Commit Graph

47 Commits

Author SHA1 Message Date
Bananymous 58f4ed22c4 Kernel: Don't map interrupt stack as userspace accessable 2023-09-28 21:58:24 +03:00
Bananymous 245f58cc3a Kernel: mmap regions are now demand paged
mmap will not actually take any memory unless you use the given
memory.
2023-09-28 21:07:14 +03:00
Bananymous 921d95d18f All: Clear lines with only whitspace in them 2023-09-10 00:31:42 +03:00
Bananymous 643e87a076 Kernel: Threads are deleted sooner and cleaner
We now delete threads when
    1. it is marked as Terminated and is the current thread
    2. it tries to start execution in Terminated state

This allows us to never have thread executing in Terminated state
2023-08-04 10:29:42 +03:00
Bananymous 20eafb4cc4 Kernel: align userspace stacks
I was getting #GP on sse instructions
2023-08-03 18:09:48 +03:00
Bananymous 3139391e06 Kernel: Change PageTable API
Getting free pages not reserves them, so you don't have to hold
the page table lock :)
2023-08-02 22:09:14 +03:00
Bananymous a78a7ed156 Kernel: Cleanup returns from any kind on interrupts 2023-08-01 14:23:50 +03:00
Bananymous c4f6c859c1 Kernel: Generally improve signals 2023-07-31 22:28:18 +03:00
Bananymous de927b6b05 Kernel: Remove is_in_syscall from Thread 2023-07-30 14:49:51 +03:00
Bananymous abe8810d47 Kernel: raise() now force sends signal 2023-07-30 14:17:07 +03:00
Bananymous acf125c853 Kernel: Signals are not queued anymore
Posix doesn't require signal queing if you don't use sigqueue() which
we don't support. Process also has its own pending signal mask.
2023-07-29 16:54:31 +03:00
Bananymous 3c6be319b1 Kernel: Restructure process and thread termination
When we want to kill a process, we mark its threads as Terminating
or Terminated. If the thread is in critical section that has to be
finished, it will be in Terminating state until done. Once Scheduler
is trying to execute Terminated thread it will instead delete it.

Once processes last thread is marked Terminated, the processes will
turn it into a cleanup thread, that will allow blocks and memory
cleanup to be done.
2023-07-28 18:06:20 +03:00
Bananymous b9dd1895bb Kernel: add Thread::queue_signal() 2023-07-24 22:26:10 +03:00
Bananymous 9729e5a05b Kernel: Change signal kernel API
return rsp and rip are now stored per thread
2023-07-23 18:33:10 +03:00
Bananymous 2dce0a0415 Kernel: Userspace signal handlers are now called one at a time
I added a syscall for telling the kernel when signal execution has
finished. We should send a random hash or id to the signal trampoline
that we would include in the syscall, so validity of signal exit can
be confirmed.
2023-07-23 13:34:53 +03:00
Bananymous c12f4fb40f Kernel: Make signals more POSIX 2023-07-21 20:01:12 +03:00
Bananymous b78596dcf4 Kernel: Scheduler now sends queued signals 2023-07-21 19:27:10 +03:00
Bananymous 40f7c6b8fa Kernel: Make signals thread specific 2023-07-21 19:00:59 +03:00
Bananymous 5d2bfc858e Kernel: SYS_FORK can now fail instead of panicing on error 2023-07-19 17:47:12 +03:00
Bananymous a45f9ee76b Kernel: Remove spammy process/thread exit printing 2023-07-06 10:34:46 +03:00
Bananymous 801025ad7b Kernel: Big commit. Rewrite ELF loading code
We now load ELF files to VirtualRanges instead of using kmalloc.
We have only a fixed 1 MiB kmalloc for big allocations and this
allows loading files even when they don't fit in there.

This caused me to rewrite the whole ELF loading process since the
loaded ELF is not in memory mapped by every process.

