From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751349AbbDHGXX (ORCPT ); Wed, 8 Apr 2015 02:23:23 -0400 Received: from conssluserg004.nifty.com ([202.248.44.42]:60012 "EHLO conssluserg004-v.nifty.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1751177AbbDHGXW (ORCPT ); Wed, 8 Apr 2015 02:23:22 -0400 X-Nifty-SrcIP: [209.85.212.179] MIME-Version: 1.0 In-Reply-To: <19177883.oJbWsfhyVG@wuerfel> References: <3196164.8b31S2QWFd@wuerfel> <19177883.oJbWsfhyVG@wuerfel> Date: Wed, 8 Apr 2015 15:23:14 +0900 Message-ID: Subject: Re: not syncing: Attempted to kill init! exitcode=0x00000004 ? From: Masahiro Yamada To: Arnd Bergmann Cc: linux-arm-kernel@lists.infradead.org, Linux Kernel Mailing List Content-Type: text/plain; charset=UTF-8 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hi Arnd, 2015-04-07 21:22 GMT+09:00 Arnd Bergmann : > On Tuesday 07 April 2015 17:22:32 Masahiro Yamada wrote: >> Hi Arnd, >> >> Thanks for your comment! >> >> 2015-04-07 16:41 GMT+09:00 Arnd Bergmann : >> > On Tuesday 07 April 2015 12:34:30 Masahiro Yamada wrote: >> >> >> >> What is the cause of the kernel panic? How to fix it? >> >> Any hint is very appreciated. >> >> >> > >> > Cortex-A9 usually need to set up the L2 Cache controller, and you >> > don't have a node for that. >> > >> > Can you try adding a DT node for it? You may also have to specify >> > the overrides for its aux control register in DT if that is not >> > set up right by the boot loader or the power-on defaults. >> >> I have not checked the L2 (outer) cache yet. >> >> The L2 cache on our SoC is not a famous IP, but our own implementation. >> I will have to implement L2 code as well as the device tree node. >> Is there any workaround to boot Linux without L2 cache? > > It should work normally, I think the only dangerous combination is > when the L2 controller is enabled but not configured correctly. OK, I must check this area. > It's also quite likely that there is some other problem with your > user space image. Are you able to boot the same image inside of > qemu or on another hardware that is supported with multi_v7_defconfig? Yes, I have a Zynq Zedboard. It successfully boots Linux with the same kernel image and initramdisk. -- Best Regards Masahiro Yamada