summaryrefslogtreecommitdiff
path: root/fs
diff options
context:
space:
mode:
Diffstat (limited to 'fs')
-rw-r--r--fs/block_dev.c7
-rw-r--r--fs/btrfs/disk-io.c2
-rw-r--r--fs/btrfs/extent_io.c4
-rw-r--r--fs/btrfs/inode.c60
-rw-r--r--fs/btrfs/print-tree.c10
-rw-r--r--fs/btrfs/print-tree.h2
-rw-r--r--fs/btrfs/relocation.c7
-rw-r--r--fs/btrfs/space-info.c4
-rw-r--r--fs/btrfs/tree-checker.c7
-rw-r--r--fs/io_uring.c256
-rw-r--r--fs/nfs/delegation.c12
-rw-r--r--fs/nfs/internal.h38
-rw-r--r--fs/nfs/nfs4proc.c28
-rw-r--r--fs/nfs/nfs4super.c4
-rw-r--r--fs/nfs/pnfs.c67
-rw-r--r--fs/nfs/pnfs.h8
-rw-r--r--fs/nfs/pnfs_nfs.c22
-rw-r--r--fs/nfsd/nfs4proc.c5
-rw-r--r--fs/nfsd/nfs4xdr.c56
-rw-r--r--fs/nfsd/nfssvc.c6
-rw-r--r--fs/nfsd/xdr4.h1
-rw-r--r--fs/notify/fanotify/fanotify_user.c17
-rw-r--r--fs/zonefs/Kconfig1
23 files changed, 386 insertions, 238 deletions
diff --git a/fs/block_dev.c b/fs/block_dev.c
index 3e5b02f6606c..3b8963e228a1 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -605,6 +605,8 @@ int thaw_bdev(struct block_device *bdev)
error = thaw_super(sb);
if (error)
bdev->bd_fsfreeze_count++;
+ else
+ bdev->bd_fsfreeze_sb = NULL;
out:
mutex_unlock(&bdev->bd_fsfreeze_mutex);
return error;
@@ -774,8 +776,11 @@ static struct kmem_cache * bdev_cachep __read_mostly;
static struct inode *bdev_alloc_inode(struct super_block *sb)
{
struct bdev_inode *ei = kmem_cache_alloc(bdev_cachep, GFP_KERNEL);
+
if (!ei)
return NULL;
+ memset(&ei->bdev, 0, sizeof(ei->bdev));
+ ei->bdev.bd_bdi = &noop_backing_dev_info;
return &ei->vfs_inode;
}
@@ -869,14 +874,12 @@ struct block_device *bdev_alloc(struct gendisk *disk, u8 partno)
mapping_set_gfp_mask(&inode->i_data, GFP_USER);
bdev = I_BDEV(inode);
- memset(bdev, 0, sizeof(*bdev));
mutex_init(&bdev->bd_mutex);
mutex_init(&bdev->bd_fsfreeze_mutex);
spin_lock_init(&bdev->bd_size_lock);
bdev->bd_disk = disk;
bdev->bd_partno = partno;
bdev->bd_inode = inode;
- bdev->bd_bdi = &noop_backing_dev_info;
#ifdef CONFIG_SYSFS
INIT_LIST_HEAD(&bdev->bd_holder_disks);
#endif
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index 1dfd4b2d0e1e..6b35b7e88136 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -1457,7 +1457,7 @@ void btrfs_check_leaked_roots(struct btrfs_fs_info *fs_info)
root = list_first_entry(&fs_info->allocated_roots,
struct btrfs_root, leak_list);
btrfs_err(fs_info, "leaked root %s refcount %d",
- btrfs_root_name(root->root_key.objectid, buf),
+ btrfs_root_name(&root->root_key, buf),
refcount_read(&root->refs));
while (refcount_read(&root->refs) > 1)
btrfs_put_root(root);
diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index 6e3b72e63e42..c9cee458e001 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -676,9 +676,7 @@ alloc_extent_state_atomic(struct extent_state *prealloc)
static void extent_io_tree_panic(struct extent_io_tree *tree, int err)
{
- struct inode *inode = tree->private_data;
-
- btrfs_panic(btrfs_sb(inode->i_sb), err,
+ btrfs_panic(tree->fs_info, err,
"locking error: extent tree was modified by another thread while locked");
}
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 070716650df8..a8e0a6b038d3 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -9390,7 +9390,8 @@ static struct btrfs_delalloc_work *btrfs_alloc_delalloc_work(struct inode *inode
* some fairly slow code that needs optimization. This walks the list
* of all the inodes with pending delalloc and forces them to disk.
*/
-static int start_delalloc_inodes(struct btrfs_root *root, u64 *nr, bool snapshot,
+static int start_delalloc_inodes(struct btrfs_root *root,
+ struct writeback_control *wbc, bool snapshot,
bool in_reclaim_context)
{
struct btrfs_inode *binode;
@@ -9399,6 +9400,7 @@ static int start_delalloc_inodes(struct btrfs_root *root, u64 *nr, bool snapshot
struct list_head works;
struct list_head splice;
int ret = 0;
+ bool full_flush = wbc->nr_to_write == LONG_MAX;
INIT_LIST_HEAD(&works);
INIT_LIST_HEAD(&splice);
@@ -9427,18 +9429,24 @@ static int start_delalloc_inodes(struct btrfs_root *root, u64 *nr, bool snapshot
if (snapshot)
set_bit(BTRFS_INODE_SNAPSHOT_FLUSH,
&binode->runtime_flags);
- work = btrfs_alloc_delalloc_work(inode);
- if (!work) {
- iput(inode);
- ret = -ENOMEM;
- goto out;
- }
- list_add_tail(&work->list, &works);
- btrfs_queue_work(root->fs_info->flush_workers,
- &work->work);
- if (*nr != U64_MAX) {
- (*nr)--;
- if (*nr == 0)
+ if (full_flush) {
+ work = btrfs_alloc_delalloc_work(inode);
+ if (!work) {
+ iput(inode);
+ ret = -ENOMEM;
+ goto out;
+ }
+ list_add_tail(&work->list, &works);
+ btrfs_queue_work(root->fs_info->flush_workers,
+ &work->work);
+ } else {
+ ret = sync_inode(inode, wbc);
+ if (!ret &&
+ test_bit(BTRFS_INODE_HAS_ASYNC_EXTENT,
+ &BTRFS_I(inode)->runtime_flags))
+ ret = sync_inode(inode, wbc);
+ btrfs_add_delayed_iput(inode);
+ if (ret || wbc->nr_to_write <= 0)
goto out;
}
cond_resched();
@@ -9464,18 +9472,29 @@ out:
int btrfs_start_delalloc_snapshot(struct btrfs_root *root)
{
+ struct writeback_control wbc = {
+ .nr_to_write = LONG_MAX,
+ .sync_mode = WB_SYNC_NONE,
+ .range_start = 0,
+ .range_end = LLONG_MAX,
+ };
struct btrfs_fs_info *fs_info = root->fs_info;
- u64 nr = U64_MAX;
if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state))
return -EROFS;
- return start_delalloc_inodes(root, &nr, true, false);
+ return start_delalloc_inodes(root, &wbc, true, false);
}
int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, u64 nr,
bool in_reclaim_context)
{
+ struct writeback_control wbc = {
+ .nr_to_write = (nr == U64_MAX) ? LONG_MAX : (unsigned long)nr,
+ .sync_mode = WB_SYNC_NONE,
+ .range_start = 0,
+ .range_end = LLONG_MAX,
+ };
struct btrfs_root *root;
struct list_head splice;
int ret;
@@ -9489,6 +9508,13 @@ int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, u64 nr,
spin_lock(&fs_info->delalloc_root_lock);
list_splice_init(&fs_info->delalloc_roots, &splice);
while (!list_empty(&splice) && nr) {
+ /*
+ * Reset nr_to_write here so we know that we're doing a full
+ * flush.
+ */
+ if (nr == U64_MAX)
+ wbc.nr_to_write = LONG_MAX;
+
root = list_first_entry(&splice, struct btrfs_root,
delalloc_root);
root = btrfs_grab_root(root);
@@ -9497,9 +9523,9 @@ int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, u64 nr,
&fs_info->delalloc_roots);
spin_unlock(&fs_info->delalloc_root_lock);
- ret = start_delalloc_inodes(root, &nr, false, in_reclaim_context);
+ ret = start_delalloc_inodes(root, &wbc, false, in_reclaim_context);
btrfs_put_root(root);
- if (ret < 0)
+ if (ret < 0 || wbc.nr_to_write <= 0)
goto out;
spin_lock(&fs_info->delalloc_root_lock);
}
diff --git a/fs/btrfs/print-tree.c b/fs/btrfs/print-tree.c
index fe5e0026129d..aae1027bd76a 100644
--- a/fs/btrfs/print-tree.c
+++ b/fs/btrfs/print-tree.c
@@ -26,22 +26,22 @@ static const struct root_name_map root_map[] = {
{ BTRFS_DATA_RELOC_TREE_OBJECTID, "DATA_RELOC_TREE" },
};
-const char *btrfs_root_name(u64 objectid, char *buf)
+const char *btrfs_root_name(const struct btrfs_key *key, char *buf)
{
int i;
- if (objectid == BTRFS_TREE_RELOC_OBJECTID) {
+ if (key->objectid == BTRFS_TREE_RELOC_OBJECTID) {
snprintf(buf, BTRFS_ROOT_NAME_BUF_LEN,
- "TREE_RELOC offset=%llu", objectid);
+ "TREE_RELOC offset=%llu", key->offset);
return buf;
}
for (i = 0; i < ARRAY_SIZE(root_map); i++) {
- if (root_map[i].id == objectid)
+ if (root_map[i].id == key->objectid)
return root_map[i].name;
}
- snprintf(buf, BTRFS_ROOT_NAME_BUF_LEN, "%llu", objectid);
+ snprintf(buf, BTRFS_ROOT_NAME_BUF_LEN, "%llu", key->objectid);
return buf;
}
diff --git a/fs/btrfs/print-tree.h b/fs/btrfs/print-tree.h
index 78b99385a503..8c3e9319ec4e 100644
--- a/fs/btrfs/print-tree.h
+++ b/fs/btrfs/print-tree.h
@@ -11,6 +11,6 @@
void btrfs_print_leaf(struct extent_buffer *l);
void btrfs_print_tree(struct extent_buffer *c, bool follow);
-const char *btrfs_root_name(u64 objectid, char *buf);
+const char *btrfs_root_name(const struct btrfs_key *key, char *buf);
#endif
diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c
index 19b7db8b2117..df63ef64c5c0 100644
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -2975,11 +2975,16 @@ static int delete_v1_space_cache(struct extent_buffer *leaf,
return 0;
for (i = 0; i < btrfs_header_nritems(leaf); i++) {
+ u8 type;
+
btrfs_item_key_to_cpu(leaf, &key, i);
if (key.type != BTRFS_EXTENT_DATA_KEY)
continue;
ei = btrfs_item_ptr(leaf, i, struct btrfs_file_extent_item);
- if (btrfs_file_extent_type(leaf, ei) == BTRFS_FILE_EXTENT_REG &&
+ type = btrfs_file_extent_type(leaf, ei);
+
+ if ((type == BTRFS_FILE_EXTENT_REG ||
+ type == BTRFS_FILE_EXTENT_PREALLOC) &&
btrfs_file_extent_disk_bytenr(leaf, ei) == data_bytenr) {
found = true;
space_cache_ino = key.objectid;
diff --git a/fs/btrfs/space-info.c b/fs/btrfs/space-info.c
index 67e55c5479b8..e8347461c8dd 100644
--- a/fs/btrfs/space-info.c
+++ b/fs/btrfs/space-info.c
@@ -532,7 +532,9 @@ static void shrink_delalloc(struct btrfs_fs_info *fs_info,
loops = 0;
while ((delalloc_bytes || dio_bytes) && loops < 3) {
- btrfs_start_delalloc_roots(fs_info, items, true);
+ u64 nr_pages = min(delalloc_bytes, to_reclaim) >> PAGE_SHIFT;
+
+ btrfs_start_delalloc_roots(fs_info, nr_pages, true);
loops++;
if (wait_ordered && !trans) {
diff --git a/fs/btrfs/tree-checker.c b/fs/btrfs/tree-checker.c
index 028e733e42f3..582061c7b547 100644
--- a/fs/btrfs/tree-checker.c
+++ b/fs/btrfs/tree-checker.c
@@ -760,6 +760,7 @@ int btrfs_check_chunk_valid(struct extent_buffer *leaf,
{
struct btrfs_fs_info *fs_info = leaf->fs_info;
u64 length;
+ u64 chunk_end;
u64 stripe_len;
u16 num_stripes;
u16 sub_stripes;
@@ -814,6 +815,12 @@ int btrfs_check_chunk_valid(struct extent_buffer *leaf,
"invalid chunk length, have %llu", length);
return -EUCLEAN;
}
+ if (unlikely(check_add_overflow(logical, length, &chunk_end))) {
+ chunk_err(leaf, chunk, logical,
+"invalid chunk logical start and length, have logical start %llu length %llu",
+ logical, length);
+ return -EUCLEAN;
+ }
if (unlikely(!is_power_of_2(stripe_len) || stripe_len != BTRFS_STRIPE_LEN)) {
chunk_err(leaf, chunk, logical,
"invalid chunk stripe length: %llu",
diff --git a/fs/io_uring.c b/fs/io_uring.c
index ca46f314640b..2f305c097bd5 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -262,6 +262,7 @@ struct io_ring_ctx {
unsigned int drain_next: 1;
unsigned int eventfd_async: 1;
unsigned int restricted: 1;
+ unsigned int sqo_dead: 1;
/*
* Ring buffer of indices into array of io_uring_sqe, which is
@@ -992,6 +993,9 @@ enum io_mem_account {
ACCT_PINNED,
};
+static void __io_uring_cancel_task_requests(struct io_ring_ctx *ctx,
+ struct task_struct *task);
+
static void destroy_fixed_file_ref_node(struct fixed_file_ref_node *ref_node);
static struct fixed_file_ref_node *alloc_fixed_file_ref_node(
struct io_ring_ctx *ctx);
@@ -1342,11 +1346,6 @@ static void __io_commit_cqring(struct io_ring_ctx *ctx)
/* order cqe stores with ring update */
smp_store_release(&rings->cq.tail, ctx->cached_cq_tail);
-
- if (wq_has_sleeper(&ctx->cq_wait)) {
- wake_up_interruptible(&ctx->cq_wait);
- kill_fasync(&ctx->cq_fasync, SIGIO, POLL_IN);
- }
}
static void io_put_identity(struct io_uring_task *tctx, struct io_kiocb *req)
@@ -1520,10 +1519,8 @@ static void io_prep_async_work(struct io_kiocb *req)
{
const struct io_op_def *def = &io_op_defs[req->opcode];
struct io_ring_ctx *ctx = req->ctx;
- struct io_identity *id;
io_req_init_async(req);
- id = req->work.identity;
if (req->flags & REQ_F_FORCE_ASYNC)
req->work.flags |= IO_WQ_WORK_CONCURRENT;
@@ -1704,18 +1701,42 @@ static inline unsigned __io_cqring_events(struct io_ring_ctx *ctx)
static void io_cqring_ev_posted(struct io_ring_ctx *ctx)
{
+ /* see waitqueue_active() comment */
+ smp_mb();
+
if (waitqueue_active(&ctx->wait))
wake_up(&ctx->wait);
if (ctx->sq_data && waitqueue_active(&ctx->sq_data->wait))
wake_up(&ctx->sq_data->wait);
if (io_should_trigger_evfd(ctx))
eventfd_signal(ctx->cq_ev_fd, 1);
+ if (waitqueue_active(&ctx->cq_wait)) {
+ wake_up_interruptible(&ctx->cq_wait);
+ kill_fasync(&ctx->cq_fasync, SIGIO, POLL_IN);
+ }
+}
+
+static void io_cqring_ev_posted_iopoll(struct io_ring_ctx *ctx)
+{
+ /* see waitqueue_active() comment */
+ smp_mb();
+
+ if (ctx->flags & IORING_SETUP_SQPOLL) {
+ if (waitqueue_active(&ctx->wait))
+ wake_up(&ctx->wait);
+ }
+ if (io_should_trigger_evfd(ctx))
+ eventfd_signal(ctx->cq_ev_fd, 1);
+ if (waitqueue_active(&ctx->cq_wait)) {
+ wake_up_interruptible(&ctx->cq_wait);
+ kill_fasync(&ctx->cq_fasync, SIGIO, POLL_IN);
+ }
}
/* Returns true if there are no backlogged entries after the flush */
-static bool io_cqring_overflow_flush(struct io_ring_ctx *ctx, bool force,
- struct task_struct *tsk,
- struct files_struct *files)
+static bool __io_cqring_overflow_flush(struct io_ring_ctx *ctx, bool force,
+ struct task_struct *tsk,
+ struct files_struct *files)
{
struct io_rings *rings = ctx->rings;
struct io_kiocb *req, *tmp;
@@ -1768,6 +1789,20 @@ static bool io_cqring_overflow_flush(struct io_ring_ctx *ctx, bool force,
return all_flushed;
}
+static void io_cqring_overflow_flush(struct io_ring_ctx *ctx, bool force,
+ struct task_struct *tsk,
+ struct files_struct *files)
+{
+ if (test_bit(0, &ctx->cq_check_overflow)) {
+ /* iopoll syncs against uring_lock, not completion_lock */
+ if (ctx->flags & IORING_SETUP_IOPOLL)
+ mutex_lock(&ctx->uring_lock);
+ __io_cqring_overflow_flush(ctx, force, tsk, files);
+ if (ctx->flags & IORING_SETUP_IOPOLL)
+ mutex_unlock(&ctx->uring_lock);
+ }
+}
+
static void __io_cqring_fill_event(struct io_kiocb *req, long res, long cflags)
{
struct io_ring_ctx *ctx = req->ctx;
@@ -2127,14 +2162,14 @@ static void __io_req_task_submit(struct io_kiocb *req)
{
struct io_ring_ctx *ctx = req->ctx;
- if (!__io_sq_thread_acquire_mm(ctx) &&
- !__io_sq_thread_acquire_files(ctx)) {
- mutex_lock(&ctx->uring_lock);
+ mutex_lock(&ctx->uring_lock);
+ if (!ctx->sqo_dead &&
+ !__io_sq_thread_acquire_mm(ctx) &&
+ !__io_sq_thread_acquire_files(ctx))
__io_queue_sqe(req, NULL);
- mutex_unlock(&ctx->uring_lock);
- } else {
+ else
__io_req_task_cancel(req, -EFAULT);
- }
+ mutex_unlock(&ctx->uring_lock);
}
static void io_req_task_submit(struct callback_head *cb)
@@ -2313,20 +2348,8 @@ static void io_double_put_req(struct io_kiocb *req)
io_free_req(req);
}
-static unsigned io_cqring_events(struct io_ring_ctx *ctx, bool noflush)
+static unsigned io_cqring_events(struct io_ring_ctx *ctx)
{
- if (test_bit(0, &ctx->cq_check_overflow)) {
- /*
- * noflush == true is from the waitqueue handler, just ensure
- * we wake up the task, and the next invocation will flush the
- * entries. We cannot safely to it from here.
- */
- if (noflush)
- return -1U;
-
- io_cqring_overflow_flush(ctx, false, NULL, NULL);
- }
-
/* See comment at the top of this file */
smp_rmb();
return __io_cqring_events(ctx);
@@ -2424,8 +2447,7 @@ static void io_iopoll_complete(struct io_ring_ctx *ctx, unsigned int *nr_events,
}
io_commit_cqring(ctx);
- if (ctx->flags & IORING_SETUP_SQPOLL)
- io_cqring_ev_posted(ctx);
+ io_cqring_ev_posted_iopoll(ctx);
io_req_free_batch_finish(ctx, &rb);
if (!list_empty(&again))
@@ -2551,7 +2573,9 @@ static int io_iopoll_check(struct io_ring_ctx *ctx, long min)
* If we do, we can potentially be spinning for commands that
* already triggered a CQE (eg in error).
*/
- if (io_cqring_events(ctx, false))
+ if (test_bit(0, &ctx->cq_check_overflow))
+ __io_cqring_overflow_flush(ctx, false, NULL, NULL);
+ if (io_cqring_events(ctx))
break;
/*
@@ -2668,6 +2692,8 @@ static bool io_rw_reissue(struct io_kiocb *req, long res)
if ((res != -EAGAIN && res != -EOPNOTSUPP) || io_wq_current_is_worker())
return false;
+ lockdep_assert_held(&req->ctx->uring_lock);
+
ret = io_sq_thread_acquire_mm_files(req->ctx, req);
if (io_resubmit_prep(req, ret)) {
@@ -6826,7 +6852,7 @@ static int io_submit_sqes(struct io_ring_ctx *ctx, unsigned int nr)
/* if we have a backlog and couldn't flush it all, return BUSY */
if (test_bit(0, &ctx->sq_check_overflow)) {
- if (!io_cqring_overflow_flush(ctx, false, NULL, NULL))
+ if (!__io_cqring_overflow_flush(ctx, false, NULL, NULL))
return -EBUSY;
}
@@ -6928,7 +6954,8 @@ static int __io_sq_thread(struct io_ring_ctx *ctx, bool cap_entries)
if (!list_empty(&ctx->iopoll_list))
io_do_iopoll(ctx, &nr_events, 0);
- if (to_submit && likely(!percpu_ref_is_dying(&ctx->refs)))
+ if (to_submit && !ctx->sqo_dead &&
+ likely(!percpu_ref_is_dying(&ctx->refs)))
ret = io_submit_sqes(ctx, to_submit);
mutex_unlock(&ctx->uring_lock);
}
@@ -7089,7 +7116,7 @@ struct io_wait_queue {
unsigned nr_timeouts;
};
-static inline bool io_should_wake(struct io_wait_queue *iowq, bool noflush)
+static inline bool io_should_wake(struct io_wait_queue *iowq)
{
struct io_ring_ctx *ctx = iowq->ctx;
@@ -7098,7 +7125,7 @@ static inline bool io_should_wake(struct io_wait_queue *iowq, bool noflush)
* started waiting. For timeouts, we always want to return to userspace,
* regardless of event count.
*/
- return io_cqring_events(ctx, noflush) >= iowq->to_wait ||
+ return io_cqring_events(ctx) >= iowq->to_wait ||
atomic_read(&ctx->cq_timeouts) != iowq->nr_timeouts;
}
@@ -7108,11 +7135,13 @@ static int io_wake_function(struct wait_queue_entry *curr, unsigned int mode,
struct io_wait_queue *iowq = container_of(curr, struct io_wait_queue,
wq);
- /* use noflush == true, as we can't safely rely on locking context */
- if (!io_should_wake(iowq, true))
- return -1;
-
- return autoremove_wake_function(curr, mode, wake_flags, key);
+ /*
+ * Cannot safely flush overflowed CQEs from here, ensure we wake up
+ * the task, and the next invocation will do it.
+ */
+ if (io_should_wake(iowq) || test_bit(0, &iowq->ctx->cq_check_overflow))
+ return autoremove_wake_function(curr, mode, wake_flags, key);
+ return -1;
}
static int io_run_task_work_sig(void)
@@ -7149,7 +7178,8 @@ static int io_cqring_wait(struct io_ring_ctx *ctx, int min_events,
int ret = 0;
do {
- if (io_cqring_events(ctx, false) >= min_events)
+ io_cqring_overflow_flush(ctx, false, NULL, NULL);
+ if (io_cqring_events(ctx) >= min_events)
return 0;
if (!io_run_task_work())
break;
@@ -7177,6 +7207,7 @@ static int io_cqring_wait(struct io_ring_ctx *ctx, int min_events,
iowq.nr_timeouts = atomic_read(&ctx->cq_timeouts);
trace_io_uring_cqring_wait(ctx, min_events);
do {
+ io_cqring_overflow_flush(ctx, false, NULL, NULL);
prepare_to_wait_exclusive(&ctx->wait, &iowq.wq,
TASK_INTERRUPTIBLE);
/* make sure we run task_work before checking for signals */
@@ -7185,8 +7216,10 @@ static int io_cqring_wait(struct io_ring_ctx *ctx, int min_events,
continue;
else if (ret < 0)
break;
- if (io_should_wake(&iowq, false))
+ if (io_should_wake(&iowq))
break;
+ if (test_bit(0, &ctx->cq_check_overflow))
+ continue;
if (uts) {
timeout = schedule_timeout(timeout);
if (timeout == 0) {
@@ -7684,12 +7717,12 @@ static struct fixed_file_ref_node *alloc_fixed_file_ref_node(
ref_node = kzalloc(sizeof(*ref_node), GFP_KERNEL);
if (!ref_node)
- return ERR_PTR(-ENOMEM);
+ return NULL;
if (percpu_ref_init(&ref_node->refs, io_file_data_ref_zero,
0, GFP_KERNEL)) {
kfree(ref_node);
- return ERR_PTR(-ENOMEM);
+ return NULL;
}
INIT_LIST_HEAD(&ref_node->node);
INIT_LIST_HEAD(&ref_node->file_list);
@@ -7783,9 +7816,9 @@ static int io_sqe_files_register(struct io_ring_ctx *ctx, void __user *arg,
}
ref_node = alloc_fixed_file_ref_node(ctx);
- if (IS_ERR(ref_node)) {
+ if (!ref_node) {
io_sqe_files_unregister(ctx);
- return PTR_ERR(ref_node);
+ return -ENOMEM;
}
io_sqe_files_set_node(file_data, ref_node);
@@ -7885,8 +7918,8 @@ static int __io_sqe_files_update(struct io_ring_ctx *ctx,
return -EINVAL;
ref_node = alloc_fixed_file_ref_node(ctx);
- if (IS_ERR(ref_node))
- return PTR_ERR(ref_node);
+ if (!ref_node)
+ return -ENOMEM;
done = 0;
fds = u64_to_user_ptr(up->fds);
@@ -8624,7 +8657,8 @@ static __poll_t io_uring_poll(struct file *file, poll_table *wait)
smp_rmb();
if (!io_sqring_full(ctx))
mask |= EPOLLOUT | EPOLLWRNORM;
- if (io_cqring_events(ctx, false))
+ io_cqring_overflow_flush(ctx, false, NULL, NULL);
+ if (io_cqring_events(ctx))
mask |= EPOLLIN | EPOLLRDNORM;
return mask;
@@ -8663,7 +8697,7 @@ static void io_ring_exit_work(struct work_struct *work)
* as nobody else will be looking for them.
*/
do {
- io_iopoll_try_reap_events(ctx);
+ __io_uring_cancel_task_requests(ctx, NULL);
} while (!wait_for_completion_timeout(&ctx->ref_comp, HZ/20));
io_ring_ctx_free(ctx);
}
@@ -8679,10 +8713,14 @@ static void io_ring_ctx_wait_and_kill(struct io_ring_ctx *ctx)
{
mutex_lock(&ctx->uring_lock);
percpu_ref_kill(&ctx->refs);
+
+ if (WARN_ON_ONCE((ctx->flags & IORING_SETUP_SQPOLL) && !ctx->sqo_dead))
+ ctx->sqo_dead = 1;
+
/* if force is set, the ring is going away. always drop after that */
ctx->cq_overflow_flushed = 1;
if (ctx->rings)
- io_cqring_overflow_flush(ctx, true, NULL, NULL);
+ __io_cqring_overflow_flush(ctx, true, NULL, NULL);
mutex_unlock(&ctx->uring_lock);
io_kill_timeouts(ctx, NULL, NULL);
@@ -8818,9 +8856,11 @@ static void __io_uring_cancel_task_requests(struct io_ring_ctx *ctx,
enum io_wq_cancel cret;
bool ret = false;
- cret = io_wq_cancel_cb(ctx->io_wq, io_cancel_task_cb, &cancel, true);
- if (cret != IO_WQ_CANCEL_NOTFOUND)
- ret = true;
+ if (ctx->io_wq) {
+ cret = io_wq_cancel_cb(ctx->io_wq, io_cancel_task_cb,
+ &cancel, true);
+ ret |= (cret != IO_WQ_CANCEL_NOTFOUND);
+ }
/* SQPOLL thread does its own polling */
if (!(ctx->flags & IORING_SETUP_SQPOLL)) {
@@ -8839,6 +8879,18 @@ static void __io_uring_cancel_task_requests(struct io_ring_ctx *ctx,
}
}
+static void io_disable_sqo_submit(struct io_ring_ctx *ctx)
+{
+ WARN_ON_ONCE(ctx->sqo_task != current);
+
+ mutex_lock(&ctx->uring_lock);
+ ctx->sqo_dead = 1;
+ mutex_unlock(&ctx->uring_lock);
+
+ /* make sure callers enter the ring to get error */
+ io_ring_set_wakeup_flag(ctx);
+}
+
/*
* We need to iteratively cancel requests, in case a request has dependent
* hard links. These persist even for failure of cancelations, hence keep
@@ -8850,15 +8902,15 @@ static void io_uring_cancel_task_requests(struct io_ring_ctx *ctx,
struct task_struct *task = current;
if ((ctx->flags & IORING_SETUP_SQPOLL) && ctx->sq_data) {
+ /* for SQPOLL only sqo_task has task notes */
+ io_disable_sqo_submit(ctx);
task = ctx->sq_data->thread;
atomic_inc(&task->io_uring->in_idle);
io_sq_thread_park(ctx->sq_data);
}
io_cancel_defer_files(ctx, task, files);
- io_ring_submit_lock(ctx, (ctx->flags & IORING_SETUP_IOPOLL));
io_cqring_overflow_flush(ctx, true, task, files);
- io_ring_submit_unlock(ctx, (ctx->flags & IORING_SETUP_IOPOLL));
if (!files)
__io_uring_cancel_task_requests(ctx, task);
@@ -8931,20 +8983,12 @@ static void io_uring_del_task_file(struct file *file)
fput(file);
}
-/*
- * Drop task note for this file if we're the only ones that hold it after
- * pending fput()
- */
-static void io_uring_attempt_task_drop(struct file *file)
+static void io_uring_remove_task_files(struct io_uring_task *tctx)
{
- if (!current->io_uring)
- return;
- /*
- * fput() is pending, will be 2 if the only other ref is our potential
- * task file note. If the task is exiting, drop regardless of count.
- */
- if (fatal_signal_pending(current) || (current->flags & PF_EXITING) ||
- atomic_long_read(&file->f_count) == 2)
+ struct file *file;
+ unsigned long index;
+
+ xa_for_each(&tctx->xa, index, file)
io_uring_del_task_file(file);
}
@@ -8956,16 +9000,12 @@ void __io_uring_files_cancel(struct files_struct *files)
/* make sure overflow events are dropped */
atomic_inc(&tctx->in_idle);
-
- xa_for_each(&tctx->xa, index, file) {
- struct io_ring_ctx *ctx = file->private_data;
-
- io_uring_cancel_task_requests(ctx, files);
- if (files)
- io_uring_del_task_file(file);
- }
-
+ xa_for_each(&tctx->xa, index, file)
+ io_uring_cancel_task_requests(file->private_data, files);
atomic_dec(&tctx->in_idle);
+
+ if (files)
+ io_uring_remove_task_files(tctx);
}
static s64 tctx_inflight(struct io_uring_task *tctx)
@@ -9028,11 +9068,39 @@ void __io_uring_task_cancel(void)
} while (1);
atomic_dec(&tctx->in_idle);
+
+ io_uring_remove_task_files(tctx);
}
static int io_uring_flush(struct file *file, void *data)
{
- io_uring_attempt_task_drop(file);
+ struct io_uring_task *tctx = current->io_uring;
+ struct io_ring_ctx *ctx = file->private_data;
+
+ if (!tctx)
+ return 0;
+
+ /* we should have cancelled and erased it before PF_EXITING */
+ WARN_ON_ONCE((current->flags & PF_EXITING) &&
+ xa_load(&tctx->xa, (unsigned long)file));
+
+ /*
+ * fput() is pending, will be 2 if the only other ref is our potential
+ * task file note. If the task is exiting, drop regardless of count.
+ */
+ if (atomic_long_read(&file->f_count) != 2)
+ return 0;
+
+ if (ctx->flags & IORING_SETUP_SQPOLL) {
+ /* there is only one file note, which is owned by sqo_task */
+ WARN_ON_ONCE((ctx->sqo_task == current) ==
+ !xa_load(&tctx->xa, (unsigned long)file));
+
+ io_disable_sqo_submit(ctx);
+ }
+
+ if (!(ctx->flags & IORING_SETUP_SQPOLL) || ctx->sqo_task == current)
+ io_uring_del_task_file(file);
return 0;
}
@@ -9106,8 +9174,9 @@ static unsigned long io_uring_nommu_get_unmapped_area(struct file *file,
#endif /* !CONFIG_MMU */
-static void io_sqpoll_wait_sq(struct io_ring_ctx *ctx)
+static int io_sqpoll_wait_sq(struct io_ring_ctx *ctx)
{
+ int ret = 0;
DEFINE_WAIT(wait);
do {
@@ -9116,6 +9185,11 @@ static void io_sqpoll_wait_sq(struct io_ring_ctx *ctx)
prepare_to_wait(&ctx->sqo_sq_wait, &wait, TASK_INTERRUPTIBLE);
+ if (unlikely(ctx->sqo_dead)) {
+ ret = -EOWNERDEAD;
+ goto out;
+ }
+
if (!io_sqring_full(ctx))
break;
@@ -9123,6 +9197,8 @@ static void io_sqpoll_wait_sq(struct io_ring_ctx *ctx)
} while (!signal_pending(current));
finish_wait(&ctx->sqo_sq_wait, &wait);
+out:
+ return ret;
}
static int io_get_ext_arg(unsigned flags, const void __user *argp, size_t *argsz,
@@ -9194,17 +9270,18 @@ SYSCALL_DEFINE6(io_uring_enter, unsigned int, fd, u32, to_submit,
*/
ret = 0;
if (ctx->flags & IORING_SETUP_SQPOLL) {
- if (!list_empty_careful(&ctx->cq_overflow_list)) {
- bool needs_lock = ctx->flags & IORING_SETUP_IOPOLL;
+ io_cqring_overflow_flush(ctx, false, NULL, NULL);
- io_ring_submit_lock(ctx, needs_lock);
- io_cqring_overflow_flush(ctx, false, NULL, NULL);
- io_ring_submit_unlock(ctx, needs_lock);
- }
+ ret = -EOWNERDEAD;
+ if (unlikely(ctx->sqo_dead))
+ goto out;
if (flags & IORING_ENTER_SQ_WAKEUP)
wake_up(&ctx->sq_data->wait);
- if (flags & IORING_ENTER_SQ_WAIT)
- io_sqpoll_wait_sq(ctx);
+ if (flags & IORING_ENTER_SQ_WAIT) {
+ ret = io_sqpoll_wait_sq(ctx);
+ if (ret)
+ goto out;
+ }
submitted = to_submit;
} else if (to_submit) {
ret = io_uring_add_task_file(ctx, f.file);
@@ -9631,6 +9708,7 @@ static int io_uring_create(unsigned entries, struct io_uring_params *p,
trace_io_uring_create(ret, ctx, p->sq_entries, p->cq_entries, p->flags);
return ret;
err:
+ io_disable_sqo_submit(ctx);
io_ring_ctx_wait_and_kill(ctx);
return ret;
}
diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c
index 816e1427f17e..04bf8066980c 100644
--- a/fs/nfs/delegation.c
+++ b/fs/nfs/delegation.c
@@ -1011,22 +1011,24 @@ nfs_delegation_find_inode_server(struct nfs_server *server,
const struct nfs_fh *fhandle)
{
struct nfs_delegation *delegation;
- struct inode *freeme, *res = NULL;
+ struct super_block *freeme = NULL;
+ struct inode *res = NULL;
list_for_each_entry_rcu(delegation, &server->delegations, super_list) {
spin_lock(&delegation->lock);
if (delegation->inode != NULL &&
!test_bit(NFS_DELEGATION_REVOKED, &delegation->flags) &&
nfs_compare_fh(fhandle, &NFS_I(delegation->inode)->fh) == 0) {
- freeme = igrab(delegation->inode);
- if (freeme && nfs_sb_active(freeme->i_sb))
- res = freeme;
+ if (nfs_sb_active(server->super)) {
+ freeme = server->super;
+ res = igrab(delegation->inode);
+ }
spin_unlock(&delegation->lock);
if (res != NULL)
return res;
if (freeme) {
rcu_read_unlock();
- iput(freeme);
+ nfs_sb_deactive(freeme);
rcu_read_lock();
}
return ERR_PTR(-EAGAIN);
diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h
index b840d0a91c9d..62d3189745cd 100644
--- a/fs/nfs/internal.h
+++ b/fs/nfs/internal.h
@@ -136,9 +136,29 @@ struct nfs_fs_context {
} clone_data;
};
-#define nfs_errorf(fc, fmt, ...) errorf(fc, fmt, ## __VA_ARGS__)
-#define nfs_invalf(fc, fmt, ...) invalf(fc, fmt, ## __VA_ARGS__)
-#define nfs_warnf(fc, fmt, ...) warnf(fc, fmt, ## __VA_ARGS__)
+#define nfs_errorf(fc, fmt, ...) ((fc)->log.log ? \
+ errorf(fc, fmt, ## __VA_ARGS__) : \
+ ({ dprintk(fmt "\n", ## __VA_ARGS__); }))
+
+#define nfs_ferrorf(fc, fac, fmt, ...) ((fc)->log.log ? \
+ errorf(fc, fmt, ## __VA_ARGS__) : \
+ ({ dfprintk(fac, fmt "\n", ## __VA_ARGS__); }))
+
+#define nfs_invalf(fc, fmt, ...) ((fc)->log.log ? \
+ invalf(fc, fmt, ## __VA_ARGS__) : \
+ ({ dprintk(fmt "\n", ## __VA_ARGS__); -EINVAL; }))
+
+#define nfs_finvalf(fc, fac, fmt, ...) ((fc)->log.log ? \
+ invalf(fc, fmt, ## __VA_ARGS__) : \
+ ({ dfprintk(fac, fmt "\n", ## __VA_ARGS__); -EINVAL; }))
+
+#define nfs_warnf(fc, fmt, ...) ((fc)->log.log ? \
+ warnf(fc, fmt, ## __VA_ARGS__) : \
+ ({ dprintk(fmt "\n", ## __VA_ARGS__); }))
+
+#define nfs_fwarnf(fc, fac, fmt, ...) ((fc)->log.log ? \
+ warnf(fc, fmt, ## __VA_ARGS__) : \
+ ({ dfprintk(fac, fmt "\n", ## __VA_ARGS__); }))
static inline struct nfs_fs_context *nfs_fc2context(const struct fs_context *fc)
{
@@ -579,12 +599,14 @@ extern void nfs4_test_session_trunk(struct rpc_clnt *clnt,
static inline struct inode *nfs_igrab_and_active(struct inode *inode)
{
- inode = igrab(inode);
- if (inode != NULL && !nfs_sb_active(inode->i_sb)) {
- iput(inode);
- inode = NULL;
+ struct super_block *sb = inode->i_sb;
+
+ if (sb && nfs_sb_active(sb)) {
+ if (igrab(inode))
+ return inode;
+ nfs_sb_deactive(sb);
}
- return inode;
+ return NULL;
}
static inline void nfs_iput_and_deactive(struct inode *inode)
diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
index 0ce04e0e5d82..2f4679a62712 100644
--- a/fs/nfs/nfs4proc.c
+++ b/fs/nfs/nfs4proc.c
@@ -3536,10 +3536,8 @@ static void nfs4_close_done(struct rpc_task *task, void *data)
trace_nfs4_close(state, &calldata->arg, &calldata->res, task->tk_status);
/* Handle Layoutreturn errors */
- if (pnfs_roc_done(task, calldata->inode,
- &calldata->arg.lr_args,
- &calldata->res.lr_res,
- &calldata->res.lr_ret) == -EAGAIN)
+ if (pnfs_roc_done(task, &calldata->arg.lr_args, &calldata->res.lr_res,
+ &calldata->res.lr_ret) == -EAGAIN)
goto out_restart;
/* hmm. we are done with the inode, and in the process of freeing
@@ -6384,10 +6382,8 @@ static void nfs4_delegreturn_done(struct rpc_task *task, void *calldata)
trace_nfs4_delegreturn_exit(&data->args, &data->res, task->tk_status);
/* Handle Layoutreturn errors */
- if (pnfs_roc_done(task, data->inode,
- &data->args.lr_args,
- &data->res.lr_res,
- &data->res.lr_ret) == -EAGAIN)
+ if (pnfs_roc_done(task, &data->args.lr_args, &data->res.lr_res,
+ &data->res.lr_ret) == -EAGAIN)
goto out_restart;
switch (task->tk_status) {
@@ -6441,10 +6437,10 @@ static void nfs4_delegreturn_release(void *calldata)
struct nfs4_delegreturndata *data = calldata;
struct inode *inode = data->inode;
+ if (data->lr.roc)
+ pnfs_roc_release(&data->lr.arg, &data->lr.res,
+ data->res.lr_ret);
if (inode) {
- if (data->lr.roc)
- pnfs_roc_release(&data->lr.arg, &data->lr.res,
- data->res.lr_ret);
nfs_post_op_update_inode_force_wcc(inode, &data->fattr);
nfs_iput_and_deactive(inode);
}
@@ -6520,16 +6516,14 @@ static int _nfs4_proc_delegreturn(struct inode *inode, const struct cred *cred,
nfs_fattr_init(data->res.fattr);
data->timestamp = jiffies;
data->rpc_status = 0;
- data->lr.roc = pnfs_roc(inode, &data->lr.arg, &data->lr.res, cred);
data->inode = nfs_igrab_and_active(inode);
- if (data->inode) {
+ if (data->inode || issync) {
+ data->lr.roc = pnfs_roc(inode, &data->lr.arg, &data->lr.res,
+ cred);
if (data->lr.roc) {
data->args.lr_args = &data->lr.arg;
data->res.lr_res = &data->lr.res;
}
- } else if (data->lr.roc) {
- pnfs_roc_release(&data->lr.arg, &data->lr.res, 0);
- data->lr.roc = false;
}
task_setup_data.callback_data = data;
@@ -7111,9 +7105,9 @@ static int _nfs4_do_setlk(struct nfs4_state *state, int cmd, struct file_lock *f
data->arg.new_lock_owner, ret);
} else
data->cancelled = true;
+ trace_nfs4_set_lock(fl, state, &data->res.stateid, cmd, ret);
rpc_put_task(task);
dprintk("%s: done, ret = %d!\n", __func__, ret);
- trace_nfs4_set_lock(fl, state, &data->res.stateid, cmd, ret);
return ret;
}
diff --git a/fs/nfs/nfs4super.c b/fs/nfs/nfs4super.c
index 984cc42ee54d..d09bcfd7db89 100644
--- a/fs/nfs/nfs4super.c
+++ b/fs/nfs/nfs4super.c
@@ -227,7 +227,7 @@ int nfs4_try_get_tree(struct fs_context *fc)
fc, ctx->nfs_server.hostname,
ctx->nfs_server.export_path);
if (err) {
- nfs_errorf(fc, "NFS4: Couldn't follow remote path");
+ nfs_ferrorf(fc, MOUNT, "NFS4: Couldn't follow remote path");
dfprintk(MOUNT, "<-- nfs4_try_get_tree() = %d [error]\n", err);
} else {
dfprintk(MOUNT, "<-- nfs4_try_get_tree() = 0\n");
@@ -250,7 +250,7 @@ int nfs4_get_referral_tree(struct fs_context *fc)
fc, ctx->nfs_server.hostname,
ctx->nfs_server.export_path);
if (err) {
- nfs_errorf(fc, "NFS4: Couldn't follow remote path");
+ nfs_ferrorf(fc, MOUNT, "NFS4: Couldn't follow remote path");
dfprintk(MOUNT, "<-- nfs4_get_referral_tree() = %d [error]\n", err);
} else {
dfprintk(MOUNT, "<-- nfs4_get_referral_tree() = 0\n");
diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c
index 07f59dc8cb2e..4f274f21c4ab 100644
--- a/fs/nfs/pnfs.c
+++ b/fs/nfs/pnfs.c
@@ -1152,7 +1152,7 @@ void pnfs_layoutreturn_free_lsegs(struct pnfs_layout_hdr *lo,
LIST_HEAD(freeme);
spin_lock(&inode->i_lock);
- if (!pnfs_layout_is_valid(lo) || !arg_stateid ||
+ if (!pnfs_layout_is_valid(lo) ||
!nfs4_stateid_match_other(&lo->plh_stateid, arg_stateid))
goto out_unlock;
if (stateid) {
@@ -1509,10 +1509,8 @@ out_noroc:
return false;
}
-int pnfs_roc_done(struct rpc_task *task, struct inode *inode,
- struct nfs4_layoutreturn_args **argpp,
- struct nfs4_layoutreturn_res **respp,
- int *ret)
+int pnfs_roc_done(struct rpc_task *task, struct nfs4_layoutreturn_args **argpp,
+ struct nfs4_layoutreturn_res **respp, int *ret)
{
struct nfs4_layoutreturn_args *arg = *argpp;
int retval = -EAGAIN;
@@ -1545,7 +1543,7 @@ int pnfs_roc_done(struct rpc_task *task, struct inode *inode,
return 0;
case -NFS4ERR_OLD_STATEID:
if (!nfs4_layout_refresh_old_stateid(&arg->stateid,
- &arg->range, inode))
+ &arg->range, arg->inode))
break;
*ret = -NFS4ERR_NOMATCHING_LAYOUT;
return -EAGAIN;
@@ -1560,23 +1558,28 @@ void pnfs_roc_release(struct nfs4_layoutreturn_args *args,
int ret)
{
struct pnfs_layout_hdr *lo = args->layout;
- const nfs4_stateid *arg_stateid = NULL;
+ struct inode *inode = args->inode;
const nfs4_stateid *res_stateid = NULL;
struct nfs4_xdr_opaque_data *ld_private = args->ld_private;
switch (ret) {
case -NFS4ERR_NOMATCHING_LAYOUT:
+ spin_lock(&inode->i_lock);
+ if (pnfs_layout_is_valid(lo) &&
+ nfs4_stateid_match_other(&args->stateid, &lo->plh_stateid))
+ pnfs_set_plh_return_info(lo, args->range.iomode, 0);
+ pnfs_clear_layoutreturn_waitbit(lo);
+ spin_unlock(&inode->i_lock);
break;
case 0:
if (res->lrs_present)
res_stateid = &res->stateid;
fallthrough;
default:
- arg_stateid = &args->stateid;
+ pnfs_layoutreturn_free_lsegs(lo, &args->stateid, &args->range,
+ res_stateid);
}
trace_nfs4_layoutreturn_on_close(args->inode, &args->stateid, ret);
- pnfs_layoutreturn_free_lsegs(lo, arg_stateid, &args->range,
- res_stateid);
if (ld_private && ld_private->ops && ld_private->ops->free)
ld_private->ops->free(ld_private);
pnfs_put_layout_hdr(lo);
@@ -2015,6 +2018,27 @@ lookup_again:
goto lookup_again;
}
+ /*
+ * Because we free lsegs when sending LAYOUTRETURN, we need to wait
+ * for LAYOUTRETURN.
+ */
+ if (test_bit(NFS_LAYOUT_RETURN, &lo->plh_flags)) {
+ spin_unlock(&ino->i_lock);
+ dprintk("%s wait for layoutreturn\n", __func__);
+ lseg = ERR_PTR(pnfs_prepare_to_retry_layoutget(lo));
+ if (!IS_ERR(lseg)) {
+ pnfs_put_layout_hdr(lo);
+ dprintk("%s retrying\n", __func__);
+ trace_pnfs_update_layout(ino, pos, count, iomode, lo,
+ lseg,
+ PNFS_UPDATE_LAYOUT_RETRY);
+ goto lookup_again;
+ }
+ trace_pnfs_update_layout(ino, pos, count, iomode, lo, lseg,
+ PNFS_UPDATE_LAYOUT_RETURN);
+ goto out_put_layout_hdr;
+ }
+
lseg = pnfs_find_lseg(lo, &arg, strict_iomode);
if (lseg) {
trace_pnfs_update_layout(ino, pos, count, iomode, lo, lseg,
@@ -2067,28 +2091,6 @@ lookup_again:
nfs4_stateid_copy(&stateid, &lo->plh_stateid);
}
- /*
- * Because we free lsegs before sending LAYOUTRETURN, we need to wait
- * for LAYOUTRETURN even if first is true.
- */
- if (test_bit(NFS_LAYOUT_RETURN, &lo->plh_flags)) {
- spin_unlock(&ino->i_lock);
- dprintk("%s wait for layoutreturn\n", __func__);
- lseg = ERR_PTR(pnfs_prepare_to_retry_layoutget(lo));
- if (!IS_ERR(lseg)) {
- if (first)
- pnfs_clear_first_layoutget(lo);
- pnfs_put_layout_hdr(lo);
- dprintk("%s retrying\n", __func__);
- trace_pnfs_update_layout(ino, pos, count, iomode, lo,
- lseg, PNFS_UPDATE_LAYOUT_RETRY);
- goto lookup_again;
- }
- trace_pnfs_update_layout(ino, pos, count, iomode, lo, lseg,
- PNFS_UPDATE_LAYOUT_RETURN);
- goto out_put_layout_hdr;
- }
-
if (pnfs_layoutgets_blocked(lo)) {
trace_pnfs_update_layout(ino, pos, count, iomode, lo, lseg,
PNFS_UPDATE_LAYOUT_BLOCKED);
@@ -2242,6 +2244,7 @@ static void _lgopen_prepare_attached(struct nfs4_opendata *data,
&rng, GFP_KERNEL);
if (!lgp) {
pnfs_clear_first_layoutget(lo);
+ nfs_layoutget_end(lo);
pnfs_put_layout_hdr(lo);
return;
}
diff --git a/fs/nfs/pnfs.h b/fs/nfs/pnfs.h
index bbd3de1025f2..d810ae674f4e 100644
--- a/fs/nfs/pnfs.h
+++ b/fs/nfs/pnfs.h
@@ -297,10 +297,8 @@ bool pnfs_roc(struct inode *ino,
struct nfs4_layoutreturn_args *args,
struct nfs4_layoutreturn_res *res,
const struct cred *cred);
-int pnfs_roc_done(struct rpc_task *task, struct inode *inode,
- struct nfs4_layoutreturn_args **argpp,
- struct nfs4_layoutreturn_res **respp,
- int *ret);
+int pnfs_roc_done(struct rpc_task *task, struct nfs4_layoutreturn_args **argpp,
+ struct nfs4_layoutreturn_res **respp, int *ret);
void pnfs_roc_release(struct nfs4_layoutreturn_args *args,
struct nfs4_layoutreturn_res *res,
int ret);
@@ -772,7 +770,7 @@ pnfs_roc(struct inode *ino,
}
static inline int
-pnfs_roc_done(struct rpc_task *task, struct inode *inode,
+pnfs_roc_done(struct rpc_task *task,
struct nfs4_layoutreturn_args **argpp,
struct nfs4_layoutreturn_res **respp,
int *ret)
diff --git a/fs/nfs/pnfs_nfs.c b/fs/nfs/pnfs_nfs.c
index 2efcfdd348a1..49d3389bd813 100644
--- a/fs/nfs/pnfs_nfs.c
+++ b/fs/nfs/pnfs_nfs.c
@@ -78,22 +78,18 @@ void
pnfs_generic_clear_request_commit(struct nfs_page *req,
struct nfs_commit_info *cinfo)
{
- struct pnfs_layout_segment *freeme = NULL;
+ struct pnfs_commit_bucket *bucket = NULL;
if (!test_and_clear_bit(PG_COMMIT_TO_DS, &req->wb_flags))
goto out;
cinfo->ds->nwritten--;
- if (list_is_singular(&req->wb_list)) {
- struct pnfs_commit_bucket *bucket;
-
+ if (list_is_singular(&req->wb_list))
bucket = list_first_entry(&req->wb_list,
- struct pnfs_commit_bucket,
- written);
- freeme = pnfs_free_bucket_lseg(bucket);
- }
+ struct pnfs_commit_bucket, written);
out:
nfs_request_remove_commit_list(req, cinfo);
- pnfs_put_lseg(freeme);
+ if (bucket)
+ pnfs_put_lseg(pnfs_free_bucket_lseg(bucket));
}
EXPORT_SYMBOL_GPL(pnfs_generic_clear_request_commit);
@@ -407,12 +403,16 @@ pnfs_bucket_get_committing(struct list_head *head,
struct pnfs_commit_bucket *bucket,
struct nfs_commit_info *cinfo)
{
+ struct pnfs_layout_segment *lseg;
struct list_head *pos;
list_for_each(pos, &bucket->committing)
cinfo->ds->ncommitting--;
list_splice_init(&bucket->committing, head);
- return pnfs_free_bucket_lseg(bucket);
+ lseg = pnfs_free_bucket_lseg(bucket);
+ if (!lseg)
+ lseg = pnfs_get_lseg(bucket->lseg);
+ return lseg;
}
static struct nfs_commit_data *
@@ -424,8 +424,6 @@ pnfs_bucket_fetch_commitdata(struct pnfs_commit_bucket *bucket,
if (!data)
return NULL;
data->lseg = pnfs_bucket_get_committing(&data->pages, bucket, cinfo);
- if (!data->lseg)
- data->lseg = pnfs_get_lseg(bucket->lseg);
return data;
}
diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c
index 4727b7f03c5b..8d6d2678abad 100644
--- a/fs/nfsd/nfs4proc.c
+++ b/fs/nfsd/nfs4proc.c
@@ -50,6 +50,11 @@
#include "pnfs.h"
#include "trace.h"
+static bool inter_copy_offload_enable;
+module_param(inter_copy_offload_enable, bool, 0644);
+MODULE_PARM_DESC(inter_copy_offload_enable,
+ "Enable inter server to server copy offload. Default: false");
+
#ifdef CONFIG_NFSD_V4_SECURITY_LABEL
#include <linux/security.h>
diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c
index 45ee6b12ce5b..eaaa1605b5b5 100644
--- a/fs/nfsd/nfs4xdr.c
+++ b/fs/nfsd/nfs4xdr.c
@@ -147,6 +147,25 @@ svcxdr_dupstr(struct nfsd4_compoundargs *argp, void *buf, u32 len)
return p;
}
+static void *
+svcxdr_savemem(struct nfsd4_compoundargs *argp, __be32 *p, u32 len)
+{
+ __be32 *tmp;
+
+ /*
+ * The location of the decoded data item is stable,
+ * so @p is OK to use. This is the common case.
+ */
+ if (p != argp->xdr->scratch.iov_base)
+ return p;
+
+ tmp = svcxdr_tmpalloc(argp, len);
+ if (!tmp)
+ return NULL;
+ memcpy(tmp, p, len);
+ return tmp;
+}
+
/*
* NFSv4 basic data type decoders
*/
@@ -183,11 +202,10 @@ nfsd4_decode_opaque(struct nfsd4_compoundargs *argp, struct xdr_netobj *o)
p = xdr_inline_decode(argp->xdr, len);
if (!p)
return nfserr_bad_xdr;
- o->data = svcxdr_tmpalloc(argp, len);
+ o->data = svcxdr_savemem(argp, p, len);
if (!o->data)
return nfserr_jukebox;
o->len = len;
- memcpy(o->data, p, len);
return nfs_ok;
}
@@ -205,10 +223,9 @@ nfsd4_decode_component4(struct nfsd4_compoundargs *argp, char **namp, u32 *lenp)
status = check_filename((char *)p, *lenp);
if (status)
return status;
- *namp = svcxdr_tmpalloc(argp, *lenp);
+ *namp = svcxdr_savemem(argp, p, *lenp);
if (!*namp)
return nfserr_jukebox;
- memcpy(*namp, p, *lenp);
return nfs_ok;
}
@@ -1200,10 +1217,9 @@ nfsd4_decode_putfh(struct nfsd4_compoundargs *argp, struct nfsd4_putfh *putfh)
p = xdr_inline_decode(argp->xdr, putfh->pf_fhlen);
if (!p)
return nfserr_bad_xdr;
- putfh->pf_fhval = svcxdr_tmpalloc(argp, putfh->pf_fhlen);
+ putfh->pf_fhval = svcxdr_savemem(argp, p, putfh->pf_fhlen);
if (!putfh->pf_fhval)
return nfserr_jukebox;
- memcpy(putfh->pf_fhval, p, putfh->pf_fhlen);
return nfs_ok;
}
@@ -1318,24 +1334,20 @@ nfsd4_decode_setclientid(struct nfsd4_compoundargs *argp, struct nfsd4_setclient
p = xdr_inline_decode(argp->xdr, setclientid->se_callback_netid_len);
if (!p)
return nfserr_bad_xdr;
- setclientid->se_callback_netid_val = svcxdr_tmpalloc(argp,
+ setclientid->se_callback_netid_val = svcxdr_savemem(argp, p,
setclientid->se_callback_netid_len);
if (!setclientid->se_callback_netid_val)
return nfserr_jukebox;
- memcpy(setclientid->se_callback_netid_val, p,
- setclientid->se_callback_netid_len);
if (xdr_stream_decode_u32(argp->xdr, &setclientid->se_callback_addr_len) < 0)
return nfserr_bad_xdr;
p = xdr_inline_decode(argp->xdr, setclientid->se_callback_addr_len);
if (!p)
return nfserr_bad_xdr;
- setclientid->se_callback_addr_val = svcxdr_tmpalloc(argp,
+ setclientid->se_callback_addr_val = svcxdr_savemem(argp, p,
setclientid->se_callback_addr_len);
if (!setclientid->se_callback_addr_val)
return nfserr_jukebox;
- memcpy(setclientid->se_callback_addr_val, p,
- setclientid->se_callback_addr_len);
if (xdr_stream_decode_u32(argp->xdr, &setclientid->se_callback_ident) < 0)
return nfserr_bad_xdr;
@@ -1375,10 +1387,9 @@ nfsd4_decode_verify(struct nfsd4_compoundargs *argp, struct nfsd4_verify *verify
p = xdr_inline_decode(argp->xdr, verify->ve_attrlen);
if (!p)
return nfserr_bad_xdr;
- verify->ve_attrval = svcxdr_tmpalloc(argp, verify->ve_attrlen);
+ verify->ve_attrval = svcxdr_savemem(argp, p, verify->ve_attrlen);
if (!verify->ve_attrval)
return nfserr_jukebox;
- memcpy(verify->ve_attrval, p, verify->ve_attrlen);
return nfs_ok;
}
@@ -2333,10 +2344,9 @@ nfsd4_decode_compound(struct nfsd4_compoundargs *argp)
p = xdr_inline_decode(argp->xdr, argp->taglen);
if (!p)
return 0;
- argp->tag = svcxdr_tmpalloc(argp, argp->taglen);
+ argp->tag = svcxdr_savemem(argp, p, argp->taglen);
if (!argp->tag)
return 0;
- memcpy(argp->tag, p, argp->taglen);
max_reply += xdr_align_size(argp->taglen);
}
@@ -4756,6 +4766,7 @@ nfsd4_encode_read_plus_data(struct nfsd4_compoundres *resp,
resp->rqstp->rq_vec, read->rd_vlen, maxcount, eof);
if (nfserr)
return nfserr;
+ xdr_truncate_encode(xdr, starting_len + 16 + xdr_align_size(*maxcount));
tmp = htonl(NFS4_CONTENT_DATA);
write_bytes_to_xdr_buf(xdr->buf, starting_len, &tmp, 4);
@@ -4763,6 +4774,10 @@ nfsd4_encode_read_plus_data(struct nfsd4_compoundres *resp,
write_bytes_to_xdr_buf(xdr->buf, starting_len + 4, &tmp64, 8);
tmp = htonl(*maxcount);
write_bytes_to_xdr_buf(xdr->buf, starting_len + 12, &tmp, 4);
+
+ tmp = xdr_zero;
+ write_bytes_to_xdr_buf(xdr->buf, starting_len + 16 + *maxcount, &tmp,
+ xdr_pad_size(*maxcount));
return nfs_ok;
}
@@ -4855,14 +4870,15 @@ out:
if (nfserr && segments == 0)
xdr_truncate_encode(xdr, starting_len);
else {
- tmp = htonl(eof);
- write_bytes_to_xdr_buf(xdr->buf, starting_len, &tmp, 4);
- tmp = htonl(segments);
- write_bytes_to_xdr_buf(xdr->buf, starting_len + 4, &tmp, 4);
if (nfserr) {
xdr_truncate_encode(xdr, last_segment);
nfserr = nfs_ok;
+ eof = 0;
}
+ tmp = htonl(eof);
+ write_bytes_to_xdr_buf(xdr->buf, starting_len, &tmp, 4);
+ tmp = htonl(segments);
+ write_bytes_to_xdr_buf(xdr->buf, starting_len + 4, &tmp, 4);
}
return nfserr;
diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c
index 00384c332f9b..f9c9f4c63cc7 100644
--- a/fs/nfsd/nfssvc.c
+++ b/fs/nfsd/nfssvc.c
@@ -33,12 +33,6 @@
#define NFSDDBG_FACILITY NFSDDBG_SVC
-bool inter_copy_offload_enable;
-EXPORT_SYMBOL_GPL(inter_copy_offload_enable);
-module_param(inter_copy_offload_enable, bool, 0644);
-MODULE_PARM_DESC(inter_copy_offload_enable,
- "Enable inter server to server copy offload. Default: false");
-
extern struct svc_program nfsd_program;
static int nfsd(void *vrqstp);
#if defined(CONFIG_NFSD_V2_ACL) || defined(CONFIG_NFSD_V3_ACL)
diff --git a/fs/nfsd/xdr4.h b/fs/nfsd/xdr4.h
index a60ff5ce1a37..c300885ae75d 100644
--- a/fs/nfsd/xdr4.h
+++ b/fs/nfsd/xdr4.h
@@ -568,7 +568,6 @@ struct nfsd4_copy {
struct nfs_fh c_fh;
nfs4_stateid stateid;
};
-extern bool inter_copy_offload_enable;
struct nfsd4_seek {
/* request */
diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c
index 3e01d8f2ab90..dcab112e1f00 100644
--- a/fs/notify/fanotify/fanotify_user.c
+++ b/fs/notify/fanotify/fanotify_user.c
@@ -1285,26 +1285,23 @@ fput_and_out:
return ret;
}
+#ifndef CONFIG_ARCH_SPLIT_ARG64
SYSCALL_DEFINE5(fanotify_mark, int, fanotify_fd, unsigned int, flags,
__u64, mask, int, dfd,
const char __user *, pathname)
{
return do_fanotify_mark(fanotify_fd, flags, mask, dfd, pathname);
}
+#endif
-#ifdef CONFIG_COMPAT
-COMPAT_SYSCALL_DEFINE6(fanotify_mark,
+#if defined(CONFIG_ARCH_SPLIT_ARG64) || defined(CONFIG_COMPAT)
+SYSCALL32_DEFINE6(fanotify_mark,
int, fanotify_fd, unsigned int, flags,
- __u32, mask0, __u32, mask1, int, dfd,
+ SC_ARG64(mask), int, dfd,
const char __user *, pathname)
{
- return do_fanotify_mark(fanotify_fd, flags,
-#ifdef __BIG_ENDIAN
- ((__u64)mask0 << 32) | mask1,
-#else
- ((__u64)mask1 << 32) | mask0,
-#endif
- dfd, pathname);
+ return do_fanotify_mark(fanotify_fd, flags, SC_VAL64(__u64, mask),
+ dfd, pathname);
}
#endif
diff --git a/fs/zonefs/Kconfig b/fs/zonefs/Kconfig
index ef2697b78820..827278f937fe 100644
--- a/fs/zonefs/Kconfig
+++ b/fs/zonefs/Kconfig
@@ -3,6 +3,7 @@ config ZONEFS_FS
depends on BLOCK
depends on BLK_DEV_ZONED
select FS_IOMAP
+ select CRC32
help
zonefs is a simple file system which exposes zones of a zoned block
device (e.g. host-managed or host-aware SMR disk drives) as files.