diff options
Diffstat (limited to 'fs')
79 files changed, 1000 insertions, 592 deletions
diff --git a/fs/anon_inodes.c b/fs/anon_inodes.c index 2c994591f4d..9f0bf13291e 100644 --- a/fs/anon_inodes.c +++ b/fs/anon_inodes.c @@ -121,13 +121,13 @@ struct file *anon_inode_getfile(const char *name, d_instantiate(path.dentry, anon_inode_inode); error = -ENFILE; - file = alloc_file(&path, FMODE_READ | FMODE_WRITE, fops); + file = alloc_file(&path, OPEN_FMODE(flags), fops); if (!file) goto err_dput; file->f_mapping = anon_inode_inode->i_mapping; file->f_pos = 0; - file->f_flags = O_RDWR | (flags & O_NONBLOCK); + file->f_flags = flags & (O_ACCMODE | O_NONBLOCK); file->f_version = 0; file->private_data = priv; diff --git a/fs/binfmt_elf_fdpic.c b/fs/binfmt_elf_fdpic.c index c25256a5c5b..c57d9ce5ff7 100644 --- a/fs/binfmt_elf_fdpic.c +++ b/fs/binfmt_elf_fdpic.c @@ -171,6 +171,9 @@ static int load_elf_fdpic_binary(struct linux_binprm *bprm, #ifdef ELF_FDPIC_PLAT_INIT unsigned long dynaddr; #endif +#ifndef CONFIG_MMU + unsigned long stack_prot; +#endif struct file *interpreter = NULL; /* to shut gcc up */ char *interpreter_name = NULL; int executable_stack; @@ -316,6 +319,8 @@ static int load_elf_fdpic_binary(struct linux_binprm *bprm, * defunct, deceased, etc. after this point we have to exit via * error_kill */ set_personality(PER_LINUX_FDPIC); + if (elf_read_implies_exec(&exec_params.hdr, executable_stack)) + current->personality |= READ_IMPLIES_EXEC; set_binfmt(&elf_fdpic_format); current->mm->start_code = 0; @@ -377,9 +382,13 @@ static int load_elf_fdpic_binary(struct linux_binprm *bprm, if (stack_size < PAGE_SIZE * 2) stack_size = PAGE_SIZE * 2; + stack_prot = PROT_READ | PROT_WRITE; + if (executable_stack == EXSTACK_ENABLE_X || + (executable_stack == EXSTACK_DEFAULT && VM_STACK_FLAGS & VM_EXEC)) + stack_prot |= PROT_EXEC; + down_write(¤t->mm->mmap_sem); - current->mm->start_brk = do_mmap(NULL, 0, stack_size, - PROT_READ | PROT_WRITE | PROT_EXEC, + current->mm->start_brk = do_mmap(NULL, 0, stack_size, stack_prot, MAP_PRIVATE | MAP_ANONYMOUS | MAP_UNINITIALIZED | MAP_GROWSDOWN, 0); @@ -1798,11 +1807,11 @@ static int elf_fdpic_core_dump(struct coredump_params *cprm) ELF_CORE_WRITE_EXTRA_DATA; #endif - if (file->f_pos != offset) { + if (cprm->file->f_pos != offset) { /* Sanity check */ printk(KERN_WARNING "elf_core_dump: file->f_pos (%lld) != offset (%lld)\n", - file->f_pos, offset); + cprm->file->f_pos, offset); } end_coredump: diff --git a/fs/cifs/CHANGES b/fs/cifs/CHANGES index 094ea65afc8..7b2600b380d 100644 --- a/fs/cifs/CHANGES +++ b/fs/cifs/CHANGES @@ -5,7 +5,9 @@ have duplicated data). Fix oops in cifs_lookup. Workaround problem mounting to OS/400 Netserve. Fix oops in cifs_get_tcp_session. Disable use of server inode numbers when server only partially supports them (e.g. for one server querying inode numbers on -FindFirst fails but QPathInfo queries works). +FindFirst fails but QPathInfo queries works). Fix oops with dfs in +cifs_put_smb_ses. Fix mmap to work on directio mounts (needed +for OpenOffice when on forcedirectio mount e.g.) Version 1.60 ------------- diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index 29f1da761bb..8c6a0362717 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -758,7 +758,7 @@ const struct file_operations cifs_file_ops = { }; const struct file_operations cifs_file_direct_ops = { - /* no mmap, no aio, no readv - + /* no aio, no readv - BB reevaluate whether they can be done with directio, no cache */ .read = cifs_user_read, .write = cifs_user_write, @@ -767,6 +767,7 @@ const struct file_operations cifs_file_direct_ops = { .lock = cifs_lock, .fsync = cifs_fsync, .flush = cifs_flush, + .mmap = cifs_file_mmap, .splice_read = generic_file_splice_read, #ifdef CONFIG_CIFS_POSIX .unlocked_ioctl = cifs_ioctl, diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 63ea83ff687..3bbcaa716b3 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -2287,12 +2287,12 @@ int cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, char *mount_data_global, const char *devname) { - int rc = 0; + int rc; int xid; struct smb_vol *volume_info; - struct cifsSesInfo *pSesInfo = NULL; - struct cifsTconInfo *tcon = NULL; - struct TCP_Server_Info *srvTcp = NULL; + struct cifsSesInfo *pSesInfo; + struct cifsTconInfo *tcon; + struct TCP_Server_Info *srvTcp; char *full_path; char *mount_data = mount_data_global; #ifdef CONFIG_CIFS_DFS_UPCALL @@ -2301,6 +2301,10 @@ cifs_mount(struct super_block *sb, struct cifs_sb_info *cifs_sb, int referral_walks_count = 0; try_mount_again: #endif + rc = 0; + tcon = NULL; + pSesInfo = NULL; + srvTcp = NULL; full_path = NULL; xid = GetXid(); @@ -2597,6 +2601,7 @@ remote_path_check: cleanup_volume_info(&volume_info); referral_walks_count++; + FreeXid(xid); goto try_mount_again; } #else /* No DFS support, return error on mount */ diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c index 14cbc831422..332dd00f089 100644 --- a/fs/compat_ioctl.c +++ b/fs/compat_ioctl.c @@ -1600,8 +1600,6 @@ static long do_ioctl_trans(int fd, unsigned int cmd, case KDSKBMETA: case KDSKBLED: case KDSETLED: - /* SG stuff */ - case SG_SET_TRANSFORM: /* AUTOFS */ case AUTOFS_IOC_READY: case AUTOFS_IOC_FAIL: diff --git a/fs/eventfd.c b/fs/eventfd.c index 8b47e4200e6..d26402ff06e 100644 --- a/fs/eventfd.c +++ b/fs/eventfd.c @@ -339,7 +339,7 @@ struct file *eventfd_file_create(unsigned int count, int flags) ctx->flags = flags; file = anon_inode_getfile("[eventfd]", &eventfd_fops, ctx, - flags & EFD_SHARED_FCNTL_FLAGS); + O_RDWR | (flags & EFD_SHARED_FCNTL_FLAGS)); if (IS_ERR(file)) eventfd_free_ctx(ctx); diff --git a/fs/eventpoll.c b/fs/eventpoll.c index 366c503f965..bd056a5b4ef 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -1206,7 +1206,7 @@ SYSCALL_DEFINE1(epoll_create1, int, flags) * a file structure and a free file descriptor. */ error = anon_inode_getfd("[eventpoll]", &eventpoll_fops, ep, - flags & O_CLOEXEC); + O_RDWR | (flags & O_CLOEXEC)); if (error < 0) ep_free(ep); diff --git a/fs/exofs/inode.c b/fs/exofs/inode.c index 698a8636d39..2afbcebeda7 100644 --- a/fs/exofs/inode.c +++ b/fs/exofs/inode.c @@ -738,13 +738,28 @@ static int exofs_write_begin_export(struct file *file, fsdata); } +static int exofs_write_end(struct file *file, struct address_space *mapping, + loff_t pos, unsigned len, unsigned copied, + struct page *page, void *fsdata) +{ + struct inode *inode = mapping->host; + /* According to comment in simple_write_end i_mutex is held */ + loff_t i_size = inode->i_size; + int ret; + + ret = simple_write_end(file, mapping,pos, len, copied, page, fsdata); + if (i_size != inode->i_size) + mark_inode_dirty(inode); + return ret; +} + const struct address_space_operations exofs_aops = { .readpage = exofs_readpage, .readpages = exofs_readpages, .writepage = exofs_writepage, .writepages = exofs_writepages, .write_begin = exofs_write_begin_export, - .write_end = simple_write_end, + .write_end = exofs_write_end, }; /****************************************************************************** diff --git a/fs/exofs/pnfs.h b/fs/exofs/pnfs.h index 423033addd1..c52e9888b8a 100644 --- a/fs/exofs/pnfs.h +++ b/fs/exofs/pnfs.h @@ -15,13 +15,7 @@ #ifndef __EXOFS_PNFS_H__ #define __EXOFS_PNFS_H__ -#if defined(CONFIG_PNFS) - - -/* FIXME: move this file to: linux/exportfs/pnfs_osd_xdr.h */ -#include "../nfs/objlayout/pnfs_osd_xdr.h" - -#else /* defined(CONFIG_PNFS) */ +#if ! defined(__PNFS_OSD_XDR_H__) enum pnfs_iomode { IOMODE_READ = 1, @@ -46,6 +40,6 @@ struct pnfs_osd_data_map { u32 odm_raid_algorithm; }; -#endif /* else defined(CONFIG_PNFS) */ +#endif /* ! defined(__PNFS_OSD_XDR_H__) */ #endif /* __EXOFS_PNFS_H__ */ diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c index ad14227f509..455e6e6e5cb 100644 --- a/fs/ext3/inode.c +++ b/fs/ext3/inode.c @@ -970,7 +970,7 @@ static int ext3_get_block(struct inode *inode, sector_t iblock, if (max_blocks > DIO_MAX_BLOCKS) max_blocks = DIO_MAX_BLOCKS; handle = ext3_journal_start(inode, DIO_CREDITS + - 2 * EXT3_QUOTA_TRANS_BLOCKS(inode->i_sb)); + EXT3_MAXQUOTAS_TRANS_BLOCKS(inode->i_sb)); if (IS_ERR(handle)) { ret = PTR_ERR(handle); goto out; @@ -3146,8 +3146,8 @@ int ext3_setattr(struct dentry *dentry, struct iattr *attr) /* (user+group)*(old+new) structure, inode write (sb, * inode block, ? - but truncate inode update has it) */ - handle = ext3_journal_start(inode, 2*(EXT3_QUOTA_INIT_BLOCKS(inode->i_sb)+ - EXT3_QUOTA_DEL_BLOCKS(inode->i_sb))+3); + handle = ext3_journal_start(inode, EXT3_MAXQUOTAS_INIT_BLOCKS(inode->i_sb)+ + EXT3_MAXQUOTAS_DEL_BLOCKS(inode->i_sb)+3); if (IS_ERR(handle)) { error = PTR_ERR(handle); goto err_out; @@ -3239,7 +3239,7 @@ static int ext3_writepage_trans_blocks(struct inode *inode) #ifdef CONFIG_QUOTA /* We know that structure was already allocated during vfs_dq_init so * we will be updating only the data blocks + inodes */ - ret += 2*EXT3_QUOTA_TRANS_BLOCKS(inode->i_sb); + ret += EXT3_MAXQUOTAS_TRANS_BLOCKS(inode->i_sb); #endif return ret; diff --git a/fs/ext3/namei.c b/fs/ext3/namei.c index aad6400c9b7..7b0e44f7d66 100644 --- a/fs/ext3/namei.c +++ b/fs/ext3/namei.c @@ -1699,7 +1699,7 @@ static int ext3_create (struct inode * dir, struct dentry * dentry, int mode, retry: handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) + EXT3_INDEX_EXTRA_TRANS_BLOCKS + 3 + - 2*EXT3_QUOTA_INIT_BLOCKS(dir->i_sb)); + EXT3_MAXQUOTAS_INIT_BLOCKS(dir->i_sb)); if (IS_ERR(handle)) return PTR_ERR(handle); @@ -1733,7 +1733,7 @@ static int ext3_mknod (struct inode * dir, struct dentry *dentry, retry: handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) + EXT3_INDEX_EXTRA_TRANS_BLOCKS + 3 + - 2*EXT3_QUOTA_INIT_BLOCKS(dir->i_sb)); + EXT3_MAXQUOTAS_INIT_BLOCKS(dir->i_sb)); if (IS_ERR(handle)) return PTR_ERR(handle); @@ -1769,7 +1769,7 @@ static int ext3_mkdir(struct inode * dir, struct dentry * dentry, int mode) retry: handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) + EXT3_INDEX_EXTRA_TRANS_BLOCKS + 3 + - 2*EXT3_QUOTA_INIT_BLOCKS(dir->i_sb)); + EXT3_MAXQUOTAS_INIT_BLOCKS(dir->i_sb)); if (IS_ERR(handle)) return PTR_ERR(handle); @@ -1920,7 +1920,7 @@ int ext3_orphan_add(handle_t *handle, struct inode *inode) struct ext3_iloc iloc; int err = 0, rc; - lock_super(sb); + mutex_lock(&EXT3_SB(sb)->s_orphan_lock); if (!list_empty(&EXT3_I(inode)->i_orphan)) goto out_unlock; @@ -1929,9 +1929,13 @@ int ext3_orphan_add(handle_t *handle, struct inode *inode) /* @@@ FIXME: Observation from aviro: * I think I can trigger J_ASSERT in ext3_orphan_add(). We block - * here (on lock_super()), so race with ext3_link() which might bump + * here (on s_orphan_lock), so race with ext3_link() which might bump * ->i_nlink. For, say it, character device. Not a regular file, * not a directory, not a symlink and ->i_nlink > 0. + * + * tytso, 4/25/2009: I'm not sure how that could happen; + * shouldn't the fs core protect us from these sort of + * unlink()/link() races? */ J_ASSERT ((S_ISREG(inode->i_mode) || S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode)) || inode->i_nlink == 0); @@ -1968,7 +1972,7 @@ int ext3_orphan_add(handle_t *handle, struct inode *inode) jbd_debug(4, "orphan inode %lu will point to %d\n", inode->i_ino, NEXT_ORPHAN(inode)); out_unlock: - unlock_super(sb); + mutex_unlock(&EXT3_SB(sb)->s_orphan_lock); ext3_std_error(inode->i_sb, err); return err; } @@ -1986,11 +1990,9 @@ int ext3_orphan_del(handle_t *handle, struct inode *inode) struct ext3_iloc iloc; int err = 0; - lock_super(inode->i_sb); - if (list_empty(&ei->i_orphan)) { - unlock_super(inode->i_sb); - return 0; - } + mutex_lock(&EXT3_SB(inode->i_sb)->s_orphan_lock); + if (list_empty(&ei->i_orphan)) + goto out; ino_next = NEXT_ORPHAN(inode); prev = ei->i_orphan.prev; @@ -2040,7 +2042,7 @@ int ext3_orphan_del(handle_t *handle, struct inode *inode) out_err: ext3_std_error(inode->i_sb, err); out: - unlock_super(inode->i_sb); + mutex_unlock(&EXT3_SB(inode->i_sb)->s_orphan_lock); return err; out_brelse: @@ -2175,7 +2177,7 @@ static int ext3_symlink (struct inode * dir, retry: handle = ext3_journal_start(dir, EXT3_DATA_TRANS_BLOCKS(dir->i_sb) + EXT3_INDEX_EXTRA_TRANS_BLOCKS + 5 + - 2*EXT3_QUOTA_INIT_BLOCKS(dir->i_sb)); + EXT3_MAXQUOTAS_INIT_BLOCKS(dir->i_sb)); if (IS_ERR(handle)) return PTR_ERR(handle); diff --git a/fs/ext3/resize.c b/fs/ext3/resize.c index 5f83b617917..54351ac7cef 100644 --- a/fs/ext3/resize.c +++ b/fs/ext3/resize.c @@ -209,7 +209,7 @@ static int setup_new_group_blocks(struct super_block *sb, if (IS_ERR(handle)) return PTR_ERR(handle); - lock_super(sb); + mutex_lock(&sbi->s_resize_lock); if (input->group != sbi->s_groups_count) { err = -EBUSY; goto exit_journal; @@ -324,7 +324,7 @@ exit_bh: brelse(bh); exit_journal: - unlock_super(sb); + mutex_unlock(&sbi->s_resize_lock); if ((err2 = ext3_journal_stop(handle)) && !err) err = err2; @@ -662,11 +662,12 @@ exit_free: * important part is that the new block and inode counts are in the backup * superblocks, and the location of the new group metadata in the GDT backups. * - * We do not need lock_super() for this, because these blocks are not - * otherwise touched by the filesystem code when it is mounted. We don't - * need to worry about last changing from sbi->s_groups_count, because the - * worst that can happen is that we do not copy the full number of backups - * at this time. The resize which changed s_groups_count will backup again. + * We do not need take the s_resize_lock for this, because these + * blocks are not otherwise touched by the filesystem code when it is + * mounted. We don't need to worry about last changing from + * sbi->s_groups_count, because the worst that can happen is that we + * do not copy the full number of backups at this time. The resize + * which changed s_groups_count will backup again. */ static void update_backups(struct super_block *sb, int blk_off, char *data, int size) @@ -825,7 +826,7 @@ int ext3_group_add(struct super_block *sb, struct ext3_new_group_data *input) goto exit_put; } - lock_super(sb); + mutex_lock(&sbi->s_resize_lock); if (input->group != sbi->s_groups_count) { ext3_warning(sb, __func__, "multiple resizers run on filesystem!"); @@ -856,7 +857,7 @@ int ext3_group_add(struct super_block *sb, struct ext3_new_group_data *input) /* * OK, now we've set up the new group. Time to make it active. * - * Current kernels don't lock all allocations via lock_super(), + * We do not lock all allocations via s_resize_lock * so we have to be safe wrt. concurrent accesses the group * data. So we need to be careful to set all of the relevant * group descriptor data etc. *before* we enable the group. @@ -900,12 +901,12 @@ int ext3_group_add(struct super_block *sb, struct ext3_new_group_data *input) * * The precise rules we use are: * - * * Writers of s_groups_count *must* hold lock_super + * * Writers of s_groups_count *must* hold s_resize_lock * AND * * Writers must perform a smp_wmb() after updating all dependent * data and before modifying the groups count * - * * Readers must hold lock_super() over the access + * * Readers must hold s_resize_lock over the access * OR * * Readers must perform an smp_rmb() after reading the groups count * and before reading any dependent data. @@ -936,7 +937,7 @@ int ext3_group_add(struct super_block *sb, struct ext3_new_group_data *input) ext3_journal_dirty_metadata(handle, sbi->s_sbh); exit_journal: - unlock_super(sb); + mutex_unlock(&sbi->s_resize_lock); if ((err2 = ext3_journal_stop(handle)) && !err) err = err2; if (!err) { @@ -973,7 +974,7 @@ int ext3_group_extend(struct super_block *sb, struct ext3_super_block *es, /* We don't need to worry about locking wrt other resizers just * yet: we're going to revalidate es->s_blocks_count after - * taking lock_super() below. */ + * taking the s_resize_lock below. */ o_blocks_count = le32_to_cpu(es->s_blocks_count); o_groups_count = EXT3_SB(sb)->s_groups_count; @@ -1045,11 +1046,11 @@ int ext3_group_extend(struct super_block *sb, struct ext3_super_block *es, goto exit_put; } - lock_super(sb); + mutex_lock(&EXT3_SB(sb)->s_resize_lock); if (o_blocks_count != le32_to_cpu(es->s_blocks_count)) { ext3_warning(sb, __func__, "multiple resizers run on filesystem!"); - unlock_super(sb); + mutex_unlock(&EXT3_SB(sb)->s_resize_lock); ext3_journal_stop(handle); err = -EBUSY; goto exit_put; @@ -1059,13 +1060,13 @@ int ext3_group_extend(struct super_block *sb, struct ext3_super_block *es, EXT3_SB(sb)->s_sbh))) { ext3_warning(sb, __func__, "error %d on journal write access", err); - unlock_super(sb); + mutex_unlock(&EXT3_SB(sb)->s_resize_lock); ext3_journal_stop(handle); goto exit_put; } es->s_blocks_count = cpu_to_le32(o_blocks_count + add); ext3_journal_dirty_metadata(handle, EXT3_SB(sb)->s_sbh); - unlock_super(sb); + mutex_unlock(&EXT3_SB(sb)->s_resize_lock); ext3_debug("freeing blocks %lu through "E3FSBLK"\n", o_blocks_count, o_blocks_count + add); ext3_free_blocks_sb(handle, sb, o_blocks_count, add, &freed_blocks); diff --git a/fs/ext3/super.c b/fs/ext3/super.c index 7ad1e8c30bd..afa2b569da1 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -1928,6 +1928,8 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent) sb->dq_op = &ext3_quota_operations; #endif INIT_LIST_HEAD(&sbi->s_orphan); /* unlinked but open files */ + mutex_init(&sbi->s_orphan_lock); + mutex_init(&sbi->s_resize_lock); sb->s_root = NULL; @@ -2014,14 +2016,7 @@ static int ext3_fill_super (struct super_block *sb, void *data, int silent) } ext3_setup_super (sb, es, sb->s_flags & MS_RDONLY); - /* - * akpm: core read_super() calls in here with the superblock locked. - * That deadlocks, because orphan cleanup needs to lock the superblock - * in numerous places. Here we just pop the lock - it's relatively - * harmless, because we are now ready to accept write_super() requests, - * and aviro says that's the only reason for hanging onto the - * superblock lock. - */ + EXT3_SB(sb)->s_mount_state |= EXT3_ORPHAN_FS; ext3_orphan_cleanup(sb, es); EXT3_SB(sb)->s_mount_state &= ~EXT3_ORPHAN_FS; @@ -2403,13 +2398,11 @@ static void ext3_mark_recovery_complete(struct super_block * sb, if (journal_flush(journal) < 0) goto out; - lock_super(sb); if (EXT3_HAS_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_RECOVER) && sb->s_flags & MS_RDONLY) { EXT3_CLEAR_INCOMPAT_FEATURE(sb, EXT3_FEATURE_INCOMPAT_RECOVER); ext3_commit_super(sb, es, 1); } - unlock_super(sb); out: journal_unlock_updates(journal); @@ -2601,13 +2594,7 @@ static int ext3_remount (struct super_block * sb, int * flags, char * data) (sbi->s_mount_state & EXT3_VALID_FS)) es->s_state = cpu_to_le16(sbi->s_mount_state); - /* - * We have to unlock super so that we can wait for - * transactions. - */ - unlock_super(sb); ext3_mark_recovery_complete(sb, es); - lock_super(sb); } else { __le32 ret; if ((ret = EXT3_HAS_RO_COMPAT_FEATURE(sb, diff --git a/fs/ext4/Kconfig b/fs/ext4/Kconfig index 9acf7e80813..9ed1bb1f319 100644 --- a/fs/ext4/Kconfig +++ b/fs/ext4/Kconfig @@ -28,6 +28,7 @@ config EXT4_FS config EXT4_USE_FOR_EXT23 bool "Use ext4 for ext2/ext3 file systems" + depends on EXT4_FS depends on EXT3_FS=n || EXT2_FS=n default y help diff --git a/fs/ext4/block_validity.c b/fs/ext4/block_validity.c index 4df8621ec31..a60ab9aad57 100644 --- a/fs/ext4/block_validity.c +++ b/fs/ext4/block_validity.c @@ -16,7 +16,6 @@ #include <linux/module.h> #include <linux/swap.h> #include <linux/pagemap.h> -#include <linux/version.h> #include <linux/blkdev.h> #include <linux/mutex.h> #include "ext4.h" diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index ab31e65d46d..af7b62699ea 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -699,11 +699,17 @@ struct ext4_inode_info { unsigned int i_reserved_meta_blocks; unsigned int i_allocated_meta_blocks; unsigned short i_delalloc_reserved_flag; + sector_t i_da_metadata_calc_last_lblock; + int i_da_metadata_calc_len; /* on-disk additional length */ __u16 i_extra_isize; spinlock_t i_block_reservation_lock; +#ifdef CONFIG_QUOTA + /* quota space reservation, managed internally by quota code */ + qsize_t i_reserved_quota; +#endif /* completed async DIOs that might need unwritten extents handling */ struct list_head i_aio_dio_complete_list; @@ -1435,7 +1441,7 @@ extern int ext4_chunk_trans_blocks(struct inode *, int nrblocks); extern int ext4_block_truncate_page(handle_t *handle, struct address_space *mapping, loff_t from); extern int ext4_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf); -extern qsize_t ext4_get_reserved_space(struct inode *inode); +extern qsize_t *ext4_get_reserved_space(struct inode *inode); extern int flush_aio_dio_completed_IO(struct inode *inode); /* ioctl.c */ extern long ext4_ioctl(struct file *, unsigned int, unsigned long); diff --git a/fs/ext4/ext4_extents.h b/fs/ext4/ext4_extents.h index 2ca686454e8..bdb6ce7e2eb 100644 --- a/fs/ext4/ext4_extents.h +++ b/fs/ext4/ext4_extents.h @@ -225,7 +225,8 @@ static inline void ext4_ext_mark_initialized(struct ext4_extent *ext) ext->ee_len = cpu_to_le16(ext4_ext_get_actual_len(ext)); } -extern int ext4_ext_calc_metadata_amount(struct inode *inode, int blocks); +extern int ext4_ext_calc_metadata_amount(struct inode *inode, + sector_t lblocks); extern ext4_fsblk_t ext_pblock(struct ext4_extent *ex); extern ext4_fsblk_t idx_pblock(struct ext4_extent_idx *); extern void ext4_ext_store_pblock(struct ext4_extent *, ext4_fsblk_t); diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 3a7928f825e..7d7b74e9468 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -296,29 +296,44 @@ static inline int ext4_ext_space_root_idx(struct inode *inode, int check) * to allocate @blocks * Worse case is one block per extent */ -int ext4_ext_calc_metadata_amount(struct inode *inode, int blocks) +int ext4_ext_calc_metadata_amount(struct inode *inode, sector_t lblock) { - int lcap, icap, rcap, leafs, idxs, num; - int newextents = blocks; - - rcap = ext4_ext_space_root_idx(inode, 0); - lcap = ext4_ext_space_block(inode, 0); - icap = ext4_ext_space_block_idx(inode, 0); + struct ext4_inode_info *ei = EXT4_I(inode); + int idxs, num = 0; - /* number of new leaf blocks needed */ - num = leafs = (newextents + lcap - 1) / lcap; + idxs = ((inode->i_sb->s_blocksize - sizeof(struct ext4_extent_header)) + / sizeof(struct ext4_extent_idx)); /* - * Worse case, we need separate index block(s) - * to link all new leaf blocks + * If the new delayed allocation block is contiguous with the + * previous da block, it can share index blocks with the + * previous block, so we only need to allocate a new index + * block every idxs leaf blocks. At ldxs**2 blocks, we need + * an additional index block, and at ldxs**3 blocks, yet + * another index blocks. */ - idxs = (leafs + icap - 1) / icap; - do { - num += idxs; - idxs = (idxs + icap - 1) / icap; - } while (idxs > rcap); + if (ei->i_da_metadata_calc_len && + ei->i_da_metadata_calc_last_lblock+1 == lblock) { + if ((ei->i_da_metadata_calc_len % idxs) == 0) + num++; + if ((ei->i_da_metadata_calc_len % (idxs*idxs)) == 0) + num++; + if ((ei->i_da_metadata_calc_len % (idxs*idxs*idxs)) == 0) { + num++; + ei->i_da_metadata_calc_len = 0; + } else + ei->i_da_metadata_calc_len++; + ei->i_da_metadata_calc_last_lblock++; + return num; + } - return num; + /* + * In the worst case we need a new set of index blocks at + * every level of the inode's extent tree. + */ + ei->i_da_metadata_calc_len = 1; + ei->i_da_metadata_calc_last_lblock = lblock; + return ext_depth(inode) + 1; } static int @@ -3023,6 +3038,14 @@ out: return err; } +static void unmap_underlying_metadata_blocks(struct block_device *bdev, + sector_t block, int count) +{ + int i; + for (i = 0; i < count; i++) + unmap_underlying_metadata(bdev, block + i); +} + static int ext4_ext_handle_uninitialized_extents(handle_t *handle, struct inode *inode, ext4_lblk_t iblock, unsigned int max_blocks, @@ -3098,6 +3121,18 @@ out: } else allocated = ret; set_buffer_new(bh_result); + /* + * if we allocated more blocks than requested + * we need to make sure we unmap the extra block + * allocated. The actual needed block will get + * unmapped later when we find the buffer_head marked + * new. + */ + if (allocated > max_blocks) { + unmap_underlying_metadata_blocks(inode->i_sb->s_bdev, + newblock + max_blocks, + allocated - max_blocks); + } map_out: set_buffer_mapped(bh_result); out1: @@ -3190,7 +3225,13 @@ int ext4_ext_get_blocks(handle_t *handle, struct inode *inode, * this situation is possible, though, _during_ tree modification; * this is why assert can't be put in ext4_ext_find_extent() */ - BUG_ON(path[depth].p_ext == NULL && depth != 0); + if (path[depth].p_ext == NULL && depth != 0) { + ext4_error(inode->i_sb, __func__, "bad extent address " + "inode: %lu, iblock: %d, depth: %d", + inode->i_ino, iblock, depth); + err = -EIO; + goto out2; + } eh = path[depth].p_hdr; ex = path[depth].p_ext; diff --git a/fs/ext4/fsync.c b/fs/ext4/fsync.c index 0b22497d92e..98bd140aad0 100644 --- a/fs/ext4/fsync.c +++ b/fs/ext4/fsync.c @@ -88,9 +88,21 @@ int ext4_sync_file(struct file *file, struct dentry *dentry, int datasync) return ext4_force_commit(inode->i_sb); commit_tid = datasync ? ei->i_datasync_tid : ei->i_sync_tid; - if (jbd2_log_start_commit(journal, commit_tid)) + if (jbd2_log_start_commit(journal, commit_tid)) { + /* + * When the journal is on a different device than the + * fs data disk, we need to issue the barrier in + * writeback mode. (In ordered mode, the jbd2 layer + * will take care of issuing the barrier. In + * data=journal, all of the data blocks are written to + * the journal device.) + */ + if (ext4_should_writeback_data(inode) && + (journal->j_fs_dev != journal->j_dev) && + (journal->j_flags & JBD2_BARRIER)) + blkdev_issue_flush(inode->i_sb->s_bdev, NULL); jbd2_log_wait_commit(journal, commit_tid); - else if (journal->j_flags & JBD2_BARRIER) + } else if (journal->j_flags & JBD2_BARRIER) blkdev_issue_flush(inode->i_sb->s_bdev, NULL); return ret; } diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 5352db1a308..c818972c830 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -1003,83 +1003,94 @@ out: return err; } -qsize_t ext4_get_reserved_space(struct inode *inode) +#ifdef CONFIG_QUOTA +qsize_t *ext4_get_reserved_space(struct inode *inode) { - unsigned long long total; - - spin_lock(&EXT4_I(inode)->i_block_reservation_lock); - total = EXT4_I(inode)->i_reserved_data_blocks + - EXT4_I(inode)->i_reserved_meta_blocks; - spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); - - return (total << inode->i_blkbits); + return &EXT4_I(inode)->i_reserved_quota; } +#endif + /* * Calculate the number of metadata blocks need to reserve - * to allocate @blocks for non extent file based file + * to allocate a new block at @lblocks for non extent file based file */ -static int ext4_indirect_calc_metadata_amount(struct inode *inode, int blocks) +static int ext4_indirect_calc_metadata_amount(struct inode *inode, + sector_t lblock) { - int icap = EXT4_ADDR_PER_BLOCK(inode->i_sb); - int ind_blks, dind_blks, tind_blks; - - /* number of new indirect blocks needed */ - ind_blks = (blocks + icap - 1) / icap; + struct ext4_inode_info *ei = EXT4_I(inode); + int dind_mask = EXT4_ADDR_PER_BLOCK(inode->i_sb) - 1; + int blk_bits; - dind_blks = (ind_blks + icap - 1) / icap; + if (lblock < EXT4_NDIR_BLOCKS) + return 0; - tind_blks = 1; + lblock -= EXT4_NDIR_BLOCKS; - return ind_blks + dind_blks + tind_blks; + if (ei->i_da_metadata_calc_len && + (lblock & dind_mask) == ei->i_da_metadata_calc_last_lblock) { + ei->i_da_metadata_calc_len++; + return 0; + } + ei->i_da_metadata_calc_last_lblock = lblock & dind_mask; + ei->i_da_metadata_calc_len = 1; + blk_bits = roundup_pow_of_two(lblock + 1); + return (blk_bits / EXT4_ADDR_PER_BLOCK_BITS(inode->i_sb)) + 1; } /* * Calculate the number of metadata blocks need to reserve - * to allocate given number of blocks + * to allocate a block located at @lblock */ -static int ext4_calc_metadata_amount(struct inode *inode, int blocks) +static int ext4_calc_metadata_amount(struct inode *inode, sector_t lblock) { - if (!blocks) - return 0; - if (EXT4_I(inode)->i_flags & EXT4_EXTENTS_FL) - return ext4_ext_calc_metadata_amount(inode, blocks); + return ext4_ext_calc_metadata_amount(inode, lblock); - return ext4_indirect_calc_metadata_amount(inode, blocks); + return ext4_indirect_calc_metadata_amount(inode, lblock); } +/* + * Called with i_data_sem down, which is important since we can call + * ext4_discard_preallocations() from here. + */ static void ext4_da_update_reserve_space(struct inode *inode, int used) { struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); - int total, mdb, mdb_free; - - spin_lock(&EXT4_I(inode)->i_block_reservation_lock); - /* recalculate the number of metablocks still need to be reserved */ - total = EXT4_I(inode)->i_reserved_data_blocks - used; - mdb = ext4_calc_metadata_amount(inode, total); - - /* figure out how many metablocks to release */ - BUG_ON(mdb > EXT4_I(inode)->i_reserved_meta_blocks); - mdb_free = EXT4_I(inode)->i_reserved_meta_blocks - mdb; - - if (mdb_free) { - /* Account for allocated meta_blocks */ - mdb_free -= EXT4_I(inode)->i_allocated_meta_blocks; - - /* update fs dirty blocks counter */ + struct ext4_inode_info *ei = EXT4_I(inode); + int mdb_free = 0; + + spin_lock(&ei->i_block_reservation_lock); + if (unlikely(used > ei->i_reserved_data_blocks)) { + ext4_msg(inode->i_sb, KERN_NOTICE, "%s: ino %lu, used %d " + "with only %d reserved data blocks\n", + __func__, inode->i_ino, used, + ei->i_reserved_data_blocks); + WARN_ON(1); + used = ei->i_reserved_data_blocks; + } + + /* Update per-inode reservations */ + ei->i_reserved_data_blocks -= used; + used += ei->i_allocated_meta_blocks; + ei->i_reserved_meta_blocks -= ei->i_allocated_meta_blocks; + ei->i_allocated_meta_blocks = 0; + percpu_counter_sub(&sbi->s_dirtyblocks_counter, used); + + if (ei->i_reserved_data_blocks == 0) { + /* + * We can release all of the reserved metadata blocks + * only when we have written all of the delayed + * allocation blocks. + */ + mdb_free = ei->i_reserved_meta_blocks; + ei->i_reserved_meta_blocks = 0; + ei->i_da_metadata_calc_len = 0; percpu_counter_sub(&sbi->s_dirtyblocks_counter, mdb_free); - EXT4_I(inode)->i_allocated_meta_blocks = 0; - EXT4_I(inode)->i_reserved_meta_blocks = mdb; } - - /* update per-inode reservations */ - BUG_ON(used > EXT4_I(inode)->i_reserved_data_blocks); - EXT4_I(inode)->i_reserved_data_blocks -= used; spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); - /* - * free those over-booking quota for metadata blocks - */ + /* Update quota subsystem */ + vfs_dq_claim_block(inode, used); if (mdb_free) vfs_dq_release_reservation_block(inode, mdb_free); @@ -1088,7 +1099,8 @@ static void ext4_da_update_reserve_space(struct inode *inode, int used) * there aren't any writers on the inode, we can discard the * inode's preallocations. */ - if (!total && (atomic_read(&inode->i_writecount) == 0)) + if ((ei->i_reserved_data_blocks == 0) && + (atomic_read(&inode->i_writecount) == 0)) ext4_discard_preallocations(inode); } @@ -1797,11 +1809,15 @@ static int ext4_journalled_write_end(struct file *file, return ret ? ret : copied; } -static int ext4_da_reserve_space(struct inode *inode, int nrblocks) +/* + * Reserve a single block located at lblock + */ +static int ext4_da_reserve_space(struct inode *inode, sector_t lblock) { int retries = 0; struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); - unsigned long md_needed, mdblocks, total = 0; + struct ext4_inode_info *ei = EXT4_I(inode); + unsigned long md_needed, md_reserved; /* * recalculate the amount of metadata blocks to reserve @@ -1809,86 +1825,90 @@ static int ext4_da_reserve_space(struct inode *inode, int nrblocks) * worse case is one extent per block */ repeat: - spin_lock(&EXT4_I(inode)->i_block_reservation_lock); - total = EXT4_I(inode)->i_reserved_data_blocks + nrblocks; - mdblocks = ext4_calc_metadata_amount(inode, total); - BUG_ON(mdblocks < EXT4_I(inode)->i_reserved_meta_blocks); - - md_needed = mdblocks - EXT4_I(inode)->i_reserved_meta_blocks; - total = md_needed + nrblocks; + spin_lock(&ei->i_block_reservation_lock); + md_reserved = ei->i_reserved_meta_blocks; + md_needed = ext4_calc_metadata_amount(inode, lblock); + spin_unlock(&ei->i_block_reservation_lock); /* * Make quota reservation here to prevent quota overflow * later. Real quota accounting is done at pages writeout * time. */ - if (vfs_dq_reserve_block(inode, total)) { - spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); + if (vfs_dq_reserve_block(inode, md_needed + 1)) { + /* + * We tend to badly over-estimate the amount of + * metadata blocks which are needed, so if we have + * reserved any metadata blocks, try to force out the + * inode and see if we have any better luck. + */ + if (md_reserved && retries++ <= 3) + goto retry; return -EDQUOT; } - if (ext4_claim_free_blocks(sbi, total)) { - spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); - vfs_dq_release_reservation_block(inode, total); + if (ext4_claim_free_blocks(sbi, md_needed + 1)) { + vfs_dq_release_reservation_block(inode, md_needed + 1); if (ext4_should_retry_alloc(inode->i_sb, &retries)) { + retry: + if (md_reserved) + write_inode_now(inode, (retries == 3)); yield(); goto repeat; } return -ENOSPC; } - EXT4_I(inode)->i_reserved_data_blocks += nrblocks; - EXT4_I(inode)->i_reserved_meta_blocks = mdblocks; + spin_lock(&ei->i_block_reservation_lock); + ei->i_reserved_data_blocks++; + ei->i_reserved_meta_blocks += md_needed; + spin_unlock(&ei->i_block_reservation_lock); - spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); return 0; /* success */ } static void ext4_da_release_space(struct inode *inode, int to_free) { struct ext4_sb_info *sbi = EXT4_SB(inode->i_sb); - int total, mdb, mdb_free, release; + struct ext4_inode_info *ei = EXT4_I(inode); if (!to_free) return; /* Nothing to release, exit */ spin_lock(&EXT4_I(inode)->i_block_reservation_lock); - if (!EXT4_I(inode)->i_reserved_data_blocks) { + if (unlikely(to_free > ei->i_reserved_data_blocks)) { /* - * if there is no reserved blocks, but we try to free some - * then the counter is messed up somewhere. - * but since this function is called from invalidate - * page, it's harmless to return without any action + * if there aren't enough reserved blocks, then the + * counter is messed up somewhere. Since this + * function is called from invalidate page, it's + * harmless to return without any action. */ - printk(KERN_INFO "ext4 delalloc try to release %d reserved " - "blocks for inode %lu, but there is no reserved " - "data blocks\n", to_free, inode->i_ino); - spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); - return; + ext4_msg(inode->i_sb, KERN_NOTICE, "ext4_da_release_space: " + "ino %lu, to_free %d with only %d reserved " + "data blocks\n", inode->i_ino, to_free, + ei->i_reserved_data_blocks); + WARN_ON(1); + to_free = ei->i_reserved_data_blocks; } + ei->i_reserved_data_blocks -= to_free; - /* recalculate the number of metablocks still need to be reserved */ - total = EXT4_I(inode)->i_reserved_data_blocks - to_free; - mdb = ext4_calc_metadata_amount(inode, total); - - /* figure out how many metablocks to release */ - BUG_ON(mdb > EXT4_I(inode)->i_reserved_meta_blocks); - mdb_free = EXT4_I(inode)->i_reserved_meta_blocks - mdb; - - release = to_free + mdb_free; - - /* update fs dirty blocks counter for truncate case */ - percpu_counter_sub(&sbi->s_dirtyblocks_counter, release); + if (ei->i_reserved_data_blocks == 0) { + /* + * We can release all of the reserved metadata blocks + * only when we have written all of the delayed + * allocation blocks. + */ + to_free += ei->i_reserved_meta_blocks; + ei->i_reserved_meta_blocks = 0; + ei->i_da_metadata_calc_len = 0; + } - /* update per-inode reservations */ - BUG_ON(to_free > EXT4_I(inode)->i_reserved_data_blocks); - EXT4_I(inode)->i_reserved_data_blocks -= to_free; + /* update fs dirty blocks counter */ + percpu_counter_sub(&sbi->s_dirtyblocks_counter, to_free); - BUG_ON(mdb > EXT4_I(inode)->i_reserved_meta_blocks); - EXT4_I(inode)->i_reserved_meta_blocks = mdb; spin_unlock(&EXT4_I(inode)->i_block_reservation_lock); - vfs_dq_release_reservation_block(inode, release); + vfs_dq_release_reservation_block(inode, to_free); } static void ext4_da_page_release_reservation(struct page *page, @@ -2494,7 +2514,7 @@ static int ext4_da_get_block_prep(struct inode *inode, sector_t iblock, * XXX: __block_prepare_write() unmaps passed block, * is it OK? */ - ret = ext4_da_reserve_space(inode, 1); + ret = ext4_da_reserve_space(inode, iblock); if (ret) /* not enough space to reserve */ return ret; @@ -2968,8 +2988,7 @@ retry: out_writepages: if (!no_nrwrite_index_update) wbc->no_nrwrite_index_update = 0; - if (wbc->nr_to_write > nr_to_writebump) - wbc->nr_to_write -= nr_to_writebump; + wbc->nr_to_write -= nr_to_writebump; wbc->range_start = range_start; trace_ext4_da_writepages_result(inode, wbc, ret, pages_written); return ret; @@ -2994,11 +3013,18 @@ static int ext4_nonda_switch(struct super_block *sb) if (2 * free_blocks < 3 * dirty_blocks || free_blocks < (dirty_blocks + EXT4_FREEBLOCKS_WATERMARK)) { /* - * free block count is less that 150% of dirty blocks - * or free blocks is less that watermark + * free block count is less than 150% of dirty blocks + * or free blocks is less than watermark */ return 1; } + /* + * Even if we don't switch but are nearing capacity, + * start pushing delalloc when 1/2 of free blocks are dirty. + */ + if (free_blocks < 2 * dirty_blocks) + writeback_inodes_sb_if_idle(sb); + return 0; } @@ -4794,6 +4820,9 @@ struct inode *ext4_iget(struct super_block *sb, unsigned long ino) ((__u64)le16_to_cpu(raw_inode->i_file_acl_high)) << 32; inode->i_size = ext4_isize(raw_inode); ei->i_disksize = inode->i_size; +#ifdef CONFIG_QUOTA + ei->i_reserved_quota = 0; +#endif inode->i_generation = le32_to_cpu(raw_inode->i_generation); ei->i_block_group = iloc.block_group; ei->i_last_alloc_group = ~0; diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index b1fd3daadc9..d34afad3e13 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -2755,12 +2755,6 @@ ext4_mb_mark_diskspace_used(struct ext4_allocation_context *ac, if (!(ac->ac_flags & EXT4_MB_DELALLOC_RESERVED)) /* release all the reserved blocks if non delalloc */ percpu_counter_sub(&sbi->s_dirtyblocks_counter, reserv_blks); - else { - percpu_counter_sub(&sbi->s_dirtyblocks_counter, - ac->ac_b_ex.fe_len); - /* convert reserved quota blocks to real quota blocks */ - vfs_dq_claim_block(ac->ac_inode, ac->ac_b_ex.fe_len); - } if (sbi->s_log_groups_per_flex) { ext4_group_t flex_group = ext4_flex_group(sbi, diff --git a/fs/ext4/mballoc.h b/fs/ext4/mballoc.h index 0ca811061bc..436521cae45 100644 --- a/fs/ext4/mballoc.h +++ b/fs/ext4/mballoc.h @@ -17,7 +17,6 @@ #include <linux/proc_fs.h> #include <linux/pagemap.h> #include <linux/seq_file.h> -#include <linux/version.h> #include <linux/blkdev.h> #include <linux/mutex.h> #include "ext4_jbd2.h" diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 827bde1f259..735c20d5fd5 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -702,8 +702,12 @@ static struct inode *ext4_alloc_inode(struct super_block *sb) ei->i_reserved_data_blocks = 0; ei->i_reserved_meta_blocks = 0; ei->i_allocated_meta_blocks = 0; + ei->i_da_metadata_calc_len = 0; ei->i_delalloc_reserved_flag = 0; spin_lock_init(&(ei->i_block_reservation_lock)); +#ifdef CONFIG_QUOTA + ei->i_reserved_quota = 0; +#endif INIT_LIST_HEAD(&ei->i_aio_dio_complete_list); ei->cur_aio_dio = NULL; ei->i_sync_tid = 0; @@ -1014,7 +1018,9 @@ static const struct dquot_operations ext4_quota_operations = { .reserve_space = dquot_reserve_space, .claim_space = dquot_claim_space, .release_rsv = dquot_release_reserved_space, +#ifdef CONFIG_QUOTA .get_reserved_space = ext4_get_reserved_space, +#endif .alloc_inode = dquot_alloc_inode, .free_space = dquot_free_space, .free_inode = dquot_free_inode, @@ -2169,9 +2175,9 @@ static ssize_t lifetime_write_kbytes_show(struct ext4_attr *a, struct super_block *sb = sbi->s_buddy_cache->i_sb; return snprintf(buf, PAGE_SIZE, "%llu\n", - sbi->s_kbytes_written + + (unsigned long long)(sbi->s_kbytes_written + ((part_stat_read(sb->s_bdev->bd_part, sectors[1]) - - EXT4_SB(sb)->s_sectors_written_start) >> 1)); + EXT4_SB(sb)->s_sectors_written_start) >> 1))); } static ssize_t inode_readahead_blks_store(struct ext4_attr *a, @@ -4000,6 +4006,7 @@ static inline void unregister_as_ext2(void) { unregister_filesystem(&ext2_fs_type); } +MODULE_ALIAS("ext2"); #else static inline void register_as_ext2(void) { } static inline void unregister_as_ext2(void) { } @@ -4026,6 +4033,7 @@ static inline void unregister_as_ext3(void) { unregister_filesystem(&ext3_fs_type); } +MODULE_ALIAS("ext3"); #else static inline void register_as_ext3(void) { } static inline void unregister_as_ext3(void) { } diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c index 83218bebbc7..f3a2f7ed45a 100644 --- a/fs/ext4/xattr.c +++ b/fs/ext4/xattr.c @@ -1332,6 +1332,8 @@ retry: goto cleanup; kfree(b_entry_name); kfree(buffer); + b_entry_name = NULL; + buffer = NULL; brelse(is->iloc.bh); kfree(is); kfree(bs); diff --git a/fs/file_table.c b/fs/file_table.c index 0afacf65439..69652c5bd5f 100644 --- a/fs/file_table.c +++ b/fs/file_table.c @@ -186,10 +186,8 @@ struct file *alloc_file(struct path *path, fmode_t mode, * that we can do debugging checks at __fput() */ if ((mode & FMODE_WRITE) && !special_file(path->dentry->d_inode->i_mode)) { - int error = 0; file_take_write(file); - error = mnt_clone_write(path->mnt); - WARN_ON(error); + WARN_ON(mnt_clone_write(path->mnt)); } ima_counts_get(file); return file; diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 49bc1b8e8f1..1a7c42c64ff 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -242,6 +242,7 @@ static void bdi_sync_writeback(struct backing_dev_info *bdi, /** * bdi_start_writeback - start writeback * @bdi: the backing device to write from + * @sb: write inodes from this super_block * @nr_pages: the number of pages to write * * Description: @@ -1187,6 +1188,23 @@ void writeback_inodes_sb(struct super_block *sb) EXPORT_SYMBOL(writeback_inodes_sb); /** + * writeback_inodes_sb_if_idle - start writeback if none underway + * @sb: the superblock + * + * Invoke writeback_inodes_sb if no writeback is currently underway. + * Returns 1 if writeback was started, 0 if not. + */ +int writeback_inodes_sb_if_idle(struct super_block *sb) +{ + if (!writeback_in_progress(sb->s_bdi)) { + writeback_inodes_sb(sb); + return 1; + } else + return 0; +} +EXPORT_SYMBOL(writeback_inodes_sb_if_idle); + +/** * sync_inodes_sb - sync sb inode pages * @sb: the superblock * diff --git a/fs/internal.h b/fs/internal.h index f67cd141d9a..e96a1667d74 100644 --- a/fs/internal.h +++ b/fs/internal.h @@ -85,3 +85,10 @@ extern struct file *get_empty_filp(void); * super.c */ extern int do_remount_sb(struct super_block *, int, void *, int); + +/* + * open.c + */ +struct nameidata; +extern struct file *nameidata_to_filp(struct nameidata *); +extern void release_open_intent(struct nameidata *); diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c index 4160afad6d0..bd224eec9b0 100644 --- a/fs/jbd/journal.c +++ b/fs/jbd/journal.c @@ -1913,7 +1913,7 @@ static void __init jbd_create_debugfs_entry(void) { jbd_debugfs_dir = debugfs_create_dir("jbd", NULL); if (jbd_debugfs_dir) - jbd_debug = debugfs_create_u8("jbd-debug", S_IRUGO, + jbd_debug = debugfs_create_u8("jbd-debug", S_IRUGO | S_IWUSR, jbd_debugfs_dir, &journal_enable_debug); } diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c index ca0f5eb62b2..88684937095 100644 --- a/fs/jbd2/checkpoint.c +++ b/fs/jbd2/checkpoint.c @@ -22,6 +22,7 @@ #include <linux/jbd2.h> #include <linux/errno.h> #include <linux/slab.h> +#include <linux/blkdev.h> #include <trace/events/jbd2.h> /* @@ -515,6 +516,20 @@ int jbd2_cleanup_journal_tail(journal_t *journal) journal->j_tail_sequence = first_tid; journal->j_tail = blocknr; spin_unlock(&journal->j_state_lock); + + /* + * If there is an external journal, we need to make sure that + * any data blocks that were recently written out --- perhaps + * by jbd2_log_do_checkpoint() --- are flushed out before we + * drop the transactions from the external journal. It's + * unlikely this will be necessary, especially with a + * appropriately sized journal, but we need this to guarantee + * correctness. Fortunately jbd2_cleanup_journal_tail() + * doesn't get called all that often. + */ + if ((journal->j_fs_dev != journal->j_dev) && + (journal->j_flags & JBD2_BARRIER)) + blkdev_issue_flush(journal->j_fs_dev, NULL); if (!