net: phy: phy_interface_is_rgmii update, drop phy_interface_is_sgmii
Subsystems (2)
v3 cover net: phy: phy_interface_is_rgmii update, drop phy_interface_is_sgmii
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 V2:
* Dropped fixing up sgmii api, instead: just fixup the phy driver, drop the api
* picked reviewed by for the rgmii
V2: https://lore.kernel.org/all/[email protected]/
V1: https://lore.kernel.org/all/[email protected]/
Nishanth Menon (3):
net: phy: dp83867: Explicitly check against sgmii
net: phy: Drop phy_interface_is_sgmii
net: phy: Make phy_interface_is_rgmii a switch statement
drivers/net/phy/dp83867.c | 2 +- include/phy.h | 23 +++++++++-------------- 2 files changed, 10 insertions(+), 15 deletions(-)
Signature · 1 line
2.40.0
PATCH v3 1/3 net: phy: dp83867: Explicitly check against sgmii
dp83867 driver only supports sgmii and not all the newer protocols. Drop the usage of the generic phy_interface_is_sgmii function and just matchup to the specific mode supported.
---
Changes since v2:
* New patch.
drivers/net/phy/dp83867.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/phy/dp83867.c +1 -1
@@ -330,7 +330,7 @@ static int dp83867_config(struct phy_device *phydev) DP83867_RGMIIDCTL, delay); } -if (phy_interface_is_sgmii (phydev)) { +if (phydev->interface ==PHY_INTERFACE_MODE_SGMII ) {if (dp83867->sgmii_ref_clk_en)phy_write_mmd (phydev,DP83867_DEVADDR ,DP83867_SGMIICTL ,DP83867_SGMII_TYPE );
Signature · 1 line
2.40.0
Re: [PATCH V3 1/3] net: phy: dp83867: Explicitly check against sgmii
Quoted 31 lines · Nishanth Menon
dp83867 driver only supports sgmii and not all the newer protocols. Drop the usage of the generic
phy_interface_is_sgmiifunction and just matchup to the specific mode supported.Reported-by Tom Rini <[email protected]>Suggested-by Marek Vasut <[email protected]>Suggested-by Marek Behún <[email protected]>Signed-off-by Nishanth Menon <[email protected]>---
Changes since v2:
* New patch.drivers/net/phy/dp83867.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)drivers/net/phy/dp83867.c +1
-2@@ -330,7 +330,7 @@ static int dp83867_config(struct phy_device *phydev) DP83867_RGMIIDCTL, delay); } -if (phy_interface_is_sgmii (phydev)) { +if (phydev->interface ==PHY_INTERFACE_MODE_SGMII ) {if (dp83867->sgmii_ref_clk_en)phy_write_mmd (phydev,DP83867_DEVADDR ,DP83867_SGMIICTL ,DP83867_SGMII_TYPE ); --
Quoted 2 lines
2.40.0
Re: [PATCH V3 1/3] net: phy: dp83867: Explicitly check against sgmii
In reply to Nishanth MenonQuoted 10 lines · Nishanth Menon
dp83867 driver only supports sgmii and not all the newer protocols. Drop the usage of the generic
phy_interface_is_sgmiifunction and just matchup to the specific mode supported.Reported-by Tom Rini <[email protected]>Suggested-by Marek Vasut <[email protected]>Suggested-by Marek Behún <[email protected]>Signed-off-by Nishanth Menon <[email protected]>Reviewed-by Ramon Fried <[email protected]>
Applied to u-boot/master, thanks!
Signature · 1 line
Tom
PATCH v3 2/3 net: phy: Drop phy_interface_is_sgmii
In reply to Nishanth MenonRecent 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.
There aren't anymore users of phy_interface_is_sgmii, so, just drop it. Also the protocols are so very different that it makes no sense to provide a helper wrapper in the hope of reuse for phy drivers.
