64c6d1d2dc
- 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.
311 lines
8.4 KiB
C
311 lines
8.4 KiB
C
/* $NetBSD: kern_physio.c,v 1.50 2001/09/15 20:36:37 chs Exp $ */
|
|
|
|
/*-
|
|
* Copyright (c) 1994 Christopher G. Demetriou
|
|
* Copyright (c) 1982, 1986, 1990, 1993
|
|
* The Regents of the University of California. All rights reserved.
|
|
* (c) UNIX System Laboratories, Inc.
|
|
* All or some portions of this file are derived from material licensed
|
|
* to the University of California by American Telephone and Telegraph
|
|
* Co. or Unix System Laboratories, Inc. and are reproduced herein with
|
|
* the permission of UNIX System Laboratories, Inc.
|
|
*
|
|
* 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.
|
|
* 3. All advertising materials mentioning features or use of this software
|
|
* must display the following acknowledgement:
|
|
* This product includes software developed by the University of
|
|
* California, Berkeley and its contributors.
|
|
* 4. Neither the name of the University nor the names of its contributors
|
|
* may be used to endorse or promote products derived from this software
|
|
* without specific prior written permission.
|
|
*
|
|
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``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 REGENTS OR CONTRIBUTORS 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.
|
|
*
|
|
* @(#)kern_physio.c 8.1 (Berkeley) 6/10/93
|
|
*/
|
|
|
|
#include <sys/param.h>
|
|
#include <sys/systm.h>
|
|
#include <sys/buf.h>
|
|
#include <sys/malloc.h>
|
|
#include <sys/proc.h>
|
|
|
|
#include <uvm/uvm_extern.h>
|
|
|
|
/*
|
|
* The routines implemented in this file are described in:
|
|
* Leffler, et al.: The Design and Implementation of the 4.3BSD
|
|
* UNIX Operating System (Addison Welley, 1989)
|
|
* on pages 231-233.
|
|
*
|
|
* The routines "getphysbuf" and "putphysbuf" steal and return a swap
|
|
* buffer. Leffler, et al., says that swap buffers are used to do the
|
|
* I/O, so raw I/O requests don't have to be single-threaded.
|
|
*/
|
|
|
|
struct buf *getphysbuf __P((void));
|
|
void putphysbuf __P((struct buf *bp));
|
|
|
|
/*
|
|
* Do "physical I/O" on behalf of a user. "Physical I/O" is I/O directly
|
|
* from the raw device to user buffers, and bypasses the buffer cache.
|
|
*
|
|
* Comments in brackets are from Leffler, et al.'s pseudo-code implementation.
|
|
*/
|
|
int
|
|
physio(strategy, bp, dev, flags, minphys, uio)
|
|
void (*strategy) __P((struct buf *));
|
|
struct buf *bp;
|
|
dev_t dev;
|
|
int flags;
|
|
void (*minphys) __P((struct buf *));
|
|
struct uio *uio;
|
|
{
|
|
struct iovec *iovp;
|
|
struct proc *p = curproc;
|
|
int error, done, i, nobuf, s;
|
|
long todo;
|
|
|
|
error = 0;
|
|
flags &= B_READ | B_WRITE | B_ORDERED;
|
|
|
|
/* Make sure we have a buffer, creating one if necessary. */
|
|
if ((nobuf = (bp == NULL)) != 0) {
|
|
|
|
bp = getphysbuf();
|
|
/* bp was just malloc'd so can't already be busy */
|
|
bp->b_flags |= B_BUSY;
|
|
|
|
} else {
|
|
|
|
/* [raise the processor priority level to splbio;] */
|
|
s = splbio();
|
|
|
|
/* [while the buffer is marked busy] */
|
|
while (bp->b_flags & B_BUSY) {
|
|
/* [mark the buffer wanted] */
|
|
bp->b_flags |= B_WANTED;
|
|
/* [wait until the buffer is available] */
|
|
tsleep((caddr_t)bp, PRIBIO+1, "physbuf", 0);
|
|
}
|
|
|
|
/* Mark it busy, so nobody else will use it. */
|
|
bp->b_flags |= B_BUSY;
|
|
|
|
/* [lower the priority level] */
|
|
splx(s);
|
|
}
|
|
|
|
/* [set up the fixed part of the buffer for a transfer] */
|
|
bp->b_dev = dev;
|
|
bp->b_error = 0;
|
|
bp->b_proc = p;
|
|
LIST_INIT(&bp->b_dep);
|
|
|
|
/*
|
|
* [while there are data to transfer and no I/O error]
|
|
* Note that I/O errors are handled with a 'goto' at the bottom
|
|
* of the 'while' loop.
|
|
*/
|
|
for (i = 0; i < uio->uio_iovcnt; i++) {
|
|
iovp = &uio->uio_iov[i];
|
|
while (iovp->iov_len > 0) {
|
|
|
|
/*
|
|
* [mark the buffer busy for physical I/O]
|
|
* (i.e. set B_PHYS (because it's an I/O to user
|
|
* memory, and B_RAW, because B_RAW is to be
|
|
* "Set by physio for raw transfers.", in addition
|
|
* to the "busy" and read/write flag.)
|
|
*/
|
|
bp->b_flags = B_BUSY | B_PHYS | B_RAW | flags;
|
|
|
|
/* [set up the buffer for a maximum-sized transfer] */
|
|
bp->b_blkno = btodb(uio->uio_offset);
|
|
bp->b_bcount = iovp->iov_len;
|
|
bp->b_data = iovp->iov_base;
|
|
|
|
/*
|
|
* [call minphys to bound the transfer size]
|
|
* and remember the amount of data to transfer,
|
|
* for later comparison.
