syscall_handler entry point is wrong, causes page fault in release mode

[kevinaboos]

the page fault occurs in switch_to(). Haven't yet found why it occurs, but it is very consistent. Here is the output:

EXCEPTION: PAGE FAULT while accessing 0x8
error code: CAUSED_BY_WRITE
ExceptionStackFrame {
    instruction_pointer: 0xffffffff80108a01,
    code_segment: 8,
    cpu_flags: 0x2,
    stack_pointer: 0xffffffff80154018,
    stack_segment: 16
}

We should also find out which commit caused this to occur, and then diff it with the next one. Who wants to do some differential debugging?

[kevinaboos]

Update: syscall_handler causes a problem in release mode because positive offsets from the stack pointer are used. The problem and solutions are discussed here in my SO post: https://stackoverflow.com/questions/49125151/code-generation-for-local-stack-variables-in-release-vs-debug-mode

Basically, we need to subtract from rsp to make space for it. We also need to set rbp properly (and restore it properly at the end), just like the compiler would do for a non-naked function.

[kevinaboos]

Closing for now, because Theseus is not focused on supporting userspace in Ring 3.