Is it possible to get this working on the Acer A500 with the ICS leak? I've been able to use it on all 3.2 ROMs and Thors ICS, but when moving to the leak it will not execute "usb_ethernet", it recognizes it as a command but no output is shown. system is set to RW but it still doesn't load.
Just curious.
Thanks everyone!
On the ICS roms, usb1 is not showing up as a network interface
When I look at my dmesg, it shows "USB Ethernet Ver .2" but never loads the ASISX driver.
On Thors ICS it registers "asix 1-1.0: eth0: register "asix" at usb-tegra-ehci.2-1, ASIX AX8877 USB 2.0 Ethernet"
Perhaps its due to the leaked builds using the stock kernel?
I'm using Flexreaper R10 and here is what I get
I flashed the zip provided HERE
I insert the USB adapter and dmesg shows
Code:
<6>[ 146.691880] usb 1-1.3: new high speed USB device number 5 using tegra-ehci
<6>[ 146.763606] usb 1-1.3: New USB device found, idVendor=13b1, idProduct=0018
<6>[ 146.691880] usb 1-1.3: new high speed USB device number 5 using tegra-ehci
<6>[ 146.763606] usb 1-1.3: New USB device found, idVendor=13b1, idProduct=0018
<6>[ 146.764309] usb 1-1.3: New USB device strings: Mfr=1, Product=2, SerialNumber=3
<6>[ 146.764980] usb 1-1.3: Product: USB 2.0 Network Adapter ver.2
<6>[ 146.765345] usb 1-1.3: Manufacturer:
<6>[ 146.765721] usb 1-1.3: SerialNumber: 01E821 device strings: Mfr=1, Product=2, SerialNumber=3
<6>[ 146.764980] usb 1-1.3: Product: USB 2.0 Network Adapter ver.2
<6>[ 146.765345] usb 1-1.3: Manufacturer:
<6>[ 146.765721] usb 1-1.3: SerialNumber: 01E821
I try to manually load the module I need, "asix410.ko" and I get
Code:
<3>[ 463.740506] asix410: version magic '2.6.36.3 SMP preempt mod_unload ARMv7 ' should be '2.6.39.4+ SMP preempt mod_unload ARMv7 '
Any ideas?
I see that it says that the kernel is not the proper version, but I was under the impression that as log as the kernel is 2.6.39.x, this driver should work.
Thanks!
Try using a hex editor to change the version info in the .ko file to match what the system is expecting and give it a try. I've done that successfully before with a cifs.ko module.
saeba said:
Try using a hex editor to change the version info in the .ko file to match what the system is expecting and give it a try. I've done that successfully before with a cifs.ko module.
Click to expand...
Click to collapse
I did that, having seen you do that with cifs, but the kernel is was made for has like 6 characters whereas this one has 7, so because of that the kernel finds that there is a mismatch and STILL won't load the module.
I am attempting to connect to a serial device from a Android tablet, via
USB-Serial Converter. In order to interface with this particular serial
device, a linux library must be used as an interface.
This library takes an integer as an argument, and tries to connect to the
device at the following location : /dev/ttySx, where x is the arguement. If
2 is provided to the library, it will look for the device at /dev/ttyS2.
Very simple.
When I connect the device to Ubuntu, there's is one simple extra step for
this device to work with the USB-Serial convert. The converter is exposed
at the location /dev/ttyUSB1. I just need to create a simple soft-link with
the serial port pattern (etc "ln -s /dev/ttyUSV1 /dev/ttyS99"). It works
perfectly.
Now the problem arises for Android. The converter is seen at
/dev/bus/usb/01/01. I never seen this convention in other project. The
question is how can I direct the I/O from /dev/ttySX to /dev/bus/usb/01/01
ehpaul said:
I am attempting to connect to a serial device from a Android tablet, via
USB-Serial Converter. In order to interface with this particular serial
device, a linux library must be used as an interface.
This library takes an integer as an argument, and tries to connect to the
device at the following location : /dev/ttySx, where x is the arguement. If
2 is provided to the library, it will look for the device at /dev/ttyS2.
Very simple.
When I connect the device to Ubuntu, there's is one simple extra step for
this device to work with the USB-Serial convert. The converter is exposed
at the location /dev/ttyUSB1. I just need to create a simple soft-link with
the serial port pattern (etc "ln -s /dev/ttyUSV1 /dev/ttyS99"). It works
perfectly.
Now the problem arises for Android. The converter is seen at
/dev/bus/usb/01/01. I never seen this convention in other project. The
question is how can I direct the I/O from /dev/ttySX to /dev/bus/usb/01/01
Click to expand...
Click to collapse
Which android device are you using and with which kernel and android build? In my case I am using kernel 3.0.+ with CM9 on a Kindle Fire and I do see the /dev/ttyUSBx getting created along with the /dev/bus/01/01. It has been a month since I tried it but I was able to see the problem you see and here is what I remember. When you plug the serial convertor both /dev get created but since the Android USB support only uses /dev/bus/usb/01/01, the kernel in order to save power "unmounts" the unused /dev/ttyUSBx after a certain time period. You can try using dmesg to see if that is happening in your case also. There is a kernel build flag that tells the kernel not to "unmount" it but I dont remember it off-hand. Best would be to use the Android USB support to talk to your device instead of the linux library if you can.
--------Update after testing it out again and checking dmesg etc---------------------------
When I plug the FTDI cable this is what shows up in dmesg:
<6>usb 1-1: new full speed USB device number 8 using musb-hdrc
<3>usb 1-1: device v0403 p6001 is not supported
<6>ftdi_sio 1-1:1.0: FTDI USB Serial Device converter detected
<6>usb 1-1: Detected FT232RL
<6>usb 1-1: Number of endpoints 2
<6>usb 1-1: Endpoint 1 MaxPacketSize 64
<6>usb 1-1: Endpoint 2 MaxPacketSize 64
<6>usb 1-1: Setting MaxPacketSize 64
<6>usb 1-1: FTDI USB Serial Device converter now attached to ttyUSB0
At this point if I check the filesystem I see both /dev/ttyUSB0 and /dev/bus/usb/001/0xx [ xx being the current device number ]
I can work with /dev/ttyUSB0 and all is well. Now as soon as I invoke any app that uses Android USB Host Mode API to connect
to USB (for enumeration, connection or whatever) the /dev/ttyUSB0 gets disconnected with following message from dmesg:
<6>ftdi_sio ttyUSB0: FTDI USB Serial Device converter now disconnected from ttyUSB0
<6>ftdi_sio 1-1:1.0: device disconnected
At this point /dev/bus/usb/001/0xx is still alive and if I enumerate the USB devices using android API I see the same as the adaptor
name. So looks like till the Android USB API takes over the FTDI is available as /dev/ttyUSB0 but not afterwards.
------------------------Update ends--------------------------------
(a) If you see a /dev/ttyUSB1 device node created, then you can try the following in an Android terminal:
Code:
stty -F /dev/ttyUSB1 raw
cat /dev/ttyUSB1
If this shows data from your device, then you can use directly the device node to read()/write(), POSIX-style, without needing any library.
(b) If you cannot see a /dev/ttyUSB1 in your Android device, this means that you need to load the ftdi_sio kernel module (check android.serverbox.ch/?p=285 for instructions).
Let me link that for you:
"How to enable FTDI Support for your USB Host featuring Honeycomb Tablet (including sample native application)"
Also have a look at FTDI's Android Paper:
"White Paper: Connecting Peripherals to an Android Platform"
and this relevant blog, from this XDA thread.
pankaj013 said:
Which android device are you using and with which kernel and android build? In my case I am using kernel 3.0.+ with CM9 on a Kindle Fire and I do see the /dev/ttyUSBx getting created along with the /dev/bus/01/01. It has been a month since I tried it but I was able to see the problem you see and here is what I remember. When you plug the serial convertor both /dev get created but since the Android USB support only uses /dev/bus/usb/01/01, the kernel in order to save power "unmounts" the unused /dev/ttyUSBx after a certain time period. You can try using dmesg to see if that is happening in your case also. There is a kernel build flag that tells the kernel not to "unmount" it but I dont remember it off-hand. Best would be to use the Android USB support to talk to your device instead of the linux library if you can.
--------Update after testing it out again and checking dmesg etc---------------------------
When I plug the FTDI cable this is what shows up in dmesg:
<6>usb 1-1: new full speed USB device number 8 using musb-hdrc
<3>usb 1-1: device v0403 p6001 is not supported
<6>ftdi_sio 1-1:1.0: FTDI USB Serial Device converter detected
<6>usb 1-1: Detected FT232RL
<6>usb 1-1: Number of endpoints 2
<6>usb 1-1: Endpoint 1 MaxPacketSize 64
<6>usb 1-1: Endpoint 2 MaxPacketSize 64
<6>usb 1-1: Setting MaxPacketSize 64
<6>usb 1-1: FTDI USB Serial Device converter now attached to ttyUSB0
At this point if I check the filesystem I see both /dev/ttyUSB0 and /dev/bus/usb/001/0xx [ xx being the current device number ]
I can work with /dev/ttyUSB0 and all is well. Now as soon as I invoke any app that uses Android USB Host Mode API to connect
to USB (for enumeration, connection or whatever) the /dev/ttyUSB0 gets disconnected with following message from dmesg:
<6>ftdi_sio ttyUSB0: FTDI USB Serial Device converter now disconnected from ttyUSB0
<6>ftdi_sio 1-1:1.0: device disconnected
At this point /dev/bus/usb/001/0xx is still alive and if I enumerate the USB devices using android API I see the same as the adaptor
name. So looks like till the Android USB API takes over the FTDI is available as /dev/ttyUSB0 but not afterwards.
------------------------Update ends--------------------------------
Click to expand...
Click to collapse
are you using ftdi's code to modify it?
change /system/etc/permissions
this is from a question on stackoverflow titled "Android USB host and hidden devices" (sorry, i can't add external URLs yet). it worked like a charm on a micromax A120 canvas 2 phone (kitkat 4.4.2). now i can control my Arduino! i used busybox tools to do all the command line work (otherwise chmod wouldn't work). my steps (perhaps some were not required):
(0) install PDAnet drivers on my Windows 8 computer.
(1) root the phone using Vroot (now called iRoot). Very simple, only catch is that the su grant/deny page is partly in chinese, no big deal.
(2) install busybox and jackpal's Android-Terminal-Emulator available on github and the google play store (free).
(3) open a terminal window and become the superuser:
# su
(4) the file system may be read-only, so you might have to remount it:
# mount -o rw,remount -t yaffs2 /
or
# mount -o rw,remount -t rootfs /
or
# mount -o rw,remount -t rootfs rootfs /system
(5) make default.prop read/write:
chmod 666 /default.prop
(6) edit /default.prop, make the following changes:
ro.secure=0
ro.debuggable=1
persist.service.adb.enable=1
... And this is the real meat of it:
(7) To enable USB host API support you should add a file named
android.hardware.usb.host.xml and containing the following lines:
<permissions>
<feature name="android.hardware.usb.host"/>
</permissions>
into folder
/system/etc/permissions
in that folder find file named
handheld_core_hardware.xml or tablet_core_hardware.xml
and add
<feature name="android.hardware.usb.host" />
into <permissions> section.
(8) Reboot your device. Usb host api should work.
This is my first thread/post in the forum, so I am not quite sure if this is the right place for this question. I recently happened to see a schematic diagram of the GT-S5360. I noticed that the main chipset includes support for USBOTG HS(480 Mbps).
http://forum.xda-developers.com/showthread.php?t=1489767
If this is indeed true, is there any way we can enable this feature in the GT-S5360? I mean, is there a custom ROM/Kernel which enables this feature?
Apologies if I have posted this in the wrong section.
If you have usb otg cable then try kuro and merruk kernel....
And search before posting
Sent from my GT-S5360 using xda premium
Hi, please look in the dev section. We already have 2-3 kernels which enable us to use this feature, if i'm not wrong.
Sent from my GT-S5360 using xda premium
that is a block diagram not schematic
alkesh95 said:
If you have usb otg cable then try kuro and merruk kernel....
And search before posting
Sent from my GT-S5360 using xda premium
Click to expand...
Click to collapse
Thanks for the response Alkesh, I will try those. I had infact searched for something similar in the forum but could not come up with any names - maybe I did something wrong. I found this though: http://forum.xda-developers.com/showthread.php?t=1493325
my kernel didn't support it
my kernel is based on samsung's sourcode. on other hand, our "beloved" samsung have messed up the source code. thus, make the OTG stuff on the kernel didn't work.
kurotsugi said:
my kernel didn't support it
my kernel is based on samsung's sourcode. on other hand, our "beloved" samsung have messed up the source code. thus, make the OTG stuff on the kernel didn't work.
Click to expand...
Click to collapse
Thanks for clarifying Kurotsugi. So is there any kernel which supports this feature currently? I searched the development forums - seems Merruk also doesnt support this feature currently, Maroc-OS is working on it as I understand from http://forum.xda-developers.com/showthread.php?t=1550065
xdroid_noob said:
Thanks for clarifying Kurotsugi. So is there any kernel which supports this feature currently? I searched the development forums - seems Merruk also doesnt support this feature currently, Maroc-OS is working on it as I understand from http://forum.xda-developers.com/showthread.php?t=1550065
Click to expand...
Click to collapse
if you have the cable you can try that.
ask samsung
works now
Hi guys.
Sorry for necroposting.
Also this post should probably go to development section, but I don't have access to post there (yet?).
First of all, BCM21553 seems to lack ID pin. ID pin from USB connector goes straight to PMU chip instead.
It is used to detect the type of the charger as far as I can tell.
It's not a big deal to extend PMU isr's to support OTG cable as a different type of charger (yes, external +5v source is required).
The real trick (for me) was to find a way to force DWC core into Host mode. Looks like Synopsis is no longer sharing datasheets with public.
Luckily Broadcom is not the only company which uses the same USB IP.
It looks like DWC driver was designed for an older kernel version.
It was partly ported by Samsung, but only as much as required to support the device mode.
It's still very flaky and fragile. For instance, it you try to cat /sys/devices/lm-2/mode, it will panic the stock kernel.
All HC code is non-functional and does not even build out of the box.
It took me almost a week to fix all the bugs and to be able to build standalone dwc_otg.ko with working HC code.
At the moment I managed to successfully mount a memory stick on my phone. Device mode is not there yet, but I don't really need it:
I'm building a telemetry project and I need the cheapest solution.
If anypony is still interested in HC on SGY, please reply and I will follow up with more details.
thats pretty awesome news.....this should be moved to development section......
Sent from my Typewriter
A bit of logs to support my statement
Code:
<6>[ 1185.825469] dwc_otg: version 2.60a 22-NOV-2006
<6>[ 1185.825531] lm_driver_register()
<6>[ 1185.825592] lm_match()
<6>[ 1185.825622] lm_bus_probe()
<3>[ 1185.825653] Debug level= 17
<3>[ 1185.825683] dwc_otg lm-2: dwc_otg_driver_probe, version 020-10-01-09
<6>[ 1185.826019] It is Ok to create charging current work queue...
<4>[ 1185.826080] >> create: c483f5e0
<6>[ 1185.826324] start_usb_clk_phy
<4>[ 1185.826354] DWC_otg: 1
<4>[ 1185.826385] DWC_otg: 2
<4>[ 1185.826416] DWC_otg: dwc_otg_core_reset() GRSTCTL=80000000
<4>[ 1185.826446] DWC_otg: dwc_otg_core_reset() value to set=80000001
<4>[ 1185.826477] DWC_otg: 3
<4>[ 1185.931610] DWC_otg: 1
<4>[ 1185.931640] DWC_otg: 2
<4>[ 1185.931671] DWC_otg: dwc_otg_core_reset() GRSTCTL=80000000
<4>[ 1185.931701] DWC_otg: dwc_otg_core_reset() value to set=80000001
<4>[ 1185.931701] DWC_otg: 3
<4>[ 1186.853454] >>>> unicorn 1
<4>[ 1186.853485] dwc_otg_pcd_init(c483f5e0)
<4>[ 1186.853485] Dedicated Tx FIFOs mode
<4>[ 1186.853607] registering handler for irq21
<6>[ 1186.853698] DWC_otg: pcd->setup_pkt=0xff6b3000, pcd->setup_pkt_dma_handle=0x82982000
<6>[ 1186.853729] DWC_otg: pcd->status_buf=0xff6b4000, pcd->status_buf_dma_handle=0x85d33000
<4>[ 1186.853729] pcd->setup_pkt=0xff6b3000, pcd->setup_pkt_dma_handle=0x82982000
<4>[ 1186.853759] pcd->status_buf=0xff6b4000, pcd->status_buf_dma_handle=0x85d33000
<6>[ 1186.853790] DWC_otg: Using DMA mode
<6>[ 1186.853820] dwc_otg lm-2: DWC OTG Controller
<6>[ 1186.853851] dwc_otg lm-2: new USB bus registered, assigned bus number 1
<6>[ 1186.853912] dwc_otg lm-2: irq 21, io mem 0x00000000
<6>[ 1186.853973] DWC_otg: Init: Port Power? op_state=1
<6>[ 1186.853973] DWC_otg: Init: Power Port (0)
<7>[ 1186.854064] usb usb1: default language 0x0409
<7>[ 1186.854125] usb usb1: udev 1, busnum 1, minor = 0
<6>[ 1186.854156] usb usb1: New USB device found, idVendor=1d6b, idProduct=0002
<6>[ 1186.854156] usb usb1: New USB device strings: Mfr=3, Product=2, SerialNumber=1
<6>[ 1186.854187] usb usb1: Product: DWC OTG Controller
<6>[ 1186.854217] usb usb1: Manufacturer: Linux 2.6.35.7 dwc_otg_hcd
<6>[ 1186.854217] usb usb1: SerialNumber: lm-2
<7>[ 1186.856048] usb usb1: usb_probe_device
<7>[ 1186.856109] usb usb1: configuration #1 chosen from 1 choice
<7>[ 1186.856170] usb usb1: adding 1-0:1.0 (config #1, interface 0)
<7>[ 1186.857482] hub 1-0:1.0: usb_probe_interface
<7>[ 1186.857513] hub 1-0:1.0: usb_probe_interface - got id
<6>[ 1186.857543] hub 1-0:1.0: USB hub found
<6>[ 1186.857604] hub 1-0:1.0: 1 port detected
<7>[ 1186.857604] hub 1-0:1.0: standalone hub
<7>[ 1186.857635] hub 1-0:1.0: ganged power switching
<7>[ 1186.857666] hub 1-0:1.0: individual port over-current protection
<7>[ 1186.857666] hub 1-0:1.0: Single TT
<7>[ 1186.857696] hub 1-0:1.0: TT requires at most 8 FS bit times (666 ns)
<7>[ 1186.857696] hub 1-0:1.0: power on to power good time: 2ms
<7>[ 1186.857757] hub 1-0:1.0: local power source is good
<7>[ 1186.857757] hub 1-0:1.0: enabling power on all ports
<7>[ 1186.955078] hub 1-0:1.0: port 1: status 0101 change 0001
<7>[ 1187.056640] hub 1-0:1.0: state 7 ports 1 chg 0002 evt 0000
<7>[ 1187.056701] hub 1-0:1.0: port 1, status 0101, change 0000, 12 Mb/s
<6>[ 1187.251861] usb 1-1: new high speed USB device using dwc_otg and address 2
<7>[ 1187.472076] usb 1-1: default language 0x0409
<7>[ 1187.473388] usb 1-1: udev 2, busnum 1, minor = 1
<6>[ 1187.473419] usb 1-1: New USB device found, idVendor=8564, idProduct=1000
<6>[ 1187.473449] usb 1-1: New USB device strings: Mfr=1, Product=2, SerialNumber=3
<6>[ 1187.473480] usb 1-1: Product: Mass Storage Device
<6>[ 1187.473510] usb 1-1: Manufacturer: JetFlash
<6>[ 1187.473541] usb 1-1: SerialNumber: 5550N3UJAP62NU0R
<7>[ 1187.475555] usb 1-1: usb_probe_device
<7>[ 1187.475616] usb 1-1: configuration #1 chosen from 1 choice
<7>[ 1187.475830] usb 1-1: adding 1-1:1.0 (config #1, interface 0)
<7>[ 1187.477294] libusual 1-1:1.0: usb_probe_interface
<7>[ 1187.477355] libusual 1-1:1.0: usb_probe_interface - got id
<7>[ 1187.477447] usb-storage 1-1:1.0: usb_probe_interface
<7>[ 1187.477478] usb-storage 1-1:1.0: usb_probe_interface - got id
<6>[ 1187.477874] scsi3 : usb-storage 1-1:1.0
<7>[ 1187.482482] hub 1-0:1.0: state 7 ports 1 chg 0000 evt 0002
<7>[ 1187.482543] hub 1-0:1.0: port 1 enable change, status 00000503
<5>[ 1188.822296] scsi 3:0:0:0: Direct-Access JetFlash Transcend 8GB 1100 PQ: 0 ANSI: 0 CCS
<5>[ 1188.829681] sd 3:0:0:0: [sda] 15826944 512-byte logical blocks: (8.10 GB/7.54 GiB)
<5>[ 1188.830444] sd 3:0:0:0: [sda] Write Protect is off
<7>[ 1188.830505] sd 3:0:0:0: [sda] Mode Sense: 43 00 00 00
<3>[ 1188.830535] sd 3:0:0:0: [sda] Assuming drive cache: write through
<3>[ 1188.834808] sd 3:0:0:0: [sda] Assuming drive cache: write through
<6>[ 1188.834930] sda:
<3>[ 1188.841308] sd 3:0:0:0: [sda] Assuming drive cache: write through
<5>[ 1188.841339] sd 3:0:0:0: [sda] Attached SCSI removable disk
<4>[ 1192.035308] wl_iw_event: dev=eth0 event=20
/mnt/sdcard/mod # mount -r -t vfat /dev/block/sda /sdcard/media
/mnt/sdcard/mod # busybox df /sdcard/media
Filesystem 1K-blocks Used Available Use% Mounted on
/dev/block/sda 7898024 3906992 3991032 49% /mnt/sdcard/media
/mnt/sdcard/mod # ls /sdcard/media
ubninit
EFI
ubnkern
boot
casper
dists
install
isolinux
pics
pool
preseed
ubnpathl.txt
README.diskdefines
autorun.inf
md5sum.txt
wubi.exe
ubnfilel.txt
ldlinux.sys
syslinux.cfg
menu.c32
HBCD
/mnt/sdcard/mod #
/mnt/sdcard/media # lsmod
dwc_otg 125404 0 - Live 0xbf262000
sd_mod 26464 1 - Live 0xbf136000
usb_storage 34867 1 - Live 0xbf257000
usb_libusual 9839 1 usb_storage, Live 0xbf09c000
scsi_mod 124018 2 sd_mod,usb_storage, Live 0xbf228000
lm 2661 1 dwc_otg, Live 0xbf055000
usbcore 146797 3 dwc_otg,usb_storage,usb_libusual, Live 0xbf18d000
bcm4330 275939 0 - Live 0xbf147000
brcm_headsetsw 7603 0 - Live 0xbf13f000
gememalloc 661 0 - Live 0xbf133000 (P)
h6270enc 3469 0 - Live 0xbf12d000
hx170dec 5362 0 - Live 0xbf125000
sec_param 8980 0 - Live 0xbf11d000
j4fs 70799 1 - Live 0xbf105000 (P)
rfs_fat 247794 3 - Live 0xbf0bd000 (P)
rfs_glue 79181 1 rfs_fat, Live 0xbf0a1000 (P)
fsr_stl 255325 5 j4fs, Live 0xbf05b000 (P)
fsr 338224 2 sec_param,fsr_stl, Live 0xbf000000 (P)
/mnt/sdcard/media #
Currently I'm working on PMU driver to allow the phone to charge while connected via Y-cable.
Once ready I'll post some binaries and patches against stock kernel sources so you will be able to try it out, or take over if you would like to.
After that I could probably try to restore gadget functionality as loadable module and probably add something to switch device/host mode automatically based on cable type.
But personally I don't need gadget functionality at all, so I'm not sure when I will be able to find enough spare time to do this.
PS.
Please don't move it to development section just yet - I won't be able to post there (need 10 posts or something like that).
Here we go, as promised.
I've decided to test whether SGY is powerful enough to play movies from external USB HDD drive with truecrypted ext4 fs.
It turns out to be good enough.
Code:
/mnt/sdcard/mod # busybox cp ../dmsetup /dev
/mnt/sdcard/mod # busybox chmod +x /dev/dmsetup
/mnt/sdcard/mod # insmod usbcore.ko
/mnt/sdcard/mod # insmod lm.ko
/mnt/sdcard/mod # insmod dwc_otg.ko
/mnt/sdcard/mod # insmod scsi_mod.ko
/mnt/sdcard/mod # insmod sd_mod.ko
/mnt/sdcard/mod # insmod usb-libusual.ko
/mnt/sdcard/mod # insmod usb-storage.ko
/mnt/sdcard/mod # busybox ls -l /dev/block/sda
brw------- 1 0 0 8, 0 Jul 20 04:18 /dev/block/sda
/mnt/sdcard/mod # insmod gf128mul.ko
/mnt/sdcard/mod # insmod xts.ko
/mnt/sdcard/mod # /dev/dmsetup create tc0 --table "0 976772656 crypt aes-xts-pla
in64 XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX 256 8:0 256"
/mnt/sdcard/mod # insmod jbd2.ko
/mnt/sdcard/mod # insmod ext4.ko
/mnt/sdcard/mod # mount -t ext4 -r /dev/mapper/tc0 /sdcard/media
/mnt/sdcard/mod # am start -n com.mxtech.videoplayer.ad/.ActivityScreen -d /mnt/sdcard/media/Videos/cartoon/mlp/My.Little.Pony.Friendship.Is.Magic.S01E26.The.Best.Night.Ever.480p.WEB-DL.DD5.1.h.264-ETP.mkv
Attachment contains kernel update which you can flash using recovery mode.
