Hi all,
      i am using TMS320C55x DSPLIB Release Version 2.40.00 , when i trie d
to implement some functions like 
1)mul32()
2)sqrt_16
3)add()
4)power()
from the library and i am not able to trace out the results.

In case of mul32(): As it is implemnted for single MAC (17x17 bits), 
it has to choose 17 bits from long datatype(32 bit) for multiplication,
whether it does any other processing on remaining bits? I am not able to
verify the results 

In case of sqrt_16: also I am not able to trace the results. 

In case of add(): results are exactly half of the required results (scale
has been set to 1)

In case of power() results are exactly twice of the expected results.

has anyone implemented the function or know the algorithm actually being
used for these function? 

Thanks in advance,
Rahul.
------------------------------------

OMAP35x EVM jump-starts low-power apps
------------------------------------
The modular and extensible OMAP35x Evaluation Module (EVM) enables developers to start building
applications based on the OMAP35x architecture:http://www.DSPRelated.com/omap35x

I remember the releases always come with a folder with examples for 
each function. The result is verified in the examples. It will be a good 
starting point to try.

Wei

--- In c...@yahoogroups.com, rahul.jadhav@... wrote:
>
> Hi all,
>       i am using TMS320C55x DSPLIB Release Version 2.40.00 , 
when i trie d
> to implement some functions like 
> 1)mul32()
> 2)sqrt_16
> 3)add()
> 4)power()
> from the library and i am not able to trace out the results.
> 
> In case of mul32(): As it is implemnted for single MAC (17x17 bits), 
> it has to choose 17 bits from long datatype(32 bit) for multiplication,
> whether it does any other processing on remaining bits? I am not able 
to
> verify the results 
> 
> In case of sqrt_16: also I am not able to trace the results. 
> 
> In case of add(): results are exactly half of the required results (scale
> has been set to 1)
> 
> In case of power() results are exactly twice of the expected results.
> 
> has anyone implemented the function or know the algorithm actually 
being
> used for these function? 
> 
> Thanks in advance,
> Rahul.
>

------------------------------------

OMAP35x EVM jump-starts low-power apps
------------------------------------
The modular and extensible OMAP35x Evaluation Module (EVM) enables developers to start building
applications based on the OMAP35x architecture:http://www.DSPRelated.com/omap35x
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==

Hello Chenkaishiyi,
               Thanks for the reply, I have gone through ASM codes but not able to get the
significance of some step for example in case of multiplication mul32( ) how to map 32 bit
result with actual 64 bit result?

 Thanks in advance.
Rahul Jadhav
Junior Design Engineer
Spectross Digital System (P) Ltd.
No. 4, Siri Fort Road | New Delhi - 110049
Phone    : +91-9990865914 | 011-26264077
Email       : r...@spectross.com
Web         : www.spectross.com

  ----- Original Message ----- 
  From: chenkaishiyi 
  To: c...@yahoogroups.com 
  Sent: Thursday, August 07, 2008 11:21 AM
  Subject: Re: [c55x] problem in TMS320C55x DSPLIB Release Version 2.40.00
  £Éadvise you to look the assembly code of DSPLIB functions,then you will find out how the
functions works.

  2008-08-07 

  chenkaishiyi 

  ·¢¼þÈË£º r...@spectross.com 
  ·¢ËÍʱ¼ä£º 2008-08-06 20:28:29 
  ÊÕ¼þÈË£º c...@yahoogroups.com 
  ³­ËÍ£º 
  Ö÷Ì⣺ [c55x] problem in TMS320C55x DSPLIB Release Version 2.40.00 

  Hi all,
  i am using TMS320C55x DSPLIB Release Version 2.40.00 , when i trie d
  to implement some functions like 
  1)mul32()
  2)sqrt_16
  3)add()
  4)power()
  from the library and i am not able to trace out the results.

  In case of mul32(): As it is implemnted for single MAC (17x17 bits), 
  it has to choose 17 bits from long datatype(32 bit) for multiplication,
  whether it does any other processing on remaining bits? I am not able to
  verify the results 

  In case of sqrt_16: also I am not able to trace the results. 

  In case of add(): results are exactly half of the required results (scale
  has been set to 1)

  In case of power() results are exactly twice of the expected results.

  has anyone implemented the function or know the algorithm actually being
  used for these function? 

  Thanks in advance,
  Rahul.

------------------------------------------------------------------------------
  No virus found in this incoming message.
  Checked by AVG. 
  Version: 7.5.526 / Virus Database: 270.5.12/1596 - Release Date: 8/6/2008 4:55 PM