[RFC PATCH net-next v3 0/6] ethtool: Add ability to control transceiver modules' power mode

[RFC PATCH net-next v3 0/6] ethtool: Add ability to control transceiver modules' power mode

[Ido Schimmel]

From: Ido Schimmel <idosch@nvidia.com>

This patchset extends the ethtool netlink API to allow user space to
control transceiver modules. Two specific APIs are added, but the plan
is to extend the interface with more APIs in the future (see "Future
plans").

This submission is a complete rework of a previous submission [1] that
tried to achieve the same goal by allowing user space to write to the
EEPROMs of these modules. It was rejected as it could have enabled user
space binary blob drivers.

However, the main issue is that by directly writing to some pages of
these EEPROMs, we are interfering with the entity that is controlling
the modules (kernel / device firmware). In addition, some functionality
cannot be implemented solely by writing to the EEPROM, as it requires
the assertion / de-assertion of hardware signals (e.g., "ResetL" pin in
SFF-8636).

Motivation
==========

The kernel can currently dump the contents of module EEPROMs to user
space via the ethtool legacy ioctl API or the new netlink API. These
dumps can then be parsed by ethtool(8) according to the specification
that defines the memory map of the EEPROM. For example, SFF-8636 [2] for
QSFP and CMIS [3] for QSFP-DD.

In addition to read-only elements, these specifications also define
writeable elements that can be used to control the behavior of the
module. For example, controlling whether the module is put in low or
high power mode to limit its power consumption.

The CMIS specification even defines a message exchange mechanism (CDB,
Command Data Block) on top of the module's memory map. This allows the
host to send various commands to the module. For example, to update its
firmware.

Implementation
==============

The ethtool netlink API is extended with two new messages,
'ETHTOOL_MSG_MODULE_SET' and 'ETHTOOL_MSG_MODULE_GET', that allow user
space to set and get transceiver module parameters. Specifically, the
'ETHTOOL_A_MODULE_POWER_MODE_POLICY' attribute allows user space to
control the power mode policy of the module in order to limit its power
consumption. See detailed description in patch #1.

The user API is designed to be generic enough so that it could be used
for modules with different memory maps (e.g., SFF-8636, CMIS).

The only implementation of the device driver API in this series is for a
MAC driver (mlxsw) where the module is controlled by the device's
firmware, but it is designed to be generic enough so that it could also
be used by implementations where the module is controlled by the kernel.

Testing and introspection
=========================

See detailed description in patches #1 and #5.

Patchset overview
=================

Patch #1 adds the initial infrastructure in ethtool along with the
ability to control transceiver modules' power mode.

Patches #2-#3 add required device registers in mlxsw.

Patch #4 implements in mlxsw the ethtool operations added in patch #1.

Patch #5 adds extended link states in order to allow user space to
troubleshoot link down issues related to transceiver modules.

Patch #6 adds support for these extended states in mlxsw.

Future plans
============

* Extend 'ETHTOOL_MSG_MODULE_SET' to control Tx output among other
attributes.

* Add new ethtool message(s) to update firmware on transceiver modules.

* Extend ethtool(8) to parse more diagnostic information from CMIS
modules. No kernel changes required.

Changes since RFC v2 [4]:

* Change 'high-on-up' power mode policy to 'auto'
* Remove 'low' power mode policy
* Document that default power mode policy is driver-dependent
* Remove restrictions regarding interface administrative state when
  setting the power mode policy

Changes since RFC v1 [5]:

* Split 'ETHTOOL_A_MODULE_LOW_POWER_ENABLED' into two attributes.
* Add 'high-on-up' power mode policy.
* Drop 'ETHTOOL_MSG_MODULE_RESET_ACT' in favor of existing ioctl
  interface.
* Add extended link states to help in troubleshooting.

Note: This series does not apply to net-next (it is RFC, anyway) because
I left out mundane infra work in mlxsw. Will submit it with reset
support as part of a separate set.

[1] https://lore.kernel.org/netdev/20210623075925.2610908-1-idosch@idosch.org/
[2] https://members.snia.org/document/dl/26418
[3] http://www.qsfp-dd.com/wp-content/uploads/2021/05/CMIS5p0.pdf
[4] https://lore.kernel.org/netdev/20210818155202.1278177-1-idosch@idosch.org/
[5] https://lore.kernel.org/netdev/20210809102152.719961-1-idosch@idosch.org/

Ido Schimmel (6):
  ethtool: Add ability to control transceiver modules' power mode
  mlxsw: reg: Add Port Module Memory Map Properties register
  mlxsw: reg: Add Management Cable IO and Notifications register
  mlxsw: Add ability to control transceiver modules' power mode
  ethtool: Add transceiver module extended states
  mlxsw: Add support for transceiver module extended states

 Documentation/networking/ethtool-netlink.rst  |  76 ++++++-
 .../net/ethernet/mellanox/mlxsw/core_env.c    | 194 +++++++++++++++++-
 .../net/ethernet/mellanox/mlxsw/core_env.h    |  10 +
 drivers/net/ethernet/mellanox/mlxsw/minimal.c |  26 +++
 drivers/net/ethernet/mellanox/mlxsw/reg.h     |  78 +++++++
 .../mellanox/mlxsw/spectrum_ethtool.c         |  73 ++++++-
 include/linux/ethtool.h                       |  26 +++
 include/uapi/linux/ethtool.h                  |  30 +++
 include/uapi/linux/ethtool_netlink.h          |  17 ++
 net/ethtool/Makefile                          |   2 +-
 net/ethtool/module.c                          | 184 +++++++++++++++++
 net/ethtool/netlink.c                         |  19 ++
 net/ethtool/netlink.h                         |   4 +
 13 files changed, 731 insertions(+), 8 deletions(-)
 create mode 100644 net/ethtool/module.c

-- 
2.31.1

[RFC PATCH net-next v3 1/6] ethtool: Add ability to control transceiver modules' power mode

[Ido Schimmel]

From: Ido Schimmel <idosch@nvidia.com>

Add a pair of new ethtool messages, 'ETHTOOL_MSG_MODULE_SET' and
'ETHTOOL_MSG_MODULE_GET', that can be used to control transceiver
modules parameters and retrieve their status.

The first parameter to control is the power mode of the module. It is
only relevant for paged memory modules, as flat memory modules always
operate in low power mode.

When a paged memory module is in low power mode, its power consumption
is reduced to the minimum, the management interface towards the host is
available and the data path is deactivated.

User space can choose to put modules that are not currently in use in
low power mode and transition them to high power mode before putting the
associated ports administratively up. This is useful for user space that
favors reduced power consumption and lower temperatures over reduced
link up times. In QSFP-DD modules the transition from low power mode to
high power mode can take a few seconds and this transition is only
expected to get longer with future / more complex modules.

User space can control the power mode of the module via the power mode
policy attribute ('ETHTOOL_A_MODULE_POWER_MODE_POLICY'). Possible
values:

* high: Module is always in high power mode.

* auto: Module is transitioned by the host to high power mode when the
  first port using it is put administratively up and to low power mode
  when the last port using it is put administratively down.

The operational power mode of the module is available to user space via
the 'ETHTOOL_A_MODULE_POWER_MODE' attribute. The attribute is not
reported to user space when a module is not plugged-in.

The user API is designed to be generic enough so that it could be used
for modules with different memory maps (e.g., SFF-8636, CMIS).

The only implementation of the device driver API in this series is for a
MAC driver (mlxsw) where the module is controlled by the device's
firmware, but it is designed to be generic enough so that it could also
be used by implementations where the module is controlled by the CPU.

CMIS testing
============

 # ethtool -m swp11
 Identifier                                : 0x18 (QSFP-DD Double Density 8X Pluggable Transceiver (INF-8628))
 ...
 Module State                              : 0x03 (ModuleReady)
 LowPwrAllowRequestHW                      : Off
 LowPwrRequestSW                           : Off

The module is not in low power mode, as it is not forced by hardware
(LowPwrAllowRequestHW is off) or by software (LowPwrRequestSW is off).

The power mode can be queried from the kernel. In case
LowPwrAllowRequestHW was on, the kernel would need to take into account
the state of the LowPwrRequestHW signal, which is not visible to user
space.

 $ ethtool --show-module swp11
 Module parameters for swp11:
 power-mode-policy high
 power-mode high

Change the power mode policy to 'auto':

 # ethtool --set-module swp11 power-mode-policy auto

Query the power mode again:

 $ ethtool --show-module swp11
 Module parameters for swp11:
 power-mode-policy auto
 power-mode low

Verify with the data read from the EEPROM:

 # ethtool -m swp11
 Identifier                                : 0x18 (QSFP-DD Double Density 8X Pluggable Transceiver (INF-8628))
 ...
 Module State                              : 0x01 (ModuleLowPwr)
 LowPwrAllowRequestHW                      : Off
 LowPwrRequestSW                           : On

Put the associated port administratively up which will instruct the host
to transition the module to high power mode:

 # ip link set dev swp11 up

Query the power mode again:

 $ ethtool --show-module swp11
 Module parameters for swp11:
 power-mode-policy auto
 power-mode high

Verify with the data read from the EEPROM:

 # ethtool -m swp11
 Identifier                                : 0x18 (QSFP-DD Double Density 8X Pluggable Transceiver (INF-8628))
 ...
 Module State                              : 0x03 (ModuleReady)
 LowPwrAllowRequestHW                      : Off
 LowPwrRequestSW                           : Off

Put the associated port administratively down which will instruct the
host to transition the module to low power mode:

 # ip link set dev swp11 down

Query the power mode again:

 $ ethtool --show-module swp11
 Module parameters for swp11:
 power-mode-policy auto
 power-mode low

Verify with the data read from the EEPROM:

 # ethtool -m swp11
 Identifier                                : 0x18 (QSFP-DD Double Density 8X Pluggable Transceiver (INF-8628))
 ...
 Module State                              : 0x01 (ModuleLowPwr)
 LowPwrAllowRequestHW                      : Off
 LowPwrRequestSW                           : On

SFF-8636 testing
================

 # ethtool -m swp13
 Identifier                                : 0x11 (QSFP28)
 ...
 Extended identifier description           : 5.0W max. Power consumption,  High Power Class (> 3.5 W) enabled
 Power set                                 : Off
 Power override                            : On
 ...
 Transmit avg optical power (Channel 1)    : 0.7733 mW / -1.12 dBm
 Transmit avg optical power (Channel 2)    : 0.7649 mW / -1.16 dBm
 Transmit avg optical power (Channel 3)    : 0.7790 mW / -1.08 dBm
 Transmit avg optical power (Channel 4)    : 0.7837 mW / -1.06 dBm
 Rcvr signal avg optical power(Channel 1)  : 0.9302 mW / -0.31 dBm
 Rcvr signal avg optical power(Channel 2)  : 0.9079 mW / -0.42 dBm
 Rcvr signal avg optical power(Channel 3)  : 0.8993 mW / -0.46 dBm
 Rcvr signal avg optical power(Channel 4)  : 0.8778 mW / -0.57 dBm

The module is not in low power mode, as it is not forced by hardware
(Power override is on) or by software (Power set is off).

