All of lore.kernel.org
 help / color / mirror / Atom feed
From: Wayne Lin <Wayne.Lin@amd.com>
To: <amd-gfx@lists.freedesktop.org>
Cc: <Harry.Wentland@amd.com>, <Sunpeng.Li@amd.com>,
	<Rodrigo.Siqueira@amd.com>, <Aurabindo.Pillai@amd.com>,
	<roman.li@amd.com>, <wayne.lin@amd.com>,
	<agustin.gutierrez@amd.com>, <chiahsuan.chung@amd.com>,
	<hersenxs.wu@amd.com>, <jerry.zuo@amd.com>,
	Alvin Lee <alvin.lee2@amd.com>, Samson Tam <samson.tam@amd.com>,
	Chaitanya Dhere <chaitanya.dhere@amd.com>
Subject: [PATCH 35/46] drm/amd/display: For FPO + Vactive check that all pipes support VA
Date: Wed, 24 Apr 2024 16:49:20 +0800	[thread overview]
Message-ID: <20240424084931.2656128-36-Wayne.Lin@amd.com> (raw)
In-Reply-To: <20240424084931.2656128-1-Wayne.Lin@amd.com>

From: Alvin Lee <alvin.lee2@amd.com>

[Description]
For FPO + Vactive scenarios we must check that all non-FPO pipes
have VACTIVE margin to allow it. The previous check only confirmed
that there is at least one pipe that has vactive margin, but this
is incorrect as the vactive display could be using two pipes (MPO)
where the desktop plane has vactive margin, and the video plane
does not.

Reviewed-by: Samson Tam <samson.tam@amd.com>
Reviewed-by: Chaitanya Dhere <chaitanya.dhere@amd.com>
Acked-by: Wayne Lin <wayne.lin@amd.com>
Signed-off-by: Alvin Lee <alvin.lee2@amd.com>
---
 .../display/dc/dcn32/dcn32_resource_helpers.c |  2 +-
 .../drm/amd/display/dc/dml/dcn32/dcn32_fpu.c  | 20 ++++++++++++++-----
 .../drm/amd/display/dc/dml/dcn32/dcn32_fpu.h  |  2 +-
 3 files changed, 17 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c
index a8c36eda1d09..eba7bfc7e4af 100644
--- a/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c
+++ b/drivers/gpu/drm/amd/display/dc/dcn32/dcn32_resource_helpers.c
@@ -545,7 +545,7 @@ struct dc_stream_state *dcn32_can_support_mclk_switch_using_fw_based_vblank_stre
 		if (fpo_candidate_stream)
 			fpo_stream_status = dc_state_get_stream_status(context, fpo_candidate_stream);
 		DC_FP_START();
-		is_fpo_vactive = dcn32_find_vactive_pipe(dc, context, dc->debug.fpo_vactive_min_active_margin_us);
+		is_fpo_vactive = dcn32_find_vactive_pipe(dc, context, fpo_candidate_stream, dc->debug.fpo_vactive_min_active_margin_us);
 		DC_FP_END();
 		if (!is_fpo_vactive || dc->debug.disable_fpo_vactive)
 			return NULL;
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
index 5be976fa44f9..8912475f01e2 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.c
@@ -3521,15 +3521,16 @@ void dcn32_assign_fpo_vactive_candidate(struct dc *dc, const struct dc_state *co
  *
  * @dc: current dc state
  * @context: new dc state
+ * @fpo_candidate_stream: candidate stream to be chosen for FPO
  * @vactive_margin_req_us: The vactive marign required for a vactive pipe to be considered "found"
  *
  * Return: True if VACTIVE display is found, false otherwise
  */
-bool dcn32_find_vactive_pipe(struct dc *dc, const struct dc_state *context, uint32_t vactive_margin_req_us)
+bool dcn32_find_vactive_pipe(struct dc *dc, const struct dc_state *context, struct dc_stream_state *fpo_candidate_stream, uint32_t vactive_margin_req_us)
 {
 	unsigned int i, pipe_idx;
 	const struct vba_vars_st *vba = &context->bw_ctx.dml.vba;
-	bool vactive_found = false;
+	bool vactive_found = true;
 	unsigned int blank_us = 0;
 
 	for (i = 0, pipe_idx = 0; i < dc->res_pool->pipe_count; i++) {
@@ -3538,11 +3539,20 @@ bool dcn32_find_vactive_pipe(struct dc *dc, const struct dc_state *context, uint
 		if (!pipe->stream)
 			continue;
 
+		/* Don't need to check for vactive margin on the FPO candidate stream */
+		if (fpo_candidate_stream && pipe->stream == fpo_candidate_stream) {
+			pipe_idx++;
+			continue;
+		}
+
+		/* Every plane (apart from the ones driven by the FPO pipes) needs to have active margin
+		 * in order for us to have found a valid "vactive" config for FPO + Vactive
+		 */
 		blank_us = ((pipe->stream->timing.v_total - pipe->stream->timing.v_addressable) * pipe->stream->timing.h_total /
 				(double)(pipe->stream->timing.pix_clk_100hz * 100)) * 1000000;
-		if (vba->ActiveDRAMClockChangeLatencyMarginPerState[vba->VoltageLevel][vba->maxMpcComb][vba->pipe_plane[pipe_idx]] >= vactive_margin_req_us &&
-				!(pipe->stream->vrr_active_variable || pipe->stream->vrr_active_fixed) && blank_us < dc->debug.fpo_vactive_max_blank_us) {
-			vactive_found = true;
+		if (vba->ActiveDRAMClockChangeLatencyMarginPerState[vba->VoltageLevel][vba->maxMpcComb][vba->pipe_plane[pipe_idx]] < vactive_margin_req_us ||
+				pipe->stream->vrr_active_variable || pipe->stream->vrr_active_fixed || blank_us >= dc->debug.fpo_vactive_max_blank_us) {
+			vactive_found = false;
 			break;
 		}
 		pipe_idx++;
diff --git a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.h b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.h
index d25c3f730a59..276e90e4e0ce 100644
--- a/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.h
+++ b/drivers/gpu/drm/amd/display/dc/dml/dcn32/dcn32_fpu.h
@@ -71,7 +71,7 @@ void dcn32_zero_pipe_dcc_fraction(display_e2e_pipe_params_st *pipes,
 
 void dcn32_assign_fpo_vactive_candidate(struct dc *dc, const struct dc_state *context, struct dc_stream_state **fpo_candidate_stream);
 
-bool dcn32_find_vactive_pipe(struct dc *dc, const struct dc_state *context, uint32_t vactive_margin_req);
+bool dcn32_find_vactive_pipe(struct dc *dc, const struct dc_state *context, struct dc_stream_state *fpo_candidate_stream, uint32_t vactive_margin_req);
 
 void dcn32_override_min_req_memclk(struct dc *dc, struct dc_state *context);
 
-- 
2.37.3


  parent reply	other threads:[~2024-04-24  8:53 UTC|newest]

