From mboxrd@z Thu Jan 1 00:00:00 1970 From: Cong Wang Subject: Re: [Patch net-next v2 0/5] fou: some fixes and updates Date: Fri, 10 Apr 2015 13:50:55 -0700 Message-ID: References: <1428692430-14465-1-git-send-email-xiyou.wangcong@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Cc: Cong Wang , Linux Kernel Network Developers To: Tom Herbert Return-path: Received: from mail-wi0-f173.google.com ([209.85.212.173]:37460 "EHLO mail-wi0-f173.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753864AbbDJUu4 (ORCPT ); Fri, 10 Apr 2015 16:50:56 -0400 Received: by wiaa2 with SMTP id a2so10286772wia.0 for ; Fri, 10 Apr 2015 13:50:55 -0700 (PDT) In-Reply-To: Sender: netdev-owner@vger.kernel.org List-ID: On Fri, Apr 10, 2015 at 1:38 PM, Tom Herbert wrote: > On Fri, Apr 10, 2015 at 12:00 PM, Cong Wang wrote: >> Patch 1~3 fix some minor bugs in net/ipv4/fou.c, the only >> thing I am not sure is if it's too late to change the >> byte order of FOU_ATTR_PORT, if so we have to fix iproute2 >> instead of kernel. >> > Is this a relic from previous patch set, or is there a reason you want > to make this hbo? I was referring to this change: - u16 port = nla_get_u16(info->attrs[FOU_ATTR_PORT]); + __be16 port = nla_get_be16(info->attrs[FOU_ATTR_PORT]); I meant I am not sure if this breaks ABI here even in v2, we just use the raw value passed from user-space, so probably we are safe. > >> Patch 4~5 add some new features to make it complete. >> >> v2: make fou->port be16 too >> >> Cong Wang (5): >> fou: avoid calling udp_del_offload() twice >> fou: exit early when parsing config fails >> fou: always use be16 for port >> fou: add network namespace support >> fou: implement FOU_CMD_GET >> > > I assume you didn't see any adverse performance impact with these patches? > I don't think I touch any hot path, foo_create() and foo_destroy() are both slow path. Thanks.