/*
  FUSE: Filesystem in Userspace
  Copyright (C) 2001-2008  Miklos Szeredi <miklos@szeredi.hu>

  This program can be distributed under the terms of the GNU GPL.
  See the file COPYING.
*/

#include "fuse_i.h"

#include <linux/pagemap.h>
#include <linux/slab.h>
#include <linux/kernel.h>
#include <linux/sched.h>
#include <linux/sched/signal.h>
#include <linux/module.h>
#include <linux/compat.h>
#include <linux/swap.h>
#include <linux/falloc.h>
#include <linux/uio.h>
#include <linux/fs.h>

static struct page **fuse_pages_alloc(unsigned int npages, gfp_t flags,
                                      struct fuse_page_desc **desc)
{
        struct page **pages;

        pages = kzalloc(npages * (sizeof(struct page *) +
                                  sizeof(struct fuse_page_desc)), flags);
        *desc = (void *) (pages + npages);

        return pages;
}

static int fuse_send_open(struct fuse_mount *fm, u64 nodeid, struct file *file,
                          int opcode, struct fuse_open_out *outargp)
{
        struct fuse_open_in inarg;
        FUSE_ARGS(args);

        memset(&inarg, 0, sizeof(inarg));
        inarg.flags = file->f_flags & ~(O_CREAT | O_EXCL | O_NOCTTY);
        if (!fm->fc->atomic_o_trunc)
                inarg.flags &= ~O_TRUNC;

        if (fm->fc->handle_killpriv_v2 &&
            (inarg.flags & O_TRUNC) && !capable(CAP_FSETID)) {
                inarg.open_flags |= FUSE_OPEN_KILL_SUIDGID;
        }

        args.opcode = opcode;
        args.nodeid = nodeid;
        args.in_numargs = 1;
        args.in_args[0].size = sizeof(inarg);
        args.in_args[0].value = &inarg;
        args.out_numargs = 1;
        args.out_args[0].size = sizeof(*outargp);
        args.out_args[0].value = outargp;

        return fuse_simple_request(fm, &args);
}

struct fuse_release_args {
        struct fuse_args args;
        struct fuse_release_in inarg;
        struct inode *inode;
};

struct fuse_file *fuse_file_alloc(struct fuse_mount *fm)
{
        struct fuse_file *ff;

        ff = kzalloc(sizeof(struct fuse_file), GFP_KERNEL_ACCOUNT);
        if (unlikely(!ff))
                return NULL;

        ff->fm = fm;
        ff->release_args = kzalloc(sizeof(*ff->release_args),
                                   GFP_KERNEL_ACCOUNT);
        if (!ff->release_args) {
                kfree(ff);
                return NULL;
        }

        INIT_LIST_HEAD(&ff->write_entry);
        mutex_init(&ff->readdir.lock);
        refcount_set(&ff->count, 1);
        RB_CLEAR_NODE(&ff->polled_node);
        init_waitqueue_head(&ff->poll_wait);

        ff->kh = atomic64_inc_return(&fm->fc->khctr);

        return ff;
}

void fuse_file_free(struct fuse_file *ff)
{
        kfree(ff->release_args);
        mutex_destroy(&ff->readdir.lock);
        kfree(ff);
}

static struct fuse_file *fuse_file_get(struct fuse_file *ff)
{
        refcount_inc(&ff->count);
        return ff;
}

static void fuse_release_end(struct fuse_mount *fm, struct fuse_args *args,
                             int error)
{
        struct fuse_release_args *ra = container_of(args, typeof(*ra), args);

        iput(ra->inode);
        kfree(ra);
}

static void fuse_file_put(struct fuse_file *ff, bool sync, bool isdir)
{
        if (refcount_dec_and_test(&ff->count)) {
                struct fuse_args *args = &ff->release_args->args;

                if (isdir ? ff->fm->fc->no_opendir : ff->fm->fc->no_open) {
                        /* Do nothing when client does not implement 'open' */
                        fuse_release_end(ff->fm, args, 0);
                } else if (sync) {
                        fuse_simple_request(ff->fm, args);
                        fuse_release_end(ff->fm, args, 0);
                } else {
                        args->end = fuse_release_end;
                        if (fuse_simple_background(ff->fm, args,
                                                   GFP_KERNEL | __GFP_NOFAIL))
                                fuse_release_end(ff->fm, args, -ENOTCONN);
                }
                kfree(ff);
        }
}

int fuse_do_open(struct fuse_mount *fm, u64 nodeid, struct file *file,
                 bool isdir)
{
        struct fuse_conn *fc = fm->fc;
        struct fuse_file *ff;
        int opcode = isdir ? FUSE_OPENDIR : FUSE_OPEN;

        ff = fuse_file_alloc(fm);
        if (!ff)
                return -ENOMEM;

        ff->fh = 0;
        /* Default for no-open */
        ff->open_flags = FOPEN_KEEP_CACHE | (isdir ? FOPEN_CACHE_DIR : 0);
        if (isdir ? !fc->no_opendir : !fc->no_open) {
                struct fuse_open_out outarg;
                int err;

                err = fuse_send_open(fm, nodeid, file, opcode, &outarg);
                if (!err) {
                        ff->fh = outarg.fh;
                        ff->open_flags = outarg.open_flags;

                } else if (err != -ENOSYS) {
                        fuse_file_free(ff);
                        return err;
                } else {
                        if (isdir)
                                fc->no_opendir = 1;
                        else
                                fc->no_open = 1;
                }
        }

        if (isdir)
                ff->open_flags &= ~FOPEN_DIRECT_IO;

        ff->nodeid = nodeid;
        file->private_data = ff;

        return 0;
}
EXPORT_SYMBOL_GPL(fuse_do_open);

static void fuse_link_write_file(struct file *file)
{
        struct inode *inode = file_inode(file);
        struct fuse_inode *fi = get_fuse_inode(inode);
        struct fuse_file *ff = file->private_data;
        /*
         * file may be written through mmap, so chain it onto the
         * inodes's write_file list
         */
        spin_lock(&fi->lock);
        if (list_empty(&ff->write_entry))
                list_add(&ff->write_entry, &fi->write_files);
        spin_unlock(&fi->lock);
}

void fuse_finish_open(struct inode *inode, struct file *file)
{
        struct fuse_file *ff = file->private_data;
        struct fuse_conn *fc = get_fuse_conn(inode);

        if (!(ff->open_flags & FOPEN_KEEP_CACHE))
                invalidate_inode_pages2(inode->i_mapping);
        if (ff->open_flags & FOPEN_STREAM)
                stream_open(inode, file);
        else if (ff->open_flags & FOPEN_NONSEEKABLE)
                nonseekable_open(inode, file);
        if (fc->atomic_o_trunc && (file->f_flags & O_TRUNC)) {
                struct fuse_inode *fi = get_fuse_inode(inode);

                spin_lock(&fi->lock);
                fi->attr_version = atomic64_inc_return(&fc->attr_version);
                i_size_write(inode, 0);
                spin_unlock(&fi->lock);
                fuse_invalidate_attr(inode);
                if (fc->writeback_cache)
                        file_update_time(file);
        }
        if ((file->f_mode & FMODE_WRITE) && fc->writeback_cache)
                fuse_link_write_file(file);
}

int fuse_open_common(struct inode *inode, struct file *file, bool isdir)
{
        struct fuse_mount *fm = get_fuse_mount(inode);
        struct fuse_conn *fc = fm->fc;
        int err;
        bool is_wb_truncate = (file->f_flags & O_TRUNC) &&
                          fc->atomic_o_trunc &&
                          fc->writeback_cache;
        bool dax_truncate = (file->f_flags & O_TRUNC) &&
                          fc->atomic_o_trunc && FUSE_IS_DAX(inode);

        if (fuse_is_bad(inode))
                return -EIO;

        err = generic_file_open(inode, file);
        if (err)
                return err;

        if (is_wb_truncate || dax_truncate) {
                inode_lock(inode);
                fuse_set_nowrite(inode);
        }

        if (dax_truncate) {
                down_write(&get_fuse_inode(inode)->i_mmap_sem);
                err = fuse_dax_break_layouts(inode, 0, 0);
                if (err)
                        goto out;
        }

        err = fuse_do_open(fm, get_node_id(inode), file, isdir);
        if (!err)
                fuse_finish_open(inode, file);

out:
        if (dax_truncate)
                up_write(&get_fuse_inode(inode)->i_mmap_sem);

        if (is_wb_truncate | dax_truncate) {
                fuse_release_nowrite(inode);
                inode_unlock(inode);
        }

        return err;
}

static void fuse_prepare_release(struct fuse_inode *fi, struct fuse_file *ff,
                                 int flags, int opcode)
{
        struct fuse_conn *fc = ff->fm->fc;
        struct fuse_release_args *ra = ff->release_args;

        /* Inode is NULL on error path of fuse_create_open() */
        if (likely(fi)) {
                spin_lock(&fi->lock);
                list_del(&ff->write_entry);
                spin_unlock(&fi->lock);
        }
        spin_lock(&fc->lock);
        if (!RB_EMPTY_NODE(&ff->polled_node))
                rb_erase(&ff->polled_node, &fc->polled_files);
        spin_unlock(&fc->lock);

        wake_up_interruptible_all(&ff->poll_wait);

        ra->inarg.fh = ff->fh;
        ra->inarg.flags = flags;
        ra->args.in_numargs = 1;
        ra->args.in_args[0].size = sizeof(struct fuse_release_in);
        ra->args.in_args[0].value = &ra->inarg;
        ra->args.opcode = opcode;
        ra->args.nodeid = ff->nodeid;
        ra->args.force = true;
        ra->args.nocreds = true;
}

void fuse_release_common(struct file *file, bool isdir)
{
        struct fuse_inode *fi = get_fuse_inode(file_inode(file));
        struct fuse_file *ff = file->private_data;
        struct fuse_release_args *ra = ff->release_args;
        int opcode = isdir ? FUSE_RELEASEDIR : FUSE_RELEASE;

        fuse_prepare_release(fi, ff, file->f_flags, opcode);

        if (ff->flock) {
                ra->inarg.release_flags |= FUSE_RELEASE_FLOCK_UNLOCK;
                ra->inarg.lock_owner = fuse_lock_owner_id(ff->fm->fc,
                                                          (fl_owner_t) file);
        }
        /* Hold inode until release is finished */
        ra->inode = igrab(file_inode(file));

        /*
         * Normally this will send the RELEASE request, however if
         * some asynchronous READ or WRITE requests are outstanding,
         * the sending will be delayed.
         *
         * Make the release synchronous if this is a fuseblk mount,
         * synchronous RELEASE is allowed (and desirable) in this case
         * because the server can be trusted not to screw up.
         */
        fuse_file_put(ff, ff->fm->fc->destroy, isdir);
}

static int fuse_open(struct inode *inode, struct file *file)
{
        return fuse_open_common(inode, file, false);
}

static int fuse_release(struct inode *inode, struct file *file)
{
        struct fuse_conn *fc = get_fuse_conn(inode);

        /* see fuse_vma_close() for !writeback_cache case */
        if (fc->writeback_cache)
                write_inode_now(inode, 1);

        fuse_release_common(file, false);

        /* return value is ignored by VFS */
        return 0;
}

void fuse_sync_release(struct fuse_inode *fi, struct fuse_file *ff, int flags)
{
        WARN_ON(refcount_read(&ff->count) > 1);
        fuse_prepare_release(fi, ff, flags, FUSE_RELEASE);
        /*
         * iput(NULL) is a no-op and since the refcount is 1 and everything's
         * synchronous, we are fine with not doing igrab() here"
         */
        fuse_file_put(ff, true, false);
}
EXPORT_SYMBOL_GPL(fuse_sync_release);

