anyone running android sdk to talk to a real phone on fedora 14?

[Tom Horsley]

I can't figure out if this is a kernel problem or what
(though I lean towards kernel since it works one way
immediately after powering up the system and a different
way after that).

If I plug in my samsung intercept phone right after powering off then powering back up
and booting fedora 14, I see this show up in /var/log/messages:

Apr 23 22:31:52 zooty kernel: [  242.131064] usb 2-3: new high speed USB device using
ehci_hcd and address 4
Apr 23 22:31:52 zooty kernel: [  242.354901] usb 2-3: New USB device found, idVendor=04e8,
idProduct=681c
Apr 23 22:31:52 zooty kernel: [  242.354905] usb 2-3: New USB device strings: Mfr=1,
Product=2, SerialNumber=3
Apr 23 22:31:52 zooty kernel: [  242.374157] usb 2-3: can't set config #1, error -71

And lsusb shows this:

Bus 002 Device 004: ID 04e8:681c Samsung Electronics Co., Ltd Galaxy Portal/Spica Android
Phone

which isn't really right as it is an Intercept, not a Galaxy.

Now I unplug the phone, and leaving the system up, plug the phone back in again:

unplug:

Apr 23 22:38:48 zooty kernel: [  658.159285] usb 2-3: USB disconnect, address 4

replug:

Apr 23 22:39:26 zooty kernel: [  696.061086] usb 2-3: new high speed USB device using
ehci_hcd and address 5
Apr 23 22:39:26 zooty kernel: [  696.257079] usb 2-3: device descriptor read/64, error
-71
Apr 23 22:39:26 zooty kernel: [  696.559082] usb 2-3: device descriptor read/64, error
-71
Apr 23 22:39:26 zooty kernel: [  696.762074] usb 2-3: new high speed USB device using
ehci_hcd and address 6
Apr 23 22:39:27 zooty kernel: [  696.957078] usb 2-3: device descriptor read/64, error
-71
Apr 23 22:39:27 zooty kernel: [  697.249077] usb 2-3: device descriptor read/64, error
-71
Apr 23 22:39:27 zooty kernel: [  697.452060] usb 2-3: new high speed USB device using
ehci_hcd and address 7
Apr 23 22:39:38 zooty kernel: [  707.854076] usb 2-3: device not accepting address 7,
error -110
Apr 23 22:39:38 zooty kernel: [  707.956083] usb 2-3: new high speed USB device using
ehci_hcd and address 8
Apr 23 22:39:38 zooty kernel: [  707.956083] usb 2-3: new high speed USB device using
ehci_hcd and address 8
Apr 23 22:39:48 zooty kernel: [  718.358072] usb 2-3: device not accepting address 8,
error -110
Apr 23 22:39:48 zooty kernel: [  718.358083] hub 2-0:1.0: unable to enumerate USB device
on port 3
Apr 23 22:39:48 zooty kernel: [  718.599078] usb 10-1: new full speed USB device using
uhci_hcd and address 2
Apr 23 22:40:03 zooty kernel: [  733.704071] usb 10-1: device descriptor read/64, error
-110
Apr 23 22:40:19 zooty kernel: [  748.911074] usb 10-1: device descriptor read/64, error
-110
Apr 23 22:40:19 zooty kernel: [  749.114059] usb 10-1: new full speed USB device using
uhci_hcd and address 3
Apr 23 22:40:20 zooty kernel: [  750.534082] hub 10-0:1.0: unable to enumerate USB device
on port 1

It just goes completely wacko after the first time I plug it in.
No matter how many times I plug in in after that, I'll see the same
thing.

On the other hand, an ancient crappy system I have with only
usb 1 ports seems to work fine. It detects a communications
device and creates a /dev/ttyACM0 (or something like that)
as well as a storage device for the SD card.

Does this sound like something I should report as a kernel
bug?

[Bill Davidsen]

Tim wrote:
>  Tim:
> >> Hmm, I wonder if that means it really does draw more current than it
> >> should from a USB port
>
>  Tom:
> > I'm pretty sure it really is drawing too much power. I had 3 different
> > systems that all showed the same symptoms, and lots of other Samsung
> > Intercept users have complained of similar problems.
>
>  Sounds like you're now relying on having a hub that will supply more
>  current than it should.  Cross your fingers that hub builders don't
>  start adhering to the USB specifications...  ;-\
>
> > If there was some way to just make it talk USB without also insisting
> > on trying to charge the battery when plugged in, it might be OK :-).
>
>  Take out the battery?
>  
My thought would be to be sure the battery was fully charged, so you provide 
only run power.

-- 
Bill Davidsen <davidsen(a)tmr.com&gt;
   "We have more to fear from the bungling of the incompetent than from
the machinations of the wicked."  - from Slashdot

[Tom Horsley]

On Fri, 29 Apr 2011 14:24:16 +0100
Alan Cox wrote:

>  > > some way to just make it talk USB without also insisting on
> trying to
>  > > charge the battery when plugged in, it might be OK :-).  
>  > 
>  > You could buy a cheap USB cable (monoprice.com?) and snip the power wires...  
>  
>  That isn't going to work, you still need vbus and ground. Powered hubs
>  are not exactly expensive 
And as further evidence you need the vbus, look at the wacky stuff I
do with relays to toggle the USB power on and off to trigger
the cameras in my book scanner :-).

http://home.comcast.net/~tomhorsley/hardware/scanner/scanner.html

[Tim]

Tim:
> > Sounds like you're now relying on having a hub that will
> supply more
> > current than it should.  Cross your fingers that hub builders don't
> > start adhering to the USB specifications...  ;-\ 
Alan Cox:
>  Or have a PC which is underpowered for some reason - that happens
> too. 
Or one that has many USB ports on the same internal hub, and between all
the peripherals you have attached, you've over-reached the limit.

I don't like USB-powered disc drives, for those reasons.  They need just
all the power that's possible for a port to supply, and anything else
taking power pushes it over the limit.  Some drives come with a USB
Y-lead, to take power from two ports, but that's not going to help in
some cases, and it bothers me that I don't know whether they'd dumbly
wired the power lines directly together, or have safely joined them
(through something like diodes).

-- 
[tim@localhost ~]$ uname -r
2.6.27.25-78.2.56.fc9.i686

Don't send private replies to my address, the mailbox is ignored.  I
read messages from the public lists.

[Alan Cox]

>  Sounds like you're now relying on having a hub that will supply
> more
>  current than it should.  Cross your fingers that hub builders don't
>  start adhering to the USB specifications...  ;-\ 
Or have a PC which is underpowered for some reason - that happens too.

[Tim]

Tim:
> > Hmm, I wonder if that means it really does draw more current than
> it
> > should from a USB port 
Tom:
>  I'm pretty sure it really is drawing too much power. I had 3
> different
>  systems that all showed the same symptoms, and lots of other Samsung
>  Intercept users have complained of similar problems. 
Sounds like you're now relying on having a hub that will supply more
current than it should.  Cross your fingers that hub builders don't
start adhering to the USB specifications...  ;-\

>  If there was some way to just make it talk USB without also
> insisting
>  on trying to charge the battery when plugged in, it might be OK :-). 
Take out the battery?

-- 
[tim@localhost ~]$ uname -r
2.6.27.25-78.2.56.fc9.i686

Don't send private replies to my address, the mailbox is ignored.  I
read messages from the public lists.

[Tom Horsley]

On Fri, 29 Apr 2011 19:53:45 +0930
Tim wrote:

>  > If I connect it via an external powered hub, it works perfectly.
>  
>  
>  Hmm, I wonder if that means it really does draw more current than it
>  should from a USB port 
I'm pretty sure it really is drawing too much power. I had 3 different
systems that all showed the same symptoms, and lots of other Samsung
Intercept users have complained of similar problems. If there was
some way to just make it talk USB without also insisting on trying to
charge the battery when plugged in, it might be OK :-).

[Tim]

On Thu, 2011-04-28 at 12:57 -0400, Tom Horsley wrote:
>  So, I finally tracked it down: What these sorts of messages
>  really mean are: "Ouch! This USB device is drawing too much
>  power!".
>   
>  If I connect it via an external powered hub, it works perfectly. 
Hmm, I wonder if that means it really does draw more current than it
should from a USB port, or whether the current hub can't supply enough
power?  If the port is on a hub with several things demanding power,
that can be a problem (whether that be internal or external hubs).

-- 
[tim@localhost ~]$ uname -r
2.6.27.25-78.2.56.fc9.i686

Don't send private replies to my address, the mailbox is ignored.  I
read messages from the public lists.

[Richard Shaw]

On Fri, Apr 29, 2011 at 6:13 AM, Tom Horsley <horsley1953(a)gmail.com&gt; wrote:
>  On Fri, 29 Apr 2011 19:53:45 +0930
>  Tim wrote:
>
> > > If I connect it via an external powered hub, it works perfectly.
> >
> > Hmm, I wonder if that means it really does draw more current than it
> > should from a USB port
>
>  I'm pretty sure it really is drawing too much power. I had 3 different
>  systems that all showed the same symptoms, and lots of other Samsung
>  Intercept users have complained of similar problems. If there was
>  some way to just make it talk USB without also insisting on trying to
>  charge the battery when plugged in, it might be OK :-). 
You could buy a cheap USB cable (monoprice.com?) and snip the power wires...

