netdev - Re: [PATCH net-next] net: phy: add phylib-internal.h

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: <4b2ebf2e-28b3-4b6a-9772-e5495c18b1d6@lunn.ch>
Date: Wed, 19 Feb 2025 14:35:57 +0100
From: Andrew Lunn <andrew@...n.ch>
To: Heiner Kallweit <hkallweit1@...il.com>
Cc: Russell King - ARM Linux <linux@...linux.org.uk>,
	Jakub Kicinski <kuba@...nel.org>, Paolo Abeni <pabeni@...hat.com>,
	Eric Dumazet <edumazet@...gle.com>,
	David Miller <davem@...emloft.net>,
	"netdev@...r.kernel.org" <netdev@...r.kernel.org>
Subject: Re: [PATCH net-next] net: phy: add phylib-internal.h

On Wed, Feb 19, 2025 at 07:37:52AM +0100, Heiner Kallweit wrote:
> On 18.02.2025 23:43, Heiner Kallweit wrote:
> > This patch is a starting point for moving phylib-internal
> > declarations to a private header file.
> > 
> > Signed-off-by: Heiner Kallweit <hkallweit1@...il.com>
> > ---
> >  drivers/net/phy/phy-c45.c          |  1 +
> >  drivers/net/phy/phy-core.c         |  3 ++-
> >  drivers/net/phy/phy.c              |  2 ++
> >  drivers/net/phy/phy_device.c       |  2 ++
> >  drivers/net/phy/phy_led_triggers.c |  2 ++
> >  drivers/net/phy/phylib-internal.h  | 25 +++++++++++++++++++++++++
> >  include/linux/phy.h                | 13 -------------
> >  7 files changed, 34 insertions(+), 14 deletions(-)
> >  create mode 100644 drivers/net/phy/phylib-internal.h

Maybe we should discuss a little where we are going with this...

I like the idea, we want to limit the scope of some functions so they
don't get abused. MAC drivers are the main abusers here, they should
have a small set of methods they can use.

If you look at some other subsystems they have a header for consumers,
and a header for providers. include/linux/gpio/{consumer|driver}.h,
clk-provider.h and clk.h, etc.

Do we want include/linux/phy.h to contain the upper API for phylib,
which MAC drivers can use? Should phylib-internal.h be just the
internal API between parts of the core? There will be another header
which has everything a PHY driver needs for the lower interface of
phylib?

Is this what you are thinking?

	Andrew