Re: [PATCH v1] powerpc: Make setjmp/longjump signature standard

From: Nathan Chancellor <natechancellor@gmail.com>
To: Nick Desaulniers <ndesaulniers@google.com>
Cc: Clement Courbet <courbet@google.com>,
	Michael Ellerman <mpe@ellerman.id.au>,
	Benjamin Herrenschmidt <benh@kernel.crashing.org>,
	Paul Mackerras <paulus@samba.org>,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	Segher Boessenkool <segher@kernel.crashing.org>,
	Christophe Leroy <christophe.leroy@c-s.fr>,
	linuxppc-dev <linuxppc-dev@lists.ozlabs.org>,
	LKML <linux-kernel@vger.kernel.org>,
	clang-built-linux <clang-built-linux@googlegroups.com>
Subject: Re: [PATCH v1] powerpc: Make setjmp/longjump signature standard
Date: Fri, 27 Mar 2020 10:27:03 -0700	[thread overview]
Message-ID: <20200327172703.GA28580@ubuntu-m2-xlarge-x86> (raw)
In-Reply-To: <CAKwvOdmLmfJY4Uk-Atd9dT5+zQTPeoagjMZMcDqdVfKCU7_BuA@mail.gmail.com>

On Fri, Mar 27, 2020 at 10:10:44AM -0700, Nick Desaulniers wrote:
> On Fri, Mar 27, 2020 at 3:08 AM Clement Courbet <courbet@google.com> wrote:
> >
> > Declaring setjmp()/longjmp() as taking longs makes the signature
> > non-standard, and makes clang complain. In the past, this has been
> > worked around by adding -ffreestanding to the compile flags.
> >
> > The implementation looks like it only ever propagates the value
> > (in longjmp) or sets it to 1 (in setjmp), and we only call longjmp
> > with integer parameters.
> >
> > This allows removing -ffreestanding from the compilation flags.
> >
> > Context:
> > https://lore.kernel.org/patchwork/patch/1214060
> > https://lore.kernel.org/patchwork/patch/1216174
> >
> > Signed-off-by: Clement Courbet <courbet@google.com>

Thanks for fixing this properly, not really sure why I did not think of
this in the first place. I guess my thought was the warning makes it
seem like clang is going to ignore the kernel's implementation of
setjmp/longjmp but I can't truly remember.

> Hi Clement, thanks for the patch! Would you mind sending a V2 that
> included a similar fix to arch/powerpc/xmon/Makefile?

Agreed.

> For context, this was the original patch:
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=aea447141c7e7824b81b49acd1bc78
> which was then modified to:
> https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=c9029ef9c95765e7b63c4d9aa780674447db1ec0
> 
> So on your V2, if you include in the commit message, the line:
> 
> Fixes c9029ef9c957 ("powerpc: Avoid clang warnings around setjmp and longjmp")
> 
> then that will help our LTS branch maintainers back port it to the
> appropriate branches.

The tags should be:

Cc: stable@vger.kernel.org # v4.14+
Fixes: c9029ef9c957 ("powerpc: Avoid clang warnings around setjmp and longjmp")

that way it explicitly gets picked up for stable, rather than Sasha's
AUTOSEL process, which could miss it.

With the xmon/Makefile -ffreestanding removed and the tags updated,
consider this:

Reviewed-by: Nathan Chancellor <natechancellor@gmail.com>
Tested-by: Nathan Chancellor <natechancellor@gmail.com>

Cheers,
Nathan