Related
In my Captivate I noticed a kernel module with name Si4709_driver.ko
Si4709 is a FM radio receiver www .silabs. com /products/audiovideo/fmreceivers/Pages/Si470809.aspx (sorry, I can't post URL)
By default it is not loaded, but when I tried to load it, it did. Now we need to know what API is used for user mode programs.
It would also help if those who put i9000 ROM into Captivate (but have stock kernel) tried load module and then run FM radio app.
# cd /lib
# ls
modules
# cd modules
# ls
dpram.ko
rfs_glue.ko
j4fs.ko
multipdp.ko
param.ko
vibrator.ko
Si4709_driver.ko
dhd.ko
rfs_fat.ko
stgloc
s3c_bc.ko
pvrsrvkm.ko
scsi_wait_scan.ko
fsr.ko
drop_caches.ko
fsr_stl.ko
s3c_lcd.ko
# insmod /lib/modules/Si4709_driver.ko
# lsmod
Si4709_driver 15744 0 - Live 0xbf195000
vibrator 6744 2 - Live 0xbf191000
s3c_bc 2880 0 - Live 0xbf18b000
s3c_lcd 5060 0 - Live 0xbf184000
pvrsrvkm 205996 48 s3c_bc,s3c_lcd, Live 0xbf14c000
multipdp 30968 1 - Live 0xbf13f000
dpram 194152 10 multipdp, Live 0xbf10a000
param 10132 0 - Live 0xbf105000 (P)
j4fs 65604 1 - Live 0xbf0ef000 (P)
rfs_fat 235456 5 - Live 0xbf0b0000 (P)
rfs_glue 74328 1 rfs_fat, Live 0xbf098000 (P)
fsr_stl 248012 6 j4fs, Live 0xbf056000 (P)
fsr 328428 3 dpram,fsr_stl, Live 0xbf000000 (P)
#
Nice find. This may turn into something.
With the module loaded the FM radio app no longer has the error 'Unable to turn on FM radio." However, the FM app has no sound and cannot seek for channels.
dalingrin said:
With the module loaded the FM radio app no longer has the error 'Unable to turn on FM radio." However, the FM app has no sound and cannot seek for channels.
Click to expand...
Click to collapse
Looks like the hardware may be there after all, you were using the headphones right? If so then maybe the antenna isn't wired to the headset port.
Another note: Its is possible that hardware is not there. I found another tool that seems to test FM Radio
# ./si4709_test -h
How to control FM Radio (Si4709) .........
================================================================
Usage: ./si4709_test [controls] [value]
-p power, up|down si4709 power control
-s search, up|down|auto seek channel
-rstate, show si4709 radio state
-rssi, show si4709 rssi status
-r, on|off RDS/RBDS Enable/Disable
-rds, show Show RDS/RBDS data
e.g) si4709_test -p (up/down)
================================================================
# ./si4709_test -p up
# ./si4709_test -rstate
[FM RADIO] Radio State
================================================================
Power State = 0
Seek State = 0
================================================================
I tried to do the same on my wife's I9000 and it reported "Power State = 1"
So we may be out of luck with Captivate.
Not judging anyone but why worry about an FM radio when we have apps that do it all, about 4 of my local stations are on iheartradio and another 20 are from immediate areas around me.
When I used FM on my Nokia it was a battery drain more so than 3G.
I tied to run test with headphones connected. Attempt to power up device results in reboot.
Below is kernel log related to loading driver and connecting headphones:
<7>[ 197.626426] device: 'fmradio': device_add
<7>[ 197.628218] PM: Adding info for No Bus:fmradio
<7>[ 197.635228] bus: 'i2c': add driver Si4709_driver
<7>[ 197.665086] i2c-adapter i2c-8: NAK from device addr 0x10 msg #0
<6>[ 204.231953] [ JACK_DRIVER (jack_type_detect_change,404) ] 3 pole headset or TV-out attatched : adc = 46
I also noticed that after I load driver, I see device file called /dev/fmradio. It disappers when I unload driver. It looks like /dev/fmradio is a control device for radio chip (i found string /dev/fmradio inside si4709_test tool).
oh noes! I want the fm radio function.
for_ap1 said:
I tied to run test with headphones connected. Attempt to power up device results in reboot.
Below is kernel log related to loading driver and connecting headphones:
<7>[ 197.626426] device: 'fmradio': device_add
<7>[ 197.628218] PM: Adding info for No Bus:fmradio
<7>[ 197.635228] bus: 'i2c': add driver Si4709_driver
<7>[ 197.665086] i2c-adapter i2c-8: NAK from device addr 0x10 msg #0
<6>[ 204.231953] [ JACK_DRIVER (jack_type_detect_change,404) ] 3 pole headset or TV-out attatched : adc = 46
I also noticed that after I load driver, I see device file called /dev/fmradio. It disappers when I unload driver. It looks like /dev/fmradio is a control device for radio chip (i found string /dev/fmradio inside si4709_test tool).
Click to expand...
Click to collapse
Any way to compare pics of the Captivate mainboard with a Galaxy S? In the general Captivate forum there are pics of the board. Compare chip numbers and see if there are any unmatching chips.
zerosource said:
oh noes! I want the fm radio function.
Click to expand...
Click to collapse
lol I can see the use for it in a natural disaster, during Katrina the radio was all we really had but all the good channels were on the AM side.
decepticon said:
Any way to compare pics of the Captivate mainboard with a Galaxy S? In the general Captivate forum there are pics of the board. Compare chip numbers and see if there are any unmatching chips.
Click to expand...
Click to collapse
I looked at these photos and also at photos from FCC site, but resolution is too low to read chip markings. FM chip is very small in size, so it is really hard to find it.
It seems that receiver talks via I2S bus and also needs few GPIO lines to work. If I understand correctly, I2S bus has no explicit acknowledgment, so driver might not now that device is not present on the bus. I couldn't find detailed application notes for the chip, just brief description is available. I need to look at driver sources to understand how it works. Hopefully Samsung releases it.
Will Samsung or AT&T tell us if there is an FM chip in there or not?
cesierra said:
Will Samsung or AT&T tell us if there is an FM chip in there or not?
Click to expand...
