--- zzzz-none-000/linux-3.10.107/fs/ocfs2/file.c 2017-06-27 09:49:32.000000000 +0000 +++ scorpion-7490-727/linux-3.10.107/fs/ocfs2/file.c 2021-02-04 17:41:59.000000000 +0000 @@ -37,6 +37,7 @@ #include #include #include +#include #include @@ -104,8 +105,11 @@ file->f_path.dentry->d_name.len, file->f_path.dentry->d_name.name, mode); - if (file->f_mode & FMODE_WRITE) - dquot_initialize(inode); + if (file->f_mode & FMODE_WRITE) { + status = dquot_initialize(inode); + if (status) + goto leave; + } spin_lock(&oi->ip_lock); @@ -175,9 +179,13 @@ int datasync) { int err = 0; - journal_t *journal; struct inode *inode = file->f_mapping->host; struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); + struct ocfs2_inode_info *oi = OCFS2_I(inode); + journal_t *journal = osb->journal->j_journal; + int ret; + tid_t commit_tid; + bool needs_barrier = false; trace_ocfs2_sync_file(inode, file, file->f_path.dentry, OCFS2_I(inode)->ip_blkno, @@ -185,33 +193,26 @@ file->f_path.dentry->d_name.name, (unsigned long long)datasync); + if (ocfs2_is_hard_readonly(osb) || ocfs2_is_soft_readonly(osb)) + return -EROFS; + err = filemap_write_and_wait_range(inode->i_mapping, start, end); if (err) return err; - /* - * Probably don't need the i_mutex at all in here, just putting it here - * to be consistent with how fsync used to be called, someone more - * familiar with the fs could possibly remove it. - */ - mutex_lock(&inode->i_mutex); - if (datasync && !(inode->i_state & I_DIRTY_DATASYNC)) { - /* - * We still have to flush drive's caches to get data to the - * platter - */ - if (osb->s_mount_opt & OCFS2_MOUNT_BARRIER) - blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, NULL); - goto bail; + commit_tid = datasync ? oi->i_datasync_tid : oi->i_sync_tid; + if (journal->j_flags & JBD2_BARRIER && + !jbd2_trans_will_send_data_barrier(journal, commit_tid)) + needs_barrier = true; + err = jbd2_complete_transaction(journal, commit_tid); + if (needs_barrier) { + ret = blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL, NULL); + if (!err) + err = ret; } - journal = osb->journal->j_journal; - err = jbd2_journal_force_commit(journal); - -bail: if (err) mlog_errno(err); - mutex_unlock(&inode->i_mutex); return (err < 0) ? -EIO : 0; } @@ -289,6 +290,7 @@ inode->i_atime = CURRENT_TIME; di->i_atime = cpu_to_le64(inode->i_atime.tv_sec); di->i_atime_nsec = cpu_to_le32(inode->i_atime.tv_nsec); + ocfs2_update_inode_fsync_trans(handle, inode, 0); ocfs2_journal_dirty(handle, bh); out_commit: @@ -297,7 +299,7 @@ return ret; } -static int ocfs2_set_inode_size(handle_t *handle, +int ocfs2_set_inode_size(handle_t *handle, struct inode *inode, struct buffer_head *fe_bh, u64 new_i_size) @@ -338,6 +340,7 @@ if (ret < 0) mlog_errno(ret); + ocfs2_update_inode_fsync_trans(handle, inode, 0); ocfs2_commit_trans(osb, handle); out: return ret; @@ -370,7 +373,7 @@ if (!