summaryrefslogtreecommitdiff
path: root/fs
diff options
context:
space:
mode:
Diffstat (limited to 'fs')
-rw-r--r--fs/btrfs/inode.c10
-rw-r--r--fs/cifs/cifsglob.h5
-rw-r--r--fs/cifs/dir.c2
-rw-r--r--fs/cifs/file.c35
-rw-r--r--fs/cifs/inode.c19
-rw-r--r--fs/cifs/misc.c50
-rw-r--r--fs/cifs/smb2pdu.c2
-rw-r--r--fs/configfs/file.c18
-rw-r--r--fs/dax.c2
-rw-r--r--fs/fuse/dax.c6
-rw-r--r--fs/io-wq.c26
-rw-r--r--fs/io_uring.c58
-rw-r--r--fs/pipe.c15
13 files changed, 154 insertions, 94 deletions
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c
index 0117d867ecf8..06f9f167222b 100644
--- a/fs/btrfs/inode.c
+++ b/fs/btrfs/inode.c
@@ -9226,8 +9226,14 @@ static int btrfs_rename_exchange(struct inode *old_dir,
bool dest_log_pinned = false;
bool need_abort = false;
- /* we only allow rename subvolume link between subvolumes */
- if (old_ino != BTRFS_FIRST_FREE_OBJECTID && root != dest)
+ /*
+ * For non-subvolumes allow exchange only within one subvolume, in the
+ * same inode namespace. Two subvolumes (represented as directory) can
+ * be exchanged as they're a logical link and have a fixed inode number.
+ */
+ if (root != dest &&
+ (old_ino != BTRFS_FIRST_FREE_OBJECTID ||
+ new_ino != BTRFS_FIRST_FREE_OBJECTID))
return -EXDEV;
/* close the race window with snapshot create/destroy ioctl */
diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h
index c0bfc2f01030..c6a9542ca281 100644
--- a/fs/cifs/cifsglob.h
+++ b/fs/cifs/cifsglob.h
@@ -1611,6 +1611,11 @@ struct dfs_info3_param {
int ttl;
};
+struct file_list {
+ struct list_head list;
+ struct cifsFileInfo *cfile;
+};
+
/*
* common struct for holding inode info when searching for or updating an
* inode with new info
diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c
index 79402ca0ddfa..5f8a302ffcb2 100644
--- a/fs/cifs/dir.c
+++ b/fs/cifs/dir.c
@@ -100,7 +100,7 @@ build_path_from_dentry_optional_prefix(struct dentry *direntry, void *page,
if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_USE_PREFIX_PATH)
pplen = cifs_sb->prepath ? strlen(cifs_sb->prepath) + 1 : 0;
- s = dentry_path_raw(direntry, page, PAGE_SIZE);
+ s = dentry_path_raw(direntry, page, PATH_MAX);
if (IS_ERR(s))
return s;
if (!s[1]) // for root we want "", not "/"
diff --git a/fs/cifs/file.c b/fs/cifs/file.c
index 0a72840a88f1..bb98fbdd22a9 100644
--- a/fs/cifs/file.c
+++ b/fs/cifs/file.c
@@ -4848,34 +4848,33 @@ void cifs_oplock_break(struct work_struct *work)
oplock_break_ack:
/*
- * releasing stale oplock after recent reconnect of smb session using
- * a now incorrect file handle is not a data integrity issue but do
- * not bother sending an oplock release if session to server still is
- * disconnected since oplock already released by the server
- */
- if (!cfile->oplock_break_cancelled) {
- rc = tcon->ses->server->ops->oplock_response(tcon, &cfile->fid,
- cinode);
- cifs_dbg(FYI, "Oplock release rc = %d\n", rc);
- }
- /*
* When oplock break is received and there are no active
* file handles but cached, then schedule deferred close immediately.
* So, new open will not use cached handle.
