From: Ioana Ciornei <ioana.ciornei@nxp.com>
This patch set provides ACPI support to DPAA2 network drivers.
It also introduces new fwnode based APIs to support phylink and phy
layers
Following functions are defined:
phylink_fwnode_phy_connect()
fwnode_mdiobus_register_phy()
fwnode_get_phy_id()
fwnode_phy_find_device()
device_phy_find_device()
fwnode_get_phy_node()
fwnode_mdio_find_device()
acpi_get_local_address()
First one helps in connecting phy to phylink instance.
Next three helps in getting phy_id and registering phy to mdiobus
Next two help in finding a phy on a mdiobus.
Next one helps in getting phy_node from a fwnode.
Last one is used to get local address from _ADR object.
Corresponding OF functions are refactored.
Tested-on: LX2160ARDB
Changes in v9:
- merged some minimal changes requested in the wording of the commit
messages
- fixed some build problems in patch 8/15 by moving the removal of
of_find_mii_timestamper from patch 8/15 to 9/15.
Changes in v8:
- fixed some checkpatch warnings/checks
- included linux/fwnode_mdio.h in fwnode_mdio.c (fixed the build warnings)
- added fwnode_find_mii_timestamper() and
fwnode_mdiobus_phy_device_register() in order to get rid of the cycle
dependency.
- change to 'depends on (ACPI || OF) || COMPILE_TEST (for FWNODE_MDIO)
- remove the fwnode_mdiobus_register from fwnode_mdio.c since it
introduces a cycle of dependencies.
Changes in v7:
- correct fwnode_mdio_find_device() description
- check NULL in unregister_mii_timestamper()
- Call unregister_mii_timestamper() without NULL check
- Create fwnode_mdio.c and move fwnode_mdiobus_register_phy()
- include fwnode_mdio.h
- Include headers directly used in acpi_mdio.c
- Move fwnode_mdiobus_register() to fwnode_mdio.c
- Include fwnode_mdio.h
- Alphabetically sort header inclusions
- remove unnecassary checks
Changes in v6:
- Minor cleanup
- fix warning for function parameter of fwnode_mdio_find_device()
- Initialize mii_ts to NULL
- use GENMASK() and ACPI_COMPANION_SET()
- some cleanup
- remove unwanted header inclusion
- remove OF check for fixed-link
- use dev_fwnode()
- remove useless else
- replace of_device_is_available() to fwnode_device_is_available()
Changes in v5:
- More cleanup
- Replace fwnode_get_id() with acpi_get_local_address()
- add missing MODULE_LICENSE()
- replace fwnode_get_id() with OF and ACPI function calls
- replace fwnode_get_id() with OF and ACPI function calls
Changes in v4:
- More cleanup
- Improve code structure to handle all cases
- Remove redundant else from fwnode_mdiobus_register()
- Cleanup xgmac_mdio_probe()
- call phy_device_free() before returning
Changes in v3:
- Add more info on legacy DT properties "phy" and "phy-device"
- Redefine fwnode_phy_find_device() to follow of_phy_find_device()
- Use traditional comparison pattern
- Use GENMASK
- Modified to retrieve reg property value for ACPI as well
- Resolved compilation issue with CONFIG_ACPI = n
- Added more info into documentation
- Use acpi_mdiobus_register()
- Avoid unnecessary line removal
- Remove unused inclusion of acpi.h
Changes in v2:
- Updated with more description in document
- use reverse christmas tree ordering for local variables
- Refactor OF functions to use fwnode functions
Calvin Johnson (15):
Documentation: ACPI: DSD: Document MDIO PHY
net: phy: Introduce fwnode_mdio_find_device()
net: phy: Introduce phy related fwnode functions
of: mdio: Refactor of_phy_find_device()
net: phy: Introduce fwnode_get_phy_id()
of: mdio: Refactor of_get_phy_id()
net: mii_timestamper: check NULL in unregister_mii_timestamper()
net: mdiobus: Introduce fwnode_mdiobus_register_phy()
of: mdio: Refactor of_mdiobus_register_phy()
ACPI: utils: Introduce acpi_get_local_address()
net: mdio: Add ACPI support code for mdio
net/fsl: Use [acpi|of]_mdiobus_register
net: phylink: introduce phylink_fwnode_phy_connect()
net: phylink: Refactor phylink_of_phy_connect()
net: dpaa2-mac: Add ACPI support for DPAA2 MAC driver
Documentation/firmware-guide/acpi/dsd/phy.rst | 133 ++++++++++++++++
MAINTAINERS | 2 +
drivers/acpi/utils.c | 14 ++
.../net/ethernet/freescale/dpaa2/dpaa2-mac.c | 88 ++++++-----
.../net/ethernet/freescale/dpaa2/dpaa2-mac.h | 2 +-
drivers/net/ethernet/freescale/xgmac_mdio.c | 30 ++--
drivers/net/mdio/Kconfig | 14 ++
drivers/net/mdio/Makefile | 4 +-
drivers/net/mdio/acpi_mdio.c | 58 +++++++
drivers/net/mdio/fwnode_mdio.c | 144 ++++++++++++++++++
drivers/net/mdio/of_mdio.c | 138 ++---------------
drivers/net/phy/mii_timestamper.c | 3 +
drivers/net/phy/phy_device.c | 109 ++++++++++++-
drivers/net/phy/phylink.c | 41 +++--
include/linux/acpi.h | 7 +
include/linux/acpi_mdio.h | 26 ++++
include/linux/fwnode_mdio.h | 35 +++++
include/linux/phy.h | 32 ++++
include/linux/phylink.h | 3 +
19 files changed, 693 insertions(+), 190 deletions(-)
create mode 100644 Documentation/firmware-guide/acpi/dsd/phy.rst
create mode 100644 drivers/net/mdio/acpi_mdio.c
create mode 100644 drivers/net/mdio/fwnode_mdio.c
create mode 100644 include/linux/acpi_mdio.h
create mode 100644 include/linux/fwnode_mdio.h
--
2.31.1
From: Calvin Johnson <redacted>
Extract phy_id from compatible string. This will be used by
fwnode_mdiobus_register_phy() to create phy device using the
phy_id.
Signed-off-by: Calvin Johnson <redacted>
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Acked-by: Grant Likely <redacted>
---
Changes in v9: None
Changes in v8: None
Changes in v7: None
Changes in v6: None
Changes in v5: None
Changes in v4: None
Changes in v3:
- Use traditional comparison pattern
- Use GENMASK
drivers/net/phy/phy_device.c | 21 +++++++++++++++++++++
include/linux/phy.h | 5 +++++
2 files changed, 26 insertions(+)
@@ -834,6 +834,27 @@ static int get_phy_c22_id(struct mii_bus *bus, int addr, u32 *phy_id)return0;}+/* Extract the phy ID from the compatible string of the form+*ethernet-phy-idAAAA.BBBB.+*/+intfwnode_get_phy_id(structfwnode_handle*fwnode,u32*phy_id)+{+unsignedintupper,lower;+constchar*cp;+intret;++ret=fwnode_property_read_string(fwnode,"compatible",&cp);+if(ret)+returnret;++if(sscanf(cp,"ethernet-phy-id%4x.%4x",&upper,&lower)!=2)+return-EINVAL;++*phy_id=((upper&GENMASK(15,0))<<16)|(lower&GENMASK(15,0));+return0;+}+EXPORT_SYMBOL(fwnode_get_phy_id);+/***get_phy_device-readsthespecifiedPHYdeviceandreturnsits@phy_device*struct
From: Calvin Johnson <redacted>
With the introduction of fwnode_get_phy_id(), refactor of_get_phy_id()
to use fwnode equivalent.
Signed-off-by: Calvin Johnson <redacted>
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Acked-by: Grant Likely <redacted>
---
Changes in v9: None
Changes in v8: None
Changes in v7: None
Changes in v6: None
Changes in v5: None
Changes in v4: None
Changes in v3: None
Changes in v2: None
drivers/net/mdio/of_mdio.c | 12 +-----------
1 file changed, 1 insertion(+), 11 deletions(-)
From: Calvin Johnson <redacted>
Introduce a wrapper around the _ADR evaluation.
