amforth-devel — Development discussion and user support

Re: [Amforth] Easy Start without Atmel Studio 6 possible?

[<dev...@de...>]

Am 08.10.2013 20:57, schrieb Matthias Trute:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
>
>> Are this appnotes missing?
> Yes. They are copyrighted by Atmel and thus not part of amforth.
> For the Atmega32 you may have luck with the files from avra, but
> I don't know about their legal status.
>

After downloading and installing of 1 GB compressed stuff i could extract at least the 18MB needed for the assembler!

What i can found within the assembler is this License.rtf for the assembler.

"Redistribution and use in source and binary forms, with or without

modification, are permitted provided that the following conditions are met:
1. Redistributions of source code must retain the above copyright notice,
this list of conditions and the following disclaimer... "

At least the redistribution is PERMITTED when this License and disclaimer is added!
So please DO IT!

But this does not solve the compilation problems ...

Cheers
Karsten

Re: [Amforth] Easy Start without Atmel Studio 6 possible?

[<dev...@de...>]

Am 08.10.2013 20:57, schrieb Matthias Trute:
>
>> Are this appnotes missing?
> Yes. They are copyrighted by Atmel and thus not part of amforth.
> For the Atmega32 you may have luck with the files from avra, but
> I don't know about their legal status.
>
> Matthias
>

O.K. Now i tried to compile with the correct assembler and appnotes.
The standard makefile runs through now! :-)

But there are much errors for the ATMega32:

@/srv/test/amforth-5.1/appl/template$ make template.hex
wine ../../Atmel/avrasm2.exe -I ../../Atmel/Appnotes -I ../../core -I ../../core/devices/atmega32 -fI -v0 -e 
template.eep.hex -l template.lst template.asm
err:winedevice:ServiceMain driver L"IOPort" failed to load
../../core\drivers/usart_0.asm(1): error: Undefined symbol: UBRR0L
../../core\drivers/usart_0.asm(2): error: Undefined symbol: UBRR0H
../../core\drivers/usart_0.asm(3): error: Undefined symbol: UCSR0C
../../core\drivers/usart_0.asm(4): error: Undefined symbol: UCSR0B
../../core\drivers/usart_0.asm(5): error: Undefined symbol: UCSR0A
../../core\drivers/usart_0.asm(12): error: Undefined symbol: RXC0
../../core\drivers/usart_0.asm(13): error: Undefined symbol: UDRE0
../../core\drivers/usart_0.asm(14): error: Undefined symbol: TXEN0
../../core\drivers/usart_0.asm(15): error: Undefined symbol: RXEN0
../../core\drivers/usart_0.asm(16): error: Undefined symbol: RXCIE0
../../core\drivers/usart_0.asm(17): error: Undefined symbol: UDRIE0

Assembly failed, 11 errors, 12 warnings
make: *** [template.hex] Fehler 1


Maybe not the correct definition file for the ATMega32?
Where are this symbols defined?

Cheers
Karsten


Am 08.10.2013 20:53, schrieb Matthias Trute:
> Compiling amforth for a particular setup is not simple. For quite a few
> reason (see documentation). The dependency on the Atmel Studio
> Assembler is only one of them. But not an important one. IMHO.
>
> wrt other assemblers I could tell a long story about avra and
> others (e.g. binutils from gnu)... But things are as they are.

Yes - i can understand this.
It was just an idea / opinion.
You can see the problems with the missing files.

Re: [Amforth] Easy Start without Atmel Studio 6 possible?

[Matthias T. <mt...@we...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1


> Are this appnotes missing?

Yes. They are copyrighted by Atmel and thus not part of amforth.
For the Atmega32 you may have luck with the files from avra, but
I don't know about their legal status.

Matthias

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/

iEYEARECAAYFAlJUVbQACgkQ9bEHdGEMFjOMDwCfT93slwrBApvuq1ieJi8mGpx1
2XwAoL0ieYjddSmCOy6CWoXs8xEO4ClR
=El8D
-----END PGP SIGNATURE-----

Re: [Amforth] Easy Start without Atmel Studio 6 possible?

[Matthias T. <mt...@we...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Karsten,

> i just want to try Amforth and downloaded the sources.