/*
 * Scramble the ID space with XTEA, so that the value of the files_struct
 * pointer is not exposed to userspace.
 */
u64 fuse_lock_owner_id(struct fuse_conn *fc, fl_owner_t id)
{
        u32 *k = fc->scramble_key;
        u64 v = (unsigned long) id;
        u32 v0 = v;
        u32 v1 = v >> 32;
        u32 sum = 0;
        int i;

        for (i = 0; i < 32; i++) {
                v0 += ((v1 << 4 ^ v1 >> 5) + v1) ^ (sum + k[sum & 3]);
                sum += 0x9E3779B9;
                v1 += ((v0 << 4 ^ v0 >> 5) + v0) ^ (sum + k[sum>>11 & 3]);
        }

        return (u64) v0 + ((u64) v1 << 32);
}

struct fuse_writepage_args {
        struct fuse_io_args ia;
        struct rb_node writepages_entry;
        struct list_head queue_entry;
        struct fuse_writepage_args *next;
        struct inode *inode;
};

static struct fuse_writepage_args *fuse_find_writeback(struct fuse_inode *fi,
                                            pgoff_t idx_from, pgoff_t idx_to)
{
        struct rb_node *n;

        n = fi->writepages.rb_node;

        while (n) {
                struct fuse_writepage_args *wpa;
                pgoff_t curr_index;

                wpa = rb_entry(n, struct fuse_writepage_args, writepages_entry);
                WARN_ON(get_fuse_inode(wpa->inode) != fi);
                curr_index = wpa->ia.write.in.offset >> PAGE_SHIFT;
                if (idx_from >= curr_index + wpa->ia.ap.num_pages)
                        n = n->rb_right;
                else if (idx_to < curr_index)
                        n = n->rb_left;
                else
                        return wpa;
        }
        return NULL;
}

/*
 * Check if any page in a range is under writeback
 *
 * This is currently done by walking the list of writepage requests
 * for the inode, which can be pretty inefficient.
 */
static bool fuse_range_is_writeback(struct inode *inode, pgoff_t idx_from,
                                   pgoff_t idx_to)
{
        struct fuse_inode *fi = get_fuse_inode(inode);
        bool found;

        spin_lock(&fi->lock);
        found = fuse_find_writeback(fi, idx_from, idx_to);
        spin_unlock(&fi->lock);

        return found;
}

static inline bool fuse_page_is_writeback(struct inode *inode, pgoff_t index)
{
        return fuse_range_is_writeback(inode, index, index);
}

/*
 * Wait for page writeback to be completed.
 *
 * Since fuse doesn't rely on the VM writeback tracking, this has to
 * use some other means.
 */
static void fuse_wait_on_page_writeback(struct inode *inode, pgoff_t index)
{
        struct fuse_inode *fi = get_fuse_inode(inode);

        wait_event(fi->page_waitq, !fuse_page_is_writeback(inode, index));
}

/*
 * Wait for all pending writepages on the inode to finish.
 *
 * This is currently done by blocking further writes with FUSE_NOWRITE
 * and waiting for all sent writes to complete.
 *
 * This must be called under i_mutex, otherwise the FUSE_NOWRITE usage
 * could conflict with truncation.
 */
static void fuse_sync_writes(struct inode *inode)
{
        fuse_set_nowrite(inode);
        fuse_release_nowrite(inode);
}

static int fuse_flush(struct file *file, fl_owner_t id)
{
        struct inode *inode = file_inode(file);
        struct fuse_mount *fm = get_fuse_mount(inode);
        struct fuse_file *ff = file->private_data;
        struct fuse_flush_in inarg;
        FUSE_ARGS(args);
        int err;

        if (fuse_is_bad(inode))
                return -EIO;

        err = write_inode_now(inode, 1);
        if (err)
                return err;

        inode_lock(inode);
        fuse_sync_writes(inode);
        inode_unlock(inode);

        err = filemap_check_errors(file->f_mapping);
        if (err)
                return err;

        err = 0;
        if (fm->fc->no_flush)
                goto inval_attr_out;

        memset(&inarg, 0, sizeof(inarg));
        inarg.fh = ff->fh;
        inarg.lock_owner = fuse_lock_owner_id(fm->fc, id);
        args.opcode = FUSE_FLUSH;
        args.nodeid = get_node_id(inode);
        args.in_numargs = 1;
        args.in_args[0].size = sizeof(inarg);
        args.in_args[0].value = &inarg;
        args.force = true;

        err = fuse_simple_request(fm, &args);
        if (err == -ENOSYS) {
                fm->fc->no_flush = 1;
                err = 0;
        }

inval_attr_out:
        /*
         * In memory i_blocks is not maintained by fuse, if writeback cache is
         * enabled, i_blocks from cached attr may not be accurate.
         */
        if (!err && fm->fc->writeback_cache)
                fuse_invalidate_attr(inode);
        return err;
}

int fuse_fsync_common(struct file *file, loff_t start, loff_t end,
                      int datasync, int opcode)
{
        struct inode *inode = file->f_mapping->host;
        struct fuse_mount *fm = get_fuse_mount(inode);
        struct fuse_file *ff = file->private_data;
        FUSE_ARGS(args);
        struct fuse_fsync_in inarg;

        memset(&inarg, 0, sizeof(inarg));
        inarg.fh = ff->fh;
        inarg.fsync_flags = datasync ? FUSE_FSYNC_FDATASYNC : 0;
        args.opcode = opcode;
        args.nodeid = get_node_id(inode);
        args.in_numargs = 1;
        args.in_args[0].size = sizeof(inarg);
        args.in_args[0].value = &inarg;
        return fuse_simple_request(fm, &args);
}

static int fuse_fsync(struct file *file, loff_t start, loff_t end,
                      int datasync)
{
        struct inode *inode = file->f_mapping->host;
        struct fuse_conn *fc = get_fuse_conn(inode);
        int err;

        if (fuse_is_bad(inode))
                return -EIO;

        inode_lock(inode);

        /*
         * Start writeback against all dirty pages of the inode, then
         * wait for all outstanding writes, before sending the FSYNC
         * request.
         */
        err = file_write_and_wait_range(file, start, end);
        if (err)
                goto out;

        fuse_sync_writes(inode);

        /*
         * Due to implementation of fuse writeback
         * file_write_and_wait_range() does not catch errors.
         * We have to do this directly after fuse_sync_writes()
         */
        err = file_check_and_advance_wb_err(file);
        if (err)
                goto out;

        err = sync_inode_metadata(inode, 1);
        if (err)
                goto out;

        if (fc->no_fsync)
                goto out;

        err = fuse_fsync_common(file, start, end, datasync, FUSE_FSYNC);
        if (err == -ENOSYS) {
                fc->no_fsync = 1;
                err = 0;
        }
out:
        inode_unlock(inode);

        return err;
}

void fuse_read_args_fill(struct fuse_io_args *ia, struct file *file, loff_t pos,
                         size_t count, int opcode)
{
        struct fuse_file *ff = file->private_data;
        struct fuse_args *args = &ia->ap.args;

        ia->read.in.fh = ff->fh;
        ia->read.in.offset = pos;
        ia->read.in.size = count;
        ia->read.in.flags = file->f_flags;
        args->opcode = opcode;
        args->nodeid = ff->nodeid;
        args->in_numargs = 1;
        args->in_args[0].size = sizeof(ia->read.in);
        args->in_args[0].value = &ia->read.in;
        args->out_argvar = true;
        args->out_numargs = 1;
        args->out_args[0].size = count;
}

static void fuse_release_user_pages(struct fuse_args_pages *ap,
                                    bool should_dirty)
{
        unsigned int i;

        for (i = 0; i < ap->num_pages; i++) {
                if (should_dirty)
                        set_page_dirty_lock(ap->pages[i]);
                put_page(ap->pages[i]);
        }
}

static void fuse_io_release(struct kref *kref)
{
        kfree(container_of(kref, struct fuse_io_priv, refcnt));
}

static ssize_t fuse_get_res_by_io(struct fuse_io_priv *io)
{
        if (io->err)
                return io->err;

        if (io->bytes >= 0 && io->write)
                return -EIO;

        return io->bytes < 0 ? io->size : io->bytes;
}

/**
 * In case of short read, the caller sets 'pos' to the position of
 * actual end of fuse request in IO request. Otherwise, if bytes_requested
 * == bytes_transferred or rw == WRITE, the caller sets 'pos' to -1.
 *
 * An example:
 * User requested DIO read of 64K. It was splitted into two 32K fuse requests,
 * both submitted asynchronously. The first of them was ACKed by userspace as
 * fully completed (req->out.args[0].size == 32K) resulting in pos == -1. The
 * second request was ACKed as short, e.g. only 1K was read, resulting in
 * pos == 33K.
 *
 * Thus, when all fuse requests are completed, the minimal non-negative 'pos'
 * will be equal to the length of the longest contiguous fragment of
 * transferred data starting from the beginning of IO request.
 */
static void fuse_aio_complete(struct fuse_io_priv *io, int err, ssize_t pos)
{
        int left;

        spin_lock(&io->lock);
        if (err)
                io->err = io->err ? : err;
        else if (pos >= 0 && (io->bytes < 0 || pos < io->bytes))
                io->bytes = pos;

        left = --io->reqs;
        if (!left && io->blocking)
                complete(io->done);
        spin_unlock(&io->lock);

        if (!left && !io->blocking) {
                ssize_t res = fuse_get_res_by_io(io);

                if (res >= 0) {
                        struct inode *inode = file_inode(io->iocb->ki_filp);
                        struct fuse_conn *fc = get_fuse_conn(inode);
                        struct fuse_inode *fi = get_fuse_inode(inode);

                        spin_lock(&fi->lock);
                        fi->attr_version = atomic64_inc_return(&fc->attr_version);
                        spin_unlock(&fi->lock);
                }

                io->iocb->ki_complete(io->iocb, res, 0);
        }

        kref_put(&io->refcnt, fuse_io_release);
}

static struct fuse_io_args *fuse_io_alloc(struct fuse_io_priv *io,
                                          unsigned int npages)
{
        struct fuse_io_args *ia;

        ia = kzalloc(sizeof(*ia), GFP_KERNEL);
        if (ia) {
                ia->io = io;
                ia->ap.pages = fuse_pages_alloc(npages, GFP_KERNEL,
                                                &ia->ap.descs);
                if (!ia->ap.pages) {
                        kfree(ia);
                        ia = NULL;
                }
        }
        return ia;
}

static void fuse_io_free(struct fuse_io_args *ia)
{
        kfree(ia->ap.pages);
        kfree(ia);
}

static void fuse_aio_complete_req(struct fuse_mount *fm, struct fuse_args *args,
                                  int err)
{
        struct fuse_io_args *ia = container_of(args, typeof(*ia), ap.args);
        struct fuse_io_priv *io = ia->io;
        ssize_t pos = -1;

        fuse_release_user_pages(&ia->ap, io->should_dirty);

        if (err) {
                /* Nothing */
        } else if (io->write) {
                if (ia->write.out.size > ia->write.in.size) {
                        err = -EIO;
                } else if (ia->write.in.size != ia->write.out.size) {
                        pos = ia->write.in.offset - io->offset +
                                ia->write.out.size;
                }
        } else {
                u32 outsize = args->out_args[0].size;

                if (ia->read.in.size != outsize)
                        pos = ia->read.in.offset - io->offset + outsize;
        }

        fuse_aio_complete(io, err, pos);
        fuse_io_free(ia);
}

static ssize_t fuse_async_req_send(struct fuse_mount *fm,
                                   struct fuse_io_args *ia, size_t num_bytes)
{
        ssize_t err;
        struct fuse_io_priv *io = ia->io;

        spin_lock(&io->lock);
        kref_get(&io->refcnt);
        io->size += num_bytes;
        io->reqs++;
        spin_unlock(&io->lock);

        ia->ap.args.end = fuse_aio_complete_req;
        ia->ap.args.may_block = io->should_dirty;
        err = fuse_simple_background(fm, &ia->ap.args, GFP_KERNEL);
        if (err)
                fuse_aio_complete_req(fm, &ia->ap.args, err);

        return num_bytes;
}

static ssize_t fuse_send_read(struct fuse_io_args *ia, loff_t pos, size_t count,
                              fl_owner_t owner)
{
        struct file *file = ia->io->iocb->ki_filp;
        struct fuse_file *ff = file->private_data;
        struct fuse_mount *fm = ff->fm;

