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: <20230331030204.1179524-1-treapking@chromium.org>
In-Reply-To: <20230331030204.1179524-1-treapking@chromium.org>
From:   Doug Anderson <dianders@chromium.org>
Date:   Fri, 31 Mar 2023 13:29:11 -0700
Message-ID: <CAD=FV=XdthxVh=wTR5hLj3sy=X__nYQge5=SBXB=eAENe0bd1g@mail.gmail.com>
Subject: Re: [PATCH] drm/bridge: ps8640: Use constant sleep time for polling hpd
To:     Pin-yen Lin <treapking@chromium.org>
Cc:     Andrzej Hajda <andrzej.hajda@intel.com>,
        Neil Armstrong <neil.armstrong@linaro.org>,
        Robert Foss <rfoss@kernel.org>,
        Laurent Pinchart <Laurent.pinchart@ideasonboard.com>,
        Jonas Karlman <jonas@kwiboo.se>,
        Jernej Skrabec <jernej.skrabec@gmail.com>,
        dri-devel@lists.freedesktop.org, linux-kernel@vger.kernel.org,
        Daniel Vetter <daniel@ffwll.ch>,
        David Airlie <airlied@gmail.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

Hi,

On Thu, Mar 30, 2023 at 8:02=E2=80=AFPM Pin-yen Lin <treapking@chromium.org=
> wrote:
>
> The default hpd_wait_us in panel_edp.c is 2 seconds. This makes the
> sleep time in the polling of _ps8640_wait_hpd_asserted become 200ms.
> Change it to a constant 20ms to speed up the function.

Ah, I see why I never ran into this. All the panels I worked with
specified "hpd_absent" of 200 and thus I've always been using 20.


> Signed-off-by: Pin-yen Lin <treapking@chromium.org>
> ---
>
>  drivers/gpu/drm/bridge/parade-ps8640.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/bridge/parade-ps8640.c b/drivers/gpu/drm/bri=
dge/parade-ps8640.c
> index b823e55650b1..c3eb45179405 100644
> --- a/drivers/gpu/drm/bridge/parade-ps8640.c
> +++ b/drivers/gpu/drm/bridge/parade-ps8640.c
> @@ -184,7 +184,7 @@ static int _ps8640_wait_hpd_asserted(struct ps8640 *p=
s_bridge, unsigned long wai
>          * actually connected to GPIO9).
>          */
>         ret =3D regmap_read_poll_timeout(map, PAGE2_GPIO_H, status,
> -                                      status & PS_GPIO9, wait_us / 10, w=
ait_us);
> +                                      status & PS_GPIO9, 20000, wait_us)=
;

I'd have been tempted to go even lower at 10ms. Waiting for HPD isn't
something that we do all the time during a normal running system and
thus it's not something we have to optimize every last bit of power
out of. The user would generally rather have the system boot up or
switch modes 10ms faster. ;-)

In any case, either at 10ms or 20ms:

Reviewed-by: Douglas Anderson <dianders@chromium.org>

-Doug