diff mbox series

[v3,2/2] USB: cdc-acm: fix rounding error in TIOCSSERIAL

Message ID 20200310101027.32152-2-anthony.mallet@laas.fr (mailing list archive)
State Superseded
Headers show
Series [v3,1/2] USB: cdc-acm: fix close_delay and closing_wait units in TIOCSSERIAL | expand

Commit Message

Anthony Mallet March 10, 2020, 10:10 a.m. UTC 
Quoting the bug reporter:

By default, tty_port_init() initializes those parameters to a multiple
of HZ. For instance in line 69 of tty_port.c:
   port->close_delay = (50 * HZ) / 100;
https://github.com/torvalds/linux/blob/master/drivers/tty/tty_port.c#L69

With e.g. CONFIG_HZ = 250 (as this is the case for Ubuntu 18.04
linux-image-4.15.0-37-generic), the default setting for close_delay is
thus 125.

When ioctl(fd, TIOCGSERIAL, &s) is executed, the setting returned in
user space is '12' (125/10). When ioctl(fd, TIOCSSERIAL, &s) is then
executed with the same setting '12', the value is interpreted as '120'
which is different from the current setting and a EPERM error may be
raised by set_serial_info() if !CAP_SYS_ADMIN.
https://github.com/torvalds/linux/blob/master/drivers/usb/class/cdc-acm.c#L919

Fixes: ba2d8ce9db0a6 ("cdc-acm: implement TIOCSSERIAL to avoid blocking close(2)")
Signed-off-by: Anthony Mallet <anthony.mallet@laas.fr>
---
Changed in v2: fix typo (extra closing brace)
Changed in v3: version this patch series
---
 drivers/usb/class/cdc-acm.c | 27 +++++++++++++++++----------
 1 file changed, 17 insertions(+), 10 deletions(-)

2.17.1

Comments

Greg KH March 12, 2020, 8:38 a.m. UTC | #1 
On Tue, Mar 10, 2020 at 11:10:27AM +0100, Anthony Mallet wrote:
> Quoting the bug reporter:
> 
> By default, tty_port_init() initializes those parameters to a multiple
> of HZ. For instance in line 69 of tty_port.c:
>    port->close_delay = (50 * HZ) / 100;
> https://github.com/torvalds/linux/blob/master/drivers/tty/tty_port.c#L69
> 
> With e.g. CONFIG_HZ = 250 (as this is the case for Ubuntu 18.04
> linux-image-4.15.0-37-generic), the default setting for close_delay is
> thus 125.
> 
> When ioctl(fd, TIOCGSERIAL, &s) is executed, the setting returned in
> user space is '12' (125/10). When ioctl(fd, TIOCSSERIAL, &s) is then
> executed with the same setting '12', the value is interpreted as '120'
> which is different from the current setting and a EPERM error may be
> raised by set_serial_info() if !CAP_SYS_ADMIN.
> https://github.com/torvalds/linux/blob/master/drivers/usb/class/cdc-acm.c#L919
> 
> Fixes: ba2d8ce9db0a6 ("cdc-acm: implement TIOCSSERIAL to avoid blocking close(2)")
> Signed-off-by: Anthony Mallet <anthony.mallet@laas.fr>
> Cc: stable <stable@vger.kernel.org>
> ---
> Changed in v2: fix typo (extra closing brace)
> Changed in v3: version this patch series
> ---
>  drivers/usb/class/cdc-acm.c | 27 +++++++++++++++++----------
>  1 file changed, 17 insertions(+), 10 deletions(-)
> 
> diff --git a/drivers/usb/class/cdc-acm.c b/drivers/usb/class/cdc-acm.c
> index da619176deca..a41a3d27016c 100644
> --- a/drivers/usb/class/cdc-acm.c
> +++ b/drivers/usb/class/cdc-acm.c
> @@ -907,6 +907,7 @@ static int set_serial_info(struct tty_struct *tty, struct serial_struct *ss)
>  {
>  	struct acm *acm = tty->driver_data;
>  	unsigned int closing_wait, close_delay;
> +	unsigned int old_closing_wait, old_close_delay;
>  	int retval = 0;
>  
>  	close_delay = msecs_to_jiffies(ss->close_delay * 10);
> @@ -914,19 +915,24 @@ static int set_serial_info(struct tty_struct *tty, struct serial_struct *ss)
>  			ASYNC_CLOSING_WAIT_NONE :
>  			msecs_to_jiffies(ss->closing_wait * 10);
>  
> +	/* we must redo the rounding here, so that the values match */
> +	old_close_delay	= jiffies_to_msecs(acm->port.close_delay) / 10;
> +	old_closing_wait = acm->port.closing_wait == ASYNC_CLOSING_WAIT_NONE ?
> +				ASYNC_CLOSING_WAIT_NONE :
> +				jiffies_to_msecs(acm->port.closing_wait) / 10;
> +
>  	mutex_lock(&acm->port.mutex);
>  
> -	if (!capable(CAP_SYS_ADMIN)) {
> -		if ((close_delay != acm->port.close_delay) ||
> -		    (closing_wait != acm->port.closing_wait))
> +	if ((ss->close_delay != old_close_delay) ||
> +            (ss->closing_wait != old_closing_wait)) {
> +		if (!capable(CAP_SYS_ADMIN))
>  			retval = -EPERM;
> -		else
> -			retval = -EOPNOTSUPP;
> -	} else {
> -		acm->port.close_delay  = close_delay;
> -		acm->port.closing_wait = closing_wait;
> -	}
> +		else {
> +			acm->port.close_delay  = close_delay;
> +			acm->port.closing_wait = closing_wait;
> +		}
> +	} else
> +		retval = -EOPNOTSUPP;
>  
>  	mutex_unlock(&acm->port.mutex);
>  	return retval;
> -- 
> 2.17.1
> 