        fuse_read_args_fill(ia, file, pos, count, FUSE_READ);
        if (owner != NULL) {
                ia->read.in.read_flags |= FUSE_READ_LOCKOWNER;
                ia->read.in.lock_owner = fuse_lock_owner_id(fm->fc, owner);
        }

        if (ia->io->async)
                return fuse_async_req_send(fm, ia, count);

        return fuse_simple_request(fm, &ia->ap.args);
}

static void fuse_read_update_size(struct inode *inode, loff_t size,
                                  u64 attr_ver)
{
        struct fuse_conn *fc = get_fuse_conn(inode);
        struct fuse_inode *fi = get_fuse_inode(inode);

        spin_lock(&fi->lock);
        if (attr_ver == fi->attr_version && size < inode->i_size &&
            !test_bit(FUSE_I_SIZE_UNSTABLE, &fi->state)) {
                fi->attr_version = atomic64_inc_return(&fc->attr_version);
                i_size_write(inode, size);
        }
        spin_unlock(&fi->lock);
}

static void fuse_short_read(struct inode *inode, u64 attr_ver, size_t num_read,
                            struct fuse_args_pages *ap)
{
        struct fuse_conn *fc = get_fuse_conn(inode);

        if (fc->writeback_cache) {
                /*
                 * A hole in a file. Some data after the hole are in page cache,
                 * but have not reached the client fs yet. So, the hole is not
                 * present there.
                 */
                int i;
                int start_idx = num_read >> PAGE_SHIFT;
                size_t off = num_read & (PAGE_SIZE - 1);

                for (i = start_idx; i < ap->num_pages; i++) {
                        zero_user_segment(ap->pages[i], off, PAGE_SIZE);
                        off = 0;
                }
        } else {
                loff_t pos = page_offset(ap->pages[0]) + num_read;
                fuse_read_update_size(inode, pos, attr_ver);
        }
}

static int fuse_do_readpage(struct file *file, struct page *page)
{
        struct inode *inode = page->mapping->host;
        struct fuse_mount *fm = get_fuse_mount(inode);
        loff_t pos = page_offset(page);
        struct fuse_page_desc desc = { .length = PAGE_SIZE };
        struct fuse_io_args ia = {
                .ap.args.page_zeroing = true,
                .ap.args.out_pages = true,
                .ap.num_pages = 1,
                .ap.pages = &page,
                .ap.descs = &desc,
        };
        ssize_t res;
        u64 attr_ver;

        /*
         * Page writeback can extend beyond the lifetime of the
         * page-cache page, so make sure we read a properly synced
         * page.
         */
        fuse_wait_on_page_writeback(inode, page->index);

        attr_ver = fuse_get_attr_version(fm->fc);

        /* Don't overflow end offset */
        if (pos + (desc.length - 1) == LLONG_MAX)
                desc.length--;

        fuse_read_args_fill(&ia, file, pos, desc.length, FUSE_READ);
        res = fuse_simple_request(fm, &ia.ap.args);
        if (res < 0)
                return res;
        /*
         * Short read means EOF.  If file size is larger, truncate it
         */
        if (res < desc.length)
                fuse_short_read(inode, attr_ver, res, &ia.ap);

        SetPageUptodate(page);

        return 0;
}

static int fuse_readpage(struct file *file, struct page *page)
{
        struct inode *inode = page->mapping->host;
        int err;

        err = -EIO;
        if (fuse_is_bad(inode))
                goto out;

        err = fuse_do_readpage(file, page);
        fuse_invalidate_atime(inode);
 out:
        unlock_page(page);
        return err;
}

static void fuse_readpages_end(struct fuse_mount *fm, struct fuse_args *args,
                               int err)
{
        int i;
        struct fuse_io_args *ia = container_of(args, typeof(*ia), ap.args);
        struct fuse_args_pages *ap = &ia->ap;
        size_t count = ia->read.in.size;
        size_t num_read = args->out_args[0].size;
        struct address_space *mapping = NULL;

        for (i = 0; mapping == NULL && i < ap->num_pages; i++)
                mapping = ap->pages[i]->mapping;

        if (mapping) {
                struct inode *inode = mapping->host;

                /*
                 * Short read means EOF. If file size is larger, truncate it
                 */
                if (!err && num_read < count)
                        fuse_short_read(inode, ia->read.attr_ver, num_read, ap);

                fuse_invalidate_atime(inode);
        }

        for (i = 0; i < ap->num_pages; i++) {
                struct page *page = ap->pages[i];

                if (!err)
                        SetPageUptodate(page);
                else
                        SetPageError(page);
                unlock_page(page);
                put_page(page);
        }
        if (ia->ff)
                fuse_file_put(ia->ff, false, false);

        fuse_io_free(ia);
}

static void fuse_send_readpages(struct fuse_io_args *ia, struct file *file)
{
        struct fuse_file *ff = file->private_data;
        struct fuse_mount *fm = ff->fm;
        struct fuse_args_pages *ap = &ia->ap;
        loff_t pos = page_offset(ap->pages[0]);
        size_t count = ap->num_pages << PAGE_SHIFT;
        ssize_t res;
        int err;

        ap->args.out_pages = true;
        ap->args.page_zeroing = true;
        ap->args.page_replace = true;

        /* Don't overflow end offset */
        if (pos + (count - 1) == LLONG_MAX) {
                count--;
                ap->descs[ap->num_pages - 1].length--;
        }
        WARN_ON((loff_t) (pos + count) < 0);

        fuse_read_args_fill(ia, file, pos, count, FUSE_READ);
        ia->read.attr_ver = fuse_get_attr_version(fm->fc);
        if (fm->fc->async_read) {
                ia->ff = fuse_file_get(ff);
                ap->args.end = fuse_readpages_end;
                err = fuse_simple_background(fm, &ap->args, GFP_KERNEL);
                if (!err)
                        return;
        } else {
                res = fuse_simple_request(fm, &ap->args);
                err = res < 0 ? res : 0;
        }
        fuse_readpages_end(fm, &ap->args, err);
}

static void fuse_readahead(struct readahead_control *rac)
{
        struct inode *inode = rac->mapping->host;
        struct fuse_conn *fc = get_fuse_conn(inode);
        unsigned int i, max_pages, nr_pages = 0;

        if (fuse_is_bad(inode))
                return;

        max_pages = min_t(unsigned int, fc->max_pages,
                        fc->max_read / PAGE_SIZE);

        for (;;) {
                struct fuse_io_args *ia;
                struct fuse_args_pages *ap;

                nr_pages = readahead_count(rac) - nr_pages;
                if (nr_pages > max_pages)
                        nr_pages = max_pages;
                if (nr_pages == 0)
                        break;
                ia = fuse_io_alloc(NULL, nr_pages);
                if (!ia)
                        return;
                ap = &ia->ap;
                nr_pages = __readahead_batch(rac, ap->pages, nr_pages);
                for (i = 0; i < nr_pages; i++) {
                        fuse_wait_on_page_writeback(inode,
                                                    readahead_index(rac) + i);
                        ap->descs[i].length = PAGE_SIZE;
                }
                ap->num_pages = nr_pages;
                fuse_send_readpages(ia, rac->file);
        }
}

static ssize_t fuse_cache_read_iter(struct kiocb *iocb, struct iov_iter *to)
{
        struct inode *inode = iocb->ki_filp->f_mapping->host;
        struct fuse_conn *fc = get_fuse_conn(inode);

        /*
         * In auto invalidate mode, always update attributes on read.

         * Otherwise, only update if we attempt to read past EOF (to ensure
         * i_size is up to date).
         */
        if (fc->auto_inval_data ||
            (iocb->ki_pos + iov_iter_count(to) > i_size_read(inode))) {
                int err;
                err = fuse_update_attributes(inode, iocb->ki_filp);
                if (err)
                        return err;
        }

        return generic_file_read_iter(iocb, to);
}

static void fuse_write_args_fill(struct fuse_io_args *ia, struct fuse_file *ff,
                                 loff_t pos, size_t count)
{
        struct fuse_args *args = &ia->ap.args;

        ia->write.in.fh = ff->fh;
        ia->write.in.offset = pos;
        ia->write.in.size = count;
        args->opcode = FUSE_WRITE;
        args->nodeid = ff->nodeid;
        args->in_numargs = 2;
        if (ff->fm->fc->minor < 9)
                args->in_args[0].size = FUSE_COMPAT_WRITE_IN_SIZE;
        else
                args->in_args[0].size = sizeof(ia->write.in);
        args->in_args[0].value = &ia->write.in;
        args->in_args[1].size = count;
        args->out_numargs = 1;
        args->out_args[0].size = sizeof(ia->write.out);
        args->out_args[0].value = &ia->write.out;
}

static unsigned int fuse_write_flags(struct kiocb *iocb)
{
        unsigned int flags = iocb->ki_filp->f_flags;

        if (iocb->ki_flags & IOCB_DSYNC)
                flags |= O_DSYNC;
        if (iocb->ki_flags & IOCB_SYNC)
                flags |= O_SYNC;

        return flags;
}

static ssize_t fuse_send_write(struct fuse_io_args *ia, loff_t pos,
                               size_t count, fl_owner_t owner)
{
        struct kiocb *iocb = ia->io->iocb;
        struct file *file = iocb->ki_filp;
        struct fuse_file *ff = file->private_data;
        struct fuse_mount *fm = ff->fm;
        struct fuse_write_in *inarg = &ia->write.in;
        ssize_t err;

        fuse_write_args_fill(ia, ff, pos, count);
        inarg->flags = fuse_write_flags(iocb);
        if (owner != NULL) {
                inarg->write_flags |= FUSE_WRITE_LOCKOWNER;
                inarg->lock_owner = fuse_lock_owner_id(fm->fc, owner);
        }

        if (ia->io->async)
                return fuse_async_req_send(fm, ia, count);

        err = fuse_simple_request(fm, &ia->ap.args);
        if (!err && ia->write.out.size > count)
                err = -EIO;

        return err ?: ia->write.out.size;
}

bool fuse_write_update_size(struct inode *inode, loff_t pos)
{
        struct fuse_conn *fc = get_fuse_conn(inode);
        struct fuse_inode *fi = get_fuse_inode(inode);
        bool ret = false;

        spin_lock(&fi->lock);
        fi->attr_version = atomic64_inc_return(&fc->attr_version);
        if (pos > inode->i_size) {
                i_size_write(inode, pos);
                ret = true;
        }
        spin_unlock(&fi->lock);

        return ret;
}

static ssize_t fuse_send_write_pages(struct fuse_io_args *ia,
                                     struct kiocb *iocb, struct inode *inode,
                                     loff_t pos, size_t count)
{
        struct fuse_args_pages *ap = &ia->ap;
        struct file *file = iocb->ki_filp;
        struct fuse_file *ff = file->private_data;
        struct fuse_mount *fm = ff->fm;
        unsigned int offset, i;
        int err;

        for (i = 0; i < ap->num_pages; i++)
                fuse_wait_on_page_writeback(inode, ap->pages[i]->index);

        fuse_write_args_fill(ia, ff, pos, count);
        ia->write.in.flags = fuse_write_flags(iocb);
        if (fm->fc->handle_killpriv_v2 && !capable(CAP_FSETID))
                ia->write.in.write_flags |= FUSE_WRITE_KILL_SUIDGID;

        err = fuse_simple_request(fm, &ap->args);
        if (!err && ia->write.out.size > count)
                err = -EIO;

        offset = ap->descs[0].offset;
        count = ia->write.out.size;
        for (i = 0; i < ap->num_pages; i++) {
                struct page *page = ap->pages[i];

                if (!err && !offset && count >= PAGE_SIZE)
                        SetPageUptodate(page);

                if (count > PAGE_SIZE - offset)
                        count -= PAGE_SIZE - offset;
                else
                        count = 0;
                offset = 0;

