[Neganur]

So, I think I made a mistake while copy-pasting from the shell to the code box. Those dates didn't make sense, why should the kernel in LfD's FW be older while using the newer SDK.Checked the dates again while looking at the dmesg output and confirmed my mistake.Sorry for the confusion 1.7.2:

/ $ dmesg | headdmesg | headLinux version 2.6.29-00236-g4f8dbbb-dirty ([email protected]) (gcc version 4.2.0 20070126 (prerelease) (MontaVista 4.2.0-3.0.0.0702771 2007-03-10)) #15 Fri May 21 18:24:36 HKT 2010CPU: ARM926EJ-S [41069265] revision 5 (ARMv5TEJ), cr=00053177CPU: VIVT data cache, VIVT instruction cacheMachine: WMTMemory policy: ECC disabled, Data cache writethroughOn node 0 totalpages: 27904free_area_init_node: node 0, pgdat c08a1be8, node_mem_map c0932000  Normal zone: 218 pages used for memmap  Normal zone: 0 pages reserved  Normal zone: 27686 pages, LIFO [URL=batch:7/]batch:7/[/URL] $ uname -auname -aLinux localhost 2.6.29-00236-g4f8dbbb-dirty #15 Fri May 21 18:24:36 HKT 2010 armv5tejl GNU/Linux/ $

-simenxie1.9.1:

/ $ dmesg | headdmesg | headLinux version 2.6.29-00236-g4f8dbbb-dirty ([email protected]) (gcc version 4.3.2 (Debian 4.3.2-1.1) ) #117 Thu Sep 16 22:50:30 EST 2010CPU: ARM926EJ-S [41069265] revision 5 (ARMv5TEJ), cr=00053177CPU: VIVT data cache, VIVT instruction cacheMachine: WMTMemory policy: ECC disabled, Data cache writethroughOn node 0 totalpages: 27904free_area_init_node: node 0, pgdat c050ead8, node_mem_map c059f000  Normal zone: 218 pages used for memmap  Normal zone: 0 pages reserved  Normal zone: 27686 pages, LIFO [URL=batch:7/]batch:7/[/URL] $ uname -auname -aLinux localhost 2.6.29-00236-g4f8dbbb-dirty #117 Thu Sep 16 22:50:30 EST 2010 armv5tejl GNU/Linux/ $

-gus (that's you right? The CPU frequency scaling you enabled)I'm not sure why, but I can only connect with ADB to my devices via TCP/IP. Took me a while to realise why both would still show up on 'ADB devices' even though I had unplugged them from USB. WiFi was still enabled...is this Windows related?Also, when I saved the output of zcat and diff to a file, the file would not show up from within the Windows file explorer until I remounted the sdcard (pull and replug the usb cable).diff.txtconfig_172.txtconfig_191.txt(the forum ATTACH won't let me upload compressed files?)Ack I'm not used to look at such a diff output, I'm spoiled by git and notepad++.Here's the file though. I can see a lot of differences, mainly things that were added to the 1.9.1.There's a few lines the 1.7.2 had but 1.9.1 notI guess # marks a remark?)# CONFIG_KEYBOARD_WMT is not set# CONFIG_SPI is not set# CONFIG_USB_SERIAL_CONSOLE is not set# CONFIG_USB_SERIAL_GENERIC is not set# CONFIG_USB_SERIAL_PL2303 is not set# CONFIG_USB_USBNET is not setCONFIG_EXTRA_FIRMWARE="" (1.9.1 has "rt2870.dat")CONFIG_VIA_VELOCITY=y (1.9.1 is set to m)CONFIG_VIDEO_BUFFER_SIZE=32768 (1.9.1 has 16384)The only thing, in my eyes, that make a difference is the buffer size.What is that velocity thing?

 

[projectgus]

Yes' date=' sorry. I obviously wasn't paying enough attention in this thread because I didn't realise you were using one of the custom firmwares w/ "my" kernel.

Ack I'm not used to look at such a diff output, I'm spoiled by git and notepad++.

