Make sure we handle all possible ways regions can overlap when EX_CONFLICTOK
is specified.

ok miod@

Fix userland regression tests by providing a #define for pool_setipl()
(just a noop since that doesn't matter in userland).

Pointed out by a couple of people, thanks.

don't use splvm ourselves, use pool_setipl on the pool we're allocating
from and it will do it for us.

ok miod@

Add a new EX_FILLED flag to make extent_create() create an extent map that
has all space allocated such that we can make holes in it using extent_free().

ok miod@

Add EX_CONFLICTOK flag that allows an allocation to conflict with existing
allocations, making sure that the union of all space is allocated.

ok deraadt@

Make extent_print() use printf to avoid the pagination that db_printf
provides, while keeping this behaviour for extent_print_all() which is
only called by ddb. Based on a diff from deraadt@.

First pass at removing clauses 3 and 4 from NetBSD licenses.

Not sure what's more surprising: how long it took for NetBSD to
catch up to the rest of the BSDs (including UCB), or the amount of
code t

First pass at removing clauses 3 and 4 from NetBSD licenses.

Not sure what's more surprising: how long it took for NetBSD to
catch up to the rest of the BSDs (including UCB), or the amount of
code that NetBSD has claimed for itself without attributing to the
actual authors.

OK deraadt@

show more ...

In extent_destroy(), only remove the extent from the global extent list if
the list is really used, i.e. if option DIAGNOSTIC || option DDB.

Turn EXTENT_ALIGN into an inline function; shaves about 50 to 100 bytes
on RISC arches, but m68k loses a few bytes; ok deraadt@

Remove unused extent_find(). Then, we can keep the extent list only if
option DIAGNOSTIC || option DDB, saves some room on installation media.

ansi and deregister. No binary change.

Remove unused extent_insert_and_optimize() parameter 'flags'. Spotted
by lint.

ok art@ miod@

Correctly handle allocations of size 1 at an extent boundary; from NetBSD.

Make the non-_KERNEL compilation usable again, in order to add a regress test.

Use list and queue macros where applicable to make the code easier to read;
no change in compiler assembly output.

compile non-DIAGNOSTIC kernels

- Use LIST_ macros.
- Make sure that extent_register is not called more than once on an extent.

Use %b and print the bitmask of the flags in the debugging message for the extent.

splvm, not splimp.

Two patches from NetBSD:
- extent_alloc_subregion shouldn't allocate a region pasts the end of subregion.
- extent_alloc_subregion1(): when a candidate region doesn't fit after being
rounded up to re

Two patches from NetBSD:
- extent_alloc_subregion shouldn't allocate a region pasts the end of subregion.
- extent_alloc_subregion1(): when a candidate region doesn't fit after being
rounded up to respect boundary limits, adjust newstart and last before
skiping to the next region. Otherwise we may check the same candidate
region against the start of the next region, no the one immediatly following
the hole, leading to corrupted map.

show more ...

First round of __P removal in sys

Pool deals fairly well with physical memory shortage, but it doesn't deal
well (not at all) with shortages of the vm_map where the pages are mapped
(usually kmem_map).

Try to deal with it:
- group

Pool deals fairly well with physical memory shortage, but it doesn't deal
well (not at all) with shortages of the vm_map where the pages are mapped
(usually kmem_map).

Try to deal with it:
- group all information the backend allocator for a pool in a separate
struct. The pool will only have a pointer to that struct.
- change the pool_init API to reflect that.
- link all pools allocating from the same allocator on a linked list.
- Since an allocator is responsible to wait for physical memory it will
only fail (waitok) when it runs out of its backing vm_map, carefully
drain pools using the same allocator so that va space is freed.
(see comments in code for caveats and details).
- change pool_reclaim to return if it actually succeeded to free some
memory, use that information to make draining easier and more efficient.
- get rid of PR_URGENT, noone uses it.

show more ...

Use pool to allocate extent region descriptors.

In extent_destroy, remove the extent from the list of
all extents.

Get rid of the wrapper macros around extent_alloc*1
Pass the right amount of arguments and rename them back to their right names.