d/freebsd-skq
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
fcc7dccff6
freebsd-skq/sys/ufs/ffs/ffs_vnops.c

224 lines
6.2 KiB
C
Raw Normal View History

BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
/*
* Copyright (c) 1982, 1986, 1989, 1993
* The Regents of the University of California. All rights reserved.
*
* 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.
*
This is the kernel Lite/2 commit. There are some requisite userland changes, so don't expect to be able to run the kernel as-is (very well) without the appropriate Lite/2 userland changes. The system boots and can mount UFS filesystems. Untested: ext2fs, msdosfs, NFS Known problems: Incorrect Berkeley ID strings in some files. Mount_std mounts will not work until the getfsent library routine is changed. Reviewed by: various people Submitted by: Jeffery Hsu <hsu@freebsd.org>
1997-02-10 02:22:35 +00:00
* @(#)ffs_vnops.c 8.15 (Berkeley) 5/14/95
Support more intelligent sync operations for MNT_NOATIME. PR: kern/5577 Submitted by: Craig Leres <leres@ee.lbl.gov>
1998-02-01 01:59:12 +00:00
* $Id: ffs_vnops.c,v 1.38 1998/01/06 05:23:44 dyson Exp $
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
*/
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/resourcevar.h>
Cosmetics. make gcc less noisy. Still some way to go here.
1994-10-10 01:04:55 +00:00
#include <sys/signalvar.h>
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
#include <sys/kernel.h>
#include <sys/stat.h>
#include <sys/buf.h>
#include <sys/proc.h>
#include <sys/mount.h>
Renamed doclusterread/write to unique names (ffs_doclusterread/write), and staticize them. Move the #include of <sys/sysctl.h> to the top of the file. Pointed out by: Bruce Evans <bde@zeta.org.au>
1997-08-25 08:18:39 +00:00
#include <sys/sysctl.h>
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
#include <sys/vnode.h>
#include <vm/vm.h>
Untangled the vm.h include file spaghetti.
1995-12-07 12:48:31 +00:00
#include <vm/vm_prot.h>
Changes from John Dyson and myself: Fixed remaining known bugs in the buffer IO and VM system. vfs_bio.c: Fixed some race conditions and locking bugs. Improved performance by removing some (now) unnecessary code and fixing some broken logic. Fixed process accounting of # of FS outputs. Properly handle NFS interrupts (B_EINTR). (various) Replaced calls to clrbuf() with calls to an optimized routine called vfs_bio_clrbuf(). (various FS sync) Sync out modified vnode_pager backed pages. ffs_vnops.c: Do two passes: Sync out file data first, then indirect blocks. vm_fault.c: Fixed deadly embrace caused by acquiring locks in the wrong order. vnode_pager.c: Changed to use buffer I/O system for writing out modified pages. This should fix the problem with the modification date previous not getting updated. Also dramatically simplifies the code. Note that this is going to change in the future and be implemented via VOP_PUTPAGES(). vm_object.c: Fixed a pile of bugs related to cleaning (vnode) objects. The performance of vm_object_page_clean() is terrible when dealing with huge objects, but this will change when we implement a binary tree to keep the object pages sorted. vm_pageout.c: Fixed broken clustering of pageouts. Fixed race conditions and other lockup style bugs in the scanning of pages. Improved performance.
1995-04-09 06:03:56 +00:00
#include <vm/vm_page.h>
#include <vm/vm_object.h>
Untangled the vm.h include file spaghetti.
1995-12-07 12:48:31 +00:00
#include <vm/vm_extern.h>
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
#include <ufs/ufs/quota.h>
#include <ufs/ufs/inode.h>
This is the kernel Lite/2 commit. There are some requisite userland changes, so don't expect to be able to run the kernel as-is (very well) without the appropriate Lite/2 userland changes. The system boots and can mount UFS filesystems. Untested: ext2fs, msdosfs, NFS Known problems: Incorrect Berkeley ID strings in some files. Mount_std mounts will not work until the getfsent library routine is changed. Reviewed by: various people Submitted by: Jeffery Hsu <hsu@freebsd.org>
1997-02-10 02:22:35 +00:00
#include <ufs/ufs/ufsmount.h>
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
#include <ufs/ufs/ufs_extern.h>
#include <ufs/ffs/fs.h>
#include <ufs/ffs/ffs_extern.h>
Staticize.
1995-12-17 21:14:36 +00:00
static int ffs_fsync __P((struct vop_fsync_args *));
static int ffs_getpages __P((struct vop_getpages_args *));
static int ffs_read __P((struct vop_read_args *));
static int ffs_write __P((struct vop_write_args *));
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
/* Global vfs data structures for ufs. */
Introduced a type `vop_t' for vnode operation functions and used it 1138 times (:-() in casts and a few more times in declarations. This change is null for the i386. The type has to be `typedef int vop_t(void *)' and not `typedef int vop_t()' because `gcc -Wstrict-prototypes' warns about the latter. Since vnode op functions are called with args of different (struct pointer) types, neither of these function types is any use for type checking of the arg, so it would be preferable not to use the complete function type, especially since using the complete type requires adding 1138 casts to avoid compiler warnings and another 40+ casts to reverse the function pointer conversions before calling the functions.
1995-11-09 08:17:23 +00:00
vop_t **ffs_vnodeop_p;
Staticize.