Signed-off-by: Calvin Johnson <redacted>
Reviewed-by: Andy Shevchenko <redacted>
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Acked-by: Rafael J. Wysocki <rafael@kernel.org>
Acked-by: Grant Likely <redacted>
---
Changes in v9: None
Changes in v8: None
Changes in v7: None
Changes in v6: None
Changes in v5:
- Replace fwnode_get_id() with acpi_get_local_address()
Changes in v4:
- Improve code structure to handle all cases
Changes in v3:
- Modified to retrieve reg property value for ACPI as well
- Resolved compilation issue with CONFIG_ACPI = n
- Added more info into documentation
drivers/acpi/utils.c | 14 ++++++++++++++
include/linux/acpi.h | 7 +++++++
2 files changed, 21 insertions(+)
From: Calvin Johnson <redacted>
Callers of unregister_mii_timestamper() currently check for NULL
value of mii_ts before calling it.
Place the NULL check inside unregister_mii_timestamper() and update
the callers accordingly.
Signed-off-by: Calvin Johnson <redacted>
Reviewed-by: Andy Shevchenko <redacted>
Suggested-by: Andy Shevchenko <redacted>
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Acked-by: Grant Likely <redacted>
---
Changes in v9: None
Changes in v8: None
Changes in v7:
- check NULL in unregister_mii_timestamper()
Changes in v6: None
Changes in v5: None
Changes in v4: None
Changes in v3: None
Changes in v2: None
drivers/net/mdio/of_mdio.c | 6 ++----
drivers/net/phy/mii_timestamper.c | 3 +++
drivers/net/phy/phy_device.c | 3 +--
3 files changed, 6 insertions(+), 6 deletions(-)
From: Calvin Johnson <redacted>
Refactor of_mdiobus_register_phy() to use fwnode_mdiobus_register_phy().
Also, remove the of_find_mii_timestamper() since the fwnode variant is
used instead.
Signed-off-by: Calvin Johnson <redacted>
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Acked-by: Grant Likely <redacted>
---
Changes in v9:
- remove the of_find_mii_timestamper() in this patch rather the previous
one
Changes in v8: None
Changes in v7:
- include fwnode_mdio.h
Changes in v6: None
Changes in v5: None
Changes in v4: None
Changes in v3: None
Changes in v2: None
drivers/net/mdio/of_mdio.c | 56 +-------------------------------------
1 file changed, 1 insertion(+), 55 deletions(-)
@@ -63,43 +45,7 @@ EXPORT_SYMBOL(of_mdiobus_phy_device_register);staticintof_mdiobus_register_phy(structmii_bus*mdio,structdevice_node*child,u32addr){-structmii_timestamper*mii_ts;-structphy_device*phy;-boolis_c45;-intrc;-u32phy_id;--mii_ts=of_find_mii_timestamper(child);-if(IS_ERR(mii_ts))-returnPTR_ERR(mii_ts);--is_c45=of_device_is_compatible(child,-"ethernet-phy-ieee802.3-c45");--if(!is_c45&&!of_get_phy_id(child,&phy_id))-phy=phy_device_create(mdio,addr,phy_id,0,NULL);-else-phy=get_phy_device(mdio,addr,is_c45);-if(IS_ERR(phy)){-unregister_mii_timestamper(mii_ts);-returnPTR_ERR(phy);-}--rc=of_mdiobus_phy_device_register(mdio,phy,child,addr);-if(rc){-unregister_mii_timestamper(mii_ts);-phy_device_free(phy);-returnrc;-}--/* phy->mii_ts may already be defined by the PHY driver. A-*mii_timestamperprobedviathedevicetreewillstillhave-*precedence.-*/-if(mii_ts)-phy->mii_ts=mii_ts;--return0;+returnfwnode_mdiobus_register_phy(mdio,of_fwnode_handle(child),addr);}staticintof_mdiobus_register_device(structmii_bus*mdio,
From: Calvin Johnson <redacted>
Define acpi_mdiobus_register() to Register mii_bus and create PHYs for
each ACPI child node.
Signed-off-by: Calvin Johnson <redacted>
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Acked-by: Rafael J. Wysocki <rafael@kernel.org>
Acked-by: Grant Likely <redacted>
---
Changes in v9:
- Add some more info about what is expected to be passed to
acpi_mdiobus_register() as fwnode
Changes in v8: None
Changes in v7:
- Include headers directly used in acpi_mdio.c
Changes in v6:
- use GENMASK() and ACPI_COMPANION_SET()
- some cleanup
- remove unwanted header inclusion
Changes in v5:
- add missing MODULE_LICENSE()
- replace fwnode_get_id() with OF and ACPI function calls
Changes in v4: None
Changes in v3: None
Changes in v2: None
MAINTAINERS | 1 +
drivers/net/mdio/Kconfig | 7 +++++
drivers/net/mdio/Makefile | 1 +
drivers/net/mdio/acpi_mdio.c | 58 ++++++++++++++++++++++++++++++++++++
include/linux/acpi_mdio.h | 26 ++++++++++++++++
5 files changed, 93 insertions(+)
create mode 100644 drivers/net/mdio/acpi_mdio.c
create mode 100644 include/linux/acpi_mdio.h
@@ -1,6 +1,7 @@# SPDX-License-Identifier: GPL-2.0# Makefile for Linux MDIO bus drivers+obj-$(CONFIG_ACPI_MDIO)+=acpi_mdio.oobj-$(CONFIG_FWNODE_MDIO)+=fwnode_mdio.oobj-$(CONFIG_OF_MDIO)+=of_mdio.o
@@ -0,0 +1,58 @@+// SPDX-License-Identifier: GPL-2.0-only+/*+*ACPIhelpersfortheMDIO(EthernetPHY)API+*+*ThisfileprovideshelperfunctionsforextractingPHYdeviceinformation+*outoftheACPIASLandusingittopopulateanmii_bus.+*/++#include<linux/acpi.h>+#include<linux/acpi_mdio.h>+#include<linux/bits.h>+#include<linux/dev_printk.h>+#include<linux/fwnode_mdio.h>+#include<linux/module.h>+#include<linux/types.h>++MODULE_AUTHOR("Calvin Johnson <calvin.johnson@oss.nxp.com>");+MODULE_LICENSE("GPL");++/**+*acpi_mdiobus_register-Registermii_busandcreatePHYsfromtheACPIASL.+*@mdio:pointertomii_busstructure+*@fwnode:pointertofwnodeofMDIObus.Thisfwnodeisexpectedtorepresent+*anACPIdeviceobjectcorrespondingtotheMDIObusanditschildrenare+*expectedtocorrespondtothePHYdevicesonthatbus.+*+*Thisfunctionregistersthemii_busstructureandregistersaphy_device+*foreachchildnodeof@fwnode.+*/+intacpi_mdiobus_register(structmii_bus*mdio,structfwnode_handle*fwnode)+{+structfwnode_handle*child;+u32addr;+intret;++/* Mask out all PHYs from auto probing. */+mdio->phy_mask=GENMASK(31,0);+ret=mdiobus_register(mdio);+if(ret)+returnret;++ACPI_COMPANION_SET(&mdio->dev,to_acpi_device_node(fwnode));++/* Loop over the child nodes and register a phy_device for each PHY */+fwnode_for_each_child_node(fwnode,child){+ret=acpi_get_local_address(ACPI_HANDLE_FWNODE(child),&addr);+if(ret||addr>=PHY_MAX_ADDR)+continue;++ret=fwnode_mdiobus_register_phy(mdio,child,addr);+if(ret==-ENODEV)+dev_err(&mdio->dev,+"MDIO device at address %d is missing.\n",+addr);+}+return0;+}+EXPORT_SYMBOL(acpi_mdiobus_register);
From: Calvin Johnson <redacted>
Introduce a mechanism based on generic ACPI _DSD device properties
definition [1] to get PHYs registered on a MDIO bus and provide them to
be connected to MAC.
[1] http://www.uefi.org/sites/default/files/resources/_DSD-device-properties-UUID.pdf
Describe properties "phy-handle" and "phy-mode".
Signed-off-by: Calvin Johnson <redacted>
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Acked-by: Rafael J. Wysocki <rafael@kernel.org>
Acked-by: Grant Likely <redacted>
---
Changes in v9:
- Reworded the commit message
Changes in v8: None
Changes in v7: None
Changes in v6:
- Minor cleanup
Changes in v5:
- More cleanup
Changes in v4:
- More cleanup
Changes in v3: None
Changes in v2:
- Updated with more description in document
Documentation/firmware-guide/acpi/dsd/phy.rst | 133 ++++++++++++++++++
1 file changed, 133 insertions(+)
create mode 100644 Documentation/firmware-guide/acpi/dsd/phy.rst
@@ -0,0 +1,133 @@+.. SPDX-License-Identifier: GPL-2.0++=========================+MDIO bus and PHYs in ACPI+=========================++The PHYs on an MDIO bus [1] are probed and registered using+fwnode_mdiobus_register_phy().++Later, for connecting these PHYs to their respective MACs, the PHYs registered+on the MDIO bus have to be referenced.++This document introduces two _DSD properties that are to be used+for connecting PHYs on the MDIO bus [3] to the MAC layer.++These properties are defined in accordance with the "Device+Properties UUID For _DSD" [2] document and the+daffd814-6eba-4d8c-8a91-bc9bbf4aa301 UUID must be used in the Device+Data Descriptors containing them.++phy-handle+----------+For each MAC node, a device property "phy-handle" is used to reference+the PHY that is registered on an MDIO bus. This is mandatory for+network interfaces that have PHYs connected to MAC via MDIO bus.++During the MDIO bus driver initialization, PHYs on this bus are probed+using the _ADR object as shown below and are registered on the MDIO bus.++::+ Scope(\_SB.MDI0)+ {+ Device(PHY1) {+ Name (_ADR, 0x1)+ } // end of PHY1++ Device(PHY2) {+ Name (_ADR, 0x2)+ } // end of PHY2+ }++Later, during the MAC driver initialization, the registered PHY devices+have to be retrieved from the MDIO bus. For this, the MAC driver needs+references to the previously registered PHYs which are provided+as device object references (e.g. \_SB.MDI0.PHY1).++phy-mode+--------+The "phy-mode" _DSD property is used to describe the connection to+the PHY. The valid values for "phy-mode" are defined in [4].++The following ASL example illustrates the usage of these properties.++DSDT entry for MDIO node+------------------------++The MDIO bus has an SoC component (MDIO controller) and a platform+component (PHYs on the MDIO bus).++a) Silicon Component+This node describes the MDIO controller, MDI0+---------------------------------------------+::+ Scope(_SB)+ {+ Device(MDI0) {+ Name(_HID, "NXP0006")+ Name(_CCA, 1)+ Name(_UID, 0)+ Name(_CRS, ResourceTemplate() {+ Memory32Fixed(ReadWrite, MDI0_BASE, MDI_LEN)+ Interrupt(ResourceConsumer, Level, ActiveHigh, Shared)+ {+ MDI0_IT+ }+ }) // end of _CRS for MDI0+ } // end of MDI0+ }++b) Platform Component+The PHY1 and PHY2 nodes represent the PHYs connected to MDIO bus MDI0+---------------------------------------------------------------------+::+ Scope(\_SB.MDI0)+ {+ Device(PHY1) {+ Name (_ADR, 0x1)+ } // end of PHY1++ Device(PHY2) {+ Name (_ADR, 0x2)+ } // end of PHY2+ }++DSDT entries representing MAC nodes+-----------------------------------++Below are the MAC nodes where PHY nodes are referenced.+phy-mode and phy-handle are used as explained earlier.+------------------------------------------------------+::+ Scope(\_SB.MCE0.PR17)+ {+ Name (_DSD, Package () {+ ToUUID("daffd814-6eba-4d8c-8a91-bc9bbf4aa301"),+ Package () {+ Package (2) {"phy-mode", "rgmii-id"},+ Package (2) {"phy-handle", \_SB.MDI0.PHY1}+ }+ })+ }++ Scope(\_SB.MCE0.PR18)+ {+ Name (_DSD, Package () {+ ToUUID("daffd814-6eba-4d8c-8a91-bc9bbf4aa301"),+ Package () {+ Package (2) {"phy-mode", "rgmii-id"},+ Package (2) {"phy-handle", \_SB.MDI0.PHY2}}+ }+ })+ }++References+==========++[1] Documentation/networking/phy.rst++[2] https://www.uefi.org/sites/default/files/resources/_DSD-device-properties-UUID.pdf++[3] Documentation/firmware-guide/acpi/DSD-properties-rules.rst++[4] Documentation/devicetree/bindings/net/ethernet-controller.yaml
From: Calvin Johnson <redacted>
Define fwnode_mdio_find_device() to get a pointer to the
mdio_device from fwnode passed to the function.
Refactor of_mdio_find_device() to use fwnode_mdio_find_device().
Signed-off-by: Calvin Johnson <redacted>
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Acked-by: Grant Likely <redacted>
---
Changes in v9: None
Changes in v8: None
Changes in v7:
- correct fwnode_mdio_find_device() description
Changes in v6:
- fix warning for function parameter of fwnode_mdio_find_device()
Changes in v5: None
Changes in v4: None
Changes in v3: None
Changes in v2: None
drivers/net/mdio/of_mdio.c | 11 +----------
drivers/net/phy/phy_device.c | 23 +++++++++++++++++++++++
include/linux/phy.h | 7 +++++++
3 files changed, 31 insertions(+), 10 deletions(-)
From: Calvin Johnson <redacted>
Modify dpaa2_mac_get_node() to get the dpmac fwnode from either
DT or ACPI.
Modify dpaa2_mac_get_if_mode() to get interface mode from dpmac_node
which is a fwnode.
Modify dpaa2_pcs_create() to create pcs from dpmac_node fwnode.
Modify dpaa2_mac_connect() to support ACPI along with DT.
Signed-off-by: Calvin Johnson <redacted>
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Acked-by: Rafael J. Wysocki <rafael@kernel.org> # from the ACPI side
Acked-by: Grant Likely <redacted>
---
Changes in v9: None
Changes in v8:
- adjust code over latest changes applied on the driver
Changes in v7:
- remove unnecassary checks
Changes in v6:
- use dev_fwnode()
- remove useless else
- replace of_device_is_available() to fwnode_device_is_available()
Changes in v5:
- replace fwnode_get_id() with OF and ACPI function calls
Changes in v4: None
Changes in v3: None
Changes in v2:
- Refactor OF functions to use fwnode functions
.../net/ethernet/freescale/dpaa2/dpaa2-mac.c | 88 +++++++++++--------
.../net/ethernet/freescale/dpaa2/dpaa2-mac.h | 2 +-
2 files changed, 53 insertions(+), 37 deletions(-)
@@ -34,39 +37,51 @@ static int phy_mode(enum dpmac_eth_if eth_if, phy_interface_t *if_mode)return0;}-/* Caller must call of_node_put on the returned value */-staticstructdevice_node*dpaa2_mac_get_node(u16dpmac_id)+staticstructfwnode_handle*dpaa2_mac_get_node(structdevice*dev,+u16dpmac_id){-structdevice_node*dpmacs,*dpmac=NULL;-u32id;+structfwnode_handle*fwnode,*parent,*child=NULL;+structdevice_node*dpmacs=NULL;interr;+u32id;-dpmacs=of_find_node_by_name(NULL,"dpmacs");-if(!dpmacs)-returnNULL;+fwnode=dev_fwnode(dev->parent);+if(is_of_node(fwnode)){+dpmacs=of_find_node_by_name(NULL,"dpmacs");+if(!dpmacs)+returnNULL;+parent=of_fwnode_handle(dpmacs);+}elseif(is_acpi_node(fwnode)){+parent=fwnode;+}-while((dpmac=of_get_next_child(dpmacs,dpmac))!=NULL){-err=of_property_read_u32(dpmac,"reg",&id);+fwnode_for_each_child_node(parent,child){+err=-EINVAL;+if(is_acpi_device_node(child))+err=acpi_get_local_address(ACPI_HANDLE_FWNODE(child),&id);+elseif(is_of_node(child))+err=of_property_read_u32(to_of_node(child),"reg",&id);if(err)continue;-if(id==dpmac_id)-break;-}+if(id==dpmac_id){+of_node_put(dpmacs);+returnchild;+}+}of_node_put(dpmacs);--returndpmac;+returnNULL;}-staticintdpaa2_mac_get_if_mode(structdevice_node*node,+staticintdpaa2_mac_get_if_mode(structfwnode_handle*dpmac_node,structdpmac_attrattr){phy_interface_tif_mode;interr;-err=of_get_phy_mode(node,&if_mode);-if(!err)-returnif_mode;+err=fwnode_get_phy_mode(dpmac_node);+if(err>0)+returnerr;err=phy_mode(attr.eth_if,&if_mode);if(!err)
@@ -235,26 +250,27 @@ static const struct phylink_mac_ops dpaa2_mac_phylink_ops = {};staticintdpaa2_pcs_create(structdpaa2_mac*mac,-structdevice_node*dpmac_node,intid)+structfwnode_handle*dpmac_node,+intid){structmdio_device*mdiodev;-structdevice_node*node;+structfwnode_handle*node;-node=of_parse_phandle(dpmac_node,"pcs-handle",0);-if(!node){+node=fwnode_find_reference(dpmac_node,"pcs-handle",0);+if(IS_ERR(node)){/* do not error out on old DTS files */netdev_warn(mac->net_dev,"pcs-handle node not found\n");return0;}-if(!of_device_is_available(node)){+if(!fwnode_device_is_available(node)){netdev_err(mac->net_dev,"pcs-handle node not available\n");-of_node_put(node);+fwnode_handle_put(node);return-ENODEV;}-mdiodev=of_mdio_find_device(node);-of_node_put(node);+mdiodev=fwnode_mdio_find_device(node);+fwnode_handle_put(node);if(!mdiodev)return-EPROBE_DEFER;
@@ -384,8 +400,8 @@ int dpaa2_mac_open(struct dpaa2_mac *mac)/* Find the device node representing the MAC device and link the device*behindtheassociatednetdevtoit.*/-mac->of_node=dpaa2_mac_get_node(mac->attr.id);-net_dev->dev.of_node=mac->of_node;+mac->fw_node=dpaa2_mac_get_node(&mac->mc_dev->dev,mac->attr.id);+net_dev->dev.of_node=to_of_node(mac->fw_node);return0;
From: Calvin Johnson <redacted>
Refactor of_phy_find_device() to use fwnode_phy_find_device().
