linux-kernel - Re: [PATCH v3 1/2] leds: trigger: netdev: extend speeds up to 10G

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: <ZXH_OdREkrt-7Bwf@makrotopia.org>
Date:   Thu, 7 Dec 2023 17:22:01 +0000
From:   Daniel Golle <daniel@...rotopia.org>
To:     Andrew Lunn <andrew@...n.ch>
Cc:     Marek Behún <kabel@...nel.org>,
        Lee Jones <lee@...nel.org>, Pavel Machek <pavel@....cz>,
        Jakub Kicinski <kuba@...nel.org>,
        Christian Marangi <ansuelsmth@...il.com>,
        "David S. Miller" <davem@...emloft.net>,
        Li Zetao <lizetao1@...wei.com>, linux-kernel@...r.kernel.org,
        linux-leds@...r.kernel.org
Subject: Re: [PATCH v3 1/2] leds: trigger: netdev: extend speeds up to 10G

On Thu, Dec 07, 2023 at 06:11:29PM +0100, Andrew Lunn wrote:
> On Thu, Dec 07, 2023 at 05:29:23PM +0100, Marek Behún wrote:
> > On Tue, 28 Nov 2023 04:00:10 +0000
> > Daniel Golle <daniel@...rotopia.org> wrote:
> > 
> > > Add 2.5G, 5G and 10G as available speeds to the netdev LED trigger.
> > > 
> > > Signed-off-by: Daniel Golle <daniel@...rotopia.org>
> > > Reviewed-by: Andrew Lunn <andrew@...n.ch>
> > 
> > So what will happen when there are more speeds? Will we create a
> > separate file for each speed?
> > 
> > Will we have a separate sysfs file for 10, 100, 1000, 2500, 5000,
> > 10000, 20000, 25000, 40000, 50000, 56000, 100000, 200000, 400000,
> > 800000 ?

Yes, why not?

I also doubt we are doing to have 800GBit/s on RJ-45 copper any time
soon. And speed-indication is generally only needed for media which
supports more than one speed (ie. twisted copper pair).

> > 
> > These are all speeds from include/uapi/linux/ethtool.h.
> > 
> > Maybe we should have reused ethtool link mode bits, or something...
> 
> That gets pretty ugly. The bits are not in any logical order, since
> they just get appended onto the end as needed.
> 
> > Also, the files should only be present if the requested speed is
> > supported by the net device. So if 2500 mbps is not supported, there
> > should no be link_2500.
> 
> Yes, this would be nice. We have the information in the phy_setting
> settings[] table in phy-core.c.

I agree on that, it would be an additional patch though because that
obviously goes beyond adding some more speeds.