Subject: Re: CVS commit: src/sys/fs/filecorefs
To: Antti Kantee <pooka@cs.hut.fi>
From: Jason Thorpe <thorpej@shagadelic.org>
List: source-changes
Date: 03/04/2007 20:06:06
On Mar 4, 2007, at 2:06 PM, Antti Kantee wrote:

> On Sun Mar 04 2007 at 21:18:08 +0000, John Nemeth wrote:
>>
>> Module Name:	src
>> Committed By:	jnemeth
>> Date:		Sun Mar  4 21:18:08 UTC 2007
>>
>> Modified Files:
>> 	src/sys/fs/filecorefs: filecore_vfsops.c
>>
>> Log Message:
>> more caddr_t -> void * fallout
>>
>>
>> To generate a diff of this commit:
>> cvs rdiff -r1.31 -r1.32 src/sys/fs/filecorefs/filecore_vfsops.c
>
> Is there any reason to not make b_un.b_addr char * or uint8_t * =20
> instead
> of adding these casts to a gazillion places?

Is it because of the pointer arithmetic?  How about "vaddr_t" ?

>
> There might also be other places where it would be good to stop and
> think about the new type instead of converting everything to void *.
> This one just caught my attention.
>
> --=20
> Antti Kantee <pooka@iki.fi>                     Of course he runs =20
> NetBSD
> http://www.iki.fi/pooka/                          =
http://www.NetBSD.org/
>     "la qualit=E9 la plus indispensable du cuisinier est l'exactitude"

-- thorpej