1995-12-17 21:14:36 +00:00
static struct vnodeopv_entry_desc ffs_vnodeop_entries[] = {
vnops megacommit 1. Use the default function to access all the specfs operations. 2. Use the default function to access all the fifofs operations. 3. Use the default function to access all the ufs operations. 4. Fix VCALL usage in vfs_cache.c 5. Use VOCALL to access specfs functions in devfs_vnops.c 6. Staticize most of the spec and fifofs vnops functions. 7. Make UFS panic if it lacks bits of the underlying storage handling.
1997-10-15 13:24:07 +00:00
{ &vop_default_desc, (vop_t *) ufs_vnoperate },
Stylistic overhaul of vnops tables. 1. Remove comment stating the blatantly obvious. 2. Align in two columns. 3. Sort all but the default element alphabetically. 4. Remove XXX comments pointing out entries not needed.
1997-10-15 09:22:02 +00:00
{ &vop_fsync_desc, (vop_t *) ffs_fsync },
Hmm, realign the vnops into two columns.
1997-10-15 10:05:29 +00:00
{ &vop_getpages_desc, (vop_t *) ffs_getpages },
Stylistic overhaul of vnops tables. 1. Remove comment stating the blatantly obvious. 2. Align in two columns. 3. Sort all but the default element alphabetically. 4. Remove XXX comments pointing out entries not needed.
1997-10-15 09:22:02 +00:00
{ &vop_read_desc, (vop_t *) ffs_read },
{ &vop_reallocblks_desc, (vop_t *) ffs_reallocblks },
{ &vop_write_desc, (vop_t *) ffs_write },
Introduced a type `vop_t' for vnode operation functions and used it 1138 times (:-() in casts and a few more times in declarations. This change is null for the i386. The type has to be `typedef int vop_t(void *)' and not `typedef int vop_t()' because `gcc -Wstrict-prototypes' warns about the latter. Since vnode op functions are called with args of different (struct pointer) types, neither of these function types is any use for type checking of the arg, so it would be preferable not to use the complete function type, especially since using the complete type requires adding 1138 casts to avoid compiler warnings and another 40+ casts to reverse the function pointer conversions before calling the functions.
1995-11-09 08:17:23 +00:00
{ NULL, NULL }
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
};
Staticize.
1995-12-17 21:14:36 +00:00
static struct vnodeopv_desc ffs_vnodeop_opv_desc =
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
{ &ffs_vnodeop_p, ffs_vnodeop_entries };
Introduced a type `vop_t' for vnode operation functions and used it 1138 times (:-() in casts and a few more times in declarations. This change is null for the i386. The type has to be `typedef int vop_t(void *)' and not `typedef int vop_t()' because `gcc -Wstrict-prototypes' warns about the latter. Since vnode op functions are called with args of different (struct pointer) types, neither of these function types is any use for type checking of the arg, so it would be preferable not to use the complete function type, especially since using the complete type requires adding 1138 casts to avoid compiler warnings and another 40+ casts to reverse the function pointer conversions before calling the functions.
1995-11-09 08:17:23 +00:00
vop_t **ffs_specop_p;
Staticize.
1995-12-17 21:14:36 +00:00
static struct vnodeopv_entry_desc ffs_specop_entries[] = {
vnops megacommit 1. Use the default function to access all the specfs operations. 2. Use the default function to access all the fifofs operations. 3. Use the default function to access all the ufs operations. 4. Fix VCALL usage in vfs_cache.c 5. Use VOCALL to access specfs functions in devfs_vnops.c 6. Staticize most of the spec and fifofs vnops functions. 7. Make UFS panic if it lacks bits of the underlying storage handling.
1997-10-15 13:24:07 +00:00
{ &vop_default_desc, (vop_t *) ufs_vnoperatespec },
Stylistic overhaul of vnops tables. 1. Remove comment stating the blatantly obvious. 2. Align in two columns. 3. Sort all but the default element alphabetically. 4. Remove XXX comments pointing out entries not needed.
1997-10-15 09:22:02 +00:00
{ &vop_fsync_desc, (vop_t *) ffs_fsync },
Introduced a type `vop_t' for vnode operation functions and used it 1138 times (:-() in casts and a few more times in declarations. This change is null for the i386. The type has to be `typedef int vop_t(void *)' and not `typedef int vop_t()' because `gcc -Wstrict-prototypes' warns about the latter. Since vnode op functions are called with args of different (struct pointer) types, neither of these function types is any use for type checking of the arg, so it would be preferable not to use the complete function type, especially since using the complete type requires adding 1138 casts to avoid compiler warnings and another 40+ casts to reverse the function pointer conversions before calling the functions.
1995-11-09 08:17:23 +00:00
{ NULL, NULL }
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
};
Staticize.
1995-12-17 21:14:36 +00:00
static struct vnodeopv_desc ffs_specop_opv_desc =
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
{ &ffs_specop_p, ffs_specop_entries };
Introduced a type `vop_t' for vnode operation functions and used it 1138 times (:-() in casts and a few more times in declarations. This change is null for the i386. The type has to be `typedef int vop_t(void *)' and not `typedef int vop_t()' because `gcc -Wstrict-prototypes' warns about the latter. Since vnode op functions are called with args of different (struct pointer) types, neither of these function types is any use for type checking of the arg, so it would be preferable not to use the complete function type, especially since using the complete type requires adding 1138 casts to avoid compiler warnings and another 40+ casts to reverse the function pointer conversions before calling the functions.
