From mboxrd@z Thu Jan  1 00:00:00 1970
From: Kay Sievers <kay.sievers@vrfy.org>
Date: Wed, 07 Oct 2009 20:06:13 +0000
Subject: Re: [BUG] No ttyS0 - Strange udev 8250_pnp interaction
Message-Id: <ac3eb2510910071306r4ad6dc9btb7503bc3ab0ec815@mail.gmail.com>
List-Id: <linux-hotplug.vger.kernel.org>
References: <20091007175403.GA335@trillian.comsick.at> <9b2b86520910071246r393f2afvacaa4d8f544508a8@mail.gmail.com>
In-Reply-To: <9b2b86520910071246r393f2afvacaa4d8f544508a8@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
To: Alan Jenkins <sourcejedi.lkml@googlemail.com>
Cc: Michael Guntsche <mike@it-loops.com>, linux-kernel <linux-kernel@vger.kernel.org>, linux-hotplug <linux-hotplug@vger.kernel.org>

On Wed, Oct 7, 2009 at 21:46, Alan Jenkins
<sourcejedi.lkml@googlemail.com> wrote:
> Udev would have avoided the race prior to
>
> 82c785e "udevd: remove check for dev_t, DEVPATH_OLD takes care of that"
>
> (the "check" removed here used to serialize events based on the device
> major:minor number).
>
> So - udev probably needs to add the check back.

What are the two devices with the same major/minor but a different DEVPATH?

> It might also be the case that future kernels could set DEVPATH_OLD as
> suggested. =C2=A0I don't know the gory details of 8250_pnp though.

The DEVPATH_OLD is only added when one and the same device was renamed
or moved, which isn't the case here, I guess?

Kay