RE: [coldsync-hackers] modem sync success -- details for contribution

["John Joseph Roets" <jjr at sunflower dot com>]

Sorry for the delay.  You've probably forgotten the entire matter, but I've
finally had the chance to place the calls to cfget[io]speed() into the code
directly preceding the explicit change to 115200 (see below).  They show that
the port is set to 4098 when it's not been explicitly set.  That is, if we don't
set it to 9600 (as normal), or 115200 (as I've done for the modem).

log:
input speed: 4098
output speed: 4098
pconn->speed == 115200


Thus, I already have code delivering use of a modem; tag in the listen block
[from 2.2.5].
This does work, however, I'd still like to know if it'd be better to have an
init_speed within the listen block (which if unset would default to the palm
standard of 9600).
What do you think?
Where should I go from here to get this code contributed?  I'd like to be able
to upgrade later without rework.

Thank you.

> -----Original Message-----
> From: owner-coldsync-hackers@lusars.net
> [mailto:owner-coldsync-hackers@lusars.net]On Behalf Of Andrew
> Arensburger
> Sent: Tuesday, January 29, 2002 11:14 AM
> To: coldsync-hackers@lusars.net
> Subject: Re: [coldsync-hackers] modem sync success -- details for
> contribution
>
>
> On Tue, Jan 29, 2002 at 10:52:46AM -0600, John Joseph Roets wrote:
> > > Andrew Arensburger wrote:
> > > 	How about a "modem" listen block flag and "-M" command-line
> > > option?
> >
> > Andrew, since it appears that ColdSync need be set to the
> appropriate initial
> > speed, and that speed _does_ need to be set (and not just deleted as I
> > originally thought/posted, how about this solution:  A listen block
> directive
> > [init_speed] is provided to allow one to override the default B9600 initial
> > connect speed.  This would solve the problem for me, and likely any
> other device
> > with strange comm settings to come.  This would make the change
> more generic,
> > and avoid more need for like minded changes later.
>
> 	First of all, I'm confused as to why it wouldn't be sufficient
> to delete the cfset{i,o}speed() calls. When the device is opened,
> isn't it already set to some speed? Why not just leave it at that
> speed? (Or don't you know? Perhaps this is a deep mystery.)
>
> 	Secondly, the CMP speed of 9600 bps is mandated by the HotSync
> protocol for serial connections. The only exception to this is what
> you've encountered, in which the Right Thing is to set it to whatever
> speed the modem is already set to.
> 	Hence, one solution would be to make the 'modem' flag mean
>
> 	ispeed = cfgetispeed();
> 	ospeed = cfgetospeed();
> 	cfsetispeed(ispeed);
> 	cfsetospeed(ospeed);
>
> A yet more generic solution would be the 'init_speed' solution you
> suggested, and make 'init_speed: 0' mean "set it to the current
> speed," as above.
>
> 	But I'm still curious as to why you can't just delete the two
> cfset[io]speed() lines. Could you add calls to cfget[io]speed() just
> before, and see what they report?
>


-- 
This message was sent through the coldsync-hackers mailing list.  To remove
yourself from this mailing list, send a message to majordomo@thedotin.net
with the words "unsubscribe coldsync-hackers" in the message body.  For more
information on Coldsync, send mail to coldsync-hackers-owner@thedotin.net.