freebsd-skq/sys/coda/coda_subr.c
1998-09-02 19:09:53 +00:00

737 lines
20 KiB
C

/*
*
* Coda: an Experimental Distributed File System
* Release 3.1
*
* Copyright (c) 1987-1998 Carnegie Mellon University
* All Rights Reserved
*
* Permission to use, copy, modify and distribute this software and its
* documentation is hereby granted, provided that both the copyright
* notice and this permission notice appear in all copies of the
* software, derivative works or modified versions, and any portions
* thereof, and that both notices appear in supporting documentation, and
* that credit is given to Carnegie Mellon University in all documents
* and publicity pertaining to direct or indirect use of this code or its
* derivatives.
*
* CODA IS AN EXPERIMENTAL SOFTWARE SYSTEM AND IS KNOWN TO HAVE BUGS,
* SOME OF WHICH MAY HAVE SERIOUS CONSEQUENCES. CARNEGIE MELLON ALLOWS
* FREE USE OF THIS SOFTWARE IN ITS "AS IS" CONDITION. CARNEGIE MELLON
* DISCLAIMS ANY LIABILITY OF ANY KIND FOR ANY DAMAGES WHATSOEVER
* RESULTING DIRECTLY OR INDIRECTLY FROM THE USE OF THIS SOFTWARE OR OF
* ANY DERIVATIVE WORK.
*
* Carnegie Mellon encourages users of this software to return any
* improvements or extensions that they make, and to grant Carnegie
* Mellon the rights to redistribute these changes without encumbrance.
*
* @(#) src/sys/cfs/cfs_subr.c,v 1.1.1.1 1998/08/29 21:14:52 rvb Exp $
* $Id: $
*
*/
/*
* Mach Operating System
* Copyright (c) 1989 Carnegie-Mellon University
* All rights reserved. The CMU software License Agreement specifies
* the terms and conditions for use and redistribution.
*/
/*
* This code was written for the Coda file system at Carnegie Mellon
* University. Contributers include David Steere, James Kistler, and
* M. Satyanarayanan. */
/*
* HISTORY
* $Log: cfs_subr.c,v $
* Revision 1.1.1.1 1998/08/29 21:14:52 rvb
* Very Preliminary Coda
*
* Revision 1.11 1998/08/28 18:12:18 rvb
* Now it also works on FreeBSD -current. This code will be
* committed to the FreeBSD -current and NetBSD -current
* trees. It will then be tailored to the particular platform
* by flushing conditional code.
*
* Revision 1.10 1998/08/18 17:05:16 rvb
* Don't use __RCSID now
*
* Revision 1.9 1998/08/18 16:31:41 rvb
* Sync the code for NetBSD -current; test on 1.3 later
*
* Revision 1.8 98/01/31 20:53:12 rvb
* First version that works on FreeBSD 2.2.5
*
* Revision 1.7 98/01/23 11:53:42 rvb
* Bring RVB_CFS1_1 to HEAD
*
* Revision 1.6.2.3 98/01/23 11:21:05 rvb
* Sync with 2.2.5
*
* Revision 1.6.2.2 97/12/16 12:40:06 rvb
* Sync with 1.3
*
* Revision 1.6.2.1 97/12/06 17:41:21 rvb
* Sync with peters coda.h
*
* Revision 1.6 97/12/05 10:39:17 rvb
* Read CHANGES
*
* Revision 1.5.4.8 97/11/26 15:28:58 rvb
* Cant make downcall pbuf == union cfs_downcalls yet
*
* Revision 1.5.4.7 97/11/20 11:46:42 rvb
* Capture current cfs_venus
*
* Revision 1.5.4.6 97/11/18 10:27:16 rvb
* cfs_nbsd.c is DEAD!!!; integrated into cfs_vf/vnops.c
* cfs_nb_foo and cfs_foo are joined
*
* Revision 1.5.4.5 97/11/13 22:03:00 rvb
* pass2 cfs_NetBSD.h mt
*
* Revision 1.5.4.4 97/11/12 12:09:39 rvb
* reorg pass1
*
* Revision 1.5.4.3 97/11/06 21:02:38 rvb
* first pass at ^c ^z
*
* Revision 1.5.4.2 97/10/29 16:06:27 rvb
* Kill DYING
*
* Revision 1.5.4.1 97/10/28 23:10:16 rvb
* >64Meg; venus can be killed!