The currently easiest possibility to test amforth is probably
the arduino. For this platform are the hex files precompiled and
there are sketches available to use a 2nd arduino as a programmer
for the amforth programming itself. Details are on the net somehwere.

Compiling amforth for a particular setup is not simple. For quite a few
reason (see documentation). The dependency on the Atmel Studio
Assembler is only one of them. But not an important one. IMHO.

wrt other assemblers I could tell a long story about avra and
others (e.g. binutils from gnu)... But things are as they are.

Matthias

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/

iEYEARECAAYFAlJUVKwACgkQ9bEHdGEMFjOZzgCfT2N9JWJD/zDSi/AUEG2oQLqT
0x0AnRs7RVpFmbYouazOkqxUeqAAOaO/
=AQz0
-----END PGP SIGNATURE-----

[Amforth] gavrasm try (was: Easy Start without Atmel Studio 6 possible?)

[Erich W. <ew....@na...>]

Hi,

On 10/08/2013 07:00 PM, dev...@de... wrote:
> Is this not possible with a smaller solution like this here ?
> http://www.avr-asm-tutorial.net/gavrasm/index_de.html
> http://www.avr-asm-tutorial.net/gavrasm/index_en.html
> 

So I look at this.

1.
This software is written in FreePascal.

2.
./gavrasm -h
+------------------------------------------------------------+
| gavrasm gerd's AVR assembler Version 3.3 (C)2012 by DG4FAC |
+------------------------------------------------------------+
                                                              
Calls: gavrasm [-ABEJLMQSWX] source[.asm]
       gavrasm [-?DHT]                   
                      
List of options:
  -L Listfile off            -M List macros
  -Q Quiet, no screen output -S Symbol list in listfile
  -B Beginners error comment -A Toggle ANSI output on command line
  -E Longer error comments   -D List all supported directives     
  -W Enable wrapping         -T List all supported AVR types 
  -X Internal def.inc off    -J Switch jump PC+X+1 to PC+X  
  -H -? List options (shows this)                         

gavrasm lacks the -I switch to specify a search path for files.

3.
So I cooked together a directory

mkdir tmp.gavrasm
cd tmp.gavrasm
cp ~/Forth/atmega/55_avra_4.6/. .  # this is a working amforth setup from appl/template
cp -a ~/Forth/amforth/releases/4.6/core/. .
cp -a devices/atmega32/. .

./gavrasm main.asm

gavrasm also lacks an option to specify the target controller directly.
I'm not sure that it guessed the correct (any?) target device.

This produces lengthy output, the interesting snippets being in main.err:

gavrasm Gerd's AVR assembler version 3.3 (C)2012 by DG4FAC
----------------------------------------------------------
Source file: main.asm
Hex file:    main.hex
Eeprom file: main.eep
Compiled:    08.10.2013, 19:24:02
Pass:        1
Error ==> .overlap
[device.asm,60] 067: Unknown directive!
Error ==> .nooverlap
[device.asm,102] 067: Unknown directive!
Error ==> .org $0000
[amforth.asm,6] 046: Origin adress (0) points backwards in code segment!
Error ==> .org pc_
[amforth.asm,8] 046: Origin adress (46) points backwards in code segment!
Error ==>     .db "\",0
[words/backslash.asm,6] 088: Illegal string constant ("\")!
Error ==>     .dw 'A'
[words/to-lower.asm,14] 049: No literals allowed in DW directive! Use DB instead!
Error ==>     .dw 'A'
[words/to-lower.asm,14] 050: No parameters found, expected words!
Error ==>     .dw 'Z'
[words/to-lower.asm,16] 049: No literals allowed in DW directive! Use DB instead!
Error ==>     .dw 'Z'
[words/to-lower.asm,16] 050: No parameters found, expected words!
Error ==> .org URXCaddr
[drivers/usart-isr-rx.asm,4] 046: Origin adress (26) points backwards in code segment!
Error ==> .org pc_
[drivers/usart-isr-rx.asm,6] 046: Origin adress (2860) points backwards in code segment!
Error ==>   
[words/umslashmod.asm,16] 079: Line not starting with a label, a directive or a separator!
Error ==>   adiw y, 1
[words/store-i_nrww.asm,96] 017: Register value (0) out of range (R24/26/28/30!)!
Error ==>   sbiw x, 1
[words/store-i_nrww.asm,97] 017: Register value (0) out of range (R24/26/28/30

so, this is as far as I got in 20 Minutes ...

