From: Arnd Bergmann <arnd@arndb.de>
To: Joseph Myers <joseph@codesourcery.com>
Cc: Chris Metcalf <cmetcalf@ezchip.com>, linux-arm-kernel@lists.infradead.org,
        bamvor.zhangjian@huawei.com, Andrew Pinski <pinskia@gmail.com>,
        "Kapoor, Prasun" <Prasun.Kapoor@caviumnetworks.com>,
        Andreas Schwab <schwab@suse.de>,
        Nathan Lynch <Nathan_Lynch@mentor.com>,
        LKML <linux-kernel@vger.kernel.org>, Alexander Graf <agraf@suse.de>,
        Alexey Klimov <klimov.linux@gmail.com>, broonie@kernel.org,
        Yury Norov <ynorov@caviumnetworks.com>,
        Andrew Pinski <apinski@cavium.com>,
        David Daney <ddaney.cavm@gmail.com>,
        Catalin Marinas <catalin.marinas@arm.com>,
        Jan Dakinevich <jan.dakinevich@gmail.com>,
        Philipp Tomsich <philipp.tomsich@theobroma-systems.com>,
        Andrey Konovalov <andrey.konovalov@linaro.org>,
        christoph.muellner@theobroma-systems.com,
        Mike Frysinger <vapier@gentoo.org>, Paul Eggert <eggert@cs.ucla.edu>,
        Rich Felker <dalias@libc.org>
Subject: Re: [PATCH v6 13/17] arm64:ilp32: add sys_ilp32.c and a separate table (in entry.S) to use it
Date: Mon, 16 Nov 2015 12:00:17 +0100
Message-ID: <4814795.z9jdqoW8IP@wuerfel>
User-Agent: KMail/4.11.5 (Linux/3.16.0-10-generic; KDE/4.11.5; x86_64; ; )
In-Reply-To: <alpine.DEB.2.10.1511161001250.25844@digraph.polyomino.org.uk>
References: <1446507046-24604-1-git-send-email-ynorov@caviumnetworks.com> <5648B5E3.3060700@ezchip.com> <alpine.DEB.2.10.1511161001250.25844@digraph.polyomino.org.uk>
MIME-Version: 1.0
Content-Transfer-Encoding: 7Bit
Content-Type: text/plain; charset="us-ascii"
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1940
Lines: 37

On Monday 16 November 2015 10:16:35 Joseph Myers wrote:
> On Sun, 15 Nov 2015, Chris Metcalf wrote:
> 
> > I've added Rich, Paul, Joseph, and Mike to the cc's as they are probably 
> > a good subset of libc-alpha to help comment on these issues.  My sense 
> > is that right now, it wouldn't be possible to add a 32-bit architecture 
> > with a non-32-bit default for _FILE_OFFSET_BITS.  And, obviously, this 
> > is why, when I added the tilegx32 APIs to glibc in 2011, I needed to 
> > provide _FILE_OFFSET_BITS=32 support.
> 
> x32 uses 64-bit off_t only.  That's not a problem; the problems are 
> tv_nsec not of type long, a bug we should avoid for all new ports (padding 
> on tv_nsec is fine; treating that padding as a significant high part of a 
> 64-bit value on input to glibc / kernel interfaces isn't), and maybe some 
> other types being 64-bit unnecessarily, although as far as I know the 
> suggested issues there 
> <https://sourceware.org/bugzilla/show_bug.cgi?id=16438> are all 
> theoretical.

Let's not get into the tv_nsec discussion today, that is not thankfully
not relevant for arm64 any more at this point. The system call ABI for
arm64/ilp32 is now the same as for any other 32-bit architecture using
the generic ABI, the question we're trying to solve here is only whether it
is ok for new 32-bit glibc ports to only offer a 64-bit off_t as the kernel
currently does (using __kernel_loff_t) or if we still need to support the
_FILE_OFFSET_BITS=32 case.

If I got you right, we can use 64-bit off_t now, so we just need someone
to figure out how to make that the default in glibc for new architectures
while keeping the existing 32-bit architectures unchanged.

	Arnd
--
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/