From: Jani Nikula <jani.nikula@linux.intel.com>
To: Alex Ivanov <gnidorah@ya.ru>, intel-gfx@lists.freedesktop.org
Subject: Re: [PATCH] drm/i915: Allow user to override PWM backlight frequency and duty cycle
Date: Wed, 17 Jan 2018 15:12:51 +0200 [thread overview]
Message-ID: <87y3kw39yk.fsf@intel.com> (raw)
In-Reply-To: <20180116134039.24178-1-gnidorah@ya.ru>
On Tue, 16 Jan 2018, Alex Ivanov <gnidorah@ya.ru> wrote:
> Since vendor may set bad defaults or user just generally may prefer her health over
> display life.
> 1. Allow overriding of PWM frequency otherwise people use solutions like
> http://devbraindom.blogspot.ru/2013/03/eliminate-led-screen-flicker-with-intel.html
> which are bad because
> I. frequency convertation logic varies from chip to chip, so people set totally wrong
> frequences in most cases
> II. writing to backlight control registers directly goes into conflict with
> i915 driver doing the same
> 2. Allow to override minimal brightness. Even if vendor limit was totally correct,
> it may be too bright to use in night, plus the default will become most likely wrong
> now, since frequency was overriden
I'm divided. Clearly, the patch at hand is a technically better solution
than what the blog post has.
But I also think the folks over at the blog know they're directly
hacking on graphics registers, and whatever it is they're doing is not
supported. There's also warranted speculation this might be harmful to
their displays.
The modulation frequency and the minimum duty cycle have been chosen by
the OEM to work with their board design and display specs. We don't have
enough data to validate the values given by the user are within spec. If
we add this, it'll get used, it'll be expected to be supported, the
"unsafe" there will go unnoticed by folks copy-pasting the parameters
from blogs and forums, and if it breaks displays for folks, they'll be
angry at us, not at some random blog poster.
I guess damned if you do, damned if you don't.
BR,
Jani.
>
> ---
> drivers/gpu/drm/i915/i915_params.c | 6 ++++++
> drivers/gpu/drm/i915/i915_params.h | 2 ++
> drivers/gpu/drm/i915/intel_panel.c | 17 +++++++++++++++++
> 3 files changed, 25 insertions(+)
>
> diff --git a/drivers/gpu/drm/i915/i915_params.c b/drivers/gpu/drm/i915/i915_params.c
> index b4faeb6aa2bd..ba37257846b0 100644
> --- a/drivers/gpu/drm/i915/i915_params.c
> +++ b/drivers/gpu/drm/i915/i915_params.c
> @@ -190,3 +190,9 @@ i915_param_named(enable_dpcd_backlight, bool, 0600,
>
> i915_param_named(enable_gvt, bool, 0400,
> "Enable support for Intel GVT-g graphics virtualization host support(default:false)");
> +
> +i915_param_named_unsafe(backlight_freq, uint, 0400,
> + "Override PWM backlight frequency (set in Hz, 0=ignore [default])");
> +
> +i915_param_named_unsafe(backlight_min_level, uint, 0400,
> + "Override PWM backlight minimum level (duty cycle, 0=ignore [default])");
> diff --git a/drivers/gpu/drm/i915/i915_params.h b/drivers/gpu/drm/i915/i915_params.h
> index c7292268ed43..7aba85db3da2 100644
> --- a/drivers/gpu/drm/i915/i915_params.h
> +++ b/drivers/gpu/drm/i915/i915_params.h
> @@ -53,6 +53,8 @@
> param(int, edp_vswing, 0) \
> param(int, reset, 2) \
> param(unsigned int, inject_load_failure, 0) \
> + param(unsigned int, backlight_freq, 0) \
> + param(unsigned int, backlight_min_level, 0) \
> /* leave bools at the end to not create holes */ \
> param(bool, alpha_support, IS_ENABLED(CONFIG_DRM_I915_ALPHA_SUPPORT)) \
> param(bool, enable_cmd_parser, true) \
> diff --git a/drivers/gpu/drm/i915/intel_panel.c b/drivers/gpu/drm/i915/intel_panel.c
> index adc51e452e3e..f188d7c1d2de 100644
> --- a/drivers/gpu/drm/i915/intel_panel.c
> +++ b/drivers/gpu/drm/i915/intel_panel.c
> @@ -1479,6 +1479,12 @@ static u32 get_backlight_min_vbt(struct intel_connector *connector)
>
> WARN_ON(panel->backlight.max == 0);
>
> + if (i915_modparams.backlight_min_level) {
> + DRM_DEBUG_KMS("Override backlight duty cycle %u\n",
> + i915_modparams.backlight_min_level);
> + return i915_modparams.backlight_min_level;
> + }
> +
> /*
> * XXX: If the vbt value is 255, it makes min equal to max, which leads
> * to problems. There are such machines out there. Either our
> @@ -1795,6 +1801,7 @@ int intel_panel_setup_backlight(struct drm_connector *connector, enum pipe pipe)
> struct intel_connector *intel_connector = to_intel_connector(connector);
> struct intel_panel *panel = &intel_connector->panel;
> int ret;
> + u32 pwm;
>
> if (!dev_priv->vbt.backlight.present) {
> if (dev_priv->quirks & QUIRK_BACKLIGHT_PRESENT) {
> @@ -1812,6 +1819,16 @@ int intel_panel_setup_backlight(struct drm_connector *connector, enum pipe pipe)
> /* set level and max in panel struct */
> mutex_lock(&dev_priv->backlight_lock);
> ret = panel->backlight.setup(intel_connector, pipe);
> + if (i915_modparams.backlight_freq) {
> + if (panel->backlight.hz_to_pwm &&
> + (pwm = panel->backlight.hz_to_pwm(intel_connector, i915_modparams.backlight_freq))) {
> + panel->backlight.max = pwm;
> + DRM_DEBUG_KMS("Override backlight frequency %u Hz\n",
> + i915_modparams.backlight_freq);
> + } else {
> + DRM_DEBUG_KMS("failed to override backlight frequency\n");
> + }
> + }
> mutex_unlock(&dev_priv->backlight_lock);
>
> if (ret) {
--
Jani Nikula, Intel Open Source Technology Center
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
next prev parent reply other threads:[~2018-01-17 13:12 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-01-16 13:40 [PATCH] drm/i915: Allow user to override PWM backlight frequency and duty cycle Alex Ivanov
2018-01-16 14:41 ` ✗ Fi.CI.BAT: failure for " Patchwork
2018-01-16 15:36 ` gnidorah
2018-01-16 19:26 ` Saarinen, Jani
2018-01-17 7:22 ` [PATCH] " Alex Ivanov
2018-01-17 7:46 ` ✓ Fi.CI.BAT: success for drm/i915: Allow user to override PWM backlight frequency and duty cycle (rev2) Patchwork
2018-01-17 8:59 ` ✗ Fi.CI.IGT: warning " Patchwork
2018-01-17 13:12 ` Jani Nikula [this message]
2018-01-17 17:49 ` [PATCH] drm/i915: Allow user to override PWM backlight frequency and duty cycle Alex Ivanov
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=87y3kw39yk.fsf@intel.com \
--to=jani.nikula@linux.intel.com \
--cc=gnidorah@ya.ru \
--cc=intel-gfx@lists.freedesktop.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox