Intel-GFX Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
To: intel-xe@lists.freedesktop.org
Cc: intel-gfx@lists.freedesktop.org,
	"Ville Syrjälä" <ville.syrjala@linux.intel.com>,
	"Jani Nikula" <jani.nikula@intel.com>,
	"Maarten Lankhorst" <maarten.lankhorst@linux.intel.com>
Subject: [PATCH v2.1 03/12] drm/i915/display: Use async flip when available for initial plane config
Date: Mon,  7 Oct 2024 12:23:55 +0200	[thread overview]
Message-ID: <20241007102355.134080-1-maarten.lankhorst@linux.intel.com> (raw)
In-Reply-To: <20241003154421.33805-4-maarten.lankhorst@linux.intel.com>

I'm planning to reorder readout in the Xe sequence in such a way that
interrupts will not be available, so just use an async flip.

Since the new FB points to the same pages, it will not tear. It also
has the benefit of perhaps being slightly faster.

Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
---
Fix compiler fails..
Change intel_de_wait_custom to normal variant.

I still believe we should be fine with async flips. The buffer will not
be a standard RGBX8888 on the first plane. If we violate a constraint, it
will be from alignment, and for that it would be interesting to find any
border cases we missed.

 drivers/gpu/drm/i915/display/skl_universal_plane.c | 13 +++++++++++--
 1 file changed, 11 insertions(+), 2 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/skl_universal_plane.c b/drivers/gpu/drm/i915/display/skl_universal_plane.c
index fdb141cfa4274..c7ba8fcff20c9 100644
--- a/drivers/gpu/drm/i915/display/skl_universal_plane.c
+++ b/drivers/gpu/drm/i915/display/skl_universal_plane.c
@@ -2800,7 +2800,7 @@ bool skl_fixup_initial_plane_config(struct intel_crtc *crtc,
 		to_intel_plane_state(plane->base.state);
 	enum plane_id plane_id = plane->id;
 	enum pipe pipe = crtc->pipe;
-	u32 base;
+	u32 base, plane_ctl;
 
 	if (!plane_state->uapi.visible)
 		return false;
@@ -2814,7 +2814,16 @@ bool skl_fixup_initial_plane_config(struct intel_crtc *crtc,
 	if (plane_config->base == base)
 		return false;
 
+	/* Perform an async flip to the new surface. */
+	plane_ctl = intel_de_read(i915, PLANE_CTL(pipe, plane_id));
+	plane_ctl |= PLANE_CTL_ASYNC_FLIP;
+
+	intel_de_write(i915, PLANE_CTL(pipe, plane_id), plane_ctl);
 	intel_de_write(i915, PLANE_SURF(pipe, plane_id), base);
 
-	return true;
+	if (intel_de_wait(i915, PLANE_SURFLIVE(pipe, plane_id), ~0U, base, 40) < 0)
+		drm_warn(&i915->drm, "async flip timed out\n");
+
+	/* No need to vblank wait either */
+	return false;
 }
-- 
2.45.2


  parent reply	other threads:[~2024-10-07 10:23 UTC|newest]

Thread overview: 33+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-10-03 15:44 [PATCH v3 00/12] drm/xe: Reduce flickering when inheriting BIOS fb Maarten Lankhorst
2024-10-03 15:44 ` [PATCH v3 01/12] drm/xe/display: Handle stolen bar readout in the same way as lmem Maarten Lankhorst
2024-10-07 18:52   ` Lucas De Marchi
2024-10-03 15:44 ` [PATCH v3 02/12] drm/xe: Remove double pageflip Maarten Lankhorst
2024-10-11  9:38   ` Govindapillai, Vinod
2024-10-03 15:44 ` [PATCH v3 03/12] drm/i915/display: Use async flip when available for initial plane config Maarten Lankhorst
2024-10-03 15:58   ` Jani Nikula
2024-10-03 20:51     ` Maarten Lankhorst
2024-10-03 16:14   ` Ville Syrjälä
2024-10-03 20:50     ` Maarten Lankhorst
2024-10-03 20:59       ` Ville Syrjälä
2024-10-04 12:48   ` kernel test robot
2024-10-07 10:23   ` Maarten Lankhorst [this message]
2024-10-03 15:44 ` [PATCH v3 04/12] drm/xe/display: Remove single wait for vblank Maarten Lankhorst
2024-10-03 15:44 ` [PATCH v3 05/12] drm/xe: Move suballocator init to after display init Maarten Lankhorst
2024-10-03 15:44 ` [PATCH v3 06/12] drm/xe: Use xe_ggtt_map_bo_unlocked for resume Maarten Lankhorst
2024-10-04  6:27   ` Matthew Brost
2024-10-03 15:44 ` [PATCH v3 07/12] drm/xe: Add xe_ggtt_might_lock Maarten Lankhorst
2024-10-04  6:28   ` Matthew Brost
2024-10-03 15:44 ` [PATCH v3 08/12] drm/xe: Add xe_ggtt_alloc Maarten Lankhorst
2024-10-04  6:33   ` Matthew Brost
2024-10-03 15:44 ` [PATCH v3 09/12] drm/xe: Abstract read/write functions for GGTT PTEs Maarten Lankhorst
2024-10-04  6:35   ` Matthew Brost
2024-10-03 15:44 ` [PATCH v3 10/12] drm/xe: Make xe_ggtt_pt_ops private Maarten Lankhorst
2024-10-04  6:38   ` Matthew Brost
2024-10-04 13:30   ` kernel test robot
2024-10-07 18:40   ` Lucas De Marchi
2024-10-03 15:44 ` [PATCH v3 11/12] drm/xe/display: Stop dereferencing ggtt in xe_fb_pin Maarten Lankhorst
2024-10-07 16:58   ` Matthew Brost
2024-10-03 15:44 ` [PATCH v3 12/12] drm/xe: Move struct xe_ggtt to xe_ggtt.c Maarten Lankhorst
2024-10-04  6:41   ` Matthew Brost
2024-10-03 19:53 ` ✗ Fi.CI.BUILD: failure for drm/xe: Reduce flickering when inheriting BIOS fb Patchwork
2024-10-08  3:16 ` ✗ Fi.CI.BUILD: failure for drm/xe: Reduce flickering when inheriting BIOS fb. (rev2) Patchwork

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=20241007102355.134080-1-maarten.lankhorst@linux.intel.com \
    --to=maarten.lankhorst@linux.intel.com \
    --cc=intel-gfx@lists.freedesktop.org \
    --cc=intel-xe@lists.freedesktop.org \
    --cc=jani.nikula@intel.com \
    --cc=ville.syrjala@linux.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