[Bluez-devel] Multiple Bluetooth Dongles

[Bluez-devel] Multiple Bluetooth Dongles

[Williams, Richard]

[-- Attachment #1.1: Type: text/plain, Size: 681 bytes --]

Hello,
 
I have a system that uses multiple Bluetooth dongles and works just
fine. I've had my system running with several USB hubs and up to the 16
maximum devices that Bluez supports. Now I have a changing requirement
from my customer to handle more than 16 Bluetooth devices. They want
support for up to 30 Bluetooth devices.
 
I changed the HCI_MAX_DEV in bluez_libs/include/hci.h from 16 to 30,
recompiled and installed bluez_libs and bluez_utils. No problem.
 
My program works fine with up to 16 Bluetooth devices, but the 17th
device is not recognized and sometimes the entire USB will no longer
work.
 
Suggestions??
 
Thank you. Best Regards.
 
Rich 

[-- Attachment #1.2: Type: text/html, Size: 1987 bytes --]

[-- Attachment #2: Type: text/plain, Size: 315 bytes --]

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/

[-- Attachment #3: Type: text/plain, Size: 164 bytes --]

_______________________________________________
Bluez-devel mailing list
Bluez-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bluez-devel

Re: [Bluez-devel] Multiple Bluetooth Dongles

[Jeffrey Cuenco]

[-- Attachment #1.1: Type: text/plain, Size: 1527 bytes --]

Hi,

We are doing something similar in that we are using multiple dongles as
well, and are also interested in doing something similar to what your
customer wants to do by increasing the # of dongles.

Out of curiosity, how are you using these dongles?

On 8/23/07, Williams, Richard <williams_richard_b@bah.com> wrote:
>
>  Hello,
>
> I have a system that uses multiple Bluetooth dongles and works just fine.
> I've had my system running with several USB hubs and up to the 16 maximum
> devices that Bluez supports. Now I have a changing requirement from my
> customer to handle more than 16 Bluetooth devices. They want support for up
> to 30 Bluetooth devices.
>
> I changed the HCI_MAX_DEV in bluez_libs/include/hci.h from 16 to
> 30, recompiled and installed bluez_libs and bluez_utils. No problem.
>
> My program works fine with up to 16 Bluetooth devices, but the 17th device
> is not recognized and sometimes the entire USB will no longer work.
>
> Suggestions??
>
> Thank you. Best Regards.
>
> *Rich *
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc.
> Still grepping through log files to find problems?  Stop.
> Now Search log events and configuration files using AJAX and a browser.
> Download your FREE copy of Splunk now >>  http://get.splunk.com/
> _______________________________________________
> Bluez-devel mailing list
> Bluez-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/bluez-devel
>
>


-- 
-Jeff

[-- Attachment #1.2: Type: text/html, Size: 3064 bytes --]

[-- Attachment #2: Type: text/plain, Size: 315 bytes --]

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/

[-- Attachment #3: Type: text/plain, Size: 164 bytes --]

_______________________________________________
Bluez-devel mailing list
Bluez-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bluez-devel

Re: [Bluez-devel] Multiple Bluetooth Dongles

[Marcel Holtmann]

Hi Jeffrey,

> We are doing something similar in that we are using multiple dongles
> as well, and are also interested in doing something similar to what
> your customer wants to do by increasing the # of dongles.

the kernel can handles as many dongles as you want and as long as you
have memory for the hci_dev structure.

For the userspace it is limited to 16, because of the ioctl() design.

Regards

Marcel



-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
_______________________________________________
Bluez-devel mailing list
Bluez-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bluez-devel

Re: [Bluez-devel] Multiple Bluetooth Dongles

[Marcel Holtmann]

Hi Richard,

> I have a system that uses multiple Bluetooth dongles and works just
> fine. I've had my system running with several USB hubs and up to the
> 16 maximum devices that Bluez supports. Now I have a changing
> requirement from my customer to handle more than 16 Bluetooth devices.
> They want support for up to 30 Bluetooth devices.
>  
> I changed the HCI_MAX_DEV in bluez_libs/include/hci.h from 16 to
> 30, recompiled and installed bluez_libs and bluez_utils. No problem.
>  
> My program works fine with up to 16 Bluetooth devices, but the 17th
> device is not recognized and sometimes the entire USB will no longer
> work.

we might broke something here since I haven't tested this in quite a
while. Mainly because I don't care so much anymore. Having these kind of
test setups are nasty.

