info->so_timestamping = SOF_TIMESTAMPING_RX_SOFTWARE |
+ SOF_TIMESTAMPING_SOFTWARE;
+ info->phc_index = -1;
+
+ return 0;
+}
+
int __ethtool_get_ts_info(struct net_device *dev, struct kernel_ethtool_ts_info *info)
{
const struct ethtool_ops *ops = dev->ethtool_ops;
struct phy_device *phydev = dev->phydev;
+ struct hwtstamp_provider *hwtstamp;
+
+ hwtstamp = rtnl_dereference(dev->hwtstamp);
+ if (hwtstamp)
+ return ethtool_get_ts_info_by_phc(dev, info, hwtstamp);
memset(info, 0, sizeof(*info));
info->cmd = ETHTOOL_GET_TS_INFO;
diff --git a/net/ethtool/common.h b/net/ethtool/common.h
index b9daeecbd84d..83bf82a4e75a 100644
--- a/net/ethtool/common.h
+++ b/net/ethtool/common.h
@@ -45,6 +45,9 @@ bool convert_legacy_settings_to_link_ksettings(
int ethtool_get_max_rxfh_channel(struct net_device *dev, u32 *max);
int ethtool_get_max_rxnfc_channel(struct net_device *dev, u64 *max);
int __ethtool_get_ts_info(struct net_device *dev, struct kernel_ethtool_ts_info *info);
+int ethtool_get_ts_info_by_phc(struct net_device *dev,
+ struct kernel_ethtool_ts_info *info,
+ struct hwtstamp_provider *hwtst);
extern const struct ethtool_phy_ops *ethtool_phy_ops;
extern const struct ethtool_pse_ops *ethtool_pse_ops;
diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c
index cb1eea00e349..8ebc9afdb245 100644
--- a/net/ethtool/netlink.c
+++ b/net/ethtool/netlink.c
@@ -1021,9 +1021,9 @@ static const struct genl_ops ethtool_genl_ops[] = {
{
.cmd = ETHTOOL_MSG_TSINFO_GET,
.doit = ethnl_default_doit,
- .start = ethnl_default_start,
- .dumpit = ethnl_default_dumpit,
- .done = ethnl_default_done,
+ .start = ethnl_tsinfo_start,
+ .dumpit = ethnl_tsinfo_dumpit,
+ .done = ethnl_tsinfo_done,
.policy = ethnl_tsinfo_get_policy,
.maxattr = ARRAY_SIZE(ethnl_tsinfo_get_policy) - 1,
},
diff --git a/net/ethtool/netlink.h b/net/ethtool/netlink.h
index 46ec273a87c5..0ebe42dc1bf0 100644
--- a/net/ethtool/netlink.h
+++ b/net/ethtool/netlink.h
@@ -436,7 +436,7 @@ extern const struct nla_policy ethnl_pause_get_policy[ETHTOOL_A_PAUSE_STATS_SRC
extern const struct nla_policy ethnl_pause_set_policy[ETHTOOL_A_PAUSE_TX + 1];
extern const struct nla_policy ethnl_eee_get_policy[ETHTOOL_A_EEE_HEADER + 1];
extern const struct nla_policy ethnl_eee_set_policy[ETHTOOL_A_EEE_TX_LPI_TIMER + 1];
-extern const struct nla_policy ethnl_tsinfo_get_policy[ETHTOOL_A_TSINFO_HEADER + 1];
+extern const struct nla_policy ethnl_tsinfo_get_policy[ETHTOOL_A_TSINFO_MAX + 1];
extern const struct nla_policy ethnl_cable_test_act_policy[ETHTOOL_A_CABLE_TEST_HEADER + 1];
extern const struct nla_policy ethnl_cable_test_tdr_act_policy[ETHTOOL_A_CABLE_TEST_TDR_CFG + 1];
extern const struct nla_policy ethnl_tunnel_info_get_policy[ETHTOOL_A_TUNNEL_INFO_HEADER + 1];
@@ -464,6 +464,9 @@ int ethnl_tunnel_info_doit(struct sk_buff *skb, struct genl_info *info);
int ethnl_tunnel_info_start(struct netlink_callback *cb);
int ethnl_tunnel_info_dumpit(struct sk_buff *skb, struct netlink_callback *cb);
int ethnl_act_module_fw_flash(struct sk_buff *skb, struct genl_info *info);
+int ethnl_tsinfo_start(struct netlink_callback *cb);
+int ethnl_tsinfo_dumpit(struct sk_buff *skb, struct netlink_callback *cb);
+int ethnl_tsinfo_done(struct netlink_callback *cb);
extern const char stats_std_names[__ETHTOOL_STATS_CNT][ETH_GSTRING_LEN];
extern const char stats_eth_phy_names[__ETHTOOL_A_STATS_ETH_PHY_CNT][ETH_GSTRING_LEN];
diff --git a/net/ethtool/ts.h b/net/ethtool/ts.h
new file mode 100644
index 000000000000..1fb7b6d9d99a
--- /dev/null
+++ b/net/ethtool/ts.h
@@ -0,0 +1,52 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+
+#ifndef _NET_ETHTOOL_TS_H
+#define _NET_ETHTOOL_TS_H
+
+#include "netlink.h"
+
+struct hwtst_provider {
+ int index;
+ u32 qualifier;
+};
+
+static const struct nla_policy
+ethnl_ts_hwtst_prov_policy[ETHTOOL_A_TS_HWTSTAMP_PROVIDER_MAX + 1] = {
+ [ETHTOOL_A_TS_HWTSTAMP_PROVIDER_INDEX] =
+ NLA_POLICY_MIN(NLA_S32, 0),
+ [ETHTOOL_A_TS_HWTSTAMP_PROVIDER_QUALIFIER] =
+ NLA_POLICY_MAX(NLA_U32, HWTSTAMP_PROVIDER_QUALIFIER_CNT - 1)
+};
+
+static inline int ts_parse_hwtst_provider(const struct nlattr *nest,
+ struct hwtst_provider *hwtst,
+ struct netlink_ext_ack *extack,
+ bool *mod)
+{
+ struct nlattr *tb[ARRAY_SIZE(ethnl_ts_hwtst_prov_policy)];
+ int ret;
+
+ ret = nla_parse_nested(tb,
+ ARRAY_SIZE(ethnl_ts_hwtst_prov_policy) - 1,
+ nest,
+ ethnl_ts_hwtst_prov_policy, extack);
+ if (ret < 0)
+ return ret;
+
+ if (NL_REQ_ATTR_CHECK(extack, nest, tb,
+ ETHTOOL_A_TS_HWTSTAMP_PROVIDER_INDEX) ||
+ NL_REQ_ATTR_CHECK(extack, nest, tb,
+ ETHTOOL_A_TS_HWTSTAMP_PROVIDER_QUALIFIER))
+ return -EINVAL;
+
+ ethnl_update_u32(&hwtst->index,
+ tb[ETHTOOL_A_TS_HWTSTAMP_PROVIDER_INDEX],
+ mod);
+ ethnl_update_u32(&hwtst->qualifier,
+ tb[ETHTOOL_A_TS_HWTSTAMP_PROVIDER_QUALIFIER],
+ mod);
+
+ return 0;
+}
+
+#endif /* _NET_ETHTOOL_TS_H */
diff --git a/net/ethtool/tsinfo.c b/net/ethtool/tsinfo.c
index 03d12d6f79ca..1ff992fcaf6a 100644
--- a/net/ethtool/tsinfo.c
+++ b/net/ethtool/tsinfo.c
@@ -1,13 +1,16 @@
// SPDX-License-Identifier: GPL-2.0-only
#include <linux/net_tstamp.h>
+#include <linux/ptp_clock_kernel.h>
#include "netlink.h"
#include "common.h"
#include "bitset.h"
+#include "ts.h"
struct tsinfo_req_info {
struct ethnl_req_info base;
+ struct hwtst_provider hwtst;
};
struct tsinfo_reply_data {
@@ -16,35 +19,76 @@ struct tsinfo_reply_data {
struct ethtool_ts_stats stats;
};
+#define TSINFO_REQINFO(__req_base) \
+ container_of(__req_base, struct tsinfo_req_info, base)
+
#define TSINFO_REPDATA(__reply_base) \
container_of(__reply_base, struct tsinfo_reply_data, base)
#define ETHTOOL_TS_STAT_CNT \
(__ETHTOOL_A_TS_STAT_CNT - (ETHTOOL_A_TS_STAT_UNSPEC + 1))
-const struct nla_policy ethnl_tsinfo_get_policy[] = {
+const struct nla_policy ethnl_tsinfo_get_policy[ETHTOOL_A_TSINFO_MAX + 1] = {
[ETHTOOL_A_TSINFO_HEADER] =
NLA_POLICY_NESTED(ethnl_header_policy_stats),
+ [ETHTOOL_A_TSINFO_HWTSTAMP_PROVIDER] =
+ NLA_POLICY_NESTED(ethnl_ts_hwtst_prov_policy),
};
+static int
+tsinfo_parse_request(struct ethnl_req_info *req_base, struct nlattr **tb,
+ struct netlink_ext_ack *extack)
+{
+ struct tsinfo_req_info *req = TSINFO_REQINFO(req_base);
+ bool mod = false;
+
+ req->hwtst.index = -1;
+
+ if (!tb[ETHTOOL_A_TSINFO_HWTSTAMP_PROVIDER])
+ return 0;
+
+ return ts_parse_hwtst_provider(tb[ETHTOOL_A_TSINFO_HWTSTAMP_PROVIDER],
+ &req->hwtst, extack, &mod);
+}
+
static int tsinfo_prepare_data(const struct ethnl_req_info *req_base,
struct ethnl_reply_data *reply_base,
const struct genl_info *info)
{
struct tsinfo_reply_data *data = TSINFO_REPDATA(reply_base);
+ struct tsinfo_req_info *req = TSINFO_REQINFO(req_base);
struct net_device *dev = reply_base->dev;
int ret;
ret = ethnl_ops_begin(dev);
if (ret < 0)
return ret;
+
if (req_base->flags & ETHTOOL_FLAG_STATS) {
ethtool_stats_init((u64 *)&data->stats,
sizeof(data->stats) / sizeof(u64));
if (dev->ethtool_ops->get_ts_stats)
dev->ethtool_ops->get_ts_stats(dev, &data->stats);
}
- ret = __ethtool_get_ts_info(dev, &data->ts_info);
+
+ if (req->hwtst.index != -1) {
+ struct hwtstamp_provider hwtstamp;
+
+ hwtstamp.ptp = ptp_clock_get_by_index(&dev->dev, req->hwtst.index);
+ if (!hwtstamp.ptp) {
+ ret = -ENODEV;
+ goto out;
+ }
+ hwtstamp.qualifier = req->hwtst.qualifier;
+
+ ret = ethtool_get_ts_info_by_phc(dev, &data->ts_info,
+ &hwtstamp);
+ ptp_clock_put(&dev->dev, hwtstamp.ptp);
+ } else {
+ ret = __ethtool_get_ts_info(dev, &data->ts_info);
+ }
+
+out:
ethnl_ops_complete(dev);
return ret;
@@ -87,8 +131,11 @@ static int tsinfo_reply_size(const struct ethnl_req_info *req_base,
return ret;
len += ret; /* _TSINFO_RX_FILTERS */
}
- if (ts_info->phc_index >= 0)
+ if (ts_info->phc_index >= 0) {
+ /* _TSINFO_HWTSTAMP_PROVIDER */
+ len += 2 * nla_total_size(sizeof(u32));
len += nla_total_size(sizeof(u32)); /* _TSINFO_PHC_INDEX */
+ }
if (req_base->flags & ETHTOOL_FLAG_STATS)
len += nla_total_size(0) + /* _TSINFO_STATS */
nla_total_size_64bit(sizeof(u64)) * ETHTOOL_TS_STAT_CNT;
@@ -163,9 +210,29 @@ static int tsinfo_fill_reply(struct sk_buff *skb,
if (ret < 0)
return ret;
}
- if (ts_info->phc_index >= 0 &&
- nla_put_u32(skb, ETHTOOL_A_TSINFO_PHC_INDEX, ts_info->phc_index))
- return -EMSGSIZE;
+ if (ts_info->phc_index >= 0) {
+ struct nlattr *nest;
+
+ ret = nla_put_u32(skb, ETHTOOL_A_TSINFO_PHC_INDEX,
+ ts_info->phc_index);
+ if (ret)
+ return -EMSGSIZE;
+
+ nest = nla_nest_start(skb, ETHTOOL_A_TSINFO_HWTSTAMP_PROVIDER);
+ if (!nest)
+ return -EMSGSIZE;
+
+ if (nla_put_u32(skb, ETHTOOL_A_TS_HWTSTAMP_PROVIDER_INDEX,
+ ts_info->phc_index) ||
+ nla_put_u32(skb,
+ ETHTOOL_A_TS_HWTSTAMP_PROVIDER_QUALIFIER,
+ ts_info->phc_qualifier)) {
+ nla_nest_cancel(skb, nest);
+ return -EMSGSIZE;
+ }
+
+ nla_nest_end(skb, nest);
+ }
if (req_base->flags & ETHTOOL_FLAG_STATS &&
tsinfo_put_stats(skb, &data->stats))
return -EMSGSIZE;
@@ -173,6 +240,165 @@ static int tsinfo_fill_reply(struct sk_buff *skb,
return 0;
}
+struct ethnl_tsinfo_dump_ctx {
+ struct tsinfo_req_info *req_info;
+ struct tsinfo_reply_data *reply_data;
+ unsigned long pos_ifindex;
+ unsigned long pos_phcindex;
+ enum hwtstamp_provider_qualifier pos_phcqualifier;
+};
+
+static int ethnl_tsinfo_dump_one_ptp(struct sk_buff *skb, struct net_device *dev,
+ struct netlink_callback *cb,
+ struct ptp_clock *ptp)
+{
+ struct ethnl_tsinfo_dump_ctx *ctx = (void *)cb->ctx;
+ struct tsinfo_reply_data *reply_data;
+ struct tsinfo_req_info *req_info;
+ void *ehdr = NULL;
+ int ret = 0;
+
+ reply_data = ctx->reply_data;
+ req_info = ctx->req_info;
+ req_info->hwtst.index = ptp_clock_index(ptp);
+
+ for (; ctx->pos_phcqualifier < HWTSTAMP_PROVIDER_QUALIFIER_CNT;
+ ctx->pos_phcqualifier++) {
+ if (!netdev_support_hwtstamp_qualifier(dev,
+ ctx->pos_phcqualifier))
+ continue;
+
+ ehdr = ethnl_dump_put(skb, cb,
+ ETHTOOL_MSG_TSINFO_GET_REPLY);
+ if (!ehdr)
+ return -EMSGSIZE;
+
+ memset(reply_data, 0, sizeof(*reply_data));
+ reply_data->base.dev = dev;
+ req_info->hwtst.qualifier = ctx->pos_phcqualifier;
+ ret = tsinfo_prepare_data(&req_info->base,
+ &reply_data->base,
+ genl_info_dump(cb));
+ if (ret < 0)
+ break;
+
+ ret = ethnl_fill_reply_header(skb, dev,
+ ETHTOOL_A_TSINFO_HEADER);
+ if (ret < 0)
+ break;
+
+ ret = tsinfo_fill_reply(skb, &req_info->base,
+ &reply_data->base);
+ if (ret < 0)
+ break;
+ }
+
+ reply_data->base.dev = NULL;
+ if (!ret && ehdr)
+ genlmsg_end(skb, ehdr);
+ else
+ genlmsg_cancel(skb, ehdr);
+ return ret;
+}
+
+static int ethnl_tsinfo_dump_one_dev(struct sk_buff *skb, struct net_device *dev,
+ struct netlink_callback *cb)
+{
+ struct ethnl_tsinfo_dump_ctx *ctx = (void *)cb->ctx;
+ struct ptp_clock *ptp;
+ int ret = 0;
+
+ netdev_for_each_ptp_clock_start(dev, ctx->pos_phcindex, ptp,
+ ctx->pos_phcindex) {
+ ret = ethnl_tsinfo_dump_one_ptp(skb, dev, cb, ptp);
+ if (ret < 0 && ret != -EOPNOTSUPP)
+ break;
+ ctx->pos_phcqualifier = HWTSTAMP_PROVIDER_QUALIFIER_PRECISE;
+ }
+
+ return ret;
+}
+
+int ethnl_tsinfo_dumpit(struct sk_buff *skb, struct netlink_callback *cb)
+{
+ struct ethnl_tsinfo_dump_ctx *ctx = (void *)cb->ctx;
+ struct net *net = sock_net(skb->sk);
+ struct net_device *dev;
+ int ret = 0;
+
+ rtnl_lock();
+ if (ctx->req_info->base.dev) {
+ ret = ethnl_tsinfo_dump_one_dev(skb,
+ ctx->req_info->base.dev,
+ cb);
+ } else {
+ for_each_netdev_dump(net, dev, ctx->pos_ifindex) {
+ ret = ethnl_tsinfo_dump_one_dev(skb, dev, cb);
+ if (ret < 0 && ret != -EOPNOTSUPP)
+ break;
+ ctx->pos_phcindex = 0;
+ }
+ }
+ rtnl_unlock();
+
+ return ret;
+}
+
+int ethnl_tsinfo_start(struct netlink_callback *cb)
+{
+ const struct genl_dumpit_info *info = genl_dumpit_info(cb);
+ struct ethnl_tsinfo_dump_ctx *ctx = (void *)cb->ctx;
+ struct nlattr **tb = info->info.attrs;
+ struct tsinfo_reply_data *reply_data;
+ struct tsinfo_req_info *req_info;
+ int ret;
+
+ BUILD_BUG_ON(sizeof(*ctx) > sizeof(cb->ctx));
+
+ req_info = kzalloc(sizeof(*req_info), GFP_KERNEL);
+ if (!req_info)
+ return -ENOMEM;
+ reply_data = kzalloc(sizeof(*reply_data), GFP_KERNEL);
+ if (!reply_data) {
+ ret = -ENOMEM;
+ goto free_req_info;
+ }
+
+ ret = ethnl_parse_header_dev_get(&req_info->base,
+ tb[ETHTOOL_A_TSINFO_HEADER],
+ sock_net(cb->skb->sk), cb->extack,
+ false);
+ if (ret < 0)
+ goto free_reply_data;
+
+ ctx->req_info = req_info;
+ ctx->reply_data = reply_data;
+ ctx->pos_ifindex = 0;
+ ctx->pos_phcindex = 0;
+ ctx->pos_phcqualifier = HWTSTAMP_PROVIDER_QUALIFIER_PRECISE;
+
+ return 0;
+
+free_reply_data:
+ kfree(reply_data);
+free_req_info:
+ kfree(req_info);
+
+ return ret;
+}
+
+int ethnl_tsinfo_done(struct netlink_callback *cb)
+{
+ struct ethnl_tsinfo_dump_ctx *ctx = (void *)cb->ctx;
+ struct tsinfo_req_info *req_info = ctx->req_info;
+
+ ethnl_parse_header_dev_put(&req_info->base);
+ kfree(ctx->reply_data);
+ kfree(ctx->req_info);
+
+ return 0;
+}
+
const struct ethnl_request_ops ethnl_tsinfo_request_ops = {
.request_cmd = ETHTOOL_MSG_TSINFO_GET,
.reply_cmd = ETHTOOL_MSG_TSINFO_GET_REPLY,
@@ -180,6 +406,7 @@ const struct ethnl_request_ops ethnl_tsinfo_request_ops = {
.req_info_size = sizeof(struct tsinfo_req_info),
.reply_data_size = sizeof(struct tsinfo_reply_data),
+ .parse_request = tsinfo_parse_request,
.prepare_data = tsinfo_prepare_data,
.reply_size = tsinfo_reply_size,
.fill_reply = tsinfo_fill_reply,
--
2.34.1
^ permalink raw reply related [flat|nested] 39+ messages in thread
* [PATCH net-next v17 13/14] net: ethtool: Add support for tsconfig command to get/set hwtstamp config
2024-07-09 13:53 [PATCH net-next v17 00/14] net: Make timestamping selectable Kory Maincent
` (11 preceding siblings ...)