1995-11-09 08:17:23 +00:00
vop_t **ffs_fifoop_p;
Staticize.
1995-12-17 21:14:36 +00:00
static struct vnodeopv_entry_desc ffs_fifoop_entries[] = {
vnops megacommit 1. Use the default function to access all the specfs operations. 2. Use the default function to access all the fifofs operations. 3. Use the default function to access all the ufs operations. 4. Fix VCALL usage in vfs_cache.c 5. Use VOCALL to access specfs functions in devfs_vnops.c 6. Staticize most of the spec and fifofs vnops functions. 7. Make UFS panic if it lacks bits of the underlying storage handling.
1997-10-15 13:24:07 +00:00
{ &vop_default_desc, (vop_t *) ufs_vnoperatefifo },
Stylistic overhaul of vnops tables. 1. Remove comment stating the blatantly obvious. 2. Align in two columns. 3. Sort all but the default element alphabetically. 4. Remove XXX comments pointing out entries not needed.
1997-10-15 09:22:02 +00:00
{ &vop_fsync_desc, (vop_t *) ffs_fsync },
Introduced a type `vop_t' for vnode operation functions and used it 1138 times (:-() in casts and a few more times in declarations. This change is null for the i386. The type has to be `typedef int vop_t(void *)' and not `typedef int vop_t()' because `gcc -Wstrict-prototypes' warns about the latter. Since vnode op functions are called with args of different (struct pointer) types, neither of these function types is any use for type checking of the arg, so it would be preferable not to use the complete function type, especially since using the complete type requires adding 1138 casts to avoid compiler warnings and another 40+ casts to reverse the function pointer conversions before calling the functions.
1995-11-09 08:17:23 +00:00
{ NULL, NULL }
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
};
Staticize.
1995-12-17 21:14:36 +00:00
static struct vnodeopv_desc ffs_fifoop_opv_desc =
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
{ &ffs_fifoop_p, ffs_fifoop_entries };
Implemented loadable VFS modules, and made most existing filesystems loadable. (NFS is a notable exception.)
1994-09-21 03:47:43 +00:00
VNODEOP_SET(ffs_vnodeop_opv_desc);
VNODEOP_SET(ffs_specop_opv_desc);
VNODEOP_SET(ffs_fifoop_opv_desc);
Fixed connection of vfs.ffs node to the sysctl tree.
1997-03-04 18:35:15 +00:00
SYSCTL_NODE(_vfs, MOUNT_UFS, ffs, CTLFLAG_RW, 0, "FFS filesystem");
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
#include <ufs/ufs/ufs_readwrite.c>
Support more intelligent sync operations for MNT_NOATIME. PR: kern/5577 Submitted by: Craig Leres <leres@ee.lbl.gov>
1998-02-01 01:59:12 +00:00
int ffs_log_sync = 0;
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
/*
* Synch an open file.
*/
/* ARGSUSED */
Staticize.
1995-12-17 21:14:36 +00:00
static int
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
ffs_fsync(ap)
struct vop_fsync_args /* {
struct vnode *a_vp;
struct ucred *a_cred;
int a_waitfor;
struct proc *a_p;
} */ *ap;
{
register struct vnode *vp = ap->a_vp;
register struct buf *bp;
struct timeval tv;
struct buf *nbp;
Changes from John Dyson and myself: Fixed remaining known bugs in the buffer IO and VM system. vfs_bio.c: Fixed some race conditions and locking bugs. Improved performance by removing some (now) unnecessary code and fixing some broken logic. Fixed process accounting of # of FS outputs. Properly handle NFS interrupts (B_EINTR). (various) Replaced calls to clrbuf() with calls to an optimized routine called vfs_bio_clrbuf(). (various FS sync) Sync out modified vnode_pager backed pages. ffs_vnops.c: Do two passes: Sync out file data first, then indirect blocks. vm_fault.c: Fixed deadly embrace caused by acquiring locks in the wrong order. vnode_pager.c: Changed to use buffer I/O system for writing out modified pages. This should fix the problem with the modification date previous not getting updated. Also dramatically simplifies the code. Note that this is going to change in the future and be implemented via VOP_PUTPAGES(). vm_object.c: Fixed a pile of bugs related to cleaning (vnode) objects. The performance of vm_object_page_clean() is terrible when dealing with huge objects, but this will change when we implement a binary tree to keep the object pages sorted. vm_pageout.c: Fixed broken clustering of pageouts. Fixed race conditions and other lockup style bugs in the scanning of pages. Improved performance.
1995-04-09 06:03:56 +00:00
int pass;
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
int s;
Make our v_usecount vnode reference count work identically to the original BSD code. The association between the vnode and the vm_object no longer includes reference counts. The major difference is that vm_object's are no longer freed gratuitiously from the vnode, and so once an object is created for the vnode, it will last as long as the vnode does. When a vnode object reference count is incremented, then the underlying vnode reference count is incremented also. The two "objects" are now more intimately related, and so the interactions are now much less complex. When vnodes are now normally placed onto the free queue with an object still attached. The rundown of the object happens at vnode rundown time, and happens with exactly the same filesystem semantics of the original VFS code. There is absolutely no need for vnode_pager_uncache and other travesties like that anymore. A side-effect of these changes is that SMP locking should be much simpler, the I/O copyin/copyout optimizations work, NFS should be more ponderable, and further work on layered filesystems should be less frustrating, because of the totally coherent management of the vnode objects and vnodes. Please be careful with your system while running this code, but I would greatly appreciate feedback as soon a reasonably possible.
