diff options
55 files changed, 691 insertions, 861 deletions
diff --git a/Documentation/filesystems/Locking b/Documentation/filesystems/Locking index 7c3f187d..0a926e2 100644 --- a/Documentation/filesystems/Locking +++ b/Documentation/filesystems/Locking @@ -196,7 +196,7 @@ prototypes: void (*invalidatepage) (struct page *, unsigned int, unsigned int); int (*releasepage) (struct page *, int); void (*freepage)(struct page *); - int (*direct_IO)(int, struct kiocb *, struct iov_iter *iter, loff_t offset); + int (*direct_IO)(struct kiocb *, struct iov_iter *iter, loff_t offset); int (*migratepage)(struct address_space *, struct page *, struct page *); int (*launder_page)(struct page *); int (*is_partially_uptodate)(struct page *, unsigned long, unsigned long); diff --git a/Documentation/filesystems/vfs.txt b/Documentation/filesystems/vfs.txt index 207cdca..5d833b3 100644 --- a/Documentation/filesystems/vfs.txt +++ b/Documentation/filesystems/vfs.txt @@ -590,7 +590,7 @@ struct address_space_operations { void (*invalidatepage) (struct page *, unsigned int, unsigned int); int (*releasepage) (struct page *, int); void (*freepage)(struct page *); - ssize_t (*direct_IO)(int, struct kiocb *, struct iov_iter *iter, loff_t offset); + ssize_t (*direct_IO)(struct kiocb *, struct iov_iter *iter, loff_t offset); /* migrate the contents of a page to the specified target */ int (*migratepage) (struct page *, struct page *); int (*launder_page) (struct page *); diff --git a/drivers/block/loop.c b/drivers/block/loop.c index c4fd1e4..ae3fcb4 100644 --- a/drivers/block/loop.c +++ b/drivers/block/loop.c @@ -88,28 +88,6 @@ static int part_shift; static struct workqueue_struct *loop_wq; -/* - * Transfer functions - */ -static int transfer_none(struct loop_device *lo, int cmd, - struct page *raw_page, unsigned raw_off, - struct page *loop_page, unsigned loop_off, - int size, sector_t real_block) -{ - char *raw_buf = kmap_atomic(raw_page) + raw_off; - char *loop_buf = kmap_atomic(loop_page) + loop_off; - - if (cmd == READ) - memcpy(loop_buf, raw_buf, size); - else - memcpy(raw_buf, loop_buf, size); - - kunmap_atomic(loop_buf); - kunmap_atomic(raw_buf); - cond_resched(); - return 0; -} - static int transfer_xor(struct loop_device *lo, int cmd, struct page *raw_page, unsigned raw_off, struct page *loop_page, unsigned loop_off, @@ -148,14 +126,13 @@ static int xor_init(struct loop_device *lo, const struct loop_info64 *info) static struct loop_func_table none_funcs = { .number = LO_CRYPT_NONE, - .transfer = transfer_none, -}; +}; static struct loop_func_table xor_funcs = { .number = LO_CRYPT_XOR, .transfer = transfer_xor, .init = xor_init -}; +}; /* xfer_funcs[0] is special - its release function is never called */ static struct loop_func_table *xfer_funcs[MAX_LO_CRYPT] = { @@ -215,207 +192,169 @@ lo_do_transfer(struct loop_device *lo, int cmd, struct page *lpage, unsigned loffs, int size, sector_t rblock) { - if (unlikely(!lo->transfer)) + int ret; + + ret = lo->transfer(lo, cmd, rpage, roffs, lpage, loffs, size, rblock); + if (likely(!ret)) return 0; - return lo->transfer(lo, cmd, rpage, roffs, lpage, loffs, size, rblock); + printk_ratelimited(KERN_ERR + "loop: Transfer error at byte offset %llu, length %i.\n", + (unsigned long long)rblock << 9, size); + return ret; } -/** - * __do_lo_send_write - helper for writing data to a loop device - * - * This helper just factors out common code between do_lo_send_direct_write() - * and do_lo_send_write(). - */ -static int __do_lo_send_write(struct file *file, - u8 *buf, const int len, loff_t pos) +static int lo_write_bvec(struct file *file, struct bio_vec *bvec, loff_t *ppos) { - struct kvec kvec = {.iov_base = buf, .iov_len = len}; - struct iov_iter from; + struct iov_iter i; ssize_t bw; - iov_iter_kvec(&from, ITER_KVEC | WRITE, &kvec, 1, len); + iov_iter_bvec(&i, ITER_BVEC, bvec, 1, bvec->bv_len); file_start_write(file); - bw = vfs_iter_write(file, &from, &pos); + bw = vfs_iter_write(file, &i, ppos); file_end_write(file); - if (likely(bw == len)) + + if (likely(bw == bvec->bv_len)) return 0; - printk_ratelimited(KERN_ERR "loop: Write error at byte offset %llu, length %i.\n", - (unsigned long long)pos, len); + + printk_ratelimited(KERN_ERR + "loop: Write error at byte offset %llu, length %i.\n", + (unsigned long long)*ppos, bvec->bv_len); if (bw >= 0) bw = -EIO; return bw; } -/** - * do_lo_send_direct_write - helper for writing data to a loop device - * - * This is the fast, non-transforming version that does not need double - * buffering. - */ -static int do_lo_send_direct_write(struct loop_device *lo, - struct bio_vec *bvec, loff_t pos, struct page *page) +static int lo_write_simple(struct loop_device *lo, struct request *rq, + loff_t pos) { - ssize_t bw = __do_lo_send_write(lo->lo_backing_file, - kmap(bvec->bv_page) + bvec->bv_offset, - bvec->bv_len, pos); - kunmap(bvec->bv_page); - cond_resched(); - return bw; + struct bio_vec bvec; + struct req_iterator iter; + int ret = 0; + + rq_for_each_segment(bvec, rq, iter) { + ret = lo_write_bvec(lo->lo_backing_file, &bvec, &pos); + if (ret < 0) + break; + cond_resched(); + } + + return ret; } -/** - * do_lo_send_write - helper for writing data to a loop device - * +/* * This is the slow, transforming version that needs to double buffer the * data as it cannot do the transformations in place without having direct * access to the destination pages of the backing file. */ -static int do_lo_send_write(struct loop_device *lo, struct bio_vec *bvec, - loff_t pos, struct page *page) +static int lo_write_transfer(struct loop_device *lo, struct request *rq, + loff_t pos) { - int ret = lo_do_transfer(lo, WRITE, page, 0, bvec->bv_page, - bvec->bv_offset, bvec->bv_len, pos >> 9); - if (likely(!ret)) - return __do_lo_send_write(lo->lo_backing_file, - page_address(page), bvec->bv_len, - pos); - printk_ratelimited(KERN_ERR "loop: Transfer error at byte offset %llu, " - "length %i.\n", (unsigned long long)pos, bvec->bv_len); - if (ret > 0) - ret = -EIO; - return ret; -} - -static int lo_send(struct loop_device *lo, struct request *rq, loff_t pos) -{ - int (*do_lo_send)(struct loop_device *, struct bio_vec *, loff_t, - struct page *page); - struct bio_vec bvec; + struct bio_vec bvec, b; struct req_iterator iter; - struct page *page = NULL; + struct page *page; int ret = 0; - if (lo->transfer != transfer_none) { - page = alloc_page(GFP_NOIO | __GFP_HIGHMEM); - if (unlikely(!page)) - goto fail; - kmap(page); - do_lo_send = do_lo_send_write; - } else { - do_lo_send = do_lo_send_direct_write; - } + page = alloc_page(GFP_NOIO); + if (unlikely(!page)) + return -ENOMEM; rq_for_each_segment(bvec, rq, iter) { - ret = do_lo_send(lo, &bvec, pos, page); + ret = lo_do_transfer(lo, WRITE, page, 0, bvec.bv_page, + bvec.bv_offset, bvec.bv_len, pos >> 9); + if (unlikely(ret)) + break; + + b.bv_page = page; + b.bv_offset = 0; + b.bv_len = bvec.bv_len; + ret = lo_write_bvec(lo->lo_backing_file, &b, &pos); if (ret < 0) break; - pos += bvec.bv_len; } - if (page) { - kunmap(page); - __free_page(page); - } -out: + + __free_page(page); return ret; -fail: - printk_ratelimited(KERN_ERR "loop: Failed to allocate temporary page for write.\n"); - ret = -ENOMEM; - goto out; } -struct lo_read_data { - struct loop_device *lo; - struct page *page; - unsigned offset; - int bsize; -}; +static int lo_read_simple(struct loop_device *lo, struct request *rq, + loff_t pos) +{ + struct bio_vec bvec; + struct req_iterator iter; + struct iov_iter i; + ssize_t len; -static int -lo_splice_actor(struct pipe_inode_info *pipe, struct pipe_buffer *buf, - struct splice_desc *sd) -{ - struct lo_read_data *p = sd->u.data; - struct loop_device *lo = p->lo; - struct page *page = buf->page; - sector_t IV; - int size; - - IV = ((sector_t) page->index << (PAGE_CACHE_SHIFT - 9)) + - (buf->offset >> 9); - size = sd->len; - if (size > p->bsize) - size = p->bsize; - - if (lo_do_transfer(lo, READ, page, buf->offset, p->page, p->offset, size, IV)) { - printk_ratelimited(KERN_ERR "loop: transfer error block %ld\n", - page->index); - size = -EINVAL; - } + rq_for_each_segment(bvec, rq, iter) { + iov_iter_bvec(&i, ITER_BVEC, &bvec, 1, bvec.bv_len); + len = vfs_iter_read(lo->lo_backing_file, &i, &pos); + if (len < 0) + return len; - flush_dcache_page(p->page); + flush_dcache_page(bvec.bv_page); - if (size > 0) - p->offset += size; + if (len != bvec.bv_len) { + struct bio *bio; - return size; -} + __rq_for_each_bio(bio, rq) + zero_fill_bio(bio); + break; + } + cond_resched(); + } -static int -lo_direct_splice_actor(struct pipe_inode_info *pipe, struct splice_desc *sd) -{ - return __splice_from_pipe(pipe, sd, lo_splice_actor); + return 0; } -static ssize_t -do_lo_receive(struct loop_device *lo, - struct bio_vec *bvec, int bsize, loff_t pos) +static int lo_read_transfer(struct loop_device *lo, struct request *rq, + loff_t pos) { - struct lo_read_data cookie; - struct splice_desc sd; - struct file *file; - ssize_t retval; + struct bio_vec bvec, b; + struct req_iterator iter; + struct iov_iter i; + struct page *page; + ssize_t len; + int ret = 0; - cookie.lo = lo; - cookie.page = bvec->bv_page; - cookie.offset = bvec->bv_offset; - cookie.bsize = bsize; + page = alloc_page(GFP_NOIO); + if (unlikely(!page)) + return -ENOMEM; - sd.len = 0; - sd.total_len = bvec->bv_len; - sd.flags = 0; - sd.pos = pos; - sd.u.data = &cookie; + rq_for_each_segment(bvec, rq, iter) { + loff_t offset = pos; - file = lo->lo_backing_file; - retval = splice_direct_to_actor(file, &sd, lo_direct_splice_actor); + b.bv_page = page; + b.bv_offset = 0; + b.bv_len = bvec.bv_len; - return retval; -} + iov_iter_bvec(&i, ITER_BVEC, &b, 1, b.bv_len); + len = vfs_iter_read(lo->lo_backing_file, &i, &pos); + if (len < 0) { + ret = len; + goto out_free_page; + } -static int -lo_receive(struct loop_device *lo, struct request *rq, int bsize, loff_t pos) -{ - struct bio_vec bvec; - struct req_iterator iter; - ssize_t s; + ret = lo_do_transfer(lo, READ, page, 0, bvec.bv_page, + bvec.bv_offset, len, offset >> 9); + if (ret) + goto out_free_page; - rq_for_each_segment(bvec, rq, iter) { - s = do_lo_receive(lo, &bvec, bsize, pos); - if (s < 0) - return s; + flush_dcache_page(bvec.bv_page); - if (s != bvec.bv_len) { + if (len != bvec.bv_len) { struct bio *bio; __rq_for_each_bio(bio, rq) zero_fill_bio(bio); break; } - pos += bvec.bv_len; } - return 0; + + ret = 0; +out_free_page: + __free_page(page); + return ret; } static int lo_discard(struct loop_device *lo, struct request *rq, loff_t pos) @@ -464,10 +403,17 @@ static int do_req_filebacked(struct loop_device *lo, struct request *rq) ret = lo_req_flush(lo, rq); else if (rq->cmd_flags & REQ_DISCARD) ret = lo_discard(lo, rq, pos); + else if (lo->transfer) + ret = lo_write_transfer(lo, rq, pos); else - ret = lo_send(lo, rq, pos); - } else - ret = lo_receive(lo, rq, lo->lo_blocksize, pos); + ret = lo_write_simple(lo, rq, pos); + + } else { + if (lo->transfer) + ret = lo_read_transfer(lo, rq, pos); + else + ret = lo_read_simple(lo, rq, pos); + } return ret; } @@ -788,7 +734,7 @@ static int loop_set_fd(struct loop_device *lo, fmode_t mode, lo->lo_device = bdev; lo->lo_flags = lo_flags; lo->lo_backing_file = file; - lo->transfer = transfer_none; + lo->transfer = NULL; lo->ioctl = NULL; lo->lo_sizelimit = 0; lo->old_gfp_mask = mapping_gfp_mask(mapping); @@ -1007,7 +953,7 @@ loop_set_status(struct loop_device *lo, const struct loop_info64 *info) memcpy(lo->lo_encrypt_key, info->lo_encrypt_key, info->lo_encrypt_key_size); lo->lo_key_owner = uid; - } + } return 0; } diff --git a/drivers/staging/lustre/lustre/llite/rw26.c b/drivers/staging/lustre/lustre/llite/rw26.c index 91442fa..c6c8243 100644 --- a/drivers/staging/lustre/lustre/llite/rw26.c +++ b/drivers/staging/lustre/lustre/llite/rw26.c @@ -359,8 +359,8 @@ static ssize_t ll_direct_IO_26_seg(const struct lu_env *env, struct cl_io *io, * up to 22MB for 128kB kmalloc and up to 682MB for 4MB kmalloc. */ #define MAX_DIO_SIZE ((MAX_MALLOC / sizeof(struct brw_page) * PAGE_CACHE_SIZE) & \ ~(DT_MAX_BRW_SIZE - 1)) -static ssize_t ll_direct_IO_26(int rw, struct kiocb *iocb, - struct iov_iter *iter, loff_t file_offset) +static ssize_t ll_direct_IO_26(struct kiocb *iocb, struct iov_iter *iter, + loff_t file_offset) { struct lu_env *env; struct cl_io *io; @@ -399,7 +399,7 @@ static ssize_t ll_direct_IO_26(int rw, struct kiocb *iocb, * size changing by concurrent truncates and writes. * 1. Need inode mutex to operate transient pages. */ - if (rw == READ) + if (iov_iter_rw(iter) == READ) mutex_lock(&inode->i_mutex); LASSERT(obj->cob_transient_pages == 0); @@ -408,7 +408,7 @@ static ssize_t ll_direct_IO_26(int rw, struct kiocb *iocb, size_t offs; count = min_t(size_t, iov_iter_count(iter), size); - if (rw == READ) { + if (iov_iter_rw(iter) == READ) { if (file_offset >= i_size_read(inode)) break; if (file_offset + count > i_size_read(inode)) @@ -418,11 +418,11 @@ static ssize_t ll_direct_IO_26(int rw, struct kiocb *iocb, result = iov_iter_get_pages_alloc(iter, &pages, count, &offs); if (likely(result > 0)) { int n = DIV_ROUND_UP(result + offs, PAGE_SIZE); - result = ll_direct_IO_26_seg(env, io, rw, inode, - file->f_mapping, - result, file_offset, - pages, n); - ll_free_user_pages(pages, n, rw==READ); + result = ll_direct_IO_26_seg(env, io, iov_iter_rw(iter), + inode, file->f_mapping, + result, file_offset, pages, + n); + ll_free_user_pages(pages, n, iov_iter_rw(iter) == READ); } if (unlikely(result <= 0)) { /* If we can't allocate a large enough buffer @@ -449,11 +449,11 @@ static ssize_t ll_direct_IO_26(int rw, struct kiocb *iocb, } out: LASSERT(obj->cob_transient_pages == 0); - if (rw == READ) + if (iov_iter_rw(iter) == READ) mutex_unlock(&inode->i_mutex); if (tot_bytes > 0) { - if (rw == WRITE) { + if (iov_iter_rw(iter) == WRITE) { struct lov_stripe_md *lsm; lsm = ccc_inode_lsm_get(inode); diff --git a/fs/9p/vfs_addr.c b/fs/9p/vfs_addr.c index 2e38f9a..be35d05 100644 --- a/fs/9p/vfs_addr.c +++ b/fs/9p/vfs_addr.c @@ -230,7 +230,6 @@ static int v9fs_launder_page(struct page *page) /** * v9fs_direct_IO - 9P address space operation for direct I/O - * @rw: direction (read or write) * @iocb: target I/O control block * @iov: array of vectors that define I/O buffer * @pos: offset in file to begin the operation @@ -248,12 +247,12 @@ static int v9fs_launder_page(struct page *page) * */ static ssize_t -v9fs_direct_IO(int rw, struct kiocb *iocb, struct iov_iter *iter, loff_t pos) +v9fs_direct_IO(struct kiocb *iocb, struct iov_iter *iter, loff_t pos) { struct file *file = iocb->ki_filp; ssize_t n; int err = 0; - if (rw & WRITE) { + if (iov_iter_rw(iter) == WRITE) { n = p9_client_write(file->private_data, pos, iter, &err); if (n) { struct inode *inode = file_inode(file); diff --git a/fs/9p/vfs_file.c b/fs/9p/vfs_file.c index d7fcb77..2a9dd37 100644 --- a/fs/9p/vfs_file.c +++ b/fs/9p/vfs_file.c @@ -404,21 +404,16 @@ static ssize_t v9fs_file_write_iter(struct kiocb *iocb, struct iov_iter *from) { struct file *file = iocb->ki_filp; - ssize_t retval = 0; - loff_t origin = iocb->ki_pos; - size_t count = iov_iter_count(from); + ssize_t retval; + loff_t origin; int err = 0; - retval = generic_write_checks(file, &origin, &count, 0); - if (retval) + retval = generic_write_checks(iocb, from); + if (retval <= 0) return retval; - iov_iter_truncate(from, count); - - if (!count) - return 0; - - retval = p9_client_write(file->private_data, origin, from, &err); + origin = iocb->ki_pos; + retval = p9_client_write(file->private_data, iocb->ki_pos, from, &err); if (retval > 0) { struct inode *inode = file_inode(file); loff_t i_size; @@ -428,12 +423,11 @@ v9fs_file_write_iter(struct kiocb *iocb, struct iov_iter *from) if (inode->i_mapping && inode->i_mapping->nrpages) invalidate_inode_pages2_range(inode->i_mapping, pg_start, pg_end); - origin += retval; + iocb->ki_pos += retval; i_size = i_size_read(inode); - iocb->ki_pos = origin; - if (origin > i_size) { - inode_add_bytes(inode, origin - i_size); - i_size_write(inode, origin); + if (iocb->ki_pos > i_size) { + inode_add_bytes(inode, iocb->ki_pos - i_size); + i_size_write(inode, iocb->ki_pos); } return retval; } diff --git a/fs/affs/file.c b/fs/affs/file.c index 7c1a3d4..dcf2795 100644 --- a/fs/affs/file.c +++ b/fs/affs/file.c @@ -389,8 +389,7 @@ static void affs_write_failed(struct address_space *mapping, loff_t to) } static ssize_t -affs_direct_IO(int rw, struct kiocb *iocb, struct iov_iter *iter, - loff_t offset) +affs_direct_IO(struct kiocb *iocb, struct iov_iter *iter, loff_t offset) { struct file *file = iocb->ki_filp; struct address_space *mapping = file->f_mapping; @@ -398,15 +397,15 @@ affs_direct_IO(int rw, struct kiocb *iocb, struct iov_iter *iter, size_t count = iov_iter_count(iter); ssize_t ret; - if (rw == WRITE) { + if (iov_iter_rw(iter) == WRITE) { loff_t size = offset + count; if (AFFS_I(inode)->mmu_private < size) return 0; } - ret = blockdev_direct_IO(rw, iocb, inode, iter, offset, affs_get_block); - if (ret < 0 && (rw & WRITE)) + ret = blockdev_direct_IO(iocb, inode, iter, offset, affs_get_block); + if (ret < 0 && iov_iter_rw(iter) == WRITE) affs_write_failed(mapping, offset + count); return ret; } @@ -1517,7 +1517,7 @@ static int io_submit_one(struct kioctx *ctx, struct iocb __user *user_iocb, } req->common.ki_pos = iocb->aio_offset; req->common.ki_complete = aio_complete; - req->common.ki_flags = 0; + req->common.ki_flags = iocb_flags(req->common.ki_filp); if (iocb->aio_flags & IOCB_FLAG_RESFD) { /* diff --git a/fs/block_dev.c b/fs/block_dev.c index b5e8789..897ee05 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -146,15 +146,13 @@ blkdev_get_block(struct inode *inode, sector_t iblock, } static ssize_t -blkdev_direct_IO(int rw, struct kiocb *iocb, struct iov_iter *iter, - loff_t offset) +blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, loff_t offset) { struct file *file = iocb->ki_filp; struct inode *inode = file->f_mapping->host; - return __blockdev_direct_IO(rw, iocb, inode, I_BDEV(inode), iter, - offset, blkdev_get_block, - NULL, NULL, 0); + return __blockdev_direct_IO(iocb, inode, I_BDEV(inode), iter, offset, + blkdev_get_block, NULL, NULL, 0); } int __sync_blockdev(struct block_device *bdev, int wait) @@ -1597,9 +1595,22 @@ static long block_ioctl(struct file *file, unsigned cmd, unsigned long arg) ssize_t blkdev_write_iter(struct kiocb *iocb, struct iov_iter *from) { struct file *file = iocb->ki_filp; + struct inode *bd_inode = file->f_mapping->host; + loff_t size = i_size_read(bd_inode); struct blk_plug plug; ssize_t ret; + if (bdev_read_only(I_BDEV(bd_inode))) + return -EPERM; + + if (!iov_iter_count(from)) + return 0; + + if (iocb->ki_pos >= size) + return -ENOSPC; + + iov_iter_truncate(from, size - iocb->ki_pos); + blk_start_plug(&plug); ret = __generic_file_write_iter(iocb, from); if (ret > 0) { diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index cdc801c..faa7d39 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -1739,27 +1739,19 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb, u64 start_pos; u64 end_pos; ssize_t num_written = 0; - ssize_t err = 0; - size_t count = iov_iter_count(from); bool sync = (file->f_flags & O_DSYNC) || IS_SYNC(file->f_mapping->host); - loff_t pos = iocb->ki_pos; + ssize_t err; + loff_t pos; + size_t count; mutex_lock(&inode->i_mutex); - - current->backing_dev_info = inode_to_bdi(inode); - err = generic_write_checks(file, &pos, &count, S_ISBLK(inode->i_mode)); - if (err) { + err = generic_write_checks(iocb, from); + if (err <= 0) { mutex_unlock(&inode->i_mutex); - goto out; - } - - if (count == 0) { - mutex_unlock(&inode->i_mutex); - goto out; + return err; } - iov_iter_truncate(from, count); - + current->backing_dev_info = inode_to_bdi(inode); err = file_remove_suid(file); if (err) { mutex_unlock(&inode->i_mutex); @@ -1786,6 +1778,8 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb, */ update_time_for_write(inode); + pos = iocb->ki_pos; + count = iov_iter_count(from); start_pos = round_down(pos, root->sectorsize); if (start_pos > i_size_read(inode)) { /* Expand hole size to cover write data, preventing empty gap */ @@ -1800,7 +1794,7 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb, if (sync) atomic_inc(&BTRFS_I(inode)->sync_writers); - if (file->f_flags & O_DIRECT) { + if (iocb->ki_flags & IOCB_DIRECT) { num_written = __btrfs_direct_write(iocb, from, pos); } else { num_written = __btrfs_buffered_write(file, from, pos); diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 686331f..43192e1 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -8081,7 +8081,7 @@ free_ordered: bio_endio(dio_bio, ret); } -static ssize_t check_direct_IO(struct btrfs_root *root, int rw, struct kiocb *iocb, +static ssize_t check_direct_IO(struct btrfs_root *root, struct kiocb *iocb, const struct iov_iter *iter, loff_t offset) { int seg; @@ -8096,7 +8096,7 @@ static ssize_t check_direct_IO(struct btrfs_root *root, int rw, struct kiocb *io goto out; /* If this is a write we don't need to check anymore */ - if (rw & WRITE) + if (iov_iter_rw(iter) == WRITE) return 0; /* * Check to make sure we don't have duplicate iov_base's in this @@ -8114,8 +8114,8 @@ out: return retval; } -static ssize_t btrfs_direct_IO(int rw, struct kiocb *iocb, - struct iov_iter *iter, loff_t offset) +static ssize_t btrfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter, + loff_t offset) { struct file *file = iocb->ki_filp; struct inode *inode = file->f_mapping->host; @@ -8126,7 +8126,7 @@ static ssize_t btrfs_direct_IO(int rw, struct kiocb *iocb, bool relock = false; ssize_t ret; - if (check_direct_IO(BTRFS_I(inode)->root, rw, iocb, iter, offset)) + if (check_direct_IO(BTRFS_I(inode)->root, iocb, iter, offset)) return 0; atomic_inc(&inode->i_dio_count); @@ -8144,7 +8144,7 @@ static ssize_t btrfs_direct_IO(int rw, struct kiocb *iocb, filemap_fdatawrite_range(inode->i_mapping, offset, offset + count - 1); - if (rw & WRITE) { + if (iov_iter_rw(iter) == WRITE) { /* * If the write DIO is beyond the EOF, we need update * the isize, but it is protected by i_mutex. So we can @@ -8174,11 +8174,11 @@ static ssize_t btrfs_direct_IO(int rw, struct kiocb *iocb, wakeup = false; } - ret = __blockdev_direct_IO(rw, iocb, inode, - BTRFS_I(inode)->root->fs_info->fs_devices->latest_bdev, - iter, offset, btrfs_get_blocks_direct, NULL, - btrfs_submit_direct, flags); - if (rw & WRITE) { + ret = __blockdev_direct_IO(iocb, inode, + BTRFS_I(inode)->root->fs_info->fs_devices->latest_bdev, + iter, offset, btrfs_get_blocks_direct, NULL, + btrfs_submit_direct, flags); + if (iov_iter_rw(iter) == WRITE) { current->journal_info = NULL; if (ret < 0 && ret != -EIOCBQUEUED) btrfs_delalloc_release_space(inode, count); diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c index fd5599d..155ab9c 100644 --- a/fs/ceph/addr.c +++ b/fs/ceph/addr.c @@ -1198,8 +1198,7 @@ static int ceph_write_end(struct file *file, struct address_space *mapping, * intercept O_DIRECT reads and writes early, this function should * never get called. */ -static ssize_t ceph_direct_io(int rw, struct kiocb *iocb, - struct iov_iter *iter, +static ssize_t ceph_direct_io(struct kiocb *iocb, struct iov_iter *iter, loff_t pos) { WARN_ON(1); diff --git a/fs/ceph/file.c b/fs/ceph/file.c index 56237ea..b9b8eb2 100644 --- a/fs/ceph/file.c +++ b/fs/ceph/file.c @@ -457,7 +457,7 @@ static ssize_t ceph_sync_read(struct kiocb *iocb, struct iov_iter *i, if (ret < 0) return ret; - if (file->f_flags & O_DIRECT) { + if (iocb->ki_flags & IOCB_DIRECT) { while (iov_iter_count(i)) { size_t start; ssize_t n; @@ -828,7 +828,7 @@ again: return ret; if ((got & (CEPH_CAP_FILE_CACHE|CEPH_CAP_FILE_LAZYIO)) == 0 || - (iocb->ki_filp->f_flags & O_DIRECT) || + (iocb->ki_flags & IOCB_DIRECT) || (fi->flags & CEPH_F_SYNC)) { dout("aio_sync_read %p %llx.%llx %llu~%u got cap refs on %s\n", @@ -941,9 +941,9 @@ static ssize_t ceph_write_iter(struct kiocb *iocb, struct iov_iter *from) struct ceph_inode_info *ci = ceph_inode(inode); struct ceph_osd_client *osdc = &ceph_sb_to_client(inode->i_sb)->client->osdc; - ssize_t count = iov_iter_count(from), written = 0; + ssize_t count, written = 0; int err, want, got; - loff_t pos = iocb->ki_pos; + loff_t pos; if (ceph_snap(inode) != CEPH_NOSNAP) return -EROFS; @@ -953,14 +953,12 @@ static ssize_t ceph_write_iter(struct kiocb *iocb, struct iov_iter *from) /* We can write back this queue in page reclaim */ current->backing_dev_info = inode_to_bdi(inode); - err = generic_write_checks(file, &pos, &count, S_ISBLK(inode->i_mode)); - if (err) - goto out; - - if (count == 0) + err = generic_write_checks(iocb, from); + if (err <= 0) goto out; - iov_iter_truncate(from, count); + pos = iocb->ki_pos; + count = iov_iter_count(from); err = file_remove_suid(file); if (err) goto out; @@ -997,12 +995,12 @@ retry_snap: inode, ceph_vinop(inode), pos, count, ceph_cap_string(got)); if ((got & (CEPH_CAP_FILE_BUFFER|CEPH_CAP_FILE_LAZYIO)) == 0 || - (file->f_flags & O_DIRECT) || (fi->flags & CEPH_F_SYNC)) { + (iocb->ki_flags & IOCB_DIRECT) || (fi->flags & CEPH_F_SYNC)) { struct iov_iter data; mutex_unlock(&inode->i_mutex); /* we might need to revert back to that point */ data = *from; - if (file->f_flags & O_DIRECT) + if (iocb->ki_flags & IOCB_DIRECT) written = ceph_sync_direct_write(iocb, &data, pos); else written = ceph_sync_write(iocb, &data, pos); diff --git a/fs/cifs/file.c b/fs/cifs/file.c index ca30c39..ca2bc54 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -2560,10 +2560,9 @@ cifs_write_from_iter(loff_t offset, size_t len, struct iov_iter *from, return rc; } -static ssize_t -cifs_iovec_write(struct file *file, struct iov_iter *from, loff_t *poffset) +ssize_t cifs_user_writev(struct kiocb *iocb, struct iov_iter *from) { - size_t len; + struct file *file = iocb->ki_filp; ssize_t total_written = 0; struct cifsFileInfo *open_file; struct cifs_tcon *tcon; @@ -2573,15 +2572,15 @@ cifs_iovec_write(struct file *file, struct iov_iter *from, loff_t *poffset) struct iov_iter saved_from; int rc; - len = iov_iter_count(from); - rc = generic_write_checks(file, poffset, &len, 0); - if (rc) - return rc; - - if (!len) - return 0; + /* + * BB - optimize the way when signing is disabled. We can drop this + * extra memory-to-memory copying and use iovec buffers for constructing + * write request. + */ - iov_iter_truncate(from, len); + rc = generic_write_checks(iocb, from); + if (rc <= 0) + return rc; INIT_LIST_HEAD(&wdata_list); cifs_sb = CIFS_FILE_SB(file); @@ -2593,8 +2592,8 @@ cifs_iovec_write(struct file *file, struct iov_iter *from, loff_t *poffset) memcpy(&saved_from, from, sizeof(struct iov_iter)); - rc = cifs_write_from_iter(*poffset, len, from, open_file, cifs_sb, - &wdata_list); + rc = cifs_write_from_iter(iocb->ki_pos, iov_iter_count(from), from, + open_file, cifs_sb, &wdata_list); /* * If at least one write was successfully sent, then discard any rc @@ -2633,7 +2632,7 @@ restart_loop: memcpy(&tmp_from, &saved_from, sizeof(struct iov_iter)); iov_iter_advance(&tmp_from, - wdata->offset - *poffset); + wdata->offset - iocb->ki_pos); rc = cifs_write_from_iter(wdata->offset, wdata->bytes, &tmp_from, @@ -2650,34 +2649,13 @@ restart_loop: kref_put(&wdata->refcount, cifs_uncached_writedata_release); } - if (total_written > 0) - *poffset += total_written; + if (unlikely(!total_written)) + return rc; + iocb->ki_pos += total_written; + set_bit(CIFS_INO_INVALID_MAPPING, &CIFS_I(file_inode(file))->flags); cifs_stats_bytes_written(tcon, total_written); - return total_written ? total_written : (ssize_t)rc; -} - -ssize_t cifs_user_writev(struct kiocb *iocb, struct iov_iter *from) -{ - ssize_t written; - struct inode *inode; - loff_t pos = iocb->ki_pos; - - inode = file_inode(iocb->ki_filp); - - /* - * BB - optimize the way when signing is disabled. We can drop this - * extra memory-to-memory copying and use iovec buffers for constructing - * write request. - */ - - written = cifs_iovec_write(iocb->ki_filp, from, &pos); - if (written > 0) { - set_bit(CIFS_INO_INVALID_MAPPING, &CIFS_I(inode)->flags); - iocb->ki_pos = pos; - } - - return written; + return total_written; } static ssize_t @@ -2688,8 +2666,7 @@ cifs_writev(struct kiocb *iocb, struct iov_iter *from) struct inode *inode = file->f_mapping->host; struct cifsInodeInfo *cinode = CIFS_I(inode); struct TCP_Server_Info *server = tlink_tcon(cfile->tlink)->ses->server; - ssize_t rc = -EACCES; - loff_t lock_pos = iocb->ki_pos; + ssize_t rc; /* * We need to hold the sem to be sure nobody modifies lock list @@ -2697,23 +2674,24 @@ cifs_writev(struct kiocb *iocb, struct iov_iter *from) */ down_read(&cinode->lock_sem); mutex_lock(&inode->i_mutex); - if (file->f_flags & O_APPEND) - lock_pos = i_size_read(inode); - if (!cifs_find_lock_conflict(cfile, lock_pos, iov_iter_count(from), + + rc = generic_write_checks(iocb, from); + if (rc <= 0) + goto out; + + if (!cifs_find_lock_conflict(cfile, iocb->ki_pos, iov_iter_count(from), server->vals->exclusive_lock_type, NULL, - CIFS_WRITE_OP)) { + CIFS_WRITE_OP)) rc = __generic_file_write_iter(iocb, from); - mutex_unlock(&inode->i_mutex); - - if (rc > 0) { - ssize_t err; + else + rc = -EACCES; +out: + mutex_unlock(&inode->i_mutex); - err = generic_write_sync(file, iocb->ki_pos - rc, rc); - if (err < 0) - rc = err; - } - } else { - mutex_unlock(&inode->i_mutex); + if (rc > 0) { + ssize_t err = generic_write_sync(file, iocb->ki_pos - rc, rc); + if (err < 0) + rc = err; } up_read(&cinode->lock_sem); return rc; @@ -3877,8 +3855,7 @@ void cifs_oplock_break(struct work_struct *work) * Direct IO is not yet supported in the cached mode. */ static ssize_t -cifs_direct_io(int rw, struct kiocb *iocb, struct iov_iter *iter, - loff_t pos) +cifs_direct_io(struct kiocb *iocb, struct iov_iter *iter, loff_t pos) { /* * FIXME diff --git a/fs/configfs/dir.c b/fs/configfs/dir.c index cf0db00..acb3d63 100644 --- a/fs/configfs/dir.c +++ b/fs/configfs/dir.c @@ -1598,7 +1598,7 @@ static loff_t configfs_dir_lseek(struct file *file, loff_t offset, int whence) if (offset >= 0) break; default: - mutex_unlock(&file_inode(file)->i_mutex); + mutex_unlock(&dentry->d_inode->i_mutex); return -EINVAL; } if (offset != file->f_pos) { @@ -98,9 +98,9 @@ static bool buffer_size_valid(struct buffer_head *bh) return bh->b_state != 0; } -static ssize_t dax_io(int rw, struct inode *inode, struct iov_iter *iter, - loff_t start, loff_t end, get_block_t get_block, - struct buffer_head *bh) +static ssize_t dax_io(struct inode *inode, struct iov_iter *iter, + loff_t start, loff_t end, get_block_t get_block, + struct buffer_head *bh) { ssize_t retval = 0; loff_t pos = start; @@ -109,7 +109,7 @@ static ssize_t dax_io(int rw, struct inode *inode, struct iov_iter *iter, void *addr; bool hole = false; - if (rw != WRITE) + if (iov_iter_rw(iter) != WRITE) end = min(end, i_size_read(inode)); while (pos < end) { @@ -124,7 +124,7 @@ static ssize_t dax_io(int rw, struct inode *inode, struct iov_iter *iter, bh->b_size = PAGE_ALIGN(end - pos); bh->b_state = 0; retval = get_block(inode, block, bh, - rw == WRITE); + iov_iter_rw(iter) == WRITE); if (retval) break; if (!buffer_size_valid(bh)) @@ -137,7 +137,7 @@ static ssize_t dax_io(int rw, struct inode *inode, struct iov_iter *iter, bh->b_size -= done; } - hole = (rw != WRITE) && !buffer_written(bh); + hole = iov_iter_rw(iter) != WRITE && !buffer_written(bh); if (hole) { addr = NULL; size = bh->b_size - first; @@ -154,7 +154,7 @@ static ssize_t dax_io(int rw, struct inode *inode, struct iov_iter *iter, max = min(pos + size, end); } - if (rw == WRITE) + if (iov_iter_rw(iter) == WRITE) len = copy_from_iter(addr, max - pos, iter); else if (!hole) len = copy_to_iter(addr, max - pos, iter); @@ -173,7 +173,6 @@ static ssize_t dax_io(int rw, struct inode *inode, struct iov_iter *iter, /** * dax_do_io - Perform I/O to a DAX file - * @rw: READ to read or WRITE to write * @iocb: The control block for this I/O * @inode: The file which the I/O is directed at * @iter: The addresses to do I/O from or to @@ -189,9 +188,9 @@ static ssize_t dax_io(int rw, struct inode *inode, struct iov_iter *iter, * As with do_blockdev_direct_IO(), we increment i_dio_count while the I/O * is in progress. */ -ssize_t dax_do_io(int rw, struct kiocb *iocb, struct inode *inode, - struct iov_iter *iter, loff_t pos, - get_block_t get_block, dio_iodone_t end_io, int flags) +ssize_t dax_do_io(struct kiocb *iocb, struct inode *inode, + struct iov_iter *iter, loff_t pos, get_block_t get_block, + dio_iodone_t end_io, int flags) { struct buffer_head bh; ssize_t retval = -EINVAL; @@ -199,7 +198,7 @@ ssize_t dax_do_io(int rw, struct kiocb *iocb, struct inode *inode, memset(&bh, 0, sizeof(bh)); - if ((flags & DIO_LOCKING) && (rw == READ)) { + if ((flags & DIO_LOCKING) && iov_iter_rw(iter) == READ) { struct address_space *mapping = inode->i_mapping; mutex_lock(&inode->i_mutex); retval = filemap_write_and_wait_range(mapping, pos, end - 1); @@ -212,9 +211,9 @@ ssize_t dax_do_io(int rw, struct kiocb *iocb, struct inode *inode, /* Protects against truncate */ atomic_inc(&inode->i_dio_count); - retval = dax_io(rw, inode, iter, pos, end, get_block, &bh); + retval = dax_io(inode, iter, pos, end, get_block, &bh); - if ((flags & DIO_LOCKING) && (rw == READ)) + if ((flags & DIO_LOCKING) && iov_iter_rw(iter) == READ) mutex_unlock(&inode->i_mutex); if ((retval > 0) && end_io) diff --git a/fs/dcache.c b/fs/dcache.c index d99736a..656ce52 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -269,6 +269,41 @@ static inline int dname_external(const struct dentry *dentry) return dentry->d_name.name != dentry->d_iname; } +/* + * Make sure other CPUs see the inode attached before the type is set. + */ +static inline void __d_set_inode_and_type(struct dentry *dentry, + struct inode *inode, + unsigned type_flags) +{ + unsigned flags; + + dentry->d_inode = inode; + smp_wmb(); + flags = READ_ONCE(dentry->d_flags); + flags &= ~(DCACHE_ENTRY_TYPE | DCACHE_FALLTHRU); + flags |= type_flags; + WRITE_ONCE(dentry->d_flags, flags); +} + +/* + * Ideally, we want to make sure that other CPUs see the flags cleared before + * the inode is detached, but this is really a violation of RCU principles + * since the ordering suggests we should always set inode before flags. + * + * We should instead replace or discard the entire dentry - but that sucks + * performancewise on mass deletion/rename. + */ +static inline void __d_clear_type_and_inode(struct dentry *dentry) +{ + unsigned flags = READ_ONCE(dentry->d_flags); + + flags &= ~(DCACHE_ENTRY_TYPE | DCACHE_FALLTHRU); + WRITE_ONCE(dentry->d_flags, flags); + smp_wmb(); + dentry->d_inode = NULL; +} + static void dentry_free(struct dentry *dentry) { WARN_ON(!hlist_unhashed(&dentry->d_u.d_alias)); @@ -311,7 +346,7 @@ static void dentry_iput(struct dentry * dentry) { struct inode *inode = dentry->d_inode; if (inode) { - dentry->d_inode = NULL; + __d_clear_type_and_inode(dentry); hlist_del_init(&dentry->d_u.d_alias); spin_unlock(&dentry->d_lock); spin_unlock(&inode->i_lock); @@ -335,8 +370,7 @@ static void dentry_unlink_inode(struct dentry * dentry) __releases(dentry->d_inode->i_lock) { struct inode *inode = dentry->d_inode; - __d_clear_type(dentry); - dentry->d_inode = NULL; + __d_clear_type_and_inode(dentry); hlist_del_init(&dentry->d_u.d_alias); dentry_rcuwalk_barrier(dentry); spin_unlock(&dentry->d_lock); @@ -1715,11 +1749,9 @@ static void __d_instantiate(struct dentry *dentry, struct inode *inode) unsigned add_flags = d_flags_for_inode(inode); spin_lock(&dentry->d_lock); - dentry->d_flags &= ~(DCACHE_ENTRY_TYPE | DCACHE_FALLTHRU); - dentry->d_flags |= add_flags; if (inode) hlist_add_head(&dentry->d_u.d_alias, &inode->i_dentry); - dentry->d_inode = inode; + __d_set_inode_and_type(dentry, inode, add_flags); dentry_rcuwalk_barrier(dentry); spin_unlock(&dentry->d_lock); fsnotify_d_instantiate(dentry, inode); @@ -1937,8 +1969,7 @@ static struct dentry *__d_obtain_alias(struct inode *inode, int disconnected) add_flags |= DCACHE_DISCONNECTED; spin_lock(&tmp->d_lock); - tmp->d_inode = inode; - tmp->d_flags |= add_flags; + __d_set_inode_and_type(tmp, inode, add_flags); hlist_add_head(&tmp->d_u.d_alias, &inode->i_dentry); hlist_bl_lock(&tmp->d_sb->s_anon); hlist_bl_add_head(&tmp->d_hash, &tmp->d_sb->s_anon); diff --git a/fs/debugfs/inode.c b/fs/debugfs/inode.c index 61e72d4..c9ee0df 100644 --- a/fs/debugfs/inode.c +++ b/fs/debugfs/inode.c @@ -524,7 +524,7 @@ static int __debugfs_remove(struct dentry *dentry, struct dentry *parent) if (debugfs_positive(dentry)) { dget(dentry); - if (S_ISDIR(dentry->d_inode->i_mode)) + if (d_is_dir(dentry)) ret = simple_rmdir(parent->d_inode, dentry); else simple_unlink(parent->d_inode, dentry); diff --git a/fs/direct-io.c b/fs/direct-io.c index 6fb00e3..c3b560b 100644 --- a/fs/direct-io.c +++ b/fs/direct-io.c @@ -1093,10 +1093,10 @@ static inline int drop_refcount(struct dio *dio) * for the whole file. */ static inline ssize_t -do_blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode, - struct block_device *bdev, struct iov_iter *iter, loff_t offset, - get_block_t get_block, dio_iodone_t end_io, - dio_submit_t submit_io, int flags) +do_blockdev_direct_IO(struct kiocb *iocb, struct inode *inode, + struct block_device *bdev, struct iov_iter *iter, + loff_t offset, get_block_t get_block, dio_iodone_t end_io, + dio_submit_t submit_io, int flags) { unsigned i_blkbits = ACCESS_ONCE(inode->i_blkbits); unsigned blkbits = i_blkbits; @@ -1110,9 +1110,6 @@ do_blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode, struct blk_plug plug; unsigned long align = offset | iov_iter_alignment(iter); - if (rw & WRITE) - rw = WRITE_ODIRECT; - /* * Avoid references to bdev if not absolutely needed to give * the early prefetch in the caller enough time. @@ -1127,7 +1124,7 @@ do_blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode, } /* watch out for a 0 len io from a tricksy fs */ - if (rw == READ && !iov_iter_count(iter)) + if (iov_iter_rw(iter) == READ && !iov_iter_count(iter)) return 0; dio = kmem_cache_alloc(dio_cache, GFP_KERNEL); @@ -1143,7 +1140,7 @@ do_blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode, dio->flags = flags; if (dio->flags & DIO_LOCKING) { - if (rw == READ) { + if (iov_iter_rw(iter) == READ) { struct address_space *mapping = iocb->ki_filp->f_mapping; @@ -1169,19 +1166,19 @@ do_blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode, if (is_sync_kiocb(iocb)) dio->is_async = false; else if (!(dio->flags & DIO_ASYNC_EXTEND) && - (rw & WRITE) && end > i_size_read(inode)) + iov_iter_rw(iter) == WRITE && end > i_size_read(inode)) dio->is_async = false; else dio->is_async = true; dio->inode = inode; - dio->rw = rw; + dio->rw = iov_iter_rw(iter) == WRITE ? WRITE_ODIRECT : READ; /* * For AIO O_(D)SYNC writes we need to defer completions to a workqueue * so that we can call ->fsync. */ - if (dio->is_async && (rw & WRITE) && + if (dio->is_async && iov_iter_rw(iter) == WRITE && ((iocb->ki_filp->f_flags & O_DSYNC) || IS_SYNC(iocb->ki_filp->f_mapping->host))) { retval = dio_set_defer_completion(dio); @@ -1274,7 +1271,7 @@ do_blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode, * we can let i_mutex go now that its achieved its purpose * of protecting us from looking up uninitialized blocks. */ - if (rw == READ && (dio->flags & DIO_LOCKING)) + if (iov_iter_rw(iter) == READ && (dio->flags & DIO_LOCKING)) mutex_unlock(&dio->inode->i_mutex); /* @@ -1286,7 +1283,7 @@ do_blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode, */ BUG_ON(retval == -EIOCBQUEUED); if (dio->is_async && retval == 0 && dio->result && - (rw == READ || dio->result == count)) + (iov_iter_rw(iter) == READ || dio->result == count)) retval = -EIOCBQUEUED; else dio_await_completion(dio); @@ -1300,11 +1297,11 @@ out: return retval; } -ssize_t -__blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode, - struct block_device *bdev, struct iov_iter *iter, loff_t offset, - get_block_t get_block, dio_iodone_t end_io, - dio_submit_t submit_io, int flags) +ssize_t __blockdev_direct_IO(struct kiocb *iocb, struct inode *inode, + struct block_device *bdev, struct iov_iter *iter, + loff_t offset, get_block_t get_block, + dio_iodone_t end_io, dio_submit_t submit_io, + int flags) { /* * The block device state is needed in the end to finally @@ -1318,8 +1315,8 @@ __blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode, prefetch(bdev->bd_queue); prefetch((char *)bdev->bd_queue + SMP_CACHE_BYTES); - return do_blockdev_direct_IO(rw, iocb, inode, bdev, iter, offset, - get_block, end_io, submit_io, flags); + return do_blockdev_direct_IO(iocb, inode, bdev, iter, offset, get_block, + end_io, submit_io, flags); } EXPORT_SYMBOL(__blockdev_direct_IO); diff --git a/fs/exofs/inode.c b/fs/exofs/inode.c index a198e94..35073aa 100644 --- a/fs/exofs/inode.c +++ b/fs/exofs/inode.c @@ -963,8 +963,8 @@ static void exofs_invalidatepage(struct page *page, unsigned int offset, /* TODO: Should be easy enough to do proprly */ -static ssize_t exofs_direct_IO(int rw, struct kiocb *iocb, - struct iov_iter *iter, loff_t offset) +static ssize_t exofs_direct_IO(struct kiocb *iocb, struct iov_iter *iter, + loff_t offset) { return 0; } diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c index b29eb67..5d92139 100644 --- a/fs/ext2/inode.c +++ b/fs/ext2/inode.c @@ -851,8 +851,7 @@ static sector_t ext2_bmap(struct address_space *mapping, sector_t block) } static ssize_t -ext2_direct_IO(int rw, struct kiocb *iocb, struct iov_iter *iter, - loff_t offset) +ext2_direct_IO(struct kiocb *iocb, struct iov_iter *iter, loff_t offset) { struct file *file = iocb->ki_filp; struct