All of lore.kernel.org
 help / color / mirror / Atom feed
From: Ville Syrjala <ville.syrjala@linux.intel.com>
To: intel-gfx@lists.freedesktop.org
Subject: [PATCH v2 01/17] drm/i915: Update pipes in reverse order for bigjoiner
Date: Fri,  5 Apr 2024 00:34:25 +0300	[thread overview]
Message-ID: <20240404213441.17637-2-ville.syrjala@linux.intel.com> (raw)
In-Reply-To: <20240404213441.17637-1-ville.syrjala@linux.intel.com>

From: Ville Syrjälä <ville.syrjala@linux.intel.com>

With bigjoiner the master crtc is the one that will send out the
uapi event/etc. We want that to happen after all the slaves are
done, so let's try to do the commits in reverse order so that
the master comes last.

Even worse, the modeset helper will simply complete the commit
on the slave pipe immediately as it consider the crtc to be inactive
(it can't see our crtc_state->hw.active/etc.).

With regular sync updates this generally doesn't matter all that
much as the slave pipe should typically finish its work during the
same frame as the master pipe. However in case the slave pipe's commit
slips into the next frame we end up in a bit of trouble. This is most
visible with either async flips (currently disabled with bigjoiner
exactly for this reason), and DSB gamma updates. With DSB the problem
happens because the DSB itself will wait until the next start vblank
before starting to execute. So if the master pipe already finished its
commit and the DSB on the slave pipe is still waiting for the next
vblank we will assume the DSB as gotten stuck and terminate it.

Reversing the commit order should ameliarate this for the most part
as the master pipe is guaranteed to start its commit after the slave
pipe started. The one thing that can still screw us over is the fact
that we aren't necessarily going to commit the pipes in the reverse
order as the actual order is dictated by the DDB overlap avoidance.
But that can only happen while other pipes are being enabled/disabled,
and so in the normal steady state we should be safe.

The full fix will involve making the commit machinery aware of the
slave pipes and not finish their commits prematurely. But that
will involve a bit more work than this. And this commit order
reversal will still be beneficial to avoid userspace getting an
-EBUSY from the following page flip if the second pipe's commit
does stretch into the next frame.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
---
 drivers/gpu/drm/i915/display/intel_display.c | 14 +++++++++++---
 drivers/gpu/drm/i915/display/intel_display.h |  8 ++++++++
 2 files changed, 19 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/display/intel_display.c b/drivers/gpu/drm/i915/display/intel_display.c
index a481c9218138..0086a7422e86 100644
--- a/drivers/gpu/drm/i915/display/intel_display.c
+++ b/drivers/gpu/drm/i915/display/intel_display.c
@@ -6956,8 +6956,12 @@ static void skl_commit_modeset_enables(struct intel_atomic_state *state)
 	intel_dbuf_mbus_pre_ddb_update(state);
 
 	while (update_pipes) {
-		for_each_oldnew_intel_crtc_in_state(state, crtc, old_crtc_state,
-						    new_crtc_state, i) {
+		/*
+		 * Commit in reverse order to make bigjoiner master
+		 * send the uapi events after slaves are done.
+		 */
+		for_each_oldnew_intel_crtc_in_state_reverse(state, crtc, old_crtc_state,
+							    new_crtc_state, i) {
 			enum pipe pipe = crtc->pipe;
 
 			if ((update_pipes & BIT(pipe)) == 0)
@@ -7036,7 +7040,11 @@ static void skl_commit_modeset_enables(struct intel_atomic_state *state)
 		intel_pre_update_crtc(state, crtc);
 	}
 
-	for_each_new_intel_crtc_in_state(state, crtc, new_crtc_state, i) {
+	/*
+	 * Commit in reverse order to make bigjoiner master
+	 * send the uapi events after slaves are done.
+	 */
+	for_each_new_intel_crtc_in_state_reverse(state, crtc, new_crtc_state, i) {
 		enum pipe pipe = crtc->pipe;
 
 		if ((update_pipes & BIT(pipe)) == 0)
diff --git a/drivers/gpu/drm/i915/display/intel_display.h b/drivers/gpu/drm/i915/display/intel_display.h
index 986ec77490de..423074d6947a 100644
--- a/drivers/gpu/drm/i915/display/intel_display.h
+++ b/drivers/gpu/drm/i915/display/intel_display.h
@@ -344,6 +344,14 @@ enum phy_fia {
 	     (__i)++) \
 		for_each_if(crtc)
 
