From mboxrd@z Thu Jan 1 00:00:00 1970 From: Cong Wang Subject: Re: [Patch net 0/5] net_sched: tc action fixes and updates Date: Wed, 10 Aug 2016 13:06:30 -0700 Message-ID: References: <1470689209-17682-1-git-send-email-xiyou.wangcong@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Cc: Linux Kernel Network Developers , David Miller To: Jamal Hadi Salim Return-path: Received: from mail-io0-f173.google.com ([209.85.223.173]:33503 "EHLO mail-io0-f173.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S935183AbcHJUGw (ORCPT ); Wed, 10 Aug 2016 16:06:52 -0400 Received: by mail-io0-f173.google.com with SMTP id 38so52123470iol.0 for ; Wed, 10 Aug 2016 13:06:51 -0700 (PDT) In-Reply-To: Sender: netdev-owner@vger.kernel.org List-ID: On Wed, Aug 10, 2016 at 7:34 AM, Jamal Hadi Salim wrote: > On 16-08-08 04:46 PM, Cong Wang wrote: >> >> This patchset fixes several regressions caused by the previous >> code refactor. Thanks to Jamal for catching them! >> > > Cong, > > Good news: oops gone. I havent done more testing than I did > before; but looks good so far. > > Bad news: You have introduced a performance regression which is > noticeable at high speed. > > tcf_exts_exec() is the culprit - and conversion to from flexarray > to linked list in the fast problem to be specific. Ah, this reminds me that I don't have to use flex_array, initially I thought the tcf_exts could hold as many actions as it wants, but actually there is a upper bound, TCA_ACT_MAX_PRIO. IOW, a regular dynamic array is just enough here. I just replaced the flex_array with a regular one, it works fine for me too, at least no crash with all of my test cases. Please try v2, since you have more test cases that I do. Or it would be great if you can share your test cases with me or us. Be patient, every big change could have regression. :) Thanks.