The power mode can be queried from the kernel. In case Power override
was off, the kernel would need to take into account the state of the
LPMode signal, which is not visible to user space.

 $ ethtool --show-module swp13
 Module parameters for swp13:
 power-mode-policy high
 power-mode high

Change the power mode policy to 'auto':

 # ethtool --set-module swp13 power-mode-policy auto

Query the power mode again:

 $ ethtool --show-module swp13
 Module parameters for swp13:
 power-mode-policy auto
 power-mode low

Verify with the data read from the EEPROM:

 # ethtool -m swp13
 Identifier                                : 0x11 (QSFP28)
 Extended identifier description           : 5.0W max. Power consumption,  High Power Class (> 3.5 W) not enabled
 Power set                                 : On
 Power override                            : On
 ...
 Transmit avg optical power (Channel 1)    : 0.0000 mW / -inf dBm
 Transmit avg optical power (Channel 2)    : 0.0000 mW / -inf dBm
 Transmit avg optical power (Channel 3)    : 0.0000 mW / -inf dBm
 Transmit avg optical power (Channel 4)    : 0.0000 mW / -inf dBm
 Rcvr signal avg optical power(Channel 1)  : 0.0000 mW / -inf dBm
 Rcvr signal avg optical power(Channel 2)  : 0.0000 mW / -inf dBm
 Rcvr signal avg optical power(Channel 3)  : 0.0000 mW / -inf dBm
 Rcvr signal avg optical power(Channel 4)  : 0.0000 mW / -inf dBm

Put the associated port administratively up which will instruct the host
to transition the module to high power mode:

 # ip link set dev swp13 up

Query the power mode again:

 $ ethtool --show-module swp13
 Module parameters for swp13:
 power-mode-policy auto
 power-mode high

Verify with the data read from the EEPROM:

 # ethtool -m swp13
 Identifier                                : 0x11 (QSFP28)
 ...
 Extended identifier description           : 5.0W max. Power consumption,  High Power Class (> 3.5 W) enabled
 Power set                                 : Off
 Power override                            : On
 ...
 Transmit avg optical power (Channel 1)    : 0.7934 mW / -1.01 dBm
 Transmit avg optical power (Channel 2)    : 0.7859 mW / -1.05 dBm
 Transmit avg optical power (Channel 3)    : 0.7885 mW / -1.03 dBm
 Transmit avg optical power (Channel 4)    : 0.7985 mW / -0.98 dBm
 Rcvr signal avg optical power(Channel 1)  : 0.9325 mW / -0.30 dBm
 Rcvr signal avg optical power(Channel 2)  : 0.9034 mW / -0.44 dBm
 Rcvr signal avg optical power(Channel 3)  : 0.9086 mW / -0.42 dBm
 Rcvr signal avg optical power(Channel 4)  : 0.8885 mW / -0.51 dBm

Put the associated port administratively down which will instruct the
host to transition the module to low power mode:

 # ip link set dev swp13 down

Query the power mode again:

 $ ethtool --show-module swp13
 Module parameters for swp13:
 power-mode-policy auto
 power-mode low

Verify with the data read from the EEPROM:

 # ethtool -m swp13
 Identifier                                : 0x11 (QSFP28)
 ...
 Extended identifier description           : 5.0W max. Power consumption,  High Power Class (> 3.5 W) not enabled
 Power set                                 : On
 Power override                            : On
 ...
 Transmit avg optical power (Channel 1)    : 0.0000 mW / -inf dBm
 Transmit avg optical power (Channel 2)    : 0.0000 mW / -inf dBm
 Transmit avg optical power (Channel 3)    : 0.0000 mW / -inf dBm
 Transmit avg optical power (Channel 4)    : 0.0000 mW / -inf dBm
 Rcvr signal avg optical power(Channel 1)  : 0.0000 mW / -inf dBm
 Rcvr signal avg optical power(Channel 2)  : 0.0000 mW / -inf dBm
 Rcvr signal avg optical power(Channel 3)  : 0.0000 mW / -inf dBm
 Rcvr signal avg optical power(Channel 4)  : 0.0000 mW / -inf dBm

Signed-off-by: Ido Schimmel <idosch@nvidia.com>
---
 Documentation/networking/ethtool-netlink.rst |  64 ++++++-
 include/linux/ethtool.h                      |  25 +++
 include/uapi/linux/ethtool.h                 |  23 +++
 include/uapi/linux/ethtool_netlink.h         |  17 ++
 net/ethtool/Makefile                         |   2 +-
 net/ethtool/module.c                         | 184 +++++++++++++++++++
 net/ethtool/netlink.c                        |  19 ++
 net/ethtool/netlink.h                        |   4 +
 8 files changed, 335 insertions(+), 3 deletions(-)
 create mode 100644 net/ethtool/module.c

diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst
index c690bb37430d..245ce2eab9a5 100644
--- a/Documentation/networking/ethtool-netlink.rst
+++ b/Documentation/networking/ethtool-netlink.rst
@@ -179,7 +179,7 @@ according to message purpose:
 
 Userspace to kernel:
 
-  ===================================== ================================
+  ===================================== =================================
   ``ETHTOOL_MSG_STRSET_GET``            get string set
   ``ETHTOOL_MSG_LINKINFO_GET``          get link settings
   ``ETHTOOL_MSG_LINKINFO_SET``          set link settings
@@ -213,7 +213,9 @@ Userspace to kernel:
   ``ETHTOOL_MSG_MODULE_EEPROM_GET``     read SFP module EEPROM
   ``ETHTOOL_MSG_STATS_GET``             get standard statistics
   ``ETHTOOL_MSG_PHC_VCLOCKS_GET``       get PHC virtual clocks info
-  ===================================== ================================
+  ``ETHTOOL_MSG_MODULE_SET``            set transceiver module parameters
+  ``ETHTOOL_MSG_MODULE_GET``            get transceiver module parameters
+  ===================================== =================================
 
 Kernel to userspace:
 
@@ -252,6 +254,7 @@ Kernel to userspace:
   ``ETHTOOL_MSG_MODULE_EEPROM_GET_REPLY``  read SFP module EEPROM
   ``ETHTOOL_MSG_STATS_GET_REPLY``          standard statistics
   ``ETHTOOL_MSG_PHC_VCLOCKS_GET_REPLY``    PHC virtual clocks info
+  ``ETHTOOL_MSG_MODULE_GET_REPLY``         transceiver module parameters
   ======================================== =================================
 
 ``GET`` requests are sent by userspace applications to retrieve device
@@ -1506,6 +1509,61 @@ Kernel response contents:
   ``ETHTOOL_A_PHC_VCLOCKS_INDEX``       s32     PHC index array
   ====================================  ======  ==========================
 
+MODULE_GET
+==========
+
+Gets transceiver module parameters.
+
+Request contents:
+
+  =====================================  ======  ==========================
+  ``ETHTOOL_A_MODULE_HEADER``            nested  request header
+  =====================================  ======  ==========================
+
+Kernel response contents:
+
+  ======================================  ======  ==========================
+  ``ETHTOOL_A_MODULE_HEADER``             nested  reply header
+  ``ETHTOOL_A_MODULE_POWER_MODE_POLICY``  u8      power mode policy
+  ``ETHTOOL_A_MODULE_POWER_MODE``         u8      operational power mode
+  ======================================  ======  ==========================
+
+The optional ``ETHTOOL_A_MODULE_POWER_MODE_POLICY`` attribute encodes the
+transceiver module power mode policy enforced by the host. The default policy
+is driver-dependent and can be queried using this attribute.
+
+The optional ``ETHTHOOL_A_MODULE_POWER_MODE`` attribute encodes the operational
+power mode policy of the transceiver module. It is only reported when a module
+is plugged-in. Possible values are:
+
+.. kernel-doc:: include/uapi/linux/ethtool.h
+    :identifiers: ethtool_module_power_mode
+
+MODULE_SET
+==========
+
+Sets transceiver module parameters.
+
+Request contents:
+
+  ======================================  ======  ==========================
+  ``ETHTOOL_A_MODULE_HEADER``             nested  request header
+  ``ETHTOOL_A_MODULE_POWER_MODE_POLICY``  u8      power mode policy
+  ======================================  ======  ==========================
+
+When set, the optional ``ETHTOOL_A_MODULE_POWER_MODE_POLICY`` attribute is used
+to set the transceiver module power policy enforced by the host. Possible
+values are:
+
+.. kernel-doc:: include/uapi/linux/ethtool.h
+    :identifiers: ethtool_module_power_mode_policy
+
+For SFF-8636 modules, low power mode is forced by the host according to table
+6-10 in revision 2.10a of the specification.
+
+For CMIS modules, low power mode is forced by the host according to table 6-12
+in revision 5.0 of the specification.
+
 Request translation
 ===================
 
@@ -1605,4 +1663,6 @@ are netlink only.
   n/a                                 ``ETHTOOL_MSG_CABLE_TEST_TDR_ACT``
   n/a                                 ``ETHTOOL_MSG_TUNNEL_INFO_GET``
   n/a                                 ``ETHTOOL_MSG_PHC_VCLOCKS_GET``
+  n/a                                 ``ETHTOOL_MSG_MODULE_GET``
+  n/a                                 ``ETHTOOL_MSG_MODULE_SET``
   =================================== =====================================
diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index 4711b96dae0c..07d40dc20ca4 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -405,6 +405,20 @@ struct ethtool_module_eeprom {
 	u8	*data;
 };
 
+/**
+ * struct ethtool_module_power_mode_params - module power mode parameters
+ * @policy: The power mode policy enforced by the host for the plug-in module.
+ * @mode: The operational power mode of the plug-in module. Should be filled by
+ * device drivers on get operations.
+ * @mode_valid: Indicates the validity of the @mode field. Should be set by
+ * device drivers on get operations when a module is plugged-in.
+ */
+struct ethtool_module_power_mode_params {
+	enum ethtool_module_power_mode_policy policy;
+	enum ethtool_module_power_mode mode;
+	u8 mode_valid:1;
+};
+
 /**
  * struct ethtool_ops - optional netdev operations
  * @cap_link_lanes_supported: indicates if the driver supports lanes
@@ -570,6 +584,11 @@ struct ethtool_module_eeprom {
  * @get_eth_ctrl_stats: Query some of the IEEE 802.3 MAC Ctrl statistics.
  * @get_rmon_stats: Query some of the RMON (RFC 2819) statistics.
  *	Set %ranges to a pointer to zero-terminated array of byte ranges.
+ * @get_module_power_mode: Get the power mode policy for the plug-in module
+ *	used by the network device and its operational power mode, if
+ *	plugged-in.
+ * @set_module_power_mode: Set the power mode policy for the plug-in module
+ *	used by the network device.
  *
  * All operations are optional (i.e. the function pointer may be set
  * to %NULL) and callers must take this into account.  Callers must
@@ -689,6 +708,12 @@ struct ethtool_ops {
 	void	(*get_rmon_stats)(struct net_device *dev,
 				  struct ethtool_rmon_stats *rmon_stats,
 				  const struct ethtool_rmon_hist_range **ranges);
+	int	(*get_module_power_mode)(struct net_device *dev,
+					 struct ethtool_module_power_mode_params *params,
+					 struct netlink_ext_ack *extack);
+	int	(*set_module_power_mode)(struct net_device *dev,
+					 const struct ethtool_module_power_mode_params *params,
+					 struct netlink_ext_ack *extack);
 };
 
 int ethtool_check_ops(const struct ethtool_ops *ops);
diff --git a/include/uapi/linux/ethtool.h b/include/uapi/linux/ethtool.h
index b6db6590baf0..8cc79811ee5d 100644
--- a/include/uapi/linux/ethtool.h
+++ b/include/uapi/linux/ethtool.h
@@ -706,6 +706,29 @@ enum ethtool_stringset {
 	ETH_SS_COUNT
 };
 
+/**
+ * enum ethtool_module_power_mode_policy - plug-in module power mode policy
+ * @ETHTOOL_MODULE_POWER_MODE_POLICY_HIGH: Module is always in high power mode.
+ * @ETHTOOL_MODULE_POWER_MODE_POLICY_AUTO: Module is transitioned by the host
+ *	to high power mode when the first port using it is put administratively
+ *	up and to low power mode when the last port using it is put
+ *	administratively down.
+ */
+enum ethtool_module_power_mode_policy {
+	ETHTOOL_MODULE_POWER_MODE_POLICY_HIGH,
+	ETHTOOL_MODULE_POWER_MODE_POLICY_AUTO,
+};
+
+/**
+ * enum ethtool_module_power_mode - plug-in module power mode
+ * @ETHTOOL_MODULE_POWER_MODE_LOW: Module is in low power mode.
+ * @ETHTOOL_MODULE_POWER_MODE_HIGH: Module is in high power mode.
+ */
+enum ethtool_module_power_mode {
+	ETHTOOL_MODULE_POWER_MODE_LOW,
+	ETHTOOL_MODULE_POWER_MODE_HIGH,
+};
+
 /**
  * struct ethtool_gstrings - string set for data tagging
  * @cmd: Command number = %ETHTOOL_GSTRINGS
diff --git a/include/uapi/linux/ethtool_netlink.h b/include/uapi/linux/ethtool_netlink.h
index b3b93710eff7..c22f5c902060 100644
--- a/include/uapi/linux/ethtool_netlink.h
+++ b/include/uapi/linux/ethtool_netlink.h
@@ -47,6 +47,8 @@ enum {
 	ETHTOOL_MSG_MODULE_EEPROM_GET,
 	ETHTOOL_MSG_STATS_GET,
 	ETHTOOL_MSG_PHC_VCLOCKS_GET,
+	ETHTOOL_MSG_MODULE_GET,
+	ETHTOOL_MSG_MODULE_SET,
 
 	/* add new constants above here */
 	__ETHTOOL_MSG_USER_CNT,
@@ -90,6 +92,8 @@ enum {
 	ETHTOOL_MSG_MODULE_EEPROM_GET_REPLY,
 	ETHTOOL_MSG_STATS_GET_REPLY,
 	ETHTOOL_MSG_PHC_VCLOCKS_GET_REPLY,
+	ETHTOOL_MSG_MODULE_GET_REPLY,
+	ETHTOOL_MSG_MODULE_NTF,
 
 	/* add new constants above here */
 	__ETHTOOL_MSG_KERNEL_CNT,
@@ -831,6 +835,19 @@ enum {
 	ETHTOOL_A_STATS_RMON_MAX = (__ETHTOOL_A_STATS_RMON_CNT - 1)
 };
 
+/* MODULE */
+
+enum {
+	ETHTOOL_A_MODULE_UNSPEC,
+	ETHTOOL_A_MODULE_HEADER,		/* nest - _A_HEADER_* */
+	ETHTOOL_A_MODULE_POWER_MODE_POLICY,	/* u8 */
+	ETHTOOL_A_MODULE_POWER_MODE,		/* u8 */
+
+	/* add new constants above here */
+	__ETHTOOL_A_MODULE_CNT,
+	ETHTOOL_A_MODULE_MAX = (__ETHTOOL_A_MODULE_CNT - 1)
+};
+
 /* generic netlink info */
 #define ETHTOOL_GENL_NAME "ethtool"
 #define ETHTOOL_GENL_VERSION 1
diff --git a/net/ethtool/Makefile b/net/ethtool/Makefile
index 0a19470efbfb..b76432e70e6b 100644
--- a/net/ethtool/Makefile
+++ b/net/ethtool/Makefile
@@ -7,4 +7,4 @@ obj-$(CONFIG_ETHTOOL_NETLINK)	+= ethtool_nl.o
 ethtool_nl-y	:= netlink.o bitset.o strset.o linkinfo.o linkmodes.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
+		   tunnels.o fec.o eeprom.o stats.o phc_vclocks.o module.o
diff --git a/net/ethtool/module.c b/net/ethtool/module.c
new file mode 100644
index 000000000000..ae9ade24f59a
--- /dev/null
+++ b/net/ethtool/module.c
@@ -0,0 +1,184 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#include <linux/ethtool.h>
+
+#include "netlink.h"
+#include "common.h"
+#include "bitset.h"
+
+struct module_req_info {
+	struct ethnl_req_info base;
+};
+
+struct module_reply_data {
+	struct ethnl_reply_data	base;
+	struct ethtool_module_power_mode_params power;
+	u8 power_valid:1;
+};
+
+#define MODULE_REPDATA(__reply_base) \
+	container_of(__reply_base, struct module_reply_data, base)
+
+/* MODULE_GET */
+
+const struct nla_policy ethnl_module_get_policy[ETHTOOL_A_MODULE_HEADER + 1] = {
+	[ETHTOOL_A_MODULE_HEADER] = NLA_POLICY_NESTED(ethnl_header_policy),
+};
+
+static int module_get_power_mode(struct net_device *dev,
+				 struct module_reply_data *data,
+				 struct netlink_ext_ack *extack)
+{
+	const struct ethtool_ops *ops = dev->ethtool_ops;
+	int ret;
+
+	if (!ops->get_module_power_mode)
+		return 0;
+
+	ret = ops->get_module_power_mode(dev, &data->power, extack);
+	if (ret < 0)
+		return ret;
+
+	data->power_valid = true;
+
+	return 0;
+}
+
+static int module_prepare_data(const struct ethnl_req_info *req_base,
+			       struct ethnl_reply_data *reply_base,
+			       struct genl_info *info)
+{
+	struct module_reply_data *data = MODULE_REPDATA(reply_base);
+	struct netlink_ext_ack *extack = info ? info->extack : NULL;
+	struct net_device *dev = reply_base->dev;
+	int ret;
+
+	ret = ethnl_ops_begin(dev);
+	if (ret < 0)
+		return ret;
+
+	ret = module_get_power_mode(dev, data, extack);
+	if (ret < 0)
+		goto out_complete;
+
+out_complete:
+	ethnl_ops_complete(dev);
+	return ret;
+}
+
+static int module_reply_size(const struct ethnl_req_info *req_base,
+			     const struct ethnl_reply_data *reply_base)
+{
+	struct module_reply_data *data = MODULE_REPDATA(reply_base);
+	int len = 0;
+
+	if (data->power_valid)
+		len += nla_total_size(sizeof(u8));	/* _MODULE_POWER_MODE_POLICY */
+
+	if (data->power_valid && data->power.mode_valid)
+		len += nla_total_size(sizeof(u8));	/* _MODULE_POWER_MODE */
+
+	return len;
+}
+
+static int module_fill_reply(struct sk_buff *skb,
+			     const struct ethnl_req_info *req_base,
+			     const struct ethnl_reply_data *reply_base)
+{
+	const struct module_reply_data *data = MODULE_REPDATA(reply_base);
+
+	if (data->power_valid &&
+	    nla_put_u8(skb, ETHTOOL_A_MODULE_POWER_MODE_POLICY,
+		       data->power.policy))
+		return -EMSGSIZE;
+
+	if (data->power_valid && data->power.mode_valid &&
+	    nla_put_u8(skb, ETHTOOL_A_MODULE_POWER_MODE, data->power.mode))
+		return -EMSGSIZE;
+
+	return 0;
+}
+
+const struct ethnl_request_ops ethnl_module_request_ops = {
+	.request_cmd		= ETHTOOL_MSG_MODULE_GET,
+	.reply_cmd		= ETHTOOL_MSG_MODULE_GET_REPLY,
+	.hdr_attr		= ETHTOOL_A_MODULE_HEADER,
+	.req_info_size		= sizeof(struct module_req_info),
+	.reply_data_size	= sizeof(struct module_reply_data),
+
+	.prepare_data		= module_prepare_data,
+	.reply_size		= module_reply_size,
+	.fill_reply		= module_fill_reply,
+};
+
+/* MODULE_SET */
+
+const struct nla_policy ethnl_module_set_policy[ETHTOOL_A_MODULE_POWER_MODE_POLICY + 1] = {
+	[ETHTOOL_A_MODULE_HEADER] = NLA_POLICY_NESTED(ethnl_header_policy),
+	[ETHTOOL_A_MODULE_POWER_MODE_POLICY] =
+		NLA_POLICY_MAX(NLA_U8, ETHTOOL_MODULE_POWER_MODE_POLICY_AUTO),
+};
+
+static int module_set_power_mode(struct net_device *dev, struct nlattr **tb,
+				 bool *p_mod, struct netlink_ext_ack *extack)
+{
+	struct ethtool_module_power_mode_params power = {};
+	struct ethtool_module_power_mode_params power_new;
+	const struct ethtool_ops *ops = dev->ethtool_ops;
+	int ret;
+
+	if (!tb[ETHTOOL_A_MODULE_POWER_MODE_POLICY])
+		return 0;
+
+	if (!ops->get_module_power_mode || !ops->set_module_power_mode) {
+		NL_SET_ERR_MSG_ATTR(extack,
+				    tb[ETHTOOL_A_MODULE_POWER_MODE_POLICY],
+				    "Setting power mode policy is not supported by this device");
+		return -EOPNOTSUPP;
+	}
+
+	power_new.policy = nla_get_u8(tb[ETHTOOL_A_MODULE_POWER_MODE_POLICY]);
+	ret = ops->get_module_power_mode(dev, &power, extack);
+	if (ret < 0)
+		return ret;
+	*p_mod = power_new.policy != power.policy;
+
+	return ops->set_module_power_mode(dev, &power_new, extack);
+}
+
+int ethnl_set_module(struct sk_buff *skb, struct genl_info *info)
+{
+	struct ethnl_req_info req_info = {};
+	struct nlattr **tb = info->attrs;
+	struct net_device *dev;
+	bool mod = false;
+	int ret;
+
+	ret = ethnl_parse_header_dev_get(&req_info, tb[ETHTOOL_A_MODULE_HEADER],
+					 genl_info_net(info), info->extack,
+					 true);
+	if (ret < 0)
+		return ret;
+	dev = req_info.dev;
+
+	rtnl_lock();
+	ret = ethnl_ops_begin(dev);
+	if (ret < 0)
+		goto out_rtnl;
+
+	ret = module_set_power_mode(dev, tb, &mod, info->extack);
+	if (ret < 0)
+		goto out_ops;
+
+	if (!mod)
+		goto out_ops;
+
+	ethtool_notify(dev, ETHTOOL_MSG_MODULE_NTF, NULL);
+
+out_ops:
+	ethnl_ops_complete(dev);
+out_rtnl:
+	rtnl_unlock();
+	dev_put(dev);
+	return ret;
+}
diff --git a/net/ethtool/netlink.c b/net/ethtool/netlink.c
index 1797a0a90019..38b44c0291b1 100644
--- a/net/ethtool/netlink.c
+++ b/net/ethtool/netlink.c
@@ -282,6 +282,7 @@ ethnl_default_requests[__ETHTOOL_MSG_USER_CNT] = {
 	[ETHTOOL_MSG_MODULE_EEPROM_GET]	= &ethnl_module_eeprom_request_ops,
 	[ETHTOOL_MSG_STATS_GET]		= &ethnl_stats_request_ops,
 	[ETHTOOL_MSG_PHC_VCLOCKS_GET]	= &ethnl_phc_vclocks_request_ops,
+	[ETHTOOL_MSG_MODULE_GET]	= &ethnl_module_request_ops,
 };
 
 static struct ethnl_dump_ctx *ethnl_dump_context(struct netlink_callback *cb)
@@ -593,6 +594,7 @@ ethnl_default_notify_ops[ETHTOOL_MSG_KERNEL_MAX + 1] = {
 	[ETHTOOL_MSG_PAUSE_NTF]		= &ethnl_pause_request_ops,
 	[ETHTOOL_MSG_EEE_NTF]		= &ethnl_eee_request_ops,
 	[ETHTOOL_MSG_FEC_NTF]		= &ethnl_fec_request_ops,
+	[ETHTOOL_MSG_MODULE_NTF]	= &ethnl_module_request_ops,
 };
 
 /* default notification handler */
@@ -686,6 +688,7 @@ static const ethnl_notify_handler_t ethnl_notify_handlers[] = {
 	[ETHTOOL_MSG_PAUSE_NTF]		= ethnl_default_notify,
 	[ETHTOOL_MSG_EEE_NTF]		= ethnl_default_notify,
 	[ETHTOOL_MSG_FEC_NTF]		= ethnl_default_notify,
+	[ETHTOOL_MSG_MODULE_NTF]	= ethnl_default_notify,
 };
 
 void ethtool_notify(struct net_device *dev, unsigned int cmd, const void *data)
@@ -999,6 +1002,22 @@ static const struct genl_ops ethtool_genl_ops[] = {
 		.policy = ethnl_phc_vclocks_get_policy,
 		.maxattr = ARRAY_SIZE(ethnl_phc_vclocks_get_policy) - 1,
 	},
+	{
+		.cmd	= ETHTOOL_MSG_MODULE_GET,
+		.doit	= ethnl_default_doit,
+		.start	= ethnl_default_start,
+		.dumpit	= ethnl_default_dumpit,
+		.done	= ethnl_default_done,
+		.policy = ethnl_module_get_policy,
+		.maxattr = ARRAY_SIZE(ethnl_module_get_policy) - 1,
+	},
+	{
+		.cmd	= ETHTOOL_MSG_MODULE_SET,
+		.flags	= GENL_UNS_ADMIN_PERM,
+		.doit	= ethnl_set_module,
+		.policy = ethnl_module_set_policy,
+		.maxattr = ARRAY_SIZE(ethnl_module_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 077aac3929a8..1aafba5b67bb 100644
--- a/net/ethtool/netlink.h
+++ b/net/ethtool/netlink.h
@@ -337,6 +337,7 @@ extern const struct ethnl_request_ops ethnl_fec_request_ops;
 extern const struct ethnl_request_ops ethnl_module_eeprom_request_ops;
 extern const struct ethnl_request_ops ethnl_stats_request_ops;
 extern const struct ethnl_request_ops ethnl_phc_vclocks_request_ops;
+extern const struct ethnl_request_ops ethnl_module_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];
@@ -373,6 +374,8 @@ extern const struct nla_policy ethnl_fec_set_policy[ETHTOOL_A_FEC_AUTO + 1];
 extern const struct nla_policy ethnl_module_eeprom_get_policy[ETHTOOL_A_MODULE_EEPROM_I2C_ADDRESS + 1];
 extern const struct nla_policy ethnl_stats_get_policy[ETHTOOL_A_STATS_GROUPS + 1];
 extern const struct nla_policy ethnl_phc_vclocks_get_policy[ETHTOOL_A_PHC_VCLOCKS_HEADER + 1];
+extern const struct nla_policy ethnl_module_get_policy[ETHTOOL_A_MODULE_HEADER + 1];
+extern const struct nla_policy ethnl_module_set_policy[ETHTOOL_A_MODULE_POWER_MODE_POLICY + 1];
 
 int ethnl_set_linkinfo(struct sk_buff *skb, struct genl_info *info);
 int ethnl_set_linkmodes(struct sk_buff *skb, struct genl_info *info);
@@ -391,6 +394,7 @@ 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_set_fec(struct sk_buff *skb, struct genl_info *info);
+int ethnl_set_module(struct sk_buff *skb, struct genl_info *info);
 
 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];
-- 
2.31.1

[RFC PATCH net-next v3 2/6] mlxsw: reg: Add Port Module Memory Map Properties register

[Ido Schimmel]

From: Ido Schimmel <idosch@nvidia.com>

Add the Port Module Memory Map Properties register. It will be used to
set the power mode of a module in subsequent patches.

Signed-off-by: Ido Schimmel <idosch@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlxsw/reg.h | 44 +++++++++++++++++++++++
 1 file changed, 44 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h
index 6c96e124e4c8..fe9bf6ce3508 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/reg.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h
@@ -5946,6 +5946,49 @@ static inline void mlxsw_reg_pddr_pack(char *payload, u8 local_port,
 	mlxsw_reg_pddr_page_select_set(payload, page_select);
 }
 
+/* PMMP - Port Module Memory Map Properties Register
+ * -------------------------------------------------
+ * The PMMP register allows to override the module memory map advertisement.
+ * The register can only be set when the module is disabled by PMAOS register.
+ */
+#define MLXSW_REG_PMMP_ID 0x5044
+#define MLXSW_REG_PMMP_LEN 0x2C
+
+MLXSW_REG_DEFINE(pmmp, MLXSW_REG_PMMP_ID, MLXSW_REG_PMMP_LEN);
+
+/* reg_pmmp_module
+ * Module number.
+ * Access: Index
+ */
+MLXSW_ITEM32(reg, pmmp, module, 0x00, 16, 8);
+
+/* reg_pmmp_eeprom_override_mask
+ * Write mask bit (negative polarity).
+ * 0 - Allow write
+ * 1 - Ignore write
+ * On write, indicates which of the bits from eeprom_override field are
+ * updated.
+ * Access: WO
+ */
+MLXSW_ITEM32(reg, pmmp, eeprom_override_mask, 0x04, 16, 16);
+
+enum {
+	/* Set module to low power mode */
+	MLXSW_REG_PMMP_EEPROM_OVERRIDE_LOW_POWER_MASK = BIT(8),
+};
+
+/* reg_pmmp_eeprom_override
+ * Override / ignore EEPROM advertisement properties bitmask
+ * Access: RW
+ */
+MLXSW_ITEM32(reg, pmmp, eeprom_override, 0x04, 0, 16);
+
+static inline void mlxsw_reg_pmmp_pack(char *payload, u8 module)
+{
+	MLXSW_REG_ZERO(pmmp, payload);
+	mlxsw_reg_pmmp_module_set(payload, module);
+}
+
 /* PLLP - Port Local port to Label Port mapping Register
  * -----------------------------------------------------
  * The PLLP register returns the mapping from Local Port into Label Port.
@@ -12273,6 +12316,7 @@ static const struct mlxsw_reg_info *mlxsw_reg_infos[] = {
 	MLXSW_REG(pmtdb),
 	MLXSW_REG(pmpe),
 	MLXSW_REG(pddr),
+	MLXSW_REG(pmmp),
 	MLXSW_REG(pllp),
 	MLXSW_REG(htgt),
 	MLXSW_REG(hpkt),
-- 
2.31.1

[RFC PATCH net-next v3 3/6] mlxsw: reg: Add Management Cable IO and Notifications register

[Ido Schimmel]

From: Ido Schimmel <idosch@nvidia.com>

Add the Management Cable IO and Notifications register. It will be used
to retrieve the power mode status of a module in subsequent patches and
whether a module is present in a cage or not.

Signed-off-by: Ido Schimmel <idosch@nvidia.com>
---
 drivers/net/ethernet/mellanox/mlxsw/reg.h | 34 +++++++++++++++++++++++
 1 file changed, 34 insertions(+)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/reg.h b/drivers/net/ethernet/mellanox/mlxsw/reg.h
index fe9bf6ce3508..b862e56c1f9e 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/reg.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/reg.h
@@ -10321,6 +10321,39 @@ static inline void mlxsw_reg_mlcr_pack(char *payload, u8 local_port,
 					   MLXSW_REG_MLCR_DURATION_MAX : 0);
 }
 
+/* MCION - Management Cable IO and Notifications Register
+ * ------------------------------------------------------
+ * The MCION register is used to query transceiver modules' IO pins and other
+ * notifications.
+ */
+#define MLXSW_REG_MCION_ID 0x9052
+#define MLXSW_REG_MCION_LEN 0x18
+
+MLXSW_REG_DEFINE(mcion, MLXSW_REG_MCION_ID, MLXSW_REG_MCION_LEN);
+
+/* reg_mcion_module
+ * Module number.
+ * Access: Index
+ */
+MLXSW_ITEM32(reg, mcion, module, 0x00, 16, 8);
+
+enum {
+	MLXSW_REG_MCION_MODULE_STATUS_BITS_PRESENT_MASK = BIT(0),
+	MLXSW_REG_MCION_MODULE_STATUS_BITS_LOW_POWER_MASK = BIT(8),
+};
+
+/* reg_mcion_module_status_bits
+ * Module IO status as defined by SFF.
+ * Access: RO
+ */
+MLXSW_ITEM32(reg, mcion, module_status_bits, 0x04, 0, 16);
+
+static inline void mlxsw_reg_mcion_pack(char *payload, u8 module)
+{
+	MLXSW_REG_ZERO(mcion, payload);
+	mlxsw_reg_mcion_module_set(payload, module);
+}
+
 /* MTPPS - Management Pulse Per Second Register
  * --------------------------------------------
  * This register provides the device PPS capabilities, configure the PPS in and
@@ -12364,6 +12397,7 @@ static const struct mlxsw_reg_info *mlxsw_reg_infos[] = {
 	MLXSW_REG(mgir),
 	MLXSW_REG(mrsr),
 	MLXSW_REG(mlcr),
+	MLXSW_REG(mcion),
 	MLXSW_REG(mtpps),
 	MLXSW_REG(mtutc),
 	MLXSW_REG(mpsc),
-- 
2.31.1

[RFC PATCH net-next v3 4/6] mlxsw: Add ability to control transceiver modules' power mode

[Ido Schimmel]

From: Ido Schimmel <idosch@nvidia.com>

Implement support for ethtool_ops::.get_module_power_mode and
ethtool_ops::set_module_power_mode.

