diff options
Diffstat (limited to 'fs')
592 files changed, 18752 insertions, 11265 deletions
diff --git a/fs/9p/acl.c b/fs/9p/acl.c index 6261719f6f2a..bb1b286c49ae 100644 --- a/fs/9p/acl.c +++ b/fs/9p/acl.c @@ -239,6 +239,7 @@ static int v9fs_xattr_get_acl(const struct xattr_handler *handler, } static int v9fs_xattr_set_acl(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *dentry, struct inode *inode, const char *name, const void *value, size_t size, int flags) @@ -258,7 +259,7 @@ static int v9fs_xattr_set_acl(const struct xattr_handler *handler, if (S_ISLNK(inode->i_mode)) return -EOPNOTSUPP; - if (!inode_owner_or_capable(inode)) + if (!inode_owner_or_capable(&init_user_ns, inode)) return -EPERM; if (value) { /* update the cached acl value */ @@ -279,7 +280,8 @@ static int v9fs_xattr_set_acl(const struct xattr_handler *handler, struct iattr iattr = { 0 }; struct posix_acl *old_acl = acl; - retval = posix_acl_update_mode(inode, &iattr.ia_mode, &acl); + retval = posix_acl_update_mode(&init_user_ns, inode, + &iattr.ia_mode, &acl); if (retval) goto err_out; if (!acl) { @@ -297,7 +299,7 @@ static int v9fs_xattr_set_acl(const struct xattr_handler *handler, * What is the following setxattr update the * mode ? */ - v9fs_vfs_setattr_dotl(dentry, &iattr); + v9fs_vfs_setattr_dotl(&init_user_ns, dentry, &iattr); } break; case ACL_TYPE_DEFAULT: diff --git a/fs/9p/v9fs.h b/fs/9p/v9fs.h index 7b763776306e..4ca56c5dd637 100644 --- a/fs/9p/v9fs.h +++ b/fs/9p/v9fs.h @@ -135,7 +135,8 @@ extern struct dentry *v9fs_vfs_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags); extern int v9fs_vfs_unlink(struct inode *i, struct dentry *d); extern int v9fs_vfs_rmdir(struct inode *i, struct dentry *d); -extern int v9fs_vfs_rename(struct inode *old_dir, struct dentry *old_dentry, +extern int v9fs_vfs_rename(struct user_namespace *mnt_userns, + struct inode *old_dir, struct dentry *old_dentry, struct inode *new_dir, struct dentry *new_dentry, unsigned int flags); extern struct inode *v9fs_inode_from_fid(struct v9fs_session_info *v9ses, diff --git a/fs/9p/v9fs_vfs.h b/fs/9p/v9fs_vfs.h index fd2a2b040250..d44ade76966a 100644 --- a/fs/9p/v9fs_vfs.h +++ b/fs/9p/v9fs_vfs.h @@ -59,7 +59,8 @@ void v9fs_inode2stat(struct inode *inode, struct p9_wstat *stat); int v9fs_uflags2omode(int uflags, int extended); void v9fs_blank_wstat(struct p9_wstat *wstat); -int v9fs_vfs_setattr_dotl(struct dentry *, struct iattr *); +int v9fs_vfs_setattr_dotl(struct user_namespace *, struct dentry *, + struct iattr *); int v9fs_file_fsync_dotl(struct file *filp, loff_t start, loff_t end, int datasync); int v9fs_refresh_inode(struct p9_fid *fid, struct inode *inode); diff --git a/fs/9p/vfs_inode.c b/fs/9p/vfs_inode.c index 4a937fac1acb..8d97f0b45e9c 100644 --- a/fs/9p/vfs_inode.c +++ b/fs/9p/vfs_inode.c @@ -251,7 +251,7 @@ int v9fs_init_inode(struct v9fs_session_info *v9ses, { int err = 0; - inode_init_owner(inode, NULL, mode); + inode_init_owner(&init_user_ns,inode, NULL, mode); inode->i_blocks = 0; inode->i_rdev = rdev; inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode); @@ -676,8 +676,8 @@ error: */ static int -v9fs_vfs_create(struct inode *dir, struct dentry *dentry, umode_t mode, - bool excl) +v9fs_vfs_create(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, bool excl) { struct v9fs_session_info *v9ses = v9fs_inode2v9ses(dir); u32 perm = unixmode2p9mode(v9ses, mode); @@ -702,7 +702,8 @@ v9fs_vfs_create(struct inode *dir, struct dentry *dentry, umode_t mode, * */ -static int v9fs_vfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) +static int v9fs_vfs_mkdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { int err; u32 perm; @@ -907,9 +908,9 @@ int v9fs_vfs_rmdir(struct inode *i, struct dentry *d) */ int -v9fs_vfs_rename(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry, - unsigned int flags) +v9fs_vfs_rename(struct user_namespace *mnt_userns, struct inode *old_dir, + struct dentry *old_dentry, struct inode *new_dir, + struct dentry *new_dentry, unsigned int flags) { int retval; struct inode *old_inode; @@ -1016,8 +1017,8 @@ done: */ static int -v9fs_vfs_getattr(const struct path *path, struct kstat *stat, - u32 request_mask, unsigned int flags) +v9fs_vfs_getattr(struct user_namespace *mnt_userns, const struct path *path, + struct kstat *stat, u32 request_mask, unsigned int flags) { struct dentry *dentry = path->dentry; struct v9fs_session_info *v9ses; @@ -1027,7 +1028,7 @@ v9fs_vfs_getattr(const struct path *path, struct kstat *stat, p9_debug(P9_DEBUG_VFS, "dentry: %p\n", dentry); v9ses = v9fs_dentry2v9ses(dentry); if (v9ses->cache == CACHE_LOOSE || v9ses->cache == CACHE_FSCACHE) { - generic_fillattr(d_inode(dentry), stat); + generic_fillattr(&init_user_ns, d_inode(dentry), stat); return 0; } fid = v9fs_fid_lookup(dentry); @@ -1040,7 +1041,7 @@ v9fs_vfs_getattr(const struct path *path, struct kstat *stat, return PTR_ERR(st); v9fs_stat2inode(st, d_inode(dentry), dentry->d_sb, 0); - generic_fillattr(d_inode(dentry), stat); + generic_fillattr(&init_user_ns, d_inode(dentry), stat); p9stat_free(st); kfree(st); @@ -1054,7 +1055,8 @@ v9fs_vfs_getattr(const struct path *path, struct kstat *stat, * */ -static int v9fs_vfs_setattr(struct dentry *dentry, struct iattr *iattr) +static int v9fs_vfs_setattr(struct user_namespace *mnt_userns, + struct dentry *dentry, struct iattr *iattr) { int retval, use_dentry = 0; struct v9fs_session_info *v9ses; @@ -1062,7 +1064,7 @@ static int v9fs_vfs_setattr(struct dentry *dentry, struct iattr *iattr) struct p9_wstat wstat; p9_debug(P9_DEBUG_VFS, "\n"); - retval = setattr_prepare(dentry, iattr); + retval = setattr_prepare(&init_user_ns, dentry, iattr); if (retval) return retval; @@ -1118,7 +1120,7 @@ static int v9fs_vfs_setattr(struct dentry *dentry, struct iattr *iattr) v9fs_invalidate_inode_attr(d_inode(dentry)); - setattr_copy(d_inode(dentry), iattr); + setattr_copy(&init_user_ns, d_inode(dentry), iattr); mark_inode_dirty(d_inode(dentry)); return 0; } @@ -1137,9 +1139,6 @@ v9fs_stat2inode(struct p9_wstat *stat, struct inode *inode, struct super_block *sb, unsigned int flags) { umode_t mode; - char ext[32]; - char tag_name[14]; - unsigned int i_nlink; struct v9fs_session_info *v9ses = sb->s_fs_info; struct v9fs_inode *v9inode = V9FS_I(inode); @@ -1157,18 +1156,18 @@ v9fs_stat2inode(struct p9_wstat *stat, struct inode *inode, inode->i_gid = stat->n_gid; } if ((S_ISREG(inode->i_mode)) || (S_ISDIR(inode->i_mode))) { - if (v9fs_proto_dotu(v9ses) && (stat->extension[0] != '\0')) { + if (v9fs_proto_dotu(v9ses)) { + unsigned int i_nlink; /* - * Hadlink support got added later to - * to the .u extension. So there can be - * server out there that doesn't support - * this even with .u extension. So check - * for non NULL stat->extension + * Hadlink support got added later to the .u extension. + * So there can be a server out there that doesn't + * support this even with .u extension. That would + * just leave us with stat->extension being an empty + * string, though. */ - strlcpy(ext, stat->extension, sizeof(ext)); /* HARDLINKCOUNT %u */ - sscanf(ext, "%13s %u", tag_name, &i_nlink); - if (!strncmp(tag_name, "HARDLINKCOUNT", 13)) + if (sscanf(stat->extension, + " HARDLINKCOUNT %u", &i_nlink) == 1) set_nlink(inode, i_nlink); } } @@ -1295,7 +1294,8 @@ static int v9fs_vfs_mkspecial(struct inode *dir, struct dentry *dentry, */ static int -v9fs_vfs_symlink(struct inode *dir, struct dentry *dentry, const char *symname) +v9fs_vfs_symlink(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, const char *symname) { p9_debug(P9_DEBUG_VFS, " %lu,%pd,%s\n", dir->i_ino, dentry, symname); @@ -1348,7 +1348,8 @@ v9fs_vfs_link(struct dentry *old_dentry, struct inode *dir, */ static int -v9fs_vfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t rdev) +v9fs_vfs_mknod(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, dev_t rdev) { struct v9fs_session_info *v9ses = v9fs_inode2v9ses(dir); int retval; diff --git a/fs/9p/vfs_inode_dotl.c b/fs/9p/vfs_inode_dotl.c index 823c2eb5f1bf..1dc7af046615 100644 --- a/fs/9p/vfs_inode_dotl.c +++ b/fs/9p/vfs_inode_dotl.c @@ -33,8 +33,8 @@ #include "acl.h" static int -v9fs_vfs_mknod_dotl(struct inode *dir, struct dentry *dentry, umode_t omode, - dev_t rdev); +v9fs_vfs_mknod_dotl(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t omode, dev_t rdev); /** * v9fs_get_fsgid_for_create - Helper function to get the gid for creating a @@ -218,10 +218,10 @@ int v9fs_open_to_dotl_flags(int flags) */ static int -v9fs_vfs_create_dotl(struct inode *dir, struct dentry *dentry, umode_t omode, - bool excl) +v9fs_vfs_create_dotl(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t omode, bool excl) { - return v9fs_vfs_mknod_dotl(dir, dentry, omode, 0); + return v9fs_vfs_mknod_dotl(mnt_userns, dir, dentry, omode, 0); } static int @@ -367,8 +367,9 @@ err_clunk_old_fid: * */ -static int v9fs_vfs_mkdir_dotl(struct inode *dir, - struct dentry *dentry, umode_t omode) +static int v9fs_vfs_mkdir_dotl(struct user_namespace *mnt_userns, + struct inode *dir, struct dentry *dentry, + umode_t omode) { int err; struct v9fs_session_info *v9ses; @@ -457,8 +458,9 @@ error: } static int -v9fs_vfs_getattr_dotl(const struct path *path, struct kstat *stat, - u32 request_mask, unsigned int flags) +v9fs_vfs_getattr_dotl(struct user_namespace *mnt_userns, + const struct path *path, struct kstat *stat, + u32 request_mask, unsigned int flags) { struct dentry *dentry = path->dentry; struct v9fs_session_info *v9ses; @@ -468,7 +470,7 @@ v9fs_vfs_getattr_dotl(const struct path *path, struct kstat *stat, p9_debug(P9_DEBUG_VFS, "dentry: %p\n", dentry); v9ses = v9fs_dentry2v9ses(dentry); if (v9ses->cache == CACHE_LOOSE || v9ses->cache == CACHE_FSCACHE) { - generic_fillattr(d_inode(dentry), stat); + generic_fillattr(&init_user_ns, d_inode(dentry), stat); return 0; } fid = v9fs_fid_lookup(dentry); @@ -485,7 +487,7 @@ v9fs_vfs_getattr_dotl(const struct path *path, struct kstat *stat, return PTR_ERR(st); v9fs_stat2inode_dotl(st, d_inode(dentry), 0); - generic_fillattr(d_inode(dentry), stat); + generic_fillattr(&init_user_ns, d_inode(dentry), stat); /* Change block size to what the server returned */ stat->blksize = st->st_blksize; @@ -540,7 +542,8 @@ static int v9fs_mapped_iattr_valid(int iattr_valid) * */ -int v9fs_vfs_setattr_dotl(struct dentry *dentry, struct iattr *iattr) +int v9fs_vfs_setattr_dotl(struct user_namespace *mnt_userns, + struct dentry *dentry, struct iattr *iattr) { int retval, use_dentry = 0; struct p9_fid *fid = NULL; @@ -549,7 +552,7 @@ int v9fs_vfs_setattr_dotl(struct dentry *dentry, struct iattr *iattr) p9_debug(P9_DEBUG_VFS, "\n"); - retval = setattr_prepare(dentry, iattr); + retval = setattr_prepare(&init_user_ns, dentry, iattr); if (retval) return retval; @@ -590,7 +593,7 @@ int v9fs_vfs_setattr_dotl(struct dentry *dentry, struct iattr *iattr) truncate_setsize(inode, iattr->ia_size); v9fs_invalidate_inode_attr(inode); - setattr_copy(inode, iattr); + setattr_copy(&init_user_ns, inode, iattr); mark_inode_dirty(inode); if (iattr->ia_valid & ATTR_MODE) { /* We also want to update ACL when we update mode bits */ @@ -684,8 +687,8 @@ v9fs_stat2inode_dotl(struct p9_stat_dotl *stat, struct inode *inode, } static int -v9fs_vfs_symlink_dotl(struct inode *dir, struct dentry *dentry, - const char *symname) +v9fs_vfs_symlink_dotl(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, const char *symname) { int err; kgid_t gid; @@ -824,8 +827,8 @@ v9fs_vfs_link_dotl(struct dentry *old_dentry, struct inode *dir, * */ static int -v9fs_vfs_mknod_dotl(struct inode *dir, struct dentry *dentry, umode_t omode, - dev_t rdev) +v9fs_vfs_mknod_dotl(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t omode, dev_t rdev) { int err; kgid_t gid; diff --git a/fs/9p/xattr.c b/fs/9p/xattr.c index 87217dd0433e..ee331845e2c7 100644 --- a/fs/9p/xattr.c +++ b/fs/9p/xattr.c @@ -157,6 +157,7 @@ static int v9fs_xattr_handler_get(const struct xattr_handler *handler, } static int v9fs_xattr_handler_set(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *dentry, struct inode *inode, const char *name, const void *value, size_t size, int flags) diff --git a/fs/Kconfig b/fs/Kconfig index da524c4d7b7e..a55bda4233bb 100644 --- a/fs/Kconfig +++ b/fs/Kconfig @@ -203,7 +203,7 @@ config TMPFS_XATTR config TMPFS_INODE64 bool "Use 64-bit ino_t by default in tmpfs" - depends on TMPFS && 64BIT && !(S390 || ALPHA) + depends on TMPFS && 64BIT default n help tmpfs has historically used only inode numbers as wide as an unsigned @@ -333,6 +333,10 @@ config NFS_COMMON depends on NFSD || NFS_FS || LOCKD default y +config NFS_V4_2_SSC_HELPER + tristate + default y if NFS_V4=y || NFS_FS=y + source "net/sunrpc/Kconfig" source "fs/ceph/Kconfig" source "fs/cifs/Kconfig" diff --git a/fs/Kconfig.binfmt b/fs/Kconfig.binfmt index 885da6d983b4..c6f1c8c1934e 100644 --- a/fs/Kconfig.binfmt +++ b/fs/Kconfig.binfmt @@ -29,7 +29,7 @@ config BINFMT_ELF latest version). config COMPAT_BINFMT_ELF - bool + def_bool y depends on COMPAT && BINFMT_ELF select ELFCORE @@ -45,7 +45,7 @@ config ARCH_USE_GNU_PROPERTY config BINFMT_ELF_FDPIC bool "Kernel support for FDPIC ELF binaries" default y if !BINFMT_ELF - depends on (ARM || (SUPERH && !MMU) || C6X) + depends on (ARM || (SUPERH && !MMU)) select ELFCORE help ELF FDPIC binaries are based on ELF, but allow the individual load diff --git a/fs/Makefile b/fs/Makefile index 999d1a23f036..3215fe205256 100644 --- a/fs/Makefile +++ b/fs/Makefile @@ -64,7 +64,6 @@ obj-$(CONFIG_SYSFS) += sysfs/ obj-$(CONFIG_CONFIGFS_FS) += configfs/ obj-y += devpts/ -obj-$(CONFIG_PROFILING) += dcookies.o obj-$(CONFIG_DLM) += dlm/ # Do not add any filesystems before this line diff --git a/fs/adfs/adfs.h b/fs/adfs/adfs.h index 699c4fa8b78b..06b7c92343ad 100644 --- a/fs/adfs/adfs.h +++ b/fs/adfs/adfs.h @@ -144,7 +144,8 @@ struct adfs_discmap { /* Inode stuff */ struct inode *adfs_iget(struct super_block *sb, struct object_info *obj); int adfs_write_inode(struct inode *inode, struct writeback_control *wbc); -int adfs_notify_change(struct dentry *dentry, struct iattr *attr); +int adfs_notify_change(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *attr); /* map.c */ int adfs_map_lookup(struct super_block *sb, u32 frag_id, unsigned int offset); diff --git a/fs/adfs/inode.c b/fs/adfs/inode.c index 32620f4a7623..fb7ee026d101 100644 --- a/fs/adfs/inode.c +++ b/fs/adfs/inode.c @@ -292,14 +292,15 @@ out: * later. */ int -adfs_notify_change(struct dentry *dentry, struct iattr *attr) +adfs_notify_change(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *attr) { struct inode *inode = d_inode(dentry); struct super_block *sb = inode->i_sb; unsigned int ia_valid = attr->ia_valid; int error; - error = setattr_prepare(dentry, attr); + error = setattr_prepare(&init_user_ns, dentry, attr); /* * we can't change the UID or GID of any file - diff --git a/fs/affs/affs.h b/fs/affs/affs.h index a755bef7c4c7..bfa89e131ead 100644 --- a/fs/affs/affs.h +++ b/fs/affs/affs.h @@ -167,27 +167,33 @@ extern const struct export_operations affs_export_ops; extern int affs_hash_name(struct super_block *sb, const u8 *name, unsigned int len); extern struct dentry *affs_lookup(struct inode *dir, struct dentry *dentry, unsigned int); extern int affs_unlink(struct inode *dir, struct dentry *dentry); -extern int affs_create(struct inode *dir, struct dentry *dentry, umode_t mode, bool); -extern int affs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode); +extern int affs_create(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, bool); +extern int affs_mkdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode); extern int affs_rmdir(struct inode *dir, struct dentry *dentry); extern int affs_link(struct dentry *olddentry, struct inode *dir, struct dentry *dentry); -extern int affs_symlink(struct inode *dir, struct dentry *dentry, - const char *symname); -extern int affs_rename2(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry, - unsigned int flags); +extern int affs_symlink(struct user_namespace *mnt_userns, + struct inode *dir, struct dentry *dentry, + const char *symname); +extern int affs_rename2(struct user_namespace *mnt_userns, + struct inode *old_dir, struct dentry *old_dentry, + struct inode *new_dir, struct dentry *new_dentry, + unsigned int flags); /* inode.c */ extern struct inode *affs_new_inode(struct inode *dir); -extern int affs_notify_change(struct dentry *dentry, struct iattr *attr); +extern int affs_notify_change(struct user_namespace *mnt_userns, + struct dentry *dentry, struct iattr *attr); extern void affs_evict_inode(struct inode *inode); extern struct inode *affs_iget(struct super_block *sb, unsigned long ino); extern int affs_write_inode(struct inode *inode, struct writeback_control *wbc); -extern int affs_add_entry(struct inode *dir, struct inode *inode, struct dentry *dentry, s32 type); +extern int affs_add_entry(struct inode *dir, struct inode *inode, + struct dentry *dentry, s32 type); /* file.c */ diff --git a/fs/affs/inode.c b/fs/affs/inode.c index 044412110b52..2352a75bd9d6 100644 --- a/fs/affs/inode.c +++ b/fs/affs/inode.c @@ -216,14 +216,15 @@ affs_write_inode(struct inode *inode, struct writeback_control *wbc) } int -affs_notify_change(struct dentry *dentry, struct iattr *attr) +affs_notify_change(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *attr) { struct inode *inode = d_inode(dentry); int error; pr_debug("notify_change(%lu,0x%x)\n", inode->i_ino, attr->ia_valid); - error = setattr_prepare(dentry, attr); + error = setattr_prepare(&init_user_ns, dentry, attr); if (error) goto out; @@ -249,7 +250,7 @@ affs_notify_change(struct dentry *dentry, struct iattr *attr) affs_truncate(inode); } - setattr_copy(inode, attr); + setattr_copy(&init_user_ns, inode, attr); mark_inode_dirty(inode); if (attr->ia_valid & ATTR_MODE) diff --git a/fs/affs/namei.c b/fs/affs/namei.c index 41c5749f4db7..bcab18956b4f 100644 --- a/fs/affs/namei.c +++ b/fs/affs/namei.c @@ -242,7 +242,8 @@ affs_unlink(struct inode *dir, struct dentry *dentry) } int -affs_create(struct inode *dir, struct dentry *dentry, umode_t mode, bool excl) +affs_create(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, bool excl) { struct super_block *sb = dir->i_sb; struct inode *inode; @@ -273,7 +274,8 @@ affs_create(struct inode *dir, struct dentry *dentry, umode_t mode, bool excl) } int -affs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) +affs_mkdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { struct inode *inode; int error; @@ -311,7 +313,8 @@ affs_rmdir(struct inode *dir, struct dentry *dentry) } int -affs_symlink(struct inode *dir, struct dentry *dentry, const char *symname) +affs_symlink(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, const char *symname) { struct super_block *sb = dir->i_sb; struct buffer_head *bh; @@ -460,8 +463,10 @@ affs_xrename(struct inode *old_dir, struct dentry *old_dentry, return -EIO; bh_new = affs_bread(sb, d_inode(new_dentry)->i_ino); - if (!bh_new) + if (!bh_new) { + affs_brelse(bh_old); return -EIO; + } /* Remove old header from its parent directory. */ affs_lock_dir(old_dir); @@ -498,9 +503,9 @@ done: return retval; } -int affs_rename2(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry, - unsigned int flags) +int affs_rename2(struct user_namespace *mnt_userns, struct inode *old_dir, + struct dentry *old_dentry, struct inode *new_dir, + struct dentry *new_dentry, unsigned int flags) { if (flags & ~(RENAME_NOREPLACE | RENAME_EXCHANGE)) diff --git a/fs/afs/dir.c b/fs/afs/dir.c index 7bd659ad959e..17548c1faf02 100644 --- a/fs/afs/dir.c +++ b/fs/afs/dir.c @@ -28,18 +28,19 @@ static int afs_lookup_one_filldir(struct dir_context *ctx, const char *name, int loff_t fpos, u64 ino, unsigned dtype); static int afs_lookup_filldir(struct dir_context *ctx, const char *name, int nlen, loff_t fpos, u64 ino, unsigned dtype); -static int afs_create(struct inode *dir, struct dentry *dentry, umode_t mode, - bool excl); -static int afs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode); +static int afs_create(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, bool excl); +static int afs_mkdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode); static int afs_rmdir(struct inode *dir, struct dentry *dentry); static int afs_unlink(struct inode *dir, struct dentry *dentry); static int afs_link(struct dentry *from, struct inode *dir, struct dentry *dentry); -static int afs_symlink(struct inode *dir, struct dentry *dentry, - const char *content); -static int afs_rename(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry, - unsigned int flags); +static int afs_symlink(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, const char *content); +static int afs_rename(struct user_namespace *mnt_userns, struct inode *old_dir, + struct dentry *old_dentry, struct inode *new_dir, + struct dentry *new_dentry, unsigned int flags); static int afs_dir_releasepage(struct page *page, gfp_t gfp_flags); static void afs_dir_invalidatepage(struct page *page, unsigned int offset, unsigned int length); @@ -69,7 +70,6 @@ const struct inode_operations afs_dir_inode_operations = { .permission = afs_permission, .getattr = afs_getattr, .setattr = afs_setattr, - .listxattr = afs_listxattr, }; const struct address_space_operations afs_dir_aops = { @@ -1325,7 +1325,8 @@ static const struct afs_operation_ops afs_mkdir_operation = { /* * create a directory on an AFS filesystem */ -static int afs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) +static int afs_mkdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { struct afs_operation *op; struct afs_vnode *dvnode = AFS_FS_I(dir); @@ -1619,8 +1620,8 @@ static const struct afs_operation_ops afs_create_operation = { /* * create a regular file on an AFS filesystem */ -static int afs_create(struct inode *dir, struct dentry *dentry, umode_t mode, - bool excl) +static int afs_create(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, bool excl) { struct afs_operation *op; struct afs_vnode *dvnode = AFS_FS_I(dir); @@ -1741,8 +1742,8 @@ static const struct afs_operation_ops afs_symlink_operation = { /* * create a symlink in an AFS filesystem */ -static int afs_symlink(struct inode *dir, struct dentry *dentry, - const char *content) +static int afs_symlink(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, const char *content) { struct afs_operation *op; struct afs_vnode *dvnode = AFS_FS_I(dir); @@ -1876,9 +1877,9 @@ static const struct afs_operation_ops afs_rename_operation = { /* * rename a file in an AFS filesystem and/or move it between directories */ -static int afs_rename(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry, - unsigned int flags) +static int afs_rename(struct user_namespace *mnt_userns, struct inode *old_dir, + struct dentry *old_dentry, struct inode *new_dir, + struct dentry *new_dentry, unsigned int flags) { struct afs_operation *op; struct afs_vnode *orig_dvnode, *new_dvnode, *vnode; diff --git a/fs/afs/file.c b/fs/afs/file.c index 85f5adf21aa0..960b64268623 100644 --- a/fs/afs/file.c +++ b/fs/afs/file.c @@ -43,7 +43,6 @@ const struct inode_operations afs_file_inode_operations = { .getattr = afs_getattr, .setattr = afs_setattr, .permission = afs_permission, - .listxattr = afs_listxattr, }; const struct address_space_operations afs_fs_aops = { diff --git a/fs/afs/fs_operation.c b/fs/afs/fs_operation.c index 97cab12b0a6c..71c58723763d 100644 --- a/fs/afs/fs_operation.c +++ b/fs/afs/fs_operation.c @@ -181,10 +181,13 @@ void afs_wait_for_operation(struct afs_operation *op) if (test_bit(AFS_SERVER_FL_IS_YFS, &op->server->flags) && op->ops->issue_yfs_rpc) op->ops->issue_yfs_rpc(op); - else + else if (op->ops->issue_afs_rpc) op->ops->issue_afs_rpc(op); + else + op->ac.error = -ENOTSUPP; - op->error = afs_wait_for_call_to_complete(op->call, &op->ac); + if (op->call) + op->error = afs_wait_for_call_to_complete(op->call, &op->ac); } switch (op->error) { diff --git a/fs/afs/inode.c b/fs/afs/inode.c index b0d7b892090d..12be88716e4c 100644 --- a/fs/afs/inode.c +++ b/fs/afs/inode.c @@ -27,7 +27,6 @@ static const struct inode_operations afs_symlink_inode_operations = { .get_link = page_get_link, - .listxattr = afs_listxattr, }; static noinline void dump_vnode(struct afs_vnode *vnode, struct afs_vnode *parent_vnode) @@ -734,8 +733,8 @@ error_unlock: /* * read the attributes of an inode */ -int afs_getattr(const struct path *path, struct kstat *stat, - u32 request_mask, unsigned int query_flags) +int afs_getattr(struct user_namespace *mnt_userns, const struct path *path, + struct kstat *stat, u32 request_mask, unsigned int query_flags) { struct inode *inode = d_inode(path->dentry); struct afs_vnode *vnode = AFS_FS_I(inode); @@ -745,7 +744,7 @@ int afs_getattr(const struct path *path, struct kstat *stat, do { read_seqbegin_or_lock(&vnode->cb_lock, &seq); - generic_fillattr(inode, stat); + generic_fillattr(&init_user_ns, inode, stat); if (test_bit(AFS_VNODE_SILLY_DELETED, &vnode->flags) && stat->nlink > 0) stat->nlink -= 1; @@ -857,7 +856,8 @@ static const struct afs_operation_ops afs_setattr_operation = { /* * set the attributes of an inode */ -int afs_setattr(struct dentry *dentry, struct iattr *attr) +int afs_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *attr) { struct afs_operation *op; struct afs_vnode *vnode = AFS_FS_I(d_inode(dentry)); diff --git a/fs/afs/internal.h b/fs/afs/internal.h index 0d150a29e39e..1627b1872812 100644 --- a/fs/afs/internal.h +++ b/fs/afs/internal.h @@ -1149,8 +1149,9 @@ extern struct inode *afs_iget(struct afs_operation *, struct afs_vnode_param *); extern struct inode *afs_root_iget(struct super_block *, struct key *); extern bool afs_check_validity(struct afs_vnode *); extern int afs_validate(struct afs_vnode *, struct key *); -extern int afs_getattr(const struct path *, struct kstat *, u32, unsigned int); -extern int afs_setattr(struct dentry *, struct iattr *); +extern int afs_getattr(struct user_namespace *mnt_userns, const struct path *, + struct kstat *, u32, unsigned int); +extern int afs_setattr(struct user_namespace *mnt_userns, struct dentry *, struct iattr *); extern void afs_evict_inode(struct inode *); extern int afs_drop_inode(struct inode *); @@ -1361,7 +1362,7 @@ extern void afs_zap_permits(struct rcu_head *); extern struct key *afs_request_key(struct afs_cell *); extern struct key *afs_request_key_rcu(struct afs_cell *); extern int afs_check_permit(struct afs_vnode *, struct key *, afs_access_t *); -extern int afs_permission(struct inode *, int); +extern int afs_permission(struct user_namespace *, struct inode *, int); extern void __exit afs_clean_up_permit_cache(void); /* @@ -1508,7 +1509,6 @@ extern int afs_launder_page(struct page *); * xattr.c */ extern const struct xattr_handler *afs_xattr_handlers[]; -extern ssize_t afs_listxattr(struct dentry *, char *, size_t); /* * yfsclient.c diff --git a/fs/afs/mntpt.c b/fs/afs/mntpt.c index 052dab2f5c03..bbb2c210d139 100644 --- a/fs/afs/mntpt.c +++ b/fs/afs/mntpt.c @@ -32,7 +32,6 @@ const struct inode_operations afs_mntpt_inode_operations = { .lookup = afs_mntpt_lookup, .readlink = page_readlink, .getattr = afs_getattr, - .listxattr = afs_listxattr, }; const struct inode_operations afs_autocell_inode_operations = { diff --git a/fs/afs/security.c b/fs/afs/security.c index 9cf3102f370c..3c7a8fc4f93f 100644 --- a/fs/afs/security.c +++ b/fs/afs/security.c @@ -396,7 +396,8 @@ int afs_check_permit(struct afs_vnode *vnode, struct key *key, * - AFS ACLs are attached to directories only, and a file is controlled by its * parent directory's ACL */ -int afs_permission(struct inode *inode, int mask) +int afs_permission(struct user_namespace *mnt_userns, struct inode *inode, + int mask) { struct afs_vnode *vnode = AFS_FS_I(inode); afs_access_t access; diff --git a/fs/afs/xattr.c b/fs/afs/xattr.c index 95c573dcda11..7751b0b3f81d 100644 --- a/fs/afs/xattr.c +++ b/fs/afs/xattr.c @@ -11,29 +11,6 @@ #include <linux/xattr.h> #include "internal.h" -static const char afs_xattr_list[] = - "afs.acl\0" - "afs.cell\0" - "afs.fid\0" - "afs.volume\0" - "afs.yfs.acl\0" - "afs.yfs.acl_inherited\0" - "afs.yfs.acl_num_cleaned\0" - "afs.yfs.vol_acl"; - -/* - * Retrieve a list of the supported xattrs. - */ -ssize_t afs_listxattr(struct dentry *dentry, char *buffer, size_t size) -{ - if (size == 0) - return sizeof(afs_xattr_list); - if (size < sizeof(afs_xattr_list)) - return -ERANGE; - memcpy(buffer, afs_xattr_list, sizeof(afs_xattr_list)); - return sizeof(afs_xattr_list); -} - /* * Deal with the result of a successful fetch ACL operation. */ @@ -120,6 +97,7 @@ static const struct afs_operation_ops afs_store_acl_operation = { * Set a file's AFS3 ACL. */ static int afs_xattr_set_acl(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *dentry, struct inode *inode, const char *name, const void *buffer, size_t size, int flags) @@ -230,6 +208,8 @@ static int afs_xattr_get_yfs(const struct xattr_handler *handler, else ret = -ERANGE; } + } else if (ret == -ENOTSUPP) { + ret = -ENODATA; } error_yacl: @@ -248,12 +228,14 @@ static const struct afs_operation_ops yfs_store_opaque_acl2_operation = { * Set a file's YFS ACL. */ static int afs_xattr_set_yfs(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *dentry, struct inode *inode, const char *name, const void *buffer, size_t size, int flags) { struct afs_operation *op; struct afs_vnode *vnode = AFS_FS_I(inode); + int ret; if (flags == XATTR_CREATE || strcmp(name, "acl") != 0) @@ -268,7 +250,10 @@ static int afs_xattr_set_yfs(const struct xattr_handler *handler, return afs_put_operation(op); op->ops = &yfs_store_opaque_acl2_operation; - return afs_do_sync_operation(op); + ret = afs_do_sync_operation(op); + if (ret == -ENOTSUPP) + ret = -ENODATA; + return ret; } static const struct xattr_handler afs_xattr_yfs_handler = { diff --git a/fs/anon_inodes.c b/fs/anon_inodes.c index 89714308c25b..a280156138ed 100644 --- a/fs/anon_inodes.c +++ b/fs/anon_inodes.c @@ -55,61 +55,79 @@ static struct file_system_type anon_inode_fs_type = { .kill_sb = kill_anon_super, }; -/** - * anon_inode_getfile - creates a new file instance by hooking it up to an - * anonymous inode, and a dentry that describe the "class" - * of the file - * - * @name: [in] name of the "class" of the new file - * @fops: [in] file operations for the new file - * @priv: [in] private data for the new file (will be file's private_data) - * @flags: [in] flags - * - * Creates a new file by hooking it on a single inode. This is useful for files - * that do not need to have a full-fledged inode in order to operate correctly. - * All the files created with anon_inode_getfile() will share a single inode, - * hence saving memory and avoiding code duplication for the file/inode/dentry - * setup. Returns the newly created file* or an error pointer. - */ -struct file *anon_inode_getfile(const char *name, - const struct file_operations *fops, - void *priv, int flags) +static struct inode *anon_inode_make_secure_inode( + const char *name, + const struct inode *context_inode) { - struct file *file; + struct inode *inode; + const struct qstr qname = QSTR_INIT(name, strlen(name)); + int error; + + inode = alloc_anon_inode(anon_inode_mnt->mnt_sb); + if (IS_ERR(inode)) + return inode; + inode->i_flags &= ~S_PRIVATE; + error = security_inode_init_security_anon(inode, &qname, context_inode); + if (error) { + iput(inode); + return ERR_PTR(error); + } + return inode; +} - if (IS_ERR(anon_inode_inode)) - return ERR_PTR(-ENODEV); +static struct file *__anon_inode_getfile(const char *name, + const struct file_operations *fops, + void *priv, int flags, + const struct inode *context_inode, + bool secure) +{ + struct inode *inode; + struct file *file; if (fops->owner && !try_module_get(fops->owner)) return ERR_PTR(-ENOENT); - /* - * We know the anon_inode inode count is always greater than zero, - * so ihold() is safe. - */ - ihold(anon_inode_inode); - file = alloc_file_pseudo(anon_inode_inode, anon_inode_mnt, name, + if (secure) { + inode = anon_inode_make_secure_inode(name, context_inode); + if (IS_ERR(inode)) { + file = ERR_CAST(inode); + goto err; + } + } else { + inode = anon_inode_inode; + if (IS_ERR(inode)) { + file = ERR_PTR(-ENODEV); + goto err; + } + /* + * We know the anon_inode inode count is always + * greater than zero, so ihold() is safe. + */ + ihold(inode); + } + + file = alloc_file_pseudo(inode, anon_inode_mnt, name, flags & (O_ACCMODE | O_NONBLOCK), fops); if (IS_ERR(file)) - goto err; + goto err_iput; - file->f_mapping = anon_inode_inode->i_mapping; + file->f_mapping = inode->i_mapping; file->private_data = priv; return file; +err_iput: + iput(inode); err: - iput(anon_inode_inode); module_put(fops->owner); return file; } -EXPORT_SYMBOL_GPL(anon_inode_getfile); /** - * anon_inode_getfd - creates a new file instance by hooking it up to an - * anonymous inode, and a dentry that describe the "class" - * of the file + * anon_inode_getfile - creates a new file instance by hooking it up to an + * anonymous inode, and a dentry that describe the "class" + * of the file * * @name: [in] name of the "class" of the new file * @fops: [in] file operations for the new file @@ -118,12 +136,23 @@ EXPORT_SYMBOL_GPL(anon_inode_getfile); * * Creates a new file by hooking it on a single inode. This is useful for files * that do not need to have a full-fledged inode in order to operate correctly. - * All the files created with anon_inode_getfd() will share a single inode, + * All the files created with anon_inode_getfile() will share a single inode, * hence saving memory and avoiding code duplication for the file/inode/dentry - * setup. Returns new descriptor or an error code. + * setup. Returns the newly created file* or an error pointer. */ -int anon_inode_getfd(const char *name, const struct file_operations *fops, - void *priv, int flags) +struct file *anon_inode_getfile(const char *name, + const struct file_operations *fops, + void *priv, int flags) +{ + return __anon_inode_getfile(name, fops, priv, flags, NULL, false); +} +EXPORT_SYMBOL_GPL(anon_inode_getfile); + +static int __anon_inode_getfd(const char *name, + const struct file_operations *fops, + void *priv, int flags, + const struct inode *context_inode, + bool secure) { int error, fd; struct file *file; @@ -133,7 +162,8 @@ int anon_inode_getfd(const char *name, const struct file_operations *fops, return error; fd = error; - file = anon_inode_getfile(name, fops, priv, flags); + file = __anon_inode_getfile(name, fops, priv, flags, context_inode, + secure); if (IS_ERR(file)) { error = PTR_ERR(file); goto err_put_unused_fd; @@ -146,8 +176,55 @@ err_put_unused_fd: put_unused_fd(fd); return error; } + +/** + * anon_inode_getfd - creates a new file instance by hooking it up to + * an anonymous inode and a dentry that describe + * the "class" of the file + * + * @name: [in] name of the "class" of the new file + * @fops: [in] file operations for the new file + * @priv: [in] private data for the new file (will be file's private_data) + * @flags: [in] flags + * + * Creates a new file by hooking it on a single inode. This is + * useful for files that do not need to have a full-fledged inode in + * order to operate correctly. All the files created with + * anon_inode_getfd() will use the same singleton inode, reducing + * memory use and avoiding code duplication for the file/inode/dentry + * setup. Returns a newly created file descriptor or an error code. + */ +int anon_inode_getfd(const char *name, const struct file_operations *fops, + void *priv, int flags) +{ + return __anon_inode_getfd(name, fops, priv, flags, NULL, false); +} EXPORT_SYMBOL_GPL(anon_inode_getfd); +/** + * anon_inode_getfd_secure - Like anon_inode_getfd(), but creates a new + * !S_PRIVATE anon inode rather than reuse the singleton anon inode, and calls + * the inode_init_security_anon() LSM hook. This allows the inode to have its + * own security context and for a LSM to reject creation of the inode. + * + * @name: [in] name of the "class" of the new file + * @fops: [in] file operations for the new file + * @priv: [in] private data for the new file (will be file's private_data) + * @flags: [in] flags + * @context_inode: + * [in] the logical relationship with the new inode (optional) + * + * The LSM may use @context_inode in inode_init_security_anon(), but a + * reference to it is not held. + */ +int anon_inode_getfd_secure(const char *name, const struct file_operations *fops, + void *priv, int flags, + const struct inode *context_inode) +{ + return __anon_inode_getfd(name, fops, priv, flags, context_inode, true); +} +EXPORT_SYMBOL_GPL(anon_inode_getfd_secure); + static int __init anon_inode_init(void) { anon_inode_mnt = kern_mount(&anon_inode_fs_type); diff --git a/fs/attr.c b/fs/attr.c index b4bbdbd4c8ca..87ef39db1c34 100644 --- a/fs/attr.c +++ b/fs/attr.c @@ -18,27 +18,55 @@ #include <linux/evm.h> #include <linux/ima.h> -static bool chown_ok(const struct inode *inode, kuid_t uid) +/** + * chown_ok - verify permissions to chown inode + * @mnt_userns: user namespace of the mount @inode was found from + * @inode: inode to check permissions on + * @uid: uid to chown @inode to + * + * If the inode has been found through an idmapped mount the user namespace of + * the vfsmount must be passed through @mnt_userns. This function will then + * take care to map the inode according to @mnt_userns before checking + * permissions. On non-idmapped mounts or if permission checking is to be + * performed on the raw inode simply passs init_user_ns. + */ +static bool chown_ok(struct user_namespace *mnt_userns, + const struct inode *inode, + kuid_t uid) { - if (uid_eq(current_fsuid(), inode->i_uid) && - uid_eq(uid, inode->i_uid)) + kuid_t kuid = i_uid_into_mnt(mnt_userns, inode); + if (uid_eq(current_fsuid(), kuid) && uid_eq(uid, kuid)) return true; - if (capable_wrt_inode_uidgid(inode, CAP_CHOWN)) + if (capable_wrt_inode_uidgid(mnt_userns, inode, CAP_CHOWN)) return true; - if (uid_eq(inode->i_uid, INVALID_UID) && + if (uid_eq(kuid, INVALID_UID) && ns_capable(inode->i_sb->s_user_ns, CAP_CHOWN)) return true; return false; } -static bool chgrp_ok(const struct inode *inode, kgid_t gid) +/** + * chgrp_ok - verify permissions to chgrp inode + * @mnt_userns: user namespace of the mount @inode was found from + * @inode: inode to check permissions on + * @gid: gid to chown @inode to + * + * If the inode has been found through an idmapped mount the user namespace of + * the vfsmount must be passed through @mnt_userns. This function will then + * take care to map the inode according to @mnt_userns before checking + * permissions. On non-idmapped mounts or if permission checking is to be + * performed on the raw inode simply passs init_user_ns. + */ +static bool chgrp_ok(struct user_namespace *mnt_userns, + const struct inode *inode, kgid_t gid) { - if (uid_eq(current_fsuid(), inode->i_uid) && - (in_group_p(gid) || gid_eq(gid, inode->i_gid))) + kgid_t kgid = i_gid_into_mnt(mnt_userns, inode); + if (uid_eq(current_fsuid(), i_uid_into_mnt(mnt_userns, inode)) && + (in_group_p(gid) || gid_eq(gid, kgid))) return true; - if (capable_wrt_inode_uidgid(inode, CAP_CHOWN)) + if (capable_wrt_inode_uidgid(mnt_userns, inode, CAP_CHOWN)) return true; - if (gid_eq(inode->i_gid, INVALID_GID) && + if (gid_eq(kgid, INVALID_GID) && ns_capable(inode->i_sb->s_user_ns, CAP_CHOWN)) return true; return false; @@ -46,6 +74,7 @@ static bool chgrp_ok(const struct inode *inode, kgid_t gid) /** * setattr_prepare - check if attribute changes to a dentry are allowed + * @mnt_userns: user namespace of the mount the inode was found from * @dentry: dentry to check * @attr: attributes to change * @@ -55,10 +84,17 @@ static bool chgrp_ok(const struct inode *inode, kgid_t gid) * SGID bit from mode if user is not allowed to set it. Also file capabilities * and IMA extended attributes are cleared if ATTR_KILL_PRIV is set. * + * If the inode has been found through an idmapped mount the user namespace of + * the vfsmount must be passed through @mnt_userns. This function will then + * take care to map the inode according to @mnt_userns before checking + * permissions. On non-idmapped mounts or if permission checking is to be + * performed on the raw inode simply passs init_user_ns. + * * Should be called as the first thing in ->setattr implementations, * possibly after taking additional locks. */ -int setattr_prepare(struct dentry *dentry, struct iattr *attr) +int setattr_prepare(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *attr) { struct inode *inode = d_inode(dentry); unsigned int ia_valid = attr->ia_valid; @@ -78,27 +114,27 @@ int setattr_prepare(struct dentry *dentry, struct iattr *attr) goto kill_priv; /* Make sure a caller can chown. */ - if ((ia_valid & ATTR_UID) && !chown_ok(inode, attr->ia_uid)) + if ((ia_valid & ATTR_UID) && !chown_ok(mnt_userns, inode, attr->ia_uid)) return -EPERM; /* Make sure caller can chgrp. */ - if ((ia_valid & ATTR_GID) && !chgrp_ok(inode, attr->ia_gid)) + if ((ia_valid & ATTR_GID) && !chgrp_ok(mnt_userns, inode, attr->ia_gid)) return -EPERM; /* Make sure a caller can chmod. */ if (ia_valid & ATTR_MODE) { - if (!inode_owner_or_capable(inode)) + if (!inode_owner_or_capable(mnt_userns, inode)) return -EPERM; /* Also check the setgid bit! */ - if (!in_group_p((ia_valid & ATTR_GID) ? attr->ia_gid : - inode->i_gid) && - !capable_wrt_inode_uidgid(inode, CAP_FSETID)) + if (!in_group_p((ia_valid & ATTR_GID) ? attr->ia_gid : + i_gid_into_mnt(mnt_userns, inode)) && + !capable_wrt_inode_uidgid(mnt_userns, inode, CAP_FSETID)) attr->ia_mode &= ~S_ISGID; } /* Check for setting the inode time. */ if (ia_valid & (ATTR_MTIME_SET | ATTR_ATIME_SET | ATTR_TIMES_SET)) { - if (!inode_owner_or_capable(inode)) + if (!inode_owner_or_capable(mnt_userns, inode)) return -EPERM; } @@ -107,7 +143,7 @@ kill_priv: if (ia_valid & ATTR_KILL_PRIV) { int error; - error = security_inode_killpriv(dentry); + error = security_inode_killpriv(mnt_userns, dentry); if (error) return error; } @@ -162,20 +198,33 @@ EXPORT_SYMBOL(inode_newsize_ok); /** * setattr_copy - copy simple metadata updates into the generic inode + * @mnt_userns: user namespace of the mount the inode was found from * @inode: the inode to be updated * @attr: the new attributes * * setattr_copy must be called with i_mutex held. * * setattr_copy updates the inode's metadata with that specified - * in attr. Noticeably missing is inode size update, which is more complex + * in attr on idmapped mounts. If file ownership is changed setattr_copy + * doesn't map ia_uid and ia_gid. It will asssume the caller has already + * provided the intended values. Necessary permission checks to determine + * whether or not the S_ISGID property needs to be removed are performed with + * the correct idmapped mount permission helpers. + * Noticeably missing is inode size update, which is more complex * as it requires pagecache updates. * + * If the inode has been found through an idmapped mount the user namespace of + * the vfsmount must be passed through @mnt_userns. This function will then + * take care to map the inode according to @mnt_userns before checking + * permissions. On non-idmapped mounts or if permission checking is to be + * performed on the raw inode simply passs init_user_ns. + * * The inode is not marked as dirty after this operation. The rationale is * that for "simple" filesystems, the struct inode is the inode storage. * The caller is free to mark the inode dirty afterwards if needed. */ -void setattr_copy(struct inode *inode, const struct iattr *attr) +void setattr_copy(struct user_namespace *mnt_userns, struct inode *inode, + const struct iattr *attr) { unsigned int ia_valid = attr->ia_valid; @@ -191,9 +240,9 @@ void setattr_copy(struct inode *inode, const struct iattr *attr) inode->i_ctime = attr->ia_ctime; if (ia_valid & ATTR_MODE) { umode_t mode = attr->ia_mode; - - if (!in_group_p(inode->i_gid) && - !capable_wrt_inode_uidgid(inode, CAP_FSETID)) + kgid_t kgid = i_gid_into_mnt(mnt_userns, inode); + if (!in_group_p(kgid) && + !capable_wrt_inode_uidgid(mnt_userns, inode, CAP_FSETID)) mode &= ~S_ISGID; inode->i_mode = mode; } @@ -202,6 +251,7 @@ EXPORT_SYMBOL(setattr_copy); /** * notify_change - modify attributes of a filesytem object + * @mnt_userns: user namespace of the mount the inode was found from * @dentry: object affected * @attr: new attributes * @delegated_inode: returns inode, if the inode is delegated @@ -214,13 +264,23 @@ EXPORT_SYMBOL(setattr_copy); * retry. Because breaking a delegation may take a long time, the * caller should drop the i_mutex before doing so. * + * If file ownership is changed notify_change() doesn't map ia_uid and + * ia_gid. It will asssume the caller has already provided the intended values. + * * Alternatively, a caller may pass NULL for delegated_inode. This may * be appropriate for callers that expect the underlying filesystem not * to be NFS exported. Also, passing NULL is fine for callers holding * the file open for write, as there can be no conflicting delegation in * that case. + * + * If the inode has been found through an idmapped mount the user namespace of + * the vfsmount must be passed through @mnt_userns. This function will then + * take care to map the inode according to @mnt_userns before checking + * permissions. On non-idmapped mounts or if permission checking is to be + * performed on the raw inode simply passs init_user_ns. */ -int notify_change(struct dentry * dentry, struct iattr * attr, struct inode **delegated_inode) +int notify_change(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *attr, struct inode **delegated_inode) { struct inode *inode = dentry->d_inode; umode_t mode = inode->i_mode; @@ -243,8 +303,8 @@ int notify_change(struct dentry * dentry, struct iattr * attr, struct inode **de if (IS_IMMUTABLE(inode)) return -EPERM; - if (!inode_owner_or_capable(inode)) { - error = inode_permission(inode, MAY_WRITE); + if (!inode_owner_or_capable(mnt_userns, inode)) { + error = inode_permission(mnt_userns, inode, MAY_WRITE); if (error) return error; } @@ -320,9 +380,11 @@ int notify_change(struct dentry * dentry, struct iattr * attr, struct inode **de /* Don't allow modifications of files with invalid uids or * gids unless those uids & gids are being made valid. */ - if (!(ia_valid & ATTR_UID) && !uid_valid(inode->i_uid)) + if (!(ia_valid & ATTR_UID) && + !uid_valid(i_uid_into_mnt(mnt_userns, inode))) return -EOVERFLOW; - if (!(ia_valid & ATTR_GID) && !gid_valid(inode->i_gid)) + if (!(ia_valid & ATTR_GID) && + !gid_valid(i_gid_into_mnt(mnt_userns, inode))) return -EOVERFLOW; error = security_inode_setattr(dentry, attr); @@ -333,13 +395,13 @@ int notify_change(struct dentry * dentry, struct iattr * attr, struct inode **de return error; if (inode->i_op->setattr) - error = inode->i_op->setattr(dentry, attr); + error = inode->i_op->setattr(mnt_userns, dentry, attr); else - error = simple_setattr(dentry, attr); + error = simple_setattr(mnt_userns, dentry, attr); if (!error) { fsnotify_change(dentry, ia_valid); - ima_inode_post_setattr(dentry); + ima_inode_post_setattr(mnt_userns, dentry); evm_inode_post_setattr(dentry, ia_valid); } diff --git a/fs/autofs/root.c b/fs/autofs/root.c index 5aaa1732bf1e..91fe4548c256 100644 --- a/fs/autofs/root.c +++ b/fs/autofs/root.c @@ -10,10 +10,12 @@ #include "autofs_i.h" -static int autofs_dir_symlink(struct inode *, struct dentry *, const char *); +static int autofs_dir_symlink(struct user_namespace *, struct inode *, + struct dentry *, const char *); static int autofs_dir_unlink(struct inode *, struct dentry *); static int autofs_dir_rmdir(struct inode *, struct dentry *); -static int autofs_dir_mkdir(struct inode *, struct dentry *, umode_t); +static int autofs_dir_mkdir(struct user_namespace *, struct inode *, + struct dentry *, umode_t); static long autofs_root_ioctl(struct file *, unsigned int, unsigned long); #ifdef CONFIG_COMPAT static long autofs_root_compat_ioctl(struct file *, @@ -524,9 +526,9 @@ static struct dentry *autofs_lookup(struct inode *dir, return NULL; } -static int autofs_dir_symlink(struct inode *dir, - struct dentry *dentry, - const char *symname) +static int autofs_dir_symlink(struct user_namespace *mnt_userns, + struct inode *dir, struct dentry *dentry, + const char *symname) { struct autofs_sb_info *sbi = autofs_sbi(dir->i_sb); struct autofs_info *ino = autofs_dentry_ino(dentry); @@ -715,8 +717,9 @@ static int autofs_dir_rmdir(struct inode *dir, struct dentry *dentry) return 0; } -static int autofs_dir_mkdir(struct inode *dir, - struct dentry *dentry, umode_t mode) +static int autofs_dir_mkdir(struct user_namespace *mnt_userns, + struct inode *dir, struct dentry *dentry, + umode_t mode) { struct autofs_sb_info *sbi = autofs_sbi(dir->i_sb); struct autofs_info *ino = autofs_dentry_ino(dentry); diff --git a/fs/bad_inode.c b/fs/bad_inode.c index 54f0ce444272..48e16144c1f7 100644 --- a/fs/bad_inode.c +++ b/fs/bad_inode.c @@ -27,8 +27,9 @@ static const struct file_operations bad_file_ops = .open = bad_file_open, }; -static int bad_inode_create (struct inode *dir, struct dentry *dentry, - umode_t mode, bool excl) +static int bad_inode_create(struct user_namespace *mnt_userns, + struct inode *dir, struct dentry *dentry, + umode_t mode, bool excl) { return -EIO; } @@ -50,14 +51,15 @@ static int bad_inode_unlink(struct inode *dir, struct dentry *dentry) return -EIO; } -static int bad_inode_symlink (struct inode *dir, struct dentry *dentry, - const char *symname) +static int bad_inode_symlink(struct user_namespace *mnt_userns, + struct inode *dir, struct dentry *dentry, + const char *symname) { return -EIO; } -static int bad_inode_mkdir(struct inode *dir, struct dentry *dentry, - umode_t mode) +static int bad_inode_mkdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { return -EIO; } @@ -67,13 +69,14 @@ static int bad_inode_rmdir (struct inode *dir, struct dentry *dentry) return -EIO; } -static int bad_inode_mknod (struct inode *dir, struct dentry *dentry, - umode_t mode, dev_t rdev) +static int bad_inode_mknod(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, dev_t rdev) { return -EIO; } -static int bad_inode_rename2(struct inode *old_dir, struct dentry *old_dentry, +static int bad_inode_rename2(struct user_namespace *mnt_userns, + struct inode *old_dir, struct dentry *old_dentry, struct inode *new_dir, struct dentry *new_dentry, unsigned int flags) { @@ -86,18 +89,21 @@ static int bad_inode_readlink(struct dentry *dentry, char __user *buffer, return -EIO; } -static int bad_inode_permission(struct inode *inode, int mask) +static int bad_inode_permission(struct user_namespace *mnt_userns, + struct inode *inode, int mask) { return -EIO; } -static int bad_inode_getattr(const struct path *path, struct kstat *stat, +static int bad_inode_getattr(struct user_namespace *mnt_userns, + const struct path *path, struct kstat *stat, u32 request_mask, unsigned int query_flags) { return -EIO; } -static int bad_inode_setattr(struct dentry *direntry, struct iattr *attrs) +static int bad_inode_setattr(struct user_namespace *mnt_userns, + struct dentry *direntry, struct iattr *attrs) { return -EIO; } @@ -140,13 +146,15 @@ static int bad_inode_atomic_open(struct inode *inode, struct dentry *dentry, return -EIO; } -static int bad_inode_tmpfile(struct inode *inode, struct dentry *dentry, +static int bad_inode_tmpfile(struct user_namespace *mnt_userns, + struct inode *inode, struct dentry *dentry, umode_t mode) { return -EIO; } -static int bad_inode_set_acl(struct inode *inode, struct posix_acl *acl, +static int bad_inode_set_acl(struct user_namespace *mnt_userns, + struct inode *inode, struct posix_acl *acl, int type) { return -EIO; diff --git a/fs/bfs/dir.c b/fs/bfs/dir.c index d8dfe3a0cb39..34d4f68f786b 100644 --- a/fs/bfs/dir.c +++ b/fs/bfs/dir.c @@ -75,8 +75,8 @@ const struct file_operations bfs_dir_operations = { .llseek = generic_file_llseek, }; -static int bfs_create(struct inode *dir, struct dentry *dentry, umode_t mode, - bool excl) +static int bfs_create(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, bool excl) { int err; struct inode *inode; @@ -96,7 +96,7 @@ static int bfs_create(struct inode *dir, struct dentry *dentry, umode_t mode, } set_bit(ino, info->si_imap); info->si_freei--; - inode_init_owner(inode, dir, mode); + inode_init_owner(&init_user_ns, inode, dir, mode); inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode); inode->i_blocks = 0; inode->i_op = &bfs_file_inops; @@ -199,9 +199,9 @@ out_brelse: return error; } -static int bfs_rename(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry, - unsigned int flags) +static int bfs_rename(struct user_namespace *mnt_userns, struct inode *old_dir, + struct dentry *old_dentry, struct inode *new_dir, + struct dentry *new_dentry, unsigned int flags) { struct inode *old_inode, *new_inode; struct buffer_head *old_bh, *new_bh; diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index 950bc177238a..b12ba98ae9f5 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -186,6 +186,7 @@ create_elf_tables(struct linux_binprm *bprm, const struct elfhdr *exec, unsigned char k_rand_bytes[16]; int items; elf_addr_t *elf_info; + elf_addr_t flags = 0; int ei_index; const struct cred *cred = current_cred(); struct vm_area_struct *vma; @@ -260,7 +261,9 @@ create_elf_tables(struct linux_binprm *bprm, const struct elfhdr *exec, NEW_AUX_ENT(AT_PHENT, sizeof(struct elf_phdr)); NEW_AUX_ENT(AT_PHNUM, exec->e_phnum); NEW_AUX_ENT(AT_BASE, interp_load_addr); - NEW_AUX_ENT(AT_FLAGS, 0); + if (bprm->interp_flags & BINPRM_FLAGS_PRESERVE_ARGV0) + flags |= AT_FLAGS_PRESERVE_ARGV0; + NEW_AUX_ENT(AT_FLAGS, flags); NEW_AUX_ENT(AT_ENTRY, e_entry); NEW_AUX_ENT(AT_UID, from_kuid_munged(cred->user_ns, cred->uid)); NEW_AUX_ENT(AT_EUID, from_kuid_munged(cred->user_ns, cred->euid)); @@ -1495,7 +1498,7 @@ static void fill_note(struct memelfnote *note, const char *name, int type, * fill up all the fields in prstatus from the given task struct, except * registers which need to be filled up separately. */ -static void fill_prstatus(struct elf_prstatus *prstatus, +static void fill_prstatus(struct elf_prstatus_common *prstatus, struct task_struct *p, long signr) { prstatus->pr_info.si_signo = prstatus->pr_cursig = signr; @@ -1717,11 +1720,11 @@ static void do_thread_regset_writeback(struct task_struct *task, } #ifndef PRSTATUS_SIZE -#define PRSTATUS_SIZE(S, R) sizeof(S) +#define PRSTATUS_SIZE sizeof(struct elf_prstatus) #endif #ifndef SET_PR_FPVALID -#define SET_PR_FPVALID(S, V, R) ((S)->pr_fpvalid = (V)) +#define SET_PR_FPVALID(S) ((S)->pr_fpvalid = 1) #endif static int fill_thread_core_info(struct elf_thread_core_info *t, @@ -1729,7 +1732,6 @@ static int fill_thread_core_info(struct elf_thread_core_info *t, long signr, size_t *total) { unsigned int i; - int regset0_size; /* * NT_PRSTATUS is the one special case, because the regset data @@ -1737,14 +1739,12 @@ static int fill_thread_core_info(struct elf_thread_core_info *t, * than being the whole note contents. We fill the reset in here. * We assume that regset 0 is NT_PRSTATUS. */ - fill_prstatus(&t->prstatus, t->task, signr); - regset0_size = regset_get(t->task, &view->regsets[0], + fill_prstatus(&t->prstatus.common, t->task, signr); + regset_get(t->task, &view->regsets[0], sizeof(t->prstatus.pr_reg), &t->prstatus.pr_reg); - if (regset0_size < 0) - return 0; fill_note(&t->notes[0], "CORE", NT_PRSTATUS, - PRSTATUS_SIZE(t->prstatus, regset0_size), &t->prstatus); + PRSTATUS_SIZE, &t->prstatus); *total += notesize(&t->notes[0]); do_thread_regset_writeback(t->task, &view->regsets[0]); @@ -1772,7 +1772,7 @@ static int fill_thread_core_info(struct elf_thread_core_info *t, continue; if (is_fpreg) - SET_PR_FPVALID(&t->prstatus, 1, regset0_size); + SET_PR_FPVALID(&t->prstatus); fill_note(&t->notes[i], is_fpreg ? "CORE" : "LINUX", note_type, ret, data); @@ -1961,7 +1961,7 @@ static int elf_dump_thread_status(long signr, struct elf_thread_status *t) struct task_struct *p = t->thread; t->num_notes = 0; - fill_prstatus(&t->prstatus, p, signr); + fill_prstatus(&t->prstatus.common, p, signr); elf_core_copy_task_regs(p, &t->prstatus.pr_reg); fill_note(&t->notes[0], "CORE", NT_PRSTATUS, sizeof(t->prstatus), @@ -2040,7 +2040,7 @@ static int fill_note_info(struct elfhdr *elf, int phdrs, } /* now collect the dump for the current */ memset(info->prstatus, 0, sizeof(*info->prstatus)); - fill_prstatus(info->prstatus, current, siginfo->si_signo); + fill_prstatus(&info->prstatus->common, current, siginfo->si_signo); elf_core_copy_regs(&info->prstatus->pr_reg, regs); /* Set up header */ diff --git a/fs/binfmt_elf_fdpic.c b/fs/binfmt_elf_fdpic.c index be4062b8ba75..3cfd6cd46f26 100644 --- a/fs/binfmt_elf_fdpic.c +++ b/fs/binfmt_elf_fdpic.c @@ -506,6 +506,7 @@ static int create_elf_fdpic_tables(struct linux_binprm *bprm, char __user *u_platform, *u_base_platform, *p; int loop; int nr; /* reset for each csp adjustment */ + unsigned long flags = 0; #ifdef CONFIG_MMU /* In some cases (e.g. Hyper-Threading), we want to avoid L1 evictions @@ -648,7 +649,9 @@ static int create_elf_fdpic_tables(struct linux_binprm *bprm, NEW_AUX_ENT(AT_PHENT, sizeof(struct elf_phdr)); NEW_AUX_ENT(AT_PHNUM, exec_params->hdr.e_phnum); NEW_AUX_ENT(AT_BASE, interp_params->elfhdr_addr); - NEW_AUX_ENT(AT_FLAGS, 0); + if (bprm->interp_flags & BINPRM_FLAGS_PRESERVE_ARGV0) + flags |= AT_FLAGS_PRESERVE_ARGV0; + NEW_AUX_ENT(AT_FLAGS, flags); NEW_AUX_ENT(AT_ENTRY, exec_params->entry_addr); NEW_AUX_ENT(AT_UID, (elf_addr_t) from_kuid_munged(cred->user_ns, cred->uid)); NEW_AUX_ENT(AT_EUID, (elf_addr_t) from_kuid_munged(cred->user_ns, cred->euid)); @@ -1191,18 +1194,7 @@ static int elf_fdpic_map_file_by_direct_mmap(struct elf_fdpic_params *params, struct elf_prstatus_fdpic { - struct elf_siginfo pr_info; /* Info associated with signal */ - short pr_cursig; /* Current signal */ - unsigned long pr_sigpend; /* Set of pending signals */ - unsigned long pr_sighold; /* Set of held signals */ - pid_t pr_pid; - pid_t pr_ppid; - pid_t pr_pgrp; - pid_t pr_sid; - struct __kernel_old_timeval pr_utime; /* User time */ - struct __kernel_old_timeval pr_stime; /* System time */ - struct __kernel_old_timeval pr_cutime; /* Cumulative user time */ - struct __kernel_old_timeval pr_cstime; /* Cumulative system time */ + struct elf_prstatus_common common; elf_gregset_t pr_reg; /* GP registers */ /* When using FDPIC, the loadmap addresses need to be communicated * to GDB in order for GDB to do the necessary relocations. The @@ -1301,7 +1293,7 @@ static inline void fill_note(struct memelfnote *note, const char *name, int type * fill up all the fields in prstatus from the given task struct, except * registers which need to be filled up separately. */ -static void fill_prstatus(struct elf_prstatus_fdpic *prstatus, +static void fill_prstatus(struct elf_prstatus_common *prstatus, struct task_struct *p, long signr) { prstatus->pr_info.si_signo = prstatus->pr_cursig = signr; @@ -1332,9 +1324,6 @@ static void fill_prstatus(struct elf_prstatus_fdpic *prstatus, } prstatus->pr_cutime = ns_to_kernel_old_timeval(p->signal->cutime); prstatus->pr_cstime = ns_to_kernel_old_timeval(p->signal->cstime); - - prstatus->pr_exec_fdpic_loadmap = p->mm->context.exec_fdpic_loadmap; - prstatus->pr_interp_fdpic_loadmap = p->mm->context.interp_fdpic_loadmap; } static int fill_psinfo(struct elf_prpsinfo *psinfo, struct task_struct *p, @@ -1405,7 +1394,9 @@ static struct elf_thread_status *elf_dump_thread_status(long signr, struct task_ if (!t) return t; - fill_prstatus(&t->prstatus, p, signr); + fill_prstatus(&t->prstatus.common, p, signr); + t->prstatus.pr_exec_fdpic_loadmap = p->mm->context.exec_fdpic_loadmap; + t->prstatus.pr_interp_fdpic_loadmap = p->mm->context.interp_fdpic_loadmap; regset_get(p, &view->regsets[0], sizeof(t->prstatus.pr_reg), &t->prstatus.pr_reg); diff --git a/fs/binfmt_misc.c b/fs/binfmt_misc.c index 3880a82da1dc..e1eae7ea823a 100644 --- a/fs/binfmt_misc.c +++ b/fs/binfmt_misc.c @@ -153,7 +153,9 @@ static int load_misc_binary(struct linux_binprm *bprm) if (bprm->interp_flags & BINPRM_FLAGS_PATH_INACCESSIBLE) goto ret; - if (!(fmt->flags & MISC_FMT_PRESERVE_ARGV0)) { + if (fmt->flags & MISC_FMT_PRESERVE_ARGV0) { + bprm->interp_flags |= BINPRM_FLAGS_PRESERVE_ARGV0; + } else { retval = remove_arg_zero(bprm); if (retval) goto ret; @@ -647,12 +649,24 @@ static ssize_t bm_register_write(struct file *file, const char __user *buffer, struct super_block *sb = file_inode(file)->i_sb; struct dentry *root = sb->s_root, *dentry; int err = 0; + struct file *f = NULL; e = create_entry(buffer, count); if (IS_ERR(e)) return PTR_ERR(e); + if (e->flags & MISC_FMT_OPEN_FILE) { + f = open_exec(e->interpreter); + if (IS_ERR(f)) { + pr_notice("register: failed to install interpreter file %s\n", + e->interpreter); + kfree(e); + return PTR_ERR(f); + } + e->interp_file = f; + } + inode_lock(d_inode(root)); dentry = lookup_one_len(e->name, root, strlen(e->name)); err = PTR_ERR(dentry); @@ -676,21 +690,6 @@ static ssize_t bm_register_write(struct file *file, const char __user *buffer, goto out2; } - if (e->flags & MISC_FMT_OPEN_FILE) { - struct file *f; - - f = open_exec(e->interpreter); - if (IS_ERR(f)) { - err = PTR_ERR(f); - pr_notice("register: failed to install interpreter file %s\n", e->interpreter); - simple_release_fs(&bm_mnt, &entry_count); - iput(inode); - inode = NULL; - goto out2; - } - e->interp_file = f; - } - e->dentry = dget(dentry); inode->i_private = e; inode->i_fop = &bm_entry_operations; @@ -707,6 +706,8 @@ out: inode_unlock(d_inode(root)); if (err) { + if (f) + filp_close(f, NULL); kfree(e); return err; } diff --git a/fs/block_dev.c b/fs/block_dev.c index 235b5042672e..92ed7d5df677 100644 --- a/fs/block_dev.c +++ b/fs/block_dev.c @@ -118,15 +118,23 @@ int truncate_bdev_range(struct block_device *bdev, fmode_t mode, if (!(mode & FMODE_EXCL)) { int err = bd_prepare_to_claim(bdev, truncate_bdev_range); if (err) - return err; + goto invalidate; } truncate_inode_pages_range(bdev->bd_inode->i_mapping, lstart, lend); if (!(mode & FMODE_EXCL)) bd_abort_claiming(bdev, truncate_bdev_range); return 0; + +invalidate: + /* + * Someone else has handle exclusively open. Try invalidating instead. + * The 'end' argument is inclusive so the rounding is safe. + */ + return invalidate_inode_pages2_range(bdev->bd_inode->i_mapping, + lstart >> PAGE_SHIFT, + lend >> PAGE_SHIFT); } -EXPORT_SYMBOL(truncate_bdev_range); static void set_init_blocksize(struct block_device *bdev) { @@ -222,7 +230,7 @@ static void blkdev_bio_end_io_simple(struct bio *bio) static ssize_t __blkdev_direct_IO_simple(struct kiocb *iocb, struct iov_iter *iter, - int nr_pages) + unsigned int nr_pages) { struct file *file = iocb->ki_filp; struct block_device *bdev = I_BDEV(bdev_file_inode(file)); @@ -356,8 +364,8 @@ static void blkdev_bio_end_io(struct bio *bio) } } -static ssize_t -__blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, int nr_pages) +static ssize_t __blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, + unsigned int nr_pages) { struct file *file = iocb->ki_filp; struct inode *inode = bdev_file_inode(file); @@ -424,7 +432,7 @@ __blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, int nr_pages) dio->size += bio->bi_iter.bi_size; pos += bio->bi_iter.bi_size; - nr_pages = iov_iter_npages(iter, BIO_MAX_PAGES); + nr_pages = bio_iov_vecs_to_alloc(iter, BIO_MAX_VECS); if (!nr_pages) { bool polled = false; @@ -487,15 +495,16 @@ __blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter, int nr_pages) static ssize_t blkdev_direct_IO(struct kiocb *iocb, struct iov_iter *iter) { - int nr_pages; + unsigned int nr_pages; - nr_pages = iov_iter_npages(iter, BIO_MAX_PAGES + 1); - if (!nr_pages) + if (!iov_iter_count(iter)) return 0; - if (is_sync_kiocb(iocb) && nr_pages <= BIO_MAX_PAGES) + + nr_pages = bio_iov_vecs_to_alloc(iter, BIO_MAX_VECS + 1); + if (is_sync_kiocb(iocb) && nr_pages <= BIO_MAX_VECS) return __blkdev_direct_IO_simple(iocb, iter, nr_pages); - return __blkdev_direct_IO(iocb, iter, min(nr_pages, BIO_MAX_PAGES)); + return __blkdev_direct_IO(iocb, iter, bio_max_segs(nr_pages)); } static __init int blkdev_init(void) @@ -688,7 +697,7 @@ int blkdev_fsync(struct file *filp, loff_t start, loff_t end, int datasync) * i_mutex and doing so causes performance issues with concurrent * O_SYNC writers to a block device. */ - error = blkdev_issue_flush(bdev, GFP_KERNEL); + error = blkdev_issue_flush(bdev); if (error == -EOPNOTSUPP) error = 0; @@ -1270,7 +1279,7 @@ rescan: return ret; } /* - * Only exported for for loop and dasd for historic reasons. Don't use in new + * Only exported for loop and dasd for historic reasons. Don't use in new * code! */ EXPORT_SYMBOL_GPL(bdev_disk_changed); @@ -1808,13 +1817,11 @@ static long blkdev_fallocate(struct file *file, int mode, loff_t start, return error; /* - * Invalidate again; if someone wandered in and dirtied a page, - * the caller will be given -EBUSY. The third argument is - * inclusive, so the rounding here is safe. + * Invalidate the page cache again; if someone wandered in and dirtied + * a page, we just discard it - userspace has no way of knowing whether + * the write happened before or after discard completing... */ - return invalidate_inode_pages2_range(bdev->bd_inode->i_mapping, - start >> PAGE_SHIFT, - end >> PAGE_SHIFT); + return truncate_bdev_range(bdev, file->f_mode, start, end); } const struct file_operations def_blk_fops = { diff --git a/fs/btrfs/Makefile b/fs/btrfs/Makefile index 9f1b1a88e317..b634c42115ea 100644 --- a/fs/btrfs/Makefile +++ b/fs/btrfs/Makefile @@ -1,5 +1,21 @@ # SPDX-License-Identifier: GPL-2.0 +# Subset of W=1 warnings +subdir-ccflags-y += -Wextra -Wunused -Wno-unused-parameter +subdir-ccflags-y += -Wmissing-declarations +subdir-ccflags-y += -Wmissing-format-attribute +subdir-ccflags-y += -Wmissing-prototypes +subdir-ccflags-y += -Wold-style-definition +subdir-ccflags-y += -Wmissing-include-dirs +subdir-ccflags-y += $(call cc-option, -Wunused-but-set-variable) +subdir-ccflags-y += $(call cc-option, -Wunused-const-variable) +subdir-ccflags-y += $(call cc-option, -Wpacked-not-aligned) +subdir-ccflags-y += $(call cc-option, -Wstringop-truncation) +# The following turn off the warnings enabled by -Wextra +subdir-ccflags-y += -Wno-missing-field-initializers +subdir-ccflags-y += -Wno-sign-compare +subdir-ccflags-y += -Wno-type-limits + obj-$(CONFIG_BTRFS_FS) := btrfs.o btrfs-y += super.o ctree.o extent-tree.o print-tree.o root-tree.o dir-item.o \ @@ -11,7 +27,8 @@ btrfs-y += super.o ctree.o extent-tree.o print-tree.o root-tree.o dir-item.o \ compression.o delayed-ref.o relocation.o delayed-inode.o scrub.o \ reada.o backref.o ulist.o qgroup.o send.o dev-replace.o raid56.o \ uuid-tree.o props.o free-space-tree.o tree-checker.o space-info.o \ - block-rsv.o delalloc-space.o block-group.o discard.o reflink.o + block-rsv.o delalloc-space.o block-group.o discard.o reflink.o \ + subpage.o btrfs-$(CONFIG_BTRFS_FS_POSIX_ACL) += acl.o btrfs-$(CONFIG_BTRFS_FS_CHECK_INTEGRITY) += check-integrity.o diff --git a/fs/btrfs/acl.c b/fs/btrfs/acl.c index a0af1b952c4d..d95eb5c8cb37 100644 --- a/fs/btrfs/acl.c +++ b/fs/btrfs/acl.c @@ -107,13 +107,15 @@ out: return ret; } -int btrfs_set_acl(struct inode *inode, struct posix_acl *acl, int type) +int btrfs_set_acl(struct user_namespace *mnt_userns, struct inode *inode, + struct posix_acl *acl, int type) { int ret; umode_t old_mode = inode->i_mode; if (type == ACL_TYPE_ACCESS && acl) { - ret = posix_acl_update_mode(inode, &inode->i_mode, &acl); + ret = posix_acl_update_mode(&init_user_ns, inode, + &inode->i_mode, &acl); if (ret) return ret; } diff --git a/fs/btrfs/backref.c b/fs/btrfs/backref.c index 9cadacf3ec27..f47c1528eb9a 100644 --- a/fs/btrfs/backref.c +++ b/fs/btrfs/backref.c @@ -1501,7 +1501,13 @@ int btrfs_find_all_roots(struct btrfs_trans_handle *trans, } /** - * btrfs_check_shared - tell us whether an extent is shared + * Check if an extent is shared or not + * + * @root: root inode belongs to + * @inum: inode number of the inode whose extent we are checking + * @bytenr: logical bytenr of the extent we are checking + * @roots: list of roots this extent is shared among + * @tmp: temporary list used for iteration * * btrfs_check_shared uses the backref walking code but will short * circuit as soon as it finds a root or inode that doesn't match the @@ -2541,13 +2547,6 @@ void btrfs_backref_cleanup_node(struct btrfs_backref_cache *cache, list_del(&edge->list[UPPER]); btrfs_backref_free_edge(cache, edge); - if (RB_EMPTY_NODE(&upper->rb_node)) { - BUG_ON(!list_empty(&node->upper)); - btrfs_backref_drop_node(cache, node); - node = upper; - node->lowest = 1; - continue; - } /* * Add the node to leaf node list if no other child block * cached. @@ -2624,7 +2623,7 @@ static int handle_direct_tree_backref(struct btrfs_backref_cache *cache, /* Only reloc backref cache cares about a specific root */ if (cache->is_reloc) { root = find_reloc_root(cache->fs_info, cur->bytenr); - if (WARN_ON(!root)) + if (!root) return -ENOENT; cur->root = root; } else { diff --git a/fs/btrfs/backref.h b/fs/btrfs/backref.h index ff705cc564a9..17abde7f794c 100644 --- a/fs/btrfs/backref.h +++ b/fs/btrfs/backref.h @@ -296,6 +296,9 @@ static inline void btrfs_backref_free_node(struct btrfs_backref_cache *cache, struct btrfs_backref_node *node) { if (node) { + ASSERT(list_empty(&node->list)); + ASSERT(list_empty(&node->lower)); + ASSERT(node->eb == NULL); cache->nr_nodes--; btrfs_put_root(node->root); kfree(node); @@ -340,11 +343,11 @@ static inline void btrfs_backref_drop_node_buffer( static inline void btrfs_backref_drop_node(struct btrfs_backref_cache *tree, struct btrfs_backref_node *node) { - BUG_ON(!list_empty(&node->upper)); + ASSERT(list_empty(&node->upper)); btrfs_backref_drop_node_buffer(node); - list_del(&node->list); - list_del(&node->lower); + list_del_init(&node->list); + list_del_init(&node->lower); if (!RB_EMPTY_NODE(&node->rb_node)) rb_erase(&node->rb_node, &tree->rb_root); btrfs_backref_free_node(tree, node); diff --git a/fs/btrfs/block-group.c b/fs/btrfs/block-group.c index 48ebc106a606..744b99ddc28c 100644 --- a/fs/btrfs/block-group.c +++ b/fs/btrfs/block-group.c @@ -15,6 +15,7 @@ #include "delalloc-space.h" #include "discard.h" #include "raid56.h" +#include "zoned.h" /* * Return target flags in extended format or 0 if restripe for this chunk_type @@ -724,6 +725,10 @@ int btrfs_cache_block_group(struct btrfs_block_group *cache, int load_cache_only struct btrfs_caching_control *caching_ctl = NULL; int ret = 0; + /* Allocator for zoned filesystems does not use the cache at all */ + if (btrfs_is_zoned(fs_info)) + return 0; + caching_ctl = kzalloc(sizeof(*caching_ctl), GFP_NOFS); if (!caching_ctl) return -ENOMEM; @@ -896,6 +901,8 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans, btrfs_return_cluster_to_free_space(block_group, cluster); spin_unlock(&cluster->refill_lock); + btrfs_clear_treelog_bg(block_group); + path = btrfs_alloc_path(); if (!path) { ret = -ENOMEM; @@ -1008,12 +1015,17 @@ int btrfs_remove_block_group(struct btrfs_trans_handle *trans, WARN_ON(block_group->space_info->total_bytes < block_group->length); WARN_ON(block_group->space_info->bytes_readonly - < block_group->length); + < block_group->length - block_group->zone_unusable); + WARN_ON(block_group->space_info->bytes_zone_unusable + < block_group->zone_unusable); WARN_ON(block_group->space_info->disk_total < block_group->length * factor); } block_group->space_info->total_bytes -= block_group->length; - block_group->space_info->bytes_readonly -= block_group->length; + block_group->space_info->bytes_readonly -= + (block_group->length - block_group->zone_unusable); + block_group->space_info->bytes_zone_unusable -= + block_group->zone_unusable; block_group->space_info->disk_total -= block_group->length * factor; spin_unlock(&block_group->space_info->lock); @@ -1150,6 +1162,11 @@ static int inc_block_group_ro(struct btrfs_block_group *cache, int force) spin_lock(&sinfo->lock); spin_lock(&cache->lock); + if (cache->swap_extents) { + ret = -ETXTBSY; + goto out; + } + if (cache->ro) { cache->ro++; ret = 0; @@ -1157,7 +1174,7 @@ static int inc_block_group_ro(struct btrfs_block_group *cache, int force) } num_bytes = cache->length - cache->reserved - cache->pinned - - cache->bytes_super - cache->used; + cache->bytes_super - cache->zone_unusable - cache->used; /* * Data never overcommits, even in mixed mode, so do just the straight @@ -1188,6 +1205,12 @@ static int inc_block_group_ro(struct btrfs_block_group *cache, int force) if (!ret) { sinfo->bytes_readonly += num_bytes; + if (btrfs_is_zoned(cache->fs_info)) { + /* Migrate zone_unusable bytes to readonly */ + sinfo->bytes_readonly += cache->zone_unusable; + sinfo->bytes_zone_unusable -= cache->zone_unusable; + cache->zone_unusable = 0; + } cache->ro++; list_add_tail(&cache->ro_list, &sinfo->ro_bgs); } @@ -1262,6 +1285,13 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info) if (!test_bit(BTRFS_FS_OPEN, &fs_info->flags)) return; + /* + * Long running balances can keep us blocked here for eternity, so + * simply skip deletion if we're unable to get the mutex. + */ + if (!mutex_trylock(&fs_info->delete_unused_bgs_mutex)) + return; + spin_lock(&fs_info->unused_bgs_lock); while (!list_empty(&fs_info->unused_bgs)) { int trimming; @@ -1281,8 +1311,6 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info) btrfs_discard_cancel_work(&fs_info->discard_ctl, block_group); - mutex_lock(&fs_info->delete_unused_bgs_mutex); - /* Don't want to race with allocators so take the groups_sem */ down_write(&space_info->groups_sem); @@ -1371,9 +1399,7 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info) btrfs_space_info_update_bytes_pinned(fs_info, space_info, -block_group->pinned); space_info->bytes_readonly += block_group->pinned; - percpu_counter_add_batch(&space_info->total_bytes_pinned, - -block_group->pinned, - BTRFS_TOTAL_BYTES_PINNED_BATCH); + __btrfs_mod_total_bytes_pinned(space_info, -block_group->pinned); block_group->pinned = 0; spin_unlock(&block_group->lock); @@ -1389,8 +1415,12 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info) if (!async_trim_enabled && btrfs_test_opt(fs_info, DISCARD_ASYNC)) goto flip_async; - /* DISCARD can flip during remount */ - trimming = btrfs_test_opt(fs_info, DISCARD_SYNC); + /* + * DISCARD can flip during remount. On zoned filesystems, we + * need to reset sequential-required zones. + */ + trimming = btrfs_test_opt(fs_info, DISCARD_SYNC) || + btrfs_is_zoned(fs_info); /* Implicit trim during transaction commit. */ if (trimming) @@ -1428,11 +1458,11 @@ void btrfs_delete_unused_bgs(struct btrfs_fs_info *fs_info) end_trans: btrfs_end_transaction(trans); next: - mutex_unlock(&fs_info->delete_unused_bgs_mutex); btrfs_put_block_group(block_group); spin_lock(&fs_info->unused_bgs_lock); } spin_unlock(&fs_info->unused_bgs_lock); + mutex_unlock(&fs_info->delete_unused_bgs_mutex); return; flip_async: @@ -1561,8 +1591,11 @@ static void set_avail_alloc_bits(struct btrfs_fs_info *fs_info, u64 flags) } /** - * btrfs_rmap_block - Map a physical disk address to a list of logical addresses + * Map a physical disk address to a list of logical addresses + * + * @fs_info: the filesystem * @chunk_start: logical address of block group + * @bdev: physical device to resolve, can be NULL to indicate any device * @physical: physical address to map to logical addresses * @logical: return array of logical addresses which map to @physical * @naddrs: length of @logical @@ -1572,9 +1605,9 @@ static void set_avail_alloc_bits(struct btrfs_fs_info *fs_info, u64 flags) * Used primarily to exclude those portions of a block group that contain super * block copies. */ -EXPORT_FOR_TESTS int btrfs_rmap_block(struct btrfs_fs_info *fs_info, u64 chunk_start, - u64 physical, u64 **logical, int *naddrs, int *stripe_len) + struct block_device *bdev, u64 physical, u64 **logical, + int *naddrs, int *stripe_len) { struct extent_map *em; struct map_lookup *map; @@ -1592,6 +1625,7 @@ int btrfs_rmap_block(struct btrfs_fs_info *fs_info, u64 chunk_start, map = em->map_lookup; data_stripe_length = em->orig_block_len; io_stripe_size = map->stripe_len; + chunk_start = em->start; /* For RAID5/6 adjust to a full IO stripe length */ if (map->type & BTRFS_BLOCK_GROUP_RAID56_MASK) @@ -1606,14 +1640,18 @@ int btrfs_rmap_block(struct btrfs_fs_info *fs_info, u64 chunk_start, for (i = 0; i < map->num_stripes; i++) { bool already_inserted = false; u64 stripe_nr; + u64 offset; int j; if (!in_range(physical, map->stripes[i].physical, data_stripe_length)) continue; + if (bdev && map->stripes[i].dev->bdev != bdev) + continue; + stripe_nr = physical - map->stripes[i].physical; - stripe_nr = div64_u64(stripe_nr, map->stripe_len); + stripe_nr = div64_u64_rem(stripe_nr, map->stripe_len, &offset); if (map->type & BTRFS_BLOCK_GROUP_RAID10) { stripe_nr = stripe_nr * map->num_stripes + i; @@ -1627,7 +1665,7 @@ int btrfs_rmap_block(struct btrfs_fs_info *fs_info, u64 chunk_start, * instead of map->stripe_len */ - bytenr = chunk_start + stripe_nr * io_stripe_size; + bytenr = chunk_start + stripe_nr * io_stripe_size + offset; /* Ensure we don't add duplicate addresses */ for (j = 0; j < nr; j++) { @@ -1669,7 +1707,7 @@ static int exclude_super_stripes(struct btrfs_block_group *cache) for (i = 0; i < BTRFS_SUPER_MIRROR_MAX; i++) { bytenr = btrfs_sb_offset(i); - ret = btrfs_rmap_block(fs_info, cache->start, + ret = btrfs_rmap_block(fs_info, cache->start, NULL, bytenr, &logical, &nr, &stripe_len); if (ret) return ret; @@ -1805,24 +1843,8 @@ static int check_chunk_block_group_mappings(struct btrfs_fs_info *fs_info) return ret; } -static void read_block_group_item(struct btrfs_block_group *cache, - struct btrfs_path *path, - const struct btrfs_key *key) -{ - struct extent_buffer *leaf = path->nodes[0]; - struct btrfs_block_group_item bgi; - int slot = path->slots[0]; - - cache->length = key->offset; - - read_extent_buffer(leaf, &bgi, btrfs_item_ptr_offset(leaf, slot), - sizeof(bgi)); - cache->used = btrfs_stack_block_group_used(&bgi); - cache->flags = btrfs_stack_block_group_flags(&bgi); -} - static int read_one_block_group(struct btrfs_fs_info *info, - struct btrfs_path *path, + struct btrfs_block_group_item *bgi, const struct btrfs_key *key, int need_clear) { @@ -1837,7 +1859,9 @@ static int read_one_block_group(struct btrfs_fs_info *info, if (!cache) return -ENOMEM; - read_block_group_item(cache, path, key); + cache->length = key->offset; + cache->used = btrfs_stack_block_group_used(bgi); + cache->flags = btrfs_stack_block_group_flags(bgi); set_free_space_tree_thresholds(cache); @@ -1864,6 +1888,13 @@ static int read_one_block_group(struct btrfs_fs_info *info, goto error; } + ret = btrfs_load_block_group_zone_info(cache, false); + if (ret) { + btrfs_err(info, "zoned: failed to load zone info of bg %llu", + cache->start); + goto error; + } + /* * We need to exclude the super stripes now so that the space info has * super bytes accounted for, otherwise we'll think we have more space @@ -1877,12 +1908,20 @@ static int read_one_block_group(struct btrfs_fs_info *info, } /* - * Check for two cases, either we are full, and therefore don't need - * to bother with the caching work since we won't find any space, or we - * are empty, and we can just add all the space in and be done with it. - * This saves us _a_lot_ of time, particularly in the full case. + * For zoned filesystem, space after the allocation offset is the only + * free space for a block group. So, we don't need any caching work. + * btrfs_calc_zone_unusable() will set the amount of free space and + * zone_unusable space. + * + * For regular filesystem, check for two cases, either we are full, and + * therefore don't need to bother with the caching work since we won't + * find any space, or we are empty, and we can just add all the space + * in and be done with it. This saves us _a_lot_ of time, particularly + * in the full case. */ - if (cache->length == cache->used) { + if (btrfs_is_zoned(info)) { + btrfs_calc_zone_unusable(cache); + } else if (cache->length == cache->used) { cache->last_byte_to_unpin = (u64)-1; cache->cached = BTRFS_CACHE_FINISHED; btrfs_free_excluded_extents(cache); @@ -1901,7 +1940,8 @@ static int read_one_block_group(struct btrfs_fs_info *info, } trace_btrfs_add_block_group(info, cache, 0); btrfs_update_space_info(info, cache->flags, cache->length, - cache->used, cache->bytes_super, &space_info); + cache->used, cache->bytes_super, + cache->zone_unusable, &space_info); cache->space_info = space_info; @@ -1957,7 +1997,7 @@ static int fill_dummy_bgs(struct btrfs_fs_info *fs_info) break; } btrfs_update_space_info(fs_info, bg->flags, em->len, em->len, - 0, &space_info); + 0, 0, &space_info); bg->space_info = space_info; link_block_group(bg); @@ -1996,19 +2036,29 @@ int btrfs_read_block_groups(struct btrfs_fs_info *info) need_clear = 1; while (1) { + struct btrfs_block_group_item bgi; + struct extent_buffer *leaf; + int slot; + ret = find_first_block_group(info, path, &key); if (ret > 0) break; if (ret != 0) goto error; - btrfs_item_key_to_cpu(path->nodes[0], &key, path->slots[0]); - ret = read_one_block_group(info, path, &key, need_clear); + leaf = path->nodes[0]; + slot = path->slots[0]; + + read_extent_buffer(leaf, &bgi, btrfs_item_ptr_offset(leaf, slot), + sizeof(bgi)); + + btrfs_item_key_to_cpu(leaf, &key, slot); + btrfs_release_path(path); + ret = read_one_block_group(info, &bgi, &key, need_clear); if (ret < 0) goto error; key.objectid += key.offset; key.offset = 0; - btrfs_release_path(path); } btrfs_release_path(path); @@ -2140,6 +2190,13 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans, u64 bytes_used, cache->cached = BTRFS_CACHE_FINISHED; if (btrfs_fs_compat_ro(fs_info, FREE_SPACE_TREE)) cache->needs_free_space = 1; + + ret = btrfs_load_block_group_zone_info(cache, true); + if (ret) { + btrfs_put_block_group(cache); + return ret; + } + ret = exclude_super_stripes(cache); if (ret) { /* We may have excluded something, so call this just in case */ @@ -2181,7 +2238,7 @@ int btrfs_make_block_group(struct btrfs_trans_handle *trans, u64 bytes_used, */ trace_btrfs_add_block_group(fs_info, cache, 1); btrfs_update_space_info(fs_info, cache->flags, size, bytes_used, - cache->bytes_super, &cache->space_info); + cache->bytes_super, 0, &cache->space_info); btrfs_update_global_block_rsv(fs_info); link_block_group(cache); @@ -2255,7 +2312,7 @@ again: } ret = inc_block_group_ro(cache, 0); - if (!do_chunk_alloc) + if (!do_chunk_alloc || ret == -ETXTBSY) goto unlock_out; if (!ret) goto out; @@ -2264,6 +2321,8 @@ again: if (ret < 0) goto out; ret = inc_block_group_ro(cache, 0); + if (ret == -ETXTBSY) + goto unlock_out; out: if (cache->flags & BTRFS_BLOCK_GROUP_SYSTEM) { alloc_flags = btrfs_get_alloc_profile(fs_info, cache->flags); @@ -2289,8 +2348,15 @@ void btrfs_dec_block_group_ro(struct btrfs_block_group *cache) spin_lock(&cache->lock); if (!--cache->ro) { num_bytes = cache->length - cache->reserved - - cache->pinned - cache->bytes_super - cache->used; + cache->pinned - cache->bytes_super - + cache->zone_unusable - cache->used; sinfo->bytes_readonly -= num_bytes; + if (btrfs_is_zoned(cache->fs_info)) { + /* Migrate zone_unusable bytes back */ + cache->zone_unusable = cache->alloc_offset - cache->used; + sinfo->bytes_zone_unusable += cache->zone_unusable; + sinfo->bytes_readonly -= cache->zone_unusable; + } list_del_init(&cache->ro_list); } spin_unlock(&cache->lock); @@ -2564,8 +2630,10 @@ again: if (!path) { path = btrfs_alloc_path(); - if (!path) - return -ENOMEM; + if (!path) { + ret = -ENOMEM; + goto out; + } } /* @@ -2659,16 +2727,14 @@ again: btrfs_put_block_group(cache); if (drop_reserve) btrfs_delayed_refs_rsv_release(fs_info, 1); - - if (ret) - break; - /* * Avoid blocking other tasks for too long. It might even save * us from writing caches for block groups that are going to be * removed. */ mutex_unlock(&trans->transaction->cache_write_mutex); + if (ret) + goto out; mutex_lock(&trans->transaction->cache_write_mutex); } mutex_unlock(&trans->transaction->cache_write_mutex); @@ -2692,7 +2758,12 @@ again: goto again; } spin_unlock(&cur_trans->dirty_bgs_lock); - } else if (ret < 0) { + } +out: + if (ret < 0) { + spin_lock(&cur_trans->dirty_bgs_lock); + list_splice_init(&dirty, &cur_trans->dirty_bgs); + spin_unlock(&cur_trans->dirty_bgs_lock); btrfs_cleanup_dirty_bgs(cur_trans, fs_info); } @@ -2896,10 +2967,8 @@ int btrfs_update_block_group(struct btrfs_trans_handle *trans, spin_unlock(&cache->lock); spin_unlock(&cache->space_info->lock); - percpu_counter_add_batch( - &cache->space_info->total_bytes_pinned, - num_bytes, - BTRFS_TOTAL_BYTES_PINNED_BATCH); + __btrfs_mod_total_bytes_pinned(cache->space_info, + num_bytes); set_extent_dirty(&trans->transaction->pinned_extents, bytenr, bytenr + num_bytes - 1, GFP_NOFS | __GFP_NOFAIL); @@ -3344,6 +3413,7 @@ int btrfs_free_block_groups(struct btrfs_fs_info *info) ASSERT(list_empty(&block_group->io_list)); ASSERT(list_empty(&block_group->bg_list)); ASSERT(refcount_read(&block_group->refs) == 1); + ASSERT(block_group->swap_extents == 0); btrfs_put_block_group(block_group); spin_lock(&info->block_group_cache_lock); @@ -3410,3 +3480,26 @@ void btrfs_unfreeze_block_group(struct btrfs_block_group *block_group) __btrfs_remove_free_space_cache(block_group->free_space_ctl); } } + +bool btrfs_inc_block_group_swap_extents(struct btrfs_block_group *bg) +{ + bool ret = true; + + spin_lock(&bg->lock); + if (bg->ro) + ret = false; + else + bg->swap_extents++; + spin_unlock(&bg->lock); + + return ret; +} + +void btrfs_dec_block_group_swap_extents(struct btrfs_block_group *bg, int amount) +{ + spin_lock(&bg->lock); + ASSERT(!bg->ro); + ASSERT(bg->swap_extents >= amount); + bg->swap_extents -= amount; + spin_unlock(&bg->lock); +} diff --git a/fs/btrfs/block-group.h b/fs/btrfs/block-group.h index 8f74a96074f7..3ecc3372a5ce 100644 --- a/fs/btrfs/block-group.h +++ b/fs/btrfs/block-group.h @@ -95,6 +95,8 @@ struct btrfs_block_group { unsigned int iref:1; unsigned int has_caching_ctl:1; unsigned int removed:1; + unsigned int to_copy:1; + unsigned int relocating_repair:1; int disk_cache_state; @@ -181,8 +183,25 @@ struct btrfs_block_group { */ int needs_free_space; + /* Flag indicating this block group is placed on a sequential zone */ + bool seq_zone; + + /* + * Number of extents in this block group used for swap files. + * All accesses protected by the spinlock 'lock'. + */ + int swap_extents; + /* Record locked full stripes for RAID5/6 block group */ struct btrfs_full_stripe_locks_tree full_stripe_locks_root; + + /* + * Allocation offset for the block group to implement sequential + * allocation. This is used only on a zoned filesystem. + */ + u64 alloc_offset; + u64 zone_unusable; + u64 meta_write_pointer; }; static inline u64 btrfs_block_group_end(struct btrfs_block_group *block_group) @@ -270,6 +289,9 @@ void btrfs_put_block_group_cache(struct btrfs_fs_info *info); int btrfs_free_block_groups(struct btrfs_fs_info *info); void btrfs_wait_space_cache_v1_finished(struct btrfs_block_group *cache, struct btrfs_caching_control *caching_ctl); +int btrfs_rmap_block(struct btrfs_fs_info *fs_info, u64 chunk_start, + struct block_device *bdev, u64 physical, u64 **logical, + int *naddrs, int *stripe_len); static inline u64 btrfs_data_alloc_profile(struct btrfs_fs_info *fs_info) { @@ -296,9 +318,7 @@ static inline int btrfs_block_group_done(struct btrfs_block_group *cache) void btrfs_freeze_block_group(struct btrfs_block_group *cache); void btrfs_unfreeze_block_group(struct btrfs_block_group *cache); -#ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS -int btrfs_rmap_block(struct btrfs_fs_info *fs_info, u64 chunk_start, - u64 physical, u64 **logical, int *naddrs, int *stripe_len); -#endif +bool btrfs_inc_block_group_swap_extents(struct btrfs_block_group *bg); +void btrfs_dec_block_group_swap_extents(struct btrfs_block_group *bg, int amount); #endif /* BTRFS_BLOCK_GROUP_H */ diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h index d9bf53d9ff90..28e202e89660 100644 --- a/fs/btrfs/btrfs_inode.h +++ b/fs/btrfs/btrfs_inode.h @@ -325,7 +325,8 @@ struct btrfs_dio_private { struct inode *inode; u64 logical_offset; u64 disk_bytenr; - u64 bytes; + /* Used for bio::bi_size */ + u32 bytes; /* * References to this structure. There is one reference per in-flight diff --git a/fs/btrfs/check-integrity.c b/fs/btrfs/check-integrity.c index 6ff44e53814c..113cb85c1fd4 100644 --- a/fs/btrfs/check-integrity.c +++ b/fs/btrfs/check-integrity.c @@ -2674,7 +2674,7 @@ static void __btrfsic_submit_bio(struct bio *bio) mutex_lock(&btrfsic_mutex); /* since btrfsic_submit_bio() is also called before * btrfsic_mount(), this might return NULL */ - dev_state = btrfsic_dev_state_lookup(bio_dev(bio) + bio->bi_partno); + dev_state = btrfsic_dev_state_lookup(bio->bi_bdev->bd_dev); if (NULL != dev_state && (bio_op(bio) == REQ_OP_WRITE) && bio_has_data(bio)) { unsigned int i = 0; @@ -2690,9 +2690,9 @@ static void __btrfsic_submit_bio(struct bio *bio) bio_is_patched = 0; if (dev_state->state->print_mask & BTRFSIC_PRINT_MASK_SUBMIT_BIO_BH) - pr_info("submit_bio(rw=%d,0x%x, bi_vcnt=%u, bi_sector=%llu (bytenr %llu), bi_disk=%p)\n", + pr_info("submit_bio(rw=%d,0x%x, bi_vcnt=%u, bi_sector=%llu (bytenr %llu), bi_bdev=%p)\n", bio_op(bio), bio->bi_opf, segs, - bio->bi_iter.bi_sector, dev_bytenr, bio->bi_disk); + bio->bi_iter.bi_sector, dev_bytenr, bio->bi_bdev); mapped_datav = kmalloc_array(segs, sizeof(*mapped_datav), GFP_NOFS); @@ -2721,8 +2721,8 @@ static void __btrfsic_submit_bio(struct bio *bio) } else if (NULL != dev_state && (bio->bi_opf & REQ_PREFLUSH)) { if (dev_state->state->print_mask & BTRFSIC_PRINT_MASK_SUBMIT_BIO_BH) - pr_info("submit_bio(rw=%d,0x%x FLUSH, disk=%p)\n", - bio_op(bio), bio->bi_opf, bio->bi_disk); + pr_info("submit_bio(rw=%d,0x%x FLUSH, bdev=%p)\n", + bio_op(bio), bio->bi_opf, bio->bi_bdev); if (!dev_state->dummy_block_for_bio_bh_flush.is_iodone) { if ((dev_state->state->print_mask & (BTRFSIC_PRINT_MASK_SUBMIT_BIO_BH | diff --git a/fs/btrfs/compression.c b/fs/btrfs/compression.c index 5ae3fa0386b7..3f4c832abfed 100644 --- a/fs/btrfs/compression.c +++ b/fs/btrfs/compression.c @@ -141,6 +141,7 @@ static int check_compressed_csum(struct btrfs_inode *inode, struct bio *bio, struct btrfs_fs_info *fs_info = inode->root->fs_info; SHASH_DESC_ON_STACK(shash, fs_info->csum_shash); const u32 csum_size = fs_info->csum_size; + const u32 sectorsize = fs_info->sectorsize; struct page *page; unsigned long i; char *kaddr; @@ -154,22 +155,34 @@ static int check_compressed_csum(struct btrfs_inode *inode, struct bio *bio, shash->tfm = fs_info->csum_shash; for (i = 0; i < cb->nr_pages; i++) { + u32 pg_offset; + u32 bytes_left = PAGE_SIZE; page = cb->compressed_pages[i]; - kaddr = kmap_atomic(page); - crypto_shash_digest(shash, kaddr, PAGE_SIZE, csum); - kunmap_atomic(kaddr); - - if (memcmp(&csum, cb_sum, csum_size)) { - btrfs_print_data_csum_error(inode, disk_start, - csum, cb_sum, cb->mirror_num); - if (btrfs_io_bio(bio)->device) - btrfs_dev_stat_inc_and_print( - btrfs_io_bio(bio)->device, - BTRFS_DEV_STAT_CORRUPTION_ERRS); - return -EIO; + /* Determine the remaining bytes inside the page first */ + if (i == cb->nr_pages - 1) + bytes_left = cb->compressed_len - i * PAGE_SIZE; + + /* Hash through the page sector by sector */ + for (pg_offset = 0; pg_offset < bytes_left; + pg_offset += sectorsize) { + kaddr = kmap_atomic(page); + crypto_shash_digest(shash, kaddr + pg_offset, + sectorsize, csum); + kunmap_atomic(kaddr); + + if (memcmp(&csum, cb_sum, csum_size) != 0) { + btrfs_print_data_csum_error(inode, disk_start, + csum, cb_sum, cb->mirror_num); + if (btrfs_io_bio(bio)->device) + btrfs_dev_stat_inc_and_print( + btrfs_io_bio(bio)->device, + BTRFS_DEV_STAT_CORRUPTION_ERRS); + return -EIO; + } + cb_sum += csum_size; + disk_start += sectorsize; } - cb_sum += csum_size; } return 0; } @@ -542,13 +555,19 @@ static noinline int add_ra_bio_pages(struct inode *inode, goto next; } - end = last_offset + PAGE_SIZE - 1; /* * at this point, we have a locked page in the page cache * for these bytes in the file. But, we have to make * sure they map to this compressed extent on disk. */ - set_page_extent_mapped(page); + ret = set_page_extent_mapped(page); + if (ret < 0) { + unlock_page(page); + put_page(page); + break; + } + + end = last_offset + PAGE_SIZE - 1; lock_extent(tree, last_offset, end); read_lock(&em_tree->lock); em = lookup_extent_mapping(em_tree, last_offset, @@ -634,7 +653,7 @@ blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio, read_lock(&em_tree->lock); em = lookup_extent_mapping(em_tree, page_offset(bio_first_page_all(bio)), - PAGE_SIZE); + fs_info->sectorsize); read_unlock(&em_tree->lock); if (!em) return BLK_STS_IOERR; @@ -692,19 +711,30 @@ blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio, refcount_set(&cb->pending_bios, 1); for (pg_index = 0; pg_index < nr_pages; pg_index++) { + u32 pg_len = PAGE_SIZE; int submit = 0; + /* + * To handle subpage case, we need to make sure the bio only + * covers the range we need. + * + * If we're at the last page, truncate the length to only cover + * the remaining part. + */ + if (pg_index == nr_pages - 1) + pg_len = min_t(u32, PAGE_SIZE, + compressed_len - pg_index * PAGE_SIZE); + page = cb->compressed_pages[pg_index]; page->mapping = inode->i_mapping; page->index = em_start >> PAGE_SHIFT; if (comp_bio->bi_iter.bi_size) - submit = btrfs_bio_fits_in_stripe(page, PAGE_SIZE, + submit = btrfs_bio_fits_in_stripe(page, pg_len, comp_bio, 0); page->mapping = NULL; - if (submit || bio_add_page(comp_bio, page, PAGE_SIZE, 0) < - PAGE_SIZE) { + if (submit || bio_add_page(comp_bio, page, pg_len, 0) < pg_len) { unsigned int nr_sectors; ret = btrfs_bio_wq_end_io(fs_info, comp_bio, @@ -737,9 +767,9 @@ blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio, comp_bio->bi_private = cb; comp_bio->bi_end_io = end_compressed_bio_read; - bio_add_page(comp_bio, page, PAGE_SIZE, 0); + bio_add_page(comp_bio, page, pg_len, 0); } - cur_disk_byte += PAGE_SIZE; + cur_disk_byte += pg_len; } ret = btrfs_bio_wq_end_io(fs_info, comp_bio, BTRFS_WQ_ENDIO_DATA); @@ -1231,7 +1261,6 @@ int btrfs_decompress_buf2page(const char *buf, unsigned long buf_start, unsigned long prev_start_byte; unsigned long working_bytes = total_out - buf_start; unsigned long bytes; - char *kaddr; struct bio_vec bvec = bio_iter_iovec(bio, bio->bi_iter); /* @@ -1262,9 +1291,8 @@ int btrfs_decompress_buf2page(const char *buf, unsigned long buf_start, PAGE_SIZE - (buf_offset % PAGE_SIZE)); bytes = min(bytes, working_bytes); - kaddr = kmap_atomic(bvec.bv_page); - memcpy(kaddr + bvec.bv_offset, buf + buf_offset, bytes); - kunmap_atomic(kaddr); + memcpy_to_page(bvec.bv_page, bvec.bv_offset, buf + buf_offset, + bytes); flush_dcache_page(bvec.bv_page); buf_offset += bytes; diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c index cc89b63d65a4..34b929bd5c1a 100644 --- a/fs/btrfs/ctree.c +++ b/fs/btrfs/ctree.c @@ -221,9 +221,12 @@ int btrfs_copy_root(struct btrfs_trans_handle *trans, ret = btrfs_inc_ref(trans, root, cow, 1); else ret = btrfs_inc_ref(trans, root, cow, 0); - - if (ret) + if (ret) { + btrfs_tree_unlock(cow); + free_extent_buffer(cow); + btrfs_abort_transaction(trans, ret); return ret; + } btrfs_mark_buffer_dirty(cow); *cow_ret = cow; @@ -1362,7 +1365,9 @@ get_old_root(struct btrfs_root *root, u64 time_seq) "failed to read tree block %llu from get_old_root", logical); } else { + btrfs_tree_read_lock(old); eb = btrfs_clone_extent_buffer(old); + btrfs_tree_read_unlock(old); free_extent_buffer(old); } } else if (old_root) { @@ -1494,6 +1499,7 @@ noinline int btrfs_cow_block(struct btrfs_trans_handle *trans, return ret; } +ALLOW_ERROR_INJECTION(btrfs_cow_block, ERRNO); /* * helper function for defrag to decide if two blocks pointed to by a @@ -2821,6 +2827,7 @@ done: btrfs_release_path(p); return ret; } +ALLOW_ERROR_INJECTION(btrfs_search_slot, ERRNO); /* * Like btrfs_search_slot, this looks for a key in the given tree. It uses the diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h index 4debdbdde2ab..9ae776ab3967 100644 --- a/fs/btrfs/ctree.h +++ b/fs/btrfs/ctree.h @@ -298,7 +298,8 @@ struct btrfs_super_block { BTRFS_FEATURE_INCOMPAT_SKINNY_METADATA | \ BTRFS_FEATURE_INCOMPAT_NO_HOLES | \ BTRFS_FEATURE_INCOMPAT_METADATA_UUID | \ - BTRFS_FEATURE_INCOMPAT_RAID1C34) + BTRFS_FEATURE_INCOMPAT_RAID1C34 | \ + BTRFS_FEATURE_INCOMPAT_ZONED) #define BTRFS_FEATURE_INCOMPAT_SAFE_SET \ (BTRFS_FEATURE_INCOMPAT_EXTENDED_IREF) @@ -523,6 +524,11 @@ struct btrfs_swapfile_pin { * points to a struct btrfs_device. */ bool is_block_group; + /* + * Only used when 'is_block_group' is true and it is the number of + * extents used by a swapfile for this block group ('ptr' field). + */ + int bg_extent_count; }; bool btrfs_pinned_by_swapfile(struct btrfs_fs_info *fs_info, void *ptr); @@ -797,7 +803,7 @@ struct btrfs_fs_info { /* used to keep from writing metadata until there is a nice batch */ struct percpu_counter dirty_metadata_bytes; struct percpu_counter delalloc_bytes; - struct percpu_counter dio_bytes; + struct percpu_counter ordered_bytes; s32 dirty_metadata_batch; s32 delalloc_batch; @@ -933,6 +939,7 @@ struct btrfs_fs_info { /* Used to reclaim the metadata space in the background. */ struct work_struct async_reclaim_work; struct work_struct async_data_reclaim_work; + struct work_struct preempt_reclaim_work; spinlock_t unused_bgs_lock; struct list_head unused_bgs; @@ -974,6 +981,9 @@ struct btrfs_fs_info { /* Max size to emit ZONE_APPEND write command */ u64 max_zone_append_size; + struct mutex zoned_meta_io_lock; + spinlock_t treelog_bg_lock; + u64 treelog_bg; #ifdef CONFIG_BTRFS_FS_REF_VERIFY spinlock_t ref_verify_lock; @@ -1104,7 +1114,7 @@ struct btrfs_root { u32 type; - u64 highest_objectid; + u64 free_objectid; struct btrfs_key defrag_progress; struct btrfs_key defrag_max; @@ -2740,6 +2750,7 @@ enum btrfs_flush_state { ALLOC_CHUNK_FORCE = 8, RUN_DELAYED_IPUTS = 9, COMMIT_TRANS = 10, + FORCE_COMMIT_TRANS = 11, }; int btrfs_subvolume_reserve_metadata(struct btrfs_root *root, @@ -3100,15 +3111,14 @@ int btrfs_truncate_inode_items(struct btrfs_trans_handle *trans, u32 min_type); int btrfs_start_delalloc_snapshot(struct btrfs_root *root); -int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, u64 nr, +int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, long nr, bool in_reclaim_context); int btrfs_set_extent_delalloc(struct btrfs_inode *inode, u64 start, u64 end, unsigned int extra_bits, struct extent_state **cached_state); int btrfs_create_subvol_root(struct btrfs_trans_handle *trans, struct btrfs_root *new_root, - struct btrfs_root *parent_root, - u64 new_dirid); + struct btrfs_root *parent_root); void btrfs_set_delalloc_extent(struct inode *inode, struct extent_state *state, unsigned *bits); void btrfs_clear_delalloc_extent(struct inode *inode, @@ -3119,6 +3129,8 @@ void btrfs_split_delalloc_extent(struct inode *inode, struct extent_state *orig, u64 split); int btrfs_bio_fits_in_stripe(struct page *page, size_t size, struct bio *bio, unsigned long bio_flags); +bool btrfs_bio_fits_in_ordered_extent(struct page *page, struct bio *bio, + unsigned int size); void btrfs_set_range_writeback(struct extent_io_tree *tree, u64 start, u64 end); vm_fault_t btrfs_page_mkwrite(struct vm_fault *vmf); int btrfs_readpage(struct file *file, struct page *page); @@ -3628,7 +3640,8 @@ static inline int __btrfs_fs_compat_ro(struct btrfs_fs_info *fs_info, u64 flag) /* acl.c */ #ifdef CONFIG_BTRFS_FS_POSIX_ACL struct posix_acl *btrfs_get_acl(struct inode *inode, int type); -int btrfs_set_acl(struct inode *inode, struct posix_acl *acl, int type); +int btrfs_set_acl(struct user_namespace *mnt_userns, struct inode *inode, + struct posix_acl *acl, int type); int btrfs_init_acl(struct btrfs_trans_handle *trans, struct inode *inode, struct inode *dir); #else diff --git a/fs/btrfs/delalloc-space.c b/fs/btrfs/delalloc-space.c index bacee09b7bfd..56642ca7af10 100644 --- a/fs/btrfs/delalloc-space.c +++ b/fs/btrfs/delalloc-space.c @@ -191,12 +191,14 @@ void btrfs_free_reserved_data_space(struct btrfs_inode *inode, } /** - * btrfs_inode_rsv_release - release any excessive reservation. - * @inode - the inode we need to release from. - * @qgroup_free - free or convert qgroup meta. - * Unlike normal operation, qgroup meta reservation needs to know if we are - * freeing qgroup reservation or just converting it into per-trans. Normally - * @qgroup_free is true for error handling, and false for normal release. + * Release any excessive reservation + * + * @inode: the inode we need to release from + * @qgroup_free: free or convert qgroup meta. Unlike normal operation, qgroup + * meta reservation needs to know if we are freeing qgroup + * reservation or just converting it into per-trans. Normally + * @qgroup_free is true for error handling, and false for normal + * release. * * This is the same as btrfs_block_rsv_release, except that it handles the * tracepoint for the reservation. @@ -361,7 +363,8 @@ int btrfs_delalloc_reserve_metadata(struct btrfs_inode *inode, u64 num_bytes) } /** - * btrfs_delalloc_release_metadata - release a metadata reservation for an inode + * Release a metadata reservation for an inode + * * @inode: the inode to release the reservation for. * @num_bytes: the number of bytes we are releasing. * @qgroup_free: free qgroup reservation or convert it to per-trans reservation @@ -455,11 +458,13 @@ int btrfs_delalloc_reserve_space(struct btrfs_inode *inode, } /** - * btrfs_delalloc_release_space - release data and metadata space for delalloc - * @inode: inode we're releasing space for - * @start: start position of the space already reserved - * @len: the len of the space already reserved - * @release_bytes: the len of the space we consumed or didn't use + * Release data and metadata space for delalloc + * + * @inode: inode we're releasing space for + * @reserved: list of changed/reserved ranges + * @start: start position of the space already reserved + * @len: length of the space already reserved + * @qgroup_free: should qgroup reserved-space also be freed * * This function will release the metadata space that was not used and will * decrement ->delalloc_bytes and remove it from the fs_info delalloc_inodes diff --git a/fs/btrfs/delayed-inode.c b/fs/btrfs/delayed-inode.c index 70c0340d839c..bf25401c9768 100644 --- a/fs/btrfs/delayed-inode.c +++ b/fs/btrfs/delayed-inode.c @@ -627,7 +627,8 @@ static int btrfs_delayed_inode_reserve_metadata( */ if (!src_rsv || (!trans->bytes_reserved && src_rsv->type != BTRFS_BLOCK_RSV_DELALLOC)) { - ret = btrfs_qgroup_reserve_meta_prealloc(root, num_bytes, true); + ret = btrfs_qgroup_reserve_meta(root, num_bytes, + BTRFS_QGROUP_RSV_META_PREALLOC, true); if (ret < 0) return ret; ret = btrfs_block_rsv_add(root, dst_rsv, num_bytes, @@ -649,7 +650,7 @@ static int btrfs_delayed_inode_reserve_metadata( btrfs_ino(inode), num_bytes, 1); } else { - btrfs_qgroup_free_meta_prealloc(root, fs_info->nodesize); + btrfs_qgroup_free_meta_prealloc(root, num_bytes); } return ret; } @@ -1154,7 +1155,7 @@ static int __btrfs_run_delayed_items(struct btrfs_trans_handle *trans, int nr) delayed_root = fs_info->delayed_root; curr_node = btrfs_first_delayed_node(delayed_root); - while (curr_node && (!count || (count && nr--))) { + while (curr_node && (!count || nr--)) { ret = __btrfs_commit_inode_delayed_items(trans, path, curr_node); if (ret) { diff --git a/fs/btrfs/delayed-ref.c b/fs/btrfs/delayed-ref.c index 353cc2994d10..63be7d01a9a3 100644 --- a/fs/btrfs/delayed-ref.c +++ b/fs/btrfs/delayed-ref.c @@ -69,9 +69,10 @@ int btrfs_should_throttle_delayed_refs(struct btrfs_trans_handle *trans) } /** - * btrfs_delayed_refs_rsv_release - release a ref head's reservation. - * @fs_info - the fs_info for our fs. - * @nr - the number of items to drop. + * Release a ref head's reservation + * + * @fs_info: the filesystem + * @nr: number of items to drop * * This drops the delayed ref head's count from the delayed refs rsv and frees * any excess reservation we had. @@ -114,10 +115,11 @@ void btrfs_update_delayed_refs_rsv(struct btrfs_trans_handle *trans) } /** - * btrfs_migrate_to_delayed_refs_rsv - transfer bytes to our delayed refs rsv. - * @fs_info - the fs info for our fs. - * @src - the source block rsv to transfer from. - * @num_bytes - the number of bytes to transfer. + * Transfer bytes to our delayed refs rsv + * + * @fs_info: the filesystem + * @src: source block rsv to transfer from + * @num_bytes: number of bytes to transfer * * This transfers up to the num_bytes amount from the src rsv to the * delayed_refs_rsv. Any extra bytes are returned to the space info. @@ -162,9 +164,10 @@ void btrfs_migrate_to_delayed_refs_rsv(struct btrfs_fs_info *fs_info, } /** - * btrfs_delayed_refs_rsv_refill - refill based on our delayed refs usage. - * @fs_info - the fs_info for our fs. - * @flush - control how we can flush for this reservation. + * Refill based on our delayed refs usage + * + * @fs_info: the filesystem + * @flush: control how we can flush for this reservation. * * This will refill the delayed block_rsv up to 1 items size worth of space and * will return -ENOSPC if we can't make the reservation. @@ -648,12 +651,12 @@ inserted: */ static noinline void update_existing_head_ref(struct btrfs_trans_handle *trans, struct btrfs_delayed_ref_head *existing, - struct btrfs_delayed_ref_head *update, - int *old_ref_mod_ret) + struct btrfs_delayed_ref_head *update) { struct btrfs_delayed_ref_root *delayed_refs = &trans->transaction->delayed_refs; struct btrfs_fs_info *fs_info = trans->fs_info; + u64 flags = btrfs_ref_head_to_space_flags(existing); int old_ref_mod; BUG_ON(existing->is_data != update->is_data); @@ -701,8 +704,6 @@ static noinline void update_existing_head_ref(struct btrfs_trans_handle *trans, * currently, for refs we just added we know we're a-ok. */ old_ref_mod = existing->total_ref_mod; - if (old_ref_mod_ret) - *old_ref_mod_ret = old_ref_mod; existing->ref_mod += update->ref_mod; existing->total_ref_mod += update->ref_mod; @@ -724,6 +725,27 @@ static noinline void update_existing_head_ref(struct btrfs_trans_handle *trans, trans->delayed_ref_updates += csum_leaves; } } + + /* + * This handles the following conditions: + * + * 1. We had a ref mod of 0 or more and went negative, indicating that + * we may be freeing space, so add our space to the + * total_bytes_pinned counter. + * 2. We were negative and went to 0 or positive, so no longer can say + * that the space would be pinned, decrement our counter from the + * total_bytes_pinned counter. + * 3. We are now at 0 and have ->must_insert_reserved set, which means + * this was a new allocation and then we dropped it, and thus must + * add our space to the total_bytes_pinned counter. + */ + if (existing->total_ref_mod < 0 && old_ref_mod >= 0) + btrfs_mod_total_bytes_pinned(fs_info, flags, existing->num_bytes); + else if (existing->total_ref_mod >= 0 && old_ref_mod < 0) + btrfs_mod_total_bytes_pinned(fs_info, flags, -existing->num_bytes); + else if (existing->total_ref_mod == 0 && existing->must_insert_reserved) + btrfs_mod_total_bytes_pinned(fs_info, flags, existing->num_bytes); + spin_unlock(&existing->lock); } @@ -798,8 +820,7 @@ static noinline struct btrfs_delayed_ref_head * add_delayed_ref_head(struct btrfs_trans_handle *trans, struct btrfs_delayed_ref_head *head_ref, struct btrfs_qgroup_extent_record *qrecord, - int action, int *qrecord_inserted_ret, - int *old_ref_mod, int *new_ref_mod) + int action, int *qrecord_inserted_ret) { struct btrfs_delayed_ref_head *existing; struct btrfs_delayed_ref_root *delayed_refs; @@ -821,8 +842,7 @@ add_delayed_ref_head(struct btrfs_trans_handle *trans, existing = htree_insert(&delayed_refs->href_root, &head_ref->href_node); if (existing) { - update_existing_head_ref(trans, existing, head_ref, - old_ref_mod); + update_existing_head_ref(trans, existing, head_ref); /* * we've updated the existing ref, free the newly * allocated ref @@ -830,14 +850,17 @@ add_delayed_ref_head(struct btrfs_trans_handle *trans, kmem_cache_free(btrfs_delayed_ref_head_cachep, head_ref); head_ref = existing; } else { - if (old_ref_mod) - *old_ref_mod = 0; + u64 flags = btrfs_ref_head_to_space_flags(head_ref); + if (head_ref->is_data && head_ref->ref_mod < 0) { delayed_refs->pending_csums += head_ref->num_bytes; trans->delayed_ref_updates += btrfs_csum_bytes_to_leaves(trans->fs_info, head_ref->num_bytes); } + if (head_ref->ref_mod < 0) + btrfs_mod_total_bytes_pinned(trans->fs_info, flags, + head_ref->num_bytes); delayed_refs->num_heads++; delayed_refs->num_heads_ready++; atomic_inc(&delayed_refs->num_entries); @@ -845,8 +868,6 @@ add_delayed_ref_head(struct btrfs_trans_handle *trans, } if (qrecord_inserted_ret) *qrecord_inserted_ret = qrecord_inserted; - if (new_ref_mod) - *new_ref_mod = head_ref->total_ref_mod; return head_ref; } @@ -909,8 +930,7 @@ static void init_delayed_ref_common(struct btrfs_fs_info *fs_info, */ int btrfs_add_delayed_tree_ref(struct btrfs_trans_handle *trans, struct btrfs_ref *generic_ref, - struct btrfs_delayed_extent_op *extent_op, - int *old_ref_mod, int *new_ref_mod) + struct btrfs_delayed_extent_op *extent_op) { struct btrfs_fs_info *fs_info = trans->fs_info; struct btrfs_delayed_tree_ref *ref; @@ -977,8 +997,7 @@ int btrfs_add_delayed_tree_ref(struct btrfs_trans_handle *trans, * the spin lock */ head_ref = add_delayed_ref_head(trans, head_ref, record, - action, &qrecord_inserted, - old_ref_mod, new_ref_mod); + action, &qrecord_inserted); ret = insert_delayed_ref(trans, delayed_refs, head_ref, &ref->node); spin_unlock(&delayed_refs->lock); @@ -1006,8 +1025,7 @@ int btrfs_add_delayed_tree_ref(struct btrfs_trans_handle *trans, */ int btrfs_add_delayed_data_ref(struct btrfs_trans_handle *trans, struct btrfs_ref *generic_ref, - u64 reserved, int *old_ref_mod, - int *new_ref_mod) + u64 reserved) { struct btrfs_fs_info *fs_info = trans->fs_info; struct btrfs_delayed_data_ref *ref; @@ -1073,8 +1091,7 @@ int btrfs_add_delayed_data_ref(struct btrfs_trans_handle *trans, * the spin lock */ head_ref = add_delayed_ref_head(trans, head_ref, record, - action, &qrecord_inserted, - old_ref_mod, new_ref_mod); + action, &qrecord_inserted); ret = insert_delayed_ref(trans, delayed_refs, head_ref, &ref->node); spin_unlock(&delayed_refs->lock); @@ -1117,7 +1134,7 @@ int btrfs_add_delayed_extent_op(struct btrfs_trans_handle *trans, spin_lock(&delayed_refs->lock); add_delayed_ref_head(trans, head_ref, NULL, BTRFS_UPDATE_DELAYED_HEAD, - NULL, NULL, NULL); + NULL); spin_unlock(&delayed_refs->lock); diff --git a/fs/btrfs/delayed-ref.h b/fs/btrfs/delayed-ref.h index 1c977e6d45dc..e22fba272e4f 100644 --- a/fs/btrfs/delayed-ref.h +++ b/fs/btrfs/delayed-ref.h @@ -135,6 +135,11 @@ struct btrfs_delayed_data_ref { u64 offset; }; +enum btrfs_delayed_ref_flags { + /* Indicate that we are flushing delayed refs for the commit */ + BTRFS_DELAYED_REFS_FLUSHING, +}; + struct btrfs_delayed_ref_root { /* head ref rbtree */ struct rb_root_cached href_root; @@ -158,12 +163,7 @@ struct btrfs_delayed_ref_root { u64 pending_csums; - /* - * set when the tree is flushing before a transaction commit, - * used by the throttling code to decide if new updates need - * to be run right away - */ - int flushing; + unsigned long flags; u64 run_delayed_start; @@ -326,6 +326,16 @@ static inline void btrfs_put_delayed_ref(struct btrfs_delayed_ref_node *ref) } } +static inline u64 btrfs_ref_head_to_space_flags( + struct btrfs_delayed_ref_head *head_ref) +{ + if (head_ref->is_data) + return BTRFS_BLOCK_GROUP_DATA; + else if (head_ref->is_system) + return BTRFS_BLOCK_GROUP_SYSTEM; + return BTRFS_BLOCK_GROUP_METADATA; +} + static inline void btrfs_put_delayed_ref_head(struct btrfs_delayed_ref_head *head) { if (refcount_dec_and_test(&head->refs)) @@ -334,12 +344,10 @@ static inline void btrfs_put_delayed_ref_head(struct btrfs_delayed_ref_head *hea int btrfs_add_delayed_tree_ref(struct btrfs_trans_handle *trans, struct btrfs_ref *generic_ref, - struct btrfs_delayed_extent_op *extent_op, - int *old_ref_mod, int *new_ref_mod); + struct btrfs_delayed_extent_op *extent_op); int btrfs_add_delayed_data_ref(struct btrfs_trans_handle *trans, struct btrfs_ref *generic_ref, - u64 reserved, int *old_ref_mod, - int *new_ref_mod); + u64 reserved); int btrfs_add_delayed_extent_op(struct btrfs_trans_handle *trans, u64 bytenr, u64 num_bytes, struct btrfs_delayed_extent_op *extent_op); diff --git a/fs/btrfs/dev-replace.c b/fs/btrfs/dev-replace.c index 324f646d6e5e..3a9c1e046ebe 100644 --- a/fs/btrfs/dev-replace.c +++ b/fs/btrfs/dev-replace.c @@ -22,6 +22,7 @@ #include "dev-replace.h" #include "sysfs.h" #include "zoned.h" +#include "block-group.h" /* * Device replace overview @@ -459,6 +460,185 @@ static char* btrfs_dev_name(struct btrfs_device *device) return rcu_str_deref(device->name); } +static int mark_block_group_to_copy(struct btrfs_fs_info *fs_info, + struct btrfs_device *src_dev) +{ + struct btrfs_path *path; + struct btrfs_key key; + struct btrfs_key found_key; + struct btrfs_root *root = fs_info->dev_root; + struct btrfs_dev_extent *dev_extent = NULL; + struct btrfs_block_group *cache; + struct btrfs_trans_handle *trans; + int ret = 0; + u64 chunk_offset; + + /* Do not use "to_copy" on non zoned filesystem for now */ + if (!btrfs_is_zoned(fs_info)) + return 0; + + mutex_lock(&fs_info->chunk_mutex); + + /* Ensure we don't have pending new block group */ + spin_lock(&fs_info->trans_lock); + while (fs_info->running_transaction && + !list_empty(&fs_info->running_transaction->dev_update_list)) { + spin_unlock(&fs_info->trans_lock); + mutex_unlock(&fs_info->chunk_mutex); + trans = btrfs_attach_transaction(root); + if (IS_ERR(trans)) { + ret = PTR_ERR(trans); + mutex_lock(&fs_info->chunk_mutex); + if (ret == -ENOENT) { + spin_lock(&fs_info->trans_lock); + continue; + } else { + goto unlock; + } + } + + ret = btrfs_commit_transaction(trans); + mutex_lock(&fs_info->chunk_mutex); + if (ret) + goto unlock; + + spin_lock(&fs_info->trans_lock); + } + spin_unlock(&fs_info->trans_lock); + + path = btrfs_alloc_path(); + if (!path) { + ret = -ENOMEM; + goto unlock; + } + + path->reada = READA_FORWARD; + path->search_commit_root = 1; + path->skip_locking = 1; + + key.objectid = src_dev->devid; + key.type = BTRFS_DEV_EXTENT_KEY; + key.offset = 0; + + ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); + if (ret < 0) + goto free_path; + if (ret > 0) { + if (path->slots[0] >= + btrfs_header_nritems(path->nodes[0])) { + ret = btrfs_next_leaf(root, path); + if (ret < 0) + goto free_path; + if (ret > 0) { + ret = 0; + goto free_path; + } + } else { + ret = 0; + } + } + + while (1) { + struct extent_buffer *leaf = path->nodes[0]; + int slot = path->slots[0]; + + btrfs_item_key_to_cpu(leaf, &found_key, slot); + + if (found_key.objectid != src_dev->devid) + break; + + if (found_key.type != BTRFS_DEV_EXTENT_KEY) + break; + + if (found_key.offset < key.offset) + break; + + dev_extent = btrfs_item_ptr(leaf, slot, struct btrfs_dev_extent); + + chunk_offset = btrfs_dev_extent_chunk_offset(leaf, dev_extent); + + cache = btrfs_lookup_block_group(fs_info, chunk_offset); + if (!cache) + goto skip; + + spin_lock(&cache->lock); + cache->to_copy = 1; + spin_unlock(&cache->lock); + + btrfs_put_block_group(cache); + +skip: + ret = btrfs_next_item(root, path); + if (ret != 0) { + if (ret > 0) + ret = 0; + break; + } + } + +free_path: + btrfs_free_path(path); +unlock: + mutex_unlock(&fs_info->chunk_mutex); + + return ret; +} + +bool btrfs_finish_block_group_to_copy(struct btrfs_device *srcdev, + struct btrfs_block_group *cache, + u64 physical) +{ + struct btrfs_fs_info *fs_info = cache->fs_info; + struct extent_map *em; + struct map_lookup *map; + u64 chunk_offset = cache->start; + int num_extents, cur_extent; + int i; + + /* Do not use "to_copy" on non zoned filesystem for now */ + if (!btrfs_is_zoned(fs_info)) + return true; + + spin_lock(&cache->lock); + if (cache->removed) { + spin_unlock(&cache->lock); + return true; + } + spin_unlock(&cache->lock); + + em = btrfs_get_chunk_map(fs_info, chunk_offset, 1); + ASSERT(!IS_ERR(em)); + map = em->map_lookup; + + num_extents = cur_extent = 0; + for (i = 0; i < map->num_stripes; i++) { + /* We have more device extent to copy */ + if (srcdev != map->stripes[i].dev) + continue; + + num_extents++; + if (physical == map->stripes[i].physical) + cur_extent = i; + } + + free_extent_map(em); + + if (num_extents > 1 && cur_extent < num_extents - 1) { + /* + * Has more stripes on this