| ↑back Search ←Prev date Next date→ Show only urls | (Click on time to select a line by its url) |
| Who | What | When |
|---|---|---|
| montjoie | hello, in OUTPUT mode, how to know the size of data in a vb2 ? payload ? in my case I got payload=0 via vb2_get_plane_payload(vb, 0) | [07:48] |
| hverkuil | montjoie: does userspace set bytesused to the payload size? | [07:50] |
| montjoie | it is ffmpeg, I believe yes
quick grep in sources say yes | [07:52] |
| ................. (idle for 1h20mn) | ||
| hverkuil | mchehab: ping | [09:14] |
| montjoie | another question, input and output should be in two separate device ? | [09:24] |
| mchehab | hverkuil: pong | [09:26] |
| hverkuil | mchehab: Can you update smatch to the latest from the smatch git repo? It looks like it fixes the incorrect error it gives for imx290.c | [09:30] |
| pH5: thanks for reviewing those three patches from Krzysztof! | [09:39] | |
| ................ (idle for 1h17mn) | ||
| jernej | mripard: paulk-leonov: hverkuil: What is status of Cedrus VP8 patch (https://patchwork.linuxtv.org/patch/63962/)? | [10:56] |
| paulk-leonov | jernej: I have yet to review it, sorry
along with other stateless decoding and cedrus related patches | [10:58] |
| ........................ (idle for 1h59mn) | ||
| arnd | hverkuil: can you send me a patch to soc@kernel.org to remove CONFIG_SOC_CAMERA from all arch/arm/config/*defconfig files? | [12:57] |
| hverkuil | arnd: will do. | [13:01] |
| arnd | thanks! | [13:01] |
| hverkuil | arnd: done. (Also in a separate patch for arch/sh/configs) | [13:12] |
| ....................... (idle for 1h53mn) | ||
| dafna2 | hverkuil: ping | [15:05] |
| .... (idle for 17mn) | ||
| *** | benjiG has left | [15:22] |
| ...... (idle for 27mn) | ||
| ezequielg | Kwiboo: I'm seeing some nice stuff in your rkvdec work-in-progress github branch. do you have any plans to submit that?
there are a few lines for work there. some patches could be picked more easy, e.g. support for high-10. | [15:49] |
| .... (idle for 18mn) | ||
| Kwiboo | ezequielg: I was planing on submitting something some weeks ago but work and other stuff got in the way, I am hoping to find some time this weekend to get a high 10/422 series out (and a v2 for initial drm part) | [16:08] |
| ezequielg | Kwiboo: I'll be posting a patchset to refuse unsupported bitstreams. Given currently high 10/422 is not supported, the patch will reject them. | [16:16] |
| Kwiboo | from my pov there is two parts missing related to the high10/422 support: pixel format validation (block try_format of 8-bit pixel format for 10-bit content etc) and validation of pic_width_in_mbs_minus1/pic_height_in_map_units_minus1 so they can be used instead of buffer width/heith
sounds great, then that will be something to build on :-) | [16:16] |
| ezequielg | Cool. | [16:17] |
| ................................... (idle for 2h53mn) | ||
| ndufresne | hverkuil: ezequielg: Not sure who maintains it, but got a crash with fwht decoder (vicodec) when running the flushing sequence, gst-play-1.0 testbin://video --videosink="v4l2fwhtenc ! v4l2fwtdec ! fakevideosink" and then press 0 to see to the start
(I'm not sure which gst version is needed, but it only occure with the decoder, and when doing a flush, so I suspect it does not liek a quick streamoff, requeue, streamon) If you care, I can likely repro with a virtme that has text console, so I could copy paste the crash report from the kernel in theory, if I had the know how, this is crashed in qemu, so one could use gdb to look it up | [19:10] |
| ↑back Search ←Prev date Next date→ Show only urls | (Click on time to select a line by its url) |