diff options
author | David S. Miller <davem@davemloft.net> | 2009-01-08 22:05:59 +0300 |
---|---|---|
committer | David S. Miller <davem@davemloft.net> | 2009-01-08 22:05:59 +0300 |
commit | 7f46b1343f723f98634a5dcee47856b2000079ed (patch) | |
tree | ed22b6298c8dd2f687890a0d79abcd1d273b5f81 /fs | |
parent | b8c31da64165b8566fc6e1c9c826f76e7b98ff02 (diff) | |
parent | 9e42d0cf5020aaf217433cad1a224745241d212a (diff) | |
download | linux-7f46b1343f723f98634a5dcee47856b2000079ed.tar.xz |
Merge branch 'master' of master.kernel.org:/pub/scm/linux/kernel/git/torvalds/linux-2.6
Diffstat (limited to 'fs')
79 files changed, 3662 insertions, 1245 deletions
diff --git a/fs/Kconfig b/fs/Kconfig index f9b6e2979aaa..32883589ee54 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -721,7 +721,20 @@ config CONFIGFS_FS endmenu -menu "Miscellaneous filesystems" +menuconfig MISC_FILESYSTEMS + bool "Miscellaneous filesystems" + default y + ---help--- + Say Y here to get to see options for various miscellaneous + filesystems, such as filesystems that came from other + operating systems. + + This option alone does not add any kernel code. + + If you say N, all options in this submenu will be skipped and + disabled; if unsure, say Y here. + +if MISC_FILESYSTEMS config ADFS_FS tristate "ADFS file system support (EXPERIMENTAL)" @@ -1091,7 +1104,7 @@ config UFS_DEBUG Y here. This will result in _many_ additional debugging messages to be written to the system log. -endmenu +endif # MISC_FILESYSTEMS menuconfig NETWORK_FILESYSTEMS bool "Network File Systems" diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4/autofs_i.h index e0f16da00e54..a76803108d06 100644 --- a/fs/autofs4/autofs_i.h +++ b/fs/autofs4/autofs_i.h @@ -25,8 +25,6 @@ #define AUTOFS_DEV_IOCTL_IOC_FIRST (AUTOFS_DEV_IOCTL_VERSION) #define AUTOFS_DEV_IOCTL_IOC_COUNT (AUTOFS_IOC_COUNT - 11) -#define AUTOFS_TYPE_TRIGGER (AUTOFS_TYPE_DIRECT|AUTOFS_TYPE_OFFSET) - #include <linux/kernel.h> #include <linux/slab.h> #include <linux/time.h> diff --git a/fs/autofs4/dev-ioctl.c b/fs/autofs4/dev-ioctl.c index 63b7c7afe8df..025e105bffea 100644 --- a/fs/autofs4/dev-ioctl.c +++ b/fs/autofs4/dev-ioctl.c @@ -124,7 +124,7 @@ static inline void free_dev_ioctl(struct autofs_dev_ioctl *param) /* * Check sanity of parameter control fields and if a path is present - * check that it has a "/" and is terminated. + * check that it is terminated and contains at least one "/". */ static int validate_dev_ioctl(int cmd, struct autofs_dev_ioctl *param) { @@ -138,15 +138,16 @@ static int validate_dev_ioctl(int cmd, struct autofs_dev_ioctl *param) } if (param->size > sizeof(*param)) { - err = check_name(param->path); + err = invalid_str(param->path, + (void *) ((size_t) param + param->size)); if (err) { - AUTOFS_WARN("invalid path supplied for cmd(0x%08x)", - cmd); + AUTOFS_WARN( + "path string terminator missing for cmd(0x%08x)", + cmd); goto out; } - err = invalid_str(param->path, - (void *) ((size_t) param + param->size)); + err = check_name(param->path); if (err) { AUTOFS_WARN("invalid path supplied for cmd(0x%08x)", cmd); @@ -180,7 +181,7 @@ static int autofs_dev_ioctl_protover(struct file *fp, struct autofs_sb_info *sbi, struct autofs_dev_ioctl *param) { - param->arg1 = sbi->version; + param->protover.version = sbi->version; return 0; } @@ -189,7 +190,7 @@ static int autofs_dev_ioctl_protosubver(struct file *fp, struct autofs_sb_info *sbi, struct autofs_dev_ioctl *param) { - param->arg1 = sbi->sub_version; + param->protosubver.sub_version = sbi->sub_version; return 0; } @@ -335,13 +336,13 @@ static int autofs_dev_ioctl_openmount(struct file *fp, int err, fd; /* param->path has already been checked */ - if (!param->arg1) + if (!param->openmount.devid) return -EINVAL; param->ioctlfd = -1; path = param->path; - devid = param->arg1; + devid = param->openmount.devid; err = 0; fd = autofs_dev_ioctl_open_mountpoint(path, devid); @@ -373,7 +374,7 @@ static int autofs_dev_ioctl_ready(struct file *fp, { autofs_wqt_t token; - token = (autofs_wqt_t) param->arg1; + token = (autofs_wqt_t) param->ready.token; return autofs4_wait_release(sbi, token, 0); } @@ -388,8 +389,8 @@ static int autofs_dev_ioctl_fail(struct file *fp, autofs_wqt_t token; int status; - token = (autofs_wqt_t) param->arg1; - status = param->arg2 ? param->arg2 : -ENOENT; + token = (autofs_wqt_t) param->fail.token; + status = param->fail.status ? param->fail.status : -ENOENT; return autofs4_wait_release(sbi, token, status); } @@ -412,10 +413,10 @@ static int autofs_dev_ioctl_setpipefd(struct file *fp, int pipefd; int err = 0; - if (param->arg1 == -1) + if (param->setpipefd.pipefd == -1) return -EINVAL; - pipefd = param->arg1; + pipefd = param->setpipefd.pipefd; mutex_lock(&sbi->wq_mutex); if (!sbi->catatonic) { @@ -457,8 +458,8 @@ static int autofs_dev_ioctl_timeout(struct file *fp, { unsigned long timeout; - timeout = param->arg1; - param->arg1 = sbi->exp_timeout / HZ; + timeout = param->timeout.timeout; + param->timeout.timeout = sbi->exp_timeout / HZ; sbi->exp_timeout = timeout * HZ; return 0; } @@ -489,7 +490,7 @@ static int autofs_dev_ioctl_requester(struct file *fp, path = param->path; devid = sbi->sb->s_dev; - param->arg1 = param->arg2 = -1; + param->requester.uid = param->requester.gid = -1; /* Get nameidata of the parent directory */ err = path_lookup(path, LOOKUP_PARENT, &nd); @@ -505,8 +506,8 @@ static int autofs_dev_ioctl_requester(struct file *fp, err = 0; autofs4_expire_wait(nd.path.dentry); spin_lock(&sbi->fs_lock); - param->arg1 = ino->uid; - param->arg2 = ino->gid; + param->requester.uid = ino->uid; + param->requester.gid = ino->gid; spin_unlock(&sbi->fs_lock); } @@ -529,10 +530,10 @@ static int autofs_dev_ioctl_expire(struct file *fp, int err = -EAGAIN; int how; - how = param->arg1; + how = param->expire.how; mnt = fp->f_path.mnt; - if (sbi->type & AUTOFS_TYPE_TRIGGER) + if (autofs_type_trigger(sbi->type)) dentry = autofs4_expire_direct(sbi->sb, mnt, sbi, how); else dentry = autofs4_expire_indirect(sbi->sb, mnt, sbi, how); @@ -565,9 +566,9 @@ static int autofs_dev_ioctl_askumount(struct file *fp, struct autofs_sb_info *sbi, struct autofs_dev_ioctl *param) { - param->arg1 = 0; + param->askumount.may_umount = 0; if (may_umount(fp->f_path.mnt)) - param->arg1 = 1; + param->askumount.may_umount = 1; return 0; } @@ -600,6 +601,7 @@ static int autofs_dev_ioctl_ismountpoint(struct file *fp, struct nameidata nd; const char *path; unsigned int type; + unsigned int devid, magic; int err = -ENOENT; if (param->size <= sizeof(*param)) { @@ -608,13 +610,13 @@ static int autofs_dev_ioctl_ismountpoint(struct file *fp, } path = param->path; - type = param->arg1; + type = param->ismountpoint.in.type; - param->arg1 = 0; - param->arg2 = 0; + param->ismountpoint.out.devid = devid = 0; + param->ismountpoint.out.magic = magic = 0; if (!fp || param->ioctlfd == -1) { - if (type == AUTOFS_TYPE_ANY) { + if (autofs_type_any(type)) { struct super_block *sb; err = path_lookup(path, LOOKUP_FOLLOW, &nd); @@ -622,7 +624,7 @@ static int autofs_dev_ioctl_ismountpoint(struct file *fp, goto out; sb = nd.path.dentry->d_sb; - param->arg1 = new_encode_dev(sb->s_dev); + devid = new_encode_dev(sb->s_dev); } else { struct autofs_info *ino; @@ -635,38 +637,41 @@ static int autofs_dev_ioctl_ismountpoint(struct file *fp, goto out_release; ino = autofs4_dentry_ino(nd.path.dentry); - param->arg1 = autofs4_get_dev(ino->sbi); + devid = autofs4_get_dev(ino->sbi); } err = 0; if (nd.path.dentry->d_inode && nd.path.mnt->mnt_root == nd.path.dentry) { err = 1; - param->arg2 = nd.path.dentry->d_inode->i_sb->s_magic; + magic = nd.path.dentry->d_inode->i_sb->s_magic; } } else { - dev_t devid = new_encode_dev(sbi->sb->s_dev); + dev_t dev = autofs4_get_dev(sbi); err = path_lookup(path, LOOKUP_PARENT, &nd); if (err) goto out; - err = autofs_dev_ioctl_find_super(&nd, devid); + err = autofs_dev_ioctl_find_super(&nd, dev); if (err) goto out_release; - param->arg1 = autofs4_get_dev(sbi); + devid = dev; err = have_submounts(nd.path.dentry); if (nd.path.mnt->mnt_mountpoint != nd.path.mnt->mnt_root) { if (follow_down(&nd.path.mnt, &nd.path.dentry)) { struct inode *inode = nd.path.dentry->d_inode; - param->arg2 = inode->i_sb->s_magic; + magic = inode->i_sb->s_magic; } } } + param->ismountpoint.out.devid = devid; + param->ismountpoint.out.magic = magic; + out_release: path_put(&nd.path); out: diff --git a/fs/autofs4/expire.c b/fs/autofs4/expire.c index 4b6fb3f628c0..e3bd50776f9e 100644 --- a/fs/autofs4/expire.c +++ b/fs/autofs4/expire.c @@ -63,7 +63,7 @@ static int autofs4_mount_busy(struct vfsmount *mnt, struct dentry *dentry) struct autofs_sb_info *sbi = autofs4_sbi(dentry->d_sb); /* This is an autofs submount, we can't expire it */ - if (sbi->type == AUTOFS_TYPE_INDIRECT) + if (autofs_type_indirect(sbi->type)) goto done; /* @@ -490,7 +490,7 @@ int autofs4_expire_multi(struct super_block *sb, struct vfsmount *mnt, if (arg && get_user(do_now, arg)) return -EFAULT; - if (sbi->type & AUTOFS_TYPE_TRIGGER) + if (autofs_type_trigger(sbi->type)) dentry = autofs4_expire_direct(sb, mnt, sbi, do_now); else dentry = autofs4_expire_indirect(sb, mnt, sbi, do_now); diff --git a/fs/autofs4/inode.c b/fs/autofs4/inode.c index cfc23e53b6f4..716e12b627b2 100644 --- a/fs/autofs4/inode.c +++ b/fs/autofs4/inode.c @@ -197,9 +197,9 @@ static int autofs4_show_options(struct seq_file *m, struct vfsmount *mnt) seq_printf(m, ",minproto=%d", sbi->min_proto); seq_printf(m, ",maxproto=%d", sbi->max_proto); - if (sbi->type & AUTOFS_TYPE_OFFSET) + if (autofs_type_offset(sbi->type)) seq_printf(m, ",offset"); - else if (sbi->type & AUTOFS_TYPE_DIRECT) + else if (autofs_type_direct(sbi->type)) seq_printf(m, ",direct"); else seq_printf(m, ",indirect"); @@ -284,13 +284,13 @@ static int parse_options(char *options, int *pipefd, uid_t *uid, gid_t *gid, *maxproto = option; break; case Opt_indirect: - *type = AUTOFS_TYPE_INDIRECT; + set_autofs_type_indirect(type); break; case Opt_direct: - *type = AUTOFS_TYPE_DIRECT; + set_autofs_type_direct(type); break; case Opt_offset: - *type = AUTOFS_TYPE_OFFSET; + set_autofs_type_offset(type); break; default: return 1; @@ -338,7 +338,7 @@ int autofs4_fill_super(struct super_block *s, void *data, int silent) sbi->sb = s; sbi->version = 0; sbi->sub_version = 0; - sbi->type = AUTOFS_TYPE_INDIRECT; + set_autofs_type_indirect(&sbi->type); sbi->min_proto = 0; sbi->max_proto = 0; mutex_init(&sbi->wq_mutex); @@ -380,7 +380,7 @@ int autofs4_fill_super(struct super_block *s, void *data, int silent) } root_inode->i_fop = &autofs4_root_operations; - root_inode->i_op = sbi->type & AUTOFS_TYPE_TRIGGER ? + root_inode->i_op = autofs_type_trigger(sbi->type) ? &autofs4_direct_root_inode_operations : &autofs4_indirect_root_inode_operations; diff --git a/fs/autofs4/waitq.c b/fs/autofs4/waitq.c index e02cc8ae5eb3..eeb246845909 100644 --- a/fs/autofs4/waitq.c +++ b/fs/autofs4/waitq.c @@ -337,7 +337,7 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry, * is very similar for indirect mounts except only dentrys * in the root of the autofs file system may be negative. */ - if (sbi->type & AUTOFS_TYPE_TRIGGER) + if (autofs_type_trigger(sbi->type)) return -ENOENT; else if (!IS_ROOT(dentry->d_parent)) return -ENOENT; @@ -348,7 +348,7 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry, return -ENOMEM; /* If this is a direct mount request create a dummy name */ - if (IS_ROOT(dentry) && sbi->type & AUTOFS_TYPE_TRIGGER) + if (IS_ROOT(dentry) && autofs_type_trigger(sbi->type)) qstr.len = sprintf(name, "%p", dentry); else { qstr.len = autofs4_getpath(sbi, dentry, &name); @@ -406,11 +406,11 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry, type = autofs_ptype_expire_multi; } else { if (notify == NFY_MOUNT) - type = (sbi->type & AUTOFS_TYPE_TRIGGER) ? + type = autofs_type_trigger(sbi->type) ? autofs_ptype_missing_direct : autofs_ptype_missing_indirect; else - type = (sbi->type & AUTOFS_TYPE_TRIGGER) ? + type = autofs_type_trigger(sbi->type) ? autofs_ptype_expire_direct : autofs_ptype_expire_indirect; } diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c index 0ed57b5ee012..cc4062d12ca2 100644 --- a/fs/bfs/inode.c +++ b/fs/bfs/inode.c @@ -213,6 +213,9 @@ static void bfs_put_super(struct super_block *s) { struct bfs_sb_info *info = BFS_SB(s); + if (!info) + return; + brelse(info->si_sbh); mutex_destroy(&info->bfs_lock); kfree(info->si_imap); @@ -327,6 +330,7 @@ static int bfs_fill_super(struct super_block *s, void *data, int silent) unsigned i, imap_len; struct bfs_sb_info *info; long ret = -EINVAL; + unsigned long i_sblock, i_eblock, i_eoff, s_size; info = kzalloc(sizeof(*info), GFP_KERNEL); if (!info) @@ -350,6 +354,12 @@ static int bfs_fill_super(struct super_block *s, void *data, int silent) s->s_magic = BFS_MAGIC; info->si_sbh = bh; + + if (le32_to_cpu(bfs_sb->s_start) > le32_to_cpu(bfs_sb->s_end)) { + printf("Superblock is corrupted\n"); + goto out; + } + info->si_lasti = (le32_to_cpu(bfs_sb->s_start) - BFS_BSIZE) / sizeof(struct bfs_inode) + BFS_ROOT_INO - 1; @@ -380,6 +390,18 @@ static int bfs_fill_super(struct super_block *s, void *data, int silent) - le32_to_cpu(bfs_sb->s_start)) >> BFS_BSIZE_BITS; info->si_freei = 0; info->si_lf_eblk = 0; + + /* can we read the last block? */ + bh = sb_bread(s, info->si_blocks - 1); + if (!bh) { + printf("Last block not available: %lu\n", info->si_blocks - 1); + iput(inode); + ret = -EIO; + kfree(info->si_imap); + goto out; + } + brelse(bh); + bh = NULL; for (i = BFS_ROOT_INO; i <= info->si_lasti; i++) { struct bfs_inode *di; @@ -397,6 +419,29 @@ static int bfs_fill_super(struct super_block *s, void *data, int silent) di = (struct bfs_inode *)bh->b_data + off; + /* test if filesystem is not corrupted */ + + i_eoff = le32_to_cpu(di->i_eoffset); + i_sblock = le32_to_cpu(di->i_sblock); + i_eblock = le32_to_cpu(di->i_eblock); + s_size = le32_to_cpu(bfs_sb->s_end); + + if (i_sblock > info->si_blocks || + i_eblock > info->si_blocks || + i_sblock > i_eblock || + i_eoff > s_size || + i_sblock * BFS_BSIZE > i_eoff) { + + printf("Inode 0x%08x corrupted\n", i); + + brelse(bh); + s->s_root = NULL; + kfree(info->si_imap); + kfree(info); + s->s_fs_info = NULL; + return -EIO; + } + if (!di->i_ino) { info->si_freei++; continue; diff --git a/fs/binfmt_misc.c b/fs/binfmt_misc.c index e1158cb4fbd6..c4e83537ead7 100644 --- a/fs/binfmt_misc.c +++ b/fs/binfmt_misc.c @@ -649,7 +649,7 @@ static const struct file_operations bm_register_operations = { static ssize_t bm_status_read(struct file *file, char __user *buf, size_t nbytes, loff_t *ppos) { - char *s = enabled ? "enabled" : "disabled"; + char *s = enabled ? "enabled\n" : "disabled\n"; return simple_read_from_buffer(buf, nbytes, ppos, s, strlen(s)); } diff --git a/fs/block_dev.c b/fs/block_dev.c index 349a26c10001..b957717e25ab 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -1262,7 +1262,7 @@ EXPORT_SYMBOL(ioctl_by_bdev); /** * lookup_bdev - lookup a struct block_device by name - * @path: special file representing the block device + * @pathname: special file representing the block device * * Get a reference to the blockdevice at @pathname in the current * namespace if possible and return it. Return ERR_PTR(error) diff --git a/fs/buffer.c b/fs/buffer.c index a13f09b696f7..c26da785938a 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -2022,7 +2022,6 @@ int block_write_begin(struct file *file, struct address_space *mapping, if (pos + len > inode->i_size) vmtruncate(inode, inode->i_size); } - goto out; } out: diff --git a/fs/char_dev.c b/fs/char_dev.c index 700697a72618..38f71222a552 100644 --- a/fs/char_dev.c +++ b/fs/char_dev.c @@ -120,7 +120,7 @@ __register_chrdev_region(unsigned int major, unsigned int baseminor, cd->major = major; cd->baseminor = baseminor; cd->minorct = minorct; - strncpy(cd->name,name, 64); + strlcpy(cd->name, name, sizeof(cd->name)); i = major_to_index(major); diff --git a/fs/compat.c b/fs/compat.c index d1ece79b6411..30f2faa22f5c 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -1187,6 +1187,9 @@ compat_sys_readv(unsigned long fd, const struct compat_iovec __user *vec, unsign ret = compat_do_readv_writev(READ, file, vec, vlen, &file->f_pos); out: + if (ret > 0) + add_rchar(current, ret); + inc_syscr(current); fput(file); return ret; } @@ -1210,6 +1213,9 @@ compat_sys_writev(unsigned long fd, const struct compat_iovec __user *vec, unsig ret = compat_do_readv_writev(WRITE, file, vec, vlen, &file->f_pos); out: + if (ret > 0) + add_wchar(current, ret); + inc_syscw(current); fput(file); return ret; } diff --git a/fs/debugfs/file.c b/fs/debugfs/file.c index 159a5efd6a8a..33a90120f6ad 100644 --- a/fs/debugfs/file.c +++ b/fs/debugfs/file.c @@ -294,6 +294,38 @@ struct dentry *debugfs_create_x32(const char *name, mode_t mode, } EXPORT_SYMBOL_GPL(debugfs_create_x32); + +static int debugfs_size_t_set(void *data, u64 val) +{ + *(size_t *)data = val; + return 0; +} +static int debugfs_size_t_get(void *data, u64 *val) +{ + *val = *(size_t *)data; + return 0; +} +DEFINE_SIMPLE_ATTRIBUTE(fops_size_t, debugfs_size_t_get, debugfs_size_t_set, + "%llu\n"); /* %llu and %zu are more or less the same */ + +/** + * debugfs_create_size_t - create a debugfs file that is used to read and write an size_t value + * @name: a pointer to a string containing the name of the file to create. + * @mode: the permission that the file should have + * @parent: a pointer to the parent dentry for this file. This should be a + * directory dentry if set. If this parameter is %NULL, then the + * file will be created in the root of the debugfs filesystem. + * @value: a pointer to the variable that the file should read to and write + * from. + */ +struct dentry *debugfs_create_size_t(const char *name, mode_t mode, + struct dentry *parent, size_t *value) +{ + return debugfs_create_file(name, mode, parent, value, &fops_size_t); +} +EXPORT_SYMBOL_GPL(debugfs_create_size_t); + + static ssize_t read_file_bool(struct file *file, char __user *user_buf, size_t count, loff_t *ppos) { diff --git a/fs/direct-io.c b/fs/direct-io.c index af0558dbe8b7..b6d43908ff7a 100644 --- a/fs/direct-io.c +++ b/fs/direct-io.c @@ -1209,6 +1209,19 @@ __blockdev_direct_IO(int rw, struct kiocb *iocb, struct inode *inode, retval = direct_io_worker(rw, iocb, inode, iov, offset, nr_segs, blkbits, get_block, end_io, dio); + /* + * In case of error extending write may have instantiated a few + * blocks outside i_size. Trim these off again for DIO_LOCKING. + * NOTE: DIO_NO_LOCK/DIO_OWN_LOCK callers have to handle this by + * it's own meaner. + */ + if (unlikely(retval < 0 && (rw & WRITE))) { + loff_t isize = i_size_read(inode); + + if (end > isize && dio_lock_type == DIO_LOCKING) + vmtruncate(inode, isize); + } + if (rw == READ && dio_lock_type == DIO_LOCKING) release_i_mutex = 0; diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c index 6046239465a1..c01e043670e2 100644 --- a/fs/ecryptfs/crypto.c +++ b/fs/ecryptfs/crypto.c @@ -175,8 +175,8 @@ out: * * Returns zero on success; non-zero on error. */ -static int ecryptfs_derive_iv(char *iv, struct ecryptfs_crypt_stat *crypt_stat, - loff_t offset) +int ecryptfs_derive_iv(char *iv, struct ecryptfs_crypt_stat *crypt_stat, + loff_t offset) { int rc = 0; char dst[MD5_DIGEST_SIZE]; @@ -924,6 +924,15 @@ static void ecryptfs_copy_mount_wide_flags_to_inode_flags( crypt_stat->flags |= ECRYPTFS_METADATA_IN_XATTR; if (mount_crypt_stat->flags & ECRYPTFS_ENCRYPTED_VIEW_ENABLED) crypt_stat->flags |= ECRYPTFS_VIEW_AS_ENCRYPTED; + if (mount_crypt_stat->flags & ECRYPTFS_GLOBAL_ENCRYPT_FILENAMES) { + crypt_stat->flags |= ECRYPTFS_ENCRYPT_FILENAMES; + if (mount_crypt_stat->flags + & ECRYPTFS_GLOBAL_ENCFN_USE_MOUNT_FNEK) + crypt_stat->flags |= ECRYPTFS_ENCFN_USE_MOUNT_FNEK; + else if (mount_crypt_stat->flags + & ECRYPTFS_GLOBAL_ENCFN_USE_FEK) + crypt_stat->flags |= ECRYPTFS_ENCFN_USE_FEK; + } } static int ecryptfs_copy_mount_wide_sigs_to_inode_sigs( @@ -1060,7 +1069,8 @@ struct ecryptfs_flag_map_elem { static struct ecryptfs_flag_map_elem ecryptfs_flag_map[] = { {0x00000001, ECRYPTFS_ENABLE_HMAC}, {0x00000002, ECRYPTFS_ENCRYPTED}, - {0x00000004, ECRYPTFS_METADATA_IN_XATTR} + {0x00000004, ECRYPTFS_METADATA_IN_XATTR}, + {0x00000008, ECRYPTFS_ENCRYPT_FILENAMES} }; /** @@ -1149,19 +1159,20 @@ ecryptfs_cipher_code_str_map[] = { /** * ecryptfs_code_for_cipher_string - * @crypt_stat: The cryptographic context + * @cipher_name: The string alias for the cipher + * @key_bytes: Length of key in bytes; used for AES code selection * * Returns zero on no match, or the cipher code on match */ -u8 ecryptfs_code_for_cipher_string(struct ecryptfs_crypt_stat *crypt_stat) +u8 ecryptfs_code_for_cipher_string(char *cipher_name, size_t key_bytes) { int i; u8 code = 0; struct ecryptfs_cipher_code_str_map_elem *map = ecryptfs_cipher_code_str_map; - if (strcmp(crypt_stat->cipher, "aes") == 0) { - switch (crypt_stat->key_size) { + if (strcmp(cipher_name, "aes") == 0) { + switch (key_bytes) { case 16: code = RFC2440_CIPHER_AES_128; break; @@ -1173,7 +1184,7 @@ u8 ecryptfs_code_for_cipher_string(struct ecryptfs_crypt_stat *crypt_stat) } } else { for (i = 0; i < ARRAY_SIZE(ecryptfs_cipher_code_str_map); i++) - if (strcmp(crypt_stat->cipher, map[i].cipher_str) == 0){ + if (strcmp(cipher_name, map[i].cipher_str) == 0) { code = map[i].cipher_code; break; } @@ -1212,6 +1223,8 @@ int ecryptfs_read_and_validate_header_region(char *data, &(ecryptfs_inode_to_private(ecryptfs_inode)->crypt_stat); int rc; + if (crypt_stat->extent_size == 0) + crypt_stat->extent_size = ECRYPTFS_DEFAULT_EXTENT_SIZE; rc = ecryptfs_read_lower(data, 0, crypt_stat->extent_size, ecryptfs_inode); if (rc) { @@ -1221,7 +1234,6 @@ int ecryptfs_read_and_validate_header_region(char *data, } if (!contains_ecryptfs_marker(data + ECRYPTFS_FILE_SIZE_BYTES)) { rc = -EINVAL; - ecryptfs_printk(KERN_DEBUG, "Valid marker not found\n"); } out: return rc; @@ -1628,95 +1640,95 @@ out: } /** - * ecryptfs_encode_filename - converts a plaintext file name to cipher text - * @crypt_stat: The crypt_stat struct associated with the file anem to encode - * @name: The plaintext name - * @length: The length of the plaintext - * @encoded_name: The encypted name + * ecryptfs_encrypt_filename - encrypt filename * - * Encrypts and encodes a filename into something that constitutes a - * valid filename for a filesystem, with printable characters. + * CBC-encrypts the filename. We do not want to encrypt the same + * filename with the same key and IV, which may happen with hard + * links, so we prepend random bits to each filename. * - * We assume that we have a properly initialized crypto context, - * pointed to by crypt_stat->tfm. - * - * TODO: Implement filename decoding and decryption here, in place of - * memcpy. We are keeping the framework around for now to (1) - * facilitate testing of the components needed to implement filename - * encryption and (2) to provide a code base from which other - * developers in the community can easily implement this feature. - * - * Returns the length of encoded filename; negative if error + * Returns zero on success; non-zero otherwise */ -int -ecryptfs_encode_filename(struct ecryptfs_crypt_stat *crypt_stat, - const char *name, int length, char **encoded_name) +static int +ecryptfs_encrypt_filename(struct ecryptfs_filename *filename, + struct ecryptfs_crypt_stat *crypt_stat, + struct ecryptfs_mount_crypt_stat *mount_crypt_stat) { - int error = 0; + int rc = 0; - (*encoded_name) = kmalloc(length + 2, GFP_KERNEL); - if (!(*encoded_name)) { - error = -ENOMEM; + filename->encrypted_filename = NULL; + filename->encrypted_filename_size = 0; + if ((crypt_stat && (crypt_stat->flags & ECRYPTFS_ENCFN_USE_MOUNT_FNEK)) + || (mount_crypt_stat && (mount_crypt_stat->flags + & ECRYPTFS_GLOBAL_ENCFN_USE_MOUNT_FNEK))) { + size_t packet_size; + size_t remaining_bytes; + + rc = ecryptfs_write_tag_70_packet( + NULL, NULL, + &filename->encrypted_filename_size, + mount_crypt_stat, NULL, + filename->filename_size); + if (rc) { + printk(KERN_ERR "%s: Error attempting to get packet " + "size for tag 72; rc = [%d]\n", __func__, + rc); + filename->encrypted_filename_size = 0; + goto out; + } + filename->encrypted_filename = + kmalloc(filename->encrypted_filename_size, GFP_KERNEL); + if (!filename->encrypted_filename) { + printk(KERN_ERR "%s: Out of memory whilst attempting " + "to kmalloc [%zd] bytes\n", __func__, + filename->encrypted_filename_size); + rc = -ENOMEM; + goto out; + } + remaining_bytes = filename->encrypted_filename_size; + rc = ecryptfs_write_tag_70_packet(filename->encrypted_filename, + &remaining_bytes, + &packet_size, + mount_crypt_stat, + filename->filename, + filename->filename_size); + if (rc) { + printk(KERN_ERR "%s: Error attempting to generate " + "tag 70 packet; rc = [%d]\n", __func__, + rc); + kfree(filename->encrypted_filename); + filename->encrypted_filename = NULL; + filename->encrypted_filename_size = 0; + goto out; + } + filename->encrypted_filename_size = packet_size; + } else { + printk(KERN_ERR "%s: No support for requested filename " + "encryption method in this release\n", __func__); + rc = -ENOTSUPP; goto out; } - /* TODO: Filename encryption is a scheduled feature for a - * future version of eCryptfs. This function is here only for - * the purpose of providing a framework for other developers - * to easily implement filename encryption. Hint: Replace this - * memcpy() with a call to encrypt and encode the - * filename, the set the length accordingly. */ - memcpy((void *)(*encoded_name), (void *)name, length); - (*encoded_name)[length] = '\0'; - error = length + 1; out: - return error; + return rc; } -/** - * ecryptfs_decode_filename - converts the cipher text name to plaintext - * @crypt_stat: The crypt_stat struct associated with the file - * @name: The filename in cipher text - * @length: The length of the cipher text name - * @decrypted_name: The plaintext name - * - * Decodes and decrypts the filename. - * - * We assume that we have a properly initialized crypto context, - * pointed to by crypt_stat->tfm. - * - * TODO: Implement filename decoding and decryption here, in place of - * memcpy. We are keeping the framework around for now to (1) - * facilitate testing of the components needed to implement filename - * encryption and (2) to provide a code base from which other - * developers in the community can easily implement this feature. - * - * Returns the length of decoded filename; negative if error - */ -int -ecryptfs_decode_filename(struct ecryptfs_crypt_stat *crypt_stat, - const char *name, int length, char **decrypted_name) +static int ecryptfs_copy_filename(char **copied_name, size_t *copied_name_size, + const char *name, size_t name_size) { - int error = 0; + int rc = 0; - (*decrypted_name) = kmalloc(length + 2, GFP_KERNEL); - if (!(*decrypted_name)) { - error = -ENOMEM; + (*copied_name) = kmalloc((name_size + 2), GFP_KERNEL); + if (!(*copied_name)) { + rc = -ENOMEM; goto out; } - /* TODO: Filename encryption is a scheduled feature for a - * future version of eCryptfs. This function is here only for - * the purpose of providing a framework for other developers - * to easily implement filename encryption. Hint: Replace this - * memcpy() with a call to decode and decrypt the - * filename, the set the length accordingly. */ - memcpy((void *)(*decrypted_name), (void *)name, length); - (*decrypted_name)[length + 1] = '\0'; /* Only for convenience + memcpy((void *)(*copied_name), (void *)name, name_size); + (*copied_name)[(name_size)] = '\0'; /* Only for convenience * in printing out the * string in debug * messages */ - error = length; + (*copied_name_size) = (name_size + 1); out: - return error; + return rc; } /** @@ -1740,7 +1752,7 @@ ecryptfs_process_key_cipher(struct crypto_blkcipher **key_tfm, *key_tfm = NULL; if (*key_size > ECRYPTFS_MAX_KEY_BYTES) { rc = -EINVAL; - printk(KERN_ERR "Requested key size is [%Zd] bytes; maximum " + printk(KERN_ERR "Requested key size is [%zd] bytes; maximum " "allowable is [%d]\n", *key_size, ECRYPTFS_MAX_KEY_BYTES); goto out; } @@ -1765,7 +1777,7 @@ ecryptfs_process_key_cipher(struct crypto_blkcipher **key_tfm, get_random_bytes(dummy_key, *key_size); rc = crypto_blkcipher_setkey(*key_tfm, dummy_key, *key_size); if (rc) { - printk(KERN_ERR "Error attempting to set key of size [%Zd] for " + printk(KERN_ERR "Error attempting to set key of size [%zd] for " "cipher [%s]; rc = [%d]\n", *key_size, cipher_name, rc); rc = -EINVAL; goto out; @@ -1910,3 +1922,341 @@ out: mutex_unlock(&key_tfm_list_mutex); return rc; } + +/* 64 characters forming a 6-bit target field */ +static unsigned char *portable_filename_chars = ("-.0123456789ABCD" + "EFGHIJKLMNOPQRST" + "UVWXYZabcdefghij" + "klmnopqrstuvwxyz"); + +/* We could either offset on every reverse map or just pad some 0x00's + * at the front here */ +static const unsigned char filename_rev_map[] = { + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 7 */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 15 */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 23 */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 31 */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 39 */ + 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x01, 0x00, /* 47 */ + 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08, 0x09, /* 55 */ + 0x0A, 0x0B, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, /* 63 */ + 0x00, 0x0C, 0x0D, 0x0E, 0x0F, 0x10, 0x11, 0x12, /* 71 */ + 0x13, 0x14, 0x15, 0x16, 0x17, 0x18, 0x19, 0x1A, /* 79 */ + 0x1B, 0x1C, 0x1D, 0x1E, 0x1F, 0x20, 0x21, 0x22, /* 87 */ + 0x23, 0x24, 0x25, 0x00, 0x00, 0x00, 0x00, 0x00, /* 95 */ + 0x00, 0x26, 0x27, 0x28, 0x29, 0x2A, 0x2B, 0x2C, /* 103 */ + 0x2D, 0x2E, 0x2F, 0x30, 0x31, 0x32, 0x33, 0x34, /* 111 */ + 0x35, 0x36, 0x37, 0x38, 0x39, 0x3A, 0x3B, 0x3C, /* 119 */ + 0x3D, 0x3E, 0x3F +}; + +/** + * ecryptfs_encode_for_filename + * @dst: Destination location for encoded filename + * @dst_size: Size of the encoded filename in bytes + * @src: Source location for the filename to encode + * @src_size: Size of the source in bytes + */ +void ecryptfs_encode_for_filename(unsigned char *dst, size_t *dst_size, + unsigned char *src, size_t src_size) +{ + size_t num_blocks; + size_t block_num = 0; + size_t dst_offset = 0; + unsigned char last_block[3]; + + if (src_size == 0) { + (*dst_size) = 0; + goto out; + } + num_blocks = (src_size / 3); + if ((src_size % 3) == 0) { + memcpy(last_block, (&src[src_size - 3]), 3); + } else { + num_blocks++; + last_block[2] = 0x00; + switch (src_size % 3) { + case 1: + last_block[0] = src[src_size - 1]; + last_block[1] = 0x00; + break; + case 2: + last_block[0] = src[src_size - 2]; + last_block[1] = src[src_size - 1]; + } + } + (*dst_size) = (num_blocks * 4); + if (!dst) + goto out; + while (block_num < num_blocks) { + unsigned char *src_block; + unsigned char dst_block[4]; + + if (block_num == (num_blocks - 1)) + src_block = last_block; + else + src_block = &src[block_num * 3]; + dst_block[0] = ((src_block[0] >> 2) & 0x3F); + dst_block[1] = (((src_block[0] << 4) & 0x30) + | ((src_block[1] >> 4) & 0x0F)); + dst_block[2] = (((src_block[1] << 2) & 0x3C) + | ((src_block[2] >> 6) & 0x03)); + dst_block[3] = (src_block[2] & 0x3F); + dst[dst_offset++] = portable_filename_chars[dst_block[0]]; + dst[dst_offset++] = portable_filename_chars[dst_block[1]]; + dst[dst_offset++] = portable_filename_chars[dst_block[2]]; + dst[dst_offset++] = portable_filename_chars[dst_block[3]]; + block_num++; + } +out: + return; +} + +/** + * ecryptfs_decode_from_filename + * @dst: If NULL, this function only sets @dst_size and returns. If + * non-NULL, this function decodes the encoded octets in @src + * into the memory that @dst points to. + * @dst_size: Set to the size of the decoded string. + * @src: The encoded set of octets to decode. + * @src_size: The size of the encoded set of octets to decode. + */ +static void +ecryptfs_decode_from_filename(unsigned char *dst, size_t *dst_size, + const unsigned char *src, size_t src_size) +{ + u8 current_bit_offset = 0; + size_t src_byte_offset = 0; + size_t dst_byte_offset = 0; + + if (dst == NULL) { + /* Not exact; conservatively long. Every block of 4 + * encoded characters decodes into a block of 3 + * decoded characters. This segment of code provides + * the caller with the maximum amount of allocated + * space that @dst will need to point to in a + * subsequent call. */ + (*dst_size) = (((src_size + 1) * 3) / 4); + goto out; + } + while (src_byte_offset < src_size) { + unsigned char src_byte = + filename_rev_map[(int)src[src_byte_offset]]; + + switch (current_bit_offset) { + case 0: + dst[dst_byte_offset] = (src_byte << 2); + current_bit_offset = 6; + break; + case 6: + dst[dst_byte_offset++] |= (src_byte >> 4); + dst[dst_byte_offset] = ((src_byte & 0xF) + << 4); + current_bit_offset = 4; + break; + case 4: + dst[dst_byte_offset++] |= (src_byte >> 2); + dst[dst_byte_offset] = (src_byte << 6); + current_bit_offset = 2; + break; + case 2: + dst[dst_byte_offset++] |= (src_byte); + dst[dst_byte_offset] = 0; + current_bit_offset = 0; + break; + } + src_byte_offset++; + } + (*dst_size) = dst_byte_offset; +out: + return; +} + +/** + * ecryptfs_encrypt_and_encode_filename - converts a plaintext file name to cipher text + * @crypt_stat: The crypt_stat struct associated with the file anem to encode + * @name: The plaintext name + * @length: The length of the plaintext + * @encoded_name: The encypted name + * + * Encrypts and encodes a filename into something that constitutes a + * valid filename for a filesystem, with printable characters. + * + * We assume that we have a properly initialized crypto context, + * pointed to by crypt_stat->tfm. + * + * Returns zero on success; non-zero on otherwise + */ +int ecryptfs_encrypt_and_encode_filename( + char **encoded_name, + size_t *encoded_name_size, + struct ecryptfs_crypt_stat *crypt_stat, + struct ecryptfs_mount_crypt_stat *mount_crypt_stat, + const char *name, size_t name_size) +{ + size_t encoded_name_no_prefix_size; + int rc = 0; + + (*encoded_name) = NULL; + (*encoded_name_size) = 0; + if ((crypt_stat && (crypt_stat->flags & ECRYPTFS_ENCRYPT_FILENAMES)) + || (mount_crypt_stat && (mount_crypt_stat->flags + & ECRYPTFS_GLOBAL_ENCRYPT_FILENAMES))) { + struct ecryptfs_filename *filename; + + filename = kzalloc(sizeof(*filename), GFP_KERNEL); + if (!filename) { + printk(KERN_ERR "%s: Out of memory whilst attempting " + "to kzalloc [%zd] bytes\n", __func__, + sizeof(*filename)); + rc = -ENOMEM; + goto out; + } + filename->filename = (char *)name; + filename->filename_size = name_size; + rc = ecryptfs_encrypt_filename(filename, crypt_stat, + mount_crypt_stat); + if (rc) { + printk(KERN_ERR "%s: Error attempting to encrypt " + "filename; rc = [%d]\n", __func__, rc); + kfree(filename); + goto out; + } + ecryptfs_encode_for_filename( + NULL, &encoded_name_no_prefix_size, + filename->encrypted_filename, + filename->encrypted_filename_size); + if ((crypt_stat && (crypt_stat->flags + & ECRYPTFS_ENCFN_USE_MOUNT_FNEK)) + || (mount_crypt_stat + && (mount_crypt_stat->flags + & ECRYPTFS_GLOBAL_ENCFN_USE_MOUNT_FNEK))) + (*encoded_name_size) = + (ECRYPTFS_FNEK_ENCRYPTED_FILENAME_PREFIX_SIZE + + encoded_name_no_prefix_size); + else + (*encoded_name_size) = + (ECRYPTFS_FEK_ENCRYPTED_FILENAME_PREFIX_SIZE + + encoded_name_no_prefix_size); + (*encoded_name) = kmalloc((*encoded_name_size) + 1, GFP_KERNEL); + if (!