linux-kernel - Re: [PATCH v4 24/27] drm/panel: panel-simple: Cache the EDID as long as we retain power

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <YILx/iODs+DFWWwm@builder.lan>
Date:   Fri, 23 Apr 2021 11:12:46 -0500
From:   Bjorn Andersson <bjorn.andersson@...aro.org>
To:     Douglas Anderson <dianders@...omium.org>
Cc:     Andrzej Hajda <a.hajda@...sung.com>,
        Neil Armstrong <narmstrong@...libre.com>,
        Laurent Pinchart <Laurent.pinchart@...asonboard.com>,
        Jonas Karlman <jonas@...boo.se>,
        Jernej Skrabec <jernej.skrabec@...l.net>,
        Sam Ravnborg <sam@...nborg.org>, Wolfram Sang <wsa@...nel.org>,
        Stephen Boyd <swboyd@...omium.org>, robdclark@...omium.org,
        Maarten Lankhorst <maarten.lankhorst@...ux.intel.com>,
        Stanislav Lisovskiy <stanislav.lisovskiy@...el.com>,
        Steev Klimaszewski <steev@...i.org>,
        linux-arm-msm@...r.kernel.org, Linus W <linus.walleij@...aro.org>,
        Daniel Vetter <daniel@...ll.ch>,
        David Airlie <airlied@...ux.ie>,
        Thierry Reding <thierry.reding@...il.com>,
        dri-devel@...ts.freedesktop.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v4 24/27] drm/panel: panel-simple: Cache the EDID as long
 as we retain power

On Fri 16 Apr 17:39 CDT 2021, Douglas Anderson wrote:

> It doesn't make sense to go out to the bus and read the EDID over and
> over again. Let's cache it and throw away the cache when we turn power
> off from the panel. Autosuspend means that even if there are several
> calls to read the EDID before we officially turn the power on then we
> should get good use out of this cache.
> 
> Signed-off-by: Douglas Anderson <dianders@...omium.org>
> ---
> 
> (no changes since v1)
> 
>  drivers/gpu/drm/panel/panel-simple.c | 17 ++++++++++-------
>  1 file changed, 10 insertions(+), 7 deletions(-)
> 
> diff --git a/drivers/gpu/drm/panel/panel-simple.c b/drivers/gpu/drm/panel/panel-simple.c
> index 40382c1be692..5a2953c4ca44 100644
> --- a/drivers/gpu/drm/panel/panel-simple.c
> +++ b/drivers/gpu/drm/panel/panel-simple.c
> @@ -189,6 +189,8 @@ struct panel_simple {
>  	struct gpio_desc *enable_gpio;
>  	struct gpio_desc *hpd_gpio;
>  
> +	struct edid *edid;
> +
>  	struct drm_display_mode override_mode;
>  
>  	enum drm_panel_orientation orientation;
> @@ -345,6 +347,9 @@ static int panel_simple_suspend(struct device *dev)
>  	regulator_disable(p->supply);
>  	p->unprepared_time = ktime_get();
>  
> +	kfree(p->edid);
> +	p->edid = NULL;

Reviewed-by: Bjorn Andersson <bjorn.andersson@...aro.org>


But separate of this, shouldn't the driver have a pm_runtime_disable()
in the remove path to synchronize the autosleep? Or is that not how that
works?

Regards,
Bjorn

> +
>  	return 0;
>  }
>  
> @@ -510,15 +515,13 @@ static int panel_simple_get_modes(struct drm_panel *panel,
>  
>  	/* probe EDID if a DDC bus is available */
>  	if (p->ddc) {
> -		struct edid *edid;
> -
>  		pm_runtime_get_sync(panel->dev);
>  
> -		edid = drm_get_edid(connector, p->ddc);
> -		if (edid) {
> -			num += drm_add_edid_modes(connector, edid);
> -			kfree(edid);
> -		}
> +		if (!p->edid)
> +			p->edid = drm_get_edid(connector, p->ddc);
> +
> +		if (p->edid)
> +			num += drm_add_edid_modes(connector, p->edid);
>  
>  		pm_runtime_mark_last_busy(panel->dev);
>  		pm_runtime_put_autosuspend(panel->dev);
> -- 
> 2.31.1.368.gbe11c130af-goog
>