<tp>
Ive started testing compatibility with Zeptoforth and my libraries and it's looking good
<tabemann>
bbit@, hbit@, and bit@ as I implemented them return -1 if any masked bit is 1
<tp>
tabemann, what word controls compiling to flash ?
<tabemann>
compile-to-flash
<tabemann>
and compile-to-ram
<tp>
i see I'm compiling to ram by default
<tabemann>
yes
<tabemann>
there's also compiling-to-flash?
<tp>
so my next step is to see if I can load a ton of dev words
<tp>
ah yes
<tp>
I can remember those :)
<tabemann>
if you want to load lots of code, I suggest compiling to flash
<tp>
Zeptoforth is looking good so far
<tp>
yeah, of course
<tp>
1MB of flash
<tabemann>
I remember recently I just tried to load some test code, and it took up like $D00
<tabemann>
and it seemed like the code itself was pretty small
<tabemann>
this was compiling to ram
<tabemann>
for some reason compiling to flash seems more efficient
<tp>
ill be writing some Forth code and testing it then dissasembling and adding to your kernel code
<tabemann>
if you have compressing mode enabled
<tp>
well there is just a lot more flash, 1MB vs 190KB of ram on my F407 iirc
<tabemann>
because compressing mode turns off the padding after each word unless you call commit-flash
<tp>
hmm
<tp>
thats new to me
<tabemann>
note that if you want to refer to a word compiled since the last commit-flash, you should do another commit-flash
<tp>
I dont want to fit everything in the flash anyway, just the peripherals I'll be developing for
<tp>
ill probably be fine with non compression
<tp>
if I run into problems I'll try it
<tp>
my first effort is to add RTS handshaking as you know
<tabemann>
compress-flash (followed when done by end-compress-flash) can save significant amounts of space when compiling many words
<tp>
if I get it working I'll send you a patch for 0.5.4
<tabemann>
cool
<tp>
oh ok, I'll test compress-flash and see what I get
<tp>
I have tons of ready made words I can play with
<tabemann>
and remember to use commit-flash if you're using compress-flash
<tp>
see-for-gas looks awesome, I'll be using that soon I hope
<tp>
ok, I've made a note of that, thanks
<tabemann>
and you have to have to use end-compress-flash when you're done, as if you reboot without compiling anything more to flash afterwards, your code won't show up when you reboot
<tp>
got it
<tp>
I'm used to 10001 screw ups, it's no biggie
<tp>
I learn by screwing up
<tabemann>
in essence compress-flash disables padding at the end of words and the DEADBEEF marker indicating the end of the last word in flash
<tp>
how does Mecrisp-Stellaris tell the end of the dictionary ?
<tp>
you use DEADBEEF
<tabemann>
the padding is to ensure that the flash write buffer gets dumped out to flash, due to the block size of the flash
<tp>
well I know how Mecrisp-Stellaris does it from my ihex words
<tp>
ahh, flash again!
<tabemann>
how does Mecrisp-Stellaris do it?
<tp>
it looks for the last BL etc
<tabemann>
also, at the end of the last word, there is a pointer to the head of the last word
<tp>
tabemann, now crest is threatening to buy a Numworks calculator
<tabemann>
I don't remember if compress-flash removes that
<tabemann>
so I saw about :)
<tp>
tabemann, youve started a thing ...
<tabemann>
well it's a neat calculator that just happens to need a neat kernel and operating environment
<tabemann>
:D
<tp>
it is neat hardware
<tp>
tabemann, plus Ive started documenting forth upload speeds
<tp>
it's made me realise that Mecrisp-Stellaris needs a in-kernel Systick to be useful
<tp>
so I'll make one up in assembly and begin the long process of trying to persuade matthias to include it
<tabemann>
hmm zeptoforth has a systick but not an in-kernel one
<tabemann>
but its systick is sufficiently light-weight that it should function fine IMO
<tp>
a systick facility would be very handy for timings
<tp>
and it needs to be in the kernel to work at boot up
<tp>
it's a small thing, but I think a essential tool
<tabemann>
in zeptoforth, once compiled in, it's initialized at boot time
<tp>
yeah
<tp>
oh
<tp>
whereas Mecrisp-Stellaris need a init word
<tabemann>
well zeptoforth does use an init word
<tabemann>
when the kernel boots, it searchs the dictionary for the last word named "init" and then executes it
<tabemann>
or do you mean that he user has to invoke an initialization word manually?
<tp>
no, it's the same as you
<tp>
if 'init' is there it will be executed at bootup
<tabemann>
I just checked the banlist for some reason, and noticed there were a ton of people all banned within the same second on May 11th, 2020
<tp>
the channel banlist ?
<tabemann>
yeah
<tp>
probably a list was uploaded or something like that ?
<tp>
or perhaps freenode has some auto banner ?
<tabemann>
something like that
<tabemann>
maybe it was a bot maintaining its own banlist
<tp>
maybe
<tabemann>
like ChanServ
<tp>
no end of trolls that need banning tho
<tp>
it's ongoing, kinda sad really that some people waste their lives and the lives of others that way
<tabemann>
yeah
<tp>
it may have just banned ip's with some bonafide problem
<tp>
so when soes the Numworks arrive ?
<tp>
does
<tabemann>
it says it's "on its way" today, doesn't give an arrival date ATM
<tp>
wow, stuff happens so fast in the USA
<tp>
see the inaugural Demo2 flight was scrubbed due to weather ?
<tabemann>
so I hear
<tp>
only had 30 minutes to lift off remaining
<tp>
moved to saturday 4pm ET
<tabemann>
yeah I heard that too
<tp>
I guess the whole world has :)
<tabemann>
okay, by on its way they seem to mean that it has left the warehouse and not reached USPS yet... even though that means that by this point they have reached USPS but have not updated their tracking yet
<tp>
usual stuff
<tp>
i was pleased to see that my print formatting all worked fine on Zeptoforth
<tp>
makes it easy to test the rest of my setup
<tabemann>
nice
<tp>
good work :)
<tabemann>
if one sticks to ANS it shouldn't be hard to write compatible code, aside from a few words named differently like b@ b! b, rather than c@ c! c,
<tabemann>
which for some reason I decided to name differently - lol
<tabemann>
(well, it's because I didn't want to imply that characters were one byte long, because in this day and age of UTF-8 they're not)
<tp>
easy for a user to alter if they want
<tabemann>
okay, I'm gonna hit the sack now
<tabemann>
g'night
<tp>
nighto!
reepca has quit [Ping timeout: 272 seconds]
jsoft has joined #forth
gravicappa has joined #forth
va_ has quit [Remote host closed the connection]
reepca has joined #forth
Zarutian_HTC has quit [Read error: No route to host]
Zarutian_HTC has joined #forth
WickedShell has quit [Remote host closed the connection]
dys has joined #forth
reepca has quit [Remote host closed the connection]
reepca has joined #forth
karswell_ has quit [Remote host closed the connection]
karswell_ has joined #forth
boru has quit [Quit: brb; patches]
reepca has quit [Ping timeout: 260 seconds]
boru has joined #forth
andrei-n has joined #forth
iyzsong has joined #forth
dave0 has quit [Quit: dave's not here]
jsoft has quit [Ping timeout: 246 seconds]
reepca has joined #forth
reepca has quit [Remote host closed the connection]