Database Info

[INFO][DEV?] Dock Charge Level

Noticed a number of logcat entries which look to detail charging level of both the dock and keyboard.
For example:
D/BatteryService( 149): level:88 scale:100 status:4 health:2 present:true dock_status:4 dock_level:3 voltage: 8 temperature: 267 technology: Li-ion AC powered:false USB powered:false icon:17302721 invalid charger:0
Click to expand...
Click to collapse
In the above extract, the dock & charger are running off internal power. Dock level is 3% and tablet level is 88%. Dock LED was flashing red which indicates low charge.
D/BatteryService( 149): level:78 scale:100 status:2 health:2 present:true dock_status:2 dock_level:4 voltage: 8 temperature: 277 technology: Li-ion AC powered:true USB powered:false icon:17302735 invalid charger:0
Click to expand...
Click to collapse
At this point, charger was added (note the extract was taken a good 20+ minutes after the first, hence the lower tablet charge level).
D/BatteryService( 149): level:99 scale:100 status:2 health:2 present:true dock_status:2 dock_level:11 voltage: 8 temperature: 263 technology: Li-ion AC powered:true USB powered:false icon:17302735 invalid charger:0
Click to expand...
Click to collapse
Above extract (taken even later) shows tablet at 99%, with Dock now charging and at 11%.
Not sure if the above information is available anywhere via the Android API - http://hi-android.info/src/com/android/server/BatteryService.java.html seems to suggest it might be (or a customised version is being used to cater for the dock perhaps?? - the output from the above seem to have additional entries for "dock")
This may help someone (or just be a repost ) work out a way to hack up a quick UI to show dock battery level (and charging status)
If you want to view the info, fire up a terminal emulator on your tablet and run the following command:
Code:
logcat BatteryService:D *:S
This should give you JUST entries for the Battery status.

It seems that you need to be rooted to perform a logcat (as a su) on BatteryService.
Do you confirm ?

flocoon said:
It seems that you need to be rooted to perform a logcat (as a su) on BatteryService.
Do you confirm ?
Click to expand...
Click to collapse
Don't have a non-rooted device to test on. However, I believe you're probably correct and root is required as i need to "login" with SU on my phone before logcat works.
<edit>
Theory tested via a basic widget -
http://forum.xda-developers.com/showthread.php?t=1112391
</edit>

[Q] Official ICS - support for USB OTG?

Hi there,
I know that some of you were able to connect peripheral devices to their phones.
Has anyone of you tested this with the ICS stock kernel? When looking at /proc/config.gz it seems to me that all things neccessary for USB OTG are contained in the kernel.
Did you always use a special "OTG cable" or were you able to force the controller into host mode?
I'm playing with several USB hubs at the moment but I'm stuck.
Have a nice day,
Markus

You need Y-cable for getting the VBUS to both phone and hub.

Well, I removed a diode from my hub and put a label on it saying "do not connect to computer" .
Now, as soon as I connect my phone to the hub's upstream port, it starts charging. But still dmesg doesn't say anything about a switch to host mode .

Check ID pin value on the cable you're using. You're most likely using the standard A-to-microB cable, meant for device. You need A-to-microA, or just hack your cable and tie ID pin low in the microB plug, turning it into microA.

Now I made sure I'm using the right cable, but still it doesn't work.
This has to do with the stock kernel. If you look at "arch/arm/mach-msm/board-saga.c", line 3168, it says "USB_PERIPHERAL". If you change this to "USB_OTG" and compile it using the default config file, the kernel still won't recognize a OTG cable. But you can mount debugfs: "mount -t debugfs none /sys/kernel/debug" and look at "/sys/kernel/debug/msm_otg/mode". This file contains "none", if no cable is plugged in, or "peripheral" else. If you try to overwrite it with "echo host > mode", something happens according to dmesg:
[USB] msm_otg_mode_write: host
[USBH] msm_otg_sm_work: state:b_peripheral bit:0x00000002
[USB] OTG_STATE_B_PERIPHERAL state
[USBH] gadget off
[USB] OTG runtime idle
[USB] OTG runtime suspend
[USBH] msm_otg_suspend
[USBH] USB in low power mode
[USBH] msm_otg_sm_work: state:b_idle bit:0x00000002
[USB] OTG runtime resume
[USBH] msm_otg_resume
[USBH] USB exited from low power mode
[USB] OTG_STATE_B_IDLE state
[USBH] b_sess_vld
[USBH] gadget on
[USB] CI13XXX_CONTROLLER_RESET_EVENT received
[USBH] send connect type 0
[TP]cable change to 0
[TP]ts->status change to 0
[USB] suspend
[CABLE] usb_status_notifier_func: cable_type = 0
[CABLE] send_cable_connect_notify: cable_type = 0
Click to expand...
Click to collapse
It seems like something is immediately switching back to "peripheral", probably it's got to do with the "carkit driver":
[CABLE] phy_id_detect: usb ID pin = 0
config_saga_usb_id_gpios 145 output high
[USBH] carkit_phy_reset
[USBH] phy_reset: success
[USBH] ulpi: write 0x06 to 0x36
[USBH] ulpi: write 0x0c to 0x31
[USBH] ulpi: write 0x31 to 0x32
[USBH] ulpi: write 0x01 to 0x0d
[USBH] ulpi: write 0x01 to 0x10
[USBH] =============htc_get_accessory_state otgsc = 0x202e20
config_saga_usb_id_gpios 145 input none pull
[CABLE] phy_id_detect: type = 1
[CABLE] cradle inserted
[CABLE] cable_detect_handler ID pin 0, type 1
Click to expand...
Click to collapse
The file "drivers/usb/otg/msm_otg.c" has been adjusted by HTC and I'm afraid you'd have to revert these changes to make host mode work. Unfortunately, I don't think I have enough knowledge about this to do it myself. But probably this is interesting to some of you who are working on custom ICS kernels .

Hello,
Did you manage to make it work with ICS?
Thank you.

Well, no. And currently I've got neither the time nor the knowledge to I'm afraid. But it was possible in Android 2.3.5 by using a kernel hack. As far as I know amidabuddha is playing around with this, too, but wasn't successful yet as we were not able to get the source code of the modified kernel.

Finally, I did it! :victory: ... At least partially.
Code:
<6>[ 1117.578247] [USB] msm_otg_mode_write: host
<6>[ 1117.578735] [USBH] msm_otg_sm_work: state:b_peripheral bit:0x00000002
<6>[ 1117.579437] [USB] OTG_STATE_B_PERIPHERAL state
<6>[ 1117.579833] [USBH] send connect type 0
<6>[ 1117.580169] [TP]cable change to 0
<6>[ 1117.582244] [TP]ts->status change to 0
<7>[ 1117.582702] [USBH] gadget off
<6>[ 1117.583679] [USBH] msm_otg_sm_work: state:b_idle bit:0x00000002
<6>[ 1117.584350] [USB] OTG_STATE_B_IDLE state
<6>[ 1117.584716] [USBH] !id || id_a
<7>[ 1117.585083] [USBH] host on
<6>[ 1117.585754] msm_hsusb_host msm_hsusb_host.0: Qualcomm On-Chip EHCI Host Controller
<6>[ 1117.586242] msm_hsusb_host msm_hsusb_host.0: new USB bus registered, assigned bus number 1
<6>[ 1117.591003] [CABLE] usb_status_notifier_func: cable_type = 0
<7>[ 1117.591400] [CABLE] send_cable_connect_notify: cable_type = 0
<6>[ 1117.592071] [BATT] peripheral_cable_update(0) from_owe=0
<6>[ 1117.592468] [BATT] Update SMEM: cable type 0 at 1111762850068 (2013-01-13 14:46:15.556571647 UTC)
<6>[ 1117.606140] [BATT] ID=2, level=86, vol=4063, temp=256, batt_current=84, chg_src=0, chg_en=1, full_bat=1600000, over_vchg=0, smem_vbus=1 at 1111776216768 (2013-01-13 14:46:15.570213004 UTC)
<6>[ 1117.611083] msm_hsusb_host msm_hsusb_host.0: irq 60, io mem 0xa3600000
<6>[ 1117.630737] msm_hsusb_host msm_hsusb_host.0: USB 2.0 started, EHCI 1.00
<6>[ 1117.632110] hub 1-0:1.0: USB hub found
<6>[ 1117.632324] hub 1-0:1.0: 1 port detected
<6>[ 1117.950714] usb 1-1: new high speed USB device number 2 using msm_hsusb_host
<6>[ 1118.120849] hub 1-1:1.0: USB hub found
<6>[ 1118.121124] hub 1-1:1.0: 4 ports detected
<6>[ 1136.541107] usb 1-1.2: new low speed USB device number 3 using msm_hsusb_host
<6>[ 1136.701995] input: BTC USB Multimedia Keyboard as /devices/platform/msm_hsusb_host.0/usb1/1-1/1-1.2/1-1.2:1.0/input/input7
<6>[ 1136.705749] using input dev BTC USB Multimedia Keyboard for key reset
<6>[ 1136.706481] keychord: using input dev BTC USB Multimedia Keyboard for fevent
<6>[ 1136.707763] generic-usb 0003:046E:52CF.0001: input: USB HID v1.10 Keyboard [BTC USB Multimedia Keyboard] on usb-msm_hsusb_host.0-1.2/input0
<6>[ 1136.817230] input: BTC USB Multimedia Keyboard as /devices/platform/msm_hsusb_host.0/usb1/1-1/1-1.2/1-1.2:1.1/input/input8
<6>[ 1136.825622] using input dev BTC USB Multimedia Keyboard for key reset
<6>[ 1136.825836] keychord: using input dev BTC USB Multimedia Keyboard for fevent
<6>[ 1136.835693] generic-usb 0003:046E:52CF.0002: input: USB HID v1.10 Device [BTC USB Multimedia Keyboard] on usb-msm_hsusb_host.0-1.2/input1
<6>[ 1161.881195] usb 1-1.3: new low speed USB device number 4 using msm_hsusb_host
<6>[ 1162.014099] input: Logitech USB-PS/2 Optical Mouse as /devices/platform/msm_hsusb_host.0/usb1/1-1/1-1.3/1-1.3:1.0/input/input9
<6>[ 1162.020507] generic-usb 0003:046D:C01E.0003: input: USB HID v1.10 Mouse [Logitech USB-PS/2 Optical Mouse] on usb-msm_hsusb_host.0-1.3/input0
At the moment, you have to switch between peripheral mode and host mode manually, as I explained in my post above. When switching to host mode, the phone stops charging, so this isn't really OTG now.
Patching is much easier than we thought, as all necessary code already exists in the stock source code. HTC just seems to have "forgotten" one line...
I'll append a patch for HTC's ICS stock source code and the .config file I used.

Using the patch and the kernel config I appended OTG mode finally works as it should. The phone switches between peripheral mode and host mode automatically depending on the cable's ID pin and starts charging even when in host mode. It is possible now to connect usb hid devices as well as usb sticks, card readers and so on, I even tried remote triggering a Canon DSLR using "Remote Release".
The appended kernel config file is based on saga_defconfig, the only changes I made are those necessary for OTG. It should still be possible to connect the phone to a car kit or a desktop dock, although I can't test that.

Wow. I worked on this thing for three days and got it working a couple hours ago. And now I find out that you made the the same thing. I patched exactly the same lines. What a coincidence

So your phone starts charging in host mode, too? A few seconds after you plugged in the otg cable?
There's still something wrong I'm afraid:
Code:
<6>[ 1061.087646] [USBH] !id
<6>[ 1061.087738] [USBH] Avail curr from USB = 0
<6>[ 1061.284545] [BATT][tps65200] Set VDPM back to original V: regh 0x03=83
<7>[ 1063.090393] [K] __msm_rpc_read: prog = 0x30100000, procedure = 1, type = 0, xid = 18
<6>[ 1063.090698] [BATT] M2A_RPC: set_charging: 0 at 1057797432594 (2013-01-18 22:12:40.474011248 UTC)
<6>[ 1063.091400] [BATT][tps65200] Switch charger OFF
Although the LED lights up the charger is switched off. You can see that either in the kernel log or in Settings -> Power. There it says "discharging" for me.
There is not even ACA support for the "CI_45NM_INTEGRATED_PHY" as I expected in the first place, so I think we must find a way to force it into ac charging mode when the id pin is grounded. "drivers/usb/otg/msm_otg.c" should be the right place to start...

Hi guys.
I have pulled the request by astro757 and combined it with some more entries from the defconfig by engelmarkus that *could* eventually make the storage host operational. See here. There are some differences in your patches.
Has anyone of you tried that? I do not have a cable to test it and before release I would like to be sure that it is working. Credits will be given to you both of course.
Btw engelmarkus, I see you defconfig is from CM. Am I right?

amidabuddha said:
Has anyone of you tried that?
Click to expand...
Click to collapse
Well, I downloaded your repository and compiled it. HID devices do not work:
...
<3>[ 479.265350] generic-usb 0003:046E:52CF.0002: can't reset device, msm_hsusb_host.0-1.3/input1, status -71
<3>[ 479.272186] generic-usb 0003:046E:52CF.0001: can't reset device, msm_hsusb_host.0-1.3/input0, status -71
...
Click to expand...
Click to collapse
This sounds almost like https://bugzilla.novell.com/show_bug.cgi?id=734754 .
EDIT:
HID devices do work, but you have to plug them in before you power the hub. HID devices plugged in afterwards are not recognized at all. I think there was some change in the patches between 3.0.16 and 3.0.59 that is responsible for this weird behaviour.
amidabuddha said:
Btw engelmarkus, I see you defconfig is from CM. Am I right?
Click to expand...
Click to collapse
I got my saga_defconfig from "saga-ics-crc-3.0.16-62a67df" and made a few changes. I never looked into the CM sources.

Hey, I was wondering what you changed in msm_otg.c. I'm trying to do similar with a Nexus 4 right now, and have modified msm_otg.c for crude support with /sys/kernel/debug/msm_otg/mode (thanks CaptainMuon), but on unplug of the otg cable, mode gets locked to "none" until the device is rebooted. There's also no auto-switching between host and peripheral when an otg cable/device is inserted (externally powered y-cable).
Thanks for any help

Hi,
ziddey said:
Hey, I was wondering what you changed in msm_otg.c.
Click to expand...
Click to collapse
We didn't change this file at all.
Here you can see all the changes that were necessary: https://github.com/amidabuddha/saga-ics-crc-3.0.16/commit/8f04057be6eda67b40eb1bf3354294138f9c2d24
Markus

OTG cable + Card reader = not working.

Hey Im running US retail on my XT 1032, with root and unlocked bootloader. I have the last OTA update with OTG fix.
Since I updated I've been trying to connect my Card reader to my moto g with the OTG cable, but it doesn't work, it doesn't recognize it. I tried with two differents readers and two micro sd, both in NTFS, FAT and FAT32.
I tried a pen drive and it works fine. Its weird because meenova its a card reader too and it works perfectly (I havent got one, and its really hard to buy it in my country because of the costumes policy).
I have a Sandisk 16gb Ultra, so, I really want to fix this.

Any idea?

tatitar said:
Any idea?
Click to expand...
Click to collapse
If you have root, try run the following command from a shell after pluging in the usb device
Code:
echo -n 1 | tee /sys/bus/usb/devices/1-1/bConfigurationValue
I used to have a problem with my HP TP, some USB devices asked to much juice and the default settings reject the device.
Have a look at the following post for more info.
http://rootzwiki.com/topic/33033-otg-usb-insufficient-available-bus-toolthingy/
A "dmesg" output will help you to identify why the device is not been detected.

dubidu said:
If you have root, try run the following command from a shell after pluging in the usb device
Code:
echo -n 1 | tee /sys/bus/usb/devices/1-1/bConfigurationValue
I used to have a problem with my HP TP, some USB devices asked to much juice and the default settings reject the device.
Have a look at the following post for more info.
http://rootzwiki.com/topic/33033-otg-usb-insufficient-available-bus-toolthingy/
A "dmesg" output will help you to identify why the device is not been detected.
Click to expand...
Click to collapse
I tried that and got "tee not found" Maybe I'll sell my microsd and buy a pendrive with that money, I tried almost everything

tatitar said:
I tried that and got "tee not found" Maybe I'll sell my microsd and buy a pendrive with that money, I tried almost everything
Click to expand...
Click to collapse
you don't really need tee, an echo will also change the parameter.
post a dmesg output when you plug the device, that may tell us why is failing to mount

dubidu said:
you don't really need tee, an echo will also change the parameter.
post a dmesg output when you plug the device, that may tell us why is failing to mount
Click to expand...
Click to collapse
http://www.sendspace.com/file/fhj00w There's my dmesg

Can't see anything related to been rejected due power, so the bConfig trick will not work in this case.
Code:
<6>[64952.092782,0] msm_otg f9a55000.usb: USB exited from low power mode
<6>[64952.151006,0] healthd: battery l=96 v=4229 t=27.4 h=2 st=3 c=136 chg=
<6>[64952.182920,0] healthd: battery l=96 v=4224 t=27.4 h=2 st=3 c=188 chg=
<6>[64952.206377,0] msm_hsusb_host msm_hsusb_host: Qualcomm On-Chip EHCI Host Controller
<6>[64952.206931,0] msm_hsusb_host msm_hsusb_host: new USB bus registered, assigned bus number 1
<6>[64952.238571,0] msm_hsusb_host msm_hsusb_host: irq 166, io mem 0xf9a55000
<6>[64952.258317,0] msm_hsusb_host msm_hsusb_host: USB 2.0 started, EHCI 1.00
<6>[64952.258647,0] usb usb1: New USB device found, idVendor=1d6b, idProduct=0002
<6>[64952.258750,0] usb usb1: New USB device strings: Mfr=3, Product=2, SerialNumber=1
<6>[64952.258924,0] usb usb1: Product: Qualcomm On-Chip EHCI Host Controller
<6>[64952.259099,0] usb usb1: Manufacturer: Linux 3.4.0-gc6fc9e1-00001-gf3bf628 ehci_hcd
<6>[64952.259197,0] usb usb1: SerialNumber: msm_hsusb_host
<6>[64952.261601,0] hub 1-0:1.0: USB hub found
<6>[64952.261731,0] hub 1-0:1.0: 1 port detected
<3>[64952.518160,0] hub 1-0:1.0: unable to enumerate USB device on port 1
<6>[64955.061905,0] msm_otg f9a55000.usb: USB in low power mode
I guess that reader is not supported by the current kernel. It looks like it is detected as a USB Hub, not as a simple USB storage device.
Code:
<6>[64952.261601,0] hub 1-0:1.0: USB hub found
Get a cheap one (e.g: like this http://www.ebay.co.uk/itm/OTG-USB-M...aphy_MemoryCardReaders_RL&hash=item5d46cbed5b
I have a couple and they do the job fine.

http://forum.xda-developers.com/moto-x/general/otg-supported-4-4-2-unrooted-via-t2709553
My fix for the X.

Question related to OTG, for Moto G running Kitkat is it safe to directly unplug the OTG cable with the USB?

hypertx said:
Question related to OTG, for Moto G running Kitkat is it safe to directly unplug the OTG cable with the USB?
Click to expand...
Click to collapse
I do it all the time, BUT IF YOU FORGET YOU HAVE A WRITE OPERATION IN PROGRESS IT WILL TRASH the card.
Btw do all the G's have /storage/usbdisk on 442?http://forum.xda-developers.com/attachment.php?attachmentid=2675348&stc=1&d=1396975449

dubidu said:
Can't see anything related to been rejected due power, so the bConfig trick will not work in this case.
Code:
<6>[64952.092782,0] msm_otg f9a55000.usb: USB exited from low power mode
<6>[64952.151006,0] healthd: battery l=96 v=4229 t=27.4 h=2 st=3 c=136 chg=
<6>[64952.182920,0] healthd: battery l=96 v=4224 t=27.4 h=2 st=3 c=188 chg=
<6>[64952.206377,0] msm_hsusb_host msm_hsusb_host: Qualcomm On-Chip EHCI Host Controller
<6>[64952.206931,0] msm_hsusb_host msm_hsusb_host: new USB bus registered, assigned bus number 1
<6>[64952.238571,0] msm_hsusb_host msm_hsusb_host: irq 166, io mem 0xf9a55000
<6>[64952.258317,0] msm_hsusb_host msm_hsusb_host: USB 2.0 started, EHCI 1.00
<6>[64952.258647,0] usb usb1: New USB device found, idVendor=1d6b, idProduct=0002
<6>[64952.258750,0] usb usb1: New USB device strings: Mfr=3, Product=2, SerialNumber=1
<6>[64952.258924,0] usb usb1: Product: Qualcomm On-Chip EHCI Host Controller
<6>[64952.259099,0] usb usb1: Manufacturer: Linux 3.4.0-gc6fc9e1-00001-gf3bf628 ehci_hcd
<6>[64952.259197,0] usb usb1: SerialNumber: msm_hsusb_host
<6>[64952.261601,0] hub 1-0:1.0: USB hub found
<6>[64952.261731,0] hub 1-0:1.0: 1 port detected
<3>[64952.518160,0] hub 1-0:1.0: unable to enumerate USB device on port 1
<6>[64955.061905,0] msm_otg f9a55000.usb: USB in low power mode
I guess that reader is not supported by the current kernel. It looks like it is detected as a USB Hub, not as a simple USB storage device.
Code:
<6>[64952.261601,0] hub 1-0:1.0: USB hub found
Get a cheap one (e.g: like this http://www.ebay.co.uk/itm/OTG-USB-M...aphy_MemoryCardReaders_RL&hash=item5d46cbed5b
I have a couple and they do the job fine.
Click to expand...
Click to collapse
Thanks, Im from Argentina and I couldn't find a micro usb card reader, and if I buy it in ebay I should wait 3 - 4 months because of post office delay, maybe I'll buy a pendrive. My reader is similar to yours, but with usb not micro

I have this problem too... Any other ideas?

[Q] GTP31xx charging issues

(Samsung GT-P3110 Galaxy Tab 2 7.0")
Recently the tab stopped charging while plugged to AC and in use...
Ok, it's not losing power as fast as when unplugged, but almost!
I have to turn it off and AC charge for hours and hours as the phone seem to think that it's connected to a PC all of a sudden!
Wasn't an issue up until last week then for no apparent reason it says "not charging" in Battery..
After a lot of reading around and confirming that it was not the charger or cable i dug deeper and it seems that USB charging is turned of by default on the Tab 2's except all custom kernels have it turned on and now that's always on!
Since the tab wont change charger state from USB to AC then all I'm getting is a slow trickle charge at the same slow pace, and it don't whether I am plugged in to the AC or into the PC!
So i figured hey, it's just a terminal command to force it to switch and that's when i realized that, no it's not that easy!
As USB charge is not only the default setting in custom kernels, its the ONLY option available!
The ability to have the tab swiftly reach a full charge while in use seems to currently be impossible with custom kernels it seems!
Is there a work around, or have everyone been to busy implementing USB charging to remember that sometimes you don't want to spend 8+ hours with tab turned off to get a full charge?
Below is a logcat from connecting and disconnecting charger cable to my GTP3110 WHILE connected to an AC adapter!
Code:
Android Tuner (logcat)
======================
V/ 96.627105 c1 17 usb ta_nconnected_irq : VBUS ON
V/ 96.627899 c1 17 [BAT_MANAGER] Charger Connected
V/ 96.627960 c1 58 musb-hdrc musb-hdrc: USB/TA Connect
V/ 97.256317 c1 1662 check_charger_type : Charger type is [USB], adc=594
V/ 97.256317 c1 1662 charger_detect_work: cable flag : 1, cable_type : 1
V/ 97.256347 c1 1662 smb136-charger 5-004d: smb136_set_charging_state: set charger current USB & Default
V/ 97.262725 c1 1662 cable detect:USB attach, current device = 0x0080
V/ 97.262756 c1 1662 espresso_con_usb_charger_attached, USB_EVENT_VBUS
V/ 97.262756 c1 1662 otg: otgwl_hold
V/ 97.262847 c0 58 musb-hdrc musb-hdrc: VBUS Connect
V/ 97.262908 c0 58 musb-omap2430 musb-omap2430: runtime resume
V/ 97.263061 c0 58 omap4430_phy_set_clk: clock (0 --> 1)
V/ 97.504699 c0 58 omap4430_phy_power : clock 1, power ON
W/ 103.476745 c0 15 smb136-charger 5-004d: smb136_i2c_read: err -6
W/ 103.476806 c0 15 smb136-charger 5-004d: smb136_read_status : I2C read fail addr: 0x35
V/ 103.535827 c0 15 smb136-charger 5-004d: addr : 0x35, data : 0x40, addr : 0x36, data : 0x31
V/ 103.806243 c1 17 usb ta_nconnected_irq : VBUS OFF
V/ 103.806793 c1 17 otg: otgwl_temporary_hold
V/ 103.806793 c0 78 musb-hdrc musb-hdrc: VBUS Disconnect
V/ 103.806793 c1 17 [BAT_MANAGER] Charger Disconnect
V/ 103.806823 c0 78 omap4430_phy is : ON
V/ 103.806854 c0 78 omap4430_phy_power : clock 1, power OFF
V/ 103.806854 c1 1662 charger_detect_work: cable flag : 0, cable_type : 0
V/ 103.806854 c1 1662 smb136-charger 5-004d: smb136_set_charging_state: Set discharging default
V/ 103.806884 c0 78 musb-hdrc musb-hdrc: ID float
V/ 103.806884 c0 605 android_work: did not send uevent (0 0 (null))
V/ 103.806915 c0 605 otg espresso_otg_work(920)
current device 0080
V/ 103.806915 c0 605 otg current device is not USB Host.
V/ 103.812805 c1 1662 cable detect:USB detach, current device = 0x0000
V/ 104.002899 c0 658 musb-omap2430 musb-omap2430: runtime suspend
V/ 104.002899 c0 658 omap4430_phy_set_clk: clock (1 --> 0)

I use cyanogenmod, and I'm fairly sure that my tablet charges faster on ac than when plugged into my PC. Have you tried a different kernel/rom?
Sent from my Galaxy S4

What kernel are you on, been trying with 5-6 different ones and they all display that bug that they are incapable in detecting if you are really on AC or USB...
It IS a kernel issue I am 100% sure of that since the Tab has USB charging turned off by default exactly to avoid this issue...

USB C with OTG with charging: Portables as Computers in 2019 and on

I'm seeing multiple kickstarters for "turn your cell phone into a computer" type appliances. Typically a keyboard thing that includes an OTG adapter, an HDMI output, etc... Most /claim/ that you can charge your cell phone while working. They are all overfunded (and shaky, I'm not recommending them)
I've tried many times to use phones / tablets as OTG devices to interface to hardware of different types. Everything from diagnostic tools to robot arms (no really, I'd post links but I'm a newbie here and linking is always suspect; I don't want to be accused of spamming).
One major issue always comes up: You can NOT charge while running OTG on /most/ phones. I've done a lot of research on this, and found a few older phones / tablets that seem to support it. I can't find a /single/ USB C based device that is known to do it. Perhaps it's just not an advertising point? MFGRs don't see most users doing that, so they don't support it?
So what am I asking?
1. Is anyone with a modern phone, especially USB C based, actually using OTG and charging at the same time? Please note: I'm NOT talking about charging the OTG device /from/ the phone, I'm talking about charging the phone while using OTG devices. If you are doing this, please: What phone, what OTG adapter?
2. Is this something people have an interest in doing? Why yes, why no? For me, I don't have any interest in using my cell as a computer, but being able to plug it into equipment and monitor / control operation while not running the battery flat would be great!
3. Is anyone aware of the actual specs, technical issues, reasons for the current state of affairs? I've started a web page to track this issue (again, not linking, not spamming) and would love to hear from anyone with direct knowledge of the technical issues involved. Why doesn't it always work? What specification needs to be met? What future is there for this sort of feature?
Here are the devices I understand DO support it:
RaspBerry Pi
Amazon FireStick
Amazon Kindle Fire 10???
Dell Venue 8 pro
HTC One M9
Lenovo Miix 2 tablet?
Lenovo Yoga Book
Moto G 4?
NuVision Tab 8 (Windows 10) after restart only?
Nvidia Shield K1 table??
OnePlus One
Samsung Galaxy S3
Samsung Galaxy S4
Samsung Galaxy S5
Samsung Galaxy Note 3
Samsung Galaxy Tab A7?
Samsung Galaxy Tab A8.0
Samsung Galaxy Tab S2??
Samsung Galaxy Tab S 10.5?
Samsung Galaxy View
Snes Classic
And here are some that support OTG, but will not charge at the same time:
Amazon Kindle Fire 7
Asus T100TAM
LG Nexus 5
LG Nexus 7
Samsung Galaxy Note 4
Samsung Galaxy S7
Samsung Galaxy S8+
Samsung Galaxy S9
Samsung Galaxy TabPro 10.1
Please do add on your info?

How's the research going? I've been looking for tabs that support USB-OTG and charging when connected to a powered USB hub as I'm planning a car install. Sort of interested in whether the newer tabs like the Samsung Tab S5E support this or not. That would be sweet installed into the dash!

Prostheta said:
How's the research going? I've been looking for tabs that support USB-OTG and charging when connected to a powered USB hub as I'm planning a car install. Sort of interested in whether the newer tabs like the Samsung Tab S5E support this or not. That would be sweet installed into the dash!
Click to expand...
Click to collapse
..
Sent from my SM-S767VL using Tapatalk

I decided to buy a Galaxy Tab A 10.1 2019 (SM-T515) and will report back with findings once I test with a USB C OTG Y cable adapter. It's possible that I may need to fabricate some discrete switch into the tablet's USB to alternate between OTG and charging. Not an ideal state of affairs.
Additionally, I've pondered whether devices with pogo pin dock connectors such as the Samsung Galaxy Tab S5E can charge that way whilst using the USB port for OTG devices. Whilst not strictly meeting the requirement of the list (one USB port charging whilst using OTG) it does present a compelling alternative.

It seems that with USB C, the "standard" includes OTG host charging but called pass through charging, dual role data or something along those lines. I connected my Tab A 10,1 to a Kingston Nucleum non-powered hub via the connected USB C lead, then plugged in a charger to another USB C port. Charging! Pretty much the same works whichever combination of connections is used. It seems the devices negotiate who is the power delivery and at what level. There are some settings in android to set also.

Prostheta said:
It seems that with USB C, the "standard" includes OTG host charging but called pass through charging, dual role data or something along those lines. I connected my Tab A 10,1 to a Kingston Nucleum non-powered hub via the connected USB C lead, then plugged in a charger to another USB C port. Charging! Pretty much the same works whichever combination of connections is used. It seems the devices negotiate who is the power delivery and at what level. There are some settings in android to set also.
Click to expand...
Click to collapse
Yes, but does it charge while also having other external devices such as a USB hard drive connected to the hub(which also requires being powered) at the same time? The USB hard drive receives power from the hub like it should and the device charges simultaneously?
Sent from my SM-S767VL using Tapatalk

I've tested with a USB mouse dongle, and when it arrives, my Meizu DAC pro. So far it seems positive. It does however default to the 5v charging level, but that's fine.

