We are currently restricted when it comes to supporting DSI on devices that have a non-DSI control bus. For example, DSI encoder chips are available in the market that are configured via i2c. Configuring their registers via DSI bus is either optional or not available at all.
These devices still need to pass DSI parameters (data lanes, mode flags etc) to the DSI host they are connected to. We don't have a way to do that at the moment.
After some discussions on the previous RFC[1], we decided to support this by providing additional API in drm_mipi_dsi which lets us create new DSI devices without the need of them to have a DT node.
There are a couple of concerns I have about this patchset which I couldn't figure out a solution to (they are addressed in the individual patches too):
- Matching of non-DT devices: buses like i2c/spi populate id_table in their drivers to perform a match for the non-DT case. Is this something that we should do too?
- Race between host_unregister and device_unregister: The new API provides peripheral drivers to unregister the DSI devices they created by using mipi_dsi_device_unregister. If the dsi host unregisters first, the peripheral driver might have an unregistered device pointer without being aware of it.
Some comments on these would help.
[1]: https://lkml.org/lkml/2015/6/30/42
Archit Taneja (5): drm/dsi: Refactor device creation drm/dsi: Try to match non-DT dsi devices drm/dsi: Check for used channels drm/dsi: Add routine to unregister dsi device drm/dsi: Get DSI host by DT device node
drivers/gpu/drm/drm_mipi_dsi.c | 135 ++++++++++++++++++++++++++++++++--------- include/drm/drm_mipi_dsi.h | 27 +++++++++ 2 files changed, 132 insertions(+), 30 deletions(-)