2024-07-09 13:53 ` [PATCH net-next v17 12/14] net: ethtool: tsinfo: Add support for reading tsinfo for a specific hwtstamp provider Kory Maincent
@ 2024-07-09 13:53 ` Kory Maincent
2024-07-15 14:59 ` Jakub Kicinski
2024-07-17 17:43 ` Jacob Keller
2024-07-09 13:53 ` [PATCH net-next v17 14/14] netlink: specs: Enhance tsinfo netlink attributes and add a tsconfig set command Kory Maincent
2024-07-15 15:30 ` [PATCH net-next v17 00/14] net: Make timestamping selectable patchwork-bot+netdevbpf
14 siblings, 2 replies; 39+ messages in thread
From: Kory Maincent @ 2024-07-09 13:53 UTC (permalink / raw)
To: Florian Fainelli, Broadcom internal kernel review list,
Andrew Lunn, Heiner Kallweit, Russell King, David S. Miller,
Eric Dumazet, Jakub Kicinski, Paolo Abeni, Richard Cochran,
Radu Pirea, Jay Vosburgh, Andy Gospodarek, Nicolas Ferre,
Claudiu Beznea, Willem de Bruijn, Jonathan Corbet, Horatiu Vultur,
UNGLinuxDriver, Simon Horman, Vladimir Oltean, donald.hunter,
danieller, ecree.xilinx
Cc: Thomas Petazzoni, linux-kernel, netdev, linux-doc,
Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter, Kory Maincent
Introduce support for ETHTOOL_MSG_TSCONFIG_GET/SET ethtool netlink socket
to read and configure hwtstamp configuration of a PHC provider. Note that
simultaneous hwtstamp isn't supported; configuring a new one disables the
previous setting.
Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
---
Changes in v16:
- Add a new patch to separate tsinfo into a new tsconfig command to get
and set the hwtstamp config.
Changes in v17:
- Fix a doc misalignment.
---
Documentation/networking/timestamping.rst | 33 +--
include/uapi/linux/ethtool_netlink.h | 18 ++
net/ethtool/Makefile | 3 +-
net/ethtool/netlink.c | 20 ++
net/ethtool/netlink.h | 3 +
net/ethtool/tsconfig.c | 347 ++++++++++++++++++++++++++++++
6 files changed, 411 insertions(+), 13 deletions(-)
diff --git a/Documentation/networking/timestamping.rst b/Documentation/networking/timestamping.rst
index 5e93cd71f99f..fccf8656adba 100644
--- a/Documentation/networking/timestamping.rst
+++ b/Documentation/networking/timestamping.rst
@@ -493,8 +493,8 @@ implicitly defined. ts[0] holds a software timestamp if set, ts[1]
is again deprecated and ts[2] holds a hardware timestamp if set.
-3. Hardware Timestamping configuration: SIOCSHWTSTAMP and SIOCGHWTSTAMP
-=======================================================================
+3. Hardware Timestamping configuration: ETHTOOL_MSG_TSCONFIG_SET/GET
+====================================================================
Hardware time stamping must also be initialized for each device driver
that is expected to do hardware time stamping. The parameter is defined in
@@ -507,10 +507,15 @@ include/uapi/linux/net_tstamp.h as::
};
Desired behavior is passed into the kernel and to a specific device by
-calling ioctl(SIOCSHWTSTAMP) with a pointer to a struct ifreq whose
-ifr_data points to a struct hwtstamp_config. The tx_type and
-rx_filter are hints to the driver what it is expected to do. If
-the requested fine-grained filtering for incoming packets is not
+calling the tsconfig netlink socket ETHTOOL_MSG_TSCONFIG_SET.
+The ``ETHTOOL_A_TSCONFIG_TX_TYPES``, ``ETHTOOL_A_TSCONFIG_RX_FILTERS`` and
+``ETHTOOL_A_TSCONFIG_HWTSTAMP_FLAGS`` netlink attributes are then used to set
+the struct hwtstamp_config accordingly.
+
+The legacy configuration is the use of the ioctl(SIOCSHWTSTAMP) with a pointer
+to a struct ifreq whose ifr_data points to a struct hwtstamp_config.
+The tx_type and rx_filter are hints to the driver what it is expected to do.
+If the requested fine-grained filtering for incoming packets is not
supported, the driver may time stamp more than just the requested types
of packets.
@@ -531,9 +536,12 @@ Only a processes with admin rights may change the configuration. User
space is responsible to ensure that multiple processes don't interfere
with each other and that the settings are reset.
-Any process can read the actual configuration by passing this
-structure to ioctl(SIOCGHWTSTAMP) in the same way. However, this has
-not been implemented in all drivers.
+Any process can read the actual configuration by requesting tsconfig netlink
+socket ETHTOOL_MSG_TSCONFIG_GET.
+
+The legacy usage is to pass this structure to ioctl(SIOCGHWTSTAMP) in the
+same way as the ioctl(SIOCSHWTSTAMP). However, this has not been implemented
+in all drivers.
::
@@ -578,9 +586,10 @@ not been implemented in all drivers.
--------------------------------------------------------
A driver which supports hardware time stamping must support the
-SIOCSHWTSTAMP ioctl and update the supplied struct hwtstamp_config with
-the actual values as described in the section on SIOCSHWTSTAMP. It
-should also support SIOCGHWTSTAMP.
+ndo_hwtstamp_set NDO or the legacy SIOCSHWTSTAMP ioctl and update the
+supplied struct hwtstamp_config with the actual values as described in
+the section on SIOCSHWTSTAMP. It should also support ndo_hwtstamp_get or
+the legacy SIOCGHWTSTAMP.
Time stamps for received packets must be stored in the skb. To get a pointer
to the shared time stamp structure of the skb call skb_hwtstamps(). Then
diff --git a/include/uapi/linux/ethtool_netlink.h b/include/uapi/linux/ethtool_netlink.h
index 0660e5a58b34..b2b04c0c9535 100644
--- a/include/uapi/linux/ethtool_netlink.h
+++ b/include/uapi/linux/ethtool_netlink.h
@@ -58,6 +58,8 @@ enum {
ETHTOOL_MSG_MM_GET,
ETHTOOL_MSG_MM_SET,
ETHTOOL_MSG_MODULE_FW_FLASH_ACT,
+ ETHTOOL_MSG_TSCONFIG_GET,
+ ETHTOOL_MSG_TSCONFIG_SET,
/* add new constants above here */
__ETHTOOL_MSG_USER_CNT,
@@ -111,6 +113,8 @@ enum {
ETHTOOL_MSG_MM_GET_REPLY,
ETHTOOL_MSG_MM_NTF,
ETHTOOL_MSG_MODULE_FW_FLASH_NTF,
+ ETHTOOL_MSG_TSCONFIG_GET_REPLY,
+ ETHTOOL_MSG_TSCONFIG_NTF,
/* add new constants above here */
__ETHTOOL_MSG_KERNEL_CNT,
@@ -530,6 +534,20 @@ enum {
};
+/* TSCONFIG */
+enum {
+ ETHTOOL_A_TSCONFIG_UNSPEC,
+ ETHTOOL_A_TSCONFIG_HEADER, /* nest - _A_HEADER_* */
+ ETHTOOL_A_TSCONFIG_HWTSTAMP_PROVIDER, /* nest - *_TS_HWTSTAMP_PROVIDER_* */
+ ETHTOOL_A_TSCONFIG_TX_TYPES, /* bitset */
+ ETHTOOL_A_TSCONFIG_RX_FILTERS, /* bitset */
+ ETHTOOL_A_TSCONFIG_HWTSTAMP_FLAGS, /* u32 */
+
+ /* add new constants above here */
+ __ETHTOOL_A_TSCONFIG_CNT,
+ ETHTOOL_A_TSCONFIG_MAX = (__ETHTOOL_A_TSCONFIG_CNT - 1)
+};
+
/* PHC VCLOCKS */
enum {
diff --git a/net/ethtool/Makefile b/net/ethtool/Makefile
index 9a190635fe95..9b6d86b10c58 100644
--- a/net/ethtool/Makefile
+++ b/net/ethtool/Makefile
@@ -8,4 +8,5 @@ ethtool_nl-y := netlink.o bitset.o strset.o linkinfo.o linkmodes.o rss.o \
linkstate.o debug.o wol.o features.o privflags.o rings.o \
channels.o coalesce.o pause.o eee.o tsinfo.o cabletest.o \
tunnels.o fec.o eeprom.o stats.o phc_vclocks.o mm.o \
- module.o cmis_fw_update.o cmis_cdb.o pse-pd.o plca.o mm.o
+ module.o cmis_fw_update.o cmis_cdb.o pse-pd.o plca.o \
+ tsconfig.o
diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c
index 8ebc9afdb245..f9a9cceaf01d 100644
--- a/net/ethtool/netlink.c
+++ b/net/ethtool/netlink.c
@@ -341,6 +341,8 @@ ethnl_default_requests[__ETHTOOL_MSG_USER_CNT] = {
[ETHTOOL_MSG_PLCA_GET_STATUS] = ðnl_plca_status_request_ops,
[ETHTOOL_MSG_MM_GET] = ðnl_mm_request_ops,
[ETHTOOL_MSG_MM_SET] = ðnl_mm_request_ops,
+ [ETHTOOL_MSG_TSCONFIG_GET] = ðnl_tsconfig_request_ops,
+ [ETHTOOL_MSG_TSCONFIG_SET] = ðnl_tsconfig_request_ops,
};
static struct ethnl_dump_ctx *ethnl_dump_context(struct netlink_callback *cb)
@@ -670,6 +672,7 @@ ethnl_default_notify_ops[ETHTOOL_MSG_KERNEL_MAX + 1] = {
[ETHTOOL_MSG_MODULE_NTF] = ðnl_module_request_ops,
[ETHTOOL_MSG_PLCA_NTF] = ðnl_plca_cfg_request_ops,
[ETHTOOL_MSG_MM_NTF] = ðnl_mm_request_ops,
+ [ETHTOOL_MSG_TSCONFIG_NTF] = ðnl_tsconfig_request_ops,
};
/* default notification handler */
@@ -768,6 +771,7 @@ static const ethnl_notify_handler_t ethnl_notify_handlers[] = {
[ETHTOOL_MSG_MODULE_NTF] = ethnl_default_notify,
[ETHTOOL_MSG_PLCA_NTF] = ethnl_default_notify,
[ETHTOOL_MSG_MM_NTF] = ethnl_default_notify,
+ [ETHTOOL_MSG_TSCONFIG_NTF] = ethnl_default_notify,
};
void ethtool_notify(struct net_device *dev, unsigned int cmd, const void *data)
@@ -1179,6 +1183,22 @@ static const struct genl_ops ethtool_genl_ops[] = {
.policy = ethnl_module_fw_flash_act_policy,
.maxattr = ARRAY_SIZE(ethnl_module_fw_flash_act_policy) - 1,
},
+ {
+ .cmd = ETHTOOL_MSG_TSCONFIG_GET,
+ .doit = ethnl_default_doit,
+ .start = ethnl_default_start,
+ .dumpit = ethnl_default_dumpit,
+ .done = ethnl_default_done,
+ .policy = ethnl_tsconfig_get_policy,
+ .maxattr = ARRAY_SIZE(ethnl_tsconfig_get_policy) - 1,
+ },
+ {
+ .cmd = ETHTOOL_MSG_TSCONFIG_SET,
+ .flags = GENL_UNS_ADMIN_PERM,
+ .doit = ethnl_default_set_doit,
+ .policy = ethnl_tsconfig_set_policy,
+ .maxattr = ARRAY_SIZE(ethnl_tsconfig_set_policy) - 1,
+ },
};
static const struct genl_multicast_group ethtool_nl_mcgrps[] = {
diff --git a/net/ethtool/netlink.h b/net/ethtool/netlink.h
index 0ebe42dc1bf0..a616f5deff7f 100644
--- a/net/ethtool/netlink.h
+++ b/net/ethtool/netlink.h
@@ -409,6 +409,7 @@ extern const struct ethnl_request_ops ethnl_rss_request_ops;
extern const struct ethnl_request_ops ethnl_plca_cfg_request_ops;
extern const struct ethnl_request_ops ethnl_plca_status_request_ops;
extern const struct ethnl_request_ops ethnl_mm_request_ops;
+extern const struct ethnl_request_ops ethnl_tsconfig_request_ops;
extern const struct nla_policy ethnl_header_policy[ETHTOOL_A_HEADER_FLAGS + 1];
extern const struct nla_policy ethnl_header_policy_stats[ETHTOOL_A_HEADER_FLAGS + 1];
@@ -456,6 +457,8 @@ extern const struct nla_policy ethnl_plca_get_status_policy[ETHTOOL_A_PLCA_HEADE
extern const struct nla_policy ethnl_mm_get_policy[ETHTOOL_A_MM_HEADER + 1];
extern const struct nla_policy ethnl_mm_set_policy[ETHTOOL_A_MM_MAX + 1];
extern const struct nla_policy ethnl_module_fw_flash_act_policy[ETHTOOL_A_MODULE_FW_FLASH_PASSWORD + 1];
+extern const struct nla_policy ethnl_tsconfig_get_policy[ETHTOOL_A_TSCONFIG_HEADER + 1];
+extern const struct nla_policy ethnl_tsconfig_set_policy[ETHTOOL_A_TSCONFIG_MAX + 1];
int ethnl_set_features(struct sk_buff *skb, struct genl_info *info);
int ethnl_act_cable_test(struct sk_buff *skb, struct genl_info *info);
diff --git a/net/ethtool/tsconfig.c b/net/ethtool/tsconfig.c
new file mode 100644
index 000000000000..518ee9e420f5
--- /dev/null
+++ b/net/ethtool/tsconfig.c
@@ -0,0 +1,347 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#include <linux/net_tstamp.h>
+#include <linux/ptp_clock_kernel.h>
+
+#include "netlink.h"
+#include "common.h"
+#include "bitset.h"
+#include "../core/dev.h"
+#include "ts.h"
+
+struct tsconfig_req_info {
+ struct ethnl_req_info base;
+};
+
+struct tsconfig_reply_data {
+ struct ethnl_reply_data base;
+ struct hwtst_provider hwtst;
+ struct {
+ u32 tx_type;
+ u32 rx_filter;
+ u32 flags;
+ } hwtst_config;
+};
+
+#define TSCONFIG_REPDATA(__reply_base) \
+ container_of(__reply_base, struct tsconfig_reply_data, base)
+
+const struct nla_policy ethnl_tsconfig_get_policy[ETHTOOL_A_TSCONFIG_HEADER + 1] = {
+ [ETHTOOL_A_TSCONFIG_HEADER] =
+ NLA_POLICY_NESTED(ethnl_header_policy),
+};
+
+static int tsconfig_prepare_data(const struct ethnl_req_info *req_base,
+ struct ethnl_reply_data *reply_base,
+ const struct genl_info *info)
+{
+ struct tsconfig_reply_data *data = TSCONFIG_REPDATA(reply_base);
+ struct hwtstamp_provider *hwtstamp = NULL;
+ struct net_device *dev = reply_base->dev;
+ struct kernel_hwtstamp_config cfg = {};
+ int ret;
+
+ if (!dev->netdev_ops->ndo_hwtstamp_get)
+ return -EOPNOTSUPP;
+
+ ret = ethnl_ops_begin(dev);
+ if (ret < 0)
+ return ret;
+
+ ret = dev_get_hwtstamp_phylib(dev, &cfg);
+ if (ret)
+ goto out;
+
+ data->hwtst_config.tx_type = BIT(cfg.tx_type);
+ data->hwtst_config.rx_filter = BIT(cfg.rx_filter);
+ data->hwtst_config.flags = BIT(cfg.flags);
+
+ data->hwtst.index = -1;
+ hwtstamp = rtnl_dereference(dev->hwtstamp);
+ if (hwtstamp) {
+ data->hwtst.index = ptp_clock_index(hwtstamp->ptp);
+ data->hwtst.qualifier = hwtstamp->qualifier;
+ }
+
+out:
+ ethnl_ops_complete(dev);
+ return ret;
+}
+
+static int tsconfig_reply_size(const struct ethnl_req_info *req_base,
+ const struct ethnl_reply_data *reply_base)
+{
+ const struct tsconfig_reply_data *data = TSCONFIG_REPDATA(reply_base);
+ bool compact = req_base->flags & ETHTOOL_FLAG_COMPACT_BITSETS;
+ int len = 0;
+ int ret;
+
+ BUILD_BUG_ON(__HWTSTAMP_TX_CNT > 32);
+ BUILD_BUG_ON(__HWTSTAMP_FILTER_CNT > 32);
+
+ if (data->hwtst_config.flags)
+ /* _TSCONFIG_HWTSTAMP_FLAGS */
+ len += nla_total_size(sizeof(u32));
+
+ if (data->hwtst_config.tx_type) {
+ ret = ethnl_bitset32_size(&data->hwtst_config.tx_type,
+ NULL, __HWTSTAMP_TX_CNT,
+ ts_tx_type_names, compact);
+ if (ret < 0)
+ return ret;
+ len += ret; /* _TSCONFIG_TX_TYPES */
+ }
+ if (data->hwtst_config.rx_filter) {
+ ret = ethnl_bitset32_size(&data->hwtst_config.rx_filter,
+ NULL, __HWTSTAMP_FILTER_CNT,
+ ts_rx_filter_names, compact);
+ if (ret < 0)
+ return ret;
+ len += ret; /* _TSCONFIG_RX_FILTERS */
+ }
+
+ if (data->hwtst.index >= 0)
+ /* _TSINFO_HWTSTAMP_PROVIDER */
+ len += 2 * nla_total_size(sizeof(u32));
+
+ return len;
+}
+
+static int tsconfig_fill_reply(struct sk_buff *skb,
+ const struct ethnl_req_info *req_base,
+ const struct ethnl_reply_data *reply_base)
+{
+ const struct tsconfig_reply_data *data = TSCONFIG_REPDATA(reply_base);
+ bool compact = req_base->flags & ETHTOOL_FLAG_COMPACT_BITSETS;
+ int ret;
+
+ if (data->hwtst_config.flags) {
+ ret = nla_put_u32(skb, ETHTOOL_A_TSCONFIG_HWTSTAMP_FLAGS,
+ data->hwtst_config.flags);
+ if (ret < 0)
+ return ret;
+ }
+
+ if (data->hwtst_config.tx_type) {
+ ret = ethnl_put_bitset32(skb, ETHTOOL_A_TSCONFIG_TX_TYPES,
+ &data->hwtst_config.tx_type, NULL,
+ __HWTSTAMP_TX_CNT,
+ ts_tx_type_names, compact);
+ if (ret < 0)
+ return ret;
+ }
+
+ if (data->hwtst_config.rx_filter) {
+ ret = ethnl_put_bitset32(skb, ETHTOOL_A_TSCONFIG_RX_FILTERS,
+ &data->hwtst_config.rx_filter,
+ NULL, __HWTSTAMP_FILTER_CNT,
+ ts_rx_filter_names, compact);
+ if (ret < 0)
+ return ret;
+ }
+
+ if (data->hwtst.index >= 0) {
+ struct nlattr *nest;
+
+ nest = nla_nest_start(skb, ETHTOOL_A_TSCONFIG_HWTSTAMP_PROVIDER);
+ if (!nest)
+ return -EMSGSIZE;
+
+ if (nla_put_u32(skb, ETHTOOL_A_TS_HWTSTAMP_PROVIDER_INDEX,
+ data->hwtst.index) ||
+ nla_put_u32(skb,
+ ETHTOOL_A_TS_HWTSTAMP_PROVIDER_QUALIFIER,
+ data->hwtst.qualifier)) {
+ nla_nest_cancel(skb, nest);
+ return -EMSGSIZE;
+ }
+
+ nla_nest_end(skb, nest);
+ }
+ return 0;
+}
+
+/* TSCONFIG_SET */
+const struct nla_policy ethnl_tsconfig_set_policy[ETHTOOL_A_TSCONFIG_MAX + 1] = {
+ [ETHTOOL_A_TSCONFIG_HEADER] = NLA_POLICY_NESTED(ethnl_header_policy),
+ [ETHTOOL_A_TSCONFIG_HWTSTAMP_PROVIDER] = { .type = NLA_NESTED },
+ [ETHTOOL_A_TSCONFIG_HWTSTAMP_FLAGS] = { .type = NLA_U32 },
+ [ETHTOOL_A_TSCONFIG_RX_FILTERS] = { .type = NLA_NESTED },
+ [ETHTOOL_A_TSCONFIG_TX_TYPES] = { .type = NLA_NESTED },
+};
+
+static int ethnl_set_tsconfig_validate(struct ethnl_req_info *req_base,
+ struct genl_info *info)
+{
+ const struct net_device_ops *ops = req_base->dev->netdev_ops;
+
+ if (!ops->ndo_hwtstamp_set || !ops->ndo_hwtstamp_get)
+ return -EOPNOTSUPP;
+
+ return 1;
+}
+
+static int ethnl_set_tsconfig(struct ethnl_req_info *req_base,
+ struct genl_info *info)
+{
+ unsigned long mask = 0, req_rx_filter, req_tx_type;
+ struct kernel_hwtstamp_config hwtst_config = {0};
+ struct hwtstamp_provider *hwtstamp = NULL;
+ struct net_device *dev = req_base->dev;
+ struct nlattr **tb = info->attrs;
+ bool mod = false;
+ int ret;
+
+ BUILD_BUG_ON(__HWTSTAMP_TX_CNT > 32);
+ BUILD_BUG_ON(__HWTSTAMP_FILTER_CNT > 32);
+
+ if (!netif_device_present(dev))
+ return -ENODEV;
+
+ if (tb[ETHTOOL_A_TSCONFIG_HWTSTAMP_PROVIDER]) {
+ struct hwtst_provider __hwtst = {.index = -1};
+ struct hwtstamp_provider *__hwtstamp;
+
+ __hwtstamp = rtnl_dereference(dev->hwtstamp);
+ if (__hwtstamp) {
+ __hwtst.index = ptp_clock_index(__hwtstamp->ptp);
+ __hwtst.qualifier = __hwtstamp->qualifier;
+ }
+
+ ret = ts_parse_hwtst_provider(tb[ETHTOOL_A_TSCONFIG_HWTSTAMP_PROVIDER],
+ &__hwtst, info->extack,
+ &mod);
+ if (ret < 0)
+ return ret;
+
+ if (mod) {
+ hwtstamp = devm_kzalloc(&dev->dev, sizeof(*hwtstamp),
+ GFP_KERNEL);
+ if (!hwtstamp)
+ return -ENOMEM;
+
+ hwtstamp->ptp = ptp_clock_get_by_index(&dev->dev,
+ __hwtst.index);
+ if (!hwtstamp->ptp) {
+ NL_SET_ERR_MSG_ATTR(info->extack,
+ tb[ETHTOOL_A_TSCONFIG_HWTSTAMP_PROVIDER],
+ "no phc at such index");
+ ret = -ENODEV;
+ goto err_free_hwtstamp;
+ }
+ hwtstamp->qualifier = __hwtst.qualifier;
+ hwtstamp->dev = &dev->dev;
+
+ /* Does the hwtstamp supported in the netdev topology */
+ if (!netdev_support_hwtstamp(dev, hwtstamp)) {
+ NL_SET_ERR_MSG_ATTR(info->extack,
+ tb[ETHTOOL_A_TSCONFIG_HWTSTAMP_PROVIDER],
+ "phc not in this net device topology");
+ ret = -ENODEV;
+ goto err_clock_put;
+ }
+ }
+ }
+
+ /* Get current hwtstamp config if we are not changing the hwtstamp
+ * source
+ */
+ if (!mod) {
+ ret = dev_get_hwtstamp_phylib(dev, &hwtst_config);
+ if (ret < 0 && ret != -EOPNOTSUPP)
+ goto err_clock_put;
+ }
+
+ /* Get the hwtstamp config from netlink */
+ if (tb[ETHTOOL_A_TSCONFIG_TX_TYPES]) {
+ ret = ethnl_parse_bitset(&req_tx_type, &mask,
+ __HWTSTAMP_TX_CNT,
+ tb[ETHTOOL_A_TSCONFIG_TX_TYPES],
+ ts_tx_type_names, info->extack);
+ if (ret < 0)
+ goto err_clock_put;
+
+ /* Select only one tx type at a time */
+ if (ffs(req_tx_type) != fls(req_tx_type)) {
+ ret = -EINVAL;
+ goto err_clock_put;
+ }
+
+ hwtst_config.tx_type = ffs(req_tx_type) - 1;
+ }
+ if (tb[ETHTOOL_A_TSCONFIG_RX_FILTERS]) {
+ ret = ethnl_parse_bitset(&req_rx_filter, &mask,
+ __HWTSTAMP_FILTER_CNT,
+ tb[ETHTOOL_A_TSCONFIG_RX_FILTERS],
+ ts_rx_filter_names, info->extack);
+ if (ret < 0)
+ goto err_clock_put;
+
+ /* Select only one rx filter at a time */
+ if (ffs(req_rx_filter) != fls(req_rx_filter)) {
+ ret = -EINVAL;
+ goto err_clock_put;
+ }
+
+ hwtst_config.rx_filter = ffs(req_rx_filter) - 1;
+ }
+ if (tb[ETHTOOL_A_TSCONFIG_HWTSTAMP_FLAGS]) {
+ ret = nla_get_u32(tb[ETHTOOL_A_TSCONFIG_HWTSTAMP_FLAGS]);
+ if (ret < 0)
+ goto err_clock_put;
+ hwtst_config.flags = ret;
+ }
+
+ ret = net_hwtstamp_validate(&hwtst_config);
+ if (ret)
+ goto err_clock_put;
+
+ /* Disable current time stamping if we try to enable another one */
+ if (mod && (hwtst_config.tx_type || hwtst_config.rx_filter)) {
+ struct kernel_hwtstamp_config zero_config = {0};
+
+ ret = dev_set_hwtstamp_phylib(dev, &zero_config, info->extack);
+ if (ret < 0)
+ goto err_clock_put;
+ }
+
+ /* Changed the selected hwtstamp source if needed */
+ if (mod) {
+ struct hwtstamp_provider *__hwtstamp;
+
+ __hwtstamp = rcu_replace_pointer_rtnl(dev->hwtstamp, hwtstamp);
+ if (__hwtstamp)
+ call_rcu(&__hwtstamp->rcu_head,
+ remove_hwtstamp_provider);
+ }
+
+ ret = dev_set_hwtstamp_phylib(dev, &hwtst_config, info->extack);
+ if (ret < 0)
+ return ret;
+
+ return 1;
+
+err_clock_put:
+ if (hwtstamp)
+ ptp_clock_put(&dev->dev, hwtstamp->ptp);
+err_free_hwtstamp:
+ devm_kfree(&dev->dev, hwtstamp);
+
+ return ret;
+}
+
+const struct ethnl_request_ops ethnl_tsconfig_request_ops = {
+ .request_cmd = ETHTOOL_MSG_TSCONFIG_GET,
+ .reply_cmd = ETHTOOL_MSG_TSCONFIG_GET_REPLY,
+ .hdr_attr = ETHTOOL_A_TSCONFIG_HEADER,
+ .req_info_size = sizeof(struct tsconfig_req_info),
+ .reply_data_size = sizeof(struct tsconfig_reply_data),
+
+ .prepare_data = tsconfig_prepare_data,
+ .reply_size = tsconfig_reply_size,
+ .fill_reply = tsconfig_fill_reply,
+
+ .set_validate = ethnl_set_tsconfig_validate,
+ .set = ethnl_set_tsconfig,
+ .set_ntf_cmd = ETHTOOL_MSG_TSCONFIG_NTF,
+};
--
2.34.1
^ permalink raw reply related [flat|nested] 39+ messages in thread
* [PATCH net-next v17 14/14] netlink: specs: Enhance tsinfo netlink attributes and add a tsconfig set command
2024-07-09 13:53 [PATCH net-next v17 00/14] net: Make timestamping selectable Kory Maincent
` (12 preceding siblings ...)
2024-07-09 13:53 ` [PATCH net-next v17 13/14] net: ethtool: Add support for tsconfig command to get/set hwtstamp config Kory Maincent
@ 2024-07-09 13:53 ` Kory Maincent
2024-07-17 17:44 ` Jacob Keller
2024-07-15 15:30 ` [PATCH net-next v17 00/14] net: Make timestamping selectable patchwork-bot+netdevbpf
14 siblings, 1 reply; 39+ messages in thread
From: Kory Maincent @ 2024-07-09 13:53 UTC (permalink / raw)
To: Florian Fainelli, Broadcom internal kernel review list,
Andrew Lunn, Heiner Kallweit, Russell King, David S. Miller,
Eric Dumazet, Jakub Kicinski, Paolo Abeni, Richard Cochran,
Radu Pirea, Jay Vosburgh, Andy Gospodarek, Nicolas Ferre,
Claudiu Beznea, Willem de Bruijn, Jonathan Corbet, Horatiu Vultur,
UNGLinuxDriver, Simon Horman, Vladimir Oltean, donald.hunter,
danieller, ecree.xilinx
Cc: Thomas Petazzoni, linux-kernel, netdev, linux-doc,
Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter, Kory Maincent
Add new attributed to tsinfo allowing to get the tsinfo from a phc provider
(composed by a phc index and a phc qualifier) on a netdevice's link.
Add simultaneously a tsconfig command to be able to get and set hwtstamp
configuration for a specified phc provider.
Here is few examples:
./ynl/cli.py --spec netlink/specs/ethtool.yaml --no-schema
--dump tsinfo-get
--json '{"header":{"dev-name":"eth0"}}'
[{'header': {'dev-index': 3, 'dev-name': 'eth0'},
'hwtst-provider': {'index': 0, 'qualifier': 0},
'phc-index': 0,
'rx-filters': {'bits': {'bit': [{'index': 0, 'name': 'none'},
{'index': 2, 'name': 'some'}]},
'nomask': True,
'size': 16},
'timestamping': {'bits': {'bit': [{'index': 0, 'name': 'hardware-transmit'},
{'index': 2, 'name': 'hardware-receive'},
{'index': 6,
'name': 'hardware-raw-clock'}]},
'nomask': True,
'size': 17},
'tx-types': {'bits': {'bit': [{'index': 0, 'name': 'off'},
{'index': 1, 'name': 'on'}]},
'nomask': True,
'size': 4}},
{'header': {'dev-index': 3, 'dev-name': 'eth0'},
'hwtst-provider': {'index': 2, 'qualifier': 0},
'phc-index': 2,
'rx-filters': {'bits': {'bit': [{'index': 0, 'name': 'none'},
{'index': 1, 'name': 'all'}]},
'nomask': True,
'size': 16},
'timestamping': {'bits': {'bit': [{'index': 0, 'name': 'hardware-transmit'},
{'index': 1, 'name': 'software-transmit'},
{'index': 2, 'name': 'hardware-receive'},
{'index': 3, 'name': 'software-receive'},
{'index': 4,
'name': 'software-system-clock'},
{'index': 6,
'name': 'hardware-raw-clock'}]},
'nomask': True,
'size': 17},
'tx-types': {'bits': {'bit': [{'index': 0, 'name': 'off'},
{'index': 1, 'name': 'on'},
{'index': 2, 'name': 'onestep-sync'}]},
'nomask': True,
'size': 4}}]
./ynl/cli.py --spec netlink/specs/ethtool.yaml --no-schema --do tsinfo-get
--json '{"header":{"dev-name":"eth0"},
"hwtst-provider":{"index":0, "qualifier":0 }
}'
{'header': {'dev-index': 3, 'dev-name': 'eth0'},
'hwtst-provider': {'index': 0, 'qualifier': 0},
'phc-index': 0,
'rx-filters': {'bits': {'bit': [{'index': 0, 'name': 'none'},
{'index': 2, 'name': 'some'}]},
'nomask': True,
'size': 16},
'timestamping': {'bits': {'bit': [{'index': 0, 'name': 'hardware-transmit'},
{'index': 2, 'name': 'hardware-receive'},
{'index': 6, 'name': 'hardware-raw-clock'}]},
'nomask': True,
'size': 17},
'tx-types': {'bits': {'bit': [{'index': 0, 'name': 'off'},
{'index': 1, 'name': 'on'}]},
'nomask': True,
'size': 4}}
./ynl/cli.py --spec netlink/specs/ethtool.yaml --no-schema --do tsinfo-set
--json '{"header":{"dev-name":"eth0"},
"hwtst-provider":{"index":2, "qualifier":0}}'
None
./ynl/cli.py --spec netlink/specs/ethtool.yaml --no-schema --do tsconfig-get
--json '{"header":{"dev-name":"eth0"}}'
{'header': {'dev-index': 3, 'dev-name': 'eth0'},
'hwtstamp-flags': 1,
'hwtstamp-provider': {'index': 1, 'qualifier': 0},
'rx-filters': {'bits': {'bit': [{'index': 12, 'name': 'ptpv2-event'}]},
'nomask': True,
'size': 16},
'tx-types': {'bits': {'bit': [{'index': 1, 'name': 'on'}]},
'nomask': True,
'size': 4}}
./ynl/cli.py --spec netlink/specs/ethtool.yaml --no-schema --do tsconfig-set
--json '{"header":{"dev-name":"eth0"},
"hwtstamp-provider":{"index":1, "qualifier":0 },
"rx-filters":{"bits": {"bit": {"name":"ptpv2-l4-event"}},
"nomask": 1},
"tx-types":{"bits": {"bit": {"name":"on"}},
"nomask": 1}}'
None
Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
---
Changes in v8:
- New patch
Changes in v10:
- Add ghwtstamp attributes
- Add tsinfo ntf command
Changes in v11:
- Add examples in the commit message.
Changes in v13:
- Replace shorter name by real name.
- Fix an issue reported by "make -C tools/net/ynl" on the namings.
Changes in v16:
- Move to tsconfig command to get and set hwtstamp configuration.
---
Documentation/netlink/specs/ethtool.yaml | 73 ++++++++++++++++++++++++++++++++
1 file changed, 73 insertions(+)
diff --git a/Documentation/netlink/specs/ethtool.yaml b/Documentation/netlink/specs/ethtool.yaml
index 495e35fcfb21..9bd1f8a7bd3f 100644
--- a/Documentation/netlink/specs/ethtool.yaml
+++ b/Documentation/netlink/specs/ethtool.yaml
@@ -624,6 +624,15 @@ attribute-sets:
-
name: tx-err
type: uint
+ -
+ name: ts-hwtstamp-provider
+ attributes:
+ -
+ name: index
+ type: u32
+ -
+ name: qualifier
+ type: u32
-
name: tsinfo
attributes:
@@ -650,6 +659,10 @@ attribute-sets:
name: stats
type: nest
nested-attributes: ts-stat
+ -
+ name: hwtstamp-provider
+ type: nest
+ nested-attributes: ts-hwtstamp-provider
-
name: cable-result
attributes:
@@ -1085,6 +1098,28 @@ attribute-sets:
-
name: total
type: uint
+ -
+ name: tsconfig
+ attributes:
+ -
+ name: header
+ type: nest
+ nested-attributes: header
+ -
+ name: hwtstamp-provider
+ type: nest
+ nested-attributes: ts-hwtstamp-provider
+ -
+ name: tx-types
+ type: nest
+ nested-attributes: bitset
+ -
+ name: rx-filters
+ type: nest
+ nested-attributes: bitset
+ -
+ name: hwtstamp-flags
+ type: u32
operations:
enum-model: directional
@@ -1526,6 +1561,7 @@ operations:
request:
attributes:
- header
+ - hwtstamp-provider
reply:
attributes:
- header
@@ -1534,6 +1570,7 @@ operations:
- rx-filters
- phc-index
- stats
+ - hwtstamp-provider
dump: *tsinfo-get-op
-
name: cable-test-act
@@ -1877,3 +1914,39 @@ operations:
- status-msg
- done
- total
+ -
+ name: tsconfig-get
+ doc: Get hwtstamp config.
+
+ attribute-set: tsconfig
+
+ do: &tsconfig-get-op
+ request:
+ attributes:
+ - header
+ reply:
+ attributes:
+ - header
+ - hwtstamp-provider
+ - tx-types
+ - rx-filters
+ - hwtstamp-flags
+ dump: *tsconfig-get-op
+ -
+ name: tsconfig-set
+ doc: Set hwtstamp config.
+
+ attribute-set: tsconfig
+
+ do:
+ request:
+ attributes:
+ - header
+ - hwtstamp-provider
+ - tx-types
+ - rx-filters
+ - hwtstamp-flags
+ -
+ name: tsconfig-ntf
+ doc: Notification for change in tsconfig configuration.
+ notify: tsconfig-get
--
2.34.1
^ permalink raw reply related [flat|nested] 39+ messages in thread
* Re: [PATCH net-next v17 13/14] net: ethtool: Add support for tsconfig command to get/set hwtstamp config
2024-07-09 13:53 ` [PATCH net-next v17 13/14] net: ethtool: Add support for tsconfig command to get/set hwtstamp config Kory Maincent
@ 2024-07-15 14:59 ` Jakub Kicinski
2024-09-26 8:47 ` Kory Maincent
2024-07-17 17:43 ` Jacob Keller
1 sibling, 1 reply; 39+ messages in thread
From: Jakub Kicinski @ 2024-07-15 14:59 UTC (permalink / raw)
To: Kory Maincent
Cc: Florian Fainelli, Broadcom internal kernel review list,
Andrew Lunn, Heiner Kallweit, Russell King, David S. Miller,
Eric Dumazet, Paolo Abeni, Richard Cochran, Radu Pirea,
Jay Vosburgh, Andy Gospodarek, Nicolas Ferre, Claudiu Beznea,
Willem de Bruijn, Jonathan Corbet, Horatiu Vultur, UNGLinuxDriver,
Simon Horman, Vladimir Oltean, donald.hunter, danieller,
ecree.xilinx, Thomas Petazzoni, linux-kernel, netdev, linux-doc,
Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter
On Tue, 09 Jul 2024 15:53:45 +0200 Kory Maincent wrote:
> + /* Get the hwtstamp config from netlink */
> + if (tb[ETHTOOL_A_TSCONFIG_TX_TYPES]) {
> + ret = ethnl_parse_bitset(&req_tx_type, &mask,
> + __HWTSTAMP_TX_CNT,
> + tb[ETHTOOL_A_TSCONFIG_TX_TYPES],
> + ts_tx_type_names, info->extack);
> + if (ret < 0)
> + goto err_clock_put;
> +
> + /* Select only one tx type at a time */
> + if (ffs(req_tx_type) != fls(req_tx_type)) {
> + ret = -EINVAL;
> + goto err_clock_put;
> + }
> +
> + hwtst_config.tx_type = ffs(req_tx_type) - 1;
> + }
> + if (tb[ETHTOOL_A_TSCONFIG_RX_FILTERS]) {
> + ret = ethnl_parse_bitset(&req_rx_filter, &mask,
> + __HWTSTAMP_FILTER_CNT,
> + tb[ETHTOOL_A_TSCONFIG_RX_FILTERS],
> + ts_rx_filter_names, info->extack);
> + if (ret < 0)
> + goto err_clock_put;
> +
> + /* Select only one rx filter at a time */
> + if (ffs(req_rx_filter) != fls(req_rx_filter)) {
> + ret = -EINVAL;
> + goto err_clock_put;
> + }
> +
> + hwtst_config.rx_filter = ffs(req_rx_filter) - 1;
> + }
> + if (tb[ETHTOOL_A_TSCONFIG_HWTSTAMP_FLAGS]) {
> + ret = nla_get_u32(tb[ETHTOOL_A_TSCONFIG_HWTSTAMP_FLAGS]);
> + if (ret < 0)
> + goto err_clock_put;
> + hwtst_config.flags = ret;
> + }
We should be tracking mod on these, too. Separately from the provider
mod bit, let's not call the driver and send notification if nothing
changed.
> + ret = net_hwtstamp_validate(&hwtst_config);
> + if (ret)
> + goto err_clock_put;
> +
> + /* Disable current time stamping if we try to enable another one */
> + if (mod && (hwtst_config.tx_type || hwtst_config.rx_filter)) {
> + struct kernel_hwtstamp_config zero_config = {0};
> +
> + ret = dev_set_hwtstamp_phylib(dev, &zero_config, info->extack);
> + if (ret < 0)
> + goto err_clock_put;
> + }
> +
> + /* Changed the selected hwtstamp source if needed */
> + if (mod) {
> + struct hwtstamp_provider *__hwtstamp;
> +
> + __hwtstamp = rcu_replace_pointer_rtnl(dev->hwtstamp, hwtstamp);
> + if (__hwtstamp)
> + call_rcu(&__hwtstamp->rcu_head,
> + remove_hwtstamp_provider);
> + }
> +
> + ret = dev_set_hwtstamp_phylib(dev, &hwtst_config, info->extack);
> + if (ret < 0)
> + return ret;
We can't unwind to old state here?
> + return 1;
Driver can change hwtst_config right? "upgrade" the rx_filter
to a broader one, IIRC. Shouldn't we reply to the set command with
the resulting configuration, in case it changed? Basically provide
the same info as the notification would.
^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: [PATCH net-next v17 00/14] net: Make timestamping selectable
2024-07-09 13:53 [PATCH net-next v17 00/14] net: Make timestamping selectable Kory Maincent
` (13 preceding siblings ...)
2024-07-09 13:53 ` [PATCH net-next v17 14/14] netlink: specs: Enhance tsinfo netlink attributes and add a tsconfig set command Kory Maincent
@ 2024-07-15 15:30 ` patchwork-bot+netdevbpf
14 siblings, 0 replies; 39+ messages in thread
From: patchwork-bot+netdevbpf @ 2024-07-15 15:30 UTC (permalink / raw)
To: Kory Maincent
Cc: florian.fainelli, bcm-kernel-feedback-list, andrew, hkallweit1,
linux, davem, edumazet, kuba, pabeni, richardcochran,
radu-nicolae.pirea, j.vosburgh, andy, nicolas.ferre,
claudiu.beznea, willemdebruijn.kernel, corbet, horatiu.vultur,
UNGLinuxDriver, horms, vladimir.oltean, donald.hunter, danieller,
ecree.xilinx, thomas.petazzoni, linux-kernel, netdev, linux-doc,
maxime.chevallier, rrameshbabu, willemb, shannon.nelson, wintera
Hello:
This series was applied to netdev/net-next.git (main)
by Jakub Kicinski <kuba@kernel.org>:
On Tue, 09 Jul 2024 15:53:32 +0200 you wrote:
> Up until now, there was no way to let the user select the hardware
> PTP provider at which time stamping occurs. The stack assumed that PHY time
> stamping is always preferred, but some MAC/PHY combinations were buggy.
>
> This series updates the default MAC/PHY default timestamping and aims to
> allow the user to select the desired hwtstamp provider administratively.
>
> [...]
Here is the summary with links:
- [net-next,v17,01/14] net_tstamp: Add TIMESTAMPING SOFTWARE and HARDWARE mask
https://git.kernel.org/netdev/net-next/c/e50bfd6bb231
- [net-next,v17,02/14] net: Make dev_get_hwtstamp_phylib accessible
(no matching commit)
- [net-next,v17,03/14] net: Make net_hwtstamp_validate accessible
(no matching commit)
- [net-next,v17,04/14] net: Change the API of PHY default timestamp to MAC
(no matching commit)
- [net-next,v17,05/14] net: net_tstamp: Add unspec field to hwtstamp_source enumeration
https://git.kernel.org/netdev/net-next/c/bc5a07ed15a3
- [net-next,v17,06/14] net: Add struct kernel_ethtool_ts_info
https://git.kernel.org/netdev/net-next/c/2111375b85ad
- [net-next,v17,07/14] ptp: Add phc source and helpers to register specific PTP clock or get information
(no matching commit)
- [net-next,v17,08/14] net: Add the possibility to support a selected hwtstamp in netdevice
(no matching commit)
- [net-next,v17,09/14] net: netdevsim: ptp_mock: Convert to netdev_ptp_clock_register
(no matching commit)
- [net-next,v17,10/14] net: macb: Convert to netdev_ptp_clock_register
(no matching commit)
- [net-next,v17,11/14] net: ptp: Move ptp_clock_index() to builtin symbol
(no matching commit)
- [net-next,v17,12/14] net: ethtool: tsinfo: Add support for reading tsinfo for a specific hwtstamp provider
(no matching commit)
- [net-next,v17,13/14] net: ethtool: Add support for tsconfig command to get/set hwtstamp config
(no matching commit)
- [net-next,v17,14/14] netlink: specs: Enhance tsinfo netlink attributes and add a tsconfig set command
(no matching commit)
You are awesome, thank you!
--
Deet-doot-dot, I am a bot.
https://korg.docs.kernel.org/patchwork/pwbot.html
^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: [PATCH net-next v17 01/14] net_tstamp: Add TIMESTAMPING SOFTWARE and HARDWARE mask
2024-07-09 13:53 ` [PATCH net-next v17 01/14] net_tstamp: Add TIMESTAMPING SOFTWARE and HARDWARE mask Kory Maincent
@ 2024-07-15 23:28 ` Jacob Keller
0 siblings, 0 replies; 39+ messages in thread
From: Jacob Keller @ 2024-07-15 23:28 UTC (permalink / raw)
To: Kory Maincent, Florian Fainelli,
Broadcom internal kernel review list, Andrew Lunn,
Heiner Kallweit, Russell King, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, Richard Cochran, Radu Pirea,
Jay Vosburgh, Andy Gospodarek, Nicolas Ferre, Claudiu Beznea,
Willem de Bruijn, Jonathan Corbet, Horatiu Vultur, UNGLinuxDriver,
Simon Horman, Vladimir Oltean, donald.hunter, danieller,
ecree.xilinx
Cc: Thomas Petazzoni, linux-kernel, netdev, linux-doc,
Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter
On 7/9/2024 6:53 AM, Kory Maincent wrote:
> Timestamping software or hardware flags are often used as a group,
> therefore adding these masks will ease future use.
>
> I did not use SOF_TIMESTAMPING_SYS_HARDWARE flag as it is deprecated and
> not used at all.
+1. Is there any hope of completely removing it? I'm not certain if this
is part of uAPI or not. Even so, we could make it more clear with
deprecation naming or similar.
>
> Reviewed-by: Willem de Bruijn <willemb@google.com>
> Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
> Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
> ---
>
> Changes in v7:
> - Move the masks out of uapi to include/linux/net_tstamp.h
>
> Changes in v9:
> - Fix commit message typos
> ---
> include/linux/net_tstamp.h | 8 ++++++++
> 1 file changed, 8 insertions(+)
>
> diff --git a/include/linux/net_tstamp.h b/include/linux/net_tstamp.h
> index eb01c37e71e0..3799c79b6c83 100644
> --- a/include/linux/net_tstamp.h
> +++ b/include/linux/net_tstamp.h
> @@ -5,6 +5,14 @@
>
> #include <uapi/linux/net_tstamp.h>
>
> +#define SOF_TIMESTAMPING_SOFTWARE_MASK (SOF_TIMESTAMPING_RX_SOFTWARE | \
> + SOF_TIMESTAMPING_TX_SOFTWARE | \
> + SOF_TIMESTAMPING_SOFTWARE)
> +
> +#define SOF_TIMESTAMPING_HARDWARE_MASK (SOF_TIMESTAMPING_RX_HARDWARE | \
> + SOF_TIMESTAMPING_TX_HARDWARE | \
> + SOF_TIMESTAMPING_RAW_HARDWARE)
> +
We can't drop _MASK because there already is SOF_TIMESTAMPING_SOFTWARE.
Makes sense.
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
> enum hwtstamp_source {
> HWTSTAMP_SOURCE_NETDEV,
> HWTSTAMP_SOURCE_PHYLIB,
>
^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: [PATCH net-next v17 02/14] net: Make dev_get_hwtstamp_phylib accessible
2024-07-09 13:53 ` [PATCH net-next v17 02/14] net: Make dev_get_hwtstamp_phylib accessible Kory Maincent
@ 2024-07-15 23:29 ` Jacob Keller
0 siblings, 0 replies; 39+ messages in thread
From: Jacob Keller @ 2024-07-15 23:29 UTC (permalink / raw)
To: Kory Maincent, Florian Fainelli,
Broadcom internal kernel review list, Andrew Lunn,
Heiner Kallweit, Russell King, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, Richard Cochran, Radu Pirea,
Jay Vosburgh, Andy Gospodarek, Nicolas Ferre, Claudiu Beznea,
Willem de Bruijn, Jonathan Corbet, Horatiu Vultur, UNGLinuxDriver,
Simon Horman, Vladimir Oltean, donald.hunter, danieller,
ecree.xilinx
Cc: Thomas Petazzoni, linux-kernel, netdev, linux-doc,
Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter
On 7/9/2024 6:53 AM, Kory Maincent wrote:
> Make the dev_get_hwtstamp_phylib function accessible in prevision to use
> it from ethtool to read the hwtstamp current configuration.
>
> Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
> Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
> ---
>
> Change in v8:
> - New patch
>
> Change in v10:
> - Remove export symbol as ethtool can't be built as a module.
> - Move the declaration to net/core/dev.h instead of netdevice.h
> ---
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
> net/core/dev.h | 2 ++
> net/core/dev_ioctl.c | 4 ++--
> 2 files changed, 4 insertions(+), 2 deletions(-)
>
> diff --git a/net/core/dev.h b/net/core/dev.h
> index 5654325c5b71..9d4ceaf9bdc0 100644
> --- a/net/core/dev.h
> +++ b/net/core/dev.h
> @@ -187,5 +187,7 @@ static inline void dev_xmit_recursion_dec(void)
> int dev_set_hwtstamp_phylib(struct net_device *dev,
> struct kernel_hwtstamp_config *cfg,
> struct netlink_ext_ack *extack);
> +int dev_get_hwtstamp_phylib(struct net_device *dev,
> + struct kernel_hwtstamp_config *cfg);
>
> #endif
> diff --git a/net/core/dev_ioctl.c b/net/core/dev_ioctl.c
> index b9719ed3c3fd..b8cf8c55fa2d 100644
> --- a/net/core/dev_ioctl.c
> +++ b/net/core/dev_ioctl.c
> @@ -268,8 +268,8 @@ static int dev_eth_ioctl(struct net_device *dev,
> * -EOPNOTSUPP for phylib for now, which is still more accurate than letting
> * the netdev handle the GET request.
