[linux-2.6-block.git] / fs / nfsd / nfs4layouts.c

/*
 * Copyright (c) 2014 Christoph Hellwig.
 */
#include <linux/kmod.h>
#include <linux/file.h>
#include <linux/jhash.h>
#include <linux/sched.h>
#include <linux/sunrpc/addr.h>

#include "pnfs.h"
#include "netns.h"
#include "trace.h"

#define NFSDDBG_FACILITY                NFSDDBG_PNFS

struct nfs4_layout {
	struct list_head		lo_perstate;
	struct nfs4_layout_stateid	*lo_state;
	struct nfsd4_layout_seg		lo_seg;
};

static struct kmem_cache *nfs4_layout_cache;
static struct kmem_cache *nfs4_layout_stateid_cache;

static const struct nfsd4_callback_ops nfsd4_cb_layout_ops;
static const struct lock_manager_operations nfsd4_layouts_lm_ops;

const struct nfsd4_layout_ops *nfsd4_layout_ops[LAYOUT_TYPE_MAX] =  {
	[LAYOUT_BLOCK_VOLUME]	= &bl_layout_ops,
};

/* pNFS device ID to export fsid mapping */
#define DEVID_HASH_BITS	8
#define DEVID_HASH_SIZE	(1 << DEVID_HASH_BITS)
#define DEVID_HASH_MASK	(DEVID_HASH_SIZE - 1)
static u64 nfsd_devid_seq = 1;
static struct list_head nfsd_devid_hash[DEVID_HASH_SIZE];
static DEFINE_SPINLOCK(nfsd_devid_lock);

static inline u32 devid_hashfn(u64 idx)
{
	return jhash_2words(idx, idx >> 32, 0) & DEVID_HASH_MASK;
}

static void
nfsd4_alloc_devid_map(const struct svc_fh *fhp)
{
	const struct knfsd_fh *fh = &fhp->fh_handle;
	size_t fsid_len = key_len(fh->fh_fsid_type);
	struct nfsd4_deviceid_map *map, *old;
	int i;

	map = kzalloc(sizeof(*map) + fsid_len, GFP_KERNEL);
	if (!map)
		return;

	map->fsid_type = fh->fh_fsid_type;
	memcpy(&map->fsid, fh->fh_fsid, fsid_len);

	spin_lock(&nfsd_devid_lock);
	if (fhp->fh_export->ex_devid_map)
		goto out_unlock;

	for (i = 0; i < DEVID_HASH_SIZE; i++) {
		list_for_each_entry(old, &nfsd_devid_hash[i], hash) {
			if (old->fsid_type != fh->fh_fsid_type)
				continue;
			if (memcmp(old->fsid, fh->fh_fsid,
					key_len(old->fsid_type)))
				continue;

			fhp->fh_export->ex_devid_map = old;
			goto out_unlock;
		}
	}

	map->idx = nfsd_devid_seq++;
	list_add_tail_rcu(&map->hash, &nfsd_devid_hash[devid_hashfn(map->idx)]);
	fhp->fh_export->ex_devid_map = map;
	map = NULL;

out_unlock:
	spin_unlock(&nfsd_devid_lock);
	kfree(map);
}

struct nfsd4_deviceid_map *
nfsd4_find_devid_map(int idx)
{
	struct nfsd4_deviceid_map *map, *ret = NULL;

	rcu_read_lock();
	list_for_each_entry_rcu(map, &nfsd_devid_hash[devid_hashfn(idx)], hash)
		if (map->idx == idx)
			ret = map;
	rcu_read_unlock();

	return ret;
}

int
nfsd4_set_deviceid(struct nfsd4_deviceid *id, const struct svc_fh *fhp,
		u32 device_generation)
{
	if (!fhp->fh_export->ex_devid_map) {
		nfsd4_alloc_devid_map(fhp);
		if (!fhp->fh_export->ex_devid_map)
			return -ENOMEM;
	}

	id->fsid_idx = fhp->fh_export->ex_devid_map->idx;
	id->generation = device_generation;
	id->pad = 0;
	return 0;
}

void nfsd4_setup_layout_type(struct svc_export *exp)
{
	struct super_block *sb = exp->ex_path.mnt->mnt_sb;

	if (!(exp->ex_flags & NFSEXP_PNFS))
		return;

	if (sb->s_export_op->get_uuid &&
	    sb->s_export_op->map_blocks &&
	    sb->s_export_op->commit_blocks)
		exp->ex_layout_type = LAYOUT_BLOCK_VOLUME;
}

static void
nfsd4_free_layout_stateid(struct nfs4_stid *stid)
{
	struct nfs4_layout_stateid *ls = layoutstateid(stid);
	struct nfs4_client *clp = ls->ls_stid.sc_client;
	struct nfs4_file *fp = ls->ls_stid.sc_file;

	trace_layoutstate_free(&ls->ls_stid.sc_stateid);

	spin_lock(&clp->cl_lock);
	list_del_init(&ls->ls_perclnt);
	spin_unlock(&clp->cl_lock);

	spin_lock(&fp->fi_lock);
	list_del_init(&ls->ls_perfile);
	spin_unlock(&fp->fi_lock);

	vfs_setlease(ls->ls_file, F_UNLCK, NULL, (void **)&ls);
	fput(ls->ls_file);

	if (ls->ls_recalled)
		atomic_dec(&ls->ls_stid.sc_file->fi_lo_recalls);

	kmem_cache_free(nfs4_layout_stateid_cache, ls);
}

static int
nfsd4_layout_setlease(struct nfs4_layout_stateid *ls)
{
	struct file_lock *fl;
	int status;

	fl = locks_alloc_lock();
	if (!fl)
		return -ENOMEM;
	locks_init_lock(fl);
	fl->fl_lmops = &nfsd4_layouts_lm_ops;
	fl->fl_flags = FL_LAYOUT;
	fl->fl_type = F_RDLCK;
	fl->fl_end = OFFSET_MAX;
	fl->fl_owner = ls;
	fl->fl_pid = current->tgid;
	fl->fl_file = ls->ls_file;

	status = vfs_setlease(fl->fl_file, fl->fl_type, &fl, NULL);
	if (status) {
		locks_free_lock(fl);
		return status;
	}
	BUG_ON(fl != NULL);
	return 0;
}

static struct nfs4_layout_stateid *
nfsd4_alloc_layout_stateid(struct nfsd4_compound_state *cstate,
		struct nfs4_stid *parent, u32 layout_type)
{
	struct nfs4_client *clp = cstate->clp;
	struct nfs4_file *fp = parent->sc_file;
	struct nfs4_layout_stateid *ls;
	struct nfs4_stid *stp;

	stp = nfs4_alloc_stid(cstate->clp, nfs4_layout_stateid_cache);
	if (!stp)
		return NULL;
	stp->sc_free = nfsd4_free_layout_stateid;
	get_nfs4_file(fp);
	stp->sc_file = fp;

	ls = layoutstateid(stp);
	INIT_LIST_HEAD(&ls->ls_perclnt);
	INIT_LIST_HEAD(&ls->ls_perfile);
	spin_lock_init(&ls->ls_lock);
	INIT_LIST_HEAD(&ls->ls_layouts);
	mutex_init(&ls->ls_mutex);
	ls->ls_layout_type = layout_type;
	nfsd4_init_cb(&ls->ls_recall, clp, &nfsd4_cb_layout_ops,
			NFSPROC4_CLNT_CB_LAYOUT);

