public inbox for intel-gfx@lists.freedesktop.org
 help / color / mirror / Atom feed
From: Paulo Zanoni <przanoni@gmail.com>
To: intel-gfx@lists.freedesktop.org
Cc: Paulo Zanoni <paulo.r.zanoni@intel.com>
Subject: [PATCH 18/19] drm/i915: save some time when waiting the eDP timings
Date: Thu, 21 Nov 2013 13:47:32 -0200	[thread overview]
Message-ID: <1385048853-1579-19-git-send-email-przanoni@gmail.com> (raw)
In-Reply-To: <1385048853-1579-1-git-send-email-przanoni@gmail.com>

From: Paulo Zanoni <paulo.r.zanoni@intel.com>

The eDP spec defines some points where after you do action A, you have
to wait some time before action B. The thing is that in our driver
action B does not happen exactly after action A, but we still use
msleep() calls directly. What this patch happens is that we record the
timestamp of when action A happened, then, just before action B, we
look at how much time has passed and only sleep the remaining amount
needed.

With this change, I am able to save about 5-20ms (out of the total
200ms) of the backlight_off delay and completely skip the 1ms
backlight_on delay. The 600ms vdd_off delay doesn't happen during
normal usage anymore due to a previous patch.

Signed-off-by: Paulo Zanoni <paulo.r.zanoni@intel.com>
---
 drivers/gpu/drm/i915/intel_dp.c  | 38 +++++++++++++++++++++++++++++++++++---
 drivers/gpu/drm/i915/intel_drv.h |  3 +++
 2 files changed, 38 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index b438e76..3a1ca80 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -1051,12 +1051,41 @@ static void ironlake_wait_panel_off(struct intel_dp *intel_dp)
 	ironlake_wait_panel_status(intel_dp, IDLE_OFF_MASK, IDLE_OFF_VALUE);
 }
 
+static void ironlake_wait_jiffies_delay(unsigned long timestamp,
+					int to_wait_ms)
+{
+	unsigned long target = timestamp + msecs_to_jiffies(to_wait_ms);
+	unsigned long diff;
+
+	if (time_after(target, jiffies)) {
+		diff = (long)target - (long)jiffies;
+		msleep(diff);
+	}
+}
+
 static void ironlake_wait_panel_power_cycle(struct intel_dp *intel_dp)
 {
 	DRM_DEBUG_KMS("Wait for panel power cycle\n");
+
+	/* When we disable the VDD override bit last we have to do the manual
+	 * wait. */
+	ironlake_wait_jiffies_delay(intel_dp->last_power_cycle,
+				    intel_dp->panel_power_cycle_delay);
+
 	ironlake_wait_panel_status(intel_dp, IDLE_CYCLE_MASK, IDLE_CYCLE_VALUE);
 }
 
+static void ironlake_wait_backlight_on(struct intel_dp *intel_dp)
+{
+	ironlake_wait_jiffies_delay(intel_dp->last_power_on,
+				    intel_dp->backlight_on_delay);
+}
+
+static void ironlake_edp_wait_backlight_off(struct intel_dp *intel_dp)
+{
+	ironlake_wait_jiffies_delay(intel_dp->last_backlight_off,
+				    intel_dp->backlight_off_delay);
+}
 
 /* Read the current pp_control value, unlocking the register if it
  * is locked
@@ -1144,7 +1173,7 @@ static void ironlake_panel_vdd_off_sync(struct intel_dp *intel_dp)
 		I915_READ(pp_stat_reg), I915_READ(pp_ctrl_reg));
 
 		if ((pp & POWER_TARGET_ON) == 0)
-			msleep(intel_dp->panel_power_cycle_delay);
+			intel_dp->last_power_cycle = jiffies;
 
 		intel_runtime_pm_put(dev_priv);
 	}
@@ -1219,6 +1248,7 @@ void ironlake_edp_panel_on(struct intel_dp *intel_dp)
 	POSTING_READ(pp_ctrl_reg);
 
 	ironlake_wait_panel_on(intel_dp);
+	intel_dp->last_power_on = jiffies;
 
 	if (IS_GEN5(dev)) {
 		pp |= PANEL_POWER_RESET; /* restore panel reset bit */
@@ -1239,6 +1269,8 @@ void ironlake_edp_panel_off(struct intel_dp *intel_dp)
 
 	DRM_DEBUG_KMS("Turn eDP power off\n");
 
