d/freebsd-dev
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
ee35caa0b3
freebsd-dev/sys/kern/vfs_vnops.c

1255 lines
28 KiB
C
Raw Normal View History

/* -> /*- for copyright notices, minor format tweaks as necessary
2005-01-06 23:35:40 +00:00
/*-
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.
* (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.
* 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.
*
* @(#)vfs_vnops.c 8.2 (Berkeley) 1/21/94
*/
Use __FBSDID().
2003-06-11 00:56:59 +00:00
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
Introduce support for Mandatory Access Control and extensible kernel access control. Restructure the vn_open_cred() access control checks to invoke the MAC entry point for open authorization. Note that MAC can reject open requests where existing DAC code skips the open authorization check due to O_CREAT. However, the failure mode here is the same as other failure modes following creation, wherein an empty file may be left behind. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-01 17:14:28 +00:00
#include "opt_mac.h"
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
#include <sys/param.h>
#include <sys/systm.h>
Don't #include <sys/fcntl.h> in <sys/file.h> if KERNEL is defined. Fixed everything that depended on getting fcntl.h stuff from the wrong place. Most things don't depend on file.h stuff at all.
1997-03-23 03:37:54 +00:00
#include <sys/fcntl.h>
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
#include <sys/file.h>
Update for the KDB framework: o Call kdb_enter() instead of Debugger().
2004-07-10 21:47:53 +00:00
#include <sys/kdb.h>
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
#include <sys/stat.h>
#include <sys/proc.h>
Deprecate machine/limits.h in favor of new sys/limits.h. Change all in-tree consumers to include <sys/limits.h> Discussed on: standards@ Partially submitted by: Craig Rodrigues <rodrigc@attbi.com>
2003-04-29 13:36:06 +00:00
#include <sys/limits.h>
Add missing includes of sys/lock.h.
2001-10-11 17:52:20 +00:00
#include <sys/lock.h>
Introduce support for Mandatory Access Control and extensible kernel access control. Restructure the vn_open_cred() access control checks to invoke the MAC entry point for open authorization. Note that MAC can reject open requests where existing DAC code skips the open authorization check due to O_CREAT. However, the failure mode here is the same as other failure modes following creation, wherein an empty file may be left behind. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-01 17:14:28 +00:00
#include <sys/mac.h>
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
#include <sys/mount.h>
Catch up to moving headers: - machine/ipl.h -> sys/ipl.h - machine/mutex.h -> sys/mutex.h
2000-10-20 07:58:15 +00:00
#include <sys/mutex.h>
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
#include <sys/namei.h>
#include <sys/vnode.h>
Separate the struct bio related stuff out of <sys/buf.h> into <sys/bio.h>. <sys/bio.h> is now a prerequisite for <sys/buf.h> but it shall not be made a nested include according to bdes teachings on the subject of nested includes. Diskdrivers and similar stuff below specfs::strategy() should no longer need to include <sys/buf.> unless they need caching of data. Still a few bogus uses of struct buf to track down. Repocopy by: peter
2000-05-05 09:59:14 +00:00
#include <sys/bio.h>
Fix fd race conditions (during shared fd table usage.) Badfileops is now used in f_ops in place of NULL, and modifications to the files are more carefully ordered. f_ops should also be set to &badfileops upon "close" of a file. This does not fix other problems mentioned in this PR than the first one. PR: 11629 Reviewed by: peter
1999-08-04 18:53:50 +00:00
#include <sys/buf.h>
Don't include <sys/ioctl.h> in the kernel. Stage 4: include <sys/ttycom.h> and sometimes <sys/filio.h> instead of <sys/ioctl.h> in miscellaneous files. Most of these files have nothing to do with ttys but need to include <sys/ttycom.h> to get the definitions of TIOC[SG]PGRP which are (ab)used to convert F[SG]ETOWN fcntls into ioctls.
1997-03-24 11:52:29 +00:00
#include <sys/filio.h>
Added used include of <sys/sx.h>. Don't depend on namespace pollution in <sys/file.h>.
2002-03-26 01:09:51 +00:00
#include <sys/sx.h>
Don't include <sys/ioctl.h> in the kernel. Stage 4: include <sys/ttycom.h> and sometimes <sys/filio.h> instead of <sys/ioctl.h> in miscellaneous files. Most of these files have nothing to do with ttys but need to include <sys/ttycom.h> to get the definitions of TIOC[SG]PGRP which are (ab)used to convert F[SG]ETOWN fcntls into ioctls.
1997-03-24 11:52:29 +00:00
#include <sys/ttycom.h>
oops: Add missing include.
1999-08-13 11:22:48 +00:00
#include <sys/conf.h>
This is a forward port of Peter's vlrureclaim() fix, with some minor mods by me to make it more efficient. The original code had serious balancing problems and could also deadlock easily. This code relegates the vnode reclamation to its own kproc and relaxes the vnode reclamation requirements to better maintain kern.maxvnodes. This code still doesn't balance as well as it could, but it does a much better job then the original code. Approved by: re@freebsd.org Obtained from: ps, peter, dillon MFS Assuming: Assuming no problems crop up in Yahoo testing MFC after: 7 days
2001-12-18 20:48:54 +00:00
#include <sys/syslog.h>
Push the VOP_ADVLOCK() call to release advisory locks on vnode file descriptors out of fdrop_locked() and into vn_closefile(). This removes all knowledge of vnodes from fdrop_locked(), since the lock behavior was specific to vnodes. This also removes the specific requirement for Giant in fdrop_locked(), it's now only required by code that it calls into. Add GIANT_REQUIRED to vn_closefile() since VFS requires Giant.
2004-06-01 18:03:20 +00:00
#include <sys/unistd.h>
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
White-space changes.
2002-12-24 09:44:51 +00:00
static fo_rdwr_t vn_read;
static fo_rdwr_t vn_write;
static fo_ioctl_t vn_ioctl;
static fo_poll_t vn_poll;
static fo_kqfilter_t vn_kqfilter;
static fo_stat_t vn_statfile;
static fo_close_t vn_closefile;
Staticize. Unstaticize a function in scsi/scsi_base that was used, with an undocumented option. My last count on the LINT kernel shows: Total symbols: 3647 unref symbols: 463 undef symbols: 4 1 ref symbols: 1751 2 ref symbols: 485 Approaching the pain threshold now.
1995-12-17 21:23:44 +00:00
Extend kqueue down to the device layer. Backwards compatible approach suggested by: peter
2001-02-15 16:34:11 +00:00
struct fileops vnops = {
Initialize struct fileops with C99 sparse initialization.
2003-06-18 18:16:40 +00:00
.fo_read = vn_read,
.fo_write = vn_write,
.fo_ioctl = vn_ioctl,
.fo_poll = vn_poll,
.fo_kqfilter = vn_kqfilter,
.fo_stat = vn_statfile,
.fo_close = vn_closefile,
Introduce a new flag on a file descriptor: DFLAG_SEEKABLE and use that rather than assume that only DTYPE_VNODE is seekable.
2003-06-18 19:53:59 +00:00
.fo_flags = DFLAG_PASSABLE | DFLAG_SEEKABLE
Introduce kqueue() and kevent(), a kernel event notification facility.
2000-04-16 18:53:38 +00:00
};
turn vn_open() into a wrapper around vn_open_cred() which allows one to perform a vn_open using temporary/other/fake credentials. Modify the nfs client side locking code to use vn_open_cred() passing proc0's ucred instead of the old way which was to temporary raise privs while running vn_open(). This should close the race hopefully.
2001-11-11 22:39:07 +00:00
int
Add fdidx argument to vn_open() and vn_open_cred() and pass -1 throughout.
2003-07-27 17:04:56 +00:00
vn_open(ndp, flagp, cmode, fdidx)
Pass the fdidx argument from vn_open{_cred}() onto VOP_OPEN()
2003-07-27 20:05:36 +00:00
struct nameidata *ndp;
Add fdidx argument to vn_open() and vn_open_cred() and pass -1 throughout.
2003-07-27 17:04:56 +00:00
int *flagp, cmode, fdidx;
turn vn_open() into a wrapper around vn_open_cred() which allows one to perform a vn_open using temporary/other/fake credentials. Modify the nfs client side locking code to use vn_open_cred() passing proc0's ucred instead of the old way which was to temporary raise privs while running vn_open(). This should close the race hopefully.
2001-11-11 22:39:07 +00:00
{
struct thread *td = ndp->ni_cnd.cn_thread;
Add fdidx argument to vn_open() and vn_open_cred() and pass -1 throughout.
2003-07-27 17:04:56 +00:00
return (vn_open_cred(ndp, flagp, cmode, td->td_ucred, fdidx));
turn vn_open() into a wrapper around vn_open_cred() which allows one to perform a vn_open using temporary/other/fake credentials. Modify the nfs client side locking code to use vn_open_cred() passing proc0's ucred instead of the old way which was to temporary raise privs while running vn_open(). This should close the race hopefully.
2001-11-11 22:39:07 +00:00
}
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
/*
* Common code for vnode open operations.
* Check permissions, and call the VOP_OPEN or VOP_CREATE routine.
Introduce NDFREE (and remove VOP_ABORTOP)
1999-12-15 23:02:35 +00:00
*
Fix comment typo. Submitted by: nrahlstr
2000-05-12 16:06:49 +00:00
* Note that this does NOT free nameidata for the successful case,
Introduce NDFREE (and remove VOP_ABORTOP)
1999-12-15 23:02:35 +00:00
* due to the NDINIT being done elsewhere.
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
*/
The big 4.4BSD Lite to FreeBSD 2.0.0 (Development) patch. Reviewed by: Rodney W. Grimes Submitted by: John Dyson and David Greenman
1994-05-25 09:21:21 +00:00
int
Add fdidx argument to vn_open() and vn_open_cred() and pass -1 throughout.
2003-07-27 17:04:56 +00:00
vn_open_cred(ndp, flagp, cmode, cred, fdidx)
Pass the fdidx argument from vn_open{_cred}() onto VOP_OPEN()
2003-07-27 20:05:36 +00:00
struct nameidata *ndp;
Move the truncation code out of vn_open and into the open system call after the acquisition of any advisory locks. This fix corrects a case in which a process tries to open a file with a non-blocking exclusive lock. Even if it fails to get the lock it would still truncate the file even though its open failed. With this change, the truncation is done only after the lock is successfully acquired. Obtained from: BSD/OS
2000-07-04 03:34:11 +00:00
int *flagp, cmode;
turn vn_open() into a wrapper around vn_open_cred() which allows one to perform a vn_open using temporary/other/fake credentials. Modify the nfs client side locking code to use vn_open_cred() passing proc0's ucred instead of the old way which was to temporary raise privs while running vn_open(). This should close the race hopefully.
2001-11-11 22:39:07 +00:00
struct ucred *cred;
Add fdidx argument to vn_open() and vn_open_cred() and pass -1 throughout.
2003-07-27 17:04:56 +00:00
int fdidx;
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
{
Move the truncation code out of vn_open and into the open system call after the acquisition of any advisory locks. This fix corrects a case in which a process tries to open a file with a non-blocking exclusive lock. Even if it fails to get the lock it would still truncate the file even though its open failed. With this change, the truncation is done only after the lock is successfully acquired. Obtained from: BSD/OS
2000-07-04 03:34:11 +00:00
struct vnode *vp;
Add snapshots to the fast filesystem. Most of the changes support the gating of system calls that cause modifications to the underlying filesystem. The gating can be enabled by any filesystem that needs to consistently suspend operations by adding the vop_stdgetwritemount to their set of vnops. Once gating is enabled, the function vfs_write_suspend stops all new write operations to a filesystem, allows any filesystem modifying system calls already in progress to complete, then sync's the filesystem to disk and returns. The function vfs_write_resume allows the suspended write operations to begin again. Gating is not added by default for all filesystems as for SMP systems it adds two extra locks to such critical kernel paths as the write system call. Thus, gating should only be added as needed. Details on the use and current status of snapshots in FFS can be found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness is not included here. Unless and until you create a snapshot file, these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
struct mount *mp;
KSE Milestone 2 Note ALL MODULES MUST BE RECOMPILED make the kernel aware that there are smaller units of scheduling than the process. (but only allow one thread per process at this time). This is functionally equivalent to teh previousl -current except that there is a thread associated with each process. Sorry john! (your next MFC will be a doosie!) Reviewed by: peter@freebsd.org, dillon@freebsd.org X-MFC after: ha ha ha ha
2001-09-12 08:38:13 +00:00
struct thread *td = ndp->ni_cnd.cn_thread;
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
struct vattr vat;
struct vattr *vap = &vat;
Move the truncation code out of vn_open and into the open system call after the acquisition of any advisory locks. This fix corrects a case in which a process tries to open a file with a non-blocking exclusive lock. Even if it fails to get the lock it would still truncate the file even though its open failed. With this change, the truncation is done only after the lock is successfully acquired. Obtained from: BSD/OS
2000-07-04 03:34:11 +00:00
int mode, fmode, error;
Change vn_open() to honor the MPSAFE flag in the passed in nameidata object and use that instead of testing fdidx against -1 to determine if it should release Giant if Giant was locked due to the requested file residing on a non-MPSAFE VFS. Discussed with: jeff
2006-03-28 21:22:08 +00:00
int vfslocked, mpsafe;
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
Change vn_open() to honor the MPSAFE flag in the passed in nameidata object and use that instead of testing fdidx against -1 to determine if it should release Giant if Giant was locked due to the requested file residing on a non-MPSAFE VFS. Discussed with: jeff
2006-03-28 21:22:08 +00:00
mpsafe = ndp->ni_cnd.cn_flags & MPSAFE;
Add snapshots to the fast filesystem. Most of the changes support the gating of system calls that cause modifications to the underlying filesystem. The gating can be enabled by any filesystem that needs to consistently suspend operations by adding the vop_stdgetwritemount to their set of vnops. Once gating is enabled, the function vfs_write_suspend stops all new write operations to a filesystem, allows any filesystem modifying system calls already in progress to complete, then sync's the filesystem to disk and returns. The function vfs_write_resume allows the suspended write operations to begin again. Gating is not added by default for all filesystems as for SMP systems it adds two extra locks to such critical kernel paths as the write system call. Thus, gating should only be added as needed. Details on the use and current status of snapshots in FFS can be found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness is not included here. Unless and until you create a snapshot file, these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
restart:
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
vfslocked = 0;
Move the truncation code out of vn_open and into the open system call after the acquisition of any advisory locks. This fix corrects a case in which a process tries to open a file with a non-blocking exclusive lock. Even if it fails to get the lock it would still truncate the file even though its open failed. With this change, the truncation is done only after the lock is successfully acquired. Obtained from: BSD/OS
2000-07-04 03:34:11 +00:00
fmode = *flagp;
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
if (fmode & O_CREAT) {
ndp->ni_cnd.cn_nameiop = CREATE;
Make sure that we are adding a path token to the audit record in open(2). Do this by making sure we are using the AUDITVNODE1 mask in the namei flags. Obtained from: TrustedBSD Project
2006-03-11 17:14:05 +00:00
ndp->ni_cnd.cn_flags = ISOPEN | LOCKPARENT | LOCKLEAF |
MPSAFE | AUDITVNODE1;
Rather than let users get fd's to symlink files, make O_NOFOLLOW cause an error if it gets a link (like it does if it gets a socket). The implications of letting users try and do file operations on symlinks themselves were too worrying.
1998-04-06 18:25:21 +00:00
if ((fmode & O_EXCL) == 0 && (fmode & O_NOFOLLOW) == 0)
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
ndp->ni_cnd.cn_flags |= FOLLOW;
Add bwillwrite to all system calls that create things in the filesystem. Benchmarks that create huge trees of empty files overwhelm the buffer cache.
2000-01-10 00:08:53 +00:00
bwillwrite();
Move the truncation code out of vn_open and into the open system call after the acquisition of any advisory locks. This fix corrects a case in which a process tries to open a file with a non-blocking exclusive lock. Even if it fails to get the lock it would still truncate the file even though its open failed. With this change, the truncation is done only after the lock is successfully acquired. Obtained from: BSD/OS
2000-07-04 03:34:11 +00:00
if ((error = namei(ndp)) != 0)
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
return (error);
Change vn_open() to honor the MPSAFE flag in the passed in nameidata object and use that instead of testing fdidx against -1 to determine if it should release Giant if Giant was locked due to the requested file residing on a non-MPSAFE VFS. Discussed with: jeff
2006-03-28 21:22:08 +00:00
vfslocked = NDHASGIANT(ndp);
if (!mpsafe)
ndp->ni_cnd.cn_flags &= ~MPSAFE;
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
if (ndp->ni_vp == NULL) {
VATTR_NULL(vap);
vap->va_type = VREG;
vap->va_mode = cmode;
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
if (fmode & O_EXCL)
vap->va_vaflags |= VA_EXCLUSIVE;
Add snapshots to the fast filesystem. Most of the changes support the gating of system calls that cause modifications to the underlying filesystem. The gating can be enabled by any filesystem that needs to consistently suspend operations by adding the vop_stdgetwritemount to their set of vnops. Once gating is enabled, the function vfs_write_suspend stops all new write operations to a filesystem, allows any filesystem modifying system calls already in progress to complete, then sync's the filesystem to disk and returns. The function vfs_write_resume allows the suspended write operations to begin again. Gating is not added by default for all filesystems as for SMP systems it adds two extra locks to such critical kernel paths as the write system call. Thus, gating should only be added as needed. Details on the use and current status of snapshots in FFS can be found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness is not included here. Unless and until you create a snapshot file, these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
if (vn_start_write(ndp->ni_dvp, &mp, V_NOWAIT) != 0) {
NDFREE(ndp, NDF_ONLY_PNBUF);
vput(ndp->ni_dvp);
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
VFS_UNLOCK_GIANT(vfslocked);
Add snapshots to the fast filesystem. Most of the changes support the gating of system calls that cause modifications to the underlying filesystem. The gating can be enabled by any filesystem that needs to consistently suspend operations by adding the vop_stdgetwritemount to their set of vnops. Once gating is enabled, the function vfs_write_suspend stops all new write operations to a filesystem, allows any filesystem modifying system calls already in progress to complete, then sync's the filesystem to disk and returns. The function vfs_write_resume allows the suspended write operations to begin again. Gating is not added by default for all filesystems as for SMP systems it adds two extra locks to such critical kernel paths as the write system call. Thus, gating should only be added as needed. Details on the use and current status of snapshots in FFS can be found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness is not included here. Unless and until you create a snapshot file, these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
if ((error = vn_start_write(NULL, &mp,
V_XSLEEP | PCATCH)) != 0)
return (error);
goto restart;
}
Drop in the MAC check for file creation as part of open(). Approved by: re Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2002-10-19 20:56:44 +00:00
#ifdef MAC
error = mac_check_vnode_create(cred, ndp->ni_dvp,
&ndp->ni_cnd, vap);
if (error == 0) {
#endif
VOP_LEASE(ndp->ni_dvp, td, cred, LEASE_WRITE);
error = VOP_CREATE(ndp->ni_dvp, &ndp->ni_vp,
&ndp->ni_cnd, vap);
#ifdef MAC
}
#endif
Move the truncation code out of vn_open and into the open system call after the acquisition of any advisory locks. This fix corrects a case in which a process tries to open a file with a non-blocking exclusive lock. Even if it fails to get the lock it would still truncate the file even though its open failed. With this change, the truncation is done only after the lock is successfully acquired. Obtained from: BSD/OS
2000-07-04 03:34:11 +00:00
vput(ndp->ni_dvp);
Add snapshots to the fast filesystem. Most of the changes support the gating of system calls that cause modifications to the underlying filesystem. The gating can be enabled by any filesystem that needs to consistently suspend operations by adding the vop_stdgetwritemount to their set of vnops. Once gating is enabled, the function vfs_write_suspend stops all new write operations to a filesystem, allows any filesystem modifying system calls already in progress to complete, then sync's the filesystem to disk and returns. The function vfs_write_resume allows the suspended write operations to begin again. Gating is not added by default for all filesystems as for SMP systems it adds two extra locks to such critical kernel paths as the write system call. Thus, gating should only be added as needed. Details on the use and current status of snapshots in FFS can be found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness is not included here. Unless and until you create a snapshot file, these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
vn_finished_write(mp);
Introduce NDFREE (and remove VOP_ABORTOP)
1999-12-15 23:02:35 +00:00
if (error) {
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
VFS_UNLOCK_GIANT(vfslocked);
Introduce NDFREE (and remove VOP_ABORTOP)
1999-12-15 23:02:35 +00:00
NDFREE(ndp, NDF_ONLY_PNBUF);
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
return (error);
Introduce NDFREE (and remove VOP_ABORTOP)
1999-12-15 23:02:35 +00:00
}
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
fmode &= ~O_TRUNC;
vp = ndp->ni_vp;
} else {
if (ndp->ni_dvp == ndp->ni_vp)
vrele(ndp->ni_dvp);
else
vput(ndp->ni_dvp);
ndp->ni_dvp = NULL;
vp = ndp->ni_vp;
if (fmode & O_EXCL) {
error = EEXIST;
goto bad;
}
fmode &= ~O_CREAT;
}
} else {
ndp->ni_cnd.cn_nameiop = LOOKUP;
- Pass the ISOPEN flag to namei so filesystems will know we're about to open them or otherwise access the data.
2005-04-27 09:05:19 +00:00
ndp->ni_cnd.cn_flags = ISOPEN |
This patch adds the "LOCKSHARED" option to namei which causes it to only acquire shared locks on leafs. The stat() and open() calls have been changed to make use of this new functionality. Using shared locks in these cases is sufficient and can significantly reduce their latency if IO is pending to these vnodes. Also, this reduces the number of exclusive locks that are floating around in the system, which helps reduce the number of deadlocks that occur. A new kernel option "LOOKUP_SHARED" has been added. It defaults to off so this patch can be turned on for testing, and should eventually go away once it is proven to be stable. I have personally been running this patch for over a year now, so it is believed to be fully stable. Reviewed by: jake, obrien Approved by: jake
2002-03-12 04:00:11 +00:00
((fmode & O_NOFOLLOW) ? NOFOLLOW : FOLLOW) |
Make sure that we are adding a path token to the audit record in open(2). Do this by making sure we are using the AUDITVNODE1 mask in the namei flags. Obtained from: TrustedBSD Project
2006-03-11 17:14:05 +00:00
LOCKSHARED | LOCKLEAF | MPSAFE | AUDITVNODE1;
Move the truncation code out of vn_open and into the open system call after the acquisition of any advisory locks. This fix corrects a case in which a process tries to open a file with a non-blocking exclusive lock. Even if it fails to get the lock it would still truncate the file even though its open failed. With this change, the truncation is done only after the lock is successfully acquired. Obtained from: BSD/OS
2000-07-04 03:34:11 +00:00
if ((error = namei(ndp)) != 0)
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
return (error);
Change vn_open() to honor the MPSAFE flag in the passed in nameidata object and use that instead of testing fdidx against -1 to determine if it should release Giant if Giant was locked due to the requested file residing on a non-MPSAFE VFS. Discussed with: jeff
2006-03-28 21:22:08 +00:00
if (!mpsafe)
ndp->ni_cnd.cn_flags &= ~MPSAFE;
vfslocked = NDHASGIANT(ndp);
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
vp = ndp->ni_vp;
}
Use a different errno (ELOOP (as sef mentioned) since the text that goes with the error sounds ok for the condition) if O_NOFOLLOW gets a link.
1998-04-06 18:43:28 +00:00
if (vp->v_type == VLNK) {
Today is not my lucky day. Fix missing brace and I got a request to use EMLINK instead.
1998-04-06 19:32:37 +00:00
error = EMLINK;
Use a different errno (ELOOP (as sef mentioned) since the text that goes with the error sounds ok for the condition) if O_NOFOLLOW gets a link.
1998-04-06 18:43:28 +00:00
goto bad;
}
Today is not my lucky day. Fix missing brace and I got a request to use EMLINK instead.
1998-04-06 19:32:37 +00:00
if (vp->v_type == VSOCK) {
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
error = EOPNOTSUPP;
goto bad;
}
Introduce support for Mandatory Access Control and extensible kernel access control. Restructure the vn_open_cred() access control checks to invoke the MAC entry point for open authorization. Note that MAC can reject open requests where existing DAC code skips the open authorization check due to O_CREAT. However, the failure mode here is the same as other failure modes following creation, wherein an empty file may be left behind. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-01 17:14:28 +00:00
mode = 0;
if (fmode & (FWRITE | O_TRUNC)) {
if (vp->v_type == VDIR) {
error = EISDIR;
goto bad;
}
mode |= VWRITE;
}
if (fmode & FREAD)
mode |= VREAD;
if (fmode & O_APPEND)
mode |= VAPPEND;
#ifdef MAC
error = mac_check_vnode_open(cred, vp, mode);
if (error)
goto bad;
#endif
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
if ((fmode & O_CREAT) == 0) {
Introduce support for Mandatory Access Control and extensible kernel access control. Restructure the vn_open_cred() access control checks to invoke the MAC entry point for open authorization. Note that MAC can reject open requests where existing DAC code skips the open authorization check due to O_CREAT. However, the failure mode here is the same as other failure modes following creation, wherein an empty file may be left behind. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-01 17:14:28 +00:00
if (mode & VWRITE) {
All of this is cosmetic. prototypes, #includes, printfs and so on. Makes GCC a lot more silent.
1994-10-02 17:35:40 +00:00
error = vn_writechk(vp);
Unlock the vnode before sleeping on an OBJ_DEAD object. Should fix Bruce's hang. Fixed some formatting anomolies and removed some unneeded casts.
1995-05-10 18:59:11 +00:00
if (error)
All of this is cosmetic. prototypes, #includes, printfs and so on. Makes GCC a lot more silent.
1994-10-02 17:35:40 +00:00
goto bad;
Only do one VOP_ACCESS() per open() instead of two. This should reduce the NFSv3 ACCESS RPC problems a little for busy clients that do a lot of open/close. The nfs code could probably cache the results, but I'm not sure whether this would be legal or useful. The problem is that with a CPU farm, on each open there would be a lookup, getattr then access RPC then the read/write RPC activity. Caching the access results probably isn't going to help much if the clients access lots of files. Having the nfs_access() routine interpret the getattr results is a bit of a hack, but it's how NFSv2 is done and it might be OK for a mount attribute for v3.
1998-11-02 02:36:16 +00:00
}
if (mode) {
KSE Milestone 2 Note ALL MODULES MUST BE RECOMPILED make the kernel aware that there are smaller units of scheduling than the process. (but only allow one thread per process at this time). This is functionally equivalent to teh previousl -current except that there is a thread associated with each process. Sorry john! (your next MFC will be a doosie!) Reviewed by: peter@freebsd.org, dillon@freebsd.org X-MFC after: ha ha ha ha
2001-09-12 08:38:13 +00:00
error = VOP_ACCESS(vp, mode, cred, td);
Unlock the vnode before sleeping on an OBJ_DEAD object. Should fix Bruce's hang. Fixed some formatting anomolies and removed some unneeded casts.
1995-05-10 18:59:11 +00:00
if (error)
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
goto bad;
}
}
Pass the fdidx argument from vn_open{_cred}() onto VOP_OPEN()
2003-07-27 20:05:36 +00:00
if ((error = VOP_OPEN(vp, fmode, cred, td, fdidx)) != 0)
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
goto bad;
Even though this looks like it, this is not a complex code change. The interface into the "VMIO" system has changed to be more consistant and robust. Essentially, it is now no longer necessary to call vn_open to get merged VM/Buffer cache operation, and exceptional conditions such as merged operation of VBLK devices is simpler and more correct. This code corrects a potentially large set of problems including the problems with ktrace output and loaded systems, file create/deletes, etc. Most of the changes to NFS are cosmetic and name changes, eliminating a layer of subroutine calls. The direct calls to vput/vrele have been re-instituted for better cross platform compatibility. Reviewed by: davidg
1996-08-21 21:56:23 +00:00
NOTE: libkvm, w, ps, 'top', and any other utility which depends on struct proc or any VM system structure will have to be rebuilt!!! Much needed overhaul of the VM system. Included in this first round of changes: 1) Improved pager interfaces: init, alloc, dealloc, getpages, putpages, haspage, and sync operations are supported. The haspage interface now provides information about clusterability. All pager routines now take struct vm_object's instead of "pagers". 2) Improved data structures. In the previous paradigm, there is constant confusion caused by pagers being both a data structure ("allocate a pager") and a collection of routines. The idea of a pager structure has escentially been eliminated. Objects now have types, and this type is used to index the appropriate pager. In most cases, items in the pager structure were duplicated in the object data structure and thus were unnecessary. In the few cases that remained, a un_pager structure union was created in the object to contain these items. 3) Because of the cleanup of #1 & #2, a lot of unnecessary layering can now be removed. For instance, vm_object_enter(), vm_object_lookup(), vm_object_remove(), and the associated object hash list were some of the things that were removed. 4) simple_lock's removed. Discussion with several people reveals that the SMP locking primitives used in the VM system aren't likely the mechanism that we'll be adopting. Even if it were, the locking that was in the code was very inadequate and would have to be mostly re-done anyway. The locking in a uni-processor kernel was a no-op but went a long way toward making the code difficult to read and debug. 5) Places that attempted to kludge-up the fact that we don't have kernel thread support have been fixed to reflect the reality that we are really dealing with processes, not threads. The VM system didn't have complete thread support, so the comments and mis-named routines were just wrong. We now use tsleep and wakeup directly in the lock routines, for instance. 6) Where appropriate, the pagers have been improved, especially in the pager_alloc routines. Most of the pager_allocs have been rewritten and are now faster and easier to maintain. 7) The pagedaemon pageout clustering algorithm has been rewritten and now tries harder to output an even number of pages before and after the requested page. This is sort of the reverse of the ideal pagein algorithm and should provide better overall performance. 8) Unnecessary (incorrect) casts to caddr_t in calls to tsleep & wakeup have been removed. Some other unnecessary casts have also been removed. 9) Some almost useless debugging code removed. 10) Terminology of shadow objects vs. backing objects straightened out. The fact that the vm_object data structure escentially had this backwards really confused things. The use of "shadow" and "backing object" throughout the code is now internally consistent and correct in the Mach terminology. 11) Several minor bug fixes, including one in the vm daemon that caused 0 RSS objects to not get purged as intended. 12) A "default pager" has now been created which cleans up the transition of objects to the "swap" type. The previous checks throughout the code for swp->pg_data != NULL were really ugly. This change also provides the rudiments for future backing of "anonymous" memory by something other than the swap pager (via the vnode pager, for example), and it allows the decision about which of these pagers to use to be made dynamically (although will need some additional decision code to do this, of course). 13) (dyson) MAP_COPY has been deprecated and the corresponding "copy object" code has been removed. MAP_COPY was undocumented and non- standard. It was furthermore broken in several ways which caused its behavior to degrade to MAP_PRIVATE. Binaries that use MAP_COPY will continue to work correctly, but via the slightly different semantics of MAP_PRIVATE. 14) (dyson) Sharing maps have been removed. It's marginal usefulness in a threads design can be worked around in other ways. Both #12 and #13 were done to simplify the code and improve readability and maintain- ability. (As were most all of these changes) TODO: 1) Rewrite most of the vnode pager to use VOP_GETPAGES/PUTPAGES. Doing this will reduce the vnode pager to a mere fraction of its current size. 2) Rewrite vm_fault and the swap/vnode pagers to use the clustering information provided by the new haspage pager interface. This will substantially reduce the overhead by eliminating a large number of VOP_BMAP() calls. The VOP_BMAP() filesystem interface should be improved to provide both a "behind" and "ahead" indication of contiguousness. 3) Implement the extended features of pager_haspage in swap_pager_haspage(). It currently just says 0 pages ahead/behind. 4) Re-implement the swap device (swstrategy) in a more elegant way, perhaps via a much more general mechanism that could also be used for disk striping of regular filesystems. 5) Do something to improve the architecture of vm_object_collapse(). The fact that it makes calls into the swap pager and knows too much about how the swap pager operates really bothers me. It also doesn't allow for collapsing of non-swap pager objects ("unnamed" objects backed by other pagers).
1995-07-13 08:48:48 +00:00
if (fmode & FWRITE)
vp->v_writecount++;
Move the truncation code out of vn_open and into the open system call after the acquisition of any advisory locks. This fix corrects a case in which a process tries to open a file with a non-blocking exclusive lock. Even if it fails to get the lock it would still truncate the file even though its open failed. With this change, the truncation is done only after the lock is successfully acquired. Obtained from: BSD/OS
2000-07-04 03:34:11 +00:00
*flagp = fmode;
Assert the vnode lock when returning successfully from vn_open_cred().
2003-06-04 00:54:27 +00:00
ASSERT_VOP_LOCKED(vp, "vn_open_cred");
Change vn_open() to honor the MPSAFE flag in the passed in nameidata object and use that instead of testing fdidx against -1 to determine if it should release Giant if Giant was locked due to the requested file residing on a non-MPSAFE VFS. Discussed with: jeff
2006-03-28 21:22:08 +00:00
if (!mpsafe)
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
VFS_UNLOCK_GIANT(vfslocked);
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
return (0);
bad:
Introduce NDFREE (and remove VOP_ABORTOP)
1999-12-15 23:02:35 +00:00
NDFREE(ndp, NDF_ONLY_PNBUF);
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
vput(vp);
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
VFS_UNLOCK_GIANT(vfslocked);
Move the truncation code out of vn_open and into the open system call after the acquisition of any advisory locks. This fix corrects a case in which a process tries to open a file with a non-blocking exclusive lock. Even if it fails to get the lock it would still truncate the file even though its open failed. With this change, the truncation is done only after the lock is successfully acquired. Obtained from: BSD/OS
2000-07-04 03:34:11 +00:00
*flagp = fmode;
In vn_open(), unset ndp->ni_vp when returning failure so that code which expects it to be NULL unless the return value was 0 will work. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-01-07 20:59:55 +00:00
ndp->ni_vp = NULL;
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
return (error);
}
/*
* Check for write permissions on the specified vnode.
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
* Prototype text segments cannot be written.
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
*/
The big 4.4BSD Lite to FreeBSD 2.0.0 (Development) patch. Reviewed by: Rodney W. Grimes Submitted by: John Dyson and David Greenman
1994-05-25 09:21:21 +00:00
int
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
vn_writechk(vp)
register struct vnode *vp;
{
- Replace v_flag with v_iflag and v_vflag - v_vflag is protected by the vnode lock and is used when synchronization with VOP calls is needed. - v_iflag is protected by interlock and is used for dealing with vnode management issues. These flags include X/O LOCK, FREE, DOOMED, etc. - All accesses to v_iflag and v_vflag have either been locked or marked with mp_fixme's. - Many ASSERT_VOP_LOCKED calls have been added where the locking was not clear. - Many functions in vfs_subr.c were restructured to provide for stronger locking. Idea stolen from: BSD/OS
2002-08-04 10:29:36 +00:00
ASSERT_VOP_LOCKED(vp, "vn_writechk");
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
/*
* If there's shared text associated with
* the vnode, try to free it up once. If
* we fail, we can't allow writing.
*/
- Replace v_flag with v_iflag and v_vflag - v_vflag is protected by the vnode lock and is used when synchronization with VOP calls is needed. - v_iflag is protected by interlock and is used for dealing with vnode management issues. These flags include X/O LOCK, FREE, DOOMED, etc. - All accesses to v_iflag and v_vflag have either been locked or marked with mp_fixme's. - Many ASSERT_VOP_LOCKED calls have been added where the locking was not clear. - Many functions in vfs_subr.c were restructured to provide for stronger locking. Idea stolen from: BSD/OS
2002-08-04 10:29:36 +00:00
if (vp->v_vflag & VV_TEXT)
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
return (ETXTBSY);
- Replace v_flag with v_iflag and v_vflag - v_vflag is protected by the vnode lock and is used when synchronization with VOP calls is needed. - v_iflag is protected by interlock and is used for dealing with vnode management issues. These flags include X/O LOCK, FREE, DOOMED, etc. - All accesses to v_iflag and v_vflag have either been locked or marked with mp_fixme's. - Many ASSERT_VOP_LOCKED calls have been added where the locking was not clear. - Many functions in vfs_subr.c were restructured to provide for stronger locking. Idea stolen from: BSD/OS
2002-08-04 10:29:36 +00:00
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
return (0);
}
/*
* Vnode close call
*/
The big 4.4BSD Lite to FreeBSD 2.0.0 (Development) patch. Reviewed by: Rodney W. Grimes Submitted by: John Dyson and David Greenman
1994-05-25 09:21:21 +00:00
int
Make similar changes to fo_stat() and fo_poll() as made earlier to fo_read() and fo_write(): explicitly use the cred argument to fo_poll() as "active_cred" using the passed file descriptor's f_cred reference to provide access to the file credential. Add an active_cred argument to fo_stat() so that implementers have access to the active credential as well as the file credential. Generally modify callers of fo_stat() to pass in td->td_ucred rather than fp->f_cred, which was redundantly provided via the fp argument. This set of modifications also permits threads to perform these operations on behalf of another thread without modifying their credential. Trickle this change down into fo_stat/poll() implementations: - badfo_poll(), badfo_stat(): modify/add arguments. - kqueue_poll(), kqueue_stat(): modify arguments. - pipe_poll(), pipe_stat(): modify/add arguments, pass active_cred to MAC checks rather than td->td_ucred. - soo_poll(), soo_stat(): modify/add arguments, pass fp->f_cred rather than cred to pru_sopoll() to maintain current semantics. - sopoll(): moidfy arguments. - vn_poll(), vn_statfile(): modify/add arguments, pass new arguments to vn_stat(). Pass active_cred to MAC and fp->f_cred to VOP_POLL() to maintian current semantics. - vn_close(): rename cred to file_cred to reflect reality while I'm here. - vn_stat(): Add active_cred and file_cred arguments to vn_stat() and consumers so that this distinction is maintained at the VFS as well as 'struct file' layer. Pass active_cred instead of td->td_ucred to MAC and to VOP_GETATTR() to maintain current semantics. - fifofs: modify the creation of a "filetemp" so that the file credential is properly initialized and can be used in the socket code if desired. Pass ap->a_td->td_ucred as the active credential to soo_poll(). If we teach the vnop interface about the distinction between file and active credentials, we would use the active credential here. Note that current inconsistent passing of active_cred vs. file_cred to VOP's is maintained. It's not clear why GETATTR would be authorized using active_cred while POLL would be authorized using file_cred at the file system level. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-16 12:52:03 +00:00
vn_close(vp, flags, file_cred, td)
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
register struct vnode *vp;
int flags;
Make similar changes to fo_stat() and fo_poll() as made earlier to fo_read() and fo_write(): explicitly use the cred argument to fo_poll() as "active_cred" using the passed file descriptor's f_cred reference to provide access to the file credential. Add an active_cred argument to fo_stat() so that implementers have access to the active credential as well as the file credential. Generally modify callers of fo_stat() to pass in td->td_ucred rather than fp->f_cred, which was redundantly provided via the fp argument. This set of modifications also permits threads to perform these operations on behalf of another thread without modifying their credential. Trickle this change down into fo_stat/poll() implementations: - badfo_poll(), badfo_stat(): modify/add arguments. - kqueue_poll(), kqueue_stat(): modify arguments. - pipe_poll(), pipe_stat(): modify/add arguments, pass active_cred to MAC checks rather than td->td_ucred. - soo_poll(), soo_stat(): modify/add arguments, pass fp->f_cred rather than cred to pru_sopoll() to maintain current semantics. - sopoll(): moidfy arguments. - vn_poll(), vn_statfile(): modify/add arguments, pass new arguments to vn_stat(). Pass active_cred to MAC and fp->f_cred to VOP_POLL() to maintian current semantics. - vn_close(): rename cred to file_cred to reflect reality while I'm here. - vn_stat(): Add active_cred and file_cred arguments to vn_stat() and consumers so that this distinction is maintained at the VFS as well as 'struct file' layer. Pass active_cred instead of td->td_ucred to MAC and to VOP_GETATTR() to maintain current semantics. - fifofs: modify the creation of a "filetemp" so that the file credential is properly initialized and can be used in the socket code if desired. Pass ap->a_td->td_ucred as the active credential to soo_poll(). If we teach the vnop interface about the distinction between file and active credentials, we would use the active credential here. Note that current inconsistent passing of active_cred vs. file_cred to VOP's is maintained. It's not clear why GETATTR would be authorized using active_cred while POLL would be authorized using file_cred at the file system level. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-16 12:52:03 +00:00
struct ucred *file_cred;
KSE Milestone 2 Note ALL MODULES MUST BE RECOMPILED make the kernel aware that there are smaller units of scheduling than the process. (but only allow one thread per process at this time). This is functionally equivalent to teh previousl -current except that there is a thread associated with each process. Sorry john! (your next MFC will be a doosie!) Reviewed by: peter@freebsd.org, dillon@freebsd.org X-MFC after: ha ha ha ha
2001-09-12 08:38:13 +00:00
struct thread *td;
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
{
- Do a vn_start_write in vn_close, we may write if this is the last ref on an unlinked file. We can't know if this is the case until after we have the lock. - Lock the vnode in vn_close, many filesystems had code which was unsafe without the lock held, and holding it greatly simplifies vgone(). - Adjust vn_lock() to check for the VI_DOOMED flag where appropriate. Sponsored by: Isilon Systems, Inc.
2005-03-13 11:56:28 +00:00
struct mount *mp;
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
int error;
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
VFS_ASSERT_GIANT(vp->v_mount);
Flag a broad range of VFS operations as GIANT_REQUIRED in order to catch leaking into VFS without Giant. Inch Giant a little lower in several file descriptor operations on vnodes to cover only VFS operations that need it, rather than file flag reading, etc.
2004-08-06 22:25:35 +00:00
- Do a vn_start_write in vn_close, we may write if this is the last ref on an unlinked file. We can't know if this is the case until after we have the lock. - Lock the vnode in vn_close, many filesystems had code which was unsafe without the lock held, and holding it greatly simplifies vgone(). - Adjust vn_lock() to check for the VI_DOOMED flag where appropriate. Sponsored by: Isilon Systems, Inc.
2005-03-13 11:56:28 +00:00
vn_start_write(vp, &mp, V_WAIT);
vn_lock(vp, LK_EXCLUSIVE | LK_RETRY, td);
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
if (flags & FWRITE)
vp->v_writecount--;
Make similar changes to fo_stat() and fo_poll() as made earlier to fo_read() and fo_write(): explicitly use the cred argument to fo_poll() as "active_cred" using the passed file descriptor's f_cred reference to provide access to the file credential. Add an active_cred argument to fo_stat() so that implementers have access to the active credential as well as the file credential. Generally modify callers of fo_stat() to pass in td->td_ucred rather than fp->f_cred, which was redundantly provided via the fp argument. This set of modifications also permits threads to perform these operations on behalf of another thread without modifying their credential. Trickle this change down into fo_stat/poll() implementations: - badfo_poll(), badfo_stat(): modify/add arguments. - kqueue_poll(), kqueue_stat(): modify arguments. - pipe_poll(), pipe_stat(): modify/add arguments, pass active_cred to MAC checks rather than td->td_ucred. - soo_poll(), soo_stat(): modify/add arguments, pass fp->f_cred rather than cred to pru_sopoll() to maintain current semantics. - sopoll(): moidfy arguments. - vn_poll(), vn_statfile(): modify/add arguments, pass new arguments to vn_stat(). Pass active_cred to MAC and fp->f_cred to VOP_POLL() to maintian current semantics. - vn_close(): rename cred to file_cred to reflect reality while I'm here. - vn_stat(): Add active_cred and file_cred arguments to vn_stat() and consumers so that this distinction is maintained at the VFS as well as 'struct file' layer. Pass active_cred instead of td->td_ucred to MAC and to VOP_GETATTR() to maintain current semantics. - fifofs: modify the creation of a "filetemp" so that the file credential is properly initialized and can be used in the socket code if desired. Pass ap->a_td->td_ucred as the active credential to soo_poll(). If we teach the vnop interface about the distinction between file and active credentials, we would use the active credential here. Note that current inconsistent passing of active_cred vs. file_cred to VOP's is maintained. It's not clear why GETATTR would be authorized using active_cred while POLL would be authorized using file_cred at the file system level. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-16 12:52:03 +00:00
error = VOP_CLOSE(vp, flags, file_cred, td);
- Do a vn_start_write in vn_close, we may write if this is the last ref on an unlinked file. We can't know if this is the case until after we have the lock. - Lock the vnode in vn_close, many filesystems had code which was unsafe without the lock held, and holding it greatly simplifies vgone(). - Adjust vn_lock() to check for the VI_DOOMED flag where appropriate. Sponsored by: Isilon Systems, Inc.
2005-03-13 11:56:28 +00:00
vput(vp);
vn_finished_write(mp);
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
return (error);
}
Document all functions, global and static variables, and sysctls. Includes some minor whitespace changes, and re-ordering to be able to document properly (e.g, grouping of variables and the SYSCTL macro calls for them, where the documentation has been added.) Reviewed by: phk (but all errors are mine)
2002-03-05 15:38:49 +00:00
/*
* Sequential heuristic - detect sequential operation
*/
Change the write-behind code to take more care when starting async I/O's. The sequential read heuristic has been extended to cover writes as well. We continue to call cluster_write() normally, thus blocks in the file will still be reallocated for large (but still random) I/O's, but I/O will only be initiated for truely sequential writes. This solves a number of annoying situations, especially with DBM (hash method) writes, and also has the side effect of fixing a number of (stupid) benchmarks. Reviewed-by: mckusick
2000-04-02 00:55:28 +00:00
static __inline
int
sequential_heuristic(struct uio *uio, struct file *fp)
{
SMP Lock struct file, filedesc and the global file list. Seigo Tanimura (tanimura) posted the initial delta. I've polished it quite a bit reducing the need for locking and adapting it for KSE. Locks: 1 mutex in each filedesc protects all the fields. protects "struct file" initialization, while a struct file is being changed from &badfileops -> &pipeops or something the filedesc should be locked. 1 mutex in each struct file protects the refcount fields. doesn't protect anything else. the flags used for garbage collection have been moved to f_gcflag which was the FILLER short, this doesn't need locking because the garbage collection is a single threaded container. could likely be made to use a pool mutex. 1 sx lock for the global filelist. struct file * fhold(struct file *fp); /* increments reference count on a file */ struct file * fhold_locked(struct file *fp); /* like fhold but expects file to locked */ struct file * ffind_hold(struct thread *, int fd); /* finds the struct file in thread, adds one reference and returns it unlocked */ struct file * ffind_lock(struct thread *, int fd); /* ffind_hold, but returns file locked */ I still have to smp-safe the fget cruft, I'll get to that asap.
2002-01-13 11:58:06 +00:00
Change the write-behind code to take more care when starting async I/O's. The sequential read heuristic has been extended to cover writes as well. We continue to call cluster_write() normally, thus blocks in the file will still be reallocated for large (but still random) I/O's, but I/O will only be initiated for truely sequential writes. This solves a number of annoying situations, especially with DBM (hash method) writes, and also has the side effect of fixing a number of (stupid) benchmarks. Reviewed-by: mckusick
2000-04-02 00:55:28 +00:00
if ((uio->uio_offset == 0 && fp->f_seqcount > 0) ||
uio->uio_offset == fp->f_nextoff) {
/*
* XXX we assume that the filesystem block size is
* the default. Not true, but still gives us a pretty
* good indicator of how sequential the read operations
* are.
*/
fp->f_seqcount += (uio->uio_resid + BKVASIZE - 1) / BKVASIZE;
Abstract-out the constants for the sequential heuristic. No operational changes. MFC after: 1 day
2002-12-28 20:28:10 +00:00
if (fp->f_seqcount > IO_SEQMAX)
fp->f_seqcount = IO_SEQMAX;
return(fp->f_seqcount << IO_SEQSHIFT);
Change the write-behind code to take more care when starting async I/O's. The sequential read heuristic has been extended to cover writes as well. We continue to call cluster_write() normally, thus blocks in the file will still be reallocated for large (but still random) I/O's, but I/O will only be initiated for truely sequential writes. This solves a number of annoying situations, especially with DBM (hash method) writes, and also has the side effect of fixing a number of (stupid) benchmarks. Reviewed-by: mckusick
2000-04-02 00:55:28 +00:00
}
/*
* Not sequential, quick draw-down of seqcount
*/
if (fp->f_seqcount > 1)
fp->f_seqcount = 1;
else
fp->f_seqcount = 0;
return(0);
}
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
/*
* Package up an I/O request on a vnode into a uio and do it.
*/
The big 4.4BSD Lite to FreeBSD 2.0.0 (Development) patch. Reviewed by: Rodney W. Grimes Submitted by: John Dyson and David Greenman
1994-05-25 09:21:21 +00:00
int
In order to better support flexible and extensible access control, make a series of modifications to the credential arguments relating to file read and write operations to cliarfy which credential is used for what: - Change fo_read() and fo_write() to accept "active_cred" instead of "cred", and change the semantics of consumers of fo_read() and fo_write() to pass the active credential of the thread requesting an operation rather than the cached file cred. The cached file cred is still available in fo_read() and fo_write() consumers via fp->f_cred. These changes largely in sys_generic.c. For each implementation of fo_read() and fo_write(), update cred usage to reflect this change and maintain current semantics: - badfo_readwrite() unchanged - kqueue_read/write() unchanged pipe_read/write() now authorize MAC using active_cred rather than td->td_ucred - soo_read/write() unchanged - vn_read/write() now authorize MAC using active_cred but VOP_READ/WRITE() with fp->f_cred Modify vn_rdwr() to accept two credential arguments instead of a single credential: active_cred and file_cred. Use active_cred for MAC authorization, and select a credential for use in VOP_READ/WRITE() based on whether file_cred is NULL or not. If file_cred is provided, authorize the VOP using that cred, otherwise the active credential, matching current semantics. Modify current vn_rdwr() consumers to pass a file_cred if used in the context of a struct file, and to always pass active_cred. When vn_rdwr() is used without a file_cred, pass NOCRED. These changes should maintain current semantics for read/write, but avoid a redundant passing of fp->f_cred, as well as making it more clear what the origin of each credential is in file descriptor read/write operations. Follow-up commits will make similar changes to other file descriptor operations, and modify the MAC framework to pass both credentials to MAC policy modules so they can implement either semantic for revocation. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-15 20:55:08 +00:00
vn_rdwr(rw, vp, base, len, offset, segflg, ioflg, active_cred, file_cred,
aresid, td)
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
enum uio_rw rw;
struct vnode *vp;
Eradicate caddr_t from the VFS API.
2005-12-14 00:49:52 +00:00
void *base;
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
int len;
off_t offset;
enum uio_seg segflg;
int ioflg;
In order to better support flexible and extensible access control, make a series of modifications to the credential arguments relating to file read and write operations to cliarfy which credential is used for what: - Change fo_read() and fo_write() to accept "active_cred" instead of "cred", and change the semantics of consumers of fo_read() and fo_write() to pass the active credential of the thread requesting an operation rather than the cached file cred. The cached file cred is still available in fo_read() and fo_write() consumers via fp->f_cred. These changes largely in sys_generic.c. For each implementation of fo_read() and fo_write(), update cred usage to reflect this change and maintain current semantics: - badfo_readwrite() unchanged - kqueue_read/write() unchanged pipe_read/write() now authorize MAC using active_cred rather than td->td_ucred - soo_read/write() unchanged - vn_read/write() now authorize MAC using active_cred but VOP_READ/WRITE() with fp->f_cred Modify vn_rdwr() to accept two credential arguments instead of a single credential: active_cred and file_cred. Use active_cred for MAC authorization, and select a credential for use in VOP_READ/WRITE() based on whether file_cred is NULL or not. If file_cred is provided, authorize the VOP using that cred, otherwise the active credential, matching current semantics. Modify current vn_rdwr() consumers to pass a file_cred if used in the context of a struct file, and to always pass active_cred. When vn_rdwr() is used without a file_cred, pass NOCRED. These changes should maintain current semantics for read/write, but avoid a redundant passing of fp->f_cred, as well as making it more clear what the origin of each credential is in file descriptor read/write operations. Follow-up commits will make similar changes to other file descriptor operations, and modify the MAC framework to pass both credentials to MAC policy modules so they can implement either semantic for revocation. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-15 20:55:08 +00:00
struct ucred *active_cred;
struct ucred *file_cred;
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
int *aresid;
KSE Milestone 2 Note ALL MODULES MUST BE RECOMPILED make the kernel aware that there are smaller units of scheduling than the process. (but only allow one thread per process at this time). This is functionally equivalent to teh previousl -current except that there is a thread associated with each process. Sorry john! (your next MFC will be a doosie!) Reviewed by: peter@freebsd.org, dillon@freebsd.org X-MFC after: ha ha ha ha
2001-09-12 08:38:13 +00:00
struct thread *td;
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
{
struct uio auio;
struct iovec aiov;
Add snapshots to the fast filesystem. Most of the changes support the gating of system calls that cause modifications to the underlying filesystem. The gating can be enabled by any filesystem that needs to consistently suspend operations by adding the vop_stdgetwritemount to their set of vnops. Once gating is enabled, the function vfs_write_suspend stops all new write operations to a filesystem, allows any filesystem modifying system calls already in progress to complete, then sync's the filesystem to disk and returns. The function vfs_write_resume allows the suspended write operations to begin again. Gating is not added by default for all filesystems as for SMP systems it adds two extra locks to such critical kernel paths as the write system call. Thus, gating should only be added as needed. Details on the use and current status of snapshots in FFS can be found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness is not included here. Unless and until you create a snapshot file, these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
struct mount *mp;
In order to better support flexible and extensible access control, make a series of modifications to the credential arguments relating to file read and write operations to cliarfy which credential is used for what: - Change fo_read() and fo_write() to accept "active_cred" instead of "cred", and change the semantics of consumers of fo_read() and fo_write() to pass the active credential of the thread requesting an operation rather than the cached file cred. The cached file cred is still available in fo_read() and fo_write() consumers via fp->f_cred. These changes largely in sys_generic.c. For each implementation of fo_read() and fo_write(), update cred usage to reflect this change and maintain current semantics: - badfo_readwrite() unchanged - kqueue_read/write() unchanged pipe_read/write() now authorize MAC using active_cred rather than td->td_ucred - soo_read/write() unchanged - vn_read/write() now authorize MAC using active_cred but VOP_READ/WRITE() with fp->f_cred Modify vn_rdwr() to accept two credential arguments instead of a single credential: active_cred and file_cred. Use active_cred for MAC authorization, and select a credential for use in VOP_READ/WRITE() based on whether file_cred is NULL or not. If file_cred is provided, authorize the VOP using that cred, otherwise the active credential, matching current semantics. Modify current vn_rdwr() consumers to pass a file_cred if used in the context of a struct file, and to always pass active_cred. When vn_rdwr() is used without a file_cred, pass NOCRED. These changes should maintain current semantics for read/write, but avoid a redundant passing of fp->f_cred, as well as making it more clear what the origin of each credential is in file descriptor read/write operations. Follow-up commits will make similar changes to other file descriptor operations, and modify the MAC framework to pass both credentials to MAC policy modules so they can implement either semantic for revocation. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-15 20:55:08 +00:00
struct ucred *cred;
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
int error;
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
VFS_ASSERT_GIANT(vp->v_mount);
Flag a broad range of VFS operations as GIANT_REQUIRED in order to catch leaking into VFS without Giant. Inch Giant a little lower in several file descriptor operations on vnodes to cover only VFS operations that need it, rather than file flag reading, etc.
2004-08-06 22:25:35 +00:00
Add snapshots to the fast filesystem. Most of the changes support the gating of system calls that cause modifications to the underlying filesystem. The gating can be enabled by any filesystem that needs to consistently suspend operations by adding the vop_stdgetwritemount to their set of vnops. Once gating is enabled, the function vfs_write_suspend stops all new write operations to a filesystem, allows any filesystem modifying system calls already in progress to complete, then sync's the filesystem to disk and returns. The function vfs_write_resume allows the suspended write operations to begin again. Gating is not added by default for all filesystems as for SMP systems it adds two extra locks to such critical kernel paths as the write system call. Thus, gating should only be added as needed. Details on the use and current status of snapshots in FFS can be found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness is not included here. Unless and until you create a snapshot file, these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
if ((ioflg & IO_NODELOCKED) == 0) {
mp = NULL;
Clean up vn_rdwr locking. - Do shared locks on read. - Only do vn_{start,finished}_write when writing.
2002-06-28 17:51:11 +00:00
if (rw == UIO_WRITE) {
if (vp->v_type != VCHR &&
(error = vn_start_write(vp, &mp, V_WAIT | PCATCH))
!= 0)
return (error);
vn_lock(vp, LK_EXCLUSIVE | LK_RETRY, td);
} else {
- Return two shared locks to exclusive locks. This was premature. - Document the problems that prevent us from using shared locks.
2002-08-22 07:26:18 +00:00
/*
* XXX This should be LK_SHARED but I don't trust VFS
* enough to leave it like that until it has been
* reviewed further.
*/
vn_lock(vp, LK_EXCLUSIVE | LK_RETRY, td);
Clean up vn_rdwr locking. - Do shared locks on read. - Only do vn_{start,finished}_write when writing.
2002-06-28 17:51:11 +00:00
}
Add snapshots to the fast filesystem. Most of the changes support the gating of system calls that cause modifications to the underlying filesystem. The gating can be enabled by any filesystem that needs to consistently suspend operations by adding the vop_stdgetwritemount to their set of vnops. Once gating is enabled, the function vfs_write_suspend stops all new write operations to a filesystem, allows any filesystem modifying system calls already in progress to complete, then sync's the filesystem to disk and returns. The function vfs_write_resume allows the suspended write operations to begin again. Gating is not added by default for all filesystems as for SMP systems it adds two extra locks to such critical kernel paths as the write system call. Thus, gating should only be added as needed. Details on the use and current status of snapshots in FFS can be found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness is not included here. Unless and until you create a snapshot file, these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
}
Assert that the vnode is locked. This is meant to catch bugs or mis-use of the vnode API in conditions where IO_NODELOCKED has been used without the vnode actually being locked.
