mirror of
https://git.FreeBSD.org/src.git
synced 2024-12-22 11:17:19 +00:00
f438ae02f5
existing mechanism uses a global queue for some buffers and the vp->b_dirtyblkhd queue for others. This turns sequential writes into randomly ordered writes to the server, affecting both read and write performance. The existing mechanism also copes badly with hung servers, tending to block accesses to other servers when all the iods are waiting for a hung server. The new mechanism uses a queue for each mount point. All asynchronous i/o goes through this queue which preserves the ordering of requests. A simple mechanism ensures that the iods are shared out fairly between active mount points. This removes the sysctl variable vfs.nfs.dwrite since the new queueing mechanism removes the old delayed write code completely. This should go into the 2.2 branch.
202 lines
7.2 KiB
C
202 lines
7.2 KiB
C
/*
|
|
* Copyright (c) 1989, 1993
|
|
* The Regents of the University of California. All rights reserved.
|
|
*
|
|
* This code is derived from software contributed to Berkeley by
|
|
* Rick Macklem at The University of Guelph.
|
|
*
|
|
* Redistribution and use in source and binary forms, with or without
|
|
* modification, are permitted provided that the following conditions
|
|
* are met:
|
|
* 1. Redistributions of source code must retain the above copyright
|
|
* notice, this list of conditions and the following disclaimer.
|
|
* 2. Redistributions in binary form must reproduce the above copyright
|
|
* notice, this list of conditions and the following disclaimer in the
|
|
* documentation and/or other materials provided with the distribution.
|
|
* 3. All advertising materials mentioning features or use of this software
|
|
* must display the following acknowledgement:
|
|
* This product includes software developed by the University of
|
|
* California, Berkeley and its contributors.
|
|
* 4. Neither the name of the University nor the names of its contributors
|
|
* may be used to endorse or promote products derived from this software
|
|
* without specific prior written permission.
|
|
*
|
|
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
|
|
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
|
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
|
|
* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
|
|
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
|
|
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
|
|
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
|
|
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
|
|
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
|
|
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
|
|
* SUCH DAMAGE.
|
|
*
|
|
* @(#)nfsnode.h 8.4 (Berkeley) 2/13/94
|
|
* $Id: nfsnode.h,v 1.15 1995/12/17 21:12:37 phk Exp $
|
|
*/
|
|
|
|
#ifndef _NFS_NFSNODE_H_
|
|
#define _NFS_NFSNODE_H_
|
|
|
|
#ifndef _NFS_NFS_H_
|
|
#include <nfs/nfs.h>
|
|
#endif
|
|
|
|
/*
|
|
* Silly rename structure that hangs off the nfsnode until the name
|
|
* can be removed by nfs_inactive()
|
|
*/
|
|
struct sillyrename {
|
|
struct ucred *s_cred;
|
|
struct vnode *s_dvp;
|
|
long s_namlen;
|
|
char s_name[20];
|
|
};
|
|
|
|
/*
|
|
* This structure is used to save the logical directory offset to
|
|
* NFS cookie mappings.
|
|
* The mappings are stored in a list headed
|
|
* by n_cookies, as required.
|
|
* There is one mapping for each NFS_DIRBLKSIZ bytes of directory information
|
|
* stored in increasing logical offset byte order.
|
|
*/
|
|
#define NFSNUMCOOKIES 31
|
|
|
|
struct nfsdmap {
|
|
LIST_ENTRY(nfsdmap) ndm_list;
|
|
int ndm_eocookie;
|
|
nfsuint64 ndm_cookies[NFSNUMCOOKIES];
|
|
};
|
|
|
|
/*
|
|
* The nfsnode is the nfs equivalent to ufs's inode. Any similarity
|
|
* is purely coincidental.
|
|
* There is a unique nfsnode allocated for each active file,
|
|
* each current directory, each mounted-on file, text file, and the root.
|
|
* An nfsnode is 'named' by its file handle. (nget/nfs_node.c)
|
|
* If this structure exceeds 256 bytes (it is currently 256 using 4.4BSD-Lite
|
|
* type definitions), file handles of > 32 bytes should probably be split out
|
|
* into a separate MALLOC()'d data structure. (Reduce the size of nfsfh_t by
|
|
* changing the definition in sys/mount.h of NFS_SMALLFH.)
|
|
* NB: Hopefully the current order of the fields is such that everything will
|
|
* be well aligned and, therefore, tightly packed.
