[PATCH] cs5535 mfgpt - fix wrong if condition

[PATCH] cs5535 mfgpt - fix wrong if condition

[Christian Gmeiner]

From: Christian Gmeiner <christian.gmeiner@gmail.com>

This patch fixes the wrong if condition for the check if the
requested timer is available.

The bitmap avail is used to store if a timer is used already.
test_bit() is used to check if the requested timer is
available. If a bit in the avail bitmap is set it means that the
timer is available.

Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
---
 drivers/misc/cs5535-mfgpt.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/drivers/misc/cs5535-mfgpt.c b/drivers/misc/cs5535-mfgpt.c
index e01e08c..bc685bf 100644
--- a/drivers/misc/cs5535-mfgpt.c
+++ b/drivers/misc/cs5535-mfgpt.c
@@ -174,7 +174,7 @@ struct cs5535_mfgpt_timer
*cs5535_mfgpt_alloc_timer(int timer_nr, int domain)
                timer_nr = t < max ? (int) t : -1;
        } else {
                /* check if the requested timer's available */
-               if (test_bit(timer_nr, mfgpt->avail))
+               if (!test_bit(timer_nr, mfgpt->avail))
                        timer_nr = -1;
        }

-- 
1.7.5.rc3

--
Christian Gmeiner, MSc

Re: [PATCH] cs5535 mfgpt - fix wrong if condition

[Andrew Morton]

On Wed, 25 May 2011 19:08:06 +0000
Christian Gmeiner <christian.gmeiner@gmail.com> wrote:

> From: Christian Gmeiner <christian.gmeiner@gmail.com>
> 
> This patch fixes the wrong if condition for the check if the
> requested timer is available.
> 
> The bitmap avail is used to store if a timer is used already.
> test_bit() is used to check if the requested timer is
> available. If a bit in the avail bitmap is set it means that the
> timer is available.
> 
> Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
> ---
>  drivers/misc/cs5535-mfgpt.c |    2 +-
>  1 files changed, 1 insertions(+), 1 deletions(-)
> 
> diff --git a/drivers/misc/cs5535-mfgpt.c b/drivers/misc/cs5535-mfgpt.c
> index e01e08c..bc685bf 100644
> --- a/drivers/misc/cs5535-mfgpt.c
> +++ b/drivers/misc/cs5535-mfgpt.c
> @@ -174,7 +174,7 @@ struct cs5535_mfgpt_timer
> *cs5535_mfgpt_alloc_timer(int timer_nr, int domain)
>                 timer_nr = t < max ? (int) t : -1;
>         } else {
>                 /* check if the requested timer's available */
> -               if (test_bit(timer_nr, mfgpt->avail))
> +               if (!test_bit(timer_nr, mfgpt->avail))
>                         timer_nr = -1;
>         }
> 

What were the runtime effects of this bug?

(Please always include this information in the changelog when fixing things)

Thanks.

Re: [PATCH] cs5535 mfgpt - fix wrong if condition

[Andres Salomon]

You're right, it's reversed.

Acked-by: Andres Salomon <dilinger@queued.net>



On Fri, 3 Jun 2011 16:14:15 -0700
Andrew Morton <akpm@linux-foundation.org> wrote:

> On Wed, 25 May 2011 19:08:06 +0000
> Christian Gmeiner <christian.gmeiner@gmail.com> wrote:
> 
> > From: Christian Gmeiner <christian.gmeiner@gmail.com>
> > 
> > This patch fixes the wrong if condition for the check if the
> > requested timer is available.
> > 
> > The bitmap avail is used to store if a timer is used already.
> > test_bit() is used to check if the requested timer is
> > available. If a bit in the avail bitmap is set it means that the
> > timer is available.
> > 
> > Signed-off-by: Christian Gmeiner <christian.gmeiner@gmail.com>
> > ---
> >  drivers/misc/cs5535-mfgpt.c |    2 +-
> >  1 files changed, 1 insertions(+), 1 deletions(-)
> > 
> > diff --git a/drivers/misc/cs5535-mfgpt.c
> > b/drivers/misc/cs5535-mfgpt.c index e01e08c..bc685bf 100644
> > --- a/drivers/misc/cs5535-mfgpt.c
> > +++ b/drivers/misc/cs5535-mfgpt.c
> > @@ -174,7 +174,7 @@ struct cs5535_mfgpt_timer
> > *cs5535_mfgpt_alloc_timer(int timer_nr, int domain)
> >                 timer_nr = t < max ? (int) t : -1;
> >         } else {
> >                 /* check if the requested timer's available */
> > -               if (test_bit(timer_nr, mfgpt->avail))
> > +               if (!test_bit(timer_nr, mfgpt->avail))
> >                         timer_nr = -1;
> >         }
> > 
> 
> What were the runtime effects of this bug?
> 

The runtime effect would be that allocating a specific timer always
fails (versus telling cs5535_mfgpt_alloc_timer to allocate the first
available timer, which works).


> (Please always include this information in the changelog when fixing
> things)
> 
> Thanks.