2005-04-05 01:11:43 +00:00
ASSERT_VOP_LOCKED(vp, "IO_NODELOCKED with no vp lock held");
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
auio.uio_iov = &aiov;
auio.uio_iovcnt = 1;
aiov.iov_base = base;
aiov.iov_len = len;
auio.uio_resid = len;
auio.uio_offset = offset;
auio.uio_segflg = segflg;
auio.uio_rw = rw;
KSE Milestone 2 Note ALL MODULES MUST BE RECOMPILED make the kernel aware that there are smaller units of scheduling than the process. (but only allow one thread per process at this time). This is functionally equivalent to teh previousl -current except that there is a thread associated with each process. Sorry john! (your next MFC will be a doosie!) Reviewed by: peter@freebsd.org, dillon@freebsd.org X-MFC after: ha ha ha ha
2001-09-12 08:38:13 +00:00
auio.uio_td = td;
Implement IO_NOMACCHECK in vn_rdwr() -- perform MAC checks (assuming 'options MAC') as long as IO_NOMACCHECK is not set in the IO flags. If IO_NOMACCHECK is set, bypass MAC checks in vn_rdwr(). This allows vn_rdwr() to be used as a utility function inside of file systems where MAC checks have already been performed, or where the operation is being done on behalf of the kernel not the user. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI LAbs
2002-08-12 16:15:34 +00:00
error = 0;
#ifdef MAC
if ((ioflg & IO_NOMACCHECK) == 0) {
if (rw == UIO_READ)
Pass active_cred and file_cred into the MAC framework explicitly for mac_check_vnode_{poll,read,stat,write}(). Pass in fp->f_cred when calling these checks with a struct file available. Otherwise, pass NOCRED. All currently MAC policies use active_cred, but could now offer the cached credential semantic used for the base system security model. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 19:04:53 +00:00
error = mac_check_vnode_read(active_cred, file_cred,
vp);
Implement IO_NOMACCHECK in vn_rdwr() -- perform MAC checks (assuming 'options MAC') as long as IO_NOMACCHECK is not set in the IO flags. If IO_NOMACCHECK is set, bypass MAC checks in vn_rdwr(). This allows vn_rdwr() to be used as a utility function inside of file systems where MAC checks have already been performed, or where the operation is being done on behalf of the kernel not the user. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI LAbs
2002-08-12 16:15:34 +00:00
else
Pass active_cred and file_cred into the MAC framework explicitly for mac_check_vnode_{poll,read,stat,write}(). Pass in fp->f_cred when calling these checks with a struct file available. Otherwise, pass NOCRED. All currently MAC policies use active_cred, but could now offer the cached credential semantic used for the base system security model. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 19:04:53 +00:00
error = mac_check_vnode_write(active_cred, file_cred,
vp);
Implement IO_NOMACCHECK in vn_rdwr() -- perform MAC checks (assuming 'options MAC') as long as IO_NOMACCHECK is not set in the IO flags. If IO_NOMACCHECK is set, bypass MAC checks in vn_rdwr(). This allows vn_rdwr() to be used as a utility function inside of file systems where MAC checks have already been performed, or where the operation is being done on behalf of the kernel not the user. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI LAbs
2002-08-12 16:15:34 +00:00
}
#endif
if (error == 0) {
In order to better support flexible and extensible access control, make a series of modifications to the credential arguments relating to file read and write operations to cliarfy which credential is used for what: - Change fo_read() and fo_write() to accept "active_cred" instead of "cred", and change the semantics of consumers of fo_read() and fo_write() to pass the active credential of the thread requesting an operation rather than the cached file cred. The cached file cred is still available in fo_read() and fo_write() consumers via fp->f_cred. These changes largely in sys_generic.c. For each implementation of fo_read() and fo_write(), update cred usage to reflect this change and maintain current semantics: - badfo_readwrite() unchanged - kqueue_read/write() unchanged pipe_read/write() now authorize MAC using active_cred rather than td->td_ucred - soo_read/write() unchanged - vn_read/write() now authorize MAC using active_cred but VOP_READ/WRITE() with fp->f_cred Modify vn_rdwr() to accept two credential arguments instead of a single credential: active_cred and file_cred. Use active_cred for MAC authorization, and select a credential for use in VOP_READ/WRITE() based on whether file_cred is NULL or not. If file_cred is provided, authorize the VOP using that cred, otherwise the active credential, matching current semantics. Modify current vn_rdwr() consumers to pass a file_cred if used in the context of a struct file, and to always pass active_cred. When vn_rdwr() is used without a file_cred, pass NOCRED. These changes should maintain current semantics for read/write, but avoid a redundant passing of fp->f_cred, as well as making it more clear what the origin of each credential is in file descriptor read/write operations. Follow-up commits will make similar changes to other file descriptor operations, and modify the MAC framework to pass both credentials to MAC policy modules so they can implement either semantic for revocation. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-15 20:55:08 +00:00
if (file_cred)
cred = file_cred;
else
cred = active_cred;
Implement IO_NOMACCHECK in vn_rdwr() -- perform MAC checks (assuming 'options MAC') as long as IO_NOMACCHECK is not set in the IO flags. If IO_NOMACCHECK is set, bypass MAC checks in vn_rdwr(). This allows vn_rdwr() to be used as a utility function inside of file systems where MAC checks have already been performed, or where the operation is being done on behalf of the kernel not the user. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI LAbs
2002-08-12 16:15:34 +00:00
if (rw == UIO_READ)
error = VOP_READ(vp, &auio, ioflg, cred);
else
error = VOP_WRITE(vp, &auio, ioflg, cred);
}
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
if (aresid)
*aresid = auio.uio_resid;
else
if (auio.uio_resid && error == 0)
error = EIO;
Add snapshots to the fast filesystem. Most of the changes support the gating of system calls that cause modifications to the underlying filesystem. The gating can be enabled by any filesystem that needs to consistently suspend operations by adding the vop_stdgetwritemount to their set of vnops. Once gating is enabled, the function vfs_write_suspend stops all new write operations to a filesystem, allows any filesystem modifying system calls already in progress to complete, then sync's the filesystem to disk and returns. The function vfs_write_resume allows the suspended write operations to begin again. Gating is not added by default for all filesystems as for SMP systems it adds two extra locks to such critical kernel paths as the write system call. Thus, gating should only be added as needed. Details on the use and current status of snapshots in FFS can be found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness is not included here. Unless and until you create a snapshot file, these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
if ((ioflg & IO_NODELOCKED) == 0) {
vn_start_write() is called only when v_type != VCHR, so corresponding vn_finished_write() should also be called only then. BTW. I fixed two functions here: vn_rdwr() and vn_write(). The latter seems to be unused. MFC after: 3 weeks
2006-04-28 21:54:05 +00:00
if (rw == UIO_WRITE && vp->v_type != VCHR)
Clean up vn_rdwr locking. - Do shared locks on read. - Only do vn_{start,finished}_write when writing.
2002-06-28 17:51:11 +00:00
vn_finished_write(mp);
KSE Milestone 2 Note ALL MODULES MUST BE RECOMPILED make the kernel aware that there are smaller units of scheduling than the process. (but only allow one thread per process at this time). This is functionally equivalent to teh previousl -current except that there is a thread associated with each process. Sorry john! (your next MFC will be a doosie!) Reviewed by: peter@freebsd.org, dillon@freebsd.org X-MFC after: ha ha ha ha
2001-09-12 08:38:13 +00:00
VOP_UNLOCK(vp, 0, td);
Add snapshots to the fast filesystem. Most of the changes support the gating of system calls that cause modifications to the underlying filesystem. The gating can be enabled by any filesystem that needs to consistently suspend operations by adding the vop_stdgetwritemount to their set of vnops. Once gating is enabled, the function vfs_write_suspend stops all new write operations to a filesystem, allows any filesystem modifying system calls already in progress to complete, then sync's the filesystem to disk and returns. The function vfs_write_resume allows the suspended write operations to begin again. Gating is not added by default for all filesystems as for SMP systems it adds two extra locks to such critical kernel paths as the write system call. Thus, gating should only be added as needed. Details on the use and current status of snapshots in FFS can be found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness is not included here. Unless and until you create a snapshot file, these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
}
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
return (error);
}
This brings in a Yahoo coredump patch from Paul, with additional mods by me (addition of vn_rdwr_inchunks). The problem Yahoo is solving is that if you have large process images core dumping, or you have a large number of forked processes all core dumping at the same time, the original coredump code would leave the vnode locked throughout. This can cause the directory vnode to get locked up, which can cause the parent directory vnode to get locked up, and so on all the way to the root node, locking the entire machine up for extremely long periods of time. This patch solves the problem in two ways. First it uses an advisory non-blocking lock to abort multiple processes trying to core to the same file. Second (my contribution) it chunks up the writes and uses bwillwrite() to avoid holding the vnode locked while blocking in the buffer cache. Submitted by: ps Reviewed by: dillon MFC after: 2 weeks
2001-09-08 20:02:33 +00:00
/*
* Package up an I/O request on a vnode into a uio and do it. The I/O
* request is split up into smaller chunks and we try to avoid saturating
* the buffer cache while potentially holding a vnode locked, so we
Make uio_yield() a global. Call uio_yield() between chunks in vn_rdwr_inchunks(), allowing other processes to gain an exclusive lock on the vnode. Specifically: directory scanning, to avoid a race to the root directory, and multiple child processes coring simultaniously so they can figure out that some other core'ing child has an exclusive adv lock and just exit instead. This completely fixes performance problems when large programs core. You can have hundreds of copies (forked children) of the same binary core all at once and not notice. MFC after: 3 days
2001-09-26 06:54:32 +00:00
* check bwillwrite() before calling vn_rdwr(). We also call uio_yield()
* to give other processes a chance to lock the vnode (either other processes
* core'ing the same binary, or unrelated processes scanning the directory).
This brings in a Yahoo coredump patch from Paul, with additional mods by me (addition of vn_rdwr_inchunks). The problem Yahoo is solving is that if you have large process images core dumping, or you have a large number of forked processes all core dumping at the same time, the original coredump code would leave the vnode locked throughout. This can cause the directory vnode to get locked up, which can cause the parent directory vnode to get locked up, and so on all the way to the root node, locking the entire machine up for extremely long periods of time. This patch solves the problem in two ways. First it uses an advisory non-blocking lock to abort multiple processes trying to core to the same file. Second (my contribution) it chunks up the writes and uses bwillwrite() to avoid holding the vnode locked while blocking in the buffer cache. Submitted by: ps Reviewed by: dillon MFC after: 2 weeks
2001-09-08 20:02:33 +00:00
*/
int
In order to better support flexible and extensible access control, make a series of modifications to the credential arguments relating to file read and write operations to cliarfy which credential is used for what: - Change fo_read() and fo_write() to accept "active_cred" instead of "cred", and change the semantics of consumers of fo_read() and fo_write() to pass the active credential of the thread requesting an operation rather than the cached file cred. The cached file cred is still available in fo_read() and fo_write() consumers via fp->f_cred. These changes largely in sys_generic.c. For each implementation of fo_read() and fo_write(), update cred usage to reflect this change and maintain current semantics: - badfo_readwrite() unchanged - kqueue_read/write() unchanged pipe_read/write() now authorize MAC using active_cred rather than td->td_ucred - soo_read/write() unchanged - vn_read/write() now authorize MAC using active_cred but VOP_READ/WRITE() with fp->f_cred Modify vn_rdwr() to accept two credential arguments instead of a single credential: active_cred and file_cred. Use active_cred for MAC authorization, and select a credential for use in VOP_READ/WRITE() based on whether file_cred is NULL or not. If file_cred is provided, authorize the VOP using that cred, otherwise the active credential, matching current semantics. Modify current vn_rdwr() consumers to pass a file_cred if used in the context of a struct file, and to always pass active_cred. When vn_rdwr() is used without a file_cred, pass NOCRED. These changes should maintain current semantics for read/write, but avoid a redundant passing of fp->f_cred, as well as making it more clear what the origin of each credential is in file descriptor read/write operations. Follow-up commits will make similar changes to other file descriptor operations, and modify the MAC framework to pass both credentials to MAC policy modules so they can implement either semantic for revocation. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-15 20:55:08 +00:00
vn_rdwr_inchunks(rw, vp, base, len, offset, segflg, ioflg, active_cred,
file_cred, aresid, td)
This brings in a Yahoo coredump patch from Paul, with additional mods by me (addition of vn_rdwr_inchunks). The problem Yahoo is solving is that if you have large process images core dumping, or you have a large number of forked processes all core dumping at the same time, the original coredump code would leave the vnode locked throughout. This can cause the directory vnode to get locked up, which can cause the parent directory vnode to get locked up, and so on all the way to the root node, locking the entire machine up for extremely long periods of time. This patch solves the problem in two ways. First it uses an advisory non-blocking lock to abort multiple processes trying to core to the same file. Second (my contribution) it chunks up the writes and uses bwillwrite() to avoid holding the vnode locked while blocking in the buffer cache. Submitted by: ps Reviewed by: dillon MFC after: 2 weeks
2001-09-08 20:02:33 +00:00
enum uio_rw rw;
struct vnode *vp;
Eradicate caddr_t from the VFS API.
2005-12-14 00:49:52 +00:00
void *base;
Change the types of vn_rdwr_inchunks()'s len and aresid arguments to size_t and size_t *, respectively. Update callers for the new interface. This is a better fix for overflows that occurred when dumping segments larger than 2GB to core files.
2004-06-05 02:18:28 +00:00
size_t len;
This brings in a Yahoo coredump patch from Paul, with additional mods by me (addition of vn_rdwr_inchunks). The problem Yahoo is solving is that if you have large process images core dumping, or you have a large number of forked processes all core dumping at the same time, the original coredump code would leave the vnode locked throughout. This can cause the directory vnode to get locked up, which can cause the parent directory vnode to get locked up, and so on all the way to the root node, locking the entire machine up for extremely long periods of time. This patch solves the problem in two ways. First it uses an advisory non-blocking lock to abort multiple processes trying to core to the same file. Second (my contribution) it chunks up the writes and uses bwillwrite() to avoid holding the vnode locked while blocking in the buffer cache. Submitted by: ps Reviewed by: dillon MFC after: 2 weeks
2001-09-08 20:02:33 +00:00
off_t offset;
enum uio_seg segflg;
int ioflg;
In order to better support flexible and extensible access control, make a series of modifications to the credential arguments relating to file read and write operations to cliarfy which credential is used for what: - Change fo_read() and fo_write() to accept "active_cred" instead of "cred", and change the semantics of consumers of fo_read() and fo_write() to pass the active credential of the thread requesting an operation rather than the cached file cred. The cached file cred is still available in fo_read() and fo_write() consumers via fp->f_cred. These changes largely in sys_generic.c. For each implementation of fo_read() and fo_write(), update cred usage to reflect this change and maintain current semantics: - badfo_readwrite() unchanged - kqueue_read/write() unchanged pipe_read/write() now authorize MAC using active_cred rather than td->td_ucred - soo_read/write() unchanged - vn_read/write() now authorize MAC using active_cred but VOP_READ/WRITE() with fp->f_cred Modify vn_rdwr() to accept two credential arguments instead of a single credential: active_cred and file_cred. Use active_cred for MAC authorization, and select a credential for use in VOP_READ/WRITE() based on whether file_cred is NULL or not. If file_cred is provided, authorize the VOP using that cred, otherwise the active credential, matching current semantics. Modify current vn_rdwr() consumers to pass a file_cred if used in the context of a struct file, and to always pass active_cred. When vn_rdwr() is used without a file_cred, pass NOCRED. These changes should maintain current semantics for read/write, but avoid a redundant passing of fp->f_cred, as well as making it more clear what the origin of each credential is in file descriptor read/write operations. Follow-up commits will make similar changes to other file descriptor operations, and modify the MAC framework to pass both credentials to MAC policy modules so they can implement either semantic for revocation. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-15 20:55:08 +00:00
struct ucred *active_cred;
struct ucred *file_cred;
Change the types of vn_rdwr_inchunks()'s len and aresid arguments to size_t and size_t *, respectively. Update callers for the new interface. This is a better fix for overflows that occurred when dumping segments larger than 2GB to core files.
2004-06-05 02:18:28 +00:00
size_t *aresid;
KSE Milestone 2 Note ALL MODULES MUST BE RECOMPILED make the kernel aware that there are smaller units of scheduling than the process. (but only allow one thread per process at this time). This is functionally equivalent to teh previousl -current except that there is a thread associated with each process. Sorry john! (your next MFC will be a doosie!) Reviewed by: peter@freebsd.org, dillon@freebsd.org X-MFC after: ha ha ha ha
2001-09-12 08:38:13 +00:00
struct thread *td;
This brings in a Yahoo coredump patch from Paul, with additional mods by me (addition of vn_rdwr_inchunks). The problem Yahoo is solving is that if you have large process images core dumping, or you have a large number of forked processes all core dumping at the same time, the original coredump code would leave the vnode locked throughout. This can cause the directory vnode to get locked up, which can cause the parent directory vnode to get locked up, and so on all the way to the root node, locking the entire machine up for extremely long periods of time. This patch solves the problem in two ways. First it uses an advisory non-blocking lock to abort multiple processes trying to core to the same file. Second (my contribution) it chunks up the writes and uses bwillwrite() to avoid holding the vnode locked while blocking in the buffer cache. Submitted by: ps Reviewed by: dillon MFC after: 2 weeks
2001-09-08 20:02:33 +00:00
{
int error = 0;
Change the types of vn_rdwr_inchunks()'s len and aresid arguments to size_t and size_t *, respectively. Update callers for the new interface. This is a better fix for overflows that occurred when dumping segments larger than 2GB to core files.
2004-06-05 02:18:28 +00:00
int iaresid;
This brings in a Yahoo coredump patch from Paul, with additional mods by me (addition of vn_rdwr_inchunks). The problem Yahoo is solving is that if you have large process images core dumping, or you have a large number of forked processes all core dumping at the same time, the original coredump code would leave the vnode locked throughout. This can cause the directory vnode to get locked up, which can cause the parent directory vnode to get locked up, and so on all the way to the root node, locking the entire machine up for extremely long periods of time. This patch solves the problem in two ways. First it uses an advisory non-blocking lock to abort multiple processes trying to core to the same file. Second (my contribution) it chunks up the writes and uses bwillwrite() to avoid holding the vnode locked while blocking in the buffer cache. Submitted by: ps Reviewed by: dillon MFC after: 2 weeks
2001-09-08 20:02:33 +00:00
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
VFS_ASSERT_GIANT(vp->v_mount);
Flag a broad range of VFS operations as GIANT_REQUIRED in order to catch leaking into VFS without Giant. Inch Giant a little lower in several file descriptor operations on vnodes to cover only VFS operations that need it, rather than file flag reading, etc.
2004-08-06 22:25:35 +00:00
This brings in a Yahoo coredump patch from Paul, with additional mods by me (addition of vn_rdwr_inchunks). The problem Yahoo is solving is that if you have large process images core dumping, or you have a large number of forked processes all core dumping at the same time, the original coredump code would leave the vnode locked throughout. This can cause the directory vnode to get locked up, which can cause the parent directory vnode to get locked up, and so on all the way to the root node, locking the entire machine up for extremely long periods of time. This patch solves the problem in two ways. First it uses an advisory non-blocking lock to abort multiple processes trying to core to the same file. Second (my contribution) it chunks up the writes and uses bwillwrite() to avoid holding the vnode locked while blocking in the buffer cache. Submitted by: ps Reviewed by: dillon MFC after: 2 weeks
2001-09-08 20:02:33 +00:00
do {
Align the offset in vn_rdwr_inchunks() so that at most the first and the last chunk are misaligned relative to a MAXBSIZE byte boundary. vn_rdwr_inchunks() is used mainly for elf core dumps, and elf sections are usually perfectly misaligned relative to MAXBSIZE, and chunking prevents the file system from doing much realigning. This gives a surprisingly large speedup for core dumps -- from 50 to 13 seconds for a 512MB core dump here. The pessimization was mostly from an interaction of the misalignment with IO_DIRECT. It increased the number of i/o's for each chunk by a factor of 5 (3 writes and 2 read-before-writes instead of 1 write).
2004-03-13 02:56:27 +00:00
int chunk;
This brings in a Yahoo coredump patch from Paul, with additional mods by me (addition of vn_rdwr_inchunks). The problem Yahoo is solving is that if you have large process images core dumping, or you have a large number of forked processes all core dumping at the same time, the original coredump code would leave the vnode locked throughout. This can cause the directory vnode to get locked up, which can cause the parent directory vnode to get locked up, and so on all the way to the root node, locking the entire machine up for extremely long periods of time. This patch solves the problem in two ways. First it uses an advisory non-blocking lock to abort multiple processes trying to core to the same file. Second (my contribution) it chunks up the writes and uses bwillwrite() to avoid holding the vnode locked while blocking in the buffer cache. Submitted by: ps Reviewed by: dillon MFC after: 2 weeks
2001-09-08 20:02:33 +00:00
Align the offset in vn_rdwr_inchunks() so that at most the first and the last chunk are misaligned relative to a MAXBSIZE byte boundary. vn_rdwr_inchunks() is used mainly for elf core dumps, and elf sections are usually perfectly misaligned relative to MAXBSIZE, and chunking prevents the file system from doing much realigning. This gives a surprisingly large speedup for core dumps -- from 50 to 13 seconds for a 512MB core dump here. The pessimization was mostly from an interaction of the misalignment with IO_DIRECT. It increased the number of i/o's for each chunk by a factor of 5 (3 writes and 2 read-before-writes instead of 1 write).
2004-03-13 02:56:27 +00:00
/*
* Force `offset' to a multiple of MAXBSIZE except possibly
* for the first chunk, so that filesystems only need to
* write full blocks except possibly for the first and last
* chunks.
*/
chunk = MAXBSIZE - (uoff_t)offset % MAXBSIZE;
if (chunk > len)
chunk = len;
This brings in a Yahoo coredump patch from Paul, with additional mods by me (addition of vn_rdwr_inchunks). The problem Yahoo is solving is that if you have large process images core dumping, or you have a large number of forked processes all core dumping at the same time, the original coredump code would leave the vnode locked throughout. This can cause the directory vnode to get locked up, which can cause the parent directory vnode to get locked up, and so on all the way to the root node, locking the entire machine up for extremely long periods of time. This patch solves the problem in two ways. First it uses an advisory non-blocking lock to abort multiple processes trying to core to the same file. Second (my contribution) it chunks up the writes and uses bwillwrite() to avoid holding the vnode locked while blocking in the buffer cache. Submitted by: ps Reviewed by: dillon MFC after: 2 weeks
2001-09-08 20:02:33 +00:00
if (rw != UIO_READ && vp->v_type == VREG)
bwillwrite();
Change the types of vn_rdwr_inchunks()'s len and aresid arguments to size_t and size_t *, respectively. Update callers for the new interface. This is a better fix for overflows that occurred when dumping segments larger than 2GB to core files.
2004-06-05 02:18:28 +00:00
iaresid = 0;
This brings in a Yahoo coredump patch from Paul, with additional mods by me (addition of vn_rdwr_inchunks). The problem Yahoo is solving is that if you have large process images core dumping, or you have a large number of forked processes all core dumping at the same time, the original coredump code would leave the vnode locked throughout. This can cause the directory vnode to get locked up, which can cause the parent directory vnode to get locked up, and so on all the way to the root node, locking the entire machine up for extremely long periods of time. This patch solves the problem in two ways. First it uses an advisory non-blocking lock to abort multiple processes trying to core to the same file. Second (my contribution) it chunks up the writes and uses bwillwrite() to avoid holding the vnode locked while blocking in the buffer cache. Submitted by: ps Reviewed by: dillon MFC after: 2 weeks
2001-09-08 20:02:33 +00:00
error = vn_rdwr(rw, vp, base, chunk, offset, segflg,
Change the types of vn_rdwr_inchunks()'s len and aresid arguments to size_t and size_t *, respectively. Update callers for the new interface. This is a better fix for overflows that occurred when dumping segments larger than 2GB to core files.
2004-06-05 02:18:28 +00:00
ioflg, active_cred, file_cred, &iaresid, td);
This brings in a Yahoo coredump patch from Paul, with additional mods by me (addition of vn_rdwr_inchunks). The problem Yahoo is solving is that if you have large process images core dumping, or you have a large number of forked processes all core dumping at the same time, the original coredump code would leave the vnode locked throughout. This can cause the directory vnode to get locked up, which can cause the parent directory vnode to get locked up, and so on all the way to the root node, locking the entire machine up for extremely long periods of time. This patch solves the problem in two ways. First it uses an advisory non-blocking lock to abort multiple processes trying to core to the same file. Second (my contribution) it chunks up the writes and uses bwillwrite() to avoid holding the vnode locked while blocking in the buffer cache. Submitted by: ps Reviewed by: dillon MFC after: 2 weeks
2001-09-08 20:02:33 +00:00
len -= chunk; /* aresid calc already includes length */
if (error)
break;
offset += chunk;
Eradicate caddr_t from the VFS API.
2005-12-14 00:49:52 +00:00
base = (char *)base + chunk;
Make uio_yield() a global. Call uio_yield() between chunks in vn_rdwr_inchunks(), allowing other processes to gain an exclusive lock on the vnode. Specifically: directory scanning, to avoid a race to the root directory, and multiple child processes coring simultaniously so they can figure out that some other core'ing child has an exclusive adv lock and just exit instead. This completely fixes performance problems when large programs core. You can have hundreds of copies (forked children) of the same binary core all at once and not notice. MFC after: 3 days
2001-09-26 06:54:32 +00:00
uio_yield();
This brings in a Yahoo coredump patch from Paul, with additional mods by me (addition of vn_rdwr_inchunks). The problem Yahoo is solving is that if you have large process images core dumping, or you have a large number of forked processes all core dumping at the same time, the original coredump code would leave the vnode locked throughout. This can cause the directory vnode to get locked up, which can cause the parent directory vnode to get locked up, and so on all the way to the root node, locking the entire machine up for extremely long periods of time. This patch solves the problem in two ways. First it uses an advisory non-blocking lock to abort multiple processes trying to core to the same file. Second (my contribution) it chunks up the writes and uses bwillwrite() to avoid holding the vnode locked while blocking in the buffer cache. Submitted by: ps Reviewed by: dillon MFC after: 2 weeks
2001-09-08 20:02:33 +00:00
} while (len);
if (aresid)
Change the types of vn_rdwr_inchunks()'s len and aresid arguments to size_t and size_t *, respectively. Update callers for the new interface. This is a better fix for overflows that occurred when dumping segments larger than 2GB to core files.
2004-06-05 02:18:28 +00:00
*aresid = len + iaresid;
This brings in a Yahoo coredump patch from Paul, with additional mods by me (addition of vn_rdwr_inchunks). The problem Yahoo is solving is that if you have large process images core dumping, or you have a large number of forked processes all core dumping at the same time, the original coredump code would leave the vnode locked throughout. This can cause the directory vnode to get locked up, which can cause the parent directory vnode to get locked up, and so on all the way to the root node, locking the entire machine up for extremely long periods of time. This patch solves the problem in two ways. First it uses an advisory non-blocking lock to abort multiple processes trying to core to the same file. Second (my contribution) it chunks up the writes and uses bwillwrite() to avoid holding the vnode locked while blocking in the buffer cache. Submitted by: ps Reviewed by: dillon MFC after: 2 weeks
2001-09-08 20:02:33 +00:00
return (error);
}
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
/*
* File table vnode read routine.
*/
Staticize. Unstaticize a function in scsi/scsi_base that was used, with an undocumented option. My last count on the LINT kernel shows: Total symbols: 3647 unref symbols: 463 undef symbols: 4 1 ref symbols: 1751 2 ref symbols: 485 Approaching the pain threshold now.
1995-12-17 21:23:44 +00:00
static int
In order to better support flexible and extensible access control, make a series of modifications to the credential arguments relating to file read and write operations to cliarfy which credential is used for what: - Change fo_read() and fo_write() to accept "active_cred" instead of "cred", and change the semantics of consumers of fo_read() and fo_write() to pass the active credential of the thread requesting an operation rather than the cached file cred. The cached file cred is still available in fo_read() and fo_write() consumers via fp->f_cred. These changes largely in sys_generic.c. For each implementation of fo_read() and fo_write(), update cred usage to reflect this change and maintain current semantics: - badfo_readwrite() unchanged - kqueue_read/write() unchanged pipe_read/write() now authorize MAC using active_cred rather than td->td_ucred - soo_read/write() unchanged - vn_read/write() now authorize MAC using active_cred but VOP_READ/WRITE() with fp->f_cred Modify vn_rdwr() to accept two credential arguments instead of a single credential: active_cred and file_cred. Use active_cred for MAC authorization, and select a credential for use in VOP_READ/WRITE() based on whether file_cred is NULL or not. If file_cred is provided, authorize the VOP using that cred, otherwise the active credential, matching current semantics. Modify current vn_rdwr() consumers to pass a file_cred if used in the context of a struct file, and to always pass active_cred. When vn_rdwr() is used without a file_cred, pass NOCRED. These changes should maintain current semantics for read/write, but avoid a redundant passing of fp->f_cred, as well as making it more clear what the origin of each credential is in file descriptor read/write operations. Follow-up commits will make similar changes to other file descriptor operations, and modify the MAC framework to pass both credentials to MAC policy modules so they can implement either semantic for revocation. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-15 20:55:08 +00:00
vn_read(fp, uio, active_cred, flags, td)
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
struct file *fp;
struct uio *uio;
In order to better support flexible and extensible access control, make a series of modifications to the credential arguments relating to file read and write operations to cliarfy which credential is used for what: - Change fo_read() and fo_write() to accept "active_cred" instead of "cred", and change the semantics of consumers of fo_read() and fo_write() to pass the active credential of the thread requesting an operation rather than the cached file cred. The cached file cred is still available in fo_read() and fo_write() consumers via fp->f_cred. These changes largely in sys_generic.c. For each implementation of fo_read() and fo_write(), update cred usage to reflect this change and maintain current semantics: - badfo_readwrite() unchanged - kqueue_read/write() unchanged pipe_read/write() now authorize MAC using active_cred rather than td->td_ucred - soo_read/write() unchanged - vn_read/write() now authorize MAC using active_cred but VOP_READ/WRITE() with fp->f_cred Modify vn_rdwr() to accept two credential arguments instead of a single credential: active_cred and file_cred. Use active_cred for MAC authorization, and select a credential for use in VOP_READ/WRITE() based on whether file_cred is NULL or not. If file_cred is provided, authorize the VOP using that cred, otherwise the active credential, matching current semantics. Modify current vn_rdwr() consumers to pass a file_cred if used in the context of a struct file, and to always pass active_cred. When vn_rdwr() is used without a file_cred, pass NOCRED. These changes should maintain current semantics for read/write, but avoid a redundant passing of fp->f_cred, as well as making it more clear what the origin of each credential is in file descriptor read/write operations. Follow-up commits will make similar changes to other file descriptor operations, and modify the MAC framework to pass both credentials to MAC policy modules so they can implement either semantic for revocation. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-15 20:55:08 +00:00
struct ucred *active_cred;
KSE Milestone 2 Note ALL MODULES MUST BE RECOMPILED make the kernel aware that there are smaller units of scheduling than the process. (but only allow one thread per process at this time). This is functionally equivalent to teh previousl -current except that there is a thread associated with each process. Sorry john! (your next MFC will be a doosie!) Reviewed by: peter@freebsd.org, dillon@freebsd.org X-MFC after: ha ha ha ha
2001-09-12 08:38:13 +00:00
struct thread *td;
Add standard padding argument to pread and pwrite syscall. That should make them NetBSD compatible. Add parameter to fo_read and fo_write. (The only flag FOF_OFFSET mean that the offset is set in the struct uio). Factor out some common code from read/pread/write/pwrite syscalls.
1999-04-04 21:41:28 +00:00
int flags;
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
{
This is what was "fdfix2.patch," a fix for fd sharing. It's pretty far-reaching in fd-land, so you'll want to consult the code for changes. The biggest change is that now, you don't use fp->f_ops->fo_foo(fp, bar) but instead fo_foo(fp, bar), which increments and decrements the fp refcount upon entry and exit. Two new calls, fhold() and fdrop(), are provided. Each does what it seems like it should, and if fdrop() brings the refcount to zero, the fd is freed as well. Thanks to peter ("to hell with it, it looks ok to me.") for his review. Thanks to msmith for keeping me from putting locks everywhere :) Reviewed by: peter
1999-09-19 17:00:25 +00:00
struct vnode *vp;
Address several problems in vn_read and vn_write: 1. Make read-ahead work for pread and aio_read. 2. Fix one place where a comparison of uio_offset with -1 wasn't updated to use FOF_OFFSET. 3. Honor O_APPEND in the FOF_OFFSET case. In addition, use the variable name "ioflag" in both vn_read and vn_write to avoid possible confusion between the variable "flag" and the parameter "flags". Submitted by: Bruce Evans <bde@zeta.org.au> and me
1999-04-21 05:56:45 +00:00
int error, ioflag;
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
int vfslocked;
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
KSE Milestone 2 Note ALL MODULES MUST BE RECOMPILED make the kernel aware that there are smaller units of scheduling than the process. (but only allow one thread per process at this time). This is functionally equivalent to teh previousl -current except that there is a thread associated with each process. Sorry john! (your next MFC will be a doosie!) Reviewed by: peter@freebsd.org, dillon@freebsd.org X-MFC after: ha ha ha ha
2001-09-12 08:38:13 +00:00
KASSERT(uio->uio_td == td, ("uio_td %p is not td %p",
uio->uio_td, td));
Add a f_vnode field to struct file. Several of the subtypes have an associated vnode which is used for stuff like the f*() functions. By giving the vnode a speparate field, a number of checks for the specific subtype can be replaced simply with a check for f_vnode != NULL, and we can later free f_data up to subtype specific use. At this point in time, f_data still points to the vnode, so any code I might have overlooked will still work.
