2008-10-21 00:24:26

by Andrew Morton

[permalink] [raw]
Subject: Re: [Bugme-new] [Bug 11779] New: No way to disable bluetooth


(switched to email. Please respond via emailed reply-to-all, not via the
bugzilla web interface).

(tap, tap - is this thing turned on?)

On Sat, 18 Oct 2008 02:55:03 -0700 (PDT)
[email protected] wrote:

> http://bugzilla.kernel.org/show_bug.cgi?id=11779
>
> Summary: No way to disable bluetooth
> Product: Drivers
> Version: 2.5
> KernelVersion: 2.6.27
> Platform: All
> OS/Version: Linux
> Tree: Mainline
> Status: NEW
> Severity: normal
> Priority: P1
> Component: Bluetooth
> AssignedTo: [email protected]
> ReportedBy: [email protected]
>
>
> Distribution: Ubuntu intrepid
> Hardware Environment: Notebook Itautec N8320
> Software Environment: Ubuntu userland
> Problem Description:
>
> (This mail was sent to the [email protected] on the 8th of
> October no answer was received).
>
> Dear developers,
>
> I got this past july a notebook that has what seems to be a strange way
> of dealing with wireless devices: there is a button (which I suspect is
> called rfkill) that, when pressed, enables *both* bluetooth and wifi,
> with the respective leds being lit.
>
> I don't have any bluetooth device. Saving energy here is paramount as
> the battery life is quite short.
>
> According to the manual, under Windows Vista (which is what came with
> the box, but which I never used), this button is supposed to cycle
> between 4 states:
>
> * both disabled;
> * bluetooth enabled & wifi enabled;
> * bluetooth disabled & wifi enabled;
> * bluetooth enabled & wifi disabled.
>
> Unfortunately, under Linux (tested with Ubuntu's 8.10 pre-releases,
> which includes a kernel based on 2.6.27-rc*, if I understand it
> correctly), it is all or nothing: both disabled or both enabled.
>
> If I launch powertop, it recommends to disable bluetooth with
> "hciconfig hci down; rmmod hci_usb", but the led of bluetooth is still
> lit and I fear that the device is still drawing power from the battery.
> (Actually, it seems that the modules for bluetooth go renamed also).
>
> Is there any way to get it to disable completely the device?
>
> Just as extra information, this is an Itautec Note N8320 notebook, with
> a Core 2 Duo T7250 CPU, ICH8 chipset and Intel's 3945abg wifi and a
> Realtek GigE driven by r8169 module (which I still have not had the
> opportunity to use).
>
> If any further information is desired, please let me know. I can post
> the dmesg logs, lspci output, lsusb, as I am really willing to get this
> situation settled (and possibly working as on Windows as it was reported
> by the support people from the manufacturer).
>
> Again, if any further information is needed, please let me know. I can
> recompile the kernel to use a vanilla kernel, if needed.
>



2008-10-21 08:01:40

by David Sainty

[permalink] [raw]
Subject: Re: [Bugme-new] [Bug 11779] New: No way to disable bluetooth

Rog?rio Brito wrote:

> I may test it latter, but I'm scared. And this is what is preventing me
> from using this notebook (as well as bug
> http://bugzilla.kernel.org/show_bug.cgi?id=11717).
>
> For this latter one I even provided a *video* of me pressing the power
> button and the machine not waking up from suspend. :-( I can't think of
> anything else that I can provice, but I'm willing to help here.
>
>

It sounds like you have much worse issues with it than whether the
bluetooth adapter is drawing idle current - which should be dwarfed by
more heavyweight sinks like the WiFi transceiver.

Sorry, I don't have any suggestions - though you could try asking the
manufacturer for support.


2008-10-21 07:00:56

by Rogério Brito

[permalink] [raw]
Subject: Re: [Bugme-new] [Bug 11779] New: No way to disable bluetooth

Hi, Andrew, David and others.

On Oct 21 2008, David Sainty wrote:
> Andrew Morton wrote:
> >>http://bugzilla.kernel.org/show_bug.cgi?id=11779
> >>
> >>According to the manual, under Windows Vista (which is what came with
> >>the box, but which I never used), this button is supposed to cycle
> >>between 4 states:
> >>
> >>* both disabled;
> >>* bluetooth enabled & wifi enabled;
> >>* bluetooth disabled & wifi enabled;
> >>* bluetooth enabled & wifi disabled.
> >>
> >>Unfortunately, under Linux (tested with Ubuntu's 8.10 pre-releases,
> >>which includes a kernel based on 2.6.27-rc*, if I understand it
> >>correctly), it is all or nothing: both disabled or both enabled.
> >>
> >>If I launch powertop, it recommends to disable bluetooth with
> >>"hciconfig hci down; rmmod hci_usb", but the led of bluetooth is still
> >>lit and I fear that the device is still drawing power from the battery.
> >>(Actually, it seems that the modules for bluetooth go renamed also).
> >>
> >>
> If the LED is on, it's obviously drawing at least enough power to light
> the LED :)

Indeed. :-)

> >>Is there any way to get it to disable completely the device?
>
> This is a guess, but probably turning it on and off has nothing to do
> with the bluetooth support, it's probably intended to be powered on and
> off through ACPI or something - it's probably very specific to that
> particular laptop.

