Kazuyuki Sato wrote:
> Hi Michael-san,
>  Thank you for your great work.  It worked with Apple MacBook Late 
> 2008 (MACBOOK 5,1) and Kensington SlimBlade
>  Trackball Mouse.( http://us.kensington.com/html/13880.html )
>  Wheel works, too.
>
>  There is a question.  Is it correct in this as the reconnecting method?
>
>     1. mouse is changed to the pairing mode.
>     2. execute btdevctl
To reconnect, the first step is detaching mouse driver from system manually.
#btdevctl -a yourmouse -d ubt0 -s hid -D
And then repeat the connect steps.
That's current situation, things will be changed in the future...
>  Please support the automatic reconnect of the device.
See above steps, no auto-reconnect supported for now.
>  
>  Moreover, After package is installed, a wrong permission for the / 
> directory  is set uid 148931 and gid staff.
>  Because, wificonfig createprofile was not able to be executed, it 
> knew it.
Ok, I will investigate how my patch change it. It is not obvious to me.
>
> -- 
> macbook(i86pc):/
> $ ls -la
> total 17962
> drwxr-xr-x  23 148931   staff         26 Feb 26 23:01 ./
> drwxr-xr-x  23 148931   staff         26 Feb 26 23:01 ../
> lrwxrwxrwx   1 root     other         27 Feb 24 02:54 .make.machines 
> -> opt/onbld/gk/.make.machines
> lrwxrwxrwx   1 root     root           9 Feb 23 14:08 bin -> ./usr/bin/
> -rw-r--r--   1 root     root     8563200 Feb 26 17:46 
> bluetooth_patch_for_snv_108.tar
> drwxr-xr-x   7 root     sys           10 Feb 23 15:10 boot/
> -- 
>
> Thanks, - Kazuyuki Sato 
>
> under becomes device information here.:
Well, you have a standard 2.1 Bluetooth usb controller!

Thanks for your feedback!

