[PATCH] serial: 8250: Make ISA ports optional

Peter Hurley peter at hurleysoftware.com
Thu Jan 8 15:25:03 PST 2015

On 01/08/2015 05:36 PM, One Thousand Gnomes wrote:
>> One idea that has come up in the past but never saw an implementation
>> is to make the ttyS namespace and minor numbers completely generic and
>> let any serial port driver use it. This would be a major rework, but
>> have the added advantage of cleaning up a number of other namespace
>> issues as well. There also lots of open question, in particular how
>> to maintain compatibility with existing drivers. One could imagine
>> that each uart always gets a ttyS device and optionally also gets
>> a device node for the same port with a driver specific chardev as
>> most of them do today. Or it could be an either/or decision that is
>> made at compile time or as a module parameter.
> I'm not sure we should use ttyS for it. ttyS means 8250, a lot of
> existing code knows what ttyS is and what sort of ioctls and the like are
> assumed.

That genie has already escaped the bottle.

./arch/powerpc/platforms/powermac/setup.c:	char *devname = "ttyS";
./arch/powerpc/platforms/chrp/setup.c:		add_preferred_console("ttyS", 0, NULL);
./arch/powerpc/kernel/legacy_serial.c:		return add_preferred_console("ttyS", offset, opt);
./arch/xtensa/platforms/iss/console.c:	serial_driver->name = "ttyS";
./arch/mips/sni/setup.c:			add_preferred_console("ttyS", port,
./arch/mips/sgi-ip32/ip32-setup.c:			add_preferred_console("ttyS", *(con + 1) == '2' ? 1 : 0,
./arch/mips/jazz/setup.c:	add_preferred_console("ttyS", 0, "9600");
./arch/mips/sgi-ip22/ip22-setup.c:		add_preferred_console("ttyS", *(ctype + 1) == '2' ? 1 : 0,
./arch/arm/mach-davinci/board-da830-evm.c:	return add_preferred_console("ttyS", 2, "115200");
./arch/arm/mach-davinci/board-mityomapl138.c:	return add_preferred_console("ttyS", 1, "115200");
./arch/arm/mach-davinci/board-da850-evm.c:	return add_preferred_console("ttyS", 2, "115200");
./arch/arm/mach-davinci/board-omapl138-hawk.c:	return add_preferred_console("ttyS", 2, "115200");
./arch/cris/arch-v10/kernel/debugport.c:	name : "ttyS",
./arch/ia64/hp/sim/simserial.c:	hp_simserial_driver->name = "ttyS";
./arch/um/drivers/ssl.c:	.device_name 		= "ttyS",
./arch/s390/kernel/setup.c:			add_preferred_console("ttyS", 1, NULL);
./drivers/tty/amiserial.c:	.name =		"ttyS",
./drivers/tty/serial/serial_txx9.c:#define TXX9_TTY_NAME "ttyS"
./drivers/tty/serial/sunsab.c:	.dev_name		= "ttyS",
./drivers/tty/serial/mcf.c:	.dev_name	= "ttyS",
./drivers/tty/serial/atmel_serial.c:#define ATMEL_DEVICENAME	"ttyS"
./drivers/tty/serial/bcm63xx_uart.c:	.dev_name	= "ttyS",
./drivers/tty/serial/sunsu.c:	.dev_name		= "ttyS",
./drivers/tty/serial/mrst_max3110.c:	.dev_name	= "ttyS",
./drivers/tty/serial/apbuart.c:	.name = "ttyS",
./drivers/tty/serial/apbuart.c:	.dev_name = "ttyS",
./drivers/tty/serial/tilegx.c:#define TILEGX_UART_NAME	"ttyS"
./drivers/tty/serial/pmac_zilog.c:#define PMACZILOG_NAME		"ttyS"
./drivers/tty/serial/pnx8xxx_uart.c:	.dev_name		= "ttyS",
./drivers/tty/serial/m32r_sio.c:	.dev_name		= "ttyS",
./drivers/tty/serial/ip22zilog.c:	.dev_name	= "ttyS",
./drivers/tty/serial/zs.c:	.dev_name		= "ttyS",
./drivers/tty/serial/68328serial.c:	serial_driver->name = "ttyS";
./drivers/tty/serial/crisv10.c:	driver->name = "ttyS";
/drivers/tty/serial/8250/8250_core.c:	.dev_name		= "ttyS",
./drivers/tty/serial/sunzilog.c:	.dev_name	=	"ttyS",
./drivers/tty/serial/pxa.c:	.dev_name	= "ttyS",
./drivers/tty/serial/dz.c:	.dev_name		= "ttyS",
./drivers/tty/serial/sunhv.c:	.dev_name		= "ttyS",
./drivers/s390/char/sclp_vt220.c:#define SCLP_VT220_CONSOLE_NAME		"ttyS"
./drivers/s390/char/sclp_con.c:#define sclp_console_name  "ttyS"
./drivers/s390/char/con3215.c:	driver->name = "ttyS";

Besides, despite its pedigree, omap is no closer to 8250 than anything
else in drivers/tty/serial/

> A generic namespace needs not to be attached to a legacy naming 
> (eg also put all serial ports in /dev/serial/0 /dev/serial/1 etc either
> via the tty layer or via systemd/udev)

This doesn't address pre-init.

Peter Hurley

More information about the linux-arm-kernel mailing list