Re: [RFC PATCH net-next 5/8] net: phy: realtek: use phy_read_paged instead of open coding

[Heiner Kallweit]

On 23.04.2023 20:01, Daniel Golle wrote:
> On Sat, Apr 22, 2023 at 05:11:57PM +0200, Heiner Kallweit wrote:
>> On 22.04.2023 13:48, Daniel Golle wrote:
>>> Instead of open coding a paged read, use the phy_read_paged function
>>> in rtlgen_supports_2_5gbps.
>>>
>>> Signed-off-by: Daniel Golle <daniel@xxxxxxxxxxxxxx>
>>> ---
>>>  drivers/net/phy/realtek.c | 4 +---
>>>  1 file changed, 1 insertion(+), 3 deletions(-)
>>>
>>> diff --git a/drivers/net/phy/realtek.c b/drivers/net/phy/realtek.c
>>> index f97b5e49fae58..62fb965b6d338 100644
>>> --- a/drivers/net/phy/realtek.c
>>> +++ b/drivers/net/phy/realtek.c
>>> @@ -735,9 +735,7 @@ static bool rtlgen_supports_2_5gbps(struct phy_device *phydev)
>>>  {
>>>  	int val;
>>>  
>>> -	phy_write(phydev, RTL821x_PAGE_SELECT, 0xa61);
>>> -	val = phy_read(phydev, 0x13);
>>> -	phy_write(phydev, RTL821x_PAGE_SELECT, 0);
>>> +	val = phy_read_paged(phydev, 0xa61, 0x13);
>>>  
>>>  	return val >= 0 && val & RTL_SUPPORTS_2500FULL;
>>>  }
>>
>> I remember I had a reason to open-code it, it took me some minutes
>> to recall it.
>> phy_read_paged() calls __phy_read_page() that relies on phydev->drv
>> being set. phydev->drv is set in phy_probe(). And probing is done
>> after matching. __phy_read_paged() should have given you a warning.
>> Did you test this patch? If yes and you didn't get the warning,
>> then apparently I miss something.
>>
> 
> Yes, you are right, this change was a bit too naive and causes a
> NULL pointer dereference e.g. for the r8169 driver which also uses
> the RealTek Ethernet PHY driver.
> My main concern and original motivation was the lack of mutex protection
> for the paged read operation. I suggest to rather make this change
> instead:
> 
>>From 4dd2cc9b91ecb25f278a2c55e07e6455e9000e6b Mon Sep 17 00:00:00 2001
> From: Daniel Golle <daniel@xxxxxxxxxxxxxx>
> Date: Sun, 23 Apr 2023 18:47:45 +0100
> Subject: [PATCH] net: phy: realtek: make sure paged read is protected by mutex
> 
> As we cannot rely on phy_read_paged function before the PHY is
> identified, the paged read in rtlgen_supports_2_5gbps needs to be open
> coded as it is being called by the match_phy_device function, ie. before
> .read_page and .write_page have been populated.
> 
> Make sure it is also protected by the MDIO bus mutex and use
> rtl821x_write_page instead of 3 individually locked MDIO bus operations.
> 
> Signed-off-by: Daniel Golle <daniel@xxxxxxxxxxxxxx>
> ---
>  drivers/net/phy/realtek.c | 8 +++++---
>  1 file changed, 5 insertions(+), 3 deletions(-)
> 
> diff --git a/drivers/net/phy/realtek.c b/drivers/net/phy/realtek.c
> index f97b5e49fae5..c27ec4e99fc2 100644
> --- a/drivers/net/phy/realtek.c
> +++ b/drivers/net/phy/realtek.c
> @@ -735,9 +735,11 @@ static bool rtlgen_supports_2_5gbps(struct phy_device *phydev)
>  {
>  	int val;
>  
> -	phy_write(phydev, RTL821x_PAGE_SELECT, 0xa61);
> -	val = phy_read(phydev, 0x13);
> -	phy_write(phydev, RTL821x_PAGE_SELECT, 0);
> +	mutex_lock(&phydev->mdio.bus->mdio_lock);

We have helpers phy_(un)lock_mdio_bus() for this.
Apart from that: LGTM

> +	rtl821x_write_page(phydev, 0xa61);
> +	val = __phy_read(phydev, 0x13);
> +	rtl821x_write_page(phydev, 0);
> +	mutex_unlock(&phydev->mdio.bus->mdio_lock);
>  
>  	return val >= 0 && val & RTL_SUPPORTS_2500FULL;
>  }