AMD-GFX Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: <sunpeng.li-5C7GfCeVMHo@public.gmane.org>
To: amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org
Cc: Su Sung Chung <Su.Chung-5C7GfCeVMHo@public.gmane.org>
Subject: [PATCH 14/26] drm/amd/display: fix calculation of total_data_read_bandwidth
Date: Mon, 27 May 2019 15:50:53 -0400	[thread overview]
Message-ID: <1558986665-12964-15-git-send-email-sunpeng.li@amd.com> (raw)
In-Reply-To: <1558986665-12964-1-git-send-email-sunpeng.li-5C7GfCeVMHo@public.gmane.org>

From: Su Sung Chung <Su.Chung@amd.com>

[why]
by adding fast_validate flag, we are skipping some portion of
dcn_validate_bandwidth code that is not necessary for mode validation.
However we have a bug where it does not calculate
v->total_data_read_bandwidth, which is one of the factors determines the
result of the validation, and therefore report false positive during
mode validation.

[how]
add calculation of v->total_data_read_bandwidth outside of the region
that is guarded by fast_validate flag

Signed-off-by: Su Sung Chung <Su.Chung@amd.com>
Reviewed-by: Dmytro Laktyushkin <Dmytro.Laktyushkin@amd.com>
Acked-by: Leo Li <sunpeng.li@amd.com>
---
 drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c | 39 +++++++++++++++++++++++-
 1 file changed, 38 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c b/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c
index d53306d..0a336c8 100644
--- a/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c
+++ b/drivers/gpu/drm/amd/display/dc/calcs/dcn_calcs.c
@@ -711,7 +711,7 @@ bool dcn_validate_bandwidth(
 
 	const struct resource_pool *pool = dc->res_pool;
 	struct dcn_bw_internal_vars *v = &context->dcn_bw_vars;
-	int i, input_idx;
+	int i, input_idx, k;
 	int vesa_sync_start, asic_blank_end, asic_blank_start;
 	bool bw_limit_pass;
 	float bw_limit;
@@ -1022,6 +1022,43 @@ bool dcn_validate_bandwidth(
 		mode_support_and_system_configuration(v);
 	}
 
+	display_pipe_configuration(v);
+
+	for (k = 0; k <= v->number_of_active_planes - 1; k++) {
+		if (v->source_scan[k] == dcn_bw_hor)
+			v->swath_width_y[k] = v->viewport_width[k] / v->dpp_per_plane[k];
+		else
+			v->swath_width_y[k] = v->viewport_height[k] / v->dpp_per_plane[k];
+	}
+	for (k = 0; k <= v->number_of_active_planes - 1; k++) {
+		if (v->source_pixel_format[k] == dcn_bw_rgb_sub_64) {
+			v->byte_per_pixel_dety[k] = 8.0;
+			v->byte_per_pixel_detc[k] = 0.0;
+		} else if (v->source_pixel_format[k] == dcn_bw_rgb_sub_32) {
+			v->byte_per_pixel_dety[k] = 4.0;
+			v->byte_per_pixel_detc[k] = 0.0;
+		} else if (v->source_pixel_format[k] == dcn_bw_rgb_sub_16) {
+			v->byte_per_pixel_dety[k] = 2.0;
+			v->byte_per_pixel_detc[k] = 0.0;
+		} else if (v->source_pixel_format[k] == dcn_bw_yuv420_sub_8) {
+			v->byte_per_pixel_dety[k] = 1.0;
+			v->byte_per_pixel_detc[k] = 2.0;
+		} else {
+			v->byte_per_pixel_dety[k] = 4.0f / 3.0f;
+			v->byte_per_pixel_detc[k] = 8.0f / 3.0f;
+		}
+	}
+
+	v->total_data_read_bandwidth = 0.0;
+	for (k = 0; k <= v->number_of_active_planes - 1; k++) {
+		v->read_bandwidth_plane_luma[k] = v->swath_width_y[k] * v->dpp_per_plane[k] *
+				dcn_bw_ceil2(v->byte_per_pixel_dety[k], 1.0) / (v->htotal[k] / v->pixel_clock[k]) * v->v_ratio[k];
+		v->read_bandwidth_plane_chroma[k] = v->swath_width_y[k] / 2.0 * v->dpp_per_plane[k] *
+				dcn_bw_ceil2(v->byte_per_pixel_detc[k], 2.0) / (v->htotal[k] / v->pixel_clock[k]) * v->v_ratio[k] / 2.0;
+		v->total_data_read_bandwidth = v->total_data_read_bandwidth +
+				v->read_bandwidth_plane_luma[k] + v->read_bandwidth_plane_chroma[k];
+	}
+
 	BW_VAL_TRACE_END_VOLTAGE_LEVEL();
 
 	if (v->voltage_level != number_of_states_plus_one && !fast_validate) {
-- 
2.7.4

_______________________________________________
amd-gfx mailing list
amd-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx

  parent reply	other threads:[~2019-05-27 19:50 UTC|newest]

