Hi all,
I tried to get this all to work about 6 months ago and could not.
I have now tried again from scratch after the recent mail regarding this
subject. I have everything installed and a brand spanking new 2.4.22
kernel. I find it very difficult to connect and when I do connect it will
not maintain the connection. I know these devices go into sleep mode, but
talking to other people on this list, they come out of it. This is not
possible for me. Frequently, after I try to wiggle the mouse in sleep
mode, bthid start to take up 100% of the CPU. I find it unlikely that this
is the bthid code itself.
It is more than possible that this is a hardware problem with my
mouse etc but I would really like to know this before trying to exchange
them. I do have 1 more piece of information with the new kernel. As you
can see at the end of the message below there is a message which may help.
I am willing and able to provide whatever information will help solve this
problem. I really do not want to lose against MS ;)
Any help is greatly appreciated,
Gareth
PS thanks for those slides Marcel, they were informative
Nov 5 17:17:48 somme bthid[2266]: Bluetooth HID daemon by Peter Klausler
<[email protected]>
Nov 5 17:17:48 somme bthid[2266]: socket 5 bound to BD 00:00:00:00:00:00
PSM 17
Nov 5 17:17:48 somme bthid[2266]: socket 6 bound to BD 00:00:00:00:00:00
PSM 19
Nov 5 17:19:26 somme bthid[2273]: Bluetooth HID daemon by Peter Klausler
<[email protected]>
Nov 5 17:19:26 somme bthid[2274]: attempting connection to BD
00:50:F2:7F:AA:A5
Nov 5 17:19:26 somme bthid[2274]: L2CAP socket 5 connected to BD
00:50:F2:7F:AA:A5 PSM 17
Nov 5 17:19:26 somme bthid[2274]: L2CAP socket 6 connected to BD
00:50:F2:7F:AA:A5 PSM 19
Nov 5 17:19:26 somme bthid[2274]: Serving Bluetooth HID at BD
00:50:F2:7F:AA:A5
Nov 5 17:19:26 somme bthid[2274]: L2CAP socket 7 connected to BD
00:50:F2:7F:AA:A5 PSM 1
Nov 5 17:19:26 somme bthid[2274]: Wireless IntelliMouse Explorer for
Bluetooth
Nov 5 17:19:26 somme bthid[2274]: made by Microsoft
Nov 5 17:19:26 somme bthid[2274]: device version 1.0
Nov 5 17:31:45 somme bthid[2274]: EOF on Interrupt channel
Nov 5 17:37:08 somme kernel: hci_acldata_packet: hci0 ACL packet for
unknown connection handle 41
Nov 5 17:41:24 somme last message repeated 166 times
hciconfig -a
[gareth@somme gareth]$ /sbin/hciconfig -a
hci0: Type: USB
BD Address: 00:50:F2:7E:29:D0 ACL MTU: 192:8 SCO MTU: 64:8
UP RUNNING PSCAN ISCAN
RX bytes:17308 acl:1125 sco:0 events:123 errors:0
TX bytes:1225 acl:39 sco:0 commands:27 errors:0
Features: 0xff 0xff 0x0f 0x00
Packet type: DM1 DM3 DM5 DH1 DH3 DH5 HV1 HV2 HV3
Link policy: HOLD SNIFF PARK
Link mode: SLAVE ACCEPT
Name: 'BlueZ (0)'
Class: 0x000100
Service Classes: Unspecified
Device Class: Computer, Uncategorized
HCI Ver: 1.1 (0x1) HCI Rev: 0x1f9 LMP Ver: 1.1 (0x1) LMP Subver:
0x1f9
Manufacturer: Cambridge Silicon Radio (10)
-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive? Does it
help you create better code? SHARE THE LOVE, and help us help
YOU! Click Here: http://sourceforge.net/donate/
_______________________________________________
Bluez-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/bluez-users
Hi Michal,
> maybe Marcel should fetch and release from CVS his
>
> hid-0.0.0.1-verysoon-notstable-onlyforinsane-prealfa version :)
>
> Then will much more people test it and we'll get more feedback from users...
>
> Marcel what do you think about it? :)
I won't do this, because the code depends on the new Bluetooth library
and this is not released.
Regards
Marcel
-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive? Does it
help you create better code? SHARE THE LOVE, and help us help
YOU! Click Here: http://sourceforge.net/donate/
_______________________________________________
Bluez-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/bluez-users
Hi Gareth,
maybe Marcel should fetch and release from CVS his=20
hid-0.0.0.1-verysoon-notstable-onlyforinsane-prealfa version :)
Then will much more people test it and we'll get more feedback from users...
