c7aebda8a1
Unify the 2 concept into a real, minimal, sxlock where the shared acquisition represent the soft busy and the exclusive acquisition represent the hard busy. The old VPO_WANTED mechanism becames the hard-path for this new lock and it becomes per-page rather than per-object. The vm_object lock becames an interlock for this functionality: it can be held in both read or write mode. However, if the vm_object lock is held in read mode while acquiring or releasing the busy state, the thread owner cannot make any assumption on the busy state unless it is also busying it. Also: - Add a new flag to directly shared busy pages while vm_page_alloc and vm_page_grab are being executed. This will be very helpful once these functions happen under a read object lock. - Move the swapping sleep into its own per-object flag The KPI is heavilly changed this is why the version is bumped. It is very likely that some VM ports users will need to change their own code. Sponsored by: EMC / Isilon storage division Discussed with: alc Reviewed by: jeff, kib Tested by: gavin, bapt (older version) Tested by: pho, scottl
168 lines
5.1 KiB
Groff
168 lines
5.1 KiB
Groff
.\" -*- nroff -*-
|
|
.\"
|
|
.\" Copyright (c) 1996 Doug Rabson
|
|
.\" Copyright 2003, Garrett A. Wollman
|
|
.\"
|
|
.\" All rights reserved.
|
|
.\"
|
|
.\" This program is free software.
|
|
.\"
|
|
.\" Redistribution and use in source and binary forms, with or without
|
|
.\" modification, are permitted provided that the following conditions
|
|
.\" are met:
|
|
.\" 1. Redistributions of source code must retain the above copyright
|
|
.\" notice, this list of conditions and the following disclaimer.
|
|
.\" 2. Redistributions in binary form must reproduce the above copyright
|
|
.\" notice, this list of conditions and the following disclaimer in the
|
|
.\" documentation and/or other materials provided with the distribution.
|
|
.\"
|
|
.\" THIS SOFTWARE IS PROVIDED BY THE DEVELOPERS ``AS IS'' AND ANY EXPRESS OR
|
|
.\" IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
|
|
.\" OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
|
|
.\" IN NO EVENT SHALL THE DEVELOPERS BE LIABLE FOR ANY DIRECT, INDIRECT,
|
|
.\" INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
|
|
.\" NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
|
|
.\" DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
|
|
.\" THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
|
|
.\" (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
|
|
.\" THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
|
.\"
|
|
.\" $FreeBSD$
|
|
.\"
|
|
.Dd September 27, 2003
|
|
.Dt VOP_GETPAGES 9
|
|
.Os
|
|
.Sh NAME
|
|
.Nm VOP_GETPAGES ,
|
|
.Nm VOP_PUTPAGES
|
|
.Nd read or write VM pages from a file
|
|
.Sh SYNOPSIS
|
|
.In sys/param.h
|
|
.In sys/vnode.h
|
|
.In vm/vm.h
|
|
.Ft int
|
|
.Fn VOP_GETPAGES "struct vnode *vp" "vm_page_t *ma" "int count" "int reqpage" "vm_ooffset_t offset"
|
|
.Ft int
|
|
.Fn VOP_PUTPAGES "struct vnode *vp" "vm_page_t *ma" "int count" "int sync" "int *rtvals" "vm_ooffset_t offset"
|
|
.Sh DESCRIPTION
|
|
The
|
|
.Fn VOP_GETPAGES
|
|
method is called to read in pages of virtual memory which are backed by
|
|
ordinary files.
|
|
If other adjacent pages are backed by adjacent regions of the same file,
|
|
.Fn VOP_GETPAGES
|
|
is requested to read those pages as well, although it is not required to
|
|
do so.
|
|
The
|
|
.Fn VOP_PUTPAGES
|
|
method does the converse; that is to say, it writes out adjacent dirty
|
|
pages of virtual memory.
|
|
.Pp
|
|
On entry, the vnode lock is held but neither the page queue nor VM object
|
|
locks are held.
|
|
Both methods return in the same state on both success and error returns.
|
|
.Pp
|
|
The arguments are:
|
|
.Bl -tag -width reqpage
|
|
.It Fa vp
|
|
The file to access.
|
|
.It Fa ma
|
|
Pointer to the first element of an array of pages representing a
|
|
contiguous region of the file to be read or written.
|
|
.It Fa count
|
|
The number of bytes that should be read into the pages of the array.
|
|
.It Fa sync
|
|
.Dv VM_PAGER_PUT_SYNC
|
|
if the write should be synchronous.
|
|
.It Fa rtvals
|
|
An array of VM system result codes indicating the status of each
|
|
page written by
|
|
.Fn VOP_PUTPAGES .
|
|
.It Fa reqpage
|
|
The index in the page array of the requested page; i.e., the one page which
|
|
the implementation of this method must handle.
|
|
.It Fa offset
|
|
Offset in the file at which the mapped pages begin.
|
|
.El
|
|
.Pp
|
|
The status of the
|
|
.Fn VOP_PUTPAGES
|
|
method is returned on a page-by-page basis in the array
|
|
.Fa rtvals[] .
|
|
The possible status values are as follows:
|
|
.Bl -tag -width VM_PAGER_ERROR
|
|
.It Dv VM_PAGER_OK
|
|
The page was successfully written.
|
|
The implementation must call
|
|
.Xr vm_page_undirty 9
|
|
to mark the page as clean.
|
|
.It Dv VM_PAGER_PEND
|
|
The page was scheduled to be written asynchronously.
|
|
When the write completes, the completion callback should
|
|
call
|
|
.Xr vm_object_pip_wakeup 9
|
|
and
|
|
.Xr vm_page_sunbusy 9
|
|
to clear the busy flag and awaken any other threads waiting for this page,
|
|
in addition to calling
|
|
.Xr vm_page_undirty 9 .
|
|
.It Dv VM_PAGER_BAD
|
|
The page was entirely beyond the end of the backing file.
|
|
This condition should not be possible if the vnode's file system
|
|
is correctly implemented.
|
|
.It Dv VM_PAGER_ERROR
|
|
The page could not be written because of an error on the underlying storage
|
|
medium or protocol.
|
|
.It Dv VM_PAGER_FAIL
|
|
Treated identically to
|
|
.Dv VM_PAGER_ERROR .
|
|
.It Dv VM_PAGER_AGAIN
|
|
The page was not handled by this request.
|
|
.El
|
|
.Pp
|
|
The
|
|
.Fn VOP_GETPAGES
|
|
method is expected to release any pages in
|
|
.Fa ma
|
|
that it does not successfully handle, by calling
|
|
.Xr vm_page_free 9 .
|
|
When it succeeds,
|
|
.Fn VOP_GETPAGES
|
|
must set the valid bits appropriately.
|
|
.Fn VOP_GETPAGES
|
|
must keep
|
|
.Fa reqpage
|
|
busy.
|
|
It must unbusy all other successfully handled pages and put them
|
|
on appropriate page queue(s).
|
|
For example,
|
|
.Fn VOP_GETPAGES
|
|
may either activate a page (if its wanted bit is set)
|
|
or deactivate it (otherwise), and finally call
|
|
.Xr vm_page_xunbusy 9
|
|
to arouse any threads currently waiting for the page to be faulted in.
|
|
.Sh RETURN VALUES
|
|
If it successfully reads
|
|
.Fa ma[reqpage] ,
|
|
.Fn VOP_GETPAGES
|
|
returns
|
|
.Dv VM_PAGER_OK ;
|
|
otherwise,
|
|
.Dv VM_PAGER_ERROR .
|
|
By convention, the return value of
|
|
.Fn VOP_PUTPAGES
|
|
is
|
|
.Fa rtvals[0] .
|
|
.Sh SEE ALSO
|
|
.Xr vm_object_pip_wakeup 9 ,
|
|
.Xr vm_page_free 9 ,
|
|
.Xr vm_pagge_sunbusy 9 ,
|
|
.Xr vm_page_undirty 9 ,
|
|
.Xr vm_page_xunbusy 9 ,
|
|
.Xr vnode 9
|
|
.Sh AUTHORS
|
|
This manual page was written by
|
|
.An Doug Rabson
|
|
and then substantially rewritten by
|
|
.An Garrett Wollman .
|