	if (parent->sc_type == NFS4_DELEG_STID)
		ls->ls_file = get_file(fp->fi_deleg_file);
	else
		ls->ls_file = find_any_file(fp);
	BUG_ON(!ls->ls_file);

	if (nfsd4_layout_setlease(ls)) {
		fput(ls->ls_file);
		put_nfs4_file(fp);
		kmem_cache_free(nfs4_layout_stateid_cache, ls);
		return NULL;
	}

	spin_lock(&clp->cl_lock);
	stp->sc_type = NFS4_LAYOUT_STID;
	list_add(&ls->ls_perclnt, &clp->cl_lo_states);
	spin_unlock(&clp->cl_lock);

	spin_lock(&fp->fi_lock);
	list_add(&ls->ls_perfile, &fp->fi_lo_states);
	spin_unlock(&fp->fi_lock);

	trace_layoutstate_alloc(&ls->ls_stid.sc_stateid);
	return ls;
}

__be32
nfsd4_preprocess_layout_stateid(struct svc_rqst *rqstp,
		struct nfsd4_compound_state *cstate, stateid_t *stateid,
		bool create, u32 layout_type, struct nfs4_layout_stateid **lsp)
{
	struct nfs4_layout_stateid *ls;
	struct nfs4_stid *stid;
	unsigned char typemask = NFS4_LAYOUT_STID;
	__be32 status;

	if (create)
		typemask |= (NFS4_OPEN_STID | NFS4_LOCK_STID | NFS4_DELEG_STID);

	status = nfsd4_lookup_stateid(cstate, stateid, typemask, &stid,
			net_generic(SVC_NET(rqstp), nfsd_net_id));
	if (status)
		goto out;

	if (!fh_match(&cstate->current_fh.fh_handle,
		      &stid->sc_file->fi_fhandle)) {
		status = nfserr_bad_stateid;
		goto out_put_stid;
	}

	if (stid->sc_type != NFS4_LAYOUT_STID) {
		ls = nfsd4_alloc_layout_stateid(cstate, stid, layout_type);
		nfs4_put_stid(stid);

		status = nfserr_jukebox;
		if (!ls)
			goto out;
		mutex_lock(&ls->ls_mutex);
	} else {
		ls = container_of(stid, struct nfs4_layout_stateid, ls_stid);

		status = nfserr_bad_stateid;
		mutex_lock(&ls->ls_mutex);
		if (stateid->si_generation > stid->sc_stateid.si_generation)
			goto out_unlock_stid;
		if (layout_type != ls->ls_layout_type)
			goto out_unlock_stid;
	}

	*lsp = ls;
	return 0;

out_unlock_stid:
	mutex_unlock(&ls->ls_mutex);
out_put_stid:
	nfs4_put_stid(stid);
out:
	return status;
}

static void
nfsd4_recall_file_layout(struct nfs4_layout_stateid *ls)
{
	spin_lock(&ls->ls_lock);
	if (ls->ls_recalled)
		goto out_unlock;

	ls->ls_recalled = true;
	atomic_inc(&ls->ls_stid.sc_file->fi_lo_recalls);
	if (list_empty(&ls->ls_layouts))
		goto out_unlock;

	trace_layout_recall(&ls->ls_stid.sc_stateid);

	atomic_inc(&ls->ls_stid.sc_count);
	nfsd4_run_cb(&ls->ls_recall);

out_unlock:
	spin_unlock(&ls->ls_lock);
}

static inline u64
layout_end(struct nfsd4_layout_seg *seg)
{
	u64 end = seg->offset + seg->length;
	return end >= seg->offset ? end : NFS4_MAX_UINT64;
}

static void
layout_update_len(struct nfsd4_layout_seg *lo, u64 end)
{
	if (end == NFS4_MAX_UINT64)
		lo->length = NFS4_MAX_UINT64;
	else
		lo->length = end - lo->offset;
}

static bool
layouts_overlapping(struct nfs4_layout *lo, struct nfsd4_layout_seg *s)
{
	if (s->iomode != IOMODE_ANY && s->iomode != lo->lo_seg.iomode)
		return false;
	if (layout_end(&lo->lo_seg) <= s->offset)
		return false;
	if (layout_end(s) <= lo->lo_seg.offset)
		return false;
	return true;
}

static bool
layouts_try_merge(struct nfsd4_layout_seg *lo, struct nfsd4_layout_seg *new)
{
	if (lo->iomode != new->iomode)
		return false;
	if (layout_end(new) < lo->offset)
		return false;
	if (layout_end(lo) < new->offset)
		return false;

	lo->offset = min(lo->offset, new->offset);
	layout_update_len(lo, max(layout_end(lo), layout_end(new)));
	return true;
}

static __be32
nfsd4_recall_conflict(struct nfs4_layout_stateid *ls)
{
	struct nfs4_file *fp = ls->ls_stid.sc_file;
	struct nfs4_layout_stateid *l, *n;
	__be32 nfserr = nfs_ok;

	assert_spin_locked(&fp->fi_lock);

	list_for_each_entry_safe(l, n, &fp->fi_lo_states, ls_perfile) {
		if (l != ls) {
			nfsd4_recall_file_layout(l);
			nfserr = nfserr_recallconflict;
		}
	}

	return nfserr;
}

__be32
nfsd4_insert_layout(struct nfsd4_layoutget *lgp, struct nfs4_layout_stateid *ls)
{
	struct nfsd4_layout_seg *seg = &lgp->lg_seg;
	struct nfs4_file *fp = ls->ls_stid.sc_file;
	struct nfs4_layout *lp, *new = NULL;
	__be32 nfserr;

	spin_lock(&fp->fi_lock);
	nfserr = nfsd4_recall_conflict(ls);
	if (nfserr)
		goto out;
	spin_lock(&ls->ls_lock);
	list_for_each_entry(lp, &ls->ls_layouts, lo_perstate) {
		if (layouts_try_merge(&lp->lo_seg, seg))
			goto done;
	}
	spin_unlock(&ls->ls_lock);
	spin_unlock(&fp->fi_lock);

	new = kmem_cache_alloc(nfs4_layout_cache, GFP_KERNEL);
	if (!new)
		return nfserr_jukebox;
	memcpy(&new->lo_seg, seg, sizeof(lp->lo_seg));
	new->lo_state = ls;

	spin_lock(&fp->fi_lock);
	nfserr = nfsd4_recall_conflict(ls);
	if (nfserr)
		goto out;
	spin_lock(&ls->ls_lock);
	list_for_each_entry(lp, &ls->ls_layouts, lo_perstate) {
		if (layouts_try_merge(&lp->lo_seg, seg))
			goto done;
	}

	atomic_inc(&ls->ls_stid.sc_count);
	list_add_tail(&new->lo_perstate, &ls->ls_layouts);
	new = NULL;
done:
	nfs4_inc_and_copy_stateid(&lgp->lg_sid, &ls->ls_stid);
	spin_unlock(&ls->ls_lock);
out:
	spin_unlock(&fp->fi_lock);
	if (new)
		kmem_cache_free(nfs4_layout_cache, new);
	return nfserr;
}

