diff options
31 files changed, 802 insertions, 442 deletions
diff --git a/fs/lockd/Makefile b/fs/lockd/Makefile index 6d5e83ed4476..ac9f9d84510e 100644 --- a/fs/lockd/Makefile +++ b/fs/lockd/Makefile @@ -3,10 +3,12 @@ # Makefile for the linux lock manager stuff # +ccflags-y += -I$(src) # needed for trace events + obj-$(CONFIG_LOCKD) += lockd.o -lockd-objs-y := clntlock.o clntproc.o clntxdr.o host.o svc.o svclock.o \ - svcshare.o svcproc.o svcsubs.o mon.o xdr.o +lockd-objs-y += clntlock.o clntproc.o clntxdr.o host.o svc.o svclock.o \ + svcshare.o svcproc.o svcsubs.o mon.o trace.o xdr.o lockd-objs-$(CONFIG_LOCKD_V4) += clnt4xdr.o xdr4.o svc4proc.o lockd-objs-$(CONFIG_PROC_FS) += procfs.o lockd-objs := $(lockd-objs-y) diff --git a/fs/lockd/clntlock.c b/fs/lockd/clntlock.c index 82b19a30e0f0..e3972aa3045a 100644 --- a/fs/lockd/clntlock.c +++ b/fs/lockd/clntlock.c @@ -14,9 +14,12 @@ #include <linux/nfs_fs.h> #include <linux/sunrpc/addr.h> #include <linux/sunrpc/svc.h> +#include <linux/sunrpc/svc_xprt.h> #include <linux/lockd/lockd.h> #include <linux/kthread.h> +#include "trace.h" + #define NLMDBG_FACILITY NLMDBG_CLIENT /* @@ -29,18 +32,6 @@ static int reclaimer(void *ptr); * client perspective. */ -/* - * This is the representation of a blocked client lock. - */ -struct nlm_wait { - struct list_head b_list; /* linked list */ - wait_queue_head_t b_wait; /* where to wait on */ - struct nlm_host * b_host; - struct file_lock * b_lock; /* local file lock */ - unsigned short b_reclaim; /* got to reclaim lock */ - __be32 b_status; /* grant callback status */ -}; - static LIST_HEAD(nlm_blocked); static DEFINE_SPINLOCK(nlm_blocked_lock); @@ -94,41 +85,42 @@ void nlmclnt_done(struct nlm_host *host) } EXPORT_SYMBOL_GPL(nlmclnt_done); +void nlmclnt_prepare_block(struct nlm_wait *block, struct nlm_host *host, struct file_lock *fl) +{ + block->b_host = host; + block->b_lock = fl; + init_waitqueue_head(&block->b_wait); + block->b_status = nlm_lck_blocked; +} + /* * Queue up a lock for blocking so that the GRANTED request can see it */ -struct nlm_wait *nlmclnt_prepare_block(struct nlm_host *host, struct file_lock *fl) +void nlmclnt_queue_block(struct nlm_wait *block) { - struct nlm_wait *block; - - block = kmalloc(sizeof(*block), GFP_KERNEL); - if (block != NULL) { - block->b_host = host; - block->b_lock = fl; - init_waitqueue_head(&block->b_wait); - block->b_status = nlm_lck_blocked; - - spin_lock(&nlm_blocked_lock); - list_add(&block->b_list, &nlm_blocked); - spin_unlock(&nlm_blocked_lock); - } - return block; + spin_lock(&nlm_blocked_lock); + list_add(&block->b_list, &nlm_blocked); + spin_unlock(&nlm_blocked_lock); } -void nlmclnt_finish_block(struct nlm_wait *block) +/* + * Dequeue the block and return its final status + */ +__be32 nlmclnt_dequeue_block(struct nlm_wait *block) { - if (block == NULL) - return; + __be32 status; + spin_lock(&nlm_blocked_lock); list_del(&block->b_list); + status = block->b_status; spin_unlock(&nlm_blocked_lock); - kfree(block); + return status; } /* * Block