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: <1536036087-15260-1-git-send-email-firoz.khan@linaro.org>
 <1536036087-15260-2-git-send-email-firoz.khan@linaro.org> <CAK8P3a1JCnSPi+D+CozhqL0dS9WwXtONDHp4=Kg5kCHrpdYwrw@mail.gmail.com>
 <CALxhOngCL4DEcMD7NaxbUiksByzoxaaozzZGPmB0JFziFEWojw@mail.gmail.com>
In-Reply-To: <CALxhOngCL4DEcMD7NaxbUiksByzoxaaozzZGPmB0JFziFEWojw@mail.gmail.com>
From:   Arnd Bergmann <arnd@arndb.de>
Date:   Mon, 24 Sep 2018 23:06:25 +0200
Message-ID: <CAK8P3a0_omFwid2NZzs_T=A_TY9ddfGOsVKV2vtiaVodWswbmQ@mail.gmail.com>
Subject: Re: [PATCH 1/4] sparc: Move __IGNORE* entries to non uapi header
To:     Firoz Khan <firoz.khan@linaro.org>
Cc:     David Miller <davem@davemloft.net>,
        sparclinux <sparclinux@vger.kernel.org>,
        gregkh <gregkh@linuxfoundation.org>,
        Philippe Ombredanne <pombredanne@nexb.com>,
        Thomas Gleixner <tglx@linutronix.de>,
        Kate Stewart <kstewart@linuxfoundation.org>,
        y2038 Mailman List <y2038@lists.linaro.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        linux-arch <linux-arch@vger.kernel.org>,
        Deepa Dinamani <deepa.kernel@gmail.com>,
        Marcin Juszkiewicz <marcin.juszkiewicz@linaro.org>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Tue, Sep 18, 2018 at 1:53 PM Firoz Khan <firoz.khan@linaro.org> wrote:
>
> On 6 September 2018 at 20:58, Arnd Bergmann <arnd@arndb.de> wrote:
> > On Tue, Sep 4, 2018 at 6:42 AM Firoz Khan <firoz.khan@linaro.org> wrote:
> >
> >> +++ b/arch/sparc/include/uapi/asm/unistd.h
> >> @@ -15,12 +15,6 @@
> >>  #ifndef _UAPI_SPARC_UNISTD_H
> >>  #define _UAPI_SPARC_UNISTD_H
> >>
> >> -#ifndef __32bit_syscall_numbers__
> >> -#ifndef __arch64__
> >> -#define __32bit_syscall_numbers__
> >> -#endif
> >> -#endif
> >
> > This is certainly required in the uapi header as of this patch,
> > without it all the numbers are wrong when you include the
> > file from user space.
> >
> > I suppose it can be removed later once the header is replaced
> > with the two generated versions,
>
> The script will generate 2 versions, ie, unistd_32.h and unistd_64.h.
> Please give few more pointer here.

You still need to check at least for __arch64__ in asm/unistd.h in order
to pick which header to include, right? Since the
__32bit_syscall_numbers__ check was already in the public header
file, it may also be possible that there is some user space application
that sets this in order to get the 32-bit definitions. This might e.g.
be needed for strace.

     Arnd