---
Changes Since v2:
* Redone the patch, but original versions below (did'nt make sense to carry reviewed-by)
* Just drop the phy_interface_is_sgmii
V2: https://lore.kernel.org/r/[email protected]
V1: https://lore.kernel.org/r/[email protected]
include/phy.h | 12 ------------ 1 file changed, 12 deletions(-)
include/phy.h +0 -12
@@ -365,18 +365,6 @@ static inline bool phy_interface_is_rgmii(struct phy_device *phydev) phydev->interface <=PHY_INTERFACE_MODE_RGMII_TXID ; } -/** -* phy_interface_is_sgmii - Convenience function for testing if a PHY interface -* is SGMII (all variants) -* @phydev: the phy_device struct -* @return: true if MII bus is SGMII or false if it is not -*/ -static inline bool phy_interface_is_sgmii (struct phy_device *phydev) -{ -return phydev->interface >=PHY_INTERFACE_MODE_SGMII && - phydev->interface <=PHY_INTERFACE_MODE_QSGMII ; -} -bool phy_interface_is_ncsi (void );/* PHY UIDs for various PHYs that are referenced in external code */
Signature · 1 line
2.40.0
Re: [PATCH V3 2/3] net: phy: Drop phy_interface_is_sgmii
Quoted 49 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.There aren't anymore users of
phy_interface_is_sgmii, so, just drop it. Also the protocols are so very different that it makes no sense to provide a helper wrapper in the hope of reuse for phy drivers.Reported-by Tom Rini <[email protected]>Suggested-by Marek Vasut <[email protected]>Suggested-by Marek Behún <[email protected]>Signed-off-by Nishanth Menon <[email protected]>---
Changes Since v2:
* Redone the patch, but original versions below (did'nt make sense to carry reviewed-by)
* Just drop thephy_interface_is_sgmii
V2: https://lore.kernel.org/r/[email protected]
V1: https://lore.kernel.org/r/[email protected]include/phy.h | 12 ------------ 1 file changed, 12 deletions(-)include/phy.h +0
-13@@ -365,18 +365,6 @@ static inline bool phy_interface_is_rgmii(struct phy_device *phydev) phydev->interface <=PHY_INTERFACE_MODE_RGMII_TXID ; } -/** -* phy_interface_is_sgmii - Convenience function for testing if a PHY interface -* is SGMII (all variants) -* @phydev: the phy_device struct -* @return: true if MII bus is SGMII or false if it is not -*/ -static inline bool phy_interface_is_sgmii (struct phy_device *phydev) -{ -return phydev->interface >=PHY_INTERFACE_MODE_SGMII && - phydev->interface <=PHY_INTERFACE_MODE_QSGMII ; -} -bool phy_interface_is_ncsi (void );/* PHY UIDs for various PHYs that are referenced in external code */ --
Quoted 2 lines
2.40.0
Re: [PATCH V3 2/3] net: phy: Drop phy_interface_is_sgmii
In reply to Nishanth MenonQuoted 8 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.There aren't anymore users of
phy_interface_is_sgmii, so, just drop it. Also the protocols are so very different that it makes no sense to provide a helper wrapper in the hope of reuse for phy drivers.
Quoted 6 lines
Reported-by Tom Rini <[email protected]>Suggested-by Marek Vasut <[email protected]>Suggested-by Marek Behún <[email protected]>Signed-off-by Nishanth Menon <[email protected]>Reviewed-by Ramon Fried <[email protected]>
Applied to u-boot/master, thanks!
Signature · 1 line
Tom
PATCH v3 3/3 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 wasn't 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 use the enums directly, drivers such as dp83867, dp83869, marvell, micrel_ksz90x1 etc use this api.
---
Changes since V2:
* picked up reviewed-by
* s/was'nt/wasn't/g in commit message, minor rewording.
V2: https://lore.kernel.org/r/[email protected]
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 ; + } }bool phy_interface_is_ncsi (void );
Signature · 1 line
2.40.0
Re: [PATCH V3 3/3] net: phy: Make phy_interface_is_rgmii a switch statement
Quoted 48 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 wasn't 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 use the enums directly, drivers such as dp83867, dp83869, marvell,
micrel_ksz90x1etc use this api.Reported-by Tom Rini <[email protected]>Reviewed-by Marek Vasut <[email protected]>Reviewed-by Marek Behún <[email protected]>Signed-off-by Nishanth Menon <[email protected]>---
Changes since V2:
* picked up reviewed-by
* s/was'nt/wasn't/g in commit message, minor rewording.
V2: https://lore.kernel.org/r/[email protected]
V1: https://lore.kernel.org/r/[email protected]include/phy.h | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-)include/phy.h +9
-3@@ -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 ; + } }bool phy_interface_is_ncsi (void ); --
Quoted 2 lines
2.40.0
Re: [PATCH V3 3/3] net: phy: Make phy_interface_is_rgmii a switch statement
In reply to Nishanth MenonQuoted 11 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 wasn't 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 use the enums directly, drivers such as dp83867, dp83869, marvell,
micrel_ksz90x1etc use this api.
Quoted 5 lines
Reported-by Tom Rini <[email protected]>Reviewed-by Marek Vasut <[email protected]>Reviewed-by Marek Behún <[email protected]>Signed-off-by Nishanth Menon <[email protected]>Reviewed-by Ramon Fried <[email protected]>
Applied to u-boot/master, thanks!
Signature · 1 line
Tom