static void
nfsd4_free_layouts(struct list_head *reaplist)
{
	while (!list_empty(reaplist)) {
		struct nfs4_layout *lp = list_first_entry(reaplist,
				struct nfs4_layout, lo_perstate);

		list_del(&lp->lo_perstate);
		nfs4_put_stid(&lp->lo_state->ls_stid);
		kmem_cache_free(nfs4_layout_cache, lp);
	}
}

static void
nfsd4_return_file_layout(struct nfs4_layout *lp, struct nfsd4_layout_seg *seg,
		struct list_head *reaplist)
{
	struct nfsd4_layout_seg *lo = &lp->lo_seg;
	u64 end = layout_end(lo);

	if (seg->offset <= lo->offset) {
		if (layout_end(seg) >= end) {
			list_move_tail(&lp->lo_perstate, reaplist);
			return;
		}
		lo->offset = layout_end(seg);
	} else {
		/* retain the whole layout segment on a split. */
		if (layout_end(seg) < end) {
			dprintk("%s: split not supported\n", __func__);
			return;
		}
		end = seg->offset;
	}

	layout_update_len(lo, end);
}

__be32
nfsd4_return_file_layouts(struct svc_rqst *rqstp,
		struct nfsd4_compound_state *cstate,
		struct nfsd4_layoutreturn *lrp)
{
	struct nfs4_layout_stateid *ls;
	struct nfs4_layout *lp, *n;
	LIST_HEAD(reaplist);
	__be32 nfserr;
	int found = 0;

	nfserr = nfsd4_preprocess_layout_stateid(rqstp, cstate, &lrp->lr_sid,
						false, lrp->lr_layout_type,
						&ls);
	if (nfserr) {
		trace_layout_return_lookup_fail(&lrp->lr_sid);
		return nfserr;
	}

	spin_lock(&ls->ls_lock);
	list_for_each_entry_safe(lp, n, &ls->ls_layouts, lo_perstate) {
		if (layouts_overlapping(lp, &lrp->lr_seg)) {
			nfsd4_return_file_layout(lp, &lrp->lr_seg, &reaplist);
			found++;
		}
	}
	if (!list_empty(&ls->ls_layouts)) {
		if (found)
			nfs4_inc_and_copy_stateid(&lrp->lr_sid, &ls->ls_stid);
		lrp->lrs_present = 1;
	} else {
		trace_layoutstate_unhash(&ls->ls_stid.sc_stateid);
		nfs4_unhash_stid(&ls->ls_stid);
		lrp->lrs_present = 0;
	}
	spin_unlock(&ls->ls_lock);

	mutex_unlock(&ls->ls_mutex);
	nfs4_put_stid(&ls->ls_stid);
	nfsd4_free_layouts(&reaplist);
	return nfs_ok;
}

__be32
nfsd4_return_client_layouts(struct svc_rqst *rqstp,
		struct nfsd4_compound_state *cstate,
		struct nfsd4_layoutreturn *lrp)
{
	struct nfs4_layout_stateid *ls, *n;
	struct nfs4_client *clp = cstate->clp;
	struct nfs4_layout *lp, *t;
	LIST_HEAD(reaplist);

	lrp->lrs_present = 0;

	spin_lock(&clp->cl_lock);
	list_for_each_entry_safe(ls, n, &clp->cl_lo_states, ls_perclnt) {
		if (ls->ls_layout_type != lrp->lr_layout_type)
			continue;

		if (lrp->lr_return_type == RETURN_FSID &&
		    !fh_fsid_match(&ls->ls_stid.sc_file->fi_fhandle,
				   &cstate->current_fh.fh_handle))
			continue;

		spin_lock(&ls->ls_lock);
		list_for_each_entry_safe(lp, t, &ls->ls_layouts, lo_perstate) {
			if (lrp->lr_seg.iomode == IOMODE_ANY ||
			    lrp->lr_seg.iomode == lp->lo_seg.iomode)
				list_move_tail(&lp->lo_perstate, &reaplist);
		}
		spin_unlock(&ls->ls_lock);
	}
	spin_unlock(&clp->cl_lock);

	nfsd4_free_layouts(&reaplist);
	return 0;
}

static void
nfsd4_return_all_layouts(struct nfs4_layout_stateid *ls,
		struct list_head *reaplist)
{
	spin_lock(&ls->ls_lock);
	list_splice_init(&ls->ls_layouts, reaplist);
	spin_unlock(&ls->ls_lock);
}

void
nfsd4_return_all_client_layouts(struct nfs4_client *clp)
{
	struct nfs4_layout_stateid *ls, *n;
	LIST_HEAD(reaplist);

	spin_lock(&clp->cl_lock);
	list_for_each_entry_safe(ls, n, &clp->cl_lo_states, ls_perclnt)
		nfsd4_return_all_layouts(ls, &reaplist);
	spin_unlock(&clp->cl_lock);

	nfsd4_free_layouts(&reaplist);
}

void
nfsd4_return_all_file_layouts(struct nfs4_client *clp, struct nfs4_file *fp)
{
	struct nfs4_layout_stateid *ls, *n;
	LIST_HEAD(reaplist);

	spin_lock(&fp->fi_lock);
	list_for_each_entry_safe(ls, n, &fp->fi_lo_states, ls_perfile) {
		if (ls->ls_stid.sc_client == clp)
			nfsd4_return_all_layouts(ls, &reaplist);
	}
	spin_unlock(&fp->fi_lock);

	nfsd4_free_layouts(&reaplist);
}

static void
nfsd4_cb_layout_fail(struct nfs4_layout_stateid *ls)
{
	struct nfs4_client *clp = ls->ls_stid.sc_client;
	char addr_str[INET6_ADDRSTRLEN];
	static char *envp[] = {
		"HOME=/",
		"TERM=linux",
		"PATH=/sbin:/usr/sbin:/bin:/usr/bin",
		NULL
	};
	char *argv[8];
	int error;

	rpc_ntop((struct sockaddr *)&clp->cl_addr, addr_str, sizeof(addr_str));

	trace_layout_recall_fail(&ls->ls_stid.sc_stateid);

	printk(KERN_WARNING
		"nfsd: client %s failed to respond to layout recall. "
		"  Fencing..\n", addr_str);

	argv[0] = "/sbin/nfsd-recall-failed";
	argv[1] = addr_str;
	argv[2] = ls->ls_file->f_path.mnt->mnt_sb->s_id;
	argv[3] = NULL;

	error = call_usermodehelper(argv[0], argv, envp, UMH_WAIT_PROC);
	if (error) {
		printk(KERN_ERR "nfsd: fence failed for client %s: %d!\n",
			addr_str, error);
	}
}

static void
nfsd4_cb_layout_prepare(struct nfsd4_callback *cb)
{
	struct nfs4_layout_stateid *ls =
		container_of(cb, struct nfs4_layout_stateid, ls_recall);

	mutex_lock(&ls->ls_mutex);
	nfs4_inc_and_copy_stateid(&ls->ls_recall_sid, &ls->ls_stid);
	mutex_unlock(&ls->ls_mutex);
}

static int
nfsd4_cb_layout_done(struct nfsd4_callback *cb, struct rpc_task *task)
{
	struct nfs4_layout_stateid *ls =
		container_of(cb, struct nfs4_layout_stateid, ls_recall);
	struct nfsd_net *nn;
	ktime_t now, cutoff;
	LIST_HEAD(reaplist);


	switch (task->tk_status) {
	case 0:
	case -NFS4ERR_DELAY:
		/*
		 * Anything left? If not, then call it done. Note that we don't
		 * take the spinlock since this is an optimization and nothing
		 * should get added until the cb counter goes to zero.
		 */
		if (list_empty(&ls->ls_layouts))
			return 1;

