Alexis DOMJAN <[EMAIL PROTECTED]> writes:
> Because I didn't have an old kernel I didn't tried to increase the slot timeout
> immediately. But after hours of different test I tried to set up the timeout
> to 90, and then it works...
>
> Here's my configuration:
>
> * irda-common/irda-tools : 0.9.5-1.1 (debian potato)
> * kernel 2.2.14-pre14
>
> I tried with a slot timeout of 89 and it does not work. So the smaller value
> is 90.
Yes, 89 is the same as 80 when you divide it by 10 ;-) I assume your kernel
is using a HZ of 100 (default). If you _really_ want to use a timeout of
89, then you must increase the HZ to say 1000. The Linux kernel cannot
generate a timout of 89 ms in any good way when using a HZ of 100.
> The weird thing is that with the same phone from a friend it works with a
> 80 ms as timeout...
Not weird, since there all small margins here. My phone uses about 87 ms
(see below) from the XID is sent til the answer is received. So if I start
the transmission of a new frame earlier than 87 ms, then it will destroy
the incomming frame. Just run irdadump -d and check out how fast your and
your friends phone is reponding:
19:19:34.898291 (0090.01 ms) xid:cmd 18088056 > ffffffff S=6 s=1 (14)
19:19:34.985651 (0087.36 ms) xid:rsp 18088056 < 24757385 S=6 s=1 I 888 WORLD
hint=9104 [ PnP Modem IrCOMM ] (31)
19:19:34.988214 (0002.56 ms) xid:cmd 18088056 > ffffffff S=6 s=2 (14)
19:19:35.078199 (0089.99 ms) xid:cmd 18088056 > ffffffff S=6 s=3 (14)
19:19:35.168223 (0090.02 ms) xid:cmd 18088056 > ffffffff S=6 s=4 (14)
19:19:35.258218 (0090.00 ms) xid:cmd 18088056 > ffffffff S=6 s=5 (14)
19:19:35.348264 (0090.05 ms) xid:cmd 18088056 > ffffffff S=6 s=* Linux hint=0400 [
Computer ] (21)
-- Dag
--
/ Dag Brattli | The Linux-IrDA Project /
// University of Tromsoe, Norway | Infrared communication for Linux //
/// http://www.cs.uit.no/~dagb | http://www.cs.uit.no/linux-irda/ ///
_______________________________________________
Linux-IrDA mailing list - [EMAIL PROTECTED]
http://www4.pasta.cs.UiT.No/mailman/listinfo/linux-irda
