Commit Graph

714 Commits

Author SHA1 Message Date
Bananymous eee0537053 Kernel: Writes to disk are not synchronous anymore
Implement "proper" DiskCache syncing
2023-09-11 01:25:16 +03:00
Bananymous 7ec860a3d4 Kernel: Print stack trace on isr 2023-09-11 01:20:55 +03:00
Bananymous 9b9a6b2cfc Kernel: Fix ext2 file write 2023-09-11 01:20:39 +03:00
Bananymous 8fb5f97a18 Kernel: Improve open() POSIX compatability
Also rename Process::sys_creat -> create_file since it is not
actually a syscall and only used by open as a healper.
2023-09-10 00:31:42 +03:00
Bananymous 921d95d18f All: Clear lines with only whitspace in them 2023-09-10 00:31:42 +03:00
Bananymous 1fcf122c50 Kernel: Add basic ZeroDevice to /dev/zero 2023-09-10 00:31:42 +03:00
Bananymous dd9af56e21 Kernel: Start work on making inodes more thread safe
All inode operations are now locked and thread blocked
2023-09-10 00:31:42 +03:00
Bananymous 39a5c52088 Kernel: Fix directory permissions
We did not care about X bit in directories and instead used only the
R bit for search/read.
2023-09-08 11:46:53 +03:00
Bananymous 660f7cbfeb Kernel: Ext2 directories can allocate new blocks if needed 2023-09-08 10:29:26 +03:00
Bananymous 1abf54d652 Kernel: Fix bug in disk writing
I have used two weeks in locating a bug in my ext2 implementation
while the bug was actually in disk write. If you called write_sectors
on disk it would write the first sector_size bytes repeatedly to all
asked sectors and this corrupted the disk...
2023-09-08 02:43:08 +03:00
Bananymous 3750d29b2b Kernel: Fix possible out-of-bounds disk access in ext2 2023-09-08 02:42:53 +03:00
Bananymous 4648f6718e Kernel: Cleanup TTY::read() 2023-09-07 15:27:21 +03:00
Bananymous b30af0edca Kernel: TTY now reads input byte by byte
This allows correct behaviour for character streams / keyboard
handling. Serial input can now send working ^C :D
2023-09-07 15:06:27 +03:00
Bananymous 5e1725abb2 Kernel: Remove bitmap debug printing from ext2 inode allocation 2023-09-06 01:25:09 +03:00
Bananymous 6ad2f23259 Kernel: Correct inode links count 2023-09-05 14:46:56 +03:00
Bananymous ba7e1b9ca5 Kernel: Add /dev/tty symlink that targets the current tty 2023-09-05 01:07:52 +03:00
Bananymous f0772e385c Kernel: Implement basic RamSymlinkInode 2023-09-05 01:07:52 +03:00
Bananymous 3fd94b1acb Kernel: You can now read serial output from the /dev/ttyS* 2023-09-05 01:07:52 +03:00
Bananymous 93c5755012 Kernel: Receive interrupts for COM1 and COM2 input
The input has to still be attached to terminal
2023-09-04 22:07:40 +03:00
Bananymous 323de3c866 Kernel: Rework whole Terminal structure
Serial monitors can now be used as a output. This requires editing
init code for the stdio opening. Serial input is not supported, so qemu
still needs graphical window for ps/2 keyboard.
2023-09-04 19:34:18 +03:00
Bananymous 52b9fddfd7 Kernel: Make serial devices to their own class 2023-09-04 14:57:05 +03:00
Bananymous 04cfbca336 Kernel: Add ANSI support for hiding/showing cursor to TTY 2023-09-04 14:30:09 +03:00
Bananymous 9901f95d78 Kernel: Add basic nanosleep, only millisecond percision 2023-09-04 12:59:50 +03:00
Bananymous b2139c0b1e Kernel: Add basic support for O_NONBLOCK (only for tty) 2023-09-04 12:57:52 +03:00
Bananymous 6f002c926a Kernel: add basic fcntl() with couple of commands and no validation 2023-09-04 12:57:09 +03:00
Bananymous a711462ef4 Kernel: Split ext2 implementation to multiple files 2023-09-01 15:10:23 +03:00
Bananymous f31c6b847f Kernel: Start work on writable ext2 filesystem 2023-08-31 23:40:31 +03:00
Bananymous 195ccf4f53 Kernel: Add missing ififo() to Inode::Mode 2023-08-31 21:38:31 +03:00
Bananymous 073edd0b8e Kernel: Stack trace dumping validates pointers before using them 2023-08-31 21:36:23 +03:00
Bananymous 250789aa20 Kernel: better handle kernel errors 2023-08-29 00:13:21 +03:00
Bananymous 8032824054 BuildSystem: use -a with rsync
This allows cmake to not rebuild whole project every time
2023-08-28 11:38:17 +03:00
Bananymous 80e7a89f67 BuildSystem: Base sysroot is now distributed as a tar ball
This allows file and directory permissions work as intended.

