> BTW, the problem I mentioned some time ago of pppd hanging at the second
> login after t39m power cycle is fixed after the upgrade of the firmware
> of the phone too.  but before doing the upgrade I found a way to
> workaround it on the linux side too: you only need to comment out
> default-asyncmap in the ppp/options.

Thanks very much for the tip about avoiding the pppd hang on the
second connect attempt.

Following your lead, I found that adding "receive-all" to my pppd
options (/etc/ppp/peers/t39-gprs) solved the problem.

BTW, I have the following T39 firmware version:
  at+cgmm : "1130102-BVT39m"
  at+cgmr : "R2G         CXC125268"


However, even with the max_baud_rate = 115200 I'm afraid things
still aren't entirely happy with my Sony Vaio N505X talking to
the T39 over IrDA.

Using the RS232 cable, I get TCP throughput of around 3.6KB/s
(28.8Kb/s). Using IrDA with an identical setup (and phone
position) I get only 1.2KB/s. I've tried all the obvious things
like changing the distance between the Ir sensors etc.

It looks like the IrDA link introduces substantially higher
latency, along with relatively frequent blocks of total loss.
This suggests to me that there's loads of retransmissions going
on at the IrDA level.

With a 'ping' running over the IrDA/GPRS link, I recorded the
following with 'irdadump':


18:33:46.064764 i:cmd  > ca=98 pf=1 nr=7 ns=5 LM slsap=45 dlsap=02 (95) 
18:33:46.085226 i:rsp  < ca=98 pf=0 nr=6 ns=7 LM slsap=02 dlsap=45 (95) 
18:33:46.094658 i:rsp  < ca=98 pf=1 nr=6 ns=0 LM slsap=02 dlsap=45 (95) 
18:33:46.594598 rr:cmd > ca=98 pf=1 nr=1 (2) 
18:33:46.607002 rr:rsp < ca=98 pf=1 nr=6 (2) 
18:33:47.064808 i:cmd  > ca=98 pf=1 nr=1 ns=6 LM slsap=45 dlsap=02 (95) 
18:33:47.087454 rr:rsp < ca=98 pf=1 nr=7 (2) 
18:33:47.584623 rr:cmd > ca=98 pf=1 nr=1 (2) 
18:33:47.595684 rr:rsp < ca=98 pf=1 nr=7 (2) 
18:33:48.064818 i:cmd  > ca=98 pf=1 nr=1 ns=7 LM slsap=45 dlsap=02 (96) 
18:33:48.085401 rr:rsp < ca=98 pf=1 nr=0 (2) 
18:33:48.584653 rr:cmd > ca=98 pf=1 nr=1 (2) 
18:33:48.596680 i:rsp  < ca=98 pf=0 nr=0 ns=1 LM slsap=02 dlsap=45 (95) 
18:33:48.606083 i:rsp  < ca=98 pf=1 nr=0 ns=2 LM slsap=02 dlsap=45 (95) 
18:33:49.064868 i:cmd  > ca=98 pf=1 nr=3 ns=0 LM slsap=45 dlsap=02 (95) 
18:33:49.084164 rr:rsp < ca=98 pf=1 nr=1 (2) 
18:33:49.574680 rr:cmd > ca=98 pf=1 nr=3 (2) 
18:33:49.588086 rr:rsp < ca=98 pf=1 nr=1 (2) 
18:33:50.064875 i:cmd  > ca=98 pf=1 nr=3 ns=1 LM slsap=45 dlsap=02 (95) 
18:33:50.085890 i:rsp  < ca=98 pf=0 nr=2 ns=3 LM slsap=02 dlsap=45 (95) 
18:33:50.096202 i:rsp  < ca=98 pf=1 nr=2 ns=4 LM slsap=02 dlsap=45 (96) 

