dev-cpp-users

Re: [Dev-C++] Use of const qualifier

[Abhijit S. <mu...@gm...>]

> > Yes. Also if you can make it easy to be inlined things will be more
> > better.
>
> Inlining a function is decided by the compiler. Functions with loops, etc.
> are generally not inlined, so this won't be of much help. You can achieve
> the same end by simply eliminating the function (moving all of the code
into
> the main routine). This will remove the overhead of function calls and
> memory allocation (which should help a lot in this case). This should be a
> case of code-elegance versus execution-speed. Also consider generating
> look-up tables of sine and cos. There are a lot of articles on the web
about
> optimizing programs. I remember one that spoke in explicit detail about
how
> memory can be optimized... unfortunately, I don't remember saving it.
Google
> helps.

This site:

http://www.numerix-dsp.com/c_coding.html

says that static variables help.

___________________________________________________________

Abhijit Shylanath

E-mail:   mu...@gm... || ibr...@bi...
Web-site: http://mudeth.tripod.com/

Re: [Dev-C++] Use of const qualifier

[Per W. <pw...@ia...>]

1 - Declaring a function inline doesn't work for function pointers,
or when writing mixed-language programss.

2 - Eliminating the function to remove the function call obviously doesn't
work either for same reason as above.

3 - The link referenced below is written with DSP processors in mind. Some
optimization work. A lot of the suggestions are questionable on a
general-purpose processor.

/Per W

On Sat, 2 Nov 2002, Abhijit Shylanath wrote:

> 
> 
> > > Yes. Also if you can make it easy to be inlined things will be more
> > > better.
> >
> > Inlining a function is decided by the compiler. Functions with loops, etc.
> > are generally not inlined, so this won't be of much help. You can achieve
> > the same end by simply eliminating the function (moving all of the code
> into
> > the main routine). This will remove the overhead of function calls and
> > memory allocation (which should help a lot in this case). This should be a
> > case of code-elegance versus execution-speed. Also consider generating
> > look-up tables of sine and cos. There are a lot of articles on the web
> about
> > optimizing programs. I remember one that spoke in explicit detail about
> how
> > memory can be optimized... unfortunately, I don't remember saving it.
> Google
> > helps.
> 
> This site:
> 
> http://www.numerix-dsp.com/c_coding.html
> 
> says that static variables help.
> 
> ___________________________________________________________
> 
> Abhijit Shylanath
> 
> E-mail:   mu...@gm... || ibr...@bi...
> Web-site: http://mudeth.tripod.com/
> 
> 
> 
> 
> -------------------------------------------------------
> This sf.net email is sponsored by: See the NEW Palm 
> Tungsten T handheld. Power & Color in a compact size!
> http://ads.sourceforge.net/cgi-bin/redirect.pl?palm0001en
> _______________________________________________
> Dev-cpp-users mailing list
> Dev...@li...
> TO UNSUBSCRIBE: http://www.noicys.cjb.net/devcpp/ub.htm
> https://lists.sourceforge.net/lists/listinfo/dev-cpp-users
> 

Re: [Dev-C++] Use of const qualifier

[Abhijit S. <mu...@gm...>]

> 1 - Declaring a function inline doesn't work for function pointers,
> or when writing mixed-language programss.
>
> 2 - Eliminating the function to remove the function call obviously doesn't
> work either for same reason as above.
>
> 3 - The link referenced below is written with DSP processors in mind. Some
> optimization work. A lot of the suggestions are questionable on a
> general-purpose processor.

Most of the stuff is about C in general, so it should work on any C
compiler. At least the point about inlining is with regard to C, so I
thought I'd link to my source.

___________________________________________________________

Abhijit Shylanath

E-mail:   mu...@gm... || ibr...@bi...
Web-site: http://mudeth.tripod.com/

RE: [Dev-C++] Use of const qualifier

[Ioannis V. <no...@ho...>]

> -----Original Message-----
> From: dev...@li... 
> [mailto:dev...@li...] On Behalf 
> Of Per Westermark
> Sent: Saturday, November 02, 2002 12:27 PM
> To: Abhijit Shylanath
> Cc: 'Dev-C++ Mailing List'
> Subject: Re: [Dev-C++] Use of const qualifier
> 
> 
> 1 - Declaring a function inline doesn't work for function pointers,
> or when writing mixed-language programss.

Usually yes. There is no rule however. For this reason, in C++ it is
better to pass a type argument with static member functions. Or use
function objects.



> 2 - Eliminating the function to remove the function call 
> obviously doesn't
> work either for same reason as above.
> 
> 3 - The link referenced below is written with DSP processors 
> in mind. Some
> optimization work. A lot of the suggestions are questionable on a
> general-purpose processor.


I am very busy these last days so i have no much time to post messages
on the list. Regarding efficiency & C++, the best thing one can do is to
use the high level constructs and do not bother about low level details.
So for example use strings and pass them by reference and do not bother
for the details.

Regarding C, the compiler can perform whatever optimisations it likes.
As a general rule one must try to avoid repeative function calls (like
calling the same function 1 million times). Regarding variables with
exception of large struct types, the time cost of creating and
destroying variables in nowadays machine can be ignored. For a large
struct, using a static struct variable can help. Again, one must think
of run time optimisations when he sees that his code is unacceptable
slow.

Like 

for(size_t  i=0; i<strlen(s); i++)
/* ... */


If the compiler is clever, it may avoid calling strlen(), strlen(s)
times. If we see that the code crawls (and only then), one can think of
optimisations like:

const size_t SLENGTH=strlen(s);

for(size_t i=0; i<SLENGTH; i++)
/* ... */


In general, do not think of optimising something without having a
problem with it. 

Also usually run-time optimisation adds space cost, and space
optimisation adds runtime cost.


Ioannis

* Ioannis Vranos
* Programming pages: http://www.noicys.freeurl.com
* Alternative URL: http://run.to/noicys