Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751354AbbFKGoT (ORCPT ); Thu, 11 Jun 2015 02:44:19 -0400 Received: from 8bytes.org ([81.169.241.247]:36749 "EHLO theia.8bytes.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750715AbbFKGoM (ORCPT ); Thu, 11 Jun 2015 02:44:12 -0400 Date: Thu, 11 Jun 2015 08:44:10 +0200 From: Joerg Roedel To: Yinghai Lu Cc: Baoquan He , Borislav Petkov , Ingo Molnar , LKML Subject: Re: [PATCH 3/3] x86/crash: Allocate enough low memory when crashkernel=high Message-ID: <20150611064410.GF20384@8bytes.org> References: <1433749795-5796-1-git-send-email-bp@alien8.de> <1433749795-5796-4-git-send-email-bp@alien8.de> <20150610070412.GA16351@dhcp-128-28.nay.redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1551 Lines: 35 On Wed, Jun 10, 2015 at 10:48:14AM -0700, Yinghai Lu wrote: > So under 4G low ram is tight, with changing default to 256M is just lazy > and punishing the other system that does not need that. > those systems do not need to have "crashkernel=xx,low" before. The definition of 'default' is that it should be a good value to work (ideally) everywhere. The previous value did not fulfill this definition, as on systems without (an enabled) hardware IOMMU and some 32bit-DMA-only devices the crashkernel quickly ran out of low-mem with it. And the new default value is not just one that we guessed as being a better one. Both Baoquan and me did testing on affected systems and after a discussion we agreed on it. > should have other smart way to detect and update the default value. > even dmi blacklist way is better than just changing default value. DMI information is not enough, it is about the availability of a hardware IOMMU and the amount and kind of devices needing low-mem for DMA. This means introducing heuristics, which is often the worst solution, because heuristics are complex and if we fix it for one machine we might break many others. So changing the default to a more workable value is the best solution imho. But if you have a better idea I am happy to review your patches. Joerg -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/