/* $NetBSD: nfs.h,v 1.81 2024/12/07 02:05:55 riastradh Exp $ */ /* * Copyright (c) 1989, 1993, 1995 * 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. 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. * * @(#)nfs.h 8.4 (Berkeley) 5/1/95 */ #ifndef _NFS_NFS_H_ #define _NFS_NFS_H_ #ifdef _KERNEL #include #include #include #include #include #endif #include /* * Tunable constants for nfs */ #define NFS_MAXIOVEC 34 #define NFS_TICKINTVL 5 /* Desired time for a tick (msec) */ #define NFS_HZ (hz / nfs_ticks) /* Ticks/sec */ #define NFS_TIMEO (3 * NFS_HZ) /* Default timeout = 3 seconds */ #define NFS_MINTIMEO (1 * NFS_HZ) /* Min timeout to use */ #define NFS_MAXTIMEO (60 * NFS_HZ) /* Max timeout to backoff to */ #define NFS_MINIDEMTIMEO (5 * NFS_HZ) /* Min timeout for non-idempotent ops*/ #define NFS_MAXREXMIT 100 /* Stop counting after this many */ #define NFS_MAXWINDOW 1024 /* Max number of outstanding requests */ #define NFS_RETRANS 10 /* Num of retrans for soft mounts */ #define NFS_MAXGRPS 16 /* Max. size of groups list */ #ifndef NFS_MINATTRTIMO #define NFS_MINATTRTIMO 5 /* Attribute cache timeout in sec */ #endif #ifndef NFS_MAXATTRTIMO #define NFS_MAXATTRTIMO 60 #endif #define NFS_TRYLATERDEL 1 /* Initial try later delay (sec) */ #define NFS_TRYLATERDELMAX (1*60) /* Maximum try later delay (sec) */ #define NFS_TRYLATERDELMUL 2 /* Exponential backoff multiplier */ #define NFS_CWNDSCALE 256 #define NFS_MAXCWND (NFS_CWNDSCALE * 32) /* * These can be overridden through , included via * . This means that should always be * included before this file. */ #ifndef NFS_WSIZE #define NFS_WSIZE 8192 /* Def. write data size */ #endif #ifndef NFS_RSIZE #define NFS_RSIZE 8192 /* Def. read data size */ #endif #ifndef NFS_READDIRSIZE #define NFS_READDIRSIZE 8192 /* Def. readdir size */ #endif /* * NFS client IO daemon threads. May be overridden by config options. */ #ifndef NFS_MAXASYNCDAEMON #define NFS_MAXASYNCDAEMON 128 /* Max. number async_daemons runnable */ #endif /* * NFS client read-ahead. May be overridden by config options. * Should be no more than NFS_MAXASYNCDAEMON as each read-ahead operation * requires one IO thread. */ #ifndef NFS_MAXRAHEAD #define NFS_MAXRAHEAD 32 /* Max. read ahead # blocks */ #endif #define NFS_DEFRAHEAD 2 /* Def. read ahead # blocks */ #define NFS_MAXUIDHASH 64 /* Max. # of hashed uid entries/mp */ #define NFS_DEFDEADTHRESH NFS_NEVERDEAD /* Default nm_deadthresh */ #define NFS_NEVERDEAD 9 /* Greater than max. nm_timeouts */ #ifdef _KERNEL extern int nfs_niothreads; /* Number of async_daemons desired */ #ifndef NFS_DEFAULT_NIOTHREADS #define NFS_DEFAULT_NIOTHREADS 4 #endif #endif #define NFS_MAXGATHERDELAY 100 /* Max. write gather delay (msec) */ #ifndef NFS_GATHERDELAY #define NFS_GATHERDELAY 10 /* Default write gather delay (msec) */ #endif /* * NFS_DIRBLKSIZ is the size of buffers in the buffer cache used for * NFS directory vnodes. NFS_DIRFRAGSIZ is the minimum aligned amount * of data in those buffers, and thus the minimum amount of data * that you can request. NFS_DIRFRAGSIZ should be no smaller than * DIRBLKSIZ. */ #define NFS_DIRBLKSIZ 8192 /* Must be a multiple of DIRBLKSIZ */ #define NFS_DIRFRAGSIZ 512 /* Same as DIRBLKSIZ, generally */ /* * Maximum number of directory entries cached per NFS node, to avoid * having this grow without bounds on very large directories. The * minimum size to get reasonable performance for emulated