Cheers,
Erich

Re: [Amforth] Easy Start without Atmel Studio 6 possible?

[<dev...@de...>]

Are this appnotes missing?


karsten@PC10:/srv/test/amforth-5.1/appl/template$ make template.hex
wine ../../Atmel/avrasm2.exe -I ../../Atmel/Appnotes2 -I ../../core -I ../../core/devices/atmega1284p -fI -v0 -e 
template.eep.hex -l template.lst template.asm
err:winedevice:ServiceMain driver L"IOPort" failed to load
../../core\macros.asm(133): error: .set directive illegal in macro definition
template.asm(13): ../../core\macros.asm included from here
../../core\macros.asm(138): error: .set directive illegal in macro definition
template.asm(13): ../../core\macros.asm included from here
../../core\macros.asm(144): error: .set directive illegal in macro definition
template.asm(13): ../../core\macros.asm included from here
../../core/devices/atmega1284p\device.asm(6): error: Cannot find include file: m1284Pdef.inc
template.asm(17): ../../core/devices/atmega1284p\device.asm included from here
../../core/devices/atmega1284p\device.asm(48): error: Invalid directive: .overlap
template.asm(17): ../../core/devices/atmega1284p\device.asm included from here

Assembly failed, 5 errors, 0 warnings
make: *** [template.hex] Fehler 1
karsten@PC10:/srv/test/amforth-5.1/appl/template$ err:menubuilder:convert_to_native_icon error 0x80004005 getting frame 6
err:menubuilder:convert_to_native_icon error 0x88982F04 committing encoder


Cheers
Karsten

Re: [Amforth] Easy Start without Atmel Studio 6 possible?

[<dev...@de...>]

Atmel Studio is a pity.
You have to download and install nearly 1 GB of software simply for an assembler. :-(
Atmel Studio requires dotnet and more.
I want to try FORTH and not C!

I could find the assembler here:
http://www.avrfreaks.net/index.php?module=Freaks%20Files&func=viewFile&id=859&showinfo=1
What's about this application notes?

Is it not possible to ask Atmel if it is possible to ship the needed files for this opensource project?

Cheers
Karsten


Am 08.10.2013 19:22, schrieb Artem Kim:
> I only can do this using studio4
> On Tue, 08 Oct 2013 19:00:07 +0200
> "dev...@de..." <dev...@de...> wrote:
>
>> Hello,
>>
>> i just want to try Amforth and downloaded the sources.
>> Now i read that i must have Atmel Studio 6 just for compiling Forth.
>>
>> It seems that only a Assembler is needed, so why download an install a monster package of 638 MB?
>>
>> Is this not possible with a smaller solution like this here ?
>> http://www.avr-asm-tutorial.net/gavrasm/index_de.html
>> http://www.avr-asm-tutorial.net/gavrasm/index_en.html
>>
>> I will try to get it compiled fot an ATMega32, but i cannot understand this way for a small solution.
>> It would be fine just to have a ready compiled image for some standard microcontrollers.
>>
>> Best regards
>> Karsten
>>

Re: [Amforth] Easy Start without Atmel Studio 6 possible?

[Artem K. <art...@gm...>]

I only can do this using studio4
On Tue, 08 Oct 2013 19:00:07 +0200
"dev...@de..." <dev...@de...> wrote:

> Hello,
> 
> i just want to try Amforth and downloaded the sources.
> Now i read that i must have Atmel Studio 6 just for compiling Forth.
> 
> It seems that only a Assembler is needed, so why download an install a monster package of 638 MB?
> 
> Is this not possible with a smaller solution like this here ?
> http://www.avr-asm-tutorial.net/gavrasm/index_de.html
> http://www.avr-asm-tutorial.net/gavrasm/index_en.html
> 
> I will try to get it compiled fot an ATMega32, but i cannot understand this way for a small solution.
> It would be fine just to have a ready compiled image for some standard microcontrollers.
> 
> Best regards
> Karsten
> 

-- 
Artem Kim <art...@gm...>

Re: [Amforth] Easy Start without Atmel Studio 6 possible?

