Commit Graph

214 Commits

Author SHA1 Message Date
Bananymous 58ce907327 Kernel: Usespace threads now have an interrupt stack 2023-04-25 14:49:18 +03:00
Bananymous 6ecc8cac0e Kernel: TSS stack can be set after initialization 2023-04-25 14:48:12 +03:00
Bananymous bd95f17426 Kernel: Stack pointer out of bounds now panics with a message 2023-04-25 13:40:55 +03:00
Bananymous b15deb420f LibC: Write mostly functioning stdio 2023-04-23 14:32:37 +03:00
Bananymous 8a5608df91 Kernel: d{print,warn,error}ln(...) now has a spinlock 2023-04-22 17:58:51 +03:00
Bananymous 3f1c0ec91b Kernel: Process now frees up its pages on destruction 2023-04-22 16:54:46 +03:00
Bananymous 1406a75a92 Kernel: Cleanup process creation for userspace 2023-04-22 16:51:50 +03:00
Bananymous ec8b9640e2 Kernel: Usespace programs are now ran through ELF files
only 64 bit elf files are supported for now.
2023-04-22 15:35:32 +03:00
Bananymous be14a6c239 Kernel: Stack pointer is validated when updated
This allows us not to fail stack pointer when in syscall since
interrupts use their own stack
2023-04-21 10:40:24 +03:00
Bananymous 3aa0eeb4a3 Kernel: Add barebones per process virtual addresses
We now assign every (userspace) process its own MMU which we load
in scheduler. This allows every process to have separate virtual
address space.

This is very hackish implementations but it works for now
2023-04-20 00:45:41 +03:00
Bananymous b3eeb6412f Kernel: Add some bareboness functionality to map virtual addresses 2023-04-19 23:51:36 +03:00
Bananymous d38470c8e2 Kernel: Rename MMU::{un,}allocate... to MMU::{un,}map
This is more appropriate name for the behaviour :D
2023-04-19 21:50:30 +03:00
Bananymous 4475e3e184 Kernel: ATAController will fail to initialize in native mode 2023-04-19 17:29:36 +03:00
Bananymous cf0320e47d Kernel: PCI devices now report their prog_if 2023-04-19 16:43:05 +03:00
Bananymous 59a682c720 Kernel: init2 is now launched as a process instead of thread
Also only process can now add threads to scheduler. Nobody should
have raw access to scheduler and everything should be through
Process::current() or irqs (reschedules)
2023-04-19 00:39:06 +03:00
Bananymous 7bd4593748 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 5494e2c125 Kernel: Heap allows us to take free pages.
The API is kinda weird and will probably be reworked in near future
but this will work for now :)
2023-04-18 10:18:15 +03:00
Bananymous aba82564f5 Kernel: Panic wont print stacktrace if it has already paniced
This prevents stack trace dump to panic and loop
2023-04-18 10:18:15 +03:00
Bananymous 93abee9c7c Kernel: Map all ACPI tables on initialization
This makes their usage much easier and less error prone

They won't mapped be processes when we get to there, so this won't be
a problem
2023-04-18 10:18:15 +03:00
Bananymous 2614437ba0 Kernel: Reorder boot initialization
We now create the TTY as soon as possible so we can show console
output without serial port.
2023-04-18 10:18:15 +03:00
Bananymous 1aac3a0425 Kernel: Heap implementation can now give free pages from all of RAM 2023-04-18 10:18:15 +03:00
Bananymous a180e72b6f Kernel: Start working on heap 2023-04-18 10:18:15 +03:00
Bananymous 9c0f3dd996 Kernel: Move kmalloc and MMU to Memory directory 2023-04-18 10:18:15 +03:00
Bananymous 8637959289 Kernel: We can create basic userspace processes
These are still allocated on the kernel memory
2023-04-12 17:52:36 +03:00
Bananymous 6be53668b9 Kernel: Scheduler can now terminate processes threads 2023-04-12 17:49:04 +03:00
Bananymous 12779cdef8 Kernel: Threads now use only 4 KiB stack :) 2023-04-12 00:22:08 +03:00
Bananymous 809eb2fe3e Kernel: Mark Scheduler::start() as noreturn as appropriate 2023-04-11 23:33:20 +03:00
Bananymous 8d6db168d6 Kernel: remove message from BAN::Error
We don't store the error message anymore in BAN::Error.
Instead we store a error code that can be mapped into a string.
This allows BAN::Error to only take 4 bytes instead of 128.

