*/
#include <linux/module.h>
-
+#include <linux/sched.h>
#include <linux/time.h>
#include <linux/errno.h>
#include <linux/nfs.h>
#include <linux/slab.h>
#include <linux/smp.h>
#include <linux/smp_lock.h>
+#include <linux/freezer.h>
#include <linux/fs_struct.h>
+#include <linux/kthread.h>
#include <linux/sunrpc/types.h>
#include <linux/sunrpc/stats.h>
#define NFSDDBG_FACILITY NFSDDBG_SVC
-/* these signals will be delivered to an nfsd thread
- * when handling a request
- */
-#define ALLOWED_SIGS (sigmask(SIGKILL))
-/* these signals will be delivered to an nfsd thread
- * when not handling a request. i.e. when waiting
- */
-#define SHUTDOWN_SIGS (sigmask(SIGKILL) | sigmask(SIGHUP) | sigmask(SIGINT) | sigmask(SIGQUIT))
-/* if the last thread dies with SIGHUP, then the exports table is
- * left unchanged ( like 2.4-{0-9} ). Any other signal will clear
- * the exports table (like 2.2).
- */
-#define SIG_NOCLEAN SIGHUP
-
extern struct svc_program nfsd_program;
-static void nfsd(struct svc_rqst *rqstp);
+static int nfsd(void *vrqstp);
struct timeval nfssvc_boot;
- struct svc_serv *nfsd_serv;
static atomic_t nfsd_busy;
static unsigned long nfsd_last_call;
static DEFINE_SPINLOCK(nfsd_call_lock);
-struct nfsd_list {
- struct list_head list;
- struct task_struct *task;
-};
-static struct list_head nfsd_list = LIST_HEAD_INIT(nfsd_list);
+/*
+ * nfsd_mutex protects nfsd_serv -- both the pointer itself and the members
+ * of the svc_serv struct. In particular, ->sv_nrthreads but also to some
+ * extent ->sv_temp_socks and ->sv_permsocks. It also protects nfsdstats.th_cnt
+ *
+ * If (out side the lock) nfsd_serv is non-NULL, then it must point to a
+ * properly initialised 'struct svc_serv' with ->sv_nrthreads > 0. That number
+ * of nfsd threads must exist and each must listed in ->sp_all_threads in each
+ * entry of ->sv_pools[].
+ *
+ * Transitions of the thread count between zero and non-zero are of particular
+ * interest since the svc_serv needs to be created and initialized at that
+ * point, or freed.
+ *
+ * Finally, the nfsd_mutex also protects some of the global variables that are
+ * accessed when nfsd starts and that are settable via the write_* routines in
+ * nfsctl.c. In particular:
+ *
+ * user_recovery_dirname
+ * user_lease_time
+ * nfsd_versions
+ */
+DEFINE_MUTEX(nfsd_mutex);
+struct svc_serv *nfsd_serv;
#if defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL)
static struct svc_stat nfsd_acl_svcstats;
.pg_prog = NFS_ACL_PROGRAM,
.pg_nvers = NFSD_ACL_NRVERS,
.pg_vers = nfsd_acl_versions,
- .pg_name = "nfsd",
+ .pg_name = "nfsacl",
.pg_class = "nfsd",
.pg_stats = &nfsd_acl_svcstats,
.pg_authenticate = &svc_set_client,
switch(change) {
case NFSD_SET:
nfsd_versions[vers] = nfsd_version[vers];
- break;
#if defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL)
if (vers < NFSD_ACL_NRVERS)
- nfsd_acl_version[vers] = nfsd_acl_version[vers];
+ nfsd_acl_versions[vers] = nfsd_acl_version[vers];
#endif
+ break;
case NFSD_CLEAR:
nfsd_versions[vers] = NULL;
#if defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL)
if (vers < NFSD_ACL_NRVERS)
- nfsd_acl_version[vers] = NULL;
+ nfsd_acl_versions[vers] = NULL;
#endif
break;
case NFSD_TEST:
int nfsd_nrthreads(void)
{
- if (nfsd_serv == NULL)
- return 0;
- else
- return nfsd_serv->sv_nrthreads;
+ int rv = 0;
+ mutex_lock(&nfsd_mutex);
+ if (nfsd_serv)
+ rv = nfsd_serv->sv_nrthreads;
+ mutex_unlock(&nfsd_mutex);
+ return rv;
}
-static int killsig; /* signal that was used to kill last nfsd */
static void nfsd_last_thread(struct svc_serv *serv)
{
/* When last nfsd thread exits we need to do some clean-up */
- struct svc_sock *svsk;
- list_for_each_entry(svsk, &serv->sv_permsocks, sk_list)
+ struct svc_xprt *xprt;
+ list_for_each_entry(xprt, &serv->sv_permsocks, xpt_list)
lockd_down();
nfsd_serv = NULL;
nfsd_racache_shutdown();
nfs4_state_shutdown();
- printk(KERN_WARNING "nfsd: last server has exited\n");
- if (killsig != SIG_NOCLEAN) {
- printk(KERN_WARNING "nfsd: unexporting all filesystems\n");
- nfsd_export_flush();
- }
+ printk(KERN_WARNING "nfsd: last server has exited, flushing export "
+ "cache\n");
+ nfsd_export_flush();
}
void nfsd_reset_versions(void)
}
}
+
int nfsd_create_serv(void)
{
int err = 0;
- lock_kernel();
+
+ WARN_ON(!mutex_is_locked(&nfsd_mutex));
if (nfsd_serv) {
svc_get(nfsd_serv);
- unlock_kernel();
return 0;
}
+ if (nfsd_max_blksize == 0) {
+ /* choose a suitable default */
+ struct sysinfo i;
+ si_meminfo(&i);
+ /* Aim for 1/4096 of memory per thread
+ * This gives 1MB on 4Gig machines
+ * But only uses 32K on 128M machines.
+ * Bottom out at 8K on 32M and smaller.
+ * Of course, this is only a default.
+ */
+ nfsd_max_blksize = NFSSVC_MAXBLKSIZE;
+ i.totalram <<= PAGE_SHIFT - 12;
+ while (nfsd_max_blksize > i.totalram &&
+ nfsd_max_blksize >= 8*1024*2)
+ nfsd_max_blksize /= 2;
+ }
atomic_set(&nfsd_busy, 0);
- nfsd_serv = svc_create(&nfsd_program, NFSD_BUFSIZE,
- nfsd_last_thread);
+ nfsd_serv = svc_create_pooled(&nfsd_program, nfsd_max_blksize,
+ AF_INET,
+ nfsd_last_thread, nfsd, THIS_MODULE);
if (nfsd_serv == NULL)
err = -ENOMEM;
- unlock_kernel();
+
do_gettimeofday(&nfssvc_boot); /* record boot time */
return err;
}
if (!list_empty(&nfsd_serv->sv_permsocks))
return 0;
- error = lockd_up(IPPROTO_UDP);
- if (error >= 0) {
- error = svc_makesock(nfsd_serv, IPPROTO_UDP, port);
- if (error < 0)
- lockd_down();
- }
+ error = svc_create_xprt(nfsd_serv, "udp", port,
+ SVC_SOCK_DEFAULTS);
if (error < 0)
return error;
-#ifdef CONFIG_NFSD_TCP
- error = lockd_up(IPPROTO_TCP);
- if (error >= 0) {
- error = svc_makesock(nfsd_serv, IPPROTO_TCP, port);
- if (error < 0)
- lockd_down();
- }
+ error = svc_create_xprt(nfsd_serv, "tcp", port,
+ SVC_SOCK_DEFAULTS);
if (error < 0)
return error;
-#endif
+
+ error = lockd_up();
+ if (error < 0)
+ return error;
+
+ return 0;
+}
+
+int nfsd_nrpools(void)
+{
+ if (nfsd_serv == NULL)
+ return 0;
+ else
+ return nfsd_serv->sv_nrpools;
+}
+
+int nfsd_get_nrthreads(int n, int *nthreads)
+{
+ int i = 0;
+
+ if (nfsd_serv != NULL) {
+ for (i = 0; i < nfsd_serv->sv_nrpools && i < n; i++)
+ nthreads[i] = nfsd_serv->sv_pools[i].sp_nrthreads;
+ }
+
return 0;
}
+int nfsd_set_nrthreads(int n, int *nthreads)
+{
+ int i = 0;
+ int tot = 0;
+ int err = 0;
+
+ WARN_ON(!mutex_is_locked(&nfsd_mutex));
+
+ if (nfsd_serv == NULL || n <= 0)
+ return 0;
+
+ if (n > nfsd_serv->sv_nrpools)
+ n = nfsd_serv->sv_nrpools;
+
+ /* enforce a global maximum number of threads */
+ tot = 0;
+ for (i = 0; i < n; i++) {
+ if (nthreads[i] > NFSD_MAXSERVS)
+ nthreads[i] = NFSD_MAXSERVS;
+ tot += nthreads[i];
+ }
+ if (tot > NFSD_MAXSERVS) {
+ /* total too large: scale down requested numbers */
+ for (i = 0; i < n && tot > 0; i++) {
+ int new = nthreads[i] * NFSD_MAXSERVS / tot;
+ tot -= (nthreads[i] - new);
+ nthreads[i] = new;
+ }
+ for (i = 0; i < n && tot > 0; i++) {
+ nthreads[i]--;
+ tot--;
+ }
+ }
+
+ /*
+ * There must always be a thread in pool 0; the admin
+ * can't shut down NFS completely using pool_threads.
+ */
+ if (nthreads[0] == 0)
+ nthreads[0] = 1;
+
+ /* apply the new numbers */
+ svc_get(nfsd_serv);
+ for (i = 0; i < n; i++) {
+ err = svc_set_num_threads(nfsd_serv, &nfsd_serv->sv_pools[i],
+ nthreads[i]);
+ if (err)
+ break;
+ }
+ svc_destroy(nfsd_serv);
+
+ return err;
+}
+
int
nfsd_svc(unsigned short port, int nrservs)
{
int error;
- struct list_head *victim;
-
- lock_kernel();
+
+ mutex_lock(&nfsd_mutex);
dprintk("nfsd: creating service\n");
error = -EINVAL;
if (nrservs <= 0)
error = nfsd_racache_init(2*nrservs);
if (error<0)
goto out;
- error = nfs4_state_start();
- if (error<0)
- goto out;
+ nfs4_state_start();
nfsd_reset_versions();
if (error)
goto failure;
- nrservs -= (nfsd_serv->sv_nrthreads-1);
- while (nrservs > 0) {
- nrservs--;
- __module_get(THIS_MODULE);
- error = svc_create_thread(nfsd, nfsd_serv);
- if (error < 0) {
- module_put(THIS_MODULE);
- break;
- }
- }
- victim = nfsd_list.next;
- while (nrservs < 0 && victim != &nfsd_list) {
- struct nfsd_list *nl =
- list_entry(victim,struct nfsd_list, list);
- victim = victim->next;
- send_sig(SIG_NOCLEAN, nl->task, 1);
- nrservs++;
- }
+ error = svc_set_num_threads(nfsd_serv, NULL, nrservs);
failure:
svc_destroy(nfsd_serv); /* Release server */
out:
- unlock_kernel();
+ mutex_unlock(&nfsd_mutex);
return error;
}
/*
* This is the NFS server kernel thread
*/
-static void
-nfsd(struct svc_rqst *rqstp)
+static int
+nfsd(void *vrqstp)
{
+ struct svc_rqst *rqstp = (struct svc_rqst *) vrqstp;
struct fs_struct *fsp;
- int err;
- struct nfsd_list me;
- sigset_t shutdown_mask, allowed_mask;
+ int err, preverr = 0;
/* Lock module and set up kernel thread */
- lock_kernel();
- daemonize("nfsd");
+ mutex_lock(&nfsd_mutex);
- /* After daemonize() this kernel thread shares current->fs
+ /* At this point, the thread shares current->fs
* with the init process. We need to create files with a
* umask of 0 instead of init's umask. */
fsp = copy_fs_struct(current->fs);
current->fs = fsp;
current->fs->umask = 0;
- siginitsetinv(&shutdown_mask, SHUTDOWN_SIGS);
- siginitsetinv(&allowed_mask, ALLOWED_SIGS);
+ /*
+ * thread is spawned with all signals set to SIG_IGN, re-enable
+ * the ones that will bring down the thread
+ */
+ allow_signal(SIGKILL);
+ allow_signal(SIGHUP);
+ allow_signal(SIGINT);
+ allow_signal(SIGQUIT);
nfsdstats.th_cnt++;
-
- me.task = current;
- list_add(&me.list, &nfsd_list);
-
- unlock_kernel();
+ mutex_unlock(&nfsd_mutex);
/*
* We want less throttling in balance_dirty_pages() so that nfs to
* dirty pages.
*/
current->flags |= PF_LESS_THROTTLE;
+ set_freezable();
/*
* The main request loop
*/
for (;;) {
- /* Block all but the shutdown signals */
- sigprocmask(SIG_SETMASK, &shutdown_mask, NULL);
-
/*
* Find a socket with data available and call its
* recvfrom routine.
*/
while ((err = svc_recv(rqstp, 60*60*HZ)) == -EAGAIN)
;
- if (err < 0)
+ if (err == -EINTR)
break;
+ else if (err < 0) {
+ if (err != preverr) {
+ printk(KERN_WARNING "%s: unexpected error "
+ "from svc_recv (%d)\n", __func__, -err);
+ preverr = err;
+ }
+ schedule_timeout_uninterruptible(HZ);
+ continue;
+ }
+
update_thread_usage(atomic_read(&nfsd_busy));
atomic_inc(&nfsd_busy);
/* Lock the export hash tables for reading. */
exp_readlock();
- /* Process request with signals blocked. */
- sigprocmask(SIG_SETMASK, &allowed_mask, NULL);
-
svc_process(rqstp);
/* Unlock export hash tables */
atomic_dec(&nfsd_busy);
}
- if (err != -EINTR) {
- printk(KERN_WARNING "nfsd: terminating on error %d\n", -err);
- } else {
- unsigned int signo;
-
- for (signo = 1; signo <= _NSIG; signo++)
- if (sigismember(¤t->pending.signal, signo) &&
- !sigismember(¤t->blocked, signo))
- break;
- killsig = signo;
- }
/* Clear signals before calling svc_exit_thread() */
flush_signals(current);
- lock_kernel();
-
- list_del(&me.list);
+ mutex_lock(&nfsd_mutex);
nfsdstats.th_cnt --;
out:
svc_exit_thread(rqstp);
/* Release module */
- unlock_kernel();
+ mutex_unlock(&nfsd_mutex);
module_put_and_exit(0);
+ return 0;
+}
+
+static __be32 map_new_errors(u32 vers, __be32 nfserr)
+{
+ if (nfserr == nfserr_jukebox && vers == 2)
+ return nfserr_dropit;
+ if (nfserr == nfserr_wrongsec && vers < 4)
+ return nfserr_acces;
+ return nfserr;
}
int
-nfsd_dispatch(struct svc_rqst *rqstp, u32 *statp)
+nfsd_dispatch(struct svc_rqst *rqstp, __be32 *statp)
{
struct svc_procedure *proc;
kxdrproc_t xdr;
- u32 nfserr;
- u32 *nfserrp;
+ __be32 nfserr;
+ __be32 *nfserrp;
dprintk("nfsd_dispatch: vers %d proc %d\n",
rqstp->rq_vers, rqstp->rq_proc);
/* Decode arguments */
xdr = proc->pc_decode;
- if (xdr && !xdr(rqstp, (u32*)rqstp->rq_arg.head[0].iov_base,
+ if (xdr && !xdr(rqstp, (__be32*)rqstp->rq_arg.head[0].iov_base,
rqstp->rq_argp)) {
dprintk("nfsd: failed to decode arguments!\n");
nfsd_cache_update(rqstp, RC_NOCACHE, NULL);
*/
nfserrp = rqstp->rq_res.head[0].iov_base
+ rqstp->rq_res.head[0].iov_len;
- rqstp->rq_res.head[0].iov_len += sizeof(u32);
+ rqstp->rq_res.head[0].iov_len += sizeof(__be32);
/* Now call the procedure handler, and encode NFS status. */
nfserr = proc->pc_func(rqstp, rqstp->rq_argp, rqstp->rq_resp);
- if (nfserr == nfserr_jukebox && rqstp->rq_vers == 2)
- nfserr = nfserr_dropit;
+ nfserr = map_new_errors(rqstp->rq_vers, nfserr);
if (nfserr == nfserr_dropit) {
- dprintk("nfsd: Dropping request due to malloc failure!\n");
+ dprintk("nfsd: Dropping request; may be revisited later\n");
nfsd_cache_update(rqstp, RC_NOCACHE, NULL);
return 0;
}