linux_dsm_epyc7002/include/linux/pipe_fs_i.h
Linus Torvalds 472e5b056f pipe: remove pipe_wait() and fix wakeup race with splice
The pipe splice code still used the old model of waiting for pipe IO by
using a non-specific "pipe_wait()" that waited for any pipe event to
happen, which depended on all pipe IO being entirely serialized by the
pipe lock.  So by checking the state you were waiting for, and then
adding yourself to the wait queue before dropping the lock, you were
guaranteed to see all the wakeups.

Strictly speaking, the actual wakeups were not done under the lock, but
the pipe_wait() model still worked, because since the waiter held the
lock when checking whether it should sleep, it would always see the
current state, and the wakeup was always done after updating the state.

However, commit 0ddad21d3e ("pipe: use exclusive waits when reading or
writing") split the single wait-queue into two, and in the process also
made the "wait for event" code wait for _two_ wait queues, and that then
showed a race with the wakers that were not serialized by the pipe lock.

It's only splice that used that "pipe_wait()" model, so the problem
wasn't obvious, but Josef Bacik reports:

 "I hit a hang with fstest btrfs/187, which does a btrfs send into
  /dev/null. This works by creating a pipe, the write side is given to
  the kernel to write into, and the read side is handed to a thread that
  splices into a file, in this case /dev/null.

  The box that was hung had the write side stuck here [pipe_write] and
  the read side stuck here [splice_from_pipe_next -> pipe_wait].

  [ more details about pipe_wait() scenario ]

  The problem is we're doing the prepare_to_wait, which sets our state
  each time, however we can be woken up either with reads or writes. In
  the case above we race with the WRITER waking us up, and re-set our
  state to INTERRUPTIBLE, and thus never break out of schedule"

Josef had a patch that avoided the issue in pipe_wait() by just making
it set the state only once, but the deeper problem is that pipe_wait()
depends on a level of synchonization by the pipe mutex that it really
shouldn't.  And the whole "wait for any pipe state change" model really
isn't very good to begin with.

So rather than trying to work around things in pipe_wait(), remove that
legacy model of "wait for arbitrary pipe event" entirely, and actually
create functions that wait for the pipe actually being readable or
writable, and can do so without depending on the pipe lock serializing
everything.

Fixes: 0ddad21d3e ("pipe: use exclusive waits when reading or writing")
Link: https://lore.kernel.org/linux-fsdevel/bfa88b5ad6f069b2b679316b9e495a970130416c.1601567868.git.josef@toxicpanda.com/
Reported-by: Josef Bacik <josef@toxicpanda.com>
Reviewed-and-tested-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2020-10-01 19:14:36 -07:00

