[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20220622161229.3a08de6b@kernel.org>
Date: Wed, 22 Jun 2022 16:12:29 -0700
From: Jakub Kicinski <kuba@...nel.org>
To: Paolo Abeni <pabeni@...hat.com>
Cc: <linux@...linux.org.uk>, netdev@...r.kernel.org,
Andrew Lunn <andrew@...n.ch>,
Heiner Kallweit <hkallweit1@...il.com>, <davem@...emloft.net>,
Eric Dumazet <edumazet@...gle.com>,
Ong Boon Leong <boon.leong.ong@...el.com>,
Masahiro Yamada <masahiroy@...nel.org>
Subject: Re: [PATCH net-next] net: pcs: xpcs: depends on PHYLINK in Kconfig
On Wed, 22 Jun 2022 17:42:13 +0200 Paolo Abeni wrote:
> @Jakub: please let me know if you prefer to go ahead yourself, or me
> sending a v3 with 'depends PHYLINK' + the above (or any other option ;)
[resending, sorry, Russell let me know that my MUA broke the headers]
Well, IDK. You said "depends PHYLINK" which makes me feel like I
haven't convinced you at all :) Unless you mean add the dependency
on the consumers not on PCS_XPCS itself, but that's awkward.
What I was saying is that "depends" in a symbol which is only
"select"ed by other symbols makes no sense. IIUC "select" does not
visit dependencies, so putting "depends" on an user-invisible symbol
(i.e. symbol without a prompt) achieves nothing.
So PCS_XPCS can have no "depends" if we hide it.
The way I see it - PHYLINK already selects MDIO_DEVICE. So we can drop
the MDIO business from PCS_XPCS, add "select PHYLINK", hide it by
removing the prompt, and we're good. Then again, I admit I have not
tested this at all so I could be speaking gibberish...
diff --git a/drivers/net/pcs/Kconfig b/drivers/net/pcs/Kconfig
index 22ba7b0b476d..f778e5155fae 100644
--- a/drivers/net/pcs/Kconfig
+++ b/drivers/net/pcs/Kconfig
@@ -6,8 +6,8 @@
menu "PCS device drivers"
config PCS_XPCS
- tristate "Synopsys DesignWare XPCS controller"
- depends on MDIO_DEVICE && MDIO_BUS
+ tristate
+ select PHYLINK
help
This module provides helper functions for Synopsys DesignWare XPCS
Powered by blists - more mailing lists