device. Keep this block group + * readonly until we finish all the stripes. + */ + return false; + } + + /* Last stripe on this device */ + spin_lock(&cache->lock); + cache->to_copy = 0; + spin_unlock(&cache->lock); + + return true; +} + static int btrfs_dev_replace_start(struct btrfs_fs_info *fs_info, const char *tgtdev_name, u64 srcdevid, const char *srcdev_name, int read_src) @@ -500,6 +680,10 @@ static int btrfs_dev_replace_start(struct btrfs_fs_info *fs_info, if (ret) return ret; + ret = mark_block_group_to_copy(fs_info, src_device); + if (ret) + return ret; + down_write(&dev_replace->rwsem); switch (dev_replace->replace_state) { case BTRFS_IOCTL_DEV_REPLACE_STATE_NEVER_STARTED: @@ -715,7 +899,7 @@ static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info, * flush all outstanding I/O and inode extent mappings before the * copy operation is declared as being finished */ - ret = btrfs_start_delalloc_roots(fs_info, U64_MAX, false); + ret = btrfs_start_delalloc_roots(fs_info, LONG_MAX, false); if (ret) { mutex_unlock(&dev_replace->lock_finishing_cancel_unmount); return ret; diff --git a/fs/btrfs/dev-replace.h b/fs/btrfs/dev-replace.h index 60b70dacc299..3911049a5f23 100644 --- a/fs/btrfs/dev-replace.h +++ b/fs/btrfs/dev-replace.h @@ -18,5 +18,8 @@ int btrfs_dev_replace_cancel(struct btrfs_fs_info *fs_info); void btrfs_dev_replace_suspend_for_unmount(struct btrfs_fs_info *fs_info); int btrfs_resume_dev_replace_async(struct btrfs_fs_info *fs_info); int __pure btrfs_dev_replace_is_ongoing(struct btrfs_dev_replace *dev_replace); +bool btrfs_finish_block_group_to_copy(struct btrfs_device *srcdev, + struct btrfs_block_group *cache, + u64 physical); #endif diff --git a/fs/btrfs/discard.c b/fs/btrfs/discard.c index 2b8383d41144..306ff20af70f 100644 --- a/fs/btrfs/discard.c +++ b/fs/btrfs/discard.c @@ -185,10 +185,12 @@ static struct btrfs_block_group *find_next_block_group( } /** - * peek_discard_list - wrap find_next_block_group() - * @discard_ctl: discard control + * Wrap find_next_block_group() + * + * @discard_ctl: discard control * @discard_state: the discard_state of the block_group after state management * @discard_index: the discard_index of the block_group after state management + * @now: time when discard was invoked, in ns * * This wraps find_next_block_group() and sets the block_group to be in use. * discard_state's control flow is managed here. Variables related to diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 07a2b4f69b10..41b718cfea40 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -459,6 +459,12 @@ static int csum_dirty_buffer(struct btrfs_fs_info *fs_info, struct bio_vec *bvec return 0; found_start = btrfs_header_bytenr(eb); + + if (test_bit(EXTENT_BUFFER_NO_CHECK, &eb->bflags)) { + WARN_ON(found_start != 0); + return 0; + } + /* * Please do not consolidate these warnings into a single if. * It is useful to know what went wrong. @@ -591,6 +597,59 @@ out: return ret; } +static int validate_subpage_buffer(struct page *page, u64 start, u64 end, + int mirror) +{ + struct btrfs_fs_info *fs_info = btrfs_sb(page->mapping->host->i_sb); + struct extent_buffer *eb; + bool reads_done; + int ret = 0; + + /* + * We don't allow bio merge for subpage metadata read, so we should + * only get one eb for each endio hook. + */ + ASSERT(end == start + fs_info->nodesize - 1); + ASSERT(PagePrivate(page)); + + eb = find_extent_buffer(fs_info, start); + /* + * When we are reading one tree block, eb must have been inserted into + * the radix tree. If not, something is wrong. + */ + ASSERT(eb); + + reads_done = atomic_dec_and_test(&eb->io_pages); + /* Subpage read must finish in page read */ + ASSERT(reads_done); + + eb->read_mirror = mirror; + if (test_bit(EXTENT_BUFFER_READ_ERR, &eb->bflags)) { + ret = -EIO; + goto err; + } + ret = validate_extent_buffer(eb); + if (ret < 0) + goto err; + + if (test_and_clear_bit(EXTENT_BUFFER_READAHEAD, &eb->bflags)) + btree_readahead_hook(eb, ret); + + set_extent_buffer_uptodate(eb); + + free_extent_buffer(eb); + return ret; +err: + /* + * end_bio_extent_readpage decrements io_pages in case of error, + * make sure it has something to decrement. + */ + atomic_inc(&eb->io_pages); + clear_extent_buffer_uptodate(eb); + free_extent_buffer(eb); + return ret; +} + int btrfs_validate_metadata_buffer(struct btrfs_io_bio *io_bio, struct page *page, u64 start, u64 end, int mirror) @@ -600,6 +659,10 @@ int btrfs_validate_metadata_buffer(struct btrfs_io_bio *io_bio, int reads_done; ASSERT(page->private); + + if (btrfs_sb(page->mapping->host->i_sb)->sectorsize < PAGE_SIZE) + return validate_subpage_buffer(page, start, end, mirror); + eb = (struct extent_buffer *)page->private; /* @@ -646,7 +709,7 @@ static void end_workqueue_bio(struct bio *bio) fs_info = end_io_wq->info; end_io_wq->status = bio->bi_status; - if (bio_op(bio) == REQ_OP_WRITE) { + if (btrfs_op(bio) == BTRFS_MAP_WRITE) { if (end_io_wq->metadata == BTRFS_WQ_ENDIO_METADATA) wq = fs_info->endio_meta_write_workers; else if (end_io_wq->metadata == BTRFS_WQ_ENDIO_FREE_SPACE) @@ -808,6 +871,8 @@ static blk_status_t btree_submit_bio_start(struct inode *inode, struct bio *bio, static int check_async_write(struct btrfs_fs_info *fs_info, struct btrfs_inode *bi) { + if (btrfs_is_zoned(fs_info)) + return 0; if (atomic_read(&bi->sync_writers)) return 0; if (test_bit(BTRFS_FS_CSUM_IMPL_FAST, &fs_info->flags)) @@ -822,7 +887,7 @@ blk_status_t btrfs_submit_metadata_bio(struct inode *inode, struct bio *bio, int async = check_async_write(fs_info, BTRFS_I(inode)); blk_status_t ret; - if (bio_op(bio) != REQ_OP_WRITE) { + if (btrfs_op(bio) != BTRFS_MAP_WRITE) { /* * called for a read, do the setup so that checksum validation * can happen in the async kernel threads @@ -1016,7 +1081,7 @@ static void __setup_root(struct btrfs_root *root, struct btrfs_fs_info *fs_info, root->orphan_cleanup_state = 0; root->last_trans = 0; - root->highest_objectid = 0; + root->free_objectid = 0; root->nr_delalloc_inodes = 0; root->nr_ordered_extents = 0; root->inode_tree = RB_ROOT; @@ -1189,7 +1254,6 @@ static struct btrfs_root *alloc_log_tree(struct btrfs_trans_handle *trans, struct btrfs_fs_info *fs_info) { struct btrfs_root *root; - struct extent_buffer *leaf; root = btrfs_alloc_root(fs_info, BTRFS_TREE_LOG_OBJECTID, GFP_NOFS); if (!root) @@ -1199,6 +1263,14 @@ static struct btrfs_root *alloc_log_tree(struct btrfs_trans_handle *trans, root->root_key.type = BTRFS_ROOT_ITEM_KEY; root->root_key.offset = BTRFS_TREE_LOG_OBJECTID; + return root; +} + +int btrfs_alloc_log_tree_node(struct btrfs_trans_handle *trans, + struct btrfs_root *root) +{ + struct extent_buffer *leaf; + /* * DON'T set SHAREABLE bit for log trees. * @@ -1211,16 +1283,15 @@ static struct btrfs_root *alloc_log_tree(struct btrfs_trans_handle *trans, leaf = btrfs_alloc_tree_block(trans, root, 0, BTRFS_TREE_LOG_OBJECTID, NULL, 0, 0, 0, BTRFS_NESTING_NORMAL); - if (IS_ERR(leaf)) { - btrfs_put_root(root); - return ERR_CAST(leaf); - } + if (IS_ERR(leaf)) + return PTR_ERR(leaf); root->node = leaf; btrfs_mark_buffer_dirty(root->node); btrfs_tree_unlock(root->node); - return root; + + return 0; } int btrfs_init_log_root_tree(struct btrfs_trans_handle *trans, @@ -1231,6 +1302,16 @@ int btrfs_init_log_root_tree(struct btrfs_trans_handle *trans, log_root = alloc_log_tree(trans, fs_info); if (IS_ERR(log_root)) return PTR_ERR(log_root); + + if (!btrfs_is_zoned(fs_info)) { + int ret = btrfs_alloc_log_tree_node(trans, log_root); + + if (ret) { + btrfs_put_root(log_root); + return ret; + } + } + WARN_ON(fs_info->log_root_tree); fs_info->log_root_tree = log_root; return 0; @@ -1242,11 +1323,18 @@ int btrfs_add_log_tree(struct btrfs_trans_handle *trans, struct btrfs_fs_info *fs_info = root->fs_info; struct btrfs_root *log_root; struct btrfs_inode_item *inode_item; + int ret; log_root = alloc_log_tree(trans, fs_info); if (IS_ERR(log_root)) return PTR_ERR(log_root); + ret = btrfs_alloc_log_tree_node(trans, log_root); + if (ret) { + btrfs_put_root(log_root); + return ret; + } + log_root->last_trans = trans->transid; log_root->root_key.offset = root->root_key.objectid; @@ -1367,14 +1455,13 @@ static int btrfs_init_fs_root(struct btrfs_root *root, dev_t anon_dev) } mutex_lock(&root->objectid_mutex); - ret = btrfs_find_highest_objectid(root, - &root->highest_objectid); + ret = btrfs_init_root_free_objectid(root); if (ret) { mutex_unlock(&root->objectid_mutex); goto fail; } - ASSERT(root->highest_objectid <= BTRFS_LAST_FREE_OBJECTID); + ASSERT(root->free_objectid <= BTRFS_LAST_FREE_OBJECTID); mutex_unlock(&root->objectid_mutex); @@ -1470,7 +1557,7 @@ void btrfs_free_fs_info(struct btrfs_fs_info *fs_info) { percpu_counter_destroy(&fs_info->dirty_metadata_bytes); percpu_counter_destroy(&fs_info->delalloc_bytes); - percpu_counter_destroy(&fs_info->dio_bytes); + percpu_counter_destroy(&fs_info->ordered_bytes); percpu_counter_destroy(&fs_info->dev_replace.bio_counter); btrfs_free_csum_hash(fs_info); btrfs_free_stripe_hash_table(fs_info); @@ -2427,13 +2514,21 @@ static int validate_super(struct btrfs_fs_info *fs_info, btrfs_err(fs_info, "invalid sectorsize %llu", sectorsize); ret = -EINVAL; } - /* Only PAGE SIZE is supported yet */ - if (sectorsize != PAGE_SIZE) { + + /* + * For 4K page size, we only support 4K sector size. + * For 64K page size, we support read-write for 64K sector size, and + * read-only for 4K sector size. + */ + if ((PAGE_SIZE == SZ_4K && sectorsize != PAGE_SIZE) || + (PAGE_SIZE == SZ_64K && (sectorsize != SZ_4K && + sectorsize != SZ_64K))) { btrfs_err(fs_info, - "sectorsize %llu not supported yet, only support %lu", + "sectorsize %llu not yet supported for page size %lu", sectorsize, PAGE_SIZE); ret = -EINVAL; } + if (!is_power_of_2(nodesize) || nodesize < sectorsize || nodesize > BTRFS_MAX_METADATA_BLOCKSIZE) { btrfs_err(fs_info, "invalid nodesize %llu", nodesize); @@ -2646,14 +2741,13 @@ static int __cold init_tree_roots(struct btrfs_fs_info *fs_info) * No need to hold btrfs_root::objectid_mutex since the fs * hasn't been fully initialised and we are the only user */ - ret = btrfs_find_highest_objectid(tree_root, - &tree_root->highest_objectid); + ret = btrfs_init_root_free_objectid(tree_root); if (ret < 0) { handle_error = true; continue; } - ASSERT(tree_root->highest_objectid <= BTRFS_LAST_FREE_OBJECTID); + ASSERT(tree_root->free_objectid <= BTRFS_LAST_FREE_OBJECTID); ret = btrfs_read_roots(fs_info); if (ret < 0) { @@ -2695,11 +2789,13 @@ void btrfs_init_fs_info(struct btrfs_fs_info *fs_info) spin_lock_init(&fs_info->super_lock); spin_lock_init(&fs_info->buffer_lock); spin_lock_init(&fs_info->unused_bgs_lock); + spin_lock_init(&fs_info->treelog_bg_lock); rwlock_init(&fs_info->tree_mod_log_lock); mutex_init(&fs_info->unused_bg_unpin_mutex); mutex_init(&fs_info->delete_unused_bgs_mutex); mutex_init(&fs_info->reloc_mutex); mutex_init(&fs_info->delalloc_root_mutex); + mutex_init(&fs_info->zoned_meta_io_lock); seqlock_init(&fs_info->profiles_lock); INIT_LIST_HEAD(&fs_info->dirty_cowonly_roots); @@ -2804,7 +2900,7 @@ static int init_mount_fs_info(struct btrfs_fs_info *fs_info, struct super_block sb->s_blocksize = BTRFS_BDEV_BLOCKSIZE; sb->s_blocksize_bits = blksize_bits(BTRFS_BDEV_BLOCKSIZE); - ret = percpu_counter_init(&fs_info->dio_bytes, 0, GFP_KERNEL); + ret = percpu_counter_init(&fs_info->ordered_bytes, 0, GFP_KERNEL); if (ret) return ret; @@ -3140,8 +3236,6 @@ int __cold open_ctree(struct super_block *sb, struct btrfs_fs_devices *fs_device if (features & BTRFS_FEATURE_INCOMPAT_SKINNY_METADATA) btrfs_info(fs_info, "has skinny extents"); - fs_info->zoned = (features & BTRFS_FEATURE_INCOMPAT_ZONED); - /* * flag our filesystem as having big metadata blocks if * they are bigger than the page size @@ -3194,6 +3288,17 @@ int __cold open_ctree(struct super_block *sb, struct btrfs_fs_devices *fs_device goto fail_alloc; } + /* For 4K sector size support, it's only read-only */ + if (PAGE_SIZE == SZ_64K && sectorsize == SZ_4K) { + if (!sb_rdonly(sb) || btrfs_super_log_root(disk_super)) { + btrfs_err(fs_info, + "subpage sectorsize %u only supported read-only for page size %lu", + sectorsize, PAGE_SIZE); + err = -EINVAL; + goto fail_alloc; + } + } + ret = btrfs_init_workqueues(fs_info, fs_devices); if (ret) { err = ret; @@ -3261,6 +3366,19 @@ int __cold open_ctree(struct super_block *sb, struct btrfs_fs_devices *fs_device goto fail_tree_roots; /* + * Get zone type information of zoned block devices. This will also + * handle emulation of a zoned filesystem if a regular device has the + * zoned incompat feature flag set. + */ + ret = btrfs_get_dev_zone_info_all_devices(fs_info); + if (ret) { + btrfs_err(fs_info, + "zoned: failed to read device zone info: %d", + ret); + goto fail_block_groups; + } + + /* * If we have a uuid root and we're not being told to rescan we need to * check the generation here so we can set the * BTRFS_FS_UPDATE_UUID_TREE_GEN bit. Otherwise we could commit the @@ -4114,6 +4232,7 @@ void __cold close_ctree(struct btrfs_fs_info *fs_info) cancel_work_sync(&fs_info->async_reclaim_work); cancel_work_sync(&fs_info->async_data_reclaim_work); + cancel_work_sync(&fs_info->preempt_reclaim_work); /* Cancel or finish ongoing discard work */ btrfs_discard_cleanup(fs_info); @@ -4166,9 +4285,9 @@ void __cold close_ctree(struct btrfs_fs_info *fs_info) percpu_counter_sum(&fs_info->delalloc_bytes)); } - if (percpu_counter_sum(&fs_info->dio_bytes)) + if (percpu_counter_sum(&fs_info->ordered_bytes)) btrfs_info(fs_info, "at unmount dio bytes count %lld", - percpu_counter_sum(&fs_info->dio_bytes)); + percpu_counter_sum(&fs_info->ordered_bytes)); btrfs_sysfs_remove_mounted(fs_info); btrfs_sysfs_remove_fsid(fs_info->fs_devices); @@ -4689,6 +4808,8 @@ void btrfs_cleanup_one_transaction(struct btrfs_transaction *cur_trans, EXTENT_DIRTY); btrfs_destroy_pinned_extent(fs_info, &cur_trans->pinned_extents); + btrfs_free_redirty_list(cur_trans); + cur_trans->state =TRANS_STATE_COMPLETED; wake_up(&cur_trans->commit_wait); } @@ -4746,7 +4867,7 @@ static int btrfs_cleanup_transaction(struct btrfs_fs_info *fs_info) return 0; } -int btrfs_find_highest_objectid(struct btrfs_root *root, u64 *objectid) +int btrfs_init_root_free_objectid(struct btrfs_root *root) { struct btrfs_path *path; int ret; @@ -4770,10 +4891,10 @@ int btrfs_find_highest_objectid(struct btrfs_root *root, u64 *objectid) slot = path->slots[0] - 1; l = path->nodes[0]; btrfs_item_key_to_cpu(l, &found_key, slot); - *objectid = max_t(u64, found_key.objectid, - BTRFS_FIRST_FREE_OBJECTID - 1); + root->free_objectid = max_t(u64, found_key.objectid + 1, + BTRFS_FIRST_FREE_OBJECTID); } else { - *objectid = BTRFS_FIRST_FREE_OBJECTID - 1; + root->free_objectid = BTRFS_FIRST_FREE_OBJECTID; } ret = 0; error: @@ -4781,12 +4902,12 @@ error: return ret; } -int btrfs_find_free_objectid(struct btrfs_root *root, u64 *objectid) +int btrfs_get_free_objectid(struct btrfs_root *root, u64 *objectid) { int ret; mutex_lock(&root->objectid_mutex); - if (unlikely(root->highest_objectid >= BTRFS_LAST_FREE_OBJECTID)) { + if (unlikely(root->free_objectid >= BTRFS_LAST_FREE_OBJECTID)) { btrfs_warn(root->fs_info, "the objectid of root %llu reaches its highest value", root->root_key.objectid); @@ -4794,7 +4915,7 @@ int btrfs_find_free_objectid(struct btrfs_root *root, u64 *objectid) goto out; } - *objectid = ++root->highest_objectid; + *objectid = root->free_objectid++; ret = 0; out: mutex_unlock(&root->objectid_mutex); diff --git a/fs/btrfs/disk-io.h b/fs/btrfs/disk-io.h index e45057c0c016..0e7e9526b6a8 100644 --- a/fs/btrfs/disk-io.h +++ b/fs/btrfs/disk-io.h @@ -120,6 +120,8 @@ blk_status_t btrfs_wq_submit_bio(struct inode *inode, struct bio *bio, extent_submit_bio_start_t *submit_bio_start); blk_status_t btrfs_submit_bio_done(void *private_data, struct bio *bio, int mirror_num); +int btrfs_alloc_log_tree_node(struct btrfs_trans_handle *trans, + struct btrfs_root *root); int btrfs_init_log_root_tree(struct btrfs_trans_handle *trans, struct btrfs_fs_info *fs_info); int btrfs_add_log_tree(struct btrfs_trans_handle *trans, @@ -133,8 +135,8 @@ struct btrfs_root *btrfs_create_tree(struct btrfs_trans_handle *trans, int btree_lock_page_hook(struct page *page, void *data, void (*flush_fn)(void *)); int btrfs_get_num_tolerated_disk_barrier_failures(u64 flags); -int btrfs_find_free_objectid(struct btrfs_root *root, u64 *objectid); -int btrfs_find_highest_objectid(struct btrfs_root *root, u64 *objectid); +int btrfs_get_free_objectid(struct btrfs_root *root, u64 *objectid); +int btrfs_init_root_free_objectid(struct btrfs_root *root); int __init btrfs_end_io_wq_init(void); void __cold btrfs_end_io_wq_exit(void); diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 0c335dae5af7..36a3c973fda1 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -34,6 +34,8 @@ #include "block-group.h" #include "discard.h" #include "rcu-string.h" +#include "zoned.h" +#include "dev-replace.h" #undef SCRAMBLE_DELAYED_REFS @@ -82,41 +84,6 @@ void btrfs_free_excluded_extents(struct btrfs_block_group *cache) EXTENT_UPTODATE); } -static u64 generic_ref_to_space_flags(struct btrfs_ref *ref) -{ - if (ref->type == BTRFS_REF_METADATA) { - if (ref->tree_ref.root == BTRFS_CHUNK_TREE_OBJECTID) - return BTRFS_BLOCK_GROUP_SYSTEM; - else - return BTRFS_BLOCK_GROUP_METADATA; - } - return BTRFS_BLOCK_GROUP_DATA; -} - -static void add_pinned_bytes(struct btrfs_fs_info *fs_info, - struct btrfs_ref *ref) -{ - struct btrfs_space_info *space_info; - u64 flags = generic_ref_to_space_flags(ref); - - space_info = btrfs_find_space_info(fs_info, flags); - ASSERT(space_info); - percpu_counter_add_batch(&space_info->total_bytes_pinned, ref->len, - BTRFS_TOTAL_BYTES_PINNED_BATCH); -} - -static void sub_pinned_bytes(struct btrfs_fs_info *fs_info, - struct btrfs_ref *ref) -{ - struct btrfs_space_info *space_info; - u64 flags = generic_ref_to_space_flags(ref); - - space_info = btrfs_find_space_info(fs_info, flags); - ASSERT(space_info); - percpu_counter_add_batch(&space_info->total_bytes_pinned, -ref->len, - BTRFS_TOTAL_BYTES_PINNED_BATCH); -} - /* simple helper to search for an existing data extent at a given offset */ int btrfs_lookup_data_extent(struct btrfs_fs_info *fs_info, u64 start, u64 len) { @@ -1299,6 +1266,46 @@ static int btrfs_issue_discard(struct block_device *bdev, u64 start, u64 len, return ret; } +static int do_discard_extent(struct btrfs_bio_stripe *stripe, u64 *bytes) +{ + struct btrfs_device *dev = stripe->dev; + struct btrfs_fs_info *fs_info = dev->fs_info; + struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace; + u64 phys = stripe->physical; + u64 len = stripe->length; + u64 discarded = 0; + int ret = 0; + + /* Zone reset on a zoned filesystem */ + if (btrfs_can_zone_reset(dev, phys, len)) { + u64 src_disc; + + ret = btrfs_reset_device_zone(dev, phys, len, &discarded); + if (ret) + goto out; + + if (!btrfs_dev_replace_is_ongoing(dev_replace) || + dev != dev_replace->srcdev) + goto out; + + src_disc = discarded; + + /* Send to replace target as well */ + ret = btrfs_reset_device_zone(dev_replace->tgtdev, phys, len, + &discarded); + discarded += src_disc; + } else if (blk_queue_discard(bdev_get_queue(stripe->dev->bdev))) { + ret = btrfs_issue_discard(dev->bdev, phys, len, &discarded); + } else { + ret = 0; + *bytes = 0; + } + +out: + *bytes = discarded; + return ret; +} + int btrfs_discard_extent(struct btrfs_fs_info *fs_info, u64 bytenr, u64 num_bytes, u64 *actual_bytes) { @@ -1333,20 +1340,13 @@ int btrfs_discard_extent(struct btrfs_fs_info *fs_info, u64 bytenr, stripe = bbio->stripes; for (i = 0; i < bbio->num_stripes; i++, stripe++) { u64 bytes; - struct request_queue *req_q; if (!stripe->dev->bdev) { ASSERT(btrfs_test_opt(fs_info, DEGRADED)); continue; } - req_q = bdev_get_queue(stripe->dev->bdev); - if (!blk_queue_discard(req_q)) - continue; - ret = btrfs_issue_discard(stripe->dev->bdev, - stripe->physical, - stripe->length, - &bytes); + ret = do_discard_extent(stripe, &bytes); if (!ret) { discarded_bytes += bytes; } else if (ret != -EOPNOTSUPP) { @@ -1388,7 +1388,6 @@ int btrfs_inc_extent_ref(struct btrfs_trans_handle *trans, struct btrfs_ref *generic_ref) { struct btrfs_fs_info *fs_info = trans->fs_info; - int old_ref_mod, new_ref_mod; int ret; ASSERT(generic_ref->type != BTRFS_REF_NOT_SET && @@ -1397,17 +1396,12 @@ int btrfs_inc_extent_ref(struct btrfs_trans_handle *trans, generic_ref->tree_ref.root == BTRFS_TREE_LOG_OBJECTID); if (generic_ref->type == BTRFS_REF_METADATA) - ret = btrfs_add_delayed_tree_ref(trans, generic_ref, - NULL, &old_ref_mod, &new_ref_mod); + ret = btrfs_add_delayed_tree_ref(trans, generic_ref, NULL); else - ret = btrfs_add_delayed_data_ref(trans, generic_ref, 0, - &old_ref_mod, &new_ref_mod); + ret = btrfs_add_delayed_data_ref(trans, generic_ref, 0); btrfs_ref_tree_mod(fs_info, generic_ref); - if (ret == 0 && old_ref_mod < 0 && new_ref_mod >= 0) - sub_pinned_bytes(fs_info, generic_ref); - return ret; } @@ -1795,34 +1789,28 @@ void btrfs_cleanup_ref_head_accounting(struct btrfs_fs_info *fs_info, { int nr_items = 1; /* Dropping this ref head update. */ - if (head->total_ref_mod < 0) { - struct btrfs_space_info *space_info; - u64 flags; + /* + * We had csum deletions accounted for in our delayed refs rsv, we need + * to drop the csum leaves for this update from our delayed_refs_rsv. + */ + if (head->total_ref_mod < 0 && head->is_data) { + spin_lock(&delayed_refs->lock); + delayed_refs->pending_csums -= head->num_bytes; + spin_unlock(&delayed_refs->lock); + nr_items += btrfs_csum_bytes_to_leaves(fs_info, head->num_bytes); + } - if (head->is_data) - flags = BTRFS_BLOCK_GROUP_DATA; - else if (head->is_system) - flags = BTRFS_BLOCK_GROUP_SYSTEM; - else - flags = BTRFS_BLOCK_GROUP_METADATA; - space_info = btrfs_find_space_info(fs_info, flags); - ASSERT(space_info); - percpu_counter_add_batch(&space_info->total_bytes_pinned, - -head->num_bytes, - BTRFS_TOTAL_BYTES_PINNED_BATCH); + /* + * We were dropping refs, or had a new ref and dropped it, and thus must + * adjust down our total_bytes_pinned, the space may or may not have + * been pinned and so is accounted for properly in the pinned space by + * now. + */ + if (head->total_ref_mod < 0 || + (head->total_ref_mod == 0 && head->must_insert_reserved)) { + u64 flags = btrfs_ref_head_to_space_flags(head); - /* - * We had csum deletions accounted for in our delayed refs rsv, - * we need to drop the csum leaves for this update from our - * delayed_refs_rsv. - */ - if (head->is_data) { - spin_lock(&delayed_refs->lock); - delayed_refs->pending_csums -= head->num_bytes; - spin_unlock(&delayed_refs->lock); - nr_items += btrfs_csum_bytes_to_leaves(fs_info, - head->num_bytes); - } + btrfs_mod_total_bytes_pinned(fs_info, flags, -head->num_bytes); } btrfs_delayed_refs_rsv_release(fs_info, nr_items); @@ -2160,7 +2148,7 @@ int btrfs_run_delayed_refs(struct btrfs_trans_handle *trans, delayed_refs = &trans->transaction->delayed_refs; if (count == 0) - count = atomic_read(&delayed_refs->num_entries) * 2; + count = delayed_refs->num_heads_ready; again: #ifdef SCRAMBLE_DELAYED_REFS @@ -2572,8 +2560,7 @@ static int pin_down_extent(struct btrfs_trans_handle *trans, spin_unlock(&cache->lock); spin_unlock(&cache->space_info->lock); - percpu_counter_add_batch(&cache->space_info->total_bytes_pinned, - num_bytes, BTRFS_TOTAL_BYTES_PINNED_BATCH); + __btrfs_mod_total_bytes_pinned(cache->space_info, num_bytes); set_extent_dirty(&trans->transaction->pinned_extents, bytenr, bytenr + num_bytes - 1, GFP_NOFS | __GFP_NOFAIL); return 0; @@ -2784,11 +2771,14 @@ static int unpin_extent_range(struct btrfs_fs_info *fs_info, cache->pinned -= len; btrfs_space_info_update_bytes_pinned(fs_info, space_info, -len); space_info->max_extent_size = 0; - percpu_counter_add_batch(&space_info->total_bytes_pinned, - -len, BTRFS_TOTAL_BYTES_PINNED_BATCH); + __btrfs_mod_total_bytes_pinned(space_info, -len); if (cache->ro) { space_info->bytes_readonly += len; readonly = true; + } else if (btrfs_is_zoned(fs_info)) { + /* Need reset before reusing in a zoned block group */ + space_info->bytes_zone_unusable += len; + readonly = true; } spin_unlock(&cache->lock); if (!readonly && return_free_space && @@ -3318,7 +3308,6 @@ void btrfs_free_tree_block(struct btrfs_trans_handle *trans, { struct btrfs_fs_info *fs_info = root->fs_info; struct btrfs_ref generic_ref = { 0 }; - int pin = 1; int ret; btrfs_init_generic_ref(&generic_ref, BTRFS_DROP_DELAYED_REF, @@ -3327,25 +3316,23 @@ void btrfs_free_tree_block(struct btrfs_trans_handle *trans, root->root_key.objectid); if (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID) { - int old_ref_mod, new_ref_mod; - btrfs_ref_tree_mod(fs_info, &generic_ref); - ret = btrfs_add_delayed_tree_ref(trans, &generic_ref, NULL, - &old_ref_mod, &new_ref_mod); + ret = btrfs_add_delayed_tree_ref(trans, &generic_ref, NULL); BUG_ON(ret); /* -ENOMEM */ - pin = old_ref_mod >= 0 && new_ref_mod < 0; } if (last_ref && btrfs_header_generation(buf) == trans->transid) { struct btrfs_block_group *cache; + bool must_pin = false; if (root->root_key.objectid != BTRFS_TREE_LOG_OBJECTID) { ret = check_ref_cleanup(trans, buf->start); - if (!ret) + if (!ret) { + btrfs_redirty_list_add(trans->transaction, buf); goto out; + } } - pin = 0; cache = btrfs_lookup_block_group(fs_info, buf->start); if (btrfs_header_flag(buf, BTRFS_HEADER_FLAG_WRITTEN)) { @@ -3354,6 +3341,33 @@ void btrfs_free_tree_block(struct btrfs_trans_handle *trans, goto out; } + /* + * If this is a leaf and there are tree mod log users, we may + * have recorded mod log operations that point to this leaf. + * So we must make sure no one reuses this leaf's extent before + * mod log operations are applied to a node, otherwise after + * rewinding a node using the mod log operations we get an + * inconsistent btree, as the leaf's extent may now be used as + * a node or leaf for another different btree. + * We are safe from races here because at this point no other + * node or root points to this extent buffer, so if after this + * check a new tree mod log user joins, it will not be able to + * find a node pointing to this leaf and record operations that + * point to this leaf. + */ + if (btrfs_header_level(buf) == 0) { + read_lock(&fs_info->tree_mod_log_lock); + must_pin = !list_empty(&fs_info->tree_mod_seq_list); + read_unlock(&fs_info->tree_mod_log_lock); + } + + if (must_pin || btrfs_is_zoned(fs_info)) { + btrfs_redirty_list_add(trans->transaction, buf); + pin_down_extent(trans, cache, buf->start, buf->len, 1); + btrfs_put_block_group(cache); + goto out; + } + WARN_ON(test_bit(EXTENT_BUFFER_DIRTY, &buf->bflags)); btrfs_add_free_space(cache, buf->start, buf->len); @@ -3362,9 +3376,6 @@ void btrfs_free_tree_block(struct btrfs_trans_handle *trans, trace_btrfs_reserved_extent_free(fs_info, buf->start, buf->len); } out: - if (pin) - add_pinned_bytes(fs_info, &generic_ref); - if (last_ref) { /* * Deleting the buffer, clear the corrupt flag since it doesn't @@ -3378,7 +3389,6 @@ out: int btrfs_free_extent(struct btrfs_trans_handle *trans, struct btrfs_ref *ref) { struct btrfs_fs_info *fs_info = trans->fs_info; - int old_ref_mod, new_ref_mod; int ret; if (btrfs_is_testing(fs_info)) @@ -3394,14 +3404,11 @@ int btrfs_free_extent(struct btrfs_trans_handle *trans, struct btrfs_ref *ref) ref->data_ref.ref_root == BTRFS_TREE_LOG_OBJECTID)) { /* unlocks the pinned mutex */ btrfs_pin_extent(trans, ref->bytenr, ref->len, 1); - old_ref_mod = new_ref_mod = 0; ret = 0; } else if (ref->type == BTRFS_REF_METADATA) { - ret = btrfs_add_delayed_tree_ref(trans, ref, NULL, - &old_ref_mod, &new_ref_mod); + ret = btrfs_add_delayed_tree_ref(trans, ref, NULL); } else { - ret = btrfs_add_delayed_data_ref(trans, ref, 0, - &old_ref_mod, &new_ref_mod); + ret = btrfs_add_delayed_data_ref(trans, ref, 0); } if (!((ref->type == BTRFS_REF_METADATA && @@ -3410,9 +3417,6 @@ int btrfs_free_extent(struct btrfs_trans_handle *trans, struct btrfs_ref *ref) ref->data_ref.ref_root == BTRFS_TREE_LOG_OBJECTID))) btrfs_ref_tree_mod(fs_info, ref); - if (ret == 0 && old_ref_mod >= 0 && new_ref_mod < 0) - add_pinned_bytes(fs_info, ref); - return ret; } @@ -3489,6 +3493,7 @@ btrfs_release_block_group(struct btrfs_block_group *cache, enum btrfs_extent_allocation_policy { BTRFS_EXTENT_ALLOC_CLUSTERED, + BTRFS_EXTENT_ALLOC_ZONED, }; /* @@ -3513,6 +3518,9 @@ struct find_free_extent_ctl { bool have_caching_bg; bool orig_have_caching_bg; + /* Allocation is called for tree-log */ + bool for_treelog; + /* RAID index, converted from flags */ int index; @@ -3741,6 +3749,118 @@ static int do_allocation_clustered(struct btrfs_block_group *block_group, return find_free_extent_unclustered(block_group, ffe_ctl); } +/* + * Tree-log block group locking + * ============================ + * + * fs_info::treelog_bg_lock protects the fs_info::treelog_bg which + * indicates the starting address of a block group, which is reserved only + * for tree-log metadata. + * + * Lock nesting + * ============ + * + * space_info::lock + * block_group::lock + * fs_info::treelog_bg_lock + */ + +/* + * Simple allocator for sequential-only block group. It only allows sequential + * allocation. No need to play with trees. This function also reserves the + * bytes as in btrfs_add_reserved_bytes. + */ +static int do_allocation_zoned(struct btrfs_block_group *block_group, + struct find_free_extent_ctl *ffe_ctl, + struct btrfs_block_group **bg_ret) +{ + struct btrfs_fs_info *fs_info = block_group->fs_info; + struct btrfs_space_info *space_info = block_group->space_info; + struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; + u64 start = block_group->start; + u64 num_bytes = ffe_ctl->num_bytes; + u64 avail; + u64 bytenr = block_group->start; + u64 log_bytenr; + int ret = 0; + bool skip; + + ASSERT(btrfs_is_zoned(block_group->fs_info)); + + /* + * Do not allow non-tree-log blocks in the dedicated tree-log block + * group, and vice versa. + */ + spin_lock(&fs_info->treelog_bg_lock); + log_bytenr = fs_info->treelog_bg; + skip = log_bytenr && ((ffe_ctl->for_treelog && bytenr != log_bytenr) || + (!ffe_ctl->for_treelog && bytenr == log_bytenr)); + spin_unlock(&fs_info->treelog_bg_lock); + if (skip) + return 1; + + spin_lock(&space_info->lock); + spin_lock(&block_group->lock); + spin_lock(&fs_info->treelog_bg_lock); + + ASSERT(!ffe_ctl->for_treelog || + block_group->start == fs_info->treelog_bg || + fs_info->treelog_bg == 0); + + if (block_group->ro) { + ret = 1; + goto out; + } + + /* + * Do not allow currently using block group to be tree-log dedicated + * block group. + */ + if (ffe_ctl->for_treelog && !fs_info->treelog_bg && + (block_group->used || block_group->reserved)) { + ret = 1; + goto out; + } + + avail = block_group->length - block_group->alloc_offset; + if (avail < num_bytes) { + if (ffe_ctl->max_extent_size < avail) { + /* + * With sequential allocator, free space is always + * contiguous + */ + ffe_ctl->max_extent_size = avail; + ffe_ctl->total_free_space = avail; + } + ret = 1; + goto out; + } + + if (ffe_ctl->for_treelog && !fs_info->treelog_bg) + fs_info->treelog_bg = block_group->start; + + ffe_ctl->found_offset = start + block_group->alloc_offset; + block_group->alloc_offset += num_bytes; + spin_lock(&ctl->tree_lock); + ctl->free_space -= num_bytes; + spin_unlock(&ctl->tree_lock); + + /* + * We do not check if found_offset is aligned to stripesize. The + * address is anyway rewritten when using zone append writing. + */ + + ffe_ctl->search_start = ffe_ctl->found_offset; + +out: + if (ret && ffe_ctl->for_treelog) + fs_info->treelog_bg = 0; + spin_unlock(&fs_info->treelog_bg_lock); + spin_unlock(&block_group->lock); + spin_unlock(&space_info->lock); + return ret; +} + static int do_allocation(struct btrfs_block_group *block_group, struct find_free_extent_ctl *ffe_ctl, struct btrfs_block_group **bg_ret) @@ -3748,6 +3868,8 @@ static int do_allocation(struct btrfs_block_group *block_group, switch (ffe_ctl->policy) { case BTRFS_EXTENT_ALLOC_CLUSTERED: return do_allocation_clustered(block_group, ffe_ctl, bg_ret); + case BTRFS_EXTENT_ALLOC_ZONED: + return do_allocation_zoned(block_group, ffe_ctl, bg_ret); default: BUG(); } @@ -3762,6 +3884,9 @@ static void release_block_group(struct btrfs_block_group *block_group, ffe_ctl->retry_clustered = false; ffe_ctl->retry_unclustered = false; break; + case BTRFS_EXTENT_ALLOC_ZONED: + /* Nothing to do */ + break; default: BUG(); } @@ -3790,6 +3915,9 @@ static void found_extent(struct find_free_extent_ctl *ffe_ctl, case BTRFS_EXTENT_ALLOC_CLUSTERED: found_extent_clustered(ffe_ctl, ins); break; + case BTRFS_EXTENT_ALLOC_ZONED: + /* Nothing to do */ + break; default: BUG(); } @@ -3805,6 +3933,9 @@ static int chunk_allocation_failed(struct find_free_extent_ctl *ffe_ctl) */ ffe_ctl->loop = LOOP_NO_EMPTY_SIZE; return 0; + case BTRFS_EXTENT_ALLOC_ZONED: + /* Give up here */ + return -ENOSPC; default: BUG(); } @@ -3973,6 +4104,14 @@ static int prepare_allocation(struct btrfs_fs_info *fs_info, case BTRFS_EXTENT_ALLOC_CLUSTERED: return prepare_allocation_clustered(fs_info, ffe_ctl, space_info, ins); + case BTRFS_EXTENT_ALLOC_ZONED: + if (ffe_ctl->for_treelog) { + spin_lock(&fs_info->treelog_bg_lock); + if (fs_info->treelog_bg) + ffe_ctl->hint_byte = fs_info->treelog_bg; + spin_unlock(&fs_info->treelog_bg_lock); + } + return 0; default: BUG(); } @@ -4015,6 +4154,7 @@ static noinline int find_free_extent(struct btrfs_root *root, struct find_free_extent_ctl ffe_ctl = {0}; struct btrfs_space_info *space_info; bool full_search = false; + bool for_treelog = (root->root_key.objectid == BTRFS_TREE_LOG_OBJECTID); WARN_ON(num_bytes < fs_info->sectorsize); @@ -4028,6 +4168,7 @@ static noinline int find_free_extent(struct btrfs_root *root, ffe_ctl.orig_have_caching_bg = false; ffe_ctl.found_offset = 0; ffe_ctl.hint_byte = hint_byte_orig; + ffe_ctl.for_treelog = for_treelog; ffe_ctl.policy = BTRFS_EXTENT_ALLOC_CLUSTERED; /* For clustered allocation */ @@ -4036,6 +4177,9 @@ static noinline int find_free_extent(struct btrfs_root *root, ffe_ctl.last_ptr = NULL; ffe_ctl.use_cluster = true; + if (btrfs_is_zoned(fs_info)) + ffe_ctl.policy = BTRFS_EXTENT_ALLOC_ZONED; + ins->type = BTRFS_EXTENT_ITEM_KEY; ins->objectid = 0; ins->offset = 0; @@ -4099,8 +4243,11 @@ search: struct btrfs_block_group *bg_ret; /* If the block group is read-only, we can skip it entirely. */ - if (unlikely(block_group->ro)) + if (unlikely(block_group->ro)) { + if (for_treelog) + btrfs_clear_treelog_bg(block_group); continue; + } btrfs_grab_block_group(block_group, delalloc); ffe_ctl.search_start = block_group->start; @@ -4178,20 +4325,21 @@ have_block_group: /* move on to the next group */ if (ffe_ctl.search_start + num_bytes > block_group->start + block_group->length) { - btrfs_add_free_space(block_group, ffe_ctl.found_offset, - num_bytes); + btrfs_add_free_space_unused(block_group, + ffe_ctl.found_offset, num_bytes); goto loop; } if (ffe_ctl.found_offset < ffe_ctl.search_start) - btrfs_add_free_space(block_group, ffe_ctl.found_offset, - ffe_ctl.search_start - ffe_ctl.found_offset); + btrfs_add_free_space_unused(block_group, + ffe_ctl.found_offset, + ffe_ctl.search_start - ffe_ctl.found_offset); ret = btrfs_add_reserved_bytes(block_group, ram_bytes, num_bytes, delalloc); if (ret == -EAGAIN) { - btrfs_add_free_space(block_group, ffe_ctl.found_offset, - num_bytes); + btrfs_add_free_space_unused(block_group, + ffe_ctl.found_offset, num_bytes); goto loop; } btrfs_inc_block_group_reservations(block_group); @@ -4285,6 +4433,7 @@ int btrfs_reserve_extent(struct btrfs_root *root, u64 ram_bytes, bool final_tried = num_bytes == min_alloc_size; u64 flags; int ret; + bool for_treelog = (root->root_key.objectid == BTRFS_TREE_LOG_OBJECTID); flags = get_alloc_profile_by_root(root, is_data); again: @@ -4308,8 +4457,8 @@ again: sinfo = btrfs_find_space_info(fs_info, flags); btrfs_err(fs_info, - "allocation failed flags %llu, wanted %llu", - flags, num_bytes); + "allocation failed flags %llu, wanted %llu tree-log %d", + flags, num_bytes, for_treelog); if (sinfo) btrfs_dump_space_info(fs_info, sinfo, num_bytes, 1); @@ -4491,7 +4640,6 @@ static int alloc_reserved_tree_block(struct btrfs_trans_handle *trans, } if (node->type == BTRFS_SHARED_BLOCK_REF_KEY) { - BUG_ON(!(flags & BTRFS_BLOCK_FLAG_FULL_BACKREF)); btrfs_set_extent_inline_ref_type(leaf, iref, BTRFS_SHARED_BLOCK_REF_KEY); btrfs_set_extent_inline_ref_offset(leaf, iref, ref->parent); @@ -4528,7 +4676,6 @@ int btrfs_alloc_reserved_file_extent(struct btrfs_trans_handle *trans, struct btrfs_key *ins) { struct btrfs_ref generic_ref = { 0 }; - int ret; BUG_ON(root->root_key.objectid == BTRFS_TREE_LOG_OBJECTID); @@ -4536,9 +4683,8 @@ int btrfs_alloc_reserved_file_extent(struct btrfs_trans_handle *trans, ins->objectid, ins->offset, 0); btrfs_init_data_ref(&generic_ref, root->root_key.objectid, owner, offset); btrfs_ref_tree_mod(root->fs_info, &generic_ref); - ret = btrfs_add_delayed_data_ref(trans, &generic_ref, - ram_bytes, NULL, NULL); - return ret; + + return btrfs_add_delayed_data_ref(trans, &generic_ref, ram_bytes); } /* @@ -4620,6 +4766,7 @@ btrfs_init_new_buffer(struct btrfs_trans_handle *trans, struct btrfs_root *root, __btrfs_tree_lock(buf, nest); btrfs_clean_tree_block(buf); clear_bit(EXTENT_BUFFER_STALE, &buf->bflags); + clear_bit(EXTENT_BUFFER_NO_CHECK, &buf->bflags); set_extent_buffer_uptodate(buf); @@ -4730,8 +4877,7 @@ struct extent_buffer *btrfs_alloc_tree_block(struct btrfs_trans_handle *trans, generic_ref.real_root = root->root_key.objectid; btrfs_init_tree_ref(&generic_ref, level, root_objectid); btrfs_ref_tree_mod(fs_info, &generic_ref); - ret = btrfs_add_delayed_tree_ref(trans, &generic_ref, - extent_op, NULL, NULL); + ret = btrfs_add_delayed_tree_ref(trans, &generic_ref, extent_op); if (ret) goto out_free_delayed; } diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index c9cee458e001..910769d5fcdb 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -24,6 +24,9 @@ #include "rcu-string.h" #include "backref.h" #include "disk-io.h" +#include "subpage.h" +#include "zoned.h" +#include "block-group.h" static struct kmem_cache *extent_state_cache; static struct kmem_cache *extent_buffer_cache; @@ -389,16 +392,16 @@ do_insert: } /** - * __etree_search - searche @tree for an entry that contains @offset. Such - * entry would have entry->start <= offset && entry->end >= offset. + * Search @tree for an entry that contains @offset. Such entry would have + * entry->start <= offset && entry->end >= offset. * - * @tree - the tree to search - * @offset - offset that should fall within an entry in @tree - * @next_ret - pointer to the first entry whose range ends after @offset - * @prev - pointer to the first entry whose range begins before @offset - * @p_ret - pointer where new node should be anchored (used when inserting an - * entry in the tree) - * @parent_ret - points to entry which would have been the parent of the entry, + * @tree: the tree to search + * @offset: offset that should fall within an entry in @tree + * @next_ret: pointer to the first entry whose range ends after @offset + * @prev_ret: pointer to the first entry whose range begins before @offset + * @p_ret: pointer where new node should be anchored (used when inserting an + * entry in the tree) + * @parent_ret: points to entry which would have been the parent of the entry, * containing @offset * * This function returns a pointer to the entry that contains @offset byte @@ -1588,12 +1591,13 @@ out: } /** - * find_contiguous_extent_bit: find a contiguous area of bits - * @tree - io tree to check - * @start - offset to start the search from - * @start_ret - the first offset we found with the bits set - * @end_ret - the final contiguous range of the bits that were set - * @bits - bits to look for + * Find a contiguous area of bits + * + * @tree: io tree to check + * @start: offset to start the search from + * @start_ret: the first offset we found with the bits set + * @end_ret: the final contiguous range of the bits that were set + * @bits: bits to look for * * set_extent_bit and clear_extent_bit can temporarily split contiguous ranges * to set bits appropriately, and then merge them again. During this time it @@ -1625,14 +1629,14 @@ int find_contiguous_extent_bit(struct extent_io_tree *tree, u64 start, } /** - * find_first_clear_extent_bit - find the first range that has @bits not set. - * This range could start before @start. + * Find the first range that has @bits not set. This range could start before + * @start. * - * @tree - the tree to search - * @start - the offset at/after which the found extent should start - * @start_ret - records the beginning of the range - * @end_ret - records the end of the range (inclusive) - * @bits - the set of bits which must be unset + * @tree: the tree to search + * @start: offset at/after which the found extent should start + * @start_ret: records the beginning of the range + * @end_ret: records the end of the range (inclusive) + * @bits: the set of bits which must be unset * * Since unallocated range is also considered one which doesn't have the bits * set it's possible that @end_ret contains -1, this happens in case the range @@ -1975,10 +1979,10 @@ static int __process_pages_contig(struct address_space *mapping, pages_processed++; continue; } - if (page_ops & PAGE_CLEAR_DIRTY) + if (page_ops & PAGE_START_WRITEBACK) { clear_page_dirty_for_io(pages[i]); - if (page_ops & PAGE_SET_WRITEBACK) set_page_writeback(pages[i]); + } if (page_ops & PAGE_SET_ERROR) SetPageError(pages[i]); if (page_ops & PAGE_END_WRITEBACK) @@ -2256,6 +2260,9 @@ int repair_io_failure(struct btrfs_fs_info *fs_info, u64 ino, u64 start, ASSERT(!(fs_info->sb->s_flags & SB_RDONLY)); BUG_ON(!mirror_num); + if (btrfs_is_zoned(fs_info)) + return btrfs_repair_one_zone(fs_info, logical); + bio = btrfs_io_bio_alloc(1); bio->bi_iter.bi_size = 0; map_length = length; @@ -2732,6 +2739,7 @@ static void end_bio_extent_writepage(struct bio *bio) u64 start; u64 end; struct bvec_iter_all iter_all; + bool first_bvec = true; ASSERT(!bio_flagged(bio, BIO_CLONED)); bio_for_each_segment_all(bvec, bio, iter_all) { @@ -2758,6 +2766,11 @@ static void end_bio_extent_writepage(struct bio *bio) start = page_offset(page); end = start + bvec->bv_offset + bvec->bv_len - 1; + if (first_bvec) { + btrfs_record_physical_zoned(inode, start, bio); + first_bvec = false; + } + end_extent_writepage(page, error, start, end); end_page_writeback(page); } @@ -2775,7 +2788,7 @@ struct processed_extent { struct btrfs_inode *inode; /* Start of the range in @inode */ u64 start; - /* End of the range in in @inode */ + /* End of the range in @inode */ u64 end; bool uptodate; }; @@ -2838,15 +2851,67 @@ update: processed->uptodate = uptodate; } -static void endio_readpage_update_page_status(struct page *page, bool uptodate) +static void begin_page_read(struct btrfs_fs_info *fs_info, struct page *page) { + ASSERT(PageLocked(page)); + if (fs_info->sectorsize == PAGE_SIZE) + return; + + ASSERT(PagePrivate(page)); + btrfs_subpage_start_reader(fs_info, page, page_offset(page), PAGE_SIZE); +} + +static void end_page_read(struct page *page, bool uptodate, u64 start, u32 len) +{ + struct btrfs_fs_info *fs_info = btrfs_sb(page->mapping->host->i_sb); + + ASSERT(page_offset(page) <= start && + start + len <= page_offset(page) + PAGE_SIZE); + if (uptodate) { - SetPageUptodate(page); + btrfs_page_set_uptodate(fs_info, page, start, len); } else { - ClearPageUptodate(page); - SetPageError(page); + btrfs_page_clear_uptodate(fs_info, page, start, len); + btrfs_page_set_error(fs_info, page, start, len); } - unlock_page(page); + + if (fs_info->sectorsize == PAGE_SIZE) + unlock_page(page); + else if (is_data_inode(page->mapping->host)) + /* + * For subpage data, unlock the page if we're the last reader. + * For subpage metadata, page lock is not utilized for read. + */ + btrfs_subpage_end_reader(fs_info, page, start, len); +} + +/* + * Find extent buffer for a givne bytenr. + * + * This is for end_bio_extent_readpage(), thus we can't do any unsafe locking + * in endio context. + */ +static struct extent_buffer *find_extent_buffer_readpage( + struct btrfs_fs_info *fs_info, struct page *page, u64 bytenr) +{ + struct extent_buffer *eb; + + /* + * For regular sectorsize, we can use page->private to grab extent + * buffer + */ + if (fs_info->sectorsize == PAGE_SIZE) { + ASSERT(PagePrivate(page) && page->private); + return (struct extent_buffer *)page->private; + } + + /* For subpage case, we need to lookup buffer radix tree */ + rcu_read_lock(); + eb = radix_tree_lookup(&fs_info->buffer_radix, + bytenr >> fs_info->sectorsize_bits); + rcu_read_unlock(); + ASSERT(eb); + return eb; } /* @@ -2960,7 +3025,7 @@ static void end_bio_extent_readpage(struct bio *bio) } else { struct extent_buffer *eb; - eb = (struct extent_buffer *)page->private; + eb = find_extent_buffer_readpage(fs_info, page, start); set_bit(EXTENT_BUFFER_READ_ERR, &eb->bflags); eb->read_mirror = mirror; atomic_dec(&eb->io_pages); @@ -2972,18 +3037,29 @@ readpage_ok: if (likely(uptodate)) { loff_t i_size = i_size_read(inode); pgoff_t end_index = i_size >> PAGE_SHIFT; - unsigned off; - /* Zero out the end if this page straddles i_size */ - off = offset_in_page(i_size); - if (page->index == end_index && off) - zero_user_segment(page, off, PAGE_SIZE); + /* + * Zero out the remaining part if this range straddles + * i_size. + * + * Here we should only zero the range inside the bvec, + * not touch anything else. + * + * NOTE: i_size is exclusive while end is inclusive. + */ + if (page->index == end_index && i_size <= end) { + u32 zero_start = max(offset_in_page(i_size), + offset_in_page(start)); + + zero_user_segment(page, zero_start, + offset_in_page(end) + 1); + } } ASSERT(bio_offset + len > bio_offset); bio_offset += len; /* Update page status and unlock */ - endio_readpage_update_page_status(page, uptodate); + end_page_read(page, uptodate, start, len); endio_readpage_release_extent(&processed, BTRFS_I(inode), start, end, uptodate); } @@ -3012,7 +3088,7 @@ struct bio *btrfs_bio_alloc(u64 first_byte) { struct bio *bio; - bio = bio_alloc_bioset(GFP_NOFS, BIO_MAX_PAGES, &btrfs_bioset); + bio = bio_alloc_bioset(GFP_NOFS, BIO_MAX_VECS, &btrfs_bioset); bio->bi_iter.bi_sector = first_byte >> 9; btrfs_io_bio_init(btrfs_io_bio(bio)); return bio; @@ -3058,14 +3134,67 @@ struct bio *btrfs_bio_clone_partial(struct bio *orig, int offset, int size) return bio; } +/** + * Attempt to add a page to bio + * + * @bio: destination bio + * @page: page to add to the bio + * @disk_bytenr: offset of the new bio or to check whether we are adding + * a contiguous page to the previous one + * @pg_offset: starting offset in the page + * @size: portion of page that we want to write + * @prev_bio_flags: flags of previous bio to see if we can merge the current one + * @bio_flags: flags of the current bio to see if we can merge them + * @return: true if page was added, false otherwise + * + * Attempt to add a page to bio considering stripe alignment etc. + * + * Return true if successfully page added. Otherwise, return false. + */ +static bool btrfs_bio_add_page(struct bio *bio, struct page *page, + u64 disk_bytenr, unsigned int size, + unsigned int pg_offset, + unsigned long prev_bio_flags, + unsigned long bio_flags) +{ + const sector_t sector = disk_bytenr >> SECTOR_SHIFT; + bool contig; + int ret; + + if (prev_bio_flags != bio_flags) + return false; + + if (prev_bio_flags & EXTENT_BIO_COMPRESSED) + contig = bio->bi_iter.bi_sector == sector; + else + contig = bio_end_sector(bio) == sector; + if (!contig) + return false; + + if (btrfs_bio_fits_in_stripe(page, size, bio, bio_flags)) + return false; + + if (bio_op(bio) == REQ_OP_ZONE_APPEND) { + struct page *first_page = bio_first_bvec_all(bio)->bv_page; + + if (!btrfs_bio_fits_in_ordered_extent(first_page, bio, size)) + return false; + ret = bio_add_zone_append_page(bio, page, size, pg_offset); + } else { + ret = bio_add_page(bio, page, size, pg_offset); + } + + return ret == size; +} + /* * @opf: bio REQ_OP_* and REQ_* flags as one value * @wbc: optional writeback control for io accounting * @page: page to add to the bio + * @disk_bytenr: logical bytenr where the write will be + * @size: portion of page that we want to write to * @pg_offset: offset of the new bio or to check whether we are adding * a contiguous page to the previous one - * @size: portion of page that we want to write - * @offset: starting offset in the page * @bio_ret: must be valid pointer, newly allocated bio will be stored there * @end_io_func: end_io callback for new bio * @mirror_num: desired mirror to read/write @@ -3074,7 +3203,7 @@ struct bio *btrfs_bio_clone_partial(struct bio *orig, int offset, int size) */ static int submit_extent_page(unsigned int opf, struct writeback_control *wbc, - struct page *page, u64 offset, + struct page *page, u64 disk_bytenr, size_t size, unsigned long pg_offset, struct bio **bio_ret, bio_end_io_t end_io_func, @@ -3086,27 +3215,17 @@ static int submit_extent_page(unsigned int opf, int ret = 0; struct bio *bio; size_t io_size = min_t(size_t, size, PAGE_SIZE); - sector_t sector = offset >> 9; - struct extent_io_tree *tree = &BTRFS_I(page->mapping->host)->io_tree; + struct btrfs_inode *inode = BTRFS_I(page->mapping->host); + struct extent_io_tree *tree = &inode->io_tree; + struct btrfs_fs_info *fs_info = inode->root->fs_info; ASSERT(bio_ret); if (*bio_ret) { - bool contig; - bool can_merge = true; - bio = *bio_ret; - if (prev_bio_flags & EXTENT_BIO_COMPRESSED) - contig = bio->bi_iter.bi_sector == sector; - else - contig = bio_end_sector(bio) == sector; - - if (btrfs_bio_fits_in_stripe(page, io_size, bio, bio_flags)) - can_merge = false; - - if (prev_bio_flags != bio_flags || !contig || !can_merge || - force_bio_submit || - bio_add_page(bio, page, io_size, pg_offset) < io_size) { + if (force_bio_submit || + !btrfs_bio_add_page(bio, page, disk_bytenr, io_size, + pg_offset, prev_bio_flags, bio_flags)) { ret = submit_one_bio(bio, mirror_num, prev_bio_flags); if (ret < 0) { *bio_ret = NULL; @@ -3120,7 +3239,7 @@ static int submit_extent_page(unsigned int opf, } } - bio = btrfs_bio_alloc(offset); + bio = btrfs_bio_alloc(disk_bytenr); bio_add_page(bio, page, io_size, pg_offset); bio->bi_end_io = end_io_func; bio->bi_private = tree; @@ -3129,20 +3248,39 @@ static int submit_extent_page(unsigned int opf, if (wbc) { struct block_device *bdev; - bdev = BTRFS_I(page->mapping->host)->root->fs_info->fs_devices->latest_bdev; + bdev = fs_info->fs_devices->latest_bdev; bio_set_dev(bio, bdev); wbc_init_bio(wbc, bio); wbc_account_cgroup_owner(wbc, page, io_size); } + if (btrfs_is_zoned(fs_info) && bio_op(bio) == REQ_OP_ZONE_APPEND) { + struct extent_map *em; + struct map_lookup *map; + + em = btrfs_get_chunk_map(fs_info, disk_bytenr, io_size); + if (IS_ERR(em)) + return PTR_ERR(em); + + map = em->map_lookup; + /* We only support single profile for now */ + ASSERT(map->num_stripes == 1); + btrfs_io_bio(bio)->device = map->stripes[0].dev; + + free_extent_map(em); + } *bio_ret = bio; return ret; } -static void attach_extent_buffer_page(struct extent_buffer *eb, - struct page *page) +static int attach_extent_buffer_page(struct extent_buffer *eb, + struct page *page, + struct btrfs_subpage *prealloc) { + struct btrfs_fs_info *fs_info = eb->fs_info; + int ret = 0; + /* * If the page is mapped to btree inode, we should hold the private * lock to prevent race. @@ -3152,16 +3290,62 @@ static void attach_extent_buffer_page(struct extent_buffer *eb, if (page->mapping) lockdep_assert_held(&page->mapping->private_lock); - if (!PagePrivate(page)) - attach_page_private(page, eb); + if (fs_info->sectorsize == PAGE_SIZE) { + if (!PagePrivate(page)) + attach_page_private(page, eb); + else + WARN_ON(page->private != (unsigned long)eb); + return 0; + } + + /* Already mapped, just free prealloc */ + if (PagePrivate(page)) { + btrfs_free_subpage(prealloc); + return 0; + } + + if (prealloc) + /* Has preallocated memory for subpage */ + attach_page_private(page, prealloc); else - WARN_ON(page->private != (unsigned long)eb); + /* Do new allocation to attach subpage */ + ret = btrfs_attach_subpage(fs_info, page, + BTRFS_SUBPAGE_METADATA); + return ret; +} + +int set_page_extent_mapped(struct page *page) +{ + struct btrfs_fs_info *fs_info; + + ASSERT(page->mapping); + + if (PagePrivate(page)) + return 0; + + fs_info = btrfs_sb(page->mapping->host->i_sb); + + if (fs_info->sectorsize < PAGE_SIZE) + return btrfs_attach_subpage(fs_info, page, BTRFS_SUBPAGE_DATA); + + attach_page_private(page, (void *)EXTENT_PAGE_PRIVATE); + return 0; } -void set_page_extent_mapped(struct page *page) +void clear_page_extent_mapped(struct page *page) { + struct btrfs_fs_info *fs_info; + + ASSERT(page->mapping); + if (!PagePrivate(page)) - attach_page_private(page, (void *)EXTENT_PAGE_PRIVATE); + return; + + fs_info = btrfs_sb(page->mapping->host->i_sb); + if (fs_info->sectorsize < PAGE_SIZE) + return btrfs_detach_subpage(fs_info, page); + + detach_page_private(page); } static struct extent_map * @@ -3202,6 +3386,7 @@ int btrfs_do_readpage(struct page *page, struct extent_map **em_cached, unsigned int read_flags, u64 *prev_em_start) { struct inode *inode = page->mapping->host; + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); u64 start = page_offset(page); const u64 end = start + PAGE_SIZE - 1; u64 cur = start; @@ -3218,12 +3403,19 @@ int btrfs_do_readpage(struct page *page, struct extent_map **em_cached, unsigned long this_bio_flag = 0; struct extent_io_tree *tree = &BTRFS_I(inode)->io_tree; - set_page_extent_mapped(page); + ret = set_page_extent_mapped(page); + if (ret < 0) { + unlock_extent(tree, start, end); + btrfs_page_set_error(fs_info, page, start, PAGE_SIZE); + unlock_page(page); + goto out; + } if (!PageUptodate(page)) { if (cleancache_get_page(page) == 0) { BUG_ON(blocksize != PAGE_SIZE); unlock_extent(tree, start, end); + unlock_page(page); goto out; } } @@ -3240,9 +3432,10 @@ int btrfs_do_readpage(struct page *page, struct extent_map **em_cached, kunmap_atomic(userpage); } } + begin_page_read(fs_info, page); while (cur <= end) { bool force_bio_submit = false; - u64 offset; + u64 disk_bytenr; if (cur >= last_byte) { char *userpage; @@ -3257,13 +3450,14 @@ int btrfs_do_readpage(struct page *page, struct extent_map **em_cached, &cached, GFP_NOFS); unlock_extent_cached(tree, cur, cur + iosize - 1, &cached); + end_page_read(page, true, cur, iosize); break; } em = __get_extent_map(inode, page, pg_offset, cur, end - cur + 1, em_cached); if (IS_ERR_OR_NULL(em)) { - SetPageError(page); unlock_extent(tree, cur, end); + end_page_read(page, false, cur, end + 1 - cur); break; } extent_offset = cur - em->start; @@ -3280,9 +3474,9 @@ int btrfs_do_readpage(struct page *page, struct extent_map **em_cached, cur_end = min(extent_map_end(em) - 1, end); iosize = ALIGN(iosize, blocksize); if (this_bio_flag & EXTENT_BIO_COMPRESSED) - offset = em->block_start; + disk_bytenr = em->block_start; else - offset = em->block_start + extent_offset; + disk_bytenr = em->block_start + extent_offset; block_start = em->block_start; if (test_bit(EXTENT_FLAG_PREALLOC, &em->flags)) block_start = EXTENT_MAP_HOLE; @@ -3346,6 +3540,7 @@ int btrfs_do_readpage(struct page *page, struct extent_map **em_cached, &cached, GFP_NOFS); unlock_extent_cached(tree, cur, cur + iosize - 1, &cached); + end_page_read(page, true, cur, iosize); cur = cur + iosize; pg_offset += iosize; continue; @@ -3355,6 +3550,7 @@ int btrfs_do_readpage(struct page *page, struct extent_map **em_cached, EXTENT_UPTODATE, 1, NULL)) { check_page_uptodate(tree, page); unlock_extent(tree, cur, cur + iosize - 1); + end_page_read(page, true, cur, iosize); cur = cur + iosize; pg_offset += iosize; continue; @@ -3363,15 +3559,15 @@ int btrfs_do_readpage(struct page *page, struct extent_map **em_cached, * to date. Error out */ if (block_start == EXTENT_MAP_INLINE) { - SetPageError(page); unlock_extent(tree, cur, cur + iosize - 1); + end_page_read(page, false, cur, iosize); cur = cur + iosize; pg_offset += iosize; continue; } ret = submit_extent_page(REQ_OP_READ | read_flags, NULL, - page, offset, iosize, + page, disk_bytenr, iosize, pg_offset, bio, end_bio_extent_readpage, 0, *bio_flags, @@ -3381,19 +3577,14 @@ int btrfs_do_readpage(struct page *page, struct extent_map **em_cached, nr++; *bio_flags = this_bio_flag; } else { - SetPageError(page); unlock_extent(tree, cur, cur + iosize - 1); + end_page_read(page, false, cur, iosize); goto out; } cur = cur + iosize; pg_offset += iosize; } out: - if (!nr) { - if (!PageError(page)) - SetPageUptodate(page); - unlock_page(page); - } return ret; } @@ -3513,23 +3704,21 @@ static noinline_for_stack int __extent_writepage_io(struct btrfs_inode *inode, unsigned long nr_written, int *nr_ret) { + struct btrfs_fs_info *fs_info = inode->root->fs_info; struct extent_io_tree *tree = &inode->io_tree; u64 start = page_offset(page); - u64 page_end = start + PAGE_SIZE - 1; - u64 end; + u64 end = start + PAGE_SIZE - 1; u64 cur = start; u64 extent_offset; u64 block_start; - u64 iosize; struct extent_map *em; - size_t pg_offset = 0; - size_t blocksize; int ret = 0; int nr = 0; + u32 opf = REQ_OP_WRITE; const unsigned int write_flags = wbc_to_write_flags(wbc); bool compressed; - ret = btrfs_writepage_cow_fixup(page, start, page_end); + ret = btrfs_writepage_cow_fixup(page, start, end); if (ret) { /* Fixup worker will requeue */ redirty_page_for_writepage(wbc, page); @@ -3544,16 +3733,13 @@ static noinline_for_stack int __extent_writepage_io(struct btrfs_inode *inode, */ update_nr_written(wbc, nr_written + 1); - end = page_end; - blocksize = inode->vfs_inode.i_sb->s_blocksize; - while (cur <= end) { + u64 disk_bytenr; u64 em_end; - u64 offset; + u32 iosize; if (cur >= i_size) { - btrfs_writepage_endio_finish_ordered(page, cur, - page_end, 1); + btrfs_writepage_endio_finish_ordered(page, cur, end, 1); break; } em = btrfs_get_extent(inode, NULL, 0, cur, end - cur + 1); @@ -3565,13 +3751,20 @@ static noinline_for_stack int __extent_writepage_io(struct btrfs_inode *inode, extent_offset = cur - em->start; em_end = extent_map_end(em); - BUG_ON(em_end <= cur); - BUG_ON(end < cur); - iosize = min(em_end - cur, end - cur + 1); - iosize = ALIGN(iosize, blocksize); - offset = em->block_start + extent_offset; + ASSERT(cur <= em_end); + ASSERT(cur < end); + ASSERT(IS_ALIGNED(em->start, fs_info->sectorsize)); + ASSERT(IS_ALIGNED(em->len, fs_info->sectorsize)); block_start = em->block_start; compressed = test_bit(EXTENT_FLAG_COMPRESSED, &em->flags); + disk_bytenr = em->block_start + extent_offset; + + /* Note that em_end from extent_map_end() is exclusive */ + iosize = min(em_end, end + 1) - cur; + + if (btrfs_use_zone_append(inode, em)) + opf = REQ_OP_ZONE_APPEND; + free_extent_map(em); em = NULL; @@ -3587,7 +3780,6 @@ static noinline_for_stack int __extent_writepage_io(struct btrfs_inode *inode, btrfs_writepage_endio_finish_ordered(page, cur, cur + iosize - 1, 1); cur += iosize; - pg_offset += iosize; continue; } @@ -3598,9 +3790,9 @@ static noinline_for_stack int __extent_writepage_io(struct btrfs_inode *inode, page->index, cur, end); } - ret = submit_extent_page(REQ_OP_WRITE | write_flags, wbc, - page, offset, iosize, pg_offset, - &epd->bio, + ret = submit_extent_page(opf | write_flags, wbc, page, + disk_bytenr, iosize, + cur - page_offset(page), &epd->bio, end_bio_extent_writepage, 0, 0, 0, false); if (ret) { @@ -3609,8 +3801,7 @@ static noinline_for_stack int __extent_writepage_io(struct btrfs_inode *inode, end_page_writeback(page); } - cur = cur + iosize; - pg_offset += iosize; + cur += iosize; nr++; } *nr_ret = nr; @@ -3663,7 +3854,11 @@ static int __extent_writepage(struct page *page, struct writeback_control *wbc, flush_dcache_page(page); } - set_page_extent_mapped(page); + ret = set_page_extent_mapped(page); + if (ret < 0) { + SetPageError(page); + goto done; + } if (!epd->extent_locked) { ret = writepage_delalloc(BTRFS_I(inode), page, wbc, start, @@ -3923,7 +4118,7 @@ static noinline_for_stack int write_one_eb(struct extent_buffer *eb, struct writeback_control *wbc, struct extent_page_data *epd) { - u64 offset = eb->start; + u64 disk_bytenr = eb->start; u32 nritems; int i, num_pages; unsigned long start, end; @@ -3956,7 +4151,7 @@ static noinline_for_stack int write_one_eb(struct extent_buffer *eb, clear_page_dirty_for_io(p); set_page_writeback(p); ret = submit_extent_page(REQ_OP_WRITE | write_flags, wbc, - p, offset, PAGE_SIZE, 0, + p, disk_bytenr, PAGE_SIZE, 0, &epd->bio, end_bio_extent_buffer_writepage, 0, 0, 0, false); @@ -3969,7 +4164,7 @@ static noinline_for_stack int write_one_eb(struct extent_buffer *eb, ret = -EIO; break; } - offset += PAGE_SIZE; + disk_bytenr += PAGE_SIZE; update_nr_written(wbc, 1); unlock_page(p); } @@ -4010,6 +4205,7 @@ static int submit_eb_page(struct page *page, struct writeback_control *wbc, struct extent_buffer **eb_context) { struct address_space *mapping = page->mapping; + struct btrfs_block_group *cache = NULL; struct extent_buffer *eb; int ret; @@ -4042,13 +4238,31 @@ static int submit_eb_page(struct page *page, struct writeback_control *wbc, if (!ret) return 0; + if (!btrfs_check_meta_write_pointer(eb->fs_info, eb, &cache)) { + /* + * If for_sync, this hole will be filled with + * trasnsaction commit. + */ + if (wbc->sync_mode == WB_SYNC_ALL && !wbc->for_sync) + ret = -EAGAIN; + else + ret = 0; + free_extent_buffer(eb); + return ret; + } + *eb_context = eb; ret = lock_extent_buffer_for_io(eb, epd); if (ret <= 0) { + btrfs_revert_meta_write_pointer(cache, eb); + if (cache) + btrfs_put_block_group(cache); free_extent_buffer(eb); return ret; } + if (cache) + btrfs_put_block_group(cache); ret = write_one_eb(eb, wbc, epd); free_extent_buffer(eb); if (ret < 0) @@ -4094,6 +4308,7 @@ int btree_write_cache_pages(struct address_space *mapping, tag = PAGECACHE_TAG_TOWRITE; else tag = PAGECACHE_TAG_DIRTY; + btrfs_zoned_meta_io_lock(fs_info); retry: if (wbc->sync_mode == WB_SYNC_ALL) tag_pages_for_writeback(mapping, index, end); @@ -4134,7 +4349,7 @@ retry: } if (ret < 0) { end_write_bio(&epd, ret); - return ret; + goto out; } /* * If something went wrong, don't allow any metadata write bio to be @@ -4169,14 +4384,17 @@ retry: ret = -EROFS; end_write_bio(&epd, ret); } +out: + btrfs_zoned_meta_io_unlock(fs_info); return ret; } /** - * write_cache_pages - walk the list of dirty pages of the given address space and write all of them. + * Walk the list of dirty pages of the given address space and write all of them. + * * @mapping: address space structure to write - * @wbc: subtract the number of written pages from *@wbc->nr_to_write - * @data: data passed to __extent_writepage function + * @wbc: subtract the number of written pages from *@wbc->nr_to_write + * @epd: holds context for the write, namely the bio * * If a page is already under I/O, write_cache_pages() skips it, even * if it's dirty. This is desirable behaviour for memory-cleaning writeback, @@ -4975,25 +5193,39 @@ int extent_buffer_under_io(const struct extent_buffer *eb) test_bit(EXTENT_BUFFER_DIRTY, &eb->bflags)); } -/* - * Release all pages attached to the extent buffer. - */ -static void btrfs_release_extent_buffer_pages(struct extent_buffer *eb) +static bool page_range_has_eb(struct btrfs_fs_info *fs_info, struct page *page) { - int i; - int num_pages; - int mapped = !test_bit(EXTENT_BUFFER_UNMAPPED, &eb->bflags); + struct btrfs_subpage *subpage; - BUG_ON(extent_buffer_under_io(eb)); + lockdep_assert_held(&page->mapping->private_lock); - num_pages = num_extent_pages(eb); - for (i = 0; i < num_pages; i++) { - struct page *page = eb->pages[i]; + if (PagePrivate(page)) { + subpage = (struct btrfs_subpage *)page->private; + if (atomic_read(&subpage->eb_refs)) + return true; + } + return false; +} - if (!page) - continue; +static void detach_extent_buffer_page(struct extent_buffer *eb, struct page *page) +{ + struct btrfs_fs_info *fs_info = eb->fs_info; + const bool mapped = !test_bit(EXTENT_BUFFER_UNMAPPED, &eb->bflags); + + /* + * For mapped eb, we're going to change the page private, which should + * be done under the private_lock. + */ + if (mapped) + spin_lock(&page->mapping->private_lock); + + if (!PagePrivate(page)) { if (mapped) - spin_lock(&page->mapping->private_lock); + spin_unlock(&page->mapping->private_lock); + return; + } + + if (fs_info->sectorsize == PAGE_SIZE) { /* * We do this since we'll remove the pages after we've * removed the eb from the radix tree, so we could race @@ -5012,9 +5244,49 @@ static void btrfs_release_extent_buffer_pages(struct extent_buffer *eb) */ detach_page_private(page); } - if (mapped) spin_unlock(&page->mapping->private_lock); + return; + } + + /* + * For subpage, we can have dummy eb with page private. In this case, + * we can directly detach the private as such page is only attached to + * one dummy eb, no sharing. + */ + if (!mapped) { + btrfs_detach_subpage(fs_info, page); + return; + } + + btrfs_page_dec_eb_refs(fs_info, page); + + /* + * We can only detach the page private if there are no other ebs in the + * page range. + */ + if (!page_range_has_eb(fs_info, page)) + btrfs_detach_subpage(fs_info, page); + + spin_unlock(&page->mapping->private_lock); +} + +/* Release all pages attached to the extent buffer */ +static void btrfs_release_extent_buffer_pages(struct extent_buffer *eb) +{ + int i; + int num_pages; + + ASSERT(!extent_buffer_under_io(eb)); + + num_pages = num_extent_pages(eb); + for (i = 0; i < num_pages; i++) { + struct page *page = eb->pages[i]; + + if (!page) + continue; + + detach_extent_buffer_page(eb, page); /* One for when we allocated the page */ put_page(page); @@ -5046,6 +5318,7 @@ __alloc_extent_buffer(struct btrfs_fs_info *fs_info, u64 start, btrfs_leak_debug_add(&fs_info->eb_leak_lock, &eb->leak_list, &fs_info->allocated_ebs); + INIT_LIST_HEAD(&eb->release_list); spin_lock_init(&eb->refs_lock); atomic_set(&eb->refs, 1); @@ -5067,21 +5340,32 @@ struct extent_buffer *btrfs_clone_extent_buffer(const struct extent_buffer *src) if (new == NULL) return NULL; + /* + * Set UNMAPPED before calling btrfs_release_extent_buffer(), as + * btrfs_release_extent_buffer() have different behavior for + * UNMAPPED subpage extent buffer. + */ + set_bit(EXTENT_BUFFER_UNMAPPED, &new->bflags); + for (i = 0; i < num_pages; i++) { + int ret; + p = alloc_page(GFP_NOFS); if (!p) { btrfs_release_extent_buffer(new); return NULL; } - attach_extent_buffer_page(new, p); + ret = attach_extent_buffer_page(new, p, NULL); + if (ret < 0) { + put_page(p); + btrfs_release_extent_buffer(new); + return NULL; + } WARN_ON(PageDirty(p)); - SetPageUptodate(p); new->pages[i] = p; copy_page(page_address(p), page_address(src->pages[i])); } - - set_bit(EXTENT_BUFFER_UPTODATE, &new->bflags); - set_bit(EXTENT_BUFFER_UNMAPPED, &new->bflags); + set_extent_buffer_uptodate(new); return new; } @@ -5099,9 +5383,14 @@ struct extent_buffer *__alloc_dummy_extent_buffer(struct btrfs_fs_info *fs_info, num_pages = num_extent_pages(eb); for (i = 0; i < num_pages; i++) { + int ret; + eb->pages[i] = alloc_page(GFP_NOFS); if (!eb->pages[i]) goto err; + ret = attach_extent_buffer_page(eb, eb->pages[i], NULL); + if (ret < 0) + goto err; } set_extent_buffer_uptodate(eb); btrfs_set_header_nritems(eb, 0); @@ -5109,8 +5398,10 @@ struct extent_buffer *__alloc_dummy_extent_buffer(struct btrfs_fs_info *fs_info, return eb; err: - for (; i > 0; i--) + for (; i > 0; i--) { + detach_extent_buffer_page(eb, eb->pages[i - 1]); __free_page(eb->pages[i - 1]); + } __free_extent_buffer(eb); return NULL; } @@ -5252,6 +5543,38 @@ free_eb: } #endif +static struct extent_buffer *grab_extent_buffer( + struct btrfs_fs_info *fs_info, struct page *page) +{ + struct extent_buffer *exists; + + /* + * For subpage case, we completely rely on radix tree to ensure we + * don't try to insert two ebs for the same bytenr. So here we always + * return NULL and just continue. + */ + if (fs_info->sectorsize < PAGE_SIZE) + return NULL; + + /* Page not yet attached to an extent buffer */ + if (!PagePrivate(page)) + return NULL; + + /* + * We could have already allocated an eb for this page and attached one + * so lets see if we can get a ref on the existing eb, and if we can we + * know it's good and we can just return that one, else we know we can + * just overwrite page->private. + */ + exists = (struct extent_buffer *)page->private; + if (atomic_inc_not_zero(&exists->refs)) + return exists; + + WARN_ON(PageDirty(page)); + detach_page_private(page); + return NULL; +} + struct extent_buffer *alloc_extent_buffer(struct btrfs_fs_info *fs_info, u64 start, u64 owner_root, int level) { @@ -5290,36 +5613,58 @@ struct extent_buffer *alloc_extent_buffer(struct btrfs_fs_info *fs_info, num_pages = num_extent_pages(eb); for (i = 0; i < num_pages; i++, index++) { + struct btrfs_subpage *prealloc = NULL; + p = find_or_create_page(mapping, index, GFP_NOFS|__GFP_NOFAIL); if (!p) { exists = ERR_PTR(-ENOMEM); goto free_eb; } - spin_lock(&mapping->private_lock); - if (PagePrivate(p)) { - /* - * We could have already allocated an eb for this page - * and attached one so lets see if we can get a ref on - * the existing eb, and if we can we know it's good and - * we can just return that one, else we know we can just - * overwrite page->private. - */ - exists = (struct extent_buffer *)p->private; - if (atomic_inc_not_zero(&exists->refs)) { - spin_unlock(&mapping->private_lock); - unlock_page(p); - put_page(p); - mark_extent_buffer_accessed(exists, p); - goto free_eb; - } - exists = NULL; + /* + * Preallocate page->private for subpage case, so that we won't + * allocate memory with private_lock hold. The memory will be + * freed by attach_extent_buffer_page() or freed manually if + * we exit earlier. + * + * Although we have ensured one subpage eb can only have one + * page, but it may change in the future for 16K page size + * support, so we still preallocate the memory in the loop. + */ + ret = btrfs_alloc_subpage(fs_info, &prealloc, + BTRFS_SUBPAGE_METADATA); + if (ret < 0) { + unlock_page(p); + put_page(p); + exists = ERR_PTR(ret); + goto free_eb; + } - WARN_ON(PageDirty(p)); - detach_page_private(p); + spin_lock(&mapping->private_lock); + exists = grab_extent_buffer(fs_info, p); + if (exists) { + spin_unlock(&mapping->private_lock); + unlock_page(p); + put_page(p); + mark_extent_buffer_accessed(exists, p); + btrfs_free_subpage(prealloc); + goto free_eb; } - attach_extent_buffer_page(eb, p); + /* Should not fail, as we have preallocated the memory */ + ret = attach_extent_buffer_page(eb, p, prealloc); + ASSERT(!ret); + /* + * To inform we have extra eb under allocation, so that + * detach_extent_buffer_page() won't release the page private + * when the eb hasn't yet been inserted into radix tree. + * + * The ref will be decreased when the eb released the page, in + * detach_extent_buffer_page(). + * Thus needs no special handling in error path. + */ + btrfs_page_inc_eb_refs(fs_info, p); spin_unlock(&mapping->private_lock); + WARN_ON(PageDirty(p)); eb->pages[i] = p; if (!PageUptodate(p)) @@ -5525,33 +5870,103 @@ bool set_extent_buffer_dirty(struct extent_buffer *eb) void clear_extent_buffer_uptodate(struct extent_buffer *eb) { - int i; + struct btrfs_fs_info *fs_info = eb->fs_info; struct page *page; int num_pages; + int i; clear_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags); num_pages = num_extent_pages(eb); for (i = 0; i < num_pages; i++) { page = eb->pages[i]; if (page) - ClearPageUptodate(page); + btrfs_page_clear_uptodate(fs_info, page, + eb->start, eb->len); } } void set_extent_buffer_uptodate(struct extent_buffer *eb) { - int i; + struct btrfs_fs_info *fs_info = eb->fs_info; struct page *page; int num_pages; + int i; set_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags); num_pages = num_extent_pages(eb); for (i = 0; i < num_pages; i++) { page = eb->pages[i]; - SetPageUptodate(page); + btrfs_page_set_uptodate(fs_info, page, eb->start, eb->len); } } +static int read_extent_buffer_subpage(struct extent_buffer *eb, int wait, + int mirror_num) +{ + struct btrfs_fs_info *fs_info = eb->fs_info; + struct extent_io_tree *io_tree; + struct page *page = eb->pages[0]; + struct bio *bio = NULL; + int ret = 0; + + ASSERT(!test_bit(EXTENT_BUFFER_UNMAPPED, &eb->bflags)); + ASSERT(PagePrivate(page)); + io_tree = &BTRFS_I(fs_info->btree_inode)->io_tree; + + if (wait == WAIT_NONE) { + ret = try_lock_extent(io_tree, eb->start, + eb->start + eb->len - 1); + if (ret <= 0) + return ret; + } else { + ret = lock_extent(io_tree, eb->start, eb->start + eb->len - 1); + if (ret < 0) + return ret; + } + + ret = 0; + if (test_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags) || + PageUptodate(page) || + btrfs_subpage_test_uptodate(fs_info, page, eb->start, eb->len)) { + set_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags); + unlock_extent(io_tree, eb->start, eb->start + eb->len - 1); + return ret; + } + + clear_bit(EXTENT_BUFFER_READ_ERR, &eb->bflags); + eb->read_mirror = 0; + atomic_set(&eb->io_pages, 1); + check_buffer_tree_ref(eb); + btrfs_subpage_clear_error(fs_info, page, eb->start, eb->len); + + ret = submit_extent_page(REQ_OP_READ | REQ_META, NULL, page, eb->start, + eb->len, eb->start - page_offset(page), &bio, + end_bio_extent_readpage, mirror_num, 0, 0, + true); + if (ret) { + /* + * In the endio function, if we hit something wrong we will + * increase the io_pages, so here we need to decrease it for + * error path. + */ + atomic_dec(&eb->io_pages); + } + if (bio) { + int tmp; + + tmp = submit_one_bio(bio, mirror_num, 0); + if (tmp < 0) + return tmp; + } + if (ret || wait != WAIT_COMPLETE) + return ret; + + wait_extent_bit(io_tree, eb->start, eb->start + eb->len - 1, EXTENT_LOCKED); + if (!test_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags)) + ret = -EIO; + return ret; +} + int read_extent_buffer_pages(struct extent_buffer *eb, int wait, int mirror_num) { int i; @@ -5568,10 +5983,20 @@ int read_extent_buffer_pages(struct extent_buffer *eb, int wait, int mirror_num) if (test_bit(EXTENT_BUFFER_UPTODATE, &eb->bflags)) return 0; + if (eb->fs_info->sectorsize < PAGE_SIZE) + return read_extent_buffer_subpage(eb, wait, mirror_num); + num_pages = num_extent_pages(eb); for (i = 0; i < num_pages; i++) { page = eb->pages[i]; if (wait == WAIT_NONE) { + /* + * WAIT_NONE is only utilized by readahead. If we can't + * acquire the lock atomically it means either the eb + * is being read out or under modification. + * Either way the eb will be or has been cached, + * readahead can exit safely. + */ if (!trylock_page(page)) goto unlock_exit; } else { @@ -5823,6 +6248,8 @@ void write_extent_buffer(const struct extent_buffer *eb, const void *srcv, char *src = (char *)srcv; unsigned long i = get_eb_page_index(start); + WARN_ON(test_bit(EXTENT_BUFFER_NO_CHECK, &eb->bflags)); + if (check_eb_range(eb, start, len)) return; @@ -6169,13 +6596,115 @@ void memmove_extent_buffer(const struct extent_buffer *dst, } } +static struct extent_buffer *get_next_extent_buffer( + struct btrfs_fs_info *fs_info, struct page *page, u64 bytenr) +{ + struct extent_buffer *gang[BTRFS_SUBPAGE_BITMAP_SIZE]; + struct extent_buffer *found = NULL; + u64 page_start = page_offset(page); + int ret; + int i; + + ASSERT(in_range(bytenr, page_start, PAGE_SIZE)); + ASSERT(PAGE_SIZE / fs_info->nodesize <= BTRFS_SUBPAGE_BITMAP_SIZE); + lockdep_assert_held(&fs_info->buffer_lock); + + ret = radix_tree_gang_lookup(&fs_info->buffer_radix, (void **)gang, + bytenr >> fs_info->sectorsize_bits, + PAGE_SIZE / fs_info->nodesize); + for (i = 0; i < ret; i++) { + /* Already beyond page end */ + if (gang[i]->start >= page_start + PAGE_SIZE) + break; + /* Found one */ + if (gang[i]->start >= bytenr) { + found = gang[i]; + break; + } + } + return found; +} + +static int try_release_subpage_extent_buffer(struct page *page) +{ + struct btrfs_fs_info *fs_info = btrfs_sb(page->mapping->host->i_sb); + u64 cur = page_offset(page); + const u64 end = page_offset(page) + PAGE_SIZE; + int ret; + + while (cur < end) { + struct extent_buffer *eb = NULL; + + /* + * Unlike try_release_extent_buffer() which uses page->private + * to grab buffer, for subpage case we rely on radix tree, thus + * we need to ensure radix tree consistency. + * + * We also want an atomic snapshot of the radix tree, thus go + * with spinlock rather than RCU. + */ + spin_lock(&fs_info->buffer_lock); + eb = get_next_extent_buffer(fs_info, page, cur); + if (!eb) { + /* No more eb in the page range after or at cur */ + spin_unlock(&fs_info->buffer_lock); + break; + } + cur = eb->start + eb->len; + + /* + * The same as try_release_extent_buffer(), to ensure the eb + * won't disappear out from under us. + */ + spin_lock(&eb->refs_lock); + if (atomic_read(&eb->refs) != 1 || extent_buffer_under_io(eb)) { + spin_unlock(&eb->refs_lock); + spin_unlock(&fs_info->buffer_lock); + break; + } + spin_unlock(&fs_info->buffer_lock); + + /* + * If tree ref isn't set then we know the ref on this eb is a + * real ref, so just return, this eb will likely be freed soon + * anyway. + */ + if (!test_and_clear_bit(EXTENT_BUFFER_TREE_REF, &eb->bflags)) { + spin_unlock(&eb->refs_lock); + break; + } + + /* + * Here we don't care about the return value, we will always + * check the page private at the end. And + * release_extent_buffer() will release the refs_lock. + */ + release_extent_buffer(eb); + } + /* + * Finally to check if we have cleared page private, as if we have + * released all ebs in the page, the page private should be cleared now. + */ + spin_lock(&page->mapping->private_lock); + if (!PagePrivate(page)) + ret = 1; + else + ret = 0; + spin_unlock(&page->mapping->private_lock); + return ret; + +} + int try_release_extent_buffer(struct page *page) { struct extent_buffer *eb; + if (btrfs_sb(page->mapping->host->i_sb)->sectorsize < PAGE_SIZE) + return try_release_subpage_extent_buffer(page); + /* - * We need to make sure nobody is attaching this page to an eb right - * now. + * We need to make sure nobody is changing page->private, as we rely on + * page->private as the pointer to extent buffer. */ spin_lock(&page->mapping->private_lock); if (!PagePrivate(page)) { diff --git a/fs/btrfs/extent_io.h b/fs/btrfs/extent_io.h index 19221095c635..824640cb0ace 100644 --- a/fs/btrfs/extent_io.h +++ b/fs/btrfs/extent_io.h @@ -31,16 +31,17 @@ enum { EXTENT_BUFFER_IN_TREE, /* write IO error */ EXTENT_BUFFER_WRITE_ERR, + EXTENT_BUFFER_NO_CHECK, }; /* these are flags for __process_pages_contig */ #define PAGE_UNLOCK (1 << 0) -#define PAGE_CLEAR_DIRTY (1 << 1) -#define PAGE_SET_WRITEBACK (1 << 2) -#define PAGE_END_WRITEBACK (1 << 3) -#define PAGE_SET_PRIVATE2 (1 << 4) -#define PAGE_SET_ERROR (1 << 5) -#define PAGE_LOCK (1 << 6) +/* Page starts writeback, clear dirty bit and set writeback bit */ +#define PAGE_START_WRITEBACK (1 << 1) +#define PAGE_END_WRITEBACK (1 << 2) +#define PAGE_SET_PRIVATE2 (1 << 3) +#define PAGE_SET_ERROR (1 << 4) +#define PAGE_LOCK (1 << 5) /* * page->private values. Every page that is controlled by the extent @@ -93,6 +94,7 @@ struct extent_buffer { struct rw_semaphore lock; struct page *pages[INLINE_EXTENT_BUFFER_PAGES]; + struct list_head release_list; #ifdef CONFIG_BTRFS_DEBUG struct list_head leak_list; #endif @@ -178,7 +180,8 @@ int btree_write_cache_pages(struct address_space *mapping, void extent_readahead(struct readahead_control *rac); int extent_fiemap(struct btrfs_inode *inode, struct fiemap_extent_info *fieinfo, u64 start, u64 len); -void set_page_extent_mapped(struct page *page); +int set_page_extent_mapped(struct page *page); +void clear_page_extent_mapped(struct page *page); struct extent_buffer *alloc_extent_buffer(struct btrfs_fs_info *fs_info, u64 start, u64 owner_root, int level); diff --git a/fs/btrfs/extent_map.c b/fs/btrfs/extent_map.c index bd6229fb2b6f..4a8e02f7b6c7 100644 --- a/fs/btrfs/extent_map.c +++ b/fs/btrfs/extent_map.c @@ -385,9 +385,12 @@ static void extent_map_device_clear_bits(struct extent_map *em, unsigned bits) } /** - * add_extent_mapping - add new extent map to the extent tree + * Add new extent map to the extent tree + * * @tree: tree to insert new map in * @em: map to insert + * @modified: indicate whether the given @em should be added to the + * modified list, which indicates the extent needs to be logged * * Insert @em into @tree or perform a simple forward/backward merge with * existing mappings. The extent_map struct passed in will be inserted @@ -574,12 +577,13 @@ static noinline int merge_extent_mapping(struct extent_map_tree *em_tree, } /** - * btrfs_add_extent_mapping - add extent mapping into em_tree - * @fs_info - used for tracepoint - * @em_tree - the extent tree into which we want to insert the extent mapping - * @em_in - extent we are inserting - * @start - start of the logical range btrfs_get_extent() is requesting - * @len - length of the logical range btrfs_get_extent() is requesting + * Add extent mapping into em_tree + * + * @fs_info: the filesystem + * @em_tree: extent tree into which we want to insert the extent mapping + * @em_in: extent we are inserting + * @start: start of the logical range btrfs_get_extent() is requesting + * @len: length of the logical range btrfs_get_extent() is requesting * * Note that @em_in's range may be different from [start, start+len), * but they must be overlapped. diff --git a/fs/btrfs/file-item.c b/fs/btrfs/file-item.c index 6ccfc019ad90..47cd3a6dc635 100644 --- a/fs/btrfs/file-item.c +++ b/fs/btrfs/file-item.c @@ -24,8 +24,10 @@ PAGE_SIZE)) /** - * @inode - the inode we want to update the disk_i_size for - * @new_i_size - the i_size we want to set to, 0 if we use i_size + * Set inode's size according to filesystem options + * + * @inode: inode we want to update the disk_i_size for + * @new_i_size: i_size we want to set to, 0 if we use i_size * * With NO_HOLES set this simply sets the disk_is_size to whatever i_size_read() * returns as it is perfectly fine with a file that has holes without hole file @@ -62,9 +64,11 @@ void btrfs_inode_safe_disk_i_size_write(struct btrfs_inode *inode, u64 new_i_siz } /** - * @inode - the inode we're modifying - * @start - the start file offset of the file extent we've inserted - * @len - the logical length of the file extent item + * Mark range within a file as having a new extent inserted + * + * @inode: inode being modified + * @start: start file offset of the file extent we've inserted + * @len: logical length of the file extent item * * Call when we are inserting a new file extent where there was none before. * Does not need to call this in the case where we're replacing an existing file @@ -88,9 +92,11 @@ int btrfs_inode_set_file_extent_range(struct btrfs_inode *inode, u64 start, } /** - * @inode - the inode we're modifying - * @start - the start file offset of the file extent we've inserted - * @len - the logical length of the file extent item + * Marks an inode range as not having a backing extent + * + * @inode: inode being modified + * @start: start file offset of the file extent we've inserted + * @len: logical length of the file extent item * * Called when we drop a file extent, for example when we truncate. Doesn't * need to be called for cases where we're replacing a file extent, like when diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index 0e41459b8de6..0e155f013839 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -453,12 +453,11 @@ static void btrfs_drop_pages(struct page **pages, size_t num_pages) } /* - * after copy_from_user, pages need to be dirtied and we need to make - * sure holes are created between the current EOF and the start of - * any next extents (if required). - * - * this also makes the decision about creating an inline extent vs - * doing real data extents, marking pages dirty and delalloc as required. + * After btrfs_copy_from_user(), update the following things for delalloc: + * - Mark newly dirtied pages as DELALLOC in the io tree. + * Used to advise which range is to be written back. + * - Mark modified pages as Uptodate/Dirty and not needing COW fixup + * - Update inode size for past EOF write */ int btrfs_dirty_pages(struct btrfs_inode *inode, struct page **pages, size_t num_pages, loff_t pos, size_t write_bytes, @@ -1370,6 +1369,12 @@ again: goto fail; } + err = set_page_extent_mapped(pages[i]); + if (err < 0) { + faili = i; + goto fail; + } + if (i == 0) err = prepare_uptodate_page(inode, pages[i], pos, force_uptodate); @@ -1454,23 +1459,11 @@ lock_and_cleanup_extent_if_need(struct btrfs_inode *inode, struct page **pages, } /* - * It's possible the pages are dirty right now, but we don't want - * to clean them yet because copy_from_user may catch a page fault - * and we might have to fall back to one page at a time. If that - * happens, we'll unlock these pages and we'd have a window where - * reclaim could sneak in and drop the once-dirty page on the floor - * without writing it. - * - * We have the pages locked and the extent range locked, so there's - * no way someone can start IO on any dirty pages in this range. - * - * We'll call btrfs_dirty_pages() later on, and that will flip around - * delalloc bits and dirty the pages as required. + * We should be called after prepare_pages() which should have locked + * all pages in the range. */ - for (i = 0; i < num_pages; i++) { - set_page_extent_mapped(pages[i]); + for (i = 0; i < num_pages; i++) WARN_ON(!PageLocked(pages[i])); - } return ret; } @@ -1949,8 +1942,8 @@ relock: goto buffered; } - dio = __iomap_dio_rw(iocb, from, &btrfs_dio_iomap_ops, - &btrfs_dio_ops, is_sync_kiocb(iocb)); + dio = __iomap_dio_rw(iocb, from, &btrfs_dio_iomap_ops, &btrfs_dio_ops, + 0); btrfs_inode_unlock(inode, ilock_flags); @@ -1997,9 +1990,7 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb, struct iov_iter *from) { struct file *file = iocb->ki_filp; - struct inode *inode = file_inode(file); - struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); - struct btrfs_root *root = BTRFS_I(inode)->root; + struct btrfs_inode *inode = BTRFS_I(file_inode(file)); ssize_t num_written = 0; const bool sync = iocb->ki_flags & IOCB_DSYNC; @@ -2008,7 +1999,7 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb, * have opened a file as writable, we have to stop this write operation * to ensure consistency. */ - if (test_bit(BTRFS_FS_STATE_ERROR, &fs_info->fs_state)) + if (test_bit(BTRFS_FS_STATE_ERROR, &inode->root->fs_info->fs_state)) return -EROFS; if (!(iocb->ki_flags & IOCB_DIRECT) && @@ -2016,7 +2007,7 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb, return -EOPNOTSUPP; if (sync) - atomic_inc(&BTRFS_I(inode)->sync_writers); + atomic_inc(&inode->sync_writers); if (iocb->ki_flags & IOCB_DIRECT) num_written = btrfs_direct_write(iocb, from); @@ -2028,14 +2019,14 @@ static ssize_t btrfs_file_write_iter(struct kiocb *iocb, * otherwise subsequent syncs to a file that's been synced in this * transaction will appear to have already occurred. */ - spin_lock(&BTRFS_I(inode)->lock); - BTRFS_I(inode)->last_sub_trans = root->log_transid; - spin_unlock(&BTRFS_I(inode)->lock); + spin_lock(&inode->lock); + inode->last_sub_trans = inode->root->log_transid; + spin_unlock(&inode->lock); if (num_written > 0) num_written = generic_write_sync(iocb, num_written); if (sync) - atomic_dec(&BTRFS_I(inode)->sync_writers); + atomic_dec(&inode->sync_writers); current->backing_dev_info = NULL; return num_written; @@ -2177,8 +2168,12 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync) * commit waits for their completion, to avoid data loss if we fsync, * the current transaction commits before the ordered extents complete * and a power failure happens right after that. + * + * For zoned filesystem, if a write IO uses a ZONE_APPEND command, the + * logical address recorded in the ordered extent may change. We need + * to wait for the IO to stabilize the logical address. */ - if (full_sync) { + if (full_sync || btrfs_is_zoned(fs_info)) { ret = btrfs_wait_ordered_range(inode, start, len); } else { /* @@ -2241,6 +2236,7 @@ int btrfs_sync_file(struct file *file, loff_t start, loff_t end, int datasync) ret = PTR_ERR(trans); goto out_release_extents; } + trans->in_fsync = true; ret = btrfs_log_dentry_safe(trans, dentry, &ctx); btrfs_release_log_ctx_extents(&ctx); @@ -3264,8 +3260,11 @@ reserve_space: goto out; ret = btrfs_qgroup_reserve_data(BTRFS_I(inode), &data_reserved, alloc_start, bytes_to_reserve); - if (ret) + if (ret) { + unlock_extent_cached(&BTRFS_I(inode)->io_tree, lockstart, + lockend, &cached_state); goto out; + } ret = btrfs_prealloc_file_range(inode, mode, alloc_start, alloc_end - alloc_start, i_blocksize(inode), @@ -3622,8 +3621,7 @@ static ssize_t btrfs_direct_read(struct kiocb *iocb, struct iov_iter *to) return 0; btrfs_inode_lock(inode, BTRFS_ILOCK_SHARED); - ret = iomap_dio_rw(iocb, to, &btrfs_dio_iomap_ops, &btrfs_dio_ops, - is_sync_kiocb(iocb)); + ret = iomap_dio_rw(iocb, to, &btrfs_dio_iomap_ops, &btrfs_dio_ops, 0); btrfs_inode_unlock(inode, BTRFS_ILOCK_SHARED); return ret; } @@ -3639,7 +3637,7 @@ static ssize_t btrfs_file_read_iter(struct kiocb *iocb, struct iov_iter *to) return ret; } - return generic_file_buffered_read(iocb, to, ret); + return filemap_read(iocb, to, ret); } const struct file_operations btrfs_file_operations = { diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c index 4d8897879c9c..9988decd5717 100644 --- a/fs/btrfs/free-space-cache.c +++ b/fs/btrfs/free-space-cache.c @@ -198,7 +198,7 @@ int create_free_space_inode(struct btrfs_trans_handle *trans, int ret; u64 ino; - ret = btrfs_find_free_objectid(trans->fs_info->tree_root, &ino); + ret = btrfs_get_free_objectid(trans->fs_info->tree_root, &ino); if (ret < 0) return ret; @@ -431,11 +431,22 @@ static int io_ctl_prepare_pages(struct btrfs_io_ctl *io_ctl, bool uptodate) int i; for (i = 0; i < io_ctl->num_pages; i++) { + int ret; + page = find_or_create_page(inode->i_mapping, i, mask); if (!page) { io_ctl_drop_pages(io_ctl); return -ENOMEM; } + + ret = set_page_extent_mapped(page); + if (ret < 0) { + unlock_page(page); + put_page(page); + io_ctl_drop_pages(io_ctl); + return ret; + } + io_ctl->pages[i] = page; if (uptodate && !PageUptodate(page)) { btrfs_readpage(NULL, page); @@ -455,10 +466,8 @@ static int io_ctl_prepare_pages(struct btrfs_io_ctl *io_ctl, bool uptodate) } } - for (i = 0; i < io_ctl->num_pages; i++) { + for (i = 0; i < io_ctl->num_pages; i++) clear_page_dirty_for_io(io_ctl->pages[i]); - set_page_extent_mapped(io_ctl->pages[i]); - } return 0; } @@ -775,8 +784,10 @@ static int __load_free_space_cache(struct btrfs_root *root, struct inode *inode, while (num_entries) { e = kmem_cache_zalloc(btrfs_free_space_cachep, GFP_NOFS); - if (!e) + if (!e) { + ret = -ENOMEM; goto free_cache; + } ret = io_ctl_read_entry(&io_ctl, e, &type); if (ret) { @@ -785,6 +796,7 @@ static int __load_free_space_cache(struct btrfs_root *root, struct inode *inode, } if (!e->bytes) { + ret = -1; kmem_cache_free(btrfs_free_space_cachep, e); goto free_cache; } @@ -805,6 +817,7 @@ static int __load_free_space_cache(struct btrfs_root *root, struct inode *inode, e->bitmap = kmem_cache_zalloc( btrfs_free_space_bitmap_cachep, GFP_NOFS); if (!e->bitmap) { + ret = -ENOMEM; kmem_cache_free( btrfs_free_space_cachep, e); goto free_cache; @@ -1295,11 +1308,14 @@ int btrfs_wait_cache_io(struct btrfs_trans_handle *trans, } /** - * __btrfs_write_out_cache - write out cached info to an inode - * @root - the root the inode belongs to - * @ctl - the free space cache we are going to write out - * @block_group - the block_group for this cache if it belongs to a block_group - * @trans - the trans handle + * Write out cached info to an inode + * + * @root: root the inode belongs to + * @inode: freespace inode we are writing out + * @ctl: free space cache we are going to write out + * @block_group: block_group for this cache if it belongs to a block_group + * @io_ctl: holds context for the io + * @trans: the trans handle * * This function writes out a free space cache struct to disk for quick recovery * on mount. This will return 0 if it was successful in writing the cache out, @@ -2461,6 +2477,8 @@ int __btrfs_add_free_space(struct btrfs_fs_info *fs_info, int ret = 0; u64 filter_bytes = bytes; + ASSERT(!btrfs_is_zoned(fs_info)); + info = kmem_cache_zalloc(btrfs_free_space_cachep, GFP_NOFS); if (!info) return -ENOMEM; @@ -2518,11 +2536,54 @@ out: return ret; } +static int __btrfs_add_free_space_zoned(struct btrfs_block_group *block_group, + u64 bytenr, u64 size, bool used) +{ + struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl; + u64 offset = bytenr - block_group->start; + u64 to_free, to_unusable; + + spin_lock(&ctl->tree_lock); + if (!used) + to_free = size; + else if (offset >= block_group->alloc_offset) + to_free = size; + else if (offset + size <= block_group->alloc_offset) + to_free = 0; + else + to_free = offset + size - block_group->alloc_offset; + to_unusable = size - to_free; + + ctl->free_space += to_free; + /* + * If the block group is read-only, we should account freed space into + * bytes_readonly. + */ + if (!block_group->ro) + block_group->zone_unusable += to_unusable; + spin_unlock(&ctl->tree_lock); + if (!used) { + spin_lock(&block_group->lock); + block_group->alloc_offset -= size; + spin_unlock(&block_group->lock); + } + + /* All the region is now unusable. Mark it as unused and reclaim */ + if (block_group->zone_unusable == block_group->length) + btrfs_mark_bg_unused(block_group); + + return 0; +} + int btrfs_add_free_space(struct btrfs_block_group *block_group, u64 bytenr, u64 size) { enum btrfs_trim_state trim_state = BTRFS_TRIM_STATE_UNTRIMMED; + if (btrfs_is_zoned(block_group->fs_info)) + return __btrfs_add_free_space_zoned(block_group, bytenr, size, + true); + if (btrfs_test_opt(block_group->fs_info, DISCARD_SYNC)) trim_state = BTRFS_TRIM_STATE_TRIMMED; @@ -2531,6 +2592,16 @@ int btrfs_add_free_space(struct btrfs_block_group *block_group, bytenr, size, trim_state); } +int btrfs_add_free_space_unused(struct btrfs_block_group *block_group, + u64 bytenr, u64 size) +{ + if (btrfs_is_zoned(block_group->fs_info)) + return __btrfs_add_free_space_zoned(block_group, bytenr, size, + false); + + return btrfs_add_free_space(block_group, bytenr, size); +} + /* * This is a subtle distinction because when adding free space back in general, * we want it to be added as untrimmed for async. But in the case where we add @@ -2541,6 +2612,10 @@ int btrfs_add_free_space_async_trimmed(struct btrfs_block_group *block_group, { enum btrfs_trim_state trim_state = BTRFS_TRIM_STATE_UNTRIMMED; + if (btrfs_is_zoned(block_group->fs_info)) + return __btrfs_add_free_space_zoned(block_group, bytenr, size, + true); + if (btrfs_test_opt(block_group->fs_info, DISCARD_SYNC) || btrfs_test_opt(block_group->fs_info, DISCARD_ASYNC)) trim_state = BTRFS_TRIM_STATE_TRIMMED; @@ -2558,6 +2633,23 @@ int btrfs_remove_free_space(struct btrfs_block_group *block_group, int ret; bool re_search = false; + if (btrfs_is_zoned(block_group->fs_info)) { + /* + * This can happen with conventional zones when replaying log. + * Since the allocation info of tree-log nodes are not recorded + * to the extent-tree, calculate_alloc_pointer() failed to + * advance the allocation pointer after last allocated tree log + * node blocks. + * + * This function is called from + * btrfs_pin_extent_for_log_replay() when replaying the log. + * Advance the pointer not to overwrite the tree-log nodes. + */ + if (block_group->alloc_offset < offset + bytes) + block_group->alloc_offset = offset + bytes; + return 0; + } + spin_lock(&ctl->tree_lock); again: @@ -2652,6 +2744,16 @@ void btrfs_dump_free_space(struct btrfs_block_group *block_group, struct rb_node *n; int count = 0; + /* + * Zoned btrfs does not use free space tree and cluster. Just print + * out the free space after the allocation offset. + */ + if (btrfs_is_zoned(fs_info)) { + btrfs_info(fs_info, "free space %llu", + block_group->length - block_group->alloc_offset); + return; + } + spin_lock(&ctl->tree_lock); for (n = rb_first(&ctl->free_space_offset); n; n = rb_next(n)) { info = rb_entry(n, struct btrfs_free_space, offset_index); @@ -2704,8 +2806,10 @@ static void __btrfs_return_cluster_to_free_space( struct rb_node *node; spin_lock(&cluster->lock); - if (cluster->block_group != block_group) - goto out; + if (cluster->block_group != block_group) { + spin_unlock(&cluster->lock); + return; + } cluster->block_group = NULL; cluster->window_start = 0; @@ -2743,8 +2847,6 @@ static void __btrfs_return_cluster_to_free_space( entry->offset, &entry->offset_index, bitmap); } cluster->root = RB_ROOT; - -out: spin_unlock(&cluster->lock); btrfs_put_block_group(block_group); } @@ -2845,6 +2947,8 @@ u64 btrfs_find_space_for_alloc(struct btrfs_block_group *block_group, u64 align_gap_len = 0; enum btrfs_trim_state align_gap_trim_state = BTRFS_TRIM_STATE_UNTRIMMED; + ASSERT(!btrfs_is_zoned(block_group->fs_info)); + spin_lock(&ctl->tree_lock); entry = find_free_space(ctl, &offset, &bytes_search, block_group->full_stripe_len, max_extent_size); @@ -2976,6 +3080,8 @@ u64 btrfs_alloc_from_cluster(struct btrfs_block_group *block_group, struct rb_node *node; u64 ret = 0; + ASSERT(!btrfs_is_zoned(block_group->fs_info)); + spin_lock(&cluster->lock); if (bytes > cluster->max_size) goto out; @@ -3024,8 +3130,6 @@ u64 btrfs_alloc_from_cluster(struct btrfs_block_group *block_group, entry->bytes -= bytes; } - if (entry->bytes == 0) - rb_erase(&entry->offset_index, &cluster->root); break; } out: @@ -3042,7 +3146,10 @@ out: ctl->free_space -= bytes; if (!entry->bitmap && !btrfs_free_space_trimmed(entry)) ctl->discardable_bytes[BTRFS_STAT_CURR] -= bytes; + + spin_lock(&cluster->lock); if (entry->bytes == 0) { + rb_erase(&entry->offset_index, &cluster->root); ctl->free_extents--; if (entry->bitmap) { kmem_cache_free(btrfs_free_space_bitmap_cachep, @@ -3055,6 +3162,7 @@ out: kmem_cache_free(btrfs_free_space_cachep, entry); } + spin_unlock(&cluster->lock); spin_unlock(&ctl->tree_lock); return ret; @@ -3752,6 +3860,8 @@ int btrfs_trim_block_group(struct btrfs_block_group *block_group, int ret; u64 rem = 0; + ASSERT(!btrfs_is_zoned(block_group->fs_info)); + *trimmed = 0; spin_lock(&block_group->lock); diff --git a/fs/btrfs/free-space-cache.h b/fs/btrfs/free-space-cache.h index ecb09a02d544..1f23088d43f9 100644 --- a/fs/btrfs/free-space-cache.h +++ b/fs/btrfs/free-space-cache.h @@ -107,6 +107,8 @@ int __btrfs_add_free_space(struct btrfs_fs_info *fs_info, enum btrfs_trim_state trim_state); int btrfs_add_free_space(struct btrfs_block_group *block_group, u64 bytenr, u64 size); +int btrfs_add_free_space_unused(struct btrfs_block_group *block_group, + u64 bytenr, u64 size); int btrfs_add_free_space_async_trimmed(struct btrfs_block_group *block_group, u64 bytenr, u64 size); int btrfs_remove_free_space(struct btrfs_block_group *block_group, diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index a8e0a6b038d3..7cdf65be3707 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -50,6 +50,7 @@ #include "delalloc-space.h" #include "block-group.h" #include "space-info.h" +#include "zoned.h" struct btrfs_iget_args { u64 ino; @@ -692,8 +693,7 @@ cont: NULL, clear_flags, PAGE_UNLOCK | - PAGE_CLEAR_DIRTY | - PAGE_SET_WRITEBACK | + PAGE_START_WRITEBACK | page_error_op | PAGE_END_WRITEBACK); @@ -917,7 +917,6 @@ retry: ins.objectid, async_extent->ram_size, ins.offset, - BTRFS_ORDERED_COMPRESSED, async_extent->compress_type); if (ret) { btrfs_drop_extent_cache(inode, async_extent->start, @@ -934,8 +933,7 @@ retry: async_extent->start + async_extent->ram_size - 1, NULL, EXTENT_LOCKED | EXTENT_DELALLOC, - PAGE_UNLOCK | PAGE_CLEAR_DIRTY | - PAGE_SET_WRITEBACK); + PAGE_UNLOCK | PAGE_START_WRITEBACK); if (btrfs_submit_compressed_write(inode, async_extent->start, async_extent->ram_size, ins.objectid, @@ -971,9 +969,8 @@ out_free: NULL, EXTENT_LOCKED | EXTENT_DELALLOC | EXTENT_DELALLOC_NEW | EXTENT_DEFRAG | EXTENT_DO_ACCOUNTING, - PAGE_UNLOCK | PAGE_CLEAR_DIRTY | - PAGE_SET_WRITEBACK | PAGE_END_WRITEBACK | - PAGE_SET_ERROR); + PAGE_UNLOCK | PAGE_START_WRITEBACK | + PAGE_END_WRITEBACK | PAGE_SET_ERROR); free_async_extent_pages(async_extent); kfree(async_extent); goto again; @@ -1071,8 +1068,7 @@ static noinline int cow_file_range(struct btrfs_inode *inode, EXTENT_LOCKED | EXTENT_DELALLOC | EXTENT_DELALLOC_NEW | EXTENT_DEFRAG | EXTENT_DO_ACCOUNTING, PAGE_UNLOCK | - PAGE_CLEAR_DIRTY | PAGE_SET_WRITEBACK | - PAGE_END_WRITEBACK); + PAGE_START_WRITEBACK | PAGE_END_WRITEBACK); *nr_written = *nr_written + (end - start + PAGE_SIZE) / PAGE_SIZE; *page_started = 1; @@ -1127,7 +1123,8 @@ static noinline int cow_file_range(struct btrfs_inode *inode, free_extent_map(em); ret = btrfs_add_ordered_extent(inode, start, ins.objectid, - ram_size, cur_alloc_size, 0); + ram_size, cur_alloc_size, + BTRFS_ORDERED_REGULAR); if (ret) goto out_drop_extent_cache; @@ -1194,8 +1191,7 @@ out_reserve: out_unlock: clear_bits = EXTENT_LOCKED | EXTENT_DELALLOC | EXTENT_DELALLOC_NEW | EXTENT_DEFRAG | EXTENT_CLEAR_META_RESV; - page_ops = PAGE_UNLOCK | PAGE_CLEAR_DIRTY | PAGE_SET_WRITEBACK | - PAGE_END_WRITEBACK; + page_ops = PAGE_UNLOCK | PAGE_START_WRITEBACK | PAGE_END_WRITEBACK; /* * If we reserved an extent for our delalloc range (or a subrange) and * failed to create the respective ordered extent, then it means that @@ -1320,9 +1316,8 @@ static int cow_file_range_async(struct btrfs_inode *inode, unsigned clear_bits = EXTENT_LOCKED | EXTENT_DELALLOC | EXTENT_DELALLOC_NEW | EXTENT_DEFRAG | EXTENT_DO_ACCOUNTING; - unsigned long page_ops = PAGE_UNLOCK | PAGE_CLEAR_DIRTY | - PAGE_SET_WRITEBACK | PAGE_END_WRITEBACK | - PAGE_SET_ERROR; + unsigned long page_ops = PAGE_UNLOCK | PAGE_START_WRITEBACK | + PAGE_END_WRITEBACK | PAGE_SET_ERROR; extent_clear_unlock_delalloc(inode, start, end, locked_page, clear_bits, page_ops); @@ -1399,6 +1394,29 @@ static int cow_file_range_async(struct btrfs_inode *inode, return 0; } +static noinline int run_delalloc_zoned(struct btrfs_inode *inode, + struct page *locked_page, u64 start, + u64 end, int *page_started, + unsigned long *nr_written) +{ + int ret; + + ret = cow_file_range(inode, locked_page, start, end, page_started, + nr_written, 0); + if (ret) + return ret; + + if (*page_started) + return 0; + + __set_page_dirty_nobuffers(locked_page); + account_page_redirty(locked_page); + extent_write_locked_range(&inode->vfs_inode, start, end, WB_SYNC_ALL); + *page_started = 1; + + return 0; +} + static noinline int csum_exist_in_range(struct btrfs_fs_info *fs_info, u64 bytenr, u64 num_bytes) { @@ -1519,8 +1537,7 @@ static noinline int run_delalloc_nocow(struct btrfs_inode *inode, EXTENT_LOCKED | EXTENT_DELALLOC | EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG, PAGE_UNLOCK | - PAGE_CLEAR_DIRTY | - PAGE_SET_WRITEBACK | + PAGE_START_WRITEBACK | PAGE_END_WRITEBACK); return -ENOMEM; } @@ -1657,9 +1674,6 @@ next_slot: */ btrfs_release_path(path); - /* If extent is RO, we must COW it */ - if (btrfs_extent_readonly(fs_info, disk_bytenr)) - goto out_check; ret = btrfs_cross_ref_exist(root, ino, found_key.offset - extent_offset, disk_bytenr, false); @@ -1706,6 +1720,7 @@ next_slot: WARN_ON_ONCE(freespace_inode); goto out_check; } + /* If the extent's block group is RO, we must COW */ if (!btrfs_inc_nocow_writers(fs_info, disk_bytenr)) goto out_check; nocow = true; @@ -1842,8 +1857,7 @@ error: locked_page, EXTENT_LOCKED | EXTENT_DELALLOC | EXTENT_DEFRAG | EXTENT_DO_ACCOUNTING, PAGE_UNLOCK | - PAGE_CLEAR_DIRTY | - PAGE_SET_WRITEBACK | + PAGE_START_WRITEBACK | PAGE_END_WRITEBACK); btrfs_free_path(path); return ret; @@ -1878,17 +1892,24 @@ int btrfs_run_delalloc_range(struct btrfs_inode *inode, struct page *locked_page { int ret; int force_cow = need_force_cow(inode, start, end); + const bool zoned = btrfs_is_zoned(inode->root->fs_info); if (inode->flags & BTRFS_INODE_NODATACOW && !force_cow) { + ASSERT(!zoned); ret = run_delalloc_nocow(inode, locked_page, start, end, page_started, 1, nr_written); } else if (inode->flags & BTRFS_INODE_PREALLOC && !force_cow) { + ASSERT(!zoned); ret = run_delalloc_nocow(inode, locked_page, start, end, page_started, 0, nr_written); } else if (!inode_can_compress(inode) || !inode_need_compress(inode, start, end)) { - ret = cow_file_range(inode, locked_page, start, end, - page_started, nr_written, 1); + if (zoned) + ret = run_delalloc_zoned(inode, locked_page, start, end, + page_started, nr_written); + else + ret = cow_file_range(inode, locked_page, start, end, + page_started, nr_written, 1); } else { set_bit(BTRFS_INODE_HAS_ASYNC_EXTENT, &inode->runtime_flags); ret = cow_file_range_async(inode, wbc, locked_page, start, end, @@ -2183,9 +2204,10 @@ int btrfs_bio_fits_in_stripe(struct page *page, size_t size, struct bio *bio, struct inode *inode = page->mapping->host; struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); u64 logical = bio->bi_iter.bi_sector << 9; + struct extent_map *em; u64 length = 0; u64 map_length; - int ret; + int ret = 0; struct btrfs_io_geometry geom; if (bio_flags & EXTENT_BIO_COMPRESSED) @@ -2193,14 +2215,19 @@ int btrfs_bio_fits_in_stripe(struct page *page, size_t size, struct bio *bio, length = bio->bi_iter.bi_size; map_length = length; - ret = btrfs_get_io_geometry(fs_info, btrfs_op(bio), logical, map_length, - &geom); + em = btrfs_get_chunk_map(fs_info, logical, map_length); + if (IS_ERR(em)) + return PTR_ERR(em); + ret = btrfs_get_io_geometry(fs_info, em, btrfs_op(bio), logical, + map_length, &geom); if (ret < 0) - return ret; + goto out; if (geom.len < length + size) - return 1; - return 0; + ret = 1; +out: + free_extent_map(em); + return ret; } /* @@ -2217,6 +2244,119 @@ static blk_status_t btrfs_submit_bio_start(struct inode *inode, struct bio *bio, return btrfs_csum_one_bio(BTRFS_I(inode), bio, 0, 0); } +bool btrfs_bio_fits_in_ordered_extent(struct page *page, struct bio *bio, + unsigned int size) +{ + struct btrfs_inode *inode = BTRFS_I(page->mapping->host); + struct btrfs_fs_info *fs_info = inode->root->fs_info; + struct btrfs_ordered_extent *ordered; + u64 len = bio->bi_iter.bi_size + size; + bool ret = true; + + ASSERT(btrfs_is_zoned(fs_info)); + ASSERT(fs_info->max_zone_append_size > 0); + ASSERT(bio_op(bio) == REQ_OP_ZONE_APPEND); + + /* Ordered extent not yet created, so we're good */ + ordered = btrfs_lookup_ordered_extent(inode, page_offset(page)); + if (!ordered) + return ret; + + if ((bio->bi_iter.bi_sector << SECTOR_SHIFT) + len > + ordered->disk_bytenr + ordered->disk_num_bytes) + ret = false; + + btrfs_put_ordered_extent(ordered); + + return ret; +} + +static blk_status_t extract_ordered_extent(struct btrfs_inode *inode, + struct bio *bio, loff_t file_offset) +{ + struct btrfs_ordered_extent *ordered; + struct extent_map *em = NULL, *em_new = NULL; + struct extent_map_tree *em_tree = &inode->extent_tree; + u64 start = (u64)bio->bi_iter.bi_sector << SECTOR_SHIFT; + u64 len = bio->bi_iter.bi_size; + u64 end = start + len; + u64 ordered_end; + u64 pre, post; + int ret = 0; + + ordered = btrfs_lookup_ordered_extent(inode, file_offset); + if (WARN_ON_ONCE(!ordered)) + return BLK_STS_IOERR; + + /* No need to split */ + if (ordered->disk_num_bytes == len) + goto out; + + /* We cannot split once end_bio'd ordered extent */ + if (WARN_ON_ONCE(ordered->bytes_left != ordered->disk_num_bytes)) { + ret = -EINVAL; + goto out; + } + + /* We cannot split a compressed ordered extent */ + if (WARN_ON_ONCE(ordered->disk_num_bytes != ordered->num_bytes)) { + ret = -EINVAL; + goto out; + } + + ordered_end = ordered->disk_bytenr + ordered->disk_num_bytes; + /* bio must be in one ordered extent */ + if (WARN_ON_ONCE(start < ordered->disk_bytenr || end > ordered_end)) { + ret = -EINVAL; + goto out; + } + + /* Checksum list should be empty */ + if (WARN_ON_ONCE(!list_empty(&ordered->list))) { + ret = -EINVAL; + goto out; + } + + pre = start - ordered->disk_bytenr; + post = ordered_end - end; + + ret = btrfs_split_ordered_extent(ordered, pre, post); + if (ret) + goto out; + + read_lock(&em_tree->lock); + em = lookup_extent_mapping(em_tree, ordered->file_offset, len); + if (!em) { + read_unlock(&em_tree->lock); + ret = -EIO; + goto out; + } + read_unlock(&em_tree->lock); + + ASSERT(!test_bit(EXTENT_FLAG_COMPRESSED, &em->flags)); + /* + * We cannot reuse em_new here but have to create a new one, as + * unpin_extent_cache() expects the start of the extent map to be the + * logical offset of the file, which does not hold true anymore after + * splitting. + */ + em_new = create_io_em(inode, em->start + pre, len, + em->start + pre, em->block_start + pre, len, + len, len, BTRFS_COMPRESS_NONE, + BTRFS_ORDERED_REGULAR); + if (IS_ERR(em_new)) { + ret = PTR_ERR(em_new); + goto out; + } + free_extent_map(em_new); + +out: + free_extent_map(em); + btrfs_put_ordered_extent(ordered); + + return errno_to_blk_status(ret); +} + /* * extent_io.c submission hook. This does the right thing for csum calculation * on write, or reading the csums from the tree before a read. @@ -2252,7 +2392,16 @@ blk_status_t btrfs_submit_data_bio(struct inode *inode, struct bio *bio, if (btrfs_is_free_space_inode(BTRFS_I(inode))) metadata = BTRFS_WQ_ENDIO_FREE_SPACE; - if (bio_op(bio) != REQ_OP_WRITE) { + if (bio_op(bio) == REQ_OP_ZONE_APPEND) { + struct page *page = bio_first_bvec_all(bio)->bv_page; + loff_t file_offset = page_offset(page); + + ret = extract_ordered_extent(BTRFS_I(inode), bio, file_offset); + if (ret) + goto out; + } + + if (btrfs_op(bio) != BTRFS_MAP_WRITE) { ret = btrfs_bio_wq_end_io(fs_info, bio, metadata); if (ret) goto out; @@ -2754,6 +2903,9 @@ static int btrfs_finish_ordered_io(struct btrfs_ordered_extent *ordered_extent) goto out; } + if (ordered_extent->disk) + btrfs_rewrite_logical_zoned(ordered_extent); + btrfs_free_io_failure_record(inode, start, end); if (test_bit(BTRFS_ORDERED_TRUNCATED, &ordered_extent->flags)) { @@ -3103,14 +3255,16 @@ void btrfs_run_delayed_iputs(struct btrfs_fs_info *fs_info) } /** - * btrfs_wait_on_delayed_iputs - wait on the delayed iputs to be done running - * @fs_info - the fs_info for this fs - * @return - EINTR if we were killed, 0 if nothing's pending + * Wait for flushing all delayed iputs + * + * @fs_info: the filesystem * * This will wait on any delayed iputs that are currently running with KILLABLE * set. Once they are all done running we will return, unless we are killed in * which case we return EINTR. This helps in user operations like fallocate etc * that might get blocked on the iputs. + * + * Return EINTR if we were killed, 0 if nothing's pending */ int btrfs_wait_on_delayed_iputs(struct btrfs_fs_info *fs_info) { @@ -4720,6 +4874,9 @@ again: ret = -ENOMEM; goto out; } + ret = set_page_extent_mapped(page); + if (ret < 0) + goto out_unlock; if (!PageUptodate(page)) { ret = btrfs_readpage(NULL, page); @@ -4737,7 +4894,6 @@ again: wait_on_page_writeback(page); lock_extent_bits(io_tree, block_start, block_end, &cached_state); - set_page_extent_mapped(page); ordered = btrfs_lookup_ordered_extent(inode, block_start); if (ordered) { @@ -5011,6 +5167,15 @@ static int btrfs_setsize(struct inode *inode, struct iattr *attr) btrfs_drew_write_unlock(&root->snapshot_lock); btrfs_end_transaction(trans); } else { + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); + + if (btrfs_is_zoned(fs_info)) { + ret = btrfs_wait_ordered_range(inode, + ALIGN(newsize, fs_info->sectorsize), + (u64)-1); + if (ret) + return ret; + } /* * We're truncating a file that used to have good data down to @@ -5045,7 +5210,8 @@ static int btrfs_setsize(struct inode *inode, struct iattr *attr) return ret; } -static int btrfs_setattr(struct dentry *dentry, struct iattr *attr) +static int btrfs_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *attr) { struct inode *inode = d_inode(dentry); struct btrfs_root *root = BTRFS_I(inode)->root; @@ -5054,7 +5220,7 @@ static int btrfs_setattr(struct dentry *dentry, struct iattr *attr) if (btrfs_root_readonly(root)) return -EROFS; - err = setattr_prepare(dentry, attr); + err = setattr_prepare(&init_user_ns, dentry, attr); if (err) return err; @@ -5065,12 +5231,13 @@ static int btrfs_setattr(struct dentry *dentry, struct iattr *attr) } if (attr->ia_valid) { - setattr_copy(inode, attr); + setattr_copy(&init_user_ns, inode, attr); inode_inc_iversion(inode); err = btrfs_dirty_inode(inode); if (!err && attr->ia_valid & ATTR_MODE) - err = posix_acl_chmod(inode, inode->i_mode); + err = posix_acl_chmod(&init_user_ns, inode, + inode->i_mode); } return err; @@ -5916,7 +6083,7 @@ static int btrfs_dirty_inode(struct inode *inode) return PTR_ERR(trans); ret = btrfs_update_inode(trans, root, BTRFS_I(inode)); - if (ret && ret == -ENOSPC) { + if (ret && (ret == -ENOSPC || ret == -EDQUOT)) { /* whoops, lets try again with the full transaction */ btrfs_end_transaction(trans); trans = btrfs_start_transaction(root, 1); @@ -6190,7 +6357,7 @@ static struct inode *btrfs_new_inode(struct btrfs_trans_handle *trans, if (ret != 0) goto fail_unlock; - inode_init_owner(inode, dir, mode); + inode_init_owner(&init_user_ns, inode, dir, mode); inode_set_bytes(inode, 0); inode->i_mtime = current_time(inode); @@ -6351,8 +6518,8 @@ static int btrfs_add_nondir(struct btrfs_trans_handle *trans, return err; } -static int btrfs_mknod(struct inode *dir, struct dentry *dentry, - umode_t mode, dev_t rdev) +static int btrfs_mknod(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, dev_t rdev) { struct btrfs_fs_info *fs_info = btrfs_sb(dir->i_sb); struct btrfs_trans_handle *trans; @@ -6371,7 +6538,7 @@ static int btrfs_mknod(struct inode *dir, struct dentry *dentry, if (IS_ERR(trans)) return PTR_ERR(trans); - err = btrfs_find_free_objectid(root, &objectid); + err = btrfs_get_free_objectid(root, &objectid); if (err) goto out_unlock; @@ -6415,8 +6582,8 @@ out_unlock: return err; } -static int btrfs_create(struct inode *dir, struct dentry *dentry, - umode_t mode, bool excl) +static int btrfs_create(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, bool excl) { struct btrfs_fs_info *fs_info = btrfs_sb(dir->i_sb); struct btrfs_trans_handle *trans; @@ -6435,7 +6602,7 @@ static int btrfs_create(struct inode *dir, struct dentry *dentry, if (IS_ERR(trans)) return PTR_ERR(trans); - err = btrfs_find_free_objectid(root, &objectid); + err = btrfs_get_free_objectid(root, &objectid); if (err) goto out_unlock; @@ -6560,7 +6727,8 @@ fail: return err; } -static int btrfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) +static int btrfs_mkdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { struct btrfs_fs_info *fs_info = btrfs_sb(dir->i_sb); struct inode *inode = NULL; @@ -6579,7 +6747,7 @@ static int btrfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) if (IS_ERR(trans)) return PTR_ERR(trans); - err = btrfs_find_free_objectid(root, &objectid); + err = btrfs_get_free_objectid(root, &objectid); if (err) goto out_fail; @@ -7103,9 +7271,6 @@ static struct extent_map *btrfs_new_extent_direct(struct btrfs_inode *inode, * @strict: if true, omit optimizations that might force us into unnecessary * cow. e.g., don't trust generation number. * - * This function will flush ordered extents in the range to ensure proper - * nocow checks for (nowait == false) case. - * * Return: * >0 and update @len if we can do nocow write * 0 if we can't do nocow write @@ -7613,6 +7778,9 @@ static int btrfs_dio_iomap_begin(struct inode *inode, loff_t start, iomap->bdev = fs_info->fs_devices->latest_bdev; iomap->length = len; + if (write && btrfs_use_zone_append(BTRFS_I(inode), em)) + iomap->flags |= IOMAP_F_ZONE_APPEND; + free_extent_map(em); return 0; @@ -7682,7 +7850,7 @@ static void btrfs_dio_private_put(struct btrfs_dio_private *dip) if (!refcount_dec_and_test(&dip->refs)) return; - if (bio_op(dip->dio_bio) == REQ_OP_WRITE) { + if (btrfs_op(dip->dio_bio) == BTRFS_MAP_WRITE) { __endio_write_update_ordered(BTRFS_I(dip->inode), dip->logical_offset, dip->bytes, @@ -7797,10 +7965,8 @@ static void __endio_write_update_ordered(struct btrfs_inode *inode, NULL); btrfs_queue_work(wq, &ordered->work); } - /* - * If btrfs_dec_test_ordered_pending does not find any ordered - * extent in the range, we can exit. - */ + + /* No ordered extent found in the range, exit */ if (ordered_offset == last_offset) return; /* @@ -7841,6 +8007,8 @@ static void btrfs_end_dio_bio(struct bio *bio) if (err) dip->dio_bio->bi_status = err; + btrfs_record_physical_zoned(dip->inode, dip->logical_offset, bio); + bio_put(bio); btrfs_dio_private_put(dip); } @@ -7850,7 +8018,7 @@ static inline blk_status_t btrfs_submit_dio_bio(struct bio *bio, { struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); struct btrfs_dio_private *dip = bio->bi_private; - bool write = bio_op(bio) == REQ_OP_WRITE; + bool write = btrfs_op(bio) == BTRFS_MAP_WRITE; blk_status_t ret; /* Check btrfs_submit_bio_hook() for rules about async submit. */ @@ -7900,7 +8068,7 @@ static struct btrfs_dio_private *btrfs_create_dio_private(struct bio *dio_bio, struct inode *inode, loff_t file_offset) { - const bool write = (bio_op(dio_bio) == REQ_OP_WRITE); + const bool write = (btrfs_op(dio_bio) == BTRFS_MAP_WRITE); const bool csum = !(BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM); size_t dip_size; struct btrfs_dio_private *dip; @@ -7930,7 +8098,7 @@ static struct btrfs_dio_private *btrfs_create_dio_private(struct bio *dio_bio, static blk_qc_t btrfs_submit_direct(struct inode *inode, struct iomap *iomap, struct bio *dio_bio, loff_t file_offset) { - const bool write = (bio_op(dio_bio) == REQ_OP_WRITE); + const bool write = (btrfs_op(dio_bio) == BTRFS_MAP_WRITE); struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); const bool raid56 = (btrfs_data_alloc_profile(fs_info) & BTRFS_BLOCK_GROUP_RAID56_MASK); @@ -7941,10 +8109,12 @@ static blk_qc_t btrfs_submit_direct(struct inode *inode, struct iomap *iomap, u64 submit_len; int clone_offset = 0; int clone_len; + u64 logical; int ret; blk_status_t status; struct btrfs_io_geometry geom; struct btrfs_dio_data *dio_data = iomap->private; + struct extent_map *em = NULL; dip = btrfs_create_dio_private(dio_bio, inode, file_offset); if (!dip) { @@ -7973,12 +8143,18 @@ static blk_qc_t btrfs_submit_direct(struct inode *inode, struct iomap *iomap, submit_len = dio_bio->bi_iter.bi_size; do { - ret = btrfs_get_io_geometry(fs_info, btrfs_op(dio_bio), - start_sector << 9, submit_len, - &geom); + logical = start_sector << 9; + em = btrfs_get_chunk_map(fs_info, logical, submit_len); + if (IS_ERR(em)) { + status = errno_to_blk_status(PTR_ERR(em)); + em = NULL; + goto out_err_em; + } + ret = btrfs_get_io_geometry(fs_info, em, btrfs_op(dio_bio), + logical, submit_len, &geom); if (ret) { status = errno_to_blk_status(ret); - goto out_err; + goto out_err_em; } ASSERT(geom.len <= INT_MAX); @@ -7993,6 +8169,19 @@ static blk_qc_t btrfs_submit_direct(struct inode *inode, struct iomap *iomap, bio->bi_end_io = btrfs_end_dio_bio; btrfs_io_bio(bio)->logical = file_offset; + WARN_ON_ONCE(write && btrfs_is_zoned(fs_info) && + fs_info->max_zone_append_size && + bio_op(bio) != REQ_OP_ZONE_APPEND); + + if (bio_op(bio) == REQ_OP_ZONE_APPEND) { + status = extract_ordered_extent(BTRFS_I(inode), bio, + file_offset); + if (status) { + bio_put(bio); + goto out_err; + } + } + ASSERT(submit_len >= clone_len); submit_len -= clone_len; @@ -8023,19 +8212,24 @@ static blk_qc_t btrfs_submit_direct(struct inode *inode, struct iomap *iomap, bio_put(bio); if (submit_len > 0) refcount_dec(&dip->refs); - goto out_err; + goto out_err_em; } dio_data->submitted += clone_len; clone_offset += clone_len; start_sector += clone_len >> 9; file_offset += clone_len; + + free_extent_map(em); } while (submit_len > 0); return BLK_QC_T_NONE; +out_err_em: + free_extent_map(em); out_err: dip->dio_bio->bi_status = status; btrfs_dio_private_put(dip); + return BLK_QC_T_NONE; } @@ -8117,7 +8311,7 @@ static int __btrfs_releasepage(struct page *page, gfp_t gfp_flags) { int ret = try_release_extent_mapping(page, gfp_flags); if (ret == 1) - detach_page_private(page); + clear_page_extent_mapped(page); return ret; } @@ -8186,8 +8380,9 @@ static void btrfs_invalidatepage(struct page *page, unsigned int offset, if (!inode_evicting) lock_extent_bits(tree, page_start, page_end, &cached_state); -again: + start = page_start; +again: ordered = btrfs_lookup_ordered_range(inode, start, page_end - start + 1); if (ordered) { found_ordered = true; @@ -8276,7 +8471,7 @@ again: } ClearPageChecked(page); - detach_page_private(page); + clear_page_extent_mapped(page); } /* @@ -8355,7 +8550,12 @@ again: wait_on_page_writeback(page); lock_extent_bits(io_tree, page_start, page_end, &cached_state); - set_page_extent_mapped(page); + ret2 = set_page_extent_mapped(page); + if (ret2 < 0) { + ret = vmf_error(ret2); + unlock_extent_cached(io_tree, page_start, page_end, &cached_state); + goto out_unlock; + } /* * we can't set the delalloc bits if there are pending ordered @@ -8592,15 +8792,18 @@ out: */ int btrfs_create_subvol_root(struct btrfs_trans_handle *trans, struct btrfs_root *new_root, - struct btrfs_root *parent_root, - u64 new_dirid) + struct btrfs_root *parent_root) { struct inode *inode; int err; u64 index = 0; + u64 ino; + + err = btrfs_get_free_objectid(new_root, &ino); + if (err < 0) + return err; - inode = btrfs_new_inode(trans, new_root, NULL, "..", 2, - new_dirid, new_dirid, + inode = btrfs_new_inode(trans, new_root, NULL, "..", 2, ino, ino, S_IFDIR | (~current_umask() & S_IRWXUGO), &index); if (IS_ERR(inode)) @@ -8805,7 +9008,7 @@ int __init btrfs_init_cachep(void) btrfs_free_space_bitmap_cachep = kmem_cache_create("btrfs_free_space_bitmap", PAGE_SIZE, PAGE_SIZE, - SLAB_RED_ZONE, NULL); + SLAB_MEM_SPREAD, NULL); if (!btrfs_free_space_bitmap_cachep) goto fail; @@ -8815,7 +9018,8 @@ fail: return -ENOMEM; } -static int btrfs_getattr(const struct path *path, struct kstat *stat, +static int btrfs_getattr(struct user_namespace *mnt_userns, + const struct path *path, struct kstat *stat, u32 request_mask, unsigned int flags) { u64 delalloc_bytes; @@ -8841,7 +9045,7 @@ static int btrfs_getattr(const struct path *path, struct kstat *stat, STATX_ATTR_IMMUTABLE | STATX_ATTR_NODUMP); - generic_fillattr(inode, stat); + generic_fillattr(&init_user_ns, inode, stat); stat->dev = BTRFS_I(inode)->root->anon_dev; spin_lock(&BTRFS_I(inode)->lock); @@ -9079,7 +9283,7 @@ static int btrfs_whiteout_for_rename(struct btrfs_trans_handle *trans, u64 objectid; u64 index; - ret = btrfs_find_free_objectid(root, &objectid); + ret = btrfs_get_free_objectid(root, &objectid); if (ret) return ret; @@ -9332,9 +9536,9 @@ out_notrans: return ret; } -static int btrfs_rename2(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry, - unsigned int flags) +static int btrfs_rename2(struct user_namespace *mnt_userns, struct inode *old_dir, + struct dentry *old_dentry, struct inode *new_dir, + struct dentry *new_dentry, unsigned int flags) { if (flags & ~(RENAME_NOREPLACE | RENAME_EXCHANGE | RENAME_WHITEOUT)) return -EINVAL; @@ -9486,11 +9690,11 @@ int btrfs_start_delalloc_snapshot(struct btrfs_root *root) return start_delalloc_inodes(root, &wbc, true, false); } -int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, u64 nr, +int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, long nr, bool in_reclaim_context) { struct writeback_control wbc = { - .nr_to_write = (nr == U64_MAX) ? LONG_MAX : (unsigned long)nr, + .nr_to_write = nr, .sync_mode = WB_SYNC_NONE, .range_start = 0, .range_end = LLONG_MAX, @@ -9507,12 +9711,12 @@ int btrfs_start_delalloc_roots(struct btrfs_fs_info *fs_info, u64 nr, mutex_lock(&fs_info->delalloc_root_mutex); spin_lock(&fs_info->delalloc_root_lock); list_splice_init(&fs_info->delalloc_roots, &splice); - while (!list_empty(&splice) && nr) { + while (!list_empty(&splice)) { /* * Reset nr_to_write here so we know that we're doing a full * flush. */ - if (nr == U64_MAX) + if (nr == LONG_MAX) wbc.nr_to_write = LONG_MAX; root = list_first_entry(&splice, struct btrfs_root, @@ -9542,8 +9746,8 @@ out: return ret; } -static int btrfs_symlink(struct inode *dir, struct dentry *dentry, - const char *symname) +static int btrfs_symlink(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, const char *symname) { struct btrfs_fs_info *fs_info = btrfs_sb(dir->i_sb); struct btrfs_trans_handle *trans; @@ -9575,7 +9779,7 @@ static int btrfs_symlink(struct inode *dir, struct dentry *dentry, if (IS_ERR(trans)) return PTR_ERR(trans); - err = btrfs_find_free_objectid(root, &objectid); + err = btrfs_get_free_objectid(root, &objectid); if (err) goto out_unlock; @@ -9673,6 +9877,7 @@ static struct btrfs_trans_handle *insert_prealloc_file_extent( struct btrfs_path *path; u64 start = ins->objectid; u64 len = ins->offset; + int qgroup_released; int ret; memset(&stack_fi, 0, sizeof(stack_fi)); @@ -9685,16 +9890,16 @@ static struct btrfs_trans_handle *insert_prealloc_file_extent( btrfs_set_stack_file_extent_compression(&stack_fi, BTRFS_COMPRESS_NONE); /* Encryption and other encoding is reserved and all 0 */ - ret = btrfs_qgroup_release_data(inode, file_offset, len); - if (ret < 0) - return ERR_PTR(ret); + qgroup_released = btrfs_qgroup_release_data(inode, file_offset, len); + if (qgroup_released < 0) + return ERR_PTR(qgroup_released); if (trans) { ret = insert_reserved_file_extent(trans, inode, file_offset, &stack_fi, - true, ret); + true, qgroup_released); if (ret) - return ERR_PTR(ret); + goto free_qgroup; return trans; } @@ -9705,21 +9910,35 @@ static struct btrfs_trans_handle *insert_prealloc_file_extent( extent_info.file_offset = file_offset; extent_info.extent_buf = (char *)&stack_fi; extent_info.is_new_extent = true; - extent_info.qgroup_reserved = ret; + extent_info.qgroup_reserved = qgroup_released; extent_info.insertions = 0; path = btrfs_alloc_path(); - if (!path) - return ERR_PTR(-ENOMEM); + if (!path) { + ret = -ENOMEM; + goto free_qgroup; + } ret = btrfs_replace_file_extents(&inode->vfs_inode, path, file_offset, file_offset + len - 1, &extent_info, &trans); btrfs_free_path(path); if (ret) - return ERR_PTR(ret); - + goto free_qgroup; return trans; + +free_qgroup: + /* + * We have released qgroup data range at the beginning of the function, + * and normally qgroup_released bytes will be freed when committing + * transaction. + * But if we error out early, we have to free what we have released + * or we leak qgroup data reservation. + */ + btrfs_qgroup_free_refroot(inode->root->fs_info, + inode->root->root_key.objectid, qgroup_released, + BTRFS_QGROUP_RSV_DATA); + return ERR_PTR(ret); } static int __btrfs_prealloc_file_range(struct inode *inode, int mode, @@ -9877,7 +10096,8 @@ static int btrfs_set_page_dirty(struct page *page) return __set_page_dirty_nobuffers(page); } -static int btrfs_permission(struct inode *inode, int mask) +static int btrfs_permission(struct user_namespace *mnt_userns, + struct inode *inode, int mask) { struct btrfs_root *root = BTRFS_I(inode)->root; umode_t mode = inode->i_mode; @@ -9889,10 +10109,11 @@ static int btrfs_permission(struct inode *inode, int mask) if (BTRFS_I(inode)->flags & BTRFS_INODE_READONLY) return -EACCES; } - return generic_permission(inode, mask); + return generic_permission(&init_user_ns, inode, mask); } -static int btrfs_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode) +static int btrfs_tmpfile(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { struct btrfs_fs_info *fs_info = btrfs_sb(dir->i_sb); struct btrfs_trans_handle *trans; @@ -9909,7 +10130,7 @@ static int btrfs_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode) if (IS_ERR(trans)) return PTR_ERR(trans); - ret = btrfs_find_free_objectid(root, &objectid); + ret = btrfs_get_free_objectid(root, &objectid); if (ret) goto out; @@ -9992,6 +10213,7 @@ static int btrfs_add_swapfile_pin(struct inode *inode, void *ptr, sp->ptr = ptr; sp->inode = inode; sp->is_block_group = is_block_group; + sp->bg_extent_count = 1; spin_lock(&fs_info->swapfile_pins_lock); p = &fs_info->swapfile_pins.rb_node; @@ -10005,6 +10227,8 @@ static int btrfs_add_swapfile_pin(struct inode *inode, void *ptr, (sp->ptr == entry->ptr && sp->inode > entry->inode)) { p = &(*p)->rb_right; } else { + if (is_block_group) + entry->bg_extent_count++; spin_unlock(&fs_info->swapfile_pins_lock); kfree(sp); return 1; @@ -10030,8 +10254,11 @@ static void btrfs_free_swapfile_pins(struct inode *inode) sp = rb_entry(node, struct btrfs_swapfile_pin, node); if (sp->inode == inode) { rb_erase(&sp->node, &fs_info->swapfile_pins); - if (sp->is_block_group) + if (sp->is_block_group) { + btrfs_dec_block_group_swap_extents(sp->ptr, + sp->bg_extent_count); btrfs_put_block_group(sp->ptr); + } kfree(sp); } node = next; @@ -10092,7 +10319,8 @@ static int btrfs_swap_activate(struct swap_info_struct *sis, struct file *file, sector_t *span) { struct inode *inode = file_inode(file); - struct btrfs_fs_info *fs_info = BTRFS_I(inode)->root->fs_info; + struct btrfs_root *root = BTRFS_I(inode)->root; + struct btrfs_fs_info *fs_info = root->fs_info; struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree; struct extent_state *cached_state = NULL; struct extent_map *em = NULL; @@ -10143,13 +10371,27 @@ static int btrfs_swap_activate(struct swap_info_struct *sis, struct file *file, "cannot activate swapfile while exclusive operation is running"); return -EBUSY; } + + /* + * Prevent snapshot creation while we are activating the swap file. + * We do not want to race with snapshot creation. If snapshot creation + * already started before we bumped nr_swapfiles from 0 to 1 and + * completes before the first write into the swap file after it is + * activated, than that write would fallback to COW. + */ + if (!btrfs_drew_try_write_lock(&root->snapshot_lock)) { + btrfs_exclop_finish(fs_info); + btrfs_warn(fs_info, + "cannot activate swapfile because snapshot creation is in progress"); + return -EINVAL; + } /* * Snapshots can create extents which require COW even if NODATACOW is * set. We use this counter to prevent snapshots. We must increment it * before walking the extents because we don't want a concurrent * snapshot to run after we've already checked the extents. */ - atomic_inc(&BTRFS_I(inode)->root->nr_swapfiles); + atomic_inc(&root->nr_swapfiles); isize = ALIGN_DOWN(inode->i_size, fs_info->sectorsize); @@ -10246,6 +10488,17 @@ static int btrfs_swap_activate(struct swap_info_struct *sis, struct file *file, goto out; } + if (!btrfs_inc_block_group_swap_extents(bg)) { + btrfs_warn(fs_info, + "block group for swapfile at %llu is read-only%s", + bg->start, + atomic_read(&fs_info->scrubs_running) ? + " (scrub running)" : ""); + btrfs_put_block_group(bg); + ret = -EINVAL; + goto out; + } + ret = btrfs_add_swapfile_pin(inode, bg, true); if (ret) { btrfs_put_block_group(bg); @@ -10284,6 +10537,8 @@ out: if (ret) btrfs_swap_deactivate(file); + btrfs_drew_write_unlock(&root->snapshot_lock); + btrfs_exclop_finish(fs_info); if (ret) diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index dde49a791f3e..e8d53fea4c61 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -213,7 +213,7 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg) const char *comp = NULL; u32 binode_flags; - if (!inode_owner_or_capable(inode)) + if (!inode_owner_or_capable(&init_user_ns, inode)) return -EPERM; if (btrfs_root_readonly(root)) @@ -429,7 +429,7 @@ static int btrfs_ioctl_fssetxattr(struct file *file, void __user *arg) unsigned old_i_flags; int ret = 0; - if (!inode_owner_or_capable(inode)) + if (!inode_owner_or_capable(&init_user_ns, inode)) return -EPERM; if (btrfs_root_readonly(root)) @@ -528,6 +528,14 @@ static noinline int btrfs_ioctl_fitrim(struct btrfs_fs_info *fs_info, return -EPERM; /* + * btrfs_trim_block_group() depends on space cache, which is not + * available in zoned filesystem. So, disallow fitrim on a zoned + * filesystem for now. + */ + if (btrfs_is_zoned(fs_info)) + return -EOPNOTSUPP; + + /* * If the fs is mounted with nologreplay, which requires it to be * mounted in RO mode as well, we can not allow discard on free space * inside block groups, because log trees refer to extents that are not @@ -606,14 +614,13 @@ static noinline int create_subvol(struct inode *dir, int err; dev_t anon_dev = 0; u64 objectid; - u64 new_dirid = BTRFS_FIRST_FREE_OBJECTID; u64 index = 0; root_item = kzalloc(sizeof(*root_item), GFP_KERNEL); if (!root_item) return -ENOMEM; - ret = btrfs_find_free_objectid(fs_info->tree_root, &objectid); + ret = btrfs_get_free_objectid(fs_info->tree_root, &objectid); if (ret) goto fail_free; @@ -693,7 +700,7 @@ static noinline int create_subvol(struct inode *dir, free_extent_buffer(leaf); leaf = NULL; - btrfs_set_root_dirid(root_item, new_dirid); + btrfs_set_root_dirid(root_item, BTRFS_FIRST_FREE_OBJECTID); key.objectid = objectid; key.offset = 0; @@ -716,7 +723,7 @@ static noinline int create_subvol(struct inode *dir, btrfs_record_root_in_trans(trans, new_root); - ret = btrfs_create_subvol_root(trans, new_root, root, new_dirid); + ret = btrfs_create_subvol_root(trans, new_root, root); btrfs_put_root(new_root); if (ret) { /* We potentially lose an unused inode item here */ @@ -724,10 +731,6 @@ static noinline int create_subvol(struct inode *dir, goto fail; } - mutex_lock(&new_root->objectid_mutex); - new_root->highest_objectid = new_dirid; - mutex_unlock(&new_root->objectid_mutex); - /* * insert the directory item */ @@ -922,13 +925,14 @@ static int btrfs_may_delete(struct inode *dir, struct dentry *victim, int isdir) BUG_ON(d_inode(victim->d_parent) != dir); audit_inode_child(dir, victim, AUDIT_TYPE_CHILD_DELETE); - error = inode_permission(dir, MAY_WRITE | MAY_EXEC); + error = inode_permission(&init_user_ns, dir, MAY_WRITE | MAY_EXEC); if (error) return error; if (IS_APPEND(dir)) return -EPERM; - if (check_sticky(dir, d_inode(victim)) || IS_APPEND(d_inode(victim)) || - IS_IMMUTABLE(d_inode(victim)) || IS_SWAPFILE(d_inode(victim))) + if (check_sticky(&init_user_ns, dir, d_inode(victim)) || + IS_APPEND(d_inode(victim)) || IS_IMMUTABLE(d_inode(victim)) || + IS_SWAPFILE(d_inode(victim))) return -EPERM; if (isdir) { if (!d_is_dir(victim)) @@ -951,7 +955,7 @@ static inline int btrfs_may_create(struct inode *dir, struct dentry *child) return -EEXIST; if (IS_DEADDIR(dir)) return -ENOENT; - return inode_permission(dir, MAY_WRITE | MAY_EXEC); + return inode_permission(&init_user_ns, dir, MAY_WRITE | MAY_EXEC); } /* @@ -1319,6 +1323,13 @@ again: if (!page) break; + ret = set_page_extent_mapped(page); + if (ret < 0) { + unlock_page(page); + put_page(page); + break; + } + page_start = page_offset(page); page_end = page_start + PAGE_SIZE - 1; while (1) { @@ -1440,7 +1451,6 @@ again: for (i = 0; i < i_done; i++) { clear_page_dirty_for_io(pages[i]); ClearPageChecked(pages[i]); - set_page_extent_mapped(pages[i]); set_page_dirty(pages[i]); unlock_page(pages[i]); put_page(pages[i]); @@ -1862,7 +1872,7 @@ static noinline int __btrfs_ioctl_snap_create(struct file *file, btrfs_info(BTRFS_I(file_inode(file))->root->fs_info, "Snapshot src from another FS"); ret = -EXDEV; - } else if (!inode_owner_or_capable(src_inode)) { + } else if (!inode_owner_or_capable(&init_user_ns, src_inode)) { /* * Subvolume creation is not restricted, but snapshots * are limited to own subvolumes only @@ -1926,7 +1936,10 @@ static noinline int btrfs_ioctl_snap_create_v2(struct file *file, if (vol_args->flags & BTRFS_SUBVOL_RDONLY) readonly = true; if (vol_args->flags & BTRFS_SUBVOL_QGROUP_INHERIT) { - if (vol_args->size > PAGE_SIZE) { + u64 nums; + + if (vol_args->size < sizeof(*inherit) || + vol_args->size > PAGE_SIZE) { ret = -EINVAL; goto free_args; } @@ -1935,6 +1948,20 @@ static noinline int btrfs_ioctl_snap_create_v2(struct file *file, ret = PTR_ERR(inherit); goto free_args; } + + if (inherit->num_qgroups > PAGE_SIZE || + inherit->num_ref_copies > PAGE_SIZE || + inherit->num_excl_copies > PAGE_SIZE) { + ret = -EINVAL; + goto free_inherit; + } + + nums = inherit->num_qgroups + 2 * inherit->num_ref_copies + + 2 * inherit->num_excl_copies; + if (vol_args->size != struct_size(inherit, qgroups, nums)) { + ret = -EINVAL; + goto free_inherit; + } } ret = __btrfs_ioctl_snap_create(file, vol_args->name, vol_args->fd, @@ -1982,7 +2009,7 @@ static noinline int btrfs_ioctl_subvol_setflags(struct file *file, u64 flags; int ret = 0; - if (!inode_owner_or_capable(inode)) + if (!inode_owner_or_capable(&init_user_ns, inode)) return -EPERM; ret = mnt_want_write_file(file); @@ -2538,7 +2565,8 @@ static int btrfs_search_path_in_tree_user(struct inode *inode, ret = PTR_ERR(temp_inode); goto out_put; } - ret = inode_permission(temp_inode, MAY_READ | MAY_EXEC); + ret = inode_permission(&init_user_ns, temp_inode, + MAY_READ | MAY_EXEC); iput(temp_inode); if (ret) { ret = -EACCES; @@ -3068,7 +3096,8 @@ static noinline int btrfs_ioctl_snap_destroy(struct file *file, if (root == dest) goto out_dput; - err = inode_permission(inode, MAY_WRITE | MAY_EXEC); + err = inode_permission(&init_user_ns, inode, + MAY_WRITE | MAY_EXEC); if (err) goto out_dput; } @@ -3139,7 +3168,7 @@ static int btrfs_ioctl_defrag(struct file *file, void __user *argp) * running and allows defrag on files open in read-only mode. */ if (!capable(CAP_SYS_ADMIN) && - inode_permission(inode, MAY_WRITE)) { + inode_permission(&init_user_ns, inode, MAY_WRITE)) { ret = -EPERM; goto out; } @@ -4451,7 +4480,7 @@ static long _btrfs_ioctl_set_received_subvol(struct file *file, int ret = 0; int received_uuid_changed; - if (!inode_owner_or_capable(inode)) + if (!inode_owner_or_capable(&init_user_ns, inode)) return -EPERM; ret = mnt_want_write_file(file); @@ -4951,7 +4980,7 @@ long btrfs_ioctl(struct file *file, unsigned int case BTRFS_IOC_SYNC: { int ret; - ret = btrfs_start_delalloc_roots(fs_info, U64_MAX, false); + ret = btrfs_start_delalloc_roots(fs_info, LONG_MAX, false); if (ret) return ret; ret = btrfs_sync_fs(inode->i_sb, 1); diff --git a/fs/btrfs/lzo.c b/fs/btrfs/lzo.c index aa9cd11f4b78..9084a950dc09 100644 --- a/fs/btrfs/lzo.c +++ b/fs/btrfs/lzo.c @@ -467,7 +467,7 @@ int lzo_decompress(struct list_head *ws, unsigned char *data_in, destlen = min_t(unsigned long, destlen, PAGE_SIZE); bytes = min_t(unsigned long, destlen, out_len - start_byte); - kaddr = kmap_atomic(dest_page); + kaddr = kmap_local_page(dest_page); memcpy(kaddr, workspace->buf + start_byte, bytes); /* @@ -477,7 +477,7 @@ int lzo_decompress(struct list_head *ws, unsigned char *data_in, */ if (bytes < destlen) memset(kaddr+bytes, 0, destlen-bytes); - kunmap_atomic(kaddr); + kunmap_local(kaddr); out: return ret; } diff --git a/fs/btrfs/ordered-data.c b/fs/btrfs/ordered-data.c index 79d366a36223..985a21558437 100644 --- a/fs/btrfs/ordered-data.c +++ b/fs/btrfs/ordered-data.c @@ -199,14 +199,21 @@ static int __btrfs_add_ordered_extent(struct btrfs_inode *inode, u64 file_offset entry->compress_type = compress_type; entry->truncated_len = (u64)-1; entry->qgroup_rsv = ret; - if (type != BTRFS_ORDERED_IO_DONE && type != BTRFS_ORDERED_COMPLETE) - set_bit(type, &entry->flags); + entry->physical = (u64)-1; + entry->disk = NULL; + entry->partno = (u8)-1; - if (dio) { - percpu_counter_add_batch(&fs_info->dio_bytes, num_bytes, - fs_info->delalloc_batch); + ASSERT(type == BTRFS_ORDERED_REGULAR || + type == BTRFS_ORDERED_NOCOW || + type == BTRFS_ORDERED_PREALLOC || + type == BTRFS_ORDERED_COMPRESSED); + set_bit(type, &entry->flags); + + percpu_counter_add_batch(&fs_info->ordered_bytes, num_bytes, + fs_info->delalloc_batch); + + if (dio) set_bit(BTRFS_ORDERED_DIRECT, &entry->flags); - } /* one ref for the tree */ refcount_set(&entry->refs, 1); @@ -256,6 +263,9 @@ int btrfs_add_ordered_extent(struct btrfs_inode *inode, u64 file_offset, u64 disk_bytenr, u64 num_bytes, u64 disk_num_bytes, int type) { + ASSERT(type == BTRFS_ORDERED_REGULAR || + type == BTRFS_ORDERED_NOCOW || + type == BTRFS_ORDERED_PREALLOC); return __btrfs_add_ordered_extent(inode, file_offset, disk_bytenr, num_bytes, disk_num_bytes, type, 0, BTRFS_COMPRESS_NONE); @@ -265,6 +275,9 @@ int btrfs_add_ordered_extent_dio(struct btrfs_inode *inode, u64 file_offset, u64 disk_bytenr, u64 num_bytes, u64 disk_num_bytes, int type) { + ASSERT(type == BTRFS_ORDERED_REGULAR || + type == BTRFS_ORDERED_NOCOW || + type == BTRFS_ORDERED_PREALLOC); return __btrfs_add_ordered_extent(inode, file_offset, disk_bytenr, num_bytes, disk_num_bytes, type, 1, BTRFS_COMPRESS_NONE); @@ -272,11 +285,12 @@ int btrfs_add_ordered_extent_dio(struct btrfs_inode *inode, u64 file_offset, int btrfs_add_ordered_extent_compress(struct btrfs_inode *inode, u64 file_offset, u64 disk_bytenr, u64 num_bytes, - u64 disk_num_bytes, int type, - int compress_type) + u64 disk_num_bytes, int compress_type) { + ASSERT(compress_type != BTRFS_COMPRESS_NONE); return __btrfs_add_ordered_extent(inode, file_offset, disk_bytenr, - num_bytes, disk_num_bytes, type, 0, + num_bytes, disk_num_bytes, + BTRFS_ORDERED_COMPRESSED, 0, compress_type); } @@ -297,26 +311,33 @@ void btrfs_add_ordered_sum(struct btrfs_ordered_extent *entry, } /* - * this is used to account for finished IO across a given range - * of the file. The IO may span ordered extents. If - * a given ordered_extent is completely done, 1 is returned, otherwise - * 0. + * Finish IO for one ordered extent across a given range. The range can + * contain several ordered extents. + * + * @found_ret: Return the finished ordered extent + * @file_offset: File offset for the finished IO + * Will also be updated to one byte past the range that is + * recordered as finished. This allows caller to walk forward. + * @io_size: Length of the finish IO range + * @uptodate: If the IO finished without problem * - * test_and_set_bit on a flag in the struct btrfs_ordered_extent is used - * to make sure this function only returns 1 once for a given ordered extent. + * Return true if any ordered extent is finished in the range, and update + * @found_ret and @file_offset. + * Return false otherwise. * - * file_offset is updated to one byte past the range that is recorded as - * complete. This allows you to walk forward in the file. + * NOTE: Although The range can cross multiple ordered extents, only one + * ordered extent will be updated during one call. The caller is responsible to + * iterate all ordered extents in the range. */ -int btrfs_dec_test_first_ordered_pending(struct btrfs_inode *inode, - struct btrfs_ordered_extent **cached, +bool btrfs_dec_test_first_ordered_pending(struct btrfs_inode *inode, + struct btrfs_ordered_extent **finished_ret, u64 *file_offset, u64 io_size, int uptodate) { struct btrfs_fs_info *fs_info = inode->root->fs_info; struct btrfs_ordered_inode_tree *tree = &inode->ordered_tree; struct rb_node *node; struct btrfs_ordered_extent *entry = NULL; - int ret; + bool finished = false; unsigned long flags; u64 dec_end; u64 dec_start; @@ -324,16 +345,12 @@ int btrfs_dec_test_first_ordered_pending(struct btrfs_inode *inode, spin_lock_irqsave(&tree->lock, flags); node = tree_search(tree, *file_offset); - if (!node) { - ret = 1; + if (!node) goto out; - } entry = rb_entry(node, struct btrfs_ordered_extent, rb_node); - if (!offset_in_entry(entry, *file_offset)) { - ret = 1; + if (!offset_in_entry(entry, *file_offset)) goto out; - } dec_start = max(*file_offset, entry->file_offset); dec_end = min(*file_offset + io_size, @@ -354,39 +371,50 @@ int btrfs_dec_test_first_ordered_pending(struct btrfs_inode *inode, set_bit(BTRFS_ORDERED_IOERR, &entry->flags); if (entry->bytes_left == 0) { - ret = test_and_set_bit(BTRFS_ORDERED_IO_DONE, &entry->flags); + /* + * Ensure only one caller can set the flag and finished_ret + * accordingly + */ + finished = !test_and_set_bit(BTRFS_ORDERED_IO_DONE, &entry->flags); /* test_and_set_bit implies a barrier */ cond_wake_up_nomb(&entry->wait); - } else { - ret = 1; } out: - if (!ret && cached && entry) { - *cached = entry; + if (finished && finished_ret && entry) { + *finished_ret = entry; refcount_inc(&entry->refs); } spin_unlock_irqrestore(&tree->lock, flags); - return ret == 0; + return finished; } /* - * this is used to account for finished IO across a given range - * of the file. The IO should not span ordered extents. If - * a given ordered_extent is completely done, 1 is returned, otherwise - * 0. + * Finish IO for one ordered extent across a given range. The range can only + * contain one ordered extent. + * + * @cached: The cached ordered extent. If not NULL, we can skip the tree + * search and use the ordered extent directly. + * Will be also used to store the finished ordered extent. + * @file_offset: File offset for the finished IO + * @io_size: Length of the finish IO range + * @uptodate: If the IO finishes without problem * - * test_and_set_bit on a flag in the struct btrfs_ordered_extent is used - * to make sure this function only returns 1 once for a given ordered extent. + * Return true if the ordered extent is finished in the range, and update + * @cached. + * Return false otherwise. + * + * NOTE: The range can NOT cross multiple ordered extents. + * Thus caller should ensure the range doesn't cross ordered extents. */ -int btrfs_dec_test_ordered_pending(struct btrfs_inode *inode, - struct btrfs_ordered_extent **cached, - u64 file_offset, u64 io_size, int uptodate) +bool btrfs_dec_test_ordered_pending(struct btrfs_inode *inode, + struct btrfs_ordered_extent **cached, + u64 file_offset, u64 io_size, int uptodate) { struct btrfs_ordered_inode_tree *tree = &inode->ordered_tree; struct rb_node *node; struct btrfs_ordered_extent *entry = NULL; unsigned long flags; - int ret; + bool finished = false; spin_lock_irqsave(&tree->lock, flags); if (cached && *cached) { @@ -395,41 +423,39 @@ int btrfs_dec_test_ordered_pending(struct btrfs_inode *inode, } node = tree_search(tree, file_offset); - if (!node) { - ret = 1; + if (!node) goto out; - } entry = rb_entry(node, struct btrfs_ordered_extent, rb_node); have_entry: - if (!offset_in_entry(entry, file_offset)) { - ret = 1; + if (!offset_in_entry(entry, file_offset)) goto out; - } - if (io_size > entry->bytes_left) { + if (io_size > entry->bytes_left) btrfs_crit(inode->root->fs_info, "bad ordered accounting left %llu size %llu", entry->bytes_left, io_size); - } + entry->bytes_left -= io_size; if (!uptodate) set_bit(BTRFS_ORDERED_IOERR, &entry->flags); if (entry->bytes_left == 0) { - ret = test_and_set_bit(BTRFS_ORDERED_IO_DONE, &entry->flags); + /* + * Ensure only one caller can set the flag and finished_ret + * accordingly + */ + finished = !test_and_set_bit(BTRFS_ORDERED_IO_DONE, &entry->flags); /* test_and_set_bit implies a barrier */ cond_wake_up_nomb(&entry->wait); - } else { - ret = 1; } out: - if (!ret && cached && entry) { + if (finished && cached && entry) { *cached = entry; refcount_inc(&entry->refs); } spin_unlock_irqrestore(&tree->lock, flags); - return ret == 0; + return finished; } /* @@ -480,9 +506,8 @@ void btrfs_remove_ordered_extent(struct btrfs_inode *btrfs_inode, btrfs_delalloc_release_metadata(btrfs_inode, entry->num_bytes, false); - if (test_bit(BTRFS_ORDERED_DIRECT, &entry->flags)) - percpu_counter_add_batch(&fs_info->dio_bytes, -entry->num_bytes, - fs_info->delalloc_batch); + percpu_counter_add_batch(&fs_info->ordered_bytes, -entry->num_bytes, + fs_info->delalloc_batch); tree = &btrfs_inode->ordered_tree; spin_lock_irq(&tree->lock); @@ -745,9 +770,10 @@ struct btrfs_ordered_extent *btrfs_lookup_ordered_extent(struct btrfs_inode *ino struct btrfs_ordered_inode_tree *tree; struct rb_node *node; struct btrfs_ordered_extent *entry = NULL; + unsigned long flags; tree = &inode->ordered_tree; - spin_lock_irq(&tree->lock); + spin_lock_irqsave(&tree->lock, flags); node = tree_search(tree, file_offset); if (!node) goto out; @@ -758,7 +784,7 @@ struct btrfs_ordered_extent *btrfs_lookup_ordered_extent(struct btrfs_inode *ino if (entry) refcount_inc(&entry->refs); out: - spin_unlock_irq(&tree->lock); + spin_unlock_irqrestore(&tree->lock, flags); return entry; } @@ -898,6 +924,84 @@ void btrfs_lock_and_flush_ordered_range(struct btrfs_inode *inode, u64 start, } } +static int clone_ordered_extent(struct btrfs_ordered_extent *ordered, u64 pos, + u64 len) +{ + struct inode *inode = ordered->inode; + u64 file_offset = ordered->file_offset + pos; + u64 disk_bytenr = ordered->disk_bytenr + pos; + u64 num_bytes = len; + u64 disk_num_bytes = len; + int type; + unsigned long flags_masked = ordered->flags & ~(1 << BTRFS_ORDERED_DIRECT); + int compress_type = ordered->compress_type; + unsigned long weight; + int ret; + + weight = hweight_long(flags_masked); + WARN_ON_ONCE(weight > 1); + if (!weight) + type = 0; + else + type = __ffs(flags_masked); + + if (test_bit(BTRFS_ORDERED_COMPRESSED, &ordered->flags)) { + WARN_ON_ONCE(1); + ret = btrfs_add_ordered_extent_compress(BTRFS_I(inode), + file_offset, disk_bytenr, num_bytes, + disk_num_bytes, compress_type); + } else if (test_bit(BTRFS_ORDERED_DIRECT, &ordered->flags)) { + ret = btrfs_add_ordered_extent_dio(BTRFS_I(inode), file_offset, + disk_bytenr, num_bytes, disk_num_bytes, type); + } else { + ret = btrfs_add_ordered_extent(BTRFS_I(inode), file_offset, + disk_bytenr, num_bytes, disk_num_bytes, type); + } + + return ret; +} + +int btrfs_split_ordered_extent(struct btrfs_ordered_extent *ordered, u64 pre, + u64 post) +{ + struct inode *inode = ordered->inode; + struct btrfs_ordered_inode_tree *tree = &BTRFS_I(inode)->ordered_tree; + struct rb_node *node; + struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb); + int ret = 0; + + spin_lock_irq(&tree->lock); + /* Remove from tree once */ + node = &ordered->rb_node; + rb_erase(node, &tree->tree); + RB_CLEAR_NODE(node); + if (tree->last == node) + tree->last = NULL; + + ordered->file_offset += pre; + ordered->disk_bytenr += pre; + ordered->num_bytes -= (pre + post); + ordered->disk_num_bytes -= (pre + post); + ordered->bytes_left -= (pre + post); + + /* Re-insert the node */ + node = tree_insert(&tree->tree, ordered->file_offset, &ordered->rb_node); + if (node) + btrfs_panic(fs_info, -EEXIST, + "zoned: inconsistency in ordered tree at offset %llu", + ordered->file_offset); + + spin_unlock_irq(&tree->lock); + + if (pre) + ret = clone_ordered_extent(ordered, 0, pre); + if (post) + ret = clone_ordered_extent(ordered, pre + ordered->disk_num_bytes, + post); + + return ret; +} + int __init ordered_data_init(void) { btrfs_ordered_extent_cache = kmem_cache_create("btrfs_ordered_extent", diff --git a/fs/btrfs/ordered-data.h b/fs/btrfs/ordered-data.h index 0bfa82b58e23..99e0853e4d3b 100644 --- a/fs/btrfs/ordered-data.h +++ b/fs/btrfs/ordered-data.h @@ -27,7 +27,7 @@ struct btrfs_ordered_sum { }; /* - * bits for the flags field: + * Bits for btrfs_ordered_extent::flags. * * BTRFS_ORDERED_IO_DONE is set when all of the blocks are written. * It is used to make sure metadata is inserted into the tree only once @@ -38,24 +38,36 @@ struct btrfs_ordered_sum { * IO is done and any metadata is inserted into the tree. */ enum { + /* + * Different types for direct io, one and only one of the 4 type can + * be set when creating ordered extent. + * + * REGULAR: For regular non-compressed COW write + * NOCOW: For NOCOW write into existing non-hole extent + * PREALLOC: For NOCOW write into preallocated extent + * COMPRESSED: For compressed COW write + */ + BTRFS_ORDERED_REGULAR, + BTRFS_ORDERED_NOCOW, + BTRFS_ORDERED_PREALLOC, + BTRFS_ORDERED_COMPRESSED, + + /* + * Extra bit for direct io, can only be set for + * REGULAR/NOCOW/PREALLOC. No direct io for compressed extent. + */ + BTRFS_ORDERED_DIRECT, + + /* Extra status bits for ordered extents */ + /* set when all the pages are written */ BTRFS_ORDERED_IO_DONE, /* set when removed from the tree */ BTRFS_ORDERED_COMPLETE, - /* set when we want to write in place */ - BTRFS_ORDERED_NOCOW, - /* writing a zlib compressed extent */ - BTRFS_ORDERED_COMPRESSED, - /* set when writing to preallocated extent */ - BTRFS_ORDERED_PREALLOC, - /* set when we're doing DIO with this extent */ - BTRFS_ORDERED_DIRECT, /* We had an io error when writing this out */ BTRFS_ORDERED_IOERR, /* Set when we have to truncate an extent */ BTRFS_ORDERED_TRUNCATED, - /* Regular IO for COW */ - BTRFS_ORDERED_REGULAR, /* Used during fsync to track already logged extents */ BTRFS_ORDERED_LOGGED, /* We have already logged all the csums of the ordered extent */ @@ -127,6 +139,14 @@ struct btrfs_ordered_extent { struct completion completion; struct btrfs_work flush_work; struct list_head work_list; + + /* + * Used to reverse-map physical address returned from ZONE_APPEND write + * command in a workqueue context + */ + u64 physical; + struct gendisk *disk; + u8 partno; }; /* @@ -152,11 +172,11 @@ btrfs_ordered_inode_tree_init(struct btrfs_ordered_inode_tree *t) void btrfs_put_ordered_extent(struct btrfs_ordered_extent *entry); void btrfs_remove_ordered_extent(struct btrfs_inode *btrfs_inode, struct btrfs_ordered_extent *entry); -int btrfs_dec_test_ordered_pending(struct btrfs_inode *inode, - struct btrfs_ordered_extent **cached, - u64 file_offset, u64 io_size, int uptodate); -int btrfs_dec_test_first_ordered_pending(struct btrfs_inode *inode, - struct btrfs_ordered_extent **cached, +bool btrfs_dec_test_ordered_pending(struct btrfs_inode *inode, + struct btrfs_ordered_extent **cached, + u64 file_offset, u64 io_size, int uptodate); +bool btrfs_dec_test_first_ordered_pending(struct btrfs_inode *inode, + struct btrfs_ordered_extent **finished_ret, u64 *file_offset, u64 io_size, int uptodate); int btrfs_add_ordered_extent(struct btrfs_inode *inode, u64 file_offset, @@ -167,8 +187,7 @@ int btrfs_add_ordered_extent_dio(struct btrfs_inode *inode, u64 file_offset, u64 disk_num_bytes, int type); int btrfs_add_ordered_extent_compress(struct btrfs_inode *inode, u64 file_offset, u64 disk_bytenr, u64 num_bytes, - u64 disk_num_bytes, int type, - int compress_type); + u64 disk_num_bytes, int compress_type); void btrfs_add_ordered_sum(struct btrfs_ordered_extent *entry, struct btrfs_ordered_sum *sum); struct btrfs_ordered_extent *btrfs_lookup_ordered_extent(struct btrfs_inode *inode, @@ -190,6 +209,8 @@ void btrfs_wait_ordered_roots(struct btrfs_fs_info *fs_info, u64 nr, void btrfs_lock_and_flush_ordered_range(struct btrfs_inode *inode, u64 start, u64 end, struct extent_state **cached_state); +int btrfs_split_ordered_extent(struct btrfs_ordered_extent *ordered, u64 pre, + u64 post); int __init ordered_data_init(void); void __cold ordered_data_exit(void); diff --git a/fs/btrfs/qgroup.c b/fs/btrfs/qgroup.c index 808370ada888..14ff388fd3bd 100644 --- a/fs/btrfs/qgroup.c +++ b/fs/btrfs/qgroup.c @@ -3841,8 +3841,8 @@ static int sub_root_meta_rsv(struct btrfs_root *root, int num_bytes, return num_bytes; } -static int qgroup_reserve_meta(struct btrfs_root *root, int num_bytes, - enum btrfs_qgroup_rsv_type type, bool enforce) +int btrfs_qgroup_reserve_meta(struct btrfs_root *root, int num_bytes, + enum btrfs_qgroup_rsv_type type, bool enforce) { struct btrfs_fs_info *fs_info = root->fs_info; int ret; @@ -3873,14 +3873,14 @@ int __btrfs_qgroup_reserve_meta(struct btrfs_root *root, int num_bytes, { int ret; - ret = qgroup_reserve_meta(root, num_bytes, type, enforce); + ret = btrfs_qgroup_reserve_meta(root, num_bytes, type, enforce); if (ret <= 0 && ret != -EDQUOT) return ret; ret = try_flush_qgroup(root); if (ret < 0) return ret; - return qgroup_reserve_meta(root, num_bytes, type, enforce); + return btrfs_qgroup_reserve_meta(root, num_bytes, type, enforce); } void btrfs_qgroup_free_meta_all_pertrans(struct btrfs_root *root) diff --git a/fs/btrfs/qgroup.h b/fs/btrfs/qgroup.h index 50dea9a2d8fb..7283e4f549af 100644 --- a/fs/btrfs/qgroup.h +++ b/fs/btrfs/qgroup.h @@ -361,6 +361,8 @@ int btrfs_qgroup_release_data(struct btrfs_inode *inode, u64 start, u64 len); int btrfs_qgroup_free_data(struct btrfs_inode *inode, struct extent_changeset *reserved, u64 start, u64 len); +int btrfs_qgroup_reserve_meta(struct btrfs_root *root, int num_bytes, + enum btrfs_qgroup_rsv_type type, bool enforce); int __btrfs_qgroup_reserve_meta(struct btrfs_root *root, int num_bytes, enum btrfs_qgroup_rsv_type type, bool enforce); /* Reserve metadata space for pertrans and prealloc type */ diff --git a/fs/btrfs/raid56.c b/fs/btrfs/raid56.c index 93fbf87bdc8d..8c31357f08ed 100644 --- a/fs/btrfs/raid56.c +++ b/fs/btrfs/raid56.c @@ -233,8 +233,7 @@ int btrfs_alloc_stripe_hash_table(struct btrfs_fs_info *info) } x = cmpxchg(&info->stripe_hash_table, NULL, table); - if (x) - kvfree(x); + kvfree(x); return 0; } @@ -250,8 +249,6 @@ int btrfs_alloc_stripe_hash_table(struct btrfs_fs_info *info) static void cache_rbio_pages(struct btrfs_raid_bio *rbio) { int i; - char *s; - char *d; int ret; ret = alloc_rbio_pages(rbio); @@ -262,13 +259,7 @@ static void cache_rbio_pages(struct btrfs_raid_bio *rbio) if (!rbio->bio_pages[i]) continue; - s = kmap(rbio->bio_pages[i]); - d = kmap(rbio->stripe_pages[i]); - - copy_page(d, s); - - kunmap(rbio->bio_pages[i]); - kunmap(rbio->stripe_pages[i]); + copy_highpage(rbio->stripe_pages[i], rbio->bio_pages[i]); SetPageUptodate(rbio->stripe_pages[i]); } set_bit(RBIO_CACHE_READY_BIT, &rbio->flags); @@ -1105,8 +1096,7 @@ static int rbio_add_io_page(struct btrfs_raid_bio *rbio, * devices or if they are not contiguous */ if (last_end == disk_start && !last->bi_status && - last->bi_disk == stripe->dev->bdev->bd_disk && - last->bi_partno == stripe->dev->bdev->bd_partno) { + last->bi_bdev == stripe->dev->bdev) { ret = bio_add_page(last, page, PAGE_SIZE, 0); if (ret == PAGE_SIZE) return 0; @@ -1357,9 +1347,7 @@ static int find_bio_stripe(struct btrfs_raid_bio *rbio, for (i = 0; i < rbio->bbio->num_stripes; i++) { stripe = &rbio->bbio->stripes[i]; if (in_range(physical, stripe->physical, rbio->stripe_len) && - stripe->dev->bdev && - bio->bi_disk == stripe->dev->bdev->bd_disk && - bio->bi_partno == stripe->dev->bdev->bd_partno) { + stripe->dev->bdev && bio->bi_bdev == stripe->dev->bdev) { return i; } } @@ -2363,16 +2351,21 @@ static noinline void finish_parity_scrub(struct btrfs_raid_bio *rbio, SetPageUptodate(p_page); if (has_qstripe) { + /* RAID6, allocate and map temp space for the Q stripe */ q_page = alloc_page(GFP_NOFS | __GFP_HIGHMEM); if (!q_page) { __free_page(p_page); goto cleanup; } SetPageUptodate(q_page); + pointers[rbio->real_stripes - 1] = kmap(q_page); } atomic_set(&rbio->error, 0); + /* Map the parity stripe just once */ + pointers[nr_data] = kmap(p_page); + for_each_set_bit(pagenr, rbio->dbitmap, rbio->stripe_npages) { struct page *p; void *parity; @@ -2382,16 +2375,8 @@ static noinline void finish_parity_scrub(struct btrfs_raid_bio *rbio, pointers[stripe] = kmap(p); } - /* then add the parity stripe */ - pointers[stripe++] = kmap(p_page); - if (has_qstripe) { - /* - * raid6, add the qstripe and call the - * library function to fill in our p/q - */ - pointers[stripe++] = kmap(q_page); - + /* RAID6, call the library function to fill in our P/Q */ raid6_call.gen_syndrome(rbio->real_stripes, PAGE_SIZE, pointers); } else { @@ -2412,12 +2397,14 @@ static noinline void finish_parity_scrub(struct btrfs_raid_bio *rbio, for (stripe = 0; stripe < nr_data; stripe++) kunmap(page_in_rbio(rbio, stripe, pagenr, 0)); - kunmap(p_page); } + kunmap(p_page); __free_page(p_page); - if (q_page) + if (q_page) { + kunmap(q_page); __free_page(q_page); + } writeback: /* diff --git a/fs/btrfs/reada.c b/fs/btrfs/reada.c index 20fd4aa48a8c..06713a8fe26b 100644 --- a/fs/btrfs/reada.c +++ b/fs/btrfs/reada.c @@ -209,7 +209,7 @@ int btree_readahead_hook(struct extent_buffer *eb, int err) /* find extent */ spin_lock(&fs_info->reada_lock); re = radix_tree_lookup(&fs_info->reada_tree, - eb->start >> PAGE_SHIFT); + eb->start >> fs_info->sectorsize_bits); if (re) re->refcnt++; spin_unlock(&fs_info->reada_lock); @@ -240,7 +240,7 @@ static struct reada_zone *reada_find_zone(struct btrfs_device *dev, u64 logical, zone = NULL; spin_lock(&fs_info->reada_lock); ret = radix_tree_gang_lookup(&dev->reada_zones, (void **)&zone, - logical >> PAGE_SHIFT, 1); + logical >> fs_info->sectorsize_bits, 1); if (ret == 1 && logical >= zone->start && logical <= zone->end) { kref_get(&zone->refcnt); spin_unlock(&fs_info->reada_lock); @@ -283,13 +283,13 @@ static struct reada_zone *reada_find_zone(struct btrfs_device *dev, u64 logical, spin_lock(&fs_info->reada_lock); ret = radix_tree_insert(&dev->reada_zones, - (unsigned long)(zone->end >> PAGE_SHIFT), - zone); + (unsigned long)(zone->end >> fs_info->sectorsize_bits), + zone); if (ret == -EEXIST) { kfree(zone); ret = radix_tree_gang_lookup(&dev->reada_zones, (void **)&zone, - logical >> PAGE_SHIFT, 1); + logical >> fs_info->sectorsize_bits, 1); if (ret == 1 && logical >= zone->start && logical <= zone->end) kref_get(&zone->refcnt); else @@ -315,7 +315,7 @@ static struct reada_extent *reada_find_extent(struct btrfs_fs_info *fs_info, u64 length; int real_stripes; int nzones = 0; - unsigned long index = logical >> PAGE_SHIFT; + unsigned long index = logical >> fs_info->sectorsize_bits; int dev_replace_is_ongoing; int have_zone = 0; @@ -497,7 +497,7 @@ static void reada_extent_put(struct btrfs_fs_info *fs_info, struct reada_extent *re) { int i; - unsigned long index = re->logical >> PAGE_SHIFT; + unsigned long index = re->logical >> fs_info->sectorsize_bits; spin_lock(&fs_info->reada_lock); if (--re->refcnt) { @@ -538,11 +538,12 @@ static void reada_extent_put(struct btrfs_fs_info *fs_info, static void reada_zone_release(struct kref *kref) { struct reada_zone *zone = container_of(kref, struct reada_zone, refcnt); + struct btrfs_fs_info *fs_info = zone->device->fs_info; - lockdep_assert_held(&zone->device->fs_info->reada_lock); + lockdep_assert_held(&fs_info->reada_lock); radix_tree_delete(&zone->device->reada_zones, - zone->end >> PAGE_SHIFT); + zone->end >> fs_info->sectorsize_bits); kfree(zone); } @@ -593,7 +594,7 @@ static int reada_add_block(struct reada_control *rc, u64 logical, static void reada_peer_zones_set_lock(struct reada_zone *zone, int lock) { int i; - unsigned long index = zone->end >> PAGE_SHIFT; + unsigned long index = zone->end >> zone->device->fs_info->sectorsize_bits; for (i = 0; i < zone->ndevs; ++i) { struct reada_zone *peer; @@ -628,7 +629,7 @@ static int reada_pick_zone(struct btrfs_device *dev) (void **)&zone, index, 1); if (ret == 0) break; - index = (zone->end >> PAGE_SHIFT) + 1; + index = (zone->end >> dev->fs_info->sectorsize_bits) + 1; if (zone->locked) { if (zone->elems > top_locked_elems) { top_locked_elems = zone->elems; @@ -709,7 +710,7 @@ static int reada_start_machine_dev(struct btrfs_device *dev) * plugging to speed things up */ ret = radix_tree_gang_lookup(&dev->reada_extents, (void **)&re, - dev->reada_next >> PAGE_SHIFT, 1); + dev->reada_next >> fs_info->sectorsize_bits, 1); if (ret == 0 || re->logical > dev->reada_curr_zone->end) { ret = reada_pick_zone(dev); if (!ret) { @@ -718,7 +719,7 @@ static int reada_start_machine_dev(struct btrfs_device *dev) } re = NULL; ret = radix_tree_gang_lookup(&dev->reada_extents, (void **)&re, - dev->reada_next >> PAGE_SHIFT, 1); + dev->reada_next >> fs_info->sectorsize_bits, 1); } if (ret == 0) { spin_unlock(&fs_info->reada_lock); @@ -885,7 +886,7 @@ static void dump_devs(struct btrfs_fs_info *fs_info, int all) pr_cont(" curr off %llu", device->reada_next - zone->start); pr_cont("\n"); - index = (zone->end >> PAGE_SHIFT) + 1; + index = (zone->end >> fs_info->sectorsize_bits) + 1; } cnt = 0; index = 0; @@ -910,7 +911,7 @@ static void dump_devs(struct btrfs_fs_info *fs_info, int all) } } pr_cont("\n"); - index = (re->logical >> PAGE_SHIFT) + 1; + index = (re->logical >> fs_info->sectorsize_bits) + 1; if (++cnt > 15) break; } @@ -926,7 +927,7 @@ static void dump_devs(struct btrfs_fs_info *fs_info, int all) if (ret == 0) break; if (!re->scheduled) { - index = (re->logical >> PAGE_SHIFT) + 1; + index = (re->logical >> fs_info->sectorsize_bits) + 1; continue; } pr_debug("re: logical %llu size %u list empty %d scheduled %d", @@ -942,7 +943,7 @@ static void dump_devs(struct btrfs_fs_info *fs_info, int all) } } pr_cont("\n"); - index = (re->logical >> PAGE_SHIFT) + 1; + index = (re->logical >> fs_info->sectorsize_bits) + 1; } spin_unlock(&fs_info->reada_lock); } diff --git a/fs/btrfs/ref-verify.c b/fs/btrfs/ref-verify.c index 4b9b6c52a83b..8e026de74c44 100644 --- a/fs/btrfs/ref-verify.c +++ b/fs/btrfs/ref-verify.c @@ -218,11 +218,11 @@ static void __print_stack_trace(struct btrfs_fs_info *fs_info, stack_trace_print(ra->trace, ra->trace_len, 2); } #else -static void inline __save_stack_trace(struct ref_action *ra) +static inline void __save_stack_trace(struct ref_action *ra) { } -static void inline __print_stack_trace(struct btrfs_fs_info *fs_info, +static inline void __print_stack_trace(struct btrfs_fs_info *fs_info, struct ref_action *ra) { btrfs_err(fs_info, " ref-verify: no stacktrace support"); @@ -495,14 +495,15 @@ static int process_extent_item(struct btrfs_fs_info *fs_info, } static int process_leaf(struct btrfs_root *root, - struct btrfs_path *path, u64 *bytenr, u64 *num_bytes) + struct btrfs_path *path, u64 *bytenr, u64 *num_bytes, + int *tree_block_level) { struct btrfs_fs_info *fs_info = root->fs_info; struct extent_buffer *leaf = path->nodes[0]; struct btrfs_extent_data_ref *dref; struct btrfs_shared_data_ref *sref; u32 count; - int i = 0, tree_block_level = 0, ret = 0; + int i = 0, ret = 0; struct btrfs_key key; int nritems = btrfs_header_nritems(leaf); @@ -515,15 +516,15 @@ static int process_leaf(struct btrfs_root *root, case BTRFS_METADATA_ITEM_KEY: *bytenr = key.objectid; ret = process_extent_item(fs_info, path, &key, i, - &tree_block_level); + tree_block_level); break; case BTRFS_TREE_BLOCK_REF_KEY: ret = add_tree_block(fs_info, key.offset, 0, - key.objectid, tree_block_level); + key.objectid, *tree_block_level); break; case BTRFS_SHARED_BLOCK_REF_KEY: ret = add_tree_block(fs_info, 0, key.offset, - key.objectid, tree_block_level); + key.objectid, *tree_block_level); break; case BTRFS_EXTENT_DATA_REF_KEY: dref = btrfs_item_ptr(leaf, i, @@ -549,7 +550,8 @@ static int process_leaf(struct btrfs_root *root, /* Walk down to the leaf from the given level */ static int walk_down_tree(struct btrfs_root *root, struct btrfs_path *path, - int level, u64 *bytenr, u64 *num_bytes) + int level, u64 *bytenr, u64 *num_bytes, + int *tree_block_level) { struct extent_buffer *eb; int ret = 0; @@ -565,7 +567,8 @@ static int walk_down_tree(struct btrfs_root *root, struct btrfs_path *path, path->slots[level-1] = 0; path->locks[level-1] = BTRFS_READ_LOCK; } else { - ret = process_leaf(root, path, bytenr, num_bytes); + ret = process_leaf(root, path, bytenr, num_bytes, + tree_block_level); if (ret) break; } @@ -666,18 +669,18 @@ int btrfs_ref_tree_mod(struct btrfs_fs_info *fs_info, u64 bytenr = generic_ref->bytenr; u64 num_bytes = generic_ref->len; u64 parent = generic_ref->parent; - u64 ref_root; - u64 owner; - u64 offset; + u64 ref_root = 0; + u64 owner = 0; + u64 offset = 0; if (!btrfs_test_opt(fs_info, REF_VERIFY)) return 0; if (generic_ref->type == BTRFS_REF_METADATA) { - ref_root = generic_ref->tree_ref.root; + if (!parent) + ref_root = generic_ref->tree_ref.root; owner = generic_ref->tree_ref.level; - offset = 0; - } else { + } else if (!parent) { ref_root = generic_ref->data_ref.ref_root; owner = generic_ref->data_ref.ino; offset = generic_ref->data_ref.offset; @@ -693,13 +696,10 @@ int btrfs_ref_tree_mod(struct btrfs_fs_info *fs_info, goto out; } - if (parent) { - ref->parent = parent; - } else { - ref->root_objectid = ref_root; - ref->owner = owner; - ref->offset = offset; - } + ref->parent = parent; + ref->owner = owner; + ref->root_objectid = ref_root; + ref->offset = offset; ref->num_refs = (action == BTRFS_DROP_DELAYED_REF) ? -1 : 1; memcpy(&ra->ref, ref, sizeof(struct ref_entry)); @@ -974,6 +974,7 @@ int btrfs_build_ref_tree(struct btrfs_fs_info *fs_info) { struct btrfs_path *path; struct extent_buffer *eb; + int tree_block_level = 0; u64 bytenr = 0, num_bytes = 0; int ret, level; @@ -998,7 +999,7 @@ int btrfs_build_ref_tree(struct btrfs_fs_info *fs_info) * different leaf from the original extent item. */ ret = walk_down_tree(fs_info->extent_root, path, level, - &bytenr, &num_bytes); + &bytenr, &num_bytes, &tree_block_level); if (ret) break; ret = walk_up_tree(path, &level); diff --git a/fs/btrfs/reflink.c b/fs/btrfs/reflink.c index b03e7891394e..762881b777b3 100644 --- a/fs/btrfs/reflink.c +++ b/fs/btrfs/reflink.c @@ -81,7 +81,10 @@ static int copy_inline_to_page(struct btrfs_inode *inode, goto out_unlock; } - set_page_extent_mapped(page); + ret = set_page_extent_mapped(page); + if (ret < 0) + goto out_unlock; + clear_extent_bit(&inode->io_tree, file_offset, range_end, EXTENT_DELALLOC | EXTENT_DO_ACCOUNTING | EXTENT_DEFRAG, 0, 0, NULL); @@ -103,12 +106,8 @@ static int copy_inline_to_page(struct btrfs_inode *inode, set_bit(BTRFS_INODE_NO_DELALLOC_FLUSH, &inode->runtime_flags); if (comp_type == BTRFS_COMPRESS_NONE) { - char *map; - - map = kmap(page); - memcpy(map, data_start, datal); + memcpy_to_page(page, 0, data_start, datal); flush_dcache_page(page); - kunmap(page); } else { ret = btrfs_decompress(comp_type, data_start, page, 0, inline_size, datal); @@ -550,6 +549,24 @@ process_slot: */ btrfs_release_path(path); + /* + * When using NO_HOLES and we are cloning a range that covers + * only a hole (no extents) into a range beyond the current + * i_size, punching a hole in the target range will not create + * an extent map defining a hole, because the range starts at or + * beyond current i_size. If the file previously had an i_size + * greater than the new i_size set by this clone operation, we + * need to make sure the next fsync is a full fsync, so that it + * detects and logs a hole covering a range from the current + * i_size to the new i_size. If the clone range covers extents, + * besides a hole, then we know the full sync flag was already + * set by previous calls to btrfs_replace_file_extents() that + * replaced file extent items. + */ + if (last_dest_end >= i_size_read(inode)) + set_bit(BTRFS_INODE_NEEDS_FULL_SYNC, + &BTRFS_I(inode)->runtime_flags); + ret = btrfs_replace_file_extents(inode, path, last_dest_end, destoff + len - 1, NULL, &trans); if (ret) diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index df63ef64c5c0..232d5da7b7be 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -97,6 +97,7 @@ struct tree_block { struct rb_node rb_node; u64 bytenr; }; /* Use rb_simple_node for search/insert */ + u64 owner; struct btrfs_key key; unsigned int level:8; unsigned int key_ready:1; @@ -668,9 +669,7 @@ static void __del_reloc_root(struct btrfs_root *root) RB_CLEAR_NODE(&node->rb_node); } spin_unlock(&rc->reloc_root_tree.lock); - if (!node) - return; - BUG_ON((struct btrfs_root *)node->data != root); + ASSERT(!node || (struct btrfs_root *)node->data == root); } /* @@ -2393,8 +2392,8 @@ static int get_tree_block_key(struct btrfs_fs_info *fs_info, { struct extent_buffer *eb; - eb = read_tree_block(fs_info, block->bytenr, 0, block->key.offset, - block->level, NULL); + eb = read_tree_block(fs_info, block->bytenr, block->owner, + block->key.offset, block->level, NULL); if (IS_ERR(eb)) { return PTR_ERR(eb); } else if (!extent_buffer_uptodate(eb)) { @@ -2493,7 +2492,8 @@ int relocate_tree_blocks(struct btrfs_trans_handle *trans, /* Kick in readahead for tree blocks with missing keys */ rbtree_postorder_for_each_entry_safe(block, next, blocks, rb_node) { if (!block->key_ready) - btrfs_readahead_tree_block(fs_info, block->bytenr, 0, 0, + btrfs_readahead_tree_block(fs_info, block->bytenr, + block->owner, 0, block->level); } @@ -2553,6 +2553,31 @@ static noinline_for_stack int prealloc_file_extent_cluster( if (ret) return ret; + /* + * On a zoned filesystem, we cannot preallocate the file region. + * Instead, we dirty and fiemap_write the region. + */ + if (btrfs_is_zoned(inode->root->fs_info)) { + struct btrfs_root *root = inode->root; + struct btrfs_trans_handle *trans; + + end = cluster->end - offset + 1; + trans = btrfs_start_transaction(root, 1); + if (IS_ERR(trans)) + return PTR_ERR(trans); + + inode->vfs_inode.i_ctime = current_time(&inode->vfs_inode); + i_size_write(&inode->vfs_inode, end); + ret = btrfs_update_inode(trans, root, inode); + if (ret) { + btrfs_abort_transaction(trans, ret); + btrfs_end_transaction(trans); + return ret; + } + + return btrfs_end_transaction(trans); + } + inode_lock(&inode->vfs_inode); for (nr = 0; nr < cluster->nr; nr++) { start = cluster->boundary[nr] - offset; @@ -2615,7 +2640,7 @@ int setup_extent_mapping(struct inode *inode, u64 start, u64 end, /* * Allow error injection to test balance cancellation */ -int btrfs_should_cancel_balance(struct btrfs_fs_info *fs_info) +noinline int btrfs_should_cancel_balance(struct btrfs_fs_info *fs_info) { return atomic_read(&fs_info->balance_cancel_req) || fatal_signal_pending(current); @@ -2679,6 +2704,15 @@ static int relocate_file_extent_cluster(struct inode *inode, goto out; } } + ret = set_page_extent_mapped(page); + if (ret < 0) { + btrfs_delalloc_release_metadata(BTRFS_I(inode), + PAGE_SIZE, true); + btrfs_delalloc_release_extents(BTRFS_I(inode), PAGE_SIZE); + unlock_page(page); + put_page(page); + goto out; + } if (PageReadahead(page)) { page_cache_async_readahead(inode->i_mapping, @@ -2706,8 +2740,6 @@ static int relocate_file_extent_cluster(struct inode *inode, lock_extent(&BTRFS_I(inode)->io_tree, page_start, page_end); - set_page_extent_mapped(page); - if (nr < cluster->nr && page_start + offset == cluster->boundary[nr]) { set_extent_bits(&BTRFS_I(inode)->io_tree, @@ -2749,6 +2781,8 @@ static int relocate_file_extent_cluster(struct inode *inode, } } WARN_ON(nr != cluster->nr); + if (btrfs_is_zoned(fs_info) && !ret) + ret = btrfs_wait_ordered_range(inode, 0, (u64)-1); out: kfree(ra); return ret; @@ -2801,21 +2835,58 @@ static int add_tree_block(struct reloc_control *rc, u32 item_size; int level = -1; u64 generation; + u64 owner = 0; eb = path->nodes[0]; item_size = btrfs_item_size_nr(eb, path->slots[0]); if (extent_key->type == BTRFS_METADATA_ITEM_KEY || item_size >= sizeof(*ei) + sizeof(*bi)) { + unsigned long ptr = 0, end; + ei = btrfs_item_ptr(eb, path->slots[0], struct btrfs_extent_item); + end = (unsigned long)ei + item_size; if (extent_key->type == BTRFS_EXTENT_ITEM_KEY) { bi = (struct btrfs_tree_block_info *)(ei + 1); level = btrfs_tree_block_level(eb, bi); + ptr = (unsigned long)(bi + 1); } else { level = (int)extent_key->offset; + ptr = (unsigned long)(ei + 1); } generation = btrfs_extent_generation(eb, ei); + + /* + * We're reading random blocks without knowing their owner ahead + * of time. This is ok most of the time, as all reloc roots and + * fs roots have the same lock type. However normal trees do + * not, and the only way to know ahead of time is to read the + * inline ref offset. We know it's an fs root if + * + * 1. There's more than one ref. + * 2. There's a SHARED_DATA_REF_KEY set. + * 3. FULL_BACKREF is set on the flags. + * + * Otherwise it's safe to assume that the ref offset == the + * owner of this block, so we can use that when calling + * read_tree_block. + */ + if (btrfs_extent_refs(eb, ei) == 1 && + !(btrfs_extent_flags(eb, ei) & + BTRFS_BLOCK_FLAG_FULL_BACKREF) && + ptr < end) { + struct btrfs_extent_inline_ref *iref; + int type; + + iref = (struct btrfs_extent_inline_ref *)ptr; + type = btrfs_get_extent_inline_ref_type(eb, iref, + BTRFS_REF_TYPE_BLOCK); + if (type == BTRFS_REF_TYPE_INVALID) + return -EINVAL; + if (type == BTRFS_TREE_BLOCK_REF_KEY) + owner = btrfs_extent_inline_ref_offset(eb, iref); + } } else if (unlikely(item_size == sizeof(struct btrfs_extent_item_v0))) { btrfs_print_v0_err(eb->fs_info); btrfs_handle_fs_error(eb->fs_info, -EINVAL, NULL); @@ -2837,6 +2908,7 @@ static int add_tree_block(struct reloc_control *rc, block->key.offset = generation; block->level = level; block->key_ready = 0; + block->owner = owner; rb_node = rb_simple_insert(blocks, block->bytenr, &block->rb_node); if (rb_node) @@ -3389,8 +3461,12 @@ static int __insert_orphan_inode(struct btrfs_trans_handle *trans, struct btrfs_path *path; struct btrfs_inode_item *item; struct extent_buffer *leaf; + u64 flags = BTRFS_INODE_NOCOMPRESS | BTRFS_INODE_PREALLOC; int ret; + if (btrfs_is_zoned(trans->fs_info)) + flags &= ~BTRFS_INODE_PREALLOC; + path = btrfs_alloc_path(); if (!path) return -ENOMEM; @@ -3405,8 +3481,7 @@ static int __insert_orphan_inode(struct btrfs_trans_handle *trans, btrfs_set_inode_generation(leaf, item, 1); btrfs_set_inode_size(leaf, item, 0); btrfs_set_inode_mode(leaf, item, S_IFREG | 0600); - btrfs_set_inode_flags(leaf, item, BTRFS_INODE_NOCOMPRESS | - BTRFS_INODE_PREALLOC); + btrfs_set_inode_flags(leaf, item, flags); btrfs_mark_buffer_dirty(leaf); out: btrfs_free_path(path); @@ -3434,7 +3509,7 @@ struct inode *create_reloc_inode(struct btrfs_fs_info *fs_info, return ERR_CAST(trans); } - err = btrfs_find_free_objectid(root, &objectid); + err = btrfs_get_free_objectid(root, &objectid); if (err) goto out; diff --git a/fs/btrfs/scrub.c b/fs/btrfs/scrub.c index 5f4f88a4d2c8..3d9088eab2fc 100644 --- a/fs/btrfs/scrub.c +++ b/fs/btrfs/scrub.c @@ -166,6 +166,7 @@ struct scrub_ctx { int pages_per_rd_bio; int is_dev_replace; + u64 write_pointer; struct scrub_bio *wr_curr_bio; struct mutex wr_lock; @@ -856,6 +857,9 @@ static int scrub_handle_errored_block(struct scrub_block *sblock_to_check) have_csum = sblock_to_check->pagev[0]->have_csum; dev = sblock_to_check->pagev[0]->dev; + if (btrfs_is_zoned(fs_info) && !sctx->is_dev_replace) + return btrfs_repair_one_zone(fs_info, logical); + /* * We must use GFP_NOFS because the scrub task might be waiting for a * worker task executing this function and in turn a transaction commit @@ -1424,7 +1428,7 @@ static void scrub_recheck_block_on_raid56(struct btrfs_fs_info *fs_info, if (!first_page->dev->bdev) goto out; - bio = btrfs_io_bio_alloc(BIO_MAX_PAGES); + bio = btrfs_io_bio_alloc(BIO_MAX_VECS); bio_set_dev(bio, first_page->dev->bdev); for (page_num = 0; page_num < sblock->page_count; page_num++) { @@ -1619,6 +1623,28 @@ static int scrub_write_page_to_dev_replace(struct scrub_block *sblock, return scrub_add_page_to_wr_bio(sblock->sctx, spage); } +static int fill_writer_pointer_gap(struct scrub_ctx *sctx, u64 physical) +{ + int ret = 0; + u64 length; + + if (!btrfs_is_zoned(sctx->fs_info)) + return 0; + + if (!btrfs_dev_is_sequential(sctx->wr_tgtdev, physical)) + return 0; + + if (sctx->write_pointer < physical) { + length = physical - sctx->write_pointer; + + ret = btrfs_zoned_issue_zeroout(sctx->wr_tgtdev, + sctx->write_pointer, length); + if (!ret) + sctx->write_pointer = physical; + } + return ret; +} + static int scrub_add_page_to_wr_bio(struct scrub_ctx *sctx, struct scrub_page *spage) { @@ -1641,6 +1667,13 @@ again: if (sbio->page_count == 0) { struct bio *bio; + ret = fill_writer_pointer_gap(sctx, + spage->physical_for_dev_replace); + if (ret) { + mutex_unlock(&sctx->wr_lock); + return ret; + } + sbio->physical = spage->physical_for_dev_replace; sbio->logical = spage->logical; sbio->dev = sctx->wr_tgtdev; @@ -1695,13 +1728,16 @@ static void scrub_wr_submit(struct scrub_ctx *sctx) sbio = sctx->wr_curr_bio; sctx->wr_curr_bio = NULL; - WARN_ON(!sbio->bio->bi_disk); + WARN_ON(!sbio->bio->bi_bdev); scrub_pending_bio_inc(sctx); /* process all writes in a single worker thread. Then the block layer * orders the requests before sending them to the driver which * doubled the write performance on spinning disks when measured * with Linux 3.5 */ btrfsic_submit_bio(sbio->bio); + + if (btrfs_is_zoned(sctx->fs_info)) + sctx->write_pointer = sbio->physical + sbio->page_count * PAGE_SIZE; } static void scrub_wr_bio_end_io(struct bio *bio) @@ -3025,6 +3061,46 @@ out: return ret < 0 ? ret : 0; } +static void sync_replace_for_zoned(struct scrub_ctx *sctx) +{ + if (!btrfs_is_zoned(sctx->fs_info)) + return; + + sctx->flush_all_writes = true; + scrub_submit(sctx); + mutex_lock(&sctx->wr_lock); + scrub_wr_submit(sctx); + mutex_unlock(&sctx->wr_lock); + + wait_event(sctx->list_wait, atomic_read(&sctx->bios_in_flight) == 0); +} + +static int sync_write_pointer_for_zoned(struct scrub_ctx *sctx, u64 logical, + u64 physical, u64 physical_end) +{ + struct btrfs_fs_info *fs_info = sctx->fs_info; + int ret = 0; + + if (!btrfs_is_zoned(fs_info)) + return 0; + + wait_event(sctx->list_wait, atomic_read(&sctx->bios_in_flight) == 0); + + mutex_lock(&sctx->wr_lock); + if (sctx->write_pointer < physical_end) { + ret = btrfs_sync_zone_write_pointer(sctx->wr_tgtdev, logical, + physical, + sctx->write_pointer); + if (ret) + btrfs_err(fs_info, + "zoned: failed to recover write pointer"); + } + mutex_unlock(&sctx->wr_lock); + btrfs_dev_clear_zone_empty(sctx->wr_tgtdev, physical); + + return ret; +} + static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx, struct map_lookup *map, struct btrfs_device *scrub_dev, @@ -3165,6 +3241,14 @@ static noinline_for_stack int scrub_stripe(struct scrub_ctx *sctx, */ blk_start_plug(&plug); + if (sctx->is_dev_replace && + btrfs_dev_is_sequential(sctx->wr_tgtdev, physical)) { + mutex_lock(&sctx->wr_lock); + sctx->write_pointer = physical; + mutex_unlock(&sctx->wr_lock); + sctx->flush_all_writes = true; + } + /* * now find all extents for each stripe and scrub them */ @@ -3353,6 +3437,9 @@ again: if (ret) goto out; + if (sctx->is_dev_replace) + sync_replace_for_zoned(sctx); + if (extent_logical + extent_len < key.objectid + bytes) { if (map->type & BTRFS_BLOCK_GROUP_RAID56_MASK) { @@ -3420,6 +3507,17 @@ out: blk_finish_plug(&plug); btrfs_free_path(path); btrfs_free_path(ppath); + + if (sctx->is_dev_replace && ret >= 0) { + int ret2; + + ret2 = sync_write_pointer_for_zoned(sctx, base + offset, + map->stripes[num].physical, + physical_end); + if (ret2) + ret = ret2; + } + return ret < 0 ? ret : 0; } @@ -3475,6 +3573,25 @@ out: return ret; } +static int finish_extent_writes_for_zoned(struct btrfs_root *root, + struct btrfs_block_group *cache) +{ + struct btrfs_fs_info *fs_info = cache->fs_info; + struct btrfs_trans_handle *trans; + + if (!btrfs_is_zoned(fs_info)) + return 0; + + btrfs_wait_block_group_reservations(cache); + btrfs_wait_nocow_writers(cache); + btrfs_wait_ordered_roots(fs_info, U64_MAX, cache->start, cache->length); + + trans = btrfs_join_transaction(root); + if (IS_ERR(trans)) + return PTR_ERR(trans); + return btrfs_commit_transaction(trans); +} + static noinline_for_stack int scrub_enumerate_chunks(struct scrub_ctx *sctx, struct btrfs_device *scrub_dev, u64 start, u64 end) @@ -3561,6 +3678,16 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx, if (!cache) goto skip; + if (sctx->is_dev_replace && btrfs_is_zoned(fs_info)) { + spin_lock(&cache->lock); + if (!cache->to_copy) { + spin_unlock(&cache->lock); + ro_set = 0; + goto done; + } + spin_unlock(&cache->lock); + } + /* * Make sure that while we are scrubbing the corresponding block * group doesn't get its logical address and its device extents @@ -3619,6 +3746,16 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx, * group is not RO. */ ret = btrfs_inc_block_group_ro(cache, sctx->is_dev_replace); + if (!ret && sctx->is_dev_replace) { + ret = finish_extent_writes_for_zoned(root, cache); + if (ret) { + btrfs_dec_block_group_ro(cache); + scrub_pause_off(fs_info); + btrfs_put_block_group(cache); + break; + } + } + if (ret == 0) { ro_set = 1; } else if (ret == -ENOSPC && !sctx->is_dev_replace) { @@ -3630,6 +3767,13 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx, * commit_transactions. */ ro_set = 0; + } else if (ret == -ETXTBSY) { + btrfs_warn(fs_info, + "skipping scrub of block group %llu due to active swapfile", + cache->start); + scrub_pause_off(fs_info); + ret = 0; + goto skip_unfreeze; } else { btrfs_warn(fs_info, "failed setting block group ro: %d", ret); @@ -3692,6 +3836,12 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx, scrub_pause_off(fs_info); + if (sctx->is_dev_replace && + !btrfs_finish_block_group_to_copy(dev_replace->srcdev, + cache, found_key.offset)) + ro_set = 0; + +done: down_write(&dev_replace->rwsem); dev_replace->cursor_left = dev_replace->cursor_right; dev_replace->item_needs_writeback = 1; @@ -3719,7 +3869,7 @@ int scrub_enumerate_chunks(struct scrub_ctx *sctx, } else { spin_unlock(&cache->lock); } - +skip_unfreeze: btrfs_unfreeze_block_group(cache); btrfs_put_block_group(cache); if (ret) diff --git a/fs/btrfs/send.c b/fs/btrfs/send.c index 78a35374d492..8f323859156b 100644 --- a/fs/btrfs/send.c +++ b/fs/btrfs/send.c @@ -1191,9 +1191,6 @@ struct backref_ctx { /* may be truncated in case it's the last extent in a file */ u64 extent_len; - /* data offset in the file extent item */ - u64 data_offset; - /* Just to check for bugs in backref resolving */ int found_itself; }; @@ -1401,19 +1398,6 @@ static int find_extent_clone(struct send_ctx *sctx, backref_ctx->cur_offset = data_offset; backref_ctx->found_itself = 0; backref_ctx->extent_len = num_bytes; - /* - * For non-compressed extents iterate_extent_inodes() gives us extent - * offsets that already take into account the data offset, but not for - * compressed extents, since the offset is logical and not relative to - * the physical extent locations. We must take this into account to - * avoid sending clone offsets that go beyond the source file's size, - * which would result in the clone ioctl failing with -EINVAL on the - * receiving end. - */ - if (compressed == BTRFS_COMPRESS_NONE) - backref_ctx->data_offset = 0; - else - backref_ctx->data_offset = btrfs_file_extent_offset(eb, fi); /* * The last extent of a file may be too large due to page alignment. @@ -4948,7 +4932,6 @@ static int put_file_data(struct send_ctx *sctx, u64 offset, u32 len) struct btrfs_fs_info *fs_info = root->fs_info; struct inode *inode; struct page *page; - char *addr; pgoff_t index = offset >> PAGE_SHIFT; pgoff_t last_index; unsigned pg_offset = offset_in_page(offset); @@ -5001,10 +4984,8 @@ static int put_file_data(struct send_ctx *sctx, u64 offset, u32 len) } } - addr = kmap(page); - memcpy(sctx->send_buf + sctx->send_size, addr + pg_offset, - cur_len); - kunmap(page); + memcpy_from_page(sctx->send_buf + sctx->send_size, page, + pg_offset, cur_len); unlock_page(page); put_page(page); index++; @@ -6607,10 +6588,9 @@ static int changed_cb(struct btrfs_path *left_path, struct btrfs_path *right_path, struct btrfs_key *key, enum btrfs_compare_tree_result result, - void *ctx) + struct send_ctx *sctx) { int ret = 0; - struct send_ctx *sctx = ctx; if (result == BTRFS_COMPARE_TREE_SAME) { if (key->type == BTRFS_INODE_REF_KEY || @@ -6815,7 +6795,7 @@ static int tree_compare_item(struct btrfs_path *left_path, * If it detects a change, it aborts immediately. */ static int btrfs_compare_trees(struct btrfs_root *left_root, - struct btrfs_root *right_root, void *ctx) + struct btrfs_root *right_root, struct send_ctx *sctx) { struct btrfs_fs_info *fs_info = left_root->fs_info; int ret; @@ -6967,7 +6947,7 @@ static int btrfs_compare_trees(struct btrfs_root *left_root, ret = changed_cb(left_path, right_path, &right_key, BTRFS_COMPARE_TREE_DELETED, - ctx); + sctx); if (ret < 0) goto out; } @@ -6978,7 +6958,7 @@ static int btrfs_compare_trees(struct btrfs_root *left_root, ret = changed_cb(left_path, right_path, &left_key, BTRFS_COMPARE_TREE_NEW, - ctx); + sctx); if (ret < 0) goto out; } @@ -6992,7 +6972,7 @@ static int btrfs_compare_trees(struct btrfs_root *left_root, ret = changed_cb(left_path, right_path, &left_key, BTRFS_COMPARE_TREE_NEW, - ctx); + sctx); if (ret < 0) goto out; advance_left = ADVANCE; @@ -7000,7 +6980,7 @@ static int btrfs_compare_trees(struct btrfs_root *left_root, ret = changed_cb(left_path, right_path, &right_key, BTRFS_COMPARE_TREE_DELETED, - ctx); + sctx); if (ret < 0) goto out; advance_right = ADVANCE; @@ -7015,7 +6995,7 @@ static int btrfs_compare_trees(struct btrfs_root *left_root, else result = BTRFS_COMPARE_TREE_SAME; ret = changed_cb(left_path, right_path, - &left_key, result, ctx); + &left_key, result, sctx); if (ret < 0) goto out; advance_left = ADVANCE; diff --git a/fs/btrfs/space-info.c b/fs/btrfs/space-info.c index e8347461c8dd..2da6177f4b0b 100644 --- a/fs/btrfs/space-info.c +++ b/fs/btrfs/space-info.c @@ -140,6 +140,12 @@ * be freed, plus any delayed work we may not have gotten rid of in the case * of metadata. * + * FORCE_COMMIT_TRANS + * For use by the preemptive flusher. We use this to bypass the ticketing + * checks in may_commit_transaction, as we have more information about the + * overall state of the system and may want to commit the transaction ahead + * of actual ENOSPC conditions. + * * OVERCOMMIT * * Because we hold so many reservations for metadata we will allow you to @@ -163,6 +169,7 @@ u64 __pure btrfs_space_info_used(struct btrfs_space_info *s_info, ASSERT(s_info); return s_info->bytes_used + s_info->bytes_reserved + s_info->bytes_pinned + s_info->bytes_readonly + + s_info->bytes_zone_unusable + (may_use_included ? s_info->bytes_may_use : 0); } @@ -206,6 +213,7 @@ static int create_space_info(struct btrfs_fs_info *info, u64 flags) INIT_LIST_HEAD(&space_info->ro_bgs); INIT_LIST_HEAD(&space_info->tickets); INIT_LIST_HEAD(&space_info->priority_tickets); + space_info->clamp = 1; ret = btrfs_sysfs_add_space_info_type(info, space_info); if (ret) @@ -257,7 +265,7 @@ out: void btrfs_update_space_info(struct btrfs_fs_info *info, u64 flags, u64 total_bytes, u64 bytes_used, - u64 bytes_readonly, + u64 bytes_readonly, u64 bytes_zone_unusable, struct btrfs_space_info **space_info) { struct btrfs_space_info *found; @@ -273,6 +281,7 @@ void btrfs_update_space_info(struct btrfs_fs_info *info, u64 flags, found->bytes_used += bytes_used; found->disk_used += bytes_used * factor; found->bytes_readonly += bytes_readonly; + found->bytes_zone_unusable += bytes_zone_unusable; if (total_bytes > 0) found->full = 0; btrfs_try_granting_tickets(info, found); @@ -422,10 +431,10 @@ static void __btrfs_dump_space_info(struct btrfs_fs_info *fs_info, info->total_bytes - btrfs_space_info_used(info, true), info->full ? "" : "not "); btrfs_info(fs_info, - "space_info total=%llu, used=%llu, pinned=%llu, reserved=%llu, may_use=%llu, readonly=%llu", + "space_info total=%llu, used=%llu, pinned=%llu, reserved=%llu, may_use=%llu, readonly=%llu zone_unusable=%llu", info->total_bytes, info->bytes_used, info->bytes_pinned, info->bytes_reserved, info->bytes_may_use, - info->bytes_readonly); + info->bytes_readonly, info->bytes_zone_unusable); DUMP_BLOCK_RSV(fs_info, global_block_rsv); DUMP_BLOCK_RSV(fs_info, trans_block_rsv); @@ -454,9 +463,10 @@ again: list_for_each_entry(cache, &info->block_groups[index], list) { spin_lock(&cache->lock); btrfs_info(fs_info, - "block group %llu has %llu bytes, %llu used %llu pinned %llu reserved %s", + "block group %llu has %llu bytes, %llu used %llu pinned %llu reserved %llu zone_unusable %s", cache->start, cache->length, cache->used, cache->pinned, - cache->reserved, cache->ro ? "[readonly]" : ""); + cache->reserved, cache->zone_unusable, + cache->ro ? "[readonly]" : ""); spin_unlock(&cache->lock); btrfs_dump_free_space(cache, bytes); } @@ -489,7 +499,7 @@ static void shrink_delalloc(struct btrfs_fs_info *fs_info, { struct btrfs_trans_handle *trans; u64 delalloc_bytes; - u64 dio_bytes; + u64 ordered_bytes; u64 items; long time_left; int loops; @@ -513,26 +523,22 @@ static void shrink_delalloc(struct btrfs_fs_info *fs_info, delalloc_bytes = percpu_counter_sum_positive( &fs_info->delalloc_bytes); - dio_bytes = percpu_counter_sum_positive(&fs_info->dio_bytes); - if (delalloc_bytes == 0 && dio_bytes == 0) { - if (trans) - return; - if (wait_ordered) - btrfs_wait_ordered_roots(fs_info, items, 0, (u64)-1); + ordered_bytes = percpu_counter_sum_positive(&fs_info->ordered_bytes); + if (delalloc_bytes == 0 && ordered_bytes == 0) return; - } /* * If we are doing more ordered than delalloc we need to just wait on * ordered extents, otherwise we'll waste time trying to flush delalloc * that likely won't give us the space back we need. */ - if (dio_bytes > delalloc_bytes) + if (ordered_bytes > delalloc_bytes) wait_ordered = true; loops = 0; - while ((delalloc_bytes || dio_bytes) && loops < 3) { - u64 nr_pages = min(delalloc_bytes, to_reclaim) >> PAGE_SHIFT; + while ((delalloc_bytes || ordered_bytes) && loops < 3) { + u64 temp = min(delalloc_bytes, to_reclaim) >> PAGE_SHIFT; + long nr_pages = min_t(u64, temp, LONG_MAX); btrfs_start_delalloc_roots(fs_info, nr_pages, true); @@ -555,15 +561,16 @@ static void shrink_delalloc(struct btrfs_fs_info *fs_info, delalloc_bytes = percpu_counter_sum_positive( &fs_info->delalloc_bytes); - dio_bytes = percpu_counter_sum_positive(&fs_info->dio_bytes); + ordered_bytes = percpu_counter_sum_positive( + &fs_info->ordered_bytes); } } /** - * maybe_commit_transaction - possibly commit the transaction if its ok to - * @root - the root we're allocating for - * @bytes - the number of bytes we want to reserve - * @force - force the commit + * Possibly commit the transaction if its ok to + * + * @fs_info: the filesystem + * @space_info: space_info we are checking for commit, either data or metadata * * This will check to make sure that committing the transaction will actually * get us somewhere and then commit the transaction if it does. Otherwise it @@ -669,7 +676,7 @@ enospc: */ static void flush_space(struct btrfs_fs_info *fs_info, struct btrfs_space_info *space_info, u64 num_bytes, - int state) + enum btrfs_flush_state state, bool for_preempt) { struct btrfs_root *root = fs_info->extent_root; struct btrfs_trans_handle *trans; @@ -738,13 +745,21 @@ static void flush_space(struct btrfs_fs_info *fs_info, case COMMIT_TRANS: ret = may_commit_transaction(fs_info, space_info); break; + case FORCE_COMMIT_TRANS: + trans = btrfs_join_transaction(root); + if (IS_ERR(trans)) { + ret = PTR_ERR(trans); + break; + } + ret = btrfs_commit_transaction(trans); + break; default: ret = -ENOSPC; break; } trace_btrfs_flush_space(fs_info, space_info->flags, num_bytes, state, - ret); + ret, for_preempt); return; } @@ -754,7 +769,6 @@ btrfs_calc_reclaim_metadata_size(struct btrfs_fs_info *fs_info, { u64 used; u64 avail; - u64 expected; u64 to_reclaim = space_info->reclaim_size; lockdep_assert_held(&space_info->lock); @@ -772,43 +786,88 @@ btrfs_calc_reclaim_metadata_size(struct btrfs_fs_info *fs_info, if (space_info->total_bytes + avail < used) to_reclaim += used - (space_info->total_bytes + avail); - if (to_reclaim) - return to_reclaim; - - to_reclaim = min_t(u64, num_online_cpus() * SZ_1M, SZ_16M); - if (btrfs_can_overcommit(fs_info, space_info, to_reclaim, - BTRFS_RESERVE_FLUSH_ALL)) - return 0; - - used = btrfs_space_info_used(space_info, true); - - if (btrfs_can_overcommit(fs_info, space_info, SZ_1M, - BTRFS_RESERVE_FLUSH_ALL)) - expected = div_factor_fine(space_info->total_bytes, 95); - else - expected = div_factor_fine(space_info->total_bytes, 90); - - if (used > expected) - to_reclaim = used - expected; - else - to_reclaim = 0; - to_reclaim = min(to_reclaim, space_info->bytes_may_use + - space_info->bytes_reserved); return to_reclaim; } -static inline int need_do_async_reclaim(struct btrfs_fs_info *fs_info, - struct btrfs_space_info *space_info, - u64 used) +static bool need_preemptive_reclaim(struct btrfs_fs_info *fs_info, + struct btrfs_space_info *space_info) { + u64 ordered, delalloc; u64 thresh = div_factor_fine(space_info->total_bytes, 98); + u64 used; /* If we're just plain full then async reclaim just slows us down. */ if ((space_info->bytes_used + space_info->bytes_reserved) >= thresh) - return 0; + return false; - if (!btrfs_calc_reclaim_metadata_size(fs_info, space_info)) - return 0; + /* + * We have tickets queued, bail so we don't compete with the async + * flushers. + */ + if (space_info->reclaim_size) + return false; + + /* + * If we have over half of the free space occupied by reservations or + * pinned then we want to start flushing. + * + * We do not do the traditional thing here, which is to say + * + * if (used >= ((total_bytes + avail) / 2)) + * return 1; + * + * because this doesn't quite work how we want. If we had more than 50% + * of the space_info used by bytes_used and we had 0 available we'd just + * constantly run the background flusher. Instead we want it to kick in + * if our reclaimable space exceeds our clamped free space. + * + * Our clamping range is 2^1 -> 2^8. Practically speaking that means + * the following: + * + * Amount of RAM Minimum threshold Maximum threshold + * + * 256GiB 1GiB 128GiB + * 128GiB 512MiB 64GiB + * 64GiB 256MiB 32GiB + * 32GiB 128MiB 16GiB + * 16GiB 64MiB 8GiB + * + * These are the range our thresholds will fall in, corresponding to how + * much delalloc we need for the background flusher to kick in. + */ + + thresh = calc_available_free_space(fs_info, space_info, + BTRFS_RESERVE_FLUSH_ALL); + thresh += (space_info->total_bytes - space_info->bytes_used - + space_info->bytes_reserved - space_info->bytes_readonly); + thresh >>= space_info->clamp; + + used = space_info->bytes_pinned; + + /* + * If we have more ordered bytes than delalloc bytes then we're either + * doing a lot of DIO, or we simply don't have a lot of delalloc waiting + * around. Preemptive flushing is only useful in that it can free up + * space before tickets need to wait for things to finish. In the case + * of ordered extents, preemptively waiting on ordered extents gets us + * nothing, if our reservations are tied up in ordered extents we'll + * simply have to slow down writers by forcing them to wait on ordered + * extents. + * + * In the case that ordered is larger than delalloc, only include the + * block reserves that we would actually be able to directly reclaim + * from. In this case if we're heavy on metadata operations this will + * clearly be heavy enough to warrant preemptive flushing. In the case + * of heavy DIO or ordered reservations, preemptive flushing will just + * waste time and cause us to slow down. + */ + ordered = percpu_counter_sum_positive(&fs_info->ordered_bytes); + delalloc = percpu_counter_sum_positive(&fs_info->delalloc_bytes); + if (ordered >= delalloc) + used += fs_info->delayed_refs_rsv.reserved + + fs_info->delayed_block_rsv.reserved; + else + used += space_info->bytes_may_use; return (used >= thresh && !btrfs_fs_closing(fs_info) && !test_bit(BTRFS_FS_STATE_REMOUNTING, &fs_info->fs_state)); @@ -922,7 +981,7 @@ static void btrfs_async_reclaim_metadata_space(struct work_struct *work) struct btrfs_fs_info *fs_info; struct btrfs_space_info *space_info; u64 to_reclaim; - int flush_state; + enum btrfs_flush_state flush_state; int commit_cycles = 0; u64 last_tickets_id; @@ -941,7 +1000,7 @@ static void btrfs_async_reclaim_metadata_space(struct work_struct *work) flush_state = FLUSH_DELAYED_ITEMS_NR; do { - flush_space(fs_info, space_info, to_reclaim, flush_state); + flush_space(fs_info, space_info, to_reclaim, flush_state, false); spin_lock(&space_info->lock); if (list_empty(&space_info->tickets)) { space_info->flush = 0; @@ -990,6 +1049,105 @@ static void btrfs_async_reclaim_metadata_space(struct work_struct *work) } /* + * This handles pre-flushing of metadata space before we get to the point that + * we need to start blocking threads on tickets. The logic here is different + * from the other flush paths because it doesn't rely on tickets to tell us how + * much we need to flush, instead it attempts to keep us below the 80% full + * watermark of space by flushing whichever reservation pool is currently the + * largest. + */ +static void btrfs_preempt_reclaim_metadata_space(struct work_struct *work) +{ + struct btrfs_fs_info *fs_info; + struct btrfs_space_info *space_info; + struct btrfs_block_rsv *delayed_block_rsv; + struct btrfs_block_rsv *delayed_refs_rsv; + struct btrfs_block_rsv *global_rsv; + struct btrfs_block_rsv *trans_rsv; + int loops = 0; + + fs_info = container_of(work, struct btrfs_fs_info, + preempt_reclaim_work); + space_info = btrfs_find_space_info(fs_info, BTRFS_BLOCK_GROUP_METADATA); + delayed_block_rsv = &fs_info->delayed_block_rsv; + delayed_refs_rsv = &fs_info->delayed_refs_rsv; + global_rsv = &fs_info->global_block_rsv; + trans_rsv = &fs_info->trans_block_rsv; + + spin_lock(&space_info->lock); + while (need_preemptive_reclaim(fs_info, space_info)) { + enum btrfs_flush_state flush; + u64 delalloc_size = 0; + u64 to_reclaim, block_rsv_size; + u64 global_rsv_size = global_rsv->reserved; + + loops++; + + /* + * We don't have a precise counter for the metadata being + * reserved for delalloc, so we'll approximate it by subtracting + * out the block rsv's space from the bytes_may_use. If that + * amount is higher than the individual reserves, then we can + * assume it's tied up in delalloc reservations. + */ + block_rsv_size = global_rsv_size + + delayed_block_rsv->reserved + + delayed_refs_rsv->reserved + + trans_rsv->reserved; + if (block_rsv_size < space_info->bytes_may_use) + delalloc_size = space_info->bytes_may_use - block_rsv_size; + spin_unlock(&space_info->lock); + + /* + * We don't want to include the global_rsv in our calculation, + * because that's space we can't touch. Subtract it from the + * block_rsv_size for the next checks. + */ + block_rsv_size -= global_rsv_size; + + /* + * We really want to avoid flushing delalloc too much, as it + * could result in poor allocation patterns, so only flush it if + * it's larger than the rest of the pools combined. + */ + if (delalloc_size > block_rsv_size) { + to_reclaim = delalloc_size; + flush = FLUSH_DELALLOC; + } else if (space_info->bytes_pinned > + (delayed_block_rsv->reserved + + delayed_refs_rsv->reserved)) { + to_reclaim = space_info->bytes_pinned; + flush = FORCE_COMMIT_TRANS; + } else if (delayed_block_rsv->reserved > + delayed_refs_rsv->reserved) { + to_reclaim = delayed_block_rsv->reserved; + flush = FLUSH_DELAYED_ITEMS_NR; + } else { + to_reclaim = delayed_refs_rsv->reserved; + flush = FLUSH_DELAYED_REFS_NR; + } + + /* + * We don't want to reclaim everything, just a portion, so scale + * down the to_reclaim by 1/4. If it takes us down to 0, + * reclaim 1 items worth. + */ + to_reclaim >>= 2; + if (!to_reclaim) + to_reclaim = btrfs_calc_insert_metadata_size(fs_info, 1); + flush_space(fs_info, space_info, to_reclaim, flush, true); + cond_resched(); + spin_lock(&space_info->lock); + } + + /* We only went through once, back off our clamping. */ + if (loops == 1 && !space_info->reclaim_size) + space_info->clamp = max(1, space_info->clamp - 1); + trace_btrfs_done_preemptive_reclaim(fs_info, space_info); + spin_unlock(&space_info->lock); +} + +/* * FLUSH_DELALLOC_WAIT: * Space is freed from flushing delalloc in one of two ways. * @@ -1054,7 +1212,7 @@ static void btrfs_async_reclaim_data_space(struct work_struct *work) struct btrfs_fs_info *fs_info; struct btrfs_space_info *space_info; u64 last_tickets_id; - int flush_state = 0; + enum btrfs_flush_state flush_state = 0; fs_info = container_of(work, struct btrfs_fs_info, async_data_reclaim_work); space_info = fs_info->data_sinfo; @@ -1069,7 +1227,7 @@ static void btrfs_async_reclaim_data_space(struct work_struct *work) spin_unlock(&space_info->lock); while (!space_info->full) { - flush_space(fs_info, space_info, U64_MAX, ALLOC_CHUNK_FORCE); + flush_space(fs_info, space_info, U64_MAX, ALLOC_CHUNK_FORCE, false); spin_lock(&space_info->lock); if (list_empty(&space_info->tickets)) { space_info->flush = 0; @@ -1082,7 +1240,7 @@ static void btrfs_async_reclaim_data_space(struct work_struct *work) while (flush_state < ARRAY_SIZE(data_flush_states)) { flush_space(fs_info, space_info, U64_MAX, - data_flush_states[flush_state]); + data_flush_states[flush_state], false); spin_lock(&space_info->lock); if (list_empty(&space_info->tickets)) { space_info->flush = 0; @@ -1115,6 +1273,8 @@ void btrfs_init_async_reclaim_work(struct btrfs_fs_info *fs_info) { INIT_WORK(&fs_info->async_reclaim_work, btrfs_async_reclaim_metadata_space); INIT_WORK(&fs_info->async_data_reclaim_work, btrfs_async_reclaim_data_space); + INIT_WORK(&fs_info->preempt_reclaim_work, + btrfs_preempt_reclaim_metadata_space); } static const enum btrfs_flush_state priority_flush_states[] = { @@ -1153,7 +1313,8 @@ static void priority_reclaim_metadata_space(struct btrfs_fs_info *fs_info, flush_state = 0; do { - flush_space(fs_info, space_info, to_reclaim, states[flush_state]); + flush_space(fs_info, space_info, to_reclaim, states[flush_state], + false); flush_state++; spin_lock(&space_info->lock); if (ticket->bytes == 0) { @@ -1169,7 +1330,7 @@ static void priority_reclaim_data_space(struct btrfs_fs_info *fs_info, struct reserve_ticket *ticket) { while (!space_info->full) { - flush_space(fs_info, space_info, U64_MAX, ALLOC_CHUNK_FORCE); + flush_space(fs_info, space_info, U64_MAX, ALLOC_CHUNK_FORCE, false); spin_lock(&space_info->lock); if (ticket->bytes == 0) { spin_unlock(&space_info->lock); @@ -1214,11 +1375,14 @@ static void wait_reserve_ticket(struct btrfs_fs_info *fs_info, } /** - * handle_reserve_ticket - do the appropriate flushing and waiting for a ticket - * @fs_info - the fs - * @space_info - the space_info for the reservation - * @ticket - the ticket for the reservation - * @flush - how much we can flush + * Do the appropriate flushing and waiting for a ticket + * + * @fs_info: the filesystem + * @space_info: space info for the reservation + * @ticket: ticket for the reservation + * @start_ns: timestamp when the reservation started + * @orig_bytes: amount of bytes originally reserved + * @flush: how much we can flush * * This does the work of figuring out how to flush for the ticket, waiting for * the reservation, and returning the appropriate error if there is one. @@ -1226,6 +1390,7 @@ static void wait_reserve_ticket(struct btrfs_fs_info *fs_info, static int handle_reserve_ticket(struct btrfs_fs_info *fs_info, struct btrfs_space_info *space_info, struct reserve_ticket *ticket, + u64 start_ns, u64 orig_bytes, enum btrfs_reserve_flush_enum flush) { int ret; @@ -1281,6 +1446,8 @@ static int handle_reserve_ticket(struct btrfs_fs_info *fs_info, * space wasn't reserved at all). */ ASSERT(!(ticket->bytes == 0 && ticket->error)); + trace_btrfs_reserve_ticket(fs_info, space_info->flags, orig_bytes, + start_ns, flush, ticket->error); return ret; } @@ -1294,12 +1461,31 @@ static inline bool is_normal_flushing(enum btrfs_reserve_flush_enum flush) (flush == BTRFS_RESERVE_FLUSH_ALL_STEAL); } +static inline void maybe_clamp_preempt(struct btrfs_fs_info *fs_info, + struct btrfs_space_info *space_info) +{ + u64 ordered = percpu_counter_sum_positive(&fs_info->ordered_bytes); + u64 delalloc = percpu_counter_sum_positive(&fs_info->delalloc_bytes); + + /* + * If we're heavy on ordered operations then clamping won't help us. We + * need to clamp specifically to keep up with dirty'ing buffered + * writers, because there's not a 1:1 correlation of writing delalloc + * and freeing space, like there is with flushing delayed refs or + * delayed nodes. If we're already more ordered than delalloc then + * we're keeping up, otherwise we aren't and should probably clamp. + */ + if (ordered < delalloc) + space_info->clamp = min(space_info->clamp + 1, 8); +} + /** - * reserve_metadata_bytes - try to reserve bytes from the block_rsv's space - * @root - the root we're allocating for - * @space_info - the space info we want to allocate from - * @orig_bytes - the number of bytes we want - * @flush - whether or not we can flush to make our reservation + * Try to reserve bytes from the block_rsv's space + * + * @fs_info: the filesystem + * @space_info: space info we want to allocate from + * @orig_bytes: number of bytes we want + * @flush: whether or not we can flush to make our reservation * * This will reserve orig_bytes number of bytes from the space info associated * with the block_rsv. If there is not enough space it will make an attempt to @@ -1314,6 +1500,7 @@ static int __reserve_bytes(struct btrfs_fs_info *fs_info, { struct work_struct *async_work; struct reserve_ticket ticket; + u64 start_ns = 0; u64 used; int ret = 0; bool pending_tickets; @@ -1366,6 +1553,9 @@ static int __reserve_bytes(struct btrfs_fs_info *fs_info, space_info->reclaim_size += ticket.bytes; init_waitqueue_head(&ticket.wait); ticket.steal = (flush == BTRFS_RESERVE_FLUSH_ALL_STEAL); + if (trace_btrfs_reserve_ticket_enabled()) + start_ns = ktime_get_ns(); + if (flush == BTRFS_RESERVE_FLUSH_ALL || flush == BTRFS_RESERVE_FLUSH_ALL_STEAL || flush == BTRFS_RESERVE_FLUSH_DATA) { @@ -1382,6 +1572,14 @@ static int __reserve_bytes(struct btrfs_fs_info *fs_info, list_add_tail(&ticket.list, &space_info->priority_tickets); } + + /* + * We were forced to add a reserve ticket, so our preemptive + * flushing is unable to keep up. Clamp down on the threshold + * for the preemptive flushing in order to keep up with the + * workload. + */ + maybe_clamp_preempt(fs_info, space_info); } else if (!ret && space_info->flags & BTRFS_BLOCK_GROUP_METADATA) { used += orig_bytes; /* @@ -1390,27 +1588,29 @@ static int __reserve_bytes(struct btrfs_fs_info *fs_info, * the async reclaim as we will panic. */ if (!test_bit(BTRFS_FS_LOG_RECOVERING, &fs_info->flags) && - need_do_async_reclaim(fs_info, space_info, used) && - !work_busy(&fs_info->async_reclaim_work)) { + need_preemptive_reclaim(fs_info, space_info) && + !work_busy(&fs_info->preempt_reclaim_work)) { trace_btrfs_trigger_flush(fs_info, space_info->flags, orig_bytes, flush, "preempt"); queue_work(system_unbound_wq, - &fs_info->async_reclaim_work); + &fs_info->preempt_reclaim_work); } } spin_unlock(&space_info->lock); if (!ret || flush == BTRFS_RESERVE_NO_FLUSH) return ret; - return handle_reserve_ticket(fs_info, space_info, &ticket, flush); + return handle_reserve_ticket(fs_info, space_info, &ticket, start_ns, + orig_bytes, flush); } /** - * reserve_metadata_bytes - try to reserve bytes from the block_rsv's space - * @root - the root we're allocating for - * @block_rsv - the block_rsv we're allocating for - * @orig_bytes - the number of bytes we want - * @flush - whether or not we can flush to make our reservation + * Trye to reserve metadata bytes from the block_rsv's space + * + * @root: the root we're allocating for + * @block_rsv: block_rsv we're allocating for + * @orig_bytes: number of bytes we want + * @flush: whether or not we can flush to make our reservation * * This will reserve orig_bytes number of bytes from the space info associated * with the block_rsv. If there is not enough space it will make an attempt to @@ -1448,10 +1648,11 @@ int btrfs_reserve_metadata_bytes(struct btrfs_root *root, } /** - * btrfs_reserve_data_bytes - try to reserve data bytes for an allocation - * @fs_info - the filesystem - * @bytes - the number of bytes we need - * @flush - how we are allowed to flush + * Try to reserve data bytes for an allocation + * + * @fs_info: the filesystem + * @bytes: number of bytes we need + * @flush: how we are allowed to flush * * This will reserve bytes from the data space info. If there is not enough * space then we will attempt to flush space as specified by flush. diff --git a/fs/btrfs/space-info.h b/fs/btrfs/space-info.h index 5646393b928c..b1a8ffb03b3e 100644 --- a/fs/btrfs/space-info.h +++ b/fs/btrfs/space-info.h @@ -17,11 +17,17 @@ struct btrfs_space_info { u64 bytes_may_use; /* number of bytes that may be used for delalloc/allocations */ u64 bytes_readonly; /* total bytes that are read only */ + u64 bytes_zone_unusable; /* total bytes that are unusable until + resetting the device zone */ u64 max_extent_size; /* This will hold the maximum extent size of the space info if we had an ENOSPC in the allocator. */ + int clamp; /* Used to scale our threshold for preemptive + flushing. The value is >> clamp, so turns + out to be a 2^clamp divisor. */ + unsigned int full:1; /* indicates that we cannot allocate any more chunks for this space */ unsigned int chunk_alloc:1; /* set if we are allocating a chunk */ @@ -119,7 +125,7 @@ DECLARE_SPACE_INFO_UPDATE(bytes_pinned, "pinned"); int btrfs_init_space_info(struct btrfs_fs_info *fs_info); void btrfs_update_space_info(struct btrfs_fs_info *info, u64 flags, u64 total_bytes, u64 bytes_used, - u64 bytes_readonly, + u64 bytes_readonly, u64 bytes_zone_unusable, struct btrfs_space_info **space_info); struct btrfs_space_info *btrfs_find_space_info(struct btrfs_fs_info *info, u64 flags); @@ -152,4 +158,21 @@ static inline void btrfs_space_info_free_bytes_may_use( int btrfs_reserve_data_bytes(struct btrfs_fs_info *fs_info, u64 bytes, enum btrfs_reserve_flush_enum flush); +static inline void __btrfs_mod_total_bytes_pinned( + struct btrfs_space_info *space_info, + s64 mod) +{ + percpu_counter_add_batch(&space_info->total_bytes_pinned, mod, + BTRFS_TOTAL_BYTES_PINNED_BATCH); +} + +static inline void btrfs_mod_total_bytes_pinned(struct btrfs_fs_info *fs_info, + u64 flags, s64 mod) +{ + struct btrfs_space_info *space_info = btrfs_find_space_info(fs_info, flags); + + ASSERT(space_info); + __btrfs_mod_total_bytes_pinned(space_info, mod); +} + #endif /* BTRFS_SPACE_INFO_H */ diff --git a/fs/btrfs/subpage.c b/fs/btrfs/subpage.c new file mode 100644 index 000000000000..c69049e7daa9 --- /dev/null +++ b/fs/btrfs/subpage.c @@ -0,0 +1,278 @@ +// SPDX-License-Identifier: GPL-2.0 + +#include <linux/slab.h> +#include "ctree.h" +#include "subpage.h" + +int btrfs_attach_subpage(const struct btrfs_fs_info *fs_info, + struct page *page, enum btrfs_subpage_type type) +{ + struct btrfs_subpage *subpage = NULL; + int ret; + + /* + * We have cases like a dummy extent buffer page, which is not mappped + * and doesn't need to be locked. + */ + if (page->mapping) + ASSERT(PageLocked(page)); + /* Either not subpage, or the page already has private attached */ + if (fs_info->sectorsize == PAGE_SIZE || PagePrivate(page)) + return 0; + + ret = btrfs_alloc_subpage(fs_info, &subpage, type); + if (ret < 0) + return ret; + attach_page_private(page, subpage); + return 0; +} + +void btrfs_detach_subpage(const struct btrfs_fs_info *fs_info, + struct page *page) +{ + struct btrfs_subpage *subpage; + + /* Either not subpage, or already detached */ + if (fs_info->sectorsize == PAGE_SIZE || !PagePrivate(page)) + return; + + subpage = (struct btrfs_subpage *)detach_page_private(page); + ASSERT(subpage); + btrfs_free_subpage(subpage); +} + +int btrfs_alloc_subpage(const struct btrfs_fs_info *fs_info, + struct btrfs_subpage **ret, + enum btrfs_subpage_type type) +{ + if (fs_info->sectorsize == PAGE_SIZE) + return 0; + + *ret = kzalloc(sizeof(struct btrfs_subpage), GFP_NOFS); + if (!*ret) + return -ENOMEM; + spin_lock_init(&(*ret)->lock); + if (type == BTRFS_SUBPAGE_METADATA) + atomic_set(&(*ret)->eb_refs, 0); + else + atomic_set(&(*ret)->readers, 0); + return 0; +} + +void btrfs_free_subpage(struct btrfs_subpage *subpage) +{ + kfree(subpage); +} + +/* + * Increase the eb_refs of current subpage. + * + * This is important for eb allocation, to prevent race with last eb freeing + * of the same page. + * With the eb_refs increased before the eb inserted into radix tree, + * detach_extent_buffer_page() won't detach the page private while we're still + * allocating the extent buffer. + */ +void btrfs_page_inc_eb_refs(const struct btrfs_fs_info *fs_info, + struct page *page) +{ + struct btrfs_subpage *subpage; + + if (fs_info->sectorsize == PAGE_SIZE) + return; + + ASSERT(PagePrivate(page) && page->mapping); + lockdep_assert_held(&page->mapping->private_lock); + + subpage = (struct btrfs_subpage *)page->private; + atomic_inc(&subpage->eb_refs); +} + +void btrfs_page_dec_eb_refs(const struct btrfs_fs_info *fs_info, + struct page *page) +{ + struct btrfs_subpage *subpage; + + if (fs_info->sectorsize == PAGE_SIZE) + return; + + ASSERT(PagePrivate(page) && page->mapping); + lockdep_assert_held(&page->mapping->private_lock); + + subpage = (struct btrfs_subpage *)page->private; + ASSERT(atomic_read(&subpage->eb_refs)); + atomic_dec(&subpage->eb_refs); +} + +static void btrfs_subpage_assert(const struct btrfs_fs_info *fs_info, + struct page *page, u64 start, u32 len) +{ + /* Basic checks */ + ASSERT(PagePrivate(page) && page->private); + ASSERT(IS_ALIGNED(start, fs_info->sectorsize) && + IS_ALIGNED(len, fs_info->sectorsize)); + /* + * The range check only works for mapped page, we can still have + * unmapped page like dummy extent buffer pages. + */ + if (page->mapping) + ASSERT(page_offset(page) <= start && + start + len <= page_offset(page) + PAGE_SIZE); +} + +void btrfs_subpage_start_reader(const struct btrfs_fs_info *fs_info, + struct page *page, u64 start, u32 len) +{ + struct btrfs_subpage *subpage = (struct btrfs_subpage *)page->private; + const int nbits = len >> fs_info->sectorsize_bits; + int ret; + + btrfs_subpage_assert(fs_info, page, start, len); + + ret = atomic_add_return(nbits, &subpage->readers); + ASSERT(ret == nbits); +} + +void btrfs_subpage_end_reader(const struct btrfs_fs_info *fs_info, + struct page *page, u64 start, u32 len) +{ + struct btrfs_subpage *subpage = (struct btrfs_subpage *)page->private; + const int nbits = len >> fs_info->sectorsize_bits; + + btrfs_subpage_assert(fs_info, page, start, len); + ASSERT(atomic_read(&subpage->readers) >= nbits); + if (atomic_sub_and_test(nbits, &subpage->readers)) + unlock_page(page); +} + +/* + * Convert the [start, start + len) range into a u16 bitmap + * + * For example: if start == page_offset() + 16K, len = 16K, we get 0x00f0. + */ +static u16 btrfs_subpage_calc_bitmap(const struct btrfs_fs_info *fs_info, + struct page *page, u64 start, u32 len) +{ + const int bit_start = offset_in_page(start) >> fs_info->sectorsize_bits; + const int nbits = len >> fs_info->sectorsize_bits; + + btrfs_subpage_assert(fs_info, page, start, len); + + /* + * Here nbits can be 16, thus can go beyond u16 range. We make the + * first left shift to be calculate in unsigned long (at least u32), + * then truncate the result to u16. + */ + return (u16)(((1UL << nbits) - 1) << bit_start); +} + +void btrfs_subpage_set_uptodate(const struct btrfs_fs_info *fs_info, + struct page *page, u64 start, u32 len) +{ + struct btrfs_subpage *subpage = (struct btrfs_subpage *)page->private; + const u16 tmp = btrfs_subpage_calc_bitmap(fs_info, page, start, len); + unsigned long flags; + + spin_lock_irqsave(&subpage->lock, flags); + subpage->uptodate_bitmap |= tmp; + if (subpage->uptodate_bitmap == U16_MAX) + SetPageUptodate(page); + spin_unlock_irqrestore(&subpage->lock, flags); +} + +void btrfs_subpage_clear_uptodate(const struct btrfs_fs_info *fs_info, + struct page *page, u64 start, u32 len) +{ + struct btrfs_subpage *subpage = (struct btrfs_subpage *)page->private; + const u16 tmp = btrfs_subpage_calc_bitmap(fs_info, page, start, len); + unsigned long flags; + + spin_lock_irqsave(&subpage->lock, flags); + subpage->uptodate_bitmap &= ~tmp; + ClearPageUptodate(page); + spin_unlock_irqrestore(&subpage->lock, flags); +} + +void btrfs_subpage_set_error(const struct btrfs_fs_info *fs_info, + struct page *page, u64 start, u32 len) +{ + struct btrfs_subpage *subpage = (struct btrfs_subpage *)page->private; + const u16 tmp = btrfs_subpage_calc_bitmap(fs_info, page, start, len); + unsigned long flags; + + spin_lock_irqsave(&subpage->lock, flags); + subpage->error_bitmap |= tmp; + SetPageError(page); + spin_unlock_irqrestore(&subpage->lock, flags); +} + +void btrfs_subpage_clear_error(const struct btrfs_fs_info *fs_info, + struct page *page, u64 start, u32 len) +{ + struct btrfs_subpage *subpage = (struct btrfs_subpage *)page->private; + const u16 tmp = btrfs_subpage_calc_bitmap(fs_info, page, start, len); + unsigned long flags; + + spin_lock_irqsave(&subpage->lock, flags); + subpage->error_bitmap &= ~tmp; + if (subpage->error_bitmap == 0) + ClearPageError(page); + spin_unlock_irqrestore(&subpage->lock, flags); +} + +/* + * Unlike set/clear which is dependent on each page status, for test all bits + * are tested in the same way. + */ +#define IMPLEMENT_BTRFS_SUBPAGE_TEST_OP(name) \ +bool btrfs_subpage_test_##name(const struct btrfs_fs_info *fs_info, \ + struct page *page, u64 start, u32 len) \ +{ \ + struct btrfs_subpage *subpage = (struct btrfs_subpage *)page->private; \ + const u16 tmp = btrfs_subpage_calc_bitmap(fs_info, page, start, len); \ + unsigned long flags; \ + bool ret; \ + \ + spin_lock_irqsave(&subpage->lock, flags); \ + ret = ((subpage->name##_bitmap & tmp) == tmp); \ + spin_unlock_irqrestore(&subpage->lock, flags); \ + return ret; \ +} +IMPLEMENT_BTRFS_SUBPAGE_TEST_OP(uptodate); +IMPLEMENT_BTRFS_SUBPAGE_TEST_OP(error); + +/* + * Note that, in selftests (extent-io-tests), we can have empty fs_info passed + * in. We only test sectorsize == PAGE_SIZE cases so far, thus we can fall + * back to regular sectorsize branch. + */ +#define IMPLEMENT_BTRFS_PAGE_OPS(name, set_page_func, clear_page_func, \ + test_page_func) \ +void btrfs_page_set_##name(const struct btrfs_fs_info *fs_info, \ + struct page *page, u64 start, u32 len) \ +{ \ + if (unlikely(!fs_info) || fs_info->sectorsize == PAGE_SIZE) { \ + set_page_func(page); \ + return; \ + } \ + btrfs_subpage_set_##name(fs_info, page, start, len); \ +} \ +void btrfs_page_clear_##name(const struct btrfs_fs_info *fs_info, \ + struct page *page, u64 start, u32 len) \ +{ \ + if (unlikely(!fs_info) || fs_info->sectorsize == PAGE_SIZE) { \ + clear_page_func(page); \ + return; \ + } \ + btrfs_subpage_clear_##name(fs_info, page, start, len); \ +} \ +bool btrfs_page_test_##name(const struct btrfs_fs_info *fs_info, \ + struct page *page, u64 start, u32 len) \ +{ \ + if (unlikely(!fs_info) || fs_info->sectorsize == PAGE_SIZE) \ + return test_page_func(page); \ + return btrfs_subpage_test_##name(fs_info, page, start, len); \ +} +IMPLEMENT_BTRFS_PAGE_OPS(uptodate, SetPageUptodate, ClearPageUptodate, + PageUptodate); +IMPLEMENT_BTRFS_PAGE_OPS(error, SetPageError, ClearPageError, PageError); diff --git a/fs/btrfs/subpage.h b/fs/btrfs/subpage.h new file mode 100644 index 000000000000..b86a4881475d --- /dev/null +++ b/fs/btrfs/subpage.h @@ -0,0 +1,91 @@ +/* SPDX-License-Identifier: GPL-2.0 */ + +#ifndef BTRFS_SUBPAGE_H +#define BTRFS_SUBPAGE_H + +#include <linux/spinlock.h> + +/* + * Maximum page size we support is 64K, minimum sector size is 4K, u16 bitmap + * is sufficient. Regular bitmap_* is not used due to size reasons. + */ +#define BTRFS_SUBPAGE_BITMAP_SIZE 16 + +/* + * Structure to trace status of each sector inside a page, attached to + * page::private for both data and metadata inodes. + */ +struct btrfs_subpage { + /* Common members for both data and metadata pages */ + spinlock_t lock; + u16 uptodate_bitmap; + u16 error_bitmap; + union { + /* + * Structures only used by metadata + * + * @eb_refs should only be operated under private_lock, as it + * manages whether the subpage can be detached. + */ + atomic_t eb_refs; + /* Structures only used by data */ + struct { + atomic_t readers; + }; + }; +}; + +enum btrfs_subpage_type { + BTRFS_SUBPAGE_METADATA, + BTRFS_SUBPAGE_DATA, +}; + +int btrfs_attach_subpage(const struct btrfs_fs_info *fs_info, + struct page *page, enum btrfs_subpage_type type); +void btrfs_detach_subpage(const struct btrfs_fs_info *fs_info, + struct page *page); + +/* Allocate additional data where page represents more than one sector */ +int btrfs_alloc_subpage(const struct btrfs_fs_info *fs_info, + struct btrfs_subpage **ret, + enum btrfs_subpage_type type); +void btrfs_free_subpage(struct btrfs_subpage *subpage); + +void btrfs_page_inc_eb_refs(const struct btrfs_fs_info *fs_info, + struct page *page); +void btrfs_page_dec_eb_refs(const struct btrfs_fs_info *fs_info, + struct page *page); + +void btrfs_subpage_start_reader(const struct btrfs_fs_info *fs_info, + struct page *page, u64 start, u32 len); +void btrfs_subpage_end_reader(const struct btrfs_fs_info *fs_info, + struct page *page, u64 start, u32 len); + +/* + * Template for subpage related operations. + * + * btrfs_subpage_*() are for call sites where the page has subpage attached and + * the range is ensured to be inside the page. + * + * btrfs_page_*() are for call sites where the page can either be subpage + * specific or regular page. The function will handle both cases. + * But the range still needs to be inside the page. + */ +#define DECLARE_BTRFS_SUBPAGE_OPS(name) \ +void btrfs_subpage_set_##name(const struct btrfs_fs_info *fs_info, \ + struct page *page, u64 start, u32 len); \ +void btrfs_subpage_clear_##name(const struct btrfs_fs_info *fs_info, \ + struct page *page, u64 start, u32 len); \ +bool btrfs_subpage_test_##name(const struct btrfs_fs_info *fs_info, \ + struct page *page, u64 start, u32 len); \ +void btrfs_page_set_##name(const struct btrfs_fs_info *fs_info, \ + struct page *page, u64 start, u32 len); \ +void btrfs_page_clear_##name(const struct btrfs_fs_info *fs_info, \ + struct page *page, u64 start, u32 len); \ +bool btrfs_page_test_##name(const struct btrfs_fs_info *fs_info, \ + struct page *page, u64 start, u32 len); + +DECLARE_BTRFS_SUBPAGE_OPS(uptodate); +DECLARE_BTRFS_SUBPAGE_OPS(error); + +#endif diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index 12d7d3be7cd4..f7a4ad86adee 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -48,7 +48,6 @@ #include "tests/btrfs-tests.h" #include "block-group.h" #include "discard.h" - #include "qgroup.h" #define CREATE_TRACE_POINTS #include <trace/events/btrfs.h> @@ -1919,8 +1918,8 @@ static int btrfs_remount(struct super_block *sb, int *flags, char *data) btrfs_resize_thread_pool(fs_info, fs_info->thread_pool_size, old_thread_pool_size); - if (btrfs_test_opt(fs_info, FREE_SPACE_TREE) != - btrfs_fs_compat_ro(fs_info, FREE_SPACE_TREE) && + if ((bool)btrfs_test_opt(fs_info, FREE_SPACE_TREE) != + (bool)btrfs_fs_compat_ro(fs_info, FREE_SPACE_TREE) && (!sb_rdonly(sb) || (*flags & SB_RDONLY))) { btrfs_warn(fs_info, "remount supports changing free space tree only from ro to rw"); @@ -2028,6 +2027,13 @@ static int btrfs_remount(struct super_block *sb, int *flags, char *data) ret = -EINVAL; goto restore; } + if (fs_info->sectorsize < PAGE_SIZE) { + btrfs_warn(fs_info, + "read-write mount is not yet allowed for sectorsize %u page size %lu", + fs_info->sectorsize, PAGE_SIZE); + ret = -EINVAL; + goto restore; + } /* * NOTE: when remounting with a change that does writes, don't diff --git a/fs/btrfs/sysfs.c b/fs/btrfs/sysfs.c index 19b9fffa2c9c..6eb1c50fa98c 100644 --- a/fs/btrfs/sysfs.c +++ b/fs/btrfs/sysfs.c @@ -666,6 +666,7 @@ SPACE_INFO_ATTR(bytes_pinned); SPACE_INFO_ATTR(bytes_reserved); SPACE_INFO_ATTR(bytes_may_use); SPACE_INFO_ATTR(bytes_readonly); +SPACE_INFO_ATTR(bytes_zone_unusable); SPACE_INFO_ATTR(disk_used); SPACE_INFO_ATTR(disk_total); BTRFS_ATTR(space_info, total_bytes_pinned, @@ -679,6 +680,7 @@ static struct attribute *space_info_attrs[] = { BTRFS_ATTR_PTR(space_info, bytes_reserved), BTRFS_ATTR_PTR(space_info, bytes_may_use), BTRFS_ATTR_PTR(space_info, bytes_readonly), + BTRFS_ATTR_PTR(space_info, bytes_zone_unusable), BTRFS_ATTR_PTR(space_info, disk_used), BTRFS_ATTR_PTR(space_info, disk_total), BTRFS_ATTR_PTR(space_info, total_bytes_pinned), diff --git a/fs/btrfs/tests/btrfs-tests.c b/fs/btrfs/tests/btrfs-tests.c index 6bd97bd4cb37..3a4099a2bf05 100644 --- a/fs/btrfs/tests/btrfs-tests.c +++ b/fs/btrfs/tests/btrfs-tests.c @@ -62,7 +62,7 @@ struct inode *btrfs_new_test_inode(void) BTRFS_I(inode)->location.type = BTRFS_INODE_ITEM_KEY; BTRFS_I(inode)->location.objectid = BTRFS_FIRST_FREE_OBJECTID; BTRFS_I(inode)->location.offset = 0; - inode_init_owner(inode, NULL, S_IFREG); + inode_init_owner(&init_user_ns, inode, NULL, S_IFREG); return inode; } diff --git a/fs/btrfs/tests/extent-map-tests.c b/fs/btrfs/tests/extent-map-tests.c index 57379e96ccc9..c0aefe6dee0b 100644 --- a/fs/btrfs/tests/extent-map-tests.c +++ b/fs/btrfs/tests/extent-map-tests.c @@ -507,7 +507,7 @@ static int test_rmap_block(struct btrfs_fs_info *fs_info, goto out_free; } - ret = btrfs_rmap_block(fs_info, em->start, btrfs_sb_offset(1), + ret = btrfs_rmap_block(fs_info, em->start, NULL, btrfs_sb_offset(1), &logical, &out_ndaddrs, &out_stripe_len); if (ret || (out_ndaddrs == 0 && test->expected_mapped_addr)) { test_err("didn't rmap anything but expected %d", diff --git a/fs/btrfs/transaction.c b/fs/btrfs/transaction.c index 6af7f2bf92de..acff6bb49a97 100644 --- a/fs/btrfs/transaction.c +++ b/fs/btrfs/transaction.c @@ -21,6 +21,7 @@ #include "qgroup.h" #include "block-group.h" #include "space-info.h" +#include "zoned.h" #define BTRFS_ROOT_TRANS_TAG 0 @@ -107,6 +108,11 @@ static const unsigned int btrfs_blocked_trans_types[TRANS_STATE_MAX] = { __TRANS_JOIN | __TRANS_JOIN_NOLOCK | __TRANS_JOIN_NOSTART), + [TRANS_STATE_SUPER_COMMITTED] = (__TRANS_START | + __TRANS_ATTACH | + __TRANS_JOIN | + __TRANS_JOIN_NOLOCK | + __TRANS_JOIN_NOSTART), [TRANS_STATE_COMPLETED] = (__TRANS_START | __TRANS_ATTACH | __TRANS_JOIN | @@ -375,6 +381,8 @@ loop: spin_lock_init(&cur_trans->dirty_bgs_lock); INIT_LIST_HEAD(&cur_trans->deleted_bgs); spin_lock_init(&cur_trans->dropped_roots_lock); + INIT_LIST_HEAD(&cur_trans->releasing_ebs); + spin_lock_init(&cur_trans->releasing_ebs_lock); list_add_tail(&cur_trans->list, &fs_info->trans_list); extent_io_tree_init(fs_info, &cur_trans->dirty_pages, IO_TREE_TRANS_DIRTY_PAGES, fs_info->btree_inode); @@ -826,10 +834,11 @@ btrfs_attach_transaction_barrier(struct btrfs_root *root) return trans; } -/* wait for a transaction commit to be fully complete */ -static noinline void wait_for_commit(struct btrfs_transaction *commit) +/* Wait for a transaction commit to reach at least the given state. */ +static noinline void wait_for_commit(struct btrfs_transaction *commit, + const enum btrfs_trans_state min_state) { - wait_event(commit->commit_wait, commit->state == TRANS_STATE_COMPLETED); + wait_event(commit->commit_wait, commit->state >= min_state); } int btrfs_wait_for_commit(struct btrfs_fs_info *fs_info, u64 transid) @@ -884,7 +893,7 @@ int btrfs_wait_for_commit(struct btrfs_fs_info *fs_info, u64 transid) goto out; /* nothing committing|committed */ } - wait_for_commit(cur_trans); + wait_for_commit(cur_trans, TRANS_STATE_COMPLETED); btrfs_put_transaction(cur_trans); out: return ret; @@ -909,9 +918,8 @@ bool btrfs_should_end_transaction(struct btrfs_trans_handle *trans) { struct btrfs_transaction *cur_trans = trans->transaction; - smp_mb(); if (cur_trans->state >= TRANS_STATE_COMMIT_START || - cur_trans->delayed_refs.flushing) + test_bit(BTRFS_DELAYED_REFS_FLUSHING, &cur_trans->delayed_refs.flags)) return true; return should_end_transaction(trans); @@ -1230,10 +1238,6 @@ static noinline int commit_cowonly_roots(struct btrfs_trans_handle *trans) if (ret) return ret; - ret = btrfs_run_delayed_refs(trans, (unsigned long)-1); - if (ret) - return ret; - ret = btrfs_run_dev_stats(trans); if (ret) return ret; @@ -1248,10 +1252,6 @@ static noinline int commit_cowonly_roots(struct btrfs_trans_handle *trans) if (ret) return ret; - /* run_qgroups might have added some more refs */ - ret = btrfs_run_delayed_refs(trans, (unsigned long)-1); - if (ret) - return ret; again: while (!list_empty(&fs_info->dirty_cowonly_roots)) { struct btrfs_root *root; @@ -1266,15 +1266,24 @@ again: ret = update_cowonly_root(trans, root); if (ret) return ret; - ret = btrfs_run_delayed_refs(trans, (unsigned long)-1); - if (ret) - return ret; } + /* Now flush any delayed refs generated by updating all of the roots */ + ret = btrfs_run_delayed_refs(trans, (unsigned long)-1); + if (ret) + return ret; + while (!list_empty(dirty_bgs) || !list_empty(io_bgs)) { ret = btrfs_write_dirty_block_groups(trans); if (ret) return ret; + + /* + * We're writing the dirty block groups, which could generate + * delayed refs, which could generate more dirty block groups, + * so we want to keep this flushing in this loop to make sure + * everything gets run. + */ ret = btrfs_run_delayed_refs(trans, (unsigned long)-1); if (ret) return ret; @@ -1319,7 +1328,6 @@ static noinline int commit_fs_roots(struct btrfs_trans_handle *trans) struct btrfs_root *gang[8]; int i; int ret; - int err = 0; spin_lock(&fs_info->fs_roots_radix_lock); while (1) { @@ -1331,6 +1339,8 @@ static noinline int commit_fs_roots(struct btrfs_trans_handle *trans) break; for (i = 0; i < ret; i++) { struct btrfs_root *root = gang[i]; + int ret2; + radix_tree_tag_clear(&fs_info->fs_roots_radix, (unsigned long)root->root_key.objectid, BTRFS_ROOT_TRANS_TAG); @@ -1350,17 +1360,17 @@ static noinline int commit_fs_roots(struct btrfs_trans_handle *trans) root->node); } - err = btrfs_update_root(trans, fs_info->tree_root, + ret2 = btrfs_update_root(trans, fs_info->tree_root, &root->root_key, &root->root_item); + if (ret2) + return ret2; spin_lock(&fs_info->fs_roots_radix_lock); - if (err) - break; btrfs_qgroup_free_meta_all_pertrans(root); } } spin_unlock(&fs_info->fs_roots_radix_lock); - return err; + return 0; } /* @@ -1433,6 +1443,23 @@ static int qgroup_account_snapshot(struct btrfs_trans_handle *trans, record_root_in_trans(trans, src, 1); /* + * btrfs_qgroup_inherit relies on a consistent view of the usage for the + * src root, so we must run the delayed refs here. + * + * However this isn't particularly fool proof, because there's no + * synchronization keeping us from changing the tree after this point + * before we do the qgroup_inherit, or even from making changes while + * we're doing the qgroup_inherit. But that's a problem for the future, + * for now flush the delayed refs to narrow the race window where the + * qgroup counters could end up wrong. + */ + ret = btrfs_run_delayed_refs(trans, (unsigned long)-1); + if (ret) { + btrfs_abort_transaction(trans, ret); + goto out; + } + + /* * We are going to commit transaction, see btrfs_commit_transaction() * comment for reason locking tree_log_mutex */ @@ -1525,7 +1552,7 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans, ASSERT(pending->root_item); new_root_item = pending->root_item; - pending->error = btrfs_find_free_objectid(tree_root, &objectid); + pending->error = btrfs_get_free_objectid(tree_root, &objectid); if (pending->error) goto no_free_objectid; @@ -1685,12 +1712,6 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans, goto fail; } - ret = btrfs_run_delayed_refs(trans, (unsigned long)-1); - if (ret) { - btrfs_abort_transaction(trans, ret); - goto fail; - } - /* * Do special qgroup accounting for snapshot, as we do some qgroup * snapshot hack to do fast snapshot. @@ -1738,12 +1759,6 @@ static noinline int create_pending_snapshot(struct btrfs_trans_handle *trans, } } - ret = btrfs_run_delayed_refs(trans, (unsigned long)-1); - if (ret) { - btrfs_abort_transaction(trans, ret); - goto fail; - } - fail: pending->error = ret; dir_item_existed: @@ -2042,32 +2057,25 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans) btrfs_trans_release_metadata(trans); trans->block_rsv = NULL; - /* make a pass through all the delayed refs we have so far - * any runnings procs may add more while we are here - */ - ret = btrfs_run_delayed_refs(trans, 0); - if (ret) { - btrfs_end_transaction(trans); - return ret; - } - - cur_trans = trans->transaction; - /* - * set the flushing flag so procs in this transaction have to - * start sending their work down. + * We only want one transaction commit doing the flushing so we do not + * waste a bunch of time on lock contention on the extent root node. */ - cur_trans->delayed_refs.flushing = 1; - smp_wmb(); + if (!test_and_set_bit(BTRFS_DELAYED_REFS_FLUSHING, + &cur_trans->delayed_refs.flags)) { + /* + * Make a pass through all the delayed refs we have so far. + * Any running threads may add more while we are here. + */ + ret = btrfs_run_delayed_refs(trans, 0); + if (ret) { + btrfs_end_transaction(trans); + return ret; + } + } btrfs_create_pending_block_groups(trans); - ret = btrfs_run_delayed_refs(trans, 0); - if (ret) { - btrfs_end_transaction(trans); - return ret; - } - if (!test_bit(BTRFS_TRANS_DIRTY_BG_RUN, &cur_trans->flags)) { int run_it = 0; @@ -2101,11 +2109,15 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans) spin_lock(&fs_info->trans_lock); if (cur_trans->state >= TRANS_STATE_COMMIT_START) { + enum btrfs_trans_state want_state = TRANS_STATE_COMPLETED; + spin_unlock(&fs_info->trans_lock); refcount_inc(&cur_trans->use_count); - ret = btrfs_end_transaction(trans); - wait_for_commit(cur_trans); + if (trans->in_fsync) + want_state = TRANS_STATE_SUPER_COMMITTED; + ret = btrfs_end_transaction(trans); + wait_for_commit(cur_trans, want_state); if (TRANS_ABORTED(cur_trans)) ret = cur_trans->aborted; @@ -2119,13 +2131,19 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans) wake_up(&fs_info->transaction_blocked_wait); if (cur_trans->list.prev != &fs_info->trans_list) { + enum btrfs_trans_state want_state = TRANS_STATE_COMPLETED; + + if (trans->in_fsync) + want_state = TRANS_STATE_SUPER_COMMITTED; + prev_trans = list_entry(cur_trans->list.prev, struct btrfs_transaction, list); - if (prev_trans->state != TRANS_STATE_COMPLETED) { + if (prev_trans->state < want_state) { refcount_inc(&prev_trans->use_count); spin_unlock(&fs_info->trans_lock); - wait_for_commit(prev_trans); + wait_for_commit(prev_trans, want_state); + ret = READ_ONCE(prev_trans->aborted); btrfs_put_transaction(prev_trans); @@ -2335,6 +2353,13 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans) goto scrub_continue; } + /* + * At this point, we should have written all the tree blocks allocated + * in this transaction. So it's now safe to free the redirtyied extent + * buffers. + */ + btrfs_free_redirty_list(cur_trans); + ret = write_all_supers(fs_info, 0); /* * the super is written, we can safely allow the tree-loggers @@ -2344,6 +2369,13 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans) if (ret) goto scrub_continue; + /* + * We needn't acquire the lock here because there is no other task + * which can change it. + */ + cur_trans->state = TRANS_STATE_SUPER_COMMITTED; + wake_up(&cur_trans->commit_wait); + btrfs_finish_extent_commit(trans); if (test_bit(BTRFS_TRANS_HAVE_FREE_BGS, &cur_trans->flags)) diff --git a/fs/btrfs/transaction.h b/fs/btrfs/transaction.h index 31ca81bad822..6335716e513f 100644 --- a/fs/btrfs/transaction.h +++ b/fs/btrfs/transaction.h @@ -16,6 +16,7 @@ enum btrfs_trans_state { TRANS_STATE_COMMIT_START, TRANS_STATE_COMMIT_DOING, TRANS_STATE_UNBLOCKED, + TRANS_STATE_SUPER_COMMITTED, TRANS_STATE_COMPLETED, TRANS_STATE_MAX, }; @@ -92,6 +93,9 @@ struct btrfs_transaction { */ atomic_t pending_ordered; wait_queue_head_t pending_wait; + + spinlock_t releasing_ebs_lock; + struct list_head releasing_ebs; }; #define __TRANS_FREEZABLE (1U << 0) @@ -133,6 +137,7 @@ struct btrfs_trans_handle { bool can_flush_pending_bgs; bool reloc_reserved; bool dirty; + bool in_fsync; struct btrfs_root *root; struct btrfs_fs_info *fs_info; struct list_head new_bgs; diff --git a/fs/btrfs/tree-checker.c b/fs/btrfs/tree-checker.c index 582061c7b547..f4ade821307d 100644 --- a/fs/btrfs/tree-checker.c +++ b/fs/btrfs/tree-checker.c @@ -1453,22 +1453,14 @@ static int check_extent_data_ref(struct extent_buffer *leaf, return -EUCLEAN; } for (; ptr < end; ptr += sizeof(*dref)) { - u64 root_objectid; - u64 owner; u64 offset; - u64 hash; + /* + * We cannot check the extent_data_ref hash due to possible + * overflow from the leaf due to hash collisions. + */ dref = (struct btrfs_extent_data_ref *)ptr; - root_objectid = btrfs_extent_data_ref_root(leaf, dref); - owner = btrfs_extent_data_ref_objectid(leaf, dref); offset = btrfs_extent_data_ref_offset(leaf, dref); - hash = hash_extent_data_ref(root_objectid, owner, offset); - if (unlikely(hash != key->offset)) { - extent_err(leaf, slot, - "invalid extent data ref hash, item has 0x%016llx key has 0x%016llx", - hash, key->offset); - return -EUCLEAN; - } if (unlikely(!IS_ALIGNED(offset, leaf->fs_info->sectorsize))) { extent_err(leaf, slot, "invalid extent data backref offset, have %llu expect aligned to %u", diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index 254c2ee43aae..92a368627791 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -19,6 +19,7 @@ #include "qgroup.h" #include "block-group.h" #include "space-info.h" +#include "zoned.h" /* magic values for the inode_only field in btrfs_log_inode: * @@ -104,6 +105,7 @@ static noinline int replay_dir_deletes(struct btrfs_trans_handle *trans, struct btrfs_root *log, struct btrfs_path *path, u64 dirid, int del_all); +static void wait_log_commit(struct btrfs_root *root, int transid); /* * tree logging is a special write ahead log used to make sure that @@ -139,7 +141,9 @@ static int start_log_trans(struct btrfs_trans_handle *trans, { struct btrfs_fs_info *fs_info = root->fs_info; struct btrfs_root *tree_root = fs_info->tree_root; + const bool zoned = btrfs_is_zoned(fs_info); int ret = 0; + bool created = false; /* * First check if the log root tree was already created. If not, create @@ -149,8 +153,10 @@ static int start_log_trans(struct btrfs_trans_handle *trans, mutex_lock(&tree_root->log_mutex); if (!fs_info->log_root_tree) { ret = btrfs_init_log_root_tree(trans, fs_info); - if (!ret) + if (!ret) { set_bit(BTRFS_ROOT_HAS_LOG_TREE, &tree_root->state); + created = true; + } } mutex_unlock(&tree_root->log_mutex); if (ret) @@ -159,12 +165,20 @@ static int start_log_trans(struct btrfs_trans_handle *trans, mutex_lock(&root->log_mutex); +again: if (root->log_root) { + int index = (root->log_transid + 1) % 2; + if (btrfs_need_log_full_commit(trans)) { ret = -EAGAIN; goto out; } + if (zoned && atomic_read(&root->log_commit[index])) { + wait_log_commit(root, root->log_transid - 1); + goto again; + } + if (!root->log_start_pid) { clear_bit(BTRFS_ROOT_MULTI_LOG_TASKS, &root->state); root->log_start_pid = current->pid; @@ -172,6 +186,17 @@ static int start_log_trans(struct btrfs_trans_handle *trans, set_bit(BTRFS_ROOT_MULTI_LOG_TASKS, &root->state); } } else { + /* + * This means fs_info->log_root_tree was already created + * for some other FS trees. Do the full commit not to mix + * nodes from multiple log transactions to do sequential + * writing. + */ + if (zoned && !created) { + ret = -EAGAIN; + goto out; + } + ret = btrfs_add_log_tree(trans, root); if (ret) goto out; @@ -200,14 +225,22 @@ out: */ static int join_running_log_trans(struct btrfs_root *root) { + const bool zoned = btrfs_is_zoned(root->fs_info); int ret = -ENOENT; if (!test_bit(BTRFS_ROOT_HAS_LOG_TREE, &root->state)) return ret; mutex_lock(&root->log_mutex); +again: if (root->log_root) { + int index = (root->log_transid + 1) % 2; + ret = 0; + if (zoned && atomic_read(&root->log_commit[index])) { + wait_log_commit(root, root->log_transid - 1); + goto again; + } atomic_inc(&root->log_writers); } mutex_unlock(&root->log_mutex); @@ -2752,6 +2785,8 @@ static noinline int walk_down_log_tree(struct btrfs_trans_handle *trans, free_extent_buffer(next); return ret; } + btrfs_redirty_list_add( + trans->transaction, next); } else { if (test_and_clear_bit(EXTENT_BUFFER_DIRTY, &next->bflags)) clear_extent_buffer_dirty(next); @@ -3085,6 +3120,17 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans, */ blk_start_plug(&plug); ret = btrfs_write_marked_extents(fs_info, &log->dirty_log_pages, mark); + /* + * -EAGAIN happens when someone, e.g., a concurrent transaction + * commit, writes a dirty extent in this tree-log commit. This + * concurrent write will create a hole writing out the extents, + * and we cannot proceed on a zoned filesystem, requiring + * sequential writing. While we can bail out to a full commit + * here, but we can continue hoping the concurrent writing fills + * the hole. + */ + if (ret == -EAGAIN && btrfs_is_zoned(fs_info)) + ret = 0; if (ret) { blk_finish_plug(&plug); btrfs_abort_transaction(trans, ret); @@ -3123,6 +3169,16 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans, mutex_lock(&log_root_tree->log_mutex); + if (btrfs_is_zoned(fs_info)) { + if (!log_root_tree->node) { + ret = btrfs_alloc_log_tree_node(trans, log_root_tree); + if (ret) { + mutex_unlock(&log_root_tree->log_mutex); + goto out; + } + } + } + index2 = log_root_tree->log_transid % 2; list_add_tail(&root_log_ctx.list, &log_root_tree->log_ctxs[index2]); root_log_ctx.log_transid = log_root_tree->log_transid; @@ -3194,7 +3250,17 @@ int btrfs_sync_log(struct btrfs_trans_handle *trans, &log_root_tree->dirty_log_pages, EXTENT_DIRTY | EXTENT_NEW); blk_finish_plug(&plug); - if (ret) { + /* + * As described above, -EAGAIN indicates a hole in the extents. We + * cannot wait for these write outs since the waiting cause a + * deadlock. Bail out to the full commit instead. + */ + if (ret == -EAGAIN && btrfs_is_zoned(fs_info)) { + btrfs_set_log_full_commit(trans); + btrfs_wait_tree_log_extents(log, mark); + mutex_unlock(&log_root_tree->log_mutex); + goto out_wake_log_root; + } else if (ret) { btrfs_set_log_full_commit(trans); btrfs_abort_transaction(trans, ret); mutex_unlock(&log_root_tree->log_mutex); @@ -3285,17 +3351,22 @@ static void free_log_tree(struct btrfs_trans_handle *trans, .process_func = process_one_buffer }; - ret = walk_log_tree(trans, log, &wc); - if (ret) { - if (trans) - btrfs_abort_transaction(trans, ret); - else - btrfs_handle_fs_error(log->fs_info, ret, NULL); + if (log->node) { + ret = walk_log_tree(trans, log, &wc); + if (ret) { + if (trans) + btrfs_abort_transaction(trans, ret); + else + btrfs_handle_fs_error(log->fs_info, ret, NULL); + } } clear_extent_bits(&log->dirty_log_pages, 0, (u64)-1, EXTENT_DIRTY | EXTENT_NEW | EXTENT_NEED_WAIT); extent_io_tree_release(&log->log_csum_range); + + if (trans && log->node) + btrfs_redirty_list_add(trans->transaction, log->node); btrfs_put_root(log); } @@ -3379,7 +3450,6 @@ int btrfs_del_dir_entries_in_log(struct btrfs_trans_handle *trans, struct btrfs_path *path; int ret; int err = 0; - int bytes_del = 0; u64 dir_ino = btrfs_ino(dir); if (!inode_logged(trans, dir)) @@ -3406,7 +3476,6 @@ int btrfs_del_dir_entries_in_log(struct btrfs_trans_handle *trans, } if (di) { ret = btrfs_delete_one_dir_name(trans, log, path, di); - bytes_del += name_len; if (ret) { err = ret; goto fail; @@ -3421,46 +3490,17 @@ int btrfs_del_dir_entries_in_log(struct btrfs_trans_handle *trans, } if (di) { ret = btrfs_delete_one_dir_name(trans, log, path, di); - bytes_del += name_len; if (ret) { err = ret; goto fail; } } - /* update the directory size in the log to reflect the names - * we have removed + /* + * We do not need to update the size field of the directory's inode item + * because on log replay we update the field to reflect all existing + * entries in the directory (see overwrite_item()). */ - if (bytes_del) { - struct btrfs_key key; - - key.objectid = dir_ino; - key.offset = 0; - key.type = BTRFS_INODE_ITEM_KEY; - btrfs_release_path(path); - - ret = btrfs_search_slot(trans, log, &key, path, 0, 1); - if (ret < 0) { - err = ret; - goto fail; - } - if (ret == 0) { - struct btrfs_inode_item *item; - u64 i_size; - - item = btrfs_item_ptr(path->nodes[0], path->slots[0], - struct btrfs_inode_item); - i_size = btrfs_inode_size(path->nodes[0], item); - if (i_size > bytes_del) - i_size -= bytes_del; - else - i_size = 0; - btrfs_set_inode_size(path->nodes[0], item, i_size); - btrfs_mark_buffer_dirty(path->nodes[0]); - } else - ret = 0; - btrfs_release_path(path); - } fail: btrfs_free_path(path); out_unlock: @@ -3889,7 +3929,14 @@ static void fill_inode_item(struct btrfs_trans_handle *trans, btrfs_set_token_timespec_nsec(&token, &item->ctime, inode->i_ctime.tv_nsec); - btrfs_set_token_inode_nbytes(&token, item, inode_get_bytes(inode)); + /* + * We do not need to set the nbytes field, in fact during a fast fsync + * its value may not even be correct, since a fast fsync does not wait + * for ordered extent completion, which is where we update nbytes, it + * only waits for writeback to complete. During log replay as we find + * file extent items and replay them, we adjust the nbytes field of the + * inode item in subvolume tree as needed (see overwrite_item()). + */ btrfs_set_token_inode_sequence(&token, item, inode_peek_iversion(inode)); btrfs_set_token_inode_transid(&token, item, trans->transid); @@ -5290,12 +5337,28 @@ static int btrfs_log_inode(struct btrfs_trans_handle *trans, } /* + * This is for cases where logging a directory could result in losing a + * a file after replaying the log. For example, if we move a file from a + * directory A to a directory B, then fsync directory A, we have no way + * to known the file was moved from A to B, so logging just A would + * result in losing the file after a log replay. + */ + if (S_ISDIR(inode->vfs_inode.i_mode) && + inode_only == LOG_INODE_ALL && + inode->last_unlink_trans >= trans->transid) { + btrfs_set_log_full_commit(trans); + err = 1; + goto out_unlock; + } + + /* * a brute force approach to making sure we get the most uptodate * copies of everything. */ if (S_ISDIR(inode->vfs_inode.i_mode)) { int max_key_type = BTRFS_DIR_LOG_INDEX_KEY; + clear_bit(BTRFS_INODE_COPY_EVERYTHING, &inode->runtime_flags); if (inode_only == LOG_INODE_EXISTS) max_key_type = BTRFS_XATTR_ITEM_KEY; ret = drop_objectid_items(trans, log, path, ino, max_key_type); @@ -5452,96 +5515,31 @@ out_unlock: } /* - * Check if we must fallback to a transaction commit when logging an inode. - * This must be called after logging the inode and is used only in the context - * when fsyncing an inode requires the need to log some other inode - in which - * case we can't lock the i_mutex of each other inode we need to log as that - * can lead to deadlocks with concurrent fsync against other inodes (as we can - * log inodes up or down in the hierarchy) or rename operations for example. So - * we take the log_mutex of the inode after we have logged it and then check for - * its last_unlink_trans value - this is safe because any task setting - * last_unlink_trans must take the log_mutex and it must do this before it does - * the actual unlink operation, so if we do this check before a concurrent task - * sets last_unlink_trans it means we've logged a consistent version/state of - * all the inode items, otherwise we are not sure and must do a transaction - * commit (the concurrent task might have only updated last_unlink_trans before - * we logged the inode or it might have also done the unlink). + * Check if we need to log an inode. This is used in contexts where while + * logging an inode we need to log another inode (either that it exists or in + * full mode). This is used instead of btrfs_inode_in_log() because the later + * requires the inode to be in the log and have the log transaction committed, + * while here we do not care if the log transaction was already committed - our + * caller will commit the log later - and we want to avoid logging an inode + * multiple times when multiple tasks have joined the same log transaction. */ -static bool btrfs_must_commit_transaction(struct btrfs_trans_handle *trans, - struct btrfs_inode *inode) +static bool need_log_inode(struct btrfs_trans_handle *trans, + struct btrfs_inode *inode) { - bool ret = false; - - mutex_lock(&inode->log_mutex); - if (inode->last_unlink_trans >= trans->transid) { - /* - * Make sure any commits to the log are forced to be full - * commits. - */ - btrfs_set_log_full_commit(trans); - ret = true; - } - mutex_unlock(&inode->log_mutex); - - return ret; -} - -/* - * follow the dentry parent pointers up the chain and see if any - * of the directories in it require a full commit before they can - * be logged. Returns zero if nothing special needs to be done or 1 if - * a full commit is required. - */ -static noinline int check_parent_dirs_for_sync(struct btrfs_trans_handle *trans, - struct btrfs_inode *inode, - struct dentry *parent, - struct super_block *sb) -{ - int ret = 0; - struct dentry *old_parent = NULL; - /* - * for regular files, if its inode is already on disk, we don't - * have to worry about the parents at all. This is because - * we can use the last_unlink_trans field to record renames - * and other fun in this file. + * If this inode does not have new/updated/deleted xattrs since the last + * time it was logged and is flagged as logged in the current transaction, + * we can skip logging it. As for new/deleted names, those are updated in + * the log by link/unlink/rename operations. + * In case the inode was logged and then evicted and reloaded, its + * logged_trans will be 0, in which case we have to fully log it since + * logged_trans is a transient field, not persisted. */ - if (S_ISREG(inode->vfs_inode.i_mode) && - inode->generation < trans->transid && - inode->last_unlink_trans < trans->transid) - goto out; - - if (!S_ISDIR(inode->vfs_inode.i_mode)) { - if (!parent || d_really_is_negative(parent) || sb != parent->d_sb) - goto out; - inode = BTRFS_I(d_inode(parent)); - } - - while (1) { - if (btrfs_must_commit_transaction(trans, inode)) { - ret = 1; - break; - } - - if (!parent || d_really_is_negative(parent) || sb != parent->d_sb) - break; - - if (IS_ROOT(parent)) { - inode = BTRFS_I(d_inode(parent)); - if (btrfs_must_commit_transaction(trans, inode)) - ret = 1; - break; - } - - parent = dget_parent(parent); - dput(old_parent); - old_parent = parent; - inode = BTRFS_I(d_inode(parent)); + if (inode->logged_trans == trans->transid && + !test_bit(BTRFS_INODE_COPY_EVERYTHING, &inode->runtime_flags)) + return false; - } - dput(old_parent); -out: - return ret; + return true; } struct btrfs_dir_list { @@ -5671,7 +5669,7 @@ process_leaf: goto next_dir_inode; } - if (btrfs_inode_in_log(BTRFS_I(di_inode), trans->transid)) { + if (!need_log_inode(trans, BTRFS_I(di_inode))) { btrfs_add_delayed_iput(di_inode); break; } @@ -5681,9 +5679,6 @@ process_leaf: log_mode = LOG_INODE_ALL; ret = btrfs_log_inode(trans, root, BTRFS_I(di_inode), log_mode, ctx); - if (!ret && - btrfs_must_commit_transaction(trans, BTRFS_I(di_inode))) - ret = 1; btrfs_add_delayed_iput(di_inode); if (ret) goto next_dir_inode; @@ -5821,13 +5816,15 @@ static int btrfs_log_all_parents(struct btrfs_trans_handle *trans, goto out; } + if (!need_log_inode(trans, BTRFS_I(dir_inode))) { + btrfs_add_delayed_iput(dir_inode); + continue; + } + if (ctx) ctx->log_new_dentries = false; ret = btrfs_log_inode(trans, root, BTRFS_I(dir_inode), LOG_INODE_ALL, ctx); - if (!ret && - btrfs_must_commit_transaction(trans, BTRFS_I(dir_inode))) - ret = 1; if (!ret && ctx && ctx->log_new_dentries) ret = log_new_dir_dentries(trans, root, BTRFS_I(dir_inode), ctx); @@ -5872,7 +5869,8 @@ static int log_new_ancestors(struct btrfs_trans_handle *trans, if (IS_ERR(inode)) return PTR_ERR(inode); - if (BTRFS_I(inode)->generation >= trans->transid) + if (BTRFS_I(inode)->generation >= trans->transid && + need_log_inode(trans, BTRFS_I(inode))) ret = btrfs_log_inode(trans, root, BTRFS_I(inode), LOG_INODE_EXISTS, ctx); btrfs_add_delayed_iput(inode); @@ -5926,7 +5924,8 @@ static int log_new_ancestors_fast(struct btrfs_trans_handle *trans, if (root != inode->root) break; - if (inode->generation >= trans->transid) { + if (inode->generation >= trans->transid && + need_log_inode(trans, inode)) { ret = btrfs_log_inode(trans, root, inode, LOG_INODE_EXISTS, ctx); if (ret) @@ -6041,12 +6040,9 @@ static int btrfs_log_inode_parent(struct btrfs_trans_handle *trans, { struct btrfs_root *root = inode->root; struct btrfs_fs_info *fs_info = root->fs_info; - struct super_block *sb; int ret = 0; bool log_dentries = false; - sb = inode->vfs_inode.i_sb; - if (btrfs_test_opt(fs_info, NOTREELOG)) { ret = 1; goto end_no_trans; @@ -6057,10 +6053,6 @@ static int btrfs_log_inode_parent(struct btrfs_trans_handle *trans, goto end_no_trans; } - ret = check_parent_dirs_for_sync(trans, inode, parent, sb); - if (ret) - goto end_no_trans; - /* * Skip already logged inodes or inodes corresponding to tmpfiles * (since logging them is pointless, a link count of 0 means they @@ -6307,8 +6299,7 @@ again: * root->objectid_mutex is not acquired as log replay * could only happen during mount. */ - ret = btrfs_find_highest_objectid(root, - &root->highest_objectid); + ret = btrfs_init_root_free_objectid(root); } wc.replay_dest->log_root = NULL; diff --git a/fs/btrfs/volumes.c b/fs/btrfs/volumes.c index d6c24c8ad749..bc3b33efddc5 100644 --- a/fs/btrfs/volumes.c +++ b/fs/btrfs/volumes.c @@ -421,7 +421,7 @@ static struct btrfs_device *__alloc_device(struct btrfs_fs_info *fs_info) * Preallocate a bio that's always going to be used for flushing device * barriers and matches the device lifespan */ - dev->flush_bio = bio_alloc_bioset(GFP_KERNEL, 0, NULL); + dev->flush_bio = bio_kmalloc(GFP_KERNEL, 0); if (!dev->flush_bio) { kfree(dev); return ERR_PTR(-ENOMEM); @@ -669,10 +669,6 @@ static int btrfs_open_one_device(struct btrfs_fs_devices *fs_devices, clear_bit(BTRFS_DEV_STATE_IN_FS_METADATA, &device->dev_state); device->mode = flags; - ret = btrfs_get_dev_zone_info(device); - if (ret != 0) - goto error_free_page; - fs_devices->open_devices++; if (test_bit(BTRFS_DEV_STATE_WRITEABLE, &device->dev_state) && device->devid != BTRFS_DEV_REPLACE_DEVID) { @@ -1418,11 +1414,62 @@ static u64 dev_extent_search_start(struct btrfs_device *device, u64 start) * make sure to start at an offset of at least 1MB. */ return max_t(u64, start, SZ_1M); + case BTRFS_CHUNK_ALLOC_ZONED: + /* + * We don't care about the starting region like regular + * allocator, because we anyway use/reserve the first two zones + * for superblock logging. + */ + return ALIGN(start, device->zone_info->zone_size); default: BUG(); } } +static bool dev_extent_hole_check_zoned(struct btrfs_device *device, + u64 *hole_start, u64 *hole_size, + u64 num_bytes) +{ + u64 zone_size = device->zone_info->zone_size; + u64 pos; + int ret; + bool changed = false; + + ASSERT(IS_ALIGNED(*hole_start, zone_size)); + + while (*hole_size > 0) { + pos = btrfs_find_allocatable_zones(device, *hole_start, + *hole_start + *hole_size, + num_bytes); + if (pos != *hole_start) { + *hole_size = *hole_start + *hole_size - pos; + *hole_start = pos; + changed = true; + if (*hole_size < num_bytes) + break; + } + + ret = btrfs_ensure_empty_zones(device, pos, num_bytes); + + /* Range is ensured to be empty */ + if (!ret) + return changed; + + /* Given hole range was invalid (outside of device) */ + if (ret == -ERANGE) { + *hole_start += *hole_size; + *hole_size = 0; + return 1; + } + + *hole_start += zone_size; + *hole_size -= zone_size; + changed = true; + } + + return changed; +} + /** * dev_extent_hole_check - check if specified hole is suitable for allocation * @device: the device which we have the hole @@ -1430,7 +1477,7 @@ static u64 dev_extent_search_start(struct btrfs_device *device, u64 start) * @hole_size: the size of the hole * @num_bytes: the size of the free space that we need * - * This function may modify @hole_start and @hole_end to reflect the suitable + * This function may modify @hole_start and @hole_size to reflect the suitable * position for allocation. Returns 1 if hole position is updated, 0 otherwise. */ static bool dev_extent_hole_check(struct btrfs_device *device, u64 *hole_start, @@ -1439,24 +1486,39 @@ static bool dev_extent_hole_check(struct btrfs_device *device, u64 *hole_start, bool changed = false; u64 hole_end = *hole_start + *hole_size; - /* - * Check before we set max_hole_start, otherwise we could end up - * sending back this offset anyway. - */ - if (contains_pending_extent(device, hole_start, *hole_size)) { - if (hole_end >= *hole_start) - *hole_size = hole_end - *hole_start; - else - *hole_size = 0; - changed = true; - } + for (;;) { + /* + * Check before we set max_hole_start, otherwise we could end up + * sending back this offset anyway. + */ + if (contains_pending_extent(device, hole_start, *hole_size)) { + if (hole_end >= *hole_start) + *hole_size = hole_end - *hole_start; + else + *hole_size = 0; + changed = true; + } + + switch (device->fs_devices->chunk_alloc_policy) { + case BTRFS_CHUNK_ALLOC_REGULAR: + /* No extra check */ + break; + case BTRFS_CHUNK_ALLOC_ZONED: + if (dev_extent_hole_check_zoned(device, hole_start, + hole_size, num_bytes)) { + changed = true; + /* + * The changed hole can contain pending extent. + * Loop again to check that. + */ + continue; + } + break; + default: + BUG(); + } - switch (device->fs_devices->chunk_alloc_policy) { - case BTRFS_CHUNK_ALLOC_REGULAR: - /* No extra check */ break; - default: - BUG(); } return changed; @@ -1509,6 +1571,9 @@ static int find_free_dev_extent_start(struct btrfs_device *device, search_start = dev_extent_search_start(device, search_start); + WARN_ON(device->zone_info && + !IS_ALIGNED(num_bytes, device->zone_info->zone_size)); + path = btrfs_alloc_path(); if (!path) return -ENOMEM; @@ -4668,11 +4733,10 @@ again: } ret = btrfs_previous_item(root, path, 0, key.type); - if (ret) - mutex_unlock(&fs_info->delete_unused_bgs_mutex); - if (ret < 0) - goto done; if (ret) { + mutex_unlock(&fs_info->delete_unused_bgs_mutex); + if (ret < 0) + goto done; ret = 0; btrfs_release_path(path); break; @@ -4904,6 +4968,37 @@ static void init_alloc_chunk_ctl_policy_regular( ctl->dev_extent_min = BTRFS_STRIPE_LEN * ctl->dev_stripes; } +static void init_alloc_chunk_ctl_policy_zoned( + struct btrfs_fs_devices *fs_devices, + struct alloc_chunk_ctl *ctl) +{ + u64 zone_size = fs_devices->fs_info->zone_size; + u64 limit; + int min_num_stripes = ctl->devs_min * ctl->dev_stripes; + int min_data_stripes = (min_num_stripes - ctl->nparity) / ctl->ncopies; + u64 min_chunk_size = min_data_stripes * zone_size; + u64 type = ctl->type; + + ctl->max_stripe_size = zone_size; + if (type & BTRFS_BLOCK_GROUP_DATA) { + ctl->max_chunk_size = round_down(BTRFS_MAX_DATA_CHUNK_SIZE, + zone_size); + } else if (type & BTRFS_BLOCK_GROUP_METADATA) { + ctl->max_chunk_size = ctl->max_stripe_size; + } else if (type & BTRFS_BLOCK_GROUP_SYSTEM) { + ctl->max_chunk_size = 2 * ctl->max_stripe_size; + ctl->devs_max = min_t(int, ctl->devs_max, + BTRFS_MAX_DEVS_SYS_CHUNK); + } + + /* We don't want a chunk larger than 10% of writable space */ + limit = max(round_down(div_factor(fs_devices->total_rw_bytes, 1), + zone_size), + min_chunk_size); + ctl->max_chunk_size = min(limit, ctl->max_chunk_size); + ctl->dev_extent_min = zone_size * ctl->dev_stripes; +} + static void init_alloc_chunk_ctl(struct btrfs_fs_devices *fs_devices, struct alloc_chunk_ctl *ctl) { @@ -4924,6 +5019,9 @@ static void init_alloc_chunk_ctl(struct btrfs_fs_devices *fs_devices, case BTRFS_CHUNK_ALLOC_REGULAR: init_alloc_chunk_ctl_policy_regular(fs_devices, ctl); break; + case BTRFS_CHUNK_ALLOC_ZONED: + init_alloc_chunk_ctl_policy_zoned(fs_devices, ctl); + break; default: BUG(); } @@ -5050,6 +5148,38 @@ static int decide_stripe_size_regular(struct alloc_chunk_ctl *ctl, return 0; } +static int decide_stripe_size_zoned(struct alloc_chunk_ctl *ctl, + struct btrfs_device_info *devices_info) +{ + u64 zone_size = devices_info[0].dev->zone_info->zone_size; + /* Number of stripes that count for block group size */ + int data_stripes; + + /* + * It should hold because: + * dev_extent_min == dev_extent_want == zone_size * dev_stripes + */ + ASSERT(devices_info[ctl->ndevs - 1].max_avail == ctl->dev_extent_min); + + ctl->stripe_size = zone_size; + ctl->num_stripes = ctl->ndevs * ctl->dev_stripes; + data_stripes = (ctl->num_stripes - ctl->nparity) / ctl->ncopies; + + /* stripe_size is fixed in zoned filesysmte. Reduce ndevs instead. */ + if (ctl->stripe_size * data_stripes > ctl->max_chunk_size) { + ctl->ndevs = div_u64(div_u64(ctl->max_chunk_size * ctl->ncopies, + ctl->stripe_size) + ctl->nparity, + ctl->dev_stripes); + ctl->num_stripes = ctl->ndevs * ctl->dev_stripes; + data_stripes = (ctl->num_stripes - ctl->nparity) / ctl->ncopies; + ASSERT(ctl->stripe_size * data_stripes <= ctl->max_chunk_size); + } + + ctl->chunk_size = ctl->stripe_size * data_stripes; + + return 0; +} + static int decide_stripe_size(struct btrfs_fs_devices *fs_devices, struct alloc_chunk_ctl *ctl, struct btrfs_device_info *devices_info) @@ -5077,6 +5207,8 @@ static int decide_stripe_size(struct btrfs_fs_devices *fs_devices, switch (fs_devices->chunk_alloc_policy) { case BTRFS_CHUNK_ALLOC_REGULAR: return decide_stripe_size_regular(ctl, devices_info); + case BTRFS_CHUNK_ALLOC_ZONED: + return decide_stripe_size_zoned(ctl, devices_info); default: BUG(); } @@ -5841,9 +5973,29 @@ static int get_extra_mirror_from_replace(struct btrfs_fs_info *fs_info, return ret; } +static bool is_block_group_to_copy(struct btrfs_fs_info *fs_info, u64 logical) +{ + struct btrfs_block_group *cache; + bool ret; + + /* Non zoned filesystem does not use "to_copy" flag */ + if (!btrfs_is_zoned(fs_info)) + return false; + + cache = btrfs_lookup_block_group(fs_info, logical); + + spin_lock(&cache->lock); + ret = cache->to_copy; + spin_unlock(&cache->lock); + + btrfs_put_block_group(cache); + return ret; +} + static void handle_ops_on_dev_replace(enum btrfs_map_op op, struct btrfs_bio **bbio_ret, struct btrfs_dev_replace *dev_replace, + u64 logical, int *num_stripes_ret, int *max_errors_ret) { struct btrfs_bio *bbio = *bbio_ret; @@ -5857,6 +6009,13 @@ static void handle_ops_on_dev_replace(enum btrfs_map_op op, int index_where_to_add; /* + * A block group which have "to_copy" set will eventually + * copied by dev-replace process. We can avoid cloning IO here. + */ + if (is_block_group_to_copy(dev_replace->srcdev->fs_info, logical)) + return; + + /* * duplicate the write operations while the dev replace * procedure is running. Since the copying of the old disk to * the new disk takes place at run time while the filesystem is @@ -5941,23 +6100,24 @@ static bool need_full_stripe(enum btrfs_map_op op) } /* - * btrfs_get_io_geometry - calculates the geomery of a particular (address, len) - * tuple. This information is used to calculate how big a - * particular bio can get before it straddles a stripe. + * Calculate the geometry of a particular (address, len) tuple. This + * information is used to calculate how big a particular bio can get before it + * straddles a stripe. * - * @fs_info - the filesystem - * @logical - address that we want to figure out the geometry of - * @len - the length of IO we are going to perform, starting at @logical - * @op - type of operation - write or read - * @io_geom - pointer used to return values + * @fs_info: the filesystem + * @em: mapping containing the logical extent + * @op: type of operation - write or read + * @logical: address that we want to figure out the geometry of + * @len: the length of IO we are going to perform, starting at @logical + * @io_geom: pointer used to return values * * Returns < 0 in case a chunk for the given logical address cannot be found, * usually shouldn't happen unless @logical is corrupted, 0 otherwise. */ -int btrfs_get_io_geometry(struct btrfs_fs_info *fs_info, enum btrfs_map_op op, - u64 logical, u64 len, struct btrfs_io_geometry *io_geom) +int btrfs_get_io_geometry(struct btrfs_fs_info *fs_info, struct extent_map *em, + enum btrfs_map_op op, u64 logical, u64 len, + struct btrfs_io_geometry *io_geom) { - struct extent_map *em; struct map_lookup *map; u64 offset; u64 stripe_offset; @@ -5965,14 +6125,9 @@ int btrfs_get_io_geometry(struct btrfs_fs_info *fs_info, enum btrfs_map_op op, u64 stripe_len; u64 raid56_full_stripe_start = (u64)-1; int data_stripes; - int ret = 0; ASSERT(op != BTRFS_MAP_DISCARD); - em = btrfs_get_chunk_map(fs_info, logical, len); - if (IS_ERR(em)) - return PTR_ERR(em); - map = em->map_lookup; /* Offset of this logical address in the chunk */ offset = logical - em->start; @@ -5986,8 +6141,7 @@ int btrfs_get_io_geometry(struct btrfs_fs_info *fs_info, enum btrfs_map_op op, btrfs_crit(fs_info, "stripe math has gone wrong, stripe_offset=%llu offset=%llu start=%llu logical=%llu stripe_len=%llu", stripe_offset, offset, em->start, logical, stripe_len); - ret = -EINVAL; - goto out; + return -EINVAL; } /* stripe_offset is the offset of this block in its stripe */ @@ -6034,10 +6188,7 @@ int btrfs_get_io_geometry(struct btrfs_fs_info *fs_info, enum btrfs_map_op op, io_geom->stripe_offset = stripe_offset; io_geom->raid56_stripe_offset = raid56_full_stripe_start; -out: - /* once for us */ - free_extent_map(em); - return ret; + return 0; } static int __btrfs_map_block(struct btrfs_fs_info *fs_info, @@ -6070,12 +6221,13 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info, ASSERT(bbio_ret); ASSERT(op != BTRFS_MAP_DISCARD); - ret = btrfs_get_io_geometry(fs_info, op, logical, *length, &geom); + em = btrfs_get_chunk_map(fs_info, logical, *length); + ASSERT(!IS_ERR(em)); + + ret = btrfs_get_io_geometry(fs_info, em, op, logical, *length, &geom); if (ret < 0) return ret; - em = btrfs_get_chunk_map(fs_info, logical, *length); - ASSERT(!IS_ERR(em)); map = em->map_lookup; *length = geom.len; @@ -6251,8 +6403,8 @@ static int __btrfs_map_block(struct btrfs_fs_info *fs_info, if (dev_replace_is_ongoing && dev_replace->tgtdev != NULL && need_full_stripe(op)) { - handle_ops_on_dev_replace(op, &bbio, dev_replace, &num_stripes, - &max_errors); + handle_ops_on_dev_replace(op, &bbio, dev_replace, logical, + &num_stripes, &max_errors); } *bbio_ret = bbio; @@ -6323,7 +6475,7 @@ static void btrfs_end_bio(struct bio *bio) struct btrfs_device *dev = btrfs_io_bio(bio)->device; ASSERT(dev->bdev); - if (bio_op(bio) == REQ_OP_WRITE) + if (btrfs_op(bio) == BTRFS_MAP_WRITE) btrfs_dev_stat_inc_and_print(dev, BTRFS_DEV_STAT_WRITE_ERRS); else if (!(bio->bi_opf & REQ_RAHEAD)) @@ -6375,6 +6527,20 @@ static void submit_stripe_bio(struct btrfs_bio *bbio, struct bio *bio, btrfs_io_bio(bio)->device = dev; bio->bi_end_io = btrfs_end_bio; bio->bi_iter.bi_sector = physical >> 9; + /* + * For zone append writing, bi_sector must point the beginning of the + * zone + */ + if (bio_op(bio) == REQ_OP_ZONE_APPEND) { + if (btrfs_dev_is_sequential(dev, physical)) { + u64 zone_start = round_down(physical, fs_info->zone_size); + + bio->bi_iter.bi_sector = zone_start >> SECTOR_SHIFT; + } else { + bio->bi_opf &= ~REQ_OP_ZONE_APPEND; + bio->bi_opf |= REQ_OP_WRITE; + } + } btrfs_debug_in_rcu(fs_info, "btrfs_map_bio: rw %d 0x%x, sector=%llu, dev=%lu (%s id %llu), size=%u", bio_op(bio), bio->bi_opf, bio->bi_iter.bi_sector, @@ -6436,10 +6602,10 @@ blk_status_t btrfs_map_bio(struct btrfs_fs_info *fs_info, struct bio *bio, atomic_set(&bbio->stripes_pending, bbio->num_stripes); if ((bbio->map_type & BTRFS_BLOCK_GROUP_RAID56_MASK) && - ((bio_op(bio) == REQ_OP_WRITE) || (mirror_num > 1))) { + ((btrfs_op(bio) == BTRFS_MAP_WRITE) || (mirror_num > 1))) { /* In this case, map_length has been set to the length of a single stripe; not the whole write */ - if (bio_op(bio) == REQ_OP_WRITE) { + if (btrfs_op(bio) == BTRFS_MAP_WRITE) { ret = raid56_parity_write(fs_info, bio, bbio, map_length); } else { @@ -6462,7 +6628,7 @@ blk_status_t btrfs_map_bio(struct btrfs_fs_info *fs_info, struct bio *bio, dev = bbio->stripes[dev_nr].dev; if (!dev || !dev->bdev || test_bit(BTRFS_DEV_STATE_MISSING, &dev->dev_state) || - (bio_op(first_bio) == REQ_OP_WRITE && + (btrfs_op(first_bio) == BTRFS_MAP_WRITE && !test_bit(BTRFS_DEV_STATE_WRITEABLE, &dev->dev_state))) { bbio_error(bbio, first_bio, logical); continue; @@ -7644,6 +7810,20 @@ static int verify_one_dev_extent(struct btrfs_fs_info *fs_info, ret = -EUCLEAN; goto out; } + + if (dev->zone_info) { + u64 zone_size = dev->zone_info->zone_size; + + if (!IS_ALIGNED(physical_offset, zone_size) || + !IS_ALIGNED(physical_len, zone_size)) { + btrfs_err(fs_info, +"zoned: dev extent devid %llu physical offset %llu len %llu is not aligned to device zone", + devid, physical_offset, physical_len); + ret = -EUCLEAN; + goto out; + } + } + out: free_extent_map(em); return ret; @@ -7800,3 +7980,75 @@ bool btrfs_pinned_by_swapfile(struct btrfs_fs_info *fs_info, void *ptr) spin_unlock(&fs_info->swapfile_pins_lock); return node != NULL; } + +static int relocating_repair_kthread(void *data) +{ + struct btrfs_block_group *cache = (struct btrfs_block_group *)data; + struct btrfs_fs_info *fs_info = cache->fs_info; + u64 target; + int ret = 0; + + target = cache->start; + btrfs_put_block_group(cache); + + if (!btrfs_exclop_start(fs_info, BTRFS_EXCLOP_BALANCE)) { + btrfs_info(fs_info, + "zoned: skip relocating block group %llu to repair: EBUSY", + target); + return -EBUSY; + } + + mutex_lock(&fs_info->delete_unused_bgs_mutex); + + /* Ensure block group still exists */ + cache = btrfs_lookup_block_group(fs_info, target); + if (!cache) + goto out; + + if (!cache->relocating_repair) + goto out; + + ret = btrfs_may_alloc_data_chunk(fs_info, target); + if (ret < 0) + goto out; + + btrfs_info(fs_info, + "zoned: relocating block group %llu to repair IO failure", + target); + ret = btrfs_relocate_chunk(fs_info, target); + +out: + if (cache) + btrfs_put_block_group(cache); + mutex_unlock(&fs_info->delete_unused_bgs_mutex); + btrfs_exclop_finish(fs_info); + + return ret; +} + +int btrfs_repair_one_zone(struct btrfs_fs_info *fs_info, u64 logical) +{ + struct btrfs_block_group *cache; + + /* Do not attempt to repair in degraded state */ + if (btrfs_test_opt(fs_info, DEGRADED)) + return 0; + + cache = btrfs_lookup_block_group(fs_info, logical); + if (!cache) + return 0; + + spin_lock(&cache->lock); + if (cache->relocating_repair) { + spin_unlock(&cache->lock); + btrfs_put_block_group(cache); + return 0; + } + cache->relocating_repair = 1; + spin_unlock(&cache->lock); + + kthread_run(relocating_repair_kthread, cache, + "btrfs-relocating-repair"); + + return 0; +} diff --git a/fs/btrfs/volumes.h b/fs/btrfs/volumes.h index c43663d9c22e..d4c3e0dd32b8 100644 --- a/fs/btrfs/volumes.h +++ b/fs/btrfs/volumes.h @@ -214,6 +214,7 @@ BTRFS_DEVICE_GETSET_FUNCS(bytes_used); enum btrfs_chunk_allocation_policy { BTRFS_CHUNK_ALLOC_REGULAR, + BTRFS_CHUNK_ALLOC_ZONED, }; /* @@ -423,6 +424,7 @@ static inline enum btrfs_map_op btrfs_op(struct bio *bio) case REQ_OP_DISCARD: return BTRFS_MAP_DISCARD; case REQ_OP_WRITE: + case REQ_OP_ZONE_APPEND: return BTRFS_MAP_WRITE; default: WARN_ON_ONCE(1); @@ -440,8 +442,9 @@ int btrfs_map_block(struct btrfs_fs_info *fs_info, enum btrfs_map_op op, int btrfs_map_sblock(struct btrfs_fs_info *fs_info, enum btrfs_map_op op, u64 logical, u64 *length, struct btrfs_bio **bbio_ret); -int btrfs_get_io_geometry(struct btrfs_fs_info *fs_info, enum btrfs_map_op op, - u64 logical, u64 len, struct btrfs_io_geometry *io_geom); +int btrfs_get_io_geometry(struct btrfs_fs_info *fs_info, struct extent_map *map, + enum btrfs_map_op op, u64 logical, u64 len, + struct btrfs_io_geometry *io_geom); int btrfs_read_sys_array(struct btrfs_fs_info *fs_info); int btrfs_read_chunk_tree(struct btrfs_fs_info *fs_info); int btrfs_alloc_chunk(struct btrfs_trans_handle *trans, u64 type); @@ -596,5 +599,6 @@ void btrfs_scratch_superblocks(struct btrfs_fs_info *fs_info, int btrfs_bg_type_to_factor(u64 flags); const char *btrfs_bg_type_to_raid_name(u64 flags); int btrfs_verify_dev_extents(struct btrfs_fs_info *fs_info); +int btrfs_repair_one_zone(struct btrfs_fs_info *fs_info, u64 logical); #endif diff --git a/fs/btrfs/xattr.c b/fs/btrfs/xattr.c index af6246f36a9e..8a4514283a4b 100644 --- a/fs/btrfs/xattr.c +++ b/fs/btrfs/xattr.c @@ -229,11 +229,33 @@ int btrfs_setxattr_trans(struct inode *inode, const char *name, { struct btrfs_root *root = BTRFS_I(inode)->root; struct btrfs_trans_handle *trans; + const bool start_trans = (current->journal_info == NULL); int ret; - trans = btrfs_start_transaction(root, 2); - if (IS_ERR(trans)) - return PTR_ERR(trans); + if (start_trans) { + /* + * 1 unit for inserting/updating/deleting the xattr + * 1 unit for the inode item update + */ + trans = btrfs_start_transaction(root, 2); + if (IS_ERR(trans)) + return PTR_ERR(trans); + } else { + /* + * This can happen when smack is enabled and a directory is being + * created. It happens through d_instantiate_new(), which calls + * smack_d_instantiate(), which in turn calls __vfs_setxattr() to + * set the transmute xattr (XATTR_NAME_SMACKTRANSMUTE) on the + * inode. We have already reserved space for the xattr and inode + * update at btrfs_mkdir(), so just use the transaction handle. + * We don't join or start a transaction, as that will reset the + * block_rsv of the handle and trigger a warning for the start + * case. + */ + ASSERT(strncmp(name, XATTR_SECURITY_PREFIX, + XATTR_SECURITY_PREFIX_LEN) == 0); + trans = current->journal_info; + } ret = btrfs_setxattr(trans, inode, name, value, size, flags); if (ret) @@ -244,7 +266,8 @@ int btrfs_setxattr_trans(struct inode *inode, const char *name, ret = btrfs_update_inode(trans, root, BTRFS_I(inode)); BUG_ON(ret); out: - btrfs_end_transaction(trans); + if (start_trans) + btrfs_end_transaction(trans); return ret; } @@ -362,6 +385,7 @@ static int btrfs_xattr_handler_get(const struct xattr_handler *handler, } static int btrfs_xattr_handler_set(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *unused, struct inode *inode, const char *name, const void *buffer, size_t size, int flags) @@ -371,6 +395,7 @@ static int btrfs_xattr_handler_set(const struct xattr_handler *handler, } static int btrfs_xattr_handler_set_prop(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *unused, struct inode *inode, const char *name, const void *value, size_t size, int flags) diff --git a/fs/btrfs/zlib.c b/fs/btrfs/zlib.c index 05615a1099db..d524acf7b3e5 100644 --- a/fs/btrfs/zlib.c +++ b/fs/btrfs/zlib.c @@ -432,9 +432,8 @@ int zlib_decompress(struct list_head *ws, unsigned char *data_in, PAGE_SIZE - (buf_offset % PAGE_SIZE)); bytes = min(bytes, bytes_left); - kaddr = kmap_atomic(dest_page); - memcpy(kaddr + pg_offset, workspace->buf + buf_offset, bytes); - kunmap_atomic(kaddr); + memcpy_to_page(dest_page, pg_offset, + workspace->buf + buf_offset, bytes); pg_offset += bytes; bytes_left -= bytes; diff --git a/fs/btrfs/zoned.c b/fs/btrfs/zoned.c index c38846659019..1f972b75a9ab 100644 --- a/fs/btrfs/zoned.c +++ b/fs/btrfs/zoned.c @@ -1,14 +1,25 @@ // SPDX-License-Identifier: GPL-2.0 +#include <linux/bitops.h> #include <linux/slab.h> #include <linux/blkdev.h> +#include <linux/sched/mm.h> #include "ctree.h" #include "volumes.h" #include "zoned.h" #include "rcu-string.h" +#include "disk-io.h" +#include "block-group.h" +#include "transaction.h" +#include "dev-replace.h" +#include "space-info.h" /* Maximum number of zones to report per blkdev_report_zones() call */ #define BTRFS_REPORT_NR_ZONES 4096 +/* Invalid allocation pointer value for missing devices */ +#define WP_MISSING_DEV ((u64)-1) +/* Pseudo write pointer value for conventional zone */ +#define WP_CONVENTIONAL ((u64)-2) /* Number of superblock log zones */ #define BTRFS_NR_SB_LOG_ZONES 2 @@ -119,6 +130,36 @@ static inline u32 sb_zone_number(int shift, int mirror) return 0; } +/* + * Emulate blkdev_report_zones() for a non-zoned device. It slices up the block + * device into static sized chunks and fake a conventional zone on each of + * them. + */ +static int emulate_report_zones(struct btrfs_device *device, u64 pos, + struct blk_zone *zones, unsigned int nr_zones) +{ + const sector_t zone_sectors = device->fs_info->zone_size >> SECTOR_SHIFT; + sector_t bdev_size = bdev_nr_sectors(device->bdev); + unsigned int i; + + pos >>= SECTOR_SHIFT; + for (i = 0; i < nr_zones; i++) { + zones[i].start = i * zone_sectors + pos; + zones[i].len = zone_sectors; + zones[i].capacity = zone_sectors; + zones[i].wp = zones[i].start + zone_sectors; + zones[i].type = BLK_ZONE_TYPE_CONVENTIONAL; + zones[i].cond = BLK_ZONE_COND_NOT_WP; + + if (zones[i].wp >= bdev_size) { + i++; + break; + } + } + + return i; +} + static int btrfs_get_dev_zones(struct btrfs_device *device, u64 pos, struct blk_zone *zones, unsigned int *nr_zones) { @@ -127,6 +168,12 @@ static int btrfs_get_dev_zones(struct btrfs_device *device, u64 pos, if (!*nr_zones) return 0; + if (!bdev_is_zoned(device->bdev)) { + ret = emulate_report_zones(device, pos, zones, *nr_zones); + *nr_zones = ret; + return 0; + } + ret = blkdev_report_zones(device->bdev, pos >> SECTOR_SHIFT, *nr_zones, copy_zone_info_cb, zones); if (ret < 0) { @@ -143,8 +190,78 @@ static int btrfs_get_dev_zones(struct btrfs_device *device, u64 pos, return 0; } +/* The emulated zone size is determined from the size of device extent */ +static int calculate_emulated_zone_size(struct btrfs_fs_info *fs_info) +{ + struct btrfs_path *path; + struct btrfs_root *root = fs_info->dev_root; + struct btrfs_key key; + struct extent_buffer *leaf; + struct btrfs_dev_extent *dext; + int ret = 0; + + key.objectid = 1; + key.type = BTRFS_DEV_EXTENT_KEY; + key.offset = 0; + + path = btrfs_alloc_path(); + if (!path) + return -ENOMEM; + + ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); + if (ret < 0) + goto out; + + if (path->slots[0] >= btrfs_header_nritems(path->nodes[0])) { + ret = btrfs_next_item(root, path); + if (ret < 0) + goto out; + /* No dev extents at all? Not good */ + if (ret > 0) { + ret = -EUCLEAN; + goto out; + } + } + + leaf = path->nodes[0]; + dext = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_dev_extent); + fs_info->zone_size = btrfs_dev_extent_length(leaf, dext); + ret = 0; + +out: + btrfs_free_path(path); + + return ret; +} + +int btrfs_get_dev_zone_info_all_devices(struct btrfs_fs_info *fs_info) +{ + struct btrfs_fs_devices *fs_devices = fs_info->fs_devices; + struct btrfs_device *device; + int ret = 0; + + /* fs_info->zone_size might not set yet. Use the incomapt flag here. */ + if (!btrfs_fs_incompat(fs_info, ZONED)) + return 0; + + mutex_lock(&fs_devices->device_list_mutex); + list_for_each_entry(device, &fs_devices->devices, dev_list) { + /* We can skip reading of zone info for missing devices */ + if (!device->bdev) + continue; + + ret = btrfs_get_dev_zone_info(device); + if (ret) + break; + } + mutex_unlock(&fs_devices->device_list_mutex); + + return ret; +} + int btrfs_get_dev_zone_info(struct btrfs_device *device) { + struct btrfs_fs_info *fs_info = device->fs_info; struct btrfs_zoned_device_info *zone_info = NULL; struct block_device *bdev = device->bdev; struct request_queue *queue = bdev_get_queue(bdev); @@ -152,10 +269,15 @@ int btrfs_get_dev_zone_info(struct btrfs_device *device) sector_t sector = 0; struct blk_zone *zones = NULL; unsigned int i, nreported = 0, nr_zones; - unsigned int zone_sectors; + sector_t zone_sectors; + char *model, *emulated; int ret; - if (!bdev_is_zoned(bdev)) + /* + * Cannot use btrfs_is_zoned here, since fs_info::zone_size might not + * yet be set. + */ + if (!btrfs_fs_incompat(fs_info, ZONED)) return 0; if (device->zone_info) @@ -165,8 +287,20 @@ int btrfs_get_dev_zone_info(struct btrfs_device *device) if (!zone_info) return -ENOMEM; + if (!bdev_is_zoned(bdev)) { + if (!fs_info->zone_size) { + ret = calculate_emulated_zone_size(fs_info); + if (ret) + goto out; + } + + ASSERT(fs_info->zone_size); + zone_sectors = fs_info->zone_size >> SECTOR_SHIFT; + } else { + zone_sectors = bdev_zone_sectors(bdev); + } + nr_sectors = bdev_nr_sectors(bdev); - zone_sectors = bdev_zone_sectors(bdev); /* Check if it's power of 2 (see is_power_of_2) */ ASSERT(zone_sectors != 0 && (zone_sectors & (zone_sectors - 1)) == 0); zone_info->zone_size = zone_sectors << SECTOR_SHIFT; @@ -272,20 +406,42 @@ int btrfs_get_dev_zone_info(struct btrfs_device *device) device->zone_info = zone_info; - /* device->fs_info is not safe to use for printing messages */ - btrfs_info_in_rcu(NULL, - "host-%s zoned block device %s, %u zones of %llu bytes", - bdev_zoned_model(bdev) == BLK_ZONED_HM ? "managed" : "aware", - rcu_str_deref(device->name), zone_info->nr_zones, - zone_info->zone_size); + switch (bdev_zoned_model(bdev)) { + case BLK_ZONED_HM: + model = "host-managed zoned"; + emulated = ""; + break; + case BLK_ZONED_HA: + model = "host-aware zoned"; + emulated = ""; + break; + case BLK_ZONED_NONE: + model = "regular"; + emulated = "emulated "; + break; + default: + /* Just in case */ + btrfs_err_in_rcu(fs_info, "zoned: unsupported model %d on %s", + bdev_zoned_model(bdev), + rcu_str_deref(device->name)); + ret = -EOPNOTSUPP; + goto out_free_zone_info; + } + + btrfs_info_in_rcu(fs_info, + "%s block device %s, %u %szones of %llu bytes", + model, rcu_str_deref(device->name), zone_info->nr_zones, + emulated, zone_info->zone_size); return 0; out: kfree(zones); +out_free_zone_info: bitmap_free(zone_info->empty_zones); bitmap_free(zone_info->seq_zones); kfree(zone_info); + device->zone_info = NULL; return ret; } @@ -324,7 +480,7 @@ int btrfs_check_zoned_mode(struct btrfs_fs_info *fs_info) u64 nr_devices = 0; u64 zone_size = 0; u64 max_zone_append_size = 0; - const bool incompat_zoned = btrfs_is_zoned(fs_info); + const bool incompat_zoned = btrfs_fs_incompat(fs_info, ZONED); int ret = 0; /* Count zoned devices */ @@ -335,9 +491,17 @@ int btrfs_check_zoned_mode(struct btrfs_fs_info *fs_info) continue; model = bdev_zoned_model(device->bdev); + /* + * A Host-Managed zoned device must be used as a zoned device. + * A Host-Aware zoned device and a non-zoned devices can be + * treated as a zoned device, if ZONED flag is enabled in the + * superblock. + */ if (model == BLK_ZONED_HM || - (model == BLK_ZONED_HA && incompat_zoned)) { - struct btrfs_zoned_device_info *zone_info; + (model == BLK_ZONED_HA && incompat_zoned) || + (model == BLK_ZONED_NONE && incompat_zoned)) { + struct btrfs_zoned_device_info *zone_info = + device->zone_info; zone_info = device->zone_info; zoned_devices++; @@ -406,6 +570,15 @@ int btrfs_check_zoned_mode(struct btrfs_fs_info *fs_info) fs_info->zone_size = zone_size; fs_info->max_zone_append_size = max_zone_append_size; + fs_info->fs_devices->chunk_alloc_policy = BTRFS_CHUNK_ALLOC_ZONED; + + /* + * Check mount options here, because we might change fs_info->zoned + * from fs_info->zone_size. + */ + ret = btrfs_check_mountopts_zoned(fs_info); + if (ret) + goto out; btrfs_info(fs_info, "zoned mode enabled with zone size %llu", zone_size); out: @@ -485,10 +658,9 @@ int btrfs_sb_log_location_bdev(struct block_device *bdev, int mirror, int rw, u64 *bytenr_ret) { struct blk_zone zones[BTRFS_NR_SB_LOG_ZONES]; - unsigned int zone_sectors; + sector_t zone_sectors; u32 sb_zone; int ret; - u64 zone_size; u8 zone_sectors_shift; sector_t nr_sectors; u32 nr_zones; @@ -503,7 +675,6 @@ int btrfs_sb_log_location_bdev(struct block_device *bdev, int mirror, int rw, zone_sectors = bdev_zone_sectors(bdev); if (!is_power_of_2(zone_sectors)) return -EINVAL; - zone_size = zone_sectors << SECTOR_SHIFT; zone_sectors_shift = ilog2(zone_sectors); nr_sectors = bdev_nr_sectors(bdev); nr_zones = nr_sectors >> zone_sectors_shift; @@ -529,7 +700,13 @@ int btrfs_sb_log_location(struct btrfs_device *device, int mirror, int rw, struct btrfs_zoned_device_info *zinfo = device->zone_info; u32 zone_num; - if (!zinfo) { + /* + * For a zoned filesystem on a non-zoned block device, use the same + * super block locations as regular filesystem. Doing so, the super + * block can always be retrieved and the zoned flag of the volume + * detected from the super block information. + */ + if (!bdev_is_zoned(device->bdev)) { *bytenr_ret = btrfs_sb_offset(mirror); return 0; } @@ -614,3 +791,671 @@ int btrfs_reset_sb_log_zones(struct block_device *bdev, int mirror) sb_zone << zone_sectors_shift, zone_sectors * BTRFS_NR_SB_LOG_ZONES, GFP_NOFS); } + +/** + * btrfs_find_allocatable_zones - find allocatable zones within a given region + * + * @device: the device to allocate a region on + * @hole_start: the position of the hole to allocate the region + * @num_bytes: size of wanted region + * @hole_end: the end of the hole + * @return: position of allocatable zones + * + * Allocatable region should not contain any superblock locations. + */ +u64 btrfs_find_allocatable_zones(struct btrfs_device *device, u64 hole_start, + u64 hole_end, u64 num_bytes) +{ + struct btrfs_zoned_device_info *zinfo = device->zone_info; + const u8 shift = zinfo->zone_size_shift; + u64 nzones = num_bytes >> shift; + u64 pos = hole_start; + u64 begin, end; + bool have_sb; + int i; + + ASSERT(IS_ALIGNED(hole_start, zinfo->zone_size)); + ASSERT(IS_ALIGNED(num_bytes, zinfo->zone_size)); + + while (pos < hole_end) { + begin = pos >> shift; + end = begin + nzones; + + if (end > zinfo->nr_zones) + return hole_end; + + /* Check if zones in the region are all empty */ + if (btrfs_dev_is_sequential(device, pos) && + find_next_zero_bit(zinfo->empty_zones, end, begin) != end) { + pos += zinfo->zone_size; + continue; + } + + have_sb = false; + for (i = 0; i < BTRFS_SUPER_MIRROR_MAX; i++) { + u32 sb_zone; + u64 sb_pos; + + sb_zone = sb_zone_number(shift, i); + if (!(end <= sb_zone || + sb_zone + BTRFS_NR_SB_LOG_ZONES <= begin)) { + have_sb = true; + pos = ((u64)sb_zone + BTRFS_NR_SB_LOG_ZONES) << shift; + break; + } + + /* We also need to exclude regular superblock positions */ + sb_pos = btrfs_sb_offset(i); + if (!(pos + num_bytes <= sb_pos || + sb_pos + BTRFS_SUPER_INFO_SIZE <= pos)) { + have_sb = true; + pos = ALIGN(sb_pos + BTRFS_SUPER_INFO_SIZE, + zinfo->zone_size); + break; + } + } + if (!have_sb) + break; + } + + return pos; +} + +int btrfs_reset_device_zone(struct btrfs_device *device, u64 physical, + u64 length, u64 *bytes) +{ + int ret; + + *bytes = 0; + ret = blkdev_zone_mgmt(device->bdev, REQ_OP_ZONE_RESET, + physical >> SECTOR_SHIFT, length >> SECTOR_SHIFT, + GFP_NOFS); + if (ret) + return ret; + + *bytes = length; + while (length) { + btrfs_dev_set_zone_empty(device, physical); + physical += device->zone_info->zone_size; + length -= device->zone_info->zone_size; + } + + return 0; +} + +int btrfs_ensure_empty_zones(struct btrfs_device *device, u64 start, u64 size) +{ + struct btrfs_zoned_device_info *zinfo = device->zone_info; + const u8 shift = zinfo->zone_size_shift; + unsigned long begin = start >> shift; + unsigned long end = (start + size) >> shift; + u64 pos; + int ret; + + ASSERT(IS_ALIGNED(start, zinfo->zone_size)); + ASSERT(IS_ALIGNED(size, zinfo->zone_size)); + + if (end > zinfo->nr_zones) + return -ERANGE; + + /* All the zones are conventional */ + if (find_next_bit(zinfo->seq_zones, begin, end) == end) + return 0; + + /* All the zones are sequential and empty */ + if (find_next_zero_bit(zinfo->seq_zones, begin, end) == end && + find_next_zero_bit(zinfo->empty_zones, begin, end) == end) + return 0; + + for (pos = start; pos < start + size; pos += zinfo->zone_size) { + u64 reset_bytes; + + if (!btrfs_dev_is_sequential(device, pos) || + btrfs_dev_is_empty_zone(device, pos)) + continue; + + /* Free regions should be empty */ + btrfs_warn_in_rcu( + device->fs_info, + "zoned: resetting device %s (devid %llu) zone %llu for allocation", + rcu_str_deref(device->name), device->devid, pos >> shift); + WARN_ON_ONCE(1); + + ret = btrfs_reset_device_zone(device, pos, zinfo->zone_size, + &reset_bytes); + if (ret) + return ret; + } + + return 0; +} + +/* + * Calculate an allocation pointer from the extent allocation information + * for a block group consist of conventional zones. It is pointed to the + * end of the highest addressed extent in the block group as an allocation + * offset. + */ +static int calculate_alloc_pointer(struct btrfs_block_group *cache, + u64 *offset_ret) +{ + struct btrfs_fs_info *fs_info = cache->fs_info; + struct btrfs_root *root = fs_info->extent_root; + struct btrfs_path *path; + struct btrfs_key key; + struct btrfs_key found_key; + int ret; + u64 length; + + path = btrfs_alloc_path(); + if (!path) + return -ENOMEM; + + key.objectid = cache->start + cache->length; + key.type = 0; + key.offset = 0; + + ret = btrfs_search_slot(NULL, root, &key, path, 0, 0); + /* We should not find the exact match */ + if (!ret) + ret = -EUCLEAN; + if (ret < 0) + goto out; + + ret = btrfs_previous_extent_item(root, path, cache->start); + if (ret) { + if (ret == 1) { + ret = 0; + *offset_ret = 0; + } + goto out; + } + + btrfs_item_key_to_cpu(path->nodes[0], &found_key, path->slots[0]); + + if (found_key.type == BTRFS_EXTENT_ITEM_KEY) + length = found_key.offset; + else + length = fs_info->nodesize; + + if (!(found_key.objectid >= cache->start && + found_key.objectid + length <= cache->start + cache->length)) { + ret = -EUCLEAN; + goto out; + } + *offset_ret = found_key.objectid + length - cache->start; + ret = 0; + +out: + btrfs_free_path(path); + return ret; +} + +int btrfs_load_block_group_zone_info(struct btrfs_block_group *cache, bool new) +{ + struct btrfs_fs_info *fs_info = cache->fs_info; + struct extent_map_tree *em_tree = &fs_info->mapping_tree; + struct extent_map *em; + struct map_lookup *map; + struct btrfs_device *device; + u64 logical = cache->start; + u64 length = cache->length; + u64 physical = 0; + int ret; + int i; + unsigned int nofs_flag; + u64 *alloc_offsets = NULL; + u64 last_alloc = 0; + u32 num_sequential = 0, num_conventional = 0; + + if (!btrfs_is_zoned(fs_info)) + return 0; + + /* Sanity check */ + if (!IS_ALIGNED(length, fs_info->zone_size)) { + btrfs_err(fs_info, + "zoned: block group %llu len %llu unaligned to zone size %llu", + logical, length, fs_info->zone_size); + return -EIO; + } + + /* Get the chunk mapping */ + read_lock(&em_tree->lock); + em = lookup_extent_mapping(em_tree, logical, length); + read_unlock(&em_tree->lock); + + if (!em) + return -EINVAL; + + map = em->map_lookup; + + alloc_offsets = kcalloc(map->num_stripes, sizeof(*alloc_offsets), GFP_NOFS); + if (!alloc_offsets) { + free_extent_map(em); + return -ENOMEM; + } + + for (i = 0; i < map->num_stripes; i++) { + bool is_sequential; + struct blk_zone zone; + struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace; + int dev_replace_is_ongoing = 0; + + device = map->stripes[i].dev; + physical = map->stripes[i].physical; + + if (device->bdev == NULL) { + alloc_offsets[i] = WP_MISSING_DEV; + continue; + } + + is_sequential = btrfs_dev_is_sequential(device, physical); + if (is_sequential) + num_sequential++; + else + num_conventional++; + + if (!is_sequential) { + alloc_offsets[i] = WP_CONVENTIONAL; + continue; + } + + /* + * This zone will be used for allocation, so mark this zone + * non-empty. + */ + btrfs_dev_clear_zone_empty(device, physical); + + down_read(&dev_replace->rwsem); + dev_replace_is_ongoing = btrfs_dev_replace_is_ongoing(dev_replace); + if (dev_replace_is_ongoing && dev_replace->tgtdev != NULL) + btrfs_dev_clear_zone_empty(dev_replace->tgtdev, physical); + up_read(&dev_replace->rwsem); + + /* + * The group is mapped to a sequential zone. Get the zone write + * pointer to determine the allocation offset within the zone. + */ + WARN_ON(!IS_ALIGNED(physical, fs_info->zone_size)); + nofs_flag = memalloc_nofs_save(); + ret = btrfs_get_dev_zone(device, physical, &zone); + memalloc_nofs_restore(nofs_flag); + if (ret == -EIO || ret == -EOPNOTSUPP) { + ret = 0; + alloc_offsets[i] = WP_MISSING_DEV; + continue; + } else if (ret) { + goto out; + } + + switch (zone.cond) { + case BLK_ZONE_COND_OFFLINE: + case BLK_ZONE_COND_READONLY: + btrfs_err(fs_info, + "zoned: offline/readonly zone %llu on device %s (devid %llu)", + physical >> device->zone_info->zone_size_shift, + rcu_str_deref(device->name), device->devid); + alloc_offsets[i] = WP_MISSING_DEV; + break; + case BLK_ZONE_COND_EMPTY: + alloc_offsets[i] = 0; + break; + case BLK_ZONE_COND_FULL: + alloc_offsets[i] = fs_info->zone_size; + break; + default: + /* Partially used zone */ + alloc_offsets[i] = + ((zone.wp - zone.start) << SECTOR_SHIFT); + break; + } + } + + if (num_sequential > 0) + cache->seq_zone = true; + + if (num_conventional > 0) { + /* + * Avoid calling calculate_alloc_pointer() for new BG. It + * is no use for new BG. It must be always 0. + * + * Also, we have a lock chain of extent buffer lock -> + * chunk mutex. For new BG, this function is called from + * btrfs_make_block_group() which is already taking the + * chunk mutex. Thus, we cannot call + * calculate_alloc_pointer() which takes extent buffer + * locks to avoid deadlock. + */ + if (new) { + cache->alloc_offset = 0; + goto out; + } + ret = calculate_alloc_pointer(cache, &last_alloc); + if (ret || map->num_stripes == num_conventional) { + if (!ret) + cache->alloc_offset = last_alloc; + else + btrfs_err(fs_info, + "zoned: failed to determine allocation offset of bg %llu", + cache->start); + goto out; + } + } + + switch (map->type & BTRFS_BLOCK_GROUP_PROFILE_MASK) { + case 0: /* single */ + cache->alloc_offset = alloc_offsets[0]; + break; + case BTRFS_BLOCK_GROUP_DUP: + case BTRFS_BLOCK_GROUP_RAID1: + case BTRFS_BLOCK_GROUP_RAID0: + case BTRFS_BLOCK_GROUP_RAID10: + case BTRFS_BLOCK_GROUP_RAID5: + case BTRFS_BLOCK_GROUP_RAID6: + /* non-single profiles are not supported yet */ + default: + btrfs_err(fs_info, "zoned: profile %s not yet supported", + btrfs_bg_type_to_raid_name(map->type)); + ret = -EINVAL; + goto out; + } + +out: + /* An extent is allocated after the write pointer */ + if (!ret && num_conventional && last_alloc > cache->alloc_offset) { + btrfs_err(fs_info, + "zoned: got wrong write pointer in BG %llu: %llu > %llu", + logical, last_alloc, cache->alloc_offset); + ret = -EIO; + } + + if (!ret) + cache->meta_write_pointer = cache->alloc_offset + cache->start; + + kfree(alloc_offsets); + free_extent_map(em); + + return ret; +} + +void btrfs_calc_zone_unusable(struct btrfs_block_group *cache) +{ + u64 unusable, free; + + if (!btrfs_is_zoned(cache->fs_info)) + return; + + WARN_ON(cache->bytes_super != 0); + unusable = cache->alloc_offset - cache->used; + free = cache->length - cache->alloc_offset; + + /* We only need ->free_space in ALLOC_SEQ block groups */ + cache->last_byte_to_unpin = (u64)-1; + cache->cached = BTRFS_CACHE_FINISHED; + cache->free_space_ctl->free_space = free; + cache->zone_unusable = unusable; + + /* Should not have any excluded extents. Just in case, though */ + btrfs_free_excluded_extents(cache); +} + +void btrfs_redirty_list_add(struct btrfs_transaction *trans, + struct extent_buffer *eb) +{ + struct btrfs_fs_info *fs_info = eb->fs_info; + + if (!btrfs_is_zoned(fs_info) || + btrfs_header_flag(eb, BTRFS_HEADER_FLAG_WRITTEN) || + !list_empty(&eb->release_list)) + return; + + set_extent_buffer_dirty(eb); + set_extent_bits_nowait(&trans->dirty_pages, eb->start, + eb->start + eb->len - 1, EXTENT_DIRTY); + memzero_extent_buffer(eb, 0, eb->len); + set_bit(EXTENT_BUFFER_NO_CHECK, &eb->bflags); + + spin_lock(&trans->releasing_ebs_lock); + list_add_tail(&eb->release_list, &trans->releasing_ebs); + spin_unlock(&trans->releasing_ebs_lock); + atomic_inc(&eb->refs); +} + +void btrfs_free_redirty_list(struct btrfs_transaction *trans) +{ + spin_lock(&trans->releasing_ebs_lock); + while (!list_empty(&trans->releasing_ebs)) { + struct extent_buffer *eb; + + eb = list_first_entry(&trans->releasing_ebs, + struct extent_buffer, release_list); + list_del_init(&eb->release_list); + free_extent_buffer(eb); + } + spin_unlock(&trans->releasing_ebs_lock); +} + +bool btrfs_use_zone_append(struct btrfs_inode *inode, struct extent_map *em) +{ + struct btrfs_fs_info *fs_info = inode->root->fs_info; + struct btrfs_block_group *cache; + bool ret = false; + + if (!btrfs_is_zoned(fs_info)) + return false; + + if (!fs_info->max_zone_append_size) + return false; + + if (!is_data_inode(&inode->vfs_inode)) + return false; + + cache = btrfs_lookup_block_group(fs_info, em->block_start); + ASSERT(cache); + if (!cache) + return false; + + ret = cache->seq_zone; + btrfs_put_block_group(cache); + + return ret; +} + +void btrfs_record_physical_zoned(struct inode *inode, u64 file_offset, + struct bio *bio) +{ + struct btrfs_ordered_extent *ordered; + const u64 physical = bio->bi_iter.bi_sector << SECTOR_SHIFT; + + if (bio_op(bio) != REQ_OP_ZONE_APPEND) + return; + + ordered = btrfs_lookup_ordered_extent(BTRFS_I(inode), file_offset); + if (WARN_ON(!ordered)) + return; + + ordered->physical = physical; + ordered->disk = bio->bi_bdev->bd_disk; + ordered->partno = bio->bi_bdev->bd_partno; + + btrfs_put_ordered_extent(ordered); +} + +void btrfs_rewrite_logical_zoned(struct btrfs_ordered_extent *ordered) +{ + struct btrfs_inode *inode = BTRFS_I(ordered->inode); + struct btrfs_fs_info *fs_info = inode->root->fs_info; + struct extent_map_tree *em_tree; + struct extent_map *em; + struct btrfs_ordered_sum *sum; + struct block_device *bdev; + u64 orig_logical = ordered->disk_bytenr; + u64 *logical = NULL; + int nr, stripe_len; + + /* Zoned devices should not have partitions. So, we can assume it is 0 */ + ASSERT(ordered->partno == 0); + bdev = bdgrab(ordered->disk->part0); + if (WARN_ON(!bdev)) + return; + + if (WARN_ON(btrfs_rmap_block(fs_info, orig_logical, bdev, + ordered->physical, &logical, &nr, + &stripe_len))) + goto out; + + WARN_ON(nr != 1); + + if (orig_logical == *logical) + goto out; + + ordered->disk_bytenr = *logical; + + em_tree = &inode->extent_tree; + write_lock(&em_tree->lock); + em = search_extent_mapping(em_tree, ordered->file_offset, + ordered->num_bytes); + em->block_start = *logical; + free_extent_map(em); + write_unlock(&em_tree->lock); + + list_for_each_entry(sum, &ordered->list, list) { + if (*logical < orig_logical) + sum->bytenr -= orig_logical - *logical; + else + sum->bytenr += *logical - orig_logical; + } + +out: + kfree(logical); + bdput(bdev); +} + +bool btrfs_check_meta_write_pointer(struct btrfs_fs_info *fs_info, + struct extent_buffer *eb, + struct btrfs_block_group **cache_ret) +{ + struct btrfs_block_group *cache; + bool ret = true; + + if (!btrfs_is_zoned(fs_info)) + return true; + + cache = *cache_ret; + + if (cache && (eb->start < cache->start || + cache->start + cache->length <= eb->start)) { + btrfs_put_block_group(cache); + cache = NULL; + *cache_ret = NULL; + } + + if (!cache) + cache = btrfs_lookup_block_group(fs_info, eb->start); + + if (cache) { + if (cache->meta_write_pointer != eb->start) { + btrfs_put_block_group(cache); + cache = NULL; + ret = false; + } else { + cache->meta_write_pointer = eb->start + eb->len; + } + + *cache_ret = cache; + } + + return ret; +} + +void btrfs_revert_meta_write_pointer(struct btrfs_block_group *cache, + struct extent_buffer *eb) +{ + if (!btrfs_is_zoned(eb->fs_info) || !cache) + return; + + ASSERT(cache->meta_write_pointer == eb->start + eb->len); + cache->meta_write_pointer = eb->start; +} + +int btrfs_zoned_issue_zeroout(struct btrfs_device *device, u64 physical, u64 length) +{ + if (!btrfs_dev_is_sequential(device, physical)) + return -EOPNOTSUPP; + + return blkdev_issue_zeroout(device->bdev, physical >> SECTOR_SHIFT, + length >> SECTOR_SHIFT, GFP_NOFS, 0); +} + +static int read_zone_info(struct btrfs_fs_info *fs_info, u64 logical, + struct blk_zone *zone) +{ + struct btrfs_bio *bbio = NULL; + u64 mapped_length = PAGE_SIZE; + unsigned int nofs_flag; + int nmirrors; + int i, ret; + + ret = btrfs_map_sblock(fs_info, BTRFS_MAP_GET_READ_MIRRORS, logical, + &mapped_length, &bbio); + if (ret || !bbio || mapped_length < PAGE_SIZE) { + btrfs_put_bbio(bbio); + return -EIO; + } + + if (bbio->map_type & BTRFS_BLOCK_GROUP_RAID56_MASK) + return -EINVAL; + + nofs_flag = memalloc_nofs_save(); + nmirrors = (int)bbio->num_stripes; + for (i = 0; i < nmirrors; i++) { + u64 physical = bbio->stripes[i].physical; + struct btrfs_device *dev = bbio->stripes[i].dev; + + /* Missing device */ + if (!dev->bdev) + continue; + + ret = btrfs_get_dev_zone(dev, physical, zone); + /* Failing device */ + if (ret == -EIO || ret == -EOPNOTSUPP) + continue; + break; + } + memalloc_nofs_restore(nofs_flag); + + return ret; +} + +/* + * Synchronize write pointer in a zone at @physical_start on @tgt_dev, by + * filling zeros between @physical_pos to a write pointer of dev-replace + * source device. + */ +int btrfs_sync_zone_write_pointer(struct btrfs_device *tgt_dev, u64 logical, + u64 physical_start, u64 physical_pos) +{ + struct btrfs_fs_info *fs_info = tgt_dev->fs_info; + struct blk_zone zone; + u64 length; + u64 wp; + int ret; + + if (!btrfs_dev_is_sequential(tgt_dev, physical_pos)) + return 0; + + ret = read_zone_info(fs_info, logical, &zone); + if (ret) + return ret; + + wp = physical_start + ((zone.wp - zone.start) << SECTOR_SHIFT); + + if (physical_pos == wp) + return 0; + + if (physical_pos > wp) + return -EUCLEAN; + + length = wp - physical_pos; + return btrfs_zoned_issue_zeroout(tgt_dev, physical_pos, length); +} diff --git a/fs/btrfs/zoned.h b/fs/btrfs/zoned.h index 8abe2f83272b..61e969652fe1 100644 --- a/fs/btrfs/zoned.h +++ b/fs/btrfs/zoned.h @@ -7,6 +7,7 @@ #include <linux/blkdev.h> #include "volumes.h" #include "disk-io.h" +#include "block-group.h" struct btrfs_zoned_device_info { /* @@ -25,6 +26,7 @@ struct btrfs_zoned_device_info { #ifdef CONFIG_BLK_DEV_ZONED int btrfs_get_dev_zone(struct btrfs_device *device, u64 pos, struct blk_zone *zone); +int btrfs_get_dev_zone_info_all_devices(struct btrfs_fs_info *fs_info); int btrfs_get_dev_zone_info(struct btrfs_device *device); void btrfs_destroy_dev_zone_info(struct btrfs_device *device); int btrfs_check_zoned_mode(struct btrfs_fs_info *fs_info); @@ -35,6 +37,28 @@ int btrfs_sb_log_location(struct btrfs_device *device, int mirror, int rw, u64 *bytenr_ret); void btrfs_advance_sb_log(struct btrfs_device *device, int mirror); int btrfs_reset_sb_log_zones(struct block_device *bdev, int mirror); +u64 btrfs_find_allocatable_zones(struct btrfs_device *device, u64 hole_start, + u64 hole_end, u64 num_bytes); +int btrfs_reset_device_zone(struct btrfs_device *device, u64 physical, + u64 length, u64 *bytes); +int btrfs_ensure_empty_zones(struct btrfs_device *device, u64 start, u64 size); +int btrfs_load_block_group_zone_info(struct btrfs_block_group *cache, bool new); +void btrfs_calc_zone_unusable(struct btrfs_block_group *cache); +void btrfs_redirty_list_add(struct btrfs_transaction *trans, + struct extent_buffer *eb); +void btrfs_free_redirty_list(struct btrfs_transaction *trans); +bool btrfs_use_zone_append(struct btrfs_inode *inode, struct extent_map *em); +void btrfs_record_physical_zoned(struct inode *inode, u64 file_offset, + struct bio *bio); +void btrfs_rewrite_logical_zoned(struct btrfs_ordered_extent *ordered); +bool btrfs_check_meta_write_pointer(struct btrfs_fs_info *fs_info, + struct extent_buffer *eb, + struct btrfs_block_group **cache_ret); +void btrfs_revert_meta_write_pointer(struct btrfs_block_group *cache, + struct extent_buffer *eb); +int btrfs_zoned_issue_zeroout(struct btrfs_device *device, u64 physical, u64 length); +int btrfs_sync_zone_write_pointer(struct btrfs_device *tgt_dev, u64 logical, + u64 physical_start, u64 physical_pos); #else /* CONFIG_BLK_DEV_ZONED */ static inline int btrfs_get_dev_zone(struct btrfs_device *device, u64 pos, struct blk_zone *zone) @@ -42,6 +66,11 @@ static inline int btrfs_get_dev_zone(struct btrfs_device *device, u64 pos, return 0; } +static inline int btrfs_get_dev_zone_info_all_devices(struct btrfs_fs_info *fs_info) +{ + return 0; +} + static inline int btrfs_get_dev_zone_info(struct btrfs_device *device) { return 0; @@ -85,6 +114,78 @@ static inline int btrfs_reset_sb_log_zones(struct block_device *bdev, int mirror return 0; } +static inline u64 btrfs_find_allocatable_zones(struct btrfs_device *device, + u64 hole_start, u64 hole_end, + u64 num_bytes) +{ + return hole_start; +} + +static inline int btrfs_reset_device_zone(struct btrfs_device *device, + u64 physical, u64 length, u64 *bytes) +{ + *bytes = 0; + return 0; +} + +static inline int btrfs_ensure_empty_zones(struct btrfs_device *device, + u64 start, u64 size) +{ + return 0; +} + +static inline int btrfs_load_block_group_zone_info( + struct btrfs_block_group *cache, bool new) +{ + return 0; +} + +static inline void btrfs_calc_zone_unusable(struct btrfs_block_group *cache) { } + +static inline void btrfs_redirty_list_add(struct btrfs_transaction *trans, + struct extent_buffer *eb) { } +static inline void btrfs_free_redirty_list(struct btrfs_transaction *trans) { } + +static inline bool btrfs_use_zone_append(struct btrfs_inode *inode, + struct extent_map *em) +{ + return false; +} + +static inline void btrfs_record_physical_zoned(struct inode *inode, + u64 file_offset, struct bio *bio) +{ +} + +static inline void btrfs_rewrite_logical_zoned( + struct btrfs_ordered_extent *ordered) { } + +static inline bool btrfs_check_meta_write_pointer(struct btrfs_fs_info *fs_info, + struct extent_buffer *eb, + struct btrfs_block_group **cache_ret) +{ + return true; +} + +static inline void btrfs_revert_meta_write_pointer( + struct btrfs_block_group *cache, + struct extent_buffer *eb) +{ +} + +static inline int btrfs_zoned_issue_zeroout(struct btrfs_device *device, + u64 physical, u64 length) +{ + return -EOPNOTSUPP; +} + +static inline int btrfs_sync_zone_write_pointer(struct btrfs_device *tgt_dev, + u64 logical, u64 physical_start, + u64 physical_pos) +{ + return -EOPNOTSUPP; +} + #endif static inline bool btrfs_dev_is_sequential(struct btrfs_device *device, u64 pos) @@ -136,12 +237,16 @@ static inline void btrfs_dev_clear_zone_empty(struct btrfs_device *device, u64 p static inline bool btrfs_check_device_zone_type(const struct btrfs_fs_info *fs_info, struct block_device *bdev) { - u64 zone_size; - if (btrfs_is_zoned(fs_info)) { - zone_size = bdev_zone_sectors(bdev) << SECTOR_SHIFT; - /* Do not allow non-zoned device */ - return bdev_is_zoned(bdev) && fs_info->zone_size == zone_size; + /* + * We can allow a regular device on a zoned filesystem, because + * we will emulate the zoned capabilities. + */ + if (!bdev_is_zoned(bdev)) + return true; + + return fs_info->zone_size == + (bdev_zone_sectors(bdev) << SECTOR_SHIFT); } /* Do not allow Host Manged zoned device */ @@ -157,4 +262,46 @@ static inline bool btrfs_check_super_location(struct btrfs_device *device, u64 p return device->zone_info == NULL || !btrfs_dev_is_sequential(device, pos); } +static inline bool btrfs_can_zone_reset(struct btrfs_device *device, + u64 physical, u64 length) +{ + u64 zone_size; + + if (!btrfs_dev_is_sequential(device, physical)) + return false; + + zone_size = device->zone_info->zone_size; + if (!IS_ALIGNED(physical, zone_size) || !IS_ALIGNED(length, zone_size)) + return false; + + return true; +} + +static inline void btrfs_zoned_meta_io_lock(struct btrfs_fs_info *fs_info) +{ + if (!btrfs_is_zoned(fs_info)) + return; + mutex_lock(&fs_info->zoned_meta_io_lock); +} + +static inline void btrfs_zoned_meta_io_unlock(struct btrfs_fs_info *fs_info) +{ + if (!btrfs_is_zoned(fs_info)) + return; + mutex_unlock(&fs_info->zoned_meta_io_lock); +} + +static inline void btrfs_clear_treelog_bg(struct btrfs_block_group *bg) +{ + struct btrfs_fs_info *fs_info = bg->fs_info; + + if (!btrfs_is_zoned(fs_info)) + return; + + spin_lock(&fs_info->treelog_bg_lock); + if (fs_info->treelog_bg == bg->start) + fs_info->treelog_bg = 0; + spin_unlock(&fs_info->treelog_bg_lock); +} + #endif diff --git a/fs/btrfs/zstd.c b/fs/btrfs/zstd.c index 9a4871636c6c..8e9626d63976 100644 --- a/fs/btrfs/zstd.c +++ b/fs/btrfs/zstd.c @@ -688,10 +688,8 @@ int zstd_decompress(struct list_head *ws, unsigned char *data_in, bytes = min_t(unsigned long, destlen - pg_offset, workspace->out_buf.size - buf_offset); - kaddr = kmap_atomic(dest_page); - memcpy(kaddr + pg_offset, workspace->out_buf.dst + buf_offset, - bytes); - kunmap_atomic(kaddr); + memcpy_to_page(dest_page, pg_offset, + workspace->out_buf.dst + buf_offset, bytes); pg_offset += bytes; } diff --git a/fs/buffer.c b/fs/buffer.c index 32647d2011df..0cb7ffd4977c 100644 --- a/fs/buffer.c +++ b/fs/buffer.c @@ -847,7 +847,8 @@ struct buffer_head *alloc_page_buffers(struct page *page, unsigned long size, if (retry) gfp |= __GFP_NOFAIL; - memcg = get_mem_cgroup_from_page(page); + /* The page lock pins the memcg */ + memcg = page_memcg(page); old_memcg = set_active_memcg(memcg); head = NULL; @@ -868,7 +869,6 @@ struct buffer_head *alloc_page_buffers(struct page *page, unsigned long size, } out: set_active_memcg(old_memcg); - mem_cgroup_put(memcg); return head; /* * In case anything failed, we just free everything we got. @@ -2083,7 +2083,8 @@ static int __block_commit_write(struct inode *inode, struct page *page, set_buffer_uptodate(bh); mark_buffer_dirty(bh); } - clear_buffer_new(bh); + if (buffer_new(bh)) + clear_buffer_new(bh); block_start = block_end; bh = bh->b_this_page; diff --git a/fs/cachefiles/interface.c b/fs/cachefiles/interface.c index 4cea5fbf695e..5efa6a3702c0 100644 --- a/fs/cachefiles/interface.c +++ b/fs/cachefiles/interface.c @@ -470,14 +470,14 @@ static int cachefiles_attr_changed(struct fscache_object *_object) _debug("discard tail %llx", oi_size); newattrs.ia_valid = ATTR_SIZE; newattrs.ia_size = oi_size & PAGE_MASK; - ret = notify_change(object->backer, &newattrs, NULL); + ret = notify_change(&init_user_ns, object->backer, &newattrs, NULL); if (ret < 0) goto truncate_failed; } newattrs.ia_valid = ATTR_SIZE; newattrs.ia_size = ni_size; - ret = notify_change(object->backer, &newattrs, NULL); + ret = notify_change(&init_user_ns, object->backer, &newattrs, NULL); truncate_failed: inode_unlock(d_inode(object->backer)); diff --git a/fs/cachefiles/namei.c b/fs/cachefiles/namei.c index ecc8ecbbfa5a..7bf0732ae25c 100644 --- a/fs/cachefiles/namei.c +++ b/fs/cachefiles/namei.c @@ -311,7 +311,8 @@ static int cachefiles_bury_object(struct cachefiles_cache *cache, cachefiles_io_error(cache, "Unlink security error"); } else { trace_cachefiles_unlink(object, rep, why); - ret = vfs_unlink(d_inode(dir), rep, NULL); + ret = vfs_unlink(&init_user_ns, d_inode(dir), rep, + NULL); if (preemptive) cachefiles_mark_object_buried(cache, rep, why); @@ -412,9 +413,16 @@ try_again: if (ret < 0) { cachefiles_io_error(cache, "Rename security error %d", ret); } else { + struct renamedata rd = { + .old_mnt_userns = &init_user_ns, + .old_dir = d_inode(dir), + .old_dentry = rep, + .new_mnt_userns = &init_user_ns, + .new_dir = d_inode(cache->graveyard), + .new_dentry = grave, + }; trace_cachefiles_rename(object, rep, grave, why); - ret = vfs_rename(d_inode(dir), rep, - d_inode(cache->graveyard), grave, NULL, 0); + ret = vfs_rename(&rd); if (ret != 0 && ret != -ENOMEM) cachefiles_io_error(cache, "Rename failed with error %d", ret); @@ -561,7 +569,7 @@ lookup_again: if (ret < 0) goto create_error; start = jiffies; - ret = vfs_mkdir(d_inode(dir), next, 0); + ret = vfs_mkdir(&init_user_ns, d_inode(dir), next, 0); cachefiles_hist(cachefiles_mkdir_histogram, start); if (!key) trace_cachefiles_mkdir(object, next, ret); @@ -597,7 +605,8 @@ lookup_again: if (ret < 0) goto create_error; start = jiffies; - ret = vfs_create(d_inode(dir), next, S_IFREG, true); + ret = vfs_create(&init_user_ns, d_inode(dir), next, + S_IFREG, true); cachefiles_hist(cachefiles_create_histogram, start); trace_cachefiles_create(object, next, ret); if (ret < 0) @@ -791,7 +800,7 @@ retry: ret = security_path_mkdir(&path, subdir, 0700); if (ret < 0) goto mkdir_error; - ret = vfs_mkdir(d_inode(dir), subdir, 0700); + ret = vfs_mkdir(&init_user_ns, d_inode(dir), subdir, 0700); if (ret < 0) goto mkdir_error; diff --git a/fs/cachefiles/xattr.c b/fs/cachefiles/xattr.c index 72e42438f3d7..a591b5e09637 100644 --- a/fs/cachefiles/xattr.c +++ b/fs/cachefiles/xattr.c @@ -39,8 +39,8 @@ int cachefiles_check_object_type(struct cachefiles_object *object) _enter("%p{%s}", object, type); /* attempt to install a type label directly */ - ret = vfs_setxattr(dentry, cachefiles_xattr_cache, type, 2, - XATTR_CREATE); + ret = vfs_setxattr(&init_user_ns, dentry, cachefiles_xattr_cache, type, + 2, XATTR_CREATE); if (ret == 0) { _debug("SET"); /* we succeeded */ goto error; @@ -54,7 +54,8 @@ int cachefiles_check_object_type(struct cachefiles_object *object) } /* read the current type label */ - ret = vfs_getxattr(dentry, cachefiles_xattr_cache, xtype, 3); + ret = vfs_getxattr(&init_user_ns, dentry, cachefiles_xattr_cache, xtype, + 3); if (ret < 0) { if (ret == -ERANGE) goto bad_type_length; @@ -110,9 +111,8 @@ int cachefiles_set_object_xattr(struct cachefiles_object *object, _debug("SET #%u", auxdata->len); clear_bit(FSCACHE_COOKIE_AUX_UPDATED, &object->fscache.cookie->flags); - ret = vfs_setxattr(dentry, cachefiles_xattr_cache, - &auxdata->type, auxdata->len, - XATTR_CREATE); + ret = vfs_setxattr(&init_user_ns, dentry, cachefiles_xattr_cache, + &auxdata->type, auxdata->len, XATTR_CREATE); if (ret < 0 && ret != -ENOMEM) cachefiles_io_error_obj( object, @@ -140,9 +140,8 @@ int cachefiles_update_object_xattr(struct cachefiles_object *object, _debug("SET #%u", auxdata->len); clear_bit(FSCACHE_COOKIE_AUX_UPDATED, &object->fscache.cookie->flags); - ret = vfs_setxattr(dentry, cachefiles_xattr_cache, - &auxdata->type, auxdata->len, - XATTR_REPLACE); + ret = vfs_setxattr(&init_user_ns, dentry, cachefiles_xattr_cache, + &auxdata->type, auxdata->len, XATTR_REPLACE); if (ret < 0 && ret != -ENOMEM) cachefiles_io_error_obj( object, @@ -171,7 +170,7 @@ int cachefiles_check_auxdata(struct cachefiles_object *object) if (!auxbuf) return -ENOMEM; - xlen = vfs_getxattr(dentry, cachefiles_xattr_cache, + xlen = vfs_getxattr(&init_user_ns, dentry, cachefiles_xattr_cache, &auxbuf->type, 512 + 1); ret = -ESTALE; if (xlen < 1 || @@ -213,7 +212,7 @@ int cachefiles_check_object_xattr(struct cachefiles_object *object, } /* read the current type label */ - ret = vfs_getxattr(dentry, cachefiles_xattr_cache, + ret = vfs_getxattr(&init_user_ns, dentry, cachefiles_xattr_cache, &auxbuf->type, 512 + 1); if (ret < 0) { if (ret == -ENODATA) @@ -270,9 +269,9 @@ int cachefiles_check_object_xattr(struct cachefiles_object *object, } /* update the current label */ - ret = vfs_setxattr(dentry, cachefiles_xattr_cache, - &auxdata->type, auxdata->len, - XATTR_REPLACE); + ret = vfs_setxattr(&init_user_ns, dentry, + cachefiles_xattr_cache, &auxdata->type, + auxdata->len, XATTR_REPLACE); if (ret < 0) { cachefiles_io_error_obj(object, "Can't update xattr on %lu" @@ -309,7 +308,7 @@ int cachefiles_remove_object_xattr(struct cachefiles_cache *cache, { int ret; - ret = vfs_removexattr(dentry, cachefiles_xattr_cache); + ret = vfs_removexattr(&init_user_ns, dentry, cachefiles_xattr_cache); if (ret < 0) { if (ret == -ENOENT || ret == -ENODATA) ret = 0; diff --git a/fs/ceph/acl.c b/fs/ceph/acl.c index e0465741c591..529af59d9fd3 100644 --- a/fs/ceph/acl.c +++ b/fs/ceph/acl.c @@ -82,7 +82,8 @@ retry: return acl; } -int ceph_set_acl(struct inode *inode, struct posix_acl *acl, int type) +int ceph_set_acl(struct user_namespace *mnt_userns, struct inode *inode, + struct posix_acl *acl, int type) { int ret = 0, size = 0; const char *name = NULL; @@ -100,7 +101,8 @@ int ceph_set_acl(struct inode *inode, struct posix_acl *acl, int type) case ACL_TYPE_ACCESS: name = XATTR_NAME_POSIX_ACL_ACCESS; if (acl) { - ret = posix_acl_update_mode(inode, &new_mode, &acl); + ret = posix_acl_update_mode(&init_user_ns, inode, + &new_mode, &acl); if (ret) goto out; } diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c index 950552944436..26e66436f005 100644 --- a/fs/ceph/addr.c +++ b/fs/ceph/addr.c @@ -1662,7 +1662,7 @@ static vm_fault_t ceph_page_mkwrite(struct vm_fault *vmf) dout("page_mkwrite %p %llu~%zd dropping cap refs on %s ret %x\n", inode, off, len, ceph_cap_string(got), ret); - ceph_put_cap_refs(ci, got); + ceph_put_cap_refs_async(ci, got); out_free: ceph_restore_sigs(&oldset); sb_end_pagefault(inode->i_sb); diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c index 255a512f1277..570731c4d019 100644 --- a/fs/ceph/caps.c +++ b/fs/ceph/caps.c @@ -3027,6 +3027,12 @@ static int ceph_try_drop_cap_snap(struct ceph_inode_info *ci, return 0; } +enum put_cap_refs_mode { + PUT_CAP_REFS_SYNC = 0, + PUT_CAP_REFS_NO_CHECK, + PUT_CAP_REFS_ASYNC, +}; + /* * Release cap refs. * @@ -3037,10 +3043,11 @@ static int ceph_try_drop_cap_snap(struct ceph_inode_info *ci, * cap_snap, and wake up any waiters. */ static void __ceph_put_cap_refs(struct ceph_inode_info *ci, int had, - bool skip_checking_caps) + enum put_cap_refs_mode mode) { struct inode *inode = &ci->vfs_inode; int last = 0, put = 0, flushsnaps = 0, wake = 0; + bool check_flushsnaps = false; spin_lock(&ci->i_ceph_lock); if (had & CEPH_CAP_PIN) @@ -3057,26 +3064,17 @@ static void __ceph_put_cap_refs(struct ceph_inode_info *ci, int had, if (had & CEPH_CAP_FILE_BUFFER) { if (--ci->i_wb_ref == 0) { last++; + /* put the ref held by ceph_take_cap_refs() */ put++; + check_flushsnaps = true; } dout("put_cap_refs %p wb %d -> %d (?)\n", inode, ci->i_wb_ref+1, ci->i_wb_ref); } - if (had & CEPH_CAP_FILE_WR) + if (had & CEPH_CAP_FILE_WR) { if (--ci->i_wr_ref == 0) { last++; - if (__ceph_have_pending_cap_snap(ci)) { - struct ceph_cap_snap *capsnap = - list_last_entry(&ci->i_cap_snaps, - struct ceph_cap_snap, - ci_item); - capsnap->writing = 0; - if (ceph_try_drop_cap_snap(ci, capsnap)) - put++; - else if (__ceph_finish_cap_snap(ci, capsnap)) - flushsnaps = 1; - wake = 1; - } + check_flushsnaps = true; if (ci->i_wrbuffer_ref_head == 0 && ci->i_dirty_caps == 0 && ci->i_flushing_caps == 0) { @@ -3088,15 +3086,42 @@ static void __ceph_put_cap_refs(struct ceph_inode_info *ci, int had, if (!__ceph_is_any_real_caps(ci) && ci->i_snap_realm) drop_inode_snap_realm(ci); } + } + if (check_flushsnaps && __ceph_have_pending_cap_snap(ci)) { + struct ceph_cap_snap *capsnap = + list_last_entry(&ci->i_cap_snaps, + struct ceph_cap_snap, + ci_item); + + capsnap->writing = 0; + if (ceph_try_drop_cap_snap(ci, capsnap)) + /* put the ref held by ceph_queue_cap_snap() */ + put++; + else if (__ceph_finish_cap_snap(ci, capsnap)) + flushsnaps = 1; + wake = 1; + } spin_unlock(&ci->i_ceph_lock); dout("put_cap_refs %p had %s%s%s\n", inode, ceph_cap_string(had), last ? " last" : "", put ? " put" : ""); - if (last && !skip_checking_caps) - ceph_check_caps(ci, 0, NULL); - else if (flushsnaps) - ceph_flush_snaps(ci, NULL); + switch (mode) { + case PUT_CAP_REFS_SYNC: + if (last) + ceph_check_caps(ci, 0, NULL); + else if (flushsnaps) + ceph_flush_snaps(ci, NULL); + break; + case PUT_CAP_REFS_ASYNC: + if (last) + ceph_queue_check_caps(inode); + else if (flushsnaps) + ceph_queue_flush_snaps(inode); + break; + default: + break; + } if (wake) wake_up_all(&ci->i_cap_wq); while (put-- > 0) @@ -3105,12 +3130,17 @@ static void __ceph_put_cap_refs(struct ceph_inode_info *ci, int had, void ceph_put_cap_refs(struct ceph_inode_info *ci, int had) { - __ceph_put_cap_refs(ci, had, false); + __ceph_put_cap_refs(ci, had, PUT_CAP_REFS_SYNC); +} + +void ceph_put_cap_refs_async(struct ceph_inode_info *ci, int had) +{ + __ceph_put_cap_refs(ci, had, PUT_CAP_REFS_ASYNC); } void ceph_put_cap_refs_no_check_caps(struct ceph_inode_info *ci, int had) { - __ceph_put_cap_refs(ci, had, true); + __ceph_put_cap_refs(ci, had, PUT_CAP_REFS_NO_CHECK); } /* diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c index 858ee7362ff5..83d9358854fb 100644 --- a/fs/ceph/dir.c +++ b/fs/ceph/dir.c @@ -823,8 +823,8 @@ int ceph_handle_notrace_create(struct inode *dir, struct dentry *dentry) return PTR_ERR(result); } -static int ceph_mknod(struct inode *dir, struct dentry *dentry, - umode_t mode, dev_t rdev) +static int ceph_mknod(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, dev_t rdev) { struct ceph_mds_client *mdsc = ceph_sb_to_mdsc(dir->i_sb); struct ceph_mds_request *req; @@ -878,14 +878,14 @@ out: return err; } -static int ceph_create(struct inode *dir, struct dentry *dentry, umode_t mode, - bool excl) +static int ceph_create(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, bool excl) { - return ceph_mknod(dir, dentry, mode, 0); + return ceph_mknod(mnt_userns, dir, dentry, mode, 0); } -static int ceph_symlink(struct inode *dir, struct dentry *dentry, - const char *dest) +static int ceph_symlink(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, const char *dest) { struct ceph_mds_client *mdsc = ceph_sb_to_mdsc(dir->i_sb); struct ceph_mds_request *req; @@ -937,7 +937,8 @@ out: return err; } -static int ceph_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) +static int ceph_mkdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { struct ceph_mds_client *mdsc = ceph_sb_to_mdsc(dir->i_sb); struct ceph_mds_request *req; @@ -1183,9 +1184,9 @@ out: return err; } -static int ceph_rename(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry, - unsigned int flags) +static int ceph_rename(struct user_namespace *mnt_userns, struct inode *old_dir, + struct dentry *old_dentry, struct inode *new_dir, + struct dentry *new_dentry, unsigned int flags) { struct ceph_mds_client *mdsc = ceph_sb_to_mdsc(old_dir->i_sb); struct ceph_mds_request *req; diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c index adc8fc3c5d85..156f849f5385 100644 --- a/fs/ceph/inode.c +++ b/fs/ceph/inode.c @@ -1816,60 +1816,17 @@ void ceph_async_iput(struct inode *inode) } } -/* - * Write back inode data in a worker thread. (This can't be done - * in the message handler context.) - */ -void ceph_queue_writeback(struct inode *inode) -{ - struct ceph_inode_info *ci = ceph_inode(inode); - set_bit(CEPH_I_WORK_WRITEBACK, &ci->i_work_mask); - - ihold(inode); - if (queue_work(ceph_inode_to_client(inode)->inode_wq, - &ci->i_work)) { - dout("ceph_queue_writeback %p\n", inode); - } else { - dout("ceph_queue_writeback %p already queued, mask=%lx\n", - inode, ci->i_work_mask); - iput(inode); - } -} - -/* - * queue an async invalidation - */ -void ceph_queue_invalidate(struct inode *inode) -{ - struct ceph_inode_info *ci = ceph_inode(inode); - set_bit(CEPH_I_WORK_INVALIDATE_PAGES, &ci->i_work_mask); - - ihold(inode); - if (queue_work(ceph_inode_to_client(inode)->inode_wq, - &ceph_inode(inode)->i_work)) { - dout("ceph_queue_invalidate %p\n", inode); - } else { - dout("ceph_queue_invalidate %p already queued, mask=%lx\n", - inode, ci->i_work_mask); - iput(inode); - } -} - -/* - * Queue an async vmtruncate. If we fail to queue work, we will handle - * the truncation the next time we call __ceph_do_pending_vmtruncate. - */ -void ceph_queue_vmtruncate(struct inode *inode) +void ceph_queue_inode_work(struct inode *inode, int work_bit) { + struct ceph_fs_client *fsc = ceph_inode_to_client(inode); struct ceph_inode_info *ci = ceph_inode(inode); - set_bit(CEPH_I_WORK_VMTRUNCATE, &ci->i_work_mask); + set_bit(work_bit, &ci->i_work_mask); ihold(inode); - if (queue_work(ceph_inode_to_client(inode)->inode_wq, - &ci->i_work)) { - dout("ceph_queue_vmtruncate %p\n", inode); + if (queue_work(fsc->inode_wq, &ci->i_work)) { + dout("queue_inode_work %p, mask=%lx\n", inode, ci->i_work_mask); } else { - dout("ceph_queue_vmtruncate %p already queued, mask=%lx\n", + dout("queue_inode_work %p already queued, mask=%lx\n", inode, ci->i_work_mask); iput(inode); } @@ -2008,6 +1965,12 @@ static void ceph_inode_work(struct work_struct *work) if (test_and_clear_bit(CEPH_I_WORK_VMTRUNCATE, &ci->i_work_mask)) __ceph_do_pending_vmtruncate(inode); + if (test_and_clear_bit(CEPH_I_WORK_CHECK_CAPS, &ci->i_work_mask)) + ceph_check_caps(ci, 0, NULL); + + if (test_and_clear_bit(CEPH_I_WORK_FLUSH_SNAPS, &ci->i_work_mask)) + ceph_flush_snaps(ci, NULL); + iput(inode); } @@ -2238,7 +2201,8 @@ int __ceph_setattr(struct inode *inode, struct iattr *attr) /* * setattr */ -int ceph_setattr(struct dentry *dentry, struct iattr *attr) +int ceph_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *attr) { struct inode *inode = d_inode(dentry); struct ceph_fs_client *fsc = ceph_inode_to_client(inode); @@ -2247,7 +2211,7 @@ int ceph_setattr(struct dentry *dentry, struct iattr *attr) if (ceph_snap(inode) != CEPH_NOSNAP) return -EROFS; - err = setattr_prepare(dentry, attr); + err = setattr_prepare(&init_user_ns, dentry, attr); if (err != 0) return err; @@ -2262,7 +2226,7 @@ int ceph_setattr(struct dentry *dentry, struct iattr *attr) err = __ceph_setattr(inode, attr); if (err >= 0 && (attr->ia_valid & ATTR_MODE)) - err = posix_acl_chmod(inode, attr->ia_mode); + err = posix_acl_chmod(&init_user_ns, inode, attr->ia_mode); return err; } @@ -2321,7 +2285,8 @@ int __ceph_do_getattr(struct inode *inode, struct page *locked_page, * Check inode permissions. We verify we have a valid value for * the AUTH cap, then call the generic handler. */ -int ceph_permission(struct inode *inode, int mask) +int ceph_permission(struct user_namespace *mnt_userns, struct inode *inode, + int mask) { int err; @@ -2331,7 +2296,7 @@ int ceph_permission(struct inode *inode, int mask) err = ceph_do_getattr(inode, CEPH_CAP_AUTH_SHARED, false); if (!err) - err = generic_permission(inode, mask); + err = generic_permission(&init_user_ns, inode, mask); return err; } @@ -2368,8 +2333,8 @@ static int statx_to_caps(u32 want, umode_t mode) * Get all the attributes. If we have sufficient caps for the requested attrs, * then we can avoid talking to the MDS at all. */ -int ceph_getattr(const struct path *path, struct kstat *stat, - u32 request_mask, unsigned int flags) +int ceph_getattr(struct user_namespace *mnt_userns, const struct path *path, + struct kstat *stat, u32 request_mask, unsigned int flags) { struct inode *inode = d_inode(path->dentry); struct ceph_inode_info *ci = ceph_inode(inode); @@ -2385,7 +2350,7 @@ int ceph_getattr(const struct path *path, struct kstat *stat, return err; } - generic_fillattr(inode, stat); + generic_fillattr(&init_user_ns, inode, stat); stat->ino = ceph_present_inode(inode); /* diff --git a/fs/ceph/snap.c b/fs/ceph/snap.c index b611f829cb61..0728b01d4d43 100644 --- a/fs/ceph/snap.c +++ b/fs/ceph/snap.c @@ -623,6 +623,16 @@ int __ceph_finish_cap_snap(struct ceph_inode_info *ci, return 0; } + /* Fb cap still in use, delay it */ + if (ci->i_wb_ref) { + dout("finish_cap_snap %p cap_snap %p snapc %p %llu %s s=%llu " + "used WRBUFFER, delaying\n", inode, capsnap, + capsnap->context, capsnap->context->seq, + ceph_cap_string(capsnap->dirty), capsnap->size); + capsnap->writing = 1; + return 0; + } + ci->i_ceph_flags |= CEPH_I_FLUSH_SNAPS; dout("finish_cap_snap %p cap_snap %p snapc %p %llu %s s=%llu\n", inode, capsnap, capsnap->context, diff --git a/fs/ceph/super.h b/fs/ceph/super.h index b62d8fee3b86..c48bb30c8d70 100644 --- a/fs/ceph/super.h +++ b/fs/ceph/super.h @@ -562,9 +562,11 @@ static inline struct inode *ceph_find_inode(struct super_block *sb, /* * Masks of ceph inode work. */ -#define CEPH_I_WORK_WRITEBACK 0 /* writeback */ -#define CEPH_I_WORK_INVALIDATE_PAGES 1 /* invalidate pages */ -#define CEPH_I_WORK_VMTRUNCATE 2 /* vmtruncate */ +#define CEPH_I_WORK_WRITEBACK 0 +#define CEPH_I_WORK_INVALIDATE_PAGES 1 +#define CEPH_I_WORK_VMTRUNCATE 2 +#define CEPH_I_WORK_CHECK_CAPS 3 +#define CEPH_I_WORK_FLUSH_SNAPS 4 /* * We set the ERROR_WRITE bit when we start seeing write errors on an inode @@ -962,21 +964,49 @@ extern int ceph_inode_holds_cap(struct inode *inode, int mask); extern bool ceph_inode_set_size(struct inode *inode, loff_t size); extern void __ceph_do_pending_vmtruncate(struct inode *inode); -extern void ceph_queue_vmtruncate(struct inode *inode); -extern void ceph_queue_invalidate(struct inode *inode); -extern void ceph_queue_writeback(struct inode *inode); + extern void ceph_async_iput(struct inode *inode); +void ceph_queue_inode_work(struct inode *inode, int work_bit); + +static inline void ceph_queue_vmtruncate(struct inode *inode) +{ + ceph_queue_inode_work(inode, CEPH_I_WORK_VMTRUNCATE); +} + +static inline void ceph_queue_invalidate(struct inode *inode) +{ + ceph_queue_inode_work(inode, CEPH_I_WORK_INVALIDATE_PAGES); +} + +static inline void ceph_queue_writeback(struct inode *inode) +{ + ceph_queue_inode_work(inode, CEPH_I_WORK_WRITEBACK); +} + +static inline void ceph_queue_check_caps(struct inode *inode) +{ + ceph_queue_inode_work(inode, CEPH_I_WORK_CHECK_CAPS); +} + +static inline void ceph_queue_flush_snaps(struct inode *inode) +{ + ceph_queue_inode_work(inode, CEPH_I_WORK_FLUSH_SNAPS); +} + extern int __ceph_do_getattr(struct inode *inode, struct page *locked_page, int mask, bool force); static inline int ceph_do_getattr(struct inode *inode, int mask, bool force) { return __ceph_do_getattr(inode, NULL, mask, force); } -extern int ceph_permission(struct inode *inode, int mask); +extern int ceph_permission(struct user_namespace *mnt_userns, + struct inode *inode, int mask); extern int __ceph_setattr(struct inode *inode, struct iattr *attr); -extern int ceph_setattr(struct dentry *dentry, struct iattr *attr); -extern int ceph_getattr(const struct path *path, struct kstat *stat, +extern int ceph_setattr(struct user_namespace *mnt_userns, + struct dentry *dentry, struct iattr *attr); +extern int ceph_getattr(struct user_namespace *mnt_userns, + const struct path *path, struct kstat *stat, u32 request_mask, unsigned int flags); /* xattr.c */ @@ -1037,7 +1067,8 @@ void ceph_release_acl_sec_ctx(struct ceph_acl_sec_ctx *as_ctx); #ifdef CONFIG_CEPH_FS_POSIX_ACL struct posix_acl *ceph_get_acl(struct inode *, int); -int ceph_set_acl(struct inode *inode, struct posix_acl *acl, int type); +int ceph_set_acl(struct user_namespace *mnt_userns, + struct inode *inode, struct posix_acl *acl, int type); int ceph_pre_init_acls(struct inode *dir, umode_t *mode, struct ceph_acl_sec_ctx *as_ctx); void ceph_init_inode_acls(struct inode *inode, @@ -1105,6 +1136,7 @@ extern void ceph_take_cap_refs(struct ceph_inode_info *ci, int caps, bool snap_rwsem_locked); extern void ceph_get_cap_refs(struct ceph_inode_info *ci, int caps); extern void ceph_put_cap_refs(struct ceph_inode_info *ci, int had); +extern void ceph_put_cap_refs_async(struct ceph_inode_info *ci, int had); extern void ceph_put_cap_refs_no_check_caps(struct ceph_inode_info *ci, int had); extern void ceph_put_wrbuffer_cap_refs(struct ceph_inode_info *ci, int nr, diff --git a/fs/ceph/xattr.c b/fs/ceph/xattr.c index 24997982de01..02f59bcb4f27 100644 --- a/fs/ceph/xattr.c +++ b/fs/ceph/xattr.c @@ -1238,6 +1238,7 @@ static int ceph_get_xattr_handler(const struct xattr_handler *handler, } static int ceph_set_xattr_handler(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *unused, struct inode *inode, const char *name, const void *value, size_t size, int flags) diff --git a/fs/cifs/cifs_debug.c b/fs/cifs/cifs_debug.c index b231dcf1d1f9..88a7958170ee 100644 --- a/fs/cifs/cifs_debug.c +++ b/fs/cifs/cifs_debug.c @@ -133,11 +133,12 @@ cifs_dump_channel(struct seq_file *m, int i, struct cifs_chan *chan) { struct TCP_Server_Info *server = chan->server; - seq_printf(m, "\t\tChannel %d Number of credits: %d Dialect 0x%x " - "TCP status: %d Instance: %d Local Users To Server: %d " - "SecMode: 0x%x Req On Wire: %d In Send: %d " - "In MaxReq Wait: %d\n", - i+1, + seq_printf(m, "\n\n\t\tChannel: %d ConnectionId: 0x%llx" + "\n\t\tNumber of credits: %d Dialect 0x%x" + "\n\t\tTCP status: %d Instance: %d" + "\n\t\tLocal Users To Server: %d SecMode: 0x%x Req On Wire: %d" + "\n\t\tIn Send: %d In MaxReq Wait: %d", + i+1, server->conn_id, server->credits, server->dialect, server->tcpStatus, @@ -197,16 +198,16 @@ static int cifs_debug_files_proc_show(struct seq_file *m, void *v) cfile = list_entry(tmp2, struct cifsFileInfo, tlist); seq_printf(m, - "0x%x 0x%llx 0x%x %d %d %d %s", + "0x%x 0x%llx 0x%x %d %d %d %pd", tcon->tid, cfile->fid.persistent_fid, cfile->f_flags, cfile->count, cfile->pid, from_kuid(&init_user_ns, cfile->uid), - cfile->dentry->d_name.name); + cfile->dentry); #ifdef CONFIG_CIFS_DEBUG2 - seq_printf(m, " 0x%llx\n", cfile->fid.mid); + seq_printf(m, " %llu\n", cfile->fid.mid); #else seq_printf(m, "\n"); #endif /* CIFS_DEBUG2 */ @@ -227,7 +228,7 @@ static int cifs_debug_data_proc_show(struct seq_file *m, void *v) struct TCP_Server_Info *server; struct cifs_ses *ses; struct cifs_tcon *tcon; - int i, j; + int c, i, j; seq_puts(m, "Display Internal CIFS Data Structures for Debugging\n" @@ -275,14 +276,25 @@ static int cifs_debug_data_proc_show(struct seq_file *m, void *v) seq_putc(m, '\n'); seq_printf(m, "CIFSMaxBufSize: %d\n", CIFSMaxBufSize); seq_printf(m, "Active VFS Requests: %d\n", GlobalTotalActiveXid); - seq_printf(m, "Servers:"); - i = 0; + seq_printf(m, "\nServers: "); + + c = 0; spin_lock(&cifs_tcp_ses_lock); list_for_each(tmp1, &cifs_tcp_ses_list) { server = list_entry(tmp1, struct TCP_Server_Info, tcp_ses_list); + /* channel info will be printed as a part of sessions below */ + if (server->is_channel) + continue; + + c++; + seq_printf(m, "\n%d) ConnectionId: 0x%llx ", + c, server->conn_id); + + if (server->hostname) + seq_printf(m, "Hostname: %s ", server->hostname); #ifdef CONFIG_CIFS_SMB_DIRECT if (!server->rdma) goto skip_rdma; @@ -362,46 +374,48 @@ skip_rdma: if (server->posix_ext_supported) seq_printf(m, " posix"); - i++; + if (server->rdma) + seq_printf(m, "\nRDMA "); + seq_printf(m, "\nTCP status: %d Instance: %d" + "\nLocal Users To Server: %d SecMode: 0x%x Req On Wire: %d", + server->tcpStatus, + server->reconnect_instance, + server->srv_count, + server->sec_mode, in_flight(server)); + + seq_printf(m, "\nIn Send: %d In MaxReq Wait: %d", + atomic_read(&server->in_send), + atomic_read(&server->num_waiters)); + + seq_printf(m, "\n\n\tSessions: "); + i = 0; list_for_each(tmp2, &server->smb_ses_list) { ses = list_entry(tmp2, struct cifs_ses, smb_ses_list); + i++; if ((ses->serverDomain == NULL) || (ses->serverOS == NULL) || (ses->serverNOS == NULL)) { - seq_printf(m, "\n%d) Name: %s Uses: %d Capability: 0x%x\tSession Status: %d ", - i, ses->serverName, ses->ses_count, + seq_printf(m, "\n\t%d) Address: %s Uses: %d Capability: 0x%x\tSession Status: %d ", + i, ses->ip_addr, ses->ses_count, ses->capabilities, ses->status); if (ses->session_flags & SMB2_SESSION_FLAG_IS_GUEST) - seq_printf(m, "Guest\t"); + seq_printf(m, "Guest "); else if (ses->session_flags & SMB2_SESSION_FLAG_IS_NULL) - seq_printf(m, "Anonymous\t"); + seq_printf(m, "Anonymous "); } else { seq_printf(m, - "\n%d) Name: %s Domain: %s Uses: %d OS:" - " %s\n\tNOS: %s\tCapability: 0x%x\n\tSMB" - " session status: %d ", - i, ses->serverName, ses->serverDomain, + "\n\t%d) Name: %s Domain: %s Uses: %d OS: %s " + "\n\tNOS: %s\tCapability: 0x%x" + "\n\tSMB session status: %d ", + i, ses->ip_addr, ses->serverDomain, ses->ses_count, ses->serverOS, ses->serverNOS, ses->capabilities, ses->status); } - seq_printf(m,"Security type: %s\n", + seq_printf(m, "\n\tSecurity type: %s ", get_security_type_str(server->ops->select_sectype(server, ses->sectype))); - if (server->rdma) - seq_printf(m, "RDMA\n\t"); - seq_printf(m, "TCP status: %d Instance: %d\n\tLocal Users To " - "Server: %d SecMode: 0x%x Req On Wire: %d", - server->tcpStatus, - server->reconnect_instance, - server->srv_count, - server->sec_mode, in_flight(server)); - - seq_printf(m, " In Send: %d In MaxReq Wait: %d", - atomic_read(&server->in_send), - atomic_read(&server->num_waiters)); - /* dump session id helpful for use with network trace */ seq_printf(m, " SessionId: 0x%llx", ses->Suid); if (ses->session_flags & SMB2_SESSION_FLAG_ENCRYPT_DATA) @@ -414,13 +428,13 @@ skip_rdma: from_kuid(&init_user_ns, ses->cred_uid)); if (ses->chan_count > 1) { - seq_printf(m, "\n\n\tExtra Channels: %zu\n", + seq_printf(m, "\n\n\tExtra Channels: %zu ", ses->chan_count-1); for (j = 1; j < ses->chan_count; j++) cifs_dump_channel(m, j, &ses->chans[j]); } - seq_puts(m, "\n\n\tShares:"); + seq_puts(m, "\n\n\tShares: "); j = 0; seq_printf(m, "\n\t%d) IPC: ", j); @@ -437,38 +451,43 @@ skip_rdma: cifs_debug_tcon(m, tcon); } - seq_puts(m, "\n\tMIDs:\n"); - - spin_lock(&GlobalMid_Lock); - list_for_each(tmp3, &server->pending_mid_q) { - mid_entry = list_entry(tmp3, struct mid_q_entry, - qhead); - seq_printf(m, "\tState: %d com: %d pid:" - " %d cbdata: %p mid %llu\n", - mid_entry->mid_state, - le16_to_cpu(mid_entry->command), - mid_entry->pid, - mid_entry->callback_data, - mid_entry->mid); - } - spin_unlock(&GlobalMid_Lock); - spin_lock(&ses->iface_lock); if (ses->iface_count) - seq_printf(m, "\n\tServer interfaces: %zu\n", + seq_printf(m, "\n\n\tServer interfaces: %zu", ses->iface_count); for (j = 0; j < ses->iface_count; j++) { struct cifs_server_iface *iface; iface = &ses->iface_list[j]; - seq_printf(m, "\t%d)", j); + seq_printf(m, "\n\t%d)", j+1); cifs_dump_iface(m, iface); if (is_ses_using_iface(ses, iface)) seq_puts(m, "\t\t[CONNECTED]\n"); } spin_unlock(&ses->iface_lock); } + if (i == 0) + seq_printf(m, "\n\t\t[NONE]"); + + seq_puts(m, "\n\n\tMIDs: "); + spin_lock(&GlobalMid_Lock); + list_for_each(tmp3, &server->pending_mid_q) { + mid_entry = list_entry(tmp3, struct mid_q_entry, + qhead); + seq_printf(m, "\n\tState: %d com: %d pid:" + " %d cbdata: %p mid %llu\n", + mid_entry->mid_state, + le16_to_cpu(mid_entry->command), + mid_entry->pid, + mid_entry->callback_data, + mid_entry->mid); + } + spin_unlock(&GlobalMid_Lock); + seq_printf(m, "\n--\n"); } + if (c == 0) + seq_printf(m, "\n\t[NONE]"); + spin_unlock(&cifs_tcp_ses_lock); seq_putc(m, '\n'); diff --git a/fs/cifs/cifs_swn.c b/fs/cifs/cifs_swn.c index d35f599aa00e..d829b8bf833e 100644 --- a/fs/cifs/cifs_swn.c +++ b/fs/cifs/cifs_swn.c @@ -248,7 +248,7 @@ nlmsg_fail: /* * Try to find a matching registration for the tcon's server name and share name. - * Calls to this funciton must be protected by cifs_swnreg_idr_mutex. + * Calls to this function must be protected by cifs_swnreg_idr_mutex. * TODO Try to avoid memory allocations */ static struct cifs_swn_reg *cifs_find_swn_reg(struct cifs_tcon *tcon) @@ -272,7 +272,7 @@ static struct cifs_swn_reg *cifs_find_swn_reg(struct cifs_tcon *tcon) if (IS_ERR(share_name)) { int ret; - ret = PTR_ERR(net_name); + ret = PTR_ERR(share_name); cifs_dbg(VFS, "%s: failed to extract share name from target '%s': %d\n", __func__, tcon->treeName, ret); kfree(net_name); diff --git a/fs/cifs/cifsacl.c b/fs/cifs/cifsacl.c index 562913e2b3f2..2be22a5c690f 100644 --- a/fs/cifs/cifsacl.c +++ b/fs/cifs/cifsacl.c @@ -267,10 +267,11 @@ is_well_known_sid(const struct cifs_sid *psid, uint32_t *puid, bool is_group) return true; /* well known sid found, uid returned */ } -static void +static __u16 cifs_copy_sid(struct cifs_sid *dst, const struct cifs_sid *src) { int i; + __u16 size = 1 + 1 + 6; dst->revision = src->revision; dst->num_subauth = min_t(u8, src->num_subauth, SID_MAX_SUB_AUTHORITIES); @@ -278,6 +279,9 @@ cifs_copy_sid(struct cifs_sid *dst, const struct cifs_sid *src) dst->authority[i] = src->authority[i]; for (i = 0; i < dst->num_subauth; ++i) dst->sub_auth[i] = src->sub_auth[i]; + size += (dst->num_subauth * 4); + + return size; } static int @@ -521,8 +525,11 @@ exit_cifs_idmap(void) } /* copy ntsd, owner sid, and group sid from a security descriptor to another */ -static void copy_sec_desc(const struct cifs_ntsd *pntsd, - struct cifs_ntsd *pnntsd, __u32 sidsoffset) +static __u32 copy_sec_desc(const struct cifs_ntsd *pntsd, + struct cifs_ntsd *pnntsd, + __u32 sidsoffset, + struct cifs_sid *pownersid, + struct cifs_sid *pgrpsid) { struct cifs_sid *owner_sid_ptr, *group_sid_ptr; struct cifs_sid *nowner_sid_ptr, *ngroup_sid_ptr; @@ -536,19 +543,25 @@ static void copy_sec_desc(const struct cifs_ntsd *pntsd, pnntsd->gsidoffset = cpu_to_le32(sidsoffset + sizeof(struct cifs_sid)); /* copy owner sid */ - owner_sid_ptr = (struct cifs_sid *)((char *)pntsd + + if (pownersid) + owner_sid_ptr = pownersid; + else + owner_sid_ptr = (struct cifs_sid *)((char *)pntsd + le32_to_cpu(pntsd->osidoffset)); nowner_sid_ptr = (struct cifs_sid *)((char *)pnntsd + sidsoffset); cifs_copy_sid(nowner_sid_ptr, owner_sid_ptr); /* copy group sid */ - group_sid_ptr = (struct cifs_sid *)((char *)pntsd + + if (pgrpsid) + group_sid_ptr = pgrpsid; + else + group_sid_ptr = (struct cifs_sid *)((char *)pntsd + le32_to_cpu(pntsd->gsidoffset)); ngroup_sid_ptr = (struct cifs_sid *)((char *)pnntsd + sidsoffset + sizeof(struct cifs_sid)); cifs_copy_sid(ngroup_sid_ptr, group_sid_ptr); - return; + return sidsoffset + (2 * sizeof(struct cifs_sid)); } @@ -663,6 +676,25 @@ static void mode_to_access_flags(umode_t mode, umode_t bits_to_use, return; } +static __u16 cifs_copy_ace(struct cifs_ace *dst, struct cifs_ace *src, struct cifs_sid *psid) +{ + __u16 size = 1 + 1 + 2 + 4; + + dst->type = src->type; + dst->flags = src->flags; + dst->access_req = src->access_req; + + /* Check if there's a replacement sid specified */ + if (psid) + size += cifs_copy_sid(&dst->sid, psid); + else + size += cifs_copy_sid(&dst->sid, &src->sid); + + dst->size = cpu_to_le16(size); + + return size; +} + static __u16 fill_ace_for_sid(struct cifs_ace *pntace, const struct cifs_sid *psid, __u64 nmode, umode_t bits, __u8 access_type, @@ -907,29 +939,30 @@ unsigned int setup_special_user_owner_ACE(struct cifs_ace *pntace) return ace_size; } -static int set_chmod_dacl(struct cifs_acl *pndacl, struct cifs_sid *pownersid, - struct cifs_sid *pgrpsid, __u64 *pnmode, bool modefromsid) +static void populate_new_aces(char *nacl_base, + struct cifs_sid *pownersid, + struct cifs_sid *pgrpsid, + __u64 *pnmode, u32 *pnum_aces, u16 *pnsize, + bool modefromsid) { - u16 size = 0; - u32 num_aces = 0; - struct cifs_acl *pnndacl; __u64 nmode; + u32 num_aces = 0; + u16 nsize = 0; __u64 user_mode; __u64 group_mode; __u64 other_mode; __u64 deny_user_mode = 0; __u64 deny_group_mode = 0; bool sticky_set = false; - - pnndacl = (struct cifs_acl *)((char *)pndacl + sizeof(struct cifs_acl)); + struct cifs_ace *pnntace = NULL; nmode = *pnmode; + num_aces = *pnum_aces; + nsize = *pnsize; if (modefromsid) { - struct cifs_ace *pntace = - (struct cifs_ace *)((char *)pnndacl + size); - - size += setup_special_mode_ACE(pntace, nmode); + pnntace = (struct cifs_ace *) (nacl_base + nsize); + nsize += setup_special_mode_ACE(pnntace, nmode); num_aces++; goto set_size; } @@ -966,40 +999,173 @@ static int set_chmod_dacl(struct cifs_acl *pndacl, struct cifs_sid *pownersid, sticky_set = true; if (deny_user_mode) { - size += fill_ace_for_sid((struct cifs_ace *)((char *)pnndacl + size), - pownersid, deny_user_mode, 0700, ACCESS_DENIED, false); + pnntace = (struct cifs_ace *) (nacl_base + nsize); + nsize += fill_ace_for_sid(pnntace, pownersid, deny_user_mode, + 0700, ACCESS_DENIED, false); num_aces++; } + /* Group DENY ACE does not conflict with owner ALLOW ACE. Keep in preferred order*/ if (deny_group_mode && !