2003-06-22 08:41:43 +00:00
vp = fp->f_vnode;
Address several problems in vn_read and vn_write: 1. Make read-ahead work for pread and aio_read. 2. Fix one place where a comparison of uio_offset with -1 wasn't updated to use FOF_OFFSET. 3. Honor O_APPEND in the FOF_OFFSET case. In addition, use the variable name "ioflag" in both vn_read and vn_write to avoid possible confusion between the variable "flag" and the parameter "flags". Submitted by: Bruce Evans <bde@zeta.org.au> and me
1999-04-21 05:56:45 +00:00
ioflag = 0;
if (fp->f_flag & FNONBLOCK)
ioflag |= IO_NDELAY;
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
if (fp->f_flag & O_DIRECT)
ioflag |= IO_DIRECT;
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
vfslocked = VFS_LOCK_GIANT(vp->v_mount);
In order to better support flexible and extensible access control, make a series of modifications to the credential arguments relating to file read and write operations to cliarfy which credential is used for what: - Change fo_read() and fo_write() to accept "active_cred" instead of "cred", and change the semantics of consumers of fo_read() and fo_write() to pass the active credential of the thread requesting an operation rather than the cached file cred. The cached file cred is still available in fo_read() and fo_write() consumers via fp->f_cred. These changes largely in sys_generic.c. For each implementation of fo_read() and fo_write(), update cred usage to reflect this change and maintain current semantics: - badfo_readwrite() unchanged - kqueue_read/write() unchanged pipe_read/write() now authorize MAC using active_cred rather than td->td_ucred - soo_read/write() unchanged - vn_read/write() now authorize MAC using active_cred but VOP_READ/WRITE() with fp->f_cred Modify vn_rdwr() to accept two credential arguments instead of a single credential: active_cred and file_cred. Use active_cred for MAC authorization, and select a credential for use in VOP_READ/WRITE() based on whether file_cred is NULL or not. If file_cred is provided, authorize the VOP using that cred, otherwise the active credential, matching current semantics. Modify current vn_rdwr() consumers to pass a file_cred if used in the context of a struct file, and to always pass active_cred. When vn_rdwr() is used without a file_cred, pass NOCRED. These changes should maintain current semantics for read/write, but avoid a redundant passing of fp->f_cred, as well as making it more clear what the origin of each credential is in file descriptor read/write operations. Follow-up commits will make similar changes to other file descriptor operations, and modify the MAC framework to pass both credentials to MAC policy modules so they can implement either semantic for revocation. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-15 20:55:08 +00:00
VOP_LEASE(vp, td, fp->f_cred, LEASE_READ);
- Return two shared locks to exclusive locks. This was premature. - Document the problems that prevent us from using shared locks.
2002-08-22 07:26:18 +00:00
/*
Allow concurrent read(2)/readv(2) access to a file. Lock file offset against multiple read calls. Submitted by: ups Obtained from: Yahoo! MFC after: 2 weeks
2006-05-16 07:50:54 +00:00
* According to McKusick the vn lock was protecting f_offset here.
* It is now protected by the FOFFSET_LOCKED flag.
- Return two shared locks to exclusive locks. This was premature. - Document the problems that prevent us from using shared locks.
2002-08-22 07:26:18 +00:00
*/
fp->f_offset doesn't need any protection when it isn't accessed.
2003-03-26 19:21:12 +00:00
if ((flags & FOF_OFFSET) == 0) {
Allow concurrent read(2)/readv(2) access to a file. Lock file offset against multiple read calls. Submitted by: ups Obtained from: Yahoo! MFC after: 2 weeks
2006-05-16 07:50:54 +00:00
FILE_LOCK(fp);
while(fp->f_vnread_flags & FOFFSET_LOCKED) {
fp->f_vnread_flags |= FOFFSET_LOCK_WAITING;
msleep(&fp->f_vnread_flags,fp->f_mtxp,PUSER -1,"vnread offlock",0);
}
fp->f_vnread_flags |= FOFFSET_LOCKED;
FILE_UNLOCK(fp);
vn_lock(vp, LK_SHARED | LK_RETRY, td);
Address several problems in vn_read and vn_write: 1. Make read-ahead work for pread and aio_read. 2. Fix one place where a comparison of uio_offset with -1 wasn't updated to use FOF_OFFSET. 3. Honor O_APPEND in the FOF_OFFSET case. In addition, use the variable name "ioflag" in both vn_read and vn_write to avoid possible confusion between the variable "flag" and the parameter "flags". Submitted by: Bruce Evans <bde@zeta.org.au> and me
1999-04-21 05:56:45 +00:00
uio->uio_offset = fp->f_offset;
fp->f_offset doesn't need any protection when it isn't accessed.
2003-03-26 19:21:12 +00:00
} else
- LK_NOPAUSE is a nop now. Sponsored by: Isilon Systems, Inc.
2005-03-31 04:37:09 +00:00
vn_lock(vp, LK_SHARED | LK_RETRY, td);
Changed vn_read/write such that fp->f_offset isn't touched if uio->uio_offset != -1. This fixes a problem with aio_read/write and permits a straightforward implementation of pread/pwrite. PR: kern/8669 Submitted by: John Plevyak <jplevyak@inktomi.com> Reviewed by: Matthew Dillon <dillon@apollo.backplane.com>
1999-03-26 20:25:21 +00:00
Change the write-behind code to take more care when starting async I/O's. The sequential read heuristic has been extended to cover writes as well. We continue to call cluster_write() normally, thus blocks in the file will still be reallocated for large (but still random) I/O's, but I/O will only be initiated for truely sequential writes. This solves a number of annoying situations, especially with DBM (hash method) writes, and also has the side effect of fixing a number of (stupid) benchmarks. Reviewed-by: mckusick
2000-04-02 00:55:28 +00:00
ioflag |= sequential_heuristic(uio, fp);
Introduce support for Mandatory Access Control and extensible kernel access control Invoke appropriate MAC framework entry points to authorize a number of vnode operations, including read, write, stat, poll. This permits MAC policies to revoke access to files following label changes, and to limit information spread about the file to user processes. Note: currently the file cached credential is used for some of these authorization check. We will need to expand some of the MAC entry point APIs to permit multiple creds to be passed to the access control check to allow diverse policy behavior. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-01 17:23:22 +00:00
#ifdef MAC
Pass active_cred and file_cred into the MAC framework explicitly for mac_check_vnode_{poll,read,stat,write}(). Pass in fp->f_cred when calling these checks with a struct file available. Otherwise, pass NOCRED. All currently MAC policies use active_cred, but could now offer the cached credential semantic used for the base system security model. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 19:04:53 +00:00
error = mac_check_vnode_read(active_cred, fp->f_cred, vp);
Introduce support for Mandatory Access Control and extensible kernel access control Invoke appropriate MAC framework entry points to authorize a number of vnode operations, including read, write, stat, poll. This permits MAC policies to revoke access to files following label changes, and to limit information spread about the file to user processes. Note: currently the file cached credential is used for some of these authorization check. We will need to expand some of the MAC entry point APIs to permit multiple creds to be passed to the access control check to allow diverse policy behavior. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-01 17:23:22 +00:00
if (error == 0)
#endif
In order to better support flexible and extensible access control, make a series of modifications to the credential arguments relating to file read and write operations to cliarfy which credential is used for what: - Change fo_read() and fo_write() to accept "active_cred" instead of "cred", and change the semantics of consumers of fo_read() and fo_write() to pass the active credential of the thread requesting an operation rather than the cached file cred. The cached file cred is still available in fo_read() and fo_write() consumers via fp->f_cred. These changes largely in sys_generic.c. For each implementation of fo_read() and fo_write(), update cred usage to reflect this change and maintain current semantics: - badfo_readwrite() unchanged - kqueue_read/write() unchanged pipe_read/write() now authorize MAC using active_cred rather than td->td_ucred - soo_read/write() unchanged - vn_read/write() now authorize MAC using active_cred but VOP_READ/WRITE() with fp->f_cred Modify vn_rdwr() to accept two credential arguments instead of a single credential: active_cred and file_cred. Use active_cred for MAC authorization, and select a credential for use in VOP_READ/WRITE() based on whether file_cred is NULL or not. If file_cred is provided, authorize the VOP using that cred, otherwise the active credential, matching current semantics. Modify current vn_rdwr() consumers to pass a file_cred if used in the context of a struct file, and to always pass active_cred. When vn_rdwr() is used without a file_cred, pass NOCRED. These changes should maintain current semantics for read/write, but avoid a redundant passing of fp->f_cred, as well as making it more clear what the origin of each credential is in file descriptor read/write operations. Follow-up commits will make similar changes to other file descriptor operations, and modify the MAC framework to pass both credentials to MAC policy modules so they can implement either semantic for revocation. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-15 20:55:08 +00:00
error = VOP_READ(vp, uio, ioflag, fp->f_cred);
Allow concurrent read(2)/readv(2) access to a file. Lock file offset against multiple read calls. Submitted by: ups Obtained from: Yahoo! MFC after: 2 weeks
2006-05-16 07:50:54 +00:00
if ((flags & FOF_OFFSET) == 0) {
Address several problems in vn_read and vn_write: 1. Make read-ahead work for pread and aio_read. 2. Fix one place where a comparison of uio_offset with -1 wasn't updated to use FOF_OFFSET. 3. Honor O_APPEND in the FOF_OFFSET case. In addition, use the variable name "ioflag" in both vn_read and vn_write to avoid possible confusion between the variable "flag" and the parameter "flags". Submitted by: Bruce Evans <bde@zeta.org.au> and me
1999-04-21 05:56:45 +00:00
fp->f_offset = uio->uio_offset;
Allow concurrent read(2)/readv(2) access to a file. Lock file offset against multiple read calls. Submitted by: ups Obtained from: Yahoo! MFC after: 2 weeks
2006-05-16 07:50:54 +00:00
FILE_LOCK(fp);
if (fp->f_vnread_flags & FOFFSET_LOCK_WAITING)
wakeup(&fp->f_vnread_flags);
fp->f_vnread_flags = 0;
FILE_UNLOCK(fp);
}
Change the write-behind code to take more care when starting async I/O's. The sequential read heuristic has been extended to cover writes as well. We continue to call cluster_write() normally, thus blocks in the file will still be reallocated for large (but still random) I/O's, but I/O will only be initiated for truely sequential writes. This solves a number of annoying situations, especially with DBM (hash method) writes, and also has the side effect of fixing a number of (stupid) benchmarks. Reviewed-by: mckusick
2000-04-02 00:55:28 +00:00
fp->f_nextoff = uio->uio_offset;
KSE Milestone 2 Note ALL MODULES MUST BE RECOMPILED make the kernel aware that there are smaller units of scheduling than the process. (but only allow one thread per process at this time). This is functionally equivalent to teh previousl -current except that there is a thread associated with each process. Sorry john! (your next MFC will be a doosie!) Reviewed by: peter@freebsd.org, dillon@freebsd.org X-MFC after: ha ha ha ha
2001-09-12 08:38:13 +00:00
VOP_UNLOCK(vp, 0, td);
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
VFS_UNLOCK_GIANT(vfslocked);
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
return (error);
}
/*
* File table vnode write routine.
*/
Staticize. Unstaticize a function in scsi/scsi_base that was used, with an undocumented option. My last count on the LINT kernel shows: Total symbols: 3647 unref symbols: 463 undef symbols: 4 1 ref symbols: 1751 2 ref symbols: 485 Approaching the pain threshold now.
1995-12-17 21:23:44 +00:00
static int
In order to better support flexible and extensible access control, make a series of modifications to the credential arguments relating to file read and write operations to cliarfy which credential is used for what: - Change fo_read() and fo_write() to accept "active_cred" instead of "cred", and change the semantics of consumers of fo_read() and fo_write() to pass the active credential of the thread requesting an operation rather than the cached file cred. The cached file cred is still available in fo_read() and fo_write() consumers via fp->f_cred. These changes largely in sys_generic.c. For each implementation of fo_read() and fo_write(), update cred usage to reflect this change and maintain current semantics: - badfo_readwrite() unchanged - kqueue_read/write() unchanged pipe_read/write() now authorize MAC using active_cred rather than td->td_ucred - soo_read/write() unchanged - vn_read/write() now authorize MAC using active_cred but VOP_READ/WRITE() with fp->f_cred Modify vn_rdwr() to accept two credential arguments instead of a single credential: active_cred and file_cred. Use active_cred for MAC authorization, and select a credential for use in VOP_READ/WRITE() based on whether file_cred is NULL or not. If file_cred is provided, authorize the VOP using that cred, otherwise the active credential, matching current semantics. Modify current vn_rdwr() consumers to pass a file_cred if used in the context of a struct file, and to always pass active_cred. When vn_rdwr() is used without a file_cred, pass NOCRED. These changes should maintain current semantics for read/write, but avoid a redundant passing of fp->f_cred, as well as making it more clear what the origin of each credential is in file descriptor read/write operations. Follow-up commits will make similar changes to other file descriptor operations, and modify the MAC framework to pass both credentials to MAC policy modules so they can implement either semantic for revocation. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-15 20:55:08 +00:00
vn_write(fp, uio, active_cred, flags, td)
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
struct file *fp;
struct uio *uio;
In order to better support flexible and extensible access control, make a series of modifications to the credential arguments relating to file read and write operations to cliarfy which credential is used for what: - Change fo_read() and fo_write() to accept "active_cred" instead of "cred", and change the semantics of consumers of fo_read() and fo_write() to pass the active credential of the thread requesting an operation rather than the cached file cred. The cached file cred is still available in fo_read() and fo_write() consumers via fp->f_cred. These changes largely in sys_generic.c. For each implementation of fo_read() and fo_write(), update cred usage to reflect this change and maintain current semantics: - badfo_readwrite() unchanged - kqueue_read/write() unchanged pipe_read/write() now authorize MAC using active_cred rather than td->td_ucred - soo_read/write() unchanged - vn_read/write() now authorize MAC using active_cred but VOP_READ/WRITE() with fp->f_cred Modify vn_rdwr() to accept two credential arguments instead of a single credential: active_cred and file_cred. Use active_cred for MAC authorization, and select a credential for use in VOP_READ/WRITE() based on whether file_cred is NULL or not. If file_cred is provided, authorize the VOP using that cred, otherwise the active credential, matching current semantics. Modify current vn_rdwr() consumers to pass a file_cred if used in the context of a struct file, and to always pass active_cred. When vn_rdwr() is used without a file_cred, pass NOCRED. These changes should maintain current semantics for read/write, but avoid a redundant passing of fp->f_cred, as well as making it more clear what the origin of each credential is in file descriptor read/write operations. Follow-up commits will make similar changes to other file descriptor operations, and modify the MAC framework to pass both credentials to MAC policy modules so they can implement either semantic for revocation. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-15 20:55:08 +00:00
struct ucred *active_cred;
KSE Milestone 2 Note ALL MODULES MUST BE RECOMPILED make the kernel aware that there are smaller units of scheduling than the process. (but only allow one thread per process at this time). This is functionally equivalent to teh previousl -current except that there is a thread associated with each process. Sorry john! (your next MFC will be a doosie!) Reviewed by: peter@freebsd.org, dillon@freebsd.org X-MFC after: ha ha ha ha
2001-09-12 08:38:13 +00:00
struct thread *td;
Add standard padding argument to pread and pwrite syscall. That should make them NetBSD compatible. Add parameter to fo_read and fo_write. (The only flag FOF_OFFSET mean that the offset is set in the struct uio). Factor out some common code from read/pread/write/pwrite syscalls.
1999-04-04 21:41:28 +00:00
int flags;
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
{
These changes appear to give us benefits with both small (32MB) and large (1G) memory machine configurations. I was able to run 'dbench 32' on a 32MB system without bring the machine to a grinding halt. * buffer cache hash table now dynamically allocated. This will have no effect on memory consumption for smaller systems and will help scale the buffer cache for larger systems. * minor enhancement to pmap_clearbit(). I noticed that all the calls to it used constant arguments. Making it an inline allows the constants to propogate to deeper inlines and should produce better code. * removal of inherent vfs_ioopt support through the emplacement of appropriate #ifdef's, with John's permission. If we do not find a use for it by the end of the year we will remove it entirely. * removal of getnewbufloops* counters & sysctl's - no longer necessary for debugging, getnewbuf() is now optimal. * buffer hash table functions removed from sys/buf.h and localized to vfs_bio.c * VFS_BIO_NEED_DIRTYFLUSH flag and support code added ( bwillwrite() ), allowing processes to block when too many dirty buffers are present in the system. * removal of a softdep test in bdwrite() that is no longer necessary now that bdwrite() no longer attempts to flush dirty buffers. * slight optimization added to bqrelse() - there is no reason to test for available buffer space on B_DELWRI buffers. * addition of reverse-scanning code to vfs_bio_awrite(). vfs_bio_awrite() will attempt to locate clusterable areas in both the forward and reverse direction relative to the offset of the buffer passed to it. This will probably not make much of a difference now, but I believe we will start to rely on it heavily in the future if we decide to shift some of the burden of the clustering closer to the actual I/O initiation. * Removal of the newbufcnt and lastnewbuf counters that Kirk added. They do not fix any race conditions that haven't already been fixed by the gbincore() test done after the only call to getnewbuf(). getnewbuf() is a static, so there is no chance of it being misused by other modules. ( Unless Kirk can think of a specific thing that this code fixes. I went through it very carefully and didn't see anything ). * removal of VOP_ISLOCKED() check in flushbufqueues(). I do not think this check is necessary, the buffer should flush properly whether the vnode is locked or not. ( yes? ). * removal of extra arguments passed to getnewbuf() that are not necessary. * missed cluster_wbuild() that had to be a cluster_wbuild_wb() in vfs_cluster.c * vn_write() now calls bwillwrite() *PRIOR* to locking the vnode, which should greatly aid flushing operations in heavy load situations - both the pageout and update daemons will be able to operate more efficiently. * removal of b_usecount. We may add it back in later but for now it is useless. Prior implementations of the buffer cache never had enough buffers for it to be useful, and current implementations which make more buffers available might not benefit relative to the amount of sophistication required to implement a b_usecount. Straight LRU should work just as well, especially when most things are VMIO backed. I expect that (even though John will not like this assumption) directories will become VMIO backed some point soon. Submitted by: Matthew Dillon <dillon@backplane.com> Reviewed by: Kirk McKusick <mckusick@mckusick.com>
1999-07-08 06:06:00 +00:00
struct vnode *vp;
Add snapshots to the fast filesystem. Most of the changes support the gating of system calls that cause modifications to the underlying filesystem. The gating can be enabled by any filesystem that needs to consistently suspend operations by adding the vop_stdgetwritemount to their set of vnops. Once gating is enabled, the function vfs_write_suspend stops all new write operations to a filesystem, allows any filesystem modifying system calls already in progress to complete, then sync's the filesystem to disk and returns. The function vfs_write_resume allows the suspended write operations to begin again. Gating is not added by default for all filesystems as for SMP systems it adds two extra locks to such critical kernel paths as the write system call. Thus, gating should only be added as needed. Details on the use and current status of snapshots in FFS can be found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness is not included here. Unless and until you create a snapshot file, these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
struct mount *mp;
Address several problems in vn_read and vn_write: 1. Make read-ahead work for pread and aio_read. 2. Fix one place where a comparison of uio_offset with -1 wasn't updated to use FOF_OFFSET. 3. Honor O_APPEND in the FOF_OFFSET case. In addition, use the variable name "ioflag" in both vn_read and vn_write to avoid possible confusion between the variable "flag" and the parameter "flags". Submitted by: Bruce Evans <bde@zeta.org.au> and me
1999-04-21 05:56:45 +00:00
int error, ioflag;
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
int vfslocked;
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
KSE Milestone 2 Note ALL MODULES MUST BE RECOMPILED make the kernel aware that there are smaller units of scheduling than the process. (but only allow one thread per process at this time). This is functionally equivalent to teh previousl -current except that there is a thread associated with each process. Sorry john! (your next MFC will be a doosie!) Reviewed by: peter@freebsd.org, dillon@freebsd.org X-MFC after: ha ha ha ha
2001-09-12 08:38:13 +00:00
KASSERT(uio->uio_td == td, ("uio_td %p is not td %p",
uio->uio_td, td));
Add a f_vnode field to struct file. Several of the subtypes have an associated vnode which is used for stuff like the f*() functions. By giving the vnode a speparate field, a number of checks for the specific subtype can be replaced simply with a check for f_vnode != NULL, and we can later free f_data up to subtype specific use. At this point in time, f_data still points to the vnode, so any code I might have overlooked will still work.
2003-06-22 08:41:43 +00:00
vp = fp->f_vnode;
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
vfslocked = VFS_LOCK_GIANT(vp->v_mount);
These changes appear to give us benefits with both small (32MB) and large (1G) memory machine configurations. I was able to run 'dbench 32' on a 32MB system without bring the machine to a grinding halt. * buffer cache hash table now dynamically allocated. This will have no effect on memory consumption for smaller systems and will help scale the buffer cache for larger systems. * minor enhancement to pmap_clearbit(). I noticed that all the calls to it used constant arguments. Making it an inline allows the constants to propogate to deeper inlines and should produce better code. * removal of inherent vfs_ioopt support through the emplacement of appropriate #ifdef's, with John's permission. If we do not find a use for it by the end of the year we will remove it entirely. * removal of getnewbufloops* counters & sysctl's - no longer necessary for debugging, getnewbuf() is now optimal. * buffer hash table functions removed from sys/buf.h and localized to vfs_bio.c * VFS_BIO_NEED_DIRTYFLUSH flag and support code added ( bwillwrite() ), allowing processes to block when too many dirty buffers are present in the system. * removal of a softdep test in bdwrite() that is no longer necessary now that bdwrite() no longer attempts to flush dirty buffers. * slight optimization added to bqrelse() - there is no reason to test for available buffer space on B_DELWRI buffers. * addition of reverse-scanning code to vfs_bio_awrite(). vfs_bio_awrite() will attempt to locate clusterable areas in both the forward and reverse direction relative to the offset of the buffer passed to it. This will probably not make much of a difference now, but I believe we will start to rely on it heavily in the future if we decide to shift some of the burden of the clustering closer to the actual I/O initiation. * Removal of the newbufcnt and lastnewbuf counters that Kirk added. They do not fix any race conditions that haven't already been fixed by the gbincore() test done after the only call to getnewbuf(). getnewbuf() is a static, so there is no chance of it being misused by other modules. ( Unless Kirk can think of a specific thing that this code fixes. I went through it very carefully and didn't see anything ). * removal of VOP_ISLOCKED() check in flushbufqueues(). I do not think this check is necessary, the buffer should flush properly whether the vnode is locked or not. ( yes? ). * removal of extra arguments passed to getnewbuf() that are not necessary. * missed cluster_wbuild() that had to be a cluster_wbuild_wb() in vfs_cluster.c * vn_write() now calls bwillwrite() *PRIOR* to locking the vnode, which should greatly aid flushing operations in heavy load situations - both the pageout and update daemons will be able to operate more efficiently. * removal of b_usecount. We may add it back in later but for now it is useless. Prior implementations of the buffer cache never had enough buffers for it to be useful, and current implementations which make more buffers available might not benefit relative to the amount of sophistication required to implement a b_usecount. Straight LRU should work just as well, especially when most things are VMIO backed. I expect that (even though John will not like this assumption) directories will become VMIO backed some point soon. Submitted by: Matthew Dillon <dillon@backplane.com> Reviewed by: Kirk McKusick <mckusick@mckusick.com>
1999-07-08 06:06:00 +00:00
if (vp->v_type == VREG)
bwillwrite();
Address several problems in vn_read and vn_write: 1. Make read-ahead work for pread and aio_read. 2. Fix one place where a comparison of uio_offset with -1 wasn't updated to use FOF_OFFSET. 3. Honor O_APPEND in the FOF_OFFSET case. In addition, use the variable name "ioflag" in both vn_read and vn_write to avoid possible confusion between the variable "flag" and the parameter "flags". Submitted by: Bruce Evans <bde@zeta.org.au> and me
1999-04-21 05:56:45 +00:00
ioflag = IO_UNIT;
if (vp->v_type == VREG && (fp->f_flag & O_APPEND))
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
ioflag |= IO_APPEND;
if (fp->f_flag & FNONBLOCK)
ioflag |= IO_NDELAY;
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
if (fp->f_flag & O_DIRECT)
ioflag |= IO_DIRECT;
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
if ((fp->f_flag & O_FSYNC) ||
(vp->v_mount && (vp->v_mount->mnt_flag & MNT_SYNCHRONOUS)))
ioflag |= IO_SYNC;
Add snapshots to the fast filesystem. Most of the changes support the gating of system calls that cause modifications to the underlying filesystem. The gating can be enabled by any filesystem that needs to consistently suspend operations by adding the vop_stdgetwritemount to their set of vnops. Once gating is enabled, the function vfs_write_suspend stops all new write operations to a filesystem, allows any filesystem modifying system calls already in progress to complete, then sync's the filesystem to disk and returns. The function vfs_write_resume allows the suspended write operations to begin again. Gating is not added by default for all filesystems as for SMP systems it adds two extra locks to such critical kernel paths as the write system call. Thus, gating should only be added as needed. Details on the use and current status of snapshots in FFS can be found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness is not included here. Unless and until you create a snapshot file, these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
mp = NULL;
Take VBLK devices further out of their missery. This should fix the panic I introduced in my previous commit on this topic.
2000-11-02 21:14:13 +00:00
if (vp->v_type != VCHR &&
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
(error = vn_start_write(vp, &mp, V_WAIT | PCATCH)) != 0)
goto unlock;
In order to better support flexible and extensible access control, make a series of modifications to the credential arguments relating to file read and write operations to cliarfy which credential is used for what: - Change fo_read() and fo_write() to accept "active_cred" instead of "cred", and change the semantics of consumers of fo_read() and fo_write() to pass the active credential of the thread requesting an operation rather than the cached file cred. The cached file cred is still available in fo_read() and fo_write() consumers via fp->f_cred. These changes largely in sys_generic.c. For each implementation of fo_read() and fo_write(), update cred usage to reflect this change and maintain current semantics: - badfo_readwrite() unchanged - kqueue_read/write() unchanged pipe_read/write() now authorize MAC using active_cred rather than td->td_ucred - soo_read/write() unchanged - vn_read/write() now authorize MAC using active_cred but VOP_READ/WRITE() with fp->f_cred Modify vn_rdwr() to accept two credential arguments instead of a single credential: active_cred and file_cred. Use active_cred for MAC authorization, and select a credential for use in VOP_READ/WRITE() based on whether file_cred is NULL or not. If file_cred is provided, authorize the VOP using that cred, otherwise the active credential, matching current semantics. Modify current vn_rdwr() consumers to pass a file_cred if used in the context of a struct file, and to always pass active_cred. When vn_rdwr() is used without a file_cred, pass NOCRED. These changes should maintain current semantics for read/write, but avoid a redundant passing of fp->f_cred, as well as making it more clear what the origin of each credential is in file descriptor read/write operations. Follow-up commits will make similar changes to other file descriptor operations, and modify the MAC framework to pass both credentials to MAC policy modules so they can implement either semantic for revocation. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-15 20:55:08 +00:00
VOP_LEASE(vp, td, fp->f_cred, LEASE_WRITE);
KSE Milestone 2 Note ALL MODULES MUST BE RECOMPILED make the kernel aware that there are smaller units of scheduling than the process. (but only allow one thread per process at this time). This is functionally equivalent to teh previousl -current except that there is a thread associated with each process. Sorry john! (your next MFC will be a doosie!) Reviewed by: peter@freebsd.org, dillon@freebsd.org X-MFC after: ha ha ha ha
2001-09-12 08:38:13 +00:00
vn_lock(vp, LK_EXCLUSIVE | LK_RETRY, td);
Address several problems in vn_read and vn_write: 1. Make read-ahead work for pread and aio_read. 2. Fix one place where a comparison of uio_offset with -1 wasn't updated to use FOF_OFFSET. 3. Honor O_APPEND in the FOF_OFFSET case. In addition, use the variable name "ioflag" in both vn_read and vn_write to avoid possible confusion between the variable "flag" and the parameter "flags". Submitted by: Bruce Evans <bde@zeta.org.au> and me
1999-04-21 05:56:45 +00:00
if ((flags & FOF_OFFSET) == 0)
uio->uio_offset = fp->f_offset;
Change the write-behind code to take more care when starting async I/O's. The sequential read heuristic has been extended to cover writes as well. We continue to call cluster_write() normally, thus blocks in the file will still be reallocated for large (but still random) I/O's, but I/O will only be initiated for truely sequential writes. This solves a number of annoying situations, especially with DBM (hash method) writes, and also has the side effect of fixing a number of (stupid) benchmarks. Reviewed-by: mckusick
2000-04-02 00:55:28 +00:00
ioflag |= sequential_heuristic(uio, fp);
Introduce support for Mandatory Access Control and extensible kernel access control Invoke appropriate MAC framework entry points to authorize a number of vnode operations, including read, write, stat, poll. This permits MAC policies to revoke access to files following label changes, and to limit information spread about the file to user processes. Note: currently the file cached credential is used for some of these authorization check. We will need to expand some of the MAC entry point APIs to permit multiple creds to be passed to the access control check to allow diverse policy behavior. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-01 17:23:22 +00:00
#ifdef MAC
Pass active_cred and file_cred into the MAC framework explicitly for mac_check_vnode_{poll,read,stat,write}(). Pass in fp->f_cred when calling these checks with a struct file available. Otherwise, pass NOCRED. All currently MAC policies use active_cred, but could now offer the cached credential semantic used for the base system security model. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 19:04:53 +00:00
error = mac_check_vnode_write(active_cred, fp->f_cred, vp);
Introduce support for Mandatory Access Control and extensible kernel access control Invoke appropriate MAC framework entry points to authorize a number of vnode operations, including read, write, stat, poll. This permits MAC policies to revoke access to files following label changes, and to limit information spread about the file to user processes. Note: currently the file cached credential is used for some of these authorization check. We will need to expand some of the MAC entry point APIs to permit multiple creds to be passed to the access control check to allow diverse policy behavior. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-01 17:23:22 +00:00
if (error == 0)
#endif
In order to better support flexible and extensible access control, make a series of modifications to the credential arguments relating to file read and write operations to cliarfy which credential is used for what: - Change fo_read() and fo_write() to accept "active_cred" instead of "cred", and change the semantics of consumers of fo_read() and fo_write() to pass the active credential of the thread requesting an operation rather than the cached file cred. The cached file cred is still available in fo_read() and fo_write() consumers via fp->f_cred. These changes largely in sys_generic.c. For each implementation of fo_read() and fo_write(), update cred usage to reflect this change and maintain current semantics: - badfo_readwrite() unchanged - kqueue_read/write() unchanged pipe_read/write() now authorize MAC using active_cred rather than td->td_ucred - soo_read/write() unchanged - vn_read/write() now authorize MAC using active_cred but VOP_READ/WRITE() with fp->f_cred Modify vn_rdwr() to accept two credential arguments instead of a single credential: active_cred and file_cred. Use active_cred for MAC authorization, and select a credential for use in VOP_READ/WRITE() based on whether file_cred is NULL or not. If file_cred is provided, authorize the VOP using that cred, otherwise the active credential, matching current semantics. Modify current vn_rdwr() consumers to pass a file_cred if used in the context of a struct file, and to always pass active_cred. When vn_rdwr() is used without a file_cred, pass NOCRED. These changes should maintain current semantics for read/write, but avoid a redundant passing of fp->f_cred, as well as making it more clear what the origin of each credential is in file descriptor read/write operations. Follow-up commits will make similar changes to other file descriptor operations, and modify the MAC framework to pass both credentials to MAC policy modules so they can implement either semantic for revocation. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-15 20:55:08 +00:00
error = VOP_WRITE(vp, uio, ioflag, fp->f_cred);
Address several problems in vn_read and vn_write: 1. Make read-ahead work for pread and aio_read. 2. Fix one place where a comparison of uio_offset with -1 wasn't updated to use FOF_OFFSET. 3. Honor O_APPEND in the FOF_OFFSET case. In addition, use the variable name "ioflag" in both vn_read and vn_write to avoid possible confusion between the variable "flag" and the parameter "flags". Submitted by: Bruce Evans <bde@zeta.org.au> and me
1999-04-21 05:56:45 +00:00
if ((flags & FOF_OFFSET) == 0)
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
fp->f_offset = uio->uio_offset;
Change the write-behind code to take more care when starting async I/O's. The sequential read heuristic has been extended to cover writes as well. We continue to call cluster_write() normally, thus blocks in the file will still be reallocated for large (but still random) I/O's, but I/O will only be initiated for truely sequential writes. This solves a number of annoying situations, especially with DBM (hash method) writes, and also has the side effect of fixing a number of (stupid) benchmarks. Reviewed-by: mckusick
2000-04-02 00:55:28 +00:00
fp->f_nextoff = uio->uio_offset;
KSE Milestone 2 Note ALL MODULES MUST BE RECOMPILED make the kernel aware that there are smaller units of scheduling than the process. (but only allow one thread per process at this time). This is functionally equivalent to teh previousl -current except that there is a thread associated with each process. Sorry john! (your next MFC will be a doosie!) Reviewed by: peter@freebsd.org, dillon@freebsd.org X-MFC after: ha ha ha ha
2001-09-12 08:38:13 +00:00
VOP_UNLOCK(vp, 0, td);
vn_start_write() is called only when v_type != VCHR, so corresponding vn_finished_write() should also be called only then. BTW. I fixed two functions here: vn_rdwr() and vn_write(). The latter seems to be unused. MFC after: 3 weeks
2006-04-28 21:54:05 +00:00
if (vp->v_type != VCHR)
vn_finished_write(mp);
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
unlock:
VFS_UNLOCK_GIANT(vfslocked);
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
return (error);
}
/*
* File table vnode stat routine.