address_space *mapping = file->f_mapping; @@ -861,12 +860,12 @@ ext2_direct_IO(int rw, struct kiocb *iocb, struct iov_iter *iter, ssize_t ret; if (IS_DAX(inode)) - ret = dax_do_io(rw, iocb, inode, iter, offset, ext2_get_block, - NULL, DIO_LOCKING); + ret = dax_do_io(iocb, inode, iter, offset, ext2_get_block, NULL, + DIO_LOCKING); else - ret = blockdev_direct_IO(rw, iocb, inode, iter, offset, + ret = blockdev_direct_IO(iocb, inode, iter, offset, ext2_get_block); - if (ret < 0 && (rw & WRITE)) + if (ret < 0 && iov_iter_rw(iter) == WRITE) ext2_write_failed(mapping, offset + count); return ret; } diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c index db07ffb..13c0868 100644 --- a/fs/ext3/inode.c +++ b/fs/ext3/inode.c @@ -1820,8 +1820,8 @@ static int ext3_releasepage(struct page *page, gfp_t wait) * crashes then stale disk data _may_ be exposed inside the file. But current * VFS code falls back into buffered path in that case so we are safe. */ -static ssize_t ext3_direct_IO(int rw, struct kiocb *iocb, - struct iov_iter *iter, loff_t offset) +static ssize_t ext3_direct_IO(struct kiocb *iocb, struct iov_iter *iter, + loff_t offset) { struct file *file = iocb->ki_filp; struct inode *inode = file->f_mapping->host; @@ -1832,9 +1832,9 @@ static ssize_t ext3_direct_IO(int rw, struct kiocb *iocb, size_t count = iov_iter_count(iter); int retries = 0; - trace_ext3_direct_IO_enter(inode, offset, count, rw); + trace_ext3_direct_IO_enter(inode, offset, count, iov_iter_rw(iter)); - if (rw == WRITE) { + if (iov_iter_rw(iter) == WRITE) { loff_t final_size = offset + count; if (final_size > inode->i_size) { @@ -1856,12 +1856,12 @@ static ssize_t ext3_direct_IO(int rw, struct kiocb *iocb, } retry: - ret = blockdev_direct_IO(rw, iocb, inode, iter, offset, ext3_get_block); + ret = blockdev_direct_IO(iocb, inode, iter, offset, ext3_get_block); /* * In case of error extending write may have instantiated a few * blocks outside i_size. Trim these off again. */ - if (unlikely((rw & WRITE) && ret < 0)) { + if (unlikely(iov_iter_rw(iter) == WRITE && ret < 0)) { loff_t isize = i_size_read(inode); loff_t end = offset + count; @@ -1908,7 +1908,7 @@ retry: ret = err; } out: - trace_ext3_direct_IO_exit(inode, offset, count, rw, ret); + trace_ext3_direct_IO_exit(inode, offset, count, iov_iter_rw(iter), ret); return ret; } diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index 8a3981e..c8eb32e 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -2152,8 +2152,8 @@ extern void ext4_da_update_reserve_space(struct inode *inode, /* indirect.c */ extern int ext4_ind_map_blocks(handle_t *handle, struct inode *inode, struct ext4_map_blocks *map, int flags); -extern ssize_t ext4_ind_direct_IO(int rw, struct kiocb *iocb, - struct iov_iter *iter, loff_t offset); +extern ssize_t ext4_ind_direct_IO(struct kiocb *iocb, struct iov_iter *iter, + loff_t offset); extern int ext4_ind_calc_metadata_amount(struct inode *inode, sector_t lblock); extern int ext4_ind_trans_blocks(struct inode *inode, int nrblocks); extern void ext4_ind_truncate(handle_t *, struct inode *inode); diff --git a/fs/ext4/file.c b/fs/ext4/file.c index 7a6defc..e576d68 100644 --- a/fs/ext4/file.c +++ b/fs/ext4/file.c @@ -95,11 +95,9 @@ ext4_file_write_iter(struct kiocb *iocb, struct iov_iter *from) struct inode *inode = file_inode(iocb->ki_filp); struct mutex *aio_mutex = NULL; struct blk_plug plug; - int o_direct = io_is_direct(file); + int o_direct = iocb->ki_flags & IOCB_DIRECT; int overwrite = 0; - size_t length = iov_iter_count(from); ssize_t ret; - loff_t pos = iocb->ki_pos; /* * Unaligned direct AIO must be serialized; see comment above @@ -108,16 +106,17 @@ ext4_file_write_iter(struct kiocb *iocb, struct iov_iter *from) if (o_direct && ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS) && !is_sync_kiocb(iocb) && - (file->f_flags & O_APPEND || - ext4_unaligned_aio(inode, from, pos))) { + (iocb->ki_flags & IOCB_APPEND || + ext4_unaligned_aio(inode, from, iocb->ki_pos))) { aio_mutex = ext4_aio_mutex(inode); mutex_lock(aio_mutex); ext4_unwritten_wait(inode); } mutex_lock(&inode->i_mutex); - if (file->f_flags & O_APPEND) - iocb->ki_pos = pos = i_size_read(inode); + ret = generic_write_checks(iocb, from); + if (ret <= 0) + goto out; /* * If we have encountered a bitmap-format file, the size limit @@ -126,22 +125,19 @@ ext4_file_write_iter(struct kiocb *iocb, struct iov_iter *from) if (!(ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))) { struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); - if ((pos > sbi->s_bitmap_maxbytes) || - (pos == sbi->s_bitmap_maxbytes && length > 0)) { - mutex_unlock(&inode->i_mutex); + if (iocb->ki_pos >= sbi->s_bitmap_maxbytes) { ret = -EFBIG; - goto errout; + goto out; } - - if (pos + length > sbi->s_bitmap_maxbytes) - iov_iter_truncate(from, sbi->s_bitmap_maxbytes - pos); + iov_iter_truncate(from, sbi->s_bitmap_maxbytes - iocb->ki_pos); } iocb->private = &overwrite; if (o_direct) { + size_t length = iov_iter_count(from); + loff_t pos = iocb->ki_pos; blk_start_plug(&plug); - /* check whether we do a DIO overwrite or not */ if (ext4_should_dioread_nolock(inode) && !aio_mutex && !file->f_mapping->nrpages && pos + length <= i_size_read(inode)) { @@ -185,7 +181,12 @@ ext4_file_write_iter(struct kiocb *iocb, struct iov_iter *from) if (o_direct) blk_finish_plug(&plug); -errout: + if (aio_mutex) + mutex_unlock(aio_mutex); + return ret; + +out: + mutex_unlock(&inode->i_mutex); if (aio_mutex) mutex_unlock(aio_mutex); return ret; diff --git a/fs/ext4/indirect.c b/fs/ext4/indirect.c index 740c787..3580629e 100644 --- a/fs/ext4/indirect.c +++ b/fs/ext4/indirect.c @@ -642,8 +642,8 @@ out: * crashes then stale disk data _may_ be exposed inside the file. But current * VFS code falls back into buffered path in that case so we are safe. */ -ssize_t ext4_ind_direct_IO(int rw, struct kiocb *iocb, - struct iov_iter *iter, loff_t offset) +ssize_t ext4_ind_direct_IO(struct kiocb *iocb, struct iov_iter *iter, + loff_t offset) { struct file *file = iocb->ki_filp; struct inode *inode = file->f_mapping->host; @@ -654,7 +654,7 @@ ssize_t ext4_ind_direct_IO(int rw, struct kiocb *iocb, size_t count = iov_iter_count(iter); int retries = 0; - if (rw == WRITE) { + if (iov_iter_rw(iter) == WRITE) { loff_t final_size = offset + count; if (final_size > inode->i_size) { @@ -676,7 +676,7 @@ ssize_t ext4_ind_direct_IO(int rw, struct kiocb *iocb, } retry: - if (rw == READ && ext4_should_dioread_nolock(inode)) { + if (iov_iter_rw(iter) == READ && ext4_should_dioread_nolock(inode)) { /* * Nolock dioread optimization may be dynamically disabled * via ext4_inode_block_unlocked_dio(). Check inode's state @@ -690,23 +690,24 @@ retry: goto locked; } if (IS_DAX(inode)) - ret = dax_do_io(rw, iocb, inode, iter, offset, + ret = dax_do_io(iocb, inode, iter, offset, ext4_get_block, NULL, 0); else - ret = __blockdev_direct_IO(rw, iocb, inode, - inode->i_sb->s_bdev, iter, offset, - ext4_get_block, NULL, NULL, 0); + ret = __blockdev_direct_IO(iocb, inode, + inode->i_sb->s_bdev, iter, + offset, ext4_get_block, NULL, + NULL, 0); inode_dio_done(inode); } else { locked: if (IS_DAX(inode)) - ret = dax_do_io(rw, iocb, inode, iter, offset, + ret = dax_do_io(iocb, inode, iter, offset, ext4_get_block, NULL, DIO_LOCKING); else - ret = blockdev_direct_IO(rw, iocb, inode, iter, - offset, ext4_get_block); + ret = blockdev_direct_IO(iocb, inode, iter, offset, + ext4_get_block); - if (unlikely((rw & WRITE) && ret < 0)) { + if (unlikely(iov_iter_rw(iter) == WRITE && ret < 0)) { loff_t isize = i_size_read(inode); loff_t end = offset + count; diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 035b7a0..b49cf6e 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -2952,8 +2952,8 @@ static void ext4_end_io_dio(struct kiocb *iocb, loff_t offset, * if the machine crashes during the write. * */ -static ssize_t ext4_ext_direct_IO(int rw, struct kiocb *iocb, - struct iov_iter *iter, loff_t offset) +static ssize_t ext4_ext_direct_IO(struct kiocb *iocb, struct iov_iter *iter, + loff_t offset) { struct file *file = iocb->ki_filp; struct inode *inode = file->f_mapping->host; @@ -2966,8 +2966,8 @@ static ssize_t ext4_ext_direct_IO(int rw, struct kiocb *iocb, ext4_io_end_t *io_end = NULL; /* Use the old path for reads and writes beyond i_size. */ - if (rw != WRITE || final_size > inode->i_size) - return ext4_ind_direct_IO(rw, iocb, iter, offset); + if (iov_iter_rw(iter) != WRITE || final_size > inode->i_size) + return ext4_ind_direct_IO(iocb, iter, offset); BUG_ON(iocb->private == NULL); @@ -2976,7 +2976,7 @@ static ssize_t ext4_ext_direct_IO(int rw, struct kiocb *iocb, * conversion. This also disallows race between truncate() and * overwrite DIO as i_dio_count needs to be incremented under i_mutex. */ - if (rw == WRITE) + if (iov_iter_rw(iter) == WRITE) atomic_inc(&inode->i_dio_count); /* If we do a overwrite dio, i_mutex locking can be released */ @@ -3034,10 +3034,10 @@ static ssize_t ext4_ext_direct_IO(int rw, struct kiocb *iocb, dio_flags = DIO_LOCKING; } if (IS_DAX(inode)) - ret = dax_do_io(rw, iocb, inode, iter, offset, get_block_func, + ret = dax_do_io(iocb, inode, iter, offset, get_block_func, ext4_end_io_dio, dio_flags); else - ret = __blockdev_direct_IO(rw, iocb, inode, + ret = __blockdev_direct_IO(iocb, inode, inode->i_sb->s_bdev, iter, offset, get_block_func, ext4_end_io_dio, NULL, dio_flags); @@ -3078,7 +3078,7 @@ static ssize_t ext4_ext_direct_IO(int rw, struct kiocb *iocb, } retake_lock: - if (rw == WRITE) + if (iov_iter_rw(iter) == WRITE) inode_dio_done(inode); /* take i_mutex locking again if we do a ovewrite dio */ if (overwrite) { @@ -3089,8 +3089,8 @@ retake_lock: return ret; } -static ssize_t ext4_direct_IO(int rw, struct kiocb *iocb, - struct iov_iter *iter, loff_t offset) +static ssize_t ext4_direct_IO(struct kiocb *iocb, struct iov_iter *iter, + loff_t offset) { struct file *file = iocb->ki_filp; struct inode *inode = file->f_mapping->host; @@ -3107,12 +3107,12 @@ static ssize_t ext4_direct_IO(int rw, struct kiocb *iocb, if (ext4_has_inline_data(inode)) return 0; - trace_ext4_direct_IO_enter(inode, offset, count, rw); + trace_ext4_direct_IO_enter(inode, offset, count, iov_iter_rw(iter)); if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS)) - ret = ext4_ext_direct_IO(rw, iocb, iter, offset); + ret = ext4_ext_direct_IO(iocb, iter, offset); else - ret = ext4_ind_direct_IO(rw, iocb, iter, offset); - trace_ext4_direct_IO_exit(inode, offset, count, rw, ret); + ret = ext4_ind_direct_IO(iocb, iter, offset); + trace_ext4_direct_IO_exit(inode, offset, count, iov_iter_rw(iter), ret); return ret; } diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index 497f851..319eda5 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -1118,12 +1118,12 @@ static int f2fs_write_end(struct file *file, return copied; } -static int check_direct_IO(struct inode *inode, int rw, - struct iov_iter *iter, loff_t offset) +static int check_direct_IO(struct inode *inode, struct iov_iter *iter, + loff_t offset) { unsigned blocksize_mask = inode->i_sb->s_blocksize - 1; - if (rw == READ) + if (iov_iter_rw(iter) == READ) return 0; if (offset & blocksize_mask) @@ -1135,8 +1135,8 @@ static int check_direct_IO(struct inode *inode, int rw, return 0; } -static ssize_t f2fs_direct_IO(int rw, struct kiocb *iocb, - struct iov_iter *iter, loff_t offset) +static ssize_t f2fs_direct_IO(struct kiocb *iocb, struct iov_iter *iter, + loff_t offset) { struct file *file = iocb->ki_filp; struct address_space *mapping = file->f_mapping; @@ -1151,19 +1151,19 @@ static ssize_t f2fs_direct_IO(int rw, struct kiocb *iocb, return err; } - if (check_direct_IO(inode, rw, iter, offset)) + if (check_direct_IO(inode, iter, offset)) return 0; - trace_f2fs_direct_IO_enter(inode, offset, count, rw); + trace_f2fs_direct_IO_enter(inode, offset, count, iov_iter_rw(iter)); - if (rw & WRITE) + if (iov_iter_rw(iter) == WRITE) __allocate_data_blocks(inode, offset, count); - err = blockdev_direct_IO(rw, iocb, inode, iter, offset, get_data_block); - if (err < 0 && (rw & WRITE)) + err = blockdev_direct_IO(iocb, inode, iter, offset, get_data_block); + if (err < 0 && iov_iter_rw(iter) == WRITE) f2fs_write_failed(mapping, offset + count); - trace_f2fs_direct_IO_exit(inode, offset, count, rw, err); + trace_f2fs_direct_IO_exit(inode, offset, count, iov_iter_rw(iter), err); return err; } diff --git a/fs/fat/inode.c b/fs/fat/inode.c index 8521207..41b7299 100644 --- a/fs/fat/inode.c +++ b/fs/fat/inode.c @@ -245,8 +245,7 @@ static int fat_write_end(struct file *file, struct address_space *mapping, return err; } -static ssize_t fat_direct_IO(int rw, struct kiocb *iocb, - struct iov_iter *iter, +static ssize_t fat_direct_IO(struct kiocb *iocb, struct iov_iter *iter, loff_t offset) { struct file *file = iocb->ki_filp; @@ -255,7 +254,7 @@ static ssize_t fat_direct_IO(int rw, struct kiocb *iocb, size_t count = iov_iter_count(iter); ssize_t ret; - if (rw == WRITE) { + if (iov_iter_rw(iter) == WRITE) { /* * FIXME: blockdev_direct_IO() doesn't use ->write_begin(), * so we need to update the ->mmu_private to block boundary. @@ -274,8 +273,8 @@ static ssize_t fat_direct_IO(int rw, struct kiocb *iocb, * FAT need to use the DIO_LOCKING for avoiding the race * condition of fat_get_block() and ->truncate(). */ - ret = blockdev_direct_IO(rw, iocb, inode, iter, offset, fat_get_block); - if (ret < 0 && (rw & WRITE)) + ret = blockdev_direct_IO(iocb, inode, iter, offset, fat_get_block); + if (ret < 0 && iov_iter_rw(iter) == WRITE) fat_write_failed(mapping, offset + count); return ret; diff --git a/fs/fuse/file.c b/fs/fuse/file.c index e1afdd7..5ef05b5 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -1145,13 +1145,11 @@ static ssize_t fuse_file_write_iter(struct kiocb *iocb, struct iov_iter *from) { struct file *file = iocb->ki_filp; struct