All of lore.kernel.org
 help / color / mirror / Atom feed
From: Chris Wilson <chris@chris-wilson.co.uk>
To: jbarnes@virtuousgeek.org
Cc: intel-gfx@lists.freedesktop.org
Subject: [PATCH 03/11] drm/i915: Fix offset page-flips on i965+
Date: Wed, 11 Aug 2010 10:31:28 +0100	[thread overview]
Message-ID: <1281519096-5152-4-git-send-email-chris@chris-wilson.co.uk> (raw)
In-Reply-To: <1281519096-5152-1-git-send-email-chris@chris-wilson.co.uk>

i965 uses the Display Registers to compute the offset from the display
base so the new base does not need adjusting when flipping. The older
chipsets use a fence to access the display and so do perceive the
surface as linear and have a single base register which is reprogrammed
using the flip.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Jesse Barnes <jbarnes@virtuousgeek.org>
Reported-by: Marty Jack <martyj19@comcast.net>
---
 drivers/gpu/drm/i915/intel_display.c |   67 ++++++++++++++++++++++++----------
 1 files changed, 48 insertions(+), 19 deletions(-)

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 6ae4f69..d5dd86f 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -5063,9 +5063,9 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
 	struct intel_unpin_work *work;
 	unsigned long flags, offset;
-	int pipesrc_reg = (intel_crtc->pipe == 0) ? PIPEASRC : PIPEBSRC;
-	int ret, pipesrc;
-	u32 flip_mask;
+	int pipe = intel_crtc->pipe;
+	u32 pf, pipesrc;
+	int ret;
 
 	work = kzalloc(sizeof *work, GFP_KERNEL);
 	if (work == NULL)
@@ -5114,12 +5114,14 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	atomic_inc(&obj_priv->pending_flip);
 	work->pending_flip_obj = obj;
 
-	if (intel_crtc->plane)
-		flip_mask = MI_WAIT_FOR_PLANE_B_FLIP;
-	else
-		flip_mask = MI_WAIT_FOR_PLANE_A_FLIP;
-
 	if (IS_GEN3(dev) || IS_GEN2(dev)) {
+		u32 flip_mask;
+
+		if (intel_crtc->plane)
+			flip_mask = MI_WAIT_FOR_PLANE_B_FLIP;
+		else
+			flip_mask = MI_WAIT_FOR_PLANE_A_FLIP;
+
 		BEGIN_LP_RING(2);
 		OUT_RING(MI_WAIT_FOR_EVENT | flip_mask);
 		OUT_RING(0);
@@ -5127,29 +5129,56 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	}
 
 	/* Offset into the new buffer for cases of shared fbs between CRTCs */
-	offset = obj_priv->gtt_offset;
-	offset += (crtc->y * fb->pitch) + (crtc->x * (fb->bits_per_pixel) / 8);
+	offset = crtc->y * fb->pitch + crtc->x * fb->bits_per_pixel/8;
 