Signed-off-by: Calvin Johnson <redacted>
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Acked-by: Grant Likely <redacted>
---
Changes in v9: None
Changes in v8: None
Changes in v7: None
Changes in v6: None
Changes in v5: None
Changes in v4: None
Changes in v3: None
Changes in v2: None
drivers/net/mdio/of_mdio.c | 13 +------------
1 file changed, 1 insertion(+), 12 deletions(-)
From: Calvin Johnson <redacted>
Define fwnode_phy_find_device() to iterate an mdiobus and find the
phy device of the provided phy fwnode. Additionally define
device_phy_find_device() to find phy device of provided device.
Define fwnode_get_phy_node() to get phy_node using named reference.
Signed-off-by: Calvin Johnson <redacted>
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Acked-by: Grant Likely <redacted>
---
Changes in v9: None
Changes in v8: None
Changes in v7: None
Changes in v6: None
Changes in v5: None
Changes in v4: None
Changes in v3:
- Add more info on legacy DT properties "phy" and "phy-device"
- Redefine fwnode_phy_find_device() to follow of_phy_find_device()
Changes in v2:
- use reverse christmas tree ordering for local variables
drivers/net/phy/phy_device.c | 62 ++++++++++++++++++++++++++++++++++++
include/linux/phy.h | 20 ++++++++++++
2 files changed, 82 insertions(+)
@@ -2898,6 +2899,67 @@ struct mdio_device *fwnode_mdio_find_device(struct fwnode_handle *fwnode)}EXPORT_SYMBOL(fwnode_mdio_find_device);+/**+*fwnode_phy_find_device-Forprovidedphy_fwnode,findphy_device.+*+*@phy_fwnode:Pointertothephy'sfwnode.+*+*Ifsuccessful,returnsapointertothephy_devicewiththeembedded+*structdevicerefcountincrementedbyone,orNULLonfailure.+*/+structphy_device*fwnode_phy_find_device(structfwnode_handle*phy_fwnode)+{+structmdio_device*mdiodev;++mdiodev=fwnode_mdio_find_device(phy_fwnode);+if(!mdiodev)+returnNULL;++if(mdiodev->flags&MDIO_DEVICE_FLAG_PHY)+returnto_phy_device(&mdiodev->dev);++put_device(&mdiodev->dev);++returnNULL;+}+EXPORT_SYMBOL(fwnode_phy_find_device);++/**+*device_phy_find_device-Forthegivendevice,getthephy_device+*@dev:Pointertothegivendevice+*+*Referreturnconditionsoffwnode_phy_find_device().+*/+structphy_device*device_phy_find_device(structdevice*dev)+{+returnfwnode_phy_find_device(dev_fwnode(dev));+}+EXPORT_SYMBOL_GPL(device_phy_find_device);++/**+*fwnode_get_phy_node-Getthephy_nodeusingthenamedreference.+*@fwnode:Pointertofwnodefromwhichphy_nodehastobeobtained.+*+*Referreturnconditionsoffwnode_find_reference().+*ForACPI,only"phy-handle"issupported.LegacyDTproperties"phy"+*and"phy-device"arenotsupportedinACPI.DTsupportsallthethree+*namedreferencestothephynode.+*/+structfwnode_handle*fwnode_get_phy_node(structfwnode_handle*fwnode)+{+structfwnode_handle*phy_node;++/* Only phy-handle is used for ACPI */+phy_node=fwnode_find_reference(fwnode,"phy-handle",0);+if(is_acpi_node(fwnode)||!IS_ERR(phy_node))+returnphy_node;+phy_node=fwnode_find_reference(fwnode,"phy",0);+if(IS_ERR(phy_node))+phy_node=fwnode_find_reference(fwnode,"phy-device",0);+returnphy_node;+}+EXPORT_SYMBOL_GPL(fwnode_get_phy_node);+/***phy_probe-probeandinitaPHYdevice*@dev:devicetoprobeandinit
From: Calvin Johnson <redacted>
Introduce fwnode_mdiobus_register_phy() to register PHYs on the
mdiobus. From the compatible string, identify whether the PHY is
c45 and based on this create a PHY device instance which is
registered on the mdiobus.
Along with fwnode_mdiobus_register_phy() also introduce
fwnode_find_mii_timestamper() and fwnode_mdiobus_phy_device_register()
since they are needed.
While at it, also use the newly introduced fwnode operation in
of_mdiobus_phy_device_register().
Signed-off-by: Calvin Johnson <redacted>
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Acked-by: Grant Likely <redacted>
---
Changes in v9:
- fixed the build issue by not removing of_find_mii_timestamper() in
this patch but rather in the next one
Changes in v8:
- fixed some checkpatch warnings/checks
- included linux/fwnode_mdio.h in fwnode_mdio.c (fixed the build warnings)
- added fwnode_find_mii_timestamper() and fwnode_mdiobus_phy_device_register()
in order to get rid of the cycle dependency.