(journal->j_flags & JBD2_ABORT)) jbd2_journal_update_superblock(journal, 1); return 0; diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index 6a10238d2c6..1bc74b6f26d 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -259,6 +259,7 @@ static int journal_submit_data_buffers(journal_t *journal, ret = err; spin_lock(&journal->j_list_lock); J_ASSERT(jinode->i_transaction == commit_transaction); + commit_transaction->t_flushed_data_blocks = 1; jinode->i_flags &= ~JI_COMMIT_RUNNING; wake_up_bit(&jinode->i_flags, __JI_COMMIT_RUNNING); } @@ -708,8 +709,17 @@ start_journal_io: } } - /* Done it all: now write the commit record asynchronously. */ + /* + * If the journal is not located on the file system device, + * then we must flush the file system device before we issue + * the commit record + */ + if (commit_transaction->t_flushed_data_blocks && + (journal->j_fs_dev != journal->j_dev) && + (journal->j_flags & JBD2_BARRIER)) + blkdev_issue_flush(journal->j_fs_dev, NULL); + /* Done it all: now write the commit record asynchronously. */ if (JBD2_HAS_INCOMPAT_FEATURE(journal, JBD2_FEATURE_INCOMPAT_ASYNC_COMMIT)) { err = journal_submit_commit_record(journal, commit_transaction, @@ -720,13 +730,6 @@ start_journal_io: blkdev_issue_flush(journal->j_dev, NULL); } - /* - * This is the right place to wait for data buffers both for ASYNC - * and !ASYNC commit. If commit is ASYNC, we need to wait only after - * the commit block went to disk (which happens above). If commit is - * SYNC, we need to wait for data buffers before we start writing - * commit block, which happens below in such setting. - */ err = journal_finish_inode_data_buffers(journal, commit_transaction); if (err) { printk(KERN_WARNING diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index b7ca3a92a4d..ac0d027595d 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -814,7 +814,7 @@ static journal_t * journal_init_common (void) journal_t *journal; int err; - journal = kzalloc(sizeof(*journal), GFP_KERNEL|__GFP_NOFAIL); + journal = kzalloc(sizeof(*journal), GFP_KERNEL); if (!journal) goto fail; @@ -2115,7 +2115,8 @@ static void __init jbd2_create_debugfs_entry(void) { jbd2_debugfs_dir = debugfs_create_dir("jbd2", NULL); if (jbd2_debugfs_dir) - jbd2_debug = debugfs_create_u8(JBD2_DEBUG_NAME, S_IRUGO, + jbd2_debug = debugfs_create_u8(JBD2_DEBUG_NAME, + S_IRUGO | S_IWUSR, jbd2_debugfs_dir, &jbd2_journal_enable_debug); } diff --git a/fs/jfs/super.c b/fs/jfs/super.c index 2234c73fc57..d929a822a74 100644 --- a/fs/jfs/super.c +++ b/fs/jfs/super.c @@ -524,7 +524,7 @@ static int jfs_fill_super(struct super_block *sb, void *data, int silent) * Page cache is indexed by long. * I would use MAX_LFS_FILESIZE, but it's only half as big */ - sb->s_maxbytes = min(((u64) PAGE_CACHE_SIZE << 32) - 1, sb->s_maxbytes); + sb->s_maxbytes = min(((u64) PAGE_CACHE_SIZE << 32) - 1, (u64)sb->s_maxbytes); #endif sb->s_time_gran = 1; return 0; diff --git a/fs/namei.c b/fs/namei.c index dad4b80257d..b55440baf7a 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -37,8 +37,6 @@ #include "internal.h" -#define ACC_MODE(x) ("\000\004\002\006"[(x)&O_ACCMODE]) - /* [Feb-1997 T. Schoebel-Theuer] * Fundamental changes in the pathname lookup mechanisms (namei) * were necessary because of omirr. The reason is that omirr needs @@ -234,6 +232,7 @@ int generic_permission(struct inode *inode, int mask, /* * Searching includes executable on directories, else just read. */ + mask &= MAY_READ | MAY_WRITE | MAY_EXEC; if (mask == MAY_READ || (S_ISDIR(inode->i_mode) && !(mask & MAY_WRITE))) if (capable(CAP_DAC_READ_SEARCH)) return 0; @@ -1640,6 +1639,7 @@ struct file *do_filp_open(int dfd, const char *pathname, if (filp == NULL) return ERR_PTR(-ENFILE); nd.intent.open.file = filp; + filp->f_flags = open_flag; nd.intent.open.flags = flag; nd.intent.open.create_mode = 0; error = do_path_lookup(dfd, pathname, @@ -1685,6 +1685,7 @@ struct file *do_filp_open(int dfd, const char *pathname, if (filp == NULL) goto exit_parent; nd.intent.open.file = filp; + filp->f_flags = open_flag; nd.intent.open.flags = flag; nd.intent.open.create_mode = mode; dir = nd.path.dentry; @@ -1725,7 +1726,7 @@ do_last: mnt_drop_write(nd.path.mnt); goto exit; } - filp = nameidata_to_filp(&nd, open_flag); + filp = nameidata_to_filp(&nd); mnt_drop_write(nd.path.mnt); if (nd.root.mnt) path_put(&nd.root); @@ -1789,7 +1790,7 @@ ok: mnt_drop_write(nd.path.mnt); goto exit; } - filp = nameidata_to_filp(&nd, open_flag); + filp = nameidata_to_filp(&nd); if (!IS_ERR(filp)) { error = ima_path_check(&filp->f_path, filp->f_mode & (MAY_READ | MAY_WRITE | MAY_EXEC)); diff --git a/fs/nfsd/nfsfh.c b/fs/nfsd/nfsfh.c index 1c12177b908..55c8e63af0b 100644 --- a/fs/nfsd/nfsfh.c +++ b/fs/nfsd/nfsfh.c @@ -89,7 +89,7 @@ static __be32 nfsd_setuser_and_check_port(struct svc_rqst *rqstp, int flags = nfsexp_flags(rqstp, exp); /* Check if the request originated from a secure port. */ - if (!rqstp->rq_secure && (flags & NFSEXP_INSECURE_PORT)) { + if (!rqstp->rq_secure && !(flags & NFSEXP_INSECURE_PORT)) { RPC_IFDEBUG(char buf[RPC_MAX_ADDRBUFLEN]); dprintk(KERN_WARNING "nfsd: request from insecure port %s!\n", diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 7c2e337d05a..c194793b642 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -780,12 +780,9 @@ static inline int nfsd_dosync(struct file *filp, struct dentry *dp, int (*fsync) (struct file *, struct dentry *, int); int err; - err = filemap_fdatawrite(inode->i_mapping); + err = filemap_write_and_wait(inode->i_mapping); if (err == 0 && fop && (fsync = fop->fsync)) err = fsync(filp, dp, 0); - if (err == 0) - err = filemap_fdatawait(inode->i_mapping); - return err; } diff --git a/fs/nilfs2/bmap.c b/fs/nilfs2/bmap.c index f4a14ea2ed9..effdbdbe6c1 100644 --- a/fs/nilfs2/bmap.c +++ b/fs/nilfs2/bmap.c @@ -417,8 +417,8 @@ __u64 nilfs_bmap_data_get_key(const struct nilfs_bmap *bmap, key = page_index(bh->b_page) << (PAGE_CACHE_SHIFT - bmap->b_inode->i_blkbits); - for (pbh = page_buffers(bh->b_page); pbh != bh; - pbh = pbh->b_this_page, key++); + for (pbh = page_buffers(bh->b_page); pbh != bh; pbh = pbh->b_this_page) + key++; return key; } diff --git a/fs/nilfs2/cpfile.c b/fs/nilfs2/cpfile.c index d5ad54e204a..18737818db6 100644 --- a/fs/nilfs2/cpfile.c +++ b/fs/nilfs2/cpfile.c @@ -328,19 +328,24 @@ int nilfs_cpfile_delete_checkpoints(struct inode *cpfile, tnicps += nicps; nilfs_mdt_mark_buffer_dirty(cp_bh); nilfs_mdt_mark_dirty(cpfile); - if (!nilfs_cpfile_is_in_first(cpfile, cno) && - (count = nilfs_cpfile_block_sub_valid_checkpoints( - cpfile, cp_bh, kaddr, nicps)) == 0) { - /* make hole */ - kunmap_atomic(kaddr, KM_USER0); - brelse(cp_bh); - ret = nilfs_cpfile_delete_checkpoint_block( - cpfile, cno); - if (ret == 0) - continue; - printk(KERN_ERR "%s: cannot delete block\n", - __func__); - break; + if (!nilfs_cpfile_is_in_first(cpfile, cno)) { + count = + nilfs_cpfile_block_sub_valid_checkpoints( + cpfile, cp_bh, kaddr, nicps); + if (count == 0) { + /* make hole */ + kunmap_atomic(kaddr, KM_USER0); + brelse(cp_bh); + ret = + nilfs_cpfile_delete_checkpoint_block( + cpfile, cno); + if (ret == 0) + continue; + printk(KERN_ERR + "%s: cannot delete block\n", + __func__); + break; + } } } diff --git a/fs/nilfs2/direct.c b/fs/nilfs2/direct.c index d369ac71827..236753df5cd 100644 --- a/fs/nilfs2/direct.c +++ b/fs/nilfs2/direct.c @@ -51,11 +51,11 @@ static int nilfs_direct_lookup(const struct nilfs_bmap *bmap, struct nilfs_direct *direct; __u64 ptr; - direct = (struct nilfs_direct *)bmap; - if ((key > NILFS_DIRECT_KEY_MAX) || - (level != 1) || /* XXX: use macro for level 1 */ - ((ptr = nilfs_direct_get_ptr(direct, key)) == - NILFS_BMAP_INVALID_PTR)) + direct = (struct nilfs_direct *)bmap; /* XXX: use macro for level 1 */ + if (key > NILFS_DIRECT_KEY_MAX || level != 1) + return -ENOENT; + ptr = nilfs_direct_get_ptr(direct, key); + if (ptr == NILFS_BMAP_INVALID_PTR) return -ENOENT; if (ptrp != NULL) @@ -73,9 +73,10 @@ static int nilfs_direct_lookup_contig(const struct nilfs_bmap *bmap, sector_t blocknr; int ret, cnt; - if (key > NILFS_DIRECT_KEY_MAX || - (ptr = nilfs_direct_get_ptr(direct, key)) == - NILFS_BMAP_INVALID_PTR) + if (key > NILFS_DIRECT_KEY_MAX) + return -ENOENT; + ptr = nilfs_direct_get_ptr(direct, key); + if (ptr == NILFS_BMAP_INVALID_PTR) return -ENOENT; if (NILFS_BMAP_USE_VBN(bmap)) { diff --git a/fs/nilfs2/ioctl.c b/fs/nilfs2/ioctl.c index f6af76042d8..d6b2b83de36 100644 --- a/fs/nilfs2/ioctl.c +++ b/fs/nilfs2/ioctl.c @@ -480,7 +480,7 @@ static int nilfs_ioctl_clean_segments(struct inode *inode, struct file *filp, unsigned int cmd, void __user *argp) { struct nilfs_argv argv[5]; - const static size_t argsz[5] = { + static const size_t argsz[5] = { sizeof(struct nilfs_vdesc), sizeof(struct nilfs_period), sizeof(__u64), diff --git a/fs/ocfs2/Kconfig b/fs/ocfs2/Kconfig index 701b7a3a872..0d840669698 100644 --- a/fs/ocfs2/Kconfig +++ b/fs/ocfs2/Kconfig @@ -6,6 +6,7 @@ config OCFS2_FS select CRC32 select QUOTA select QUOTA_TREE + select FS_POSIX_ACL help OCFS2 is a general purpose extent based shared disk cluster file system with many similarities to ext3. It supports 64 bit inode @@ -74,12 +75,3 @@ config OCFS2_DEBUG_FS This option will enable expensive consistency checks. Enable this option for debugging only as it is likely to decrease performance of the filesystem. - -config OCFS2_FS_POSIX_ACL - bool "OCFS2 POSIX Access Control Lists" - depends on OCFS2_FS - select FS_POSIX_ACL - default n - help - Posix Access Control Lists (ACLs) support permissions for users and - groups beyond the owner/group/world scheme. diff --git a/fs/ocfs2/Makefile b/fs/ocfs2/Makefile index 31f25ce32c9..600d2d2ade1 100644 --- a/fs/ocfs2/Makefile +++ b/fs/ocfs2/Makefile @@ -39,11 +39,8 @@ ocfs2-objs := \ ver.o \ quota_local.o \ quota_global.o \ - xattr.o - -ifeq ($(CONFIG_OCFS2_FS_POSIX_ACL),y) -ocfs2-objs += acl.o -endif + xattr.o \ + acl.o ocfs2_stackglue-objs := stackglue.o ocfs2_stack_o2cb-objs := stack_o2cb.o diff --git a/fs/ocfs2/acl.c b/fs/ocfs2/acl.c index e3e47415d85..0501974bedd 100644 --- a/fs/ocfs2/acl.c +++ b/fs/ocfs2/acl.c @@ -98,15 +98,11 @@ static struct posix_acl *ocfs2_get_acl_nolock(struct inode *inode, int type, struct buffer_head *di_bh) { - struct ocfs2_super *osb = OCFS2_SB(inode->i_sb); int name_index; char *value = NULL; struct posix_acl *acl; int retval; - if (!(osb->s_mount_opt & OCFS2_MOUNT_POSIX_ACL)) - return NULL; - switch (type) { case ACL_TYPE_ACCESS: name_index = OCFS2_XATTR_INDEX_POSIX_ACL_ACCESS; diff --git a/fs/ocfs2/acl.h b/fs/ocfs2/acl.h index 8f6389ed4da..5c5d31f0585 100644 --- a/fs/ocfs2/acl.h +++ b/fs/ocfs2/acl.h @@ -26,8 +26,6 @@ struct ocfs2_acl_entry { __le32 e_id; }; -#ifdef CONFIG_OCFS2_FS_POSIX_ACL - extern int ocfs2_check_acl(struct inode *, int); extern int ocfs2_acl_chmod(struct inode *); extern int ocfs2_init_acl(handle_t *, struct inode *, struct inode *, @@ -35,24 +33,4 @@ extern int ocfs2_init_acl(handle_t *, struct inode *, struct inode *, struct ocfs2_alloc_context *, struct ocfs2_alloc_context *); -#else /* CONFIG_OCFS2_FS_POSIX_ACL*/ - -#define ocfs2_check_acl NULL -static inline int ocfs2_acl_chmod(struct inode *inode) -{ - return 0; -} -static inline int ocfs2_init_acl(handle_t *handle, - struct inode *inode, - struct inode *dir, - struct buffer_head *di_bh, - struct buffer_head *dir_bh, - struct ocfs2_alloc_context *meta_ac, - struct ocfs2_alloc_context *data_ac) -{ - return 0; -} - -#endif /* CONFIG_OCFS2_FS_POSIX_ACL*/ - #endif /* OCFS2_ACL_H */ diff --git a/fs/ocfs2/alloc.c b/fs/ocfs2/alloc.c index fb4e672579b..d17bdc718f7 100644 --- a/fs/ocfs2/alloc.c +++ b/fs/ocfs2/alloc.c @@ -1765,9 +1765,9 @@ set_and_inc: * * The array index of the subtree root is passed back. */ -static int ocfs2_find_subtree_root(struct ocfs2_extent_tree *et, - struct ocfs2_path *left, - struct ocfs2_path *right) +int ocfs2_find_subtree_root(struct ocfs2_extent_tree *et, + struct ocfs2_path *left, + struct ocfs2_path *right) { int i = 0; @@ -2872,8 +2872,8 @@ out: * This looks similar, but is subtly different to * ocfs2_find_cpos_for_left_leaf(). */ -static int ocfs2_find_cpos_for_right_leaf(struct super_block *sb, - struct ocfs2_path *path, u32 *cpos) +int ocfs2_find_cpos_for_right_leaf(struct super_block *sb, + struct ocfs2_path *path, u32 *cpos) { int i, j, ret = 0; u64 blkno; diff --git a/fs/ocfs2/alloc.h b/fs/ocfs2/alloc.h index 9c122d57446..1db4359ccb9 100644 --- a/fs/ocfs2/alloc.h +++ b/fs/ocfs2/alloc.h @@ -317,4 +317,9 @@ int ocfs2_path_bh_journal_access(handle_t *handle, int ocfs2_journal_access_path(struct ocfs2_caching_info *ci, handle_t *handle, struct ocfs2_path *path); +int ocfs2_find_cpos_for_right_leaf(struct super_block *sb, + struct ocfs2_path *path, u32 *cpos); +int ocfs2_find_subtree_root(struct ocfs2_extent_tree *et, + struct ocfs2_path *left, + struct ocfs2_path *right); #endif /* OCFS2_ALLOC_H */ diff --git a/fs/ocfs2/cluster/heartbeat.c b/fs/ocfs2/cluster/heartbeat.c index c452d116b89..eda5b8bcddd 100644 --- a/fs/ocfs2/cluster/heartbeat.c +++ b/fs/ocfs2/cluster/heartbeat.c @@ -176,7 +176,8 @@ static void o2hb_write_timeout(struct work_struct *work) static void o2hb_arm_write_timeout(struct o2hb_region *reg) { - mlog(0, "Queue write timeout for %u ms\n", O2HB_MAX_WRITE_TIMEOUT_MS); + mlog(ML_HEARTBEAT, "Queue write timeout for %u ms\n", + O2HB_MAX_WRITE_TIMEOUT_MS); cancel_delayed_work(®->hr_write_timeout_work); reg->hr_last_timeout_start = jiffies; @@ -874,7 +875,8 @@ static int o2hb_thread(void *data) do_gettimeofday(&after_hb); elapsed_msec = o2hb_elapsed_msecs(&before_hb, &after_hb); - mlog(0, "start = %lu.