Homework: Jump Tables and Switch Cases x86 64

I'm stuck on a problem where I have to provide x86 code for a switch statement given in C. My instructor gives an example in one of his lectures, but I am having difficulty understanding this example. We use Red Hat Linux for this course.



In the first section (which is red)
The code checks if a is less than 1, and jumps to the default if it is which makes sense.

In the second section (which is green)
The code checks if a is greater than 4, and jumps to the default if it is which makes sense.

In the third section (which is red)
.T gets put into register %rbx
1 is subtracted from the register, %eax, which has a (why?)
8 is put into %ecx
the product 8 * (a - 1) is stored into %eax (why?)

The instruction "movslq" is used, but I do not know what it does. I've used movl and the conditional move instructions, but I'm not quite sure what movslq is and am having a hard time finding this out with google.

And everything after that point is uncertain to me because I do not understand why some things are being done, for the most part, or what movslq does.

.T is an array and EAX will be used to index that array. But EAX contains a value in 1..4 instead of starting at 0.

The elements of .T are quads (8 bytes long). Presumably your instructor wants to show you explicitly how you index an array rather than using the indexing constructs.

EAX is a 32-bit register, RAX is a 64-bit register. So you need an instruction that extends a 32-bit value to a 64-bit one for this operation. MOVSLQ is move and sign-extend a value from a 32-bit source to a 64-bit destination.