Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 20 Apr 2007 09:45:43 -0700
From:      Marcel Moolenaar <xcllnt@mac.com>
To:        Jeremy Chadwick <koitsu@FreeBSD.org>
Cc:        freebsd-stable@FreeBSD.org, Stefan Lambrev <stefan.lambrev@sun-fish.com>
Subject:   Re: sio0: port may not be enabled
Message-ID:  <049954BE-364B-4897-87C3-342D0A824C00@mac.com>
In-Reply-To: <20070420152329.GA16702@icarus.home.lan>
References:  <4628D63A.3050909@sun-fish.com> <20070420152329.GA16702@icarus.home.lan>

next in thread | previous in thread | raw e-mail | index | archive | help

On Apr 20, 2007, at 8:23 AM, Jeremy Chadwick wrote:

> Look closely at the dmesg line, note what device sio0 is claiming =20
> to be
> associated with (acpi0, not isa0):
>
>> sio0: <16550A-compatible COM port> port 0x3f8-0x3ff irq 4 flags =20
>> 0x10 on acpi0
>
> This is one of the drawbacks to using ACPI.

This is not a drawback. It's partly why ACPI was designed and =20
implemented:
to describe legacy hardware.

> Some systems apparently tie the serial port to ACPI functionality in a
> different way.  For example, I have a couple boxes which have sio0
> attached to acpi0 that work fine.  In some other cases, I have ones
> which result in a non-working serial port unless I disable ACPI (thus
> sio0 shows up as being attached to isa0).

Could you try uart(4) instead. It seems quite excessive to have to
disable ACPI just to get a serial port working. I'd like to know
if this is related to the sio(4) driver or something else.

=04Thanks,

--=20
Marcel Moolenaar
xcllnt@mac.com





Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?049954BE-364B-4897-87C3-342D0A824C00>