Laurent, you are just remarkable!

Camera worked, just after the patch was applied!
I tested it using lucview, though no controls are working...

here's dmesg;

[ 2910.636000] usb 2-1: new full speed USB device using uhci_hcd and 
address 5
[ 2910.812000] usb 2-1: configuration #1 chosen from 1 choice
[ 2910.816000] uvcvideo: Probing generic UVC device 1
[ 2910.816000] uvcvideo: Found format MJPEG.
[ 2910.816000] uvcvideo: - 320x240 (15.0 fps)
[ 2910.816000] uvcvideo: Found UVC 1.10 device au W33SA (0474:0722)
[ 2910.816000] uvcvideo: Scanning UVC chain: OT 3 <- PU 2 <- IT 1
[ 2910.816000] uvcvideo: Found a valid video chain (1 -> 3).
[ 2910.820000] uvcvideo: UVC device initialized.
[ 2914.108000] uvcvideo: uvc_v4l2_open
[ 2914.108000] uvcvideo: uvc_v4l2_ioctl
[ 2914.108000] v4l2 ioctl VIDIOC_QUERYCAP, dir=r- (0x80685600)
[ 2914.108000] uvcvideo: uvc_v4l2_ioctl
[ 2914.108000] v4l2 ioctl VIDIOC_S_FMT, dir=rw (0xc0cc5605)
[ 2914.108000] uvcvideo: Trying format 0x47504a4d (MJPG): 320x240.
[ 2914.108000] uvcvideo: Using default frame interval 66666.6 us (15.0 fps).
[ 2914.124000] uvcvideo: uvc_v4l2_ioctl
[ 2914.124000] v4l2 ioctl VIDIOC_S_PARM, dir=rw (0xc0cc5616)
[ 2914.124000] uvcvideo: Setting frame interval to 1/15 (666666).
[ 2914.160000] uvcvideo: uvc_v4l2_ioctl
[ 2914.160000] v4l2 ioctl VIDIOC_REQBUFS, dir=rw (0xc0145608)
[ 2914.160000] uvcvideo: uvc_v4l2_ioctl
[ 2914.160000] v4l2 ioctl VIDIOC_QUERYBUF, dir=rw (0xc0445609)
[ 2914.160000] uvcvideo: uvc_v4l2_mmap
[ 2914.160000] uvcvideo: uvc_v4l2_ioctl
[ 2914.160000] v4l2 ioctl VIDIOC_QUERYBUF, dir=rw (0xc0445609)
[ 2914.160000] uvcvideo: uvc_v4l2_mmap
[ 2914.160000] uvcvideo: uvc_v4l2_ioctl
[ 2914.160000] v4l2 ioctl VIDIOC_QUERYBUF, dir=rw (0xc0445609)
[ 2914.160000] uvcvideo: uvc_v4l2_mmap
[ 2914.160000] uvcvideo: uvc_v4l2_ioctl
[ 2914.160000] v4l2 ioctl VIDIOC_QUERYBUF, dir=rw (0xc0445609)
[ 2914.164000] uvcvideo: uvc_v4l2_mmap
[ 2914.164000] uvcvideo: uvc_v4l2_ioctl
[ 2914.164000] v4l2 ioctl VIDIOC_QBUF, dir=rw (0xc044560f)
[ 2914.164000] uvcvideo: Queuing buffer 0.
[ 2914.164000] uvcvideo: uvc_v4l2_ioctl
[ 2914.164000] v4l2 ioctl VIDIOC_QBUF, dir=rw (0xc044560f)
[ 2914.164000] uvcvideo: Queuing buffer 1.
[ 2914.164000] uvcvideo: uvc_v4l2_ioctl
[ 2914.164000] v4l2 ioctl VIDIOC_QBUF, dir=rw (0xc044560f)
[ 2914.164000] uvcvideo: Queuing buffer 2.
[ 2914.164000] uvcvideo: uvc_v4l2_ioctl
[ 2914.164000] v4l2 ioctl VIDIOC_QBUF, dir=rw (0xc044560f)
[ 2914.164000] uvcvideo: Queuing buffer 3.
[ 2914.380000] uvcvideo: uvc_v4l2_ioctl
[ 2914.380000] v4l2 ioctl VIDIOC_STREAMON, dir=-w (0x40045612)
[ 2914.380000] uvcvideo: uvc_v4l2_ioctl
[ 2914.380000] v4l2 ioctl VIDIOC_DQBUF, dir=rw (0xc0445611)
[ 2914.380000] uvcvideo: Dequeuing buffer 0.
[ 2914.676000] uvcvideo: Frame complete (EOF found).
...

Why all these details in kernel messages? Just for the sake of the test?

How can i set it up to work in gstreamer?

I have the following error in gstreamer-properties;

gstreamer-properties-Message: Error running pipeline 'Video for Linux 2 
(v4l2)': Could not negotiate format [gstbasesrc.c(2062): 
gst_base_src_start (): /pipeline0/v4l2src3:
Check your filtered caps, if any]

Thanks again for your hard work! This means one more camera can work in 
Linux! Actually, there is a whole line of Sanyo (au) cellphones 
available here in japan, who probably use same or very similar camera 
USB interface...


Samir


Laurent Pinchart wrote:
> Hi Samir,
> 
> On Wednesday 28 November 2007, Samir Ibradzic wrote:
>> Hello Laurent,
>>
>> I've tried the USB capture thing... Used SnoopyPro, and because of the
>> size of the files i will attach them only to you personally... The
>> reason for the big file size is that i started capture before plugging
>> the cable, then i tested the video stream in Media Player Classics for a
>> secund or two... The reason because there are two files is that there
>> were two devices in SnoopyPro, with similar IDs, but somewhat similar
>> capture results;
> 
> Thanks for the logs. The binary format scared me, but I found a tool to dump 
> the logs to text (http://www.stillhq.com/usblogdump/).
> 
>> USB\Vid_0474&Pid_0722&Rev_0000&MI_00,USB\Vid_0474&Pid_0722&MI_00  -  USB
>> Video Device
>>
>> and
>>
>> USB\Vid_0474&Pid_0722&Rev_0000,USB\Vid_0474&Pid_0722  -  USB Composite
>> Device
>>
>> What is the deal with these two?
> 
> That's just the way Windows deals with devices that have more than one 
> function. Sniffing the composite device will probably log all URBs regardless 
> of the function (camera, ...), while sniffing the video device will only log 
> video-related URBs. Don't take this for granted as I'm no Windows expert, but 
> I expect it to work like that.
> 
>> I also have to mention that when i plug this phone into USB, I get the
>> menu on it, where i select if it should work as USB camera, Storage
>> Device, USB Print device or Direct Comm. Transfer mode. Prior to
>> selecting USB camera mode there is no activity on USB, afaik...
>>
>> Thanks again! I just hope we will come up with something...
> 
> Could you please try the attached patch ? You can discard the previous ones.
> 
> Best regards,
> 
> Laurent Pinchart
> 
> 
> --- AV & Spam Filtering by M+Guardian - Risk Free Email (TM) ---
> 
_______________________________________________
Linux-uvc-devel mailing list
[email protected]
https://lists.berlios.de/mailman/listinfo/linux-uvc-devel

Reply via email to