netdev - Re: [EXT] [PATCH net-next] net: aquantia: fix error handling in aq_nic

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] [day] [month] [year] [list]

Message-ID: <16c059aa-04a9-1301-fa72-eb475f3f3c6f@embeddedor.com>
Date:   Mon, 28 Oct 2019 07:00:00 -0500
From:   "Gustavo A. R. Silva" <gustavo@...eddedor.com>
To:     Igor Russkikh <irusskikh@...vell.com>,
        Igor Russkikh <Igor.Russkikh@...antia.com>,
        "David S. Miller" <davem@...emloft.net>,
        Dmitry Bezrukov <Dmitry.Bezrukov@...antia.com>,
        Nikita Danilov <Nikita.Danilov@...antia.com>,
        Andrew Lunn <andrew@...n.ch>
Cc:     "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [EXT] [PATCH net-next] net: aquantia: fix error handling in
 aq_nic_init



On 10/28/19 05:53, Igor Russkikh wrote:
> 
>> Fix currenty ignored returned error by properly error checking
>> aq_phy_init().
>>
>> Addresses-Coverity-ID: 1487376 ("Unused value")
>> Fixes: dbcd6806af42 ("net: aquantia: add support for Phy access")
>> Signed-off-by: Gustavo A. R. Silva <gustavo@...eddedor.com>
>> ---
>>  drivers/net/ethernet/aquantia/atlantic/aq_nic.c | 3 ++-
>>  1 file changed, 2 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c
>> index 433adc099e44..1914aa0a19d0 100644
>> --- a/drivers/net/ethernet/aquantia/atlantic/aq_nic.c
>> +++ b/drivers/net/ethernet/aquantia/atlantic/aq_nic.c
>> @@ -341,7 +341,8 @@ int aq_nic_init(struct aq_nic_s *self)
>>  
>>  	if (self->aq_nic_cfg.aq_hw_caps->media_type == AQ_HW_MEDIA_TYPE_TP) {
>>  		self->aq_hw->phy_id = HW_ATL_PHY_ID_MAX;
>> -		err = aq_phy_init(self->aq_hw);
>> +		if (!aq_phy_init(self->aq_hw))
>> +			goto err_exit;
>>  	}
>>  
>>  	for (i = 0U, aq_vec = self->aq_vec[0];
>>
> 
> Hi Gustavo,
> 

Hi!

> I'd say the intention here was to ignore the error, as driver may still live if
> something unexpected happened on Phy access path.
> 

I see. Please, see my comments below...

> Notice in the above fix you leave `err` as zero but do error path return -
> that'll break the datapath.
> 

You're right. I totally missed that.

> I'd prefer to fix this with simple
> 
>      (void)aq_phy_init(self->aq_hw);
> 

Yep. This is much better. Otherwise, both static analyzers and developers
get confused. :p

Something even better is to add both a cast and a comment explaining why
the return value is not being checked in this case (as you do above).

Thanks
--
Gustavo