Click to collapse
Maybe a brave soul, will tear the phone apart and finally give us a straight answer
wont be me though
In the source code for the captivate, there's is a folder called fm_si4709 containing many files pertaining to that radio. Perhaps someone astute w/ kernel stuff could work the magic and enable the HW if it is there...
Well, I looked at sources of the driver (i9000 and I897 are the same). It does not do anything with hardware on driver init. So successfully loaded driver does not mean that hardware is present. I need to find a way to compile driver with debug mode on. Another option will be to build a tool that probes I2C bus (i2cdump and i2cdetect) from shell.
Ok, I was able to build modules using Samsung sources and toolchain from Sourcery. I do not know if I can use it with stock kernel though. Will try to play with it, if I have spare time on weekend.
I just built the kernel using the sourcery toolchain: http://www.codesourcery.com/sgpp/lite/arm/portal/release1033
I enabled the debug flag for the FM radio driver (uncomment #define Si4709_DEBUG in kernel/modules/fm_si4709/Si4709_common.h)
I did an insmod of the compiled module (Si4709_driver.ko). Note the last few lines from dmesg (i2c_read failed & Si4709_dev_init failed) . Doesn't look promising.
<2>[277139.763458] --------Si4709_driver_init called
<7>[277139.766871] device: 'fmradio': device_add
<7>[277139.768313] PM: Adding info for No Bus:fmradio
<2>[277139.774798] --------Si4709_driver_init request_irq success 192
<2>[277139.779520] --------Si4709 i2c driver Si4709_i2c_driver_init called
<7>[277139.785779] bus: 'i2c': add driver Si4709_driver
<4>[277139.799406] ----- si4709_i2c_probe 94
<7>[277139.801782] device: '8-0010': device_add
<7>[277139.801979] bus: 'i2c': add device 8-0010
<7>[277139.802269] PM: Adding info for i2c:8-0010
<7>[277139.802442] driver: '8-0010': driver_bound: bound to device 'Si4709_driver'
<2>[277139.802555] --------Si4709 i2c driver Si4709_probe called
<2>[277139.807314] --------Si4709_dev_init called
<7>[277139.847935] i2c-adapter i2c-8: NAK from device addr 0x10 msg #0
<2>[277139.848866] --------i2c_read failed
<2>[277139.851148] --------Si4709_dev_init call over
<2>[277139.855568] --------Si4709_dev_init failed
<2>[277139.859757] --------Si4709_driver_init successful
Damn that sucks :-(. Wish we knew what the chip looked like so we could just see if its on the phone.
why would they put the drivers in, if there is no chip? This is just a hope.
I was think the same thing. I don't know much about kernel building but I would imagine if I was the guy at Samsung making it I wouldn't add sh!t that was useless
They didn't add a potentially useless driver, they just didn't remove it. Big difference
Thanks to the trinity kernel, morfic and a very nice FiiO E10 USB dac, I have excellent audio quality in my car through my dash mounted nexus 7. It has worked flawlessly for weeks without issue.
Unfortunately as morfic pointed out in his kernel thread, with a USB DAC enabled and his asound.conf changes, we destroy the audio input. No IP phone calls/ video recording... they crash since the audio input configuration is messed up.
We need to merge the asound.conf USB DAC output changes with the stock asound.conf. I've tried all permutations of this that I think make sense and I've had zero luck. Heck, if I go back to the stock asound.conf my USB DAC still works and audio input is still broken.
If there are any alsa gurus out there, please speak up and give us 10 minutes of your time. I've spent hours on this, reading alsa docs and trying asound.conf changes.
Not sure if it matters, but I'm using the stock JB rom with just the upgraded trinity kernel.
for reference: the stock asound.conf is:
#
# ALSA library configuration file
#
pcm.!default {
type plug
slave {
pcm "hw:tegrart5640,0"
channels 2
rate 48000
}
}
ctl.!default {
type hw
card tegrart5640
}
pcm.music {
type plug
slave {
pcm "hw:tegrart5640,0"
channels 2
rate 48000
}
}
pcm.aux {
type hw
card "Tegra"
device 3
}
pcm.voice {
type plug
slave {
pcm "hw:tegrart5640,2"
channels 1
rate 8000
}
}
pcm.aux_plug {
type plug
slave.pcm "aux"
}
pcm.music_and_voice {
type asym
playback.pcm {
type plug
slave.pcm {
type multi
slaves.a.pcm music
slaves.b.pcm voice
slaves.a.channels 2
slaves.b.channels 1
bindings.0.slave a
bindings.0.channel 0
bindings.1.slave a
bindings.1.channel 1
bindings.2.slave b
bindings.2.channel 0
}
route_policy duplicate
}
capture.pcm "voice"
}
Click to expand...
Click to collapse
The updated changes from morfic are:
pcm.usbdac {
type hw;
card Audio;
}
ctl.usbdac {
type hw;
card Audio;
}
pcm.!default usbdac
Click to expand...
Click to collapse
Thanks!
did you have a look at this yet?
http://rootzwiki.com/topic/30129-kernel-nexus-7-cifsnfsusb-otg-gpu-oc-cpu-oc-072912/
Tried that kernel, same results as trinity.
I'm pretty well stuck in the same situation. My main reason for wanting usb audio is for connecting a better mic to the nexus 7.
I tried some variations in the asound.conf with no impact either.
It is supposed to be asound.conf in /system/etc correct? I ask, because I completely emptied the file, and everything still works exactly the same. Normal mic and internal sound, and usb sound out after a reboot.... my asound.conf still currently sits empty.
Maybe this information will help get things started. Here's stuff from my /proc/asound directory with the usb device plugged in.
Code:
sh-3.2# ls -a /proc/asound/
. Tegra card2 hwdep timers
.. card0 cards pcm version
Device card1 devices tegrart5640
Code:
sh-3.2# cat /proc/asound/devices
2: [ 0- 3]: digital audio playback
3: [ 0] : control
4: [ 1- 0]: digital audio playback
5: [ 1- 0]: digital audio capture
6: [ 1] : control
7: [ 2- 2]: digital audio playback
8: [ 2- 2]: digital audio capture
9: [ 2- 1]: digital audio playback
10: [ 2- 1]: digital audio capture
11: [ 2- 0]: digital audio playback
12: [ 2- 0]: digital audio capture
13: [ 2- 0]: hardware dependent
14: [ 2] : control
33: : timer
Code:
sh-3.2# cat cards
0 [Tegra ]: HDA-Intel - HDA NVIDIA Tegra
HDA NVIDIA Tegra at 0x70038000 irq 113
1 [Device ]: USB-Audio - USB PnP Sound Device
C-Media Electronics Inc. USB PnP Sound Device at usb-tegra-ehci.0-1, full speed
2 [tegrart5640 ]: tegra-rt5640 - tegra-rt5640
tegra-rt5640
Code:
sh-3.2# ls Device
id pcm0c pcm0p stream0 usbbus usbid usbmixer
Code:
sh-3.2# cat pcm
00-03: HDMI 0 : HDMI 0 : playback 1
01-00: USB Audio : USB Audio : playback 1 : capture 1
02-00: RT5640 PCM rt5640-aif1-0 : : playback 1 : capture 1
02-01: SPDIF PCM dit-hifi-1 : : playback 1 : capture 1
02-02: BT SCO PCM dit-hifi-2 : : playback 1 : capture 1
More information.
Code:
bash-3.2# lsusb
Bus 002 Device 001: ID 1d6b:0002
Bus 001 Device 001: ID 1d6b:0002
Bus 002 Device 002: ID 0d8c:0139
First device is present in lsusb before I plug the usb device in. Dmesg seems to show the second as the host controller..
Some dmesg output right after plugging in.
Code:
[ 586.426776] usb 2-1: new full speed USB device number 2 using tegra-ehci
[ 586.473084] usb 2-1: New USB device found, idVendor=0d8c, idProduct=0139
[ 586.473493] usb 2-1: New USB device strings: Mfr=1, Product=2, SerialNumber=0
[ 586.473712] usb 2-1: Product: USB PnP Sound Device
[ 586.474095] usb 2-1: Manufacturer: C-Media Electronics Inc.
[ 586.499589] input: C-Media Electronics Inc. USB PnP Sound Device as /devices/platform/tegra-ehci.0/usb2/2-1/2-1:1.3/input/input4
[ 586.500399] keychord: using input dev C-Media Electronics Inc. USB PnP Sound Device for fevent
[ 586.500701] generic-usb 0003:0D8C:0139.0002: input: USB HID v1.00 Device [C-Media Electronics Inc. USB PnP Sound Device] on usb-tegra-ehci.0-1/input3
This is my first time participating in any forum. I've always been an irc geek. So I'm not entirely sure how to promote an interest in this topic, or if we need a new specific thread. Also.. excuse any newb annoyances on my part as I adjust...
I'm pretty well willing to try anything with my nexus 7. No consern for loosing data or starting from scratch... any rom, kernel, tweak.. if anybody has suggestions or want me to try/test something.
I have a feeling this is configuration related, maybe not asound.conf... I messed with audio_policy.conf no luck, but it did have an impact on my input when I tried a few input lines similar to top lines for internal, but in USB section at bottom. I ended up with a continuous buzz on input, and it was still buzzing on input when I pulled the USB device. No change in wave form when tapping mics.. so no good, but I was excited that I had an impact on the input channel at all. Might just be bad lines I used.. was just guessing at config based on rest of file.
I'll try to remember to post what I put in audio_policy.conf later.
Wish I had aplay and arecord to test each device. Tried tinycap with no luck.
Sent from my Nexus 7 using XDA Premium HD app
Glad someone else is excited about getting this working!
I believe you can grab alsa binaries from the play store for free. I played with it a few weeks back and the alsa tools seem to work.
I haven't seen those? I was looking for alsa tools for android tools the other day and could only come up with tinycap, tinymix, and tinyplay with a libtinyalsa.so on another forum... but they aren't utilizing the right devices when I give them device numbers and only throw errors. If you could point me at the tools, that'd be cool.
Do you think we need to start a thread on a different part of the forum? like in a development section? or... ionno...
I honestly don't really know how this works. I am seeing the importance of threads vs just sharing info on irc as I get older... and if/when I do a little bit of work towards something or digging of information.. I do like that it gets passed along to somebody else, is discoverable by others using google, and that it is a contribution to the geek community that I have always been proud to consider myself a part of.
I think here is a good place to start. Someone out there definitely knows exactly how to do this and the more exposure the better. XDA is awesome for that.
As per your previous question, search for alsa mixer in the market. You'll need to install busy box as well.
Can you post your audio_policy.conf changes? I'll try to fiddle with it as well.
OP, the problem that I'm seeing that you now have two default devices, i.e. two definitions of "pcm.!default" (BTW, the exclamation mark overrides built-in defaults) and the latter one takes precedence, I think. Hence, any application using the sound card using the default parameters, as most apps (minus built-in ones) would probably do will grab your DAC which does not have a microphone input. If you had a DAC+ADC you can make this probably work fairly easily. I'm not an ALSA "guru", but someone who has had quite some exposure. I hesitate giving you blind suggestions, but a good place to start is the asoundrc documentation on the alsa-project web pages (can't post links yet)
You may want to look into setting up "multi", "dmix", and/or "dnsoop" devices that combine your DAC with the built-in one. Good luck.
Thanks for the pointers. I've read through much of that documentation but I'll be honest I really got lost. I'll try it again.
I only have one pcm.!default definition in each config file. I do not append the changes from morfic, I replace the entire file.
I'm a little concerned that as we've discovered, we can delete the asound.conf file or put completely invalid junk in it and we are unable to change the behavior... I don't want to spend hours perhaps crossing over a correct asound.conf only to realize something else was preventing my configuration from being successful...
Right now, I'm just trying to get any change from asound.conf to reflect the audio configuration of android. Maybe asound.conf is ignored if there is an invalid config and I've yet to put a valid config?
I'll post the audio_policy.conf tomorrow when I get to work. I'll have a little extra time, I really just based what I added on how the entries above it looked... I'm probably way off.
Also... I do honestly believe the asound.conf has no impact on the audio at all... Mine's still empty... works the same...
papatel said:
I think here is a good place to start. Someone out there definitely knows exactly how to do this and the more exposure the better. XDA is awesome for that.
As per your previous question, search for alsa mixer in the market. You'll need to install busy box as well.
Can you post your audio_policy.conf changes? I'll try to fiddle with it as well.
Click to expand...
Click to collapse
The changes I made, just the bottom usb section of audio_policy.conf. I left everything above it the same.