(*encoded_name)) { + printk(KERN_ERR "%s: Out of memory whilst attempting " + "to kzalloc [%zd] bytes\n", __func__, + (*encoded_name_size)); + rc = -ENOMEM; + kfree(filename->encrypted_filename); + kfree(filename); + goto out; + } + if ((crypt_stat && (crypt_stat->flags + & ECRYPTFS_ENCFN_USE_MOUNT_FNEK)) + || (mount_crypt_stat + && (mount_crypt_stat->flags + & ECRYPTFS_GLOBAL_ENCFN_USE_MOUNT_FNEK))) { + memcpy((*encoded_name), + ECRYPTFS_FNEK_ENCRYPTED_FILENAME_PREFIX, + ECRYPTFS_FNEK_ENCRYPTED_FILENAME_PREFIX_SIZE); + ecryptfs_encode_for_filename( + ((*encoded_name) + + ECRYPTFS_FNEK_ENCRYPTED_FILENAME_PREFIX_SIZE), + &encoded_name_no_prefix_size, + filename->encrypted_filename, + filename->encrypted_filename_size); + (*encoded_name_size) = + (ECRYPTFS_FNEK_ENCRYPTED_FILENAME_PREFIX_SIZE + + encoded_name_no_prefix_size); + (*encoded_name)[(*encoded_name_size)] = '\0'; + (*encoded_name_size)++; + } else { + rc = -ENOTSUPP; + } + if (rc) { + printk(KERN_ERR "%s: Error attempting to encode " + "encrypted filename; rc = [%d]\n", __func__, + rc); + kfree((*encoded_name)); + (*encoded_name) = NULL; + (*encoded_name_size) = 0; + } + kfree(filename->encrypted_filename); + kfree(filename); + } else { + rc = ecryptfs_copy_filename(encoded_name, + encoded_name_size, + name, name_size); + } +out: + return rc; +} + +/** + * ecryptfs_decode_and_decrypt_filename - converts the encoded cipher text name to decoded plaintext + * @plaintext_name: The plaintext name + * @plaintext_name_size: The plaintext name size + * @ecryptfs_dir_dentry: eCryptfs directory dentry + * @name: The filename in cipher text + * @name_size: The cipher text name size + * + * Decrypts and decodes the filename. + * + * Returns zero on error; non-zero otherwise + */ +int ecryptfs_decode_and_decrypt_filename(char **plaintext_name, + size_t *plaintext_name_size, + struct dentry *ecryptfs_dir_dentry, + const char *name, size_t name_size) +{ + char *decoded_name; + size_t decoded_name_size; + size_t packet_size; + int rc = 0; + + if ((name_size > ECRYPTFS_FNEK_ENCRYPTED_FILENAME_PREFIX_SIZE) + && (strncmp(name, ECRYPTFS_FNEK_ENCRYPTED_FILENAME_PREFIX, + ECRYPTFS_FNEK_ENCRYPTED_FILENAME_PREFIX_SIZE) == 0)) { + struct ecryptfs_mount_crypt_stat *mount_crypt_stat = + &ecryptfs_superblock_to_private( + ecryptfs_dir_dentry->d_sb)->mount_crypt_stat; + const char *orig_name = name; + size_t orig_name_size = name_size; + + name += ECRYPTFS_FNEK_ENCRYPTED_FILENAME_PREFIX_SIZE; + name_size -= ECRYPTFS_FNEK_ENCRYPTED_FILENAME_PREFIX_SIZE; + ecryptfs_decode_from_filename(NULL, &decoded_name_size, + name, name_size); + decoded_name = kmalloc(decoded_name_size, GFP_KERNEL); + if (!decoded_name) { + printk(KERN_ERR "%s: Out of memory whilst attempting " + "to kmalloc [%zd] bytes\n", __func__, + decoded_name_size); + rc = -ENOMEM; + goto out; + } + ecryptfs_decode_from_filename(decoded_name, &decoded_name_size, + name, name_size); + rc = ecryptfs_parse_tag_70_packet(plaintext_name, + plaintext_name_size, + &packet_size, + mount_crypt_stat, + decoded_name, + decoded_name_size); + if (rc) { + printk(KERN_INFO "%s: Could not parse tag 70 packet " + "from filename; copying through filename " + "as-is\n", __func__); + rc = ecryptfs_copy_filename(plaintext_name, + plaintext_name_size, + orig_name, orig_name_size); + goto out_free; + } + } else { + rc = ecryptfs_copy_filename(plaintext_name, + plaintext_name_size, + name, name_size); + goto out; + } +out_free: + kfree(decoded_name); +out: + return rc; +} diff --git a/fs/ecryptfs/ecryptfs_kernel.h b/fs/ecryptfs/ecryptfs_kernel.h index a75026d35d16..c11fc95714ab 100644 --- a/fs/ecryptfs/ecryptfs_kernel.h +++ b/fs/ecryptfs/ecryptfs_kernel.h @@ -51,12 +51,16 @@ #define ECRYPTFS_VERSIONING_XATTR 0x00000010 #define ECRYPTFS_VERSIONING_MULTKEY 0x00000020 #define ECRYPTFS_VERSIONING_DEVMISC 0x00000040 +#define ECRYPTFS_VERSIONING_HMAC 0x00000080 +#define ECRYPTFS_VERSIONING_FILENAME_ENCRYPTION 0x00000100 +#define ECRYPTFS_VERSIONING_GCM 0x00000200 #define ECRYPTFS_VERSIONING_MASK (ECRYPTFS_VERSIONING_PASSPHRASE \ | ECRYPTFS_VERSIONING_PLAINTEXT_PASSTHROUGH \ | ECRYPTFS_VERSIONING_PUBKEY \ | ECRYPTFS_VERSIONING_XATTR \ | ECRYPTFS_VERSIONING_MULTKEY \ - | ECRYPTFS_VERSIONING_DEVMISC) + | ECRYPTFS_VERSIONING_DEVMISC \ + | ECRYPTFS_VERSIONING_FILENAME_ENCRYPTION) #define ECRYPTFS_MAX_PASSWORD_LENGTH 64 #define ECRYPTFS_MAX_PASSPHRASE_BYTES ECRYPTFS_MAX_PASSWORD_LENGTH #define ECRYPTFS_SALT_SIZE 8 @@ -199,6 +203,7 @@ ecryptfs_get_key_payload_data(struct key *key) #define ECRYPTFS_DEFAULT_CIPHER "aes" #define ECRYPTFS_DEFAULT_KEY_BYTES 16 #define ECRYPTFS_DEFAULT_HASH "md5" +#define ECRYPTFS_TAG_70_DIGEST ECRYPTFS_DEFAULT_HASH #define ECRYPTFS_TAG_1_PACKET_TYPE 0x01 #define ECRYPTFS_TAG_3_PACKET_TYPE 0x8C #define ECRYPTFS_TAG_11_PACKET_TYPE 0xED @@ -206,30 +211,64 @@ ecryptfs_get_key_payload_data(struct key *key) #define ECRYPTFS_TAG_65_PACKET_TYPE 0x41 #define ECRYPTFS_TAG_66_PACKET_TYPE 0x42 #define ECRYPTFS_TAG_67_PACKET_TYPE 0x43 +#define ECRYPTFS_TAG_70_PACKET_TYPE 0x46 /* FNEK-encrypted filename + * as dentry name */ +#define ECRYPTFS_TAG_71_PACKET_TYPE 0x47 /* FNEK-encrypted filename in + * metadata */ +#define ECRYPTFS_TAG_72_PACKET_TYPE 0x48 /* FEK-encrypted filename as + * dentry name */ +#define ECRYPTFS_TAG_73_PACKET_TYPE 0x49 /* FEK-encrypted filename as + * metadata */ +/* Constraint: ECRYPTFS_FILENAME_MIN_RANDOM_PREPEND_BYTES >= + * ECRYPTFS_MAX_IV_BYTES */ +#define ECRYPTFS_FILENAME_MIN_RANDOM_PREPEND_BYTES 16 +#define ECRYPTFS_NON_NULL 0x42 /* A reasonable substitute for NULL */ #define MD5_DIGEST_SIZE 16 +#define ECRYPTFS_TAG_70_DIGEST_SIZE MD5_DIGEST_SIZE +#define ECRYPTFS_FEK_ENCRYPTED_FILENAME_PREFIX "ECRYPTFS_FEK_ENCRYPTED." +#define ECRYPTFS_FEK_ENCRYPTED_FILENAME_PREFIX_SIZE 23 +#define ECRYPTFS_FNEK_ENCRYPTED_FILENAME_PREFIX "ECRYPTFS_FNEK_ENCRYPTED." +#define ECRYPTFS_FNEK_ENCRYPTED_FILENAME_PREFIX_SIZE 24 +#define ECRYPTFS_ENCRYPTED_DENTRY_NAME_LEN (18 + 1 + 4 + 1 + 32) struct ecryptfs_key_sig { struct list_head crypt_stat_list; char keysig[ECRYPTFS_SIG_SIZE_HEX]; }; +struct ecryptfs_filename { + struct list_head crypt_stat_list; +#define ECRYPTFS_FILENAME_CONTAINS_DECRYPTED 0x00000001 + u32 flags; + u32 seq_no; + char *filename; + char *encrypted_filename; + size_t filename_size; + size_t encrypted_filename_size; + char fnek_sig[ECRYPTFS_SIG_SIZE_HEX]; + char dentry_name[ECRYPTFS_ENCRYPTED_DENTRY_NAME_LEN + 1]; +}; + /** * This is the primary struct associated with each encrypted file. * * TODO: cache align/pack? */ struct ecryptfs_crypt_stat { -#define ECRYPTFS_STRUCT_INITIALIZED 0x00000001 -#define ECRYPTFS_POLICY_APPLIED 0x00000002 -#define ECRYPTFS_NEW_FILE 0x00000004 -#define ECRYPTFS_ENCRYPTED 0x00000008 -#define ECRYPTFS_SECURITY_WARNING 0x00000010 -#define ECRYPTFS_ENABLE_HMAC 0x00000020 -#define ECRYPTFS_ENCRYPT_IV_PAGES 0x00000040 -#define ECRYPTFS_KEY_VALID 0x00000080 -#define ECRYPTFS_METADATA_IN_XATTR 0x00000100 -#define ECRYPTFS_VIEW_AS_ENCRYPTED 0x00000200 -#define ECRYPTFS_KEY_SET 0x00000400 +#define ECRYPTFS_STRUCT_INITIALIZED 0x00000001 +#define ECRYPTFS_POLICY_APPLIED 0x00000002 +#define ECRYPTFS_NEW_FILE 0x00000004 +#define ECRYPTFS_ENCRYPTED 0x00000008 +#define ECRYPTFS_SECURITY_WARNING 0x00000010 +#define ECRYPTFS_ENABLE_HMAC 0x00000020 +#define ECRYPTFS_ENCRYPT_IV_PAGES 0x00000040 +#define ECRYPTFS_KEY_VALID 0x00000080 +#define ECRYPTFS_METADATA_IN_XATTR 0x00000100 +#define ECRYPTFS_VIEW_AS_ENCRYPTED 0x00000200 +#define ECRYPTFS_KEY_SET 0x00000400 +#define ECRYPTFS_ENCRYPT_FILENAMES 0x00000800 +#define ECRYPTFS_ENCFN_USE_MOUNT_FNEK 0x00001000 +#define ECRYPTFS_ENCFN_USE_FEK 0x00002000 u32 flags; unsigned int file_version; size_t iv_bytes; @@ -332,13 +371,20 @@ struct ecryptfs_mount_crypt_stat { #define ECRYPTFS_XATTR_METADATA_ENABLED 0x00000002 #define ECRYPTFS_ENCRYPTED_VIEW_ENABLED 0x00000004 #define ECRYPTFS_MOUNT_CRYPT_STAT_INITIALIZED 0x00000008 +#define ECRYPTFS_GLOBAL_ENCRYPT_FILENAMES 0x00000010 +#define ECRYPTFS_GLOBAL_ENCFN_USE_MOUNT_FNEK 0x00000020 +#define ECRYPTFS_GLOBAL_ENCFN_USE_FEK 0x00000040 u32 flags; struct list_head global_auth_tok_list; struct mutex global_auth_tok_list_mutex; size_t num_global_auth_toks; size_t global_default_cipher_key_size; + size_t global_default_fn_cipher_key_bytes; unsigned char global_default_cipher_name[ECRYPTFS_MAX_CIPHER_NAME_SIZE + 1]; + unsigned char global_default_fn_cipher_name[ + ECRYPTFS_MAX_CIPHER_NAME_SIZE + 1]; + char global_default_fnek_sig[ECRYPTFS_SIG_SIZE_HEX + 1]; }; /* superblock private data. */ @@ -571,13 +617,22 @@ struct ecryptfs_open_req { int ecryptfs_interpose(struct dentry *hidden_dentry, struct dentry *this_dentry, struct super_block *sb, u32 flags); +int ecryptfs_lookup_and_interpose_lower(struct dentry *ecryptfs_dentry, + struct dentry *lower_dentry, + struct ecryptfs_crypt_stat *crypt_stat, + struct inode *ecryptfs_dir_inode, + struct nameidata *ecryptfs_nd); +int ecryptfs_decode_and_decrypt_filename(char **decrypted_name, + size_t *decrypted_name_size, + struct dentry *ecryptfs_dentry, + const char *name, size_t name_size); int ecryptfs_fill_zeros(struct file *file, loff_t new_length); -int ecryptfs_decode_filename(struct ecryptfs_crypt_stat *crypt_stat, - const char *name, int length, - char **decrypted_name); -int ecryptfs_encode_filename(struct ecryptfs_crypt_stat *crypt_stat, - const char *name, int length, - char **encoded_name); +int ecryptfs_encrypt_and_encode_filename( + char **encoded_name, + size_t *encoded_name_size, + struct ecryptfs_crypt_stat *crypt_stat, + struct ecryptfs_mount_crypt_stat *mount_crypt_stat, + const char *name, size_t name_size); struct dentry *ecryptfs_lower_dentry(struct dentry *this_dentry); void ecryptfs_dump_hex(char *data, int bytes); int virt_to_scatterlist(const void *addr, int size, struct scatterlist *sg, @@ -599,7 +654,7 @@ int ecryptfs_read_and_validate_header_region(char *data, struct inode *ecryptfs_inode); int ecryptfs_read_and_validate_xattr_region(char *page_virt, struct dentry *ecryptfs_dentry); -u8 ecryptfs_code_for_cipher_string(struct ecryptfs_crypt_stat *crypt_stat); +u8 ecryptfs_code_for_cipher_string(char *cipher_name, size_t key_bytes); int ecryptfs_cipher_code_to_string(char *str, u8 cipher_code); void ecryptfs_set_default_sizes(struct ecryptfs_crypt_stat *crypt_stat); int ecryptfs_generate_key_packet_set(char *dest_base, @@ -694,5 +749,17 @@ int ecryptfs_privileged_open(struct file **lower_file, struct vfsmount *lower_mnt, const struct cred *cred); int ecryptfs_init_persistent_file(struct dentry *ecryptfs_dentry); +int +ecryptfs_write_tag_70_packet(char *dest, size_t *remaining_bytes, + size_t *packet_size, + struct ecryptfs_mount_crypt_stat *mount_crypt_stat, + char *filename, size_t filename_size); +int +ecryptfs_parse_tag_70_packet(char **filename, size_t *filename_size, + size_t *packet_size, + struct ecryptfs_mount_crypt_stat *mount_crypt_stat, + char *data, size_t max_packet_size); +int ecryptfs_derive_iv(char *iv, struct ecryptfs_crypt_stat *crypt_stat, + loff_t offset); #endif /* #ifndef ECRYPTFS_KERNEL_H */ diff --git a/fs/ecryptfs/file.c b/fs/ecryptfs/file.c index 713834371229..9e944057001b 100644 --- a/fs/ecryptfs/file.c +++ b/fs/ecryptfs/file.c @@ -77,27 +77,27 @@ struct ecryptfs_getdents_callback { /* Inspired by generic filldir in fs/readdir.c */ static int -ecryptfs_filldir(void *dirent, const char *name, int namelen, loff_t offset, - u64 ino, unsigned int d_type) +ecryptfs_filldir(void *dirent, const char *lower_name, int lower_namelen, + loff_t offset, u64 ino, unsigned int d_type) { - struct ecryptfs_crypt_stat *crypt_stat; struct ecryptfs_getdents_callback *buf = (struct ecryptfs_getdents_callback *)dirent; + size_t name_size; + char *name; int rc; - int decoded_length; - char *decoded_name; - crypt_stat = ecryptfs_dentry_to_private(buf->dentry)->crypt_stat; buf->filldir_called++; - decoded_length = ecryptfs_decode_filename(crypt_stat, name, namelen, - &decoded_name); - if (decoded_length < 0) { - rc = decoded_length; + rc = ecryptfs_decode_and_decrypt_filename(&name, &name_size, + buf->dentry, lower_name, + lower_namelen); + if (rc) { + printk(KERN_ERR "%s: Error attempting to decode and decrypt " + "filename [%s]; rc = [%d]\n", __func__, lower_name, + rc); goto out; } - rc = buf->filldir(buf->dirent, decoded_name, decoded_length, offset, - ino, d_type); - kfree(decoded_name); + rc = buf->filldir(buf->dirent, name, name_size, offset, ino, d_type); + kfree(name); if (rc >= 0) buf->entries_written++; out: @@ -106,8 +106,8 @@ out: /** * ecryptfs_readdir - * @file: The ecryptfs file struct - * @dirent: Directory entry + * @file: The eCryptfs directory file + * @dirent: Directory entry handle * @filldir: The filldir callback function */ static int ecryptfs_readdir(struct file *file, void *dirent, filldir_t filldir) diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index 0111906a8877..5697899a168d 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -52,8 +52,7 @@ static void unlock_dir(struct dentry *dir) /** * ecryptfs_create_underlying_file * @lower_dir_inode: inode of the parent in the lower fs of the new file - * @lower_dentry: New file's dentry in the lower fs - * @ecryptfs_dentry: New file's dentry in ecryptfs + * @dentry: New file's dentry * @mode: The mode of the new file * @nd: nameidata of ecryptfs' parent's dentry & vfsmount * @@ -228,8 +227,7 @@ ecryptfs_create(struct inode *directory_inode, struct dentry *ecryptfs_dentry, { int rc; - /* ecryptfs_do_create() calls ecryptfs_interpose(), which opens - * the crypt_stat->lower_file (persistent file) */ + /* ecryptfs_do_create() calls ecryptfs_interpose() */ rc = ecryptfs_do_create(directory_inode, ecryptfs_dentry, mode, nd); if (unlikely(rc)) { ecryptfs_printk(KERN_WARNING, "Failed to create file in" @@ -244,141 +242,91 @@ out: } /** - * ecryptfs_lookup - * @dir: inode - * @dentry: The dentry - * @nd: nameidata, may be NULL - * - * Find a file on disk. If the file does not exist, then we'll add it to the - * dentry cache and continue on to read it from the disk. + * ecryptfs_lookup_and_interpose_lower - Perform a lookup */ -static struct dentry *ecryptfs_lookup(struct inode *dir, struct dentry *dentry, - struct nameidata *nd) +int ecryptfs_lookup_and_interpose_lower(struct dentry *ecryptfs_dentry, + struct dentry *lower_dentry, + struct ecryptfs_crypt_stat *crypt_stat, + struct inode *ecryptfs_dir_inode, + struct nameidata *ecryptfs_nd) { - int rc = 0; struct dentry *lower_dir_dentry; - struct dentry *lower_dentry; struct vfsmount *lower_mnt; - char *encoded_name; - int encoded_namelen; - struct ecryptfs_crypt_stat *crypt_stat = NULL; + struct inode *lower_inode; struct ecryptfs_mount_crypt_stat *mount_crypt_stat; char *page_virt = NULL; - struct inode *lower_inode; u64 file_size; + int rc = 0; - lower_dir_dentry = ecryptfs_dentry_to_lower(dentry->d_parent); - dentry->d_op = &ecryptfs_dops; - if ((dentry->d_name.len == 1 && !strcmp(dentry->d_name.name, ".")) - || (dentry->d_name.len == 2 - && !strcmp(dentry->d_name.name, ".."))) { - d_drop(dentry); - goto out; - } - encoded_namelen = ecryptfs_encode_filename(crypt_stat, - dentry->d_name.name, - dentry->d_name.len, - &encoded_name); - if (encoded_namelen < 0) { - rc = encoded_namelen; - d_drop(dentry); - goto out; - } - ecryptfs_printk(KERN_DEBUG, "encoded_name = [%s]; encoded_namelen " - "= [%d]\n", encoded_name, encoded_namelen); - lower_dentry = lookup_one_len(encoded_name, lower_dir_dentry, - encoded_namelen - 1); - kfree(encoded_name); - if (IS_ERR(lower_dentry)) { - ecryptfs_printk(KERN_ERR, "ERR from lower_dentry\n"); - rc = PTR_ERR(lower_dentry); - d_drop(dentry); - goto out; - } - lower_mnt = mntget(ecryptfs_dentry_to_lower_mnt(dentry->d_parent)); - ecryptfs_printk(KERN_DEBUG, "lower_dentry = [%p]; lower_dentry->" - "d_name.name = [%s]\n", lower_dentry, - lower_dentry->d_name.name); + lower_dir_dentry = lower_dentry->d_parent; + lower_mnt = mntget(ecryptfs_dentry_to_lower_mnt( + ecryptfs_dentry->d_parent)); lower_inode = lower_dentry->d_inode; - fsstack_copy_attr_atime(dir, lower_dir_dentry->d_inode); + fsstack_copy_attr_atime(ecryptfs_dir_inode, lower_dir_dentry->d_inode); BUG_ON(!atomic_read(&lower_dentry->d_count)); - ecryptfs_set_dentry_private(dentry, + ecryptfs_set_dentry_private(ecryptfs_dentry, kmem_cache_alloc(ecryptfs_dentry_info_cache, GFP_KERNEL)); - if (!ecryptfs_dentry_to_private(dentry)) { + if (!ecryptfs_dentry_to_private(ecryptfs_dentry)) { rc = -ENOMEM; - ecryptfs_printk(KERN_ERR, "Out of memory whilst attempting " - "to allocate ecryptfs_dentry_info struct\n"); + printk(KERN_ERR "%s: Out of memory whilst attempting " + "to allocate ecryptfs_dentry_info struct\n", + __func__); goto out_dput; } - ecryptfs_set_dentry_lower(dentry, lower_dentry); - ecryptfs_set_dentry_lower_mnt(dentry, lower_mnt); + ecryptfs_set_dentry_lower(ecryptfs_dentry, lower_dentry); + ecryptfs_set_dentry_lower_mnt(ecryptfs_dentry, lower_mnt); if (!lower_dentry->d_inode) { /* We want to add because we couldn't find in lower */ - d_add(dentry, NULL); + d_add(ecryptfs_dentry, NULL); goto out; } - rc = ecryptfs_interpose(lower_dentry, dentry, dir->i_sb, - ECRYPTFS_INTERPOSE_FLAG_D_ADD); + rc = ecryptfs_interpose(lower_dentry, ecryptfs_dentry, + ecryptfs_dir_inode->i_sb, 1); if (rc) { - ecryptfs_printk(KERN_ERR, "Error interposing\n"); + printk(KERN_ERR "%s: Error interposing; rc = [%d]\n", + __func__, rc); goto out; } - if (S_ISDIR(lower_inode->i_mode)) { - ecryptfs_printk(KERN_DEBUG, "Is a directory; returning\n"); + if (S_ISDIR(lower_inode->i_mode)) goto out; - } - if (S_ISLNK(lower_inode->i_mode)) { - ecryptfs_printk(KERN_DEBUG, "Is a symlink; returning\n"); + if (S_ISLNK(lower_inode->i_mode)) goto out; - } - if (special_file(lower_inode->i_mode)) { - ecryptfs_printk(KERN_DEBUG, "Is a special file; returning\n"); + if (special_file(lower_inode->i_mode)) goto out; - } - if (!nd) { - ecryptfs_printk(KERN_DEBUG, "We have a NULL nd, just leave" - "as we *think* we are about to unlink\n"); + if (!ecryptfs_nd) goto out; - } /* Released in this function */ - page_virt = kmem_cache_zalloc(ecryptfs_header_cache_2, - GFP_USER); + page_virt = kmem_cache_zalloc(ecryptfs_header_cache_2, GFP_USER); if (!page_virt) { + printk(KERN_ERR "%s: Cannot kmem_cache_zalloc() a page\n", + __func__); rc = -ENOMEM; - ecryptfs_printk(KERN_ERR, - "Cannot ecryptfs_kmalloc a page\n"); goto out; } - crypt_stat = &ecryptfs_inode_to_private(dentry->d_inode)->crypt_stat; - if (!(crypt_stat->flags & ECRYPTFS_POLICY_APPLIED)) - ecryptfs_set_default_sizes(crypt_stat); - if (!ecryptfs_inode_to_private(dentry->d_inode)->lower_file) { - rc = ecryptfs_init_persistent_file(dentry); + if (!ecryptfs_inode_to_private(ecryptfs_dentry->d_inode)->lower_file) { + rc = ecryptfs_init_persistent_file(ecryptfs_dentry); if (rc) { printk(KERN_ERR "%s: Error attempting to initialize " "the persistent file for the dentry with name " "[%s]; rc = [%d]\n", __func__, - dentry->d_name.name, rc); - goto out; + ecryptfs_dentry->d_name.name, rc); + goto out_free_kmem; } } rc = ecryptfs_read_and_validate_header_region(page_virt, - dentry->d_inode); + ecryptfs_dentry->d_inode); if (rc) { - rc = ecryptfs_read_and_validate_xattr_region(page_virt, dentry); + rc = ecryptfs_read_and_validate_xattr_region(page_virt, + ecryptfs_dentry); if (rc) { - printk(KERN_DEBUG "Valid metadata not found in header " - "region or xattr region; treating file as " - "unencrypted\n"); rc = 0; - kmem_cache_free(ecryptfs_header_cache_2, page_virt); - goto out; + goto out_free_kmem; } crypt_stat->flags |= ECRYPTFS_METADATA_IN_XATTR; } mount_crypt_stat = &ecryptfs_superblock_to_private( - dentry->d_sb)->mount_crypt_stat; + ecryptfs_dentry->d_sb)->mount_crypt_stat; if (mount_crypt_stat->flags & ECRYPTFS_ENCRYPTED_VIEW_ENABLED) { if (crypt_stat->flags & ECRYPTFS_METADATA_IN_XATTR) file_size = (crypt_stat->num_header_bytes_at_front @@ -388,14 +336,103 @@ static struct dentry *ecryptfs_lookup(struct inode *dir, struct dentry *dentry, } else { file_size = get_unaligned_be64(page_virt); } - i_size_write(dentry->d_inode, (loff_t)file_size); + i_size_write(ecryptfs_dentry->d_inode, (loff_t)file_size); +out_free_kmem: kmem_cache_free(ecryptfs_header_cache_2, page_virt); goto out; - out_dput: dput(lower_dentry); - d_drop(dentry); + d_drop(ecryptfs_dentry); +out: + return rc; +} + +/** + * ecryptfs_lookup + * @ecryptfs_dir_inode: The eCryptfs directory inode + * @ecryptfs_dentry: The eCryptfs dentry that we are looking up + * @ecryptfs_nd: nameidata; may be NULL + * + * Find a file on disk. If the file does not exist, then we'll add it to the + * dentry cache and continue on to read it from the disk. + */ +static struct dentry *ecryptfs_lookup(struct inode *ecryptfs_dir_inode, + struct dentry *ecryptfs_dentry, + struct nameidata *ecryptfs_nd) +{ + char *encrypted_and_encoded_name = NULL; + size_t encrypted_and_encoded_name_size; + struct ecryptfs_crypt_stat *crypt_stat = NULL; + struct ecryptfs_mount_crypt_stat *mount_crypt_stat = NULL; + struct ecryptfs_inode_info *inode_info; + struct dentry *lower_dir_dentry, *lower_dentry; + int rc = 0; + + ecryptfs_dentry->d_op = &ecryptfs_dops; + if ((ecryptfs_dentry->d_name.len == 1 + && !strcmp(ecryptfs_dentry->d_name.name, ".")) + || (ecryptfs_dentry->d_name.len == 2 + && !strcmp(ecryptfs_dentry->d_name.name, ".."))) { + goto out_d_drop; + } + lower_dir_dentry = ecryptfs_dentry_to_lower(ecryptfs_dentry->d_parent); + lower_dentry = lookup_one_len(ecryptfs_dentry->d_name.name, + lower_dir_dentry, + ecryptfs_dentry->d_name.len); + if (IS_ERR(lower_dentry)) { + rc = PTR_ERR(lower_dentry); + printk(KERN_ERR "%s: lookup_one_len() returned [%d] on " + "lower_dentry = [%s]\n", __func__, rc, + ecryptfs_dentry->d_name.name); + goto out_d_drop; + } + if (lower_dentry->d_inode) + goto lookup_and_interpose; + inode_info = ecryptfs_inode_to_private(ecryptfs_dentry->d_inode); + if (inode_info) { + crypt_stat = &inode_info->crypt_stat; + /* TODO: lock for crypt_stat comparison */ + if (!(crypt_stat->flags & ECRYPTFS_POLICY_APPLIED)) + ecryptfs_set_default_sizes(crypt_stat); + } + if (crypt_stat) + mount_crypt_stat = crypt_stat->mount_crypt_stat; + else + mount_crypt_stat = &ecryptfs_superblock_to_private( + ecryptfs_dentry->d_sb)->mount_crypt_stat; + if (!(crypt_stat && (crypt_stat->flags & ECRYPTFS_ENCRYPT_FILENAMES)) + && !(mount_crypt_stat && (mount_crypt_stat->flags + & ECRYPTFS_GLOBAL_ENCRYPT_FILENAMES))) + goto lookup_and_interpose; + dput(lower_dentry); + rc = ecryptfs_encrypt_and_encode_filename( + &encrypted_and_encoded_name, &encrypted_and_encoded_name_size, + crypt_stat, mount_crypt_stat, ecryptfs_dentry->d_name.name, + ecryptfs_dentry->d_name.len); + if (rc) { + printk(KERN_ERR "%s: Error attempting to encrypt and encode " + "filename; rc = [%d]\n", __func__, rc); + goto out_d_drop; + } + lower_dentry = lookup_one_len(encrypted_and_encoded_name, + lower_dir_dentry, + encrypted_and_encoded_name_size - 1); + if (IS_ERR(lower_dentry)) { + rc = PTR_ERR(lower_dentry); + printk(KERN_ERR "%s: lookup_one_len() returned [%d] on " + "lower_dentry = [%s]\n", __func__, rc, + encrypted_and_encoded_name); + goto out_d_drop; + } +lookup_and_interpose: + rc = ecryptfs_lookup_and_interpose_lower(ecryptfs_dentry, lower_dentry, + crypt_stat, ecryptfs_dir_inode, + ecryptfs_nd); + goto out; +out_d_drop: + d_drop(ecryptfs_dentry); out: + kfree(encrypted_and_encoded_name); return ERR_PTR(rc); } @@ -466,19 +503,21 @@ static int ecryptfs_symlink(struct inode *dir, struct dentry *dentry, struct dentry *lower_dentry; struct dentry *lower_dir_dentry; char *encoded_symname; - int encoded_symlen; - struct ecryptfs_crypt_stat *crypt_stat = NULL; + size_t encoded_symlen; + struct ecryptfs_mount_crypt_stat *mount_crypt_stat = NULL; lower_dentry = ecryptfs_dentry_to_lower(dentry); dget(lower_dentry); lower_dir_dentry = lock_parent(lower_dentry); - encoded_symlen = ecryptfs_encode_filename(crypt_stat, symname, - strlen(symname), - &encoded_symname); - if (encoded_symlen < 0) { - rc = encoded_symlen; + mount_crypt_stat = &ecryptfs_superblock_to_private( + dir->i_sb)->mount_crypt_stat; + rc = ecryptfs_encrypt_and_encode_filename(&encoded_symname, + &encoded_symlen, + NULL, + mount_crypt_stat, symname, + strlen(symname)); + if (rc) goto out_lock; - } rc = vfs_symlink(lower_dir_dentry->d_inode, lower_dentry, encoded_symname); kfree(encoded_symname); @@ -602,52 +641,54 @@ out_lock: } static int -ecryptfs_readlink(struct dentry *dentry, char __user * buf, int bufsiz) +ecryptfs_readlink(struct dentry *dentry, char __user *buf, int bufsiz) { - int rc; - struct dentry *lower_dentry; - char *decoded_name; char *lower_buf; - mm_segment_t old_fs; + struct dentry *lower_dentry; struct ecryptfs_crypt_stat *crypt_stat; + char *plaintext_name; + size_t plaintext_name_size; + mm_segment_t old_fs; + int rc; lower_dentry = ecryptfs_dentry_to_lower(dentry); if (!lower_dentry->d_inode->i_op->readlink) { rc = -EINVAL; goto out; } + crypt_stat = &ecryptfs_inode_to_private(dentry->d_inode)->crypt_stat; /* Released in this function */ lower_buf = kmalloc(bufsiz, GFP_KERNEL); if (lower_buf == NULL) { - ecryptfs_printk(KERN_ERR, "Out of memory\n"); + printk(KERN_ERR "%s: Out of memory whilst attempting to " + "kmalloc [%d] bytes\n", __func__, bufsiz); rc = -ENOMEM; goto out; } old_fs = get_fs(); set_fs(get_ds()); - ecryptfs_printk(KERN_DEBUG, "Calling readlink w/ " - "lower_dentry->d_name.name = [%s]\n", - lower_dentry->d_name.name); rc = lower_dentry->d_inode->i_op->readlink(lower_dentry, (char __user *)lower_buf, bufsiz); set_fs(old_fs); if (rc >= 0) { - crypt_stat = NULL; - rc = ecryptfs_decode_filename(crypt_stat, lower_buf, rc, - &decoded_name); - if (rc == -ENOMEM) + rc = ecryptfs_decode_and_decrypt_filename(&plaintext_name, + &plaintext_name_size, + dentry, lower_buf, + rc); + if (rc) { + printk(KERN_ERR "%s: Error attempting to decode and " + "decrypt filename; rc = [%d]\n", __func__, + rc); goto out_free_lower_buf; - if (rc > 0) { - ecryptfs_printk(KERN_DEBUG, "Copying [%d] bytes " - "to userspace: [%*s]\n", rc, - decoded_name); - if (copy_to_user(buf, decoded_name, rc)) - rc = -EFAULT; } - kfree(decoded_name); - fsstack_copy_attr_atime(dentry->d_inode, - lower_dentry->d_inode); + rc = copy_to_user(buf, plaintext_name, plaintext_name_size); + if (rc) + rc = -EFAULT; + else + rc = plaintext_name_size; + kfree(plaintext_name); + fsstack_copy_attr_atime(dentry->d_inode, lower_dentry->d_inode); } out_free_lower_buf: kfree(lower_buf); @@ -669,8 +710,6 @@ static void *ecryptfs_follow_link(struct dentry *dentry, struct nameidata *nd) } old_fs = get_fs(); set_fs(get_ds()); - ecryptfs_printk(KERN_DEBUG, "Calling readlink w/ " - "dentry->d_name.name = [%s]\n", dentry->d_name.name); rc = dentry->d_inode->i_op->readlink(dentry, (char __user *)buf, len); set_fs(old_fs); if (rc < 0) diff --git a/fs/ecryptfs/keystore.c b/fs/ecryptfs/keystore.c index 0d713b691941..ff539420cc6f 100644 --- a/fs/ecryptfs/keystore.c +++ b/fs/ecryptfs/keystore.c @@ -358,7 +358,7 @@ parse_tag_67_packet(struct ecryptfs_key_record *key_rec, /* verify that everything through the encrypted FEK size is present */ if (message_len < 4) { rc = -EIO; - printk(KERN_ERR "%s: message_len is [%Zd]; minimum acceptable " + printk(KERN_ERR "%s: message_len is [%zd]; minimum acceptable " "message length is [%d]\n", __func__, message_len, 4); goto out; } @@ -385,13 +385,13 @@ parse_tag_67_packet(struct ecryptfs_key_record *key_rec, i += data_len; if (message_len < (i + key_rec->enc_key_size)) { rc = -EIO; - printk(KERN_ERR "%s: message_len [%Zd]; max len is [%Zd]\n", + printk(KERN_ERR "%s: message_len [%zd]; max len is [%zd]\n", __func__, message_len, (i + key_rec->enc_key_size)); goto out; } if (key_rec->enc_key_size > ECRYPTFS_MAX_ENCRYPTED_KEY_BYTES) { rc = -EIO; - printk(KERN_ERR "%s: Encrypted key_size [%Zd] larger than " + printk(KERN_ERR "%s: Encrypted key_size [%zd] larger than " "the maximum key size [%d]\n", __func__, key_rec->enc_key_size, ECRYPTFS_MAX_ENCRYPTED_KEY_BYTES); @@ -403,6 +403,580 @@ out: } static int +ecryptfs_find_global_auth_tok_for_sig( + struct ecryptfs_global_auth_tok **global_auth_tok, + struct ecryptfs_mount_crypt_stat *mount_crypt_stat, char *sig) +{ + struct ecryptfs_global_auth_tok *walker; + int rc = 0; + + (*global_auth_tok) = NULL; + mutex_lock(&mount_crypt_stat->global_auth_tok_list_mutex); + list_for_each_entry(walker, + &mount_crypt_stat->global_auth_tok_list, + mount_crypt_stat_list) { + if (memcmp(walker->sig, sig, ECRYPTFS_SIG_SIZE_HEX) == 0) { + (*global_auth_tok) = walker; + goto out; + } + } + rc = -EINVAL; +out: + mutex_unlock(&mount_crypt_stat->global_auth_tok_list_mutex); + return rc; +} + +/** + * ecryptfs_find_auth_tok_for_sig + * @auth_tok: Set to the matching auth_tok; NULL if not found + * @crypt_stat: inode crypt_stat crypto context + * @sig: Sig of auth_tok to find + * + * For now, this function simply looks at the registered auth_tok's + * linked off the mount_crypt_stat, so all the auth_toks that can be + * used must be registered at mount time. This function could + * potentially try a lot harder to find auth_tok's (e.g., by calling + * out to ecryptfsd to dynamically retrieve an auth_tok object) so + * that static registration of auth_tok's will no longer be necessary. + * + * Returns zero on no error; non-zero on error + */ +static int +ecryptfs_find_auth_tok_for_sig( + struct ecryptfs_auth_tok **auth_tok, + struct ecryptfs_mount_crypt_stat *mount_crypt_stat, + char *sig) +{ + struct ecryptfs_global_auth_tok *global_auth_tok; + int rc = 0; + + (*auth_tok) = NULL; + if (ecryptfs_find_global_auth_tok_for_sig(&global_auth_tok, + mount_crypt_stat, sig)) { + struct key *auth_tok_key; + + rc = ecryptfs_keyring_auth_tok_for_sig(&auth_tok_key, auth_tok, + sig); + } else + (*auth_tok) = global_auth_tok->global_auth_tok; + return rc; +} + +/** + * write_tag_70_packet can gobble a lot of stack space. We stuff most + * of the function's parameters in a kmalloc'd struct to help reduce + * eCryptfs' overall stack usage. + */ +struct ecryptfs_write_tag_70_packet_silly_stack { + u8 cipher_code; + size_t max_packet_size; + size_t packet_size_len; + size_t block_aligned_filename_size; + size_t block_size; + size_t i; + size_t j; + size_t num_rand_bytes; + struct mutex *tfm_mutex; + char *block_aligned_filename; + struct ecryptfs_auth_tok *auth_tok; + struct scatterlist src_sg; + struct scatterlist dst_sg; + struct blkcipher_desc desc; + char iv[ECRYPTFS_MAX_IV_BYTES]; + char hash[ECRYPTFS_TAG_70_DIGEST_SIZE]; + char tmp_hash[ECRYPTFS_TAG_70_DIGEST_SIZE]; + struct hash_desc hash_desc; + struct scatterlist hash_sg; +}; + +/** + * write_tag_70_packet - Write encrypted filename (EFN) packet against FNEK + * @filename: NULL-terminated filename string + * + * This is the simplest mechanism for achieving filename encryption in + * eCryptfs. It encrypts the given filename with the mount-wide + * filename encryption key (FNEK) and stores it in a packet to @dest, + * which the callee will encode and write directly into the dentry + * name. + */ +int +ecryptfs_write_tag_70_packet(char *dest, size_t *remaining_bytes, + size_t *packet_size, + struct ecryptfs_mount_crypt_stat *mount_crypt_stat, + char *filename, size_t filename_size) +{ + struct ecryptfs_write_tag_70_packet_silly_stack *s; + int rc = 0; + + s = kmalloc(sizeof(*s), GFP_KERNEL); + if (!s) { + printk(KERN_ERR "%s: Out of memory whilst trying to kmalloc " + "[%zd] bytes of kernel memory\n", __func__, sizeof(*s)); + goto out; + } + s->desc.flags = CRYPTO_TFM_REQ_MAY_SLEEP; + (*packet_size) = 0; + rc = ecryptfs_get_tfm_and_mutex_for_cipher_name( + &s->desc.tfm, + &s->tfm_mutex, mount_crypt_stat->global_default_fn_cipher_name); + if (unlikely(rc)) { + printk(KERN_ERR "Internal error whilst attempting to get " + "tfm and mutex for cipher name [%s]; rc = [%d]\n", + mount_crypt_stat->global_default_fn_cipher_name, rc); + goto out; + } + mutex_lock(s->tfm_mutex); + s->block_size = crypto_blkcipher_blocksize(s->desc.tfm); + /* Plus one for the \0 separator between the random prefix + * and the plaintext filename */ + s->num_rand_bytes = (ECRYPTFS_FILENAME_MIN_RANDOM_PREPEND_BYTES + 1); + s->block_aligned_filename_size = (s->num_rand_bytes + filename_size); + if ((s->block_aligned_filename_size % s->block_size) != 0) { + s->num_rand_bytes += (s->block_size + - (s->block_aligned_filename_size + % s->block_size)); + s->block_aligned_filename_size = (s->num_rand_bytes + + filename_size); + } + /* Octet 0: Tag 70 identifier + * Octets 1-N1: Tag 70 packet size (includes cipher identifier + * and block-aligned encrypted filename size) + * Octets N1-N2: FNEK sig (ECRYPTFS_SIG_SIZE) + * Octet N2-N3: Cipher identifier (1 octet) + * Octets N3-N4: Block-aligned encrypted filename + * - Consists of a minimum number of random characters, a \0 + * separator, and then the filename */ + s->max_packet_size = (1 /* Tag 70 identifier */ + + 3 /* Max Tag 70 packet size */ + + ECRYPTFS_SIG_SIZE /* FNEK sig */ + + 1 /* Cipher identifier */ + + s->block_aligned_filename_size); + if (dest == NULL) { + (*packet_size) = s->max_packet_size; + goto out_unlock; + } + if (s->max_packet_size > (*remaining_bytes)) { + printk(KERN_WARNING "%s: Require [%zd] bytes to write; only " + "[%zd] available\n", __func__, s->max_packet_size, + (*remaining_bytes)); + rc = -EINVAL; + goto out_unlock; + } + s->block_aligned_filename = kzalloc(s->block_aligned_filename_size, + GFP_KERNEL); + if (!s->block_aligned_filename) { + printk(KERN_ERR "%s: Out of kernel memory whilst attempting to " + "kzalloc [%zd] bytes\n", __func__, + s->block_aligned_filename_size); + rc = -ENOMEM; + goto out_unlock; + } + s->i = 0; + dest[s->i++] = ECRYPTFS_TAG_70_PACKET_TYPE; + rc = ecryptfs_write_packet_length(&dest[s->i], + (ECRYPTFS_SIG_SIZE + + 1 /* Cipher code */ + + s->block_aligned_filename_size), + &s->packet_size_len); + if (rc) { + printk(KERN_ERR "%s: Error generating tag 70 packet " + "header; cannot generate packet length; rc = [%d]\n", + __func__, rc); + goto out_free_unlock; + } + s->i += s->packet_size_len; + ecryptfs_from_hex(&dest[s->i], + mount_crypt_stat->global_default_fnek_sig, + ECRYPTFS_SIG_SIZE); + s->i += ECRYPTFS_SIG_SIZE; + s->cipher_code = ecryptfs_code_for_cipher_string( + mount_crypt_stat->global_default_fn_cipher_name, + mount_crypt_stat->global_default_fn_cipher_key_bytes); + if (s->cipher_code == 0) { + printk(KERN_WARNING "%s: Unable to generate code for " + "cipher [%s] with key bytes [%zd]\n", __func__, + mount_crypt_stat->global_default_fn_cipher_name, + mount_crypt_stat->global_default_fn_cipher_key_bytes); + rc = -EINVAL; + goto out_free_unlock; + } + dest[s->i++] = s->cipher_code; + rc = ecryptfs_find_auth_tok_for_sig( + &s->auth_tok, mount_crypt_stat, + mount_crypt_stat->global_default_fnek_sig); + if (rc) { + printk(KERN_ERR "%s: Error attempting to find auth tok for " + "fnek sig [%s]; rc = [%d]\n", __func__, + mount_crypt_stat->global_default_fnek_sig, rc); + goto out_free_unlock; + } + /* TODO: Support other key modules than passphrase for + * filename encryption */ + BUG_ON(s->auth_tok->token_type != ECRYPTFS_PASSWORD); + sg_init_one( + &s->hash_sg, + (u8 *)s->auth_tok->token.password.session_key_encryption_key, + s->auth_tok->token.password.session_key_encryption_key_bytes); + s->hash_desc.flags = CRYPTO_TFM_REQ_MAY_SLEEP; + s->hash_desc.tfm = crypto_alloc_hash(ECRYPTFS_TAG_70_DIGEST, 0, + CRYPTO_ALG_ASYNC); + if (IS_ERR(s->hash_desc.tfm)) { + rc = PTR_ERR(s->hash_desc.tfm); + printk(KERN_ERR "%s: Error attempting to " + "allocate hash crypto context; rc = [%d]\n", + __func__, rc); + goto out_free_unlock; + } + rc = crypto_hash_init(&s->hash_desc); + if (rc) { + printk(KERN_ERR + "%s: Error initializing crypto hash; rc = [%d]\n", + __func__, rc); + goto out_release_free_unlock; + } + rc = crypto_hash_update( + &s->hash_desc, &s->hash_sg, + s->auth_tok->token.password.session_key_encryption_key_bytes); + if (rc) { + printk(KERN_ERR + "%s: Error updating crypto hash; rc = [%d]\n", + __func__, rc); + goto out_release_free_unlock; + } + rc = crypto_hash_final(&s->hash_desc, s->hash); + if (rc) { + printk(KERN_ERR + "%s: Error finalizing crypto hash; rc = [%d]\n", + __func__, rc); + goto out_release_free_unlock; + } + for (s->j = 0; s->j < (s->num_rand_bytes - 1); s->j++) { + s->block_aligned_filename[s->j] = + s->hash[(s->j % ECRYPTFS_TAG_70_DIGEST_SIZE)]; + if ((s->j % ECRYPTFS_TAG_70_DIGEST_SIZE) + == (ECRYPTFS_TAG_70_DIGEST_SIZE - 1)) { + sg_init_one(&s->hash_sg, (u8 *)s->hash, + ECRYPTFS_TAG_70_DIGEST_SIZE); + rc = crypto_hash_init(&s->hash_desc); + if (rc) { + printk(KERN_ERR + "%s: Error initializing crypto hash; " + "rc = [%d]\n", __func__, rc); + goto out_release_free_unlock; + } + rc = crypto_hash_update(&s->hash_desc, &s->hash_sg, + ECRYPTFS_TAG_70_DIGEST_SIZE); + if (rc) { + printk(KERN_ERR + "%s: Error updating crypto hash; " + "rc = [%d]\n", __func__, rc); + goto out_release_free_unlock; + } + rc = crypto_hash_final(&s->hash_desc, s->tmp_hash); + if (rc) { + printk(KERN_ERR + "%s: Error finalizing crypto hash; " + "rc = [%d]\n", __func__, rc); + goto out_release_free_unlock; + } + memcpy(s->hash, s->tmp_hash, + ECRYPTFS_TAG_70_DIGEST_SIZE); + } + if (s->block_aligned_filename[s->j] == '\0') + s->block_aligned_filename[s->j] = ECRYPTFS_NON_NULL; + } + memcpy(&s->block_aligned_filename[s->num_rand_bytes], filename, + filename_size); + rc = virt_to_scatterlist(s->block_aligned_filename, + s->block_aligned_filename_size, &s->src_sg, 1); + if (rc != 1) { + printk(KERN_ERR "%s: Internal error whilst attempting to " + "convert filename memory to scatterlist; " + "expected rc = 1; got rc = [%d]. " + "block_aligned_filename_size = [%zd]\n", __func__, rc, + s->block_aligned_filename_size); + goto out_release_free_unlock; + } + rc = virt_to_scatterlist(&dest[s->i], s->block_aligned_filename_size, + &s->dst_sg, 1); + if (rc != 1) { + printk(KERN_ERR "%s: Internal error whilst attempting to " + "convert encrypted filename memory to scatterlist; " + "expected rc = 1; got rc = [%d]. " + "block_aligned_filename_size = [%zd]\n", __func__, rc, + s->block_aligned_filename_size); + goto out_release_free_unlock; + } + /* The characters in the first block effectively do the job + * of the IV here, so we just use 0's for the IV. Note the + * constraint that ECRYPTFS_FILENAME_MIN_RANDOM_PREPEND_BYTES + * >= ECRYPTFS_MAX_IV_BYTES. */ + memset(s->iv, 0, ECRYPTFS_MAX_IV_BYTES); + s->desc.info = s->iv; + rc = crypto_blkcipher_setkey( + s->desc.tfm, + s->auth_tok->token.password.session_key_encryption_key, + mount_crypt_stat->global_default_fn_cipher_key_bytes); + if (rc < 0) { + printk(KERN_ERR "%s: Error setting key for crypto context; " + "rc = [%d]. s->auth_tok->token.password.session_key_" + "encryption_key = [0x%p]; mount_crypt_stat->" + "global_default_fn_cipher_key_bytes = [%zd]\n", __func__, + rc, + s->auth_tok->token.password.session_key_encryption_key, + mount_crypt_stat->global_default_fn_cipher_key_bytes); + goto out_release_free_unlock; + } + rc = crypto_blkcipher_encrypt_iv(&s->desc, &s->dst_sg, &s->src_sg, + s->block_aligned_filename_size); + if (rc) { + printk(KERN_ERR "%s: Error attempting to encrypt filename; " + "rc = [%d]\n", __func__, rc); + goto out_release_free_unlock; + } + s->i += s->block_aligned_filename_size; + (*packet_size) = s->i; + (*remaining_bytes) -= (*packet_size); +out_release_free_unlock: + crypto_free_hash(s->hash_desc.tfm); +out_free_unlock: + memset(s->block_aligned_filename, 0, s->block_aligned_filename_size); + kfree(s->block_aligned_filename); +out_unlock: + mutex_unlock(s->tfm_mutex); +out: + kfree(s); + return rc; +} + +struct ecryptfs_parse_tag_70_packet_silly_stack { + u8 cipher_code; + size_t max_packet_size; + size_t packet_size_len; + size_t parsed_tag_70_packet_size; + size_t block_aligned_filename_size; + size_t block_size; + size_t i; + struct mutex *tfm_mutex; + char *decrypted_filename; + struct ecryptfs_auth_tok *auth_tok; + struct scatterlist src_sg; + struct scatterlist dst_sg; + struct blkcipher_desc desc; + char fnek_sig_hex[ECRYPTFS_SIG_SIZE_HEX + 1]; + char iv[ECRYPTFS_MAX_IV_BYTES]; + char cipher_string[ECRYPTFS_MAX_CIPHER_NAME_SIZE]; +}; + +/** + * parse_tag_70_packet - Parse and process FNEK-encrypted passphrase packet + * @filename: This function kmalloc's the memory for the filename + * @filename_size: This function sets this to the amount of memory + * kmalloc'd for the filename + * @packet_size: This function sets this to the the number of octets + * in the packet parsed + * @mount_crypt_stat: The mount-wide cryptographic context + * @data: The memory location containing the start of the tag 70 + * packet + * @max_packet_size: The maximum legal size of the packet to be parsed + * from @data + * + * Returns zero on success; non-zero otherwise + */ +int +ecryptfs_parse_tag_70_packet(char **filename, size_t *filename_size, + size_t *packet_size, + struct ecryptfs_mount_crypt_stat *mount_crypt_stat, + char *data, size_t max_packet_size) +{ + struct ecryptfs_parse_tag_70_packet_silly_stack *s; + int rc = 0; + + (*packet_size) = 0; + (*filename_size) = 0; + (*filename) = NULL; + s = kmalloc(sizeof(*s), GFP_KERNEL); + if (!s) { + printk(KERN_ERR "%s: Out of memory whilst trying to kmalloc " + "[%zd] bytes of kernel memory\n", __func__, sizeof(*s)); + goto out; + } + s->desc.flags = CRYPTO_TFM_REQ_MAY_SLEEP; + if (max_packet_size < (1 + 1 + ECRYPTFS_SIG_SIZE + 1 + 1)) { + printk(KERN_WARNING "%s: max_packet_size is [%zd]; it must be " + "at least [%d]\n", __func__, max_packet_size, + (1 + 1 + ECRYPTFS_SIG_SIZE + 1 + 1)); + rc = -EINVAL; + goto out; + } + /* Octet 0: Tag 70 identifier + * Octets 1-N1: Tag 70 packet size (includes cipher identifier + * and block-aligned encrypted filename size) + * Octets N1-N2: FNEK sig (ECRYPTFS_SIG_SIZE) + * Octet N2-N3: Cipher identifier (1 octet) + * Octets N3-N4: Block-aligned encrypted filename + * - Consists of a minimum number of random numbers, a \0 + * separator, and then the filename */ + if (data[(*packet_size)++] != ECRYPTFS_TAG_70_PACKET_TYPE) { + printk(KERN_WARNING "%s: Invalid packet tag [0x%.2x]; must be " + "tag [0x%.2x]\n", __func__, + data[((*packet_size) - 1)], ECRYPTFS_TAG_70_PACKET_TYPE); + rc = -EINVAL; + goto out; + } + rc = ecryptfs_parse_packet_length(&data[(*packet_size)], + &s->parsed_tag_70_packet_size, + &s->packet_size_len); + if (rc) { + printk(KERN_WARNING "%s: Error parsing packet length; " + "rc = [%d]\n", __func__, rc); + goto out; + } + s->block_aligned_filename_size = (s->parsed_tag_70_packet_size + - ECRYPTFS_SIG_SIZE - 1); + if ((1 + s->packet_size_len + s->parsed_tag_70_packet_size) + > max_packet_size) { + printk(KERN_WARNING "%s: max_packet_size is [%zd]; real packet " + "size is [%zd]\n", __func__, max_packet_size, + (1 + s->packet_size_len + 1 + + s->block_aligned_filename_size)); + rc = -EINVAL; + goto out; + } + (*packet_size) += s->packet_size_len; + ecryptfs_to_hex(s->fnek_sig_hex, &data[(*packet_size)], + ECRYPTFS_SIG_SIZE); + s->fnek_sig_hex[ECRYPTFS_SIG_SIZE_HEX] = '\0'; + (*packet_size) += ECRYPTFS_SIG_SIZE; + s->cipher_code = data[(*packet_size)++]; + rc = ecryptfs_cipher_code_to_string(s->cipher_string, s->cipher_code); + if (rc) { + printk(KERN_WARNING "%s: Cipher code [%d] is invalid\n", + __func__, s->cipher_code); + goto out; + } + rc = ecryptfs_get_tfm_and_mutex_for_cipher_name(&s->desc.tfm, + &s->tfm_mutex, + s->cipher_string); + if (unlikely(rc)) { + printk(KERN_ERR "Internal error whilst attempting to get " + "tfm and mutex for cipher name [%s]; rc = [%d]\n", + s->cipher_string, rc); + goto out; + } + mutex_lock(s->tfm_mutex); + rc = virt_to_scatterlist(&data[(*packet_size)], + s->block_aligned_filename_size, &s->src_sg, 1); + if (rc != 1) { + printk(KERN_ERR "%s: Internal error whilst attempting to " + "convert encrypted filename memory to scatterlist; " + "expected rc = 1; got rc = [%d]. " + "block_aligned_filename_size = [%zd]\n", __func__, rc, + s->block_aligned_filename_size); + goto out_unlock; + } + (*packet_size) += s->block_aligned_filename_size; + s->decrypted_filename = kmalloc(s->block_aligned_filename_size, + GFP_KERNEL); + if (!s->decrypted_filename) { + printk(KERN_ERR "%s: Out of memory whilst attempting to " + "kmalloc [%zd] bytes\n", __func__, + s->block_aligned_filename_size); + rc = -ENOMEM; + goto out_unlock; + } + rc = virt_to_scatterlist(s->decrypted_filename, + s->block_aligned_filename_size, &s->dst_sg, 1); + if (rc != 1) { + printk(KERN_ERR "%s: Internal error whilst attempting to " + "convert decrypted filename memory to scatterlist; " + "expected rc = 1; got rc = [%d]. " + "block_aligned_filename_size = [%zd]\n", __func__, rc, + s->block_aligned_filename_size); + goto out_free_unlock; + } + /* The characters in the first block effectively do the job of + * the IV here, so we just use 0's for the IV. Note the + * constraint that ECRYPTFS_FILENAME_MIN_RANDOM_PREPEND_BYTES + * >= ECRYPTFS_MAX_IV_BYTES. */ + memset(s->iv, 0, ECRYPTFS_MAX_IV_BYTES); + s->desc.info = s->iv; + rc = ecryptfs_find_auth_tok_for_sig(&s->auth_tok, mount_crypt_stat, + s->fnek_sig_hex); + if (rc) { + printk(KERN_ERR "%s: Error attempting to find auth tok for " + "fnek sig [%s]; rc = [%d]\n", __func__, s->fnek_sig_hex, + rc); + goto out_free_unlock; + } + /* TODO: Support other key modules than passphrase for + * filename encryption */ + BUG_ON(s->auth_tok->token_type != ECRYPTFS_PASSWORD); + rc = crypto_blkcipher_setkey( + s->desc.tfm, + s->auth_tok->token.password.session_key_encryption_key, + mount_crypt_stat->global_default_fn_cipher_key_bytes); + if (rc < 0) { + printk(KERN_ERR "%s: Error setting key for crypto context; " + "rc = [%d]. s->auth_tok->token.password.session_key_" + "encryption_key = [0x%p]; mount_crypt_stat->" + "global_default_fn_cipher_key_bytes = [%zd]\n", __func__, + rc, + s->auth_tok->token.password.session_key_encryption_key, + mount_crypt_stat->global_default_fn_cipher_key_bytes); + goto out_free_unlock; + } + rc = crypto_blkcipher_decrypt_iv(&s->desc, &s->dst_sg, &s->src_sg, + s->block_aligned_filename_size); + if (rc) { + printk(KERN_ERR "%s: Error attempting to decrypt filename; " + "rc = [%d]\n", __func__, rc); + goto out_free_unlock; + } + s->i = 0; + while (s->decrypted_filename[s->i] != '\0' + && s->i < s->block_aligned_filename_size) + s->i++; + if (s->i == s->block_aligned_filename_size) { + printk(KERN_WARNING "%s: Invalid tag 70 packet; could not " + "find valid separator between random characters and " + "the filename\n", __func__); + rc = -EINVAL; + goto out_free_unlock; + } + s->i++; + (*filename_size) = (s->block_aligned_filename_size - s->i); + if (!((*filename_size) > 0 && (*filename_size < PATH_MAX))) { + printk(KERN_WARNING "%s: Filename size is [%zd], which is " + "invalid\n", __func__, (*filename_size)); + rc = -EINVAL; + goto out_free_unlock; + } + (*filename) = kmalloc(((*filename_size) + 1), GFP_KERNEL); + if (!(*filename)) { + printk(KERN_ERR "%s: Out of memory whilst attempting to " + "kmalloc [%zd] bytes\n", __func__, + ((*filename_size) + 1)); + rc = -ENOMEM; + goto out_free_unlock; + } + memcpy((*filename), &s->decrypted_filename[s->i], (*filename_size)); + (*filename)[(*filename_size)] = '\0'; +out_free_unlock: + kfree(s->decrypted_filename); +out_unlock: + mutex_unlock(s->tfm_mutex); +out: + if (rc) { + (*packet_size) = 0; + (*filename_size) = 0; + (*filename) = NULL; + } + kfree(s); + return rc; +} + +static int ecryptfs_get_auth_tok_sig(char **sig, struct ecryptfs_auth_tok *auth_tok) { int rc = 0; @@ -897,30 +1471,6 @@ out: return rc; } -static int -ecryptfs_find_global_auth_tok_for_sig( - struct ecryptfs_global_auth_tok **global_auth_tok, - struct ecryptfs_mount_crypt_stat *mount_crypt_stat, char *sig) -{ - struct ecryptfs_global_auth_tok *walker; - int rc = 0; - - (*global_auth_tok) = NULL; - mutex_lock(&mount_crypt_stat->global_auth_tok_list_mutex); - list_for_each_entry(walker, - &mount_crypt_stat->global_auth_tok_list, - mount_crypt_stat_list) { - if (memcmp(walker->sig, sig, ECRYPTFS_SIG_SIZE_HEX) == 0) { - (*global_auth_tok) = walker; - goto out; - } - } - rc = -EINVAL; -out: - mutex_unlock(&mount_crypt_stat->global_auth_tok_list_mutex); - return rc; -} - /** * ecryptfs_verify_version * @version: The version number to confirm @@ -990,43 +1540,6 @@ out: } /** - * ecryptfs_find_auth_tok_for_sig - * @auth_tok: Set to the matching auth_tok; NULL if not found - * @crypt_stat: inode crypt_stat crypto context - * @sig: Sig of auth_tok to find - * - * For now, this function simply looks at the registered auth_tok's - * linked off the mount_crypt_stat, so all the auth_toks that can be - * used must be registered at mount time. This function could - * potentially try a lot harder to find auth_tok's (e.g., by calling - * out to ecryptfsd to dynamically retrieve an auth_tok object) so - * that static registration of auth_tok's will no longer be necessary. - * - * Returns zero on no error; non-zero on error - */ -static int -ecryptfs_find_auth_tok_for_sig( - struct ecryptfs_auth_tok **auth_tok, - struct ecryptfs_crypt_stat *crypt_stat, char *sig) -{ - struct ecryptfs_mount_crypt_stat *mount_crypt_stat = - crypt_stat->mount_crypt_stat; - struct ecryptfs_global_auth_tok *global_auth_tok; - int rc = 0; - - (*auth_tok) = NULL; - if (ecryptfs_find_global_auth_tok_for_sig(&global_auth_tok, - mount_crypt_stat, sig)) { - struct key *auth_tok_key; - - rc = ecryptfs_keyring_auth_tok_for_sig(&auth_tok_key, auth_tok, - sig); - } else - (*auth_tok) = global_auth_tok->global_auth_tok; - return rc; -} - -/** * decrypt_passphrase_encrypted_session_key - Decrypt the session key with the given auth_tok. * @auth_tok: The passphrase authentication token to use to encrypt the FEK * @crypt_stat: The cryptographic context @@ -1256,7 +1769,8 @@ find_next_matching_auth_tok: rc = -EINVAL; goto out_wipe_list; } - ecryptfs_find_auth_tok_for_sig(&matching_auth_tok, crypt_stat, + ecryptfs_find_auth_tok_for_sig(&matching_auth_tok, + crypt_stat->mount_crypt_stat, candidate_auth_tok_sig); if (matching_auth_tok) { found_auth_tok = 1; @@ -1336,7 +1850,9 @@ pki_encrypt_session_key(struct ecryptfs_auth_tok *auth_tok, int rc; rc = write_tag_66_packet(auth_tok->token.private_key.signature, - ecryptfs_code_for_cipher_string(crypt_stat), + ecryptfs_code_for_cipher_string( + crypt_stat->cipher, + crypt_stat->key_size), crypt_stat, &payload, &payload_len); if (rc) { ecryptfs_printk(KERN_ERR, "Error generating tag 66 packet\n"); @@ -1696,7 +2212,8 @@ encrypted_session_key_set: dest[(*packet_size)++] = 0x04; /* version 4 */ /* TODO: Break from RFC2440 so that arbitrary ciphers can be * specified with strings */ - cipher_code = ecryptfs_code_for_cipher_string(crypt_stat); + cipher_code = ecryptfs_code_for_cipher_string(crypt_stat->cipher, + crypt_stat->key_size); if (cipher_code == 0) { ecryptfs_printk(KERN_WARNING, "Unable to generate code for " "cipher [%s]\n", crypt_stat->cipher); diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c index fd630713c5c7..789cf2e1be1e 100644 --- a/fs/ecryptfs/main.c +++ b/fs/ecryptfs/main.c @@ -206,7 +206,9 @@ enum { ecryptfs_opt_sig, ecryptfs_opt_ecryptfs_sig, ecryptfs_opt_cipher, ecryptfs_opt_ecryptfs_cipher, ecryptfs_opt_ecryptfs_key_bytes, ecryptfs_opt_passthrough, ecryptfs_opt_xattr_metadata, - ecryptfs_opt_encrypted_view, ecryptfs_opt_err }; + ecryptfs_opt_encrypted_view, ecryptfs_opt_fnek_sig, + ecryptfs_opt_fn_cipher, ecryptfs_opt_fn_cipher_key_bytes, + ecryptfs_opt_err }; static const match_table_t tokens = { {ecryptfs_opt_sig, "sig=%s"}, @@ -217,6 +219,9 @@ static const match_table_t tokens = { {ecryptfs_opt_passthrough, "ecryptfs_passthrough"}, {ecryptfs_opt_xattr_metadata, "ecryptfs_xattr_metadata"}, {ecryptfs_opt_encrypted_view, "ecryptfs_encrypted_view"}, + {ecryptfs_opt_fnek_sig, "ecryptfs_fnek_sig=%s"}, + {ecryptfs_opt_fn_cipher, "ecryptfs_fn_cipher=%s"}, + {ecryptfs_opt_fn_cipher_key_bytes, "ecryptfs_fn_key_bytes=%u"}, {ecryptfs_opt_err, NULL} }; @@ -281,8 +286,11 @@ static int ecryptfs_parse_options(struct super_block *sb, char *options) int rc = 0; int sig_set = 0; int cipher_name_set = 0; + int fn_cipher_name_set = 0; int cipher_key_bytes; int cipher_key_bytes_set = 0; + int fn_cipher_key_bytes; + int fn_cipher_key_bytes_set = 0; struct ecryptfs_mount_crypt_stat *mount_crypt_stat = &ecryptfs_superblock_to_private(sb)->mount_crypt_stat; substring_t args[MAX_OPT_ARGS]; @@ -290,7 +298,12 @@ static int ecryptfs_parse_options(struct super_block *sb, char *options) char *sig_src; char *cipher_name_dst; char *cipher_name_src; + char *fn_cipher_name_dst; + char *fn_cipher_name_src; + char *fnek_dst; + char *fnek_src; char *cipher_key_bytes_src; + char *fn_cipher_key_bytes_src; if (!options) { rc = -EINVAL; @@ -322,10 +335,7 @@ static int ecryptfs_parse_options(struct super_block *sb, char *options) global_default_cipher_name; strncpy(cipher_name_dst, cipher_name_src, ECRYPTFS_MAX_CIPHER_NAME_SIZE); - ecryptfs_printk(KERN_DEBUG, - "The mount_crypt_stat " - "global_default_cipher_name set to: " - "[%s]\n", cipher_name_dst); + cipher_name_dst[ECRYPTFS_MAX_CIPHER_NAME_SIZE] = '\0'; cipher_name_set = 1; break; case ecryptfs_opt_ecryptfs_key_bytes: @@ -335,11 +345,6 @@ static int ecryptfs_parse_options(struct super_block *sb, char *options) &cipher_key_bytes_src, 0); mount_crypt_stat->global_default_cipher_key_size = cipher_key_bytes; - ecryptfs_printk(KERN_DEBUG, - "The mount_crypt_stat " - "global_default_cipher_key_size " - "set to: [%d]\n", mount_crypt_stat-> - global_default_cipher_key_size); cipher_key_bytes_set = 1; break; case ecryptfs_opt_passthrough: @@ -356,11 +361,51 @@ static int ecryptfs_parse_options(struct super_block *sb, char *options) mount_crypt_stat->flags |= ECRYPTFS_ENCRYPTED_VIEW_ENABLED; break; + case ecryptfs_opt_fnek_sig: + fnek_src = args[0].from; + fnek_dst = + mount_crypt_stat->global_default_fnek_sig; + strncpy(fnek_dst, fnek_src, ECRYPTFS_SIG_SIZE_HEX); + mount_crypt_stat->global_default_fnek_sig[ + ECRYPTFS_SIG_SIZE_HEX] = '\0'; + rc = ecryptfs_add_global_auth_tok( + mount_crypt_stat, + mount_crypt_stat->global_default_fnek_sig); + if (rc) { + printk(KERN_ERR "Error attempting to register " + "global fnek sig [%s]; rc = [%d]\n", + mount_crypt_stat->global_default_fnek_sig, + rc); + goto out; + } + mount_crypt_stat->flags |= + (ECRYPTFS_GLOBAL_ENCRYPT_FILENAMES + | ECRYPTFS_GLOBAL_ENCFN_USE_MOUNT_FNEK); + break; + case ecryptfs_opt_fn_cipher: + fn_cipher_name_src = args[0].from; + fn_cipher_name_dst = + mount_crypt_stat->global_default_fn_cipher_name; + strncpy(fn_cipher_name_dst, fn_cipher_name_src, + ECRYPTFS_MAX_CIPHER_NAME_SIZE); + mount_crypt_stat->global_default_fn_cipher_name[ + ECRYPTFS_MAX_CIPHER_NAME_SIZE] = '\0'; + fn_cipher_name_set = 1; + break; + case ecryptfs_opt_fn_cipher_key_bytes: + fn_cipher_key_bytes_src = args[0].from; + fn_cipher_key_bytes = + (int)simple_strtol(fn_cipher_key_bytes_src, + &fn_cipher_key_bytes_src, 0); + mount_crypt_stat->global_default_fn_cipher_key_bytes = + fn_cipher_key_bytes; + fn_cipher_key_bytes_set = 1; + break; case ecryptfs_opt_err: default: - ecryptfs_printk(KERN_WARNING, - "eCryptfs: unrecognized option '%s'\n", - p); + printk(KERN_WARNING + "%s: eCryptfs: unrecognized option [%s]\n", + __func__, p); } } if (!sig_set) { @@ -374,33 +419,60 @@ static int ecryptfs_parse_options(struct super_block *sb, char *options) int cipher_name_len = strlen(ECRYPTFS_DEFAULT_CIPHER); BUG_ON(cipher_name_len >= ECRYPTFS_MAX_CIPHER_NAME_SIZE); - strcpy(mount_crypt_stat->global_default_cipher_name, ECRYPTFS_DEFAULT_CIPHER); } - if (!cipher_key_bytes_set) { + if ((mount_crypt_stat->flags & ECRYPTFS_GLOBAL_ENCRYPT_FILENAMES) + && !fn_cipher_name_set) + strcpy(mount_crypt_stat->global_default_fn_cipher_name, + mount_crypt_stat->global_default_cipher_name); + if (!cipher_key_bytes_set) mount_crypt_stat->global_default_cipher_key_size = 0; - } + if ((mount_crypt_stat->flags & ECRYPTFS_GLOBAL_ENCRYPT_FILENAMES) + && !fn_cipher_key_bytes_set) + mount_crypt_stat->global_default_fn_cipher_key_bytes = + mount_crypt_stat->global_default_cipher_key_size; mutex_lock(&key_tfm_list_mutex); if (!ecryptfs_tfm_exists(mount_crypt_stat->global_default_cipher_name, - NULL)) + NULL)) { rc = ecryptfs_add_new_key_tfm( NULL, mount_crypt_stat->global_default_cipher_name, mount_crypt_stat->global_default_cipher_key_size); - mutex_unlock(&key_tfm_list_mutex); - if (rc) { - printk(KERN_ERR "Error attempting to initialize cipher with " - "name = [%s] and key size = [%td]; rc = [%d]\n", - mount_crypt_stat->global_default_cipher_name, - mount_crypt_stat->global_default_cipher_key_size, rc); - rc = -EINVAL; - goto out; + if (rc) { + printk(KERN_ERR "Error attempting to initialize " + "cipher with name = [%s] and key size = [%td]; " + "rc = [%d]\n", + mount_crypt_stat->global_default_cipher_name, + mount_crypt_stat->global_default_cipher_key_size, + rc); + rc = -EINVAL; + mutex_unlock(&key_tfm_list_mutex); + goto out; + } } + if ((mount_crypt_stat->flags & ECRYPTFS_GLOBAL_ENCRYPT_FILENAMES) + && !ecryptfs_tfm_exists( + mount_crypt_stat->global_default_fn_cipher_name, NULL)) { + rc = ecryptfs_add_new_key_tfm( + NULL, mount_crypt_stat->global_default_fn_cipher_name, + mount_crypt_stat->global_default_fn_cipher_key_bytes); + if (rc) { + printk(KERN_ERR "Error attempting to initialize " + "cipher with name = [%s] and key size = [%td]; " + "rc = [%d]\n", + mount_crypt_stat->global_default_fn_cipher_name, + mount_crypt_stat->global_default_fn_cipher_key_bytes, + rc); + rc = -EINVAL; + mutex_unlock(&key_tfm_list_mutex); + goto out; + } + } + mutex_unlock(&key_tfm_list_mutex); rc = ecryptfs_init_global_auth_toks(mount_crypt_stat); - if (rc) { + if (rc) printk(KERN_WARNING "One or more global auth toks could not " "properly register; rc = [%d]\n", rc); - } out: return rc; } diff --git a/fs/ecryptfs/messaging.c b/fs/ecryptfs/messaging.c index 6913f727624d..96ef51489e01 100644 --- a/fs/ecryptfs/messaging.c +++ b/fs/ecryptfs/messaging.c @@ -193,7 +193,7 @@ ecryptfs_spawn_daemon(struct ecryptfs_daemon **daemon, uid_t euid, (*daemon) = kzalloc(sizeof(**daemon), GFP_KERNEL); if (!(*daemon)) { rc = -ENOMEM; - printk(KERN_ERR "%s: Failed to allocate [%Zd] bytes of " + printk(KERN_ERR "%s: Failed to allocate [%zd] bytes of " "GFP_KERNEL memory\n", __func__, sizeof(**daemon)); goto out; } @@ -435,7 +435,7 @@ int ecryptfs_process_response(struct ecryptfs_message *msg, uid_t euid, msg_ctx->msg = kmalloc(msg_size, GFP_KERNEL); if (!msg_ctx->msg) { rc = -ENOMEM; - printk(KERN_ERR "%s: Failed to allocate [%Zd] bytes of " + printk(KERN_ERR "%s: Failed to allocate [%zd] bytes of " "GFP_KERNEL memory\n", __func__, msg_size); goto unlock; } diff --git a/fs/ecryptfs/miscdev.c b/fs/ecryptfs/miscdev.c index efd95a0ed1ea..a67fea655f49 100644 --- a/fs/ecryptfs/miscdev.c +++ b/fs/ecryptfs/miscdev.c @@ -199,7 +199,7 @@ int ecryptfs_send_miscdev(char *data, size_t data_size, if (!msg_ctx->msg) { rc = -ENOMEM; printk(KERN_ERR "%s: Out of memory whilst attempting " - "to kmalloc(%Zd, GFP_KERNEL)\n", __func__, + "to kmalloc(%zd, GFP_KERNEL)\n", __func__, (sizeof(*msg_ctx->msg) + data_size)); goto out_unlock; } @@ -322,7 +322,7 @@ check_list: if (count < total_length) { rc = 0; printk(KERN_WARNING "%s: Only given user buffer of " - "size [%Zd], but we need [%Zd] to read the " + "size [%zd], but we need [%zd] to read the " "pending message\n", __func__, count, total_length); goto out_unlock_msg_ctx; } @@ -376,7 +376,7 @@ static int ecryptfs_miscdev_response(char *data, size_t data_size, if ((sizeof(*msg) + msg->data_len) != data_size) { printk(KERN_WARNING "%s: (sizeof(*msg) + msg->data_len) = " - "[%Zd]; data_size = [%Zd]. Invalid packet.\n", __func__, + "[%zd]; data_size = [%zd]. Invalid packet.