| ↑back Search ←Prev date Next date→ Show only urls | (Click on time to select a line by its url) |
| Who | What | When |
|---|---|---|
| awalls | yeah the gnuradio channel had the full one spamming for about a week
A carefully crafted ban seems to have kept it out. | [00:00] |
| pinchartl | most hosting companies don't seem to deal with abuse reports in any efficient way :-( | [00:01] |
| ......................................... (idle for 3h21mn) | ||
| *** | YuGiOhJCJ has quit IRC (Quit: YuGiOhJCJ) | [03:22] |
| ......................... (idle for 2h0mn) | ||
| koike_ is now known as koike
DaRock has quit IRC (Ping timeout: 245 seconds) | [05:22] | |
| .................... (idle for 1h38mn) | ||
| Nikhil_D has quit IRC (Remote host closed the connection) | [07:01] | |
| .............. (idle for 1h5mn) | ||
| sailus | pinchartl: Or at all... | [08:06] |
| ............ (idle for 56mn) | ||
| larsc | it's their job to do the best to ignore abuse reports | [09:02] |
| .............. (idle for 1h8mn) | ||
| *** | prabhakarlad has quit IRC (Ping timeout: 256 seconds) | [10:10] |
| ...................... (idle for 1h45mn) | ||
| sic_ has quit IRC (*.net *.split)
luisbg has quit IRC (*.net *.split) jj- has quit IRC (*.net *.split) toso has quit IRC (*.net *.split) Juan has quit IRC (*.net *.split) binchen has quit IRC (Ping timeout: 245 seconds) | [11:55] | |
| ........... (idle for 53mn) | ||
| benjiG | hverkuil: hello | [12:50] |
| ........ (idle for 37mn) | ||
| *** | awalls has quit IRC (Ping timeout: 245 seconds) | [13:27] |
| ...... (idle for 27mn) | ||
| faLUCE | hello. A buggy program occasionally hangs on an usb v4l device If I kill the program, even with SIGKILL, its pid remains active, until I phisically disconnect the usb device. Is there a way to detach the usb link without unplugging the device ? I tried: http://askubuntu.com/questions/645/how-do-you-reset-a-usb-device-from-the-command-line but it's quite unportable, and invasive | [13:54] |
| pinchartl | what USB device ? | [13:58] |
| larsc | faLUCE: is it a uvc? | [13:58] |
| faLUCE | larsc: pinchartl Bus 002 Device 003: ID 046d:0825 Logitech, Inc. Webcam C270 | [14:03] |
| pinchartl | that shouldn't happen :-/ | [14:05] |
| larsc | pinchartl: blame the driver author ;) | [14:05] |
| pinchartl | do you know what ioctl the application is blocked on ? | [14:05] |
| faLUCE | pinchartl: no | [14:06] |
| larsc | faLUCE: you can simulate a unplug by going to /sys/bus/usb/drivers/uvcvideo and then do echo your_device > unbind
to simulate replug echo ... > bind but that should only be a workaround the real issue should be fixed | [14:06] |
| faLUCE | larsc: yes, I know.... but it's the only solution, since I don't know if it's a program issue or a driver issue
in each case it's not my issue | [14:07] |
| pinchartl | it can be a driver issue
can you attach gdb to the application after it hangs and get a backtrace ? | [14:07] |
| faLUCE | ok: larsc, what is the token for "your_device" ? the v4l id ?
or bus-dev ? | [14:08] |
| larsc | faLUCE: do a ls in that folder it lists all the devices
if you only have one it is easy to figure out which one it is | [14:09] |
| faLUCE | larsc: it's listed as 2-1 <---- what are 2 and 1 ? | [14:11] |
| larsc | bus-device | [14:11] |
| faLUCE | larsc: then these id is different from what lsusb shows: lsusb shows bus-device 2-4 for the camera
and 2-1 is a root hub more precisely: I have 2-1:1.0 2-1:1.1 in /sys/bus/usb/drivers/uvcvideo | [14:13] |
| larsc | faLUCE: it's the full path
matches the output in lsusb -t | [14:15] |
| faLUCE | larsc: I see. Is there a simple way to retrieve this id from
ls /dev/v4l/by-id/mydeviceid ? | [14:17] |
| larsc | maybe
I don't know | [14:18] |
| faLUCE | larsc: ok thanks | [14:18] |
| pinchartl | faLUCE: can you attach gdb to the application after it hangs and get a backtrace ? | [14:19] |
| faLUCE | pinchartl: not easy, sorry
pinchartl: I could do that if it is a serious issue, but it could have been corrected meanwhile it's hard to reproduce the bug | [14:19] |
| pinchartl | I'd like to fix it if it hasn't been fixed already | [14:21] |
| faLUCE | in addition, the file /sys/bus/usb/drivers/uvcvideo/unbind is not readable even by root
pinchartl: I'll test it and I'll write to you a report when it will hang the next time larsc: are you sure I can write to that file? it has very restrictive permissions | [14:23] |
| pinchartl | faLUCE: thank you. make sure you compile your application with debug symbols if you can to make the backtrace more useful | [14:25] |
| faLUCE | the permissions are --w------- | [14:26] |
| pinchartl | faLUCE: you can write as root, but you can't read | [14:26] |
| faLUCE | pinchartl: I don't want to mess stuff if I write bad things to this file | [14:26] |
| larsc | all you can do is simulate a unplug
nothing else will happen | [14:26] |
| faLUCE | larsc: pinchartl: echo "2-1:1.0" > /sys/bus/usb/drivers/uvcvideo/unbind <---- is it the right syntax ?