		/* Poll the client until it's done with the layout */
		now = ktime_get();
		nn = net_generic(ls->ls_stid.sc_client->net, nfsd_net_id);

		/* Client gets 2 lease periods to return it */
		cutoff = ktime_add_ns(task->tk_start,
					 nn->nfsd4_lease * NSEC_PER_SEC * 2);

		if (ktime_before(now, cutoff)) {
			rpc_delay(task, HZ/100); /* 10 mili-seconds */
			return 0;
		}
		/* Fallthrough */
	case -NFS4ERR_NOMATCHING_LAYOUT:
		trace_layout_recall_done(&ls->ls_stid.sc_stateid);
		task->tk_status = 0;
		return 1;
	default:
		/*
		 * Unknown error or non-responding client, we'll need to fence.
		 */
		nfsd4_cb_layout_fail(ls);
		return -1;
	}
}

static void
nfsd4_cb_layout_release(struct nfsd4_callback *cb)
{
	struct nfs4_layout_stateid *ls =
		container_of(cb, struct nfs4_layout_stateid, ls_recall);
	LIST_HEAD(reaplist);

	trace_layout_recall_release(&ls->ls_stid.sc_stateid);

	nfsd4_return_all_layouts(ls, &reaplist);
	nfsd4_free_layouts(&reaplist);
	nfs4_put_stid(&ls->ls_stid);
}

static const struct nfsd4_callback_ops nfsd4_cb_layout_ops = {
	.prepare	= nfsd4_cb_layout_prepare,
	.done		= nfsd4_cb_layout_done,
	.release	= nfsd4_cb_layout_release,
};

static bool
nfsd4_layout_lm_break(struct file_lock *fl)
{
	/*
	 * We don't want the locks code to timeout the lease for us;
	 * we'll remove it ourself if a layout isn't returned
	 * in time:
	 */
	fl->fl_break_time = 0;
	nfsd4_recall_file_layout(fl->fl_owner);
	return false;
}

static int
nfsd4_layout_lm_change(struct file_lock *onlist, int arg,
		struct list_head *dispose)
{
	BUG_ON(!(arg & F_UNLCK));
	return lease_modify(onlist, arg, dispose);
}

static const struct lock_manager_operations nfsd4_layouts_lm_ops = {
	.lm_break	= nfsd4_layout_lm_break,
	.lm_change	= nfsd4_layout_lm_change,
};

int
nfsd4_init_pnfs(void)
{
	int i;

	for (i = 0; i < DEVID_HASH_SIZE; i++)
		INIT_LIST_HEAD(&nfsd_devid_hash[i]);

	nfs4_layout_cache = kmem_cache_create("nfs4_layout",
			sizeof(struct nfs4_layout), 0, 0, NULL);
	if (!nfs4_layout_cache)
		return -ENOMEM;

	nfs4_layout_stateid_cache = kmem_cache_create("nfs4_layout_stateid",
			sizeof(struct nfs4_layout_stateid), 0, 0, NULL);
	if (!nfs4_layout_stateid_cache) {
		kmem_cache_destroy(nfs4_layout_cache);
		return -ENOMEM;
	}
	return 0;
}

void
nfsd4_exit_pnfs(void)
{
	int i;

	kmem_cache_destroy(nfs4_layout_cache);
	kmem_cache_destroy(nfs4_layout_stateid_cache);

	for (i = 0; i < DEVID_HASH_SIZE; i++) {
		struct nfsd4_deviceid_map *map, *n;

