Removed dependency on hard-coded pager id. Pager id is now passed
as an environment string `pagerid' to tasks. Alternatively, this
could take space in the utcb of each task.
- Need to remove old versions
- Need to merge the two.
- Need to investigate occasional page fault on NMOP sequence. (resembles an error ipc_extended test)
Could be related to new page cache read/write routines.
Any thread that touches a utcb inside the kernel now properly checks
whether the utcb is mapped on its owner, and whether the mapped physical
address matches that of the current thread's tables. If not the tables
are updated.
This way, even though page tables become incoherent on utcb address
change situations (such as fork() exit(), execve()) they get updated
as they are referenced.
Since mappings are added only conditionally, caches are flushed only
when an update is necessary.
Added UTCB functions for full ipc that copy buffer or string payloads
in secondary message registers.
Some posix syscalls now use these utcb functions to copy pathnames.
test0 now successfully runs its beginning.
test0 SConscript has a dependency problem.
Issues to be investigated:
- vm_file and vnodes need to be merged fully in all functions.
- libposix shared page references need to be removed.
- Any references to VFS_TID, PAGER_TID need to be removed.
Changes:
It is now possible to use do_mmap() from within mm0.
- pager_new_virtual()/delete_virtual() return addresses that are
disjoint from find_unmapped_area() used by mmap() interface for
anonymous or not-fixed areas.
- find_unmapped_area() now uses task->map_start task->map_end instead
of task->start and task->end. task->start/end are still valid task
space addresses for mmap(), but finding a new address is limited to
map_start/map_end.
- We have both interfaces because mmap() is only useful for backed-files.
When the pager needs to access a user memory range for example, that is
not backed by a file and thus we need to use pager_new_virtual() instead
of mmap() for mapping.
Now worked around.
CML2 seems to forget supressing derived symbols that are derived from
suppressed symbols. (Expected behaviour) This caused the configuration parser
to recognize the illegally visible symbols to be from those containers.
Now worked around by adding the extra check for CONFIG_CONTAINERS have been parsed.
A real solution requires fixing CML2 symbol visibility function.
It might be a good idea to simply use mmap() from inside the pager and prefault it
as a conventional way of mapping internal buffers. Will be investigated.
- This would open the way that all internal buffer mapping is
done in a standard way
- Standard syscalls would be possible to use from within the pager.
CML boolean type compiles into #define and #undefs.
This is not useful for capabilities since a value of 0 or 1 can be
used automatically for setting or unsetting bits in permission words.
Meanwhile using an integer type with 0 - 1 range has a worse usability
experience than the boolean Y, N.
Also moved parsing of container.template.cml to newer Python argument
parsing format.