or do I need to add some path ? | [14:27] |
| *** | rshanmu has quit IRC (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client) | [14:29] |
| javier__ | faLUCE: yes, it is | [14:29] |
| larsc | that's correct | [14:29] |
| javier__ | faLUCE: if you have strace in your target device, you can also attach to the process and see on which v4l2 ioctl it got hang | [14:30] |
| *** | benjiG has left | [14:31] |
| rshanmu | faLUCE: (or) /proc/<pid>/stack or /proc/<pid>/task/<tid>/stack should give the backtrace. no? | [14:32] |
| faLUCE | rshanmu: now the process is killed... when I have it again with the bug I'll do a backtrace | [14:33] |
| ........ (idle for 39mn) | ||
| the problem is that I can't execute sudo echo "2-1:1.0" > /sys/bus/usb/drivers/uvcvideo/unbind . I need to switch to root user.... why ? | [15:12] | |
| larsc | try '| tee' instead of '>'
'| sudo tee' or sudo sh -c '...' if you do sudo foo > bar foo runs with root rights but the redirect is still done by the shell which is running as your user | [15:15] |
| *** | mmattice has quit IRC (Remote host closed the connection) | [15:17] |
| faLUCE | larsc: sorry, I have to write unbind file inside a python script, which is run by sudo ....
inside the script, I write the to the file using file.write() | [15:17] |
| larsc | yeah, that should work | [15:18] |
| faLUCE | larsc: instead, with cli, which is the syntax for tee? sudo echo "2-1:1.0" | tee /sys/bus/usb/drivers/uvcvideo/bin ? | [15:19] |
| larsc | sudo before tee instead of echo | [15:21] |
| faLUCE | sudo tee "2-1:1.0" | /sys/bus/usb/drivers/uvcvideo/unbind
bash: /sys/bus/usb/drivers/uvcvideo/unbind: Permission denied | [15:22] |
| pinchartl | echo "2-1:1.0" | sudo tee /sys/bus/usb/drivers/uvcvideo/unbind | [15:22] |
| javier__ | faLUCE: sudo sh -c "echo 2-1:1.0 > /sys/bus/usb/drivers/uvcvideo/unbind" works as well | [15:22] |
| jmleo | pinchartl, hi there !
best wishes !! I have a question regarding DRM and video modes. I have two touchscreen tablets with HDMI input (well, VGA input, but a HDMI->VGA converter between those and the device), and I want to specify the correct fb definition for the screen. I am right now using the video= parameter, but I need two resolutions (one is 800x600@60 and the other is 1280x800@60) | [15:27] |
| pinchartl | jmleo: onnellista uutta vuotta ! | [15:29] |
| jmleo | pinchartl, the issue is the following: when I am not specifying this parameter, the device (an i.MX6 with etnaviv GPU) is sometimes crashing odly | [15:33] |
| pinchartl | jmleo: debug & fix ? :-) | [15:35] |
| jmleo | I don't have time (the client won't pay for it) and so, this workaround is good enough... :/
but I am stuck with this two display devices thing... | [15:36] |
| *** | koike has quit IRC (Quit: ZNC - 1.6.0 - http://znc.in) | [15:39] |
| pinchartl | jmleo: I don't know what to tell you, if you don't have time to fix the problem but still need it to be fixed... hope ? :-) | [15:42] |
| jmleo | pinchartl, what would you activate to debug this part in the kernel ?
I will try to explain that the only way to correct the issue is to find the bug :p | [15:46] |
| pinchartl | how does it "crash oddly" ? | [15:48] |
| jmleo | I have a midori browser (don't ask me why... :/) and when I move around the window, resize it, move it, etc. the kernel will stuck with a freezed screen
I is based on X11+fluxbox It* | [15:50] |
| pinchartl | does the whole system freeze, or just the display ? | [15:50] |
| jmleo | well, I can go to a text console, but then, nothing more is possible (even the magic sysrq)
but I need to verify the serial console | [15:50] |
| pinchartl | please do | [15:52] |
| ....... (idle for 32mn) | ||
| jmleo | pinchartl, can't reproduce it... this bug sucks, I knew it... | [16:24] |
| ...... (idle for 29mn) | ||
| faLUCE | I searched.... but could not find an easy way to obtain the tuple bus-port-device for a given id of a v4l device.... do you know any way for obtaining that ?
the problem is that lsusb -t doesn't work with -d option | [16:53] |
| pinchartl | faLUCE: basename $(readlink /sys/class/video4linux/video0/device) | [16:59] |
| faLUCE | pinchartl: my ids are the ids listed in /dev/v4l/by-id or devicesId in lsusb
I could combine two commands yes I can do that, thanks pinchartl | [17:02] |
| pinchartl | basename $(readlink /sys/class/video4linux/$(basename $(readlink /dev/v4l/by-id/usb-046d_080a-video-index0))/device) | [17:04] |
| faLUCE | yes | [17:07] |
| *** | cybrNaut has quit IRC (Changing host) | [17:09] |
| .... (idle for 18mn) | ||
| _abbenormal has quit IRC (Quit: Yup Im Leaving) | [17:27] | |
| ........ (idle for 39mn) | ||
| awalls1 has left | [18:06] | |
| .............................. (idle for 2h27mn) | ||
| mchehab has quit IRC (Ping timeout: 260 seconds) | [20:33] | |
| ........ (idle for 36mn) | ||
| kbingham has quit IRC (Quit: ZNC - http://znc.in) | [21:09] | |
| ↑back Search ←Prev date Next date→ Show only urls | (Click on time to select a line by its url) |