[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20250118012632.qf4jmwz2bry43qqe@skbuf>
Date: Sat, 18 Jan 2025 03:26:32 +0200
From: Vladimir Oltean <olteanv@...il.com>
To: Tristram.Ha@...rochip.com
Cc: andrew@...n.ch, maxime.chevallier@...tlin.com,
Woojung.Huh@...rochip.com, davem@...emloft.net, edumazet@...gle.com,
kuba@...nel.org, pabeni@...hat.com, UNGLinuxDriver@...rochip.com,
netdev@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH net-next v2] net: dsa: microchip: Add SGMII port support
to KSZ9477 switch
On Sat, Jan 18, 2025 at 12:59:25AM +0000, Tristram.Ha@...rochip.com wrote:
> Some of the register definitions are not present in the XPCS driver so I
> need to add them.
Not a problem.
> Some register bits programmed by the XPCS driver do not have effect.
Like what?
> Actually KSZ9477 has a bug in SGMII implementation and needs a software
> workaround. I am not sure if the generic XCPS driver can cover that.
What kind of bug? In the integration or in the IP itself? Anyway,
SJA1105 had what you could call an integration bug too, and that's why
nxp_sja1105_sgmii_pma_config() exists. I am not sure either until you
are more specific.
It is a widespread hardware IP. I don't think it's unreasonable to have
a central driver for it with many quirks, as long as they're well documented
and clearly understood.
Powered by blists - more mailing lists