*
* Revision 1.5 97/08/05 11:08:17 lily
* Removed cfsnc_replace, replaced it with a cfs_find, unhash, and
* rehash. This fixes a cnode leak and a bug in which the fid is
* not actually replaced. (cfs_namecache.c, cfsnc.h, cfs_subr.c)
*
* Revision 1.4 96/12/12 22:10:59 bnoble
* Fixed the "downcall invokes venus operation" deadlock in all known cases.
* There may be more
*
* Revision 1.3 1996/12/05 16:20:15 bnoble
* Minor debugging aids
*
* Revision 1.2 1996/01/02 16:57:01 bnoble
* Added support for Coda MiniCache and raw inode calls (final commit)
*
* Revision 1.1.2.1 1995/12/20 01:57:27 bnoble
* Added CFS-specific files
*
* Revision 3.1.1.1 1995/03/04 19:07:59 bnoble
* Branch for NetBSD port revisions
*
* Revision 3.1 1995/03/04 19:07:58 bnoble
* Bump to major revision 3 to prepare for NetBSD port
*
* Revision 2.8 1995/03/03 17:00:04 dcs
* Fixed kernel bug involving sleep and upcalls. Basically if you killed
* a job waiting on venus, the venus upcall queues got trashed. Depending
* on luck, you could kill the kernel or not.
* (mods to cfs_subr.c and cfs_mach.d)
*
* Revision 2.7 95/03/02 22:45:21 dcs
* Sun4 compatibility
*
* Revision 2.6 95/02/17 16:25:17 dcs
* These versions represent several changes:
* 1. Allow venus to restart even if outstanding references exist.
* 2. Have only one ctlvp per client, as opposed to one per mounted cfs device.d
* 3. Allow ody_expand to return many members, not just one.
*
* Revision 2.5 94/11/09 15:56:26 dcs
* Had the thread sleeping on the wrong thing!
*
* Revision 2.4 94/10/14 09:57:57 dcs
* Made changes 'cause sun4s have braindead compilers
*
* Revision 2.3 94/10/12 16:46:26 dcs
* Cleaned kernel/venus interface by removing XDR junk, plus
* so cleanup to allow this code to be more easily ported.
*
* Revision 1.2 92/10/27 17:58:22 lily
* merge kernel/latest and alpha/src/cfs
*
* Revision 2.4 92/09/30 14:16:26 mja
* Incorporated Dave Steere's fix for the GNU-Emacs bug.
* Also, included his cfs_flush routine in place of the former cfsnc_flush.
* [91/02/07 jjk]
*
* Added contributors blurb.
* [90/12/13 jjk]
*
* Hack to allow users to keep coda venus calls uninterruptible. THis
* basically prevents the Gnu-emacs bug from appearing, in which a call
* was being interrupted, and return EINTR, but gnu didn't check for the
* error and figured the file was buggered.
* [90/12/09 dcs]
*
* Revision 2.3 90/08/10 10:23:20 mrt
* Removed include of vm/vm_page.h as it no longer exists.
* [90/08/10 mrt]
*
* Revision 2.2 90/07/05 11:26:35 mrt
* Initialize name cache on first call to vcopen.
* [90/05/23 dcs]
*
* Created for the Coda File System.
* [90/05/23 dcs]
*
* Revision 1.5 90/05/31 17:01:35 dcs
* Prepare for merge with facilities kernel.
*
* Revision 1.2 90/03/19 15:56:25 dcs
* Initialize name cache on first call to vcopen.
*
* Revision 1.1 90/03/15 10:43:26 jjk
* Initial revision
*
*/
/* NOTES: rvb
* 1. Added cfs_unmounting to mark all cnodes as being UNMOUNTING. This has to
* be done before dounmount is called. Because some of the routines that
* dounmount calls before cfs_unmounted might try to force flushes to venus.
* The vnode pager does this.
* 2. cfs_unmounting marks all cnodes scanning cfs_cache.
* 3. cfs_checkunmounting (under DEBUG) checks all cnodes by chasing the vnodes
* under the /coda mount point.
* 4. cfs_cacheprint (under DEBUG) prints names with vnode/cnode address
*/
#include <vcfs.h>
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/proc.h>
#include <sys/malloc.h>
#include <sys/select.h>
#include <sys/mount.h>
#include <cfs/coda.h>
#include <cfs/cnode.h>
#include <cfs/cfs_subr.h>
#include <cfs/cfsnc.h>
int cfs_active = 0;
int cfs_reuse = 0;
int cfs_new = 0;
struct cnode *cfs_freelist = NULL;
struct cnode *cfs_cache[CFS_CACHESIZE];
#define cfshash(fid) (((fid)->Volume + (fid)->Vnode) & (CFS_CACHESIZE-1))
#define CNODE_NEXT(cp) ((cp)->c_next)
#define ODD(vnode) ((vnode) & 0x1)
/*
* Allocate a cnode.
*/
struct cnode *
cfs_alloc(void)
{
struct cnode *cp;
if (cfs_freelist) {
cp = cfs_freelist;
cfs_freelist = CNODE_NEXT(cp);
cfs_reuse++;
}
else {
CFS_ALLOC(cp, struct cnode *, sizeof(struct cnode));
/* NetBSD vnodes don't have any Pager info in them ('cause there are
no external pagers, duh!) */
#define VNODE_VM_INFO_INIT(vp) /* MT */
VNODE_VM_INFO_INIT(CTOV(cp));
cfs_new++;
}
bzero(cp, sizeof (struct cnode));
return(cp);
}
/*
* Deallocate a cnode.
*/
void
cfs_free(cp)
register struct cnode *cp;
{
CNODE_NEXT(cp) = cfs_freelist;
cfs_freelist = cp;
}
/*
* Put a cnode in the hash table
*/
void
cfs_save(cp)
struct cnode *cp;
{
CNODE_NEXT(cp) = cfs_cache[cfshash(&cp->c_fid)];
cfs_cache[cfshash(&cp->c_fid)] = cp;
}
/*
* Remove a cnode from the hash table
*/
void
cfs_unsave(cp)
struct cnode *cp;
{
struct cnode *ptr;
struct cnode *ptrprev = NULL;
ptr = cfs_cache[cfshash(&cp->c_fid)];
while (ptr != NULL) {
if (ptr == cp) {
if (ptrprev == NULL) {
cfs_cache[cfshash(&cp->c_fid)]
= CNODE_NEXT(ptr);
} else {
CNODE_NEXT(ptrprev) = CNODE_NEXT(ptr);
}
CNODE_NEXT(cp) = (struct cnode *)NULL;
return;
}
ptrprev = ptr;
ptr = CNODE_NEXT(ptr);
}
}
/*
* Lookup a cnode by fid. If the cnode is dying, it is bogus so skip it.
* NOTE: this allows multiple cnodes with same fid -- dcs 1/25/95
*/
struct cnode *
cfs_find(fid)
ViceFid *fid;
{
struct cnode *cp;
cp = cfs_cache[cfshash(fid)];
while (cp) {
if ((cp->c_fid.Vnode == fid->Vnode) &&
(cp->c_fid.Volume == fid->Volume) &&
(cp->c_fid.Unique == fid->Unique) &&
(!IS_UNMOUNTING(cp)))
{
cfs_active++;
return(cp);
}
cp = CNODE_NEXT(cp);
}
return(NULL);
}
/*
* cfs_kill is called as a side effect to vcopen. To prevent any
* cnodes left around from an earlier run of a venus or warden from
* causing problems with the new instance, mark any outstanding cnodes
* as dying. Future operations on these cnodes should fail (excepting
* cfs_inactive of course!). Since multiple venii/wardens can be
* running, only kill the cnodes for a particular entry in the
* cfs_mnttbl. -- DCS 12/1/94 */
int
cfs_kill(whoIam, dcstat)
struct mount *whoIam;
enum dc_status dcstat;
{
int hash, count = 0;
struct cnode *cp;
/*
* Algorithm is as follows:
* Second, flush whatever vnodes we can from the name cache.
*
* Finally, step through whatever is left and mark them dying.
* This prevents any operation at all.
*/
/* This is slightly overkill, but should work. Eventually it'd be
* nice to only flush those entries from the namecache that
* reference a vnode in this vfs. */
cfsnc_flush(dcstat);
for (hash = 0; hash < CFS_CACHESIZE; hash++) {
for (cp = cfs_cache[hash]; cp != NULL; cp = CNODE_NEXT(cp)) {
if (CTOV(cp)->v_mount == whoIam) {
#ifdef DEBUG
printf("cfs_kill: vp %p, cp %p\n", CTOV(cp), cp);
#endif
count++;
CFSDEBUG(CFS_FLUSH,
myprintf(("Live cnode fid %lx.%lx.%lx flags %d count %d\n",
(cp->c_fid).Volume,
(cp->c_fid).Vnode,
(cp->c_fid).Unique,
cp->c_flags,
CTOV(cp)->v_usecount)); );
}
}
}
return count;
}
/*
* There are two reasons why a cnode may be in use, it may be in the
* name cache or it may be executing.
*/
void
cfs_flush(dcstat)
enum dc_status dcstat;
{
int hash;
struct cnode *cp;
cfs_clstat.ncalls++;
cfs_clstat.reqs[CFS_FLUSH]++;
cfsnc_flush(dcstat); /* flush files from the name cache */
for (hash = 0; hash < CFS_CACHESIZE; hash++) {
for (cp = cfs_cache[hash]; cp != NULL; cp = CNODE_NEXT(cp)) {
if (!ODD(cp->c_fid.Vnode)) /* only files can be executed */
cfs_vmflush(cp);
}
}
}
/*
* As a debugging measure, print out any cnodes that lived through a
* name cache flush.
*/
void
cfs_testflush(void)
{
int hash;
struct cnode *cp;
for (hash = 0; hash < CFS_CACHESIZE; hash++) {
for (cp = cfs_cache[hash];
cp != NULL;
cp = CNODE_NEXT(cp)) {
myprintf(("Live cnode fid %lx.%lx.%lx count %d\n",
(cp->c_fid).Volume,(cp->c_fid).Vnode,
(cp->c_fid).Unique, CTOV(cp)->v_usecount));
}
}
}
/*
* First, step through all cnodes and mark them unmounting.
* NetBSD kernels may try to fsync them now that venus
* is dead, which would be a bad thing.
*
*/
void
cfs_unmounting(whoIam)
struct mount *whoIam;
{
int hash;
struct cnode *cp;
for (hash = 0; hash < CFS_CACHESIZE; hash++) {
for (cp = cfs_cache[hash]; cp != NULL; cp = CNODE_NEXT(cp)) {
if (CTOV(cp)->v_mount == whoIam) {
if (cp->c_flags & (C_LOCKED|C_WANTED)) {
printf("cfs_unmounting: Unlocking %p\n", cp);
cp->c_flags &= ~(C_LOCKED|C_WANTED);
wakeup((caddr_t) cp);
}
cp->c_flags |= C_UNMOUNTING;
}
}
}
}
#ifdef DEBUG
cfs_checkunmounting(mp)
struct mount *mp;
{
register struct vnode *vp, *nvp;
struct cnode *cp;
int count = 0, bad = 0;
loop:
for (vp = mp->mnt_vnodelist.lh_first; vp; vp = nvp) {
if (vp->v_mount != mp)
goto loop;
nvp = vp->v_mntvnodes.le_next;
cp = VTOC(vp);
count++;
if (!(cp->c_flags & C_UNMOUNTING)) {
bad++;
printf("vp %p, cp %p missed\n", vp, cp);
cp->c_flags |= C_UNMOUNTING;
}
}
}
int
cfs_cacheprint(whoIam)
struct mount *whoIam;
{
int hash;
struct cnode *cp;
int count = 0;
printf("cfs_cacheprint: cfs_ctlvp %p, cp %p", cfs_ctlvp, VTOC(cfs_ctlvp));
cfsnc_name(cfs_ctlvp);
printf("\n");
for (hash = 0; hash < CFS_CACHESIZE; hash++) {
for (cp = cfs_cache[hash]; cp != NULL; cp = CNODE_NEXT(cp)) {
if (CTOV(cp)->v_mount == whoIam) {
printf("cfs_cacheprint: vp %p, cp %p", CTOV(cp), cp);
cfsnc_name(cp);
printf("\n");
count++;
}
}
}
printf("cfs_cacheprint: count %d\n", count);
}
#endif
/*
* There are 6 cases where invalidations occur. The semantics of each
* is listed here.
