[PATCH] serial port in use bug - discussion please

Ed Vance (EdV@macrolink.com)
Tue, 26 Mar 2002 17:40:49 -0800

This patch adds code to function set_serial_info() to fix two issues:

1. Function returned -EADDRINUSE when attempt was made to change fields
such as "baud_base" on a memory based serial port with the setserial
command. A check was added to bypass the port-in-use test loop if the
existing "iomem_base" field is nonzero, indicating a memory based
serial port that cannot be moved by this function.

2. Function allowed attempts to set "port" field on memory based serial
ports to a nonzero value. The iomem_base field cannot be changed by this
function, so a test was added to return -EINVALID when new "port" field
is nonzero and existing "iomem_base" field is also nonzero.

For kernel files rev 2.14.19-pre3

Contributor: Roman Kurakin <rik@cronyx.ru>

Subject: Serial.c Bug
Date: Wed, 14 Nov 2001 13:02:47 +0300
From: Roman Kurakin <rik@cronyx.ru>
To: linux-kernel@vger.kernel.org

I have found a bug. It is in support of serial cards which uses
memory for I/O instead of ports. I made a patch for serial.c and
fix one place, but probably the problem like this one could be
somewhere else.

If you try to use setserial with such cards you will get "Address in use"

Best regards,
Roman Kurakin

diff -urN -X dontdiff.txt linux-2.4.19-pre3/drivers/char/serial.c
--- linux-2.4.19-pre3/drivers/char/serial.c Thu Mar 14 16:19:02 2002
+++ patched/drivers/char/serial.c Tue Mar 26 16:06:06 2002
@@ -2131,6 +2131,7 @@
if ((new_serial.irq >= NR_IRQS) || (new_serial.irq < 0) ||
(new_serial.baud_base < 9600)|| (new_serial.type < PORT_UNKNOWN)
(new_serial.type > PORT_MAX) || (new_serial.type == PORT_CIRRUS)
+ (new_port && state->iomem_base) ||
(new_serial.type == PORT_STARTECH)) {
return -EINVAL;
@@ -2141,7 +2142,7 @@

/* Make sure address is not already in use */
- if (new_serial.type) {
+ if (!state->iomem_base && new_serial.type) {
for (i = 0 ; i < NR_PORTS; i++)
if ((state != &rs_table[i]) &&
(rs_table[i].port == new_port) &&

Ed Vance serial24@macrolink.com
Macrolink, Inc. 1500 N. Kellogg Dr Anaheim, CA 92807

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/