> */
> -static int dev_get_hwtstamp_phylib(struct net_device *dev,
> - struct kernel_hwtstamp_config *cfg)
> +int dev_get_hwtstamp_phylib(struct net_device *dev,
> + struct kernel_hwtstamp_config *cfg)
> {
> if (phy_has_hwtstamp(dev->phydev))
> return phy_hwtstamp_get(dev->phydev, cfg);
>
^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: [PATCH net-next v17 03/14] net: Make net_hwtstamp_validate accessible
2024-07-09 13:53 ` [PATCH net-next v17 03/14] net: Make net_hwtstamp_validate accessible Kory Maincent
@ 2024-07-15 23:30 ` Jacob Keller
0 siblings, 0 replies; 39+ messages in thread
From: Jacob Keller @ 2024-07-15 23:30 UTC (permalink / raw)
To: Kory Maincent, Florian Fainelli,
Broadcom internal kernel review list, Andrew Lunn,
Heiner Kallweit, Russell King, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, Richard Cochran, Radu Pirea,
Jay Vosburgh, Andy Gospodarek, Nicolas Ferre, Claudiu Beznea,
Willem de Bruijn, Jonathan Corbet, Horatiu Vultur, UNGLinuxDriver,
Simon Horman, Vladimir Oltean, donald.hunter, danieller,
ecree.xilinx
Cc: Thomas Petazzoni, linux-kernel, netdev, linux-doc,
Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter
On 7/9/2024 6:53 AM, Kory Maincent wrote:
> Make the net_hwtstamp_validate function accessible in prevision to use
> it from ethtool to validate the hwtstamp configuration before setting it.
>
> Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
> Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
> ---
>
> Change in v8:
> - New patch
>
> Change in v10:
> - Remove export symbol as ethtool can't be built as a module.
> - Move the declaration to net/core/dev.h instead of netdevice.h
> ---
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
> net/core/dev.h | 1 +
> net/core/dev_ioctl.c | 2 +-
> 2 files changed, 2 insertions(+), 1 deletion(-)
>
> diff --git a/net/core/dev.h b/net/core/dev.h
> index 9d4ceaf9bdc0..df085c3e510c 100644
> --- a/net/core/dev.h
> +++ b/net/core/dev.h
> @@ -189,5 +189,6 @@ int dev_set_hwtstamp_phylib(struct net_device *dev,
> struct netlink_ext_ack *extack);
> int dev_get_hwtstamp_phylib(struct net_device *dev,
> struct kernel_hwtstamp_config *cfg);
> +int net_hwtstamp_validate(const struct kernel_hwtstamp_config *cfg);
>
> #endif
> diff --git a/net/core/dev_ioctl.c b/net/core/dev_ioctl.c
> index b8cf8c55fa2d..6aaa8326bf8f 100644
> --- a/net/core/dev_ioctl.c
> +++ b/net/core/dev_ioctl.c
> @@ -184,7 +184,7 @@ static int dev_ifsioc_locked(struct net *net, struct ifreq *ifr, unsigned int cm
> return err;
> }
>
> -static int net_hwtstamp_validate(const struct kernel_hwtstamp_config *cfg)
> +int net_hwtstamp_validate(const struct kernel_hwtstamp_config *cfg)
> {
> enum hwtstamp_tx_types tx_type;
> enum hwtstamp_rx_filters rx_filter;
>
^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: [PATCH net-next v17 04/14] net: Change the API of PHY default timestamp to MAC
2024-07-09 13:53 ` [PATCH net-next v17 04/14] net: Change the API of PHY default timestamp to MAC Kory Maincent
@ 2024-07-15 23:37 ` Jacob Keller
2024-07-27 13:44 ` Kory Maincent
0 siblings, 1 reply; 39+ messages in thread
From: Jacob Keller @ 2024-07-15 23:37 UTC (permalink / raw)
To: Kory Maincent, Florian Fainelli,
Broadcom internal kernel review list, Andrew Lunn,
Heiner Kallweit, Russell King, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, Richard Cochran, Radu Pirea,
Jay Vosburgh, Andy Gospodarek, Nicolas Ferre, Claudiu Beznea,
Willem de Bruijn, Jonathan Corbet, Horatiu Vultur, UNGLinuxDriver,
Simon Horman, Vladimir Oltean, donald.hunter, danieller,
ecree.xilinx
Cc: Thomas Petazzoni, linux-kernel, netdev, linux-doc,
Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter
On 7/9/2024 6:53 AM, Kory Maincent wrote:
> Change the API to select MAC default time stamping instead of the PHY.
> Indeed the PHY is closer to the wire therefore theoretically it has less
> delay than the MAC timestamping but the reality is different. Due to lower
> time stamping clock frequency, latency in the MDIO bus and no PHC hardware
> synchronization between different PHY, the PHY PTP is often less precise
> than the MAC. The exception is for PHY designed specially for PTP case but
> these devices are not very widespread. For not breaking the compatibility
> default_timestamp flag has been introduced in phy_device that is set by
> the phy driver to know we are using the old API behavior.
>
This description feels like it is making a pretty broad generalization
about devices. The specifics of whether MAC or PHY timestamping is
better will be device dependent.
It looks like you introduce a default_timestamp flag to ensure existing
devices default to PHY? I assume your goal here is to discourage this
and not allow setting it for new devices? Or do we want to let device
driver authors decide which is a better default?
> Reviewed-by: Rahul Rameshbabu <rrameshbabu@nvidia.com>
> Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
Overall this makes sense, with a couple questions I had during review.
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
> ---
> diff --git a/include/linux/phy.h b/include/linux/phy.h
> index bd68f9d8e74f..e7a38137211c 100644
> --- a/include/linux/phy.h
> +++ b/include/linux/phy.h
> @@ -616,6 +616,8 @@ struct macsec_ops;
> * handling shall be postponed until PHY has resumed
> * @irq_rerun: Flag indicating interrupts occurred while PHY was suspended,
> * requiring a rerun of the interrupt handler after resume
> + * @default_timestamp: Flag indicating whether we are using the phy
> + * timestamp as the default one
This is clearly intended to ensure existing drivers maintain legacy
behavior. But what is our policy going forward for new devices? Do we
want to leave it up to PHY driver authors?
> * @interface: enum phy_interface_t value
> * @possible_interfaces: bitmap if interface modes that the attached PHY
> * will switch between depending on media speed.
> @@ -681,6 +683,8 @@ struct phy_device {
> unsigned irq_suspended:1;
> unsigned irq_rerun:1;
>
> + unsigned default_timestamp:1;
> +
> int rate_matching;
>
> enum phy_state state;
> @@ -1625,6 +1629,21 @@ static inline void phy_txtstamp(struct phy_device *phydev, struct sk_buff *skb,
> phydev->mii_ts->txtstamp(phydev->mii_ts, skb, type);
> }
>
> +/**
> + * phy_is_default_hwtstamp - Is the PHY hwtstamp the default timestamp
> + * @phydev: Pointer to phy_device
> + *
> + * This is used to get default timestamping device taking into account
> + * the new API choice, which is selecting the timestamping from MAC by
> + * default if the phydev does not have default_timestamp flag enabled.
> + *
> + * Return: True if phy is the default hw timestamp, false otherwise.
> + */
> +static inline bool phy_is_default_hwtstamp(struct phy_device *phydev)
> +{
> + return phy_has_hwtstamp(phydev) && phydev->default_timestamp;
> +}
> +
> /**
> * phy_is_internal - Convenience function for testing if a PHY is internal
> * @phydev: the phy_device struct
> diff --git a/net/core/dev_ioctl.c b/net/core/dev_ioctl.c
> index 6aaa8326bf8f..36cea843381f 100644
> --- a/net/core/dev_ioctl.c
> +++ b/net/core/dev_ioctl.c
> @@ -259,9 +259,7 @@ static int dev_eth_ioctl(struct net_device *dev,
> * @dev: Network device
> * @cfg: Timestamping configuration structure
> *
> - * Helper for enforcing a common policy that phylib timestamping, if available,
> - * should take precedence in front of hardware timestamping provided by the
> - * netdev.
> + * Helper for calling the default hardware provider timestamping.
> *
> * Note: phy_mii_ioctl() only handles SIOCSHWTSTAMP (not SIOCGHWTSTAMP), and
> * there only exists a phydev->mii_ts->hwtstamp() method. So this will return
> @@ -271,7 +269,7 @@ static int dev_eth_ioctl(struct net_device *dev,
> int dev_get_hwtstamp_phylib(struct net_device *dev,
> struct kernel_hwtstamp_config *cfg)
> {
> - if (phy_has_hwtstamp(dev->phydev))
> + if (phy_is_default_hwtstamp(dev->phydev))
> return phy_hwtstamp_get(dev->phydev, cfg);
>
> return dev->netdev_ops->ndo_hwtstamp_get(dev, cfg);
> @@ -327,7 +325,7 @@ int dev_set_hwtstamp_phylib(struct net_device *dev,
> struct netlink_ext_ack *extack)
> {
> const struct net_device_ops *ops = dev->netdev_ops;
> - bool phy_ts = phy_has_hwtstamp(dev->phydev);
> + bool phy_ts = phy_is_default_hwtstamp(dev->phydev);
> struct kernel_hwtstamp_config old_cfg = {};
> bool changed = false;
> int err;
> diff --git a/net/core/timestamping.c b/net/core/timestamping.c
> index 04840697fe79..3717fb152ecc 100644
> --- a/net/core/timestamping.c
> +++ b/net/core/timestamping.c
> @@ -25,7 +25,8 @@ void skb_clone_tx_timestamp(struct sk_buff *skb)
> struct sk_buff *clone;
> unsigned int type;
>
> - if (!skb->sk)
> + if (!skb->sk || !skb->dev ||
> + !phy_is_default_hwtstamp(skb->dev->phydev))
I don't follow why this check is added and its not calling something
like "phy_is_current_hwtstamp"? I guess because we don't yet have a way
to select between MAC/PHY at this point in the series? Ok.
> return;
>
> type = classify(skb);
> @@ -47,7 +48,7 @@ bool skb_defer_rx_timestamp(struct sk_buff *skb)
> struct mii_timestamper *mii_ts;
> unsigned int type;
>
> - if (!skb->dev || !skb->dev->phydev || !skb->dev->phydev->mii_ts)
> + if (!skb->dev || !phy_is_default_hwtstamp(skb->dev->phydev))
> return false;
>
> if (skb_headroom(skb) < ETH_HLEN)
> diff --git a/net/ethtool/common.c b/net/ethtool/common.c
> index 6b2a360dcdf0..01b7550f12c6 100644
> --- a/net/ethtool/common.c
> +++ b/net/ethtool/common.c
> @@ -637,7 +637,7 @@ int __ethtool_get_ts_info(struct net_device *dev, struct ethtool_ts_info *info)
> memset(info, 0, sizeof(*info));
> info->cmd = ETHTOOL_GET_TS_INFO;
>
> - if (phy_has_tsinfo(phydev))
> + if (phy_is_default_hwtstamp(phydev) && phy_has_tsinfo(phydev))
> return phy_ts_info(phydev, info);
> if (ops->get_ts_info)
> return ops->get_ts_info(dev, info);
>
^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: [PATCH net-next v17 05/14] net: net_tstamp: Add unspec field to hwtstamp_source enumeration
2024-07-09 13:53 ` [PATCH net-next v17 05/14] net: net_tstamp: Add unspec field to hwtstamp_source enumeration Kory Maincent
@ 2024-07-15 23:37 ` Jacob Keller
0 siblings, 0 replies; 39+ messages in thread
From: Jacob Keller @ 2024-07-15 23:37 UTC (permalink / raw)
To: Kory Maincent, Florian Fainelli,
Broadcom internal kernel review list, Andrew Lunn,
Heiner Kallweit, Russell King, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, Richard Cochran, Radu Pirea,
Jay Vosburgh, Andy Gospodarek, Nicolas Ferre, Claudiu Beznea,
Willem de Bruijn, Jonathan Corbet, Horatiu Vultur, UNGLinuxDriver,
Simon Horman, Vladimir Oltean, donald.hunter, danieller,
ecree.xilinx
Cc: Thomas Petazzoni, linux-kernel, netdev, linux-doc,
Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter
On 7/9/2024 6:53 AM, Kory Maincent wrote:
> Prepare for future support of saving hwtstamp source in PTP xarray by
> introducing HWTSTAMP_SOURCE_UNSPEC to hwtstamp_source enum, setting it
> to 0 to match old behavior of no source defined.
>
> Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
> ---
>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
> Change in v8:
> - New patch
> ---
> include/linux/net_tstamp.h | 1 +
> 1 file changed, 1 insertion(+)
>
> diff --git a/include/linux/net_tstamp.h b/include/linux/net_tstamp.h
> index 3799c79b6c83..662074b08c94 100644
> --- a/include/linux/net_tstamp.h
> +++ b/include/linux/net_tstamp.h
> @@ -14,6 +14,7 @@
> SOF_TIMESTAMPING_RAW_HARDWARE)
>
> enum hwtstamp_source {
> + HWTSTAMP_SOURCE_UNSPEC,
> HWTSTAMP_SOURCE_NETDEV,
> HWTSTAMP_SOURCE_PHYLIB,
> };
>
^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: [PATCH net-next v17 06/14] net: Add struct kernel_ethtool_ts_info
2024-07-09 13:53 ` [PATCH net-next v17 06/14] net: Add struct kernel_ethtool_ts_info Kory Maincent
@ 2024-07-15 23:41 ` Jacob Keller
0 siblings, 0 replies; 39+ messages in thread
From: Jacob Keller @ 2024-07-15 23:41 UTC (permalink / raw)
To: Kory Maincent, Florian Fainelli,
Broadcom internal kernel review list, Andrew Lunn,
Heiner Kallweit, Russell King, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, Richard Cochran, Radu Pirea,
Jay Vosburgh, Andy Gospodarek, Nicolas Ferre, Claudiu Beznea,
Willem de Bruijn, Jonathan Corbet, Horatiu Vultur, UNGLinuxDriver,
Simon Horman, Vladimir Oltean, donald.hunter, danieller,
ecree.xilinx
Cc: Thomas Petazzoni, linux-kernel, netdev, linux-doc,
Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter
On 7/9/2024 6:53 AM, Kory Maincent wrote:
> In prevision to add new UAPI for hwtstamp we will be limited to the struct
> ethtool_ts_info that is currently passed in fixed binary format through the
> ETHTOOL_GET_TS_INFO ethtool ioctl. It would be good if new kernel code
> already started operating on an extensible kernel variant of that
> structure, similar in concept to struct kernel_hwtstamp_config vs struct
> hwtstamp_config.
>
> Since struct ethtool_ts_info is in include/uapi/linux/ethtool.h, here
> we introduce the kernel-only structure in include/linux/ethtool.h.
> The manual copy is then made in the function called by ETHTOOL_GET_TS_INFO.
>
> Acked-by: Shannon Nelson <shannon.nelson@amd.com>
> Acked-by: Alexandra Winter <wintera@linux.ibm.com>
> Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
> ---
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
Also for the Intel driver changes specifically:
Acked-by: Jacob Keller <jacob.e.keller@intel.com>
^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: [PATCH net-next v17 07/14] ptp: Add phc source and helpers to register specific PTP clock or get information
2024-07-09 13:53 ` [PATCH net-next v17 07/14] ptp: Add phc source and helpers to register specific PTP clock or get information Kory Maincent
@ 2024-07-15 23:44 ` Jacob Keller
0 siblings, 0 replies; 39+ messages in thread
From: Jacob Keller @ 2024-07-15 23:44 UTC (permalink / raw)
To: Kory Maincent, Florian Fainelli,
Broadcom internal kernel review list, Andrew Lunn,
Heiner Kallweit, Russell King, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, Richard Cochran, Radu Pirea,
Jay Vosburgh, Andy Gospodarek, Nicolas Ferre, Claudiu Beznea,
Willem de Bruijn, Jonathan Corbet, Horatiu Vultur, UNGLinuxDriver,
Simon Horman, Vladimir Oltean, donald.hunter, danieller,
ecree.xilinx
Cc: Thomas Petazzoni, linux-kernel, netdev, linux-doc,
Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter
On 7/9/2024 6:53 AM, Kory Maincent wrote:
> Prepare for future hardware timestamp selection by adding source and
> corresponding pointers to ptp_clock structure.
> Additionally, introduce helpers for registering specific phydev or netdev
> PTP clocks, retrieving PTP clock information such as hwtstamp source or
> phydev/netdev pointers, and obtaining the ptp_clock structure from the
> phc index.
> These helpers are added to a new ptp_clock_consumer.c file, built as
> builtin. This is necessary because these helpers will be called by
> ethtool or net timestamping, which are builtin code.
>
> Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
> ---
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: [PATCH net-next v17 08/14] net: Add the possibility to support a selected hwtstamp in netdevice
2024-07-09 13:53 ` [PATCH net-next v17 08/14] net: Add the possibility to support a selected hwtstamp in netdevice Kory Maincent
@ 2024-07-17 17:22 ` Jacob Keller
0 siblings, 0 replies; 39+ messages in thread
From: Jacob Keller @ 2024-07-17 17:22 UTC (permalink / raw)
To: Kory Maincent, Florian Fainelli,
Broadcom internal kernel review list, Andrew Lunn,
Heiner Kallweit, Russell King, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, Richard Cochran, Radu Pirea,
Jay Vosburgh, Andy Gospodarek, Nicolas Ferre, Claudiu Beznea,
Willem de Bruijn, Jonathan Corbet, Horatiu Vultur, UNGLinuxDriver,
Simon Horman, Vladimir Oltean, donald.hunter, danieller,
ecree.xilinx
Cc: Thomas Petazzoni, linux-kernel, netdev, linux-doc,
Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter
On 7/9/2024 6:53 AM, Kory Maincent wrote:
> Introduce the description of a hwtstamp provider, mainly defined with a
> ptp_clock pointer and a qualifier value.
