From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752721AbaBGOqf (ORCPT ); Fri, 7 Feb 2014 09:46:35 -0500 Received: from moutng.kundenserver.de ([212.227.126.186]:54070 "EHLO moutng.kundenserver.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751901AbaBGOqd (ORCPT ); Fri, 7 Feb 2014 09:46:33 -0500 Message-ID: <1391784386.5475.103.camel@marge.simpson.net> Subject: Re: [ANNOUNCE] 3.12.9-rt13 From: Mike Galbraith To: Sebastian Andrzej Siewior Cc: linux-rt-users , LKML , Thomas Gleixner , rostedt@goodmis.org, John Kacur Date: Fri, 07 Feb 2014 15:46:26 +0100 In-Reply-To: <20140207085701.GB23668@linutronix.de> References: <20140203194841.GA17108@linutronix.de> <1391526218.5532.33.camel@marge.simpson.net> <20140207085701.GB23668@linutronix.de> Content-Type: text/plain; charset="UTF-8" X-Mailer: Evolution 3.2.3 Content-Transfer-Encoding: 7bit Mime-Version: 1.0 X-Provags-ID: V02:K0:esQ+a7OVkupeNtosEVL5GZJpdIEdg8EVgOuaOEmsWDV Exa8FTw4Sdyop2lSh8WsOIqtTcQrG92PVEuZOOfwylWTMkzH8r ao7vboTY5tcd4vA6MT1DaoI8RnUPRV88wFGdpbtXY2xim1hyBG gUcrNKTDxNRvGA4JaBAWkmtmqs/Howcxbb1U2Myw1il7VRlMsx n8d4Yz/XuCiI+2zdYmfl5CC4auntKLxxS2TM9B96RSOP5vd1kr Ws5P2inCYqJb3+cu/+JY2lvz2fOhAhktH2wD0mRRho2gcs0jOA 5ooUTm3xnItLGQ4b2TRTlrTSeUVZf0XaMwpBk93E4AqjyU+91H tawMrBIYGLIXlIY/a/UVh93oUeWl77RH9g75g3+wo9WpalUQQ1 tM69pqnpXI26A== Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Fri, 2014-02-07 at 09:57 +0100, Sebastian Andrzej Siewior wrote: > * Mike Galbraith | 2014-02-04 16:03:38 [+0100]: > > >On Mon, 2014-02-03 at 20:48 +0100, Sebastian Andrzej Siewior wrote: > > > >> Known issues: > > > >> Mike said that it caused two of his 64 CPUs to spin for five minutes > >> doing nothing. Therefore it is disabled in the queue. Everyone > >> who wants test it may do so :) > > > >I put rt13 on an 8 socket Opetron (48 core) box, with patch enabled. It > >box did NOT reproduce the hangs the 8 socket (1 node, bleh) DL980 did. > > > >Nor did it explode running hotplug stress, but then it deadlocks before > >BOOM would have happened on the DL980. kdump isn't working properly on > >this box (not an rt issue, userspace), so.. > > Okay, so this is at some point good news. Thank you. I've been letting other big boxen play around the clock distro build slave and whatnot, with no sign of trouble. I'm raising an eyebrow in the general direction of DL980 test box. Its microscopic, and more importantly, unsupported, memory configuration might be more than a mere performance hindrance. -Mike