|
|
*/
|
|
(*minphys)(bp);
|
|
todo = bp->b_bcount;
|
|
#ifdef DIAGNOSTIC
|
|
if (todo <= 0)
|
|
panic("todo(%ld) <= 0; minphys broken", todo);
|
|
if (todo > MAXPHYS)
|
|
panic("todo(%ld) > MAXPHYS; minphys broken",
|
|
todo);
|
|
#endif
|
|
|
|
/*
|
|
* [lock the part of the user address space involved
|
|
* in the transfer]
|
|
* Beware vmapbuf(); it clobbers b_data and
|
|
* saves it in b_saveaddr. However, vunmapbuf()
|
|
* restores it.
|
|
*/
|
|
PHOLD(p);
|
|
error = uvm_vslock(p, bp->b_data, todo,
|
|
(flags & B_READ) ?
|
|
VM_PROT_READ | VM_PROT_WRITE :
|
|
VM_PROT_READ);
|
|
if (error) {
|
|
bp->b_flags |= B_ERROR;
|
|
bp->b_error = error;
|
|
goto after_vsunlock;
|
|
}
|
|
vmapbuf(bp, todo);
|
|
|
|
/* [call strategy to start the transfer] */
|
|
(*strategy)(bp);
|
|
|
|
/*
|
|
* Note that the raise/wait/lower/get error
|
|
* steps below would be done by biowait(), but
|
|
* we want to unlock the address space before
|
|
* we lower the priority.
|
|
*
|
|
* [raise the priority level to splbio]
|
|
*/
|
|
s = splbio();
|
|
|
|
/* [wait for the transfer to complete] */
|
|
while ((bp->b_flags & B_DONE) == 0)
|
|
tsleep((caddr_t) bp, PRIBIO + 1, "physio", 0);
|
|
|
|
/* Mark it busy again, so nobody else will use it. */
|
|
bp->b_flags |= B_BUSY;
|
|
|
|
/* [lower the priority level] */
|
|
splx(s);
|
|
|
|
/*
|
|
* [unlock the part of the address space previously
|
|
* locked]
|
|
*/
|
|
vunmapbuf(bp, todo);
|
|
uvm_vsunlock(p, bp->b_data, todo);
|
|
after_vsunlock:
|
|
PRELE(p);
|
|
|
|
/* remember error value (save a splbio/splx pair) */
|
|
if (bp->b_flags & B_ERROR)
|
|
error = (bp->b_error ? bp->b_error : EIO);
|
|
|
|
/*
|
|
* [deduct the transfer size from the total number
|
|
* of data to transfer]
|
|
*/
|
|
done = bp->b_bcount - bp->b_resid;
|
|
KASSERT(done >= 0);
|
|
KASSERT(done <= todo);
|
|
|
|
iovp->iov_len -= done;
|
|
iovp->iov_base = (caddr_t)iovp->iov_base + done;
|
|
uio->uio_offset += done;
|
|
uio->uio_resid -= done;
|
|
|
|
/*
|
|
* Now, check for an error.
|
|
* Also, handle weird end-of-disk semantics.
|
|
*/
|
|
if (error || done < todo)
|
|
goto done;
|
|
}
|
|
}
|
|
|
|
done:
|
|
/*
|
|
* [clean up the state of the buffer]
|
|
* Remember if somebody wants it, so we can wake them up below.
|
|
* Also, if we had to steal it, give it back.
|
|
*/
|
|
s = splbio();
|
|
bp->b_flags &= ~(B_BUSY | B_PHYS | B_RAW);
|
|
if (nobuf)
|
|
putphysbuf(bp);
|
|
else {
|
|
/*
|
|
* [if another process is waiting for the raw I/O buffer,
|
|
* wake up processes waiting to do physical I/O;
|
|
*/
|
|
if (bp->b_flags & B_WANTED) {
|
|
bp->b_flags &= ~B_WANTED;
|
|
wakeup(bp);
|
|
}
|
|
}
|
|
splx(s);
|
|
|
|
return (error);
|
|
}
|
|
|
|
/*
|
|
* allocate a buffer structure for use in physical I/O.
|
|
*/
|
|
struct buf *
|
|
getphysbuf()
|
|
{
|
|
struct buf *bp;
|
|
int s;
|
|
|
|
s = splbio();
|
|
bp = pool_get(&bufpool, PR_WAITOK);
|
|
splx(s);
|
|
memset(bp, 0, sizeof(*bp));
|
|
return(bp);
|
|
}
|
|
|
|
/*
|
|
* get rid of a swap buffer structure which has been used in physical I/O.
|
|
*/
|
|
void
|
|
putphysbuf(bp)
|
|
struct buf *bp;
|
|
{
|
|
int s;
|
|
|
|
if (__predict_false(bp->b_flags & B_WANTED))
|
|
panic("putphysbuf: private buf B_WANTED");
|
|
s = splbio();
|
|
pool_put(&bufpool, bp);
|
|
splx(s);
|
|
}
|
|
|
|
/*
|
|
* Leffler, et al., says on p. 231:
|
|
* "The minphys() routine is called by physio() to adjust the
|
|
* size of each I/O transfer before the latter is passed to
|
|
* the strategy routine..."
|
|
*
|
|
* so, just adjust the buffer's count accounting to MAXPHYS here,
|
|
* and return the new count;
|
|
*/
|
|
void
|
|
minphys(bp)
|
|
struct buf *bp;
|
|
{
|
|
|
|
if (bp->b_bcount > MAXPHYS)
|
|
bp->b_bcount = MAXPHYS;
|
|
}
|