I normally use meld or diffuse, but I'm on a *cough* different *cough* platform to you. I believe (from my brief time developing for Windows) that Beyond Compare is the diff tool of choice, although I think it's a paid product.[quote name='Neganur;79392]The only thing' date=' in my eyes, that make a difference is the buffer size.What is that velocity thing?[/QUOTE'] I think Velocity is a VIA network driver. I don't see anything there that would mean any difference (the buffer size is for video out, not video in.)Which leaves the possibility that the source release from VIA is missing something, but that doesn't seem likely to me. These are standard USB webcams, right?I can try and build you a kernel with all of the v4l options turned on, and see if it makes any difference. Not sure what else would/could have an effect.

 

[Neganur]

*giggle* well I do have Ubuntu here, but the thing keeps killing my graphic driver (nvidia) everytime I let it update stuff.I wish I could say more about the webcams. I bought 5 of them for 1 EUR each in Hong Kong...they're 0.3 megapixel 640*480 YUY2, have a condenser microphone and 6 LED's. I saccrificed one of them to find out what cmos chip they use - it's impossible to get any chinese documentation.

I can try and build you a kernel with all of the v4l options turned on

I'd appreciate this, if it's not too much trouble!

 

[projectgus]

I actually just realised that a while back I compiled all the v4l modules for someone else, actually in the other thread(!) So that should still work, I think.I also realised we're a little ahead of ourselves, we don't know if the problem is in the Linux layer or the Android layer. What output do you get if you plug a webcam in and then immediately run 'dmesg | tail'? Do you have a /dev/video0 device? I also posted some advice about lsusb in the other thread.

 

[Neganur]

I do have a /dev/video0here's some dmesg: (191)

just after booting/ $ dmesg | taildmesg | tailselect 377 (app_process), adj 15, size 2592, to killselect 442 (app_process), adj 15, size 2864, to killsend sigkill to 442 (app_process), adj 15, size 2864select 377 (app_process), adj 15, size 2369, to killsend sigkill to 377 (app_process), adj 15, size 2369select 428 (app_process), adj 15, size 1975, to killsend sigkill to 428 (app_process), adj 15, size 1975Rcv Wcid(1) AddBAReqStart Seq = 00000000ERROR!!! CntlEnqueueForRecv: BlockAck Request frame length size = 44 incorrect/ $usb hub plugged in (keyboard, usb cam)/ $ dmesg | taildmesg | tailuvcvideo: Added control 00000000-0000-0000-0000-000000000001/15 to device 2.2 entity 1uvcvideo: Scanning UVC chain: OT 3 <- PU 2 <- IT 1uvcvideo: Found a valid video chain (1 -> 3).uvcvideo: UVC non compliance - GET_DEF(PROBE) not supported. Enabling workaround.input: USB2.0 PC CAMERA as /devices/pci0000:00/0000:00:04.0/usb1/1-2/1-2.2/1-2.2:1.0/input/input6uvcvideo: UVC device initialized.select 358 (app_process), adj 14, size 2239, to killselect 510 (app_process), adj 14, size 2642, to killselect 540 (app_process), adj 14, size 3627, to killsend sigkill to 540 (app_process), adj 14, size 3627/ $boot with [URL=hubC:UsersMaster]hubC:UsersMaster[/URL] Lou>adb shell/ $ dmesg | taildmesg | tailselect 436 (app_process), adj 15, size 2793, to killsend sigkill to 436 (app_process), adj 15, size 2793select 361 (app_process), adj 15, size 2546, to killselect 409 (app_process), adj 15, size 2559, to killselect 449 (app_process), adj 15, size 2679, to killsend sigkill to 449 (app_process), adj 15, size 2679Rcv Wcid(1) AddBAReqStart Seq = 00000000ERROR!!! CntlEnqueueForRecv: BlockAck Request frame length size = 44 incorrectinit: sys_prop: permission denied [URL=uid:2000]uid:2000[/URL]  [URL=name:adb.connected/]name:adb.connected/[/URL] $starting camera apk (force close)/ $ dmesg | taildmesg | tail633 free pages5924 reserved pages1218 slab pages37262 pages shared766 pages swap cachedehci_hcd 0000:00:04.0: dma_pool_free buffer-32, ffc56000/b20000 (bad dma)ehci_hcd 0000:00:04.0: dma_pool_free buffer-32, ffc74000/b40000 (bad dma)ehci_hcd 0000:00:04.0: dma_pool_free buffer-32, ffc92000/b60000 (bad dma)select 224 (app_process), adj 14, size 3835, to killsend sigkill to 224 (app_process), adj 14, size 3835/ $ dmesg | taildmesg | tailselect 224 (app_process), adj 14, size 3835, to killsend sigkill to 224 (app_process), adj 14, size 3835uvcvideo: uvc_v4l2_openuvcvideo: uvc_v4l2_ioctl(VIDIOC_QUERYCAP)uvcvideo: uvc_v4l2_ioctl(VIDIOC_S_FMT)ERROR!!! CntlEnqueueForRecv: BlockAck Request frame length size = 44 incorrectERROR!!! CntlEnqueueForRecv: BlockAck Request frame length size = 44 incorrect19, flush one!1a, flush one!1f, flush one!/ $ dmesg | taildmesg | tailuvcvideo: uvc_v4l2_releaseuvcvideo: uvc_v4l2_openuvcvideo: uvc_v4l2_read: not implemented.uvcvideo: uvc_v4l2_read: not implemented.uvcvideo: uvc_v4l2_read: not implemented.uvcvideo: uvc_v4l2_read: not implemented.uvcvideo: uvc_v4l2_read: not implemented.uvcvideo: uvc_v4l2_releasebinder: BC_ATTEMPT_ACQUIRE not supportedbinder: 602:603 ioctl c0186201 100ffd70 returned -22unplugging the camusb 1-2.2: USB disconnect, address 4plugging back indmesg | tailuvcvideo: Added control 00000000-0000-0000-0000-000000000101/4 to device 2.4 entity 2uvcvideo: Added control 00000000-0000-0000-0000-000000000101/5 to device 2.4 entity 2uvcvideo: Added control 00000000-0000-0000-0000-000000000001/2 to device 2.4 entity 1uvcvideo: Added control 00000000-0000-0000-0000-000000000001/4 to device 2.4 entity 1uvcvideo: Added control 00000000-0000-0000-0000-000000000001/15 to device 2.4 entity 1uvcvideo: Scanning UVC chain: OT 3 <- PU 2 <- IT 1uvcvideo: Found a valid video chain (1 -> 3).uvcvideo: UVC non compliance - GET_DEF(PROBE) not supported. Enabling workaround.input: USB2.0 PC CAMERA as /devices/pci0000:00/0000:00:04.0/usb1/1-2/1-2.4/1-2.4:1.0/input/input7uvcvideo: UVC device initialized.