*/
Add a vnode fo_stat() entry point.
1999-11-08 03:32:15 +00:00
static int
Make similar changes to fo_stat() and fo_poll() as made earlier to fo_read() and fo_write(): explicitly use the cred argument to fo_poll() as "active_cred" using the passed file descriptor's f_cred reference to provide access to the file credential. Add an active_cred argument to fo_stat() so that implementers have access to the active credential as well as the file credential. Generally modify callers of fo_stat() to pass in td->td_ucred rather than fp->f_cred, which was redundantly provided via the fp argument. This set of modifications also permits threads to perform these operations on behalf of another thread without modifying their credential. Trickle this change down into fo_stat/poll() implementations: - badfo_poll(), badfo_stat(): modify/add arguments. - kqueue_poll(), kqueue_stat(): modify arguments. - pipe_poll(), pipe_stat(): modify/add arguments, pass active_cred to MAC checks rather than td->td_ucred. - soo_poll(), soo_stat(): modify/add arguments, pass fp->f_cred rather than cred to pru_sopoll() to maintain current semantics. - sopoll(): moidfy arguments. - vn_poll(), vn_statfile(): modify/add arguments, pass new arguments to vn_stat(). Pass active_cred to MAC and fp->f_cred to VOP_POLL() to maintian current semantics. - vn_close(): rename cred to file_cred to reflect reality while I'm here. - vn_stat(): Add active_cred and file_cred arguments to vn_stat() and consumers so that this distinction is maintained at the VFS as well as 'struct file' layer. Pass active_cred instead of td->td_ucred to MAC and to VOP_GETATTR() to maintain current semantics. - fifofs: modify the creation of a "filetemp" so that the file credential is properly initialized and can be used in the socket code if desired. Pass ap->a_td->td_ucred as the active credential to soo_poll(). If we teach the vnop interface about the distinction between file and active credentials, we would use the active credential here. Note that current inconsistent passing of active_cred vs. file_cred to VOP's is maintained. It's not clear why GETATTR would be authorized using active_cred while POLL would be authorized using file_cred at the file system level. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-16 12:52:03 +00:00
vn_statfile(fp, sb, active_cred, td)
Add a vnode fo_stat() entry point.
1999-11-08 03:32:15 +00:00
struct file *fp;
struct stat *sb;
Make similar changes to fo_stat() and fo_poll() as made earlier to fo_read() and fo_write(): explicitly use the cred argument to fo_poll() as "active_cred" using the passed file descriptor's f_cred reference to provide access to the file credential. Add an active_cred argument to fo_stat() so that implementers have access to the active credential as well as the file credential. Generally modify callers of fo_stat() to pass in td->td_ucred rather than fp->f_cred, which was redundantly provided via the fp argument. This set of modifications also permits threads to perform these operations on behalf of another thread without modifying their credential. Trickle this change down into fo_stat/poll() implementations: - badfo_poll(), badfo_stat(): modify/add arguments. - kqueue_poll(), kqueue_stat(): modify arguments. - pipe_poll(), pipe_stat(): modify/add arguments, pass active_cred to MAC checks rather than td->td_ucred. - soo_poll(), soo_stat(): modify/add arguments, pass fp->f_cred rather than cred to pru_sopoll() to maintain current semantics. - sopoll(): moidfy arguments. - vn_poll(), vn_statfile(): modify/add arguments, pass new arguments to vn_stat(). Pass active_cred to MAC and fp->f_cred to VOP_POLL() to maintian current semantics. - vn_close(): rename cred to file_cred to reflect reality while I'm here. - vn_stat(): Add active_cred and file_cred arguments to vn_stat() and consumers so that this distinction is maintained at the VFS as well as 'struct file' layer. Pass active_cred instead of td->td_ucred to MAC and to VOP_GETATTR() to maintain current semantics. - fifofs: modify the creation of a "filetemp" so that the file credential is properly initialized and can be used in the socket code if desired. Pass ap->a_td->td_ucred as the active credential to soo_poll(). If we teach the vnop interface about the distinction between file and active credentials, we would use the active credential here. Note that current inconsistent passing of active_cred vs. file_cred to VOP's is maintained. It's not clear why GETATTR would be authorized using active_cred while POLL would be authorized using file_cred at the file system level. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-16 12:52:03 +00:00
struct ucred *active_cred;
KSE Milestone 2 Note ALL MODULES MUST BE RECOMPILED make the kernel aware that there are smaller units of scheduling than the process. (but only allow one thread per process at this time). This is functionally equivalent to teh previousl -current except that there is a thread associated with each process. Sorry john! (your next MFC will be a doosie!) Reviewed by: peter@freebsd.org, dillon@freebsd.org X-MFC after: ha ha ha ha
2001-09-12 08:38:13 +00:00
struct thread *td;
Add a vnode fo_stat() entry point.
1999-11-08 03:32:15 +00:00
{
Add a f_vnode field to struct file. Several of the subtypes have an associated vnode which is used for stuff like the f*() functions. By giving the vnode a speparate field, a number of checks for the specific subtype can be replaced simply with a check for f_vnode != NULL, and we can later free f_data up to subtype specific use. At this point in time, f_data still points to the vnode, so any code I might have overlooked will still work.
2003-06-22 08:41:43 +00:00
struct vnode *vp = fp->f_vnode;
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
int vfslocked;
Make sure to hold vnode lock when calling into VOP_GETATTR(). Discussed with: mckusick, phk
2002-02-10 21:44:30 +00:00
int error;
Add a vnode fo_stat() entry point.
1999-11-08 03:32:15 +00:00
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
vfslocked = VFS_LOCK_GIANT(vp->v_mount);
Make sure to hold vnode lock when calling into VOP_GETATTR(). Discussed with: mckusick, phk
2002-02-10 21:44:30 +00:00
vn_lock(vp, LK_EXCLUSIVE | LK_RETRY, td);
Make similar changes to fo_stat() and fo_poll() as made earlier to fo_read() and fo_write(): explicitly use the cred argument to fo_poll() as "active_cred" using the passed file descriptor's f_cred reference to provide access to the file credential. Add an active_cred argument to fo_stat() so that implementers have access to the active credential as well as the file credential. Generally modify callers of fo_stat() to pass in td->td_ucred rather than fp->f_cred, which was redundantly provided via the fp argument. This set of modifications also permits threads to perform these operations on behalf of another thread without modifying their credential. Trickle this change down into fo_stat/poll() implementations: - badfo_poll(), badfo_stat(): modify/add arguments. - kqueue_poll(), kqueue_stat(): modify arguments. - pipe_poll(), pipe_stat(): modify/add arguments, pass active_cred to MAC checks rather than td->td_ucred. - soo_poll(), soo_stat(): modify/add arguments, pass fp->f_cred rather than cred to pru_sopoll() to maintain current semantics. - sopoll(): moidfy arguments. - vn_poll(), vn_statfile(): modify/add arguments, pass new arguments to vn_stat(). Pass active_cred to MAC and fp->f_cred to VOP_POLL() to maintian current semantics. - vn_close(): rename cred to file_cred to reflect reality while I'm here. - vn_stat(): Add active_cred and file_cred arguments to vn_stat() and consumers so that this distinction is maintained at the VFS as well as 'struct file' layer. Pass active_cred instead of td->td_ucred to MAC and to VOP_GETATTR() to maintain current semantics. - fifofs: modify the creation of a "filetemp" so that the file credential is properly initialized and can be used in the socket code if desired. Pass ap->a_td->td_ucred as the active credential to soo_poll(). If we teach the vnop interface about the distinction between file and active credentials, we would use the active credential here. Note that current inconsistent passing of active_cred vs. file_cred to VOP's is maintained. It's not clear why GETATTR would be authorized using active_cred while POLL would be authorized using file_cred at the file system level. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-16 12:52:03 +00:00
error = vn_stat(vp, sb, active_cred, fp->f_cred, td);
Make sure to hold vnode lock when calling into VOP_GETATTR(). Discussed with: mckusick, phk
2002-02-10 21:44:30 +00:00
VOP_UNLOCK(vp, 0, td);
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
VFS_UNLOCK_GIANT(vfslocked);
Make sure to hold vnode lock when calling into VOP_GETATTR(). Discussed with: mckusick, phk
2002-02-10 21:44:30 +00:00
return (error);
Add a vnode fo_stat() entry point.
1999-11-08 03:32:15 +00:00
}
Document all functions, global and static variables, and sysctls. Includes some minor whitespace changes, and re-ordering to be able to document properly (e.g, grouping of variables and the SYSCTL macro calls for them, where the documentation has been added.) Reviewed by: phk (but all errors are mine)
2002-03-05 15:38:49 +00:00
/*
* Stat a vnode; implementation for the stat syscall
*/
The big 4.4BSD Lite to FreeBSD 2.0.0 (Development) patch. Reviewed by: Rodney W. Grimes Submitted by: John Dyson and David Greenman
1994-05-25 09:21:21 +00:00
int
Make similar changes to fo_stat() and fo_poll() as made earlier to fo_read() and fo_write(): explicitly use the cred argument to fo_poll() as "active_cred" using the passed file descriptor's f_cred reference to provide access to the file credential. Add an active_cred argument to fo_stat() so that implementers have access to the active credential as well as the file credential. Generally modify callers of fo_stat() to pass in td->td_ucred rather than fp->f_cred, which was redundantly provided via the fp argument. This set of modifications also permits threads to perform these operations on behalf of another thread without modifying their credential. Trickle this change down into fo_stat/poll() implementations: - badfo_poll(), badfo_stat(): modify/add arguments. - kqueue_poll(), kqueue_stat(): modify arguments. - pipe_poll(), pipe_stat(): modify/add arguments, pass active_cred to MAC checks rather than td->td_ucred. - soo_poll(), soo_stat(): modify/add arguments, pass fp->f_cred rather than cred to pru_sopoll() to maintain current semantics. - sopoll(): moidfy arguments. - vn_poll(), vn_statfile(): modify/add arguments, pass new arguments to vn_stat(). Pass active_cred to MAC and fp->f_cred to VOP_POLL() to maintian current semantics. - vn_close(): rename cred to file_cred to reflect reality while I'm here. - vn_stat(): Add active_cred and file_cred arguments to vn_stat() and consumers so that this distinction is maintained at the VFS as well as 'struct file' layer. Pass active_cred instead of td->td_ucred to MAC and to VOP_GETATTR() to maintain current semantics. - fifofs: modify the creation of a "filetemp" so that the file credential is properly initialized and can be used in the socket code if desired. Pass ap->a_td->td_ucred as the active credential to soo_poll(). If we teach the vnop interface about the distinction between file and active credentials, we would use the active credential here. Note that current inconsistent passing of active_cred vs. file_cred to VOP's is maintained. It's not clear why GETATTR would be authorized using active_cred while POLL would be authorized using file_cred at the file system level. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-16 12:52:03 +00:00
vn_stat(vp, sb, active_cred, file_cred, td)
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
struct vnode *vp;
register struct stat *sb;
Make similar changes to fo_stat() and fo_poll() as made earlier to fo_read() and fo_write(): explicitly use the cred argument to fo_poll() as "active_cred" using the passed file descriptor's f_cred reference to provide access to the file credential. Add an active_cred argument to fo_stat() so that implementers have access to the active credential as well as the file credential. Generally modify callers of fo_stat() to pass in td->td_ucred rather than fp->f_cred, which was redundantly provided via the fp argument. This set of modifications also permits threads to perform these operations on behalf of another thread without modifying their credential. Trickle this change down into fo_stat/poll() implementations: - badfo_poll(), badfo_stat(): modify/add arguments. - kqueue_poll(), kqueue_stat(): modify arguments. - pipe_poll(), pipe_stat(): modify/add arguments, pass active_cred to MAC checks rather than td->td_ucred. - soo_poll(), soo_stat(): modify/add arguments, pass fp->f_cred rather than cred to pru_sopoll() to maintain current semantics. - sopoll(): moidfy arguments. - vn_poll(), vn_statfile(): modify/add arguments, pass new arguments to vn_stat(). Pass active_cred to MAC and fp->f_cred to VOP_POLL() to maintian current semantics. - vn_close(): rename cred to file_cred to reflect reality while I'm here. - vn_stat(): Add active_cred and file_cred arguments to vn_stat() and consumers so that this distinction is maintained at the VFS as well as 'struct file' layer. Pass active_cred instead of td->td_ucred to MAC and to VOP_GETATTR() to maintain current semantics. - fifofs: modify the creation of a "filetemp" so that the file credential is properly initialized and can be used in the socket code if desired. Pass ap->a_td->td_ucred as the active credential to soo_poll(). If we teach the vnop interface about the distinction between file and active credentials, we would use the active credential here. Note that current inconsistent passing of active_cred vs. file_cred to VOP's is maintained. It's not clear why GETATTR would be authorized using active_cred while POLL would be authorized using file_cred at the file system level. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-16 12:52:03 +00:00
struct ucred *active_cred;
struct ucred *file_cred;
KSE Milestone 2 Note ALL MODULES MUST BE RECOMPILED make the kernel aware that there are smaller units of scheduling than the process. (but only allow one thread per process at this time). This is functionally equivalent to teh previousl -current except that there is a thread associated with each process. Sorry john! (your next MFC will be a doosie!) Reviewed by: peter@freebsd.org, dillon@freebsd.org X-MFC after: ha ha ha ha
2001-09-12 08:38:13 +00:00
struct thread *td;
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
{
struct vattr vattr;
register struct vattr *vap;
int error;
u_short mode;
Introduce support for Mandatory Access Control and extensible kernel access control Invoke appropriate MAC framework entry points to authorize a number of vnode operations, including read, write, stat, poll. This permits MAC policies to revoke access to files following label changes, and to limit information spread about the file to user processes. Note: currently the file cached credential is used for some of these authorization check. We will need to expand some of the MAC entry point APIs to permit multiple creds to be passed to the access control check to allow diverse policy behavior. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-01 17:23:22 +00:00
#ifdef MAC
Pass active_cred and file_cred into the MAC framework explicitly for mac_check_vnode_{poll,read,stat,write}(). Pass in fp->f_cred when calling these checks with a struct file available. Otherwise, pass NOCRED. All currently MAC policies use active_cred, but could now offer the cached credential semantic used for the base system security model. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 19:04:53 +00:00
error = mac_check_vnode_stat(active_cred, file_cred, vp);
Introduce support for Mandatory Access Control and extensible kernel access control Invoke appropriate MAC framework entry points to authorize a number of vnode operations, including read, write, stat, poll. This permits MAC policies to revoke access to files following label changes, and to limit information spread about the file to user processes. Note: currently the file cached credential is used for some of these authorization check. We will need to expand some of the MAC entry point APIs to permit multiple creds to be passed to the access control check to allow diverse policy behavior. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-01 17:23:22 +00:00
if (error)
return (error);
#endif
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
vap = &vattr;
Make similar changes to fo_stat() and fo_poll() as made earlier to fo_read() and fo_write(): explicitly use the cred argument to fo_poll() as "active_cred" using the passed file descriptor's f_cred reference to provide access to the file credential. Add an active_cred argument to fo_stat() so that implementers have access to the active credential as well as the file credential. Generally modify callers of fo_stat() to pass in td->td_ucred rather than fp->f_cred, which was redundantly provided via the fp argument. This set of modifications also permits threads to perform these operations on behalf of another thread without modifying their credential. Trickle this change down into fo_stat/poll() implementations: - badfo_poll(), badfo_stat(): modify/add arguments. - kqueue_poll(), kqueue_stat(): modify arguments. - pipe_poll(), pipe_stat(): modify/add arguments, pass active_cred to MAC checks rather than td->td_ucred. - soo_poll(), soo_stat(): modify/add arguments, pass fp->f_cred rather than cred to pru_sopoll() to maintain current semantics. - sopoll(): moidfy arguments. - vn_poll(), vn_statfile(): modify/add arguments, pass new arguments to vn_stat(). Pass active_cred to MAC and fp->f_cred to VOP_POLL() to maintian current semantics. - vn_close(): rename cred to file_cred to reflect reality while I'm here. - vn_stat(): Add active_cred and file_cred arguments to vn_stat() and consumers so that this distinction is maintained at the VFS as well as 'struct file' layer. Pass active_cred instead of td->td_ucred to MAC and to VOP_GETATTR() to maintain current semantics. - fifofs: modify the creation of a "filetemp" so that the file credential is properly initialized and can be used in the socket code if desired. Pass ap->a_td->td_ucred as the active credential to soo_poll(). If we teach the vnop interface about the distinction between file and active credentials, we would use the active credential here. Note that current inconsistent passing of active_cred vs. file_cred to VOP's is maintained. It's not clear why GETATTR would be authorized using active_cred while POLL would be authorized using file_cred at the file system level. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-16 12:52:03 +00:00
error = VOP_GETATTR(vp, vap, active_cred, td);
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
if (error)
return (error);
Ensure that garbage from the kernel stack does not wind up being returned to user mode in the spare fields of the stat structure. PR: kern/14966 Reviewed by: dillon@freebsd.org Submitted by: Kelly Yancey kbyanc@posi.net
1999-11-18 08:14:20 +00:00
/*
* Zero the spare stat fields
*/
Use proper size in bzero of stat structure. Submitted by: Jake Burkholder <jake@locore.ca> Sponsored by: DARPA & NAI Labs.
2002-06-24 07:14:44 +00:00
bzero(sb, sizeof *sb);
Ensure that garbage from the kernel stack does not wind up being returned to user mode in the spare fields of the stat structure. PR: kern/14966 Reviewed by: dillon@freebsd.org Submitted by: Kelly Yancey kbyanc@posi.net
1999-11-18 08:14:20 +00:00
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
/*
* Copy from vattr table
*/
Make sure that stat(2) and friends always return a valid st_dev field. Pseudo-FS need not fill in the va_fsid anymore, the syscall code will use the first half of the fsid, which now looks like a udev_t with major 255.
1999-07-02 16:29:47 +00:00
if (vap->va_fsid != VNOVAL)
sb->st_dev = vap->va_fsid;
else
sb->st_dev = vp->v_mount->mnt_stat.f_fsid.val[0];
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
sb->st_ino = vap->va_fileid;
mode = vap->va_mode;
Report the mode as the result of the VOP_GETATTR rather than the vnodes type, they may not correspond.
1998-06-27 06:43:09 +00:00
switch (vap->va_type) {
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
case VREG:
mode |= S_IFREG;
break;
case VDIR:
mode |= S_IFDIR;
break;
case VBLK:
mode |= S_IFBLK;
break;
case VCHR:
mode |= S_IFCHR;
break;
case VLNK:
mode |= S_IFLNK;
Grammar police.
1998-04-10 00:09:04 +00:00
/* This is a cosmetic change, symlinks do not have a mode. */
New mount option nosymfollow. If enabled, the kernel lookup() function will not follow symbolic links on the mounted file system and return EACCES (Permission denied).
1998-04-08 18:31:59 +00:00
if (vp->v_mount->mnt_flag & MNT_NOSYMFOLLOW)
sb->st_mode &= ~ACCESSPERMS; /* 0000 */
else
sb->st_mode |= ACCESSPERMS; /* 0777 */
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
break;
case VSOCK:
mode |= S_IFSOCK;
break;
case VFIFO:
mode |= S_IFIFO;
break;
default:
return (EBADF);
};
sb->st_mode = mode;
sb->st_nlink = vap->va_nlink;
sb->st_uid = vap->va_uid;
sb->st_gid = vap->va_gid;
sb->st_rdev = vap->va_rdev;
vn_stat(): if va_size (u_quad_t) > OFF_MAX, return EOVERFLOW, don't copy it blindly to st_size
2001-08-23 17:56:48 +00:00
if (vap->va_size > OFF_MAX)
return (EOVERFLOW);
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
sb->st_size = vap->va_size;
sb->st_atimespec = vap->va_atime;
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
sb->st_mtimespec = vap->va_mtime;
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
sb->st_ctimespec = vap->va_ctime;
Change the name of st_createtime to st_birthtime. This change is made to reduce confusion between st_ctime and st_createtime. Submitted by: Eric Allman <eric@sendmail.org> Sponsored by: DARPA & NAI Labs.
2002-07-16 22:36:00 +00:00
sb->st_birthtimespec = vap->va_birthtime;
Changes to centralise the default blocksize behaviour. More likely to follow. Submitted by: phk@freebsd.org
1999-09-09 19:08:44 +00:00
/*
* According to www.opengroup.org, the meaning of st_blksize is
* "a filesystem-specific preferred I/O block size for this
* object. In some filesystem types, this may vary from file
* to file"
Make st_blksize default to PAGE_SIZE instead of zero.
2002-01-25 16:39:57 +00:00
* Default to PAGE_SIZE after much discussion.
Put the I/O block size in bufobj->bo_bsize. We keep si_bsize_phys around for now as that is the simplest way to pull the number out of disk device drivers in devfs_open(). The correct solution would be to do an ioctl(DIOCGSECTORSIZE), but the point is probably mooth when filesystems sit on GEOM, so don't bother for now.
2004-10-26 07:39:12 +00:00
* XXX: min(PAGE_SIZE, vp->v_bufobj.bo_bsize) may be more correct.
Move the special-casing of stat(2)->st_blksize for device files from UFS to the generic level. For chr/blk devices we don't care about the blocksize of the filesystem, we want what the device asked for.
1999-08-13 10:56:07 +00:00
*/
Changes to centralise the default blocksize behaviour. More likely to follow. Submitted by: phk@freebsd.org
1999-09-09 19:08:44 +00:00
Put the I/O block size in bufobj->bo_bsize. We keep si_bsize_phys around for now as that is the simplest way to pull the number out of disk device drivers in devfs_open(). The correct solution would be to do an ioctl(DIOCGSECTORSIZE), but the point is probably mooth when filesystems sit on GEOM, so don't bother for now.
2004-10-26 07:39:12 +00:00
sb->st_blksize = PAGE_SIZE;
Changes to centralise the default blocksize behaviour. More likely to follow. Submitted by: phk@freebsd.org
1999-09-09 19:08:44 +00:00
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
sb->st_flags = vap->va_flags;
Change the suser() API to take advantage of td_ucred as well as do a general cleanup of the API. The entire API now consists of two functions similar to the pre-KSE API. The suser() function takes a thread pointer as its only argument. The td_ucred member of this thread must be valid so the only valid thread pointers are curthread and a few kernel threads such as thread0. The suser_cred() function takes a pointer to a struct ucred as its first argument and an integer flag as its second argument. The flag is currently only used for the PRISON_ROOT flag. Discussed on: smp@
2002-04-01 21:31:13 +00:00
if (suser(td))
Fix style bugs and other bugs in the NFS fix.
1997-03-08 15:14:30 +00:00
sb->st_gen = 0;
else
sb->st_gen = vap->va_gen;
Fix (I hope) the NFS hole. This is only compile tested. Submitted by: (partly) davids@SECNET.COM via BUGTRAQ
1997-03-07 07:42:41 +00:00
A little hack to avoid a 64bit divide. Can go away if Gcc ever learns to optimise 64bit stuff...
1995-10-06 09:43:32 +00:00
#if (S_BLKSIZE == 512)
/* Optimize this case */
sb->st_blocks = vap->va_bytes >> 9;
#else
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
sb->st_blocks = vap->va_bytes / S_BLKSIZE;
A little hack to avoid a 64bit divide. Can go away if Gcc ever learns to optimise 64bit stuff...
1995-10-06 09:43:32 +00:00
#endif
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
return (0);
}
/*
* File table vnode ioctl routine.
*/
Staticize. Unstaticize a function in scsi/scsi_base that was used, with an undocumented option. My last count on the LINT kernel shows: Total symbols: 3647 unref symbols: 463 undef symbols: 4 1 ref symbols: 1751 2 ref symbols: 485 Approaching the pain threshold now.
1995-12-17 21:23:44 +00:00
static int
In continuation of early fileop credential changes, modify fo_ioctl() to accept an 'active_cred' argument reflecting the credential of the thread initiating the ioctl operation. - Change fo_ioctl() to accept active_cred; change consumers of the fo_ioctl() interface to generally pass active_cred from td->td_ucred. - In fifofs, initialize filetmp.f_cred to ap->a_cred so that the invocations of soo_ioctl() are provided access to the calling f_cred. Pass ap->a_td->td_ucred as the active_cred, but note that this is required because we don't yet distinguish file_cred and active_cred in invoking VOP's. - Update kqueue_ioctl() for its new argument. - Update pipe_ioctl() for its new argument, pass active_cred rather than td_ucred to MAC for authorization. - Update soo_ioctl() for its new argument. - Update vn_ioctl() for its new argument, use active_cred rather than td->td_ucred to authorize VOP_IOCTL() and the associated VOP_GETATTR(). Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-17 02:36:16 +00:00
vn_ioctl(fp, com, data, active_cred, td)
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
struct file *fp;
This commit fixes various 64bit portability problems required for FreeBSD/alpha. The most significant item is to change the command argument to ioctl functions from int to u_long. This change brings us inline with various other BSD versions. Driver writers may like to use (__FreeBSD_version == 300003) to detect this change. The prototype FreeBSD/alpha machdep will follow in a couple of days time.
1998-06-07 17:13:14 +00:00
u_long com;
More caddr_t removal, make fo_ioctl take a void * instead of a caddr_t.
2002-06-29 01:50:25 +00:00
void *data;
In continuation of early fileop credential changes, modify fo_ioctl() to accept an 'active_cred' argument reflecting the credential of the thread initiating the ioctl operation. - Change fo_ioctl() to accept active_cred; change consumers of the fo_ioctl() interface to generally pass active_cred from td->td_ucred. - In fifofs, initialize filetmp.f_cred to ap->a_cred so that the invocations of soo_ioctl() are provided access to the calling f_cred. Pass ap->a_td->td_ucred as the active_cred, but note that this is required because we don't yet distinguish file_cred and active_cred in invoking VOP's. - Update kqueue_ioctl() for its new argument. - Update pipe_ioctl() for its new argument, pass active_cred rather than td_ucred to MAC for authorization. - Update soo_ioctl() for its new argument. - Update vn_ioctl() for its new argument, use active_cred rather than td->td_ucred to authorize VOP_IOCTL() and the associated VOP_GETATTR(). Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-17 02:36:16 +00:00
struct ucred *active_cred;
KSE Milestone 2 Note ALL MODULES MUST BE RECOMPILED make the kernel aware that there are smaller units of scheduling than the process. (but only allow one thread per process at this time). This is functionally equivalent to teh previousl -current except that there is a thread associated with each process. Sorry john! (your next MFC will be a doosie!) Reviewed by: peter@freebsd.org, dillon@freebsd.org X-MFC after: ha ha ha ha
2001-09-12 08:38:13 +00:00
struct thread *td;
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
{
Add a f_vnode field to struct file. Several of the subtypes have an associated vnode which is used for stuff like the f*() functions. By giving the vnode a speparate field, a number of checks for the specific subtype can be replaced simply with a check for f_vnode != NULL, and we can later free f_data up to subtype specific use. At this point in time, f_data still points to the vnode, so any code I might have overlooked will still work.
