| ↑back Search ←Prev date Next date→ Show only urls | (Click on time to select a line by its url) |
| Who | What | When |
|---|---|---|
| paulk-leonov | hi
a general (and quite basic) question about the v4l2 api: is userspace expected to fill num_planes on s_fmt (in mplane context), or should the driver set it based on the pixel format provided by userspace? | [13:26] |
| hverkuil | The driver sets it based on the pixel format.
Happens in the queue_setup vb2 callback. | [13:29] |
| paulk-leonov | ok, looks like this might be a bug with hantro then
thanks hverkuil | [13:31] |
| tfiga | hverkuil: how does it happen in queue_setup? That is already much later than S_FMT, when buffers are being created.
That said, for S_FMT it's always the driver responsibility to determine the final format returned to the application | [13:43] |
| .......... (idle for 46mn) | ||
| *** | benjiG has left | [14:30] |
| hverkuil | paulk-leonov: tfiga: sorry, num_planes is set in queue_setup, but that's for struct v4l2_buffer. Sorry, got confused.
Drivers set this in g/s/try_format. | [14:32] |
| ............... (idle for 1h13mn) | ||
| paulk-leonov | my mistake, no bug in hantro, only that it resets output format after setting capture format, which makes sense | [15:46] |
| ↑back Search ←Prev date Next date→ Show only urls | (Click on time to select a line by its url) |