At the same time, the irda_debug output was:

ircomm_tty_data_indication()
irlmp_discovery_timer_expired()
irlap_do_event(), event = POLL_TIMER_EXPIRED, state = LAP_XMIT_P
irlap_state_xmit_p(), POLL_TIMER_EXPIRED (10339452)
async_wrap_skb_R6cd9cf2b(), xbofs=0
nsc_ircc_pio_write()
nsc_ircc_pio_write(), fifo_size 25 ; 6 sent of 6
next LAP state = LAP_NRM_P
irlap_do_event(), event = RECV_I_RSP, state = LAP_NRM_P
irlap_validate_nr_received(), expected!
next LAP state = LAP_XMIT_P
irlmp_link_data_indication()
irttp_run_rx_queue() send=6,avail=2,remote=11
ircomm_ttp_data_indication()
ircomm_do_event: state=IRCOMM_CONN, event=IRCOMM_TTP_DATA_INDICATION
ircomm_data_indication()
ircomm_tty_data_indication()
ircomm_tty_write(), count=89, hw_stopped=0
ircomm_tty_do_softint()
ircomm_tty_do_event: state=IRCOMM_TTY_READY, event=IRCOMM_TTY_DATA_REQUEST
ircomm_data_request()
ircomm_do_event: state=IRCOMM_CONN, event=IRCOMM_DATA_REQUEST
ircomm_ttp_data_request(), clen=0
irttp_run_tx_queue() : Got SKB not attached to a socket.
irlmp_do_lsap_event(), EVENT = LM_DATA_REQUEST, STATE = LSAP_DATA_TRANSFER_READY
irlmp_state_dtr()
irlap_data_request()
irlap_do_event(), event = SEND_I_CMD, state = LAP_XMIT_P
async_wrap_skb_R6cd9cf2b(), xbofs=0
nsc_ircc_pio_write()
nsc_ircc_pio_write(), fifo_size -1 ; 32 sent of 99
next LAP state = LAP_NRM_P
nsc_ircc_pio_write()
nsc_ircc_pio_write(), warning, FIFO not empty yet!
nsc_ircc_pio_write(), fifo_size -1 ; 15 sent of 67
nsc_ircc_pio_write()
nsc_ircc_pio_write(), warning, FIFO not empty yet!
nsc_ircc_pio_write(), fifo_size -1 ; 15 sent of 52
nsc_ircc_pio_write()
nsc_ircc_pio_write(), warning, FIFO not empty yet!
nsc_ircc_pio_write(), fifo_size -1 ; 15 sent of 37
nsc_ircc_pio_write()
nsc_ircc_pio_write(), warning, FIFO not empty yet!
nsc_ircc_pio_write(), fifo_size -1 ; 15 sent of 22
nsc_ircc_pio_write()
nsc_ircc_pio_write(), warning, FIFO not empty yet!
nsc_ircc_pio_write(), fifo_size 7 ; 7 sent of 7
irlap_do_event(), event = RECV_RR_RSP, state = LAP_NRM_P
irlap_validate_nr_received(), expected!
next LAP state = LAP_XMIT_P
irlap_do_event(), event = POLL_TIMER_EXPIRED, state = LAP_XMIT_P
irlap_state_xmit_p(), POLL_TIMER_EXPIRED (10339550)
async_wrap_skb_R6cd9cf2b(), xbofs=0
nsc_ircc_pio_write()
nsc_ircc_pio_write(), fifo_size 25 ; 6 sent of 6
next LAP state = LAP_NRM_P
irlap_do_event(), event = RECV_RR_RSP, state = LAP_NRM_P
irlap_validate_nr_received(), expected!
next LAP state = LAP_XMIT_P
ircomm_tty_write(), count=89, hw_stopped=0
ircomm_tty_do_softint()
ircomm_tty_do_event: state=IRCOMM_TTY_READY, event=IRCOMM_TTY_DATA_REQUEST
ircomm_data_request()
ircomm_do_event: state=IRCOMM_CONN, event=IRCOMM_DATA_REQUEST
ircomm_ttp_data_request(), clen=0
irttp_run_tx_queue() : Got SKB not attached to a socket.
irlmp_do_lsap_event(), EVENT = LM_DATA_REQUEST, STATE = LSAP_DATA_TRANSFER_READY


I'm afraid I haven't yet figured out how to interpret this, but
thought I'd post it in case an expert eye could spot if there's
anything strange happening and point me in the right direction.

I'm using "nsc-ircc" configure to use the "IBM31T1100 or Temic
TFDS6000/TFDS6500" dongle. I beginning to wander if this setting
is correct for the Sony N505X...


Many Thanks,
Ian

_______________________________________________
Linux-IrDA mailing list  -  [EMAIL PROTECTED]
http://www.pasta.cs.UiT.No/mailman/listinfo/linux-irda

Reply via email to