Prostheta said:
I've tested with a USB mouse dongle, and when it arrives, my Meizu DAC pro. So far it seems positive. It does however default to the 5v charging level, but that's fine.
Click to expand...
Click to collapse
Hey did you ever get this working? Does USB-C charge and connect devices simultaneously?
thx

Yes, everything works fine....sometimes. The Kingston Nucleum USB C hub has a 5v supply plugged into a 12v car charger port that has USB PD and QC3.0 options. The tablet is problematic in that once the power to the charger (and hence the hub) is removed after turning the car off, the tablet decides that it'll take over the role of supplying power to the hub which discharges the tablet battery whilst the car is off. Also, the hub doesn't always arrange the attached devices correctly, leaving the USB SSD unmounted and the tablet sat alone charging. I'm unsure what is happening, however I'll blame USB C for being a dumpster fire.

Prostheta said:
Yes, everything works fine....sometimes. The Kingston Nucleum USB C hub has a 5v supply plugged into a 12v car charger port that has USB PD and QC3.0 options. The tablet is problematic in that once the power to the charger (and hence the hub) is removed after turning the car off, the tablet decides that it'll take over the role of supplying power to the hub which discharges the tablet battery whilst the car is off. Also, the hub doesn't always arrange the attached devices correctly, leaving the USB SSD unmounted and the tablet sat alone charging. I'm unsure what is happening, however I'll blame USB C for being a dumpster fire.
Click to expand...
Click to collapse
Thanks, I've been looking for a tablet for car infotainment and gauges/data logging, finally decided on a Lenovo tab 4 8 plus Verizon model, precisely because it had a USB-C 3.0 port, and it is rootable. Then I started reading about the complicated mess that the USB-C standard is. Anyway I'll make a post in the General Discussion>Connected Car forums in a month or so when I get everything connected... work and kids always stand in the way of progress

Is that the place to discuss these things? I've had a lot of difficulty trying to find the best place for this, as most are either too specific or not quite on-topic.

I'm updating all of my comments in various threads on this. I've found that with my specific setup with the Kingston Nucleum hub, there are issues. I'm unsure as to the cause, but hey. Firstly, the tablet needs to be powered up before I turn on the car or at least supply power to the hub. It seems that if this is done the other way around, that is, to power on the car accessories (and hence the hub) then it seems to end up staying as the "device in charge" of the USB network. The tablet charges (slowly) but doesn't mount the USB SSD drive or the USB DAC. If the tablet is powering the hub (in charge, so to speak) then applying power to the hub via the car accessory supply has it working as expected. This does mean that the tablet ends up with a power drain when the car/USB power supply is powered down which isn't ideal. I've gotten into the habit of turning the tablet off when I park for long periods of time, however I need to boot it up to "take charge" of the USB hub network before I can start the car. It's a bit of a bummer.

Prostheta said:
It seems that with USB C, the "standard" includes OTG host charging but called pass through charging, dual role data or something along those lines.
Click to expand...
Click to collapse
Yes, my notebook has an USB-C port and can be charged via USB-C PD pass trough by a USB-C dock (USB hub with LAN). Why the Android Smartphones/Tablets don't do the same?
JamesMNewton said:
3. Is anyone aware of the actual specs, technical issues, reasons for the current state of affairs? I've started a web page to track this issue (again, not linking, not spamming) and would love to hear from anyone with direct knowledge of the technical issues involved. Why doesn't it always work? What specification needs to be met? What future is there for this sort of feature?
Click to expand...
Click to collapse
I found this diagram of a "accessory charger adapter" (USB 2.0 spec, not 3.0+?): https://superuser.com/questions/1124237/windows-tablet-otg-simultaneously-charge/1124354#1124354
JamesMNewton said:
Supports OTG, but will not charge at the same time:
LG Nexus 7
Click to expand...
Click to collapse
Charging+OTG was possible on Nexus 7 2012+2013 using a custom kernel!

Yeah, the Nexus 7 was one of the rarities amongst pre-USB3.0 spec ports that could be manipulated into simultaneous data and charging.
My ongoing concern is with ennumeration; I can get my Tab A10,1 charging through the hub easily. However, to access devices attached to the hub, the tablet needs to be active first and then power applied to the hub. Applying power to the hub first, then starting the tablet leads to charging only. The problem as I see it is the negotiation over host rights on a USB network, and whether re-negotiation or forcing priority is possible. It could be a hardware issue built into the USB-PD negotiation specs, a hardware issue with the tablet or simply a software issue in the tablet. I'm not really wanting to wade into the USB specs since they're notoriously vague and contradictary these days....

Prostheta said:
It does however default to the 5v charging level, but that's fine.
Click to expand...
Click to collapse
I would want to know why? It was the same with the Pixel C, but Notebooks will charge at faster speed via PD pass through.
skally said:
I tried powering the PixelC and it works but not at the same speed as the same charger connected directly. I estimate about 1-1.2A
Click to expand...
Click to collapse
PD charging of a Pixel 1 via pass trough (interesting read on PD negotiations):
"The USB Power Delivery behavior seen when the Plugable USB-C 3-Port Hub is passing through power from Google Pixel's own USB Power Delivery charger is a lot different from what you get when you plug the charger directly into the Google Pixel." https://gtrusted.com/review/how-the...google-pixel-phone-using-usb-power-delivery-2

