From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([2001:4830:134:3::10]:45601)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <Paul.Burton@imgtec.com>) id 1VEEud-0000Hh-5E
	for qemu-devel@nongnu.org; Tue, 27 Aug 2013 04:46:13 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <Paul.Burton@imgtec.com>) id 1VEEuX-0002qv-5M
	for qemu-devel@nongnu.org; Tue, 27 Aug 2013 04:46:07 -0400
Received: from multi.imgtec.com ([194.200.65.239]:26304)
	by eggs.gnu.org with esmtp (Exim 4.71)
	(envelope-from <Paul.Burton@imgtec.com>) id 1VEEuW-0002qe-VI
	for qemu-devel@nongnu.org; Tue, 27 Aug 2013 04:46:01 -0400
Message-ID: <521C671C.1000805@imgtec.com>
Date: Tue, 27 Aug 2013 09:45:16 +0100
From: Paul Burton <paul.burton@imgtec.com>
MIME-Version: 1.0
References: <1377270586-14753-1-git-send-email-yongbok.kim@imgtec.com>
	<521B980F.4040208@twiddle.net>
In-Reply-To: <521B980F.4040208@twiddle.net>
Content-Type: text/plain; charset="UTF-8"; format=flowed
Content-Transfer-Encoding: 7bit
Subject: Re: [Qemu-devel] [PATCH] mips_malta: support up to 2GiB RAM
List-Id: <qemu-devel.nongnu.org>
List-Unsubscribe: <https://lists.nongnu.org/mailman/options/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=unsubscribe>
List-Archive: <http://lists.nongnu.org/archive/html/qemu-devel>
List-Post: <mailto:qemu-devel@nongnu.org>
List-Help: <mailto:qemu-devel-request@nongnu.org?subject=help>
List-Subscribe: <https://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=subscribe>
To: Richard Henderson <rth@twiddle.net>
Cc: james.hogan@imgtec.com, qemu-devel@nongnu.org, markos.chandras@imgtec.com, Yongbok Kim <yongbok.kim@imgtec.com>, leon.alrae@imgtec.com, aurelien@aurel32.net

Thanks Richard. To be clear, is that a "this would be a nice future 
improvement" or a "NAK until this is done"?

Paul

On Mon 26 Aug 2013 19:01:51 BST, Richard Henderson wrote:
> On 08/23/2013 08:09 AM, Yongbok Kim wrote:
>> +    /* alias for pre IO hole access */
>> +    memory_region_init_alias(ram_low_preio, NULL, "mips_malta_low_preio.ram",
>> +                             ram_high, 0, MIN(ram_size, (256 << 20)));
>> +    memory_region_add_subregion(system_memory, 0, ram_low_preio);
>> +
>> +    /* alias for post IO hole access, if there is enough RAM */
>> +    if (ram_size > (512 << 20)) {
>> +        ram_low_postio = g_new(MemoryRegion, 1);
>> +        memory_region_init_alias(ram_low_postio, NULL,
>> +                                 "mips_malta_low_postio.ram",
>> +                                 ram_high, 512 << 20,
>> +                                 ram_size - (512 << 20));
>> +        memory_region_add_subregion(system_memory, 512 << 20, ram_low_postio);
>> +    }
>
> This sort of thing is what subregion prioirities are for.  Ideally you'd have
> one alias region, with lower priority than the IO hole.
>
> Of course, it looks like most of gt64xxx_pci.c needs cleaning up for proper
> usage of the memory region infrastructure before that can happen...
>
>
> r~