From mboxrd@z Thu Jan 1 00:00:00 1970 From: lijiang Date: Tue, 27 Nov 2018 02:58:38 +0000 Subject: Re: [PATCH 0/2 RESEND v7] add reserved e820 ranges to the kdump kernel e820 table Message-Id: <463c284b-115e-0784-ca07-a430127e6176@redhat.com> List-Id: References: <20181124051223.19994-1-lijiang@redhat.com> <35e8ade5-d5a8-b5e4-180b-44b4ff0a83b1@intel.com> In-Reply-To: <35e8ade5-d5a8-b5e4-180b-44b4ff0a83b1@intel.com> MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 8bit To: Dave Hansen , linux-kernel@vger.kernel.org Cc: kexec@lists.infradead.org, x86@kernel.org, linux-ia64@vger.kernel.org, linux-efi@vger.kernel.org, tglx@linutronix.de, mingo@redhat.com, bp@alien8.de, akpm@linux-foundation.org, dave.hansen@linux.intel.com, luto@kernel.org, peterz@infradead.org, ard.biesheuvel@linaro.org, tony.luck@intel.com, fenghua.yu@intel.com, dyoung@redhat.com, bhe@redhat.com 在 2018年11月27日 02:54, Dave Hansen 写道: > On 11/23/18 9:12 PM, Lianbo Jiang wrote: >> These patches add the new I/O resource descriptor 'IORES_DESC_RESERVED' >> for the iomem resources search interfaces, and in order to make it still >> work after the new descriptor is added, these codes originally related >> to 'IORES_DESC_NONE' have been updated. > > This is rather anemic "0/" text. Could you please include some more > background in here? The 2/2 patch is pretty good in this regard, but it > needs to be here, too. > Thanks for your comment. Originally, this patch added the e820 reserved ranges by accurately comparing a string. It only modifies fewer code paths. Please refer to patch v6. https://lore.kernel.org/lkml/20181114072926.13312-2-lijiang@redhat.com/ Because the string comparison is fragile and error-prone, this patch used the solution that adds a new descriptor 'IORES_DESC_RESERVED'. Please refer to this link: https://lore.kernel.org/lkml/20181120192935.GK2527@zn.tnic/ When passing the e820 reserved ranges to the second kernel, why does it need to compare strings accurately or add a new descriptor 'IORES_DESC_RESERVED'? -The reason is that it can not exactly match the e820 reserved resource ranges when walking through iomem resources with the descriptor 'IORES_DESC_NONE'. Thanks. Lianbo