- change to 'depends on (ACPI || OF) || COMPILE_TEST
Changes in v7:
- Call unregister_mii_timestamper() without NULL check
- Create fwnode_mdio.c and move fwnode_mdiobus_register_phy()
Changes in v6:
- Initialize mii_ts to NULL
Changes in v5: None
Changes in v4: None
Changes in v3: None
Changes in v2: None
MAINTAINERS | 1 +
drivers/net/mdio/Kconfig | 7 ++
drivers/net/mdio/Makefile | 3 +-
drivers/net/mdio/fwnode_mdio.c | 144 +++++++++++++++++++++++++++++++++
drivers/net/mdio/of_mdio.c | 44 ++--------
include/linux/fwnode_mdio.h | 35 ++++++++
6 files changed, 194 insertions(+), 40 deletions(-)
create mode 100644 drivers/net/mdio/fwnode_mdio.c
create mode 100644 include/linux/fwnode_mdio.h
@@ -1,7 +1,8 @@# SPDX-License-Identifier: GPL-2.0# Makefile for Linux MDIO bus drivers-obj-$(CONFIG_OF_MDIO)+=of_mdio.o+obj-$(CONFIG_FWNODE_MDIO)+=fwnode_mdio.o+obj-$(CONFIG_OF_MDIO)+=of_mdio.oobj-$(CONFIG_MDIO_ASPEED)+=mdio-aspeed.oobj-$(CONFIG_MDIO_BCM_IPROC)+=mdio-bcm-iproc.o
@@ -0,0 +1,144 @@+// SPDX-License-Identifier: GPL-2.0-only+/*+*fwnodehelpersfortheMDIO(EthernetPHY)API+*+*ThisfileprovideshelperfunctionsforextractingPHYdeviceinformation+*outofthefwnodeandusingittopopulateanmii_bus.+*/++#include<linux/acpi.h>+#include<linux/fwnode_mdio.h>+#include<linux/of.h>+#include<linux/phy.h>++MODULE_AUTHOR("Calvin Johnson <calvin.johnson@oss.nxp.com>");+MODULE_LICENSE("GPL");++staticstructmii_timestamper*+fwnode_find_mii_timestamper(structfwnode_handle*fwnode)+{+structof_phandle_argsarg;+interr;++if(is_acpi_node(fwnode))+returnNULL;++err=of_parse_phandle_with_fixed_args(to_of_node(fwnode),+"timestamper",1,0,&arg);+if(err==-ENOENT)+returnNULL;+elseif(err)+returnERR_PTR(err);++if(arg.args_count!=1)+returnERR_PTR(-EINVAL);++returnregister_mii_timestamper(arg.np,arg.args[0]);+}++intfwnode_mdiobus_phy_device_register(structmii_bus*mdio,+structphy_device*phy,+structfwnode_handle*child,u32addr)+{+intrc;++rc=fwnode_irq_get(child,0);+if(rc==-EPROBE_DEFER)+returnrc;++if(rc>0){+phy->irq=rc;+mdio->irq[addr]=rc;+}else{+phy->irq=mdio->irq[addr];+}++if(fwnode_property_read_bool(child,"broken-turn-around"))+mdio->phy_ignore_ta_mask|=1<<addr;++fwnode_property_read_u32(child,"reset-assert-us",+&phy->mdio.reset_assert_delay);+fwnode_property_read_u32(child,"reset-deassert-us",+&phy->mdio.reset_deassert_delay);++/* Associate the fwnode with the device structure so it+*canbelookeduplater+*/+fwnode_handle_get(child);+phy->mdio.dev.fwnode=child;++/* All data is now stored in the phy struct;+*registerit+*/+rc=phy_device_register(phy);+if(rc){+fwnode_handle_put(child);+returnrc;+}++dev_dbg(&mdio->dev,"registered phy %p fwnode at address %i\n",+child,addr);+return0;+}+EXPORT_SYMBOL(fwnode_mdiobus_phy_device_register);++intfwnode_mdiobus_register_phy(structmii_bus*bus,+structfwnode_handle*child,u32addr)+{+structmii_timestamper*mii_ts=NULL;+structphy_device*phy;+boolis_c45=false;+u32phy_id;+intrc;++mii_ts=fwnode_find_mii_timestamper(child);+if(IS_ERR(mii_ts))+returnPTR_ERR(mii_ts);++rc=fwnode_property_match_string(child,"compatible",+"ethernet-phy-ieee802.3-c45");+if(rc>=0)+is_c45=true;++if(is_c45||fwnode_get_phy_id(child,&phy_id))+phy=get_phy_device(bus,addr,is_c45);+else+phy=phy_device_create(bus,addr,phy_id,0,NULL);+if(IS_ERR(phy)){+unregister_mii_timestamper(mii_ts);+returnPTR_ERR(phy);+}++if(is_acpi_node(child)){+phy->irq=bus->irq[addr];++/* Associate the fwnode with the device structure so it+*canbelookeduplater.+*/+phy->mdio.dev.fwnode=child;++/* All data is now stored in the phy struct, so register it */+rc=phy_device_register(phy);+if(rc){+phy_device_free(phy);+fwnode_handle_put(phy->mdio.dev.fwnode);+returnrc;+}+}elseif(is_of_node(child)){+rc=fwnode_mdiobus_phy_device_register(bus,phy,child,addr);+if(rc){+unregister_mii_timestamper(mii_ts);+phy_device_free(phy);+returnrc;+}+}++/* phy->mii_ts may already be defined by the PHY driver. A+*mii_timestamperprobedviathedevicetreewillstillhave+*precedence.+*/+if(mii_ts)+phy->mii_ts=mii_ts;+return0;+}+EXPORT_SYMBOL(fwnode_mdiobus_register_phy);
@@ -51,46 +52,11 @@ static struct mii_timestamper *of_find_mii_timestamper(struct device_node *node)}intof_mdiobus_phy_device_register(structmii_bus*mdio,structphy_device*phy,-structdevice_node*child,u32addr)+structdevice_node*child,u32addr){-intrc;--rc=of_irq_get(child,0);-if(rc==-EPROBE_DEFER)-returnrc;--if(rc>0){-phy->irq=rc;-mdio->irq[addr]=rc;-}else{-phy->irq=mdio->irq[addr];-}--if(of_property_read_bool(child,"broken-turn-around"))-mdio->phy_ignore_ta_mask|=1<<addr;--of_property_read_u32(child,"reset-assert-us",-&phy->mdio.reset_assert_delay);-of_property_read_u32(child,"reset-deassert-us",-&phy->mdio.reset_deassert_delay);--/* Associate the OF node with the device structure so it-*canbelookeduplater*/-of_node_get(child);-phy->mdio.dev.of_node=child;-phy->mdio.dev.fwnode=of_fwnode_handle(child);--/* All data is now stored in the phy struct;-*registerit*/-rc=phy_device_register(phy);-if(rc){-of_node_put(child);-returnrc;-}--dev_dbg(&mdio->dev,"registered phy %pOFn at address %i\n",-child,addr);-return0;+returnfwnode_mdiobus_phy_device_register(mdio,phy,+of_fwnode_handle(child),+addr);}EXPORT_SYMBOL(of_mdiobus_phy_device_register);
From: Calvin Johnson <redacted>
Define phylink_fwnode_phy_connect() to connect phy specified by
a fwnode to a phylink instance.
Signed-off-by: Calvin Johnson <redacted>
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Acked-by: Grant Likely <redacted>
---
Changes in v9: None
Changes in v8: None
Changes in v7: None
Changes in v6:
- remove OF check for fixed-link
Changes in v5: None
Changes in v4:
- call phy_device_free() before returning
Changes in v3: None
Changes in v2: None
drivers/net/phy/phylink.c | 54 +++++++++++++++++++++++++++++++++++++++
include/linux/phylink.h | 3 +++
2 files changed, 57 insertions(+)
@@ -1125,6 +1126,59 @@ int phylink_of_phy_connect(struct phylink *pl, struct device_node *dn,}EXPORT_SYMBOL_GPL(phylink_of_phy_connect);+/**+*phylink_fwnode_phy_connect()-connectthePHYspecifiedinthefwnode.+*@pl:apointertoa&structphylinkreturnedfromphylink_create()+*@fwnode:apointertoa&structfwnode_handle.+*@flags:PHY-specificflagstocommunicatetothePHYdevicedriver+*+*Connectthephyspecified@fwnodetothephylinkinstancespecified+*by@pl.+*+*Returns0onsuccessoranegativeerrno.+*/+intphylink_fwnode_phy_connect(structphylink*pl,+structfwnode_handle*fwnode,+u32flags)+{+structfwnode_handle*phy_fwnode;+structphy_device*phy_dev;+intret;++/* Fixed links and 802.3z are handled without needing a PHY */+if(pl->cfg_link_an_mode==MLO_AN_FIXED||+(pl->cfg_link_an_mode==MLO_AN_INBAND&&+phy_interface_mode_is_8023z(pl->link_interface)))+return0;++phy_fwnode=fwnode_get_phy_node(fwnode);+if(IS_ERR(phy_fwnode)){+if(pl->cfg_link_an_mode==MLO_AN_PHY)+return-ENODEV;+return0;+}++phy_dev=fwnode_phy_find_device(phy_fwnode);+/* We're done with the phy_node handle */+fwnode_handle_put(phy_fwnode);+if(!phy_dev)+return-ENODEV;++ret=phy_attach_direct(pl->netdev,phy_dev,flags,+pl->link_interface);+if(ret){+phy_device_free(phy_dev);+returnret;+}++ret=phylink_bringup_phy(pl,phy_dev,pl->link_config.interface);+if(ret)+phy_detach(phy_dev);++returnret;+}+EXPORT_SYMBOL_GPL(phylink_fwnode_phy_connect);+/***phylink_disconnect_phy()-disconnectanyPHYattachedtothephylink*instance.
