1) The PML2 was not initialized to zeroes, thus leaving some bits behind that
caused the fork code to go crazy, forking the unforkable, and mapping addresses
that never, ever, should have been mapped, leaving behind a trail of page faults
and general protection faults on some computers, while other computers worked
because the uninitalized memory just wasn't uninitialized enough. Yep, this was
a schrödinbug!
2) Fixed a time bomb. The kernel heap was accidentally put such that whenever a
few megabytes were allocated, it would begin overwriting the physical page stack
causing unthinkable events to unfold and would probably be even more obscure to
debug than 1).
Oh, and some string errors fixed and removed RunApplication from kernel.cpp,
funny thing that even linked in the first place. Guess, the optimizer actually
did work for once. :)
Physical paging have been extended with Page::Insert() and Page::GetStats()
which allows the physical paging system to add new pages to the physical page
allocator, and still keep the "free/used pages" count accurate, and providing
this information to the kernel (and user-space at some point).
The virtual memory API has been extended with RemapKernel(), RemapUser(),
MapRangeKernel(), UnmapRangeKernel(), MapRangeUser(), and UnmapRangeUser().
This huge number of related functions have been created in the hope that it
hides the internal complexity of portable virtual memory management and avoid
bugs. It is crucial that the correct group of functions are used when solving
a problem and that they are not mixed in a manner not documented.
I probably overdocumented the code - hopefully it should help avoiding making
stupid or bothersome code.
Another problem is that code calling Page::Get() often should call something
like Page::AlwaysGetPageEvenIfYouHaveToSwap(). I'd be swell to have a function
that always gets a page under heavily-swapping conditions. Possibly Page::Get()
could become that?
uintptr_t is now replaced with addr_t when referring to physical memory
addresses in Sortix. Many bugs in the previous memory management code have been
fixed. The new interface is less flexible - but should prove more solid as the
nasty internals have been hidden away. The current interface design should also
make the code more 64-bit ready/friendly. And so on.
Jonas 'Sortie' Termansen