On 3/9/20 11:14 AM, Marco Felsch wrote: > Hi all, > > here is the new and _hopefully_ last version :) This version contains > the discussion results with Sakari and some minor fixes from the v11. > > Each patch has a dedicate log, so I avoid the details here and give only > an overview. > > Patch 1-2: > Those where previously one patch. > > Patch 3-6: > Prepare and implement the common v4l2-fwnode parsing code. I changed the > code from a single parse_and_add_links/free to parse/add_links/free. I > did the split to allow adding multiple links from different devices to a > _maybe_ coming global connector device. > > Patch 8: > I converted the parsing code to the new v4l2-fwnode-connector parsing > behaviour and fixed a two bugs during remove() > > Patch 13: > This one is new due to the switch from the s_power() to the pm_runtime > as Sakari suggested. > > Patch 14: > Also a new patch. > > Patch 15: > New patch.. > > I've tested the series on a custom imx6-based board which uses the chip > directly (no-usb). I would appreciate if someone with a usb device can > test this too. I'm getting these sparse/smatch warnings: sparse: WARNINGS SPARSE:drivers/media/i2c/tvp5150.c:1071:14: warning: symbol 'tvp5150_get_hmax' was not declared. Should it be static? drivers/media/i2c/tvp5150.c:1071:14: warning: no previous prototype for 'tvp5150_get_hmax' [-Wmissing-prototypes] smatch: WARNINGS drivers/media/v4l2-core/v4l2-fwnode.c:744 v4l2_fwnode_connector_add_link() warn: possible memory leak of 'link' drivers/media/i2c/tvp5150.c:1071:14: warning: no previous prototype for 'tvp5150_get_hmax' [-Wmissing-prototypes] Please look at these: the tvp5150.c is probably a missing static, so easy to fix, but the 'possible memory leak' definitely needs closer attention. Based on a quick look of that function it appears a correct warning, and the error path there needs to be fixed. Regards, Hans > > Regards, > Marco > > Javier Martinez Canillas (1): > partial revert of "[media] tvp5150: add HW input connectors support" > > Marco Felsch (17): > dt-bindings: connector: analog: add sdtv standards property > dt-bindings: display: add sdtv-standards defines > media: v4l: link dt-bindings and uapi > media: v4l2-fwnode: add endpoint id field to v4l2_fwnode_link > media: v4l2-fwnode: add v4l2_fwnode_connector > media: v4l2-fwnode: add initial connector parsing support > media: tvp5150: add input source selection of_graph support > media: dt-bindings: tvp5150: Add input port connectors DT bindings > media: tvp5150: fix set_selection rectangle handling > media: tvp5150: add FORMAT_TRY support for get/set selection handlers > media: tvp5150: move irq en-/disable into runtime-pm ops > media: tvp5150: add v4l2-event support > media: tvp5150: add subdev open/close callbacks > media: dt-bindings: tvp5150: cleanup bindings stlye > media: dt-bindings: tvp5150: add optional sdtv standards documentation > media: tvp5150: add support to limit sdtv standards > media: tvp5150: make debug output more readable > > Michael Tretter (1): > media: tvp5150: initialize subdev before parsing device tree > > .../display/connector/analog-tv-connector.txt | 6 + > .../devicetree/bindings/media/i2c/tvp5150.txt | 146 +++- > drivers/media/i2c/tvp5150.c | 802 ++++++++++++++---- > drivers/media/v4l2-core/v4l2-fwnode.c | 166 ++++ > include/dt-bindings/display/sdtv-standards.h | 76 ++ > include/dt-bindings/media/tvp5150.h | 2 - > include/media/v4l2-fwnode.h | 143 ++++ > include/uapi/linux/videodev2.h | 4 + > 8 files changed, 1146 insertions(+), 199 deletions(-) > create mode 100644 include/dt-bindings/display/sdtv-standards.h >