netdev - Re: [PATCH net-next 06/19] net: usb: aqc111: Introduce link management

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]

Date:   Mon, 8 Oct 2018 09:29:26 +0000
From:   Igor Russkikh <Igor.Russkikh@...antia.com>
To:     Andrew Lunn <andrew@...n.ch>
CC:     "David S . Miller" <davem@...emloft.net>,
        "linux-usb@...r.kernel.org" <linux-usb@...r.kernel.org>,
        "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
        Dmitry Bezrukov <Dmitry.Bezrukov@...antia.com>
Subject: Re: [PATCH net-next 06/19] net: usb: aqc111: Introduce link
 management

Hi Andrew,

>>  	aqc111_read_fw_version(dev, aqc111_data);
>> +	aqc111_data->autoneg = AUTONEG_ENABLE;
>> +	aqc111_data->advertised_speed = (usb_speed == USB_SPEED_SUPER) ?
>> +					 SPEED_5000 : SPEED_1000;
> 

> USB 3 has a raw bandwidth of 5Gbps. But it is a shared bus. So you
> have no guaranteed you are actually going to get the needed bandwidth
> to support line rate.
> 
> USB 2.0 only gives you 480Mbps. So it won't even give you the full
> 1G. So using the same reasoning for USB3, maybe you should limit it to
> 100Mbps?
> 
> I personally would not apply restrictions on the PHY depending on what
> USB is being used.

First argument here is to reduce power consumption on USB2.
2.5G/5G uses OCSGMII/XFI serdes which consumes more power.
Of course in normal conditions usb2 is capable to feed that, but
the risk still exists on legacy usb2 hardware.

> This becomes more important when using SFPs. If i have an SFP peer
> which is expecting 2500Base-X, but because the device is plugged into
> USB 2 port it is forced to use 1000Base-X, it is not going to get
> link.

Do you mean here 2500Base-T? This particular device is an integrated
mac+phy, thus we can't easily link it with -X SFP endpoint.

Although its not a common usecase for the consumer dongle to connect to SFP
endpoints, think your comment is quite reasonable.
We'll clarify this internally.


Regards,
  Igor