Thread overview: 49+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-04-24  8:48 [PATCH 00/46] DC Patches April 29, 2024 Wayne Lin
2024-04-24  8:48 ` [PATCH 01/46] drm/amd/display: Do cursor programming with rest of pipe Wayne Lin
2024-04-24  8:48 ` [PATCH 02/46] drm/amd/display: Always use legacy way of setting cursor on DCE Wayne Lin
2024-04-24  8:48 ` [PATCH 03/46] drm/amd/display: Add NULL pointer check for kzalloc Wayne Lin
2024-04-24  8:48 ` [PATCH 04/46] drm/amd/display: Check index msg_id before read or write Wayne Lin
2024-04-24  8:48 ` [PATCH 05/46] drm/amd/display: Check pipe offset before setting vblank Wayne Lin
2024-04-24  8:48 ` [PATCH 06/46] drm/amd/display: Skip finding free audio for unknown engine_id Wayne Lin
2024-04-24  8:48 ` [PATCH 07/46] drm/amd/display: Fix overlapping copy within dml_core_mode_programming Wayne Lin
2024-04-24  8:48 ` [PATCH 08/46] drm/amd/display: Do not return negative stream id for array Wayne Lin
2024-04-24  8:48 ` [PATCH 09/46] drm/amd/display: ASSERT when failing to find index by plane/stream id Wayne Lin
2024-04-24  8:48 ` [PATCH 10/46] drm/amd/display: Remove unnecessary files Wayne Lin
2024-04-24  8:48 ` [PATCH 11/46] drm/amd/display: Improve registers write Wayne Lin
2024-04-24  8:48 ` [PATCH 12/46] drm/amd/display: Add missing SMU version Wayne Lin
2024-04-24  8:48 ` [PATCH 13/46] drm/amd/display: Adjust codestyle for dcn31 and hdcp_msg Wayne Lin
2024-04-24  8:48 ` [PATCH 14/46] drm/amd/display: Add VCO speed parameter for DCN31 FPU Wayne Lin
2024-04-24  8:49 ` [PATCH 15/46] drm/amd/display: Adjust functions prefix for some of the dcn301 fpu functions Wayne Lin
2024-04-24  8:49 ` [PATCH 16/46] drm/amd/display: Enable legacy fast update for dcn301 Wayne Lin
2024-04-24  8:49 ` [PATCH 17/46] drm/amd/display: Update some of the dcn303 parameters Wayne Lin
2024-04-24  8:49 ` [PATCH 18/46] drm/amd/display: Remove legacy code in DC Wayne Lin
2024-04-24  8:49 ` [PATCH 19/46] drm/amd/display: Add log_color_state callback to multiple DCNs Wayne Lin
2024-04-24  8:49 ` [PATCH 20/46] drm/amd/display: Handle the case which quad_part is equal 0 Wayne Lin
2024-04-24  8:49 ` [PATCH 21/46] drm/amd/display: Refactor for Replay Link off frame count Wayne Lin
2024-04-24  8:49 ` [PATCH 22/46] drm/amd/display: Restrict multi-disp support for in-game FAMS Wayne Lin
2024-04-24  8:49 ` [PATCH 23/46] drm/amd/display: Defer handling mst up request in resume Wayne Lin
2024-04-24  8:49 ` [PATCH 24/46] drm/amd/display: Fix DC mode screen flickering on DCN321 Wayne Lin
2024-04-24  8:49 ` [PATCH 25/46] drm/amd/display: take ODM slice count into account when deciding DSC slice Wayne Lin
2024-04-24  8:49 ` [PATCH 26/46] drm/amd/display: Re-enable IPS2 for static screen Wayne Lin
2024-04-24  8:49 ` [PATCH 27/46] drm/amd/display: Add trigger FIFO resync path for DCN35 Wayne Lin
2024-04-24  8:49 ` [PATCH 28/46] drm/amd/display: Enable RCO for PHYSYMCLK in DCN35 Wayne Lin
2024-04-24 13:48   ` Li, Roman
2024-04-24  8:49 ` [PATCH 29/46] drm/amd/display: Revert "dc: Keep VBios pixel rate div setting util next mode set" Wayne Lin
2024-04-24  8:49 ` [PATCH 30/46] drm/amd/display: Only program P-State force if pipe config changed Wayne Lin
2024-04-24  8:49 ` [PATCH 31/46] drm/amd/display: Remove redundant include file Wayne Lin
2024-04-24  8:49 ` [PATCH 32/46] drm/amd/display: Refactor HUBBUB into component folder Wayne Lin
2024-04-24  8:49 ` [PATCH 33/46] drm/amd/display: Assign linear_pitch_alignment even for VM Wayne Lin
2024-04-24  8:49 ` [PATCH 34/46] drm/amd/display: gpuvm handling in DML21 Wayne Lin
2024-04-24  8:49 ` Wayne Lin [this message]
2024-04-24  8:49 ` [PATCH 36/46] drm/amd/display: Fix uninitialized variables in DM Wayne Lin
2024-04-24  8:49 ` [PATCH 37/46] drm/amd/display: Fix uninitialized variables in DC Wayne Lin
2024-04-24  8:49 ` [PATCH 38/46] drm/amd/display: Fix FEC_READY write on DP LT Wayne Lin
2024-04-24  8:49 ` [PATCH 39/46] drm/amd/display: use even ODM slice width for two pixels per container Wayne Lin
2024-04-24  8:49 ` [PATCH 40/46] drm/amd/display: Enable Replay for DCN315 Wayne Lin
2024-04-24  8:49 ` [PATCH 41/46] drm/amd/display: Notify idle link detection through shared state Wayne Lin
2024-04-24  8:49 ` [PATCH 42/46] drm/amd/display: Add periodic detection for IPS Wayne Lin
2024-04-24  8:49 ` [PATCH 43/46] drm/amd/display: Change ASSR disable sequence Wayne Lin
2024-04-24  8:49 ` [PATCH 44/46] drm/amd/display: Fix uninitialized variables in DC Wayne Lin
2024-04-24  8:49 ` [PATCH 45/46] drm/amd/display: Disable seamless boot on 128b/132b encoding Wayne Lin
2024-04-24  8:49 ` [PATCH 46/46] drm/amd/display: 3.2.283 Wayne Lin
2024-04-29 13:17 ` [PATCH 00/46] DC Patches April 29, 2024 Wheeler, Daniel

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=20240424084931.2656128-36-Wayne.Lin@amd.com \
    --to=wayne.lin@amd.com \
    --cc=Aurabindo.Pillai@amd.com \
    --cc=Harry.Wentland@amd.com \
    --cc=Rodrigo.Siqueira@amd.com \
    --cc=Sunpeng.Li@amd.com \
    --cc=agustin.gutierrez@amd.com \
    --cc=alvin.lee2@amd.com \
    --cc=amd-gfx@lists.freedesktop.org \
    --cc=chaitanya.dhere@amd.com \
    --cc=chiahsuan.chung@amd.com \
    --cc=hersenxs.wu@amd.com \
    --cc=jerry.zuo@amd.com \
    --cc=roman.li@amd.com \
    --cc=samson.tam@amd.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 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.