Commit Graph

1192 Commits

Author SHA1 Message Date
Bananymous 08cbd009ac Kernel: PCI don't report multi function bit with header_type() 2024-06-29 22:11:55 +03:00
Bananymous 3d4219bfee Kernel: Don't panic if nvme initialization fails 2024-06-29 22:07:55 +03:00
Bananymous 1a844426c3 Kernel: Add fast path for framembuffer sync if bpp == 32 2024-06-28 23:34:11 +03:00
Bananymous 42237a3bc8 Kernel: Implement fast scrolling for TTY 2024-06-28 23:15:03 +03:00
Bananymous 48a76426e7 BAN: Add more APIs for Atomic and make compare_exchage take a reference 2024-06-28 21:47:47 +03:00
Bananymous 31568fc5a1 Kernel: Rewrite Sockets to not be TmpInodes
TmpInodes just caused issues because TmpFS kept them alive. There was
really no reason for sockets to even be stored inside a TmpFS...
2024-06-27 00:35:19 +03:00
Bananymous cb07142832 Kernel: ACPI allow more calling or \_S5
Spec says that \_Sx must have 4 fields, but virtual box seems to only
have the two defined once. This patch allows shutodown on virtual box
2024-06-25 23:25:10 +03:00
Bananymous 60a05412c9 Kernel: ACPI implement SizeOf 2024-06-25 23:24:51 +03:00
Bananymous 0179f5ea09 Kernel: ACPI add \_OS string 2024-06-25 23:24:19 +03:00
Bananymous f671ed7e3f Kernel: ACPI implement integer stores to registers as copies
Before storing const integer and then modifying the register it would
error.
2024-06-25 23:23:52 +03:00
Bananymous 2fccff5a35 Kernel: Implement ACPI IndexOp into Strings 2024-06-25 23:23:00 +03:00
Bananymous af4b138094 Kernel/LibC: Implement realpath
realpath is implemented as a syscall. This is not really required but it
was the easiest way to get it working as there is already path
canonicalization at kernel level.
2024-06-25 19:32:40 +03:00
Bananymous 25099b4c98 Kernel: Don't validate O_SEARCH for non-directories 2024-06-25 19:27:55 +03:00
Bananymous d7b8458a56 Kernel: Fix TCP sending
TCP send was effectively always waiting for connection to close and then
return a value of 0.
2024-06-25 11:04:03 +03:00
Bananymous 67dfe0bcf3 BAN: Allow String::formatted to fail 2024-06-25 11:04:03 +03:00
Bananymous 9e1b5cbaab BuildSystem: Cleanup CMake code to allow libc only installation
There was no way to just install libc which is required for stdlibc++
2024-06-21 01:45:14 +03:00
Bananymous 4f0457a268 Kernel: Rewrite a lot of TCP code and implement TCP server sockets
TCP stack is now implemented much closer to spec
2024-06-20 13:26:50 +03:00
Bananymous bce16cdd6e Kernel: Fix how socket closing works
Sockets are now closed only when they are not referenced any more. This
allows child process to close socket and still keep it open for the
parent.
2024-06-19 10:39:44 +03:00
Bananymous ad6d95ba52 BuildSystem: Rework the whole cmake build system
Now files are installed using the install() command instead of manually
copying files to their destinations. This allows automatic recompilation
of headers that did not work previously
2024-06-19 09:40:03 +03:00
Bananymous 318ce5dec8 All: Fix a lot of compiler warnings from header files
While reworking build system, header files started to report warnings.
2024-06-18 23:02:10 +03:00
Bananymous c69919738b BuildSystem: Move all userpace libraries under the userspace directory
As the number of libraries is increasing, root directory starts to
expand. This adds better organization for libraries
2024-06-18 13:14:35 +03:00
Bananymous f1e366d36f Kernel: Free keyboard mutex while waiting for data to read
This was making select hang if one thread was trying to read from
keyboard.
2024-06-17 23:04:37 +03:00
Bananymous be7ed8e74a Kernel/LibC: Implement {get,set}sockopt()
These are pretty much dummy functions in the kernel side. Only case that
is handled is SOL_SOCKET with SO_ERROR. This is hard coded to return no
error. Network stack is currently synchronous, so all errors are already
reported through synchronous network functions.
2024-06-17 20:56:48 +03:00
Bananymous cad55a4da5 Kernel/LibC: Implement getsockname for ipv4 sockets 2024-06-17 20:54:45 +03:00
Bananymous 511fc870a1 BAN: Mark RefPtr and WeakPtr operator bool() as explicit 2024-06-17 20:19:36 +03:00
Bananymous ea7fc7f6c4 Kernel: Implement read-only FAT12/16/32 driver with long name support
You can now mount FAT filesystems! This code might not work perfectly
but my quick testing seemed to work on all (FAT12/16/32) variants.
2024-06-14 01:04:12 +03:00
Bananymous 6b1d5d28be Kernel: VFS root now has to be block device instead of partition 2024-06-14 00:19:12 +03:00
Bananymous 766439db6d Kernel: Start work on adding support for new filesystems
Old code tried to create ext2 filesystem from all devices.
2024-06-11 10:50:26 +03:00
Bananymous d4903caafa Kernel: Combine consecutive mouse move and scroll events
This makes mouse work much smoother when running without kvm.
2024-06-11 00:07:31 +03:00
Bananymous ffacff67cf LibFont: Move PSF code to separate file 2024-06-10 16:10:05 +03:00
Bananymous 530c259e71 Kernel: Close unix domain socket when it gets destoyed
Inode closing is something that needs a complete rework. Currently all
sockets are closed when close() is called, which leads to connection
closing if you fork()/exec() with socket being marked as CLOEXEC.

