UserOp_IF had its own way of extracting the XN bit from page tables.
This is now unified with the existing functions in ADT_AI, which also
means that the proof for XN bit equality is basically the same as for
pt_rights and pt_lift.
Some of the noninterference results depend on executions at the haskell level starting at a valid initial state. This file demonstrates this condition being realised.
The fact that the C infoflow adt refines the abstract infoflow adt now only requires that given user operation is nonempty and not sane (nonempty and doesn't return an interrupt).
Also added some more general lemmas about fw_sim and refinement to lib/Simulation.thy.
...and clean up some fallout.
In particular, we now say that the output of "nat_to_bl sz n" is taken
to be the bitlist of "n mod 2^sz", so the output is always defined.
The idea is to remove the undefinedness of "the_nat_to_bl" so that it is
easier to generate simp rules for it; some of these are developed in the
theory below, and simplify some of the more concrete infoflow proofs.
Use the previously-added "distinct" command to simplify the
"Example_Valid_State" proof. This brings quite significant speedups as
it means that raw definitions need not be unfolded, and hence automated
tactics don't get side-tracked with their numerical definitions.
The global-object pointer validity assertion is now created at
export time, and the graph refine mechanism now proves them. It
seems they were forgotten about once again in adjusting the globals
logic.
IRQ Nodes are now their own object type in capDL. This makes it much easier
to distinguish between "real" CNodes and IRQ Nodes.
Updated:
* the capDL refinement,
* the access proofs, and
* the system initialiser.
These final changes complete the SEL4GraphRefine process. Some
cleanup remains to be done, especially in SEL4GlobalsSwap, but the
process is now mature and working, and the testing code
in SEL4GraphRefine can be discarded.
Success depends on seL4 commit 97d6bc96d54f1f0beafb25033b03b57ba54a5113
which is compatible with crefine and will be included in the repo
manifest immediately.
In detail:
- add a general user-specified exception to c_exntype
(for use in tools like Substitute)
- wrap calls to 'halt' in Guard {}, making it clearer that
halt is never called, simplifying asmrefine
- repair halt changes in crefine
- avoid use of some suspicious 'modifies' properties in crefine
which were generated by the parser for functions where inline
ASM blocks have been elided, and which may be inaccurate.
* The separation algebra for capDL is also a cancellative separation algebra.
* The arrows are strictly_exact, meaning they describe only a single heap.
* Since we have a cancellative separation algebra, this means the arrows are also precise.
* The definitions of the separation "arrows" is slightly nicer and more consistent.
- We have a nicer correspondence between sep_map_c and sep_map_s.
- sep_map_irq now specifies exactly what the IRQ table contains
(that it *only* has one entry, not that it contains at least that entry).
- Nicer LaTeX output for the arrows.
* A number of minor renaming of constants and types.
- cdl_component => cdl_component_id
- sep_entity => cdl_component
- state_sep_projection => sep_state_projection
- obj_to_sep_state => object_to_sep_state
* Removed a few unused lemmas.