| Age | Commit message (Collapse) | Author |
|---|
|
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
The map callback within the MAC ops returns ssize_t, therefore the
mac_map() wrapper should return the same
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
This commit introduces initial support for mandatory access control. As
one may recall, L5 follows "everything is memory". In order to interact
with a resource, a process must request it from the kernel in the form
of a (sometimes) syncable memory buffer.
Each resource as well as processes have an access level, if a process
attempts to request a resource with a higher access level than it, the
request is rejected by the kernel. However, if a process has a greater
than or equal access level as a resource, the request can be granted.
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Typically filesystems are tree-like, however certain filesystems in the
form of images have fixed paths baked into them. This optimizes handling
so there isn't a need to unpack it into a rootfs/tmpfs
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
This commit introduces a new framework we call "clkdev". It is used to
keep track of the clock sources on the system and be able to fetch them
later based on specific attributes.
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Here we add a write(2) stub and add it to the UNIX syscall interface. We
also move the UNIX syscall numbers into compat/unix/syscall.h
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Add routine to check if an address is valid within a process's address
space. More work will need to be done with this but this introduces a
great starting point
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
This commit adds a virtual memory range queue to the process descriptor
in order to keep track of mapped pages and free their respective frames
upon program exit.
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Introduce a namei() prototype function to build upon when implementing
core namei() logic. The prototype simply implemenets the path parsing
and mimicks a UNIX-like (shell) namei output.
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Add helper function to lookup mountpoints such as '/'
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
This commit introduces the groundwork for mountpoints, filesystems and
the VFS as a whole. OMAR is now initialized as its own filesystem by the
VFS
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Introduce routines to allocate (create) and deallocate (destroy) virtual
file nodes.
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
A syscall domain in the L5 kernel is a fixed list of "syscall windows",
each syscall window represents a specific platform and/or syscall model.
A platform latch within each domain determines which window / platform
should be visible. Since syscall domains are per-process, these changes
are local to their respective processes.
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Add a new function to kill processes and clean them up. They will then
be marked as EXITING and it is up to the parent to do the rest.
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
This commit introduces load balancing of processes between cores by
using a scalable "CPU arbiter" which decides how to fetch the next core
descriptor.
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Use the OSMORA kernel configuration framework to add options for L5Lunos
to parse during early startup.
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Allocate more than one frame that exists and increment by one if zero.
We also did some style fix ups, added an #undef and a break.
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
This commit introduces the initial logic for putting process descriptors
in an initialized state. This commit also introduces the process control
block definition.
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
This commit implements the task state segment and splits up processor
initialization into two seperate stages. The cpu_conf() function is
apart of the first stage and sets up things that should be going by the
time the kernel is started / early init. The cpu_init() function runs
later functions that initialize further platform specific subsystems.
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
Signed-off-by: Ian Moffett <ian@osmora.org>
|
|
- Wrap on X overflow
- Wrap on Y overflow
- Keep track of max console width and height
- Add console enable/disable control
Signed-off-by: Ian Moffett <ian@osmora.org>
|