1998-01-06 05:26:17 +00:00
daddr_t lbn;
if (vp->v_type == VBLK) {
lbn = INT_MAX;
} else {
struct inode *ip;
ip = VTOI(vp);
lbn = lblkno(ip->i_fs, (ip->i_size + ip->i_fs->fs_bsize - 1));
}
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
Support more intelligent sync operations for MNT_NOATIME. PR: kern/5577 Submitted by: Craig Leres <leres@ee.lbl.gov>
1998-02-01 01:59:12 +00:00
if (ffs_log_sync) {
struct inode* ip = VTOI(vp);
printf("fsync i %u iflags 0x%x vn 0x%x vtype %d tag %d vflags 0x%x\n",
ip->i_number, ip->i_flag, vp, vp->v_type, vp->v_tag, vp->v_flag);
}
Changes from John Dyson and myself: Fixed remaining known bugs in the buffer IO and VM system. vfs_bio.c: Fixed some race conditions and locking bugs. Improved performance by removing some (now) unnecessary code and fixing some broken logic. Fixed process accounting of # of FS outputs. Properly handle NFS interrupts (B_EINTR). (various) Replaced calls to clrbuf() with calls to an optimized routine called vfs_bio_clrbuf(). (various FS sync) Sync out modified vnode_pager backed pages. ffs_vnops.c: Do two passes: Sync out file data first, then indirect blocks. vm_fault.c: Fixed deadly embrace caused by acquiring locks in the wrong order. vnode_pager.c: Changed to use buffer I/O system for writing out modified pages. This should fix the problem with the modification date previous not getting updated. Also dramatically simplifies the code. Note that this is going to change in the future and be implemented via VOP_PUTPAGES(). vm_object.c: Fixed a pile of bugs related to cleaning (vnode) objects. The performance of vm_object_page_clean() is terrible when dealing with huge objects, but this will change when we implement a binary tree to keep the object pages sorted. vm_pageout.c: Fixed broken clustering of pageouts. Fixed race conditions and other lockup style bugs in the scanning of pages. Improved performance.
1995-04-09 06:03:56 +00:00
pass = 0;
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
/*
* Flush all dirty buffers associated with a vnode.
*/
loop:
s = splbio();
for (bp = vp->v_dirtyblkhd.lh_first; bp; bp = nbp) {
nbp = bp->b_vnbufs.le_next;
Make our v_usecount vnode reference count work identically to the original BSD code. The association between the vnode and the vm_object no longer includes reference counts. The major difference is that vm_object's are no longer freed gratuitiously from the vnode, and so once an object is created for the vnode, it will last as long as the vnode does. When a vnode object reference count is incremented, then the underlying vnode reference count is incremented also. The two "objects" are now more intimately related, and so the interactions are now much less complex. When vnodes are now normally placed onto the free queue with an object still attached. The rundown of the object happens at vnode rundown time, and happens with exactly the same filesystem semantics of the original VFS code. There is absolutely no need for vnode_pager_uncache and other travesties like that anymore. A side-effect of these changes is that SMP locking should be much simpler, the I/O copyin/copyout optimizations work, NFS should be more ponderable, and further work on layered filesystems should be less frustrating, because of the totally coherent management of the vnode objects and vnodes. Please be careful with your system while running this code, but I would greatly appreciate feedback as soon a reasonably possible.
