From mboxrd@z Thu Jan  1 00:00:00 1970
From: Scott Wood <scottwood@freescale.com>
Date: Thu, 18 Aug 2011 10:14:18 -0500
Subject: [U-Boot] [PATCH 1/2 v3] fdt: Add a do_fixup_by_path_string()
 function
In-Reply-To: <4E4C5CFF.9020307@gmail.com>
References: <1313562246-24627-1-git-send-email-Chunhe.Lan@freescale.com>
	<4E4C5CFF.9020307@gmail.com>
Message-ID: <4E4D2C4A.7050204@freescale.com>
List-Id: <u-boot.lists.denx.de>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
To: u-boot@lists.denx.de

On 08/17/2011 07:29 PM, Jerry Van Baren wrote:
> Hi Chunhe Lan,
> 
> On 08/17/2011 02:24 AM, Chunhe Lan wrote:
> 
> [snip]
> 
>> +
>> +static inline void do_fixup_by_path_string(void *fdt, const char *path,
>> +                       const char *prop, const char *status)
>> +{
>> +    do_fixup_by_path(fdt, path, prop, status, strlen(status) + 1, 1);
>> +}
>> +
> 
> After all the good advice from Scott et al., the patch turns into a
> pretty trivial one-liner.  I am questioning the advantage of calling
>   do_fixup_by_path_string(fdt, path, prop, status);
> vs. simply calling
>   do_fixup_by_path(fdt, path, prop, status, strlen(status) + 1, 1);
> 
> The do_fixup_by_path_string() saves two parameters
>   "strlen(status) + 1, 1"
> at the cost of Yet Another Function.  Is it worth it?

I think it's a nice convenience, with no runtime cost.  It avoids any
chance of a mismatch between the string passed to do_fixup_by_path and
the string passed to strlen.

More functions versus open-coding is not generally a bad thing.

-Scott