+	ironlake_edp_wait_backlight_off(intel_dp);
+
 	pp = ironlake_get_pp_control(intel_dp);
 	/* We need to switch off panel power _and_ force vdd, for otherwise some
 	 * panels get very unhappy and cease to work. */
@@ -1270,7 +1302,7 @@ void ironlake_edp_backlight_on(struct intel_dp *intel_dp)
 	 * link.  So delay a bit to make sure the image is solid before
 	 * allowing it to appear.
 	 */
-	msleep(intel_dp->backlight_on_delay);
+	ironlake_wait_backlight_on(intel_dp);
 	pp = ironlake_get_pp_control(intel_dp);
 	pp |= EDP_BLC_ENABLE;
 
@@ -1302,7 +1334,7 @@ void ironlake_edp_backlight_off(struct intel_dp *intel_dp)
 
 	I915_WRITE(pp_ctrl_reg, pp);
 	POSTING_READ(pp_ctrl_reg);
-	msleep(intel_dp->backlight_off_delay);
+	intel_dp->last_backlight_off = jiffies;
 }
 
 static void ironlake_edp_pll_on(struct intel_dp *intel_dp)
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index 5596498..778a7be 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -484,6 +484,9 @@ struct intel_dp {
 	int backlight_off_delay;
 	struct delayed_work panel_vdd_work;
 	bool want_panel_vdd;
+	unsigned long last_power_cycle;
+	unsigned long last_power_on;
+	unsigned long last_backlight_off;
 	bool psr_setup_done;
 	struct intel_connector *attached_connector;
 };
-- 
1.8.3.1

  parent reply	other threads:[~2013-11-21 15:48 UTC|newest]