The USB part can be an issue if you use the SCO support of the hci_usb
driver. Load the driver with isoc=0 to disable it. The ISOC frames for
the SCO support consume to much USB bandwidth.

Regards

Marcel



-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
_______________________________________________
Bluez-devel mailing list
Bluez-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bluez-devel

Re: [Bluez-devel] multithreaded name resolution

[Peter Wippich]

Hi Matthias,

On Fri, 24 Aug 2007, Matthias Becker wrote:

> Hi,
> 
> I tried to read the name of different devices parallel using multi
> threading, but without success. If there is one bluetooth device I works
> (as there is no real multi threading). If there are who devices, the
> same name is reported for both devices. In another test with 6 devices
> hci_read_remote_name returned something <0 so I did not get any name at
> all.

this won't work. Asking for the name of a device is some sort of 
connection setup, and most (all ?) Bt chips do not allow / can not handle 
parallel connection setups (even though parallel connections are 
possible). 

So don't try further, the problem is outside your code (and even BlueZ). 

Ciao, 

Peter 




-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
_______________________________________________
Bluez-devel mailing list
Bluez-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bluez-devel

[Bluez-devel] multithreaded name resolution

[Matthias Becker]

Hi,

I tried to read the name of different devices parallel using multi
threading, but without success. If there is one bluetooth device I works
(as there is no real multi threading). If there are who devices, the
same name is reported for both devices. In another test with 6 devices
hci_read_remote_name returned something <0 so I did not get any name at
all.

hcitool scan --flush
Scanning ...
        00:1A:89:6F:DE:EE       Nokia 5300
        00:19:63:9D:09:16       K800i

My multithreaded tool:

00:19:63:9D:09:16       K800i
00:1A:89:6F:DE:EE       K800i


When I don't start multiple threads but do it single threaded everything
works fine, and my code should be absolutly threadsafe.

Is this a bug in BlueZ or was it never supposed to work parallel?

When I run hcidump -V  I see that the right names are recived:
> HCI Event: Remote Name Req Complete (0x07) plen 255
    status 0x00 bdaddr 00:1A:89:6F:DE:EE name 'Nokia 5300'
> HCI Event: Remote Name Req Complete (0x07) plen 255
    status 0x00 bdaddr 00:19:63:9D:09:16 name 'K800i'


-- Matthias


-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
_______________________________________________
Bluez-devel mailing list
Bluez-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bluez-devel

Re: [Bluez-devel] Multiple Bluetooth Dongles

[Williams, Richard]

Hello Marcel,

In my case, I am not using the SCO feature. I haven't gotten that far in
my development. 
The steps I followed were:
- boot my PC
- modified the HCI_MAX_DEV to 30
- ran make & make install
- reboot ( probably not necessary, but what the heck ??)
- plug in 16 Bluetooth USB dongles. All appear to work OK. For each of
these, my Fedora Core 6 system pops up a small dialog box with "BlueZ
(X) Switched device into connected mode"
- plug in USB flash device. Works OK. 
- plug in USB mouse. Works OK.
- plug in Bluetooth dongle 17. No obvious response from the system.
After a few seconds, the USB mouse no longer works. The keyboard no
longer works. A power down restart is required.

I realize that the test setup is tedious, but maybe you can point me
towards what to look for ??

Thanks for all your help. Best Regards,

Rich


-----Original Message-----
From: bluez-devel-bounces@lists.sourceforge.net
[mailto:bluez-devel-bounces@lists.sourceforge.net] On Behalf Of Marcel
Holtmann
Sent: Friday, August 24, 2007 8:15 AM
To: BlueZ development
Subject: Re: [Bluez-devel] Multiple Bluetooth Dongles

Hi Richard,

> I have a system that uses multiple Bluetooth dongles and works just 
> fine. I've had my system running with several USB hubs and up to the
> 16 maximum devices that Bluez supports. Now I have a changing 
> requirement from my customer to handle more than 16 Bluetooth devices.
> They want support for up to 30 Bluetooth devices.
>  
> I changed the HCI_MAX_DEV in bluez_libs/include/hci.h from 16 to 30, 
> recompiled and installed bluez_libs and bluez_utils. No problem.
>  
> My program works fine with up to 16 Bluetooth devices, but the 17th 
> device is not recognized and sometimes the entire USB will no longer 
> work.

we might broke something here since I haven't tested this in quite a
while. Mainly because I don't care so much anymore. Having these kind of
test setups are nasty.

