| ↑back Search ←Prev date Next date→ Show only urls | (Click on time to select a line by its url) |
| Who | What | When |
|---|---|---|
| *** | _bingbu_ has left | [09:15] |
| ................................................................................................................................................ (idle for 11h59mn) | ||
| vitoriha has left | [21:14] | |
| ............................ (idle for 2h17mn) | ||
| koike | media API question 1: in a topology, if the subdevice node is exposed to userspace as /dev/v4l-subdev0, is the implementation of S_/G_FMT in the whole topology mandatory? Or can /dev/v4l-subdev0 only be exposed to allow userspace to subscribe to an event? | [23:31] |
| media API question 2: the sensor is usually a diferent driver from the ISP (this is because the same sensor could be used with different ISPs in theory), what happen if one ISP expects the sensor to expose /dev/v4l-subdevX with S_/G_FMT, but another ISP doesn't? | [23:40] | |
| pinchartl | koike: from a subdev point of view, whether the get/set format operations are called from userspace or from the ISP driver is irrelevant, subdevices shouldn't see the difference
the5~ ISP can decide to expose subdev device nodes to userspace by calling v4l2_device_register_subdev_nodes() so ISP drivers can decide already this however prevents not exposing the set format operation to userspace, but exposing events for that use case, Jacopo has posted a patch series that introduces a v4l2_device_register_ro_subdev_nodes() the subdev nodes will then be read-only you will be able to perform any operation that doesn't change the state of the subdev get format, set format(TRY), but not set format(ACTIVE) and unless I'm mistaken events are available through ro subdev nodes | [23:52] |
| ↑back Search ←Prev date Next date→ Show only urls | (Click on time to select a line by its url) |