Marcel what do you think about it? :)
Michal
Dne st=F8eda 26 listopad 2003 23:13 jste napsal(a):
> Hi,
>
> On Wed, 26 Nov 2003, Marcel Holtmann wrote:
> > Hi Gareth,
> >
> > > I know the ping thing works, however I don't believe that it
> > > should be necessary. I think there is either something wrong with the
> > > hardware, bthid or bluez. If it is bthid or bluez then I would like to
> > > try and fix it so we can all benefit :)
> >
> > look at my bthid code and see how I have handled it. This was working
> > for me. You need a sleep() and the correct order of closing the L2CAP
> > channels. Otherwise the ACL connection would not be terminated.
>
> I modified my bthid code after you posted the other week. It still does
> not work. Ah well, this just gives me more incentive to work on the
> kernel HID parser :) I will speak again after I have read your posting on
> devel and had a go at implementing it.
>
>
> Cheers,
>
> Gareth
-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive? Does it
help you create better code? SHARE THE LOVE, and help us help
YOU! Click Here: http://sourceforge.net/donate/
_______________________________________________
Bluez-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/bluez-users
Hi,
On Wed, 26 Nov 2003, Marcel Holtmann wrote:
> Hi Gareth,
>
> > I know the ping thing works, however I don't believe that it
> > should be necessary. I think there is either something wrong with the
> > hardware, bthid or bluez. If it is bthid or bluez then I would like to try
> > and fix it so we can all benefit :)
>
> look at my bthid code and see how I have handled it. This was working
> for me. You need a sleep() and the correct order of closing the L2CAP
> channels. Otherwise the ACL connection would not be terminated.
I modified my bthid code after you posted the other week. It still does
not work. Ah well, this just gives me more incentive to work on the
kernel HID parser :) I will speak again after I have read your posting on
devel and had a go at implementing it.
Cheers,
Gareth
Hi Gareth,
> I know the ping thing works, however I don't believe that it
> should be necessary. I think there is either something wrong with the
> hardware, bthid or bluez. If it is bthid or bluez then I would like to try
> and fix it so we can all benifit :)
look at my bthid code and see how I have handled it. This was working
for me. You need a sleep() and the correct order of closing the L2CAP
channels. Otherwise the ACL connection would not be terminated.
Regards
Marcel
-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive? Does it
help you create better code? SHARE THE LOVE, and help us help
YOU! Click Here: http://sourceforge.net/donate/
_______________________________________________
Bluez-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/bluez-users
Hi Gareth,
> thanks for the response. Is there someway I can simulate this with
> hcitool (or some other tool)? If I hcitool cc to this what would you
> expect the behaveiour to be? The mouse will go to sleep in 12 mins, but if
> I move it and to a hcitool con after that should the device appear? Or
> would I have to connect again? I seem to recall that there was some way of
> showing the incoming input from the connection. If I did this without
> bthid, then would this establish if the problem was with bthid or bluez?
you can't use hcitool, because you have to establish the two HID L2CAP
connections for PSM 17 and 19.
Regards
Marcel
-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive? Does it
help you create better code? SHARE THE LOVE, and help us help
YOU! Click Here: http://sourceforge.net/donate/
_______________________________________________
Bluez-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/bluez-users
Hey Michal,
I know the ping thing works, however I don't believe that it
should be necessary. I think there is either something wrong with the
hardware, bthid or bluez. If it is bthid or bluez then I would like to try
and fix it so we can all benifit :)
Gareth
Hey,
thanks for the response. Is there someway I can simulate this with
hcitool (or some other tool)? If I hcitool cc to this what would you
expect the behaveiour to be? The mouse will go to sleep in 12 mins, but if
I move it and to a hcitool con after that should the device appear? Or
would I have to connect again? I seem to recall that there was some way of
showing the incoming input from the connection. If I did this without
bthid, then would this establish if the problem was with bthid or bluez?
Once again, thanks for your help,
Gareth
Hi,
this switches works for me perfectly
/usr/local/bin/bthid -m -s -p 600 -l "$MYS" "$KLAVESNICE"
Michal
Dne st=F8eda 26 listopad 2003 22:37 Marcel Holtmann napsal(a):
> Hi Gareth,
>
> > I was really hoping it was not going to work. Unfortunately it does :( I
> > would really appreciate any help in sorting this out. I have rebuilt and
> > installed the bluez modules against my patched 2.4.22 kernel and all is
> > the same. The output from the 0.8 bthid is:
> >
> > Nov 26 21:05:34 zebedee bthid[1747]: Serving Bluetooth HID at BD
> > 00:50:F2:E2:EB:E6
> > Nov 26 21:06:26 zebedee bthid[1753]: Bluetooth HID daemon by Peter
> > Klausler <[email protected]>
> > Nov 26 21:06:26 zebedee bthid[1754]: attempting connection to BD
> > 00:50:F2:E2:EB:E6
> > Nov 26 21:06:27 zebedee bthid[1754]: L2CAP socket 5 connected to BD
> > 00:50:F2:E2:EB:E6 PSM 1
> > Nov 26 21:06:28 zebedee bthid[1754]: L2CAP socket 6 connected to BD
> > 00:50:F2:E2:EB:E6 PSM 17
> > Nov 26 21:06:28 zebedee bthid[1754]: L2CAP socket 7 connected to BD
> > 00:50:F2:E2:EB:E6 PSM 19
> > Nov 26 21:06:28 zebedee bthid[1754]: Serving Bluetooth HID at BD
> > 00:50:F2:E2:EB:E6
> > Nov 26 21:06:28 zebedee bthid[1754]: device: Wireless IntelliMouse
> > Explorer for Bluetooth
> > Nov 26 21:06:28 zebedee bthid[1754]: made by Microsoft
> > Nov 26 21:06:28 zebedee bthid[1754]: device version 1.0
> > Nov 26 21:06:28 zebedee bthid[1754]: is virtually cabled
> > Nov 26 21:06:28 zebedee bthid[1754]: will initiate reconnections
> > Nov 26 21:06:28 zebedee bthid[1754]: is battery-powered
> > Nov 26 21:07:15 zebedee bthid[1754]: vendor HID usage 0xff00fe01 value
> > 0x1 Nov 26 21:07:15 zebedee bthid[1754]: adequate battery on Wireless
> > IntelliMouse Explorer for Bluetooth
> > Nov 26 21:17:40 zebedee bthid[1754]: EOF on Interrupt channel
> >
> >
> > After the EOF on interrupt channel it will not reconnect. If there is a
> > bthid -l listening then 100% of the CPU is consumed. I have cvs checked
> > out and am happy to put debug in etc to find out the problem and
> > hopefuly get this working.
>
> this is the implementation from Peter and I can't help you with it. I
> have only used it once.
>
> Regards
>
> Marcel
>
>
>
>
> -------------------------------------------------------
> This SF.net email is sponsored by: SF.net Giveback Program.
> Does SourceForge.net help you be more productive? Does it
> help you create better code? SHARE THE LOVE, and help us help
> YOU! Click Here: http://sourceforge.net/donate/
> _______________________________________________
> Bluez-users mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/bluez-users
Hi Gareth,
> I was really hoping it was not going to work. Unfortunately it does :( I
> would really appreciate any help in sorting this out. I have rebuilt and
> installed the bluez modules against my patched 2.4.22 kernel and all is
> the same. The output from the 0.8 bthid is:
>
> Nov 26 21:05:34 zebedee bthid[1747]: Serving Bluetooth HID at BD
> 00:50:F2:E2:EB:E6
> Nov 26 21:06:26 zebedee bthid[1753]: Bluetooth HID daemon by Peter
> Klausler <[email protected]>
> Nov 26 21:06:26 zebedee bthid[1754]: attempting connection to BD
> 00:50:F2:E2:EB:E6
> Nov 26 21:06:27 zebedee bthid[1754]: L2CAP socket 5 connected to BD
> 00:50:F2:E2:EB:E6 PSM 1
> Nov 26 21:06:28 zebedee bthid[1754]: L2CAP socket 6 connected to BD
> 00:50:F2:E2:EB:E6 PSM 17
> Nov 26 21:06:28 zebedee bthid[1754]: L2CAP socket 7 connected to BD
> 00:50:F2:E2:EB:E6 PSM 19
> Nov 26 21:06:28 zebedee bthid[1754]: Serving Bluetooth HID at BD
> 00:50:F2:E2:EB:E6
> Nov 26 21:06:28 zebedee bthid[1754]: device: Wireless IntelliMouse
> Explorer for Bluetooth
> Nov 26 21:06:28 zebedee bthid[1754]: made by Microsoft
> Nov 26 21:06:28 zebedee bthid[1754]: device version 1.0
> Nov 26 21:06:28 zebedee bthid[1754]: is virtually cabled
> Nov 26 21:06:28 zebedee bthid[1754]: will initiate reconnections
> Nov 26 21:06:28 zebedee bthid[1754]: is battery-powered
> Nov 26 21:07:15 zebedee bthid[1754]: vendor HID usage 0xff00fe01 value 0x1
> Nov 26 21:07:15 zebedee bthid[1754]: adequate battery on Wireless
> IntelliMouse Explorer for Bluetooth
> Nov 26 21:17:40 zebedee bthid[1754]: EOF on Interrupt channel
>
>
> After the EOF on interrupt channel it will not reconnect. If there is a
> bthid -l listening then 100% of the CPU is consumed. I have cvs checked
> out and am happy to put debug in etc to find out the problem and
> hopefuly get this working.
this is the implementation from Peter and I can't help you with it. I
have only used it once.
Regards
Marcel
-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive? Does it
help you create better code? SHARE THE LOVE, and help us help
YOU! Click Here: http://sourceforge.net/donate/
_______________________________________________
Bluez-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/bluez-users
Hey again,
> > please check that it works with the Windows XP Bluetooth stack.
>
> will do, have to wait until the week as I don't own a copy.
I was really hoping it was not going to work. Unfortunately it does :( I
would really appreciate any help in sorting this out. I have rebuilt and
installed the bluez modules against my patched 2.4.22 kernel and all is
the same. The output from the 0.8 bthid is:
Nov 26 21:05:34 zebedee bthid[1747]: Serving Bluetooth HID at BD
00:50:F2:E2:EB:E6
Nov 26 21:06:26 zebedee bthid[1753]: Bluetooth HID daemon by Peter
Klausler <[email protected]>
Nov 26 21:06:26 zebedee bthid[1754]: attempting connection to BD
00:50:F2:E2:EB:E6
Nov 26 21:06:27 zebedee bthid[1754]: L2CAP socket 5 connected to BD
00:50:F2:E2:EB:E6 PSM 1
Nov 26 21:06:28 zebedee bthid[1754]: L2CAP socket 6 connected to BD
00:50:F2:E2:EB:E6 PSM 17
Nov 26 21:06:28 zebedee bthid[1754]: L2CAP socket 7 connected to BD
00:50:F2:E2:EB:E6 PSM 19
Nov 26 21:06:28 zebedee bthid[1754]: Serving Bluetooth HID at BD
00:50:F2:E2:EB:E6
Nov 26 21:06:28 zebedee bthid[1754]: device: Wireless IntelliMouse
Explorer for Bluetooth
Nov 26 21:06:28 zebedee bthid[1754]: made by Microsoft
Nov 26 21:06:28 zebedee bthid[1754]: device version 1.0
Nov 26 21:06:28 zebedee bthid[1754]: is virtually cabled
Nov 26 21:06:28 zebedee bthid[1754]: will initiate reconnections
Nov 26 21:06:28 zebedee bthid[1754]: is battery-powered
Nov 26 21:07:15 zebedee bthid[1754]: vendor HID usage 0xff00fe01 value 0x1
Nov 26 21:07:15 zebedee bthid[1754]: adequate battery on Wireless
IntelliMouse Explorer for Bluetooth
Nov 26 21:17:40 zebedee bthid[1754]: EOF on Interrupt channel
After the EOF on interrupt channel it will not reconnect. If there is a
bthid -l listening then 100% of the CPU is consumed. I have cvs checked
out and am happy to put debug in etc to find out the problem and
hopefuly get this working.
Thanks for any help,
Gareth
Hi Gareth,
> > Do anyone have interest in working on the HID kernel driver. My idea of
> > the interface for the bthid daemon is very easy. You will see it if you
> > look at the bthid source code. On request I can post some comments about
> > it.
>
> I am happy to put some time in. If you post some comments then I will try
> and take a look later next week or on the weekend.
I will prepare a separate posting on the developer mailing list.
Regards
Marcel
-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive? Does it
help you create better code? SHARE THE LOVE, and help us help
YOU! Click Here: http://sourceforge.net/donate/
_______________________________________________
Bluez-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/bluez-users
Hi,
> please check that it works with the Windows XP Bluetooth stack.
will do, have to wait until the week as I don't own a copy.
> And here is the problem. This code is not available at the moment. I
> have only one machine with a hacked USB HID driver and never found the
> time to get this clean. You also have to install the new Bluetooth
> library to make my bthid working, because the old SDP code can't get the
> complete HID descriptor from the HID device.
