dillon a179ee09ab This patch implements O_DIRECT about 80% of the way. It takes a patchset ... Tor created a while ago, removes the raw I/O piece (that has cache coherency problems), and adds a buffer cache / VM freeing piece. Essentially this patch causes O_DIRECT I/O to not be left in the cache, but does not prevent it from going through the cache, hence the 80%. For the last 20% we need a method by which the I/O can be issued directly to buffer supplied by the user process and bypass the buffer cache entirely, but still maintain cache coherency. I also have the code working under -stable but the changes made to sys/file.h may not be MFCable, so an MFC is not on the table yet. Submitted by: tegge, dillon		2001-05-24 07:22:27 +00:00
..
default_pager.c	Introduce a global lock for the vm subsystem (vm_mtx).	2001-05-19 01:28:09 +00:00
device_pager.c	Assert Giant is held for the device pager alloc and getpages methods since	2001-05-23 22:27:52 +00:00
phys_pager.c	Set the phys_pager_alloc_lock to 1 when it is acquired so that it is	2001-05-23 19:52:23 +00:00
pmap.h	Unrevert the pmap_map() changes. They weren't broken on x86.	2001-03-07 05:29:21 +00:00
swap_pager.c	- Fix the sw_alloc_interlock to actually lock itself when the lock is	2001-05-23 22:31:15 +00:00
swap_pager.h	- If swap metadata does not fit into the KVM, reduce the number of	2000-12-13 10:01:00 +00:00
vm_extern.h
vm_fault.c	Take a more conservative approach and still lock Giant around VM faults	2001-05-23 22:09:18 +00:00
vm_glue.c	- Lock the VM around the pmap_swapin_proc() call in faultin().	2001-05-23 22:35:45 +00:00
vm_init.c	Sort includes from previous commit.	2001-05-22 05:35:45 +00:00
vm_kern.c	Introduce a global lock for the vm subsystem (vm_mtx).	2001-05-19 01:28:09 +00:00
vm_kern.h
vm_map.c	- Add lots of vm_mtx assertions.	2001-05-23 22:38:00 +00:00
vm_map.h	Introduce a global lock for the vm subsystem (vm_mtx).	2001-05-19 01:28:09 +00:00
vm_meter.c	Introduce a global lock for the vm subsystem (vm_mtx).	2001-05-19 01:28:09 +00:00
vm_mmap.c	- Obtain Giant in mmap() syscall while messing with file descriptors and	2001-05-23 22:17:43 +00:00
vm_object.c	- Assert that the vm lock is held for all of _vm_object_allocate().	2001-05-23 22:42:10 +00:00
vm_object.h	Introduce a global lock for the vm subsystem (vm_mtx).	2001-05-19 01:28:09 +00:00
vm_page.c	This patch implements O_DIRECT about 80% of the way. It takes a patchset	2001-05-24 07:22:27 +00:00
vm_page.h	This patch implements O_DIRECT about 80% of the way. It takes a patchset	2001-05-24 07:22:27 +00:00
vm_pageout.c	- Add in several asserts of vm_mtx.	2001-05-23 22:48:28 +00:00
vm_pageout.h
vm_pager.c	Introduce a global lock for the vm subsystem (vm_mtx).	2001-05-19 01:28:09 +00:00
vm_pager.h	Introduce a global lock for the vm subsystem (vm_mtx).	2001-05-19 01:28:09 +00:00
vm_param.h
vm_swap.c	Introduce per-swap area accounting in the VM system, and export	2001-02-23 18:46:21 +00:00
vm_unix.c	Acquire Giant around vm_map_remove() inside of the obreak() syscall for	2001-05-23 22:13:10 +00:00
vm_zone.c	Introduce a global lock for the vm subsystem (vm_mtx).	2001-05-19 01:28:09 +00:00
vm_zone.h	Undo part of the tangle of having sys/lock.h and sys/mutex.h included in	2001-05-01 08:13:21 +00:00
vm.h	Introduce a global lock for the vm subsystem (vm_mtx).	2001-05-19 01:28:09 +00:00
vnode_pager.c	- Assert Giant is held in the vnode pager methods.	2001-05-23 22:51:23 +00:00
vnode_pager.h	vnode_pager_freepage() is really vm_page_free() in disguise,	2001-04-19 06:18:23 +00:00