Code:
usb {
inputs {
usb_accessory {
sampling_rates 44100
channel_masks AUDIO_CHANNEL_IN_MONO
formats AUDIO_FORMAT_PCM_16_BIT
devices AUDIO_DEVICE_IN_USB_ACCESSORY
}
usb_device {
sampling_rates 44100
channel_masks AUDIO_CHANNEL_IN_MONO
formats AUDIO_FORMAT_PCM_16_BIT
devices AUDIO_DEVICE_IN_USB_DEVICE
}
}
outputs {
usb_accessory {
sampling_rates 44100
channel_masks AUDIO_CHANNEL_OUT_STEREO
formats AUDIO_FORMAT_PCM_16_BIT
devices AUDIO_DEVICE_OUT_USB_ACCESSORY
}
usb_device {
sampling_rates 44100
channel_masks AUDIO_CHANNEL_OUT_STEREO
formats AUDIO_FORMAT_PCM_16_BIT
devices AUDIO_DEVICE_OUT_USB_DEVICE
}
}
}
}
Like I said, this did nothing except create a continuous buzz on the input channel, even after checking it with usb device in, and then pulling the device and checking again without the device.... still buzzing... but as I said, I was just pleased that I managed to find something that actually had an impact on the input channel at all. Maybe with a little more fiddling, the right settings, maybe this is the file to focus on... until somebody more informed tells me otherwise.. lol.
---------- Post added at 03:43 PM ---------- Previous post was at 03:35 PM ----------
ah... sorry. There's an extra close bracket, it's cause I pasted the whole bottom of the file, and that last bracket closes off the configuration for the entire file, not just the usb section. Just noticed that and didn't want anybody to get messed up or confused... just make sure each section is opened and closed I guess when you're done editing.
Just in case anybody is interested, I just built the necessary kernel modules (and the snd-aloop device that we may need to get this to work) for the stock kernel (3.1.10-g52027f9) to allow USB audio without having to install a custom kernel. The modules load fine.
I currently don't have access to my USB sound card to test further, but I'll play around with it once I get it back.
(I'm not sure I can attach files yet, so PM me if you are interested.)
chdloc said:
Just in case anybody is interested, I just build the necessary kernel modules (and the snd-aloop device that we may need to get this to work) for the stock kernel (3.1.10-g52027f9) to allow USB audio without having to install a custom kernel. The modules load fine.
I currently don't have access to my USB sound card to test further, but I'll play around with it once I get it back.
(I'm not sure I can attach files yet, so PM me if you are interested.)
Click to expand...
Click to collapse
That sounds good, I'd be willing to give it a shot. I will hit you up when I have a little time. Perhaps you could consider hosting them on dropbox or something similar. I find myself using it a lot these days.
Any update on this? Getting my N7 prepped for car use.
Have it set up so I can do full USB OTG with USB audio through a DAC, and charge simultaneously (I can even run the N7 in this configuration with the battery connector disconnected). I thought this was the hard part.
But with this installation going into my car, having the ability to control functions with my voice will be very critical! So... has anybody successfully enabled hooking a microphone up to the DAC for audio-in to their Nexus 7?
MetalMan2 said:
Any update on this? Getting my N7 prepped for car use.
Have it set up so I can do full USB OTG with USB audio through a DAC, and charge simultaneously (I can even run the N7 in this configuration with the battery connector disconnected). I thought this was the hard part.
But with this installation going into my car, having the ability to control functions with my voice will be very critical! So... has anybody successfully enabled hooking a microphone up to the DAC for audio-in to their Nexus 7?
Click to expand...
Click to collapse
I've been juggling a lot of things lately and can't even get to my "GOOD" pc... but as soon as I get myself organized, I'm jumping right back into trying to solve this.
I really hope some other people are interested enough to be working on it still too.
It's a pretty important issue to me as a musician... I'm thinking the N7 could be the first android device to do some light multi-tracking on... Not looking for a pocket studio, but it sure would be nice to sit down with just my guitar and a good mic hooked to the n7 to work out some ideas.
So, I compiled the modules for usb-audio for my tf101 too, loaded them into kernel and played some noise via aplay. Then I checked lsof and noticed that the sound is handled by the "mediaserver" process, then I went to grep the android sources.
I found that android has hardcoded it's audio devices files, so it's pretty hard to make usb audio work with asoundrc (i.e. impossible), but also I noticed that there is actually support of usb audio (that has hardcoded to the card number 1 though, but this is something and better than nothing) and the system has to be built with the QCOM_USBAUDIO_ENABLED flag to get this work.
Who will test it first?
hm. this code is specific for the qcom hardware... so need dive deeper to the code...
jbinary said:
So, I compiled the modules for usb-audio for my tf101 too, loaded them into kernel and played some noise via aplay. Then I checked lsof and noticed that the sound is handled by the "mediaserver" process, then I went to grep the android sources.
I found that android has hardcoded it's audio devices files, so it's pretty hard to make usb audio work with asoundrc (i.e. impossible), but also I noticed that there is actually support of usb audio (that has hardcoded to the card number 1 though, but this is something and better than nothing) and the system has to be built with the QCOM_USBAUDIO_ENABLED flag to get this work.
Who will test it first?
Click to expand...
Click to collapse
I was doing some filework on my A7+ and thought "Wouldn't it be so much better if I could just plug this damn thing directly into my computer?" So I put it in device mode, rebooted, and used a male-to-male USB cable to connect it to my PC. It didn't work, as expected, but what I did find out was the filesystem or driver system that it uses. Windows XP was trying to find drivers for a "UT-PB1" device. It couldn't, of course, so I went to the internet. I literally searched "UT PB1 USB drivers" and I got this article:
http://thetechjournal.com/electroni...raboo-ut-pb1-e-book-tablet-now-official.xhtml
(Other source: http://techcrunch.com/2011/07/22/raboo-panasonic-japan-to-launch-android-e-book-readertablet-video/ )
Look closely at the bevel, the camera and light sensor positions, and the three soft keys. (I know that the menu key is rotated 90 degrees) This weird tablet has most of the same specs as the A7+.
7-inch 1024×600 color touchscreen
Dual Core CPU (the article says that the actual chip is unknown, but I bet that it's Tegra 2)
1.3Mpix Camera
Wifi (BGN), Bluetooth
USB port (but no HDMI)
It seems that this Panasonic tablet was never released in the US. Maybe StreamTV bought a factory off of this Chinese company, and used the same basic hardware.