Thread overview: 27+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-05-27 19:50 [PATCH 00/26] DC Patches 27 May 2019 sunpeng.li-5C7GfCeVMHo
     [not found] ` <1558986665-12964-1-git-send-email-sunpeng.li-5C7GfCeVMHo@public.gmane.org>
2019-05-27 19:50   ` [PATCH 01/26] drm/amd/display: make clk mgr soc specific sunpeng.li-5C7GfCeVMHo
2019-05-27 19:50   ` [PATCH 02/26] drm/amd/display: Don't use ROM for output TF if GAMMA_CS_TFM_1D sunpeng.li-5C7GfCeVMHo
2019-05-27 19:50   ` [PATCH 03/26] drm/amd/display: Implement CM dealpha and bias interfaces sunpeng.li-5C7GfCeVMHo
2019-05-27 19:50   ` [PATCH 04/26] drm/amd/display: Move CLK_BASE_INNER macro sunpeng.li-5C7GfCeVMHo
2019-05-27 19:50   ` [PATCH 05/26] drm/amd/display: assign new stream id in dc_copy_stream sunpeng.li-5C7GfCeVMHo
2019-05-27 19:50   ` [PATCH 06/26] drm/amd/display: Disable audio stream only if it's currently enabled sunpeng.li-5C7GfCeVMHo
2019-05-27 19:50   ` [PATCH 07/26] drm/amd/display: Add GSL source select registers sunpeng.li-5C7GfCeVMHo
2019-05-27 19:50   ` [PATCH 08/26] drm/amd/display: Ensure DRR triggers in BP sunpeng.li-5C7GfCeVMHo
2019-05-27 19:50   ` [PATCH 09/26] drm/amd/display: disable PSR/ABM before destroy DMCU struct sunpeng.li-5C7GfCeVMHo
2019-05-27 19:50   ` [PATCH 10/26] drm/amd/display: move clk_mgr files to right place sunpeng.li-5C7GfCeVMHo
2019-05-27 19:50   ` [PATCH 11/26] drm/amd/display: Add min_dcfclk_mhz field to bb overrides sunpeng.li-5C7GfCeVMHo
2019-05-27 19:50   ` [PATCH 12/26] drm/amd/display: Move link functions from dc to dc_link sunpeng.li-5C7GfCeVMHo
2019-05-27 19:50   ` [PATCH 13/26] drm/amd/display: 3.2.32 sunpeng.li-5C7GfCeVMHo
2019-05-27 19:50   ` sunpeng.li-5C7GfCeVMHo [this message]
2019-05-27 19:50   ` [PATCH 15/26] drm/amd/display: fix crash on setmode when mode is close to bw limit sunpeng.li-5C7GfCeVMHo
2019-05-27 19:50   ` [PATCH 16/26] drm/amd/display: do not power on eDP power rail early sunpeng.li-5C7GfCeVMHo
2019-05-27 19:50   ` [PATCH 17/26] drm/amd/display: Fix type of pp_smu_wm_set_range struct sunpeng.li-5C7GfCeVMHo
2019-05-27 19:50   ` [PATCH 18/26] drm/amd/display: Refactor clk_mgr functions sunpeng.li-5C7GfCeVMHo
2019-05-27 19:50   ` [PATCH 19/26] drm/amd/display: Add writeback_config to VBA vars sunpeng.li-5C7GfCeVMHo
2019-05-27 19:50   ` [PATCH 20/26] drm/amd/display: Modified AUX_DPHY_RX_CONTROL0 sunpeng.li-5C7GfCeVMHo
2019-05-27 19:51   ` [PATCH 21/26] drm/amd/display: Refactor DIO stream encoder sunpeng.li-5C7GfCeVMHo
2019-05-27 19:51   ` [PATCH 22/26] drm/amd/display: fix issues with bad AUX reply on some displays sunpeng.li-5C7GfCeVMHo
2019-05-27 19:51   ` [PATCH 23/26] drm/amd/display: Increase Backlight Gain Step Size sunpeng.li-5C7GfCeVMHo
2019-05-27 19:51   ` [PATCH 24/26] drm/amd/display: Reset planes for color management changes sunpeng.li-5C7GfCeVMHo
2019-05-27 19:51   ` [PATCH 25/26] drm/amd/display: CS_TFM_1D only applied post EOTF sunpeng.li-5C7GfCeVMHo
2019-05-27 19:51   ` [PATCH 26/26] drm/amd/display: program manual trigger only for bottom most pipe sunpeng.li-5C7GfCeVMHo

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=1558986665-12964-15-git-send-email-sunpeng.li@amd.com \
    --to=sunpeng.li-5c7gfcevmho@public.gmane.org \
    --cc=Su.Chung-5C7GfCeVMHo@public.gmane.org \
    --cc=amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox