AVR Freaks

Wow!  I didn't expect a response on New Year's day.  Thanks Guys - I had imagined you all did this as a profession, and would be off work today.

The chip is an ATtiny85.  I should also have mentioned that I'm programming in assembler.

Thanks for the info, meolsen.  This takes me into unfamiliar territory, but I'll read your link and let you know where it gets me.

Mike.

Well, I've spent some time trying to figure it out, but I'm not getting anywhere.

The article linked to by meolsen talks about writing stuff into the ELF file.  At first, I examined the files in my project to see if there was a <project name>.elf, then I looked up ELF as an acronym and discovered that it is an Executable Linker File.  The article talks about including .h header files, which is a C-language concept, but in fairness, I had not said at that point that I was programming in assembler.

Clawson, you asked which assembler - and in truth, I don't know.  Maybe it would help if I show the code.  I'm trying to write a byte into flash,using a routine I picked up by searching for it.  I'm pretty sure I know how to do it, or can work it out from what I already have, but it's not happening in the AS7 simulator.  I haven't tried writing it to a chip yet (AT85), because when I get it working, I will be building it into my project.

The routine doesn't actually run - it always ends in an error in rdloop:  but I don't care about that.  I can single-step enough of it to show me what to do, and I can see that the page buffer does not get written to flash.  I may be wrong, but I surmise that it's because the simulator thinks the SELFPRGEN fuse is not set in the extended fuse byte.  Otherwise, there's something wrong with the program, but as I found it on this site, I think it probably works.

The definitions file (tn85def.inc)  contains ...

; EXTENDED fuse bits
.equ    SELFPRGEN    = 0    ; Self-Programming Enable

but the '0' is not a value - just the number of the bit in the Extended Fuse byte.  The file also equates

; LOW fuse bits
.equ    CKSEL0    = 0    ; Select Clock source

; HIGH fuse bits
.equ    BODLEVEL0    = 0    ; Brown-out Detector trigger level

However, I have not been able to find an equate for the fuse bytes themselves - only the bits.  If I could say something like

EFUSE = 0 << SELFPRGEN    

or simply

EFUSE = 0              ... that would help, but EFUSE is not in the file.

Note:  I'm not asking or help with the program.  I only included it to answer the question "Which assembler"

Also, to allay questions about writing to flash, the application is a mains on-timer to prevent me leaving the garage light on for a week at a time till I need the car again,or my coffee machine switched on for hours after use.  Once the timer has been programmed, the number of seconds (2 bytes - 7FFF secs = 9 hours) will be written to NVR so that it will survive a power down.  The limit on the number of writes to NVR doesn't come into it.  Once it is programmed, I expect it to be used for a long time without being changed again.

I also know that timers can be bought - but I want to make one.

 ;This example shows SPM write of one page for devices with page write
 ;- the routine writes one page of data from RAM to Flash
 ; the first data location in RAM is pointed to by the Y pointer
 ; the first data location in Flash is pointed to by the Z pointer
 ;- error handling is not included
 ;- the routine must be placed inside the boot space
 ; (at least the do_spm sub routine)
 ;- registers used: r0, r1, temp1, temp2, looplo, loophi, spmcrval
 ; (temp1, temp2, looplo, loophi, spmcrval must be defined by the user)
 ; storing and restoring of registers is not included in the routine
 ; register usage can be optimized at the expense of code size
 .equ PAGESIZEB = (2*PAGESIZE) ;PAGESIZEB is page size in BYTES, not words
 
.org 0x00
ldi R28,0x00 ; YL - addresses data memory (0x0100)
ldi R29,0x01 ; YH

ldi R30,0x00 ; ZL - addresses Flash (0x0600)
ldi R31,0x06 ; ZH

ldi r16,0x69
st Y,R16     ; write to data mem add 0x100

 write_page:
;page erase
ldi R22, (1<<PGERS) + (1<<SPMEN)  ; PGERS is Page Erase
rcall do_spm
;transfer data from RAM to Flash page buffer
ldi R20, low(PAGESIZEB) ;init loop variable
ldi R21, high(PAGESIZEB) ;not required for PAGESIZEB<=256
wrloop:
ld r0, Y+
ld r1, Y+
ldi R22, (1<<SPMEN)
rcall do_spm
adiw ZH:ZL, 2
subi R20, 2
sbci R21,0
brne wrloop
 ;execute page write
subi ZL, low(PAGESIZEB) ;restore pointer
sbci ZH, high(PAGESIZEB) ;not required for PAGESIZEB<=256
ldi R22, (1<<PGWRT) + (1<<SPMEN)
rcall do_spm
;read back and check, optional
ldi R20, low(PAGESIZEB) ;init loop variable
ldi R21, high(PAGESIZEB) ;not required for PAGESIZEB<=256
subi YL, low(PAGESIZEB) ;restore pointer
sbci YH, high(PAGESIZEB)
rdloop:
lpm r0,Z+
ld r1, Y+
cpse r0, r1
rjmp error
subi R20,2
sbci R21,0
brne rdloop
 ;return
ret
 do_spm:
 ;input: R22 determines SPM action
 ;disable interrupts if enabled, store status
in R15, SREG
cli
;check for previous SPM complete
 wait:
in R16, SPMCSR
sbrc R16, SPMEN
rjmp wait
;SPM timed sequence
out SPMCSR, R22
spm
 ;restore SREG (to enable interrupts if originally enabled)
out SREG, R15
ret

error:
rjmp error

I think it may be possible to set fuses with a Stimuli file.   See   http://www.atmel.com/webdoc/simu...

Under Assignments, it says the general statement syntax is

target = value     where Target is the numerical memory address of an I/O register in the I/O map.  For simple devices with flat I/O structure such as tinyAVR® and megaAVR®, the register name as found in the data sheet can also be used.

Under Directives, it gives a directive "$fuse" with an argument of "address value" and a comment "Set fuse byte at address to value.  Fuse addresses generally start at 0."

If there is something in the AT85 datasheet that gives an address of the fuse bytes, then I must have missed it.  It mentions the I/O map in paragraph 3.2 of the datasheet, but very briefly, with no helpful information.

Under "Known Issues in Simulator", it says "Setting fuses in the code of the simulated program may interfere with the operation of the simulator. " - so it seems to be possible, but I can't find out how.

Can anybody tell me what to do to set this fuse?  It's probably right in front of my face, but I'm not seeing it.

Thank you, John.

When I came to the AVR scene, it seemed intuitive that AS7 would be the more capable package - 7 is a higher number than 4 or 6, so must be later software.

However, software does sometimes lose its way.  I often wish I could go back to Windows XP.  It seems to me that every level since then has added bells & whistles at the expense of usability.

If the experts have stuck with AS4, then it's clearly the way to go.  I'll load it.

BTW, is there a development history of Atmel Studio, where I could read what the revisions do?

Edit: On second thoughts, don't answer that - Google finds quite a number of links to that question.