>
> Add a hwtstamp provider description within the netdev structure to
> allow saving the hwtstamp we want to use. This prepares for future
> support of an ethtool netlink command to select the desired hwtstamp
> provider. By default, the old API that does not support hwtstamp
> selectability is used, meaning the hwtstamp ptp_clock pointer is unset.
>
> Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
> ---
>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: [PATCH net-next v17 09/14] net: netdevsim: ptp_mock: Convert to netdev_ptp_clock_register
2024-07-09 13:53 ` [PATCH net-next v17 09/14] net: netdevsim: ptp_mock: Convert to netdev_ptp_clock_register Kory Maincent
@ 2024-07-17 17:23 ` Jacob Keller
0 siblings, 0 replies; 39+ messages in thread
From: Jacob Keller @ 2024-07-17 17:23 UTC (permalink / raw)
To: Kory Maincent, Florian Fainelli,
Broadcom internal kernel review list, Andrew Lunn,
Heiner Kallweit, Russell King, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, Richard Cochran, Radu Pirea,
Jay Vosburgh, Andy Gospodarek, Nicolas Ferre, Claudiu Beznea,
Willem de Bruijn, Jonathan Corbet, Horatiu Vultur, UNGLinuxDriver,
Simon Horman, Vladimir Oltean, donald.hunter, danieller,
ecree.xilinx
Cc: Thomas Petazzoni, linux-kernel, netdev, linux-doc,
Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter
On 7/9/2024 6:53 AM, Kory Maincent wrote:
> The hardware registration clock for net device is now using
> netdev_ptp_clock_register to save the net_device pointer within the PTP
> clock xarray. netdevsim is registering its ptp through the mock driver.
> It is the only driver using the mock driver to register a ptp clock.
> Convert the mock driver to the new API.
>
> Reviewed-by: Rahul Rameshbabu <rrameshbabu@nvidia.com>
> Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
> ---
> Changes in v8:
> - New patch
> ---
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: [PATCH net-next v17 10/14] net: macb: Convert to netdev_ptp_clock_register
2024-07-09 13:53 ` [PATCH net-next v17 10/14] net: macb: " Kory Maincent
@ 2024-07-17 17:24 ` Jacob Keller
0 siblings, 0 replies; 39+ messages in thread
From: Jacob Keller @ 2024-07-17 17:24 UTC (permalink / raw)
To: Kory Maincent, Florian Fainelli,
Broadcom internal kernel review list, Andrew Lunn,
Heiner Kallweit, Russell King, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, Richard Cochran, Radu Pirea,
Jay Vosburgh, Andy Gospodarek, Nicolas Ferre, Claudiu Beznea,
Willem de Bruijn, Jonathan Corbet, Horatiu Vultur, UNGLinuxDriver,
Simon Horman, Vladimir Oltean, donald.hunter, danieller,
ecree.xilinx
Cc: Thomas Petazzoni, linux-kernel, netdev, linux-doc,
Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter
On 7/9/2024 6:53 AM, Kory Maincent wrote:
> The hardware registration clock for net device is now using
> netdev_ptp_clock_register to save the net_device pointer within the ptp
> clock xarray. Convert the macb driver to the new API.
>
> Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
> ---
>
> Change in v8:
> - New patch
> ---
> drivers/net/ethernet/cadence/macb_ptp.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/net/ethernet/cadence/macb_ptp.c b/drivers/net/ethernet/cadence/macb_ptp.c
> index a63bf29c4fa8..50fa62a0ddc5 100644
> --- a/drivers/net/ethernet/cadence/macb_ptp.c
> +++ b/drivers/net/ethernet/cadence/macb_ptp.c
> @@ -332,7 +332,7 @@ void gem_ptp_init(struct net_device *dev)
> bp->tsu_rate = bp->ptp_info->get_tsu_rate(bp);
> bp->ptp_clock_info.max_adj = bp->ptp_info->get_ptp_max_adj();
> gem_ptp_init_timer(bp);
> - bp->ptp_clock = ptp_clock_register(&bp->ptp_clock_info, &dev->dev);
> + bp->ptp_clock = netdev_ptp_clock_register(&bp->ptp_clock_info, dev);
> if (IS_ERR(bp->ptp_clock)) {
> pr_err("ptp clock register failed: %ld\n",
> PTR_ERR(bp->ptp_clock));
>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: [PATCH net-next v17 11/14] net: ptp: Move ptp_clock_index() to builtin symbol
2024-07-09 13:53 ` [PATCH net-next v17 11/14] net: ptp: Move ptp_clock_index() to builtin symbol Kory Maincent
@ 2024-07-17 17:24 ` Jacob Keller
0 siblings, 0 replies; 39+ messages in thread
From: Jacob Keller @ 2024-07-17 17:24 UTC (permalink / raw)
To: Kory Maincent, Florian Fainelli,
Broadcom internal kernel review list, Andrew Lunn,
Heiner Kallweit, Russell King, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, Richard Cochran, Radu Pirea,
Jay Vosburgh, Andy Gospodarek, Nicolas Ferre, Claudiu Beznea,
Willem de Bruijn, Jonathan Corbet, Horatiu Vultur, UNGLinuxDriver,
Simon Horman, Vladimir Oltean, donald.hunter, danieller,
ecree.xilinx
Cc: Thomas Petazzoni, linux-kernel, netdev, linux-doc,
Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter
On 7/9/2024 6:53 AM, Kory Maincent wrote:
> Move ptp_clock_index() to builtin symbols to prepare for supporting get
> and set hardware timestamps from ethtool, which is builtin.
>
> Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
> ---
>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
> Change in v13:
> - New patch
> ---
> drivers/ptp/ptp_clock.c | 6 ------
> drivers/ptp/ptp_clock_consumer.c | 6 ++++++
> 2 files changed, 6 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/ptp/ptp_clock.c b/drivers/ptp/ptp_clock.c
> index 593b5c906314..fc4b266abe1d 100644
> --- a/drivers/ptp/ptp_clock.c
> +++ b/drivers/ptp/ptp_clock.c
> @@ -460,12 +460,6 @@ void ptp_clock_event(struct ptp_clock *ptp, struct ptp_clock_event *event)
> }
> EXPORT_SYMBOL(ptp_clock_event);
>
> -int ptp_clock_index(struct ptp_clock *ptp)
> -{
> - return ptp->index;
> -}
> -EXPORT_SYMBOL(ptp_clock_index);
> -
> int ptp_find_pin(struct ptp_clock *ptp,
> enum ptp_pin_function func, unsigned int chan)
> {
> diff --git a/drivers/ptp/ptp_clock_consumer.c b/drivers/ptp/ptp_clock_consumer.c
> index f5fab1c14b47..f521b07da231 100644
> --- a/drivers/ptp/ptp_clock_consumer.c
> +++ b/drivers/ptp/ptp_clock_consumer.c
> @@ -108,3 +108,9 @@ void remove_hwtstamp_provider(struct rcu_head *rcu_head)
> kfree(hwtstamp);
> }
> EXPORT_SYMBOL(remove_hwtstamp_provider);
> +
> +int ptp_clock_index(struct ptp_clock *ptp)
> +{
> + return ptp->index;
> +}
> +EXPORT_SYMBOL(ptp_clock_index);
>
^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: [PATCH net-next v17 12/14] net: ethtool: tsinfo: Add support for reading tsinfo for a specific hwtstamp provider
2024-07-09 13:53 ` [PATCH net-next v17 12/14] net: ethtool: tsinfo: Add support for reading tsinfo for a specific hwtstamp provider Kory Maincent
@ 2024-07-17 17:35 ` Jacob Keller
2024-07-26 19:04 ` Kory Maincent
0 siblings, 1 reply; 39+ messages in thread
From: Jacob Keller @ 2024-07-17 17:35 UTC (permalink / raw)
To: Kory Maincent, Florian Fainelli,
Broadcom internal kernel review list, Andrew Lunn,
Heiner Kallweit, Russell King, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, Richard Cochran, Radu Pirea,
Jay Vosburgh, Andy Gospodarek, Nicolas Ferre, Claudiu Beznea,
Willem de Bruijn, Jonathan Corbet, Horatiu Vultur, UNGLinuxDriver,
Simon Horman, Vladimir Oltean, donald.hunter, danieller,
ecree.xilinx
Cc: Thomas Petazzoni, linux-kernel, netdev, linux-doc,
Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter
On 7/9/2024 6:53 AM, Kory Maincent wrote:
> Either the MAC or the PHY can provide hwtstamp, so we should be able to
> read the tsinfo for any hwtstamp provider.
>
> Enhance 'get' command to retrieve tsinfo of hwtstamp providers within a
> network topology.
>
> Add support for a specific dump command to retrieve all hwtstamp
> providers within the network topology, with added functionality for
> filtered dump to target a single interface.
>
> Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
> ---
>
> Pointer attached_dev is used to know if the phy is on the net topology.
> This might not be enough and might need Maxime Chevallier link topology
> patch series:
> https://lore.kernel.org/netdev/20240213150431.1796171-1-maxime.chevallier@bootlin.com/
>
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
One thing which applies more broadly to the whole series, but I see the
focus right now is on selecting between NETDEV and PHYLIB.
For ice (E800 series) hardware, the timestamps are captured by the PHY,
but its not managed by phylib, its managed by firmware. In our case we
would obviously report NETDEV in this case. The hardware only has one
timestamp point and the fact that it happens at the PHY layer is not
relevant since you can't select or change it.
There are some future plans in the work for hardware based on the ixgbe
driver which could timestamp at either the MAC or PHY (with varying
trade-offs in precision vs what can be timestamped), and (perhaps
unfortunately), the PHY would likely not manageable by phylib.
There is also the possibility of something like DMA or completion
timestamps which are distinct from MAC timestamps. But again can have
varying trade offs.
I'm hopeful this work can be extended somehow to enable selection
between the different mechanisms, even when the kernel device being
represented is the same netdev.
^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: [PATCH net-next v17 13/14] net: ethtool: Add support for tsconfig command to get/set hwtstamp config
2024-07-09 13:53 ` [PATCH net-next v17 13/14] net: ethtool: Add support for tsconfig command to get/set hwtstamp config Kory Maincent
2024-07-15 14:59 ` Jakub Kicinski
@ 2024-07-17 17:43 ` Jacob Keller
2024-07-27 13:00 ` Kory Maincent
1 sibling, 1 reply; 39+ messages in thread
From: Jacob Keller @ 2024-07-17 17:43 UTC (permalink / raw)
To: Kory Maincent, Florian Fainelli,
Broadcom internal kernel review list, Andrew Lunn,
Heiner Kallweit, Russell King, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, Richard Cochran, Radu Pirea,
Jay Vosburgh, Andy Gospodarek, Nicolas Ferre, Claudiu Beznea,
Willem de Bruijn, Jonathan Corbet, Horatiu Vultur, UNGLinuxDriver,
Simon Horman, Vladimir Oltean, donald.hunter, danieller,
ecree.xilinx
Cc: Thomas Petazzoni, linux-kernel, netdev, linux-doc,
Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter
On 7/9/2024 6:53 AM, Kory Maincent wrote:
> Introduce support for ETHTOOL_MSG_TSCONFIG_GET/SET ethtool netlink socket
> to read and configure hwtstamp configuration of a PHC provider. Note that
> simultaneous hwtstamp isn't supported; configuring a new one disables the
> previous setting.
>
> Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
> ---
>
> Changes in v16:
> - Add a new patch to separate tsinfo into a new tsconfig command to get
> and set the hwtstamp config.
>
> Changes in v17:
> - Fix a doc misalignment.
> ---
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
> +The legacy configuration is the use of the ioctl(SIOCSHWTSTAMP) with a pointer
> +to a struct ifreq whose ifr_data points to a struct hwtstamp_config.
> +The tx_type and rx_filter are hints to the driver what it is expected to do.
> +If the requested fine-grained filtering for incoming packets is not
> supported, the driver may time stamp more than just the requested types
> of packets.
>
Does the core automatically handle SIOCSHWTSTAMP and SIOCGHWTSTAMP in
terms of the new API? I'm guessing yes because of the new
.ndo_set_hwtstamp ops?
> @@ -531,9 +536,12 @@ Only a processes with admin rights may change the configuration. User
> space is responsible to ensure that multiple processes don't interfere
> with each other and that the settings are reset.
>
> -Any process can read the actual configuration by passing this
> -structure to ioctl(SIOCGHWTSTAMP) in the same way. However, this has
> -not been implemented in all drivers.
> +Any process can read the actual configuration by requesting tsconfig netlink
> +socket ETHTOOL_MSG_TSCONFIG_GET.
> +
> +The legacy usage is to pass this structure to ioctl(SIOCGHWTSTAMP) in the
> +same way as the ioctl(SIOCSHWTSTAMP). However, this has not been implemented
> +in all drivers.
>
> ::
>
> @@ -578,9 +586,10 @@ not been implemented in all drivers.
> --------------------------------------------------------
>
> A driver which supports hardware time stamping must support the
> -SIOCSHWTSTAMP ioctl and update the supplied struct hwtstamp_config with
> -the actual values as described in the section on SIOCSHWTSTAMP. It
> -should also support SIOCGHWTSTAMP.
> +ndo_hwtstamp_set NDO or the legacy SIOCSHWTSTAMP ioctl and update the
> +supplied struct hwtstamp_config with the actual values as described in
> +the section on SIOCSHWTSTAMP. It should also support ndo_hwtstamp_get or
> +the legacy SIOCGHWTSTAMP.
Can we simply drop the mention of implementing the legacy implementation
on the kernel side? I guess not all existing drivers have converted yet...?
I have a similar thought about the other legacy PTP hooks.. it is good
to completely remove the legacy/deprecated implementations as it means
drivers can't be published which don't update to new APIs. That
ultimately just wastes reviewer/maintainer time to point out that it
must be updated to new APIs.
Obviously this will require some effort to make sure all existing
drivers get refactored.
^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: [PATCH net-next v17 14/14] netlink: specs: Enhance tsinfo netlink attributes and add a tsconfig set command
2024-07-09 13:53 ` [PATCH net-next v17 14/14] netlink: specs: Enhance tsinfo netlink attributes and add a tsconfig set command Kory Maincent
@ 2024-07-17 17:44 ` Jacob Keller
0 siblings, 0 replies; 39+ messages in thread
From: Jacob Keller @ 2024-07-17 17:44 UTC (permalink / raw)
To: Kory Maincent, Florian Fainelli,
Broadcom internal kernel review list, Andrew Lunn,
Heiner Kallweit, Russell King, David S. Miller, Eric Dumazet,
Jakub Kicinski, Paolo Abeni, Richard Cochran, Radu Pirea,
Jay Vosburgh, Andy Gospodarek, Nicolas Ferre, Claudiu Beznea,
Willem de Bruijn, Jonathan Corbet, Horatiu Vultur, UNGLinuxDriver,
Simon Horman, Vladimir Oltean, donald.hunter, danieller,
ecree.xilinx
Cc: Thomas Petazzoni, linux-kernel, netdev, linux-doc,
Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter
On 7/9/2024 6:53 AM, Kory Maincent wrote:
> Add new attributed to tsinfo allowing to get the tsinfo from a phc provider
> (composed by a phc index and a phc qualifier) on a netdevice's link.
> Add simultaneously a tsconfig command to be able to get and set hwtstamp
> configuration for a specified phc provider.