*/
spin_lock(&CIFS_I(inode)->deferred_lock);
is_deferred = cifs_is_deferred_close(cfile, &dclose);
+ spin_unlock(&CIFS_I(inode)->deferred_lock);
if (is_deferred &&
cfile->deferred_close_scheduled &&
delayed_work_pending(&cfile->deferred)) {
- /*
- * If there is no pending work, mod_delayed_work queues new work.
- * So, Increase the ref count to avoid use-after-free.
- */
- if (!mod_delayed_work(deferredclose_wq, &cfile->deferred, 0))
- cifsFileInfo_get(cfile);
+ if (cancel_delayed_work(&cfile->deferred)) {
+ _cifsFileInfo_put(cfile, false, false);
+ goto oplock_break_done;
+ }
}
- spin_unlock(&CIFS_I(inode)->deferred_lock);
+ /*
+ * releasing stale oplock after recent reconnect of smb session using
+ * a now incorrect file handle is not a data integrity issue but do
+ * not bother sending an oplock release if session to server still is
+ * disconnected since oplock already released by the server
+ */
+ if (!cfile->oplock_break_cancelled) {
+ rc = tcon->ses->server->ops->oplock_response(tcon, &cfile->fid,
+ cinode);
+ cifs_dbg(FYI, "Oplock release rc = %d\n", rc);
+ }
+oplock_break_done:
_cifsFileInfo_put(cfile, false /* do not wait for ourself */, false);
cifs_done_oplock_break(cinode);
}
diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c
index b96b253e7635..65f8a70cece3 100644
--- a/fs/cifs/inode.c
+++ b/fs/cifs/inode.c
@@ -1625,7 +1625,7 @@ int cifs_unlink(struct inode *dir, struct dentry *dentry)
goto unlink_out;
}
- cifs_close_all_deferred_files(tcon);
+ cifs_close_deferred_file(CIFS_I(inode));
if (cap_unix(tcon->ses) && (CIFS_UNIX_POSIX_PATH_OPS_CAP &
le64_to_cpu(tcon->fsUnixInfo.Capability))) {
rc = CIFSPOSIXDelFile(xid, tcon, full_path,
@@ -2084,6 +2084,7 @@ cifs_rename2(struct user_namespace *mnt_userns, struct inode *source_dir,
FILE_UNIX_BASIC_INFO *info_buf_target;
unsigned int xid;
int rc, tmprc;
+ int retry_count = 0;
if (flags & ~RENAME_NOREPLACE)
return -EINVAL;
@@ -2113,10 +2114,24 @@ cifs_rename2(struct user_namespace *mnt_userns, struct inode *source_dir,
goto cifs_rename_exit;
}
- cifs_close_all_deferred_files(tcon);
+ cifs_close_deferred_file(CIFS_I(d_inode(source_dentry)));
+ if (d_inode(target_dentry) != NULL)
+ cifs_close_deferred_file(CIFS_I(d_inode(target_dentry)));
+
rc = cifs_do_rename(xid, source_dentry, from_name, target_dentry,
to_name);
+ if (rc == -EACCES) {
+ while (retry_count < 3) {
+ cifs_close_all_deferred_files(tcon);
+ rc = cifs_do_rename(xid, source_dentry, from_name, target_dentry,
+ to_name);
+ if (rc != -EACCES)
+ break;
+ retry_count++;
+ }
+ }
+
/*
* No-replace is the natural behavior for CIFS, so skip unlink hacks.