1998-01-06 05:26:17 +00:00
if ((bp->b_flags & B_BUSY) || (pass == 0 && (bp->b_lblkno < 0)))
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
continue;
if ((bp->b_flags & B_DELWRI) == 0)
panic("ffs_fsync: not dirty");
These changes embody the support of the fully coherent merged VM buffer cache, much higher filesystem I/O performance, and much better paging performance. It represents the culmination of over 6 months of R&D. The majority of the merged VM/cache work is by John Dyson. The following highlights the most significant changes. Additionally, there are (mostly minor) changes to the various filesystem modules (nfs, msdosfs, etc) to support the new VM/buffer scheme. vfs_bio.c: Significant rewrite of most of vfs_bio to support the merged VM buffer cache scheme. The scheme is almost fully compatible with the old filesystem interface. Significant improvement in the number of opportunities for write clustering. vfs_cluster.c, vfs_subr.c Upgrade and performance enhancements in vfs layer code to support merged VM/buffer cache. Fixup of vfs_cluster to eliminate the bogus pagemove stuff. vm_object.c: Yet more improvements in the collapse code. Elimination of some windows that can cause list corruption. vm_pageout.c: Fixed it, it really works better now. Somehow in 2.0, some "enhancements" broke the code. This code has been reworked from the ground-up. vm_fault.c, vm_page.c, pmap.c, vm_object.c Support for small-block filesystems with merged VM/buffer cache scheme. pmap.c vm_map.c Dynamic kernel VM size, now we dont have to pre-allocate excessive numbers of kernel PTs. vm_glue.c Much simpler and more effective swapping code. No more gratuitous swapping. proc.h Fixed the problem that the p_lock flag was not being cleared on a fork. swap_pager.c, vnode_pager.c Removal of old vfs_bio cruft to support the past pseudo-coherency. Now the code doesn't need it anymore. machdep.c Changes to better support the parameter values for the merged VM/buffer cache scheme. machdep.c, kern_exec.c, vm_glue.c Implemented a seperate submap for temporary exec string space and another one to contain process upages. This eliminates all map fragmentation problems that previously existed. ffs_inode.c, ufs_inode.c, ufs_readwrite.c Changes for merged VM/buffer cache. Add "bypass" support for sneaking in on busy buffers. Submitted by: John Dyson and David Greenman
1995-01-09 16:06:02 +00:00
Support more intelligent sync operations for MNT_NOATIME. PR: kern/5577 Submitted by: Craig Leres <leres@ee.lbl.gov>
1998-02-01 01:59:12 +00:00
if (ffs_log_sync) {
printf(" blk %u (%u) flags 0x%x vn 0x%x\n", bp->b_lblkno, bp->b_blkno,
bp->b_flags, bp->b_vp);
}
Make our v_usecount vnode reference count work identically to the original BSD code. The association between the vnode and the vm_object no longer includes reference counts. The major difference is that vm_object's are no longer freed gratuitiously from the vnode, and so once an object is created for the vnode, it will last as long as the vnode does. When a vnode object reference count is incremented, then the underlying vnode reference count is incremented also. The two "objects" are now more intimately related, and so the interactions are now much less complex. When vnodes are now normally placed onto the free queue with an object still attached. The rundown of the object happens at vnode rundown time, and happens with exactly the same filesystem semantics of the original VFS code. There is absolutely no need for vnode_pager_uncache and other travesties like that anymore. A side-effect of these changes is that SMP locking should be much simpler, the I/O copyin/copyout optimizations work, NFS should be more ponderable, and further work on layered filesystems should be less frustrating, because of the totally coherent management of the vnode objects and vnodes. Please be careful with your system while running this code, but I would greatly appreciate feedback as soon a reasonably possible.
1998-01-06 05:26:17 +00:00
if (((bp->b_vp != vp) || (ap->a_waitfor != MNT_NOWAIT)) ||
((vp->v_type != VREG) && (vp->v_type != VBLK))) {
Remove trailing whitespace.
1995-05-30 08:16:23 +00:00
These changes embody the support of the fully coherent merged VM buffer cache, much higher filesystem I/O performance, and much better paging performance. It represents the culmination of over 6 months of R&D. The majority of the merged VM/cache work is by John Dyson. The following highlights the most significant changes. Additionally, there are (mostly minor) changes to the various filesystem modules (nfs, msdosfs, etc) to support the new VM/buffer scheme. vfs_bio.c: Significant rewrite of most of vfs_bio to support the merged VM buffer cache scheme. The scheme is almost fully compatible with the old filesystem interface. Significant improvement in the number of opportunities for write clustering. vfs_cluster.c, vfs_subr.c Upgrade and performance enhancements in vfs layer code to support merged VM/buffer cache. Fixup of vfs_cluster to eliminate the bogus pagemove stuff. vm_object.c: Yet more improvements in the collapse code. Elimination of some windows that can cause list corruption. vm_pageout.c: Fixed it, it really works better now. Somehow in 2.0, some "enhancements" broke the code. This code has been reworked from the ground-up. vm_fault.c, vm_page.c, pmap.c, vm_object.c Support for small-block filesystems with merged VM/buffer cache scheme. pmap.c vm_map.c Dynamic kernel VM size, now we dont have to pre-allocate excessive numbers of kernel PTs. vm_glue.c Much simpler and more effective swapping code. No more gratuitous swapping. proc.h Fixed the problem that the p_lock flag was not being cleared on a fork. swap_pager.c, vnode_pager.c Removal of old vfs_bio cruft to support the past pseudo-coherency. Now the code doesn't need it anymore. machdep.c Changes to better support the parameter values for the merged VM/buffer cache scheme. machdep.c, kern_exec.c, vm_glue.c Implemented a seperate submap for temporary exec string space and another one to contain process upages. This eliminates all map fragmentation problems that previously existed. ffs_inode.c, ufs_inode.c, ufs_readwrite.c Changes for merged VM/buffer cache. Add "bypass" support for sneaking in on busy buffers. Submitted by: John Dyson and David Greenman
1995-01-09 16:06:02 +00:00
bremfree(bp);
bp->b_flags |= B_BUSY;
splx(s);
Make our v_usecount vnode reference count work identically to the original BSD code. The association between the vnode and the vm_object no longer includes reference counts. The major difference is that vm_object's are no longer freed gratuitiously from the vnode, and so once an object is created for the vnode, it will last as long as the vnode does. When a vnode object reference count is incremented, then the underlying vnode reference count is incremented also. The two "objects" are now more intimately related, and so the interactions are now much less complex. When vnodes are now normally placed onto the free queue with an object still attached. The rundown of the object happens at vnode rundown time, and happens with exactly the same filesystem semantics of the original VFS code. There is absolutely no need for vnode_pager_uncache and other travesties like that anymore. A side-effect of these changes is that SMP locking should be much simpler, the I/O copyin/copyout optimizations work, NFS should be more ponderable, and further work on layered filesystems should be less frustrating, because of the totally coherent management of the vnode objects and vnodes. Please be careful with your system while running this code, but I would greatly appreciate feedback as soon a reasonably possible.
