From mboxrd@z Thu Jan 1 00:00:00 1970 From: Alexei Starovoitov Subject: Re: [PATCH net-next 5/5] s390/bpf: recache skb->data/hlen for skb_vlan_push/pop Date: Tue, 28 Jul 2015 10:19:28 -0700 Message-ID: <55B7B9A0.5060606@plumgrid.com> References: <1438092600-16221-1-git-send-email-holzheu@linux.vnet.ibm.com> <1438092600-16221-6-git-send-email-holzheu@linux.vnet.ibm.com> Mime-Version: 1.0 Content-Type: text/plain; charset=windows-1252; format=flowed Content-Transfer-Encoding: 7bit Cc: Martin Schwidefsky , Daniel Borkmann , netdev@vger.kernel.org, linux-s390@vger.kernel.org To: Michael Holzheu , David Miller Return-path: Received: from mail-pd0-f175.google.com ([209.85.192.175]:33346 "EHLO mail-pd0-f175.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752190AbbG1RTb (ORCPT ); Tue, 28 Jul 2015 13:19:31 -0400 Received: by pdbnt7 with SMTP id nt7so73854403pdb.0 for ; Tue, 28 Jul 2015 10:19:31 -0700 (PDT) In-Reply-To: <1438092600-16221-6-git-send-email-holzheu@linux.vnet.ibm.com> Sender: netdev-owner@vger.kernel.org List-ID: On 7/28/15 7:10 AM, Michael Holzheu wrote: > Allow eBPF programs attached to TC qdiscs call skb_vlan_push/pop > via helper functions. These functions may change skb->data/hlen. > This data is cached by s390 JIT to improve performance of ld_abs/ld_ind > instructions. Therefore after a change we have to reload the data. > > In case of usage of skb_vlan_push/pop, in the prologue we store > the SKB pointer on the stack and restore it after BPF_JMP_CALL > to skb_vlan_push/pop. > > Signed-off-by: Michael Holzheu Thanks! Acked-by: Alexei Starovoitov