Retrieving 32bit Multiply Result

[Nicmus]

Thank you Darrel,

I will add the new code to my program and report the results.

On the refreshing the big_number question:
I’m using a Viniculun device to read an USB storage media that lets me read one byte at the time, n bytes or the entire file in one shot.
The very first line in my data .txt file is the total number of records, 0 - 2,000,000, in fixed length (10 digit left padded with zeros.I used 10 digits to allow for up to 31 bit numbers if need be in the future), decimal format. The rest of the file are the records in fixed length (78 bytes) starting with "#" character, right padded with "~" character and ending with CR/LF. The data is only numerical and I do not process it, I just pass it to the Hserout. Every few records I must pause about 1 second to let the receiver process the information.

I can read the 10 digit number byte by byte, strip the leading zeros and transfer into my big_number variable. How am I going to do it? This is the $10,000,000 question.

I will fix my code right the way and work on the other problem.

I must commend you guys for taking my “what if?” question and make sure I get the best solution. I learned a lot with your help.

Regards,

Nick

[Nicmus]

I added the changes Darrel suggested and it all works fine.
I need to make small changes to the code to better fit my application and i will post it tomorrow.

Thanks again and good night!

Nick

[Darrel Taylor]

If you come back and have it counting from 0 to <strike>1,000,000</strike> File Length ...

I will ..., uh ...,

oh hell, I don't know what I'd do.
<br>

[Nicmus]

Hi Darrel,

Sorry for the delay. I had some hardware problems which I had to fix.
Also I’m using a USB to serial interface to communicate with HT and every now and then Windows will see my device as a mouse.
Here is what I have so far:
I am able to count down and send out the big_number. I made few changes to your code to format the output as a 8 digit number (kept the leading zeros).
I’m also receiving the big_number as two 4 digit decimal numbers.
The only thing left is to make one big_number from the two word size variable (in_high and in_low).
I was thinking to use your PutMultResult?D macro Din but I’m not quite sure how to do it.
Here is the latest working code:

INCLUDE "modedefs.bas"
@ device pic16F876a,hs_osc,pwrt_on, protect_on
DEFINE OSC 16
DEFINE LOADER_USED 1
DEFIne HSER_RCSTA 90h
DEFINE HSER_TXSTA 20h
DEFINE HSER_BAUD 4800
define HSER_CLROERR 1

HighDigits VAR WORD
LowDigits VAR WORD
big_number VAR WORD[2] ; 32-bit variable
in_low var word
in_high var word
led var PORTA.2

ADCON1=6
CMCON=7

ASM
MOVE?CN macro Cin, Nout
MOVE?CW Cin & 0xFFFF, Nout ; Low Word
MOVE?CW (Cin >> 16), Nout + 2 ; High Word
endm
ENDASM
ASM
PutMulResult?N macro Nin
MOVE?WW Nin, R2
MOVE?WW Nin + 2, R0
endm
ENDASM
pause 10000 ' Allow for the USB device to enumerate
hserout [13,10,"Waiting for Big_number",13,10] ' Add time out later
hserin [dec4 in_high,dec4 in_low]
main:
@ MOVE?CN 20000, _big_number
WHILE (big_number[1] > 0) OR (big_number[0] > 0)
big_number[0] = big_number[0] - 1
IF big_number[0] = $ffff THEN big_number[1] = big_number[1] - 1
@ PutMulResult?N _big_number
GOSUB SendBigNum
toggle led ' Show if any action
WEND
pause 5000
goto main
SendBigNum:
HighDigits = DIV32 10000
LowDigits = R2
'IF (HighDigits > 0) THEN
HSEROUT [DEC4 HighDigits, DEC4 LowDigits," and ",dec4 in_high,dec4 in_low,13,10]
'ELSE
'HSEROUT [DEC4 LowDigits,13,10]
'ENDIF
RETURN

End

Thanks again for your patience.

Nick

BTW: I'm practising to post code the right way.

[Darrel Taylor]

I’m also receiving the big_number as two 4 digit decimal numbers.
The only thing left is to make one big_number from the two word size variable (in_high and in_low).

This one I'm not so sure about. And I don't have anything handy to test it on.
I'm pretty sure it will work, but not Positive...

