Date: Sun, 4 Sep 2005 11:19:01 +0100
From: Russell King <rmk+lkml@arm.linux.org.uk>
To: Linux Kernel List <linux-kernel@vger.kernel.org>, jongk@linux-m68k.org
Subject: 8250_hp300: initialisation ordering bug
Message-ID: <20050904111901.A30509@flint.arm.linux.org.uk>
Mail-Followup-To: Linux Kernel List <linux-kernel@vger.kernel.org>,
	jongk@linux-m68k.org
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1188
Lines: 34

Hi,

I've noticed that 8250_hp300 is buggy wrt the ordering of hardware
initialisation to the visibility of devices to user space.  Namely,
8250_hp300 does the following:

        line = serial8250_register_port(&port);
...
        /* Enable board-interrupts */
        out_8(d->resource.start + DIO_VIRADDRBASE + DCA_IC, DCA_IC_IE);
        dio_set_drvdata(d, (void *)line);

        /* Reset the DCA */
        out_8(d->resource.start + DIO_VIRADDRBASE + DCA_ID, 0xff);
        udelay(100);

serial8250_register_port() makes the port visible to userspace, so
from that point on it could be opened.  However, if it's opened
prior to the remainder of the above completing, we will be missing
interrupts (and what effect does "reset the DCA" have?)

Surely this hardware fiddling should be completed before we register
the port?


-- 
Russell King
 Linux kernel    2.6 ARM Linux   - http://www.arm.linux.org.uk/
 maintainer of:  2.6 Serial core
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/