Received: from [140.186.70.92] (port=39599 helo=eggs.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1Ovdu4-00062v-6j for qemu-devel@nongnu.org; Tue, 14 Sep 2010 18:23:05 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.69) (envelope-from ) id 1Ovdu1-0006A0-NW for qemu-devel@nongnu.org; Tue, 14 Sep 2010 18:23:03 -0400 Received: from mail-ew0-f45.google.com ([209.85.215.45]:54394) by eggs.gnu.org with esmtp (Exim 4.69) (envelope-from ) id 1Ovdu1-00069v-I7 for qemu-devel@nongnu.org; Tue, 14 Sep 2010 18:23:01 -0400 Received: by ewy27 with SMTP id 27so3888526ewy.4 for ; Tue, 14 Sep 2010 15:23:00 -0700 (PDT) Date: Wed, 15 Sep 2010 00:22:57 +0200 From: "Edgar E. Iglesias" Subject: Re: [Qemu-devel] [PATCH] [slirp] Make ARP replies at least 64 bytes long Message-ID: <20100914222257.GB16690@laped.lan> References: <1284411762-3687-1-git-send-email-hpoussin@reactos.org> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: <1284411762-3687-1-git-send-email-hpoussin@reactos.org> List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: =?iso-8859-1?Q?Herv=E9?= Poussineau Cc: qemu-devel@nongnu.org On Mon, Sep 13, 2010 at 11:02:42PM +0200, Hervé Poussineau wrote: > IEEE 802.3 standard requires Ethernet frames to be at least 64 bytes long. > If it is not the case, they will be considered as runt frames, and may be ignored by netcard and/or OS > > Signed-off-by: Hervé Poussineau > --- > slirp/slirp.c | 12 ++++++++---- > 1 files changed, 8 insertions(+), 4 deletions(-) > > diff --git a/slirp/slirp.c b/slirp/slirp.c > index 82fd9b4..2e8c017 100644 > --- a/slirp/slirp.c > +++ b/slirp/slirp.c > @@ -599,9 +599,12 @@ static void arp_input(Slirp *slirp, const uint8_t *pkt, int pkt_len) > { > struct ethhdr *eh = (struct ethhdr *)pkt; > struct arphdr *ah = (struct arphdr *)(pkt + ETH_HLEN); > - uint8_t arp_reply[ETH_HLEN + sizeof(struct arphdr)]; > - struct ethhdr *reh = (struct ethhdr *)arp_reply; > - struct arphdr *rah = (struct arphdr *)(arp_reply + ETH_HLEN); > + union { > + uint8_t data[ETH_HLEN + sizeof(struct arphdr)]; > + uint8_t payload[64]; /* Minimum Ethernet frame size */ > + } arp_reply; > + struct ethhdr *reh = (struct ethhdr *)arp_reply.data; > + struct arphdr *rah = (struct arphdr *)(arp_reply.data + ETH_HLEN); Hi, Would you mind explaning the point of the union here? Why not just do something like: - uint8_t arp_reply[ETH_HLEN + sizeof(struct arphdr)]; + uint8_t arp_reply[MAX(ETH_HLEN + sizeof(struct arphdr), 64)]; Cheers > int ar_op; > struct ex_list *ex_ptr; > > @@ -619,6 +622,7 @@ static void arp_input(Slirp *slirp, const uint8_t *pkt, int pkt_len) > } > return; > arp_ok: > + memset(&arp_reply, 0, sizeof(arp_reply)); > /* XXX: make an ARP request to have the client address */ > memcpy(slirp->client_ethaddr, eh->h_source, ETH_ALEN); > > @@ -637,7 +641,7 @@ static void arp_input(Slirp *slirp, const uint8_t *pkt, int pkt_len) > rah->ar_sip = ah->ar_tip; > memcpy(rah->ar_tha, ah->ar_sha, ETH_ALEN); > rah->ar_tip = ah->ar_sip; > - slirp_output(slirp->opaque, arp_reply, sizeof(arp_reply)); > + slirp_output(slirp->opaque, (const uint8_t *)&arp_reply, sizeof(arp_reply)); > } > break; > case ARPOP_REPLY: > -- > 1.7.1.GIT > >