CX51: USING 24-BIT MATH WITH FAR POINTERS

Information in this article applies to:

• C51 Version 6.23

QUESTION

Is there a way to access the linear address space of the Philips 51MX devices using pointers? I've tried the following code and the pointer math is truncated to 16-bits.

void func (void)
{
unsigned char far *ptr;

for (ptr = 0x020000; ptr != 0x080000; ptr++)
  {
  *ptr = 0xFF;
  }
}

Is there a way to do what I want?

ANSWER

Yes. This features is added in C51 Version 7. You may download the latest updates from the Keil Website.

For 24-bit pointer arithmetic, you must use long types in your pointer calculations. For example, the following code works:

void func (void)
{
unsigned char far *ptr;

for (ptr = (void far*) 0x020000; ptr != (void far*) 0x080000; ptr+=1UL)
  {
  *ptr = 0xFF;
  }
}

The Keil CX51 C Compiler limits individual objects to 64K in size. So, pointer accesses using normal data objects are restricted to a single segment (64K) and 16-bit math is used.

By forcing long pointer arithmetic, you signal the compiler that you want 24-bit address calculations to be performed.

MORE INFORMATION

• Application Note 160: Programming the Philips 51MX Architecture with the Keil PK51.
• Refer to Far Memory in the Cx51 User's Guide.

SEE ALSO

• CX51: ARRAY INDEX ARITHMETIC
• C51: CONSTANT ARRAYS LARGER THAN 64KB

FORUM THREADS

The following Discussion Forum threads may provide information related to this topic.

• Reinhard, Jon can you answer: Struct across 64 boundary
• passing 64k boundary

Last Reviewed: Monday, January 28, 2008