This code has very ugly file parsing code. I have to create API
for reading files line by line in kernel space...
This allows users to open framebuffer/input files without root.
Mounting has to be moved to userspace soon. It makes no sense to
hard code permissions for every (device) file.
Use this new FramebufferTerminalDriver for terminal instead of the
old VesaTerminalDriver. Only drawback with this is that framebuffer
device can only be intialized after DevFS is initialized.
SATA drives can now be used with banan-os. This allows much faster
disk access (writing 10 MiB from 30s to 1.5s). This can definitely
be optimized but the main slow down is probably the whole disk
structure in the os.
AHCI drive is now the default when running qemu.
MemoryBackedRegion now inherits from this and is used for private
anonymous mappigs. This will make shared mappings and file backed
mappings much easier to implement.
All executable files are now read from disk and paged on demand.
This was a big rewrite of the old ELF library but in the end
everything seems much cleaner, since all the old functionality was
not actually needed for execution.
I have to do some measurements, but I feel like memory usage dropped
quite a bit after this change.
I could delete the whole FixedWidthAllocator as it was now obsolete.
GeneralAllocator is still used by kmalloc. Kmalloc cannot actually
use it since, GeneralAllocator depends on SpinLock and kmalloc runs
without interrupts.
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.
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.
We now have DevFileSystem which is derived from RamFileSystem. All
devices are RamInodes. We don't have separate DeviceManager anymore.
To iterate over devices, you can loop througn every inode in devfs.
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.
We used to copy all headers everytime to sysroot which caused
rebuild of the whole os. Now we use the cmake command
'copy_directory_if_different' which seemed to fix this issue :)
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.
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
I have been annoyed for a while since I had to build everything
when running the os since the buildsystem was really bad.
I wanted to rewrite the whole build system and changed to using cmake
:)
Bananymous
Bananymous