cmake is now filled with 'sudo' but with sudo timeout this should be
fine.
2023-08-23 10:38:21 +03:00
Bananymous 38e72019c7 Kernel: kill() with negative pid actually matches pgid 2023-08-22 14:54:15 +03:00
Bananymous d745fca86a Kernel: Process keeps track if forked process has called exec*() 2023-08-22 14:53:46 +03:00
Bananymous 60a2185ee6 Kernel/LibC: implement proper getpgid and setpgid 2023-08-22 14:53:12 +03:00
Bananymous d634fec8dc Kernel: Add function to enumerate processes in session 2023-08-22 14:52:28 +03:00
Bananymous c9243f0d1e Kernel/LibC: Add {get,set}pgrp() 2023-08-22 11:36:33 +03:00
Bananymous 0f6c19a1b7 Kernel: Barebones implementation sessions and process groups 2023-08-22 11:35:40 +03:00
Bananymous 79f3aa5419 Kernel/LibC: add dup() syscall and function 2023-08-17 12:03:29 +03:00
Bananymous 871c792976 Kernel: O_SEARCH doesn't require any permissions
Except read permissions for all directories in the path

This allows us to stat e.g. device files for which we don't have
the permissions.
2023-08-11 12:26:07 +03:00
Bananymous 8480ffe108 Kernel: open() now validates file access mode 2023-08-11 11:53:38 +03:00
Bananymous e1400f9680 Kernel: Remove unused syscall 2023-08-11 11:43:48 +03:00
Bananymous 6e5bce3c57 Kernel: Zero initialize threads sse_storage 2023-08-11 00:26:43 +03:00
Bananymous f75adab9d8 Kernel: Move structures in boot.S to .data section 2023-08-11 00:26:12 +03:00
Bananymous 5d0a6e7b08 Kernel: HPET is now used in legacy mode when PIC is forced 2023-08-10 22:01:30 +03:00
Bananymous b0c8a9cdc4 Kernel: Fix slave PIC
we dont mask interrupt 2 in PIC since it corresponds to the
slave PIC. Also cleanup PIC code :)
2023-08-10 21:52:31 +03:00
Bananymous ebe0adb3b5 Kernel: HPET doesn't use the legacy mapping anymore 2023-08-10 21:08:32 +03:00
Bananymous 4842d5e2b9 Kernel: APIC now uses MMIO namespace functions for mmio 2023-08-10 21:07:23 +03:00
Bananymous 0b8396b1a6 Kernel: IDT now sends EOI for every interrupt
EOI is sent before calling the interrupt handler. This should be fine
since all interrupts are processed with interrupts disabled
2023-08-09 09:57:02 +03:00
Bananymous 55ef793a74 Kernel: Validate HPET tick period 2023-08-09 09:50:38 +03:00
Bananymous 96c0ad0d3d Kernel: Remove get_unix_timestamp from SystemTimer
Kernel now uses the percise real_time() and time_since_boot()
2023-08-09 08:57:50 +03:00
Bananymous 6fcb191ca0 Kernel: Add PageTable::Flags::CacheDisable
Also fix multiple places where we were using uint8_t as page table
flags instead of PageTable::flags_t which we promoted to uint16_t
while back.
2023-08-06 23:59:30 +03:00
Bananymous 3df97c36e6 Kerne: SystemTimer can now output the current real time 2023-08-04 16:06:47 +03:00
Bananymous c732297623 Kernel: Rename TimerHandler to SystemTimer
I changed SystemTimer to only handle the "best" supported timer
it can initialize.
2023-08-04 16:06:47 +03:00
Bananymous ca5a097ef5 Kernel: Add function to retrieve boot time as timespec 2023-08-04 16:06:47 +03:00
Bananymous 1f8a5f0ce5 Kernel: F11 now prints current time to debug output
This allows better timer percision and system freeze checks
2023-08-04 15:22:51 +03:00
Bananymous 1fa7a1cac4 Kernel: Add basic HPET support to replace PIT if exists
This works same way as the PIT implementation; calls Scheduler every
milli second.