1998-01-06 05:26:17 +00:00
Changes from John Dyson and myself: Fixed remaining known bugs in the buffer IO and VM system. vfs_bio.c: Fixed some race conditions and locking bugs. Improved performance by removing some (now) unnecessary code and fixing some broken logic. Fixed process accounting of # of FS outputs. Properly handle NFS interrupts (B_EINTR). (various) Replaced calls to clrbuf() with calls to an optimized routine called vfs_bio_clrbuf(). (various FS sync) Sync out modified vnode_pager backed pages. ffs_vnops.c: Do two passes: Sync out file data first, then indirect blocks. vm_fault.c: Fixed deadly embrace caused by acquiring locks in the wrong order. vnode_pager.c: Changed to use buffer I/O system for writing out modified pages. This should fix the problem with the modification date previous not getting updated. Also dramatically simplifies the code. Note that this is going to change in the future and be implemented via VOP_PUTPAGES(). vm_object.c: Fixed a pile of bugs related to cleaning (vnode) objects. The performance of vm_object_page_clean() is terrible when dealing with huge objects, but this will change when we implement a binary tree to keep the object pages sorted. vm_pageout.c: Fixed broken clustering of pageouts. Fixed race conditions and other lockup style bugs in the scanning of pages. Improved performance.
1995-04-09 06:03:56 +00:00
/*
* Wait for I/O associated with indirect blocks to complete,
* since there is no way to quickly wait for them below.
*/
Make our v_usecount vnode reference count work identically to the original BSD code. The association between the vnode and the vm_object no longer includes reference counts. The major difference is that vm_object's are no longer freed gratuitiously from the vnode, and so once an object is created for the vnode, it will last as long as the vnode does. When a vnode object reference count is incremented, then the underlying vnode reference count is incremented also. The two "objects" are now more intimately related, and so the interactions are now much less complex. When vnodes are now normally placed onto the free queue with an object still attached. The rundown of the object happens at vnode rundown time, and happens with exactly the same filesystem semantics of the original VFS code. There is absolutely no need for vnode_pager_uncache and other travesties like that anymore. A side-effect of these changes is that SMP locking should be much simpler, the I/O copyin/copyout optimizations work, NFS should be more ponderable, and further work on layered filesystems should be less frustrating, because of the totally coherent management of the vnode objects and vnodes. Please be careful with your system while running this code, but I would greatly appreciate feedback as soon a reasonably possible.
1998-01-06 05:26:17 +00:00
if ((bp->b_vp == vp) && (ap->a_waitfor == MNT_NOWAIT)) {
if (bp->b_flags & B_CLUSTEROK) {
bdwrite(bp);
(void) vfs_bio_awrite(bp);
} else {
(void) bawrite(bp);
}
} else {
These changes embody the support of the fully coherent merged VM buffer cache, much higher filesystem I/O performance, and much better paging performance. It represents the culmination of over 6 months of R&D. The majority of the merged VM/cache work is by John Dyson. The following highlights the most significant changes. Additionally, there are (mostly minor) changes to the various filesystem modules (nfs, msdosfs, etc) to support the new VM/buffer scheme. vfs_bio.c: Significant rewrite of most of vfs_bio to support the merged VM buffer cache scheme. The scheme is almost fully compatible with the old filesystem interface. Significant improvement in the number of opportunities for write clustering. vfs_cluster.c, vfs_subr.c Upgrade and performance enhancements in vfs layer code to support merged VM/buffer cache. Fixup of vfs_cluster to eliminate the bogus pagemove stuff. vm_object.c: Yet more improvements in the collapse code. Elimination of some windows that can cause list corruption. vm_pageout.c: Fixed it, it really works better now. Somehow in 2.0, some "enhancements" broke the code. This code has been reworked from the ground-up. vm_fault.c, vm_page.c, pmap.c, vm_object.c Support for small-block filesystems with merged VM/buffer cache scheme. pmap.c vm_map.c Dynamic kernel VM size, now we dont have to pre-allocate excessive numbers of kernel PTs. vm_glue.c Much simpler and more effective swapping code. No more gratuitous swapping. proc.h Fixed the problem that the p_lock flag was not being cleared on a fork. swap_pager.c, vnode_pager.c Removal of old vfs_bio cruft to support the past pseudo-coherency. Now the code doesn't need it anymore. machdep.c Changes to better support the parameter values for the merged VM/buffer cache scheme. machdep.c, kern_exec.c, vm_glue.c Implemented a seperate submap for temporary exec string space and another one to contain process upages. This eliminates all map fragmentation problems that previously existed. ffs_inode.c, ufs_inode.c, ufs_readwrite.c Changes for merged VM/buffer cache. Add "bypass" support for sneaking in on busy buffers. Submitted by: John Dyson and David Greenman
1995-01-09 16:06:02 +00:00
(void) bwrite(bp);
Make our v_usecount vnode reference count work identically to the original BSD code. The association between the vnode and the vm_object no longer includes reference counts. The major difference is that vm_object's are no longer freed gratuitiously from the vnode, and so once an object is created for the vnode, it will last as long as the vnode does. When a vnode object reference count is incremented, then the underlying vnode reference count is incremented also. The two "objects" are now more intimately related, and so the interactions are now much less complex. When vnodes are now normally placed onto the free queue with an object still attached. The rundown of the object happens at vnode rundown time, and happens with exactly the same filesystem semantics of the original VFS code. There is absolutely no need for vnode_pager_uncache and other travesties like that anymore. A side-effect of these changes is that SMP locking should be much simpler, the I/O copyin/copyout optimizations work, NFS should be more ponderable, and further work on layered filesystems should be less frustrating, because of the totally coherent management of the vnode objects and vnodes. Please be careful with your system while running this code, but I would greatly appreciate feedback as soon a reasonably possible.
