[PATCH v1 1/1] tty: Drop duplicate NULL check in TTY port functions

[PATCH v1 1/1] tty: Drop duplicate NULL check in TTY port functions

[Andy Shevchenko]

The free_page(addr), which becomes free_pages(addr, 0) checks addr against 0.
No need to repeat this check in the callers, i.e.  tty_port_free_xmit_buf()
and tty_port_destructor().

Note, INIT_KFIFO() is safe without that check, because it's aware of kfifo PTR
versus embedded kfifo.

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
---
 drivers/tty/tty_port.c | 11 ++++-------
 1 file changed, 4 insertions(+), 7 deletions(-)

diff --git a/drivers/tty/tty_port.c b/drivers/tty/tty_port.c
index 4282895ede9e..880608a65773 100644
--- a/drivers/tty/tty_port.c
+++ b/drivers/tty/tty_port.c
@@ -240,11 +240,9 @@ EXPORT_SYMBOL(tty_port_alloc_xmit_buf);
 void tty_port_free_xmit_buf(struct tty_port *port)
 {
 	mutex_lock(&port->buf_mutex);
-	if (port->xmit_buf != NULL) {
-		free_page((unsigned long)port->xmit_buf);
-		port->xmit_buf = NULL;
-		INIT_KFIFO(port->xmit_fifo);
-	}
+	free_page((unsigned long)port->xmit_buf);
+	port->xmit_buf = NULL;
+	INIT_KFIFO(port->xmit_fifo);
 	mutex_unlock(&port->buf_mutex);
 }
 EXPORT_SYMBOL(tty_port_free_xmit_buf);
@@ -271,8 +269,7 @@ static void tty_port_destructor(struct kref *kref)
 	/* check if last port ref was dropped before tty release */
 	if (WARN_ON(port->itty))
 		return;
-	if (port->xmit_buf)
-		free_page((unsigned long)port->xmit_buf);
+	free_page((unsigned long)port->xmit_buf);
 	tty_port_destroy(port);
 	if (port->ops && port->ops->destruct)
 		port->ops->destruct(port);
-- 
2.34.1

Re: [PATCH v1 1/1] tty: Drop duplicate NULL check in TTY port functions

[Jiri Slaby]

On 02. 02. 22, 17:57, Andy Shevchenko wrote:
> The free_page(addr), which becomes free_pages(addr, 0) checks addr against 0.
> No need to repeat this check in the callers, i.e.  tty_port_free_xmit_buf()
> and tty_port_destructor().
> 
> Note, INIT_KFIFO() is safe without that check, because it's aware of kfifo PTR
> versus embedded kfifo.

Not sure what you mean here ^^^? But it might be one of the morning 
brain parser errors.

Acked-by: Jiri Slaby <jirislaby@kernel.org>

> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> ---
>   drivers/tty/tty_port.c | 11 ++++-------
>   1 file changed, 4 insertions(+), 7 deletions(-)
> 
> diff --git a/drivers/tty/tty_port.c b/drivers/tty/tty_port.c
> index 4282895ede9e..880608a65773 100644
> --- a/drivers/tty/tty_port.c
> +++ b/drivers/tty/tty_port.c
> @@ -240,11 +240,9 @@ EXPORT_SYMBOL(tty_port_alloc_xmit_buf);
>   void tty_port_free_xmit_buf(struct tty_port *port)
>   {
>   	mutex_lock(&port->buf_mutex);
> -	if (port->xmit_buf != NULL) {
> -		free_page((unsigned long)port->xmit_buf);
> -		port->xmit_buf = NULL;
> -		INIT_KFIFO(port->xmit_fifo);
> -	}
> +	free_page((unsigned long)port->xmit_buf);
> +	port->xmit_buf = NULL;
> +	INIT_KFIFO(port->xmit_fifo);
>   	mutex_unlock(&port->buf_mutex);
>   }
>   EXPORT_SYMBOL(tty_port_free_xmit_buf);
> @@ -271,8 +269,7 @@ static void tty_port_destructor(struct kref *kref)
>   	/* check if last port ref was dropped before tty release */
>   	if (WARN_ON(port->itty))
>   		return;
> -	if (port->xmit_buf)
> -		free_page((unsigned long)port->xmit_buf);
> +	free_page((unsigned long)port->xmit_buf);
>   	tty_port_destroy(port);
>   	if (port->ops && port->ops->destruct)
>   		port->ops->destruct(port);