- Michael
>
> -- 
>> ::prtusb -i 9 -t
> INDEX   DRIVER      INST  NODE            VID.PID     
> PRODUCT             9       ubt         0     device          
> 05ac.8213   Bluetooth USB Host Controller
>
> usba_device: 0xbd6f5500
> mfg_prod_sn: Apple, Inc. - Bluetooth USB Host Controller - 00236CA19318
>
> device, instance #0 (driver name: ubt)
>  /pci at 0,0/pci10de,cb79 at 6/hub at 1/device at 1
>  dip: 0xbb034ee0
>  ubt_statep: 0xbd8a4a00
> -- 
>
> -- 
> Feb 26 17:51:59 macbook usba: [ID 912658 kern.info] USB 2.0 device 
> (usb5ac,8213) operating at full speed (USB 1.x) on USB 2.0 external 
> hub: device at 1, ubt0 at bus address 3
> Feb 26 17:51:59 macbook usba: [ID 349649 kern.info]     Apple, Inc. 
> Bluetooth USB Host Controller 00236CA19318
> Feb 26 17:51:59 macbook genunix: [ID 936769 kern.info] ubt0 is 
> /pci at 0,0/pci10de,cb79 at 6/hub at 1/device at 1
> Feb 26 17:51:59 macbook genunix: [ID 408114 kern.info] 
> /pci at 0,0/pci10de,cb79 at 6/hub at 1/device at 1 (ubt0) online
>
> Feb 27 12:54:22 macbook netbt: [ID 759129 kern.notice] Bluetooth HCI 
> ver 2.1|251
> Feb 27 12:54:22 macbook netbt: [ID 999797 kern.notice] Bluetooth LMP 
> ver 2.1|8610, from company 15
> Feb 27 12:54:22 macbook netbt: [ID 334930 kern.notice] 
> bthub_create_node child =ffffff01b9755ae8
> Feb 27 12:54:22 macbook pseudo: [ID 129642 kern.info] pseudo-device: 
> bthub0
> Feb 27 12:54:22 macbook genunix: [ID 936769 kern.info] bthub0 is 
> /pseudo/bthub at 0
> Feb 27 12:54:22 macbook genunix: [ID 408114 kern.info] /pseudo/bthub at 0 
> (bthub0) online
> Feb 27 12:54:47 macbook bthub: [ID 610781 kern.notice] bthub-device: 
> bthidev0
> Feb 27 12:54:47 macbook genunix: [ID 936769 kern.info] bthidev0 is 
> /pseudo/bthub at 0/bthidev at 0
> Feb 27 12:54:47 macbook genunix: [ID 408114 kern.info] 
> /pseudo/bthub at 0/bthidev at 0 (bthidev0) online
> Feb 27 12:54:47 macbook bthidev: [ID 963222 kern.notice] 
> bthidev_int_connected connected
> Feb 27 12:54:47 macbook bthidev: [ID 997010 kern.notice] report id 3, 
> len = 1 ignored
> Feb 27 14:43:31 macbook bthidev: [ID 997010 kern.notice] report id 3, 
> len = 1 ignored
>
> -- 
>
>
> Michael Li wrote:
>> Hi, Cyril,
>>
>> I said too fast, md5sum error could be part of the issue, but the 
>> root cause is version mismatch between "unix" and "genunix".
>>
>> After more experience with the patched system, same crash happened, 
>> the stack backtrace is the almost the same as the one you and Bernd 
>> reported.
>> The "unix" module is non-debug version shipped with SXCE, while 
>> "genunix" was debug version with Bluetooth patch. bp->b_queue is set 
>> in putnex() debug code section. This panic could be triggered by 
>> ethernet or wifi link up/down event handling.
>>
>> The fix is then very simple, I've uploaded non-debug version tarballs 
>> for b107/b108/b109 because SXCE DVD and Indiana Free CD build_X are 
>> non-debug version.
>> (If the system has been bfu(ed) to debug kernel, please bfu back to 
>> none-debug kernel before run btms_installer.sh).
>>
>> SXCE debug and non-debug kernel could be downloaded here: 
>> http://dlc.sun.com/osol/on/downloads.
>>
>> As you suggested, md5sum will be added into script to double ensure 
>> the download stage. And, I will invetigate pending [1] and contrib 
>> [2] later!
>>
>> Welcome and appriciate your suggestions!
>>
>> - Michael
>>
>> Cyril Plisko ??:
>>> Michael,
>>>
>>>  
>>>> Did you installed other speicial packages into your system? Another 
>>>> issue
>>>> might be from download stage, the bluetooth_mouse_snv_108.tar.gz 
>>>> md5sum is:
>>>> 197bd5c019d6c6513f9bde7beb270110, you may need to check it your side.
>>>>     
>>>
>>> Indeed, I've just checked my file and md5sum was wrong !
>>> I have to admit it is the first time it happens to me, but I guess
>>> there is always a first time :-/
>>>
>>> Anyway, I am glad it is borken download, rather than borken bits :)
>>>
>>> May I suggest add md5sum verification in the btms_install.sh script
>>> after downloading the archive ?
>>> That will prevent such problems in the future.
>>>
>>> I am going to rerun tests as soon as I can !
>>>
>>>  
>>>> btw, the bluetooth mouse packages test for Indiana0906-b107 is 
>>>> ongoing, but,
>>>> these packages will not be delivered to OpenSolaris Bluetooth 
>>>> project until
>>>> we find good method for people experience. The packages.tar.gz is 
>>>> about
>>>> 17MB, it can not be uploaded to OpenSolaris due to 15MB/per-file 
>>>> limitation,
>>>> any thoughts?
>>>>     
>>>
>>> I think you may want to investigate publishing your bits to pending
>>> [1] or contrib [2] repository. Folks at pkg-discuss mailing list
>>> should give you guidance on what is needed to publish there.
>>>
>>> [1] http://pkg.opensolaris.org/pending/en/index.shtml
>>> [2] http://pkg.opensolaris.org/contrib/en/index.shtml
>>>
>>>   
>>
>>
>>
>> ------------------------------------------------------------------------
>>
>> _______________________________________________
>> laptop-discuss mailing list
>> laptop-discuss at opensolaris.org


Reply via email to