                unlock_page(page);
                put_page(page);
        }

        return err;
}

static ssize_t fuse_fill_write_pages(struct fuse_args_pages *ap,
                                     struct address_space *mapping,
                                     struct iov_iter *ii, loff_t pos,
                                     unsigned int max_pages)
{
        struct fuse_conn *fc = get_fuse_conn(mapping->host);
        unsigned offset = pos & (PAGE_SIZE - 1);
        size_t count = 0;
        int err;

        ap->args.in_pages = true;
        ap->descs[0].offset = offset;

        do {
                size_t tmp;
                struct page *page;
                pgoff_t index = pos >> PAGE_SHIFT;
                size_t bytes = min_t(size_t, PAGE_SIZE - offset,
                                     iov_iter_count(ii));

                bytes = min_t(size_t, bytes, fc->max_write - count);

 again:
                err = -EFAULT;
                if (iov_iter_fault_in_readable(ii, bytes))
                        break;

                err = -ENOMEM;
                page = grab_cache_page_write_begin(mapping, index, 0);
                if (!page)
                        break;

                if (mapping_writably_mapped(mapping))
                        flush_dcache_page(page);

                tmp = iov_iter_copy_from_user_atomic(page, ii, offset, bytes);
                flush_dcache_page(page);

                iov_iter_advance(ii, tmp);
                if (!tmp) {
                        unlock_page(page);
                        put_page(page);
                        bytes = min(bytes, iov_iter_single_seg_count(ii));
                        goto again;
                }

                err = 0;
                ap->pages[ap->num_pages] = page;
                ap->descs[ap->num_pages].length = tmp;
                ap->num_pages++;

                count += tmp;
                pos += tmp;
                offset += tmp;
                if (offset == PAGE_SIZE)
                        offset = 0;

                if (!fc->big_writes)
                        break;
        } while (iov_iter_count(ii) && count < fc->max_write &&
                 ap->num_pages < max_pages && offset == 0);

        return count > 0 ? count : err;
}

static inline unsigned int fuse_wr_pages(loff_t pos, size_t len,
                                     unsigned int max_pages)
{
        return min_t(unsigned int,
                     ((pos + len - 1) >> PAGE_SHIFT) -
                     (pos >> PAGE_SHIFT) + 1,
                     max_pages);
}

static ssize_t fuse_perform_write(struct kiocb *iocb,
                                  struct address_space *mapping,
                                  struct iov_iter *ii, loff_t pos)
{
        struct inode *inode = mapping->host;
        struct fuse_conn *fc = get_fuse_conn(inode);
        struct fuse_inode *fi = get_fuse_inode(inode);
        int err = 0;
        ssize_t res = 0;

        if (inode->i_size < pos + iov_iter_count(ii))
                set_bit(FUSE_I_SIZE_UNSTABLE, &fi->state);

        do {
                ssize_t count;
                struct fuse_io_args ia = {};
                struct fuse_args_pages *ap = &ia.ap;
                unsigned int nr_pages = fuse_wr_pages(pos, iov_iter_count(ii),
                                                      fc->max_pages);

                ap->pages = fuse_pages_alloc(nr_pages, GFP_KERNEL, &ap->descs);
                if (!ap->pages) {
                        err = -ENOMEM;
                        break;
                }

                count = fuse_fill_write_pages(ap, mapping, ii, pos, nr_pages);
                if (count <= 0) {
                        err = count;
                } else {
                        err = fuse_send_write_pages(&ia, iocb, inode,
                                                    pos, count);
                        if (!err) {
                                size_t num_written = ia.write.out.size;

                                res += num_written;
                                pos += num_written;

                                /* break out of the loop on short write */
                                if (num_written != count)
                                        err = -EIO;
                        }
                }
                kfree(ap->pages);
        } while (!err && iov_iter_count(ii));

        if (res > 0)
                fuse_write_update_size(inode, pos);

        clear_bit(FUSE_I_SIZE_UNSTABLE, &fi->state);
        fuse_invalidate_attr(inode);

        return res > 0 ? res : err;
}

static ssize_t fuse_cache_write_iter(struct kiocb *iocb, struct iov_iter *from)
{
        struct file *file = iocb->ki_filp;
        struct address_space *mapping = file->f_mapping;
        ssize_t written = 0;
        ssize_t written_buffered = 0;
        struct inode *inode = mapping->host;
        ssize_t err;
        struct fuse_conn *fc = get_fuse_conn(inode);
        loff_t endbyte = 0;

        if (fc->writeback_cache) {
                /* Update size (EOF optimization) and mode (SUID clearing) */
                err = fuse_update_attributes(mapping->host, file);
                if (err)
                        return err;

                if (fc->handle_killpriv_v2 &&
                    should_remove_suid(file_dentry(file))) {
                        goto writethrough;
                }

                return generic_file_write_iter(iocb, from);
        }

writethrough:
        inode_lock(inode);

        /* We can write back this queue in page reclaim */
        current->backing_dev_info = inode_to_bdi(inode);

        err = generic_write_checks(iocb, from);
        if (err <= 0)
                goto out;

        err = file_remove_privs(file);
        if (err)
                goto out;

        err = file_update_time(file);
        if (err)
                goto out;

        if (iocb->ki_flags & IOCB_DIRECT) {
                loff_t pos = iocb->ki_pos;
                written = generic_file_direct_write(iocb, from);
                if (written < 0 || !iov_iter_count(from))
                        goto out;

                pos += written;

                written_buffered = fuse_perform_write(iocb, mapping, from, pos);
                if (written_buffered < 0) {
                        err = written_buffered;
                        goto out;
                }
                endbyte = pos + written_buffered - 1;

                err = filemap_write_and_wait_range(file->f_mapping, pos,
                                                   endbyte);
                if (err)
                        goto out;

                invalidate_mapping_pages(file->f_mapping,
                                         pos >> PAGE_SHIFT,
                                         endbyte >> PAGE_SHIFT);

                written += written_buffered;
                iocb->ki_pos = pos + written_buffered;
        } else {
                written = fuse_perform_write(iocb, mapping, from, iocb->ki_pos);
                if (written >= 0)
                        iocb->ki_pos += written;
        }
out:
        current->backing_dev_info = NULL;
        inode_unlock(inode);
        if (written > 0)
                written = generic_write_sync(iocb, written);

        return written ? written : err;
}

static inline void fuse_page_descs_length_init(struct fuse_page_desc *descs,
                                               unsigned int index,
                                               unsigned int nr_pages)
{
        int i;

        for (i = index; i < index + nr_pages; i++)
                descs[i].length = PAGE_SIZE - descs[i].offset;
}

static inline unsigned long fuse_get_user_addr(const struct iov_iter *ii)
{
        return (unsigned long)ii->iov->iov_base + ii->iov_offset;
}

static inline size_t fuse_get_frag_size(const struct iov_iter *ii,
                                        size_t max_size)
{
        return min(iov_iter_single_seg_count(ii), max_size);
}

static int fuse_get_user_pages(struct fuse_args_pages *ap, struct iov_iter *ii,
                               size_t *nbytesp, int write,
                               unsigned int max_pages)
{
        size_t nbytes = 0;  /* # bytes already packed in req */
        ssize_t ret = 0;

        /* Special case for kernel I/O: can copy directly into the buffer */
        if (iov_iter_is_kvec(ii)) {
                unsigned long user_addr = fuse_get_user_addr(ii);
                size_t frag_size = fuse_get_frag_size(ii, *nbytesp);

                if (write)
                        ap->args.in_args[1].value = (void *) user_addr;
                else
                        ap->args.out_args[0].value = (void *) user_addr;

                iov_iter_advance(ii, frag_size);
                *nbytesp = frag_size;
                return 0;
        }

        while (nbytes < *nbytesp && ap->num_pages < max_pages) {
                unsigned npages;
                size_t start;
                ret = iov_iter_get_pages(ii, &ap->pages[ap->num_pages],
                                        *nbytesp - nbytes,
                                        max_pages - ap->num_pages,
                                        &start);
                if (ret < 0)
                        break;

                iov_iter_advance(ii, ret);
                nbytes += ret;

                ret += start;
                npages = (ret + PAGE_SIZE - 1) / PAGE_SIZE;

                ap->descs[ap->num_pages].offset = start;
                fuse_page_descs_length_init(ap->descs, ap->num_pages, npages);

                ap->num_pages += npages;
                ap->descs[ap->num_pages - 1].length -=
                        (PAGE_SIZE - ret) & (PAGE_SIZE - 1);
        }

        if (write)
                ap->args.in_pages = true;
        else
                ap->args.out_pages = true;

        *nbytesp = nbytes;

        return ret < 0 ? ret : 0;
}

ssize_t fuse_direct_io(struct fuse_io_priv *io, struct iov_iter *iter,
                       loff_t *ppos, int flags)
{
        int write = flags & FUSE_DIO_WRITE;
        int cuse = flags & FUSE_DIO_CUSE;
        struct file *file = io->iocb->ki_filp;
        struct inode *inode = file->f_mapping->host;
        struct fuse_file *ff = file->private_data;
        struct fuse_conn *fc = ff->fm->fc;
        size_t nmax = write ? fc->max_write : fc->max_read;
        loff_t pos = *ppos;
        size_t count = iov_iter_count(iter);
        pgoff_t idx_from = pos >> PAGE_SHIFT;
        pgoff_t idx_to = (pos + count - 1) >> PAGE_SHIFT;
        ssize_t res = 0;
        int err = 0;
        struct fuse_io_args *ia;
        unsigned int max_pages;

        max_pages = iov_iter_npages(iter, fc->max_pages);
        ia = fuse_io_alloc(io, max_pages);
        if (!ia)
                return -ENOMEM;

        ia->io = io;
        if (!cuse && fuse_range_is_writeback(inode, idx_from, idx_to)) {
                if (!write)
                        inode_lock(inode);
                fuse_sync_writes(inode);
                if (!write)
                        inode_unlock(inode);
        }

        io->should_dirty = !write && iter_is_iovec(iter);
        while (count) {
                ssize_t nres;
                fl_owner_t owner = current->files;
                size_t nbytes = min(count, nmax);

                err = fuse_get_user_pages(&ia->ap, iter, &nbytes, write,
                                          max_pages);
                if (err && !nbytes)
                        break;

                if (write) {
                        if (!capable(CAP_FSETID))
                                ia->write.in.write_flags |= FUSE_WRITE_KILL_SUIDGID;

                        nres = fuse_send_write(ia, pos, nbytes, owner);
                } else {
                        nres = fuse_send_read(ia, pos, nbytes, owner);
                }

                if (!io->async || nres < 0) {
                        fuse_release_user_pages(&ia->ap, io->should_dirty);
                        fuse_io_free(ia);
                }
                ia = NULL;
                if (nres < 0) {
                        iov_iter_revert(iter, nbytes);
                        err = nres;
                        break;
                }
                WARN_ON(nres > nbytes);

                count -= nres;
                res += nres;
                pos += nres;
                if (nres != nbytes) {
                        iov_iter_revert(iter, nbytes - nres);
                        break;
                }
                if (count) {
                        max_pages = iov_iter_npages(iter, fc->max_pages);
                        ia = fuse_io_alloc(io, max_pages);
                        if (!ia)
                                break;
                }
        }
        if (ia)
                fuse_io_free(ia);
        if (res > 0)
                *ppos = pos;

        return res > 0 ? res : err;
}
EXPORT_SYMBOL_GPL(fuse_direct_io);

static ssize_t __fuse_direct_read(struct fuse_io_priv *io,
                                  struct iov_iter *iter,
                                  loff_t *ppos)
{
        ssize_t res;
        struct inode *inode = file_inode(io->iocb->ki_filp);

        res = fuse_direct_io(io, iter, ppos, 0);

        fuse_invalidate_atime(inode);

        return res;
}

static ssize_t fuse_direct_IO(struct kiocb *iocb, struct iov_iter *iter);

static ssize_t fuse_direct_read_iter(struct kiocb *iocb, struct iov_iter *to)
{
        ssize_t res;

        if (!is_sync_kiocb(iocb) && iocb->ki_flags & IOCB_DIRECT) {
                res = fuse_direct_IO(iocb, to);
        } else {
                struct fuse_io_priv io = FUSE_IO_PRIV_SYNC(iocb);

                res = __fuse_direct_read(&io, to, &iocb->ki_pos);
        }

        return res;
}

static ssize_t fuse_direct_write_iter(struct kiocb *iocb, struct iov_iter *from)
{
        struct inode *inode = file_inode(iocb->ki_filp);
        struct fuse_io_priv io = FUSE_IO_PRIV_SYNC(iocb);
        ssize_t res;

