Now I've changed the fuses to:
l=F7 h=D9 e=FF (full swing osc., slow rising)
crystal = 20MHz (comm at 19k2 as the usart settings routine is quite
intolerant as you know..):
> ver
amforth 4.0 ATmega1284P ok
> primes
1 iterations.
6412 primes found, ok
>
Sieve: size 32000, 1 iteration done in 6s
11.05920 MHz. I may do experiments with overclocking today.. Pito
- PŮVODNÍ ZPRÁVA -
Od: [email protected]
Komu: [email protected]
Předmět: Re: [Amforth-devel] 1284p compilation
Datum: 28.8.2010 - 1:03:51
> OH something I forgot to ask. What crystal
> frequency are you
OH something I forgot to ask. What crystal frequency are you running it
at ??
Cheers
Andy Kirby
On 27/08/10 20:38, pito wrote:
>> words
> TIMER3_OVFAddr TIMER3_COMPBAddr TIMER3_COMPAAddr TIMER3_CAPTAddr
> USART1_TXAddr USART1_UDREAddr USART1_RXAddr SPM_READYAddr TWIAddr
> EE_READYAddr ADCAddr AN
Wow well done pito. That's one job sorted.
On 27/08/10 20:38, pito wrote:
>> words
> TIMER3_OVFAddr TIMER3_COMPBAddr TIMER3_COMPAAddr TIMER3_CAPTAddr
> USART1_TXAddr USART1_UDREAddr USART1_RXAddr SPM_READYAddr TWIAddr
> EE_READYAddr ADCAddr ANALOG_COMPAddr USART0_TXAddr USART0_UDREAddr
> USART0_
Leon,
e.g.?
result = ((1/3) * 3) + 1.0
> 1 s>f 3 s>f f/ 3 s>f f* 1 s>f f+ f>s .
^
?? -42 45
>
P.
- PŮVODNÍ ZPRÁVA -
Od: "Leon Nathaniel Maurer"
Komu: [email protected]
Předmět: Re: [Amforth-devel] Leon's float
Datum: 27.8.2010 - 23:32:34
> Got it.
>
Hi,
is this stack address correct (decimal)?
> .s
ok
> 123
ok
> .s
0 16557 123
ok
> 456
ok
> .s
0 16555 456
1 16557 123
ok
>
p.
--
Sell apps to millions through the Intel(R) Atom(Tm) Developer Program
Be part of this
Got it.
If all you want to test is that f+, f-, f*, and f/ do what they're supposed to
do, then make floats from single length integers with "s>f", then do what you
want and convert them back to integers with "f>s" (that doesn't round -- it
just truncates the part after the decimal point).
For
s>f defined after the first usageYou have to relocate the
Conversion words on the better place within the code(:-)). After the
second upload it has been taken. So now how to test??? I have no
idea... Float, float..P/
- PŮVODNÍ ZPRÁVA -
Od: "pito"
Komu: [email protected]
f/ isn't ready yet -- I'm actually working on it now (I wrote it when away
from my device).
I'd just remove f/ from the upload, I think the rest is good to go.
-Leon
On Friday, August 27, 2010 04:03:31 pm pito wrote:
> Hi Leon,
> I've tried to upload to my new 1284p, but getting this error:
>
>
Hi Leon,
I've tried to upload to my new 1284p, but getting this error:
.
> : f/ ( f1 f2 -- f1/f2 )
okfdup f0= abort" division by zero "
okfzeroexponent >r ( f1 f2, R: n2 )
okfswap fzeroexponent r> - >r fswap ( f1 f2, R: n1-n2 )
okf0 frot frot ( sum remainder divisor, R: exponent )
ok[ 1 s>f ]
> words
TIMER3_OVFAddr TIMER3_COMPBAddr TIMER3_COMPAAddr TIMER3_CAPTAddr
USART1_TXAddr USART1_UDREAddr USART1_RXAddr SPM_READYAddr TWIAddr
EE_READYAddr ADCAddr ANALOG_COMPAddr USART0_TXAddr USART0_UDREAddr
USART0_RXAddr SPI_STCAddr TIMER0_OVFAddr TIMER0_COMPBAddr
TIMER0_COMPAAddr TIMER1_OVFAddr TIM
size 16384, ~5sek @11.059MHz
> primes
1 iterations.
3511 primes found, ok
>
size 32000, ~11sek @11.059MHz
> primes
1 iterations.
6412 primes found, ok
>
- PŮVODNÍ ZPRÁVA -
Od: "pito"
Komu: [email protected], [email protected],
[email protected]
Předmět: Re: [Amforth-de
> .res
free FLASH cells 58231
free RAM cells16080
used EEPROM cells 62
used data stack cells 0
used return stack 10
free return stack 70
ok
> Cool
PS: shall be "free RAM bytes"
Pito.
- PŮVODNÍ ZPRÁVA -
Od: "pito"
Komu: [email protected], [email protected],
amforth
RESULT:
Fuses: l=FF h=D9 e=FF
source burned with pony-stk200 via lpt1 and sinaprog (avrdude gui):
>
ok
> ver
amforth 4.0 ATmega1284P ok
> words
i@ (i!) i! e@ e! not s>d up! up@ >< cmove> unloop i sp! sp@ rp! rp@
+! rshift lshift 1- 1+ xor or and 2* 2/ invert um* um/mod m* + -
log2 d< d> 0> u> u< t
With (template.asm based on Matthias):
; initial baud rate of terminal
.equ BAUD = 115200
.set USART_B_VALUE = (1<
Komu: "Everything around amforth"
Předmět: Re: [Amforth-devel] 1284p compilation
Datum: 27.8.2010 - 19:43:36
> Am Freitag, den 27.08.2010, 17:24 +0200 schrieb
> pito:
> > FYI - res
Am Freitag, den 27.08.2010, 17:24 +0200 schrieb pito:
> FYI - result of the compilation for 1284p (avr studio 4.18), device
> from trunk:
>
> AVRASM: AVR macro assembler 2.1.42 (build 1796 Sep 15 2009 10:48:36)
> Copyright (C) 1995-2009 ATMEL Corporation
Thats the version I use too. so far ok
>
Looks similar to some of the initial errors that ere weeded out with the
arduino boards.
Basicaly mismatched naming of constants form for uart reg's and bit fields.
Will have a look when i get to it.
I seem to recal it meant adding numbers to uart defs and a minor rename.
Avoid altering the m
Just noticed the memory defs are a little out too.
The way i would be tacklingb this is by using the rlevant files for the nearest
device (periphs) and then patchin the mem values for something sensible from
the datasheets. Plus inthis casde stealing the mem access macro statments from
another
FYI - result of the compilation for 1284p (avr studio 4.18), device
from trunk:
AVRASM: AVR macro assembler 2.1.42 (build 1796 Sep 15 2009 10:48:36)
Copyright (C) 1995-2009 ATMEL Corporation
...
C:\MyCode\AVR\WINAVR\projects\AMFORTH\core\drivers/usart.asm(2):
warning: Use of undefined or forward
19 matches
Mail list logo