(deny_group_mode & (user_mode >> 3))) { - size += fill_ace_for_sid((struct cifs_ace *)((char *)pnndacl + size), - pgrpsid, deny_group_mode, 0070, ACCESS_DENIED, false); + pnntace = (struct cifs_ace *) (nacl_base + nsize); + nsize += fill_ace_for_sid(pnntace, pgrpsid, deny_group_mode, + 0070, ACCESS_DENIED, false); num_aces++; } - size += fill_ace_for_sid((struct cifs_ace *) ((char *)pnndacl + size), - pownersid, user_mode, 0700, ACCESS_ALLOWED, true); + + pnntace = (struct cifs_ace *) (nacl_base + nsize); + nsize += fill_ace_for_sid(pnntace, pownersid, user_mode, + 0700, ACCESS_ALLOWED, true); num_aces++; + /* Group DENY ACE conflicts with owner ALLOW ACE. So keep it after. */ if (deny_group_mode && (deny_group_mode & (user_mode >> 3))) { - size += fill_ace_for_sid((struct cifs_ace *)((char *)pnndacl + size), - pgrpsid, deny_group_mode, 0070, ACCESS_DENIED, false); + pnntace = (struct cifs_ace *) (nacl_base + nsize); + nsize += fill_ace_for_sid(pnntace, pgrpsid, deny_group_mode, + 0070, ACCESS_DENIED, false); num_aces++; } - size += fill_ace_for_sid((struct cifs_ace *)((char *)pnndacl + size), - pgrpsid, group_mode, 0070, ACCESS_ALLOWED, !sticky_set); + + pnntace = (struct cifs_ace *) (nacl_base + nsize); + nsize += fill_ace_for_sid(pnntace, pgrpsid, group_mode, + 0070, ACCESS_ALLOWED, !sticky_set); num_aces++; - size += fill_ace_for_sid((struct cifs_ace *)((char *)pnndacl + size), - &sid_everyone, other_mode, 0007, ACCESS_ALLOWED, !sticky_set); + + pnntace = (struct cifs_ace *) (nacl_base + nsize); + nsize += fill_ace_for_sid(pnntace, &sid_everyone, other_mode, + 0007, ACCESS_ALLOWED, !sticky_set); num_aces++; set_size: + *pnum_aces = num_aces; + *pnsize = nsize; +} + +static __u16 replace_sids_and_copy_aces(struct cifs_acl *pdacl, struct cifs_acl *pndacl, + struct cifs_sid *pownersid, struct cifs_sid *pgrpsid, + struct cifs_sid *pnownersid, struct cifs_sid *pngrpsid) +{ + int i; + u16 size = 0; + struct cifs_ace *pntace = NULL; + char *acl_base = NULL; + u32 src_num_aces = 0; + u16 nsize = 0; + struct cifs_ace *pnntace = NULL; + char *nacl_base = NULL; + u16 ace_size = 0; + + acl_base = (char *)pdacl; + size = sizeof(struct cifs_acl); + src_num_aces = le32_to_cpu(pdacl->num_aces); + + nacl_base = (char *)pndacl; + nsize = sizeof(struct cifs_acl); + + /* Go through all the ACEs */ + for (i = 0; i < src_num_aces; ++i) { + pntace = (struct cifs_ace *) (acl_base + size); + pnntace = (struct cifs_ace *) (nacl_base + nsize); + + if (pnownersid && compare_sids(&pntace->sid, pownersid) == 0) + ace_size = cifs_copy_ace(pnntace, pntace, pnownersid); + else if (pngrpsid && compare_sids(&pntace->sid, pgrpsid) == 0) + ace_size = cifs_copy_ace(pnntace, pntace, pngrpsid); + else + ace_size = cifs_copy_ace(pnntace, pntace, NULL); + + size += le16_to_cpu(pntace->size); + nsize += ace_size; + } + + return nsize; +} + +static int set_chmod_dacl(struct cifs_acl *pdacl, struct cifs_acl *pndacl, + struct cifs_sid *pownersid, struct cifs_sid *pgrpsid, + __u64 *pnmode, bool mode_from_sid) +{ + int i; + u16 size = 0; + struct cifs_ace *pntace = NULL; + char *acl_base = NULL; + u32 src_num_aces = 0; + u16 nsize = 0; + struct cifs_ace *pnntace = NULL; + char *nacl_base = NULL; + u32 num_aces = 0; + __u64 nmode; + bool new_aces_set = false; + + /* Assuming that pndacl and pnmode are never NULL */ + nmode = *pnmode; + nacl_base = (char *)pndacl; + nsize = sizeof(struct cifs_acl); + + /* If pdacl is NULL, we don't have a src. Simply populate new ACL. */ + if (!pdacl) { + populate_new_aces(nacl_base, + pownersid, pgrpsid, + pnmode, &num_aces, &nsize, + mode_from_sid); + goto finalize_dacl; + } + + acl_base = (char *)pdacl; + size = sizeof(struct cifs_acl); + src_num_aces = le32_to_cpu(pdacl->num_aces); + + /* Retain old ACEs which we can retain */ + for (i = 0; i < src_num_aces; ++i) { + pntace = (struct cifs_ace *) (acl_base + size); + + if (!new_aces_set && (pntace->flags & INHERITED_ACE)) { + /* Place the new ACEs in between existing explicit and inherited */ + populate_new_aces(nacl_base, + pownersid, pgrpsid, + pnmode, &num_aces, &nsize, + mode_from_sid); + + new_aces_set = true; + } + + /* If it's any one of the ACE we're replacing, skip! */ + if (!mode_from_sid && + ((compare_sids(&pntace->sid, &sid_unix_NFS_mode) == 0) || + (compare_sids(&pntace->sid, pownersid) == 0) || + (compare_sids(&pntace->sid, pgrpsid) == 0) || + (compare_sids(&pntace->sid, &sid_everyone) == 0) || + (compare_sids(&pntace->sid, &sid_authusers) == 0))) { + goto next_ace; + } + + /* update the pointer to the next ACE to populate*/ + pnntace = (struct cifs_ace *) (nacl_base + nsize); + + nsize += cifs_copy_ace(pnntace, pntace, NULL); + num_aces++; + +next_ace: + size += le16_to_cpu(pntace->size); + } + + /* If inherited ACEs are not present, place the new ones at the tail */ + if (!new_aces_set) { + populate_new_aces(nacl_base, + pownersid, pgrpsid, + pnmode, &num_aces, &nsize, + mode_from_sid); + + new_aces_set = true; + } + +finalize_dacl: pndacl->num_aces = cpu_to_le32(num_aces); - pndacl->size = cpu_to_le16(size + sizeof(struct cifs_acl)); + pndacl->size = cpu_to_le16(nsize); return 0; } - static int parse_sid(struct cifs_sid *psid, char *end_of_acl) { /* BB need to add parm so we can store the SID BB */ @@ -1094,7 +1260,7 @@ static int parse_sec_desc(struct cifs_sb_info *cifs_sb, /* Convert permission bits from mode to equivalent CIFS ACL */ static int build_sec_desc(struct cifs_ntsd *pntsd, struct cifs_ntsd *pnntsd, - __u32 secdesclen, __u64 *pnmode, kuid_t uid, kgid_t gid, + __u32 secdesclen, __u32 *pnsecdesclen, __u64 *pnmode, kuid_t uid, kgid_t gid, bool mode_from_sid, bool id_from_sid, int *aclflag) { int rc = 0; @@ -1102,39 +1268,59 @@ static int build_sec_desc(struct cifs_ntsd *pntsd, struct cifs_ntsd *pnntsd, __u32 ndacloffset; __u32 sidsoffset; struct cifs_sid *owner_sid_ptr, *group_sid_ptr; - struct cifs_sid *nowner_sid_ptr, *ngroup_sid_ptr; + struct cifs_sid *nowner_sid_ptr = NULL, *ngroup_sid_ptr = NULL; struct cifs_acl *dacl_ptr = NULL; /* no need for SACL ptr */ struct cifs_acl *ndacl_ptr = NULL; /* no need for SACL ptr */ + char *end_of_acl = ((char *)pntsd) + secdesclen; + u16 size = 0; - if (pnmode && *pnmode != NO_CHANGE_64) { /* chmod */ - owner_sid_ptr = (struct cifs_sid *)((char *)pntsd + - le32_to_cpu(pntsd->osidoffset)); - group_sid_ptr = (struct cifs_sid *)((char *)pntsd + - le32_to_cpu(pntsd->gsidoffset)); - dacloffset = le32_to_cpu(pntsd->dacloffset); + dacloffset = le32_to_cpu(pntsd->dacloffset); + if (dacloffset) { dacl_ptr = (struct cifs_acl *)((char *)pntsd + dacloffset); + if (end_of_acl < (char *)dacl_ptr + le16_to_cpu(dacl_ptr->size)) { + cifs_dbg(VFS, "Server returned illegal ACL size\n"); + return -EINVAL; + } + } + + owner_sid_ptr = (struct cifs_sid *)((char *)pntsd + + le32_to_cpu(pntsd->osidoffset)); + group_sid_ptr = (struct cifs_sid *)((char *)pntsd + + le32_to_cpu(pntsd->gsidoffset)); + + if (pnmode && *pnmode != NO_CHANGE_64) { /* chmod */ ndacloffset = sizeof(struct cifs_ntsd); ndacl_ptr = (struct cifs_acl *)((char *)pnntsd + ndacloffset); - ndacl_ptr->revision = dacl_ptr->revision; - ndacl_ptr->size = 0; - ndacl_ptr->num_aces = 0; + ndacl_ptr->revision = + dacloffset ? dacl_ptr->revision : cpu_to_le16(ACL_REVISION); - rc = set_chmod_dacl(ndacl_ptr, owner_sid_ptr, group_sid_ptr, + ndacl_ptr->size = cpu_to_le16(0); + ndacl_ptr->num_aces = cpu_to_le32(0); + + rc = set_chmod_dacl(dacl_ptr, ndacl_ptr, owner_sid_ptr, group_sid_ptr, pnmode, mode_from_sid); + sidsoffset = ndacloffset + le16_to_cpu(ndacl_ptr->size); - /* copy sec desc control portion & owner and group sids */ - copy_sec_desc(pntsd, pnntsd, sidsoffset); - *aclflag = CIFS_ACL_DACL; + /* copy the non-dacl portion of secdesc */ + *pnsecdesclen = copy_sec_desc(pntsd, pnntsd, sidsoffset, + NULL, NULL); + + *aclflag |= CIFS_ACL_DACL; } else { - memcpy(pnntsd, pntsd, secdesclen); + ndacloffset = sizeof(struct cifs_ntsd); + ndacl_ptr = (struct cifs_acl *)((char *)pnntsd + ndacloffset); + ndacl_ptr->revision = + dacloffset ? dacl_ptr->revision : cpu_to_le16(ACL_REVISION); + ndacl_ptr->num_aces = dacl_ptr->num_aces; + if (uid_valid(uid)) { /* chown */ uid_t id; - owner_sid_ptr = (struct cifs_sid *)((char *)pnntsd + - le32_to_cpu(pnntsd->osidoffset)); nowner_sid_ptr = kmalloc(sizeof(struct cifs_sid), GFP_KERNEL); - if (!nowner_sid_ptr) - return -ENOMEM; + if (!nowner_sid_ptr) { + rc = -ENOMEM; + goto chown_chgrp_exit; + } id = from_kuid(&init_user_ns, uid); if (id_from_sid) { struct owner_sid *osid = (struct owner_sid *)nowner_sid_ptr; @@ -1145,27 +1331,25 @@ static int build_sec_desc(struct cifs_ntsd *pntsd, struct cifs_ntsd *pnntsd, osid->SubAuthorities[0] = cpu_to_le32(88); osid->SubAuthorities[1] = cpu_to_le32(1); osid->SubAuthorities[2] = cpu_to_le32(id); + } else { /* lookup sid with upcall */ rc = id_to_sid(id, SIDOWNER, nowner_sid_ptr); if (rc) { cifs_dbg(FYI, "%s: Mapping error %d for owner id %d\n", __func__, rc, id); - kfree(nowner_sid_ptr); - return rc; + goto chown_chgrp_exit; } } - cifs_copy_sid(owner_sid_ptr, nowner_sid_ptr); - kfree(nowner_sid_ptr); - *aclflag = CIFS_ACL_OWNER; + *aclflag |= CIFS_ACL_OWNER; } if (gid_valid(gid)) { /* chgrp */ gid_t id; - group_sid_ptr = (struct cifs_sid *)((char *)pnntsd + - le32_to_cpu(pnntsd->gsidoffset)); ngroup_sid_ptr = kmalloc(sizeof(struct cifs_sid), GFP_KERNEL); - if (!ngroup_sid_ptr) - return -ENOMEM; + if (!ngroup_sid_ptr) { + rc = -ENOMEM; + goto chown_chgrp_exit; + } id = from_kgid(&init_user_ns, gid); if (id_from_sid) { struct owner_sid *gsid = (struct owner_sid *)ngroup_sid_ptr; @@ -1176,19 +1360,35 @@ static int build_sec_desc(struct cifs_ntsd *pntsd, struct cifs_ntsd *pnntsd, gsid->SubAuthorities[0] = cpu_to_le32(88); gsid->SubAuthorities[1] = cpu_to_le32(2); gsid->SubAuthorities[2] = cpu_to_le32(id); + } else { /* lookup sid with upcall */ rc = id_to_sid(id, SIDGROUP, ngroup_sid_ptr); if (rc) { cifs_dbg(FYI, "%s: Mapping error %d for group id %d\n", __func__, rc, id); - kfree(ngroup_sid_ptr); - return rc; + goto chown_chgrp_exit; } } - cifs_copy_sid(group_sid_ptr, ngroup_sid_ptr); - kfree(ngroup_sid_ptr); - *aclflag = CIFS_ACL_GROUP; + *aclflag |= CIFS_ACL_GROUP; + } + + if (dacloffset) { + /* Replace ACEs for old owner with new one */ + size = replace_sids_and_copy_aces(dacl_ptr, ndacl_ptr, + owner_sid_ptr, group_sid_ptr, + nowner_sid_ptr, ngroup_sid_ptr); + ndacl_ptr->size = cpu_to_le16(size); } + + sidsoffset = ndacloffset + le16_to_cpu(ndacl_ptr->size); + /* copy the non-dacl portion of secdesc */ + *pnsecdesclen = copy_sec_desc(pntsd, pnntsd, sidsoffset, + nowner_sid_ptr, ngroup_sid_ptr); + +chown_chgrp_exit: + /* errors could jump here. So make sure we return soon after this */ + kfree(nowner_sid_ptr); + kfree(ngroup_sid_ptr); } return rc; @@ -1384,6 +1584,9 @@ id_mode_to_cifs_acl(struct inode *inode, const char *path, __u64 *pnmode, int rc = 0; int aclflag = CIFS_ACL_DACL; /* default flag to set */ __u32 secdesclen = 0; + __u32 nsecdesclen = 0; + __u32 dacloffset = 0; + struct cifs_acl *dacl_ptr = NULL; struct cifs_ntsd *pntsd = NULL; /* acl obtained from server */ struct cifs_ntsd *pnntsd = NULL; /* modified acl to be sent to server */ struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); @@ -1414,31 +1617,52 @@ id_mode_to_cifs_acl(struct inode *inode, const char *path, __u64 *pnmode, return rc; } + if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MODE_FROM_SID) + mode_from_sid = true; + else + mode_from_sid = false; + + if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UID_FROM_ACL) + id_from_sid = true; + else + id_from_sid = false; + + /* Potentially, five new ACEs can be added to the ACL for U,G,O mapping */ + nsecdesclen = secdesclen; + if (pnmode && *pnmode != NO_CHANGE_64) { /* chmod */ + if (mode_from_sid) + nsecdesclen += sizeof(struct cifs_ace); + else /* cifsacl */ + nsecdesclen += 5 * sizeof(struct cifs_ace); + } else { /* chown */ + /* When ownership changes, changes new owner sid length could be different */ + nsecdesclen = sizeof(struct cifs_ntsd) + (sizeof(struct cifs_sid) * 2); + dacloffset = le32_to_cpu(pntsd->dacloffset); + if (dacloffset) { + dacl_ptr = (struct cifs_acl *)((char *)pntsd + dacloffset); + if (mode_from_sid) + nsecdesclen += + le32_to_cpu(dacl_ptr->num_aces) * sizeof(struct cifs_ace); + else /* cifsacl */ + nsecdesclen += le16_to_cpu(dacl_ptr->size); + } + } + /* * Add three ACEs for owner, group, everyone getting rid of other ACEs * as chmod disables ACEs and set the security descriptor. Allocate * memory for the smb header, set security descriptor request security * descriptor parameters, and secuirty descriptor itself */ - secdesclen = max_t(u32, secdesclen, DEFAULT_SEC_DESC_LEN); - pnntsd = kmalloc(secdesclen, GFP_KERNEL); + nsecdesclen = max_t(u32, nsecdesclen, DEFAULT_SEC_DESC_LEN); + pnntsd = kmalloc(nsecdesclen, GFP_KERNEL); if (!pnntsd) { kfree(pntsd); cifs_put_tlink(tlink); return -ENOMEM; } - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_MODE_FROM_SID) - mode_from_sid = true; - else - mode_from_sid = false; - - if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_UID_FROM_ACL) - id_from_sid = true; - else - id_from_sid = false; - - rc = build_sec_desc(pntsd, pnntsd, secdesclen, pnmode, uid, gid, + rc = build_sec_desc(pntsd, pnntsd, secdesclen, &nsecdesclen, pnmode, uid, gid, mode_from_sid, id_from_sid, &aclflag); cifs_dbg(NOISY, "build_sec_desc rc: %d\n", rc); @@ -1448,7 +1672,7 @@ id_mode_to_cifs_acl(struct inode *inode, const char *path, __u64 *pnmode, if (!rc) { /* Set the security descriptor */ - rc = ops->set_acl(pnntsd, secdesclen, inode, path, aclflag); + rc = ops->set_acl(pnntsd, nsecdesclen, inode, path, aclflag); cifs_dbg(NOISY, "set_cifs_acl rc: %d\n", rc); } cifs_put_tlink(tlink); diff --git a/fs/cifs/cifsacl.h b/fs/cifs/cifsacl.h index ff7fd0862e28..d9e704979d99 100644 --- a/fs/cifs/cifsacl.h +++ b/fs/cifs/cifsacl.h @@ -31,8 +31,8 @@ #define EXEC_BIT 0x1 #define ACL_OWNER_MASK 0700 -#define ACL_GROUP_MASK 0770 -#define ACL_EVERYONE_MASK 0777 +#define ACL_GROUP_MASK 0070 +#define ACL_EVERYONE_MASK 0007 #define UBITSHIFT 6 #define GBITSHIFT 3 diff --git a/fs/cifs/cifsencrypt.c b/fs/cifs/cifsencrypt.c index 51d53e4bdf6b..b8f1ff9a83f3 100644 --- a/fs/cifs/cifsencrypt.c +++ b/fs/cifs/cifsencrypt.c @@ -568,15 +568,15 @@ static int calc_ntlmv2_hash(struct cifs_ses *ses, char *ntlmv2_hash, return rc; } } else { - /* We use ses->serverName if no domain name available */ - len = strlen(ses->serverName); + /* We use ses->ip_addr if no domain name available */ + len = strlen(ses->ip_addr); server = kmalloc(2 + (len * 2), GFP_KERNEL); if (server == NULL) { rc = -ENOMEM; return rc; } - len = cifs_strtoUTF16((__le16 *)server, ses->serverName, len, + len = cifs_strtoUTF16((__le16 *)server, ses->ip_addr, len, nls_cp); rc = crypto_shash_update(&ses->server->secmech.sdeschmacmd5->shash, diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index ab883e84e116..099ad9f3660b 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -290,7 +290,7 @@ cifs_statfs(struct dentry *dentry, struct kstatfs *buf) rc = server->ops->queryfs(xid, tcon, cifs_sb, buf); free_xid(xid); - return 0; + return rc; } static long cifs_fallocate(struct file *file, int mode, loff_t off, loff_t len) @@ -305,7 +305,8 @@ static long cifs_fallocate(struct file *file, int mode, loff_t off, loff_t len) return -EOPNOTSUPP; } -static int cifs_permission(struct inode *inode, int mask) +static int cifs_permission(struct user_namespace *mnt_userns, + struct inode *inode, int mask) { struct cifs_sb_info *cifs_sb; @@ -320,7 +321,7 @@ static int cifs_permission(struct inode *inode, int mask) on the client (above and beyond ACL on servers) for servers which do not support setting and viewing mode bits, so allowing client to check permissions is useful */ - return generic_permission(inode, mask); + return generic_permission(&init_user_ns, inode, mask); } static struct kmem_cache *cifs_inode_cachep; @@ -637,8 +638,18 @@ cifs_show_options(struct seq_file *s, struct dentry *root) seq_printf(s, ",snapshot=%llu", tcon->snapshot_time); if (tcon->handle_timeout) seq_printf(s, ",handletimeout=%u", tcon->handle_timeout); - /* convert actimeo and display it in seconds */ - seq_printf(s, ",actimeo=%lu", cifs_sb->ctx->actimeo / HZ); + + /* + * Display file and directory attribute timeout in seconds. + * If file and directory attribute timeout the same then actimeo + * was likely specified on mount + */ + if (cifs_sb->ctx->acdirmax == cifs_sb->ctx->acregmax) + seq_printf(s, ",actimeo=%lu", cifs_sb->ctx->acregmax / HZ); + else { + seq_printf(s, ",acdirmax=%lu", cifs_sb->ctx->acdirmax / HZ); + seq_printf(s, ",acregmax=%lu", cifs_sb->ctx->acregmax / HZ); + } if (tcon->ses->chan_max > 1) seq_printf(s, ",multichannel,max_channels=%zu", @@ -1525,6 +1536,7 @@ init_cifs(void) */ atomic_set(&sesInfoAllocCount, 0); atomic_set(&tconInfoAllocCount, 0); + atomic_set(&tcpSesNextId, 0); atomic_set(&tcpSesAllocCount, 0); atomic_set(&tcpSesReconnectCount, 0); atomic_set(&tconInfoReconnectCount, 0); diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h index 2307bb0f6147..0d7ef150dbb2 100644 --- a/fs/cifs/cifsfs.h +++ b/fs/cifs/cifsfs.h @@ -62,19 +62,22 @@ extern void cifs_sb_deactive(struct super_block *sb); /* Functions related to inodes */ extern const struct inode_operations cifs_dir_inode_ops; extern struct inode *cifs_root_iget(struct super_block *); -extern int cifs_create(struct inode *, struct dentry *, umode_t, - bool excl); +extern int cifs_create(struct user_namespace *, struct inode *, + struct dentry *, umode_t, bool excl); extern int cifs_atomic_open(struct inode *, struct dentry *, struct file *, unsigned, umode_t); extern struct dentry *cifs_lookup(struct inode *, struct dentry *, unsigned int); extern int cifs_unlink(struct inode *dir, struct dentry *dentry); extern int cifs_hardlink(struct dentry *, struct inode *, struct dentry *); -extern int cifs_mknod(struct inode *, struct dentry *, umode_t, dev_t); -extern int cifs_mkdir(struct inode *, struct dentry *, umode_t); +extern int cifs_mknod(struct user_namespace *, struct inode *, struct dentry *, + umode_t, dev_t); +extern int cifs_mkdir(struct user_namespace *, struct inode *, struct dentry *, + umode_t); extern int cifs_rmdir(struct inode *, struct dentry *); -extern int cifs_rename2(struct inode *, struct dentry *, struct inode *, - struct dentry *, unsigned int); +extern int cifs_rename2(struct user_namespace *, struct inode *, + struct dentry *, struct inode *, struct dentry *, + unsigned int); extern int cifs_revalidate_file_attr(struct file *filp); extern int cifs_revalidate_dentry_attr(struct dentry *); extern int cifs_revalidate_file(struct file *filp); @@ -82,8 +85,10 @@ extern int cifs_revalidate_dentry(struct dentry *); extern int cifs_invalidate_mapping(struct inode *inode); extern int cifs_revalidate_mapping(struct inode *inode); extern int cifs_zap_mapping(struct inode *inode); -extern int cifs_getattr(const struct path *, struct kstat *, u32, unsigned int); -extern int cifs_setattr(struct dentry *, struct iattr *); +extern int cifs_getattr(struct user_namespace *, const struct path *, + struct kstat *, u32, unsigned int); +extern int cifs_setattr(struct user_namespace *, struct dentry *, + struct iattr *); extern int cifs_fiemap(struct inode *, struct fiemap_extent_info *, u64 start, u64 len); @@ -132,8 +137,8 @@ extern struct vfsmount *cifs_dfs_d_automount(struct path *path); /* Functions related to symlinks */ extern const char *cifs_get_link(struct dentry *, struct inode *, struct delayed_call *); -extern int cifs_symlink(struct inode *inode, struct dentry *direntry, - const char *symname); +extern int cifs_symlink(struct user_namespace *mnt_userns, struct inode *inode, + struct dentry *direntry, const char *symname); #ifdef CONFIG_CIFS_XATTR extern const struct xattr_handler *cifs_xattr_handlers[]; @@ -160,5 +165,5 @@ extern struct dentry *cifs_smb3_do_mount(struct file_system_type *fs_type, extern const struct export_operations cifs_export_ops; #endif /* CONFIG_CIFS_NFSD_EXPORT */ -#define CIFS_VERSION "2.30" +#define CIFS_VERSION "2.31" #endif /* _CIFSFS_H */ diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index 50fcb65920e8..31fc8695abd6 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -21,6 +21,7 @@ #include <linux/in.h> #include <linux/in6.h> +#include <linux/inet.h> #include <linux/slab.h> #include <linux/mempool.h> #include <linux/workqueue.h> @@ -256,7 +257,7 @@ struct smb_version_operations { /* verify the message */ int (*check_message)(char *, unsigned int, struct TCP_Server_Info *); bool (*is_oplock_break)(char *, struct TCP_Server_Info *); - int (*handle_cancelled_mid)(char *, struct TCP_Server_Info *); + int (*handle_cancelled_mid)(struct mid_q_entry *, struct TCP_Server_Info *); void (*downgrade_oplock)(struct TCP_Server_Info *server, struct cifsInodeInfo *cinode, __u32 oplock, unsigned int epoch, bool *purge_cache); @@ -504,6 +505,8 @@ struct smb_version_operations { loff_t (*llseek)(struct file *, struct cifs_tcon *, loff_t, int); /* Check for STATUS_IO_TIMEOUT */ bool (*is_status_io_timeout)(char *buf); + /* Check for STATUS_NETWORK_NAME_DELETED */ + void (*is_network_name_deleted)(char *buf, struct TCP_Server_Info *srv); }; struct smb_version_values { @@ -577,6 +580,7 @@ inc_rfc1001_len(void *buf, int count) struct TCP_Server_Info { struct list_head tcp_ses_list; struct list_head smb_ses_list; + __u64 conn_id; /* connection identifier (useful for debugging) */ int srv_count; /* reference counter */ /* 15 character server name + 0x20 16th byte indicating type = srv */ char server_RFC1001_name[RFC1001_NAME_LEN_WITH_NULL]; @@ -901,7 +905,7 @@ struct cifs_ses { kuid_t linux_uid; /* overriding owner of files on the mount */ kuid_t cred_uid; /* owner of credentials */ unsigned int capabilities; - char serverName[SERVER_NAME_LEN_WITH_NULL]; + char ip_addr[INET6_ADDRSTRLEN + 1]; /* Max ipv6 (or v4) addr string len */ char *user_name; /* must not be null except during init of sess and after mount option parsing we fill it */ char *domainName; @@ -1701,14 +1705,17 @@ static inline bool is_retryable_error(int error) #define CIFS_NO_RSP_BUF 0x040 /* no response buffer required */ /* Type of request operation */ -#define CIFS_ECHO_OP 0x080 /* echo request */ -#define CIFS_OBREAK_OP 0x0100 /* oplock break request */ -#define CIFS_NEG_OP 0x0200 /* negotiate request */ -#define CIFS_OP_MASK 0x0380 /* mask request type */ - -#define CIFS_HAS_CREDITS 0x0400 /* already has credits */ -#define CIFS_TRANSFORM_REQ 0x0800 /* transform request before sending */ -#define CIFS_NO_SRV_RSP 0x1000 /* there is no server response */ +#define CIFS_ECHO_OP 0x080 /* echo request */ +#define CIFS_OBREAK_OP 0x0100 /* oplock break request */ +#define CIFS_NEG_OP 0x0200 /* negotiate request */ +#define CIFS_CP_CREATE_CLOSE_OP 0x0400 /* compound create+close request */ +/* Lower bitmask values are reserved by others below. */ +#define CIFS_SESS_OP 0x2000 /* session setup request */ +#define CIFS_OP_MASK 0x2780 /* mask request type */ + +#define CIFS_HAS_CREDITS 0x0400 /* already has credits */ +#define CIFS_TRANSFORM_REQ 0x0800 /* transform request before sending */ +#define CIFS_NO_SRV_RSP 0x1000 /* there is no server response */ /* Security Flags: indicate type of session setup needed */ #define CIFSSEC_MAY_SIGN 0x00001 @@ -1844,6 +1851,7 @@ GLOBAL_EXTERN spinlock_t GlobalMid_Lock; /* protects above & list operations */ */ GLOBAL_EXTERN atomic_t sesInfoAllocCount; GLOBAL_EXTERN atomic_t tconInfoAllocCount; +GLOBAL_EXTERN atomic_t tcpSesNextId; GLOBAL_EXTERN atomic_t tcpSesAllocCount; GLOBAL_EXTERN atomic_t tcpSesReconnectCount; GLOBAL_EXTERN atomic_t tconInfoReconnectCount; diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h index 32f7a013402e..75ce6f742b8d 100644 --- a/fs/cifs/cifsproto.h +++ b/fs/cifs/cifsproto.h @@ -232,6 +232,8 @@ extern unsigned int setup_special_user_owner_ACE(struct cifs_ace *pace); extern void dequeue_mid(struct mid_q_entry *mid, bool malformed); extern int cifs_read_from_socket(struct TCP_Server_Info *server, char *buf, unsigned int to_read); +extern ssize_t cifs_discard_from_socket(struct TCP_Server_Info *server, + size_t to_read); extern int cifs_read_page_from_socket(struct TCP_Server_Info *server, struct page *page, unsigned int page_offset, diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index 0496934feecb..c279527aae92 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -1451,9 +1451,9 @@ cifs_discard_remaining_data(struct TCP_Server_Info *server) while (remaining > 0) { int length; - length = cifs_read_from_socket(server, server->bigbuf, - min_t(unsigned int, remaining, - CIFSMaxBufSize + MAX_HEADER_SIZE(server))); + length = cifs_discard_from_socket(server, + min_t(size_t, remaining, + CIFSMaxBufSize + MAX_HEADER_SIZE(server))); if (length < 0) return length; server->total_read += length; diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 4bb9decbbf27..eec8a2052da2 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -242,7 +242,7 @@ cifs_reconnect(struct TCP_Server_Info *server) server->max_read = 0; cifs_dbg(FYI, "Mark tcp session as need reconnect\n"); - trace_smb3_reconnect(server->CurrentMid, server->hostname); + trace_smb3_reconnect(server->CurrentMid, server->conn_id, server->hostname); /* before reconnecting the tcp session, mark the smb session (uid) and the tid bad so they are not used until reconnected */ @@ -564,6 +564,23 @@ cifs_read_from_socket(struct TCP_Server_Info *server, char *buf, return cifs_readv_from_socket(server, &smb_msg); } +ssize_t +cifs_discard_from_socket(struct TCP_Server_Info *server, size_t to_read) +{ + struct msghdr smb_msg; + + /* + * iov_iter_discard already sets smb_msg.type and count and iov_offset + * and cifs_readv_from_socket sets msg_control and msg_controllen + * so little to initialize in struct msghdr + */ + smb_msg.msg_name = NULL; + smb_msg.msg_namelen = 0; + iov_iter_discard(&smb_msg.msg_iter, READ, to_read); + + return cifs_readv_from_socket(server, &smb_msg); +} + int cifs_read_page_from_socket(struct TCP_Server_Info *server, struct page *page, unsigned int page_offset, unsigned int to_read) @@ -724,7 +741,7 @@ static void clean_demultiplex_info(struct TCP_Server_Info *server) spin_lock(&GlobalMid_Lock); list_for_each_safe(tmp, tmp2, &server->pending_mid_q) { mid_entry = list_entry(tmp, struct mid_q_entry, qhead); - cifs_dbg(FYI, "Clearing mid 0x%llx\n", mid_entry->mid); + cifs_dbg(FYI, "Clearing mid %llu\n", mid_entry->mid); kref_get(&mid_entry->refcount); mid_entry->mid_state = MID_SHUTDOWN; list_move(&mid_entry->qhead, &dispose_list); @@ -735,7 +752,7 @@ static void clean_demultiplex_info(struct TCP_Server_Info *server) /* now walk dispose list and issue callbacks */ list_for_each_safe(tmp, tmp2, &dispose_list) { mid_entry = list_entry(tmp, struct mid_q_entry, qhead); - cifs_dbg(FYI, "Callback mid 0x%llx\n", mid_entry->mid); + cifs_dbg(FYI, "Callback mid %llu\n", mid_entry->mid); list_del_init(&mid_entry->qhead); mid_entry->callback(mid_entry); cifs_mid_q_entry_release(mid_entry); @@ -846,7 +863,7 @@ static void smb2_add_credits_from_hdr(char *buffer, struct TCP_Server_Info *server) { struct smb2_sync_hdr *shdr = (struct smb2_sync_hdr *)buffer; - int scredits = server->credits; + int scredits, in_flight; /* * SMB1 does not use credits. @@ -857,12 +874,14 @@ smb2_add_credits_from_hdr(char *buffer, struct TCP_Server_Info *server) if (shdr->CreditRequest) { spin_lock(&server->req_lock); server->credits += le16_to_cpu(shdr->CreditRequest); + scredits = server->credits; + in_flight = server->in_flight; spin_unlock(&server->req_lock); wake_up(&server->request_q); trace_smb3_add_credits(server->CurrentMid, - server->hostname, scredits, - le16_to_cpu(shdr->CreditRequest)); + server->conn_id, server->hostname, scredits, + le16_to_cpu(shdr->CreditRequest), in_flight); cifs_server_dbg(FYI, "%s: added %u credits total=%d\n", __func__, le16_to_cpu(shdr->CreditRequest), scredits); @@ -993,6 +1012,10 @@ next_pdu: if (mids[i] != NULL) { mids[i]->resp_buf_size = server->pdu_size; + if (bufs[i] && server->ops->is_network_name_deleted) + server->ops->is_network_name_deleted(bufs[i], + server); + if (!mids[i]->multiRsp || mids[i]->multiEnd) mids[i]->callback(mids[i]); @@ -1317,6 +1340,7 @@ cifs_get_tcp_session(struct smb3_fs_context *ctx) goto out_err_crypto_release; } + tcp_ses->conn_id = atomic_inc_return(&tcpSesNextId); tcp_ses->noblockcnt = ctx->rootfs; tcp_ses->noblocksnd = ctx->noblocksnd || ctx->rootfs; tcp_ses->noautotune = ctx->noautotune; @@ -1405,6 +1429,11 @@ smbd_connected: tcp_ses->min_offload = ctx->min_offload; tcp_ses->tcpStatus = CifsNeedNegotiate; + if ((ctx->max_credits < 20) || (ctx->max_credits > 60000)) + tcp_ses->max_credits = SMB2_MAX_CREDITS_AVAILABLE; + else + tcp_ses->max_credits = ctx->max_credits; + tcp_ses->nr_targets = 1; tcp_ses->ignore_signature = ctx->ignore_signature; /* thread spawned, put it on the list */ @@ -1838,9 +1867,9 @@ cifs_get_smb_ses(struct TCP_Server_Info *server, struct smb3_fs_context *ctx) /* new SMB session uses our server ref */ ses->server = server; if (server->dstaddr.ss_family == AF_INET6) - sprintf(ses->serverName, "%pI6", &addr6->sin6_addr); + sprintf(ses->ip_addr, "%pI6", &addr6->sin6_addr); else - sprintf(ses->serverName, "%pI4", &addr->sin_addr); + sprintf(ses->ip_addr, "%pI4", &addr->sin_addr); if (ctx->username) { ses->user_name = kstrdup(ctx->username, GFP_KERNEL); @@ -2269,7 +2298,9 @@ compare_mount_options(struct super_block *sb, struct cifs_mnt_data *mnt_data) if (strcmp(old->local_nls->charset, new->local_nls->charset)) return 0; - if (old->ctx->actimeo != new->ctx->actimeo) + if (old->ctx->acregmax != new->ctx->acregmax) + return 0; + if (old->ctx->acdirmax != new->ctx->acdirmax) return 0; return 1; @@ -2806,11 +2837,6 @@ static int mount_get_conns(struct smb3_fs_context *ctx, struct cifs_sb_info *cif *nserver = server; - if ((ctx->max_credits < 20) || (ctx->max_credits > 60000)) - server->max_credits = SMB2_MAX_CREDITS_AVAILABLE; - else - server->max_credits = ctx->max_credits; - /* get a reference to a SMB session */ ses = cifs_get_smb_ses(server, ctx); if (IS_ERR(ses)) { @@ -2911,7 +2937,7 @@ static int mount_setup_tlink(struct cifs_sb_info *cifs_sb, struct cifs_ses *ses, #ifdef CONFIG_CIFS_DFS_UPCALL /* * cifs_build_path_to_root returns full path to root when we do not have an - * exiting connection (tcon) + * existing connection (tcon) */ static char * build_unc_path_to_root(const struct smb3_fs_context *ctx, @@ -3038,96 +3064,91 @@ static int update_vol_info(const struct dfs_cache_tgt_iterator *tgt_it, return 0; } -static int setup_dfs_tgt_conn(const char *path, const char *full_path, - const struct dfs_cache_tgt_iterator *tgt_it, - struct cifs_sb_info *cifs_sb, struct smb3_fs_context *ctx, - unsigned int *xid, struct TCP_Server_Info **server, - struct cifs_ses **ses, struct cifs_tcon **tcon) -{ - int rc; - struct dfs_info3_param ref = {0}; - char *mdata = NULL; - struct smb3_fs_context fake_ctx = {NULL}; - char *fake_devname = NULL; - - cifs_dbg(FYI, "%s: dfs path: %s\n", __func__, path); - - rc = dfs_cache_get_tgt_referral(path, tgt_it, &ref); - if (rc) - return rc; - - mdata = cifs_compose_mount_options(cifs_sb->ctx->mount_options, - full_path + 1, &ref, - &fake_devname); - free_dfs_info_param(&ref); - - if (IS_ERR(mdata)) { - rc = PTR_ERR(mdata); - mdata = NULL; - } else - rc = cifs_setup_volume_info(&fake_ctx, mdata, fake_devname); - - kfree(mdata); - kfree(fake_devname); - - if (!rc) { - /* - * We use a 'fake_ctx' here because we need pass it down to the - * mount_{get,put} functions to test connection against new DFS - * targets. - */ - mount_put_conns(cifs_sb, *xid, *server, *ses, *tcon); - rc = mount_get_conns(&fake_ctx, cifs_sb, xid, server, ses, - tcon); - if (!rc || (*server && *ses)) { - /* - * We were able to connect to new target server. - * Update current context with new target server. - */ - rc = update_vol_info(tgt_it, &fake_ctx, ctx); - } - } - smb3_cleanup_fs_context_contents(&fake_ctx); - return rc; -} - static int do_dfs_failover(const char *path, const char *full_path, struct cifs_sb_info *cifs_sb, struct smb3_fs_context *ctx, struct cifs_ses *root_ses, unsigned int *xid, struct TCP_Server_Info **server, struct cifs_ses **ses, struct cifs_tcon **tcon) { int rc; - struct dfs_cache_tgt_list tgt_list; + struct dfs_cache_tgt_list tgt_list = {0}; struct dfs_cache_tgt_iterator *tgt_it = NULL; + struct smb3_fs_context tmp_ctx = {NULL}; if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_DFS) return -EOPNOTSUPP; + cifs_dbg(FYI, "%s: path=%s full_path=%s\n", __func__, path, full_path); + rc = dfs_cache_noreq_find(path, NULL, &tgt_list); if (rc) return rc; + /* + * We use a 'tmp_ctx' here because we need pass it down to the mount_{get,put} functions to + * test connection against new DFS targets. + */ + rc = smb3_fs_context_dup(&tmp_ctx, ctx); + if (rc) + goto out; for (;;) { + struct dfs_info3_param ref = {0}; + char *fake_devname = NULL, *mdata = NULL; + /* Get next DFS target server - if any */ rc = get_next_dfs_tgt(path, &tgt_list, &tgt_it); if (rc) break; - /* Connect to next DFS target */ - rc = setup_dfs_tgt_conn(path, full_path, tgt_it, cifs_sb, ctx, xid, server, ses, - tcon); - if (!rc || (*server && *ses)) + + rc = dfs_cache_get_tgt_referral(path, tgt_it, &ref); + if (rc) break; + + cifs_dbg(FYI, "%s: old ctx: UNC=%s prepath=%s\n", __func__, tmp_ctx.UNC, + tmp_ctx.prepath); + + mdata = cifs_compose_mount_options(cifs_sb->ctx->mount_options, full_path + 1, &ref, + &fake_devname); + free_dfs_info_param(&ref); + + if (IS_ERR(mdata)) { + rc = PTR_ERR(mdata); + mdata = NULL; + } else + rc = cifs_setup_volume_info(&tmp_ctx, mdata, fake_devname); + + kfree(mdata); + kfree(fake_devname); + + if (rc) + break; + + cifs_dbg(FYI, "%s: new ctx: UNC=%s prepath=%s\n", __func__, tmp_ctx.UNC, + tmp_ctx.prepath); + + mount_put_conns(cifs_sb, *xid, *server, *ses, *tcon); + rc = mount_get_conns(&tmp_ctx, cifs_sb, xid, server, ses, tcon); + if (!rc || (*server && *ses)) { + /* + * We were able to connect to new target server. Update current context with + * new target server. + */ + rc = update_vol_info(tgt_it, &tmp_ctx, ctx); + break; + } } if (!rc) { + cifs_dbg(FYI, "%s: final ctx: UNC=%s prepath=%s\n", __func__, tmp_ctx.UNC, + tmp_ctx.prepath); /* - * Update DFS target hint in DFS referral cache with the target - * server we successfully reconnected to. + * Update DFS target hint in DFS referral cache with the target server we + * successfully reconnected to. */ - rc = dfs_cache_update_tgthint(*xid, root_ses ? root_ses : *ses, - cifs_sb->local_nls, - cifs_remap(cifs_sb), path, - tgt_it); + rc = dfs_cache_update_tgthint(*xid, root_ses ? root_ses : *ses, cifs_sb->local_nls, + cifs_remap(cifs_sb), path, tgt_it); } + +out: + smb3_cleanup_fs_context_contents(&tmp_ctx); dfs_cache_free_tgts(&tgt_list); return rc; } @@ -3285,77 +3306,77 @@ static void put_root_ses(struct cifs_ses *ses) cifs_put_smb_ses(ses); } -/* Check if a path component is remote and then update @dfs_path accordingly */ -static int check_dfs_prepath(struct cifs_sb_info *cifs_sb, struct smb3_fs_context *ctx, - const unsigned int xid, struct TCP_Server_Info *server, - struct cifs_tcon *tcon, char **dfs_path) +/* Set up next dfs prefix path in @dfs_path */ +static int next_dfs_prepath(struct cifs_sb_info *cifs_sb, struct smb3_fs_context *ctx, + const unsigned int xid, struct TCP_Server_Info *server, + struct cifs_tcon *tcon, char **dfs_path) { - char *path, *s; - char sep = CIFS_DIR_SEP(cifs_sb), tmp; - char *npath; - int rc = 0; - int added_treename = tcon->Flags & SMB_SHARE_IS_IN_DFS; - int skip = added_treename; + char *path, *npath; + int added_treename = is_tcon_dfs(tcon); + int rc; path = cifs_build_path_to_root(ctx, cifs_sb, tcon, added_treename); if (!path) return -ENOMEM; - /* - * Walk through the path components in @path and check if they're accessible. In case any of - * the components is -EREMOTE, then update @dfs_path with the next DFS referral request path - * (NOT including the remaining components). - */ - s = path; - do { - /* skip separators */ - while (*s && *s == sep) - s++; - if (!*s) - break; - /* next separator */ - while (*s && *s != sep) - s++; - /* - * if the treename is added, we then have to skip the first - * part within the separators - */ - if (skip) { - skip = 0; - continue; + rc = is_path_remote(cifs_sb, ctx, xid, server, tcon); + if (rc == -EREMOTE) { + struct smb3_fs_context v = {NULL}; + /* if @path contains a tree name, skip it in the prefix path */ + if (added_treename) { + rc = smb3_parse_devname(path, &v); + if (rc) + goto out; + npath = build_unc_path_to_root(&v, cifs_sb, true); + smb3_cleanup_fs_context_contents(&v); + } else { + v.UNC = ctx->UNC; + v.prepath = path + 1; + npath = build_unc_path_to_root(&v, cifs_sb, true); } - tmp = *s; - *s = 0; - rc = server->ops->is_path_accessible(xid, tcon, cifs_sb, path); - if (rc && rc == -EREMOTE) { - struct smb3_fs_context v = {NULL}; - /* if @path contains a tree name, skip it in the prefix path */ - if (added_treename) { - rc = smb3_parse_devname(path, &v); - if (rc) - break; - rc = -EREMOTE; - npath = build_unc_path_to_root(&v, cifs_sb, true); - smb3_cleanup_fs_context_contents(&v); - } else { - v.UNC = ctx->UNC; - v.prepath = path + 1; - npath = build_unc_path_to_root(&v, cifs_sb, true); - } - if (IS_ERR(npath)) { - rc = PTR_ERR(npath); - break; - } - kfree(*dfs_path); - *dfs_path = npath; + + if (IS_ERR(npath)) { + rc = PTR_ERR(npath); + goto out; } - *s = tmp; - } while (rc == 0); + kfree(*dfs_path); + *dfs_path = npath; + rc = -EREMOTE; + } + +out: kfree(path); return rc; } +/* Check if resolved targets can handle any DFS referrals */ +static int is_referral_server(const char *ref_path, struct cifs_tcon *tcon, bool *ref_server) +{ + int rc; + struct dfs_info3_param ref = {0}; + + if (is_tcon_dfs(tcon)) { + *ref_server = true; + } else { + cifs_dbg(FYI, "%s: ref_path=%s\n", __func__, ref_path); + + rc = dfs_cache_noreq_find(ref_path, &ref, NULL); + if (rc) { + cifs_dbg(VFS, "%s: dfs_cache_noreq_find: failed (rc=%d)\n", __func__, rc); + return rc; + } + cifs_dbg(FYI, "%s: ref.flags=0x%x\n", __func__, ref.flags); + /* + * Check if all targets are capable of handling DFS referrals as per + * MS-DFSC 2.2.4 RESP_GET_DFS_REFERRAL. + */ + *ref_server = !!(ref.flags & DFSREF_REFERRAL_SERVER); + free_dfs_info_param(&ref); + } + return 0; +} + int cifs_mount(struct cifs_sb_info *cifs_sb, struct smb3_fs_context *ctx) { int rc = 0; @@ -3367,18 +3388,19 @@ int cifs_mount(struct cifs_sb_info *cifs_sb, struct smb3_fs_context *ctx) char *ref_path = NULL, *full_path = NULL; char *oldmnt = NULL; char *mntdata = NULL; + bool ref_server = false; rc = mount_get_conns(ctx, cifs_sb, &xid, &server, &ses, &tcon); /* - * Unconditionally try to get an DFS referral (even cached) to determine whether it is an - * DFS mount. + * If called with 'nodfs' mount option, then skip DFS resolving. Otherwise unconditionally + * try to get an DFS referral (even cached) to determine whether it is an DFS mount. * * Skip prefix path to provide support for DFS referrals from w2k8 servers which don't seem * to respond with PATH_NOT_COVERED to requests that include the prefix. */ - if (dfs_cache_find(xid, ses, cifs_sb->local_nls, cifs_remap(cifs_sb), ctx->UNC + 1, NULL, + if ((cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_DFS) || + dfs_cache_find(xid, ses, cifs_sb->local_nls, cifs_remap(cifs_sb), ctx->UNC + 1, NULL, NULL)) { - /* No DFS referral was returned. Looks like a regular share. */ if (rc) goto error; /* Check if it is fully accessible and then mount it */ @@ -3432,13 +3454,18 @@ int cifs_mount(struct cifs_sb_info *cifs_sb, struct smb3_fs_context *ctx) break; if (!tcon) continue; + /* Make sure that requests go through new root servers */ - if (is_tcon_dfs(tcon)) { + rc = is_referral_server(ref_path + 1, tcon, &ref_server); + if (rc) + break; + if (ref_server) { put_root_ses(root_ses); set_root_ses(cifs_sb, ses, &root_ses); } - /* Check for remaining path components and then continue chasing them (-EREMOTE) */ - rc = check_dfs_prepath(cifs_sb, ctx, xid, server, tcon, &ref_path); + + /* Get next dfs path and then continue chasing them if -EREMOTE */ + rc = next_dfs_prepath(cifs_sb, ctx, xid, server, tcon, &ref_path); /* Prevent recursion on broken link referrals */ if (rc == -EREMOTE && ++count > MAX_NESTED_LINKS) rc = -ELOOP; diff --git a/fs/cifs/dfs_cache.c b/fs/cifs/dfs_cache.c index 4950ab0486ae..098b4bc8da59 100644 --- a/fs/cifs/dfs_cache.c +++ b/fs/cifs/dfs_cache.c @@ -37,11 +37,12 @@ struct cache_dfs_tgt { struct cache_entry { struct hlist_node hlist; const char *path; - int ttl; - int srvtype; - int flags; + int hdr_flags; /* RESP_GET_DFS_REFERRAL.ReferralHeaderFlags */ + int ttl; /* DFS_REREFERRAL_V3.TimeToLive */ + int srvtype; /* DFS_REREFERRAL_V3.ServerType */ + int ref_flags; /* DFS_REREFERRAL_V3.ReferralEntryFlags */ struct timespec64 etime; - int path_consumed; + int path_consumed; /* RESP_GET_DFS_REFERRAL.PathConsumed */ int numtgts; struct list_head tlist; struct cache_dfs_tgt *tgthint; @@ -166,14 +167,11 @@ static int dfscache_proc_show(struct seq_file *m, void *v) continue; seq_printf(m, - "cache entry: path=%s,type=%s,ttl=%d,etime=%ld," - "interlink=%s,path_consumed=%d,expired=%s\n", - ce->path, - ce->srvtype == DFS_TYPE_ROOT ? "root" : "link", - ce->ttl, ce->etime.tv_nsec, - IS_INTERLINK_SET(ce->flags) ? "yes" : "no", - ce->path_consumed, - cache_entry_expired(ce) ? "yes" : "no"); + "cache entry: path=%s,type=%s,ttl=%d,etime=%ld,hdr_flags=0x%x,ref_flags=0x%x,interlink=%s,path_consumed=%d,expired=%s\n", + ce->path, ce->srvtype == DFS_TYPE_ROOT ? "root" : "link", + ce->ttl, ce->etime.tv_nsec, ce->ref_flags, ce->hdr_flags, + IS_INTERLINK_SET(ce->hdr_flags) ? "yes" : "no", + ce->path_consumed, cache_entry_expired(ce) ? "yes" : "no"); list_for_each_entry(t, &ce->tlist, list) { seq_printf(m, " %s%s\n", @@ -236,11 +234,12 @@ static inline void dump_tgts(const struct cache_entry *ce) static inline void dump_ce(const struct cache_entry *ce) { - cifs_dbg(FYI, "cache entry: path=%s,type=%s,ttl=%d,etime=%ld,interlink=%s,path_consumed=%d,expired=%s\n", + cifs_dbg(FYI, "cache entry: path=%s,type=%s,ttl=%d,etime=%ld,hdr_flags=0x%x,ref_flags=0x%x,interlink=%s,path_consumed=%d,expired=%s\n", ce->path, ce->srvtype == DFS_TYPE_ROOT ? "root" : "link", ce->ttl, ce->etime.tv_nsec, - IS_INTERLINK_SET(ce->flags) ? "yes" : "no", + ce->hdr_flags, ce->ref_flags, + IS_INTERLINK_SET(ce->hdr_flags) ? "yes" : "no", ce->path_consumed, cache_entry_expired(ce) ? "yes" : "no"); dump_tgts(ce); @@ -381,7 +380,8 @@ static int copy_ref_data(const struct dfs_info3_param *refs, int numrefs, ce->ttl = refs[0].ttl; ce->etime = get_expire_time(ce->ttl); ce->srvtype = refs[0].server_type; - ce->flags = refs[0].ref_flag; + ce->hdr_flags = refs[0].flags; + ce->ref_flags = refs[0].ref_flag; ce->path_consumed = refs[0].path_consumed; for (i = 0; i < numrefs; i++) { @@ -799,7 +799,8 @@ static int setup_referral(const char *path, struct cache_entry *ce, ref->path_consumed = ce->path_consumed; ref->ttl = ce->ttl; ref->server_type = ce->srvtype; - ref->ref_flag = ce->flags; + ref->ref_flag = ce->ref_flags; + ref->flags = ce->hdr_flags; return 0; diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c index 97ac363b5df1..a3fb81e0ba17 100644 --- a/fs/cifs/dir.c +++ b/fs/cifs/dir.c @@ -567,8 +567,8 @@ out_free_xid: return rc; } -int cifs_create(struct inode *inode, struct dentry *direntry, umode_t mode, - bool excl) +int cifs_create(struct user_namespace *mnt_userns, struct inode *inode, + struct dentry *direntry, umode_t mode, bool excl) { int rc; unsigned int xid = get_xid(); @@ -611,8 +611,8 @@ out_free_xid: return rc; } -int cifs_mknod(struct inode *inode, struct dentry *direntry, umode_t mode, - dev_t device_number) +int cifs_mknod(struct user_namespace *mnt_userns, struct inode *inode, + struct dentry *direntry, umode_t mode, dev_t device_number) { int rc = -EPERM; unsigned int xid; diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 6d001905c8e5..26de4329d161 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -580,7 +580,7 @@ int cifs_open(struct inode *inode, struct file *file) } else if ((rc == -EINVAL) || (rc == -EOPNOTSUPP)) { if (tcon->ses->serverNOS) cifs_dbg(VFS, "server %s of type %s returned unexpected error on SMB posix open, disabling posix open support. Check if server update available.\n", - tcon->ses->serverName, + tcon->ses->ip_addr, tcon->ses->serverNOS); tcon->broken_posix_open = true; } else if ((rc != -EIO) && (rc != -EREMOTE) && diff --git a/fs/cifs/fs_context.c b/fs/cifs/fs_context.c index 12a5da0230b5..78889024a7ed 100644 --- a/fs/cifs/fs_context.c +++ b/fs/cifs/fs_context.c @@ -140,6 +140,8 @@ const struct fs_parameter_spec smb3_fs_parameters[] = { fsparam_u32("rsize", Opt_rsize), fsparam_u32("wsize", Opt_wsize), fsparam_u32("actimeo", Opt_actimeo), + fsparam_u32("acdirmax", Opt_acdirmax), + fsparam_u32("acregmax", Opt_acregmax), fsparam_u32("echo_interval", Opt_echo_interval), fsparam_u32("max_credits", Opt_max_credits), fsparam_u32("handletimeout", Opt_handletimeout), @@ -397,7 +399,7 @@ cifs_parse_smb_version(char *value, struct smb3_fs_context *ctx, bool is_smb3) ctx->vals = &smb3any_values; break; case Smb_default: - ctx->ops = &smb30_operations; /* currently identical with 3.0 */ + ctx->ops = &smb30_operations; ctx->vals = &smbdefault_values; break; default: @@ -542,20 +544,37 @@ static int smb3_fs_context_parse_monolithic(struct fs_context *fc, /* BB Need to add support for sep= here TBD */ while ((key = strsep(&options, ",")) != NULL) { - if (*key) { - size_t v_len = 0; - char *value = strchr(key, '='); - - if (value) { - if (value == key) - continue; - *value++ = 0; - v_len = strlen(value); - } - ret = vfs_parse_fs_string(fc, key, value, v_len); - if (ret < 0) - break; + size_t len; + char *value; + + if (*key == 0) + break; + + /* Check if following character is the deliminator If yes, + * we have encountered a double deliminator reset the NULL + * character to the deliminator + */ + while (options && options[0] == ',') { + len = strlen(key); + strcpy(key + len, options); + options = strchr(options, ','); + if (options) + *options++ = 0; + } + + + len = 0; + value = strchr(key, '='); + if (value) { + if (value == key) + continue; + *value++ = 0; + len = strlen(value); } + + ret = vfs_parse_fs_string(fc, key, value, len); + if (ret < 0) + break; } return ret; @@ -929,12 +948,31 @@ static int smb3_fs_context_parse_param(struct fs_context *fc, ctx->wsize = result.uint_32; ctx->got_wsize = true; break; + case Opt_acregmax: + ctx->acregmax = HZ * result.uint_32; + if (ctx->acregmax > CIFS_MAX_ACTIMEO) { + cifs_dbg(VFS, "acregmax too large\n"); + goto cifs_parse_mount_err; + } + break; + case Opt_acdirmax: + ctx->acdirmax = HZ * result.uint_32; + if (ctx->acdirmax > CIFS_MAX_ACTIMEO) { + cifs_dbg(VFS, "acdirmax too large\n"); + goto cifs_parse_mount_err; + } + break; case Opt_actimeo: - ctx->actimeo = HZ * result.uint_32; - if (ctx->actimeo > CIFS_MAX_ACTIMEO) { - cifs_dbg(VFS, "attribute cache timeout too large\n"); + if (HZ * result.uint_32 > CIFS_MAX_ACTIMEO) { + cifs_dbg(VFS, "timeout too large\n"); goto cifs_parse_mount_err; } + if ((ctx->acdirmax != CIFS_DEF_ACTIMEO) || + (ctx->acregmax != CIFS_DEF_ACTIMEO)) { + cifs_dbg(VFS, "actimeo ignored since acregmax or acdirmax specified\n"); + break; + } + ctx->acdirmax = ctx->acregmax = HZ * result.uint_32; break; case Opt_echo_interval: ctx->echo_interval = result.uint_32; @@ -1158,9 +1196,11 @@ static int smb3_fs_context_parse_param(struct fs_context *fc, pr_warn_once("Witness protocol support is experimental\n"); break; case Opt_rootfs: -#ifdef CONFIG_CIFS_ROOT - ctx->rootfs = true; +#ifndef CONFIG_CIFS_ROOT + cifs_dbg(VFS, "rootfs support requires CONFIG_CIFS_ROOT config option\n"); + goto cifs_parse_mount_err; #endif + ctx->rootfs = true; break; case Opt_posixpaths: if (result.negated) @@ -1361,7 +1401,8 @@ int smb3_init_fs_context(struct fs_context *fc) /* default is to use strict cifs caching semantics */ ctx->strict_io = true; - ctx->actimeo = CIFS_DEF_ACTIMEO; + ctx->acregmax = CIFS_DEF_ACTIMEO; + ctx->acdirmax = CIFS_DEF_ACTIMEO; /* Most clients set timeout to 0, allows server to use its default */ ctx->handle_timeout = 0; /* See MS-SMB2 spec section 2.2.14.2.12 */ diff --git a/fs/cifs/fs_context.h b/fs/cifs/fs_context.h index 1c44a460e2c0..87dd1f7168f2 100644 --- a/fs/cifs/fs_context.h +++ b/fs/cifs/fs_context.h @@ -118,6 +118,8 @@ enum cifs_param { Opt_rsize, Opt_wsize, Opt_actimeo, + Opt_acdirmax, + Opt_acregmax, Opt_echo_interval, Opt_max_credits, Opt_snapshot, @@ -232,7 +234,9 @@ struct smb3_fs_context { unsigned int wsize; unsigned int min_offload; bool sockopt_tcp_nodelay:1; - unsigned long actimeo; /* attribute cache timeout (jiffies) */ + /* attribute cache timemout for files and directories in jiffies */ + unsigned long acregmax; + unsigned long acdirmax; struct smb_version_operations *ops; struct smb_version_values *vals; char *prepath; diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index a83b3a8ffaac..f2df4422e54a 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -1857,7 +1857,8 @@ posix_mkdir_get_info: goto posix_mkdir_out; } -int cifs_mkdir(struct inode *inode, struct dentry *direntry, umode_t mode) +int cifs_mkdir(struct user_namespace *mnt_userns, struct inode *inode, + struct dentry *direntry, umode_t mode) { int rc = 0; unsigned int xid; @@ -2067,9 +2068,9 @@ do_rename_exit: } int -cifs_rename2(struct inode *source_dir, struct dentry *source_dentry, - struct inode *target_dir, struct dentry *target_dentry, - unsigned int flags) +cifs_rename2(struct user_namespace *mnt_userns, struct inode *source_dir, + struct dentry *source_dentry, struct inode *target_dir, + struct dentry *target_dentry, unsigned int flags) { char *from_name = NULL; char *to_name = NULL; @@ -2198,12 +2199,23 @@ cifs_inode_needs_reval(struct inode *inode) if (!lookupCacheEnabled) return true; - if (!cifs_sb->ctx->actimeo) - return true; - - if (!time_in_range(jiffies, cifs_i->time, - cifs_i->time + cifs_sb->ctx->actimeo)) - return true; + /* + * depending on inode type, check if attribute caching disabled for + * files or directories + */ + if (S_ISDIR(inode->i_mode)) { + if (!cifs_sb->ctx->acdirmax) + return true; + if (!time_in_range(jiffies, cifs_i->time, + cifs_i->time + cifs_sb->ctx->acdirmax)) + return true; + } else { /* file */ + if (!cifs_sb->ctx->acregmax) + return true; + if (!time_in_range(jiffies, cifs_i->time, + cifs_i->time + cifs_sb->ctx->acregmax)) + return true; + } /* hardlinked files w/ noserverino get "special" treatment */ if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_SERVER_INUM) && @@ -2370,8 +2382,8 @@ int cifs_revalidate_dentry(struct dentry *dentry) return cifs_revalidate_mapping(inode); } -int cifs_getattr(const struct path *path, struct kstat *stat, - u32 request_mask, unsigned int flags) +int cifs_getattr(struct user_namespace *mnt_userns, const struct path *path, + struct kstat *stat, u32 request_mask, unsigned int flags) { struct dentry *dentry = path->dentry; struct cifs_sb_info *cifs_sb = CIFS_SB(dentry->d_sb); @@ -2383,7 +2395,7 @@ int cifs_getattr(const struct path *path, struct kstat *stat, * We need to be sure that all dirty pages are written and the server * has actual ctime, mtime and file length. */ - if ((request_mask & (STATX_CTIME | STATX_MTIME | STATX_SIZE)) && + if ((request_mask & (STATX_CTIME | STATX_MTIME | STATX_SIZE | STATX_BLOCKS)) && !CIFS_CACHE_READ(CIFS_I(inode)) && inode->i_mapping && inode->i_mapping->nrpages != 0) { rc = filemap_fdatawait(inode->i_mapping); @@ -2408,7 +2420,7 @@ int cifs_getattr(const struct path *path, struct kstat *stat, return rc; } - generic_fillattr(inode, stat); + generic_fillattr(&init_user_ns, inode, stat); stat->blksize = cifs_sb->ctx->bsize; stat->ino = CIFS_I(inode)->uniqueid; @@ -2573,6 +2585,14 @@ set_size_out: if (rc == 0) { cifsInode->server_eof = attrs->ia_size; cifs_setsize(inode, attrs->ia_size); + /* + * i_blocks is not related to (i_size / i_blksize), but instead + * 512 byte (2**9) size is required for calculating num blocks. + * Until we can query the server for actual allocation size, + * this is best estimate we have for blocks allocated for a file + * Number of blocks must be rounded up so size 1 is not 0 blocks + */ + inode->i_blocks = (512 - 1 + attrs->ia_size) >> 9; /* * The man page of truncate says if the size changed, @@ -2610,7 +2630,7 @@ cifs_setattr_unix(struct dentry *direntry, struct iattr *attrs) if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_PERM) attrs->ia_valid |= ATTR_FORCE; - rc = setattr_prepare(direntry, attrs); + rc = setattr_prepare(&init_user_ns, direntry, attrs); if (rc < 0) goto out; @@ -2715,7 +2735,7 @@ cifs_setattr_unix(struct dentry *direntry, struct iattr *attrs) attrs->ia_size != i_size_read(inode)) truncate_setsize(inode, attrs->ia_size); - setattr_copy(inode, attrs); + setattr_copy(&init_user_ns, inode, attrs); mark_inode_dirty(inode); /* force revalidate when any of these times are set since some @@ -2757,7 +2777,7 @@ cifs_setattr_nounix(struct dentry *direntry, struct iattr *attrs) if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_NO_PERM) attrs->ia_valid |= ATTR_FORCE; - rc = setattr_prepare(direntry, attrs); + rc = setattr_prepare(&init_user_ns, direntry, attrs); if (rc < 0) { free_xid(xid); return rc; @@ -2913,7 +2933,7 @@ cifs_setattr_nounix(struct dentry *direntry, struct iattr *attrs) attrs->ia_size != i_size_read(inode)) truncate_setsize(inode, attrs->ia_size); - setattr_copy(inode, attrs); + setattr_copy(&init_user_ns, inode, attrs); mark_inode_dirty(inode); cifs_setattr_exit: @@ -2923,7 +2943,8 @@ cifs_setattr_exit: } int -cifs_setattr(struct dentry *direntry, struct iattr *attrs) +cifs_setattr(struct user_namespace *mnt_userns, struct dentry *direntry, + struct iattr *attrs) { struct cifs_sb_info *cifs_sb = CIFS_SB(direntry->d_sb); struct cifs_tcon *pTcon = cifs_sb_master_tcon(cifs_sb); diff --git a/fs/cifs/link.c b/fs/cifs/link.c index 94dab4309fbb..7c5878a645d9 100644 --- a/fs/cifs/link.c +++ b/fs/cifs/link.c @@ -661,7 +661,8 @@ cifs_get_link(struct dentry *direntry, struct inode *inode, } int -cifs_symlink(struct inode *inode, struct dentry *direntry, const char *symname) +cifs_symlink(struct user_namespace *mnt_userns, struct inode *inode, + struct dentry *direntry, const char *symname) { int rc = -EOPNOTSUPP; unsigned int xid; diff --git a/fs/cifs/sess.c b/fs/cifs/sess.c index 213465718fa8..63d517b9f2ff 100644 --- a/fs/cifs/sess.c +++ b/fs/cifs/sess.c @@ -218,7 +218,7 @@ cifs_ses_add_channel(struct cifs_sb_info *cifs_sb, struct cifs_ses *ses, /* UNC and paths */ /* XXX: Use ses->server->hostname? */ - sprintf(unc, unc_fmt, ses->serverName); + sprintf(unc, unc_fmt, ses->ip_addr); ctx.UNC = unc; ctx.prepath = ""; @@ -230,6 +230,7 @@ cifs_ses_add_channel(struct cifs_sb_info *cifs_sb, struct cifs_ses *ses, ctx.noautotune = ses->server->noautotune; ctx.sockopt_tcp_nodelay = ses->server->tcp_nodelay; ctx.echo_interval = ses->server->echo_interval / HZ; + ctx.max_credits = ses->server->max_credits; /* * This will be used for encoding/decoding user/domain/pw diff --git a/fs/cifs/smb2inode.c b/fs/cifs/smb2inode.c index 1f900b81c34a..a718dc77e604 100644 --- a/fs/cifs/smb2inode.c +++ b/fs/cifs/smb2inode.c @@ -358,6 +358,7 @@ smb2_compound_op(const unsigned int xid, struct cifs_tcon *tcon, if (cfile) goto after_close; /* Close */ + flags |= CIFS_CP_CREATE_CLOSE_OP; rqst[num_rqst].rq_iov = &vars->close_iov[0]; rqst[num_rqst].rq_nvec = 1; rc = SMB2_close_init(tcon, server, diff --git a/fs/cifs/smb2misc.c b/fs/cifs/smb2misc.c index 60d4bd1eae2b..b50164e2c88d 100644 --- a/fs/cifs/smb2misc.c +++ b/fs/cifs/smb2misc.c @@ -767,7 +767,7 @@ smb2_cancelled_close_fid(struct work_struct *work) int rc; if (cancelled->mid) - cifs_tcon_dbg(VFS, "Close unmatched open for MID:%llx\n", + cifs_tcon_dbg(VFS, "Close unmatched open for MID:%llu\n", cancelled->mid); else cifs_tcon_dbg(VFS, "Close interrupted close\n"); @@ -844,14 +844,14 @@ smb2_handle_cancelled_close(struct cifs_tcon *tcon, __u64 persistent_fid, } int -smb2_handle_cancelled_mid(char *buffer, struct TCP_Server_Info *server) +smb2_handle_cancelled_mid(struct mid_q_entry *mid, struct TCP_Server_Info *server) { - struct smb2_sync_hdr *sync_hdr = (struct smb2_sync_hdr *)buffer; - struct smb2_create_rsp *rsp = (struct smb2_create_rsp *)buffer; + struct smb2_sync_hdr *sync_hdr = mid->resp_buf; + struct smb2_create_rsp *rsp = mid->resp_buf; struct cifs_tcon *tcon; int rc; - if (sync_hdr->Command != SMB2_CREATE || + if ((mid->optype & CIFS_CP_CREATE_CLOSE_OP) || sync_hdr->Command != SMB2_CREATE || sync_hdr->Status != STATUS_SUCCESS) return 0; diff --git a/fs/cifs/smb2ops.c b/fs/cifs/smb2ops.c index f19274857292..9bae7e8deb09 100644 --- a/fs/cifs/smb2ops.c +++ b/fs/cifs/smb2ops.c @@ -63,17 +63,19 @@ smb2_add_credits(struct TCP_Server_Info *server, const struct cifs_credits *credits, const int optype) { int *val, rc = -1; + int scredits, in_flight; unsigned int add = credits->value; unsigned int instance = credits->instance; bool reconnect_detected = false; + bool reconnect_with_invalid_credits = false; spin_lock(&server->req_lock); val = server->ops->get_credits_field(server, optype); /* eg found case where write overlapping reconnect messed up credits */ if (((optype & CIFS_OP_MASK) == CIFS_NEG_OP) && (*val != 0)) - trace_smb3_reconnect_with_invalid_credits(server->CurrentMid, - server->hostname, *val, add); + reconnect_with_invalid_credits = true; + if ((instance == 0) || (instance == server->reconnect_instance)) *val += add; else @@ -84,7 +86,9 @@ smb2_add_credits(struct TCP_Server_Info *server, pr_warn_once("server overflowed SMB3 credits\n"); } server->in_flight--; - if (server->in_flight == 0 && (optype & CIFS_OP_MASK) != CIFS_NEG_OP) + if (server->in_flight == 0 && + ((optype & CIFS_OP_MASK) != CIFS_NEG_OP) && + ((optype & CIFS_OP_MASK) != CIFS_SESS_OP)) rc = change_conf(server); /* * Sometimes server returns 0 credits on oplock break ack - we need to @@ -97,14 +101,26 @@ smb2_add_credits(struct TCP_Server_Info *server, server->oplock_credits++; } } + scredits = *val; + in_flight = server->in_flight; spin_unlock(&server->req_lock); wake_up(&server->request_q); if (reconnect_detected) { + trace_smb3_reconnect_detected(server->CurrentMid, + server->conn_id, server->hostname, scredits, add, in_flight); + cifs_dbg(FYI, "trying to put %d credits from the old server instance %d\n", add, instance); } + if (reconnect_with_invalid_credits) { + trace_smb3_reconnect_with_invalid_credits(server->CurrentMid, + server->conn_id, server->hostname, scredits, add, in_flight); + cifs_dbg(FYI, "Negotiate operation when server credits is non-zero. Optype: %d, server credits: %d, credits added: %d\n", + optype, scredits, add); + } + if (server->tcpStatus == CifsNeedReconnect || server->tcpStatus == CifsExiting) return; @@ -123,23 +139,30 @@ smb2_add_credits(struct TCP_Server_Info *server, cifs_dbg(FYI, "disabling oplocks\n"); break; default: - trace_smb3_add_credits(server->CurrentMid, - server->hostname, rc, add); - cifs_dbg(FYI, "%s: added %u credits total=%d\n", __func__, add, rc); + /* change_conf rebalanced credits for different types */ + break; } + + trace_smb3_add_credits(server->CurrentMid, + server->conn_id, server->hostname, scredits, add, in_flight); + cifs_dbg(FYI, "%s: added %u credits total=%d\n", __func__, add, scredits); } static void smb2_set_credits(struct TCP_Server_Info *server, const int val) { + int scredits, in_flight; + spin_lock(&server->req_lock); server->credits = val; if (val == 1) server->reconnect_instance++; + scredits = server->credits; + in_flight = server->in_flight; spin_unlock(&server->req_lock); trace_smb3_set_credits(server->CurrentMid, - server->hostname, val, val); + server->conn_id, server->hostname, scredits, val, in_flight); cifs_dbg(FYI, "%s: set %u credits\n", __func__, val); /* don't log while holding the lock */ @@ -171,7 +194,7 @@ smb2_wait_mtu_credits(struct TCP_Server_Info *server, unsigned int size, unsigned int *num, struct cifs_credits *credits) { int rc = 0; - unsigned int scredits; + unsigned int scredits, in_flight; spin_lock(&server->req_lock); while (1) { @@ -208,17 +231,18 @@ smb2_wait_mtu_credits(struct TCP_Server_Info *server, unsigned int size, DIV_ROUND_UP(*num, SMB2_MAX_BUFFER_SIZE); credits->instance = server->reconnect_instance; server->credits -= credits->value; - scredits = server->credits; server->in_flight++; if (server->in_flight > server->max_in_flight) server->max_in_flight = server->in_flight; break; } } + scredits = server->credits; + in_flight = server->in_flight; spin_unlock(&server->req_lock); trace_smb3_add_credits(server->CurrentMid, - server->hostname, scredits, -(credits->value)); + server->conn_id, server->hostname, scredits, -(credits->value), in_flight); cifs_dbg(FYI, "%s: removed %u credits total=%d\n", __func__, credits->value, scredits); @@ -231,14 +255,14 @@ smb2_adjust_credits(struct TCP_Server_Info *server, const unsigned int payload_size) { int new_val = DIV_ROUND_UP(payload_size, SMB2_MAX_BUFFER_SIZE); - int scredits; + int scredits, in_flight; if (!credits->value || credits->value == new_val) return 0; if (credits->value < new_val) { trace_smb3_too_many_credits(server->CurrentMid, - server->hostname, 0, credits->value - new_val); + server->conn_id, server->hostname, 0, credits->value - new_val, 0); cifs_server_dbg(VFS, "request has less credits (%d) than required (%d)", credits->value, new_val); @@ -248,9 +272,13 @@ smb2_adjust_credits(struct TCP_Server_Info *server, spin_lock(&server->req_lock); if (server->reconnect_instance != credits->instance) { + scredits = server->credits; + in_flight = server->in_flight; spin_unlock(&server->req_lock); + trace_smb3_reconnect_detected(server->CurrentMid, - server->hostname, 0, 0); + server->conn_id, server->hostname, scredits, + credits->value - new_val, in_flight); cifs_server_dbg(VFS, "trying to return %d credits to old session\n", credits->value - new_val); return -EAGAIN; @@ -258,15 +286,18 @@ smb2_adjust_credits(struct TCP_Server_Info *server, server->credits += credits->value - new_val; scredits = server->credits; + in_flight = server->in_flight; spin_unlock(&server->req_lock); wake_up(&server->request_q); - credits->value = new_val; trace_smb3_add_credits(server->CurrentMid, - server->hostname, scredits, credits->value - new_val); + server->conn_id, server->hostname, scredits, + credits->value - new_val, in_flight); cifs_dbg(FYI, "%s: adjust added %u credits total=%d\n", __func__, credits->value - new_val, scredits); + credits->value = new_val; + return 0; } @@ -1164,7 +1195,7 @@ smb2_set_ea(const unsigned int xid, struct cifs_tcon *tcon, struct TCP_Server_Info *server = cifs_pick_channel(ses); __le16 *utf16_path = NULL; int ea_name_len = strlen(ea_name); - int flags = 0; + int flags = CIFS_CP_CREATE_CLOSE_OP; int len; struct smb_rqst rqst[3]; int resp_buftype[3]; @@ -1542,7 +1573,7 @@ smb2_ioctl_query_info(const unsigned int xid, struct smb_query_info qi; struct smb_query_info __user *pqi; int rc = 0; - int flags = 0; + int flags = CIFS_CP_CREATE_CLOSE_OP; struct smb2_query_info_rsp *qi_rsp = NULL; struct smb2_ioctl_rsp *io_rsp = NULL; void *buffer = NULL; @@ -2369,7 +2400,7 @@ static bool smb2_is_status_pending(char *buf, struct TCP_Server_Info *server) { struct smb2_sync_hdr *shdr = (struct smb2_sync_hdr *)buf; - int scredits; + int scredits, in_flight; if (shdr->Status != STATUS_PENDING) return false; @@ -2378,11 +2409,13 @@ smb2_is_status_pending(char *buf, struct TCP_Server_Info *server) spin_lock(&server->req_lock); server->credits += le16_to_cpu(shdr->CreditRequest); scredits = server->credits; + in_flight = server->in_flight; spin_unlock(&server->req_lock); wake_up(&server->request_q); trace_smb3_add_credits(server->CurrentMid, - server->hostname, scredits, le16_to_cpu(shdr->CreditRequest)); + server->conn_id, server->hostname, scredits, + le16_to_cpu(shdr->CreditRequest), in_flight); cifs_dbg(FYI, "%s: status pending add %u credits total=%d\n", __func__, le16_to_cpu(shdr->CreditRequest), scredits); } @@ -2418,6 +2451,34 @@ smb2_is_status_io_timeout(char *buf) return false; } +static void +smb2_is_network_name_deleted(char *buf, struct TCP_Server_Info *server) +{ + struct smb2_sync_hdr *shdr = (struct smb2_sync_hdr *)buf; + struct list_head *tmp, *tmp1; + struct cifs_ses *ses; + struct cifs_tcon *tcon; + + if (shdr->Status != STATUS_NETWORK_NAME_DELETED) + return; + + spin_lock(&cifs_tcp_ses_lock); + list_for_each(tmp, &server->smb_ses_list) { + ses = list_entry(tmp, struct cifs_ses, smb_ses_list); + list_for_each(tmp1, &ses->tcon_list) { + tcon = list_entry(tmp1, struct cifs_tcon, tcon_list); + if (tcon->tid == shdr->TreeId) { + tcon->need_reconnect = true; + spin_unlock(&cifs_tcp_ses_lock); + pr_warn_once("Server share %s deleted.\n", + tcon->treeName); + return; + } + } + } + spin_unlock(&cifs_tcp_ses_lock); +} + static int smb2_oplock_response(struct cifs_tcon *tcon, struct cifs_fid *fid, struct cifsInodeInfo *cinode) @@ -2516,7 +2577,7 @@ smb2_query_info_compound(const unsigned int xid, struct cifs_tcon *tcon, { struct cifs_ses *ses = tcon->ses; struct TCP_Server_Info *server = cifs_pick_channel(ses); - int flags = 0; + int flags = CIFS_CP_CREATE_CLOSE_OP; struct smb_rqst rqst[3]; int resp_buftype[3]; struct kvec rsp_iov[3]; @@ -2914,7 +2975,7 @@ smb2_query_symlink(const unsigned int xid, struct cifs_tcon *tcon, unsigned int sub_offset; unsigned int print_len; unsigned int print_offset; - int flags = 0; + int flags = CIFS_CP_CREATE_CLOSE_OP; struct smb_rqst rqst[3]; int resp_buftype[3]; struct kvec rsp_iov[3]; @@ -3096,7 +3157,7 @@ smb2_query_reparse_tag(const unsigned int xid, struct cifs_tcon *tcon, struct cifs_open_parms oparms; struct cifs_fid fid; struct TCP_Server_Info *server = cifs_pick_channel(tcon->ses); - int flags = 0; + int flags = CIFS_CP_CREATE_CLOSE_OP; struct smb_rqst rqst[3]; int resp_buftype[3]; struct kvec rsp_iov[3]; @@ -4605,6 +4666,10 @@ static void smb2_decrypt_offload(struct work_struct *work) #ifdef CONFIG_CIFS_STATS2 mid->when_received = jiffies; #endif + if (dw->server->ops->is_network_name_deleted) + dw->server->ops->is_network_name_deleted(dw->buf, + dw->server); + mid->callback(mid); } else { spin_lock(&GlobalMid_Lock); @@ -4723,6 +4788,12 @@ non_offloaded_decrypt: rc = handle_read_data(server, *mid, buf, server->vals->read_rsp_size, pages, npages, len, false); + if (rc >= 0) { + if (server->ops->is_network_name_deleted) { + server->ops->is_network_name_deleted(buf, + server); + } + } } free_pages: @@ -5072,6 +5143,7 @@ struct smb_version_operations smb20_operations = { .fiemap = smb3_fiemap, .llseek = smb3_llseek, .is_status_io_timeout = smb2_is_status_io_timeout, + .is_network_name_deleted = smb2_is_network_name_deleted, }; struct smb_version_operations smb21_operations = { @@ -5173,6 +5245,7 @@ struct smb_version_operations smb21_operations = { .fiemap = smb3_fiemap, .llseek = smb3_llseek, .is_status_io_timeout = smb2_is_status_io_timeout, + .is_network_name_deleted = smb2_is_network_name_deleted, }; struct smb_version_operations smb30_operations = { @@ -5286,6 +5359,7 @@ struct smb_version_operations smb30_operations = { .fiemap = smb3_fiemap, .llseek = smb3_llseek, .is_status_io_timeout = smb2_is_status_io_timeout, + .is_network_name_deleted = smb2_is_network_name_deleted, }; struct smb_version_operations smb311_operations = { @@ -5399,6 +5473,7 @@ struct smb_version_operations smb311_operations = { .fiemap = smb3_fiemap, .llseek = smb3_llseek, .is_status_io_timeout = smb2_is_status_io_timeout, + .is_network_name_deleted = smb2_is_network_name_deleted, }; struct smb_version_values smb20_values = { diff --git a/fs/cifs/smb2pdu.c b/fs/cifs/smb2pdu.c index 794fc3b68b4f..2199a9bfae8f 100644 --- a/fs/cifs/smb2pdu.c +++ b/fs/cifs/smb2pdu.c @@ -814,8 +814,9 @@ SMB2_negotiate(const unsigned int xid, struct cifs_ses *ses) SMB3ANY_VERSION_STRING) == 0) { req->Dialects[0] = cpu_to_le16(SMB30_PROT_ID); req->Dialects[1] = cpu_to_le16(SMB302_PROT_ID); - req->DialectCount = cpu_to_le16(2); - total_len += 4; + req->Dialects[2] = cpu_to_le16(SMB311_PROT_ID); + req->DialectCount = cpu_to_le16(3); + total_len += 6; } else if (strcmp(server->vals->version_string, SMBDEFAULT_VERSION_STRING) == 0) { req->Dialects[0] = cpu_to_le16(SMB21_PROT_ID); @@ -849,6 +850,8 @@ SMB2_negotiate(const unsigned int xid, struct cifs_ses *ses) SMB2_CLIENT_GUID_SIZE); if ((server->vals->protocol_id == SMB311_PROT_ID) || (strcmp(server->vals->version_string, + SMB3ANY_VERSION_STRING) == 0) || + (strcmp(server->vals->version_string, SMBDEFAULT_VERSION_STRING) == 0)) assemble_neg_contexts(req, server, &total_len); } @@ -883,6 +886,10 @@ SMB2_negotiate(const unsigned int xid, struct cifs_ses *ses) cifs_server_dbg(VFS, "SMB2.1 dialect returned but not requested\n"); return -EIO; + } else if (rsp->DialectRevision == cpu_to_le16(SMB311_PROT_ID)) { + /* ops set to 3.0 by default for default so update */ + server->ops = &smb311_operations; + server->vals = &smb311_values; } } else if (strcmp(server->vals->version_string, SMBDEFAULT_VERSION_STRING) == 0) { @@ -1042,10 +1049,11 @@ int smb3_validate_negotiate(const unsigned int xid, struct cifs_tcon *tcon) SMB3ANY_VERSION_STRING) == 0) { pneg_inbuf->Dialects[0] = cpu_to_le16(SMB30_PROT_ID); pneg_inbuf->Dialects[1] = cpu_to_le16(SMB302_PROT_ID); - pneg_inbuf->DialectCount = cpu_to_le16(2); - /* structure is big enough for 3 dialects, sending only 2 */ + pneg_inbuf->Dialects[2] = cpu_to_le16(SMB311_PROT_ID); + pneg_inbuf->DialectCount = cpu_to_le16(3); + /* SMB 2.1 not included so subtract one dialect from len */ inbuflen = sizeof(*pneg_inbuf) - - (2 * sizeof(pneg_inbuf->Dialects[0])); + (sizeof(pneg_inbuf->Dialects[0])); } else if (strcmp(server->vals->version_string, SMBDEFAULT_VERSION_STRING) == 0) { pneg_inbuf->Dialects[0] = cpu_to_le16(SMB21_PROT_ID); @@ -1053,7 +1061,7 @@ int smb3_validate_negotiate(const unsigned int xid, struct cifs_tcon *tcon) pneg_inbuf->Dialects[2] = cpu_to_le16(SMB302_PROT_ID); pneg_inbuf->Dialects[3] = cpu_to_le16(SMB311_PROT_ID); pneg_inbuf->DialectCount = cpu_to_le16(4); - /* structure is big enough for 3 dialects */ + /* structure is big enough for 4 dialects */ inbuflen = sizeof(*pneg_inbuf); } else { /* otherwise specific dialect was requested */ @@ -1253,7 +1261,7 @@ SMB2_sess_sendreceive(struct SMB2_sess_data *sess_data) cifs_ses_server(sess_data->ses), &rqst, &sess_data->buf0_type, - CIFS_LOG_ERROR | CIFS_NEG_OP, &rsp_iov); + CIFS_LOG_ERROR | CIFS_SESS_OP, &rsp_iov); cifs_small_buf_release(sess_data->iov[0].iov_base); memcpy(&sess_data->iov[0], &rsp_iov, sizeof(struct kvec)); @@ -4033,8 +4041,7 @@ smb2_async_readv(struct cifs_readdata *rdata) if (rdata->credits.value > 0) { shdr->CreditCharge = cpu_to_le16(DIV_ROUND_UP(rdata->bytes, SMB2_MAX_BUFFER_SIZE)); - shdr->CreditRequest = - cpu_to_le16(le16_to_cpu(shdr->CreditCharge) + 1); + shdr->CreditRequest = cpu_to_le16(le16_to_cpu(shdr->CreditCharge) + 8); rc = adjust_credits(server, &rdata->credits, rdata->bytes); if (rc) @@ -4340,8 +4347,7 @@ smb2_async_writev(struct cifs_writedata *wdata, if (wdata->credits.value > 0) { shdr->CreditCharge = cpu_to_le16(DIV_ROUND_UP(wdata->bytes, SMB2_MAX_BUFFER_SIZE)); - shdr->CreditRequest = - cpu_to_le16(le16_to_cpu(shdr->CreditCharge) + 1); + shdr->CreditRequest = cpu_to_le16(le16_to_cpu(shdr->CreditCharge) + 8); rc = adjust_credits(server, &wdata->credits, wdata->bytes); if (rc) diff --git a/fs/cifs/smb2proto.h b/fs/cifs/smb2proto.h index 9565e27681a5..a2eb34a8d9c9 100644 --- a/fs/cifs/smb2proto.h +++ b/fs/cifs/smb2proto.h @@ -246,8 +246,7 @@ extern int SMB2_oplock_break(const unsigned int xid, struct cifs_tcon *tcon, extern int smb2_handle_cancelled_close(struct cifs_tcon *tcon, __u64 persistent_fid, __u64 volatile_fid); -extern int smb2_handle_cancelled_mid(char *buffer, - struct TCP_Server_Info *server); +extern int smb2_handle_cancelled_mid(struct mid_q_entry *mid, struct TCP_Server_Info *server); void smb2_cancelled_close_fid(struct work_struct *work); extern int SMB2_QFS_info(const unsigned int xid, struct cifs_tcon *tcon, u64 persistent_file_id, u64 volatile_file_id, diff --git a/fs/cifs/trace.h b/fs/cifs/trace.h index c3d1a584f251..d6df908dccad 100644 --- a/fs/cifs/trace.h +++ b/fs/cifs/trace.h @@ -851,17 +851,21 @@ DEFINE_SMB3_LEASE_ERR_EVENT(lease_err); DECLARE_EVENT_CLASS(smb3_reconnect_class, TP_PROTO(__u64 currmid, + __u64 conn_id, char *hostname), - TP_ARGS(currmid, hostname), + TP_ARGS(currmid, conn_id, hostname), TP_STRUCT__entry( __field(__u64, currmid) + __field(__u64, conn_id) __field(char *, hostname) ), TP_fast_assign( __entry->currmid = currmid; + __entry->conn_id = conn_id; __entry->hostname = hostname; ), - TP_printk("server=%s current_mid=0x%llx", + TP_printk("conn_id=0x%llx server=%s current_mid=%llu", + __entry->conn_id, __entry->hostname, __entry->currmid) ) @@ -869,44 +873,56 @@ DECLARE_EVENT_CLASS(smb3_reconnect_class, #define DEFINE_SMB3_RECONNECT_EVENT(name) \ DEFINE_EVENT(smb3_reconnect_class, smb3_##name, \ TP_PROTO(__u64 currmid, \ - char *hostname), \ - TP_ARGS(currmid, hostname)) + __u64 conn_id, \ + char *hostname), \ + TP_ARGS(currmid, conn_id, hostname)) DEFINE_SMB3_RECONNECT_EVENT(reconnect); DEFINE_SMB3_RECONNECT_EVENT(partial_send_reconnect); DECLARE_EVENT_CLASS(smb3_credit_class, TP_PROTO(__u64 currmid, + __u64 conn_id, char *hostname, int credits, - int credits_to_add), - TP_ARGS(currmid, hostname, credits, credits_to_add), + int credits_to_add, + int in_flight), + TP_ARGS(currmid, conn_id, hostname, credits, credits_to_add, in_flight), TP_STRUCT__entry( __field(__u64, currmid) + __field(__u64, conn_id) __field(char *, hostname) __field(int, credits) __field(int, credits_to_add) + __field(int, in_flight) ), TP_fast_assign( __entry->currmid = currmid; + __entry->conn_id = conn_id; __entry->hostname = hostname; __entry->credits = credits; __entry->credits_to_add = credits_to_add; + __entry->in_flight = in_flight; ), - TP_printk("server=%s current_mid=0x%llx credits=%d credits_to_add=%d", + TP_printk("conn_id=0x%llx server=%s current_mid=%llu " + "credits=%d credit_change=%d in_flight=%d", + __entry->conn_id, __entry->hostname, __entry->currmid, __entry->credits, - __entry->credits_to_add) + __entry->credits_to_add, + __entry->in_flight) ) #define DEFINE_SMB3_CREDIT_EVENT(name) \ DEFINE_EVENT(smb3_credit_class, smb3_##name, \ TP_PROTO(__u64 currmid, \ + __u64 conn_id, \ char *hostname, \ int credits, \ - int credits_to_add), \ - TP_ARGS(currmid, hostname, credits, credits_to_add)) + int credits_to_add, \ + int in_flight), \ + TP_ARGS(currmid, conn_id, hostname, credits, credits_to_add, in_flight)) DEFINE_SMB3_CREDIT_EVENT(reconnect_with_invalid_credits); DEFINE_SMB3_CREDIT_EVENT(reconnect_detected); diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c index 4a2b836eb017..c1725b55f364 100644 --- a/fs/cifs/transport.c +++ b/fs/cifs/transport.c @@ -101,7 +101,7 @@ static void _cifs_mid_q_entry_release(struct kref *refcount) if (midEntry->resp_buf && (midEntry->mid_flags & MID_WAIT_CANCELLED) && midEntry->mid_state == MID_RESPONSE_RECEIVED && server->ops->handle_cancelled_mid) - server->ops->handle_cancelled_mid(midEntry->resp_buf, server); + server->ops->handle_cancelled_mid(midEntry, server); midEntry->mid_state = MID_FREE; atomic_dec(&midCount); @@ -445,7 +445,7 @@ unmask: */ server->tcpStatus = CifsNeedReconnect; trace_smb3_partial_send_reconnect(server->CurrentMid, - server->hostname); + server->conn_id, server->hostname); } smbd_done: if (rc < 0 && rc != -EINTR) @@ -527,7 +527,7 @@ wait_for_free_credits(struct TCP_Server_Info *server, const int num_credits, int *credits; int optype; long int t; - int scredits = server->credits; + int scredits, in_flight; if (timeout < 0) t = MAX_JIFFY_OFFSET; @@ -551,23 +551,39 @@ wait_for_free_credits(struct TCP_Server_Info *server, const int num_credits, server->max_in_flight = server->in_flight; *credits -= 1; *instance = server->reconnect_instance; + scredits = *credits; + in_flight = server->in_flight; spin_unlock(&server->req_lock); + + trace_smb3_add_credits(server->CurrentMid, + server->conn_id, server->hostname, scredits, -1, in_flight); + cifs_dbg(FYI, "%s: remove %u credits total=%d\n", + __func__, 1, scredits); + return 0; } while (1) { if (*credits < num_credits) { + scredits = *credits; spin_unlock(&server->req_lock); + cifs_num_waiters_inc(server); rc = wait_event_killable_timeout(server->request_q, has_credits(server, credits, num_credits), t); cifs_num_waiters_dec(server); if (!rc) { + spin_lock(&server->req_lock); + scredits = *credits; + in_flight = server->in_flight; + spin_unlock(&server->req_lock); + trace_smb3_credit_timeout(server->CurrentMid, - server->hostname, num_credits, 0); + server->conn_id, server->hostname, scredits, + num_credits, in_flight); cifs_server_dbg(VFS, "wait timed out after %d ms\n", - timeout); - return -ENOTSUPP; + timeout); + return -EBUSY; } if (rc == -ERESTARTSYS) return -ERESTARTSYS; @@ -595,6 +611,7 @@ wait_for_free_credits(struct TCP_Server_Info *server, const int num_credits, server->in_flight > 2 * MAX_COMPOUND && *credits <= MAX_COMPOUND) { spin_unlock(&server->req_lock); + cifs_num_waiters_inc(server); rc = wait_event_killable_timeout( server->request_q, @@ -603,13 +620,18 @@ wait_for_free_credits(struct TCP_Server_Info *server, const int num_credits, t); cifs_num_waiters_dec(server); if (!rc) { + spin_lock(&server->req_lock); + scredits = *credits; + in_flight = server->in_flight; + spin_unlock(&server->req_lock); + trace_smb3_credit_timeout( - server->CurrentMid, - server->hostname, num_credits, - 0); + server->CurrentMid, + server->conn_id, server->hostname, + scredits, num_credits, in_flight); cifs_server_dbg(VFS, "wait timed out after %d ms\n", - timeout); - return -ENOTSUPP; + timeout); + return -EBUSY; } if (rc == -ERESTARTSYS) return -ERESTARTSYS; @@ -625,16 +647,18 @@ wait_for_free_credits(struct TCP_Server_Info *server, const int num_credits, /* update # of requests on the wire to server */ if ((flags & CIFS_TIMEOUT_MASK) != CIFS_BLOCKING_OP) { *credits -= num_credits; - scredits = *credits; server->in_flight += num_credits; if (server->in_flight > server->max_in_flight) server->max_in_flight = server->in_flight; *instance = server->reconnect_instance; } + scredits = *credits; + in_flight = server->in_flight; spin_unlock(&server->req_lock); trace_smb3_add_credits(server->CurrentMid, - server->hostname, scredits, -(num_credits)); + server->conn_id, server->hostname, scredits, + -(num_credits), in_flight); cifs_dbg(FYI, "%s: remove %u credits total=%d\n", __func__, num_credits, scredits); break; @@ -656,13 +680,13 @@ wait_for_compound_request(struct TCP_Server_Info *server, int num, const int flags, unsigned int *instance) { int *credits; - int scredits, sin_flight; + int scredits, in_flight; credits = server->ops->get_credits_field(server, flags & CIFS_OP_MASK); spin_lock(&server->req_lock); scredits = *credits; - sin_flight = server->in_flight; + in_flight = server->in_flight; if (*credits < num) { /* @@ -684,10 +708,11 @@ wait_for_compound_request(struct TCP_Server_Info *server, int num, if (server->in_flight == 0) { spin_unlock(&server->req_lock); trace_smb3_insufficient_credits(server->CurrentMid, - server->hostname, scredits, sin_flight); + server->conn_id, server->hostname, scredits, + num, in_flight); cifs_dbg(FYI, "%s: %d requests in flight, needed %d total=%d\n", - __func__, sin_flight, num, scredits); - return -ENOTSUPP; + __func__, in_flight, num, scredits); + return -EDEADLK; } } spin_unlock(&server->req_lock); @@ -1171,9 +1196,12 @@ compound_send_recv(const unsigned int xid, struct cifs_ses *ses, /* * Compounding is never used during session establish. */ - if ((ses->status == CifsNew) || (optype & CIFS_NEG_OP)) + if ((ses->status == CifsNew) || (optype & CIFS_NEG_OP) || (optype & CIFS_SESS_OP)) { + mutex_lock(&server->srv_mutex); smb311_update_preauth_hash(ses, rqst[0].rq_iov, rqst[0].rq_nvec); + mutex_unlock(&server->srv_mutex); + } for (i = 0; i < num_rqst; i++) { rc = wait_for_response(server, midQ[i]); @@ -1182,7 +1210,7 @@ compound_send_recv(const unsigned int xid, struct cifs_ses *ses, } if (rc != 0) { for (; i < num_rqst; i++) { - cifs_server_dbg(VFS, "Cancelling wait for mid %llu cmd: %d\n", + cifs_server_dbg(FYI, "Cancelling wait for mid %llu cmd: %d\n", midQ[i]->mid, le16_to_cpu(midQ[i]->command)); send_cancel(server, &rqst[i], midQ[i]); spin_lock(&GlobalMid_Lock); @@ -1236,12 +1264,14 @@ compound_send_recv(const unsigned int xid, struct cifs_ses *ses, /* * Compounding is never used during session establish. */ - if ((ses->status == CifsNew) || (optype & CIFS_NEG_OP)) { + if ((ses->status == CifsNew) || (optype & CIFS_NEG_OP) || (optype & CIFS_SESS_OP)) { struct kvec iov = { .iov_base = resp_iov[0].iov_base, .iov_len = resp_iov[0].iov_len }; + mutex_lock(&server->srv_mutex); smb311_update_preauth_hash(ses, &iov, 1); + mutex_unlock(&server->srv_mutex); } out: diff --git a/fs/cifs/xattr.c b/fs/cifs/xattr.c index 6b658a1172ef..41a611e76bb7 100644 --- a/fs/cifs/xattr.c +++ b/fs/cifs/xattr.c @@ -101,6 +101,7 @@ static int cifs_creation_time_set(unsigned int xid, struct cifs_tcon *pTcon, } static int cifs_xattr_set(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *dentry, struct inode *inode, const char *name, const void *value, size_t size, int flags) diff --git a/fs/coda/coda_linux.h b/fs/coda/coda_linux.h index d5ebd36fb2cc..e7b27754ce78 100644 --- a/fs/coda/coda_linux.h +++ b/fs/coda/coda_linux.h @@ -46,10 +46,12 @@ extern const struct file_operations coda_ioctl_operations; /* operations shared over more than one file */ int coda_open(struct inode *i, struct file *f); int coda_release(struct inode *i, struct file *f); -int coda_permission(struct inode *inode, int mask); +int coda_permission(struct user_namespace *mnt_userns, struct inode *inode, + int mask); int coda_revalidate_inode(struct inode *); -int coda_getattr(const struct path *, struct kstat *, u32, unsigned int); -int coda_setattr(struct dentry *, struct iattr *); +int coda_getattr(struct user_namespace *, const struct path *, struct kstat *, + u32, unsigned int); +int coda_setattr(struct user_namespace *, struct dentry *, struct iattr *); /* this file: heloers */ char *coda_f2s(struct CodaFid *f); diff --git a/fs/coda/dir.c b/fs/coda/dir.c index ca40c2556ba6..d69989c1bac3 100644 --- a/fs/coda/dir.c +++ b/fs/coda/dir.c @@ -73,7 +73,8 @@ static struct dentry *coda_lookup(struct inode *dir, struct dentry *entry, unsig } -int coda_permission(struct inode *inode, int mask) +int coda_permission(struct user_namespace *mnt_userns, struct inode *inode, + int mask) { int error; @@ -132,7 +133,8 @@ static inline void coda_dir_drop_nlink(struct inode *dir) } /* creation routines: create, mknod, mkdir, link, symlink */ -static int coda_create(struct inode *dir, struct dentry *de, umode_t mode, bool excl) +static int coda_create(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *de, umode_t mode, bool excl) { int error; const char *name=de->d_name.name; @@ -164,7 +166,8 @@ err_out: return error; } -static int coda_mkdir(struct inode *dir, struct dentry *de, umode_t mode) +static int coda_mkdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *de, umode_t mode) { struct inode *inode; struct coda_vattr attrs; @@ -225,7 +228,8 @@ static int coda_link(struct dentry *source_de, struct inode *dir_inode, } -static int coda_symlink(struct inode *dir_inode, struct dentry *de, +static int coda_symlink(struct user_namespace *mnt_userns, + struct inode *dir_inode, struct dentry *de, const char *symname) { const char *name = de->d_name.name; @@ -291,9 +295,9 @@ static int coda_rmdir(struct inode *dir, struct dentry *de) } /* rename */ -static int coda_rename(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry, - unsigned int flags) +static int coda_rename(struct user_namespace *mnt_userns, struct inode *old_dir, + struct dentry *old_dentry, struct inode *new_dir, + struct dentry *new_dentry, unsigned int flags) { const char *old_name = old_dentry->d_name.name; const char *new_name = new_dentry->d_name.name; diff --git a/fs/coda/inode.c b/fs/coda/inode.c index b1c70e2b9b1e..d9f1bd7153df 100644 --- a/fs/coda/inode.c +++ b/fs/coda/inode.c @@ -251,16 +251,17 @@ static void coda_evict_inode(struct inode *inode) coda_cache_clear_inode(inode); } -int coda_getattr(const struct path *path, struct kstat *stat, - u32 request_mask, unsigned int flags) +int coda_getattr(struct user_namespace *mnt_userns, const struct path *path, + struct kstat *stat, u32 request_mask, unsigned int flags) { int err = coda_revalidate_inode(d_inode(path->dentry)); if (!err) - generic_fillattr(d_inode(path->dentry), stat); + generic_fillattr(&init_user_ns, d_inode(path->dentry), stat); return err; } -int coda_setattr(struct dentry *de, struct iattr *iattr) +int coda_setattr(struct user_namespace *mnt_userns, struct dentry *de, + struct iattr *iattr) { struct inode *inode = d_inode(de); struct coda_vattr vattr; diff --git a/fs/coda/pioctl.c b/fs/coda/pioctl.c index 3aec27e5eb82..cb9fd59a688c 100644 --- a/fs/coda/pioctl.c +++ b/fs/coda/pioctl.c @@ -24,7 +24,8 @@ #include "coda_linux.h" /* pioctl ops */ -static int coda_ioctl_permission(struct inode *inode, int mask); +static int coda_ioctl_permission(struct user_namespace *mnt_userns, + struct inode *inode, int mask); static long coda_pioctl(struct file *filp, unsigned int cmd, unsigned long user_data); @@ -40,7 +41,8 @@ const struct file_operations coda_ioctl_operations = { }; /* the coda pioctl inode ops */ -static int coda_ioctl_permission(struct inode *inode, int mask) +static int coda_ioctl_permission(struct user_namespace *mnt_userns, + struct inode *inode, int mask) { return (mask & MAY_EXEC) ? -EACCES : 0; } diff --git a/fs/compat_binfmt_elf.c b/fs/compat_binfmt_elf.c index 2c557229696a..95e72d271b95 100644 --- a/fs/compat_binfmt_elf.c +++ b/fs/compat_binfmt_elf.c @@ -50,6 +50,7 @@ * which requires asm/elf.h to define compat_elf_gregset_t et al. */ #define elf_prstatus compat_elf_prstatus +#define elf_prstatus_common compat_elf_prstatus_common #define elf_prpsinfo compat_elf_prpsinfo #undef ns_to_kernel_old_timeval @@ -61,7 +62,6 @@ * differ from the native ones, or omitted when they match. */ -#undef ELF_ARCH #undef elf_check_arch #define elf_check_arch compat_elf_check_arch @@ -90,11 +90,6 @@ #define ELF_ET_DYN_BASE COMPAT_ELF_ET_DYN_BASE #endif -#ifdef COMPAT_ELF_EXEC_PAGESIZE -#undef ELF_EXEC_PAGESIZE -#define ELF_EXEC_PAGESIZE COMPAT_ELF_EXEC_PAGESIZE -#endif - #ifdef COMPAT_ELF_PLAT_INIT #undef ELF_PLAT_INIT #define ELF_PLAT_INIT COMPAT_ELF_PLAT_INIT diff --git a/fs/configfs/configfs_internal.h b/fs/configfs/configfs_internal.h index 22dce2d35a4b..9a3aed249692 100644 --- a/fs/configfs/configfs_internal.h +++ b/fs/configfs/configfs_internal.h @@ -79,7 +79,8 @@ extern void configfs_hash_and_remove(struct dentry * dir, const char * name); extern const unsigned char * configfs_get_name(struct configfs_dirent *sd); extern void configfs_drop_dentry(struct configfs_dirent *sd, struct dentry *parent); -extern int configfs_setattr(struct dentry *dentry, struct iattr *iattr); +extern int configfs_setattr(struct user_namespace *mnt_userns, + struct dentry *dentry, struct iattr *iattr); extern struct dentry *configfs_pin_fs(void); extern void configfs_release_fs(void); @@ -92,7 +93,8 @@ extern const struct inode_operations configfs_root_inode_operations; extern const struct inode_operations configfs_symlink_inode_operations; extern const struct dentry_operations configfs_dentry_ops; -extern int configfs_symlink(struct inode *dir, struct dentry *dentry, +extern int configfs_symlink(struct user_namespace *mnt_userns, + struct inode *dir, struct dentry *dentry, const char *symname); extern int configfs_unlink(struct inode *dir, struct dentry *dentry); diff --git a/fs/configfs/dir.c b/fs/configfs/dir.c index b839dd1b459f..b6098e02e20b 100644 --- a/fs/configfs/dir.c +++ b/fs/configfs/dir.c @@ -1268,7 +1268,8 @@ out_root_unlock: } EXPORT_SYMBOL(configfs_depend_item_unlocked); -static int configfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) +static int configfs_mkdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { int ret = 0; int module_got = 0; diff --git a/fs/configfs/file.c b/fs/configfs/file.c index 1f0270229d7b..da8351d1e455 100644 --- a/fs/configfs/file.c +++ b/fs/configfs/file.c @@ -378,7 +378,7 @@ static int __configfs_open_file(struct inode *inode, struct file *file, int type attr = to_attr(dentry); if (!attr) - goto out_put_item; + goto out_free_buffer; if (type & CONFIGFS_ITEM_BIN_ATTR) { buffer->bin_attr = to_bin_attr(dentry); @@ -391,7 +391,7 @@ static int __configfs_open_file(struct inode *inode, struct file *file, int type /* Grab the module reference for this attribute if we have one */ error = -ENODEV; if (!try_module_get(buffer->owner)) - goto out_put_item; + goto out_free_buffer; error = -EACCES; if (!buffer->item->ci_type) @@ -435,8 +435,6 @@ static int __configfs_open_file(struct inode *inode, struct file *file, int type out_put_module: module_put(buffer->owner); -out_put_item: - config_item_put(buffer->item); out_free_buffer: up_read(&frag->frag_sem); kfree(buffer); diff --git a/fs/configfs/inode.c b/fs/configfs/inode.c index 8bd6a883c94c..42c348bb2903 100644 --- a/fs/configfs/inode.c +++ b/fs/configfs/inode.c @@ -40,7 +40,8 @@ static const struct inode_operations configfs_inode_operations ={ .setattr = configfs_setattr, }; -int configfs_setattr(struct dentry * dentry, struct iattr * iattr) +int configfs_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *iattr) { struct inode * inode = d_inode(dentry); struct configfs_dirent * sd = dentry->d_fsdata; @@ -67,7 +68,7 @@ int configfs_setattr(struct dentry * dentry, struct iattr * iattr) } /* attributes were changed atleast once in past */ - error = simple_setattr(dentry, iattr); + error = simple_setattr(mnt_userns, dentry, iattr); if (error) return error; diff --git a/fs/configfs/symlink.c b/fs/configfs/symlink.c index cb61467478ca..77c854364e60 100644 --- a/fs/configfs/symlink.c +++ b/fs/configfs/symlink.c @@ -139,7 +139,8 @@ static int get_target(const char *symname, struct path *path, } -int configfs_symlink(struct inode *dir, struct dentry *dentry, const char *symname) +int configfs_symlink(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, const char *symname) { int ret; struct path path; @@ -197,7 +198,8 @@ int configfs_symlink(struct inode *dir, struct dentry *dentry, const char *symna if (dentry->d_inode || d_unhashed(dentry)) ret = -EEXIST; else - ret = inode_permission(dir, MAY_WRITE | MAY_EXEC); + ret = inode_permission(&init_user_ns, dir, + MAY_WRITE | MAY_EXEC); if (!ret) ret = type->ct_item_ops->allow_link(parent_item, target_item); if (!ret) { diff --git a/fs/coredump.c b/fs/coredump.c index a2f6ecc8e345..1c0fdc1aa70b 100644 --- a/fs/coredump.c +++ b/fs/coredump.c @@ -703,6 +703,7 @@ void do_coredump(const kernel_siginfo_t *siginfo) goto close_fail; } } else { + struct user_namespace *mnt_userns; struct inode *inode; int open_flags = O_CREAT | O_RDWR | O_NOFOLLOW | O_LARGEFILE | O_EXCL; @@ -780,13 +781,15 @@ void do_coredump(const kernel_siginfo_t *siginfo) * a process dumps core while its cwd is e.g. on a vfat * filesystem. */ - if (!uid_eq(inode->i_uid, current_fsuid())) + mnt_userns = file_mnt_user_ns(cprm.file); + if (!uid_eq(i_uid_into_mnt(mnt_userns, inode), current_fsuid())) goto close_fail; if ((inode->i_mode & 0677) != 0600) goto close_fail; if (!(cprm.file->f_mode & FMODE_CAN_WRITE)) goto close_fail; - if (do_truncate(cprm.file->f_path.dentry, 0, 0, cprm.file)) + if (do_truncate(mnt_userns, cprm.file->f_path.dentry, + 0, 0, cprm.file)) goto close_fail; } @@ -894,10 +897,10 @@ int dump_user_range(struct coredump_params *cprm, unsigned long start, */ page = get_dump_page(addr); if (page) { - void *kaddr = kmap(page); + void *kaddr = kmap_local_page(page); stop = !dump_emit(cprm, kaddr, PAGE_SIZE); - kunmap(page); + kunmap_local(kaddr); put_page(page); } else { stop = !dump_skip(cprm, PAGE_SIZE); @@ -931,7 +934,8 @@ void dump_truncate(struct coredump_params *cprm) if (file->f_op->llseek && file->f_op->llseek != no_llseek) { offset = file->f_op->llseek(file, 0, SEEK_CUR); if (i_size_read(file->f_mapping->host) < offset) - do_truncate(file->f_path.dentry, offset, 0, file); + do_truncate(file_mnt_user_ns(file), file->f_path.dentry, + offset, 0, file); } } EXPORT_SYMBOL(dump_truncate); diff --git a/fs/cramfs/inode.c b/fs/cramfs/inode.c index 4b90cfd1ec36..2be65269a987 100644 --- a/fs/cramfs/inode.c +++ b/fs/cramfs/inode.c @@ -392,8 +392,7 @@ static int cramfs_physmem_mmap(struct file *file, struct vm_area_struct *vma) /* Don't map the last page if it contains some other data */ if (pgoff + pages == max_pages && cramfs_last_page_is_shared(inode)) { - pr_debug("mmap: %s: last page is shared\n", - file_dentry(file)->d_name.name); + pr_debug("mmap: %pD: last page is shared\n", file); pages--; } @@ -430,16 +429,15 @@ static int cramfs_physmem_mmap(struct file *file, struct vm_area_struct *vma) } if (!ret) - pr_debug("mapped %s[%lu] at 0x%08lx (%u/%lu pages) " - "to vma 0x%08lx, page_prot 0x%llx\n", - file_dentry(file)->d_name.name, pgoff, - address, pages, vma_pages(vma), vma->vm_start, + pr_debug("mapped %pD[%lu] at 0x%08lx (%u/%lu pages) " + "to vma 0x%08lx, page_prot 0x%llx\n", file, + pgoff, address, pages, vma_pages(vma), vma->vm_start, (unsigned long long)pgprot_val(vma->vm_page_prot)); return ret; bailout: - pr_debug("%s[%lu]: direct mmap impossible: %s\n", - file_dentry(file)->d_name.name, pgoff, bailout_reason); + pr_debug("%pD[%lu]: direct mmap impossible: %s\n", + file, pgoff, bailout_reason); /* Didn't manage any direct map, but normal paging is still possible */ return 0; } @@ -469,8 +467,8 @@ static unsigned long cramfs_physmem_get_unmapped_area(struct file *file, if (!offset || block_pages != pages) return -ENOSYS; addr = sbi->linear_phys_addr + offset; - pr_debug("get_unmapped for %s ofs %#lx siz %lu at 0x%08lx\n", - file_dentry(file)->d_name.name, pgoff*PAGE_SIZE, len, addr); + pr_debug("get_unmapped for %pD ofs %#lx siz %lu at 0x%08lx\n", + file, pgoff*PAGE_SIZE, len, addr); return addr; } diff --git a/fs/crypto/bio.c b/fs/crypto/bio.c index b048a0e38516..68a2de6b5a9b 100644 --- a/fs/crypto/bio.c +++ b/fs/crypto/bio.c @@ -52,7 +52,7 @@ static int fscrypt_zeroout_range_inline_crypt(const struct inode *inode, int num_pages = 0; /* This always succeeds since __GFP_DIRECT_RECLAIM is set. */ - bio = bio_alloc(GFP_NOFS, BIO_MAX_PAGES); + bio = bio_alloc(GFP_NOFS, BIO_MAX_VECS); while (len) { unsigned int blocks_this_page = min(len, blocks_per_page); @@ -74,7 +74,7 @@ static int fscrypt_zeroout_range_inline_crypt(const struct inode *inode, len -= blocks_this_page; lblk += blocks_this_page; pblk += blocks_this_page; - if (num_pages == BIO_MAX_PAGES || !len || + if (num_pages == BIO_MAX_VECS || !len || !fscrypt_mergeable_bio(bio, inode, lblk)) { err = submit_bio_wait(bio); if (err) @@ -126,7 +126,7 @@ int fscrypt_zeroout_range(const struct inode *inode, pgoff_t lblk, return fscrypt_zeroout_range_inline_crypt(inode, lblk, pblk, len); - BUILD_BUG_ON(ARRAY_SIZE(pages) > BIO_MAX_PAGES); + BUILD_BUG_ON(ARRAY_SIZE(pages) > BIO_MAX_VECS); nr_pages = min_t(unsigned int, ARRAY_SIZE(pages), (len + blocks_per_page - 1) >> blocks_per_page_bits); diff --git a/fs/crypto/policy.c b/fs/crypto/policy.c index a51cef6bd27f..ed3d623724cd 100644 --- a/fs/crypto/policy.c +++ b/fs/crypto/policy.c @@ -465,7 +465,7 @@ int fscrypt_ioctl_set_policy(struct file *filp, const void __user *arg) return -EFAULT; policy.version = version; - if (!inode_owner_or_capable(inode)) + if (!inode_owner_or_capable(&init_user_ns, inode)) return -EACCES; ret = mnt_want_write_file(filp); @@ -810,11 +810,12 @@ static void dax_entry_mkclean(struct address_space *mapping, pgoff_t index, address = pgoff_address(index, vma); /* - * Note because we provide range to follow_pte it will call + * follow_invalidate_pte() will use the range to call * mmu_notifier_invalidate_range_start() on our behalf before * taking any lock. */ - if (follow_pte(vma->vm_mm, address, &range, &ptep, &pmdp, &ptl)) + if (follow_invalidate_pte(vma->vm_mm, address, &range, &ptep, + &pmdp, &ptl)) continue; /* diff --git a/fs/dcache.c b/fs/dcache.c index 97e81a844a96..7d24ff7eb206 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -456,23 +456,6 @@ static void d_lru_shrink_move(struct list_lru_one *lru, struct dentry *dentry, list_lru_isolate_move(lru, &dentry->d_lru, list); } -/** - * d_drop - drop a dentry - * @dentry: dentry to drop - * - * d_drop() unhashes the entry from the parent dentry hashes, so that it won't - * be found through a VFS lookup any more. Note that this is different from - * deleting the dentry - d_delete will try to mark the dentry negative if - * possible, giving a successful _negative_ lookup, while d_drop will - * just make the cache lookup fail. - * - * d_drop() is used mainly for stuff that wants to invalidate a dentry for some - * reason (NFS timeouts or autofs deletes). - * - * __d_drop requires dentry->d_lock - * ___d_drop doesn't mark dentry as "unhashed" - * (dentry->d_hash.pprev will be LIST_POISON2, not NULL). - */ static void ___d_drop(struct dentry *dentry) { struct hlist_bl_head *b; @@ -501,6 +484,24 @@ void __d_drop(struct dentry *dentry) } EXPORT_SYMBOL(__d_drop); +/** + * d_drop - drop a dentry + * @dentry: dentry to drop + * + * d_drop() unhashes the entry from the parent dentry hashes, so that it won't + * be found through a VFS lookup any more. Note that this is different from + * deleting the dentry - d_delete will try to mark the dentry negative if + * possible, giving a successful _negative_ lookup, while d_drop will + * just make the cache lookup fail. + * + * d_drop() is used mainly for stuff that wants to invalidate a dentry for some + * reason (NFS timeouts or autofs deletes). + * + * __d_drop requires dentry->d_lock + * + * ___d_drop doesn't mark dentry as "unhashed" + * (dentry->d_hash.pprev will be LIST_POISON2, not NULL). + */ void d_drop(struct dentry *dentry) { spin_lock(&dentry->d_lock); @@ -996,20 +997,6 @@ struct dentry *d_find_any_alias(struct inode *inode) } EXPORT_SYMBOL(d_find_any_alias); -/** - * d_find_alias - grab a hashed alias of inode - * @inode: inode in question - * - * If inode has a hashed alias, or is a directory and has any alias, - * acquire the reference to alias and return it. Otherwise return NULL. - * Notice that if inode is a directory there can be only one alias and - * it can be unhashed only if it has no children, or if it is the root - * of a filesystem, or if the directory was renamed and d_revalidate - * was the first vfs operation to notice. - * - * If the inode has an IS_ROOT, DCACHE_DISCONNECTED alias, then prefer - * any other hashed alias over that one. - */ static struct dentry *__d_find_alias(struct inode *inode) { struct dentry *alias; @@ -1029,6 +1016,20 @@ static struct dentry *__d_find_alias(struct inode *inode) return NULL; } +/** + * d_find_alias - grab a hashed alias of inode + * @inode: inode in question + * + * If inode has a hashed alias, or is a directory and has any alias, + * acquire the reference to alias and return it. Otherwise return NULL. + * Notice that if inode is a directory there can be only one alias and + * it can be unhashed only if it has no children, or if it is the root + * of a filesystem, or if the directory was renamed and d_revalidate + * was the first vfs operation to notice. + * + * If the inode has an IS_ROOT, DCACHE_DISCONNECTED alias, then prefer + * any other hashed alias over that one. + */ struct dentry *d_find_alias(struct inode *inode) { struct dentry *de = NULL; @@ -1043,6 +1044,31 @@ struct dentry *d_find_alias(struct inode *inode) EXPORT_SYMBOL(d_find_alias); /* + * Caller MUST be holding rcu_read_lock() and be guaranteed + * that inode won't get freed until rcu_read_unlock(). + */ +struct dentry *d_find_alias_rcu(struct inode *inode) +{ + struct hlist_head *l = &inode->i_dentry; + struct dentry *de = NULL; + + spin_lock(&inode->i_lock); + // ->i_dentry and ->i_rcu are colocated, but the latter won't be + // used without having I_FREEING set, which means no aliases left + if (likely(!(inode->i_state & I_FREEING) && !hlist_empty(l))) { + if (S_ISDIR(inode->i_mode)) { + de = hlist_entry(l->first, struct dentry, d_u.d_alias); + } else { + hlist_for_each_entry(de, l, d_u.d_alias) + if (!d_unhashed(de)) + break; + } + } + spin_unlock(&inode->i_lock); + return de; +} + +/* * Try to kill dentries associated with this inode. * WARNING: you must own a reference to inode. */ @@ -2150,8 +2176,8 @@ EXPORT_SYMBOL(d_obtain_root); * same inode, only the actual correct case is stored in the dcache for * case-insensitive filesystems. * - * For a case-insensitive lookup match and if the the case-exact dentry - * already exists in in the dcache, use it and return it. + * For a case-insensitive lookup match and if the case-exact dentry + * already exists in the dcache, use it and return it. * * If no entry exists with the exact case name, allocate new dentry with * the exact case, and return the spliced entry. diff --git a/fs/dcookies.c b/fs/dcookies.c deleted file mode 100644 index 6eeb61100a09..000000000000 --- a/fs/dcookies.c +++ /dev/null @@ -1,356 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0-only -/* - * dcookies.c - * - * Copyright 2002 John Levon <levon@movementarian.org> - * - * Persistent cookie-path mappings. These are used by - * profilers to convert a per-task EIP value into something - * non-transitory that can be processed at a later date. - * This is done by locking the dentry/vfsmnt pair in the - * kernel until released by the tasks needing the persistent - * objects. The tag is simply an unsigned long that refers - * to the pair and can be looked up from userspace. - */ - -#include <linux/syscalls.h> -#include <linux/export.h> -#include <linux/slab.h> -#include <linux/list.h> -#include <linux/mount.h> -#include <linux/capability.h> -#include <linux/dcache.h> -#include <linux/mm.h> -#include <linux/err.h> -#include <linux/errno.h> -#include <linux/dcookies.h> -#include <linux/mutex.h> -#include <linux/path.h> -#include <linux/compat.h> -#include <linux/uaccess.h> - -/* The dcookies are allocated from a kmem_cache and - * hashed onto a small number of lists. None of the - * code here is particularly performance critical - */ -struct dcookie_struct { - struct path path; - struct list_head hash_list; -}; - -static LIST_HEAD(dcookie_users); -static DEFINE_MUTEX(dcookie_mutex); -static struct kmem_cache *dcookie_cache __read_mostly; -static struct list_head *dcookie_hashtable __read_mostly; -static size_t hash_size __read_mostly; - -static inline int is_live(void) -{ - return !(list_empty(&dcookie_users)); -} - - -/* The dentry is locked, its address will do for the cookie */ -static inline unsigned long dcookie_value(struct dcookie_struct * dcs) -{ - return (unsigned long)dcs->path.dentry; -} - - -static size_t dcookie_hash(unsigned long dcookie) -{ - return (dcookie >> L1_CACHE_SHIFT) & (hash_size - 1); -} - - -static struct dcookie_struct * find_dcookie(unsigned long dcookie) -{ - struct dcookie_struct *found = NULL; - struct dcookie_struct * dcs; - struct list_head * pos; - struct list_head * list; - - list = dcookie_hashtable + dcookie_hash(dcookie); - - list_for_each(pos, list) { - dcs = list_entry(pos, struct dcookie_struct, hash_list); - if (dcookie_value(dcs) == dcookie) { - found = dcs; - break; - } - } - - return found; -} - - -static void hash_dcookie(struct dcookie_struct * dcs) -{ - struct list_head * list = dcookie_hashtable + dcookie_hash(dcookie_value(dcs)); - list_add(&dcs->hash_list, list); -} - - -static struct dcookie_struct *alloc_dcookie(const struct path *path) -{ - struct dcookie_struct *dcs = kmem_cache_alloc(dcookie_cache, - GFP_KERNEL); - struct dentry *d; - if (!dcs) - return NULL; - - d = path->dentry; - spin_lock(&d->d_lock); - d->d_flags |= DCACHE_COOKIE; - spin_unlock(&d->d_lock); - - dcs->path = *path; - path_get(path); - hash_dcookie(dcs); - return dcs; -} - - -/* This is the main kernel-side routine that retrieves the cookie - * value for a dentry/vfsmnt pair. - */ -int get_dcookie(const struct path *path, unsigned long *cookie) -{ - int err = 0; - struct dcookie_struct * dcs; - - mutex_lock(&dcookie_mutex); - - if (!is_live()) { - err = -EINVAL; - goto out; - } - - if (path->dentry->d_flags & DCACHE_COOKIE) { - dcs = find_dcookie((unsigned long)path->dentry); - } else { - dcs = alloc_dcookie(path); - if (!dcs) { - err = -ENOMEM; - goto out; - } - } - - *cookie = dcookie_value(dcs); - -out: - mutex_unlock(&dcookie_mutex); - return err; -} - - -/* And here is where the userspace process can look up the cookie value - * to retrieve the path. - */ -static int do_lookup_dcookie(u64 cookie64, char __user *buf, size_t len) -{ - unsigned long cookie = (unsigned long)cookie64; - int err = -EINVAL; - char * kbuf; - char * path; - size_t pathlen; - struct dcookie_struct * dcs; - - /* we could leak path information to users - * without dir read permission without this - */ - if (!capable(CAP_SYS_ADMIN)) - return -EPERM; - - mutex_lock(&dcookie_mutex); - - if (!is_live()) { - err = -EINVAL; - goto out; - } - - if (!(dcs = find_dcookie(cookie))) - goto out; - - err = -ENOMEM; - kbuf = kmalloc(PAGE_SIZE, GFP_KERNEL); - if (!kbuf) - goto out; - - /* FIXME: (deleted) ? */ - path = d_path(&dcs->path, kbuf, PAGE_SIZE); - - mutex_unlock(&dcookie_mutex); - - if (IS_ERR(path)) { - err = PTR_ERR(path); - goto out_free; - } - - err = -ERANGE; - - pathlen = kbuf + PAGE_SIZE - path; - if (pathlen <= len) { - err = pathlen; - if (copy_to_user(buf, path, pathlen)) - err = -EFAULT; - } - -out_free: - kfree(kbuf); - return err; -out: - mutex_unlock(&dcookie_mutex); - return err; -} - -SYSCALL_DEFINE3(lookup_dcookie, u64, cookie64, char __user *, buf, size_t, len) -{ - return do_lookup_dcookie(cookie64, buf, len); -} - -#ifdef CONFIG_COMPAT -COMPAT_SYSCALL_DEFINE4(lookup_dcookie, u32, w0, u32, w1, char __user *, buf, compat_size_t, len) -{ -#ifdef __BIG_ENDIAN - return do_lookup_dcookie(((u64)w0 << 32) | w1, buf, len); -#else - return do_lookup_dcookie(((u64)w1 << 32) | w0, buf, len); -#endif -} -#endif - -static int dcookie_init(void) -{ - struct list_head * d; - unsigned int i, hash_bits; - int err = -ENOMEM; - - dcookie_cache = kmem_cache_create("dcookie_cache", - sizeof(struct dcookie_struct), - 0, 0, NULL); - - if (!dcookie_cache) - goto out; - - dcookie_hashtable = kmalloc(PAGE_SIZE, GFP_KERNEL); - if (!dcookie_hashtable) - goto out_kmem; - - err = 0; - - /* - * Find the power-of-two list-heads that can fit into the allocation.. - * We don't guarantee that "sizeof(struct list_head)" is necessarily - * a power-of-two. - */ - hash_size = PAGE_SIZE / sizeof(struct list_head); - hash_bits = 0; - do { - hash_bits++; - } while ((hash_size >> hash_bits) != 0); - hash_bits--; - - /* - * Re-calculate the actual number of entries and the mask - * from the number of bits we can fit. - */ - hash_size = 1UL << hash_bits; - - /* And initialize the newly allocated array */ - d = dcookie_hashtable; - i = hash_size; - do { - INIT_LIST_HEAD(d); - d++; - i--; - } while (i); - -out: - return err; -out_kmem: - kmem_cache_destroy(dcookie_cache); - goto out; -} - - -static void free_dcookie(struct dcookie_struct * dcs) -{ - struct dentry *d = dcs->path.dentry; - - spin_lock(&d->d_lock); - d->d_flags &= ~DCACHE_COOKIE; - spin_unlock(&d->d_lock); - - path_put(&dcs->path); - kmem_cache_free(dcookie_cache, dcs); -} - - -static void dcookie_exit(void) -{ - struct list_head * list; - struct list_head * pos; - struct list_head * pos2; - struct dcookie_struct * dcs; - size_t i; - - for (i = 0; i < hash_size; ++i) { - list = dcookie_hashtable + i; - list_for_each_safe(pos, pos2, list) { - dcs = list_entry(pos, struct dcookie_struct, hash_list); - list_del(&dcs->hash_list); - free_dcookie(dcs); - } - } - - kfree(dcookie_hashtable); - kmem_cache_destroy(dcookie_cache); -} - - -struct dcookie_user { - struct list_head next; -}; - -struct dcookie_user * dcookie_register(void) -{ - struct dcookie_user * user; - - mutex_lock(&dcookie_mutex); - - user = kmalloc(sizeof(struct dcookie_user), GFP_KERNEL); - if (!user) - goto out; - - if (!is_live() && dcookie_init()) - goto out_free; - - list_add(&user->next, &dcookie_users); - -out: - mutex_unlock(&dcookie_mutex); - return user; -out_free: - kfree(user); - user = NULL; - goto out; -} - - -void dcookie_unregister(struct dcookie_user * user) -{ - mutex_lock(&dcookie_mutex); - - list_del(&user->next); - kfree(user); - - if (!is_live()) - dcookie_exit(); - - mutex_unlock(&dcookie_mutex); -} - -EXPORT_SYMBOL_GPL(dcookie_register); -EXPORT_SYMBOL_GPL(dcookie_unregister); -EXPORT_SYMBOL_GPL(get_dcookie); diff --git a/fs/debugfs/inode.c b/fs/debugfs/inode.c index 2fcf66473436..22e86ae4dd5a 100644 --- a/fs/debugfs/inode.c +++ b/fs/debugfs/inode.c @@ -42,13 +42,14 @@ static unsigned int debugfs_allow = DEFAULT_DEBUGFS_ALLOW_BITS; * so that we can use the file mode as part of a heuristic to determine whether * to lock down individual files. */ -static int debugfs_setattr(struct dentry *dentry, struct iattr *ia) +static int debugfs_setattr(struct user_namespace *mnt_userns, + struct dentry *dentry, struct iattr *ia) { int ret = security_locked_down(LOCKDOWN_DEBUGFS); if (ret && (ia->ia_valid & (ATTR_MODE | ATTR_UID | ATTR_GID))) return ret; - return simple_setattr(dentry, ia); + return simple_setattr(&init_user_ns, dentry, ia); } static const struct inode_operations debugfs_file_inode_operations = { @@ -297,7 +298,7 @@ struct dentry *debugfs_lookup(const char *name, struct dentry *parent) { struct dentry *dentry; - if (IS_ERR(parent)) + if (!debugfs_initialized() || IS_ERR_OR_NULL(name) || IS_ERR(parent)) return NULL; if (!parent) @@ -318,6 +319,9 @@ static struct dentry *start_creating(const char *name, struct dentry *parent) if (!(debugfs_allow & DEBUGFS_ALLOW_API)) return ERR_PTR(-EPERM); + if (!debugfs_initialized()) + return ERR_PTR(-ENOENT); + pr_debug("creating file '%s'\n", name); if (IS_ERR(parent)) @@ -775,8 +779,8 @@ struct dentry *debugfs_rename(struct dentry *old_dir, struct dentry *old_dentry, take_dentry_name_snapshot(&old_name, old_dentry); - error = simple_rename(d_inode(old_dir), old_dentry, d_inode(new_dir), - dentry, 0); + error = simple_rename(&init_user_ns, d_inode(old_dir), old_dentry, + d_inode(new_dir), dentry, 0); if (error) { release_dentry_name_snapshot(&old_name); goto exit; diff --git a/fs/direct-io.c b/fs/direct-io.c index d53fa92a1ab6..b61491bf3166 100644 --- a/fs/direct-io.c +++ b/fs/direct-io.c @@ -426,6 +426,8 @@ static inline void dio_bio_submit(struct dio *dio, struct dio_submit *sdio) unsigned long flags; bio->bi_private = dio; + /* don't account direct I/O as memory stall */ + bio_clear_flag(bio, BIO_WORKINGSET); spin_lock_irqsave(&dio->bio_lock, flags); dio->refcount++; @@ -434,7 +436,7 @@ static inline void dio_bio_submit(struct dio *dio, struct dio_submit *sdio) if (dio->is_async && dio->op == REQ_OP_READ && dio->should_dirty) bio_set_pages_dirty(bio); - dio->bio_disk = bio->bi_disk; + dio->bio_disk = bio->bi_bdev->bd_disk; if (sdio->submit_io) { sdio->submit_io(bio, dio->inode, sdio->logical_offset_in_bio); @@ -460,7 +462,7 @@ static inline void dio_cleanup(struct dio *dio, struct dio_submit *sdio) * Wait for the next BIO to complete. Remove it and return it. NULL is * returned once all BIOs have been completed. This must only be called once * all bios have been issued so that dio->refcount can only decrease. This - * requires that that the caller hold a reference on the dio. + * requires that the caller hold a reference on the dio. */ static struct bio *dio_await_one(struct dio *dio) { @@ -693,7 +695,7 @@ static inline int dio_new_bio(struct dio *dio, struct dio_submit *sdio, if (ret) goto out; sector = start_sector << (sdio->blkbits - 9); - nr_pages = min(sdio->pages_in_io, BIO_MAX_PAGES); + nr_pages = bio_max_segs(sdio->pages_in_io); BUG_ON(nr_pages <= 0); dio_bio_alloc(dio, sdio, map_bh->b_bdev, sector, nr_pages); sdio->boundary = 0; @@ -1277,7 +1279,7 @@ do_blockdev_direct_IO(struct kiocb *iocb, struct inode *inode, if (retval == -ENOTBLK) { /* * The remaining part of the request will be - * be handled by buffered I/O when we return + * handled by buffered I/O when we return */ retval = 0; } diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c index 0681540c48d9..943e523f4c9d 100644 --- a/fs/ecryptfs/crypto.c +++ b/fs/ecryptfs/crypto.c @@ -1110,8 +1110,8 @@ ecryptfs_write_metadata_to_xattr(struct dentry *ecryptfs_dentry, } inode_lock(lower_inode); - rc = __vfs_setxattr(lower_dentry, lower_inode, ECRYPTFS_XATTR_NAME, - page_virt, size, 0); + rc = __vfs_setxattr(&init_user_ns, lower_dentry, lower_inode, + ECRYPTFS_XATTR_NAME, page_virt, size, 0); if (!rc && ecryptfs_inode) fsstack_copy_attr_all(ecryptfs_inode, lower_inode); inode_unlock(lower_inode); diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index 58d0f7187997..18e9285fbb4c 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -141,7 +141,8 @@ static int ecryptfs_do_unlink(struct inode *dir, struct dentry *dentry, else if (d_unhashed(lower_dentry)) rc = -EINVAL; else - rc = vfs_unlink(lower_dir_inode, lower_dentry, NULL); + rc = vfs_unlink(&init_user_ns, lower_dir_inode, lower_dentry, + NULL); if (rc) { printk(KERN_ERR "Error in vfs_unlink; rc = [%d]\n", rc); goto out_unlock; @@ -180,7 +181,8 @@ ecryptfs_do_create(struct inode *directory_inode, lower_dentry = ecryptfs_dentry_to_lower(ecryptfs_dentry); lower_dir_dentry = lock_parent(lower_dentry); - rc = vfs_create(d_inode(lower_dir_dentry), lower_dentry, mode, true); + rc = vfs_create(&init_user_ns, d_inode(lower_dir_dentry), lower_dentry, + mode, true); if (rc) { printk(KERN_ERR "%s: Failure to create dentry in lower fs; " "rc = [%d]\n", __func__, rc); @@ -190,7 +192,8 @@ ecryptfs_do_create(struct inode *directory_inode, inode = __ecryptfs_get_inode(d_inode(lower_dentry), directory_inode->i_sb); if (IS_ERR(inode)) { - vfs_unlink(d_inode(lower_dir_dentry), lower_dentry, NULL); + vfs_unlink(&init_user_ns, d_inode(lower_dir_dentry), + lower_dentry, NULL); goto out_lock; } fsstack_copy_attr_times(directory_inode, d_inode(lower_dir_dentry)); @@ -254,7 +257,8 @@ out: * Returns zero on success; non-zero on error condition */ static int -ecryptfs_create(struct inode *directory_inode, struct dentry *ecryptfs_dentry, +ecryptfs_create(struct user_namespace *mnt_userns, + struct inode *directory_inode, struct dentry *ecryptfs_dentry, umode_t mode, bool excl) { struct inode *ecryptfs_inode; @@ -436,8 +440,8 @@ static int ecryptfs_link(struct dentry *old_dentry, struct inode *dir, dget(lower_old_dentry); dget(lower_new_dentry); lower_dir_dentry = lock_parent(lower_new_dentry); - rc = vfs_link(lower_old_dentry, d_inode(lower_dir_dentry), - lower_new_dentry, NULL); + rc = vfs_link(lower_old_dentry, &init_user_ns, + d_inode(lower_dir_dentry), lower_new_dentry, NULL); if (rc || d_really_is_negative(lower_new_dentry)) goto out_lock; rc = ecryptfs_interpose(lower_new_dentry, new_dentry, dir->i_sb); @@ -460,7 +464,8 @@ static int ecryptfs_unlink(struct inode *dir, struct dentry *dentry) return ecryptfs_do_unlink(dir, dentry, d_inode(dentry)); } -static int ecryptfs_symlink(struct inode *dir, struct dentry *dentry, +static int ecryptfs_symlink(struct user_namespace *mnt_userns, + struct inode *dir, struct dentry *dentry, const char *symname) { int rc; @@ -481,7 +486,7 @@ static int ecryptfs_symlink(struct inode *dir, struct dentry *dentry, strlen(symname)); if (rc) goto out_lock; - rc = vfs_symlink(d_inode(lower_dir_dentry), lower_dentry, + rc = vfs_symlink(&init_user_ns, d_inode(lower_dir_dentry), lower_dentry, encoded_symname); kfree(encoded_symname); if (rc || d_really_is_negative(lower_dentry)) @@ -499,7 +504,8 @@ out_lock: return rc; } -static int ecryptfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) +static int ecryptfs_mkdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { int rc; struct dentry *lower_dentry; @@ -507,7 +513,8 @@ static int ecryptfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode lower_dentry = ecryptfs_dentry_to_lower(dentry); lower_dir_dentry = lock_parent(lower_dentry); - rc = vfs_mkdir(d_inode(lower_dir_dentry), lower_dentry, mode); + rc = vfs_mkdir(&init_user_ns, d_inode(lower_dir_dentry), lower_dentry, + mode); if (rc || d_really_is_negative(lower_dentry)) goto out; rc = ecryptfs_interpose(lower_dentry, dentry, dir->i_sb); @@ -541,7 +548,7 @@ static int ecryptfs_rmdir(struct inode *dir, struct dentry *dentry) else if (d_unhashed(lower_dentry)) rc = -EINVAL; else - rc = vfs_rmdir(lower_dir_inode, lower_dentry); + rc = vfs_rmdir(&init_user_ns, lower_dir_inode, lower_dentry); if (!rc) { clear_nlink(d_inode(dentry)); fsstack_copy_attr_times(dir, lower_dir_inode); @@ -555,7 +562,8 @@ static int ecryptfs_rmdir(struct inode *dir, struct dentry *dentry) } static int -ecryptfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t dev) +ecryptfs_mknod(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, dev_t dev) { int rc; struct dentry *lower_dentry; @@ -563,7 +571,8 @@ ecryptfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t dev lower_dentry = ecryptfs_dentry_to_lower(dentry); lower_dir_dentry = lock_parent(lower_dentry); - rc = vfs_mknod(d_inode(lower_dir_dentry), lower_dentry, mode, dev); + rc = vfs_mknod(&init_user_ns, d_inode(lower_dir_dentry), lower_dentry, + mode, dev); if (rc || d_really_is_negative(lower_dentry)) goto out; rc = ecryptfs_interpose(lower_dentry, dentry, dir->i_sb); @@ -579,9 +588,9 @@ out: } static int -ecryptfs_rename(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry, - unsigned int flags) +ecryptfs_rename(struct user_namespace *mnt_userns, struct inode *old_dir, + struct dentry *old_dentry, struct inode *new_dir, + struct dentry *new_dentry, unsigned int flags) { int rc; struct dentry *lower_old_dentry; @@ -590,6 +599,7 @@ ecryptfs_rename(struct inode *old_dir, struct dentry *old_dentry, struct dentry *lower_new_dir_dentry; struct dentry *trap; struct inode *target_inode; + struct renamedata rd = {}; if (flags) return -EINVAL; @@ -619,9 +629,14 @@ ecryptfs_rename(struct inode *old_dir, struct dentry *old_dentry, rc = -ENOTEMPTY; goto out_lock; } - rc = vfs_rename(d_inode(lower_old_dir_dentry), lower_old_dentry, - d_inode(lower_new_dir_dentry), lower_new_dentry, - NULL, 0); + + rd.old_mnt_userns = &init_user_ns; + rd.old_dir = d_inode(lower_old_dir_dentry); + rd.old_dentry = lower_old_dentry; + rd.new_mnt_userns = &init_user_ns; + rd.new_dir = d_inode(lower_new_dir_dentry); + rd.new_dentry = lower_new_dentry; + rc = vfs_rename(&rd); if (rc) goto out_lock; if (target_inode) @@ -855,16 +870,19 @@ int ecryptfs_truncate(struct dentry *dentry, loff_t new_length) struct dentry *lower_dentry = ecryptfs_dentry_to_lower(dentry); inode_lock(d_inode(lower_dentry)); - rc = notify_change(lower_dentry, &lower_ia, NULL); + rc = notify_change(&init_user_ns, lower_dentry, + &lower_ia, NULL); inode_unlock(d_inode(lower_dentry)); } return rc; } static int -ecryptfs_permission(struct inode *inode, int mask) +ecryptfs_permission(struct user_namespace *mnt_userns, struct inode *inode, + int mask) { - return inode_permission(ecryptfs_inode_to_lower(inode), mask); + return inode_permission(&init_user_ns, + ecryptfs_inode_to_lower(inode), mask); } /** @@ -879,7 +897,8 @@ ecryptfs_permission(struct inode *inode, int mask) * All other metadata changes will be passed right to the lower filesystem, * and we will just update our inode to look like the lower. */ -static int ecryptfs_setattr(struct dentry *dentry, struct iattr *ia) +static int ecryptfs_setattr(struct user_namespace *mnt_userns, + struct dentry *dentry, struct iattr *ia) { int rc = 0; struct dentry *lower_dentry; @@ -933,7 +952,7 @@ static int ecryptfs_setattr(struct dentry *dentry, struct iattr *ia) } mutex_unlock(&crypt_stat->cs_mutex); - rc = setattr_prepare(dentry, ia); + rc = setattr_prepare(&init_user_ns, dentry, ia); if (rc) goto out; if (ia->ia_valid & ATTR_SIZE) { @@ -959,14 +978,15 @@ static int ecryptfs_setattr(struct dentry *dentry, struct iattr *ia) lower_ia.ia_valid &= ~ATTR_MODE; inode_lock(d_inode(lower_dentry)); - rc = notify_change(lower_dentry, &lower_ia, NULL); + rc = notify_change(&init_user_ns, lower_dentry, &lower_ia, NULL); inode_unlock(d_inode(lower_dentry)); out: fsstack_copy_attr_all(inode, lower_inode); return rc; } -static int ecryptfs_getattr_link(const struct path *path, struct kstat *stat, +static int ecryptfs_getattr_link(struct user_namespace *mnt_userns, + const struct path *path, struct kstat *stat, u32 request_mask, unsigned int flags) { struct dentry *dentry = path->dentry; @@ -975,7 +995,7 @@ static int ecryptfs_getattr_link(const struct path *path, struct kstat *stat, mount_crypt_stat = &ecryptfs_superblock_to_private( dentry->d_sb)->mount_crypt_stat; - generic_fillattr(d_inode(dentry), stat); + generic_fillattr(&init_user_ns, d_inode(dentry), stat); if (mount_crypt_stat->flags & ECRYPTFS_GLOBAL_ENCRYPT_FILENAMES) { char *target; size_t targetsiz; @@ -991,7 +1011,8 @@ static int ecryptfs_getattr_link(const struct path *path, struct kstat *stat, return rc; } -static int ecryptfs_getattr(const struct path *path, struct kstat *stat, +static int ecryptfs_getattr(struct user_namespace *mnt_userns, + const struct path *path, struct kstat *stat, u32 request_mask, unsigned int flags) { struct dentry *dentry = path->dentry; @@ -1003,7 +1024,7 @@ static int ecryptfs_getattr(const struct path *path, struct kstat *stat, if (!rc) { fsstack_copy_attr_all(d_inode(dentry), ecryptfs_inode_to_lower(d_inode(dentry))); - generic_fillattr(d_inode(dentry), stat); + generic_fillattr(&init_user_ns, d_inode(dentry), stat); stat->blocks = lower_stat.blocks; } return rc; @@ -1025,7 +1046,7 @@ ecryptfs_setxattr(struct dentry *dentry, struct inode *inode, goto out; } inode_lock(lower_inode); - rc = __vfs_setxattr_locked(lower_dentry, name, value, size, flags, NULL); + rc = __vfs_setxattr_locked(&init_user_ns, lower_dentry, name, value, size, flags, NULL); inode_unlock(lower_inode); if (!rc && inode) fsstack_copy_attr_all(inode, lower_inode); @@ -1091,7 +1112,7 @@ static int ecryptfs_removexattr(struct dentry *dentry, struct inode *inode, goto out; } inode_lock(lower_inode); - rc = __vfs_removexattr(lower_dentry, name); + rc = __vfs_removexattr(&init_user_ns, lower_dentry, name); inode_unlock(lower_inode); out: return rc; @@ -1135,6 +1156,7 @@ static int ecryptfs_xattr_get(const struct xattr_handler *handler, } static int ecryptfs_xattr_set(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *dentry, struct inode *inode, const char *name, const void *value, size_t size, int flags) diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c index e63259fdef28..cdf40a54a35d 100644 --- a/fs/ecryptfs/main.c +++ b/fs/ecryptfs/main.c @@ -531,6 +531,12 @@ static struct dentry *ecryptfs_mount(struct file_system_type *fs_type, int flags goto out_free; } + if (mnt_user_ns(path.mnt) != &init_user_ns) { + rc = -EINVAL; + printk(KERN_ERR "Mounting on idmapped mounts currently disallowed\n"); + goto out_free; + } + if (check_ruid && !uid_eq(d_inode(path.dentry)->i_uid, current_uid())) { rc = -EPERM; printk(KERN_ERR "Mount of device (uid: %d) not owned by " diff --git a/fs/ecryptfs/mmap.c b/fs/ecryptfs/mmap.c index 019572c6b39a..2f333a40ff4d 100644 --- a/fs/ecryptfs/mmap.c +++ b/fs/ecryptfs/mmap.c @@ -426,8 +426,8 @@ static int ecryptfs_write_inode_size_to_xattr(struct inode *ecryptfs_inode) if (size < 0) size = 8; put_unaligned_be64(i_size_read(ecryptfs_inode), xattr_virt); - rc = __vfs_setxattr(lower_dentry, lower_inode, ECRYPTFS_XATTR_NAME, - xattr_virt, size, 0); + rc = __vfs_setxattr(&init_user_ns, lower_dentry, lower_inode, + ECRYPTFS_XATTR_NAME, xattr_virt, size, 0); inode_unlock(lower_inode); if (rc) printk(KERN_ERR "Error whilst attempting to write inode size " diff --git a/fs/efivarfs/file.c b/fs/efivarfs/file.c index feaa5e182b7b..e6bc0302643b 100644 --- a/fs/efivarfs/file.c +++ b/fs/efivarfs/file.c @@ -137,7 +137,7 @@ efivarfs_ioc_setxflags(struct file *file, void __user *arg) unsigned int oldflags = efivarfs_getflags(inode); int error; - if (!inode_owner_or_capable(inode)) + if (!inode_owner_or_capable(&init_user_ns, inode)) return -EACCES; if (copy_from_user(&flags, arg, sizeof(flags))) diff --git a/fs/efivarfs/inode.c b/fs/efivarfs/inode.c index 0297ad95eb5c..14e2947975fd 100644 --- a/fs/efivarfs/inode.c +++ b/fs/efivarfs/inode.c @@ -66,8 +66,8 @@ bool efivarfs_valid_name(const char *str, int len) return uuid_is_valid(s); } -static int efivarfs_create(struct inode *dir, struct dentry *dentry, - umode_t mode, bool excl) +static int efivarfs_create(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, bool excl) { struct inode *inode = NULL; struct efivar_entry *var; diff --git a/fs/erofs/data.c b/fs/erofs/data.c index ea4f693bee22..1249e74b3bf0 100644 --- a/fs/erofs/data.c +++ b/fs/erofs/data.c @@ -129,6 +129,7 @@ static inline struct bio *erofs_read_raw_page(struct bio *bio, struct page *page, erofs_off_t *last_block, unsigned int nblocks, + unsigned int *eblks, bool ra) { struct inode *const inode = mapping->host; @@ -145,8 +146,7 @@ static inline struct bio *erofs_read_raw_page(struct bio *bio, /* note that for readpage case, bio also equals to NULL */ if (bio && - /* not continuous */ - *last_block + 1 != current_block) { + (*last_block + 1 != current_block || !