        /* Don't allow parallel writes to the same file */
        inode_lock(inode);
        res = generic_write_checks(iocb, from);
        if (res > 0) {
                if (!is_sync_kiocb(iocb) && iocb->ki_flags & IOCB_DIRECT) {
                        res = fuse_direct_IO(iocb, from);
                } else {
                        res = fuse_direct_io(&io, from, &iocb->ki_pos,
                                             FUSE_DIO_WRITE);
                }
        }
        fuse_invalidate_attr(inode);
        if (res > 0)
                fuse_write_update_size(inode, iocb->ki_pos);
        inode_unlock(inode);

        return res;
}

static ssize_t fuse_file_read_iter(struct kiocb *iocb, struct iov_iter *to)
{
        struct file *file = iocb->ki_filp;
        struct fuse_file *ff = file->private_data;
        struct inode *inode = file_inode(file);

        if (fuse_is_bad(inode))
                return -EIO;

        if (FUSE_IS_DAX(inode))
                return fuse_dax_read_iter(iocb, to);

        if (!(ff->open_flags & FOPEN_DIRECT_IO))
                return fuse_cache_read_iter(iocb, to);
        else
                return fuse_direct_read_iter(iocb, to);
}

static ssize_t fuse_file_write_iter(struct kiocb *iocb, struct iov_iter *from)
{
        struct file *file = iocb->ki_filp;
        struct fuse_file *ff = file->private_data;
        struct inode *inode = file_inode(file);

        if (fuse_is_bad(inode))
                return -EIO;

        if (FUSE_IS_DAX(inode))
                return fuse_dax_write_iter(iocb, from);

        if (!(ff->open_flags & FOPEN_DIRECT_IO))
                return fuse_cache_write_iter(iocb, from);
        else
                return fuse_direct_write_iter(iocb, from);
}

static void fuse_writepage_free(struct fuse_writepage_args *wpa)
{
        struct fuse_args_pages *ap = &wpa->ia.ap;
        int i;

        for (i = 0; i < ap->num_pages; i++)
                __free_page(ap->pages[i]);

        if (wpa->ia.ff)
                fuse_file_put(wpa->ia.ff, false, false);

        kfree(ap->pages);
        kfree(wpa);
}

static void fuse_writepage_finish(struct fuse_mount *fm,
                                  struct fuse_writepage_args *wpa)
{
        struct fuse_args_pages *ap = &wpa->ia.ap;
        struct inode *inode = wpa->inode;
        struct fuse_inode *fi = get_fuse_inode(inode);
        struct backing_dev_info *bdi = inode_to_bdi(inode);
        int i;

        for (i = 0; i < ap->num_pages; i++) {
                dec_wb_stat(&bdi->wb, WB_WRITEBACK);
                dec_node_page_state(ap->pages[i], NR_WRITEBACK_TEMP);
                wb_writeout_inc(&bdi->wb);
        }
        wake_up(&fi->page_waitq);
}

/* Called under fi->lock, may release and reacquire it */
static void fuse_send_writepage(struct fuse_mount *fm,
                                struct fuse_writepage_args *wpa, loff_t size)
__releases(fi->lock)
__acquires(fi->lock)
{
        struct fuse_writepage_args *aux, *next;
        struct fuse_inode *fi = get_fuse_inode(wpa->inode);
        struct fuse_write_in *inarg = &wpa->ia.write.in;
        struct fuse_args *args = &wpa->ia.ap.args;
        __u64 data_size = wpa->ia.ap.num_pages * PAGE_SIZE;
        int err;

        fi->writectr++;
        if (inarg->offset + data_size <= size) {
                inarg->size = data_size;
        } else if (inarg->offset < size) {
                inarg->size = size - inarg->offset;
        } else {
                /* Got truncated off completely */
                goto out_free;
        }

        args->in_args[1].size = inarg->size;
        args->force = true;
        args->nocreds = true;

        err = fuse_simple_background(fm, args, GFP_ATOMIC);
        if (err == -ENOMEM) {
                spin_unlock(&fi->lock);
                err = fuse_simple_background(fm, args, GFP_NOFS | __GFP_NOFAIL);
                spin_lock(&fi->lock);
        }

        /* Fails on broken connection only */
        if (unlikely(err))
                goto out_free;

        return;

 out_free:
        fi->writectr--;
        rb_erase(&wpa->writepages_entry, &fi->writepages);
        fuse_writepage_finish(fm, wpa);
        spin_unlock(&fi->lock);

        /* After fuse_writepage_finish() aux request list is private */
        for (aux = wpa->next; aux; aux = next) {
                next = aux->next;
                aux->next = NULL;
                fuse_writepage_free(aux);
        }

        fuse_writepage_free(wpa);
        spin_lock(&fi->lock);
}

/*
 * If fi->writectr is positive (no truncate or fsync going on) send
 * all queued writepage requests.
 *
 * Called with fi->lock
 */
void fuse_flush_writepages(struct inode *inode)
__releases(fi->lock)
__acquires(fi->lock)
{
        struct fuse_mount *fm = get_fuse_mount(inode);
        struct fuse_inode *fi = get_fuse_inode(inode);
        loff_t crop = i_size_read(inode);
        struct fuse_writepage_args *wpa;

        while (fi->writectr >= 0 && !list_empty(&fi->queued_writes)) {
                wpa = list_entry(fi->queued_writes.next,
                                 struct fuse_writepage_args, queue_entry);
                list_del_init(&wpa->queue_entry);
                fuse_send_writepage(fm, wpa, crop);
        }
}

static struct fuse_writepage_args *fuse_insert_writeback(struct rb_root *root,
                                                struct fuse_writepage_args *wpa)
{
        pgoff_t idx_from = wpa->ia.write.in.offset >> PAGE_SHIFT;
        pgoff_t idx_to = idx_from + wpa->ia.ap.num_pages - 1;
        struct rb_node **p = &root->rb_node;
        struct rb_node  *parent = NULL;

        WARN_ON(!wpa->ia.ap.num_pages);
        while (*p) {
                struct fuse_writepage_args *curr;
                pgoff_t curr_index;

                parent = *p;
                curr = rb_entry(parent, struct fuse_writepage_args,
                                writepages_entry);
                WARN_ON(curr->inode != wpa->inode);
                curr_index = curr->ia.write.in.offset >> PAGE_SHIFT;

                if (idx_from >= curr_index + curr->ia.ap.num_pages)
                        p = &(*p)->rb_right;
                else if (idx_to < curr_index)
                        p = &(*p)->rb_left;
                else
                        return curr;
        }

        rb_link_node(&wpa->writepages_entry, parent, p);
        rb_insert_color(&wpa->writepages_entry, root);
        return NULL;
}

static void tree_insert(struct rb_root *root, struct fuse_writepage_args *wpa)
{
        WARN_ON(fuse_insert_writeback(root, wpa));
}

static void fuse_writepage_end(struct fuse_mount *fm, struct fuse_args *args,
                               int error)
{
        struct fuse_writepage_args *wpa =
                container_of(args, typeof(*wpa), ia.ap.args);
        struct inode *inode = wpa->inode;
        struct fuse_inode *fi = get_fuse_inode(inode);

        mapping_set_error(inode->i_mapping, error);
        spin_lock(&fi->lock);
        rb_erase(&wpa->writepages_entry, &fi->writepages);
        while (wpa->next) {
                struct fuse_mount *fm = get_fuse_mount(inode);
                struct fuse_write_in *inarg = &wpa->ia.write.in;
                struct fuse_writepage_args *next = wpa->next;

                wpa->next = next->next;
                next->next = NULL;
                next->ia.ff = fuse_file_get(wpa->ia.ff);
                tree_insert(&fi->writepages, next);

                /*
                 * Skip fuse_flush_writepages() to make it easy to crop requests
                 * based on primary request size.
                 *
                 * 1st case (trivial): there are no concurrent activities using
                 * fuse_set/release_nowrite.  Then we're on safe side because
                 * fuse_flush_writepages() would call fuse_send_writepage()
                 * anyway.
                 *
                 * 2nd case: someone called fuse_set_nowrite and it is waiting
                 * now for completion of all in-flight requests.  This happens
                 * rarely and no more than once per page, so this should be
                 * okay.
                 *
                 * 3rd case: someone (e.g. fuse_do_setattr()) is in the middle
                 * of fuse_set_nowrite..fuse_release_nowrite section.  The fact
                 * that fuse_set_nowrite returned implies that all in-flight
                 * requests were completed along with all of their secondary
                 * requests.  Further primary requests are blocked by negative
                 * writectr.  Hence there cannot be any in-flight requests and
                 * no invocations of fuse_writepage_end() while we're in
                 * fuse_set_nowrite..fuse_release_nowrite section.
                 */
                fuse_send_writepage(fm, next, inarg->offset + inarg->size);
        }
        fi->writectr--;
        fuse_writepage_finish(fm, wpa);
        spin_unlock(&fi->lock);
        fuse_writepage_free(wpa);
}

static struct fuse_file *__fuse_write_file_get(struct fuse_conn *fc,
                                               struct fuse_inode *fi)
{
        struct fuse_file *ff = NULL;

        spin_lock(&fi->lock);
        if (!list_empty(&fi->write_files)) {
                ff = list_entry(fi->write_files.next, struct fuse_file,
                                write_entry);
                fuse_file_get(ff);
        }
        spin_unlock(&fi->lock);

        return ff;
}

static struct fuse_file *fuse_write_file_get(struct fuse_conn *fc,
                                             struct fuse_inode *fi)
{
        struct fuse_file *ff = __fuse_write_file_get(fc, fi);
        WARN_ON(!ff);
        return ff;
}

int fuse_write_inode(struct inode *inode, struct writeback_control *wbc)
{
        struct fuse_conn *fc = get_fuse_conn(inode);
        struct fuse_inode *fi = get_fuse_inode(inode);
        struct fuse_file *ff;
        int err;

        ff = __fuse_write_file_get(fc, fi);
        err = fuse_flush_times(inode, ff);
        if (ff)
                fuse_file_put(ff, false, false);

        return err;
}

static struct fuse_writepage_args *fuse_writepage_args_alloc(void)
{
        struct fuse_writepage_args *wpa;
        struct fuse_args_pages *ap;

        wpa = kzalloc(sizeof(*wpa), GFP_NOFS);
        if (wpa) {
                ap = &wpa->ia.ap;
                ap->num_pages = 0;
                ap->pages = fuse_pages_alloc(1, GFP_NOFS, &ap->descs);
                if (!ap->pages) {
                        kfree(wpa);
                        wpa = NULL;
                }
        }
        return wpa;

}

static int fuse_writepage_locked(struct page *page)
{
        struct address_space *mapping = page->mapping;
        struct inode *inode = mapping->host;
        struct fuse_conn *fc = get_fuse_conn(inode);
        struct fuse_inode *fi = get_fuse_inode(inode);
        struct fuse_writepage_args *wpa;
        struct fuse_args_pages *ap;
        struct page *tmp_page;
        int error = -ENOMEM;

        set_page_writeback(page);

        wpa = fuse_writepage_args_alloc();
        if (!wpa)
                goto err;
        ap = &wpa->ia.ap;

        tmp_page = alloc_page(GFP_NOFS | __GFP_HIGHMEM);
        if (!tmp_page)
                goto err_free;

        error = -EIO;
        wpa->ia.ff = fuse_write_file_get(fc, fi);
        if (!wpa->ia.ff)
                goto err_nofile;

        fuse_write_args_fill(&wpa->ia, wpa->ia.ff, page_offset(page), 0);

        copy_highpage(tmp_page, page);
        wpa->ia.write.in.write_flags |= FUSE_WRITE_CACHE;
        wpa->next = NULL;
        ap->args.in_pages = true;
        ap->num_pages = 1;
        ap->pages[0] = tmp_page;
        ap->descs[0].offset = 0;
        ap->descs[0].length = PAGE_SIZE;
        ap->args.end = fuse_writepage_end;
        wpa->inode = inode;

        inc_wb_stat(&inode_to_bdi(inode)->wb, WB_WRITEBACK);
        inc_node_page_state(tmp_page, NR_WRITEBACK_TEMP);

        spin_lock(&fi->lock);
        tree_insert(&fi->writepages, wpa);
        list_add_tail(&wpa->queue_entry, &fi->queued_writes);
        fuse_flush_writepages(inode);
        spin_unlock(&fi->lock);

        end_page_writeback(page);

        return 0;

err_nofile:
        __free_page(tmp_page);
err_free:
        kfree(wpa);
err:
        mapping_set_error(page->mapping, error);
        end_page_writeback(page);
        return error;
}

static int fuse_writepage(struct page *page, struct writeback_control *wbc)
{
        int err;

        if (fuse_page_is_writeback(page->mapping->host, page->index)) {
                /*
                 * ->writepages() should be called for sync() and friends.  We
                 * should only get here on direct reclaim and then we are
                 * allowed to skip a page which is already in flight
                 */
                WARN_ON(wbc->sync_mode == WB_SYNC_ALL);

                redirty_page_for_writepage(wbc, page);
                unlock_page(page);
                return 0;
        }

        err = fuse_writepage_locked(page);
        unlock_page(page);

        return err;
}

struct fuse_fill_wb_data {
        struct fuse_writepage_args *wpa;
        struct fuse_file *ff;
        struct inode *inode;
        struct page **orig_pages;
        unsigned int max_pages;
};

static bool fuse_pages_realloc(struct fuse_fill_wb_data *data)
{
        struct fuse_args_pages *ap = &data->wpa->ia.ap;
        struct fuse_conn *fc = get_fuse_conn(data->inode);
        struct page **pages;
        struct fuse_page_desc *descs;
        unsigned int npages = min_t(unsigned int,
                                    max_t(unsigned int, data->max_pages * 2,
                                          FUSE_DEFAULT_MAX_PAGES_PER_REQ),
                                    fc->max_pages);
        WARN_ON(npages <= data->max_pages);

        pages = fuse_pages_alloc(npages, GFP_NOFS, &descs);
        if (!pages)
                return false;

        memcpy(pages, ap->pages, sizeof(struct page *) * ap->num_pages);
        memcpy(descs, ap->descs, sizeof(struct fuse_page_desc) * ap->num_pages);
        kfree(ap->pages);
        ap->pages = pages;
        ap->descs = descs;
        data->max_pages = npages;

        return true;
}

static void fuse_writepages_send(struct fuse_fill_wb_data *data)
{
        struct fuse_writepage_args *wpa = data->wpa;
        struct inode *inode = data->inode;
        struct fuse_inode *fi = get_fuse_inode(inode);
        int num_pages = wpa->ia.ap.num_pages;
        int i;

        wpa->ia.ff = fuse_file_get(data->ff);
        spin_lock(&fi->lock);
        list_add_tail(&wpa->queue_entry, &fi->queued_writes);
        fuse_flush_writepages(inode);
        spin_unlock(&fi->lock);

        for (i = 0; i < num_pages; i++)
                end_page_writeback(data->orig_pages[i]);

}

/*
 * Check under fi->lock if the page is under writeback, and insert it onto the
 * rb_tree if not. Otherwise iterate auxiliary write requests, to see if there's
 * one already added for a page at this offset.  If there's none, then insert
 * this new request onto the auxiliary list, otherwise reuse the existing one by
 * swapping the new temp page with the old one.
 */
static bool fuse_writepage_add(struct fuse_writepage_args *new_wpa,
                               struct page *page)
{
        struct fuse_inode *fi = get_fuse_inode(new_wpa->inode);
        struct fuse_writepage_args *tmp;
        struct fuse_writepage_args *old_wpa;
        struct fuse_args_pages *new_ap = &new_wpa->ia.ap;

        WARN_ON(new_ap->num_pages != 0);
        new_ap->num_pages = 1;

        spin_lock(&fi->lock);
        old_wpa = fuse_insert_writeback(&fi->writepages, new_wpa);
        if (!old_wpa) {
                spin_unlock(&fi->lock);
                return true;
        }

        for (tmp = old_wpa->next; tmp; tmp = tmp->next) {
                pgoff_t curr_index;

                WARN_ON(tmp->inode != new_wpa->inode);
                curr_index = tmp->ia.write.in.offset >> PAGE_SHIFT;
                if (curr_index == page->index) {
                        WARN_ON(tmp->ia.ap.num_pages != 1);
                        swap(tmp->ia.ap.pages[0], new_ap->pages[0]);
                        break;
                }
        }

        if (!tmp) {
                new_wpa->next = old_wpa->next;
                old_wpa->next = new_wpa;
        }

        spin_unlock(&fi->lock);

        if (tmp) {
                struct backing_dev_info *bdi = inode_to_bdi(new_wpa->inode);

                dec_wb_stat(&bdi->wb, WB_WRITEBACK);
                dec_node_page_state(new_ap->pages[0], NR_WRITEBACK_TEMP);
                wb_writeout_inc(&bdi->wb);
                fuse_writepage_free(new_wpa);
        }

        return false;
}

static bool fuse_writepage_need_send(struct fuse_conn *fc, struct page *page,
                                     struct fuse_args_pages *ap,
                                     struct fuse_fill_wb_data *data)
{
        WARN_ON(!ap->num_pages);

        /*
         * Being under writeback is unlikely but possible.  For example direct
         * read to an mmaped fuse file will set the page dirty twice; once when
         * the pages are faulted with get_user_pages(), and then after the read
         * completed.
         */
        if (fuse_page_is_writeback(data->inode, page->index))
                return true;

        /* Reached max pages */
        if (ap->num_pages == fc->max_pages)
                return true;

        /* Reached max write bytes */
        if ((ap->num_pages + 1) * PAGE_SIZE > fc->max_write)
                return true;

        /* Discontinuity */
        if (data->orig_pages[ap->num_pages - 1]->index + 1 != page->index)
                return true;

        /* Need to grow the pages array?  If so, did the expansion fail? */
        if (ap->num_pages == data->max_pages && !fuse_pages_realloc(data))
                return true;

        return false;
}

static int fuse_writepages_fill(struct page *page,
                struct writeback_control *wbc, void *_data)
{
        struct fuse_fill_wb_data *data = _data;
        struct fuse_writepage_args *wpa = data->wpa;
        struct fuse_args_pages *ap = &wpa->ia.ap;
        struct inode *inode = data->inode;
        struct fuse_inode *fi = get_fuse_inode(inode);
        struct fuse_conn *fc = get_fuse_conn(inode);
        struct page *tmp_page;
        int err;

        if (!data->ff) {
                err = -EIO;
                data->ff = fuse_write_file_get(fc, fi);
                if (!data->ff)
                        goto out_unlock;
        }

        if (wpa && fuse_writepage_need_send(fc, page, ap, data)) {
                fuse_writepages_send(data);
                data->wpa = NULL;
        }

        err = -ENOMEM;
        tmp_page = alloc_page(GFP_NOFS | __GFP_HIGHMEM);
        if (!tmp_page)
                goto out_unlock;

        /*
         * The page must not be redirtied until the writeout is completed
         * (i.e. userspace has sent a reply to the write request).  Otherwise
         * there could be more than one temporary page instance for each real
         * page.
         *
         * This is ensured by holding the page lock in page_mkwrite() while
         * checking fuse_page_is_writeback().  We already hold the page lock
         * since clear_page_dirty_for_io() and keep it held until we add the
         * request to the fi->writepages list and increment ap->num_pages.
         * After this fuse_page_is_writeback() will indicate that the page is
         * under writeback, so we can release the page lock.
         */
        if (data->wpa == NULL) {
                err = -ENOMEM;
                wpa = fuse_writepage_args_alloc();
                if (!wpa) {
                        __free_page(tmp_page);
                        goto out_unlock;
                }
                data->max_pages = 1;

                ap = &wpa->ia.ap;
                fuse_write_args_fill(&wpa->ia, data->ff, page_offset(page), 0);
                wpa->ia.write.in.write_flags |= FUSE_WRITE_CACHE;
                wpa->next = NULL;
                ap->args.in_pages = true;
                ap->args.end = fuse_writepage_end;
                ap->num_pages = 0;
                wpa->inode = inode;
        }
        set_page_writeback(page);

        copy_highpage(tmp_page, page);
        ap->pages[ap->num_pages] = tmp_page;
        ap->descs[ap->num_pages].offset = 0;
        ap->descs[ap->num_pages].length = PAGE_SIZE;
        data->orig_pages[ap->num_pages] = page;

        inc_wb_stat(&inode_to_bdi(inode)->wb, WB_WRITEBACK);
        inc_node_page_state(tmp_page, NR_WRITEBACK_TEMP);

        err = 0;
        if (data->wpa) {
                /*
                 * Protected by fi->lock against concurrent access by
                 * fuse_page_is_writeback().
                 */
                spin_lock(&fi->lock);
                ap->num_pages++;
                spin_unlock(&fi->lock);
        } else if (fuse_writepage_add(wpa, page)) {
                data->wpa = wpa;
        } else {
                end_page_writeback(page);
        }
out_unlock:
        unlock_page(page);

        return err;
}

static int fuse_writepages(struct address_space *mapping,
                           struct writeback_control *wbc)
{
        struct inode *inode = mapping->host;
        struct fuse_conn *fc = get_fuse_conn(inode);
        struct fuse_fill_wb_data data;
        int err;

        err = -EIO;
        if (fuse_is_bad(inode))
                goto out;

        data.inode = inode;
        data.wpa = NULL;
        data.ff = NULL;

        err = -ENOMEM;
        data.orig_pages = kcalloc(fc->max_pages,
                                  sizeof(struct page *),
                                  GFP_NOFS);
        if (!data.orig_pages)
                goto out;

        err = write_cache_pages(mapping, wbc, fuse_writepages_fill, &data);
        if (data.wpa) {
                WARN_ON(!data.wpa->ia.ap.num_pages);
                fuse_writepages_send(&data);
        }
        if (data.ff)
                fuse_file_put(data.ff, false, false);

        kfree(data.orig_pages);
out:
        return err;
}

/*
 * It's worthy to make sure that space is reserved on disk for the write,
 * but how to implement it without killing performance need more thinking.
 */
static int fuse_write_begin(struct file *file, struct address_space *mapping,
                loff_t pos, unsigned len, unsigned flags,
                struct page **pagep, void **fsdata)
{
        pgoff_t index = pos >> PAGE_SHIFT;
        struct fuse_conn *fc = get_fuse_conn(file_inode(file));
        struct page *page;
        loff_t fsize;
        int err = -ENOMEM;

        WARN_ON(!fc->writeback_cache);

        page = grab_cache_page_write_begin(mapping, index, flags);
        if (!page)
                goto error;

        fuse_wait_on_page_writeback(mapping->host, page->index);

        if (PageUptodate(page) || len == PAGE_SIZE)
                goto success;
        /*
         * Check if the start this page comes after the end of file, in which
         * case the readpage can be optimized away.
         */
        fsize = i_size_read(mapping->host);
        if (fsize <= (pos & PAGE_MASK)) {
                size_t off = pos & ~PAGE_MASK;
                if (off)
                        zero_user_segment(page, 0, off);
                goto success;
        }
        err = fuse_do_readpage(file, page);
        if (err)
                goto cleanup;
success:
        *pagep = page;
        return 0;

cleanup:
        unlock_page(page);
        put_page(page);
error:
        return err;
}

static int fuse_write_end(struct file *file, struct address_space *mapping,
                loff_t pos, unsigned len, unsigned copied,
                struct page *page, void *fsdata)
{
        struct inode *inode = page->mapping->host;

