Thanks to visit codestin.com
Credit goes to www.scribd.com

Scribd
Upload
68 views4 pages

Example Floating Point Problems: Problem 1

This document presents three example floating point problems. Problem 1 describes a C program that stores different data types in a union and prints the resulting bytes, asking the reader to fill in the output. Problem 2 describes a 7-bit floating point format and asks the reader to determine exponent and significand values for normalized and denormalized numbers, and fill in a table encoding various values. Problem 3 describes an 8-bit floating point format and asks the reader to fill in exponent and significand values in a table for several examples values.

Uploaded by

siva
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
68 views4 pages

Example Floating Point Problems: Problem 1

This document presents three example floating point problems. Problem 1 describes a C program that stores different data types in a union and prints the resulting bytes, asking the reader to fill in the output. Problem 2 describes a 7-bit floating point format and asks the reader to determine exponent and significand values for normalized and denormalized numbers, and fill in a table encoding various values. Problem 3 describes an 8-bit floating point format and asks the reader to fill in exponent and significand values in a table for several examples values.

Uploaded by

siva
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 4

Example Floating Point Problems

Problem 1:
Consider the following program:

struct s {
char c;
double d;
float f;
short s;
};

union u {
unsigned char buf[24];
struct s a;
int i;
} u1;

int main()
{
int i,j;

memset(&u1.a, 0, sizeof(struct s));

u1.a.c = 0xab;
u1.a.d = -3.5;
u1.a.f = 0x1;
u1.a.s = 0xcdef;
u1.i = 0x12345678;

/* print out the bytes of u1.buf as 2 digit


hexidecimal numbers wit ha line break after
every 8 bytes */
for(i = 0; i < 3; i++)
{
for(j = 0; j < 8; j++)
printf("0x\%.2x ",u1.buf[i*8+j]);
printf("\n");
}
}

Page 1 of 4
You can use the following template as scratch space.

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+
| |
+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+--+

This program is compiled and run on a Linux/x86 machine. Fill in the output below. Write ?? if the value
cannot be determined from the information provided.

0x____ 0x____ 0x____ 0x____ 0x____ 0x____ 0x____ 0x____

0x____ 0x____ 0x____ 0x____ 0x____ 0x____ 0x____ 0x____

0x____ 0x____ 0x____ 0x____ 0x____ 0x____ 0x____ 0x____

Page 2 of 4
Problem 2:
Consider the following 7-bit floating point representation based on the IEEE floating point format:

• There is a sign bit in the most significant bit.

• The next k = 3 bits are the exponent. The exponent bias is 3.

• The last n = 3 bits are the fractional part.

Numeric values are encoded in this format as a value of the form V = (−1)s × M × 2E , where s is the sign
bit, E is exponent after biasing, and M is the significand.

Part I
Answer the following problems using either decimal (e.g., 1.375) or fractional (e.g., 11/8) representations
for numbers that are not integers.

A. For denormalized numbers:

(a) What is the value E of the exponent after biasing? ________


(b) What is the largest value M of the significand? ________

B. For normalized numbers:

(a) What is the smallest value E of the exponent after biasing? ________
(b) What is the largest value E of the exponent after biasing? ________
(c) What is the largest value M of the significand? ________

Part II
Fill in the blank entries in the following table giving the encodings for some interesting numbers.

Description E M V Binary Encoding


Zero 0 0 0 000 0000
Smallest Positive (nonzero)
Largest denormalized
Smallest positive normalized
One 1
Largest finite number
NaN — — NaN
Infinity — — +∞

Page 3 of 4
Problem 3:
Consider a 8 bit floating point representation with a three-bit significand, four bits of exponent, a sign bit,
and a bias value of 7. Assume that the implementation supports the IEEE standard (both normalized and
denormalized values, and uses “nearest even” rounding). Fill in the empty boxes in the following table:

Description Value s exponent significand

zero 0.0

closest positive to zero

largest positive

-5 −5.0

2−4 − 2−6
9
2.0 − 16

6 7
4 · 4

Recommended Book Practice Problems: 2.33, 2.34, 2.37


Solutions are at the end of the chapter.

Page 4 of 4

You might also like