2003-06-22 08:41:43 +00:00
struct vnode *vp = fp->f_vnode;
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
struct vattr vattr;
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
int vfslocked;
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
int error;
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
vfslocked = VFS_LOCK_GIANT(vp->v_mount);
Push Giant down through ioctl. Don't grab Giant in the upper syscall/wrapper code NET_LOCK_GIANT in the socket code (sockets/fifos). mtx_lock(&Giant) in the vnode code. mtx_lock(&Giant) in the opencrypto code. (This may actually not be needed, but better safe than sorry). Devfs grabs Giant if the driver is marked as needing Giant.
2004-11-17 09:09:55 +00:00
error = ENOTTY;
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
switch (vp->v_type) {
case VREG:
case VDIR:
if (com == FIONREAD) {
Make sure to hold vnode lock when calling into VOP_GETATTR(). Discussed with: mckusick, phk
2002-02-10 21:44:30 +00:00
vn_lock(vp, LK_EXCLUSIVE | LK_RETRY, td);
In continuation of early fileop credential changes, modify fo_ioctl() to accept an 'active_cred' argument reflecting the credential of the thread initiating the ioctl operation. - Change fo_ioctl() to accept active_cred; change consumers of the fo_ioctl() interface to generally pass active_cred from td->td_ucred. - In fifofs, initialize filetmp.f_cred to ap->a_cred so that the invocations of soo_ioctl() are provided access to the calling f_cred. Pass ap->a_td->td_ucred as the active_cred, but note that this is required because we don't yet distinguish file_cred and active_cred in invoking VOP's. - Update kqueue_ioctl() for its new argument. - Update pipe_ioctl() for its new argument, pass active_cred rather than td_ucred to MAC for authorization. - Update soo_ioctl() for its new argument. - Update vn_ioctl() for its new argument, use active_cred rather than td->td_ucred to authorize VOP_IOCTL() and the associated VOP_GETATTR(). Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-17 02:36:16 +00:00
error = VOP_GETATTR(vp, &vattr, active_cred, td);
Make sure to hold vnode lock when calling into VOP_GETATTR(). Discussed with: mckusick, phk
2002-02-10 21:44:30 +00:00
VOP_UNLOCK(vp, 0, td);
Push Giant down through ioctl. Don't grab Giant in the upper syscall/wrapper code NET_LOCK_GIANT in the socket code (sockets/fifos). mtx_lock(&Giant) in the vnode code. mtx_lock(&Giant) in the opencrypto code. (This may actually not be needed, but better safe than sorry). Devfs grabs Giant if the driver is marked as needing Giant.
2004-11-17 09:09:55 +00:00
if (!error)
*(int *)data = vattr.va_size - fp->f_offset;
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
}
if (com == FIONBIO || com == FIOASYNC) /* XXX */
Push Giant down through ioctl. Don't grab Giant in the upper syscall/wrapper code NET_LOCK_GIANT in the socket code (sockets/fifos). mtx_lock(&Giant) in the vnode code. mtx_lock(&Giant) in the opencrypto code. (This may actually not be needed, but better safe than sorry). Devfs grabs Giant if the driver is marked as needing Giant.
2004-11-17 09:09:55 +00:00
error = 0;
Ok, first blunder: ioctls are not entirely unused on vnodes anymore :-) Add dropped call to VOP_IOCTL().
2004-11-18 17:15:04 +00:00
else
error = VOP_IOCTL(vp, com, data, fp->f_flag,
active_cred, td);
Push Giant down through ioctl. Don't grab Giant in the upper syscall/wrapper code NET_LOCK_GIANT in the socket code (sockets/fifos). mtx_lock(&Giant) in the vnode code. mtx_lock(&Giant) in the opencrypto code. (This may actually not be needed, but better safe than sorry). Devfs grabs Giant if the driver is marked as needing Giant.
2004-11-17 09:09:55 +00:00
break;
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
default:
Push Giant down through ioctl. Don't grab Giant in the upper syscall/wrapper code NET_LOCK_GIANT in the socket code (sockets/fifos). mtx_lock(&Giant) in the vnode code. mtx_lock(&Giant) in the opencrypto code. (This may actually not be needed, but better safe than sorry). Devfs grabs Giant if the driver is marked as needing Giant.
2004-11-17 09:09:55 +00:00
break;
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
}
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
VFS_UNLOCK_GIANT(vfslocked);
Push Giant down through ioctl. Don't grab Giant in the upper syscall/wrapper code NET_LOCK_GIANT in the socket code (sockets/fifos). mtx_lock(&Giant) in the vnode code. mtx_lock(&Giant) in the opencrypto code. (This may actually not be needed, but better safe than sorry). Devfs grabs Giant if the driver is marked as needing Giant.
2004-11-17 09:09:55 +00:00
return (error);
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
}
/*
vn_select -> vn_poll
1997-09-14 02:51:16 +00:00
* File table vnode poll routine.
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
*/
Staticize. Unstaticize a function in scsi/scsi_base that was used, with an undocumented option. My last count on the LINT kernel shows: Total symbols: 3647 unref symbols: 463 undef symbols: 4 1 ref symbols: 1751 2 ref symbols: 485 Approaching the pain threshold now.
1995-12-17 21:23:44 +00:00
static int
Make similar changes to fo_stat() and fo_poll() as made earlier to fo_read() and fo_write(): explicitly use the cred argument to fo_poll() as "active_cred" using the passed file descriptor's f_cred reference to provide access to the file credential. Add an active_cred argument to fo_stat() so that implementers have access to the active credential as well as the file credential. Generally modify callers of fo_stat() to pass in td->td_ucred rather than fp->f_cred, which was redundantly provided via the fp argument. This set of modifications also permits threads to perform these operations on behalf of another thread without modifying their credential. Trickle this change down into fo_stat/poll() implementations: - badfo_poll(), badfo_stat(): modify/add arguments. - kqueue_poll(), kqueue_stat(): modify arguments. - pipe_poll(), pipe_stat(): modify/add arguments, pass active_cred to MAC checks rather than td->td_ucred. - soo_poll(), soo_stat(): modify/add arguments, pass fp->f_cred rather than cred to pru_sopoll() to maintain current semantics. - sopoll(): moidfy arguments. - vn_poll(), vn_statfile(): modify/add arguments, pass new arguments to vn_stat(). Pass active_cred to MAC and fp->f_cred to VOP_POLL() to maintian current semantics. - vn_close(): rename cred to file_cred to reflect reality while I'm here. - vn_stat(): Add active_cred and file_cred arguments to vn_stat() and consumers so that this distinction is maintained at the VFS as well as 'struct file' layer. Pass active_cred instead of td->td_ucred to MAC and to VOP_GETATTR() to maintain current semantics. - fifofs: modify the creation of a "filetemp" so that the file credential is properly initialized and can be used in the socket code if desired. Pass ap->a_td->td_ucred as the active credential to soo_poll(). If we teach the vnop interface about the distinction between file and active credentials, we would use the active credential here. Note that current inconsistent passing of active_cred vs. file_cred to VOP's is maintained. It's not clear why GETATTR would be authorized using active_cred while POLL would be authorized using file_cred at the file system level. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-16 12:52:03 +00:00
vn_poll(fp, events, active_cred, td)
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
struct file *fp;
vn_select -> vn_poll
1997-09-14 02:51:16 +00:00
int events;
Make similar changes to fo_stat() and fo_poll() as made earlier to fo_read() and fo_write(): explicitly use the cred argument to fo_poll() as "active_cred" using the passed file descriptor's f_cred reference to provide access to the file credential. Add an active_cred argument to fo_stat() so that implementers have access to the active credential as well as the file credential. Generally modify callers of fo_stat() to pass in td->td_ucred rather than fp->f_cred, which was redundantly provided via the fp argument. This set of modifications also permits threads to perform these operations on behalf of another thread without modifying their credential. Trickle this change down into fo_stat/poll() implementations: - badfo_poll(), badfo_stat(): modify/add arguments. - kqueue_poll(), kqueue_stat(): modify arguments. - pipe_poll(), pipe_stat(): modify/add arguments, pass active_cred to MAC checks rather than td->td_ucred. - soo_poll(), soo_stat(): modify/add arguments, pass fp->f_cred rather than cred to pru_sopoll() to maintain current semantics. - sopoll(): moidfy arguments. - vn_poll(), vn_statfile(): modify/add arguments, pass new arguments to vn_stat(). Pass active_cred to MAC and fp->f_cred to VOP_POLL() to maintian current semantics. - vn_close(): rename cred to file_cred to reflect reality while I'm here. - vn_stat(): Add active_cred and file_cred arguments to vn_stat() and consumers so that this distinction is maintained at the VFS as well as 'struct file' layer. Pass active_cred instead of td->td_ucred to MAC and to VOP_GETATTR() to maintain current semantics. - fifofs: modify the creation of a "filetemp" so that the file credential is properly initialized and can be used in the socket code if desired. Pass ap->a_td->td_ucred as the active credential to soo_poll(). If we teach the vnop interface about the distinction between file and active credentials, we would use the active credential here. Note that current inconsistent passing of active_cred vs. file_cred to VOP's is maintained. It's not clear why GETATTR would be authorized using active_cred while POLL would be authorized using file_cred at the file system level. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-16 12:52:03 +00:00
struct ucred *active_cred;
KSE Milestone 2 Note ALL MODULES MUST BE RECOMPILED make the kernel aware that there are smaller units of scheduling than the process. (but only allow one thread per process at this time). This is functionally equivalent to teh previousl -current except that there is a thread associated with each process. Sorry john! (your next MFC will be a doosie!) Reviewed by: peter@freebsd.org, dillon@freebsd.org X-MFC after: ha ha ha ha
2001-09-12 08:38:13 +00:00
struct thread *td;
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
{
Introduce support for Mandatory Access Control and extensible kernel access control Invoke appropriate MAC framework entry points to authorize a number of vnode operations, including read, write, stat, poll. This permits MAC policies to revoke access to files following label changes, and to limit information spread about the file to user processes. Note: currently the file cached credential is used for some of these authorization check. We will need to expand some of the MAC entry point APIs to permit multiple creds to be passed to the access control check to allow diverse policy behavior. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-01 17:23:22 +00:00
struct vnode *vp;
- Remove explicit giant acquires and replace it with VFS_LOCK_GIANT. Sponsored by: Isilon Systems, Inc.
2006-03-22 00:00:05 +00:00
int vfslocked;
Introduce support for Mandatory Access Control and extensible kernel access control Invoke appropriate MAC framework entry points to authorize a number of vnode operations, including read, write, stat, poll. This permits MAC policies to revoke access to files following label changes, and to limit information spread about the file to user processes. Note: currently the file cached credential is used for some of these authorization check. We will need to expand some of the MAC entry point APIs to permit multiple creds to be passed to the access control check to allow diverse policy behavior. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-01 17:23:22 +00:00
int error;
Add a f_vnode field to struct file. Several of the subtypes have an associated vnode which is used for stuff like the f*() functions. By giving the vnode a speparate field, a number of checks for the specific subtype can be replaced simply with a check for f_vnode != NULL, and we can later free f_data up to subtype specific use. At this point in time, f_data still points to the vnode, so any code I might have overlooked will still work.
2003-06-22 08:41:43 +00:00
vp = fp->f_vnode;
- Remove explicit giant acquires and replace it with VFS_LOCK_GIANT. Sponsored by: Isilon Systems, Inc.
2006-03-22 00:00:05 +00:00
vfslocked = VFS_LOCK_GIANT(vp->v_mount);
Introduce support for Mandatory Access Control and extensible kernel access control Invoke appropriate MAC framework entry points to authorize a number of vnode operations, including read, write, stat, poll. This permits MAC policies to revoke access to files following label changes, and to limit information spread about the file to user processes. Note: currently the file cached credential is used for some of these authorization check. We will need to expand some of the MAC entry point APIs to permit multiple creds to be passed to the access control check to allow diverse policy behavior. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-01 17:23:22 +00:00
#ifdef MAC
vn_lock(vp, LK_EXCLUSIVE | LK_RETRY, td);
Pass active_cred and file_cred into the MAC framework explicitly for mac_check_vnode_{poll,read,stat,write}(). Pass in fp->f_cred when calling these checks with a struct file available. Otherwise, pass NOCRED. All currently MAC policies use active_cred, but could now offer the cached credential semantic used for the base system security model. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-19 19:04:53 +00:00
error = mac_check_vnode_poll(active_cred, fp->f_cred, vp);
Introduce support for Mandatory Access Control and extensible kernel access control Invoke appropriate MAC framework entry points to authorize a number of vnode operations, including read, write, stat, poll. This permits MAC policies to revoke access to files following label changes, and to limit information spread about the file to user processes. Note: currently the file cached credential is used for some of these authorization check. We will need to expand some of the MAC entry point APIs to permit multiple creds to be passed to the access control check to allow diverse policy behavior. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-01 17:23:22 +00:00
VOP_UNLOCK(vp, 0, td);
Give vn_poll single exit point (to make it easier to insert "mtx_unlock(&Giant)" real soon now).
2004-11-15 21:56:42 +00:00
if (!error)
Introduce support for Mandatory Access Control and extensible kernel access control Invoke appropriate MAC framework entry points to authorize a number of vnode operations, including read, write, stat, poll. This permits MAC policies to revoke access to files following label changes, and to limit information spread about the file to user processes. Note: currently the file cached credential is used for some of these authorization check. We will need to expand some of the MAC entry point APIs to permit multiple creds to be passed to the access control check to allow diverse policy behavior. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-08-01 17:23:22 +00:00
#endif
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
Give vn_poll single exit point (to make it easier to insert "mtx_unlock(&Giant)" real soon now).
2004-11-15 21:56:42 +00:00
error = VOP_POLL(vp, events, fp->f_cred, td);
- Remove explicit giant acquires and replace it with VFS_LOCK_GIANT. Sponsored by: Isilon Systems, Inc.
2006-03-22 00:00:05 +00:00
VFS_UNLOCK_GIANT(vfslocked);
Give vn_poll single exit point (to make it easier to insert "mtx_unlock(&Giant)" real soon now).
2004-11-15 21:56:42 +00:00
return (error);
BSD 4.4 Lite Kernel Sources
1994-05-24 10:09:53 +00:00
}
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
/*
* Check that the vnode is still valid, and if so
* acquire requested lock.
*/
int
KSE Milestone 2 Note ALL MODULES MUST BE RECOMPILED make the kernel aware that there are smaller units of scheduling than the process. (but only allow one thread per process at this time). This is functionally equivalent to teh previousl -current except that there is a thread associated with each process. Sorry john! (your next MFC will be a doosie!) Reviewed by: peter@freebsd.org, dillon@freebsd.org X-MFC after: ha ha ha ha
2001-09-12 08:38:13 +00:00
vn_lock(vp, flags, td)
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
struct vnode *vp;
int flags;
KSE Milestone 2 Note ALL MODULES MUST BE RECOMPILED make the kernel aware that there are smaller units of scheduling than the process. (but only allow one thread per process at this time). This is functionally equivalent to teh previousl -current except that there is a thread associated with each process. Sorry john! (your next MFC will be a doosie!) Reviewed by: peter@freebsd.org, dillon@freebsd.org X-MFC after: ha ha ha ha
2001-09-12 08:38:13 +00:00
struct thread *td;
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
{
int error;
Previous commit broke interlock locking for !LK_RETRY case.
2001-03-26 12:45:35 +00:00
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
do {
Use 127 instead of CHAR_MAX for the limit on the sequence count. The limit doesn't have anything to do with characters. The count mainly needs to fit in the VOP_READ() ioflag after being left shifted by 16. Moved vn_lock() before vn_closefile(). vn_lock() was mismerged from Lite2. Removed some gratuitous braces.
1997-10-27 15:26:23 +00:00
if ((flags & LK_INTERLOCK) == 0)
- Replace v_flag with v_iflag and v_vflag - v_vflag is protected by the vnode lock and is used when synchronization with VOP calls is needed. - v_iflag is protected by interlock and is used for dealing with vnode management issues. These flags include X/O LOCK, FREE, DOOMED, etc. - All accesses to v_iflag and v_vflag have either been locked or marked with mp_fixme's. - Many ASSERT_VOP_LOCKED calls have been added where the locking was not clear. - Many functions in vfs_subr.c were restructured to provide for stronger locking. Idea stolen from: BSD/OS
2002-08-04 10:29:36 +00:00
VI_LOCK(vp);
- Assert that we're no longer doing recursive vn_locks in inactive/reclaim as I'd like to get rid of the vxthread. - Handle lock requests which don't actually want a lock as this is a much more convenient place to handle this condition than in vget(). These requests simply want to know that VI_DOOMED isn't set. - Correct a test at the end of vn_lock, if error !=0 should be if error == 0, this has been broken since I comitted the VI_DOOMED changes, but no one ran into it because vget() duplicated this functionality. Sponsored by: Isilon Systems, Inc.
2005-04-11 09:23:56 +00:00
if ((flags & LK_NOWAIT || (flags & LK_TYPE_MASK) == 0) &&
- Stop checking vxthread, we've asserted that it was useless for several weeks.
2005-04-27 09:17:11 +00:00
vp->v_iflag & VI_DOOMED) {
- Do a vn_start_write in vn_close, we may write if this is the last ref on an unlinked file. We can't know if this is the case until after we have the lock. - Lock the vnode in vn_close, many filesystems had code which was unsafe without the lock held, and holding it greatly simplifies vgone(). - Adjust vn_lock() to check for the VI_DOOMED flag where appropriate. Sponsored by: Isilon Systems, Inc.
2005-03-13 11:56:28 +00:00
VI_UNLOCK(vp);
return (ENOENT);
}
- Assert that we're no longer doing recursive vn_locks in inactive/reclaim as I'd like to get rid of the vxthread. - Handle lock requests which don't actually want a lock as this is a much more convenient place to handle this condition than in vget(). These requests simply want to know that VI_DOOMED isn't set. - Correct a test at the end of vn_lock, if error !=0 should be if error == 0, this has been broken since I comitted the VI_DOOMED changes, but no one ran into it because vget() duplicated this functionality. Sponsored by: Isilon Systems, Inc.
2005-04-11 09:23:56 +00:00
/*
* Just polling to check validity.
*/
if ((flags & LK_TYPE_MASK) == 0) {
VI_UNLOCK(vp);
return (0);
}
- Closer inspection revealed a possible deadlock situation in vn_lock() that was introduced by my last commit but not caught by stress testing. Fix that and slightly restructure the code so that it is more readable.
2002-08-22 07:57:43 +00:00
/*
* lockmgr drops interlock before it will return for
* any reason. So force the code above to relock it.
*/
- LK_NOPAUSE is a nop now. Sponsored by: Isilon Systems, Inc.
2005-03-31 04:37:09 +00:00
error = VOP_LOCK(vp, flags | LK_INTERLOCK, td);
- Closer inspection revealed a possible deadlock situation in vn_lock() that was introduced by my last commit but not caught by stress testing. Fix that and slightly restructure the code so that it is more readable.
2002-08-22 07:57:43 +00:00
flags &= ~LK_INTERLOCK;
- It has long been my suspicion that we don't actually need a loop in vn_lock(). Add an assert that will help me gain more confidence that this is correct. Sponsored by: Isilon Systems, Inc.
2005-06-13 00:47:29 +00:00
KASSERT((flags & LK_RETRY) == 0 || error == 0,
("LK_RETRY set with incompatible flags %d\n", flags));
- Do a vn_start_write in vn_close, we may write if this is the last ref on an unlinked file. We can't know if this is the case until after we have the lock. - Lock the vnode in vn_close, many filesystems had code which was unsafe without the lock held, and holding it greatly simplifies vgone(). - Adjust vn_lock() to check for the VI_DOOMED flag where appropriate. Sponsored by: Isilon Systems, Inc.
2005-03-13 11:56:28 +00:00
/*
* Callers specify LK_RETRY if they wish to get dead vnodes.
* If RETRY is not set, we return ENOENT instead.
*/
- Assert that we're no longer doing recursive vn_locks in inactive/reclaim as I'd like to get rid of the vxthread. - Handle lock requests which don't actually want a lock as this is a much more convenient place to handle this condition than in vget(). These requests simply want to know that VI_DOOMED isn't set. - Correct a test at the end of vn_lock, if error !=0 should be if error == 0, this has been broken since I comitted the VI_DOOMED changes, but no one ran into it because vget() duplicated this functionality. Sponsored by: Isilon Systems, Inc.
2005-04-11 09:23:56 +00:00
if (error == 0 && vp->v_iflag & VI_DOOMED &&
- Stop checking vxthread, we've asserted that it was useless for several weeks.
2005-04-27 09:17:11 +00:00
(flags & LK_RETRY) == 0) {
- Do a vn_start_write in vn_close, we may write if this is the last ref on an unlinked file. We can't know if this is the case until after we have the lock. - Lock the vnode in vn_close, many filesystems had code which was unsafe without the lock held, and holding it greatly simplifies vgone(). - Adjust vn_lock() to check for the VI_DOOMED flag where appropriate. Sponsored by: Isilon Systems, Inc.
2005-03-13 11:56:28 +00:00
VOP_UNLOCK(vp, 0, td);
error = ENOENT;
break;
}
- Fix interlock handling in vn_lock(). Previously, vn_lock() could return with interlock held in error conditions when the caller did not specify LK_INTERLOCK. - Add several comments to vn_lock() describing the rational behind the code flow since it was not immediately obvious.
2002-08-22 06:58:11 +00:00
} while (flags & LK_RETRY && error != 0);
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
return (error);
}
Use 127 instead of CHAR_MAX for the limit on the sequence count. The limit doesn't have anything to do with characters. The count mainly needs to fit in the VOP_READ() ioflag after being left shifted by 16. Moved vn_lock() before vn_closefile(). vn_lock() was mismerged from Lite2. Removed some gratuitous braces.
1997-10-27 15:26:23 +00:00
/*
* File table vnode close routine.
*/
static int
KSE Milestone 2 Note ALL MODULES MUST BE RECOMPILED make the kernel aware that there are smaller units of scheduling than the process. (but only allow one thread per process at this time). This is functionally equivalent to teh previousl -current except that there is a thread associated with each process. Sorry john! (your next MFC will be a doosie!) Reviewed by: peter@freebsd.org, dillon@freebsd.org X-MFC after: ha ha ha ha
2001-09-12 08:38:13 +00:00
vn_closefile(fp, td)
Use 127 instead of CHAR_MAX for the limit on the sequence count. The limit doesn't have anything to do with characters. The count mainly needs to fit in the VOP_READ() ioflag after being left shifted by 16. Moved vn_lock() before vn_closefile(). vn_lock() was mismerged from Lite2. Removed some gratuitous braces.
1997-10-27 15:26:23 +00:00
struct file *fp;
KSE Milestone 2 Note ALL MODULES MUST BE RECOMPILED make the kernel aware that there are smaller units of scheduling than the process. (but only allow one thread per process at this time). This is functionally equivalent to teh previousl -current except that there is a thread associated with each process. Sorry john! (your next MFC will be a doosie!) Reviewed by: peter@freebsd.org, dillon@freebsd.org X-MFC after: ha ha ha ha
2001-09-12 08:38:13 +00:00
struct thread *td;
Use 127 instead of CHAR_MAX for the limit on the sequence count. The limit doesn't have anything to do with characters. The count mainly needs to fit in the VOP_READ() ioflag after being left shifted by 16. Moved vn_lock() before vn_closefile(). vn_lock() was mismerged from Lite2. Removed some gratuitous braces.
1997-10-27 15:26:23 +00:00
{
Push the VOP_ADVLOCK() call to release advisory locks on vnode file descriptors out of fdrop_locked() and into vn_closefile(). This removes all knowledge of vnodes from fdrop_locked(), since the lock behavior was specific to vnodes. This also removes the specific requirement for Giant in fdrop_locked(), it's now only required by code that it calls into. Add GIANT_REQUIRED to vn_closefile() since VFS requires Giant.
2004-06-01 18:03:20 +00:00
struct vnode *vp;
struct flock lf;
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
int vfslocked;
Push acquisition of Giant from fdrop_closed() into fo_close() so that individual file object implementations can optionally acquire Giant if they require it: - soo_close(): depends on debug.mpsafenet - pipe_close(): Giant not acquired - kqueue_close(): Giant required - vn_close(): Giant required - cryptof_close(): Giant required (conservative) Notes: Giant is still acquired in close() even when closing MPSAFE objects due to kqueue requiring Giant in the calling closef() code. Microbenchmarks indicate that this removal of Giant cuts 3%-3% off of pipe create/destroy pairs from user space with SMP compiled into the kernel. The cryptodev and opencrypto code appears MPSAFE, but I'm unable to test it extensively and so have left Giant over fo_close(). It can probably be removed given some testing and review.
2004-07-22 18:35:43 +00:00
int error;
Push the VOP_ADVLOCK() call to release advisory locks on vnode file descriptors out of fdrop_locked() and into vn_closefile(). This removes all knowledge of vnodes from fdrop_locked(), since the lock behavior was specific to vnodes. This also removes the specific requirement for Giant in fdrop_locked(), it's now only required by code that it calls into. Add GIANT_REQUIRED to vn_closefile() since VFS requires Giant.
2004-06-01 18:03:20 +00:00
vp = fp->f_vnode;
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
vfslocked = VFS_LOCK_GIANT(vp->v_mount);
Rather than assert f_type==DTYPE_VNODE, conditionally perform the file lock release based on f_type==DTYPE_VNODE. vn_closefile() is used by non-vnode types as well (fifo).
2004-06-01 23:36:47 +00:00
if (fp->f_type == DTYPE_VNODE && fp->f_flag & FHASLOCK) {
Push the VOP_ADVLOCK() call to release advisory locks on vnode file descriptors out of fdrop_locked() and into vn_closefile(). This removes all knowledge of vnodes from fdrop_locked(), since the lock behavior was specific to vnodes. This also removes the specific requirement for Giant in fdrop_locked(), it's now only required by code that it calls into. Add GIANT_REQUIRED to vn_closefile() since VFS requires Giant.
2004-06-01 18:03:20 +00:00
lf.l_whence = SEEK_SET;
lf.l_start = 0;
lf.l_len = 0;
lf.l_type = F_UNLCK;
Eradicate caddr_t from the VFS API.
2005-12-14 00:49:52 +00:00
(void) VOP_ADVLOCK(vp, fp, F_UNLCK, &lf, F_FLOCK);
Push the VOP_ADVLOCK() call to release advisory locks on vnode file descriptors out of fdrop_locked() and into vn_closefile(). This removes all knowledge of vnodes from fdrop_locked(), since the lock behavior was specific to vnodes. This also removes the specific requirement for Giant in fdrop_locked(), it's now only required by code that it calls into. Add GIANT_REQUIRED to vn_closefile() since VFS requires Giant.
2004-06-01 18:03:20 +00:00
}
Use 127 instead of CHAR_MAX for the limit on the sequence count. The limit doesn't have anything to do with characters. The count mainly needs to fit in the VOP_READ() ioflag after being left shifted by 16. Moved vn_lock() before vn_closefile(). vn_lock() was mismerged from Lite2. Removed some gratuitous braces.
1997-10-27 15:26:23 +00:00
Fix fd race conditions (during shared fd table usage.) Badfileops is now used in f_ops in place of NULL, and modifications to the files are more carefully ordered. f_ops should also be set to &badfileops upon "close" of a file. This does not fix other problems mentioned in this PR than the first one. PR: 11629 Reviewed by: peter
1999-08-04 18:53:50 +00:00
fp->f_ops = &badfileops;
Push the VOP_ADVLOCK() call to release advisory locks on vnode file descriptors out of fdrop_locked() and into vn_closefile(). This removes all knowledge of vnodes from fdrop_locked(), since the lock behavior was specific to vnodes. This also removes the specific requirement for Giant in fdrop_locked(), it's now only required by code that it calls into. Add GIANT_REQUIRED to vn_closefile() since VFS requires Giant.
2004-06-01 18:03:20 +00:00
Push acquisition of Giant from fdrop_closed() into fo_close() so that individual file object implementations can optionally acquire Giant if they require it: - soo_close(): depends on debug.mpsafenet - pipe_close(): Giant not acquired - kqueue_close(): Giant required - vn_close(): Giant required - cryptof_close(): Giant required (conservative) Notes: Giant is still acquired in close() even when closing MPSAFE objects due to kqueue requiring Giant in the calling closef() code. Microbenchmarks indicate that this removal of Giant cuts 3%-3% off of pipe create/destroy pairs from user space with SMP compiled into the kernel. The cryptodev and opencrypto code appears MPSAFE, but I'm unable to test it extensively and so have left Giant over fo_close(). It can probably be removed given some testing and review.
2004-07-22 18:35:43 +00:00
error = vn_close(vp, fp->f_flag, fp->f_cred, td);
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
VFS_UNLOCK_GIANT(vfslocked);
Push acquisition of Giant from fdrop_closed() into fo_close() so that individual file object implementations can optionally acquire Giant if they require it: - soo_close(): depends on debug.mpsafenet - pipe_close(): Giant not acquired - kqueue_close(): Giant required - vn_close(): Giant required - cryptof_close(): Giant required (conservative) Notes: Giant is still acquired in close() even when closing MPSAFE objects due to kqueue requiring Giant in the calling closef() code. Microbenchmarks indicate that this removal of Giant cuts 3%-3% off of pipe create/destroy pairs from user space with SMP compiled into the kernel. The cryptodev and opencrypto code appears MPSAFE, but I'm unable to test it extensively and so have left Giant over fo_close(). It can probably be removed given some testing and review.
2004-07-22 18:35:43 +00:00
return (error);
Use 127 instead of CHAR_MAX for the limit on the sequence count. The limit doesn't have anything to do with characters. The count mainly needs to fit in the VOP_READ() ioflag after being left shifted by 16. Moved vn_lock() before vn_closefile(). vn_lock() was mismerged from Lite2. Removed some gratuitous braces.
1997-10-27 15:26:23 +00:00
}
Introduce kqueue() and kevent(), a kernel event notification facility.