1998-01-06 05:26:17 +00:00
}
} else if ((vp->v_type == VREG) && (bp->b_lblkno >= lbn)) {
bremfree(bp);
bp->b_flags |= B_BUSY | B_INVAL | B_NOCACHE;
brelse(bp);
splx(s);
These changes embody the support of the fully coherent merged VM buffer cache, much higher filesystem I/O performance, and much better paging performance. It represents the culmination of over 6 months of R&D. The majority of the merged VM/cache work is by John Dyson. The following highlights the most significant changes. Additionally, there are (mostly minor) changes to the various filesystem modules (nfs, msdosfs, etc) to support the new VM/buffer scheme. vfs_bio.c: Significant rewrite of most of vfs_bio to support the merged VM buffer cache scheme. The scheme is almost fully compatible with the old filesystem interface. Significant improvement in the number of opportunities for write clustering. vfs_cluster.c, vfs_subr.c Upgrade and performance enhancements in vfs layer code to support merged VM/buffer cache. Fixup of vfs_cluster to eliminate the bogus pagemove stuff. vm_object.c: Yet more improvements in the collapse code. Elimination of some windows that can cause list corruption. vm_pageout.c: Fixed it, it really works better now. Somehow in 2.0, some "enhancements" broke the code. This code has been reworked from the ground-up. vm_fault.c, vm_page.c, pmap.c, vm_object.c Support for small-block filesystems with merged VM/buffer cache scheme. pmap.c vm_map.c Dynamic kernel VM size, now we dont have to pre-allocate excessive numbers of kernel PTs. vm_glue.c Much simpler and more effective swapping code. No more gratuitous swapping. proc.h Fixed the problem that the p_lock flag was not being cleared on a fork. swap_pager.c, vnode_pager.c Removal of old vfs_bio cruft to support the past pseudo-coherency. Now the code doesn't need it anymore. machdep.c Changes to better support the parameter values for the merged VM/buffer cache scheme. machdep.c, kern_exec.c, vm_glue.c Implemented a seperate submap for temporary exec string space and another one to contain process upages. This eliminates all map fragmentation problems that previously existed. ffs_inode.c, ufs_inode.c, ufs_readwrite.c Changes for merged VM/buffer cache. Add "bypass" support for sneaking in on busy buffers. Submitted by: John Dyson and David Greenman
1995-01-09 16:06:02 +00:00
} else {
vfs_bio_awrite(bp);
splx(s);
}
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
goto loop;
}
Changes from John Dyson and myself: Fixed remaining known bugs in the buffer IO and VM system. vfs_bio.c: Fixed some race conditions and locking bugs. Improved performance by removing some (now) unnecessary code and fixing some broken logic. Fixed process accounting of # of FS outputs. Properly handle NFS interrupts (B_EINTR). (various) Replaced calls to clrbuf() with calls to an optimized routine called vfs_bio_clrbuf(). (various FS sync) Sync out modified vnode_pager backed pages. ffs_vnops.c: Do two passes: Sync out file data first, then indirect blocks. vm_fault.c: Fixed deadly embrace caused by acquiring locks in the wrong order. vnode_pager.c: Changed to use buffer I/O system for writing out modified pages. This should fix the problem with the modification date previous not getting updated. Also dramatically simplifies the code. Note that this is going to change in the future and be implemented via VOP_PUTPAGES(). vm_object.c: Fixed a pile of bugs related to cleaning (vnode) objects. The performance of vm_object_page_clean() is terrible when dealing with huge objects, but this will change when we implement a binary tree to keep the object pages sorted. vm_pageout.c: Fixed broken clustering of pageouts. Fixed race conditions and other lockup style bugs in the scanning of pages. Improved performance.
