diff options
Diffstat (limited to 'fs/ext4/file.c')
| -rw-r--r-- | fs/ext4/file.c | 1252 |
1 files changed, 806 insertions, 446 deletions
diff --git a/fs/ext4/file.c b/fs/ext4/file.c index 6f4cc567c382..7a8b30932189 100644 --- a/fs/ext4/file.c +++ b/fs/ext4/file.c @@ -1,3 +1,4 @@ +// SPDX-License-Identifier: GPL-2.0 /* * linux/fs/ext4/file.c * @@ -20,16 +21,142 @@ #include <linux/time.h> #include <linux/fs.h> -#include <linux/jbd2.h> +#include <linux/iomap.h> #include <linux/mount.h> #include <linux/path.h> -#include <linux/aio.h> +#include <linux/dax.h> #include <linux/quotaops.h> #include <linux/pagevec.h> +#include <linux/uio.h> +#include <linux/mman.h> +#include <linux/backing-dev.h> #include "ext4.h" #include "ext4_jbd2.h" #include "xattr.h" #include "acl.h" +#include "truncate.h" + +/* + * Returns %true if the given DIO request should be attempted with DIO, or + * %false if it should fall back to buffered I/O. + * + * DIO isn't well specified; when it's unsupported (either due to the request + * being misaligned, or due to the file not supporting DIO at all), filesystems + * either fall back to buffered I/O or return EINVAL. For files that don't use + * any special features like encryption or verity, ext4 has traditionally + * returned EINVAL for misaligned DIO. iomap_dio_rw() uses this convention too. + * In this case, we should attempt the DIO, *not* fall back to buffered I/O. + * + * In contrast, in cases where DIO is unsupported due to ext4 features, ext4 + * traditionally falls back to buffered I/O. + * + * This function implements the traditional ext4 behavior in all these cases. + */ +static bool ext4_should_use_dio(struct kiocb *iocb, struct iov_iter *iter) +{ + struct inode *inode = file_inode(iocb->ki_filp); + u32 dio_align = ext4_dio_alignment(inode); + + if (dio_align == 0) + return false; + + if (dio_align == 1) + return true; + + return IS_ALIGNED(iocb->ki_pos | iov_iter_alignment(iter), dio_align); +} + +static ssize_t ext4_dio_read_iter(struct kiocb *iocb, struct iov_iter *to) +{ + ssize_t ret; + struct inode *inode = file_inode(iocb->ki_filp); + + if (iocb->ki_flags & IOCB_NOWAIT) { + if (!inode_trylock_shared(inode)) + return -EAGAIN; + } else { + inode_lock_shared(inode); + } + + if (!ext4_should_use_dio(iocb, to)) { + inode_unlock_shared(inode); + /* + * Fallback to buffered I/O if the operation being performed on + * the inode is not supported by direct I/O. The IOCB_DIRECT + * flag needs to be cleared here in order to ensure that the + * direct I/O path within generic_file_read_iter() is not + * taken. + */ + iocb->ki_flags &= ~IOCB_DIRECT; + return generic_file_read_iter(iocb, to); + } + + ret = iomap_dio_rw(iocb, to, &ext4_iomap_ops, NULL, 0, NULL, 0); + inode_unlock_shared(inode); + + file_accessed(iocb->ki_filp); + return ret; +} + +#ifdef CONFIG_FS_DAX +static ssize_t ext4_dax_read_iter(struct kiocb *iocb, struct iov_iter *to) +{ + struct inode *inode = file_inode(iocb->ki_filp); + ssize_t ret; + + if (iocb->ki_flags & IOCB_NOWAIT) { + if (!inode_trylock_shared(inode)) + return -EAGAIN; + } else { + inode_lock_shared(inode); + } + /* + * Recheck under inode lock - at this point we are sure it cannot + * change anymore + */ + if (!IS_DAX(inode)) { + inode_unlock_shared(inode); + /* Fallback to buffered IO in case we cannot support DAX */ + return generic_file_read_iter(iocb, to); + } + ret = dax_iomap_rw(iocb, to, &ext4_iomap_ops); + inode_unlock_shared(inode); + + file_accessed(iocb->ki_filp); + return ret; +} +#endif + +static ssize_t ext4_file_read_iter(struct kiocb *iocb, struct iov_iter *to) +{ + struct inode *inode = file_inode(iocb->ki_filp); + + if (unlikely(ext4_forced_shutdown(inode->i_sb))) + return -EIO; + + if (!iov_iter_count(to)) + return 0; /* skip atime */ + +#ifdef CONFIG_FS_DAX + if (IS_DAX(inode)) + return ext4_dax_read_iter(iocb, to); +#endif + if (iocb->ki_flags & IOCB_DIRECT) + return ext4_dio_read_iter(iocb, to); + + return generic_file_read_iter(iocb, to); +} + +static ssize_t ext4_file_splice_read(struct file *in, loff_t *ppos, + struct pipe_inode_info *pipe, + size_t len, unsigned int flags) +{ + struct inode *inode = file_inode(in); + + if (unlikely(ext4_forced_shutdown(inode->i_sb))) + return -EIO; + return filemap_splice_read(in, ppos, pipe, len, flags); +} /* * Called when an inode is released. Note that this is different @@ -45,8 +172,7 @@ static int ext4_release_file(struct inode *inode, struct file *filp) /* if we are the last writer on the inode, drop the block reservation */ if ((filp->f_mode & FMODE_WRITE) && (atomic_read(&inode->i_writecount) == 1) && - !EXT4_I(inode)->i_reserved_data_blocks) - { + !EXT4_I(inode)->i_reserved_data_blocks) { down_write(&EXT4_I(inode)->i_data_sem); ext4_discard_preallocations(inode); up_write(&EXT4_I(inode)->i_data_sem); @@ -57,13 +183,6 @@ static int ext4_release_file(struct inode *inode, struct file *filp) return 0; } -void ext4_unwritten_wait(struct inode *inode) -{ - wait_queue_head_t *wq = ext4_ioend_wq(inode); - - wait_event(*wq, (atomic_read(&EXT4_I(inode)->i_unwritten) == 0)); -} - /* * This tests whether the IO in question is block-aligned or not. * Ext4 utilizes unwritten extents when hole-filling during direct IO, and they @@ -73,504 +192,739 @@ void ext4_unwritten_wait(struct inode *inode) * threads are at work on the same unwritten block, they must be synchronized * or one thread will zero the other's data, causing corruption. */ -static int -ext4_unaligned_aio(struct inode *inode, const struct iovec *iov, - unsigned long nr_segs, loff_t pos) +static bool +ext4_unaligned_io(struct inode *inode, struct iov_iter *from, loff_t pos) { struct super_block *sb = inode->i_sb; - int blockmask = sb->s_blocksize - 1; - size_t count = iov_length(iov, nr_segs); - loff_t final_size = pos + count; + unsigned long blockmask = sb->s_blocksize - 1; - if (pos >= inode->i_size) - return 0; + if ((pos | iov_iter_alignment(from)) & blockmask) + return true; - if ((pos & blockmask) || (final_size & blockmask)) - return 1; + return false; +} - return 0; +static bool +ext4_extending_io(struct inode *inode, loff_t offset, size_t len) +{ + if (offset + len > i_size_read(inode) || + offset + len > EXT4_I(inode)->i_disksize) + return true; + return false; } -static ssize_t -ext4_file_dio_write(struct kiocb *iocb, const struct iovec *iov, - unsigned long nr_segs, loff_t pos) +/* Is IO overwriting allocated or initialized blocks? */ +static bool ext4_overwrite_io(struct inode *inode, + loff_t pos, loff_t len, bool *unwritten) { - struct file *file = iocb->ki_filp; - struct inode *inode = file->f_mapping->host; - struct blk_plug plug; - int unaligned_aio = 0; - ssize_t ret; - int overwrite = 0; - size_t length = iov_length(iov, nr_segs); + struct ext4_map_blocks map; + unsigned int blkbits = inode->i_blkbits; + int err, blklen; - if (ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS) && - !is_sync_kiocb(iocb)) - unaligned_aio = ext4_unaligned_aio(inode, iov, nr_segs, pos); + if (pos + len > i_size_read(inode)) + return false; - /* Unaligned direct AIO must be serialized; see comment above */ - if (unaligned_aio) { - mutex_lock(ext4_aio_mutex(inode)); - ext4_unwritten_wait(inode); - } + map.m_lblk = pos >> blkbits; + map.m_len = EXT4_MAX_BLOCKS(len, pos, blkbits); + blklen = map.m_len; - BUG_ON(iocb->ki_pos != pos); + err = ext4_map_blocks(NULL, inode, &map, 0); + if (err != blklen) + return false; + /* + * 'err==len' means that all of the blocks have been preallocated, + * regardless of whether they have been initialized or not. We need to + * check m_flags to distinguish the unwritten extents. + */ + *unwritten = !(map.m_flags & EXT4_MAP_MAPPED); + return true; +} - mutex_lock(&inode->i_mutex); - blk_start_plug(&plug); +static ssize_t ext4_generic_write_checks(struct kiocb *iocb, + struct iov_iter *from) +{ + struct inode *inode = file_inode(iocb->ki_filp); + ssize_t ret; - iocb->private = &overwrite; + if (unlikely(IS_IMMUTABLE(inode))) + return -EPERM; - /* check whether we do a DIO overwrite or not */ - if (ext4_should_dioread_nolock(inode) && !unaligned_aio && - !file->f_mapping->nrpages && pos + length <= i_size_read(inode)) { - struct ext4_map_blocks map; - unsigned int blkbits = inode->i_blkbits; - int err, len; + ret = generic_write_checks(iocb, from); + if (ret <= 0) + return ret; - map.m_lblk = pos >> blkbits; - map.m_len = (EXT4_BLOCK_ALIGN(pos + length, blkbits) >> blkbits) - - map.m_lblk; - len = map.m_len; + /* + * If we have encountered a bitmap-format file, the size limit + * is smaller than s_maxbytes, which is for extent-mapped files. + */ + if (!(ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))) { + struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); - err = ext4_map_blocks(NULL, inode, &map, 0); - /* - * 'err==len' means that all of blocks has been preallocated no - * matter they are initialized or not. For excluding - * uninitialized extents, we need to check m_flags. There are - * two conditions that indicate for initialized extents. - * 1) If we hit extent cache, EXT4_MAP_MAPPED flag is returned; - * 2) If we do a real lookup, non-flags are returned. - * So we should check these two conditions. - */ - if (err == len && (map.m_flags & EXT4_MAP_MAPPED)) - overwrite = 1; + if (iocb->ki_pos >= sbi->s_bitmap_maxbytes) + return -EFBIG; + iov_iter_truncate(from, sbi->s_bitmap_maxbytes - iocb->ki_pos); } - ret = __generic_file_aio_write(iocb, iov, nr_segs, &iocb->ki_pos); - mutex_unlock(&inode->i_mutex); - - if (ret > 0 || ret == -EIOCBQUEUED) { - ssize_t err; + return iov_iter_count(from); +} - err = generic_write_sync(file, pos, ret); - if (err < 0 && ret > 0) - ret = err; - } - blk_finish_plug(&plug); +static ssize_t ext4_write_checks(struct kiocb *iocb, struct iov_iter *from) +{ + ssize_t ret, count; - if (unaligned_aio) - mutex_unlock(ext4_aio_mutex(inode)); + count = ext4_generic_write_checks(iocb, from); + if (count <= 0) + return count; - return ret; + ret = file_modified(iocb->ki_filp); + if (ret) + return ret; + return count; } -static ssize_t -ext4_file_write(struct kiocb *iocb, const struct iovec *iov, - unsigned long nr_segs, loff_t pos) +static ssize_t ext4_buffered_write_iter(struct kiocb *iocb, + struct iov_iter *from) { - struct inode *inode = file_inode(iocb->ki_filp); ssize_t ret; + struct inode *inode = file_inode(iocb->ki_filp); - /* - * If we have encountered a bitmap-format file, the size limit - * is smaller than s_maxbytes, which is for extent-mapped files. - */ + if (iocb->ki_flags & IOCB_NOWAIT) + return -EOPNOTSUPP; - if (!(ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))) { - struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); - size_t length = iov_length(iov, nr_segs); + inode_lock(inode); + ret = ext4_write_checks(iocb, from); + if (ret <= 0) + goto out; - if ((pos > sbi->s_bitmap_maxbytes || - (pos == sbi->s_bitmap_maxbytes && length > 0))) - return -EFBIG; + ret = generic_perform_write(iocb, from); - if (pos + length > sbi->s_bitmap_maxbytes) { - nr_segs = iov_shorten((struct iovec *)iov, nr_segs, - sbi->s_bitmap_maxbytes - pos); - } - } +out: + inode_unlock(inode); + if (unlikely(ret <= 0)) + return ret; + return generic_write_sync(iocb, ret); +} - if (unlikely(iocb->ki_filp->f_flags & O_DIRECT)) - ret = ext4_file_dio_write(iocb, iov, nr_segs, pos); - else - ret = generic_file_aio_write(iocb, iov, nr_segs, pos); +static ssize_t ext4_handle_inode_extension(struct inode *inode, loff_t offset, + ssize_t written, ssize_t count) +{ + handle_t *handle; - return ret; -} + lockdep_assert_held_write(&inode->i_rwsem); + handle = ext4_journal_start(inode, EXT4_HT_INODE, 2); + if (IS_ERR(handle)) + return PTR_ERR(handle); -static const struct vm_operations_struct ext4_file_vm_ops = { - .fault = filemap_fault, - .page_mkwrite = ext4_page_mkwrite, - .remap_pages = generic_file_remap_pages, -}; + if (ext4_update_inode_size(inode, offset + written)) { + int ret = ext4_mark_inode_dirty(handle, inode); + if (unlikely(ret)) { + ext4_journal_stop(handle); + return ret; + } + } -static int ext4_file_mmap(struct file *file, struct vm_area_struct *vma) -{ - struct address_space *mapping = file->f_mapping; + if ((written == count) && inode->i_nlink) + ext4_orphan_del(handle, inode); + ext4_journal_stop(handle); - if (!mapping->a_ops->readpage) - return -ENOEXEC; - file_accessed(file); - vma->vm_ops = &ext4_file_vm_ops; - return 0; + return written; } -static int ext4_file_open(struct inode * inode, struct file * filp) +/* + * Clean up the inode after DIO or DAX extending write has completed and the + * inode size has been updated using ext4_handle_inode_extension(). + */ +static void ext4_inode_extension_cleanup(struct inode *inode, bool need_trunc) { - struct super_block *sb = inode->i_sb; - struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); - struct ext4_inode_info *ei = EXT4_I(inode); - struct vfsmount *mnt = filp->f_path.mnt; - struct path path; - char buf[64], *cp; - - if (unlikely(!(sbi->s_mount_flags & EXT4_MF_MNTDIR_SAMPLED) && - !(sb->s_flags & MS_RDONLY))) { - sbi->s_mount_flags |= EXT4_MF_MNTDIR_SAMPLED; + lockdep_assert_held_write(&inode->i_rwsem); + if (need_trunc) { + ext4_truncate_failed_write(inode); /* - * Sample where the filesystem has been mounted and - * store it in the superblock for sysadmin convenience - * when trying to sort through large numbers of block - * devices or filesystem images. + * If the truncate operation failed early, then the inode may + * still be on the orphan list. In that case, we need to try + * remove the inode from the in-memory linked list. */ - memset(buf, 0, sizeof(buf)); - path.mnt = mnt; - path.dentry = mnt->mnt_root; - cp = d_path(&path, buf, sizeof(buf)); - if (!IS_ERR(cp)) { - handle_t *handle; - int err; - - handle = ext4_journal_start_sb(sb, EXT4_HT_MISC, 1); - if (IS_ERR(handle)) - return PTR_ERR(handle); - err = ext4_journal_get_write_access(handle, sbi->s_sbh); - if (err) { - ext4_journal_stop(handle); - return err; - } - strlcpy(sbi->s_es->s_last_mounted, cp, - sizeof(sbi->s_es->s_last_mounted)); - ext4_handle_dirty_super(handle, sb); - ext4_journal_stop(handle); - } + if (inode->i_nlink) + ext4_orphan_del(NULL, inode); + return; } /* - * Set up the jbd2_inode if we are opening the inode for - * writing and the journal is present + * If i_disksize got extended either due to writeback of delalloc + * blocks or extending truncate while the DIO was running we could fail + * to cleanup the orphan list in ext4_handle_inode_extension(). Do it + * now. */ - if (sbi->s_journal && !ei->jinode && (filp->f_mode & FMODE_WRITE)) { - struct jbd2_inode *jinode = jbd2_alloc_inode(GFP_KERNEL); - - spin_lock(&inode->i_lock); - if (!ei->jinode) { - if (!jinode) { - spin_unlock(&inode->i_lock); - return -ENOMEM; - } - ei->jinode = jinode; - jbd2_journal_init_jbd_inode(ei->jinode, inode); - jinode = NULL; + if (ext4_inode_orphan_tracked(inode) && inode->i_nlink) { + handle_t *handle = ext4_journal_start(inode, EXT4_HT_INODE, 2); + + if (IS_ERR(handle)) { + /* + * The write has successfully completed. Not much to + * do with the error here so just cleanup the orphan + * list and hope for the best. + */ + ext4_orphan_del(NULL, inode); + return; } - spin_unlock(&inode->i_lock); - if (unlikely(jinode != NULL)) - jbd2_free_inode(jinode); + ext4_orphan_del(handle, inode); + ext4_journal_stop(handle); } - return dquot_file_open(inode, filp); } -/* - * Here we use ext4_map_blocks() to get a block mapping for a extent-based - * file rather than ext4_ext_walk_space() because we can introduce - * SEEK_DATA/SEEK_HOLE for block-mapped and extent-mapped file at the same - * function. When extent status tree has been fully implemented, it will - * track all extent status for a file and we can directly use it to - * retrieve the offset for SEEK_DATA/SEEK_HOLE. - */ +static int ext4_dio_write_end_io(struct kiocb *iocb, ssize_t size, + int error, unsigned int flags) +{ + loff_t pos = iocb->ki_pos; + struct inode *inode = file_inode(iocb->ki_filp); + + + if (!error && size && (flags & IOMAP_DIO_UNWRITTEN) && + (iocb->ki_flags & IOCB_ATOMIC)) + error = ext4_convert_unwritten_extents_atomic(NULL, inode, pos, + size); + else if (!error && size && flags & IOMAP_DIO_UNWRITTEN) + error = ext4_convert_unwritten_extents(NULL, inode, pos, size); + if (error) + return error; + /* + * Note that EXT4_I(inode)->i_disksize can get extended up to + * inode->i_size while the I/O was running due to writeback of delalloc + * blocks. But the code in ext4_iomap_alloc() is careful to use + * zeroed/unwritten extents if this is possible; thus we won't leave + * uninitialized blocks in a file even if we didn't succeed in writing + * as much as we intended. Also we can race with truncate or write + * expanding the file so we have to be a bit careful here. + */ + if (pos + size <= READ_ONCE(EXT4_I(inode)->i_disksize) && + pos + size <= i_size_read(inode)) + return 0; + error = ext4_handle_inode_extension(inode, pos, size, size); + return error < 0 ? error : 0; +} + +static const struct iomap_dio_ops ext4_dio_write_ops = { + .end_io = ext4_dio_write_end_io, +}; /* - * When we retrieve the offset for SEEK_DATA/SEEK_HOLE, we would need to - * lookup page cache to check whether or not there has some data between - * [startoff, endoff] because, if this range contains an unwritten extent, - * we determine this extent as a data or a hole according to whether the - * page cache has data or not. + * The intention here is to start with shared lock acquired then see if any + * condition requires an exclusive inode lock. If yes, then we restart the + * whole operation by releasing the shared lock and acquiring exclusive lock. + * + * - For unaligned_io we never take shared lock as it may cause data corruption + * when two unaligned IO tries to modify the same block e.g. while zeroing. + * + * - For extending writes case we don't take the shared lock, since it requires + * updating inode i_disksize and/or orphan handling with exclusive lock. + * + * - shared locking will only be true mostly with overwrites, including + * initialized blocks and unwritten blocks. For overwrite unwritten blocks + * we protect splitting extents by i_data_sem in ext4_inode_info, so we can + * also release exclusive i_rwsem lock. + * + * - Otherwise we will switch to exclusive i_rwsem lock. */ -static int ext4_find_unwritten_pgoff(struct inode *inode, - int whence, - struct ext4_map_blocks *map, - loff_t *offset) +static ssize_t ext4_dio_write_checks(struct kiocb *iocb, struct iov_iter *from, + bool *ilock_shared, bool *extend, + bool *unwritten, int *dio_flags) { - struct pagevec pvec; - unsigned int blkbits; - pgoff_t index; - pgoff_t end; - loff_t endoff; - loff_t startoff; - loff_t lastoff; - int found = 0; - - blkbits = inode->i_sb->s_blocksize_bits; - startoff = *offset; - lastoff = startoff; - endoff = (loff_t)(map->m_lblk + map->m_len) << blkbits; - - index = startoff >> PAGE_CACHE_SHIFT; - end = endoff >> PAGE_CACHE_SHIFT; - - pagevec_init(&pvec, 0); - do { - int i, num; - unsigned long nr_pages; - - num = min_t(pgoff_t, end - index, PAGEVEC_SIZE); - nr_pages = pagevec_lookup(&pvec, inode->i_mapping, index, - (pgoff_t)num); - if (nr_pages == 0) { - if (whence == SEEK_DATA) - break; - - BUG_ON(whence != SEEK_HOLE); - /* - * If this is the first time to go into the loop and - * offset is not beyond the end offset, it will be a - * hole at this offset - */ - if (lastoff == startoff || lastoff < endoff) - found = 1; - break; - } + struct file *file = iocb->ki_filp; + struct inode *inode = file_inode(file); + loff_t offset; + size_t count; + ssize_t ret; + bool overwrite, unaligned_io; - /* - * If this is the first time to go into the loop and - * offset is smaller than the first page offset, it will be a - * hole at this offset. - */ - if (lastoff == startoff && whence == SEEK_HOLE && - lastoff < page_offset(pvec.pages[0])) { - found = 1; - break; - } +restart: + ret = ext4_generic_write_checks(iocb, from); + if (ret <= 0) + goto out; - for (i = 0; i < nr_pages; i++) { - struct page *page = pvec.pages[i]; - struct buffer_head *bh, *head; + offset = iocb->ki_pos; + count = ret; - /* - * If the current offset is not beyond the end of given - * range, it will be a hole. - */ - if (lastoff < endoff && whence == SEEK_HOLE && - page->index > end) { - found = 1; - *offset = lastoff; - goto out; - } - - lock_page(page); - - if (unlikely(page->mapping != inode->i_mapping)) { - unlock_page(page); - continue; - } - - if (!page_has_buffers(page)) { - unlock_page(page); - continue; - } - - if (page_has_buffers(page)) { - lastoff = page_offset(page); - bh = head = page_buffers(page); - do { - if (buffer_uptodate(bh) || - buffer_unwritten(bh)) { - if (whence == SEEK_DATA) - found = 1; - } else { - if (whence == SEEK_HOLE) - found = 1; - } - if (found) { - *offset = max_t(loff_t, - startoff, lastoff); - unlock_page(page); - goto out; - } - lastoff += bh->b_size; - bh = bh->b_this_page; - } while (bh != head); - } - - lastoff = page_offset(page) + PAGE_SIZE; - unlock_page(page); + unaligned_io = ext4_unaligned_io(inode, from, offset); + *extend = ext4_extending_io(inode, offset, count); + overwrite = ext4_overwrite_io(inode, offset, count, unwritten); + + /* + * Determine whether we need to upgrade to an exclusive lock. This is + * required to change security info in file_modified(), for extending + * I/O, any form of non-overwrite I/O, and unaligned I/O to unwritten + * extents (as partial block zeroing may be required). + * + * Note that unaligned writes are allowed under shared lock so long as + * they are pure overwrites. Otherwise, concurrent unaligned writes risk + * data corruption due to partial block zeroing in the dio layer, and so + * the I/O must occur exclusively. + */ + if (*ilock_shared && + ((!IS_NOSEC(inode) || *extend || !overwrite || + (unaligned_io && *unwritten)))) { + if (iocb->ki_flags & IOCB_NOWAIT) { + ret = -EAGAIN; + goto out; } + inode_unlock_shared(inode); + *ilock_shared = false; + inode_lock(inode); + goto restart; + } - /* - * The no. of pages is less than our desired, that would be a - * hole in there. - */ - if (nr_pages < num && whence == SEEK_HOLE) { - found = 1; - *offset = lastoff; - break; + /* + * Now that locking is settled, determine dio flags and exclusivity + * requirements. We don't use DIO_OVERWRITE_ONLY because we enforce + * behavior already. The inode lock is already held exclusive if the + * write is non-overwrite or extending, so drain all outstanding dio and + * set the force wait dio flag. + */ + if (!*ilock_shared && (unaligned_io || *extend)) { + if (iocb->ki_flags & IOCB_NOWAIT) { + ret = -EAGAIN; + goto out; } + if (unaligned_io && (!overwrite || *unwritten)) + inode_dio_wait(inode); + *dio_flags = IOMAP_DIO_FORCE_WAIT; + } - index = pvec.pages[i - 1]->index + 1; - pagevec_release(&pvec); - } while (index <= end); + ret = file_modified(file); + if (ret < 0) + goto out; + return count; out: - pagevec_release(&pvec); - return found; + if (*ilock_shared) + inode_unlock_shared(inode); + else + inode_unlock(inode); + return ret; } -/* - * ext4_seek_data() retrieves the offset for SEEK_DATA. - */ -static loff_t ext4_seek_data(struct file *file, loff_t offset, loff_t maxsize) +static ssize_t ext4_dio_write_iter(struct kiocb *iocb, struct iov_iter *from) { - struct inode *inode = file->f_mapping->host; - struct ext4_map_blocks map; - struct extent_status es; - ext4_lblk_t start, last, end; - loff_t dataoff, isize; - int blkbits; - int ret = 0; - - mutex_lock(&inode->i_mutex); - - isize = i_size_read(inode); - if (offset >= isize) { - mutex_unlock(&inode->i_mutex); - return -ENXIO; + ssize_t ret; + handle_t *handle; + struct inode *inode = file_inode(iocb->ki_filp); + loff_t offset = iocb->ki_pos; + size_t count = iov_iter_count(from); + const struct iomap_ops *iomap_ops = &ext4_iomap_ops; + bool extend = false, unwritten = false; + bool ilock_shared = true; + int dio_flags = 0; + + /* + * Quick check here without any i_rwsem lock to see if it is extending + * IO. A more reliable check is done in ext4_dio_write_checks() with + * proper locking in place. + */ + if (offset + count > i_size_read(inode)) + ilock_shared = false; + + if (iocb->ki_flags & IOCB_NOWAIT) { + if (ilock_shared) { + if (!inode_trylock_shared(inode)) + return -EAGAIN; + } else { + if (!inode_trylock(inode)) + return -EAGAIN; + } + } else { + if (ilock_shared) + inode_lock_shared(inode); + else + inode_lock(inode); + } + + /* Fallback to buffered I/O if the inode does not support direct I/O. */ + if (!ext4_should_use_dio(iocb, from)) { + if (ilock_shared) + inode_unlock_shared(inode); + else + inode_unlock(inode); + return ext4_buffered_write_iter(iocb, from); } - blkbits = inode->i_sb->s_blocksize_bits; - start = offset >> blkbits; - last = start; - end = isize >> blkbits; - dataoff = offset; - - do { - map.m_lblk = last; - map.m_len = end - last + 1; - ret = ext4_map_blocks(NULL, inode, &map, 0); - if (ret > 0 && !