serial: 8250_exar: Read INT0 from slave device, too

The sleep wake-up refactoring that I introduced in

  commit c7e1b405 ("tty: serial: exar: Relocate sleep wake-up handling")

did not account for devices with a slave device on the expansion port.
This patch pokes the INT0 register in the slave device, if present, in
order to ensure that MSI interrupts don't get permanently "stuck"
because of a sleep wake-up interrupt as described here:

  commit 2c0ac5b4 ("serial: exar: Fix stuck MSIs")

This also converts an ioread8() to readb() in order to provide visual
consistency with the MMIO-only accessors used elsewhere in the driver.
Reported-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Signed-off-by: NAaron Sierra <asierra@xes-inc.com>
Fixes: c7e1b405 ("tty: serial: exar: Relocate sleep wake-up handling")
Reviewed-by: NAndy Shevchenko <andriy.shevchenko@linux.intel.com>
Cc: stable <stable@vger.kernel.org>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

drivers/tty/serial/8250/8250_exar.c

@@ -445,7 +445,11 @@ static irqreturn_t exar_misc_handler(int irq, void *data)
 	struct exar8250 *priv = data;
 
 	/* Clear all PCI interrupts by reading INT0. No effect on IIR */
-	ioread8(priv->virt + UART_EXAR_INT0);
+	readb(priv->virt + UART_EXAR_INT0);
+
+	/* Clear INT0 for Expansion Interface slave ports, too */
+	if (priv->board->num_ports > 8)
+		readb(priv->virt + 0x2000 + UART_EXAR_INT0);
 
 	return IRQ_HANDLED;
 }