%lu, end = %lu.%lu, msec = %u\n", + mlog(ML_HEARTBEAT, + "start = %lu.%lu, end = %lu.%lu, msec = %u\n", before_hb.tv_sec, (unsigned long) before_hb.tv_usec, after_hb.tv_sec, (unsigned long) after_hb.tv_usec, elapsed_msec); diff --git a/fs/ocfs2/cluster/nodemanager.c b/fs/ocfs2/cluster/nodemanager.c index 7ee6188bc79..c81142e3ef8 100644 --- a/fs/ocfs2/cluster/nodemanager.c +++ b/fs/ocfs2/cluster/nodemanager.c @@ -35,6 +35,10 @@ * cluster references throughout where nodes are looked up */ struct o2nm_cluster *o2nm_single_cluster = NULL; +char *o2nm_fence_method_desc[O2NM_FENCE_METHODS] = { + "reset", /* O2NM_FENCE_RESET */ + "panic", /* O2NM_FENCE_PANIC */ +}; struct o2nm_node *o2nm_get_node_by_num(u8 node_num) { @@ -579,6 +583,43 @@ static ssize_t o2nm_cluster_attr_reconnect_delay_ms_write( return o2nm_cluster_attr_write(page, count, &cluster->cl_reconnect_delay_ms); } + +static ssize_t o2nm_cluster_attr_fence_method_read( + struct o2nm_cluster *cluster, char *page) +{ + ssize_t ret = 0; + + if (cluster) + ret = sprintf(page, "%s\n", + o2nm_fence_method_desc[cluster->cl_fence_method]); + return ret; +} + +static ssize_t o2nm_cluster_attr_fence_method_write( + struct o2nm_cluster *cluster, const char *page, size_t count) +{ + unsigned int i; + + if (page[count - 1] != '\n') + goto bail; + + for (i = 0; i < O2NM_FENCE_METHODS; ++i) { + if (count != strlen(o2nm_fence_method_desc[i]) + 1) + continue; + if (strncasecmp(page, o2nm_fence_method_desc[i], count - 1)) + continue; + if (cluster->cl_fence_method != i) { + printk(KERN_INFO "ocfs2: Changing fence method to %s\n", + o2nm_fence_method_desc[i]); + cluster->cl_fence_method = i; + } + return count; + } + +bail: + return -EINVAL; +} + static struct o2nm_cluster_attribute o2nm_cluster_attr_idle_timeout_ms = { .attr = { .ca_owner = THIS_MODULE, .ca_name = "idle_timeout_ms", @@ -603,10 +644,19 @@ static struct o2nm_cluster_attribute o2nm_cluster_attr_reconnect_delay_ms = { .store = o2nm_cluster_attr_reconnect_delay_ms_write, }; +static struct o2nm_cluster_attribute o2nm_cluster_attr_fence_method = { + .attr = { .ca_owner = THIS_MODULE, + .ca_name = "fence_method", + .ca_mode = S_IRUGO | S_IWUSR }, + .show = o2nm_cluster_attr_fence_method_read, + .store = o2nm_cluster_attr_fence_method_write, +}; + static struct configfs_attribute *o2nm_cluster_attrs[] = { &o2nm_cluster_attr_idle_timeout_ms.attr, &o2nm_cluster_attr_keepalive_delay_ms.attr, &o2nm_cluster_attr_reconnect_delay_ms.attr, + &o2nm_cluster_attr_fence_method.attr, NULL, }; static ssize_t o2nm_cluster_show(struct config_item *item, @@ -778,6 +828,7 @@ static struct config_group *o2nm_cluster_group_make_group(struct config_group *g cluster->cl_reconnect_delay_ms = O2NET_RECONNECT_DELAY_MS_DEFAULT; cluster->cl_idle_timeout_ms = O2NET_IDLE_TIMEOUT_MS_DEFAULT; cluster->cl_keepalive_delay_ms = O2NET_KEEPALIVE_DELAY_MS_DEFAULT; + cluster->cl_fence_method = O2NM_FENCE_RESET; ret = &cluster->cl_group; o2nm_single_cluster = cluster; diff --git a/fs/ocfs2/cluster/nodemanager.h b/fs/ocfs2/cluster/nodemanager.h index c992ea0da4a..09ea2d388bb 100644 --- a/fs/ocfs2/cluster/nodemanager.h +++ b/fs/ocfs2/cluster/nodemanager.h @@ -33,6 +33,12 @@ #include <linux/configfs.h> #include <linux/rbtree.h> +enum o2nm_fence_method { + O2NM_FENCE_RESET = 0, + O2NM_FENCE_PANIC, + O2NM_FENCE_METHODS, /* Number of fence methods */ +}; + struct o2nm_node { spinlock_t nd_lock; struct config_item nd_item; @@ -58,6 +64,7 @@ struct o2nm_cluster { unsigned int cl_idle_timeout_ms; unsigned int cl_keepalive_delay_ms; unsigned int cl_reconnect_delay_ms; + enum o2nm_fence_method cl_fence_method; /* this bitmap is part of a hack for disk bitmap.. will go eventually. - zab */ unsigned long cl_nodes_bitmap[BITS_TO_LONGS(O2NM_MAX_NODES)]; diff --git a/fs/ocfs2/cluster/quorum.c b/fs/ocfs2/cluster/quorum.c index bbacf7da48a..639024033fc 100644 --- a/fs/ocfs2/cluster/quorum.c +++ b/fs/ocfs2/cluster/quorum.c @@ -74,8 +74,20 @@ static void o2quo_fence_self(void) * threads can still schedule, etc, etc */ o2hb_stop_all_regions(); - printk("ocfs2 is very sorry to be fencing this system by restarting\n"); - emergency_restart(); + switch (o2nm_single_cluster->cl_fence_method) { + case O2NM_FENCE_PANIC: + panic("*** ocfs2 is very sorry to be fencing this system by " + "panicing ***\n"); + break; + default: + WARN_ON(o2nm_single_cluster->cl_fence_method >= + O2NM_FENCE_METHODS); + case O2NM_FENCE_RESET: + printk(KERN_ERR "*** ocfs2 is very sorry to be fencing this " + "system by restarting ***\n"); + emergency_restart(); + break; + }; } /* Indicate that a timeout occured on a hearbeat region write. The diff --git a/fs/ocfs2/dlm/dlmrecovery.c b/fs/ocfs2/dlm/dlmrecovery.c index d9fa3d22e17..2f9e4e19a4f 100644 --- a/fs/ocfs2/dlm/dlmrecovery.c +++ b/fs/ocfs2/dlm/dlmrecovery.c @@ -2589,6 +2589,14 @@ retry: "begin reco msg (%d)\n", dlm->name, nodenum, ret); ret = 0; } + if (ret == -EAGAIN) { + mlog(0, "%s: trying to start recovery of node " + "%u, but node %u is waiting for last recovery " + "to complete, backoff for a bit\n", dlm->name, + dead_node, nodenum); + msleep(100); + goto retry; + } if (ret < 0) { struct dlm_lock_resource *res; /* this is now a serious problem, possibly ENOMEM @@ -2608,14 +2616,6 @@ retry: * another ENOMEM */ msleep(100); goto retry; - } else if (ret == EAGAIN) { - mlog(0, "%s: trying to start recovery of node " - "%u, but node %u is waiting for last recovery " - "to complete, backoff for a bit\n", dlm->name, - dead_node, nodenum); - /* TODO Look into replacing msleep with cond_resched() */ - msleep(100); - goto retry; } } @@ -2639,7 +2639,7 @@ int dlm_begin_reco_handler(struct o2net_msg *msg, u32 len, void *data, dlm->name, br->node_idx, br->dead_node, dlm->reco.dead_node, dlm->reco.new_master); spin_unlock(&dlm->spinlock); - return EAGAIN; + return -EAGAIN; } spin_unlock(&dlm->spinlock); diff --git a/fs/ocfs2/extent_map.c b/fs/ocfs2/extent_map.c index 843db64e9d4..d35a27f4523 100644 --- a/fs/ocfs2/extent_map.c +++ b/fs/ocfs2/extent_map.c @@ -37,6 +37,7 @@ #include "extent_map.h" #include "inode.h" #include "super.h" +#include "symlink.h" #include "buffer_head_io.h" @@ -703,6 +704,12 @@ out: return ret; } +/* + * The ocfs2_fiemap_inline() may be a little bit misleading, since + * it not only handles the fiemap for inlined files, but also deals + * with the fast symlink, cause they have no difference for extent + * mapping per se. + */ static int ocfs2_fiemap_inline(struct inode *inode, struct buffer_head *di_bh, struct fiemap_extent_info *fieinfo, u64 map_start) @@ -715,11 +722,18 @@ static int ocfs2_fiemap_inline(struct inode *inode, struct buffer_head *di_bh, struct ocfs2_inode_info *oi = OCFS2_I(inode); di = (struct ocfs2_dinode *)di_bh->b_data; - id_count = le16_to_cpu(di->id2.i_data.id_count); + if (ocfs2_inode_is_fast_symlink(inode)) + id_count = ocfs2_fast_symlink_chars(inode->i_sb); + else + id_count = le16_to_cpu(di->id2.i_data.id_count); if (map_start < id_count) { phys = oi->ip_blkno << inode->i_sb->s_blocksize_bits; - phys += offsetof(struct ocfs2_dinode, id2.i_data.id_data); + if (ocfs2_inode_is_fast_symlink(inode)) + phys += offsetof(struct ocfs2_dinode, id2.i_symlink); + else + phys += offsetof(struct ocfs2_dinode, + id2.i_data.id_data); ret = fiemap_fill_next_extent(fieinfo, 0, phys, id_count, flags); @@ -756,9 +770,10 @@ int ocfs2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, down_read(&OCFS2_I(inode)->ip_alloc_sem); /* - * Handle inline-data separately. + * Handle inline-data and fast symlink separately. */ - if (OCFS2_I(inode)->ip_dyn_features & OCFS2_INLINE_DATA_FL) { + if ((OCFS2_I(inode)->ip_dyn_features & OCFS2_INLINE_DATA_FL) || + ocfs2_inode_is_fast_symlink(inode)) { ret = ocfs2_fiemap_inline(inode, di_bh, fieinfo, map_start); goto out_unlock; } @@ -786,6 +801,8 @@ int ocfs2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, fe_flags = 0; if (rec.e_flags & OCFS2_EXT_UNWRITTEN) fe_flags |= FIEMAP_EXTENT_UNWRITTEN; + if (rec.e_flags & OCFS2_EXT_REFCOUNTED) + fe_flags |= FIEMAP_EXTENT_SHARED; if (is_last) fe_flags |= FIEMAP_EXTENT_LAST; len_bytes = (u64)le16_to_cpu(rec.e_leaf_clusters) << osb->s_clustersize_bits; diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index 3d30a1c974a..06ccf6a86d3 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -1772,7 +1772,8 @@ static int ocfs2_prepare_inode_for_write(struct dentry *dentry, loff_t *ppos, size_t count, int appending, - int *direct_io) + int *direct_io, + int *has_refcount) { int ret = 0, meta_level = 0; struct inode *inode = dentry->d_inode; @@ -1833,6 +1834,8 @@ static int ocfs2_prepare_inode_for_write(struct dentry *dentry, saved_pos, count, &meta_level); + if (has_refcount) + *has_refcount = 1; } if (ret < 0) { @@ -1856,6 +1859,10 @@ static int ocfs2_prepare_inode_for_write(struct dentry *dentry, break; } + if (has_refcount && *has_refcount == 1) { + *direct_io = 0; + break; + } /* * Allowing concurrent direct writes means * i_size changes wouldn't be synchronized, so @@ -1899,7 +1906,7 @@ static ssize_t ocfs2_file_aio_write(struct kiocb *iocb, loff_t pos) { int ret, direct_io, appending, rw_level, have_alloc_sem = 0; - int can_do_direct; + int can_do_direct, has_refcount = 0; ssize_t written = 0; size_t ocount; /* original count */ size_t count; /* after file limit checks */ @@ -1942,7 +1949,7 @@ relock: can_do_direct = direct_io; ret = ocfs2_prepare_inode_for_write(file->f_path.dentry, ppos, iocb->ki_left, appending, - &can_do_direct); + &can_do_direct, &has_refcount); if (ret < 0) { mlog_errno(ret); goto out; @@ -2006,14 +2013,16 @@ out_dio: /* buffered aio wouldn't have proper lock coverage today */ BUG_ON(ret == -EIOCBQUEUED && !(file->f_flags & O_DIRECT)); - if ((file->f_flags & O_DSYNC && !direct_io) || IS_SYNC(inode)) { + if ((file->f_flags & O_DSYNC && !direct_io) || IS_SYNC(inode) || + (file->f_flags & O_DIRECT && has_refcount)) { ret = filemap_fdatawrite_range(file->f_mapping, pos, pos + count - 1); if (ret < 0) written = ret; if (!ret && (old_size != i_size_read(inode) || - old_clusters != OCFS2_I(inode)->ip_clusters)) { + old_clusters != OCFS2_I(inode)->ip_clusters || + has_refcount)) { ret = jbd2_journal_force_commit(osb->journal->j_journal); if (ret < 0) written = ret; @@ -2062,7 +2071,7 @@ static int ocfs2_splice_to_file(struct pipe_inode_info *pipe, int ret; ret = ocfs2_prepare_inode_for_write(out->f_path.dentry, &sd->pos, - sd->total_len, 0, NULL); + sd->total_len, 0, NULL, NULL); if (ret < 0) { mlog_errno(ret); return ret; diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c index f010b22b1c4..50fb26a6a5f 100644 --- a/fs/ocfs2/namei.c +++ b/fs/ocfs2/namei.c @@ -2108,6 +2108,7 @@ int ocfs2_create_inode_in_orphan(struct inode *dir, } did_quota_inode = 1; + inode->i_nlink = 0; /* do the real work now. */ status = ocfs2_mknod_locked(osb, dir, inode, 0, &new_di_bh, parent_di_bh, handle, @@ -2136,6 +2137,7 @@ int ocfs2_create_inode_in_orphan(struct inode *dir, if (status < 0) mlog_errno(status); + insert_inode_hash(inode); leave: if (status < 0 && did_quota_inode) vfs_dq_free_inode(inode); @@ -2267,6 +2269,8 @@ int ocfs2_mv_orphaned_inode_to_new(struct inode *dir, di = (struct ocfs2_dinode *)di_bh->b_data; le32_add_cpu(&di->i_flags, -OCFS2_ORPHANED_FL); di->i_orphaned_slot = 0; + inode->i_nlink = 1; + ocfs2_set_links_count(di, inode->i_nlink); ocfs2_journal_dirty(handle, di_bh); status = ocfs2_add_entry(handle, dentry, inode, @@ -2284,7 +2288,6 @@ int ocfs2_mv_orphaned_inode_to_new(struct inode *dir, goto out_commit; } - insert_inode_hash(inode); dentry->d_op = &ocfs2_dentry_ops; d_instantiate(dentry, inode); status = 0; @@ -2326,4 +2329,5 @@ const struct inode_operations ocfs2_dir_iops = { .getxattr = generic_getxattr, .listxattr = ocfs2_listxattr, .removexattr = generic_removexattr, + .fiemap = ocfs2_fiemap, }; diff --git a/fs/ocfs2/ocfs2.h b/fs/ocfs2/ocfs2.h index d963d863870..9362eea7424 100644 --- a/fs/ocfs2/ocfs2.h +++ b/fs/ocfs2/ocfs2.h @@ -245,9 +245,11 @@ enum ocfs2_mount_options OCFS2_MOUNT_LOCALFLOCKS = 1 << 5, /* No cluster aware user file locks */ OCFS2_MOUNT_NOUSERXATTR = 1 << 6, /* No user xattr */ OCFS2_MOUNT_INODE64 = 1 << 7, /* Allow inode numbers > 2^32 */ - OCFS2_MOUNT_POSIX_ACL = 1 << 8, /* POSIX access control lists */ - OCFS2_MOUNT_USRQUOTA = 1 << 9, /* We support user quotas */ - OCFS2_MOUNT_GRPQUOTA = 1 << 10, /* We support group quotas */ + OCFS2_MOUNT_POSIX_ACL = 1 << 8, /* Force POSIX access control lists */ + OCFS2_MOUNT_NO_POSIX_ACL = 1 << 9, /* Disable POSIX access + control lists */ + OCFS2_MOUNT_USRQUOTA = 1 << 10, /* We support user quotas */ + OCFS2_MOUNT_GRPQUOTA = 1 << 11, /* We support group quotas */ }; #define OCFS2_OSB_SOFT_RO 0x0001 diff --git a/fs/ocfs2/ocfs2_fs.h b/fs/ocfs2/ocfs2_fs.h index e9431e4a5e7..1a1a679e51b 100644 --- a/fs/ocfs2/ocfs2_fs.h +++ b/fs/ocfs2/ocfs2_fs.h @@ -1202,7 +1202,7 @@ struct ocfs2_local_disk_dqinfo { /* Header of one chunk of a quota file */ struct ocfs2_local_disk_chunk { __le32 dqc_free; /* Number of free entries in the bitmap */ - u8 dqc_bitmap[0]; /* Bitmap of entries in the corresponding + __u8 dqc_bitmap[0]; /* Bitmap of entries in the corresponding * chunk of quota file */ }; diff --git a/fs/ocfs2/refcounttree.c b/fs/ocfs2/refcounttree.c index 30967e3f5e4..74db2be75dd 100644 --- a/fs/ocfs2/refcounttree.c +++ b/fs/ocfs2/refcounttree.c @@ -276,7 +276,7 @@ static void ocfs2_erase_refcount_tree_from_list(struct ocfs2_super *osb, spin_unlock(&osb->osb_lock); } -void ocfs2_kref_remove_refcount_tree(struct kref *kref) +static void ocfs2_kref_remove_refcount_tree(struct kref *kref) { struct ocfs2_refcount_tree *tree = container_of(kref, struct ocfs2_refcount_tree, rf_getcnt); @@ -524,23 +524,6 @@ out: return ret; } -int ocfs2_lock_refcount_tree_by_inode(struct inode *inode, int rw, - struct ocfs2_refcount_tree **ret_tree, - struct buffer_head **ref_bh) -{ - int ret; - u64 ref_blkno; - - ret = ocfs2_get_refcount_block(inode, &ref_blkno); - if (ret) { - mlog_errno(ret); - return ret; - } - - return ocfs2_lock_refcount_tree(OCFS2_SB(inode->i_sb), ref_blkno, - rw, ret_tree, ref_bh); -} - void