Basically, it is an ordinary Samsung kernel without any USB code and with some minor hw-related tweaks.
modules.tar.bz2 contains all modules from snippet above plus HID and FTDI support.
It's huge because of ext4.ko. You probably won't need it.
dwc_otg.ko is USB driver itself. At the moment it has host-only functionality.
It means - it is a very bad idea to connect your phone to PC with this module loaded. Be warned.
Also, I'm using Y-cabe with 30k resistor between D+ and D-. This way MUIC thinks that phone is plugged into 1A charger.
At the same time, D+/D- is routed USB PHY. This hack is statically linked into the kernel.
So - don't try to charge your phone with this kernel with chargers with D+/D- shortened.
Oh, yeah. ID pin connection is irrelevant at the moment, so you can hack any micro-usb cable in the middle - no complex soldering required.
Disclaimer: You may brick your phone. Do it on your own risk. This post is developer-oriented.
If you do not fully understand what's written above - better don't do it at all.
uunicorn said:
Here we go, as promised.
I've decided to test whether SGY is powerful enough to play movies from external USB HDD drive with truecrypted ext4 fs.
It turns out to be good enough.
Code:
/mnt/sdcard/mod # busybox cp ../dmsetup /dev
/mnt/sdcard/mod # busybox chmod +x /dev/dmsetup
/mnt/sdcard/mod # insmod usbcore.ko
/mnt/sdcard/mod # insmod lm.ko
/mnt/sdcard/mod # insmod dwc_otg.ko
/mnt/sdcard/mod # insmod scsi_mod.ko
/mnt/sdcard/mod # insmod sd_mod.ko
/mnt/sdcard/mod # insmod usb-libusual.ko
/mnt/sdcard/mod # insmod usb-storage.ko
/mnt/sdcard/mod # busybox ls -l /dev/block/sda
brw------- 1 0 0 8, 0 Jul 20 04:18 /dev/block/sda
/mnt/sdcard/mod # insmod gf128mul.ko
/mnt/sdcard/mod # insmod xts.ko
/mnt/sdcard/mod # /dev/dmsetup create tc0 --table "0 976772656 crypt aes-xts-pla
in64 XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX 256 8:0 256"
/mnt/sdcard/mod # insmod jbd2.ko
/mnt/sdcard/mod # insmod ext4.ko
/mnt/sdcard/mod # mount -t ext4 -r /dev/mapper/tc0 /sdcard/media
/mnt/sdcard/mod # am start -n com.mxtech.videoplayer.ad/.ActivityScreen -d /mnt/sdcard/media/Videos/cartoon/mlp/My.Little.Pony.Friendship.Is.Magic.S01E26.The.Best.Night.Ever.480p.WEB-DL.DD5.1.h.264-ETP.mkv
Attachment contains kernel update which you can flash using recovery mode.
Basically, it is an ordinary Samsung kernel without any USB code and with some minor hw-related tweaks.
modules.tar.bz2 contains all modules from snippet above plus HID and FTDI support.
It's huge because of ext4.ko. You probably won't need it.
dwc_otg.ko is USB driver itself. At the moment it has host-only functionality.
It means - it is a very bad idea to connect your phone to PC with this module loaded. Be warned.
Also, I'm using Y-cabe with 30k resistor between D+ and D-. This way MUIC thinks that phone is plugged into 1A charger.
At the same time, D+/D- is routed USB PHY. This hack is statically linked into the kernel.
So - don't try to charge your phone with this kernel with chargers with D+/D- shortened.
Oh, yeah. ID pin connection is irrelevant at the moment, so you can hack any micro-usb cable in the middle - no complex soldering required.
Disclaimer: You may brick your phone. Do it on your own risk. This post is developer-oriented.
If you do not fully understand what's written above - better don't do it at all.
Click to expand...
Click to collapse
Dude what you are doing is great but the problem is its too much difficult for average member to understand what all this means....you should give a better discription about what you have done with the kernel and what is the advantage of that and how it will work.....uploading kernels like this could lead to someone bricking their phone....
According to me you should team up with someone and make a thread in dev section.....
Sent from my Dishwasher
-HellRaiser- said:
you should give a better discription about what you have done with the kernel and what is the advantage of that and how it will work.....
Click to expand...
Click to collapse
Yes, I know, but there is nothing user-friendly to show yet.
There is not much profit for an avg phone user until automatic switching between host and device mode is added.
After this feature is implemented in usb driver, it could be incorporated into someones custom kernel distribution and OTG will just work out of the box.
At the moment it is just an ongoing development, which I would like to share with other developers, because I do not think I have enough motivation to finish everything by myself.
-HellRaiser- said:
uploading kernels like this could lead to someone bricking their phone....
Click to expand...
Click to collapse
Hence all the warnings.
-HellRaiser- said:
According to me you should team up with someone and make a thread in dev section.....
Click to expand...
Click to collapse
There is a thread about OTG on SGY already in development section, but I can't even bump it - no rights to post there.