		list_for_each_entry_safe(map, n, &nfsd_devid_hash[i], hash)
			kfree(map);
	}
}
Commit	Line	Data
9cf514cc CH	1	/*
	2	* Copyright (c) 2014 Christoph Hellwig.
	3	*/
c5c707f9 CH	4	#include <linux/kmod.h>
c5c707f9 CH	5	#include <linux/file.h>
9cf514cc CH	6	#include <linux/jhash.h>
9cf514cc CH	7	#include <linux/sched.h>
c5c707f9	8	#include <linux/sunrpc/addr.h>
9cf514cc CH	9
	10	#include "pnfs.h"
	11	#include "netns.h"
31ef83dc	12	#include "trace.h"
9cf514cc CH	13
	14	#define NFSDDBG_FACILITY NFSDDBG_PNFS
	15
	16	struct nfs4_layout {
	17	struct list_head lo_perstate;
	18	struct nfs4_layout_stateid *lo_state;
	19	struct nfsd4_layout_seg lo_seg;
	20	};
	21
	22	static struct kmem_cache *nfs4_layout_cache;
	23	static struct kmem_cache *nfs4_layout_stateid_cache;
	24
c4cb8974	25	static const struct nfsd4_callback_ops nfsd4_cb_layout_ops;
c5c707f9 CH	26	static const struct lock_manager_operations nfsd4_layouts_lm_ops;
c5c707f9 CH	27
9cf514cc	28	const struct nfsd4_layout_ops *nfsd4_layout_ops[LAYOUT_TYPE_MAX] = {
8650b8a0	29	[LAYOUT_BLOCK_VOLUME] = &bl_layout_ops,
9cf514cc CH	30	};
	31
	32	/* pNFS device ID to export fsid mapping */
	33	#define DEVID_HASH_BITS 8
	34	#define DEVID_HASH_SIZE (1 << DEVID_HASH_BITS)
	35	#define DEVID_HASH_MASK (DEVID_HASH_SIZE - 1)
	36	static u64 nfsd_devid_seq = 1;
	37	static struct list_head nfsd_devid_hash[DEVID_HASH_SIZE];
	38	static DEFINE_SPINLOCK(nfsd_devid_lock);
	39
	40	static inline u32 devid_hashfn(u64 idx)
	41	{
	42	return jhash_2words(idx, idx >> 32, 0) & DEVID_HASH_MASK;
	43	}
	44
	45	static void
	46	nfsd4_alloc_devid_map(const struct svc_fh *fhp)
	47	{
	48	const struct knfsd_fh *fh = &fhp->fh_handle;
	49	size_t fsid_len = key_len(fh->fh_fsid_type);
	50	struct nfsd4_deviceid_map map, old;
	51	int i;
	52
	53	map = kzalloc(sizeof(*map) + fsid_len, GFP_KERNEL);
	54	if (!map)
	55	return;
	56
	57	map->fsid_type = fh->fh_fsid_type;
	58	memcpy(&map->fsid, fh->fh_fsid, fsid_len);
	59
	60	spin_lock(&nfsd_devid_lock);
	61	if (fhp->fh_export->ex_devid_map)
	62	goto out_unlock;
	63
	64	for (i = 0; i < DEVID_HASH_SIZE; i++) {
	65	list_for_each_entry(old, &nfsd_devid_hash[i], hash) {
	66	if (old->fsid_type != fh->fh_fsid_type)
	67	continue;
	68	if (memcmp(old->fsid, fh->fh_fsid,
	69	key_len(old->fsid_type)))
	70	continue;
	71
	72	fhp->fh_export->ex_devid_map = old;
	73	goto out_unlock;
	74	}
	75	}
	76
	77	map->idx = nfsd_devid_seq++;
	78	list_add_tail_rcu(&map->hash, &nfsd_devid_hash[devid_hashfn(map->idx)]);
	79	fhp->fh_export->ex_devid_map = map;
	80	map = NULL;
	81
	82	out_unlock:
	83	spin_unlock(&nfsd_devid_lock);
	84	kfree(map);
	85	}
	86
	87	struct nfsd4_deviceid_map *
	88	nfsd4_find_devid_map(int idx)
	89	{
	90	struct nfsd4_deviceid_map map, ret = NULL;
	91
	92	rcu_read_lock();
	93	list_for_each_entry_rcu(map, &nfsd_devid_hash[devid_hashfn(idx)], hash)
94	if (map->idx == idx)
95	ret = map;
96	rcu_read_unlock();
97
98	return ret;
99	}
100
101	int
102	nfsd4_set_deviceid(struct nfsd4_deviceid id, const struct svc_fh fhp,
103	u32 device_generation)
104	{
105	if (!fhp->fh_export->ex_devid_map) {
106	nfsd4_alloc_devid_map(fhp);
107	if (!fhp->fh_export->ex_devid_map)
108	return -ENOMEM;
109	}
110
111	id->fsid_idx = fhp->fh_export->ex_devid_map->idx;
112	id->generation = device_generation;
113	id->pad = 0;
114	return 0;
115	}
116
117	void nfsd4_setup_layout_type(struct svc_export *exp)
118	{
8650b8a0 CH	119	struct super_block *sb = exp->ex_path.mnt->mnt_sb;
8650b8a0 CH	120
f3f03330	121	if (!(exp->ex_flags & NFSEXP_PNFS))
9cf514cc	122	return;
8650b8a0 CH	123
	124	if (sb->s_export_op->get_uuid &&
	125	sb->s_export_op->map_blocks &&
	126	sb->s_export_op->commit_blocks)
	127	exp->ex_layout_type = LAYOUT_BLOCK_VOLUME;
9cf514cc CH	128	}
	129
	130	static void
	131	nfsd4_free_layout_stateid(struct nfs4_stid *stid)
	132	{
	133	struct nfs4_layout_stateid *ls = layoutstateid(stid);
	134	struct nfs4_client *clp = ls->ls_stid.sc_client;
	135	struct nfs4_file *fp = ls->ls_stid.sc_file;
	136
31ef83dc CH	137	trace_layoutstate_free(&ls->ls_stid.sc_stateid);
31ef83dc CH	138
9cf514cc CH	139	spin_lock(&clp->cl_lock);
	140	list_del_init(&ls->ls_perclnt);
	141	spin_unlock(&clp->cl_lock);
	142
	143	spin_lock(&fp->fi_lock);
	144	list_del_init(&ls->ls_perfile);
	145	spin_unlock(&fp->fi_lock);
	146
c5c707f9 CH	147	vfs_setlease(ls->ls_file, F_UNLCK, NULL, (void **)&ls);
	148	fput(ls->ls_file);
	149
	150	if (ls->ls_recalled)
	151	atomic_dec(&ls->ls_stid.sc_file->fi_lo_recalls);
	152
9cf514cc CH	153	kmem_cache_free(nfs4_layout_stateid_cache, ls);
	154	}
	155
c5c707f9 CH	156	static int
	157	nfsd4_layout_setlease(struct nfs4_layout_stateid *ls)
	158	{
	159	struct file_lock *fl;
	160	int status;
	161
	162	fl = locks_alloc_lock();
	163	if (!fl)
	164	return -ENOMEM;
	165	locks_init_lock(fl);
	166	fl->fl_lmops = &nfsd4_layouts_lm_ops;
	167	fl->fl_flags = FL_LAYOUT;
	168	fl->fl_type = F_RDLCK;
	169	fl->fl_end = OFFSET_MAX;
	170	fl->fl_owner = ls;
	171	fl->fl_pid = current->tgid;
	172	fl->fl_file = ls->ls_file;
	173
	174	status = vfs_setlease(fl->fl_file, fl->fl_type, &fl, NULL);
	175	if (status) {
	176	locks_free_lock(fl);
	177	return status;
	178	}
	179	BUG_ON(fl != NULL);
	180	return 0;
	181	}
	182
9cf514cc CH	183	static struct nfs4_layout_stateid *
	184	nfsd4_alloc_layout_stateid(struct nfsd4_compound_state *cstate,
	185	struct nfs4_stid *parent, u32 layout_type)
	186	{
	187	struct nfs4_client *clp = cstate->clp;
	188	struct nfs4_file *fp = parent->sc_file;
	189	struct nfs4_layout_stateid *ls;