*eblks)) { submit_bio_retry: submit_bio(bio); bio = NULL; @@ -215,10 +215,9 @@ submit_bio_retry: /* max # of continuous pages */ if (nblocks > DIV_ROUND_UP(map.m_plen, PAGE_SIZE)) nblocks = DIV_ROUND_UP(map.m_plen, PAGE_SIZE); - if (nblocks > BIO_MAX_PAGES) - nblocks = BIO_MAX_PAGES; - bio = bio_alloc(GFP_NOIO, nblocks); + *eblks = bio_max_segs(nblocks); + bio = bio_alloc(GFP_NOIO, *eblks); bio->bi_end_io = erofs_readendio; bio_set_dev(bio, sb->s_bdev); @@ -231,16 +230,8 @@ submit_bio_retry: /* out of the extent or bio is full */ if (err < PAGE_SIZE) goto submit_bio_retry; - + --*eblks; *last_block = current_block; - - /* shift in advance in case of it followed by too many gaps */ - if (bio->bi_iter.bi_size >= bio->bi_max_vecs * PAGE_SIZE) { - /* err should reassign to 0 after submitting */ - err = 0; - goto submit_bio_out; - } - return bio; err_out: @@ -254,7 +245,6 @@ has_updated: /* if updated manually, continuous pages has a gap */ if (bio) -submit_bio_out: submit_bio(bio); return err ? ERR_PTR(err) : NULL; } @@ -266,23 +256,26 @@ submit_bio_out: static int erofs_raw_access_readpage(struct file *file, struct page *page) { erofs_off_t last_block; + unsigned int eblks; struct bio *bio; trace_erofs_readpage(page, true); bio = erofs_read_raw_page(NULL, page->mapping, - page, &last_block, 1, false); + page, &last_block, 1, &eblks, false); if (IS_ERR(bio)) return PTR_ERR(bio); - DBG_BUGON(bio); /* since we have only one bio -- must be NULL */ + if (bio) + submit_bio(bio); return 0; } static void erofs_raw_access_readahead(struct readahead_control *rac) { erofs_off_t last_block; + unsigned int eblks; struct bio *bio = NULL; struct page *page; @@ -293,7 +286,7 @@ static void erofs_raw_access_readahead(struct readahead_control *rac) prefetchw(&page->flags); bio = erofs_read_raw_page(bio, rac->mapping, page, &last_block, - readahead_count(rac), true); + readahead_count(rac), &eblks, true); /* all the page errors are ignored when readahead */ if (IS_ERR(bio)) { @@ -307,7 +300,6 @@ static void erofs_raw_access_readahead(struct readahead_control *rac) put_page(page); } - /* the rare case (end in gaps) */ if (bio) submit_bio(bio); } diff --git a/fs/erofs/inode.c b/fs/erofs/inode.c index 3e21c0e8adae..119fdce1b520 100644 --- a/fs/erofs/inode.c +++ b/fs/erofs/inode.c @@ -331,8 +331,9 @@ struct inode *erofs_iget(struct super_block *sb, return inode; } -int erofs_getattr(const struct path *path, struct kstat *stat, - u32 request_mask, unsigned int query_flags) +int erofs_getattr(struct user_namespace *mnt_userns, const struct path *path, + struct kstat *stat, u32 request_mask, + unsigned int query_flags) { struct inode *const inode = d_inode(path->dentry); @@ -343,7 +344,7 @@ int erofs_getattr(const struct path *path, struct kstat *stat, stat->attributes_mask |= (STATX_ATTR_COMPRESSED | STATX_ATTR_IMMUTABLE); - generic_fillattr(inode, stat); + generic_fillattr(&init_user_ns, inode, stat); return 0; } diff --git a/fs/erofs/internal.h b/fs/erofs/internal.h index 67a7ec945686..351dae524a0c 100644 --- a/fs/erofs/internal.h +++ b/fs/erofs/internal.h @@ -373,8 +373,9 @@ extern const struct inode_operations erofs_symlink_iops; extern const struct inode_operations erofs_fast_symlink_iops; struct inode *erofs_iget(struct super_block *sb, erofs_nid_t nid, bool dir); -int erofs_getattr(const struct path *path, struct kstat *stat, - u32 request_mask, unsigned int query_flags); +int erofs_getattr(struct user_namespace *mnt_userns, const struct path *path, + struct kstat *stat, u32 request_mask, + unsigned int query_flags); /* namei.c */ extern const struct inode_operations erofs_dir_iops; diff --git a/fs/erofs/namei.c b/fs/erofs/namei.c index 5f8cc7346c69..3a81e1f7fc06 100644 --- a/fs/erofs/namei.c +++ b/fs/erofs/namei.c @@ -234,8 +234,8 @@ static struct dentry *erofs_lookup(struct inode *dir, } else if (err) { inode = ERR_PTR(err); } else { - erofs_dbg("%s, %s (nid %llu) found, d_type %u", __func__, - dentry->d_name.name, nid, d_type); + erofs_dbg("%s, %pd (nid %llu) found, d_type %u", __func__, + dentry, nid, d_type); inode = erofs_iget(dir->i_sb, nid, d_type == FT_DIR); } return d_splice_alias(inode, dentry); diff --git a/fs/erofs/super.c b/fs/erofs/super.c index be10b16ea66e..d5a6b9b888a5 100644 --- a/fs/erofs/super.c +++ b/fs/erofs/super.c @@ -158,8 +158,8 @@ static int erofs_read_superblock(struct super_block *sb) blkszbits = dsb->blkszbits; /* 9(512 bytes) + LOG_SECTORS_PER_BLOCK == LOG_BLOCK_SIZE */ if (blkszbits != LOG_BLOCK_SIZE) { - erofs_err(sb, "blksize %u isn't supported on this platform", - 1 << blkszbits); + erofs_err(sb, "blkszbits %u isn't supported on this platform", + blkszbits); goto out; } diff --git a/fs/erofs/xattr.c b/fs/erofs/xattr.c index 5bde77d70852..47314a26767a 100644 --- a/fs/erofs/xattr.c +++ b/fs/erofs/xattr.c @@ -48,8 +48,14 @@ static int init_inode_xattrs(struct inode *inode) int ret = 0; /* the most case is that xattrs of this inode are initialized. */ - if (test_bit(EROFS_I_EA_INITED_BIT, &vi->flags)) + if (test_bit(EROFS_I_EA_INITED_BIT, &vi->flags)) { + /* + * paired with smp_mb() at the end of the function to ensure + * fields will only be observed after the bit is set. + */ + smp_mb(); return 0; + } if (wait_on_bit_lock(&vi->flags, EROFS_I_BL_XATTR_BIT, TASK_KILLABLE)) return -ERESTARTSYS; @@ -137,6 +143,8 @@ static int init_inode_xattrs(struct inode *inode) } xattr_iter_end(&it, atomic_map); + /* paired with smp_mb() at the beginning of the function. */ + smp_mb(); set_bit(EROFS_I_EA_INITED_BIT, &vi->flags); out_unlock: diff --git a/fs/erofs/zdata.c b/fs/erofs/zdata.c index 6cb356c4217b..3851e1a64f73 100644 --- a/fs/erofs/zdata.c +++ b/fs/erofs/zdata.c @@ -1235,7 +1235,7 @@ submit_bio_retry: } if (!bio) { - bio = bio_alloc(GFP_NOIO, BIO_MAX_PAGES); + bio = bio_alloc(GFP_NOIO, BIO_MAX_VECS); bio->bi_end_io = z_erofs_decompressqueue_endio; bio_set_dev(bio, sb->s_bdev); diff --git a/fs/erofs/zmap.c b/fs/erofs/zmap.c index ae325541884e..14d2de35110c 100644 --- a/fs/erofs/zmap.c +++ b/fs/erofs/zmap.c @@ -36,8 +36,14 @@ static int z_erofs_fill_inode_lazy(struct inode *inode) void *kaddr; struct z_erofs_map_header *h; - if (test_bit(EROFS_I_Z_INITED_BIT, &vi->flags)) + if (test_bit(EROFS_I_Z_INITED_BIT, &vi->flags)) { + /* + * paired with smp_mb() at the end of the function to ensure + * fields will only be observed after the bit is set. + */ + smp_mb(); return 0; + } if (wait_on_bit_lock(&vi->flags, EROFS_I_BL_Z_BIT, TASK_KILLABLE)) return -ERESTARTSYS; @@ -83,6 +89,8 @@ static int z_erofs_fill_inode_lazy(struct inode *inode) vi->z_physical_clusterbits[1] = vi->z_logical_clusterbits + ((h->h_clusterbits >> 5) & 7); + /* paired with smp_mb() at the beginning of the function */ + smp_mb(); set_bit(EROFS_I_Z_INITED_BIT, &vi->flags); unmap_done: kunmap_atomic(kaddr); diff --git a/fs/eventpoll.c b/fs/eventpoll.c index a829af074eb5..3196474cbe24 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -979,7 +979,7 @@ static struct epitem *ep_find(struct eventpoll *ep, struct file *file, int fd) return epir; } -#ifdef CONFIG_CHECKPOINT_RESTORE +#ifdef CONFIG_KCMP static struct epitem *ep_find_tfd(struct eventpoll *ep, int tfd, unsigned long toff) { struct rb_node *rbp; @@ -1021,7 +1021,7 @@ struct file *get_epoll_tfile_raw_ptr(struct file *file, int tfd, return file_raw; } -#endif /* CONFIG_CHECKPOINT_RESTORE */ +#endif /* CONFIG_KCMP */ /** * Adds a new entry to the tail of the list in a lockless way, i.e. diff --git a/fs/exec.c b/fs/exec.c index 5d4d52039105..18594f11c31f 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -708,7 +708,7 @@ static int shift_arg_pages(struct vm_area_struct *vma, unsigned long shift) return -ENOMEM; lru_add_drain(); - tlb_gather_mmu(&tlb, mm, old_start, old_end); + tlb_gather_mmu(&tlb, mm); if (new_end > old_start) { /* * when the old and new regions overlap clear from new_end. @@ -725,7 +725,7 @@ static int shift_arg_pages(struct vm_area_struct *vma, unsigned long shift) free_pgd_range(&tlb, old_start, old_end, new_end, vma->vm_next ? vma->vm_next->vm_start : USER_PGTABLES_CEILING); } - tlb_finish_mmu(&tlb, old_start, old_end); + tlb_finish_mmu(&tlb); /* * Shrink the vma to just the new range. Always succeeds. @@ -1404,14 +1404,15 @@ EXPORT_SYMBOL(begin_new_exec); void would_dump(struct linux_binprm *bprm, struct file *file) { struct inode *inode = file_inode(file); - if (inode_permission(inode, MAY_READ) < 0) { + struct user_namespace *mnt_userns = file_mnt_user_ns(file); + if (inode_permission(mnt_userns, inode, MAY_READ) < 0) { struct user_namespace *old, *user_ns; bprm->interp_flags |= BINPRM_FLAGS_ENFORCE_NONDUMP; /* Ensure mm->user_ns contains the executable */ user_ns = old = bprm->mm->user_ns; while ((user_ns != &init_user_ns) && - !privileged_wrt_inode_uidgid(user_ns, inode)) + !privileged_wrt_inode_uidgid(user_ns, mnt_userns, inode)) user_ns = user_ns->parent; if (old != user_ns) { @@ -1454,7 +1455,7 @@ EXPORT_SYMBOL(finalize_exec); /* * Prepare credentials and lock ->cred_guard_mutex. * setup_new_exec() commits the new creds and drops the lock. - * Or, if exec fails before, free_bprm() should release ->cred and + * Or, if exec fails before, free_bprm() should release ->cred * and unlock. */ static int prepare_bprm_creds(struct linux_binprm *bprm) @@ -1579,6 +1580,7 @@ static void check_unsafe_exec(struct linux_binprm *bprm) static void bprm_fill_uid(struct linux_binprm *bprm, struct file *file) { /* Handle suid and sgid on files */ + struct user_namespace *mnt_userns; struct inode *inode; unsigned int mode; kuid_t uid; @@ -1595,13 +1597,15 @@ static void bprm_fill_uid(struct linux_binprm *bprm, struct file *file) if (!(mode & (S_ISUID|S_ISGID))) return; + mnt_userns = file_mnt_user_ns(file); + /* Be careful if suid/sgid is set */ inode_lock(inode); /* reload atomically mode/uid/gid now that lock held */ mode = inode->i_mode; - uid = inode->i_uid; - gid = inode->i_gid; + uid = i_uid_into_mnt(mnt_userns, inode); + gid = i_gid_into_mnt(mnt_userns, inode); inode_unlock(inode); /* We ignore suid/sgid if there are no mappings for them in the ns */ @@ -1837,7 +1841,7 @@ static int bprm_execve(struct linux_binprm *bprm, out: /* - * If past the point of no return ensure the the code never + * If past the point of no return ensure the code never * returns to the userspace process. Use an existing fatal * signal if present otherwise terminate the process with * SIGSEGV. diff --git a/fs/exfat/balloc.c b/fs/exfat/balloc.c index a987919686c0..761c79c3a4ba 100644 --- a/fs/exfat/balloc.c +++ b/fs/exfat/balloc.c @@ -166,7 +166,7 @@ int exfat_set_bitmap(struct inode *inode, unsigned int clu) * If the value of "clu" is 0, it means cluster 2 which is the first cluster of * the cluster heap. */ -void exfat_clear_bitmap(struct inode *inode, unsigned int clu) +void exfat_clear_bitmap(struct inode *inode, unsigned int clu, bool sync) { int i, b; unsigned int ent_idx; @@ -180,7 +180,7 @@ void exfat_clear_bitmap(struct inode *inode, unsigned int clu) b = BITMAP_OFFSET_BIT_IN_SECTOR(sb, ent_idx); clear_bit_le(b, sbi->vol_amap[i]->b_data); - exfat_update_bh(sbi->vol_amap[i], IS_DIRSYNC(inode)); + exfat_update_bh(sbi->vol_amap[i], sync); if (opts->discard) { int ret_discard; diff --git a/fs/exfat/exfat_fs.h b/fs/exfat/exfat_fs.h index b8f0e829ecbd..fa21421a14d9 100644 --- a/fs/exfat/exfat_fs.h +++ b/fs/exfat/exfat_fs.h @@ -408,7 +408,7 @@ int exfat_count_num_clusters(struct super_block *sb, int exfat_load_bitmap(struct super_block *sb); void exfat_free_bitmap(struct exfat_sb_info *sbi); int exfat_set_bitmap(struct inode *inode, unsigned int clu); -void exfat_clear_bitmap(struct inode *inode, unsigned int clu); +void exfat_clear_bitmap(struct inode *inode, unsigned int clu, bool sync); unsigned int exfat_find_free_bitmap(struct super_block *sb, unsigned int clu); int exfat_count_used_clusters(struct super_block *sb, unsigned int *ret_count); @@ -416,9 +416,11 @@ int exfat_count_used_clusters(struct super_block *sb, unsigned int *ret_count); extern const struct file_operations exfat_file_operations; int __exfat_truncate(struct inode *inode, loff_t new_size); void exfat_truncate(struct inode *inode, loff_t size); -int exfat_setattr(struct dentry *dentry, struct iattr *attr); -int exfat_getattr(const struct path *path, struct kstat *stat, - unsigned int request_mask, unsigned int query_flags); +int exfat_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *attr); +int exfat_getattr(struct user_namespace *mnt_userns, const struct path *path, + struct kstat *stat, unsigned int request_mask, + unsigned int query_flags); int exfat_file_fsync(struct file *file, loff_t start, loff_t end, int datasync); /* namei.c */ diff --git a/fs/exfat/exfat_raw.h b/fs/exfat/exfat_raw.h index 6aec6288e1f2..7f39b1c6469c 100644 --- a/fs/exfat/exfat_raw.h +++ b/fs/exfat/exfat_raw.h @@ -77,6 +77,10 @@ #define EXFAT_FILE_NAME_LEN 15 +#define EXFAT_MIN_SECT_SIZE_BITS 9 +#define EXFAT_MAX_SECT_SIZE_BITS 12 +#define EXFAT_MAX_SECT_PER_CLUS_BITS(x) (25 - (x)->sect_size_bits) + /* EXFAT: Main and Backup Boot Sector (512 bytes) */ struct boot_sector { __u8 jmp_boot[BOOTSEC_JUMP_BOOT_LEN]; diff --git a/fs/exfat/fatent.c b/fs/exfat/fatent.c index c3c9afee7418..7b2e8af17193 100644 --- a/fs/exfat/fatent.c +++ b/fs/exfat/fatent.c @@ -157,6 +157,7 @@ int exfat_free_cluster(struct inode *inode, struct exfat_chain *p_chain) unsigned int clu; struct super_block *sb = inode->i_sb; struct exfat_sb_info *sbi = EXFAT_SB(sb); + int cur_cmap_i, next_cmap_i; /* invalid cluster number */ if (p_chain->dir == EXFAT_FREE_CLUSTER || @@ -176,21 +177,51 @@ int exfat_free_cluster(struct inode *inode, struct exfat_chain *p_chain) clu = p_chain->dir; + cur_cmap_i = next_cmap_i = + BITMAP_OFFSET_SECTOR_INDEX(sb, CLUSTER_TO_BITMAP_ENT(clu)); + if (p_chain->flags == ALLOC_NO_FAT_CHAIN) { + unsigned int last_cluster = p_chain->dir + p_chain->size - 1; do { - exfat_clear_bitmap(inode, clu); - clu++; + bool sync = false; + + if (clu < last_cluster) + next_cmap_i = + BITMAP_OFFSET_SECTOR_INDEX(sb, CLUSTER_TO_BITMAP_ENT(clu+1)); + /* flush bitmap only if index would be changed or for last cluster */ + if (clu == last_cluster || cur_cmap_i != next_cmap_i) { + sync = true; + cur_cmap_i = next_cmap_i; + } + + exfat_clear_bitmap(inode, clu, (sync && IS_DIRSYNC(inode))); + clu++; num_clusters++; } while (num_clusters < p_chain->size); } else { do { - exfat_clear_bitmap(inode, clu); - - if (exfat_get_next_cluster(sb, &clu)) - goto dec_used_clus; + bool sync = false; + unsigned int n_clu = clu; + int err = exfat_get_next_cluster(sb, &n_clu); + + if (err || n_clu == EXFAT_EOF_CLUSTER) + sync = true; + else + next_cmap_i = + BITMAP_OFFSET_SECTOR_INDEX(sb, CLUSTER_TO_BITMAP_ENT(n_clu)); + + if (cur_cmap_i != next_cmap_i) { + sync = true; + cur_cmap_i = next_cmap_i; + } + exfat_clear_bitmap(inode, clu, (sync && IS_DIRSYNC(inode))); + clu = n_clu; num_clusters++; + + if (err) + goto dec_used_clus; } while (clu != EXFAT_EOF_CLUSTER); } diff --git a/fs/exfat/file.c b/fs/exfat/file.c index a92478eabfa4..f783cf38dd8e 100644 --- a/fs/exfat/file.c +++ b/fs/exfat/file.c @@ -267,13 +267,14 @@ write_size: mutex_unlock(&sbi->s_lock); } -int exfat_getattr(const struct path *path, struct kstat *stat, - unsigned int request_mask, unsigned int query_flags) +int exfat_getattr(struct user_namespace *mnt_uerns, const struct path *path, + struct kstat *stat, unsigned int request_mask, + unsigned int query_flags) { struct inode *inode = d_backing_inode(path->dentry); struct exfat_inode_info *ei = EXFAT_I(inode); - generic_fillattr(inode, stat); + generic_fillattr(&init_user_ns, inode, stat); exfat_truncate_atime(&stat->atime); stat->result_mask |= STATX_BTIME; stat->btime.tv_sec = ei->i_crtime.tv_sec; @@ -282,7 +283,8 @@ int exfat_getattr(const struct path *path, struct kstat *stat, return 0; } -int exfat_setattr(struct dentry *dentry, struct iattr *attr) +int exfat_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *attr) { struct exfat_sb_info *sbi = EXFAT_SB(dentry->d_sb); struct inode *inode = dentry->d_inode; @@ -305,7 +307,7 @@ int exfat_setattr(struct dentry *dentry, struct iattr *attr) ATTR_TIMES_SET); } - error = setattr_prepare(dentry, attr); + error = setattr_prepare(&init_user_ns, dentry, attr); attr->ia_valid = ia_valid; if (error) goto out; @@ -340,7 +342,7 @@ int exfat_setattr(struct dentry *dentry, struct iattr *attr) up_write(&EXFAT_I(inode)->truncate_lock); } - setattr_copy(inode, attr); + setattr_copy(&init_user_ns, inode, attr); exfat_truncate_atime(&inode->i_atime); mark_inode_dirty(inode); @@ -361,7 +363,7 @@ int exfat_file_fsync(struct file *filp, loff_t start, loff_t end, int datasync) if (err) return err; - return blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL); + return blkdev_issue_flush(inode->i_sb->s_bdev); } const struct file_operations exfat_file_operations = { diff --git a/fs/exfat/namei.c b/fs/exfat/namei.c index 2932b23a3b6c..d9e8ec689c55 100644 --- a/fs/exfat/namei.c +++ b/fs/exfat/namei.c @@ -541,8 +541,8 @@ out: return ret; } -static int exfat_create(struct inode *dir, struct dentry *dentry, umode_t mode, - bool excl) +static int exfat_create(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, bool excl) { struct super_block *sb = dir->i_sb; struct inode *inode; @@ -827,7 +827,8 @@ unlock: return err; } -static int exfat_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) +static int exfat_mkdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { struct super_block *sb = dir->i_sb; struct inode *inode; @@ -1318,9 +1319,10 @@ out: return ret; } -static int exfat_rename(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry, - unsigned int flags) +static int exfat_rename(struct user_namespace *mnt_userns, + struct inode *old_dir, struct dentry *old_dentry, + struct inode *new_dir, struct dentry *new_dentry, + unsigned int flags) { struct inode *old_inode, *new_inode; struct super_block *sb = old_dir->i_sb; diff --git a/fs/exfat/super.c b/fs/exfat/super.c index 87be5bfc31eb..c6d8d2e53486 100644 --- a/fs/exfat/super.c +++ b/fs/exfat/super.c @@ -381,8 +381,7 @@ static int exfat_calibrate_blocksize(struct super_block *sb, int logical_sect) { struct exfat_sb_info *sbi = EXFAT_SB(sb); - if (!is_power_of_2(logical_sect) || - logical_sect < 512 || logical_sect > 4096) { + if (!is_power_of_2(logical_sect)) { exfat_err(sb, "bogus logical sector size %u", logical_sect); return -EIO; } @@ -451,6 +450,25 @@ static int exfat_read_boot_sector(struct super_block *sb) return -EINVAL; } + /* + * sect_size_bits could be at least 9 and at most 12. + */ + if (p_boot->sect_size_bits < EXFAT_MIN_SECT_SIZE_BITS || + p_boot->sect_size_bits > EXFAT_MAX_SECT_SIZE_BITS) { + exfat_err(sb, "bogus sector size bits : %u\n", + p_boot->sect_size_bits); + return -EINVAL; + } + + /* + * sect_per_clus_bits could be at least 0 and at most 25 - sect_size_bits. + */ + if (p_boot->sect_per_clus_bits > EXFAT_MAX_SECT_PER_CLUS_BITS(p_boot)) { + exfat_err(sb, "bogus sectors bits per cluster : %u\n", + p_boot->sect_per_clus_bits); + return -EINVAL; + } + sbi->sect_per_clus = 1 << p_boot->sect_per_clus_bits; sbi->sect_per_clus_bits = p_boot->sect_per_clus_bits; sbi->cluster_size_bits = p_boot->sect_per_clus_bits + @@ -477,16 +495,19 @@ static int exfat_read_boot_sector(struct super_block *sb) sbi->used_clusters = EXFAT_CLUSTERS_UNTRACKED; /* check consistencies */ - if (sbi->num_FAT_sectors << p_boot->sect_size_bits < - sbi->num_clusters * 4) { + if ((u64)sbi->num_FAT_sectors << p_boot->sect_size_bits < + (u64)sbi->num_clusters * 4) { exfat_err(sb, "bogus fat length"); return -EINVAL; } + if (sbi->data_start_sector < - sbi->FAT1_start_sector + sbi->num_FAT_sectors * p_boot->num_fats) { + (u64)sbi->FAT1_start_sector + + (u64)sbi->num_FAT_sectors * p_boot->num_fats) { exfat_err(sb, "bogus data start sector"); return -EINVAL; } + if (sbi->vol_flags & VOLUME_DIRTY) exfat_warn(sb, "Volume was not properly unmounted. Some data may be corrupt. Please run fsck."); if (sbi->vol_flags & MEDIA_FAILURE) diff --git a/fs/ext2/acl.c b/fs/ext2/acl.c index cf4c77f8dd08..b9a9db98e94b 100644 --- a/fs/ext2/acl.c +++ b/fs/ext2/acl.c @@ -216,14 +216,16 @@ __ext2_set_acl(struct inode *inode, struct posix_acl *acl, int type) * inode->i_mutex: down */ int -ext2_set_acl(struct inode *inode, struct posix_acl *acl, int type) +ext2_set_acl(struct user_namespace *mnt_userns, struct inode *inode, + struct posix_acl *acl, int type) { int error; int update_mode = 0; umode_t mode = inode->i_mode; if (type == ACL_TYPE_ACCESS && acl) { - error = posix_acl_update_mode(inode, &mode, &acl); + error = posix_acl_update_mode(&init_user_ns, inode, &mode, + &acl); if (error) return error; update_mode = 1; diff --git a/fs/ext2/acl.h b/fs/ext2/acl.h index 0f01c759daac..917db5f6630a 100644 --- a/fs/ext2/acl.h +++ b/fs/ext2/acl.h @@ -56,7 +56,8 @@ static inline int ext2_acl_count(size_t size) /* acl.c */ extern struct posix_acl *ext2_get_acl(struct inode *inode, int type); -extern int ext2_set_acl(struct inode *inode, struct posix_acl *acl, int type); +extern int ext2_set_acl(struct user_namespace *mnt_userns, struct inode *inode, + struct posix_acl *acl, int type); extern int ext2_init_acl (struct inode *, struct inode *); #else diff --git a/fs/ext2/ext2.h b/fs/ext2/ext2.h index 2a4175fbaf5e..3309fb2d327a 100644 --- a/fs/ext2/ext2.h +++ b/fs/ext2/ext2.h @@ -764,8 +764,9 @@ extern struct inode *ext2_iget (struct super_block *, unsigned long); extern int ext2_write_inode (struct inode *, struct writeback_control *); extern void ext2_evict_inode(struct inode *); extern int ext2_get_block(struct inode *, sector_t, struct buffer_head *, int); -extern int ext2_setattr (struct dentry *, struct iattr *); -extern int ext2_getattr (const struct path *, struct kstat *, u32, unsigned int); +extern int ext2_setattr (struct user_namespace *, struct dentry *, struct iattr *); +extern int ext2_getattr (struct user_namespace *, const struct path *, + struct kstat *, u32, unsigned int); extern void ext2_set_inode_flags(struct inode *inode); extern int ext2_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo, u64 start, u64 len); diff --git a/fs/ext2/ialloc.c b/fs/ext2/ialloc.c index 432c3febea6d..df14e750e9fe 100644 --- a/fs/ext2/ialloc.c +++ b/fs/ext2/ialloc.c @@ -551,7 +551,7 @@ got: inode->i_uid = current_fsuid(); inode->i_gid = dir->i_gid; } else - inode_init_owner(inode, dir, mode); + inode_init_owner(&init_user_ns, inode, dir, mode); inode->i_ino = ino; inode->i_blocks = 0; diff --git a/fs/ext2/inode.c b/fs/ext2/inode.c index 78c417d3c898..68178b2234bd 100644 --- a/fs/ext2/inode.c +++ b/fs/ext2/inode.c @@ -1638,8 +1638,8 @@ int ext2_write_inode(struct inode *inode, struct writeback_control *wbc) return __ext2_write_inode(inode, wbc->sync_mode == WB_SYNC_ALL); } -int ext2_getattr(const struct path *path, struct kstat *stat, - u32 request_mask, unsigned int query_flags) +int ext2_getattr(struct user_namespace *mnt_userns, const struct path *path, + struct kstat *stat, u32 request_mask, unsigned int query_flags) { struct inode *inode = d_inode(path->dentry); struct ext2_inode_info *ei = EXT2_I(inode); @@ -1660,16 +1660,17 @@ int ext2_getattr(const struct path *path, struct kstat *stat, STATX_ATTR_IMMUTABLE | STATX_ATTR_NODUMP); - generic_fillattr(inode, stat); + generic_fillattr(&init_user_ns, inode, stat); return 0; } -int ext2_setattr(struct dentry *dentry, struct iattr *iattr) +int ext2_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *iattr) { struct inode *inode = d_inode(dentry); int error; - error = setattr_prepare(dentry, iattr); + error = setattr_prepare(&init_user_ns, dentry, iattr); if (error) return error; @@ -1689,9 +1690,9 @@ int ext2_setattr(struct dentry *dentry, struct iattr *iattr) if (error) return error; } - setattr_copy(inode, iattr); + setattr_copy(&init_user_ns, inode, iattr); if (iattr->ia_valid & ATTR_MODE) - error = posix_acl_chmod(inode, inode->i_mode); + error = posix_acl_chmod(&init_user_ns, inode, inode->i_mode); mark_inode_dirty(inode); return error; diff --git a/fs/ext2/ioctl.c b/fs/ext2/ioctl.c index 32a8d10b579d..b399cbb7022d 100644 --- a/fs/ext2/ioctl.c +++ b/fs/ext2/ioctl.c @@ -39,7 +39,7 @@ long ext2_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) if (ret) return ret; - if (!inode_owner_or_capable(inode)) { + if (!inode_owner_or_capable(&init_user_ns, inode)) { ret = -EACCES; goto setflags_out; } @@ -84,7 +84,7 @@ setflags_out: case EXT2_IOC_SETVERSION: { __u32 generation; - if (!inode_owner_or_capable(inode)) + if (!inode_owner_or_capable(&init_user_ns, inode)) return -EPERM; ret = mnt_want_write_file(filp); if (ret) @@ -117,7 +117,7 @@ setversion_out: if (!test_opt(inode->i_sb, RESERVATION) ||!S_ISREG(inode->i_mode)) return -ENOTTY; - if (!inode_owner_or_capable(inode)) + if (!inode_owner_or_capable(&init_user_ns, inode)) return -EACCES; if (get_user(rsv_window_size, (int __user *)arg)) diff --git a/fs/ext2/namei.c b/fs/ext2/namei.c index ea980f1e2e99..3367384d344d 100644 --- a/fs/ext2/namei.c +++ b/fs/ext2/namei.c @@ -100,7 +100,9 @@ struct dentry *ext2_get_parent(struct dentry *child) * If the create succeeds, we fill in the inode information * with d_instantiate(). */ -static int ext2_create (struct inode * dir, struct dentry * dentry, umode_t mode, bool excl) +static int ext2_create (struct user_namespace * mnt_userns, + struct inode * dir, struct dentry * dentry, + umode_t mode, bool excl) { struct inode *inode; int err; @@ -118,7 +120,8 @@ static int ext2_create (struct inode * dir, struct dentry * dentry, umode_t mode return ext2_add_nondir(dentry, inode); } -static int ext2_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode) +static int ext2_tmpfile(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { struct inode *inode = ext2_new_inode(dir, mode, NULL); if (IS_ERR(inode)) @@ -131,7 +134,8 @@ static int ext2_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode) return 0; } -static int ext2_mknod (struct inode * dir, struct dentry *dentry, umode_t mode, dev_t rdev) +static int ext2_mknod (struct user_namespace * mnt_userns, struct inode * dir, + struct dentry *dentry, umode_t mode, dev_t rdev) { struct inode * inode; int err; @@ -151,8 +155,8 @@ static int ext2_mknod (struct inode * dir, struct dentry *dentry, umode_t mode, return err; } -static int ext2_symlink (struct inode * dir, struct dentry * dentry, - const char * symname) +static int ext2_symlink (struct user_namespace * mnt_userns, struct inode * dir, + struct dentry * dentry, const char * symname) { struct super_block * sb = dir->i_sb; int err = -ENAMETOOLONG; @@ -225,7 +229,8 @@ static int ext2_link (struct dentry * old_dentry, struct inode * dir, return err; } -static int ext2_mkdir(struct inode * dir, struct dentry * dentry, umode_t mode) +static int ext2_mkdir(struct user_namespace * mnt_userns, + struct inode * dir, struct dentry * dentry, umode_t mode) { struct inode * inode; int err; @@ -315,8 +320,9 @@ static int ext2_rmdir (struct inode * dir, struct dentry *dentry) return err; } -static int ext2_rename (struct inode * old_dir, struct dentry * old_dentry, - struct inode * new_dir, struct dentry * new_dentry, +static int ext2_rename (struct user_namespace * mnt_userns, + struct inode * old_dir, struct dentry * old_dentry, + struct inode * new_dir, struct dentry * new_dentry, unsigned int flags) { struct inode * old_inode = d_inode(old_dentry); diff --git a/fs/ext2/xattr_security.c b/fs/ext2/xattr_security.c index 9a682e440acb..ebade1f52451 100644 --- a/fs/ext2/xattr_security.c +++ b/fs/ext2/xattr_security.c @@ -19,6 +19,7 @@ ext2_xattr_security_get(const struct xattr_handler *handler, static int ext2_xattr_security_set(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *unused, struct inode *inode, const char *name, const void *value, size_t size, int flags) diff --git a/fs/ext2/xattr_trusted.c b/fs/ext2/xattr_trusted.c index 49add1107850..18a87d5dd1ab 100644 --- a/fs/ext2/xattr_trusted.c +++ b/fs/ext2/xattr_trusted.c @@ -26,6 +26,7 @@ ext2_xattr_trusted_get(const struct xattr_handler *handler, static int ext2_xattr_trusted_set(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *unused, struct inode *inode, const char *name, const void *value, size_t size, int flags) diff --git a/fs/ext2/xattr_user.c b/fs/ext2/xattr_user.c index c243a3b4d69d..58092449f8ff 100644 --- a/fs/ext2/xattr_user.c +++ b/fs/ext2/xattr_user.c @@ -30,6 +30,7 @@ ext2_xattr_user_get(const struct xattr_handler *handler, static int ext2_xattr_user_set(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *unused, struct inode *inode, const char *name, const void *value, size_t size, int flags) diff --git a/fs/ext4/.kunitconfig b/fs/ext4/.kunitconfig new file mode 100644 index 000000000000..bf51da7cd9fc --- /dev/null +++ b/fs/ext4/.kunitconfig @@ -0,0 +1,3 @@ +CONFIG_KUNIT=y +CONFIG_EXT4_FS=y +CONFIG_EXT4_KUNIT_TESTS=y diff --git a/fs/ext4/Kconfig b/fs/ext4/Kconfig index 619dd35ddd48..86699c8cab28 100644 --- a/fs/ext4/Kconfig +++ b/fs/ext4/Kconfig @@ -103,8 +103,7 @@ config EXT4_DEBUG config EXT4_KUNIT_TESTS tristate "KUnit tests for ext4" if !KUNIT_ALL_TESTS - select EXT4_FS - depends on KUNIT + depends on EXT4_FS && KUNIT default KUNIT_ALL_TESTS help This builds the ext4 KUnit tests. diff --git a/fs/ext4/acl.c b/fs/ext4/acl.c index 68aaed48315f..c5eaffccecc3 100644 --- a/fs/ext4/acl.c +++ b/fs/ext4/acl.c @@ -222,7 +222,8 @@ __ext4_set_acl(handle_t *handle, struct inode *inode, int type, } int -ext4_set_acl(struct inode *inode, struct posix_acl *acl, int type) +ext4_set_acl(struct user_namespace *mnt_userns, struct inode *inode, + struct posix_acl *acl, int type) { handle_t *handle; int error, credits, retries = 0; @@ -245,7 +246,7 @@ retry: ext4_fc_start_update(inode); if ((type == ACL_TYPE_ACCESS) && acl) { - error = posix_acl_update_mode(inode, &mode, &acl); + error = posix_acl_update_mode(mnt_userns, inode, &mode, &acl); if (error) goto out_stop; if (mode != inode->i_mode) diff --git a/fs/ext4/acl.h b/fs/ext4/acl.h index 9b63f5416a2f..84b8942a57f2 100644 --- a/fs/ext4/acl.h +++ b/fs/ext4/acl.h @@ -56,7 +56,8 @@ static inline int ext4_acl_count(size_t size) /* acl.c */ struct posix_acl *ext4_get_acl(struct inode *inode, int type); -int ext4_set_acl(struct inode *inode, struct posix_acl *acl, int type); +int ext4_set_acl(struct user_namespace *mnt_userns, struct inode *inode, + struct posix_acl *acl, int type); extern int ext4_init_acl(handle_t *, struct inode *, struct inode *); #else /* CONFIG_EXT4_FS_POSIX_ACL */ diff --git a/fs/ext4/balloc.c b/fs/ext4/balloc.c index f45f9feebe59..74a5172c2d83 100644 --- a/fs/ext4/balloc.c +++ b/fs/ext4/balloc.c @@ -626,27 +626,41 @@ int ext4_claim_free_clusters(struct ext4_sb_info *sbi, /** * ext4_should_retry_alloc() - check if a block allocation should be retried - * @sb: super block - * @retries: number of attemps has been made + * @sb: superblock + * @retries: number of retry attempts made so far * - * ext4_should_retry_alloc() is called when ENOSPC is returned, and if - * it is profitable to retry the operation, this function will wait - * for the current or committing transaction to complete, and then - * return TRUE. We will only retry once. + * ext4_should_retry_alloc() is called when ENOSPC is returned while + * attempting to allocate blocks. If there's an indication that a pending + * journal transaction might free some space and allow another attempt to + * succeed, this function will wait for the current or committing transaction + * to complete and then return TRUE. */ int ext4_should_retry_alloc(struct super_block *sb, int *retries) { - if (!ext4_has_free_clusters(EXT4_SB(sb), 1, 0) || - (*retries)++ > 1 || - !EXT4_SB(sb)->s_journal) + struct ext4_sb_info *sbi = EXT4_SB(sb); + + if (!sbi->s_journal) return 0; - smp_mb(); - if (EXT4_SB(sb)->s_mb_free_pending == 0) + if (++(*retries) > 3) { + percpu_counter_inc(&sbi->s_sra_exceeded_retry_limit); return 0; + } + /* + * if there's no indication that blocks are about to be freed it's + * possible we just missed a transaction commit that did so + */ + smp_mb(); + if (sbi->s_mb_free_pending == 0) + return ext4_has_free_clusters(sbi, 1, 0); + + /* + * it's possible we've just missed a transaction commit here, + * so ignore the returned status + */ jbd_debug(1, "%s: retrying operation after ENOSPC\n", sb->s_id); - jbd2_journal_force_commit_nested(EXT4_SB(sb)->s_journal); + (void) jbd2_journal_force_commit_nested(sbi->s_journal); return 1; } diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h index 2866d249f3d2..826a56e3bbd2 100644 --- a/fs/ext4/ext4.h +++ b/fs/ext4/ext4.h @@ -1484,6 +1484,7 @@ struct ext4_sb_info { struct percpu_counter s_freeinodes_counter; struct percpu_counter s_dirs_counter; struct percpu_counter s_dirtyclusters_counter; + struct percpu_counter s_sra_exceeded_retry_limit; struct blockgroup_lock *s_blockgroup_lock; struct proc_dir_entry *s_proc; struct kobject s_kobj; @@ -2755,18 +2756,19 @@ extern int ext4fs_dirhash(const struct inode *dir, const char *name, int len, /* ialloc.c */ extern int ext4_mark_inode_used(struct super_block *sb, int ino); -extern struct inode *__ext4_new_inode(handle_t *, struct inode *, umode_t, +extern struct inode *__ext4_new_inode(struct user_namespace *, handle_t *, + struct inode *, umode_t, const struct qstr *qstr, __u32 goal, uid_t *owner, __u32 i_flags, int handle_type, unsigned int line_no, int nblocks); -#define ext4_new_inode(handle, dir, mode, qstr, goal, owner, i_flags) \ - __ext4_new_inode((handle), (dir), (mode), (qstr), (goal), (owner), \ - i_flags, 0, 0, 0) -#define ext4_new_inode_start_handle(dir, mode, qstr, goal, owner, \ +#define ext4_new_inode(handle, dir, mode, qstr, goal, owner, i_flags) \ + __ext4_new_inode(&init_user_ns, (handle), (dir), (mode), (qstr), \ + (goal), (owner), i_flags, 0, 0, 0) +#define ext4_new_inode_start_handle(mnt_userns, dir, mode, qstr, goal, owner, \ type, nblocks) \ - __ext4_new_inode(NULL, (dir), (mode), (qstr), (goal), (owner), \ + __ext4_new_inode((mnt_userns), NULL, (dir), (mode), (qstr), (goal), (owner), \ 0, (type), __LINE__, (nblocks)) @@ -2792,6 +2794,8 @@ void __ext4_fc_track_link(handle_t *handle, struct inode *inode, struct dentry *dentry); void ext4_fc_track_unlink(handle_t *handle, struct dentry *dentry); void ext4_fc_track_link(handle_t *handle, struct dentry *dentry); +void __ext4_fc_track_create(handle_t *handle, struct inode *inode, + struct dentry *dentry); void ext4_fc_track_create(handle_t *handle, struct dentry *dentry); void ext4_fc_track_inode(handle_t *handle, struct inode *inode); void ext4_fc_mark_ineligible(struct super_block *sb, int reason); @@ -2877,11 +2881,14 @@ extern struct inode *__ext4_iget(struct super_block *sb, unsigned long ino, __ext4_iget((sb), (ino), (flags), __func__, __LINE__) extern int ext4_write_inode(struct inode *, struct writeback_control *); -extern int ext4_setattr(struct dentry *, struct iattr *); -extern int ext4_getattr(const struct path *, struct kstat *, u32, unsigned int); +extern int ext4_setattr(struct user_namespace *, struct dentry *, + struct iattr *); +extern int ext4_getattr(struct user_namespace *, const struct path *, + struct kstat *, u32, unsigned int); extern void ext4_evict_inode(struct inode *); extern void ext4_clear_inode(struct inode *); -extern int ext4_file_getattr(const struct path *, struct kstat *, u32, unsigned int); +extern int ext4_file_getattr(struct user_namespace *, const struct path *, + struct kstat *, u32, unsigned int); extern int ext4_sync_inode(handle_t *, struct inode *); extern void ext4_dirty_inode(struct inode *, int); extern int ext4_change_inode_journal_flag(struct inode *, int); diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 3960b7ec3ab7..77c84d6f1af6 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -4382,8 +4382,7 @@ static int ext4_alloc_file_blocks(struct file *file, ext4_lblk_t offset, { struct inode *inode = file_inode(file); handle_t *handle; - int ret = 0; - int ret2 = 0, ret3 = 0; + int ret = 0, ret2 = 0, ret3 = 0; int retries = 0; int depth = 0; struct ext4_map_blocks map; @@ -4408,7 +4407,7 @@ static int ext4_alloc_file_blocks(struct file *file, ext4_lblk_t offset, depth = ext_depth(inode); retry: - while (ret >= 0 && len) { + while (len) { /* * Recalculate credits when extent tree depth changes. */ @@ -4430,9 +4429,13 @@ retry: inode->i_ino, map.m_lblk, map.m_len, ret); ext4_mark_inode_dirty(handle, inode); - ret2 = ext4_journal_stop(handle); + ext4_journal_stop(handle); break; } + /* + * allow a full retry cycle for any remaining allocations + */ + retries = 0; map.m_lblk += ret; map.m_len = len = len - ret; epos = (loff_t)map.m_lblk << inode->i_blkbits; @@ -4450,11 +4453,8 @@ retry: if (unlikely(ret2)) break; } - if (ret == -ENOSPC && - ext4_should_retry_alloc(inode->i_sb, &retries)) { - ret = 0; + if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries)) goto retry; - } return ret > 0 ? ret2 : ret; } diff --git a/fs/ext4/fast_commit.c b/fs/ext4/fast_commit.c index 0a14a7c87bf8..7541d0b5d706 100644 --- a/fs/ext4/fast_commit.c +++ b/fs/ext4/fast_commit.c @@ -513,10 +513,10 @@ void ext4_fc_track_link(handle_t *handle, struct dentry *dentry) __ext4_fc_track_link(handle, d_inode(dentry), dentry); } -void ext4_fc_track_create(handle_t *handle, struct dentry *dentry) +void __ext4_fc_track_create(handle_t *handle, struct inode *inode, + struct dentry *dentry) { struct __track_dentry_update_args args; - struct inode *inode = d_inode(dentry); int ret; args.dentry = dentry; @@ -527,6 +527,11 @@ void ext4_fc_track_create(handle_t *handle, struct dentry *dentry) trace_ext4_fc_track_create(inode, dentry, ret); } +void ext4_fc_track_create(handle_t *handle, struct dentry *dentry) +{ + __ext4_fc_track_create(handle, d_inode(dentry), dentry); +} + /* __track_fn for inode tracking */ static int __track_inode(struct inode *inode, void *arg, bool update) { @@ -915,13 +920,11 @@ static int ext4_fc_submit_inode_data_all(journal_t *journal) struct super_block *sb = (struct super_block *)(journal->j_private); struct ext4_sb_info *sbi = EXT4_SB(sb); struct ext4_inode_info *ei; - struct list_head *pos; int ret = 0; spin_lock(&sbi->s_fc_lock); ext4_set_mount_flag(sb, EXT4_MF_FC_COMMITTING); - list_for_each(pos, &sbi->s_fc_q[FC_Q_MAIN]) { - ei = list_entry(pos, struct ext4_inode_info, i_fc_list); + list_for_each_entry(ei, &sbi->s_fc_q[FC_Q_MAIN], i_fc_list) { ext4_set_inode_state(&ei->vfs_inode, EXT4_STATE_FC_COMMITTING); while (atomic_read(&ei->i_fc_updates)) { DEFINE_WAIT(wait); @@ -978,17 +981,15 @@ __releases(&sbi->s_fc_lock) { struct super_block *sb = (struct super_block *)(journal->j_private); struct ext4_sb_info *sbi = EXT4_SB(sb); - struct ext4_fc_dentry_update *fc_dentry; + struct ext4_fc_dentry_update *fc_dentry, *fc_dentry_n; struct inode *inode; - struct list_head *pos, *n, *fcd_pos, *fcd_n; - struct ext4_inode_info *ei; + struct ext4_inode_info *ei, *ei_n; int ret; if (list_empty(&sbi->s_fc_dentry_q[FC_Q_MAIN])) return 0; - list_for_each_safe(fcd_pos, fcd_n, &sbi->s_fc_dentry_q[FC_Q_MAIN]) { - fc_dentry = list_entry(fcd_pos, struct ext4_fc_dentry_update, - fcd_list); + list_for_each_entry_safe(fc_dentry, fc_dentry_n, + &sbi->s_fc_dentry_q[FC_Q_MAIN], fcd_list) { if (fc_dentry->fcd_op != EXT4_FC_TAG_CREAT) { spin_unlock(&sbi->s_fc_lock); if (!ext4_fc_add_dentry_tlv( @@ -1004,8 +1005,8 @@ __releases(&sbi->s_fc_lock) } inode = NULL; - list_for_each_safe(pos, n, &sbi->s_fc_q[FC_Q_MAIN]) { - ei = list_entry(pos, struct ext4_inode_info, i_fc_list); + list_for_each_entry_safe(ei, ei_n, &sbi->s_fc_q[FC_Q_MAIN], + i_fc_list) { if (ei->vfs_inode.i_ino == fc_dentry->fcd_ino) { inode = &ei->vfs_inode; break; @@ -1057,7 +1058,6 @@ static int ext4_fc_perform_commit(journal_t *journal) struct ext4_sb_info *sbi = EXT4_SB(sb); struct ext4_inode_info *iter; struct ext4_fc_head head; - struct list_head *pos; struct inode *inode; struct blk_plug plug; int ret = 0; @@ -1076,7 +1076,7 @@ static int ext4_fc_perform_commit(journal_t *journal) * flush before we start writing fast commit blocks. */ if (journal->j_fs_dev != journal->j_dev) - blkdev_issue_flush(journal->j_fs_dev, GFP_NOFS); + blkdev_issue_flush(journal->j_fs_dev); blk_start_plug(&plug); if (sbi->s_fc_bytes == 0) { @@ -1099,8 +1099,7 @@ static int ext4_fc_perform_commit(journal_t *journal) goto out; } - list_for_each(pos, &sbi->s_fc_q[FC_Q_MAIN]) { - iter = list_entry(pos, struct ext4_inode_info, i_fc_list); + list_for_each_entry(iter, &sbi->s_fc_q[FC_Q_MAIN], i_fc_list) { inode = &iter->vfs_inode; if (!ext4_test_inode_state(inode, EXT4_STATE_FC_COMMITTING)) continue; @@ -1226,9 +1225,8 @@ static void ext4_fc_cleanup(journal_t *journal, int full) { struct super_block *sb = journal->j_private; struct ext4_sb_info *sbi = EXT4_SB(sb); - struct ext4_inode_info *iter; + struct ext4_inode_info *iter, *iter_n; struct ext4_fc_dentry_update *fc_dentry; - struct list_head *pos, *n; if (full && sbi->s_fc_bh) sbi->s_fc_bh = NULL; @@ -1236,8 +1234,8 @@ static void ext4_fc_cleanup(journal_t *journal, int full) jbd2_fc_release_bufs(journal); spin_lock(&sbi->s_fc_lock); - list_for_each_safe(pos, n, &sbi->s_fc_q[FC_Q_MAIN]) { - iter = list_entry(pos, struct ext4_inode_info, i_fc_list); + list_for_each_entry_safe(iter, iter_n, &sbi->s_fc_q[FC_Q_MAIN], + i_fc_list) { list_del_init(&iter->i_fc_list); ext4_clear_inode_state(&iter->vfs_inode, EXT4_STATE_FC_COMMITTING); @@ -1535,7 +1533,7 @@ static int ext4_fc_replay_inode(struct super_block *sb, struct ext4_fc_tl *tl) out: iput(inode); if (!ret) - blkdev_issue_flush(sb->s_bdev, GFP_KERNEL); + blkdev_issue_flush(sb->s_bdev); return 0; } diff --git a/fs/ext4/file.c b/fs/ext4/file.c index 349b27f0dda0..194f5d00fa32 100644 --- a/fs/ext4/file.c +++ b/fs/ext4/file.c @@ -74,8 +74,7 @@ static ssize_t ext4_dio_read_iter(struct kiocb *iocb, struct iov_iter *to) return generic_file_read_iter(iocb, to); } - ret = iomap_dio_rw(iocb, to, &ext4_iomap_ops, NULL, - is_sync_kiocb(iocb)); + ret = iomap_dio_rw(iocb, to, &ext4_iomap_ops, NULL, 0); inode_unlock_shared(inode); file_accessed(iocb->ki_filp); @@ -550,7 +549,7 @@ static ssize_t ext4_dio_write_iter(struct kiocb *iocb, struct iov_iter *from) if (ilock_shared) iomap_ops = &ext4_iomap_overwrite_ops; ret = iomap_dio_rw(iocb, from, iomap_ops, &ext4_dio_write_ops, - is_sync_kiocb(iocb) || unaligned_io || extend); + (unaligned_io || extend) ? IOMAP_DIO_FORCE_WAIT : 0); if (ret == -ENOTBLK) ret = 0; diff --git a/fs/ext4/fsync.c b/fs/ext4/fsync.c index 113bfb023a4a..027a7d7037a0 100644 --- a/fs/ext4/fsync.c +++ b/fs/ext4/fsync.c @@ -174,7 +174,7 @@ int ext4_sync_file(struct file *file, loff_t start, loff_t end, int datasync) ret = ext4_fsync_journal(inode, datasync, &needs_barrier); if (needs_barrier) { - err = blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL); + err = blkdev_issue_flush(inode->i_sb->s_bdev); if (!ret) ret = err; } diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c index b215c564bc31..633ae7becd61 100644 --- a/fs/ext4/ialloc.c +++ b/fs/ext4/ialloc.c @@ -919,7 +919,8 @@ static int ext4_xattr_credits_for_new_inode(struct inode *dir, mode_t mode, * For other inodes, search forward from the parent directory's block * group to find a free inode. */ -struct inode *__ext4_new_inode(handle_t *handle, struct inode *dir, +struct inode *__ext4_new_inode(struct user_namespace *mnt_userns, + handle_t *handle, struct inode *dir, umode_t mode, const struct qstr *qstr, __u32 goal, uid_t *owner, __u32 i_flags, int handle_type, unsigned int line_no, @@ -969,10 +970,10 @@ struct inode *__ext4_new_inode(handle_t *handle, struct inode *dir, i_gid_write(inode, owner[1]); } else if (test_opt(sb, GRPID)) { inode->i_mode = mode; - inode->i_uid = current_fsuid(); + inode->i_uid = fsuid_into_mnt(mnt_userns); inode->i_gid = dir->i_gid; } else - inode_init_owner(inode, dir, mode); + inode_init_owner(mnt_userns, inode, dir, mode); if (ext4_has_feature_project(sb) && ext4_test_inode_flag(dir, EXT4_INODE_PROJINHERIT)) @@ -1583,7 +1584,7 @@ int ext4_init_inode_table(struct super_block *sb, ext4_group_t group, if (ret < 0) goto err_out; if (barrier) - blkdev_issue_flush(sb->s_bdev, GFP_NOFS); + blkdev_issue_flush(sb->s_bdev); skip_zeroout: ext4_lock_group(sb, group); diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index c173c8405856..0948a43f1b3d 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -20,6 +20,7 @@ */ #include <linux/fs.h> +#include <linux/mount.h> #include <linux/time.h> #include <linux/highuid.h> #include <linux/pagemap.h> @@ -1937,13 +1938,13 @@ static int __ext4_journalled_writepage(struct page *page, if (!ret) ret = err; - if (!ext4_has_inline_data(inode)) - ext4_walk_page_buffers(NULL, page_bufs, 0, len, - NULL, bput_one); ext4_set_inode_state(inode, EXT4_STATE_JDATA); out: unlock_page(page); out_no_pagelock: + if (!inline_data && page_bufs) + ext4_walk_page_buffers(NULL, page_bufs, 0, len, + NULL, bput_one); brelse(inode_bh); return ret; } @@ -4961,15 +4962,11 @@ static void __ext4_update_other_inode_time(struct super_block *sb, if (!inode) return; - if ((inode->i_state & (I_FREEING | I_WILL_FREE | I_NEW | - I_DIRTY_INODE)) || - ((inode->i_state & I_DIRTY_TIME) == 0)) + if (!inode_is_dirtytime_only(inode)) return; spin_lock(&inode->i_lock); - if (((inode->i_state & (I_FREEING | I_WILL_FREE | I_NEW | - I_DIRTY_INODE)) == 0) && - (inode->i_state & I_DIRTY_TIME)) { + if (inode_is_dirtytime_only(inode)) { struct ext4_inode_info *ei = EXT4_I(inode); inode->i_state &= ~I_DIRTY_TIME; @@ -5029,7 +5026,7 @@ static int ext4_do_update_inode(handle_t *handle, struct ext4_inode_info *ei = EXT4_I(inode); struct buffer_head *bh = iloc->bh; struct super_block *sb = inode->i_sb; - int err = 0, rc, block; + int err = 0, block; int need_datasync = 0, set_large_file = 0; uid_t i_uid; gid_t i_gid; @@ -5141,9 +5138,9 @@ static int ext4_do_update_inode(handle_t *handle, bh->b_data); BUFFER_TRACE(bh, "call ext4_handle_dirty_metadata"); - rc = ext4_handle_dirty_metadata(handle, NULL, bh); - if (!err) - err = rc; + err = ext4_handle_dirty_metadata(handle, NULL, bh); + if (err) + goto out_brelse; ext4_clear_inode_state(inode, EXT4_STATE_NEW); if (set_large_file) { BUFFER_TRACE(EXT4_SB(sb)->s_sbh, "get write access"); @@ -5319,7 +5316,8 @@ static void ext4_wait_for_tail_page_commit(struct inode *inode) * * Called with inode->i_mutex down. */ -int ext4_setattr(struct dentry *dentry, struct iattr *attr) +int ext4_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *attr) { struct inode *inode = d_inode(dentry); int error, rc = 0; @@ -5337,7 +5335,7 @@ int ext4_setattr(struct dentry *dentry, struct iattr *attr) ATTR_GID | ATTR_TIMES_SET)))) return -EPERM; - error = setattr_prepare(dentry, attr); + error = setattr_prepare(mnt_userns, dentry, attr); if (error) return error; @@ -5389,8 +5387,10 @@ int ext4_setattr(struct dentry *dentry, struct iattr *attr) inode->i_gid = attr->ia_gid; error = ext4_mark_inode_dirty(handle, inode); ext4_journal_stop(handle); - if (unlikely(error)) + if (unlikely(error)) { + ext4_fc_stop_update(inode); return error; + } } if (attr->ia_valid & ATTR_SIZE) { @@ -5512,7 +5512,7 @@ out_mmap_sem: } if (!error) { - setattr_copy(inode, attr); + setattr_copy(mnt_userns, inode, attr); mark_inode_dirty(inode); } @@ -5524,7 +5524,7 @@ out_mmap_sem: ext4_orphan_del(NULL, inode); if (!error && (ia_valid & ATTR_MODE)) - rc = posix_acl_chmod(inode, inode->i_mode); + rc = posix_acl_chmod(mnt_userns, inode, inode->i_mode); err_out: if (error) @@ -5535,8 +5535,8 @@ err_out: return error; } -int ext4_getattr(const struct path *path, struct kstat *stat, - u32 request_mask, unsigned int query_flags) +int ext4_getattr(struct user_namespace *mnt_userns, const struct path *path, + struct kstat *stat, u32 request_mask, unsigned int query_flags) { struct inode *inode = d_inode(path->dentry); struct ext4_inode *raw_inode; @@ -5571,17 +5571,18 @@ int ext4_getattr(const struct path *path, struct kstat *stat, STATX_ATTR_NODUMP | STATX_ATTR_VERITY); - generic_fillattr(inode, stat); + generic_fillattr(mnt_userns, inode, stat); return 0; } -int ext4_file_getattr(const struct path *path, struct kstat *stat, +int ext4_file_getattr(struct user_namespace *mnt_userns, + const struct path *path, struct kstat *stat, u32 request_mask, unsigned int query_flags) { struct inode *inode = d_inode(path->dentry); u64 delalloc_blocks; - ext4_getattr(path, stat, request_mask, query_flags); + ext4_getattr(mnt_userns, path, stat, request_mask, query_flags); /* * If there is inline data in the inode, the inode will normally not @@ -5937,26 +5938,16 @@ out: * If the inode is marked synchronous, we don't honour that here - doing * so would cause a commit on atime updates, which we don't bother doing. * We handle synchronous inodes at the highest possible level. - * - * If only the I_DIRTY_TIME flag is set, we can skip everything. If - * I_DIRTY_TIME and I_DIRTY_SYNC is set, the only inode fields we need - * to copy into the on-disk inode structure are the timestamp files. */ void ext4_dirty_inode(struct inode *inode, int flags) { handle_t *handle; - if (flags == I_DIRTY_TIME) - return; handle = ext4_journal_start(inode, EXT4_HT_INODE, 2); if (IS_ERR(handle)) - goto out; - + return; ext4_mark_inode_dirty(handle, inode); - ext4_journal_stop(handle); -out: - return; } int ext4_change_inode_journal_flag(struct inode *inode, int val) diff --git a/fs/ext4/ioctl.c b/fs/ext4/ioctl.c index d9665d2f82db..a2cf35066f46 100644 --- a/fs/ext4/ioctl.c +++ b/fs/ext4/ioctl.c @@ -107,10 +107,12 @@ void ext4_reset_inode_seed(struct inode *inode) * important fields of the inodes. * * @sb: the super block of the filesystem + * @mnt_userns: user namespace of the mount the inode was found from * @inode: the inode to swap with EXT4_BOOT_LOADER_INO * */ static long swap_inode_boot_loader(struct super_block *sb, + struct user_namespace *mnt_userns, struct inode *inode) { handle_t *handle; @@ -139,7 +141,8 @@ static long swap_inode_boot_loader(struct super_block *sb, } if (IS_RDONLY(inode) || IS_APPEND(inode) || IS_IMMUTABLE(inode) || - !inode_owner_or_capable(inode) || !capable(CAP_SYS_ADMIN)) { + !inode_owner_or_capable(mnt_userns, inode) || + !capable(CAP_SYS_ADMIN)) { err = -EPERM; goto journal_err_out; } @@ -814,6 +817,7 @@ static long __ext4_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) struct inode *inode = file_inode(filp); struct super_block *sb = inode->i_sb; struct ext4_inode_info *ei = EXT4_I(inode); + struct user_namespace *mnt_userns = file_mnt_user_ns(filp); unsigned int flags; ext4_debug("cmd = %u, arg = %lu\n", cmd, arg); @@ -829,7 +833,7 @@ static long __ext4_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) case FS_IOC_SETFLAGS: { int err; - if (!inode_owner_or_capable(inode)) + if (!inode_owner_or_capable(mnt_userns, inode)) return -EACCES; if (get_user(flags, (int __user *) arg)) @@ -871,7 +875,7 @@ static long __ext4_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) __u32 generation; int err; - if (!inode_owner_or_capable(inode)) + if (!inode_owner_or_capable(mnt_userns, inode)) return -EPERM; if (ext4_has_metadata_csum(inode->i_sb)) { @@ -1010,7 +1014,7 @@ mext_out: case EXT4_IOC_MIGRATE: { int err; - if (!inode_owner_or_capable(inode)) + if (!inode_owner_or_capable(mnt_userns, inode)) return -EACCES; err = mnt_want_write_file(filp); @@ -1032,7 +1036,7 @@ mext_out: case EXT4_IOC_ALLOC_DA_BLKS: { int err; - if (!inode_owner_or_capable(inode)) + if (!inode_owner_or_capable(mnt_userns, inode)) return -EACCES; err = mnt_want_write_file(filp); @@ -1051,7 +1055,7 @@ mext_out: err = mnt_want_write_file(filp); if (err) return err; - err = swap_inode_boot_loader(sb, inode); + err = swap_inode_boot_loader(sb, mnt_userns, inode); mnt_drop_write_file(filp); return err; } @@ -1217,7 +1221,7 @@ resizefs_out: case EXT4_IOC_CLEAR_ES_CACHE: { - if (!inode_owner_or_capable(inode)) + if (!inode_owner_or_capable(mnt_userns, inode)) return -EACCES; ext4_clear_inode_es(inode); return 0; @@ -1263,7 +1267,7 @@ resizefs_out: return -EFAULT; /* Make sure caller has proper permission */ - if (!inode_owner_or_capable(inode)) + if (!inode_owner_or_capable(mnt_userns, inode)) return -EACCES; if (fa.fsx_xflags & ~EXT4_SUPPORTED_FS_XFLAGS) @@ -1309,6 +1313,12 @@ out: return -EOPNOTSUPP; return fsverity_ioctl_measure(filp, (void __user *)arg); + case FS_IOC_READ_VERITY_METADATA: + if (!ext4_has_feature_verity(sb)) + return -EOPNOTSUPP; + return fsverity_ioctl_read_metadata(filp, + (const void __user *)arg); + default: return -ENOTTY; } @@ -1391,6 +1401,7 @@ long ext4_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg) case FS_IOC_GETFSMAP: case FS_IOC_ENABLE_VERITY: case FS_IOC_MEASURE_VERITY: + case FS_IOC_READ_VERITY_METADATA: case EXT4_IOC_CLEAR_ES_CACHE: case EXT4_IOC_GETSTATE: case EXT4_IOC_GET_ES_CACHE: diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index 99bf091fee10..a02fadf4fc84 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -2709,8 +2709,15 @@ static int ext4_mb_init_backend(struct super_block *sb) } if (ext4_has_feature_flex_bg(sb)) { - /* a single flex group is supposed to be read by a single IO */ - sbi->s_mb_prefetch = min(1 << sbi->s_es->s_log_groups_per_flex, + /* a single flex group is supposed to be read by a single IO. + * 2 ^ s_log_groups_per_flex != UINT_MAX as s_mb_prefetch is + * unsigned integer, so the maximum shift is 32. + */ + if (sbi->s_es->s_log_groups_per_flex >= 32) { + ext4_msg(sb, KERN_ERR, "too many log groups per flexible block group"); + goto err_freesgi; + } + sbi->s_mb_prefetch = min_t(uint, 1 << sbi->s_es->s_log_groups_per_flex, BLK_MAX_SEGMENT_SIZE >> (sb->s_blocksize_bits - 9)); sbi->s_mb_prefetch *= 8; /* 8 prefetch IOs in flight at most */ } else { diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index cf652ba3e74d..883e2a7cd4ab 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -731,6 +731,29 @@ struct stats dx_show_entries(struct dx_hash_info *hinfo, struct inode *dir, (space/bcount)*100/blocksize); return (struct stats) { names, space, bcount}; } + +/* + * Linear search cross check + */ +static inline void htree_rep_invariant_check(struct dx_entry *at, + struct dx_entry *target, + u32 hash, unsigned int n) +{ + while (n--) { + dxtrace(printk(KERN_CONT ",")); + if (dx_get_hash(++at) > hash) { + at--; + break; + } + } + ASSERT(at == target - 1); +} +#else /* DX_DEBUG */ +static inline void htree_rep_invariant_check(struct dx_entry *at, + struct dx_entry *target, + u32 hash, unsigned int n) +{ +} #endif /* DX_DEBUG */ /* @@ -827,20 +850,7 @@ dx_probe(struct ext4_filename *fname, struct inode *dir, p = m + 1; } - if (0) { // linear search cross check - unsigned n = count - 1; - at = entries; - while (n--) - { - dxtrace(printk(KERN_CONT ",")); - if (dx_get_hash(++at) > hash) - { - at--; - break; - } - } - ASSERT(at == p - 1); - } + htree_rep_invariant_check(entries, p, hash, count - 1); at = p - 1; dxtrace(printk(KERN_CONT " %x->%u\n", @@ -2401,11 +2411,10 @@ again: (frame - 1)->bh); if (err) goto journal_error; - if (restart) { - err = ext4_handle_dirty_dx_node(handle, dir, - frame->bh); + err = ext4_handle_dirty_dx_node(handle, dir, + frame->bh); + if (err) goto journal_error; - } } else { struct dx_root *dxroot; memcpy((char *) entries2, (char *) entries, @@ -2596,8 +2605,8 @@ static int ext4_add_nondir(handle_t *handle, * If the create succeeds, we fill in the inode information * with d_instantiate(). */ -static int ext4_create(struct inode *dir, struct dentry *dentry, umode_t mode, - bool excl) +static int ext4_create(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, bool excl) { handle_t *handle; struct inode *inode; @@ -2610,8 +2619,8 @@ static int ext4_create(struct inode *dir, struct dentry *dentry, umode_t mode, credits = (EXT4_DATA_TRANS_BLOCKS(dir->i_sb) + EXT4_INDEX_EXTRA_TRANS_BLOCKS + 3); retry: - inode = ext4_new_inode_start_handle(dir, mode, &dentry->d_name, 0, - NULL, EXT4_HT_DIR, credits); + inode = ext4_new_inode_start_handle(mnt_userns, dir, mode, &dentry->d_name, + 0, NULL, EXT4_HT_DIR, credits); handle = ext4_journal_current_handle(); err = PTR_ERR(inode); if (!IS_ERR(inode)) { @@ -2631,8 +2640,8 @@ retry: return err; } -static int ext4_mknod(struct inode *dir, struct dentry *dentry, - umode_t mode, dev_t rdev) +static int ext4_mknod(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, dev_t rdev) { handle_t *handle; struct inode *inode; @@ -2645,8 +2654,8 @@ static int ext4_mknod(struct inode *dir, struct dentry *dentry, credits = (EXT4_DATA_TRANS_BLOCKS(dir->i_sb) + EXT4_INDEX_EXTRA_TRANS_BLOCKS + 3); retry: - inode = ext4_new_inode_start_handle(dir, mode, &dentry->d_name, 0, - NULL, EXT4_HT_DIR, credits); + inode = ext4_new_inode_start_handle(mnt_userns, dir, mode, &dentry->d_name, + 0, NULL, EXT4_HT_DIR, credits); handle = ext4_journal_current_handle(); err = PTR_ERR(inode); if (!IS_ERR(inode)) { @@ -2665,7 +2674,8 @@ retry: return err; } -static int ext4_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode) +static int ext4_tmpfile(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { handle_t *handle; struct inode *inode; @@ -2676,7 +2686,7 @@ static int ext4_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode) return err; retry: - inode = ext4_new_inode_start_handle(dir, mode, + inode = ext4_new_inode_start_handle(mnt_userns, dir, mode, NULL, 0, NULL, EXT4_HT_DIR, EXT4_MAXQUOTAS_INIT_BLOCKS(dir->i_sb) + @@ -2774,7 +2784,8 @@ out: return err; } -static int ext4_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) +static int ext4_mkdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { handle_t *handle; struct inode *inode; @@ -2790,7 +2801,7 @@ static int ext4_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) credits = (EXT4_DATA_TRANS_BLOCKS(dir->i_sb) + EXT4_INDEX_EXTRA_TRANS_BLOCKS + 3); retry: - inode = ext4_new_inode_start_handle(dir, S_IFDIR | mode, + inode = ext4_new_inode_start_handle(mnt_userns, dir, S_IFDIR | mode, &dentry->d_name, 0, NULL, EXT4_HT_DIR, credits); handle = ext4_journal_current_handle(); @@ -3292,7 +3303,7 @@ out_trace: return retval; } -static int ext4_symlink(struct inode *dir, +static int ext4_symlink(struct user_namespace *mnt_userns, struct inode *dir, struct dentry *dentry, const char *symname) { handle_t *handle; @@ -3333,7 +3344,7 @@ static int ext4_symlink(struct inode *dir, EXT4_INDEX_EXTRA_TRANS_BLOCKS + 3; } - inode = ext4_new_inode_start_handle(dir, S_IFLNK|S_IRWXUGO, + inode = ext4_new_inode_start_handle(mnt_userns, dir, S_IFLNK|S_IRWXUGO, &dentry->d_name, 0, NULL, EXT4_HT_DIR, credits); handle = ext4_journal_current_handle(); @@ -3602,6 +3613,31 @@ static int ext4_setent(handle_t *handle, struct ext4_renament *ent, return retval; } +static void ext4_resetent(handle_t *handle, struct ext4_renament *ent, + unsigned ino, unsigned file_type) +{ + struct ext4_renament old = *ent; + int retval = 0; + + /* + * old->de could have moved from under us during make indexed dir, + * so the old->de may no longer valid and need to find it again + * before reset old inode info. + */ + old.bh = ext4_find_entry(old.dir, &old.dentry->d_name, &old.de, NULL); + if (IS_ERR(old.bh)) + retval = PTR_ERR(old.bh); + if (!old.bh) + retval = -ENOENT; + if (retval) { + ext4_std_error(old.dir->i_sb, retval); + return; + } + + ext4_setent(handle, &old, ino, file_type); + brelse(old.bh); +} + static int ext4_find_delete_entry(handle_t *handle, struct inode *dir, const struct qstr *d_name) { @@ -3662,7 +3698,8 @@ static void ext4_update_dir_count(handle_t *handle, struct ext4_renament *ent) } } -static struct inode *ext4_whiteout_for_rename(struct ext4_renament *ent, +static struct inode *ext4_whiteout_for_rename(struct user_namespace *mnt_userns, + struct ext4_renament *ent, int credits, handle_t **h) { struct inode *wh; @@ -3676,7 +3713,8 @@ static struct inode *ext4_whiteout_for_rename(struct ext4_renament *ent, credits += (EXT4_MAXQUOTAS_TRANS_BLOCKS(ent->dir->i_sb) + EXT4_XATTR_TRANS_BLOCKS + 4); retry: - wh = ext4_new_inode_start_handle(ent->dir, S_IFCHR | WHITEOUT_MODE, + wh = ext4_new_inode_start_handle(mnt_userns, ent->dir, + S_IFCHR | WHITEOUT_MODE, &ent->dentry->d_name, 0, NULL, EXT4_HT_DIR, credits); @@ -3703,9 +3741,9 @@ retry: * while new_{dentry,inode) refers to the destination dentry/inode * This comes from rename(const char *oldpath, const char *newpath) */ -static int ext4_rename(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry, - unsigned int flags) +static int ext4_rename(struct user_namespace *mnt_userns, struct inode *old_dir, + struct dentry *old_dentry, struct inode *new_dir, + struct dentry *new_dentry, unsigned int flags) { handle_t *handle = NULL; struct ext4_renament old = { @@ -3761,14 +3799,14 @@ static int ext4_rename(struct inode *old_dir, struct dentry *old_dentry, */ retval = -ENOENT; if (!old.bh || le32_to_cpu(old.de->inode) != old.inode->i_ino) - goto end_rename; + goto release_bh; new.bh = ext4_find_entry(new.dir, &new.dentry->d_name, &new.de, &new.inlined); if (IS_ERR(new.bh)) { retval = PTR_ERR(new.bh); new.bh = NULL; - goto end_rename; + goto release_bh; } if (new.bh) { if (!new.inode) { @@ -3785,15 +3823,13 @@ static int ext4_rename(struct inode *old_dir, struct dentry *old_dentry, handle = ext4_journal_start(old.dir, EXT4_HT_DIR, credits); if (IS_ERR(handle)) { retval = PTR_ERR(handle); - handle = NULL; - goto end_rename; + goto release_bh; } } else { - whiteout = ext4_whiteout_for_rename(&old, credits, &handle); + whiteout = ext4_whiteout_for_rename(mnt_userns, &old, credits, &handle); if (IS_ERR(whiteout)) { retval = PTR_ERR(whiteout); - whiteout = NULL; - goto end_rename; + goto release_bh; } } @@ -3837,6 +3873,7 @@ static int ext4_rename(struct inode *old_dir, struct dentry *old_dentry, retval = ext4_mark_inode_dirty(handle, whiteout); if (unlikely(retval)) goto end_rename; + } if (!new.bh) { retval = ext4_add_entry(handle, new.dentry, old.inode); @@ -3910,6 +3947,8 @@ static int ext4_rename(struct inode *old_dir, struct dentry *old_dentry, ext4_fc_track_unlink(handle, new.dentry); __ext4_fc_track_link(handle, old.inode, new.dentry); __ext4_fc_track_unlink(handle, old.inode, old.dentry); + if (whiteout) + __ext4_fc_track_create(handle, whiteout, old.dentry); } if (new.inode) { @@ -3924,19 +3963,21 @@ static int ext4_rename(struct inode *old_dir, struct dentry *old_dentry, end_rename: if (whiteout) { if (retval) { - ext4_setent(handle, &old, - old.inode->i_ino, old_file_type); + ext4_resetent(handle, &old, + old.inode->i_ino, old_file_type); drop_nlink(whiteout); + ext4_orphan_add(handle, whiteout); } unlock_new_inode(whiteout); + ext4_journal_stop(handle); iput(whiteout); - + } else { + ext4_journal_stop(handle); } +release_bh: brelse(old.dir_bh); brelse(old.bh); brelse(new.bh); - if (handle) - ext4_journal_stop(handle); return retval; } @@ -4085,7 +4126,8 @@ end_rename: return retval; } -static int ext4_rename2(struct inode *old_dir, struct dentry *old_dentry, +static int ext4_rename2(struct user_namespace *mnt_userns, + struct inode *old_dir, struct dentry *old_dentry, struct inode *new_dir, struct dentry *new_dentry, unsigned int flags) { @@ -4107,7 +4149,7 @@ static int ext4_rename2(struct inode *old_dir, struct dentry *old_dentry, new_dir, new_dentry); } - return ext4_rename(old_dir, old_dentry, new_dir, new_dentry, flags); + return ext4_rename(mnt_userns, old_dir, old_dentry, new_dir, new_dentry, flags); } /* diff --git a/fs/ext4/page-io.c b/fs/ext4/page-io.c index 03a44a0de86a..f038d578d8d8 100644 --- a/fs/ext4/page-io.c +++ b/fs/ext4/page-io.c @@ -398,7 +398,7 @@ static void io_submit_init_bio(struct ext4_io_submit *io, * bio_alloc will _always_ be able to allocate a bio if * __GFP_DIRECT_RECLAIM is set, see comments for bio_alloc_bioset(). */ - bio = bio_alloc(GFP_NOIO, BIO_MAX_PAGES); + bio = bio_alloc(GFP_NOIO, BIO_MAX_VECS); fscrypt_set_bio_crypt_ctx_bh(bio, bh, GFP_NOIO); bio->bi_iter.bi_sector = bh->b_blocknr * (bh->b_size >> 9); bio_set_dev(bio, bh->b_bdev); diff --git a/fs/ext4/readpage.c b/fs/ext4/readpage.c index f014c5e473a9..3db923403505 100644 --- a/fs/ext4/readpage.c +++ b/fs/ext4/readpage.c @@ -371,8 +371,7 @@ int ext4_mpage_readpages(struct inode *inode, * bio_alloc will _always_ be able to allocate a bio if * __GFP_DIRECT_RECLAIM is set, see bio_alloc_bioset(). */ - bio = bio_alloc(GFP_KERNEL, - min_t(int, nr_pages, BIO_MAX_PAGES)); + bio = bio_alloc(GFP_KERNEL, bio_max_segs(nr_pages)); fscrypt_set_bio_crypt_ctx(bio, inode, next_block, GFP_KERNEL); ext4_set_bio_post_read_ctx(bio, inode, page->index); diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 9a6f9875aa34..b9693680463a 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -59,7 +59,7 @@ #include <trace/events/ext4.h> static struct ext4_lazy_init *ext4_li_info; -static struct mutex ext4_li_mtx; +static DEFINE_MUTEX(ext4_li_mtx); static struct ratelimit_state ext4_mount_msg_ratelimit; static int ext4_load_journal(struct super_block *, struct ext4_super_block *, @@ -1210,6 +1210,7 @@ static void ext4_put_super(struct super_block *sb) percpu_counter_destroy(&sbi->s_freeinodes_counter); percpu_counter_destroy(&sbi->s_dirs_counter); percpu_counter_destroy(&sbi->s_dirtyclusters_counter); + percpu_counter_destroy(&sbi->s_sra_exceeded_retry_limit); percpu_free_rwsem(&sbi->s_writepages_rwsem); #ifdef CONFIG_QUOTA for (i = 0; i < EXT4_MAXQUOTAS; i++) @@ -4875,7 +4876,6 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) set_task_ioprio(sbi->s_journal->j_task, journal_ioprio); - sbi->s_journal->j_commit_callback = ext4_journal_commit_callback; sbi->s_journal->j_submit_inode_data_buffers = ext4_journal_submit_inode_data_buffers; sbi->s_journal->j_finish_inode_data_buffers = @@ -4987,6 +4987,14 @@ no_journal: goto failed_mount5; } + /* + * We can only set up the journal commit callback once + * mballoc is initialized + */ + if (sbi->s_journal) + sbi->s_journal->j_commit_callback = + ext4_journal_commit_callback; + block = ext4_count_free_clusters(sb); ext4_free_blocks_count_set(sbi->s_es, EXT4_C2B(sbi, block)); @@ -5005,6 +5013,9 @@ no_journal: err = percpu_counter_init(&sbi->s_dirtyclusters_counter, 0, GFP_KERNEL); if (!err) + err = percpu_counter_init(&sbi->s_sra_exceeded_retry_limit, 0, + GFP_KERNEL); + if (!err) err = percpu_init_rwsem(&sbi->s_writepages_rwsem); if (err) { @@ -5117,6 +5128,7 @@ failed_mount6: percpu_counter_destroy(&sbi->s_freeinodes_counter); percpu_counter_destroy(&sbi->s_dirs_counter); percpu_counter_destroy(&sbi->s_dirtyclusters_counter); + percpu_counter_destroy(&sbi->s_sra_exceeded_retry_limit); percpu_free_rwsem(&sbi->s_writepages_rwsem); failed_mount5: ext4_ext_release(sb); @@ -5142,8 +5154,8 @@ failed_mount_wq: failed_mount3a: ext4_es_unregister_shrinker(sbi); failed_mount3: - del_timer_sync(&sbi->s_err_report); flush_work(&sbi->s_error_work); + del_timer_sync(&sbi->s_err_report); if (sbi->s_mmp_tsk) kthread_stop(sbi->s_mmp_tsk); failed_mount2: @@ -5709,7 +5721,7 @@ static int ext4_sync_fs(struct super_block *sb, int wait) needs_barrier = true; if (needs_barrier) { int err; - err = blkdev_issue_flush(sb->s_bdev, GFP_KERNEL); + err = blkdev_issue_flush(sb->s_bdev); if (!ret) ret = err; } @@ -6654,7 +6666,7 @@ static struct file_system_type ext4_fs_type = { .name = "ext4", .mount = ext4_mount, .kill_sb = kill_block_super, - .fs_flags = FS_REQUIRES_DEV, + .fs_flags = FS_REQUIRES_DEV | FS_ALLOW_IDMAP, }; MODULE_ALIAS_FS("ext4"); @@ -6667,7 +6679,6 @@ static int __init ext4_init_fs(void) ratelimit_state_init(&ext4_mount_msg_ratelimit, 30 * HZ, 64); ext4_li_info = NULL; - mutex_init(&ext4_li_mtx); /* Build-time check for flags consistency */ ext4_check_flag_values(); diff --git a/fs/ext4/sysfs.c b/fs/ext4/sysfs.c index 075aa3a19ff5..a3d08276d441 100644 --- a/fs/ext4/sysfs.c +++ b/fs/ext4/sysfs.c @@ -24,6 +24,7 @@ typedef enum { attr_session_write_kbytes, attr_lifetime_write_kbytes, attr_reserved_clusters, + attr_sra_exceeded_retry_limit, attr_inode_readahead, attr_trigger_test_error, attr_first_error_time, @@ -202,6 +203,7 @@ EXT4_ATTR_FUNC(delayed_allocation_blocks, 0444); EXT4_ATTR_FUNC(session_write_kbytes, 0444); EXT4_ATTR_FUNC(lifetime_write_kbytes, 0444); EXT4_ATTR_FUNC(reserved_clusters, 0644); +EXT4_ATTR_FUNC(sra_exceeded_retry_limit, 0444); EXT4_ATTR_OFFSET(inode_readahead_blks, 0644, inode_readahead, ext4_sb_info, s_inode_readahead_blks); @@ -251,6 +253,7 @@ static struct attribute *ext4_attrs[] = { ATTR_LIST(session_write_kbytes), ATTR_LIST(lifetime_write_kbytes), ATTR_LIST(reserved_clusters), + ATTR_LIST(sra_exceeded_retry_limit), ATTR_LIST(inode_readahead_blks), ATTR_LIST(inode_goal), ATTR_LIST(mb_stats), @@ -374,6 +377,10 @@ static ssize_t ext4_attr_show(struct kobject *kobj, return snprintf(buf, PAGE_SIZE, "%llu\n", (unsigned long long) atomic64_read(&sbi->s_resv_clusters)); + case attr_sra_exceeded_retry_limit: + return snprintf(buf, PAGE_SIZE, "%llu\n", + (unsigned long long) + percpu_counter_sum(&sbi->s_sra_exceeded_retry_limit)); case attr_inode_readahead: case attr_pointer_ui: if (!ptr) diff --git a/fs/ext4/verity.c b/fs/ext4/verity.c index 5b7ba8f71153..00e3cbde472e 100644 --- a/fs/ext4/verity.c +++ b/fs/ext4/verity.c @@ -201,55 +201,76 @@ static int ext4_end_enable_verity(struct file *filp, const void *desc, struct inode *inode = file_inode(filp); const int credits = 2; /* superblock and inode for ext4_orphan_del() */ handle_t *handle; + struct ext4_iloc iloc; int err = 0; - int err2; - if (desc != NULL) { - /* Succeeded; write the verity descriptor. */ - err = ext4_write_verity_descriptor(inode, desc, desc_size, - merkle_tree_size); - - /* Write all pages before clearing VERITY_IN_PROGRESS. */ - if (!err) - err = filemap_write_and_wait(inode->i_mapping); - } + /* + * If an error already occurred (which fs/verity/ signals by passing + * desc == NULL), then only clean-up is needed. + */ + if (desc == NULL) + goto cleanup; - /* If we failed, truncate anything we wrote past i_size. */ - if (desc == NULL || err) - ext4_truncate(inode); + /* Append the verity descriptor. */ + err = ext4_write_verity_descriptor(inode, desc, desc_size, + merkle_tree_size); + if (err) + goto cleanup; /* - * We must always clean up by clearing EXT4_STATE_VERITY_IN_PROGRESS and - * deleting the inode from the orphan list, even if something failed. - * If everything succeeded, we'll also set the verity bit in the same - * transaction. + * Write all pages (both data and verity metadata). Note that this must + * happen before clearing EXT4_STATE_VERITY_IN_PROGRESS; otherwise pages + * beyond i_size won't be written properly. For crash consistency, this + * also must happen before the verity inode flag gets persisted. */ + err = filemap_write_and_wait(inode->i_mapping); + if (err) + goto cleanup; - ext4_clear_inode_state(inode, EXT4_STATE_VERITY_IN_PROGRESS); + /* + * Finally, set the verity inode flag and remove the inode from the + * orphan list (in a single transaction). + */ handle = ext4_journal_start(inode, EXT4_HT_INODE, credits); if (IS_ERR(handle)) { - ext4_orphan_del(NULL, inode); - return PTR_ERR(handle); + err = PTR_ERR(handle); + goto cleanup; } - err2 = ext4_orphan_del(handle, inode); - if (err2) - goto out_stop; + err = ext4_orphan_del(handle, inode); + if (err) + goto stop_and_cleanup; - if (desc != NULL && !err) { - struct ext4_iloc iloc; + err = ext4_reserve_inode_write(handle, inode, &iloc); + if (err) + goto stop_and_cleanup; - err = ext4_reserve_inode_write(handle, inode, &iloc); - if (err) - goto out_stop; - ext4_set_inode_flag(inode, EXT4_INODE_VERITY); - ext4_set_inode_flags(inode, false); - err = ext4_mark_iloc_dirty(handle, inode, &iloc); - } -out_stop: + ext4_set_inode_flag(inode, EXT4_INODE_VERITY); + ext4_set_inode_flags(inode, false); + err = ext4_mark_iloc_dirty(handle, inode, &iloc); + if (err) + goto stop_and_cleanup; + + ext4_journal_stop(handle); + + ext4_clear_inode_state(inode, EXT4_STATE_VERITY_IN_PROGRESS); + return 0; + +stop_and_cleanup: ext4_journal_stop(handle); - return err ?: err2; +cleanup: + /* + * Verity failed to be enabled, so clean up by truncating any verity + * metadata that was written beyond i_size (both from cache and from + * disk), removing the inode from the orphan list (if it wasn't done + * already), and clearing EXT4_STATE_VERITY_IN_PROGRESS. + */ + truncate_inode_pages(inode->i_mapping, inode->i_size); + ext4_truncate(inode); + ext4_orphan_del(NULL, inode); + ext4_clear_inode_state(inode, EXT4_STATE_VERITY_IN_PROGRESS); + return err; } static int ext4_get_verity_descriptor_location(struct inode *inode, diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c index 372208500f4e..6c1018223c54 100644 --- a/fs/ext4/xattr.c +++ b/fs/ext4/xattr.c @@ -1462,6 +1462,9 @@ ext4_xattr_inode_cache_find(struct inode *inode, const void *value, if (!ce) return NULL; + WARN_ON_ONCE(ext4_handle_valid(journal_current_handle()) && + !(current->flags & PF_MEMALLOC_NOFS)); + ea_data = kvmalloc(value_len, GFP_KERNEL); if (!ea_data) { mb_cache_entry_put(ea_inode_cache, ce); @@ -2327,6 +2330,7 @@ ext4_xattr_set_handle(handle_t *handle, struct inode *inode, int name_index, error = -ENOSPC; goto cleanup; } + WARN_ON_ONCE(!(current->flags & PF_MEMALLOC_NOFS)); } error = ext4_reserve_inode_write(handle, inode, &is.iloc); @@ -2400,7 +2404,7 @@ retry_inode: * external inode if possible. */ if (ext4_has_feature_ea_inode(inode->i_sb) && - !i.in_inode) { + i.value_len && !i.in_inode) { i.in_inode = 1; goto retry_inode; } diff --git a/fs/ext4/xattr_hurd.c b/fs/ext4/xattr_hurd.c index 8cfa74a56361..c78df5790377 100644 --- a/fs/ext4/xattr_hurd.c +++ b/fs/ext4/xattr_hurd.c @@ -32,6 +32,7 @@ ext4_xattr_hurd_get(const struct xattr_handler *handler, static int ext4_xattr_hurd_set(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *unused, struct inode *inode, const char *name, const void *value, size_t size, int flags) diff --git a/fs/ext4/xattr_security.c b/fs/ext4/xattr_security.c index 197a9d8a15ef..8213f66f7b2d 100644 --- a/fs/ext4/xattr_security.c +++ b/fs/ext4/xattr_security.c @@ -23,6 +23,7 @@ ext4_xattr_security_get(const struct xattr_handler *handler, static int ext4_xattr_security_set(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *unused, struct inode *inode, const char *name, const void *value, size_t size, int flags) diff --git a/fs/ext4/xattr_trusted.c b/fs/ext4/xattr_trusted.c index e9389e5d75c3..7c21ffb26d25 100644 --- a/fs/ext4/xattr_trusted.c +++ b/fs/ext4/xattr_trusted.c @@ -30,6 +30,7 @@ ext4_xattr_trusted_get(const struct xattr_handler *handler, static int ext4_xattr_trusted_set(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *unused, struct inode *inode, const char *name, const void *value, size_t size, int flags) diff --git a/fs/ext4/xattr_user.c b/fs/ext4/xattr_user.c index d4546184b34b..2fe7ff0a479c 100644 --- a/fs/ext4/xattr_user.c +++ b/fs/ext4/xattr_user.c @@ -31,6 +31,7 @@ ext4_xattr_user_get(const struct xattr_handler *handler, static int ext4_xattr_user_set(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *unused, struct inode *inode, const char *name, const void *value, size_t size, int flags) diff --git a/fs/f2fs/Kconfig b/fs/f2fs/Kconfig index d13c5c6a9787..62e638a49bbf 100644 --- a/fs/f2fs/Kconfig +++ b/fs/f2fs/Kconfig @@ -76,16 +76,6 @@ config F2FS_CHECK_FS If you want to improve the performance, say N. -config F2FS_IO_TRACE - bool "F2FS IO tracer" - depends on F2FS_FS - depends on FUNCTION_TRACER - help - F2FS IO trace is based on a function trace, which gathers process - information and block IO patterns in the filesystem level. - - If unsure, say N. - config F2FS_FAULT_INJECTION bool "F2FS fault injection facility" depends on F2FS_FS @@ -119,6 +109,16 @@ config F2FS_FS_LZ4 help Support LZ4 compress algorithm, if unsure, say Y. +config F2FS_FS_LZ4HC + bool "LZ4HC compression support" + depends on F2FS_FS_COMPRESSION + depends on F2FS_FS_LZ4 + select LZ4HC_COMPRESS + default y + help + Support LZ4HC compress algorithm, LZ4HC has compatible on-disk + layout with LZ4, if unsure, say Y. + config F2FS_FS_ZSTD bool "ZSTD compression support" depends on F2FS_FS_COMPRESSION diff --git a/fs/f2fs/Makefile b/fs/f2fs/Makefile index ee7316b42f69..e5295746208b 100644 --- a/fs/f2fs/Makefile +++ b/fs/f2fs/Makefile @@ -7,6 +7,5 @@ f2fs-y += shrinker.o extent_cache.o sysfs.o f2fs-$(CONFIG_F2FS_STAT_FS) += debug.o f2fs-$(CONFIG_F2FS_FS_XATTR) += xattr.o f2fs-$(CONFIG_F2FS_FS_POSIX_ACL) += acl.o -f2fs-$(CONFIG_F2FS_IO_TRACE) += trace.o f2fs-$(CONFIG_FS_VERITY) += verity.o f2fs-$(CONFIG_F2FS_FS_COMPRESSION) += compress.o diff --git a/fs/f2fs/acl.c b/fs/f2fs/acl.c index 1e5e9b1136ee..965037a9c205 100644 --- a/fs/f2fs/acl.c +++ b/fs/f2fs/acl.c @@ -200,6 +200,27 @@ struct posix_acl *f2fs_get_acl(struct inode *inode, int type) return __f2fs_get_acl(inode, type, NULL); } +static int f2fs_acl_update_mode(struct inode *inode, umode_t *mode_p, + struct posix_acl **acl) +{ + umode_t mode = inode->i_mode; + int error; + + if (is_inode_flag_set(inode, FI_ACL_MODE)) + mode = F2FS_I(inode)->i_acl_mode; + + error = posix_acl_equiv_mode(*acl, &mode); + if (error < 0) + return error; + if (error == 0) + *acl = NULL; + if (!in_group_p(i_gid_into_mnt(&init_user_ns, inode)) && + !capable_wrt_inode_uidgid(&init_user_ns, inode, CAP_FSETID)) + mode &= ~S_ISGID; + *mode_p = mode; + return 0; +} + static int __f2fs_set_acl(struct inode *inode, int type, struct posix_acl *acl, struct page *ipage) { @@ -213,7 +234,7 @@ static int __f2fs_set_acl(struct inode *inode, int type, case ACL_TYPE_ACCESS: name_index = F2FS_XATTR_INDEX_POSIX_ACL_ACCESS; if (acl && !ipage) { - error = posix_acl_update_mode(inode, &mode, &acl); + error = f2fs_acl_update_mode(inode, &mode, &acl); if (error) return error; set_acl_inode(inode, mode); @@ -248,7 +269,8 @@ static int __f2fs_set_acl(struct inode *inode, int type, return error; } -int f2fs_set_acl(struct inode *inode, struct posix_acl *acl, int type) +int f2fs_set_acl(struct user_namespace *mnt_userns, struct inode *inode, + struct posix_acl *acl, int type) { if (unlikely(f2fs_cp_error(F2FS_I_SB(inode)))) return -EIO; diff --git a/fs/f2fs/acl.h b/fs/f2fs/acl.h index 124868c13f80..986fd1bc780b 100644 --- a/fs/f2fs/acl.h +++ b/fs/f2fs/acl.h @@ -34,7 +34,8 @@ struct f2fs_acl_header { #ifdef CONFIG_F2FS_FS_POSIX_ACL extern struct posix_acl *f2fs_get_acl(struct inode *, int); -extern int f2fs_set_acl(struct inode *, struct posix_acl *, int); +extern int f2fs_set_acl(struct user_namespace *, struct inode *, + struct posix_acl *, int); extern int f2fs_init_acl(struct inode *, struct inode *, struct page *, struct page *); #else diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c index 897edb7c951a..be5415a0dbbc 100644 --- a/fs/f2fs/checkpoint.c +++ b/fs/f2fs/checkpoint.c @@ -13,13 +13,15 @@ #include <linux/f2fs_fs.h> #include <linux/pagevec.h> #include <linux/swap.h> +#include <linux/kthread.h> #include "f2fs.h" #include "node.h" #include "segment.h" -#include "trace.h" #include <trace/events/f2fs.h> +#define DEFAULT_CHECKPOINT_IOPRIO (IOPRIO_PRIO_VALUE(IOPRIO_CLASS_BE, 3)) + static struct kmem_cache *ino_entry_slab; struct kmem_cache *f2fs_inode_entry_slab; @@ -290,7 +292,7 @@ void f2fs_ra_meta_pages_cond(struct f2fs_sb_info *sbi, pgoff_t index) f2fs_put_page(page, 0); if (readahead) - f2fs_ra_meta_pages(sbi, index, BIO_MAX_PAGES, META_POR, true); + f2fs_ra_meta_pages(sbi, index, BIO_MAX_VECS, META_POR, true); } static int __f2fs_write_meta_page(struct page *page, @@ -443,7 +445,6 @@ static int f2fs_set_meta_page_dirty(struct page *page) __set_page_dirty_nobuffers(page); inc_page_count(F2FS_P_SB(page), F2FS_DIRTY_META); f2fs_set_page_private(page, 0); - f2fs_trace_pid(page); return 1; } return 0; @@ -1017,7 +1018,6 @@ void f2fs_update_dirty_page(struct inode *inode, struct page *page) spin_unlock(&sbi->inode_lock[type]); f2fs_set_page_private(page, 0); - f2fs_trace_pid(page); } void f2fs_remove_dirty_inode(struct inode *inode) @@ -1707,3 +1707,174 @@ void f2fs_destroy_checkpoint_caches(void) kmem_cache_destroy(ino_entry_slab); kmem_cache_destroy(f2fs_inode_entry_slab); } + +static int __write_checkpoint_sync(struct f2fs_sb_info *sbi) +{ + struct cp_control cpc = { .reason = CP_SYNC, }; + int err; + + down_write(&sbi->gc_lock); + err = f2fs_write_checkpoint(sbi, &cpc); + up_write(&sbi->gc_lock); + + return err; +} + +static void __checkpoint_and_complete_reqs(struct f2fs_sb_info *sbi) +{ + struct ckpt_req_control *cprc = &sbi->cprc_info; + struct ckpt_req *req, *next; + struct llist_node *dispatch_list; + u64 sum_diff = 0, diff, count = 0; + int ret; + + dispatch_list = llist_del_all(&cprc->issue_list); + if (!dispatch_list) + return; + dispatch_list = llist_reverse_order(dispatch_list); + + ret = __write_checkpoint_sync(sbi); + atomic_inc(&cprc->issued_ckpt); + + llist_for_each_entry_safe(req, next, dispatch_list, llnode) { + diff = (u64)ktime_ms_delta(ktime_get(), req->queue_time); + req->ret = ret; + complete(&req->wait); + + sum_diff += diff; + count++; + } + atomic_sub(count, &cprc->queued_ckpt); + atomic_add(count, &cprc->total_ckpt); + + spin_lock(&cprc->stat_lock); + cprc->cur_time = (unsigned int)div64_u64(sum_diff, count); + if (cprc->peak_time < cprc->cur_time) + cprc->peak_time = cprc->cur_time; + spin_unlock(&cprc->stat_lock); +} + +static int issue_checkpoint_thread(void *data) +{ + struct f2fs_sb_info *sbi = data; + struct ckpt_req_control *cprc = &sbi->cprc_info; + wait_queue_head_t *q = &cprc->ckpt_wait_queue; +repeat: + if (kthread_should_stop()) + return 0; + + if (!llist_empty(&cprc->issue_list)) + __checkpoint_and_complete_reqs(sbi); + + wait_event_interruptible(*q, + kthread_should_stop() || !llist_empty(&cprc->issue_list)); + goto repeat; +} + +static void flush_remained_ckpt_reqs(struct f2fs_sb_info *sbi, + struct ckpt_req *wait_req) +{ + struct ckpt_req_control *cprc = &sbi->cprc_info; + + if (!llist_empty(&cprc->issue_list)) { + __checkpoint_and_complete_reqs(sbi); + } else { + /* already dispatched by issue_checkpoint_thread */ + if (wait_req) + wait_for_completion(&wait_req->wait); + } +} + +static void init_ckpt_req(struct ckpt_req *req) +{ + memset(req, 0, sizeof(struct ckpt_req)); + + init_completion(&req->wait); + req->queue_time = ktime_get(); +} + +int f2fs_issue_checkpoint(struct f2fs_sb_info *sbi) +{ + struct ckpt_req_control *cprc = &sbi->cprc_info; + struct ckpt_req req; + struct cp_control cpc; + + cpc.reason = __get_cp_reason(sbi); + if (!test_opt(sbi, MERGE_CHECKPOINT) || cpc.reason != CP_SYNC) { + int ret; + + down_write(&sbi->gc_lock); + ret = f2fs_write_checkpoint(sbi, &cpc); + up_write(&sbi->gc_lock); + + return ret; + } + + if (!cprc->f2fs_issue_ckpt) + return __write_checkpoint_sync(sbi); + + init_ckpt_req(&req); + + llist_add(&req.llnode, &cprc->issue_list); + atomic_inc(&cprc->queued_ckpt); + + /* update issue_list before we wake up issue_checkpoint thread */ + smp_mb(); + + if (waitqueue_active(&cprc->ckpt_wait_queue)) + wake_up(&cprc->ckpt_wait_queue); + + if (cprc->f2fs_issue_ckpt) + wait_for_completion(&req.wait); + else + flush_remained_ckpt_reqs(sbi, &req); + + return req.ret; +} + +int f2fs_start_ckpt_thread(struct f2fs_sb_info *sbi) +{ + dev_t dev = sbi->sb->s_bdev->bd_dev; + struct ckpt_req_control *cprc = &sbi->cprc_info; + + if (cprc->f2fs_issue_ckpt) + return 0; + + cprc->f2fs_issue_ckpt = kthread_run(issue_checkpoint_thread, sbi, + "f2fs_ckpt-%u:%u", MAJOR(dev), MINOR(dev)); + if (IS_ERR(cprc->f2fs_issue_ckpt)) { + cprc->f2fs_issue_ckpt = NULL; + return -ENOMEM; + } + + set_task_ioprio(cprc->f2fs_issue_ckpt, cprc->ckpt_thread_ioprio); + + return 0; +} + +void f2fs_stop_ckpt_thread(struct f2fs_sb_info *sbi) +{ + struct ckpt_req_control *cprc = &sbi->cprc_info; + + if (cprc->f2fs_issue_ckpt) { + struct task_struct *ckpt_task = cprc->f2fs_issue_ckpt; + + cprc->f2fs_issue_ckpt = NULL; + kthread_stop(ckpt_task); + + flush_remained_ckpt_reqs(sbi, NULL); + } +} + +void f2fs_init_ckpt_req_control(struct f2fs_sb_info *sbi) +{ + struct ckpt_req_control *cprc = &sbi->cprc_info; + + atomic_set(&cprc->issued_ckpt, 0); + atomic_set(&cprc->total_ckpt, 0); + atomic_set(&cprc->queued_ckpt, 0); + cprc->ckpt_thread_ioprio = DEFAULT_CHECKPOINT_IOPRIO; + init_waitqueue_head(&cprc->ckpt_wait_queue); + init_llist_head(&cprc->issue_list); + spin_lock_init(&cprc->stat_lock); +} diff --git a/fs/f2fs/compress.c b/fs/f2fs/compress.c index 4bcbacfe3325..77fa342de38f 100644 --- a/fs/f2fs/compress.c +++ b/fs/f2fs/compress.c @@ -252,8 +252,14 @@ static const struct f2fs_compress_ops f2fs_lzo_ops = { #ifdef CONFIG_F2FS_FS_LZ4 static int lz4_init_compress_ctx(struct compress_ctx *cc) { - cc->private = f2fs_kvmalloc(F2FS_I_SB(cc->inode), - LZ4_MEM_COMPRESS, GFP_NOFS); + unsigned int size = LZ4_MEM_COMPRESS; + +#ifdef CONFIG_F2FS_FS_LZ4HC + if (F2FS_I(cc->inode)->i_compress_flag >> COMPRESS_LEVEL_OFFSET) + size = LZ4HC_MEM_COMPRESS; +#endif + + cc->private = f2fs_kvmalloc(F2FS_I_SB(cc->inode), size, GFP_NOFS); if (!cc->private) return -ENOMEM; @@ -272,10 +278,34 @@ static void lz4_destroy_compress_ctx(struct compress_ctx *cc) cc->private = NULL; } +#ifdef CONFIG_F2FS_FS_LZ4HC +static int lz4hc_compress_pages(struct compress_ctx *cc) +{ + unsigned char level = F2FS_I(cc->inode)->i_compress_flag >> + COMPRESS_LEVEL_OFFSET; + int len; + + if (level) + len = LZ4_compress_HC(cc->rbuf, cc->cbuf->cdata, cc->rlen, + cc->clen, level, cc->private); + else + len = LZ4_compress_default(cc->rbuf, cc->cbuf->cdata, cc->rlen, + cc->clen, cc->private); + if (!len) + return -EAGAIN; + + cc->clen = len; + return 0; +} +#endif + static int lz4_compress_pages(struct compress_ctx *cc) { int len; +#ifdef CONFIG_F2FS_FS_LZ4HC + return lz4hc_compress_pages(cc); +#endif len = LZ4_compress_default(cc->rbuf, cc->cbuf->cdata, cc->rlen, cc->clen, cc->private); if (!len) @@ -325,8 +355,13 @@ static int zstd_init_compress_ctx(struct compress_ctx *cc) ZSTD_CStream *stream; void *workspace; unsigned int workspace_size; + unsigned char level = F2FS_I(cc->inode)->i_compress_flag >> + COMPRESS_LEVEL_OFFSET; - params = ZSTD_getParams(F2FS_ZSTD_DEFAULT_CLEVEL, cc->rlen, 0); + if (!level) + level = F2FS_ZSTD_DEFAULT_CLEVEL; + + params = ZSTD_getParams(level, cc->rlen, 0); workspace_size = ZSTD_CStreamWorkspaceBound(params.cParams); workspace = f2fs_kvmalloc(F2FS_I_SB(cc->inode), @@ -721,38 +756,27 @@ out: return ret; } -void f2fs_decompress_pages(struct bio *bio, struct page *page, bool verity) +static void f2fs_decompress_cluster(struct decompress_io_ctx *dic) { - struct decompress_io_ctx *dic = - (struct decompress_io_ctx *)page_private(page); struct f2fs_sb_info *sbi = F2FS_I_SB(dic->inode); - struct f2fs_inode_info *fi= F2FS_I(dic->inode); + struct f2fs_inode_info *fi = F2FS_I(dic->inode); const struct f2fs_compress_ops *cops = f2fs_cops[fi->i_compress_algorithm]; int ret; int i; - dec_page_count(sbi, F2FS_RD_DATA); - - if (bio->bi_status || PageError(page)) - dic->failed = true; - - if (atomic_dec_return(&dic->pending_pages)) - return; - trace_f2fs_decompress_pages_start(dic->inode, dic->cluster_idx, dic->cluster_size, fi->i_compress_algorithm); - /* submit partial compressed pages */ if (dic->failed) { ret = -EIO; - goto out_free_dic; + goto out_end_io; } dic->tpages = page_array_alloc(dic->inode, dic->cluster_size); if (!dic->tpages) { ret = -ENOMEM; - goto out_free_dic; + goto out_end_io; } for (i = 0; i < dic->cluster_size; i++) { @@ -764,20 +788,20 @@ void f2fs_decompress_pages(struct bio *bio, struct page *page, bool verity) dic->tpages[i] = f2fs_compress_alloc_page(); if (!dic->tpages[i]) { ret = -ENOMEM; - goto out_free_dic; + goto out_end_io; } } if (cops->init_decompress_ctx) { ret = cops->init_decompress_ctx(dic); if (ret) - goto out_free_dic; + goto out_end_io; } dic->rbuf = f2fs_vmap(dic->tpages, dic->cluster_size); if (!dic->rbuf) { ret = -ENOMEM; - goto destroy_decompress_ctx; + goto out_destroy_decompress_ctx; } dic->cbuf = f2fs_vmap(dic->cpages, dic->nr_cpages); @@ -816,18 +840,34 @@ out_vunmap_cbuf: vm_unmap_ram(dic->cbuf, dic->nr_cpages); out_vunmap_rbuf: vm_unmap_ram(dic->rbuf, dic->cluster_size); -destroy_decompress_ctx: +out_destroy_decompress_ctx: if (cops->destroy_decompress_ctx) cops->destroy_decompress_ctx(dic); -out_free_dic: - if (!verity) - f2fs_decompress_end_io(dic->rpages, dic->cluster_size, - ret, false); - +out_end_io: trace_f2fs_decompress_pages_end(dic->inode, dic->cluster_idx, dic->clen, ret); - if (!verity) - f2fs_free_dic(dic); + f2fs_decompress_end_io(dic, ret); +} + +/* + * This is called when a page of a compressed cluster has been read from disk + * (or failed to be read from disk). It checks whether this page was the last + * page being waited on in the cluster, and if so, it decompresses the cluster + * (or in the case of a failure, cleans up without actually decompressing). + */ +void f2fs_end_read_compressed_page(struct page *page, bool failed) +{ + struct decompress_io_ctx *dic = + (struct decompress_io_ctx *)page_private(page); + struct f2fs_sb_info *sbi = F2FS_I_SB(dic->inode); + + dec_page_count(sbi, F2FS_RD_DATA); + + if (failed) + WRITE_ONCE(dic->failed, true); + + if (atomic_dec_and_test(&dic->remaining_pages)) + f2fs_decompress_cluster(dic); } static bool is_page_in_cluster(struct compress_ctx *cc, pgoff_t index) @@ -1415,7 +1455,7 @@ retry_write: ret = f2fs_write_single_data_page(cc->rpages[i], &_submitted, NULL, NULL, wbc, io_type, - compr_blocks); + compr_blocks, false); if (ret) { if (ret == AOP_WRITEPAGE_ACTIVATE) { unlock_page(cc->rpages[i]); @@ -1450,6 +1490,9 @@ retry_write: *submitted += _submitted; } + + f2fs_balance_fs(F2FS_M_SB(mapping), true); + return 0; out_err: for (++i; i < cc->cluster_size; i++) { @@ -1494,6 +1537,8 @@ destroy_out: return err; } +static void f2fs_free_dic(struct decompress_io_ctx *dic); + struct decompress_io_ctx *f2fs_alloc_dic(struct compress_ctx *cc) { struct decompress_io_ctx *dic; @@ -1512,12 +1557,14 @@ struct decompress_io_ctx *f2fs_alloc_dic(struct compress_ctx *cc) dic->magic = F2FS_COMPRESSED_PAGE_MAGIC; dic->inode = cc->inode; - atomic_set(&dic->pending_pages, cc->nr_cpages); + atomic_set(&dic->remaining_pages, cc->nr_cpages); dic->cluster_idx = cc->cluster_idx; dic->cluster_size = cc->cluster_size; dic->log_cluster_size = cc->log_cluster_size; dic->nr_cpages = cc->nr_cpages; + refcount_set(&dic->refcnt, 1); dic->failed = false; + dic->need_verity = f2fs_need_verity(cc->inode, start_idx); for (i = 0; i < dic->cluster_size; i++) dic->rpages[i] = cc->rpages[i]; @@ -1546,7 +1593,7 @@ out_free: return ERR_PTR(-ENOMEM); } -void f2fs_free_dic(struct decompress_io_ctx *dic) +static void f2fs_free_dic(struct decompress_io_ctx *dic) { int i; @@ -1574,30 +1621,88 @@ void f2fs_free_dic(struct decompress_io_ctx *dic) kmem_cache_free(dic_entry_slab, dic); } -void f2fs_decompress_end_io(struct page **rpages, - unsigned int cluster_size, bool err, bool verity) +static void f2fs_put_dic(struct decompress_io_ctx *dic) +{ + if (refcount_dec_and_test(&dic->refcnt)) + f2fs_free_dic(dic); +} + +/* + * Update and unlock the cluster's pagecache pages, and release the reference to + * the decompress_io_ctx that was being held for I/O completion. + */ +static void __f2fs_decompress_end_io(struct decompress_io_ctx *dic, bool failed) { int i; - for (i = 0; i < cluster_size; i++) { - struct page *rpage = rpages[i]; + for (i = 0; i < dic->cluster_size; i++) { + struct page *rpage = dic->rpages[i]; if (!rpage) continue; - if (err || PageError(rpage)) - goto clear_uptodate; - - if (!verity || fsverity_verify_page(rpage)) { + /* PG_error was set if verity failed. */ + if (failed || PageError(rpage)) { + ClearPageUptodate(rpage); + /* will re-read again later */ + ClearPageError(rpage); + } else { SetPageUptodate(rpage); - goto unlock; } -clear_uptodate: - ClearPageUptodate(rpage); - ClearPageError(rpage); -unlock: unlock_page(rpage); } + + f2fs_put_dic(dic); +} + +static void f2fs_verify_cluster(struct work_struct *work) +{ + struct decompress_io_ctx *dic = + container_of(work, struct decompress_io_ctx, verity_work); + int i; + + /* Verify the cluster's decompressed pages with fs-verity. */ + for (i = 0; i < dic->cluster_size; i++) { + struct page *rpage = dic->rpages[i]; + + if (rpage && !fsverity_verify_page(rpage)) + SetPageError(rpage); + } + + __f2fs_decompress_end_io(dic, false); +} + +/* + * This is called when a compressed cluster has been decompressed + * (or failed to be read and/or decompressed). + */ +void f2fs_decompress_end_io(struct decompress_io_ctx *dic, bool failed) +{ + if (!failed && dic->need_verity) { + /* + * Note that to avoid deadlocks, the verity work can't be done + * on the decompression workqueue. This is because verifying + * the data pages can involve reading metadata pages from the + * file, and these metadata pages may be compressed. + */ + INIT_WORK(&dic->verity_work, f2fs_verify_cluster); + fsverity_enqueue_verify_work(&dic->verity_work); + } else { + __f2fs_decompress_end_io(dic, failed); + } +} + +/* + * Put a reference to a compressed page's decompress_io_ctx. + * + * This is called when the page is no longer needed and can be freed. + */ +void f2fs_put_page_dic(struct page *page) +{ + struct decompress_io_ctx *dic = + (struct decompress_io_ctx *)page_private(page); + + f2fs_put_dic(dic); } int f2fs_init_page_array_cache(struct f2fs_sb_info *sbi) diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index aa34d620bec9..4e5257c763d0 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -25,7 +25,6 @@ #include "f2fs.h" #include "node.h" #include "segment.h" -#include "trace.h" #include <trace/events/f2fs.h> #define NUM_PREALLOC_POST_READ_CTXS 128 @@ -50,27 +49,6 @@ void f2fs_destroy_bioset(void) bioset_exit(&f2fs_bioset); } -static inline struct bio *__f2fs_bio_alloc(gfp_t gfp_mask, - unsigned int nr_iovecs) -{ - return bio_alloc_bioset(gfp_mask, nr_iovecs, &f2fs_bioset); -} - -struct bio *f2fs_bio_alloc(struct f2fs_sb_info *sbi, int npages, bool noio) -{ - if (noio) { - /* No failure on bio allocation */ - return __f2fs_bio_alloc(GFP_NOIO, npages); - } - - if (time_to_inject(sbi, FAULT_ALLOC_BIO)) { - f2fs_show_injection_info(sbi, FAULT_ALLOC_BIO); - return NULL; - } - - return __f2fs_bio_alloc(GFP_KERNEL, npages); -} - static bool __is_cp_guaranteed(struct page *page) { struct address_space *mapping = page->mapping; @@ -115,10 +93,21 @@ static enum count_type __read_io_type(struct page *page) /* postprocessing steps for read bios */ enum bio_post_read_step { - STEP_DECRYPT, - STEP_DECOMPRESS_NOWQ, /* handle normal cluster data inplace */ - STEP_DECOMPRESS, /* handle compressed cluster data in workqueue */ - STEP_VERITY, +#ifdef CONFIG_FS_ENCRYPTION + STEP_DECRYPT = 1 << 0, +#else + STEP_DECRYPT = 0, /* compile out the decryption-related code */ +#endif +#ifdef CONFIG_F2FS_FS_COMPRESSION + STEP_DECOMPRESS = 1 << 1, +#else + STEP_DECOMPRESS = 0, /* compile out the decompression-related code */ +#endif +#ifdef CONFIG_FS_VERITY + STEP_VERITY = 1 << 2, +#else + STEP_VERITY = 0, /* compile out the verity-related code */ +#endif }; struct bio_post_read_ctx { @@ -128,25 +117,26 @@ struct bio_post_read_ctx { unsigned int enabled_steps; }; -static void __read_end_io(struct bio *bio, bool compr, bool verity) +static void f2fs_finish_read_bio(struct bio *bio) { - struct page *page; struct bio_vec *bv; struct bvec_iter_all iter_all; + /* + * Update and unlock the bio's pagecache pages, and put the + * decompression context for any compressed pages. + */ bio_for_each_segment_all(bv, bio, iter_all) { - page = bv->bv_page; + struct page *page = bv->bv_page; -#ifdef CONFIG_F2FS_FS_COMPRESSION - if (compr && f2fs_is_compressed_page(page)) { - f2fs_decompress_pages(bio, page, verity); + if (f2fs_is_compressed_page(page)) { + if (bio->bi_status) + f2fs_end_read_compressed_page(page, true); + f2fs_put_page_dic(page); continue; } - if (verity) - continue; -#endif - /* PG_error was set if any post_read step failed */ + /* PG_error was set if decryption or verity failed. */ if (bio->bi_status || PageError(page)) { ClearPageUptodate(page); /* will re-read again later */ @@ -157,181 +147,141 @@ static void __read_end_io(struct bio *bio, bool compr, bool verity) dec_page_count(F2FS_P_SB(page), __read_io_type(page)); unlock_page(page); } -} - -static void f2fs_release_read_bio(struct bio *bio); -static void __f2fs_read_end_io(struct bio *bio, bool compr, bool verity) -{ - if (!compr) - __read_end_io(bio, false, verity); - f2fs_release_read_bio(bio); -} - -static void f2fs_decompress_bio(struct bio *bio, bool verity) -{ - __read_end_io(bio, true, verity); -} - -static void bio_post_read_processing(struct bio_post_read_ctx *ctx); - -static void f2fs_decrypt_work(struct bio_post_read_ctx *ctx) -{ - fscrypt_decrypt_bio(ctx->bio); -} - -static void f2fs_decompress_work(struct bio_post_read_ctx *ctx) -{ - f2fs_decompress_bio(ctx->bio, ctx->enabled_steps & (1 << STEP_VERITY)); -} - -#ifdef CONFIG_F2FS_FS_COMPRESSION -static void f2fs_verify_pages(struct page **rpages, unsigned int cluster_size) -{ - f2fs_decompress_end_io(rpages, cluster_size, false, true); -} - -static void f2fs_verify_bio(struct bio *bio) -{ - struct bio_vec *bv; - struct bvec_iter_all iter_all; - bio_for_each_segment_all(bv, bio, iter_all) { - struct page *page = bv->bv_page; - struct decompress_io_ctx *dic; - - dic = (struct decompress_io_ctx *)page_private(page); - - if (dic) { - if (atomic_dec_return(&dic->verity_pages)) - continue; - f2fs_verify_pages(dic->rpages, - dic->cluster_size); - f2fs_free_dic(dic); - continue; - } - - if (bio->bi_status || PageError(page)) - goto clear_uptodate; - - if (fsverity_verify_page(page)) { - SetPageUptodate(page); - goto unlock; - } -clear_uptodate: - ClearPageUptodate(page); - ClearPageError(page); -unlock: - dec_page_count(F2FS_P_SB(page), __read_io_type(page)); - unlock_page(page); - } + if (bio->bi_private) + mempool_free(bio->bi_private, bio_post_read_ctx_pool); + bio_put(bio); } -#endif -static void f2fs_verity_work(struct work_struct *work) +static void f2fs_verify_bio(struct work_struct *work) { struct bio_post_read_ctx *ctx = container_of(work, struct bio_post_read_ctx, work); struct bio *bio = ctx->bio; -#ifdef CONFIG_F2FS_FS_COMPRESSION - unsigned int enabled_steps = ctx->enabled_steps; -#endif + bool may_have_compressed_pages = (ctx->enabled_steps & STEP_DECOMPRESS); /* * fsverity_verify_bio() may call readpages() again, and while verity - * will be disabled for this, decryption may still be needed, resulting - * in another bio_post_read_ctx being allocated. So to prevent - * deadlocks we need to release the current ctx to the mempool first. - * This assumes that verity is the last post-read step. + * will be disabled for this, decryption and/or decompression may still + * be needed, resulting in another bio_post_read_ctx being allocated. + * So to prevent deadlocks we need to release the current ctx to the + * mempool first. This assumes that verity is the last post-read step. */ mempool_free(ctx, bio_post_read_ctx_pool); bio->bi_private = NULL; -#ifdef CONFIG_F2FS_FS_COMPRESSION - /* previous step is decompression */ - if (enabled_steps & (1 << STEP_DECOMPRESS)) { - f2fs_verify_bio(bio); - f2fs_release_read_bio(bio); - return; + /* + * Verify the bio's pages with fs-verity. Exclude compressed pages, + * as those were handled separately by f2fs_end_read_compressed_page(). + */ + if (may_have_compressed_pages) { + struct bio_vec *bv; + struct bvec_iter_all iter_all; + + bio_for_each_segment_all(bv, bio, iter_all) { + struct page *page = bv->bv_page; + + if (!f2fs_is_compressed_page(page) && + !PageError(page) && !fsverity_verify_page(page)) + SetPageError(page); + } + } else { + fsverity_verify_bio(bio); } -#endif - fsverity_verify_bio(bio); - __f2fs_read_end_io(bio, false, false); + f2fs_finish_read_bio(bio); } -static void f2fs_post_read_work(struct work_struct *work) +/* + * If the bio's data needs to be verified with fs-verity, then enqueue the + * verity work for the bio. Otherwise finish the bio now. + * + * Note that to avoid deadlocks, the verity work can't be done on the + * decryption/decompression workqueue. This is because verifying the data pages + * can involve reading verity metadata pages from the file, and these verity + * metadata pages may be encrypted and/or compressed. + */ +static void f2fs_verify_and_finish_bio(struct bio *bio) { - struct bio_post_read_ctx *ctx = - container_of(work, struct bio_post_read_ctx, work); + struct bio_post_read_ctx *ctx = bio->bi_private; - if (ctx->enabled_steps & (1 << STEP_DECRYPT)) - f2fs_decrypt_work(ctx); - - if (ctx->enabled_steps & (1 << STEP_DECOMPRESS)) - f2fs_decompress_work(ctx); - - if (ctx->enabled_steps & (1 << STEP_VERITY)) { - INIT_WORK(&ctx->work, f2fs_verity_work); + if (ctx && (ctx->enabled_steps & STEP_VERITY)) { + INIT_WORK(&ctx->work, f2fs_verify_bio); fsverity_enqueue_verify_work(&ctx->work); - return; + } else { + f2fs_finish_read_bio(bio); } - - __f2fs_read_end_io(ctx->bio, - ctx->enabled_steps & (1 << STEP_DECOMPRESS), false); } -static void f2fs_enqueue_post_read_work(struct f2fs_sb_info *sbi, - struct work_struct *work) -{ - queue_work(sbi->post_read_wq, work); -} - -static void bio_post_read_processing(struct bio_post_read_ctx *ctx) +/* + * Handle STEP_DECOMPRESS by decompressing any compressed clusters whose last + * remaining page was read by @ctx->bio. + * + * Note that a bio may span clusters (even a mix of compressed and uncompressed + * clusters) or be for just part of a cluster. STEP_DECOMPRESS just indicates + * that the bio includes at least one compressed page. The actual decompression + * is done on a per-cluster basis, not a per-bio basis. + */ +static void f2fs_handle_step_decompress(struct bio_post_read_ctx *ctx) { - /* - * We use different work queues for decryption and for verity because - * verity may require reading metadata pages that need decryption, and - * we shouldn't recurse to the same workqueue. - */ + struct bio_vec *bv; + struct bvec_iter_all iter_all; + bool all_compressed = true; - if (ctx->enabled_steps & (1 << STEP_DECRYPT) || - ctx->enabled_steps & (1 << STEP_DECOMPRESS)) { - INIT_WORK(&ctx->work, f2fs_post_read_work); - f2fs_enqueue_post_read_work(ctx->sbi, &ctx->work); - return; - } + bio_for_each_segment_all(bv, ctx->bio, iter_all) { + struct page *page = bv->bv_page; - if (ctx->enabled_steps & (1 << STEP_VERITY)) { - INIT_WORK(&ctx->work, f2fs_verity_work); - fsverity_enqueue_verify_work(&ctx->work); - return; + /* PG_error was set if decryption failed. */ + if (f2fs_is_compressed_page(page)) + f2fs_end_read_compressed_page(page, PageError(page)); + else + all_compressed = false; } - __f2fs_read_end_io(ctx->bio, false, false); + /* + * Optimization: if all the bio's pages are compressed, then scheduling + * the per-bio verity work is unnecessary, as verity will be fully + * handled at the compression cluster level. + */ + if (all_compressed) + ctx->enabled_steps &= ~STEP_VERITY; } -static bool f2fs_bio_post_read_required(struct bio *bio) +static void f2fs_post_read_work(struct work_struct *work) { - return bio->bi_private; + struct bio_post_read_ctx *ctx = + container_of(work, struct bio_post_read_ctx, work); + + if (ctx->enabled_steps & STEP_DECRYPT) + fscrypt_decrypt_bio(ctx->bio); + + if (ctx->enabled_steps & STEP_DECOMPRESS) + f2fs_handle_step_decompress(ctx); + + f2fs_verify_and_finish_bio(ctx->bio); } static void f2fs_read_end_io(struct bio *bio) { struct f2fs_sb_info *sbi = F2FS_P_SB(bio_first_page_all(bio)); + struct bio_post_read_ctx *ctx = bio->bi_private; if (time_to_inject(sbi, FAULT_READ_IO)) { f2fs_show_injection_info(sbi, FAULT_READ_IO); bio->bi_status = BLK_STS_IOERR; } - if (f2fs_bio_post_read_required(bio)) { - struct bio_post_read_ctx *ctx = bio->bi_private; - - bio_post_read_processing(ctx); + if (bio->bi_status) { + f2fs_finish_read_bio(bio); return; } - __f2fs_read_end_io(bio, false, false); + if (ctx && (ctx->enabled_steps & (STEP_DECRYPT | STEP_DECOMPRESS))) { + INIT_WORK(&ctx->work, f2fs_post_read_work); + queue_work(ctx->sbi->post_read_wq, &ctx->work); + } else { + f2fs_verify_and_finish_bio(bio); + } } static void f2fs_write_end_io(struct bio *bio) @@ -427,22 +377,12 @@ int f2fs_target_device_index(struct f2fs_sb_info *sbi, block_t blkaddr) return 0; } -/* - * Return true, if pre_bio's bdev is same as its target device. - */ -static bool __same_bdev(struct f2fs_sb_info *sbi, - block_t blk_addr, struct bio *bio) -{ - struct block_device *b = f2fs_target_device(sbi, blk_addr, NULL); - return bio->bi_disk == b->bd_disk && bio->bi_partno == b->bd_partno; -} - static struct bio *__bio_alloc(struct f2fs_io_info *fio, int npages) { struct f2fs_sb_info *sbi = fio->sbi; struct bio *bio; - bio = f2fs_bio_alloc(sbi, npages, true); + bio = bio_alloc_bioset(GFP_NOIO, npages, &f2fs_bioset); f2fs_target_device(sbi, fio->new_blkaddr, bio); if (is_read_io(fio->op)) { @@ -499,7 +439,7 @@ static inline void __submit_bio(struct f2fs_sb_info *sbi, if (f2fs_lfs_mode(sbi) && current->plug) blk_finish_plug(current->plug); - if (F2FS_IO_ALIGNED(sbi)) + if (!F2FS_IO_ALIGNED(sbi)) goto submit_io; start = bio->bi_iter.bi_size >> F2FS_BLKSIZE_BITS; @@ -707,7 +647,6 @@ int f2fs_submit_page_bio(struct f2fs_io_info *fio) return -EFSCORRUPTED; trace_f2fs_submit_page_bio(page, fio); - f2fs_trace_ios(fio, 0); /* Allocate a new bio */ bio = __bio_alloc(fio, 1); @@ -741,7 +680,7 @@ static bool page_is_mergeable(struct f2fs_sb_info *sbi, struct bio *bio, return false; if (last_blkaddr + 1 != cur_blkaddr) return false; - return __same_bdev(sbi, cur_blkaddr, bio); + return bio->bi_bdev == f2fs_target_device(sbi, cur_blkaddr, NULL); } static bool io_type_is_mergeable(struct f2fs_bio_info *io, @@ -912,14 +851,13 @@ int f2fs_merge_page_bio(struct f2fs_io_info *fio) return -EFSCORRUPTED; trace_f2fs_submit_page_bio(page, fio); - f2fs_trace_ios(fio, 0); if (bio && !page_is_mergeable(fio->sbi, bio, *fio->last_block, fio->new_blkaddr)) f2fs_submit_merged_ipu_write(fio->sbi, &bio, NULL); alloc_new: if (!bio) { - bio = __bio_alloc(fio, BIO_MAX_PAGES); + bio = __bio_alloc(fio, BIO_MAX_VECS); __attach_io_flag(fio); f2fs_set_bio_crypt_ctx(bio, fio->page->mapping->host, fio->page->index, fio, GFP_NOIO); @@ -994,7 +932,7 @@ alloc_new: fio->retry = true; goto skip; } - io->bio = __bio_alloc(fio, BIO_MAX_PAGES); + io->bio = __bio_alloc(fio, BIO_MAX_VECS); f2fs_set_bio_crypt_ctx(io->bio, fio->page->mapping->host, bio_page->index, fio, GFP_NOIO); io->fio = *fio; @@ -1009,7 +947,6 @@ alloc_new: wbc_account_cgroup_owner(fio->io_wbc, bio_page, PAGE_SIZE); io->last_block_in_bio = fio->new_blkaddr; - f2fs_trace_ios(fio, 0); trace_f2fs_submit_page_write(fio->page, fio); skip: @@ -1022,24 +959,17 @@ out: up_write(&io->io_rwsem); } -static inline bool f2fs_need_verity(const struct inode *inode, pgoff_t idx) -{ - return fsverity_active(inode) && - idx < DIV_ROUND_UP(inode->i_size, PAGE_SIZE); -} - static struct bio *f2fs_grab_read_bio(struct inode *inode, block_t blkaddr, unsigned nr_pages, unsigned op_flag, - pgoff_t first_idx, bool for_write, - bool for_verity) + pgoff_t first_idx, bool for_write) { struct f2fs_sb_info *sbi = F2FS_I_SB(inode); struct bio *bio; struct bio_post_read_ctx *ctx; unsigned int post_read_steps = 0; - bio = f2fs_bio_alloc(sbi, min_t(int, nr_pages, BIO_MAX_PAGES), - for_write); + bio = bio_alloc_bioset(for_write ? GFP_NOIO : GFP_KERNEL, + bio_max_segs(nr_pages), &f2fs_bioset); if (!bio) return ERR_PTR(-ENOMEM); @@ -1050,13 +980,19 @@ static struct bio *f2fs_grab_read_bio(struct inode *inode, block_t blkaddr, bio_set_op_attrs(bio, REQ_OP_READ, op_flag); if (fscrypt_inode_uses_fs_layer_crypto(inode)) - post_read_steps |= 1 << STEP_DECRYPT; - if (f2fs_compressed_file(inode)) - post_read_steps |= 1 << STEP_DECOMPRESS_NOWQ; - if (for_verity && f2fs_need_verity(inode, first_idx)) - post_read_steps |= 1 << STEP_VERITY; + post_read_steps |= STEP_DECRYPT; + + if (f2fs_need_verity(inode, first_idx)) + post_read_steps |= STEP_VERITY; + + /* + * STEP_DECOMPRESS is handled specially, since a compressed file might + * contain both compressed and uncompressed clusters. We'll allocate a + * bio_post_read_ctx if the file is compressed, but the caller is + * responsible for enabling STEP_DECOMPRESS if it's actually needed. + */ - if (post_read_steps) { + if (post_read_steps || f2fs_compressed_file(inode)) { /* Due to the mempool, this never fails. */ ctx = mempool_alloc(bio_post_read_ctx_pool, GFP_NOFS); ctx->bio = bio; @@ -1068,13 +1004,6 @@ static struct bio *f2fs_grab_read_bio(struct inode *inode, block_t blkaddr, return bio; } -static void f2fs_release_read_bio(struct bio *bio) -{ - if (bio->bi_private) - mempool_free(bio->bi_private, bio_post_read_ctx_pool); - bio_put(bio); -} - /* This can handle encryption stuffs */ static int f2fs_submit_page_read(struct inode *inode, struct page *page, block_t blkaddr, int op_flags, bool for_write) @@ -1083,7 +1012,7 @@ static int f2fs_submit_page_read(struct inode *inode, struct page *page, struct bio *bio; bio = f2fs_grab_read_bio(inode, blkaddr, 1, op_flags, - page->index, for_write, true); + page->index, for_write); if (IS_ERR(bio)) return PTR_ERR(bio); @@ -1964,6 +1893,7 @@ next: } if (size) { + flags |= FIEMAP_EXTENT_MERGED; if (IS_ENCRYPTED(inode)) flags |= FIEMAP_EXTENT_DATA_ENCRYPTED; @@ -2121,7 +2051,7 @@ submit_and_realloc: if (bio == NULL) { bio = f2fs_grab_read_bio(inode, block_nr, nr_pages, is_readahead ? REQ_RAHEAD : 0, page->index, - false, true); + false); if (IS_ERR(bio)) { ret = PTR_ERR(bio); bio = NULL; @@ -2167,8 +2097,6 @@ int f2fs_read_multi_pages(struct compress_ctx *cc, struct bio **bio_ret, sector_t last_block_in_file; const unsigned blocksize = blks_to_bytes(inode, 1); struct decompress_io_ctx *dic = NULL; - struct bio_post_read_ctx *ctx; - bool for_verity = false; int i; int ret = 0; @@ -2234,29 +2162,10 @@ int f2fs_read_multi_pages(struct compress_ctx *cc, struct bio **bio_ret, goto out_put_dnode; } - /* - * It's possible to enable fsverity on the fly when handling a cluster, - * which requires complicated error handling. Instead of adding more - * complexity, let's give a rule where end_io post-processes fsverity - * per cluster. In order to do that, we need to submit bio, if previous - * bio sets a different post-process policy. - */ - if (fsverity_active(cc->inode)) { - atomic_set(&dic->verity_pages, cc->nr_cpages); - for_verity = true; - - if (bio) { - ctx = bio->bi_private; - if (!(ctx->enabled_steps & (1 << STEP_VERITY))) { - __submit_bio(sbi, bio, DATA); - bio = NULL; - } - } - } - for (i = 0; i < dic->nr_cpages; i++) { struct page *page = dic->cpages[i]; block_t blkaddr; + struct bio_post_read_ctx *ctx; blkaddr = data_blkaddr(dn.inode, dn.node_page, dn.ofs_in_node + i + 1); @@ -2272,31 +2181,10 @@ submit_and_realloc: if (!bio) { bio = f2fs_grab_read_bio(inode, blkaddr, nr_pages, is_readahead ? REQ_RAHEAD : 0, - page->index, for_write, for_verity); + page->index, for_write); if (IS_ERR(bio)) { - unsigned int remained = dic->nr_cpages - i; - bool release = false; - ret = PTR_ERR(bio); - dic->failed = true; - - if (for_verity) { - if (!atomic_sub_return(remained, - &dic->verity_pages)) - release = true; - } else { - if (!atomic_sub_return(remained, - &dic->pending_pages)) - release = true; - } - - if (release) { - f2fs_decompress_end_io(dic->rpages, - cc->cluster_size, true, - false); - f2fs_free_dic(dic); - } - + f2fs_decompress_end_io(dic, ret); f2fs_put_dnode(&dn); *bio_ret = NULL; return ret; @@ -2308,10 +2196,9 @@ submit_and_realloc: if (bio_add_page(bio, page, blocksize, 0) < blocksize) goto submit_and_realloc; - /* tag STEP_DECOMPRESS to handle IO in wq */ ctx = bio->bi_private; - if (!