Re: Change in functionality of futex() system call.

[Darren Hart <dvhart@linux.intel.com>]

On 06/12/2011 07:11 PM, KOSAKI Motohiro wrote:
>>> Off topic: current futex documentations are near terribly unclear and
>>> many futex op are completely undocumented. They are one of root cause
>>> that every change can make compatibility issue. (;_;
>>
>> What documentation are you referring to? The futex man page is a wreck,
>> and I'm not sure what to do with it since glibc removed the futex()
>> call. You now have to wrap the syscall manually anyway.
> 
> Honestly, I don't know linux man pages policy at all. example, gettid(2)
> also need to be wrap syscall manually. and it's documented and NOTES section
> describe "Glibc does not provide a wrapper for this system call; call it
> using syscall(2)".

Something like that would be good. I don't have any experience pushing
man-page updates. This page seems to document the process:

http://www.kernel.org/doc/man-pages/

> 
> Or, if nobody want to update the doc, shouldn't we just remove futex(2) man
> pages? out date docs are often wrong than nothing. I dunno.

I'd prefer to see it updated as I know of several users of the interface
outside of glibc.

> 
> 
>> If you are referring to the futex.c file itself, I have been documenting
>> functions as I modify them. If you found any of those lacking, please
>> let me know which ones and I'll try to clean them up. If you're
>> referring to those that remain undocumented, please send a doc patch and
>> I'll review and help get it upstream. I'd like to see this improved as well.
> 
> No. I think the comments of futex.c are very good, at least, than a lot of mm code. ;)
> 
> 

-- 
Darren Hart
Intel Open Source Technology Center
Yocto Project - Linux Kernel