And here the working 172

fresh boot, plugging the camera [URL=in:/]in:/[/URL] $ dmesg | taildmesg | tailuvcvideo: Added control 00000000-0000-0000-0000-000000000101/4 to device 2.4 entity 2uvcvideo: Added control 00000000-0000-0000-0000-000000000101/5 to device 2.4 entity 2uvcvideo: Added control 00000000-0000-0000-0000-000000000001/2 to device 2.4 entity 1uvcvideo: Added control 00000000-0000-0000-0000-000000000001/4 to device 2.4 entity 1uvcvideo: Added control 00000000-0000-0000-0000-000000000001/15 to device 2.4 entity 1uvcvideo: Scanning UVC chain: OT 3 <- PU 2 <- IT 1uvcvideo: Found a valid video chain (1 -> 3).uvcvideo: UVC non compliance - GET_DEF(PROBE) not supported. Enabling workaround.input: USB2.0 PC CAMERA as /devices/pci0000:00/0000:00:04.0/usb1/1-2/1-2.4/1-2.4:1.0/input/input6uvcvideo: UVC device initialized.starting Camera360 (I have a picture)/ $ dmesg | taildmesg | tailuvcvideo: Dropping payload (out of sync).uvcvideo: Dropping payload (out of sync).uvcvideo: Dropping payload (out of sync).uvcvideo: Dropping payload (out of sync).uvcvideo: Dropping payload (out of sync).uvcvideo: Dropping payload (out of sync).uvcvideo: Dropping payload (out of sync).uvcvideo: Dropping payload (out of sync).uvcvideo: Dropping payload (out of sync).uvcvideo: Dropping payload (out of sync)./ $/ $ dmesg | taildmesg | tailuvcvideo: uvc_v4l2_ioctl(VIDIOC_QBUF)uvcvideo: Queuing buffer 3.uvcvideo: uvc_v4l2_ioctl(VIDIOC_DQBUF)uvcvideo: Frame complete (EOF found).uvcvideo: EOF in empty payload.uvcvideo: Dequeuing buffer 0 (3, 614400 bytes).uvcvideo: uvc_v4l2_ioctl(VIDIOC_QBUF)uvcvideo: Queuing buffer 0.uvcvideo: Frame complete (EOF found).uvcvideo: EOF in empty payload.unpluggingusb 1-2.4: USB disconnect, address 5plugging back in/ $ dmesg | taildmesg | tailuvcvideo: Added control 00000000-0000-0000-0000-000000000101/4 to device 2.4 entity 2uvcvideo: Added control 00000000-0000-0000-0000-000000000101/5 to device 2.4 entity 2uvcvideo: Added control 00000000-0000-0000-0000-000000000001/2 to device 2.4 entity 1uvcvideo: Added control 00000000-0000-0000-0000-000000000001/4 to device 2.4 entity 1uvcvideo: Added control 00000000-0000-0000-0000-000000000001/15 to device 2.4 entity 1uvcvideo: Scanning UVC chain: OT 3 <- PU 2 <- IT 1uvcvideo: Found a valid video chain (1 -> 3).uvcvideo: UVC non compliance - GET_DEF(PROBE) not supported. Enabling workaround.input: USB2.0 PC CAMERA as /devices/pci0000:00/0000:00:04.0/usb1/1-2/1-2.4/1-2.4:1.0/input/input7uvcvideo: UVC device initialized.

