From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: Re: [PATCH] bnx2x: Fix mem leak in bnx2x_tpa_stop() if build_skb() fails. Date: Mon, 06 Feb 2012 23:53:53 +0100 Message-ID: <1328568833.2476.1.camel@edumazet-laptop> References: Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: netdev@vger.kernel.org, Eilon Greenstein , linux-kernel@vger.kernel.org, David Miller , Eliezer Tamir To: Jesper Juhl Return-path: In-Reply-To: Sender: linux-kernel-owner@vger.kernel.org List-Id: netdev.vger.kernel.org Le lundi 06 f=C3=A9vrier 2012 =C3=A0 22:28 +0100, Jesper Juhl a =C3=A9c= rit : > We allocate memory for 'new_data' with kmalloc(). If we get the memor= y > we then try to build_skb() and if that should fail (which it can) we > do not enter 'if (likely(skb)) {' and actually use 'new_data' but > instead fall through to the 'drop:' label and end up returning from > the function without ever assigning 'new'data' to anything or freeing > it. That leaks the memory allocated to 'new_data'. >=20 > This patch fixes the memory leak by doing a kfree(new_data) in the > case where build_skb() fails (or where allocation of 'new_data' itsel= f > fails, but in taht case it's just a harmless kfree(NULL)). >=20 > Signed-off-by: Jesper Juhl > --- > drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c | 3 +-- > 1 files changed, 1 insertions(+), 2 deletions(-) >=20 > No hardware to test, so compile tested only. >=20 > diff --git a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c b/driver= s/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c > index 03f3935..7aee469 100644 > --- a/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c > +++ b/drivers/net/ethernet/broadcom/bnx2x/bnx2x_cmn.c > @@ -523,7 +523,6 @@ static void bnx2x_tpa_stop(struct bnx2x *bp, stru= ct bnx2x_fastpath *fp, > skb =3D build_skb(data); > =20 > if (likely(skb)) { > - > #ifdef BNX2X_STOP_ON_ERROR > if (pad + len > fp->rx_buf_size) { > BNX2X_ERR("skb_put is about to fail... " > @@ -557,7 +556,7 @@ static void bnx2x_tpa_stop(struct bnx2x *bp, stru= ct bnx2x_fastpath *fp, > =20 > return; > } > - > + kfree(new_data); > drop: > /* drop the packet and keep the buffer in the bin */ > DP(NETIF_MSG_RX_STATUS, > --=20 > 1.7.9 >=20 >=20 Good catch, my bad. Thanks Acked-by: Eric Dumazet