Inodes should probably only be closed once they are not referenced
anywhere.
2024-06-03 18:06:01 +03:00
Bananymous 765ccfa18c Kernel: Deliver SIGCHLD on process exit and ignore it properly 2024-06-03 17:58:24 +03:00
Bananymous bd1290706a Kernel: Implement SharedMemoryObject cloning 2024-06-03 03:41:00 +03:00
Bananymous aec5a09caf Kernel/LibC: Implement SYS_ISATTY and isatty() 2024-06-03 03:36:25 +03:00
Bananymous 446220494e Kernel: Unix domain sockets close can now be detected
When a unix domain socket is closed and it has a connection to another
socket, it will make the other socket readable and recv will return 0.

This allows detection of socket closing
2024-06-02 16:48:55 +03:00
Bananymous 8bfacb0091 Kernel: Implement deletion of SMO objects 2024-05-31 13:04:23 +03:00
Bananymous 0501f3bd99 Kernel: Move font code to its own library LibFont 2024-05-31 10:47:05 +03:00
Bananymous 84b3289a2a Kernel: Move Scheduler::yield() lock check after interrupts disabled
I have no idea why this solves a bug where current processor has
scheduler lock at the beginning of yield.
2024-05-31 02:56:39 +03:00
Bananymous b760892de2 Kernel: Make pselect use nanosecods instead of milliseconds 2024-05-31 02:56:17 +03:00
Bananymous 6840a8983c Kernel: Make sure MSB is not set on SMO keys 2024-05-29 20:01:12 +03:00
Bananymous a1b3490764 Kernel: Improve random number generation for unsigned types 2024-05-29 20:00:47 +03:00
Bananymous 076f1efecb Kernel: Fix 32 bit fast page locking
I forgot to change this when changing the lock type. 32 bit boots again
fine :D
2024-05-29 19:44:39 +03:00
Bananymous 53e572f072 Kernel: Fix s_fast_page_lock type on 32 bit target 2024-05-29 18:04:23 +03:00
Bananymous d4d530e6c8 Kernel: Implement basic shared memory objects
These can allocate memory that can be shared between processes using
a global key. There is currenly no safety checks meaning anyone can
map any shared memory object just by trying to map every possible key.
2024-05-29 15:58:46 +03:00
Bananymous 99270e96a9 Kernel: Lock debug lock while printing fault details
This allows multiprocessor to dump clean output on concurrent faults
2024-05-29 15:49:24 +03:00
Bananymous 4bf7a08c80 Kernel: Allow SYS_PSELECT to work with timeout of zero 2024-05-29 15:32:18 +03:00
Bananymous 3823de6552 Kernel: Add templated get function for Random 2024-05-29 15:32:00 +03:00
Bananymous 8bc6c2eb20 Kernel: Move KeyEvent/MouseEvent from kernel to LibInput 2024-05-28 23:30:08 +03:00
Bananymous 87d52e5ebe Kernel: Fix timer early wake message
When printing early return message, current time was read twice. This
could lead to early return check failing, but when printing and reading
the time again subtraction overflow would happen.
2024-05-28 16:04:18 +03:00
Bananymous 598a09c13d Kernel: Allow select to work on any type of inode 2024-05-28 16:03:54 +03:00
Bananymous 18e2559b1e Kernel/LibC: Add SYS_TRUNCATE 2024-05-28 01:08:04 +03:00
Bananymous a1ab44d39f Kernel: Optimize disk reads to read multiple sectors at once
Old StorageDevice::read_sectors() read each sector separately if the
underlying disk had a disk cache. This patch allows multiple sectors to
be read even if the disk cache exists and contains some of the sectors.

