[1/4] USB: serial: ftdi_sio: fix IXON/IXOFF mixup

[1/4] USB: serial: ftdi_sio: fix IXON/IXOFF mixup

[Johan Hovold]

On Fri, May 18, 2018 at 03:53:54PM +0200, Greg Kroah-Hartman wrote:
> On Fri, May 18, 2018 at 03:25:47PM +0200, Johan Hovold wrote:
> > Since forever this driver has had IXON and IXOFF mixed up, and has used
> > the latter rather than the former to enable hardware-assisted software
> > flow control on output.

> Ugh, that's a huge sign that no one actually uses IXON/IXOFF anymore :(

Could be that some applications set both IXON and IXOFF "just to be
safe"; that appears to be the case with minicom at least.

> Anyway, thanks for fixing this up, should also probably be backported to
> the stable kernels.

I considered that, but since its a change in behaviour I figured it was
best that anyone affected hits this with a major (minor?) kernel version
update (rather than a stable update). But I'll add a stable tag if you
think it's warranted?

> Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Thanks for the review.

Johan
---
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[1/4] USB: serial: ftdi_sio: fix IXON/IXOFF mixup

[Greg Kroah-Hartman]

On Fri, May 18, 2018 at 04:33:50PM +0200, Johan Hovold wrote:
> On Fri, May 18, 2018 at 03:53:54PM +0200, Greg Kroah-Hartman wrote:
> > On Fri, May 18, 2018 at 03:25:47PM +0200, Johan Hovold wrote:
> > > Since forever this driver has had IXON and IXOFF mixed up, and has used
> > > the latter rather than the former to enable hardware-assisted software
> > > flow control on output.
> 
> > Ugh, that's a huge sign that no one actually uses IXON/IXOFF anymore :(
> 
> Could be that some applications set both IXON and IXOFF "just to be
> safe"; that appears to be the case with minicom at least.
> 
> > Anyway, thanks for fixing this up, should also probably be backported to
> > the stable kernels.
> 
> I considered that, but since its a change in behaviour I figured it was
> best that anyone affected hits this with a major (minor?) kernel version
> update (rather than a stable update). But I'll add a stable tag if you
> think it's warranted?

Fair enough.  If anyone notices we can easily backport it later on if
needed.

thanks,

greg k-h
---
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[1/4] USB: serial: ftdi_sio: fix IXON/IXOFF mixup

[Greg Kroah-Hartman]

On Fri, May 18, 2018 at 03:25:47PM +0200, Johan Hovold wrote:
> Since forever this driver has had IXON and IXOFF mixed up, and has used
> the latter rather than the former to enable hardware-assisted software
> flow control on output.
> 
> Signed-off-by: Johan Hovold <johan@kernel.org>
> ---
>  drivers/usb/serial/ftdi_sio.c | 8 ++------
>  1 file changed, 2 insertions(+), 6 deletions(-)
> 
> diff --git a/drivers/usb/serial/ftdi_sio.c b/drivers/usb/serial/ftdi_sio.c
> index 7ea221d42dba..62c99871863c 100644
> --- a/drivers/usb/serial/ftdi_sio.c
> +++ b/drivers/usb/serial/ftdi_sio.c
> @@ -2342,12 +2342,8 @@ static void ftdi_set_termios(struct tty_struct *tty,
>  	} else {
>  		/*
>  		 * Xon/Xoff code
> -		 *
> -		 * Check the IXOFF status in the iflag component of the
> -		 * termios structure. If IXOFF is not set, the pre-xon/xoff
> -		 * code is executed.
>  		 */
> -		if (iflag & IXOFF) {
> +		if (iflag & IXON) {
>  			dev_dbg(ddev, "%s  request to enable xonxoff iflag=%04x\n",
>  				__func__, iflag);
>  			/* Try to enable the XON/XOFF on the ftdi_sio
> @@ -2372,7 +2368,7 @@ static void ftdi_set_termios(struct tty_struct *tty,
>  			}
>  		} else {
>  			/* else clause to only run if cflag ! CRTSCTS and iflag
> -			 * ! XOFF. CHECKME Assuming XON/XOFF handled by tty
> +			 * ! XON. CHECKME Assuming XON/XOFF handled by tty
>  			 * stack - not by device */
>  			dev_dbg(ddev, "%s Turning off hardware flow control\n", __func__);
>  			if (usb_control_msg(dev,

Ugh, that's a huge sign that no one actually uses IXON/IXOFF anymore :(

Anyway, thanks for fixing this up, should also probably be backported to
the stable kernels.

Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[1/4] USB: serial: ftdi_sio: fix IXON/IXOFF mixup

[Johan Hovold]

Since forever this driver has had IXON and IXOFF mixed up, and has used
the latter rather than the former to enable hardware-assisted software
flow control on output.

Signed-off-by: Johan Hovold <johan@kernel.org>
---
 drivers/usb/serial/ftdi_sio.c | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)

diff --git a/drivers/usb/serial/ftdi_sio.c b/drivers/usb/serial/ftdi_sio.c
index 7ea221d42dba..62c99871863c 100644
--- a/drivers/usb/serial/ftdi_sio.c
+++ b/drivers/usb/serial/ftdi_sio.c
@@ -2342,12 +2342,8 @@ static void ftdi_set_termios(struct tty_struct *tty,
 	} else {
 		/*
 		 * Xon/Xoff code
-		 *
-		 * Check the IXOFF status in the iflag component of the
-		 * termios structure. If IXOFF is not set, the pre-xon/xoff
-		 * code is executed.
 		 */
-		if (iflag & IXOFF) {
+		if (iflag & IXON) {
 			dev_dbg(ddev, "%s  request to enable xonxoff iflag=%04x\n",
 				__func__, iflag);
 			/* Try to enable the XON/XOFF on the ftdi_sio
@@ -2372,7 +2368,7 @@ static void ftdi_set_termios(struct tty_struct *tty,
 			}
 		} else {
 			/* else clause to only run if cflag ! CRTSCTS and iflag
-			 * ! XOFF. CHECKME Assuming XON/XOFF handled by tty
+			 * ! XON. CHECKME Assuming XON/XOFF handled by tty
 			 * stack - not by device */
 			dev_dbg(ddev, "%s Turning off hardware flow control\n", __func__);
 			if (usb_control_msg(dev,