From d56e823ac32a793b24c857e8635def13d42b4008 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <ville.syrjala@linux.intel.com>
Date: Wed, 3 Jul 2019 23:08:22 +0300
Subject: drm/i915: Deal with cpp==8 for g4x watermarks
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Git-commit: d56e823ac32a793b24c857e8635def13d42b4008
Patch-mainline: v5.4-rc1
References: bsc#1152489

Docs tell us that on g4x we have to compute the SR watermarks
using 4 bytes per pixel. I'm going to assume that only applies
to 1 and 2 byte per pixel formats, and not 8 byte per pixel
formats. That seems like a recipe for an insufficient watermark
which could lead to underruns. Use the maximum of the two numbers
instead.

Reviewed-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20190703200824.5971-5-ville.syrjala@linux.intel.com
Acked-by: Thomas Zimmermann <tzimmermann@suse.de>
---
 drivers/gpu/drm/i915/intel_pm.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c
index d10c62d3f10c..87244d8215a7 100644
--- a/drivers/gpu/drm/i915/intel_pm.c
+++ b/drivers/gpu/drm/i915/intel_pm.c
@@ -1116,6 +1116,8 @@ static u16 g4x_compute_wm(const struct intel_crtc_state *crtc_state,
 	if (!intel_wm_plane_visible(crtc_state, plane_state))
 		return 0;
 
+	cpp = plane_state->base.fb->format->cpp[0];
+
 	/*
 	 * Not 100% sure which way ELK should go here as the
 	 * spec only says CL/CTG should assume 32bpp and BW
@@ -1129,9 +1131,7 @@ static u16 g4x_compute_wm(const struct intel_crtc_state *crtc_state,
 	 */
 	if (IS_GM45(dev_priv) && plane->id == PLANE_PRIMARY &&
 	    level != G4X_WM_LEVEL_NORMAL)
-		cpp = 4;
-	else
-		cpp = plane_state->base.fb->format->cpp[0];
+		cpp = max(cpp, 4u);
 
 	clock = adjusted_mode->crtc_clock;
 	htotal = adjusted_mode->crtc_htotal;
-- 
2.28.0