The get operation is implemented using the Management Cable IO and
Notifications (MCION) register that reports the operational power mode
of the module and its presence. In case a module is not present, its
operational power mode is not reported to ethtool and user space. If not
set before, the power mode policy is reported as "high", which is the
default on Mellanox systems.

The set operation is implemented using the Port Module Memory Map
Properties (PMMP) register. The register instructs the device's firmware
to transition a plugged-in module to / out of low power mode by writing
to its memory map.

When the power mode policy is set to 'auto', a module will not
transition to low power mode as long as any ports using it are
administratively up. Example:

 # devlink port split swp11 count 4

 # ethtool --set-module swp11s0 power-mode-policy auto

 $ ethtool --show-module swp11s0
 Module parameters for swp11s0:
 power-mode-policy auto
 power-mode low

 # ip link set dev swp11s0 up

 # ip link set dev swp11s1 up

 $ ethtool --show-module swp11s0
 Module parameters for swp11s0:
 power-mode-policy auto
 power-mode high

 # ip link set dev swp11s1 down

 $ ethtool --show-module swp11s0
 Module parameters for swp11s0:
 power-mode-policy auto
 power-mode high

 # ip link set dev swp11s0 down

 $ ethtool --show-module swp11s0
 Module parameters for swp11s0:
 power-mode-policy auto
 power-mode low

Signed-off-by: Ido Schimmel <idosch@nvidia.com>
---
 .../net/ethernet/mellanox/mlxsw/core_env.c    | 194 +++++++++++++++++-
 .../net/ethernet/mellanox/mlxsw/core_env.h    |  10 +
 drivers/net/ethernet/mellanox/mlxsw/minimal.c |  26 +++
 .../mellanox/mlxsw/spectrum_ethtool.c         |  28 +++
 4 files changed, 255 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/core_env.c b/drivers/net/ethernet/mellanox/mlxsw/core_env.c
index 9e367174743d..0911d9019246 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/core_env.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/core_env.c
@@ -17,6 +17,7 @@ struct mlxsw_env_module_info {
 	bool is_overheat;
 	int num_ports_mapped;
 	int num_ports_up;
+	enum ethtool_module_power_mode_policy power_mode_policy;
 };
 
 struct mlxsw_env {
@@ -445,6 +446,153 @@ int mlxsw_env_reset_module(struct net_device *netdev,
 }
 EXPORT_SYMBOL(mlxsw_env_reset_module);
 
+int
+mlxsw_env_get_module_power_mode(struct mlxsw_core *mlxsw_core, u8 module,
+				struct ethtool_module_power_mode_params *params,
+				struct netlink_ext_ack *extack)
+{
+	struct mlxsw_env *mlxsw_env = mlxsw_core_env(mlxsw_core);
+	char mcion_pl[MLXSW_REG_MCION_LEN];
+	u32 status_bits;
+	int err;
+
+	if (WARN_ON_ONCE(module >= mlxsw_env->module_count))
+		return -EINVAL;
+
+	mutex_lock(&mlxsw_env->module_info_lock);
+
+	params->policy = mlxsw_env->module_info[module].power_mode_policy;
+
+	mlxsw_reg_mcion_pack(mcion_pl, module);
+	err = mlxsw_reg_query(mlxsw_core, MLXSW_REG(mcion), mcion_pl);
+	if (err) {
+		NL_SET_ERR_MSG_MOD(extack, "Failed to retrieve module's power mode");
+		goto out;
+	}
+
+	status_bits = mlxsw_reg_mcion_module_status_bits_get(mcion_pl);
+	if (!(status_bits & MLXSW_REG_MCION_MODULE_STATUS_BITS_PRESENT_MASK))
+		goto out;
+
+	if (status_bits & MLXSW_REG_MCION_MODULE_STATUS_BITS_LOW_POWER_MASK)
+		params->mode = ETHTOOL_MODULE_POWER_MODE_LOW;
+	else
+		params->mode = ETHTOOL_MODULE_POWER_MODE_HIGH;
+
+	params->mode_valid = true;
+
+out:
+	mutex_unlock(&mlxsw_env->module_info_lock);
+	return err;
+}
+EXPORT_SYMBOL(mlxsw_env_get_module_power_mode);
+
+static int mlxsw_env_module_enable_set(struct mlxsw_core *mlxsw_core,
+				       u8 module, bool enable)
+{
+	enum mlxsw_reg_pmaos_admin_status admin_status;
+	char pmaos_pl[MLXSW_REG_PMAOS_LEN];
+
+	mlxsw_reg_pmaos_pack(pmaos_pl, module);
+	admin_status = enable ? MLXSW_REG_PMAOS_ADMIN_STATUS_ENABLED :
+				MLXSW_REG_PMAOS_ADMIN_STATUS_DISABLED;
+	mlxsw_reg_pmaos_admin_status_set(pmaos_pl, admin_status);
+	mlxsw_reg_pmaos_ase_set(pmaos_pl, true);
+
+	return mlxsw_reg_write(mlxsw_core, MLXSW_REG(pmaos), pmaos_pl);
+}
+
+static int mlxsw_env_module_low_power_set(struct mlxsw_core *mlxsw_core,
+					  u8 module, bool low_power)
+{
+	u16 eeprom_override_mask, eeprom_override;
+	char pmmp_pl[MLXSW_REG_PMMP_LEN];
+
+	mlxsw_reg_pmmp_pack(pmmp_pl, module);
+	/* Mask all the bits except low power mode. */
+	eeprom_override_mask = ~MLXSW_REG_PMMP_EEPROM_OVERRIDE_LOW_POWER_MASK;
+	mlxsw_reg_pmmp_eeprom_override_mask_set(pmmp_pl, eeprom_override_mask);
+	eeprom_override = low_power ? MLXSW_REG_PMMP_EEPROM_OVERRIDE_LOW_POWER_MASK :
+				      0;
+	mlxsw_reg_pmmp_eeprom_override_set(pmmp_pl, eeprom_override);
+
+	return mlxsw_reg_write(mlxsw_core, MLXSW_REG(pmmp), pmmp_pl);
+}
+
+static int __mlxsw_env_set_module_power_mode(struct mlxsw_core *mlxsw_core,
+					     u8 module, bool low_power,
+					     struct netlink_ext_ack *extack)
+{
+	int err;
+
+	err = mlxsw_env_module_enable_set(mlxsw_core, module, false);
+	if (err) {
+		NL_SET_ERR_MSG_MOD(extack, "Failed to disable module");
+		return err;
+	}
+
+	err = mlxsw_env_module_low_power_set(mlxsw_core, module, low_power);
+	if (err) {
+		NL_SET_ERR_MSG_MOD(extack, "Failed to set module's power mode");
+		goto err_module_low_power_set;
+	}
+
+	err = mlxsw_env_module_enable_set(mlxsw_core, module, true);
+	if (err) {
+		NL_SET_ERR_MSG_MOD(extack, "Failed to enable module");
+		goto err_module_enable_set;
+	}
+
+	return 0;
+
+err_module_enable_set:
+	mlxsw_env_module_low_power_set(mlxsw_core, module, !low_power);
+err_module_low_power_set:
+	mlxsw_env_module_enable_set(mlxsw_core, module, true);
+	return err;
+}
+
+int
+mlxsw_env_set_module_power_mode(struct mlxsw_core *mlxsw_core, u8 module,
+				enum ethtool_module_power_mode_policy policy,
+				struct netlink_ext_ack *extack)
+{
+	struct mlxsw_env *mlxsw_env = mlxsw_core_env(mlxsw_core);
+	bool low_power;
+	int err = 0;
+
+	if (WARN_ON_ONCE(module >= mlxsw_env->module_count))
+		return -EINVAL;
+
+	if (policy != ETHTOOL_MODULE_POWER_MODE_POLICY_HIGH &&
+	    policy != ETHTOOL_MODULE_POWER_MODE_POLICY_AUTO) {
+		NL_SET_ERR_MSG_MOD(extack, "Unsupported power mode policy");
+		return -EOPNOTSUPP;
+	}
+
+	mutex_lock(&mlxsw_env->module_info_lock);
+
+	if (mlxsw_env->module_info[module].power_mode_policy == policy)
+		goto out;
+
+	/* If any ports are up, we are already in high power mode. */
+	if (mlxsw_env->module_info[module].num_ports_up)
+		goto out_set_policy;
+
+	low_power = policy == ETHTOOL_MODULE_POWER_MODE_POLICY_AUTO;
+	err = __mlxsw_env_set_module_power_mode(mlxsw_core, module, low_power,
+						extack);
+	if (err)
+		goto out;
+
+out_set_policy:
+	mlxsw_env->module_info[module].power_mode_policy = policy;
+out:
+	mutex_unlock(&mlxsw_env->module_info_lock);
+	return err;
+}
+EXPORT_SYMBOL(mlxsw_env_set_module_power_mode);
+
 static int mlxsw_env_module_has_temp_sensor(struct mlxsw_core *mlxsw_core,
 					    u8 module,
 					    bool *p_has_temp_sensor)
@@ -794,15 +942,33 @@ EXPORT_SYMBOL(mlxsw_env_module_port_unmap);
 int mlxsw_env_module_port_up(struct mlxsw_core *mlxsw_core, u8 module)
 {
 	struct mlxsw_env *mlxsw_env = mlxsw_core_env(mlxsw_core);
+	int err = 0;
 
 	if (WARN_ON_ONCE(module >= mlxsw_env->module_count))
 		return -EINVAL;
 
 	mutex_lock(&mlxsw_env->module_info_lock);
+
+	if (mlxsw_env->module_info[module].power_mode_policy !=
+	    ETHTOOL_MODULE_POWER_MODE_POLICY_AUTO)
+		goto out_inc;
+
+	if (mlxsw_env->module_info[module].num_ports_up != 0)
+		goto out_inc;
+
+	/* Transition to high power mode following first port using the module
+	 * being put administratively up.
+	 */
+	err = __mlxsw_env_set_module_power_mode(mlxsw_core, module, false,
+						NULL);
+	if (err)
+		goto out_unlock;
+
+out_inc:
 	mlxsw_env->module_info[module].num_ports_up++;
+out_unlock:
 	mutex_unlock(&mlxsw_env->module_info_lock);
-
-	return 0;
+	return err;
 }
 EXPORT_SYMBOL(mlxsw_env_module_port_up);
 
@@ -814,7 +980,22 @@ void mlxsw_env_module_port_down(struct mlxsw_core *mlxsw_core, u8 module)
 		return;
 
 	mutex_lock(&mlxsw_env->module_info_lock);
+
 	mlxsw_env->module_info[module].num_ports_up--;
+
+	if (mlxsw_env->module_info[module].power_mode_policy !=
+	    ETHTOOL_MODULE_POWER_MODE_POLICY_AUTO)
+		goto out_unlock;
+
+	if (mlxsw_env->module_info[module].num_ports_up != 0)
+		goto out_unlock;
+
+	/* Transition to low power mode following last port using the module
+	 * being put administratively down.
+	 */
+	__mlxsw_env_set_module_power_mode(mlxsw_core, module, true, NULL);
+
+out_unlock:
 	mutex_unlock(&mlxsw_env->module_info_lock);
 }
 EXPORT_SYMBOL(mlxsw_env_module_port_down);
@@ -824,7 +1005,7 @@ int mlxsw_env_init(struct mlxsw_core *mlxsw_core, struct mlxsw_env **p_env)
 	char mgpir_pl[MLXSW_REG_MGPIR_LEN];
 	struct mlxsw_env *env;
 	u8 module_count;
-	int err;
+	int i, err;
 
 	mlxsw_reg_mgpir_pack(mgpir_pl);
 	err = mlxsw_reg_query(mlxsw_core, MLXSW_REG(mgpir), mgpir_pl);
@@ -837,6 +1018,13 @@ int mlxsw_env_init(struct mlxsw_core *mlxsw_core, struct mlxsw_env **p_env)
 	if (!env)
 		return -ENOMEM;
 