(ctx->enabled_steps & (1 << STEP_DECOMPRESS))) - ctx->enabled_steps |= 1 << STEP_DECOMPRESS; + ctx->enabled_steps |= STEP_DECOMPRESS; + refcount_inc(&dic->refcnt); inc_page_count(sbi, F2FS_RD_DATA); f2fs_update_iostat(sbi, FS_DATA_READ_IO, F2FS_BLKSIZE); @@ -2328,7 +2215,13 @@ submit_and_realloc: out_put_dnode: f2fs_put_dnode(&dn); out: - f2fs_decompress_end_io(cc->rpages, cc->cluster_size, true, false); + for (i = 0; i < cc->cluster_size; i++) { + if (cc->rpages[i]) { + ClearPageUptodate(cc->rpages[i]); + ClearPageError(cc->rpages[i]); + unlock_page(cc->rpages[i]); + } + } *bio_ret = bio; return ret; } @@ -2337,11 +2230,6 @@ out: /* * This function was originally taken from fs/mpage.c, and customized for f2fs. * Major change was from block_size == page_size in f2fs by default. - * - * Note that the aops->readpages() function is ONLY used for read-ahead. If - * this function ever deviates from doing just read-ahead, it should either - * use ->readpage() or do the necessary surgery to decouple ->readpages() - * from read-ahead. */ static int f2fs_mpage_readpages(struct inode *inode, struct readahead_control *rac, struct page *page) @@ -2364,7 +2252,6 @@ static int f2fs_mpage_readpages(struct inode *inode, unsigned nr_pages = rac ? readahead_count(rac) : 1; unsigned max_nr_pages = nr_pages; int ret = 0; - bool drop_ra = false; map.m_pblk = 0; map.m_lblk = 0; @@ -2375,26 +2262,10 @@ static int f2fs_mpage_readpages(struct inode *inode, map.m_seg_type = NO_CHECK_TYPE; map.m_may_create = false; - /* - * Two readahead threads for same address range can cause race condition - * which fragments sequential read IOs. So let's avoid each other. - */ - if (rac && readahead_count(rac)) { - if (READ_ONCE(F2FS_I(inode)->ra_offset) == readahead_index(rac)) - drop_ra = true; - else - WRITE_ONCE(F2FS_I(inode)->ra_offset, - readahead_index(rac)); - } - for (; nr_pages; nr_pages--) { if (rac) { page = readahead_page(rac); prefetchw(&page->flags); - if (drop_ra) { - f2fs_put_page(page, 1); - continue; - } } #ifdef CONFIG_F2FS_FS_COMPRESSION @@ -2457,9 +2328,6 @@ next_page: } if (bio) __submit_bio(F2FS_I_SB(inode), bio, DATA); - - if (rac && readahead_count(rac) && !drop_ra) - WRITE_ONCE(F2FS_I(inode)->ra_offset, -1); return ret; } @@ -2743,7 +2611,8 @@ int f2fs_write_single_data_page(struct page *page, int *submitted, sector_t *last_block, struct writeback_control *wbc, enum iostat_type io_type, - int compr_blocks) + int compr_blocks, + bool allow_balance) { struct inode *inode = page->mapping->host; struct f2fs_sb_info *sbi = F2FS_I_SB(inode); @@ -2881,7 +2750,7 @@ out: } unlock_page(page); if (!S_ISDIR(inode->i_mode) && !IS_NOQUOTA(inode) && - !F2FS_I(inode)->cp_task) + !F2FS_I(inode)->cp_task && allow_balance) f2fs_balance_fs(sbi, need_balance_fs); if (unlikely(f2fs_cp_error(sbi))) { @@ -2928,7 +2797,7 @@ out: #endif return f2fs_write_single_data_page(page, NULL, NULL, NULL, - wbc, FS_DATA_IO, 0); + wbc, FS_DATA_IO, 0, true); } /* @@ -3096,7 +2965,8 @@ continue_unlock: } #endif ret = f2fs_write_single_data_page(page, &submitted, - &bio, &last_block, wbc, io_type, 0); + &bio, &last_block, wbc, io_type, + 0, true); if (ret == AOP_WRITEPAGE_ACTIVATE) unlock_page(page); #ifdef CONFIG_F2FS_FS_COMPRESSION @@ -3831,7 +3701,7 @@ static sector_t f2fs_bmap(struct address_space *mapping, sector_t block) filemap_write_and_wait(mapping); /* Block number less than F2FS MAX BLOCKS */ - if (unlikely(block >= F2FS_I_SB(inode)->max_file_blocks)) + if (unlikely(block >= max_file_blocks(inode))) goto out; if (f2fs_compressed_file(inode)) { @@ -4108,12 +3978,13 @@ static int f2fs_swap_activate(struct swap_info_struct *sis, struct file *file, if (!f2fs_disable_compressed_file(inode)) return -EINVAL; + f2fs_precache_extents(inode); + ret = check_swap_activate(sis, file, span); if (ret < 0) return ret; set_inode_flag(inode, FI_PIN_FILE); - f2fs_precache_extents(inode); f2fs_update_time(F2FS_I_SB(inode), REQ_TIME); return ret; } diff --git a/fs/f2fs/debug.c b/fs/f2fs/debug.c index 197c914119da..91855d5721cd 100644 --- a/fs/f2fs/debug.c +++ b/fs/f2fs/debug.c @@ -120,6 +120,13 @@ static void update_general_status(struct f2fs_sb_info *sbi) atomic_read(&SM_I(sbi)->dcc_info->discard_cmd_cnt); si->undiscard_blks = SM_I(sbi)->dcc_info->undiscard_blks; } + si->nr_issued_ckpt = atomic_read(&sbi->cprc_info.issued_ckpt); + si->nr_total_ckpt = atomic_read(&sbi->cprc_info.total_ckpt); + si->nr_queued_ckpt = atomic_read(&sbi->cprc_info.queued_ckpt); + spin_lock(&sbi->cprc_info.stat_lock); + si->cur_ckpt_time = sbi->cprc_info.cur_time; + si->peak_ckpt_time = sbi->cprc_info.peak_time; + spin_unlock(&sbi->cprc_info.stat_lock); si->total_count = (int)sbi->user_block_count / sbi->blocks_per_seg; si->rsvd_segs = reserved_segments(sbi); si->overp_segs = overprovision_segments(sbi); @@ -417,6 +424,11 @@ static int stat_show(struct seq_file *s, void *v) si->meta_count[META_NAT]); seq_printf(s, " - ssa blocks : %u\n", si->meta_count[META_SSA]); + seq_printf(s, "CP merge (Queued: %4d, Issued: %4d, Total: %4d, " + "Cur time: %4d(ms), Peak time: %4d(ms))\n", + si->nr_queued_ckpt, si->nr_issued_ckpt, + si->nr_total_ckpt, si->cur_ckpt_time, + si->peak_ckpt_time); seq_printf(s, "GC calls: %d (BG: %d)\n", si->call_count, si->bg_gc); seq_printf(s, " - data segments : %d (%d)\n", diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index bb11759191dc..e2d302ae3a46 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -43,7 +43,6 @@ enum { FAULT_KVMALLOC, FAULT_PAGE_ALLOC, FAULT_PAGE_GET, - FAULT_ALLOC_BIO, FAULT_ALLOC_NID, FAULT_ORPHAN, FAULT_BLOCK, @@ -97,6 +96,7 @@ extern const char *f2fs_fault_name[FAULT_MAX]; #define F2FS_MOUNT_DISABLE_CHECKPOINT 0x02000000 #define F2FS_MOUNT_NORECOVERY 0x04000000 #define F2FS_MOUNT_ATGC 0x08000000 +#define F2FS_MOUNT_MERGE_CHECKPOINT 0x10000000 #define F2FS_OPTION(sbi) ((sbi)->mount_opt) #define clear_opt(sbi, option) (F2FS_OPTION(sbi).opt &= ~F2FS_MOUNT_##option) @@ -146,6 +146,7 @@ struct f2fs_mount_info { /* For compression */ unsigned char compress_algorithm; /* algorithm type */ unsigned char compress_log_size; /* cluster log size */ + unsigned char compress_level; /* compress level */ bool compress_chksum; /* compressed data chksum */ unsigned char compress_ext_cnt; /* extension count */ int compress_mode; /* compression mode */ @@ -266,6 +267,26 @@ struct fsync_node_entry { unsigned int seq_id; /* sequence id */ }; +struct ckpt_req { + struct completion wait; /* completion for checkpoint done */ + struct llist_node llnode; /* llist_node to be linked in wait queue */ + int ret; /* return code of checkpoint */ + ktime_t queue_time; /* request queued time */ +}; + +struct ckpt_req_control { + struct task_struct *f2fs_issue_ckpt; /* checkpoint task */ + int ckpt_thread_ioprio; /* checkpoint merge thread ioprio */ + wait_queue_head_t ckpt_wait_queue; /* waiting queue for wake-up */ + atomic_t issued_ckpt; /* # of actually issued ckpts */ + atomic_t total_ckpt; /* # of total ckpts */ + atomic_t queued_ckpt; /* # of queued ckpts */ + struct llist_head issue_list; /* list for command issue */ + spinlock_t stat_lock; /* lock for below checkpoint time stats */ + unsigned int cur_time; /* cur wait time in msec for currently issued checkpoint */ + unsigned int peak_time; /* peak wait time in msec until now */ +}; + /* for the bitmap indicate blocks to be discarded */ struct discard_entry { struct list_head list; /* list head */ @@ -717,7 +738,6 @@ struct f2fs_inode_info { struct list_head inmem_pages; /* inmemory pages managed by f2fs */ struct task_struct *inmem_task; /* store inmemory task */ struct mutex inmem_lock; /* lock for inmemory pages */ - pgoff_t ra_offset; /* ongoing readahead offset */ struct extent_tree *extent_tree; /* cached extent_tree entry */ /* avoid racing between foreground op and gc */ @@ -735,6 +755,7 @@ struct f2fs_inode_info { atomic_t i_compr_blocks; /* # of compressed blocks */ unsigned char i_compress_algorithm; /* algorithm type */ unsigned char i_log_cluster_size; /* log of cluster size */ + unsigned char i_compress_level; /* compress level (lz4hc,zstd) */ unsigned short i_compress_flag; /* compress flag */ unsigned int i_cluster_size; /* cluster size */ }; @@ -1310,6 +1331,8 @@ struct compress_data { #define F2FS_COMPRESSED_PAGE_MAGIC 0xF5F2C000 +#define COMPRESS_LEVEL_OFFSET 8 + /* compress context */ struct compress_ctx { struct inode *inode; /* inode the context belong to */ @@ -1337,7 +1360,7 @@ struct compress_io_ctx { atomic_t pending_pages; /* in-flight compressed page count */ }; -/* decompress io context for read IO path */ +/* Context for decompressing one cluster on the read IO path */ struct decompress_io_ctx { u32 magic; /* magic number to indicate page is compressed */ struct inode *inode; /* inode the context belong to */ @@ -1353,11 +1376,37 @@ struct decompress_io_ctx { struct compress_data *cbuf; /* virtual mapped address on cpages */ size_t rlen; /* valid data length in rbuf */ size_t clen; /* valid data length in cbuf */ - atomic_t pending_pages; /* in-flight compressed page count */ - atomic_t verity_pages; /* in-flight page count for verity */ - bool failed; /* indicate IO error during decompression */ + + /* + * The number of compressed pages remaining to be read in this cluster. + * This is initially nr_cpages. It is decremented by 1 each time a page + * has been read (or failed to be read). When it reaches 0, the cluster + * is decompressed (or an error is reported). + * + * If an error occurs before all the pages have been submitted for I/O, + * then this will never reach 0. In this case the I/O submitter is + * responsible for calling f2fs_decompress_end_io() instead. + */ + atomic_t remaining_pages; + + /* + * Number of references to this decompress_io_ctx. + * + * One reference is held for I/O completion. This reference is dropped + * after the pagecache pages are updated and unlocked -- either after + * decompression (and verity if enabled), or after an error. + * + * In addition, each compressed page holds a reference while it is in a + * bio. These references are necessary prevent compressed pages from + * being freed while they are still in a bio. + */ + refcount_t refcnt; + + bool failed; /* IO error occurred before decompression? */ + bool need_verity; /* need fs-verity verification after decompression? */ void *private; /* payload buffer for specified decompression algorithm */ void *private2; /* extra payload buffer */ + struct work_struct verity_work; /* work to verify the decompressed pages */ }; #define NULL_CLUSTER ((unsigned int)(~0)) @@ -1404,6 +1453,7 @@ struct f2fs_sb_info { wait_queue_head_t cp_wait; unsigned long last_time[MAX_TIME]; /* to store time in jiffies */ long interval_time[MAX_TIME]; /* to store thresholds */ + struct ckpt_req_control cprc_info; /* for checkpoint request control */ struct inode_management im[MAX_INO_ENTRY]; /* manage inode cache */ @@ -1444,7 +1494,6 @@ struct f2fs_sb_info { unsigned int total_sections; /* total section count */ unsigned int total_node_count; /* total node block count */ unsigned int total_valid_node_count; /* valid node block count */ - loff_t max_file_blocks; /* max block index of file */ int dir_level; /* directory level */ int readdir_ra; /* readahead inode in readdir */ u64 max_io_bytes; /* max io bytes to merge IOs */ @@ -1541,9 +1590,12 @@ struct f2fs_sb_info { unsigned int node_io_flag; /* For sysfs suppport */ - struct kobject s_kobj; + struct kobject s_kobj; /* /sys/fs/f2fs/<devname> */ struct completion s_kobj_unregister; + struct kobject s_stat_kobj; /* /sys/fs/f2fs/<devname>/stat */ + struct completion s_stat_kobj_unregister; + /* For shrinker support */ struct list_head s_list; int s_ndevs; /* number of devices */ @@ -3135,9 +3187,10 @@ void f2fs_truncate_data_blocks(struct dnode_of_data *dn); int f2fs_do_truncate_blocks(struct inode *inode, u64 from, bool lock); int f2fs_truncate_blocks(struct inode *inode, u64 from, bool lock); int f2fs_truncate(struct inode *inode); -int f2fs_getattr(const struct path *path, struct kstat *stat, - u32 request_mask, unsigned int flags); -int f2fs_setattr(struct dentry *dentry, struct iattr *attr); +int f2fs_getattr(struct user_namespace *mnt_userns, const struct path *path, + struct kstat *stat, u32 request_mask, unsigned int flags); +int f2fs_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *attr); int f2fs_truncate_hole(struct inode *inode, pgoff_t pg_start, pgoff_t pg_end); void f2fs_truncate_data_blocks_range(struct dnode_of_data *dn, int count); int f2fs_precache_extents(struct inode *inode); @@ -3232,6 +3285,7 @@ int f2fs_inode_dirtied(struct inode *inode, bool sync); void f2fs_inode_synced(struct inode *inode); int f2fs_enable_quota_files(struct f2fs_sb_info *sbi, bool rdonly); int f2fs_quota_sync(struct super_block *sb, int type); +loff_t max_file_blocks(struct inode *inode); void f2fs_quota_off_umount(struct super_block *sb); int f2fs_commit_super(struct f2fs_sb_info *sbi, bool recover); int f2fs_sync_fs(struct super_block *sb, int sync); @@ -3418,13 +3472,16 @@ int f2fs_write_checkpoint(struct f2fs_sb_info *sbi, struct cp_control *cpc); void f2fs_init_ino_entry_info(struct f2fs_sb_info *sbi); int __init f2fs_create_checkpoint_caches(void); void f2fs_destroy_checkpoint_caches(void); +int f2fs_issue_checkpoint(struct f2fs_sb_info *sbi); +int f2fs_start_ckpt_thread(struct f2fs_sb_info *sbi); +void f2fs_stop_ckpt_thread(struct f2fs_sb_info *sbi); +void f2fs_init_ckpt_req_control(struct f2fs_sb_info *sbi); /* * data.c */ int __init f2fs_init_bioset(void); void f2fs_destroy_bioset(void); -struct bio *f2fs_bio_alloc(struct f2fs_sb_info *sbi, int npages, bool noio); int f2fs_init_bio_entry_cache(void); void f2fs_destroy_bio_entry_cache(void); void f2fs_submit_bio(struct f2fs_sb_info *sbi, @@ -3469,7 +3526,7 @@ int f2fs_write_single_data_page(struct page *page, int *submitted, struct bio **bio, sector_t *last_block, struct writeback_control *wbc, enum iostat_type io_type, - int compr_blocks); + int compr_blocks, bool allow_balance); void f2fs_invalidate_page(struct page *page, unsigned int offset, unsigned int length); int f2fs_release_page(struct page *page, gfp_t wait); @@ -3530,6 +3587,8 @@ struct f2fs_stat_info { int nr_discarding, nr_discarded; int nr_discard_cmd; unsigned int undiscard_blks; + int nr_issued_ckpt, nr_total_ckpt, nr_queued_ckpt; + unsigned int cur_ckpt_time, peak_ckpt_time; int inline_xattr, inline_inode, inline_dir, append, update, orphans; int compr_inode; unsigned long long compr_blocks; @@ -3715,8 +3774,6 @@ void f2fs_update_sit_info(struct f2fs_sb_info *sbi); #define stat_dec_compr_inode(inode) do { } while (0) #define stat_add_compr_blocks(inode, blocks) do { } while (0) #define stat_sub_compr_blocks(inode, blocks) do { } while (0) -#define stat_inc_atomic_write(inode) do { } while (0) -#define stat_dec_atomic_write(inode) do { } while (0) #define stat_update_max_atomic_write(inode) do { } while (0) #define stat_inc_volatile_write(inode) do { } while (0) #define stat_dec_volatile_write(inode) do { } while (0) @@ -3876,7 +3933,7 @@ void f2fs_compress_write_end_io(struct bio *bio, struct page *page); bool f2fs_is_compress_backend_ready(struct inode *inode); int f2fs_init_compress_mempool(void); void f2fs_destroy_compress_mempool(void); -void f2fs_decompress_pages(struct bio *bio, struct page *page, bool verity); +void f2fs_end_read_compressed_page(struct page *page, bool failed); bool f2fs_cluster_is_empty(struct compress_ctx *cc); bool f2fs_cluster_can_merge_page(struct compress_ctx *cc, pgoff_t index); void f2fs_compress_ctx_add_page(struct compress_ctx *cc, struct page *page); @@ -3889,9 +3946,8 @@ int f2fs_read_multi_pages(struct compress_ctx *cc, struct bio **bio_ret, unsigned nr_pages, sector_t *last_block_in_bio, bool is_readahead, bool for_write); struct decompress_io_ctx *f2fs_alloc_dic(struct compress_ctx *cc); -void f2fs_free_dic(struct decompress_io_ctx *dic); -void f2fs_decompress_end_io(struct page **rpages, - unsigned int cluster_size, bool err, bool verity); +void f2fs_decompress_end_io(struct decompress_io_ctx *dic, bool failed); +void f2fs_put_page_dic(struct page *page); int f2fs_init_compress_ctx(struct compress_ctx *cc); void f2fs_destroy_compress_ctx(struct compress_ctx *cc); void f2fs_init_compress_info(struct f2fs_sb_info *sbi); @@ -3915,6 +3971,14 @@ static inline struct page *f2fs_compress_control_page(struct page *page) } static inline int f2fs_init_compress_mempool(void) { return 0; } static inline void f2fs_destroy_compress_mempool(void) { } +static inline void f2fs_end_read_compressed_page(struct page *page, bool failed) +{ + WARN_ON_ONCE(1); +} +static inline void f2fs_put_page_dic(struct page *page) +{ + WARN_ON_ONCE(1); +} static inline int f2fs_init_page_array_cache(struct f2fs_sb_info *sbi) { return 0; } static inline void f2fs_destroy_page_array_cache(struct f2fs_sb_info *sbi) { } static inline int __init f2fs_init_compress_cache(void) { return 0; } @@ -3934,6 +3998,11 @@ static inline void set_compress_context(struct inode *inode) 1 << COMPRESS_CHKSUM : 0; F2FS_I(inode)->i_cluster_size = 1 << F2FS_I(inode)->i_log_cluster_size; + if (F2FS_I(inode)->i_compress_algorithm == COMPRESS_LZ4 && + F2FS_OPTION(sbi).compress_level) + F2FS_I(inode)->i_compress_flag |= + F2FS_OPTION(sbi).compress_level << + COMPRESS_LEVEL_OFFSET; F2FS_I(inode)->i_flags |= F2FS_COMPR_FL; set_inode_flag(inode, FI_COMPRESSED_FILE); stat_inc_compr_inode(inode); @@ -4114,6 +4183,12 @@ static inline bool f2fs_force_buffered_io(struct inode *inode, return false; } +static inline bool f2fs_need_verity(const struct inode *inode, pgoff_t idx) +{ + return fsverity_active(inode) && + idx < DIV_ROUND_UP(inode->i_size, PAGE_SIZE); +} + #ifdef CONFIG_F2FS_FAULT_INJECTION extern void f2fs_build_fault_attr(struct f2fs_sb_info *sbi, unsigned int rate, unsigned int type); diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index f585545277d7..d26ff2ae3f5e 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -29,7 +29,6 @@ #include "xattr.h" #include "acl.h" #include "gc.h" -#include "trace.h" #include <trace/events/f2fs.h> #include <uapi/linux/f2fs.h> @@ -60,6 +59,9 @@ static vm_fault_t f2fs_vm_page_mkwrite(struct vm_fault *vmf) bool need_alloc = true; int err = 0; + if (unlikely(IS_IMMUTABLE(inode))) + return VM_FAULT_SIGBUS; + if (unlikely(f2fs_cp_error(sbi))) { err = -EIO; goto err; @@ -70,6 +72,10 @@ static vm_fault_t f2fs_vm_page_mkwrite(struct vm_fault *vmf) goto err; } + err = f2fs_convert_inline_inode(inode); + if (err) + goto err; + #ifdef CONFIG_F2FS_FS_COMPRESSION if (f2fs_compressed_file(inode)) { int ret = f2fs_is_compressed_cluster(inode, page->index); @@ -366,7 +372,6 @@ flush_out: f2fs_update_time(sbi, REQ_TIME); out: trace_f2fs_sync_file_exit(inode, cp_reason, datasync, ret); - f2fs_trace_ios(NULL, 1); return ret; } @@ -483,6 +488,9 @@ static loff_t f2fs_llseek(struct file *file, loff_t offset, int whence) struct inode *inode = file->f_mapping->host; loff_t maxbytes = inode->i_sb->s_maxbytes; + if (f2fs_compressed_file(inode)) + maxbytes = max_file_blocks(inode) << F2FS_BLKSIZE_BITS; + switch (whence) { case SEEK_SET: case SEEK_CUR: @@ -502,7 +510,6 @@ static loff_t f2fs_llseek(struct file *file, loff_t offset, int whence) static int f2fs_file_mmap(struct file *file, struct vm_area_struct *vma) { struct inode *inode = file_inode(file); - int err; if (unlikely(f2fs_cp_error(F2FS_I_SB(inode)))) return -EIO; @@ -510,11 +517,6 @@ static int f2fs_file_mmap(struct file *file, struct vm_area_struct *vma) if (!f2fs_is_compress_backend_ready(inode)) return -EOPNOTSUPP; - /* we don't need to use inline_data strictly */ - err = f2fs_convert_inline_inode(inode); - if (err) - return err; - file_accessed(file); vma->vm_ops = &f2fs_file_vm_ops; set_inode_flag(inode, FI_MMAP_FILE); @@ -667,7 +669,7 @@ int f2fs_do_truncate_blocks(struct inode *inode, u64 from, bool lock) free_from = (pgoff_t)F2FS_BLK_ALIGN(from); - if (free_from >= sbi->max_file_blocks) + if (free_from >= max_file_blocks(inode)) goto free_partial; if (lock) @@ -767,6 +769,10 @@ int f2fs_truncate(struct inode *inode) return -EIO; } + err = dquot_initialize(inode); + if (err) + return err; + /* we should check inline_data size */ if (!f2fs_may_inline_data(inode)) { err = f2fs_convert_inline_inode(inode); @@ -783,8 +789,8 @@ int f2fs_truncate(struct inode *inode) return 0; } -int f2fs_getattr(const struct path *path, struct kstat *stat, - u32 request_mask, unsigned int query_flags) +int f2fs_getattr(struct user_namespace *mnt_userns, const struct path *path, + struct kstat *stat, u32 request_mask, unsigned int query_flags) { struct inode *inode = d_inode(path->dentry); struct f2fs_inode_info *fi = F2FS_I(inode); @@ -820,7 +826,7 @@ int f2fs_getattr(const struct path *path, struct kstat *stat, STATX_ATTR_NODUMP | STATX_ATTR_VERITY); - generic_fillattr(inode, stat); + generic_fillattr(&init_user_ns, inode, stat); /* we need to show initial sectors used for inline_data/dentries */ if ((S_ISREG(inode->i_mode) && f2fs_has_inline_data(inode)) || @@ -831,7 +837,8 @@ int f2fs_getattr(const struct path *path, struct kstat *stat, } #ifdef CONFIG_F2FS_FS_POSIX_ACL -static void __setattr_copy(struct inode *inode, const struct iattr *attr) +static void __setattr_copy(struct user_namespace *mnt_userns, + struct inode *inode, const struct iattr *attr) { unsigned int ia_valid = attr->ia_valid; @@ -847,8 +854,9 @@ static void __setattr_copy(struct inode *inode, const struct iattr *attr) inode->i_ctime = attr->ia_ctime; if (ia_valid & ATTR_MODE) { umode_t mode = attr->ia_mode; + kgid_t kgid = i_gid_into_mnt(mnt_userns, inode); - if (!in_group_p(inode->i_gid) && !capable(CAP_FSETID)) + if (!in_group_p(kgid) && !capable_wrt_inode_uidgid(mnt_userns, inode, CAP_FSETID)) mode &= ~S_ISGID; set_acl_inode(inode, mode); } @@ -857,7 +865,8 @@ static void __setattr_copy(struct inode *inode, const struct iattr *attr) #define __setattr_copy setattr_copy #endif -int f2fs_setattr(struct dentry *dentry, struct iattr *attr) +int f2fs_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *attr) { struct inode *inode = d_inode(dentry); int err; @@ -865,11 +874,19 @@ int f2fs_setattr(struct dentry *dentry, struct iattr *attr) if (unlikely(f2fs_cp_error(F2FS_I_SB(inode)))) return -EIO; + if (unlikely(IS_IMMUTABLE(inode))) + return -EPERM; + + if (unlikely(IS_APPEND(inode) && + (attr->ia_valid & (ATTR_MODE | ATTR_UID | + ATTR_GID | ATTR_TIMES_SET)))) + return -EPERM; + if ((attr->ia_valid & ATTR_SIZE) && !f2fs_is_compress_backend_ready(inode)) return -EOPNOTSUPP; - err = setattr_prepare(dentry, attr); + err = setattr_prepare(&init_user_ns, dentry, attr); if (err) return err; @@ -945,12 +962,14 @@ int f2fs_setattr(struct dentry *dentry, struct iattr *attr) spin_unlock(&F2FS_I(inode)->i_size_lock); } - __setattr_copy(inode, attr); + __setattr_copy(&init_user_ns, inode, attr); if (attr->ia_valid & ATTR_MODE) { - err = posix_acl_chmod(inode, f2fs_get_inode_mode(inode)); - if (err || is_inode_flag_set(inode, FI_ACL_MODE)) { - inode->i_mode = F2FS_I(inode)->i_acl_mode; + err = posix_acl_chmod(&init_user_ns, inode, f2fs_get_inode_mode(inode)); + + if (is_inode_flag_set(inode, FI_ACL_MODE)) { + if (!err) + inode->i_mode = F2FS_I(inode)->i_acl_mode; clear_inode_flag(inode, FI_ACL_MODE); } } @@ -1961,7 +1980,7 @@ static int f2fs_ioc_setflags(struct file *filp, unsigned long arg) u32 iflags; int ret; - if (!inode_owner_or_capable(inode)) + if (!inode_owner_or_capable(&init_user_ns, inode)) return -EACCES; if (get_user(fsflags, (int __user *)arg)) @@ -2008,7 +2027,7 @@ static int f2fs_ioc_start_atomic_write(struct file *filp) struct f2fs_sb_info *sbi = F2FS_I_SB(inode); int ret; - if (!inode_owner_or_capable(inode)) + if (!inode_owner_or_capable(&init_user_ns, inode)) return -EACCES; if (!S_ISREG(inode->i_mode)) @@ -2075,7 +2094,7 @@ static int f2fs_ioc_commit_atomic_write(struct file *filp) struct inode *inode = file_inode(filp); int ret; - if (!inode_owner_or_capable(inode)) + if (!inode_owner_or_capable(&init_user_ns, inode)) return -EACCES; ret = mnt_want_write_file(filp); @@ -2117,7 +2136,7 @@ static int f2fs_ioc_start_volatile_write(struct file *filp) struct inode *inode = file_inode(filp); int ret; - if (!inode_owner_or_capable(inode)) + if (!inode_owner_or_capable(&init_user_ns, inode)) return -EACCES; if (!S_ISREG(inode->i_mode)) @@ -2152,7 +2171,7 @@ static int f2fs_ioc_release_volatile_write(struct file *filp) struct inode *inode = file_inode(filp); int ret; - if (!inode_owner_or_capable(inode)) + if (!inode_owner_or_capable(&init_user_ns, inode)) return -EACCES; ret = mnt_want_write_file(filp); @@ -2181,7 +2200,7 @@ static int f2fs_ioc_abort_volatile_write(struct file *filp) struct inode *inode = file_inode(filp); int ret; - if (!inode_owner_or_capable(inode)) + if (!inode_owner_or_capable(&init_user_ns, inode)) return -EACCES; ret = mnt_want_write_file(filp); @@ -2730,7 +2749,7 @@ static int f2fs_ioc_defragment(struct file *filp, unsigned long arg) return -EINVAL; if (unlikely((range.start + range.len) >> PAGE_SHIFT > - sbi->max_file_blocks)) + max_file_blocks(inode))) return -EINVAL; err = mnt_want_write_file(filp); @@ -3158,7 +3177,7 @@ static int f2fs_ioc_fssetxattr(struct file *filp, unsigned long arg) return -EFAULT; /* Make sure caller has proper permission */ - if (!inode_owner_or_capable(inode)) + if (!inode_owner_or_capable(&init_user_ns, inode)) return -EACCES; if (fa.fsx_xflags & ~F2FS_SUPPORTED_XFLAGS) @@ -3293,7 +3312,7 @@ int f2fs_precache_extents(struct inode *inode) map.m_next_extent = &m_next_extent; map.m_seg_type = NO_CHECK_TYPE; map.m_may_create = false; - end = F2FS_I_SB(inode)->max_file_blocks; + end = max_file_blocks(inode); while (map.m_lblk < end) { map.m_len = end - map.m_lblk; @@ -3357,6 +3376,14 @@ static int f2fs_ioc_measure_verity(struct file *filp, unsigned long arg) return fsverity_ioctl_measure(filp, (void __user *)arg); } +static int f2fs_ioc_read_verity_metadata(struct file *filp, unsigned long arg) +{ + if (!f2fs_sb_has_verity(F2FS_I_SB(file_inode(filp)))) + return -EOPNOTSUPP; + + return fsverity_ioctl_read_metadata(filp, (const void __user *)arg); +} + static int f2fs_ioc_getfslabel(struct file *filp, unsigned long arg) { struct inode *inode = file_inode(filp); @@ -4043,8 +4070,10 @@ static int redirty_blocks(struct inode *inode, pgoff_t page_idx, int len) for (i = 0; i < page_len; i++, redirty_idx++) { page = find_lock_page(mapping, redirty_idx); - if (!page) - ret = -ENOENT; + if (!page) { + ret = -ENOMEM; + break; + } set_page_dirty(page); f2fs_put_page(page, 1); f2fs_put_page(page, 0); @@ -4272,6 +4301,8 @@ static long __f2fs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) return f2fs_ioc_enable_verity(filp, arg); case FS_IOC_MEASURE_VERITY: return f2fs_ioc_measure_verity(filp, arg); + case FS_IOC_READ_VERITY_METADATA: + return f2fs_ioc_read_verity_metadata(filp, arg); case FS_IOC_GETFSLABEL: return f2fs_ioc_getfslabel(filp, arg); case FS_IOC_SETFSLABEL: @@ -4349,6 +4380,11 @@ static ssize_t f2fs_file_write_iter(struct kiocb *iocb, struct iov_iter *from) inode_lock(inode); } + if (unlikely(IS_IMMUTABLE(inode))) { + ret = -EPERM; + goto unlock; + } + ret = generic_write_checks(iocb, from); if (ret > 0) { bool preallocated = false; @@ -4413,6 +4449,7 @@ write: if (ret > 0) f2fs_update_iostat(F2FS_I_SB(inode), APP_WRITE_IO, ret); } +unlock: inode_unlock(inode); out: trace_f2fs_file_write_iter(inode, iocb->ki_pos, @@ -4523,6 +4560,7 @@ long f2fs_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg) case F2FS_IOC_RESIZE_FS: case FS_IOC_ENABLE_VERITY: case FS_IOC_MEASURE_VERITY: + case FS_IOC_READ_VERITY_METADATA: case FS_IOC_GETFSLABEL: case FS_IOC_SETFSLABEL: case F2FS_IOC_GET_COMPRESS_BLOCKS: diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c index 3ef84e6ded41..39330ad3c44e 100644 --- a/fs/f2fs/gc.c +++ b/fs/f2fs/gc.c @@ -1169,8 +1169,6 @@ static int move_data_block(struct inode *inode, block_t bidx, if (err) goto put_out; - set_summary(&sum, dn.nid, dn.ofs_in_node, ni.version); - /* read page */ fio.page = page; fio.new_blkaddr = fio.old_blkaddr = dn.data_blkaddr; @@ -1207,6 +1205,9 @@ static int move_data_block(struct inode *inode, block_t bidx, } } + set_summary(&sum, dn.nid, dn.ofs_in_node, ni.version); + + /* allocate block address */ f2fs_allocate_data_block(fio.sbi, NULL, fio.old_blkaddr, &newaddr, &sum, type, NULL); @@ -1233,9 +1234,6 @@ static int move_data_block(struct inode *inode, block_t bidx, set_page_writeback(fio.encrypted_page); ClearPageError(page); - /* allocate block address */ - f2fs_wait_on_page_writeback(dn.node_page, NODE, true, true); - fio.op = REQ_OP_WRITE; fio.op_flags = REQ_SYNC; fio.new_blkaddr = newaddr; diff --git a/fs/f2fs/inline.c b/fs/f2fs/inline.c index 806ebabf5870..993caefcd2bb 100644 --- a/fs/f2fs/inline.c +++ b/fs/f2fs/inline.c @@ -192,6 +192,10 @@ int f2fs_convert_inline_inode(struct inode *inode) f2fs_hw_is_readonly(sbi) || f2fs_readonly(sbi->sb)) return 0; + err = dquot_initialize(inode); + if (err) + return err; + page = f2fs_grab_cache_page(inode->i_mapping, 0, false); if (!page) return -ENOMEM; diff --git a/fs/f2fs/namei.c b/fs/f2fs/namei.c index 6edb1ab579a1..17bd072a5d39 100644 --- a/fs/f2fs/namei.c +++ b/fs/f2fs/namei.c @@ -46,7 +46,7 @@ static struct inode *f2fs_new_inode(struct inode *dir, umode_t mode) nid_free = true; - inode_init_owner(inode, dir, mode); + inode_init_owner(&init_user_ns, inode, dir, mode); inode->i_ino = ino; inode->i_blocks = 0; @@ -314,8 +314,8 @@ static void set_compress_inode(struct f2fs_sb_info *sbi, struct inode *inode, } } -static int f2fs_create(struct inode *dir, struct dentry *dentry, umode_t mode, - bool excl) +static int f2fs_create(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, bool excl) { struct f2fs_sb_info *sbi = F2FS_I_SB(dir); struct inode *inode; @@ -637,8 +637,8 @@ static const char *f2fs_get_link(struct dentry *dentry, return link; } -static int f2fs_symlink(struct inode *dir, struct dentry *dentry, - const char *symname) +static int f2fs_symlink(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, const char *symname) { struct f2fs_sb_info *sbi = F2FS_I_SB(dir); struct inode *inode; @@ -717,7 +717,8 @@ out_free_encrypted_link: return err; } -static int f2fs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) +static int f2fs_mkdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { struct f2fs_sb_info *sbi = F2FS_I_SB(dir); struct inode *inode; @@ -770,8 +771,8 @@ static int f2fs_rmdir(struct inode *dir, struct dentry *dentry) return -ENOTEMPTY; } -static int f2fs_mknod(struct inode *dir, struct dentry *dentry, - umode_t mode, dev_t rdev) +static int f2fs_mknod(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, dev_t rdev) { struct f2fs_sb_info *sbi = F2FS_I_SB(dir); struct inode *inode; @@ -855,7 +856,11 @@ static int __f2fs_tmpfile(struct inode *dir, struct dentry *dentry, if (whiteout) { f2fs_i_links_write(inode, false); + + spin_lock(&inode->i_lock); inode->i_state |= I_LINKABLE; + spin_unlock(&inode->i_lock); + *whiteout = inode; } else { d_tmpfile(dentry, inode); @@ -874,7 +879,8 @@ out: return err; } -static int f2fs_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode) +static int f2fs_tmpfile(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { struct f2fs_sb_info *sbi = F2FS_I_SB(dir); @@ -1041,7 +1047,11 @@ static int f2fs_rename(struct inode *old_dir, struct dentry *old_dentry, err = f2fs_add_link(old_dentry, whiteout); if (err) goto put_out_dir; + + spin_lock(&whiteout->i_lock); whiteout->i_state &= ~I_LINKABLE; + spin_unlock(&whiteout->i_lock); + iput(whiteout); } @@ -1247,7 +1257,8 @@ out: return err; } -static int f2fs_rename2(struct inode *old_dir, struct dentry *old_dentry, +static int f2fs_rename2(struct user_namespace *mnt_userns, + struct inode *old_dir, struct dentry *old_dentry, struct inode *new_dir, struct dentry *new_dentry, unsigned int flags) { diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c index 3a24423ac65f..4b0e2e3c2c88 100644 --- a/fs/f2fs/node.c +++ b/fs/f2fs/node.c @@ -17,7 +17,6 @@ #include "node.h" #include "segment.h" #include "xattr.h" -#include "trace.h" #include <trace/events/f2fs.h> #define on_f2fs_build_free_nids(nmi) mutex_is_locked(&(nm_i)->build_lock) @@ -2089,7 +2088,6 @@ static int f2fs_set_node_page_dirty(struct page *page) __set_page_dirty_nobuffers(page); inc_page_count(F2FS_P_SB(page), F2FS_DIRTY_NODES); f2fs_set_page_private(page, 0); - f2fs_trace_pid(page); return 1; } return 0; @@ -2696,7 +2694,7 @@ retry: src = F2FS_INODE(page); dst = F2FS_INODE(ipage); - memcpy(dst, src, (unsigned long)&src->i_ext - (unsigned long)src); + memcpy(dst, src, offsetof(struct f2fs_inode, i_ext)); dst->i_size = 0; dst->i_blocks = cpu_to_le64(1); dst->i_links = cpu_to_le32(1); @@ -2749,7 +2747,7 @@ int f2fs_restore_node_summary(struct f2fs_sb_info *sbi, sum_entry = &sum->entries[0]; for (i = 0; i < last_offset; i += nrpages, addr += nrpages) { - nrpages = min(last_offset - i, BIO_MAX_PAGES); + nrpages = bio_max_segs(last_offset - i); /* readahead node pages */ f2fs_ra_meta_pages(sbi, addr, nrpages, META_POR, true); diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index deca74cb17df..c2866561263e 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -20,7 +20,6 @@ #include "segment.h" #include "node.h" #include "gc.h" -#include "trace.h" #include <trace/events/f2fs.h> #define __reverse_ffz(x) __reverse_ffs(~(x)) @@ -187,8 +186,6 @@ void f2fs_register_inmem_page(struct inode *inode, struct page *page) { struct inmem_pages *new; - f2fs_trace_pid(page); - f2fs_set_page_private(page, ATOMIC_WRITTEN_PAGE); new = f2fs_kmem_cache_alloc(inmem_entry_slab, GFP_NOFS); @@ -566,17 +563,7 @@ do_sync: static int __submit_flush_wait(struct f2fs_sb_info *sbi, struct block_device *bdev) { - struct bio *bio; - int ret; - - bio = f2fs_bio_alloc(sbi, 0, false); - if (!bio) - return -ENOMEM; - - bio->bi_opf = REQ_OP_WRITE | REQ_SYNC | REQ_PREFLUSH; - bio_set_dev(bio, bdev); - ret = submit_bio_wait(bio); - bio_put(bio); + int ret = blkdev_issue_flush(bdev); trace_f2fs_issue_flush(bdev, test_opt(sbi, NOBARRIER), test_opt(sbi, FLUSH_MERGE), ret); @@ -610,8 +597,6 @@ repeat: if (kthread_should_stop()) return 0; - sb_start_intwrite(sbi->sb); - if (!llist_empty(&fcc->issue_list)) { struct flush_cmd *cmd, *next; int ret; @@ -632,8 +617,6 @@ repeat: fcc->dispatch_list = NULL; } - sb_end_intwrite(sbi->sb); - wait_event_interruptible(*q, kthread_should_stop() || !llist_empty(&fcc->issue_list)); goto repeat; @@ -4398,7 +4381,7 @@ static int build_sit_entries(struct f2fs_sb_info *sbi) block_t total_node_blocks = 0; do { - readed = f2fs_ra_meta_pages(sbi, start_blk, BIO_MAX_PAGES, + readed = f2fs_ra_meta_pages(sbi, start_blk, BIO_MAX_VECS, META_SIT, true); start = start_blk * sit_i->sents_per_block; diff --git a/fs/f2fs/segment.h b/fs/f2fs/segment.h index e81eb0748e2a..e9a7a637d688 100644 --- a/fs/f2fs/segment.h +++ b/fs/f2fs/segment.h @@ -101,11 +101,11 @@ static inline void sanity_check_seg_type(struct f2fs_sb_info *sbi, #define BLKS_PER_SEC(sbi) \ ((sbi)->segs_per_sec * (sbi)->blocks_per_seg) #define GET_SEC_FROM_SEG(sbi, segno) \ - ((segno) / (sbi)->segs_per_sec) + (((segno) == -1) ? -1: (segno) / (sbi)->segs_per_sec) #define GET_SEG_FROM_SEC(sbi, secno) \ ((secno) * (sbi)->segs_per_sec) #define GET_ZONE_FROM_SEC(sbi, secno) \ - ((secno) / (sbi)->secs_per_zone) + (((secno) == -1) ? -1: (secno) / (sbi)->secs_per_zone) #define GET_ZONE_FROM_SEG(sbi, segno) \ GET_ZONE_FROM_SEC(sbi, GET_SEC_FROM_SEG(sbi, segno)) @@ -851,7 +851,7 @@ static inline int nr_pages_to_skip(struct f2fs_sb_info *sbi, int type) else if (type == NODE) return 8 * sbi->blocks_per_seg; else if (type == META) - return 8 * BIO_MAX_PAGES; + return 8 * BIO_MAX_VECS; else return 0; } @@ -868,7 +868,7 @@ static inline long nr_pages_to_write(struct f2fs_sb_info *sbi, int type, return 0; nr_to_write = wbc->nr_to_write; - desired = BIO_MAX_PAGES; + desired = BIO_MAX_VECS; if (type == NODE) desired <<= 1; diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c index b4a07fe62d1a..82592b19b4e0 100644 --- a/fs/f2fs/super.c +++ b/fs/f2fs/super.c @@ -25,13 +25,14 @@ #include <linux/quota.h> #include <linux/unicode.h> #include <linux/part_stat.h> +#include <linux/zstd.h> +#include <linux/lz4.h> #include "f2fs.h" #include "node.h" #include "segment.h" #include "xattr.h" #include "gc.h" -#include "trace.h" #define CREATE_TRACE_POINTS #include <trace/events/f2fs.h> @@ -45,7 +46,6 @@ const char *f2fs_fault_name[FAULT_MAX] = { [FAULT_KVMALLOC] = "kvmalloc", [FAULT_PAGE_ALLOC] = "page alloc", [FAULT_PAGE_GET] = "page get", - [FAULT_ALLOC_BIO] = "alloc bio", [FAULT_ALLOC_NID] = "alloc nid", [FAULT_ORPHAN] = "orphan", [FAULT_BLOCK] = "no more block", @@ -143,6 +143,8 @@ enum { Opt_checkpoint_disable_cap, Opt_checkpoint_disable_cap_perc, Opt_checkpoint_enable, + Opt_checkpoint_merge, + Opt_nocheckpoint_merge, Opt_compress_algorithm, Opt_compress_log_size, Opt_compress_extension, @@ -213,6 +215,8 @@ static match_table_t f2fs_tokens = { {Opt_checkpoint_disable_cap, "checkpoint=disable:%u"}, {Opt_checkpoint_disable_cap_perc, "checkpoint=disable:%u%%"}, {Opt_checkpoint_enable, "checkpoint=enable"}, + {Opt_checkpoint_merge, "checkpoint_merge"}, + {Opt_nocheckpoint_merge, "nocheckpoint_merge"}, {Opt_compress_algorithm, "compress_algorithm=%s"}, {Opt_compress_log_size, "compress_log_size=%u"}, {Opt_compress_extension, "compress_extension=%s"}, @@ -464,6 +468,74 @@ static int f2fs_set_test_dummy_encryption(struct super_block *sb, return 0; } +#ifdef CONFIG_F2FS_FS_COMPRESSION +#ifdef CONFIG_F2FS_FS_LZ4 +static int f2fs_set_lz4hc_level(struct f2fs_sb_info *sbi, const char *str) +{ +#ifdef CONFIG_F2FS_FS_LZ4HC + unsigned int level; +#endif + + if (strlen(str) == 3) { + F2FS_OPTION(sbi).compress_level = 0; + return 0; + } + +#ifdef CONFIG_F2FS_FS_LZ4HC + str += 3; + + if (str[0] != ':') { + f2fs_info(sbi, "wrong format, e.g. <alg_name>:<compr_level>"); + return -EINVAL; + } + if (kstrtouint(str + 1, 10, &level)) + return -EINVAL; + + if (level < LZ4HC_MIN_CLEVEL || level > LZ4HC_MAX_CLEVEL) { + f2fs_info(sbi, "invalid lz4hc compress level: %d", level); + return -EINVAL; + } + + F2FS_OPTION(sbi).compress_level = level; + return 0; +#else + f2fs_info(sbi, "kernel doesn't support lz4hc compression"); + return -EINVAL; +#endif +} +#endif + +#ifdef CONFIG_F2FS_FS_ZSTD +static int f2fs_set_zstd_level(struct f2fs_sb_info *sbi, const char *str) +{ + unsigned int level; + int len = 4; + + if (strlen(str) == len) { + F2FS_OPTION(sbi).compress_level = 0; + return 0; + } + + str += len; + + if (str[0] != ':') { + f2fs_info(sbi, "wrong format, e.g. <alg_name>:<compr_level>"); + return -EINVAL; + } + if (kstrtouint(str + 1, 10, &level)) + return -EINVAL; + + if (!level || level > ZSTD_maxCLevel()) { + f2fs_info(sbi, "invalid zstd compress level: %d", level); + return -EINVAL; + } + + F2FS_OPTION(sbi).compress_level = level; + return 0; +} +#endif +#endif + static int parse_options(struct super_block *sb, char *options, bool is_remount) { struct f2fs_sb_info *sbi = F2FS_SB(sb); @@ -681,9 +753,9 @@ static int parse_options(struct super_block *sb, char *options, bool is_remount) case Opt_io_size_bits: if (args->from && match_int(args, &arg)) return -EINVAL; - if (arg <= 0 || arg > __ilog2_u32(BIO_MAX_PAGES)) { + if (arg <= 0 || arg > __ilog2_u32(BIO_MAX_VECS)) { f2fs_warn(sbi, "Not support %d, larger than %d", - 1 << arg, BIO_MAX_PAGES); + 1 << arg, BIO_MAX_VECS); return -EINVAL; } F2FS_OPTION(sbi).write_io_size_bits = arg; @@ -872,6 +944,12 @@ static int parse_options(struct super_block *sb, char *options, bool is_remount) case Opt_checkpoint_enable: clear_opt(sbi, DISABLE_CHECKPOINT); break; + case Opt_checkpoint_merge: + set_opt(sbi, MERGE_CHECKPOINT); + break; + case Opt_nocheckpoint_merge: + clear_opt(sbi, MERGE_CHECKPOINT); + break; #ifdef CONFIG_F2FS_FS_COMPRESSION case Opt_compress_algorithm: if (!f2fs_sb_has_compression(sbi)) { @@ -882,17 +960,45 @@ static int parse_options(struct super_block *sb, char *options, bool is_remount) if (!name) return -ENOMEM; if (!strcmp(name, "lzo")) { +#ifdef CONFIG_F2FS_FS_LZO + F2FS_OPTION(sbi).compress_level = 0; F2FS_OPTION(sbi).compress_algorithm = COMPRESS_LZO; - } else if (!strcmp(name, "lz4")) { +#else + f2fs_info(sbi, "kernel doesn't support lzo compression"); +#endif + } else if (!strncmp(name, "lz4", 3)) { +#ifdef CONFIG_F2FS_FS_LZ4 + ret = f2fs_set_lz4hc_level(sbi, name); + if (ret) { + kfree(name); + return -EINVAL; + } F2FS_OPTION(sbi).compress_algorithm = COMPRESS_LZ4; - } else if (!strcmp(name, "zstd")) { +#else + f2fs_info(sbi, "kernel doesn't support lz4 compression"); +#endif + } else if (!strncmp(name, "zstd", 4)) { +#ifdef CONFIG_F2FS_FS_ZSTD + ret = f2fs_set_zstd_level(sbi, name); + if (ret) { + kfree(name); + return -EINVAL; + } F2FS_OPTION(sbi).compress_algorithm = COMPRESS_ZSTD; +#else + f2fs_info(sbi, "kernel doesn't support zstd compression"); +#endif } else if (!strcmp(name, "lzo-rle")) { +#ifdef CONFIG_F2FS_FS_LZORLE + F2FS_OPTION(sbi).compress_level = 0; F2FS_OPTION(sbi).compress_algorithm = COMPRESS_LZORLE; +#else + f2fs_info(sbi, "kernel doesn't support lzorle compression"); +#endif } else { kfree(name); return -EINVAL; @@ -1076,8 +1182,6 @@ static struct inode *f2fs_alloc_inode(struct super_block *sb) /* Will be used by directory only */ fi->i_dir_level = F2FS_SB(sb)->dir_level; - fi->ra_offset = -1; - return &fi->vfs_inode; } @@ -1196,9 +1300,6 @@ static void f2fs_dirty_inode(struct inode *inode, int flags) inode->i_ino == F2FS_META_INO(sbi)) return; - if (flags == I_DIRTY_TIME) - return; - if (is_inode_flag_set(inode, FI_AUTO_RECOVER)) clear_inode_flag(inode, FI_AUTO_RECOVER); @@ -1246,6 +1347,12 @@ static void f2fs_put_super(struct super_block *sb) mutex_lock(&sbi->umount_mutex); /* + * flush all issued checkpoints and stop checkpoint issue thread. + * after then, all checkpoints should be done by each process context. + */ + f2fs_stop_ckpt_thread(sbi); + + /* * We don't need to do checkpoint when superblock is clean. * But, the previous checkpoint was not done by umount, it needs to do * clean checkpoint again. @@ -1343,16 +1450,8 @@ int f2fs_sync_fs(struct super_block *sb, int sync) if (unlikely(is_sbi_flag_set(sbi, SBI_POR_DOING))) return -EAGAIN; - if (sync) { - struct cp_control cpc; - - cpc.reason = __get_cp_reason(sbi); - - down_write(&sbi->gc_lock); - err = f2fs_write_checkpoint(sbi, &cpc); - up_write(&sbi->gc_lock); - } - f2fs_trace_ios(NULL, 1); + if (sync) + err = f2fs_issue_checkpoint(sbi); return err; } @@ -1369,6 +1468,10 @@ static int f2fs_freeze(struct super_block *sb) /* must be clean, since sync_filesystem() was already called */ if (is_sbi_flag_set(F2FS_SB(sb), SBI_IS_DIRTY)) return -EINVAL; + + /* ensure no checkpoint required */ + if (!llist_empty(&F2FS_SB(sb)->cprc_info.issue_list)) + return -EINVAL; return 0; } @@ -1539,6 +1642,9 @@ static inline void f2fs_show_compress_options(struct seq_file *seq, } seq_printf(seq, ",compress_algorithm=%s", algtype); + if (F2FS_OPTION(sbi).compress_level) + seq_printf(seq, ":%d", F2FS_OPTION(sbi).compress_level); + seq_printf(seq, ",compress_log_size=%u", F2FS_OPTION(sbi).compress_log_size); @@ -1674,6 +1780,10 @@ static int f2fs_show_options(struct seq_file *seq, struct dentry *root) if (test_opt(sbi, DISABLE_CHECKPOINT)) seq_printf(seq, ",checkpoint=disable:%u", F2FS_OPTION(sbi).unusable_cap); + if (test_opt(sbi, MERGE_CHECKPOINT)) + seq_puts(seq, ",checkpoint_merge"); + else + seq_puts(seq, ",nocheckpoint_merge"); if (F2FS_OPTION(sbi).fsync_mode == FSYNC_MODE_POSIX) seq_printf(seq, ",fsync_mode=%s", "posix"); else if (F2FS_OPTION(sbi).fsync_mode == FSYNC_MODE_STRICT) @@ -1796,6 +1906,9 @@ restore_flag: static void f2fs_enable_checkpoint(struct f2fs_sb_info *sbi) { + /* we should flush all the data to keep data consistency */ + sync_inodes_sb(sbi->sb); + down_write(&sbi->gc_lock); f2fs_dirty_to_prefree(sbi); @@ -1954,6 +2067,19 @@ static int f2fs_remount(struct super_block *sb, int *flags, char *data) } } + if (!test_opt(sbi, DISABLE_CHECKPOINT) && + test_opt(sbi, MERGE_CHECKPOINT)) { + err = f2fs_start_ckpt_thread(sbi); + if (err) { + f2fs_err(sbi, + "Failed to start F2FS issue_checkpoint_thread (%d)", + err); + goto restore_gc; + } + } else { + f2fs_stop_ckpt_thread(sbi); + } + /* * We stop issue flush thread if FS is mounted as RO * or if flush_merge is not passed in mount option. @@ -2638,10 +2764,10 @@ static const struct export_operations f2fs_export_ops = { .get_parent = f2fs_get_parent, }; -static loff_t max_file_blocks(void) +loff_t max_file_blocks(struct inode *inode) { loff_t result = 0; - loff_t leaf_count = DEF_ADDRS_PER_BLOCK; + loff_t leaf_count; /* * note: previously, result is equal to (DEF_ADDRS_PER_INODE - @@ -2650,6 +2776,11 @@ static loff_t max_file_blocks(void) * result as zero. */ + if (inode && f2fs_compressed_file(inode)) + leaf_count = ADDRS_PER_BLOCK(inode); + else + leaf_count = DEF_ADDRS_PER_BLOCK; + /* two direct node blocks */ result += (leaf_count * 2); @@ -3533,8 +3664,7 @@ try_onemore: if (err) goto free_options; - sbi->max_file_blocks = max_file_blocks(); - sb->s_maxbytes = sbi->max_file_blocks << + sb->s_maxbytes = max_file_blocks(NULL) << le32_to_cpu(raw_super->log_blocksize); sb->s_max_links = F2FS_LINK_MAX; @@ -3701,6 +3831,19 @@ try_onemore: f2fs_init_fsync_node_info(sbi); + /* setup checkpoint request control and start checkpoint issue thread */ + f2fs_init_ckpt_req_control(sbi); + if (!test_opt(sbi, DISABLE_CHECKPOINT) && + test_opt(sbi, MERGE_CHECKPOINT)) { + err = f2fs_start_ckpt_thread(sbi); + if (err) { + f2fs_err(sbi, + "Failed to start F2FS issue_checkpoint_thread (%d)", + err); + goto stop_ckpt_thread; + } + } + /* setup f2fs internal modules */ err = f2fs_build_segment_manager(sbi); if (err) { @@ -3786,12 +3929,10 @@ try_onemore: * previous checkpoint was not done by clean system shutdown. */ if (f2fs_hw_is_readonly(sbi)) { - if (!is_set_ckpt_flags(sbi, CP_UMOUNT_FLAG)) { - err = -EROFS; + if (!is_set_ckpt_flags(sbi, CP_UMOUNT_FLAG)) f2fs_err(sbi, "Need to recover fsync data, but write access unavailable"); - goto free_meta; - } - f2fs_info(sbi, "write access unavailable, skipping recovery"); + else + f2fs_info(sbi, "write access unavailable, skipping recovery"); goto reset_checkpoint; } @@ -3910,6 +4051,8 @@ free_nm: free_sm: f2fs_destroy_segment_manager(sbi); f2fs_destroy_post_read_wq(sbi); +stop_ckpt_thread: + f2fs_stop_ckpt_thread(sbi); free_devices: destroy_device_list(sbi); kvfree(sbi->ckpt); @@ -4024,8 +4167,6 @@ static int __init init_f2fs_fs(void) return -EINVAL; } - f2fs_build_trace_ios(); - err = init_inodecache(); if (err) goto fail; @@ -4118,7 +4259,6 @@ static void __exit exit_f2fs_fs(void) f2fs_destroy_segment_manager_caches(); f2fs_destroy_node_manager_caches(); destroy_inodecache(); - f2fs_destroy_trace_ios(); } module_init(init_f2fs_fs) diff --git a/fs/f2fs/sysfs.c b/fs/f2fs/sysfs.c index 30bae57428d1..e38a7f6921dd 100644 --- a/fs/f2fs/sysfs.c +++ b/fs/f2fs/sysfs.c @@ -11,6 +11,7 @@ #include <linux/f2fs_fs.h> #include <linux/seq_file.h> #include <linux/unicode.h> +#include <linux/ioprio.h> #include "f2fs.h" #include "segment.h" @@ -34,6 +35,7 @@ enum { FAULT_INFO_TYPE, /* struct f2fs_fault_info */ #endif RESERVED_BLOCKS, /* struct f2fs_sb_info */ + CPRC_INFO, /* struct ckpt_req_control */ }; struct f2fs_attr { @@ -70,6 +72,8 @@ static unsigned char *__struct_ptr(struct f2fs_sb_info *sbi, int struct_type) else if (struct_type == STAT_INFO) return (unsigned char *)F2FS_STAT(sbi); #endif + else if (struct_type == CPRC_INFO) + return (unsigned char *)&sbi->cprc_info; return NULL; } @@ -96,6 +100,12 @@ static ssize_t lifetime_write_kbytes_show(struct f2fs_attr *a, sbi->sectors_written_start) >> 1))); } +static ssize_t sb_status_show(struct f2fs_attr *a, + struct f2fs_sb_info *sbi, char *buf) +{ + return sprintf(buf, "%lx\n", sbi->s_flag); +} + static ssize_t features_show(struct f2fs_attr *a, struct f2fs_sb_info *sbi, char *buf) { @@ -255,6 +265,23 @@ static ssize_t f2fs_sbi_show(struct f2fs_attr *a, return len; } + if (!strcmp(a->attr.name, "ckpt_thread_ioprio")) { + struct ckpt_req_control *cprc = &sbi->cprc_info; + int len = 0; + int class = IOPRIO_PRIO_CLASS(cprc->ckpt_thread_ioprio); + int data = IOPRIO_PRIO_DATA(cprc->ckpt_thread_ioprio); + + if (class == IOPRIO_CLASS_RT) + len += scnprintf(buf + len, PAGE_SIZE - len, "rt,"); + else if (class == IOPRIO_CLASS_BE) + len += scnprintf(buf + len, PAGE_SIZE - len, "be,"); + else + return -EINVAL; + + len += scnprintf(buf + len, PAGE_SIZE - len, "%d\n", data); + return len; + } + ui = (unsigned int *)(ptr + a->offset); return sprintf(buf, "%u\n", *ui); @@ -308,6 +335,38 @@ out: return ret ? ret : count; } + if (!strcmp(a->attr.name, "ckpt_thread_ioprio")) { + const char *name = strim((char *)buf); + struct ckpt_req_control *cprc = &sbi->cprc_info; + int class; + long data; + int ret; + + if (!strncmp(name, "rt,", 3)) + class = IOPRIO_CLASS_RT; + else if (!strncmp(name, "be,", 3)) + class = IOPRIO_CLASS_BE; + else + return -EINVAL; + + name += 3; + ret = kstrtol(name, 10, &data); + if (ret) + return ret; + if (data >= IOPRIO_BE_NR || data < 0) + return -EINVAL; + + cprc->ckpt_thread_ioprio = IOPRIO_PRIO_VALUE(class, data); + if (test_opt(sbi, MERGE_CHECKPOINT)) { + ret = set_task_ioprio(cprc->f2fs_issue_ckpt, + cprc->ckpt_thread_ioprio); + if (ret) + return ret; + } + + return count; + } + ui = (unsigned int *)(ptr + a->offset); ret = kstrtoul(skip_spaces(buf), 0, &t); @@ -567,6 +626,7 @@ F2FS_RW_ATTR(FAULT_INFO_TYPE, f2fs_fault_info, inject_type, inject_type); #endif F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, data_io_flag, data_io_flag); F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, node_io_flag, node_io_flag); +F2FS_RW_ATTR(CPRC_INFO, ckpt_req_control, ckpt_thread_ioprio, ckpt_thread_ioprio); F2FS_GENERAL_RO_ATTR(dirty_segments); F2FS_GENERAL_RO_ATTR(free_segments); F2FS_GENERAL_RO_ATTR(lifetime_write_kbytes); @@ -652,6 +712,7 @@ static struct attribute *f2fs_attrs[] = { #endif ATTR_LIST(data_io_flag), ATTR_LIST(node_io_flag), + ATTR_LIST(ckpt_thread_ioprio), ATTR_LIST(dirty_segments), ATTR_LIST(free_segments), ATTR_LIST(unusable), @@ -702,6 +763,13 @@ static struct attribute *f2fs_feat_attrs[] = { }; ATTRIBUTE_GROUPS(f2fs_feat); +F2FS_GENERAL_RO_ATTR(sb_status); +static struct attribute *f2fs_stat_attrs[] = { + ATTR_LIST(sb_status), + NULL, +}; +ATTRIBUTE_GROUPS(f2fs_stat); + static const struct sysfs_ops f2fs_attr_ops = { .show = f2fs_attr_show, .store = f2fs_attr_store, @@ -730,6 +798,44 @@ static struct kobject f2fs_feat = { .kset = &f2fs_kset, }; +static ssize_t f2fs_stat_attr_show(struct kobject *kobj, + struct attribute *attr, char *buf) +{ + struct f2fs_sb_info *sbi = container_of(kobj, struct f2fs_sb_info, + s_stat_kobj); + struct f2fs_attr *a = container_of(attr, struct f2fs_attr, attr); + + return a->show ? a->show(a, sbi, buf) : 0; +} + +static ssize_t f2fs_stat_attr_store(struct kobject *kobj, struct attribute *attr, + const char *buf, size_t len) +{ + struct f2fs_sb_info *sbi = container_of(kobj, struct f2fs_sb_info, + s_stat_kobj); + struct f2fs_attr *a = container_of(attr, struct f2fs_attr, attr); + + return a->store ? a->store(a, sbi, buf, len) : 0; +} + +static void f2fs_stat_kobj_release(struct kobject *kobj) +{ + struct f2fs_sb_info *sbi = container_of(kobj, struct f2fs_sb_info, + s_stat_kobj); + complete(&sbi->s_stat_kobj_unregister); +} + +static const struct sysfs_ops f2fs_stat_attr_ops = { + .show = f2fs_stat_attr_show, + .store = f2fs_stat_attr_store, +}; + +static struct kobj_type f2fs_stat_ktype = { + .default_groups = f2fs_stat_groups, + .sysfs_ops = &f2fs_stat_attr_ops, + .release = f2fs_stat_kobj_release, +}; + static int __maybe_unused segment_info_seq_show(struct seq_file *seq, void *offset) { @@ -936,11 +1042,15 @@ int f2fs_register_sysfs(struct f2fs_sb_info *sbi) init_completion(&sbi->s_kobj_unregister); err = kobject_init_and_add(&sbi->s_kobj, &f2fs_sb_ktype, NULL, "%s", sb->s_id); - if (err) { - kobject_put(&sbi->s_kobj); - wait_for_completion(&sbi->s_kobj_unregister); - return err; - } + if (err) + goto put_sb_kobj; + + sbi->s_stat_kobj.kset = &f2fs_kset; + init_completion(&sbi->s_stat_kobj_unregister); + err = kobject_init_and_add(&sbi->s_stat_kobj, &f2fs_stat_ktype, + &sbi->s_kobj, "stat"); + if (err) + goto put_stat_kobj; if (f2fs_proc_root) sbi->s_proc = proc_mkdir(sb->s_id, f2fs_proc_root); @@ -956,6 +1066,13 @@ int f2fs_register_sysfs(struct f2fs_sb_info *sbi) victim_bits_seq_show, sb); } return 0; +put_stat_kobj: + kobject_put(&sbi->s_stat_kobj); + wait_for_completion(&sbi->s_stat_kobj_unregister); +put_sb_kobj: + kobject_put(&sbi->s_kobj); + wait_for_completion(&sbi->s_kobj_unregister); + return err; } void f2fs_unregister_sysfs(struct f2fs_sb_info *sbi) @@ -967,6 +1084,11 @@ void f2fs_unregister_sysfs(struct f2fs_sb_info *sbi) remove_proc_entry("victim_bits", sbi->s_proc); remove_proc_entry(sbi->sb->s_id, f2fs_proc_root); } + + kobject_del(&sbi->s_stat_kobj); + kobject_put(&sbi->s_stat_kobj); + wait_for_completion(&sbi->s_stat_kobj_unregister); + kobject_del(&sbi->s_kobj); kobject_put(&sbi->s_kobj); wait_for_completion(&sbi->s_kobj_unregister); diff --git a/fs/f2fs/trace.c b/fs/f2fs/trace.c deleted file mode 100644 index d0ab533a9ce8..000000000000 --- a/fs/f2fs/trace.c +++ /dev/null @@ -1,165 +0,0 @@ -// SPDX-License-Identifier: GPL-2.0 -/* - * f2fs IO tracer - * - * Copyright (c) 2014 Motorola Mobility - * Copyright (c) 2014 Jaegeuk Kim <jaegeuk@kernel.org> - */ -#include <linux/fs.h> -#include <linux/f2fs_fs.h> -#include <linux/sched.h> -#include <linux/radix-tree.h> - -#include "f2fs.h" -#include "trace.h" - -static RADIX_TREE(pids, GFP_ATOMIC); -static spinlock_t pids_lock; -static struct last_io_info last_io; - -static inline void __print_last_io(void) -{ - if (!last_io.len) - return; - - trace_printk("%3x:%3x %4x %-16s %2x %5x %5x %12x %4x\n", - last_io.major, last_io.minor, - last_io.pid, "----------------", - last_io.type, - last_io.fio.op, last_io.fio.op_flags, - last_io.fio.new_blkaddr, - last_io.len); - memset(&last_io, 0, sizeof(last_io)); -} - -static int __file_type(struct inode *inode, pid_t pid) -{ - if (f2fs_is_atomic_file(inode)) - return __ATOMIC_FILE; - else if (f2fs_is_volatile_file(inode)) - return __VOLATILE_FILE; - else if (S_ISDIR(inode->i_mode)) - return __DIR_FILE; - else if (inode->i_ino == F2FS_NODE_INO(F2FS_I_SB(inode))) - return __NODE_FILE; - else if (inode->i_ino == F2FS_META_INO(F2FS_I_SB(inode))) - return __META_FILE; - else if (pid) - return __NORMAL_FILE; - else - return __MISC_FILE; -} - -void f2fs_trace_pid(struct page *page) -{ - struct inode *inode = page->mapping->host; - pid_t pid = task_pid_nr(current); - void *p; - - set_page_private(page, (unsigned long)pid); - -retry: - if (radix_tree_preload(GFP_NOFS)) - return; - - spin_lock(&pids_lock); - p = radix_tree_lookup(&pids, pid); - if (p == current) - goto out; - if (p) - radix_tree_delete(&pids, pid); - - if (radix_tree_insert(&pids, pid, current)) { - spin_unlock(&pids_lock); - radix_tree_preload_end(); - cond_resched(); - goto retry; - } - - trace_printk("%3x:%3x %4x %-16s\n", - MAJOR(inode->i_sb->s_dev), MINOR(inode->i_sb->s_dev), - pid, current->comm); -out: - spin_unlock(&pids_lock); - radix_tree_preload_end(); -} - -void f2fs_trace_ios(struct f2fs_io_info *fio, int flush) -{ - struct inode *inode; - pid_t pid; - int major, minor; - - if (flush) { - __print_last_io(); - return; - } - - inode = fio->page->mapping->host; - pid = page_private(fio->page); - - major = MAJOR(inode->i_sb->s_dev); - minor = MINOR(inode->i_sb->s_dev); - - if (last_io.major == major && last_io.minor == minor && - last_io.pid == pid && - last_io.type == __file_type(inode, pid) && - last_io.fio.op == fio->op && - last_io.fio.op_flags == fio->op_flags && - last_io.fio.new_blkaddr + last_io.len == - fio->new_blkaddr) { - last_io.len++; - return; - } - - __print_last_io(); - - last_io.major = major; - last_io.minor = minor; - last_io.pid = pid; - last_io.type = __file_type(inode, pid); - last_io.fio = *fio; - last_io.len = 1; - return; -} - -void f2fs_build_trace_ios(void) -{ - spin_lock_init(&pids_lock); -} - -#define PIDVEC_SIZE 128 -static unsigned int gang_lookup_pids(pid_t *results, unsigned long first_index, - unsigned int max_items) -{ - struct radix_tree_iter iter; - void **slot; - unsigned int ret = 0; - - if (unlikely(!max_items)) - return 0; - - radix_tree_for_each_slot(slot, &pids, &iter, first_index) { - results[ret] = iter.index; - if (++ret == max_items) - break; - } - return ret; -} - -void f2fs_destroy_trace_ios(void) -{ - pid_t pid[PIDVEC_SIZE]; - pid_t next_pid = 0; - unsigned int found; - - spin_lock(&pids_lock); - while ((found = gang_lookup_pids(pid, next_pid, PIDVEC_SIZE))) { - unsigned idx; - - next_pid = pid[found - 1] + 1; - for (idx = 0; idx < found; idx++) - radix_tree_delete(&pids, pid[idx]); - } - spin_unlock(&pids_lock); -} diff --git a/fs/f2fs/trace.h b/fs/f2fs/trace.h deleted file mode 100644 index 789f6aa727fc..000000000000 --- a/fs/f2fs/trace.h +++ /dev/null @@ -1,43 +0,0 @@ -/* SPDX-License-Identifier: GPL-2.0 */ -/* - * f2fs IO tracer - * - * Copyright (c) 2014 Motorola Mobility - * Copyright (c) 2014 Jaegeuk Kim <jaegeuk@kernel.org> - */ -#ifndef __F2FS_TRACE_H__ -#define __F2FS_TRACE_H__ - -#ifdef CONFIG_F2FS_IO_TRACE -#include <trace/events/f2fs.h> - -enum file_type { - __NORMAL_FILE, - __DIR_FILE, - __NODE_FILE, - __META_FILE, - __ATOMIC_FILE, - __VOLATILE_FILE, - __MISC_FILE, -}; - -struct last_io_info { - int major, minor; - pid_t pid; - enum file_type type; - struct f2fs_io_info fio; - block_t len; -}; - -extern void f2fs_trace_pid(struct page *); -extern void f2fs_trace_ios(struct f2fs_io_info *, int); -extern void f2fs_build_trace_ios(void); -extern void f2fs_destroy_trace_ios(void); -#else -#define f2fs_trace_pid(p) -#define f2fs_trace_ios(i, n) -#define f2fs_build_trace_ios() -#define f2fs_destroy_trace_ios() - -#endif -#endif /* __F2FS_TRACE_H__ */ diff --git a/fs/f2fs/xattr.c b/fs/f2fs/xattr.c index 65afcc3cc68a..490f843ec3bf 100644 --- a/fs/f2fs/xattr.c +++ b/fs/f2fs/xattr.c @@ -64,6 +64,7 @@ static int f2fs_xattr_generic_get(const struct xattr_handler *handler, } static int f2fs_xattr_generic_set(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *unused, struct inode *inode, const char *name, const void *value, size_t size, int flags) @@ -107,6 +108,7 @@ static int f2fs_xattr_advise_get(const struct xattr_handler *handler, } static int f2fs_xattr_advise_set(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *unused, struct inode *inode, const char *name, const void *value, size_t size, int flags) @@ -114,7 +116,7 @@ static int f2fs_xattr_advise_set(const struct xattr_handler *handler, unsigned char old_advise = F2FS_I(inode)->i_advise; unsigned char new_advise; - if (!inode_owner_or_capable(inode)) + if (!inode_owner_or_capable(&init_user_ns, inode)) return -EPERM; if (value == NULL) return -EINVAL; @@ -327,7 +329,7 @@ static int lookup_all_xattrs(struct inode *inode, struct page *ipage, void *last_addr = NULL; nid_t xnid = F2FS_I(inode)->i_xattr_nid; unsigned int inline_size = inline_xattr_size(inode); - int err = 0; + int err; if (!xnid && !inline_size) return -ENODATA; @@ -515,7 +517,7 @@ int f2fs_getxattr(struct inode *inode, int index, const char *name, void *buffer, size_t buffer_size, struct page *ipage) { struct f2fs_xattr_entry *entry = NULL; - int error = 0; + int error; unsigned int size, len; void *base_addr = NULL; int base_size; @@ -562,7 +564,7 @@ ssize_t f2fs_listxattr(struct dentry *dentry, char *buffer, size_t buffer_size) struct inode *inode = d_inode(dentry); struct f2fs_xattr_entry *entry; void *base_addr, *last_base_addr; - int error = 0; + int error; size_t rest = buffer_size; down_read(&F2FS_I(inode)->i_xattr_sem); @@ -632,7 +634,7 @@ static int __f2fs_setxattr(struct inode *inode, int index, int found, newsize; size_t len; __u32 new_hsize; - int error = 0; + int error; if (name == NULL) return -EINVAL; @@ -673,7 +675,7 @@ static int __f2fs_setxattr(struct inode *inode, int index, } if (value && f2fs_xattr_value_same(here, value, size)) - goto exit; + goto same; } else if ((flags & XATTR_REPLACE)) { error = -ENODATA; goto exit; @@ -738,17 +740,20 @@ static int __f2fs_setxattr(struct inode *inode, int index, if (error) goto exit; - if (is_inode_flag_set(inode, FI_ACL_MODE)) { - inode->i_mode = F2FS_I(inode)->i_acl_mode; - inode->i_ctime = current_time(inode); - clear_inode_flag(inode, FI_ACL_MODE); - } if (index == F2FS_XATTR_INDEX_ENCRYPTION && !strcmp(name, F2FS_XATTR_NAME_ENCRYPTION_CONTEXT)) f2fs_set_encrypted_inode(inode); f2fs_mark_inode_dirty_sync(inode, true); if (!error && S_ISDIR(inode->i_mode)) set_sbi_flag(F2FS_I_SB(inode), SBI_NEED_CP); + +same: + if (is_inode_flag_set(inode, FI_ACL_MODE)) { + inode->i_mode = F2FS_I(inode)->i_acl_mode; + inode->i_ctime = current_time(inode); + clear_inode_flag(inode, FI_ACL_MODE); + } + exit: kfree(base_addr); return error; diff --git a/fs/fat/fat.h b/fs/fat/fat.h index 922a0c6ba46c..02d4d4234956 100644 --- a/fs/fat/fat.h +++ b/fs/fat/fat.h @@ -397,9 +397,11 @@ extern long fat_generic_ioctl(struct file *filp, unsigned int cmd, unsigned long arg); extern const struct file_operations fat_file_operations; extern const struct inode_operations fat_file_inode_operations; -extern int fat_setattr(struct dentry *dentry, struct iattr *attr); +extern int fat_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *attr); extern void fat_truncate_blocks(struct inode *inode, loff_t offset); -extern int fat_getattr(const struct path *path, struct kstat *stat, +extern int fat_getattr(struct user_namespace *mnt_userns, + const struct path *path, struct kstat *stat, u32 request_mask, unsigned int flags); extern int fat_file_fsync(struct file *file, loff_t start, loff_t end, int datasync); diff --git a/fs/fat/file.c b/fs/fat/file.c index f9ee27cf4d7c..13855ba49cd9 100644 --- a/fs/fat/file.c +++ b/fs/fat/file.c @@ -95,7 +95,7 @@ static int fat_ioctl_set_attributes(struct file *file, u32 __user *user_attr) goto out_unlock_inode; /* This MUST be done before doing anything irreversible... */ - err = fat_setattr(file->f_path.dentry, &ia); + err = fat_setattr(file_mnt_user_ns(file), file->f_path.dentry, &ia); if (err) goto out_unlock_inode; @@ -195,7 +195,7 @@ int fat_file_fsync(struct file *filp, loff_t start, loff_t end, int datasync) if (err) return err; - return blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL); + return blkdev_issue_flush(inode->i_sb->s_bdev); } @@ -394,11 +394,11 @@ void fat_truncate_blocks(struct inode *inode, loff_t offset) fat_flush_inodes(inode->i_sb, inode, NULL); } -int fat_getattr(const struct path *path, struct kstat *stat, - u32 request_mask, unsigned int flags) +int fat_getattr(struct user_namespace *mnt_userns, const struct path *path, + struct kstat *stat, u32 request_mask, unsigned int flags) { struct inode *inode = d_inode(path->dentry); - generic_fillattr(inode, stat); + generic_fillattr(mnt_userns, inode, stat); stat->blksize = MSDOS_SB(inode->i_sb)->cluster_size; if (MSDOS_SB(inode->i_sb)->options.nfs == FAT_NFS_NOSTALE_RO) { @@ -447,12 +447,13 @@ static int fat_sanitize_mode(const struct msdos_sb_info *sbi, return 0; } -static int fat_allow_set_time(struct msdos_sb_info *sbi, struct inode *inode) +static int fat_allow_set_time(struct user_namespace *mnt_userns, + struct msdos_sb_info *sbi, struct inode *inode) { umode_t allow_utime = sbi->options.allow_utime; - if (!uid_eq(current_fsuid(), inode->i_uid)) { - if (in_group_p(inode->i_gid)) + if (!uid_eq(current_fsuid(), i_uid_into_mnt(mnt_userns, inode))) { + if (in_group_p(i_gid_into_mnt(mnt_userns, inode))) allow_utime >>= 3; if (allow_utime & MAY_WRITE) return 1; @@ -466,7 +467,8 @@ static int fat_allow_set_time(struct msdos_sb_info *sbi, struct inode *inode) /* valid file mode bits */ #define FAT_VALID_MODE (S_IFREG | S_IFDIR | S_IRWXUGO) -int fat_setattr(struct dentry *dentry, struct iattr *attr) +int fat_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *attr) { struct msdos_sb_info *sbi = MSDOS_SB(dentry->d_sb); struct inode *inode = d_inode(dentry); @@ -476,11 +478,11 @@ int fat_setattr(struct dentry *dentry, struct iattr *attr) /* Check for setting the inode time. */ ia_valid = attr->ia_valid; if (ia_valid & TIMES_SET_FLAGS) { - if (fat_allow_set_time(sbi, inode)) + if (fat_allow_set_time(mnt_userns, sbi, inode)) attr->ia_valid &= ~TIMES_SET_FLAGS; } - error = setattr_prepare(dentry, attr); + error = setattr_prepare(mnt_userns, dentry, attr); attr->ia_valid = ia_valid; if (error) { if (sbi->options.quiet) @@ -550,7 +552,7 @@ int fat_setattr(struct dentry *dentry, struct iattr *attr) fat_truncate_time(inode, &attr->ia_mtime, S_MTIME); attr->ia_valid &= ~(ATTR_ATIME|ATTR_CTIME|ATTR_MTIME); - setattr_copy(inode, attr); + setattr_copy(mnt_userns, inode, attr); mark_inode_dirty(inode); out: return error; diff --git a/fs/fat/misc.c b/fs/fat/misc.c index f1b2a1fc2a6a..18a50a46b57f 100644 --- a/fs/fat/misc.c +++ b/fs/fat/misc.c @@ -329,22 +329,23 @@ EXPORT_SYMBOL_GPL(fat_truncate_time); int fat_update_time(struct inode *inode, struct timespec64 *now, int flags) { - int iflags = I_DIRTY_TIME; - bool dirty = false; + int dirty_flags = 0; if (inode->i_ino == MSDOS_ROOT_INO) return 0; - fat_truncate_time(inode, now, flags); - if (flags & S_VERSION) - dirty = inode_maybe_inc_iversion(inode, false); - if ((flags & (S_ATIME | S_CTIME | S_MTIME)) && - !(inode->i_sb->s_flags & SB_LAZYTIME)) - dirty = true; + if (flags & (S_ATIME | S_CTIME | S_MTIME)) { + fat_truncate_time(inode, now, flags); + if (inode->i_sb->s_flags & SB_LAZYTIME) + dirty_flags |= I_DIRTY_TIME; + else + dirty_flags |= I_DIRTY_SYNC; + } + + if ((flags & S_VERSION) && inode_maybe_inc_iversion(inode, false)) + dirty_flags |= I_DIRTY_SYNC; - if (dirty) - iflags |= I_DIRTY_SYNC; - __mark_inode_dirty(inode, iflags); + __mark_inode_dirty(inode, dirty_flags); return 0; } EXPORT_SYMBOL_GPL(fat_update_time); diff --git a/fs/fat/namei_msdos.c b/fs/fat/namei_msdos.c index 9d062886fbc1..efba301d68ae 100644 --- a/fs/fat/namei_msdos.c +++ b/fs/fat/namei_msdos.c @@ -261,8 +261,8 @@ static int msdos_add_entry(struct inode *dir, const unsigned char *name, } /***** Create a file */ -static int msdos_create(struct inode *dir, struct dentry *dentry, umode_t mode, - bool excl) +static int msdos_create(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, bool excl) { struct super_block *sb = dir->i_sb; struct inode *inode = NULL; @@ -339,7 +339,8 @@ out: } /***** Make a directory */ -static int msdos_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) +static int msdos_mkdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { struct super_block *sb = dir->i_sb; struct fat_slot_info sinfo; @@ -593,7 +594,8 @@ error_inode: } /***** Rename, a wrapper for rename_same_dir & rename_diff_dir */ -static int msdos_rename(struct inode *old_dir, struct dentry *old_dentry, +static int msdos_rename(struct user_namespace *mnt_userns, + struct inode *old_dir, struct dentry *old_dentry, struct inode *new_dir, struct dentry *new_dentry, unsigned int flags) { @@ -665,7 +667,7 @@ static struct file_system_type msdos_fs_type = { .name = "msdos", .mount = msdos_mount, .kill_sb = kill_block_super, - .fs_flags = FS_REQUIRES_DEV, + .fs_flags = FS_REQUIRES_DEV | FS_ALLOW_IDMAP, }; MODULE_ALIAS_FS("msdos"); diff --git a/fs/fat/namei_vfat.c b/fs/fat/namei_vfat.c index 0cdd0fb9f742..5369d82e0bfb 100644 --- a/fs/fat/namei_vfat.c +++ b/fs/fat/namei_vfat.c @@ -756,8 +756,8 @@ error: return ERR_PTR(err); } -static int vfat_create(struct inode *dir, struct dentry *dentry, umode_t mode, - bool excl) +static int vfat_create(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, bool excl) { struct super_block *sb = dir->i_sb; struct inode *inode; @@ -846,7 +846,8 @@ out: return err; } -static int vfat_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) +static int vfat_mkdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { struct super_block *sb = dir->i_sb; struct inode *inode; @@ -892,9 +893,9 @@ out: return err; } -static int vfat_rename(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry, - unsigned int flags) +static int vfat_rename(struct user_namespace *mnt_userns, struct inode *old_dir, + struct dentry *old_dentry, struct inode *new_dir, + struct dentry *new_dentry, unsigned int flags) { struct buffer_head *dotdot_bh; struct msdos_dir_entry *dotdot_de; @@ -1062,7 +1063,7 @@ static struct file_system_type vfat_fs_type = { .name = "vfat", .mount = vfat_mount, .kill_sb = kill_block_super, - .fs_flags = FS_REQUIRES_DEV, + .fs_flags = FS_REQUIRES_DEV | FS_ALLOW_IDMAP, }; MODULE_ALIAS_FS("vfat"); diff --git a/fs/fcntl.c b/fs/fcntl.c index 05b36b28f2e8..dfc72f15be7f 100644 --- a/fs/fcntl.c +++ b/fs/fcntl.c @@ -25,6 +25,7 @@ #include <linux/user_namespace.h> #include <linux/memfd.h> #include <linux/compat.h> +#include <linux/mount.h> #include <linux/poll.h> #include <asm/siginfo.h> @@ -46,7 +47,7 @@ static int setfl(int fd, struct file * filp, unsigned long arg) /* O_NOATIME can only be set by the owner or superuser */ if ((arg & O_NOATIME) && !(filp->f_flags & O_NOATIME)) - if (!inode_owner_or_capable(inode)) + if (!inode_owner_or_capable(file_mnt_user_ns(filp), inode)) return -EPERM; /* required for strict SunOS emulation */ @@ -148,11 +149,15 @@ void f_delown(struct file *filp) pid_t f_getown(struct file *filp) { - pid_t pid; + pid_t pid = 0; read_lock(&filp->f_owner.lock); - pid = pid_vnr(filp->f_owner.pid); - if (filp->f_owner.pid_type == PIDTYPE_PGID) - pid = -pid; + rcu_read_lock(); + if (pid_task(filp->f_owner.pid, filp->f_owner.pid_type)) { + pid = pid_vnr(filp->f_owner.pid); + if (filp->f_owner.pid_type == PIDTYPE_PGID) + pid = -pid; + } + rcu_read_unlock(); read_unlock(&filp->f_owner.lock); return pid; } @@ -200,11 +205,14 @@ static int f_setown_ex(struct file *filp, unsigned long arg) static int f_getown_ex(struct file *filp, unsigned long arg) { struct f_owner_ex __user *owner_p = (void __user *)arg; - struct f_owner_ex owner; + struct f_owner_ex owner = {}; int ret = 0; read_lock(&filp->f_owner.lock); - owner.pid = pid_vnr(filp->f_owner.pid); + rcu_read_lock(); + if (pid_task(filp->f_owner.pid, filp->f_owner.pid_type)) + owner.pid = pid_vnr(filp->f_owner.pid); + rcu_read_unlock(); switch (filp->f_owner.pid_type) { case PIDTYPE_PID: owner.type = F_OWNER_TID; diff --git a/fs/fhandle.c b/fs/fhandle.c index 01263ffbc4c0..ec6feeccc276 100644 --- a/fs/fhandle.c +++ b/fs/fhandle.c @@ -173,7 +173,7 @@ static int handle_to_path(int mountdirfd, struct file_handle __user *ufh, /* * With handle we don't look at the execute bit on the - * the directory. Ideally we would like CAP_DAC_SEARCH. + * directory. Ideally we would like CAP_DAC_SEARCH. * But we don't have that */ if (!capable(CAP_DAC_READ_SEARCH)) { diff --git a/fs/file.c b/fs/file.c index dab120b71e44..f3a4bac2cbe9 100644 --- a/fs/file.c +++ b/fs/file.c @@ -22,6 +22,8 @@ #include <linux/close_range.h> #include <net/sock.h> +#include "internal.h" + unsigned int sysctl_nr_open __read_mostly = 1024*1024; unsigned int sysctl_nr_open_min = BITS_PER_LONG; /* our min() is unusable in constant expressions ;-/ */ @@ -732,36 +734,48 @@ int __close_range(unsigned fd, unsigned max_fd, unsigned int flags) } /* - * variant of close_fd that gets a ref on the file for later fput. - * The caller must ensure that filp_close() called on the file, and then - * an fput(). + * See close_fd_get_file() below, this variant assumes current->files->file_lock + * is held. */ -int close_fd_get_file(unsigned int fd, struct file **res) +int __close_fd_get_file(unsigned int fd, struct file **res) { struct files_struct *files = current->files; struct file *file; struct fdtable *fdt; - spin_lock(&files->file_lock); fdt = files_fdtable(files); if (fd >= fdt->max_fds) - goto out_unlock; + goto out_err; file = fdt->fd[fd]; if (!file) - goto out_unlock; + goto out_err; rcu_assign_pointer(fdt->fd[fd], NULL); __put_unused_fd(files, fd); - spin_unlock(&files->file_lock); get_file(file); *res = file; return 0; - -out_unlock: - spin_unlock(&files->file_lock); +out_err: *res = NULL; return -ENOENT; } +/* + * variant of close_fd that gets a ref on the file for later fput. + * The caller must ensure that filp_close() called on the file, and then + * an fput(). + */ +int close_fd_get_file(unsigned int fd, struct file **res) +{ + struct files_struct *files = current->files; + int ret; + + spin_lock(&files->file_lock); + ret = __close_fd_get_file(fd, res); + spin_unlock(&files->file_lock); + + return ret; +} + void do_close_on_exec(struct files_struct *files) { unsigned i; diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index c41cb887eb7d..e91980f49388 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -1442,9 +1442,15 @@ static void requeue_inode(struct inode *inode, struct bdi_writeback *wb, } /* - * Write out an inode and its dirty pages. Do not update the writeback list - * linkage. That is left to the caller. The caller is also responsible for - * setting I_SYNC flag and calling inode_sync_complete() to clear it. + * Write out an inode and its dirty pages (or some of its dirty pages, depending + * on @wbc->nr_to_write), and clear the relevant dirty flags from i_state. + * + * This doesn't remove the inode from the writeback list it is on, except + * potentially to move it from b_dirty_time to b_dirty due to timestamp + * expiration. The caller is otherwise responsible for writeback list handling. + * + * The caller is also responsible for setting the I_SYNC flag beforehand and + * calling inode_sync_complete() to clear it afterwards. */ static int __writeback_single_inode(struct inode *inode, struct writeback_control *wbc) @@ -1479,7 +1485,7 @@ __writeback_single_inode(struct inode *inode, struct writeback_control *wbc) * change I_DIRTY_TIME into I_DIRTY_SYNC. */ if ((inode->i_state & I_DIRTY_TIME) && - (wbc->sync_mode == WB_SYNC_ALL || wbc->for_sync || + (wbc->sync_mode == WB_SYNC_ALL || time_after(jiffies, inode->dirtied_time_when + dirtytime_expire_interval * HZ))) { trace_writeback_lazytime(inode); @@ -1487,9 +1493,10 @@ __writeback_single_inode(struct inode *inode, struct writeback_control *wbc) } /* - * Some filesystems may redirty the inode during the writeback - * due to delalloc, clear dirty metadata flags right before - * write_inode() + * Get and clear the dirty flags from i_state. This needs to be done + * after calling writepages because some filesystems may redirty the + * inode during writepages due to delalloc. It also needs to be done + * after handling timestamp expiration, as that may dirty the inode too. */ spin_lock(&inode->i_lock); dirty = inode->i_state & I_DIRTY; @@ -1524,12 +1531,13 @@ __writeback_single_inode(struct inode *inode, struct writeback_control *wbc) } /* - * Write out an inode's dirty pages. Either the caller has an active reference - * on the inode or the inode has I_WILL_FREE set. + * Write out an inode's dirty data and metadata on-demand, i.e. separately from + * the regular batched writeback done by the flusher threads in + * writeback_sb_inodes(). @wbc controls various aspects of the write, such as + * whether it is a data-integrity sync (%WB_SYNC_ALL) or not (%WB_SYNC_NONE). * - * This function is designed to be called for writing back one inode which - * we go e.g. from filesystem. Flusher thread uses __writeback_single_inode() - * and does more profound writeback list handling in writeback_sb_inodes(). + * To prevent the inode from going away, either the caller must have a reference + * to the inode, or the inode must have I_WILL_FREE or I_FREEING set. */ static int writeback_single_inode(struct inode *inode, struct writeback_control *wbc) @@ -1544,23 +1552,23 @@ static int writeback_single_inode(struct inode *inode, WARN_ON(inode->i_state & I_WILL_FREE); if (inode->i_state & I_SYNC) { - if (wbc->sync_mode != WB_SYNC_ALL) - goto out; /* - * It's a data-integrity sync. We must wait. Since callers hold - * inode reference or inode has I_WILL_FREE set, it cannot go - * away under us. + * Writeback is already running on the inode. For WB_SYNC_NONE, + * that's enough and we can just return. For WB_SYNC_ALL, we + * must wait for the existing writeback to complete, then do + * writeback again if there's anything left. */ + if (wbc->sync_mode != WB_SYNC_ALL) + goto out; __inode_wait_for_writeback(inode); } WARN_ON(inode->i_state & I_SYNC); /* - * Skip inode if it is clean and we have no outstanding writeback in - * WB_SYNC_ALL mode. We don't want to mess with writeback lists in this - * function since flusher thread may be doing for example sync in - * parallel and if we move the inode, it could get skipped. So here we - * make sure inode is on some writeback list and leave it there unless - * we have completely cleaned the inode. + * If the inode is already fully clean, then there's nothing to do. + * + * For data-integrity syncs we also need to check whether any pages are + * still under writeback, e.g. due to prior WB_SYNC_NONE writeback. If + * there are any such pages, we'll need to wait for them. */ if (!(inode->i_state & I_DIRTY_ALL) && (wbc->sync_mode != WB_SYNC_ALL || @@ -1576,8 +1584,9 @@ static int writeback_single_inode(struct inode *inode, wb = inode_to_wb_and_lock_list(inode); spin_lock(&inode->i_lock); /* - * If inode is clean, remove it from writeback lists. Otherwise don't - * touch it. See comment above for explanation. + * If the inode is now fully clean, then it can be safely removed from + * its writeback list (if any). Otherwise the flusher threads are + * responsible for the writeback lists. */ if (!(inode->i_state & I_DIRTY_ALL)) inode_io_list_del_locked(inode, wb); @@ -2219,23 +2228,24 @@ static noinline void block_dump___mark_inode_dirty(struct inode *inode) } /** - * __mark_inode_dirty - internal function + * __mark_inode_dirty - internal function to mark an inode dirty * * @inode: inode to mark - * @flags: what kind of dirty (i.e. I_DIRTY_SYNC) + * @flags: what kind of dirty, e.g. I_DIRTY_SYNC. This can be a combination of + * multiple I_DIRTY_* flags, except that I_DIRTY_TIME can't be combined + * with I_DIRTY_PAGES. * - * Mark an inode as dirty. Callers should use mark_inode_dirty or - * mark_inode_dirty_sync. + * Mark an inode as dirty. We notify the filesystem, then update the inode's + * dirty flags. Then, if needed we add the inode to the appropriate dirty list. * - * Put the inode on the super block's dirty list. + * Most callers should use mark_inode_dirty() or mark_inode_dirty_sync() + * instead of calling this directly. * - * CAREFUL! We mark it dirty unconditionally, but move it onto the - * dirty list only if it is hashed or if it refers to a blockdev. - * If it was not hashed, it will never be added to the dirty list - * even if it is later hashed, as it will have been marked dirty already. + * CAREFUL! We only add the inode to the dirty list if it is hashed or if it + * refers to a blockdev. Unhashed inodes will never be added to the dirty list + * even if they are later hashed, as they will have been marked dirty already. * - * In short, make sure you hash any inodes _before_ you start marking - * them dirty. + * In short, ensure you hash any inodes _before_ you start marking them dirty. * * Note that for blockdevs, inode->dirtied_when represents the dirtying time of * the block-special inode (/dev/hda1) itself. And the ->dirtied_when field of @@ -2247,25 +2257,34 @@ static noinline void block_dump___mark_inode_dirty(struct inode *inode) void __mark_inode_dirty(struct inode *inode, int flags) { struct super_block *sb = inode->i_sb; - int dirtytime; + int dirtytime = 0; trace_writeback_mark_inode_dirty(inode, flags); - /* - * Don't do this for I_DIRTY_PAGES - that doesn't actually - * dirty the inode itself - */ - if (flags & (I_DIRTY_INODE | I_DIRTY_TIME)) { + if (flags & I_DIRTY_INODE) { + /* + * Notify the filesystem about the inode being dirtied, so that + * (if needed) it can update on-disk fields and journal the + * inode. This is only needed when the inode itself is being + * dirtied now. I.e. it's only needed for I_DIRTY_INODE, not + * for just I_DIRTY_PAGES or I_DIRTY_TIME. + */ trace_writeback_dirty_inode_start(inode, flags); - if (sb->s_op->dirty_inode) - sb->s_op->dirty_inode(inode, flags); - + sb->s_op->dirty_inode(inode, flags & I_DIRTY_INODE); trace_writeback_dirty_inode(inode, flags); - } - if (flags & I_DIRTY_INODE) + + /* I_DIRTY_INODE supersedes I_DIRTY_TIME. */ flags &= ~I_DIRTY_TIME; - dirtytime = flags & I_DIRTY_TIME; + } else { + /* + * Else it's either I_DIRTY_PAGES, I_DIRTY_TIME, or nothing. + * (We don't support setting both I_DIRTY_PAGES and I_DIRTY_TIME + * in one call to __mark_inode_dirty().) + */ + dirtytime = flags & I_DIRTY_TIME; + WARN_ON_ONCE(dirtytime && flags != I_DIRTY_TIME); + } /* * Paired with smp_mb() in __writeback_single_inode() for the @@ -2288,6 +2307,7 @@ void __mark_inode_dirty(struct inode *inode, int flags) inode_attach_wb(inode, NULL); + /* I_DIRTY_INODE supersedes I_DIRTY_TIME. */ if (flags & I_DIRTY_INODE) inode->i_state &= ~I_DIRTY_TIME; inode->i_state |= flags; diff --git a/fs/fuse/acl.c b/fs/fuse/acl.c index f529075a2ce8..e9c0f916349d 100644 --- a/fs/fuse/acl.c +++ b/fs/fuse/acl.c @@ -50,7 +50,8 @@ struct posix_acl *fuse_get_acl(struct inode *inode, int type) return acl; } -int fuse_set_acl(struct inode *inode, struct posix_acl *acl, int type) +int fuse_set_acl(struct user_namespace *mnt_userns, struct inode *inode, + struct posix_acl *acl, int type) { struct fuse_conn *fc = get_fuse_conn(inode); const char *name; diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c index 588f8d1240aa..c0fee830a34e 100644 --- a/fs/fuse/dev.c +++ b/fs/fuse/dev.c @@ -844,11 +844,7 @@ static int fuse_try_move_page(struct fuse_copy_state *cs, struct page **pagep) if (WARN_ON(PageMlocked(oldpage))) goto out_fallback_unlock; - err = replace_page_cache_page(oldpage, newpage, GFP_KERNEL); - if (err) { - unlock_page(newpage); - goto out_put_old; - } + replace_page_cache_page(oldpage, newpage); get_page(newpage); @@ -2233,19 +2229,21 @@ static int fuse_device_clone(struct fuse_conn *fc, struct file *new) static long fuse_dev_ioctl(struct file *file, unsigned int cmd, unsigned long arg) { - int err = -ENOTTY; + int res; + int oldfd; + struct fuse_dev *fud = NULL; - if (cmd == FUSE_DEV_IOC_CLONE) { - int oldfd; + if (_IOC_TYPE(cmd) != FUSE_DEV_IOC_MAGIC) + return -ENOTTY; - err = -EFAULT; - if (!get_user(oldfd, (__u32 __user *) arg)) { + switch (_IOC_NR(cmd)) { + case _IOC_NR(FUSE_DEV_IOC_CLONE): + res = -EFAULT; + if (!get_user(oldfd, (__u32 __user *)arg)) { struct file *old = fget(oldfd); - err = -EINVAL; + res = -EINVAL; if (old) { - struct fuse_dev *fud = NULL; - /* * Check against file->f_op because CUSE * uses the same ioctl handler. @@ -2256,14 +2254,18 @@ static long fuse_dev_ioctl(struct file *file, unsigned int cmd, if (fud) { mutex_lock(&fuse_mutex); - err = fuse_device_clone(fud->fc, file); + res = fuse_device_clone(fud->fc, file); mutex_unlock(&fuse_mutex); } fput(old); } } + break; + default: + res = -ENOTTY; + break; } - return err; + return res; } const struct file_operations fuse_dev_operations = { diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c index 78f9f209078c..06a18700a845 100644 --- a/fs/fuse/dir.c +++ b/fs/fuse/dir.c @@ -605,7 +605,8 @@ out_err: return err; } -static int fuse_mknod(struct inode *, struct dentry *, umode_t, dev_t); +static int fuse_mknod(struct user_namespace *, struct inode *, struct dentry *, + umode_t, dev_t); static int fuse_atomic_open(struct inode *dir, struct dentry *entry, struct file *file, unsigned flags, umode_t mode) @@ -645,7 +646,7 @@ out_dput: return err; mknod: - err = fuse_mknod(dir, entry, mode, 0); + err = fuse_mknod(&init_user_ns, dir, entry, mode, 0); if (err) goto out_dput; no_open: @@ -715,8 +716,8 @@ static int create_new_entry(struct fuse_mount *fm, struct fuse_args *args, return err; } -static int fuse_mknod(struct inode *dir, struct dentry *entry, umode_t mode, - dev_t rdev) +static int fuse_mknod(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *entry, umode_t mode, dev_t rdev) { struct fuse_mknod_in inarg; struct fuse_mount *fm = get_fuse_mount(dir); @@ -738,13 +739,14 @@ static int fuse_mknod(struct inode *dir, struct dentry *entry, umode_t mode, return create_new_entry(fm, &args, dir, entry, mode); } -static int fuse_create(struct inode *dir, struct dentry *entry, umode_t mode, - bool excl) +static int fuse_create(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *entry, umode_t mode, bool excl) { - return fuse_mknod(dir, entry, mode, 0); + return fuse_mknod(&init_user_ns, dir, entry, mode, 0); } -static int fuse_mkdir(struct inode *dir, struct dentry *entry, umode_t mode) +static int fuse_mkdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *entry, umode_t mode) { struct fuse_mkdir_in inarg; struct fuse_mount *fm = get_fuse_mount(dir); @@ -765,8 +767,8 @@ static int fuse_mkdir(struct inode *dir, struct dentry *entry, umode_t mode) return create_new_entry(fm, &args, dir, entry, S_IFDIR); } -static int fuse_symlink(struct inode *dir, struct dentry *entry, - const char *link) +static int fuse_symlink(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *entry, const char *link) { struct fuse_mount *fm = get_fuse_mount(dir); unsigned len = strlen(link) + 1; @@ -908,9 +910,9 @@ static int fuse_rename_common(struct inode *olddir, struct dentry *oldent, return err; } -static int fuse_rename2(struct inode *olddir, struct dentry *oldent, - struct inode *newdir, struct dentry *newent, - unsigned int flags) +static int fuse_rename2(struct user_namespace *mnt_userns, struct inode *olddir, + struct dentry *oldent, struct inode *newdir, + struct dentry *newent, unsigned int flags) { struct fuse_conn *fc = get_fuse_conn(olddir); int err; @@ -1087,7 +1089,7 @@ static int fuse_update_get_attr(struct inode *inode, struct file *file, forget_all_cached_acls(inode); err = fuse_do_getattr(inode, stat, file); } else if (stat) { - generic_fillattr(inode, stat); + generic_fillattr(&init_user_ns, inode, stat); stat->mode = fi->orig_i_mode; stat->ino = fi->orig_ino; } @@ -1249,7 +1251,8 @@ static int fuse_perm_getattr(struct inode *inode, int mask) * access request is sent. Execute permission is still checked * locally based on file mode. */ -static int fuse_permission(struct inode *inode, int mask) +static int fuse_permission(struct user_namespace *mnt_userns, + struct inode *inode, int mask) { struct fuse_conn *fc = get_fuse_conn(inode); bool refreshed = false; @@ -1280,7 +1283,7 @@ static int fuse_permission(struct inode *inode, int mask) } if (fc->default_permissions) { - err = generic_permission(inode, mask); + err = generic_permission(&init_user_ns, inode, mask); /* If permission is denied, try to refresh file attributes. This is also needed, because the root @@ -1288,7 +1291,8 @@ static int fuse_permission(struct inode *inode, int mask) if (err == -EACCES && !refreshed) { err = fuse_perm_getattr(inode, mask); if (!err) - err = generic_permission(inode, mask); + err = generic_permission(&init_user_ns, + inode, mask); } /* Note: the opposite of the above test does not @@ -1610,7 +1614,7 @@ int fuse_do_setattr(struct dentry *dentry, struct iattr *attr, if (!fc->default_permissions) attr->ia_valid |= ATTR_FORCE; - err = setattr_prepare(dentry, attr); + err = setattr_prepare(&init_user_ns, dentry, attr); if (err) return err; @@ -1756,7 +1760,8 @@ error: return err; } -static int fuse_setattr(struct dentry *entry, struct iattr *attr) +static int fuse_setattr(struct user_namespace *mnt_userns, struct dentry *entry, + struct iattr *attr) { struct inode *inode = d_inode(entry); struct fuse_conn *fc = get_fuse_conn(inode); @@ -1818,7 +1823,8 @@ static int fuse_setattr(struct dentry *entry, struct iattr *attr) return ret; } -static int fuse_getattr(const struct path *path, struct kstat *stat, +static int fuse_getattr(struct user_namespace *mnt_userns, + const struct path *path, struct kstat *stat, u32 request_mask, unsigned int flags) { struct inode *inode = d_inode(path->dentry); diff --git a/fs/fuse/fuse_i.h b/fs/fuse/fuse_i.h index 7c4b8cb93f9f..63d97a15ffde 100644 --- a/fs/fuse/fuse_i.h +++ b/fs/fuse/fuse_i.h @@ -863,6 +863,7 @@ static inline u64 fuse_get_attr_version(struct fuse_conn *fc) static inline void fuse_make_bad(struct inode *inode) { + remove_inode_hash(inode); set_bit(FUSE_I_BAD, &get_fuse_inode(inode)->state); } @@ -1180,8 +1181,8 @@ extern const struct xattr_handler *fuse_no_acl_xattr_handlers[]; struct posix_acl; struct posix_acl *fuse_get_acl(struct inode *inode, int type); -int fuse_set_acl(struct inode *inode, struct posix_acl *acl, int type); - +int fuse_set_acl(struct user_namespace *mnt_userns, struct inode *inode, + struct posix_acl *acl, int type); /* readdir.c */ int fuse_readdir(struct file *file, struct dir_context *ctx); diff --git a/fs/fuse/virtio_fs.c b/fs/fuse/virtio_fs.c index 8868ac31a3c0..4ee6f734ba83 100644 --- a/fs/fuse/virtio_fs.c +++ b/fs/fuse/virtio_fs.c @@ -1324,8 +1324,15 @@ static int virtio_fs_fill_super(struct super_block *sb, struct fs_context *fsc) /* virtiofs allocates and installs its own fuse devices */ ctx->fudptr = NULL; - if (ctx->dax) + if (ctx->dax) { + if (!fs->dax_dev) { + err = -EINVAL; + pr_err("virtio-fs: dax can't be enabled as filesystem" + " device does not support it.\n"); + goto err_free_fuse_devs; + } ctx->dax_dev = fs->dax_dev; + } err = fuse_fill_super_common(sb, ctx); if (err < 0) goto err_free_fuse_devs; diff --git a/fs/fuse/xattr.c b/fs/fuse/xattr.c index cdea18de94f7..1a7d7ace54e1 100644 --- a/fs/fuse/xattr.c +++ b/fs/fuse/xattr.c @@ -188,6 +188,7 @@ static int fuse_xattr_get(const struct xattr_handler *handler, } static int fuse_xattr_set(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *dentry, struct inode *inode, const char *name, const void *value, size_t size, int flags) @@ -214,6 +215,7 @@ static int no_xattr_get(const struct xattr_handler *handler, } static int no_xattr_set(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *dentry, struct inode *nodee, const char *name, const void *value, size_t size, int flags) diff --git a/fs/gfs2/acl.c b/fs/gfs2/acl.c index 2e939f5fe751..9165d70ead07 100644 --- a/fs/gfs2/acl.c +++ b/fs/gfs2/acl.c @@ -106,7 +106,8 @@ out: return error; } -int gfs2_set_acl(struct inode *inode, struct posix_acl *acl, int type) +int gfs2_set_acl(struct user_namespace *mnt_userns, struct inode *inode, + struct posix_acl *acl, int type) { struct gfs2_inode *ip = GFS2_I(inode); struct gfs2_holder gh; @@ -130,7 +131,7 @@ int gfs2_set_acl(struct inode *inode, struct posix_acl *acl, int type) mode = inode->i_mode; if (type == ACL_TYPE_ACCESS && acl) { - ret = posix_acl_update_mode(inode, &mode, &acl); + ret = posix_acl_update_mode(&init_user_ns, inode, &mode, &acl); if (ret) goto unlock; } diff --git a/fs/gfs2/acl.h b/fs/gfs2/acl.h index 61353a1501c5..eccc6a43326c 100644 --- a/fs/gfs2/acl.h +++ b/fs/gfs2/acl.h @@ -13,6 +13,7 @@ extern struct posix_acl *gfs2_get_acl(struct inode *inode, int type); extern int __gfs2_set_acl(struct inode *inode, struct posix_acl *acl, int type); -extern int gfs2_set_acl(struct inode *inode, struct posix_acl *acl, int type); +extern int gfs2_set_acl(struct user_namespace *mnt_userns, struct inode *inode, + struct posix_acl *acl, int type); #endif /* __ACL_DOT_H__ */ diff --git a/fs/gfs2/bmap.c b/fs/gfs2/bmap.c index 62d9081d1e26..7a358ae05185 100644 --- a/fs/gfs2/bmap.c +++ b/fs/gfs2/bmap.c @@ -1230,6 +1230,9 @@ static int gfs2_iomap_end(struct inode *inode, loff_t pos, loff_t length, gfs2_inplace_release(ip); + if (ip->i_qadata && ip->i_qadata->qa_qd_num) + gfs2_quota_unlock(ip); + if (length != written && (iomap->flags & IOMAP_F_NEW)) { /* Deallocate blocks that were just allocated. */ loff_t blockmask = i_blocksize(inode) - 1; @@ -1242,9 +1245,6 @@ static int gfs2_iomap_end(struct inode *inode, loff_t pos, loff_t length, } } - if (ip->i_qadata && ip->i_qadata->qa_qd_num) - gfs2_quota_unlock(ip); - if (unlikely(!written)) goto out_unlock; @@ -1538,13 +1538,13 @@ more_rgrps: goto out; } ret = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, - 0, rd_gh); + LM_FLAG_NODE_SCOPE, rd_gh); if (ret) goto out; /* Must be done with the rgrp glock held: */ if (gfs2_rs_active(&ip->i_res) && - rgd == ip->i_res.rs_rbm.rgd) + rgd == ip->i_res.rs_rgd) gfs2_rs_deltree(&ip->i_res); } diff --git a/fs/gfs2/file.c b/fs/gfs2/file.c index b39b339feddc..2d500f90cdac 100644 --- a/fs/gfs2/file.c +++ b/fs/gfs2/file.c @@ -238,7 +238,7 @@ static int do_gfs2_set_flags(struct file *filp, u32 reqflags, u32 mask, goto out; error = -EACCES; - if (!inode_owner_or_capable(inode)) + if (!inode_owner_or_capable(&init_user_ns, inode)) goto out; error = 0; @@ -256,7 +256,7 @@ static int do_gfs2_set_flags(struct file *filp, u32 reqflags, u32 mask, !capable(CAP_LINUX_IMMUTABLE)) goto out; if (!IS_IMMUTABLE(inode)) { - error = gfs2_permission(inode, MAY_WRITE); + error = gfs2_permission(&init_user_ns, inode, MAY_WRITE); if (error) goto out; } @@ -716,10 +716,10 @@ static int gfs2_release(struct inode *inode, struct file *file) kfree(file->private_data); file->private_data = NULL; - if (file->f_mode & FMODE_WRITE) { + if (gfs2_rs_active(&ip->i_res)) gfs2_rs_delete(ip, &inode->i_writecount); + if (file->f_mode & FMODE_WRITE) gfs2_qa_put(ip); - } return 0; } @@ -749,7 +749,7 @@ static int gfs2_fsync(struct file *file, loff_t start, loff_t end, { struct address_space *mapping = file->f_mapping; struct inode *inode = mapping->host; - int sync_state = inode->i_state & I_DIRTY_ALL; + int sync_state = inode->i_state & I_DIRTY; struct gfs2_inode *ip = GFS2_I(inode); int ret = 0, ret1 = 0; @@ -762,7 +762,7 @@ static int gfs2_fsync(struct file *file, loff_t start, loff_t end, if (!gfs2_is_jdata(ip)) sync_state &= ~I_DIRTY_PAGES; if (datasync) - sync_state &= ~(I_DIRTY_SYNC | I_DIRTY_TIME); + sync_state &= ~I_DIRTY_SYNC; if (sync_state) { ret = sync_inode_metadata(inode, 1); @@ -797,9 +797,7 @@ static ssize_t gfs2_file_direct_read(struct kiocb *iocb, struct iov_iter *to, if (ret) goto out_uninit; - ret = iomap_dio_rw(iocb, to, &gfs2_iomap_ops, NULL, - is_sync_kiocb(iocb)); - + ret = iomap_dio_rw(iocb, to, &gfs2_iomap_ops, NULL, 0); gfs2_glock_dq(gh); out_uninit: gfs2_holder_uninit(gh); @@ -833,8 +831,7 @@ static ssize_t gfs2_file_direct_write(struct kiocb *iocb, struct iov_iter *from, if (offset + len > i_size_read(&ip->i_inode)) goto out; - ret = iomap_dio_rw(iocb, from, &gfs2_iomap_ops, NULL, - is_sync_kiocb(iocb)); + ret = iomap_dio_rw(iocb, from, &gfs2_iomap_ops, NULL, 0); if (ret == -ENOTBLK) ret = 0; out: @@ -1115,8 +1112,8 @@ static long __gfs2_fallocate(struct file *file, int mode, loff_t offset, loff_t goto out_qunlock; /* check if the selected rgrp limits our max_blks further */ - if (ap.allowed && ap.allowed < max_blks) - max_blks = ap.allowed; + if (ip->i_res.rs_reserved < max_blks) + max_blks = ip->i_res.rs_reserved; /* Almost done. Calculate bytes that can be written using * max_blks. We also recompute max_bytes, data_blocks and diff --git a/fs/gfs2/glock.c b/fs/gfs2/glock.c index d87a5bc3607b..9567520d79f7 100644 --- a/fs/gfs2/glock.c +++ b/fs/gfs2/glock.c @@ -313,9 +313,23 @@ void gfs2_glock_put(struct gfs2_glock *gl) static inline int may_grant(const struct gfs2_glock *gl, const struct gfs2_holder *gh) { const struct gfs2_holder *gh_head = list_first_entry(&gl->gl_holders, const struct gfs2_holder, gh_list); - if ((gh->gh_state == LM_ST_EXCLUSIVE || - gh_head->gh_state == LM_ST_EXCLUSIVE) && gh != gh_head) - return 0; + + if (gh != gh_head) { + /** + * Here we make a special exception to grant holders who agree + * to share the EX lock with other holders who also have the + * bit set. If the original holder has the LM_FLAG_NODE_SCOPE bit + * is set, we grant more holders with the bit set. + */ + if (gh_head->gh_state == LM_ST_EXCLUSIVE && + (gh_head->gh_flags & LM_FLAG_NODE_SCOPE) && + gh->gh_state == LM_ST_EXCLUSIVE && + (gh->gh_flags & LM_FLAG_NODE_SCOPE)) + return 1; + if ((gh->gh_state == LM_ST_EXCLUSIVE || + gh_head->gh_state == LM_ST_EXCLUSIVE)) + return 0; + } if (gl->gl_state == gh->gh_state) return 1; if (gh->gh_flags & GL_EXACT) @@ -2030,6 +2044,8 @@ static const char *hflags2str(char *buf, u16 flags, unsigned long iflags) *p++ = 'A'; if (flags & LM_FLAG_PRIORITY) *p++ = 'p'; + if (flags & LM_FLAG_NODE_SCOPE) + *p++ = 'n'; if (flags & GL_ASYNC) *p++ = 'a'; if (flags & GL_EXACT) diff --git a/fs/gfs2/glock.h b/fs/gfs2/glock.h index 53813364517b..31a8f2f649b5 100644 --- a/fs/gfs2/glock.h +++ b/fs/gfs2/glock.h @@ -75,6 +75,11 @@ enum { * request and directly join the other shared lock. A shared lock request * without the priority flag might be forced to wait until the deferred * requested had acquired and released the lock. + * + * LM_FLAG_NODE_SCOPE + * This holder agrees to share the lock within this node. In other words, + * the glock is held in EX mode according to DLM, but local holders on the + * same node can share it. */ #define LM_FLAG_TRY 0x0001 @@ -82,6 +87,7 @@ enum { #define LM_FLAG_NOEXP 0x0004 #define LM_FLAG_ANY 0x0008 #define LM_FLAG_PRIORITY 0x0010 +#define LM_FLAG_NODE_SCOPE 0x0020 #define GL_ASYNC 0x0040 #define GL_EXACT 0x0080 #define GL_SKIP 0x0100 diff --git a/fs/gfs2/glops.c b/fs/gfs2/glops.c index 3faa421568b0..8e32d569c8bf 100644 --- a/fs/gfs2/glops.c +++ b/fs/gfs2/glops.c @@ -86,16 +86,12 @@ static int gfs2_ail_empty_gl(struct gfs2_glock *gl) { struct gfs2_sbd *sdp = gl->gl_name.ln_sbd; struct gfs2_trans tr; + unsigned int revokes; int ret; - memset(&tr, 0, sizeof(tr)); - INIT_LIST_HEAD(&tr.tr_buf); - INIT_LIST_HEAD(&tr.tr_databuf); - INIT_LIST_HEAD(&tr.tr_ail1_list); - INIT_LIST_HEAD(&tr.tr_ail2_list); - tr.tr_revokes = atomic_read(&gl->gl_ail_count); + revokes = atomic_read(&gl->gl_ail_count); - if (!tr.tr_revokes) { + if (!revokes) { bool have_revokes; bool log_in_flight; @@ -122,20 +118,14 @@ static int gfs2_ail_empty_gl(struct gfs2_glock *gl) return 0; } - /* A shortened, inline version of gfs2_trans_begin() - * tr->alloced is not set since the transaction structure is - * on the stack */ - tr.tr_reserved = 1 + gfs2_struct2blk(sdp, tr.tr_revokes); - tr.tr_ip = _RET_IP_; - ret = gfs2_log_reserve(sdp, tr.tr_reserved); - if (ret < 0) - return ret; - WARN_ON_ONCE(current->journal_info); - current->journal_info = &tr; - - __gfs2_ail_flush(gl, 0, tr.tr_revokes); - + memset(&tr, 0, sizeof(tr)); + set_bit(TR_ONSTACK, &tr.tr_flags); + ret = __gfs2_trans_begin(&tr, sdp, 0, revokes, _RET_IP_); + if (ret) + goto flush; + __gfs2_ail_flush(gl, 0, revokes); gfs2_trans_end(sdp); + flush: gfs2_log_flush(sdp, NULL, GFS2_LOG_HEAD_FLUSH_NORMAL | GFS2_LFC_AIL_EMPTY_GL); @@ -146,19 +136,15 @@ void gfs2_ail_flush(struct gfs2_glock *gl, bool fsync) { struct gfs2_sbd *sdp = gl->gl_name.ln_sbd; unsigned int revokes = atomic_read(&gl->gl_ail_count); - unsigned int max_revokes = (sdp->sd_sb.sb_bsize - sizeof(struct gfs2_log_descriptor)) / sizeof(u64); int ret; if (!revokes) return; - while (revokes > max_revokes) - max_revokes += (sdp->sd_sb.sb_bsize - sizeof(struct gfs2_meta_header)) / sizeof(u64); - - ret = gfs2_trans_begin(sdp, 0, max_revokes); + ret = gfs2_trans_begin(sdp, 0, revokes); if (ret) return; - __gfs2_ail_flush(gl, fsync, max_revokes); + __gfs2_ail_flush(gl, fsync, revokes); gfs2_trans_end(sdp); gfs2_log_flush(sdp, NULL, GFS2_LOG_HEAD_FLUSH_NORMAL | GFS2_LFC_AIL_FLUSH); diff --git a/fs/gfs2/incore.h b/fs/gfs2/incore.h index 8e1ab8ed4abc..0957119f7744 100644 --- a/fs/gfs2/incore.h +++ b/fs/gfs2/incore.h @@ -20,6 +20,7 @@ #include <linux/percpu.h> #include <linux/lockref.h> #include <linux/rhashtable.h> +#include <linux/mutex.h> #define DIO_WAIT 0x00000010 #define DIO_METADATA 0x00000020 @@ -106,7 +107,8 @@ struct gfs2_rgrpd { u32 rd_data; /* num of data blocks in rgrp */ u32 rd_bitbytes; /* number of bytes in data bitmaps */ u32 rd_free; - u32 rd_reserved; /* number of blocks reserved */ + u32 rd_requested; /* number of blocks in rd_rstree */ + u32 rd_reserved; /* number of reserved blocks */ u32 rd_free_clone; u32 rd_dinodes; u64 rd_igeneration; @@ -122,34 +124,10 @@ struct gfs2_rgrpd { #define GFS2_RDF_PREFERRED 0x80000000 /* This rgrp is preferred */ #define GFS2_RDF_MASK 0xf0000000 /* mask for internal flags */ spinlock_t rd_rsspin; /* protects reservation related vars */ + struct mutex rd_mutex; struct rb_root rd_rstree; /* multi-block reservation tree */ }; -struct gfs2_rbm { - struct gfs2_rgrpd *rgd; - u32 offset; /* The offset is bitmap relative */ - int bii; /* Bitmap index */ -}; - -static inline struct gfs2_bitmap *rbm_bi(const struct gfs2_rbm *rbm) -{ - return rbm->rgd->rd_bits + rbm->bii; -} - -static inline u64 gfs2_rbm_to_block(const struct gfs2_rbm *rbm) -{ - BUG_ON(rbm->offset >= rbm->rgd->rd_data); - return rbm->rgd->rd_data0 + (rbm_bi(rbm)->bi_start * GFS2_NBBY) + - rbm->offset; -} - -static inline bool gfs2_rbm_eq(const struct gfs2_rbm *rbm1, - const struct gfs2_rbm *rbm2) -{ - return (rbm1->rgd == rbm2->rgd) && (rbm1->bii == rbm2->bii) && - (rbm1->offset == rbm2->offset); -} - enum gfs2_state_bits { BH_Pinned = BH_PrivateStart, BH_Escaped = BH_PrivateStart + 1, @@ -313,9 +291,11 @@ struct gfs2_qadata { /* quota allocation data */ */ struct gfs2_blkreserv { - struct rb_node rs_node; /* link to other block reservations */ - struct gfs2_rbm rs_rbm; /* Start of reservation */ - u32 rs_free; /* how many blocks are still free */ + struct rb_node rs_node; /* node within rd_rstree */ + struct gfs2_rgrpd *rs_rgd; + u64 rs_start; + u32 rs_requested; + u32 rs_reserved; /* number of reserved blocks */ }; /* @@ -490,7 +470,7 @@ struct gfs2_quota_data { enum { TR_TOUCHED = 1, TR_ATTACHED = 2, - TR_ALLOCED = 3, + TR_ONSTACK = 3, }; struct gfs2_trans { @@ -506,7 +486,6 @@ struct gfs2_trans { unsigned int tr_num_buf_rm; unsigned int tr_num_databuf_rm; unsigned int tr_num_revoke; - unsigned int tr_num_revoke_rm; struct list_head tr_list; struct list_head tr_databuf; @@ -531,6 +510,7 @@ struct gfs2_jdesc { unsigned int nr_extents; struct work_struct jd_work; struct inode *jd_inode; + struct bio *jd_log_bio; unsigned long jd_flags; #define JDF_RECOVERY 1 unsigned int jd_jid; @@ -585,6 +565,7 @@ struct gfs2_args { unsigned int ar_errors:2; /* errors=withdraw | panic */ unsigned int ar_nobarrier:1; /* do not send barriers */ unsigned int ar_rgrplvb:1; /* use lvbs for rgrp info */ + unsigned int ar_got_rgrplvb:1; /* Was the rgrplvb opt given? */ unsigned int ar_loccookie:1; /* use location based readdir cookies */ s32 ar_commit; /* Commit interval */ @@ -821,7 +802,6 @@ struct gfs2_sbd { struct gfs2_trans *sd_log_tr; unsigned int sd_log_blks_reserved; - int sd_log_committed_revoke; atomic_t sd_log_pinned; unsigned int sd_log_num_revoke; @@ -834,24 +814,22 @@ struct gfs2_sbd { atomic_t sd_log_thresh2; atomic_t sd_log_blks_free; atomic_t sd_log_blks_needed; + atomic_t sd_log_revokes_available; wait_queue_head_t sd_log_waitq; wait_queue_head_t sd_logd_waitq; u64 sd_log_sequence; - unsigned int sd_log_head; - unsigned int sd_log_tail; int sd_log_idle; struct rw_semaphore sd_log_flush_lock; atomic_t sd_log_in_flight; - struct bio *sd_log_bio; wait_queue_head_t sd_log_flush_wait; int sd_log_error; /* First log error */ wait_queue_head_t sd_withdraw_wait; - atomic_t sd_reserving_log; - wait_queue_head_t sd_reserving_log_wait; - + unsigned int sd_log_tail; + unsigned int sd_log_flush_tail; + unsigned int sd_log_head; unsigned int sd_log_flush_head; spinlock_t sd_ail_lock; diff --git a/fs/gfs2/inode.c b/fs/gfs2/inode.c index c1b77e8d6b1c..c9775d5c6594 100644 --- a/fs/gfs2/inode.c +++ b/fs/gfs2/inode.c @@ -325,7 +325,7 @@ struct inode *gfs2_lookupi(struct inode *dir, const struct qstr *name, } if (!is_root) { - error = gfs2_permission(dir, MAY_EXEC); + error = gfs2_permission(&init_user_ns, dir, MAY_EXEC); if (error) goto out; } @@ -355,7 +355,8 @@ static int create_ok(struct gfs2_inode *dip, const struct qstr *name, { int error; - error = gfs2_permission(&dip->i_inode, MAY_WRITE | MAY_EXEC); + error = gfs2_permission(&init_user_ns, &dip->i_inode, + MAY_WRITE | MAY_EXEC); if (error) return error; @@ -490,8 +491,8 @@ static void init_dinode(struct gfs2_inode *dip, struct gfs2_inode *ip, di = (struct gfs2_dinode *)dibh->b_data; gfs2_dinode_out(ip, di); - di->di_major = cpu_to_be32(MAJOR(ip->i_inode.i_rdev)); - di->di_minor = cpu_to_be32(MINOR(ip->i_inode.i_rdev)); + di->di_major = cpu_to_be32(imajor(&ip->i_inode)); + di->di_minor = cpu_to_be32(iminor(&ip->i_inode)); di->__pad1 = 0; di->__pad2 = 0; di->__pad3 = 0; @@ -843,8 +844,8 @@ fail: * Returns: errno */ -static int gfs2_create(struct inode *dir, struct dentry *dentry, - umode_t mode, bool excl) +static int gfs2_create(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, bool excl) { return gfs2_create_inode(dir, dentry, NULL, S_IFREG | mode, 0, NULL, 0, excl); } @@ -951,7 +952,7 @@ static int gfs2_link(struct dentry *old_dentry, struct inode *dir, if (inode->i_nlink == 0) goto out_gunlock; - error = gfs2_permission(dir, MAY_WRITE | MAY_EXEC); + error = gfs2_permission(&init_user_ns, dir, MAY_WRITE | MAY_EXEC); if (error) goto out_gunlock; @@ -1068,7 +1069,8 @@ static int gfs2_unlink_ok(struct gfs2_inode *dip, const struct qstr *name, if (IS_APPEND(&dip->i_inode)) return -EPERM; - error = gfs2_permission(&dip->i_inode, MAY_WRITE | MAY_EXEC); + error = gfs2_permission(&init_user_ns, &dip->i_inode, + MAY_WRITE | MAY_EXEC); if (error) return error; @@ -1145,7 +1147,7 @@ static int gfs2_unlink(struct inode *dir, struct dentry *dentry) if (!rgd) goto out_inodes; - gfs2_holder_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0, ghs + 2); + gfs2_holder_init(rgd->rd_gl, LM_ST_EXCLUSIVE, LM_FLAG_NODE_SCOPE, ghs + 2); error = gfs2_glock_nq(ghs); /* parent */ @@ -1204,8 +1206,8 @@ out_inodes: * Returns: errno */ -static int gfs2_symlink(struct inode *dir, struct dentry *dentry, - const char *symname) +static int gfs2_symlink(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, const char *symname) { unsigned int size; @@ -1225,7 +1227,8 @@ static int gfs2_symlink(struct inode *dir, struct dentry *dentry, * Returns: errno */ -static int gfs2_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) +static int gfs2_mkdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { unsigned dsize = gfs2_max_stuffed_size(GFS2_I(dir)); return gfs2_create_inode(dir, dentry, NULL, S_IFDIR | mode, 0, NULL, dsize, 0); @@ -1240,8 +1243,8 @@ static int gfs2_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) * */ -static int gfs2_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, - dev_t dev) +static int gfs2_mknod(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, dev_t dev) { return gfs2_create_inode(dir, dentry, NULL, mode, dev, NULL, 0, 0); } @@ -1450,8 +1453,8 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry, error = -ENOENT; goto out_gunlock; } - error = gfs2_glock_nq_init(nrgd->rd_gl, LM_ST_EXCLUSIVE, 0, - &rd_gh); + error = gfs2_glock_nq_init(nrgd->rd_gl, LM_ST_EXCLUSIVE, + LM_FLAG_NODE_SCOPE, &rd_gh); if (error) goto out_gunlock; } @@ -1490,7 +1493,8 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry, } } } else { - error = gfs2_permission(ndir, MAY_WRITE | MAY_EXEC); + error = gfs2_permission(&init_user_ns, ndir, + MAY_WRITE | MAY_EXEC); if (error) goto out_gunlock; @@ -1525,7 +1529,8 @@ static int gfs2_rename(struct inode *odir, struct dentry *odentry, /* Check out the dir to be renamed */ if (dir_rename) { - error = gfs2_permission(d_inode(odentry), MAY_WRITE); + error = gfs2_permission(&init_user_ns, d_inode(odentry), + MAY_WRITE); if (error) goto out_gunlock; } @@ -1688,12 +1693,14 @@ static int gfs2_exchange(struct inode *odir, struct dentry *odentry, goto out_gunlock; if (S_ISDIR(old_mode)) { - error = gfs2_permission(odentry->d_inode, MAY_WRITE); + error = gfs2_permission(&init_user_ns, odentry->d_inode, + MAY_WRITE); if (error) goto out_gunlock; } if (S_ISDIR(new_mode)) { - error = gfs2_permission(ndentry->d_inode, MAY_WRITE); + error = gfs2_permission(&init_user_ns, ndentry->d_inode, + MAY_WRITE); if (error) goto out_gunlock; } @@ -1747,9 +1754,9 @@ out: return error; } -static int gfs2_rename2(struct inode *odir, struct dentry *odentry, - struct inode *ndir, struct dentry *ndentry, - unsigned int flags) +static int gfs2_rename2(struct user_namespace *mnt_userns, struct inode *odir, + struct dentry *odentry, struct inode *ndir, + struct dentry *ndentry, unsigned int flags) { flags &= ~RENAME_NOREPLACE; @@ -1833,7 +1840,8 @@ out: * Returns: errno */ -int gfs2_permission(struct inode *inode, int mask) +int gfs2_permission(struct user_namespace *mnt_userns, struct inode *inode, + int mask) { struct gfs2_inode *ip; struct gfs2_holder i_gh; @@ -1852,7 +1860,7 @@ int gfs2_permission(struct inode *inode, int mask) if ((mask & MAY_WRITE) && IS_IMMUTABLE(inode)) error = -EPERM; else - error = generic_permission(inode, mask); + error = generic_permission(&init_user_ns, inode, mask); if (gfs2_holder_initialized(&i_gh)) gfs2_glock_dq_uninit(&i_gh); @@ -1861,7 +1869,7 @@ int gfs2_permission(struct inode *inode, int mask) static int __gfs2_setattr_simple(struct inode *inode, struct iattr *attr) { - setattr_copy(inode, attr); + setattr_copy(&init_user_ns, inode, attr); mark_inode_dirty(inode); return 0; } @@ -1963,7 +1971,8 @@ out: * Returns: errno */ -static int gfs2_setattr(struct dentry *dentry, struct iattr *attr) +static int gfs2_setattr(struct user_namespace *mnt_userns, + struct dentry *dentry, struct iattr *attr) { struct inode *inode = d_inode(dentry); struct gfs2_inode *ip = GFS2_I(inode); @@ -1982,7 +1991,7 @@ static int gfs2_setattr(struct dentry *dentry, struct iattr *attr) if (IS_IMMUTABLE(inode) || IS_APPEND(inode)) goto error; - error = setattr_prepare(dentry, attr); + error = setattr_prepare(&init_user_ns, dentry, attr); if (error) goto error; @@ -1993,7 +2002,8 @@ static int gfs2_setattr(struct dentry *dentry, struct iattr *attr) else { error = gfs2_setattr_simple(inode, attr); if (!error && attr->ia_valid & ATTR_MODE) - error = posix_acl_chmod(inode, inode->i_mode); + error = posix_acl_chmod(&init_user_ns, inode, + inode->i_mode); } error: @@ -2007,6 +2017,7 @@ out: /** * gfs2_getattr - Read out an inode's attributes + * @mnt_userns: user namespace of the mount the inode was found from * @path: Object to query * @stat: The inode's stats * @request_mask: Mask of STATX_xxx flags indicating the caller's interests @@ -2021,7 +2032,8 @@ out: * Returns: errno */ -static int gfs2_getattr(const struct path *path, struct kstat *stat, +static int gfs2_getattr(struct user_namespace *mnt_userns, + const struct path *path, struct kstat *stat, u32 request_mask, unsigned int flags) { struct inode *inode = d_inode(path->dentry); @@ -2049,7 +2061,7 @@ static int gfs2_getattr(const struct path *path, struct kstat *stat, STATX_ATTR_IMMUTABLE | STATX_ATTR_NODUMP); - generic_fillattr(inode, stat); + generic_fillattr(&init_user_ns, inode, stat); if (gfs2_holder_initialized(&gh)) gfs2_glock_dq_uninit(&gh); diff --git a/fs/gfs2/inode.h b/fs/gfs2/inode.h index 8073b8d2c7fa..c447bd5b3017 100644 --- a/fs/gfs2/inode.h +++ b/fs/gfs2/inode.h @@ -99,7 +99,8 @@ extern int gfs2_inode_refresh(struct gfs2_inode *ip); extern struct inode *gfs2_lookupi(struct inode *dir, const struct qstr *name, int is_root); -extern int gfs2_permission(struct inode *inode, int mask); +extern int gfs2_permission(struct user_namespace *mnt_userns, + struct inode *inode, int mask); extern int gfs2_setattr_simple(struct inode *inode, struct iattr *attr); extern struct inode *gfs2_lookup_simple(struct inode *dip, const char *name); extern void gfs2_dinode_out(const struct gfs2_inode *ip, void *buf); diff --git a/fs/gfs2/lock_dlm.c b/fs/gfs2/lock_dlm.c index 9f2b5609f225..153272f82984 100644 --- a/fs/gfs2/lock_dlm.c +++ b/fs/gfs2/lock_dlm.c @@ -284,7 +284,6 @@ static void gdlm_put_lock(struct gfs2_glock *gl) { struct gfs2_sbd *sdp = gl->gl_name.ln_sbd; struct lm_lockstruct *ls = &sdp->sd_lockstruct; - int lvb_needs_unlock = 0; int error; if (gl->gl_lksb.sb_lkid == 0) { @@ -297,13 +296,10 @@ static void gdlm_put_lock(struct gfs2_glock *gl) gfs2_sbstats_inc(gl, GFS2_LKS_DCOUNT); gfs2_update_request_times(gl); - /* don't want to skip dlm_unlock writing the lvb when lock is ex */ - - if (gl->gl_lksb.sb_lvbptr && (gl->gl_state == LM_ST_EXCLUSIVE)) - lvb_needs_unlock = 1; + /* don't want to skip dlm_unlock writing the lvb when lock has one */ if (test_bit(SDF_SKIP_DLM_UNLOCK, &sdp->sd_flags) && - !lvb_needs_unlock) { + !gl->gl_lksb.sb_lvbptr) { gfs2_glock_free(gl); return; } diff --git a/fs/gfs2/log.c b/fs/gfs2/log.c index 2e9314091c81..6410281546f9 100644 --- a/fs/gfs2/log.c +++ b/fs/gfs2/log.c @@ -50,10 +50,12 @@ unsigned int gfs2_struct2blk(struct gfs2_sbd *sdp, unsigned int nstruct) unsigned int blks; unsigned int first, second; + /* The initial struct gfs2_log_descriptor block */ blks = 1; first = sdp->sd_ldptrs; if (nstruct > first) { + /* Subsequent struct gfs2_meta_header blocks */ second = sdp->sd_inptrs; blks += DIV_ROUND_UP(nstruct - first, second); } @@ -89,7 +91,7 @@ void gfs2_remove_from_ail(struct gfs2_bufdata *bd) static int gfs2_ail1_start_one(struct gfs2_sbd *sdp, struct writeback_control *wbc, - struct gfs2_trans *tr) + struct gfs2_trans *tr, struct blk_plug *plug) __releases(&sdp->sd_ail_lock) __acquires(&sdp->sd_ail_lock) { @@ -131,6 +133,11 @@ __acquires(&sdp->sd_ail_lock) continue; spin_unlock(&sdp->sd_ail_lock); ret = generic_writepages(mapping, wbc); + if (need_resched()) { + blk_finish_plug(plug); + cond_resched(); + blk_start_plug(plug); + } spin_lock(&sdp->sd_ail_lock); if (ret == -ENODATA) /* if a jdata write into a new hole */ ret = 0; /* ignore it */ @@ -205,7 +212,7 @@ restart: list_for_each_entry_reverse(tr, head, tr_list) { if (wbc->nr_to_write <= 0) break; - ret = gfs2_ail1_start_one(sdp, wbc, tr); + ret = gfs2_ail1_start_one(sdp, wbc, tr, &plug); if (ret) { if (ret == -EBUSY) goto restart; @@ -240,6 +247,45 @@ static void gfs2_ail1_start(struct gfs2_sbd *sdp) return gfs2_ail1_flush(sdp, &wbc); } +static void gfs2_log_update_flush_tail(struct gfs2_sbd *sdp) +{ + unsigned int new_flush_tail = sdp->sd_log_head; + struct gfs2_trans *tr; + + if (!list_empty(&sdp->sd_ail1_list)) { + tr = list_last_entry(&sdp->sd_ail1_list, + struct gfs2_trans, tr_list); + new_flush_tail = tr->tr_first; + } + sdp->sd_log_flush_tail = new_flush_tail; +} + +static void gfs2_log_update_head(struct gfs2_sbd *sdp) +{ + unsigned int new_head = sdp->sd_log_flush_head; + + if (sdp->sd_log_flush_tail == sdp->sd_log_head) + sdp->sd_log_flush_tail = new_head; + sdp->sd_log_head = new_head; +} + +/** + * gfs2_ail_empty_tr - empty one of the ail lists of a transaction + */ + +static void gfs2_ail_empty_tr(struct gfs2_sbd *sdp, struct gfs2_trans *tr, + struct list_head *head) +{ + struct gfs2_bufdata *bd; + + while (!list_empty(head)) { + bd = list_first_entry(head, struct gfs2_bufdata, + bd_ail_st_list); + gfs2_assert(sdp, bd->bd_tr == tr); + gfs2_remove_from_ail(bd); + } +} + /** * gfs2_ail1_empty_one - Check whether or not a trans in the AIL has been synced * @sdp: the filesystem @@ -315,6 +361,7 @@ static int gfs2_ail1_empty(struct gfs2_sbd *sdp, int max_revokes) else oldest_tr = 0; } + gfs2_log_update_flush_tail(sdp); ret = list_empty(&sdp->sd_ail1_list); spin_unlock(&sdp->sd_ail_lock); @@ -348,47 +395,69 @@ static void gfs2_ail1_wait(struct gfs2_sbd *sdp) spin_unlock(&sdp->sd_ail_lock); } -/** - * gfs2_ail_empty_tr - empty one of the ail lists for a transaction - */ - -static void gfs2_ail_empty_tr(struct gfs2_sbd *sdp, struct gfs2_trans *tr, - struct list_head *head) +static void __ail2_empty(struct gfs2_sbd *sdp, struct gfs2_trans *tr) { - struct gfs2_bufdata *bd; - - while (!list_empty(head)) { - bd = list_first_entry(head, struct gfs2_bufdata, - bd_ail_st_list); - gfs2_assert(sdp, bd->bd_tr == tr); - gfs2_remove_from_ail(bd); - } + gfs2_ail_empty_tr(sdp, tr, &tr->tr_ail2_list); + list_del(&tr->tr_list); + gfs2_assert_warn(sdp, list_empty(&tr->tr_ail1_list)); + gfs2_assert_warn(sdp, list_empty(&tr->tr_ail2_list)); + gfs2_trans_free(sdp, tr); } static void ail2_empty(struct gfs2_sbd *sdp, unsigned int new_tail) { - struct gfs2_trans *tr, *safe; + struct list_head *ail2_list = &sdp->sd_ail2_list; unsigned int old_tail = sdp->sd_log_tail; - int wrap = (new_tail < old_tail); - int a, b, rm; + struct gfs2_trans *tr, *safe; spin_lock(&sdp->sd_ail_lock); + if (old_tail <= new_tail) { + list_for_each_entry_safe(tr, safe, ail2_list, tr_list) { + if (old_tail <= tr->tr_first && tr->tr_first < new_tail) + __ail2_empty(sdp, tr); + } + } else { + list_for_each_entry_safe(tr, safe, ail2_list, tr_list) { + if (old_tail <= tr->tr_first || tr->tr_first < new_tail) + __ail2_empty(sdp, tr); + } + } + spin_unlock(&sdp->sd_ail_lock); +} - list_for_each_entry_safe(tr, safe, &sdp->sd_ail2_list, tr_list) { - a = (old_tail <= tr->tr_first); - b = (tr->tr_first < new_tail); - rm = (wrap) ? (a || b) : (a && b); - if (!rm) - continue; +/** + * gfs2_log_is_empty - Check if the log is empty + * @sdp: The GFS2 superblock + */ - gfs2_ail_empty_tr(sdp, tr, &tr->tr_ail2_list); - list_del(&tr->tr_list); - gfs2_assert_warn(sdp, list_empty(&tr->tr_ail1_list)); - gfs2_assert_warn(sdp, list_empty(&tr->tr_ail2_list)); - gfs2_trans_free(sdp, tr); +bool gfs2_log_is_empty(struct gfs2_sbd *sdp) { + return atomic_read(&sdp->sd_log_blks_free) == sdp->sd_jdesc->jd_blocks; +} + +static bool __gfs2_log_try_reserve_revokes(struct gfs2_sbd *sdp, unsigned int revokes) +{ + unsigned int available; + + available = atomic_read(&sdp->sd_log_revokes_available); + while (available >= revokes) { + if (atomic_try_cmpxchg(&sdp->sd_log_revokes_available, + &available, available - revokes)) + return true; } + return false; +} - spin_unlock(&sdp->sd_ail_lock); +/** + * gfs2_log_release_revokes - Release a given number of revokes + * @sdp: The GFS2 superblock + * @revokes: The number of revokes to release + * + * sdp->sd_log_flush_lock must be held. + */ +void gfs2_log_release_revokes(struct gfs2_sbd *sdp, unsigned int revokes) +{ + if (revokes) + atomic_add(revokes, &sdp->sd_log_revokes_available); } /** @@ -400,86 +469,141 @@ static void ail2_empty(struct gfs2_sbd *sdp, unsigned int new_tail) void gfs2_log_release(struct gfs2_sbd *sdp, unsigned int blks) { - atomic_add(blks, &sdp->sd_log_blks_free); trace_gfs2_log_blocks(sdp, blks); gfs2_assert_withdraw(sdp, atomic_read(&sdp->sd_log_blks_free) <= sdp->sd_jdesc->jd_blocks); - up_read(&sdp->sd_log_flush_lock); + if (atomic_read(&sdp->sd_log_blks_needed)) + wake_up(&sdp->sd_log_waitq); } /** - * gfs2_log_reserve - Make a log reservation + * __gfs2_log_try_reserve - Try to make a log reservation * @sdp: The GFS2 superblock * @blks: The number of blocks to reserve + * @taboo_blks: The number of blocks to leave free * - * Note that we never give out the last few blocks of the journal. Thats - * due to the fact that there is a small number of header blocks - * associated with each log flush. The exact number can't be known until - * flush time, so we ensure that we have just enough free blocks at all - * times to avoid running out during a log flush. + * Try to do the same as __gfs2_log_reserve(), but fail if no more log + * space is immediately available. + */ +static bool __gfs2_log_try_reserve(struct gfs2_sbd *sdp, unsigned int blks, + unsigned int taboo_blks) +{ + unsigned wanted = blks + taboo_blks; + unsigned int free_blocks; + + free_blocks = atomic_read(&sdp->sd_log_blks_free); + while (free_blocks >= wanted) { + if (atomic_try_cmpxchg(&sdp->sd_log_blks_free, &free_blocks, + free_blocks - blks)) { + trace_gfs2_log_blocks(sdp, -blks); + return true; + } + } + return false; +} + +/** + * __gfs2_log_reserve - Make a log reservation + * @sdp: The GFS2 superblock + * @blks: The number of blocks to reserve + * @taboo_blks: The number of blocks to leave free + * + * @taboo_blks is set to 0 for logd, and to GFS2_LOG_FLUSH_MIN_BLOCKS + * for all other processes. This ensures that when the log is almost full, + * logd will still be able to call gfs2_log_flush one more time without + * blocking, which will advance the tail and make some more log space + * available. * * We no longer flush the log here, instead we wake up logd to do that * for us. To avoid the thundering herd and to ensure that we deal fairly * with queued waiters, we use an exclusive wait. This means that when we * get woken with enough journal space to get our reservation, we need to * wake the next waiter on the list. - * - * Returns: errno */ -int gfs2_log_reserve(struct gfs2_sbd *sdp, unsigned int blks) +static void __gfs2_log_reserve(struct gfs2_sbd *sdp, unsigned int blks, + unsigned int taboo_blks) { - int ret = 0; - unsigned reserved_blks = 7 * (4096 / sdp->sd_vfs->s_blocksize); - unsigned wanted = blks + reserved_blks; - DEFINE_WAIT(wait); - int did_wait = 0; + unsigned wanted = blks + taboo_blks; unsigned int free_blocks; - if (gfs2_assert_warn(sdp, blks) || - gfs2_assert_warn(sdp, blks <= sdp->sd_jdesc->jd_blocks)) - return -EINVAL; atomic_add(blks, &sdp->sd_log_blks_needed); -retry: - free_blocks = atomic_read(&sdp->sd_log_blks_free); - if (unlikely(free_blocks <= wanted)) { - do { - prepare_to_wait_exclusive(&sdp->sd_log_waitq, &wait, - TASK_UNINTERRUPTIBLE); + for (;;) { + if (current != sdp->sd_logd_process) wake_up(&sdp->sd_logd_waitq); - did_wait = 1; - if (atomic_read(&sdp->sd_log_blks_free) <= wanted) - io_schedule(); - free_blocks = atomic_read(&sdp->sd_log_blks_free); - } while(free_blocks <= wanted); - finish_wait(&sdp->sd_log_waitq, &wait); - } - atomic_inc(&sdp->sd_reserving_log); - if (atomic_cmpxchg(&sdp->sd_log_blks_free, free_blocks, - free_blocks - blks) != free_blocks) { - if (atomic_dec_and_test(&sdp->sd_reserving_log)) - wake_up(&sdp->sd_reserving_log_wait); - goto retry; + io_wait_event(sdp->sd_log_waitq, + (free_blocks = atomic_read(&sdp->sd_log_blks_free), + free_blocks >= wanted)); + do { + if (atomic_try_cmpxchg(&sdp->sd_log_blks_free, + &free_blocks, + free_blocks - blks)) + goto reserved; + } while (free_blocks >= wanted); } - atomic_sub(blks, &sdp->sd_log_blks_needed); - trace_gfs2_log_blocks(sdp, -blks); - /* - * If we waited, then so might others, wake them up _after_ we get - * our share of the log. - */ - if (unlikely(did_wait)) +reserved: + trace_gfs2_log_blocks(sdp, -blks); + if (atomic_sub_return(blks, &sdp->sd_log_blks_needed)) wake_up(&sdp->sd_log_waitq); +} + +/** + * gfs2_log_try_reserve - Try to make a log reservation + * @sdp: The GFS2 superblock + * @tr: The transaction + * @extra_revokes: The number of additional revokes reserved (output) + * + * This is similar to gfs2_log_reserve, but sdp->sd_log_flush_lock must be + * held for correct revoke accounting. + */ + +bool gfs2_log_try_reserve(struct gfs2_sbd *sdp, struct gfs2_trans *tr, + unsigned int *extra_revokes) +{ + unsigned int blks = tr->tr_reserved; + unsigned int revokes = tr->tr_revokes; + unsigned int revoke_blks = 0; + + *extra_revokes = 0; + if (revokes && !