This patch doesn't apply to my usb-linus branch at all.  What
tree/branch did you make it against?

thanks,

greg k-h
Anthony Mallet March 12, 2020, 9:15 a.m. UTC | #2 
On Thursday 12 Mar 2020, at 09:38, Greg KH wrote:
> This patch doesn't apply to my usb-linus branch at all.  What
> tree/branch did you make it against?

This is against master of https://github.com/torvalds/linux.git
Greg KH March 12, 2020, 9:35 a.m. UTC | #3 
On Thu, Mar 12, 2020 at 10:15:39AM +0100, Anthony Mallet wrote:
> On Thursday 12 Mar 2020, at 09:38, Greg KH wrote:
> > This patch doesn't apply to my usb-linus branch at all.  What
> > tree/branch did you make it against?
> 
> This is against master of https://github.com/torvalds/linux.git

Odd.  Can you rebase and resend?

Also, maybe try it against my usb-linus branch of usb.git on
git.kernel.org?

thanks,

greg k-h
Anthony Mallet March 12, 2020, 1:30 p.m. UTC | #4 
On Thursday 12 Mar 2020, at 10:35, Greg KH wrote:
> On Thu, Mar 12, 2020 at 10:15:39AM +0100, Anthony Mallet wrote:
> > On Thursday 12 Mar 2020, at 09:38, Greg KH wrote:
> > > This patch doesn't apply to my usb-linus branch at all.  What
> > > tree/branch did you make it against?
> >
> > This is against master of https://github.com/torvalds/linux.git
>
> Odd.  Can you rebase and resend?

I did that: the only difference is in the "@@ -914,19 +915,24 @@" of
the second chunk of the second patch. The "19" is wrong, it should
be 18. However, I've no idea how this could happen, since I used "git
format-patch" to generate the diff, and I did not edit the diff part
at all manually.

It might be an emacs and diff-mode glitch (or me mistyping something)
when I added the comments in the patch, although this never happend to
me before.

Anyway, resending v4.

Thanks for your time,
Cheers.
diff mbox series

Patch

diff --git a/drivers/usb/class/cdc-acm.c b/drivers/usb/class/cdc-acm.c
index da619176deca..a41a3d27016c 100644
--- a/drivers/usb/class/cdc-acm.c
+++ b/drivers/usb/class/cdc-acm.c
@@ -907,6 +907,7 @@  static int set_serial_info(struct tty_struct *tty, struct serial_struct *ss)
 {
 	struct acm *acm = tty->driver_data;
 	unsigned int closing_wait, close_delay;
+	unsigned int old_closing_wait, old_close_delay;
 	int retval = 0;
 
 	close_delay = msecs_to_jiffies(ss->close_delay * 10);
@@ -914,19 +915,24 @@  static int set_serial_info(struct tty_struct *tty, struct serial_struct *ss)
 			ASYNC_CLOSING_WAIT_NONE :
 			msecs_to_jiffies(ss->closing_wait * 10);
 
+	/* we must redo the rounding here, so that the values match */
+	old_close_delay	= jiffies_to_msecs(acm->port.close_delay) / 10;
+	old_closing_wait = acm->port.closing_wait == ASYNC_CLOSING_WAIT_NONE ?
+				ASYNC_CLOSING_WAIT_NONE :
+				jiffies_to_msecs(acm->port.closing_wait) / 10;
+
 	mutex_lock(&acm->port.mutex);
 
-	if (!capable(CAP_SYS_ADMIN)) {
-		if ((close_delay != acm->port.close_delay) ||
-		    (closing_wait != acm->port.closing_wait))
+	if ((ss->close_delay != old_close_delay) ||
+            (ss->closing_wait != old_closing_wait)) {
+		if (!capable(CAP_SYS_ADMIN))
 			retval = -EPERM;
-		else
-			retval = -EOPNOTSUPP;
-	} else {
-		acm->port.close_delay  = close_delay;
-		acm->port.closing_wait = closing_wait;
-	}
+		else {
+			acm->port.close_delay  = close_delay;
+			acm->port.closing_wait = closing_wait;
+		}
+	} else
+		retval = -EOPNOTSUPP;
 
 	mutex_unlock(&acm->port.mutex);
 	return retval;
--