- remove special treatment of pager_map mappings in pmaps. this is
required now, since I've removed the globals that expose the address range.
pager_map now uses pmap_kenter_pa() instead of pmap_enter(), so there's
no longer any need to special-case it.
- eliminate struct uvm_vnode by moving its fields into struct vnode.
- rewrite the pageout path. the pager is now responsible for handling the
high-level requests instead of only getting control after a bunch of work
has already been done on its behalf. this will allow us to UBCify LFS,
which needs tighter control over its pages than other filesystems do.
writing a page to disk no longer requires making it read-only, which
allows us to write wired pages without causing all kinds of havoc.
- use a new PG_PAGEOUT flag to indicate that a page should be freed
on behalf of the pagedaemon when it's unlocked. this flag is very similar
to PG_RELEASED, but unlike PG_RELEASED, PG_PAGEOUT can be cleared if the
pageout fails due to eg. an indirect-block buffer being locked.
this allows us to remove the "version" field from struct vm_page,
and together with shrinking "loan_count" from 32 bits to 16,
struct vm_page is now 4 bytes smaller.
- no longer use PG_RELEASED for swap-backed pages. if the page is busy
because it's being paged out, we can't release the swap slot to be
reallocated until that write is complete, but unlike with vnodes we
don't keep a count of in-progress writes so there's no good way to
know when the write is done. instead, when we need to free a busy
swap-backed page, just sleep until we can get it busy ourselves.
- implement a fast-path for extending writes which allows us to avoid
zeroing new pages. this substantially reduces cpu usage.
- encapsulate the data used by the genfs code in a struct genfs_node,
which must be the first element of the filesystem-specific vnode data
for filesystems which use genfs_{get,put}pages().
- eliminate many of the UVM pagerops, since they aren't needed anymore
now that the pager "put" operation is a higher-level operation.
- enhance the genfs code to allow NFS to use the genfs_{get,put}pages
instead of a modified copy.
- clean up struct vnode by removing all the fields that used to be used by
the vfs_cluster.c code (which we don't use anymore with UBC).
- remove kmem_object and mb_object since they were useless.
instead of allocating pages to these objects, we now just allocate
pages with no object. such pages are mapped in the kernel until they
are freed, so we can use the mapping to find the page to free it.
this allows us to remove splvm() protection in several places.
The sum of all these changes improves write throughput on my
decstation 5000/200 to within 1% of the rate of NetBSD 1.5
and reduces the elapsed time for "make release" of a NetBSD 1.5
source tree on my 128MB pc to 10% less than a 1.5 kernel took.
Remove some overzealous locking of HEAD_TO_MAP
Remove a potential deadlock in pmap_copy_page
Change alloc and free l1pt to use kenter/kremove.
Update pmap_map to use kenter (only actually used by dumpsys, so no matching kremove)
This will allow improvements to the pmaps so that they can more easily defer expensive operations, eg tlb/cache flush, til the last possible moment.
Currently this is a no-op on most platforms, so they should see no difference.
Reviewed by Jason.
Thorpe as an extension to the bus_space(9) API.
Only the mainbus has a working mmap... iomd, podulebus and isa dont have
one yet. Its trivial to add the iomd and isa's memory part but am a bit
hessistant to add it without knowing its implications by heart.
Podulebus might be implemented in 32 bits EASI but otherwise its not really
transparent since there is AFAIK no function to explicitly read one byte or
so .. or is this implemented in a higher level ?
ARM26 and arch/arm32/dnard (Shark) kernels compile and RiscPC and Shark
kernels are tested and function OK.
Implications of other ARM systems like hpcarm and ebarm are propably
minimal or solved trivially.
ELF _C_LABEL(irqhandlers) == irqhandlers so we were getting self-referential
pointer which causes the IRQ to fall on itself. [With this change the
netwinder kernel can now be *ELF* and boots to multiuser]
Improved locking (not that we actually use it on a uniprocessor, but one day :)
Removed unneeded splvm's
tweaked pmap_clean_page code to only flush the cache if the page is mapped in the current pmap (based on diff from richard E)
Adopted pv entry allocation mechanism from i386.
Laid framework for returning ptp's when we've finished with them rather than holding onto them till the process exits.
ptp's are now allocated with a uvm object for the pmap, means that we can walk a list to free them off in pmap_release, until they get freed off by pmap_remove.
Also implemented a page zeroing function when the processor is idling. Note that hpcarm may wish to disable this.
I believe this code to be stable, if anyone has any problems please shout up.