The USB part can be an issue if you use the SCO support of the hci_usb
driver. Load the driver with isoc=0 to disable it. The ISOC frames for
the SCO support consume to much USB bandwidth.

Regards

Marcel



------------------------------------------------------------------------
-
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
_______________________________________________
Bluez-devel mailing list
Bluez-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bluez-devel

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
_______________________________________________
Bluez-devel mailing list
Bluez-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bluez-devel

Re: [Bluez-devel] Multiple Bluetooth Dongles

[Marcel Holtmann]

Hi Richard,

> In my case, I am not using the SCO feature. I haven't gotten that far in
> my development. 
> The steps I followed were:
> - boot my PC
> - modified the HCI_MAX_DEV to 30
> - ran make & make install
> - reboot ( probably not necessary, but what the heck ??)
> - plug in 16 Bluetooth USB dongles. All appear to work OK. For each of
> these, my Fedora Core 6 system pops up a small dialog box with "BlueZ
> (X) Switched device into connected mode"
> - plug in USB flash device. Works OK. 
> - plug in USB mouse. Works OK.
> - plug in Bluetooth dongle 17. No obvious response from the system.
> After a few seconds, the USB mouse no longer works. The keyboard no
> longer works. A power down restart is required.
> 
> I realize that the test setup is tedious, but maybe you can point me
> towards what to look for ??

load the hci_usb with isoc=0 and see how that works out. If not, then
that is an USB subsystem issue. Make sure you use 2.6.23-rc3.

Regards

Marcel



-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
_______________________________________________
Bluez-devel mailing list
Bluez-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bluez-devel

Re: [Bluez-devel] Multiple Bluetooth Dongles

[Williams, Richard]

Hello Marcel,

Thank you for the quick response and the suggestion.

With my existing 2.6.22.2-42.fc6 kernel, reloading the hci_usb with
isoc=0 does help somewhat. I can plug in more than 16 dongles and the
system remains stable. The mouse and keyboard remain usable. However,
the 17th USB device is not recognized.

I am reluctant to try the 2.6.23-rc3 kernel you suggest, since my
development shop is all standardized on the current FC release. I don't
want my machine to get out of sync with the rest of my team.

I think that I'll just tell my customer that he'll have to wait on more
than 16. After all, if the application can't accomplish what it wants
with 16 blueteeth, why do they need more ?

So, thanks for your help. I'm not going to waste any more time on this.

Regards,

Rich

e

-----Original Message-----
From: bluez-devel-bounces@lists.sourceforge.net
[mailto:bluez-devel-bounces@lists.sourceforge.net] On Behalf Of Marcel
Holtmann
Sent: Friday, August 24, 2007 10:26 AM
To: BlueZ development
Subject: Re: [Bluez-devel] Multiple Bluetooth Dongles

Hi Richard,

> In my case, I am not using the SCO feature. I haven't gotten that far 
> in my development.
> The steps I followed were:
> - boot my PC
> - modified the HCI_MAX_DEV to 30
> - ran make & make install
> - reboot ( probably not necessary, but what the heck ??)
> - plug in 16 Bluetooth USB dongles. All appear to work OK. For each of

> these, my Fedora Core 6 system pops up a small dialog box with "BlueZ
> (X) Switched device into connected mode"
> - plug in USB flash device. Works OK. 
> - plug in USB mouse. Works OK.
> - plug in Bluetooth dongle 17. No obvious response from the system.
> After a few seconds, the USB mouse no longer works. The keyboard no 
> longer works. A power down restart is required.
> 
> I realize that the test setup is tedious, but maybe you can point me 
> towards what to look for ??

load the hci_usb with isoc=0 and see how that works out. If not, then
that is an USB subsystem issue. Make sure you use 2.6.23-rc3.

Regards

Marcel



------------------------------------------------------------------------
-
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
_______________________________________________
Bluez-devel mailing list
Bluez-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bluez-devel

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
_______________________________________________
Bluez-devel mailing list
Bluez-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bluez-devel

[Bluez-devel] [PATCH]Re: multithreaded name resolution

[Andreas Gaufer]

[-- Attachment #1: Type: text/plain, Size: 1156 bytes --]

On Fri, 24 Aug 2007 14:14:53 +0200 (CEST)
Peter Wippich <pewi@gw-instruments.de> wrote:

> 
> Hi Matthias,
> 
> On Fri, 24 Aug 2007, Matthias Becker wrote:
> 
> > Hi,
> > 
> > I tried to read the name of different devices parallel using multi
> > threading, but without success. If there is one bluetooth device I works
> > (as there is no real multi threading). If there are who devices, the
> > same name is reported for both devices. In another test with 6 devices
> > hci_read_remote_name returned something <0 so I did not get any name at
> > all.
> 
> this won't work. Asking for the name of a device is some sort of 
> connection setup, and most (all ?) Bt chips do not allow / can not handle 
> parallel connection setups (even though parallel connections are 
> possible). 
> 
> So don't try further, the problem is outside your code (and even BlueZ). 
> 

That's not true, the reported behavior is produced by bluez-libs. I already
sent a patch to fix it and will attach that patch again to this mail.

Please let me know if that Patch fixes your problem Matthias.

Greetings

Andy

See also:
http://thread.gmane.org/gmane.linux.bluez.devel/12491

[-- Attachment #2: bluez-libs-3.19-EVT_REMOTE_NAME_REQ_COMPLETE.patch --]
[-- Type: application/octet-stream, Size: 797 bytes --]

diff -ur bluez-libs-3.19-ORG/src/hci.c bluez-libs-3.19/src/hci.c
--- bluez-libs-3.19-ORG/src/hci.c	2007-10-01 11:06:18.013479576 +0200
+++ bluez-libs-3.19/src/hci.c	2007-10-01 15:19:36.301988528 +0200
@@ -1010,6 +1010,8 @@
 	while (try--) {
 		evt_cmd_complete *cc;
 		evt_cmd_status   *cs;
+		evt_remote_name_req_complete *rn;
+		remote_name_req_cp *cp;
 
 		if (to) {
 			struct pollfd p;
@@ -1074,6 +1076,21 @@
 			memcpy(r->rparam, ptr, r->rlen);
 			goto done;
 
+		case EVT_REMOTE_NAME_REQ_COMPLETE:
+
+			if (hdr->evt != r->event)
+				break;
+
+			r->rlen = MIN(len, r->rlen);
+			memcpy(r->rparam, ptr, r->rlen);
+
+			rn = r->rparam;
+			cp = r->cparam;
+
+			if (bacmp(&rn->bdaddr,&cp->bdaddr) == 0)
+				goto done;
+			continue;
+
 		default:
 			if (hdr->evt != r->event)
 				break;

[-- Attachment #3: Type: text/plain, Size: 228 bytes --]

-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/

[-- Attachment #4: Type: text/plain, Size: 164 bytes --]

_______________________________________________
Bluez-devel mailing list
Bluez-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bluez-devel

[Bluez-devel] Patch mh3

[Olivier Le Pogam]

Hello,

I have a 2.6.22.12 kernel (ubuntu gutsy gibbon) and when I want to apply the 
mh3 kernel patch (dry run for now) it seems there are some modifications 
which have already been applied (partially ?) in hci_core.h, rfcomm.h, sco.h

Could you please tell me what I have to do :( ?
Or where I can get an up-to-date drivers/bluetooth/* , /net/bluetooth/* , 
/include/net/bluetooth/*  sources ?