Just a thought. Maybe someone can do something with this, I don't know.
I ran into similar results in Win 7, however it did display a generic drive icon for the device.
I think I had USB debugging and allow mock connections on at the time for experimentation.
From what I could tell the two devices were talking to each other however I couldn't access the tablet's file system.
Do you have root setup? I do not.
---------- Post added at 12:34 AM ---------- Previous post was at 12:17 AM ----------
Here is a short output from ADB:
$ adb -d devices
List of devices attached
???????????? no permissions
adb -d bugreport
error: insufficient permissions for device
As far as I can tell from this, the hardware works....
Upon further review, I think I had the USB permissions incorrect in my: /etc/udev/rules.d/51-android.rules
But after rebuilding my desktop environment for the Nth time recently, ADB is able to poll the device.
$ adb devices
List of devices attached
037c708542c0e097 device
The "$ adb -d bugreport" also works but has far too much info to post here.
I have managed to boot to rEFInd on my Toshiba Encore in preparation for booting Ubuntu from a microSD card natively. I have not written up a guide yet from my notes because I don't think there will be much demand for this here on XDA. If enough people want to do this, I will turn this post into a guide. This boots directly from the UEFI partition with no USB stick needed. You DO need a USB stick to do the actual install.
Toshiba Encore booting to rEFInd: http://postimg.org/image/4lx3mnxzr/
Toshiba Encore booting to Grub2: http://postimg.org/image/p3rgtpqkp/
__________________________________________________________________________
Installing Ubuntu 14.10 to a microSD card on the Toshiba Encore:
Note: There are many problems right now with Ubuntu on the Encore. Touchscreen driver is broken as of kernel version 3.16. It's some kind of regression bug in the atmel-maxtouch driver(here is a program to debug our touchscreen: https://github.com/atmel-maxtouch/obp-utils).
You will need to prepare with:
A USB stick to do the install
Ubuntu 14.10 64 bit ISO(Daily build) http://cdimage.ubuntu.com/daily-live/current/
A microUSB OTG cable(if you use the StarTech cable mentioned below then you don't need this)
A high speed microSD card(Partitioned: 200MB FAT32 EFI partition, a large EXT4 partition, and a swap partition). I used this(was $21 on sale) http://www.amazon.com/gp/product/B00IVPU894/ref=oh_aui_detailpage_o04_s00?ie=UTF8&psc=1
A Linux box to compile the latest versions of a few programs used to boot into Grub2 and Ubuntu. I will post the compiled versions but you should compile them yourself for security reasons.
Read this to get an idea what's needed to get rEFInd installed: http://www.rodsbooks.com/refind/installing.html#windows (there are extra steps not listed there)
The install will go from the usb stick booted from rEFIind to the microSD card. The Samsung PRO I am using only does ~20MBps so I'm guessing that is a hardware limit. I did the install without a swap partition and Ubuntu was unusable after opening up 4 or more programs so I added a swap partition to the end of my microSD card.
Getting rEFInd booting:
Turn off Secure Boot in the Encore bios. It should be possible to boot Ubuntu from a USB stick without turning off Secure Boot but I doubt it will ever be possible to do from a microSD card because the Toshiba Encore has no option to boot from a microSD card in the bios boot options.
Get rEFInd here: http://www.rodsbooks.com/refind/getting.html
Run these commands in an elevated permissions Windows command shell:
powercfg /h off
manage-bde -status c: (checks bitlocker status)
powershell
PS C:\> Disable-BitLocker -MountPoint "C:"
manage-bde -status c: (after above command finishes, makes sure your disk is really decrypted.)
Then follow this: http://www.rodsbooks.com/refind/installing.html#windows
All the commands are explained on rodsbooks website except disabling Bitlocker. This is optional but if something goes wrong along the way, you might find Windows 8 encrypted and unbootable.
Note: The command: "bcdedit /set {bootmgr} path \EFI\refind\refind_x64.efi" to get rEFInd set as the default EFI boot program did not take the first time I did it. You might have to run that commend a few times rebooting each time until your tablet boots into rEFInd. Once it does, you will see a graphic to boot Windows 8. I suggest changing rEFInd to text mode so that you can use the VOL UP, VOL DOWN, and Windows button(as enter) to scroll through the boot options. Use "textonly" option in the rEFInd config file to set this: http://www.rodsbooks.com/refind/configfile.html#adjusting
Installing Ubuntu(bleeding edge) 14.10:
Touchscreen does not work yet so I will update this section when I get it working.
_________________________________________________________________________
If you want to have power to your tablet and USB OTG access at the same time, here is what you need(both Amazon prime eligible):
http://www.amazon.com/gp/product/B0047AALS0/ref=oh_aui_detailpage_o03_s00?ie=UTF8&psc=1
http://www.amazon.com/gp/product/B00KQES55M/ref=oh_aui_detailpage_o03_s00?ie=UTF8&psc=1 (Any usb gender bender like this will do...there are cheaper ones)
Attach the USB gender changer to the black end of the StarTech cable and the attach a USB hub to the other end of the gender changer. Plug the red end into a 2.1 amp USB power source and the microUSB end into the Encore. Now the tablet will say "plugged in, not charging". You can now attach any USB device to the hub and Windows or Linux with see them. The tablet will have power but the battery will not charge.
bads3ctor said:
I have managed to boot to rEFInd on my Toshiba Encore in preparation for booting Ubuntu from a microSD card natively. I have not written up a guide yet from my notes because I don't think there will be much demand for this here on XDA. If enough people want to do this, I will turn this post into a guide. This boots directly from the UEFI partition with no USB stick needed.
Toshiba Encore booting to rEFInd: http://postimg.org/image/4lx3mnxzr/
Toshiba Encore booting to Grub2: http://postimg.org/image/p3rgtpqkp/
Click to expand...
Click to collapse
Hi
I wanna boot ubuntu on my encore Can You write the instructions how to, step by step?
Nice, not sure how many others own this tablet, the Venue 8 Pro seems more popular. But I bought an Encore for the microHDMI so I would love to see a guide.
Thank You, I would love to see a guide also, I own a Toshiba encore 32gb
Touch working? Wifi?
Thanks for posting the guide though!