address_space *mapping = file->f_mapping; - size_t count = iov_iter_count(from); ssize_t written = 0; ssize_t written_buffered = 0; struct inode *inode = mapping->host; ssize_t err; loff_t endbyte = 0; - loff_t pos = iocb->ki_pos; if (get_fuse_conn(inode)->writeback_cache) { /* Update size (EOF optimization) and mode (SUID clearing) */ @@ -1167,14 +1165,10 @@ static ssize_t fuse_file_write_iter(struct kiocb *iocb, struct iov_iter *from) /* We can write back this queue in page reclaim */ current->backing_dev_info = inode_to_bdi(inode); - err = generic_write_checks(file, &pos, &count, S_ISBLK(inode->i_mode)); - if (err) - goto out; - - if (count == 0) + err = generic_write_checks(iocb, from); + if (err <= 0) goto out; - iov_iter_truncate(from, count); err = file_remove_suid(file); if (err) goto out; @@ -1183,7 +1177,8 @@ static ssize_t fuse_file_write_iter(struct kiocb *iocb, struct iov_iter *from) if (err) goto out; - if (file->f_flags & O_DIRECT) { + if (iocb->ki_flags & IOCB_DIRECT) { + loff_t pos = iocb->ki_pos; written = generic_file_direct_write(iocb, from, pos); if (written < 0 || !iov_iter_count(from)) goto out; @@ -1209,9 +1204,9 @@ static ssize_t fuse_file_write_iter(struct kiocb *iocb, struct iov_iter *from) written += written_buffered; iocb->ki_pos = pos + written_buffered; } else { - written = fuse_perform_write(file, mapping, from, pos); + written = fuse_perform_write(file, mapping, from, iocb->ki_pos); if (written >= 0) - iocb->ki_pos = pos + written; + iocb->ki_pos += written; } out: current->backing_dev_info = NULL; @@ -1412,7 +1407,6 @@ static ssize_t fuse_direct_write_iter(struct kiocb *iocb, struct iov_iter *from) struct file *file = iocb->ki_filp; struct inode *inode = file_inode(file); struct fuse_io_priv io = { .async = 0, .file = file }; - size_t count = iov_iter_count(from); ssize_t res; if (is_bad_inode(inode)) @@ -1420,11 +1414,9 @@ static ssize_t fuse_direct_write_iter(struct kiocb *iocb, struct iov_iter *from) /* Don't allow parallel writes to the same file */ mutex_lock(&inode->i_mutex); - res = generic_write_checks(file, &iocb->ki_pos, &count, 0); - if (!res) { - iov_iter_truncate(from, count); + res = generic_write_checks(iocb, from); + if (res > 0) 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); @@ -2782,8 +2774,7 @@ static inline loff_t fuse_round_up(loff_t off) } static ssize_t -fuse_direct_IO(int rw, struct kiocb *iocb, struct iov_iter *iter, - loff_t offset) +fuse_direct_IO(struct kiocb *iocb, struct iov_iter *iter, loff_t offset) { DECLARE_COMPLETION_ONSTACK(wait); ssize_t ret = 0; @@ -2800,15 +2791,15 @@ fuse_direct_IO(int rw, struct kiocb *iocb, struct iov_iter *iter, inode = file->f_mapping->host; i_size = i_size_read(inode); - if ((rw == READ) && (offset > i_size)) + if ((iov_iter_rw(iter) == READ) && (offset > i_size)) return 0; /* optimization for short read */ - if (async_dio && rw != WRITE && offset + count > i_size) { + if (async_dio && iov_iter_rw(iter) != WRITE && offset + count > i_size) { if (offset >= i_size) return 0; - count = min_t(loff_t, count, fuse_round_up(i_size - offset)); - iov_iter_truncate(iter, count); + iov_iter_truncate(iter, fuse_round_up(i_size - offset)); + count = iov_iter_count(iter); } io = kmalloc(sizeof(struct fuse_io_priv), GFP_KERNEL); @@ -2819,7 +2810,7 @@ fuse_direct_IO(int rw, struct kiocb *iocb, struct iov_iter *iter, io->bytes = -1; io->size = 0; io->offset = offset; - io->write = (rw == WRITE); + io->write = (iov_iter_rw(iter) == WRITE); io->err = 0; io->file = file; /* @@ -2834,19 +2825,15 @@ fuse_direct_IO(int rw, struct kiocb *iocb, struct iov_iter *iter, * to wait on real async I/O requests, so we must submit this request * synchronously. */ - if (!is_sync_kiocb(iocb) && (offset + count > i_size) && rw == WRITE) + if (!is_sync_kiocb(iocb) && (offset + count > i_size) && + iov_iter_rw(iter) == WRITE) io->async = false; if (io->async && is_sync_kiocb(iocb)) io->done = &wait; - if (rw == WRITE) { - ret = generic_write_checks(file, &pos, &count, 0); - if (!ret) { - iov_iter_truncate(iter, count); - ret = fuse_direct_io(io, iter, &pos, FUSE_DIO_WRITE); - } - + if (iov_iter_rw(iter) == WRITE) { + ret = fuse_direct_io(io, iter, &pos, FUSE_DIO_WRITE); fuse_invalidate_attr(inode); } else { ret = __fuse_direct_read(io, iter, &pos); @@ -2865,7 +2852,7 @@ fuse_direct_IO(int rw, struct kiocb *iocb, struct iov_iter *iter, kfree(io); - if (rw == WRITE) { + if (iov_iter_rw(iter) == WRITE) { if (ret > 0) fuse_write_update_size(inode, pos); else if (ret < 0 && offset + count > i_size) diff --git a/fs/gfs2/aops.c b/fs/gfs2/aops.c index a6e6990..5551fea 100644 --- a/fs/gfs2/aops.c +++ b/fs/gfs2/aops.c @@ -1016,13 +1016,12 @@ out: /** * gfs2_ok_for_dio - check that dio is valid on this file * @ip: The inode - * @rw: READ or WRITE * @offset: The offset at which we are reading or writing * * Returns: 0 (to ignore the i/o request and thus fall back to buffered i/o) * 1 (to accept the i/o request) */ -static int gfs2_ok_for_dio(struct gfs2_inode *ip, int rw, loff_t offset) +static int gfs2_ok_for_dio(struct gfs2_inode *ip, loff_t offset) { /* * Should we return an error here? I can't see that O_DIRECT for @@ -1039,8 +1038,8 @@ static int gfs2_ok_for_dio(struct gfs2_inode *ip, int rw, loff_t offset) -static ssize_t gfs2_direct_IO(int rw, struct kiocb *iocb, - struct iov_iter *iter, loff_t offset) +static ssize_t gfs2_direct_IO(struct kiocb *iocb, struct iov_iter *iter, + loff_t offset) { struct file *file = iocb->ki_filp; struct inode *inode = file->f_mapping->host; @@ -1061,7 +1060,7 @@ static ssize_t gfs2_direct_IO(int rw, struct kiocb *iocb, rv = gfs2_glock_nq(&gh); if (rv) return rv; - rv = gfs2_ok_for_dio(ip, rw, offset); + rv = gfs2_ok_for_dio(ip, offset); if (rv != 1) goto out; /* dio not valid, fall back to buffered i/o */ @@ -1091,13 +1090,12 @@ static ssize_t gfs2_direct_IO(int rw, struct kiocb *iocb, rv = filemap_write_and_wait_range(mapping, lstart, end); if (rv) goto out; - if (rw == WRITE) + if (iov_iter_rw(iter) == WRITE) truncate_inode_pages_range(mapping, lstart, end); } - rv = __blockdev_direct_IO(rw, iocb, inode, inode->i_sb->s_bdev, - iter, offset, - gfs2_get_block_direct, NULL, NULL, 0); + rv = __blockdev_direct_IO(iocb, inode, inode->i_sb->s_bdev, iter, + offset, gfs2_get_block_direct, NULL, NULL, 0); out: gfs2_glock_dq(&gh); gfs2_holder_uninit(&gh); diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c index 207eb4a..3189287 100644 --- a/fs/gfs2/file.c +++ b/fs/gfs2/file.c @@ -709,7 +709,7 @@ static ssize_t gfs2_file_write_iter(struct kiocb *iocb, struct iov_iter *from) gfs2_size_hint(file, iocb->ki_pos, iov_iter_count(from)); - if (file->f_flags & O_APPEND) { + if (iocb->ki_flags & IOCB_APPEND) { struct gfs2_holder gh; ret = gfs2_glock_nq_init(ip->i_gl, LM_ST_SHARED, 0, &gh); diff --git a/fs/hfs/inode.c b/fs/hfs/inode.c index 9337065..75fd5d8 100644 --- a/fs/hfs/inode.c +++ b/fs/hfs/inode.c @@ -124,8 +124,8 @@ static int hfs_releasepage(struct page *page, gfp_t mask) return res ? try_to_free_buffers(page) : 0; } -static ssize_t hfs_direct_IO(int rw, struct kiocb *iocb, - struct iov_iter *iter, loff_t offset) +static ssize_t hfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter, + loff_t offset) { struct file *file = iocb->ki_filp; struct address_space *mapping = file->f_mapping; @@ -133,13 +133,13 @@ static ssize_t hfs_direct_IO(int rw, struct kiocb *iocb, size_t count = iov_iter_count(iter); ssize_t ret; - ret = blockdev_direct_IO(rw, iocb, inode, iter, offset, hfs_get_block); + ret = blockdev_direct_IO(iocb, inode, iter, offset, hfs_get_block); /* * In case of error extending write may have instantiated a few * blocks outside i_size. Trim these off again. */ - if (unlikely((rw & WRITE) && ret < 0)) { + if (unlikely(iov_iter_rw(iter) == WRITE && ret < 0)) { loff_t isize = i_size_read(inode); loff_t end = offset + count; diff --git a/fs/hfsplus/inode.c b/fs/hfsplus/inode.c index 5f86cadb..a43811f 100644 --- a/fs/hfsplus/inode.c +++ b/fs/hfsplus/inode.c @@ -122,8 +122,8 @@ static int hfsplus_releasepage(struct page *page, gfp_t mask) return res ? try_to_free_buffers(page) : 0; } -static ssize_t hfsplus_direct_IO(int rw, struct kiocb *iocb, - struct iov_iter *iter, loff_t offset) +static ssize_t hfsplus_direct_IO(struct kiocb *iocb, struct iov_iter *iter, + loff_t offset) { struct file *file = iocb->ki_filp; struct address_space *mapping = file->f_mapping; @@ -131,14 +131,13 @@ static ssize_t hfsplus_direct_IO(int rw, struct kiocb *iocb, size_t count = iov_iter_count(iter); ssize_t ret; - ret = blockdev_direct_IO(rw, iocb, inode, iter, offset, - hfsplus_get_block); + ret = blockdev_direct_IO(iocb, inode, iter, offset, hfsplus_get_block); /* * In case of error extending write may have instantiated a few * blocks outside i_size. Trim these off again. */ - if (unlikely((rw & WRITE) && ret < 0)) { + if (unlikely(iov_iter_rw(iter) == WRITE && ret < 0)) { loff_t isize = i_size_read(inode); loff_t end = offset + count; diff --git a/fs/jfs/inode.c b/fs/jfs/inode.c index 3197aed..070dc4b 100644 --- a/fs/jfs/inode.c +++ b/fs/jfs/inode.c @@ -330,8 +330,8 @@ static sector_t jfs_bmap(struct address_space *mapping, sector_t block) return generic_block_bmap(mapping, block, jfs_get_block); } -static ssize_t jfs_direct_IO(int rw, struct kiocb *iocb, - struct iov_iter *iter, loff_t offset) +static ssize_t jfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter, + loff_t offset) { struct file *file = iocb->ki_filp; struct address_space *mapping = file->f_mapping; @@ -339,13 +339,13 @@ static ssize_t jfs_direct_IO(int rw, struct kiocb *iocb, size_t count = iov_iter_count(iter); ssize_t ret; - ret = blockdev_direct_IO(rw, iocb, inode, iter, offset, jfs_get_block); + ret = blockdev_direct_IO(iocb, inode, iter, offset, jfs_get_block); /* * In case of error extending write may have instantiated a few * blocks outside i_size. Trim these off again. */ - if (unlikely((rw & WRITE) && ret < 0)) { + if (unlikely(iov_iter_rw(iter) == WRITE && ret < 0)) { loff_t isize = i_size_read(inode); loff_t end = offset + count; @@ -1585,7 +1585,7 @@ static inline int walk_component(struct nameidata *nd, struct path *path, inode = path->dentry->d_inode; } err = -ENOENT; - if (!inode || d_is_negative(path->dentry)) + if (d_is_negative(path->dentry)) goto out_path_put; if (should_follow_link(path->dentry, follow)) { @@ -2310,7 +2310,7 @@ mountpoint_last(struct nameidata *nd, struct path *path) mutex_unlock(&dir->d_inode->i_mutex); done: - if (!dentry->d_inode || d_is_negative(dentry)) { + if (d_is_negative(dentry)) { error = -ENOENT; dput(dentry); goto out; @@ -3038,7 +3038,7 @@ retry_lookup: finish_lookup: /* we _can_ be in RCU mode here */ error = -ENOENT; - if (!inode || d_is_negative(path->dentry)) { + if (d_is_negative(path->dentry)) { path_to_nameidata(path, nd); goto out; } @@ -3077,7 +3077,7 @@ finish_open: error = -ENOTDIR; if ((nd->flags & LOOKUP_DIRECTORY) && !d_can_lookup(nd->path.dentry)) goto out; - if (!S_ISREG(nd->inode->i_mode)) + if (!d_is_reg(nd->path.dentry)) will_truncate = false; if (will_truncate) { diff --git a/fs/ncpfs/file.c b/fs/ncpfs/file.c index 479bf8d..011324c 100644 --- a/fs/ncpfs/file.c +++ b/fs/ncpfs/file.c @@ -170,20 +170,15 @@ ncp_file_write_iter(struct kiocb *iocb, struct iov_iter *from) struct file *file = iocb->ki_filp; struct inode *inode = file_inode(file); size_t already_written = 0; - loff_t pos = iocb->ki_pos; - size_t count = iov_iter_count(from); size_t bufsize; int errno; void *bouncebuffer; + off_t pos; ncp_dbg(1, "enter %pD2\n", file); - errno = generic_write_checks(file, &pos, &count, 0); - if (errno) + errno = generic_write_checks(iocb, from); + if (errno <= 0) return errno; - iov_iter_truncate(from, count); - - if (!count) - return 0; errno = ncp_make_open(inode, O_WRONLY); if (errno) { @@ -201,10 +196,11 @@ ncp_file_write_iter(struct kiocb *iocb, struct iov_iter *from) errno = -EIO; /* -ENOMEM */ goto outrel; } + pos = iocb->ki_pos; while (iov_iter_count(from)) { int written_this_time; size_t to_write = min_t(size_t, - bufsize - ((off_t)pos % bufsize), + bufsize - (pos % bufsize), iov_iter_count(from)); if (copy_from_iter(bouncebuffer, to_write, from) != to_write) { diff --git a/fs/nfs/direct.c b/fs/nfs/direct.c index c3929fb..682f65f 100644 --- a/fs/nfs/direct.c +++ b/fs/nfs/direct.c @@ -240,7 +240,6 @@ static int nfs_direct_cmp_commit_data_verf(struct nfs_direct_req *dreq, /** * nfs_direct_IO - NFS address space operation for direct I/O - * @rw: direction (read or write) * @iocb: target I/O control block * @iov: array of vectors that define I/O buffer * @pos: offset in file to begin the operation @@ -251,7 +250,7 @@ static int nfs_direct_cmp_commit_data_verf(struct nfs_direct_req *dreq, * shunt off direct read and write requests before the VFS gets them, * so this method is only ever called for swap. */ -ssize_t nfs_direct_IO(int rw, struct kiocb *iocb, struct iov_iter *iter, loff_t pos) +ssize_t nfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter, loff_t pos) { struct inode *inode = iocb->ki_filp->f_mapping->host; @@ -267,9 +266,9 @@ ssize_t nfs_direct_IO(int rw, struct kiocb *iocb, struct iov_iter *iter, loff_t #else VM_BUG_ON(iov_iter_count(iter) != PAGE_SIZE); - if (rw == READ) + if (iov_iter_rw(iter) == READ) return nfs_file_direct_read(iocb, iter, pos); - return nfs_file_direct_write(iocb, iter, pos); + return nfs_file_direct_write(iocb, iter); #endif /* CONFIG_NFS_SWAP */ } @@ -960,8 +959,7 @@ static ssize_t nfs_direct_write_schedule_iovec(struct nfs_direct_req *dreq, * Note that O_APPEND is not supported for NFS direct writes, as there * is no atomic O_APPEND write facility in the NFS protocol. */ -ssize_t nfs_file_direct_write(struct kiocb *iocb, struct iov_iter *iter, - loff_t pos) +ssize_t nfs_file_direct_write(struct kiocb *iocb, struct iov_iter *iter) { ssize_t result = -EINVAL; struct file *file = iocb->ki_filp; @@ -969,25 +967,16 @@ ssize_t nfs_file_direct_write(struct kiocb *iocb, struct iov_iter *iter, struct inode *inode = mapping->host; struct nfs_direct_req *dreq; struct nfs_lock_context *l_ctx; - loff_t end; - size_t count = iov_iter_count(iter); - end = (pos + count - 1) >> PAGE_CACHE_SHIFT; - - nfs_add_stats(mapping->host, NFSIOS_DIRECTWRITTENBYTES, count); + loff_t pos, end; dfprintk(FILE, "NFS: direct write(%pD2, %zd@%Ld)\n", - file, count, (long long) pos); + file, iov_iter_count(iter), (long long) iocb->ki_pos); - result = generic_write_checks(file, &pos, &count, 0); - if (result) - goto out; + nfs_add_stats(mapping->host, NFSIOS_DIRECTWRITTENBYTES, + iov_iter_count(iter)); - result = -EINVAL; - if ((ssize_t) count < 0) - goto out; - result = 0; - if (!count) - goto out; + pos = iocb->ki_pos; + end = (pos + iov_iter_count(iter) - 1) >> PAGE_CACHE_SHIFT; mutex_lock(&inode->i_mutex); @@ -1002,7 +991,7 @@ ssize_t nfs_file_direct_write(struct kiocb *iocb, struct iov_iter *iter, goto out_unlock; } - task_io_account_write(count); + task_io_account_write(iov_iter_count(iter)); result = -ENOMEM; dreq = nfs_direct_req_alloc(); @@ -1010,7 +999,7 @@ ssize_t nfs_file_direct_write(struct kiocb *iocb, struct iov_iter *iter, goto out_unlock; dreq->inode = inode; - dreq->bytes_left = count; + dreq->bytes_left = iov_iter_count(iter); dreq->io_start = pos; dreq->ctx = get_nfs_open_context(nfs_file_open_context(iocb->ki_filp)); l_ctx = nfs_get_lock_context(dreq->ctx); @@ -1050,7 +1039,6 @@ out_release: nfs_direct_req_release(dreq); out_unlock: mutex_unlock(&inode->i_mutex); -out: return result; } diff --git a/fs/nfs/file.c b/fs/nfs/file.c index f6a3ade..c40e436 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -170,7 +170,7 @@ nfs_file_read(struct kiocb *iocb, struct iov_iter *to) struct inode *inode = file_inode(iocb->ki_filp); ssize_t result; - if (iocb->ki_filp->f_flags & O_DIRECT) + if (iocb->ki_flags & IOCB_DIRECT) return nfs_file_direct_read(iocb, to, iocb->ki_pos); dprintk("NFS: read(%pD2, %zu@%lu)\n", @@ -674,17 +674,20 @@ ssize_t nfs_file_write(struct kiocb *iocb, struct iov_iter *from) unsigned long written = 0; ssize_t result; size_t count = iov_iter_count(from); - loff_t pos = iocb->ki_pos; result = nfs_key_timeout_notify(file, inode); if (result) return result; - if (file->f_flags & O_DIRECT) - return nfs_file_direct_write(iocb, from, pos); + if (iocb->ki_flags & IOCB_DIRECT) { + result = generic_write_checks(iocb, from); + if (result <= 0) + return result; + return nfs_file_direct_write(iocb, from); + } dprintk("NFS: write(%pD2, %zu@%Ld)\n", - file, count, (long long) pos); + file, count, (long long) iocb->ki_pos); result = -EBUSY; if (IS_SWAPFILE(inode)) @@ -692,7 +695,7 @@ ssize_t nfs_file_write(struct kiocb *iocb, struct iov_iter *from) /* * O_APPEND implies that we must revalidate the file length. */ - if (file->f_flags & O_APPEND) { + if (iocb->ki_flags & IOCB_APPEND) { result = nfs_revalidate_file_size(inode, file); if (result) goto out; diff --git a/fs/nfs/read.c b/fs/nfs/read.c index 568ecf0..b8f5c63 100644 --- a/fs/nfs/read.c +++ b/fs/nfs/read.c @@ -117,15 +117,15 @@ int nfs_readpage_async(struct nfs_open_context *ctx, struct inode *inode, static void nfs_readpage_release(struct nfs_page *req) { - struct inode *d_inode = req->wb_context->dentry->d_inode; + struct inode *inode = req->wb_context->dentry->d_inode; - dprintk("NFS: read done (%s/%llu %d@%lld)\n", d_inode->i_sb->s_id, - (unsigned long long)NFS_FILEID(d_inode), req->wb_bytes, + dprintk("NFS: read done (%s/%llu %d@%lld)\n", inode->i_sb->s_id, + (unsigned long long)NFS_FILEID(inode), req->wb_bytes, (long long)req_offset(req)); if (nfs_page_group_sync_on_bit(req, PG_UNLOCKPAGE)) { if (PageUptodate(req->wb_page)) - nfs_readpage_to_fscache(d_inode, req->wb_page, 0); + nfs_readpage_to_fscache(inode, req->wb_page, 0); unlock_page(req->wb_page); } diff --git a/fs/nilfs2/inode.c b/fs/nilfs2/inode.c index ab4987b..36f057f 100644 --- a/fs/nilfs2/inode.c +++ b/fs/nilfs2/inode.c @@ -305,8 +305,7 @@ static int nilfs_write_end(struct file *file, struct address_space *mapping, } static ssize_t -nilfs_direct_IO(int rw, struct kiocb *iocb, struct iov_iter *iter, - loff_t offset) +nilfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter, loff_t offset) { struct file *file = iocb->ki_filp; struct address_space *mapping = file->f_mapping; @@ -314,18 +313,17 @@ nilfs_direct_IO(int rw, struct kiocb *iocb, struct iov_iter *iter, size_t count = iov_iter_count(iter); ssize_t size; - if (rw == WRITE) + if (iov_iter_rw(iter) == WRITE) return 0; /* Needs synchronization with the cleaner */ - size = blockdev_direct_IO(rw, iocb, inode, iter, offset, - nilfs_get_block); + size = blockdev_direct_IO(iocb, inode, iter, offset, nilfs_get_block); /* * In case of error extending write may have instantiated a few * blocks outside i_size. Trim these off again. */ - if (unlikely((rw & WRITE) && size < 0)) { + if (unlikely(iov_iter_rw(iter) == WRITE && size < 0)) { loff_t isize = i_size_read(inode); loff_t end = offset + count; diff --git a/fs/ntfs/file.c b/fs/ntfs/file.c index 840e95e..7bb487e 100644 --- a/fs/ntfs/file.c +++ b/fs/ntfs/file.c @@ -328,25 +328,25 @@ err_out: return err; } -static ssize_t ntfs_prepare_file_for_write(struct file *file, loff_t *ppos, - size_t *count) +static ssize_t ntfs_prepare_file_for_write(struct kiocb *iocb, + struct iov_iter *from) { loff_t pos; s64 end, ll; ssize_t err; unsigned long flags; + struct file *file = iocb->ki_filp; struct inode *vi = file_inode(file); ntfs_inode *base_ni, *ni = NTFS_I(vi); ntfs_volume *vol = ni->vol; ntfs_debug("Entering for i_ino 0x%lx, attribute type 0x%x, pos " - "0x%llx, count 0x%lx.", vi->i_ino, + "0x%llx, count 0x%zx.", vi->i_ino, (unsigned)le32_to_cpu(ni->type), - (unsigned long long)*ppos, (unsigned long)*count); - /* We can write back this queue in page reclaim. */ - current->backing_dev_info = inode_to_bdi(vi); - err = generic_write_checks(file, ppos, count, S_ISBLK(vi->i_mode)); - if (unlikely(err)) + (unsigned long long)iocb->ki_pos, + iov_iter_count(from)); + err = generic_write_checks(iocb, from); + if (unlikely(err <= 0)) goto out; /* * All checks have passed. Before we start doing any writing we want @@ -379,8 +379,6 @@ static ssize_t ntfs_prepare_file_for_write(struct file *file, loff_t *ppos, err = -EOPNOTSUPP; goto out; } - if (*count == 0) - goto out; base_ni = ni; if (NInoAttr(ni)) base_ni = ni->ext.base_ntfs_ino; @@ -392,9 +390,9 @@ static ssize_t ntfs_prepare_file_for_write(struct file *file, loff_t *ppos, * cannot fail either so there is no need to check the return code. */ file_update_time(file); - pos = *ppos; + pos = iocb->ki_pos; /* The first byte after the last cluster being written to. */ - end = (pos + *count + vol->cluster_size_mask) & + end = (pos + iov_iter_count(from) + vol->cluster_size_mask) & ~(u64)vol->cluster_size_mask; /* * If the write goes beyond the allocated size, extend the allocation @@ -422,7 +420,7 @@ static ssize_t ntfs_prepare_file_for_write(struct file *file, loff_t *ppos, "partially extended.", vi->i_ino, (unsigned) le32_to_cpu(ni->type)); - *count = ll - pos; + iov_iter_truncate(from, ll - pos); } } else { err = ll; @@ -438,7 +436,7 @@ static ssize_t ntfs_prepare_file_for_write(struct file *file, loff_t *ppos, vi->i_ino, (unsigned) le32_to_cpu(ni->type), (int)-err); - *count = ll - pos; + iov_iter_truncate(from, ll - pos); } else { if (err != -ENOSPC) ntfs_error(vi->i_sb, "Cannot perform " @@ -1930,60 +1928,36 @@ again: } /** - * ntfs_file_write_iter_nolock - write data to a file - * @iocb: IO state structure (file, offset, etc.) - * @from: iov_iter with data to write - * - * Basically the same as __generic_file_write_iter() except that it ends - * up calling ntfs_perform_write() instead of generic_perform_write() and that - * O_DIRECT is not implemented. - */ -static ssize_t ntfs_file_write_iter_nolock(struct kiocb *iocb, - struct iov_iter *from) -{ - struct file *file = iocb->ki_filp; - loff_t pos = iocb->ki_pos; - ssize_t written = 0; - ssize_t err; - size_t count = iov_iter_count(from); - - err = ntfs_prepare_file_for_write(file, &pos, &count); - if (count && !err) { - iov_iter_truncate(from, count); - written = ntfs_perform_write(file, from, pos); - if (likely(written >= 0)) - iocb->ki_pos = pos + written; - } - current->backing_dev_info = NULL; - return written ? written : err; -} - -/** * ntfs_file_write_iter - simple wrapper for ntfs_file_write_iter_nolock() * @iocb: IO state structure * @from: iov_iter with data to write * * Basically the same as generic_file_write_iter() except that it ends up - * calling ntfs_file_write_iter_nolock() instead of - * __generic_file_write_iter(). + * up calling ntfs_perform_write() instead of generic_perform_write() and that + * O_DIRECT is not implemented. */ static ssize_t ntfs_file_write_iter(struct kiocb *iocb, struct iov_iter *from) { struct file *file = iocb->ki_filp; struct inode *vi = file_inode(file); - ssize_t ret; + ssize_t written = 0; + ssize_t err; mutex_lock(&vi->i_mutex); - ret = ntfs_file_write_iter_nolock(iocb, from); + /* We can write back this queue in page reclaim. */ + current->backing_dev_info = inode_to_bdi(vi); + err = ntfs_prepare_file_for_write(iocb, from); + if (iov_iter_count(from) && !err) + written = ntfs_perform_write(file, from, iocb->ki_pos); + current->backing_dev_info = NULL; mutex_unlock(&vi->i_mutex); - if (ret > 0) { - ssize_t err; - - err = generic_write_sync(file, iocb->ki_pos - ret, ret); + if (likely(written > 0)) { + err = generic_write_sync(file, iocb->ki_pos, written); if (err < 0) - ret = err; + written = 0; } - return ret; + iocb->ki_pos += written; + return written ? written : err; } /** diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c index 8d2bc84..f906a25 100644 --- a/fs/ocfs2/aops.c +++ b/fs/ocfs2/aops.c @@ -855,10 +855,9 @@ static ssize_t ocfs2_direct_IO_write(struct kiocb *iocb, ocfs2_inode_unlock(inode, 1); } - written = __blockdev_direct_IO(WRITE, iocb, inode, inode->i_sb->s_bdev, - iter, offset, - ocfs2_direct_IO_get_blocks, - ocfs2_dio_end_io, NULL, 0); + written = __blockdev_direct_IO(iocb, inode, inode->i_sb->s_bdev, iter, + offset, ocfs2_direct_IO_get_blocks, + ocfs2_dio_end_io, NULL, 0); if (unlikely(written < 0)) { loff_t i_size = i_size_read(inode); @@ -946,9 +945,7 @@ out: return ret; } -static ssize_t ocfs2_direct_IO(int rw, - struct kiocb *iocb, - struct iov_iter *iter, +static ssize_t ocfs2_direct_IO(struct kiocb *iocb, struct iov_iter *iter, loff_t offset) { struct file *file = iocb->ki_filp; @@ -970,12 +967,11 @@ static ssize_t ocfs2_direct_IO(int rw, if (i_size_read(inode) <= offset && !full_coherency) return 0; - if (rw == READ) - return __blockdev_direct_IO(rw, iocb, inode, - inode->i_sb->s_bdev, - iter, offset, - ocfs2_direct_IO_get_blocks, - ocfs2_dio_end_io, NULL, 0); + if (iov_iter_rw(iter) == READ) + return __blockdev_direct_IO(iocb, inode, inode->i_sb->s_bdev, + iter, offset, + ocfs2_direct_IO_get_blocks, + ocfs2_dio_end_io, NULL, 0); else return ocfs2_direct_IO_write(iocb, iter, offset); } diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index 8c48e98..913fc250 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -2106,7 +2106,7 @@ out: } static int ocfs2_prepare_inode_for_write(struct file *file, - loff_t *ppos, + loff_t pos, size_t count, int appending, int *direct_io, @@ -2115,7 +2115,7 @@ static int ocfs2_prepare_inode_for_write(struct file *file, int ret = 0, meta_level = 0; struct dentry *dentry = file->f_path.dentry; struct inode *inode = dentry->d_inode; - loff_t saved_pos = 0, end; + loff_t end; struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); int full_coherency = !(osb->s_mount_opt & OCFS2_MOUNT_COHERENCY_BUFFERED); @@ -2155,23 +2155,16 @@ static int ocfs2_prepare_inode_for_write(struct file *file, } } - /* work on a copy of ppos until we're sure that we won't have - * to recalculate it due to relocking. */ - if (appending) - saved_pos = i_size_read(inode); - else - saved_pos = *ppos; - - end = saved_pos + count; + end = pos + count; - ret = ocfs2_check_range_for_refcount(inode, saved_pos, count); + ret = ocfs2_check_range_for_refcount(inode, pos, count); if (ret == 1) { ocfs2_inode_unlock(inode, meta_level); meta_level = -1; ret = ocfs2_prepare_inode_for_refcount(inode, file, - saved_pos, + pos, count, &meta_level); if (has_refcount) @@ -2227,7 +2220,7 @@ static int ocfs2_prepare_inode_for_write(struct file *file, * caller will have to retake some cluster * locks and initiate the io as buffered. */ - ret = ocfs2_check_range_for_holes(inode, saved_pos, count); + ret = ocfs2_check_range_for_holes(inode, pos, count); if (ret == 1) { /* * Fallback to old way if the feature bit is not set. @@ -2242,12 +2235,9 @@ static int ocfs2_prepare_inode_for_write(struct file *file, break; } - if (appending) - *ppos = saved_pos; - out_unlock: trace_ocfs2_prepare_inode_for_write(OCFS2_I(inode)->ip_blkno, - saved_pos, appending, count, + pos, appending, count, direct_io, has_refcount); if (meta_level >= 0) @@ -2260,19 +2250,20 @@ out: static ssize_t ocfs2_file_write_iter(struct kiocb *iocb, struct iov_iter *from) { - int ret, direct_io, appending, rw_level, have_alloc_sem = 0; + int direct_io, appending, rw_level, have_alloc_sem = 0; int can_do_direct, has_refcount = 0; ssize_t written = 0; - size_t count = iov_iter_count(from); - loff_t old_size, *ppos = &iocb->ki_pos; + ssize_t ret; + size_t count = iov_iter_count(from), orig_count; + loff_t old_size; u32 old_clusters; struct file *file = iocb->ki_filp; struct inode *inode = file_inode(file); - struct address_space *mapping = file->f_mapping; struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); int full_coherency = !