Thanks a lot,
Oli

------------------------------------------------------------------

patching file linux-source-2.6.22/include/net/bluetooth/rfcomm.h
Reversed (or previously applied) patch detected!  Assume -R? [n]
Apply anyway? [n]
Skipping patch.
1 out of 1 hunk ignored -- saving rejects to file 
linux-source-2.6.22/include/net/bluetooth/rfcomm.h.rej
patching file linux-source-2.6.22/Makefile
Hunk #1 FAILED at 1.
1 out of 1 hunk FAILED -- saving rejects to file 
linux-source-2.6.22/Makefile.rej
patching file linux-source-2.6.22/net/bluetooth/hci_conn.c
Hunk #6 FAILED at 243.
Hunk #7 succeeded at 348 (offset 28 lines).
Hunk #8 succeeded at 365 (offset 28 lines).
Hunk #9 succeeded at 403 (offset 28 lines).
Hunk #10 succeeded at 424 (offset 28 lines).
Hunk #11 succeeded at 438 (offset 28 lines).
Hunk #12 succeeded at 456 (offset 28 lines).
Hunk #13 succeeded at 478 (offset 28 lines).
Hunk #14 succeeded at 509 (offset 28 lines).
Hunk #15 succeeded at 519 (offset 28 lines).
Hunk #16 succeeded at 545 (offset 28 lines).
1 out of 16 hunks FAILED -- saving rejects to file 
linux-source-2.6.22/net/bluetooth/hci_conn.c.rej
patching file linux-source-2.6.22/net/bluetooth/hci_core.c
Hunk #8 succeeded at 831 (offset 1 line).
Hunk #9 succeeded at 856 (offset 1 line).
Hunk #10 succeeded at 871 (offset 1 line).
Hunk #11 succeeded at 898 (offset 1 line).
Hunk #12 succeeded at 910 (offset 1 line).
Hunk #13 succeeded at 937 (offset 1 line).
Hunk #14 succeeded at 1118 (offset 1 line).
Hunk #15 succeeded at 1150 (offset 1 line).
Hunk #16 succeeded at 1159 (offset 1 line).
Hunk #17 FAILED at 1260.
Hunk #18 succeeded at 1405 (offset 44 lines).
Hunk #19 succeeded at 1440 with fuzz 2 (offset 44 lines).
1 out of 19 hunks FAILED -- saving rejects to file 
linux-source-2.6.22/net/bluetooth/hci_core.c.rej
patching file linux-source-2.6.22/net/bluetooth/hci_event.c
Hunk #2 FAILED at 326.
Hunk #7 FAILED at 933.
Hunk #8 succeeded at 1125 (offset -3 lines).
Hunk #9 succeeded at 1196 (offset -3 lines).
Hunk #10 succeeded at 1221 (offset -3 lines).
Hunk #11 succeeded at 1342 (offset -3 lines).
Hunk #12 succeeded at 1378 (offset -3 lines).
Hunk #13 succeeded at 1392 (offset -3 lines).
Hunk #14 succeeded at 1464 (offset -3 lines).
Hunk #15 succeeded at 1472 (offset -3 lines).
2 out of 15 hunks FAILED -- saving rejects to file 
linux-source-2.6.22/net/bluetooth/hci_event.c.rej


-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Bluez-devel mailing list
Bluez-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bluez-devel

Re: [Bluez-devel] Patch mh3

[Marcel Holtmann]

Hi Olivier,

> I have a 2.6.22.12 kernel (ubuntu gutsy gibbon) and when I want to apply the 
> mh3 kernel patch (dry run for now) it seems there are some modifications 
> which have already been applied (partially ?) in hci_core.h, rfcomm.h, sco.h

the patch name is patch-2.6.22-mh3. This means it applies to 2.6.22 and
might not apply to 2.6.22.12. Otherwise it would have been named
patch-2.6.22.12-mh3. It is as simple as that.

Regards

Marcel



-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Bluez-devel mailing list
Bluez-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bluez-devel

Re: [Bluez-devel] Patch mh3

[Olivier Le Pogam]

Hi Marcel,

Thanks a lot, indeed it's really simple :)
Is there a way to have all of these changes applied to my current kernel 
version, or should I simply wait for a next kernel release ?

