From mboxrd@z Thu Jan  1 00:00:00 1970
From: Sasha Levin <levinsasha928@gmail.com>
Subject: Re: [RFC PATCH 13/16] kvm tools: keep track of registered memory
 banks in struct kvm
Date: Tue, 20 Nov 2012 16:01:01 -0500
Message-ID: <50ABEF8D.6050707@gmail.com>
References: <1352721450-11340-1-git-send-email-will.deacon@arm.com> <1352721450-11340-14-git-send-email-will.deacon@arm.com> <50A1CE92.6060803@gmail.com> <20121120171553.GI27765@mudshark.cambridge.arm.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Cc: "kvm@vger.kernel.org" <kvm@vger.kernel.org>,
	"penberg@kernel.org" <penberg@kernel.org>,
	Marc Zyngier <Marc.Zyngier@arm.com>,
	"c.dall@virtualopensystems.com" <c.dall@virtualopensystems.com>,
	Matt Evans <Matt.Evans@arm.com>,
	"peter.maydell@linaro.org" <peter.maydell@linaro.org>
To: Will Deacon <will.deacon@arm.com>
Return-path: <kvm-owner@vger.kernel.org>
Received: from mail-qa0-f46.google.com ([209.85.216.46]:64384 "EHLO
	mail-qa0-f46.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1751933Ab2KTVBb (ORCPT <rfc822;kvm@vger.kernel.org>);
	Tue, 20 Nov 2012 16:01:31 -0500
Received: by mail-qa0-f46.google.com with SMTP id c11so1030162qad.19
        for <kvm@vger.kernel.org>; Tue, 20 Nov 2012 13:01:30 -0800 (PST)
In-Reply-To: <20121120171553.GI27765@mudshark.cambridge.arm.com>
Sender: kvm-owner@vger.kernel.org
List-ID: <kvm.vger.kernel.org>

On 11/20/2012 12:15 PM, Will Deacon wrote:
> Hi Sasha,
> 
> On Tue, Nov 13, 2012 at 04:37:38AM +0000, Sasha Levin wrote:
>> On 11/12/2012 06:57 AM, Will Deacon wrote:
>>>  struct kvm {
>>>  	struct kvm_arch		arch;
>>>  	struct kvm_config	cfg;
>>> @@ -49,6 +56,7 @@ struct kvm {
>>>  	u64			ram_size;
>>>  	void			*ram_start;
>>>  	u64			ram_pagesize;
>>> +	struct list_head	mem_banks;
>>
>> These memory banks actually look like a perfect example to use our augmented interval rb-tree,
>> can we switch them to use it, or is it a list on purpose?
> 
> I found some time to look at this today but unfortunately they're not as
> ideally suited to the interval tree as they look: the problem being that we
> need to search for banks by both host virtual address *and* guest physical
> address depending on the translation that we're doing.
> 
> We could have two separate tress, but that seems like overkill given the
> likely number of banks.

Makes sense. We can convert it later if we need to as well.


Thanks,
Sasha