(osb->s_mount_opt & OCFS2_MOUNT_COHERENCY_BUFFERED); int unaligned_dio = 0; + int dropped_dio = 0; trace_ocfs2_file_aio_write(inode, file, file->f_path.dentry, (unsigned long long)OCFS2_I(inode)->ip_blkno, @@ -2283,8 +2274,8 @@ static ssize_t ocfs2_file_write_iter(struct kiocb *iocb, if (count == 0) return 0; - appending = file->f_flags & O_APPEND ? 1 : 0; - direct_io = file->f_flags & O_DIRECT ? 1 : 0; + appending = iocb->ki_flags & IOCB_APPEND ? 1 : 0; + direct_io = iocb->ki_flags & IOCB_DIRECT ? 1 : 0; mutex_lock(&inode->i_mutex); @@ -2329,8 +2320,17 @@ relock: ocfs2_inode_unlock(inode, 1); } + orig_count = iov_iter_count(from); + ret = generic_write_checks(iocb, from); + if (ret <= 0) { + if (ret) + mlog_errno(ret); + goto out; + } + count = ret; + can_do_direct = direct_io; - ret = ocfs2_prepare_inode_for_write(file, ppos, count, appending, + ret = ocfs2_prepare_inode_for_write(file, iocb->ki_pos, count, appending, &can_do_direct, &has_refcount); if (ret < 0) { mlog_errno(ret); @@ -2338,7 +2338,7 @@ relock: } if (direct_io && !is_sync_kiocb(iocb)) - unaligned_dio = ocfs2_is_io_unaligned(inode, count, *ppos); + unaligned_dio = ocfs2_is_io_unaligned(inode, count, iocb->ki_pos); /* * We can't complete the direct I/O as requested, fall back to @@ -2351,6 +2351,9 @@ relock: rw_level = -1; direct_io = 0; + iocb->ki_flags &= ~IOCB_DIRECT; + iov_iter_reexpand(from, orig_count); + dropped_dio = 1; goto relock; } @@ -2374,74 +2377,15 @@ relock: /* communicate with ocfs2_dio_end_io */ ocfs2_iocb_set_rw_locked(iocb, rw_level); - ret = generic_write_checks(file, ppos, &count, - S_ISBLK(inode->i_mode)); - if (ret) - goto out_dio; - - iov_iter_truncate(from, count); - if (direct_io) { - loff_t endbyte; - ssize_t written_buffered; - written = generic_file_direct_write(iocb, from, *ppos); - if (written < 0 || written == count) { - ret = written; - goto out_dio; - } - - /* - * for completing the rest of the request. - */ - count -= written; - written_buffered = generic_perform_write(file, from, *ppos); - /* - * If generic_file_buffered_write() returned a synchronous error - * then we want to return the number of bytes which were - * direct-written, or the error code if that was zero. Note - * that this differs from normal direct-io semantics, which - * will return -EFOO even if some bytes were written. - */ - if (written_buffered < 0) { - ret = written_buffered; - goto out_dio; - } - - /* We need to ensure that the page cache pages are written to - * disk and invalidated to preserve the expected O_DIRECT - * semantics. - */ - endbyte = *ppos + written_buffered - 1; - ret = filemap_write_and_wait_range(file->f_mapping, *ppos, - endbyte); - if (ret == 0) { - iocb->ki_pos = *ppos + written_buffered; - written += written_buffered; - invalidate_mapping_pages(mapping, - *ppos >> PAGE_CACHE_SHIFT, - endbyte >> PAGE_CACHE_SHIFT); - } else { - /* - * We don't know how much we wrote, so just return - * the number of bytes which were direct-written - */ - } - } else { - current->backing_dev_info = inode_to_bdi(inode); - written = generic_perform_write(file, from, *ppos); - if (likely(written >= 0)) - iocb->ki_pos = *ppos + written; - current->backing_dev_info = NULL; - } - -out_dio: + written = __generic_file_write_iter(iocb, from); /* buffered aio wouldn't have proper lock coverage today */ - BUG_ON(ret == -EIOCBQUEUED && !(file->f_flags & O_DIRECT)); + BUG_ON(written == -EIOCBQUEUED && !(iocb->ki_flags & IOCB_DIRECT)); if (unlikely(written <= 0)) goto no_sync; - if (((file->f_flags & O_DSYNC) && !direct_io) || IS_SYNC(inode) || - ((file->f_flags & O_DIRECT) && !direct_io)) { + if (((file->f_flags & O_DSYNC) && !direct_io) || + IS_SYNC(inode) || dropped_dio) { ret = filemap_fdatawrite_range(file->f_mapping, iocb->ki_pos - written, iocb->ki_pos - 1); @@ -2552,7 +2496,7 @@ static ssize_t ocfs2_file_read_iter(struct kiocb *iocb, * buffered reads protect themselves in ->readpage(). O_DIRECT reads * need locks to protect pending reads from racing with truncate. */ - if (filp->f_flags & O_DIRECT) { + if (iocb->ki_flags & IOCB_DIRECT) { have_alloc_sem = 1; ocfs2_iocb_set_sem_locked(iocb); @@ -2586,7 +2530,7 @@ static ssize_t ocfs2_file_read_iter(struct kiocb *iocb, trace_generic_file_aio_read_ret(ret); /* buffered aio wouldn't have proper lock coverage today */ - BUG_ON(ret == -EIOCBQUEUED && !(filp->f_flags & O_DIRECT)); + BUG_ON(ret == -EIOCBQUEUED && !(iocb->ki_flags & IOCB_DIRECT)); /* see ocfs2_file_write_iter */ if (ret == -EIOCBQUEUED || !ocfs2_iocb_is_rw_locked(iocb)) { diff --git a/fs/read_write.c b/fs/read_write.c index 45d583c..819ef3f 100644 --- a/fs/read_write.c +++ b/fs/read_write.c @@ -477,7 +477,8 @@ static ssize_t new_sync_write(struct file *filp, const char __user *buf, size_t ret = filp->f_op->write_iter(&kiocb, &iter); BUG_ON(ret == -EIOCBQUEUED); - *ppos = kiocb.ki_pos; + if (ret > 0) + *ppos = kiocb.ki_pos; return ret; } diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c index 9312b78..742242b 100644 --- a/fs/reiserfs/inode.c +++ b/fs/reiserfs/inode.c @@ -3278,22 +3278,22 @@ static int reiserfs_releasepage(struct page *page, gfp_t unused_gfp_flags) * We thank Mingming Cao for helping us understand in great detail what * to do in this section of the code. */ -static ssize_t reiserfs_direct_IO(int rw, struct kiocb *iocb, - struct iov_iter *iter, loff_t offset) +static ssize_t reiserfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter, + loff_t offset) { struct file *file = iocb->ki_filp; struct inode *inode = file->f_mapping->host; size_t count = iov_iter_count(iter); ssize_t ret; - ret = blockdev_direct_IO(rw, iocb, inode, iter, offset, + ret = blockdev_direct_IO(iocb, inode, iter, offset, reiserfs_get_blocks_direct_io); /* * In case of error extending write may have instantiated a few * blocks outside i_size. Trim these off again. */ - if (unlikely((rw & WRITE) && ret < 0)) { + if (unlikely(iov_iter_rw(iter) == WRITE && ret < 0)) { loff_t isize = i_size_read(inode); loff_t end = offset + count; diff --git a/fs/udf/file.c b/fs/udf/file.c index f77f768..5dadad9 100644 --- a/fs/udf/file.c +++ b/fs/udf/file.c @@ -99,8 +99,7 @@ static int udf_adinicb_write_begin(struct file *file, return 0; } -static ssize_t udf_adinicb_direct_IO(int rw, struct kiocb *iocb, - struct iov_iter *iter, +static ssize_t udf_adinicb_direct_IO(struct kiocb *iocb, struct iov_iter *iter, loff_t offset) { /* Fallback to buffered I/O. */ @@ -120,21 +119,21 @@ static ssize_t udf_file_write_iter(struct kiocb *iocb, struct iov_iter *from) ssize_t retval; struct file *file = iocb->ki_filp; struct inode *inode = file_inode(file); - int err, pos; - size_t count = iov_iter_count(from); struct udf_inode_info *iinfo = UDF_I(inode); + int err; mutex_lock(&inode->i_mutex); + + retval = generic_write_checks(iocb, from); + if (retval <= 0) + goto out; + down_write(&iinfo->i_data_sem); if (iinfo->i_alloc_type == ICBTAG_FLAG_AD_IN_ICB) { - if (file->f_flags & O_APPEND) - pos = inode->i_size; - else - pos = iocb->ki_pos; + loff_t end = iocb->ki_pos + iov_iter_count(from); if (inode->i_sb->s_blocksize < - (udf_file_entry_alloc_offset(inode) + - pos + count)) { + (udf_file_entry_alloc_offset(inode) + end)) { err = udf_expand_file_adinicb(inode); if (err) { mutex_unlock(&inode->i_mutex); @@ -142,16 +141,14 @@ static ssize_t udf_file_write_iter(struct kiocb *iocb, struct iov_iter *from) return err; } } else { - if (pos + count > inode->i_size) - iinfo->i_lenAlloc = pos + count; - else - iinfo->i_lenAlloc = inode->i_size; + iinfo->i_lenAlloc = max(end, inode->i_size); up_write(&iinfo->i_data_sem); } } else up_write(&iinfo->i_data_sem); retval = __generic_file_write_iter(iocb, from); +out: mutex_unlock(&inode->i_mutex); if (retval > 0) { diff --git a/fs/udf/inode.c b/fs/udf/inode.c index 9e3d780..6afac3d 100644 --- a/fs/udf/inode.c +++ b/fs/udf/inode.c @@ -214,8 +214,7 @@ static int udf_write_begin(struct file *file, struct address_space *mapping, return ret; } -static ssize_t udf_direct_IO(int rw, struct kiocb *iocb, - struct iov_iter *iter, +static ssize_t udf_direct_IO(struct kiocb *iocb, struct iov_iter *iter, loff_t offset) { struct file *file = iocb->ki_filp; @@ -224,8 +223,8 @@ static ssize_t udf_direct_IO(int rw, struct kiocb *iocb, size_t count = iov_iter_count(iter); ssize_t ret; - ret = blockdev_direct_IO(rw, iocb, inode, iter, offset, udf_get_block); - if (unlikely(ret < 0 && (rw & WRITE))) + ret = blockdev_direct_IO(iocb, inode, iter, offset, udf_get_block); + if (unlikely(ret < 0 && iov_iter_rw(iter) == WRITE)) udf_write_failed(mapping, offset + count); return ret; } diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c index 4f8cdc5..1d8eef9 100644 --- a/fs/xfs/xfs_aops.c +++ b/fs/xfs/xfs_aops.c @@ -1495,7 +1495,6 @@ xfs_end_io_direct_write( STATIC ssize_t xfs_vm_direct_IO( - int rw, struct kiocb *iocb, struct iov_iter *iter, loff_t offset) @@ -1503,15 +1502,14 @@ xfs_vm_direct_IO( struct inode *inode = iocb->ki_filp->f_mapping->host; struct block_device *bdev = xfs_find_bdev_for_inode(inode); - if (rw & WRITE) { - return __blockdev_direct_IO(rw, iocb, inode, bdev, iter, - offset, xfs_get_blocks_direct, + if (iov_iter_rw(iter) == WRITE) { + return __blockdev_direct_IO(iocb, inode, bdev, iter, offset, + xfs_get_blocks_direct, xfs_end_io_direct_write, NULL, DIO_ASYNC_EXTEND); } - return __blockdev_direct_IO(rw, iocb, inode, bdev, iter, - offset, xfs_get_blocks_direct, - NULL, NULL, 0); + return __blockdev_direct_IO(iocb, inode, bdev, iter, offset, + xfs_get_blocks_direct, NULL, NULL, 0); } /* diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index 44856c3..1f12ad0 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -279,7 +279,7 @@ xfs_file_read_iter( XFS_STATS_INC(xs_read_calls); - if (unlikely(file->f_flags & O_DIRECT)) + if (unlikely(iocb->ki_flags & IOCB_DIRECT)) ioflags |= XFS_IO_ISDIRECT; if (file->f_mode & FMODE_NOCMTIME) ioflags |= XFS_IO_INVIS; @@ -544,18 +544,19 @@ xfs_zero_eof( */ STATIC ssize_t xfs_file_aio_write_checks( - struct file *file, - loff_t *pos, - size_t *count, + struct kiocb *iocb, + struct iov_iter *from, int *iolock) { + struct file *file = iocb->ki_filp; struct inode *inode = file->f_mapping->host; struct xfs_inode *ip = XFS_I(inode); - int error = 0; + ssize_t error = 0; + size_t count = iov_iter_count(from); restart: - error = generic_write_checks(file, pos, count, S_ISBLK(inode->i_mode)); - if (error) + error = generic_write_checks(iocb, from); + if (error <= 0) return error; error = xfs_break_layouts(inode, iolock); @@ -569,16 +570,17 @@ restart: * iolock shared, we need to update it to exclusive which implies * having to redo all checks before. */ - if (*pos > i_size_read(inode)) { + if (iocb->ki_pos > i_size_read(inode)) { bool zero = false; if (*iolock == XFS_IOLOCK_SHARED) { xfs_rw_iunlock(ip, *iolock); *iolock = XFS_IOLOCK_EXCL; xfs_rw_ilock(ip, *iolock); + iov_iter_reexpand(from, count); goto restart; } - error = xfs_zero_eof(ip, *pos, i_size_read(inode), &zero); + error = xfs_zero_eof(ip, iocb->ki_pos, i_size_read(inode), &zero); if (error) return error; } @@ -678,10 +680,11 @@ xfs_file_dio_aio_write( xfs_rw_ilock(ip, iolock); } - ret = xfs_file_aio_write_checks(file, &pos, &count, &iolock); + ret = xfs_file_aio_write_checks(iocb, from, &iolock); if (ret) goto out; - iov_iter_truncate(from, count); + count = iov_iter_count(from); + pos = iocb->ki_pos; if (mapping->nrpages) { ret = filemap_write_and_wait_range(VFS_I(ip)->i_mapping, @@ -734,24 +737,22 @@ xfs_file_buffered_aio_write( ssize_t ret; int enospc = 0; int iolock = XFS_IOLOCK_EXCL; - loff_t pos = iocb->ki_pos; - size_t count = iov_iter_count(from); xfs_rw_ilock(ip, iolock); - ret = xfs_file_aio_write_checks(file, &pos, &count, &iolock); + ret = xfs_file_aio_write_checks(iocb, from, &iolock); if (ret) goto out; - iov_iter_truncate(from, count); /* We can write back this queue in page reclaim */ current->backing_dev_info = inode_to_bdi(inode); write_retry: - trace_xfs_file_buffered_write(ip, count, iocb->ki_pos, 0); - ret = generic_perform_write(file, from, pos); + trace_xfs_file_buffered_write(ip, iov_iter_count(from), + iocb->ki_pos, 0); + ret = generic_perform_write(file, from, iocb->ki_pos); if (likely(ret >= 0)) - iocb->ki_pos = pos + ret; + iocb->ki_pos += ret; /* * If we hit a space limit, try to free up some lingering preallocated @@ -803,7 +804,7 @@ xfs_file_write_iter( if (XFS_FORCED_SHUTDOWN(ip->i_mount)) return -EIO; - if (unlikely(file->f_flags & O_DIRECT)) + if (unlikely(iocb->ki_flags & IOCB_DIRECT)) ret = xfs_file_dio_aio_write(iocb, from); else ret = xfs_file_buffered_aio_write(iocb, from); diff --git a/include/linux/dcache.h b/include/linux/dcache.h index d835879..df334cb 100644 --- a/include/linux/dcache.h +++ b/include/linux/dcache.h @@ -404,26 +404,11 @@ static inline bool d_mountpoint(const struct dentry *dentry) /* * Directory cache entry type accessor functions. */ -static inline void __d_set_type(struct dentry *dentry, unsigned type) -{ - dentry->d_flags = (dentry->d_flags & ~DCACHE_ENTRY_TYPE) | type; -} - -static inline void __d_clear_type(struct dentry *dentry) -{ - __d_set_type(dentry, DCACHE_MISS_TYPE); -} - -static inline void d_set_type(struct dentry *dentry, unsigned type) -{ - spin_lock(&dentry->d_lock); - __d_set_type(dentry, type); - spin_unlock(&dentry->d_lock); -} - static inline unsigned __d_entry_type(const struct dentry *dentry) { - return dentry->d_flags & DCACHE_ENTRY_TYPE; + unsigned type = READ_ONCE(dentry->d_flags); + smp_rmb(); + return type & DCACHE_ENTRY_TYPE; } static inline bool d_is_miss(const struct dentry *dentry) @@ -482,6 +467,44 @@ static inline bool d_is_positive(const struct dentry *dentry) return !d_is_negative(dentry); } +/** + * d_really_is_negative - Determine if a dentry is really negative (ignoring fallthroughs) + * @dentry: The dentry in question + * + * Returns true if the dentry represents either an absent