Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::3:3 as permitted sender) client-ip=2620:137:e000::3:3;
Date:   Wed, 13 Dec 2023 15:01:09 +0000
From:   "Russell King (Oracle)" <linux@armlinux.org.uk>
To:     Andrew Lunn <andrew@lunn.ch>
Cc:     Justin Chen <justin.chen@broadcom.com>, netdev@vger.kernel.org,
        Doug Berger <opendmb@gmail.com>,
        Florian Fainelli <florian.fainelli@broadcom.com>,
        Broadcom internal kernel review list 
        <bcm-kernel-feedback-list@broadcom.com>,
        Heiner Kallweit <hkallweit1@gmail.com>,
        "David S. Miller" <davem@davemloft.net>,
        Eric Dumazet <edumazet@google.com>,
        Jakub Kicinski <kuba@kernel.org>,
        Paolo Abeni <pabeni@redhat.com>,
        open list <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH] net: mdio: mdio-bcm-unimac: Delay before first poll
Message-ID: <ZXnHNTreKY/F2Aqm@shell.armlinux.org.uk>
References: <20231213000249.2020835-1-justin.chen@broadcom.com>
 <c3cc7a9d-d464-48e7-beb7-b90b1abbcfc7@lunn.ch>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <c3cc7a9d-d464-48e7-beb7-b90b1abbcfc7@lunn.ch>
Sender: Russell King (Oracle) <linux@armlinux.org.uk>
Precedence: bulk

On Wed, Dec 13, 2023 at 11:57:52AM +0100, Andrew Lunn wrote:
> On Tue, Dec 12, 2023 at 04:02:49PM -0800, Justin Chen wrote:
> > With a clock interval of 400 nsec and a 64 bit transactions (32 bit
> > preamble & 16 bit control & 16 bit data), it is reasonable to assume
> > the mdio transaction will take 25.6 usec. Add a 30 usec delay before
> > the first poll to reduce the chance of a 1000-2000 usec sleep.
> 
> #define  MDIO_C45               0
> 
> suggests the hardware can do C45? The timing works out different then.
> Maybe add a comment by the udelay() that is assumes C22, to give a
> clue to somebody who is adding C45 support the delay needs to be
> re-evaluated.

Note, however, that the driver only supports C22 operations (it only
populates the read|write functions, not the c45 variants).

However, it doesn't explicitly set the MDIO_C22 bit in the configuration
register, so what ends up being spat out on the bus would be dependent
on the boot loader configuration.

However, I'm wondering why unimac_mdio_poll() isn't written as
(based on current code):

	return read_poll_timeout(unimac_mdio_readl(priv, MDIO_CMD), val,
				 !(val & MDIO_START_BUSY), 2000,
				 2000000);

rather than open-coding the io polling.

-- 
RMK's Patch system: https://www.armlinux.org.uk/developer/patches/
FTTP is here! 80Mbps down 10Mbps up. Decent connectivity at last!