[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20230823073743.61770fa4@kernel.org>
Date: Wed, 23 Aug 2023 07:37:43 -0700
From: Jakub Kicinski <kuba@...nel.org>
To: Ruan Jinjie <ruanjinjie@...wei.com>
Cc: <rafal@...ecki.pl>, <bcm-kernel-feedback-list@...adcom.com>,
<davem@...emloft.net>, <edumazet@...gle.com>, <pabeni@...hat.com>,
<opendmb@...il.com>, <florian.fainelli@...adcom.com>,
<bryan.whitehead@...rochip.com>, <UNGLinuxDriver@...rochip.com>,
<netdev@...r.kernel.org>, <andrew@...n.ch>
Subject: Re: [PATCH net-next v4 3/3] net: lan743x: Return PTR_ERR() for
fixed_phy_register()
On Wed, 23 Aug 2023 15:43:58 +0800 Ruan Jinjie wrote:
> > EPROBE_DEFER is not a unix error code. We can't return it to user
> > space, so propagating it from ndo_open is not correct.
>
> When the error is EPROBE_DEFER, Whether print the netdev_err is ok? And
> what should it return?
>
> How about this?
>
> if (IS_ERR(phydev)) {
> ......
> if (PTR_ERR(phydev) != -EPROBE_DEFER)
> return PTR_ERR(phydev);
> else
> return -EIO;
> }
That's too much code to copy & paste into every driver.
Someone who understands the code flow very well should tackle this.
Please leave the cases where fixed_phy_register() is called outside
of probe alone.
Powered by blists - more mailing lists