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: <cover.1567007242.git.msuchanek@suse.de> <0ad51b41aebf65b3f3fcb9922f0f00b47932725d.1567007242.git.msuchanek@suse.de>
 <20190829064624.GA28508@infradead.org>
In-Reply-To: <20190829064624.GA28508@infradead.org>
From:   Arnd Bergmann <arnd@arndb.de>
Date:   Thu, 29 Aug 2019 10:01:16 +0200
Message-ID: <CAK8P3a2qgLTbud+2Fw8Rr0RXV8-xwBMiBg3hFuqqBinN1zS90A@mail.gmail.com>
Subject: Re: [PATCH v3 3/4] powerpc/64: make buildable without CONFIG_COMPAT
To:     Christoph Hellwig <hch@infradead.org>
Cc:     Michal Suchanek <msuchanek@suse.de>,
        linuxppc-dev <linuxppc-dev@lists.ozlabs.org>,
        David Hildenbrand <david@redhat.com>,
        Heiko Carstens <heiko.carstens@de.ibm.com>,
        David Howells <dhowells@redhat.com>,
        Paul Mackerras <paulus@samba.org>,
        Breno Leitao <leitao@debian.org>,
        Michael Neuling <mikey@neuling.org>,
        Nicolai Stange <nstange@suse.de>,
        Allison Randal <allison@lohutok.net>,
        Geert Uytterhoeven <geert@linux-m68k.org>,
        Joel Stanley <joel@jms.id.au>,
        Firoz Khan <firoz.khan@linaro.org>,
        Nicholas Piggin <npiggin@gmail.com>,
        Thomas Gleixner <tglx@linutronix.de>,
        Christian Brauner <christian@brauner.io>,
        Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        "Eric W. Biederman" <ebiederm@xmission.com>,
        Andrew Donnellan <andrew.donnellan@au1.ibm.com>,
        Hari Bathini <hbathini@linux.ibm.com>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Thu, Aug 29, 2019 at 8:46 AM Christoph Hellwig <hch@infradead.org> wrote:

> > @@ -277,7 +277,7 @@ static void do_signal(struct task_struct *tsk)
> >
> >       rseq_signal_deliver(&ksig, tsk->thread.regs);
> >
> > -     if (is32) {
> > +     if ((IS_ENABLED(CONFIG_PPC32) || IS_ENABLED(CONFIG_COMPAT)) && is32) {
>
> I think we should fix the is_32bit_task definitions instead so that
> callers don't need this mess.  I'd suggest something like:
>
> #ifdef CONFIG_COMPAT
> #define is_32bit_task()         test_thread_flag(TIF_32BIT)
> #else
> #define is_32bit_task()         IS_ENABLED(CONFIG_PPC32)
> #endif

Are there actually any (correct) uses of is_32bit_task() outside of
#ifdef CONFIG_PPC64?

I suspect most if not all could be changed to the generic
in_compat_syscall() that we use outside of architecture specific
code.

       Arnd