]> bbs.cooldavid.org Git - net-next-2.6.git/commitdiff
atmel_serial: fix hang in set_termios when crtscts is enabled
authorHaavard Skinnemoen <haavard.skinnemoen@atmel.com>
Tue, 16 Jun 2009 16:02:03 +0000 (17:02 +0100)
committerLinus Torvalds <torvalds@linux-foundation.org>
Tue, 16 Jun 2009 19:01:16 +0000 (12:01 -0700)
After enabling hardware flow control, any subsequent termios call may hang
waiting for the transmitter to drain.  This appears to be caused by a
busy-loop in set_termios() waiting for the transmitter to become empty,
which may take a very long time (or hang indefinitely) if the device at
the other end is blocking us.

A quick look through the tty and serial_core code indicates that any
necessary flushing (which is optional) has already been done at this
point, so there's no need for the driver to flush the transmitter on its
own.

Fix it by removing the busy-loop altogether.

Tested-by: Eirik Aanonsen <eaa@wprmedical.com>
Signed-off-by: Haavard Skinnemoen <haavard.skinnemoen@atmel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
drivers/serial/atmel_serial.c

index b3497d7e5354dec9ab0ba49cd754f5e47867c45b..338b15c0a548e9f2d6d630d1ea3eb52954abff0a 100644 (file)
@@ -1104,11 +1104,13 @@ static void atmel_set_termios(struct uart_port *port, struct ktermios *termios,
        /* update the per-port timeout */
        uart_update_timeout(port, termios->c_cflag, baud);
 
-       /* save/disable interrupts and drain transmitter */
+       /*
+        * save/disable interrupts. The tty layer will ensure that the
+        * transmitter is empty if requested by the caller, so there's
+        * no need to wait for it here.
+        */
        imr = UART_GET_IMR(port);
        UART_PUT_IDR(port, -1);
-       while (!(UART_GET_CSR(port) & ATMEL_US_TXEMPTY))
-               cpu_relax();
 
        /* disable receiver and transmitter */
        UART_PUT_CR(port, ATMEL_US_TXDIS | ATMEL_US_RXDIS);