[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <47A9A483.6060709@fr.ibm.com>
Date: Wed, 06 Feb 2008 13:13:55 +0100
From: Daniel Lezcano <dlezcano@...ibm.com>
To: David Miller <davem@...emloft.net>
CC: yoshfuji@...ux-ipv6.org, netdev@...r.kernel.org
Subject: Re: [patch 14/14] replace __inline__ by inline in include/linux (net
related)
David Miller wrote:
> From: Daniel Lezcano <dlezcano@...ibm.com>
> Date: Wed, 06 Feb 2008 11:39:27 +0100
>
>> YOSHIFUJI Hideaki / 吉藤英明 wrote:
>>> In article <20080206092300.701741592@....toulouse-stg.fr.ibm.com> (at Wed, 06 Feb 2008 10:14:19 +0100), Daniel Lezcano <dlezcano@...ibm.com> says:
>>>
>>>> replace __inline__ by inline in include/linux (net related)
>>>>
>>>> Signed-off-by: Daniel Lezcano <dlezcano@...ibm.com>
>>>> ---
>>>> include/linux/atm.h | 4 ++--
>>>> include/linux/atmsap.h | 2 +-
>>>> include/linux/hdlc.h | 13 +++++++------
>>>> include/linux/inetdevice.h | 14 ++++++--------
>>>> include/linux/netlink.h | 5 +++--
>>>> include/linux/rtnetlink.h | 2 +-
>>>> 6 files changed, 20 insertions(+), 20 deletions(-)
>>> Please do this change within the #ifdef __KERNEL__ .. #endif only.
>> Oups ! I didn't thought about userspace program when modifying these
>> headers. Thanks for catching this.
>>
>> Hopefully, the patches can be applied separately. Is it possible to
>> discard this one ?
>
> What is the real problem though?
>
> Oh, c89 style C compilation will break with plain "inline"?
If a userspace program include one of these headers and compile with the
-ansi option, the keyword "inline" will be disabled and the compilation
will fail. For this reason when a header is exported, the __inline__
keyword should be used instead of "inline" to avoid such problem.
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists