From mboxrd@z Thu Jan 1 00:00:00 1970 From: Nathan Sullivan Subject: softirq behavior during a UDP flood Date: Fri, 1 May 2015 10:04:12 -0500 Message-ID: <20150501150412.GA15483@nathan3500-linux-VM> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii To: linux-rt-users@vger.kernel.org Return-path: Received: from skprod3.natinst.com ([130.164.80.24]:48135 "EHLO ni.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1751311AbbEAPEO (ORCPT ); Fri, 1 May 2015 11:04:14 -0400 Received: from us-aus-mgwout2.amer.corp.natinst.com (nb-snip2-1338.natinst.com [130.164.19.135]) by us-aus-skprod3.natinst.com (8.15.0.59/8.15.0.59) with ESMTP id t41F4Bcr008042 for ; Fri, 1 May 2015 10:04:12 -0500 Content-Disposition: inline Sender: linux-rt-users-owner@vger.kernel.org List-ID: Hello all, We are running 3.14.37-rt on a Xilinx Zynq based board, and have noticed some unfortunate behavior with NAPI polling during heavy incoming traffic. Since, as I understand it, softirqs are scheduled on the thread that caused them in rt, the netowrk RX softirq simply runs over and over on one CPU of the system. The network device never re-enables interupts, basically NAPI polling runs forever and weight/budget are irrelevant with preempt-rt on. Since we set IRQ affinity to CPU 0 for everything, this leads to the system live-locking and becoming unusable. With full RT preemption off, things are fine. In addition, 3.2 kernels with RT are fine as well under heavy net load. Is this behavior due to a design tradeoff, or is it a bug?