Hi Benoit,

On Mon, Jan 29, 2018 at 01:30:46PM -0600, Benoit Parrot wrote:

quoted

+++ b/drivers/media/platform/cadence/cdns-csi2tx.c

@@ -0,0 +1,586 @@
+/*
+ * Driver for Cadence MIPI-CSI2 TX Controller
+ *
+ * Copyright (C) 2017 Cadence Design Systems Inc.
+ *
+ * This program is free software; you can redistribute  it and/or modify it
+ * under  the terms of  the GNU General  Public License as published by the
+ * Free Software Foundation;  either version 2 of the  License, or (at your
+ * option) any later version.

Should use SPDX license tag line instead.

Indeed.

quoted

+struct csi2tx_priv {
+	struct device			*dev;
+	atomic_t			count;
+
+	void __iomem			*base;
+
+	struct clk			*esc_clk;
+	struct clk			*p_clk;
+	struct clk			*pixel_clk[CSI2TX_STREAMS_MAX];
+
+	struct v4l2_subdev		subdev;
+	struct v4l2_async_notifier	notifier;
+	struct media_pad		pads[CSI2TX_PAD_MAX];
+	struct v4l2_mbus_framefmt	pad_fmts[CSI2TX_PAD_MAX];
+
+	bool				has_internal_dphy;
+	unsigned int			lanes;
+	unsigned int			max_lanes;

So if I understand this correctly the actual lane number list
specified in DT is not actually used? And so the lane numbers are only
derived from the current numbers of lanes specified.

Meaning the following would produce identical setup:
	csi2tx_out: endpoint {
		remote-endpoint = <&remote_in>;
		clock-lanes = <0>;
		data-lanes = <1 2>;
	};

or
	csi2tx_out: endpoint {
		remote-endpoint = <&remote_in>;
		clock-lanes = <0>;
		data-lanes = <2 3>;
	};

But they would be interpreted differently by say the CSI2-RX side, no?

Ah, right, I should totally fix that.

I'll send a new version shortly.

Thanks!
Maxime

-- 
Maxime Ripard, Free Electrons
Embedded Linux and Kernel engineering
http://free-electrons.com