I'll try and log the terminal console while doing the same thing again.EDIT: here is the terminal log.On the third boot, it finally worked and I could take a picture.I then unplugged the camera from the hub and plugged it back but the application didn't recognise any camera anymore.It does that every now and then - I have to reboot the device once or twice before the camera works.Edit2: Another terminal log. This time everything worked without any errors. Booted, started the camera application, took a shot, left the application, started again and shutdown.I'll try downgrading the 191-tablet to an old stock FW.

 

[projectgus]

Hmm. Looks like the kernel has the right video support, at least. So it's either a problem in Android or an oddity in your webcam.The fact that yuou have the problem intermittently makes me wonder if maybe it's a physical problem with the USB connection or something, as well?If you boot into Ubuntu on your desktop machine, does the camera connect OK there?

 

[Neganur]

Yeah both Windows and Linux connect to the cams without any problems. I don't trust the hardware in my tablet though, guess I will open the new tablet just to be sure.From what I've seen so far - the connecter isn't secured properly on the pcb and will break due to wear.The connections on the board I repaired are fine though.Also, I don't see any terminal response to me rocking the cables and twisting the board Perhaps it is Android after all.They (the cameras) use generic drivers, maybe it's the microphone that upsets android?But even that is just a Generic USB Audio device.

 

[Ikarus]

Hi and thanks for Eken M001! Running it from microSD card on my Flytouch.Iam trying to get some webcams work, but i cant get them installed to /dev/video0 etc.. I got message below when i plug my webcam.input: USB2.0 PC CAMERA as /devices/pci0000:00/0000:00:04.0/usb1/1-2/1-2.4/1-2.4:1.0/input/input7uvcvideo: UVC device initialized.Ive tried webcams like some logitech,philips spc900nc and some others. They all work fine with Ubuntu.Maybe someone can help me? My goal is to get this sites.google.com/site/linuxuhvo/ program to work at minimal hardware

 

[projectgus]

It looks like the webcam is being found just fine. AFAIK the Debian build for M001 doesn't have hotplug or devfs (might be wrong) so you need to create device nodes yourself (again, might be wrong - it has been quite a while since I used it.)If you try and create one manually with

mknod /dev/video0 c 81 0

Does that help at all?

 

[lefeudedieu]

Hi,I currently have Froyo on my APAD 128MB 300Mhz.This is the feat of another (topic APAD)http://www.slatedroid.com/apad-firm...1-apad-batch-2-based-cyanogen-mod-alpha1.htmlAnd it runs very well.You can also make a Miracle ?

 

[shishanjiu]

Hi! I?m struggling with eken?s audio driver. The sound played by android?s music player is normal. But when I kill mediaserver, and play wav to /dev/snd/dsp, the sound is wild and it?s frequency is seem to be changed. Does it need some init or config to make the audio device normal and can be play wav through /dev/snd/dsp at command line?