Revert "serial/core: Initialize the console pm state"

This reverts commit a86713b.

Kevin Hilman writes:

	Multiple boot failures on ARM[1] were bisected down to this
	patch.

	How was this patch tested, and on which platforms?

	Also, the changelog states that this should be done only for
	UART_CAP_SLEEP, but the patch does it for every UART.

	Greg, I suggest this patch be dropped from tty-next until it has
	been better described and tested.

	[1] http://lists.linaro.org/pipermail/kernel-build-reports/2014-October/005550.html

Reported-by: Kevin Hilman <khilman@kernel.org>
Cc: Sudhir Sreedharan <ssreedharan@mvista.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

drivers/tty/serial/serial_core.c

@@ -2606,9 +2606,6 @@ int uart_add_one_port(struct uart_driver *drv, struct uart_port *uport)
 	if (uport->cons && uport->dev)
 		of_console_check(uport->dev->of_node, uport->cons->name, uport->line);
 
-	if (uart_console(uport))
-		state->pm_state = UART_PM_STATE_ON;
-
 	uart_configure_port(drv, state, uport);
 
 	num_groups = 2;