2000-04-16 18:53:38 +00:00
Add snapshots to the fast filesystem. Most of the changes support the gating of system calls that cause modifications to the underlying filesystem. The gating can be enabled by any filesystem that needs to consistently suspend operations by adding the vop_stdgetwritemount to their set of vnops. Once gating is enabled, the function vfs_write_suspend stops all new write operations to a filesystem, allows any filesystem modifying system calls already in progress to complete, then sync's the filesystem to disk and returns. The function vfs_write_resume allows the suspended write operations to begin again. Gating is not added by default for all filesystems as for SMP systems it adds two extra locks to such critical kernel paths as the write system call. Thus, gating should only be added as needed. Details on the use and current status of snapshots in FFS can be found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness is not included here. Unless and until you create a snapshot file, these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
/*
* Preparing to start a filesystem write operation. If the operation is
* permitted, then we bump the count of operations in progress and
* proceed. If a suspend request is in progress, we wait until the
* suspension is over, and then proceed.
*/
int
vn_start_write(vp, mpp, flags)
struct vnode *vp;
struct mount **mpp;
int flags;
{
struct mount *mp;
int error;
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
error = 0;
Add snapshots to the fast filesystem. Most of the changes support the gating of system calls that cause modifications to the underlying filesystem. The gating can be enabled by any filesystem that needs to consistently suspend operations by adding the vop_stdgetwritemount to their set of vnops. Once gating is enabled, the function vfs_write_suspend stops all new write operations to a filesystem, allows any filesystem modifying system calls already in progress to complete, then sync's the filesystem to disk and returns. The function vfs_write_resume allows the suspended write operations to begin again. Gating is not added by default for all filesystems as for SMP systems it adds two extra locks to such critical kernel paths as the write system call. Thus, gating should only be added as needed. Details on the use and current status of snapshots in FFS can be found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness is not included here. Unless and until you create a snapshot file, these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
/*
* If a vnode is provided, get and return the mount point that
* to which it will write.
*/
if (vp != NULL) {
if ((error = VOP_GETWRITEMOUNT(vp, mpp)) != 0) {
*mpp = NULL;
if (error != EOPNOTSUPP)
return (error);
return (0);
}
}
if ((mp = *mpp) == NULL)
return (0);
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
MNT_ILOCK(mp);
- Release the references acquired by VOP_GETWRITEMOUNT and vfs_getvfs(). Discussed with: tegge Tested by: kris Sponsored by: Isilon Systems, Inc.
2006-03-31 03:54:20 +00:00
if (vp == NULL)
MNT_REF(mp);
Add snapshots to the fast filesystem. Most of the changes support the gating of system calls that cause modifications to the underlying filesystem. The gating can be enabled by any filesystem that needs to consistently suspend operations by adding the vop_stdgetwritemount to their set of vnops. Once gating is enabled, the function vfs_write_suspend stops all new write operations to a filesystem, allows any filesystem modifying system calls already in progress to complete, then sync's the filesystem to disk and returns. The function vfs_write_resume allows the suspended write operations to begin again. Gating is not added by default for all filesystems as for SMP systems it adds two extra locks to such critical kernel paths as the write system call. Thus, gating should only be added as needed. Details on the use and current status of snapshots in FFS can be found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness is not included here. Unless and until you create a snapshot file, these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
/*
* Check on status of suspension.
*/
while ((mp->mnt_kern_flag & MNTK_SUSPEND) != 0) {
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
if (flags & V_NOWAIT) {
error = EWOULDBLOCK;
goto unlock;
}
error = msleep(&mp->mnt_flag, MNT_MTX(mp),
(PUSER - 1) | (flags & PCATCH), "suspfs", 0);
Add snapshots to the fast filesystem. Most of the changes support the gating of system calls that cause modifications to the underlying filesystem. The gating can be enabled by any filesystem that needs to consistently suspend operations by adding the vop_stdgetwritemount to their set of vnops. Once gating is enabled, the function vfs_write_suspend stops all new write operations to a filesystem, allows any filesystem modifying system calls already in progress to complete, then sync's the filesystem to disk and returns. The function vfs_write_resume allows the suspended write operations to begin again. Gating is not added by default for all filesystems as for SMP systems it adds two extra locks to such critical kernel paths as the write system call. Thus, gating should only be added as needed. Details on the use and current status of snapshots in FFS can be found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness is not included here. Unless and until you create a snapshot file, these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
if (error)
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
goto unlock;
Add snapshots to the fast filesystem. Most of the changes support the gating of system calls that cause modifications to the underlying filesystem. The gating can be enabled by any filesystem that needs to consistently suspend operations by adding the vop_stdgetwritemount to their set of vnops. Once gating is enabled, the function vfs_write_suspend stops all new write operations to a filesystem, allows any filesystem modifying system calls already in progress to complete, then sync's the filesystem to disk and returns. The function vfs_write_resume allows the suspended write operations to begin again. Gating is not added by default for all filesystems as for SMP systems it adds two extra locks to such critical kernel paths as the write system call. Thus, gating should only be added as needed. Details on the use and current status of snapshots in FFS can be found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness is not included here. Unless and until you create a snapshot file, these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
}
if (flags & V_XSLEEP)
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
goto unlock;
Add snapshots to the fast filesystem. Most of the changes support the gating of system calls that cause modifications to the underlying filesystem. The gating can be enabled by any filesystem that needs to consistently suspend operations by adding the vop_stdgetwritemount to their set of vnops. Once gating is enabled, the function vfs_write_suspend stops all new write operations to a filesystem, allows any filesystem modifying system calls already in progress to complete, then sync's the filesystem to disk and returns. The function vfs_write_resume allows the suspended write operations to begin again. Gating is not added by default for all filesystems as for SMP systems it adds two extra locks to such critical kernel paths as the write system call. Thus, gating should only be added as needed. Details on the use and current status of snapshots in FFS can be found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness is not included here. Unless and until you create a snapshot file, these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
mp->mnt_writeopcount++;
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
unlock:
- Release the references acquired by VOP_GETWRITEMOUNT and vfs_getvfs(). Discussed with: tegge Tested by: kris Sponsored by: Isilon Systems, Inc.
2006-03-31 03:54:20 +00:00
MNT_REL(mp);
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
MNT_IUNLOCK(mp);
return (error);
Add snapshots to the fast filesystem. Most of the changes support the gating of system calls that cause modifications to the underlying filesystem. The gating can be enabled by any filesystem that needs to consistently suspend operations by adding the vop_stdgetwritemount to their set of vnops. Once gating is enabled, the function vfs_write_suspend stops all new write operations to a filesystem, allows any filesystem modifying system calls already in progress to complete, then sync's the filesystem to disk and returns. The function vfs_write_resume allows the suspended write operations to begin again. Gating is not added by default for all filesystems as for SMP systems it adds two extra locks to such critical kernel paths as the write system call. Thus, gating should only be added as needed. Details on the use and current status of snapshots in FFS can be found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness is not included here. Unless and until you create a snapshot file, these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
}
/*
* Secondary suspension. Used by operations such as vop_inactive
* routines that are needed by the higher level functions. These
* are allowed to proceed until all the higher level functions have
* completed (indicated by mnt_writeopcount dropping to zero). At that
* time, these operations are halted until the suspension is over.
*/
int
This patch corrects the first round of panics and hangs reported with the new snapshot code. Update addaliasu to correctly implement the semantics of the old checkalias function. When a device vnode first comes into existence, check to see if an anonymous vnode for the same device was created at boot time by bdevvp(). If so, adopt the bdevvp vnode rather than creating a new vnode for the device. This corrects a problem which caused the kernel to panic when taking a snapshot of the root filesystem. Change the calling convention of vn_write_suspend_wait() to be the same as vn_start_write(). Split out softdep_flushworklist() from softdep_flushfiles() so that it can be used to clear the work queue when suspending filesystem operations. Access to buffers becomes recursive so that snapshots can recursively traverse their indirect blocks using ffs_copyonwrite() when checking for the need for copy on write when flushing one of their own indirect blocks. This eliminates a deadlock between the syncer daemon and a process taking a snapshot. Ensure that softdep_process_worklist() can never block because of a snapshot being taken. This eliminates a problem with buffer starvation. Cleanup change in ffs_sync() which did not synchronously wait when MNT_WAIT was specified. The result was an unclean filesystem panic when doing forcible unmount with heavy filesystem I/O in progress. Return a zero'ed block when reading a block that was not in use at the time that a snapshot was taken. Normally, these blocks should never be read. However, the readahead code will occationally read them which can cause unexpected behavior. Clean up the debugging code that ensures that no blocks be written on a filesystem while it is suspended. Snapshots must explicitly label the blocks that they are writing during the suspension so that they do not cause a `write on suspended filesystem' panic. Reorganize ffs_copyonwrite() to eliminate a deadlock and also to prevent a race condition that would permit the same block to be copied twice. This change eliminates an unexpected soft updates inconsistency in fsck caused by the double allocation. Use bqrelse rather than brelse for buffers that will be needed soon again by the snapshot code. This improves snapshot performance.
2000-07-24 05:28:33 +00:00
vn_write_suspend_wait(vp, mp, flags)
Add snapshots to the fast filesystem. Most of the changes support the gating of system calls that cause modifications to the underlying filesystem. The gating can be enabled by any filesystem that needs to consistently suspend operations by adding the vop_stdgetwritemount to their set of vnops. Once gating is enabled, the function vfs_write_suspend stops all new write operations to a filesystem, allows any filesystem modifying system calls already in progress to complete, then sync's the filesystem to disk and returns. The function vfs_write_resume allows the suspended write operations to begin again. Gating is not added by default for all filesystems as for SMP systems it adds two extra locks to such critical kernel paths as the write system call. Thus, gating should only be added as needed. Details on the use and current status of snapshots in FFS can be found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness is not included here. Unless and until you create a snapshot file, these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
struct vnode *vp;
This patch corrects the first round of panics and hangs reported with the new snapshot code. Update addaliasu to correctly implement the semantics of the old checkalias function. When a device vnode first comes into existence, check to see if an anonymous vnode for the same device was created at boot time by bdevvp(). If so, adopt the bdevvp vnode rather than creating a new vnode for the device. This corrects a problem which caused the kernel to panic when taking a snapshot of the root filesystem. Change the calling convention of vn_write_suspend_wait() to be the same as vn_start_write(). Split out softdep_flushworklist() from softdep_flushfiles() so that it can be used to clear the work queue when suspending filesystem operations. Access to buffers becomes recursive so that snapshots can recursively traverse their indirect blocks using ffs_copyonwrite() when checking for the need for copy on write when flushing one of their own indirect blocks. This eliminates a deadlock between the syncer daemon and a process taking a snapshot. Ensure that softdep_process_worklist() can never block because of a snapshot being taken. This eliminates a problem with buffer starvation. Cleanup change in ffs_sync() which did not synchronously wait when MNT_WAIT was specified. The result was an unclean filesystem panic when doing forcible unmount with heavy filesystem I/O in progress. Return a zero'ed block when reading a block that was not in use at the time that a snapshot was taken. Normally, these blocks should never be read. However, the readahead code will occationally read them which can cause unexpected behavior. Clean up the debugging code that ensures that no blocks be written on a filesystem while it is suspended. Snapshots must explicitly label the blocks that they are writing during the suspension so that they do not cause a `write on suspended filesystem' panic. Reorganize ffs_copyonwrite() to eliminate a deadlock and also to prevent a race condition that would permit the same block to be copied twice. This change eliminates an unexpected soft updates inconsistency in fsck caused by the double allocation. Use bqrelse rather than brelse for buffers that will be needed soon again by the snapshot code. This improves snapshot performance.
2000-07-24 05:28:33 +00:00
struct mount *mp;
Add snapshots to the fast filesystem. Most of the changes support the gating of system calls that cause modifications to the underlying filesystem. The gating can be enabled by any filesystem that needs to consistently suspend operations by adding the vop_stdgetwritemount to their set of vnops. Once gating is enabled, the function vfs_write_suspend stops all new write operations to a filesystem, allows any filesystem modifying system calls already in progress to complete, then sync's the filesystem to disk and returns. The function vfs_write_resume allows the suspended write operations to begin again. Gating is not added by default for all filesystems as for SMP systems it adds two extra locks to such critical kernel paths as the write system call. Thus, gating should only be added as needed. Details on the use and current status of snapshots in FFS can be found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness is not included here. Unless and until you create a snapshot file, these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
int flags;
{
int error;
This patch corrects the first round of panics and hangs reported with the new snapshot code. Update addaliasu to correctly implement the semantics of the old checkalias function. When a device vnode first comes into existence, check to see if an anonymous vnode for the same device was created at boot time by bdevvp(). If so, adopt the bdevvp vnode rather than creating a new vnode for the device. This corrects a problem which caused the kernel to panic when taking a snapshot of the root filesystem. Change the calling convention of vn_write_suspend_wait() to be the same as vn_start_write(). Split out softdep_flushworklist() from softdep_flushfiles() so that it can be used to clear the work queue when suspending filesystem operations. Access to buffers becomes recursive so that snapshots can recursively traverse their indirect blocks using ffs_copyonwrite() when checking for the need for copy on write when flushing one of their own indirect blocks. This eliminates a deadlock between the syncer daemon and a process taking a snapshot. Ensure that softdep_process_worklist() can never block because of a snapshot being taken. This eliminates a problem with buffer starvation. Cleanup change in ffs_sync() which did not synchronously wait when MNT_WAIT was specified. The result was an unclean filesystem panic when doing forcible unmount with heavy filesystem I/O in progress. Return a zero'ed block when reading a block that was not in use at the time that a snapshot was taken. Normally, these blocks should never be read. However, the readahead code will occationally read them which can cause unexpected behavior. Clean up the debugging code that ensures that no blocks be written on a filesystem while it is suspended. Snapshots must explicitly label the blocks that they are writing during the suspension so that they do not cause a `write on suspended filesystem' panic. Reorganize ffs_copyonwrite() to eliminate a deadlock and also to prevent a race condition that would permit the same block to be copied twice. This change eliminates an unexpected soft updates inconsistency in fsck caused by the double allocation. Use bqrelse rather than brelse for buffers that will be needed soon again by the snapshot code. This improves snapshot performance.
2000-07-24 05:28:33 +00:00
if (vp != NULL) {
if ((error = VOP_GETWRITEMOUNT(vp, &mp)) != 0) {
if (error != EOPNOTSUPP)
return (error);
return (0);
}
Add snapshots to the fast filesystem. Most of the changes support the gating of system calls that cause modifications to the underlying filesystem. The gating can be enabled by any filesystem that needs to consistently suspend operations by adding the vop_stdgetwritemount to their set of vnops. Once gating is enabled, the function vfs_write_suspend stops all new write operations to a filesystem, allows any filesystem modifying system calls already in progress to complete, then sync's the filesystem to disk and returns. The function vfs_write_resume allows the suspended write operations to begin again. Gating is not added by default for all filesystems as for SMP systems it adds two extra locks to such critical kernel paths as the write system call. Thus, gating should only be added as needed. Details on the use and current status of snapshots in FFS can be found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness is not included here. Unless and until you create a snapshot file, these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
}
/*
* If we are not suspended or have not yet reached suspended
* mode, then let the operation proceed.
*/
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
if (mp == NULL)
return (0);
MNT_ILOCK(mp);
- Release the references acquired by VOP_GETWRITEMOUNT and vfs_getvfs(). Discussed with: tegge Tested by: kris Sponsored by: Isilon Systems, Inc.
2006-03-31 03:54:20 +00:00
if (vp == NULL)
MNT_REF(mp);
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
if ((mp->mnt_kern_flag & MNTK_SUSPENDED) == 0) {
- Release the references acquired by VOP_GETWRITEMOUNT and vfs_getvfs(). Discussed with: tegge Tested by: kris Sponsored by: Isilon Systems, Inc.
2006-03-31 03:54:20 +00:00
MNT_REL(mp);
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
MNT_IUNLOCK(mp);
Add snapshots to the fast filesystem. Most of the changes support the gating of system calls that cause modifications to the underlying filesystem. The gating can be enabled by any filesystem that needs to consistently suspend operations by adding the vop_stdgetwritemount to their set of vnops. Once gating is enabled, the function vfs_write_suspend stops all new write operations to a filesystem, allows any filesystem modifying system calls already in progress to complete, then sync's the filesystem to disk and returns. The function vfs_write_resume allows the suspended write operations to begin again. Gating is not added by default for all filesystems as for SMP systems it adds two extra locks to such critical kernel paths as the write system call. Thus, gating should only be added as needed. Details on the use and current status of snapshots in FFS can be found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness is not included here. Unless and until you create a snapshot file, these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
return (0);
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
}
if (flags & V_NOWAIT) {
- Release the references acquired by VOP_GETWRITEMOUNT and vfs_getvfs(). Discussed with: tegge Tested by: kris Sponsored by: Isilon Systems, Inc.
2006-03-31 03:54:20 +00:00
MNT_REL(mp);
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
MNT_IUNLOCK(mp);
Add snapshots to the fast filesystem. Most of the changes support the gating of system calls that cause modifications to the underlying filesystem. The gating can be enabled by any filesystem that needs to consistently suspend operations by adding the vop_stdgetwritemount to their set of vnops. Once gating is enabled, the function vfs_write_suspend stops all new write operations to a filesystem, allows any filesystem modifying system calls already in progress to complete, then sync's the filesystem to disk and returns. The function vfs_write_resume allows the suspended write operations to begin again. Gating is not added by default for all filesystems as for SMP systems it adds two extra locks to such critical kernel paths as the write system call. Thus, gating should only be added as needed. Details on the use and current status of snapshots in FFS can be found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness is not included here. Unless and until you create a snapshot file, these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
return (EWOULDBLOCK);
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
}
Add snapshots to the fast filesystem. Most of the changes support the gating of system calls that cause modifications to the underlying filesystem. The gating can be enabled by any filesystem that needs to consistently suspend operations by adding the vop_stdgetwritemount to their set of vnops. Once gating is enabled, the function vfs_write_suspend stops all new write operations to a filesystem, allows any filesystem modifying system calls already in progress to complete, then sync's the filesystem to disk and returns. The function vfs_write_resume allows the suspended write operations to begin again. Gating is not added by default for all filesystems as for SMP systems it adds two extra locks to such critical kernel paths as the write system call. Thus, gating should only be added as needed. Details on the use and current status of snapshots in FFS can be found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness is not included here. Unless and until you create a snapshot file, these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
/*
* Wait for the suspension to finish.
*/
- Release the references acquired by VOP_GETWRITEMOUNT and vfs_getvfs(). Discussed with: tegge Tested by: kris Sponsored by: Isilon Systems, Inc.
2006-03-31 03:54:20 +00:00
error = msleep(&mp->mnt_flag, MNT_MTX(mp),
(PUSER - 1) | (flags & PCATCH) | PDROP, "suspfs", 0);
vfs_rel(mp);
return (error);
Add snapshots to the fast filesystem. Most of the changes support the gating of system calls that cause modifications to the underlying filesystem. The gating can be enabled by any filesystem that needs to consistently suspend operations by adding the vop_stdgetwritemount to their set of vnops. Once gating is enabled, the function vfs_write_suspend stops all new write operations to a filesystem, allows any filesystem modifying system calls already in progress to complete, then sync's the filesystem to disk and returns. The function vfs_write_resume allows the suspended write operations to begin again. Gating is not added by default for all filesystems as for SMP systems it adds two extra locks to such critical kernel paths as the write system call. Thus, gating should only be added as needed. Details on the use and current status of snapshots in FFS can be found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness is not included here. Unless and until you create a snapshot file, these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
}
Use vn_start_secondary_write() and vn_finished_secondary_write() as a replacement for vn_write_suspend_wait() to better account for secondary write processing. Close race where secondary writes could be started after ffs_sync() returned but before the file system was marked as suspended. Detect if secondary writes or softdep processing occurred during vnode sync loop in ffs_sync() and retry the loop if needed.
2006-03-08 23:43:39 +00:00
/*
* Secondary suspension. Used by operations such as vop_inactive
* routines that are needed by the higher level functions. These
* are allowed to proceed until all the higher level functions have
* completed (indicated by mnt_writeopcount dropping to zero). At that
* time, these operations are halted until the suspension is over.
*/
int
vn_start_secondary_write(vp, mpp, flags)
struct vnode *vp;
struct mount **mpp;
int flags;
{
struct mount *mp;
int error;
retry:
if (vp != NULL) {
if ((error = VOP_GETWRITEMOUNT(vp, mpp)) != 0) {
*mpp = NULL;
if (error != EOPNOTSUPP)
return (error);
return (0);
}
}
/*
* If we are not suspended or have not yet reached suspended
* mode, then let the operation proceed.
*/
if ((mp = *mpp) == NULL)
return (0);
MNT_ILOCK(mp);
- Release the references acquired by VOP_GETWRITEMOUNT and vfs_getvfs(). Discussed with: tegge Tested by: kris Sponsored by: Isilon Systems, Inc.
2006-03-31 03:54:20 +00:00
if (vp == NULL)
MNT_REF(mp);
Block secondary writes while expunging active unlinked files. Fix detection of active unlinked files by checking VI_OWEINACT and VI_DOINGINACT in addition to v_usecount. Defer inactive handling for unlinked files if the file system is mostly suspended (secondary writes being blocked). Perform deferred inactive handling after the file system is resumed.
2006-03-11 01:08:37 +00:00
if ((mp->mnt_kern_flag & (MNTK_SUSPENDED | MNTK_SUSPEND2)) == 0) {
Use vn_start_secondary_write() and vn_finished_secondary_write() as a replacement for vn_write_suspend_wait() to better account for secondary write processing. Close race where secondary writes could be started after ffs_sync() returned but before the file system was marked as suspended. Detect if secondary writes or softdep processing occurred during vnode sync loop in ffs_sync() and retry the loop if needed.
2006-03-08 23:43:39 +00:00
mp->mnt_secondary_writes++;
mp->mnt_secondary_accwrites++;
- Release the references acquired by VOP_GETWRITEMOUNT and vfs_getvfs(). Discussed with: tegge Tested by: kris Sponsored by: Isilon Systems, Inc.
2006-03-31 03:54:20 +00:00
MNT_REL(mp);
Use vn_start_secondary_write() and vn_finished_secondary_write() as a replacement for vn_write_suspend_wait() to better account for secondary write processing. Close race where secondary writes could be started after ffs_sync() returned but before the file system was marked as suspended. Detect if secondary writes or softdep processing occurred during vnode sync loop in ffs_sync() and retry the loop if needed.
2006-03-08 23:43:39 +00:00
MNT_IUNLOCK(mp);
return (0);
}
if (flags & V_NOWAIT) {
- Release the references acquired by VOP_GETWRITEMOUNT and vfs_getvfs(). Discussed with: tegge Tested by: kris Sponsored by: Isilon Systems, Inc.
2006-03-31 03:54:20 +00:00
MNT_REL(mp);
Use vn_start_secondary_write() and vn_finished_secondary_write() as a replacement for vn_write_suspend_wait() to better account for secondary write processing. Close race where secondary writes could be started after ffs_sync() returned but before the file system was marked as suspended. Detect if secondary writes or softdep processing occurred during vnode sync loop in ffs_sync() and retry the loop if needed.
2006-03-08 23:43:39 +00:00
MNT_IUNLOCK(mp);
return (EWOULDBLOCK);
}
/*
* Wait for the suspension to finish.
*/
error = msleep(&mp->mnt_flag, MNT_MTX(mp),
(PUSER - 1) | (flags & PCATCH) | PDROP, "suspfs", 0);
- Release the references acquired by VOP_GETWRITEMOUNT and vfs_getvfs(). Discussed with: tegge Tested by: kris Sponsored by: Isilon Systems, Inc.
2006-03-31 03:54:20 +00:00
vfs_rel(mp);
Use vn_start_secondary_write() and vn_finished_secondary_write() as a replacement for vn_write_suspend_wait() to better account for secondary write processing. Close race where secondary writes could be started after ffs_sync() returned but before the file system was marked as suspended. Detect if secondary writes or softdep processing occurred during vnode sync loop in ffs_sync() and retry the loop if needed.
2006-03-08 23:43:39 +00:00
if (error == 0)
goto retry;
return (error);
}
Add snapshots to the fast filesystem. Most of the changes support the gating of system calls that cause modifications to the underlying filesystem. The gating can be enabled by any filesystem that needs to consistently suspend operations by adding the vop_stdgetwritemount to their set of vnops. Once gating is enabled, the function vfs_write_suspend stops all new write operations to a filesystem, allows any filesystem modifying system calls already in progress to complete, then sync's the filesystem to disk and returns. The function vfs_write_resume allows the suspended write operations to begin again. Gating is not added by default for all filesystems as for SMP systems it adds two extra locks to such critical kernel paths as the write system call. Thus, gating should only be added as needed. Details on the use and current status of snapshots in FFS can be found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness is not included here. Unless and until you create a snapshot file, these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
/*
* Filesystem write operation has completed. If we are suspending and this
* operation is the last one, notify the suspender that the suspension is
* now in effect.
*/
void
vn_finished_write(mp)
struct mount *mp;
{
if (mp == NULL)
return;
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
MNT_ILOCK(mp);
Add snapshots to the fast filesystem. Most of the changes support the gating of system calls that cause modifications to the underlying filesystem. The gating can be enabled by any filesystem that needs to consistently suspend operations by adding the vop_stdgetwritemount to their set of vnops. Once gating is enabled, the function vfs_write_suspend stops all new write operations to a filesystem, allows any filesystem modifying system calls already in progress to complete, then sync's the filesystem to disk and returns. The function vfs_write_resume allows the suspended write operations to begin again. Gating is not added by default for all filesystems as for SMP systems it adds two extra locks to such critical kernel paths as the write system call. Thus, gating should only be added as needed. Details on the use and current status of snapshots in FFS can be found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness is not included here. Unless and until you create a snapshot file, these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
mp->mnt_writeopcount--;
if (mp->mnt_writeopcount < 0)
panic("vn_finished_write: neg cnt");
if ((mp->mnt_kern_flag & MNTK_SUSPEND) != 0 &&
mp->mnt_writeopcount <= 0)
wakeup(&mp->mnt_writeopcount);
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
MNT_IUNLOCK(mp);
Add snapshots to the fast filesystem. Most of the changes support the gating of system calls that cause modifications to the underlying filesystem. The gating can be enabled by any filesystem that needs to consistently suspend operations by adding the vop_stdgetwritemount to their set of vnops. Once gating is enabled, the function vfs_write_suspend stops all new write operations to a filesystem, allows any filesystem modifying system calls already in progress to complete, then sync's the filesystem to disk and returns. The function vfs_write_resume allows the suspended write operations to begin again. Gating is not added by default for all filesystems as for SMP systems it adds two extra locks to such critical kernel paths as the write system call. Thus, gating should only be added as needed. Details on the use and current status of snapshots in FFS can be found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness is not included here. Unless and until you create a snapshot file, these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
}
Use vn_start_secondary_write() and vn_finished_secondary_write() as a replacement for vn_write_suspend_wait() to better account for secondary write processing. Close race where secondary writes could be started after ffs_sync() returned but before the file system was marked as suspended. Detect if secondary writes or softdep processing occurred during vnode sync loop in ffs_sync() and retry the loop if needed.
2006-03-08 23:43:39 +00:00
/*
* Filesystem secondary write operation has completed. If we are
* suspending and this operation is the last one, notify the suspender
* that the suspension is now in effect.
*/
void
vn_finished_secondary_write(mp)
struct mount *mp;
{
if (mp == NULL)
return;
MNT_ILOCK(mp);
mp->mnt_secondary_writes--;
if (mp->mnt_secondary_writes < 0)
panic("vn_finished_secondary_write: neg cnt");
if ((mp->mnt_kern_flag & MNTK_SUSPEND) != 0 &&
mp->mnt_secondary_writes <= 0)
wakeup(&mp->mnt_secondary_writes);
MNT_IUNLOCK(mp);
}
Add snapshots to the fast filesystem. Most of the changes support the gating of system calls that cause modifications to the underlying filesystem. The gating can be enabled by any filesystem that needs to consistently suspend operations by adding the vop_stdgetwritemount to their set of vnops. Once gating is enabled, the function vfs_write_suspend stops all new write operations to a filesystem, allows any filesystem modifying system calls already in progress to complete, then sync's the filesystem to disk and returns. The function vfs_write_resume allows the suspended write operations to begin again. Gating is not added by default for all filesystems as for SMP systems it adds two extra locks to such critical kernel paths as the write system call. Thus, gating should only be added as needed. Details on the use and current status of snapshots in FFS can be found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness is not included here. Unless and until you create a snapshot file, these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
/*
* Request a filesystem to suspend write operations.
*/
Within ufs, the ffs_sync and ffs_fsync functions did not always check for and/or report I/O errors. The result is that a VFS_SYNC or VOP_FSYNC called with MNT_WAIT could loop infinitely on ufs in the presence of a hard error writing a disk sector or in a filesystem full condition. This patch ensures that I/O errors will always be checked and returned. This patch also ensures that every call to VFS_SYNC or VOP_FSYNC with MNT_WAIT set checks for and takes appropriate action when an error is returned. Sponsored by: DARPA & NAI Labs.
2002-10-25 00:20:37 +00:00
int
Add snapshots to the fast filesystem. Most of the changes support the gating of system calls that cause modifications to the underlying filesystem. The gating can be enabled by any filesystem that needs to consistently suspend operations by adding the vop_stdgetwritemount to their set of vnops. Once gating is enabled, the function vfs_write_suspend stops all new write operations to a filesystem, allows any filesystem modifying system calls already in progress to complete, then sync's the filesystem to disk and returns. The function vfs_write_resume allows the suspended write operations to begin again. Gating is not added by default for all filesystems as for SMP systems it adds two extra locks to such critical kernel paths as the write system call. Thus, gating should only be added as needed. Details on the use and current status of snapshots in FFS can be found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness is not included here. Unless and until you create a snapshot file, these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
vfs_write_suspend(mp)
struct mount *mp;
{
KSE Milestone 2 Note ALL MODULES MUST BE RECOMPILED make the kernel aware that there are smaller units of scheduling than the process. (but only allow one thread per process at this time). This is functionally equivalent to teh previousl -current except that there is a thread associated with each process. Sorry john! (your next MFC will be a doosie!) Reviewed by: peter@freebsd.org, dillon@freebsd.org X-MFC after: ha ha ha ha
2001-09-12 08:38:13 +00:00
struct thread *td = curthread;
Within ufs, the ffs_sync and ffs_fsync functions did not always check for and/or report I/O errors. The result is that a VFS_SYNC or VOP_FSYNC called with MNT_WAIT could loop infinitely on ufs in the presence of a hard error writing a disk sector or in a filesystem full condition. This patch ensures that I/O errors will always be checked and returned. This patch also ensures that every call to VFS_SYNC or VOP_FSYNC with MNT_WAIT set checks for and takes appropriate action when an error is returned. Sponsored by: DARPA & NAI Labs.
