b0cd20172d
o With new KPI consumers can request contiguous ranges of pages, and unlike before, all pages will be kept busied on return, like it was done before with the 'reqpage' only. Now the reqpage goes away. With new interface it is easier to implement code protected from race conditions. Such arrayed requests for now should be preceeded by a call to vm_pager_haspage() to make sure that request is possible. This could be improved later, making vm_pager_haspage() obsolete. Strenghtening the promises on the business of the array of pages allows us to remove such hacks as swp_pager_free_nrpage() and vm_pager_free_nonreq(). o New KPI accepts two integer pointers that may optionally point at values for read ahead and read behind, that a pager may do, if it can. These pages are completely owned by pager, and not controlled by the caller. This shifts the UFS-specific readahead logic from vm_fault.c, which should be file system agnostic, into vnode_pager.c. It also removes one VOP_BMAP() request per hard fault. Discussed with: kib, alc, jeff, scottl Sponsored by: Nginx, Inc. Sponsored by: Netflix
173 lines
5.3 KiB
Groff
173 lines
5.3 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 December 16, 2015
|
|
.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 *rbehind" "int *rahead"
|
|
.Ft int
|
|
.Fn VOP_PUTPAGES "struct vnode *vp" "vm_page_t *ma" "int count" "int sync" "int *rtvals"
|
|
.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 rbehind
|
|
.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 rbehind
|
|
Optional pointer to integer specifying number of pages to be read behind, if
|
|
possible.
|
|
If the filesystem supports that feature, number of actually read pages is
|
|
reported back, otherwise zero is returned.
|
|
.It Fa rahead
|
|
Optional pointer to integer specifying number of pages to be read ahead, if
|
|
possible.
|
|
If the filesystem supports that feature, number of actually read pages is
|
|
reported back, otherwise zero is returned.
|
|
.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_page_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 .
|