Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1759249Ab3ICRf4 (ORCPT ); Tue, 3 Sep 2013 13:35:56 -0400 Received: from terminus.zytor.com ([198.137.202.10]:41637 "EHLO mail.zytor.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757869Ab3ICRfy (ORCPT ); Tue, 3 Sep 2013 13:35:54 -0400 Message-ID: <52261DC8.4070307@zytor.com> Date: Tue, 03 Sep 2013 10:35:04 -0700 From: "H. Peter Anvin" User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/20130625 Thunderbird/17.0.7 MIME-Version: 1.0 To: "Li, Fei" CC: Ingo Molnar , "mingo@redhat.com" , "tglx@linutronix.de" , "Liu, Chuansheng" , "x86@kernel.org" , "akpm@linux-foundation.org" , "holt@sgi.com" , "rmk+kernel@arm.linux.org.uk" , "linux-kernel@vger.kernel.org" Subject: Re: [PATCH] reboot: perform warm/cold reset correctly for CF9 type References: <1377049949.18357.3.camel@fli24-HP-Compaq-8100-Elite-CMT-PC> <20130821072940.GB27495@gmail.com> <69b80c44-3245-4393-b6e9-4d8a7f3d3c66@email.android.com> <7233d29e-ea01-4eb6-8018-b9c8e3730a30@email.android.com> In-Reply-To: X-Enigmail-Version: 1.5.2 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1510 Lines: 38 On 08/21/2013 08:32 PM, Li, Fei wrote: > > In fact, ACPI method also implements reset by writing RESET_VALUE into RESET_REG > port. In our platform, it also uses port 0xCF9. The difference between cold and wart reset > is whether power cycle is involved or not. After all, this is another topic. > > From the source code, CF9 boot method are still used by some apple MAC and dell > platforms if I understand correctly. > > Back to our patch, if you really concern the effect of cold reboot 0x0E to current existed > platforms, we can set the default value of reboot_mode as REBOOT_WARM. > > Does it make sense? > Sort-of-kind-of. The problem here is that "cold" versus "warm" reboot seems to have slightly different meanings for BIOS (bypass POST) and the CF9 register (where I presume it is a "deeper" platform reset of some kind ... is it even well-defined what the semantics are?) I could be wrong and the cold/warm reset values in CF9 end up having exactly the same function as the magic BIOS signature does; if so, then I would like to be told so explicitly, ideally with an explanation about how it works on the hw level (or a pointer to relevant documentation, Intel internal docs are fine.) Finally, again, does this solve a real problem? -hpa -- 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/