 	BEGIN_LP_RING(4);
-	if (IS_I965G(dev)) {
+	switch(INTEL_INFO(dev)->gen) {
+	case 2:
 		OUT_RING(MI_DISPLAY_FLIP |
 			 MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
 		OUT_RING(fb->pitch);
-		OUT_RING(offset | obj_priv->tiling_mode);
-		pipesrc = I915_READ(pipesrc_reg); 
-		OUT_RING(pipesrc & 0x0fff0fff);
-	} else if (IS_GEN3(dev)) {
+		OUT_RING(obj_priv->gtt_offset + offset);
+		OUT_RING(MI_NOOP);
+		break;
+
+	case 3:
 		OUT_RING(MI_DISPLAY_FLIP_I915 |
 			 MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
 		OUT_RING(fb->pitch);
-		OUT_RING(offset);
+		OUT_RING(obj_priv->gtt_offset + offset);
 		OUT_RING(MI_NOOP);
-	} else {
+		break;
+
+	case 4:
+	case 5:
+		/* i965+ uses the linear or tiled offsets from the
+		 * Display Registers (which do not change across a page-flip)
+		 * so we need only reprogram the base address.
+		 */
 		OUT_RING(MI_DISPLAY_FLIP |
 			 MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
 		OUT_RING(fb->pitch);
-		OUT_RING(offset);
-		OUT_RING(MI_NOOP);
+		OUT_RING(obj_priv->gtt_offset | obj_priv->tiling_mode);
+
+		/* XXX Enabling the panel-fitter across page-flip is so far
+		 * untested on non-native modes, so ignore it for now.
+		 * pf = I915_READ(pipe == 0 ? PFA_CTL_1 : PFB_CTL_1) & PF_ENABLE;
+		 */
+		pf = 0;
+		pipesrc = I915_READ(pipe == 0 ? PIPEASRC : PIPEBSRC) & 0x0fff0fff;
+		OUT_RING(pf | pipesrc);
+		break;
+
+	case 6:
+		OUT_RING(MI_DISPLAY_FLIP |
+			 MI_DISPLAY_FLIP_PLANE(intel_crtc->plane));
+		OUT_RING(fb->pitch | obj_priv->tiling_mode);
+		OUT_RING(obj_priv->gtt_offset);
+
+		pf = I915_READ(pipe == 0 ? PFA_CTL_1 : PFB_CTL_1) & PF_ENABLE;
+		pipesrc = I915_READ(pipe == 0 ? PIPEASRC : PIPEBSRC) & 0x0fff0fff;
+		OUT_RING(pf | pipesrc);
+		break;
 	}
 	ADVANCE_LP_RING();
 
-- 
1.7.1

  parent reply	other threads:[~2010-08-11  9:32 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2010-08-11  9:31 Remaining set of small patches for -rc1 Chris Wilson
2010-08-11  9:31 ` [PATCH 01/11] drm/i915: Avoid using msleep under kdb and wait_for() Chris Wilson
2010-08-11 15:20   ` Jesse Barnes
2010-08-11  9:31 ` [PATCH 02/11] drm/i915: Include a generation number in the device info Chris Wilson
2010-08-11 15:22   ` Jesse Barnes
2010-08-11  9:31 ` Chris Wilson [this message]
2010-08-11  9:31 ` [PATCH 04/11] drm/i915: Clear scanline waits after disabling the pipe Chris Wilson
2010-08-11 15:23   ` Jesse Barnes
2010-08-11 16:12     ` Chris Wilson
2010-08-11  9:31 ` [PATCH 05/11] drm/i915: Sanity check user framebuffer parameters on creation Chris Wilson
2010-08-11  9:31 ` [PATCH 06/11] drm/i915: Re-use set_base_atomic to share setting of the display registers Chris Wilson
2010-08-11 15:25   ` Jesse Barnes
2010-08-11  9:31 ` [PATCH 07/11] drm/i915/sdvo: Propagate error from switching control buses Chris Wilson
2010-08-11  9:31 ` [PATCH 08/11] drm/i915: Add RING_WAIT reset to hangcheck Chris Wilson
2010-08-22  6:19   ` Eric Anholt
2010-08-22  8:50     ` [PATCH] drm/i915: Add ringbuffer wait " Chris Wilson
2010-08-11  9:31 ` [PATCH 09/11] drm/i915/crt: Flush register prior to waiting for vblank Chris Wilson
2010-08-11  9:31 ` [PATCH 10/11] drm/i915/dp: Boost timeout for enabling transcoder to 100ms Chris Wilson
2010-08-11  9:31 ` [PATCH 11/11] drm/i915: Perform initial configuration asynchronously Chris Wilson
2010-08-11  9:46 ` Remaining set of small patches for -rc1 shadi
2010-08-11 15:21   ` Alan W. Irwin

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=1281519096-5152-4-git-send-email-chris@chris-wilson.co.uk \
    --to=chris@chris-wilson.co.uk \
    --cc=intel-gfx@lists.freedesktop.org \
    --cc=jbarnes@virtuousgeek.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.