kmi - ACTIVE - Kernel with migrating threads as IPC

Age	Commit message (Collapse)	Author
2023-08-04	add initial smp bringup	Kimplul

2023-08-03	add sbi_hart_start	Kimplul

2023-08-03	fix cpu_send_ipi	Kimplul

2023-07-31	start working on irqs	Kimplul

2023-07-27	use nifty variable argument count trick	Kimplul
	+ Found it in newlib. This allows the compiler to more effectively optimize away unnecessary register operations. Applied to both kernel and init.c for a slight speed increase. I really should move init.c to some other repository though, the binary files are starting to get annoying
2023-07-27	add ipc_kick	Kimplul
	+ Tests will have to come later
2023-06-08	skip saving registers when possible	Kimplul

2023-06-08	add fast userspace return to do_ipc	Kimplul

2023-06-07	slight improvements to building	Kimplul
	+ User can now specify debug, release and assert handling
2023-06-07	change register save/load to increasing order	Kimplul

2023-06-05	give compiler more room to optimize rpc invocation	Kimplul
	+ Try to write arguments to memory as fast as possible to free up some registers that the compiler can then play with. Qemu runs ~1 700 000 rpc's per second, visionfive2 @1GHz (I think) ~800 000.
2023-06-04	slight optimization and brainfart fix	Kimplul

2023-06-04	add some docs about visionfive2	Kimplul

2023-06-04	visionfive2 boots	Kimplul
	+ Make 8250 serial driver more generic + Still TODO: write a tutorial on how to boot on the visionfive2
2023-06-04	visionfive2 boots until debugging is initialized	Kimplul

2023-05-29	start trying to boot on visionfive 2	Kimplul
	+ Not bootable quite yet. Among other things, I couldn't get the current starfive u-boot fork to boot, so try adding support for booting with precompiled u-boot via the `go` command. Initial testing with qemu shows that this should be possible, and if it works, might be useful in the (far) future with other slightly janky SBCs. Also, NS16550 is 8250-based, and I'm really only using the base 8250, so rename and add visionfive 2 uart to list of compatibles. Visionfive 2 is still completely untested.
2023-05-17	align handle_irq to four byte boundaries always	Kimplul
	+ CSR_STVEC requires four byte alignment
2023-05-12	disable LTO with LLVM	Kimplul

2023-05-12	fix LLVM compilation	Kimplul
	+ RELEASE=1 doesn't work for some reason, though
2023-05-11	arbitrary load address and RAM base detection	Kimplul
	+ Both kind of go hand in hand, made sense to do both at the same time. Some parts feel slightly hacky, the loader works by placing everything on the stack and avoiding global values. Still, seems to work?
2023-05-06	make cross compile more conformant	Kimplul

2023-05-01	update license stuff	Kimplul

2023-04-30	rename to kmi	Kimplul

2023-01-26	add req_page syscall	Kimplul

2022-12-02	initial shared memory working	Kimplul

2022-12-01	start looking into memory handling	Kimplul
	+ Now 10M alloc/frees succeed, which totals more memory than the virtual machine has, so no too obvious leaks are occuring. For future debugging speed, changed 10M to 1M. + Also quick fix to rpcs, stacks are now assigned. Not entirely sure why they worked before this, but good that I found it.
2022-11-21	fix llvm	Kimplul

2022-11-21	make syscall handlers void	Kimplul

2022-11-21	possible optimisations	Kimplul

2022-11-13	improve documentation on new features	Kimplul

2022-11-13	~1.3M ipc requests	Kimplul
	+ The secret is using gravestones. I'll have to write up full documentation for the feature but essentially riscv lets us encode whatever we want into page table entries, as long as they're not active. We use this to encode highest user address that is not a zero, in that all entries in the top level are either active or gravestones. When an active entry is removed, it is either a gravestone (if there are other active entries above it) or it starts a cascade of removing entries that have been previously removed Slight runtime overhead to page mapping, pretty major advantage in rpc calls. Feature will need to be tested more thorougly, and the init program is sort of a best scenario with just one top level userspace page table entry active at a time, leading to incredibly fast context switches. Current implementation limits a process' max virtual memory to 248 GiB (in Sv39), but I don't think the missing 8 GiB is that big of a deal.
2022-11-13	~700k ipc requests	Kimplul
	+ Eliminated need for save/load_regs, now the register save area is behind a pointer which _slightly_ increases overall syscall delay, but speeds up ipc requests by a fair bit.
2022-11-13	give uvmem_map more fitting name	Kimplul
	+ It really only works with Sv39, I suppose similar structures should be added for Sv4} etc. if I ever get around to it.
2022-11-13	write init more sensibly	Kimplul
	+ Both easier to look at and now the userspace doesn't play as big of a role in the 'benchmarking' with optimizations turned on.
2022-11-12	release mode, ~600k requests	Kimplul

2022-11-12	make 'performance' results more accurate in qemu	Kimplul

2022-11-12	slight optimization	Kimplul

2022-11-12	initial rpc implementations	Kimplul

2022-11-11	less buggy fork	Kimplul
	+ Not strictly done yet, but we can swap between two processes :D
2022-11-10	initial implementation of process copying	Kimplul
	+ Not actually working (at least fully), need to continue debugging when I have time.
2022-11-09	add basic IPI structure	Kimplul

2022-10-29	rewrite pmem	Kimplul

2022-10-22	start outlining ipc	Kimplul

2022-10-22	create impl and CAP_PROC checking	Kimplul

2022-10-22	initial high level fork imp	Kimplul

2022-10-21	beautification	Kimplul

2022-10-21	initial implementation of swap	Kimplul
	+ Slightly unsure if it should be a syscall, or if assign would be enough. Also, which thread should run in a fork/spawn? For now, old thread, though this might increase latency. Or add swap flag to these calls?
2022-10-21	initial spawn implementation	Kimplul

2022-10-21	slight optimisation to cur_tcb on riscv	Kimplul

2022-09-15	change syscalls to take 5 params and return 6	Kimplul
	+ In total, a syscall is built up of 6 values, with the first being the syscall number. Symmetrically, the first value is now a status and the following five values return "values". This allows us to cram in more info into the ipc_* functions. The performance difference is absolutely minimal, at least from my testing in qemu.