07:59 <thumbcore[m]> It seems to me like it could be simpler if the DTS simply stated `compatible = "nvidia,imx219";` and did _not_ specify camera modes. The driver would provide an interface for querying that mode information, so there's just one source of truth.

07:59 <thumbcore[m]> Why is camera data basically replicated in both a driver (the mode data tables) and the board's device tree? It's even worse in the device tree because that mode data is replicated for each instance of the camera!

08:08 <kwizart> thumbcore[m], is this upstream context ? (see topic)

15:13 <thumbcore[m]> My question is asking about implementation decisions in L4T. Topic says "NVIDIA Tegra discussion" so this seems like it fits. Did I misinterpret? Apologies if so.

15:27 <kwizart> thumbcore[m], you will have more chance to ask on the linked forum, here is mostly addressed concern for upstream kernel tegra dev...