C preprocessor

[selvesteen]

Hello All,

I came across the following macro in a program, could someone explain
what it actually means.

#define BUFFER_MAX_HPN_LEN (2>>21)-1

Is there any other way to achieve the same

Thanks for the help.
Mike

 

[Chris Dollin]

Hello All,

I came across the following macro in a program, could someone explain
what it actually means.

#define BUFFER_MAX_HPN_LEN (2>>21)-1

It means that wherever BUFFER_MAX_HPN_LEN appears in the
following text, it is replaced by `(2>>21)-1`, which looks
like a typo on someone's part to me, since that evaluates
to -1. Are you sure it wasn't `(2<<21)-1`?

The latter would be a way of writing a mask of 1's from
position 21ish rightwards.

<fx:thinks/>
<fx:assume32BitInts>

2 << 21 = 1 << 22 = 1 followed by 22 0's; -1 then gives
us a mask of 21 ones, which is probably why the macro
would be written using 2<<, as the right operand of <<
would be the number of 1's in the mask.

I'd write `((2 << 21) - 1)` myself.

</>

 

[pete]

Hello All,

I came across the following macro in a program, could someone explain
what it actually means.

#define BUFFER_MAX_HPN_LEN (2>>21)-1

Is there any other way to achieve the same

It looks more like a typographical error, than anything else.

BUFFER_MAX_HPN_LEN * 5 is -5

5 * BUFFER_MAX_HPN_LEN is -1



/* BEGIN new.c */

#include <stdio.h>

#define BUFFER_MAX_HPN_LEN (2>>21)-1

int main(void)
{
printf("\nBUFFER_MAX_HPN_LEN * 5 is %d\n\n",
BUFFER_MAX_HPN_LEN * 5);
printf("5 * BUFFER_MAX_HPN_LEN is %d\n\n",
5 * BUFFER_MAX_HPN_LEN);
return 0;
}

/* END new.c */

 

[CBFalconer]

.... snip ...

2 << 21 = 1 << 22 = 1 followed by 22 0's; -1 then gives
us a mask of 21 ones, which is probably why the macro
would be written using 2<<, as the right operand of <<
would be the number of 1's in the mask.

I'd write `((2 << 21) - 1)` myself.

<nit> or '((2uL << 21) -1)' so it works on 16 bitters. </nit>

 

[snnn]

Hello All,

I came across the following macro in a program, could someone explain
what it actually means.

#define BUFFER_MAX_HPN_LEN (2>>21)-1

Is there any other way to achieve the same

Thanks for the help.
Mike

How should the following codes work?

int x;
x=5*BUFFER_MAX_HPN_LEN
x=BUFFER_MAX_HPN_LEN *5


please define "BUFFER_MAX_HPN_LEN" as "((2>>21)-1)" instead.

 

[pete]

please define "BUFFER_MAX_HPN_LEN" as "((2>>21)-1)" instead.

That's a strange way to write negative one.