\n", __func__, (sizeof(*msg) + msg->data_len), data_size); rc = -EINVAL; goto out; @@ -421,7 +421,7 @@ ecryptfs_miscdev_write(struct file *file, const char __user *buf, data = kmalloc(count, GFP_KERNEL); if (!data) { printk(KERN_ERR "%s: Out of memory whilst attempting to " - "kmalloc([%Zd], GFP_KERNEL)\n", __func__, count); + "kmalloc([%zd], GFP_KERNEL)\n", __func__, count); goto out; } rc = copy_from_user(data, buf, count); @@ -436,8 +436,8 @@ ecryptfs_miscdev_write(struct file *file, const char __user *buf, case ECRYPTFS_MSG_RESPONSE: if (count < (1 + 4 + 1 + sizeof(struct ecryptfs_message))) { printk(KERN_WARNING "%s: Minimum acceptable packet " - "size is [%Zd], but amount of data written is " - "only [%Zd]. Discarding response packet.\n", + "size is [%zd], but amount of data written is " + "only [%zd]. Discarding response packet.\n", __func__, (1 + 4 + 1 + sizeof(struct ecryptfs_message)), count); @@ -455,9 +455,9 @@ ecryptfs_miscdev_write(struct file *file, const char __user *buf, } i += packet_size_length; if ((1 + 4 + packet_size_length + packet_size) != count) { - printk(KERN_WARNING "%s: (1 + packet_size_length([%Zd])" - " + packet_size([%Zd]))([%Zd]) != " - "count([%Zd]). Invalid packet format.\n", + printk(KERN_WARNING "%s: (1 + packet_size_length([%zd])" + " + packet_size([%zd]))([%zd]) != " + "count([%zd]). Invalid packet format.\n", __func__, packet_size_length, packet_size, (1 + packet_size_length + packet_size), count); goto out_free; diff --git a/fs/exec.c b/fs/exec.c index 9c33f542dc77..71a6efe5d8bd 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -232,13 +232,13 @@ static void flush_arg_page(struct linux_binprm *bprm, unsigned long pos, static int __bprm_mm_init(struct linux_binprm *bprm) { - int err = -ENOMEM; + int err; struct vm_area_struct *vma = NULL; struct mm_struct *mm = bprm->mm; bprm->vma = vma = kmem_cache_zalloc(vm_area_cachep, GFP_KERNEL); if (!vma) - goto err; + return -ENOMEM; down_write(&mm->mmap_sem); vma->vm_mm = mm; @@ -251,28 +251,20 @@ static int __bprm_mm_init(struct linux_binprm *bprm) */ vma->vm_end = STACK_TOP_MAX; vma->vm_start = vma->vm_end - PAGE_SIZE; - vma->vm_flags = VM_STACK_FLAGS; vma->vm_page_prot = vm_get_page_prot(vma->vm_flags); err = insert_vm_struct(mm, vma); - if (err) { - up_write(&mm->mmap_sem); + if (err) goto err; - } mm->stack_vm = mm->total_vm = 1; up_write(&mm->mmap_sem); - bprm->p = vma->vm_end - sizeof(void *); - return 0; - err: - if (vma) { - bprm->vma = NULL; - kmem_cache_free(vm_area_cachep, vma); - } - + up_write(&mm->mmap_sem); + bprm->vma = NULL; + kmem_cache_free(vm_area_cachep, vma); return err; } @@ -1694,7 +1686,7 @@ int get_dumpable(struct mm_struct *mm) return (ret >= 2) ? 2 : ret; } -int do_coredump(long signr, int exit_code, struct pt_regs * regs) +void do_coredump(long signr, int exit_code, struct pt_regs *regs) { struct core_state core_state; char corename[CORENAME_MAX_SIZE + 1]; @@ -1778,6 +1770,11 @@ int do_coredump(long signr, int exit_code, struct pt_regs * regs) if (ispipe) { helper_argv = argv_split(GFP_KERNEL, corename+1, &helper_argc); + if (!helper_argv) { + printk(KERN_WARNING "%s failed to allocate memory\n", + __func__); + goto fail_unlock; + } /* Terminate the string before the first option */ delimit = strchr(corename, ' '); if (delimit) @@ -1845,5 +1842,5 @@ fail_unlock: put_cred(cred); coredump_finish(mm); fail: - return retval; + return; } diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index b0537c827024..6c46c648430d 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -1225,11 +1225,11 @@ do { \ } while (0) #ifdef CONFIG_SMP -/* Each CPU can accumulate FBC_BATCH blocks in their local +/* Each CPU can accumulate percpu_counter_batch blocks in their local * counters. So we need to make sure we have free blocks more - * than FBC_BATCH * nr_cpu_ids. Also add a window of 4 times. + * than percpu_counter_batch * nr_cpu_ids. Also add a window of 4 times. */ -#define EXT4_FREEBLOCKS_WATERMARK (4 * (FBC_BATCH * nr_cpu_ids)) +#define EXT4_FREEBLOCKS_WATERMARK (4 * (percpu_counter_batch * nr_cpu_ids)) #else #define EXT4_FREEBLOCKS_WATERMARK 0 #endif diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 6702a49992a6..98d3fe7057ef 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -2498,7 +2498,7 @@ static int ext4_nonda_switch(struct super_block *sb) /* * switch to non delalloc mode if we are running low * on free block. The free block accounting via percpu - * counters can get slightly wrong with FBC_BATCH getting + * counters can get slightly wrong with percpu_counter_batch getting * accumulated on each CPU without updating global counters * Delalloc need an accurate free block accounting. So switch * to non delalloc when we are near to error range. diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index d0ff0b8cf309..e5eaa62fd17f 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -421,9 +421,6 @@ __writeback_single_inode(struct inode *inode, struct writeback_control *wbc) * If we're a pdlfush thread, then implement pdflush collision avoidance * against the entire list. * - * WB_SYNC_HOLD is a hack for sys_sync(): reattach the inode to sb->s_dirty so - * that it can be located for waiting on in __writeback_single_inode(). - * * If `bdi' is non-zero then we're being asked to writeback a specific queue. * This function assumes that the blockdev superblock's inodes are backed by * a variety of queues, so all inodes are searched. For other superblocks, @@ -443,6 +440,7 @@ void generic_sync_sb_inodes(struct super_block *sb, struct writeback_control *wbc) { const unsigned long start = jiffies; /* livelock avoidance */ + int sync = wbc->sync_mode == WB_SYNC_ALL; spin_lock(&inode_lock); if (!wbc->for_kupdate || list_empty(&sb->s_io)) @@ -499,10 +497,6 @@ void generic_sync_sb_inodes(struct super_block *sb, __iget(inode); pages_skipped = wbc->pages_skipped; __writeback_single_inode(inode, wbc); - if (wbc->sync_mode == WB_SYNC_HOLD) { - inode->dirtied_when = jiffies; - list_move(&inode->i_list, &sb->s_dirty); - } if (current_is_pdflush()) writeback_release(bdi); if (wbc->pages_skipped != pages_skipped) { @@ -523,7 +517,49 @@ void generic_sync_sb_inodes(struct super_block *sb, if (!list_empty(&sb->s_more_io)) wbc->more_io = 1; } - spin_unlock(&inode_lock); + + if (sync) { + struct inode *inode, *old_inode = NULL; + + /* + * Data integrity sync. Must wait for all pages under writeback, + * because there may have been pages dirtied before our sync + * call, but which had writeout started before we write it out. + * In which case, the inode may not be on the dirty list, but + * we still have to wait for that writeout. + */ + list_for_each_entry(inode, &sb->s_inodes, i_sb_list) { + struct address_space *mapping; + + if (inode->i_state & (I_FREEING|I_WILL_FREE)) + continue; + mapping = inode->i_mapping; + if (mapping->nrpages == 0) + continue; + __iget(inode); + spin_unlock(&inode_lock); + /* + * We hold a reference to 'inode' so it couldn't have + * been removed from s_inodes list while we dropped the + * inode_lock. We cannot iput the inode now as we can + * be holding the last reference and we cannot iput it + * under inode_lock. So we keep the reference and iput + * it later. + */ + iput(old_inode); + old_inode = inode; + + filemap_fdatawait(mapping); + + cond_resched(); + + spin_lock(&inode_lock); + } + spin_unlock(&inode_lock); + iput(old_inode); + } else + spin_unlock(&inode_lock); + return; /* Leave any unwritten inodes on s_io */ } EXPORT_SYMBOL_GPL(generic_sync_sb_inodes); @@ -588,8 +624,7 @@ restart: /* * writeback and wait upon the filesystem's dirty inodes. The caller will - * do this in two passes - one to write, and one to wait. WB_SYNC_HOLD is - * used to park the written inodes on sb->s_dirty for the wait pass. + * do this in two passes - one to write, and one to wait. * * A finite limit is set on the number of pages which will be written. * To prevent infinite livelock of sys_sync(). @@ -600,30 +635,21 @@ restart: void sync_inodes_sb(struct super_block *sb, int wait) { struct writeback_control wbc = { - .sync_mode = wait ? WB_SYNC_ALL : WB_SYNC_HOLD, + .sync_mode = wait ? WB_SYNC_ALL : WB_SYNC_NONE, .range_start = 0, .range_end = LLONG_MAX, }; - unsigned long nr_dirty = global_page_state(NR_FILE_DIRTY); - unsigned long nr_unstable = global_page_state(NR_UNSTABLE_NFS); - wbc.nr_to_write = nr_dirty + nr_unstable + - (inodes_stat.nr_inodes - inodes_stat.nr_unused) + - nr_dirty + nr_unstable; - wbc.nr_to_write += wbc.nr_to_write / 2; /* Bit more for luck */ - sync_sb_inodes(sb, &wbc); -} + if (!wait) { + unsigned long nr_dirty = global_page_state(NR_FILE_DIRTY); + unsigned long nr_unstable = global_page_state(NR_UNSTABLE_NFS); -/* - * Rather lame livelock avoidance. - */ -static void set_sb_syncing(int val) -{ - struct super_block *sb; - spin_lock(&sb_lock); - list_for_each_entry_reverse(sb, &super_blocks, s_list) - sb->s_syncing = val; - spin_unlock(&sb_lock); + wbc.nr_to_write = nr_dirty + nr_unstable + + (inodes_stat.nr_inodes - inodes_stat.nr_unused); + } else + wbc.nr_to_write = LONG_MAX; /* doesn't actually matter */ + + sync_sb_inodes(sb, &wbc); } /** @@ -652,9 +678,6 @@ static void __sync_inodes(int wait) spin_lock(&sb_lock); restart: list_for_each_entry(sb, &super_blocks, s_list) { - if (sb->s_syncing) - continue; - sb->s_syncing = 1; sb->s_count++; spin_unlock(&sb_lock); down_read(&sb->s_umount); @@ -672,13 +695,10 @@ restart: void sync_inodes(int wait) { - set_sb_syncing(0); __sync_inodes(0); - if (wait) { - set_sb_syncing(0); + if (wait) __sync_inodes(1); - } } /** diff --git a/fs/fuse/control.c b/fs/fuse/control.c index 4f3cab321415..99c99dfb0373 100644 --- a/fs/fuse/control.c +++ b/fs/fuse/control.c @@ -1,6 +1,6 @@ /* FUSE: Filesystem in Userspace - Copyright (C) 2001-2006 Miklos Szeredi <miklos@szeredi.hu> + Copyright (C) 2001-2008 Miklos Szeredi <miklos@szeredi.hu> This program can be distributed under the terms of the GNU GPL. See the file COPYING. @@ -48,11 +48,13 @@ static ssize_t fuse_conn_waiting_read(struct file *file, char __user *buf, size_t size; if (!*ppos) { + long value; struct fuse_conn *fc = fuse_ctl_file_conn_get(file); if (!fc) return 0; - file->private_data=(void *)(long)atomic_read(&fc->num_waiting); + value = atomic_read(&fc->num_waiting); + file->private_data = (void *)value; fuse_conn_put(fc); } size = sprintf(tmp, "%ld\n", (long)file->private_data); diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c index fba571648a8e..e0c7ada08a1f 100644 --- a/fs/fuse/dev.c +++ b/fs/fuse/dev.c @@ -1,6 +1,6 @@ /* FUSE: Filesystem in Userspace - Copyright (C) 2001-2006 Miklos Szeredi <miklos@szeredi.hu> + Copyright (C) 2001-2008 Miklos Szeredi <miklos@szeredi.hu> This program can be distributed under the terms of the GNU GPL. See the file COPYING. @@ -269,7 +269,7 @@ static void flush_bg_queue(struct fuse_conn *fc) * Called with fc->lock, unlocks it */ static void request_end(struct fuse_conn *fc, struct fuse_req *req) - __releases(fc->lock) +__releases(&fc->lock) { void (*end) (struct fuse_conn *, struct fuse_req *) = req->end; req->end = NULL; @@ -293,13 +293,13 @@ static void request_end(struct fuse_conn *fc, struct fuse_req *req) wake_up(&req->waitq); if (end) end(fc, req); - else - fuse_put_request(fc, req); + fuse_put_request(fc, req); } static void wait_answer_interruptible(struct fuse_conn *fc, struct fuse_req *req) - __releases(fc->lock) __acquires(fc->lock) +__releases(&fc->lock) +__acquires(&fc->lock) { if (signal_pending(current)) return; @@ -317,7 +317,8 @@ static void queue_interrupt(struct fuse_conn *fc, struct fuse_req *req) } static void request_wait_answer(struct fuse_conn *fc, struct fuse_req *req) - __releases(fc->lock) __acquires(fc->lock) +__releases(&fc->lock) +__acquires(&fc->lock) { if (!fc->no_interrupt) { /* Any signal may interrupt this */ @@ -380,7 +381,7 @@ static void request_wait_answer(struct fuse_conn *fc, struct fuse_req *req) } } -void request_send(struct fuse_conn *fc, struct fuse_req *req) +void fuse_request_send(struct fuse_conn *fc, struct fuse_req *req) { req->isreply = 1; spin_lock(&fc->lock); @@ -399,8 +400,8 @@ void request_send(struct fuse_conn *fc, struct fuse_req *req) spin_unlock(&fc->lock); } -static void request_send_nowait_locked(struct fuse_conn *fc, - struct fuse_req *req) +static void fuse_request_send_nowait_locked(struct fuse_conn *fc, + struct fuse_req *req) { req->background = 1; fc->num_background++; @@ -414,11 +415,11 @@ static void request_send_nowait_locked(struct fuse_conn *fc, flush_bg_queue(fc); } -static void request_send_nowait(struct fuse_conn *fc, struct fuse_req *req) +static void fuse_request_send_nowait(struct fuse_conn *fc, struct fuse_req *req) { spin_lock(&fc->lock); if (fc->connected) { - request_send_nowait_locked(fc, req); + fuse_request_send_nowait_locked(fc, req); spin_unlock(&fc->lock); } else { req->out.h.error = -ENOTCONN; @@ -426,16 +427,16 @@ static void request_send_nowait(struct fuse_conn *fc, struct fuse_req *req) } } -void request_send_noreply(struct fuse_conn *fc, struct fuse_req *req) +void fuse_request_send_noreply(struct fuse_conn *fc, struct fuse_req *req) { req->isreply = 0; - request_send_nowait(fc, req); + fuse_request_send_nowait(fc, req); } -void request_send_background(struct fuse_conn *fc, struct fuse_req *req) +void fuse_request_send_background(struct fuse_conn *fc, struct fuse_req *req) { req->isreply = 1; - request_send_nowait(fc, req); + fuse_request_send_nowait(fc, req); } /* @@ -443,10 +444,11 @@ void request_send_background(struct fuse_conn *fc, struct fuse_req *req) * * fc->connected must have been checked previously */ -void request_send_background_locked(struct fuse_conn *fc, struct fuse_req *req) +void fuse_request_send_background_locked(struct fuse_conn *fc, + struct fuse_req *req) { req->isreply = 1; - request_send_nowait_locked(fc, req); + fuse_request_send_nowait_locked(fc, req); } /* @@ -539,8 +541,8 @@ static int fuse_copy_fill(struct fuse_copy_state *cs) BUG_ON(!cs->nr_segs); cs->seglen = cs->iov[0].iov_len; cs->addr = (unsigned long) cs->iov[0].iov_base; - cs->iov ++; - cs->nr_segs --; + cs->iov++; + cs->nr_segs--; } down_read(¤t->mm->mmap_sem); err = get_user_pages(current, current->mm, cs->addr, 1, cs->write, 0, @@ -589,9 +591,11 @@ static int fuse_copy_page(struct fuse_copy_state *cs, struct page *page, kunmap_atomic(mapaddr, KM_USER1); } while (count) { - int err; - if (!cs->len && (err = fuse_copy_fill(cs))) - return err; + if (!cs->len) { + int err = fuse_copy_fill(cs); + if (err) + return err; + } if (page) { void *mapaddr = kmap_atomic(page, KM_USER1); void *buf = mapaddr + offset; @@ -631,9 +635,11 @@ static int fuse_copy_pages(struct fuse_copy_state *cs, unsigned nbytes, static int fuse_copy_one(struct fuse_copy_state *cs, void *val, unsigned size) { while (size) { - int err; - if (!cs->len && (err = fuse_copy_fill(cs))) - return err; + if (!cs->len) { + int err = fuse_copy_fill(cs); + if (err) + return err; + } fuse_copy_do(cs, &val, &size); } return 0; @@ -664,6 +670,8 @@ static int request_pending(struct fuse_conn *fc) /* Wait until a request is available on the pending list */ static void request_wait(struct fuse_conn *fc) +__releases(&fc->lock) +__acquires(&fc->lock) { DECLARE_WAITQUEUE(wait, current); @@ -691,7 +699,7 @@ static void request_wait(struct fuse_conn *fc) */ static int fuse_read_interrupt(struct fuse_conn *fc, struct fuse_req *req, const struct iovec *iov, unsigned long nr_segs) - __releases(fc->lock) +__releases(&fc->lock) { struct fuse_copy_state cs; struct fuse_in_header ih; @@ -813,6 +821,34 @@ static ssize_t fuse_dev_read(struct kiocb *iocb, const struct iovec *iov, return err; } +static int fuse_notify_poll(struct fuse_conn *fc, unsigned int size, + struct fuse_copy_state *cs) +{ + struct fuse_notify_poll_wakeup_out outarg; + int err; + + if (size != sizeof(outarg)) + return -EINVAL; + + err = fuse_copy_one(cs, &outarg, sizeof(outarg)); + if (err) + return err; + + return fuse_notify_poll_wakeup(fc, &outarg); +} + +static int fuse_notify(struct fuse_conn *fc, enum fuse_notify_code code, + unsigned int size, struct fuse_copy_state *cs) +{ + switch (code) { + case FUSE_NOTIFY_POLL: + return fuse_notify_poll(fc, size, cs); + + default: + return -EINVAL; + } +} + /* Look up request on processing list by unique ID */ static struct fuse_req *request_find(struct fuse_conn *fc, u64 unique) { @@ -876,9 +912,23 @@ static ssize_t fuse_dev_write(struct kiocb *iocb, const struct iovec *iov, err = fuse_copy_one(&cs, &oh, sizeof(oh)); if (err) goto err_finish; + + err = -EINVAL; + if (oh.len != nbytes) + goto err_finish; + + /* + * Zero oh.unique indicates unsolicited notification message + * and error contains notification code. + */ + if (!oh.unique) { + err = fuse_notify(fc, oh.error, nbytes - sizeof(oh), &cs); + fuse_copy_finish(&cs); + return err ? err : nbytes; + } + err = -EINVAL; - if (!oh.unique || oh.error <= -1000 || oh.error > 0 || - oh.len != nbytes) + if (oh.error <= -1000 || oh.error > 0) goto err_finish; spin_lock(&fc->lock); @@ -966,6 +1016,8 @@ static unsigned fuse_dev_poll(struct file *file, poll_table *wait) * This function releases and reacquires fc->lock */ static void end_requests(struct fuse_conn *fc, struct list_head *head) +__releases(&fc->lock) +__acquires(&fc->lock) { while (!list_empty(head)) { struct fuse_req *req; @@ -988,7 +1040,8 @@ static void end_requests(struct fuse_conn *fc, struct list_head *head) * locked). */ static void end_io_requests(struct fuse_conn *fc) - __releases(fc->lock) __acquires(fc->lock) +__releases(&fc->lock) +__acquires(&fc->lock) { while (!list_empty(&fc->io)) { struct fuse_req *req = @@ -1002,11 +1055,11 @@ static void end_io_requests(struct fuse_conn *fc) wake_up(&req->waitq); if (end) { req->end = NULL; - /* The end function will consume this reference */ __fuse_get_request(req); spin_unlock(&fc->lock); wait_event(req->waitq, !req->locked); end(fc, req); + fuse_put_request(fc, req); spin_lock(&fc->lock); } } diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c index 95bc22bdd060..fdff346e96fd 100644 --- a/fs/fuse/dir.c +++ b/fs/fuse/dir.c @@ -1,6 +1,6 @@ /* FUSE: Filesystem in Userspace - Copyright (C) 2001-2006 Miklos Szeredi <miklos@szeredi.hu> + Copyright (C) 2001-2008 Miklos Szeredi <miklos@szeredi.hu> This program can be distributed under the terms of the GNU GPL. See the file COPYING. @@ -189,7 +189,7 @@ static int fuse_dentry_revalidate(struct dentry *entry, struct nameidata *nd) parent = dget_parent(entry); fuse_lookup_init(fc, req, get_node_id(parent->d_inode), &entry->d_name, &outarg); - request_send(fc, req); + fuse_request_send(fc, req); dput(parent); err = req->out.h.error; fuse_put_request(fc, req); @@ -204,7 +204,7 @@ static int fuse_dentry_revalidate(struct dentry *entry, struct nameidata *nd) return 0; } spin_lock(&fc->lock); - fi->nlookup ++; + fi->nlookup++; spin_unlock(&fc->lock); } fuse_put_request(fc, forget_req); @@ -283,7 +283,7 @@ int fuse_lookup_name(struct super_block *sb, u64 nodeid, struct qstr *name, attr_version = fuse_get_attr_version(fc); fuse_lookup_init(fc, req, nodeid, name, outarg); - request_send(fc, req); + fuse_request_send(fc, req); err = req->out.h.error; fuse_put_request(fc, req); /* Zero nodeid is same as -ENOENT, but with valid timeout */ @@ -369,7 +369,7 @@ static void fuse_sync_release(struct fuse_conn *fc, struct fuse_file *ff, { fuse_release_fill(ff, nodeid, flags, FUSE_RELEASE); ff->reserved_req->force = 1; - request_send(fc, ff->reserved_req); + fuse_request_send(fc, ff->reserved_req); fuse_put_request(fc, ff->reserved_req); kfree(ff); } @@ -408,7 +408,7 @@ static int fuse_create_open(struct inode *dir, struct dentry *entry, int mode, goto out_put_forget_req; err = -ENOMEM; - ff = fuse_file_alloc(); + ff = fuse_file_alloc(fc); if (!ff) goto out_put_request; @@ -432,7 +432,7 @@ static int fuse_create_open(struct inode *dir, struct dentry *entry, int mode, req->out.args[0].value = &outentry; req->out.args[1].size = sizeof(outopen); req->out.args[1].value = &outopen; - request_send(fc, req); + fuse_request_send(fc, req); err = req->out.h.error; if (err) { if (err == -ENOSYS) @@ -502,7 +502,7 @@ static int create_new_entry(struct fuse_conn *fc, struct fuse_req *req, else req->out.args[0].size = sizeof(outarg); req->out.args[0].value = &outarg; - request_send(fc, req); + fuse_request_send(fc, req); err = req->out.h.error; fuse_put_request(fc, req); if (err) @@ -631,15 +631,17 @@ static int fuse_unlink(struct inode *dir, struct dentry *entry) req->in.numargs = 1; req->in.args[0].size = entry->d_name.len + 1; req->in.args[0].value = entry->d_name.name; - request_send(fc, req); + fuse_request_send(fc, req); err = req->out.h.error; fuse_put_request(fc, req); if (!err) { struct inode *inode = entry->d_inode; - /* Set nlink to zero so the inode can be cleared, if - the inode does have more links this will be - discovered at the next lookup/getattr */ + /* + * Set nlink to zero so the inode can be cleared, if the inode + * does have more links this will be discovered at the next + * lookup/getattr. + */ clear_nlink(inode); fuse_invalidate_attr(inode); fuse_invalidate_attr(dir); @@ -662,7 +664,7 @@ static int fuse_rmdir(struct inode *dir, struct dentry *entry) req->in.numargs = 1; req->in.args[0].size = entry->d_name.len + 1; req->in.args[0].value = entry->d_name.name; - request_send(fc, req); + fuse_request_send(fc, req); err = req->out.h.error; fuse_put_request(fc, req); if (!err) { @@ -695,7 +697,7 @@ static int fuse_rename(struct inode *olddir, struct dentry *oldent, req->in.args[1].value = oldent->d_name.name; req->in.args[2].size = newent->d_name.len + 1; req->in.args[2].value = newent->d_name.name; - request_send(fc, req); + fuse_request_send(fc, req); err = req->out.h.error; fuse_put_request(fc, req); if (!err) { @@ -811,7 +813,7 @@ static int fuse_do_getattr(struct inode *inode, struct kstat *stat, else req->out.args[0].size = sizeof(outarg); req->out.args[0].value = &outarg; - request_send(fc, req); + fuse_request_send(fc, req); err = req->out.h.error; fuse_put_request(fc, req); if (!err) { @@ -911,7 +913,7 @@ static int fuse_access(struct inode *inode, int mask) req->in.numargs = 1; req->in.args[0].size = sizeof(inarg); req->in.args[0].value = &inarg; - request_send(fc, req); + fuse_request_send(fc, req); err = req->out.h.error; fuse_put_request(fc, req); if (err == -ENOSYS) { @@ -1033,7 +1035,7 @@ static int fuse_readdir(struct file *file, void *dstbuf, filldir_t filldir) req->num_pages = 1; req->pages[0] = page; fuse_read_fill(req, file, inode, file->f_pos, PAGE_SIZE, FUSE_READDIR); - request_send(fc, req); + fuse_request_send(fc, req); nbytes = req->out.args[0].size; err = req->out.h.error; fuse_put_request(fc, req); @@ -1067,7 +1069,7 @@ static char *read_link(struct dentry *dentry) req->out.numargs = 1; req->out.args[0].size = PAGE_SIZE - 1; req->out.args[0].value = link; - request_send(fc, req); + fuse_request_send(fc, req); if (req->out.h.error) { free_page((unsigned long) link); link = ERR_PTR(req->out.h.error); @@ -1273,7 +1275,7 @@ static int fuse_do_setattr(struct dentry *entry, struct iattr *attr, else req->out.args[0].size = sizeof(outarg); req->out.args[0].value = &outarg; - request_send(fc, req); + fuse_request_send(fc, req); err = req->out.h.error; fuse_put_request(fc, req); if (err) { @@ -1367,7 +1369,7 @@ static int fuse_setxattr(struct dentry *entry, const char *name, req->in.args[1].value = name; req->in.args[2].size = size; req->in.args[2].value = value; - request_send(fc, req); + fuse_request_send(fc, req); err = req->out.h.error; fuse_put_request(fc, req); if (err == -ENOSYS) { @@ -1413,7 +1415,7 @@ static ssize_t fuse_getxattr(struct dentry *entry, const char *name, req->out.args[0].size = sizeof(outarg); req->out.args[0].value = &outarg; } - request_send(fc, req); + fuse_request_send(fc, req); ret = req->out.h.error; if (!ret) ret = size ? req->out.args[0].size : outarg.size; @@ -1463,7 +1465,7 @@ static ssize_t fuse_listxattr(struct dentry *entry, char *list, size_t size) req->out.args[0].size = sizeof(outarg); req->out.args[0].value = &outarg; } - request_send(fc, req); + fuse_request_send(fc, req); ret = req->out.h.error; if (!ret) ret = size ? req->out.args[0].size : outarg.size; @@ -1496,7 +1498,7 @@ static int fuse_removexattr(struct dentry *entry, const char *name) req->in.numargs = 1; req->in.args[0].size = strlen(name) + 1; req->in.args[0].value = name; - request_send(fc, req); + fuse_request_send(fc, req); err = req->out.h.error; fuse_put_request(fc, req); if (err == -ENOSYS) { diff --git a/fs/fuse/file.c b/fs/fuse/file.c index 4c9ee7011265..e8162646a9b5 100644 --- a/fs/fuse/file.c +++ b/fs/fuse/file.c @@ -1,6 +1,6 @@ /* FUSE: Filesystem in Userspace - Copyright (C) 2001-2006 Miklos Szeredi <miklos@szeredi.hu> + Copyright (C) 2001-2008 Miklos Szeredi <miklos@szeredi.hu> This program can be distributed under the terms of the GNU GPL. See the file COPYING. @@ -39,14 +39,14 @@ static int fuse_send_open(struct inode *inode, struct file *file, int isdir, req->out.numargs = 1; req->out.args[0].size = sizeof(*outargp); req->out.args[0].value = outargp; - request_send(fc, req); + fuse_request_send(fc, req); err = req->out.h.error; fuse_put_request(fc, req); return err; } -struct fuse_file *fuse_file_alloc(void) +struct fuse_file *fuse_file_alloc(struct fuse_conn *fc) { struct fuse_file *ff; ff = kmalloc(sizeof(struct fuse_file), GFP_KERNEL); @@ -58,7 +58,12 @@ struct fuse_file *fuse_file_alloc(void) } else { INIT_LIST_HEAD(&ff->write_entry); atomic_set(&ff->count, 0); + spin_lock(&fc->lock); + ff->kh = ++fc->khctr; + spin_unlock(&fc->lock); } + RB_CLEAR_NODE(&ff->polled_node); + init_waitqueue_head(&ff->poll_wait); } return ff; } @@ -79,7 +84,6 @@ static void fuse_release_end(struct fuse_conn *fc, struct fuse_req *req) { dput(req->misc.release.dentry); mntput(req->misc.release.vfsmount); - fuse_put_request(fc, req); } static void fuse_file_put(struct fuse_file *ff) @@ -89,7 +93,7 @@ static void fuse_file_put(struct fuse_file *ff) struct inode *inode = req->misc.release.dentry->d_inode; struct fuse_conn *fc = get_fuse_conn(inode); req->end = fuse_release_end; - request_send_background(fc, req); + fuse_request_send_background(fc, req); kfree(ff); } } @@ -109,6 +113,7 @@ void fuse_finish_open(struct inode *inode, struct file *file, int fuse_open_common(struct inode *inode, struct file *file, int isdir) { + struct fuse_conn *fc = get_fuse_conn(inode); struct fuse_open_out outarg; struct fuse_file *ff; int err; @@ -121,7 +126,7 @@ int fuse_open_common(struct inode *inode, struct file *file, int isdir) if (err) return err; - ff = fuse_file_alloc(); + ff = fuse_file_alloc(fc); if (!ff) return -ENOMEM; @@ -167,7 +172,11 @@ int fuse_release_common(struct inode *inode, struct file *file, int isdir) spin_lock(&fc->lock); list_del(&ff->write_entry); + if (!RB_EMPTY_NODE(&ff->polled_node)) + rb_erase(&ff->polled_node, &fc->polled_files); spin_unlock(&fc->lock); + + wake_up_interruptible_sync(&ff->poll_wait); /* * Normally this will send the RELEASE request, * however if some asynchronous READ or WRITE requests @@ -280,7 +289,7 @@ static int fuse_flush(struct file *file, fl_owner_t id) req->in.args[0].size = sizeof(inarg); req->in.args[0].value = &inarg; req->force = 1; - request_send(fc, req); + fuse_request_send(fc, req); err = req->out.h.error; fuse_put_request(fc, req); if (err == -ENOSYS) { @@ -344,7 +353,7 @@ int fuse_fsync_common(struct file *file, struct dentry *de, int datasync, req->in.numargs = 1; req->in.args[0].size = sizeof(inarg); req->in.args[0].value = &inarg; - request_send(fc, req); + fuse_request_send(fc, req); err = req->out.h.error; fuse_put_request(fc, req); if (err == -ENOSYS) { @@ -396,7 +405,7 @@ static size_t fuse_send_read(struct fuse_req *req, struct file *file, inarg->read_flags |= FUSE_READ_LOCKOWNER; inarg->lock_owner = fuse_lock_owner_id(fc, owner); } - request_send(fc, req); + fuse_request_send(fc, req); return req->out.args[0].size; } @@ -493,7 +502,6 @@ static void fuse_readpages_end(struct fuse_conn *fc, struct fuse_req *req) } if (req->ff) fuse_file_put(req->ff); - fuse_put_request(fc, req); } static void fuse_send_readpages(struct fuse_req *req, struct file *file, @@ -509,10 +517,11 @@ static void fuse_send_readpages(struct fuse_req *req, struct file *file, struct fuse_file *ff = file->private_data; req->ff = fuse_file_get(ff); req->end = fuse_readpages_end; - request_send_background(fc, req); + fuse_request_send_background(fc, req); } else { - request_send(fc, req); + fuse_request_send(fc, req); fuse_readpages_end(fc, req); + fuse_put_request(fc, req); } } @@ -543,7 +552,7 @@ static int fuse_readpages_fill(void *_data, struct page *page) } } req->pages[req->num_pages] = page; - req->num_pages ++; + req->num_pages++; return 0; } @@ -636,7 +645,7 @@ static size_t fuse_send_write(struct fuse_req *req, struct file *file, inarg->write_flags |= FUSE_WRITE_LOCKOWNER; inarg->lock_owner = fuse_lock_owner_id(fc, owner); } - request_send(fc, req); + fuse_request_send(fc, req); return req->misc.write.out.size; } @@ -1042,7 +1051,6 @@ static void fuse_writepage_free(struct fuse_conn *fc, struct fuse_req *req) { __free_page(req->pages[0]); fuse_file_put(req->ff); - fuse_put_request(fc, req); } static void fuse_writepage_finish(struct fuse_conn *fc, struct fuse_req *req) @@ -1060,6 +1068,8 @@ static void fuse_writepage_finish(struct fuse_conn *fc, struct fuse_req *req) /* Called under fc->lock, may release and reacquire it */ static void fuse_send_writepage(struct fuse_conn *fc, struct fuse_req *req) +__releases(&fc->lock) +__acquires(&fc->lock) { struct fuse_inode *fi = get_fuse_inode(req->inode); loff_t size = i_size_read(req->inode); @@ -1079,13 +1089,14 @@ static void fuse_send_writepage(struct fuse_conn *fc, struct fuse_req *req) req->in.args[1].size = inarg->size; fi->writectr++; - request_send_background_locked(fc, req); + fuse_request_send_background_locked(fc, req); return; out_free: fuse_writepage_finish(fc, req); spin_unlock(&fc->lock); fuse_writepage_free(fc, req); + fuse_put_request(fc, req); spin_lock(&fc->lock); } @@ -1096,6 +1107,8 @@ static void fuse_send_writepage(struct fuse_conn *fc, struct fuse_req *req) * Called with fc->lock */ void fuse_flush_writepages(struct inode *inode) +__releases(&fc->lock) +__acquires(&fc->lock) { struct fuse_conn *fc = get_fuse_conn(inode); struct fuse_inode *fi = get_fuse_inode(inode); @@ -1325,7 +1338,7 @@ static int fuse_getlk(struct file *file, struct file_lock *fl) req->out.numargs = 1; req->out.args[0].size = sizeof(outarg); req->out.args[0].value = &outarg; - request_send(fc, req); + fuse_request_send(fc, req); err = req->out.h.error; fuse_put_request(fc, req); if (!err) @@ -1357,7 +1370,7 @@ static int fuse_setlk(struct file *file, struct file_lock *fl, int flock) return PTR_ERR(req); fuse_lk_fill(req, file, fl, opcode, pid, flock); - request_send(fc, req); + fuse_request_send(fc, req); err = req->out.h.error; /* locking is restartable */ if (err == -EINTR) @@ -1433,7 +1446,7 @@ static sector_t fuse_bmap(struct address_space *mapping, sector_t block) req->out.numargs = 1; req->out.args[0].size = sizeof(outarg); req->out.args[0].value = &outarg; - request_send(fc, req); + fuse_request_send(fc, req); err = req->out.h.error; fuse_put_request(fc, req); if (err == -ENOSYS) @@ -1470,6 +1483,406 @@ static loff_t fuse_file_llseek(struct file *file, loff_t offset, int origin) return retval; } +static int fuse_ioctl_copy_user(struct page **pages, struct iovec *iov, + unsigned int nr_segs, size_t bytes, bool to_user) +{ + struct iov_iter ii; + int page_idx = 0; + + if (!bytes) + return 0; + + iov_iter_init(&ii, iov, nr_segs, bytes, 0); + + while (iov_iter_count(&ii)) { + struct page *page = pages[page_idx++]; + size_t todo = min_t(size_t, PAGE_SIZE, iov_iter_count(&ii)); + void *kaddr, *map; + + kaddr = map = kmap(page); + + while (todo) { + char __user *uaddr = ii.iov->iov_base + ii.iov_offset; + size_t iov_len = ii.iov->iov_len - ii.iov_offset; + size_t copy = min(todo, iov_len); + size_t left; + + if (!to_user) + left = copy_from_user(kaddr, uaddr, copy); + else + left = copy_to_user(uaddr, kaddr, copy); + + if (unlikely(left)) + return -EFAULT; + + iov_iter_advance(&ii, copy); + todo -= copy; + kaddr += copy; + } + + kunmap(map); + } + + return 0; +} + +/* + * For ioctls, there is no generic way to determine how much memory + * needs to be read and/or written. Furthermore, ioctls are allowed + * to dereference the passed pointer, so the parameter requires deep + * copying but FUSE has no idea whatsoever about what to copy in or + * out. + * + * This is solved by allowing FUSE server to retry ioctl with + * necessary in/out iovecs. Let's assume the ioctl implementation + * needs to read in the following structure. + * + * struct a { + * char *buf; + * size_t buflen; + * } + * + * On the first callout to FUSE server, inarg->in_size and + * inarg->out_size will be NULL; then, the server completes the ioctl + * with FUSE_IOCTL_RETRY set in out->flags, out->in_iovs set to 1 and + * the actual iov array to + * + * { { .iov_base = inarg.arg, .iov_len = sizeof(struct a) } } + * + * which tells FUSE to copy in the requested area and retry the ioctl. + * On the second round, the server has access to the structure and + * from that it can tell what to look for next, so on the invocation, + * it sets FUSE_IOCTL_RETRY, out->in_iovs to 2 and iov array to + * + * { { .iov_base = inarg.arg, .iov_len = sizeof(struct a) }, + * { .iov_base = a.buf, .iov_len = a.buflen } } + * + * FUSE will copy both struct a and the pointed buffer from the + * process doing the ioctl and retry ioctl with both struct a and the + * buffer. + * + * This time, FUSE server has everything it needs and completes ioctl + * without FUSE_IOCTL_RETRY which finishes the ioctl call. + * + * Copying data out works the same way. + * + * Note that if FUSE_IOCTL_UNRESTRICTED is clear, the kernel + * automatically initializes in and out iovs by decoding @cmd with + * _IOC_* macros and the server is not allowed to request RETRY. This + * limits ioctl data transfers to well-formed ioctls and is the forced + * behavior for all FUSE servers. + */ +static long fuse_file_do_ioctl(struct file *file, unsigned int cmd, + unsigned long arg, unsigned int flags) +{ + struct inode *inode = file->f_dentry->d_inode; + struct fuse_file *ff = file->private_data; + struct fuse_conn *fc = get_fuse_conn(inode); + struct fuse_ioctl_in inarg = { + .fh = ff->fh, + .cmd = cmd, + .arg = arg, + .flags = flags + }; + struct fuse_ioctl_out outarg; + struct fuse_req *req = NULL; + struct page **pages = NULL; + struct page *iov_page = NULL; + struct iovec *in_iov = NULL, *out_iov = NULL; + unsigned int in_iovs = 0, out_iovs = 0, num_pages = 0, max_pages; + size_t in_size, out_size, transferred; + int err; + + /* assume all the iovs returned by client always fits in a page */ + BUILD_BUG_ON(sizeof(struct iovec) * FUSE_IOCTL_MAX_IOV > PAGE_SIZE); + + if (!fuse_allow_task(fc, current)) + return -EACCES; + + err = -EIO; + if (is_bad_inode(inode)) + goto out; + + err = -ENOMEM; + pages = kzalloc(sizeof(pages[0]) * FUSE_MAX_PAGES_PER_REQ, GFP_KERNEL); + iov_page = alloc_page(GFP_KERNEL); + if (!pages || !iov_page) + goto out; + + /* + * If restricted, initialize IO parameters as encoded in @cmd. + * RETRY from server is not allowed. + */ + if (!