FAILED: patch "[PATCH] drm/i915: Do not enable FEC without DSC" failed to apply to 5.0-stable tree

The patch below does not apply to the 5.0-stable tree.
If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <[email protected]>.

thanks,

greg k-h

------------------ original commit in Linus's tree ------------------

From 5aae7832d1b4ec614996ea0f4fafc4d9855ec0b0 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Ville=20Syrj=C3=A4l=C3=A4?= <[email protected]>
Date: Tue, 26 Mar 2019 16:49:02 +0200
Subject: [PATCH] drm/i915: Do not enable FEC without DSC
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

Currently we enable FEC even when DSC is no used. While that is theoretically valid supposedly there isn't much of a benefit from this. But more importantly we do not account for the FEC link bandwidth overhead (2.4%) in the non-DSC link bandwidth computations. So the code may think we have enough bandwidth when we in fact do not.

index 8891f29a8c7f..48da4a969a0a 100644
@@ -1886,6 +1886,9 @@ static int intel_dp_dsc_compute_config(struct intel_dp *intel_dp,
 	int pipe_bpp;
 	int ret;
 
+	pipe_config->fec_enable = !intel_dp_is_edp(intel_dp) &&
+		intel_dp_supports_fec(intel_dp, pipe_config);
+
 	if (!intel_dp_supports_dsc(intel_dp, pipe_config))
 		return -EINVAL;
 

@@ -2116,9 +2119,6 @@ intel_dp_compute_config(struct intel_encoder *encoder,
 	if (adjusted_mode->flags & DRM_MODE_FLAG_DBLCLK)
 		return -EINVAL;
 
-	pipe_config->fec_enable = !intel_dp_is_edp(intel_dp) &&
-				  intel_dp_supports_fec(intel_dp, pipe_config);
-
 	ret = intel_dp_compute_link_config(encoder, pipe_config, conn_state);
 	if (ret < 0)
 		return ret;

From: Ville Syrjälä <[email protected]>

commit 5aae7832d1b4ec614996ea0f4fafc4d9855ec0b0 upstream.

Currently we enable FEC even when DSC is no used. While that is theoretically valid supposedly there isn't much of a benefit from this. But more importantly we do not account for the FEC link bandwidth overhead (2.4%) in the non-DSC link bandwidth computations. So the code may think we have enough bandwidth when we in fact do not.

---
 drivers/gpu/drm/i915/intel_dp.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

index dcd1df5322e8..21c6016ccba5 100644
@@ -1871,6 +1871,9 @@ static bool intel_dp_dsc_compute_config(struct intel_dp *intel_dp,
 	u8 dsc_max_bpc;
 	int pipe_bpp;
 
+	pipe_config->fec_enable = !intel_dp_is_edp(intel_dp) &&
+		intel_dp_supports_fec(intel_dp, pipe_config);
+
 	if (!intel_dp_supports_dsc(intel_dp, pipe_config))
 		return false;
 

@@ -2097,9 +2100,6 @@ intel_dp_compute_config(struct intel_encoder *encoder,
 	if (adjusted_mode->flags & DRM_MODE_FLAG_DBLCLK)
 		return false;
 
-	pipe_config->fec_enable = !intel_dp_is_edp(intel_dp) &&
-				  intel_dp_supports_fec(intel_dp, pipe_config);
-
 	if (!intel_dp_compute_link_config(encoder, pipe_config, conn_state))
 		return false;
 

2.21.0

Thanks for the backport, now queued up.

greg k-h