[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: [bluetooth-dev] Something really broken in the latest cvs.



As I said before, I haven't seen this problem myself and I'm running the latest version of the stack without any problems. Though, this might be some problem with some specific HW, I am running the stack on CSR HW, and that works fine. I suggest that you try to check out previous versions of the stack until you find a version that works and then do a diff, to see what have changed since then.

Regards

Mats 

>-----Original Message-----
>From: david LIBAULT [mailto:david.libault@xxxxxxx.fr]
>Sent: Tuesday, June 19, 2001 4:36 PM
>To: Mats Fridén
>Subject: Re: [bluetooth-dev] Something really broken in the latest cvs.
>
>
>I have just crashed my PC runing the bluetooth stack + btd.
>
>The only smart solution I see to fix this problem is to take 
>the cvs backward 
>step by step until we find the commit that corrupts everything.
>Can you do it ? Do you have a better idea ?
>
>BTW, has anyone seen this  on his system ?
>
>David.
>
>Le Mardi 19 Juin 2001 15:26, Mats Fridén a écrit :
>> I can't see how process_event() can possible change the count and or
>> event_len parameter ?
>>
>> /Mats
>>
>> >-----Original Message-----
>> >From: david LIBAULT [mailto:david.libault@xxxxxxx.fr]
>> >Sent: Tuesday, June 19, 2001 3:13 PM
>> >To: Mats Fridén
>> >Cc: bluetooth-dev@xxxxxxx.com
>> >Subject: Re: [bluetooth-dev] Something broken in latest cvs ?
>> >
>> >
>> >Thank you for your answer.
>> >
>> >I think you should be able to repeat this problem, as I have
>> >just tried this
>> >on my PC with an Ericsson module (that returns 1 as cmd_num),
>> >and exactly the
>> >same thing happens...
>> >
>> >David.
>> >
>> >Le Mardi 19 Juin 2001 15:02, Mats Fridén a écrit :
>> >> Hi,
>> >>
>> >> I haven't seen this before, and I can't repeat it myself either...
>> >>
>> >> There might be some problems (actually, I don't know, since
>> >
>> >I've never
>> >
>> >> tested it) if you are using a HW that may set the cmd_num to
>> >
>> >a value larger
>> >
>> >> than one...
>> >>
>> >> Though, to me this looks like some kind of race problem...
>> >>
>> >> /Mats
>> >>
>> >> >-----Original Message-----
>> >> >From: david LIBAULT [mailto:david.libault@xxxxxxx.fr]
>> >> >Sent: Tuesday, June 19, 2001 2:20 PM
>> >> >To: bluetooth-dev@xxxxxxx.com
>> >> >Subject: Re: [bluetooth-dev] Something broken in latest cvs ?
>> >> >
>> >> >
>> >> >In function hci_receive_data (in hci.c) I have printked 
>the values
>> >> >of the static variables count and event_len before and after
>> >> >the call to
>> >> >process_event . It looks like these variables are corrupted by
>> >> >the call to
>> >> >process_event. This happens for what ever HCI command sent to
>> >> >the BT device.
>> >> >
>> >> >What happens ? Is anyone having the same problem ?
>> >> >
>> >> >David.
>> >> >
>> >> >[016,563392<003330>]  -->|X| (14):
>> >> >0x04 0x0e 0x0b 0x04 0x05 0x10 0x00 0xf4 0x01 0x40 0x08 0x00
>> >
>> >0x10 0x00
>> >
>> >> >[016,572939<009547>]  hci_receive_data,  (14):
>> >> >0x04 0x0e 0x0b 0x04 0x05 0x10 0x00 0xf4 0x01 0x40 0x08 0x00
>> >
>> >0x10 0x00
>> >
>> >> >hci_receive_data, hci-14
>> >> >HCI: hci_receive_data, WAIT_FOR_PACKET_TYPE
>> >> >HCI: hci_receive_data, WAIT_FOR_EVENT_TYPE
>> >> >HCI: hci_receive_data, WAIT_FOR_EVENT_LENGTH
>> >> >--------------->count : 11
>> >> >--------------->event_len : 11
>> >> >
