From: Andrew Lunn <andrew@lunn.ch>
To: Igor Russkikh <Igor.Russkikh@aquantia.com>
Cc: "David S . Miller" <davem@davemloft.net>,
"linux-usb@vger.kernel.org" <linux-usb@vger.kernel.org>,
"netdev@vger.kernel.org" <netdev@vger.kernel.org>,
Dmitry Bezrukov <Dmitry.Bezrukov@aquantia.com>
Subject: Re: [PATCH net-next 06/19] net: usb: aqc111: Introduce link management
Date: Mon, 8 Oct 2018 14:12:45 +0200 [thread overview]
Message-ID: <20181008121245.GA6216@lunn.ch> (raw)
In-Reply-To: <9318605b-6ddc-a093-52ad-76c18097c414@aquantia.com>
On Mon, Oct 08, 2018 at 09:29:26AM +0000, Igor Russkikh wrote:
> 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.
O.K, that sounds like a sensible argument. Please add a comment. I
hope the Marketing Department also understand this. It should probably
explain this on the product packaging.
> > 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.
I only went to find the product brief after finishing the
review. Without an external SERDES interface, SFP is not possible. But
from your comment above, i does sound like internally it has such a
SERDES. So it is not out of the question a follow up device is
produced which could connect to an SFP. I actually have a no-name USB
based SFP dongle...
Andrew
WARNING: multiple messages have this Message-ID (diff)
From: Andrew Lunn <andrew@lunn.ch>
To: Igor Russkikh <Igor.Russkikh@aquantia.com>
Cc: "David S . Miller" <davem@davemloft.net>,
"linux-usb@vger.kernel.org" <linux-usb@vger.kernel.org>,
"netdev@vger.kernel.org" <netdev@vger.kernel.org>,
Dmitry Bezrukov <Dmitry.Bezrukov@aquantia.com>
Subject: [net-next,06/19] net: usb: aqc111: Introduce link management
Date: Mon, 8 Oct 2018 14:12:45 +0200 [thread overview]
Message-ID: <20181008121245.GA6216@lunn.ch> (raw)
On Mon, Oct 08, 2018 at 09:29:26AM +0000, Igor Russkikh wrote:
> 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.
O.K, that sounds like a sensible argument. Please add a comment. I
hope the Marketing Department also understand this. It should probably
explain this on the product packaging.
> > 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.
I only went to find the product brief after finishing the
review. Without an external SERDES interface, SFP is not possible. But
from your comment above, i does sound like internally it has such a
SERDES. So it is not out of the question a follow up device is
produced which could connect to an SFP. I actually have a no-name USB
based SFP dongle...
Andrew
next prev parent reply other threads:[~2018-10-08 19:24 UTC|newest]
Thread overview: 133+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-10-05 10:24 [PATCH net-next 00/19] Add support for Aquantia AQtion USB to 5/2.5GbE devices Igor Russkikh
2018-10-05 10:24 ` [PATCH net-next 01/19] net: usb: aqc111: Driver skeleton for Aquantia AQtion USB to 5GbE Igor Russkikh
2018-10-05 10:24 ` [net-next,01/19] " Igor Russkikh
2018-10-09 13:37 ` [PATCH net-next 01/19] " Bjørn Mork
2018-10-09 13:37 ` [net-next,01/19] " Bjørn Mork
2018-10-05 10:24 ` [PATCH net-next 02/19] net: usb: aqc111: Add bind and empty unbind callbacks Igor Russkikh
2018-10-05 10:24 ` [net-next,02/19] " Igor Russkikh
2018-10-05 17:39 ` [PATCH net-next 02/19] " David Miller
2018-10-05 17:39 ` [net-next,02/19] " David Miller
2018-10-05 10:24 ` [PATCH net-next 03/19] net: usb: aqc111: Add implementation of read and write commands Igor Russkikh
2018-10-05 10:24 ` [net-next,03/19] " Igor Russkikh
2018-10-05 17:40 ` [PATCH net-next 03/19] " David Miller
2018-10-05 17:40 ` [net-next,03/19] " David Miller
2018-10-08 13:44 ` [PATCH net-next 03/19] " Oliver Neukum
2018-10-08 13:44 ` [net-next,03/19] " Oliver Neukum
2018-10-09 13:33 ` [PATCH net-next 03/19] " Bjørn Mork
