From mboxrd@z Thu Jan 1 00:00:00 1970 From: Richard Cochran Subject: Re: PTP + H/W time stamping + CONFIG_PREEMPT_RT_FULL == Oops in slab Date: Thu, 21 Jun 2012 21:54:28 +0200 Message-ID: <20120621195428.GC13496@netboy.at.omicron.at> References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: netdev@vger.kernel.org To: satpal parmar Return-path: Received: from mail-bk0-f46.google.com ([209.85.214.46]:34771 "EHLO mail-bk0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1756405Ab2FUTyj (ORCPT ); Thu, 21 Jun 2012 15:54:39 -0400 Received: by bkcji2 with SMTP id ji2so964945bkc.19 for ; Thu, 21 Jun 2012 12:54:38 -0700 (PDT) Content-Disposition: inline In-Reply-To: Sender: netdev-owner@vger.kernel.org List-ID: I suspect this isn't a time stamping bug, but rather that the time stamping code just triggers the issue. My own experience has been that preempt_rt on arm is quite shaky. On Thu, Jun 21, 2012 at 08:09:11PM +0530, satpal parmar wrote: > I am not sure if this is the right mailing list. If anyone know better > forum please direct me to that. Appreciate your patience and time. Maybe lkml or rt-users would be better. However, there are really too many variables, and I recommend first narrowing the problem down. Here are a few ideas. 1. use a mainline kernel plus rt patch (without vendor/custom changes) 2. try to trigger issue without using ptpd2 (like stress testing) 3. write a simple test case (like sending a single PTP sync) and instrument your kernel Good luck, Richard