(flags & FUSE_IOCTL_UNRESTRICTED)) { + struct iovec *iov = page_address(iov_page); + + iov->iov_base = (void __user *)arg; + iov->iov_len = _IOC_SIZE(cmd); + + if (_IOC_DIR(cmd) & _IOC_WRITE) { + in_iov = iov; + in_iovs = 1; + } + + if (_IOC_DIR(cmd) & _IOC_READ) { + out_iov = iov; + out_iovs = 1; + } + } + + retry: + inarg.in_size = in_size = iov_length(in_iov, in_iovs); + inarg.out_size = out_size = iov_length(out_iov, out_iovs); + + /* + * Out data can be used either for actual out data or iovs, + * make sure there always is at least one page. + */ + out_size = max_t(size_t, out_size, PAGE_SIZE); + max_pages = DIV_ROUND_UP(max(in_size, out_size), PAGE_SIZE); + + /* make sure there are enough buffer pages and init request with them */ + err = -ENOMEM; + if (max_pages > FUSE_MAX_PAGES_PER_REQ) + goto out; + while (num_pages < max_pages) { + pages[num_pages] = alloc_page(GFP_KERNEL | __GFP_HIGHMEM); + if (!pages[num_pages]) + goto out; + num_pages++; + } + + req = fuse_get_req(fc); + if (IS_ERR(req)) { + err = PTR_ERR(req); + req = NULL; + goto out; + } + memcpy(req->pages, pages, sizeof(req->pages[0]) * num_pages); + req->num_pages = num_pages; + + /* okay, let's send it to the client */ + req->in.h.opcode = FUSE_IOCTL; + req->in.h.nodeid = get_node_id(inode); + req->in.numargs = 1; + req->in.args[0].size = sizeof(inarg); + req->in.args[0].value = &inarg; + if (in_size) { + req->in.numargs++; + req->in.args[1].size = in_size; + req->in.argpages = 1; + + err = fuse_ioctl_copy_user(pages, in_iov, in_iovs, in_size, + false); + if (err) + goto out; + } + + req->out.numargs = 2; + req->out.args[0].size = sizeof(outarg); + req->out.args[0].value = &outarg; + req->out.args[1].size = out_size; + req->out.argpages = 1; + req->out.argvar = 1; + + fuse_request_send(fc, req); + err = req->out.h.error; + transferred = req->out.args[1].size; + fuse_put_request(fc, req); + req = NULL; + if (err) + goto out; + + /* did it ask for retry? */ + if (outarg.flags & FUSE_IOCTL_RETRY) { + char *vaddr; + + /* no retry if in restricted mode */ + err = -EIO; + if (!(flags & FUSE_IOCTL_UNRESTRICTED)) + goto out; + + in_iovs = outarg.in_iovs; + out_iovs = outarg.out_iovs; + + /* + * Make sure things are in boundary, separate checks + * are to protect against overflow. + */ + err = -ENOMEM; + if (in_iovs > FUSE_IOCTL_MAX_IOV || + out_iovs > FUSE_IOCTL_MAX_IOV || + in_iovs + out_iovs > FUSE_IOCTL_MAX_IOV) + goto out; + + err = -EIO; + if ((in_iovs + out_iovs) * sizeof(struct iovec) != transferred) + goto out; + + /* okay, copy in iovs and retry */ + vaddr = kmap_atomic(pages[0], KM_USER0); + memcpy(page_address(iov_page), vaddr, transferred); + kunmap_atomic(vaddr, KM_USER0); + + in_iov = page_address(iov_page); + out_iov = in_iov + in_iovs; + + goto retry; + } + + err = -EIO; + if (transferred > inarg.out_size) + goto out; + + err = fuse_ioctl_copy_user(pages, out_iov, out_iovs, transferred, true); + out: + if (req) + fuse_put_request(fc, req); + if (iov_page) + __free_page(iov_page); + while (num_pages) + __free_page(pages[--num_pages]); + kfree(pages); + + return err ? err : outarg.result; +} + +static long fuse_file_ioctl(struct file *file, unsigned int cmd, + unsigned long arg) +{ + return fuse_file_do_ioctl(file, cmd, arg, 0); +} + +static long fuse_file_compat_ioctl(struct file *file, unsigned int cmd, + unsigned long arg) +{ + return fuse_file_do_ioctl(file, cmd, arg, FUSE_IOCTL_COMPAT); +} + +/* + * All files which have been polled are linked to RB tree + * fuse_conn->polled_files which is indexed by kh. Walk the tree and + * find the matching one. + */ +static struct rb_node **fuse_find_polled_node(struct fuse_conn *fc, u64 kh, + struct rb_node **parent_out) +{ + struct rb_node **link = &fc->polled_files.rb_node; + struct rb_node *last = NULL; + + while (*link) { + struct fuse_file *ff; + + last = *link; + ff = rb_entry(last, struct fuse_file, polled_node); + + if (kh < ff->kh) + link = &last->rb_left; + else if (kh > ff->kh) + link = &last->rb_right; + else + return link; + } + + if (parent_out) + *parent_out = last; + return link; +} + +/* + * The file is about to be polled. Make sure it's on the polled_files + * RB tree. Note that files once added to the polled_files tree are + * not removed before the file is released. This is because a file + * polled once is likely to be polled again. + */ +static void fuse_register_polled_file(struct fuse_conn *fc, + struct fuse_file *ff) +{ + spin_lock(&fc->lock); + if (RB_EMPTY_NODE(&ff->polled_node)) { + struct rb_node **link, *parent; + + link = fuse_find_polled_node(fc, ff->kh, &parent); + BUG_ON(*link); + rb_link_node(&ff->polled_node, parent, link); + rb_insert_color(&ff->polled_node, &fc->polled_files); + } + spin_unlock(&fc->lock); +} + +static unsigned fuse_file_poll(struct file *file, poll_table *wait) +{ + struct inode *inode = file->f_dentry->d_inode; + struct fuse_file *ff = file->private_data; + struct fuse_conn *fc = get_fuse_conn(inode); + struct fuse_poll_in inarg = { .fh = ff->fh, .kh = ff->kh }; + struct fuse_poll_out outarg; + struct fuse_req *req; + int err; + + if (fc->no_poll) + return DEFAULT_POLLMASK; + + poll_wait(file, &ff->poll_wait, wait); + + /* + * Ask for notification iff there's someone waiting for it. + * The client may ignore the flag and always notify. + */ + if (waitqueue_active(&ff->poll_wait)) { + inarg.flags |= FUSE_POLL_SCHEDULE_NOTIFY; + fuse_register_polled_file(fc, ff); + } + + req = fuse_get_req(fc); + if (IS_ERR(req)) + return PTR_ERR(req); + + req->in.h.opcode = FUSE_POLL; + req->in.h.nodeid = get_node_id(inode); + req->in.numargs = 1; + req->in.args[0].size = sizeof(inarg); + req->in.args[0].value = &inarg; + req->out.numargs = 1; + req->out.args[0].size = sizeof(outarg); + req->out.args[0].value = &outarg; + fuse_request_send(fc, req); + err = req->out.h.error; + fuse_put_request(fc, req); + + if (!err) + return outarg.revents; + if (err == -ENOSYS) { + fc->no_poll = 1; + return DEFAULT_POLLMASK; + } + return POLLERR; +} + +/* + * This is called from fuse_handle_notify() on FUSE_NOTIFY_POLL and + * wakes up the poll waiters. + */ +int fuse_notify_poll_wakeup(struct fuse_conn *fc, + struct fuse_notify_poll_wakeup_out *outarg) +{ + u64 kh = outarg->kh; + struct rb_node **link; + + spin_lock(&fc->lock); + + link = fuse_find_polled_node(fc, kh, NULL); + if (*link) { + struct fuse_file *ff; + + ff = rb_entry(*link, struct fuse_file, polled_node); + wake_up_interruptible_sync(&ff->poll_wait); + } + + spin_unlock(&fc->lock); + return 0; +} + static const struct file_operations fuse_file_operations = { .llseek = fuse_file_llseek, .read = do_sync_read, @@ -1484,6 +1897,9 @@ static const struct file_operations fuse_file_operations = { .lock = fuse_file_lock, .flock = fuse_file_flock, .splice_read = generic_file_splice_read, + .unlocked_ioctl = fuse_file_ioctl, + .compat_ioctl = fuse_file_compat_ioctl, + .poll = fuse_file_poll, }; static const struct file_operations fuse_direct_io_file_operations = { @@ -1496,6 +1912,9 @@ static const struct file_operations fuse_direct_io_file_operations = { .fsync = fuse_fsync, .lock = fuse_file_lock, .flock = fuse_file_flock, + .unlocked_ioctl = fuse_file_ioctl, + .compat_ioctl = fuse_file_compat_ioctl, + .poll = fuse_file_poll, /* no mmap and splice_read */ }; diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h index 35accfdd747f..5e64b815a5a1 100644 --- a/fs/fuse/fuse_i.h +++ b/fs/fuse/fuse_i.h @@ -1,6 +1,6 @@ /* FUSE: Filesystem in Userspace - Copyright (C) 2001-2006 Miklos Szeredi <miklos@szeredi.hu> + Copyright (C) 2001-2008 Miklos Szeredi <miklos@szeredi.hu> This program can be distributed under the terms of the GNU GPL. See the file COPYING. @@ -19,6 +19,8 @@ #include <linux/backing-dev.h> #include <linux/mutex.h> #include <linux/rwsem.h> +#include <linux/rbtree.h> +#include <linux/poll.h> /** Max number of pages that can be used in a single read request */ #define FUSE_MAX_PAGES_PER_REQ 32 @@ -100,6 +102,9 @@ struct fuse_file { /** Request reserved for flush and release */ struct fuse_req *reserved_req; + /** Kernel file handle guaranteed to be unique */ + u64 kh; + /** File handle used by userspace */ u64 fh; @@ -108,6 +113,12 @@ struct fuse_file { /** Entry on inode's write_files list */ struct list_head write_entry; + + /** RB node to be linked on fuse_conn->polled_files */ + struct rb_node polled_node; + + /** Wait queue head for poll */ + wait_queue_head_t poll_wait; }; /** One input argument of a request */ @@ -322,6 +333,12 @@ struct fuse_conn { /** The list of requests under I/O */ struct list_head io; + /** The next unique kernel file handle */ + u64 khctr; + + /** rbtree of fuse_files waiting for poll events indexed by ph */ + struct rb_root polled_files; + /** Number of requests currently in the background */ unsigned num_background; @@ -355,19 +372,19 @@ struct fuse_conn { /** Connection failed (version mismatch). Cannot race with setting other bitfields since it is only set once in INIT reply, before any other request, and never cleared */ - unsigned conn_error : 1; + unsigned conn_error:1; /** Connection successful. Only set in INIT */ - unsigned conn_init : 1; + unsigned conn_init:1; /** Do readpages asynchronously? Only set in INIT */ - unsigned async_read : 1; + unsigned async_read:1; /** Do not send separate SETATTR request before open(O_TRUNC) */ - unsigned atomic_o_trunc : 1; + unsigned atomic_o_trunc:1; /** Filesystem supports NFS exporting. Only set in INIT */ - unsigned export_support : 1; + unsigned export_support:1; /* * The following bitfields are only for optimization purposes @@ -375,43 +392,46 @@ struct fuse_conn { */ /** Is fsync not implemented by fs? */ - unsigned no_fsync : 1; + unsigned no_fsync:1; /** Is fsyncdir not implemented by fs? */ - unsigned no_fsyncdir : 1; + unsigned no_fsyncdir:1; /** Is flush not implemented by fs? */ - unsigned no_flush : 1; + unsigned no_flush:1; /** Is setxattr not implemented by fs? */ - unsigned no_setxattr : 1; + unsigned no_setxattr:1; /** Is getxattr not implemented by fs? */ - unsigned no_getxattr : 1; + unsigned no_getxattr:1; /** Is listxattr not implemented by fs? */ - unsigned no_listxattr : 1; + unsigned no_listxattr:1; /** Is removexattr not implemented by fs? */ - unsigned no_removexattr : 1; + unsigned no_removexattr:1; /** Are file locking primitives not implemented by fs? */ - unsigned no_lock : 1; + unsigned no_lock:1; /** Is access not implemented by fs? */ - unsigned no_access : 1; + unsigned no_access:1; /** Is create not implemented by fs? */ - unsigned no_create : 1; + unsigned no_create:1; /** Is interrupt not implemented by fs? */ - unsigned no_interrupt : 1; + unsigned no_interrupt:1; /** Is bmap not implemented by fs? */ - unsigned no_bmap : 1; + unsigned no_bmap:1; + + /** Is poll not implemented by fs? */ + unsigned no_poll:1; /** Do multi-page cached writes */ - unsigned big_writes : 1; + unsigned big_writes:1; /** The number of requests waiting for completion */ atomic_t num_waiting; @@ -445,6 +465,9 @@ struct fuse_conn { /** Version counter for attribute changes */ u64 attr_version; + + /** Called on final put */ + void (*release)(struct fuse_conn *); }; static inline struct fuse_conn *get_fuse_conn_super(struct super_block *sb) @@ -499,7 +522,7 @@ void fuse_read_fill(struct fuse_req *req, struct file *file, */ int fuse_open_common(struct inode *inode, struct file *file, int isdir); -struct fuse_file *fuse_file_alloc(void); +struct fuse_file *fuse_file_alloc(struct fuse_conn *fc); void fuse_file_free(struct fuse_file *ff); void fuse_finish_open(struct inode *inode, struct file *file, struct fuse_file *ff, struct fuse_open_out *outarg); @@ -519,6 +542,12 @@ int fuse_fsync_common(struct file *file, struct dentry *de, int datasync, int isdir); /** + * Notify poll wakeup + */ +int fuse_notify_poll_wakeup(struct fuse_conn *fc, + struct fuse_notify_poll_wakeup_out *outarg); + +/** * Initialize file operations on a regular file */ void fuse_init_file_inode(struct inode *inode); @@ -593,19 +622,20 @@ void fuse_put_request(struct fuse_conn *fc, struct fuse_req *req); /** * Send a request (synchronous) */ -void request_send(struct fuse_conn *fc, struct fuse_req *req); +void fuse_request_send(struct fuse_conn *fc, struct fuse_req *req); /** * Send a request with no reply */ -void request_send_noreply(struct fuse_conn *fc, struct fuse_req *req); +void fuse_request_send_noreply(struct fuse_conn *fc, struct fuse_req *req); /** * Send a request in the background */ -void request_send_background(struct fuse_conn *fc, struct fuse_req *req); +void fuse_request_send_background(struct fuse_conn *fc, struct fuse_req *req); -void request_send_background_locked(struct fuse_conn *fc, struct fuse_req *req); +void fuse_request_send_background_locked(struct fuse_conn *fc, + struct fuse_req *req); /* Abort all requests */ void fuse_abort_conn(struct fuse_conn *fc); @@ -623,6 +653,11 @@ void fuse_invalidate_entry_cache(struct dentry *entry); struct fuse_conn *fuse_conn_get(struct fuse_conn *fc); /** + * Initialize fuse_conn + */ +int fuse_conn_init(struct fuse_conn *fc, struct super_block *sb); + +/** * Release reference to fuse_conn */ void fuse_conn_put(struct fuse_conn *fc); diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index 2e99f34b4435..47c96fdca1ac 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -1,6 +1,6 @@ /* FUSE: Filesystem in Userspace - Copyright (C) 2001-2006 Miklos Szeredi <miklos@szeredi.hu> + Copyright (C) 2001-2008 Miklos Szeredi <miklos@szeredi.hu> This program can be distributed under the terms of the GNU GPL. See the file COPYING. @@ -37,10 +37,10 @@ struct fuse_mount_data { unsigned rootmode; unsigned user_id; unsigned group_id; - unsigned fd_present : 1; - unsigned rootmode_present : 1; - unsigned user_id_present : 1; - unsigned group_id_present : 1; + unsigned fd_present:1; + unsigned rootmode_present:1; + unsigned user_id_present:1; + unsigned group_id_present:1; unsigned flags; unsigned max_read; unsigned blksize; @@ -94,7 +94,7 @@ void fuse_send_forget(struct fuse_conn *fc, struct fuse_req *req, req->in.numargs = 1; req->in.args[0].size = sizeof(struct fuse_forget_in); req->in.args[0].value = inarg; - request_send_noreply(fc, req); + fuse_request_send_noreply(fc, req); } static void fuse_clear_inode(struct inode *inode) @@ -250,7 +250,7 @@ struct inode *fuse_iget(struct super_block *sb, u64 nodeid, fi = get_fuse_inode(inode); spin_lock(&fc->lock); - fi->nlookup ++; + fi->nlookup++; spin_unlock(&fc->lock); fuse_change_attributes(inode, attr, attr_valid, attr_version); @@ -269,7 +269,7 @@ static void fuse_send_destroy(struct fuse_conn *fc) fc->destroy_req = NULL; req->in.h.opcode = FUSE_DESTROY; req->force = 1; - request_send(fc, req); + fuse_request_send(fc, req); fuse_put_request(fc, req); } } @@ -334,7 +334,7 @@ static int fuse_statfs(struct dentry *dentry, struct kstatfs *buf) req->out.args[0].size = fc->minor < 4 ? FUSE_COMPAT_STATFS_SIZE : sizeof(outarg); req->out.args[0].value = &outarg; - request_send(fc, req); + fuse_request_send(fc, req); err = req->out.h.error; if (!err) convert_fuse_statfs(buf, &outarg.st); @@ -462,68 +462,69 @@ static int fuse_show_options(struct seq_file *m, struct vfsmount *mnt) return 0; } -static struct fuse_conn *new_conn(struct super_block *sb) +int fuse_conn_init(struct fuse_conn *fc, struct super_block *sb) { - struct fuse_conn *fc; int err; - fc = kzalloc(sizeof(*fc), GFP_KERNEL); - if (fc) { - spin_lock_init(&fc->lock); - mutex_init(&fc->inst_mutex); - atomic_set(&fc->count, 1); - init_waitqueue_head(&fc->waitq); - init_waitqueue_head(&fc->blocked_waitq); - init_waitqueue_head(&fc->reserved_req_waitq); - INIT_LIST_HEAD(&fc->pending); - INIT_LIST_HEAD(&fc->processing); - INIT_LIST_HEAD(&fc->io); - INIT_LIST_HEAD(&fc->interrupts); - INIT_LIST_HEAD(&fc->bg_queue); - atomic_set(&fc->num_waiting, 0); - fc->bdi.ra_pages = (VM_MAX_READAHEAD * 1024) / PAGE_CACHE_SIZE; - fc->bdi.unplug_io_fn = default_unplug_io_fn; - /* fuse does it's own writeback accounting */ - fc->bdi.capabilities = BDI_CAP_NO_ACCT_WB; - fc->dev = sb->s_dev; - err = bdi_init(&fc->bdi); - if (err) - goto error_kfree; - if (sb->s_bdev) { - err = bdi_register(&fc->bdi, NULL, "%u:%u-fuseblk", - MAJOR(fc->dev), MINOR(fc->dev)); - } else { - err = bdi_register_dev(&fc->bdi, fc->dev); - } - if (err) - goto error_bdi_destroy; - /* - * For a single fuse filesystem use max 1% of dirty + - * writeback threshold. - * - * This gives about 1M of write buffer for memory maps on a - * machine with 1G and 10% dirty_ratio, which should be more - * than enough. - * - * Privileged users can raise it by writing to - * - * /sys/class/bdi/<bdi>/max_ratio - */ - bdi_set_max_ratio(&fc->bdi, 1); - fc->reqctr = 0; - fc->blocked = 1; - fc->attr_version = 1; - get_random_bytes(&fc->scramble_key, sizeof(fc->scramble_key)); + memset(fc, 0, sizeof(*fc)); + spin_lock_init(&fc->lock); + mutex_init(&fc->inst_mutex); + atomic_set(&fc->count, 1); + init_waitqueue_head(&fc->waitq); + init_waitqueue_head(&fc->blocked_waitq); + init_waitqueue_head(&fc->reserved_req_waitq); + INIT_LIST_HEAD(&fc->pending); + INIT_LIST_HEAD(&fc->processing); + INIT_LIST_HEAD(&fc->io); + INIT_LIST_HEAD(&fc->interrupts); + INIT_LIST_HEAD(&fc->bg_queue); + INIT_LIST_HEAD(&fc->entry); + atomic_set(&fc->num_waiting, 0); + fc->bdi.ra_pages = (VM_MAX_READAHEAD * 1024) / PAGE_CACHE_SIZE; + fc->bdi.unplug_io_fn = default_unplug_io_fn; + /* fuse does it's own writeback accounting */ + fc->bdi.capabilities = BDI_CAP_NO_ACCT_WB; + fc->khctr = 0; + fc->polled_files = RB_ROOT; + fc->dev = sb->s_dev; + err = bdi_init(&fc->bdi); + if (err) + goto error_mutex_destroy; + if (sb->s_bdev) { + err = bdi_register(&fc->bdi, NULL, "%u:%u-fuseblk", + MAJOR(fc->dev), MINOR(fc->dev)); + } else { + err = bdi_register_dev(&fc->bdi, fc->dev); } - return fc; + if (err) + goto error_bdi_destroy; + /* + * For a single fuse filesystem use max 1% of dirty + + * writeback threshold. + * + * This gives about 1M of write buffer for memory maps on a + * machine with 1G and 10% dirty_ratio, which should be more + * than enough. + * + * Privileged users can raise it by writing to + * + * /sys/class/bdi/<bdi>/max_ratio + */ + bdi_set_max_ratio(&fc->bdi, 1); + fc->reqctr = 0; + fc->blocked = 1; + fc->attr_version = 1; + get_random_bytes(&fc->scramble_key, sizeof(fc->scramble_key)); -error_bdi_destroy: + return 0; + + error_bdi_destroy: bdi_destroy(&fc->bdi); -error_kfree: + error_mutex_destroy: mutex_destroy(&fc->inst_mutex); - kfree(fc); - return NULL; + return err; } +EXPORT_SYMBOL_GPL(fuse_conn_init); void fuse_conn_put(struct fuse_conn *fc) { @@ -532,7 +533,7 @@ void fuse_conn_put(struct fuse_conn *fc) fuse_request_free(fc->destroy_req); mutex_destroy(&fc->inst_mutex); bdi_destroy(&fc->bdi); - kfree(fc); + fc->release(fc); } } @@ -542,7 +543,7 @@ struct fuse_conn *fuse_conn_get(struct fuse_conn *fc) return fc; } -static struct inode *get_root_inode(struct super_block *sb, unsigned mode) +static struct inode *fuse_get_root_inode(struct super_block *sb, unsigned mode) { struct fuse_attr attr; memset(&attr, 0, sizeof(attr)); @@ -553,8 +554,7 @@ static struct inode *get_root_inode(struct super_block *sb, unsigned mode) return fuse_iget(sb, 1, 0, &attr, 0, 0); } -struct fuse_inode_handle -{ +struct fuse_inode_handle { u64 nodeid; u32 generation; }; @@ -761,7 +761,6 @@ static void process_init_reply(struct fuse_conn *fc, struct fuse_req *req) fc->max_write = max_t(unsigned, 4096, fc->max_write); fc->conn_init = 1; } - fuse_put_request(fc, req); fc->blocked = 0; wake_up_all(&fc->blocked_waitq); } @@ -787,7 +786,12 @@ static void fuse_send_init(struct fuse_conn *fc, struct fuse_req *req) req->out.args[0].size = sizeof(struct fuse_init_out); req->out.args[0].value = &req->misc.init_out; req->end = process_init_reply; - request_send_background(fc, req); + fuse_request_send_background(fc, req); +} + +static void fuse_free_conn(struct fuse_conn *fc) +{ + kfree(fc); } static int fuse_fill_super(struct super_block *sb, void *data, int silent) @@ -828,10 +832,17 @@ static int fuse_fill_super(struct super_block *sb, void *data, int silent) if (file->f_op != &fuse_dev_operations) return -EINVAL; - fc = new_conn(sb); + fc = kmalloc(sizeof(*fc), GFP_KERNEL); if (!fc) return -ENOMEM; + err = fuse_conn_init(fc, sb); + if (err) { + kfree(fc); + return err; + } + + fc->release = fuse_free_conn; fc->flags = d.flags; fc->user_id = d.user_id; fc->group_id = d.group_id; @@ -841,7 +852,7 @@ static int fuse_fill_super(struct super_block *sb, void *data, int silent) sb->s_fs_info = fc; err = -ENOMEM; - root = get_root_inode(sb, d.rootmode); + root = fuse_get_root_inode(sb, d.rootmode); if (!root) goto err; @@ -952,7 +963,7 @@ static inline void unregister_fuseblk(void) static void fuse_inode_init_once(void *foo) { - struct inode * inode = foo; + struct inode *inode = foo; inode_init_once(inode); } @@ -1031,7 +1042,7 @@ static int __init fuse_init(void) { int res; - printk("fuse init (API version %i.%i)\n", + printk(KERN_INFO "fuse init (API version %i.%i)\n", FUSE_KERNEL_VERSION, FUSE_KERNEL_MINOR_VERSION); INIT_LIST_HEAD(&fuse_conn_list); diff --git a/fs/gfs2/Kconfig b/fs/gfs2/Kconfig index ab2f57e3fb87..e563a6449811 100644 --- a/fs/gfs2/Kconfig +++ b/fs/gfs2/Kconfig @@ -1,6 +1,6 @@ config GFS2_FS tristate "GFS2 file system support" - depends on EXPERIMENTAL && (64BIT || (LSF && LBD)) + depends on EXPERIMENTAL && (64BIT || LBD) select FS_POSIX_ACL select CRC32 help diff --git a/fs/gfs2/ops_address.c b/fs/gfs2/ops_address.c index 6e4ea36c6605..4ddab67867eb 100644 --- a/fs/gfs2/ops_address.c +++ b/fs/gfs2/ops_address.c @@ -675,6 +675,7 @@ static int gfs2_write_begin(struct file *file, struct address_space *mapping, goto out_trans_fail; error = -ENOMEM; + flags |= AOP_FLAG_NOFS; page = grab_cache_page_write_begin(mapping, index, flags); *pagep = page; if (unlikely(!page)) diff --git a/fs/gfs2/ops_file.c b/fs/gfs2/ops_file.c index 289c5f54ba53..93fe41b67f97 100644 --- a/fs/gfs2/ops_file.c +++ b/fs/gfs2/ops_file.c @@ -342,7 +342,7 @@ static int gfs2_page_mkwrite(struct vm_area_struct *vma, struct page *page) struct gfs2_inode *ip = GFS2_I(inode); struct gfs2_sbd *sdp = GFS2_SB(inode); unsigned long last_index; - u64 pos = page->index << (PAGE_CACHE_SIZE - inode->i_blkbits); + u64 pos = page->index << PAGE_CACHE_SHIFT; unsigned int data_blocks, ind_blocks, rblocks; int alloc_required = 0; struct gfs2_holder gh; diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 0ab0c6f5f438..6903d37af037 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -252,6 +252,7 @@ static ssize_t hugetlbfs_read(struct file *filp, char __user *buf, for (;;) { struct page *page; unsigned long nr, ret; + int ra; /* nr is the maximum number of bytes to copy from this page */ nr = huge_page_size(h); @@ -274,16 +275,19 @@ static ssize_t hugetlbfs_read(struct file *filp, char __user *buf, */ ret = len < nr ? len : nr; if (clear_user(buf, ret)) - ret = -EFAULT; + ra = -EFAULT; + else + ra = 0; } else { /* * We have the page, copy it to user space buffer. */ - ret = hugetlbfs_read_actor(page, offset, buf, len, nr); + ra = hugetlbfs_read_actor(page, offset, buf, len, nr); + ret = ra; } - if (ret < 0) { + if (ra < 0) { if (retval == 0) - retval = ret; + retval = ra; if (page) page_cache_release(page); goto out; diff --git a/fs/inode.c b/fs/inode.c index bd48e5e6d3e8..0013ac1af8e7 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -22,6 +22,7 @@ #include <linux/bootmem.h> #include <linux/inotify.h> #include <linux/mount.h> +#include <linux/async.h> /* * This is needed for the following functions: @@ -110,8 +111,8 @@ static void wake_up_inode(struct inode *inode) /** * inode_init_always - perform inode structure intialisation - * @sb - superblock inode belongs to. - * @inode - inode to initialise + * @sb: superblock inode belongs to + * @inode: inode to initialise * * These are initializations that need to be done on every inode * allocation as the fields are not initialised by slab allocation. @@ -166,7 +167,7 @@ struct inode *inode_init_always(struct super_block *sb, struct inode *inode) mapping->a_ops = &empty_aops; mapping->host = inode; mapping->flags = 0; - mapping_set_gfp_mask(mapping, GFP_HIGHUSER_PAGECACHE); + mapping_set_gfp_mask(mapping, GFP_HIGHUSER_MOVABLE); mapping->assoc_mapping = NULL; mapping->backing_dev_info = &default_backing_dev_info; mapping->writeback_index = 0; @@ -576,8 +577,8 @@ __inode_add_to_lists(struct super_block *sb, struct hlist_head *head, /** * inode_add_to_lists - add a new inode to relevant lists - * @sb - superblock inode belongs to. - * @inode - inode to mark in use + * @sb: superblock inode belongs to + * @inode: inode to mark in use * * When an inode is allocated it needs to be accounted for, added to the in use * list, the owning superblock and the inode hash. This needs to be done under @@ -601,7 +602,7 @@ EXPORT_SYMBOL_GPL(inode_add_to_lists); * @sb: superblock * * Allocates a new inode for given superblock. The default gfp_mask - * for allocations related to inode->i_mapping is GFP_HIGHUSER_PAGECACHE. + * for allocations related to inode->i_mapping is GFP_HIGHUSER_MOVABLE. * If HIGHMEM pages are unsuitable or it is known that pages allocated * for the page cache are not reclaimable or migratable, * mapping_set_gfp_mask() must be called with suitable flags on the @@ -1138,16 +1139,11 @@ EXPORT_SYMBOL(remove_inode_hash); * I_FREEING is set so that no-one will take a new reference to the inode while * it is being deleted. */ -void generic_delete_inode(struct inode *inode) +static void generic_delete_inode_async(void *data, async_cookie_t cookie) { + struct inode *inode = data; const struct super_operations *op = inode->i_sb->s_op; - list_del_init(&inode->i_list); - list_del_init(&inode->i_sb_list); - inode->i_state |= I_FREEING; - inodes_stat.nr_inodes--; - spin_unlock(&inode_lock); - security_inode_delete(inode); if (op->delete_inode) { @@ -1171,6 +1167,16 @@ void generic_delete_inode(struct inode *inode) destroy_inode(inode); } +void generic_delete_inode(struct inode *inode) +{ + list_del_init(&inode->i_list); + list_del_init(&inode->i_sb_list); + inode->i_state |= I_FREEING; + inodes_stat.nr_inodes--; + spin_unlock(&inode_lock); + async_schedule_special(generic_delete_inode_async, inode, &inode->i_sb->s_async_list); +} + EXPORT_SYMBOL(generic_delete_inode); static void generic_forget_inode(struct inode *inode) diff --git a/fs/lockd/clntproc.c b/fs/lockd/clntproc.c index 31668b690e03..dd7957064a8c 100644 --- a/fs/lockd/clntproc.c +++ b/fs/lockd/clntproc.c @@ -16,7 +16,6 @@ #include <linux/sunrpc/clnt.h> #include <linux/sunrpc/svc.h> #include <linux/lockd/lockd.h> -#include <linux/lockd/sm_inter.h> #define NLMDBG_FACILITY NLMDBG_CLIENT #define NLMCLNT_GRACE_WAIT (5*HZ) @@ -518,11 +517,9 @@ nlmclnt_lock(struct nlm_rqst *req, struct file_lock *fl) unsigned char fl_type; int status = -ENOLCK; - if (nsm_monitor(host) < 0) { - printk(KERN_NOTICE "lockd: failed to monitor %s\n", - host->h_name); + if (nsm_monitor(host) < 0) goto out; - } + fl->fl_flags |= FL_ACCESS; status = do_vfs_lock(fl); fl->fl_flags = fl_flags; diff --git a/fs/lockd/host.c b/fs/lockd/host.c index abdebf76b820..99d737bd4325 100644 --- a/fs/lockd/host.c +++ b/fs/lockd/host.c @@ -15,7 +15,6 @@ #include <linux/sunrpc/clnt.h> #include <linux/sunrpc/svc.h> #include <linux/lockd/lockd.h> -#include <linux/lockd/sm_inter.h> #include <linux/mutex.h> #include <net/ipv6.h> @@ -32,11 +31,6 @@ static int nrhosts; static DEFINE_MUTEX(nlm_host_mutex); static void nlm_gc_hosts(void); -static struct nsm_handle *nsm_find(const struct sockaddr *sap, - const size_t salen, - const char *hostname, - const size_t hostname_len, - const int create); struct nlm_lookup_host_info { const int server; /* search for server|client */ @@ -105,32 +99,6 @@ static void nlm_clear_port(struct sockaddr *sap) } } -static void nlm_display_address(const struct sockaddr *sap, - char *buf, const size_t len) -{ - const struct sockaddr_in *sin = (struct sockaddr_in *)sap; - const struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)sap; - - switch (sap->sa_family) { - case AF_UNSPEC: - snprintf(buf, len, "unspecified"); - break; - case AF_INET: - snprintf(buf, len, "%pI4", &sin->sin_addr.s_addr); - break; - case AF_INET6: - if (ipv6_addr_v4mapped(&sin6->sin6_addr)) - snprintf(buf, len, "%pI4", - &sin6->sin6_addr.s6_addr32[3]); - else - snprintf(buf, len, "%pI6", &sin6->sin6_addr); - break; - default: - snprintf(buf, len, "unsupported address family"); - break; - } -} - /* * Common host lookup routine for server & client */ @@ -190,8 +158,8 @@ static struct nlm_host *nlm_lookup_host(struct nlm_lookup_host_info *ni) atomic_inc(&nsm->sm_count); else { host = NULL; - nsm = nsm_find(ni->sap, ni->salen, - ni->hostname, ni->hostname_len, 1); + nsm = nsm_get_handle(ni->sap, ni->salen, + ni->hostname, ni->hostname_len); if (!nsm) { dprintk("lockd: nlm_lookup_host failed; " "no nsm handle\n"); @@ -206,6 +174,7 @@ static struct nlm_host *nlm_lookup_host(struct nlm_lookup_host_info *ni) goto out; } host->h_name = nsm->sm_name; + host->h_addrbuf = nsm->sm_addrbuf; memcpy(nlm_addr(host), ni->sap, ni->salen); host->h_addrlen = ni->salen; nlm_clear_port(nlm_addr(host)); @@ -232,11 +201,6 @@ static struct nlm_host *nlm_lookup_host(struct nlm_lookup_host_info *ni) nrhosts++; - nlm_display_address((struct sockaddr *)&host->h_addr, - host->h_addrbuf, sizeof(host->h_addrbuf)); - nlm_display_address((struct sockaddr *)&host->h_srcaddr, - host->h_srcaddrbuf, sizeof(host->h_srcaddrbuf)); - dprintk("lockd: nlm_lookup_host created host %s\n", host->h_name); @@ -256,10 +220,8 @@ nlm_destroy_host(struct nlm_host *host) BUG_ON(!list_empty(&host->h_lockowners)); BUG_ON(atomic_read(&host->h_count)); - /* - * Release NSM handle and unmonitor host. - */ nsm_unmonitor(host); + nsm_release(host->h_nsmhandle); clnt = host->h_rpcclnt; if (clnt != NULL) @@ -378,8 +340,8 @@ nlm_bind_host(struct nlm_host *host) { struct rpc_clnt *clnt; - dprintk("lockd: nlm_bind_host %s (%s), my addr=%s\n", - host->h_name, host->h_addrbuf, host->h_srcaddrbuf); + dprintk("lockd: nlm_bind_host %s (%s)\n", + host->h_name, host->h_addrbuf); /* Lock host handle */ mutex_lock(&host->h_mutex); @@ -481,35 +443,23 @@ void nlm_release_host(struct nlm_host *host) } } -/* - * We were notified that the host indicated by address &sin - * has rebooted. - * Release all resources held by that peer. +/** + * nlm_host_rebooted - Release all resources held by rebooted host + * @info: pointer to decoded results of NLM_SM_NOTIFY call + * + * We were notified that the specified host has rebooted. Release + * all resources held by that peer. */ -void nlm_host_rebooted(const struct sockaddr_in *sin, - const char *hostname, - unsigned int hostname_len, - u32 new_state) +void nlm_host_rebooted(const struct nlm_reboot *info) { struct hlist_head *chain; struct hlist_node *pos; struct nsm_handle *nsm; struct nlm_host *host; - nsm = nsm_find((struct sockaddr *)sin, sizeof(*sin), - hostname, hostname_len, 0); - if (nsm == NULL) { - dprintk("lockd: never saw rebooted peer '%.*s' before\n", - hostname_len, hostname); + nsm = nsm_reboot_lookup(info); + if (unlikely(nsm == NULL)) return; - } - - dprintk("lockd: nlm_host_rebooted(%.*s, %s)\n", - hostname_len, hostname, nsm->sm_addrbuf); - - /* When reclaiming locks on this peer, make sure that - * we set up a new notification */ - nsm->sm_monitored = 0; /* Mark all hosts tied to this NSM state as having rebooted. * We run the loop repeatedly, because we drop the host table @@ -520,8 +470,8 @@ again: mutex_lock(&nlm_host_mutex); for (chain = nlm_hosts; chain < nlm_hosts + NLM_HOST_NRHASH; ++chain) { hlist_for_each_entry(host, pos, chain, h_hash) { if (host->h_nsmhandle == nsm - && host->h_nsmstate != new_state) { - host->h_nsmstate = new_state; + && host->h_nsmstate != info->state) { + host->h_nsmstate = info->state; host->h_state++; nlm_get_host(host); @@ -629,89 +579,3 @@ nlm_gc_hosts(void) next_gc = jiffies + NLM_HOST_COLLECT; } - - -/* - * Manage NSM handles - */ -static LIST_HEAD(nsm_handles); -static DEFINE_SPINLOCK(nsm_lock); - -static struct nsm_handle *nsm_find(const struct sockaddr *sap, - const size_t salen, - const char *hostname, - const size_t hostname_len, - const int create) -{ - struct nsm_handle *nsm = NULL; - struct nsm_handle *pos; - - if (!sap) - return NULL; - - if (hostname && memchr(hostname, '/', hostname_len) != NULL) { - if (printk_ratelimit()) { - printk(KERN_WARNING "Invalid hostname \"%.*s\" " - "in NFS lock request\n", - (int)hostname_len, hostname); - } - return NULL; - } - -retry: - spin_lock(&nsm_lock); - list_for_each_entry(pos, &nsm_handles, sm_link) { - - if (hostname && nsm_use_hostnames) { - if (strlen(pos->sm_name) != hostname_len - || memcmp(pos->sm_name, hostname, hostname_len)) - continue; - } else if (!nlm_cmp_addr(nsm_addr(pos), sap)) - continue; - atomic_inc(&pos->sm_count); - kfree(nsm); - nsm = pos; - goto found; - } - if (nsm) { - list_add(&nsm->sm_link, &nsm_handles); - goto found; - } - spin_unlock(&nsm_lock); - - if (!create) - return NULL; - - nsm = kzalloc(sizeof(*nsm) + hostname_len + 1, GFP_KERNEL); - if (nsm == NULL) - return NULL; - - memcpy(nsm_addr(nsm), sap, salen); - nsm->sm_addrlen = salen; - nsm->sm_name = (char *) (nsm + 1); - memcpy(nsm->sm_name, hostname, hostname_len); - nsm->sm_name[hostname_len] = '\0'; - nlm_display_address((struct sockaddr *)&nsm->sm_addr, - nsm->sm_addrbuf, sizeof(nsm->sm_addrbuf)); - atomic_set(&nsm->sm_count, 1); - goto retry; - -found: - spin_unlock(&nsm_lock); - return nsm; -} - -/* - * Release an NSM handle - */ -void -nsm_release(struct nsm_handle *nsm) -{ - if (!nsm) - return; - if (atomic_dec_and_lock(&nsm->sm_count, &nsm_lock)) { - list_del(&nsm->sm_link); - spin_unlock(&nsm_lock); - kfree(nsm); - } -} diff --git a/fs/lockd/mon.c b/fs/lockd/mon.c index ffd3461f75ef..5e2c4d5ac827 100644 --- a/fs/lockd/mon.c +++ b/fs/lockd/mon.c @@ -9,35 +9,123 @@ #include <linux/types.h> #include <linux/utsname.h> #include <linux/kernel.h> +#include <linux/ktime.h> + #include <linux/sunrpc/clnt.h> #include <linux/sunrpc/xprtsock.h> #include <linux/sunrpc/svc.h> #include <linux/lockd/lockd.h> -#include <linux/lockd/sm_inter.h> - #define NLMDBG_FACILITY NLMDBG_MONITOR +#define NSM_PROGRAM 100024 +#define NSM_VERSION 1 + +enum { + NSMPROC_NULL, + NSMPROC_STAT, + NSMPROC_MON, + NSMPROC_UNMON, + NSMPROC_UNMON_ALL, + NSMPROC_SIMU_CRASH, + NSMPROC_NOTIFY, +}; + +struct nsm_args { + struct nsm_private *priv; + u32 prog; /* RPC callback info */ + u32 vers; + u32 proc; -#define XDR_ADDRBUF_LEN (20) + char *mon_name; +}; -static struct rpc_clnt * nsm_create(void); +struct nsm_res { + u32 status; + u32 state; +}; static struct rpc_program nsm_program; +static LIST_HEAD(nsm_handles); +static DEFINE_SPINLOCK(nsm_lock); /* * Local NSM state */ -int nsm_local_state; +int __read_mostly nsm_local_state; +int __read_mostly nsm_use_hostnames; -/* - * Common procedure for SM_MON/SM_UNMON calls - */ -static int -nsm_mon_unmon(struct nsm_handle *nsm, u32 proc, struct nsm_res *res) +static inline struct sockaddr *nsm_addr(const struct nsm_handle *nsm) +{ + return (struct sockaddr *)&nsm->sm_addr; +} + +static void nsm_display_ipv4_address(const struct sockaddr *sap, char *buf, + const size_t len) +{ + const struct sockaddr_in *sin = (struct sockaddr_in *)sap; + snprintf(buf, len, "%pI4", &sin->sin_addr.s_addr); +} + +static void nsm_display_ipv6_address(const struct sockaddr *sap, char *buf, + const size_t len) +{ + const struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *)sap; + + if (ipv6_addr_v4mapped(&sin6->sin6_addr)) + snprintf(buf, len, "%pI4", &sin6->sin6_addr.s6_addr32[3]); + else if (sin6->sin6_scope_id != 0) + snprintf(buf, len, "%pI6%%%u", &sin6->sin6_addr, + sin6->sin6_scope_id); + else + snprintf(buf, len, "%pI6", &sin6->sin6_addr); +} + +static void nsm_display_address(const struct sockaddr *sap, + char *buf, const size_t len) +{ + switch (sap->sa_family) { + case AF_INET: + nsm_display_ipv4_address(sap, buf, len); + break; + case AF_INET6: + nsm_display_ipv6_address(sap, buf, len); + break; + default: + snprintf(buf, len, "unsupported address family"); + break; + } +} + +static struct rpc_clnt *nsm_create(void) +{ + struct sockaddr_in sin = { + .sin_family = AF_INET, + .sin_addr.s_addr = htonl(INADDR_LOOPBACK), + }; + struct rpc_create_args args = { + .protocol = XPRT_TRANSPORT_UDP, + .address = (struct sockaddr *)&sin, + .addrsize = sizeof(sin), + .servername = "rpc.statd", + .program = &nsm_program, + .version = NSM_VERSION, + .authflavor = RPC_AUTH_NULL, + }; + + return rpc_create(&args); +} + +static int nsm_mon_unmon(struct nsm_handle *nsm, u32 proc, struct nsm_res *res) { struct rpc_clnt *clnt; int status; - struct nsm_args args; + struct nsm_args args = { + .priv = &nsm->sm_priv, + .prog = NLM_PROGRAM, + .vers = 3, + .proc = NLMPROC_NSM_NOTIFY, + .mon_name = nsm->sm_mon_name, + }; struct rpc_message msg = { .rpc_argp = &args, .rpc_resp = res, @@ -46,22 +134,18 @@ nsm_mon_unmon(struct nsm_handle *nsm, u32 proc, struct nsm_res *res) clnt = nsm_create(); if (IS_ERR(clnt)) { status = PTR_ERR(clnt); + dprintk("lockd: failed to create NSM upcall transport, " + "status=%d\n", status); goto out; } - memset(&args, 0, sizeof(args)); - args.mon_name = nsm->sm_name; - args.addr = nsm_addr_in(nsm)->sin_addr.s_addr; - args.prog = NLM_PROGRAM; - args.vers = 3; - args.proc = NLMPROC_NSM_NOTIFY; memset(res, 0, sizeof(*res)); msg.rpc_proc = &clnt->cl_procinfo[proc]; status = rpc_call_sync(clnt, &msg, 0); if (status < 0) - printk(KERN_DEBUG "nsm_mon_unmon: rpc failed, status=%d\n", - status); + dprintk("lockd: NSM upcall RPC failed, status=%d\n", + status); else status = 0; rpc_shutdown_client(clnt); @@ -69,82 +153,272 @@ nsm_mon_unmon(struct nsm_handle *nsm, u32 proc, struct nsm_res *res) return status; } -/* - * Set up monitoring of a remote host +/** + * nsm_monitor - Notify a peer in case we reboot + * @host: pointer to nlm_host of peer to notify + * + * If this peer is not already monitored, this function sends an + * upcall to the local rpc.statd to record the name/address of + * the peer to notify in case we reboot. + * + * Returns zero if the peer is monitored by the local rpc.statd; + * otherwise a negative errno value is returned. */ -int -nsm_monitor(struct nlm_host *host) +int nsm_monitor(const struct nlm_host *host) { struct nsm_handle *nsm = host->h_nsmhandle; struct nsm_res res; int status; - dprintk("lockd: nsm_monitor(%s)\n", host->h_name); - BUG_ON(nsm == NULL); + dprintk("lockd: nsm_monitor(%s)\n", nsm->sm_name); if (nsm->sm_monitored) return 0; - status = nsm_mon_unmon(nsm, SM_MON, &res); + /* + * Choose whether to record the caller_name or IP address of + * this peer in the local rpc.statd's database. + */ + nsm->sm_mon_name = nsm_use_hostnames ? nsm->sm_name : nsm->sm_addrbuf; - if (status < 0 || res.status != 0) - printk(KERN_NOTICE "lockd: cannot monitor %s\n", host->h_name); + status = nsm_mon_unmon(nsm, NSMPROC_MON, &res); + if (res.status != 0) + status = -EIO; + if (status < 0) + printk(KERN_NOTICE "lockd: cannot monitor %s\n", nsm->sm_name); else nsm->sm_monitored = 1; return status; } -/* - * Cease to monitor remote host +/** + * nsm_unmonitor - Unregister peer notification + * @host: pointer to nlm_host of peer to stop monitoring + * + * If this peer is monitored, this function sends an upcall to + * tell the local rpc.statd not to send this peer a notification + * when we reboot. */ -int -nsm_unmonitor(struct nlm_host *host) +void nsm_unmonitor(const struct nlm_host *host) { struct nsm_handle *nsm = host->h_nsmhandle; struct nsm_res res; - int status = 0; - - if (nsm == NULL) - return 0; - host->h_nsmhandle = NULL; + int status; if (atomic_read(&nsm->sm_count) == 1 && nsm->sm_monitored && !nsm->sm_sticky) { - dprintk("lockd: nsm_unmonitor(%s)\n", host->h_name); + dprintk("lockd: nsm_unmonitor(%s)\n", nsm->sm_name); - status = nsm_mon_unmon(nsm, SM_UNMON, &res); + status = nsm_mon_unmon(nsm, NSMPROC_UNMON, &res); + if (res.status != 0) + status = -EIO; if (status < 0) printk(KERN_NOTICE "lockd: cannot unmonitor %s\n", - host->h_name); + nsm->sm_name); else nsm->sm_monitored = 0; } - nsm_release(nsm); - return status; +} + +static struct nsm_handle *nsm_lookup_hostname(const char *hostname, + const size_t len) +{ + struct nsm_handle *nsm; + + list_for_each_entry(nsm, &nsm_handles, sm_link) + if (strlen(nsm->sm_name) == len && + memcmp(nsm->sm_name, hostname, len) == 0) + return nsm; + return NULL; +} + +static struct nsm_handle *nsm_lookup_addr(const struct sockaddr *sap) +{ + struct nsm_handle *nsm; + + list_for_each_entry(nsm, &nsm_handles, sm_link) + if (nlm_cmp_addr(nsm_addr(nsm), sap)) + return nsm; + return NULL; +} + +static struct nsm_handle *nsm_lookup_priv(const struct nsm_private *priv) +{ + struct nsm_handle *nsm; + + list_for_each_entry(nsm, &nsm_handles, sm_link) + if (memcmp(nsm->sm_priv.data, priv->data, + sizeof(priv->data)) == 0) + return nsm; + return NULL; } /* - * Create NSM client for the local host + * Construct a unique cookie to match this nsm_handle to this monitored + * host. It is passed to the local rpc.statd via NSMPROC_MON, and + * returned via NLMPROC_SM_NOTIFY, in the "priv" field of these + * requests. + * + * The NSM protocol requires that these cookies be unique while the + * system is running. We prefer a stronger requirement of making them + * unique across reboots. If user space bugs cause a stale cookie to + * be sent to the kernel, it could cause the wrong host to lose its + * lock state if cookies were not unique across reboots. + * + * The cookies are exposed only to local user space via loopback. They + * do not appear on the physical network. If we want greater security + * for some reason, nsm_init_private() could perform a one-way hash to + * obscure the contents of the cookie. */ -static struct rpc_clnt * -nsm_create(void) +static void nsm_init_private(struct nsm_handle *nsm) { - struct sockaddr_in sin = { - .sin_family = AF_INET, - .sin_addr.s_addr = htonl(INADDR_LOOPBACK), - .sin_port = 0, - }; - struct rpc_create_args args = { - .protocol = XPRT_TRANSPORT_UDP, - .address = (struct sockaddr *)&sin, - .addrsize = sizeof(sin), - .servername = "localhost", - .program = &nsm_program, - .version = SM_VERSION, - .authflavor = RPC_AUTH_NULL, - }; + u64 *p = (u64 *)&nsm->sm_priv.data; + struct timespec ts; - return rpc_create(&args); + ktime_get_ts(&ts); + *p++ = timespec_to_ns(&ts); + *p = (unsigned long)nsm; +} + +static struct nsm_handle *nsm_create_handle(const struct sockaddr *sap, + const size_t salen, + const char *hostname, + const size_t hostname_len) +{ + struct nsm_handle *new; + + new = kzalloc(sizeof(*new) + hostname_len + 1, GFP_KERNEL); + if (unlikely(new == NULL)) + return NULL; + + atomic_set(&new->sm_count, 1); + new->sm_name = (char *)(new + 1); + memcpy(nsm_addr(new), sap, salen); + new->sm_addrlen = salen; + nsm_init_private(new); + nsm_display_address((const struct sockaddr *)&new->sm_addr, + new->sm_addrbuf, sizeof(new->sm_addrbuf)); + memcpy(new->sm_name, hostname, hostname_len); + new->sm_name[hostname_len] = '\0'; + + return new; +} + +/** + * nsm_get_handle - Find or create a cached nsm_handle + * @sap: pointer to socket address of handle to find + * @salen: length of socket address + * @hostname: pointer to C string containing hostname to find + * @hostname_len: length of C string + * + * Behavior is modulated by the global nsm_use_hostnames variable. + * + * Returns a cached nsm_handle after bumping its ref count, or + * returns a fresh nsm_handle if a handle that matches @sap and/or + * @hostname cannot be found in the handle cache. Returns NULL if + * an error occurs. + */ +struct nsm_handle *nsm_get_handle(const struct sockaddr *sap, + const size_t salen, const char *hostname, + const size_t hostname_len) +{ + struct nsm_handle *cached, *new = NULL; + + if (hostname && memchr(hostname, '/', hostname_len) != NULL) { + if (printk_ratelimit()) { + printk(KERN_WARNING "Invalid hostname \"%.*s\" " + "in NFS lock request\n", + (int)hostname_len, hostname); + } + return NULL; + } + +retry: + spin_lock(&nsm_lock); + + if (nsm_use_hostnames && hostname != NULL) + cached = nsm_lookup_hostname(hostname, hostname_len); + else + cached = nsm_lookup_addr(sap); + + if (cached != NULL) { + atomic_inc(&cached->sm_count); + spin_unlock(&nsm_lock); + kfree(new); + dprintk("lockd: found nsm_handle for %s (%s), " + "cnt %d\n", cached->sm_name, + cached->sm_addrbuf, + atomic_read(&cached->sm_count)); + return cached; + } + + if (new != NULL) { + list_add(&new->sm_link, &nsm_handles); + spin_unlock(&nsm_lock); + dprintk("lockd: created nsm_handle for %s (%s)\n", + new->sm_name, new->sm_addrbuf); + return new; + } + + spin_unlock(&nsm_lock); + + new = nsm_create_handle(sap, salen, hostname, hostname_len); + if (unlikely(new == NULL)) + return NULL; + goto retry; +} + +/** + * nsm_reboot_lookup - match NLMPROC_SM_NOTIFY arguments to an nsm_handle + * @info: pointer to NLMPROC_SM_NOTIFY arguments + * + * Returns a matching nsm_handle if found in the nsm cache; the returned + * nsm_handle's reference count is bumped and sm_monitored is cleared. + * Otherwise returns NULL if some error occurred. + */ +struct nsm_handle *nsm_reboot_lookup(const struct nlm_reboot *info) +{ + struct nsm_handle *cached; + + spin_lock(&nsm_lock); + + cached = nsm_lookup_priv(&info->priv); + if (unlikely(cached == NULL)) { + spin_unlock(&nsm_lock); + dprintk("lockd: never saw rebooted peer '%.*s' before\n", + info->len, info->mon); + return cached; + } + + atomic_inc(&cached->sm_count); + spin_unlock(&nsm_lock); + + /* + * During subsequent lock activity, force a fresh + * notification to be set up for this host. + */ + cached->sm_monitored = 0; + + dprintk("lockd: host %s (%s) rebooted, cnt %d\n", + cached->sm_name, cached->sm_addrbuf, + atomic_read(&cached->sm_count)); + return cached; +} + +/** + * nsm_release - Release an NSM handle + * @nsm: pointer to handle to be released + * + */ +void nsm_release(struct nsm_handle *nsm) +{ + if (atomic_dec_and_lock(&nsm->sm_count, &nsm_lock)) { + list_del(&nsm->sm_link); + spin_unlock(&nsm_lock); + dprintk("lockd: destroyed nsm_handle for %s (%s)\n", + nsm->sm_name, nsm->sm_addrbuf); + kfree(nsm); + } } /* @@ -154,127 +428,132 @@ nsm_create(void) * Status Monitor wire protocol. */ -static __be32 *xdr_encode_nsm_string(__be32 *p, char *string) +static int encode_nsm_string(struct xdr_stream *xdr, const char *string) { - size_t len = strlen(string); - - if (len > SM_MAXSTRLEN) - len = SM_MAXSTRLEN; - return xdr_encode_opaque(p, string, len); + const u32 len = strlen(string); + __be32 *p; + + if (unlikely(len > SM_MAXSTRLEN)) + return -EIO; + p = xdr_reserve_space(xdr, sizeof(u32) + len); + if (unlikely(p == NULL)) + return -EIO; + xdr_encode_opaque(p, string, len); + return 0; } /* * "mon_name" specifies the host to be monitored. - * - * Linux uses a text version of the IP address of the remote - * host as the host identifier (the "mon_name" argument). - * - * Linux statd always looks up the canonical hostname first for - * whatever remote hostname it receives, so this works alright. */ -static __be32 *xdr_encode_mon_name(__be32 *p, struct nsm_args *argp) +static int encode_mon_name(struct xdr_stream *xdr, const struct nsm_args *argp) { - char buffer[XDR_ADDRBUF_LEN + 1]; - char *name = argp->mon_name; - - if (!nsm_use_hostnames) { - snprintf(buffer, XDR_ADDRBUF_LEN, - "%pI4", &argp->addr); - name = buffer; - } - - return xdr_encode_nsm_string(p, name); + return encode_nsm_string(xdr, argp->mon_name); } /* * The "my_id" argument specifies the hostname and RPC procedure * to be called when the status manager receives notification - * (via the SM_NOTIFY call) that the state of host "mon_name" + * (via the NLMPROC_SM_NOTIFY call) that the state of host "mon_name" * has changed. */ -static __be32 *xdr_encode_my_id(__be32 *p, struct nsm_args *argp) +static int encode_my_id(struct xdr_stream *xdr, const struct nsm_args *argp) { - p = xdr_encode_nsm_string(p, utsname()->nodename); - if (!p) - return ERR_PTR(-EIO); - + int status; + __be32 *p; + + status = encode_nsm_string(xdr, utsname()->nodename); + if (unlikely(status != 0)) + return status; + p = xdr_reserve_space(xdr, 3 * sizeof(u32)); + if (unlikely(p == NULL)) + return -EIO; *p++ = htonl(argp->prog); *p++ = htonl(argp->vers); *p++ = htonl(argp->proc); - - return p; + return 0; } /* * The "mon_id" argument specifies the non-private arguments - * of an SM_MON or SM_UNMON call. + * of an NSMPROC_MON or NSMPROC_UNMON call. */ -static __be32 *xdr_encode_mon_id(__be32 *p, struct nsm_args *argp) +static int encode_mon_id(struct xdr_stream *xdr, const struct nsm_args *argp) { - p = xdr_encode_mon_name(p, argp); - if (!p) - return ERR_PTR(-EIO); + int status; - return xdr_encode_my_id(p, argp); + status = encode_mon_name(xdr, argp); + if (unlikely(status != 0)) + return status; + return encode_my_id(xdr, argp); } /* * The "priv" argument may contain private information required - * by the SM_MON call. This information will be supplied in the - * SM_NOTIFY call. - * - * Linux provides the raw IP address of the monitored host, - * left in network byte order. + * by the NSMPROC_MON call. This information will be supplied in the + * NLMPROC_SM_NOTIFY call. */ -static __be32 *xdr_encode_priv(__be32 *p, struct nsm_args *argp) +static int encode_priv(struct xdr_stream *xdr, const struct nsm_args *argp) { - *p++ = argp->addr; - *p++ = 0; - *p++ = 0; - *p++ = 0; + __be32 *p; - return p; + p = xdr_reserve_space(xdr, SM_PRIV_SIZE); + if (unlikely(p == NULL)) + return -EIO; + xdr_encode_opaque_fixed(p, argp->priv->data, SM_PRIV_SIZE); + return 0; } -static int -xdr_encode_mon(struct rpc_rqst *rqstp, __be32 *p, struct nsm_args *argp) +static int xdr_enc_mon(struct rpc_rqst *req, __be32 *p, + const struct nsm_args *argp) { - p = xdr_encode_mon_id(p, argp); - if (IS_ERR(p)) - return PTR_ERR(p); - - p = xdr_encode_priv(p, argp); - if (IS_ERR(p)) - return PTR_ERR(p); - - rqstp->rq_slen = xdr_adjust_iovec(rqstp->rq_svec, p); - return 0; + struct xdr_stream xdr; + int status; + + xdr_init_encode(&xdr, &req->rq_snd_buf, p); + status = encode_mon_id(&xdr, argp); + if (unlikely(status)) + return status; + return encode_priv(&xdr, argp); } -static int -xdr_encode_unmon(struct rpc_rqst *rqstp, __be32 *p, struct nsm_args *argp) +static int xdr_enc_unmon(struct rpc_rqst *req, __be32 *p, + const struct nsm_args *argp) { - p = xdr_encode_mon_id(p, argp); - if (IS_ERR(p)) - return PTR_ERR(p); - rqstp->rq_slen = xdr_adjust_iovec(rqstp->rq_svec, p); - return 0; + struct xdr_stream xdr; + + xdr_init_encode(&xdr, &req->rq_snd_buf, p); + return encode_mon_id(&xdr, argp); } -static int -xdr_decode_stat_res(struct rpc_rqst *rqstp, __be32 *p, struct nsm_res *resp) +static int xdr_dec_stat_res(struct rpc_rqst *rqstp, __be32 *p, + struct nsm_res *resp) { + struct xdr_stream xdr; + + xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p); + p = xdr_inline_decode(&xdr, 2 * sizeof(u32)); + if (unlikely(p == NULL)) + return -EIO; resp->status = ntohl(*p++); - resp->state = ntohl(*p++); - dprintk("nsm: xdr_decode_stat_res status %d state %d\n", + resp->state = ntohl(*p); + + dprintk("lockd: xdr_dec_stat_res status %d state %d\n", resp->status, resp->state); return 0; } -static int -xdr_decode_stat(struct rpc_rqst *rqstp, __be32 *p, struct nsm_res *resp) +static int xdr_dec_stat(struct rpc_rqst *rqstp, __be32 *p, + struct nsm_res *resp) { - resp->state = ntohl(*p++); + struct xdr_stream xdr; + + xdr_init_decode(&xdr, &rqstp->rq_rcv_buf, p); + p = xdr_inline_decode(&xdr, sizeof(u32)); + if (unlikely(p == NULL)) + return -EIO; + resp->state = ntohl(*p); + + dprintk("lockd: xdr_dec_stat state %d\n", resp->state); return 0; } @@ -288,22 +567,22 @@ xdr_decode_stat(struct rpc_rqst *rqstp, __be32 *p, struct nsm_res *resp) #define SM_unmonres_sz 1 static struct rpc_procinfo nsm_procedures[] = { -[SM_MON] = { - .p_proc = SM_MON, - .p_encode = (kxdrproc_t) xdr_encode_mon, - .p_decode = (kxdrproc_t) xdr_decode_stat_res, +[NSMPROC_MON] = { + .p_proc = NSMPROC_MON, + .p_encode = (kxdrproc_t)xdr_enc_mon, + .p_decode = (kxdrproc_t)xdr_dec_stat_res, .p_arglen = SM_mon_sz, .p_replen = SM_monres_sz, - .p_statidx = SM_MON, + .p_statidx = NSMPROC_MON, .p_name = "MONITOR", }, -[SM_UNMON] = { - .p_proc = SM_UNMON, - .p_encode = (kxdrproc_t) xdr_encode_unmon, - .p_decode = (kxdrproc_t) xdr_decode_stat, +[NSMPROC_UNMON] = { + .p_proc = NSMPROC_UNMON, + .p_encode = (kxdrproc_t)xdr_enc_unmon, + .p_decode = (kxdrproc_t)xdr_dec_stat, .p_arglen = SM_mon_id_sz, .p_replen = SM_unmonres_sz, - .p_statidx = SM_UNMON, + .p_statidx = NSMPROC_UNMON, .p_name = "UNMONITOR", }, }; @@ -322,7 +601,7 @@ static struct rpc_stat nsm_stats; static struct rpc_program nsm_program = { .name = "statd", - .number = SM_PROGRAM, + .number = NSM_PROGRAM, .nrvers = ARRAY_SIZE(nsm_version), .version = nsm_version, .stats = &nsm_stats diff --git a/fs/lockd/svc.c b/fs/lockd/svc.c index 252d80163d02..64f1c31b5853 100644 --- a/fs/lockd/svc.c +++ b/fs/lockd/svc.c @@ -35,7 +35,6 @@ #include <linux/sunrpc/svcsock.h> #include <net/ip.h> #include <linux/lockd/lockd.h> -#include <linux/lockd/sm_inter.h> #include <linux/nfs.h> #define NLMDBG_FACILITY NLMDBG_SVC @@ -54,13 +53,26 @@ static struct svc_rqst *nlmsvc_rqst; unsigned long nlmsvc_timeout; /* + * If the kernel has IPv6 support available, always listen for + * both AF_INET and AF_INET6 requests. + */ +#if (defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)) && \ + defined(CONFIG_SUNRPC_REGISTER_V4) +static const sa_family_t nlmsvc_family = AF_INET6; +#else /* (CONFIG_IPV6 || CONFIG_IPV6_MODULE) && CONFIG_SUNRPC_REGISTER_V4 */ +static const sa_family_t nlmsvc_family = AF_INET; +#endif /* (CONFIG_IPV6 || CONFIG_IPV6_MODULE) && CONFIG_SUNRPC_REGISTER_V4 */ + +/* * These can be set at insmod time (useful for NFS as root filesystem), * and also changed through the sysctl interface. -- Jamie Lokier, Aug 2003 */ static unsigned long nlm_grace_period; static unsigned long nlm_timeout = LOCKD_DFLT_TIMEO; static int nlm_udpport, nlm_tcpport; -int nsm_use_hostnames = 0; + +/* RLIM_NOFILE defaults to 1024. That seems like a reasonable default here. */ +static unsigned int nlm_max_connections = 1024; /* * Constants needed for the sysctl interface. @@ -143,6 +155,9 @@ lockd(void *vrqstp) long timeout = MAX_SCHEDULE_TIMEOUT; RPC_IFDEBUG(char buf[RPC_MAX_ADDRBUFLEN]); + /* update sv_maxconn if it has changed */ + rqstp->rq_server->sv_maxconn = nlm_max_connections; + if (signalled()) { flush_signals(current); if (nlmsvc_ops) { @@ -189,6 +204,19 @@ lockd(void *vrqstp) return 0; } +static int create_lockd_listener(struct svc_serv *serv, char *name, + unsigned short port) +{ + struct svc_xprt *xprt; + + xprt = svc_find_xprt(serv, name, 0, 0); + if (xprt == NULL) + return svc_create_xprt(serv, name, port, SVC_SOCK_DEFAULTS); + + svc_xprt_put(xprt); + return 0; +} + /* * Ensure there are active UDP and TCP listeners for lockd. * @@ -202,29 +230,23 @@ lockd(void *vrqstp) static int make_socks(struct svc_serv *serv) { static int warned; - struct svc_xprt *xprt; - int err = 0; + int err; - xprt = svc_find_xprt(serv, "udp", 0, 0); - if (!xprt) - err = svc_create_xprt(serv, "udp", nlm_udpport, - SVC_SOCK_DEFAULTS); - else - svc_xprt_put(xprt); - if (err >= 0) { - xprt = svc_find_xprt(serv, "tcp", 0, 0); - if (!xprt) - err = svc_create_xprt(serv, "tcp", nlm_tcpport, - SVC_SOCK_DEFAULTS); - else - svc_xprt_put(xprt); - } - if (err >= 0) { - warned = 0; - err = 0; - } else if (warned++ == 0) + err = create_lockd_listener(serv, "udp", nlm_udpport); + if (err < 0) + goto out_err; + + err = create_lockd_listener(serv, "tcp", nlm_tcpport); + if (err < 0) + goto out_err; + + warned = 0; + return 0; + +out_err: + if (warned++ == 0) printk(KERN_WARNING - "lockd_up: makesock failed, error=%d\n", err); + "lockd_up: makesock failed, error=%d\n", err); return err; } @@ -252,7 +274,7 @@ int lockd_up(void) "lockd_up: no pid, %d users??\n", nlmsvc_users); error = -ENOMEM; - serv = svc_create(&nlmsvc_program, LOCKD_BUFSIZE, AF_INET, NULL); + serv = svc_create(&nlmsvc_program, LOCKD_BUFSIZE, nlmsvc_family, NULL); if (!serv) { printk(KERN_WARNING "lockd_up: create service failed\n"); goto out; @@ -276,6 +298,7 @@ int lockd_up(void) } svc_sock_update_bufs(serv); + serv->sv_maxconn = nlm_max_connections; nlmsvc_task = kthread_run(lockd, nlmsvc_rqst, serv->sv_name); if (IS_ERR(nlmsvc_task)) { @@ -485,6 +508,7 @@ module_param_call(nlm_udpport, param_set_port, param_get_int, module_param_call(nlm_tcpport, param_set_port, param_get_int, &nlm_tcpport, 0644); module_param(nsm_use_hostnames, bool, 0644); +module_param(nlm_max_connections, uint, 0644); /* * Initialising and terminating the module. diff --git a/fs/lockd/svc4proc.c b/fs/lockd/svc4proc.c index 4dfdcbc6bf68..1725037374c5 100644 --- a/fs/lockd/svc4proc.c +++ b/fs/lockd/svc4proc.c @@ -16,8 +16,6 @@ #include <linux/nfsd/nfsd.h> #include <linux/lockd/lockd.h> #include <linux/lockd/share.h> -#include <linux/lockd/sm_inter.h> - #define NLMDBG_FACILITY NLMDBG_CLIENT @@ -419,8 +417,6 @@ static __be32 nlm4svc_proc_sm_notify(struct svc_rqst *rqstp, struct nlm_reboot *argp, void *resp) { - struct sockaddr_in saddr; - dprintk("lockd: SM_NOTIFY called\n"); if (!nlm_privileged_requester(rqstp)) { @@ -430,14 +426,7 @@ nlm4svc_proc_sm_notify(struct svc_rqst *rqstp, struct nlm_reboot *argp, return rpc_system_err; } - /* Obtain the host pointer for this NFS server and try to - * reclaim all locks we hold on this server. - */ - memset(&saddr, 0, sizeof(saddr)); - saddr.sin_family = AF_INET; - saddr.sin_addr.s_addr = argp->addr; - nlm_host_rebooted(&saddr, argp->mon, argp->len, argp->state); - + nlm_host_rebooted(argp); return rpc_success; } diff --git a/fs/lockd/svcproc.c b/fs/lockd/svcproc.c index 3ca89e2a9381..3688e55901fc 100644 --- a/fs/lockd/svcproc.c +++ b/fs/lockd/svcproc.c @@ -16,8 +16,6 @@ #include <linux/nfsd/nfsd.h> #include <linux/lockd/lockd.h> #include <linux/lockd/share.h> -#include <linux/lockd/sm_inter.h> - #define NLMDBG_FACILITY NLMDBG_CLIENT @@ -451,8 +449,6 @@ static __be32 nlmsvc_proc_sm_notify(struct svc_rqst *rqstp, struct nlm_reboot *argp, void *resp) { - struct sockaddr_in saddr; - dprintk("lockd: SM_NOTIFY called\n"); if (!nlm_privileged_requester(rqstp)) { @@ -462,14 +458,7 @@ nlmsvc_proc_sm_notify(struct svc_rqst *rqstp, struct nlm_reboot *argp, return rpc_system_err; } - /* Obtain the host pointer for this NFS server and try to - * reclaim all locks we hold on this server. - */ - memset(&saddr, 0, sizeof(saddr)); - saddr.sin_family = AF_INET; - saddr.sin_addr.s_addr = argp->addr; - nlm_host_rebooted(&saddr, argp->mon, argp->len, argp->state); - + nlm_host_rebooted(argp); return rpc_success; } diff --git a/fs/lockd/svcsubs.c b/fs/lockd/svcsubs.c index 34c2766e27c7..9e4d6aab611b 100644 --- a/fs/lockd/svcsubs.c +++ b/fs/lockd/svcsubs.c @@ -17,7 +17,6 @@ #include <linux/nfsd/export.h> #include <linux/lockd/lockd.h> #include <linux/lockd/share.h> -#include <linux/lockd/sm_inter.h> #include <linux/module.h> #include <linux/mount.h> diff --git a/fs/lockd/xdr.c b/fs/lockd/xdr.c index 1f226290c67c..0336f2beacde 100644 --- a/fs/lockd/xdr.c +++ b/fs/lockd/xdr.c @@ -16,7 +16,6 @@ #include <linux/sunrpc/svc.h> #include <linux/sunrpc/stats.h> #include <linux/lockd/lockd.h> -#include <linux/lockd/sm_inter.h> #define NLMDBG_FACILITY NLMDBG_XDR @@ -349,8 +348,8 @@ nlmsvc_decode_reboot(struct svc_rqst *rqstp, __be32 *p, struct nlm_reboot *argp) if (!(p = xdr_decode_string_inplace(p, &argp->mon, &argp->len, SM_MAXSTRLEN))) return 0; argp->state = ntohl(*p++); - /* Preserve the address in network byte order */ - argp->addr = *p++; + memcpy(&argp->priv.data, p, sizeof(argp->priv.data)); + p += XDR_QUADLEN(SM_PRIV_SIZE); return xdr_argsize_check(rqstp, p); } diff --git a/fs/lockd/xdr4.c b/fs/lockd/xdr4.c index 50c493a8ad8e..e1d528653192 100644 --- a/fs/lockd/xdr4.c +++ b/fs/lockd/xdr4.c @@ -17,7 +17,6 @@ #include <linux/sunrpc/svc.h> #include <linux/sunrpc/stats.h> #include <linux/lockd/lockd.h> -#include <linux/lockd/sm_inter.h> #define NLMDBG_FACILITY NLMDBG_XDR @@ -356,8 +355,8 @@ nlm4svc_decode_reboot(struct svc_rqst *rqstp, __be32 *p, struct nlm_reboot *argp if (!