Commit Graph

1972 Commits

Author SHA1 Message Date
Bananymous 4599e1dec5 Kernel: Make schedler not crash when unblocking active thread
This was happening when sending signals to processes that were currently
running.
2024-08-01 18:21:49 +03:00
Bananymous 6de350ce9d Kernel/LibC: Cleanup, fix and implement a lot of signal code
This patch implements sigsets and some of their usages
2024-08-01 17:01:18 +03:00
Bananymous 838d31fa41 Kernel: Implement more POSIX compliant open() and openat() syscalls 2024-08-01 15:35:02 +03:00
Bananymous 401b460d75 LibC: Implement setbuf() and setvbuf() 2024-08-01 01:30:00 +03:00
Bananymous c440204fa5 LibC: Implement dummy strcoll() 2024-07-31 23:58:10 +03:00
Bananymous adf50dffd8 LibC: Implement uname() 2024-07-31 23:54:26 +03:00
Bananymous 20d38ed28c LibC/Shell: Implement gethostname() and shell uses it for hostname 2024-07-31 23:53:55 +03:00
Bananymous edc30cd71d LibC: Compile without exceptions
This allows libc to not require __gxx_personality_v0. I can maybe add
C++ back to libc... :D I don't know why I did not research earlier what
this symbols was used for
2024-07-31 23:33:41 +03:00
Bananymous 10ce03a1e7 LibC: Implement ctime() and asctime() 2024-07-31 23:26:10 +03:00
Bananymous 5fca5c774a LibC: Implement umask() 2024-07-31 23:26:10 +03:00
Bananymous fc6c39e670 LibC: Implement gettimeofday() 2024-07-31 23:26:06 +03:00
Bananymous 7c3b6307d9 Kernel: Cleanup USB initialization code 2024-07-31 23:23:44 +03:00
Bananymous e52dac3b25 LibC: Implement tmpnam()
This implementation is not really spec compliant as rand() does not
guarantee TMP_MAX different outputs + seeding
2024-07-30 12:02:30 +03:00
Bananymous 62db9a8ef3 LibC: Implement C++ static guards 2024-07-30 12:02:05 +03:00
Bananymous f0be4f86a6 LibC/Kernel: Implement access 2024-07-30 11:51:07 +03:00
Bananymous e5bb843059 LibC/Kernel: Implement ttyname 2024-07-30 11:41:18 +03:00
Bananymous ca774dfeb5 LibC: Implement setlocale() for C locale 2024-07-30 11:33:02 +03:00
Bananymous 681d8327f5 LibC/Kernel: Cleanup termios code
This is still not correct, but much better than it used to be
2024-07-30 11:10:43 +03:00
Bananymous 9bc02c81f8 Kernel: Map DMA and PCI MMIO as uncached 2024-07-30 11:10:08 +03:00
Bananymous bb1738db8c Kernel: Make thread unblocking O(1)
This is still bit broken. VirtualBox seems to freeze sometimes, but I
could not recreate this on qemu (with and without kvm) or real hardware.
2024-07-24 00:31:01 +03:00
Bananymous 9548c592a3 Kernel: Always print basic info about xHCI controller
Its nice to see that the controller is getting initalized
2024-07-23 09:09:08 +03:00
Bananymous cda0276d39 Kernel: Force PCI irq line usage when not using APIC
afaik PIC does not support MSI
2024-07-23 09:08:00 +03:00
Bananymous 539afb329a Kernel: All processors use LAPIC timer when running with APIC
This makes scheduler preemption much cleaner as bsb does not have to
send smp messages to notify other processes about timer interrupt.

Also PIT percision is now "full" 0.8 us instead of 1 ms that I was using
before.
2024-07-23 02:28:52 +03:00
Bananymous 3e0150f847 Kernel: Pressing F1 now toggles rendering of CPU loads to the terminal
This can be nice for seeing the performance and problems on the load
balancing algorithm.
2024-07-22 00:43:13 +03:00
Bananymous f8261c60c0 Kernel: Rewrite the whole scheduler and re-architecture SMP handling
Change Semaphore -> ThreadBlocker
  This was not a semaphore, I just named it one because I didn't know
  what semaphore was. I have meant to change this sooner, but it was in
  no way urgent :D

Implement SMP events. Processors can now be sent SMP events through
IPIs. SMP events can be sent either to a single processor or broadcasted
to every processor.

PageTable::{map_page,map_range,unmap_page,unmap_range}() now send SMP
event to invalidate TLB caches for the changed pages.

Scheduler no longer uses a global run queue. Each processor has its own
scheduler that keeps track of the load on the processor. Once every
second schedulers do load balancing. Schedulers have no access to other
processors' schedulers, they just see approximate loads. If scheduler
decides that it has too much load, it will send a thread to another
processor through a SMP event.