+	/* Firmware defaults to high power mode policy where modules are
+	 * transitioned to high power mode following plug-in.
+	 */
+	for (i = 0; i < module_count; i++)
+		env->module_info[i].power_mode_policy =
+			ETHTOOL_MODULE_POWER_MODE_POLICY_HIGH;
+
 	mutex_init(&env->module_info_lock);
 	env->core = mlxsw_core;
 	env->module_count = module_count;
diff --git a/drivers/net/ethernet/mellanox/mlxsw/core_env.h b/drivers/net/ethernet/mellanox/mlxsw/core_env.h
index c486397f5dfe..da121b1a84b4 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/core_env.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/core_env.h
@@ -28,6 +28,16 @@ int mlxsw_env_reset_module(struct net_device *netdev,
 			   struct mlxsw_core *mlxsw_core, u8 module,
 			   u32 *flags);
 
+int
+mlxsw_env_get_module_power_mode(struct mlxsw_core *mlxsw_core, u8 module,
+				struct ethtool_module_power_mode_params *params,
+				struct netlink_ext_ack *extack);
+
+int
+mlxsw_env_set_module_power_mode(struct mlxsw_core *mlxsw_core, u8 module,
+				enum ethtool_module_power_mode_policy policy,
+				struct netlink_ext_ack *extack);
+
 int
 mlxsw_env_module_overheat_counter_get(struct mlxsw_core *mlxsw_core, u8 module,
 				      u64 *p_counter);
diff --git a/drivers/net/ethernet/mellanox/mlxsw/minimal.c b/drivers/net/ethernet/mellanox/mlxsw/minimal.c
index 9644e9c486b8..e0892f259adf 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/minimal.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/minimal.c
@@ -145,12 +145,38 @@ static int mlxsw_m_reset(struct net_device *netdev, u32 *flags)
 				      flags);
 }
 
+static int
+mlxsw_m_get_module_power_mode(struct net_device *netdev,
+			      struct ethtool_module_power_mode_params *params,
+			      struct netlink_ext_ack *extack)
+{
+	struct mlxsw_m_port *mlxsw_m_port = netdev_priv(netdev);
+	struct mlxsw_core *core = mlxsw_m_port->mlxsw_m->core;
+
+	return mlxsw_env_get_module_power_mode(core, mlxsw_m_port->module,
+					       params, extack);
+}
+
+static int
+mlxsw_m_set_module_power_mode(struct net_device *netdev,
+			      const struct ethtool_module_power_mode_params *params,
+			      struct netlink_ext_ack *extack)
+{
+	struct mlxsw_m_port *mlxsw_m_port = netdev_priv(netdev);
+	struct mlxsw_core *core = mlxsw_m_port->mlxsw_m->core;
+
+	return mlxsw_env_set_module_power_mode(core, mlxsw_m_port->module,
+					       params->policy, extack);
+}
+
 static const struct ethtool_ops mlxsw_m_port_ethtool_ops = {
 	.get_drvinfo		= mlxsw_m_module_get_drvinfo,
 	.get_module_info	= mlxsw_m_get_module_info,
 	.get_module_eeprom	= mlxsw_m_get_module_eeprom,
 	.get_module_eeprom_by_page = mlxsw_m_get_module_eeprom_by_page,
 	.reset			= mlxsw_m_reset,
+	.get_module_power_mode	= mlxsw_m_get_module_power_mode,
+	.set_module_power_mode	= mlxsw_m_set_module_power_mode,
 };
 
 static int
diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ethtool.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ethtool.c
index 06f1645561c6..7329bc84a8ee 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ethtool.c
@@ -1206,6 +1206,32 @@ static int mlxsw_sp_reset(struct net_device *dev, u32 *flags)
 	return mlxsw_env_reset_module(dev, mlxsw_sp->core, module, flags);
 }
 
+static int
+mlxsw_sp_get_module_power_mode(struct net_device *dev,
+			       struct ethtool_module_power_mode_params *params,
+			       struct netlink_ext_ack *extack)
+{
+	struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev);
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	u8 module = mlxsw_sp_port->mapping.module;
+
+	return mlxsw_env_get_module_power_mode(mlxsw_sp->core, module, params,
+					       extack);
+}
+
+static int
+mlxsw_sp_set_module_power_mode(struct net_device *dev,
+			       const struct ethtool_module_power_mode_params *params,
+			       struct netlink_ext_ack *extack)
+{
+	struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev);
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	u8 module = mlxsw_sp_port->mapping.module;
+
+	return mlxsw_env_set_module_power_mode(mlxsw_sp->core, module,
+					       params->policy, extack);
+}
+
 const struct ethtool_ops mlxsw_sp_port_ethtool_ops = {
 	.cap_link_lanes_supported	= true,
 	.get_drvinfo			= mlxsw_sp_port_get_drvinfo,
@@ -1228,6 +1254,8 @@ const struct ethtool_ops mlxsw_sp_port_ethtool_ops = {
 	.get_eth_ctrl_stats		= mlxsw_sp_get_eth_ctrl_stats,
 	.get_rmon_stats			= mlxsw_sp_get_rmon_stats,
 	.reset				= mlxsw_sp_reset,
+	.get_module_power_mode		= mlxsw_sp_get_module_power_mode,
+	.set_module_power_mode		= mlxsw_sp_set_module_power_mode,
 };
 
 struct mlxsw_sp1_port_link_mode {
-- 
2.31.1

[RFC PATCH net-next v3 5/6] ethtool: Add transceiver module extended states

[Ido Schimmel]

From: Ido Schimmel <idosch@nvidia.com>

Add an extended state and two extended sub-states to describe link
issues related to transceiver modules.

The first, 'ETHTOOL_LINK_EXT_SUBSTATE_MODULE_LOW_POWER_MODE', tells user
space that port is unable to gain a carrier because the associated
transceiver module is in low power mode where the data path is
deactivated. This is applicable to both SFF-8636 and CMIS modules.
Currently, user space cannot force a module to stay in low power mode
while putting the associated port administratively up, so the extended
sub-state is indicative of a problem in the module/driver.

The second, 'ETHTOOL_LINK_EXT_SUBSTATE_MODULE_CMIS_NOT_READY', tells
user space that port is unable to gain a carrier because the CMIS Module
State Machine did not reach the ModuleReady (Fully Operational) state.
For example, if the module is stuck at ModuleFault state. In which case,
user space can read the fault reason from the module's EEPROM and
potentially reset it.

For CMIS modules, the first extended sub-state is contained in the
second, but has the added advantage of being applicable to more module
types and being more specific about the nature of the problem.

Signed-off-by: Ido Schimmel <idosch@nvidia.com>
---
 Documentation/networking/ethtool-netlink.rst | 12 ++++++++++++
 include/linux/ethtool.h                      |  1 +
 include/uapi/linux/ethtool.h                 |  7 +++++++
 3 files changed, 20 insertions(+)

diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst
index 245ce2eab9a5..c258b3f30a2e 100644
--- a/Documentation/networking/ethtool-netlink.rst
+++ b/Documentation/networking/ethtool-netlink.rst
@@ -523,6 +523,8 @@ Link extended states:
                                                         power required from cable or module
 
   ``ETHTOOL_LINK_EXT_STATE_OVERHEAT``                   The module is overheated
+
+  ``ETHTOOL_LINK_EXT_STATE_MODULE``                     Transceiver module issue
   ================================================      ============================================
 
 Link extended substates:
@@ -616,6 +618,16 @@ Link extended substates:
   ``ETHTOOL_LINK_EXT_SUBSTATE_CI_CABLE_TEST_FAILURE``   Cable test failure
   ===================================================   ============================================
 
+  Transceiver module issue substates:
+
+  ===================================================   ============================================
+  ``ETHTOOL_LINK_EXT_SUBSTATE_MODULE_LOW_POWER_MODE``   The transceiver module is in low power mode
+
+  ``ETHTOOL_LINK_EXT_SUBSTATE_MODULE_CMIS_NOT_READY``   The CMIS Module State Machine did not reach
+                                                        the ModuleReady state. For example, if the
+                                                        module is stuck at ModuleFault state
+  ===================================================   ============================================
+
 DEBUG_GET
 =========
 
diff --git a/include/linux/ethtool.h b/include/linux/ethtool.h
index 07d40dc20ca4..1f71293011ca 100644
--- a/include/linux/ethtool.h
+++ b/include/linux/ethtool.h
@@ -93,6 +93,7 @@ struct ethtool_link_ext_state_info {
 		enum ethtool_link_ext_substate_link_logical_mismatch link_logical_mismatch;
 		enum ethtool_link_ext_substate_bad_signal_integrity bad_signal_integrity;
 		enum ethtool_link_ext_substate_cable_issue cable_issue;
+		enum ethtool_link_ext_substate_module module;
 		u8 __link_ext_substate;
 	};
 };
diff --git a/include/uapi/linux/ethtool.h b/include/uapi/linux/ethtool.h
index 8cc79811ee5d..7d453f0e993b 100644
--- a/include/uapi/linux/ethtool.h
+++ b/include/uapi/linux/ethtool.h
@@ -603,6 +603,7 @@ enum ethtool_link_ext_state {
 	ETHTOOL_LINK_EXT_STATE_CALIBRATION_FAILURE,
 	ETHTOOL_LINK_EXT_STATE_POWER_BUDGET_EXCEEDED,
 	ETHTOOL_LINK_EXT_STATE_OVERHEAT,
+	ETHTOOL_LINK_EXT_STATE_MODULE,
 };
 
 /* More information in addition to ETHTOOL_LINK_EXT_STATE_AUTONEG. */
@@ -649,6 +650,12 @@ enum ethtool_link_ext_substate_cable_issue {
 	ETHTOOL_LINK_EXT_SUBSTATE_CI_CABLE_TEST_FAILURE,
 };
 
+/* More information in addition to ETHTOOL_LINK_EXT_STATE_MODULE. */
+enum ethtool_link_ext_substate_module {
+	ETHTOOL_LINK_EXT_SUBSTATE_MODULE_LOW_POWER_MODE = 1,
+	ETHTOOL_LINK_EXT_SUBSTATE_MODULE_CMIS_NOT_READY,
+};
+
 #define ETH_GSTRING_LEN		32
 
 /**
-- 
2.31.1

[RFC PATCH net-next v3 6/6] mlxsw: Add support for transceiver module extended states

[Ido Schimmel]

From: Ido Schimmel <idosch@nvidia.com>

Add support for the two transceiver module extended sub-states added in
previous patch. The two extended sub-states are meant to describe link
issues related to transceiver modules.