(p = xdr_decode_string_inplace(p, &argp->mon, &argp->len, SM_MAXSTRLEN))) return 0; argp->state = ntohl(*p++); - /* Preserve the address in network byte order */ - argp->addr = *p++; + memcpy(&argp->priv.data, p, sizeof(argp->priv.data)); + p += XDR_QUADLEN(SM_PRIV_SIZE); return xdr_argsize_check(rqstp, p); } diff --git a/fs/minix/dir.c b/fs/minix/dir.c index f70433816a38..d4946c4c90e2 100644 --- a/fs/minix/dir.c +++ b/fs/minix/dir.c @@ -280,7 +280,7 @@ int minix_add_link(struct dentry *dentry, struct inode *inode) return -EINVAL; got_it: - pos = (page->index >> PAGE_CACHE_SHIFT) + p - (char*)page_address(page); + pos = page_offset(page) + p - (char *)page_address(page); err = __minix_write_begin(NULL, page->mapping, pos, sbi->s_dirsize, AOP_FLAG_UNINTERRUPTIBLE, &page, NULL); if (err) diff --git a/fs/mpage.c b/fs/mpage.c index 552b80b3facc..16c3ef37eae3 100644 --- a/fs/mpage.c +++ b/fs/mpage.c @@ -241,7 +241,6 @@ do_mpage_readpage(struct bio *bio, struct page *page, unsigned nr_pages, first_hole = page_block; page_block++; block_in_file++; - clear_buffer_mapped(map_bh); continue; } @@ -308,7 +307,10 @@ alloc_new: goto alloc_new; } - if (buffer_boundary(map_bh) || (first_hole != blocks_per_page)) + relative_block = block_in_file - *first_logical_block; + nblocks = map_bh->b_size >> blkbits; + if ((buffer_boundary(map_bh) && relative_block == nblocks) || + (first_hole != blocks_per_page)) bio = mpage_bio_submit(READ, bio); else *last_block_in_bio = blocks[blocks_per_page - 1]; diff --git a/fs/ncpfs/getopt.c b/fs/ncpfs/getopt.c index 335b003dddf9..0af3349de851 100644 --- a/fs/ncpfs/getopt.c +++ b/fs/ncpfs/getopt.c @@ -16,7 +16,6 @@ * @opts: an array of &struct option entries controlling parser operations * @optopt: output; will contain the current option * @optarg: output; will contain the value (if one exists) - * @flag: output; may be NULL; should point to a long for or'ing flags * @value: output; may be NULL; will be overwritten with the integer value * of the current argument. * diff --git a/fs/ncpfs/ioctl.c b/fs/ncpfs/ioctl.c index 6d04e050c74e..f54360f50a9c 100644 --- a/fs/ncpfs/ioctl.c +++ b/fs/ncpfs/ioctl.c @@ -98,7 +98,7 @@ struct compat_ncp_objectname_ioctl { s32 auth_type; u32 object_name_len; - compat_caddr_t object_name; /* an userspace data, in most cases user name */ + compat_caddr_t object_name; /* a userspace data, in most cases user name */ }; struct compat_ncp_fs_info_v2 { diff --git a/fs/nfsd/auth.c b/fs/nfsd/auth.c index 0184fe9b514c..c903e04aa217 100644 --- a/fs/nfsd/auth.c +++ b/fs/nfsd/auth.c @@ -76,10 +76,10 @@ int nfsd_setuser(struct svc_rqst *rqstp, struct svc_export *exp) ret = set_groups(new, gi); put_group_info(gi); - if (!ret) + if (ret < 0) goto error; - if (new->uid) + if (new->fsuid) new->cap_effective = cap_drop_nfsd_set(new->cap_effective); else new->cap_effective = cap_raise_nfsd_set(new->cap_effective, diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c index 6d7d8c02c197..c464181b5994 100644 --- a/fs/nfsd/nfs4callback.c +++ b/fs/nfsd/nfs4callback.c @@ -53,9 +53,6 @@ #define NFSPROC4_CB_NULL 0 #define NFSPROC4_CB_COMPOUND 1 -/* declarations */ -static const struct rpc_call_ops nfs4_cb_null_ops; - /* Index of predefined Linux callback client operations */ enum { diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index 669461e291ae..9fa60a3ad48c 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -946,6 +946,11 @@ encode_op: nfsd4_encode_operation(resp, op); status = op->status; } + + dprintk("nfsv4 compound op %p opcnt %d #%d: %d: status %d\n", + args->ops, args->opcnt, resp->opcnt, op->opnum, + be32_to_cpu(status)); + if (cstate->replay_owner) { nfs4_put_stateowner(cstate->replay_owner); cstate->replay_owner = NULL; diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c index 0f9d6efaa62b..74f7b67567fd 100644 --- a/fs/nfsd/nfs4recover.c +++ b/fs/nfsd/nfs4recover.c @@ -116,9 +116,9 @@ nfs4_make_rec_clidname(char *dname, struct xdr_netobj *clname) md5_to_hex(dname, cksum.data); - kfree(cksum.data); status = nfs_ok; out: + kfree(cksum.data); crypto_free_hash(desc.tfm); out_no_tfm: return status; diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 13e0e074dbb8..88db7d3ec120 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -2416,6 +2416,26 @@ out: #define LOCK_HASH_SIZE (1 << LOCK_HASH_BITS) #define LOCK_HASH_MASK (LOCK_HASH_SIZE - 1) +static inline u64 +end_offset(u64 start, u64 len) +{ + u64 end; + + end = start + len; + return end >= start ? end: NFS4_MAX_UINT64; +} + +/* last octet in a range */ +static inline u64 +last_byte_offset(u64 start, u64 len) +{ + u64 end; + + BUG_ON(!len); + end = start + len; + return end > start ? end - 1: NFS4_MAX_UINT64; +} + #define lockownerid_hashval(id) \ ((id) & LOCK_HASH_MASK) @@ -2435,13 +2455,13 @@ static struct list_head lockstateid_hashtbl[STATEID_HASH_SIZE]; static struct nfs4_stateid * find_stateid(stateid_t *stid, int flags) { - struct nfs4_stateid *local = NULL; + struct nfs4_stateid *local; u32 st_id = stid->si_stateownerid; u32 f_id = stid->si_fileid; unsigned int hashval; dprintk("NFSD: find_stateid flags 0x%x\n",flags); - if ((flags & LOCK_STATE) || (flags & RD_STATE) || (flags & WR_STATE)) { + if (flags & (LOCK_STATE | RD_STATE | WR_STATE)) { hashval = stateid_hashval(st_id, f_id); list_for_each_entry(local, &lockstateid_hashtbl[hashval], st_hash) { if ((local->st_stateid.si_stateownerid == st_id) && @@ -2449,7 +2469,8 @@ find_stateid(stateid_t *stid, int flags) return local; } } - if ((flags & OPEN_STATE) || (flags & RD_STATE) || (flags & WR_STATE)) { + + if (flags & (OPEN_STATE | RD_STATE | WR_STATE)) { hashval = stateid_hashval(st_id, f_id); list_for_each_entry(local, &stateid_hashtbl[hashval], st_hash) { if ((local->st_stateid.si_stateownerid == st_id) && @@ -2518,8 +2539,8 @@ nfs4_set_lock_denied(struct file_lock *fl, struct nfsd4_lock_denied *deny) deny->ld_clientid.cl_id = 0; } deny->ld_start = fl->fl_start; - deny->ld_length = ~(u64)0; - if (fl->fl_end != ~(u64)0) + deny->ld_length = NFS4_MAX_UINT64; + if (fl->fl_end != NFS4_MAX_UINT64) deny->ld_length = fl->fl_end - fl->fl_start + 1; deny->ld_type = NFS4_READ_LT; if (fl->fl_type != F_RDLCK) @@ -2616,7 +2637,7 @@ out: static int check_lock_length(u64 offset, u64 length) { - return ((length == 0) || ((length != ~(u64)0) && + return ((length == 0) || ((length != NFS4_MAX_UINT64) && LOFF_OVERFLOW(offset, length))); } @@ -2736,11 +2757,7 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, file_lock.fl_lmops = &nfsd_posix_mng_ops; file_lock.fl_start = lock->lk_offset; - if ((lock->lk_length == ~(u64)0) || - LOFF_OVERFLOW(lock->lk_offset, lock->lk_length)) - file_lock.fl_end = ~(u64)0; - else - file_lock.fl_end = lock->lk_offset + lock->lk_length - 1; + file_lock.fl_end = last_byte_offset(lock->lk_offset, lock->lk_length); nfs4_transform_lock_offset(&file_lock); /* @@ -2781,6 +2798,25 @@ out: } /* + * The NFSv4 spec allows a client to do a LOCKT without holding an OPEN, + * so we do a temporary open here just to get an open file to pass to + * vfs_test_lock. (Arguably perhaps test_lock should be done with an + * inode operation.) + */ +static int nfsd_test_lock(struct svc_rqst *rqstp, struct svc_fh *fhp, struct file_lock *lock) +{ + struct file *file; + int err; + + err = nfsd_open(rqstp, fhp, S_IFREG, NFSD_MAY_READ, &file); + if (err) + return err; + err = vfs_test_lock(file, lock); + nfsd_close(file); + return err; +} + +/* * LOCKT operation */ __be32 @@ -2788,7 +2824,6 @@ nfsd4_lockt(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, struct nfsd4_lockt *lockt) { struct inode *inode; - struct file file; struct file_lock file_lock; int error; __be32 status; @@ -2839,23 +2874,12 @@ nfsd4_lockt(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, file_lock.fl_lmops = &nfsd_posix_mng_ops; file_lock.fl_start = lockt->lt_offset; - if ((lockt->lt_length == ~(u64)0) || LOFF_OVERFLOW(lockt->lt_offset, lockt->lt_length)) - file_lock.fl_end = ~(u64)0; - else - file_lock.fl_end = lockt->lt_offset + lockt->lt_length - 1; + file_lock.fl_end = last_byte_offset(lockt->lt_offset, lockt->lt_length); nfs4_transform_lock_offset(&file_lock); - /* vfs_test_lock uses the struct file _only_ to resolve the inode. - * since LOCKT doesn't require an OPEN, and therefore a struct - * file may not exist, pass vfs_test_lock a struct file with - * only the dentry:inode set. - */ - memset(&file, 0, sizeof (struct file)); - file.f_path.dentry = cstate->current_fh.fh_dentry; - status = nfs_ok; - error = vfs_test_lock(&file, &file_lock); + error = nfsd_test_lock(rqstp, &cstate->current_fh, &file_lock); if (error) { status = nfserrno(error); goto out; @@ -2906,10 +2930,7 @@ nfsd4_locku(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, file_lock.fl_lmops = &nfsd_posix_mng_ops; file_lock.fl_start = locku->lu_offset; - if ((locku->lu_length == ~(u64)0) || LOFF_OVERFLOW(locku->lu_offset, locku->lu_length)) - file_lock.fl_end = ~(u64)0; - else - file_lock.fl_end = locku->lu_offset + locku->lu_length - 1; + file_lock.fl_end = last_byte_offset(locku->lu_offset, locku->lu_length); nfs4_transform_lock_offset(&file_lock); /* diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index afcdf4b76843..f65953be39c0 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -1,6 +1,4 @@ /* - * fs/nfs/nfs4xdr.c - * * Server-side XDR for NFSv4 * * Copyright (c) 2002 The Regents of the University of Michigan. diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c index 77d7b8c531a6..3d93b2064ce5 100644 --- a/fs/nfsd/nfsctl.c +++ b/fs/nfsd/nfsctl.c @@ -84,6 +84,8 @@ static ssize_t write_unexport(struct file *file, char *buf, size_t size); static ssize_t write_getfd(struct file *file, char *buf, size_t size); static ssize_t write_getfs(struct file *file, char *buf, size_t size); static ssize_t write_filehandle(struct file *file, char *buf, size_t size); +static ssize_t write_unlock_ip(struct file *file, char *buf, size_t size); +static ssize_t write_unlock_fs(struct file *file, char *buf, size_t size); static ssize_t write_threads(struct file *file, char *buf, size_t size); static ssize_t write_pool_threads(struct file *file, char *buf, size_t size); static ssize_t write_versions(struct file *file, char *buf, size_t size); @@ -94,9 +96,6 @@ static ssize_t write_leasetime(struct file *file, char *buf, size_t size); static ssize_t write_recoverydir(struct file *file, char *buf, size_t size); #endif -static ssize_t failover_unlock_ip(struct file *file, char *buf, size_t size); -static ssize_t failover_unlock_fs(struct file *file, char *buf, size_t size); - static ssize_t (*write_op[])(struct file *, char *, size_t) = { [NFSD_Svc] = write_svc, [NFSD_Add] = write_add, @@ -106,8 +105,8 @@ static ssize_t (*write_op[])(struct file *, char *, size_t) = { [NFSD_Getfd] = write_getfd, [NFSD_Getfs] = write_getfs, [NFSD_Fh] = write_filehandle, - [NFSD_FO_UnlockIP] = failover_unlock_ip, - [NFSD_FO_UnlockFS] = failover_unlock_fs, + [NFSD_FO_UnlockIP] = write_unlock_ip, + [NFSD_FO_UnlockFS] = write_unlock_fs, [NFSD_Threads] = write_threads, [NFSD_Pool_Threads] = write_pool_threads, [NFSD_Versions] = write_versions, @@ -176,10 +175,24 @@ static const struct file_operations exports_operations = { /*----------------------------------------------------------------------------*/ /* * payload - write methods - * If the method has a response, the response should be put in buf, - * and the length returned. Otherwise return 0 or and -error. */ +/** + * write_svc - Start kernel's NFSD server + * + * Deprecated. /proc/fs/nfsd/threads is preferred. + * Function remains to support old versions of nfs-utils. + * + * Input: + * buf: struct nfsctl_svc + * svc_port: port number of this + * server's listener + * svc_nthreads: number of threads to start + * size: size in bytes of passed in nfsctl_svc + * Output: + * On success: returns zero + * On error: return code is negative errno value + */ static ssize_t write_svc(struct file *file, char *buf, size_t size) { struct nfsctl_svc *data; @@ -189,6 +202,30 @@ static ssize_t write_svc(struct file *file, char *buf, size_t size) return nfsd_svc(data->svc_port, data->svc_nthreads); } +/** + * write_add - Add or modify client entry in auth unix cache + * + * Deprecated. /proc/net/rpc/auth.unix.ip is preferred. + * Function remains to support old versions of nfs-utils. + * + * Input: + * buf: struct nfsctl_client + * cl_ident: '\0'-terminated C string + * containing domain name + * of client + * cl_naddr: no. of items in cl_addrlist + * cl_addrlist: array of client addresses + * cl_fhkeytype: ignored + * cl_fhkeylen: ignored + * cl_fhkey: ignored + * size: size in bytes of passed in nfsctl_client + * Output: + * On success: returns zero + * On error: return code is negative errno value + * + * Note: Only AF_INET client addresses are passed in, since + * nfsctl_client.cl_addrlist contains only in_addr fields for addresses. + */ static ssize_t write_add(struct file *file, char *buf, size_t size) { struct nfsctl_client *data; @@ -198,6 +235,30 @@ static ssize_t write_add(struct file *file, char *buf, size_t size) return exp_addclient(data); } +/** + * write_del - Remove client from auth unix cache + * + * Deprecated. /proc/net/rpc/auth.unix.ip is preferred. + * Function remains to support old versions of nfs-utils. + * + * Input: + * buf: struct nfsctl_client + * cl_ident: '\0'-terminated C string + * containing domain name + * of client + * cl_naddr: ignored + * cl_addrlist: ignored + * cl_fhkeytype: ignored + * cl_fhkeylen: ignored + * cl_fhkey: ignored + * size: size in bytes of passed in nfsctl_client + * Output: + * On success: returns zero + * On error: return code is negative errno value + * + * Note: Only AF_INET client addresses are passed in, since + * nfsctl_client.cl_addrlist contains only in_addr fields for addresses. + */ static ssize_t write_del(struct file *file, char *buf, size_t size) { struct nfsctl_client *data; @@ -207,6 +268,33 @@ static ssize_t write_del(struct file *file, char *buf, size_t size) return exp_delclient(data); } +/** + * write_export - Export part or all of a local file system + * + * Deprecated. /proc/net/rpc/{nfsd.export,nfsd.fh} are preferred. + * Function remains to support old versions of nfs-utils. + * + * Input: + * buf: struct nfsctl_export + * ex_client: '\0'-terminated C string + * containing domain name + * of client allowed to access + * this export + * ex_path: '\0'-terminated C string + * containing pathname of + * directory in local file system + * ex_dev: fsid to use for this export + * ex_ino: ignored + * ex_flags: export flags for this export + * ex_anon_uid: UID to use for anonymous + * requests + * ex_anon_gid: GID to use for anonymous + * requests + * size: size in bytes of passed in nfsctl_export + * Output: + * On success: returns zero + * On error: return code is negative errno value + */ static ssize_t write_export(struct file *file, char *buf, size_t size) { struct nfsctl_export *data; @@ -216,6 +304,31 @@ static ssize_t write_export(struct file *file, char *buf, size_t size) return exp_export(data); } +/** + * write_unexport - Unexport a previously exported file system + * + * Deprecated. /proc/net/rpc/{nfsd.export,nfsd.fh} are preferred. + * Function remains to support old versions of nfs-utils. + * + * Input: + * buf: struct nfsctl_export + * ex_client: '\0'-terminated C string + * containing domain name + * of client no longer allowed + * to access this export + * ex_path: '\0'-terminated C string + * containing pathname of + * directory in local file system + * ex_dev: ignored + * ex_ino: ignored + * ex_flags: ignored + * ex_anon_uid: ignored + * ex_anon_gid: ignored + * size: size in bytes of passed in nfsctl_export + * Output: + * On success: returns zero + * On error: return code is negative errno value + */ static ssize_t write_unexport(struct file *file, char *buf, size_t size) { struct nfsctl_export *data; @@ -226,6 +339,30 @@ static ssize_t write_unexport(struct file *file, char *buf, size_t size) return exp_unexport(data); } +/** + * write_getfs - Get a variable-length NFS file handle by path + * + * Deprecated. /proc/fs/nfsd/filehandle is preferred. + * Function remains to support old versions of nfs-utils. + * + * Input: + * buf: struct nfsctl_fsparm + * gd_addr: socket address of client + * gd_path: '\0'-terminated C string + * containing pathname of + * directory in local file system + * gd_maxlen: maximum size of returned file + * handle + * size: size in bytes of passed in nfsctl_fsparm + * Output: + * On success: passed-in buffer filled with a knfsd_fh structure + * (a variable-length raw NFS file handle); + * return code is the size in bytes of the file handle + * On error: return code is negative errno value + * + * Note: Only AF_INET client addresses are passed in, since gd_addr + * is the same size as a struct sockaddr_in. + */ static ssize_t write_getfs(struct file *file, char *buf, size_t size) { struct nfsctl_fsparm *data; @@ -265,6 +402,29 @@ static ssize_t write_getfs(struct file *file, char *buf, size_t size) return err; } +/** + * write_getfd - Get a fixed-length NFS file handle by path (used by mountd) + * + * Deprecated. /proc/fs/nfsd/filehandle is preferred. + * Function remains to support old versions of nfs-utils. + * + * Input: + * buf: struct nfsctl_fdparm + * gd_addr: socket address of client + * gd_path: '\0'-terminated C string + * containing pathname of + * directory in local file system + * gd_version: fdparm structure version + * size: size in bytes of passed in nfsctl_fdparm + * Output: + * On success: passed-in buffer filled with nfsctl_res + * (a fixed-length raw NFS file handle); + * return code is the size in bytes of the file handle + * On error: return code is negative errno value + * + * Note: Only AF_INET client addresses are passed in, since gd_addr + * is the same size as a struct sockaddr_in. + */ static ssize_t write_getfd(struct file *file, char *buf, size_t size) { struct nfsctl_fdparm *data; @@ -309,7 +469,23 @@ static ssize_t write_getfd(struct file *file, char *buf, size_t size) return err; } -static ssize_t failover_unlock_ip(struct file *file, char *buf, size_t size) +/** + * write_unlock_ip - Release all locks used by a client + * + * Experimental. + * + * Input: + * buf: '\n'-terminated C string containing a + * presentation format IPv4 address + * size: length of C string in @buf + * Output: + * On success: returns zero if all specified locks were released; + * returns one if one or more locks were not released + * On error: return code is negative errno value + * + * Note: Only AF_INET client addresses are passed in + */ +static ssize_t write_unlock_ip(struct file *file, char *buf, size_t size) { struct sockaddr_in sin = { .sin_family = AF_INET, @@ -339,7 +515,21 @@ static ssize_t failover_unlock_ip(struct file *file, char *buf, size_t size) return nlmsvc_unlock_all_by_ip((struct sockaddr *)&sin); } -static ssize_t failover_unlock_fs(struct file *file, char *buf, size_t size) +/** + * write_unlock_fs - Release all locks on a local file system + * + * Experimental. + * + * Input: + * buf: '\n'-terminated C string containing the + * absolute pathname of a local file system + * size: length of C string in @buf + * Output: + * On success: returns zero if all specified locks were released; + * returns one if one or more locks were not released + * On error: return code is negative errno value + */ +static ssize_t write_unlock_fs(struct file *file, char *buf, size_t size) { struct path path; char *fo_path; @@ -360,21 +550,44 @@ static ssize_t failover_unlock_fs(struct file *file, char *buf, size_t size) if (error) return error; + /* + * XXX: Needs better sanity checking. Otherwise we could end up + * releasing locks on the wrong file system. + * + * For example: + * 1. Does the path refer to a directory? + * 2. Is that directory a mount point, or + * 3. Is that directory the root of an exported file system? + */ error = nlmsvc_unlock_all_by_sb(path.mnt->mnt_sb); path_put(&path); return error; } +/** + * write_filehandle - Get a variable-length NFS file handle by path + * + * On input, the buffer contains a '\n'-terminated C string comprised of + * three alphanumeric words separated by whitespace. The string may + * contain escape sequences. + * + * Input: + * buf: + * domain: client domain name + * path: export pathname + * maxsize: numeric maximum size of + * @buf + * size: length of C string in @buf + * Output: + * On success: passed-in buffer filled with '\n'-terminated C + * string containing a ASCII hex text version + * of the NFS file handle; + * return code is the size in bytes of the string + * On error: return code is negative errno value + */ static ssize_t write_filehandle(struct file *file, char *buf, size_t size) { - /* request is: - * domain path maxsize - * response is - * filehandle - * - * qword quoting is used, so filehandle will be \x.... - */ char *dname, *path; int uninitialized_var(maxsize); char *mesg = buf; @@ -391,11 +604,13 @@ static ssize_t write_filehandle(struct file *file, char *buf, size_t size) dname = mesg; len = qword_get(&mesg, dname, size); - if (len <= 0) return -EINVAL; + if (len <= 0) + return -EINVAL; path = dname+len+1; len = qword_get(&mesg, path, size); - if (len <= 0) return -EINVAL; + if (len <= 0) + return -EINVAL; len = get_int(&mesg, &maxsize); if (len) @@ -419,17 +634,43 @@ static ssize_t write_filehandle(struct file *file, char *buf, size_t size) if (len) return len; - mesg = buf; len = SIMPLE_TRANSACTION_LIMIT; + mesg = buf; + len = SIMPLE_TRANSACTION_LIMIT; qword_addhex(&mesg, &len, (char*)&fh.fh_base, fh.fh_size); mesg[-1] = '\n'; return mesg - buf; } +/** + * write_threads - Start NFSD, or report the current number of running threads + * + * Input: + * buf: ignored + * size: zero + * Output: + * On success: passed-in buffer filled with '\n'-terminated C + * string numeric value representing the number of + * running NFSD threads; + * return code is the size in bytes of the string + * On error: return code is zero + * + * OR + * + * Input: + * buf: C string containing an unsigned + * integer value representing the + * number of NFSD threads to start + * size: non-zero length of C string in @buf + * Output: + * On success: NFS service is started; + * passed-in buffer filled with '\n'-terminated C + * string numeric value representing the number of + * running NFSD threads; + * return code is the size in bytes of the string + * On error: return code is zero or a negative errno value + */ static ssize_t write_threads(struct file *file, char *buf, size_t size) { - /* if size > 0, look for a number of threads and call nfsd_svc - * then write out number of threads as reply - */ char *mesg = buf; int rv; if (size > 0) { @@ -437,9 +678,9 @@ static ssize_t write_threads(struct file *file, char *buf, size_t size) rv = get_int(&mesg, &newthreads); if (rv) return rv; - if (newthreads <0) + if (newthreads < 0) return -EINVAL; - rv = nfsd_svc(2049, newthreads); + rv = nfsd_svc(NFS_PORT, newthreads); if (rv) return rv; } @@ -447,6 +688,28 @@ static ssize_t write_threads(struct file *file, char *buf, size_t size) return strlen(buf); } +/** + * write_pool_threads - Set or report the current number of threads per pool + * + * Input: + * buf: ignored + * size: zero + * + * OR + * + * Input: + * buf: C string containing whitespace- + * separated unsigned integer values + * representing the number of NFSD + * threads to start in each pool + * size: non-zero length of C string in @buf + * Output: + * On success: passed-in buffer filled with '\n'-terminated C + * string containing integer values representing the + * number of NFSD threads in each pool; + * return code is the size in bytes of the string + * On error: return code is zero or a negative errno value + */ static ssize_t write_pool_threads(struct file *file, char *buf, size_t size) { /* if size > 0, look for an array of number of threads per node @@ -517,10 +780,6 @@ out_free: static ssize_t __write_versions(struct file *file, char *buf, size_t size) { - /* - * Format: - * [-/+]vers [-/+]vers ... - */ char *mesg = buf; char *vers, sign; int len, num; @@ -578,6 +837,38 @@ static ssize_t __write_versions(struct file *file, char *buf, size_t size) return len; } +/** + * write_versions - Set or report the available NFS protocol versions + * + * Input: + * buf: ignored + * size: zero + * Output: + * On success: passed-in buffer filled with '\n'-terminated C + * string containing positive or negative integer + * values representing the current status of each + * protocol version; + * return code is the size in bytes of the string + * On error: return code is zero or a negative errno value + * + * OR + * + * Input: + * buf: C string containing whitespace- + * separated positive or negative + * integer values representing NFS + * protocol versions to enable ("+n") + * or disable ("-n") + * size: non-zero length of C string in @buf + * Output: + * On success: status of zero or more protocol versions has + * been updated; passed-in buffer filled with + * '\n'-terminated C string containing positive + * or negative integer values representing the + * current status of each protocol version; + * return code is the size in bytes of the string + * On error: return code is zero or a negative errno value + */ static ssize_t write_versions(struct file *file, char *buf, size_t size) { ssize_t rv; @@ -687,6 +978,75 @@ static ssize_t __write_ports(struct file *file, char *buf, size_t size) return -EINVAL; } +/** + * write_ports - Pass a socket file descriptor or transport name to listen on + * + * Input: + * buf: ignored + * size: zero + * Output: + * On success: passed-in buffer filled with a '\n'-terminated C + * string containing a whitespace-separated list of + * named NFSD listeners; + * return code is the size in bytes of the string + * On error: return code is zero or a negative errno value + * + * OR + * + * Input: + * buf: C string containing an unsigned + * integer value representing a bound + * but unconnected socket that is to be + * used as an NFSD listener + * size: non-zero length of C string in @buf + * Output: + * On success: NFS service is started; + * passed-in buffer filled with a '\n'-terminated C + * string containing a unique alphanumeric name of + * the listener; + * return code is the size in bytes of the string + * On error: return code is a negative errno value + * + * OR + * + * Input: + * buf: C string containing a "-" followed + * by an integer value representing a + * previously passed in socket file + * descriptor + * size: non-zero length of C string in @buf + * Output: + * On success: NFS service no longer listens on that socket; + * passed-in buffer filled with a '\n'-terminated C + * string containing a unique name of the listener; + * return code is the size in bytes of the string + * On error: return code is a negative errno value + * + * OR + * + * Input: + * buf: C string containing a transport + * name and an unsigned integer value + * representing the port to listen on, + * separated by whitespace + * size: non-zero length of C string in @buf + * Output: + * On success: returns zero; NFS service is started + * On error: return code is a negative errno value + * + * OR + * + * Input: + * buf: C string containing a "-" followed + * by a transport name and an unsigned + * integer value representing the port + * to listen on, separated by whitespace + * size: non-zero length of C string in @buf + * Output: + * On success: returns zero; NFS service no longer listens + * on that transport + * On error: return code is a negative errno value + */ static ssize_t write_ports(struct file *file, char *buf, size_t size) { ssize_t rv; @@ -700,6 +1060,27 @@ static ssize_t write_ports(struct file *file, char *buf, size_t size) int nfsd_max_blksize; +/** + * write_maxblksize - Set or report the current NFS blksize + * + * Input: + * buf: ignored + * size: zero + * + * OR + * + * Input: + * buf: C string containing an unsigned + * integer value representing the new + * NFS blksize + * size: non-zero length of C string in @buf + * Output: + * On success: passed-in buffer filled with '\n'-terminated C string + * containing numeric value of the current NFS blksize + * setting; + * return code is the size in bytes of the string + * On error: return code is zero or a negative errno value + */ static ssize_t write_maxblksize(struct file *file, char *buf, size_t size) { char *mesg = buf; @@ -752,6 +1133,27 @@ static ssize_t __write_leasetime(struct file *file, char *buf, size_t size) return strlen(buf); } +/** + * write_leasetime - Set or report the current NFSv4 lease time + * + * Input: + * buf: ignored + * size: zero + * + * OR + * + * Input: + * buf: C string containing an unsigned + * integer value representing the new + * NFSv4 lease expiry time + * size: non-zero length of C string in @buf + * Output: + * On success: passed-in buffer filled with '\n'-terminated C + * string containing unsigned integer value of the + * current lease expiry time; + * return code is the size in bytes of the string + * On error: return code is zero or a negative errno value + */ static ssize_t write_leasetime(struct file *file, char *buf, size_t size) { ssize_t rv; @@ -788,6 +1190,27 @@ static ssize_t __write_recoverydir(struct file *file, char *buf, size_t size) return strlen(buf); } +/** + * write_recoverydir - Set or report the pathname of the recovery directory + * + * Input: + * buf: ignored + * size: zero + * + * OR + * + * Input: + * buf: C string containing the pathname + * of the directory on a local file + * system containing permanent NFSv4 + * recovery data + * size: non-zero length of C string in @buf + * Output: + * On success: passed-in buffer filled with '\n'-terminated C string + * containing the current recovery pathname setting; + * return code is the size in bytes of the string + * On error: return code is zero or a negative errno value + */ static ssize_t write_recoverydir(struct file *file, char *buf, size_t size) { ssize_t rv; diff --git a/fs/nfsd/nfsfh.c b/fs/nfsd/nfsfh.c index f0da7d9c3a92..9f1ca17293d3 100644 --- a/fs/nfsd/nfsfh.c +++ b/fs/nfsd/nfsfh.c @@ -258,14 +258,32 @@ out: return error; } -/* - * Perform sanity checks on the dentry in a client's file handle. +/** + * fh_verify - filehandle lookup and access checking + * @rqstp: pointer to current rpc request + * @fhp: filehandle to be verified + * @type: expected type of object pointed to by filehandle + * @access: type of access needed to object + * + * Look up a dentry from the on-the-wire filehandle, check the client's + * access to the export, and set the current task's credentials. + * + * Regardless of success or failure of fh_verify(), fh_put() should be + * called on @fhp when the caller is finished with the filehandle. * - * Note that the file handle dentry may need to be freed even after - * an error return. + * fh_verify() may be called multiple times on a given filehandle, for + * example, when processing an NFSv4 compound. The first call will look + * up a dentry using the on-the-wire filehandle. Subsequent calls will + * skip the lookup and just perform the other checks and possibly change + * the current task's credentials. * - * This is only called at the start of an nfsproc call, so fhp points to - * a svc_fh which is all 0 except for the over-the-wire file handle. + * @type specifies the type of object expected using one of the S_IF* + * constants defined in include/linux/stat.h. The caller may use zero + * to indicate that it doesn't care, or a negative integer to indicate + * that it expects something not of the given type. + * + * @access is formed