*
* CFS_FLUSH -- flush all entries from the name cache and the cnode cache.
* CFS_PURGEUSER -- flush all entries from the name cache for a specific user
* This call is a result of token expiration.
*
* The next two are the result of callbacks on a file or directory.
* CFS_ZAPDIR -- flush the attributes for the dir from its cnode.
* Zap all children of this directory from the namecache.
* CFS_ZAPFILE -- flush the attributes for a file.
*
* The fifth is a result of Venus detecting an inconsistent file.
* CFS_PURGEFID -- flush the attribute for the file
* If it is a dir (odd vnode), purge its
* children from the namecache
* remove the file from the namecache.
*
* The sixth allows Venus to replace local fids with global ones
* during reintegration.
*
* CFS_REPLACE -- replace one ViceFid with another throughout the name cache
*/
int handleDownCall(opcode, out)
int opcode; union outputArgs *out;
{
int error;
/* Handle invalidate requests. */
switch (opcode) {
case CFS_FLUSH : {
cfs_flush(IS_DOWNCALL);
CFSDEBUG(CFS_FLUSH,cfs_testflush();) /* print remaining cnodes */
return(0);
}
case CFS_PURGEUSER : {
cfs_clstat.ncalls++;
cfs_clstat.reqs[CFS_PURGEUSER]++;
/* XXX - need to prevent fsync's */
cfsnc_purge_user(out->cfs_purgeuser.cred.cr_uid, IS_DOWNCALL);
return(0);
}
case CFS_ZAPFILE : {
struct cnode *cp;
error = 0;
cfs_clstat.ncalls++;
cfs_clstat.reqs[CFS_ZAPFILE]++;
cp = cfs_find(&out->cfs_zapfile.CodaFid);
if (cp != NULL) {
vref(CTOV(cp));
cp->c_flags &= ~C_VATTR;
if (CTOV(cp)->v_flag & VTEXT)
error = cfs_vmflush(cp);
CFSDEBUG(CFS_ZAPFILE, myprintf(("zapfile: fid = (%lx.%lx.%lx),
refcnt = %d, error = %d\n",
cp->c_fid.Volume,
cp->c_fid.Vnode,
cp->c_fid.Unique,
CTOV(cp)->v_usecount - 1, error)););
if (CTOV(cp)->v_usecount == 1) {
cp->c_flags |= C_PURGING;
}
vrele(CTOV(cp));
}
return(error);
}
case CFS_ZAPDIR : {
struct cnode *cp;
cfs_clstat.ncalls++;
cfs_clstat.reqs[CFS_ZAPDIR]++;
cp = cfs_find(&out->cfs_zapdir.CodaFid);
if (cp != NULL) {
vref(CTOV(cp));
cp->c_flags &= ~C_VATTR;
cfsnc_zapParentfid(&out->cfs_zapdir.CodaFid, IS_DOWNCALL);
CFSDEBUG(CFS_ZAPDIR, myprintf(("zapdir: fid = (%lx.%lx.%lx),
refcnt = %d\n",cp->c_fid.Volume,
cp->c_fid.Vnode,
cp->c_fid.Unique,
CTOV(cp)->v_usecount - 1)););
if (CTOV(cp)->v_usecount == 1) {
cp->c_flags |= C_PURGING;
}
vrele(CTOV(cp));
}
return(0);
}
case CFS_ZAPVNODE : {
cfs_clstat.ncalls++;
cfs_clstat.reqs[CFS_ZAPVNODE]++;
myprintf(("CFS_ZAPVNODE: Called, but uniplemented\n"));
/*
* Not that below we must really translate the returned coda_cred to
* a netbsd cred. This is a bit muddled at present and the cfsnc_zapnode
* is further unimplemented, so punt!