>
> Here is few examples:
> ./ynl/cli.py --spec netlink/specs/ethtool.yaml --no-schema
> --dump tsinfo-get
> --json '{"header":{"dev-name":"eth0"}}'
> [{'header': {'dev-index': 3, 'dev-name': 'eth0'},
> 'hwtst-provider': {'index': 0, 'qualifier': 0},
> 'phc-index': 0,
> 'rx-filters': {'bits': {'bit': [{'index': 0, 'name': 'none'},
> {'index': 2, 'name': 'some'}]},
> 'nomask': True,
> 'size': 16},
> 'timestamping': {'bits': {'bit': [{'index': 0, 'name': 'hardware-transmit'},
> {'index': 2, 'name': 'hardware-receive'},
> {'index': 6,
> 'name': 'hardware-raw-clock'}]},
> 'nomask': True,
> 'size': 17},
> 'tx-types': {'bits': {'bit': [{'index': 0, 'name': 'off'},
> {'index': 1, 'name': 'on'}]},
> 'nomask': True,
> 'size': 4}},
> {'header': {'dev-index': 3, 'dev-name': 'eth0'},
> 'hwtst-provider': {'index': 2, 'qualifier': 0},
> 'phc-index': 2,
> 'rx-filters': {'bits': {'bit': [{'index': 0, 'name': 'none'},
> {'index': 1, 'name': 'all'}]},
> 'nomask': True,
> 'size': 16},
> 'timestamping': {'bits': {'bit': [{'index': 0, 'name': 'hardware-transmit'},
> {'index': 1, 'name': 'software-transmit'},
> {'index': 2, 'name': 'hardware-receive'},
> {'index': 3, 'name': 'software-receive'},
> {'index': 4,
> 'name': 'software-system-clock'},
> {'index': 6,
> 'name': 'hardware-raw-clock'}]},
> 'nomask': True,
> 'size': 17},
> 'tx-types': {'bits': {'bit': [{'index': 0, 'name': 'off'},
> {'index': 1, 'name': 'on'},
> {'index': 2, 'name': 'onestep-sync'}]},
> 'nomask': True,
> 'size': 4}}]
>
> ./ynl/cli.py --spec netlink/specs/ethtool.yaml --no-schema --do tsinfo-get
> --json '{"header":{"dev-name":"eth0"},
> "hwtst-provider":{"index":0, "qualifier":0 }
> }'
> {'header': {'dev-index': 3, 'dev-name': 'eth0'},
> 'hwtst-provider': {'index': 0, 'qualifier': 0},
> 'phc-index': 0,
> 'rx-filters': {'bits': {'bit': [{'index': 0, 'name': 'none'},
> {'index': 2, 'name': 'some'}]},
> 'nomask': True,
> 'size': 16},
> 'timestamping': {'bits': {'bit': [{'index': 0, 'name': 'hardware-transmit'},
> {'index': 2, 'name': 'hardware-receive'},
> {'index': 6, 'name': 'hardware-raw-clock'}]},
> 'nomask': True,
> 'size': 17},
> 'tx-types': {'bits': {'bit': [{'index': 0, 'name': 'off'},
> {'index': 1, 'name': 'on'}]},
> 'nomask': True,
> 'size': 4}}
>
> ./ynl/cli.py --spec netlink/specs/ethtool.yaml --no-schema --do tsinfo-set
> --json '{"header":{"dev-name":"eth0"},
> "hwtst-provider":{"index":2, "qualifier":0}}'
> None
> ./ynl/cli.py --spec netlink/specs/ethtool.yaml --no-schema --do tsconfig-get
> --json '{"header":{"dev-name":"eth0"}}'
> {'header': {'dev-index': 3, 'dev-name': 'eth0'},
> 'hwtstamp-flags': 1,
> 'hwtstamp-provider': {'index': 1, 'qualifier': 0},
> 'rx-filters': {'bits': {'bit': [{'index': 12, 'name': 'ptpv2-event'}]},
> 'nomask': True,
> 'size': 16},
> 'tx-types': {'bits': {'bit': [{'index': 1, 'name': 'on'}]},
> 'nomask': True,
> 'size': 4}}
>
> ./ynl/cli.py --spec netlink/specs/ethtool.yaml --no-schema --do tsconfig-set
> --json '{"header":{"dev-name":"eth0"},
> "hwtstamp-provider":{"index":1, "qualifier":0 },
> "rx-filters":{"bits": {"bit": {"name":"ptpv2-l4-event"}},
> "nomask": 1},
> "tx-types":{"bits": {"bit": {"name":"on"}},
> "nomask": 1}}'
> None
>
> Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
> ---
Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: [PATCH net-next v17 12/14] net: ethtool: tsinfo: Add support for reading tsinfo for a specific hwtstamp provider
2024-07-17 17:35 ` Jacob Keller
@ 2024-07-26 19:04 ` Kory Maincent
2024-07-29 17:58 ` Jacob Keller
0 siblings, 1 reply; 39+ messages in thread
From: Kory Maincent @ 2024-07-26 19:04 UTC (permalink / raw)
To: Jacob Keller
Cc: Florian Fainelli, Broadcom internal kernel review list,
Andrew Lunn, Heiner Kallweit, Russell King, David S. Miller,
Eric Dumazet, Jakub Kicinski, Paolo Abeni, Richard Cochran,
Radu Pirea, Jay Vosburgh, Andy Gospodarek, Nicolas Ferre,
Claudiu Beznea, Willem de Bruijn, Jonathan Corbet, Horatiu Vultur,
UNGLinuxDriver, Simon Horman, Vladimir Oltean, donald.hunter,
danieller, ecree.xilinx, Thomas Petazzoni, linux-kernel, netdev,
linux-doc, Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter
Hello Jacob,
Thanks a lot for your full review!
On Wed, 17 Jul 2024 10:35:20 -0700
Jacob Keller <jacob.e.keller@intel.com> wrote:
> On 7/9/2024 6:53 AM, Kory Maincent wrote:
> [...]
>
> Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
>
> One thing which applies more broadly to the whole series, but I see the
> focus right now is on selecting between NETDEV and PHYLIB.
>
> For ice (E800 series) hardware, the timestamps are captured by the PHY,
> but its not managed by phylib, its managed by firmware. In our case we
> would obviously report NETDEV in this case. The hardware only has one
> timestamp point and the fact that it happens at the PHY layer is not
> relevant since you can't select or change it.
>
> There are some future plans in the work for hardware based on the ixgbe
> driver which could timestamp at either the MAC or PHY (with varying
> trade-offs in precision vs what can be timestamped), and (perhaps
> unfortunately), the PHY would likely not manageable by phylib.
>
> There is also the possibility of something like DMA or completion
> timestamps which are distinct from MAC timestamps. But again can have
> varying trade offs.
As we already discussed in older version of this patch series the
hwtstamp qualifier will be used to select between IEEE 1588 timestamp or DMA
timestamp. See patch 8 :
+/*
+ * Possible type of htstamp provider. Mainly "precise" the default one
+ * is for IEEE 1588 quality and "approx" is for NICs DMA point.
+ */
We could add other enumeration values in the future if needed, to manage new
cases.
Just figured out there is a NIT in the doc. h*w*tstamp.
> I'm hopeful this work can be extended somehow to enable selection
> between the different mechanisms, even when the kernel device being
> represented is the same netdev.
Another nice features would be the support for simultaneous hardware timestamp
but I sadly won't be able to work on this.
Regards,
--
Köry Maincent, Bootlin
Embedded Linux and kernel engineering
https://bootlin.com
^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: [PATCH net-next v17 13/14] net: ethtool: Add support for tsconfig command to get/set hwtstamp config
2024-07-17 17:43 ` Jacob Keller
@ 2024-07-27 13:00 ` Kory Maincent
2024-07-29 18:02 ` Jacob Keller
0 siblings, 1 reply; 39+ messages in thread
From: Kory Maincent @ 2024-07-27 13:00 UTC (permalink / raw)
To: Jacob Keller
Cc: Florian Fainelli, Broadcom internal kernel review list,
Andrew Lunn, Heiner Kallweit, Russell King, David S. Miller,
Eric Dumazet, Jakub Kicinski, Paolo Abeni, Richard Cochran,
Radu Pirea, Jay Vosburgh, Andy Gospodarek, Nicolas Ferre,
Claudiu Beznea, Willem de Bruijn, Jonathan Corbet, Horatiu Vultur,
UNGLinuxDriver, Simon Horman, Vladimir Oltean, donald.hunter,
danieller, ecree.xilinx, Thomas Petazzoni, linux-kernel, netdev,
linux-doc, Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter
On Wed, 17 Jul 2024 10:43:05 -0700
Jacob Keller <jacob.e.keller@intel.com> wrote:
>
> > +The legacy configuration is the use of the ioctl(SIOCSHWTSTAMP) with a
> > pointer +to a struct ifreq whose ifr_data points to a struct
> > hwtstamp_config. +The tx_type and rx_filter are hints to the driver what it
> > is expected to do. +If the requested fine-grained filtering for incoming
> > packets is not supported, the driver may time stamp more than just the
> > requested types of packets.
> >
>
> Does the core automatically handle SIOCSHWTSTAMP and SIOCGHWTSTAMP in
> terms of the new API? I'm guessing yes because of the new
> .ndo_set_hwtstamp ops?
Yes.
> > A driver which supports hardware time stamping must support the
> > -SIOCSHWTSTAMP ioctl and update the supplied struct hwtstamp_config with
> > -the actual values as described in the section on SIOCSHWTSTAMP. It
> > -should also support SIOCGHWTSTAMP.
> > +ndo_hwtstamp_set NDO or the legacy SIOCSHWTSTAMP ioctl and update the
> > +supplied struct hwtstamp_config with the actual values as described in
> > +the section on SIOCSHWTSTAMP. It should also support ndo_hwtstamp_get or
> > +the legacy SIOCGHWTSTAMP.
>
> Can we simply drop the mention of implementing the legacy implementation
> on the kernel side? I guess not all existing drivers have converted yet...?
Yes indeed.
In fact, Vlad has already worked on converting all the existing drivers:
https://github.com/vladimiroltean/linux/tree/ndo-hwtstamp-v9
I can't find any patch series sent to net next. Vlad what is the status on this?
> I have a similar thought about the other legacy PTP hooks.. it is good
> to completely remove the legacy/deprecated implementations as it means
> drivers can't be published which don't update to new APIs. That
> ultimately just wastes reviewer/maintainer time to point out that it
> must be updated to new APIs.
Yes but on the userspace side linuxPTP is still using the IOCTLs uAPI that will
become legacy with this series. Maybe it is still a bit early to remove totally
their descriptions in the doc?
Regards,
--
Köry Maincent, Bootlin
Embedded Linux and kernel engineering
https://bootlin.com
^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: [PATCH net-next v17 04/14] net: Change the API of PHY default timestamp to MAC
2024-07-15 23:37 ` Jacob Keller
@ 2024-07-27 13:44 ` Kory Maincent
2024-07-29 18:08 ` Jacob Keller
0 siblings, 1 reply; 39+ messages in thread
From: Kory Maincent @ 2024-07-27 13:44 UTC (permalink / raw)
To: Jacob Keller
Cc: Florian Fainelli, Broadcom internal kernel review list,
Andrew Lunn, Heiner Kallweit, Russell King, David S. Miller,
Eric Dumazet, Jakub Kicinski, Paolo Abeni, Richard Cochran,
Radu Pirea, Jay Vosburgh, Andy Gospodarek, Nicolas Ferre,
Claudiu Beznea, Willem de Bruijn, Jonathan Corbet, Horatiu Vultur,
UNGLinuxDriver, Simon Horman, Vladimir Oltean, donald.hunter,
danieller, ecree.xilinx, Thomas Petazzoni, linux-kernel, netdev,
linux-doc, Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter
On Mon, 15 Jul 2024 16:37:01 -0700
Jacob Keller <jacob.e.keller@intel.com> wrote:
> On 7/9/2024 6:53 AM, Kory Maincent wrote:
> > Change the API to select MAC default time stamping instead of the PHY.
> > Indeed the PHY is closer to the wire therefore theoretically it has less
> > delay than the MAC timestamping but the reality is different. Due to lower
> > time stamping clock frequency, latency in the MDIO bus and no PHC hardware
> > synchronization between different PHY, the PHY PTP is often less precise
> > than the MAC. The exception is for PHY designed specially for PTP case but
> > these devices are not very widespread. For not breaking the compatibility
> > default_timestamp flag has been introduced in phy_device that is set by
> > the phy driver to know we are using the old API behavior.
> >
>
> This description feels like it is making a pretty broad generalization
> about devices. The specifics of whether MAC or PHY timestamping is
> better will be device dependent.
As explained, except for specific PTP specialized PHY, the MAC is better in
term of PTP precision.
This patch was a requisite from Russell, who wanted to add support for the PTP
in the marvell PHY. Doing so would select the PHY PTP by default which cause a
regression as the PHY hardware timestamp is less precise than the MAC.
https://lore.kernel.org/netdev/20200729105807.GZ1551@shell.armlinux.org.uk/
https://lore.kernel.org/netdev/Y%2F4DZIDm1d74MuFJ@shell.armlinux.org.uk/
There is also discussion on how to support it in older version of this series.
> It looks like you introduce a default_timestamp flag to ensure existing
> devices default to PHY? I assume your goal here is to discourage this
> and not allow setting it for new devices? Or do we want to let device
> driver authors decide which is a better default?
Yes to not change the old behavior the current PHY with PTP support will still
behave as default PTP. The point is indeed to discourage future drivers to
select the PHY as default PTP.
> > Reviewed-by: Rahul Rameshbabu <rrameshbabu@nvidia.com>
> > Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
>
> Overall this makes sense, with a couple questions I had during review.
>
> Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
>
> > ---
> > diff --git a/include/linux/phy.h b/include/linux/phy.h
> > index bd68f9d8e74f..e7a38137211c 100644
> > --- a/include/linux/phy.h
> > +++ b/include/linux/phy.h
> > @@ -616,6 +616,8 @@ struct macsec_ops;
> > * handling shall be postponed until PHY has resumed
> > * @irq_rerun: Flag indicating interrupts occurred while PHY was suspended,
> > * requiring a rerun of the interrupt handler after resume
> > + * @default_timestamp: Flag indicating whether we are using the phy
> > + * timestamp as the default one
>
> This is clearly intended to ensure existing drivers maintain legacy
> behavior. But what is our policy going forward for new devices? Do we
> want to leave it up to PHY driver authors?
Yes, new devices should not set this flag.
> > diff --git a/net/core/timestamping.c b/net/core/timestamping.c
> > index 04840697fe79..3717fb152ecc 100644
> > --- a/net/core/timestamping.c
> > +++ b/net/core/timestamping.c
> > @@ -25,7 +25,8 @@ void skb_clone_tx_timestamp(struct sk_buff *skb)
> > struct sk_buff *clone;
> > unsigned int type;
> >
> > - if (!skb->sk)
> > + if (!skb->sk || !skb->dev ||
> > + !phy_is_default_hwtstamp(skb->dev->phydev))
>
> I don't follow why this check is added and its not calling something
> like "phy_is_current_hwtstamp"? I guess because we don't yet have a way
> to select between MAC/PHY at this point in the series? Ok.
skb_clone_tx_timestamp is only used for PHY timestamping so we should do nothing
if the default PTP is the MAC.
Regards,
--
Köry Maincent, Bootlin
Embedded Linux and kernel engineering
https://bootlin.com
^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: [PATCH net-next v17 12/14] net: ethtool: tsinfo: Add support for reading tsinfo for a specific hwtstamp provider
2024-07-26 19:04 ` Kory Maincent
@ 2024-07-29 17:58 ` Jacob Keller
0 siblings, 0 replies; 39+ messages in thread
From: Jacob Keller @ 2024-07-29 17:58 UTC (permalink / raw)
To: Kory Maincent
Cc: Florian Fainelli, Broadcom internal kernel review list,
Andrew Lunn, Heiner Kallweit, Russell King, David S. Miller,
Eric Dumazet, Jakub Kicinski, Paolo Abeni, Richard Cochran,
Radu Pirea, Jay Vosburgh, Andy Gospodarek, Nicolas Ferre,
Claudiu Beznea, Willem de Bruijn, Jonathan Corbet, Horatiu Vultur,
UNGLinuxDriver, Simon Horman, Vladimir Oltean, donald.hunter,
danieller, ecree.xilinx, Thomas Petazzoni, linux-kernel, netdev,
linux-doc, Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter
On 7/26/2024 12:04 PM, Kory Maincent wrote:
> Hello Jacob,
>
> Thanks a lot for your full review!
>
> On Wed, 17 Jul 2024 10:35:20 -0700
> Jacob Keller <jacob.e.keller@intel.com> wrote:
>
>> On 7/9/2024 6:53 AM, Kory Maincent wrote:
>> [...]
>>
>> Reviewed-by: Jacob Keller <jacob.e.keller@intel.com>
>>
>> One thing which applies more broadly to the whole series, but I see the
>> focus right now is on selecting between NETDEV and PHYLIB.
>>
>> For ice (E800 series) hardware, the timestamps are captured by the PHY,
>> but its not managed by phylib, its managed by firmware. In our case we
>> would obviously report NETDEV in this case. The hardware only has one
>> timestamp point and the fact that it happens at the PHY layer is not
>> relevant since you can't select or change it.
>>
>> There are some future plans in the work for hardware based on the ixgbe
>> driver which could timestamp at either the MAC or PHY (with varying
>> trade-offs in precision vs what can be timestamped), and (perhaps
>> unfortunately), the PHY would likely not manageable by phylib.
>>
>> There is also the possibility of something like DMA or completion
>> timestamps which are distinct from MAC timestamps. But again can have
>> varying trade offs.
>
> As we already discussed in older version of this patch series the
> hwtstamp qualifier will be used to select between IEEE 1588 timestamp or DMA
> timestamp. See patch 8 :
> +/*
> + * Possible type of htstamp provider. Mainly "precise" the default one
> + * is for IEEE 1588 quality and "approx" is for NICs DMA point.
> + */
>
> We could add other enumeration values in the future if needed, to manage new
> cases.
>
> Just figured out there is a NIT in the doc. h*w*tstamp.
>
Ah, perfect, thanks for the clarification!
>> I'm hopeful this work can be extended somehow to enable selection
>> between the different mechanisms, even when the kernel device being
>> represented is the same netdev.
>
> Another nice features would be the support for simultaneous hardware timestamp
> but I sadly won't be able to work on this.
> > Regards,
Yes this would be useful, though I think we're somewhat limited by the
API that returns to userspace currently.
^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: [PATCH net-next v17 13/14] net: ethtool: Add support for tsconfig command to get/set hwtstamp config
2024-07-27 13:00 ` Kory Maincent
@ 2024-07-29 18:02 ` Jacob Keller
0 siblings, 0 replies; 39+ messages in thread
From: Jacob Keller @ 2024-07-29 18:02 UTC (permalink / raw)
To: Kory Maincent
Cc: Florian Fainelli, Broadcom internal kernel review list,
Andrew Lunn, Heiner Kallweit, Russell King, David S. Miller,
Eric Dumazet, Jakub Kicinski, Paolo Abeni, Richard Cochran,
Radu Pirea, Jay Vosburgh, Andy Gospodarek, Nicolas Ferre,
Claudiu Beznea, Willem de Bruijn, Jonathan Corbet, Horatiu Vultur,
UNGLinuxDriver, Simon Horman, Vladimir Oltean, donald.hunter,
danieller, ecree.xilinx, Thomas Petazzoni, linux-kernel, netdev,
linux-doc, Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter
On 7/27/2024 6:00 AM, Kory Maincent wrote:
> On Wed, 17 Jul 2024 10:43:05 -0700
> Jacob Keller <jacob.e.keller@intel.com> wrote:
>>> A driver which supports hardware time stamping must support the
>>> -SIOCSHWTSTAMP ioctl and update the supplied struct hwtstamp_config with
>>> -the actual values as described in the section on SIOCSHWTSTAMP. It
>>> -should also support SIOCGHWTSTAMP.