ocfs2_unlock_refcount_tree(struct ocfs2_super *osb, struct ocfs2_refcount_tree *tree, int rw) { @@ -969,6 +952,103 @@ out: } /* + * Find the end range for a leaf refcount block indicated by + * el->l_recs[index].e_blkno. + */ +static int ocfs2_get_refcount_cpos_end(struct ocfs2_caching_info *ci, + struct buffer_head *ref_root_bh, + struct ocfs2_extent_block *eb, + struct ocfs2_extent_list *el, + int index, u32 *cpos_end) +{ + int ret, i, subtree_root; + u32 cpos; + u64 blkno; + struct super_block *sb = ocfs2_metadata_cache_get_super(ci); + struct ocfs2_path *left_path = NULL, *right_path = NULL; + struct ocfs2_extent_tree et; + struct ocfs2_extent_list *tmp_el; + + if (index < le16_to_cpu(el->l_next_free_rec) - 1) { + /* + * We have a extent rec after index, so just use the e_cpos + * of the next extent rec. + */ + *cpos_end = le32_to_cpu(el->l_recs[index+1].e_cpos); + return 0; + } + + if (!eb || (eb && !eb->h_next_leaf_blk)) { + /* + * We are the last extent rec, so any high cpos should + * be stored in this leaf refcount block. + */ + *cpos_end = UINT_MAX; + return 0; + } + + /* + * If the extent block isn't the last one, we have to find + * the subtree root between this extent block and the next + * leaf extent block and get the corresponding e_cpos from + * the subroot. Otherwise we may corrupt the b-tree. + */ + ocfs2_init_refcount_extent_tree(&et, ci, ref_root_bh); + + left_path = ocfs2_new_path_from_et(&et); + if (!left_path) { + ret = -ENOMEM; + mlog_errno(ret); + goto out; + } + + cpos = le32_to_cpu(eb->h_list.l_recs[index].e_cpos); + ret = ocfs2_find_path(ci, left_path, cpos); + if (ret) { + mlog_errno(ret); + goto out; + } + + right_path = ocfs2_new_path_from_path(left_path); + if (!right_path) { + ret = -ENOMEM; + mlog_errno(ret); + goto out; + } + + ret = ocfs2_find_cpos_for_right_leaf(sb, left_path, &cpos); + if (ret) { + mlog_errno(ret); + goto out; + } + + ret = ocfs2_find_path(ci, right_path, cpos); + if (ret) { + mlog_errno(ret); + goto out; + } + + subtree_root = ocfs2_find_subtree_root(&et, left_path, + right_path); + + tmp_el = left_path->p_node[subtree_root].el; + blkno = left_path->p_node[subtree_root+1].bh->b_blocknr; + for (i = 0; i < le32_to_cpu(tmp_el->l_next_free_rec); i++) { + if (le64_to_cpu(tmp_el->l_recs[i].e_blkno) == blkno) { + *cpos_end = le32_to_cpu(tmp_el->l_recs[i+1].e_cpos); + break; + } + } + + BUG_ON(i == le32_to_cpu(tmp_el->l_next_free_rec)); + +out: + ocfs2_free_path(left_path); + ocfs2_free_path(right_path); + return ret; +} + +/* * Given a cpos and len, try to find the refcount record which contains cpos. * 1. If cpos can be found in one refcount record, return the record. * 2. If cpos can't be found, return a fake record which start from cpos @@ -983,10 +1063,10 @@ static int ocfs2_get_refcount_rec(struct ocfs2_caching_info *ci, struct buffer_head **ret_bh) { int ret = 0, i, found; - u32 low_cpos; + u32 low_cpos, uninitialized_var(cpos_end); struct ocfs2_extent_list *el; - struct ocfs2_extent_rec *tmp, *rec = NULL; - struct ocfs2_extent_block *eb; + struct ocfs2_extent_rec *rec = NULL; + struct ocfs2_extent_block *eb = NULL; struct buffer_head *eb_bh = NULL, *ref_leaf_bh = NULL; struct super_block *sb = ocfs2_metadata_cache_get_super(ci); struct ocfs2_refcount_block *rb = @@ -1034,12 +1114,16 @@ static int ocfs2_get_refcount_rec(struct ocfs2_caching_info *ci, } } - /* adjust len when we have ocfs2_extent_rec after it. */ - if (found && i < le16_to_cpu(el->l_next_free_rec) - 1) { - tmp = &el->l_recs[i+1]; + if (found) { + ret = ocfs2_get_refcount_cpos_end(ci, ref_root_bh, + eb, el, i, &cpos_end); + if (ret) { + mlog_errno(ret); + goto out; + } - if (le32_to_cpu(tmp->e_cpos) < cpos + len) - len = le32_to_cpu(tmp->e_cpos) - cpos; + if (cpos_end < low_cpos + len) + len = cpos_end - low_cpos; } ret = ocfs2_read_refcount_block(ci, le64_to_cpu(rec->e_blkno), @@ -1418,7 +1502,7 @@ static int ocfs2_divide_leaf_refcount_block(struct buffer_head *ref_leaf_bh, /* change old and new rl_used accordingly. */ le16_add_cpu(&rl->rl_used, -num_moved); - new_rl->rl_used = cpu_to_le32(num_moved); + new_rl->rl_used = cpu_to_le16(num_moved); sort(&rl->rl_recs, le16_to_cpu(rl->rl_used), sizeof(struct ocfs2_refcount_rec), @@ -1797,7 +1881,8 @@ static int ocfs2_split_refcount_rec(handle_t *handle, recs_need++; /* If the leaf block don't have enough record, expand it. */ - if (le16_to_cpu(rf_list->rl_used) + recs_need > rf_list->rl_count) { + if (le16_to_cpu(rf_list->rl_used) + recs_need > + le16_to_cpu(rf_list->rl_count)) { struct ocfs2_refcount_rec tmp_rec; u64 cpos = le64_to_cpu(orig_rec->r_cpos); len = le32_to_cpu(orig_rec->r_clusters); @@ -1859,7 +1944,7 @@ static int ocfs2_split_refcount_rec(handle_t *handle, memcpy(tail_rec, orig_rec, sizeof(struct ocfs2_refcount_rec)); le64_add_cpu(&tail_rec->r_cpos, le32_to_cpu(tail_rec->r_clusters) - len); - tail_rec->r_clusters = le32_to_cpu(len); + tail_rec->r_clusters = cpu_to_le32(len); } /* @@ -3840,8 +3925,7 @@ static int ocfs2_add_refcounted_extent(struct inode *inode, } ret = ocfs2_insert_extent(handle, et, cpos, - cpu_to_le64(ocfs2_clusters_to_blocks(inode->i_sb, - p_cluster)), + ocfs2_clusters_to_blocks(inode->i_sb, p_cluster), num_clusters, ext_flags, meta_ac); if (ret) { mlog_errno(ret); @@ -4253,8 +4337,8 @@ static int ocfs2_user_path_parent(const char __user *path, * @new_dentry: target dentry * @preserve: if true, preserve all file attributes */ -int ocfs2_vfs_reflink(struct dentry *old_dentry, struct inode *dir, - struct dentry *new_dentry, bool preserve) +static int ocfs2_vfs_reflink(struct dentry *old_dentry, struct inode *dir, + struct dentry *new_dentry, bool preserve) { struct inode *inode = old_dentry->d_inode; int error; diff --git a/fs/ocfs2/stack_user.c b/fs/ocfs2/stack_user.c index ff4c798a563..da78a2a334f 100644 --- a/fs/ocfs2/stack_user.c +++ b/fs/ocfs2/stack_user.c @@ -814,7 +814,7 @@ static int fs_protocol_compare(struct ocfs2_protocol_version *existing, static int user_cluster_connect(struct ocfs2_cluster_connection *conn) { dlm_lockspace_t *fsdlm; - struct ocfs2_live_connection *control; + struct ocfs2_live_connection *uninitialized_var(control); int rc = 0; BUG_ON(conn == NULL); diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index 14f47d2bfe0..26069917a9f 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -100,6 +100,8 @@ struct mount_options static int ocfs2_parse_options(struct super_block *sb, char *options, struct mount_options *mopt, int is_remount); +static int ocfs2_check_set_options(struct super_block *sb, + struct mount_options *options); static int ocfs2_show_options(struct seq_file *s, struct vfsmount *mnt); static void ocfs2_put_super(struct super_block *sb); static int ocfs2_mount_volume(struct super_block *sb); @@ -600,7 +602,8 @@ static int ocfs2_remount(struct super_block *sb, int *flags, char *data) lock_kernel(); - if (!ocfs2_parse_options(sb, data, &parsed_options, 1)) { + if (!ocfs2_parse_options(sb, data, &parsed_options, 1) || + !ocfs2_check_set_options(sb, &parsed_options)) { ret = -EINVAL; goto out; } @@ -691,8 +694,6 @@ unlock_osb: if (!ret) { /* Only save off the new mount options in case of a successful * remount. */ - if (!(osb->s_feature_incompat & OCFS2_FEATURE_INCOMPAT_XATTR)) - parsed_options.mount_opt &= ~OCFS2_MOUNT_POSIX_ACL; osb->s_mount_opt = parsed_options.mount_opt; osb->s_atime_quantum = parsed_options.atime_quantum; osb->preferred_slot = parsed_options.slot; @@ -701,6 +702,10 @@ unlock_osb: if (!ocfs2_is_hard_readonly(osb)) ocfs2_set_journal_params(osb); + + sb->s_flags = (sb->s_flags & ~MS_POSIXACL) | + ((osb->s_mount_opt & OCFS2_MOUNT_POSIX_ACL) ? + MS_POSIXACL : 0); } out: unlock_kernel(); @@ -1011,31 +1016,16 @@ static int ocfs2_fill_super(struct super_block *sb, void *data, int silent) brelse(bh); bh = NULL; - if (!(osb->s_feature_incompat & OCFS2_FEATURE_INCOMPAT_XATTR)) - parsed_options.mount_opt &= ~OCFS2_MOUNT_POSIX_ACL; - + if (!ocfs2_check_set_options(sb, &parsed_options)) { + status = -EINVAL; + goto read_super_error; + } osb->s_mount_opt = parsed_options.mount_opt; osb->s_atime_quantum = parsed_options.atime_quantum; osb->preferred_slot = parsed_options.slot; osb->osb_commit_interval = parsed_options.commit_interval; osb->local_alloc_default_bits = ocfs2_megabytes_to_clusters(sb, parsed_options.localalloc_opt); osb->local_alloc_bits = osb->local_alloc_default_bits; - if (osb->s_mount_opt & OCFS2_MOUNT_USRQUOTA && - !OCFS2_HAS_RO_COMPAT_FEATURE(sb, - OCFS2_FEATURE_RO_COMPAT_USRQUOTA)) { - status = -EINVAL; - mlog(ML_ERROR, "User quotas were requested, but this " - "filesystem does not have the feature enabled.\n"); - goto read_super_error; - } - if (osb->s_mount_opt & OCFS2_MOUNT_GRPQUOTA && - !OCFS2_HAS_RO_COMPAT_FEATURE(sb, - OCFS2_FEATURE_RO_COMPAT_GRPQUOTA)) { - status = -EINVAL; - mlog(ML_ERROR, "Group quotas were requested, but this " - "filesystem does not have the feature enabled.\n"); - goto read_super_error; - } status = ocfs2_verify_userspace_stack(osb, &parsed_options); if (status) @@ -1245,6 +1235,40 @@ static struct file_system_type ocfs2_fs_type = { .next = NULL }; +static int ocfs2_check_set_options(struct super_block *sb, + struct mount_options *options) +{ + if (options->mount_opt & OCFS2_MOUNT_USRQUOTA && + !OCFS2_HAS_RO_COMPAT_FEATURE(sb, + OCFS2_FEATURE_RO_COMPAT_USRQUOTA)) { + mlog(ML_ERROR, "User quotas were requested, but this " + "filesystem does not have the feature enabled.\n"); + return 0; + } + if (options->mount_opt & OCFS2_MOUNT_GRPQUOTA && + !OCFS2_HAS_RO_COMPAT_FEATURE(sb, + OCFS2_FEATURE_RO_COMPAT_GRPQUOTA)) { + mlog(ML_ERROR, "Group quotas were requested, but this " + "filesystem does not have the feature enabled.\n"); + return 0; + } + if (options->mount_opt & OCFS2_MOUNT_POSIX_ACL && + !OCFS2_HAS_INCOMPAT_FEATURE(sb, OCFS2_FEATURE_INCOMPAT_XATTR)) { + mlog(ML_ERROR, "ACL support requested but extended attributes " + "feature is not enabled\n"); + return 0; + } + /* No ACL setting specified? Use XATTR feature... */ + if (!(options->mount_opt & (OCFS2_MOUNT_POSIX_ACL | + OCFS2_MOUNT_NO_POSIX_ACL))) { + if (OCFS2_HAS_INCOMPAT_FEATURE(sb, OCFS2_FEATURE_INCOMPAT_XATTR)) + options->mount_opt |= OCFS2_MOUNT_POSIX_ACL; + else + options->mount_opt |= OCFS2_MOUNT_NO_POSIX_ACL; + } + return 1; +} + static int ocfs2_parse_options(struct super_block *sb, char *options, struct mount_options *mopt, @@ -1392,40 +1416,19 @@ static int ocfs2_parse_options(struct super_block *sb, mopt->mount_opt |= OCFS2_MOUNT_INODE64; break; case Opt_usrquota: - /* We check only on remount, otherwise features - * aren't yet initialized. */ - if (is_remount && !OCFS2_HAS_RO_COMPAT_FEATURE(sb, - OCFS2_FEATURE_RO_COMPAT_USRQUOTA)) { - mlog(ML_ERROR, "User quota requested but " - "filesystem feature is not set\n"); - status = 0; - goto bail; - } mopt->mount_opt |= OCFS2_MOUNT_USRQUOTA; break; case Opt_grpquota: - if (is_remount && !OCFS2_HAS_RO_COMPAT_FEATURE(sb, - OCFS2_FEATURE_RO_COMPAT_GRPQUOTA)) { - mlog(ML_ERROR, "Group quota requested but " - "filesystem feature is not set\n"); - status = 0; - goto bail; - } mopt->mount_opt |= OCFS2_MOUNT_GRPQUOTA; break; -#ifdef CONFIG_OCFS2_FS_POSIX_ACL case Opt_acl: mopt->mount_opt |= OCFS2_MOUNT_POSIX_ACL; + mopt->mount_opt &= ~OCFS2_MOUNT_NO_POSIX_ACL; break; case Opt_noacl: + mopt->mount_opt |= OCFS2_MOUNT_NO_POSIX_ACL; mopt->mount_opt &= ~OCFS2_MOUNT_POSIX_ACL; break; -#else - case Opt_acl: - case Opt_noacl: - printk(KERN_INFO "ocfs2 (no)acl options not supported\n"); - break; -#endif default: mlog(ML_ERROR, "Unrecognized mount option \"%s\" " @@ -1502,12 +1505,10 @@ static int ocfs2_show_options(struct seq_file *s, struct vfsmount *mnt) if (opts & OCFS2_MOUNT_INODE64) seq_printf(s, ",inode64"); -#ifdef CONFIG_OCFS2_FS_POSIX_ACL if (opts & OCFS2_MOUNT_POSIX_ACL) seq_printf(s, ",acl"); else seq_printf(s, ",noacl"); -#endif return 0; } diff --git a/fs/ocfs2/symlink.c b/fs/ocfs2/symlink.c index e3421030a69..49b133ccbf1 100644 --- a/fs/ocfs2/symlink.c +++ b/fs/ocfs2/symlink.c @@ -163,6 +163,7 @@ const struct inode_operations ocfs2_symlink_inode_operations = { .getxattr = generic_getxattr, .listxattr = ocfs2_listxattr, .removexattr = generic_removexattr, + .fiemap = ocfs2_fiemap, }; const struct inode_operations ocfs2_fast_symlink_inode_operations = { .readlink = ocfs2_readlink, @@ -174,4 +175,5 @@ const struct inode_operations ocfs2_fast_symlink_inode_operations = { .getxattr = generic_getxattr, .listxattr = ocfs2_listxattr, .removexattr = generic_removexattr, + .fiemap = ocfs2_fiemap, }; diff --git a/fs/ocfs2/xattr.c b/fs/ocfs2/xattr.c index 43c114831c0..8fc6fb071c6 100644 --- a/fs/ocfs2/xattr.c +++ b/fs/ocfs2/xattr.c @@ -98,10 +98,8 @@ static struct ocfs2_xattr_def_value_root def_xv = { struct xattr_handler *ocfs2_xattr_handlers[] = { &ocfs2_xattr_user_handler, -#ifdef CONFIG_OCFS2_FS_POSIX_ACL &ocfs2_xattr_acl_access_handler, &ocfs2_xattr_acl_default_handler, -#endif &ocfs2_xattr_trusted_handler, &ocfs2_xattr_security_handler, NULL @@ -109,12 +107,10 @@ struct xattr_handler *ocfs2_xattr_handlers[] = { static struct xattr_handler *ocfs2_xattr_handler_map[OCFS2_XATTR_MAX] = { [OCFS2_XATTR_INDEX_USER] = &ocfs2_xattr_user_handler, -#ifdef CONFIG_OCFS2_FS_POSIX_ACL [OCFS2_XATTR_INDEX_POSIX_ACL_ACCESS] = &ocfs2_xattr_acl_access_handler, [OCFS2_XATTR_INDEX_POSIX_ACL_DEFAULT] = &ocfs2_xattr_acl_default_handler, -#endif [OCFS2_XATTR_INDEX_TRUSTED] = &ocfs2_xattr_trusted_handler, [OCFS2_XATTR_INDEX_SECURITY] = &ocfs2_xattr_security_handler, }; @@ -6064,7 +6060,7 @@ static int ocfs2_value_metas_in_xattr_header(struct super_block *sb, * to the extent block, so just calculate a maximum record num. */ if (!xv->xr_list.l_tree_depth) - *num_recs += xv->xr_list.l_next_free_rec; + *num_recs += le16_to_cpu(xv->xr_list.l_next_free_rec); else *num_recs += ocfs2_clusters_for_bytes(sb, XATTR_SIZE_MAX); diff --git a/fs/ocfs2/xattr.h b/fs/ocfs2/xattr.h index 08e36389f56..abd72a47f52 100644 --- a/fs/ocfs2/xattr.h +++ b/fs/ocfs2/xattr.h @@ -40,10 +40,8 @@ struct ocfs2_security_xattr_info { extern struct xattr_handler ocfs2_xattr_user_handler; extern struct xattr_handler ocfs2_xattr_trusted_handler; extern struct xattr_handler ocfs2_xattr_security_handler; -#ifdef CONFIG_OCFS2_FS_POSIX_ACL extern struct xattr_handler ocfs2_xattr_acl_access_handler; extern struct xattr_handler ocfs2_xattr_acl_default_handler; -#endif extern struct xattr_handler *ocfs2_xattr_handlers[]; ssize_t ocfs2_listxattr(struct dentry *, char *, size_t); diff --git a/fs/open.c b/fs/open.c index ca69241796b..040cef72bc0 100644 --- a/fs/open.c +++ b/fs/open.c @@ -821,15 +821,14 @@ static inline int __get_file_write_access(struct inode *inode, } static struct file *__dentry_open(struct dentry *dentry, struct vfsmount *mnt, - int flags, struct file *f, + struct file *f, int (*open)(struct inode *, struct file *), const struct cred *cred) { struct inode *inode; int error; - f->f_flags = flags; - f->f_mode = (__force fmode_t)((flags+1) & O_ACCMODE) | FMODE_LSEEK | + f->f_mode = OPEN_FMODE(f->f_flags) | FMODE_LSEEK | FMODE_PREAD | FMODE_PWRITE; inode = dentry->d_inode; if (f->f_mode & FMODE_WRITE) { @@ -930,7 +929,6 @@ struct file *lookup_instantiate_filp(struct nameidata *nd, struct dentry *dentry if (IS_ERR(dentry)) goto out_err; nd->intent.open.file = __dentry_open(dget(dentry), mntget(nd->path.mnt), - nd->intent.open.flags - 1, nd->intent.open.file, open, cred); out: @@ -949,7 +947,7 @@ EXPORT_SYMBOL_GPL(lookup_instantiate_filp); * * Note that this function destroys the original nameidata */ -struct file *nameidata_to_filp(struct nameidata *nd, int flags) +struct file *nameidata_to_filp(struct nameidata *nd) { const struct cred *cred = current_cred(); struct file *filp; @@ -958,7 +956,7 @@ struct file *nameidata_to_filp(struct nameidata *nd, int flags) filp = nd->intent.open.file; /* Has the filesystem initialised the file for us? */ if (filp->f_path.dentry == NULL) - filp = __dentry_open(nd->path.dentry, nd->path.mnt, flags, filp, + filp = __dentry_open(nd->path.dentry, nd->path.mnt, filp, NULL, cred); else path_put(&nd->path); @@ -997,7 +995,8 @@ struct file *dentry_open(struct dentry *dentry, struct vfsmount *mnt, int flags, return ERR_PTR(error); } - return __dentry_open(dentry, mnt, flags, f, NULL, cred); + f->f_flags = flags; + return __dentry_open(dentry, mnt, f, NULL, cred); } EXPORT_SYMBOL(dentry_open); diff --git a/fs/proc/array.c b/fs/proc/array.c index 4badde179b1..f560325c444 100644 --- a/fs/proc/array.c +++ b/fs/proc/array.c @@ -134,13 +134,16 @@ static inline void task_name(struct seq_file *m, struct task_struct *p) * simple bit tests. */ static const char *task_state_array[] = { - "R (running)", /* 0 */ - "S (sleeping)", /* 1 */ - "D (disk sleep)", /* 2 */ - "T (stopped)", /* 4 */ - "T (tracing stop)", /* 8 */ - "Z (zombie)", /* 16 */ - "X (dead)" /* 32 */ + "R (running)", /* 0 */ + "S (sleeping)", /* 1 */ + "D (disk sleep)", /* 2 */ + "T (stopped)", /* 4 */ + "t (tracing stop)", /* 8 */ + "Z (zombie)", /* 16 */ + "X (dead)", /* 32 */ + "x (dead)", /* 64 */ + "K (wakekill)", /* 128 */ + "W (waking)", /* 256 */ }; static inline const char *get_task_state(struct task_struct *tsk) @@ -148,6 +151,8 @@ static inline const char *get_task_state(struct task_struct *tsk) unsigned int state = (tsk->state & TASK_REPORT) | tsk->exit_state; const char **p = &task_state_array[0]; + BUILD_BUG_ON(1 + ilog2(TASK_STATE_MAX) != ARRAY_SIZE(task_state_array)); + while (state) { p++; state >>= 1; diff --git a/fs/quota/dquot.c b/fs/quota/dquot.c index cd6bb9a33c1..dea86abdf2e 100644 --- a/fs/quota/dquot.c +++ b/fs/quota/dquot.c @@ -323,6 +323,30 @@ int dquot_mark_dquot_dirty(struct dquot *dquot) } EXPORT_SYMBOL(dquot_mark_dquot_dirty); +/* Dirtify all the dquots - this can block when journalling */ +static inline int mark_all_dquot_dirty(struct dquot * const *dquot) +{ + int ret, err, cnt; + + ret = err = 0; + for (cnt = 0; cnt < MAXQUOTAS; cnt++) { + if (dquot[cnt]) + /* Even in case of error we have to continue */ + ret = mark_dquot_dirty(dquot[cnt]); + if (!err) + err = ret; + } + return err; +} + +static inline void dqput_all(struct dquot **dquot) +{ + unsigned int cnt; + + for (cnt = 0; cnt < MAXQUOTAS; cnt++) + dqput(dquot[cnt]); +} + /* This function needs dq_list_lock */ static inline int clear_dquot_dirty(struct dquot *dquot) { @@ -1268,8 +1292,7 @@ int dquot_initialize(struct inode *inode, int type) out_err: up_write(&sb_dqopt(sb)->dqptr_sem); /* Drop unused references */ - for (cnt = 0; cnt < MAXQUOTAS; cnt++) - dqput(got[cnt]); + dqput_all(got); return ret; } EXPORT_SYMBOL(dquot_initialize); @@ -1288,9 +1311,7 @@ int dquot_drop(struct inode *inode) inode->i_dquot[cnt] = NULL; } up_write(&sb_dqopt(inode->i_sb)->dqptr_sem); - - for (cnt = 0; cnt < MAXQUOTAS; cnt++) - dqput(put[cnt]); + dqput_all(put); return 0; } EXPORT_SYMBOL(dquot_drop); @@ -1319,6 +1340,67 @@ void vfs_dq_drop(struct inode *inode) EXPORT_SYMBOL(vfs_dq_drop); /* + * inode_reserved_space is managed internally by quota, and protected by + * i_lock similar to i_blocks+i_bytes. + */ +static qsize_t *inode_reserved_space(struct inode * inode) +{ + /* Filesystem must explicitly define it's own method in order to use + * quota reservation interface */ + BUG_ON(!inode->i_sb->dq_op->get_reserved_space); + return inode->i_sb->dq_op->get_reserved_space(inode); +} + +static void inode_add_rsv_space(struct inode *inode, qsize_t number) +{ + spin_lock(&inode->i_lock); + *inode_reserved_space(inode) += number; + spin_unlock(&inode->i_lock); +} + + +static void inode_claim_rsv_space(struct inode *inode, qsize_t number) +{ + spin_lock(&inode->i_lock); + *inode_reserved_space(inode) -= number; + __inode_add_bytes(inode, number); + spin_unlock(&inode->i_lock); +} + +static void inode_sub_rsv_space(struct inode *inode, qsize_t number) +{ + spin_lock(&inode->i_lock); + *inode_reserved_space(inode) -= number; + spin_unlock(&inode->i_lock); +} + +static qsize_t inode_get_rsv_space(struct inode *inode) +{ + qsize_t ret; + spin_lock(&inode->i_lock); + ret = *inode_reserved_space(inode); + spin_unlock(&inode->i_lock); + return ret; +} + +static void inode_incr_space(struct inode *inode, qsize_t number, + int reserve) +{ + if (reserve) + inode_add_rsv_space(inode, number); + else + inode_add_bytes(inode, number); +} + +static void inode_decr_space(struct inode *inode, qsize_t number, int reserve) +{ + if (reserve) + inode_sub_rsv_space(inode, number); + else + inode_sub_bytes(inode, number); +} + +/* * Following four functions update i_blocks+i_bytes fields and * quota information (together with appropriate checks) * NOTE: We absolutely rely on the fact that caller dirties @@ -1336,6 +1418,21 @@ int __dquot_alloc_space(struct inode *inode, qsize_t number, int cnt, ret = QUOTA_OK; char warntype[MAXQUOTAS]; + /* + * First test before acquiring mutex - solves deadlocks when we + * re-enter the quota code and are already holding the mutex + */ + if (IS_NOQUOTA(inode)) { + inode_incr_space(inode, number, reserve); + goto out; + } + + down_read(&sb_dqopt(inode->i_sb)->dqptr_sem); + if (IS_NOQUOTA(inode)) { + inode_incr_space(inode, number, reserve); + goto out_unlock; + } + for (cnt = 0; cnt < MAXQUOTAS; cnt++) warntype[cnt] = QUOTA_NL_NOWARN; @@ -1346,7 +1443,8 @@ int __dquot_alloc_space(struct inode *inode, qsize_t number, if (check_bdq(inode->i_dquot[cnt], number, warn, warntype+cnt) == NO_QUOTA) { ret = NO_QUOTA; - goto out_unlock; + spin_unlock(&dq_data_lock); + goto out_flush_warn; } } for (cnt = 0; cnt < MAXQUOTAS; cnt++) { @@ -1357,64 +1455,29 @@ int __dquot_alloc_space(struct inode *inode, qsize_t number, else dquot_incr_space(inode->i_dquot[cnt], number); } - if (!reserve) - inode_add_bytes(inode, number); -out_unlock: + inode_incr_space(inode, number, reserve); spin_unlock(&dq_data_lock); + + if (reserve) + goto out_flush_warn; + mark_all_dquot_dirty(inode->i_dquot); +out_flush_warn: flush_warnings(inode->i_dquot, warntype); +out_unlock: + up_read(&sb_dqopt(inode->i_sb)->dqptr_sem); +out: return ret; } int dquot_alloc_space(struct inode *inode, qsize_t number, int warn) { - int cnt, ret = QUOTA_OK; - - /* - * First test before acquiring mutex - solves deadlocks when we - * re-enter the quota code and are already holding the mutex - */ - if (IS_NOQUOTA(inode)) { - inode_add_bytes(inode, number); - goto out; - } - - down_read(&sb_dqopt(inode->i_sb)->dqptr_sem); - if (IS_NOQUOTA(inode)) { - inode_add_bytes(inode, number); - goto out_unlock; - } - - ret = __dquot_alloc_space(inode, number, warn, 0); - if (ret == NO_QUOTA) - goto out_unlock; - - /* Dirtify all the dquots - this can block when journalling */ - for (cnt = 0; cnt < MAXQUOTAS; cnt++) - if (inode->i_dquot[cnt]) - mark_dquot_dirty(inode->i_dquot[cnt]); -out_unlock: - up_read(&sb_dqopt(inode->i_sb)->dqptr_sem); -out: - return ret; + return __dquot_alloc_space(inode, number, warn, 0); } EXPORT_SYMBOL(dquot_alloc_space); int dquot_reserve_space(struct inode *inode, qsize_t number, int warn) { - int ret = QUOTA_OK; - - if (IS_NOQUOTA(inode)) - goto out; - - down_read(&sb_dqopt(inode->i_sb)->dqptr_sem); - if (IS_NOQUOTA(inode)) - goto out_unlock; - - ret = __dquot_alloc_space(inode, number, warn, 1); -out_unlock: - up_read(&sb_dqopt(inode->i_sb)->dqptr_sem); -out: - return ret; + return __dquot_alloc_space(inode, number, warn, 1); } EXPORT_SYMBOL(dquot_reserve_space); @@ -1455,10 +1518,7 @@ int dquot_alloc_inode(const struct inode *inode, qsize_t number) warn_put_all: spin_unlock(&dq_data_lock); if (ret == QUOTA_OK) - /* Dirtify all the dquots - this can block when journalling */ - for (cnt = 0; cnt < MAXQUOTAS; cnt++) - if (inode->i_dquot[cnt]) - mark_dquot_dirty(inode->i_dquot[cnt]); + mark_all_dquot_dirty(inode->i_dquot); flush_warnings(inode->i_dquot, warntype); up_read(&sb_dqopt(inode->i_sb)->dqptr_sem); return ret; @@ -1471,14 +1531,14 @@ int dquot_claim_space(struct inode *inode, qsize_t number) int ret = QUOTA_OK; if (IS_NOQUOTA(inode)) { - inode_add_bytes(inode, number); + inode_claim_rsv_space(inode, number); goto out; } down_read(&sb_dqopt(inode->i_sb)->dqptr_sem); if (IS_NOQUOTA(inode)) { up_read(&sb_dqopt(inode->i_sb)->dqptr_sem); - inode_add_bytes(inode, number); + inode_claim_rsv_space(inode, number); goto out; } @@ -1490,12 +1550,9 @@ int dquot_claim_space(struct inode *inode, qsize_t number) number); } /* Update inode bytes */ - inode_add_bytes(inode, number); + inode_claim_rsv_space(inode, number); spin_unlock(&dq_data_lock); - /* Dirtify all the dquots - this can block when journalling */ - for (cnt = 0; cnt < MAXQUOTAS; cnt++) - if (inode->i_dquot[cnt]) - mark_dquot_dirty(inode->i_dquot[cnt]); + mark_all_dquot_dirty(inode->i_dquot); up_read(&sb_dqopt(inode->i_sb)->dqptr_sem); out: return ret; @@ -1503,38 +1560,9 @@ out: EXPORT_SYMBOL(dquot_claim_space); /* - * Release reserved quota space - */ -void dquot_release_reserved_space(struct inode *inode, qsize_t number) -{ - int cnt; - - if (IS_NOQUOTA(inode)) - goto out; - - down_read(&sb_dqopt(inode->i_sb)->dqptr_sem); - if (IS_NOQUOTA(inode)) - goto out_unlock; - - spin_lock(&dq_data_lock); - /* Release reserved dquots */ - for (cnt = 0; cnt < MAXQUOTAS; cnt++) { - if (inode->i_dquot[cnt]) - dquot_free_reserved_space(inode->i_dquot[cnt], number); - } - spin_unlock(&dq_data_lock); - -out_unlock: - up_read(&sb_dqopt(inode->i_sb)->dqptr_sem); -out: - return; -} -EXPORT_SYMBOL(dquot_release_reserved_space); - -/* * This operation can block, but only after everything is updated */ -int dquot_free_space(struct inode *inode, qsize_t number) +int __dquot_free_space(struct inode *inode, qsize_t number, int reserve) { unsigned int cnt; char warntype[MAXQUOTAS]; @@ -1543,7 +1571,7 @@ int dquot_free_space(struct inode *inode, qsize_t number) * re-enter the quota code and are already holding the mutex */ if (IS_NOQUOTA(inode)) { out_sub: - inode_sub_bytes(inode, number); + inode_decr_space(inode, number, reserve); return QUOTA_OK; } @@ -1558,21 +1586,40 @@ out_sub: if (!inode->i_dquot[cnt]) continue; warntype[cnt] = info_bdq_free(inode->i_dquot[cnt], number); - dquot_decr_space(inode->i_dquot[cnt], number); + if (reserve) + dquot_free_reserved_space(inode->i_dquot[cnt], number); + else + dquot_decr_space(inode->i_dquot[cnt], number); } - inode_sub_bytes(inode, number); + inode_decr_space(inode, number, reserve); spin_unlock(&dq_data_lock); - /* Dirtify all the dquots - this can block when journalling */ - for (cnt = 0; cnt < MAXQUOTAS; cnt++) - if (inode->i_dquot[cnt]) - mark_dquot_dirty(inode->i_dquot[cnt]); + + if (reserve) + goto out_unlock; + mark_all_dquot_dirty(inode->i_dquot); +out_unlock: flush_warnings(inode->i_dquot, warntype); up_read(&sb_dqopt(inode->i_sb)->dqptr_sem); return QUOTA_OK; } + +int dquot_free_space(struct inode *inode, qsize_t number) +{ + return __dquot_free_space(inode, number, 0); +} EXPORT_SYMBOL(dquot_free_space); /* + * Release reserved quota space + */ +void dquot_release_reserved_space(struct inode *inode, qsize_t number) +{ + __dquot_free_space(inode, number, 1); + +} +EXPORT_SYMBOL(dquot_release_reserved_space); + +/* * This operation can block, but only after everything is updated */ int dquot_free_inode(const struct inode *inode, qsize_t number) @@ -1599,10 +1646,7 @@ int dquot_free_inode(const struct inode *inode, qsize_t number) dquot_decr_inodes(inode->i_dquot[cnt], number); } spin_unlock(&dq_data_lock); - /* Dirtify all the dquots - this can block when journalling */ - for (cnt = 0; cnt < MAXQUOTAS; cnt++) - if (inode->i_dquot[cnt]) - mark_dquot_dirty(inode->i_dquot[cnt]); + mark_all_dquot_dirty(inode->i_dquot); flush_warnings(inode->i_dquot, warntype); up_read(&sb_dqopt(inode->i_sb)->dqptr_sem); return QUOTA_OK; @@ -1610,19 +1654,6 @@ int dquot_free_inode(const struct inode *inode, qsize_t number) EXPORT_SYMBOL(dquot_free_inode); /* - * call back function, get reserved quota space from underlying fs - */ -qsize_t dquot_get_reserved_space(struct inode *inode) -{ - qsize_t reserved_space = 0; - - if (sb_any_quota_active(inode->i_sb) && - inode->i_sb->dq_op->get_reserved_space) - reserved_space = inode->i_sb->dq_op->get_reserved_space(inode); - return reserved_space; -} - -/* * Transfer the number of inode and blocks from one diskquota to an other. * * This operation can block, but only after everything is updated @@ -1665,7 +1696,7 @@ int dquot_transfer(struct inode *inode, struct iattr *iattr) } spin_lock(&dq_data_lock); cur_space = inode_get_bytes(inode); - rsv_space = dquot_get_reserved_space(inode); + rsv_space = inode_get_rsv_space(inode); space = cur_space + rsv_space; /* Build the transfer_from list and check the limits */ for (cnt = 0; cnt < MAXQUOTAS; cnt++) { @@ -1709,25 +1740,18 @@ int dquot_transfer(struct inode *inode, struct iattr *iattr) spin_unlock(&dq_data_lock); up_write(&sb_dqopt(inode->i_sb)->dqptr_sem); - /* Dirtify all the dquots - this can block when journalling */ - for (cnt = 0; cnt < MAXQUOTAS; cnt++) { - if (transfer_from[cnt]) - mark_dquot_dirty(transfer_from[cnt]); - if (transfer_to[cnt]) { - mark_dquot_dirty(transfer_to[cnt]); - /* The reference we got is transferred to the inode */ - transfer_to[cnt] = NULL; - } - } + mark_all_dquot_dirty(transfer_from); + mark_all_dquot_dirty(transfer_to); + /* The reference we got is transferred to the inode */ + for (cnt = 0; cnt < MAXQUOTAS; cnt++) + transfer_to[cnt] = NULL; warn_put_all: flush_warnings(transfer_to, warntype_to); flush_warnings(transfer_from, warntype_from_inodes); flush_warnings(transfer_from, warntype_from_space); put_all: - for (cnt = 0; cnt < MAXQUOTAS; cnt++) { - dqput(transfer_from[cnt]); - dqput(transfer_to[cnt]); - } + dqput_all(transfer_from); + dqput_all(transfer_to); return ret; over_quota: spin_unlock(&dq_data_lock); diff --git a/fs/quota/quota_v2.c b/fs/quota/quota_v2.c index 3dfc23e0213..e3da02f4986 100644 --- a/fs/quota/quota_v2.c +++ b/fs/quota/quota_v2.c @@ -97,8 +97,11 @@ static int v2_read_file_info(struct super_block *sb, int type) unsigned int version; if (!v2_read_header(sb, type, &dqhead)) - return 0; + return -1; version = le32_to_cpu(dqhead.dqh_version); + if ((info->dqi_fmt_id == QFMT_VFS_V0 && version != 0) || + (info->dqi_fmt_id == QFMT_VFS_V1 && version != 1)) + return -1; size = sb->s_op->quota_read(sb, type, (char *)&dinfo, sizeof(struct v2_disk_dqinfo), V2_DQINFOOFF); @@ -120,8 +123,8 @@ static int v2_read_file_info(struct super_block *sb, int type) info->dqi_maxilimit = 0xffffffff; } else { /* used space is stored as unsigned 64-bit value */ - info->dqi_maxblimit = 0xffffffffffffffff; /* 2^64-1 */ - info->dqi_maxilimit = 0xffffffffffffffff; + info->dqi_maxblimit = 0xffffffffffffffffULL; /* 2^64-1 */ + info->dqi_maxilimit = 0xffffffffffffffffULL; } info->dqi_bgrace = le32_to_cpu(dinfo.dqi_bgrace); info->dqi_igrace = le32_to_cpu(dinfo.dqi_igrace); diff --git a/fs/reiserfs/bitmap.c b/fs/reiserfs/bitmap.c index 68549570718..65c87276117 100644 --- a/fs/reiserfs/bitmap.c +++ b/fs/reiserfs/bitmap.c @@ -1277,7 +1277,10 @@ int reiserfs_init_bitmap_cache(struct super_block *sb) struct reiserfs_bitmap_info *bitmap; unsigned int bmap_nr = reiserfs_bmap_count(sb); + /* Avoid lock recursion in fault case */ + reiserfs_write_unlock(sb); bitmap = vmalloc(sizeof(*bitmap) * bmap_nr); + reiserfs_write_lock(sb); if (bitmap == NULL) return -ENOMEM; diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c index 290ae38fca8..1150ebb2536 100644 --- a/fs/reiserfs/inode.c +++ b/fs/reiserfs/inode.c @@ -31,11 +31,12 @@ void reiserfs_delete_inode(struct inode *inode) JOURNAL_PER_BALANCE_CNT * 2 + 2 * REISERFS_QUOTA_INIT_BLOCKS(inode->i_sb); struct reiserfs_transaction_handle th; + int depth; int err; truncate_inode_pages(&inode->i_data, 0); - reiserfs_write_lock(inode->i_sb); + depth = reiserfs_write_lock_once(inode->i_sb); /* The = 0 happens when we abort creating a new inode for some reason like lack of space.. */ if (!(inode->i_state & I_NEW) && INODE_PKEY(inode)->k_objectid != 0) { /* also handles bad_inode case */ @@ -74,7 +75,7 @@ void reiserfs_delete_inode(struct inode *inode) out: clear_inode(inode); /* note this must go after the journal_end to prevent deadlock */ inode->i_blocks = 0; - reiserfs_write_unlock(inode->i_sb); + reiserfs_write_unlock_once(inode->i_sb, depth); } static void _make_cpu_key(struct cpu_key *key, int version, __u32 dirid, diff --git a/fs/reiserfs/journal.c b/fs/reiserfs/journal.c index 2f8a7e7b8da..83ac4d3b3cb 100644 --- a/fs/reiserfs/journal.c +++ b/fs/reiserfs/journal.c @@ -2009,10 +2009,11 @@ static int do_journal_release(struct reiserfs_transaction_handle *th, destroy_workqueue(commit_wq); commit_wq = NULL; } - reiserfs_write_lock(sb); free_journal_ram(sb); + reiserfs_write_lock(sb); + return 0; } @@ -2758,11 +2759,18 @@ int journal_init(struct super_block *sb, const char *j_dev_name, struct reiserfs_journal *journal; struct reiserfs_journal_list *jl; char b[BDEVNAME_SIZE]; + int ret; + /* + * Unlock here to avoid various RECLAIM-FS-ON <-> IN-RECLAIM-FS + * dependency inversion warnings. + */ + reiserfs_write_unlock(sb); journal = SB_JOURNAL(sb) = vmalloc(sizeof(struct reiserfs_journal)); if (!journal) { reiserfs_warning(sb, "journal-1256", "unable to get memory for journal structure"); + reiserfs_write_lock(sb); return 1; } memset(journal, 0, sizeof(struct reiserfs_journal)); @@ -2771,10 +2779,12 @@ int journal_init(struct super_block *sb, const char *j_dev_name, INIT_LIST_HEAD(&journal->j_working_list); INIT_LIST_HEAD(&journal->j_journal_list); journal->j_persistent_trans = 0; - if (reiserfs_allocate_list_bitmaps(sb, - journal->j_list_bitmap, - reiserfs_bmap_count(sb))) + ret = reiserfs_allocate_list_bitmaps(sb, journal->j_list_bitmap, + reiserfs_bmap_count(sb)); + reiserfs_write_lock(sb); + if (ret) goto free_and_return; + allocate_bitmap_nodes(sb); /* reserved for journal area support */ diff --git a/fs/reiserfs/lock.c b/fs/reiserfs/lock.c index ee2cfc0fd8a..b87aa2c1afc 100644 --- a/fs/reiserfs/lock.c +++ b/fs/reiserfs/lock.c @@ -86,3 +86,12 @@ void reiserfs_check_lock_depth(struct super_block *sb, char *caller) reiserfs_panic(sb, "%s called without kernel lock held %d", caller); } + +#ifdef CONFIG_REISERFS_CHECK +void reiserfs_lock_check_recursive(struct super_block *sb) +{ + struct reiserfs_sb_info *sb_i = REISERFS_SB(sb); + + WARN_ONCE((sb_i->lock_depth > 0), "Unwanted recursive reiserfs lock!\n"); +} +#endif diff --git a/fs/reiserfs/namei.c b/fs/reiserfs/namei.c index e296ff72a6c..9d4dcf0b07c 100644 --- a/fs/reiserfs/namei.c +++ b/fs/reiserfs/namei.c @@ -921,6 +921,7 @@ static int reiserfs_unlink(struct inode *dir, struct dentry *dentry) struct reiserfs_transaction_handle th; int jbegin_count; unsigned long savelink; + int depth; inode = dentry->d_inode; @@ -932,7 +933,7 @@ static int reiserfs_unlink(struct inode *dir, struct dentry *dentry) JOURNAL_PER_BALANCE_CNT * 2 + 2 + 4 * REISERFS_QUOTA_TRANS_BLOCKS(dir->i_sb); - reiserfs_write_lock(dir->i_sb); + depth = reiserfs_write_lock_once(dir->i_sb); retval = journal_begin(&th, dir->i_sb, jbegin_count); if (retval) goto out_unlink; @@ -993,7 +994,7 @@ static int reiserfs_unlink(struct inode *dir, struct dentry *dentry) retval = journal_end(&th, dir->i_sb, jbegin_count); reiserfs_check_path(&path); - reiserfs_write_unlock(dir->i_sb); + reiserfs_write_unlock_once(dir->i_sb, depth); return retval; end_unlink: @@ -1003,7 +1004,7 @@ static int reiserfs_unlink(struct inode *dir, struct dentry *dentry) if (err) retval = err; out_unlink: - reiserfs_write_unlock(dir->i_sb); + reiserfs_write_unlock_once(dir->i_sb, depth); return retval; } diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c index 8c7033a8b67..c3b004ee627 100644 --- a/fs/reiserfs/xattr.c +++ b/fs/reiserfs/xattr.c @@ -83,7 +83,8 @@ static int xattr_unlink(struct inode *dir, struct dentry *dentry) BUG_ON(!mutex_is_locked(&dir->i_mutex)); vfs_dq_init(dir); - mutex_lock_nested(&dentry->d_inode->i_mutex, I_MUTEX_CHILD); + reiserfs_mutex_lock_nested_safe(&dentry->d_inode->i_mutex, + I_MUTEX_CHILD, dir->i_sb); error = dir->i_op->unlink(dir, dentry); mutex_unlock(&dentry->d_inode->i_mutex); @@ -98,7 +99,8 @@ static int xattr_rmdir(struct inode *dir, struct dentry *dentry) BUG_ON(!mutex_is_locked(&dir->i_mutex)); vfs_dq_init(dir); - mutex_lock_nested(&dentry->d_inode->i_mutex, I_MUTEX_CHILD); + reiserfs_mutex_lock_nested_safe(&dentry->d_inode->i_mutex, + I_MUTEX_CHILD, dir->i_sb); dentry_unhash(dentry); error = dir->i_op->rmdir(dir, dentry); if (!error) @@ -235,16 +237,22 @@ static int reiserfs_for_each_xattr(struct inode *inode, if (IS_PRIVATE(inode) || get_inode_sd_version(inode) == STAT_DATA_V1) return 0; + reiserfs_write_unlock(inode->i_sb); dir = open_xa_dir(inode, XATTR_REPLACE); if (IS_ERR(dir)) { err = PTR_ERR(dir); + reiserfs_write_lock(inode->i_sb); goto out; } else if (!dir->d_inode) { err = 0; + reiserfs_write_lock(inode->i_sb); goto out_dir; } mutex_lock_nested(&dir->d_inode->i_mutex, I_MUTEX_XATTR); + + reiserfs_write_lock(inode->i_sb); + buf.xadir = dir; err = reiserfs_readdir_dentry(dir, &buf, fill_with_dentries, &pos); while ((err == 0 || err == -ENOSPC) && buf.count) { @@ -283,8 +291,9 @@ static int reiserfs_for_each_xattr(struct inode *inode, err = journal_begin(&th, inode->i_sb, blocks); if (!err) { int jerror; - mutex_lock_nested(&dir->d_parent->d_inode->i_mutex, - I_MUTEX_XATTR); + reiserfs_mutex_lock_nested_safe( + &dir->d_parent->d_inode->i_mutex, + I_MUTEX_XATTR, inode->i_sb); err = action(dir, data); jerror = journal_end(&th, inode->i_sb, blocks); mutex_unlock(&dir->d_parent->d_inode->i_mutex); @@ -480,11 +489,16 @@ reiserfs_xattr_set_handle(struct reiserfs_transaction_handle *th, if (!buffer) return lookup_and_delete_xattr(inode, name); + reiserfs_write_unlock(inode->i_sb); dentry = xattr_lookup(inode, name, flags); - if (IS_ERR(dentry)) + if (IS_ERR(dentry)) { + reiserfs_write_lock(inode->i_sb); return PTR_ERR(dentry); + } - down_write(&REISERFS_I(inode)->i_xattr_sem); + down_read(&REISERFS_I(inode)->i_xattr_sem); + + reiserfs_write_lock(inode->i_sb); xahash = xattr_hash(buffer, buffer_size); while (buffer_pos < buffer_size || buffer_pos == 0) { diff --git a/fs/signalfd.c b/fs/signalfd.c index b07565c9438..1dabe4ee02f 100644 --- a/fs/signalfd.c +++ b/fs/signalfd.c @@ -236,7 +236,7 @@ SYSCALL_DEFINE4(signalfd4, int, ufd, sigset_t __user *, user_mask, * anon_inode_getfd() will install the fd. */ ufd = anon_inode_getfd("[signalfd]", &signalfd_fops, ctx, - flags & (O_CLOEXEC | O_NONBLOCK)); + O_RDWR | (flags & (O_CLOEXEC | O_NONBLOCK))); if (ufd < 0) kfree(ctx); } else { diff --git a/fs/stat.c b/fs/stat.c index 075694e31d8..c4ecd52c573 100644 --- a/fs/stat.c +++ b/fs/stat.c @@ -401,9 +401,9 @@ SYSCALL_DEFINE4(fstatat64, int, dfd, char __user *, filename, } #endif /* __ARCH_WANT_STAT64 */ -void inode_add_bytes(struct inode *inode, loff_t bytes) +/* Caller is here responsible for sufficient locking (ie. inode->i_lock) */ +void __inode_add_bytes(struct inode *inode, loff_t bytes) { - spin_lock(&inode->i_lock); inode->i_blocks += bytes >> 9; bytes &= 511; inode->i_bytes += bytes; @@ -411,6 +411,12 @@ void inode_add_bytes(struct inode *inode, loff_t bytes) inode->i_blocks++; inode->i_bytes -= 512; } +} + +void inode_add_bytes(struct inode *inode, loff_t bytes) +{ + spin_lock(&inode->i_lock); + __inode_add_bytes(inode, bytes); spin_unlock(&inode->i_lock); } diff --git a/fs/super.c b/fs/super.c index 19eb70b374b..aff046b0fe7 100644 --- a/fs/super.c +++ b/fs/super.c @@ -901,8 +901,9 @@ int get_sb_single(struct file_system_type *fs_type, return error; } s->s_flags |= MS_ACTIVE; + } else { + do_remount_sb(s, flags, data, 0); } - do_remount_sb(s, flags, data, 0); simple_set_mnt(mnt, s); return 0; } diff --git a/fs/sysfs/bin.c b/fs/sysfs/bin.c index 60c702bc10a..a0a500af24a 100644 --- a/fs/sysfs/bin.c +++ b/fs/sysfs/bin.c @@ -483,7 +483,8 @@ void unmap_bin_file(struct sysfs_dirent *attr_sd) * @attr: attribute descriptor. */ -int sysfs_create_bin_file(struct kobject * kobj, struct bin_attribute * attr) +int sysfs_create_bin_file(struct kobject *kobj, + const struct bin_attribute *attr) { BUG_ON(!kobj || !kobj->sd || !attr); @@ -497,7 +498,8 @@ int sysfs_create_bin_file(struct kobject * kobj, struct bin_attribute * attr) * @attr: attribute descriptor. */ -void sysfs_remove_bin_file(struct kobject * kobj, struct bin_attribute * attr) +void sysfs_remove_bin_file(struct kobject *kobj, + const struct bin_attribute *attr) { sysfs_hash_and_remove(kobj->sd, attr->attr.name); } diff --git a/fs/sysfs/dir.c b/fs/sysfs/dir.c index f05f2303a8b..699f371b9f1 100644 --- a/fs/sysfs/dir.c +++ b/fs/sysfs/dir.c @@ -106,8 +106,10 @@ static struct sysfs_dirent *sysfs_get_active(struct sysfs_dirent *sd) return NULL; t = atomic_cmpxchg(&sd->s_active, v, v + 1); - if (likely(t == v)) + if (likely(t == v)) { + rwsem_acquire_read(&sd->dep_map, 0, 1, _RET_IP_); return sd; + } if (t < 0) return NULL; @@ -130,6 +132,7 @@ static void sysfs_put_active(struct sysfs_dirent *sd) if (unlikely(!sd)) return; + rwsem_release(&sd->dep_map, 1, _RET_IP_); v = atomic_dec_return(&sd->s_active); if (likely(v != SD_DEACTIVATED_BIAS)) return; @@ -194,15 +197,21 @@ static void sysfs_deactivate(struct sysfs_dirent *sd) BUG_ON(sd->s_sibling || !(sd->s_flags & SYSFS_FLAG_REMOVED)); sd->s_sibling = (void *)&wait; + rwsem_acquire(&sd->dep_map, 0, 0, _RET_IP_); /* atomic_add_return() is a mb(), put_active() will always see * the updated sd->s_sibling. */ v = atomic_add_return(SD_DEACTIVATED_BIAS, &sd->s_active); - if (v != SD_DEACTIVATED_BIAS) + if (v != SD_DEACTIVATED_BIAS) { + lock_contended(&sd->dep_map, _RET_IP_); wait_for_completion(&wait); + } sd->s_sibling = NULL; + + lock_acquired(&sd->dep_map, _RET_IP_); + rwsem_release(&sd->dep_map, 1, _RET_IP_); } static int sysfs_alloc_ino(ino_t *pino) @@ -345,6 +354,7 @@ struct sysfs_dirent *sysfs_new_dirent(const char *name, umode_t mode, int type) atomic_set(&sd->s_count, 1); atomic_set(&sd->s_active, 0); + sysfs_dirent_init_lockdep(sd); sd->s_name = name; sd->s_mode = mode; diff --git a/fs/sysfs/sysfs.h b/fs/sysfs/sysfs.h index ca52e7b9d8f..cdd9377a6e0 100644 --- a/fs/sysfs/sysfs.h +++ b/fs/sysfs/sysfs.h @@ -8,6 +8,7 @@ * This file is released under the GPLv2. */ +#include <linux/lockdep.h> #include <linux/fs.h> struct sysfs_open_dirent; @@ -50,6 +51,9 @@ struct sysfs_inode_attrs { struct sysfs_dirent { atomic_t s_count; atomic_t s_active; +#ifdef CONFIG_DEBUG_LOCK_ALLOC + struct lockdep_map dep_map; +#endif struct sysfs_dirent *s_parent; struct sysfs_dirent *s_sibling; const char *s_name; @@ -84,6 +88,17 @@ static inline unsigned int sysfs_type(struct sysfs_dirent *sd) return sd->s_flags & SYSFS_TYPE_MASK; } +#ifdef CONFIG_DEBUG_LOCK_ALLOC +#define sysfs_dirent_init_lockdep(sd) \ +do { \ + static struct lock_class_key __key; \ + \ + lockdep_init_map(&sd->dep_map, "s_active", &__key, 0); \ +} while(0) +#else +#define sysfs_dirent_init_lockdep(sd) do {} while(0) +#endif + /* * Context structure to be used while adding/removing nodes. */ diff --git a/fs/timerfd.c b/fs/timerfd.c index b042bd7034b..1bfc95ad5f7 100644 --- a/fs/timerfd.c +++ b/fs/timerfd.c @@ -200,7 +200,7 @@ SYSCALL_DEFINE2(timerfd_create, int, clockid, int, flags) hrtimer_init(&ctx->tmr, clockid, HRTIMER_MODE_ABS); ufd = anon_inode_getfd("[timerfd]", &timerfd_fops, ctx, - flags & TFD_SHARED_FCNTL_FLAGS); + O_RDWR | (flags & TFD_SHARED_FCNTL_FLAGS)); if (ufd < 0) kfree(ctx); |