Thread overview: 97+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-11-21 15:47 [PATCH 00/19] Haswell runtime PM support + D3 Paulo Zanoni
2013-11-21 15:47 ` [PATCH 01/19] drm/i915: WARN if !HAS_PC8 when enabling/disabling PC8 Paulo Zanoni
2013-11-29 11:11   ` Rodrigo Vivi
2013-11-29 12:55     ` Paulo Zanoni
2013-11-29 13:31       ` Rodrigo Vivi
2013-11-21 15:47 ` [PATCH 02/19] drm/i915: use the correct force_wake function at the PC8 code Paulo Zanoni
2013-11-27 19:57   ` Paulo Zanoni
2013-11-29 11:14     ` [Intel-gfx] " Rodrigo Vivi
2013-11-29 13:23       ` Daniel Vetter
2013-11-21 15:47 ` [PATCH 03/19] drm/i915: get a PC8 reference when enabling the power well Paulo Zanoni
2013-11-27 19:59   ` Paulo Zanoni
2013-11-29 12:35     ` Rodrigo Vivi
2013-11-29 13:34       ` Rodrigo Vivi
2013-12-10 21:29     ` Daniel Vetter
2013-11-21 15:47 ` [PATCH 04/19] drm/i915: get/put PC8 when we get/put a CRTC Paulo Zanoni
2013-11-21 16:12   ` Chris Wilson
2013-11-27 20:01     ` Paulo Zanoni
2013-11-29 12:38       ` Rodrigo Vivi
2013-11-29 13:34         ` Rodrigo Vivi
2013-12-04  9:01   ` Daniel Vetter
2013-12-04 13:44     ` Paulo Zanoni
2013-12-04 14:07       ` Daniel Vetter
2013-12-05 13:43         ` Paulo Zanoni
2013-12-05 14:40           ` Daniel Vetter
2013-12-06 22:29             ` [PATCH] drm/i915: change CRTC assertion on LCPLL disable Paulo Zanoni
2013-12-06 22:37               ` Daniel Vetter
2013-11-21 15:47 ` [PATCH 05/19] drm/i915: add initial Runtime PM functions Paulo Zanoni
2013-11-27 20:10   ` Paulo Zanoni
2013-11-29 12:54     ` Rodrigo Vivi
2013-11-29 13:33       ` Rodrigo Vivi
2013-11-29 14:05     ` Takashi Iwai
2013-12-06 22:31       ` Paulo Zanoni
2013-12-06 22:32         ` Paulo Zanoni
2013-12-08  9:06         ` Takashi Iwai
2013-12-02 12:23   ` Imre Deak
2013-11-21 15:47 ` [PATCH 06/19] drm/i915: do adapter power state notification at runtime PM Paulo Zanoni
2013-11-21 16:14   ` Chris Wilson
2013-11-27 20:13     ` Paulo Zanoni
2013-11-29 12:56       ` Rodrigo Vivi
2013-11-29 13:33         ` Rodrigo Vivi
2013-12-06 22:34         ` Paulo Zanoni
2013-11-21 15:47 ` [PATCH 07/19] drm/i915: add runtime put/get calls at the basic places Paulo Zanoni
2013-11-21 16:07   ` Chris Wilson
2013-11-25 20:55     ` Paulo Zanoni
2013-11-25 21:21       ` Chris Wilson
2013-11-27 20:20         ` Paulo Zanoni
2013-11-29 13:03           ` Rodrigo Vivi
2013-11-29 13:32             ` Rodrigo Vivi
2013-12-10 21:49             ` Daniel Vetter
2013-12-12 20:07               ` Paulo Zanoni
2013-12-12 20:54                 ` Daniel Vetter
2013-11-21 15:47 ` [PATCH 08/19] drm/i915: add some runtime PM get/put calls Paulo Zanoni
2013-11-27 20:21   ` Paulo Zanoni
2013-11-29 13:05     ` Rodrigo Vivi
2013-11-29 13:31       ` Rodrigo Vivi
2013-11-29 13:42       ` Daniel Vetter
2013-11-29 13:56         ` Paulo Zanoni
2013-11-21 15:47 ` [PATCH 09/19] drm/i915: get a runtime PM reference when the panel VDD is on Paulo Zanoni
2013-11-29 13:50   ` Rodrigo Vivi
2013-11-29 13:59     ` Paulo Zanoni
2013-11-29 14:37       ` Rodrigo Vivi
2013-12-06 22:23         ` Paulo Zanoni
2013-11-21 15:47 ` [PATCH 10/19] drm/i915: do not assert DE_PCH_EVENT_IVB enabled Paulo Zanoni
2013-11-29 14:30   ` Rodrigo Vivi
2013-12-10 21:54   ` Daniel Vetter
2013-11-21 15:47 ` [PATCH 11/19] drm/i915: disable interrupts when enabling PC8 Paulo Zanoni
2013-12-02 13:33   ` Rodrigo Vivi
2013-12-10 21:59   ` Daniel Vetter
2013-12-11 21:33     ` Paulo Zanoni
2013-11-21 15:47 ` [PATCH 12/19] drm/i915: release the GTT mmaps when going into D3 Paulo Zanoni
2013-11-21 16:02   ` Chris Wilson
2013-11-21 16:27     ` Paulo Zanoni
2013-12-10 22:03   ` Daniel Vetter
2013-11-21 15:47 ` [PATCH 13/19] drm: do not steal the display if we have a master Paulo Zanoni
2013-11-21 16:04   ` Chris Wilson
2013-11-27 20:24     ` Paulo Zanoni
2013-11-29 13:37       ` Daniel Vetter
2013-11-30 11:19       ` David Herrmann
2013-11-21 15:47 ` [PATCH 14/19] drm/i915: add runtime PM support on Haswell Paulo Zanoni
2013-12-02 13:37   ` Rodrigo Vivi
2013-12-10 22:10     ` Daniel Vetter
2013-12-10 22:06   ` Daniel Vetter
2013-11-21 15:47 ` [PATCH 15/19] drm/i915: don't enable VDD just to enable the panel Paulo Zanoni
2013-11-29 14:40   ` Rodrigo Vivi
2013-11-21 15:47 ` [PATCH 16/19] drm/i915: don't touch the VDD when disabling " Paulo Zanoni
2013-11-29 14:41   ` Rodrigo Vivi
2013-11-21 15:47 ` [PATCH 17/19] drm/i915: fix VDD override off wait Paulo Zanoni
2013-11-21 15:47 ` Paulo Zanoni [this message]
2013-11-21 16:00   ` [PATCH 18/19] drm/i915: save some time when waiting the eDP timings Chris Wilson
2013-11-25 22:17     ` Ben Widawsky
2013-11-25 23:25       ` Chris Wilson
2013-11-26  2:38         ` Ben Widawsky
2013-11-26  9:14           ` Chris Wilson
2013-11-26 15:53             ` Paulo Zanoni
2013-11-21 15:47 ` [PATCH 19/19] drm/i915: init the DP panel power seq regs earlier Paulo Zanoni
2013-12-05 15:00   ` Jani Nikula
2013-12-06 18:39     ` Paulo Zanoni

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=1385048853-1579-19-git-send-email-przanoni@gmail.com \
    --to=przanoni@gmail.com \
    --cc=intel-gfx@lists.freedesktop.org \
    --cc=paulo.r.zanoni@intel.com \
    /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