net: phy: Make phy_interface_is_sgmii|rgmii a switch statement
v2 cover net: phy: Make phy_interface_is_sgmii|rgmii a switch statement
Originally reported by Tom[1], turned out to be that recent commit 75d28899e3e9 ("net: phy: Synchronize PHY interface modes with Linux") reordered the enum definitions which in turn broke the range checks.
we are left with two options:
a) check against explicit values to help reuse as much as possible and let compiler optimize where applicable
or
b) be very explicit in phy drivers and drop these helpers.
I have chosen to go with (a) approach.
Tested on am64x, though the dp83867 is used elsewhere as well.
Changes since v1:
* Change in switch handling (Thanks Marek for the suggestion)
V1: https://lore.kernel.org/all/[email protected]/
Nishanth Menon (2):
net: phy: Make phy_interface_is_sgmii a switch statement
net: phy: Make phy_interface_is_rgmii a switch statement
include/phy.h | 22 ++++++++++++++++++---- 1 file changed, 18 insertions(+), 4 deletions(-)
Signature · 1 line
2.40.0
PATCH v2 1/2 net: phy: Make phy_interface_is_sgmii a switch statement
Recent commit 75d28899e3e9 ("net: phy: Synchronize PHY interface modes with Linux") reordered the enum definitions. This caused the range of enums that this api was checking to go bad.
While it is possible for the phy drivers to practically use the enum's directly, drivers such as dp83867 use this helper to manage the configuration of the phy correctly.
---
Changes Since v1:
* Switch update based on feedback from Marek
V1: https://lore.kernel.org/r/[email protected]
include/phy.h | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-)
include/phy.h +9 -2
@@ -373,8 +373,15 @@ static inline bool phy_interface_is_rgmii(struct phy_device *phydev)*/ static inline bool phy_interface_is_sgmii (struct phy_device *phydev) { -return phydev->interface >=PHY_INTERFACE_MODE_SGMII && - phydev->interface <=PHY_INTERFACE_MODE_QSGMII ; +switch (phydev->interface) { +case PHY_INTERFACE_MODE_SGMII : +case PHY_INTERFACE_MODE_QUSGMII : +case PHY_INTERFACE_MODE_USXGMII : +case PHY_INTERFACE_MODE_QSGMII : +return 1 ; +default : +return 0 ; + } }bool phy_interface_is_ncsi (void );
Signature · 1 line
2.40.0
Re: [PATCH V2 1/2] net: phy: Make phy_interface_is_sgmii a switch statement
Quoted 31 lines · On Thu, 13 Apr 2023 23:24:32 -0500 Nishanth Menon <[email protected]> wrote:
Recent commit
75d28899e3e9("net: phy: Synchronize PHY interface modes with Linux") reordered the enum definitions. This caused the range of enums that this api was checking to go bad.While it is possible for the phy drivers to practically use the enum's directly, drivers such as dp83867 use this helper to manage the configuration of the phy correctly.
Reported-by Tom Rini <[email protected]>Signed-off-by Nishanth Menon <[email protected]>---
Changes Since v1:
* Switch update based on feedback from MarekV1: https://lore.kernel.org/r/[email protected]
include/phy.h | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-)include/phy.h +2
-2@@ -373,8 +373,15 @@ static inline bool phy_interface_is_rgmii(struct phy_device *phydev)*/ static inline bool phy_interface_is_sgmii (struct phy_device *phydev) { -return phydev->interface >=PHY_INTERFACE_MODE_SGMII && - phydev->interface <=PHY_INTERFACE_MODE_QSGMII ; +switch (phydev->interface) { +case PHY_INTERFACE_MODE_SGMII :
include/phy.h static inline bool phy_interface_is_rgmii(struct phy_device *phydev) · line 387
Quoted 10 lines
include/phy.h +7
-0+case PHY_INTERFACE_MODE_QUSGMII : +case PHY_INTERFACE_MODE_USXGMII : +case PHY_INTERFACE_MODE_QSGMII : +return 1 ; +default : +return 0 ; + } }bool phy_interface_is_ncsi (void );
Please also consider dropping this function, since it is used only in one driver
drivers/net/phy/dp83867.c
and that device does not actually support QUSGMII, USXGMII nor QSGMII. It only supports SGMII.
With rgmii, the corresponding phy_interface_is_rgmii() function makes sense, but for SGMII the four modes are entirely too different for this to make sense.