From: Calvin Johnson <redacted>
Depending on the device node type, call the specific OF or ACPI
mdiobus_register function.
Note: For both ACPI and DT cases, endianness of MDIO controllers
need to be specified using the "little-endian" property.
Signed-off-by: Calvin Johnson <redacted>
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Acked-by: Grant Likely <redacted>
---
Changes in v9: none
Changes in v8:
- Directly call the OF or ACPI variants of registering the MDIO bus.
This is needed because the fwnode_mdio.c module should only implement
features which can be achieved without going back to the OF/ACPI
variants. Without this restrictions we directly end up in a dependency
cycle: of_mdio -> fwnode_mdio -> of_mdio.
- Changed the commit title since the fwnode_mdiobus_register() is no
longer available
Changes in v7:
- Include fwnode_mdio.h
- Alphabetically sort header inclusions
Changes in v6: None
Changes in v5: None
Changes in v4:
- Cleanup xgmac_mdio_probe()
Changes in v3:
- Avoid unnecessary line removal
- Remove unused inclusion of acpi.h
Changes in v2: None
drivers/net/ethernet/freescale/xgmac_mdio.c | 30 ++++++++++++++-------
1 file changed, 21 insertions(+), 9 deletions(-)
@@ -11,15 +12,17 @@*kind,whetherexpressorimplied.*/-#include<linux/kernel.h>-#include<linux/slab.h>+#include<linux/acpi.h>+#include<linux/acpi_mdio.h>#include<linux/interrupt.h>-#include<linux/module.h>-#include<linux/phy.h>+#include<linux/kernel.h>#include<linux/mdio.h>+#include<linux/module.h>#include<linux/of_address.h>-#include<linux/of_platform.h>#include<linux/of_mdio.h>+#include<linux/of_platform.h>+#include<linux/phy.h>+#include<linux/slab.h>/* Number of microseconds to wait for a register to respond */#define TIMEOUT 1000
@@ -243,10 +246,10 @@ static int xgmac_mdio_read(struct mii_bus *bus, int phy_id, int regnum)staticintxgmac_mdio_probe(structplatform_device*pdev){-structdevice_node*np=pdev->dev.of_node;-structmii_bus*bus;-structresource*res;+structfwnode_handle*fwnode;structmdio_fsl_priv*priv;+structresource*res;+structmii_bus*bus;intret;/* In DPAA-1, MDIO is one of the many FMan sub-devices. The FMan
@@ -279,13 +282,22 @@ static int xgmac_mdio_probe(struct platform_device *pdev)gotoerr_ioremap;}+/* For both ACPI and DT cases, endianness of MDIO controller+*needstobespecifiedusing"little-endian"property.+*/priv->is_little_endian=device_property_read_bool(&pdev->dev,"little-endian");priv->has_a011043=device_property_read_bool(&pdev->dev,"fsl,erratum-a011043");-ret=of_mdiobus_register(bus,np);+fwnode=pdev->dev.fwnode;+if(is_of_node(fwnode))+ret=of_mdiobus_register(bus,to_of_node(fwnode));+elseif(is_acpi_node(fwnode))+ret=acpi_mdiobus_register(bus,fwnode);+else+ret=-EINVAL;if(ret){dev_err(&pdev->dev,"cannot register MDIO bus\n");gotoerr_registration;
From: Calvin Johnson <redacted>
Refactor phylink_of_phy_connect() to use phylink_fwnode_phy_connect().
Signed-off-by: Calvin Johnson <redacted>
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Acked-by: Grant Likely <redacted>
---
Changes in v9: None
Changes in v8: None
Changes in v7: None
Changes in v6: None
Changes in v5: None
Changes in v4: None
Changes in v3: None
Changes in v2: None
drivers/net/phy/phylink.c | 39 +--------------------------------------
1 file changed, 1 insertion(+), 38 deletions(-)
@@ -1085,44 +1085,7 @@ EXPORT_SYMBOL_GPL(phylink_connect_phy);intphylink_of_phy_connect(structphylink*pl,structdevice_node*dn,u32flags){-structdevice_node*phy_node;-structphy_device*phy_dev;-intret;--/* Fixed links and 802.3z are handled without needing a PHY */-if(pl->cfg_link_an_mode==MLO_AN_FIXED||-(pl->cfg_link_an_mode==MLO_AN_INBAND&&-phy_interface_mode_is_8023z(pl->link_interface)))-return0;--phy_node=of_parse_phandle(dn,"phy-handle",0);-if(!phy_node)-phy_node=of_parse_phandle(dn,"phy",0);-if(!phy_node)-phy_node=of_parse_phandle(dn,"phy-device",0);--if(!phy_node){-if(pl->cfg_link_an_mode==MLO_AN_PHY)-return-ENODEV;-return0;-}--phy_dev=of_phy_find_device(phy_node);-/* We're done with the phy_node handle */-of_node_put(phy_node);-if(!phy_dev)-return-ENODEV;--ret=phy_attach_direct(pl->netdev,phy_dev,flags,-pl->link_interface);-if(ret)-returnret;--ret=phylink_bringup_phy(pl,phy_dev,pl->link_config.interface);-if(ret)-phy_detach(phy_dev);--returnret;+returnphylink_fwnode_phy_connect(pl,of_fwnode_handle(dn),flags);}EXPORT_SYMBOL_GPL(phylink_of_phy_connect);
From: "Russell King (Oracle)" <linux@armlinux.org.uk> Date: 2021-06-11 11:18:10
On Fri, Jun 11, 2021 at 01:53:59PM +0300, Ioana Ciornei wrote:
From: Calvin Johnson <redacted>
Define phylink_fwnode_phy_connect() to connect phy specified by
a fwnode to a phylink instance.
Signed-off-by: Calvin Johnson <redacted>
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Acked-by: Grant Likely <redacted>
Reviewed-by: Russell King (Oracle) <redacted>
Thanks.
quoted hunk
---
Changes in v9: None
Changes in v8: None
Changes in v7: None
Changes in v6:
- remove OF check for fixed-link
Changes in v5: None
Changes in v4:
- call phy_device_free() before returning
Changes in v3: None
Changes in v2: None
drivers/net/phy/phylink.c | 54 +++++++++++++++++++++++++++++++++++++++
include/linux/phylink.h | 3 +++
2 files changed, 57 insertions(+)
@@ -1125,6 +1126,59 @@ int phylink_of_phy_connect(struct phylink *pl, struct device_node *dn,}EXPORT_SYMBOL_GPL(phylink_of_phy_connect);+/**+*phylink_fwnode_phy_connect()-connectthePHYspecifiedinthefwnode.+*@pl:apointertoa&structphylinkreturnedfromphylink_create()+*@fwnode:apointertoa&structfwnode_handle.+*@flags:PHY-specificflagstocommunicatetothePHYdevicedriver+*+*Connectthephyspecified@fwnodetothephylinkinstancespecified+*by@pl.+*+*Returns0onsuccessoranegativeerrno.+*/+intphylink_fwnode_phy_connect(structphylink*pl,+structfwnode_handle*fwnode,+u32flags)+{+structfwnode_handle*phy_fwnode;+structphy_device*phy_dev;+intret;++/* Fixed links and 802.3z are handled without needing a PHY */+if(pl->cfg_link_an_mode==MLO_AN_FIXED||+(pl->cfg_link_an_mode==MLO_AN_INBAND&&+phy_interface_mode_is_8023z(pl->link_interface)))+return0;++phy_fwnode=fwnode_get_phy_node(fwnode);+if(IS_ERR(phy_fwnode)){+if(pl->cfg_link_an_mode==MLO_AN_PHY)+return-ENODEV;+return0;+}++phy_dev=fwnode_phy_find_device(phy_fwnode);+/* We're done with the phy_node handle */+fwnode_handle_put(phy_fwnode);+if(!phy_dev)+return-ENODEV;++ret=phy_attach_direct(pl->netdev,phy_dev,flags,+pl->link_interface);+if(ret){+phy_device_free(phy_dev);+returnret;+}++ret=phylink_bringup_phy(pl,phy_dev,pl->link_config.interface);+if(ret)+phy_detach(phy_dev);++returnret;+}+EXPORT_SYMBOL_GPL(phylink_fwnode_phy_connect);+/***phylink_disconnect_phy()-disconnectanyPHYattachedtothephylink*instance.