on a lock */ -int nlmclnt_block(struct nlm_wait *block, struct nlm_rqst *req, long timeout) +int nlmclnt_wait(struct nlm_wait *block, struct nlm_rqst *req, long timeout) { long ret; @@ -154,7 +146,6 @@ int nlmclnt_block(struct nlm_wait *block, struct nlm_rqst *req, long timeout) /* Reset the lock status after a server reboot so we resend */ if (block->b_status == nlm_lck_denied_grace_period) block->b_status = nlm_lck_blocked; - req->a_res.status = block->b_status; return 0; } @@ -198,6 +189,7 @@ __be32 nlmclnt_grant(const struct sockaddr *addr, const struct nlm_lock *lock) res = nlm_granted; } spin_unlock(&nlm_blocked_lock); + trace_nlmclnt_grant(lock, addr, svc_addr_len(addr), res); return res; } diff --git a/fs/lockd/clntproc.c b/fs/lockd/clntproc.c index 16b4de868cd2..fba6c7fa7474 100644 --- a/fs/lockd/clntproc.c +++ b/fs/lockd/clntproc.c @@ -20,6 +20,8 @@ #include <linux/sunrpc/svc.h> #include <linux/lockd/lockd.h> +#include "trace.h" + #define NLMDBG_FACILITY NLMDBG_CLIENT #define NLMCLNT_GRACE_WAIT (5*HZ) #define NLMCLNT_POLL_TIMEOUT (30*HZ) @@ -451,6 +453,9 @@ nlmclnt_test(struct nlm_rqst *req, struct file_lock *fl) status = nlm_stat_to_errno(req->a_res.status); } out: + trace_nlmclnt_test(&req->a_args.lock, + (const struct sockaddr *)&req->a_host->h_addr, + req->a_host->h_addrlen, req->a_res.status); nlmclnt_release_call(req); return status; } @@ -516,9 +521,10 @@ nlmclnt_lock(struct nlm_rqst *req, struct file_lock *fl) const struct cred *cred = nfs_file_cred(fl->fl_file); struct nlm_host *host = req->a_host; struct nlm_res *resp = &req->a_res; - struct nlm_wait *block = NULL; + struct nlm_wait block; unsigned char fl_flags = fl->fl_flags; unsigned char fl_type; + __be32 b_status; int status = -ENOLCK; if (nsm_monitor(host) < 0) @@ -531,31 +537,41 @@ nlmclnt_lock(struct nlm_rqst *req, struct file_lock *fl) if (status < 0) goto out; - block = nlmclnt_prepare_block(host, fl); + nlmclnt_prepare_block(&block, host, fl); again: /* * Initialise resp->status to a valid non-zero value, * since 0 == nlm_lck_granted */ resp->status = nlm_lck_blocked; - for(;;) { + + /* + * A GRANTED callback can come at any time -- even before the reply + * to the LOCK request arrives, so we queue the wait before + * requesting the lock. + */ + nlmclnt_queue_block(&block); + for (;;) { /* Reboot protection */ fl->fl_u.nfs_fl.state = host->h_state; status = nlmclnt_call(cred, req, NLMPROC_LOCK); if (status < 0) break; /* Did a reclaimer thread notify us of a server reboot? */ - if (resp->status == nlm_lck_denied_grace_period) + if (resp->status == nlm_lck_denied_grace_period) continue; if (resp->status != nlm_lck_blocked) break; /* Wait on an NLM blocking lock */ - status = nlmclnt_block(block, req, NLMCLNT_POLL_TIMEOUT); + status = nlmclnt_wait(&block, req, NLMCLNT_POLL_TIMEOUT); if (status < 0) break; - if (resp->status != nlm_lck_blocked) + if (block.b_status != nlm_lck_blocked) break; } + b_status = nlmclnt_dequeue_block(&block); + if (resp->status == nlm_lck_blocked) + resp->status = b_status; /* if we were interrupted while blocking, then cancel the lock request * and exit @@ -564,7 +580,7 @@ again: if (!req->a_args.block) goto out_unlock; if (nlmclnt_cancel(host, req->a_args.block, fl) == 0) - goto out_unblock; + goto out; } if (resp->status == nlm_granted) { @@ -593,16 +609,19 @@ again: status = -ENOLCK; else status = nlm_stat_to_errno(resp->status); -out_unblock: - nlmclnt_finish_block(block); out: + trace_nlmclnt_lock(&req->a_args.lock, + (const struct sockaddr *)&req->a_host->h_addr, + req->a_host->h_addrlen, req->a_res.status); nlmclnt_release_call(req); return status; out_unlock: /* Fatal error: ensure that we remove the lock altogether */ + trace_nlmclnt_lock(&req->a_args.lock, + (const struct sockaddr *)&req->a_host->h_addr, + req->a_host->h_addrlen, req->a_res.status); dprintk("lockd: lock attempt ended in fatal error.\n" " Attempting to unlock.\n"); - nlmclnt_finish_block(block); fl_type = fl->fl_type; fl->fl_type = F_UNLCK; down_read(&host->h_rwsem); @@ -696,6 +715,9 @@ nlmclnt_unlock(struct nlm_rqst *req, struct file_lock *fl) /* What to do now? I'm out of my depth... */ status = -ENOLCK; out: + trace_nlmclnt_unlock(&req->a_args.lock, + (const struct sockaddr *)&req->a_host->h_addr, + req->a_host->h_addrlen, req->a_res.status); nlmclnt_release_call(req); return status; } diff --git a/fs/lockd/host.c b/fs/lockd/host.c index cdc8e12cdac4..127a728fcbc8 100644 --- a/fs/lockd/host.c +++ b/fs/lockd/host.c @@ -629,6 +629,7 @@ nlm_shutdown_hosts_net(struct net *net) rpc_shutdown_client(host->h_rpcclnt); host->h_rpcclnt = NULL; } + nlmsvc_free_host_resources(host); } /* Then, perform a garbage collection pass */ diff --git a/fs/lockd/svclock.c b/fs/lockd/svclock.c index 4e30f3c50970..c43ccdf28ed9 100644 --- a/fs/lockd/svclock.c +++ b/fs/lockd/svclock.c @@ -954,19 +954,32 @@ void nlmsvc_grant_reply(struct nlm_cookie *cookie, __be32 status) { struct nlm_block *block; + struct file_lock *fl; + int error; dprintk("grant_reply: looking for cookie %x, s=%d \n", *(unsigned int *)(cookie->data), status); if (!(block = nlmsvc_find_block(cookie))) return; - if (status == nlm_lck_denied_grace_period) { + switch (status) { + case nlm_lck_denied_grace_period: /* Try again in a couple of seconds */ nlmsvc_insert_block(block, 10 * HZ); - } else { + break; + case nlm_lck_denied: + /* Client doesn't want it, just unlock it */ + nlmsvc_unlink_block(block); + fl = &block->b_call->a_args.lock.fl; + fl->fl_type = F_UNLCK; + error = vfs_lock_file(fl->fl_file, F_SETLK, fl, NULL); + if (error) + pr_warn("lockd: unable to unlock lock rejected by client!