*/
diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c
index 844abeb2b48f..9469f1cf0b46 100644
--- a/fs/cifs/misc.c
+++ b/fs/cifs/misc.c
@@ -723,13 +723,31 @@ void
cifs_close_deferred_file(struct cifsInodeInfo *cifs_inode)
{
struct cifsFileInfo *cfile = NULL;
- struct cifs_deferred_close *dclose;
+ struct file_list *tmp_list, *tmp_next_list;
+ struct list_head file_head;
+
+ if (cifs_inode == NULL)
+ return;
+ INIT_LIST_HEAD(&file_head);
+ spin_lock(&cifs_inode->open_file_lock);
list_for_each_entry(cfile, &cifs_inode->openFileList, flist) {
- spin_lock(&cifs_inode->deferred_lock);
- if (cifs_is_deferred_close(cfile, &dclose))
- mod_delayed_work(deferredclose_wq, &cfile->deferred, 0);
- spin_unlock(&cifs_inode->deferred_lock);
+ if (delayed_work_pending(&cfile->deferred)) {
+ if (cancel_delayed_work(&cfile->deferred)) {
+ tmp_list = kmalloc(sizeof(struct file_list), GFP_ATOMIC);
+ if (tmp_list == NULL)
+ continue;
+ tmp_list->cfile = cfile;
+ list_add_tail(&tmp_list->list, &file_head);
+ }
+ }
+ }
+ spin_unlock(&cifs_inode->open_file_lock);
+
+ list_for_each_entry_safe(tmp_list, tmp_next_list, &file_head, list) {
+ _cifsFileInfo_put(tmp_list->cfile, true, false);
+ list_del(&tmp_list->list);
+ kfree(tmp_list);
}
}
@@ -738,20 +756,30 @@ cifs_close_all_deferred_files(struct cifs_tcon *tcon)
{
struct cifsFileInfo *cfile;
struct list_head *tmp;
+ struct file_list *tmp_list, *tmp_next_list;
+ struct list_head file_head;
+ INIT_LIST_HEAD(&file_head);
spin_lock(&tcon->open_file_lock);
list_for_each(tmp, &tcon->openFileList) {
cfile = list_entry(tmp, struct cifsFileInfo, tlist);
if (delayed_work_pending(&cfile->deferred)) {
- /*
- * If there is no pending work, mod_delayed_work queues new work.
- * So, Increase the ref count to avoid use-after-free.
- */
- if (!mod_delayed_work(deferredclose_wq, &cfile->deferred, 0))
- cifsFileInfo_get(cfile);
+ if (cancel_delayed_work(&cfile->deferred)) {
+ tmp_list = kmalloc(sizeof(struct file_list), GFP_ATOMIC);
+ if (tmp_list == NULL)
+ continue;
+ tmp_list->cfile = cfile;
+ list_add_tail(&tmp_list->list, &file_head);
+ }
}
}
spin_unlock(&tcon->open_file_lock);
+
+ list_for_each_entry_safe(tmp_list, tmp_next_list, &file_head, list) {
+ _cifsFileInfo_put(tmp_list->cfile, true, false);
+ list_del(&tmp_list->list);
+ kfree(tmp_list);
+ }
}
/* parses DFS refferal V3 structure
diff --git a/fs/cifs/smb2pdu.c b/fs/cifs/smb2pdu.c
index 781d14e5f2af..b6d2e3591927 100644
--- a/fs/cifs/smb2pdu.c
+++ b/fs/cifs/smb2pdu.c
@@ -2426,7 +2426,7 @@ create_sd_buf(umode_t mode, bool set_owner, unsigned int *len)
memcpy(aclptr, &acl, sizeof(struct cifs_acl));
buf->ccontext.DataLength = cpu_to_le32(ptr - (__u8 *)&buf->sd);
- *len = ptr - (__u8 *)buf;
+ *len = roundup(ptr - (__u8 *)buf, 8);
return buf;
}
diff --git a/fs/configfs/file.c b/fs/configfs/file.c
index 5a0be9985bae..0ad32150611e 100644
--- a/fs/configfs/file.c
+++ b/fs/configfs/file.c
@@ -177,28 +177,22 @@ out:
return retval;
