netdev - Re: [PATCH net-next] net: phy: realtek: add suspend/resume callbacks for RTL8211B

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <fbaf55ac-4404-c3e6-6e5d-d3a8122ca0d2@gmail.com>
Date:   Thu, 24 May 2018 23:42:07 +0200
From:   Heiner Kallweit <hkallweit1@...il.com>
To:     Andrew Lunn <andrew@...n.ch>
Cc:     David Miller <davem@...emloft.net>,
        Realtek linux nic maintainers <nic_swsd@...ltek.com>,
        Hau <hau@...ltek.com>, Florian Fainelli <f.fainelli@...il.com>,
        "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
        Kevin Hao <haokexin@...il.com>
Subject: Re: [PATCH net-next] net: phy: realtek: add suspend/resume callbacks
 for RTL8211B

Am 24.05.2018 um 22:53 schrieb Andrew Lunn:
> On Thu, May 24, 2018 at 10:40:12PM +0200, Heiner Kallweit wrote:
>> Add RTL8211B suspend / resume callbacks.
>>
>> Signed-off-by: Heiner Kallweit <hkallweit1@...il.com>
>> ---
>> This patch is based on my knowledge of the r8169 driver, and on some
>> guessing. Therefore I'd appreciate a confirmation from Realtek.
>>
>> The integrated PHY in some chips supported by the r8169 driver uses
>> a special sequence for power-down/-up. I have a board with a RTL8168D
>> network chip (one of the chips using the special sequence) and there
>> the PHY identifies as RTL8211B. So my guess is that this applies also
>> to external RTL8211B PHY's.
>>
>> A hint for RTL8211B requiring a special sequence is that no suspend/
>> resume callbacks are defined yet in the Realtek PHY driver.
>> Last but not least the non-standard usage of register MII_MMD_DATA
>> is in line with the description of patch 0231b1a074c6.
>> ("net: phy: realtek: Use the dummy stubs for MMD register access for rtl8211b")
>> ---
>>  drivers/net/phy/realtek.c | 16 ++++++++++++++++
>>  1 file changed, 16 insertions(+)
>>
>> diff --git a/drivers/net/phy/realtek.c b/drivers/net/phy/realtek.c
>> index 9f48ecf9c..082fb40c6 100644
>> --- a/drivers/net/phy/realtek.c
>> +++ b/drivers/net/phy/realtek.c
>> @@ -145,6 +145,20 @@ static int rtl8211f_config_init(struct phy_device *phydev)
>>  	return phy_modify_paged(phydev, 0xd08, 0x11, RTL8211F_TX_DELAY, val);
>>  }
>>  
>> +static int rtl8211b_suspend(struct phy_device *phydev)
>> +{
>> +	phy_write(phydev, MII_MMD_DATA, BIT(9));
> 
> Hi Heiner
> 
> Using it like this suggests it is not actually MMD_DATA, it is
> something else which just happens to use the same address as the
> optional MMD_DATA. To make this clearer, it would be good to add
> #defines for both the register address and this BIT(9). Is there any
> vendor code you know of which might give you a clue for appropriate
> names?
> 
Vendor code (Realtek r8168 driver) just writes value 0x0200 to
register 0x0E. I also would have preferred to assign proper
names to register and bit. Maybe the Realtek people on cc can
provide some information.

> I guess this device also does not support EEE? Does phy_init_eee()
> correctly figure this out? Is there a chance calling phy_init_eee()
> might trigger a suspend?
> 
Right, EEE isn't supported. phy_init_eee() figures this out correctly
because read_mmd callback is set to new genphy_read_mmd_unsupported.

Heiner

>        Andrew
>