From: "Russell King (Oracle)" <linux@armlinux.org.uk> Date: 2021-06-11 11:18:33
On Fri, Jun 11, 2021 at 01:54:00PM +0300, Ioana Ciornei wrote:
From: Calvin Johnson <redacted>
Refactor phylink_of_phy_connect() to use phylink_fwnode_phy_connect().
Signed-off-by: Calvin Johnson <redacted>
Signed-off-by: Ioana Ciornei <ioana.ciornei@nxp.com>
Acked-by: Grant Likely <redacted>
Reviewed-by: Russell King (Oracle) <redacted>
Thanks.
--
RMK's Patch system: https://www.armlinux.org.uk/developer/patches/
FTTP is here! 40Mbps down 10Mbps up. Decent connectivity at last!
From: Andy Shevchenko <hidden> Date: 2021-06-11 11:22:27
On Fri, Jun 11, 2021 at 1:54 PM Ioana Ciornei [off-list ref] wrote:
From: Calvin Johnson <redacted>
Define fwnode_mdio_find_device() to get a pointer to the
mdio_device from fwnode passed to the function.
Refactor of_mdio_find_device() to use fwnode_mdio_find_device().
That's what I meant. Is this series converting two users to use
fwnode_mdio_find_device()? If not, it should do this and kill
of_mdio_find_device() completely.
--
With Best Regards,
Andy Shevchenko
From: Andy Shevchenko <hidden> Date: 2021-06-11 11:27:46
On Fri, Jun 11, 2021 at 1:54 PM Ioana Ciornei [off-list ref] wrote:
From: Calvin Johnson <redacted>
Define fwnode_phy_find_device() to iterate an mdiobus and find the
phy device of the provided phy fwnode. Additionally define
device_phy_find_device() to find phy device of provided device.
Define fwnode_get_phy_node() to get phy_node using named reference.
using a named
...
+struct fwnode_handle *fwnode_get_phy_node(struct fwnode_handle *fwnode)
+{
+ struct fwnode_handle *phy_node;
+
+ /* Only phy-handle is used for ACPI */
+ phy_node = fwnode_find_reference(fwnode, "phy-handle", 0);
+ if (is_acpi_node(fwnode) || !IS_ERR(phy_node))
+ return phy_node;
+ phy_node = fwnode_find_reference(fwnode, "phy", 0);
+ if (IS_ERR(phy_node))
+ phy_node = fwnode_find_reference(fwnode, "phy-device", 0);
+ return phy_node;
Looking into the patterns in this code I would perhaps refactor it the
following way:
/* First try "phy-handle" as most common in use */
phy_node = fwnode_find_reference(fwnode, "phy-handle", 0);
/* Only phy-handle is used for ACPI */
if (is_acpi_node(fwnode))
return phy_node;
if (!IS_ERR(phy_node))
return phy_node;
/* Try "phy" reference */
phy_node = fwnode_find_reference(fwnode, "phy", 0);
if (!IS_ERR(phy_node))
return phy_node;
/* At last try "phy-device" reference */
return fwnode_find_reference(fwnode, "phy-device", 0);
From: Andy Shevchenko <hidden> Date: 2021-06-11 11:30:52
On Fri, Jun 11, 2021 at 2:28 PM Andy Shevchenko
[off-list ref] wrote:
On Fri, Jun 11, 2021 at 1:54 PM Ioana Ciornei [off-list ref] wrote:
quoted
From: Calvin Johnson <redacted>
Refactor of_phy_find_device() to use fwnode_phy_find_device().
I see that there are many users of this, but I think eventually we
should kill of_phy_find_device() completely.
Looking into other examples of such I think this series may not touch
them right now, but clearly state that it's the plan in the future to
kill this kind of OF APIs that call fwnode underneath.
--
With Best Regards,
Andy Shevchenko
From: "Russell King (Oracle)" <linux@armlinux.org.uk> Date: 2021-06-11 11:35:56
On Fri, Jun 11, 2021 at 02:30:19PM +0300, Andy Shevchenko wrote:
On Fri, Jun 11, 2021 at 2:28 PM Andy Shevchenko
[off-list ref] wrote:
quoted
On Fri, Jun 11, 2021 at 1:54 PM Ioana Ciornei [off-list ref] wrote:
quoted
From: Calvin Johnson <redacted>
Refactor of_phy_find_device() to use fwnode_phy_find_device().
I see that there are many users of this, but I think eventually we
should kill of_phy_find_device() completely.
Looking into other examples of such I think this series may not touch
them right now, but clearly state that it's the plan in the future to
kill this kind of OF APIs that call fwnode underneath.
That's something I most definitely support - once we have the fwnode
APIs in place, the OF specific APIs become an additional maintenance
burden that we don't need. So, I would also like to see the old APIs
killed.
--
RMK's Patch system: https://www.armlinux.org.uk/developer/patches/
FTTP is here! 40Mbps down 10Mbps up. Decent connectivity at last!
From: "Rafael J. Wysocki" <rafael@kernel.org> Date: 2021-06-11 11:41:28
On Fri, Jun 11, 2021 at 1:26 PM Andy Shevchenko
[off-list ref] wrote:
On Fri, Jun 11, 2021 at 1:54 PM Ioana Ciornei [off-list ref] wrote:
quoted
From: Calvin Johnson <redacted>
Define fwnode_phy_find_device() to iterate an mdiobus and find the
phy device of the provided phy fwnode. Additionally define
device_phy_find_device() to find phy device of provided device.
Define fwnode_get_phy_node() to get phy_node using named reference.
using a named
...
quoted
+struct fwnode_handle *fwnode_get_phy_node(struct fwnode_handle *fwnode)
+{
+ struct fwnode_handle *phy_node;
+
+ /* Only phy-handle is used for ACPI */
+ phy_node = fwnode_find_reference(fwnode, "phy-handle", 0);
+ if (is_acpi_node(fwnode) || !IS_ERR(phy_node))
+ return phy_node;
+ phy_node = fwnode_find_reference(fwnode, "phy", 0);
+ if (IS_ERR(phy_node))
+ phy_node = fwnode_find_reference(fwnode, "phy-device", 0);
+ return phy_node;
Looking into the patterns in this code I would perhaps refactor it the
following way:
/* First try "phy-handle" as most common in use */
phy_node = fwnode_find_reference(fwnode, "phy-handle", 0);
/* Only phy-handle is used for ACPI */
if (is_acpi_node(fwnode))
return phy_node;
if (!IS_ERR(phy_node))
return phy_node;
I'm not sure why you want the above to be two if () statements instead of one?
I would change the ordering anyway, that is
if (!IS_ERR(phy_node) || is_acpi_node(fwnode))
return phy_node;
And I think that the is_acpi_node() check is there to return the error
code right away so as to avoid returning a "not found" error later.
But I'm not sure if this is really necessary. Namely, if nothing
depends on the specific error code returned by this function, it would
be somewhat cleaner to let the code below run if phy_node is an error
pointer in the ACPI case, because in that case the code below will
produce an error pointer anyway.
/* Try "phy" reference */
phy_node = fwnode_find_reference(fwnode, "phy", 0);
if (!IS_ERR(phy_node))
return phy_node;
/* At last try "phy-device" reference */
return fwnode_find_reference(fwnode, "phy-device", 0);
From: "Russell King (Oracle)" <linux@armlinux.org.uk> Date: 2021-06-11 12:09:01
On Fri, Jun 11, 2021 at 01:40:59PM +0200, Rafael J. Wysocki wrote:
I'm not sure why you want the above to be two if () statements instead of one?
I would change the ordering anyway, that is
if (!IS_ERR(phy_node) || is_acpi_node(fwnode))
return phy_node;
And I think that the is_acpi_node() check is there to return the error
code right away so as to avoid returning a "not found" error later.
But I'm not sure if this is really necessary. Namely, if nothing
depends on the specific error code returned by this function, it would
be somewhat cleaner to let the code below run if phy_node is an error
pointer in the ACPI case, because in that case the code below will
produce an error pointer anyway.
However, that opens the door to someone shipping "working" ACPI with
one of these names that we've taken the decision not to support on
ACPI firmware. Surely, it's much better that we don't accept the
legacy names so we don't allow such configurations to work.
--
RMK's Patch system: https://www.armlinux.org.uk/developer/patches/
FTTP is here! 40Mbps down 10Mbps up. Decent connectivity at last!