Touchscreen driver works but needs calibration
SebMKd said:
Touch working? Wifi?
Thanks for posting the guide though!
Click to expand...
Click to collapse
Touchscreen bug:
Code:
i2c_hid i2c-ATML1000:00: failed to retrieve report from device.
Supposed to be fixed in kernel version 3.16 but so far in Ubuntu 14.10(kernel 3.16.0-23) TS is still broken. Internal WIFI works with a few tweaks. I am using a USB Atheros WIFI device for now.
I am running Windoze 8 and every few weeks I update Ubuntu 14.10 to see if the touchscreen works. I used to hack away at bleeding kernels to fix bugs like this but I am just going to wait for the Linux kernel devs to fix the touchscreen driver.
Edit: Booted into Ubuntu 14.10 after updating to the newest kernel and:
dmesg | grep i2c-ATML
[ 8.053603] input: ATML1000:00 03EB:8C12 as /devices/platform/80860F41:05/i2c-5/i2c-ATML1000:00/0018:03EB:8C12.0005/input/input6
Touchscreen works but is badly calibrated and the Unity side bar refuses to appear when the mouse is pushed to the left side of the screen. Not a very good trade-off! On the bright side, the former useless USB keyboard windows key now brings up the Unity sidebar.
Code:
[email protected]:~$ xinput list
⎡ Virtual core pointer id=2 [master pointer (3)]
⎜ ↳ Virtual core XTEST pointer id=4 [slave pointer (2)]
⎜ ↳ MOSART Semi. 2.4G Wireless Mouse id=9 [slave pointer (2)]
⎜ ↳ ATML1000:00 03EB:8C12 id=10 [slave pointer (2)]
⎣ Virtual core keyboard id=3 [master keyboard (2)]
↳ Virtual core XTEST keyboard id=5 [slave keyboard (3)]
↳ Video Bus id=6 [slave keyboard (3)]
↳ Logitech USB Keyboard id=7 [slave keyboard (3)]
↳ Logitech USB Keyboard id=8 [slave keyboard (3)]
↳ gpio-keys id=11 [slave keyboard (3)]
↳ gpio-keys id=12 [slave keyboard (3)]
[email protected]:~$ xinput list-props 10
Device 'ATML1000:00 03EB:8C12':
Device Enabled (130): 1
Coordinate Transformation Matrix (132): 1.000000, 0.000000, 0.000000, 0.000000, 1.000000, 0.000000, 0.000000, 0.000000, 1.000000
Device Accel Profile (253): 0
Device Accel Constant Deceleration (254): 1.000000
Device Accel Adaptive Deceleration (255): 1.000000
Device Accel Velocity Scaling (256): 10.000000
Device Product ID (248): 1003, 35858
Device Node (249): "/dev/input/event5"
Evdev Axis Inversion (257): 0, 0
Evdev Axis Calibration (258): 1271, 1282, 3196, 3115
Evdev Axes Swap (259): 1
Axis Labels (260): "Abs MT Position X" (280), "Abs MT Position Y" (281), "None" (0), "None" (0)
Button Labels (267): "Button Unknown" (262), "Button Unknown" (262), "Button Unknown" (262), "Button Wheel Up" (136), "Button Wheel Down" (137)
Evdev Scrolling Distance (261): 0, 0, 0
Evdev Middle Button Emulation (268): 0
Evdev Middle Button Timeout (269): 50
Evdev Third Button Emulation (270): 0
Evdev Third Button Emulation Timeout (271): 1000
Evdev Third Button Emulation Button (272): 3
Evdev Third Button Emulation Threshold (273): 20
Evdev Wheel Emulation (274): 0
Evdev Wheel Emulation Axes (275): 0, 0, 4, 5
Evdev Wheel Emulation Inertia (276): 10
Evdev Wheel Emulation Timeout (277): 200
Evdev Wheel Emulation Button (278): 4
Evdev Drag Lock Buttons (279): 0
Edit2: Here is the problem....the coordinates for calibration are off because the tablet is attached to an external monitor:
Code:
[email protected]:~$ xrandr
Screen 0: minimum 8 x 8, current 3200 x 1080, maximum 32767 x 32767
HDMI1 connected primary 1920x1080+0+0 (normal left inverted right x axis y axis) 983mm x 576mm
1920x1080 60.0*+ 50.0 59.9 30.0 25.0 24.0 30.0 24.0
1920x1080i 60.1 50.0 60.0
1600x900 60.0
1280x1024 60.0
1360x768 60.0
1280x720 60.0 50.0 59.9
1440x576i 50.1
1024x768 60.0
1440x480i 60.1 60.1
800x600 60.3
720x576 50.0
720x480 60.0 59.9
640x480 60.0 59.9
DP1 disconnected (normal left inverted right x axis y axis)
DSI1 connected 1280x800+1920+0 right (normal left inverted right x axis y axis) 0mm x 0mm
800x1280 60.0*+
VIRTUAL1 disconnected (normal left inverted right x axis y axis)
very interesting. It's what I need :good:
Any more news on this?
I managed to install Refind. Also, I managed to install ubuntu on sdcard. But in Refind not appears Ubuntu. How I can edit Refind to search the OS on sdcard?
Thanks!!
PainToad said:
Any more news on this?
Click to expand...
Click to collapse
Sorry for not getting back earlier on this but I wanted to exhaust all options before giving up on running Ubuntu on this tablet. The deal killer turns out to be the MicroSD card bandwidth. From what I can tell from the tests I have done while running Ubuntu from the MicroSD card, the BIOS is set to run the SD bus in DDR25 mode with no option to set it to DDR50 mode. DDR25 mode maxes out at about 18 MB/Sec and this is causing Ubuntu to lockup when disk I/O gets saturated.
With no option in the BIOS to set the SD bus to DDR50 mode, it is effectively a hardware limitation preventing Ubuntu from running properly.
bads3ctor said:
If you want to have power to your tablet and USB OTG access at the same time, here is what you need(both Amazon prime eligible):
http://www.amazon.com/gp/product/B0047AALS0/ref=oh_aui_detailpage_o03_s00?ie=UTF8&psc=1
http://www.amazon.com/gp/product/B00KQES55M/ref=oh_aui_detailpage_o03_s00?ie=UTF8&psc=1 (Any usb gender bender like this will do...there are cheaper ones)
Attach the USB gender changer to the black end of the StarTech cable and the attach a USB hub to the other end of the gender changer. Plug the red end into a 2.1 amp USB power source and the microUSB end into the Encore. Now the tablet will say "plugged in, not charging". You can now attach any USB device to the hub and Windows or Linux with see them. The tablet will have power but the battery will not charge.