Richard

[Richard Shaw]

On Thu, Apr 28, 2011 at 11:57 AM, Tom Horsley <horsley1953(a)gmail.com&gt; wrote:
>  On Sat, 23 Apr 2011 22:57:19 -0400
>  Tom Horsley wrote:
>
> > Apr 23 22:39:26 zooty kernel: [  696.061086] usb 2-3: new high speed USB device using
> ehci_hcd and address 5
> > Apr 23 22:39:26 zooty kernel: [  696.257079] usb 2-3: device descriptor read/64,
> error -71
> > Apr 23 22:39:26 zooty kernel: [  696.559082] usb 2-3: device descriptor read/64,
> error -71
> > Apr 23 22:39:26 zooty kernel: [  696.762074] usb 2-3: new high speed USB device using
> ehci_hcd and address 6
> > Apr 23 22:39:27 zooty kernel: [  696.957078] usb 2-3: device descriptor read/64,
> error -71
> > Apr 23 22:39:27 zooty kernel: [  697.249077] usb 2-3: device descriptor read/64,
> error -71
> > Apr 23 22:39:27 zooty kernel: [  697.452060] usb 2-3: new high speed USB device using
> ehci_hcd and address 7
> > Apr 23 22:39:38 zooty kernel: [  707.854076] usb 2-3: device not accepting address 7,
> error -110
>
>  So, I finally tracked it down: What these sorts of messages
>  really mean are: "Ouch! This USB device is drawing too much
>  power!".
>
>  If I connect it via an external powered hub, it works perfectly. 
Good to know! I'm sorry I didn't see your first message or I would
have responded that I was able to use the android SDK on F14 with my
wifes HTC MyTouch 3g slide. It only had android 2.1 so I put cyanogen
6 on it (android 2.2).

I can't connect connect to my LG Optimus T but I think that's because
it needs a special USB driver that they only provide for windows (and
maybe Mac?)

Richard

[Tom Horsley]

On Sat, 23 Apr 2011 22:57:19 -0400
Tom Horsley wrote:

>  Apr 23 22:39:26 zooty kernel: [  696.061086] usb 2-3: new high speed
> USB device using ehci_hcd and address 5
>  Apr 23 22:39:26 zooty kernel: [  696.257079] usb 2-3: device descriptor read/64, error
> -71
>  Apr 23 22:39:26 zooty kernel: [  696.559082] usb 2-3: device descriptor read/64, error
> -71
>  Apr 23 22:39:26 zooty kernel: [  696.762074] usb 2-3: new high speed USB device using
> ehci_hcd and address 6
>  Apr 23 22:39:27 zooty kernel: [  696.957078] usb 2-3: device descriptor read/64, error
> -71
>  Apr 23 22:39:27 zooty kernel: [  697.249077] usb 2-3: device descriptor read/64, error
> -71
>  Apr 23 22:39:27 zooty kernel: [  697.452060] usb 2-3: new high speed USB device using
> ehci_hcd and address 7
>  Apr 23 22:39:38 zooty kernel: [  707.854076] usb 2-3: device not accepting address 7,
> error -110 
So, I finally tracked it down: What these sorts of messages
really mean are: "Ouch! This USB device is drawing too much
power!".

If I connect it via an external powered hub, it works perfectly.

[Tom Horsley]

On Sat, 23 Apr 2011 22:57:19 -0400
Tom Horsley wrote:

>  On the other hand, an ancient crappy system I have with only
>  usb 1 ports seems to work fine. It detects a communications
>  device and creates a /dev/ttyACM0 (or something like that)
>  as well as a storage device for the SD card. 
After a zillion google searches, I finally found this:

http://ubuntuforums.org/showthread.php?t=1631552

which sure makes it seem like a kernel problem since
disabling the high speed usb driver allows it to work.

I wonder if I can find a usb 1.1 hub somewhere and
make it work by going through that? :-).

[Alan Cox]

>  > some way to just make it talk USB without also insisting on
> trying to
>  > charge the battery when plugged in, it might be OK :-).
>  
>  You could buy a cheap USB cable (monoprice.com?) and snip the power wires... 
That isn't going to work, you still need vbus and ground. Powered hubs
are not exactly expensive