any value conversion to 32 bit vector form in vhdl.please help

Try this:

1

library IEEE;

2

use IEEE.STD_LOGIC_1164.ALL;

3

use IEEE.NUMERIC_STD.ALL;

4

:

5

signal vector std_logic_vector(31 downto 0);

6

:

7

vector <= std_logic_vector(to_unsigned(34,32)); -- here your value 34 is converted to a 32 bit vector

BTW:
WHERE does this value come from?
And whats your actual problem?
I ask this because I think you asked the wrong question...

thank you so much...
its a long program...
if u want to have a look than i can send...but u understand my logic 
right...

but the value of A changes very instant of time...
cant we put directly A inplace of 34
means
vector <= std_logic_vector(to_unsigned(A,32));
like wise...

regards

please tell me one thing more...if vector define by you, is variable..

means vector is not signal,its variable..can this convergence is 
possible..

regards

> vector is not signal,its variable..
What do you mean?
Mark with a x:
[ ] the vector is a variable.
[ ] the input value A is variable, it changes.

> if u want to have a look than i can send...
Pls post it here as a *.vhdl file attachment.

sending the word file document...
error at the last page...at K_temp

> Pls post it here as a *.vhdl file attachment.
Try it again as *.vhd  or *.vhdl, NOT AS a Microsoft Word *.docx file!

ok...
i am sending the VHD extension file...
thanks for your concern really...

regards

this is zip file, containing everything...

> .docx
> .ise
> .rar
> zip
I cannot help you with those files, because I'm on a smartphone and 
can't open either one of them...

So just attach the according *.vhd or *.vhdl file. Is it that difficult?

ok...
this is the file extension .vhd..
please sought out my problem..
problem is at the K_temp

thanks

regards

1

--error is here-"Operator <INVALID OPERATOR> must have constant operands or first operand must be power of 2

2

K_temp := "0100000000000000000000000000000000" /A ;

Oooops, a divider. Is this design for a FPGA?

You can simulate a division easily.
But if you want it to run on real hardware, you will get immediately 
into trouble, if the divisor isn't a power of 2 (like 2, 4, 8, 16, 
32...).

Because if the divisor is a power of 2, then the division is just a 
different wiring inside the FPGA. Thats easy, and its done by the 
synthesizer on its won.

If the divisor is not a power of 2, then the division gets fairly 
complex. You then usually involve a divider-core, plus you will have to 
read a manual of at least 20 pages...   :-o


BTW:

1

Use IEEE .numeric_std .All ;          -- take his one

2

3

use IEEE.std_logic_signed.all;        -- or those old osolete ones here

4

--use IEEE.std_logic_unsigned.all;

5

use IEEE.std_logic_arith.all;

6

7

                                      -- but never all of them together!!!

BTW2:

1

-- ***********************

2

--* Begin main process . *

3

-- ***********************

4

Begin

5

Process ( z_position , reset ) Is

1. There is no "main process". Every process has the very same 
proirity, because every process will be synthesized to hardware.

2. The reset is absolutely unnecessary. Because even with reset='1' the 
reset-values of x and P will be overwritten immediately in the first two 
lines after the "end if;".

BTW3:

1

x :=

2

("00000000000000010000000000000000",

3

"00000000000000010000000000000000");

For better readability try this:

1

x := (x"00010000",x"00010000");

Or this:

1

x := (to_signed(65536,32),to_signed(65536,32));

BTW4:
>>> Begin main process
This whole design looks like you are "thinking in software", this will 
not lead to a solution for a language that "describes hardware"!

So what to do now?
should i use ipcore?
or one thing is in my mind, tell me is it possible
in K_temp := "0100000000000000000000000000000000" /A ;
first i convert "0100000000000000000000000000000000" into integer
and then convert A into integer..
then divide both bcoz they are in integer now..
after that the resultant of the divison be converted back into signed...

please help me out..

thanks

best regards

> or one thing is in my mind, tell me is it possible ....
> then divide both bcoz they are in integer now..
Absolutely NO.
Because casting and converting does not change one thing at the fact:
you want to do a division.

You're tricking around and hoping for a hidden way.
That will not work here...

> should i use ipcore?
I think you will have to. And: read ALL of the documentation.

thanks a lot....

i would go for ipcore..if i have any problem, i know who can help me 
now...


best reagrds....

thanks...
i have implemented the divisor core...
the whole program is working well now...

thanks a ton really..

regards...

> the whole program is working well now...
Once more (the last time): you do not program a FPGA.
You're writing a hardware description for a FPGA.

> thanks a ton really..
You're welcome.

Yes...
now implementing on FPGA...


Regards...