276 lines
8.6 KiB
C

/* SPDX-License-Identifier: GPL-2.0 */
#ifndef _LINUX_PIPE_FS_I_H
#define _LINUX_PIPE_FS_I_H
#define PIPE_DEF_BUFFERS 16
#define PIPE_BUF_FLAG_LRU 0x01 /* page is on the LRU */
#define PIPE_BUF_FLAG_ATOMIC 0x02 /* was atomically mapped */
#define PIPE_BUF_FLAG_GIFT 0x04 /* page is a gift */
#define PIPE_BUF_FLAG_PACKET 0x08 /* read() as a packet */
#define PIPE_BUF_FLAG_CAN_MERGE 0x10 /* can merge buffers */
#define PIPE_BUF_FLAG_WHOLE 0x20 /* read() must return entire buffer or error */
#ifdef CONFIG_WATCH_QUEUE
#define PIPE_BUF_FLAG_LOSS 0x40 /* Message loss happened after this buffer */
#endif
/**
* struct pipe_buffer - a linux kernel pipe buffer
* @page: the page containing the data for the pipe buffer
* @offset: offset of data inside the @page
* @len: length of data inside the @page
* @ops: operations associated with this buffer. See @pipe_buf_operations.
* @flags: pipe buffer flags. See above.
* @private: private data owned by the ops.
**/
struct pipe_buffer {
struct page *page;
unsigned int offset, len;
const struct pipe_buf_operations *ops;
unsigned int flags;
unsigned long private;
};
/**
* struct pipe_inode_info - a linux kernel pipe
* @mutex: mutex protecting the whole thing
* @rd_wait: reader wait point in case of empty pipe
* @wr_wait: writer wait point in case of full pipe
* @head: The point of buffer production
* @tail: The point of buffer consumption
* @note_loss: The next read() should insert a data-lost message
* @max_usage: The maximum number of slots that may be used in the ring
* @ring_size: total number of buffers (should be a power of 2)
* @nr_accounted: The amount this pipe accounts for in user->pipe_bufs
* @tmp_page: cached released page
* @readers: number of current readers of this pipe
* @writers: number of current writers of this pipe
* @files: number of struct file referring this pipe (protected by ->i_lock)
* @r_counter: reader counter
* @w_counter: writer counter
* @fasync_readers: reader side fasync
* @fasync_writers: writer side fasync
* @bufs: the circular array of pipe buffers
* @user: the user who created this pipe
* @watch_queue: If this pipe is a watch_queue, this is the stuff for that
**/
struct pipe_inode_info {
struct mutex mutex;
wait_queue_head_t rd_wait, wr_wait;
unsigned int head;
unsigned int tail;
unsigned int max_usage;
unsigned int ring_size;
#ifdef CONFIG_WATCH_QUEUE
bool note_loss;
#endif
unsigned int nr_accounted;
unsigned int readers;
unsigned int writers;
unsigned int files;
unsigned int r_counter;
unsigned int w_counter;
struct page *tmp_page;
struct fasync_struct *fasync_readers;
struct fasync_struct *fasync_writers;
struct pipe_buffer *bufs;
struct user_struct *user;
#ifdef CONFIG_WATCH_QUEUE
struct watch_queue *watch_queue;
#endif
};
/*
* Note on the nesting of these functions:
*
* ->confirm()
* ->try_steal()
*
* That is, ->try_steal() must be called on a confirmed buffer. See below for
* the meaning of each operation. Also see the kerneldoc in fs/pipe.c for the
* pipe and generic variants of these hooks.
*/
struct pipe_buf_operations {
/*
* ->confirm() verifies that the data in the pipe buffer is there
* and that the contents are good. If the pages in the pipe belong
* to a file system, we may need to wait for IO completion in this
* hook. Returns 0 for good, or a negative error value in case of
* error. If not present all pages are considered good.
*/
int (*confirm)(struct pipe_inode_info *, struct pipe_buffer *);
/*
* When the contents of this pipe buffer has been completely
* consumed by a reader, ->release() is called.
*/
void (*release)(struct pipe_inode_info *, struct pipe_buffer *);
/*
* Attempt to take ownership of the pipe buffer and its contents.
* ->try_steal() returns %true for success, in which case the contents
* of the pipe (the buf->page) is locked and now completely owned by the
* caller. The page may then be transferred to a different mapping, the
* most often used case is insertion into different file address space
* cache.
*/
bool (*try_steal)(struct pipe_inode_info *, struct pipe_buffer *);
/*
* Get a reference to the pipe buffer.
*/
bool (*get)(struct pipe_inode_info *, struct pipe_buffer *);
};
/**
* pipe_empty - Return true if the pipe is empty
* @head: The pipe ring head pointer
* @tail: The pipe ring tail pointer
*/
static inline bool pipe_empty(unsigned int head, unsigned int tail)
{
return head == tail;
}
/**
* pipe_occupancy - Return number of slots used in the pipe
* @head: The pipe ring head pointer
* @tail: The pipe ring tail pointer
*/
static inline unsigned int pipe_occupancy(unsigned int head, unsigned int tail)
{
return head - tail;
}
/**
* pipe_full - Return true if the pipe is full
* @head: The pipe ring head pointer
* @tail: The pipe ring tail pointer
* @limit: The maximum amount of slots available.
*/
static inline bool pipe_full(unsigned int head, unsigned int tail,
unsigned int limit)
{
return pipe_occupancy(head, tail) >= limit;
}
/**
* pipe_space_for_user - Return number of slots available to userspace
* @head: The pipe ring head pointer
* @tail: The pipe ring tail pointer
* @pipe: The pipe info structure
*/
static inline unsigned int pipe_space_for_user(unsigned int head, unsigned int tail,
struct pipe_inode_info *pipe)
{
unsigned int p_occupancy, p_space;
p_occupancy = pipe_occupancy(head, tail);
if (p_occupancy >= pipe->max_usage)
return 0;
p_space = pipe->ring_size - p_occupancy;
if (p_space > pipe->max_usage)
p_space = pipe->max_usage;
return p_space;
}
/**
* pipe_buf_get - get a reference to a pipe_buffer
* @pipe: the pipe that the buffer belongs to
* @buf: the buffer to get a reference to
*
* Return: %true if the reference was successfully obtained.
*/
static inline __must_check bool pipe_buf_get(struct pipe_inode_info *pipe,
struct pipe_buffer *buf)
{
return buf->ops->get(pipe, buf);
}
/**
* pipe_buf_release - put a reference to a pipe_buffer
* @pipe: the pipe that the buffer belongs to
* @buf: the buffer to put a reference to
*/
static inline void pipe_buf_release(struct pipe_inode_info *pipe,
struct pipe_buffer *buf)
{
const struct pipe_buf_operations *ops = buf->ops;
buf->ops = NULL;
ops->release(pipe, buf);
}
/**
* pipe_buf_confirm - verify contents of the pipe buffer
* @pipe: the pipe that the buffer belongs to
* @buf: the buffer to confirm
*/
static inline int pipe_buf_confirm(struct pipe_inode_info *pipe,
struct pipe_buffer *buf)
{
if (!buf->ops->confirm)
return 0;
return buf->ops->confirm(pipe, buf);
}
/**
* pipe_buf_try_steal - attempt to take ownership of a pipe_buffer
* @pipe: the pipe that the buffer belongs to
* @buf: the buffer to attempt to steal
*/
static inline bool pipe_buf_try_steal(struct pipe_inode_info *pipe,
struct pipe_buffer *buf)
{
if (!buf->ops->try_steal)
return false;
return buf->ops->try_steal(pipe, buf);
}
/* Differs from PIPE_BUF in that PIPE_SIZE is the length of the actual
memory allocation, whereas PIPE_BUF makes atomicity guarantees. */
#define PIPE_SIZE PAGE_SIZE
/* Pipe lock and unlock operations */
void pipe_lock(struct pipe_inode_info *);
void pipe_unlock(struct pipe_inode_info *);
void pipe_double_lock(struct pipe_inode_info *, struct pipe_inode_info *);
extern unsigned int pipe_max_size;
extern unsigned long pipe_user_pages_hard;
extern unsigned long pipe_user_pages_soft;
/* Wait for a pipe to be readable/writable while dropping the pipe lock */
void pipe_wait_readable(struct pipe_inode_info *);
void pipe_wait_writable(struct pipe_inode_info *);
struct pipe_inode_info *alloc_pipe_info(void);
void free_pipe_info(struct pipe_inode_info *);
/* Generic pipe buffer ops functions */
bool generic_pipe_buf_get(struct pipe_inode_info *, struct pipe_buffer *);
bool generic_pipe_buf_try_steal(struct pipe_inode_info *, struct pipe_buffer *);
void generic_pipe_buf_release(struct pipe_inode_info *, struct pipe_buffer *);
extern const struct pipe_buf_operations nosteal_pipe_buf_ops;
#ifdef CONFIG_WATCH_QUEUE
unsigned long account_pipe_buffers(struct user_struct *user,
unsigned long old, unsigned long new);
bool too_many_pipe_buffers_soft(unsigned long user_bufs);
bool too_many_pipe_buffers_hard(unsigned long user_bufs);
bool pipe_is_unprivileged_user(void);
#endif
/* for F_SETPIPE_SZ and F_GETPIPE_SZ */
#ifdef CONFIG_WATCH_QUEUE
int pipe_resize_ring(struct pipe_inode_info *pipe, unsigned int nr_slots);
#endif
long pipe_fcntl(struct file *, unsigned int, unsigned long arg);
struct pipe_inode_info *get_pipe_info(struct file *file, bool for_splice);
int create_pipe_files(struct file **, int);
unsigned int round_pipe_size(unsigned long size);
#endif