>>> +ndo_hwtstamp_set NDO or the legacy SIOCSHWTSTAMP ioctl and update the
>>> +supplied struct hwtstamp_config with the actual values as described in
>>> +the section on SIOCSHWTSTAMP. It should also support ndo_hwtstamp_get or
>>> +the legacy SIOCGHWTSTAMP.
>>
>> Can we simply drop the mention of implementing the legacy implementation
>> on the kernel side? I guess not all existing drivers have converted yet...?
>
> Yes indeed.>
> In fact, Vlad has already worked on converting all the existing drivers:
> https://github.com/vladimiroltean/linux/tree/ndo-hwtstamp-v9
> I can't find any patch series sent to net next. Vlad what is the status on this?
>
Great!
>> I have a similar thought about the other legacy PTP hooks.. it is good
>> to completely remove the legacy/deprecated implementations as it means
>> drivers can't be published which don't update to new APIs. That
>> ultimately just wastes reviewer/maintainer time to point out that it
>> must be updated to new APIs.
>
> Yes but on the userspace side linuxPTP is still using the IOCTLs uAPI that will
> become legacy with this series. Maybe it is still a bit early to remove totally
> their descriptions in the doc?
>
Right, they would need to use the netlink implementation to get the new
features, but the ioctls can at least be translated to the new kAPI
thats in the drivers?
Removing the old APIs from the uAPI doc is bad, but I think we can
clarify the wording of the doc and update to make it clear where the
separation is.
I may take a pass at the doc to see if I think I can improve it.
> Regards,
^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: [PATCH net-next v17 04/14] net: Change the API of PHY default timestamp to MAC
2024-07-27 13:44 ` Kory Maincent
@ 2024-07-29 18:08 ` Jacob Keller
2024-09-25 12:46 ` Kory Maincent
0 siblings, 1 reply; 39+ messages in thread
From: Jacob Keller @ 2024-07-29 18:08 UTC (permalink / raw)
To: Kory Maincent
Cc: Florian Fainelli, Broadcom internal kernel review list,
Andrew Lunn, Heiner Kallweit, Russell King, David S. Miller,
Eric Dumazet, Jakub Kicinski, Paolo Abeni, Richard Cochran,
Radu Pirea, Jay Vosburgh, Andy Gospodarek, Nicolas Ferre,
Claudiu Beznea, Willem de Bruijn, Jonathan Corbet, Horatiu Vultur,
UNGLinuxDriver, Simon Horman, Vladimir Oltean, donald.hunter,
danieller, ecree.xilinx, Thomas Petazzoni, linux-kernel, netdev,
linux-doc, Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter
On 7/27/2024 6:44 AM, Kory Maincent wrote:
> On Mon, 15 Jul 2024 16:37:01 -0700
> Jacob Keller <jacob.e.keller@intel.com> wrote:
>
>> On 7/9/2024 6:53 AM, Kory Maincent wrote:
>>> Change the API to select MAC default time stamping instead of the PHY.
>>> Indeed the PHY is closer to the wire therefore theoretically it has less
>>> delay than the MAC timestamping but the reality is different. Due to lower
>>> time stamping clock frequency, latency in the MDIO bus and no PHC hardware
>>> synchronization between different PHY, the PHY PTP is often less precise
>>> than the MAC. The exception is for PHY designed specially for PTP case but
>>> these devices are not very widespread. For not breaking the compatibility
>>> default_timestamp flag has been introduced in phy_device that is set by
>>> the phy driver to know we are using the old API behavior.
>>>
>>
>> This description feels like it is making a pretty broad generalization
>> about devices. The specifics of whether MAC or PHY timestamping is
>> better will be device dependent.
>
> As explained, except for specific PTP specialized PHY, the MAC is better in
> term of PTP precision.
> This patch was a requisite from Russell, who wanted to add support for the PTP
> in the marvell PHY. Doing so would select the PHY PTP by default which cause a
> regression as the PHY hardware timestamp is less precise than the MAC.
> https://lore.kernel.org/netdev/20200729105807.GZ1551@shell.armlinux.org.uk/
> https://lore.kernel.org/netdev/Y%2F4DZIDm1d74MuFJ@shell.armlinux.org.uk/
> There is also discussion on how to support it in older version of this series.
>
Right. So it is a bit of a generalization, but in practice it matches up
with the available hardware on the market.
>> It looks like you introduce a default_timestamp flag to ensure existing
>> devices default to PHY? I assume your goal here is to discourage this
>> and not allow setting it for new devices? Or do we want to let device
>> driver authors decide which is a better default?
>
> Yes to not change the old behavior the current PHY with PTP support will still
> behave as default PTP. The point is indeed to discourage future drivers to
> select the PHY as default PTP.
>
Ok great!
>>> diff --git a/net/core/timestamping.c b/net/core/timestamping.c
>>> index 04840697fe79..3717fb152ecc 100644
>>> --- a/net/core/timestamping.c
>>> +++ b/net/core/timestamping.c
>>> @@ -25,7 +25,8 @@ void skb_clone_tx_timestamp(struct sk_buff *skb)
>>> struct sk_buff *clone;
>>> unsigned int type;
>>>
>>> - if (!skb->sk)
>>> + if (!skb->sk || !skb->dev ||
>>> + !phy_is_default_hwtstamp(skb->dev->phydev))
>>
>> I don't follow why this check is added and its not calling something
>> like "phy_is_current_hwtstamp"? I guess because we don't yet have a way
>> to select between MAC/PHY at this point in the series? Ok.
>
> skb_clone_tx_timestamp is only used for PHY timestamping so we should do nothing
> if the default PTP is the MAC.
>
I guess my misunderstanding is what about the case where user selects
PHY timestamping with the netlink command? Then it would still need to
do the skb_clone_tx_timestamp even though its not the default? Or does
phy_is_default_hwtstamp take that into account? In which case it would
make more sense to name it phy_is_current_hwtstamp.
Either way this is mostly bikeshedding and probably just some
misunderstanding in my reading of the code.
Thanks,
Jake
> Regards,
^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: [PATCH net-next v17 04/14] net: Change the API of PHY default timestamp to MAC
2024-07-29 18:08 ` Jacob Keller
@ 2024-09-25 12:46 ` Kory Maincent
0 siblings, 0 replies; 39+ messages in thread
From: Kory Maincent @ 2024-09-25 12:46 UTC (permalink / raw)
To: Jacob Keller
Cc: Florian Fainelli, Broadcom internal kernel review list,
Andrew Lunn, Heiner Kallweit, Russell King, David S. Miller,
Eric Dumazet, Jakub Kicinski, Paolo Abeni, Richard Cochran,
Radu Pirea, Jay Vosburgh, Andy Gospodarek, Nicolas Ferre,
Claudiu Beznea, Willem de Bruijn, Jonathan Corbet, Horatiu Vultur,
UNGLinuxDriver, Simon Horman, Vladimir Oltean, donald.hunter,
danieller, ecree.xilinx, Thomas Petazzoni, linux-kernel, netdev,
linux-doc, Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter
Hello Jacob,
On Mon, 29 Jul 2024 11:08:01 -0700
Jacob Keller <jacob.e.keller@intel.com> wrote:
Sorry for answering it so late. I was a bit busy.
> >>> --- a/net/core/timestamping.c
> >>> +++ b/net/core/timestamping.c
> >>> @@ -25,7 +25,8 @@ void skb_clone_tx_timestamp(struct sk_buff *skb)
> >>> struct sk_buff *clone;
> >>> unsigned int type;
> >>>
> >>> - if (!skb->sk)
> >>> + if (!skb->sk || !skb->dev ||
> >>> + !phy_is_default_hwtstamp(skb->dev->phydev))
> >>
> >> I don't follow why this check is added and its not calling something
> >> like "phy_is_current_hwtstamp"? I guess because we don't yet have a way
> >> to select between MAC/PHY at this point in the series? Ok.
> >
> > skb_clone_tx_timestamp is only used for PHY timestamping so we should do
> > nothing if the default PTP is the MAC.
> >
>
> I guess my misunderstanding is what about the case where user selects
> PHY timestamping with the netlink command? Then it would still need to
> do the skb_clone_tx_timestamp even though its not the default? Or does
> phy_is_default_hwtstamp take that into account? In which case it would
> make more sense to name it phy_is_current_hwtstamp.
>
> Either way this is mostly bikeshedding and probably just some
> misunderstanding in my reading of the code.
In fact the phy_is_default_hwtstamp() is only needed in case of no netlink
command used. As you can see in patch 8, we call it only if dev->hwtstamp is
null which mean that a netlink command has been sent.
Regards,
--
Köry Maincent, Bootlin
Embedded Linux and kernel engineering
https://bootlin.com
^ permalink raw reply [flat|nested] 39+ messages in thread
* Re: [PATCH net-next v17 13/14] net: ethtool: Add support for tsconfig command to get/set hwtstamp config
2024-07-15 14:59 ` Jakub Kicinski
@ 2024-09-26 8:47 ` Kory Maincent
0 siblings, 0 replies; 39+ messages in thread
From: Kory Maincent @ 2024-09-26 8:47 UTC (permalink / raw)
To: Jakub Kicinski
Cc: Florian Fainelli, Broadcom internal kernel review list,
Andrew Lunn, Heiner Kallweit, Russell King, David S. Miller,
Eric Dumazet, Paolo Abeni, Richard Cochran, Radu Pirea,
Jay Vosburgh, Andy Gospodarek, Nicolas Ferre, Claudiu Beznea,
Willem de Bruijn, Jonathan Corbet, Horatiu Vultur, UNGLinuxDriver,
Simon Horman, Vladimir Oltean, donald.hunter, danieller,
ecree.xilinx, Thomas Petazzoni, linux-kernel, netdev, linux-doc,
Maxime Chevallier, Rahul Rameshbabu, Willem de Bruijn,
Shannon Nelson, Alexandra Winter
Hello Jakub,
On Mon, 15 Jul 2024 07:59:26 -0700
Jakub Kicinski <kuba@kernel.org> wrote:
Thanks for the review and sorry for the late reply.
> On Tue, 09 Jul 2024 15:53:45 +0200 Kory Maincent wrote:
> > + /* Get the hwtstamp config from netlink */
> > + if (tb[ETHTOOL_A_TSCONFIG_TX_TYPES]) {
> > + ret = ethnl_parse_bitset(&req_tx_type, &mask,
> > + __HWTSTAMP_TX_CNT,
> > + tb[ETHTOOL_A_TSCONFIG_TX_TYPES],
> > + ts_tx_type_names, info->extack);
> > + if (ret < 0)
> > + goto err_clock_put;
> > +
> > + /* Select only one tx type at a time */
> > + if (ffs(req_tx_type) != fls(req_tx_type)) {
> > + ret = -EINVAL;
> > + goto err_clock_put;
> > + }
> > +
> > + hwtst_config.tx_type = ffs(req_tx_type) - 1;
> > + }
> > + if (tb[ETHTOOL_A_TSCONFIG_RX_FILTERS]) {
> > + ret = ethnl_parse_bitset(&req_rx_filter, &mask,
> > + __HWTSTAMP_FILTER_CNT,
> > + tb[ETHTOOL_A_TSCONFIG_RX_FILTERS],
> > + ts_rx_filter_names, info->extack);
> > + if (ret < 0)
> > + goto err_clock_put;
> > +
> > + /* Select only one rx filter at a time */
> > + if (ffs(req_rx_filter) != fls(req_rx_filter)) {
> > + ret = -EINVAL;
> > + goto err_clock_put;
> > + }
> > +
> > + hwtst_config.rx_filter = ffs(req_rx_filter) - 1;
> > + }
> > + if (tb[ETHTOOL_A_TSCONFIG_HWTSTAMP_FLAGS]) {
> > + ret = nla_get_u32(tb[ETHTOOL_A_TSCONFIG_HWTSTAMP_FLAGS]);
> > + if (ret < 0)
> > + goto err_clock_put;
> > + hwtst_config.flags = ret;
> > + }
>
> We should be tracking mod on these, too. Separately from the provider
> mod bit, let's not call the driver and send notification if nothing
> changed.
Ok
> > + ret = net_hwtstamp_validate(&hwtst_config);
> > + if (ret)
> > + goto err_clock_put;
> > +
> > + /* Disable current time stamping if we try to enable another one */
> > + if (mod && (hwtst_config.tx_type || hwtst_config.rx_filter)) {
> > + struct kernel_hwtstamp_config zero_config = {0};
> > +
> > + ret = dev_set_hwtstamp_phylib(dev, &zero_config,
> > info->extack);
> > + if (ret < 0)
> > + goto err_clock_put;
> > + }
> > +
> > + /* Changed the selected hwtstamp source if needed */
> > + if (mod) {
> > + struct hwtstamp_provider *__hwtstamp;
> > +
> > + __hwtstamp = rcu_replace_pointer_rtnl(dev->hwtstamp,
> > hwtstamp);
> > + if (__hwtstamp)
> > + call_rcu(&__hwtstamp->rcu_head,
> > + remove_hwtstamp_provider);
> > + }
> > +
> > + ret = dev_set_hwtstamp_phylib(dev, &hwtst_config, info->extack);
> > + if (ret < 0)
> > + return ret;
>
> We can't unwind to old state here?
Yes indeed we could unwind old state here. I will update it in next version.
> Driver can change hwtst_config right? "upgrade" the rx_filter
> to a broader one, IIRC. Shouldn't we reply to the set command with
> the resulting configuration, in case it changed? Basically provide
> the same info as the notification would.
Yes, the driver does that.
Indeed that's a good idea to report the resulting configuration.
I will take a look at how I can do that.
Regards,
--
Köry Maincent, Bootlin
Embedded Linux and kernel engineering
https://bootlin.com
^ permalink raw reply [flat|nested] 39+ messages in thread
end of thread, other threads:[~2024-09-26 8:47 UTC | newest]
Thread overview: 39+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2024-07-09 13:53 [PATCH net-next v17 00/14] net: Make timestamping selectable Kory Maincent
2024-07-09 13:53 ` [PATCH net-next v17 01/14] net_tstamp: Add TIMESTAMPING SOFTWARE and HARDWARE mask Kory Maincent
2024-07-15 23:28 ` Jacob Keller
2024-07-09 13:53 ` [PATCH net-next v17 02/14] net: Make dev_get_hwtstamp_phylib accessible Kory Maincent
2024-07-15 23:29 ` Jacob Keller
2024-07-09 13:53 ` [PATCH net-next v17 03/14] net: Make net_hwtstamp_validate accessible Kory Maincent
2024-07-15 23:30 ` Jacob Keller
2024-07-09 13:53 ` [PATCH net-next v17 04/14] net: Change the API of PHY default timestamp to MAC Kory Maincent
2024-07-15 23:37 ` Jacob Keller
2024-07-27 13:44 ` Kory Maincent
2024-07-29 18:08 ` Jacob Keller
2024-09-25 12:46 ` Kory Maincent
2024-07-09 13:53 ` [PATCH net-next v17 05/14] net: net_tstamp: Add unspec field to hwtstamp_source enumeration Kory Maincent
2024-07-15 23:37 ` Jacob Keller
2024-07-09 13:53 ` [PATCH net-next v17 06/14] net: Add struct kernel_ethtool_ts_info Kory Maincent
2024-07-15 23:41 ` Jacob Keller
2024-07-09 13:53 ` [PATCH net-next v17 07/14] ptp: Add phc source and helpers to register specific PTP clock or get information Kory Maincent
2024-07-15 23:44 ` Jacob Keller
2024-07-09 13:53 ` [PATCH net-next v17 08/14] net: Add the possibility to support a selected hwtstamp in netdevice Kory Maincent
2024-07-17 17:22 ` Jacob Keller
2024-07-09 13:53 ` [PATCH net-next v17 09/14] net: netdevsim: ptp_mock: Convert to netdev_ptp_clock_register Kory Maincent
2024-07-17 17:23 ` Jacob Keller
2024-07-09 13:53 ` [PATCH net-next v17 10/14] net: macb: " Kory Maincent
2024-07-17 17:24 ` Jacob Keller
2024-07-09 13:53 ` [PATCH net-next v17 11/14] net: ptp: Move ptp_clock_index() to builtin symbol Kory Maincent
2024-07-17 17:24 ` Jacob Keller
2024-07-09 13:53 ` [PATCH net-next v17 12/14] net: ethtool: tsinfo: Add support for reading tsinfo for a specific hwtstamp provider Kory Maincent
2024-07-17 17:35 ` Jacob Keller
2024-07-26 19:04 ` Kory Maincent
2024-07-29 17:58 ` Jacob Keller
2024-07-09 13:53 ` [PATCH net-next v17 13/14] net: ethtool: Add support for tsconfig command to get/set hwtstamp config Kory Maincent
2024-07-15 14:59 ` Jakub Kicinski
2024-09-26 8:47 ` Kory Maincent
2024-07-17 17:43 ` Jacob Keller
2024-07-27 13:00 ` Kory Maincent
2024-07-29 18:02 ` Jacob Keller
2024-07-09 13:53 ` [PATCH net-next v17 14/14] netlink: specs: Enhance tsinfo netlink attributes and add a tsconfig set command Kory Maincent
2024-07-17 17:44 ` Jacob Keller
2024-07-15 15:30 ` [PATCH net-next v17 00/14] net: Make timestamping selectable patchwork-bot+netdevbpf
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).