Re: [PATCH 1/2] x86: Add Isolated Memory Regions for Quark X1000

From: Darren Hart <dvhart@infradead.org>
To: "Bryan O'Donoghue" <pure.logic@nexus-software.ie>
Cc: "Ong, Boon Leong" <boon.leong.ong@intel.com>,
	"tglx@linutronix.de" <tglx@linutronix.de>,
	"mingo@redhat.com" <mingo@redhat.com>,
	"hpa@zytor.com" <hpa@zytor.com>,
	"x86@kernel.org" <x86@kernel.org>,
	"platform-driver-x86@vger.kernel.org" 
	<platform-driver-x86@vger.kernel.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	"andy.shevchenko@gmail.com" <andy.shevchenko@gmail.com>
Subject: Re: [PATCH 1/2] x86: Add Isolated Memory Regions for Quark X1000
Date: Thu, 8 Jan 2015 19:44:40 -0800	[thread overview]
Message-ID: <20150109034439.GA9311@vmdeb7> (raw)
In-Reply-To: <54AE9E27.2050307@nexus-software.ie>

On Thu, Jan 08, 2015 at 03:11:35PM +0000, Bryan O'Donoghue wrote:
> 
> >Suggest to split the imr_del() into 2 functions:-
> >(1) by address + size
> >(2) by IMR index
> >At current implementation, it does not support (2) only because it fails at
> >imr_check_range().
> 
> Hi Boon Leong.
> 
> I'll have a think about that :)
> 
> Just on imr_del() though, it does support removal by way of index.
> 
> +static void __init intel_galileo_imr_init(void)
> +{
> +	unsigned long base  = virt_to_phys(&_text);
> +	unsigned long size = virt_to_phys(&_sinittext) - base - IMR_ALIGN;
> +	int i, ret;
> +
> +	/* Tear down all existing unlocked IMRs */
> +	for (i = 0; i <= QUARK_X1000_IMR_NUM; i++)
> +		imr_del(i, 0, 0);
> 
> That's what the platform code has to do for every unlocked IMR, to make sure
> there are no stale IMRs left that could conflict with the EFI memory map !

I'm OK with a single function so long as by index works without having to
specify the address. Please update the kernel doc to describe this usage though.

-- 
Darren Hart
Intel Open Source Technology Center