        /* Haven't copied anything?  Skip zeroing, size extending, dirtying. */
        if (!copied)
                goto unlock;

        if (!PageUptodate(page)) {
                /* Zero any unwritten bytes at the end of the page */
                size_t endoff = (pos + copied) & ~PAGE_MASK;
                if (endoff)
                        zero_user_segment(page, endoff, PAGE_SIZE);
                SetPageUptodate(page);
        }

        fuse_write_update_size(inode, pos + copied);
        set_page_dirty(page);

unlock:
        unlock_page(page);
        put_page(page);

        return copied;
}

static int fuse_launder_page(struct page *page)
{
        int err = 0;
        if (clear_page_dirty_for_io(page)) {
                struct inode *inode = page->mapping->host;

                /* Serialize with pending writeback for the same page */
                fuse_wait_on_page_writeback(inode, page->index);
                err = fuse_writepage_locked(page);
                if (!err)
                        fuse_wait_on_page_writeback(inode, page->index);
        }
        return err;
}

/*
 * Write back dirty pages now, because there may not be any suitable
 * open files later
 */
static void fuse_vma_close(struct vm_area_struct *vma)
{
        filemap_write_and_wait(vma->vm_file->f_mapping);
}

/*
 * Wait for writeback against this page to complete before allowing it
 * to be marked dirty again, and hence written back again, possibly
 * before the previous writepage completed.
 *
 * Block here, instead of in ->writepage(), so that the userspace fs
 * can only block processes actually operating on the filesystem.
 *
 * Otherwise unprivileged userspace fs would be able to block
 * unrelated:
 *
 * - page migration
 * - sync(2)
 * - try_to_free_pages() with order > PAGE_ALLOC_COSTLY_ORDER
 */
static vm_fault_t fuse_page_mkwrite(struct vm_fault *vmf)
{
        struct page *page = vmf->page;
        struct inode *inode = file_inode(vmf->vma->vm_file);

        file_update_time(vmf->vma->vm_file);
        lock_page(page);
        if (page->mapping != inode->i_mapping) {
                unlock_page(page);
                return VM_FAULT_NOPAGE;
        }

        fuse_wait_on_page_writeback(inode, page->index);
        return VM_FAULT_LOCKED;
}

static const struct vm_operations_struct fuse_file_vm_ops = {
        .close          = fuse_vma_close,
        .fault          = filemap_fault,
        .map_pages      = filemap_map_pages,
        .page_mkwrite   = fuse_page_mkwrite,
};

static int fuse_file_mmap(struct file *file, struct vm_area_struct *vma)
{
        struct fuse_file *ff = file->private_data;

        /* DAX mmap is superior to direct_io mmap */
        if (FUSE_IS_DAX(file_inode(file)))
                return fuse_dax_mmap(file, vma);

        if (ff->open_flags & FOPEN_DIRECT_IO) {
                /* Can't provide the coherency needed for MAP_SHARED */
                if (vma->vm_flags & VM_MAYSHARE)
                        return -ENODEV;

                invalidate_inode_pages2(file->f_mapping);

                return generic_file_mmap(file, vma);
        }

        if ((vma->vm_flags & VM_SHARED) && (vma->vm_flags & VM_MAYWRITE))
                fuse_link_write_file(file);

        file_accessed(file);
        vma->vm_ops = &fuse_file_vm_ops;
        return 0;
}

static int convert_fuse_file_lock(struct fuse_conn *fc,
                                  const struct fuse_file_lock *ffl,
                                  struct file_lock *fl)
{
        switch (ffl->type) {
        case F_UNLCK:
                break;

        case F_RDLCK:
        case F_WRLCK:
                if (ffl->start > OFFSET_MAX || ffl->end > OFFSET_MAX ||
                    ffl->end < ffl->start)
                        return -EIO;

                fl->fl_start = ffl->start;
                fl->fl_end = ffl->end;

                /*
                 * Convert pid into init's pid namespace.  The locks API will
                 * translate it into the caller's pid namespace.
                 */
                rcu_read_lock();
                fl->fl_pid = pid_nr_ns(find_pid_ns(ffl->pid, fc->pid_ns), &init_pid_ns);
                rcu_read_unlock();
                break;

        default:
                return -EIO;
        }
        fl->fl_type = ffl->type;
        return 0;
}

static void fuse_lk_fill(struct fuse_args *args, struct file *file,
                         const struct file_lock *fl, int opcode, pid_t pid,
                         int flock, struct fuse_lk_in *inarg)
{
        struct inode *inode = file_inode(file);
        struct fuse_conn *fc = get_fuse_conn(inode);
        struct fuse_file *ff = file->private_data;

        memset(inarg, 0, sizeof(*inarg));
        inarg->fh = ff->fh;
        inarg->owner = fuse_lock_owner_id(fc, fl->fl_owner);
        inarg->lk.start = fl->fl_start;
        inarg->lk.end = fl->fl_end;
        inarg->lk.type = fl->fl_type;
        inarg->lk.pid = pid;
        if (flock)
                inarg->lk_flags |= FUSE_LK_FLOCK;
        args->opcode = opcode;
        args->nodeid = get_node_id(inode);
        args->in_numargs = 1;
        args->in_args[0].size = sizeof(*inarg);
        args->in_args[0].value = inarg;
}

static int fuse_getlk(struct file *file, struct file_lock *fl)
{
        struct inode *inode = file_inode(file);
        struct fuse_mount *fm = get_fuse_mount(inode);
        FUSE_ARGS(args);
        struct fuse_lk_in inarg;
        struct fuse_lk_out outarg;
        int err;

        fuse_lk_fill(&args, file, fl, FUSE_GETLK, 0, 0, &inarg);
        args.out_numargs = 1;
        args.out_args[0].size = sizeof(outarg);
        args.out_args[0].value = &outarg;
        err = fuse_simple_request(fm, &args);
        if (!err)
                err = convert_fuse_file_lock(fm->fc, &outarg.lk, fl);

        return err;
}

static int fuse_setlk(struct file *file, struct file_lock *fl, int flock)
{
        struct inode *inode = file_inode(file);
        struct fuse_mount *fm = get_fuse_mount(inode);
        FUSE_ARGS(args);
        struct fuse_lk_in inarg;
        int opcode = (fl->fl_flags & FL_SLEEP) ? FUSE_SETLKW : FUSE_SETLK;
        struct pid *pid = fl->fl_type != F_UNLCK ? task_tgid(current) : NULL;
        pid_t pid_nr = pid_nr_ns(pid, fm->fc->pid_ns);
        int err;

        if (fl->fl_lmops && fl->fl_lmops->lm_grant) {
                /* NLM needs asynchronous locks, which we don't support yet */
                return -ENOLCK;
        }

        /* Unlock on close is handled by the flush method */
        if ((fl->fl_flags & FL_CLOSE_POSIX) == FL_CLOSE_POSIX)
                return 0;

        fuse_lk_fill(&args, file, fl, opcode, pid_nr, flock, &inarg);
        err = fuse_simple_request(fm, &args);

        /* locking is restartable */
        if (err == -EINTR)
                err = -ERESTARTSYS;

        return err;
}

static int fuse_file_lock(struct file *file, int cmd, struct file_lock *fl)
{
        struct inode *inode = file_inode(file);
        struct fuse_conn *fc = get_fuse_conn(inode);
        int err;

        if (cmd == F_CANCELLK) {
                err = 0;
        } else if (cmd == F_GETLK) {
                if (fc->no_lock) {
                        posix_test_lock(file, fl);
                        err = 0;
                } else
                        err = fuse_getlk(file, fl);
        } else {
                if (fc->no_lock)
                        err = posix_lock_file(file, fl, NULL);
                else
                        err = fuse_setlk(file, fl, 0);
        }
        return err;
}

static int fuse_file_flock(struct file *file, int cmd, struct file_lock *fl)
{
        struct inode *inode = file_inode(file);
        struct fuse_conn *fc = get_fuse_conn(inode);
        int err;

        if (fc->no_flock) {
                err = locks_lock_file_wait(file, fl);
        } else {
                struct fuse_file *ff = file->private_data;

                /* emulate flock with POSIX locks */
                ff->flock = true;
                err = fuse_setlk(file, fl, 1);
        }

        return err;
}

static sector_t fuse_bmap(struct address_space *mapping, sector_t block)
{
        struct inode *inode = mapping->host;
        struct fuse_mount *fm = get_fuse_mount(inode);
        FUSE_ARGS(args);
        struct fuse_bmap_in inarg;
        struct fuse_bmap_out outarg;
        int err;

        if (!inode->i_sb->s_bdev || fm->fc->no_bmap)
                return 0;

        memset(&inarg, 0, sizeof(inarg));
        inarg.block = block;
        inarg.blocksize = inode->i_sb->s_blocksize;
        args.opcode = FUSE_BMAP;
        args.nodeid = get_node_id(inode);
        args.in_numargs = 1;
        args.in_args[0].size = sizeof(inarg);
        args.in_args[0].value = &inarg;
        args.out_numargs = 1;
        args.out_args[0].size = sizeof(outarg);
        args.out_args[0].value = &outarg;
        err = fuse_simple_request(fm, &args);
        if (err == -ENOSYS)
                fm->fc->no_bmap = 1;

        return err ? 0 : outarg.block;
}

static loff_t fuse_lseek(struct file *file, loff_t offset, int whence)
{
        struct inode *inode = file->f_mapping->host;
        struct fuse_mount *fm = get_fuse_mount(inode);
        struct fuse_file *ff = file->private_data;
        FUSE_ARGS(args);
        struct fuse_lseek_in inarg = {
                .fh = ff->fh,
                .offset = offset,
                .whence = whence
        };
        struct fuse_lseek_out outarg;
        int err;

        if (fm->fc->no_lseek)
                goto fallback;

        args.opcode = FUSE_LSEEK;
        args.nodeid = ff->nodeid;
        args.in_numargs = 1;
        args.in_args[0].size = sizeof(inarg);
        args.in_args[0].value = &inarg;
        args.out_numargs = 1;
        args.out_args[0].size = sizeof(outarg);
        args.out_args[0].value = &outarg;
        err = fuse_simple_request(fm, &args);
        if (err) {
                if (err == -ENOSYS) {
                        fm->fc->no_lseek = 1;
                        goto fallback;
                }
                return err;
        }

        return vfs_setpos(file, outarg.offset, inode->i_sb->s_maxbytes);

fallback:
        err = fuse_update_attributes(inode, file);
        if (!err)
                return generic_file_llseek(file, offset, whence);
        else
                return err;
}

static loff_t fuse_file_llseek(struct file *file, loff_t offset, int whence)
{
        loff_t retval;
        struct inode *inode = file_inode(file);

        switch (whence) {
        case SEEK_SET:
        case SEEK_CUR:
                 /* No i_mutex protection necessary for SEEK_CUR and SEEK_SET */
                retval = generic_file_llseek(file, offset, whence);
                break;
        case SEEK_END:
                inode_lock(inode);
                retval = fuse_update_attributes(inode, file);
                if (!retval)
                        retval = generic_file_llseek(file, offset, whence);
                inode_unlock(inode);
                break;
        case SEEK_HOLE:
        case SEEK_DATA:
                inode_lock(inode);
                retval = fuse_lseek(file, offset, whence);
                inode_unlock(inode);
                break;
        default:
                retval = -EINVAL;
        }

        return retval;
}

/*
 * CUSE servers compiled on 32bit broke on 64bit kernels because the
 * ABI was defined to be 'struct iovec' which is different on 32bit
 * and 64bit.  Fortunately we can determine which structure the server
 * used from the size of the reply.
 */
static int fuse_copy_ioctl_iovec_old(struct iovec *dst, void *src,
                                     size_t transferred, unsigned count,
                                     bool is_compat)
{
#ifdef CONFIG_COMPAT
        if (count * sizeof(struct compat_iovec) == transferred) {
                struct compat_iovec *ciov = src;
                unsigned i;

