I think so, but I don't know if it will help you.
My app does some work in response to the sysAppLaunchCmdSyncNotify and
sysAppLaunchCmdHandleSyncCallApp launch codes, and under some circumstances
a HotSync done on POSE would appear to work normally, but I would get the
"You must now reset ..." message at the end. On a real device the same test
case appeared to work fine with no reset message. Other test cases caused
the HotSync process to hang at the "cleaning up..." phase.
When v3.1 of POSE came out, for these test cases it reported that my app was
close to overflowing the stack - an error that was not reported before.
Upon further investigation with the debugger, it turns out that POSE was
right.
For those launch codes the stack is already deeper than normal when an app
is called - deep enough to cause the function call depth of my code to
overflow the stack. Upon reorganizing the code a little, I was able to
reduce the function call depth enough to avoid overflowing the stack. This
also made all the weird failure modes of HotSync on POSE and real devices go
away.
The upshot of this is that overflowing the stack can cause unpredictable
results, depending on what gets clobbered by the overflow. If code in your
app responds to HotSync launch codes, this could be your problem.
Mark Peters
in article 44386@palm-dev-forum, DeAnna Davidson at [EMAIL PROTECTED] wrote
on 4/2/01 9:23 AM:
>
> OK, so I've tried everything I can immediately think of, I'm sure the
> reset bit is not set in the pdb, but I still get a reset at HotSync.
> Mark, did you happen to find a solution?
>
> DeAnna
>
--
For information on using the Palm Developer Forums, or to unsubscribe, please see
http://www.palmos.com/dev/tech/support/forums/
