hi,
I am trying to measure the time when the frame is inserted into queue
(ath_tx_process_buffer, ath_drain_txq_list[not req actually])
and when one receives ACK from hardware in ath_tx_process_buffer()
in the descriptor, ath_tx_status and eventually the
timestamp is calculated in ath_tx_complete_
This is a sample output showing the time for ACK_from_HW-
Enque_of_frame(seq no. 3513) i=318 s less than
framesize/bitrate (685.33)
If someone can help resolve the issue
tsf txflags,retx, successful bitrate, total time,Qlen,AMPDU-Q
len,Q no, phy-type,retx rate list,seq no, frag n
Well I have understood the problem.
As a check and for other users, I would like someone to correct me (Adrian
or Felix ? )
the 32 bit tsf value returned by ath_tx_status is the timestamp when the
last of
the x retransmissions for a packet begins ?
It is not the time when the last byte of the fram
On 13 April 2013 08:09, abhinav narain wrote:
> The question following it is :
> if the tx descriptors(ath_tx_status) tsf corresponds to time when
> (a) bits start hitting the air, or
> (b) when the driver starts the process of waiting for idle slots ?
So, the MAC is the thing adding the times