from the NFSD_MAY_* constants defined in + * include/linux/nfsd/nfsd.h. */ __be32 fh_verify(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, int access) @@ -466,6 +484,8 @@ fh_compose(struct svc_fh *fhp, struct svc_export *exp, struct dentry *dentry, goto retry; break; } + } else if (exp->ex_flags & NFSEXP_FSID) { + fsid_type = FSID_NUM; } else if (exp->ex_uuid) { if (fhp->fh_maxsize >= 64) { if (root_export) @@ -478,9 +498,7 @@ fh_compose(struct svc_fh *fhp, struct svc_export *exp, struct dentry *dentry, else fsid_type = FSID_UUID4_INUM; } - } else if (exp->ex_flags & NFSEXP_FSID) - fsid_type = FSID_NUM; - else if (!old_valid_dev(ex_dev)) + } else if (!old_valid_dev(ex_dev)) /* for newer device numbers, we must use a newer fsid format */ fsid_type = FSID_ENCODE_DEV; else diff --git a/fs/nfsd/nfsproc.c b/fs/nfsd/nfsproc.c index 5cffeca7acef..6f7f26351227 100644 --- a/fs/nfsd/nfsproc.c +++ b/fs/nfsd/nfsproc.c @@ -622,6 +622,7 @@ nfserrno (int errno) { nfserr_badname, -ESRCH }, { nfserr_io, -ETXTBSY }, { nfserr_notsupp, -EOPNOTSUPP }, + { nfserr_toosmall, -ETOOSMALL }, }; int i; diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 44aa92aba891..6e50aaa56ca2 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -744,16 +744,44 @@ nfsd_close(struct file *filp) fput(filp); } +/* + * Sync a file + * As this calls fsync (not fdatasync) there is no need for a write_inode + * after it. + */ +static inline int nfsd_dosync(struct file *filp, struct dentry *dp, + const struct file_operations *fop) +{ + struct inode *inode = dp->d_inode; + int (*fsync) (struct file *, struct dentry *, int); + int err; + + err = filemap_fdatawrite(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; +} + static int nfsd_sync(struct file *filp) { - return vfs_fsync(filp, filp->f_path.dentry, 0); + int err; + struct inode *inode = filp->f_path.dentry->d_inode; + dprintk("nfsd: sync file %s\n", filp->f_path.dentry->d_name.name); + mutex_lock(&inode->i_mutex); + err=nfsd_dosync(filp, filp->f_path.dentry, filp->f_op); + mutex_unlock(&inode->i_mutex); + + return err; } int -nfsd_sync_dir(struct dentry *dentry) +nfsd_sync_dir(struct dentry *dp) { - return vfs_fsync(NULL, dentry, 0); + return nfsd_dosync(NULL, dp, dp->d_inode->i_fop); } /* diff --git a/fs/ocfs2/cluster/heartbeat.c b/fs/ocfs2/cluster/heartbeat.c index 6ebaa58e2c03..04697ba7f73e 100644 --- a/fs/ocfs2/cluster/heartbeat.c +++ b/fs/ocfs2/cluster/heartbeat.c @@ -854,7 +854,7 @@ static int o2hb_thread(void *data) while (!kthread_should_stop() && !reg->hr_unclean_stop) { /* We track the time spent inside - * o2hb_do_disk_heartbeat so that we avoid more then + * o2hb_do_disk_heartbeat so that we avoid more than * hr_timeout_ms between disk writes. On busy systems * this should result in a heartbeat which is less * likely to time itself out. */ diff --git a/fs/partitions/check.c b/fs/partitions/check.c index 6d5b213b8a9b..5198ada67398 100644 --- a/fs/partitions/check.c +++ b/fs/partitions/check.c @@ -384,9 +384,9 @@ struct hd_struct *add_partition(struct gendisk *disk, int partno, dname = dev_name(ddev); if (isdigit(dname[strlen(dname) - 1])) - snprintf(pdev->bus_id, BUS_ID_SIZE, "%sp%d", dname, partno); + dev_set_name(pdev, "%sp%d", dname, partno); else - snprintf(pdev->bus_id, BUS_ID_SIZE, "%s%d", dname, partno); + dev_set_name(pdev, "%s%d", dname, partno); device_initialize(pdev); pdev->class = &block_class; @@ -447,16 +447,11 @@ void register_disk(struct gendisk *disk) struct block_device *bdev; struct disk_part_iter piter; struct hd_struct *part; - char *s; int err; ddev->parent = disk->driverfs_dev; - strlcpy(ddev->bus_id, disk->disk_name, BUS_ID_SIZE); - /* ewww... some of these buggers have / in the name... */ - s = strchr(ddev->bus_id, '/'); - if (s) - *s = '!'; + dev_set_name(ddev, disk->disk_name); /* delay uevents, until we scanned partition table */ ddev->uevent_suppress = 1; diff --git a/fs/proc/base.c b/fs/proc/base.c index 10fd5223d600..0c9de19a1633 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -65,6 +65,7 @@ #include <linux/mm.h> #include <linux/rcupdate.h> #include <linux/kallsyms.h> +#include <linux/stacktrace.h> #include <linux/resource.h> #include <linux/module.h> #include <linux/mount.h> @@ -109,25 +110,22 @@ struct pid_entry { .op = OP, \ } -#define DIR(NAME, MODE, OTYPE) \ - NOD(NAME, (S_IFDIR|(MODE)), \ - &proc_##OTYPE##_inode_operations, &proc_##OTYPE##_operations, \ - {} ) -#define LNK(NAME, OTYPE) \ +#define DIR(NAME, MODE, iops, fops) \ + NOD(NAME, (S_IFDIR|(MODE)), &iops, &fops, {} ) +#define LNK(NAME, get_link) \ NOD(NAME, (S_IFLNK|S_IRWXUGO), \ &proc_pid_link_inode_operations, NULL, \ - { .proc_get_link = &proc_##OTYPE##_link } ) -#define REG(NAME, MODE, OTYPE) \ - NOD(NAME, (S_IFREG|(MODE)), NULL, \ - &proc_##OTYPE##_operations, {}) -#define INF(NAME, MODE, OTYPE) \ + { .proc_get_link = get_link } ) +#define REG(NAME, MODE, fops) \ + NOD(NAME, (S_IFREG|(MODE)), NULL, &fops, {}) +#define INF(NAME, MODE, read) \ NOD(NAME, (S_IFREG|(MODE)), \ NULL, &proc_info_file_operations, \ - { .proc_read = &proc_##OTYPE } ) -#define ONE(NAME, MODE, OTYPE) \ + { .proc_read = read } ) +#define ONE(NAME, MODE, show) \ NOD(NAME, (S_IFREG|(MODE)), \ NULL, &proc_single_file_operations, \ - { .proc_show = &proc_##OTYPE } ) + { .proc_show = show } ) /* * Count the number of hardlinks for the pid_entry table, excluding the . @@ -308,9 +306,9 @@ static int proc_pid_auxv(struct task_struct *task, char *buffer) struct mm_struct *mm = get_task_mm(task); if (mm) { unsigned int nwords = 0; - do + do { nwords += 2; - while (mm->saved_auxv[nwords - 2] != 0); /* AT_NULL */ + } while (mm->saved_auxv[nwords - 2] != 0); /* AT_NULL */ res = nwords * sizeof(mm->saved_auxv[0]); if (res > PAGE_SIZE) res = PAGE_SIZE; @@ -340,6 +338,37 @@ static int proc_pid_wchan(struct task_struct *task, char *buffer) } #endif /* CONFIG_KALLSYMS */ +#ifdef CONFIG_STACKTRACE + +#define MAX_STACK_TRACE_DEPTH 64 + +static int proc_pid_stack(struct seq_file *m, struct pid_namespace *ns, + struct pid *pid, struct task_struct *task) +{ + struct stack_trace trace; + unsigned long *entries; + int i; + + entries = kmalloc(MAX_STACK_TRACE_DEPTH * sizeof(*entries), GFP_KERNEL); + if (!entries) + return -ENOMEM; + + trace.nr_entries = 0; + trace.max_entries = MAX_STACK_TRACE_DEPTH; + trace.entries = entries; + trace.skip = 0; + save_stack_trace_tsk(task, &trace); + + for (i = 0; i < trace.nr_entries; i++) { + seq_printf(m, "[<%p>] %pS\n", + (void *)entries[i], (void *)entries[i]); + } + kfree(entries); + + return 0; +} +#endif + #ifdef CONFIG_SCHEDSTATS /* * Provides /proc/PID/schedstat @@ -1186,8 +1215,6 @@ static int sched_show(struct seq_file *m, void *v) struct inode *inode = m->private; struct task_struct *p; - WARN_ON(!inode); - p = get_proc_task(inode); if (!p) return -ESRCH; @@ -1205,8 +1232,6 @@ sched_write(struct file *file, const char __user *buf, struct inode *inode = file->f_path.dentry->d_inode; struct task_struct *p; - WARN_ON(!inode); - p = get_proc_task(inode); if (!p) return -ESRCH; @@ -1974,13 +1999,11 @@ static struct dentry *proc_pident_lookup(struct inode *dir, const struct pid_entry *ents, unsigned int nents) { - struct inode *inode; struct dentry *error; struct task_struct *task = get_proc_task(dir); const struct pid_entry *p, *last; error = ERR_PTR(-ENOENT); - inode = NULL; if (!task) goto out_no_task; @@ -2136,12 +2159,12 @@ static const struct file_operations proc_pid_attr_operations = { }; static const struct pid_entry attr_dir_stuff[] = { - REG("current", S_IRUGO|S_IWUGO, pid_attr), - REG("prev", S_IRUGO, pid_attr), - REG("exec", S_IRUGO|S_IWUGO, pid_attr), - REG("fscreate", S_IRUGO|S_IWUGO, pid_attr), - REG("keycreate", S_IRUGO|S_IWUGO, pid_attr), - REG("sockcreate", S_IRUGO|S_IWUGO, pid_attr), + REG("current", S_IRUGO|S_IWUGO, proc_pid_attr_operations), + REG("prev", S_IRUGO, proc_pid_attr_operations), + REG("exec", S_IRUGO|S_IWUGO, proc_pid_attr_operations), + REG("fscreate", S_IRUGO|S_IWUGO, proc_pid_attr_operations), + REG("keycreate", S_IRUGO|S_IWUGO, proc_pid_attr_operations), + REG("sockcreate", S_IRUGO|S_IWUGO, proc_pid_attr_operations), }; static int proc_attr_dir_readdir(struct file * filp, @@ -2461,74 +2484,77 @@ static const struct file_operations proc_task_operations; static const struct inode_operations proc_task_inode_operations; static const struct pid_entry tgid_base_stuff[] = { - DIR("task", S_IRUGO|S_IXUGO, task), - DIR("fd", S_IRUSR|S_IXUSR, fd), - DIR("fdinfo", S_IRUSR|S_IXUSR, fdinfo), + DIR("task", S_IRUGO|S_IXUGO, proc_task_inode_operations, proc_task_operations), + DIR("fd", S_IRUSR|S_IXUSR, proc_fd_inode_operations, proc_fd_operations), + DIR("fdinfo", S_IRUSR|S_IXUSR, proc_fdinfo_inode_operations, proc_fdinfo_operations), #ifdef CONFIG_NET - DIR("net", S_IRUGO|S_IXUGO, net), + DIR("net", S_IRUGO|S_IXUGO, proc_net_inode_operations, proc_net_operations), #endif - REG("environ", S_IRUSR, environ), - INF("auxv", S_IRUSR, pid_auxv), - ONE("status", S_IRUGO, pid_status), - ONE("personality", S_IRUSR, pid_personality), - INF("limits", S_IRUSR, pid_limits), + REG("environ", S_IRUSR, proc_environ_operations), + INF("auxv", S_IRUSR, proc_pid_auxv), + ONE("status", S_IRUGO, proc_pid_status), + ONE("personality", S_IRUSR, proc_pid_personality), + INF("limits", S_IRUSR, proc_pid_limits), #ifdef CONFIG_SCHED_DEBUG - REG("sched", S_IRUGO|S_IWUSR, pid_sched), + REG("sched", S_IRUGO|S_IWUSR, proc_pid_sched_operations), #endif #ifdef CONFIG_HAVE_ARCH_TRACEHOOK - INF("syscall", S_IRUSR, pid_syscall), + INF("syscall", S_IRUSR, proc_pid_syscall), #endif - INF("cmdline", S_IRUGO, pid_cmdline), - ONE("stat", S_IRUGO, tgid_stat), - ONE("statm", S_IRUGO, pid_statm), - REG("maps", S_IRUGO, maps), + INF("cmdline", S_IRUGO, proc_pid_cmdline), + ONE("stat", S_IRUGO, proc_tgid_stat), + ONE("statm", S_IRUGO, proc_pid_statm), + REG("maps", S_IRUGO, proc_maps_operations), #ifdef CONFIG_NUMA - REG("numa_maps", S_IRUGO, numa_maps), + REG("numa_maps", S_IRUGO, proc_numa_maps_operations), #endif - REG("mem", S_IRUSR|S_IWUSR, mem), - LNK("cwd", cwd), - LNK("root", root), - LNK("exe", exe), - REG("mounts", S_IRUGO, mounts), - REG("mountinfo", S_IRUGO, mountinfo), - REG("mountstats", S_IRUSR, mountstats), + REG("mem", S_IRUSR|S_IWUSR, proc_mem_operations), + LNK("cwd", proc_cwd_link), + LNK("root", proc_root_link), + LNK("exe", proc_exe_link), + REG("mounts", S_IRUGO, proc_mounts_operations), + REG("mountinfo", S_IRUGO, proc_mountinfo_operations), + REG("mountstats", S_IRUSR, proc_mountstats_operations), #ifdef CONFIG_PROC_PAGE_MONITOR - REG("clear_refs", S_IWUSR, clear_refs), - REG("smaps", S_IRUGO, smaps), - REG("pagemap", S_IRUSR, pagemap), + REG("clear_refs", S_IWUSR, proc_clear_refs_operations), + REG("smaps", S_IRUGO, proc_smaps_operations), + REG("pagemap", S_IRUSR, proc_pagemap_operations), #endif #ifdef CONFIG_SECURITY - DIR("attr", S_IRUGO|S_IXUGO, attr_dir), + DIR("attr", S_IRUGO|S_IXUGO, proc_attr_dir_inode_operations, proc_attr_dir_operations), #endif #ifdef CONFIG_KALLSYMS - INF("wchan", S_IRUGO, pid_wchan), + INF("wchan", S_IRUGO, proc_pid_wchan), +#endif +#ifdef CONFIG_STACKTRACE + ONE("stack", S_IRUSR, proc_pid_stack), #endif #ifdef CONFIG_SCHEDSTATS - INF("schedstat", S_IRUGO, pid_schedstat), + INF("schedstat", S_IRUGO, proc_pid_schedstat), #endif #ifdef CONFIG_LATENCYTOP - REG("latency", S_IRUGO, lstats), + REG("latency", S_IRUGO, proc_lstats_operations), #endif #ifdef CONFIG_PROC_PID_CPUSET - REG("cpuset", S_IRUGO, cpuset), + REG("cpuset", S_IRUGO, proc_cpuset_operations), #endif #ifdef CONFIG_CGROUPS - REG("cgroup", S_IRUGO, cgroup), + REG("cgroup", S_IRUGO, proc_cgroup_operations), #endif - INF("oom_score", S_IRUGO, oom_score), - REG("oom_adj", S_IRUGO|S_IWUSR, oom_adjust), + INF("oom_score", S_IRUGO, proc_oom_score), + REG("oom_adj", S_IRUGO|S_IWUSR, proc_oom_adjust_operations), #ifdef CONFIG_AUDITSYSCALL - REG("loginuid", S_IWUSR|S_IRUGO, loginuid), - REG("sessionid", S_IRUGO, sessionid), + REG("loginuid", S_IWUSR|S_IRUGO, proc_loginuid_operations), + REG("sessionid", S_IRUGO, proc_sessionid_operations), #endif #ifdef CONFIG_FAULT_INJECTION - REG("make-it-fail", S_IRUGO|S_IWUSR, fault_inject), + REG("make-it-fail", S_IRUGO|S_IWUSR, proc_fault_inject_operations), #endif #if defined(USE_ELF_CORE_DUMP) && defined(CONFIG_ELF_CORE) - REG("coredump_filter", S_IRUGO|S_IWUSR, coredump_filter), + REG("coredump_filter", S_IRUGO|S_IWUSR, proc_coredump_filter_operations), #endif #ifdef CONFIG_TASK_IO_ACCOUNTING - INF("io", S_IRUGO, tgid_io_accounting), + INF("io", S_IRUGO, proc_tgid_io_accounting), #endif }; @@ -2801,66 +2827,69 @@ out_no_task: * Tasks */ static const struct pid_entry tid_base_stuff[] = { - DIR("fd", S_IRUSR|S_IXUSR, fd), - DIR("fdinfo", S_IRUSR|S_IXUSR, fdinfo), - REG("environ", S_IRUSR, environ), - INF("auxv", S_IRUSR, pid_auxv), - ONE("status", S_IRUGO, pid_status), - ONE("personality", S_IRUSR, pid_personality), - INF("limits", S_IRUSR, pid_limits), + DIR("fd", S_IRUSR|S_IXUSR, proc_fd_inode_operations, proc_fd_operations), + DIR("fdinfo", S_IRUSR|S_IXUSR, proc_fdinfo_inode_operations, proc_fd_operations), + REG("environ", S_IRUSR, proc_environ_operations), + INF("auxv", S_IRUSR, proc_pid_auxv), + ONE("status", S_IRUGO, proc_pid_status), + ONE("personality", S_IRUSR, proc_pid_personality), + INF("limits", S_IRUSR, proc_pid_limits), #ifdef CONFIG_SCHED_DEBUG - REG("sched", S_IRUGO|S_IWUSR, pid_sched), + REG("sched", S_IRUGO|S_IWUSR, proc_pid_sched_operations), #endif #ifdef CONFIG_HAVE_ARCH_TRACEHOOK - INF("syscall", S_IRUSR, pid_syscall), + INF("syscall", S_IRUSR, proc_pid_syscall), #endif - INF("cmdline", S_IRUGO, pid_cmdline), - ONE("stat", S_IRUGO, tid_stat), - ONE("statm", S_IRUGO, pid_statm), - REG("maps", S_IRUGO, maps), + INF("cmdline", S_IRUGO, proc_pid_cmdline), + ONE("stat", S_IRUGO, proc_tid_stat), + ONE("statm", S_IRUGO, proc_pid_statm), + REG("maps", S_IRUGO, proc_maps_operations), #ifdef CONFIG_NUMA - REG("numa_maps", S_IRUGO, numa_maps), + REG("numa_maps", S_IRUGO, proc_numa_maps_operations), #endif - REG("mem", S_IRUSR|S_IWUSR, mem), - LNK("cwd", cwd), - LNK("root", root), - LNK("exe", exe), - REG("mounts", S_IRUGO, mounts), - REG("mountinfo", S_IRUGO, mountinfo), + REG("mem", S_IRUSR|S_IWUSR, proc_mem_operations), + LNK("cwd", proc_cwd_link), + LNK("root", proc_root_link), + LNK("exe", proc_exe_link), + REG("mounts", S_IRUGO, proc_mounts_operations), + REG("mountinfo", S_IRUGO, proc_mountinfo_operations), #ifdef CONFIG_PROC_PAGE_MONITOR - REG("clear_refs", S_IWUSR, clear_refs), - REG("smaps", S_IRUGO, smaps), - REG("pagemap", S_IRUSR, pagemap), + REG("clear_refs", S_IWUSR, proc_clear_refs_operations), + REG("smaps", S_IRUGO, proc_smaps_operations), + REG("pagemap", S_IRUSR, proc_pagemap_operations), #endif #ifdef CONFIG_SECURITY - DIR("attr", S_IRUGO|S_IXUGO, attr_dir), + DIR("attr", S_IRUGO|S_IXUGO, proc_attr_dir_inode_operations, proc_attr_dir_operations), #endif #ifdef CONFIG_KALLSYMS - INF("wchan", S_IRUGO, pid_wchan), + INF("wchan", S_IRUGO, proc_pid_wchan), +#endif +#ifdef CONFIG_STACKTRACE + ONE("stack", S_IRUSR, proc_pid_stack), #endif #ifdef CONFIG_SCHEDSTATS - INF("schedstat", S_IRUGO, pid_schedstat), + INF("schedstat", S_IRUGO, proc_pid_schedstat), #endif #ifdef CONFIG_LATENCYTOP - REG("latency", S_IRUGO, lstats), + REG("latency", S_IRUGO, proc_lstats_operations), #endif #ifdef CONFIG_PROC_PID_CPUSET - REG("cpuset", S_IRUGO, cpuset), + REG("cpuset", S_IRUGO, proc_cpuset_operations), #endif #ifdef CONFIG_CGROUPS - REG("cgroup", S_IRUGO, cgroup), + REG("cgroup", S_IRUGO, proc_cgroup_operations), #endif - INF("oom_score", S_IRUGO, oom_score), - REG("oom_adj", S_IRUGO|S_IWUSR, oom_adjust), + INF("oom_score", S_IRUGO, proc_oom_score), + REG("oom_adj", S_IRUGO|S_IWUSR, proc_oom_adjust_operations), #ifdef CONFIG_AUDITSYSCALL - REG("loginuid", S_IWUSR|S_IRUGO, loginuid), - REG("sessionid", S_IRUSR, sessionid), + REG("loginuid", S_IWUSR|S_IRUGO, proc_loginuid_operations), + REG("sessionid", S_IRUSR, proc_sessionid_operations), #endif #ifdef CONFIG_FAULT_INJECTION - REG("make-it-fail", S_IRUGO|S_IWUSR, fault_inject), + REG("make-it-fail", S_IRUGO|S_IWUSR, proc_fault_inject_operations), #endif #ifdef CONFIG_TASK_IO_ACCOUNTING - INF("io", S_IRUGO, tid_io_accounting), + INF("io", S_IRUGO, proc_tid_io_accounting), #endif }; diff --git a/fs/proc/generic.c b/fs/proc/generic.c index 60a359b35582..db7fa5cab988 100644 --- a/fs/proc/generic.c +++ b/fs/proc/generic.c @@ -14,7 +14,6 @@ #include <linux/stat.h> #include <linux/module.h> #include <linux/mount.h> -#include <linux/smp_lock.h> #include <linux/init.h> #include <linux/idr.h> #include <linux/namei.h> @@ -379,7 +378,6 @@ struct dentry *proc_lookup_de(struct proc_dir_entry *de, struct inode *dir, struct inode *inode = NULL; int error = -ENOENT; - lock_kernel(); spin_lock(&proc_subdir_lock); for (de = de->subdir; de ; de = de->next) { if (de->namelen != dentry->d_name.len) @@ -397,7 +395,6 @@ struct dentry *proc_lookup_de(struct proc_dir_entry *de, struct inode *dir, } spin_unlock(&proc_subdir_lock); out_unlock: - unlock_kernel(); if (inode) { dentry->d_op = &proc_dentry_operations; @@ -432,8 +429,6 @@ int proc_readdir_de(struct proc_dir_entry *de, struct file *filp, void *dirent, struct inode *inode = filp->f_path.dentry->d_inode; int ret = 0; - lock_kernel(); - ino = inode->i_ino; i = filp->f_pos; switch (i) { @@ -487,7 +482,7 @@ int proc_readdir_de(struct proc_dir_entry *de, struct file *filp, void *dirent, spin_unlock(&proc_subdir_lock); } ret = 1; -out: unlock_kernel(); +out: return ret; } @@ -504,6 +499,7 @@ int proc_readdir(struct file *filp, void *dirent, filldir_t filldir) * the /proc directory. */ static const struct file_operations proc_dir_operations = { + .llseek = generic_file_llseek, .read = generic_read_dir, .readdir = proc_readdir, }; diff --git a/fs/proc/inode.c b/fs/proc/inode.c index 2543fd00c658..3e76bb9b3ad6 100644 --- a/fs/proc/inode.c +++ b/fs/proc/inode.c @@ -35,16 +35,13 @@ struct proc_dir_entry *de_get(struct proc_dir_entry *de) */ void de_put(struct proc_dir_entry *de) { - lock_kernel(); if (!atomic_read(&de->count)) { printk("de_put: entry %s already free!\n", de->name); - unlock_kernel(); return; } if (atomic_dec_and_test(&de->count)) free_proc_entry(de); - unlock_kernel(); } /* diff --git a/fs/proc/proc_net.c b/fs/proc/proc_net.c index 7bc296f424ae..04d1270f1c38 100644 --- a/fs/proc/proc_net.c +++ b/fs/proc/proc_net.c @@ -18,7 +18,6 @@ #include <linux/sched.h> #include <linux/module.h> #include <linux/bitops.h> -#include <linux/smp_lock.h> #include <linux/mount.h> #include <linux/nsproxy.h> #include <net/net_namespace.h> @@ -172,6 +171,7 @@ static int proc_tgid_net_readdir(struct file *filp, void *dirent, } const struct file_operations proc_net_operations = { + .llseek = generic_file_llseek, .read = generic_read_dir, .readdir = proc_tgid_net_readdir, }; diff --git a/fs/proc/root.c b/fs/proc/root.c index 7761602af9de..f6299a25594e 100644 --- a/fs/proc/root.c +++ b/fs/proc/root.c @@ -16,7 +16,6 @@ #include <linux/sched.h> #include <linux/module.h> #include <linux/bitops.h> -#include <linux/smp_lock.h> #include <linux/mount.h> #include <linux/pid_namespace.h> @@ -162,17 +161,12 @@ static int proc_root_readdir(struct file * filp, unsigned int nr = filp->f_pos; int ret; - lock_kernel(); - if (nr < FIRST_PROCESS_ENTRY) { int error = proc_readdir(filp, dirent, filldir); - if (error <= 0) { - unlock_kernel(); + if (error <= 0) return error; - } filp->f_pos = FIRST_PROCESS_ENTRY; } - unlock_kernel(); ret = proc_pid_readdir(filp, dirent, filldir); return ret; diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index 3a8bdd7f5756..94063840832a 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -396,7 +396,9 @@ static int show_smap(struct seq_file *m, void *v) "Private_Clean: %8lu kB\n" "Private_Dirty: %8lu kB\n" "Referenced: %8lu kB\n" - "Swap: %8lu kB\n", + "Swap: %8lu kB\n" + "KernelPageSize: %8lu kB\n" + "MMUPageSize: %8lu kB\n", (vma->vm_end - vma->vm_start) >> 10, mss.resident >> 10, (unsigned long)(mss.pss >> (10 + PSS_SHIFT)), @@ -405,7 +407,9 @@ static int show_smap(struct seq_file *m, void *v) mss.private_clean >> 10, mss.private_dirty >> 10, mss.referenced >> 10, - mss.swap >> 10); + mss.swap >> 10, + vma_kernel_pagesize(vma) >> 10, + vma_mmu_pagesize(vma) >> 10); if (m->count < m->size) /* vma is copied successfully */ m->version = (vma != get_gate_vma(task)) ? vma->vm_start : 0; diff --git a/fs/proc/task_nommu.c b/fs/proc/task_nommu.c index 219bd79ea894..d4a8be32b902 100644 --- a/fs/proc/task_nommu.c +++ b/fs/proc/task_nommu.c @@ -9,7 +9,7 @@ /* * Logic: we've got two memory sums for each process, "shared", and - * "non-shared". Shared memory may get counted more then once, for + * "non-shared". Shared memory may get counted more than once, for * each process that owns it. Non-shared memory is counted * accurately. */ diff --git a/fs/select.c b/fs/select.c index 87df51eadcf2..08b91beed806 100644 --- a/fs/select.c +++ b/fs/select.c @@ -109,11 +109,11 @@ static void __pollwait(struct file *filp, wait_queue_head_t *wait_address, void poll_initwait(struct poll_wqueues *pwq) { init_poll_funcptr(&pwq->pt, __pollwait); + pwq->polling_task = current; pwq->error = 0; pwq->table = NULL; pwq->inline_index = 0; } - EXPORT_SYMBOL(poll_initwait); static void free_poll_entry(struct poll_table_entry *entry) @@ -142,12 +142,10 @@ void poll_freewait(struct poll_wqueues *pwq) free_page((unsigned long) old); } } - EXPORT_SYMBOL(poll_freewait); -static struct poll_table_entry *poll_get_entry(poll_table *_p) +static struct poll_table_entry *poll_get_entry(struct poll_wqueues *p) { - struct poll_wqueues *p = container_of(_p, struct poll_wqueues, pt); struct poll_table_page *table = p->table; if (p->inline_index < N_INLINE_POLL_ENTRIES) @@ -159,7 +157,6 @@ static struct poll_table_entry *poll_get_entry(poll_table *_p) new_table = (struct poll_table_page *) __get_free_page(GFP_KERNEL); if (!new_table) { p->error = -ENOMEM; - __set_current_state(TASK_RUNNING); return NULL; } new_table->entry = new_table->entries; @@ -171,20 +168,75 @@ static struct poll_table_entry *poll_get_entry(poll_table *_p) return table->entry++; } +static int pollwake(wait_queue_t *wait, unsigned mode, int sync, void *key) +{ + struct poll_wqueues *pwq = wait->private; + DECLARE_WAITQUEUE(dummy_wait, pwq->polling_task); + + /* + * Although this function is called under waitqueue lock, LOCK + * doesn't imply write barrier and the users expect write + * barrier semantics on wakeup functions. The following + * smp_wmb() is equivalent to smp_wmb() in try_to_wake_up() + * and is paired with set_mb() in poll_schedule_timeout. + */ + smp_wmb(); + pwq->triggered = 1; + + /* + * Perform the default wake up operation using a dummy + * waitqueue. + * + * TODO: This is hacky but there currently is no interface to + * pass in @sync. @sync is scheduled to be removed and once + * that happens, wake_up_process() can be used directly. + */ + return default_wake_function(&dummy_wait, mode, sync, key); +} + /* Add a new entry */ static void __pollwait(struct file *filp, wait_queue_head_t *wait_address, poll_table *p) { - struct poll_table_entry *entry = poll_get_entry(p); + struct poll_wqueues *pwq = container_of(p, struct poll_wqueues, pt); + struct poll_table_entry *entry = poll_get_entry(pwq); if (!entry) return; get_file(filp); entry->filp = filp; entry->wait_address = wait_address; - init_waitqueue_entry(&entry->wait, current); + init_waitqueue_func_entry(&entry->wait, pollwake); + entry->wait.private = pwq; add_wait_queue(wait_address, &entry->wait); } +int poll_schedule_timeout(struct poll_wqueues *pwq, int state, + ktime_t *expires, unsigned long slack) +{ + int rc = -EINTR; + + set_current_state(state); + if (!pwq->triggered) + rc = schedule_hrtimeout_range(expires, slack, HRTIMER_MODE_ABS); + __set_current_state(TASK_RUNNING); + + /* + * Prepare for the next iteration. + * + * The following set_mb() serves two purposes. First, it's + * the counterpart rmb of the wmb in pollwake() such that data + * written before wake up is always visible after wake up. + * Second, the full barrier guarantees that triggered clearing + * doesn't pass event check of the next iteration. Note that + * this problem doesn't exist for the first iteration as + * add_wait_queue() has full barrier semantics. + */ + set_mb(pwq->triggered, 0); + + return rc; +} +EXPORT_SYMBOL(poll_schedule_timeout); + /** * poll_select_set_timeout - helper function to setup the timeout value * @to: pointer to timespec variable for the final timeout @@ -340,8 +392,6 @@ int do_select(int n, fd_set_bits *fds, struct timespec *end_time) for (;;) { unsigned long *rinp, *routp, *rexp, *inp, *outp, *exp; - set_current_state(TASK_INTERRUPTIBLE); - inp = fds->in; outp = fds->out; exp = fds->ex; rinp = fds->res_in; routp = fds->res_out; rexp = fds->res_ex; @@ -411,10 +461,10 @@ int do_select(int n, fd_set_bits *fds, struct timespec *end_time) to = &expire; } - if (!schedule_hrtimeout_range(to, slack, HRTIMER_MODE_ABS)) + if (!poll_schedule_timeout(&table, TASK_INTERRUPTIBLE, + to, slack)) timed_out = 1; } - __set_current_state(TASK_RUNNING); poll_freewait(&table); @@ -666,7 +716,6 @@ static int do_poll(unsigned int nfds, struct poll_list *list, for (;;) { struct poll_list *walk; - set_current_state(TASK_INTERRUPTIBLE); for (walk = list; walk != NULL; walk = walk->next) { struct pollfd * pfd, * pfd_end; @@ -709,10 +758,9 @@ static int do_poll(unsigned int nfds, struct poll_list *list, to = &expire; } - if (!schedule_hrtimeout_range(to, slack, HRTIMER_MODE_ABS)) + if (!poll_schedule_timeout(wait, TASK_INTERRUPTIBLE, to, slack)) timed_out = 1; } - __set_current_state(TASK_RUNNING); return count; } diff --git a/fs/super.c b/fs/super.c index ddba069d7a99..cb20744ec789 100644 --- a/fs/super.c +++ b/fs/super.c @@ -38,6 +38,7 @@ #include <linux/kobject.h> #include <linux/mutex.h> #include <linux/file.h> +#include <linux/async.h> #include <asm/uaccess.h> #include "internal.h" @@ -71,6 +72,7 @@ static struct super_block *alloc_super(struct file_system_type *type) INIT_HLIST_HEAD(&s->s_anon); INIT_LIST_HEAD(&s->s_inodes); INIT_LIST_HEAD(&s->s_dentry_lru); + INIT_LIST_HEAD(&s->s_async_list); init_rwsem(&s->s_umount); mutex_init(&s->s_lock); lockdep_set_class(&s->s_umount, &type->s_umount_key); @@ -289,11 +291,18 @@ void generic_shutdown_super(struct super_block *sb) { const struct super_operations *sop = sb->s_op; + if (sb->s_root) { shrink_dcache_for_umount(sb); fsync_super(sb); lock_super(sb); sb->s_flags &= ~MS_ACTIVE; + + /* + * wait for asynchronous fs operations to finish before going further + */ + async_synchronize_full_special(&sb->s_async_list); + /* bad name - it should be evict_inodes() */ invalidate_inodes(sb); lock_kernel(); @@ -449,6 +458,7 @@ void sync_filesystems(int wait) if (sb->s_flags & MS_RDONLY) continue; sb->s_need_sync_fs = 1; + async_synchronize_full_special(&sb->s_async_list); } restart: diff --git a/fs/sync.c b/fs/sync.c index 0921d6d4b5e6..ac02b56548bc 100644 --- a/fs/sync.c +++ b/fs/sync.c @@ -295,7 +295,7 @@ int do_sync_mapping_range(struct address_space *mapping, loff_t offset, if (flags & SYNC_FILE_RANGE_WRITE) { ret = __filemap_fdatawrite_range(mapping, offset, endbyte, - WB_SYNC_NONE); + WB_SYNC_ALL); if (ret < 0) goto out; } diff --git a/fs/ubifs/Kconfig b/fs/ubifs/Kconfig index 91ceeda7e5bf..e35b54d5059d 100644 --- a/fs/ubifs/Kconfig +++ b/fs/ubifs/Kconfig @@ -40,7 +40,7 @@ config UBIFS_FS_ZLIB depends on UBIFS_FS default y help - Zlib copresses better then LZO but it is slower. Say 'Y' if unsure. + Zlib compresses better than LZO but it is slower. Say 'Y' if unsure. # Debugging-related stuff config UBIFS_FS_DEBUG diff --git a/fs/ubifs/budget.c b/fs/ubifs/budget.c index 0e5e54d82924..175f9c590b77 100644 --- a/fs/ubifs/budget.c +++ b/fs/ubifs/budget.c @@ -142,7 +142,7 @@ static long long get_liability(struct ubifs_info *c) * * This function is called when an operation cannot be budgeted because there * is supposedly no free space. But in most cases there is some free space: - * o budgeting is pessimistic, so it always budgets more then it is actually + * o budgeting is pessimistic, so it always budgets more than it is actually * needed, so shrinking the liability is one way to make free space - the * cached data will take less space then it was budgeted for; * o GC may turn some dark space into free space (budgeting treats dark space @@ -606,7 +606,7 @@ void ubifs_release_budget(struct ubifs_info *c, struct ubifs_budget_req *req) * @c: UBIFS file-system description object * * This function converts budget which was allocated for a new page of data to - * the budget of changing an existing page of data. The latter is smaller then + * the budget of changing an existing page of data. The latter is smaller than * the former, so this function only does simple re-calculation and does not * involve any write-back. */ diff --git a/fs/ubifs/gc.c b/fs/ubifs/gc.c index 0bef6501d58a..9832f9abe28e 100644 --- a/fs/ubifs/gc.c +++ b/fs/ubifs/gc.c @@ -45,7 +45,7 @@ #define SMALL_NODE_WM UBIFS_MAX_DENT_NODE_SZ /* - * GC may need to move more then one LEB to make progress. The below constants + * GC may need to move more than one LEB to make progress. The below constants * define "soft" and "hard" limits on the number of LEBs the garbage collector * may move. */ diff --git a/fs/ubifs/journal.c b/fs/ubifs/journal.c index 10ae25b7d1db..9b7c54e0cd2a 100644 --- a/fs/ubifs/journal.c +++ b/fs/ubifs/journal.c @@ -191,7 +191,7 @@ again: if (wbuf->lnum != -1 && avail >= len) { /* * Someone else has switched the journal head and we have - * enough space now. This happens when more then one process is + * enough space now. This happens when more than one process is * trying to write to the same journal head at the same time. */ dbg_jnl("return LEB %d back, already have LEB %d:%d", diff --git a/fs/ubifs/shrinker.c b/fs/ubifs/shrinker.c index f248533841a2..e7bab52a1410 100644 --- a/fs/ubifs/shrinker.c +++ b/fs/ubifs/shrinker.c @@ -151,7 +151,7 @@ static int shrink_tnc(struct ubifs_info *c, int nr, int age, int *contention) * @contention: if any contention, this is set to %1 * * This function walks the list of mounted UBIFS file-systems and frees clean - * znodes which are older then @age, until at least @nr znodes are freed. + * znodes which are older than @age, until at least @nr znodes are freed. * Returns the number of freed znodes. */ static int shrink_tnc_trees(int nr, int age, int *contention) diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c index 0d7564b95f8e..89556ee72518 100644 --- a/fs/ubifs/super.c +++ b/fs/ubifs/super.c @@ -432,12 +432,19 @@ static int ubifs_sync_fs(struct super_block *sb, int wait) int i, err; struct ubifs_info *c = sb->s_fs_info; struct writeback_control wbc = { - .sync_mode = wait ? WB_SYNC_ALL : WB_SYNC_HOLD, + .sync_mode = wait ? WB_SYNC_ALL : WB_SYNC_NONE, .range_start = 0, .range_end = LLONG_MAX, .nr_to_write = LONG_MAX, }; + /* + * Note by akpm about WB_SYNC_NONE used above: zero @wait is just an + * advisory thing to help the file system shove lots of data into the + * queues. If some gets missed then it'll be picked up on the second + * '->sync_fs()' call, with non-zero @wait. + */ + if (sb->s_flags & MS_RDONLY) return 0; diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c index 36f6cc703ef2..be846d606ae8 100644 --- a/fs/xfs/linux-2.6/xfs_super.c +++ b/fs/xfs/linux-2.6/xfs_super.c @@ -1348,7 +1348,7 @@ xfs_finish_flags( { int ronly = (mp->m_flags & XFS_MOUNT_RDONLY); - /* Fail a mount where the logbuf is smaller then the log stripe */ + /* Fail a mount where the logbuf is smaller than the log stripe */ if (xfs_sb_version_haslogv2(&mp->m_sb)) { if (mp->m_logbsize <= 0 && mp->m_sb.sb_logsunit > XLOG_BIG_RECORD_BSIZE) { |