\n"); + break; + default: /* - * Lock is now held by client, or has been rejected. - * In both cases, the block should be removed. + * Either it was accepted or the status makes no sense + * just unlink it either way. */ nlmsvc_unlink_block(block); } diff --git a/fs/lockd/trace.c b/fs/lockd/trace.c new file mode 100644 index 000000000000..d9a6ff6e673c --- /dev/null +++ b/fs/lockd/trace.c @@ -0,0 +1,3 @@ +// SPDX-License-Identifier: GPL-2.0 +#define CREATE_TRACE_POINTS +#include "trace.h" diff --git a/fs/lockd/trace.h b/fs/lockd/trace.h new file mode 100644 index 000000000000..7461b13b6e74 --- /dev/null +++ b/fs/lockd/trace.h @@ -0,0 +1,106 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +#undef TRACE_SYSTEM +#define TRACE_SYSTEM lockd + +#if !defined(_TRACE_LOCKD_H) || defined(TRACE_HEADER_MULTI_READ) +#define _TRACE_LOCKD_H + +#include <linux/tracepoint.h> +#include <linux/crc32.h> +#include <linux/nfs.h> +#include <linux/lockd/lockd.h> + +#ifdef CONFIG_LOCKD_V4 +#define NLM_STATUS_LIST \ + nlm_status_code(LCK_GRANTED) \ + nlm_status_code(LCK_DENIED) \ + nlm_status_code(LCK_DENIED_NOLOCKS) \ + nlm_status_code(LCK_BLOCKED) \ + nlm_status_code(LCK_DENIED_GRACE_PERIOD) \ + nlm_status_code(DEADLCK) \ + nlm_status_code(ROFS) \ + nlm_status_code(STALE_FH) \ + nlm_status_code(FBIG) \ + nlm_status_code_end(FAILED) +#else +#define NLM_STATUS_LIST \ + nlm_status_code(LCK_GRANTED) \ + nlm_status_code(LCK_DENIED) \ + nlm_status_code(LCK_DENIED_NOLOCKS) \ + nlm_status_code(LCK_BLOCKED) \ + nlm_status_code_end(LCK_DENIED_GRACE_PERIOD) +#endif + +#undef nlm_status_code +#undef nlm_status_code_end +#define nlm_status_code(x) TRACE_DEFINE_ENUM(NLM_##x); +#define nlm_status_code_end(x) TRACE_DEFINE_ENUM(NLM_##x); + +NLM_STATUS_LIST + +#undef nlm_status_code +#undef nlm_status_code_end +#define nlm_status_code(x) { NLM_##x, #x }, +#define nlm_status_code_end(x) { NLM_##x, #x } + +#define show_nlm_status(x) __print_symbolic(x, NLM_STATUS_LIST) + +DECLARE_EVENT_CLASS(nlmclnt_lock_event, + TP_PROTO( + const struct nlm_lock *lock, + const struct sockaddr *addr, + unsigned int addrlen, + __be32 status + ), + + TP_ARGS(lock, addr, addrlen, status), + + TP_STRUCT__entry( + __field(u32, oh) + __field(u32, svid) + __field(u32, fh) + __field(unsigned long, status) + __field(u64, start) + __field(u64, len) + __sockaddr(addr, addrlen) + ), + + TP_fast_assign( + __entry->oh = ~crc32_le(0xffffffff, lock->oh.data, lock->oh.len); + __entry->svid = lock->svid; + __entry->fh = nfs_fhandle_hash(&lock->fh); + __entry->start = lock->lock_start; + __entry->len = lock->lock_len; + __entry->status = be32_to_cpu(status); + __assign_sockaddr(addr, addr, addrlen); + ), + + TP_printk( + "addr=%pISpc oh=0x%08x svid=0x%08x fh=0x%08x start=%llu len=%llu status=%s", + __get_sockaddr(addr), __entry->oh, __entry->svid, + __entry->fh, __entry->start, __entry->len, + show_nlm_status(__entry->status) + ) +); + +#define DEFINE_NLMCLNT_EVENT(name) \ + DEFINE_EVENT(nlmclnt_lock_event, name, \ + TP_PROTO( \ + const struct nlm_lock *lock, \ + const struct sockaddr *addr, \ + unsigned int addrlen, \ + __be32 status \ + ), \ + TP_ARGS(lock, addr, addrlen, status)) + +DEFINE_NLMCLNT_EVENT(nlmclnt_test); +DEFINE_NLMCLNT_EVENT(nlmclnt_lock); +DEFINE_NLMCLNT_EVENT(nlmclnt_unlock); +DEFINE_NLMCLNT_EVENT(nlmclnt_grant); + +#endif /* _TRACE_LOCKD_H */ + +#undef TRACE_INCLUDE_PATH +#define TRACE_INCLUDE_PATH . +#define TRACE_INCLUDE_FILE trace +#include <trace/define_trace.h> diff --git a/fs/nfs/export.c b/fs/nfs/export.c index d6a6d1ebb8fd..be686b8e0c54 100644 --- a/fs/nfs/export.c +++ b/fs/nfs/export.c @@ -149,7 +149,10 @@ const struct export_operations nfs_export_ops = { .encode_fh = nfs_encode_fh, .fh_to_dentry = nfs_fh_to_dentry, .get_parent = nfs_get_parent, - .flags = EXPORT_OP_NOWCC|EXPORT_OP_NOSUBTREECHK| - EXPORT_OP_CLOSE_BEFORE_UNLINK|EXPORT_OP_REMOTE_FS| - EXPORT_OP_NOATOMIC_ATTR, + .flags = EXPORT_OP_NOWCC | + EXPORT_OP_NOSUBTREECHK | + EXPORT_OP_CLOSE_BEFORE_UNLINK | + EXPORT_OP_REMOTE_FS | + EXPORT_OP_NOATOMIC_ATTR | + EXPORT_OP_FLUSH_ON_CLOSE, }; diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index 9ba07553eab8..3cc027d3bd58 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -855,27 +855,12 @@ u64 nfs_timespec_to_change_attr(const struct timespec64 *ts) } #ifdef CONFIG_CRC32 -/** - * nfs_fhandle_hash - calculate the crc32 hash for the filehandle - * @fh - pointer to filehandle - * - * returns a crc32 hash for the filehandle that is compatible with - * the one displayed by "wireshark". - */ -static inline u32 nfs_fhandle_hash(const struct nfs_fh *fh) -{ - return ~crc32_le(0xFFFFFFFF, &fh->data[0], fh->size); -} static inline u32 nfs_stateid_hash(const nfs4_stateid *stateid) { return ~crc32_le(0xFFFFFFFF, &stateid->other[0], NFS4_STATEID_OTHER_SIZE); } #else -static inline u32 nfs_fhandle_hash(const struct nfs_fh *fh) -{ - return 0; -} static inline u32 nfs_stateid_hash(nfs4_stateid *stateid) { return 0; diff --git a/fs/nfsd/export.c b/fs/nfsd/export.c index 668c7527b17e..ae85257b4238 100644 --- a/fs/nfsd/export.c +++ b/fs/nfsd/export.c @@ -123,11 +123,11 @@ static int expkey_parse(struct cache_detail *cd, char *mesg, int mlen) /* OK, we seem to have a valid key */ key.h.flags = 0; - key.h.expiry_time = get_expiry(&mesg); - if (key.h.expiry_time == 0) + err = get_expiry(&mesg, &key.h.expiry_time); + if (err) goto out; - key.ek_client = dom; + key.ek_client = dom; key.ek_fsidtype = fsidtype; memcpy(key.ek_fsid, buf, len); @@ -439,7 +439,6 @@ static int check_export(struct path *path, int *flags, unsigned char *uuid) return -EINVAL; } return 0; - } #ifdef CONFIG_NFSD_V4 @@ -546,6 +545,29 @@ static inline int secinfo_parse(char **mesg, char *buf, struct svc_export *exp) { return 0; } #endif +static int xprtsec_parse(char **mesg, char *buf, struct svc_export *exp) +{ + unsigned int i, mode, listsize; + int err; + + err = get_uint(mesg, &listsize); + if (err) + return err; + if (listsize > NFSEXP_XPRTSEC_NUM) + return -EINVAL; + + exp->ex_xprtsec_modes = 0; + for (i = 0; i < listsize; i++) { + err = get_uint(mesg, &mode); + if (err) + return err; + if (mode > NFSEXP_XPRTSEC_MTLS) + return -EINVAL; + exp->ex_xprtsec_modes |= mode; + } + return 0; +} + static inline int nfsd_uuid_parse(char **mesg, char *buf, unsigned char **puuid) { @@ -608,11 +630,11 @@ static int svc_export_parse(struct cache_detail *cd, char *mesg, int mlen) exp.ex_client = dom; exp.cd = cd; exp.ex_devid_map = NULL; + exp.ex_xprtsec_modes = NFSEXP_XPRTSEC_ALL; /* expiry */ - err = -EINVAL; - exp.h.expiry_time = get_expiry(&mesg); - if (exp.h.expiry_time == 0) + err = get_expiry(&mesg, &exp.h.expiry_time); + if (err) goto out3; /* flags */ @@ -624,7 +646,7 @@ static int svc_export_parse(struct cache_detail *cd, char *mesg, int mlen) if (err || an_int < 0) goto out3; exp.ex_flags= an_int; - + /* anon uid */ err = get_int(&mesg, &an_int); if (err) @@ -650,6 +672,8 @@ static int svc_export_parse(struct cache_detail *cd, char *mesg, int mlen) err = nfsd_uuid_parse(&mesg, buf, &exp.ex_uuid); else if (strcmp(buf, "secinfo") == 0) err = secinfo_parse(&mesg, buf, &exp); + else if (strcmp(buf, "xprtsec") == 0) + err = xprtsec_parse(&mesg, buf, &exp); else /* quietly ignore unknown words and anything * following. Newer user-space can try to set @@ -663,6 +687,7 @@ static int svc_export_parse(struct cache_detail *cd, char *mesg, int mlen) err = check_export(&exp.ex_path, &exp.ex_flags, exp.ex_uuid); if (err) goto out4; + /* * No point caching this if it would immediately expire. * Also, this protects exportfs's dummy export from the @@ -824,6 +849,7 @@ static void export_update(struct cache_head *cnew, struct cache_head *citem) for (i = 0; i < MAX_SECINFO_LIST; i++) { new->ex_flavors[i] = item->ex_flavors[i]; } + new->ex_xprtsec_modes = item->ex_xprtsec_modes; } static struct cache_head *svc_export_alloc(void) @@ -1035,9 +1061,26 @@ static struct svc_export *exp_find(struct cache_detail *cd, __be32 check_nfsd_access(struct svc_export *exp, struct svc_rqst *rqstp) { - struct exp_flavor_info *f; - struct exp_flavor_info *end = exp->ex_flavors + exp->ex_nflavors; + struct exp_flavor_info *f, *end = exp->ex_flavors + exp->ex_nflavors; + struct svc_xprt *xprt = rqstp->rq_xprt; + + if (exp->ex_xprtsec_modes & NFSEXP_XPRTSEC_NONE) { + if (!test_bit(XPT_TLS_SESSION, &xprt->xpt_flags)) + goto ok; + } + if (exp->ex_xprtsec_modes & NFSEXP_XPRTSEC_TLS) { + if (test_bit(XPT_TLS_SESSION, &xprt->xpt_flags) && + !test_bit(XPT_PEER_AUTH, &xprt->xpt_flags)) + goto ok; + } + if (exp->ex_xprtsec_modes & NFSEXP_XPRTSEC_MTLS) { + if (test_bit(XPT_TLS_SESSION, &xprt->xpt_flags) && + test_bit(XPT_PEER_AUTH, &xprt->xpt_flags)) + goto ok; + } + goto denied; +ok: /* legacy gss-only clients are always OK: */ if (exp->ex_client == rqstp->rq_gssclient) return 0; @@ -1062,6 +1105,7 @@ __be32 check_nfsd_access(struct svc_export *exp, struct svc_rqst *rqstp) if (nfsd4_spo_must_allow(rqstp)) return 0; +denied: return rqstp->rq_vers < 4 ? nfserr_acces : nfserr_wrongsec; } diff --git a/fs/nfsd/export.h b/fs/nfsd/export.h index d03f7f6a8642..2df8ae25aad3 100644 --- a/fs/nfsd/export.h +++ b/fs/nfsd/export.h @@ -77,6 +77,7 @@ struct svc_export { struct cache_detail *cd; struct rcu_head ex_rcu; struct export_stats ex_stats; + unsigned long ex_xprtsec_modes; }; /* an "export key" (expkey) maps a filehandlefragement to an diff --git a/fs/nfsd/filecache.c b/fs/nfsd/filecache.c index 6e8712bd7c99..ee9c923192e0 100644 --- a/fs/nfsd/filecache.c +++ b/fs/nfsd/filecache.c @@ -74,70 +74,9 @@ static struct list_lru nfsd_file_lru; static unsigned long nfsd_file_flags; static struct fsnotify_group *nfsd_file_fsnotify_group; static struct delayed_work nfsd_filecache_laundrette; -static struct rhashtable nfsd_file_rhash_tbl +static struct rhltable nfsd_file_rhltable ____cacheline_aligned_in_smp; -enum nfsd_file_lookup_type { - NFSD_FILE_KEY_INODE, - NFSD_FILE_KEY_FULL, -}; - -struct nfsd_file_lookup_key { - struct inode *inode; - struct net *net; - const struct cred *cred; - unsigned char need; - bool gc; - enum nfsd_file_lookup_type type; -}; - -/* - * The returned hash value is based solely on the address of an in-code - * inode, a pointer to a slab-allocated object. The entropy in such a - * pointer is concentrated in its middle bits. - */ -static u32 nfsd_file_inode_hash(const struct inode *inode, u32 seed) -{ - unsigned long ptr = (unsigned long)inode; - u32 k; - - k = ptr >> L1_CACHE_SHIFT; - k &= 0x00ffffff; - return jhash2(&k, 1, seed); -} - -/** - * nfsd_file_key_hashfn - Compute the hash value of a lookup key - * @data: key on which to compute the hash value - * @len: rhash table's key_len parameter (unused) - * @seed: rhash table's random seed of the day - * - * Return value: - * Computed 32-bit hash value - */ -static u32 nfsd_file_key_hashfn(const void *data, u32 len, u32 seed) -{ - const struct nfsd_file_lookup_key *key = data; - - return nfsd_file_inode_hash(key->inode, seed); -} - -/** - * nfsd_file_obj_hashfn - Compute the hash value of an nfsd_file - * @data: object on which to compute the hash value - * @len: rhash table's key_len parameter (unused) - * @seed: rhash table's random seed of the day - * - * Return value: - * Computed 32-bit hash value - */ -static u32 nfsd_file_obj_hashfn(const void *data, u32 len, u32 seed) -{ - const struct nfsd_file *nf = data; - - return nfsd_file_inode_hash(nf->nf_inode, seed); -} - static bool nfsd_match_cred(const struct cred *c1, const struct cred *c2) { @@ -158,53 +97,16 @@ nfsd_match_cred(const struct cred *c1, const struct cred *c2) return true; } -/** - * nfsd_file_obj_cmpfn - Match a cache item against search criteria - * @arg: search criteria - * @ptr: cache item to check - * - * Return values: - * %0 - Item matches search criteria - * %1 - Item does not match search criteria - */ -static int nfsd_file_obj_cmpfn(struct rhashtable_compare_arg *arg, - const void *ptr) -{ - const struct nfsd_file_lookup_key *key = arg->key; - const struct nfsd_file *nf = ptr; - - switch (key->type) { - case NFSD_FILE_KEY_INODE: - if (nf->nf_inode != key->inode) - return 1; - break; - case NFSD_FILE_KEY_FULL: - if (nf->nf_inode != key->inode) - return 1; - if (nf->nf_may != key->need) - return 1; - if (nf->nf_net != key->net) - return 1; - if (!nfsd_match_cred(nf->nf_cred, key->cred)) - return 1; - if (!!test_bit(NFSD_FILE_GC, &nf->nf_flags) != key->gc) - return 1; - if (test_bit(NFSD_FILE_HASHED, &nf->nf_flags) == 0) - return 