[Erich W. <ew....@na...>]

Hello Karsten,

On 10/08/2013 06:56 PM, dev...@de... wrote:
> Hello,
> 
> i just want to try Amforth and downloaded the sources.
> Now i read that i must have Atmel Studio 6 just for compiling Forth.
> 
> It seems that only a Assembler is needed, so why download an install
> a monster package of 638 MB?
> 
> Is this not possible with a smaller solution like this here ?
> http://www.avr-asm-tutorial.net/gavrasm/index_de.html
> http://www.avr-asm-tutorial.net/gavrasm/index_en.html
> 

Well, you can extract the Assembler from AVR Studio4 as well.
This is "only" 100 MB, but I don't know, if it is still
available somewhere. That's what I use.

*Some versions* of amForth can be assembled with avra as well,
however, this is not for the faint of heart. We have contacted
the avra maintainers, but it seems the only way forward is to
really fix it ourselves. Noone has so far picked this up.

amForth cannot be assembled with as-avr (from the gcc-avr) because
it uses different syntax, unfortunately.

> I will try to get it compiled fot an ATMega32, but i cannot
> understand this way for a small solution.
> It would be fine just to have a ready compiled image for some
> standard microcontrollers.
> 
> Best regards
> Karsten

Yes, this is a weak point of amForth, help is welcome.
I don't know gavrasm, so I have to look at it first.

Cheers,
Erich

[Amforth] Easy Start without Atmel Studio 6 possible?

[<dev...@de...>]

Hello,

i just want to try Amforth and downloaded the sources.
Now i read that i must have Atmel Studio 6 just for compiling Forth.

It seems that only a Assembler is needed, so why download an install a monster package of 638 MB?

Is this not possible with a smaller solution like this here ?
http://www.avr-asm-tutorial.net/gavrasm/index_de.html
http://www.avr-asm-tutorial.net/gavrasm/index_en.html

I will try to get it compiled fot an ATMega32, but i cannot understand this way for a small solution.
It would be fine just to have a ready compiled image for some standard microcontrollers.

Best regards
Karsten

Re: [Amforth] traverse-wordlist

[Matthias T. <mt...@we...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi,


> Please explain. "name>string" seems like the beginning of something
> :-)
> 
> Where are you heading?

Its not me but the forth guru's ;)

> Is it a compiliation utility? Is this described at forth200x.org ?

Just have a look at amforth.sf.net and follow the
link to www.forth200x.org/traverse-wordlist.html

Looks a tool for tool makers, not for application
programmers. I may be wrong, however...

Matthias

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/

iEYEARECAAYFAlJQBzoACgkQ9bEHdGEMFjO2bwCfXHovbIh/kE1eq+axGFSlLIH/
pvsAoLXc8nwI9esVBIJ020D7wdaD29T3
=9hsR
-----END PGP SIGNATURE-----

Re: [Amforth] traverse-wordlist

[Enoch <ix...@ho...>]

Hello Matthias,

Matthias Trute <mt...@we...> writes:
> The name>x words and the concept of the name token
> are far more interesting, but need some more
> thoughts, IMHO.

Please explain. "name>string" seems like the beginning of something :-)

Where are you heading? 
Is it a compiliation utility?
Is this described at forth200x.org ?

Thanks, Enoch.

Re: [Amforth] traverse-wordlist

[Matthias T. <mt...@we...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello,

> While I understand that Forth "high priests" approved the new 
> traverse-wordlist what use is it really for the application
> programmer?

Dont put your fingers in an open wound ;)

The name>x words and the concept of the name token
are far more interesting, but need some more
thoughts, IMHO.

