From: "Jeff" <fyberoptic1979@...>
Jul 27, 2007
I decided to make a new topic with just a bunch of various questions,
and =
hopefully it'll be easier for folks to find in a search. HERE
GOES~
- W=
hat does "XTND: On" mean exactly in the top right corner of the
options sc=
reen?
- From my understanding, the demo units of the DET1 allowed you to
=
type in your own POP/SMTP servers. Does this mean the codeflash
between a=
demo unit and standard unit are different, or is it simply
detecting some=
hardware aspect and locking out the feature from
standard units? If the =
code is indeed different, has it ever been
dumped? If not/so, are the dum=
ps in the Files section from a
standard unit?
My own demo unit version =
seems to be 3.03a, which makes me think that
the code is in fact different=
, and that it hasn't been dumped (or at
least not uploaded), but I wanted =
to verify both things.
And related to this, I assume it would be possible =
to take a dump
from a demo unit and put it on a standard unit, so that one=
could
change the POP/SMTP servers. Not that it matters much, I don't use=
the modem functionality of the MS, but I'm curious nonetheless.
- Do we=
know what all triggers NMI interrupts?
- Do we know how exactly to write =
into the codeflash and/or dataflash
via software? Apparently the Mailstat=
ion can do it, so I'm only left
to assume one can either use the routines =
already there, or someone
has possibly figured out how to do it from scrat=
ch.
- How is a codeflash update triggered from the parallel port?
- How=
does the port "P2C =3D printer data direction" work? And related
to that=
: once set as inputs, does one still need to write 1's to the
data port i=
n order to read from it properly?
- Did we ever find out if there was a de=
vice 6? There seems to be a
pin for it on the processor pinout list. If =
nothing's attached, it
could be kind of useful for adding ones own hardwar=
e. Assuming one
could even solder to the tiny pins on the board!
I th=
ought I had more questions, but they seem to have escaped me at
the moment=
. But this is enough for now!
From: "John R. Hogerhuis" <jhoger@...>
Jul 27, 2007
Usually on devices like this I believe power-off is what triggers NMI.
This gives a small window to prepare for loss of power.
From: "Neil Morrison" <neilsmorr@...>
Jul 27, 2007
My take on this is that there must be a pin which puts the port into a
quas=
i serial I/O mode, and this is how the device is reflashed.
Neil
ailstation@yahoogroups.com, "Jeff" <fyberoptic1979@...> wrote:
a codeflash update triggered from the parallel port?
rt "P2C =3D printer data direction" work? And related
t as inputs, does one still need to write 1's to the
to read from it properly?
From: "Jeff" <fyberoptic1979@...>
Jul 29, 2007
dataflash
=
left
to assume one can either use the routines already there, or someone
possibly figured out how to do it from scratch.
I'm going to answer=
my own question with a bit of a "duh", because I
forgot that most modern =
Flash memory can be programmed in-system with
series of standard read/writ=
e command sequences. I normally do all
the ones I work with in a programm=
er, though I remembered this
helpful little fact while reading over a data=
sheet for a different
eeprom, which prompted me to look at the sheet for t=
he Mailstation's
dataflash (28SF040). It appears even simpler to do than =
some. Only
tricky part is that you'd have to write to the appropriate
Ma=
ilstation port to change the dataflash page (in whichever device
slot you =
were using) to set the upper address pins (A14-18), then
separately set A0=
-13 to the lower dataflash address during a normal
read/write z80 command,=
while making sure to still set A14-15 from
there appropriately to make su=
re you were writing to the correct 16k
Mailstation slot (aka to the eeprom=
).
Only difference I see from a quick glance at the codeflash is that it
=
has some longer command sequences, and of course a different sector
size.
=
Anyhoo, as I said, that may be a bit of a "duh" to many folks I'm
sure, b=
ut maybe it'll be useful info to somebody else. Obviously one
shouldn't s=
till need the MS's flash writing routines if you were
going to totally rep=
lace its data/codeflash.
From: "Cyrano Jones" <cyranojones_lalp@...>
Aug 1, 2007
I decided to make a new topic with just a bunch of various questions,
d hopefully it'll be easier for folks to find in a search. HERE
=
ions screen?
XTND, as best I can guess, refers to extensions to the pop
p=
rotocol that allow clipmail commands to work. I also
suspect that the "XT=
ND: ON" is what prevents the mailstation
from working with some pop server=
s.
type in your own POP/SMTP servers.
I think what you are calling demo mi=
ght really be just an
unlocked unit. Some mailstations were sold that can=
use
any isp, others are locked into using pop.mailstation.com.
One of my=
mailstations is a demo unit, and it has some
canned messages in it that c=
annot be erased. I do not
think it can be used to get email from a real a=
ccount
(but I never tried).
It can "get mail" even without plugging it in=
to a phone
line, a couple of canned messages are added to the inbox.
Yo=
u can compose new messages, but I'm pretty sure it won't
really send them.=
Whenever you power cycle, any changes
that were made are lost, inbox and=
outbox revert to
original messages.
tween a demo unit and standard unit are different, or is it simply
ting some hardware aspect and locking out the feature from
s? If the code is indeed different, has it ever been
, are the dumps in the Files section from a
I don't kn=
ow if it is a setting, or different firmware
that makes these "demo" units=
, but mine most definitly
says "loading demo data" on the screen when it b=
oots up,
before the cidco logo appears. AFAIK, all the mailstations
have=
same memory capacity (except 350).
After typing the above, it occurs to m=
e that I can't
remember if I ever opened up this "demo" unit. I did
find=
*that* one, in a box with several mis-flashed units :-]
So, I just opene=
d it up, and my first surprise was there
were no speaker wires holding boa=
rd from coming loose
from case. Second was a bit more of a shock, a bunch=
of the chips were missing from board!
The cpu/ram/codeflash/dataflash =
are all there, and same
capacity as normal units, but none of the modem
ci=
rcuitry is there!!!
So, I guess there is no chance of flashing this one
i=
nto a normal mailstation.
The firmware version shown on the settings scr=
een is
2.53YR, which is the same as the one I had been working
with. But=
in the test mode, it shows it as
"version 2.53YR Demo", which seems to in=
dicate it is
special firmware, not just a setting (the missing
components=
point that way, too!)
I put a list of the different models that I am awar=
e
of in groups "database" section. From what I have
gathered, any of the=
various firmware versions can
be locked or unlocked, where the difference=
is whether
you can change the servers or not. I have noticed
that some =
firmware versions that include a "R" at
the end cannot be set to any serve=
r, even with the
"flash test" method. All those without the "R" will
acc=
ept any server after erasing flash. I checked in
the code (of either 2.22=
r or 2.53r), and it has
hardcoded strings that it uses rather than asking =
user
for pop and smtp entries.
I only dumped a 2.53yr (and I am pretty su=
re I verified
it against the dump in file area). I have looked at the
fi=
rst 4 pages of several firmware versions, after
flashing sboot into the hi=
gher pages.
I don't intend to *post* any rom images myself, unless
I get =
permission from Earthlink. The listing for
spew.asm was supposed to make =
downloading a rom image
unnecessary.
interrupts?
Both the normal INT, and nonmaskable NMI vector to the
same =
routine. Port #03 is polled to find source. I can
only speculate on what=
is activating the interrupts, but
the ISR has been studied:
if p3.7 call=
er_id_handler
if p3.5 sets bit 0 of #D631 (whatisit???) (related to rtc???)=
if p3.6 Modem handler
if p3.4 increment time16
if p3.3 null
if p3.0 null
i=
f p3.1 KB scan (just 1 row each int), time32, and 10 timers.
if p3.2 null
=
The odd ordering of the port 3 bits reflects the priority
given to each int=
errupt. Callid is highest priority.
into the codeflash and/or dataflash
Mailbug can use the=
mailstations built in reflash.
That's how we have been installing sboot.=
can either use the routines already there, or someone
red out how to do it from scratch.
Mailbug can also write to codeflash w=
ithout using
the native reflash code, but it needs one of the
various boo=
t progs (mboot, sboot, or bootapp) to
load the remote control stuff (mbug)=
.
Mailbug cannot yet write to the dataflash (that
jogs my memory a bit, t=
hat was the reason 0.0.4
was not yet released).
pdate triggered from the parallel port?
See message #92
(URL)
.yahoo.com/group/mailstation/message/92?var=3D0&l=3D1
t "P2C =3D printer data direction" work? And related
as inputs, does one still need to write 1's to the
o read from it properly?
I think a 1 makes coresponding bit an output????
=
I don't think it matters what was written when it
is an input??? Sorry, it=
has been about three years
since I thought about this.
ind out if there was a device 6? There seems to be a
processor pinout list. If nothing's attached, it
l for adding ones own hardware. Assuming one
iny pins on the board!
It seems to be available. And yes, it would be
h=
ard to connect to it, but not impossible.
From: "Jeff" <fyberoptic1979@...>
Aug 1, 2007
Jones" <cyranojones_lalp@...> =
wrote:
ocked unit. Some mailstations were sold that can use
e locked into using pop.mailstation.com.
emo unit, and it has some
I do not
never tried).
e
compose new messages, but I'm pretty sure it won't
enever you power cycle, any changes
utbox revert to
That's pretty odd then, because I de=
finately know mine came from a
store display, and also has a sticker on it=
saying "This unit is for
demo purposes only and will not send/receive ema=
il" towards the top
left, and a big "NOT FOR RESALE" on the back. Yet I k=
now it sends
email and that the modem components are there, because I've u=
sed it
to receive email in the past.
So I dunno, perhaps this is just an =
unlocked unit as you mentioned,
just being used as a demo unit.
The one y=
ou have though still sounds perfectly good for using it as a
development s=
ystem or other useful thing, as long as it can still be
flashed the same. =
Even if I ever do get an OS onto my own, I doubt
I'll even use the modem =
functionality for anything. I haven't had
dial-up internet access to mess=
with for a while now.
m/group/mailstation/message/92?var=3D0&l=3D1
Oh hey, I totally missed t=
hat message when digging through old ones.
That one has lots of good tidb=
its! Especially what all the functions
are for in the page 1 jump list, w=
hich I asked about in another post.
t an output????
input??? Sorry, it has been about three years
is.
I wrote my own routines earlier to send out bytes/strings and found
t=
hat I never needed to set that input/output port for it to work
correctly.=
Of course it might have already been set during some MS
init code, so I'=
ll have to try setting it both ways before I run my
version of sendbyte, a=
nd see what happens.
From: "Neil Morrison" <neilsmorr@...>
Aug 1, 2007
You could upload them to Rapidshare.com and post links to them here.
If Ear=
thlink bitches Rapidshare will pull them but Earthlink is trying
to bail on=
this system now so I doubt they'll care.
If you want to know more about R=
apidshare, see
(URL)
Neil
n@yahoogroups.com, "Cyrano Jones"
...
't intend to *post* any rom images myself, unless
Earthlink. The listing for
rom image