From mboxrd@z Thu Jan 1 00:00:00 1970 From: Arnaldo Carvalho de Melo Subject: Re: [PATCH 2/2] [SKBUFF] introduce tr_hdr(skb) Date: Wed, 06 Oct 2004 16:18:49 -0300 Sender: netdev-bounce@oss.sgi.com Message-ID: <41644519.6040405@conectiva.com.br> References: <41643C40.8070605@conectiva.com.br> <20041006185835.GB19714@rei.reeler.org> <4164424E.6000504@conectiva.com.br> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: Thomas Graf , "David S. Miller" , netdev@oss.sgi.com Return-path: To: Arnaldo Carvalho de Melo In-Reply-To: <4164424E.6000504@conectiva.com.br> Errors-to: netdev-bounce@oss.sgi.com List-Id: netdev.vger.kernel.org Arnaldo Carvalho de Melo wrote: > > > Thomas Graf wrote: > >> * Arnaldo Carvalho de Melo <41643C40.8070605@conectiva.com.br> >> 2004-10-06 15:41 >> >>> - trh = (struct trh_hdr *)skb_push(skb, sizeof(*trh)); >>> + skb->mac.raw = skb_push(skb, sizeof(*trh)); >>> + trh = tr_hdr(skb); >>> trh->ac = AC; >>> trh->fc = LLC_FRAME; >>> if (sa) >>> @@ -51,7 +52,6 @@ >>> memcpy(trh->daddr, da, dev->addr_len); >>> tr_source_route(skb, trh, dev); >>> } >>> - skb->mac.raw = skb->data; >> >> >> >> Are you sure the bevhaviour is the same? There is an skb_pull >> in tr_source_route so skb->data gets modified. >> > > > Thanks for reviewing, checking... > > You're right, if tr_source_route is called, skb->mac.raw has to be set to skb->data, to maintain the old behaviour, if it is right? Oh well, as I said there aren't many people using Token Ring, let alone LLC over token ring, I'll be posting a patch in some minutes. - Arnaldo