RE: [PATCH 3/4] sound/soc/mxs/mxs-saif.c: add missing kfree

[Julia Lawall <julia@diku.dk>]

On Sun, 21 Aug 2011, Dong Aisheng-B29396 wrote:

> > -----Original Message-----
> > From: Julia Lawall [mailto:julia@diku.dk]
> > Sent: Saturday, August 20, 2011 11:19 PM
> > To: Wolfram Sang
> > Cc: Liam Girdwood; kernel-janitors@vger.kernel.org; Mark Brown; Jaroslav
> > Kysela; Takashi Iwai; Dong Aisheng-B29396; alsa-devel@alsa-project.org;
> > linux-kernel@vger.kernel.org
> > Subject: Re: [PATCH 3/4] sound/soc/mxs/mxs-saif.c: add missing kfree
> > 
> > From: Julia Lawall <julia@diku.dk>
> > 
> > Move the test on pdev->id before the kzalloc to avoid requiring kfree
> > when the test fails.  This fix was suggested by Wolfram Sang.
> > 
> > A simplified version of the semantic match that finds this problem is as
> > follows: (http://coccinelle.lip6.fr/)
> > 
> > // <smpl>
> > @@
> > identifier x;
> > expression E1!=0,E2,E3,E4;
> > statement S;
> > iterator I;
> > @@
> > 
> > (
> > if (...) { ... when != kfree(x)
> >                when != x = E3
> >                when != E3 = x
> > *  return ...;
> >  }
> > ... when != x = E2
> >     when != I(...,x,...) S
> > if (...) { ... when != x = E4
> >  kfree(x); ... return ...; }
> > )
> > // </smpl>
> > 
> > Signed-off-by: Julia Lawall <julia@diku.dk>
> > 
> > ---
> > diff --git a/sound/soc/mxs/mxs-saif.c b/sound/soc/mxs/mxs-saif.c index
> > 0b3adae..e9a90e4 100644
> > --- a/sound/soc/mxs/mxs-saif.c
> > +++ b/sound/soc/mxs/mxs-saif.c
> > @@ -519,12 +519,11 @@ static int mxs_saif_probe(struct platform_device
> > *pdev)
> >  	struct mxs_saif *saif;
> >  	int ret = 0;
> > 
> > +	if (pdev->id >= ARRAY_SIZE(mxs_saif))
> > +		return -EINVAL;
> >  	saif = kzalloc(sizeof(*saif), GFP_KERNEL);
> >  	if (!saif)
> >  		return -ENOMEM;
> > -
> > -	if (pdev->id >= ARRAY_SIZE(mxs_saif))
> > -		return -EINVAL;
> >  	mxs_saif[pdev->id] = saif;
> > 
> >  	saif->clk = clk_get(&pdev->dev, NULL);
> 
> Thanks for the finding, Julia.
> It's ok to me.
> 
> BTW I guess you may need to resend a new version patch instead of following it
> in this mail thread. :)
> If yes, one minus thing, add one blank line below the line of 'return -ENOMEM'
> may look more comfortable.

OK, I will do both.

Thanks,
julia