Active Topics
-
Install Fonts in N900 (67)
to Maemo 5 / Fremantle by panjgoori - 18 hrs, 47 mins ago -
Nokia Booklet 3G (RX-75) Drivers (60)
to Alternatives by Maemish - 5 days, 2 hrs ago - more...
| The Following 7 Users Say Thank You to SpeedEvil For This Useful Post: | ||
|
2010-05-16
, 22:50
|
|
Posts: 946 |
Thanked: 1,650 times |
Joined on Oct 2009
@ Germany
|
#872
|
FYI:
echo 5 > /sys/module/musb_hdrc/parameters/debug
enables USB debugging in dmesg!
it contains very interesting information incl. line numbers
echo 5 > /sys/module/musb_hdrc/parameters/debug
enables USB debugging in dmesg!
it contains very interesting information incl. line numbers
 |
|
2010-05-17
, 00:41
|
|
Posts: 2,355 |
Thanked: 5,249 times |
Joined on Jan 2009
@ Barcelona
|
#873
|
Using egoshin's idea, I got a DisplayLink device working:
This is the same USB monitor shown at the DisplayLink homepage, connected to a N900, in dual screen "clone" mode (more accurately x11vnc, vnc2dl and libdlo, all on the N900). vnc2dl seems to have a 1280x1024 forced resolution.
Basically the requirement of connecting the N900 to a host previously is the key. If you play your cards well you don't even need to patch the kernel (Y-cable, or fully self powered bug, etc.).
I initially though that this requirement came from a g_mass_storage bug that seemingly also happens on the beagleboard, but I was wrong.
Seemingly, DocScrutinizer is right: for some reason after repeating egoshin's steps the N900 seems incapable of closing any USB session, and thus after you unplug it from the host it still believes it's plugged into it, keeping everything powered on. This tricks the driver's OTG state machine into some weirdo state where something starts moving
.
Also, with this method you cannot unplug the hub from the N900, as per the above the N900 will keep on thinking it's still connected to it, thus causing mayhem on the kernel.
Last edited by javispedro; 2010-05-17 at 00:50.
This is the same USB monitor shown at the DisplayLink homepage, connected to a N900, in dual screen "clone" mode (more accurately x11vnc, vnc2dl and libdlo, all on the N900). vnc2dl seems to have a 1280x1024 forced resolution.
Basically the requirement of connecting the N900 to a host previously is the key. If you play your cards well you don't even need to patch the kernel (Y-cable, or fully self powered bug, etc.).
I initially though that this requirement came from a g_mass_storage bug that seemingly also happens on the beagleboard, but I was wrong.
Seemingly, DocScrutinizer is right: for some reason after repeating egoshin's steps the N900 seems incapable of closing any USB session, and thus after you unplug it from the host it still believes it's plugged into it, keeping everything powered on. This tricks the driver's OTG state machine into some weirdo state where something starts moving
.Also, with this method you cannot unplug the hub from the N900, as per the above the N900 will keep on thinking it's still connected to it, thus causing mayhem on the kernel.
Last edited by javispedro; 2010-05-17 at 00:50.
| The Following 26 Users Say Thank You to javispedro For This Useful Post: | ||
 |
|
2010-05-17
, 06:28
|
|
Moderator |
Posts: 7,109 |
Thanked: 8,820 times |
Joined on Oct 2007
@ Vancouver, BC, Canada
|
#874
|
apparently superheroes can be on multiple teams at the same time.
Keep up the good work guys!
Keep up the good work guys!
__________________
qole.org --- twitter --- Easy Debian wiki page
Please don't send me a private message, post to the appropriate thread.
Thank you all for your donations!
qole.org --- twitter --- Easy Debian wiki page
Please don't send me a private message, post to the appropriate thread.
Thank you all for your donations!
| The Following User Says Thank You to qole For This Useful Post: | ||
|
|
2010-05-17
, 17:24
|
|
Posts: 946 |
Thanked: 1,650 times |
Joined on Oct 2009
@ Germany
|
#875
|
I cannot replicate this. btw, it is /proc/driver/musb_hdrc
Originally Posted by egoshin
If you want repeat a test, some detailed explanation is here:
C. Connect to PC and answer "PC Suite" to mode question.
D. echo host >/sys/devices/platform/musb_hdrc/mode
You should see in /var/log/syslog a message "twl4030_set_host() after 4030 OTG_CTRL=0x26" - the '2' is a significant.
E. echo H >/proc/drivers/musb_hdrc
F. reattach USB cable from PC to self powered USB hub with memory stick or to external hard disk.
E. Look into /var/log/syslog - You can see a lot of messages and two are most valuable - "Initializing USB Mass Storage driver" and "
USB Mass Storage support registered". And in between you may see "scsi0 : SCSI emulation for USB Mass Storage devices".
| The Following 2 Users Say Thank You to titan For This Useful Post: | ||
|
|
2010-05-17
, 17:37
|
|
Posts: 2,355 |
Thanked: 5,249 times |
Joined on Jan 2009
@ Barcelona
|
#876
|
Originally Posted by qole
Hah, not really a superhero as I discarded all what I was doing to test egoshin's method, which is the first time I've ever seen it moving.
apparently superheroes can be on multiple teams at the same time.
Btw, you might need the patches after all -- seems I had some previous leftover on my N900 (be careful as some settings survive rebooting and reflashing, forcing you to remove the battery).
| The Following 3 Users Say Thank You to javispedro For This Useful Post: | ||
|
|
2010-05-17
, 17:52
|
|
Posts: 992 |
Thanked: 995 times |
Joined on Dec 2009
@ California
|
#877
|
Originally Posted by titan
Interesting. Sometime I can't do it too but I blamed it to some leftovers and repeat it after removing battery.
I cannot replicate this. btw, it is /proc/driver/musb_hdrc
But there is still a possibility that some process like BME or DSME intervenes and creates a problem.
BTW, did you see in log a message "twl4030_set_host() after 4030 OTG_CTRL=0x26"?
| The Following User Says Thank You to egoshin For This Useful Post: | ||
|
|
2010-05-17
, 17:59
|
|
Posts: 992 |
Thanked: 995 times |
Joined on Dec 2009
@ California
|
#878
|
Originally Posted by javispedro
To return back from host mode state to normal just do:
Using egoshin's idea, I got a DisplayLink device working:
This is the same USB monitor shown at the DisplayLink homepage, connected to a N900, in dual screen "clone" mode (more accurately x11vnc, vnc2dl and libdlo, all on the N900). vnc2dl seems to have a 1280x1024 forced resolution.
Basically the requirement of connecting the N900 to a host previously is the key. If you play your cards well you don't even need to patch the kernel (Y-cable, or fully self powered bug, etc.).
I initially though that this requirement came from a g_mass_storage bug that seemingly also happens on the beagleboard, but I was wrong.
Seemingly, DocScrutinizer is right: for some reason after repeating egoshin's steps the N900 seems incapable of closing any USB session, and thus after you unplug it from the host it still believes it's plugged into it, keeping everything powered on. This tricks the driver's OTG state machine into some weirdo state where something starts moving
Also, with this method you cannot unplug the hub from the N900, as per the above the N900 will keep on thinking it's still connected to it, thus causing mayhem on the kernel.
echo peripheral >/sys/devices/platform/musb_hdrc/mode
It will cancel VBUS and it immediately stops a session.
BUT! It doesn't do the right stuff about DISK - you should unmount it first, especially if it is not ISO9660 format but some Linux file system.
| The Following 2 Users Say Thank You to egoshin For This Useful Post: | ||
|
|
2010-05-17
, 18:03
|
|
Posts: 2,355 |
Thanked: 5,249 times |
Joined on Jan 2009
@ Barcelona
|
#879
|
Originally Posted by egoshin
I couldn't -- as I've noticed though my twl3040 was possibly in some random state -- I wasn't using your kernel.
To return back from host mode state to normal just do:
echo peripheral >/sys/devices/platform/musb_hdrc/mode
It will cancel VBUS and it immediately stops a session.
Last edited by javispedro; 2010-05-17 at 18:21.
|
|
2010-05-17
, 18:15
|
|
Posts: 75 |
Thanked: 78 times |
Joined on Jan 2010
@ Germany
|
#880
|
fyi, just found a comparison chart of USB PHY:
http://www.ebv.com/fileadmin/product...sceivers_1.pdf
and checked out the isp1704a because of its similarity to the 1707:
http://www.stericsson.com/product/222222.jsp
It is 100% pin-compatible to the chip in the N900-schematic (which only misses ID and PSW, in comparison to isp1704a, but ID should be present somehow, considering its an OTG chip…)
happy reading
http://www.ebv.com/fileadmin/product...sceivers_1.pdf
and checked out the isp1704a because of its similarity to the 1707:
http://www.stericsson.com/product/222222.jsp
It is 100% pin-compatible to the chip in the N900-schematic (which only misses ID and PSW, in comparison to isp1704a, but ID should be present somehow, considering its an OTG chip…)
happy reading
| The Following 3 Users Say Thank You to hcm For This Useful Post: | ||
 |
| Tags |
| awesomeness in the works, boulevard of broken deals, host, i am the dealbreaker, inspector gadget lies, mobidapter is a scam, nokia fanbois, otg, over 9000, usb, usbcontrol |
«
Previous Thread
|
Next Thread
»
|
All times are GMT. The time now is 10:52.
the current exposed by the bq27200 needs multiplied by (20/3.57)mA to get a current.
It's a raw value from the chip.
the '20' is arguable - but is at least more or less reasonable.
This is as the chip really reads out in units of 3.57uV, and the resistor the battery current is applied across is 20mR in the datasheet recommendation.