Kernel: optimize yielding

Doing a yield no longer raises a software interrupt. Instead it just saves all the callee saved registers, ip, sp and return value. Because yield is only called in the kernel, it can just restore registers and jump to the target address. There is never a need to use iret :)
2026-01-11 01:31:09 +02:00
parent 83e5cb81e8
commit 35c97e2ff8
13 changed files with 109 additions and 118 deletions
--- a/kernel/arch/x86_64/Yield.S
+++ b/kernel/arch/x86_64/Yield.S
@@ -0,0 +1,29 @@
+.global asm_yield_trampoline
+asm_yield_trampoline:
+	movq %rsp, %rcx
+	movq %rdi, %rsp
+
+	subq $8, %rsp
+	pushq (%rcx)
+	pushq %rcx
+	pushq %rax
+	pushq %rbx
+	pushq %rbp
+	pushq %r12
+	pushq %r13
+	pushq %r14
+	pushq %r15
+
+	movq %rsp, %rdi
+	call scheduler_on_yield
+
+	popq %r15
+	popq %r14
+	popq %r13
+	popq %r12
+	popq %rbp
+	popq %rbx
+	popq %rax
+	movq 8(%rsp), %rcx
+	movq 0(%rsp), %rsp
+	jmp *%rcx