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

RE: [bluetooth-dev] What is the effects of 'sleep(10)'


i notice that the "negative length" error occurs under the following
condition: while the sending thread is inside the send_acl_packet, the
num_of_completed_packets event is received by the receiving thread, which
calls the update_ncp function. the update_ncp function in turn calls the
send_acl_packet function. therefore the cur_len is modified twice, resulting
in the negative length error. is this a known race condition problem?

why the sending is implemented differently in kernel mode and user mode? 

i guess the sti() and cli() functions are used to enable/disable interrupts,
am i right?

ur help will be appreciated!

best regards,

liu yong

-----Original Message-----
From:	Peter Kjellerstedt
Sent:	Mon 28/05/2001 10:08 PM
To:	'???'; bluetooth-dev@xxxxxxx.com
Subject:	RE: [bluetooth-dev] What is the effects of 'sleep(10)'

First of all is this an error case (negative length...), 
and second it is only used for user mode stack.
Though I do not see any real reason why it is there...

-----Original Message-----
From: kdk@xxxxxxx.edu">mailto:kdk@xxxxxxx.edu]
Sent: 28 May 2001 16:01
To: bluetooth-dev@xxxxxxx.com
Subject: [bluetooth-dev] What is the effects of 'sleep(10)'

In the 'send_acl_data()' function in 'hci.c' file,  there is 'sleep(10)'.
Does it significantly affect the speed of the bluetooth module? I think it
cause the host-to-host controller's speed to be lowered. 
Dong Kyoo Kim
Office) 054-279-5588

To unsubscribe from this list: send the line "unsubscribe bluetooth-dev" in
the body of a message to majordomo@xxxxxxx.com