Received-SPF: softfail (google.com: domain of transitioning linux-kernel-owner@vger.kernel.org does not designate 23.128.96.19 as permitted sender) client-ip=23.128.96.19;
Date:   Wed, 6 Apr 2022 12:06:12 +0200
From:   Maxime Ripard <maxime@cerno.tech>
To:     "Zhengbin (OSKernel)" <zhengbin13@huawei.com>
Cc:     emma@anholt.net, airlied@linux.ie, daniel@ffwll.ch,
        dri-devel@lists.freedesktop.org, linux-kernel@vger.kernel.org,
        tangyizhou@huawei.com
Subject: Re: [PATCH v2 -next] drm/vc4: Fix build error when CONFIG_DRM_VC4=y
 && CONFIG_RASPBERRYPI_FIRMWARE=m
Message-ID: <20220406100612.2nuvjcc2oxo54n5j@houat>
References: <20220325021831.2812735-1-zhengbin13@huawei.com>
 <20220325130541.tjxq4kjgmkaz32en@houat>
 <601a9682-6afc-d87b-cfa7-5c683271f7a5@huawei.com>
MIME-Version: 1.0
Content-Type: multipart/signed; micalg=pgp-sha512;
        protocol="application/pgp-signature"; boundary="m2ydzgbnjwzzthnm"
Content-Disposition: inline
In-Reply-To: <601a9682-6afc-d87b-cfa7-5c683271f7a5@huawei.com>
Precedence: bulk


--m2ydzgbnjwzzthnm
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: quoted-printable

Hi,

On Mon, Mar 28, 2022 at 10:22:11AM +0800, Zhengbin (OSKernel) wrote:
> On 2022/3/25 21:05, Maxime Ripard wrote:
> > On Fri, Mar 25, 2022 at 10:18:31AM +0800, Zheng Bin wrote:
> > > If CONFIG_DRM_VC4=3Dy, CONFIG_RASPBERRYPI_FIRMWARE=3Dm, CONFIG_COMPIL=
E_TEST=3Dn,
> > > bulding fails:
> > >=20
> > > drivers/gpu/drm/vc4/vc4_drv.o: In function `vc4_drm_bind':
> > > vc4_drv.c:(.text+0x320): undefined reference to `rpi_firmware_get'
> > > vc4_drv.c:(.text+0x320): relocation truncated to fit: R_AARCH64_CALL2=
6 against undefined symbol `rpi_firmware_get'
> > > vc4_drv.c:(.text+0x34c): undefined reference to `rpi_firmware_propert=
y'
> > > vc4_drv.c:(.text+0x34c): relocation truncated to fit: R_AARCH64_CALL2=
6 against undefined symbol `rpi_firmware_property'
> > > vc4_drv.c:(.text+0x354): undefined reference to `rpi_firmware_put'
> > > vc4_drv.c:(.text+0x354): relocation truncated to fit: R_AARCH64_CALL2=
6 against undefined symbol `rpi_firmware_put'
> > >=20
> > > Make DRM_VC4 depends on RASPBERRYPI_FIRMWARE || COMPILE_TEST to fix t=
his.
> > >=20
> > > Fixes: c406ad5e4a85 ("drm/vc4: Notify the firmware when DRM is in cha=
rge")
> > > Reported-by: Hulk Robot <hulkci@huawei.com>
> > > Signed-off-by: Zheng Bin <zhengbin13@huawei.com>
> > > ---
> > > v2: follow Maxime's advise, add RASPBERRYPI_FIRMWARE as an additional=
 depends
> > >   drivers/gpu/drm/vc4/Kconfig | 1 +
> > >   1 file changed, 1 insertion(+)
> > >=20
> > > diff --git a/drivers/gpu/drm/vc4/Kconfig b/drivers/gpu/drm/vc4/Kconfig
> > > index de3424fed2fc..92d74ba7c449 100644
> > > --- a/drivers/gpu/drm/vc4/Kconfig
> > > +++ b/drivers/gpu/drm/vc4/Kconfig
> > > @@ -2,6 +2,7 @@
> > >   config DRM_VC4
> > >   	tristate "Broadcom VC4 Graphics"
> > >   	depends on ARCH_BCM || ARCH_BCM2835 || COMPILE_TEST
> > > +	depends on RASPBERRYPI_FIRMWARE || COMPILE_TEST
> > Why do we need the || COMPILE_TEST here?
> >=20
> > The rpi_firmware_get, _property and _put functions aren't define if
> > RASPBERRYPI_FIRMWARE, so we need that dependency every time, even if
> > COMPILE_TEST is set?
>=20
> include/soc/bcm2835/raspberrypi-firmware.h
>=20
> #if IS_ENABLED(CONFIG_RASPBERRYPI_FIRMWARE)
>=20
> struct rpi_firmware *rpi_firmware_get(struct device_node
> *firmware_node);/_property/_put
>=20
> #else=A0=A0=A0 -->empty function
>=20
> =A0static inline struct rpi_firmware *rpi_firmware_get(struct device_node
> *firmware_node)/_property/_put
> {
> =A0=A0=A0=A0=A0=A0=A0=A0 return NULL;
> =A0}
> #endif
>=20
>=20
> We can refer to drivers/pwm/Kconfig
>=20
> config PWM_RASPBERRYPI_POE
> =A0=A0=A0=A0=A0=A0=A0=A0 tristate "Raspberry Pi Firwmware PoE Hat PWM sup=
port"
> =A0=A0=A0=A0=A0=A0=A0=A0 # Make sure not 'y' when RASPBERRYPI_FIRMWARE is=
 'm'. This can only
> =A0=A0=A0=A0=A0=A0=A0 # happen when COMPILE_TEST=3Dy, hence the added !RA=
SPBERRYPI_FIRMWARE.
> =A0=A0=A0=A0=A0=A0=A0=A0 depends on RASPBERRYPI_FIRMWARE || (COMPILE_TEST=
 &&
> !RASPBERRYPI_FIRMWARE)=A0 -->There is no need to add (&&
> !RASPBERRYPI_FIRMWARE)

Oh, so you're fixing the case where vc4 would be built-in, but the
firmware driver a module?

I guess that makes sense, but the comment definitely helps there. And
the (COMPILE_TEST && !RASPBERRYPI_FIRMWARE) is also needed, since it
wouldn't fix the issue you're trying to fix for COMPILE_TEST otherwise.

Maxime

--m2ydzgbnjwzzthnm
Content-Type: application/pgp-signature; name="signature.asc"

-----BEGIN PGP SIGNATURE-----

iHUEABYKAB0WIQRcEzekXsqa64kGDp7j7w1vZxhRxQUCYk1mFAAKCRDj7w1vZxhR
xT80AQC7Tm+NmNDitAW/WPTe7zEurlBuKyI80CaozozVmnmeUgD/ZdkwChF2xf3c
k1aWDhJy5q1hEMUagUb6QAuZLV0lMAM=
=wzKX
-----END PGP SIGNATURE-----

--m2ydzgbnjwzzthnm--