binaries * is the maximum number of entries that fits in NFS_DIRBLKSIZ. * For NFS_DIRBLKSIZ = 512, this would be 512 / 14 = 36. */ #define NFS_MAXDIRCACHE (NFS_DIRBLKSIZ / 14) /* * Oddballs */ #define NFS_CMPFH(n, f, s) \ ((n)->n_fhsize == (s) && !memcmp((void *)(n)->n_fhp, (void *)(f), (s))) #ifdef NFS_V2_ONLY #define NFS_ISV3(v) (0) #else #define NFS_ISV3(v) (VFSTONFS((v)->v_mount)->nm_flag & NFSMNT_NFSV3) #endif #define NFS_SRVMAXDATA(n) \ (((n)->nd_flag & ND_NFSV3) ? (((n)->nd_nam2) ? \ NFS_MAXDGRAMDATA : NFS_MAXDATA) : NFS_V2MAXDATA) /* * Use the vm_page flag reserved for pager use to indicate pages * which have been written to the server but not yet committed. */ #define PG_NEEDCOMMIT PG_PAGER1 /* * The IO_METASYNC flag should be implemented for local file systems. * (Until then, it is nothin at all.) */ #ifndef IO_METASYNC #define IO_METASYNC 0 #endif /* * Export arguments for local filesystem mount calls. * Keep in mind that changing this structure modifies nfssvc(2)'s ABI (see * 'struct mountd_exports_list' below). * When modifying this structure, take care to also edit the * nfs_update_exports_30 function in nfs_export.c accordingly to convert * export_args to export_args30. */ struct export_args { int ex_flags; /* export related flags */ uid_t ex_root; /* mapping for root uid */ struct uucred ex_anon; /* mapping for anonymous user */ struct sockaddr *ex_addr; /* net address to which exported */ int ex_addrlen; /* and the net address length */ struct sockaddr *ex_mask; /* mask of valid bits in saddr */ int ex_masklen; /* and the smask length */ char *ex_indexfile; /* index file for WebNFS URLs */ }; /* * Structures for the nfssvc(2) syscall. Not that anyone but mountd, nfsd and * mount_nfs should ever try and use it. */ struct nfsd_args { int sock; /* Socket to serve */ void * name; /* Client addr for connection based sockets */ int namelen; /* Length of name */ }; struct nfsd_srvargs { struct nfsd *nsd_nfsd; /* Pointer to in kernel nfsd struct */ uid_t nsd_uid; /* Effective uid mapped to cred */ u_int32_t nsd_haddr; /* Ip address of client */ struct uucred nsd_cr; /* Cred. uid maps to */ int nsd_authlen; /* Length of auth string (ret) */ u_char *nsd_authstr; /* Auth string (ret) */ int nsd_verflen; /* and the verfier */ u_char *nsd_verfstr; struct timeval nsd_timestamp; /* timestamp from verifier */ u_int32_t nsd_ttl; /* credential ttl (sec) */ NFSKERBKEY_T nsd_key; /* Session key */ }; struct nfsd_cargs { char *ncd_dirp; /* Mount dir path */ uid_t ncd_authuid; /* Effective uid */ int ncd_authtype; /* Type of authenticator */ u_int ncd_authlen; /* Length of authenticator string */ u_char *ncd_authstr; /* Authenticator string */ u_int ncd_verflen; /* and the verifier */ u_char *ncd_verfstr; NFSKERBKEY_T ncd_key; /* Session key */ }; struct mountd_exports_list { const char *mel_path; size_t mel_nexports; struct export_args *mel_exports; }; /* * try to keep nfsstats, which is exposed to userland via sysctl, * compatible after NQNFS removal. * 26 is the old value of NFS_NPROCS, which includes NQNFS procedures. */ #define NFSSTATS_NPROCS 26 /* * Stats structure */ struct nfsstats { uint32_t attrcache_hits; uint32_t attrcache_misses; uint32_t lookupcache_hits; uint32_t lookupcache_misses; uint32_t direofcache_hits; uint32_t direofcache_misses; uint32_t biocache_reads; uint32_t read_bios; uint32_t read_physios; uint32_t biocache_writes; uint32_t write_bios; uint32_t write_physios; uint32_t biocache_readlinks; uint32_t readlink_bios; uint32_t biocache_readdirs; uint32_t readdir_bios; uint32_t rpccnt[NFSSTATS_NPROCS]; uint32_t rpcretries; uint32_t srvrpccnt[NFSSTATS_NPROCS]; uint32_t srvrpc_errs; uint32_t srv_errs; uint32_t rpcrequests; uint32_t rpctimeouts; uint32_t rpcunexpected; uint32_t rpcinvalid; uint32_t srvcache_inproghits; uint32_t srvcache_idemdonehits; uint32_t srvcache_nonidemdonehits; uint32_t srvcache_misses; uint32_t __srvnqnfs_leases; /* unused */ uint32_t __srvnqnfs_maxleases; /* unused */ uint32_t __srvnqnfs_getleases; /* unused */ uint32_t srvvop_writes; }; /* * Flags for nfssvc() system call. */ #define NFSSVC_BIOD 0x002 #define NFSSVC_NFSD 0x004 #define NFSSVC_ADDSOCK 0x008 #define NFSSVC_AUTHIN 0x010 #define NFSSVC_GOTAUTH 0x040 #define NFSSVC_AUTHINFAIL 0x080 #define NFSSVC_MNTD 0x100 #define NFSSVC_SETEXPORTSLIST 0x200 #define NFSSVC_REPLACEEXPORTSLIST 0x400 /* * fs.nfs sysctl(3) identifiers */ #define NFS_NFSSTATS 1 /* struct: struct nfsstats */ #define NFS_IOTHREADS 2 /* number of io threads */ /* * The set of signals the interrupt an I/O in progress for NFSMNT_INT mounts. * What should be in this set is open to debate, but I believe that since * I/O system calls on ufs are never interrupted by signals the set should * be minimal. My reasoning is that many current programs that use signals * such as SIGALRM will not expect file I/O system calls to be interrupted * by them and break. */ #ifdef _KERNEL struct uio; struct buf; struct vattr; struct nameidata; /* XXX */ /* * Socket errors ignored for connectionless sockets?? * For now, ignore them all */ #define NFSIGNORE_SOERROR(s, e) \ ((e) != EINTR && (e) != ERESTART && (e) != EWOULDBLOCK && \ ((s) & PR_CONNREQUIRED) == 0) /* * Nfs outstanding request list element */ struct nfsreq { TAILQ_ENTRY(nfsreq) r_chain; struct mbuf *r_mreq; struct mbuf *r_mrep; struct mbuf *r_md; void * r_dpos; struct nfsmount *r_nmp; u_int32_t r_xid; int r_flags; /* flags on request, see below */ int r_retry; /* max retransmission count */ int r_rexmit; /* current retrans count */ u_int32_t r_procnum; /* NFS procedure number */ int r_rtt; /* RTT for rpc */ struct lwp *r_lwp; /* LWP that did I/O system call */ }; /* * Queue head for nfsreq's */ extern TAILQ_HEAD(nfsreqhead, nfsreq) nfs_reqq; extern kmutex_t nfs_reqq_lock; /* Flag values for r_flags */ #define R_TIMING 0x01 /* timing request (in mntp) */ #define R_SENT 0x02 /* request has been sent */ #define R_SOFTTERM 0x04 /* soft mnt, too many retries */ #define R_INTR 0x08 /* intr mnt, signal pending */ #define R_SOCKERR 0x10 /* Fatal error on socket */ #define R_TPRINTFMSG 0x20 /* Did a tprintf msg. */ #define R_MUSTRESEND 0x40 /* Must resend request */ #define R_REXMITTED 0x100 /* retransmitted after reconnect */ /* * A list of nfssvc_sock structures is maintained with all the sockets * that require service by the nfsd. * The nfsuid structs hang off of the nfssvc_sock structs in both lru * and uid hash lists. */ #ifndef NFS_UIDHASHSIZ #define NFS_UIDHASHSIZ 29 /* Tune the size of nfssvc_sock with this */ #endif #define NUIDHASH(sock, uid) \ (&(sock)->ns_uidhashtbl[(uid) % NFS_UIDHASHSIZ]) #ifndef NFS_WDELAYHASHSIZ #define NFS_WDELAYHASHSIZ 16 /* and with this */ #endif #ifndef NFS_MUIDHASHSIZ #define NFS_MUIDHASHSIZ 63 /* Tune the size of nfsmount with this */ #endif #define NMUIDHASH(nmp, uid) \ (&(nmp)->nm_uidhashtbl[(uid) % NFS_MUIDHASHSIZ]) #ifndef NFS_DIRHASHSIZ #define NFS_DIRHASHSIZ 64 #endif #define NFSDIRHASH(np, off) \ (&np->n_dircache[(nfs_dirhash((off)) & nfsdirhashmask)]) /* * Macros for storing/retrieving cookies into directory buffers. */ #define NFS_STASHCOOKIE(dp,off) \ *((off_t *)((char *)(dp) + (dp)->d_reclen - sizeof (off_t))) = off #define NFS_GETCOOKIE(dp) \ (*((off_t *)((char *)(dp) + (dp)->d_reclen - sizeof (off_t)))) #define NFS_STASHCOOKIE32(dp, val) \ *((u_int32_t *)((char *)(dp) + (dp)->d_reclen - sizeof (off_t) - \ sizeof (int))) = val #define NFS_GETCOOKIE32(dp) \ (*((u_int32_t *)((char *)(dp) + (dp)->d_reclen - sizeof (off_t) - \ sizeof (int)))) /* * Flags passed to nfs_bioread(). */ #define NFSBIO_CACHECOOKIES 0x0001 /* Cache dir offset cookies */ /* * Network address hash list element */ union nethostaddr { u_int32_t had_inetaddr; struct mbuf *had_nam; }; struct nfsuid { TAILQ_ENTRY(nfsuid) nu_lru; /* LRU chain */ LIST_ENTRY(nfsuid) nu_hash; /* Hash list */ int nu_flag; /* Flags */ union nethostaddr nu_haddr; /* Host addr. for dgram sockets */ kauth_cred_t nu_cr; /* Cred uid mapped to */ int nu_expire; /* Expiry time (sec) */ struct timeval nu_timestamp; /* Kerb. timestamp */ u_int32_t nu_nickname; /* Nickname on server */ NFSKERBKEY_T nu_key; /* and session key */ }; #define nu_inetaddr nu_haddr.had_inetaddr #define nu_nam nu_haddr.had_nam /* Bits for nu_flag */ #define NU_INETADDR 0x1 #define NU_NAM 0x2 #define NU_NETFAM(u) \ (((u)->nu_flag & NU_INETADDR) ? AF_INET : AF_INET6) /* * b: protected by SLP_BUSY * g: protected by nfsd_lock * s: protected by ns_lock * a: protected by ns_alock */ struct nfssvc_sock { kmutex_t ns_lock; kmutex_t ns_alock; kcondvar_t ns_cv; /* s: */ TAILQ_ENTRY(nfssvc_sock) ns_chain; /* g: List of all nfssvc_sock */ TAILQ_ENTRY(nfssvc_sock) ns_pending; /* g: List of pending sockets */ TAILQ_HEAD(, nfsuid) ns_uidlruhead; struct file *ns_fp; struct socket *ns_so; struct mbuf *ns_nam; struct mbuf *ns_raw; /* b: */ struct mbuf *ns_rawend; /* b: */ struct mbuf *ns_rec; /* b: */ struct mbuf *ns_recend; /* b: */ struct mbuf *ns_frag; /* b: */ int ns_flags; /* s: */ int ns_aflags; /* a: */ int ns_gflags; /* g: */ int ns_sflags; /* b: */ int ns_cc; /* b: */ int ns_reclen; /* b: */ int ns_numuids; u_int32_t ns_sref; /* g: */ SIMPLEQ_HEAD(, nfsrv_descript) ns_sendq; /* s: send reply list */ LIST_HEAD(, nfsrv_descript) ns_tq; /* g: Write gather lists */ LIST_HEAD(, nfsuid) ns_uidhashtbl[NFS_UIDHASHSIZ]; LIST_HEAD(nfsrvw_delayhash, nfsrv_descript) ns_wdelayhashtbl[NFS_WDELAYHASHSIZ]; /* g: */ }; /* Bits for "ns_flags" */ #define SLP_VALID 0x01 #define SLP_BUSY 0x10 #define SLP_SENDING 0x80 /* Bits for "ns_aflags" */ #define SLP_A_NEEDQ 0x01 #define SLP_A_DISCONN 0x04 /* Bits for "ns_gflags" */ #define SLP_G_DOREC 0x02 /* on nfssvc_sockpending queue */ /* Bits for "ns_sflags" */ #define SLP_S_LASTFRAG 0x40 extern TAILQ_HEAD(nfssvc_sockhead, nfssvc_sock) nfssvc_sockhead; extern struct nfssvc_sockhead nfssvc_sockpending; extern int nfssvc_sockhead_flag; #define SLP_INIT 0x01 /* * One of these structures is allocated for each nfsd. */ struct nfsd { struct rb_node nfsd_node; /* Tree of all nfsd's */ SLIST_ENTRY(nfsd) nfsd_idle; /* List of idle nfsd's */ kcondvar_t nfsd_cv; int nfsd_flag; /* NFSD_ flags */ struct nfssvc_sock *nfsd_slp; /* Current socket */ int nfsd_authlen; /* Authenticator len */ u_char nfsd_authstr[RPCAUTH_MAXSIZ]; /* Authenticator data */ int nfsd_verflen; /* and the Verifier */ u_char nfsd_verfstr[RPCVERF_MAXSIZ]; struct proc *nfsd_procp; /* Proc ptr */ struct nfsrv_descript *nfsd_nd; /* Associated nfsrv_descript */ uint32_t nfsd_cookie; /* Userland cookie, fits 32bit ptr */ }; /* Bits for "nfsd_flag" */ #define NFSD_NEEDAUTH 0x04 #define NFSD_AUTHFAIL 0x08 #define NFSD_MAXFHSIZE 64 typedef struct nfsrvfh { size_t nsfh_size; union { fhandle_t u_fh; uint8_t u_opaque[NFSD_MAXFHSIZE]; } nsfh_u; } nfsrvfh_t; #define NFSRVFH_SIZE(nsfh) ((nsfh)->nsfh_size) #define NFSRVFH_DATA(nsfh) ((nsfh)->nsfh_u.u_opaque) #define NFSRVFH_FHANDLE(nsfh) (&(nsfh)->nsfh_u.u_fh) /* * This structure is used by the server for describing each request. * Some fields are used only when write request gathering is performed. */ struct nfsrv_descript { u_quad_t nd_time; /* Write deadline (usec) */ off_t nd_off; /* Start byte offset */ off_t nd_eoff; /* and end byte offset */ LIST_ENTRY(nfsrv_descript) nd_hash; /* Hash list */ LIST_ENTRY(nfsrv_descript) nd_tq; /* and timer list */ LIST_HEAD(,nfsrv_descript) nd_coalesce; /* coalesced writes */ SIMPLEQ_ENTRY(nfsrv_descript) nd_sendq; /* send reply list */ struct mbuf *nd_mrep; /* Request mbuf list */ struct mbuf *nd_md; /* Current dissect mbuf */ struct mbuf *nd_mreq; /* Reply mbuf list */ struct mbuf *nd_nam; /* and socket addr */ struct mbuf *nd_nam2; /* return socket addr */ void * nd_dpos; /* Current dissect pos */ u_int32_t nd_procnum; /* RPC # */ int nd_stable; /* storage type */ int nd_flag; /* nd_flag */ int nd_len; /* Length of this write */ int nd_repstat; /* Reply status */ u_int32_t nd_retxid; /* Reply xid */ u_int32_t nd_duration; /* Lease duration */ struct timeval nd_starttime; /* Time RPC initiated */ nfsrvfh_t nd_fh; /* File handle */ kauth_cred_t nd_cr; /* Credentials */ }; /* Bits for "nd_flag" */ #define ND_READ 0x01 /* Check lease for readers */ #define ND_WRITE 0x02 /* Check lease for modifiers */ #define ND_CHECK 0x04 #define ND_LEASE (ND_READ | ND_WRITE | ND_CHECK) #define ND_NFSV3 0x08 #define ND_KERBNICK 0x20 #define ND_KERBFULL 0x40 #define ND_KERBAUTH (ND_KERBNICK | ND_KERBFULL) extern kmutex_t nfsd_lock; extern kcondvar_t nfsd_initcv; extern SLIST_HEAD(nfsdidlehead, nfsd) nfsd_idle_head; extern int nfsd_head_flag; #define NFSD_CHECKSLP 0x01 extern struct mowner nfs_mowner; extern struct nfsstats nfsstats; extern int nfs_numasync; /* * These macros compare nfsrv_descript structures. */ #define NFSW_CONTIG(o, n) \ ((o)->nd_eoff >= (n)->nd_off && \ !memcmp((void *)&(o)->nd_fh, (void *)&(n)->nd_fh, NFSX_V3FH)) /* * Defines for WebNFS */ #define WEBNFS_ESC_CHAR '%' #define WEBNFS_SPECCHAR_START 0x80 #define WEBNFS_NATIVE_CHAR 0x80 /* * .. * Possibly more here in the future. */ /* * Macro for converting escape characters in WebNFS pathnames. * Should really be in libkern. */ #define HEXTOC(c) \ ((c) >= 'a' ? ((c) - ('a' - 10)) : \ ((c) >= 'A' ? ((c) - ('A' - 10)) : ((c) - '0'))) #define HEXSTRTOI(p) \ ((HEXTOC(p[0]) << 4) + HEXTOC(p[1])) /* * Structure holding information for a publicly exported filesystem * (WebNFS). Currently the specs allow just for one such filesystem. */ struct nfs_public { int np_valid; /* Do we hold valid information */ fhandle_t *np_handle; /* Filehandle for pub fs (internal) */ struct mount *np_mount; /* Mountpoint of exported fs */ char *np_index; /* Index file */ }; #endif /* _KERNEL */ #endif /* _NFS_NFS_H */