Matthias

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/

iEYEARECAAYFAlJMUKoACgkQ9bEHdGEMFjPoPACg6tNJaSPc46kYAunp0NQ1tpX5
biEAoLsRK8PlVRGoE/XvbLOwOUdO3OGT
=yPXB
-----END PGP SIGNATURE-----

[Amforth] traverse-wordlist

[Enoch <ix...@ho...>]

Hello Matthias,

Thank you for keeping up with the emerging Forth200x standard.

While I understand that Forth "high priests" approved the new
traverse-wordlist what use is it really for the application programmer?

Thanks, Enoch.

Re: [Amforth] AVRDUDE (AVR109 protocol) matching Bootloader

[Enoch <ix...@ho...>]

Hello Matthias,

Matthias Trute <mt...@we...> writes:

>> May I suggest making AmForth a BOOFA bootloader friendly by adding 
>> the following to core/amforth.asm:
>> 
>> .org NRWW_START_ADDR + #ifdef BOOFA			;The BOOFA bootloader + .org 
>> pc+512			; is 512 words max! + #endif .include 
>> "amforth-interpreter.asm"
>
> Looks like a hack to me.

A proud hacker I am ;-)

> First that comes to me is that the bootloader cannot
> change the content of the NRWW area. I think this will be
> troublesome if you want to change code there as well.

"The SPM (Store Program Memory) instruction that writes into the Application Flash
memory section must reside in the Boot Program section."

SPM can be used to write both to the RWW and the NRWW sections (at least
in my µC case).

> For that it would be better to use the amforth-low instead
> of the amforth.asm. This requires some glue code for the
> !i word. And here lies what I call the "hack": both
> boofa and amforth have code to write to the flash.
> But they do not cooperate here and do not share code.
> I think that this can be made in a better way, e.g.
> some forth code that implements the boot loader protocol.

BOOFA is simple code that is unlikely to change. Not so is our lovely
AmForth. It would be best to keep those two projects mindful of each
other but not dependent on each other.

> In a few earlier releases I had a customizable start address
> of the NRWW code part, seems that dropping this was
> not a smart idea ;)

In my application the Start NRWW and the Start Boot Loader
coincide. Through BOOTRST=0 the Reset Vector = Boot Loader Reset passes
control to BOOFA, BOOFA then passes control to AmForth via a JMP 0.

Anyway, BOOFA works well and made my multi µC AmForth project easier to handle
using a single JTAG Mk2 that I don't need any more to plug and replug.

Regards, Enoch.

Re: [Amforth] AVRDUDE (AVR109 protocol) matching Bootloader

[Matthias T. <mt...@we...>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Enoch,

> May I suggest making AmForth a BOOFA bootloader friendly by adding 
> the following to core/amforth.asm:
> 
> .org NRWW_START_ADDR + #ifdef BOOFA			;The BOOFA bootloader + .org 
> pc+512			; is 512 words max! + #endif .include 
> "amforth-interpreter.asm"

Looks like a hack to me.

First that comes to me is that the bootloader cannot
change the content of the NRWW area. I think this will be
troublesome if you want to change code there as well.
For that it would be better to use the amforth-low instead
of the amforth.asm. This requires some glue code for the
!i word. And here lies what I call the "hack": both
boofa and amforth have code to write to the flash.
But they do not cooperate here and do not share code.
I think that this can be made in a better way, e.g.
some forth code that implements the boot loader protocol.

In a few earlier releases I had a customizable start address
of the NRWW code part, seems that dropping this was
not a smart idea ;)

Matthias

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/

iEYEARECAAYFAlI8jYQACgkQ9bEHdGEMFjOWPACfX02onpgjKAe76I8n3Go3yTdh
8NQAnildqOz3q/os/un6WYOgsuP0Rgy1
=JhPf
-----END PGP SIGNATURE-----

[Amforth] AVRDUDE (AVR109 protocol) matching Bootloader

[Enoch <ix...@ho...>]

Hello Matthias & All:

Long time no talk, disconcerting...

May I suggest making AmForth a BOOFA bootloader friendly by adding the
following to core/amforth.asm:

  .org NRWW_START_ADDR
+ #ifdef BOOFA			;The BOOFA bootloader
+ .org	pc+512			; is 512 words max!
+ #endif
  .include "amforth-interpreter.asm"

I'd be happy if BOOFA is offered alongside AmForth on sourceforge as I
am interested in it being tested on other uC architectures. Grab it
from: https://github.com/wexi/boofa

Use BOOFA with the latest available AVRDUDE SVN as in earlier releases
AVRDUDE's butterfly.c is seriously buggy.