(map.m_flags & EXT4_MAP_UNWRITTEN)) { - if (last != start) - dataoff = (loff_t)last << blkbits; - break; + /* + * Prevent inline data from being created since we are going to allocate + * blocks for DIO. We know the inode does not currently have inline data + * because ext4_should_use_dio() checked for it, but we have to clear + * the state flag before the write checks because a lock cycle could + * introduce races with other writers. + */ + ext4_clear_inode_state(inode, EXT4_STATE_MAY_INLINE_DATA); + + ret = ext4_dio_write_checks(iocb, from, &ilock_shared, &extend, + &unwritten, &dio_flags); + if (ret <= 0) + return ret; + + offset = iocb->ki_pos; + count = ret; + + if (extend) { + handle = ext4_journal_start(inode, EXT4_HT_INODE, 2); + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + goto out; } + ret = ext4_orphan_add(handle, inode); + ext4_journal_stop(handle); + if (ret) + goto out; + } + + if (ilock_shared && !unwritten) + iomap_ops = &ext4_iomap_overwrite_ops; + ret = iomap_dio_rw(iocb, from, iomap_ops, &ext4_dio_write_ops, + dio_flags, NULL, 0); + if (ret == -ENOTBLK) + ret = 0; + if (extend) { /* - * If there is a delay extent at this offset, - * it will be as a data. + * We always perform extending DIO write synchronously so by + * now the IO is completed and ext4_handle_inode_extension() + * was called. Cleanup the inode in case of error or race with + * writeback of delalloc blocks. */ - ext4_es_find_delayed_extent_range(inode, last, last, &es); - if (es.es_len != 0 && in_range(last, es.es_lblk, es.es_len)) { - if (last != start) - dataoff = (loff_t)last << blkbits; - break; - } + WARN_ON_ONCE(ret == -EIOCBQUEUED); + ext4_inode_extension_cleanup(inode, ret < 0); + } + +out: + if (ilock_shared) + inode_unlock_shared(inode); + else + inode_unlock(inode); + + if (ret >= 0 && iov_iter_count(from)) { + ssize_t err; + loff_t endbyte; /* - * If there is a unwritten extent at this offset, - * it will be as a data or a hole according to page - * cache that has data or not. + * There is no support for atomic writes on buffered-io yet, + * we should never fallback to buffered-io for DIO atomic + * writes. */ - if (map.m_flags & EXT4_MAP_UNWRITTEN) { - int unwritten; - unwritten = ext4_find_unwritten_pgoff(inode, SEEK_DATA, - &map, &dataoff); - if (unwritten) - break; - } - - last++; - dataoff = (loff_t)last << blkbits; - } while (last <= end); + WARN_ON_ONCE(iocb->ki_flags & IOCB_ATOMIC); - mutex_unlock(&inode->i_mutex); + offset = iocb->ki_pos; + err = ext4_buffered_write_iter(iocb, from); + if (err < 0) + return err; - if (dataoff > isize) - return -ENXIO; + /* + * We need to ensure that the pages within the page cache for + * the range covered by this I/O are written to disk and + * invalidated. This is in attempt to preserve the expected + * direct I/O semantics in the case we fallback to buffered I/O + * to complete off the I/O request. + */ + ret += err; + endbyte = offset + err - 1; + err = filemap_write_and_wait_range(iocb->ki_filp->f_mapping, + offset, endbyte); + if (!err) + invalidate_mapping_pages(iocb->ki_filp->f_mapping, + offset >> PAGE_SHIFT, + endbyte >> PAGE_SHIFT); + } - return vfs_setpos(file, dataoff, maxsize); + return ret; } -/* - * ext4_seek_hole() retrieves the offset for SEEK_HOLE. - */ -static loff_t ext4_seek_hole(struct file *file, loff_t offset, loff_t maxsize) +#ifdef CONFIG_FS_DAX +static ssize_t +ext4_dax_write_iter(struct kiocb *iocb, struct iov_iter *from) { - struct inode *inode = file->f_mapping->host; - struct ext4_map_blocks map; - struct extent_status es; - ext4_lblk_t start, last, end; - loff_t holeoff, isize; - int blkbits; - int ret = 0; - - mutex_lock(&inode->i_mutex); - - isize = i_size_read(inode); - if (offset >= isize) { - mutex_unlock(&inode->i_mutex); - return -ENXIO; + ssize_t ret; + size_t count; + loff_t offset; + handle_t *handle; + bool extend = false; + struct inode *inode = file_inode(iocb->ki_filp); + + if (iocb->ki_flags & IOCB_NOWAIT) { + if (!inode_trylock(inode)) + return -EAGAIN; + } else { + inode_lock(inode); } - blkbits = inode->i_sb->s_blocksize_bits; - start = offset >> blkbits; - last = start; - end = isize >> blkbits; - holeoff = offset; - - do { - map.m_lblk = last; - map.m_len = end - last + 1; - ret = ext4_map_blocks(NULL, inode, &map, 0); - if (ret > 0 && !