Schedulers are currently run using the timer interrupt on BSB. This
should be not the case, and each processor should use its LAPIC timer
for interrupts. There is no reason to broadcast SMP event to all
processors when BSB gets timer interrupt.

Old scheduler only achieved 20% idle load on qemu. That was probably a
very inefficient implementation. This new scheduler seems to average
around 1% idle load. This is much closer to what I would expect. On my
own laptop idle load seems to be only around 0.5% on each processor.
2024-07-22 00:33:50 +03:00
Bananymous 9f90eeab05 Kernel: Fix stack pointer OOB check
i686 does not push the stack pointer on interrupt when no CPL change
happens.
2024-07-22 00:27:08 +03:00
Bananymous 1ee37cb671 Kernel: Make sure stack is aligned on interrupts for i686 target
I just realized that only x86_64 aligns stack pointer when interrupt is
triggered.
2024-07-22 00:10:42 +03:00
Bananymous 8fe798de6d Kernel: Make PIT counter atomic with spinlock 2024-07-21 18:56:41 +03:00
Bananymous e00efca170 Userspace: Pprogram template and loadfont mark install as optional 2024-07-21 17:41:40 +03:00
Bananymous 86b6714777 Kernel: Add command line option to boot without smp 2024-07-21 17:40:57 +03:00
Bananymous 77b5e6d44a Kernel: Make ATABus use BAN::Atomic<> instead of gcc builtin atomics 2024-07-21 17:39:26 +03:00
Bananymous ffe73165f9 Kernel: Fix PCI typo serail -> serial 2024-07-21 17:36:29 +03:00
Bananymous 96c7e9e29d Kernel: Cleanup VirtualRange code
Virtual range does not need to look into the page table for mapped
physcial pages. It can (and should) just keep track of them itself.
2024-07-21 17:35:07 +03:00
Bananymous 02051ed60f Kernel: Move keyboard kernel panic to the general input device
This will allow command to work without requiring it per keyboard driver
2024-07-21 17:33:38 +03:00
Bananymous f49689caac Kernel: Implement API for sending IPIs for only one processor
Also move the spin wait to check for pending IPI sends to the start of
ipi sending function. There is no need to wait for IPI send succeeding
right after it. It is enough to make sure there are not multiple IPIs
being sent at the same time.
2024-07-21 17:30:59 +03:00
Bananymous 819c130366 BAN: If format string is started with space, numbers are padded with it 2024-07-19 21:39:31 +03:00
Bananymous 974b9b992d Kernel: Fix process exiting infinite loop
If process was exiting and had a process waiting for it and that waiting
process got interrupted, the exiting process could never exit.
2024-07-18 21:40:44 +03:00
Bananymous 3ab1214012 LibC: Fix _toupper and _tolower definitions
They were using C++ global namespace, which of course does not exist for
C targets.
2024-07-18 01:10:51 +03:00
Bananymous 4b917390ac Kernel: Fix sse state saving
This was broken when I added SMP support. This patch makes sse kind of
dumb as it is saved and restored on every interrupt, but now it at least
works properly... I'll have to look into how sse can get optimized
nicely with SMP. Simple way would be pinning each thread to a specific
processor and doing pretty much what I had before, but sse thread saved
in processor rather than static global.
2024-07-16 23:15:11 +03:00
Bananymous 7a0fb9a57f Kernel: Fix TTY scroll clearing first and/or last characters from line 2024-07-16 22:49:49 +03:00
Bananymous 58fcd2b2fe Kernel: Fix multi-interface USB device initialization 2024-07-16 22:29:18 +03:00
Bananymous bdcf058e52 README: update discord link 2024-07-16 21:10:32 +03:00
Bananymous 61f0043cd8 resolver: Fix infinite debug printing loop, if client disconnects 2024-07-16 15:05:07 +03:00
Bananymous dd7bfba8d5 ports: Compile with multiple jobs and fix curl install query 2024-07-16 15:04:46 +03:00
Bananymous 5b8fdbc82d LibC: Don't expose math.h in stdlib.h 2024-07-16 12:47:38 +03:00
Bananymous 6a1c677cbc Ports: Update doom port gitignore 2024-07-16 11:54:57 +03:00
Bananymous e30952efee Kernel: Implement key repeating for USB keyboard
USB keyboards don't send repeating keys automatically, so it has to be
emulated in software.
2024-07-16 11:54:57 +03:00
Bananymous fd8dc03ae9 BAN: Default initialize optional storage
This removes maybe uninitialized warning when using default initialized
optionals.
2024-07-16 11:19:16 +03:00
Bananymous 1337758660 Kernel: Make USB HID interfaces configure all endpoints 2024-07-16 00:23:26 +03:00
Bananymous 339e8a7910 Kernel: Fix USB keyboard state updating with variable reports 2024-07-16 00:03:00 +03:00