2018-10-09 13:33 ` [net-next,03/19] " Bjørn Mork
2018-10-05 10:24 ` [PATCH net-next 04/19] net: usb: aqc111: Various callbacks implementation Igor Russkikh
2018-10-05 10:24 ` [net-next,04/19] " Igor Russkikh
2018-10-08 13:47 ` [PATCH net-next 04/19] " Oliver Neukum
2018-10-08 13:47 ` [net-next,04/19] " Oliver Neukum
2018-10-09 13:27 ` [PATCH net-next 04/19] " Bjørn Mork
2018-10-09 13:27 ` [net-next,04/19] " Bjørn Mork
2018-10-10 11:33 ` [PATCH net-next 04/19] " Oliver Neukum
2018-10-10 11:33 ` [net-next,04/19] " Oliver Neukum
2018-10-05 10:24 ` [PATCH net-next 05/19] net: usb: aqc111: Introduce PHY access Igor Russkikh
2018-10-05 10:24 ` [net-next,05/19] " Igor Russkikh
2018-10-05 22:04 ` [PATCH net-next 05/19] " Andrew Lunn
2018-10-05 22:04 ` [net-next,05/19] " Andrew Lunn
2018-10-08 9:09 ` [PATCH net-next 05/19] " Igor Russkikh
2018-10-08 9:09 ` [net-next,05/19] " Igor Russkikh
2018-10-08 12:17 ` [PATCH net-next 05/19] " Andrew Lunn
2018-10-08 12:17 ` [net-next,05/19] " Andrew Lunn
2018-10-08 14:07 ` [PATCH net-next 05/19] " Igor Russkikh
2018-10-08 14:07 ` [net-next,05/19] " Igor Russkikh
2018-10-10 0:58 ` [PATCH net-next 05/19] " Andrew Lunn
2018-10-10 0:58 ` [net-next,05/19] " Andrew Lunn
2018-10-10 7:54 ` [PATCH net-next 05/19] " Igor Russkikh
2018-10-10 7:54 ` [net-next,05/19] " Igor Russkikh
2018-10-08 13:52 ` [PATCH net-next 05/19] " Oliver Neukum
2018-10-08 13:52 ` [net-next,05/19] " Oliver Neukum
2018-10-08 14:10 ` [PATCH net-next 05/19] " Igor Russkikh
2018-10-08 14:10 ` [net-next,05/19] " Igor Russkikh
2018-10-08 14:24 ` [PATCH net-next 05/19] " Oliver Neukum
2018-10-08 14:24 ` [net-next,05/19] " Oliver Neukum
2018-10-05 10:24 ` [PATCH net-next 06/19] net: usb: aqc111: Introduce link management Igor Russkikh
2018-10-05 10:24 ` [net-next,06/19] " Igor Russkikh
2018-10-05 17:11 ` [PATCH net-next 06/19] " Andrew Lunn
2018-10-05 17:11 ` [net-next,06/19] " Andrew Lunn
2018-10-08 13:22 ` [PATCH net-next 06/19] " Igor Russkikh
2018-10-08 13:22 ` [net-next,06/19] " Igor Russkikh
2018-10-05 17:46 ` [PATCH net-next 06/19] " David Miller
2018-10-05 17:46 ` [net-next,06/19] " David Miller
2018-10-06 17:35 ` [PATCH net-next 06/19] " Andrew Lunn
2018-10-06 17:35 ` [net-next,06/19] " Andrew Lunn
2018-10-08 9:29 ` [PATCH net-next 06/19] " Igor Russkikh
2018-10-08 9:29 ` [net-next,06/19] " Igor Russkikh
2018-10-08 12:12 ` Andrew Lunn [this message]
2018-10-08 12:12 ` Andrew Lunn
2018-10-08 13:59 ` [PATCH net-next 06/19] " Oliver Neukum
2018-10-08 13:59 ` [net-next,06/19] " Oliver Neukum
2018-10-05 10:24 ` [PATCH net-next 07/19] net: usb: aqc111: Add support for getting and setting of MAC address Igor Russkikh
2018-10-05 10:24 ` [net-next,07/19] " Igor Russkikh
2018-10-06 1:03 ` [PATCH net-next 07/19] " Andrew Lunn
2018-10-06 1:03 ` [net-next,07/19] " Andrew Lunn
2018-10-09 14:34 ` [PATCH net-next 07/19] " Igor Russkikh
2018-10-09 14:34 ` [net-next,07/19] " Igor Russkikh
2018-10-09 14:46 ` [PATCH net-next 07/19] " Andrew Lunn
2018-10-09 14:46 ` [net-next,07/19] " Andrew Lunn
2018-10-05 10:25 ` [PATCH net-next 08/19] net: usb: aqc111: Implement TX data path Igor Russkikh
2018-10-05 10:25 ` [net-next,08/19] " Igor Russkikh
2018-10-06 1:13 ` [PATCH net-next 08/19] " Andrew Lunn
2018-10-06 1:13 ` [net-next,08/19] " Andrew Lunn
2018-10-08 13:43 ` [PATCH net-next 08/19] " Igor Russkikh
2018-10-08 13:43 ` [net-next,08/19] " Igor Russkikh
2018-10-08 14:07 ` [PATCH net-next 08/19] " Oliver Neukum
2018-10-08 14:07 ` [net-next,08/19] " Oliver Neukum
2018-10-09 13:50 ` [PATCH net-next 08/19] " Bjørn Mork
2018-10-09 13:50 ` [net-next,08/19] " Bjørn Mork
2018-10-05 10:25 ` [PATCH net-next 09/19] net: usb: aqc111: Implement RX " Igor Russkikh
