Dag Brattli <[EMAIL PROTECTED]> writes:

> Hi,
> 
> I've just put out patch-2.3.15-irda2 for testing. This patch must be
> applied after patch-2.3.14-irda1, since it doesn't include this patch by
> itself. It might also be possible to apply this after patch-2.2.12-irda1 as 
> well (use --dry-run first).
> 
> Some changes:
> 
> o Fixed bug in IrLMP state machine (me)
> 
> o Fixed IrCOMM to be more robust (hopefully), but most importantly it and
>   will hopefully work for more devices as well. (me)
> 
> Please rapport back if this fixes the problems some of you have rapported
> with the previous patch.

I applied it to 2.2.12+irda1 with the following addional patch (to fix
rejections - I don't know what I am doing here):

--- linux/net/irda/ircomm/ircomm_tty.c.orig     Fri Aug 27 16:55:49 1999
+++ linux/net/irda/ircomm/ircomm_tty.c  Tue Aug 31 19:33:11 1999
@@ -392,12 +390,13 @@
                self->tqueue.routine = ircomm_tty_do_softint;
                self->tqueue.data = self;
                self->max_header_size = 5;
-               self->max_data_size = 2048-self->max_header_size;
+               self->max_data_size = 64-self->max_header_size;
 
                /* Init some important stuff */
                init_timer(&self->watchdog_timer);
-               skb_queue_head_init(&self->tx_queue);
-
+               init_waitqueue(&self->open_wait);
+               init_waitqueue(&self->close_wait);
+ 
                /* Insert into hash */
                hashbin_insert(ircomm_tty, (QUEUE *) self, line, NULL);
 
--- linux/include/net/irda/ircomm_tty.h.orig    Fri Aug 27 17:13:45 1999
+++ linux/include/net/irda/ircomm_tty.h Tue Aug 31 19:23:20 1999
@@ -83,9 +83,13 @@
        struct termios    callout_termios;
 
        struct wait_queue *open_wait;
+       struct wait_queue *close_wait;
        struct timer_list watchdog_timer;
        struct tq_struct  tqueue;
 
+ 
+       unsigned short    close_delay;
+       unsigned short    closing_wait; /* time to wait before closing */
        long pgrp;              /* pgrp of opening process */
        int  open_count;
        int  blocked_open;      /* # of blocked opens */


But still: The same problem with NO CARRIER.

I will try to get hold of a SH888 and see if this is just some problem I 
am having for some reason, or if it really depends on the phone.


Bj�rn


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

Reply via email to