Jason, Thanks.
What are you referring to with regards to "making it continue to work with older kernels"? As far as understand it, the kernel on my BBAI is updated to the latest 4.14 kernel; well at least to 4.14.108-ti-r131 from a few days ago.. Is this supported on 4.19 or later now? Cheers, Jon On Fri, Mar 27, 2020 at 11:17 AM Jason Kridner <jkrid...@beagleboard.org> wrote: > > > On Fri, Mar 27, 2020 at 7:07 AM Jason Kridner <jkrid...@beagleboard.org> > wrote: > >> Any thoughts on making it more generic? Select the highest index? >> >> On Thu, Mar 26, 2020 at 10:22 PM jonnymo <jonnym...@gmail.com> wrote: >> >>> In the GitHub issue that was filed, I noted the following which seems to >>> get the camera active again: >>> >>> >>> >>> Setting '-d /dev/video1' in the common Makefile seemed to do the trick. >>> I now have video from the camera in the TIDL example. >>> >>> This is what I changed at line 170 in the Makefile at: >>> /var/lib/cloud9/common$ >>> >>> else ifeq ($(PROC),tidl) >>> ti-mct-heap-check -c >>> sudo mjpg_streamer -i "input_opencv.so -d /dev/video1 -r 640x480 >>> --filter ./$(TARGET)$(EXE)" -o "output_http.so -p 8080 -w >>> /usr/share/mjpg-streamer/www" >>> else >>> >>> >>> > Perhaps not the best, but here's my stab at making it continue to work > with older kernels: > > commit 06c70c2b6a9573a6c72968bce725d30d09cbfe8a > Author: Jason Kridner <j...@ti.com> > Date: Fri Mar 27 18:13:10 2020 +0000 > > Remove usage of VPE in running TIDL examples > > diff --git a/common/Makefile b/common/Makefile > index 5a2b886..c1a0982 100644 > --- a/common/Makefile > +++ b/common/Makefile > @@ -167,7 +167,8 @@ ifneq ($(PRU_DIR),) > @echo start > $(PRU_DIR)/state > else ifeq ($(PROC),tidl) > ti-mct-heap-check -c > - sudo mjpg_streamer -i "input_opencv.so -r 640x480 --filter > ./$(TARGET)$(EXE)" -o "output_http.so -p 8090 -w /usr/share/mjpg-streamer/www" > + sudo mjpg_streamer -i "input_opencv.so -r 640x480 -d /dev/$(shell fgrep > -v vpe /sys/class/video4linux/video*/name | perl -ne '/\/(video\d+)\/name/ && > print $$1') \ > + --filter ./$(TARGET)$(EXE)" -o "output_http.so -p 8090 -w > /usr/share/mjpg-streamer/www" > else > ./$(TARGET)$(EXE) > endif > > > As I look at this, I think it would fail if there were just 2 non-vpe > interfaces. > > There has to be a relatively easy way to pick the first webcam. > > > >> >>> On Thu, Mar 26, 2020 at 9:20 AM Robert Nelson <robertcnel...@gmail.com> >>> wrote: >>> >>>> On Thu, Mar 26, 2020 at 10:27 AM <masterdo...@gmail.com> wrote: >>>> > >>>> > Hi, >>>> > I just came across this post that is relevant to the same problem. I >>>> just started working on this Beaglebone AI two days ago. I got the latest >>>> os image and updates and upgrades everything. I tried >>>> classification.tidl.cpp and I got the same last message that Jon Morss >>>> got. And your latest message was about /dev/video0. On my board I got >>>> /dev/video0, /dev/video1 and it does recognize the camera. Do you mean that >>>> I can fix it with the change to the default dev for the camera ? And where >>>> can I change that? I worked with OpenCV and this VideoCapture::open() >>>> failed message seems to indicate a wrong /dev/video index? >>>> >>>> Please submit a bug to: >>>> >>>> https://github.com/beagleboard/cloud9-examples >>>> >>>> the classification demo should allow you to specify a video offset.. >>>> >>>> Regards, >>>> >>>> -- >>>> Robert Nelson >>>> https://rcn-ee.com/ >>>> >>>> -- >>>> For more options, visit http://beagleboard.org/discuss >>>> --- >>>> You received this message because you are subscribed to the Google >>>> Groups "BeagleBoard" group. >>>> To unsubscribe from this group and stop receiving emails from it, send >>>> an email to beagleboard+unsubscr...@googlegroups.com. >>>> To view this discussion on the web visit >>>> https://groups.google.com/d/msgid/beagleboard/CAOCHtYjyA8yTBWKs-%3DFUoE4kh20XgiQNpheJLfnzHG7_ouqVNA%40mail.gmail.com >>>> . >>>> >>> -- >>> For more options, visit http://beagleboard.org/discuss >>> --- >>> You received this message because you are subscribed to the Google >>> Groups "BeagleBoard" group. >>> To unsubscribe from this group and stop receiving emails from it, send >>> an email to beagleboard+unsubscr...@googlegroups.com. >>> To view this discussion on the web visit >>> https://groups.google.com/d/msgid/beagleboard/CAG99bko9Nkz-Jb%2BnBYd-RvhBzGhNHJMkv1%2B7bVcn6hLdaqDcTA%40mail.gmail.com >>> <https://groups.google.com/d/msgid/beagleboard/CAG99bko9Nkz-Jb%2BnBYd-RvhBzGhNHJMkv1%2B7bVcn6hLdaqDcTA%40mail.gmail.com?utm_medium=email&utm_source=footer> >>> . >>> >> -- >> https://beagleboard.org/about - a 501c3 non-profit educating around open >> hardware computing >> > > > -- > https://beagleboard.org/about - a 501c3 non-profit educating around open > hardware computing > > -- > For more options, visit http://beagleboard.org/discuss > --- > You received this message because you are subscribed to the Google Groups > "BeagleBoard" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to beagleboard+unsubscr...@googlegroups.com. > To view this discussion on the web visit > https://groups.google.com/d/msgid/beagleboard/CA%2BT6QPkAeEiaNK-ULejZqiUR6pZuTt2J5egr20BM%2BC7%2BDLfdug%40mail.gmail.com > <https://groups.google.com/d/msgid/beagleboard/CA%2BT6QPkAeEiaNK-ULejZqiUR6pZuTt2J5egr20BM%2BC7%2BDLfdug%40mail.gmail.com?utm_medium=email&utm_source=footer> > . > -- For more options, visit http://beagleboard.org/discuss --- You received this message because you are subscribed to the Google Groups "BeagleBoard" group. To unsubscribe from this group and stop receiving emails from it, send an email to beagleboard+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/beagleboard/CAG99bkr%3DPxgWX%3D0QpkFpHhAq99UsUugLHD0VQfYA7e3HxmpTBw%40mail.gmail.com.