JamesMNewton said:
You can NOT charge while running OTG on /most/ phones. I've done a lot of research on this, and found a few older phones / tablets that seem to support it. I can't find a /single/ USB C based device that is known to do it. Perhaps it's just not an advertising point? MFGRs don't see most users doing that, so they don't support it?
Click to expand...
Click to collapse
I was able to use the ethernet, PD pass through and USB host feature at the same time with:
1. Samsung Galaxy Tab A 10.5 2018 (Stock ROM)
2. Google Pixel 2 (LineageOS 17.1)
The USB-C standard really has made things easier! I was not able to use the HDMI output because these devices do not support DP alt mode! :-(
"USB C Charger RAVPower 65W Fast Wall Charger, PD3.0 & GaN Tech 2-Port Power Adapter" https://www.amazon.com/Charger-RAVPower-Adapter-Compact-Foldable/dp/B08MWG612F
"Dockteck 4-in-1 USB C Hub with [email protected] HDMI HDR, 100W PD, USB 3.0 and Gigabit Ethernet, Multiport USB Type C Adapter"
Amazon.com: dockteck 4-in-1 USB C Hub with [email protected] HDMI HDR, 100W PD, USB 3.0 and Gigabit Ethernet, Multiport USB Type C Adapter for MacBook/Pro/Air, iMac, XPS, iPad Pro 2020 2018 and More: Computers & Accessories
Amazon.com: dockteck 4-in-1 USB C Hub with [email protected] HDMI HDR, 100W PD, USB 3.0 and Gigabit Ethernet, Multiport USB Type C Adapter for MacBook/Pro/Air, iMac, XPS, iPad Pro 2020 2018 and More: Computers & Accessories
www.amazon.com
USB Device Info
---
Device Path: /dev/bus/usb/001/003
Device Class: Unknown (0x11)
Vendor ID: 2109
Vendor Name (reported): VIA Labs, Inc.
Vendor Name (from DB): not found
Product ID: 8817
Product Name (reported): PD3.0 USB-C Device
Product Name (from DB): not found
Additional Info
Interface #0
Class: Unknown (0x11)
Endpoints: none
---
Device Path: /dev/bus/usb/002/003
Device Class: Use class information in the Interface Descriptors (0x0)
Vendor ID: 0bda
Vendor Name (reported): Realtek
Vendor Name (from DB): Realtek Semiconductor Corp.
Product ID: 8153
Product Name (reported): USB 10/100/1000 LAN
Product Name (from DB): not found
Additional Info
Interface #0
Class: Vendor Specific (0xff)
Endpoint: #0
Address : 0x81 (10000001)
Number : 1
Direction : Inbound (0x80)
Type : Bulk (0x2)
Poll Interval : 0
Max Packet Size: 1024
Attributes : 00000010
Endpoint: #1
Address : 0x02 (00000010)
Number : 2
Direction : Outbound (0x0)
Type : Bulk (0x2)
Poll Interval : 0
Max Packet Size: 1024
Attributes : 00000010
Endpoint: #2
Address : 0x83 (10000011)
Number : 3
Direction : Inbound (0x80)
Type : Interrupt (0x3)
Poll Interval : 8
Max Packet Size: 2
Attributes : 00000011
Interface #1
Class: Communication Device (0x2)
Endpoint: #0
Address : 0x83 (10000011)
Number : 3
Direction : Inbound (0x80)
Type : Interrupt (0x3)
Poll Interval : 8
Max Packet Size: 16
Attributes : 00000011
Interface #2
Class: Communication Device Class (CDC) (0xa)
Endpoints: none
Interface #3
Class: Communication Device Class (CDC) (0xa)
Endpoint: #0
Address : 0x81 (10000001)
Number : 1
Direction : Inbound (0x80)
Type : Bulk (0x2)
Poll Interval : 0
Max Packet Size: 1024
Attributes : 00000010
Endpoint: #1
Address : 0x02 (00000010)
Number : 2
Direction : Outbound (0x0)
Type : Bulk (0x2)
Poll Interval : 0
Max Packet Size: 1024
Attributes : 00000010
---
Device Info
Device Path: /dev/bus/usb/002/004
Device Class: Use class information in the Interface Descriptors (0x0)
Vendor ID: 05e3
Vendor Name (reported): Generic
Vendor Name (from DB): Genesys Logic, Inc.
Product ID: 0749
Product Name (reported): USB3.0 Card Reader
Product Name (from DB): not found
Additional Info
Interface #0
Class: Mass Storage Device (0x8)
Endpoint: #0
Address : 0x81 (10000001)
Number : 1
Direction : Inbound (0x80)
Type : Bulk (0x2)
Poll Interval : 0
Max Packet Size: 1024
Attributes : 00000010
Endpoint: #1
Address : 0x02 (00000010)
Number : 2
Direction : Outbound (0x0)
Type : Bulk (0x2)
Poll Interval : 0
Max Packet Size: 1024
Attributes : 00000010
---

SGH-i200 said:
"Charging+OTG was possible on Nexus 7 2012+2013 using a custom kernel! "
Click to expand...
Click to collapse
Actually You don't need to modify anything with certain tablets to do otg and charging at the same time:
I bought a Lenovo tab 8 hd and only by using a so called "charging y cable" (please Google it) I was able to listen music through my USB dac and charge the tab simultaneously.
Maybe getting into developer mode helped? I don't know, but it's so easy to try...
Daniel