Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
MIME-Version: 1.0
References: <20191223130834.GA102399@zx2c4.com> <20191224135404.389039-1-Jason@zx2c4.com>
 <CAK8P3a1fVFDkHe=gLy55rHxwfZ8YKcUSYvnhSoMbcAgWy6Nm9w@mail.gmail.com>
 <CAHmME9o07dQV_MmWmtBFCKp=sdsO-scC6-UbXNi=dpU6umCoPg@mail.gmail.com>
 <CAK8P3a0sWObusG3xO_JE9CXCyNfFN0p6OgPjUyU2CHLBBZNpZw@mail.gmail.com> <CAHmME9o==nBONywVgSjsmjs2H_A8itgmwibqzPmvivcSocKWRQ@mail.gmail.com>
In-Reply-To: <CAHmME9o==nBONywVgSjsmjs2H_A8itgmwibqzPmvivcSocKWRQ@mail.gmail.com>
From:   Arnd Bergmann <arnd@arndb.de>
Date:   Mon, 30 Dec 2019 16:37:17 +0100
Message-ID: <CAK8P3a11g-UXcYdudDtp0TWCQAfotpc-63BqYwn-a9LDxV-b+Q@mail.gmail.com>
Subject: Re: [PATCH] mips: vdso: conditionalize 32-bit time functions on COMPAT_32BIT_TIME
To:     "Jason A. Donenfeld" <Jason@zx2c4.com>
Cc:     "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        "open list:BROADCOM NVRAM DRIVER" <linux-mips@vger.kernel.org>,
        Paul Burton <paulburton@kernel.org>,
        Vincenzo Frascino <vincenzo.frascino@arm.com>,
        Christian Brauner <christian.brauner@canonical.com>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Mon, Dec 30, 2019 at 3:37 PM Jason A. Donenfeld <Jason@zx2c4.com> wrote:
>
> On Mon, Dec 30, 2019 at 1:34 PM Arnd Bergmann <arnd@arndb.de> wrote:
> >
> > - Why does it crash in the first place rather than returning -ENOSYS?
>
> There's a bit of speculation about this in the original thread that
> prompted this patch (you're CC'd).
>
> >
> > - How does it actually work if you run an application built against
> >   an old musl version on a kernel that tries to make this not work?
> >   Do you just get a random time (uninitialized user space stack) and
> >   work with that without checking the error code?
>
> Actually, your patch fails here. The ts struct remains as it was
> before, filled with garbage. No good. My original patch in this
> thread, though, does result in the correct value being written to ts.

Ok, that is the intended behavior then, clock_gettime() needs
to fail with -EINVAL or -ENOSYS here (depending on the libc
implementation), and of course the data is not updated.

Returning success from clock_gettime() on a kernel with only
time64 support and a libc with only time32 support (or vice
versa) would be a bug.

    Arnd