* I suppose we could use just the uid.
*/
/* cfsnc_zapvnode(&out->cfs_zapvnode.VFid, &out->cfs_zapvnode.cred,
IS_DOWNCALL); */
return(0);
}
case CFS_PURGEFID : {
struct cnode *cp;
error = 0;
cfs_clstat.ncalls++;
cfs_clstat.reqs[CFS_PURGEFID]++;
cp = cfs_find(&out->cfs_purgefid.CodaFid);
if (cp != NULL) {
vref(CTOV(cp));
if (ODD(out->cfs_purgefid.CodaFid.Vnode)) { /* Vnode is a directory */
cfsnc_zapParentfid(&out->cfs_purgefid.CodaFid,
IS_DOWNCALL);
}
cp->c_flags &= ~C_VATTR;
cfsnc_zapfid(&out->cfs_purgefid.CodaFid, IS_DOWNCALL);
if (!(ODD(out->cfs_purgefid.CodaFid.Vnode))
&& (CTOV(cp)->v_flag & VTEXT)) {
error = cfs_vmflush(cp);
}
CFSDEBUG(CFS_PURGEFID, myprintf(("purgefid: fid = (%lx.%lx.%lx), refcnt = %d, error = %d\n",
cp->c_fid.Volume, cp->c_fid.Vnode,
cp->c_fid.Unique,
CTOV(cp)->v_usecount - 1, error)););
if (CTOV(cp)->v_usecount == 1) {
cp->c_flags |= C_PURGING;
}
vrele(CTOV(cp));
}
return(error);
}
case CFS_REPLACE : {
struct cnode *cp = NULL;
cfs_clstat.ncalls++;
cfs_clstat.reqs[CFS_REPLACE]++;
cp = cfs_find(&out->cfs_replace.OldFid);
if (cp != NULL) {
/* remove the cnode from the hash table, replace the fid, and reinsert */
vref(CTOV(cp));
cfs_unsave(cp);
cp->c_fid = out->cfs_replace.NewFid;
cfs_save(cp);
CFSDEBUG(CFS_REPLACE, myprintf(("replace: oldfid = (%lx.%lx.%lx), newfid = (%lx.%lx.%lx), cp = %p\n",
out->cfs_replace.OldFid.Volume,
out->cfs_replace.OldFid.Vnode,
out->cfs_replace.OldFid.Unique,
cp->c_fid.Volume, cp->c_fid.Vnode,
cp->c_fid.Unique, cp));)
vrele(CTOV(cp));
}
return (0);
}
default:
myprintf(("handleDownCall: unknown opcode %d\n", opcode));
return (EINVAL);
}
}
/* cfs_grab_vnode: lives in either cfs_mach.c or cfs_nbsd.c */
int
cfs_vmflush(cp)
struct cnode *cp;
{
return 0;
}
/*
* kernel-internal debugging switches
*/
void cfs_debugon(void)
{
cfsdebug = -1;
cfsnc_debug = -1;
cfs_vnop_print_entry = 1;
cfs_psdev_print_entry = 1;
cfs_vfsop_print_entry = 1;
}
void cfs_debugoff(void)
{
cfsdebug = 0;
cfsnc_debug = 0;
cfs_vnop_print_entry = 0;
cfs_psdev_print_entry = 0;
cfs_vfsop_print_entry = 0;
}
/*
* Utilities used by both client and server
* Standard levels:
* 0) no debugging
* 1) hard failures
* 2) soft failures
* 3) current test software
* 4) main procedure entry points
* 5) main procedure exit points
* 6) utility procedure entry points
* 7) utility procedure exit points
* 8) obscure procedure entry points
* 9) obscure procedure exit points
* 10) random stuff
* 11) all <= 1
* 12) all <= 2
* 13) all <= 3
* ...
*/