/*
- * linux/fs/nfsd/nfssvc.c
- *
* Central processing for nfsd.
*
* Authors: Olaf Kirch (okir@monad.swb.de)
* Copyright (C) 1995, 1996, 1997 Olaf Kirch <okir@monad.swb.de>
*/
-#include <linux/module.h>
#include <linux/sched.h>
-#include <linux/time.h>
-#include <linux/errno.h>
-#include <linux/nfs.h>
-#include <linux/in.h>
-#include <linux/uio.h>
-#include <linux/unistd.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/swap.h>
-#include <linux/sunrpc/types.h>
#include <linux/sunrpc/stats.h>
-#include <linux/sunrpc/svc.h>
#include <linux/sunrpc/svcsock.h>
-#include <linux/sunrpc/cache.h>
-#include <linux/nfsd/nfsd.h>
-#include <linux/nfsd/stats.h>
-#include <linux/nfsd/cache.h>
-#include <linux/nfsd/syscall.h>
#include <linux/lockd/bind.h>
#include <linux/nfsacl.h>
+#include <linux/seq_file.h>
+#include "nfsd.h"
+#include "cache.h"
+#include "vfs.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);
+
+/*
+ * 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;
+
+/*
+ * nfsd_drc_lock protects nfsd_drc_max_pages and nfsd_drc_pages_used.
+ * nfsd_drc_max_pages limits the total amount of memory available for
+ * version 4.1 DRC caches.
+ * nfsd_drc_pages_used tracks the current version 4.1 DRC memory usage.
+ */
+spinlock_t nfsd_drc_lock;
+unsigned int nfsd_drc_max_mem;
+unsigned int nfsd_drc_mem_used;
#if defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL)
static struct svc_stat nfsd_acl_svcstats;
};
+u32 nfsd_supported_minorversion;
+
int nfsd_vers(int vers, enum vers_op change)
{
if (vers < NFSD_MINVERS || vers >= NFSD_NRVERS)
}
return 0;
}
+
+int nfsd_minorversion(u32 minorversion, enum vers_op change)
+{
+ if (minorversion > NFSD_SUPPORTED_MINOR_VERSION)
+ return -1;
+ switch(change) {
+ case NFSD_SET:
+ nfsd_supported_minorversion = minorversion;
+ break;
+ case NFSD_CLEAR:
+ if (minorversion == 0)
+ return -1;
+ nfsd_supported_minorversion = minorversion - 1;
+ break;
+ case NFSD_TEST:
+ return minorversion <= nfsd_supported_minorversion;
+ case NFSD_AVAIL:
+ return minorversion <= NFSD_SUPPORTED_MINOR_VERSION;
+ }
+ return 0;
+}
+
/*
* Maximum number of nfsd processes
*/
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)
}
}
+/*
+ * Each session guarantees a negotiated per slot memory cache for replies
+ * which in turn consumes memory beyond the v2/v3/v4.0 server. A dedicated
+ * NFSv4.1 server might want to use more memory for a DRC than a machine
+ * with mutiple services.
+ *
+ * Impose a hard limit on the number of pages for the DRC which varies
+ * according to the machines free pages. This is of course only a default.
+ *
+ * For now this is a #defined shift which could be under admin control
+ * in the future.
+ */
+static void set_max_drc(void)
+{
+ #define NFSD_DRC_SIZE_SHIFT 10
+ nfsd_drc_max_mem = (nr_free_buffer_pages()
+ >> NFSD_DRC_SIZE_SHIFT) * PAGE_SIZE;
+ nfsd_drc_mem_used = 0;
+ spin_lock_init(&nfsd_drc_lock);
+ dprintk("%s nfsd_drc_max_mem %u \n", __func__, nfsd_drc_max_mem);
+}
+
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) {
nfsd_max_blksize /= 2;
}
- atomic_set(&nfsd_busy, 0);
- nfsd_serv = svc_create_pooled(&nfsd_program,
- nfsd_max_blksize,
- nfsd_last_thread,
- nfsd, SIG_NOCLEAN, THIS_MODULE);
+ nfsd_serv = svc_create_pooled(&nfsd_program, nfsd_max_blksize,
+ nfsd_last_thread, nfsd, THIS_MODULE);
if (nfsd_serv == NULL)
err = -ENOMEM;
- unlock_kernel();
+ else
+ set_max_drc();
+
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,
+ error = svc_create_xprt(nfsd_serv, "udp", PF_INET, port,
SVC_SOCK_DEFAULTS);
- if (error < 0)
- lockd_down();
- }
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,
+ error = lockd_up();
+ if (error < 0)
+ return error;
+
+ error = svc_create_xprt(nfsd_serv, "tcp", PF_INET, port,
SVC_SOCK_DEFAULTS);
- if (error < 0)
- lockd_down();
- }
if (error < 0)
return error;
-#endif
+
+ error = lockd_up();
+ if (error < 0)
+ return error;
+
return 0;
}
int tot = 0;
int err = 0;
+ WARN_ON(!mutex_is_locked(&nfsd_mutex));
+
if (nfsd_serv == NULL || n <= 0)
return 0;
nthreads[0] = 1;
/* apply the new numbers */
- lock_kernel();
svc_get(nfsd_serv);
for (i = 0; i < n; i++) {
err = svc_set_num_threads(nfsd_serv, &nfsd_serv->sv_pools[i],
break;
}
svc_destroy(nfsd_serv);
- unlock_kernel();
return err;
}
nfsd_svc(unsigned short port, int nrservs)
{
int error;
-
- lock_kernel();
+
+ mutex_lock(&nfsd_mutex);
dprintk("nfsd: creating service\n");
- error = -EINVAL;
if (nrservs <= 0)
nrservs = 0;
if (nrservs > NFSD_MAXSERVS)
nrservs = NFSD_MAXSERVS;
-
+ error = 0;
+ if (nrservs == 0 && nfsd_serv == NULL)
+ goto out;
+
/* Readahead param cache - will no-op if it already exists */
error = nfsd_racache_init(2*nrservs);
if (error<0)
goto out;
error = nfs4_state_start();
- if (error<0)
+ if (error)
goto out;
nfsd_reset_versions();
goto failure;
error = svc_set_num_threads(nfsd_serv, NULL, nrservs);
+ if (error == 0)
+ /* We are holding a reference to nfsd_serv which
+ * we don't want to count in the return value,
+ * so subtract 1
+ */
+ error = nfsd_serv->sv_nrthreads - 1;
failure:
svc_destroy(nfsd_serv); /* Release server */
out:
- unlock_kernel();
+ mutex_unlock(&nfsd_mutex);
return error;
}
-static inline void
-update_thread_usage(int busy_threads)
-{
- unsigned long prev_call;
- unsigned long diff;
- int decile;
-
- spin_lock(&nfsd_call_lock);
- prev_call = nfsd_last_call;
- nfsd_last_call = jiffies;
- decile = busy_threads*10/nfsdstats.th_cnt;
- if (decile>0 && decile <= 10) {
- diff = nfsd_last_call - prev_call;
- if ( (nfsdstats.th_usage[decile-1] += diff) >= NFSD_USAGE_WRAP)
- nfsdstats.th_usage[decile-1] -= NFSD_USAGE_WRAP;
- if (decile == 10)
- nfsdstats.th_fullcnt++;
- }
- spin_unlock(&nfsd_call_lock);
-}
/*
* This is the NFS server kernel thread
*/
-static void
-nfsd(struct svc_rqst *rqstp)
+static int
+nfsd(void *vrqstp)
{
- struct fs_struct *fsp;
- int err;
- sigset_t shutdown_mask, allowed_mask;
+ struct svc_rqst *rqstp = (struct svc_rqst *) vrqstp;
+ 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);
- if (!fsp) {
+ if (unshare_fs_struct() < 0) {
printk("Unable to start nfsd thread: out of memory\n");
goto out;
}
- exit_fs(current);
- 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++;
-
- rqstp->rq_task = current;
-
- 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;
- update_thread_usage(atomic_read(&nfsd_busy));
- atomic_inc(&nfsd_busy);
+ 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;
+ }
+
/* Lock the export hash tables for reading. */
exp_readlock();
- /* Process request with signals blocked. */
- sigprocmask(SIG_SETMASK, &allowed_mask, NULL);
-
+ validate_process_creds();
svc_process(rqstp);
+ validate_process_creds();
/* Unlock export hash tables */
exp_readunlock();
- update_thread_usage(atomic_read(&nfsd_busy));
- 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();
-
+ 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
/* 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;
}
nfsd_cache_update(rqstp, proc->pc_cachetype, statp + 1);
return 1;
}
+
+int nfsd_pool_stats_open(struct inode *inode, struct file *file)
+{
+ int ret;
+ mutex_lock(&nfsd_mutex);
+ if (nfsd_serv == NULL) {
+ mutex_unlock(&nfsd_mutex);
+ return -ENODEV;
+ }
+ /* bump up the psudo refcount while traversing */
+ svc_get(nfsd_serv);
+ ret = svc_pool_stats_open(nfsd_serv, file);
+ mutex_unlock(&nfsd_mutex);
+ return ret;
+}
+
+int nfsd_pool_stats_release(struct inode *inode, struct file *file)
+{
+ int ret = seq_release(inode, file);
+ mutex_lock(&nfsd_mutex);
+ /* this function really, really should have been called svc_put() */
+ svc_destroy(nfsd_serv);
+ mutex_unlock(&nfsd_mutex);
+ return ret;
+}