From mboxrd@z Thu Jan 1 00:00:00 1970 From: Li Yu Subject: [PATCH 0/3] skbtrace v2: A trace infrastructure for networking subsystem Date: Fri, 19 Oct 2012 14:16:04 +0800 Message-ID: <5080F024.30608@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=GB2312 Content-Transfer-Encoding: 7bit To: Linux Netdev List Return-path: Received: from mail-pa0-f46.google.com ([209.85.220.46]:56761 "EHLO mail-pa0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753263Ab2JSGQO (ORCPT ); Fri, 19 Oct 2012 02:16:14 -0400 Received: by mail-pa0-f46.google.com with SMTP id hz1so121597pad.19 for ; Thu, 18 Oct 2012 23:16:13 -0700 (PDT) Sender: netdev-owner@vger.kernel.org List-ID: Hi, This patch set introduces to the tracing infrastructure for networking subsystem, they are based on v3.6 source code. The changes from last version (v1): 1. IPv4 mapped IPv6 sockets support, I found many applications use this mean to live on double stack environment. 2. The BPF filter for connection based events, skbtrace constructs a temporarily sk_buff to run BPF program. 3. Add three TCP trace events for RTTM, show active connection information, TCP internal timers. and show more states in tcp_ca_state event. 4. Introduce to primary and secondary event support. That is secondary event only would be exported after its primary event is triggered. This could be used for reduce gathered trace points more, we can see this is another kind of trace point filter. 5. Userland utility modularizaion. 6. Code refactoring and bug fixes. I am going to add some sk_buff based events in next version (v3), so we will could know such like how a sk_buff pass stack, how long it stay in stack, how TSO/GSO/SACK/... split/merge them, and so on. You can get whole kernel source tree at github: https://github.com/Rover-Yu/skbtrace-userland https://github.com/Rover-Yu/skbtrace-kernel Welcome for suggestions. Thanks. Yu