Have added a drivers/serial folder to kernel, where all serial drivers
will be kept. The point is to have the the dts parsed and generate cmake
to include the right uart.c file prefixed with the compatibility.
Have removed all io.c from plat and includes from plat/config.cmake and
updated CHANGES file.

Currently dts support is only for arm so makes
sense to move into arm config.cmake. Will eventually add RISC-V to
the dts list too.

Prio to this change the kernel would crash if a thread with no VCPU
attempted to seL4_VMEnter. As of this commit, an unknown syscall exception
will be raised and the kernel will not crash.

Rename plat_cleanInvalidateCache to plat_cleanInvalidateL2Cache to
indicate that it is a L2 cache maintenance operation only.

Parse the /reserved-memory node per the Linux spec[1].
Ignore regions marked as 'no-map', but keep other regions
as they should be OK to use as RAM.

[1]: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/Documentation/devicetree/bindings/reserved-memory/reserved-memory.txt?h=v5.0-rc6

These values are now autogenerated.

it's not guaranteed that node names will be globally unique,
so use the full path to the node instead.

this is needed for the BOOT_RODATA attribute

exynos5 platforms have to initialise custom hardware before initialising
the generic timer. This was forcing every other platform to call a
wrapper function. We instead define initTimer that exynos5 can override.

Previously seL4_PGDBits was calculated, which is inconstent with the rest
of the composite constants in libsel4, and makes size constants harder
to parse from the C.

This commit makes seL4_PGDBits consistant with other definitions, by making the definition an integer and checking it with a compile time assert.

This file is based on foundation-v8-gicv3.dts.

These have no effect when there aren't any definitions.

This allows for mapping of consecutive pages for a single device in the
kernel.

We currently only support Sv39 on RV64. Sv39 has a cannonical address
range of -2^38 and 2^38-1. This means that bits 63-39 must be equal to
bit 38, or that we sign extend from bit 39 to 63 based on the value of
bit 38. It also means that we only have 39 bits of addressable space,
which limits our max untype size.

These shared objects have different definitions based on the cannonical
address of the virtual memory system.

base 32 implies base of 32 and mask of 32 with no sign extension which
is required for Sv32

Conventionally a newline is used to signify a system word boundary. This
is supposed to improve readability

The assert has an assumption that regions are also ordered in ascending
order.

This change fixes support for instances where we have
multiple kernel devices in the same page, or kernel devices
which aren't at page-aligned addresses.

Also use seL4_UserTop to pick the right address to start
putting the kernel device pages.

Note that the auto-generated kernel_devices will differ slightly from
the ones present in the kernel until now. When devices have
registers that aren't page-aligned we now always set the appropriate
PPTR to be the start address of the device, while previously the PPTR
was sometimes page-aligned.

Specifically, this change to PPTRs affects:
- bcm2837 intc
- bcm2837 uart
- allwinnerA20 timer

KDEV_BASE is the first address that can be used for mapping devices.
This will be used by hardware_gen to auto-generate the kernel_devices
table.

This header is going away and contains nothing of substance. Remove it.

This removes the need for the buses array in the hardware YAML

This moves the kernel back up to 0x60000000 as it was before
the HW headers were autogenerated, as this behaviour is depended on by
some user-level projects.

The CAmkES VM wants these. These definitions are incomplete, but
are enough that the kernel will pass the appropriate memory regions
through to userspace.

Merge overlapping regions to ensure that we don't expose
the same paddr more than once.

If one region is conditional and another isn't, we shouldn't merge them.
This fixes a problem where some regions wouldn't be exposed to
userspace when they were merged with a conditional region which would
cause userspace apps to fail in unexpected ways.