From: "Rafael J. Wysocki" <rafael@kernel.org> Date: 2021-06-11 12:32:05
On Fri, Jun 11, 2021 at 2:08 PM Russell King (Oracle)
[off-list ref] wrote:
On Fri, Jun 11, 2021 at 01:40:59PM +0200, Rafael J. Wysocki wrote:
quoted
I'm not sure why you want the above to be two if () statements instead of one?
I would change the ordering anyway, that is
if (!IS_ERR(phy_node) || is_acpi_node(fwnode))
return phy_node;
And I think that the is_acpi_node() check is there to return the error
code right away so as to avoid returning a "not found" error later.
But I'm not sure if this is really necessary. Namely, if nothing
depends on the specific error code returned by this function, it would
be somewhat cleaner to let the code below run if phy_node is an error
pointer in the ACPI case, because in that case the code below will
produce an error pointer anyway.
However, that opens the door to someone shipping "working" ACPI with
one of these names that we've taken the decision not to support on
ACPI firmware. Surely, it's much better that we don't accept the
legacy names so we don't allow such configurations to work.
Hello:
This series was applied to netdev/net-next.git (refs/heads/master):
On Fri, 11 Jun 2021 13:53:46 +0300 you wrote:
From: Ioana Ciornei <ioana.ciornei@nxp.com>
This patch set provides ACPI support to DPAA2 network drivers.
It also introduces new fwnode based APIs to support phylink and phy
layers
Following functions are defined:
phylink_fwnode_phy_connect()
fwnode_mdiobus_register_phy()
fwnode_get_phy_id()
fwnode_phy_find_device()
device_phy_find_device()
fwnode_get_phy_node()
fwnode_mdio_find_device()
acpi_get_local_address()
[...]
Fri, Jun 11, 2021 at 01:53:54PM +0300, Ioana Ciornei kirjoitti:
From: Calvin Johnson <redacted>
Introduce fwnode_mdiobus_register_phy() to register PHYs on the
mdiobus. From the compatible string, identify whether the PHY is
c45 and based on this create a PHY device instance which is
registered on the mdiobus.
Along with fwnode_mdiobus_register_phy() also introduce
fwnode_find_mii_timestamper() and fwnode_mdiobus_phy_device_register()
since they are needed.
While at it, also use the newly introduced fwnode operation in
of_mdiobus_phy_device_register().
I understand that this patch series is applied, but I think we have a problem
that has to be fixed before the release. See below for the details.
...
+int fwnode_mdiobus_phy_device_register(struct mii_bus *mdio,
+ struct phy_device *phy,
+ struct fwnode_handle *child, u32 addr)
+{
+ int rc;
+
+ rc = fwnode_irq_get(child, 0);
+ if (rc == -EPROBE_DEFER)
+ return rc;
+
+ if (rc > 0) {
+ phy->irq = rc;
+ mdio->irq[addr] = rc;
+ } else {
+ phy->irq = mdio->irq[addr];
+ }
+
+ if (fwnode_property_read_bool(child, "broken-turn-around"))
+ mdio->phy_ignore_ta_mask |= 1 << addr;
+
+ fwnode_property_read_u32(child, "reset-assert-us",
+ &phy->mdio.reset_assert_delay);
+ fwnode_property_read_u32(child, "reset-deassert-us",
+ &phy->mdio.reset_deassert_delay);
+
+ /* Associate the fwnode with the device structure so it
+ * can be looked up later
+ */
+ fwnode_handle_get(child);
We take a reference counting here.
Who and when should drop it?
+ phy->mdio.dev.fwnode = child;
+
+ /* All data is now stored in the phy struct;
+ * register it
+ */
+ rc = phy_device_register(phy);
+ if (rc) {
+ fwnode_handle_put(child);
+ return rc;
+ }
+
+ dev_dbg(&mdio->dev, "registered phy %p fwnode at address %i\n",
+ child, addr);
+ return 0;
+}
+EXPORT_SYMBOL(fwnode_mdiobus_phy_device_register);
+
+int fwnode_mdiobus_register_phy(struct mii_bus *bus,
+ struct fwnode_handle *child, u32 addr)
+{
+ struct mii_timestamper *mii_ts = NULL;
+ struct phy_device *phy;
+ bool is_c45 = false;
+ u32 phy_id;
+ int rc;
+
+ mii_ts = fwnode_find_mii_timestamper(child);
+ if (IS_ERR(mii_ts))
+ return PTR_ERR(mii_ts);
+
+ rc = fwnode_property_match_string(child, "compatible",
+ "ethernet-phy-ieee802.3-c45");
+ if (rc >= 0)
+ is_c45 = true;
+
+ if (is_c45 || fwnode_get_phy_id(child, &phy_id))
+ phy = get_phy_device(bus, addr, is_c45);
+ else
+ phy = phy_device_create(bus, addr, phy_id, 0, NULL);
+ if (IS_ERR(phy)) {
+ unregister_mii_timestamper(mii_ts);
+ return PTR_ERR(phy);
+ }
+
+ if (is_acpi_node(child)) {
+ phy->irq = bus->irq[addr];
+
+ /* Associate the fwnode with the device structure so it
+ * can be looked up later.
+ */
+ phy->mdio.dev.fwnode = child;
+
+ /* All data is now stored in the phy struct, so register it */
+ rc = phy_device_register(phy);
+ if (rc) {
+ phy_device_free(phy);
+ fwnode_handle_put(phy->mdio.dev.fwnode);
We dropped reference counting here, but who and when acquired it?
+ return rc;
+ }
+ } else if (is_of_node(child)) {
+ rc = fwnode_mdiobus_phy_device_register(bus, phy, child, addr);
+ if (rc) {
+ unregister_mii_timestamper(mii_ts);
+ phy_device_free(phy);
+ return rc;
+ }
+ }
+
+ /* phy->mii_ts may already be defined by the PHY driver. A
+ * mii_timestamper probed via the device tree will still have
+ * precedence.
+ */
+ if (mii_ts)
+ phy->mii_ts = mii_ts;
+ return 0;
+}
Fri, Jun 11, 2021 at 01:53:59PM +0300, Ioana Ciornei kirjoitti:
From: Calvin Johnson <redacted>
Define phylink_fwnode_phy_connect() to connect phy specified by
a fwnode to a phylink instance.
...
+/**
+ * phylink_fwnode_phy_connect() - connect the PHY specified in the fwnode.
+ * @pl: a pointer to a &struct phylink returned from phylink_create()
+ * @fwnode: a pointer to a &struct fwnode_handle.
+ * @flags: PHY-specific flags to communicate to the PHY device driver
+ *
+ * Connect the phy specified @fwnode to the phylink instance specified
+ * by @pl.
+ *
+ * Returns 0 on success or a negative errno.
+ */
+int phylink_fwnode_phy_connect(struct phylink *pl,
+ struct fwnode_handle *fwnode,
+ u32 flags)
+{
+ struct fwnode_handle *phy_fwnode;
+ struct phy_device *phy_dev;
+ int ret;
+
+ /* Fixed links and 802.3z are handled without needing a PHY */
+ if (pl->cfg_link_an_mode == MLO_AN_FIXED ||
+ (pl->cfg_link_an_mode == MLO_AN_INBAND &&
+ phy_interface_mode_is_8023z(pl->link_interface)))
+ return 0;
+
+ phy_fwnode = fwnode_get_phy_node(fwnode);
+ if (IS_ERR(phy_fwnode)) {
+ if (pl->cfg_link_an_mode == MLO_AN_PHY)
+ return -ENODEV;
+ return 0;
+ }
+
+ phy_dev = fwnode_phy_find_device(phy_fwnode);
+ /* We're done with the phy_node handle */
+ fwnode_handle_put(phy_fwnode);
As per previous mail, who and when acquire this reference counting? Can it be
possible that caller will never call phylink_fwnode_phy_connect() and hence
reference counting will become disbalances?
We need to fix all this before the release.
+ if (!phy_dev)
+ return -ENODEV;
+
+ ret = phy_attach_direct(pl->netdev, phy_dev, flags,
+ pl->link_interface);
+ if (ret) {
+ phy_device_free(phy_dev);
+ return ret;
+ }
+
+ ret = phylink_bringup_phy(pl, phy_dev, pl->link_config.interface);
+ if (ret)
+ phy_detach(phy_dev);
+
+ return ret;
+}