Virtual ranges allow you to zero out the memory and to copy into
them from arbitary byte buffers.
2023-06-09 00:37:43 +03:00
Bananymous 55ea5c5488 Kernel: add basic support for environment variables
exec functions will search files from PATH
2023-06-05 22:51:02 +03:00
Bananymous 896b919c9d Kernel/LibC: pass environ pointer to process 2023-06-05 21:12:08 +03:00
Bananymous b48b239882 Kernel: Implement Process::exec() 2023-05-31 20:56:29 +03:00
Bananymous 5fb69300ca Kernel: Move userspace entry functions to Process instead of Thread 2023-05-31 19:31:10 +03:00
Bananymous 5bb1f2a48c Kernel: Rename MMU to PageTable
This is more descriptive name for what it actually represents
2023-05-29 21:06:09 +03:00
Bananymous f2d767b799 Kernel: Add bareboness fork() function 2023-05-28 18:08:49 +03:00
Bananymous b021d3eebd Kernel: Processes and Threads use VirtualRange memory allocations 2023-05-28 17:48:34 +03:00
Bananymous e0a72defa2 Kernel: Add argc and argv to process entry 2023-05-16 00:27:49 +03:00
Bananymous 55bb0084aa Kernel: namespace and function renames
MMU moved to namespace kernel
Kernel::Memory::Heap moved to just Kernel
MMU::map_{page,range} renamed to identity_map_{page,range}

Add MMU::get_page_flags
2023-04-28 14:48:38 +03:00
Bananymous dc0f8b383f Kernel: Usespace threads now have an interrupt stack 2023-04-25 14:49:18 +03:00
Bananymous 48445f12ac Kernel: More proper paging in Elf loading 2023-04-22 18:17:44 +03:00
Bananymous 7530482cc2 Kernel: Cleanup process creation for userspace 2023-04-22 16:51:50 +03:00
Bananymous d63716db96 Kernel: Process is not reference counted any more
This was not necessary and it made things needlessly complicated
2023-04-19 00:34:18 +03:00
Bananymous fdb4eb6042 Kernel: Move kmalloc and MMU to Memory directory 2023-04-18 10:18:15 +03:00
Bananymous 8ee63f8264 Kernel: We can create basic userspace processes
These are still allocated on the kernel memory
2023-04-12 17:52:36 +03:00
Bananymous 3201c3654e Kernel: Threads now use only 4 KiB stack :) 2023-04-12 00:22:08 +03:00
Bananymous dcee92a6bc Kernel: Thread is no longer RefCounted
This makes developement with Scheduler much easier against compiler
optimizations. I could now remove the pragma GCC optimize stuff.
2023-03-30 19:16:51 +03:00
Bananymous 1b7625581d Kernel: Add basic concept of Processes
We now create Shell as a process that has its own threads
2023-03-16 12:17:04 +02:00
Bananymous af854ec9e1 Kernel: Thread creation now takes void(*)() as entry and void* as data
This simplifies threading for the future and gets rid of (undefined?)
cast of member function pointer to address
2023-03-09 15:33:42 +02:00
Bananymous d90aba0963 Kernel: Create CriticalScope and fix kmalloc
This disables interrupts for the current scope and restores them
after the scope. This is used in kmalloc, since scheduler might
call into kmalloc/kfree, but deadlock if some thread is currently
trying to allocate. This allows us to use kmalloc in Scheduler.
2023-03-08 13:55:53 +02:00
Bananymous 23b3028e15 Kernel: Rename RefCounted -> RefPtr and implement RefCounted 2023-03-08 03:26:25 +02:00
Bananymous b8ee77eb78 Kernel: Initial work on new scheduler with queues
Sleeping is definately broken
2023-03-08 02:41:44 +02:00
Bananymous 1dd61e93b6 Kernel: Threads cannot take arguments anymore 2023-03-02 01:56:09 +02:00
Bananymous 1bd8b0fe5c Kernel: Sleep now actually sleeps and allows idling 2023-02-19 18:52:25 +02:00
Bananymous 5b5e620d8a Kernel: Improve multithreading support
We can now use arbitary BAN::function<void(...)> as the Thread.
I also implemented multithreading for i386 since it was not done
on the initial multithreading commit.
2023-02-02 23:24:12 +02:00
Bananymous 6a9d60a8fb Kernel: Implement somewhat functioning multithread support
This still uses only a single cpu, but we can now have 'parallelization'

This seems to work fine in qemu and bochs, but my own computer did not
like this when I last tried.

I have absolutely no idea how multithreading should actually be
implmemented and I just thought and implemented the most simple one I
could think of. This might not be in any way correct :D
2023-02-01 01:53:35 +02:00