Received-SPF: pass (google.com: domain of linux-kernel+bounces-103049-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45e3:2400::1 as permitted sender) client-ip=2604:1380:45e3:2400::1;
Precedence: bulk
MIME-Version: 1.0
References: <CA+G9fYvG9KE15PGNoLu+SBVyShe+u5HBLQ81+kK9Zop6u=ywmw@mail.gmail.com>
 <338c89bb-a70b-4f35-b71b-f974e90e3383@app.fastmail.com> <20240304-brawny-goshawk-of-sorcery-860cef@houat>
 <85b807289ff2400ea5887ced63655862@AcuMS.aculab.com>
In-Reply-To: <85b807289ff2400ea5887ced63655862@AcuMS.aculab.com>
From: Geert Uytterhoeven <geert@linux-m68k.org>
Date: Thu, 14 Mar 2024 10:27:23 +0100
Message-ID: <CAMuHMdUuOnXVpocYU02Mx3_KrPow-=+WgLJqy_ku=AN52SvQiA@mail.gmail.com>
Subject: Re: arm: ERROR: modpost: "__aeabi_uldivmod" [drivers/gpu/drm/sun4i/sun4i-drm-hdmi.ko]
 undefined!
To: David Laight <David.Laight@aculab.com>
Cc: Maxime Ripard <mripard@kernel.org>, Arnd Bergmann <arnd@arndb.de>, 
	Naresh Kamboju <naresh.kamboju@linaro.org>, open list <linux-kernel@vger.kernel.org>, 
	Linux ARM <linux-arm-kernel@lists.infradead.org>, 
	"linux-sunxi@lists.linux.dev" <linux-sunxi@lists.linux.dev>, 
	"dri-devel@lists.freedesktop.org" <dri-devel@lists.freedesktop.org>, 
	"lkft-triage@lists.linaro.org" <lkft-triage@lists.linaro.org>, Dave Airlie <airlied@redhat.com>, 
	Dan Carpenter <dan.carpenter@linaro.org>, Ard Biesheuvel <ardb@kernel.org>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

On Sat, Mar 9, 2024 at 3:34=E2=80=AFPM David Laight <David.Laight@aculab.co=
m> wrote:
> From: Maxime Ripard
> > Sent: 04 March 2024 11:46
> >
> > On Mon, Mar 04, 2024 at 12:11:36PM +0100, Arnd Bergmann wrote:
> > > On Mon, Mar 4, 2024, at 09:07, Naresh Kamboju wrote:
> > > > The arm defconfig builds failed on today's Linux next tag next-2024=
0304.
> > > >
> > > > Build log:
> > > > ---------
> > > > ERROR: modpost: "__aeabi_uldivmod"
> > > > [drivers/gpu/drm/sun4i/sun4i-drm-hdmi.ko] undefined!
> > > >
> > >
> > > Apparently caused by the 64-bit division in 358e76fd613a
> > > ("drm/sun4i: hdmi: Consolidate atomic_check and mode_valid"):
> > >
> > >
> > > +static enum drm_mode_status
> > > +sun4i_hdmi_connector_clock_valid(const struct drm_connector *connect=
or,
> > > +                                const struct drm_display_mode *mode,
> > > +                                unsigned long long clock)
> > >  {
> > > -       struct sun4i_hdmi *hdmi =3D drm_encoder_to_sun4i_hdmi(encoder=
);
> > > -       unsigned long rate =3D mode->clock * 1000;
> > > -       unsigned long diff =3D rate / 200; /* +-0.5% allowed by HDMI =
spec */
> > > +       const struct sun4i_hdmi *hdmi =3D drm_connector_to_sun4i_hdmi=
(connector);
> > > +       unsigned long diff =3D clock / 200; /* +-0.5% allowed by HDMI=
 spec */
> > >         long rounded_rate;
> > >
> > > This used to be a 32-bit division. If the rate is never more than
> > > 4.2GHz, clock could be turned back into 'unsigned long' to avoid
> > > the expensive div_u64().
> >
> > I sent a fix for it this morning:
> > https://lore.kernel.org/r/20240304091225.366325-1-mripard@kernel.org
> >
> > The framework will pass an unsigned long long because HDMI character
> > rates can go up to 5.9GHz.
>
> You could do:
>         /* The max clock is 5.9GHz, split the divide */
>         u32 diff =3D (u32)(clock / 8) / (200/8);

+1, as the issue is still present in current next, as per the recent
nagging from the build bots.

> The code should really use u32 and u64.
> Otherwise the sizes are different on 32bit.

The code is already using a variety of types (long, unsigned long long,
unsigned long) :-(

       max_t(unsigned long, rounded_rate, clock) -
       min_t(unsigned long, rounded_rate, clock) < diff)

At least u64 should make it very clear clock does not fit in 32-bit.

Gr{oetje,eeting}s,

                        Geert

--=20
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k=
org

In personal conversations with technical people, I call myself a hacker. Bu=
t
when I'm talking to journalists I just say "programmer" or something like t=
hat.
                                -- Linus Torvalds