| ↑back Search ←Prev date Next date→ Show only urls | (Click on time to select a line by its url) |
| Who | What | When |
|---|---|---|
| *** | pinchartl has quit IRC (Ping timeout: 248 seconds) | [06:29] |
| ............... (idle for 1h10mn) | ||
| tfiga has quit IRC (Quit: http://quassel-irc.org - Chat comfortably. Anywhere.) | [07:39] | |
| ...................... (idle for 1h49mn) | ||
| hfr has left | [09:28] | |
| .............. (idle for 1h9mn) | ||
| bbrezillon | pH5: just to be sure, you have this patch applied https://github.com/bbrezillon/linux/commit/af4d014f19d0f21a9ce8ea642a8d26990c1b176d, right? | [10:37] |
| pH5 | bbrezillon: good catch, that should be the culprit. I'm missing the fixup. | [10:51] |
| bbrezillon | if you were seeing iommu page faults, that should be it
but maybe you don't have an iommu in the first place :) | [10:53] |
| ..... (idle for 22mn) | ||
| llm | i've got a freeze problem with a Cypress FX3 USB Camera (Full HD at 33fps), sometimes the select waiting for an image timeouts (2-20seconds) after that every call to select() will timeout, if i close/open the device again select will work again, any ideas why close/open of the device "reduces" the problem?
the camera hangs every 30-100sec, im using the capture example of the v4l project reopening the device is just a temporary fix because i think the cypress attached microcontroller is the problem here, but why does it help to reopen - does then everything in the driver gets reinitalized? fresh Ubuntu 18.04.6 with all updates, Kernel 4.18.0-25-generic | [11:15] |
| ......... (idle for 41mn) | ||
| *** | LazyGrizzly has left | [12:02] |
| ..... (idle for 21mn) | ||
| pinchartl | llm: any relevant message in the kernel log ? | [12:23] |
| llm | im waiting to get my changes deployed - then i can see if there are any related msg at the freeze time, any special log or just dmesg? | [12:30] |
| pinchartl | just in dmesg | [12:36] |
| *** | _jmleo has quit IRC (Ping timeout: 244 seconds) | [12:47] |
| ....... (idle for 32mn) | ||
| ezequielg | hverkuil: I was thinking about m2m-deinterlace. One option would be to add support for a simple memcopy mode (no deinterlace). That would increase the number of DMA engines you can use, and thus make testing easier. It would be a mode solely for testing the driver and make sure it's still working. | [13:19] |
| llm | uvcvideo: Non-zero status (-71) in video completion handler. on every open on the device, nearby the freeze - nothing more | [13:19] |
| pinchartl | llm: that seems to indicate a USB issue. possibly a hardware problem, or a bug in the USB controller driver (the likelyhood of that depends on the hardware, ranging from very unlikely on x86 to very likely on a raspberry pi) | [13:32] |
| llm | x86 so very unlikely | [13:34] |
| .... (idle for 17mn) | ||
| pH5 | ezequielg: do you have an updated ffmpeg branch corresponding to the yesterday's v2 VP8 patches?
bbrezillon: at least yesterday's broken test streams seem to work now that I have integrated all H.264 patches. | [13:51] |
| ezequielg | pH5: I should, yes. | [13:53] |
| pH5 | I have just seen the kernel crash in hantro_g1_vp8_dec_run (with the old, mis-matching patched ffmpeg). | [13:54] |
| bbrezillon | pH5: ezequielg's version is likely to not work with H264 though :) | [13:57] |
| ezequielg | pH5: i should integrate bbrezillon's and push ffmpeg for you to test both. | [13:57] |
| bbrezillon | ezequielg: that'd be awesome
and we really need to clean that up and submit the changes at some point | [13:58] |
| pH5 | ezequielg: that would be awesome, right now I have use bbrezillon's rk-vpu-h264-vp8 branch rebased onto 4.1.3
used* | [13:58] |
| bbrezillon | hm, maybe not submit it, since the uAPI is not definitive yet | [13:59] |
| sailus | mchehab: Are we planning something for ELc-E this year? Or just LPC? | [14:04] |
| pinchartl | llm: a bug in the device is more likely | [14:16] |
| llm | i also think its a bug, hope the responsible developer will find the cause, thx | [14:17] |
| *** | benjiG has left | [14:24] |
| ......... (idle for 43mn) | ||
| shuah | hverkuil: Hi Hans! streaming isn't working on vimc starting 5.2 - is anybody running into this or just me? | [15:07] |
| koike | shuah: I'll check in a moment | [15:08] |
| shuah | okay thanks - I isolated to this commit - commit b6c61a6c37317efd7327199bfe24770af3d7e799
Author: Helen Fornazier <helen.koike@collabora.com> Date: Wed Mar 13 14:29:37 2019 -0400 media: vimc: propagate pixel format in the stream It works if I revert it | [15:08] |
| koike | yes, I was discussing it with tonyk some minutes ago, this also needs to be reverted for design reasons | [15:09] |
| shuah | okay | [15:10] |
| koike | shuah: thanks for checking this | [15:10] |
| shuah | koike: are you planning to send revert soon, if you do can you cc me or I can send it | [15:15] |
| koike | shuah: I appreciate it if you could send it | [15:17] |
| shuah | koike: I will do that - I am stuck anyway. I can't test my vimc panic fix without this revert :) | [15:19] |
| koike | shuah: thanks a lot! | [15:19] |
| shuah | Also - don't worry about spending time to review the patch series I sent. I am looking into fixing this within vimc based on pinchartl and hverkuil feedback | [15:20] |
| ↑back Search ←Prev date Next date→ Show only urls | (Click on time to select a line by its url) |