2018-10-05 10:25 ` [net-next,09/19] " Igor Russkikh
2018-10-06 1:18 ` [PATCH net-next 09/19] " Andrew Lunn
2018-10-06 1:18 ` [net-next,09/19] " Andrew Lunn
2018-10-09 13:39 ` [PATCH net-next 09/19] " Bjørn Mork
2018-10-09 13:39 ` [net-next,09/19] " Bjørn Mork
2018-10-05 10:25 ` [PATCH net-next 10/19] net: usb: aqc111: Add checksum offload support Igor Russkikh
2018-10-05 10:25 ` [net-next,10/19] " Igor Russkikh
2018-10-05 10:25 ` [PATCH net-next 11/19] net: usb: aqc111: Add support for changing MTU Igor Russkikh
2018-10-05 10:25 ` [net-next,11/19] " Igor Russkikh
2018-10-06 16:56 ` [PATCH net-next 11/19] " Andrew Lunn
2018-10-06 16:56 ` [net-next,11/19] " Andrew Lunn
2018-10-05 10:25 ` [PATCH net-next 12/19] net: usb: aqc111: Add support for enable/disable checksum offload Igor Russkikh
2018-10-05 10:25 ` [net-next,12/19] " Igor Russkikh
2018-10-05 10:25 ` [PATCH net-next 13/19] net: usb: aqc111: Add support for TSO Igor Russkikh
2018-10-05 10:25 ` [net-next,13/19] " Igor Russkikh
2018-10-08 14:12 ` [PATCH net-next 13/19] " Oliver Neukum
2018-10-08 14:12 ` [net-next,13/19] " Oliver Neukum
2018-10-05 10:25 ` [PATCH net-next 14/19] net: usb: aqc111: Implement set_rx_mode callback Igor Russkikh
2018-10-05 10:25 ` [net-next,14/19] " Igor Russkikh
2018-10-06 17:03 ` [PATCH net-next 14/19] " Andrew Lunn
2018-10-06 17:03 ` [net-next,14/19] " Andrew Lunn
2018-10-08 13:49 ` [PATCH net-next 14/19] " Igor Russkikh
2018-10-08 13:49 ` [net-next,14/19] " Igor Russkikh
2018-10-05 10:25 ` [PATCH net-next 15/19] net: usb: aqc111: Add support for VLAN_CTAG_TX/RX offload Igor Russkikh
2018-10-05 10:25 ` [net-next,15/19] " Igor Russkikh
2018-10-08 14:14 ` [PATCH net-next 15/19] " Oliver Neukum
2018-10-08 14:14 ` [net-next,15/19] " Oliver Neukum
2018-10-05 10:25 ` [PATCH net-next 16/19] net: usb: aqc111: Add RX VLAN filtering support Igor Russkikh
2018-10-05 10:25 ` [net-next,16/19] " Igor Russkikh
2018-10-06 17:05 ` [PATCH net-next 16/19] " Andrew Lunn
2018-10-06 17:05 ` [net-next,16/19] " Andrew Lunn
2018-10-05 10:25 ` [PATCH net-next 17/19] net: usb: aqc111: Initialize ethtool_ops structure Igor Russkikh
2018-10-05 10:25 ` [net-next,17/19] " Igor Russkikh
2018-10-06 17:08 ` [PATCH net-next 17/19] " Andrew Lunn
2018-10-06 17:08 ` [net-next,17/19] " Andrew Lunn
2018-10-05 10:25 ` [PATCH net-next 18/19] net: usb: aqc111: Implement get/set_link_ksettings callbacks Igor Russkikh
2018-10-05 10:25 ` [net-next,18/19] " Igor Russkikh
2018-10-06 17:38 ` [PATCH net-next 18/19] " Andrew Lunn
2018-10-06 17:38 ` [net-next,18/19] " Andrew Lunn
2018-10-08 14:18 ` [PATCH net-next 18/19] " Oliver Neukum
2018-10-08 14:18 ` [net-next,18/19] " Oliver Neukum
2018-10-05 10:25 ` [PATCH net-next 19/19] net: usb: aqc111: Add support for wake on LAN by MAGIC packet Igor Russkikh
2018-10-05 10:25 ` [net-next,19/19] " Igor Russkikh
2018-10-06 17:49 ` [PATCH net-next 19/19] " Andrew Lunn
2018-10-06 17:49 ` [net-next,19/19] " Andrew Lunn
2018-10-08 14:12 ` [PATCH net-next 19/19] " Igor Russkikh
2018-10-08 14:12 ` [net-next,19/19] " Igor Russkikh
2018-10-08 14:47 ` [PATCH net-next 19/19] " Andrew Lunn
2018-10-08 14:47 ` [net-next,19/19] " Andrew Lunn
2018-10-06 17:51 ` [PATCH net-next 00/19] Add support for Aquantia AQtion USB to 5/2.5GbE devices Andrew Lunn
2018-10-08 7:58 ` Igor Russkikh
2018-10-08 14:21 ` Oliver Neukum
2018-10-08 14:52 ` Igor Russkikh
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20181008121245.GA6216@lunn.ch \
--to=andrew@lunn.ch \
--cc=Dmitry.Bezrukov@aquantia.com \
--cc=Igor.Russkikh@aquantia.com \
--cc=davem@davemloft.net \
--cc=linux-usb@vger.kernel.org \
--cc=netdev@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.