Only sectors that could not be found from the disk cache are actually
read from the disk. This optimization is not done for writing, which
still will write each sector separately, if disk cache has no memory to
store new sectors. It would feel kind of unnecessary optimization as you
have greater problems if disk cache cannot allocate a single page.
2024-05-27 15:52:34 +03:00
Bananymous 8b1514e575 Kernel: Make all storage devices readable and writable
I only had a {read,write}_impl defined for ATABaseDevice. This patch
moves that implmentation to general storage device.
2024-05-27 13:41:55 +03:00
Bananymous 2d3810874d Kernel: Fix thread signal handling
Threads will now only handle signals once they are not holding any
mutexes. This removes some dead locks.
2024-05-26 20:08:35 +03:00
Bananymous 2a4d986da5 Kernel: Add preliminary support for PCIe
Only segment 0 is supported, but devices can now be accessed through
mmio.

Adding more segments would require adding argument to every PCI API so
it is left for later.
2024-05-25 20:50:07 +03:00
Bananymous df260fe0e8 Kernel: Process::validate_pointer_access now maps the whole range
This fixes a bug where userspace provided address is not fully mapped
and the kernel tries to read/write it while using PageTable fast page.

In the future userspace input should be copied on syscall entry, so
userspace could not modify the input during syscall. Currently there
is change that userspace input passes kernel syscall validation and
after that userspace could modify the input before the value is
actually used.
2024-05-24 14:14:17 +03:00
Bananymous 2be4fe8404 Kernel: Make PageTable::s_fast_page_lock non-recursive
This lock is only used in wrapper of PageTable. There is no possiblity
of taking the lock outside of these wrappers.
2024-05-24 14:12:35 +03:00
Bananymous 06f4b0b29a BAN: Make String and StringView header only
This allows linking with libc without having to link ban
2024-05-23 15:43:26 +03:00
Bananymous 83e3409bd8 Kernel/LibC: Update SYS_SEEK to return new offset and implement lseek 2024-05-23 14:49:23 +03:00
Bananymous 0af74fccda Kernel/LibC: Rework dirent structure
dirent now contains statically sized d_name. This allows using
sizeof on the name and dirent properly, which some programs seem
to be using.
2024-05-22 20:19:59 +03:00
Bananymous e77de1804f Kernel: Fix some race conditions in TCP stack
Remove race condition if two acks are to be sent one after another.

Always unblock semaphore once TCP thread has done something. This
allows better chance of TCP sending to succeed.

