From mboxrd@z Thu Jan  1 00:00:00 1970
From: Yinghai <yinghai.lu@oracle.com>
Subject: Re: [PATCH 08/35] x86,lmb: Add lmb_reserve_area/lmb_free_area
Date: Thu, 13 May 2010 23:30:59 -0700
Message-ID: <4BECEE23.3080904@oracle.com>
References: <1273796396-29649-1-git-send-email-yinghai@kernel.org>	 <1273796396-29649-9-git-send-email-yinghai@kernel.org> <1273803964.21352.380.camel@pasglop>
Mime-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
Return-path: <linux-arch-owner@vger.kernel.org>
Received: from rcsinet10.oracle.com ([148.87.113.121]:59391 "EHLO
	rcsinet10.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1755650Ab0ENGdX (ORCPT
	<rfc822;linux-arch@vger.kernel.org>); Fri, 14 May 2010 02:33:23 -0400
In-Reply-To: <1273803964.21352.380.camel@pasglop>
Sender: linux-arch-owner@vger.kernel.org
List-ID: <linux-arch.vger.kernel.org>
To: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Ingo Molnar <mingo@elte.hu>, Thomas Gleixner <tglx@linutronix.de>, "H. Peter Anvin" <hpa@zytor.com>, Andrew Morton <akpm@linux-foundation.org>, David Miller <davem@davemloft.net>, Linus Torvalds <torvalds@linux-foundation.org>, Johannes Weiner <hannes@cmpxchg.org>, linux-kernel@vger.kernel.org, linux-arch@vger.kernel.org

On 05/13/2010 07:26 PM, Benjamin Herrenschmidt wrote:
> On Thu, 2010-05-13 at 17:19 -0700, Yinghai Lu wrote:
> 
>>  #endif
>> diff --git a/arch/x86/mm/lmb.c b/arch/x86/mm/lmb.c
>> index 37a05e2..0dbe05b 100644
>> --- a/arch/x86/mm/lmb.c
>> +++ b/arch/x86/mm/lmb.c
>> @@ -117,3 +117,30 @@ void __init lmb_to_bootmem(u64 start, u64 end)
>>  	lmb.reserved.cnt = 0;
>>  }
>>  #endif
>> +
>> +void __init lmb_add_memory(u64 start, u64 end)
>> +{
>> +	lmb_add_region(&lmb.memory, start, end - start);
>> +}
> 
> I completely fail the point of doing such a minor argument conversion
> as an exported function, with a naming that would cause certain
> confusion with the existing lmb_add().
> 
> In any case, the above should be done at the call sites. Just call
> lmb_add(start, end-start). You also aren't consistent since you do the
> similar conversion using the _area suffix below, but not above.
> 
>> +void __init lmb_reserve_area(u64 start, u64 end, char *name)
>> +{
>> +	if (start == end)
>> +		return;
>> +
>> +	if (WARN_ONCE(start > end, "lmb_reserve_area: wrong range [%#llx, %#llx]\n", start, end))
>> +		return;
>> +
>> +	lmb_add_region(&lmb.reserved, start, end - start);
>> +}
> 
> You seem to be fond of gratuituous bloat... 
> 
>> +void __init lmb_free_area(u64 start, u64 end)
>> +{
>> +	if (start == end)
>> +		return;
>> +
>> +	if (WARN_ONCE(start > end, "lmb_free_area: wrong range [%#llx, %#llx]\n", start, end))
>> +		return;
>> +
>> +	lmb_free(start, end - start);
>> +}
> 
> And here again.
> 
> If you -really- think there's value in the prototype conversions, then
> make those _area() variants static inlines, group the well together in
> the .h with a clear explanation saying something like "it's more
> practical for some arch to use start/end rather than start/size".
> 
> I personally don't see why you are doing that tho.

make the rebase more easy.

previous version is using those api.

also will add some debug print out with them