-- 
js
suse labs

Re: [PATCH v1 1/1] tty: Drop duplicate NULL check in TTY port functions

[Andy Shevchenko]

On Thu, Feb 03, 2022 at 09:36:55AM +0100, Jiri Slaby wrote:
> On 02. 02. 22, 17:57, Andy Shevchenko wrote:
> > The free_page(addr), which becomes free_pages(addr, 0) checks addr against 0.
> > No need to repeat this check in the callers, i.e.  tty_port_free_xmit_buf()
> > and tty_port_destructor().
> > 
> > Note, INIT_KFIFO() is safe without that check, because it's aware of kfifo PTR
> > versus embedded kfifo.
> 
> Not sure what you mean here ^^^? But it might be one of the morning brain
> parser errors.

Or maybe my evening weren't working...

Actually INIT_KFIFO() can be outside of that check from day 1 because it
operates on a separate member and does not rely on the FIFO itself to be
allocated.

I tried to explain that, while kfifo allocation goes together with buffer,
there is no dependency to any of those allocations.

-- 
With Best Regards,
Andy Shevchenko

Re: [PATCH v1 1/1] tty: Drop duplicate NULL check in TTY port functions

[Greg Kroah-Hartman]

On Thu, Feb 03, 2022 at 12:40:00PM +0200, Andy Shevchenko wrote:
> On Thu, Feb 03, 2022 at 09:36:55AM +0100, Jiri Slaby wrote:
> > On 02. 02. 22, 17:57, Andy Shevchenko wrote:
> > > The free_page(addr), which becomes free_pages(addr, 0) checks addr against 0.
> > > No need to repeat this check in the callers, i.e.  tty_port_free_xmit_buf()
> > > and tty_port_destructor().
> > > 
> > > Note, INIT_KFIFO() is safe without that check, because it's aware of kfifo PTR
> > > versus embedded kfifo.
> > 
> > Not sure what you mean here ^^^? But it might be one of the morning brain
> > parser errors.
> 
> Or maybe my evening weren't working...
> 
> Actually INIT_KFIFO() can be outside of that check from day 1 because it
> operates on a separate member and does not rely on the FIFO itself to be
> allocated.
> 
> I tried to explain that, while kfifo allocation goes together with buffer,
> there is no dependency to any of those allocations.

Can you rewrite the changelog to say this?

thanks,

greg k-h

Re: [PATCH v1 1/1] tty: Drop duplicate NULL check in TTY port functions

[Andy Shevchenko]

On Fri, Feb 04, 2022 at 03:42:15PM +0100, Greg Kroah-Hartman wrote:
> On Thu, Feb 03, 2022 at 12:40:00PM +0200, Andy Shevchenko wrote:
> > On Thu, Feb 03, 2022 at 09:36:55AM +0100, Jiri Slaby wrote:
> > > On 02. 02. 22, 17:57, Andy Shevchenko wrote:

...

> > > > Note, INIT_KFIFO() is safe without that check, because it's aware of kfifo PTR
> > > > versus embedded kfifo.
> > > 
> > > Not sure what you mean here ^^^? But it might be one of the morning brain
> > > parser errors.
> > 
> > Or maybe my evening weren't working...
> > 
> > Actually INIT_KFIFO() can be outside of that check from day 1 because it
> > operates on a separate member and does not rely on the FIFO itself to be
> > allocated.
> > 
> > I tried to explain that, while kfifo allocation goes together with buffer,
> > there is no dependency to any of those allocations.
> 
> Can you rewrite the changelog to say this?

Sure!

-- 
With Best Regards,
Andy Shevchenko