I produced the following program using gcc -S foo.c for the RISC-V architecture:
f:
addi
sp,sp,-48
sw
s0,44(sp)
addi
s0,sp,48
sw
a0,-36(s0)
sw
a1,-40(s0)
sw
zero,-20(s0)
sw
zero,-24(s0)
j
.L2
.L3:
lw
a5,-24(s0)
slli
a5,a5,2
lw
a4,-36(s0)
add
a5,a4,a5
lw
a5,0(a5)
lw
a4,-20(s0)
add
a5,a4,a5
sw
a5,-20(s0)
lw
a5,-24(s0)
addi
a5,a5,1
sw
a5,-24(s0)
.L2:
lw
a4,-24(s0)
lw
a5,-40(s0)
blt
a4,a5,.L3
lw
a5,-20(s0)
mv
a0,a5
lw
s0,44(sp)
addi
sp,sp,48
jr
ra
Produce a plausible definition (in C) of function f, one that might have produced this output. The function does return a value.
1
HOMEWORK #6
2
In lecture, we talked about array descriptors, which are data structures containing all the information one needs to access (get the address of) an array element A[i,j] in an implementation that allocates all elements of a new array contiguously. In C, multidimen-sional arrays are composed of rows of rows, so that A[i,j] (or A[i][j] in C) is located at address(A0,0) + N · S · i + S · j, where the array in A is M × N and each element has size S. Thus, the three constants data address(A0,0) (the virtual origin), N · S (the row stride), and S (the column stride) can be precomputed into an array descriptor, which the program can use to generate array accesses and can pass as a parameter to functions that expect to receive the array as a by-reference parameter. Show the RISC V code that you’d use to access array element A[i][j], assuming that the d, ti, and tj are registers containing the address of the array descriptor for A, the value of i, and the value of j, respectively.
By constructing appropriate array descriptors, one can give different views of an array. Describe how to compute the constructors to create the following views (we don’t need the actual code, just the calculations it must do).
Suppose that a certain array descriptor contains the information (VO, S1, S2) for accessing two-dimensional array B. Show how to create a new array descriptor that accesses column number j of B. This will be a one-dimensional array descriptor (hav-ing only one stride).
Show how to create a new array descriptor that accesses the transpose of B.
Show how to create a new array descriptor (for array view B’) that accesses the rows and columns of B in reverse, so that B’[0,0] is the same as the last column of the last row of B.
