On Mon, Jun 25, 2012 at 07:15:46PM +0200, Sasha Levin wrote: > On Mon, 2012-06-25 at 19:17 +0200, Samuel Ortiz wrote: > > Hi Sasha, > > > > On Tue, Jun 12, 2012 at 10:08:19PM +0200, Sasha Levin wrote: > > > Not having 'local' is a valid case when a socket was created but never > > > bound or connected to anything, so avoid putting 'local' if it was > > > never created. > > > > > > Signed-off-by: Sasha Levin <levinsasha928@xxxxxxxxx> > > > --- > > > net/nfc/llcp/sock.c | 3 ++- > > > 1 files changed, 2 insertions(+), 1 deletions(-) > > > > > > diff --git a/net/nfc/llcp/sock.c b/net/nfc/llcp/sock.c > > > index 2c0b317..54daa10 100644 > > > --- a/net/nfc/llcp/sock.c > > > +++ b/net/nfc/llcp/sock.c > > > @@ -710,7 +710,8 @@ void nfc_llcp_sock_free(struct nfc_llcp_sock *sock) > > > > > > sock->parent = NULL; > > > > > > - nfc_llcp_local_put(sock->local); > > > + if (sock->local) > > > + nfc_llcp_local_put(sock->local); > > nfc_llcp_local_put() already checks for its argument being NULL or not. > > nfc_llcp_local_put() triggers a warning in this case as well, which > means that this code path shouldn't be happening. > > Should we remove the WARN_ON from nfc_llcp_local_put() instead? Yes, that would be better. Cheers, Samuel. -- Intel Open Source Technology Centre http://oss.intel.com/ -- To unsubscribe from this list: send the line "unsubscribe linux-wireless" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