name or a name that + * doesn't map to an inode (ie. ->d_inode is NULL). The dentry could represent + * a true miss, a whiteout that isn't represented by a 0,0 chardev or a + * fallthrough marker in an opaque directory. + * + * Note! (1) This should be used *only* by a filesystem to examine its own + * dentries. It should not be used to look at some other filesystem's + * dentries. (2) It should also be used in combination with d_inode() to get + * the inode. (3) The dentry may have something attached to ->d_lower and the + * type field of the flags may be set to something other than miss or whiteout. + */ +static inline bool d_really_is_negative(const struct dentry *dentry) +{ + return dentry->d_inode == NULL; +} + +/** + * d_really_is_positive - Determine if a dentry is really positive (ignoring fallthroughs) + * @dentry: The dentry in question + * + * Returns true if the dentry represents a name that maps to an inode + * (ie. ->d_inode is not NULL). The dentry might still represent a whiteout if + * that is represented on medium as a 0,0 chardev. + * + * Note! (1) This should be used *only* by a filesystem to examine its own + * dentries. It should not be used to look at some other filesystem's + * dentries. (2) It should also be used in combination with d_inode() to get + * the inode. + */ +static inline bool d_really_is_positive(const struct dentry *dentry) +{ + return dentry->d_inode != NULL; +} + extern void d_set_fallthru(struct dentry *dentry); static inline bool d_is_fallthru(const struct dentry *dentry) diff --git a/include/linux/fs.h b/include/linux/fs.h index f4fc607..f4d6354 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -315,6 +315,8 @@ struct address_space; struct writeback_control; #define IOCB_EVENTFD (1 << 0) +#define IOCB_APPEND (1 << 1) +#define IOCB_DIRECT (1 << 2) struct kiocb { struct file *ki_filp; @@ -329,10 +331,13 @@ static inline bool is_sync_kiocb(struct kiocb *kiocb) return kiocb->ki_complete == NULL; } +static inline int iocb_flags(struct file *file); + static inline void init_sync_kiocb(struct kiocb *kiocb, struct file *filp) { *kiocb = (struct kiocb) { .ki_filp = filp, + .ki_flags = iocb_flags(filp), }; } @@ -383,7 +388,7 @@ struct address_space_operations { void (*invalidatepage) (struct page *, unsigned int, unsigned int); int (*releasepage) (struct page *, gfp_t); void (*freepage)(struct page *); - ssize_t (*direct_IO)(int, struct kiocb *, struct iov_iter *iter, loff_t offset); + ssize_t (*direct_IO)(struct kiocb *, struct iov_iter *iter, loff_t offset); /* * migrate the contents of a page to the specified target. If * migrate_mode is MIGRATE_ASYNC, it must not block. @@ -2566,7 +2571,7 @@ extern int sb_min_blocksize(struct super_block *, int); extern int generic_file_mmap(struct file *, struct vm_area_struct *); extern int generic_file_readonly_mmap(struct file *, struct vm_area_struct *); -int generic_write_checks(struct file *file, loff_t *pos, size_t *count, int isblk); +extern ssize_t generic_write_checks(struct kiocb *, struct iov_iter *); extern ssize_t generic_file_read_iter(struct kiocb *, struct iov_iter *); extern ssize_t __generic_file_write_iter(struct kiocb *, struct iov_iter *); extern ssize_t generic_file_write_iter(struct kiocb *, struct iov_iter *); @@ -2609,8 +2614,8 @@ extern loff_t fixed_size_llseek(struct file *file, loff_t offset, extern int generic_file_open(struct inode * inode, struct file * filp); extern int nonseekable_open(struct inode * inode, struct file * filp); -ssize_t dax_do_io(int rw, struct kiocb *, struct inode *, struct iov_iter *, - loff_t, get_block_t, dio_iodone_t, int flags); +ssize_t dax_do_io(struct kiocb *, struct inode *, struct iov_iter *, loff_t, + get_block_t, dio_iodone_t, int flags); int dax_clear_blocks(struct inode *, sector_t block, long size); int dax_zero_page_range(struct inode *, loff_t from, unsigned len, get_block_t); int dax_truncate_page(struct inode *, loff_t from, get_block_t); @@ -2635,16 +2640,18 @@ enum { void dio_end_io(struct bio *bio, int error); -ssize_t __blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode, - struct block_device *bdev, struct iov_iter *iter, loff_t offset, - get_block_t get_block, dio_iodone_t end_io, - dio_submit_t submit_io, int flags); +ssize_t __blockdev_direct_IO(struct kiocb *iocb, struct inode *inode, + struct block_device *bdev, struct iov_iter *iter, + loff_t offset, get_block_t get_block, + dio_iodone_t end_io, dio_submit_t submit_io, + int flags); -static inline ssize_t blockdev_direct_IO(int rw, struct kiocb *iocb, - struct inode *inode, struct iov_iter *iter, loff_t offset, - get_block_t get_block) +static inline ssize_t blockdev_direct_IO(struct kiocb *iocb, + struct inode *inode, + struct iov_iter *iter, loff_t offset, + get_block_t get_block) { - return __blockdev_direct_IO(rw, iocb, inode, inode->i_sb->s_bdev, iter, + return __blockdev_direct_IO(iocb, inode, inode->i_sb->s_bdev, iter, offset, get_block, NULL, NULL, DIO_LOCKING | DIO_SKIP_HOLES); } @@ -2777,6 +2784,16 @@ static inline bool io_is_direct(struct file *filp) return (filp->f_flags & O_DIRECT) || IS_DAX(file_inode(filp)); } +static inline int iocb_flags(struct file *file) +{ + int res = 0; + if (file->f_flags & O_APPEND) + res |= IOCB_APPEND; + if (io_is_direct(file)) + res |= IOCB_DIRECT; + return res; +} + static inline ino_t parent_ino(struct dentry *dentry) { ino_t res; diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h index b01ccf3..410abd1 100644 --- a/include/linux/nfs_fs.h +++ b/include/linux/nfs_fs.h @@ -447,13 +447,12 @@ static inline struct rpc_cred *nfs_file_cred(struct file *file) /* * linux/fs/nfs/direct.c */ -extern ssize_t nfs_direct_IO(int, struct kiocb *, struct iov_iter *, loff_t); +extern ssize_t nfs_direct_IO(struct kiocb *, struct iov_iter *, loff_t); extern ssize_t nfs_file_direct_read(struct kiocb *iocb, struct iov_iter *iter, loff_t pos); extern ssize_t nfs_file_direct_write(struct kiocb *iocb, - struct iov_iter *iter, - loff_t pos); + struct iov_iter *iter); /* * linux/fs/nfs/dir.c diff --git a/include/linux/uio.h b/include/linux/uio.h index 15f11fb..8b01e1c 100644 --- a/include/linux/uio.h +++ b/include/linux/uio.h @@ -112,6 +112,14 @@ static inline bool iter_is_iovec(struct iov_iter *i) } /* + * Get one of READ or WRITE out of iter->type without any other flags OR'd in + * with it. + * + * The ?: is just for type safety. + */ +#define iov_iter_rw(i) ((0 ? (struct iov_iter *)0 : (i))->type & RW_MASK) + +/* * Cap the iov_iter by given limit; note that the second argument is * *not* the new size - it's upper limit for such. Passing it a value * greater than the amount of data in iov_iter is fine - it'll just do diff --git a/mm/filemap.c b/mm/filemap.c index 12548d0..6bf5e42 100644 --- a/mm/filemap.c +++ b/mm/filemap.c @@ -1693,7 +1693,7 @@ generic_file_read_iter(struct kiocb *iocb, struct iov_iter *iter) loff_t *ppos = &iocb->ki_pos; loff_t pos = *ppos; - if (io_is_direct(file)) { + if (iocb->ki_flags & IOCB_DIRECT) { struct address_space *mapping = file->f_mapping; struct inode *inode = mapping->host; size_t count = iov_iter_count(iter); @@ -1706,7 +1706,7 @@ generic_file_read_iter(struct kiocb *iocb, struct iov_iter *iter) pos + count - 1); if (!retval) { struct iov_iter data = *iter; - retval = mapping->a_ops->direct_IO(READ, iocb, &data, pos); + retval = mapping->a_ops->direct_IO(iocb, &data, pos); } if (retval > 0) { @@ -2259,41 +2259,38 @@ EXPORT_SYMBOL(read_cache_page_gfp); * Returns appropriate error code that caller should return or * zero in case that write should be allowed. */ -inline int generic_write_checks(struct file *file, loff_t *pos, size_t *count, int isblk) +inline ssize_t generic_write_checks(struct kiocb *iocb, struct iov_iter *from) { + struct file *file = iocb->ki_filp; struct inode *inode = file->f_mapping->host; unsigned long limit = rlimit(RLIMIT_FSIZE); + loff_t pos; - if (unlikely(*pos < 0)) - return -EINVAL; + if (!iov_iter_count(from)) + return 0; - if (!isblk) { - /* FIXME: this is for backwards compatibility with 2.4 */ - if (file->f_flags & O_APPEND) - *pos = i_size_read(inode); + /* FIXME: this is for backwards compatibility with 2.4 */ + if (iocb->ki_flags & IOCB_APPEND) + iocb->ki_pos = i_size_read(inode); - if (limit != RLIM_INFINITY) { - if (*pos >= limit) { - send_sig(SIGXFSZ, current, 0); - return -EFBIG; - } - if (*count > limit - (typeof(limit))*pos) { - *count = limit - (typeof(limit))*pos; - } + pos = iocb->ki_pos; + + if (limit != RLIM_INFINITY) { + if (iocb->ki_pos >= limit) { + send_sig(SIGXFSZ, current, 0); + return -EFBIG; } + iov_iter_truncate(from, limit - (unsigned long)pos); } /* * LFS rule */ - if (unlikely(*pos + *count > MAX_NON_LFS && + if (unlikely(pos + iov_iter_count(from) > MAX_NON_LFS && !(file->f_flags & O_LARGEFILE))) { - if (*pos >= MAX_NON_LFS) { + if (pos >= MAX_NON_LFS) return -EFBIG; - } - if (*count > MAX_NON_LFS - (unsigned long)*pos) { - *count = MAX_NON_LFS - (unsigned long)*pos; - } + iov_iter_truncate(from, MAX_NON_LFS - (unsigned long)pos); } /* @@ -2303,34 +2300,11 @@ inline int generic_write_checks(struct file *file, loff_t *pos, size_t *count, i * exceeded without writing data we send a signal and return EFBIG. * Linus frestrict idea will clean these up nicely.. */ - if (likely(!isblk)) { - if (unlikely(*pos >= inode->i_sb->s_maxbytes)) { - if (*count || *pos > inode->i_sb->s_maxbytes) { - return -EFBIG; - } - /* zero-length writes at ->s_maxbytes are OK */ - } + if (unlikely(pos >= inode->i_sb->s_maxbytes)) + return -EFBIG; - if (unlikely(*pos + *count > inode->i_sb->s_maxbytes)) - *count = inode->i_sb->s_maxbytes - *pos; - } else { -#ifdef CONFIG_BLOCK - loff_t isize; - if (bdev_read_only(I_BDEV(inode))) - return -EPERM; - isize = i_size_read(inode); - if (*pos >= isize) { - if (*count || *pos > isize) - return -ENOSPC; - } - - if (*pos + *count > isize) - *count = isize - *pos; -#else - return -EPERM; -#endif - } - return 0; + iov_iter_truncate(from, inode->i_sb->s_maxbytes - pos); + return iov_iter_count(from); } EXPORT_SYMBOL(generic_write_checks); @@ -2394,7 +2368,7 @@ generic_file_direct_write(struct kiocb *iocb, struct iov_iter *from, loff_t pos) } data = *from; - written = mapping->a_ops->direct_IO(WRITE, iocb, &data, pos); + written = mapping->a_ops->direct_IO(iocb, &data, pos); /* * Finally, try again to invalidate clean pages which might have been @@ -2556,23 +2530,12 @@ ssize_t __generic_file_write_iter(struct kiocb *iocb, struct iov_iter *from) struct file *file = iocb->ki_filp; struct address_space * mapping = file->f_mapping; struct inode *inode = mapping->host; - loff_t pos = iocb->ki_pos; ssize_t written = 0; ssize_t err; ssize_t status; - size_t count = iov_iter_count(from); /* We can write back this queue in page reclaim */ current->backing_dev_info = inode_to_bdi(inode); - err = generic_write_checks(file, &pos, &count, S_ISBLK(inode->i_mode)); - if (err) - goto out; - - if (count == 0) - goto out; - - iov_iter_truncate(from, count); - err = file_remove_suid(file); if (err) goto out; @@ -2581,10 +2544,10 @@ ssize_t __generic_file_write_iter(struct kiocb *iocb, struct iov_iter *from) if (err) goto out; - if (io_is_direct(file)) { - loff_t endbyte; + if (iocb->ki_flags & IOCB_DIRECT) { + loff_t pos, endbyte; - written = generic_file_direct_write(iocb, from, pos); + written = generic_file_direct_write(iocb, from, iocb->ki_pos); /* * If the write stopped short of completing, fall back to * buffered writes. Some filesystems do this for writes to @@ -2592,13 +2555,10 @@ ssize_t __generic_file_write_iter(struct kiocb *iocb, struct iov_iter *from) * not succeed (even if it did, DAX does not handle dirty * page-cache pages correctly). */ - if (written < 0 || written == count || IS_DAX(inode)) + if (written < 0 || !iov_iter_count(from) || IS_DAX(inode)) goto out; - pos += written; - count -= written; - - status = generic_perform_write(file, from, pos); + status = generic_perform_write(file, from, pos = iocb->ki_pos); /* * If generic_perform_write() returned a synchronous error * then we want to return the number of bytes which were @@ -2610,15 +2570,15 @@ ssize_t __generic_file_write_iter(struct kiocb *iocb, struct iov_iter *from) err = status; goto out; } - iocb->ki_pos = pos + status; /* * We need to ensure that the page cache pages are written to * disk and invalidated to preserve the expected O_DIRECT * semantics. */ endbyte = pos + status - 1; - err = filemap_write_and_wait_range(file->f_mapping, pos, endbyte); + err = filemap_write_and_wait_range(mapping, pos, endbyte); if (err == 0) { + iocb->ki_pos = endbyte + 1; written += status; invalidate_mapping_pages(mapping, pos >> PAGE_CACHE_SHIFT, @@ -2630,9 +2590,9 @@ ssize_t __generic_file_write_iter(struct kiocb *iocb, struct iov_iter *from) */ } } else { - written = generic_perform_write(file, from, pos); - if (likely(written >= 0)) - iocb->ki_pos = pos + written; + written = generic_perform_write(file, from, iocb->ki_pos); + if (likely(written > 0)) + iocb->ki_pos += written; } out: current->backing_dev_info = NULL; @@ -2656,7 +2616,9 @@ ssize_t generic_file_write_iter(struct kiocb *iocb, struct iov_iter *from) ssize_t ret; mutex_lock(&inode->i_mutex); - ret = __generic_file_write_iter(iocb, from); + ret = generic_write_checks(iocb, from); + if (ret > 0) + ret = __generic_file_write_iter(iocb, from); mutex_unlock(&inode->i_mutex); if (ret > 0) { diff --git a/mm/page_io.c b/mm/page_io.c index a96c856..6424869 100644 --- a/mm/page_io.c +++ b/mm/page_io.c @@ -277,9 +277,7 @@ int __swap_writepage(struct page *page, struct writeback_control *wbc, set_page_writeback(page); unlock_page(page); - ret = mapping->a_ops->direct_IO(ITER_BVEC | WRITE, - &kiocb, &from, - kiocb.ki_pos); + ret = mapping->a_ops->direct_IO(&kiocb, &from, kiocb.ki_pos); if (ret == PAGE_SIZE) { count_vm_event(PSWPOUT); ret = 0; |