}
-/* Fill [buffer, buffer + pos) with data coming from @from. */
-static int fill_write_buffer(struct configfs_buffer *buffer, loff_t pos,
+/* Fill @buffer with data coming from @from. */
+static int fill_write_buffer(struct configfs_buffer *buffer,
struct iov_iter *from)
{
- loff_t to_copy;
int copied;
- u8 *to;
if (!buffer->page)
buffer->page = (char *)__get_free_pages(GFP_KERNEL, 0);
if (!buffer->page)
return -ENOMEM;
- to_copy = SIMPLE_ATTR_SIZE - 1 - pos;
- if (to_copy <= 0)
- return 0;
- to = buffer->page + pos;
- copied = copy_from_iter(to, to_copy, from);
+ copied = copy_from_iter(buffer->page, SIMPLE_ATTR_SIZE - 1, from);
buffer->needs_read_fill = 1;
/* if buf is assumed to contain a string, terminate it by \0,
* so e.g. sscanf() can scan the string easily */
- to[copied] = 0;
+ buffer->page[copied] = 0;
return copied ? : -EFAULT;
}
@@ -227,10 +221,10 @@ static ssize_t configfs_write_iter(struct kiocb *iocb, struct iov_iter *from)
{
struct file *file = iocb->ki_filp;
struct configfs_buffer *buffer = file->private_data;
- ssize_t len;
+ int len;
mutex_lock(&buffer->mutex);
- len = fill_write_buffer(buffer, iocb->ki_pos, from);
+ len = fill_write_buffer(buffer, from);
if (len > 0)
len = flush_write_buffer(file, buffer, len);
if (len > 0)
diff --git a/fs/dax.c b/fs/dax.c
index da41f9363568..99b4e78d888f 100644
--- a/fs/dax.c
+++ b/fs/dax.c
@@ -722,7 +722,7 @@ static int copy_cow_page_dax(struct block_device *bdev, struct dax_device *dax_d
return rc;
id = dax_read_lock();
- rc = dax_direct_access(dax_dev, pgoff, PHYS_PFN(PAGE_SIZE), &kaddr, NULL);
+ rc = dax_direct_access(dax_dev, pgoff, 1, &kaddr, NULL);
if (rc < 0) {
dax_read_unlock(id);
return rc;
diff --git a/fs/fuse/dax.c b/fs/fuse/dax.c
index e55723744f58..9d58371d22c2 100644
--- a/fs/fuse/dax.c
+++ b/fs/fuse/dax.c
@@ -1235,8 +1235,6 @@ void fuse_dax_conn_free(struct fuse_conn *fc)
static int fuse_dax_mem_range_init(struct fuse_conn_dax *fcd)
{
long nr_pages, nr_ranges;
- void *kaddr;
- pfn_t pfn;
struct fuse_dax_mapping *range;
int ret, id;
size_t dax_size = -1;
@@ -1248,8 +1246,8 @@ static int fuse_dax_mem_range_init(struct fuse_conn_dax *fcd)
INIT_DELAYED_WORK(&fcd->free_work, fuse_dax_free_mem_worker);
id = dax_read_lock();
- nr_pages = dax_direct_access(fcd->dev, 0, PHYS_PFN(dax_size), &kaddr,
- &pfn);
+ nr_pages = dax_direct_access(fcd->dev, 0, PHYS_PFN(dax_size), NULL,
+ NULL);
dax_read_unlock(id);
if (nr_pages < 0) {
pr_debug("dax_direct_access() returned %ld\n", nr_pages);
diff --git a/fs/io-wq.c b/fs/io-wq.c
index 12fc19353bb0..7d2ed8c7dd31 100644
--- a/fs/io-wq.c
+++ b/fs/io-wq.c
@@ -129,7 +129,7 @@ struct io_cb_cancel_data {
bool cancel_all;
};
-static void create_io_worker(struct io_wq *wq, struct io_wqe *wqe, int index);
+static void create_io_worker(struct io_wq *wq, struct io_wqe *wqe, int index, bool first);
static void io_wqe_dec_running(struct io_worker *worker);
static bool io_worker_get(struct io_worker *worker)
@@ -248,18 +248,20 @@ static void io_wqe_wake_worker(struct io_wqe *wqe, struct io_wqe_acct *acct)
rcu_read_unlock();
if (!ret) {
- bool do_create = false;
+ bool do_create = false, first = false;
raw_spin_lock_irq(&wqe->lock);
if (acct->nr_workers < acct->max_workers) {
atomic_inc(&acct->nr_running);
atomic_inc(&wqe->wq->worker_refs);
+ if (!acct->nr_workers)
+ first = true;
acct->nr_workers++;
do_create = true;
}
raw_spin_unlock_irq(&wqe->lock);
if (do_create)
- create_io_worker(wqe->wq, wqe, acct->index);
+ create_io_worker(wqe->wq, wqe, acct->index, first);
}
}
@@ -282,16 +284,26 @@ static void create_worker_cb(struct callback_head *cb)
struct io_wq *wq;
struct io_wqe *wqe;
struct io_wqe_acct *acct;
+ bool do_create = false, first = false;
cwd = container_of(cb, struct create_worker_data, work);
wqe = cwd->wqe;
wq = wqe->wq;
acct = &wqe->acct[cwd->index];
raw_spin_lock_irq(&wqe->lock);
- if (acct->nr_workers < acct->max_workers)
+ if (acct->nr_workers < acct->max_workers) {
+ if (!acct->nr_workers)
+ first = true;
acct->nr_workers++;
+ do_create = true;
+ }
raw_spin_unlock_irq(&wqe->lock);
- create_io_worker(wq, cwd->wqe, cwd->index);
+ if (do_create) {
+ create_io_worker(wq, wqe, cwd->index, first);
+ } else {
+ atomic_dec(&acct->nr_running);
+ io_worker_ref_put(wq);
+ }
kfree(cwd);
}
@@ -629,7 +641,7 @@ void io_wq_worker_sleeping(struct task_struct *tsk)
raw_spin_unlock_irq(&worker->wqe->lock);
}
-static void create_io_worker(struct io_wq *wq, struct io_wqe *wqe, int index)
+static void create_io_worker(struct io_wq *wq, struct io_wqe *wqe, int index, bool first)
{
struct io_wqe_acct *acct = &wqe->acct[index];
struct io_worker *worker;
@@ -670,7 +682,7 @@ fail:
worker->flags |= IO_WORKER_F_FREE;
if (index == IO_WQ_ACCT_BOUND)
worker->flags |= IO_WORKER_F_BOUND;
- if ((acct->nr_workers == 1) && (worker->flags & IO_WORKER_F_BOUND))
+ if (first && (worker->flags & IO_WORKER_F_BOUND))
worker->flags |= IO_WORKER_F_FIXED;
raw_spin_unlock_irq(&wqe->lock);
wake_up_new_task(tsk);
diff --git a/fs/io_uring.c b/fs/io_uring.c
index bf548af0426c..04c6d059ea94 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -78,6 +78,7 @@
#include <linux/task_work.h>
#include <linux/pagemap.h>
#include <linux/io_uring.h>
+#include <linux/tracehook.h>
#define CREATE_TRACE_POINTS
#include <trace/events/io_uring.h>
@@ -1499,7 +1500,8 @@ static bool __io_cqring_overflow_flush(struct io_ring_ctx *ctx, bool force)
all_flushed = list_empty(&ctx->cq_overflow_list);
if (all_flushed) {
clear_bit(0, &ctx->check_cq_overflow);
- ctx->rings->sq_flags &= ~IORING_SQ_CQ_OVERFLOW;
+ WRITE_ONCE(ctx->rings->sq_flags,
+ ctx->rings->sq_flags & ~IORING_SQ_CQ_OVERFLOW);
}
if (posted)
@@ -1578,7 +1580,9 @@ static bool io_cqring_event_overflow(struct io_ring_ctx *ctx, u64 user_data,
}
if (list_empty(&ctx->cq_overflow_list)) {
set_bit(0, &ctx->check_cq_overflow);
- ctx->rings->sq_flags |= IORING_SQ_CQ_OVERFLOW;
