Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753539Ab0LWRsy (ORCPT ); Thu, 23 Dec 2010 12:48:54 -0500 Received: from rcsinet10.oracle.com ([148.87.113.121]:54580 "EHLO rcsinet10.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752286Ab0LWRsx convert rfc822-to-8bit (ORCPT ); Thu, 23 Dec 2010 12:48:53 -0500 References: <153445735.411361293110429680.JavaMail.root@zimbra-corp-1-b8> In-Reply-To: <153445735.411361293110429680.JavaMail.root@zimbra-corp-1-b8> Mime-Version: 1.0 (iPhone Mail 8C148) Content-Transfer-Encoding: 8BIT Content-Type: text/plain; charset=utf-8 Message-Id: <41A13AA7-7CC7-4AA7-858A-ED00DAF24BEB@kernel.org> Cc: "linux-kernel@vger.kernel.org" , Dirk Brandewie , "sodaville@linutronix.de" , Ingo Molnar , Florian Fainelli X-Mailer: iPhone Mail (8C148) From: Yinghai Subject: Re: [PATCH] x86/trampoline: fallback to fixed trampoline address if no e820 map is provided Date: Thu, 23 Dec 2010 09:46:38 -0800 To: Florian Fainelli Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 3147 Lines: 91 Can you check if you can add ce4100 own memory_setup() and install it into x86_init.resources.memory_setup in x86_ce4100_early_setup? Thanks On Dec 23, 2010, at 5:20 AM, Florian Fainelli wrote: > > ----- "Yinghai" a écrit : > >> On Dec 23, 2010, at 1:57 AM, Florian Fainelli >> wrote: >> >>> On Tuesday 14 December 2010 22:19:39 Florian Fainelli wrote: >>>> From: Florian Fainelli >>>> >>>> Since 893f38d1 (x86: Use find_e820() instead of hard coded >> trampoline >>>> address), the trampoline base address should be provided by the >> BIOS as an >>>> e820 area. Some minimalistic BIOSes (like CEFDK on CE4100) may not >> provide >>>> any e820 area, still we should fallback to a fixed trampoline base >> address >>>> for SMP to work correctly. >> >> No e820 memmap ? > > Yes, the BIOS is only doing the very minimal setup to get a kernel in real-mode booting. > >> >> Not sure how does it work >> >> Do you have boot log with debug ? > > I would need to produce one for you, give me a couple days to be in sync. > >> >> Also recent kernel is using memblock already for finding free area. >> >> Thanks >> >> >> >>>> >>>> Cc: Dirk Brandewie >>>> Cc: sodaville@linutronix.de >>>> Cc: Ingo Molnar >>>> Cc: Yinghai Lu >>>> Signed-off-by: Florian Fainelli >>> >>> Ingo, Yinghai, anything wrong with that patch? Thank you. >>> >>>> --- >>>> diff --git a/arch/x86/include/asm/trampoline.h >>>> b/arch/x86/include/asm/trampoline.h index f4500fb..908b395 100644 >>>> --- a/arch/x86/include/asm/trampoline.h >>>> +++ b/arch/x86/include/asm/trampoline.h >>>> @@ -16,6 +16,7 @@ extern unsigned long initial_code; >>>> extern unsigned long initial_gs; >>>> >>>> #define TRAMPOLINE_SIZE roundup(trampoline_end - trampoline_data, >>>> PAGE_SIZE) +#define TRAMPOLINE_BASE 0x6000 >>>> >>>> extern unsigned long setup_trampoline(void); >>>> extern void __init reserve_trampoline_memory(void); >>>> diff --git a/arch/x86/kernel/trampoline.c >> b/arch/x86/kernel/trampoline.c >>>> index f1488a3..d025aae 100644 >>>> --- a/arch/x86/kernel/trampoline.c >>>> +++ b/arch/x86/kernel/trampoline.c >>>> @@ -21,8 +21,11 @@ void __init reserve_trampoline_memory(void) >>>> >>>> /* Has to be in very low memory so we can execute real-mode AP >> code. */ >>>> mem = find_e820_area(0, 1<<20, TRAMPOLINE_SIZE, PAGE_SIZE); >>>> - if (mem == -1L) >>>> - panic("Cannot allocate trampoline\n"); >>>> + if (mem == -1L) { >>>> + printk(KERN_INFO "BIOS did not provide e820 area for >> trampoline" >>>> + " ,using static trampoline address\n"); >>>> + mem = TRAMPOLINE_BASE; >>>> + } >>>> >>>> trampoline_base = __va(mem); >>>> reserve_early(mem, mem + TRAMPOLINE_SIZE, "TRAMPOLINE"); -- 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/