From: Szymon Janc <szymon.janc@gmail.com>
To: Marcel Holtmann <marcel@holtmann.org>
Cc: Marcin Kraglak <marcin.kraglak@tieto.com>,
linux-bluetooth@vger.kernel.org
Subject: Re: [PATCH] shared/gatt: Fix freeing uninitialized attributes
Date: Fri, 16 May 2014 18:24:22 +0200 [thread overview]
Message-ID: <1987880.BnWgK9kt0s@athlon> (raw)
In-Reply-To: <9A0C0D87-616B-4D33-A61F-85E5F954B77A@holtmann.org>
Hi Marcel,
On Friday 16 May 2014 08:17:25 Marcel Holtmann wrote:
> Hi Marcin,
>
> > Don't free uninitialized attributes. It could cause NULL pointer
> > dereference.
> > ---
> > src/shared/gatt-db.c | 4 ++++
> > 1 file changed, 4 insertions(+)
> >
> > diff --git a/src/shared/gatt-db.c b/src/shared/gatt-db.c
> > index d9f63be..36316af 100644
> > --- a/src/shared/gatt-db.c
> > +++ b/src/shared/gatt-db.c
> > @@ -99,6 +99,10 @@ static void attribute_destroy(void *data)
> > {
> >
> > struct gatt_db_attribute *attribute = data;
> >
> > + /* Attribute was not initialized by user */
> > + if (!attribute)
> > + return;
> > +
> >
> > free(attribute->value);
> > free(attribute);
>
> I think the real question is why a function with void *data is called with a
> NULL pointer in the first place. That is the thing you should fix and not
> hack around this by doing a NULL check. So this is clearly a function that
> you gave to callback somehow, fix that part. I do not remember any of the
> code in src/shared/ that has the destroy callbacks requiring any of this.
> We are expecting that data is valid.
This is due to attributes[] in struct gatt_db_service might not be fully used
by user (we preallocate handles). But since attribute_destroy() is used only
from gatt_db_service_destroy() maybe it would be better to check if pointer is
not NULL there since checking this inside gatt_db_service_destroy() would be
easier to follow.
Something like this maybe: ?
-static void attribute_destroy(void *data)
+static void attribute_destroy(struct gatt_db_attribute *attribute)
{
- struct gatt_db_attribute *attribute = data;
-
- /* Attribute was not initialized by user */
- if (!attribute)
- return;
-
free(attribute->value);
free(attribute);
}
@@ -131,8 +125,13 @@ static void gatt_db_service_destroy(void *data)
struct gatt_db_service *service = data;
int i;
- for (i = 0; i < service->num_handles; i++)
+ for (i = 0; i < service->num_handles; i++) {
+ /* Attribute was not initialized by user */
+ if (!service->attributes[i])
+ continue;
+
attribute_destroy(service->attributes[i]);
+ }
--
Szymon K. Janc
szymon.janc@gmail.com
prev parent reply other threads:[~2014-05-16 16:24 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-05-16 10:07 [PATCH] shared/gatt: Fix freeing uninitialized attributes Marcin Kraglak
2014-05-16 12:36 ` Szymon Janc
2014-05-16 15:17 ` Marcel Holtmann
2014-05-16 16:24 ` Szymon Janc [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1987880.BnWgK9kt0s@athlon \
--to=szymon.janc@gmail.com \
--cc=linux-bluetooth@vger.kernel.org \
--cc=marcel@holtmann.org \
--cc=marcin.kraglak@tieto.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox