summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2024-06-26LibC: Fix printf formattingAnton Kling
Previously %02x would not print out padding if the given value was zero.
2024-06-26TCP: Increment ACK for FIN and SYNAnton Kling
2024-06-25VFS: dup2 fixesAnton Kling
2024-06-25Optimize mouse driver to perform more work in kernelAnton Kling
Instead of sending every event to userland they now get handeld in the kernel when possible. It will now only send out events when buttons are clicked or the mouse position is requested by userland.
2024-06-24Kernel: Purge unused/bad syscallsAnton Kling
2024-06-23Kernel: Fix small out of memory issuesAnton Kling
2024-06-23LibC: Include delim in getdelim if it was seenAnton Kling
2024-06-22LibC: Fix bugs relating to fseekAnton Kling
2024-06-22Kernel: Use correct types for syscallAnton Kling
The previous type was a 64 bit which causes problems for the syscall calling interface.
2024-06-22Kernel stuffAnton Kling
2024-06-21LibC: Add ldexpAnton Kling
2024-06-21Userland: Use sys/socket.h not socket.hAnton Kling
These files only compiled due to the previous header not being cleaned up
2024-06-20Add getaddrinfo and change other libc stuffAnton Kling
2024-06-17Improve PIT timer freqAnton Kling
2024-06-17Reorganize logging codeAnton Kling
2024-06-12Improve TCP blocking of requestsAnton Kling
2024-06-11TCP stuffAnton Kling
2024-06-09Bug fix: Avoid race condition for exited process recieving interruptsAnton Kling
2024-06-09Add UDPAnton Kling
2024-05-08ext2: Use LRU for block cacheAnton Kling
2024-05-06kmalloc: Add debug kmallocAnton Kling
2024-05-05Kernel: Clear more interrupts for mouse driverAnton Kling
I am not sure why the mouse requires interrupt line 2 to be cleared.
2024-05-05LibC: fflush() should not dereference stream if it is NULLAnton Kling
2024-05-05Kernel: Mask interrupts until they have a handlerAnton Kling
2024-05-04Bug: Ext2: Properly check bounds before doing reads.Anton Kling
2024-05-02Bug: Kernel: Bounds check array accessAnton Kling
2024-05-02Kernel/MMU: Flush TLB after having made allocationsAnton Kling
2024-05-02sh: Use string view instead of C stringsAnton Kling
2024-05-01Kernel/Keyboard: Only write to buffer after it is initializedAnton Kling
2024-05-01Bug: Kernel/MMU: Initialize memory after allocationAnton Kling
The memory not being zeroed caused certain flags to be set(presumably no-cache flag) and resulted in the OS being unusually for only certain operations and only in certain cases.
2024-05-01Bug: Kernel/TCP: Initalize the tcp header.Anton Kling
2024-04-29Kernel/LibC: Remove unused syscallsAnton Kling
2024-04-29Kernel/IRC: Add setsockopt and move IRC client to use new socket interfaceAnton Kling
2024-04-29Kernel/Socket: Move sockets back to being file descriptorsAnton Kling
2024-04-29Kernel/TCP: Increment seq number after sending SYN|ACKAnton Kling
2024-04-28Kernel: Use ringbuffer for keyboard and mouse eventsAnton Kling
2024-04-28WindowServer: Make mouse_fd non blockingAnton Kling
2024-04-28Kernel: Use ringbuffer for handling incoming socket dataAnton Kling
2024-04-27Kernel: Ensure new kernel tables geet allocated for all current page directoriesAnton Kling
2024-04-27Kernel: Make "relist" able to dynamically resizeAnton Kling
2024-04-27Kernel: Make file descriptor numbers reusable.Anton Kling
Instead of using the "append only" list it now uses "relist" which allows for indexes to be removed.
2024-04-26Kernel/VFS: Change polling from variables to functionsAnton Kling
Instead of having to store state in variables functions are called to check the object directly.
2024-04-26IRC: Add PING commandAnton Kling
2024-04-26Kernel: clang format fixesAnton Kling
2024-04-26Kernel: More out of memory condition checksAnton Kling
2024-04-26Kernel: Handle userspace memory errors instead of crashing.Anton Kling
The kernel should kill the process if it has memory issues.
2024-04-26Kernel: Improve error handling for out of memory conditionsAnton Kling
2024-04-25Kernel: Add a ringbufferAnton Kling
This will be used later
2024-04-23VFS: Fix reference count of file descriptors.Anton Kling
It appears that there was a race condition where the process got closed before the server accepted the incoming request. Causing the file descriptor to have "0" in the reference count but it would still be given when calling accept.
2024-04-22MMU: Extra asserts to avoid future regressionsAnton Kling