Signed-off-by: Ido Schimmel <idosch@nvidia.com>
---
 .../mellanox/mlxsw/spectrum_ethtool.c         | 45 ++++++++++++++++++-
 1 file changed, 43 insertions(+), 2 deletions(-)

diff --git a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ethtool.c b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ethtool.c
index 7329bc84a8ee..337fa0053de9 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/spectrum_ethtool.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/spectrum_ethtool.c
@@ -96,6 +96,11 @@ mlxsw_sp_link_ext_state_opcode_map[] = {
 	{1032, ETHTOOL_LINK_EXT_STATE_POWER_BUDGET_EXCEEDED, 0},
 
 	{1030, ETHTOOL_LINK_EXT_STATE_OVERHEAT, 0},
+
+	{1042, ETHTOOL_LINK_EXT_STATE_MODULE,
+	 ETHTOOL_LINK_EXT_SUBSTATE_MODULE_CMIS_NOT_READY},
+	{2048, ETHTOOL_LINK_EXT_STATE_MODULE,
+	 ETHTOOL_LINK_EXT_SUBSTATE_MODULE_LOW_POWER_MODE},
 };
 
 static void
@@ -124,6 +129,10 @@ mlxsw_sp_port_set_link_ext_state(struct mlxsw_sp_ethtool_link_ext_state_opcode_m
 		link_ext_state_info->cable_issue =
 			link_ext_state_mapping.link_ext_substate;
 		break;
+	case ETHTOOL_LINK_EXT_STATE_MODULE:
+		link_ext_state_info->module =
+			link_ext_state_mapping.link_ext_substate;
+		break;
 	default:
 		break;
 	}
@@ -131,19 +140,46 @@ mlxsw_sp_port_set_link_ext_state(struct mlxsw_sp_ethtool_link_ext_state_opcode_m
 	link_ext_state_info->link_ext_state = link_ext_state_mapping.link_ext_state;
 }
 
+static int
+mlxsw_sp_port_status_opcode_drv_get(struct mlxsw_sp_port *mlxsw_sp_port,
+				    u32 *p_status_opcode)
+{
+	struct ethtool_module_power_mode_params params = {};
+	struct mlxsw_sp *mlxsw_sp = mlxsw_sp_port->mlxsw_sp;
+	u8 module = mlxsw_sp_port->mapping.module;
+	int err;
+
+	*p_status_opcode = 0;
+
+	err = mlxsw_env_get_module_power_mode(mlxsw_sp->core, module, &params,
+					      NULL);
+	if (err)
+		return err;
+	if (params.mode_valid && params.mode == ETHTOOL_MODULE_POWER_MODE_LOW)
+		*p_status_opcode = 2048;
+
+	return 0;
+}
+
 static int
 mlxsw_sp_port_get_link_ext_state(struct net_device *dev,
 				 struct ethtool_link_ext_state_info *link_ext_state_info)
 {
 	struct mlxsw_sp_ethtool_link_ext_state_opcode_mapping link_ext_state_mapping;
 	struct mlxsw_sp_port *mlxsw_sp_port = netdev_priv(dev);
+	u32 status_opcode, status_opcode_drv;
 	char pddr_pl[MLXSW_REG_PDDR_LEN];
 	int opcode, err, i;
-	u32 status_opcode;
 
 	if (netif_carrier_ok(dev))
 		return -ENODATA;
 
+	/* Opcodes 2048-3072 are reserved for driver use. */
+	err = mlxsw_sp_port_status_opcode_drv_get(mlxsw_sp_port,
+						  &status_opcode_drv);
+	if (err)
+		return err;
+
 	mlxsw_reg_pddr_pack(pddr_pl, mlxsw_sp_port->local_port,
 			    MLXSW_REG_PDDR_PAGE_SELECT_TROUBLESHOOTING_INFO);
 
@@ -156,9 +192,14 @@ mlxsw_sp_port_get_link_ext_state(struct net_device *dev,
 		return err;
 
 	status_opcode = mlxsw_reg_pddr_trblsh_status_opcode_get(pddr_pl);
-	if (!status_opcode)
+	if (!status_opcode && !status_opcode_drv)
 		return -ENODATA;
 
+	/* Allow driver-detected issues to take precedence, as it is likely
+	 * that they are more meaningful to user space.
+	 */
+	status_opcode = status_opcode_drv ? status_opcode_drv : status_opcode;
+
 	for (i = 0; i < ARRAY_SIZE(mlxsw_sp_link_ext_state_opcode_map); i++) {
 		link_ext_state_mapping = mlxsw_sp_link_ext_state_opcode_map[i];
 		if (link_ext_state_mapping.status_opcode == status_opcode) {
-- 
2.31.1

Re: [RFC PATCH net-next v3 1/6] ethtool: Add ability to control transceiver modules' power mode

[Jakub Kicinski]

On Tue, 24 Aug 2021 16:03:39 +0300 Ido Schimmel wrote:
> From: Ido Schimmel <idosch@nvidia.com>
> 
> Add a pair of new ethtool messages, 'ETHTOOL_MSG_MODULE_SET' and
> 'ETHTOOL_MSG_MODULE_GET', that can be used to control transceiver
> modules parameters and retrieve their status.

Lgtm! A few "take it or leave it" nit picks below.

> Signed-off-by: Ido Schimmel <idosch@nvidia.com>

> +The optional ``ETHTOOL_A_MODULE_POWER_MODE_POLICY`` attribute encodes the
> +transceiver module power mode policy enforced by the host. The default policy
> +is driver-dependent and can be queried using this attribute.

Should we make a recommendation for those who don't have to worry about
legacy behavior? Like:

  The default policy is driver-dependent (but "auto" is the recommended
  and generally assumed to be used for drivers no implementing this API).

IMHO the "and can be queried using this attribute" part can be skipped.

> +/**
> + * struct ethtool_module_power_mode_params - module power mode parameters
> + * @policy: The power mode policy enforced by the host for the plug-in module.
> + * @mode: The operational power mode of the plug-in module. Should be filled by
> + * device drivers on get operations.

Indent continuation lines by one tab.

> + * @mode_valid: Indicates the validity of the @mode field. Should be set by
> + * device drivers on get operations when a module is plugged-in.

Should we make a firm decision on whether we want to use these kind of
valid bits or choose invalid defaults? As you may guess my preference
is the latter since that's what I usually do, that way drivers don't
have to write two fields.

Actually I think this may be the first "valid" in ethtool, I thought we
already had one but I don't see it now..

> +struct ethtool_module_power_mode_params {
> +	enum ethtool_module_power_mode_policy policy;
> +	enum ethtool_module_power_mode mode;
> +	u8 mode_valid:1;
> +};

RE: [RFC PATCH net-next v3 1/6] ethtool: Add ability to control transceiver modules' power mode

[Keller, Jacob E]

> -----Original Message-----
> From: Jakub Kicinski <kuba@kernel.org>
> Sent: Tuesday, August 24, 2021 4:13 PM
> To: Ido Schimmel <idosch@idosch.org>
> Cc: netdev@vger.kernel.org; davem@davemloft.net; andrew@lunn.ch;
> mkubecek@suse.cz; pali@kernel.org; Keller, Jacob E <jacob.e.keller@intel.com>;
> jiri@nvidia.com; vadimp@nvidia.com; mlxsw@nvidia.com; Ido Schimmel
> <idosch@nvidia.com>
> Subject: Re: [RFC PATCH net-next v3 1/6] ethtool: Add ability to control
> transceiver modules' power mode
> 
> On Tue, 24 Aug 2021 16:03:39 +0300 Ido Schimmel wrote:
> > From: Ido Schimmel <idosch@nvidia.com>
> >
> > Add a pair of new ethtool messages, 'ETHTOOL_MSG_MODULE_SET' and
> > 'ETHTOOL_MSG_MODULE_GET', that can be used to control transceiver
> > modules parameters and retrieve their status.
> 
> Lgtm! A few "take it or leave it" nit picks below.
> 
> > Signed-off-by: Ido Schimmel <idosch@nvidia.com>
> 
> > +The optional ``ETHTOOL_A_MODULE_POWER_MODE_POLICY`` attribute
> encodes the
> > +transceiver module power mode policy enforced by the host. The default
> policy
> > +is driver-dependent and can be queried using this attribute.
> 
> Should we make a recommendation for those who don't have to worry about
> legacy behavior? Like:
> 
>   The default policy is driver-dependent (but "auto" is the recommended
>   and generally assumed to be used for drivers no implementing this API).
> 
> IMHO the "and can be queried using this attribute" part can be skipped.
> 
> > +/**
> > + * struct ethtool_module_power_mode_params - module power mode
> parameters
> > + * @policy: The power mode policy enforced by the host for the plug-in
> module.
> > + * @mode: The operational power mode of the plug-in module. Should be
> filled by
> > + * device drivers on get operations.
> 
> Indent continuation lines by one tab.
> 
> > + * @mode_valid: Indicates the validity of the @mode field. Should be set by
> > + * device drivers on get operations when a module is plugged-in.
> 
> Should we make a firm decision on whether we want to use these kind of
> valid bits or choose invalid defaults? As you may guess my preference
> is the latter since that's what I usually do, that way drivers don't
> have to write two fields.
> 
> Actually I think this may be the first "valid" in ethtool, I thought we
> already had one but I don't see it now..
> 


coalesce settings have a valid mode don't they? Or at least an "accepted modes"?

Thanks,
Jake

> > +struct ethtool_module_power_mode_params {
> > +	enum ethtool_module_power_mode_policy policy;
> > +	enum ethtool_module_power_mode mode;
> > +	u8 mode_valid:1;
> > +};

Re: [RFC PATCH net-next v3 1/6] ethtool: Add ability to control transceiver modules' power mode

[Michal Kubecek]

On Tue, Aug 24, 2021 at 11:18:56PM +0000, Keller, Jacob E wrote:
> > -----Original Message-----
> > From: Jakub Kicinski <kuba@kernel.org>
> > Sent: Tuesday, August 24, 2021 4:13 PM
> > To: Ido Schimmel <idosch@idosch.org>
> > Cc: netdev@vger.kernel.org; davem@davemloft.net; andrew@lunn.ch;
> > mkubecek@suse.cz; pali@kernel.org; Keller, Jacob E <jacob.e.keller@intel.com>;
> > jiri@nvidia.com; vadimp@nvidia.com; mlxsw@nvidia.com; Ido Schimmel
> > <idosch@nvidia.com>
> > Subject: Re: [RFC PATCH net-next v3 1/6] ethtool: Add ability to control
> > transceiver modules' power mode
> > 
> > On Tue, 24 Aug 2021 16:03:39 +0300 Ido Schimmel wrote:
[...]
> > > +/**
> > > + * struct ethtool_module_power_mode_params - module power mode
> > parameters
> > > + * @policy: The power mode policy enforced by the host for the plug-in
> > module.
> > > + * @mode: The operational power mode of the plug-in module. Should be
> > filled by
> > > + * device drivers on get operations.
> > 
> > Indent continuation lines by one tab.
> > 
> > > + * @mode_valid: Indicates the validity of the @mode field. Should be set by
> > > + * device drivers on get operations when a module is plugged-in.
> > 
> > Should we make a firm decision on whether we want to use these kind of
> > valid bits or choose invalid defaults? As you may guess my preference
> > is the latter since that's what I usually do, that way drivers don't
> > have to write two fields.
> > 
> > Actually I think this may be the first "valid" in ethtool, I thought we
> > already had one but I don't see it now..
> > 
> 
> 
> coalesce settings have a valid mode don't they? Or at least an
> "accepted modes"?

