CCS C Software and Maintenance Offers
FAQFAQ   FAQForum Help   FAQOfficial CCS Support   SearchSearch  RegisterRegister 

ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 

CCS does not monitor this forum on a regular basis.

Please do not post bug reports on this forum. Send them to support@ccsinfo.com
int32 variable multiplied by 32

 
Post new topic   Reply to topic    CCS Forum Index -> General CCS C Discussion
View previous topic :: View next topic  
Author Message
epitalon
Guest
int32 variable multiplied by 32
PostPosted: Thu Dec 03, 2009 11:07 am     Reply with quote
Hello,

I got a problem with compiler version v4.085

Code:

void my_fct (unsigned int volume)
{
    unsigned int32 volume_32;

    volume_32 = (unsigned int32) volume * 32;
    printf ("value : %Lu\r\n", volume_32);

    volume_32 = 32 * (unsigned int32) volume;
    printf ("value : %Lu\r\n", volume_32);

    volume_32 = (int32) 32 * volume;
    printf ("value : %Lu\r\n", volume_32);
}           


Output of the above code (when "volume" is 200)
value : 6400
value : 0
value : 0

This is obviously a bug from the compiler. This is the fifth bug or more I find in the different versions of PCWH compiler, in a period of 4 years.
For the V4.085 only, this is the second one and I am stuck with a version that seems to be incomplete regarding the USB-hid example code.

Only my license has expired for nearly 9 months. I am stuck. Do you think that CCS will consider that I deserve a new release for free ?
That would be nice.

Recently, I bought the Intel C compiler, a very comprehensive suite for about $400 and some months later, Intel released a version that fixed some bugs and they urged all their customer by email to update for free their compiler !!!

What would be nice from CCS is that they allow me access to a known stable version of their compiler.

Jean-Marie Epitalon
PCM programmer



Joined: 06 Sep 2003
Posts: 21708

View user's profile Send private message

PostPosted: Thu Dec 03, 2009 1:48 pm     Reply with quote
I made your routine into the test program shown below and ran it in
the MPLAB simulator with several versions of the compiler. I used
UART1 to display the results in the Output window. Here's what I got:

PCH 3.249:
value : 6400
value : 6400
value : 6400

PCH 4.068:
value : 6400
value : 320
value : 256

PCH 4.086 (next version after your version)
value : 6400
value : 0
value : 0

PCH 4.099
value : 6400
value : 0
value : 0

PCH 4.100
value : 6400
value : 320
value : 256

PCH 4.101
value : 6400
value : 320
value : 256

Test program:
Code:

#include <18F452.h>
#fuses XT,NOWDT,PUT,BROWNOUT,NOLVP
#use delay(clock=4000000)
#use rs232(baud=9600, xmit=PIN_C6, rcv=PIN_C7, ERRORS)

void my_fct (unsigned int volume)
{
    unsigned int32 volume_32;

    volume_32 = (unsigned int32) volume * 32;
    printf ("value : %Lu\r\n", volume_32);

    volume_32 = 32 * (unsigned int32) volume;
    printf ("value : %Lu\r\n", volume_32);

    volume_32 = (int32) 32 * volume;
    printf ("value : %Lu\r\n", volume_32);
}   

//======================================
void main(void)
{
my_fct(200);

while(1);
}


To make your version give the desired output, I changed the castings
to 'int16', as shown below. Then I got this output:
value : 6400
value : 6400
value : 6400
Code:

void my_fct (unsigned int volume)
{
    unsigned int32 volume_32;

    volume_32 = (int16)volume * 32;
    printf ("value : %Lu\r\n", volume_32);

    volume_32 = 32 * (int16)volume;
    printf ("value : %Lu\r\n", volume_32);

    volume_32 = (int16)32 * volume;
    printf ("value : %Lu\r\n", volume_32);
}           

For background information, Ttelmah has written several comments
on the CCS type promotion philosophy in this thread:
http://www.ccsinfo.com/forum/viewtopic.php?t=29749
FvM



Joined: 27 Aug 2008
Posts: 2337
Location: Germany

View user's profile Send private message

PostPosted: Thu Dec 03, 2009 4:25 pm     Reply with quote
Interestingly, the result is correct (according to the C standard) with PIC24 and PCD.

The considerations about a CCS "type promotion philosophy" sound somewhat like searching a reasonable concept
behind a sloppy designed handling of implicit type conversions. If there's a concept behind it, why has it been
introduced with V4.0, giving up the good V3.0 practice and effectively ignoring clear C rules. (But keep harping on
"full ANSI C compliance" in other regards). In my opinion, it's one of many V4.0 accidents, unfortunately not repaired
til today.

Thanks to PCM programmer for documenting the version differences!
epitalon
Guest
PostPosted: Fri Dec 04, 2009 4:10 am     Reply with quote
Thanks for all your comments.

But I would like to know:
Is CCS likely to give me a new compiler version because I found some bugs ? Did they do that to anyone of you ?
rkinchin
Guest
PostPosted: Fri Dec 04, 2009 8:08 am     Reply with quote
Implicit type promotion is haphazard. It sometimes works and mostly doesn't.
I have had to spend near a week recasting ANSI 'C' into something that PCD would understand. I think this compiler has passed its sell by date, good for quick 8 bit applications but is far too buggy and non standard to be of use for complex 16bit applications.
Display posts from previous:   
Post new topic   Reply to topic    CCS Forum Index -> General CCS C Discussion All times are GMT - 6 Hours
Page 1 of 1

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum


Powered by phpBB © 2001, 2005 phpBB Group