From mboxrd@z Thu Jan 1 00:00:00 1970 Return-path: Received: from out01.mta.xmission.com ([166.70.13.231]) by merlin.infradead.org with esmtps (Exim 4.92.3 #3 (Red Hat Linux)) id 1k6bVk-0005fs-Kc for kexec@lists.infradead.org; Fri, 14 Aug 2020 15:20:53 +0000 From: ebiederm@xmission.com (Eric W. Biederman) References: <20200814055239.47348-1-sangyan@huawei.com> Date: Fri, 14 Aug 2020 10:17:06 -0500 In-Reply-To: <20200814055239.47348-1-sangyan@huawei.com> (Sang Yan's message of "Fri, 14 Aug 2020 01:52:38 -0400") Message-ID: <87tux5i7f1.fsf@x220.int.ebiederm.org> MIME-Version: 1.0 Subject: Re: [PATCH 1/2] kexec: Add quick kexec support for kernel List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: "kexec" Errors-To: kexec-bounces+dwmw2=infradead.org@lists.infradead.org To: Sang Yan Cc: luanjianhai@huawei.com, xiexiuqi@huawei.com, kexec@lists.infradead.org, linux-kernel@vger.kernel.org, luchunhua@huawei.com, guohanjun@huawei.com, zhuling8@huawei.com Sang Yan writes: > In normal kexec, relocating kernel may cost 5 ~ 10 seconds, to > copy all segments from vmalloced memory to kernel boot memory, > because of disabled mmu. I haven't seen kexec that slow since I tested on my 16Mhz 386. That machine has an excuse it really is slow. Anything else that takes seconds is almost certainly slow because someone has misconfigured things to not cache the data copied by kexec. I humbly suggest that you fix the arm64 code so that the data gets cached. Eric _______________________________________________ kexec mailing list kexec@lists.infradead.org http://lists.infradead.org/mailman/listinfo/kexec