Thanks, Enoch.

P/S Thanks to roland-riegel.de whose work I continued.

Re: [Amforth] Making menus to LCD

[Rafael G. <ast...@ya...>]

You could use a Finite State Machine module. AFAIK, there is one included in the latests amforth release
br
Rafael


________________________________
 De: Hannu Vuolasaho <vu...@ms...>
Para: "amf...@li..." <amf...@li...> 
Enviado: Sábado 24 de agosto de 2013 15:55
Asunto: [Amforth] Making menus to LCD
 

Hi!

I'm working with 16x2 LCD and U/D/L/R/C buttons. I was wondering if someone has done the menus smart way.

I believe it could be done with some kind of CREATE DOES> trick which makes them. I'm thinking something like this as an interface :

:noname ." Top row" ; :noname ."Bottom row" ; ' up_action ' down_action ' left_action ' right_action ' center_action add-menu my_new_menu

The actions could be another menus or actions for some data.

Or, There could be array of structs and those items could be added there with add-menu. 

Anyway how this should be taken care? Currently I'm having hard time with ugly spagheti code.

Best regards,
Hannu Vuolasaho                           
------------------------------------------------------------------------------
Introducing Performance Central, a new site from SourceForge and 
AppDynamics. Performance Central is your source for news, insights, 
analysis and resources for efficient Application Performance Management. 
Visit us today!
http://pubads.g.doubleclick.net/gampad/clk?id=48897511&iu=/4140/ostg.clktrk
_______________________________________________
Amforth-devel mailing list for http://amforth.sf.net/
Amf...@li...
https://lists.sourceforge.net/lists/listinfo/amforth-devel

Re: [Amforth] Addressing with @i

[Matthias T. <mt...@we...>]

Hi David,

 > Just noticed something with the @i word with addressing.
>
>
>> : strng s" ABCDEFGHIJKLM" ;
>
> ok
>
>> ' strng 3 + @i .
>
> 4241 ok
>
>> ' strng 4 + @i .
>
> 4443 ok
>
>> ' strng 5 + @i .
>
> 4645 ok
>
>>
>
>
> It looks like there is an implied 'cells' when using @i. Is this correct?

The picture is a bit more complex. First the trivial things:
The Atmegas use a 16bit opcode schema. Every machine instruction
is basically a 16bit number. The program counter register (the
pointer for the machine code) that points to the next machine
instruction is incremented by 1 and points to the next flash cell, at 
16bit distance. So I think its ok to assume, that the flash has
16bit per address unit, there is a 8bit access too, but
that is not the "natural" size of it. EEPROM and RAM have 8bits per
address unit.

When I designed amforth, I wanted to keep the specific
elements. Having a 16bit forth system, its obvious that I use
one address unit to get the flash contents and two consecutive
address units for EEPROM and RAM (Endianess is basically determined
by the few 16bit ports in the IO region). They all end up in a
16bit number. Same with the addresses: a 16bit number is a perfect
target for the controller PC register (aka a "far jump" instruction).

Things got a little bit more complex with the 256x controllers but not
that much.


> (what I actually want is c@i - i.e. read 1 byte)

IMHO the @i should be better named a c@i, since c@ is
defined as the fetch of a single address unit content.
But...

Not the complex stuff: You compile strings into the
flash. If you'd have written a single number, nothing
complicated would have happened ;) Just look at the
example code for e.g. the table driven sine function
in the sources.

Strings consist of many bytes for the characters (the xchar
extensions from forth 2012 are not part of amforth, yet). So I
decided to combine two characters into one flash cell and store them 
that way. That makes words like itype and find(-name) a bit more
complex but the savings of flash space are more important.

HTH
Matthias

[Amforth] @i addressing again

[David W. <ins...@gm...>]

(I meant flash)

[Amforth] @i addressing again

[David W. <ins...@gm...>]

I'm guessing that on the AVR, the EEPROM is addressed as 16-bit words.... ?

