[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <4D91A941.7090207@cn.fujitsu.com>
Date: Tue, 29 Mar 2011 17:41:21 +0800
From: Lai Jiangshan <laijs@...fujitsu.com>
To: Jan Beulich <JBeulich@...ell.com>
CC: Ingo Molnar <mingo@...e.hu>,
Alexander van Heukelum <heukelum@...tmail.fm>,
Dipankar Sarma <dipankar@...ibm.com>,
Peter Zijlstra <peterz@...radead.org>,
Andrew Morton <akpm@...ux-foundation.org>,
"H. Peter Anvin" <hpa@...ux.intel.com>,
"Paul E. McKenney" <paulmck@...ux.vnet.ibm.com>,
Sam Ravnborg <sam@...nborg.org>,
David Howells <dhowells@...hat.com>,
Oleg Nesterov <oleg@...hat.com>,
Roland McGrath <roland@...hat.com>, serge.hallyn@...onical.com,
linux-kernel@...r.kernel.org
Subject: Re: [RFC PATCH 4/5] RCU: Add TASK_RCU_OFFSET
On 03/28/2011 04:35 PM, Jan Beulich wrote:
>>>> On 28.03.11 at 05:00, Lai Jiangshan <laijs@...fujitsu.com> wrote:
>>
>> We can not include linux/sched.h in linux/rcupdate.h.
>> If we access task's task_rcu_strut without linux/sched.h included,
>> We will failed with compiling error.
>>
>> So we add TASK_RCU_OFFSET, which help us access
>> task's task_rcu_strut without linux/sched.h included.
>> Now, task_rcu_struct(), current_task_rcu_struct(), task_of_task_rcu() and
>> rcu_copy_process() can be used anywhere without linux/sched.h included.
>
> Aren't the offsets-generation methods meant for assembly
> consumption only? Header dependency problems normally can
> be solved by splitting headers into a type declaration one and
> a second one carrying inline function implementations. Is that
> indeed completely impossible here?
>
That will need to split >100 header files, it is almost impossible.
Thanks,
Lai
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists