path: root/fs/nfsd/filecache.c

/*
 * Open file cache.
 *
 * (c) 2015 - Jeff Layton <jeff.layton@primarydata.com>
 */

#include <linux/hash.h>
#include <linux/slab.h>
#include <linux/file.h>
#include <linux/sched.h>
#include <linux/list_lru.h>
#include <linux/fsnotify_backend.h>
#include <linux/fsnotify.h>
#include <linux/seq_file.h>

#include "vfs.h"
#include "nfsd.h"
#include "nfsfh.h"
#include "netns.h"
#include "filecache.h"
#include "trace.h"

#define NFSDDBG_FACILITY	NFSDDBG_FH

/* FIXME: dynamically size this for the machine somehow? */
#define NFSD_FILE_HASH_BITS                   12
#define NFSD_FILE_HASH_SIZE                  (1 << NFSD_FILE_HASH_BITS)
#define NFSD_LAUNDRETTE_DELAY		     (2 * HZ)

#define NFSD_FILE_SHUTDOWN		     (1)
#define NFSD_FILE_LRU_THRESHOLD		     (4096UL)
#define NFSD_FILE_LRU_LIMIT		     (NFSD_FILE_LRU_THRESHOLD << 2)

/* We only care about NFSD_MAY_READ/WRITE for this cache */
#define NFSD_FILE_MAY_MASK	(NFSD_MAY_READ|NFSD_MAY_WRITE)

struct nfsd_fcache_bucket {
	struct hlist_head	nfb_head;
	spinlock_t		nfb_lock;
	unsigned int		nfb_count;
	unsigned int		nfb_maxcount;
};

static DEFINE_PER_CPU(unsigned long, nfsd_file_cache_hits);

struct nfsd_fcache_disposal {
	struct list_head list;
	struct work_struct work;
	struct net *net;
	spinlock_t lock;
	struct list_head freeme;
	struct rcu_head rcu;
};

static struct workqueue_struct *nfsd_filecache_wq __read_mostly;

static struct kmem_cache		*nfsd_file_slab;
static struct kmem_cache		*nfsd_file_mark_slab;
static struct nfsd_fcache_bucket	*nfsd_file_hashtbl;
static struct list_lru			nfsd_file_lru;
static long				nfsd_file_lru_flags;
static struct fsnotify_group		*nfsd_file_fsnotify_group;
static atomic_long_t			nfsd_filecache_count;
static struct delayed_work		nfsd_filecache_laundrette;
static DEFINE_SPINLOCK(laundrette_lock);
static LIST_HEAD(laundrettes);

static void nfsd_file_gc(void);

static void
nfsd_file_schedule_laundrette(void)
{
	long count = atomic_long_read(&nfsd_filecache_count);

	if (count == 0 || test_bit(NFSD_FILE_SHUTDOWN, &nfsd_file_lru_flags))
		return;

	queue_delayed_work(system_wq, &nfsd_filecache_laundrette,
			NFSD_LAUNDRETTE_DELAY);
}

static void
nfsd_file_slab_free(struct rcu_head *rcu)
{
	struct nfsd_file *nf = container_of(rcu, struct nfsd_file, nf_rcu);

	put_cred(nf->nf_cred);
	kmem_cache_free(nfsd_file_slab, nf);
}

static void
nfsd_file_mark_free(struct fsnotify_mark *mark)
{
	struct nfsd_file_mark *nfm = container_of(mark, struct nfsd_file_mark,
						  nfm_mark);

	kmem_cache_free(nfsd_file_mark_slab, nfm);
}

static struct nfsd_file_mark *
nfsd_file_mark_get(struct nfsd_file_mark *nfm)
{
	if (!refcount_inc_not_zero(&nfm->nfm_ref))
		return NULL;
	return nfm;
}

static void
nfsd_file_mark_put(struct nfsd_file_mark *nfm)
{
	if (refcount_dec_and_test(&nfm->nfm_ref)) {
		fsnotify_destroy_mark(&nfm->nfm_mark, nfsd_file_fsnotify_group);
		fsnotify_put_mark(&nfm->nfm_mark);
	}
}

static struct nfsd_file_mark *
nfsd_file_mark_find_or_create(struct nfsd_file *nf)
{
	int			err;
	struct fsnotify_mark	*mark;
	struct nfsd_file_mark	*nfm = NULL, *new;
	struct inode *inode = nf->nf_inode;

	do {
		mutex_lock(&nfsd_file_fsnotify_group->mark_mutex);
		mark = fsnotify_find_mark(&inode->i_fsnotify_marks,
				nfsd_file_fsnotify_group);
		if (mark) {
			nfm = nfsd_file_mark_get(container_of(mark,
						 struct nfsd_file_mark,
						 nfm_mark));
			mutex_unlock(&nfsd_file_fsnotify_group->mark_mutex);
			if (nfm) {
				fsnotify_put_mark(mark);
				break;
			}
			/* Avoid soft lockup race with nfsd_file_mark_put() */
			fsnotify_destroy_mark(mark, nfsd_file_fsnotify_group);
			fsnotify_put_mark(mark);
		} else
			mutex_unlock(&nfsd_file_fsnotify_group->mark_mutex);

		/* allocate a new nfm */
		new = kmem_cache_alloc(nfsd_file_mark_slab, GFP_KERNEL);
		if (!new)
			return NULL;
		fsnotify_init_mark(&new->nfm_mark, nfsd_file_fsnotify_group);
		new->nfm_mark.mask = FS_ATTRIB|FS_DELETE_SELF;
		refcount_set(&new->nfm_ref, 1);

		err = fsnotify_add_inode_mark(&new->nfm_mark, inode, 0);

		/*
		 * If the add was successful, then return the object.
		 * Otherwise, we need to put the reference we hold on the
		 * nfm_mark. The fsnotify code will take a reference and put
		 * it on failure, so we can't just free it directly. It's also
		 * not safe to call fsnotify_destroy_mark on it as the
		 * mark->group will be NULL. Thus, we can't let the nfm_ref
		 * counter drive the destruction at this point.
		 */
		if (likely(!err))
			nfm = new;
		else
			fsnotify_put_mark(&new->nfm_mark);
	} while (unlikely(err == -EEXIST));

	return nfm;
}

static struct nfsd_file *
nfsd_file_alloc(struct inode *inode, unsigned int may, unsigned int hashval,
		struct net *net)
{
	struct nfsd_file *nf;

	nf = kmem_cache_alloc(nfsd_file_slab, GFP_KERNEL);
	if</