Consider the following CUDA kernel and the corresponding host function that

calls it:

void foo$\_$kernel$($int$*$ a$,$ int$*$ b$)\{$

unsigned int $i=$ blockIdx.x$*$blockDim$.$x $+$ threadIdx.x;

if $($threadIdx$.$x $<mn\; id="EditorElement\_2957074">4</mn><mn\; id="EditorElement\_2957075">0</mn><mi\; id="EditorElement\_2957077">|</mi><mi\; id="EditorElement\_2957079">|</mi>$ threadIdx.x $104$

$b[i]=a[i]+1$;

$\}$

if

$a[i]=b[i]**2$;

$\}$

for $($unsigned int

$b[i]+=j$;

$\}$

$\}$

void foo$($int$*$ a$\_$d$,$ int$*$ b$\_$d$)\{$

unsigned int $N=\frac{1024}{b}ar($; $)$

foo$\_$kernel a$\_$d$,$ b$\_$d$)$ ;

$\}$

a$.$ What is the number of warps per block?

b$.$ What is the number of warps in the grid?

c$.$ For the statement on line $04$:

i$.$ How many warps in the grid are active?

ii$.$ How many warps in the grid are divergent?

iii. What is the SIMD efficiency $($in $\%)$ of warp $0$ of block $0?$

iv$.$ What is the SIMD efficiency $($in $\%)$ of warp $1$ of block $0?$

v$.$ What is the SIMD efficiency $($in $\%)$ of warp $3$ of block $0?$

d$.$ For the statement on line $07$:

i$.$ How many warps in the grid are active?

ii$.$ How many warps in the grid are divergent?

iii. What is the SIMD efficiency $($in $\%)$ of warp $0$ of block $0?$

e$.$ For the loop on line $09$ :

i$.$ How many iterations have no divergence?

ii$.$ How many iterations have divergence?