[Amforth] Addressing with @i

[David W. <ins...@gm...>]

Just noticed something with the @i word with addressing.


> : strng s" ABCDEFGHIJKLM" ;

ok

> ' strng 3 + @i .

4241 ok

> ' strng 4 + @i .

4443 ok

> ' strng 5 + @i .

4645 ok

>


It looks like there is an implied 'cells' when using @i. Is this correct?

(what I actually want is c@i - i.e. read 1 byte)


Many thanks,


David Wallis.

Re: [Amforth] What files are put in which locations to get 1-wire running on an ATmega2560 platform?

[Matthias T. <mt...@we...>]

hi Nitpickers and not-so-brave-ones ;)

> Hmmm. I don't know. Looking at appl/atmega2561/atmega256.asm says:
>    ; Settings for the avr butterfly demo board
>    .include "macros.asm"
>    .include "device.asm"
>    ...
> You are not using the butterfly board, do you?

It is the proof that porting amforth from one
controller to another one requires to copy
files only. Nevertheless I fixed the comment with
revision 1452 in the repository.

Matthias

[Amforth] Making menus to LCD

[Hannu V. <vu...@ms...>]

Hi!

I'm working with 16x2 LCD and U/D/L/R/C buttons. I was wondering if someone has done the menus smart way.

I believe it could be done with some kind of CREATE DOES> trick which makes them. I'm thinking something like this as an interface :

:noname ." Top row" ; :noname ."Bottom row" ; ' up_action ' down_action ' left_action ' right_action ' center_action add-menu my_new_menu

The actions could be another menus or actions for some data.

Or, There could be array of structs and those items could be added there with add-menu. 

Anyway how this should be taken care? Currently I'm having hard time with ugly spagheti code.

Best regards,
Hannu Vuolasaho 		 	   		  

Re: [Amforth] What files are put in which locations to get 1-wire running on an ATmega2560 platform?

[Michael P. <mp...@rc...>]

Forgot to mention ... I am only using the 2560 and an Uno ... no
Butterfly in use at all.

On 8/16/2013 10:39 AM, Erich Waelde wrote:
> Hello Michael,
>
> On 08/16/2013 01:01 AM, Michael Picco wrote:
>> Hello Erich,
>>
>> Thank you for the reply.  Can I safely assume the template file can be
>> copied/pasted to my new project and used as-is for 1wire?
>> The other question is about the other files required:
>> dict_appl
>> dict_appl_core
>> etc.
>> Should I be using the ones in the atmega2561 directory for creation of
>> atmega2560 project?
> Hmmm. I don't know. Looking at appl/atmega2561/atmega256.asm says:
>   ; Settings for the avr butterfly demo board
>   .include "macros.asm"
>   .include "device.asm"
>   ...
> You are not using the butterfly board, do you? Allthough I cannot
> exclude the possibility, that this is a leftover comment from
> somewhere else [1], I personally would start with a copy of
>   appl/template/
> and go from there. Crystal frequency and baud rate go into
>   appl/template/template.asm
> The device type goes into
>   appl/template/makefile
>
> If you are using Windows/Atmel Studio then consult Karl Lunt's
> documentation listet ond the amforth webpage [2] (I cannot help
> you out with Windows/Atmel Studio, sorry).
>
> Hope this helps,
> Erich
>
> [1] there is also the directory appl/avr-butterfly
>     for a project for that particular device.
> [2] http://amforth.sourceforge.net/ and
>     http://amforth.sourceforge.net/UG/amforth_user.html#user-guide
>
>
>
> ------------------------------------------------------------------------------
> Get 100% visibility into Java/.NET code with AppDynamics Lite!
> It's a free troubleshooting tool designed for production.
> Get down to code-level detail for bottlenecks, with <2% overhead. 
> Download for free and get started troubleshooting in minutes. 
> http://pubads.g.doubleclick.net/gampad/clk?id=48897031&iu=/4140/ostg.clktrk
> _______________________________________________
> Amforth-devel mailing list for http://amforth.sf.net/
> Amf...@li...
> https://lists.sourceforge.net/lists/listinfo/amforth-devel