Marek
Re: [PATCH V2 1/2] net: phy: Make phy_interface_is_sgmii a switch statement
In reply to Nishanth MenonQuoted 10 lines · Nishanth Menon
Recent commit
75d28899e3e9("net: phy: Synchronize PHY interface modes with Linux") reordered the enum definitions. This caused the range of enums that this api was checking to go bad.While it is possible for the phy drivers to practically use the enum's directly, drivers such as dp83867 use this helper to manage the configuration of the phy correctly.
Reported-by Tom Rini <[email protected]>Signed-off-by Nishanth Menon <[email protected]>
It would be nice if you could do a follow-up patch as suggested by Marek Behun and inline this function into the TI PHY driver instead.
PATCH v2 2/2 net: phy: Make phy_interface_is_rgmii a switch statement
In reply to Nishanth MenonRecent commit 75d28899e3e9 ("net: phy: Synchronize PHY interface modes with Linux") reordered the enum definitions. This exposed a problem in range checking functions to identify the interface type. Though this specific api was'nt impacted (all the RGMII definitions remained within range), this experience should be used to never to have to face this kind of challenge again.
While it is possible for the phy drivers to practically use the enum's directly, drivers such as dp83867, dp83869, marvell, micrel_ksz90x1 etc use the same.
---
Changes since V1:
* Switch update based on feedback from Marek
V1: https://lore.kernel.org/r/[email protected]
include/phy.h | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-)
include/phy.h +9 -2
@@ -361,8 +361,15 @@ int get_phy_id(struct mii_dev *bus, int addr, int devad, u32 *phy_id);*/ static inline bool phy_interface_is_rgmii (struct phy_device *phydev) { -return phydev->interface >=PHY_INTERFACE_MODE_RGMII && - phydev->interface <=PHY_INTERFACE_MODE_RGMII_TXID ; +switch (phydev->interface) { +case PHY_INTERFACE_MODE_RGMII : +case PHY_INTERFACE_MODE_RGMII_ID : +case PHY_INTERFACE_MODE_RGMII_RXID : +case PHY_INTERFACE_MODE_RGMII_TXID : +return 1 ; +default : +return 0 ; + } }/**
Signature · 1 line
2.40.0
Re: [PATCH V2 2/2] net: phy: Make phy_interface_is_rgmii a switch statement
Quoted 33 lines · On Thu, 13 Apr 2023 23:24:33 -0500 Nishanth Menon <[email protected]> wrote:
Recent commit
75d28899e3e9("net: phy: Synchronize PHY interface modes with Linux") reordered the enum definitions. This exposed a problem in range checking functions to identify the interface type. Though this specific api was'nt impacted (all the RGMII definitions remained within range), this experience should be used to never to have to face this kind of challenge again.While it is possible for the phy drivers to practically use the enum's directly, drivers such as dp83867, dp83869, marvell,
micrel_ksz90x1etc use the same.Reported-by Tom Rini <[email protected]>Signed-off-by Nishanth Menon <[email protected]>---
Changes since V1:
* Switch update based on feedback from MarekV1: https://lore.kernel.org/r/[email protected]
include/phy.h | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-)include/phy.h +2
-2@@ -361,8 +361,15 @@ int get_phy_id(struct mii_dev *bus, int addr, int devad, u32 *phy_id);*/ static inline bool phy_interface_is_rgmii (struct phy_device *phydev) { -return phydev->interface >=PHY_INTERFACE_MODE_RGMII && - phydev->interface <=PHY_INTERFACE_MODE_RGMII_TXID ; +switch (phydev->interface) { +case PHY_INTERFACE_MODE_RGMII :
include/phy.h int get_phy_id(struct mii_dev *bus, int addr, int devad, u32 *phy_id); · line 375
Quoted 10 lines
include/phy.h +7
-0+case PHY_INTERFACE_MODE_RGMII_ID : +case PHY_INTERFACE_MODE_RGMII_RXID : +case PHY_INTERFACE_MODE_RGMII_TXID : +return 1 ; +default : +return 0 ; + } }/**
Re: [PATCH V2 2/2] net: phy: Make phy_interface_is_rgmii a switch statement
In reply to Nishanth MenonQuoted 4 lines · Nishanth Menon
Recent commit
75d28899e3e9("net: phy: Synchronize PHY interface modes with Linux") reordered the enum definitions. This exposed a problem in range checking functions to identify the interface type. Though this specific api was'nt impacted (all the RGMII definitions remained within
Nit: "wasn't" , apostrophe in the wrong place .
Quoted 9 lines
range), this experience should be used to never to have to face this kind of challenge again.
While it is possible for the phy drivers to practically use the enum's directly, drivers such as dp83867, dp83869, marvell,
micrel_ksz90x1etc use the same.Reported-by Tom Rini <[email protected]>Signed-off-by Nishanth Menon <[email protected]>