I understand.
> Do anyone have interest in working on the HID kernel driver. My idea of
> the interface for the bthid daemon is very easy. You will see it if you
> look at the bthid source code. On request I can post some comments about
> it.
I am happy to put some time in. If you post some comments then I will try
and take a look later next week or on the weekend.
Once again, thanks for your help Marcel
Gareth
Hi Gareth,
> I thought I would give a quick update. M$ agreed within seconds
> that there was a problem with the keyboard and mouse and promptly sent a
> new one out! I have just recieved them and they do not work either :(
> Exactly the same problem.
please check that it works with the Windows XP Bluetooth stack.
> I thought that I would give it a try with Marcels bthid. In his
> previous mail he states that it relies on a modifcation to the USB HID
> kernel module. Where is this in CVS? Also, do I have to do anything else
> (like undo the things I did to make the original bthid work?) before this
> should work?
And here is the problem. This code is not available at the moment. I
have only one machine with a hacked USB HID driver and never found the
time to get this clean. You also have to install the new Bluetooth
library to make my bthid working, because the old SDP code can't get the
complete HID descriptor from the HID device.
Do anyone have interest in working on the HID kernel driver. My idea of
the interface for the bthid daemon is very easy. You will see it if you
look at the bthid source code. On request I can post some comments about
it.
Regards
Marcel
-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive? Does it
help you create better code? SHARE THE LOVE, and help us help
YOU! Click Here: http://sourceforge.net/donate/
_______________________________________________
Bluez-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/bluez-users
Hey all,
I thought I would give a quick update. M$ agreed within seconds
that there was a problem with the keyboard and mouse and promptly sent a
new one out! I have just recieved them and they do not work either :(
Exactly the same problem.
I thought that I would give it a try with Marcels bthid. In his
previous mail he states that it relies on a modifcation to the USB HID
kernel module. Where is this in CVS? Also, do I have to do anything else
(like undo the things I did to make the original bthid work?) before this
should work?
Thanks,
Gareth
-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive? Does it
help you create better code? SHARE THE LOVE, and help us help
YOU! Click Here: http://sourceforge.net/donate/
_______________________________________________
Bluez-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/bluez-users
Hi,
> it looks like you lost the ACL link, but the mouse didn't know about
> that and keeps sending you reports on that channel.
>
> Do you get this working under Windows XP. If not, they are broken.
Arrrrggh. They do not work properly under XP! Finally found someone with
XP. They connect and then a few minutes later they start to lock up. Off
to talk with MS to try and change them.
Thanks for all your help,
Gareth
Hi Marcel,
> it looks like you lost the ACL link, but the mouse didn't know about
> that and keeps sending you reports on that channel.
>
> Do you get this working under Windows XP. If not, they are broken.
I will fish out the drivers and find someone with XP to test them. I will
hopefully do this next week and will give a shout back about success then.
However, I assume that the 100% CPU usage is not a good thing? I will
happily build this stuff up in debug mode and try and start it under gdb
to get a backtrace if this will be helpful for bluez development. Tell me
if there is anything I can do.
Thanks for your help,
Gareth
Hi Gareth,
> I tried to get this all to work about 6 months ago and could not.
> I have now tried again from scratch after the recent mail regarding this
> subject. I have everything installed and a brand spanking new 2.4.22
> kernel. I find it very difficult to connect and when I do connect it will
> not maintain the connection. I know these devices go into sleep mode, but
> talking to other people on this list, they come out of it. This is not
> possible for me. Frequently, after I try to wiggle the mouse in sleep
> mode, bthid start to take up 100% of the CPU. I find it unlikely that this
> is the bthid code itself.
> It is more than possible that this is a hardware problem with my
> mouse etc but I would really like to know this before trying to exchange
> them. I do have 1 more piece of information with the new kernel. As you
> can see at the end of the message below there is a message which may help.
> I am willing and able to provide whatever information will help solve this
> problem. I really do not want to lose against MS ;)
it looks like you lost the ACL link, but the mouse didn't know about
that and keeps sending you reports on that channel.
Do you get this working under Windows XP. If not, they are broken.
> PS thanks for those slides Marcel, they were informative
You are welcome :)
Regards
Marcel
-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive? Does it
help you create better code? SHARE THE LOVE, and help us help
YOU! Click Here: http://sourceforge.net/donate/
_______________________________________________
Bluez-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/bluez-users