|
|
*/
|
|
struct nfsnode {
|
|
LIST_ENTRY(nfsnode) n_hash; /* Hash chain */
|
|
CIRCLEQ_ENTRY(nfsnode) n_timer; /* Nqnfs timer chain */
|
|
u_quad_t n_size; /* Current size of file */
|
|
u_quad_t n_brev; /* Modify rev when cached */
|
|
u_quad_t n_lrev; /* Modify rev for lease */
|
|
struct vattr n_vattr; /* Vnode attribute cache */
|
|
time_t n_attrstamp; /* Attr. cache timestamp */
|
|
time_t n_mtime; /* Prev modify time. */
|
|
time_t n_ctime; /* Prev create time. */
|
|
time_t n_expiry; /* Lease expiry time */
|
|
nfsfh_t *n_fhp; /* NFS File Handle */
|
|
struct vnode *n_vnode; /* associated vnode */
|
|
struct lockf *n_lockf; /* Locking record of file */
|
|
int n_error; /* Save write error value */
|
|
union {
|
|
struct timespec nf_atim; /* Special file times */
|
|
nfsuint64 nd_cookieverf; /* Cookie verifier (dir only) */
|
|
} n_un1;
|
|
union {
|
|
struct timespec nf_mtim;
|
|
off_t nd_direof; /* Dir. EOF offset cache */
|
|
} n_un2;
|
|
union {
|
|
struct sillyrename *nf_silly; /* Ptr to silly rename struct */
|
|
LIST_HEAD(, nfsdmap) nd_cook; /* cookies */
|
|
} n_un3;
|
|
short n_fhsize; /* size in bytes, of fh */
|
|
short n_flag; /* Flag for locking.. */
|
|
nfsfh_t n_fh; /* Small File Handle */
|
|
};
|
|
|
|
#define n_atim n_un1.nf_atim
|
|
#define n_mtim n_un2.nf_mtim
|
|
#define n_sillyrename n_un3.nf_silly
|
|
#define n_cookieverf n_un1.nd_cookieverf
|
|
#define n_direofoffset n_un2.nd_direof
|
|
#define n_cookies n_un3.nd_cook
|
|
|
|
/*
|
|
* Flags for n_flag
|
|
*/
|
|
#define NFLUSHWANT 0x0001 /* Want wakeup from a flush in prog. */
|
|
#define NFLUSHINPROG 0x0002 /* Avoid multiple calls to vinvalbuf() */
|
|
#define NMODIFIED 0x0004 /* Might have a modified buffer in bio */
|
|
#define NWRITEERR 0x0008 /* Flag write errors so close will know */
|
|
#define NQNFSNONCACHE 0x0020 /* Non-cachable lease */
|
|
#define NQNFSWRITE 0x0040 /* Write lease */
|
|
#define NQNFSEVICTED 0x0080 /* Has been evicted */
|
|
#define NACC 0x0100 /* Special file accessed */
|
|
#define NUPD 0x0200 /* Special file updated */
|
|
#define NCHG 0x0400 /* Special file times changed */
|
|
#define NLOCKED 0x0800 /* node is locked */
|
|
#define NWANTED 0x0100 /* someone wants to lock */
|
|
|
|
/*
|
|
* Convert between nfsnode pointers and vnode pointers
|
|
*/
|
|
#define VTONFS(vp) ((struct nfsnode *)(vp)->v_data)
|
|
#define NFSTOV(np) ((struct vnode *)(np)->n_vnode)
|
|
|
|
/*
|
|
* Queue head for nfsiod's
|
|
*/
|
|
extern TAILQ_HEAD(nfs_bufq, buf) nfs_bufq;
|
|
extern struct proc *nfs_iodwant[NFS_MAXASYNCDAEMON];
|
|
extern struct nfsmount *nfs_iodmount[NFS_MAXASYNCDAEMON];
|
|
|
|
#if defined(KERNEL) || defined(_KERNEL)
|
|
extern vop_t **fifo_nfsv2nodeop_p;
|
|
extern vop_t **nfsv2_vnodeop_p;
|
|
extern vop_t **spec_nfsv2nodeop_p;
|
|
|
|
/*
|
|
* Prototypes for NFS vnode operations
|
|
*/
|
|
int nfs_write __P((struct vop_write_args *));
|
|
#ifdef HAS_VOPLEASE
|
|
#define nfs_lease_check ((int (*) __P((struct vop_lease_args *)))nullop)
|
|
#define nqnfs_vop_lease_check lease_check
|
|
#else
|
|
#ifdef __FreeBSD__
|
|
#define nqnfs_lease_check nfs_lease_check
|
|
#else
|
|
#define nqnfs_lease_check lease_check
|
|
#endif
|
|
#endif
|
|
#ifdef HAS_VOPREVOKE
|
|
#define nfs_revoke vop_revoke
|
|
#endif
|
|
#define nfs_seek ((int (*) __P((struct vop_seek_args *)))nullop)
|
|
int nfs_abortop __P((struct vop_abortop_args *));
|
|
int nfs_inactive __P((struct vop_inactive_args *));
|
|
int nfs_reclaim __P((struct vop_reclaim_args *));
|
|
int nfs_lock __P((struct vop_lock_args *));
|
|
int nfs_unlock __P((struct vop_unlock_args *));
|
|
int nfs_islocked __P((struct vop_islocked_args *));
|
|
#define nfs_reallocblks \
|
|
((int (*) __P((struct vop_reallocblks_args *)))eopnotsupp)
|
|
|
|
/* other stuff */
|
|
int nfs_removeit __P((struct sillyrename *));
|
|
int nfs_nget __P((struct mount *,nfsfh_t *,int,struct nfsnode **));
|
|
nfsuint64 *nfs_getcookie __P((struct nfsnode *, off_t, int));
|
|
void nfs_invaldir __P((struct vnode *));
|
|
|
|
#ifdef __FreeBSD__
|
|
#define nqnfs_lease_updatetime nfs_lease_updatetime
|
|
#else
|
|
#define nqnfs_lease_updatetime lease_updatetime
|
|
#endif
|
|
|
|
#endif /* KERNEL */
|
|
|
|
#endif
|