Click to expand...
Click to collapse
I have a question. This method won't be charging my tablet, so my battery won't be destroyed?
BTW another question. Why card reader in this tablet is that slow? I have sandisk extreme micro sd (64MB/s write - ~ 60MB/s in my PC card reader) but reader in toshiba can only read/write ~26MB/s. I thought that this card will be great for my tablet, but now I fell like I wasted my money. I could better buy standard class 10 micro sd card...
rascal0pl said:
I have a question. This method won't be charging my tablet, so my battery won't be destroyed?
Click to expand...
Click to collapse
The tablet will receive a charge but the battery will not be charged. If your battery is at 100% then it will stay at 100%.
rascal0pl said:
BTW another question. Why card reader in this tablet is that slow? I have sandisk extreme micro sd (64MB/s write - ~ 60MB/s in my PC card reader) but reader in toshiba can only read/write ~26MB/s. I thought that this card will be great for my tablet, but now I fell like I wasted my money. I could better buy standard class 10 micro sd card...
Click to expand...
Click to collapse
It is an internal firmware limitation. There is supposed to be a BIOS setting to set the SD slot to a higher bandwidth level but this tablet does not have that option in the BIOS.
Card Reader of Bay Trail tablets is broken
The Card Reader of Bay Trail tablets is slow because is broken, no linux boot and is a read-only device with stock linuxes. You can patch the kernel and install some boot stuff in new partitions of Windows disk, but you can't boot from microSD. I also installed Android x86 and run fast, Android 4 and 6 stable versions, but I couldn't activate sound nor battery status. At least for my research, until today, if you have the 32 GB disk is very difficult you can play with dual boot, Windows is slow and wastes the hard disk, tablet mode of windows has no attractive (while Android is near perfection) and desktop mode is not friendly for touch-screen use, but for now it is stable and all the other tablet's devices works (touchscreen, GPS, Bluetooth, cameras, sensors, miracast), the battery management in this Atom environment has issues with linux and under Windows this x86 system doesn't seem so far away from ARM efficiency. I installed many GPL portable Windows applications in a minimal disk space, so for now linux and Android are worse for this tablet that seemed created for a Microsoft's skinhead.
UEFI boot
I forgot I have a 10 inches version of this tablet, 32 GB, Windows 10 32 bits upgrade (first upgrade) and BIOS 2.0
UEFI multi-boot is not so difficult with: rEFInd, easyUEFI Windows application (because Windows changes boot order), linux's Unetbootin for USB boot, bootia32.efi archive for some linux's installations (Ubuntu-gnome 16.04.1 64 bits requires at least 7-10 GB and swap, may be 32 bits is a bit better because it's less RAM hunger), ... before tests, you have to disable Windows's Bitlocker encryption (AES-128 Windows disk software encryption for this model of the Atom processor, I think without hardware assistance, that it's seamless for the user because UEFI includes shared key, if disk key is deleted Microsoft online user account stores a large password), secure boot, fast boot, windows session identification (Microsoft online authentication sometimes locks session start, probably cause some autocheck of a efi boot manipulation... may be was "default_selection" option of rEFInd for booting last selected OS), backup efi boot partition (copy the files after mount partition)... then I gparted the Windows disk form USB (with no data loss).
And if someone can the microSD boot, Raspberry Pi community recommends Samsung EVO+ 32 GB, for about 10 $/€ (be warned about scam versions), with the best by far random IO, far less from 20 MB/s, but may be important for linux, although I don't know if max random IO of the card reader had been reduced too (most probably yes... and a lot).
Another chance is wubi EFI, a Ubuntu Windows installer that now supports UEFI 32 bits (the non approved by free-software community secure boot only with UEFI 64 bits), this install linux in a big file (resizable) of a NTFS Windows disk (again microSD doesn't boot, it requires a rEFInd's Card Reader firmware and is not available) and linux runs in loop mode inside this file. I installed Ubuntu-gnome 64 bits and 32 bits (in this case the installer requires a command-line option). UEFI32 too requires EasyUEFI and secure boot and bitlocker disabled, I utilized a USB wifi stick (TL-WN722N) until internal wifi was configured... I found lots of issues, including slow internal-wifi, soundcard, battery status, processor stepping and hangs, and other users's solutions for different tablets have not always worked... then no usability research here. Anyway, wubi is not the best way to install linux, although it works and it's easy, and Windows complain less.
A last possibility for linux could be VirtualBox 32 bits, this supports microSD use and Atom processor includes hardware assistance supporting even linux 64 bits. Virtualize linux on this tablet is very slow and touchscreen is a pain.
After some time I returned the tablet to its original state with no data loss*, USB boot still there or the install in a fast (if well chosen) microUSB 3.0 stick or microSD-adapter... but no charger without the use of a less portable, and manipulation disconnections, cable adapter. I utilized a cheap "octopussy" cable from China (3 USB 2.0+ microUSB for charge... 3-4 $/€ or less), but this tablet is incompatible for simultaneous operation (USB OTG+charge), it don't charge while USB use and only I can do slow charge with a tricky: connect with charge mode, change to OTG mode and connect USB devices (warning: some Amazon's users report issues or damages in other tablets).
*bitlocker key changed with a new online password stored for recovery, and Toshiba recovery (Bios option) was broken, may be caused by gparted or some UEFI boot application or some of the installed OS, who knows... normal fixes solutions don't work, although windows own recovery partition worked. Toshiba utilizes a particular EFI directory and Windows doesn't repair it, I copied Toshiba BCD backed up file or, in case it doesn't work, BCD file from Windows recovery boot directory (may be a normal fixboot is required first).
Bug issues under Windows 10 update
After some time of use this tablet suffers issues, cause of its Bay Trail buggy platform, under Windows 8.1 and 10 (worse): Wifi disconnections and poor speed, laggy applications and touchscreen sometimes is unresponsive, miracast disconnections and choppy sound, although the whole system doesn´t hang. I did try linux last updates from http://linuxiumcomau.blogspot.com.es/ and now most work better (July 2017), probably in a short time I will uninstall Windows 10.