(map.m_flags & EXT4_MAP_UNWRITTEN)) { - last += ret; - holeoff = (loff_t)last << blkbits; - continue; + ret = ext4_write_checks(iocb, from); + if (ret <= 0) + goto out; + + offset = iocb->ki_pos; + count = iov_iter_count(from); + + if (offset + count > EXT4_I(inode)->i_disksize) { + handle = ext4_journal_start(inode, EXT4_HT_INODE, 2); + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + goto out; } - /* - * If there is a delay extent at this offset, - * we will skip this extent. - */ - ext4_es_find_delayed_extent_range(inode, last, last, &es); - if (es.es_len != 0 && in_range(last, es.es_lblk, es.es_len)) { - last = es.es_lblk + es.es_len; - holeoff = (loff_t)last << blkbits; - continue; + ret = ext4_orphan_add(handle, inode); + if (ret) { + ext4_journal_stop(handle); + goto out; } - /* - * If there is a unwritten extent at this offset, - * it will be as a data or a hole according to page - * cache that has data or not. - */ - if (map.m_flags & EXT4_MAP_UNWRITTEN) { - int unwritten; - unwritten = ext4_find_unwritten_pgoff(inode, SEEK_HOLE, - &map, &holeoff); - if (!unwritten) { - last += ret; - holeoff = (loff_t)last << blkbits; - continue; - } + extend = true; + ext4_journal_stop(handle); + } + + ret = dax_iomap_rw(iocb, from, &ext4_iomap_ops); + + if (extend) { + ret = ext4_handle_inode_extension(inode, offset, ret, count); + ext4_inode_extension_cleanup(inode, ret < (ssize_t)count); + } +out: + inode_unlock(inode); + if (ret > 0) + ret = generic_write_sync(iocb, ret); + return ret; +} +#endif + +static ssize_t +ext4_file_write_iter(struct kiocb *iocb, struct iov_iter *from) +{ + int ret; + struct inode *inode = file_inode(iocb->ki_filp); + + ret = ext4_emergency_state(inode->i_sb); + if (unlikely(ret)) + return ret; + +#ifdef CONFIG_FS_DAX + if (IS_DAX(inode)) + return ext4_dax_write_iter(iocb, from); +#endif + + if (iocb->ki_flags & IOCB_ATOMIC) { + size_t len = iov_iter_count(from); + + if (len < EXT4_SB(inode->i_sb)->s_awu_min || + len > EXT4_SB(inode->i_sb)->s_awu_max) + return -EINVAL; + + ret = generic_atomic_write_valid(iocb, from); + if (ret) + return ret; + } + + if (iocb->ki_flags & IOCB_DIRECT) + return ext4_dio_write_iter(iocb, from); + else + return ext4_buffered_write_iter(iocb, from); +} + +#ifdef CONFIG_FS_DAX +static vm_fault_t ext4_dax_huge_fault(struct vm_fault *vmf, unsigned int order) +{ + int error = 0; + vm_fault_t result; + int retries = 0; + handle_t *handle = NULL; + struct inode *inode = file_inode(vmf->vma->vm_file); + struct super_block *sb = inode->i_sb; + + /* + * We have to distinguish real writes from writes which will result in a + * COW page; COW writes should *not* poke the journal (the file will not + * be changed). Doing so would cause unintended failures when mounted + * read-only. + * + * We check for VM_SHARED rather than vmf->cow_page since the latter is + * unset for order != 0 (i.e. only in do_cow_fault); for + * other sizes, dax_iomap_fault will handle splitting / fallback so that + * we eventually come back with a COW page. + */ + bool write = (vmf->flags & FAULT_FLAG_WRITE) && + (vmf->vma->vm_flags & VM_SHARED); + struct address_space *mapping = vmf->vma->vm_file->f_mapping; + unsigned long pfn; + + if (write) { + sb_start_pagefault(sb); + file_update_time(vmf->vma->vm_file); + filemap_invalidate_lock_shared(mapping); +retry: + handle = ext4_journal_start_sb(sb, EXT4_HT_WRITE_PAGE, + EXT4_DATA_TRANS_BLOCKS(sb)); + if (IS_ERR(handle)) { + filemap_invalidate_unlock_shared(mapping); + sb_end_pagefault(sb); + return VM_FAULT_SIGBUS; } + } else { + filemap_invalidate_lock_shared(mapping); + } + result = dax_iomap_fault(vmf, order, &pfn, &error, &ext4_iomap_ops); + if (write) { + ext4_journal_stop(handle); + + if ((result & VM_FAULT_ERROR) && error == -ENOSPC && + ext4_should_retry_alloc(sb, &retries)) + goto retry; + /* Handling synchronous page fault? */ + if (result & VM_FAULT_NEEDDSYNC) + result = dax_finish_sync_fault(vmf, order, pfn); + filemap_invalidate_unlock_shared(mapping); + sb_end_pagefault(sb); + } else { + filemap_invalidate_unlock_shared(mapping); + } - /* find a hole */ - break; - } while (last <= end); + return result; +} - mutex_unlock(&inode->i_mutex); +static vm_fault_t ext4_dax_fault(struct vm_fault *vmf) +{ + return ext4_dax_huge_fault(vmf, 0); +} + +static const struct vm_operations_struct ext4_dax_vm_ops = { + .fault = ext4_dax_fault, + .huge_fault = ext4_dax_huge_fault, + .page_mkwrite = ext4_dax_fault, + .pfn_mkwrite = ext4_dax_fault, +}; +#else +#define ext4_dax_vm_ops ext4_file_vm_ops +#endif - if (holeoff > isize) - holeoff = isize; +static const struct vm_operations_struct ext4_file_vm_ops = { + .fault = filemap_fault, + .map_pages = filemap_map_pages, + .page_mkwrite = ext4_page_mkwrite, +}; - return vfs_setpos(file, holeoff, maxsize); +static int ext4_file_mmap_prepare(struct vm_area_desc *desc) +{ + int ret; + struct file *file = desc->file; + struct inode *inode = file->f_mapping->host; + struct dax_device *dax_dev = EXT4_SB(inode->i_sb)->s_daxdev; + + if (file->f_mode & FMODE_WRITE) + ret = ext4_emergency_state(inode->i_sb); + else + ret = ext4_forced_shutdown(inode->i_sb) ? -EIO : 0; + if (unlikely(ret)) + return ret; + + /* + * We don't support synchronous mappings for non-DAX files and + * for DAX files if underneath dax_device is not synchronous. + */ + if (!daxdev_mapping_supported(desc->vm_flags, file_inode(file), dax_dev)) + return -EOPNOTSUPP; + + file_accessed(file); + if (IS_DAX(file_inode(file))) { + desc->vm_ops = &ext4_dax_vm_ops; + desc->vm_flags |= VM_HUGEPAGE; + } else { + desc->vm_ops = &ext4_file_vm_ops; + } + return 0; +} + +static int ext4_sample_last_mounted(struct super_block *sb, + struct vfsmount *mnt) +{ + struct ext4_sb_info *sbi = EXT4_SB(sb); + struct path path; + char buf[64], *cp; + handle_t *handle; + int err; + + if (likely(ext4_test_mount_flag(sb, EXT4_MF_MNTDIR_SAMPLED))) + return 0; + + if (ext4_emergency_state(sb) || sb_rdonly(sb) || + !sb_start_intwrite_trylock(sb)) + return 0; + + ext4_set_mount_flag(sb, EXT4_MF_MNTDIR_SAMPLED); + /* + * Sample where the filesystem has been mounted and + * store it in the superblock for sysadmin convenience + * when trying to sort through large numbers of block + * devices or filesystem images. + */ + memset(buf, 0, sizeof(buf)); + path.mnt = mnt; + path.dentry = mnt->mnt_root; + cp = d_path(&path, buf, sizeof(buf)); + err = 0; + if (IS_ERR(cp)) + goto out; + + handle = ext4_journal_start_sb(sb, EXT4_HT_MISC, 1); + err = PTR_ERR(handle); + if (IS_ERR(handle)) + goto out; + BUFFER_TRACE(sbi->s_sbh, "get_write_access"); + err = ext4_journal_get_write_access(handle, sb, sbi->s_sbh, + EXT4_JTR_NONE); + if (err) + goto out_journal; + lock_buffer(sbi->s_sbh); + strtomem_pad(sbi->s_es->s_last_mounted, cp, 0); + ext4_superblock_csum_set(sb); + unlock_buffer(sbi->s_sbh); + ext4_handle_dirty_metadata(handle, NULL, sbi->s_sbh); +out_journal: + ext4_journal_stop(handle); +out: + sb_end_intwrite(sb); + return err; +} + +static int ext4_file_open(struct inode *inode, struct file *filp) +{ + int ret; + + if (filp->f_mode & FMODE_WRITE) + ret = ext4_emergency_state(inode->i_sb); + else + ret = ext4_forced_shutdown(inode->i_sb) ? -EIO : 0; + if (unlikely(ret)) + return ret; + + ret = ext4_sample_last_mounted(inode->i_sb, filp->f_path.mnt); + if (ret) + return ret; + + ret = fscrypt_file_open(inode, filp); + if (ret) + return ret; + + ret = fsverity_file_open(inode, filp); + if (ret) + return ret; + + /* + * Set up the jbd2_inode if we are opening the inode for + * writing and the journal is present + */ + if (filp->f_mode & FMODE_WRITE) { + ret = ext4_inode_attach_jinode(inode); + if (ret < 0) + return ret; + } + + if (ext4_inode_can_atomic_write(inode)) + filp->f_mode |= FMODE_CAN_ATOMIC_WRITE; + + filp->f_mode |= FMODE_NOWAIT | FMODE_CAN_ODIRECT; + return dquot_file_open(inode, filp); } /* @@ -581,55 +935,61 @@ static loff_t ext4_seek_hole(struct file *file, loff_t offset, loff_t maxsize) loff_t ext4_llseek(struct file *file, loff_t offset, int whence) { struct inode *inode = file->f_mapping->host; - loff_t maxbytes; - - if (!(ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS))) - maxbytes = EXT4_SB(inode->i_sb)->s_bitmap_maxbytes; - else - maxbytes = inode->i_sb->s_maxbytes; + loff_t maxbytes = ext4_get_maxbytes(inode); switch (whence) { - case SEEK_SET: - case SEEK_CUR: - case SEEK_END: + default: return generic_file_llseek_size(file, offset, whence, maxbytes, i_size_read(inode)); - case SEEK_DATA: - return ext4_seek_data(file, offset, maxbytes); case SEEK_HOLE: - return ext4_seek_hole(file, offset, maxbytes); + inode_lock_shared(inode); + offset = iomap_seek_hole(inode, offset, + &ext4_iomap_report_ops); + inode_unlock_shared(inode); + break; + case SEEK_DATA: + inode_lock_shared(inode); + offset = iomap_seek_data(inode, offset, + &ext4_iomap_report_ops); + inode_unlock_shared(inode); + break; } - return -EINVAL; + if (offset < 0) + return offset; + return vfs_setpos(file, offset, maxbytes); } const struct file_operations ext4_file_operations = { .llseek = ext4_llseek, - .read = do_sync_read, - .write = do_sync_write, - .aio_read = generic_file_aio_read, - .aio_write = ext4_file_write, + .read_iter = ext4_file_read_iter, + .write_iter = ext4_file_write_iter, + .iopoll = iocb_bio_iopoll, .unlocked_ioctl = ext4_ioctl, #ifdef CONFIG_COMPAT .compat_ioctl = ext4_compat_ioctl, #endif - .mmap = ext4_file_mmap, + .mmap_prepare = ext4_file_mmap_prepare, .open = ext4_file_open, .release = ext4_release_file, .fsync = ext4_sync_file, - .splice_read = generic_file_splice_read, - .splice_write = generic_file_splice_write, + .get_unmapped_area = thp_get_unmapped_area, + .splice_read = ext4_file_splice_read, + .splice_write = iter_file_splice_write, .fallocate = ext4_fallocate, + .fop_flags = FOP_MMAP_SYNC | FOP_BUFFER_RASYNC | + FOP_DIO_PARALLEL_WRITE | + FOP_DONTCACHE, }; const struct inode_operations ext4_file_inode_operations = { .setattr = ext4_setattr, - .getattr = ext4_getattr, - .setxattr = generic_setxattr, - .getxattr = generic_getxattr, + .getattr = ext4_file_getattr, .listxattr = ext4_listxattr, - .removexattr = generic_removexattr, - .get_acl = ext4_get_acl, + .get_inode_acl = ext4_get_acl, + .set_acl = ext4_set_acl, .fiemap = ext4_fiemap, + .fileattr_get = ext4_fileattr_get, + .fileattr_set = ext4_fileattr_set, }; |