2023-08-04 15:22:51 +03:00
Bananymous ea4a70c3b3 Kernel: Move sleep() implementation to TimerHandler 2023-08-04 15:15:00 +03:00
Bananymous aa0929614a Kernel: Add more structures to ACPI 2023-08-04 15:13:47 +03:00
Bananymous 3c31fc9c4b Kernel: Implement basic MMIO functions
This is equivalent to IO.h except for memory mapped io instead of
IO ports
2023-08-04 15:12:29 +03:00
Bananymous 7eb2c140fe Kernel: Move RTC to Timer directory
PIT ms counter seems to be off by multiple seconds/minute. I will
be probably changing to HPET for system time
2023-08-04 11:12:16 +03:00
Bananymous 659adb89a6 Kernel: Start work on abstracting Timers 2023-08-04 10:29:42 +03:00
Bananymous dd17124c77 Kernel: Remove unnecessary timer check from Scheduler 2023-08-04 10:29:42 +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 64ad752e73 Kernel: PageTable now supports reserved pages
You can now map pages as reserved, so that they will not be given
from get_free_page() or get_free_contiguous_pages().
2023-08-01 16:35:54 +03:00
Bananymous a78a7ed156 Kernel: Cleanup returns from any kind on interrupts 2023-08-01 14:23:50 +03:00
Bananymous 2a851b52f1 BuildSystem: Edit build flags regarding sse and warnings 2023-07-31 22:31:17 +03:00
Bananymous b245a55ea0 Kernel: store/load sse/mmx/fpu state on isr/irq/syscall
I haven't tested this yet but should be fine. This will be optimized
to only save state from threads that are using it
2023-07-31 22:28:57 +03:00
Bananymous c4f6c859c1 Kernel: Generally improve signals 2023-07-31 22:28:18 +03:00
Bananymous 1a6100f083 Kernel: Cleanup signal trampoline 2023-07-30 15:58:35 +03:00
Bananymous de927b6b05 Kernel: Remove is_in_syscall from Thread 2023-07-30 14:49:51 +03:00
Bananymous 1d7795e22c Kernel: Hardware exceptions now sends signals to userspace 2023-07-30 14:34:15 +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 9279bbbd19 Kernel: TTY now sends SIGINT on ^C to foreground process 2023-07-28 18:10:09 +03:00
Bananymous 9fd9a8b5b1 Kernel: Mark reading section from StorageDevice as terminate critical 2023-07-28 18:09:35 +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 104894c0c7 Kernel: Rewrite DiskCache
We now cache only clean pages since I don't want to think about
syncing the disk later.
2023-07-27 22:22:39 +03:00
Bananymous 2f52001c6d Kernel: Locks allow locking after locker is invalid
SpinLock and RecursiveSpinLock will now allow locking after the initial
locker is invalid. This allows us to kill threads even if they are holding
internal locks
2023-07-27 18:52:51 +03:00
Bananymous f609170a6a Kernel: DiskCache uses are now done without interrupts
This allows disk cache to not be invalid state after partial read/write
2023-07-27 18:52:51 +03:00
Bananymous 9f75b04714 Kernel: Scheduler can now check if tid is valid
Tid can become invalid if the thread is already terminated
2023-07-27 18:34:06 +03:00
Bananymous 7b4a2fe3d1 Kernel: Process::exit() unblocks the whole lock before blocking 2023-07-24 22:29:59 +03:00
Bananymous f6ee4b3496 Kernel: Implement SYS_GET_PID and SYS_TCSETPGID
we don't have consept of process groups yet
2023-07-24 22:29:59 +03:00
Bananymous 1ef0534b69 Kernel: TTY now stores foreground process pid
this will be process pgid when we have process groups
2023-07-24 22:29:59 +03:00
Bananymous b9dd1895bb Kernel: add Thread::queue_signal() 2023-07-24 22:26:10 +03:00
Bananymous be47743dfa Kernel: Scheduler can now block threads based on tid 2023-07-24 22:24:21 +03:00
Bananymous 57798e5d76 Kernel: remove Semaphore::is_blocked
this will not mean anything after semaphores can spuriously wake up
2023-07-24 22:23:13 +03:00
Bananymous cd64c1cfec Kernel: add is_tty() to inode 2023-07-24 22:19:59 +03:00
Bananymous f76d921e25 Kernel: you can now ask process if it is userspace process 2023-07-23 18:54:10 +03:00
Bananymous a2bf474013 Kernel: OpenFileDesctiptor can now return inode of fd 2023-07-23 18:52:33 +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 d560137ae6 Kernel/LibC: add SYS_SIGNAL/signal() 2023-07-21 20:08:13 +03:00
Bananymous c12f4fb40f Kernel: Make signals more POSIX 2023-07-21 20:01:12 +03:00
Bananymous 10169d773d Kernel/LibC: Add SYS_KILL/kill() 2023-07-21 19:27:38 +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 8063700d7b Kernel: add default signal actions 2023-07-21 18:02:35 +03:00