	190	struct nfs4_stid *stp;
	191
	192	stp = nfs4_alloc_stid(cstate->clp, nfs4_layout_stateid_cache);
	193	if (!stp)
	194	return NULL;
	195	stp->sc_free = nfsd4_free_layout_stateid;
	196	get_nfs4_file(fp);
	197	stp->sc_file = fp;
	198
	199	ls = layoutstateid(stp);
	200	INIT_LIST_HEAD(&ls->ls_perclnt);
	201	INIT_LIST_HEAD(&ls->ls_perfile);
	202	spin_lock_init(&ls->ls_lock);
	203	INIT_LIST_HEAD(&ls->ls_layouts);
cc8a5532	204	mutex_init(&ls->ls_mutex);
9cf514cc	205	ls->ls_layout_type = layout_type;
c5c707f9 CH	206	nfsd4_init_cb(&ls->ls_recall, clp, &nfsd4_cb_layout_ops,
	207	NFSPROC4_CLNT_CB_LAYOUT);
	208
	209	if (parent->sc_type == NFS4_DELEG_STID)
	210	ls->ls_file = get_file(fp->fi_deleg_file);
	211	else
	212	ls->ls_file = find_any_file(fp);
	213	BUG_ON(!ls->ls_file);
	214
	215	if (nfsd4_layout_setlease(ls)) {
1ca4b88e	216	fput(ls->ls_file);
c5c707f9 CH	217	put_nfs4_file(fp);
	218	kmem_cache_free(nfs4_layout_stateid_cache, ls);
	219	return NULL;
	220	}
9cf514cc CH	221
	222	spin_lock(&clp->cl_lock);
	223	stp->sc_type = NFS4_LAYOUT_STID;
	224	list_add(&ls->ls_perclnt, &clp->cl_lo_states);
	225	spin_unlock(&clp->cl_lock);
	226
	227	spin_lock(&fp->fi_lock);
	228	list_add(&ls->ls_perfile, &fp->fi_lo_states);
	229	spin_unlock(&fp->fi_lock);
	230
31ef83dc	231	trace_layoutstate_alloc(&ls->ls_stid.sc_stateid);
9cf514cc CH	232	return ls;
	233	}
	234
	235	__be32
	236	nfsd4_preprocess_layout_stateid(struct svc_rqst *rqstp,
	237	struct nfsd4_compound_state cstate, stateid_t stateid,
	238	bool create, u32 layout_type, struct nfs4_layout_stateid **lsp)
	239	{
	240	struct nfs4_layout_stateid *ls;
	241	struct nfs4_stid *stid;
	242	unsigned char typemask = NFS4_LAYOUT_STID;
	243	__be32 status;
	244
	245	if (create)
	246	typemask \|= (NFS4_OPEN_STID \| NFS4_LOCK_STID \| NFS4_DELEG_STID);
	247
	248	status = nfsd4_lookup_stateid(cstate, stateid, typemask, &stid,
	249	net_generic(SVC_NET(rqstp), nfsd_net_id));
	250	if (status)
	251	goto out;
	252
	253	if (!fh_match(&cstate->current_fh.fh_handle,
	254	&stid->sc_file->fi_fhandle)) {
	255	status = nfserr_bad_stateid;
	256	goto out_put_stid;
	257	}
	258
	259	if (stid->sc_type != NFS4_LAYOUT_STID) {
	260	ls = nfsd4_alloc_layout_stateid(cstate, stid, layout_type);
	261	nfs4_put_stid(stid);
	262
	263	status = nfserr_jukebox;
	264	if (!ls)
	265	goto out;
cc8a5532	266	mutex_lock(&ls->ls_mutex);
9cf514cc CH	267	} else {
	268	ls = container_of(stid, struct nfs4_layout_stateid, ls_stid);
	269
	270	status = nfserr_bad_stateid;
cc8a5532	271	mutex_lock(&ls->ls_mutex);
9cf514cc	272	if (stateid->si_generation > stid->sc_stateid.si_generation)
cc8a5532	273	goto out_unlock_stid;
9cf514cc	274	if (layout_type != ls->ls_layout_type)
cc8a5532	275	goto out_unlock_stid;
9cf514cc CH	276	}
	277
	278	*lsp = ls;
	279	return 0;
	280
cc8a5532 JL	281	out_unlock_stid:
cc8a5532 JL	282	mutex_unlock(&ls->ls_mutex);
9cf514cc CH	283	out_put_stid:
	284	nfs4_put_stid(stid);
	285	out:
	286	return status;
	287	}
	288
c5c707f9 CH	289	static void
	290	nfsd4_recall_file_layout(struct nfs4_layout_stateid *ls)
	291	{
	292	spin_lock(&ls->ls_lock);
	293	if (ls->ls_recalled)
	294	goto out_unlock;
	295
	296	ls->ls_recalled = true;
	297	atomic_inc(&ls->ls_stid.sc_file->fi_lo_recalls);
	298	if (list_empty(&ls->ls_layouts))
	299	goto out_unlock;
	300
31ef83dc CH	301	trace_layout_recall(&ls->ls_stid.sc_stateid);
31ef83dc CH	302
c5c707f9	303	atomic_inc(&ls->ls_stid.sc_count);
c5c707f9 CH	304	nfsd4_run_cb(&ls->ls_recall);
	305
	306	out_unlock:
	307	spin_unlock(&ls->ls_lock);
	308	}
	309
9cf514cc CH	310	static inline u64
	311	layout_end(struct nfsd4_layout_seg *seg)
	312	{
	313	u64 end = seg->offset + seg->length;
	314	return end >= seg->offset ? end : NFS4_MAX_UINT64;
	315	}
	316
	317	static void
	318	layout_update_len(struct nfsd4_layout_seg *lo, u64 end)
	319	{
	320	if (end == NFS4_MAX_UINT64)
	321	lo->length = NFS4_MAX_UINT64;
	322	else
	323	lo->length = end - lo->offset;
	324	}
	325
	326	static bool
	327	layouts_overlapping(struct nfs4_layout lo, struct nfsd4_layout_seg s)
	328	{
	329	if (s->iomode != IOMODE_ANY && s->iomode != lo->lo_seg.iomode)
	330	return false;
	331	if (layout_end(&lo->lo_seg) <= s->offset)
	332	return false;
	333	if (layout_end(s) <= lo->lo_seg.offset)
	334	return false;
	335	return true;
	336	}
	337
	338	static bool
	339	layouts_try_merge(struct nfsd4_layout_seg lo, struct nfsd4_layout_seg new)
	340	{
	341	if (lo->iomode != new->iomode)
	342	return false;
	343	if (layout_end(new) < lo->offset)
	344	return false;
	345	if (layout_end(lo) < new->offset)
	346	return false;
	347
	348	lo->offset = min(lo->offset, new->offset);
	349	layout_update_len(lo, max(layout_end(lo), layout_end(new)));
	350	return true;
	351	}
	352
c5c707f9 CH	353	static __be32
	354	nfsd4_recall_conflict(struct nfs4_layout_stateid *ls)
	355	{
	356	struct nfs4_file *fp = ls->ls_stid.sc_file;
	357	struct nfs4_layout_stateid l, n;
	358	__be32 nfserr = nfs_ok;
	359
	360	assert_spin_locked(&fp->fi_lock);
	361
	362	list_for_each_entry_safe(l, n, &fp->fi_lo_states, ls_perfile) {
	363	if (l != ls) {
	364	nfsd4_recall_file_layout(l);
	365	nfserr = nfserr_recallconflict;
	366	}
	367	}
	368
	369	return nfserr;
	370	}
	371
9cf514cc CH	372	__be32
	373	nfsd4_insert_layout(struct nfsd4_layoutget lgp, struct nfs4_layout_stateid ls)
	374	{
	375	struct nfsd4_layout_seg *seg = &lgp->lg_seg;
c5c707f9	376	struct nfs4_file *fp = ls->ls_stid.sc_file;
9cf514cc	377	struct nfs4_layout lp, new = NULL;
c5c707f9	378	__be32 nfserr;
9cf514cc	379
c5c707f9 CH	380	spin_lock(&fp->fi_lock);
	381	nfserr = nfsd4_recall_conflict(ls);
	382	if (nfserr)
	383	goto out;
9cf514cc CH	384	spin_lock(&ls->ls_lock);
	385	list_for_each_entry(lp, &ls->ls_layouts, lo_perstate) {
	386	if (layouts_try_merge(&lp->lo_seg, seg))
	387	goto done;
	388	}
	389	spin_unlock(&ls->ls_lock);
c5c707f9	390	spin_unlock(&fp->fi_lock);
9cf514cc CH	391
	392	new = kmem_cache_alloc(nfs4_layout_cache, GFP_KERNEL);
	393	if (!new)
	394	return nfserr_jukebox;
	395	memcpy(&new->lo_seg, seg, sizeof(lp->lo_seg));
	396	new->lo_state = ls;
	397
c5c707f9 CH	398	spin_lock(&fp->fi_lock);
	399	nfserr = nfsd4_recall_conflict(ls);
	400	if (nfserr)
	401	goto out;
9cf514cc CH	402	spin_lock(&ls->ls_lock);
	403	list_for_each_entry(lp, &ls->ls_layouts, lo_perstate) {
	404	if (layouts_try_merge(&lp->lo_seg, seg))
	405	goto done;
	406	}
	407
	408	atomic_inc(&ls->ls_stid.sc_count);
	409	list_add_tail(&new->lo_perstate, &ls->ls_layouts);
	410	new = NULL;
	411	done:
9767feb2	412	nfs4_inc_and_copy_stateid(&lgp->lg_sid, &ls->ls_stid);
9cf514cc	413	spin_unlock(&ls->ls_lock);
c5c707f9 CH	414	out:
c5c707f9 CH	415	spin_unlock(&fp->fi_lock);
9cf514cc CH	416	if (new)
9cf514cc CH	417	kmem_cache_free(nfs4_layout_cache, new);
c5c707f9	418	return nfserr;
9cf514cc CH	419	}
	420
	421	static void
	422	nfsd4_free_layouts(struct list_head *reaplist)
	423	{
	424	while (!list_empty(reaplist)) {
	425	struct nfs4_layout *lp = list_first_entry(reaplist,
	426	struct nfs4_layout, lo_perstate);
	427
	428	list_del(&lp->lo_perstate);
	429	nfs4_put_stid(&lp->lo_state->ls_stid);
	430	kmem_cache_free(nfs4_layout_cache, lp);
	431	}
	432	}
	433
	434	static void
	435	nfsd4_return_file_layout(struct nfs4_layout lp, struct nfsd4_layout_seg seg,
	436	struct list_head *reaplist)
	437	{
	438	struct nfsd4_layout_seg *lo = &lp->lo_seg;
	439	u64 end = layout_end(lo);
	440
	441	if (seg->offset <= lo->offset) {
	442	if (layout_end(seg) >= end) {
	443	list_move_tail(&lp->lo_perstate, reaplist);
	444	return;
	445	}
7890203d	446	lo->offset = layout_end(seg);
9cf514cc CH	447	} else {
	448	/* retain the whole layout segment on a split. */
	449	if (layout_end(seg) < end) {
	450	dprintk("%s: split not supported\n", __func__);
	451	return;
	452	}
7890203d	453	end = seg->offset;
9cf514cc CH	454	}
	455
	456	layout_update_len(lo, end);
	457	}
	458
	459	__be32
	460	nfsd4_return_file_layouts(struct svc_rqst *rqstp,
	461	struct nfsd4_compound_state *cstate,
	462	struct nfsd4_layoutreturn *lrp)
	463	{
	464	struct nfs4_layout_stateid *ls;
	465	struct nfs4_layout lp, n;
	466	LIST_HEAD(reaplist);
	467	__be32 nfserr;
	468	int found = 0;
	469
	470	nfserr = nfsd4_preprocess_layout_stateid(rqstp, cstate, &lrp->lr_sid,
	471	false, lrp->lr_layout_type,
	472	&ls);
31ef83dc CH	473	if (nfserr) {
31ef83dc CH	474	trace_layout_return_lookup_fail(&lrp->lr_sid);
9cf514cc	475	return nfserr;
31ef83dc	476	}
9cf514cc CH	477
	478	spin_lock(&ls->ls_lock);
	479	list_for_each_entry_safe(lp, n, &ls->ls_layouts, lo_perstate) {
	480	if (layouts_overlapping(lp, &lrp->lr_seg)) {
	481	nfsd4_return_file_layout(lp, &lrp->lr_seg, &reaplist);
	482	found++;
	483	}
	484	}
	485	if (!list_empty(&ls->ls_layouts)) {
9767feb2 JL	486	if (found)
9767feb2 JL	487	nfs4_inc_and_copy_stateid(&lrp->lr_sid, &ls->ls_stid);
9cf514cc CH	488	lrp->lrs_present = 1;
9cf514cc CH	489	} else {
31ef83dc	490	trace_layoutstate_unhash(&ls->ls_stid.sc_stateid);
9cf514cc CH	491	nfs4_unhash_stid(&ls->ls_stid);
	492	lrp->lrs_present = 0;
	493	}
	494	spin_unlock(&ls->ls_lock);
	495
cc8a5532	496	mutex_unlock(&ls->ls_mutex);
9cf514cc CH	497	nfs4_put_stid(&ls->ls_stid);
	498	nfsd4_free_layouts(&reaplist);
	499	return nfs_ok;
	500	}
	501
	502	__be32
	503	nfsd4_return_client_layouts(struct svc_rqst *rqstp,
	504	struct nfsd4_compound_state *cstate,
	505	struct nfsd4_layoutreturn *lrp)
	506	{
	507	struct nfs4_layout_stateid ls, n;
	508	struct nfs4_client *clp = cstate->clp;
	509	struct nfs4_layout lp, t;
	510	LIST_HEAD(reaplist);
	511
	512	lrp->lrs_present = 0;
	513
	514	spin_lock(&clp->cl_lock);
	515	list_for_each_entry_safe(ls, n, &clp->cl_lo_states, ls_perclnt) {
6f8f28ec KM	516	if (ls->ls_layout_type != lrp->lr_layout_type)
	517	continue;
	518
9cf514cc CH	519	if (lrp->lr_return_type == RETURN_FSID &&
	520	!fh_fsid_match(&ls->ls_stid.sc_file->fi_fhandle,
	521	&cstate->current_fh.fh_handle))
	522	continue;
	523
	524	spin_lock(&ls->ls_lock);
	525	list_for_each_entry_safe(lp, t, &ls->ls_layouts, lo_perstate) {
	526	if (lrp->lr_seg.iomode == IOMODE_ANY \|\|
	527	lrp->lr_seg.iomode == lp->lo_seg.iomode)
	528	list_move_tail(&lp->lo_perstate, &reaplist);
	529	}
	530	spin_unlock(&ls->ls_lock);
	531	}
	532	spin_unlock(&clp->cl_lock);
	533
	534	nfsd4_free_layouts(&reaplist);
	535	return 0;
	536	}
	537
	538	static void
	539	nfsd4_return_all_layouts(struct nfs4_layout_stateid *ls,
	540	struct list_head *reaplist)
	541	{
	542	spin_lock(&ls->ls_lock);
	543	list_splice_init(&ls->ls_layouts, reaplist);
	544	spin_unlock(&ls->ls_lock);
	545	}
	546
	547	void
	548	nfsd4_return_all_client_layouts(struct nfs4_client *clp)
	549	{
	550	struct nfs4_layout_stateid ls, n;
	551	LIST_HEAD(reaplist);
	552
	553	spin_lock(&clp->cl_lock);
	554	list_for_each_entry_safe(ls, n, &clp->cl_lo_states, ls_perclnt)
	555	nfsd4_return_all_layouts(ls, &reaplist);
	556	spin_unlock(&clp->cl_lock);
	557
	558	nfsd4_free_layouts(&reaplist);
	559	}
	560
	561	void
	562	nfsd4_return_all_file_layouts(struct nfs4_client clp, struct nfs4_file fp)
	563	{
	564	struct nfs4_layout_stateid ls, n;
