[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <20190228211126.GA28901@avx2>
Date: Fri, 1 Mar 2019 00:11:26 +0300
From: Alexey Dobriyan <adobriyan@...il.com>
To: linux-kernel@...r.kernel.org
Cc: linux-kernel@...r.kernel.org, yamada.masahiro@...ionext.com,
wang.bo116@....com.cn, wdjjwb@....com
Subject: Re: +
include-linux-typesh-use-unsigned-int-instead-of-unsigned.patch added to -mm
tree
On Thu, Feb 28, 2019 at 09:41:09AM -0800, akpm@...ux-foundation.org wrote:
> --- a/include/linux/types.h~include-linux-typesh-use-unsigned-int-instead-of-unsigned
> +++ a/include/linux/types.h
> @@ -155,9 +155,9 @@ typedef u64 dma_addr_t;
> typedef u32 dma_addr_t;
> #endif
>
> -typedef unsigned __bitwise gfp_t;
> -typedef unsigned __bitwise slab_flags_t;
> -typedef unsigned __bitwise fmode_t;
> +typedef unsigned int __bitwise gfp_t;
> +typedef unsigned int __bitwise slab_flags_t;
> +typedef unsigned int __bitwise fmode_t;
I don't know if this is desireable. Switching to "unsigned" is better:
1 less token and less 80 column pressure.
Powered by blists - more mailing lists