À (At) 10:25 -0800 10/11/02, Blair Rosser écrivait (wrote) :
> > In fact, the Sandisks should work. They do work in Peter
>> Cameron's Newton, they would probably work in mine.
>
>Thanks for the encouragement, Paul. This is good news.
>
>I'll test again with RC2 when I get back to my office next week.
The -10606 bug has been fixed, but not in RC2. It will be fixed in
RC3. And the weird -10059 error that you had was probably fixed as
well.
If you want details about the bug, here are they. You can skip to
last paragraph otherwise :)
Under some circumstances, the Newton looks for indexes on the newly
formatted store. This especially happens if you have changed the
indexes from US to German or Swedish it didn't happen on my Newton.
This explains why:
(a) Harri didn't experience the bug at first when he reformatted his
Newton, but did afterwards (I guess he restored data or configured
his Newton to Sweden locale)
(b) David and Christian didn't experience the bug when they
reformatted their Newton, as they got rid of some additional
indexes/data.
Without Anne-Kathrin Hesse's report that she experienced the problem
after installing the german patch (that tells you to change the
indexes), I probably wouldn't have discovered this issue.
So the system when formatting the store refers to an object that
doesn't exist yet (since the store wasn't fully formatted). It then
compares the NewtonScript frame of the store with other frames.
Since beta 10 (hence the 'it worked with beta 10, what have you
done?' reports), I patch this frame somehow to recall which slot the
card is in. But at this point of the code, the NewtonOS compares the
unpatched frame with the patched frame and this comparison fails.
(this is where the code is really dirty, instead, it should convert
the NewtonScript frame to the C++ object and compare the two C++
objects which would succeed).
So it asks the object that doesn't exist yet to do some job
(basically to reindex the soups). Since the object doesn't exist, it
yields to a big number error (bus error). Then bad things happen
because formatting failed and was interrupted.
Unfortunately, RC3 which fixes that bug isn't ready yet. I hope that
I'll be able to release RC3 before next week. I know that at least
two registered users suffer from this bug so I try to put as much
time I can on that task.
Paul
-- NPDS: http://newton.kallisys.net:8080/ Apache: http://www.kallisys.com/-- This is the NewtonTalk list - http://www.newtontalk.net/ for all inquiries List FAQ/Etiquette/Terms: http://www.newtontalk.net/faq.html Official Newton FAQ: http://www.chuma.org/newton/faq/
This archive was generated by hypermail 2.1.2 : Mon Dec 02 2002 - 22:02:25 EST