netdev - Re: [PATCH net] net: phy: smsc: add missed clk_disable_unprepare in smsc_phy

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: <fc03d4de-14d2-1b61-ac9b-40ea26e6fa9a@gmail.com>
Date:   Sat, 14 Nov 2020 11:45:35 -0800
From:   Florian Fainelli <f.fainelli@...il.com>
To:     Jakub Kicinski <kuba@...nel.org>,
        Zhang Changzhong <zhangchangzhong@...wei.com>
Cc:     andrew@...n.ch, hkallweit1@...il.com, linux@...linux.org.uk,
        davem@...emloft.net, m.felsch@...gutronix.de,
        netdev@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH net] net: phy: smsc: add missed clk_disable_unprepare in
 smsc_phy_probe()



On 11/14/2020 11:26 AM, Jakub Kicinski wrote:
> On Thu, 12 Nov 2020 19:23:59 +0800 Zhang Changzhong wrote:
>> Add the missing clk_disable_unprepare() before return from
>> smsc_phy_probe() in the error handling case.
>>
>> Fixes: bedd8d78aba3 ("net: phy: smsc: LAN8710/20: add phy refclk in support")
>> Reported-by: Hulk Robot <hulkci@...wei.com>
>> Signed-off-by: Zhang Changzhong <zhangchangzhong@...wei.com>
>> ---
>>  drivers/net/phy/smsc.c | 4 +++-
>>  1 file changed, 3 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/net/phy/smsc.c b/drivers/net/phy/smsc.c
>> index ec97669..0fc39ac 100644
>> --- a/drivers/net/phy/smsc.c
>> +++ b/drivers/net/phy/smsc.c
>> @@ -291,8 +291,10 @@ static int smsc_phy_probe(struct phy_device *phydev)
>>  		return ret;
>>  
>>  	ret = clk_set_rate(priv->refclk, 50 * 1000 * 1000);
>> -	if (ret)
>> +	if (ret) {
>> +		clk_disable_unprepare(priv->refclk);
>>  		return ret;
>> +	}
>>  
>>  	return 0;
>>  }
> 
> Applied, thanks!
> 
> The code right above looks highly questionable as well:
> 
>         priv->refclk = clk_get_optional(dev, NULL);
>         if (IS_ERR(priv->refclk))
>                 dev_err_probe(dev, PTR_ERR(priv->refclk), "Failed to request clock\n");
>  
>         ret = clk_prepare_enable(priv->refclk);
>         if (ret)
>                 return ret;
> 
> I don't think clk_prepare_enable() will be too happy to see an error
> pointer. This should probably be:
> 
>         priv->refclk = clk_get_optional(dev, NULL);
>         if (IS_ERR(priv->refclk))
>                 return dev_err_probe(dev, PTR_ERR(priv->refclk), 
> 				      "Failed to request clock\n");

Right, especially if EPROBE_DEFER must be returned because the clock
provider is not ready yet, we should have a chance to do that.
-- 
Florian