From e134202611a50b358c147c92bfb8a7f443030b8b Mon Sep 17 00:00:00 2001 From: Kimplul Date: Mon, 8 Jul 2024 17:43:59 +0300 Subject: fix LLVM + Anything extern is right out as LLVM doesn't produce correct code for them. Maybe if I added some extra attributes but I'm skeptical. Replaced with static variables and getters/setters. + Inline ASM is apparently a bit buggy, so use an assembly stub when jumping to init. + Minimize work done in main() to minimize chance of LLVM doing something silly. Still not 100% certain that I shouldn't just write the main() as an assembly stub in arch/riscv64 to be absolutely sure everything works as intended. + Make .kernel.start section SHF_ALLOC, otherwise lld complains about pc-relative addressing Probably some other stuff as well that I'm forgetting right now. But at least with LLVM14 LTO seems to work, which is pretty cool? --- include/arch/vmem.h | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) (limited to 'include/arch/vmem.h') diff --git a/include/arch/vmem.h b/include/arch/vmem.h index e62c112..a19f555 100644 --- a/include/arch/vmem.h +++ b/include/arch/vmem.h @@ -148,10 +148,14 @@ vm_t setup_kernel_io(struct vmem *b, vm_t paddr); * address space and the kernel address space in the other. * Called at boot, not allowed to allocate memory. * + * @param load_addr Where in physical memory kernel was loaded to. Could in + * theory be fetched from \ref get_load_addr(), but the fewer globals are + * touched during initialization the better. + * * @return The vmem node used to build the address space. Probably statically * allocated. */ -struct vmem *init_mapping(); +struct vmem *init_mapping(uintptr_t load_addr); /** * Create new virtual memory space. -- cgit v1.3