[bone] arecord works on the rev A5 bone image, but not on the rev A6

I’ve plugged in a Sony PS EYE web cam in the USB port of a rev A3 bone running the rev A5 Angstrom image. I’m surprised to see the ALSA arecord works!

However, when I upgrade to running the rev A6 image on the same bone, arecord > /tmp/arecord results in no data written to the file.

How do I go about fixing this?

–Mark

contact your TI rep and complain about USB being broken. That’s a serious recommendation, not an attempt at being funny

Let me make sure you understand. In both cases it’s the same A3 board. What’s changing is the version of the image I’m using.

When running the version that ships with the rev A5 board, it works. When running the version that ships with the rev A6 it doesn’t.

In fact, I just switched to a rev A5 board running the image that ships with rev A6 and arecord also doesn’t work.

I think it’s the software…

–Mark

Let me make sure you understand. In both cases it's the same A3 board. What's changing is the version of the image I'm using.

When running the version that ships with the rev A5 board, it works. When running the version that ships with the rev A6 it doesn't.

In fact, I just switched to a rev A5 board running the image that ships with rev A6 and arecord also doesn't work.

I think it's the software...

Yes, so:

contact your TI rep and complain about USB being broken. That's a serious recommendation, not an attempt at being funny

Maybe the modules are not loaded?

My post on video conferencing might help too. → http://blog.engine12.com/?p=507

It references unrelated hardware, but it should work just the same on the beaglebone and beagleboard.

– Mark

Just tested the Eye on my BeagleBone A6, this is output of dmesg:

[ 106.004058] usb 1-1: new high-speed USB device number 2 using musb-hdrc
[ 106.146423] usb 1-1: New USB device found, idVendor=1415, idProduct=2000
[ 106.146453] usb 1-1: New USB device strings: Mfr=1, Product=2, SerialNumber=0
[ 106.146453] usb 1-1: Product: USB Camera-B4.09.24.1
[ 106.146484] usb 1-1: Manufacturer: OmniVision Technologies, Inc.
[ 106.193176] gspca_main: v2.14.0 registered
[ 106.200866] gspca_main: ov534-2.14.0 probing 1415:2000
[ 106.350280] usbcore: registered new interface driver ov534

It is detecting as a USB video device but not the audio part, maybe the older release detected the audio part. I quick search on Google seems lot of people have the same problem on Linux x86, this is probably due udev rules changes that gives priority to the video over audio.

Sorry, brain fart.

I do have also the audio device. I’ll test further.

root@beaglebone:~# cat /proc/asound/cards
0 [CameraB409241 ]: USB-Audio - USB Camera-B4.09.24.1
OmniVision Technologies, Inc. USB Camera-B4.09.24.1 at usb-musb-hdrc.1-1, high

Tested. Same behavior also noted that MUSB interrupts aren´t increasing during the test.

I’ll try 3.2.31 RobertCNelson on my other Bone to see if there any diference.

Lots of rumors of AM3359 USB is broken but nothing concrete. But I do note that it consume much more CPU resources (even with DMA enabled) than other ARM SoC even slower than this one.