2002-10-25 00:20:37 +00:00
int error;
Add snapshots to the fast filesystem. Most of the changes support the gating of system calls that cause modifications to the underlying filesystem. The gating can be enabled by any filesystem that needs to consistently suspend operations by adding the vop_stdgetwritemount to their set of vnops. Once gating is enabled, the function vfs_write_suspend stops all new write operations to a filesystem, allows any filesystem modifying system calls already in progress to complete, then sync's the filesystem to disk and returns. The function vfs_write_resume allows the suspended write operations to begin again. Gating is not added by default for all filesystems as for SMP systems it adds two extra locks to such critical kernel paths as the write system call. Thus, gating should only be added as needed. Details on the use and current status of snapshots in FFS can be found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness is not included here. Unless and until you create a snapshot file, these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
MNT_ILOCK(mp);
Simplify the code and remove two mutex operations. MFC after: 2 weeks
2006-06-24 22:55:43 +00:00
if (mp->mnt_kern_flag & MNTK_SUSPEND) {
MNT_IUNLOCK(mp);
return (0);
}
Add snapshots to the fast filesystem. Most of the changes support the gating of system calls that cause modifications to the underlying filesystem. The gating can be enabled by any filesystem that needs to consistently suspend operations by adding the vop_stdgetwritemount to their set of vnops. Once gating is enabled, the function vfs_write_suspend stops all new write operations to a filesystem, allows any filesystem modifying system calls already in progress to complete, then sync's the filesystem to disk and returns. The function vfs_write_resume allows the suspended write operations to begin again. Gating is not added by default for all filesystems as for SMP systems it adds two extra locks to such critical kernel paths as the write system call. Thus, gating should only be added as needed. Details on the use and current status of snapshots in FFS can be found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness is not included here. Unless and until you create a snapshot file, these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
mp->mnt_kern_flag |= MNTK_SUSPEND;
if (mp->mnt_writeopcount > 0)
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
(void) msleep(&mp->mnt_writeopcount,
MNT_MTX(mp), (PUSER - 1)|PDROP, "suspwt", 0);
else
MNT_IUNLOCK(mp);
Simplify the code and remove two mutex operations. MFC after: 2 weeks
2006-06-24 22:55:43 +00:00
if ((error = VFS_SYNC(mp, MNT_SUSPEND, td)) != 0)
Within ufs, the ffs_sync and ffs_fsync functions did not always check for and/or report I/O errors. The result is that a VFS_SYNC or VOP_FSYNC called with MNT_WAIT could loop infinitely on ufs in the presence of a hard error writing a disk sector or in a filesystem full condition. This patch ensures that I/O errors will always be checked and returned. This patch also ensures that every call to VFS_SYNC or VOP_FSYNC with MNT_WAIT set checks for and takes appropriate action when an error is returned. Sponsored by: DARPA & NAI Labs.
2002-10-25 00:20:37 +00:00
vfs_write_resume(mp);
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
return (error);
Add snapshots to the fast filesystem. Most of the changes support the gating of system calls that cause modifications to the underlying filesystem. The gating can be enabled by any filesystem that needs to consistently suspend operations by adding the vop_stdgetwritemount to their set of vnops. Once gating is enabled, the function vfs_write_suspend stops all new write operations to a filesystem, allows any filesystem modifying system calls already in progress to complete, then sync's the filesystem to disk and returns. The function vfs_write_resume allows the suspended write operations to begin again. Gating is not added by default for all filesystems as for SMP systems it adds two extra locks to such critical kernel paths as the write system call. Thus, gating should only be added as needed. Details on the use and current status of snapshots in FFS can be found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness is not included here. Unless and until you create a snapshot file, these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
}
/*
* Request a filesystem to resume write operations.
*/
void
vfs_write_resume(mp)
struct mount *mp;
{
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
MNT_ILOCK(mp);
if ((mp->mnt_kern_flag & MNTK_SUSPEND) != 0) {
Block secondary writes while expunging active unlinked files. Fix detection of active unlinked files by checking VI_OWEINACT and VI_DOINGINACT in addition to v_usecount. Defer inactive handling for unlinked files if the file system is mostly suspended (secondary writes being blocked). Perform deferred inactive handling after the file system is resumed.
2006-03-11 01:08:37 +00:00
mp->mnt_kern_flag &= ~(MNTK_SUSPEND | MNTK_SUSPEND2 |
MNTK_SUSPENDED);
- Remove GIANT_REQUIRED where giant is no longer required. - Protect access to mnt_kern_flag with the mountpoint mutex. - Use the appropriate nd flags to deal with giant in vn_open_cred(). We currently determine whether the caller is mpsafe by checking for a valid fdidx. Any caller coming from user-space is now mpsafe and supplies a valid fd. No kenrel callers have been converted to mpsafe, so this check is sufficient for now. - Use VFS_LOCK_GIANT instead of manual giant acquisition where appropriate. Sponsored By: Isilon Systems, Inc.
2005-01-24 10:31:42 +00:00
wakeup(&mp->mnt_writeopcount);
wakeup(&mp->mnt_flag);
}
MNT_IUNLOCK(mp);
Add snapshots to the fast filesystem. Most of the changes support the gating of system calls that cause modifications to the underlying filesystem. The gating can be enabled by any filesystem that needs to consistently suspend operations by adding the vop_stdgetwritemount to their set of vnops. Once gating is enabled, the function vfs_write_suspend stops all new write operations to a filesystem, allows any filesystem modifying system calls already in progress to complete, then sync's the filesystem to disk and returns. The function vfs_write_resume allows the suspended write operations to begin again. Gating is not added by default for all filesystems as for SMP systems it adds two extra locks to such critical kernel paths as the write system call. Thus, gating should only be added as needed. Details on the use and current status of snapshots in FFS can be found in /sys/ufs/ffs/README.snapshot so for brevity and timelyness is not included here. Unless and until you create a snapshot file, these changes should have no effect on your system (famous last words).
2000-07-11 22:07:57 +00:00
}
Document all functions, global and static variables, and sysctls. Includes some minor whitespace changes, and re-ordering to be able to document properly (e.g, grouping of variables and the SYSCTL macro calls for them, where the documentation has been added.) Reviewed by: phk (but all errors are mine)
2002-03-05 15:38:49 +00:00
/*
* Implement kqueues for files by translating it to vnode operation.
*/
Introduce kqueue() and kevent(), a kernel event notification facility.
2000-04-16 18:53:38 +00:00
static int
Extend kqueue down to the device layer. Backwards compatible approach suggested by: peter
2001-02-15 16:34:11 +00:00
vn_kqfilter(struct file *fp, struct knote *kn)
Introduce kqueue() and kevent(), a kernel event notification facility.
2000-04-16 18:53:38 +00:00
{
- Remove explicit giant acquires and replace it with VFS_LOCK_GIANT. Sponsored by: Isilon Systems, Inc.
2006-03-22 00:00:05 +00:00
int vfslocked;
Add locking to the kqueue subsystem. This also makes the kqueue subsystem a more complete subsystem, and removes the knowlege of how things are implemented from the drivers. Include locking around filter ops, so a module like aio will know when not to be unloaded if there are outstanding knotes using it's filter ops. Currently, it uses the MTX_DUPOK even though it is not always safe to aquire duplicate locks. Witness currently doesn't support the ability to discover if a dup lock is ok (in some cases). Reviewed by: green, rwatson (both earlier versions)
2004-08-15 06:24:42 +00:00
int error;
Introduce kqueue() and kevent(), a kernel event notification facility.
2000-04-16 18:53:38 +00:00
- Remove explicit giant acquires and replace it with VFS_LOCK_GIANT. Sponsored by: Isilon Systems, Inc.
2006-03-22 00:00:05 +00:00
vfslocked = VFS_LOCK_GIANT(fp->f_vnode->v_mount);
Add locking to the kqueue subsystem. This also makes the kqueue subsystem a more complete subsystem, and removes the knowlege of how things are implemented from the drivers. Include locking around filter ops, so a module like aio will know when not to be unloaded if there are outstanding knotes using it's filter ops. Currently, it uses the MTX_DUPOK even though it is not always safe to aquire duplicate locks. Witness currently doesn't support the ability to discover if a dup lock is ok (in some cases). Reviewed by: green, rwatson (both earlier versions)
2004-08-15 06:24:42 +00:00
error = VOP_KQFILTER(fp->f_vnode, kn);
- Remove explicit giant acquires and replace it with VFS_LOCK_GIANT. Sponsored by: Isilon Systems, Inc.
2006-03-22 00:00:05 +00:00
VFS_UNLOCK_GIANT(vfslocked);
Flag a broad range of VFS operations as GIANT_REQUIRED in order to catch leaking into VFS without Giant. Inch Giant a little lower in several file descriptor operations on vnodes to cover only VFS operations that need it, rather than file flag reading, etc.
2004-08-06 22:25:35 +00:00
Add locking to the kqueue subsystem. This also makes the kqueue subsystem a more complete subsystem, and removes the knowlege of how things are implemented from the drivers. Include locking around filter ops, so a module like aio will know when not to be unloaded if there are outstanding knotes using it's filter ops. Currently, it uses the MTX_DUPOK even though it is not always safe to aquire duplicate locks. Witness currently doesn't support the ability to discover if a dup lock is ok (in some cases). Reviewed by: green, rwatson (both earlier versions)
2004-08-15 06:24:42 +00:00
return error;
Introduce kqueue() and kevent(), a kernel event notification facility.
2000-04-16 18:53:38 +00:00
}
o Introduce vn_extattr_{get,set}, wrapper routines for VOP_GETEXTATTR and VOP_SETEXTATTR to simplify calling from in-kernel consumers, such as capability code. Both accept a vnode (optionally locked, with ioflg to indicate that), attribute name, and a buffer + buffer length in UIO_SYSSPACE. Both authorize the call as a kernel request, with cred set to NULL for the actual VOP_ calls. Obtained from: TrustedBSD Project
2000-08-08 17:15:32 +00:00
/*
* Simplified in-kernel wrapper calls for extended attribute access.
* Both calls pass in a NULL credential, authorizing as "kernel" access.
* Set IO_NODELOCKED in ioflg if the vnode is already locked.
*/
int
o Rename "namespace" argument to "attrnamespace" as namespace is a C++ reserved word. Submitted by: jkh Obtained from: TrustedBSD Project
2001-03-19 05:44:15 +00:00
vn_extattr_get(struct vnode *vp, int ioflg, int attrnamespace,
KSE Milestone 2 Note ALL MODULES MUST BE RECOMPILED make the kernel aware that there are smaller units of scheduling than the process. (but only allow one thread per process at this time). This is functionally equivalent to teh previousl -current except that there is a thread associated with each process. Sorry john! (your next MFC will be a doosie!) Reviewed by: peter@freebsd.org, dillon@freebsd.org X-MFC after: ha ha ha ha
2001-09-12 08:38:13 +00:00
const char *attrname, int *buflen, char *buf, struct thread *td)
o Introduce vn_extattr_{get,set}, wrapper routines for VOP_GETEXTATTR and VOP_SETEXTATTR to simplify calling from in-kernel consumers, such as capability code. Both accept a vnode (optionally locked, with ioflg to indicate that), attribute name, and a buffer + buffer length in UIO_SYSSPACE. Both authorize the call as a kernel request, with cred set to NULL for the actual VOP_ calls. Obtained from: TrustedBSD Project
2000-08-08 17:15:32 +00:00
{
struct uio auio;
struct iovec iov;
int error;
iov.iov_len = *buflen;
iov.iov_base = buf;
auio.uio_iov = &iov;
auio.uio_iovcnt = 1;
auio.uio_rw = UIO_READ;
auio.uio_segflg = UIO_SYSSPACE;
KSE Milestone 2 Note ALL MODULES MUST BE RECOMPILED make the kernel aware that there are smaller units of scheduling than the process. (but only allow one thread per process at this time). This is functionally equivalent to teh previousl -current except that there is a thread associated with each process. Sorry john! (your next MFC will be a doosie!) Reviewed by: peter@freebsd.org, dillon@freebsd.org X-MFC after: ha ha ha ha
2001-09-12 08:38:13 +00:00
auio.uio_td = td;
o Introduce vn_extattr_{get,set}, wrapper routines for VOP_GETEXTATTR and VOP_SETEXTATTR to simplify calling from in-kernel consumers, such as capability code. Both accept a vnode (optionally locked, with ioflg to indicate that), attribute name, and a buffer + buffer length in UIO_SYSSPACE. Both authorize the call as a kernel request, with cred set to NULL for the actual VOP_ calls. Obtained from: TrustedBSD Project
2000-08-08 17:15:32 +00:00
auio.uio_offset = 0;
auio.uio_resid = *buflen;
if ((ioflg & IO_NODELOCKED) == 0)
KSE Milestone 2 Note ALL MODULES MUST BE RECOMPILED make the kernel aware that there are smaller units of scheduling than the process. (but only allow one thread per process at this time). This is functionally equivalent to teh previousl -current except that there is a thread associated with each process. Sorry john! (your next MFC will be a doosie!) Reviewed by: peter@freebsd.org, dillon@freebsd.org X-MFC after: ha ha ha ha
2001-09-12 08:38:13 +00:00
vn_lock(vp, LK_EXCLUSIVE | LK_RETRY, td);
o Introduce vn_extattr_{get,set}, wrapper routines for VOP_GETEXTATTR and VOP_SETEXTATTR to simplify calling from in-kernel consumers, such as capability code. Both accept a vnode (optionally locked, with ioflg to indicate that), attribute name, and a buffer + buffer length in UIO_SYSSPACE. Both authorize the call as a kernel request, with cred set to NULL for the actual VOP_ calls. Obtained from: TrustedBSD Project
2000-08-08 17:15:32 +00:00
Add locking assertions into vn_extattr_set, vn_extattr_get and vn_extattr_rm. This is meant to catch conditions where IO_NODELOCKED has been specified without the vnode being locked. Discussed with: rwatson MFC after: 1 week
2005-02-24 00:13:16 +00:00
ASSERT_VOP_LOCKED(vp, "IO_NODELOCKED with no vp lock held");
o Introduce vn_extattr_{get,set}, wrapper routines for VOP_GETEXTATTR and VOP_SETEXTATTR to simplify calling from in-kernel consumers, such as capability code. Both accept a vnode (optionally locked, with ioflg to indicate that), attribute name, and a buffer + buffer length in UIO_SYSSPACE. Both authorize the call as a kernel request, with cred set to NULL for the actual VOP_ calls. Obtained from: TrustedBSD Project
2000-08-08 17:15:32 +00:00
/* authorize attribute retrieval as kernel */
Part I: Update extended attribute API and ABI: o Modify the system call syntax for extattr_{get,set}_{fd,file}() so as not to use the scatter gather API (which appeared not to be used by any consumers, and be less portable), rather, accepts 'data' and 'nbytes' in the style of other simple read/write interfaces. This changes the API and ABI. o Modify system call semantics so that extattr_get_{fd,file}() return a size_t. When performing a read, the number of bytes read will be returned, unless the data pointer is NULL, in which case the number of bytes of data are returned. This changes the API only. o Modify the VOP_GETEXTATTR() vnode operation to accept a *size_t argument so as to return the size, if desirable. If set to NULL, the size will not be returned. o Update various filesystems (pseodofs, ufs) to DTRT. These changes should make extended attributes more useful and more portable. More commits to rebuild the system call files, as well as update userland utilities to follow. Obtained from: TrustedBSD Project Sponsored by: DARPA, NAI Labs
2002-02-10 04:43:22 +00:00
error = VOP_GETEXTATTR(vp, attrnamespace, attrname, &auio, NULL, NULL,
td);
o Introduce vn_extattr_{get,set}, wrapper routines for VOP_GETEXTATTR and VOP_SETEXTATTR to simplify calling from in-kernel consumers, such as capability code. Both accept a vnode (optionally locked, with ioflg to indicate that), attribute name, and a buffer + buffer length in UIO_SYSSPACE. Both authorize the call as a kernel request, with cred set to NULL for the actual VOP_ calls. Obtained from: TrustedBSD Project
2000-08-08 17:15:32 +00:00
if ((ioflg & IO_NODELOCKED) == 0)
KSE Milestone 2 Note ALL MODULES MUST BE RECOMPILED make the kernel aware that there are smaller units of scheduling than the process. (but only allow one thread per process at this time). This is functionally equivalent to teh previousl -current except that there is a thread associated with each process. Sorry john! (your next MFC will be a doosie!) Reviewed by: peter@freebsd.org, dillon@freebsd.org X-MFC after: ha ha ha ha
2001-09-12 08:38:13 +00:00
VOP_UNLOCK(vp, 0, td);
o Introduce vn_extattr_{get,set}, wrapper routines for VOP_GETEXTATTR and VOP_SETEXTATTR to simplify calling from in-kernel consumers, such as capability code. Both accept a vnode (optionally locked, with ioflg to indicate that), attribute name, and a buffer + buffer length in UIO_SYSSPACE. Both authorize the call as a kernel request, with cred set to NULL for the actual VOP_ calls. Obtained from: TrustedBSD Project
2000-08-08 17:15:32 +00:00
if (error == 0) {
*buflen = *buflen - auio.uio_resid;
}
return (error);
}
/*
* XXX failure mode if partially written?
*/
int
o Rename "namespace" argument to "attrnamespace" as namespace is a C++ reserved word. Submitted by: jkh Obtained from: TrustedBSD Project
2001-03-19 05:44:15 +00:00
vn_extattr_set(struct vnode *vp, int ioflg, int attrnamespace,
KSE Milestone 2 Note ALL MODULES MUST BE RECOMPILED make the kernel aware that there are smaller units of scheduling than the process. (but only allow one thread per process at this time). This is functionally equivalent to teh previousl -current except that there is a thread associated with each process. Sorry john! (your next MFC will be a doosie!) Reviewed by: peter@freebsd.org, dillon@freebsd.org X-MFC after: ha ha ha ha
2001-09-12 08:38:13 +00:00
const char *attrname, int buflen, char *buf, struct thread *td)
o Introduce vn_extattr_{get,set}, wrapper routines for VOP_GETEXTATTR and VOP_SETEXTATTR to simplify calling from in-kernel consumers, such as capability code. Both accept a vnode (optionally locked, with ioflg to indicate that), attribute name, and a buffer + buffer length in UIO_SYSSPACE. Both authorize the call as a kernel request, with cred set to NULL for the actual VOP_ calls. Obtained from: TrustedBSD Project
2000-08-08 17:15:32 +00:00
{
struct uio auio;
struct iovec iov;
o vn_extattr_set() will now call appropriate vn_start_write() and vn_finished_write() if IO_NODELOCKED is not set. Obtained from: TrustedBSD Project
2000-09-05 03:15:02 +00:00
struct mount *mp;
o Introduce vn_extattr_{get,set}, wrapper routines for VOP_GETEXTATTR and VOP_SETEXTATTR to simplify calling from in-kernel consumers, such as capability code. Both accept a vnode (optionally locked, with ioflg to indicate that), attribute name, and a buffer + buffer length in UIO_SYSSPACE. Both authorize the call as a kernel request, with cred set to NULL for the actual VOP_ calls. Obtained from: TrustedBSD Project
2000-08-08 17:15:32 +00:00
int error;
iov.iov_len = buflen;
iov.iov_base = buf;
auio.uio_iov = &iov;
auio.uio_iovcnt = 1;
auio.uio_rw = UIO_WRITE;
auio.uio_segflg = UIO_SYSSPACE;
KSE Milestone 2 Note ALL MODULES MUST BE RECOMPILED make the kernel aware that there are smaller units of scheduling than the process. (but only allow one thread per process at this time). This is functionally equivalent to teh previousl -current except that there is a thread associated with each process. Sorry john! (your next MFC will be a doosie!) Reviewed by: peter@freebsd.org, dillon@freebsd.org X-MFC after: ha ha ha ha
2001-09-12 08:38:13 +00:00
auio.uio_td = td;
o Introduce vn_extattr_{get,set}, wrapper routines for VOP_GETEXTATTR and VOP_SETEXTATTR to simplify calling from in-kernel consumers, such as capability code. Both accept a vnode (optionally locked, with ioflg to indicate that), attribute name, and a buffer + buffer length in UIO_SYSSPACE. Both authorize the call as a kernel request, with cred set to NULL for the actual VOP_ calls. Obtained from: TrustedBSD Project
2000-08-08 17:15:32 +00:00
auio.uio_offset = 0;
auio.uio_resid = buflen;
o vn_extattr_set() will now call appropriate vn_start_write() and vn_finished_write() if IO_NODELOCKED is not set. Obtained from: TrustedBSD Project
2000-09-05 03:15:02 +00:00
if ((ioflg & IO_NODELOCKED) == 0) {
if ((error = vn_start_write(vp, &mp, V_WAIT)) != 0)
return (error);
KSE Milestone 2 Note ALL MODULES MUST BE RECOMPILED make the kernel aware that there are smaller units of scheduling than the process. (but only allow one thread per process at this time). This is functionally equivalent to teh previousl -current except that there is a thread associated with each process. Sorry john! (your next MFC will be a doosie!) Reviewed by: peter@freebsd.org, dillon@freebsd.org X-MFC after: ha ha ha ha
2001-09-12 08:38:13 +00:00
vn_lock(vp, LK_EXCLUSIVE | LK_RETRY, td);
o vn_extattr_set() will now call appropriate vn_start_write() and vn_finished_write() if IO_NODELOCKED is not set. Obtained from: TrustedBSD Project
2000-09-05 03:15:02 +00:00
}
o Introduce vn_extattr_{get,set}, wrapper routines for VOP_GETEXTATTR and VOP_SETEXTATTR to simplify calling from in-kernel consumers, such as capability code. Both accept a vnode (optionally locked, with ioflg to indicate that), attribute name, and a buffer + buffer length in UIO_SYSSPACE. Both authorize the call as a kernel request, with cred set to NULL for the actual VOP_ calls. Obtained from: TrustedBSD Project
2000-08-08 17:15:32 +00:00
Add locking assertions into vn_extattr_set, vn_extattr_get and vn_extattr_rm. This is meant to catch conditions where IO_NODELOCKED has been specified without the vnode being locked. Discussed with: rwatson MFC after: 1 week
2005-02-24 00:13:16 +00:00
ASSERT_VOP_LOCKED(vp, "IO_NODELOCKED with no vp lock held");
o Introduce vn_extattr_{get,set}, wrapper routines for VOP_GETEXTATTR and VOP_SETEXTATTR to simplify calling from in-kernel consumers, such as capability code. Both accept a vnode (optionally locked, with ioflg to indicate that), attribute name, and a buffer + buffer length in UIO_SYSSPACE. Both authorize the call as a kernel request, with cred set to NULL for the actual VOP_ calls. Obtained from: TrustedBSD Project
2000-08-08 17:15:32 +00:00
/* authorize attribute setting as kernel */
KSE Milestone 2 Note ALL MODULES MUST BE RECOMPILED make the kernel aware that there are smaller units of scheduling than the process. (but only allow one thread per process at this time). This is functionally equivalent to teh previousl -current except that there is a thread associated with each process. Sorry john! (your next MFC will be a doosie!) Reviewed by: peter@freebsd.org, dillon@freebsd.org X-MFC after: ha ha ha ha
2001-09-12 08:38:13 +00:00
error = VOP_SETEXTATTR(vp, attrnamespace, attrname, &auio, NULL, td);
o Introduce vn_extattr_{get,set}, wrapper routines for VOP_GETEXTATTR and VOP_SETEXTATTR to simplify calling from in-kernel consumers, such as capability code. Both accept a vnode (optionally locked, with ioflg to indicate that), attribute name, and a buffer + buffer length in UIO_SYSSPACE. Both authorize the call as a kernel request, with cred set to NULL for the actual VOP_ calls. Obtained from: TrustedBSD Project
2000-08-08 17:15:32 +00:00
o vn_extattr_set() will now call appropriate vn_start_write() and vn_finished_write() if IO_NODELOCKED is not set. Obtained from: TrustedBSD Project
2000-09-05 03:15:02 +00:00
if ((ioflg & IO_NODELOCKED) == 0) {
vn_finished_write(mp);
KSE Milestone 2 Note ALL MODULES MUST BE RECOMPILED make the kernel aware that there are smaller units of scheduling than the process. (but only allow one thread per process at this time). This is functionally equivalent to teh previousl -current except that there is a thread associated with each process. Sorry john! (your next MFC will be a doosie!) Reviewed by: peter@freebsd.org, dillon@freebsd.org X-MFC after: ha ha ha ha
2001-09-12 08:38:13 +00:00
VOP_UNLOCK(vp, 0, td);
o vn_extattr_set() will now call appropriate vn_start_write() and vn_finished_write() if IO_NODELOCKED is not set. Obtained from: TrustedBSD Project
2000-09-05 03:15:02 +00:00
}
o Introduce vn_extattr_{get,set}, wrapper routines for VOP_GETEXTATTR and VOP_SETEXTATTR to simplify calling from in-kernel consumers, such as capability code. Both accept a vnode (optionally locked, with ioflg to indicate that), attribute name, and a buffer + buffer length in UIO_SYSSPACE. Both authorize the call as a kernel request, with cred set to NULL for the actual VOP_ calls. Obtained from: TrustedBSD Project
2000-08-08 17:15:32 +00:00
return (error);
}
o Introduce vn_extattr_rm(), a helper function in the style of vn_extattr_get() and vn_extattr_set(). vn_extattr_rm() removes the specified extended attribute from a vnode, authorizing the change as the kernel (NULL cred). Obtained from: TrustedBSD Project
2000-09-22 22:33:13 +00:00
int
o Rename "namespace" argument to "attrnamespace" as namespace is a C++ reserved word. Submitted by: jkh Obtained from: TrustedBSD Project
2001-03-19 05:44:15 +00:00
vn_extattr_rm(struct vnode *vp, int ioflg, int attrnamespace,
KSE Milestone 2 Note ALL MODULES MUST BE RECOMPILED make the kernel aware that there are smaller units of scheduling than the process. (but only allow one thread per process at this time). This is functionally equivalent to teh previousl -current except that there is a thread associated with each process. Sorry john! (your next MFC will be a doosie!) Reviewed by: peter@freebsd.org, dillon@freebsd.org X-MFC after: ha ha ha ha
2001-09-12 08:38:13 +00:00
const char *attrname, struct thread *td)
o Introduce vn_extattr_rm(), a helper function in the style of vn_extattr_get() and vn_extattr_set(). vn_extattr_rm() removes the specified extended attribute from a vnode, authorizing the change as the kernel (NULL cred). Obtained from: TrustedBSD Project
2000-09-22 22:33:13 +00:00
{
struct mount *mp;
int error;
if ((ioflg & IO_NODELOCKED) == 0) {
if ((error = vn_start_write(vp, &mp, V_WAIT)) != 0)
return (error);
KSE Milestone 2 Note ALL MODULES MUST BE RECOMPILED make the kernel aware that there are smaller units of scheduling than the process. (but only allow one thread per process at this time). This is functionally equivalent to teh previousl -current except that there is a thread associated with each process. Sorry john! (your next MFC will be a doosie!) Reviewed by: peter@freebsd.org, dillon@freebsd.org X-MFC after: ha ha ha ha
2001-09-12 08:38:13 +00:00
vn_lock(vp, LK_EXCLUSIVE | LK_RETRY, td);
o Introduce vn_extattr_rm(), a helper function in the style of vn_extattr_get() and vn_extattr_set(). vn_extattr_rm() removes the specified extended attribute from a vnode, authorizing the change as the kernel (NULL cred). Obtained from: TrustedBSD Project
2000-09-22 22:33:13 +00:00
}
Add locking assertions into vn_extattr_set, vn_extattr_get and vn_extattr_rm. This is meant to catch conditions where IO_NODELOCKED has been specified without the vnode being locked. Discussed with: rwatson MFC after: 1 week
2005-02-24 00:13:16 +00:00
ASSERT_VOP_LOCKED(vp, "IO_NODELOCKED with no vp lock held");
o Introduce vn_extattr_rm(), a helper function in the style of vn_extattr_get() and vn_extattr_set(). vn_extattr_rm() removes the specified extended attribute from a vnode, authorizing the change as the kernel (NULL cred). Obtained from: TrustedBSD Project
2000-09-22 22:33:13 +00:00
/* authorize attribute removal as kernel */
Rename VOP_RMEXTATTR() to VOP_DELETEEXTATTR() for consistency with the kernel ACL interfaces and system call names. Break out UFS2 and FFS extattr delete and list vnode operations from setextattr and getextattr to deleteextattr and listextattr, which cleans up the implementations, and makes the results more readable, and makes the APIs more clear. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-07-28 18:53:29 +00:00
error = VOP_DELETEEXTATTR(vp, attrnamespace, attrname, NULL, td);
Prefer the vop_rmextattr() vnode operation for removing extended attributes from objects over vop_setextattr() with a NULL uio; if the file system doesn't support the vop_rmextattr() method, fall back to the vop_setextattr() method. Obtained from: TrustedBSD Project Sponsored by: DARPA, Network Associates Laboratories
2003-06-22 23:03:07 +00:00
if (error == EOPNOTSUPP)
error = VOP_SETEXTATTR(vp, attrnamespace, attrname, NULL,
NULL, td);
o Introduce vn_extattr_rm(), a helper function in the style of vn_extattr_get() and vn_extattr_set(). vn_extattr_rm() removes the specified extended attribute from a vnode, authorizing the change as the kernel (NULL cred). Obtained from: TrustedBSD Project
2000-09-22 22:33:13 +00:00
if ((ioflg & IO_NODELOCKED) == 0) {
vn_finished_write(mp);
KSE Milestone 2 Note ALL MODULES MUST BE RECOMPILED make the kernel aware that there are smaller units of scheduling than the process. (but only allow one thread per process at this time). This is functionally equivalent to teh previousl -current except that there is a thread associated with each process. Sorry john! (your next MFC will be a doosie!) Reviewed by: peter@freebsd.org, dillon@freebsd.org X-MFC after: ha ha ha ha
2001-09-12 08:38:13 +00:00
VOP_UNLOCK(vp, 0, td);
o Introduce vn_extattr_rm(), a helper function in the style of vn_extattr_get() and vn_extattr_set(). vn_extattr_rm() removes the specified extended attribute from a vnode, authorizing the change as the kernel (NULL cred). Obtained from: TrustedBSD Project
2000-09-22 22:33:13 +00:00
}
return (error);
}
Reference in New Issue Copy Permalink