Question: You are given the following DAXPY loop, which computes the operation Y = aX + YY = aX + YY = aX + Y for

You are given the following DAXPY loop, which computes the operation Y

=

+

=

+

=

+

Y for a vector of length

100,

where aaa is a scalar, and X and Y are vectors. The loop is implemented with the following assembly code:

scss

Copy code

DADDIU R

4,

1,

800

; R

1 =

upper bound for X L

.

D F

2, 0 (

1)

; F

2 =

(

)

MUL.D F

4,

2,

0

; F

4 =

*

(

)

.

D F

6, 0 (

2)

; F

6 =

(

)

ADD.D F

6,

4,

6

; F

6 =

*

(

) +

(

)

.

0 (

2),

6

; Store Y

(

)

DADDIU R

1,

1,

8

; Increment X index DADDIU R

2,

2,

8

; Increment Y index DSLTU R

3,

1,

4

; Test: continue loop? BNEZ R

3,

foo ; Loop if needed

Assume the following:

The functional unit latencies are given as:

FP multiply:

6

cyclesFP add:

3

cyclesFP store:

2

cyclesInteger operations and loads:

2

cycles

Results are fully bypassed.

The branch has a

1 -

cycle delay and resolves in the ID stage.

Tasks:

Unroll the loop as many times as necessary to schedule it without stalls, collapsing the loop overhead instructions.

Provide the instruction schedule.

Determine the execution time per element of the result.

Step by Step Solution

There are 3 Steps involved in it

1 Expert Approved Answer

Step: 1 Unlock blur-text-image

Question Has Been Solved by an Expert!

Get step-by-step solutions from verified subject matter experts

Step: 2 Unlock

Step: 3 Unlock

Students Have Also Explored These Related Programming Questions!

Ryan could not stay awake at work and was unable to remember and keep track of key parts of his job. When questioned, he told his boss that he had sleep apnea, a sleep disorder that causes a person...

3.14 In this exercise, we look at how software techniques can extract instruction-level parallelism (ILP) in a common vector loop. The following loop is the so-called DAXPY loop (double-precision ax...

In this exercise, we look at how software techniques can extract instruction-level parallelism (ILP) in a common vector loop. The following loop is the so-called DAXPY loop (double-precision aX plus...

Problem 1: In this exercise, we look at how software techniques can extract instruction-level parallelism (ILP) in a common vector loop. The following loop is the so-called DAXPY loop...

Q1: In this exercise, we look at how software techniques can extract instruction-level parallelism (ILP) in a common vector loop. The following loop is the so-called DAXPY loop (double- precision aX...

can someone solve this Modern workstations typically have memory systems that incorporate two or three levels of caching. Explain why they are designed like this. [4 marks] In order to investigate...

re Regular Languages and Finite Automata (a) Let L be the set of all strings over the alphabet {a, b} that end in a and do not contain the substring bb. Describe a deterministic finite automaton...

Advanced Linear Algebra / Advanced Math / Matlab question need help! Some of the needed codes are attached. In the question, it talks about the HW 6.1 but it can be neglected because every thing...

Tom Bhd. needs RM 10 million to embark on a new project in Perlis. This project is expected to generate after tax cash flows of RM 3 million per year for 7 years. The company's target for debt to...

Let f(x) and g(x) be differentiable functions such that f(7) = 2, f'(7) = 5, g(7) = 3, and g'(7) = 6 f(x) then find h'(7). g(x)' If h(x) = Give the exact value. NO DECIMALS. h'(7) =

a project has the following cash flows Year 0 = - 1 6 , 8 0 0 Year 1 = 7 , 5 0 0 Year 2 = 8 , 8 0 0 Year 3 = 7 , 3 0 0 . what is the npv at a discount rate of 0 % ? At 1 1 % ? At 2 1 % ? and at 2 9 %...

Solve clearly