+#define for_each_new_intel_crtc_in_state_reverse(__state, crtc, new_crtc_state, __i) \
+	for ((__i) = (__state)->base.dev->mode_config.num_crtc - 1; \
+	     (__i) >= 0  && \
+	     ((crtc) = to_intel_crtc((__state)->base.crtcs[__i].ptr), \
+	      (new_crtc_state) = to_intel_crtc_state((__state)->base.crtcs[__i].new_state), 1); \
+	     (__i)--) \
+		for_each_if(crtc)
+
 #define for_each_oldnew_intel_plane_in_state(__state, plane, old_plane_state, new_plane_state, __i) \
 	for ((__i) = 0; \
 	     (__i) < (__state)->base.dev->mode_config.num_total_plane && \
-- 
2.43.2


  reply	other threads:[~2024-04-04 21:34 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-04-04 21:34 [PATCH v2 00/17] drm/i915: Bigjoiner modeset sequence redesign and MST support Ville Syrjala
2024-04-04 21:34 ` Ville Syrjala [this message]
2024-04-05 12:56   ` [PATCH v2 01/17] drm/i915: Update pipes in reverse order for bigjoiner Murthy, Arun R
2024-04-08 12:46     ` Ville Syrjälä
2024-04-05 13:42   ` Kulkarni, Vandita
2024-04-05 19:45     ` Ville Syrjälä
2024-04-04 21:34 ` [PATCH v2 02/17] drm/i915/psr: Disable PSR when bigjoiner is used Ville Syrjala
2024-04-05  6:58   ` Hogander, Jouni
2024-04-05 12:58     ` Murthy, Arun R
2024-04-05 13:00     ` Murthy, Arun R
2024-04-05 15:45       ` Murthy, Arun R
2024-04-05 19:34     ` Ville Syrjälä
2024-04-08  5:09       ` Hogander, Jouni
2024-04-04 21:34 ` [PATCH v2 03/17] drm/i915: Disable port sync " Ville Syrjala
2024-04-04 21:34 ` [PATCH v2 04/17] drm/i915: Disable live M/N updates when using bigjoiner Ville Syrjala
2024-04-04 21:34 ` [PATCH v2 05/17] drm/i915/vrr: Disable VRR " Ville Syrjala
2024-04-04 21:34 ` [PATCH v2 06/17] drm/i915: Fix intel_modeset_pipe_config_late() for bigjoiner Ville Syrjala
2024-04-04 21:34 ` [PATCH v2 07/17] drm/i915: s/intel_dp_can_bigjoiner()/intel_dp_has_bigjoiner()/ Ville Syrjala
2024-04-04 21:34 ` [PATCH v2 08/17] drm/i915: Extract intel_dp_joiner_needs_dsc() Ville Syrjala
2024-04-04 21:34 ` [PATCH v2 09/17] drm/i915/mst: Check intel_dp_joiner_needs_dsc() Ville Syrjala
2024-04-04 21:34 ` [PATCH v2 10/17] drm/i915: Pass connector to intel_dp_need_bigjoiner() Ville Syrjala
2024-04-04 21:34 ` [PATCH v2 11/17] drm/i915: Introduce intel_crtc_joined_pipe_mask() Ville Syrjala
2024-04-04 21:34 ` [PATCH v2 12/17] drm/i915: Extract intel_ddi_post_disable_hdmi_or_sst() Ville Syrjala
2024-04-04 21:34 ` [PATCH v2 13/17] drm/i915: Utilize intel_crtc_joined_pipe_mask() more Ville Syrjala
2024-04-04 21:34 ` [PATCH v2 14/17] drm/i915: Handle joined pipes inside hsw_crtc_disable() Ville Syrjala
2024-04-04 21:34 ` [PATCH v2 15/17] drm/i915: Handle joined pipes inside hsw_crtc_enable() Ville Syrjala
2024-04-04 21:34 ` [PATCH v2 16/17] drm/i915/mst: Add bigjoiner handling to MST modeset sequence Ville Syrjala
2024-04-04 21:34 ` [PATCH v2 17/17] drm/i915: Allow bigjoiner for MST Ville Syrjala
2024-04-04 22:16 ` ✗ Fi.CI.CHECKPATCH: warning for drm/i915: Bigjoiner modeset sequence redesign and MST support (rev2) Patchwork
2024-04-04 22:16 ` ✗ Fi.CI.SPARSE: " Patchwork
2024-04-04 22:25 ` ✗ Fi.CI.BAT: failure " Patchwork
2024-04-05  1:46 ` ✗ Fi.CI.CHECKPATCH: warning for drm/i915: Bigjoiner modeset sequence redesign and MST support (rev3) Patchwork
2024-04-05  1:46 ` ✗ Fi.CI.SPARSE: " Patchwork
2024-04-05  1:51 ` ✓ Fi.CI.BAT: success " Patchwork
2024-04-05 14:25 ` ✗ Fi.CI.IGT: failure " 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=20240404213441.17637-2-ville.syrjala@linux.intel.com \
    --to=ville.syrjala@linux.intel.com \
    --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 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.