1995-04-09 06:03:56 +00:00
splx(s);
if (pass == 0) {
pass = 1;
goto loop;
}
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
if (ap->a_waitfor == MNT_WAIT) {
Changes from John Dyson and myself: Fixed remaining known bugs in the buffer IO and VM system. vfs_bio.c: Fixed some race conditions and locking bugs. Improved performance by removing some (now) unnecessary code and fixing some broken logic. Fixed process accounting of # of FS outputs. Properly handle NFS interrupts (B_EINTR). (various) Replaced calls to clrbuf() with calls to an optimized routine called vfs_bio_clrbuf(). (various FS sync) Sync out modified vnode_pager backed pages. ffs_vnops.c: Do two passes: Sync out file data first, then indirect blocks. vm_fault.c: Fixed deadly embrace caused by acquiring locks in the wrong order. vnode_pager.c: Changed to use buffer I/O system for writing out modified pages. This should fix the problem with the modification date previous not getting updated. Also dramatically simplifies the code. Note that this is going to change in the future and be implemented via VOP_PUTPAGES(). vm_object.c: Fixed a pile of bugs related to cleaning (vnode) objects. The performance of vm_object_page_clean() is terrible when dealing with huge objects, but this will change when we implement a binary tree to keep the object pages sorted. vm_pageout.c: Fixed broken clustering of pageouts. Fixed race conditions and other lockup style bugs in the scanning of pages. Improved performance.
1995-04-09 06:03:56 +00:00
s = splbio();
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
while (vp->v_numoutput) {
vp->v_flag |= VBWAIT;
Use tsleep() rather than sleep so that 'ps' is more informative about the wait.
1994-10-06 21:07:04 +00:00
(void) tsleep((caddr_t)&vp->v_numoutput, PRIBIO + 1, "ffsfsn", 0);
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
}
Changes from John Dyson and myself: Fixed remaining known bugs in the buffer IO and VM system. vfs_bio.c: Fixed some race conditions and locking bugs. Improved performance by removing some (now) unnecessary code and fixing some broken logic. Fixed process accounting of # of FS outputs. Properly handle NFS interrupts (B_EINTR). (various) Replaced calls to clrbuf() with calls to an optimized routine called vfs_bio_clrbuf(). (various FS sync) Sync out modified vnode_pager backed pages. ffs_vnops.c: Do two passes: Sync out file data first, then indirect blocks. vm_fault.c: Fixed deadly embrace caused by acquiring locks in the wrong order. vnode_pager.c: Changed to use buffer I/O system for writing out modified pages. This should fix the problem with the modification date previous not getting updated. Also dramatically simplifies the code. Note that this is going to change in the future and be implemented via VOP_PUTPAGES(). vm_object.c: Fixed a pile of bugs related to cleaning (vnode) objects. The performance of vm_object_page_clean() is terrible when dealing with huge objects, but this will change when we implement a binary tree to keep the object pages sorted. vm_pageout.c: Fixed broken clustering of pageouts. Fixed race conditions and other lockup style bugs in the scanning of pages. Improved performance.
1995-04-09 06:03:56 +00:00
splx(s);
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
#ifdef DIAGNOSTIC
if (vp->v_dirtyblkhd.lh_first) {
vprint("ffs_fsync: dirty", vp);
goto loop;
}
#endif
}
These changes embody the support of the fully coherent merged VM buffer cache, much higher filesystem I/O performance, and much better paging performance. It represents the culmination of over 6 months of R&D. The majority of the merged VM/cache work is by John Dyson. The following highlights the most significant changes. Additionally, there are (mostly minor) changes to the various filesystem modules (nfs, msdosfs, etc) to support the new VM/buffer scheme. vfs_bio.c: Significant rewrite of most of vfs_bio to support the merged VM buffer cache scheme. The scheme is almost fully compatible with the old filesystem interface. Significant improvement in the number of opportunities for write clustering. vfs_cluster.c, vfs_subr.c Upgrade and performance enhancements in vfs layer code to support merged VM/buffer cache. Fixup of vfs_cluster to eliminate the bogus pagemove stuff. vm_object.c: Yet more improvements in the collapse code. Elimination of some windows that can cause list corruption. vm_pageout.c: Fixed it, it really works better now. Somehow in 2.0, some "enhancements" broke the code. This code has been reworked from the ground-up. vm_fault.c, vm_page.c, pmap.c, vm_object.c Support for small-block filesystems with merged VM/buffer cache scheme. pmap.c vm_map.c Dynamic kernel VM size, now we dont have to pre-allocate excessive numbers of kernel PTs. vm_glue.c Much simpler and more effective swapping code. No more gratuitous swapping. proc.h Fixed the problem that the p_lock flag was not being cleared on a fork. swap_pager.c, vnode_pager.c Removal of old vfs_bio cruft to support the past pseudo-coherency. Now the code doesn't need it anymore. machdep.c Changes to better support the parameter values for the merged VM/buffer cache scheme. machdep.c, kern_exec.c, vm_glue.c Implemented a seperate submap for temporary exec string space and another one to contain process upages. This eliminates all map fragmentation problems that previously existed. ffs_inode.c, ufs_inode.c, ufs_readwrite.c Changes for merged VM/buffer cache. Add "bypass" support for sneaking in on busy buffers. Submitted by: John Dyson and David Greenman
1995-01-09 16:06:02 +00:00
Fixed some invalid (non-atomic) accesses to `time', mostly ones of the form `tv = time'. Use a new function gettime(). The current version just forces atomicicity without fixing precision or efficiency bugs. Simplified some related valid accesses by using the central function.
1997-03-22 06:53:45 +00:00
gettime(&tv);
Support more intelligent sync operations for MNT_NOATIME. PR: kern/5577 Submitted by: Craig Leres <leres@ee.lbl.gov>
1998-02-01 01:59:12 +00:00
return (UFS_UPDATE(ap->a_vp, &tv, &tv, 1));
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
}
Reference in New Issue Copy Permalink