Date: Tue, 21 Apr 2015 17:13:32 +0200 (CEST)
From: Thomas Gleixner <tglx@linutronix.de>
To: Arnd Bergmann <arnd@arndb.de>
cc: y2038@lists.linaro.org, Baolin Wang <baolin.wang@linaro.org>,
        pang.xunlei@linaro.org, peterz@infradead.org, benh@kernel.crashing.org,
        heiko.carstens@de.ibm.com, paulus@samba.org, cl@linux.com,
        heenasirwani@gmail.com, linux-arch@vger.kernel.org,
        linux-s390@vger.kernel.org, mpe@ellerman.id.au,
        rafael.j.wysocki@intel.com, ahh@google.com, fweisbec@gmail.com,
        pjt@google.com, riel@redhat.com, richardcochran@gmail.com,
        schwidefsky@de.ibm.com, john.stultz@linaro.org, rth@twiddle.net,
        gregkh@linuxfoundation.org, linux-kernel@vger.kernel.org,
        netdev@vger.kernel.org, tj@kernel.org, linux390@de.ibm.com,
        linuxppc-dev@lists.ozlabs.org
Subject: Re: [Y2038] [PATCH 04/11] posix timers:Introduce the 64bit methods
 with timespec64 type for k_clock structure
In-Reply-To: <3231171.5TrYVVBLh4@wuerfel>
Message-ID: <alpine.DEB.2.11.1504211701140.13914@nanos>
References: <1429509459-17068-1-git-send-email-baolin.wang@linaro.org> <3755355.Xf0HbltZXg@wuerfel> <alpine.DEB.2.11.1504211528040.13914@nanos> <3231171.5TrYVVBLh4@wuerfel>
User-Agent: Alpine 2.11 (DEB 23 2013-08-11)
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 2499
Lines: 56

On Tue, 21 Apr 2015, Arnd Bergmann wrote:
> On Tuesday 21 April 2015 16:14:26 Thomas Gleixner wrote:
> > > Note the use of a separate __kernel_itimerspec64 for the user interface
> > > here, which I think will be needed to hide the differences between the
> > > normal itimerspec on 64-bit machines, and the new itimerspec on 32-bit
> > > platforms that will be defined differently (using 'long long').
> > 
> > Confused.
> > 
> > timespec64 / itimerspec64 should be the same independent of 64bit and
> > 32bit. So why do we need another variant ?
> 
> There are multiple reasons:
> 
> * On 64-bit systems, timespec64 would always be defined in the same way
>   as struct timespec { __kernel_time_t tv_sec; long tv_nsec; }, with
>   __kernel_time_t being 'long'. On 32-bit, we probably need to make both
>   members 'long long' for the user space side, in order to share the
>   syscall implementation with the kernel side, but we may also want to
>   keep the internal timespec64 using a 'long' for tv_nsec, as we do
>   today. This means that both the binary layout (padding or no padding)
>   and the basic types (long or long long) are different between 32-bit
>   and 64-bit, and between kernel and user space

So you want to avoid a compat syscall for 32bit applications on a
64bit kernel, right?

That burdens 32bit with the extra 'long long' in user space. Not sure
whether user space folks will be happy about it.

> * We should not put 'struct timespec64' into the user space namespace,
>   as applications might already use that identifier. This is similar
>   to the __u32/u32 or __kernel_time_t/time_t tuple of types for interface
>   and in-kernel uses. This is particularly important when embedding a
>   timespec in another data structure.

Fair enough.

> * My plan is to use a temporary hack where I actually define
>   __kernel_timespec64 to look like the 32-bit version of timespec,
>   as an intermediate step when converting all 32-bit architectures over
>   to use the compat_*() syscalls in place of the existing ones, so
>   I can change over the normal syscalls to use __kernel_timespec64
>   without having to change all architectures at once, or having to
>   modify each syscall multiple times.

Makes sense.

Thanks,

	tglx
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/