>> >> >[016,599933<026994>]  process_event (11):
>> >> >0x04 0x05 0x10 0x00 0xf4 0x01 0x40 0x08 0x00 0x10 0x00
>> >> >HCI: process_event, COMMAND_COMPLETE
>> >> >HCI: <CC>
>> >> >HCI: release_cmd_timer
>> >> >HCI: process_return_param, READ_BUFFER_SIZE
>> >> >
>> >> >HW module contains...
>> >> >8 ACL buffers at 500 bytes
>> >> >16 SCO buffers at 64 bytes
>> >> >
>> >> >HCI: update_nhcp, Num_HCI_Command_Packets=4
>> >> >After process event
>> >> >--------------->count : -21
>> >> >--------------->event_len : 32
>> >> >HCI: hci_receive_data, WAIT_FOR_PACKET_TYPE
>> >> >BT SYS: ERROR :hci_receive_data: discarding -21 bytes
>> >> >
>> >> >Le Mardi 19 Juin 2001 10:54, david LIBAULT a écrit :
>> >> >> Something looks broken in hci on the latest CVS as the
>> >
>> >initialisation
>> >
>> >> >> doesn't work properly anymore on my system...
>> >> >> Here is my log :
>> >> >>
>> >> >> Initialisation stuff, and
>> >> >>
>> >> >> BT SYS: hci_init: Reading buffer sizes in HW module
>> >> >> HCI: hci_read_buffer_size
>> >> >> HCI: start_cmd_timer
>> >> >> HCI: insert_cmd
>> >> >> HCI: send_cmd_queue, start: Num_HCI_Command_Packets=1
>> >> >>
>> >> >> [016,551183<16551183>]  send_cmd_queue,  (4):
>> >> >> 0x01 0x05 0x10 0x00
>> >> >> [016,557298<006115>] BT DATA <--|X|      4
>> >> >>
>> >> >> [016,561171<003873>]  <--|X| (4):
>> >> >> 0x01 0x05 0x10 0x00
>> >> >> HCI: send_cmd_queue, end : 0 cmds left in queue
>> >> >> Opening dev /dev/ttyBTC
>> >> >> Init stack
>> >> >> IOCTL result [016,580060<018889>] BT DATA -->|X|     14
>> >> >>
>> >> >> [016,583423<003363>]  -->|X| (14):
>> >> >> 0x04 0x0e 0x0b 0x04 0x05 0x10 0x00 0xf4 0x01 0x40 0x08
>> >
>> >0x00 0x10 0x00
>> >
>> >> >> [016,592970<009547>]  hci_receive_data,  (14):
>> >> >> 0x04 0x0e 0x0b 0x04 0x05 0x10 0x00 0xf4 0x01 0x40 0x08
>> >
>> >0x00 0x10 0x00
>> >
>> >> >> hci_receive_data, hci-14
>> >> >> HCI: hci_receive_data, WAIT_FOR_PACKET_TYPE
>> >> >> HCI: hci_receive_data, WAIT_FOR_EVENT_TYPE
>> >> >> HCI: hci_receive_data, WAIT_FOR_EVENT_LENGTH
>> >> >>
>> >> >> [016,617533<024563>]  process_event (11):
>> >> >> 0x04 0x05 0x10 0x00 0xf4 0x01 0x40 0x08 0x00 0x10 0x00
>> >> >> HCI: process_event, COMMAND_COMPLETE
>> >> >> HCI: <CC>
>> >> >> HCI: release_cmd_timer
>> >> >> HCI: process_return_param, READ_BUFFER_SIZE
>> >> >>
>> >> >> HW module contains...
>> >> >> 8 ACL buffers at 500 bytes
>> >> >> 16 SCO buffers at 64 bytes
>> >> >>
>> >> >> HCI: update_nhcp, Num_HCI_Command_Packets=4
>> >> >> HCI: hci_receive_data, WAIT_FOR_PACKET_TYPE
>> >> >> BT SYS: ERROR :hci_receive_data: discarding -21 bytes
>> >> >>
>> >> >> : 0
>> >> >>
>> >> >> HCI: send_cmd_queue, start: Num_HCI_Command_Packets=4
>> >> >> HCI: send_cmd_queue, end : 0 cmds left in queue
>> >> >> Initialisation OK
>> >> >>
>> >> >>
>> >> >>
>> >> >> The stack is discarding -21 bytes... strange !
>> >> >> -
>> >> >> To unsubscribe from this list: send the line "unsubscribe
>> >> >
>> >> >bluetooth-dev" in
>> >> >
>> >> >> the body of a message to majordomo@xxxxxxx.com
>> >> >
>> >> >-
>> >> >To unsubscribe from this list: send the line "unsubscribe
>> >> >bluetooth-dev" in
>> >> >the body of a message to majordomo@xxxxxxx.com
>> >>
>> >> -
>> >> To unsubscribe from this list: send the line "unsubscribe
>> >
>> >bluetooth-dev" in
>> >
>> >> the body of a message to majordomo@xxxxxxx.com
>
-
To unsubscribe from this list: send the line "unsubscribe bluetooth-dev" in
the body of a message to majordomo@xxxxxxx.com