Call configure() directly immediately after config_init().
This causes autoconfiguration to happen at the same time as before, but
creates some kernel submaps earlier, so that e.g. mbinit() can now
allocate memory.
- Protect userspace from unnecessary header inclusions (as noted on
current-users).
- Some const poisioning.
- GREATLY simplify the locking protocol, and fix potential deadlock
scenarios. In particular, assume that the back-end page allocator
provides its own locking mechanism (this is currently true for all
such allocators in the NetBSD kernel). Doing so allows us to simply
use one spin lock for serialized access to all r/w members of the pool
descriptor. The spin lock is released before calling the back-end
allocator, and re-acquired upon return from it.
- Fix a problem in pr_rmpage() where a data structure was referenced
after it was freed.
- Minor tweak to page manaement. Migrate both idle and empty pages
to the end of the page list. As soon as a page becomes un-empty
(by a pool_put()), place it at the head of the page list, and set
curpage to point to it. This reduces fragmentation as well as the
time required to find a non-empty page as soon as curpage becomes
empty again.
- Use mono_time throughout, and protect access to it w/ splclock().
- In pool_reclaim(), if freeing an idle page would reduce the number
of allocatable items to below the low water mark, don't.
- there are a couple of registers that are reset after the X server finishes,
which do not have the same values as they do when the machine is powered
on. If the two are mismatched you can't write to the video memory.
- video memory is mapped at offset 4MB to keep the X server happy
- there is now a VGA memory mapping
- DEBUG_ET4000 is now a bit more useful for debugging!
call to __main(), and therefore saves the size of the call and the
size of a stub implementation of __main().
in the primary boot block, don't bother saving/restoring the argument
passed in from the caller. There is no such argument (that we care
about, at least) to the primary. (for secondary, it's the firmware
FD being used.)
Clean up the "Region 1" related definitions, and define load addresses,
max load size, and max total size for as many boot block types as we can.
(types = unified, primary, secondary). We can't always define all
values for all boot blocks, though.
Make CPP flags selection less gross.
Use objcopy rather than headersize (yay, evil gets a stake to the heart!).
Use a little shell script to verify that the sizes of the boot blocks are OK.
Do not compile too much more of libsa than we actually have to.
size to be reduced substantially. (backward compatibility verified
by compiling one of the alpha boot blocks which uses all of the code
before and after, diffing the object files, and manually verifying that
the differences were 'correct'. some differences were "unavoidable,"
it wanting to avoid a double-commit, because e.g. local variables which
were previously used were no longer used.) a README which describes
supported options (or at least the ones mentioned below) is forthcoming.
add support for the preprocessor macro LIBSA_NO_TWIDDLE, which
causes calls to twiddle() to be omitted if it's defined.
add support for the preprocessor macros:
LIBSA_NO_FS_CLOSE
LIBSA_NO_FS_WRITE
LIBSA_NO_FS_SEEK
which, if defined, cause the corresponding file system operations
in the individual file system implementations to be omitted. (note
that all of those macros are not supported by all file systems at
this point. comments were added to individual file system files
to indicate lack of support, and should be cleaned up later. Backward
compatibility options e.g. UFS_NOCLOSE, etc., are supported.)
add support for the preprocessor macro LIBSA_NO_FS_SYMLINK, which
removes support for symbolic links from the file system support
functions. (same notes as for the macros above apply.)
add support for the preprocessor macro LIBSA_FS_SINGLECOMPONENT which
removes all subdirectory and symlink support from the file system
support functions. (same notes as for the macros above apply.)
add support for the preprocessor macro LIBSA_NO_FD_CHECKING, which
causes code relating to libsa file descriptor checks (e.g. range
checking and checking that a file descriptor is valid) to be
omitted if it's defined.
add support for the preprocessor macro LIBSA_NO_RAW_ACCESS, which
causes code relating to raw device access to be omitted if it's
defined.
change some structure copies to use bcopy() instead. that way
use of bcopy vs. memcpy() can easily be selected by
LIBSA_USE_MEMCPY. (without changes like these, you could end up
having both bcopy() and memcpy() included. eventually, all
calls to bcopy should be changed to calls to memcpy() or memmove()
as appropriate -- hopefully never the latter -- with an option to
use bcopy instead.)
add support for the preprocessor macro LIBSA_NO_DISKLABEL_MSGS, which
causes disklabel() to return '1' as msg rather than a string. Can
be used if the boot blocks don't care about the string, and need to
save the space.
add support for the preprocessor macro LIBSA_SINGLE_FILESYSTEM, which
if defined causes all of the file system switch code to be removed.
Its value should be the name of the file system supported by the
boot block, e.g. "ufs" for the FFS file system. calls to the
file system functions open, close, etc., which were previously
done through a function switch are then done via direct invocation
of <fs>_open, <fs>_close, etc. (e.g. ufs_open, ...).
add support for the preprocessor macro LIBSA_SINGLE_DEVICE, which
does the equivalent of LIBSA_SINGLE_FILESYSTEM but for the device
switch table. Device entry pointes are expected to be named
<dev>foo, e.g. the 'strategy' routine used when LIBSA_SINGLE_DEVICE
is set to 'disk' is diskstrategy.
make ufs.c f_nindir array be unsigned ints. the fact that it was signed
caused ufs.c to require signed division routines (which were otherwise
unnecessary for a small boot block).
a small implementation of memcpy(). libsa memcpy() wouldn't
do the right thing if LIBSA_USE_MEMCPY was defined, and the whole
point of that define is to get rid of either bcopy() or memcpy().
(cloned from the bcopy() code.)
NMBCLUSTERS for the mbuf cluster pool. On platforms which use direct-mapped
segments for pool pages (MIPS and Alpha), this makes NMBCLUSTERS actually
meaningful (such ports don't even allocate mb_map, as it is not used to
map mbuf cluster pages).
Improve the message logged at a maximum rate of once per second. The
new message: "WARNING: mclpool limit reached; increase NMBCLUSTERS".
In the back-end pool page allocator, remove the message about mb_map
being full. The message was not necessarily correct as the allocator
may have been starved for pages, rather than for space in the map. Also,
the hard limit on the mbuf cluster pool will be reached before the map
fills (the last cluster will always fit into the map), so the message
is redundant.
Add a comment in mbinit() about considering setting low water marks on
the mbuf and mbuf cluster pools.
- Add support for hard limits, with optional rate-limited logging of
a warning message when the pool limit is reached. (This will be used
to fix a bug in mbuf cluster allocation on the MIPS and Alpha ports.)
- Fix some locking protocol errors. This required splitting pr_flags
into pr_flags (which is protected by the spin lock) and pr_roflags (which
are `read only' flags, set when the pool is initialized, and never changed
again; these do not need to be protected by a mutex).
- Make the low water support actually mean something. When a low water
mark is set, add free items to the pool until the low water mark is
reached. When an item allocation causes the number of free items to
drop below the low water mark, make the pool catch up to it. This can
make the pool allocator more useful for several applications (e.g.
pmap `pv entry' management) and more robust for others (for e.g. mbuf
and mbuf cluster allocation, so that the pagedaemon can use NFS to clean
pages on diskless systems without completely running dry on buffers to
receive packets in during extreme memory shoratages).
- Add a comment where we sleep waiting for more pages for the back-end
page allocator. Specifically, instead of sleeping potentially forever,
perhaps we should just wake up once a second to try allocating a page
again. XXX Revisit this soon.