	565	LIST_HEAD(reaplist);
	566
	567	spin_lock(&fp->fi_lock);
	568	list_for_each_entry_safe(ls, n, &fp->fi_lo_states, ls_perfile) {
	569	if (ls->ls_stid.sc_client == clp)
	570	nfsd4_return_all_layouts(ls, &reaplist);
	571	}
	572	spin_unlock(&fp->fi_lock);
	573
	574	nfsd4_free_layouts(&reaplist);
	575	}
	576
c5c707f9 CH	577	static void
	578	nfsd4_cb_layout_fail(struct nfs4_layout_stateid *ls)
	579	{
	580	struct nfs4_client *clp = ls->ls_stid.sc_client;
	581	char addr_str[INET6_ADDRSTRLEN];
	582	static char *envp[] = {
	583	"HOME=/",
	584	"TERM=linux",
	585	"PATH=/sbin:/usr/sbin:/bin:/usr/bin",
	586	NULL
	587	};
	588	char *argv[8];
	589	int error;
	590
	591	rpc_ntop((struct sockaddr *)&clp->cl_addr, addr_str, sizeof(addr_str));
	592
715a03d2 KM	593	trace_layout_recall_fail(&ls->ls_stid.sc_stateid);
715a03d2 KM	594
c5c707f9 CH	595	printk(KERN_WARNING
	596	"nfsd: client %s failed to respond to layout recall. "
	597	" Fencing..\n", addr_str);
	598
	599	argv[0] = "/sbin/nfsd-recall-failed";
	600	argv[1] = addr_str;
	601	argv[2] = ls->ls_file->f_path.mnt->mnt_sb->s_id;
	602	argv[3] = NULL;
	603
	604	error = call_usermodehelper(argv[0], argv, envp, UMH_WAIT_PROC);
	605	if (error) {
	606	printk(KERN_ERR "nfsd: fence failed for client %s: %d!\n",
	607	addr_str, error);
	608	}
	609	}
	610
cc8a5532 JL	611	static void
	612	nfsd4_cb_layout_prepare(struct nfsd4_callback *cb)
	613	{
	614	struct nfs4_layout_stateid *ls =
	615	container_of(cb, struct nfs4_layout_stateid, ls_recall);
	616
	617	mutex_lock(&ls->ls_mutex);
9767feb2	618	nfs4_inc_and_copy_stateid(&ls->ls_recall_sid, &ls->ls_stid);
be20aa00	619	mutex_unlock(&ls->ls_mutex);
cc8a5532 JL	620	}
cc8a5532 JL	621
c5c707f9 CH	622	static int
	623	nfsd4_cb_layout_done(struct nfsd4_callback cb, struct rpc_task task)
	624	{
	625	struct nfs4_layout_stateid *ls =
	626	container_of(cb, struct nfs4_layout_stateid, ls_recall);
6b9b2107 JL	627	struct nfsd_net *nn;
6b9b2107 JL	628	ktime_t now, cutoff;
c5c707f9 CH	629	LIST_HEAD(reaplist);
c5c707f9 CH	630
6b9b2107	631
c5c707f9 CH	632	switch (task->tk_status) {
c5c707f9 CH	633	case 0:
6b9b2107 JL	634	case -NFS4ERR_DELAY:
	635	/*
	636	* Anything left? If not, then call it done. Note that we don't
	637	* take the spinlock since this is an optimization and nothing
	638	* should get added until the cb counter goes to zero.
	639	*/
	640	if (list_empty(&ls->ls_layouts))
	641	return 1;
	642
	643	/* Poll the client until it's done with the layout */
	644	now = ktime_get();
	645	nn = net_generic(ls->ls_stid.sc_client->net, nfsd_net_id);
	646
	647	/* Client gets 2 lease periods to return it */
	648	cutoff = ktime_add_ns(task->tk_start,
	649	nn->nfsd4_lease * NSEC_PER_SEC * 2);
	650
	651	if (ktime_before(now, cutoff)) {
	652	rpc_delay(task, HZ/100); /* 10 mili-seconds */
	653	return 0;
	654	}
	655	/* Fallthrough */
c5c707f9	656	case -NFS4ERR_NOMATCHING_LAYOUT:
31ef83dc	657	trace_layout_recall_done(&ls->ls_stid.sc_stateid);
c5c707f9 CH	658	task->tk_status = 0;
c5c707f9 CH	659	return 1;
c5c707f9 CH	660	default:
	661	/*
	662	* Unknown error or non-responding client, we'll need to fence.
	663	*/
	664	nfsd4_cb_layout_fail(ls);
	665	return -1;
	666	}
	667	}
	668
	669	static void
	670	nfsd4_cb_layout_release(struct nfsd4_callback *cb)
	671	{
	672	struct nfs4_layout_stateid *ls =
	673	container_of(cb, struct nfs4_layout_stateid, ls_recall);
	674	LIST_HEAD(reaplist);
	675
31ef83dc CH	676	trace_layout_recall_release(&ls->ls_stid.sc_stateid);
31ef83dc CH	677
c5c707f9 CH	678	nfsd4_return_all_layouts(ls, &reaplist);
	679	nfsd4_free_layouts(&reaplist);
	680	nfs4_put_stid(&ls->ls_stid);
	681	}
	682
c4cb8974	683	static const struct nfsd4_callback_ops nfsd4_cb_layout_ops = {
cc8a5532	684	.prepare = nfsd4_cb_layout_prepare,
c5c707f9 CH	685	.done = nfsd4_cb_layout_done,
	686	.release = nfsd4_cb_layout_release,
	687	};
	688
	689	static bool
	690	nfsd4_layout_lm_break(struct file_lock *fl)
	691	{
	692	/*
	693	* We don't want the locks code to timeout the lease for us;
	694	* we'll remove it ourself if a layout isn't returned
	695	* in time:
	696	*/
	697	fl->fl_break_time = 0;
	698	nfsd4_recall_file_layout(fl->fl_owner);
	699	return false;
	700	}
	701
	702	static int
	703	nfsd4_layout_lm_change(struct file_lock *onlist, int arg,
	704	struct list_head *dispose)
	705	{
	706	BUG_ON(!(arg & F_UNLCK));
	707	return lease_modify(onlist, arg, dispose);
	708	}
	709
	710	static const struct lock_manager_operations nfsd4_layouts_lm_ops = {
	711	.lm_break = nfsd4_layout_lm_break,
	712	.lm_change = nfsd4_layout_lm_change,
	713	};
	714
9cf514cc CH	715	int
	716	nfsd4_init_pnfs(void)
	717	{
	718	int i;
	719
	720	for (i = 0; i < DEVID_HASH_SIZE; i++)
	721	INIT_LIST_HEAD(&nfsd_devid_hash[i]);
	722
	723	nfs4_layout_cache = kmem_cache_create("nfs4_layout",
	724	sizeof(struct nfs4_layout), 0, 0, NULL);
	725	if (!nfs4_layout_cache)
	726	return -ENOMEM;
	727
	728	nfs4_layout_stateid_cache = kmem_cache_create("nfs4_layout_stateid",
	729	sizeof(struct nfs4_layout_stateid), 0, 0, NULL);
	730	if (!nfs4_layout_stateid_cache) {
	731	kmem_cache_destroy(nfs4_layout_cache);
	732	return -ENOMEM;
	733	}
	734	return 0;
	735	}
	736
	737	void
	738	nfsd4_exit_pnfs(void)
	739	{
	740	int i;
	741
	742	kmem_cache_destroy(nfs4_layout_cache);
	743	kmem_cache_destroy(nfs4_layout_stateid_cache);
	744
	745	for (i = 0; i < DEVID_HASH_SIZE; i++) {
	746	struct nfsd4_deviceid_map map, n;
	747
	748	list_for_each_entry_safe(map, n, &nfsd_devid_hash[i], hash)
	749	kfree(map);
	750	}
	751	}