How can we chase this issue? I am willing to cooperate as much as possible,
with as much data that I can to get things straight. It's got an Intel ICH8
chipset here, with a Realtek rt8169-driven wired network. It's battery is
quite short under Linux. :-(

(And it seems that the gnome power indicator is confused when reading data
about the battery, since it sometimes shows the battery as completely
drained when it is actually charged and plugged on the AC outlet).

I really, really don't want to install Windows Vista for use, but I may
install it for debugging/reverse engineering/simulating the behaviour.

> However, "hciconfig hci down" should at least minimise any drain the
> device is causing.

I may test it latter, but I'm scared. And this is what is preventing me
from using this notebook (as well as bug
http://bugzilla.kernel.org/show_bug.cgi?id=11717).

For this latter one I even provided a *video* of me pressing the power
button and the machine not waking up from suspend. :-( I can't think of
anything else that I can provice, but I'm willing to help here.


Thanks for any help,

--
Rog?rio Brito : [email protected]{mackenzie,ime.usp}.br : GPG key 1024D/7C2CAEB8
http://www.ime.usp.br/~rbrito : http://meusite.mackenzie.com.br/rbrito
Projects: algorithms.berlios.de : lame.sf.net : vrms.alioth.debian.org

2008-10-21 02:07:59

by David Sainty

[permalink] [raw]
Subject: Re: [Bugme-new] [Bug 11779] New: No way to disable bluetooth

Andrew Morton wrote:
>> http://bugzilla.kernel.org/show_bug.cgi?id=11779
>>
>> According to the manual, under Windows Vista (which is what came with
>> the box, but which I never used), this button is supposed to cycle
>> between 4 states:
>>
>> * both disabled;
>> * bluetooth enabled & wifi enabled;
>> * bluetooth disabled & wifi enabled;
>> * bluetooth enabled & wifi disabled.
>>
>> Unfortunately, under Linux (tested with Ubuntu's 8.10 pre-releases,
>> which includes a kernel based on 2.6.27-rc*, if I understand it
>> correctly), it is all or nothing: both disabled or both enabled.
>>
>> If I launch powertop, it recommends to disable bluetooth with
>> "hciconfig hci down; rmmod hci_usb", but the led of bluetooth is still
>> lit and I fear that the device is still drawing power from the battery.
>> (Actually, it seems that the modules for bluetooth go renamed also).
>>
>>
If the LED is on, it's obviously drawing at least enough power to light
the LED :)

>> Is there any way to get it to disable completely the device?
>>
>>

This is a guess, but probably turning it on and off has nothing to do
with the bluetooth support, it's probably intended to be powered on and
off through ACPI or something - it's probably very specific to that
particular laptop.

However, "hciconfig hci down" should at least minimise any drain the
device is causing.

2008-11-06 19:31:31

by Rogério Brito

[permalink] [raw]
Subject: Re: [Bugme-new] [Bug 11779] New: No way to disable bluetooth

Hi, David.

I have not received any response from you in the last few weeks.

On Oct 21 2008, David Sainty wrote:
> Rog?rio Brito wrote:
>
> >I may test it latter, but I'm scared.

(...)

> >I can't think of
> >anything else that I can provice, but I'm willing to help here.
>
> It sounds like you have much worse issues with it than whether the
> bluetooth adapter is drawing idle current - which should be dwarfed by
> more heavyweight sinks like the WiFi transceiver.

I don't know how much wifi would use of power, but I guess, in my layman
way of seeing things, that it would, indeed, use more power.

But I would like to save any corners that I can here. And just correcting
the behaviour would make me extremely happy.

Again, I can provide any information needed. Just ask and I will do my
best (I guess that you can't get many bug reporters willing to do as many
tests and experiments as I do; and I don't disappear). :-)


Regards, Rog?rio Brito.

--
Rog?rio Brito : [email protected]{mackenzie,ime.usp}.br : GPG key 1024D/7C2CAEB8
http://www.ime.usp.br/~rbrito : http://meusite.mackenzie.com.br/rbrito
Projects: algorithms.berlios.de : lame.sf.net : vrms.alioth.debian.org