Someone could probably post a link to this thread there, I suppose.
uunicorn said:
There is a thread about OTG on SGY already in development section, but I can't even bump it - no rights to post there.
Someone could probably post a link to this thread there, I suppose.
Click to expand...
Click to collapse
I bumped it for you :good:
btw, I think you need 10 posts in order to start a thread or post on development section.
anyways, keep up the good work!!!
uunicorn said:
Yes, I know, but there is nothing user-friendly to show yet.
There is not much profit for an avg phone user until automatic switching between host and device mode is added.
After this feature is implemented in usb driver, it could be incorporated into someones custom kernel distribution and OTG will just work out of the box.
At the moment it is just an ongoing development, which I would like to share with other developers, because I do not think I have enough motivation to finish everything by myself.
Hence all the warnings.
There is a thread about OTG on SGY already in development section, but I can't even bump it - no rights to post there.
Someone could probably post a link to this thread there, I suppose.
Click to expand...
Click to collapse
uunicorn said:
Yes, I know, but there is nothing user-friendly to show yet.
There is not much profit for an avg phone user until automatic switching between host and device mode is added.
After this feature is implemented in usb driver, it could be incorporated into someones custom kernel distribution and OTG will just work out of the box.
At the moment it is just an ongoing development, which I would like to share with other developers, because I do not think I have enough motivation to finish everything by myself.
Hence all the warnings.
There is a thread about OTG on SGY already in development section, but I can't even bump it - no rights to post there.
Someone could probably post a link to this thread there, I suppose.
Click to expand...
Click to collapse
what need to doing ?
flash your kernel
for what is modules and dmsetup ?
I can't understand ?
uunicorn said:
Here we go, as promised.
I've decided to test whether SGY is powerful enough to play movies from external USB HDD drive with truecrypted ext4 fs.
It turns out to be good enough.
Code:
/mnt/sdcard/mod # busybox cp ../dmsetup /dev
/mnt/sdcard/mod # busybox chmod +x /dev/dmsetup
/mnt/sdcard/mod # insmod usbcore.ko
/mnt/sdcard/mod # insmod lm.ko
/mnt/sdcard/mod # insmod dwc_otg.ko
/mnt/sdcard/mod # insmod scsi_mod.ko
/mnt/sdcard/mod # insmod sd_mod.ko
/mnt/sdcard/mod # insmod usb-libusual.ko
/mnt/sdcard/mod # insmod usb-storage.ko
/mnt/sdcard/mod # busybox ls -l /dev/block/sda
brw------- 1 0 0 8, 0 Jul 20 04:18 /dev/block/sda
/mnt/sdcard/mod # insmod gf128mul.ko
/mnt/sdcard/mod # insmod xts.ko
/mnt/sdcard/mod # /dev/dmsetup create tc0 --table "0 976772656 crypt aes-xts-pla
in64 XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX 256 8:0 256"
/mnt/sdcard/mod # insmod jbd2.ko
/mnt/sdcard/mod # insmod ext4.ko
/mnt/sdcard/mod # mount -t ext4 -r /dev/mapper/tc0 /sdcard/media
/mnt/sdcard/mod # am start -n com.mxtech.videoplayer.ad/.ActivityScreen -d /mnt/sdcard/media/Videos/cartoon/mlp/My.Little.Pony.Friendship.Is.Magic.S01E26.The.Best.Night.Ever.480p.WEB-DL.DD5.1.h.264-ETP.mkv
Attachment contains kernel update which you can flash using recovery mode.
Basically, it is an ordinary Samsung kernel without any USB code and with some minor hw-related tweaks.
modules.tar.bz2 contains all modules from snippet above plus HID and FTDI support.
It's huge because of ext4.ko. You probably won't need it.
dwc_otg.ko is USB driver itself. At the moment it has host-only functionality.
It means - it is a very bad idea to connect your phone to PC with this module loaded. Be warned.
Also, I'm using Y-cabe with 30k resistor between D+ and D-. This way MUIC thinks that phone is plugged into 1A charger.
At the same time, D+/D- is routed USB PHY. This hack is statically linked into the kernel.
So - don't try to charge your phone with this kernel with chargers with D+/D- shortened.
Oh, yeah. ID pin connection is irrelevant at the moment, so you can hack any micro-usb cable in the middle - no complex soldering required.
Disclaimer: You may brick your phone. Do it on your own risk. This post is developer-oriented.
If you do not fully understand what's written above - better don't do it at all.
Click to expand...
Click to collapse
A Biggg Thanks to you Master....
I am the one who tried for the same target before... But not yet successed..
I am very much curious to see the source or the patch for this.
I hope you will move your code to github or somethig...
Again .. You have Done a Great Job..
Sorry for reviving old post first of all guys, I'm a Galaxy Ace-I user (cooperve, not cooper) and I've found out that our phone in particular, just like the SGY has a BCM21552 (physically that's what the chip reads not 21553 lol), which also supports OTG as I've stated in this thread, I really think it'd be interesting if S5830i and SGY developers could join forces to make this go through
Whats the status of last development of OTG support?
SOLVED: defective otg cable
useful tools:
any android terminal emulator plus root
dmesg
lsusb
2nd otg cable
I cannot seem to get otg working with any of the various USB storage devices I have.
Any ideas on how I can debug this?
If it were normal Linux it would be easy enough, however it's not.
Any suggestions would be appreciated.