We should also make some kernel initialization just panic instead
of returning errors since they are required for succesfull boot
anyway.
2023-04-11 23:25:21 +03:00
Bananymous 2fabe1949c BAN: Move RefPtr to its own file and create New.h
New.h contains definitions for placement new operators and
BAN::allocator and BAN::dealloctor
2023-04-10 21:07:25 +03:00
Bananymous e704968f96 Kernel: Threads can now be terminated mid execution 2023-04-10 21:07:25 +03:00
Bananymous abbbf7ec15 Kernel: Add tty to process and make termios modifiable 2023-04-05 02:53:28 +03:00
Bananymous a423cd8bb3 Kernel: Add partial support for shell
We don't handle arrow keys, and the tty does not know how to clear
the screeen :D
2023-04-05 01:30:58 +03:00
Bananymous db076058b9 Kernel: Process can now initialize stdio
This allows the use of the fds STD{IN,OUT,ERR}_FILENO
2023-04-05 01:10:25 +03:00
Bananymous a1100624bf Kernel: Start work on making tty a file
TTY is now a file that you can read from/write to. I still have
to port shell to use this new interface
2023-04-05 00:56:09 +03:00
Bananymous 28e1497f88 Kernel: add virtual write function to inode 2023-04-03 20:29:07 +03:00
Bananymous 3ee20d1a84 Kernel: Fix typo 2023-04-03 19:56:55 +03:00
Bananymous 461a5774f8 Kernel: Device dev and rdev number is done more properly
Also hd* partitions are now 1 indexed instead of 0
2023-04-03 11:43:16 +03:00
Bananymous e322826347 Kernel: Optimize scheduler idling
Now after each interrupt we will ask the scheduler to reschedule
if the current thread is the idle thread. This allows semaphore
unblocking to be practically instant when there is only one thread
executing.

Now disk reading is back to ~3 MB/s for single threaded process
2023-04-03 01:51:05 +03:00
Bananymous 3998c5f955 Kernel: ATA now uses irqs instead of polling
Reading is now much slower at ~500 kB/s it was around 3 MB/s.
This is probably mostly due semaphore blocking taking atleast
until next reschedule (1 ms itervals). This will be a problem
as long as we are using only single processor.

I could try to use {READ/WRITE}_MULTIPLE commands, but since
most of the disk reads are 2 sectors (inode block size) this
will at most double the speed.

Most efficient speed up would of course be caching disk access
data and inodes overall.
2023-04-03 00:03:38 +03:00
Bananymous 762d22ed28 Kernel: Move ATADevice to its own file from ATAController
The API is kinda weird since device reads/writes go from
ATADevice -> ATAController -> ATADevice
but this is for now atleast necessary since ATAController has(?)
to keep all devices from using the disks at the same time
2023-04-02 18:26:19 +03:00
Bananymous 79090c2648 Kernel: cleanup includes
I went quickly went through all files since I found some weird
includes :D
2023-04-01 02:14:49 +03:00
Bananymous 7a6b1c8e47 Kernel: Fix traversing back from mount points 2023-04-01 01:54:35 +03:00
Bananymous 990887891e Kernel: Process gets absolute paths for mount 2023-04-01 00:30:11 +03:00
Bananymous 3a4557d417 Kernel: Cleanup ATA device initialization 2023-03-31 00:58:57 +03:00
Bananymous 26d9a3e253 Kernel: Move DeviceManager to its own file 2023-03-30 22:39:45 +03:00
Bananymous eef3631a5a Kernel: Improve locking in Process, VFS and ATAController
We used to block on all process access. This meant that shell
reading the keyboard input would block all VFS access making disk
accesses practically impossible. We now block only when it is
necessary :)
2023-03-30 22:02:16 +03:00
Bananymous 88ee35165f 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 c8f05b4a7a Kernel: Add Semaphore to block threads 2023-03-30 18:46:33 +03:00
Bananymous 2995a36942 Kernel: root partition is now passed from the commandline 2023-03-30 18:46:19 +03:00
Bananymous 3e8ab8271d Kernel: Shell can now mount partitions 2023-03-30 15:06:41 +03:00