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;
From:   Alexander Stein <alexander.stein@ew.tq-group.com>
To:     Andrzej Hajda <andrzej.hajda@intel.com>,
        Daniel Vetter <daniel@ffwll.ch>,
        David Airlie <airlied@gmail.com>,
        dri-devel@lists.freedesktop.org, Inki Dae <inki.dae@samsung.com>,
        Jagan Teki <jagan@amarulasolutions.com>,
        linux-kernel@vger.kernel.org,
        Marek Szyprowski <m.szyprowski@samsung.com>,
        Neil Armstrong <neil.armstrong@linaro.org>,
        Robert Foss <rfoss@kernel.org>
Cc:     Marek Vasut <marex@denx.de>,
        Laurent Pinchart <Laurent.pinchart@ideasonboard.com>,
        Jernej Skrabec <jernej.skrabec@gmail.com>,
        Frieder Schrempf <frieder.schrempf@kontron.de>,
        Jonas Karlman <jonas@kwiboo.se>,
        Frieder Schrempf <frieder@fris.de>
Subject: Re: [PATCH v2 1/2] drm: bridge: samsung-dsim: Fix i.MX8M enable flow to meet spec
Date:   Thu, 04 May 2023 11:07:24 +0200
Message-ID: <8215519.T7Z3S40VBb@steina-w>
Organization: TQ-Systems GmbH
In-Reply-To: <20230503163313.2640898-2-frieder@fris.de>
References: <20230503163313.2640898-1-frieder@fris.de> <20230503163313.2640898-2-frieder@fris.de>
MIME-Version: 1.0
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain; charset="iso-8859-1"
Precedence: bulk

Hello Frieder,

Am Mittwoch, 3. Mai 2023, 18:33:06 CEST schrieb Frieder Schrempf:
> From: Frieder Schrempf <frieder.schrempf@kontron.de>
>=20
> According to the documentation [1] the proper enable flow is:
>=20
> 1. Enable DSI link and keep data lanes in LP-11 (stop state)
> 2. Disable stop state to bring data lanes into HS mode
>=20
> Currently we do this all at once within enable(), which doesn't
> allow to meet the requirements of some downstream bridges.
>=20
> To fix this we now enable the DSI in pre_enable() and force it
> into stop state using the FORCE_STOP_STATE bit in the ESCMODE
> register until enable() is called where we reset the bit.
>=20
> We currently do this only for i.MX8M as Exynos uses a different
> init flow where samsung_dsim_init() is called from
> samsung_dsim_host_transfer().
>=20
> [1]
> https://docs.kernel.org/gpu/drm-kms-helpers.html#mipi-dsi-bridge-operation
>=20
> Signed-off-by: Frieder Schrempf <frieder.schrempf@kontron.de>
> ---
> Changes for v2:
> * Drop RFC
> ---
>  drivers/gpu/drm/bridge/samsung-dsim.c | 25 +++++++++++++++++++++++--
>  1 file changed, 23 insertions(+), 2 deletions(-)
>=20
> diff --git a/drivers/gpu/drm/bridge/samsung-dsim.c
> b/drivers/gpu/drm/bridge/samsung-dsim.c index e0a402a85787..9775779721d9
> 100644
> --- a/drivers/gpu/drm/bridge/samsung-dsim.c
> +++ b/drivers/gpu/drm/bridge/samsung-dsim.c
> @@ -859,6 +859,10 @@ static int samsung_dsim_init_link(struct samsung_dsim
> *dsi) reg =3D samsung_dsim_read(dsi, DSIM_ESCMODE_REG);
>  	reg &=3D ~DSIM_STOP_STATE_CNT_MASK;
>  	reg |=3D DSIM_STOP_STATE_CNT(driver_data->reg_values[STOP_STATE_CNT]);
> +
> +	if (!samsung_dsim_hw_is_exynos(dsi->plat_data->hw_type))
> +		reg |=3D DSIM_FORCE_STOP_STATE;
> +
>  	samsung_dsim_write(dsi, DSIM_ESCMODE_REG, reg);
>=20
>  	reg =3D DSIM_BTA_TIMEOUT(0xff) | DSIM_LPDR_TIMEOUT(0xffff);
> @@ -1340,6 +1344,9 @@ static void samsung_dsim_atomic_pre_enable(struct
> drm_bridge *bridge, ret =3D samsung_dsim_init(dsi);
>  		if (ret)
>  			return;
> +
> +		samsung_dsim_set_display_mode(dsi);
> +		samsung_dsim_set_display_enable(dsi, true);
>  	}
>  }
>=20
> @@ -1347,9 +1354,16 @@ static void samsung_dsim_atomic_enable(struct
> drm_bridge *bridge, struct drm_bridge_state *old_bridge_state)
>  {
>  	struct samsung_dsim *dsi =3D bridge_to_dsi(bridge);
> +	u32 reg;
>=20
> -	samsung_dsim_set_display_mode(dsi);
> -	samsung_dsim_set_display_enable(dsi, true);
> +	if (samsung_dsim_hw_is_exynos(dsi->plat_data->hw_type)) {
> +		samsung_dsim_set_display_mode(dsi);
> +		samsung_dsim_set_display_enable(dsi, true);
> +	} else {
> +		reg =3D samsung_dsim_read(dsi, DSIM_ESCMODE_REG);
> +		reg &=3D ~DSIM_FORCE_STOP_STATE;
> +		samsung_dsim_write(dsi, DSIM_ESCMODE_REG, reg);
> +	}
>=20
>  	dsi->state |=3D DSIM_STATE_VIDOUT_AVAILABLE;
>  }
> @@ -1358,10 +1372,17 @@ static void samsung_dsim_atomic_disable(struct
> drm_bridge *bridge, struct drm_bridge_state *old_bridge_state)
>  {
>  	struct samsung_dsim *dsi =3D bridge_to_dsi(bridge);
> +	u32 reg;
>=20
>  	if (!(dsi->state & DSIM_STATE_ENABLED))
>  		return;
>=20
> +	if (!samsung_dsim_hw_is_exynos(dsi->plat_data->hw_type)) {
> +		reg =3D samsung_dsim_read(dsi, DSIM_ESCMODE_REG);
> +		reg |=3D DSIM_FORCE_STOP_STATE;
> +		samsung_dsim_write(dsi, DSIM_ESCMODE_REG, reg);
> +	}
> +
>  	dsi->state &=3D ~DSIM_STATE_VIDOUT_AVAILABLE;
>  }

I know that this is necessary right now, but I don't like that=20
'samsung_dsim_hw_is_exynos()' checks all over the place.

Despite that:
Tested-by: Alexander Stein <alexander.stein@ew.tq-group.com> #TQMa8MxML/MBa=
8Mx

=2D-=20
TQ-Systems GmbH | M=FChlstra=DFe 2, Gut Delling | 82229 Seefeld, Germany
Amtsgericht M=FCnchen, HRB 105018
Gesch=E4ftsf=FChrer: Detlef Schneider, R=FCdiger Stahl, Stefan Schneider
http://www.tq-group.com/