There are multiple places in the networking code that would require
thread-safe entering to blocking mode. I should add some API for this
so that a lot of race conditions could be removed.
2024-05-21 01:53:45 +03:00
Bananymous e00b92225d Kernel: Fix E1000 interrupt handling condition
I had written the ICR register check backwards which lead to interrupt
handling only when it was not needed, and no handling when it was
needed. This somehow still worked, just much slower often requiring tcp
resends from the server.
2024-05-21 01:52:19 +03:00
Bananymous e7e1dd91c7 Kernel: Implement ACPI reset 2024-04-22 21:12:13 +03:00
Bananymous 195c5e92a4 Kernel: Add floating bus detection for ATA Bus 2024-04-22 21:12:13 +03:00
Bananymous 26922ebb51 Kernel: Remove stack size check for keyboard layout initialization
Keyboard layout loading can take around 1 KB of stack for i686 target
2024-04-22 21:11:04 +03:00
Bananymous 693f90449f Kernel: Rework AML package and implement indexing in to packages 2024-04-19 11:26:48 +03:00
Bananymous 34b10f61ce Kernel: Make PIT reserve its IRQ
PIT did not reserve IRQ leading to kernel panic if it was being
initialized.
2024-04-18 13:34:28 +03:00
Bananymous 1479b42112 Kernel: Don't even loop over processor when system has only BSP 2024-04-18 13:33:52 +03:00
Bananymous bb061d2a0a Kernel: Make 32bit not use large pages during boot
There is no guarantee that they are supported
2024-04-18 13:33:15 +03:00
Bananymous 061012a268 Kernel: Fix signal trampoline
32 bit did not even support it and 64 bit did not align stack
2024-04-18 13:32:40 +03:00
Bananymous a698f91db4 Kernel: Fix PriorityMutex::try_lock()
Forgot to check this when I updated Mutex::try_lock()
2024-04-18 01:36:15 +03:00
Bananymous 30d12a76bc Kernel: Fix 32 bit compilation 2024-04-18 01:35:56 +03:00
Bananymous 687fa44eff Kernel: Add support for ACPI general purpose events 2024-04-17 21:51:03 +03:00
Bananymous aefb33efff Kernel: Implement dummy AML NotifyOp 2024-04-17 15:40:11 +03:00
Bananymous da0c45b7ee Kernel: Cleanup AML debug dump output 2024-04-17 15:35:42 +03:00
Bananymous 956335e844 Kernel: Implement AML WhileOp 2024-04-17 15:00:14 +03:00
Bananymous 701fc600cd Kernel: Implement evaluate and store for BankFieldElement 2024-04-17 14:59:51 +03:00
Bananymous e38b2cff4f Kernel: Implement proper synchronization for AML opregions 2024-04-17 14:59:26 +03:00
Bananymous b268293402 Kernel: Fix Buffer evaluate and store for sizes >= byte
All reads and writes were to the first byte. This patch adds the
correct offset for reading and writing.
2024-04-17 14:57:13 +03:00
Bananymous 45b9dc8be9 Kernel: Make ACPI \\_OSI return true for Linux 2024-04-17 02:19:35 +03:00
Bananymous 0ad7025a17 Kernel: Actually map FACS (and global lock) 2024-04-17 02:14:56 +03:00
Bananymous 49b7467840 Kernel: Simplify AML method invocation API 2024-04-17 01:20:58 +03:00
Bananymous a40ef610a2 Kernel: Allow demand paging only for userspace threads 2024-04-17 01:16:43 +03:00
Bananymous f9943b60e4 Kernel: Fix mutex try_lock return value
WHY IS GCC NOT COMPLAINING FROM HEADERS... I used almost 2 hours
trying to locate why AML was broken
2024-04-17 01:16:43 +03:00
Bananymous f97fb1b35d Kernel: Evaluate _REG on devices for embedded controller if needed 2024-04-17 01:16:43 +03:00
Bananymous b959181afd Kernel: Remove ACPI namespace from itself
This removes unnecessary cyclic reference
2024-04-17 00:52:05 +03:00
Bananymous cbc27a94ac Kernel: AML cleanup IfElse code 2024-04-17 00:49:07 +03:00
Bananymous 6a4f999b88 Kernel: Cleanup AML code and add evaluate for string 2024-04-16 17:39:16 +03:00
Bananymous 7707e01352 Kernel: AML add flag to force absolute lookup for ACPI namespace 2024-04-16 17:38:23 +03:00
Bananymous e667326df5 Kernel: Fix AML if parsing
Parse else case unconditionally, even if _if_ case evaluates to true
2024-04-16 17:36:04 +03:00
Bananymous f1b2d7530d Kernel: Rework AML namespace and object hierarchy
Remove tree-like structure from AML. This allows more spec compliant
parsing of named objects inside not yet declared devices.

This also allows AML to be run thread safely. All object adds/removes
are now guarded by a mutex.
2024-04-16 16:47:45 +03:00
Bananymous b6587b32b9 Kernel: ACPI ignore events from non-existent ports 2024-04-16 00:00:07 +03:00
Bananymous b89bafa165 Kernel: Add support for power button shutdown
This patch implements basic support for power button using ACPI
fixed events. I still need to implement general purpose events
and embedded controller for full power button support.
2024-04-15 23:55:25 +03:00
Bananymous 9fac5f94ba Kernel: Fix entering ACPI mode
My condition to check wheter to enter ACPI mode was incorrect.
I only entered ACPI mode if I already was in ACPI mode :DD
2024-04-15 15:05:48 +03:00
Bananymous 5affc73ee6 Kernel: Parse ACPI namespace only after TTY is initialized
This allows better debugging when there is no serial port available.
2024-04-12 16:47:17 +03:00
Bananymous 027016ddae Kernel: Load all SSDTs and PSDTs
Spec says to load all tables with unique OEM table IDs but that seem
wrong.
2024-04-12 16:45:00 +03:00
Bananymous 8f2f98b7b4 Kernel: AML implement SleepOp 2024-04-12 16:17:14 +03:00
Bananymous 6b43d12469 Kernel: AML package doesn't mark ZeroOp as NullName reference 2024-04-12 16:10:16 +03:00
Bananymous 74940ed33c Kernel: Cleanup AML code and fix bugs
I can enter ACPI mode on my own laptop!
2024-04-12 16:03:14 +03:00
Bananymous 17871bb3ca Kernel: Fix ACPI namespace initialization
ACPI spec says that only SSDTS with unique OEM table IDs are to be
loaded.

Add loading of ACPI 1.0 PSDTs
2024-04-12 12:41:30 +03:00