From: Nicolas Schichan Subject: Re: [PATCH v2 net-next 3/3] ARM: net: bpf_jit_32: support BPF_S_ANC_SECCOMP_LD_W instruction Date: Mon, 29 Apr 2013 14:39:29 +0200 Message-ID: <517E6A01.4040407@freebox.fr> References: <1367029047-14830-1-git-send-email-xi.wang@gmail.com> <1367029047-14830-4-git-send-email-xi.wang@gmail.com> <517B6FC5.7070003@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: Daniel Borkmann , netdev@vger.kernel.org, linux-kernel@vger.kernel.org, Heiko Carstens , Will Drewry , Eric Dumazet , Russell King , David Laight , "David S. Miller" , Andrew Morton To: Xi Wang Return-path: In-Reply-To: Sender: linux-kernel-owner@vger.kernel.org List-Id: netdev.vger.kernel.org On 04/27/2013 08:32 PM, Xi Wang wrote: > On Sat, Apr 27, 2013 at 2:27 AM, Daniel Borkmann wrote: >> Arent't you doing here a similar thing in terms of getting arch as Eric >> criticized (Nicolas' implementation does not use that part btw.)? Also, >> even if it would be possible here, now your 2 JIT implementations differ >> in behaviour. I think this is unintended. > > Eric's comment was about x86, where the audit arch could change on the > fly. For ARM, the audit arch doesn't change---syscall_get_arch() > always returns AUDIT_ARCH_ARM. Hi, Indeed, syscall_get_arch() will only return AUDIT_ARCH_ARM on ARM right now. This might be more future proof to call syscall_get_arch() though. The main reason that comes to my mind would be an AArch64 kernel with support for AArch32 userland tasks. This would I expect require a different AUDIT_ARCH constant to differenciate between AArch64 and AArch32 tasks. Regards, -- Nicolas Schichan Freebox SAS