Add partial aarch64 support to libuser

[leo60228]

This is able to make libuser compile with the included target JSON. It does not yet include a crt0 or syscall stub, though preparations for that have been made (I'm not experienced enough with aarch64 assembly).

[leo60228]

I've implemented a syscall stub, it was easier than I thought.

[roblabla]

Looks really good!

[Orycterope]

Omg thanks so much

[Orycterope]

Update the comment

[Orycterope]

almost certainly not valid for aarch64.

This deserves a huge

// todo: figure out aarch64 elf TLS

[Orycterope]

@roblabla , I need your confirmation on this, but IIRC the sole reason we use a global asm for the syscall is because we need to pass ebp as our sixth argument, but trying to backup+set/restore it in an inline asm block was unfeasible, because Rust needs this register to exit the inline asm block and restore regular context (or something like that).

This struct is only used as an argument to the global asm syscall_iner. This is pretty inefficient, since to do a syscall you must:

1. push all 6 arguments on the stack, and call syscall. (this is standard i386 ABI)
2. syscall copies those arguments to fill a new Register structure, and calls syscall_inner
3. syscall_inner consumes this struct, and puts every field in a register.
4. int 0x80
5. syscall_inner saves the return registers state to the Register structure, and returns.

On aarch64, where we have plenty of registers, this whole dance seems unnecessary, and I'd like to see the syscall function simply be a wrapper around a single inline asm instruction, that define its input/output registers properly (e.g. nr -> r0) and let llvm do all the copying for us.

In this context, the Register struct is very i386 specific, and doesn't need to be made generic.

I'd like to hear roblabla's opinion this, but if he agrees with me I'd recommend dropping this commit entirely

[Orycterope]

wow, you did exactly what I was discussing in the previous comment 😅

For aarch64, I don't see the need for copying arguments to the Registers struct. Can't you just directly use nr ?

[leo60228]

I thought Registers was in the public API? If it isn't, that's unnecessary.

[Orycterope]

It was, but I think that's a mistake, it should be defined only for i386

[todo]

figure out aarch64 TLS

SunriseOS/libuser/src/threads.rs

Lines 394 to 404 in 708cb35

	// TODO: figure out aarch64 TLS
	debug!("starting from new thread, context at address {:#010x}", thread_context_addr);
	// first save the address of our context in our TLS region
	unsafe {
	// safe: - get_my_tls returns a valid 0x200 aligned ptr,
	// - .ptr_thread_context is correctly aligned in the TLS region to usize,
	// - we're a private fn, thread_context_addr is guaranteed by our caller to point to the context.
	(*get_my_tls_region()).ptr_thread_context = thread_context_addr
	};
	// use get_my_thread_context to create a ref for us

---

This comment was generated by todo based on a TODO comment in 708cb35 in #566. cc @leo60228.