Best regards,
Oli
----- Original Message ----- 
From: "Marcel Holtmann" <marcel@holtmann.org>
To: "Olivier Le Pogam" <olepogam@free.fr>; "BlueZ development" 
<bluez-devel@lists.sourceforge.net>
Sent: Monday, October 01, 2007 4:14 PM
Subject: Re: [Bluez-devel] Patch mh3


> Hi Olivier,
>
>> I have a 2.6.22.12 kernel (ubuntu gutsy gibbon) and when I want to apply 
>> the
>> mh3 kernel patch (dry run for now) it seems there are some modifications
>> which have already been applied (partially ?) in hci_core.h, rfcomm.h, 
>> sco.h
>
> the patch name is patch-2.6.22-mh3. This means it applies to 2.6.22 and
> might not apply to 2.6.22.12. Otherwise it would have been named
> patch-2.6.22.12-mh3. It is as simple as that.
>
> Regards
>
> Marcel
>
>
>
>
>
> -- 
> No virus found in this incoming message.
> Checked by AVG Free Edition.
> Version: 7.5.488 / Virus Database: 269.13.35/1040 - Release Date: 
> 30/09/2007 21:01
> 


-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Bluez-devel mailing list
Bluez-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bluez-devel

Re: [Bluez-devel] Patch mh3

[Marcel Holtmann]

Hi Olivier,

> Is there a way to have all of these changes applied to my current kernel 
> version, or should I simply wait for a next kernel release ?

there is a way to do this, but not as simple as you think. However since
2.6.23 is around the corner, you better wait for 2.6.23-mh1.

Regards

Marcel



-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Bluez-devel mailing list
Bluez-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bluez-devel

Re: [Bluez-devel] Patch mh3

[Mats Erik Andersson]

m=E5n 2007-10-01 klockan 16:12 +0200 skrev Olivier Le Pogam:
> Hello,
> =

> I have a 2.6.22.12 kernel (ubuntu gutsy gibbon) and when I want to apply =
the =

> mh3 kernel patch (dry run for now) it seems there are some modifications =

> which have already been applied (partially ?) in hci_core.h, rfcomm.h, sc=
o.h
> =

Do a manual auditing of those patch bits that fail: read the
corresponding source in your kernel-source-2.6.22.12 and cross check
aginst the suggested amending from the patch file. Remove from the
patch file those parts that the Ubuntu maintainer already applied,
and keep the new additions. Then perform a final (or intermediary)
dry-runs with the patch command to check that things apply cleanly.
Of course, you have to use your own wit to go with the Bluez maintainers
suggestion against the decision of the Ubuntu crew, but the time will
be well invested. I used this technique to patch my own kernel
sources for Debian Etch 2.6.18 and I did eliminate some suboptimal
behaviour of Bluez in doing so.

In your case, disect the following contributions:

> patching file linux-source-2.6.22/include/net/bluetooth/rfcomm.h
> patching file linux-source-2.6.22/Makefile
> patching file linux-source-2.6.22/net/bluetooth/hci_conn.c
> patching file linux-source-2.6.22/net/bluetooth/hci_core.c
> patching file linux-source-2.6.22/net/bluetooth/hci_event.c

since they only produced rejects.


Good luck on your venture

Mats Erik Andersson


-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2005.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Bluez-devel mailing list
Bluez-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bluez-devel

Re: [Bluez-devel] [PATCH]Re: multithreaded name resolution

[Matthias Becker]

> > > I tried to read the name of different devices parallel using multi
> > > threading, but without success. If there is one bluetooth device I works
> > > (as there is no real multi threading). If there are who devices, the
> > > same name is reported for both devices. In another test with 6 devices
> > > hci_read_remote_name returned something <0 so I did not get any name at
> > > all.
> > 
> > this won't work. Asking for the name of a device is some sort of 
> > connection setup, and most (all ?) Bt chips do not allow / can not handle 
> > parallel connection setups (even though parallel connections are 
> > possible). 
> > 
> > So don't try further, the problem is outside your code (and even BlueZ). 
> > 
> 
> That's not true, the reported behavior is produced by bluez-libs. I already
> sent a patch to fix it and will attach that patch again to this mail.
> 
> Please let me know if that Patch fixes your problem Matthias.
> 
> Greetings
> 
> Andy
> 
> See also:
> http://thread.gmane.org/gmane.linux.bluez.devel/12491

Sorry for the delay. I tried your patch and in my tests it works
perfectly with up to three parallel name requests using this bluetooth
adapter: http://www.trust.com/products/default.aspx?item=14693

Thanks

-- Matthias Becker


-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
Bluez-devel mailing list
Bluez-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/bluez-devel