[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 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_NOPNFS)
		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);
	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)) {
		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;
	} else {
		ls = container_of(stid, struct nfs4_layout_stateid, ls_stid);

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

	*lsp = ls;
	return 0;

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);
	update_stateid(&ls->ls_stid.sc_stateid);
	memcpy(&ls->ls_recall_sid, &ls->ls_stid.sc_stateid, sizeof(stateid_t));
	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:
	update_stateid(&ls->ls_stid.sc_stateid);
	memcpy(&lgp->lg_sid, &ls->ls_stid.sc_stateid, sizeof(stateid_t));
	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) {
			update_stateid(&ls->ls_stid.sc_stateid);
			memcpy(&lrp->lr_sid, &ls->ls_stid.sc_stateid,
				sizeof(stateid_t));
		}
		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);

	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 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);
	LIST_HEAD(reaplist);

	switch (task->tk_status) {
	case 0:
		return 1;
	case -NFS4ERR_NOMATCHING_LAYOUT:
		trace_layout_recall_done(&ls->ls_stid.sc_stateid);
		task->tk_status = 0;
		return 1;
	case -NFS4ERR_DELAY:
		/* Poll the client until it's done with the layout */
		/* FIXME: cap number of retries.
		 * The pnfs standard states that we need to only expire
		 * the client after at-least "lease time" .eg lease-time * 2
		 * when failing to communicate a recall
		 */
		rpc_delay(task, HZ/100); /* 10 mili-seconds */
		return 0;
	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 struct nfsd4_callback_ops nfsd4_cb_layout_ops = {
	.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
c5c707f9 CH	25	static struct nfsd4_callback_ops nfsd4_cb_layout_ops;
	26	static const struct lock_manager_operations nfsd4_layouts_lm_ops;
	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
9cf514cc CH	121	if (exp->ex_flags & NFSEXP_NOPNFS)
9cf514cc CH	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);
	204	ls->ls_layout_type = layout_type;
c5c707f9 CH	205	nfsd4_init_cb(&ls->ls_recall, clp, &nfsd4_cb_layout_ops,
	206	NFSPROC4_CLNT_CB_LAYOUT);
	207
	208	if (parent->sc_type == NFS4_DELEG_STID)
	209	ls->ls_file = get_file(fp->fi_deleg_file);
	210	else
	211	ls->ls_file = find_any_file(fp);
	212	BUG_ON(!ls->ls_file);
	213
	214	if (nfsd4_layout_setlease(ls)) {
	215	put_nfs4_file(fp);
	216	kmem_cache_free(nfs4_layout_stateid_cache, ls);
	217	return NULL;
	218	}
9cf514cc CH	219
	220	spin_lock(&clp->cl_lock);
	221	stp->sc_type = NFS4_LAYOUT_STID;
	222	list_add(&ls->ls_perclnt, &clp->cl_lo_states);
	223	spin_unlock(&clp->cl_lock);
	224
	225	spin_lock(&fp->fi_lock);
	226	list_add(&ls->ls_perfile, &fp->fi_lo_states);
	227	spin_unlock(&fp->fi_lock);
	228
31ef83dc	229	trace_layoutstate_alloc(&ls->ls_stid.sc_stateid);
9cf514cc CH	230	return ls;
	231	}
	232
	233	__be32
	234	nfsd4_preprocess_layout_stateid(struct svc_rqst *rqstp,
	235	struct nfsd4_compound_state cstate, stateid_t stateid,
	236	bool create, u32 layout_type, struct nfs4_layout_stateid **lsp)
	237	{
	238	struct nfs4_layout_stateid *ls;
	239	struct nfs4_stid *stid;
	240	unsigned char typemask = NFS4_LAYOUT_STID;
	241	__be32 status;
	242
	243	if (create)
	244	typemask \|= (NFS4_OPEN_STID \| NFS4_LOCK_STID \| NFS4_DELEG_STID);
	245
	246	status = nfsd4_lookup_stateid(cstate, stateid, typemask, &stid,
	247	net_generic(SVC_NET(rqstp), nfsd_net_id));
	248	if (status)
	249	goto out;
	250
	251	if (!fh_match(&cstate->current_fh.fh_handle,
	252	&stid->sc_file->fi_fhandle)) {
	253	status = nfserr_bad_stateid;
	254	goto out_put_stid;
	255	}
	256
	257	if (stid->sc_type != NFS4_LAYOUT_STID) {
	258	ls = nfsd4_alloc_layout_stateid(cstate, stid, layout_type);
	259	nfs4_put_stid(stid);
	260
	261	status = nfserr_jukebox;
	262	if (!ls)
	263	goto out;
	264	} else {
	265	ls = container_of(stid, struct nfs4_layout_stateid, ls_stid);
	266
	267	status = nfserr_bad_stateid;
	268	if (stateid->si_generation > stid->sc_stateid.si_generation)
	269	goto out_put_stid;
	270	if (layout_type != ls->ls_layout_type)
	271	goto out_put_stid;
	272	}
	273
	274	*lsp = ls;
	275	return 0;
	276
	277	out_put_stid:
	278	nfs4_put_stid(stid);
	279	out:
	280	return status;
	281	}
	282
c5c707f9 CH	283	static void
	284	nfsd4_recall_file_layout(struct nfs4_layout_stateid *ls)
	285	{
	286	spin_lock(&ls->ls_lock);
	287	if (ls->ls_recalled)
	288	goto out_unlock;
	289
	290	ls->ls_recalled = true;
	291	atomic_inc(&ls->ls_stid.sc_file->fi_lo_recalls);
	292	if (list_empty(&ls->ls_layouts))
	293	goto out_unlock;
	294
31ef83dc CH	295	trace_layout_recall(&ls->ls_stid.sc_stateid);
31ef83dc CH	296
c5c707f9 CH	297	atomic_inc(&ls->ls_stid.sc_count);
	298	update_stateid(&ls->ls_stid.sc_stateid);
	299	memcpy(&ls->ls_recall_sid, &ls->ls_stid.sc_stateid, sizeof(stateid_t));
	300	nfsd4_run_cb(&ls->ls_recall);
	301
	302	out_unlock:
	303	spin_unlock(&ls->ls_lock);
	304	}
	305
9cf514cc CH	306	static inline u64
	307	layout_end(struct nfsd4_layout_seg *seg)
	308	{
	309	u64 end = seg->offset + seg->length;
	310	return end >= seg->offset ? end : NFS4_MAX_UINT64;
	311	}
	312
	313	static void
	314	layout_update_len(struct nfsd4_layout_seg *lo, u64 end)
	315	{
	316	if (end == NFS4_MAX_UINT64)
	317	lo->length = NFS4_MAX_UINT64;
	318	else
	319	lo->length = end - lo->offset;
	320	}
	321
	322	static bool
	323	layouts_overlapping(struct nfs4_layout lo, struct nfsd4_layout_seg s)
	324	{
	325	if (s->iomode != IOMODE_ANY && s->iomode != lo->lo_seg.iomode)
	326	return false;
	327	if (layout_end(&lo->lo_seg) <= s->offset)
	328	return false;
	329	if (layout_end(s) <= lo->lo_seg.offset)
	330	return false;
	331	return true;
	332	}
	333
	334	static bool
	335	layouts_try_merge(struct nfsd4_layout_seg lo, struct nfsd4_layout_seg new)
	336	{
	337	if (lo->iomode != new->iomode)
	338	return false;
	339	if (layout_end(new) < lo->offset)
	340	return false;
	341	if (layout_end(lo) < new->offset)
	342	return false;
	343
	344	lo->offset = min(lo->offset, new->offset);
	345	layout_update_len(lo, max(layout_end(lo), layout_end(new)));
	346	return true;
	347	}
	348
c5c707f9 CH	349	static __be32
	350	nfsd4_recall_conflict(struct nfs4_layout_stateid *ls)
	351	{
	352	struct nfs4_file *fp = ls->ls_stid.sc_file;
	353	struct nfs4_layout_stateid l, n;
	354	__be32 nfserr = nfs_ok;
	355
	356	assert_spin_locked(&fp->fi_lock);
	357
	358	list_for_each_entry_safe(l, n, &fp->fi_lo_states, ls_perfile) {
	359	if (l != ls) {
	360	nfsd4_recall_file_layout(l);
	361	nfserr = nfserr_recallconflict;
	362	}
	363	}
	364
	365	return nfserr;
	366	}
	367
9cf514cc CH	368	__be32
	369	nfsd4_insert_layout(struct nfsd4_layoutget lgp, struct nfs4_layout_stateid ls)
	370	{
	371	struct nfsd4_layout_seg *seg = &lgp->lg_seg;
c5c707f9	372	struct nfs4_file *fp = ls->ls_stid.sc_file;
9cf514cc	373	struct nfs4_layout lp, new = NULL;
c5c707f9	374	__be32 nfserr;
9cf514cc	375
c5c707f9 CH	376	spin_lock(&fp->fi_lock);
	377	nfserr = nfsd4_recall_conflict(ls);
	378	if (nfserr)
	379	goto out;
9cf514cc CH	380	spin_lock(&ls->ls_lock);
	381	list_for_each_entry(lp, &ls->ls_layouts, lo_perstate) {
	382	if (layouts_try_merge(&lp->lo_seg, seg))
	383	goto done;
	384	}
	385	spin_unlock(&ls->ls_lock);
c5c707f9	386	spin_unlock(&fp->fi_lock);
9cf514cc CH	387
	388	new = kmem_cache_alloc(nfs4_layout_cache, GFP_KERNEL);
	389	if (!new)
	390	return nfserr_jukebox;
	391	memcpy(&new->lo_seg, seg, sizeof(lp->lo_seg));
	392	new->lo_state = ls;
	393
c5c707f9 CH	394	spin_lock(&fp->fi_lock);
	395	nfserr = nfsd4_recall_conflict(ls);
	396	if (nfserr)
	397	goto out;
9cf514cc CH	398	spin_lock(&ls->ls_lock);
	399	list_for_each_entry(lp, &ls->ls_layouts, lo_perstate) {
	400	if (layouts_try_merge(&lp->lo_seg, seg))
	401	goto done;
	402	}
	403
	404	atomic_inc(&ls->ls_stid.sc_count);
	405	list_add_tail(&new->lo_perstate, &ls->ls_layouts);
	406	new = NULL;
	407	done:
	408	update_stateid(&ls->ls_stid.sc_stateid);
	409	memcpy(&lgp->lg_sid, &ls->ls_stid.sc_stateid, sizeof(stateid_t));
	410	spin_unlock(&ls->ls_lock);
c5c707f9 CH	411	out:
c5c707f9 CH	412	spin_unlock(&fp->fi_lock);
9cf514cc CH	413	if (new)
9cf514cc CH	414	kmem_cache_free(nfs4_layout_cache, new);
c5c707f9	415	return nfserr;
9cf514cc CH	416	}
	417
	418	static void
	419	nfsd4_free_layouts(struct list_head *reaplist)
	420	{
	421	while (!list_empty(reaplist)) {
	422	struct nfs4_layout *lp = list_first_entry(reaplist,
	423	struct nfs4_layout, lo_perstate);
	424
	425	list_del(&lp->lo_perstate);
	426	nfs4_put_stid(&lp->lo_state->ls_stid);
	427	kmem_cache_free(nfs4_layout_cache, lp);
	428	}
	429	}
	430
	431	static void
	432	nfsd4_return_file_layout(struct nfs4_layout lp, struct nfsd4_layout_seg seg,
	433	struct list_head *reaplist)
	434	{
	435	struct nfsd4_layout_seg *lo = &lp->lo_seg;
	436	u64 end = layout_end(lo);
	437
	438	if (seg->offset <= lo->offset) {
	439	if (layout_end(seg) >= end) {
	440	list_move_tail(&lp->lo_perstate, reaplist);
	441	return;
	442	}
7890203d	443	lo->offset = layout_end(seg);
9cf514cc CH	444	} else {
	445	/* retain the whole layout segment on a split. */
	446	if (layout_end(seg) < end) {
	447	dprintk("%s: split not supported\n", __func__);
	448	return;
	449	}
7890203d	450	end = seg->offset;
9cf514cc CH	451	}
	452
	453	layout_update_len(lo, end);
	454	}
	455
	456	__be32
	457	nfsd4_return_file_layouts(struct svc_rqst *rqstp,
	458	struct nfsd4_compound_state *cstate,
	459	struct nfsd4_layoutreturn *lrp)
	460	{
	461	struct nfs4_layout_stateid *ls;
	462	struct nfs4_layout lp, n;
	463	LIST_HEAD(reaplist);
	464	__be32 nfserr;
	465	int found = 0;
	466
	467	nfserr = nfsd4_preprocess_layout_stateid(rqstp, cstate, &lrp->lr_sid,
	468	false, lrp->lr_layout_type,
	469	&ls);
31ef83dc CH	470	if (nfserr) {
31ef83dc CH	471	trace_layout_return_lookup_fail(&lrp->lr_sid);
9cf514cc	472	return nfserr;
31ef83dc	473	}
9cf514cc CH	474
	475	spin_lock(&ls->ls_lock);
	476	list_for_each_entry_safe(lp, n, &ls->ls_layouts, lo_perstate) {
	477	if (layouts_overlapping(lp, &lrp->lr_seg)) {
	478	nfsd4_return_file_layout(lp, &lrp->lr_seg, &reaplist);
	479	found++;
	480	}
	481	}
	482	if (!list_empty(&ls->ls_layouts)) {
	483	if (found) {
	484	update_stateid(&ls->ls_stid.sc_stateid);
	485	memcpy(&lrp->lr_sid, &ls->ls_stid.sc_stateid,
	486	sizeof(stateid_t));
	487	}
	488	lrp->lrs_present = 1;
	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
	496	nfs4_put_stid(&ls->ls_stid);
	497	nfsd4_free_layouts(&reaplist);
	498	return nfs_ok;
	499	}
	500
	501	__be32
	502	nfsd4_return_client_layouts(struct svc_rqst *rqstp,
	503	struct nfsd4_compound_state *cstate,
	504	struct nfsd4_layoutreturn *lrp)
	505	{
	506	struct nfs4_layout_stateid ls, n;
	507	struct nfs4_client *clp = cstate->clp;
	508	struct nfs4_layout lp, t;
	509	LIST_HEAD(reaplist);
	510
	511	lrp->lrs_present = 0;
	512
	513	spin_lock(&clp->cl_lock);
	514	list_for_each_entry_safe(ls, n, &clp->cl_lo_states, ls_perclnt) {
6f8f28ec KM	515	if (ls->ls_layout_type != lrp->lr_layout_type)
	516	continue;
	517
9cf514cc CH	518	if (lrp->lr_return_type == RETURN_FSID &&
	519	!fh_fsid_match(&ls->ls_stid.sc_file->fi_fhandle,
	520	&cstate->current_fh.fh_handle))
	521	continue;
	522
	523	spin_lock(&ls->ls_lock);
	524	list_for_each_entry_safe(lp, t, &ls->ls_layouts, lo_perstate) {
	525	if (lrp->lr_seg.iomode == IOMODE_ANY \|\|
	526	lrp->lr_seg.iomode == lp->lo_seg.iomode)
	527	list_move_tail(&lp->lo_perstate, &reaplist);
	528	}
	529	spin_unlock(&ls->ls_lock);
	530	}
	531	spin_unlock(&clp->cl_lock);
	532
	533	nfsd4_free_layouts(&reaplist);
	534	return 0;
	535	}
	536
	537	static void
	538	nfsd4_return_all_layouts(struct nfs4_layout_stateid *ls,
	539	struct list_head *reaplist)
	540	{
	541	spin_lock(&ls->ls_lock);
	542	list_splice_init(&ls->ls_layouts, reaplist);
	543	spin_unlock(&ls->ls_lock);
	544	}
	545
	546	void
	547	nfsd4_return_all_client_layouts(struct nfs4_client *clp)
	548	{
	549	struct nfs4_layout_stateid ls, n;
	550	LIST_HEAD(reaplist);
	551
	552	spin_lock(&clp->cl_lock);
	553	list_for_each_entry_safe(ls, n, &clp->cl_lo_states, ls_perclnt)
	554	nfsd4_return_all_layouts(ls, &reaplist);
	555	spin_unlock(&clp->cl_lock);
	556
	557	nfsd4_free_layouts(&reaplist);
	558	}
	559
	560	void
	561	nfsd4_return_all_file_layouts(struct nfs4_client clp, struct nfs4_file fp)
	562	{
	563	struct nfs4_layout_stateid ls, n;
	564	LIST_HEAD(reaplist);
	565
	566	spin_lock(&fp->fi_lock);
	567	list_for_each_entry_safe(ls, n, &fp->fi_lo_states, ls_perfile) {
	568	if (ls->ls_stid.sc_client == clp)
	569	nfsd4_return_all_layouts(ls, &reaplist);
	570	}
	571	spin_unlock(&fp->fi_lock);
	572
	573	nfsd4_free_layouts(&reaplist);
	574	}
	575
c5c707f9 CH	576	static void
	577	nfsd4_cb_layout_fail(struct nfs4_layout_stateid *ls)
	578	{
	579	struct nfs4_client *clp = ls->ls_stid.sc_client;
	580	char addr_str[INET6_ADDRSTRLEN];
	581	static char *envp[] = {
	582	"HOME=/",
	583	"TERM=linux",
	584	"PATH=/sbin:/usr/sbin:/bin:/usr/bin",
	585	NULL
	586	};
	587	char *argv[8];
	588	int error;
	589
	590	rpc_ntop((struct sockaddr *)&clp->cl_addr, addr_str, sizeof(addr_str));
	591
715a03d2 KM	592	trace_layout_recall_fail(&ls->ls_stid.sc_stateid);
715a03d2 KM	593
c5c707f9 CH	594	printk(KERN_WARNING
	595	"nfsd: client %s failed to respond to layout recall. "
	596	" Fencing..\n", addr_str);
	597
	598	argv[0] = "/sbin/nfsd-recall-failed";
	599	argv[1] = addr_str;
	600	argv[2] = ls->ls_file->f_path.mnt->mnt_sb->s_id;
	601	argv[3] = NULL;
	602
	603	error = call_usermodehelper(argv[0], argv, envp, UMH_WAIT_PROC);
	604	if (error) {
	605	printk(KERN_ERR "nfsd: fence failed for client %s: %d!\n",
	606	addr_str, error);
	607	}
	608	}
	609
	610	static int
	611	nfsd4_cb_layout_done(struct nfsd4_callback cb, struct rpc_task task)
	612	{
	613	struct nfs4_layout_stateid *ls =
	614	container_of(cb, struct nfs4_layout_stateid, ls_recall);
	615	LIST_HEAD(reaplist);
	616
	617	switch (task->tk_status) {
	618	case 0:
	619	return 1;
	620	case -NFS4ERR_NOMATCHING_LAYOUT:
31ef83dc	621	trace_layout_recall_done(&ls->ls_stid.sc_stateid);
c5c707f9 CH	622	task->tk_status = 0;
	623	return 1;
	624	case -NFS4ERR_DELAY:
	625	/* Poll the client until it's done with the layout */
	626	/* FIXME: cap number of retries.
	627	* The pnfs standard states that we need to only expire
	628	* the client after at-least "lease time" .eg lease-time * 2
	629	* when failing to communicate a recall
	630	*/
	631	rpc_delay(task, HZ/100); /* 10 mili-seconds */
	632	return 0;
	633	default:
	634	/*
	635	* Unknown error or non-responding client, we'll need to fence.
	636	*/
	637	nfsd4_cb_layout_fail(ls);
	638	return -1;
	639	}
	640	}
	641
	642	static void
	643	nfsd4_cb_layout_release(struct nfsd4_callback *cb)
	644	{
	645	struct nfs4_layout_stateid *ls =
	646	container_of(cb, struct nfs4_layout_stateid, ls_recall);
	647	LIST_HEAD(reaplist);
	648
31ef83dc CH	649	trace_layout_recall_release(&ls->ls_stid.sc_stateid);
31ef83dc CH	650
c5c707f9 CH	651	nfsd4_return_all_layouts(ls, &reaplist);
	652	nfsd4_free_layouts(&reaplist);
	653	nfs4_put_stid(&ls->ls_stid);
	654	}
	655
	656	static struct nfsd4_callback_ops nfsd4_cb_layout_ops = {
	657	.done = nfsd4_cb_layout_done,
	658	.release = nfsd4_cb_layout_release,
	659	};
	660
	661	static bool
	662	nfsd4_layout_lm_break(struct file_lock *fl)
	663	{
	664	/*
	665	* We don't want the locks code to timeout the lease for us;
	666	* we'll remove it ourself if a layout isn't returned
	667	* in time:
	668	*/
	669	fl->fl_break_time = 0;
	670	nfsd4_recall_file_layout(fl->fl_owner);
	671	return false;
	672	}
	673
	674	static int
	675	nfsd4_layout_lm_change(struct file_lock *onlist, int arg,
	676	struct list_head *dispose)
	677	{
	678	BUG_ON(!(arg & F_UNLCK));
	679	return lease_modify(onlist, arg, dispose);
	680	}
	681
	682	static const struct lock_manager_operations nfsd4_layouts_lm_ops = {
	683	.lm_break = nfsd4_layout_lm_break,
	684	.lm_change = nfsd4_layout_lm_change,
	685	};
	686
9cf514cc CH	687	int
	688	nfsd4_init_pnfs(void)
	689	{
	690	int i;
	691
	692	for (i = 0; i < DEVID_HASH_SIZE; i++)
	693	INIT_LIST_HEAD(&nfsd_devid_hash[i]);
	694
	695	nfs4_layout_cache = kmem_cache_create("nfs4_layout",
	696	sizeof(struct nfs4_layout), 0, 0, NULL);
	697	if (!nfs4_layout_cache)
	698	return -ENOMEM;
	699
	700	nfs4_layout_stateid_cache = kmem_cache_create("nfs4_layout_stateid",
	701	sizeof(struct nfs4_layout_stateid), 0, 0, NULL);
	702	if (!nfs4_layout_stateid_cache) {
	703	kmem_cache_destroy(nfs4_layout_cache);
	704	return -ENOMEM;
	705	}
	706	return 0;
	707	}
	708
	709	void
	710	nfsd4_exit_pnfs(void)
	711	{
	712	int i;
	713
	714	kmem_cache_destroy(nfs4_layout_cache);
	715	kmem_cache_destroy(nfs4_layout_stateid_cache);
	716
	717	for (i = 0; i < DEVID_HASH_SIZE; i++) {
	718	struct nfsd4_deviceid_map map, n;
	719
	720	list_for_each_entry_safe(map, n, &nfsd_devid_hash[i], hash)
	721	kfree(map);
	722	}
	723	}