Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
References: <CAHk-=wim8DMRzjyYTJ3UbdqZ26keQyZSU02NZb-JY1=9OpcO1w@mail.gmail.com>
 <20230102225656.GA3532398@roeck-us.net>
In-Reply-To: <20230102225656.GA3532398@roeck-us.net>
From:   Linus Torvalds <torvalds@linux-foundation.org>
Date:   Mon, 2 Jan 2023 16:21:41 -0800
Message-ID: <CAHk-=wjZPPscjDhsHQw_ttHOaQS69rADLm0KuRhbNavBiO62OQ@mail.gmail.com>
Subject: Re: Linux 6.2-rc2
To:     Guenter Roeck <linux@roeck-us.net>,
        "Jason A. Donenfeld" <Jason@zx2c4.com>,
        Yoshinori Sato <ysato@users.sourceforge.jp>,
        Rich Felker <dalias@libc.org>, Arnd Bergmann <arnd@arndb.de>
Cc:     Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

[ Adding Jason in case he has any ideas, and seeing if sh maintainer
emails are still valid, and Arnd in case they aren't ]

On Mon, Jan 2, 2023 at 2:57 PM Guenter Roeck <linux@roeck-us.net> wrote:
>
> One detail to mention, though, is that sh:rts7751r2dplus_defconfig
> no longer builds with older versions of binutils (2.32). Trying to
> do so results in the following build error.
>
> `.exit.text' referenced in section `__bug_table' of drivers/char/hw_random/core.o:
>         defined in discarded section `.exit.text' of drivers/char/hw_random/core.o
>
> To make this more interesting, kernels older than v5.10 do not boot
> (at least not in qemu) when images are built with binutils 2.27 or newer.
> That is why I had used binutils 2.32 in the first place.
>
> I didn't bother tracking this down but switched to binutils 2.39 when
> building v5.10+ images.

I have to admit that I can't really see myself carding deeply about
SH, but somebody else may. I don't think I've gotten an arch/sh pull
in a couple of years.

That said, I also don't see anything wrong with the arch/sh version of
BUG() and friends, so I don't see why this would hit arch/sh and not
somebody else.

I _assume_ it is the BUG_ON() in hwrng_modexit() that triggers this:

  static void __exit hwrng_modexit(void)
  {
        mutex_lock(&rng_mutex);
        BUG_ON(current_rng);
        kfree(rng_buffer);
        ...

but again, I don't see what's special about sh here apart from maybe
"not well maintained binutils support".

Does removing the BUG_ON() fix the build?

None of this is at all new, though. Funky.

              Linus