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]
Date:   Wed, 12 May 2021 14:38:25 +0200
From:   Andrew Lunn <andrew@...n.ch>
To:     Peter Geis <pgwipeout@...il.com>
Cc:     Heiner Kallweit <hkallweit1@...il.com>,
        Russell King <linux@...linux.org.uk>,
        "David S . Miller" <davem@...emloft.net>,
        Jakub Kicinski <kuba@...nel.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        Linux Kernel Network Developers <netdev@...r.kernel.org>,
        "open list:ARM/Rockchip SoC..." <linux-rockchip@...ts.infradead.org>
Subject: Re: [PATCH v2] net: phy: add driver for Motorcomm yt8511 phy

Hi Peter

> A lot of items should be set up via the device tree, though it seems
> this is a relatively unused concept in the net phy subsystem.

Very little should be set up via device tree, since it does not
describe the hardware. The interface mode does describe the hardware,
so that is expected to be in DT, but not too much else.

> > Do you know which one of the four RGMII modes your setup needs? Is the
> > PHY adding the Rx and Tx delays? So "rgmii-id"?
> 
> By default it implements a 500ps delay internally on the txd clock and
> a 1.2 ns delay on the rx clock.
> The controller is the snps,dwmac-4.20a, and it implements a default
> delay as well.

O.K, that is confusing. We generally recommend that the MAC does not
add delays, the PHY does it. So maybe you can implement "rgmii-id" in
the PHY, and return -EOPNOTSUP for the other three RGMII modes, as a
minimum. However dwmac is one of the oddball drivers which does
sometime add the delays, and always sets the PHY to "rgmii" so it does
not add delays. Either way, is O.K, but please avoid having some of
the delay on one side, and the rest on the other.

	Andrew

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