HighDigits = 100   ; Simulating File Length of 1,000,000
LowDigits  = 0000

big_number[0] = HighDigits * 10000
big_number[1] = R0
big_number[0] = big_number[0] + LowDigits
IF (big_number[0] < LowDigits) THEN big_number[1] = big_number[1] + 1

P.S. To put code in a scrolling box that doesn't scrunch it all to the left, use ...
[code]' --- code goes here ---[/code]

oops, nevermind. I see you're even using colors in the test area.
<br>

[Nicmus]

Hi Darrel,

I’m a bit confused by your last post.
I’m not sure were to add this code. Also I have the feeling that the big_number[0] and big_number[1] are switched from the way they are in my code.

Here is the code in “colors” format:

<font color="#000000"><b>INCLUDE </b><font color="#FF0000">&quot;modedefs.bas&quot;
</font><font color="#008000">@	device  pic16F876a,hs_osc,pwrt_on, protect_on
</font><b>DEFINE </b>OSC 16
<b>DEFINE</b>	LOADER_USED	1
<b>DEFINE </b>HSER_RCSTA 90h
<b>DEFINE </b>HSER_TXSTA 20h
<b>DEFINE </b>HSER_BAUD 4800
<b>DEFINE </b>HSER_CLROERR 1

HighDigits  <b>VAR     WORD
</b>LowDigits   <b>VAR     WORD
</b>big_number  <b>VAR     WORD</b>[2]   <font color="#000080"><i>; 32-bit varia
</i></font>in_low      <b>VAR     WORD
</b>in_high     <b>VAR     WORD
</b>led         <b>VAR     </b>PORTA.2

ADCON1=6
CMCON=7	

<b>ASM  
</b><font color="#008000">MOVE?CN  macro Cin, Nout
    MOVE?CW  Cin &amp; 0xFFFF, Nout     </font><font color="#000080"><i>; Low Word
    </i></font><font color="#008000">MOVE?CW  (Cin &gt;&gt; 16), Nout + 2  </font><font color="#000080"><i>; High Word
  </i></font><font color="#008000">endm
</font><b>ENDASM
ASM
</b><font color="#008000">PutMulResult?N  macro Nin
    MOVE?WW  Nin, R2
    MOVE?WW  Nin + 2, R0
  endm
</font><b>ENDASM
    PAUSE </b>1000
    <b>HSEROUT </b>[13,10,<font color="#FF0000">&quot;Waiting for Big_number&quot;</font>,13,10]
    <b>HSERIN </b>[<b>DEC4 </b>in_high,<b>DEC4 </b>in_low]

main:
<font color="#008000">@   MOVE?CN  20000, _big_number
</font><b>WHILE </b>(big_number[1] &gt; 0) <b>OR </b>(big_number[0] &gt; 0)
    big_number[0] = big_number[0] - 1
    <b>IF </b>big_number[0] = $ffff <b>THEN </b>big_number[1] = big_number[1] - 1
<font color="#008000">@   PutMulResult?N  _big_number
    </font><b>GOSUB </b>SendBigNum
    <b>TOGGLE </b>led
<b>WEND
    PAUSE </b>5000
    <b>GOTO </b>main
SendBigNum:
    HighDigits = <b>DIV32 </b>10000
    LowDigits  = R2
    <font color="#000080"><i>'IF (HighDigits &gt; 0) THEN
        </i></font><b>HSEROUT </b>[<b>DEC4 </b>HighDigits, <b>DEC4 </b>LowDigits,<font color="#FF0000">&quot; and &quot;</font>,<b>DEC4 </b>in_high,<b>DEC4 </b>in_low,13,10]
    <font color="#000080"><i>'ELSE
        'HSEROUT [DEC4 LowDigits,13,10]
    'ENDIF
</i></font><b>RETURN

END
</b>

How do I process the in_low and in_high variables to get the big_number before entering the main loop?
Am I pulling them right as DEC4 or should I do it differently?

Regards,

Nick

[Darrel Taylor]

Sorry. Should have used your variable names ...

big_number[0] = in_high * 10000
big_number[1] = R0
big_number[0] = big_number[0] + in_low
IF (big_number[0] < in_low) THEN big_number[1] = big_number[1] + 1

Just stick it after the HSERIN.

P.S. Nice "Colors".