(ext_flags & OCFS2_EXT_REFCOUNTED)) goto out; - return ocfs2_refcount_cow(inode, NULL, fe_bh, cpos, 1, cpos+1); + return ocfs2_refcount_cow(inode, fe_bh, cpos, 1, cpos+1); out: return status; @@ -432,6 +435,7 @@ di->i_size = cpu_to_le64(new_i_size); di->i_ctime = di->i_mtime = cpu_to_le64(inode->i_ctime.tv_sec); di->i_ctime_nsec = di->i_mtime_nsec = cpu_to_le32(inode->i_ctime.tv_nsec); + ocfs2_update_inode_fsync_trans(handle, inode, 0); ocfs2_journal_dirty(handle, fe_bh); @@ -441,7 +445,7 @@ return status; } -static int ocfs2_truncate_file(struct inode *inode, +int ocfs2_truncate_file(struct inode *inode, struct buffer_head *di_bh, u64 new_i_size) { @@ -474,11 +478,6 @@ goto bail; } - /* lets handle the simple truncate cases before doing any more - * cluster locking. */ - if (new_i_size == le64_to_cpu(fe->i_size)) - goto bail; - down_write(&OCFS2_I(inode)->ip_alloc_sem); ocfs2_resv_discard(&osb->osb_la_resmap, @@ -574,13 +573,13 @@ handle_t *handle = NULL; struct ocfs2_alloc_context *data_ac = NULL; struct ocfs2_alloc_context *meta_ac = NULL; - enum ocfs2_alloc_restarted why; + enum ocfs2_alloc_restarted why = RESTART_NONE; struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); struct ocfs2_extent_tree et; int did_quota = 0; /* - * This function only exists for file systems which don't + * Unwritten extent only exists for file systems which * support holes. */ BUG_ON(mark_unwritten && !ocfs2_sparse_alloc(osb)); @@ -603,8 +602,7 @@ goto leave; } - credits = ocfs2_calc_extend_credits(osb->sb, &fe->id2.i_list, - clusters_to_add); + credits = ocfs2_calc_extend_credits(osb->sb, &fe->id2.i_list); handle = ocfs2_start_trans(osb, credits); if (IS_ERR(handle)) { status = PTR_ERR(handle); @@ -653,7 +651,7 @@ mlog_errno(status); goto leave; } - + ocfs2_update_inode_fsync_trans(handle, inode, 1); ocfs2_journal_dirty(handle, bh); spin_lock(&OCFS2_I(inode)->ip_lock); @@ -671,11 +669,7 @@ } else { BUG_ON(why != RESTART_TRANS); - /* TODO: This can be more intelligent. */ - credits = ocfs2_calc_extend_credits(osb->sb, - &fe->id2.i_list, - clusters_to_add); - status = ocfs2_extend_trans(handle, credits); + status = ocfs2_allocate_extend_trans(handle, 1); if (status < 0) { /* handle still has to be committed at * this point. */ @@ -719,11 +713,19 @@ return status; } +int ocfs2_extend_allocation(struct inode *inode, u32 logical_start, + u32 clusters_to_add, int mark_unwritten) +{ + return __ocfs2_extend_allocation(inode, logical_start, + clusters_to_add, mark_unwritten); +} + /* * While a write will already be ordering the data, a truncate will not. * Thus, we need to explicitly order the zeroed pages. */ -static handle_t *ocfs2_zero_start_ordered_transaction(struct inode *inode) +static handle_t *ocfs2_zero_start_ordered_transaction(struct inode *inode, + struct buffer_head *di_bh) { struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); handle_t *handle = NULL; @@ -740,8 +742,16 @@ } ret = ocfs2_jbd2_file_inode(handle, inode); - if (ret < 0) + if (ret < 0) { mlog_errno(ret); + goto out; + } + + ret = ocfs2_journal_access_di(handle, INODE_CACHE(inode), di_bh, + OCFS2_JOURNAL_ACCESS_WRITE); + if (ret) + mlog_errno(ret); + ocfs2_update_inode_fsync_trans(handle, inode, 1); out: if (ret) { @@ -756,24 +766,31 @@ * to be too fragile to do exactly what we need without us having to * worry about recursive locking in ->write_begin() and ->write_end(). */ static int ocfs2_write_zero_page(struct inode *inode, u64 abs_from, - u64 abs_to) + u64 abs_to, struct buffer_head *di_bh) { struct address_space *mapping = inode->i_mapping; struct page *page; unsigned long index = abs_from >> PAGE_CACHE_SHIFT; - handle_t *handle = NULL; + handle_t *handle; int ret = 0; unsigned zero_from, zero_to, block_start, block_end; + struct ocfs2_dinode *di = (struct ocfs2_dinode *)di_bh->b_data; BUG_ON(abs_from >= abs_to); BUG_ON(abs_to > (((u64)index + 1) << PAGE_CACHE_SHIFT)); BUG_ON(abs_from & (inode->i_blkbits - 1)); + handle = ocfs2_zero_start_ordered_transaction(inode, di_bh); + if (IS_ERR(handle)) { + ret = PTR_ERR(handle); + goto out; + } + page = find_or_create_page(mapping, index, GFP_NOFS); if (!page) { ret = -ENOMEM; mlog_errno(ret); - goto out; + goto out_commit_trans; } /* Get the offsets within the page that we want to zero */ @@ -805,14 +822,6 @@ goto out_unlock; } - if (!handle) { - handle = ocfs2_zero_start_ordered_transaction(inode); - if (IS_ERR(handle)) { - ret = PTR_ERR(handle); - handle = NULL; - break; - } - } /* must not update i_size! */ ret = block_commit_write(page, block_start + 1, @@ -823,12 +832,29 @@ ret = 0; } - if (handle) - ocfs2_commit_trans(OCFS2_SB(inode->i_sb), handle); + /* + * fs-writeback will release the dirty pages without page lock + * whose offset are over inode size, the release happens at + * block_write_full_page(). + */ + i_size_write(inode, abs_to); + inode->i_blocks = ocfs2_inode_sector_count(inode); + di->i_size = cpu_to_le64((u64)i_size_read(inode)); + inode->i_mtime = inode->i_ctime = CURRENT_TIME; + di->i_mtime = di->i_ctime = cpu_to_le64(inode->i_mtime.tv_sec); + di->i_ctime_nsec = cpu_to_le32(inode->i_mtime.tv_nsec); + di->i_mtime_nsec = di->i_ctime_nsec; + if (handle) { + ocfs2_journal_dirty(handle, di_bh); + ocfs2_update_inode_fsync_trans(handle, inode, 1); + } out_unlock: unlock_page(page); page_cache_release(page); +out_commit_trans: + if (handle) + ocfs2_commit_trans(OCFS2_SB(inode->i_sb), handle); out: return ret; } @@ -899,7 +925,7 @@ zero_clusters = last_cpos - zero_cpos; if (needs_cow) { - rc = ocfs2_refcount_cow(inode, NULL, di_bh, zero_cpos, + rc = ocfs2_refcount_cow(inode, di_bh, zero_cpos, zero_clusters, UINT_MAX); if (rc) { mlog_errno(rc); @@ -920,7 +946,7 @@ * has made sure that the entire range needs zeroing. */ static int ocfs2_zero_extend_range(struct inode *inode, u64 range_start, - u64 range_end) + u64 range_end, struct buffer_head *di_bh) { int rc = 0; u64 next_pos; @@ -936,7 +962,7 @@ next_pos = (zero_pos & PAGE_CACHE_MASK) + PAGE_CACHE_SIZE; if (next_pos > range_end) next_pos = range_end; - rc = ocfs2_write_zero_page(inode, zero_pos, next_pos); + rc = ocfs2_write_zero_page(inode, zero_pos, next_pos, di_bh); if (rc < 0) { mlog_errno(rc); break; @@ -982,7 +1008,7 @@ range_end = zero_to_size; ret = ocfs2_zero_extend_range(inode, range_start, - range_end); + range_end, di_bh); if (ret) { mlog_errno(ret); break; @@ -1105,7 +1131,7 @@ { int status = 0, size_change; int inode_locked = 0; - struct inode *inode = dentry->d_inode; + struct inode *inode = d_inode(dentry); struct super_block *sb = inode->i_sb; struct ocfs2_super *osb = OCFS2_SB(sb); struct buffer_head *bh = NULL; @@ -1133,8 +1159,11 @@ if (status) return status; - if (is_quota_modification(inode, attr)) - dquot_initialize(inode); + if (is_quota_modification(inode, attr)) { + status = dquot_initialize(inode); + if (status) + return status; + } size_change = S_ISREG(inode->i_mode) && attr->ia_valid & ATTR_SIZE; if (size_change) { status = ocfs2_rw_lock(inode, 1); @@ -1152,14 +1181,14 @@ } inode_locked = 1; - if (size_change && attr->ia_size != i_size_read(inode)) { + if (size_change) { status = inode_newsize_ok(inode, attr->ia_size); if (status) goto bail_unlock; inode_dio_wait(inode); - if (i_size_read(inode) > attr->ia_size) { + if (i_size_read(inode) >= attr->ia_size) { if (ocfs2_should_order_data(inode)) { status = ocfs2_begin_ordered_truncate(inode, attr->ia_size); @@ -1188,8 +1217,8 @@ && OCFS2_HAS_RO_COMPAT_FEATURE(sb, OCFS2_FEATURE_RO_COMPAT_USRQUOTA)) { transfer_to[USRQUOTA] = dqget(sb, make_kqid_uid(attr->ia_uid)); - if (!transfer_to[USRQUOTA]) { - status = -ESRCH; + if (IS_ERR(transfer_to[USRQUOTA])) { + status = PTR_ERR(transfer_to[USRQUOTA]); goto bail_unlock; } } @@ -1197,8 +1226,8 @@ && OCFS2_HAS_RO_COMPAT_FEATURE(sb, OCFS2_FEATURE_RO_COMPAT_GRPQUOTA)) { transfer_to[GRPQUOTA] = dqget(sb, make_kqid_gid(attr->ia_gid)); - if (!transfer_to[GRPQUOTA]) { - status = -ESRCH; + if (IS_ERR(transfer_to[GRPQUOTA])) { + status = PTR_ERR(transfer_to[GRPQUOTA]); goto bail_unlock; } } @@ -1239,20 +1268,20 @@ if (size_change) ocfs2_rw_unlock(inode, 1); bail: - brelse(bh); /* Release quota pointers in case we acquired them */ - for (qtype = 0; qtype < MAXQUOTAS; qtype++) + for (qtype = 0; qtype < OCFS2_MAXQUOTAS; qtype++) dqput(transfer_to[qtype]); if (!status && attr->ia_valid & ATTR_MODE) { - status = ocfs2_acl_chmod(inode); + status = ocfs2_acl_chmod(inode, bh); if (status < 0) mlog_errno(status); } if (inode_locked) ocfs2_inode_unlock(inode, 1); + brelse(bh); return status; } @@ -1260,8 +1289,8 @@ struct dentry *dentry, struct kstat *stat) { - struct inode *inode = dentry->d_inode; - struct super_block *sb = dentry->d_inode->i_sb; + struct inode *inode = d_inode(dentry); + struct super_block *sb = d_inode(dentry)->i_sb; struct ocfs2_super *osb = sb->s_fs_info; int err; @@ -1334,6 +1363,7 @@ di = (struct ocfs2_dinode *) bh->b_data; di->i_mode = cpu_to_le16(inode->i_mode); + ocfs2_update_inode_fsync_trans(handle, inode, 0); ocfs2_journal_dirty(handle, bh); @@ -1580,6 +1610,7 @@ if (ret) mlog_errno(ret); } + ocfs2_update_inode_fsync_trans(handle, inode, 1); ocfs2_commit_trans(osb, handle); out: @@ -1807,7 +1838,7 @@ ret = ocfs2_remove_btree_range(inode, &et, trunc_cpos, phys_cpos, trunc_len, flags, - &dealloc, refcount_loc); + &dealloc, refcount_loc, false); if (ret < 0) { mlog_errno(ret); goto out; @@ -1821,6 +1852,7 @@ ocfs2_truncate_cluster_pages(inode, byte_start, byte_len); out: + ocfs2_free_path(path); ocfs2_schedule_truncate_log_flush(osb, 1); ocfs2_run_deallocs(osb, &dealloc); @@ -1894,7 +1926,8 @@ } size = sr->l_start + sr->l_len; - if (cmd == OCFS2_IOC_RESVSP || cmd == OCFS2_IOC_RESVSP64) { + if (cmd == OCFS2_IOC_RESVSP || cmd == OCFS2_IOC_RESVSP64 || + cmd == OCFS2_IOC_UNRESVSP || cmd == OCFS2_IOC_UNRESVSP64) { if (sr->l_len <= 0) { ret = -EINVAL; goto out_inode_unlock; @@ -2063,13 +2096,6 @@ return ret; } -static void ocfs2_aiodio_wait(struct inode *inode) -{ - wait_queue_head_t *wq = ocfs2_ioend_wq(inode); - - wait_event(*wq, (atomic_read(&OCFS2_I(inode)->ip_unaligned_aio) == 0)); -} - static int ocfs2_is_io_unaligned(struct inode *inode, size_t count, loff_t pos) { int blockmask = inode->i_sb->s_blocksize - 1; @@ -2099,7 +2125,7 @@ *meta_level = 1; - ret = ocfs2_refcount_cow(inode, file, di_bh, cpos, clusters, UINT_MAX); + ret = ocfs2_refcount_cow(inode, di_bh, cpos, clusters, UINT_MAX); if (ret) mlog_errno(ret); out: @@ -2108,7 +2134,7 @@ } static int ocfs2_prepare_inode_for_write(struct file *file, - loff_t *ppos, + loff_t pos, size_t count, int appending, int *direct_io, @@ -2116,8 +2142,11 @@ { 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; + struct inode *inode = d_inode(dentry); + loff_t end; + struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); + int full_coherency = !(osb->s_mount_opt & + OCFS2_MOUNT_COHERENCY_BUFFERED); /* * We start with a read level meta lock and only jump to an ex @@ -2154,23 +2183,16 @@ } } - /* 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) @@ -2206,7 +2228,16 @@ * one node could wind up truncating another * nodes writes. */ - if (end > i_size_read(inode)) { + if (end > i_size_read(inode) && !full_coherency) { + *direct_io = 0; + break; + } + + /* + * Fallback to old way if the feature bit is not set. + */ + if (end > i_size_read(inode) && + !ocfs2_supports_append_dio(osb)) { *direct_io = 0; break; } @@ -2217,21 +2248,24 @@ * 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) { - *direct_io = 0; + /* + * Fallback to old way if the feature bit is not set. + * Otherwise try dio first and then complete the rest + * request through buffer io. + */ + if (!ocfs2_supports_append_dio(osb)) + *direct_io = 0; ret = 0; } else if (ret < 0) mlog_errno(ret); 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) @@ -2241,59 +2275,50 @@ return ret; } -static ssize_t ocfs2_file_aio_write(struct kiocb *iocb, - const struct iovec *iov, - unsigned long nr_segs, - loff_t pos) +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; int can_do_direct, has_refcount = 0; ssize_t written = 0; - size_t ocount; /* original count */ - size_t count; /* after file limit checks */ - loff_t old_size, *ppos = &iocb->ki_pos; - u32 old_clusters; + ssize_t ret; + size_t count = iov_iter_count(from), orig_count; struct file *file = iocb->ki_filp; struct inode *inode = file_inode(file); 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; + int append_write = ((iocb->ki_pos + count) >= + i_size_read(inode) ? 1 : 0); trace_ocfs2_file_aio_write(inode, file, file->f_path.dentry, (unsigned long long)OCFS2_I(inode)->ip_blkno, file->f_path.dentry->d_name.len, file->f_path.dentry->d_name.name, - (unsigned int)nr_segs); + (unsigned int)from->nr_segs); /* GRRRRR */ - if (iocb->ki_left == 0) + 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); - ocfs2_iocb_clear_sem_locked(iocb); - relock: - /* to match setattr's i_mutex -> rw_lock ordering */ - if (direct_io) { - have_alloc_sem = 1; - /* communicate with ocfs2_dio_end_io */ - ocfs2_iocb_set_sem_locked(iocb); - } - /* * Concurrent O_DIRECT writes are allowed with * mount_option "coherency=buffered". + * For append write, we must take rw EX. */ - rw_level = (!direct_io || full_coherency); + rw_level = (!direct_io || full_coherency || append_write); ret = ocfs2_rw_lock(inode, rw_level); if (ret < 0) { mlog_errno(ret); - goto out_sems; + goto out_mutex; } /* @@ -2315,9 +2340,17 @@ 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, - iocb->ki_left, appending, + ret = ocfs2_prepare_inode_for_write(file, iocb->ki_pos, count, appending, &can_do_direct, &has_refcount); if (ret < 0) { mlog_errno(ret); @@ -2325,8 +2358,7 @@ } if (direct_io && !is_sync_kiocb(iocb)) - unaligned_dio = ocfs2_is_io_unaligned(inode, iocb->ki_left, - *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 @@ -2335,10 +2367,12 @@ if (direct_io && !can_do_direct) { ocfs2_rw_unlock(inode, rw_level); - have_alloc_sem = 0; rw_level = -1; direct_io = 0; + iocb->ki_flags &= ~IOCB_DIRECT; + iov_iter_reexpand(from, orig_count); + dropped_dio = 1; goto relock; } @@ -2347,66 +2381,44 @@ * Wait on previous unaligned aio to complete before * proceeding. */ - ocfs2_aiodio_wait(inode); - - /* Mark the iocb as needing a decrement in ocfs2_dio_end_io */ - atomic_inc(&OCFS2_I(inode)->ip_unaligned_aio); + mutex_lock(&OCFS2_I(inode)->ip_unaligned_aio); + /* Mark the iocb as needing an unlock in ocfs2_dio_end_io */ ocfs2_iocb_set_unaligned_aio(iocb); } - /* - * To later detect whether a journal commit for sync writes is - * necessary, we sample i_size, and cluster count here. - */ - old_size = i_size_read(inode); - old_clusters = OCFS2_I(inode)->ip_clusters; - /* communicate with ocfs2_dio_end_io */ ocfs2_iocb_set_rw_locked(iocb, rw_level); - ret = generic_segment_checks(iov, &nr_segs, &ocount, - VERIFY_READ); - if (ret) - goto out_dio; - - count = ocount; - ret = generic_write_checks(file, ppos, &count, - S_ISBLK(inode->i_mode)); - if (ret) - goto out_dio; + written = __generic_file_write_iter(iocb, from); + /* buffered aio wouldn't have proper lock coverage today */ + BUG_ON(written == -EIOCBQUEUED && !(iocb->ki_flags & IOCB_DIRECT)); - if (direct_io) { - written = generic_file_direct_write(iocb, iov, &nr_segs, *ppos, - ppos, count, ocount); - if (written < 0) { - ret = written; - goto out_dio; - } - } else { - current->backing_dev_info = file->f_mapping->backing_dev_info; - written = generic_file_buffered_write(iocb, iov, nr_segs, *ppos, - ppos, count, 0); - current->backing_dev_info = NULL; + /* + * deep in g_f_a_w_n()->ocfs2_direct_IO we pass in a ocfs2_dio_end_io + * function pointer which is called when o_direct io completes so that + * it can unlock our rw lock. + * Unfortunately there are error cases which call end_io and others + * that don't. so we don't have to unlock the rw_lock if either an + * async dio is going to do it in the future or an end_io after an + * error has already done it. + */ + if ((written == -EIOCBQUEUED) || (!ocfs2_iocb_is_rw_locked(iocb))) { + rw_level = -1; + unaligned_dio = 0; } -out_dio: - /* buffered aio wouldn't have proper lock coverage today */ - BUG_ON(ret == -EIOCBQUEUED && !(file->f_flags & O_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); if (ret < 0) written = ret; - if (!ret && ((old_size != i_size_read(inode)) || - (old_clusters != OCFS2_I(inode)->ip_clusters) || - has_refcount)) { + if (!ret) { ret = jbd2_journal_force_commit(osb->journal->j_journal); if (ret < 0) written = ret; @@ -2419,34 +2431,16 @@ } no_sync: - /* - * deep in g_f_a_w_n()->ocfs2_direct_IO we pass in a ocfs2_dio_end_io - * function pointer which is called when o_direct io completes so that - * it can unlock our rw lock. - * Unfortunately there are error cases which call end_io and others - * that don't. so we don't have to unlock the rw_lock if either an - * async dio is going to do it in the future or an end_io after an - * error has already done it. - */ - if ((ret == -EIOCBQUEUED) || (!ocfs2_iocb_is_rw_locked(iocb))) { - rw_level = -1; - have_alloc_sem = 0; - unaligned_dio = 0; - } - - if (unaligned_dio) { + if (unaligned_dio && ocfs2_iocb_is_unaligned_aio(iocb)) { ocfs2_iocb_clear_unaligned_aio(iocb); - atomic_dec(&OCFS2_I(inode)->ip_unaligned_aio); + mutex_unlock(&OCFS2_I(inode)->ip_unaligned_aio); } out: if (rw_level != -1) ocfs2_rw_unlock(inode, rw_level); -out_sems: - if (have_alloc_sem) - ocfs2_iocb_clear_sem_locked(iocb); - +out_mutex: mutex_unlock(&inode->i_mutex); if (written) @@ -2454,86 +2448,6 @@ return ret; } -static int ocfs2_splice_to_file(struct pipe_inode_info *pipe, - struct file *out, - struct splice_desc *sd) -{ - int ret; - - ret = ocfs2_prepare_inode_for_write(out, &sd->pos, - sd->total_len, 0, NULL, NULL); - if (ret < 0) { - mlog_errno(ret); - return ret; - } - - return splice_from_pipe_feed(pipe, sd, pipe_to_file); -} - -static ssize_t ocfs2_file_splice_write(struct pipe_inode_info *pipe, - struct file *out, - loff_t *ppos, - size_t len, - unsigned int flags) -{ - int ret; - struct address_space *mapping = out->f_mapping; - struct inode *inode = mapping->host; - struct splice_desc sd = { - .flags = flags, - .u.file = out, - }; - ret = generic_write_checks(out, ppos, &len, 0); - if(ret) - return ret; - sd.total_len = len; - sd.pos = *ppos; - - trace_ocfs2_file_splice_write(inode, out, out->f_path.dentry, - (unsigned long long)OCFS2_I(inode)->ip_blkno, - out->f_path.dentry->d_name.len, - out->f_path.dentry->d_name.name, len); - - pipe_lock(pipe); - - splice_from_pipe_begin(&sd); - do { - ret = splice_from_pipe_next(pipe, &sd); - if (ret <= 0) - break; - - mutex_lock_nested(&inode->i_mutex, I_MUTEX_CHILD); - ret = ocfs2_rw_lock(inode, 1); - if (ret < 0) - mlog_errno(ret); - else { - ret = ocfs2_splice_to_file(pipe, out, &sd); - ocfs2_rw_unlock(inode, 1); - } - mutex_unlock(&inode->i_mutex); - } while (ret > 0); - splice_from_pipe_end(pipe, &sd); - - pipe_unlock(pipe); - - if (sd.num_spliced) - ret = sd.num_spliced; - - if (ret > 0) { - int err; - - err = generic_write_sync(out, *ppos, ret); - if (err) - ret = err; - else - *ppos += ret; - - balance_dirty_pages_ratelimited(mapping); - } - - return ret; -} - static ssize_t ocfs2_file_splice_read(struct file *in, loff_t *ppos, struct pipe_inode_info *pipe, @@ -2549,7 +2463,7 @@ in->f_path.dentry->d_name.name, len); /* - * See the comment in ocfs2_file_aio_read() + * See the comment in ocfs2_file_read_iter() */ ret = ocfs2_inode_lock_atime(inode, in->f_path.mnt, &lock_level); if (ret < 0) { @@ -2564,19 +2478,18 @@ return ret; } -static ssize_t ocfs2_file_aio_read(struct kiocb *iocb, - const struct iovec *iov, - unsigned long nr_segs, - loff_t pos) +static ssize_t ocfs2_file_read_iter(struct kiocb *iocb, + struct iov_iter *to) { - int ret = 0, rw_level = -1, have_alloc_sem = 0, lock_level = 0; + int ret = 0, rw_level = -1, lock_level = 0; struct file *filp = iocb->ki_filp; struct inode *inode = file_inode(filp); trace_ocfs2_file_aio_read(inode, filp, filp->f_path.dentry, (unsigned long long)OCFS2_I(inode)->ip_blkno, filp->f_path.dentry->d_name.len, - filp->f_path.dentry->d_name.name, nr_segs); + filp->f_path.dentry->d_name.name, + to->nr_segs); /* GRRRRR */ if (!inode) { @@ -2585,16 +2498,11 @@ goto bail; } - ocfs2_iocb_clear_sem_locked(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) { - have_alloc_sem = 1; - ocfs2_iocb_set_sem_locked(iocb); - + if (iocb->ki_flags & IOCB_DIRECT) { ret = ocfs2_rw_lock(inode, 0); if (ret < 0) { mlog_errno(ret); @@ -2621,22 +2529,18 @@ } ocfs2_inode_unlock(inode, lock_level); - ret = generic_file_aio_read(iocb, iov, nr_segs, iocb->ki_pos); + ret = generic_file_read_iter(iocb, to); 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_aio_write */ + /* see ocfs2_file_write_iter */ if (ret == -EIOCBQUEUED || !ocfs2_iocb_is_rw_locked(iocb)) { rw_level = -1; - have_alloc_sem = 0; } bail: - if (have_alloc_sem) - ocfs2_iocb_clear_sem_locked(iocb); - if (rw_level != -1) ocfs2_rw_unlock(inode, rw_level); @@ -2655,7 +2559,16 @@ case SEEK_SET: break; case SEEK_END: - offset += inode->i_size; + /* SEEK_END requires the OCFS2 inode lock for the file + * because it references the file's size. + */ + ret = ocfs2_inode_lock(inode, NULL, 0); + if (ret < 0) { + mlog_errno(ret); + goto out; + } + offset += i_size_read(inode); + ocfs2_inode_unlock(inode, 0); break; case SEEK_CUR: if (offset == 0) { @@ -2675,17 +2588,7 @@ goto out; } - if (offset < 0 && !(file->f_mode & FMODE_UNSIGNED_OFFSET)) - ret = -EINVAL; - if (!ret && offset > inode->i_sb->s_maxbytes) - ret = -EINVAL; - if (ret) - goto out; - - if (offset != file->f_pos) { - file->f_pos = offset; - file->f_version = 0; - } + offset = vfs_setpos(file, offset, inode->i_sb->s_maxbytes); out: mutex_unlock(&inode->i_mutex); @@ -2704,6 +2607,7 @@ .removexattr = generic_removexattr, .fiemap = ocfs2_fiemap, .get_acl = ocfs2_iop_get_acl, + .set_acl = ocfs2_iop_set_acl, }; const struct inode_operations ocfs2_special_file_iops = { @@ -2711,6 +2615,7 @@ .getattr = ocfs2_getattr, .permission = ocfs2_permission, .get_acl = ocfs2_iop_get_acl, + .set_acl = ocfs2_iop_set_acl, }; /* @@ -2719,14 +2624,12 @@ */ const struct file_operations ocfs2_fops = { .llseek = ocfs2_file_llseek, - .read = do_sync_read, - .write = do_sync_write, .mmap = ocfs2_mmap, .fsync = ocfs2_sync_file, .release = ocfs2_file_release, .open = ocfs2_file_open, - .aio_read = ocfs2_file_aio_read, - .aio_write = ocfs2_file_aio_write, + .read_iter = ocfs2_file_read_iter, + .write_iter = ocfs2_file_write_iter, .unlocked_ioctl = ocfs2_ioctl, #ifdef CONFIG_COMPAT .compat_ioctl = ocfs2_compat_ioctl, @@ -2734,14 +2637,14 @@ .lock = ocfs2_lock, .flock = ocfs2_flock, .splice_read = ocfs2_file_splice_read, - .splice_write = ocfs2_file_splice_write, + .splice_write = iter_file_splice_write, .fallocate = ocfs2_fallocate, }; const struct file_operations ocfs2_dops = { .llseek = generic_file_llseek, .read = generic_read_dir, - .readdir = ocfs2_readdir, + .iterate = ocfs2_readdir, .fsync = ocfs2_sync_file, .release = ocfs2_dir_release, .open = ocfs2_dir_open, @@ -2767,28 +2670,26 @@ */ const struct file_operations ocfs2_fops_no_plocks = { .llseek = ocfs2_file_llseek, - .read = do_sync_read, - .write = do_sync_write, .mmap = ocfs2_mmap, .fsync = ocfs2_sync_file, .release = ocfs2_file_release, .open = ocfs2_file_open, - .aio_read = ocfs2_file_aio_read, - .aio_write = ocfs2_file_aio_write, + .read_iter = ocfs2_file_read_iter, + .write_iter = ocfs2_file_write_iter, .unlocked_ioctl = ocfs2_ioctl, #ifdef CONFIG_COMPAT .compat_ioctl = ocfs2_compat_ioctl, #endif .flock = ocfs2_flock, .splice_read = ocfs2_file_splice_read, - .splice_write = ocfs2_file_splice_write, + .splice_write = iter_file_splice_write, .fallocate = ocfs2_fallocate, }; const struct file_operations ocfs2_dops_no_plocks = { .llseek = generic_file_llseek, .read = generic_read_dir, - .readdir = ocfs2_readdir, + .iterate = ocfs2_readdir, .fsync = ocfs2_sync_file, .release = ocfs2_dir_release, .open = ocfs2_dir_open,