On Wed, Jul 23, 2008 at 10:00:05PM -0700, Harvey Harrison wrote: > On Thu, 2008-07-24 at 14:36 +1000, Stephen Rothwell wrote: > > On Wed, 23 Jul 2008 18:34:08 -0700 Harvey Harrison <harvey.harrison@xxxxxxxxx> wrote: > > > > > > +++ b/drivers/pcmcia/rsrc_nonstatic.c > > > @@ -275,7 +275,7 @@ static int readable(struct pcmcia_socket *s, struct resource *res, > > > destroy_cis_cache(s); > > > } > > > s->cis_mem.res = NULL; > > > - if ((ret != 0) || (count == 0)) > > > + if ((ret != 0) || (count == NULL)) > > > > I was wondering if it should be > > (*count == 0) > > > > Actually, it looks that way as in this case count can never be none. > > readable() is only called in one place where it is passed the addresses > of two local variables. Looking at pccard_validate_cis(), the number > of valid tuples found is returned through the info pointer, or zero > if invalid cis is found. > > So I'd say *count == 0 is probably right. > > Dominik? Yes, *count == 0 is what we'd need to check here. Thanks! Dominik From: Dominik Brodowski <linux@xxxxxxxxxxxxxxxxxxxx> Date: Mon, 28 Jul 2008 16:37:10 +0200 Subject: [PATCH 1/1] pcmcia: rsrc_nonstatic: check value, not pointer Bug found by Harvey Harrison and Stephen Rothwell. Signed-off-by: Dominik Brodowski <linux@xxxxxxxxxxxxxxxxxxxx> --- drivers/pcmcia/rsrc_nonstatic.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/drivers/pcmcia/rsrc_nonstatic.c b/drivers/pcmcia/rsrc_nonstatic.c index d0c1d63..203e579 100644 --- a/drivers/pcmcia/rsrc_nonstatic.c +++ b/drivers/pcmcia/rsrc_nonstatic.c @@ -275,7 +275,7 @@ static int readable(struct pcmcia_socket *s, struct resource *res, destroy_cis_cache(s); } s->cis_mem.res = NULL; - if ((ret != 0) || (count == 0)) + if ((ret != 0) || (*count == 0)) return 0; return 1; } -- 1.5.4.3 -- To unsubscribe from this list: send the line "unsubscribe linux-next" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