1; - break; - } - return 0; -} - static const struct rhashtable_params nfsd_file_rhash_params = { .key_len = sizeof_field(struct nfsd_file, nf_inode), .key_offset = offsetof(struct nfsd_file, nf_inode), - .head_offset = offsetof(struct nfsd_file, nf_rhash), - .hashfn = nfsd_file_key_hashfn, - .obj_hashfn = nfsd_file_obj_hashfn, - .obj_cmpfn = nfsd_file_obj_cmpfn, - /* Reduce resizing churn on light workloads */ - .min_size = 512, /* buckets */ + .head_offset = offsetof(struct nfsd_file, nf_rlist), + + /* + * Start with a single page hash table to reduce resizing churn + * on light workloads. + */ + .min_size = 256, .automatic_shrinking = true, }; @@ -307,27 +209,27 @@ nfsd_file_mark_find_or_create(struct nfsd_file *nf, struct inode *inode) } static struct nfsd_file * -nfsd_file_alloc(struct nfsd_file_lookup_key *key, unsigned int may) +nfsd_file_alloc(struct net *net, struct inode *inode, unsigned char need, + bool want_gc) { struct nfsd_file *nf; nf = kmem_cache_alloc(nfsd_file_slab, GFP_KERNEL); - if (nf) { - INIT_LIST_HEAD(&nf->nf_lru); - nf->nf_birthtime = ktime_get(); - nf->nf_file = NULL; - nf->nf_cred = get_current_cred(); - nf->nf_net = key->net; - nf->nf_flags = 0; - __set_bit(NFSD_FILE_HASHED, &nf->nf_flags); - __set_bit(NFSD_FILE_PENDING, &nf->nf_flags); - if (key->gc) - __set_bit(NFSD_FILE_GC, &nf->nf_flags); - nf->nf_inode = key->inode; - refcount_set(&nf->nf_ref, 1); - nf->nf_may = key->need; - nf->nf_mark = NULL; - } + if (unlikely(!nf)) + return NULL; + + INIT_LIST_HEAD(&nf->nf_lru); + nf->nf_birthtime = ktime_get(); + nf->nf_file = NULL; + nf->nf_cred = get_current_cred(); + nf->nf_net = net; + nf->nf_flags = want_gc ? + BIT(NFSD_FILE_HASHED) | BIT(NFSD_FILE_PENDING) | BIT(NFSD_FILE_GC) : + BIT(NFSD_FILE_HASHED) | BIT(NFSD_FILE_PENDING); + nf->nf_inode = inode; + refcount_set(&nf->nf_ref, 1); + nf->nf_may = need; + nf->nf_mark = NULL; return nf; } @@ -352,8 +254,8 @@ static void nfsd_file_hash_remove(struct nfsd_file *nf) { trace_nfsd_file_unhash(nf); - rhashtable_remove_fast(&nfsd_file_rhash_tbl, &nf->nf_rhash, - nfsd_file_rhash_params); + rhltable_remove(&nfsd_file_rhltable, &nf->nf_rlist, + nfsd_file_rhash_params); } static bool @@ -380,10 +282,8 @@ nfsd_file_free(struct nfsd_file *nf) if (nf->nf_mark) nfsd_file_mark_put(nf->nf_mark); if (nf->nf_file) { - get_file(nf->nf_file); - filp_close(nf->nf_file, NULL); nfsd_file_check_write_error(nf); - fput(nf->nf_file); + filp_close(nf->nf_file, NULL); } /* @@ -402,13 +302,23 @@ nfsd_file_check_writeback(struct nfsd_file *nf) struct file *file = nf->nf_file; struct address_space *mapping; - if (!file || !(file->f_mode & FMODE_WRITE)) + /* File not open for write? */ + if (!(file->f_mode & FMODE_WRITE)) + return false; + + /* + * Some filesystems (e.g. NFS) flush all dirty data on close. + * On others, there is no need to wait for writeback. + */ + if (!(file_inode(file)->i_sb->s_export_op->flags & EXPORT_OP_FLUSH_ON_CLOSE)) return false; + mapping = file->f_mapping; return mapping_tagged(mapping, P |