                /*
                 * With this interface a 32bit server cannot support
                 * non-compat (i.e. ones coming from 64bit apps) ioctl
                 * requests
                 */
                if (!is_compat)
                        return -EINVAL;

                for (i = 0; i < count; i++) {
                        dst[i].iov_base = compat_ptr(ciov[i].iov_base);
                        dst[i].iov_len = ciov[i].iov_len;
                }
                return 0;
        }
#endif

        if (count * sizeof(struct iovec) != transferred)
                return -EIO;

        memcpy(dst, src, transferred);
        return 0;
}

/* Make sure iov_length() won't overflow */
static int fuse_verify_ioctl_iov(struct fuse_conn *fc, struct iovec *iov,
                                 size_t count)
{
        size_t n;
        u32 max = fc->max_pages << PAGE_SHIFT;

        for (n = 0; n < count; n++, iov++) {
                if (iov->iov_len > (size_t) max)
                        return -ENOMEM;
                max -= iov->iov_len;
        }
        return 0;
}

static int fuse_copy_ioctl_iovec(struct fuse_conn *fc, struct iovec *dst,
                                 void *src, size_t transferred, unsigned count,
                                 bool is_compat)
{
        unsigned i;
        struct fuse_ioctl_iovec *fiov = src;

        if (fc->minor < 16) {
                return fuse_copy_ioctl_iovec_old(dst, src, transferred,
                                                 count, is_compat);
        }

        if (count * sizeof(struct fuse_ioctl_iovec) != transferred)
                return -EIO;

        for (i = 0; i < count; i++) {
                /* Did the server supply an inappropriate value? */
                if (fiov[i].base != (unsigned long) fiov[i].base ||
                    fiov[i].len != (unsigned long) fiov[i].len)
                        return -EIO;

                dst[i].iov_base = (void __user *) (unsigned long) fiov[i].base;
                dst[i].iov_len = (size_t) fiov[i].len;

#ifdef CONFIG_COMPAT
                if (is_compat &&
                    (ptr_to_compat(dst[i].iov_base) != fiov[i].base ||
                     (compat_size_t) dst[i].iov_len != fiov[i].len))
                        return -EIO;
#endif
        }

        return 0;
}


/*
 * For ioctls, there is no generic way to determine how much memory
 * needs to be read and/or written.  Furthermore, ioctls are allowed
 * to dereference the passed pointer, so the parameter requires deep
 * copying but FUSE has no idea whatsoever about what to copy in or
 * out.
 *
 * This is solved by allowing FUSE server to retry ioctl with
 * necessary in/out iovecs.  Let's assume the ioctl implementation
 * needs to read in the following structure.
 *
 * struct a {
 *      char    *buf;
 *      size_t  buflen;
 * }
 *
 * On the first callout to FUSE server, inarg->in_size and
 * inarg->out_size will be NULL; then, the server completes the ioctl
 * with FUSE_IOCTL_RETRY set in out->flags, out->in_iovs set to 1 and
 * the actual iov array to
 *
 * { { .iov_base = inarg.arg,   .iov_len = sizeof(struct a) } }
 *
 * which tells FUSE to copy in the requested area and retry the ioctl.
 * On the second round, the server has access to the structure and
 * from that it can tell what to look for next, so on the invocation,
 * it sets FUSE_IOCTL_RETRY, out->in_iovs to 2 and iov array to
 *
 * { { .iov_base = inarg.arg,   .iov_len = sizeof(struct a)     },
 *   { .iov_base = a.buf,       .iov_len = a.buflen             } }
 *
 * FUSE will copy both struct a and the pointed buffer from the
 * process doing the ioctl and retry ioctl with both struct a and the
 * buffer.
 *
 * This time, FUSE server has everything it needs and completes ioctl
 * without FUSE_IOCTL_RETRY which finishes the ioctl call.
 *
 * Copying data out works the same way.
 *
 * Note that if FUSE_IOCTL_UNRESTRICTED is clear, the kernel
 * automatically initializes in and out iovs by decoding @cmd with
 * _IOC_* macros and the server is not allowed to request RETRY.  This
 * limits ioctl data transfers to well-formed ioctls and is the forced
 * behavior for all FUSE servers.
 */
long fuse_do_ioctl(struct file *file, unsigned int cmd, unsigned long arg,
                   unsigned int flags)
{
        struct fuse_file *ff = file->private_data;
        struct fuse_mount *fm = ff->fm;
        struct fuse_ioctl_in inarg = {
                .fh = ff->fh,
                .cmd = cmd,
                .arg = arg,
                .flags = flags
        };
        struct fuse_ioctl_out outarg;
        struct iovec *iov_page = NULL;
        struct iovec *in_iov = NULL, *out_iov = NULL;
        unsigned int in_iovs = 0, out_iovs = 0, max_pages;
        size_t in_size, out_size, c;
        ssize_t transferred;
        int err, i;
        struct iov_iter ii;
        struct fuse_args_pages ap = {};

#if BITS_PER_LONG == 32
        inarg.flags |= FUSE_IOCTL_32BIT;
#else
        if (flags & FUSE_IOCTL_COMPAT) {
                inarg.flags |= FUSE_IOCTL_32BIT;
#ifdef CONFIG_X86_X32
                if (in_x32_syscall())
                        inarg.flags |= FUSE_IOCTL_COMPAT_X32;
#endif
        }
#endif

        /* assume all the iovs returned by client always fits in a page */
        BUILD_BUG_ON(sizeof(struct fuse_ioctl_iovec) * FUSE_IOCTL_MAX_IOV > PAGE_SIZE);

        err = -ENOMEM;
        ap.pages = fuse_pages_alloc(fm->fc->max_pages, GFP_KERNEL, &ap.descs);
        iov_page = (struct iovec *) __get_free_page(GFP_KERNEL);
        if (!ap.pages || !iov_page)
                goto out;

        fuse_page_descs_length_init(ap.descs, 0, fm->fc->max_pages);

        /*
         * If restricted, initialize IO parameters as encoded in @cmd.
         * RETRY from server is not allowed.
         */
        if (!(flags & FUSE_IOCTL_UNRESTRICTED)) {
                struct iovec *iov = iov_page;

                iov->iov_base = (void __user *)arg;

                switch (cmd) {
                case FS_IOC_GETFLAGS:
                case FS_IOC_SETFLAGS:
                        iov->iov_len = sizeof(int);
                        break;
                default:
                        iov->iov_len = _IOC_SIZE(cmd);
                        break;
                }

                if (_IOC_DIR(cmd) & _IOC_WRITE) {
                        in_iov = iov;
                        in_iovs = 1;
                }

                if (_IOC_DIR(cmd) & _IOC_READ) {
                        out_iov = iov;
                        out_iovs = 1;
                }
        }

 retry:
        inarg.in_size = in_size = iov_length(in_iov, in_iovs);
        inarg.out_size = out_size = iov_length(out_iov, out_iovs);

        /*
         * Out data can be used either for actual out data or iovs,
         * make sure there always is at least one page.
         */
        out_size = max_t(size_t, out_size, PAGE_SIZE);
        max_pages = DIV_ROUND_UP(max(in_size, out_size), PAGE_SIZE);

        /* make sure there are enough buffer pages and init request with them */
        err = -ENOMEM;
        if (max_pages > fm->fc->max_pages)
                goto out;
        while (ap.num_pages < max_pages) {
                ap.pages[ap.num_pages] = alloc_page(GFP_KERNEL | __GFP_HIGHMEM);
                if (!ap.pages[ap.num_pages])
                        goto out;
                ap.num_pages++;
        }


        /* okay, let's send it to the client */
        ap.args.opcode = FUSE_IOCTL;
        ap.args.nodeid = ff->nodeid;
        ap.args.in_numargs = 1;
        ap.args.in_args[0].size = sizeof(inarg);
        ap.args.in_args[0].value = &inarg;
        if (in_size) {
                ap.args.in_numargs++;
                ap.args.in_args[1].size = in_size;
                ap.args.in_pages = true;

                err = -EFAULT;
                iov_iter_init(&ii, WRITE, in_iov, in_iovs, in_size);
                for (i = 0; iov_iter_count(&ii) && !WARN_ON(i >= ap.num_pages); i++) {
                        c = copy_page_from_iter(ap.pages[i], 0, PAGE_SIZE, &ii);
                        if (c != PAGE_SIZE && iov_iter_count(&ii))
                                goto out;
                }
        }

        ap.args.out_numargs = 2;
        ap.args.out_args[0].size = sizeof(outarg);
        ap.args.out_args[0].value = &outarg;
        ap.args.out_args[1].size = out_size;
        ap.args.out_pages = true;
        ap.args.out_argvar = true;

        transferred = fuse_simple_request(fm, &ap.args);
        err = transferred;
        if (transferred < 0)
                goto out;

        /* did it ask for retry? */
        if (outarg.flags & FUSE_IOCTL_RETRY) {
                void *vaddr;

                /* no retry if in restricted mode */
                err = -EIO;
                if (!(flags & FUSE_IOCTL_UNRESTRICTED))
                        goto out;

                in_iovs = outarg.in_iovs;
                out_iovs = outarg.out_iovs;

                /*
                 * Make sure things are in boundary, separate checks
                 * are to protect against overflow.
                 */
                err = -ENOMEM;
                if (in_iovs > FUSE_IOCTL_MAX_IOV ||
                    out_iovs > FUSE_IOCTL_MAX_IOV ||
                    in_iovs + out_iovs > FUSE_IOCTL_MAX_IOV)
                        goto out;

                vaddr = kmap_atomic(ap.pages[0]);
                err = fuse_copy_ioctl_iovec(fm->fc, iov_page, vaddr,
                                            transferred, in_iovs + out_iovs,
                                            (flags & FUSE_IOCTL_COMPAT) != 0);
                kunmap_atomic(vaddr);
                if (err)
                        goto out;

                in_iov = iov_page;
                out_iov = in_iov + in_iovs;

                err = fuse_verify_ioctl_iov(fm->fc, in_iov, in_iovs);
                if (err)
                        goto out;

                err = fuse_verify_ioctl_iov(fm->fc, out_iov, out_iovs);
                if (err)
                        goto out;

                goto retry;
        }

        err = -EIO;
        if (transferred > inarg.out_size)
                goto out;

        err = -EFAULT;
        iov_iter_init(&ii, READ, out_iov, out_iovs, transferred);
        for (i = 0; iov_iter_count(&ii) && !WARN_ON(i >= ap.num_pages); i++) {
                c = copy_page_to_iter(ap.pages[i], 0, PAGE_SIZE, &ii);
                if (c != PAGE_SIZE && iov_iter_count(&ii))
                        goto out;
        }
        err = 0;
 out:
        free_page((unsigned long) iov_page);
        while (ap.num_pages)
                __free_page(ap.pages[--ap.num_pages]);
        kfree(ap.pages);

        return err ? err : outarg.result;
}
EXPORT_SYMBOL_GPL(fuse_do_ioctl);

long fuse_ioctl_common(struct file *file, unsigned int cmd,
                       unsigned long arg, unsigned int flags)
{
        struct inode *inode = file_inode(file);
        struct fuse_conn *fc = get_fuse_conn(inode);

        if (!fuse_allow_current_process(fc))
                return -EACCES;

        if (fuse_is_bad(inode))
                return -EIO;

        return fuse_do_ioctl(file, cmd, arg, flags);
}

static long fuse_file_ioctl(struct file *file, unsigned int cmd,
                            unsigned long arg)
{
        return fuse_ioctl_common(file, cmd, arg, 0);
}

static long fuse_file_compat_ioctl(struct file *file, unsigned int cmd,
                                   unsigned long arg)
{
        return fuse_ioctl_common(file, cmd, arg, FUSE_IOCTL_COMPAT);
}

/*
 * All files which have been polled are linked to RB tree
 * fuse_conn->polled_files which is indexed by kh.  Walk the tree and
 * find the matching one.
 */