[PATCH v2 1/1] Char: tty, fix tty_port_block_til_ready waiting

[PATCH v2 1/1] Char: tty, fix tty_port_block_til_ready waiting

[Jiri Slaby]

Use DEFINE_WAIT instead DECLARE_WAITQUEUE, because we use
prepare/finish_wait since commit
3e3b5c087799e536871c8261b05bc28e4783c8da
(tty: use prepare/finish_wait)

Otherwise we would oops in finish_wait.

Signed-off-by: Jiri Slaby <jirislaby@gmail.com>
---
 drivers/char/tty_port.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/drivers/char/tty_port.c b/drivers/char/tty_port.c
index 62dadfc..4e862a7 100644
--- a/drivers/char/tty_port.c
+++ b/drivers/char/tty_port.c
@@ -193,7 +193,7 @@ int tty_port_block_til_ready(struct tty_port *port,
 {
 	int do_clocal = 0, retval;
 	unsigned long flags;
-	DECLARE_WAITQUEUE(wait, current);
+	DEFINE_WAIT(wait);
 	int cd;
 
 	/* block if port is in the process of being closed */
-- 
1.6.3.2

Re: [PATCH v2 1/1] Char: tty, fix tty_port_block_til_ready waiting

[Andrew Morton]

On Thu, 18 Jun 2009 09:46:13 +0200
Jiri Slaby <jirislaby@gmail.com> wrote:

> Use DEFINE_WAIT instead DECLARE_WAITQUEUE, because we use
> prepare/finish_wait since commit
> 3e3b5c087799e536871c8261b05bc28e4783c8da
> (tty: use prepare/finish_wait)
> 
> Otherwise we would oops in finish_wait.
> 

<scratches head for a while>

OK, here's a better changelog:

   Since commit 3e3b5c087799e536871c8261b05bc28e4783c8da ("tty: use
   prepare/finish_wait"), tty_port_block_til_ready() is using
   prepare_to_wait()/finish_wait().  Those functions require that
   the wait_queue_t be initialised with .func=autoremove_wake_function,
   via DEFINE_WAIT().

   But the conversion from DECLARE_WAITQUEUE() to DEFINE_WAIT() was
   not made, so this code will oops in finish_wait().

> 
> diff --git a/drivers/char/tty_port.c b/drivers/char/tty_port.c
> index 62dadfc..4e862a7 100644
> --- a/drivers/char/tty_port.c
> +++ b/drivers/char/tty_port.c
> @@ -193,7 +193,7 @@ int tty_port_block_til_ready(struct tty_port *port,
>  {
>  	int do_clocal = 0, retval;
>  	unsigned long flags;
> -	DECLARE_WAITQUEUE(wait, current);
> +	DEFINE_WAIT(wait);
>  	int cd;
>  
>  	/* block if port is in the process of being closed */

Why is nobody reporting the oopses?

Re: [PATCH v2 1/1] Char: tty, fix tty_port_block_til_ready waiting

[Jiri Slaby]

On 06/24/2009 01:30 AM, Andrew Morton wrote:
> On Thu, 18 Jun 2009 09:46:13 +0200
>> --- a/drivers/char/tty_port.c
>> +++ b/drivers/char/tty_port.c
>> @@ -193,7 +193,7 @@ int tty_port_block_til_ready(struct tty_port *port,
>>  {
>>  	int do_clocal = 0, retval;
>>  	unsigned long flags;
>> -	DECLARE_WAITQUEUE(wait, current);
>> +	DEFINE_WAIT(wait);
>>  	int cd;
>>  
>>  	/* block if port is in the process of being closed */
> 
> Why is nobody reporting the oopses?

Myaybe because only few drivers use that function. I hit it the first
time I tested cyclades patchset.