Running AOKP buttered right now but sentinel stock and paranoid all do not work either. So it might be a cable or the tablet even.
Hmm not sure but I wonder if the kernel with this rom is missing the fat kernel modules
Thanks
Sent from my Nexus 10 using xda premium
tazmeister said:
I cannot seem to get otg working with any of the various USB storage devices I have.
Click to expand...
Click to collapse
I had the same problems with rooted Stock and various custom ROMs. Try installing "USB OTG Helper [root]" then retest. This app helped get me RW access to OTG storage from SD cards to powered HDD drives in various formats (ext2,3,4 and NTFS).
It worked so well I paid for the donate version just to support the developer. It even has exFAT support although I never tried it. Personally I think it should be part of standard Android.
Have you checked "lsusb" or "udevadm" in terminal if the device recognizes anything?
KasF said:
Have you checked "lsusb" or "udevadm" in terminal if the device recognizes anything?
Click to expand...
Click to collapse
Thanks more commands to jog my memory and try
I installed a console app and can check dmesg and other stuff.
Lsmod and related seem to be mostly obsolete.
Dmesg shows something is seen but the system keeps failing after this.
I see the same messages regardless of which device I try, powered and unpowered.
<6>[ 6171.135316] s5p-ehci s5p-ehci: S5P EHCI Host Controller
<6>[ 6171.136622] s5p-ehci s5p-ehci: new USB bus registered, assigned bus number 1
<6>[ 6171.137173] s5p-ehci s5p-ehci: irq 103, io mem 0x12110000
<6>[ 6171.145031] s5p-ehci s5p-ehci: USB 0.0 started, EHCI 1.00
<6>[ 6171.145636] usb usb1: New USB device found, idVendor=1d6b, idProduct=0002
<6>[ 6171.146150] usb usb1: New USB device strings: Mfr=3, Product=2, SerialNumber=1
<6>[ 6171.146439] usb usb1: Product: S5P EHCI Host Controller
<6>[ 6171.146925] usb usb1: Manufacturer: Linux 3.4.5-gb44fbc0 ehci_hcd
<6>[ 6171.147202] usb usb1: SerialNumber: s5p-ehci
<6>[ 6171.150947] hub 1-0:1.0: USB hub found
<6>[ 6171.151298] hub 1-0:1.0: 3 ports detected
<6>[ 6171.153725] exynos-ohci exynos-ohci: EXYNOS OHCI Host Controller
<6>[ 6171.156783] exynos-ohci exynos-ohci: new USB bus registered, assigned bus number 2
<6>[ 6171.157031] exynos-ohci exynos-ohci: irq 103, io mem 0x12120000
<6>[ 6171.214330] usb usb2: New USB device found, idVendor=1d6b, idProduct=0001
<6>[ 6171.214522] usb usb2: New USB device strings: Mfr=3, Product=2, SerialNumber=1
<6>[ 6171.214630] usb usb2: Product: EXYNOS OHCI Host Controller
<6>[ 6171.214813] usb usb2: Manufacturer: Linux 3.4.5-gb44fbc0 ohci_hcd
<6>[ 6171.215052] usb usb2: SerialNumber: exynos-ohci
<6>[ 6171.220588] hub 2-0:1.0: USB hub found
<6>[ 6171.220946] hub 2-0:1.0: 3 ports detected
<6>[ 6171.225049] manta_otg manta_otg_work: b_idle -> a_host
<6>[ 6171.705161] usb 2-1: new full-speed USB device number 2 using exynos-ohci
<3>[ 6171.835100] usb 2-1: device descriptor read/64, error -62
<3>[ 6172.070121] usb 2-1: device descriptor read/64, error -62
<6>[ 6172.300093] usb 2-1: new full-speed USB device number 3 using exynos-ohci
<3>[ 6172.430142] usb 2-1: device descriptor read/64, error -62
<3>[ 6172.665754] usb 2-1: device descriptor read/64, error -62
<6>[ 6172.895140] usb 2-1: new full-speed USB device number 4 using exynos-ohci
<3>[ 6173.305043] usb 2-1: device not accepting address 4, error -62
<6>[ 6173.430032] usb 2-1: new full-speed USB device number 5 using exynos-ohci
<3>[ 6173.840038] usb 2-1: device not accepting address 5, error -62
<3>[ 6173.840490] hub 2-0:1.0: unable to enumerate USB device on port 1
Sent from my Nexus 10 using xda premium
None of the org helper apps have helped.
I don't think the org cable is bad though maybe this detail means otherwise
one device starts charging itself of the musb port when i plug it in.
Sent from my Nexus 10 using xda premium
KasF said:
Have you checked "lsusb" or "udevadm" in terminal if the device recognizes anything?
Click to expand...
Click to collapse
udevadm doesn't seem to be present but lsusb lists 2 devices when I have something plugged in.
Bus 001 Device 001: ID 1d6b:0002
Bus 002 Device 001: ID 1d6b:0001
Sent from my Nexus 10 using xda premium
ok problem solved.
The final debugging step was to "get another otg cable and test to make sure the first otg cable isn't defective"
Since I only owned a single OTG cable at the start, it was sort of difficult to perform this ultimate debug step.
But once I tried a different cable, almost everything worked just fine.
Only my 64gb micro-sd cards gave me a little trouble, but they all work now.