That's different, IIUC. In coalesce settings, the driver declares which
parameters are allowed in "set" requests so that this part of request
checks can be done in the general ethtool code rather than in each
driver separately. Here the "valid" field says whether mode field holds
a meaningful value or should be ignored.

Michal

Re: [RFC PATCH net-next v3 1/6] ethtool: Add ability to control transceiver modules' power mode

[Jakub Kicinski]

On Tue, 24 Aug 2021 23:18:56 +0000 Keller, Jacob E wrote:
> > > + * @mode_valid: Indicates the validity of the @mode field. Should be set by
> > > + * device drivers on get operations when a module is plugged-in.  
> > 
> > Should we make a firm decision on whether we want to use these kind of
> > valid bits or choose invalid defaults? As you may guess my preference
> > is the latter since that's what I usually do, that way drivers don't
> > have to write two fields.
> > 
> > Actually I think this may be the first "valid" in ethtool, I thought we
> > already had one but I don't see it now..
> 
> coalesce settings have a valid mode don't they? Or at least an "accepted modes"?

That's a static per-driver bitmask 'cause we don't trust driver writers
to error out on all the unsupported fields. The driver code doesn't
operate on it in the callbacks.

RE: [RFC PATCH net-next v3 1/6] ethtool: Add ability to control transceiver modules' power mode

[Keller, Jacob E]

> -----Original Message-----
> From: Jakub Kicinski <kuba@kernel.org>
> Sent: Tuesday, August 24, 2021 4:48 PM
> To: Keller, Jacob E <jacob.e.keller@intel.com>
> Cc: Ido Schimmel <idosch@idosch.org>; netdev@vger.kernel.org;
> davem@davemloft.net; andrew@lunn.ch; mkubecek@suse.cz; pali@kernel.org;
> jiri@nvidia.com; vadimp@nvidia.com; mlxsw@nvidia.com; Ido Schimmel
> <idosch@nvidia.com>
> Subject: Re: [RFC PATCH net-next v3 1/6] ethtool: Add ability to control
> transceiver modules' power mode
> 
> On Tue, 24 Aug 2021 23:18:56 +0000 Keller, Jacob E wrote:
> > > > + * @mode_valid: Indicates the validity of the @mode field. Should be set
> by
> > > > + * device drivers on get operations when a module is plugged-in.
> > >
> > > Should we make a firm decision on whether we want to use these kind of
> > > valid bits or choose invalid defaults? As you may guess my preference
> > > is the latter since that's what I usually do, that way drivers don't
> > > have to write two fields.
> > >
> > > Actually I think this may be the first "valid" in ethtool, I thought we
> > > already had one but I don't see it now..
> >
> > coalesce settings have a valid mode don't they? Or at least an "accepted
> modes"?
> 
> That's a static per-driver bitmask 'cause we don't trust driver writers
> to error out on all the unsupported fields. The driver code doesn't
> operate on it in the callbacks.

Ahh. Right. Ok yea this is different here.

If we can keep it simple in the drivers, great! I usually like the valid approach, but mostly if its kernel-core code doing it since we're more likely to catch and review that as opposed to individual drivers.

If we're expecting drivers to set validity, a simpler interface would be preferable.

Thanks,
Jake

Re: [RFC PATCH net-next v3 1/6] ethtool: Add ability to control transceiver modules' power mode

[Ido Schimmel]

On Tue, Aug 24, 2021 at 04:12:31PM -0700, Jakub Kicinski wrote:
> On Tue, 24 Aug 2021 16:03:39 +0300 Ido Schimmel wrote:
> > From: Ido Schimmel <idosch@nvidia.com>
> > 
> > Add a pair of new ethtool messages, 'ETHTOOL_MSG_MODULE_SET' and
> > 'ETHTOOL_MSG_MODULE_GET', that can be used to control transceiver
> > modules parameters and retrieve their status.
> 
> Lgtm! A few "take it or leave it" nit picks below.

Thanks and thanks a lot for the reviews!

> 
> > Signed-off-by: Ido Schimmel <idosch@nvidia.com>
> 
> > +The optional ``ETHTOOL_A_MODULE_POWER_MODE_POLICY`` attribute encodes the
> > +transceiver module power mode policy enforced by the host. The default policy
> > +is driver-dependent and can be queried using this attribute.
> 
> Should we make a recommendation for those who don't have to worry about
> legacy behavior? 

Yes

> Like:
> 
>   The default policy is driver-dependent (but "auto" is the recommended
>   and generally assumed to be used for drivers no implementing this API).

I think "generally assumed to be used for drivers no implementing this
API" is problematic given that it is most likely the exact opposite of
what actually happens. I imagine most vendors supporting these modules
just went with "high" policy instead of implementing "auto" policy in
firmware.

So I suggest:

"The default policy is driver-dependent, but "auto" is the recommended
default and it should be implemented by new drivers and drivers where
conformance to a legacy behavior is not critical."

> 
> IMHO the "and can be queried using this attribute" part can be skipped.

OK

> 
> > +/**
> > + * struct ethtool_module_power_mode_params - module power mode parameters
> > + * @policy: The power mode policy enforced by the host for the plug-in module.
> > + * @mode: The operational power mode of the plug-in module. Should be filled by
> > + * device drivers on get operations.
> 
> Indent continuation lines by one tab.

Oops, I see that I did do that for other kdoc comments. Will fix.

> 
> > + * @mode_valid: Indicates the validity of the @mode field. Should be set by
> > + * device drivers on get operations when a module is plugged-in.
> 
> Should we make a firm decision on whether we want to use these kind of
> valid bits or choose invalid defaults? As you may guess my preference
> is the latter since that's what I usually do, that way drivers don't
> have to write two fields.
> 
> Actually I think this may be the first "valid" in ethtool, I thought we
> already had one but I don't see it now..

I was thinking about this as well, but I wasn't sure if it's valid to
adjust uAPI values in order to make in-kernel APIs simpler. I did see it
in some other places, but wasn't sure if it's a pattern that should be
copied.

Do you mean something like this?

diff --git a/include/uapi/linux/ethtool.h b/include/uapi/linux/ethtool.h
index 7d453f0e993b..d61049091538 100644
--- a/include/uapi/linux/ethtool.h
+++ b/include/uapi/linux/ethtool.h
@@ -732,7 +732,7 @@ enum ethtool_module_power_mode_policy {
  * @ETHTOOL_MODULE_POWER_MODE_HIGH: Module is in high power mode.
  */
 enum ethtool_module_power_mode {
-       ETHTOOL_MODULE_POWER_MODE_LOW,
+       ETHTOOL_MODULE_POWER_MODE_LOW = 1,
        ETHTOOL_MODULE_POWER_MODE_HIGH,
 };

I prefer this over memsetting a struct to 0xff.

If the above is fine, I can make the following patch:

diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst
index c258b3f30a2e..d304df39ee5c 100644
--- a/Documentation/networking/ethtool-netlink.rst
+++ b/Documentation/networking/ethtool-netlink.rst
@@ -41,6 +41,11 @@ In the message structure descriptions below, if an attribute name is suffixed
 with "+", parent nest can contain multiple attributes of the same type. This
 implements an array of entries.
 
+Attributes that need to be filled-in by device drivers and that are dumped to
+user space based on whether they are valid or not should not use zero as a
+valid value. For example, ``ETHTOOL_A_MODULE_POWER_MODE``. This avoids the need
+to explicitly signal the validity of the attribute in the device driver API.
+
 
 Request header
 ==============

> 
> > +struct ethtool_module_power_mode_params {
> > +	enum ethtool_module_power_mode_policy policy;
> > +	enum ethtool_module_power_mode mode;
> > +	u8 mode_valid:1;
> > +};

Re: [RFC PATCH net-next v3 1/6] ethtool: Add ability to control transceiver modules' power mode

[Jakub Kicinski]

On Wed, 25 Aug 2021 14:14:28 +0300 Ido Schimmel wrote:
> So I suggest:
> 
> "The default policy is driver-dependent, but "auto" is the recommended
> default and it should be implemented by new drivers and drivers where
> conformance to a legacy behavior is not critical."

SGTM

> > > + * @mode_valid: Indicates the validity of the @mode field. Should be set by
> > > + * device drivers on get operations when a module is plugged-in.  
> > 
> > Should we make a firm decision on whether we want to use these kind of
> > valid bits or choose invalid defaults? As you may guess my preference
> > is the latter since that's what I usually do, that way drivers don't
> > have to write two fields.
> > 
> > Actually I think this may be the first "valid" in ethtool, I thought we
> > already had one but I don't see it now..  
> 
> I was thinking about this as well, but I wasn't sure if it's valid to
> adjust uAPI values in order to make in-kernel APIs simpler. I did see it
> in some other places, but wasn't sure if it's a pattern that should be
> copied.
> 
> Do you mean something like this?
> 
> diff --git a/include/uapi/linux/ethtool.h b/include/uapi/linux/ethtool.h
> index 7d453f0e993b..d61049091538 100644
> --- a/include/uapi/linux/ethtool.h
> +++ b/include/uapi/linux/ethtool.h
> @@ -732,7 +732,7 @@ enum ethtool_module_power_mode_policy {
>   * @ETHTOOL_MODULE_POWER_MODE_HIGH: Module is in high power mode.
>   */
>  enum ethtool_module_power_mode {
> -       ETHTOOL_MODULE_POWER_MODE_LOW,
> +       ETHTOOL_MODULE_POWER_MODE_LOW = 1,
>         ETHTOOL_MODULE_POWER_MODE_HIGH,
>  };
> 
> I prefer this over memsetting a struct to 0xff.

Yup, I mean we mostly care about the policy but can adjust the state
enum as well ;) For stats 0 was a no-go, obviously but in general it
should work perfectly.

> If the above is fine, I can make the following patch:
> 
> diff --git a/Documentation/networking/ethtool-netlink.rst b/Documentation/networking/ethtool-netlink.rst
> index c258b3f30a2e..d304df39ee5c 100644
> --- a/Documentation/networking/ethtool-netlink.rst
> +++ b/Documentation/networking/ethtool-netlink.rst
> @@ -41,6 +41,11 @@ In the message structure descriptions below, if an attribute name is suffixed
>  with "+", parent nest can contain multiple attributes of the same type. This
>  implements an array of entries.
>  
> +Attributes that need to be filled-in by device drivers and that are dumped to
> +user space based on whether they are valid or not should not use zero as a
> +valid value. For example, ``ETHTOOL_A_MODULE_POWER_MODE``. This avoids the need
> +to explicitly signal the validity of the attribute in the device driver API.
> +

Modulo the enum in question, but 👍