I have a "Tata Sky HD" Set-top Box and I was about to throw this in garbage but before I want to know what is happening under the hood.
I search on internet and I found nothing except this. I'm noob so sorry for if say something silly.
I found this specifications.
Product : TATA SKY HD
Original Maker : Technicolor
Product Model Number : DSI729TAT
Chipset : STiH237 BHKB B3L
Type : ST40 -32 BIT
Architecture : RISC
RAM : 2GB [ SK Hynix H5TQ2G63FFR H9C
Storage : 1GB [ Spanison ML01G100
Power : 12v DC
Software: busybox 1.18.2 , mtdwrap, uclibc, Linux Kernel 2.6.32.59_stm24_0211, ST drivers: embx.ko, embxmailbox.ko, ics.ko, ics_user.ko, lxload.ko, mme.ko, mme_user.ko, LZO Decompression Library 2.03, Decompression Utility
PORTS : 1 HDMI 1.2/1.3/1.4, 1 USB 2.0, SAT-IN & 2 Audio 1 Video Out , 1 Optical S/PDIF (for Dolby Digital Plus Audio ), 1 Digi Card.
I Found 1 UART PORT Which would be used while extraction of Firmware.
AFTER SOME REASEARCH I FOUND THAT IT HAS SIMILAR TO ARM-CORTEX-A9 AND MALI-400 GPU. (MAYBE I'M WORNG)
IDEA : It has a a Good processor and ram which can run as raspberry-pi os.
so we can repurpose it as a Media Center, Gaming Console, NAS, Smart Home, Small Server or a Mini Computer.
storage is low so we have to add some storage. I'm not sure how this is possible. except swapping the NAND flash Chip.
GOAL 1 : Extract Firmware and Extract Paid Decryption key which is use to verify the sat-in signals. ( a stb which don't required subscription to watch any tv channel )
i think they modified the software which capture the unencrypted signal and if we have a signal receiver then we are good to go. but for big companies wants to earn money so they added these barriers which needs decryption. and if the satellite is sending encrypted signal then we need to find the key. ( i know it's hard that's why we are here. I'll love to hear you thoughts on these)
GOAL 2: Change the Firmware and install Linux.
Goal 3: Find a way to use it as media server with increased storage and add a wireless module for WIFI access.
I'm not sure it is possible or not. but i think its possible. just think about it a small hardware can collect signal from satellite and decrypt the signals in HD with Dolby HD audio. we just need to find a way to access this.
I SHARED MY IDEA AND I DON'T KNOW MUCH ABOUT THESE.
PROBLLY I'M GOING TO ACCESS THIS WITH UART INTERFACE AND TRYING TO ACCESS THE BOOTLOADER.
OR MAYBE DESIGN A CUSTOM KERNAL.
I'M SEARCHING FOR COMPATIBLE FIRMWARE WHICH I CAN MODIFY AS I NEED.
EXTRA : I FOUND A SIMILAR STB WHICH USED IN RUSSIA "NTV PLUS SET TOP BOX" HAS SIMILAR PROPERTIES LIKE TATASKY HD BUT WITH EXTRA I/O PORTS.
THANK YOU. IF YOU HAVE ANY ADDIONAL IDEA THEN I'LL LOVE TO HEAR THAT.
Links Used For Gathering Information
Chipset : https://www.st.com/en/digital-set-top-box-ics/stih237.html
RAM : https://www.electronicsdatasheets.com/manufacturers/sk-hynix/parts/h5tq2g63ffrh9c
OS information : https://www.technicolor.com/node/1899
Storage : https://www.qdatasheet.com/search.jsp?sWord=ML01G100&page=2&op=i
RISC BASED TOOLS AND APPS : https://www.riscosopen.org/content/downloads/common
This is probably the UART. You will most likely get a shell and U-Boot logs provided that it's not fused-off (ST microcontrollers can have debug interfaces fused off during flashing at the manufacturer)
How to find the pinout:
GND will have continuity with metallic parts of the board (heatsinks, HDMI ports, etc)
VCC will measure 1.8-5V DC depending on logic level
RX will not measure very much voltage
TX will go crazy during boot on an oscilliscope.
Try baudrate 115200 8n1
$cronos_ said:
This is probably the UART. You will most likely get a shell and U-Boot logs provided that it's not fused-off (ST microcontrollers can have debug interfaces fused off during flashing at the manufacturer)
View attachment 5877483
How to find the pinout:
GND will have continuity with metallic parts of the board (heatsinks, HDMI ports, etc)
VCC will measure 1.8-5V DC depending on logic level
RX will not measure very much voltage
TX will go crazy during boot on an oscilliscope.
Try baudrate 115200 8n1
Click to expand...
Click to collapse
Well i don't have oscilloscope yet, soon I will try your guide, thanks for guidance. I will try to update upcoming experiments.
dyal96 said:
IDEA : It has a a Good processor and ram which can run as raspberry-pi os.
so we can repurpose it as a Media Center, Gaming Console, NAS, Smart Home, Small Server or a Mini Computer.
storage is low so we have to add some storage. I'm not sure how this is possible. except swapping the NAND flash Chip.
GOAL 1 : Extract Firmware and Extract Paid Decryption key which is use to verify the sat-in signals. ( a stb which don't required subscription to watch any tv channel )
i think they modified the software which capture the unencrypted signal and if we have a signal receiver then we are good to go. but for big companies wants to earn money so they added these barriers which needs decryption. and if the satellite is sending encrypted signal then we need to find the key. ( i know it's hard that's why we are here. I'll love to hear you thoughts on these)
Click to expand...
Click to collapse
I think we can utilize the usb port on the back to add the external storage, as the usb port is used for storing the TV recording (as far as I can recall), and for the uart part, we can also use it for accessing root shell in the initial step, to figure out the operation method and framework.
I don't have any idea about the encryption keys, it would be cool if there's a way for that.
I have the same STB, would love to repurpose the old box, what's your progress on this so far ?
If you have a multimeter, you can check if the pins are for UART, RX voltage would be very low, TX voltage would be fluctuating upon boot, check continuity for GND with any grounded part like the HDMI port shield or the AV port silver port, VCC would be 3.3 or 5 volts