+ WRITE_ONCE(ctx->rings->sq_flags,
+ ctx->rings->sq_flags | IORING_SQ_CQ_OVERFLOW);
+
}
ocqe->cqe.user_data = user_data;
ocqe->cqe.res = res;
@@ -2222,9 +2226,9 @@ static inline unsigned int io_put_rw_kbuf(struct io_kiocb *req)
static inline bool io_run_task_work(void)
{
- if (current->task_works) {
+ if (test_thread_flag(TIF_NOTIFY_SIGNAL) || current->task_works) {
__set_current_state(TASK_RUNNING);
- task_work_run();
+ tracehook_notify_signal();
return true;
}
@@ -6803,14 +6807,16 @@ static inline void io_ring_set_wakeup_flag(struct io_ring_ctx *ctx)
{
/* Tell userspace we may need a wakeup call */
spin_lock_irq(&ctx->completion_lock);
- ctx->rings->sq_flags |= IORING_SQ_NEED_WAKEUP;
+ WRITE_ONCE(ctx->rings->sq_flags,
+ ctx->rings->sq_flags | IORING_SQ_NEED_WAKEUP);
spin_unlock_irq(&ctx->completion_lock);
}
static inline void io_ring_clear_wakeup_flag(struct io_ring_ctx *ctx)
{
spin_lock_irq(&ctx->completion_lock);
- ctx->rings->sq_flags &= ~IORING_SQ_NEED_WAKEUP;
+ WRITE_ONCE(ctx->rings->sq_flags,
+ ctx->rings->sq_flags & ~IORING_SQ_NEED_WAKEUP);
spin_unlock_irq(&ctx->completion_lock);
}
@@ -7132,16 +7138,6 @@ static void **io_alloc_page_table(size_t size)
return table;
}
-static inline void io_rsrc_ref_lock(struct io_ring_ctx *ctx)
-{
- spin_lock_bh(&ctx->rsrc_ref_lock);
-}
-
-static inline void io_rsrc_ref_unlock(struct io_ring_ctx *ctx)
-{
- spin_unlock_bh(&ctx->rsrc_ref_lock);
-}
-
static void io_rsrc_node_destroy(struct io_rsrc_node *ref_node)
{
percpu_ref_exit(&ref_node->refs);
@@ -7158,9 +7154,9 @@ static void io_rsrc_node_switch(struct io_ring_ctx *ctx,
struct io_rsrc_node *rsrc_node = ctx->rsrc_node;
rsrc_node->rsrc_data = data_to_kill;
- io_rsrc_ref_lock(ctx);
+ spin_lock_irq(&ctx->rsrc_ref_lock);
list_add_tail(&rsrc_node->node, &ctx->rsrc_ref_list);
- io_rsrc_ref_unlock(ctx);
+ spin_unlock_irq(&ctx->rsrc_ref_lock);
atomic_inc(&data_to_kill->refs);
percpu_ref_kill(&rsrc_node->refs);
@@ -7199,17 +7195,19 @@ static int io_rsrc_ref_quiesce(struct io_rsrc_data *data, struct io_ring_ctx *ct
/* kill initial ref, already quiesced if zero */
if (atomic_dec_and_test(&data->refs))
break;
+ mutex_unlock(&ctx->uring_lock);
flush_delayed_work(&ctx->rsrc_put_work);
ret = wait_for_completion_interruptible(&data->done);
- if (!ret)
+ if (!ret) {
+ mutex_lock(&ctx->uring_lock);
break;
+ }
atomic_inc(&data->refs);
/* wait for all works potentially completing data->done */
flush_delayed_work(&ctx->rsrc_put_work);
reinit_completion(&data->done);
- mutex_unlock(&ctx->uring_lock);
ret = io_run_task_work_sig();
mutex_lock(&ctx->uring_lock);
} while (ret >= 0);
@@ -7668,9 +7666,10 @@ static void io_rsrc_node_ref_zero(struct percpu_ref *ref)
{
struct io_rsrc_node *node = container_of(ref, struct io_rsrc_node, refs);
struct io_ring_ctx *ctx = node->rsrc_data->ctx;
+ unsigned long flags;
bool first_add = false;
- io_rsrc_ref_lock(ctx);
+ spin_lock_irqsave(&ctx->rsrc_ref_lock, flags);
node->done = true;
while (!list_empty(&ctx->rsrc_ref_list)) {
@@ -7682,7 +7681,7 @@ static void io_rsrc_node_ref_zero(struct percpu_ref *ref)
list_del(&node->node);
first_add |= llist_add(&node->llist, &ctx->rsrc_put_llist);
}
- io_rsrc_ref_unlock(ctx);
+ spin_unlock_irqrestore(&ctx->rsrc_ref_lock, flags);
if (first_add)
mod_delayed_work(system_wq, &ctx->rsrc_put_work, HZ);
@@ -8653,13 +8652,10 @@ static void io_req_caches_free(struct io_ring_ctx *ctx)
mutex_unlock(&ctx->uring_lock);
}
-static bool io_wait_rsrc_data(struct io_rsrc_data *data)
+static void io_wait_rsrc_data(struct io_rsrc_data *data)
{
- if (!data)
- return false;
- if (!atomic_dec_and_test(&data->refs))
+ if (data && !atomic_dec_and_test(&data->refs))
wait_for_completion(&data->done);
- return true;
}
static void io_ring_ctx_free(struct io_ring_ctx *ctx)
@@ -8671,10 +8667,14 @@ static void io_ring_ctx_free(struct io_ring_ctx *ctx)
ctx->mm_account = NULL;
}
+ /* __io_rsrc_put_work() may need uring_lock to progress, wait w/o it */
+ io_wait_rsrc_data(ctx->buf_data);
+ io_wait_rsrc_data(ctx->file_data);
+
mutex_lock(&ctx->uring_lock);
- if (io_wait_rsrc_data(ctx->buf_data))
+ if (ctx->buf_data)
__io_sqe_buffers_unregister(ctx);
- if (io_wait_rsrc_data(ctx->file_data))
+ if (ctx->file_data)
__io_sqe_files_unregister(ctx);
if (ctx->rings)
__io_cqring_overflow_flush(ctx, true);
diff --git a/fs/pipe.c b/fs/pipe.c
index 8e6ef62aeb1c..678dee2a8228 100644
--- a/fs/pipe.c
+++ b/fs/pipe.c
@@ -444,9 +444,6 @@ pipe_write(struct kiocb *iocb, struct iov_iter *from)
#endif
/*
- * Epoll nonsensically wants a wakeup whether the pipe
- * was already empty or not.
- *
* If it wasn't empty we try to merge new data into
* the last buffer.
*
@@ -455,9 +452,9 @@ pipe_write(struct kiocb *iocb, struct iov_iter *from)
* spanning multiple pages.
*/
head = pipe->head;
- was_empty = true;
+ was_empty = pipe_empty(head, pipe->tail);
chars = total_len & (PAGE_SIZE-1);
- if (chars && !pipe_empty(head, pipe->tail)) {
+ if (chars && !was_empty) {
unsigned int mask = pipe->ring_size - 1;
struct pipe_buffer *buf = &pipe->bufs[(head - 1) & mask];
int offset = buf->offset + buf->len;
@@ -590,8 +587,11 @@ out:
* This is particularly important for small writes, because of
* how (for example) the GNU make jobserver uses small writes to
* wake up pending jobs
+ *
+ * Epoll nonsensically wants a wakeup whether the pipe
+ * was already empty or not.
*/
- if (was_empty) {
+ if (was_empty || pipe->poll_usage) {
wake_up_interruptible_sync_poll(&pipe->rd_wait, EPOLLIN | EPOLLRDNORM);
kill_fasync(&pipe->fasync_readers, SIGIO, POLL_IN);
}
@@ -654,6 +654,9 @@ pipe_poll(struct file *filp, poll_table *wait)
struct pipe_inode_info *pipe = filp->private_data;
unsigned int head, tail;
+ /* Epoll has some historical nasty semantics, this enables them */
+ pipe->poll_usage = 1;
+
/*
* Reading pipe state only -- no need for acquiring the semaphore.
*
generated by cgit v1.2.3 (git 2.39.0) at 2024-11-22 17:13:03 +0300