From mboxrd@z Thu Jan  1 00:00:00 1970
From: rjui@broadcom.com (Ray Jui)
Date: Sun, 7 Dec 2014 17:34:18 -0800
Subject: [PATCH 2/5] gpio: Cygnus: add GPIO driver
In-Reply-To: <1417839865.31745.10.camel@perches.com>
References: <Ray Jui <rjui@broadcom.com>	
 <1417826408-1600-1-git-send-email-rjui@broadcom.com>	
 <1417826408-1600-3-git-send-email-rjui@broadcom.com>	
 <1417829282.31745.3.camel@perches.com> <5482668F.1040907@broadcom.com>	
 <1417833291.31745.5.camel@perches.com> <54827B07.3000707@broadcom.com>
 <1417839865.31745.10.camel@perches.com>
Message-ID: <5485001A.7050909@broadcom.com>
To: linux-arm-kernel@lists.infradead.org
List-Id: linux-arm-kernel.lists.infradead.org



On 12/5/2014 8:24 PM, Joe Perches wrote:
> On Fri, 2014-12-05 at 19:41 -0800, Ray Jui wrote:
>> On 12/5/2014 6:34 PM, Joe Perches wrote:
>>> On Fri, 2014-12-05 at 18:14 -0800, Ray Jui wrote:
>>>> On 12/5/2014 5:28 PM, Joe Perches wrote:
>>>>> On Fri, 2014-12-05 at 16:40 -0800, Ray Jui wrote:
>>>>>> +			for_each_set_bit(bit, &val, 32) {
> []
>> Actually, by reviewing the code more deeply, I'm not sure why using
>> for_each_set_bit here is 'endian unsafe'.
>
> It's not.  The 32 confused me as it was long
> and sizeof(long) isn't necessarily 32.
>
> Maybe the 32 should be a #define
>
>
>
Okay, to improve readability, I will change 32 to NGPIOS_PER_BANK.