On Mon, Jan 27, 2014 at 09:58:03AM +0000, Lee Jones wrote:
> On Fri, 24 Jan 2014, Thierry Reding wrote:
> 
> > On Thu, Jan 23, 2014 at 10:32:20PM +0100, Christian Engelmayer wrote:
> > > Fix a memory leak in the lp3943_pwm_request_map() error handling path.
> > > Make sure already allocated pwm map memory is freed correctly.
> > > Detected by Coverity: CID 1162829.
> > > 
> > > Signed-off-by: Christian Engelmayer <[email protected]>
> > > ---
> > >  drivers/pwm/pwm-lp3943.c | 4 +++-
> > >  1 file changed, 3 insertions(+), 1 deletion(-)
> > > 
> > > diff --git a/drivers/pwm/pwm-lp3943.c b/drivers/pwm/pwm-lp3943.c
> > > index 8a843a0..a40b9c3 100644
> > > --- a/drivers/pwm/pwm-lp3943.c
> > > +++ b/drivers/pwm/pwm-lp3943.c
> > > @@ -52,8 +52,10 @@ lp3943_pwm_request_map(struct lp3943_pwm *lp3943_pwm, 
> > > int hwpwm)
> > >           offset = pwm_map->output[i];
> > >  
> > >           /* Return an error if the pin is already assigned */
> > > -         if (test_and_set_bit(offset, &lp3943->pin_used))
> > > +         if (test_and_set_bit(offset, &lp3943->pin_used)) {
> > > +                 kfree(pwm_map);
> > >                   return ERR_PTR(-EBUSY);
> > > +         }
> > >   }
> > >  
> > >   return pwm_map;
> > 
> > Lee, since you took this driver via the MFD tree, would you mind picking
> > up this fix as well?
> 
> Everything I have is now in Mainline.
> 
> It's best for you to apply it now I think?

Sure, I could do that after -rc1 is released. It's somewhat problematic
because I can't really create a tree that's based on yours before -rc1.
After -rc1 it shouldn't be a problem and this also isn't all that
critical, so sending off another pull request to Linus after -rc1 is
probably what I'll do.

Thierry

Attachment: pgpoQT345KurR.pgp
Description: PGP signature

Reply via email to