[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CAMe9rOrxpEPjg_N7UnihFwSYTSxzGsfUCD04WsG=3xp-dYVGAw@mail.gmail.com>
Date: Fri, 16 Mar 2012 12:16:57 -0700
From: "H.J. Lu" <hjl.tools@...il.com>
To: Paul Eggert <eggert@...ucla.edu>
Cc: LKML <linux-kernel@...r.kernel.org>,
"H. Peter Anvin" <hpa@...or.com>,
GNU C Library <libc-alpha@...rceware.org>
Subject: Re: PATCH [3/n]: Add __snseconds_t and __SNSECONDS_T_TYPE
On Fri, Mar 16, 2012 at 11:54 AM, Paul Eggert <eggert@...ucla.edu> wrote:
> On 03/16/2012 08:49 AM, H.J. Lu wrote:
>> I'd prefer to change POSIX. This isn't the only place where x32 isn't
>> 100% compatible with POSIX.
>
> If this approach is taken, the compatibility issues with tv_nsec,
> blksize_t, and suseconds_t should be documented in the glibc manual,
> in areas where programmers using the interfaces are likely to find
> out about the problem. Could you propose a patch to do that?
I will do that once we know the full scope for POSIX issue under
x32.
> It should be possible to change POSIX at some point, once the POSIX folks
> decide that C90 compatibility is no longer an issue. If POSIX
> conformance is not a goal of x32, then this is not an urgent matter --
> which is probably just as well, as I think they still worry about C90.
>
> In the meantime, I guess applications will just have to know not to take
> the address of a tv_nsec member. I just now added a note
> about this in the Gnulib portability manual, where it talks about
> glitches on systems that don't conform to POSIX:
>
> http://git.savannah.gnu.org/cgit/gnulib.git/commit/?id=47834c92f8122f4ba5e6fca5199a4611425bfe69
Thanks.
--
H.J.
--
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