[LEDE-DEV] [PATCH 1/2] generic: ar8216: fix ar8xxx_is_possible check
John Crispin
john at phrozen.org
Mon Oct 3 12:12:32 PDT 2016
On 01/10/2016 18:33, Christian Lamparter wrote:
> The commit "generic: ar8216: add sanity check to ar8216_probe"
> (774da6c7a40320a320b28d71291c0e61fcf7bc8a) stated that PHY IDs
> should be checked at address 0-4. However, the PHY 4 was
> never check by the for loop... And I can't find any documents
> about why this check should be performed the way it is
> currently?!
Hi
the SDK driver checks for the phy ids if i am not mistaken. doing the
cech does make sense to ensure that the switch is actually present. why
do you want to remove this check ? imho it does no harm
John
>
> Signed-off-by: Christian Lamparter <chunkeey at gmail.com>
> ---
> .../linux/generic/files/drivers/net/phy/ar8216.c | 23 ----------------------
> 1 file changed, 23 deletions(-)
>
> diff --git a/target/linux/generic/files/drivers/net/phy/ar8216.c b/target/linux/generic/files/drivers/net/phy/ar8216.c
> index 70f4774..57130b1 100644
> --- a/target/linux/generic/files/drivers/net/phy/ar8216.c
> +++ b/target/linux/generic/files/drivers/net/phy/ar8216.c
> @@ -2110,26 +2110,6 @@ ar8xxx_phy_match(u32 phy_id)
> return false;
> }
>
> -static bool
> -ar8xxx_is_possible(struct mii_bus *bus)
> -{
> - unsigned i;
> -
> - for (i = 0; i < 4; i++) {
> - u32 phy_id;
> -
> - phy_id = mdiobus_read(bus, i, MII_PHYSID1) << 16;
> - phy_id |= mdiobus_read(bus, i, MII_PHYSID2);
> - if (!ar8xxx_phy_match(phy_id)) {
> - pr_debug("ar8xxx: unknown PHY at %s:%02x id:%08x\n",
> - dev_name(&bus->dev), i, phy_id);
> - return false;
> - }
> - }
> -
> - return true;
> -}
> -
> static int
> ar8xxx_phy_probe(struct phy_device *phydev)
> {
> @@ -2141,9 +2121,6 @@ ar8xxx_phy_probe(struct phy_device *phydev)
> if (phydev->addr != 0 && phydev->addr != 4)
> return -ENODEV;
>
> - if (!ar8xxx_is_possible(phydev->bus))
> - return -ENODEV;
> -
> mutex_lock(&ar8xxx_dev_list_lock);
> list_for_each_entry(priv, &ar8xxx_dev_list, list)
> if (priv->mii_bus == phydev->bus)
>
More information about the Lede-dev
mailing list