__gfs2_log_try_reserve_revokes(sdp, revokes)) { + revoke_blks = DIV_ROUND_UP(revokes, sdp->sd_inptrs); + *extra_revokes = revoke_blks * sdp->sd_inptrs - revokes; + blks += revoke_blks; + } + if (!blks) + return true; + if (__gfs2_log_try_reserve(sdp, blks, GFS2_LOG_FLUSH_MIN_BLOCKS)) + return true; + if (!revoke_blks) + gfs2_log_release_revokes(sdp, revokes); + return false; +} - down_read(&sdp->sd_log_flush_lock); - if (unlikely(!test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags))) { - gfs2_log_release(sdp, blks); - ret = -EROFS; +/** + * gfs2_log_reserve - Make a log reservation + * @sdp: The GFS2 superblock + * @tr: The transaction + * @extra_revokes: The number of additional revokes reserved (output) + * + * sdp->sd_log_flush_lock must not be held. + */ + +void gfs2_log_reserve(struct gfs2_sbd *sdp, struct gfs2_trans *tr, + unsigned int *extra_revokes) +{ + unsigned int blks = tr->tr_reserved; + unsigned int revokes = tr->tr_revokes; + unsigned int revoke_blks = 0; + + *extra_revokes = 0; + if (revokes) { + revoke_blks = DIV_ROUND_UP(revokes, sdp->sd_inptrs); + *extra_revokes = revoke_blks * sdp->sd_inptrs - revokes; + blks += revoke_blks; } - if (atomic_dec_and_test(&sdp->sd_reserving_log)) - wake_up(&sdp->sd_reserving_log_wait); - return ret; + __gfs2_log_reserve(sdp, blks, GFS2_LOG_FLUSH_MIN_BLOCKS); } /** @@ -507,24 +631,20 @@ static inline unsigned int log_distance(struct gfs2_sbd *sdp, unsigned int newer } /** - * calc_reserved - Calculate the number of blocks to reserve when - * refunding a transaction's unused buffers. + * calc_reserved - Calculate the number of blocks to keep reserved * @sdp: The GFS2 superblock * * This is complex. We need to reserve room for all our currently used - * metadata buffers (e.g. normal file I/O rewriting file time stamps) and - * all our journaled data buffers for journaled files (e.g. files in the + * metadata blocks (e.g. normal file I/O rewriting file time stamps) and + * all our journaled data blocks for journaled files (e.g. files in the * meta_fs like rindex, or files for which chattr +j was done.) - * If we don't reserve enough space, gfs2_log_refund and gfs2_log_flush - * will count it as free space (sd_log_blks_free) and corruption will follow. + * If we don't reserve enough space, corruption will follow. * - * We can have metadata bufs and jdata bufs in the same journal. So each - * type gets its own log header, for which we need to reserve a block. - * In fact, each type has the potential for needing more than one header - * in cases where we have more buffers than will fit on a journal page. + * We can have metadata blocks and jdata blocks in the same journal. Each + * type gets its own log descriptor, for which we need to reserve a block. + * In fact, each type has the potential for needing more than one log descriptor + * in cases where we have more blocks than will fit in a log descriptor. * Metadata journal entries take up half the space of journaled buffer entries. - * Thus, metadata entries have buf_limit (502) and journaled buffers have - * databuf_limit (251) before they cause a wrap around. * * Also, we need to reserve blocks for revoke journal entries and one for an * overall header for the lot. @@ -533,59 +653,29 @@ static inline unsigned int log_distance(struct gfs2_sbd *sdp, unsigned int newer */ static unsigned int calc_reserved(struct gfs2_sbd *sdp) { - unsigned int reserved = 0; - unsigned int mbuf; - unsigned int dbuf; + unsigned int reserved = GFS2_LOG_FLUSH_MIN_BLOCKS; + unsigned int blocks; struct gfs2_trans *tr = sdp->sd_log_tr; if (tr) { - mbuf = tr->tr_num_buf_new - tr->tr_num_buf_rm; - dbuf = tr->tr_num_databuf_new - tr->tr_num_databuf_rm; - reserved = mbuf + dbuf; - /* Account for header blocks */ - reserved += DIV_ROUND_UP(mbuf, buf_limit(sdp)); - reserved += DIV_ROUND_UP(dbuf, databuf_limit(sdp)); + blocks = tr->tr_num_buf_new - tr->tr_num_buf_rm; + reserved += blocks + DIV_ROUND_UP(blocks, buf_limit(sdp)); + blocks = tr->tr_num_databuf_new - tr->tr_num_databuf_rm; + reserved += blocks + DIV_ROUND_UP(blocks, databuf_limit(sdp)); } - - if (sdp->sd_log_committed_revoke > 0) - reserved += gfs2_struct2blk(sdp, sdp->sd_log_committed_revoke); - /* One for the overall header */ - if (reserved) - reserved++; return reserved; } -static unsigned int current_tail(struct gfs2_sbd *sdp) -{ - struct gfs2_trans *tr; - unsigned int tail; - - spin_lock(&sdp->sd_ail_lock); - - if (list_empty(&sdp->sd_ail1_list)) { - tail = sdp->sd_log_head; - } else { - tr = list_last_entry(&sdp->sd_ail1_list, struct gfs2_trans, - tr_list); - tail = tr->tr_first; - } - - spin_unlock(&sdp->sd_ail_lock); - - return tail; -} - -static void log_pull_tail(struct gfs2_sbd *sdp, unsigned int new_tail) +static void log_pull_tail(struct gfs2_sbd *sdp) { - unsigned int dist = log_distance(sdp, new_tail, sdp->sd_log_tail); + unsigned int new_tail = sdp->sd_log_flush_tail; + unsigned int dist; + if (new_tail == sdp->sd_log_tail) + return; + dist = log_distance(sdp, new_tail, sdp->sd_log_tail); ail2_empty(sdp, new_tail); - - atomic_add(dist, &sdp->sd_log_blks_free); - trace_gfs2_log_blocks(sdp, dist); - gfs2_assert_withdraw(sdp, atomic_read(&sdp->sd_log_blks_free) <= - sdp->sd_jdesc->jd_blocks); - + gfs2_log_release(sdp, dist); sdp->sd_log_tail = new_tail; } @@ -698,7 +788,7 @@ void gfs2_glock_remove_revoke(struct gfs2_glock *gl) } /** - * gfs2_write_revokes - Add as many revokes to the system transaction as we can + * gfs2_flush_revokes - Add as many revokes to the system transaction as we can * @sdp: The GFS2 superblock * * Our usual strategy is to defer writing revokes as much as we can in the hope @@ -709,38 +799,14 @@ void gfs2_glock_remove_revoke(struct gfs2_glock *gl) * been written back. This will basically come at no cost now, and will save * us from having to keep track of those blocks on the AIL2 list later. */ -void gfs2_write_revokes(struct gfs2_sbd *sdp) +void gfs2_flush_revokes(struct gfs2_sbd *sdp) { /* number of revokes we still have room for */ - int max_revokes = (sdp->sd_sb.sb_bsize - sizeof(struct gfs2_log_descriptor)) / sizeof(u64); + unsigned int max_revokes = atomic_read(&sdp->sd_log_revokes_available); gfs2_log_lock(sdp); - while (sdp->sd_log_num_revoke > max_revokes) - max_revokes += (sdp->sd_sb.sb_bsize - sizeof(struct gfs2_meta_header)) / sizeof(u64); - max_revokes -= sdp->sd_log_num_revoke; - if (!sdp->sd_log_num_revoke) { - atomic_dec(&sdp->sd_log_blks_free); - /* If no blocks have been reserved, we need to also - * reserve a block for the header */ - if (!sdp->sd_log_blks_reserved) { - atomic_dec(&sdp->sd_log_blks_free); - trace_gfs2_log_blocks(sdp, -2); - } else { - trace_gfs2_log_blocks(sdp, -1); - } - } gfs2_ail1_empty(sdp, max_revokes); gfs2_log_unlock(sdp); - - if (!sdp->sd_log_num_revoke) { - atomic_inc(&sdp->sd_log_blks_free); - if (!sdp->sd_log_blks_reserved) { - atomic_inc(&sdp->sd_log_blks_free); - trace_gfs2_log_blocks(sdp, 2); - } else { - trace_gfs2_log_blocks(sdp, 1); - } - } } /** @@ -769,7 +835,7 @@ void gfs2_write_log_header(struct gfs2_sbd *sdp, struct gfs2_jdesc *jd, u64 dblock; if (gfs2_withdrawn(sdp)) - goto out; + return; page = mempool_alloc(gfs2_page_pool, GFP_NOIO); lh = page_address(page); @@ -822,10 +888,8 @@ void gfs2_write_log_header(struct gfs2_sbd *sdp, struct gfs2_jdesc *jd, sb->s_blocksize - LH_V1_SIZE - 4); lh->lh_crc = cpu_to_be32(crc); - gfs2_log_write(sdp, page, sb->s_blocksize, 0, dblock); - gfs2_log_submit_bio(&sdp->sd_log_bio, REQ_OP_WRITE | op_flags); -out: - log_flush_wait(sdp); + gfs2_log_write(sdp, jd, page, sb->s_blocksize, 0, dblock); + gfs2_log_submit_bio(&jd->jd_log_bio, REQ_OP_WRITE | op_flags); } /** @@ -838,25 +902,24 @@ out: static void log_write_header(struct gfs2_sbd *sdp, u32 flags) { - unsigned int tail; int op_flags = REQ_PREFLUSH | REQ_FUA | REQ_META | REQ_SYNC; enum gfs2_freeze_state state = atomic_read(&sdp->sd_freeze_state); gfs2_assert_withdraw(sdp, (state != SFS_FROZEN)); - tail = current_tail(sdp); if (test_bit(SDF_NOBARRIERS, &sdp->sd_flags)) { gfs2_ordered_wait(sdp); log_flush_wait(sdp); op_flags = REQ_SYNC | REQ_META | REQ_PRIO; } - sdp->sd_log_idle = (tail == sdp->sd_log_flush_head); - gfs2_write_log_header(sdp, sdp->sd_jdesc, sdp->sd_log_sequence++, tail, - sdp->sd_log_flush_head, flags, op_flags); + sdp->sd_log_idle = (sdp->sd_log_flush_tail == sdp->sd_log_flush_head); + gfs2_write_log_header(sdp, sdp->sd_jdesc, sdp->sd_log_sequence++, + sdp->sd_log_flush_tail, sdp->sd_log_flush_head, + flags, op_flags); gfs2_log_incr_head(sdp); - - if (sdp->sd_log_tail != tail) - log_pull_tail(sdp, tail); + log_flush_wait(sdp); + log_pull_tail(sdp); + gfs2_log_update_head(sdp); } /** @@ -935,12 +998,16 @@ static void trans_drain(struct gfs2_trans *tr) while (!list_empty(head)) { bd = list_first_entry(head, struct gfs2_bufdata, bd_list); list_del_init(&bd->bd_list); + if (!list_empty(&bd->bd_ail_st_list)) + gfs2_remove_from_ail(bd); kmem_cache_free(gfs2_bufdata_cachep, bd); } head = &tr->tr_databuf; while (!list_empty(head)) { bd = list_first_entry(head, struct gfs2_bufdata, bd_list); list_del_init(&bd->bd_list); + if (!list_empty(&bd->bd_ail_st_list)) + gfs2_remove_from_ail(bd); kmem_cache_free(gfs2_bufdata_cachep, bd); } } @@ -956,42 +1023,66 @@ static void trans_drain(struct gfs2_trans *tr) void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl, u32 flags) { struct gfs2_trans *tr = NULL; + unsigned int reserved_blocks = 0, used_blocks = 0; enum gfs2_freeze_state state = atomic_read(&sdp->sd_freeze_state); + unsigned int first_log_head; + unsigned int reserved_revokes = 0; down_write(&sdp->sd_log_flush_lock); + trace_gfs2_log_flush(sdp, 1, flags); +repeat: /* * Do this check while holding the log_flush_lock to prevent new * buffers from being added to the ail via gfs2_pin() */ - if (gfs2_withdrawn(sdp)) + if (gfs2_withdrawn(sdp) || !test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags)) goto out; /* Log might have been flushed while we waited for the flush lock */ if (gl && !test_bit(GLF_LFLUSH, &gl->gl_flags)) goto out; - trace_gfs2_log_flush(sdp, 1, flags); - if (flags & GFS2_LOG_HEAD_FLUSH_SHUTDOWN) - clear_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags); + first_log_head = sdp->sd_log_head; + sdp->sd_log_flush_head = first_log_head; - sdp->sd_log_flush_head = sdp->sd_log_head; tr = sdp->sd_log_tr; - if (tr) { - sdp->sd_log_tr = NULL; - tr->tr_first = sdp->sd_log_flush_head; - if (unlikely (state == SFS_FROZEN)) - if (gfs2_assert_withdraw_delayed(sdp, - !tr->tr_num_buf_new && !tr->tr_num_databuf_new)) - goto out_withdraw; + if (tr || sdp->sd_log_num_revoke) { + if (reserved_blocks) + gfs2_log_release(sdp, reserved_blocks); + reserved_blocks = sdp->sd_log_blks_reserved; + reserved_revokes = sdp->sd_log_num_revoke; + if (tr) { + sdp->sd_log_tr = NULL; + tr->tr_first = first_log_head; + if (unlikely (state == SFS_FROZEN)) { + if (gfs2_assert_withdraw_delayed(sdp, + !tr->tr_num_buf_new && !tr->tr_num_databuf_new)) + goto out_withdraw; + } + } + } else if (!reserved_blocks) { + unsigned int taboo_blocks = GFS2_LOG_FLUSH_MIN_BLOCKS; + + reserved_blocks = GFS2_LOG_FLUSH_MIN_BLOCKS; + if (current == sdp->sd_logd_process) + taboo_blocks = 0; + + if (!__gfs2_log_try_reserve(sdp, reserved_blocks, taboo_blocks)) { + up_write(&sdp->sd_log_flush_lock); + __gfs2_log_reserve(sdp, reserved_blocks, taboo_blocks); + down_write(&sdp->sd_log_flush_lock); + goto repeat; + } + BUG_ON(sdp->sd_log_num_revoke); } + if (flags & GFS2_LOG_HEAD_FLUSH_SHUTDOWN) + clear_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags); + if (unlikely(state == SFS_FROZEN)) - if (gfs2_assert_withdraw_delayed(sdp, !sdp->sd_log_num_revoke)) + if (gfs2_assert_withdraw_delayed(sdp, !reserved_revokes)) goto out_withdraw; - if (gfs2_assert_withdraw_delayed(sdp, - sdp->sd_log_num_revoke == sdp->sd_log_committed_revoke)) - goto out_withdraw; gfs2_ordered_write(sdp); if (gfs2_withdrawn(sdp)) @@ -999,16 +1090,13 @@ void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl, u32 flags) lops_before_commit(sdp, tr); if (gfs2_withdrawn(sdp)) goto out_withdraw; - gfs2_log_submit_bio(&sdp->sd_log_bio, REQ_OP_WRITE); + gfs2_log_submit_bio(&sdp->sd_jdesc->jd_log_bio, REQ_OP_WRITE); if (gfs2_withdrawn(sdp)) goto out_withdraw; if (sdp->sd_log_head != sdp->sd_log_flush_head) { - log_flush_wait(sdp); log_write_header(sdp, flags); - } else if (sdp->sd_log_tail != current_tail(sdp) && !sdp->sd_log_idle){ - atomic_dec(&sdp->sd_log_blks_free); /* Adjust for unreserved buffer */ - trace_gfs2_log_blocks(sdp, -1); + } else if (sdp->sd_log_tail != sdp->sd_log_flush_tail && !sdp->sd_log_idle) { log_write_header(sdp, flags); } if (gfs2_withdrawn(sdp)) @@ -1016,9 +1104,7 @@ void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl, u32 flags) lops_after_commit(sdp, tr); gfs2_log_lock(sdp); - sdp->sd_log_head = sdp->sd_log_flush_head; sdp->sd_log_blks_reserved = 0; - sdp->sd_log_committed_revoke = 0; spin_lock(&sdp->sd_ail_lock); if (tr && !list_empty(&tr->tr_ail1_list)) { @@ -1033,10 +1119,7 @@ void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl, u32 flags) empty_ail1_list(sdp); if (gfs2_withdrawn(sdp)) goto out_withdraw; - atomic_dec(&sdp->sd_log_blks_free); /* Adjust for unreserved buffer */ - trace_gfs2_log_blocks(sdp, -1); log_write_header(sdp, flags); - sdp->sd_log_head = sdp->sd_log_flush_head; } if (flags & (GFS2_LOG_HEAD_FLUSH_SHUTDOWN | GFS2_LOG_HEAD_FLUSH_FREEZE)) @@ -1046,12 +1129,22 @@ void gfs2_log_flush(struct gfs2_sbd *sdp, struct gfs2_glock *gl, u32 flags) } out_end: - trace_gfs2_log_flush(sdp, 0, flags); + used_blocks = log_distance(sdp, sdp->sd_log_flush_head, first_log_head); + reserved_revokes += atomic_read(&sdp->sd_log_revokes_available); + atomic_set(&sdp->sd_log_revokes_available, sdp->sd_ldptrs); + gfs2_assert_withdraw(sdp, reserved_revokes % sdp->sd_inptrs == sdp->sd_ldptrs); + if (reserved_revokes > sdp->sd_ldptrs) + reserved_blocks += (reserved_revokes - sdp->sd_ldptrs) / sdp->sd_inptrs; out: + if (used_blocks != reserved_blocks) { + gfs2_assert_withdraw_delayed(sdp, used_blocks < reserved_blocks); + gfs2_log_release(sdp, reserved_blocks - used_blocks); + } up_write(&sdp->sd_log_flush_lock); gfs2_trans_free(sdp, tr); if (gfs2_withdrawing(sdp)) gfs2_withdraw(sdp); + trace_gfs2_log_flush(sdp, 0, flags); return; out_withdraw: @@ -1087,8 +1180,8 @@ static void gfs2_merge_trans(struct gfs2_sbd *sdp, struct gfs2_trans *new) old->tr_num_databuf_new += new->tr_num_databuf_new; old->tr_num_buf_rm += new->tr_num_buf_rm; old->tr_num_databuf_rm += new->tr_num_databuf_rm; + old->tr_revokes += new->tr_revokes; old->tr_num_revoke += new->tr_num_revoke; - old->tr_num_revoke_rm += new->tr_num_revoke_rm; list_splice_tail_init(&new->tr_databuf, &old->tr_databuf); list_splice_tail_init(&new->tr_buf, &old->tr_buf); @@ -1110,20 +1203,17 @@ static void log_refund(struct gfs2_sbd *sdp, struct gfs2_trans *tr) if (sdp->sd_log_tr) { gfs2_merge_trans(sdp, tr); } else if (tr->tr_num_buf_new || tr->tr_num_databuf_new) { - gfs2_assert_withdraw(sdp, test_bit(TR_ALLOCED, &tr->tr_flags)); + gfs2_assert_withdraw(sdp, !test_bit(TR_ONSTACK, &tr->tr_flags)); sdp->sd_log_tr = tr; set_bit(TR_ATTACHED, &tr->tr_flags); } - sdp->sd_log_committed_revoke += tr->tr_num_revoke - tr->tr_num_revoke_rm; reserved = calc_reserved(sdp); maxres = sdp->sd_log_blks_reserved + tr->tr_reserved; gfs2_assert_withdraw(sdp, maxres >= reserved); unused = maxres - reserved; - atomic_add(unused, &sdp->sd_log_blks_free); - trace_gfs2_log_blocks(sdp, unused); - gfs2_assert_withdraw(sdp, atomic_read(&sdp->sd_log_blks_free) <= - sdp->sd_jdesc->jd_blocks); + if (unused) + gfs2_log_release(sdp, unused); sdp->sd_log_blks_reserved = reserved; gfs2_log_unlock(sdp); @@ -1166,15 +1256,11 @@ static void gfs2_log_shutdown(struct gfs2_sbd *sdp) gfs2_assert_withdraw(sdp, !sdp->sd_log_num_revoke); gfs2_assert_withdraw(sdp, list_empty(&sdp->sd_ail1_list)); - sdp->sd_log_flush_head = sdp->sd_log_head; - log_write_header(sdp, GFS2_LOG_HEAD_UNMOUNT | GFS2_LFC_SHUTDOWN); + log_pull_tail(sdp); gfs2_assert_warn(sdp, sdp->sd_log_head == sdp->sd_log_tail); gfs2_assert_warn(sdp, list_empty(&sdp->sd_ail2_list)); - - sdp->sd_log_head = sdp->sd_log_flush_head; - sdp->sd_log_tail = sdp->sd_log_head; } static inline int gfs2_jrnl_flush_reqd(struct gfs2_sbd *sdp) @@ -1208,7 +1294,6 @@ int gfs2_logd(void *data) struct gfs2_sbd *sdp = data; unsigned long t = 1; DEFINE_WAIT(wait); - bool did_flush; while (!kthread_should_stop()) { @@ -1227,12 +1312,10 @@ int gfs2_logd(void *data) continue; } - did_flush = false; if (gfs2_jrnl_flush_reqd(sdp) || t == 0) { gfs2_ail1_empty(sdp, 0); gfs2_log_flush(sdp, NULL, GFS2_LOG_HEAD_FLUSH_NORMAL | - GFS2_LFC_LOGD_JFLUSH_REQD); - did_flush = true; + GFS2_LFC_LOGD_JFLUSH_REQD); } if (gfs2_ail_flush_reqd(sdp)) { @@ -1240,13 +1323,9 @@ int gfs2_logd(void *data) gfs2_ail1_wait(sdp); gfs2_ail1_empty(sdp, 0); gfs2_log_flush(sdp, NULL, GFS2_LOG_HEAD_FLUSH_NORMAL | - GFS2_LFC_LOGD_AIL_FLUSH_REQD); - did_flush = true; + GFS2_LFC_LOGD_AIL_FLUSH_REQD); } - if (!gfs2_ail_flush_reqd(sdp) || did_flush) - wake_up(&sdp->sd_log_waitq); - t = gfs2_tune_get(sdp, gt_logd_secs) * HZ; try_to_freeze(); diff --git a/fs/gfs2/log.h b/fs/gfs2/log.h index 79f97290146e..eea58015710e 100644 --- a/fs/gfs2/log.h +++ b/fs/gfs2/log.h @@ -13,6 +13,13 @@ #include "incore.h" #include "inode.h" +/* + * The minimum amount of log space required for a log flush is one block for + * revokes and one block for the log header. Log flushes other than + * GFS2_LOG_HEAD_FLUSH_NORMAL may write one or two more log headers. + */ +#define GFS2_LOG_FLUSH_MIN_BLOCKS 4 + /** * gfs2_log_lock - acquire the right to mess with the log manager * @sdp: the filesystem @@ -43,7 +50,9 @@ static inline void gfs2_log_pointers_init(struct gfs2_sbd *sdp, if (++value == sdp->sd_jdesc->jd_blocks) { value = 0; } - sdp->sd_log_head = sdp->sd_log_tail = value; + sdp->sd_log_tail = value; + sdp->sd_log_flush_tail = value; + sdp->sd_log_head = value; } static inline void gfs2_ordered_add_inode(struct gfs2_inode *ip) @@ -64,8 +73,13 @@ static inline void gfs2_ordered_add_inode(struct gfs2_inode *ip) extern void gfs2_ordered_del_inode(struct gfs2_inode *ip); extern unsigned int gfs2_struct2blk(struct gfs2_sbd *sdp, unsigned int nstruct); extern void gfs2_remove_from_ail(struct gfs2_bufdata *bd); +extern bool gfs2_log_is_empty(struct gfs2_sbd *sdp); +extern void gfs2_log_release_revokes(struct gfs2_sbd *sdp, unsigned int revokes); extern void gfs2_log_release(struct gfs2_sbd *sdp, unsigned int blks); -extern int gfs2_log_reserve(struct gfs2_sbd *sdp, unsigned int blks); +extern bool gfs2_log_try_reserve(struct gfs2_sbd *sdp, struct gfs2_trans *tr, + unsigned int *extra_revokes); +extern void gfs2_log_reserve(struct gfs2_sbd *sdp, struct gfs2_trans *tr, + unsigned int *extra_revokes); extern void gfs2_write_log_header(struct gfs2_sbd *sdp, struct gfs2_jdesc *jd, u64 seq, u32 tail, u32 lblock, u32 flags, int op_flags); @@ -78,6 +92,6 @@ extern void log_flush_wait(struct gfs2_sbd *sdp); extern int gfs2_logd(void *data); extern void gfs2_add_revoke(struct gfs2_sbd *sdp, struct gfs2_bufdata *bd); extern void gfs2_glock_remove_revoke(struct gfs2_glock *gl); -extern void gfs2_write_revokes(struct gfs2_sbd *sdp); +extern void gfs2_flush_revokes(struct gfs2_sbd *sdp); #endif /* __LOG_DOT_H__ */ diff --git a/fs/gfs2/lops.c b/fs/gfs2/lops.c index 3922b26264f5..a82f4747aa8d 100644 --- a/fs/gfs2/lops.c +++ b/fs/gfs2/lops.c @@ -76,15 +76,20 @@ static void maybe_release_space(struct gfs2_bufdata *bd) unsigned int index = bd->bd_bh->b_blocknr - gl->gl_name.ln_number; struct gfs2_bitmap *bi = rgd->rd_bits + index; + rgrp_lock_local(rgd); if (bi->bi_clone == NULL) - return; + goto out; if (sdp->sd_args.ar_discard) gfs2_rgrp_send_discards(sdp, rgd->rd_data0, bd->bd_bh, bi, 1, NULL); memcpy(bi->bi_clone + bi->bi_offset, bd->bd_bh->b_data + bi->bi_offset, bi->bi_bytes); clear_bit(GBF_FULL, &bi->bi_flags); rgd->rd_free_clone = rgd->rd_free; + BUG_ON(rgd->rd_free_clone < rgd->rd_reserved); rgd->rd_extfail_pt = rgd->rd_free; + +out: + rgrp_unlock_local(rgd); } /** @@ -262,7 +267,7 @@ static struct bio *gfs2_log_alloc_bio(struct gfs2_sbd *sdp, u64 blkno, bio_end_io_t *end_io) { struct super_block *sb = sdp->sd_vfs; - struct bio *bio = bio_alloc(GFP_NOIO, BIO_MAX_PAGES); + struct bio *bio = bio_alloc(GFP_NOIO, BIO_MAX_VECS); bio->bi_iter.bi_sector = blkno << sdp->sd_fsb2bb_shift; bio_set_dev(bio, sb->s_bdev); @@ -322,17 +327,18 @@ static struct bio *gfs2_log_get_bio(struct gfs2_sbd *sdp, u64 blkno, * then add the page segment to that. */ -void gfs2_log_write(struct gfs2_sbd *sdp, struct page *page, - unsigned size, unsigned offset, u64 blkno) +void gfs2_log_write(struct gfs2_sbd *sdp, struct gfs2_jdesc *jd, + struct page *page, unsigned size, unsigned offset, + u64 blkno) { struct bio *bio; int ret; - bio = gfs2_log_get_bio(sdp, blkno, &sdp->sd_log_bio, REQ_OP_WRITE, + bio = gfs2_log_get_bio(sdp, blkno, &jd->jd_log_bio, REQ_OP_WRITE, gfs2_end_log_write, false); ret = bio_add_page(bio, page, size, offset); if (ret == 0) { - bio = gfs2_log_get_bio(sdp, blkno, &sdp->sd_log_bio, + bio = gfs2_log_get_bio(sdp, blkno, &jd->jd_log_bio, REQ_OP_WRITE, gfs2_end_log_write, true); ret = bio_add_page(bio, page, size, offset); WARN_ON(ret == 0); @@ -355,7 +361,8 @@ static void gfs2_log_write_bh(struct gfs2_sbd *sdp, struct buffer_head *bh) dblock = gfs2_log_bmap(sdp->sd_jdesc, sdp->sd_log_flush_head); gfs2_log_incr_head(sdp); - gfs2_log_write(sdp, bh->b_page, bh->b_size, bh_offset(bh), dblock); + gfs2_log_write(sdp, sdp->sd_jdesc, bh->b_page, bh->b_size, + bh_offset(bh), dblock); } /** @@ -369,14 +376,14 @@ static void gfs2_log_write_bh(struct gfs2_sbd *sdp, struct buffer_head *bh) * the page may be freed at any time. */ -void gfs2_log_write_page(struct gfs2_sbd *sdp, struct page *page) +static void gfs2_log_write_page(struct gfs2_sbd *sdp, struct page *page) { struct super_block *sb = sdp->sd_vfs; u64 dblock; dblock = gfs2_log_bmap(sdp->sd_jdesc, sdp->sd_log_flush_head); gfs2_log_incr_head(sdp); - gfs2_log_write(sdp, page, sb->s_blocksize, 0, dblock); + gfs2_log_write(sdp, sdp->sd_jdesc, page, sb->s_blocksize, 0, dblock); } /** @@ -845,7 +852,7 @@ static void revoke_lo_before_commit(struct gfs2_sbd *sdp, struct gfs2_trans *tr) struct page *page; unsigned int length; - gfs2_write_revokes(sdp); + gfs2_flush_revokes(sdp); if (!sdp->sd_log_num_revoke) return; @@ -857,7 +864,6 @@ static void revoke_lo_before_commit(struct gfs2_sbd *sdp, struct gfs2_trans *tr) sdp->sd_log_num_revoke--; if (offset + sizeof(u64) > sdp->sd_sb.sb_bsize) { - gfs2_log_write_page(sdp, page); page = mempool_alloc(gfs2_page_pool, GFP_NOIO); mh = page_address(page); diff --git a/fs/gfs2/lops.h b/fs/gfs2/lops.h index fbdbb08dcec6..31b6dd0d2e5d 100644 --- a/fs/gfs2/lops.h +++ b/fs/gfs2/lops.h @@ -10,37 +10,24 @@ #include <linux/list.h> #include "incore.h" -#define BUF_OFFSET \ - ((sizeof(struct gfs2_log_descriptor) + sizeof(__be64) - 1) & \ - ~(sizeof(__be64) - 1)) -#define DATABUF_OFFSET \ - ((sizeof(struct gfs2_log_descriptor) + (2 * sizeof(__be64) - 1)) & \ - ~(2 * sizeof(__be64) - 1)) - extern const struct gfs2_log_operations *gfs2_log_ops[]; extern void gfs2_log_incr_head(struct gfs2_sbd *sdp); extern u64 gfs2_log_bmap(struct gfs2_jdesc *jd, unsigned int lbn); -extern void gfs2_log_write(struct gfs2_sbd *sdp, struct page *page, - unsigned size, unsigned offset, u64 blkno); -extern void gfs2_log_write_page(struct gfs2_sbd *sdp, struct page *page); +extern void gfs2_log_write(struct gfs2_sbd *sdp, struct gfs2_jdesc *jd, + struct page *page, unsigned size, unsigned offset, + u64 blkno); extern void gfs2_log_submit_bio(struct bio **biop, int opf); extern void gfs2_pin(struct gfs2_sbd *sdp, struct buffer_head *bh); extern int gfs2_find_jhead(struct gfs2_jdesc *jd, struct gfs2_log_header_host *head, bool keep_cache); static inline unsigned int buf_limit(struct gfs2_sbd *sdp) { - unsigned int limit; - - limit = (sdp->sd_sb.sb_bsize - BUF_OFFSET) / sizeof(__be64); - return limit; + return sdp->sd_ldptrs; } static inline unsigned int databuf_limit(struct gfs2_sbd *sdp) { - unsigned int limit; - - limit = (sdp->sd_sb.sb_bsize - DATABUF_OFFSET) / (2 * sizeof(__be64)); - return limit; + return sdp->sd_ldptrs / 2; } static inline void lops_before_commit(struct gfs2_sbd *sdp, diff --git a/fs/gfs2/main.c b/fs/gfs2/main.c index c7393ee9cf68..28d0eb23e18e 100644 --- a/fs/gfs2/main.c +++ b/fs/gfs2/main.c @@ -98,7 +98,7 @@ static int __init init_gfs2_fs(void) error = -ENOMEM; gfs2_glock_cachep = kmem_cache_create("gfs2_glock", sizeof(struct gfs2_glock), - 0, 0, + 0, SLAB_RECLAIM_ACCOUNT, gfs2_init_glock_once); if (!gfs2_glock_cachep) goto fail_cachep1; @@ -134,7 +134,7 @@ static int __init init_gfs2_fs(void) gfs2_quotad_cachep = kmem_cache_create("gfs2_quotad", sizeof(struct gfs2_quota_data), - 0, 0, NULL); + 0, SLAB_RECLAIM_ACCOUNT, NULL); if (!gfs2_quotad_cachep) goto fail_cachep6; diff --git a/fs/gfs2/ops_fstype.c b/fs/gfs2/ops_fstype.c index 61fce59cb4d3..aa4136055a83 100644 --- a/fs/gfs2/ops_fstype.c +++ b/fs/gfs2/ops_fstype.c @@ -136,8 +136,6 @@ static struct gfs2_sbd *init_sbd(struct super_block *sb) init_rwsem(&sdp->sd_log_flush_lock); atomic_set(&sdp->sd_log_in_flight, 0); - atomic_set(&sdp->sd_reserving_log, 0); - init_waitqueue_head(&sdp->sd_reserving_log_wait); init_waitqueue_head(&sdp->sd_log_flush_wait); atomic_set(&sdp->sd_freeze_state, SFS_UNFROZEN); mutex_init(&sdp->sd_freeze_mutex); @@ -171,7 +169,8 @@ static int gfs2_check_sb(struct gfs2_sbd *sdp, int silent) return -EINVAL; } - if (sb->sb_fs_format != GFS2_FORMAT_FS || + if (sb->sb_fs_format < GFS2_FS_FORMAT_MIN || + sb->sb_fs_format > GFS2_FS_FORMAT_MAX || sb->sb_multihost_format != GFS2_FORMAT_MULTI) { fs_warn(sdp, "Unknown on-disk format, unable to mount\n"); return -EINVAL; @@ -179,7 +178,7 @@ static int gfs2_check_sb(struct gfs2_sbd *sdp, int silent) if (sb->sb_bsize < 512 || sb->sb_bsize > PAGE_SIZE || (sb->sb_bsize & (sb->sb_bsize - 1))) { - pr_warn("Invalid superblock size\n"); + pr_warn("Invalid block size\n"); return -EINVAL; } @@ -317,6 +316,13 @@ static int gfs2_read_sb(struct gfs2_sbd *sdp, int silent) sizeof(struct gfs2_meta_header)) * GFS2_NBBY; /* not the rgrp bitmap, subsequent bitmaps only */ + /* + * We always keep at least one block reserved for revokes in + * transactions. This greatly simplifies allocating additional + * revoke blocks. + */ + atomic_set(&sdp->sd_log_revokes_available, sdp->sd_ldptrs); + /* Compute maximum reservation required to add a entry to a directory */ hash_blocks = DIV_ROUND_UP(sizeof(u64) * BIT(GFS2_DIR_MAX_DEPTH), @@ -488,6 +494,19 @@ static int init_sb(struct gfs2_sbd *sdp, int silent) goto out; } + switch(sdp->sd_sb.sb_fs_format) { + case GFS2_FS_FORMAT_MAX: + sb->s_xattr = gfs2_xattr_handlers_max; + break; + + case GFS2_FS_FORMAT_MIN: + sb->s_xattr = gfs2_xattr_handlers_min; + break; + + default: + BUG(); + } + /* Set up the buffer cache and SB for real */ if (sdp->sd_sb.sb_bsize < bdev_logical_block_size(sb->s_bdev)) { ret = -EINVAL; @@ -1032,13 +1051,14 @@ hostdata_error: } if (lm->lm_mount == NULL) { - fs_info(sdp, "Now mounting FS...\n"); + fs_info(sdp, "Now mounting FS (format %u)...\n", sdp->sd_sb.sb_fs_format); complete_all(&sdp->sd_locking_init); return 0; } ret = lm->lm_mount(sdp, table); if (ret == 0) - fs_info(sdp, "Joined cluster. Now mounting FS...\n"); + fs_info(sdp, "Joined cluster. Now mounting FS (format %u)...\n", + sdp->sd_sb.sb_fs_format); complete_all(&sdp->sd_locking_init); return ret; } @@ -1084,6 +1104,7 @@ static int gfs2_fill_super(struct super_block *sb, struct fs_context *fc) int silent = fc->sb_flags & SB_SILENT; struct gfs2_sbd *sdp; struct gfs2_holder mount_gh; + struct gfs2_holder freeze_gh; int error; sdp = init_sbd(sb); @@ -1107,7 +1128,6 @@ static int gfs2_fill_super(struct super_block *sb, struct fs_context *fc) sb->s_op = &gfs2_super_ops; sb->s_d_op = &gfs2_dops; sb->s_export_op = &gfs2_export_ops; - sb->s_xattr = gfs2_xattr_handlers; sb->s_qcop = &gfs2_quotactl_ops; sb->s_quota_types = QTYPE_MASK_USR | QTYPE_MASK_GRP; sb_dqopt(sb)->flags |= DQUOT_QUOTA_SYS_FILE; @@ -1156,6 +1176,10 @@ static int gfs2_fill_super(struct super_block *sb, struct fs_context *fc) if (error) goto fail_locking; + /* Turn rgrplvb on by default if fs format is recent enough */ + if (!sdp->sd_args.ar_got_rgrplvb && sdp->sd_sb.sb_fs_format > 1801) + sdp->sd_args.ar_rgrplvb = 1; + error = wait_on_journal(sdp); if (error) goto fail_sb; @@ -1195,25 +1219,18 @@ static int gfs2_fill_super(struct super_block *sb, struct fs_context *fc) goto fail_per_node; } - if (sb_rdonly(sb)) { - struct gfs2_holder freeze_gh; + error = gfs2_freeze_lock(sdp, &freeze_gh, 0); + if (error) + goto fail_per_node; - error = gfs2_glock_nq_init(sdp->sd_freeze_gl, LM_ST_SHARED, - LM_FLAG_NOEXP | GL_EXACT, - &freeze_gh); - if (error) { - fs_err(sdp, "can't make FS RO: %d\n", error); - goto fail_per_node; - } - gfs2_glock_dq_uninit(&freeze_gh); - } else { + if (!sb_rdonly(sb)) error = gfs2_make_fs_rw(sdp); - if (error) { - fs_err(sdp, "can't make FS RW: %d\n", error); - goto fail_per_node; - } - } + gfs2_freeze_unlock(&freeze_gh); + if (error) { + fs_err(sdp, "can't make FS RW: %d\n", error); + goto fail_per_node; + } gfs2_glock_dq_uninit(&mount_gh); gfs2_online_uevent(sdp); return 0; @@ -1456,6 +1473,7 @@ static int gfs2_parse_param(struct fs_context *fc, struct fs_parameter *param) break; case Opt_rgrplvb: args->ar_rgrplvb = result.boolean; + args->ar_got_rgrplvb = 1; break; case Opt_loccookie: args->ar_loccookie = result.boolean; @@ -1514,15 +1532,20 @@ static int gfs2_reconfigure(struct fs_context *fc) fc->sb_flags |= SB_RDONLY; if ((sb->s_flags ^ fc->sb_flags) & SB_RDONLY) { + struct gfs2_holder freeze_gh; + + error = gfs2_freeze_lock(sdp, &freeze_gh, 0); + if (error) + return -EINVAL; + if (fc->sb_flags & SB_RDONLY) { - error = gfs2_make_fs_ro(sdp); - if (error) - errorfc(fc, "unable to remount read-only"); + gfs2_make_fs_ro(sdp); } else { error = gfs2_make_fs_rw(sdp); if (error) errorfc(fc, "unable to remount read-write"); } + gfs2_freeze_unlock(&freeze_gh); } sdp->sd_args = *newargs; diff --git a/fs/gfs2/recovery.c b/fs/gfs2/recovery.c index c26c68ebd29d..282173774005 100644 --- a/fs/gfs2/recovery.c +++ b/fs/gfs2/recovery.c @@ -470,9 +470,7 @@ void gfs2_recover_func(struct work_struct *work) /* Acquire a shared hold on the freeze lock */ - error = gfs2_glock_nq_init(sdp->sd_freeze_gl, LM_ST_SHARED, - LM_FLAG_NOEXP | LM_FLAG_PRIORITY | - GL_EXACT, &thaw_gh); + error = gfs2_freeze_lock(sdp, &thaw_gh, LM_FLAG_PRIORITY); if (error) goto fail_gunlock_ji; @@ -507,22 +505,24 @@ void gfs2_recover_func(struct work_struct *work) /* We take the sd_log_flush_lock here primarily to prevent log * flushes and simultaneous journal replays from stomping on - * each other wrt sd_log_bio. */ + * each other wrt jd_log_bio. */ down_read(&sdp->sd_log_flush_lock); for (pass = 0; pass < 2; pass++) { lops_before_scan(jd, &head, pass); error = foreach_descriptor(jd, head.lh_tail, head.lh_blkno, pass); lops_after_scan(jd, error, pass); - if (error) + if (error) { + up_read(&sdp->sd_log_flush_lock); goto fail_gunlock_thaw; + } } recover_local_statfs(jd, &head); clean_journal(jd, &head); up_read(&sdp->sd_log_flush_lock); - gfs2_glock_dq_uninit(&thaw_gh); + gfs2_freeze_unlock(&thaw_gh); t_rep = ktime_get(); fs_info(sdp, "jid=%u: Journal replayed in %lldms [jlck:%lldms, " "jhead:%lldms, tlck:%lldms, replay:%lldms]\n", @@ -544,7 +544,7 @@ void gfs2_recover_func(struct work_struct *work) goto done; fail_gunlock_thaw: - gfs2_glock_dq_uninit(&thaw_gh); + gfs2_freeze_unlock(&thaw_gh); fail_gunlock_ji: if (jlocked) { gfs2_glock_dq_uninit(&ji_gh); diff --git a/fs/gfs2/rgrp.c b/fs/gfs2/rgrp.c index 5e8eef9990e3..89c37a845e64 100644 --- a/fs/gfs2/rgrp.c +++ b/fs/gfs2/rgrp.c @@ -36,6 +36,24 @@ #define BFITNOENT ((u32)~0) #define NO_BLOCK ((u64)~0) +struct gfs2_rbm { + struct gfs2_rgrpd *rgd; + u32 offset; /* The offset is bitmap relative */ + int bii; /* Bitmap index */ +}; + +static inline struct gfs2_bitmap *rbm_bi(const struct gfs2_rbm *rbm) +{ + return rbm->rgd->rd_bits + rbm->bii; +} + +static inline u64 gfs2_rbm_to_block(const struct gfs2_rbm *rbm) +{ + BUG_ON(rbm->offset >= rbm->rgd->rd_data); + return rbm->rgd->rd_data0 + (rbm_bi(rbm)->bi_start * GFS2_NBBY) + + rbm->offset; +} + /* * These routines are used by the resource group routines (rgrp.c) * to keep track of block allocation. Each block is represented by two @@ -61,7 +79,7 @@ static const char valid_change[16] = { }; static int gfs2_rbm_find(struct gfs2_rbm *rbm, u8 state, u32 *minext, - const struct gfs2_inode *ip, bool nowrap); + struct gfs2_blkreserv *rs, bool nowrap); /** @@ -175,7 +193,7 @@ static inline u64 gfs2_bit_search(const __le64 *ptr, u64 mask, u8 state) /** * rs_cmp - multi-block reservation range compare - * @blk: absolute file system block number of the new reservation + * @start: start of the new reservation * @len: number of blocks in the new reservation * @rs: existing reservation to compare against * @@ -183,13 +201,11 @@ static inline u64 gfs2_bit_search(const __le64 *ptr, u64 mask, u8 state) * -1 if the block range is before the start of the reservation * 0 if the block range overlaps with the reservation */ -static inline int rs_cmp(u64 blk, u32 len, struct gfs2_blkreserv *rs) +static inline int rs_cmp(u64 start, u32 len, struct gfs2_blkreserv *rs) { - u64 startblk = gfs2_rbm_to_block(&rs->rs_rbm); - - if (blk >= startblk + rs->rs_free) + if (start >= rs->rs_start + rs->rs_requested) return 1; - if (blk + len - 1 < startblk) + if (rs->rs_start >= start + len) return -1; return 0; } @@ -277,29 +293,38 @@ static int gfs2_rbm_from_block(struct gfs2_rbm *rbm, u64 block) } /** - * gfs2_rbm_incr - increment an rbm structure + * gfs2_rbm_add - add a number of blocks to an rbm * @rbm: The rbm with rgd already set correctly + * @blocks: The number of blocks to add to rpm * - * This function takes an existing rbm structure and increments it to the next - * viable block offset. - * - * Returns: If incrementing the offset would cause the rbm to go past the - * end of the rgrp, true is returned, otherwise false. + * This function takes an existing rbm structure and adds a number of blocks to + * it. * + * Returns: True if the new rbm would point past the end of the rgrp. */ -static bool gfs2_rbm_incr(struct gfs2_rbm *rbm) +static bool gfs2_rbm_add(struct gfs2_rbm *rbm, u32 blocks) { - if (rbm->offset + 1 < rbm_bi(rbm)->bi_blocks) { /* in the same bitmap */ - rbm->offset++; + struct gfs2_rgrpd *rgd = rbm->rgd; + struct gfs2_bitmap *bi = rgd->rd_bits + rbm->bii; + + if (rbm->offset + blocks < bi->bi_blocks) { + rbm->offset += blocks; return false; } - if (rbm->bii == rbm->rgd->rd_length - 1) /* at the last bitmap */ - return true; + blocks -= bi->bi_blocks - rbm->offset; - rbm->offset = 0; - rbm->bii++; - return false; + for(;;) { + bi++; + if (bi == rgd->rd_bits + rgd->rd_length) + return true; + if (blocks < bi->bi_blocks) { + rbm->offset = blocks; + rbm->bii = bi - rgd->rd_bits; + return false; + } + blocks -= bi->bi_blocks; + } } /** @@ -308,7 +333,8 @@ static bool gfs2_rbm_incr(struct gfs2_rbm *rbm) * @n_unaligned: Number of unaligned blocks to check * @len: Decremented for each block found (terminate on zero) * - * Returns: true if a non-free block is encountered + * Returns: true if a non-free block is encountered or the end of the resource + * group is reached. */ static bool gfs2_unaligned_extlen(struct gfs2_rbm *rbm, u32 n_unaligned, u32 *len) @@ -323,7 +349,7 @@ static bool gfs2_unaligned_extlen(struct gfs2_rbm *rbm, u32 n_unaligned, u32 *le (*len)--; if (*len == 0) return true; - if (gfs2_rbm_incr(rbm)) + if (gfs2_rbm_add(rbm, 1)) return true; } @@ -595,10 +621,11 @@ static void dump_rs(struct seq_file *seq, const struct gfs2_blkreserv *rs, { struct gfs2_inode *ip = container_of(rs, struct gfs2_inode, i_res); - gfs2_print_dbg(seq, "%s B: n:%llu s:%llu b:%u f:%u\n", fs_id_buf, + gfs2_print_dbg(seq, "%s B: n:%llu s:%llu f:%u\n", + fs_id_buf, (unsigned long long)ip->i_no_addr, - (unsigned long long)gfs2_rbm_to_block(&rs->rs_rbm), - rs->rs_rbm.offset, rs->rs_free); + (unsigned long long)rs->rs_start, + rs->rs_requested); } /** @@ -613,33 +640,22 @@ static void __rs_deltree(struct gfs2_blkreserv *rs) if (!gfs2_rs_active(rs)) return; - rgd = rs->rs_rbm.rgd; + rgd = rs->rs_rgd; trace_gfs2_rs(rs, TRACE_RS_TREEDEL); rb_erase(&rs->rs_node, &rgd->rd_rstree); RB_CLEAR_NODE(&rs->rs_node); - if (rs->rs_free) { - u64 last_block = gfs2_rbm_to_block(&rs->rs_rbm) + - rs->rs_free - 1; - struct gfs2_rbm last_rbm = { .rgd = rs->rs_rbm.rgd, }; - struct gfs2_bitmap *start, *last; + if (rs->rs_requested) { + /* return requested blocks to the rgrp */ + BUG_ON(rs->rs_rgd->rd_requested < rs->rs_requested); + rs->rs_rgd->rd_requested -= rs->rs_requested; - /* return reserved blocks to the rgrp */ - BUG_ON(rs->rs_rbm.rgd->rd_reserved < rs->rs_free); - rs->rs_rbm.rgd->rd_reserved -= rs->rs_free; /* The rgrp extent failure point is likely not to increase; it will only do so if the freed blocks are somehow contiguous with a span of free blocks that follows. Still, it will force the number to be recalculated later. */ - rgd->rd_extfail_pt += rs->rs_free; - rs->rs_free = 0; - if (gfs2_rbm_from_block(&last_rbm, last_block)) - return; - start = rbm_bi(&rs->rs_rbm); - last = rbm_bi(&last_rbm); - do - clear_bit(GBF_FULL, &start->bi_flags); - while (start++ != last); + rgd->rd_extfail_pt += rs->rs_requested; + rs->rs_requested = 0; } } @@ -652,11 +668,11 @@ void gfs2_rs_deltree(struct gfs2_blkreserv *rs) { struct gfs2_rgrpd *rgd; - rgd = rs->rs_rbm.rgd; + rgd = rs->rs_rgd; if (rgd) { spin_lock(&rgd->rd_rsspin); __rs_deltree(rs); - BUG_ON(rs->rs_free); + BUG_ON(rs->rs_requested); spin_unlock(&rgd->rd_rsspin); } } @@ -904,6 +920,7 @@ static int read_rindex_entry(struct gfs2_inode *ip) rgd->rd_data = be32_to_cpu(buf.ri_data); rgd->rd_bitbytes = be32_to_cpu(buf.ri_bitbytes); spin_lock_init(&rgd->rd_rsspin); + mutex_init(&rgd->rd_mutex); error = compute_bitstructs(rgd); if (error) @@ -1149,6 +1166,23 @@ static u32 count_unlinked(struct gfs2_rgrpd *rgd) return count; } +static void rgrp_set_bitmap_flags(struct gfs2_rgrpd *rgd) +{ + struct gfs2_bitmap *bi; + int x; + + if (rgd->rd_free) { + for (x = 0; x < rgd->rd_length; x++) { + bi = rgd->rd_bits + x; + clear_bit(GBF_FULL, &bi->bi_flags); + } + } else { + for (x = 0; x < rgd->rd_length; x++) { + bi = rgd->rd_bits + x; + set_bit(GBF_FULL, &bi->bi_flags); + } + } +} /** * gfs2_rgrp_bh_get - Read in a RG's header and bitmaps @@ -1192,11 +1226,11 @@ static int gfs2_rgrp_bh_get(struct gfs2_rgrpd *rgd) } if (!(rgd->rd_flags & GFS2_RDF_UPTODATE)) { - for (x = 0; x < length; x++) - clear_bit(GBF_FULL, &rgd->rd_bits[x].bi_flags); gfs2_rgrp_in(rgd, (rgd->rd_bits[0].bi_bh)->b_data); + rgrp_set_bitmap_flags(rgd); rgd->rd_flags |= (GFS2_RDF_UPTODATE | GFS2_RDF_CHECK); rgd->rd_free_clone = rgd->rd_free; + BUG_ON(rgd->rd_reserved); /* max out the rgrp allocation failure point */ rgd->rd_extfail_pt = rgd->rd_free; } @@ -1244,7 +1278,11 @@ static int update_rgrp_lvb(struct gfs2_rgrpd *rgd) if (rgd->rd_rgl->rl_unlinked == 0) rgd->rd_flags &= ~GFS2_RDF_CHECK; rgd->rd_free = be32_to_cpu(rgd->rd_rgl->rl_free); + rgrp_set_bitmap_flags(rgd); rgd->rd_free_clone = rgd->rd_free; + BUG_ON(rgd->rd_reserved); + /* max out the rgrp allocation failure point */ + rgd->rd_extfail_pt = rgd->rd_free; rgd->rd_dinodes = be32_to_cpu(rgd->rd_rgl->rl_dinodes); rgd->rd_igeneration = be64_to_cpu(rgd->rd_rgl->rl_igeneration); return 0; @@ -1404,7 +1442,8 @@ int gfs2_fitrim(struct file *filp, void __user *argp) while (1) { - ret = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0, &gh); + ret = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, + LM_FLAG_NODE_SCOPE, &gh); if (ret) goto out; @@ -1412,9 +1451,11 @@ int gfs2_fitrim(struct file *filp, void __user *argp) /* Trim each bitmap in the rgrp */ for (x = 0; x < rgd->rd_length; x++) { struct gfs2_bitmap *bi = rgd->rd_bits + x; + rgrp_lock_local(rgd); ret = gfs2_rgrp_send_discards(sdp, rgd->rd_data0, NULL, bi, minlen, &amt); + rgrp_unlock_local(rgd); if (ret) { gfs2_glock_dq_uninit(&gh); goto out; @@ -1426,9 +1467,11 @@ int gfs2_fitrim(struct file *filp, void __user *argp) ret = gfs2_trans_begin(sdp, RES_RG_HDR, 0); if (ret == 0) { bh = rgd->rd_bits[0].bi_bh; + rgrp_lock_local(rgd); rgd->rd_flags |= GFS2_RGF_TRIMMED; gfs2_trans_add_meta(rgd->rd_gl, bh); gfs2_rgrp_out(rgd, bh->b_data); + rgrp_unlock_local(rgd); gfs2_trans_end(sdp); } } @@ -1458,8 +1501,7 @@ static void rs_insert(struct gfs2_inode *ip) struct rb_node **newn, *parent = NULL; int rc; struct gfs2_blkreserv *rs = &ip->i_res; - struct gfs2_rgrpd *rgd = rs->rs_rbm.rgd; - u64 fsblock = gfs2_rbm_to_block(&rs->rs_rbm); + struct gfs2_rgrpd *rgd = rs->rs_rgd; BUG_ON(gfs2_rs_active(rs)); @@ -1470,7 +1512,7 @@ static void rs_insert(struct gfs2_inode *ip) rb_entry(*newn, struct gfs2_blkreserv, rs_node); parent = *newn; - rc = rs_cmp(fsblock, rs->rs_free, cur); + rc = rs_cmp(rs->rs_start, rs->rs_requested, cur); if (rc > 0) newn = &((*newn)->rb_right); else if (rc < 0) @@ -1486,7 +1528,7 @@ static void rs_insert(struct gfs2_inode *ip) rb_insert_color(&rs->rs_node, &rgd->rd_rstree); /* Do our rgrp accounting for the reservation */ - rgd->rd_reserved += rs->rs_free; /* blocks reserved */ + rgd->rd_requested += rs->rs_requested; /* blocks requested */ spin_unlock(&rgd->rd_rsspin); trace_gfs2_rs(rs, TRACE_RS_INSERT); } @@ -1507,9 +1549,9 @@ static inline u32 rgd_free(struct gfs2_rgrpd *rgd, struct gfs2_blkreserv *rs) { u32 tot_reserved, tot_free; - if (WARN_ON_ONCE(rgd->rd_reserved < rs->rs_free)) + if (WARN_ON_ONCE(rgd->rd_requested < rs->rs_requested)) return 0; - tot_reserved = rgd->rd_reserved - rs->rs_free; + tot_reserved = rgd->rd_requested - rs->rs_requested; if (rgd->rd_free_clone < tot_reserved) tot_reserved = 0; @@ -1534,17 +1576,26 @@ static void rg_mblk_search(struct gfs2_rgrpd *rgd, struct gfs2_inode *ip, u64 goal; struct gfs2_blkreserv *rs = &ip->i_res; u32 extlen; - u32 free_blocks = rgd_free(rgd, rs); + u32 free_blocks, blocks_available; int ret; struct inode *inode = &ip->i_inode; + spin_lock(&rgd->rd_rsspin); + free_blocks = rgd_free(rgd, rs); + if (rgd->rd_free_clone < rgd->rd_requested) + free_blocks = 0; + blocks_available = rgd->rd_free_clone - rgd->rd_reserved; + if (rgd == rs->rs_rgd) + blocks_available += rs->rs_reserved; + spin_unlock(&rgd->rd_rsspin); + if (S_ISDIR(inode->i_mode)) extlen = 1; else { extlen = max_t(u32, atomic_read(&ip->i_sizehint), ap->target); extlen = clamp(extlen, (u32)RGRP_RSRV_MINBLKS, free_blocks); } - if ((rgd->rd_free_clone < rgd->rd_reserved) || (free_blocks < extlen)) + if (free_blocks < extlen || blocks_available < extlen) return; /* Find bitmap block that contains bits for goal block */ @@ -1556,10 +1607,10 @@ static void rg_mblk_search(struct gfs2_rgrpd *rgd, struct gfs2_inode *ip, if (WARN_ON(gfs2_rbm_from_block(&rbm, goal))) return; - ret = gfs2_rbm_find(&rbm, GFS2_BLKST_FREE, &extlen, ip, true); + ret = gfs2_rbm_find(&rbm, GFS2_BLKST_FREE, &extlen, &ip->i_res, true); if (ret == 0) { - rs->rs_rbm = rbm; - rs->rs_free = extlen; + rs->rs_start = gfs2_rbm_to_block(&rbm); + rs->rs_requested = extlen; rs_insert(ip); } else { if (goal == rgd->rd_last_alloc + rgd->rd_data0) @@ -1572,7 +1623,7 @@ static void rg_mblk_search(struct gfs2_rgrpd *rgd, struct gfs2_inode *ip, * @rgd: The resource group * @block: The starting block * @length: The required length - * @ip: Ignore any reservations for this inode + * @ignore_rs: Reservation to ignore * * If the block does not appear in any reservation, then return the * block number unchanged. If it does appear in the reservation, then @@ -1582,7 +1633,7 @@ static void rg_mblk_search(struct gfs2_rgrpd *rgd, struct gfs2_inode *ip, static u64 gfs2_next_unreserved_block(struct gfs2_rgrpd *rgd, u64 block, u32 length, - const struct gfs2_inode *ip) + struct gfs2_blkreserv *ignore_rs) { struct gfs2_blkreserv *rs; struct rb_node *n; @@ -1602,8 +1653,8 @@ static u64 gfs2_next_unreserved_block(struct gfs2_rgrpd *rgd, u64 block, } if (n) { - while ((rs_cmp(block, length, rs) == 0) && (&ip->i_res != rs)) { - block = gfs2_rbm_to_block(&rs->rs_rbm) + rs->rs_free; + while (rs_cmp(block, length, rs) == 0 && rs != ignore_rs) { + block = rs->rs_start + rs->rs_requested; n = n->rb_right; if (n == NULL) break; @@ -1618,7 +1669,7 @@ static u64 gfs2_next_unreserved_block(struct gfs2_rgrpd *rgd, u64 block, /** * gfs2_reservation_check_and_update - Check for reservations during block alloc * @rbm: The current position in the resource group - * @ip: The inode for which we are searching for blocks + * @rs: Our own reservation * @minext: The minimum extent length * @maxext: A pointer to the maximum extent structure * @@ -1632,20 +1683,19 @@ static u64 gfs2_next_unreserved_block(struct gfs2_rgrpd *rgd, u64 block, */ static int gfs2_reservation_check_and_update(struct gfs2_rbm *rbm, - const struct gfs2_inode *ip, + struct gfs2_blkreserv *rs, u32 minext, struct gfs2_extent *maxext) { u64 block = gfs2_rbm_to_block(rbm); u32 extlen = 1; u64 nblock; - int ret; /* * If we have a minimum extent length, then skip over any extent * which is less than the min extent length in size. */ - if (minext) { + if (minext > 1) { extlen = gfs2_free_extlen(rbm, minext); if (extlen <= maxext->len) goto fail; @@ -1655,7 +1705,7 @@ static int gfs2_reservation_check_and_update(struct gfs2_rbm *rbm, * Check the extent which has been found against the reservations * and skip if parts of it are already reserved */ - nblock = gfs2_next_unreserved_block(rbm->rgd, block, extlen, ip); + nblock = gfs2_next_unreserved_block(rbm->rgd, block, extlen, rs); if (nblock == block) { if (!minext || extlen >= minext) return 0; @@ -1664,12 +1714,15 @@ static int gfs2_reservation_check_and_update(struct gfs2_rbm *rbm, maxext->len = extlen; maxext->rbm = *rbm; } -fail: - nblock = block + extlen; + } else { + u64 len = nblock - block; + if (len >= (u64)1 << 32) + return -E2BIG; + extlen = len; } - ret = gfs2_rbm_from_block(rbm, nblock); - if (ret < 0) - return ret; +fail: + if (gfs2_rbm_add(rbm, extlen)) + return -E2BIG; return 1; } @@ -1677,9 +1730,9 @@ fail: * gfs2_rbm_find - Look for blocks of a particular state * @rbm: Value/result starting position and final position * @state: The state which we want to find - * @minext: Pointer to the requested extent length (NULL for a single block) + * @minext: Pointer to the requested extent length * This is updated to be the actual reservation size. - * @ip: If set, check for reservations + * @rs: Our own reservation (NULL to skip checking for reservations) * @nowrap: Stop looking at the end of the rgrp, rather than wrapping * around until we've reached the starting point. * @@ -1693,7 +1746,7 @@ fail: */ static int gfs2_rbm_find(struct gfs2_rbm *rbm, u8 state, u32 *minext, - const struct gfs2_inode *ip, bool nowrap) + struct gfs2_blkreserv *rs, bool nowrap) { bool scan_from_start = rbm->bii == 0 && rbm->offset == 0; struct buffer_head *bh; @@ -1714,8 +1767,7 @@ static int gfs2_rbm_find(struct gfs2_rbm *rbm, u8 state, u32 *minext, while(1) { bi = rbm_bi(rbm); - if ((ip == NULL || !gfs2_rs_active(&ip->i_res)) && - test_bit(GBF_FULL, &bi->bi_flags) && + if (test_bit(GBF_FULL, &bi->bi_flags) && (state == GFS2_BLKST_FREE)) goto next_bitmap; @@ -1731,11 +1783,10 @@ static int gfs2_rbm_find(struct gfs2_rbm *rbm, u8 state, u32 *minext, goto next_bitmap; } rbm->offset = offset; - if (ip == NULL) + if (!rs) return 0; - ret = gfs2_reservation_check_and_update(rbm, ip, - minext ? *minext : 0, + ret = gfs2_reservation_check_and_update(rbm, rs, *minext, &maxext); if (ret == 0) return 0; @@ -1767,7 +1818,7 @@ next_iter: break; } - if (minext == NULL || state != GFS2_BLKST_FREE) + if (state != GFS2_BLKST_FREE) return -ENOSPC; /* If the extent was too small, and it's smaller than the smallest @@ -1775,7 +1826,7 @@ next_iter: useless to search this rgrp again for this amount or more. */ if (wrapped && (scan_from_start || rbm->bii > last_bii) && *minext < rbm->rgd->rd_extfail_pt) - rbm->rgd->rd_extfail_pt = *minext; + rbm->rgd->rd_extfail_pt = *minext - 1; /* If the maximum extent we found is big enough to fulfill the minimum requirements, use it anyway. */ @@ -1938,7 +1989,7 @@ static bool gfs2_rgrp_used_recently(const struct gfs2_blkreserv *rs, u64 tdiff; tdiff = ktime_to_ns(ktime_sub(ktime_get_real(), - rs->rs_rbm.rgd->rd_gl->gl_dstamp)); + rs->rs_rgd->rd_gl->gl_dstamp)); return tdiff > (msecs * 1000 * 1000); } @@ -1993,8 +2044,7 @@ static inline int fast_to_acquire(struct gfs2_rgrpd *rgd) * We try our best to find an rgrp that has at least ap->target blocks * available. After a couple of passes (loops == 2), the prospects of finding * such an rgrp diminish. At this stage, we return the first rgrp that has - * at least ap->min_target blocks available. Either way, we set ap->allowed to - * the number of blocks available in the chosen rgrp. + * at least ap->min_target blocks available. * * Returns: 0 on success, * -ENOMEM if a suitable rgrp can't be found @@ -2006,56 +2056,64 @@ int gfs2_inplace_reserve(struct gfs2_inode *ip, struct gfs2_alloc_parms *ap) struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); struct gfs2_rgrpd *begin = NULL; struct gfs2_blkreserv *rs = &ip->i_res; - int error = 0, rg_locked, flags = 0; + int error = 0, flags = LM_FLAG_NODE_SCOPE; + bool rg_locked; u64 last_unlinked = NO_BLOCK; + u32 target = ap->target; int loops = 0; - u32 free_blocks, skip = 0; + u32 free_blocks, blocks_available, skip = 0; + + BUG_ON(rs->rs_reserved); if (sdp->sd_args.ar_rgrplvb) flags |= GL_SKIP; - if (gfs2_assert_warn(sdp, ap->target)) + if (gfs2_assert_warn(sdp, target)) return -EINVAL; if (gfs2_rs_active(rs)) { - begin = rs->rs_rbm.rgd; - } else if (rs->rs_rbm.rgd && - rgrp_contains_block(rs->rs_rbm.rgd, ip->i_goal)) { - begin = rs->rs_rbm.rgd; + begin = rs->rs_rgd; + } else if (rs->rs_rgd && + rgrp_contains_block(rs->rs_rgd, ip->i_goal)) { + begin = rs->rs_rgd; } else { check_and_update_goal(ip); - rs->rs_rbm.rgd = begin = gfs2_blk2rgrpd(sdp, ip->i_goal, 1); + rs->rs_rgd = begin = gfs2_blk2rgrpd(sdp, ip->i_goal, 1); } if (S_ISDIR(ip->i_inode.i_mode) && (ap->aflags & GFS2_AF_ORLOV)) skip = gfs2_orlov_skip(ip); - if (rs->rs_rbm.rgd == NULL) + if (rs->rs_rgd == NULL) return -EBADSLT; while (loops < 3) { - rg_locked = 1; + struct gfs2_rgrpd *rgd; - if (!gfs2_glock_is_locked_by_me(rs->rs_rbm.rgd->rd_gl)) { - rg_locked = 0; + rg_locked = gfs2_glock_is_locked_by_me(rs->rs_rgd->rd_gl); + if (rg_locked) { + rgrp_lock_local(rs->rs_rgd); + } else { if (skip && skip--) goto next_rgrp; if (!gfs2_rs_active(rs)) { if (loops == 0 && - !fast_to_acquire(rs->rs_rbm.rgd)) + !fast_to_acquire(rs->rs_rgd)) goto next_rgrp; if ((loops < 2) && gfs2_rgrp_used_recently(rs, 1000) && - gfs2_rgrp_congested(rs->rs_rbm.rgd, loops)) + gfs2_rgrp_congested(rs->rs_rgd, loops)) goto next_rgrp; } - error = gfs2_glock_nq_init(rs->rs_rbm.rgd->rd_gl, + error = gfs2_glock_nq_init(rs->rs_rgd->rd_gl, LM_ST_EXCLUSIVE, flags, &ip->i_rgd_gh); if (unlikely(error)) return error; + rgrp_lock_local(rs->rs_rgd); if (!gfs2_rs_active(rs) && (loops < 2) && - gfs2_rgrp_congested(rs->rs_rbm.rgd, loops)) + gfs2_rgrp_congested(rs->rs_rgd, loops)) goto skip_rgrp; if (sdp->sd_args.ar_rgrplvb) { - error = update_rgrp_lvb(rs->rs_rbm.rgd); + error = update_rgrp_lvb(rs->rs_rgd); if (unlikely(error)) { + rgrp_unlock_local(rs->rs_rgd); gfs2_glock_dq_uninit(&ip->i_rgd_gh); return error; } @@ -2063,36 +2121,46 @@ int gfs2_inplace_reserve(struct gfs2_inode *ip, struct gfs2_alloc_parms *ap) } /* Skip unusable resource groups */ - if ((rs->rs_rbm.rgd->rd_flags & (GFS2_RGF_NOALLOC | + if ((rs->rs_rgd->rd_flags & (GFS2_RGF_NOALLOC | GFS2_RDF_ERROR)) || - (loops == 0 && ap->target > rs->rs_rbm.rgd->rd_extfail_pt)) + (loops == 0 && target > rs->rs_rgd->rd_extfail_pt)) goto skip_rgrp; if (sdp->sd_args.ar_rgrplvb) - gfs2_rgrp_bh_get(rs->rs_rbm.rgd); + gfs2_rgrp_bh_get(rs->rs_rgd); /* Get a reservation if we don't already have one */ if (!gfs2_rs_active(rs)) - rg_mblk_search(rs->rs_rbm.rgd, ip, ap); + rg_mblk_search(rs->rs_rgd, ip, ap); /* Skip rgrps when we can't get a reservation on first pass */ if (!gfs2_rs_active(rs) && (loops < 1)) goto check_rgrp; /* If rgrp has enough free space, use it */ - free_blocks = rgd_free(rs->rs_rbm.rgd, rs); - if (free_blocks >= ap->target || - (loops == 2 && ap->min_target && - free_blocks >= ap->min_target)) { - ap->allowed = free_blocks; - return 0; + rgd = rs->rs_rgd; + spin_lock(&rgd->rd_rsspin); + free_blocks = rgd_free(rgd, rs); + blocks_available = rgd->rd_free_clone - rgd->rd_reserved; + if (free_blocks < target || blocks_available < target) { + spin_unlock(&rgd->rd_rsspin); + goto check_rgrp; } + rs->rs_reserved = ap->target; + if (rs->rs_reserved > blocks_available) + rs->rs_reserved = blocks_available; + rgd->rd_reserved += rs->rs_reserved; + spin_unlock(&rgd->rd_rsspin); + rgrp_unlock_local(rs->rs_rgd); + return 0; check_rgrp: /* Check for unlinked inodes which can be reclaimed */ - if (rs->rs_rbm.rgd->rd_flags & GFS2_RDF_CHECK) - try_rgrp_unlink(rs->rs_rbm.rgd, &last_unlinked, + if (rs->rs_rgd->rd_flags & GFS2_RDF_CHECK) + try_rgrp_unlink(rs->rs_rgd, &last_unlinked, ip->i_no_addr); skip_rgrp: + rgrp_unlock_local(rs->rs_rgd); + /* Drop reservation, if we couldn't use reserved rgrp */ if (gfs2_rs_active(rs)) gfs2_rs_deltree(rs); @@ -2102,7 +2170,7 @@ skip_rgrp: gfs2_glock_dq_uninit(&ip->i_rgd_gh); next_rgrp: /* Find the next rgrp, and continue looking */ - if (gfs2_select_rgrp(&rs->rs_rbm.rgd, begin)) + if (gfs2_select_rgrp(&rs->rs_rgd, begin)) continue; if (skip) continue; @@ -2119,9 +2187,12 @@ next_rgrp: return error; } /* Flushing the log may release space */ - if (loops == 2) + if (loops == 2) { + if (ap->min_target) + target = ap->min_target; gfs2_log_flush(sdp, NULL, GFS2_LOG_HEAD_FLUSH_NORMAL | GFS2_LFC_INPLACE_RESERVE); + } } return -ENOSPC; @@ -2136,6 +2207,17 @@ next_rgrp: void gfs2_inplace_release(struct gfs2_inode *ip) { + struct gfs2_blkreserv *rs = &ip->i_res; + + if (rs->rs_reserved) { + struct gfs2_rgrpd *rgd = rs->rs_rgd; + + spin_lock(&rgd->rd_rsspin); + BUG_ON(rgd->rd_reserved < rs->rs_reserved); + rgd->rd_reserved -= rs->rs_reserved; + spin_unlock(&rgd->rd_rsspin); + rs->rs_reserved = 0; + } if (gfs2_holder_initialized(&ip->i_rgd_gh)) gfs2_glock_dq_uninit(&ip->i_rgd_gh); } @@ -2205,7 +2287,7 @@ static void rgblk_free(struct gfs2_sbd *sdp, struct gfs2_rgrpd *rgd, bi_prev = bi; } gfs2_setbit(&rbm, false, new_state); - gfs2_rbm_incr(&rbm); + gfs2_rbm_add(&rbm, 1); } } @@ -2223,11 +2305,12 @@ void gfs2_rgrp_dump(struct seq_file *seq, struct gfs2_rgrpd *rgd, struct gfs2_blkreserv *trs; const struct rb_node *n; - gfs2_print_dbg(seq, "%s R: n:%llu f:%02x b:%u/%u i:%u r:%u e:%u\n", + spin_lock(&rgd->rd_rsspin); + gfs2_print_dbg(seq, "%s R: n:%llu f:%02x b:%u/%u i:%u q:%u r:%u e:%u\n", fs_id_buf, (unsigned long long)rgd->rd_addr, rgd->rd_flags, rgd->rd_free, rgd->rd_free_clone, rgd->rd_dinodes, - rgd->rd_reserved, rgd->rd_extfail_pt); + rgd->rd_requested, rgd->rd_reserved, rgd->rd_extfail_pt); if (rgd->rd_sbd->sd_args.ar_rgrplvb) { struct gfs2_rgrp_lvb *rgl = rgd->rd_rgl; @@ -2236,7 +2319,6 @@ void gfs2_rgrp_dump(struct seq_file *seq, struct gfs2_rgrpd *rgd, be32_to_cpu(rgl->rl_free), be32_to_cpu(rgl->rl_dinodes)); } - spin_lock(&rgd->rd_rsspin); for (n = rb_first(&rgd->rd_rstree); n; n = rb_next(&trs->rs_node)) { trs = rb_entry(n, struct gfs2_blkreserv, rs_node); dump_rs(seq, trs, fs_id_buf); @@ -2273,29 +2355,29 @@ static void gfs2_adjust_reservation(struct gfs2_inode *ip, { struct gfs2_blkreserv *rs = &ip->i_res; struct gfs2_rgrpd *rgd = rbm->rgd; - unsigned rlen; - u64 block; - int ret; - spin_lock(&rgd->rd_rsspin); + BUG_ON(rs->rs_reserved < len); + rs->rs_reserved -= len; if (gfs2_rs_active(rs)) { - if (gfs2_rbm_eq(&rs->rs_rbm, rbm)) { - block = gfs2_rbm_to_block(rbm); - ret = gfs2_rbm_from_block(&rs->rs_rbm, block + len); - rlen = min(rs->rs_free, len); - rs->rs_free -= rlen; - rgd->rd_reserved -= rlen; + u64 start = gfs2_rbm_to_block(rbm); + + if (rs->rs_start == start) { + unsigned int rlen; + + rs->rs_start += len; + rlen = min(rs->rs_requested, len); + rs->rs_requested -= rlen; + rgd->rd_requested -= rlen; trace_gfs2_rs(rs, TRACE_RS_CLAIM); - if (rs->rs_free && !ret) - goto out; + if (rs->rs_start < rgd->rd_data0 + rgd->rd_data && + rs->rs_requested) + return; /* We used up our block reservation, so we should reserve more blocks next time. */ atomic_add(RGRP_RSRV_ADDBLKS, &ip->i_sizehint); } __rs_deltree(rs); } -out: - spin_unlock(&rgd->rd_rsspin); } /** @@ -2315,15 +2397,13 @@ static void gfs2_set_alloc_start(struct gfs2_rbm *rbm, u64 goal; if (gfs2_rs_active(&ip->i_res)) { - *rbm = ip->i_res.rs_rbm; - return; + goal = ip->i_res.rs_start; + } else { + if (!dinode && rgrp_contains_block(rbm->rgd, ip->i_goal)) + goal = ip->i_goal; + else + goal = rbm->rgd->rd_last_alloc + rbm->rgd->rd_data0; } - - if (!dinode && rgrp_contains_block(rbm->rgd, ip->i_goal)) - goal = ip->i_goal; - else - goal = rbm->rgd->rd_last_alloc + rbm->rgd->rd_data0; - if (WARN_ON_ONCE(gfs2_rbm_from_block(rbm, goal))) { rbm->bii = 0; rbm->offset = 0; @@ -2346,17 +2426,21 @@ int gfs2_alloc_blocks(struct gfs2_inode *ip, u64 *bn, unsigned int *nblocks, { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); struct buffer_head *dibh; - struct gfs2_rbm rbm = { .rgd = ip->i_res.rs_rbm.rgd, }; - unsigned int ndata; + struct gfs2_rbm rbm = { .rgd = ip->i_res.rs_rgd, }; u64 block; /* block, within the file system scope */ - int error; + u32 minext = 1; + int error = -ENOSPC; - gfs2_set_alloc_start(&rbm, ip, dinode); - error = gfs2_rbm_find(&rbm, GFS2_BLKST_FREE, NULL, ip, false); + BUG_ON(ip->i_res.rs_reserved < *nblocks); + rgrp_lock_local(rbm.rgd); + if (gfs2_rs_active(&ip->i_res)) { + gfs2_set_alloc_start(&rbm, ip, dinode); + error = gfs2_rbm_find(&rbm, GFS2_BLKST_FREE, &minext, &ip->i_res, false); + } if (error == -ENOSPC) { gfs2_set_alloc_start(&rbm, ip, dinode); - error = gfs2_rbm_find(&rbm, GFS2_BLKST_FREE, NULL, NULL, false); + error = gfs2_rbm_find(&rbm, GFS2_BLKST_FREE, &minext, NULL, false); } /* Since all blocks are reserved in advance, this shouldn't happen */ @@ -2371,14 +2455,8 @@ int gfs2_alloc_blocks(struct gfs2_inode *ip, u64 *bn, unsigned int *nblocks, gfs2_alloc_extent(&rbm, dinode, nblocks); block = gfs2_rbm_to_block(&rbm); rbm.rgd->rd_last_alloc = block - rbm.rgd->rd_data0; - if (gfs2_rs_active(&ip->i_res)) - gfs2_adjust_reservation(ip, &rbm, *nblocks); - ndata = *nblocks; - if (dinode) - ndata--; - if (!dinode) { - ip->i_goal = block + ndata - 1; + ip->i_goal = block + *nblocks - 1; error = gfs2_meta_inode_buffer(ip, &dibh); if (error == 0) { struct gfs2_dinode *di = @@ -2389,12 +2467,20 @@ int gfs2_alloc_blocks(struct gfs2_inode *ip, u64 *bn, unsigned int *nblocks, brelse(dibh); } } - if (rbm.rgd->rd_free < *nblocks) { + spin_lock(&rbm.rgd->rd_rsspin); + gfs2_adjust_reservation(ip, &rbm, *nblocks); + if (rbm.rgd->rd_free < *nblocks || rbm.rgd->rd_reserved < *nblocks) { fs_warn(sdp, "nblocks=%u\n", *nblocks); + spin_unlock(&rbm.rgd->rd_rsspin); goto rgrp_error; } - + BUG_ON(rbm.rgd->rd_reserved < *nblocks); + BUG_ON(rbm.rgd->rd_free_clone < *nblocks); + BUG_ON(rbm.rgd->rd_free < *nblocks); + rbm.rgd->rd_reserved -= *nblocks; + rbm.rgd->rd_free_clone -= *nblocks; rbm.rgd->rd_free -= *nblocks; + spin_unlock(&rbm.rgd->rd_rsspin); if (dinode) { rbm.rgd->rd_dinodes++; *generation = rbm.rgd->rd_igeneration++; @@ -2404,6 +2490,7 @@ int gfs2_alloc_blocks(struct gfs2_inode *ip, u64 *bn, unsigned int *nblocks, gfs2_trans_add_meta(rbm.rgd->rd_gl, rbm.rgd->rd_bits[0].bi_bh); gfs2_rgrp_out(rbm.rgd, rbm.rgd->rd_bits[0].bi_bh->b_data); + rgrp_unlock_local(rbm.rgd); gfs2_statfs_change(sdp, 0, -(s64)*nblocks, dinode ? 1 : 0); if (dinode) @@ -2411,13 +2498,13 @@ int gfs2_alloc_blocks(struct gfs2_inode *ip, u64 *bn, unsigned int *nblocks, gfs2_quota_change(ip, *nblocks, ip->i_inode.i_uid, ip->i_inode.i_gid); - rbm.rgd->rd_free_clone -= *nblocks; trace_gfs2_block_alloc(ip, rbm.rgd, block, *nblocks, dinode ? GFS2_BLKST_DINODE : GFS2_BLKST_USED); *bn = block; return 0; rgrp_error: + rgrp_unlock_local(rbm.rgd); gfs2_rgrp_error(rbm.rgd); return -EIO; } @@ -2437,12 +2524,14 @@ void __gfs2_free_blocks(struct gfs2_inode *ip, struct gfs2_rgrpd *rgd, { struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); + rgrp_lock_local(rgd); rgblk_free(sdp, rgd, bstart, blen, GFS2_BLKST_FREE); trace_gfs2_block_alloc(ip, rgd, bstart, blen, GFS2_BLKST_FREE); rgd->rd_free += blen; rgd->rd_flags &= ~GFS2_RGF_TRIMMED; gfs2_trans_add_meta(rgd->rd_gl, rgd->rd_bits[0].bi_bh); gfs2_rgrp_out(rgd, rgd->rd_bits[0].bi_bh->b_data); + rgrp_unlock_local(rgd); /* Directories keep their data in the metadata address space */ if (meta || ip->i_depth || gfs2_is_jdata(ip)) @@ -2478,17 +2567,20 @@ void gfs2_unlink_di(struct inode *inode) rgd = gfs2_blk2rgrpd(sdp, blkno, true); if (!rgd) return; + rgrp_lock_local(rgd); rgblk_free(sdp, rgd, blkno, 1, GFS2_BLKST_UNLINKED); trace_gfs2_block_alloc(ip, rgd, blkno, 1, GFS2_BLKST_UNLINKED); gfs2_trans_add_meta(rgd->rd_gl, rgd->rd_bits[0].bi_bh); gfs2_rgrp_out(rgd, rgd->rd_bits[0].bi_bh->b_data); be32_add_cpu(&rgd->rd_rgl->rl_unlinked, 1); + rgrp_unlock_local(rgd); } void gfs2_free_di(struct gfs2_rgrpd *rgd, struct gfs2_inode *ip) { struct gfs2_sbd *sdp = rgd->rd_sbd; + rgrp_lock_local(rgd); rgblk_free(sdp, rgd, ip->i_no_addr, 1, GFS2_BLKST_FREE); if (!rgd->rd_dinodes) gfs2_consist_rgrpd(rgd); @@ -2497,6 +2589,7 @@ void gfs2_free_di(struct gfs2_rgrpd *rgd, struct gfs2_inode *ip) gfs2_trans_add_meta(rgd->rd_gl, rgd->rd_bits[0].bi_bh); gfs2_rgrp_out(rgd, rgd->rd_bits[0].bi_bh->b_data); + rgrp_unlock_local(rgd); be32_add_cpu(&rgd->rd_rgl->rl_unlinked, -1); gfs2_statfs_change(sdp, 0, +1, -1); @@ -2511,6 +2604,10 @@ void gfs2_free_di(struct gfs2_rgrpd *rgd, struct gfs2_inode *ip) * @no_addr: The block number to check * @type: The block type we are looking for * + * The inode glock of @no_addr must be held. The @type to check for is either + * GFS2_BLKST_DINODE or GFS2_BLKST_UNLINKED; checking for type GFS2_BLKST_FREE + * or GFS2_BLKST_USED would make no sense. + * * Returns: 0 if the block type matches the expected type * -ESTALE if it doesn't match * or -ve errno if something went wrong while checking @@ -2534,6 +2631,13 @@ int gfs2_check_blk_type(struct gfs2_sbd *sdp, u64 no_addr, unsigned int type) rbm.rgd = rgd; error = gfs2_rbm_from_block(&rbm, no_addr); if (!WARN_ON_ONCE(error)) { + /* + * No need to take the local resource group lock here; the + * inode glock of @no_addr provides the necessary + * synchronization in case the block is an inode. (In case + * the block is not an inode, the block type will not match + * the @type we are looking for.) + */ if (gfs2_testbit(&rbm, false) != type) error = -ESTALE; } @@ -2578,7 +2682,7 @@ void gfs2_rlist_add(struct gfs2_inode *ip, struct gfs2_rgrp_list *rlist, return; rgd = gfs2_blk2rgrpd(sdp, block, 1); } else { - rgd = ip->i_res.rs_rbm.rgd; + rgd = ip->i_res.rs_rgd; if (!rgd || !rgrp_contains_block(rgd, block)) rgd = gfs2_blk2rgrpd(sdp, block, 1); } @@ -2633,9 +2737,8 @@ void gfs2_rlist_alloc(struct gfs2_rgrp_list *rlist) sizeof(struct gfs2_holder), GFP_NOFS | __GFP_NOFAIL); for (x = 0; x < rlist->rl_rgrps; x++) - gfs2_holder_init(rlist->rl_rgd[x]->rd_gl, - LM_ST_EXCLUSIVE, 0, - &rlist->rl_ghs[x]); + gfs2_holder_init(rlist->rl_rgd[x]->rd_gl, LM_ST_EXCLUSIVE, + LM_FLAG_NODE_SCOPE, &rlist->rl_ghs[x]); } /** @@ -2658,3 +2761,14 @@ void gfs2_rlist_free(struct gfs2_rgrp_list *rlist) } } +void rgrp_lock_local(struct gfs2_rgrpd *rgd) +{ + BUG_ON(!gfs2_glock_is_held_excl(rgd->rd_gl) && + !test_bit(SDF_NORECOVERY, &rgd->rd_sbd->sd_flags)); + mutex_lock(&rgd->rd_mutex); +} + +void rgrp_unlock_local(struct gfs2_rgrpd *rgd) +{ + mutex_unlock(&rgd->rd_mutex); +} diff --git a/fs/gfs2/rgrp.h b/fs/gfs2/rgrp.h index 9a587ada51ed..a6855fd796e0 100644 --- a/fs/gfs2/rgrp.h +++ b/fs/gfs2/rgrp.h @@ -77,7 +77,7 @@ extern int gfs2_fitrim(struct file *filp, void __user *argp); /* This is how to tell if a reservation is in the rgrp tree: */ static inline bool gfs2_rs_active(const struct gfs2_blkreserv *rs) { - return rs && !RB_EMPTY_NODE(&rs->rs_node); + return !RB_EMPTY_NODE(&rs->rs_node); } static inline int rgrp_contains_block(struct gfs2_rgrpd *rgd, u64 block) @@ -88,4 +88,8 @@ static inline int rgrp_contains_block(struct gfs2_rgrpd *rgd, u64 block) } extern void check_and_update_goal(struct gfs2_inode *ip); + +extern void rgrp_lock_local(struct gfs2_rgrpd *rgd); +extern void rgrp_unlock_local(struct gfs2_rgrpd *rgd); + #endif /* __RGRP_DOT_H__ */ diff --git a/fs/gfs2/super.c b/fs/gfs2/super.c index 2f56acc41c04..97076d3f562f 100644 --- a/fs/gfs2/super.c +++ b/fs/gfs2/super.c @@ -81,19 +81,12 @@ void gfs2_jindex_free(struct gfs2_sbd *sdp) static struct gfs2_jdesc *jdesc_find_i(struct list_head *head, unsigned int jid) { struct gfs2_jdesc *jd; - int found = 0; list_for_each_entry(jd, head, jd_list) { - if (jd->jd_jid == jid) { - found = 1; - break; - } + if (jd->jd_jid == jid) + return jd; } - - if (!found) - jd = NULL; - - return jd; + return NULL; } struct gfs2_jdesc *gfs2_jdesc_find(struct gfs2_sbd *sdp, unsigned int jid) @@ -165,7 +158,6 @@ int gfs2_make_fs_rw(struct gfs2_sbd *sdp) { struct gfs2_inode *ip = GFS2_I(sdp->sd_jdesc->jd_inode); struct gfs2_glock *j_gl = ip->i_gl; - struct gfs2_holder freeze_gh; struct gfs2_log_header_host head; int error; @@ -173,12 +165,6 @@ int gfs2_make_fs_rw(struct gfs2_sbd *sdp) if (error) return error; - error = gfs2_glock_nq_init(sdp->sd_freeze_gl, LM_ST_SHARED, - LM_FLAG_NOEXP | GL_EXACT, - &freeze_gh); - if (error) - goto fail_threads; - j_gl->gl_ops->go_inval(j_gl, DIO_METADATA); if (gfs2_withdrawn(sdp)) { error = -EIO; @@ -205,13 +191,9 @@ int gfs2_make_fs_rw(struct gfs2_sbd *sdp) set_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags); - gfs2_glock_dq_uninit(&freeze_gh); - return 0; fail: - gfs2_glock_dq_uninit(&freeze_gh); -fail_threads: if (sdp->sd_quotad_process) kthread_stop(sdp->sd_quotad_process); sdp->sd_quotad_process = NULL; @@ -452,7 +434,7 @@ static int gfs2_lock_fs_check_clean(struct gfs2_sbd *sdp) } if (error) - gfs2_glock_dq_uninit(&sdp->sd_freeze_gh); + gfs2_freeze_unlock(&sdp->sd_freeze_gh); out: while (!list_empty(&list)) { @@ -562,8 +544,6 @@ static void gfs2_dirty_inode(struct inode *inode, int flags) int need_endtrans = 0; int ret; - if (!(flags & I_DIRTY_INODE)) - return; if (unlikely(gfs2_withdrawn(sdp))) return; if (!gfs2_glock_is_locked_by_me(ip->i_gl)) { @@ -607,32 +587,10 @@ out: * Returns: errno */ -int gfs2_make_fs_ro(struct gfs2_sbd *sdp) +void gfs2_make_fs_ro(struct gfs2_sbd *sdp) { - struct gfs2_holder freeze_gh; - int error = 0; int log_write_allowed = test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags); - gfs2_holder_mark_uninitialized(&freeze_gh); - if (sdp->sd_freeze_gl && - !gfs2_glock_is_locked_by_me(sdp->sd_freeze_gl)) { - if (!log_write_allowed) { - error = gfs2_glock_nq_init(sdp->sd_freeze_gl, - LM_ST_SHARED, LM_FLAG_TRY | - LM_FLAG_NOEXP | GL_EXACT, - &freeze_gh); - if (error == GLR_TRYFAILED) - error = 0; - } else { - error = gfs2_glock_nq_init(sdp->sd_freeze_gl, - LM_ST_SHARED, - LM_FLAG_NOEXP | GL_EXACT, - &freeze_gh); - if (error && !gfs2_withdrawn(sdp)) - return error; - } - } - gfs2_flush_delete_work(sdp); if (!log_write_allowed && current == sdp->sd_quotad_process) fs_warn(sdp, "The quotad daemon is withdrawing.\n"); @@ -652,24 +610,19 @@ int gfs2_make_fs_ro(struct gfs2_sbd *sdp) gfs2_log_flush(sdp, NULL, GFS2_LOG_HEAD_FLUSH_SHUTDOWN | GFS2_LFC_MAKE_FS_RO); - wait_event(sdp->sd_reserving_log_wait, - atomic_read(&sdp->sd_reserving_log) == 0); - gfs2_assert_warn(sdp, atomic_read(&sdp->sd_log_blks_free) == - sdp->sd_jdesc->jd_blocks); + wait_event_timeout(sdp->sd_log_waitq, + gfs2_log_is_empty(sdp), + HZ * 5); + gfs2_assert_warn(sdp, gfs2_log_is_empty(sdp)); } else { - wait_event_timeout(sdp->sd_reserving_log_wait, - atomic_read(&sdp->sd_reserving_log) == 0, + wait_event_timeout(sdp->sd_log_waitq, + gfs2_log_is_empty(sdp), HZ * 5); } - if (gfs2_holder_initialized(&freeze_gh)) - gfs2_glock_dq_uninit(&freeze_gh); - gfs2_quota_cleanup(sdp); if (!log_write_allowed) sdp->sd_vfs->s_flags |= SB_RDONLY; - - return error; } /** @@ -681,7 +634,6 @@ int gfs2_make_fs_ro(struct gfs2_sbd *sdp) static void gfs2_put_super(struct super_block *sb) { struct gfs2_sbd *sdp = sb->s_fs_info; - int error; struct gfs2_jdesc *jd; /* No more recovery requests */ @@ -702,9 +654,7 @@ restart: spin_unlock(&sdp->sd_jindex_spin); if (!sb_rdonly(sb)) { - error = gfs2_make_fs_ro(sdp); - if (error) - gfs2_io_error(sdp); + gfs2_make_fs_ro(sdp); } WARN_ON(gfs2_withdrawing(sdp)); @@ -772,10 +722,8 @@ void gfs2_freeze_func(struct work_struct *work) struct super_block *sb = sdp->sd_vfs; atomic_inc(&sb->s_active); - error = gfs2_glock_nq_init(sdp->sd_freeze_gl, LM_ST_SHARED, - LM_FLAG_NOEXP | GL_EXACT, &freeze_gh); + error = gfs2_freeze_lock(sdp, &freeze_gh, 0); if (error) { - fs_info(sdp, "GFS2: couldn't get freeze lock : %d\n", error); gfs2_assert_withdraw(sdp, 0); } else { atomic_set(&sdp->sd_freeze_state, SFS_UNFROZEN); @@ -785,7 +733,7 @@ void gfs2_freeze_func(struct work_struct *work) error); gfs2_assert_withdraw(sdp, 0); } - gfs2_glock_dq_uninit(&freeze_gh); + gfs2_freeze_unlock(&freeze_gh); } deactivate_super(sb); clear_bit_unlock(SDF_FS_FROZEN, &sdp->sd_flags); @@ -853,7 +801,7 @@ static int gfs2_unfreeze(struct super_block *sb) return 0; } - gfs2_glock_dq_uninit(&sdp->sd_freeze_gh); + gfs2_freeze_unlock(&sdp->sd_freeze_gh); mutex_unlock(&sdp->sd_freeze_mutex); return wait_on_bit(&sdp->sd_flags, SDF_FS_FROZEN, TASK_INTERRUPTIBLE); } @@ -1229,7 +1177,8 @@ static int gfs2_dinode_dealloc(struct gfs2_inode *ip) goto out_qs; } - error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0, &gh); + error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, + LM_FLAG_NODE_SCOPE, &gh); if (error) goto out_qs; diff --git a/fs/gfs2/super.h b/fs/gfs2/super.h index c9fb2a654181..ec4affb33ed5 100644 --- a/fs/gfs2/super.h +++ b/fs/gfs2/super.h @@ -11,6 +11,10 @@ #include <linux/dcache.h> #include "incore.h" +/* Supported fs format version range */ +#define GFS2_FS_FORMAT_MIN (1801) +#define GFS2_FS_FORMAT_MAX (1802) + extern void gfs2_lm_unmount(struct gfs2_sbd *sdp); static inline unsigned int gfs2_jindex_size(struct gfs2_sbd *sdp) @@ -30,7 +34,7 @@ extern int gfs2_lookup_in_master_dir(struct gfs2_sbd *sdp, char *filename, struct gfs2_inode **ipp); extern int gfs2_make_fs_rw(struct gfs2_sbd *sdp); -extern int gfs2_make_fs_ro(struct gfs2_sbd *sdp); +extern void gfs2_make_fs_ro(struct gfs2_sbd *sdp); extern void gfs2_online_uevent(struct gfs2_sbd *sdp); extern int gfs2_statfs_init(struct gfs2_sbd *sdp); extern void gfs2_statfs_change(struct gfs2_sbd *sdp, s64 total, s64 free, @@ -54,7 +58,9 @@ extern struct file_system_type gfs2meta_fs_type; extern const struct export_operations gfs2_export_ops; extern const struct super_operations gfs2_super_ops; extern const struct dentry_operations gfs2_dops; -extern const struct xattr_handler *gfs2_xattr_handlers[]; + +extern const struct xattr_handler *gfs2_xattr_handlers_max[]; +extern const struct xattr_handler **gfs2_xattr_handlers_min; #endif /* __SUPER_DOT_H__ */ diff --git a/fs/gfs2/trace_gfs2.h b/fs/gfs2/trace_gfs2.h index 0b2f858d9a8c..bd6c8e9e49db 100644 --- a/fs/gfs2/trace_gfs2.h +++ b/fs/gfs2/trace_gfs2.h @@ -560,6 +560,7 @@ TRACE_EVENT(gfs2_block_alloc, __field( u8, block_state ) __field( u64, rd_addr ) __field( u32, rd_free_clone ) + __field( u32, rd_requested ) __field( u32, rd_reserved ) ), @@ -571,17 +572,20 @@ TRACE_EVENT(gfs2_block_alloc, __entry->block_state = block_state; __entry->rd_addr = rgd->rd_addr; __entry->rd_free_clone = rgd->rd_free_clone; + __entry->rd_requested = rgd->rd_requested; __entry->rd_reserved = rgd->rd_reserved; ), - TP_printk("%u,%u bmap %llu alloc %llu/%lu %s rg:%llu rf:%u rr:%lu", + TP_printk("%u,%u bmap %llu alloc %llu/%lu %s rg:%llu rf:%u rq:%u rr:%u", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->inum, (unsigned long long)__entry->start, (unsigned long)__entry->len, block_state_name(__entry->block_state), (unsigned long long)__entry->rd_addr, - __entry->rd_free_clone, (unsigned long)__entry->rd_reserved) + __entry->rd_free_clone, + __entry->rd_requested, + __entry->rd_reserved) ); /* Keep track of multi-block reservations as they are allocated/freed */ @@ -595,33 +599,40 @@ TRACE_EVENT(gfs2_rs, __field( dev_t, dev ) __field( u64, rd_addr ) __field( u32, rd_free_clone ) + __field( u32, rd_requested ) __field( u32, rd_reserved ) __field( u64, inum ) __field( u64, start ) - __field( u32, free ) + __field( u32, requested ) + __field( u32, reserved ) __field( u8, func ) ), TP_fast_assign( - __entry->dev = rs->rs_rbm.rgd->rd_sbd->sd_vfs->s_dev; - __entry->rd_addr = rs->rs_rbm.rgd->rd_addr; - __entry->rd_free_clone = rs->rs_rbm.rgd->rd_free_clone; - __entry->rd_reserved = rs->rs_rbm.rgd->rd_reserved; + __entry->dev = rs->rs_rgd->rd_sbd->sd_vfs->s_dev; + __entry->rd_addr = rs->rs_rgd->rd_addr; + __entry->rd_free_clone = rs->rs_rgd->rd_free_clone; + __entry->rd_requested = rs->rs_rgd->rd_requested; + __entry->rd_reserved = rs->rs_rgd->rd_reserved; __entry->inum = container_of(rs, struct gfs2_inode, i_res)->i_no_addr; - __entry->start = gfs2_rbm_to_block(&rs->rs_rbm); - __entry->free = rs->rs_free; + __entry->start = rs->rs_start; + __entry->requested = rs->rs_requested; + __entry->reserved = rs->rs_reserved; __entry->func = func; ), - TP_printk("%u,%u bmap %llu resrv %llu rg:%llu rf:%lu rr:%lu %s f:%lu", + TP_printk("%u,%u bmap %llu resrv %llu rg:%llu rf:%u rq:%u rr:%u %s q:%u r:%u", MAJOR(__entry->dev), MINOR(__entry->dev), (unsigned long long)__entry->inum, (unsigned long long)__entry->start, (unsigned long long)__entry->rd_addr, - (unsigned long)__entry->rd_free_clone, - (unsigned long)__entry->rd_reserved, - rs_func_name(__entry->func), (unsigned long)__entry->free) + __entry->rd_free_clone, + __entry->rd_requested, + __entry->rd_reserved, + rs_func_name(__entry->func), + __entry->requested, + __entry->reserved) ); #endif /* _TRACE_GFS2_H */ diff --git a/fs/gfs2/trans.c b/fs/gfs2/trans.c index 6d4bf7ea7b3b..63fec11ef2ce 100644 --- a/fs/gfs2/trans.c +++ b/fs/gfs2/trans.c @@ -31,17 +31,17 @@ static void gfs2_print_trans(struct gfs2_sbd *sdp, const struct gfs2_trans *tr) fs_warn(sdp, "blocks=%u revokes=%u reserved=%u touched=%u\n", tr->tr_blocks, tr->tr_revokes, tr->tr_reserved, test_bit(TR_TOUCHED, &tr->tr_flags)); - fs_warn(sdp, "Buf %u/%u Databuf %u/%u Revoke %u/%u\n", + fs_warn(sdp, "Buf %u/%u Databuf %u/%u Revoke %u\n", tr->tr_num_buf_new, tr->tr_num_buf_rm, tr->tr_num_databuf_new, tr->tr_num_databuf_rm, - tr->tr_num_revoke, tr->tr_num_revoke_rm); + tr->tr_num_revoke); } -int gfs2_trans_begin(struct gfs2_sbd *sdp, unsigned int blocks, - unsigned int revokes) +int __gfs2_trans_begin(struct gfs2_trans *tr, struct gfs2_sbd *sdp, + unsigned int blocks, unsigned int revokes, + unsigned long ip) { - struct gfs2_trans *tr; - int error; + unsigned int extra_revokes; if (current->journal_info) { gfs2_print_trans(sdp, current->journal_info); @@ -52,39 +52,72 @@ int gfs2_trans_begin(struct gfs2_sbd *sdp, unsigned int blocks, if (!test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags)) return -EROFS; - tr = kmem_cache_zalloc(gfs2_trans_cachep, GFP_NOFS); - if (!tr) - return -ENOMEM; - - tr->tr_ip = _RET_IP_; + tr->tr_ip = ip; tr->tr_blocks = blocks; tr->tr_revokes = revokes; - tr->tr_reserved = 1; - set_bit(TR_ALLOCED, &tr->tr_flags); - if (blocks) - tr->tr_reserved += 6 + blocks; - if (revokes) - tr->tr_reserved += gfs2_struct2blk(sdp, revokes); + tr->tr_reserved = GFS2_LOG_FLUSH_MIN_BLOCKS; + if (blocks) { + /* + * The reserved blocks are either used for data or metadata. + * We can have mixed data and metadata, each with its own log + * descriptor block; see calc_reserved(). + */ + tr->tr_reserved += blocks + 1 + DIV_ROUND_UP(blocks - 1, databuf_limit(sdp)); + } INIT_LIST_HEAD(&tr->tr_databuf); INIT_LIST_HEAD(&tr->tr_buf); INIT_LIST_HEAD(&tr->tr_list); INIT_LIST_HEAD(&tr->tr_ail1_list); INIT_LIST_HEAD(&tr->tr_ail2_list); + if (gfs2_assert_warn(sdp, tr->tr_reserved <= sdp->sd_jdesc->jd_blocks)) + return -EINVAL; + sb_start_intwrite(sdp->sd_vfs); - error = gfs2_log_reserve(sdp, tr->tr_reserved); - if (error) - goto fail; + /* + * Try the reservations under sd_log_flush_lock to prevent log flushes + * from creating inconsistencies between the number of allocated and + * reserved revokes. If that fails, do a full-block allocation outside + * of the lock to avoid stalling log flushes. Then, allot the + * appropriate number of blocks to revokes, use as many revokes locally + * as needed, and "release" the surplus into the revokes pool. + */ + + down_read(&sdp->sd_log_flush_lock); + if (gfs2_log_try_reserve(sdp, tr, &extra_revokes)) + goto reserved; + up_read(&sdp->sd_log_flush_lock); + gfs2_log_reserve(sdp, tr, &extra_revokes); + down_read(&sdp->sd_log_flush_lock); + +reserved: + gfs2_log_release_revokes(sdp, extra_revokes); + if (unlikely(!test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags))) { + gfs2_log_release_revokes(sdp, tr->tr_revokes); + up_read(&sdp->sd_log_flush_lock); + gfs2_log_release(sdp, tr->tr_reserved); + sb_end_intwrite(sdp->sd_vfs); + return -EROFS; + } current->journal_info = tr; return 0; +} -fail: - sb_end_intwrite(sdp->sd_vfs); - kmem_cache_free(gfs2_trans_cachep, tr); +int gfs2_trans_begin(struct gfs2_sbd *sdp, unsigned int blocks, + unsigned int revokes) +{ + struct gfs2_trans *tr; + int error; + tr = kmem_cache_zalloc(gfs2_trans_cachep, GFP_NOFS); + if (!tr) + return -ENOMEM; + error = __gfs2_trans_begin(tr, sdp, blocks, revokes, _RET_IP_); + if (error) + kmem_cache_free(gfs2_trans_cachep, tr); return error; } @@ -92,37 +125,39 @@ void gfs2_trans_end(struct gfs2_sbd *sdp) { struct gfs2_trans *tr = current->journal_info; s64 nbuf; - int alloced = test_bit(TR_ALLOCED, &tr->tr_flags); current->journal_info = NULL; if (!test_bit(TR_TOUCHED, &tr->tr_flags)) { + gfs2_log_release_revokes(sdp, tr->tr_revokes); + up_read(&sdp->sd_log_flush_lock); gfs2_log_release(sdp, tr->tr_reserved); - if (alloced) { + if (!test_bit(TR_ONSTACK, &tr->tr_flags)) gfs2_trans_free(sdp, tr); - sb_end_intwrite(sdp->sd_vfs); - } + sb_end_intwrite(sdp->sd_vfs); return; } + gfs2_log_release_revokes(sdp, tr->tr_revokes - tr->tr_num_revoke); + nbuf = tr->tr_num_buf_new + tr->tr_num_databuf_new; nbuf -= tr->tr_num_buf_rm; nbuf -= tr->tr_num_databuf_rm; - if (gfs2_assert_withdraw(sdp, (nbuf <= tr->tr_blocks) && - (tr->tr_num_revoke <= tr->tr_revokes))) + if (gfs2_assert_withdraw(sdp, nbuf <= tr->tr_blocks) || + gfs2_assert_withdraw(sdp, tr->tr_num_revoke <= tr->tr_revokes)) gfs2_print_trans(sdp, tr); gfs2_log_commit(sdp, tr); - if (alloced && !test_bit(TR_ATTACHED, &tr->tr_flags)) + if (!test_bit(TR_ONSTACK, &tr->tr_flags) && + !test_bit(TR_ATTACHED, &tr->tr_flags)) gfs2_trans_free(sdp, tr); up_read(&sdp->sd_log_flush_lock); if (sdp->sd_vfs->s_flags & SB_SYNCHRONOUS) gfs2_log_flush(sdp, NULL, GFS2_LOG_HEAD_FLUSH_NORMAL | GFS2_LFC_TRANS_END); - if (alloced) - sb_end_intwrite(sdp->sd_vfs); + sb_end_intwrite(sdp->sd_vfs); } static struct gfs2_bufdata *gfs2_alloc_bufdata(struct gfs2_glock *gl, @@ -134,6 +169,8 @@ static struct gfs2_bufdata *gfs2_alloc_bufdata(struct gfs2_glock *gl, bd->bd_bh = bh; bd->bd_gl = gl; INIT_LIST_HEAD(&bd->bd_list); + INIT_LIST_HEAD(&bd->bd_ail_st_list); + INIT_LIST_HEAD(&bd->bd_ail_gl_list); bh->b_private = bd; return bd; } @@ -262,7 +299,6 @@ void gfs2_trans_add_revoke(struct gfs2_sbd *sdp, struct gfs2_bufdata *bd) void gfs2_trans_remove_revoke(struct gfs2_sbd *sdp, u64 blkno, unsigned int len) { struct gfs2_bufdata *bd, *tmp; - struct gfs2_trans *tr = current->journal_info; unsigned int n = len; gfs2_log_lock(sdp); @@ -274,7 +310,7 @@ void gfs2_trans_remove_revoke(struct gfs2_sbd *sdp, u64 blkno, unsigned int len) if (bd->bd_gl) gfs2_glock_remove_revoke(bd->bd_gl); kmem_cache_free(gfs2_bufdata_cachep, bd); - tr->tr_num_revoke_rm++; + gfs2_log_release_revokes(sdp, 1); if (--n == 0) break; } diff --git a/fs/gfs2/trans.h b/fs/gfs2/trans.h index 83199ce5a5c5..c76ad9a4c75a 100644 --- a/fs/gfs2/trans.h +++ b/fs/gfs2/trans.h @@ -27,13 +27,16 @@ struct gfs2_glock; * block, or all of the blocks in the rg, whichever is smaller */ static inline unsigned int gfs2_rg_blocks(const struct gfs2_inode *ip, unsigned requested) { - struct gfs2_rgrpd *rgd = ip->i_res.rs_rbm.rgd; + struct gfs2_rgrpd *rgd = ip->i_res.rs_rgd; if (requested < rgd->rd_length) return requested + 1; return rgd->rd_length; } +extern int __gfs2_trans_begin(struct gfs2_trans *tr, struct gfs2_sbd *sdp, + unsigned int blocks, unsigned int revokes, + unsigned long ip); extern int gfs2_trans_begin(struct gfs2_sbd *sdp, unsigned int blocks, unsigned int revokes); diff --git a/fs/gfs2/util.c b/fs/gfs2/util.c index a374397f4273..4f034b87b427 100644 --- a/fs/gfs2/util.c +++ b/fs/gfs2/util.c @@ -91,18 +91,50 @@ out_unlock: return error; } +/** + * gfs2_freeze_lock - hold the freeze glock + * @sdp: the superblock + * @freeze_gh: pointer to the requested holder + * @caller_flags: any additional flags needed by the caller + */ +int gfs2_freeze_lock(struct gfs2_sbd *sdp, struct gfs2_holder *freeze_gh, + int caller_flags) +{ + int flags = LM_FLAG_NOEXP | GL_EXACT | caller_flags; + int error; + + error = gfs2_glock_nq_init(sdp->sd_freeze_gl, LM_ST_SHARED, flags, + freeze_gh); + if (error && error != GLR_TRYFAILED) + fs_err(sdp, "can't lock the freeze lock: %d\n", error); + return error; +} + +void gfs2_freeze_unlock(struct gfs2_holder *freeze_gh) +{ + if (gfs2_holder_initialized(freeze_gh)) + gfs2_glock_dq_uninit(freeze_gh); +} + static void signal_our_withdraw(struct gfs2_sbd *sdp) { - struct gfs2_glock *gl = sdp->sd_live_gh.gh_gl; - struct inode *inode = sdp->sd_jdesc->jd_inode; - struct gfs2_inode *ip = GFS2_I(inode); - u64 no_formal_ino = ip->i_no_formal_ino; + struct gfs2_glock *live_gl = sdp->sd_live_gh.gh_gl; + struct inode *inode; + struct gfs2_inode *ip; + struct gfs2_glock *i_gl; + u64 no_formal_ino; + int log_write_allowed = test_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags); int ret = 0; int tries; - if (test_bit(SDF_NORECOVERY, &sdp->sd_flags)) + if (test_bit(SDF_NORECOVERY, &sdp->sd_flags) || !sdp->sd_jdesc) return; + inode = sdp->sd_jdesc->jd_inode; + ip = GFS2_I(inode); + i_gl = ip->i_gl; + no_formal_ino = ip->i_no_formal_ino; + /* Prevent any glock dq until withdraw recovery is complete */ set_bit(SDF_WITHDRAW_RECOVERY, &sdp->sd_flags); /* @@ -117,8 +149,21 @@ static void signal_our_withdraw(struct gfs2_sbd *sdp) * therefore we need to clear SDF_JOURNAL_LIVE manually. */ clear_bit(SDF_JOURNAL_LIVE, &sdp->sd_flags); - if (!sb_rdonly(sdp->sd_vfs)) - ret = gfs2_make_fs_ro(sdp); + if (!sb_rdonly(sdp->sd_vfs)) { + struct gfs2_holder freeze_gh; + + gfs2_holder_mark_uninitialized(&freeze_gh); + if (sdp->sd_freeze_gl && + !gfs2_glock_is_locked_by_me(sdp->sd_freeze_gl)) { + ret = gfs2_freeze_lock(sdp, &freeze_gh, + log_write_allowed ? 0 : LM_FLAG_TRY); + if (ret == GLR_TRYFAILED) + ret = 0; + } + if (!ret) + gfs2_make_fs_ro(sdp); + gfs2_freeze_unlock(&freeze_gh); + } if (sdp->sd_lockstruct.ls_ops->lm_lock == NULL) { /* lock_nolock */ if (!ret) @@ -141,7 +186,8 @@ static void signal_our_withdraw(struct gfs2_sbd *sdp) atomic_set(&sdp->sd_freeze_state, SFS_FROZEN); thaw_super(sdp->sd_vfs); } else { - wait_on_bit(&gl->gl_flags, GLF_DEMOTE, TASK_UNINTERRUPTIBLE); + wait_on_bit(&i_gl->gl_flags, GLF_DEMOTE, + TASK_UNINTERRUPTIBLE); } /* @@ -161,15 +207,15 @@ static void signal_our_withdraw(struct gfs2_sbd *sdp) * on other nodes to be successful, otherwise we remain the owner of * the glock as far as dlm is concerned. */ - if (gl->gl_ops->go_free) { - set_bit(GLF_FREEING, &gl->gl_flags); - wait_on_bit(&gl->gl_flags, GLF_FREEING, TASK_UNINTERRUPTIBLE); + if (i_gl->gl_ops->go_free) { + set_bit(GLF_FREEING, &i_gl->gl_flags); + wait_on_bit(&i_gl->gl_flags, GLF_FREEING, TASK_UNINTERRUPTIBLE); } /* * Dequeue the "live" glock, but keep a reference so it's never freed. */ - gfs2_glock_hold(gl); + gfs2_glock_hold(live_gl); gfs2_glock_dq_wait(&sdp->sd_live_gh); /* * We enqueue the "live" glock in EX so that all other nodes @@ -208,7 +254,7 @@ static void signal_our_withdraw(struct gfs2_sbd *sdp) gfs2_glock_nq(&sdp->sd_live_gh); } - gfs2_glock_queue_put(gl); /* drop the extra reference we acquired */ + gfs2_glock_queue_put(live_gl); /* drop extra reference we acquired */ clear_bit(SDF_WITHDRAW_RECOVERY, &sdp->sd_flags); /* diff --git a/fs/gfs2/util.h b/fs/gfs2/util.h index a4443dd8a94b..69e1a0ae5a4d 100644 --- a/fs/gfs2/util.h +++ b/fs/gfs2/util.h @@ -149,6 +149,9 @@ int gfs2_io_error_i(struct gfs2_sbd *sdp, const char *function, extern int check_journal_clean(struct gfs2_sbd *sdp, struct gfs2_jdesc *jd, bool verbose); +extern int gfs2_freeze_lock(struct gfs2_sbd *sdp, + struct gfs2_holder *freeze_gh, int caller_flags); +extern void gfs2_freeze_unlock(struct gfs2_holder *freeze_gh); #define gfs2_io_error(sdp) \ gfs2_io_error_i((sdp), __func__, __FILE__, __LINE__) diff --git a/fs/gfs2/xattr.c b/fs/gfs2/xattr.c index 9d7667bc4292..124b3d5a7266 100644 --- a/fs/gfs2/xattr.c +++ b/fs/gfs2/xattr.c @@ -70,6 +70,20 @@ static int ea_check_size(struct gfs2_sbd *sdp, unsigned int nsize, size_t dsize) return 0; } +static bool gfs2_eatype_valid(struct gfs2_sbd *sdp, u8 type) +{ + switch(sdp->sd_sb.sb_fs_format) { + case GFS2_FS_FORMAT_MAX: + return true; + + case GFS2_FS_FORMAT_MIN: + return type <= GFS2_EATYPE_SECURITY; + + default: + return false; + } +} + typedef int (*ea_call_t) (struct gfs2_inode *ip, struct buffer_head *bh, struct gfs2_ea_header *ea, struct gfs2_ea_header *prev, void *private); @@ -77,6 +91,7 @@ typedef int (*ea_call_t) (struct gfs2_inode *ip, struct buffer_head *bh, static int ea_foreach_i(struct gfs2_inode *ip, struct buffer_head *bh, ea_call_t ea_call, void *data) { + struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); struct gfs2_ea_header *ea, *prev = NULL; int error = 0; @@ -89,9 +104,8 @@ static int ea_foreach_i(struct gfs2_inode *ip, struct buffer_head *bh, if (!(bh->b_data <= (char *)ea && (char *)GFS2_EA2NEXT(ea) <= bh->b_data + bh->b_size)) goto fail; - if (!GFS2_EATYPE_VALID(ea->ea_type)) + if (!gfs2_eatype_valid(sdp, ea->ea_type)) goto fail; - error = ea_call(ip, bh, ea, prev, data); if (error) return error; @@ -259,7 +273,8 @@ static int ea_dealloc_unstuffed(struct gfs2_inode *ip, struct buffer_head *bh, return -EIO; } - error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0, &rg_gh); + error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, + LM_FLAG_NODE_SCOPE, &rg_gh); if (error) return error; @@ -344,6 +359,7 @@ static int ea_list_i(struct gfs2_inode *ip, struct buffer_head *bh, struct gfs2_ea_header *ea, struct gfs2_ea_header *prev, void *private) { + struct gfs2_sbd *sdp = GFS2_SB(&ip->i_inode); struct ea_list *ei = private; struct gfs2_ea_request *er = ei->ei_er; unsigned int ea_size; @@ -353,6 +369,8 @@ static int ea_list_i(struct gfs2_inode *ip, struct buffer_head *bh, if (ea->ea_type == GFS2_EATYPE_UNUSED) return 0; + BUG_ON(ea->ea_type > GFS2_EATYPE_SECURITY && + sdp->sd_sb.sb_fs_format == GFS2_FS_FORMAT_MIN); switch (ea->ea_type) { case GFS2_EATYPE_USR: prefix = "user."; @@ -366,8 +384,12 @@ static int ea_list_i(struct gfs2_inode *ip, struct buffer_head *bh, prefix = "security."; l = 9; break; + case GFS2_EATYPE_TRUSTED: + prefix = "trusted."; + l = 8; + break; default: - BUG(); + return 0; } ea_size = l + ea->ea_name_len + 1; @@ -1214,6 +1236,7 @@ int __gfs2_xattr_set(struct inode *inode, const char *name, } static int gfs2_xattr_set(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *unused, struct inode *inode, const char *name, const void *value, size_t size, int flags) @@ -1385,7 +1408,8 @@ static int ea_dealloc_block(struct gfs2_inode *ip) return -EIO; } - error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, 0, &gh); + error = gfs2_glock_nq_init(rgd->rd_gl, LM_ST_EXCLUSIVE, + LM_FLAG_NODE_SCOPE, &gh); if (error) return error; @@ -1463,7 +1487,25 @@ static const struct xattr_handler gfs2_xattr_security_handler = { .set = gfs2_xattr_set, }; -const struct xattr_handler *gfs2_xattr_handlers[] = { +static bool +gfs2_xattr_trusted_list(struct dentry *dentry) +{ + return capable(CAP_SYS_ADMIN); +} + +static const struct xattr_handler gfs2_xattr_trusted_handler = { + .prefix = XATTR_TRUSTED_PREFIX, + .flags = GFS2_EATYPE_TRUSTED, + .list = gfs2_xattr_trusted_list, + .get = gfs2_xattr_get, + .set = gfs2_xattr_set, +}; + +const struct xattr_handler *gfs2_xattr_handlers_max[] = { + /* GFS2_FS_FORMAT_MAX */ + &gfs2_xattr_trusted_handler, + + /* GFS2_FS_FORMAT_MIN */ &gfs2_xattr_user_handler, &gfs2_xattr_security_handler, &posix_acl_access_xattr_handler, @@ -1471,3 +1513,4 @@ const struct xattr_handler *gfs2_xattr_handlers[] = { NULL, }; +const struct xattr_handler **gfs2_xattr_handlers_min = gfs2_xattr_handlers_max + 1; diff --git a/fs/hfs/attr.c b/fs/hfs/attr.c index 74fa62643136..2bd54efaf416 100644 --- a/fs/hfs/attr.c +++ b/fs/hfs/attr.c @@ -121,6 +121,7 @@ static int hfs_xattr_get(const struct xattr_handler *handler, } static int hfs_xattr_set(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *unused, struct inode *inode, const char *name, const void *value, size_t size, int flags) diff --git a/fs/hfs/dir.c b/fs/hfs/dir.c index 3bf2ae0e467c..527f6e46cbe8 100644 --- a/fs/hfs/dir.c +++ b/fs/hfs/dir.c @@ -189,8 +189,8 @@ static int hfs_dir_release(struct inode *inode, struct file *file) * a directory and return a corresponding inode, given the inode for * the directory and the name (and its length) of the new file. */ -static int hfs_create(struct inode *dir, struct dentry *dentry, umode_t mode, - bool excl) +static int hfs_create(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, bool excl) { struct inode *inode; int res; @@ -219,7 +219,8 @@ static int hfs_create(struct inode *dir, struct dentry *dentry, umode_t mode, * in a directory, given the inode for the parent directory and the * name (and its length) of the new directory. */ -static int hfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) +static int hfs_mkdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { struct inode *inode; int res; @@ -279,9 +280,9 @@ static int hfs_remove(struct inode *dir, struct dentry *dentry) * new file/directory. * XXX: how do you handle must_be dir? */ -static int hfs_rename(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry, - unsigned int flags) +static int hfs_rename(struct user_namespace *mnt_userns, struct inode *old_dir, + struct dentry *old_dentry, struct inode *new_dir, + struct dentry *new_dentry, unsigned int flags) { int res; diff --git a/fs/hfs/hfs_fs.h b/fs/hfs/hfs_fs.h index f71c384064c8..b8eb0322a3e5 100644 --- a/fs/hfs/hfs_fs.h +++ b/fs/hfs/hfs_fs.h @@ -204,7 +204,8 @@ extern const struct address_space_operations hfs_btree_aops; extern struct inode *hfs_new_inode(struct inode *, const struct qstr *, umode_t); extern void hfs_inode_write_fork(struct inode *, struct hfs_extent *, __be32 *, __be32 *); extern int hfs_write_inode(struct inode *, struct writeback_control *); -extern int hfs_inode_setattr(struct dentry *, struct iattr *); +extern int hfs_inode_setattr(struct user_namespace *, struct dentry *, + struct iattr *); extern void hfs_inode_read_fork(struct inode *inode, struct hfs_extent *ext, __be32 log_size, __be32 phys_size, u32 clump_size); extern struct inode *hfs_iget(struct super_block *, struct hfs_cat_key *, hfs_cat_rec *); diff --git a/fs/hfs/inode.c b/fs/hfs/inode.c index f35a37c65e5f..3fc5cb346586 100644 --- a/fs/hfs/inode.c +++ b/fs/hfs/inode.c @@ -602,13 +602,15 @@ static int hfs_file_release(struct inode *inode, struct file *file) * correspond to the same HFS file. */ -int hfs_inode_setattr(struct dentry *dentry, struct iattr * attr) +int hfs_inode_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *attr) { struct inode *inode = d_inode(dentry); struct hfs_sb_info *hsb = HFS_SB(inode->i_sb); int error; - error = setattr_prepare(dentry, attr); /* basic permission checks */ + error = setattr_prepare(&init_user_ns, dentry, + attr); /* basic permission checks */ if (error) return error; @@ -647,7 +649,7 @@ int hfs_inode_setattr(struct dentry *dentry, struct iattr * attr) current_time(inode); } - setattr_copy(inode, attr); + setattr_copy(&init_user_ns, inode, attr); mark_inode_dirty(inode); return 0; } diff --git a/fs/hfsplus/dir.c b/fs/hfsplus/dir.c index 29a9dcfbe81f..03e6c046faf4 100644 --- a/fs/hfsplus/dir.c +++ b/fs/hfsplus/dir.c @@ -434,8 +434,8 @@ out: return res; } -static int hfsplus_symlink(struct inode *dir, struct dentry *dentry, - const char *symname) +static int hfsplus_symlink(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, const char *symname) { struct hfsplus_sb_info *sbi = HFSPLUS_SB(dir->i_sb); struct inode *inode; @@ -476,8 +476,8 @@ out: return res; } -static int hfsplus_mknod(struct inode *dir, struct dentry *dentry, - umode_t mode, dev_t rdev) +static int hfsplus_mknod(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, dev_t rdev) { struct hfsplus_sb_info *sbi = HFSPLUS_SB(dir->i_sb); struct inode *inode; @@ -517,18 +517,20 @@ out: return res; } -static int hfsplus_create(struct inode *dir, struct dentry *dentry, umode_t mode, - bool excl) +static int hfsplus_create(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, bool excl) { - return hfsplus_mknod(dir, dentry, mode, 0); + return hfsplus_mknod(&init_user_ns, dir, dentry, mode, 0); } -static int hfsplus_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) +static int hfsplus_mkdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { - return hfsplus_mknod(dir, dentry, mode | S_IFDIR, 0); + return hfsplus_mknod(&init_user_ns, dir, dentry, mode | S_IFDIR, 0); } -static int hfsplus_rename(struct inode *old_dir, struct dentry *old_dentry, +static int hfsplus_rename(struct user_namespace *mnt_userns, + struct inode *old_dir, struct dentry *old_dentry, struct inode *new_dir, struct dentry *new_dentry, unsigned int flags) { diff --git a/fs/hfsplus/hfsplus_fs.h b/fs/hfsplus/hfsplus_fs.h index a92de5199ec3..12b20479ed2b 100644 --- a/fs/hfsplus/hfsplus_fs.h +++ b/fs/hfsplus/hfsplus_fs.h @@ -488,8 +488,9 @@ void hfsplus_inode_write_fork(struct inode *inode, struct hfsplus_fork_raw *fork); int hfsplus_cat_read_inode(struct inode *inode, struct hfs_find_data *fd); int hfsplus_cat_write_inode(struct inode *inode); -int hfsplus_getattr(const struct path *path, struct kstat *stat, - u32 request_mask, unsigned int query_flags); +int hfsplus_getattr(struct user_namespace *mnt_userns, const struct path *path, + struct kstat *stat, u32 request_mask, + unsigned int query_flags); int hfsplus_file_fsync(struct file *file, loff_t start, loff_t end, int datasync); diff --git a/fs/hfsplus/inode.c b/fs/hfsplus/inode.c index e3da9e96b835..078c5c8a5156 100644 --- a/fs/hfsplus/inode.c +++ b/fs/hfsplus/inode.c @@ -241,12 +241,13 @@ static int hfsplus_file_release(struct inode *inode, struct file *file) return 0; } -static int hfsplus_setattr(struct dentry *dentry, struct iattr *attr) +static int hfsplus_setattr(struct user_namespace *mnt_userns, + struct dentry *dentry, struct iattr *attr) { struct inode *inode = d_inode(dentry); int error; - error = setattr_prepare(dentry, attr); + error = setattr_prepare(&init_user_ns, dentry, attr); if (error) return error; @@ -264,14 +265,15 @@ static int hfsplus_setattr(struct dentry *dentry, struct iattr *attr) inode->i_mtime = inode->i_ctime = current_time(inode); } - setattr_copy(inode, attr); + setattr_copy(&init_user_ns, inode, attr); mark_inode_dirty(inode); return 0; } -int hfsplus_getattr(const struct path *path, struct kstat *stat, - u32 request_mask, unsigned int query_flags) +int hfsplus_getattr(struct user_namespace *mnt_userns, const struct path *path, + struct kstat *stat, u32 request_mask, + unsigned int query_flags) { struct inode *inode = d_inode(path->dentry); struct hfsplus_inode_info *hip = HFSPLUS_I(inode); @@ -286,7 +288,7 @@ int hfsplus_getattr(const struct path *path, struct kstat *stat, stat->attributes_mask |= STATX_ATTR_APPEND | STATX_ATTR_IMMUTABLE | STATX_ATTR_NODUMP; - generic_fillattr(inode, stat); + generic_fillattr(&init_user_ns, inode, stat); return 0; } @@ -340,7 +342,7 @@ int hfsplus_file_fsync(struct file *file, loff_t start, loff_t end, } if (!test_bit(HFSPLUS_SB_NOBARRIER, &sbi->flags)) - blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL); + blkdev_issue_flush(inode->i_sb->s_bdev); inode_unlock(inode); @@ -376,7 +378,7 @@ struct inode *hfsplus_new_inode(struct super_block *sb, struct inode *dir, return NULL; inode->i_ino = sbi->next_cnid++; - inode_init_owner(inode, dir, mode); + inode_init_owner(&init_user_ns, inode, dir, mode); set_nlink(inode, 1); inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode); diff --git a/fs/hfsplus/ioctl.c b/fs/hfsplus/ioctl.c index ce15b9496b77..3edb1926d127 100644 --- a/fs/hfsplus/ioctl.c +++ b/fs/hfsplus/ioctl.c @@ -91,7 +91,7 @@ static int hfsplus_ioctl_setflags(struct file *file, int __user *user_flags) if (err) goto out; - if (!inode_owner_or_capable(inode)) { + if (!inode_owner_or_capable(&init_user_ns, inode)) { err = -EACCES; goto out_drop_write; } diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c index 807119ae5adf..b9e3db3f855f 100644 --- a/fs/hfsplus/super.c +++ b/fs/hfsplus/super.c @@ -239,7 +239,7 @@ out: mutex_unlock(&sbi->vh_mutex); if (!test_bit(HFSPLUS_SB_NOBARRIER, &sbi->flags)) - blkdev_issue_flush(sb->s_bdev, GFP_KERNEL); + blkdev_issue_flush(sb->s_bdev); return error; } diff --git a/fs/hfsplus/xattr.c b/fs/hfsplus/xattr.c index bb0b27d88e50..4d169c5a2673 100644 --- a/fs/hfsplus/xattr.c +++ b/fs/hfsplus/xattr.c @@ -858,6 +858,7 @@ static int hfsplus_osx_getxattr(const struct xattr_handler *handler, } static int hfsplus_osx_setxattr(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *unused, struct inode *inode, const char *name, const void *buffer, size_t size, int flags) diff --git a/fs/hfsplus/xattr_security.c b/fs/hfsplus/xattr_security.c index cfbe6a3bfb1e..c1c7a16cbf21 100644 --- a/fs/hfsplus/xattr_security.c +++ b/fs/hfsplus/xattr_security.c @@ -23,6 +23,7 @@ static int hfsplus_security_getxattr(const struct xattr_handler *handler, } static int hfsplus_security_setxattr(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *unused, struct inode *inode, const char *name, const void *buffer, size_t size, int flags) diff --git a/fs/hfsplus/xattr_trusted.c b/fs/hfsplus/xattr_trusted.c index fbad91e1dada..e150372ec564 100644 --- a/fs/hfsplus/xattr_trusted.c +++ b/fs/hfsplus/xattr_trusted.c @@ -22,6 +22,7 @@ static int hfsplus_trusted_getxattr(const struct xattr_handler *handler, } static int hfsplus_trusted_setxattr(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *unused, struct inode *inode, const char *name, const void *buffer, size_t size, int flags) diff --git a/fs/hfsplus/xattr_user.c b/fs/hfsplus/xattr_user.c index 74d19faf255e..a6b60b153916 100644 --- a/fs/hfsplus/xattr_user.c +++ b/fs/hfsplus/xattr_user.c @@ -22,6 +22,7 @@ static int hfsplus_user_getxattr(const struct xattr_handler *handler, } static int hfsplus_user_setxattr(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *unused, struct inode *inode, const char *name, const void *buffer, size_t size, int flags) diff --git a/fs/hostfs/hostfs_kern.c b/fs/hostfs/hostfs_kern.c index aea35459d390..29e407762626 100644 --- a/fs/hostfs/hostfs_kern.c +++ b/fs/hostfs/hostfs_kern.c @@ -34,6 +34,8 @@ static inline struct hostfs_inode_info *HOSTFS_I(struct inode *inode) #define FILE_HOSTFS_I(file) HOSTFS_I(file_inode(file)) +static struct kmem_cache *hostfs_inode_cache; + /* Changed in hostfs_args before the kernel starts running */ static char *root_ino = ""; static int append = 0; @@ -221,7 +223,7 @@ static struct inode *hostfs_alloc_inode(struct super_block *sb) { struct hostfs_inode_info *hi; - hi = kmalloc(sizeof(*hi), GFP_KERNEL_ACCOUNT); + hi = kmem_cache_alloc(hostfs_inode_cache, GFP_KERNEL_ACCOUNT); if (hi == NULL) return NULL; hi->fd = -1; @@ -243,7 +245,7 @@ static void hostfs_evict_inode(struct inode *inode) static void hostfs_free_inode(struct inode *inode) { - kfree(HOSTFS_I(inode)); + kmem_cache_free(hostfs_inode_cache, HOSTFS_I(inode)); } static int hostfs_show_options(struct seq_file *seq, struct dentry *root) @@ -555,8 +557,8 @@ static int read_name(struct inode *ino, char *name) return 0; } -static int hostfs_create(struct inode *dir, struct dentry *dentry, umode_t mode, - bool excl) +static int hostfs_create(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, bool excl) { struct inode *inode; char *name; @@ -654,8 +656,8 @@ static int hostfs_unlink(struct inode *ino, struct dentry *dentry) return err; } -static int hostfs_symlink(struct inode *ino, struct dentry *dentry, - const char *to) +static int hostfs_symlink(struct user_namespace *mnt_userns, struct inode *ino, + struct dentry *dentry, const char *to) { char *file; int err; @@ -667,7 +669,8 @@ static int hostfs_symlink(struct inode *ino, struct dentry *dentry, return err; } -static int hostfs_mkdir(struct inode *ino, struct dentry *dentry, umode_t mode) +static int hostfs_mkdir(struct user_namespace *mnt_userns, struct inode *ino, + struct dentry *dentry, umode_t mode) { char *file; int err; @@ -691,7 +694,8 @@ static int hostfs_rmdir(struct inode *ino, struct dentry *dentry) return err; } -static int hostfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t dev) +static int hostfs_mknod(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, dev_t dev) { struct inode *inode; char *name; @@ -729,7 +733,8 @@ static int hostfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, return err; } -static int hostfs_rename2(struct inode *old_dir, struct dentry *old_dentry, +static int hostfs_rename2(struct user_namespace *mnt_userns, + struct inode *old_dir, struct dentry *old_dentry, struct inode *new_dir, struct dentry *new_dentry, unsigned int flags) { @@ -757,7 +762,8 @@ static int hostfs_rename2(struct inode *old_dir, struct dentry *old_dentry, return err; } -static int hostfs_permission(struct inode *ino, int desired) +static int hostfs_permission(struct user_namespace *mnt_userns, + struct inode *ino, int desired) { char *name; int r = 0, w = 0, x = 0, err; @@ -779,11 +785,12 @@ static int hostfs_permission(struct inode *ino, int desired) err = access_file(name, r, w, x); __putname(name); if (!err) - err = generic_permission(ino, desired); + err = generic_permission(&init_user_ns, ino, desired); return err; } -static int hostfs_setattr(struct dentry *dentry, struct iattr *attr) +static int hostfs_setattr(struct user_namespace *mnt_userns, + struct dentry *dentry, struct iattr *attr) { struct inode *inode = d_inode(dentry); struct hostfs_iattr attrs; @@ -792,7 +799,7 @@ static int hostfs_setattr(struct dentry *dentry, struct iattr *attr) int fd = HOSTFS_I(inode)->fd; - err = setattr_prepare(dentry, attr); + err = setattr_prepare(&init_user_ns, dentry, attr); if (err) return err; @@ -849,7 +856,7 @@ static int hostfs_setattr(struct dentry *dentry, struct iattr *attr) attr->ia_size != i_size_read(inode)) truncate_setsize(inode, attr->ia_size); - setattr_copy(inode, attr); + setattr_copy(&init_user_ns, inode, attr); mark_inode_dirty(inode); return 0; } @@ -986,12 +993,16 @@ MODULE_ALIAS_FS("hostfs"); static int __init init_hostfs(void) { + hostfs_inode_cache = KMEM_CACHE(hostfs_inode_info, 0); + if (!hostfs_inode_cache) + return -ENOMEM; return register_filesystem(&hostfs_type); } static void __exit exit_hostfs(void) { unregister_filesystem(&hostfs_type); + kmem_cache_destroy(hostfs_inode_cache); } module_init(init_hostfs) diff --git a/fs/hpfs/hpfs_fn.h b/fs/hpfs/hpfs_fn.h index 1cca83218fb5..167ec6884642 100644 --- a/fs/hpfs/hpfs_fn.h +++ b/fs/hpfs/hpfs_fn.h @@ -280,7 +280,7 @@ void hpfs_init_inode(struct inode *); void hpfs_read_inode(struct inode *); void hpfs_write_inode(struct inode *); void hpfs_write_inode_nolock(struct inode *); -int hpfs_setattr(struct dentry *, struct iattr *); +int hpfs_setattr(struct user_namespace *, struct dentry *, struct iattr *); void hpfs_write_if_changed(struct inode *); void hpfs_evict_inode(struct inode *); diff --git a/fs/hpfs/inode.c b/fs/hpfs/inode.c index eb8b4baf0f2e..82208cc28ebd 100644 --- a/fs/hpfs/inode.c +++ b/fs/hpfs/inode.c @@ -257,7 +257,8 @@ void hpfs_write_inode_nolock(struct inode *i) brelse(bh); } -int hpfs_setattr(struct dentry *dentry, struct iattr *attr) +int hpfs_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *attr) { struct inode *inode = d_inode(dentry); int error = -EINVAL; @@ -274,7 +275,7 @@ int hpfs_setattr(struct dentry *dentry, struct iattr *attr) if ((attr->ia_valid & ATTR_SIZE) && attr->ia_size > inode->i_size) goto out_unlock; - error = setattr_prepare(dentry, attr); + error = setattr_prepare(&init_user_ns, dentry, attr); if (error) goto out_unlock; @@ -288,7 +289,7 @@ int hpfs_setattr(struct dentry *dentry, struct iattr *attr) hpfs_truncate(inode); } - setattr_copy(inode, attr); + setattr_copy(&init_user_ns, inode, attr); hpfs_write_inode(inode); diff --git a/fs/hpfs/namei.c b/fs/hpfs/namei.c index 1aee39160ac5..d73f8a67168e 100644 --- a/fs/hpfs/namei.c +++ b/fs/hpfs/namei.c @@ -20,7 +20,8 @@ static void hpfs_update_directory_times(struct inode *dir) hpfs_write_inode_nolock(dir); } -static int hpfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) +static int hpfs_mkdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { const unsigned char *name = dentry->d_name.name; unsigned len = dentry->d_name.len; @@ -128,7 +129,8 @@ bail: return err; } -static int hpfs_create(struct inode *dir, struct dentry *dentry, umode_t mode, bool excl) +static int hpfs_create(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, bool excl) { const unsigned char *name = dentry->d_name.name; unsigned len = dentry->d_name.len; @@ -215,7 +217,8 @@ bail: return err; } -static int hpfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t rdev) +static int hpfs_mknod(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, dev_t rdev) { const unsigned char *name = dentry->d_name.name; unsigned len = dentry->d_name.len; @@ -289,7 +292,8 @@ bail: return err; } -static int hpfs_symlink(struct inode *dir, struct dentry *dentry, const char *symlink) +static int hpfs_symlink(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, const char *symlink) { const unsigned char *name = dentry->d_name.name; unsigned len = dentry->d_name.len; @@ -506,10 +510,10 @@ fail: const struct address_space_operations hpfs_symlink_aops = { .readpage = hpfs_symlink_readpage }; - -static int hpfs_rename(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry, - unsigned int flags) + +static int hpfs_rename(struct user_namespace *mnt_userns, struct inode *old_dir, + struct dentry *old_dentry, struct inode *new_dir, + struct dentry *new_dentry, unsigned int flags) { const unsigned char *old_name = old_dentry->d_name.name; unsigned old_len = old_dentry->d_name.len; diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 21c20fd5f9ee..701c82c36138 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -171,7 +171,7 @@ static int hugetlbfs_file_mmap(struct file *file, struct vm_area_struct *vma) file_accessed(file); ret = -ENOMEM; - if (hugetlb_reserve_pages(inode, + if (!hugetlb_reserve_pages(inode, vma->vm_pgoff >> huge_page_order(h), len >> huge_page_shift(h), vma, vma->vm_flags)) @@ -310,7 +310,7 @@ hugetlbfs_read_actor(struct page *page, unsigned long offset, /* * Support for read() - Find the page attached to f_mapping and copy out the - * data. Its *very* similar to do_generic_mapping_read(), we can't use that + * data. Its *very* similar to generic_file_buffered_read(), we can't use that * since it has PAGE_SIZE assumptions. */ static ssize_t hugetlbfs_read_iter(struct kiocb *iocb, struct iov_iter *to) @@ -442,15 +442,15 @@ hugetlb_vmdelete_list(struct rb_root_cached *root, pgoff_t start, pgoff_t end) * * truncation is indicated by end of range being LLONG_MAX * In this case, we first scan the range and release found pages. - * After releasing pages, hugetlb_unreserve_pages cleans up region/reserv + * After releasing pages, hugetlb_unreserve_pages cleans up region/reserve * maps and global counts. Page faults can not race with truncation * in this routine. hugetlb_no_page() holds i_mmap_rwsem and prevents * page faults in the truncated range by checking i_size. i_size is * modified while holding i_mmap_rwsem. * hole punch is indicated if end is not LLONG_MAX * In the hole punch case we scan the range and release found pages. - * Only when releasing a page is the associated region/reserv map - * deleted. The region/reserv map for ranges without associated + * Only when releasing a page is the associated region/reserve map + * deleted. The region/reserve map for ranges without associated * pages are not modified. Page faults can race with hole punch. * This is indicated if we find a mapped page. * Note: If the passed end of range value is beyond the end of file, but @@ -567,7 +567,7 @@ static void hugetlbfs_evict_inode(struct inode *inode) clear_inode(inode); } -static int hugetlb_vmtruncate(struct inode *inode, loff_t offset) +static void hugetlb_vmtruncate(struct inode *inode, loff_t offset) { pgoff_t pgoff; struct address_space *mapping = inode->i_mapping; @@ -582,7 +582,6 @@ static int hugetlb_vmtruncate(struct inode *inode, loff_t offset) hugetlb_vmdelete_list(&mapping->i_mmap, pgoff, 0); i_mmap_unlock_write(mapping); remove_inode_hugepages(inode, offset, LLONG_MAX); - return 0; } static long hugetlbfs_punch_hole(struct inode *inode, loff_t offset, loff_t len) @@ -604,7 +603,7 @@ static long hugetlbfs_punch_hole(struct inode *inode, loff_t offset, loff_t len) inode_lock(inode); - /* protected by i_mutex */ + /* protected by i_rwsem */ if (info->seals & (F_SEAL_WRITE | F_SEAL_FUTURE_WRITE)) { inode_unlock(inode); return -EPERM; @@ -680,7 +679,6 @@ static long hugetlbfs_fallocate(struct file *file, int mode, loff_t offset, */ struct page *page; unsigned long addr; - int avoid_reserve = 0; cond_resched(); @@ -716,8 +714,15 @@ static long hugetlbfs_fallocate(struct file *file, int mode, loff_t offset, continue; } - /* Allocate page and add to page cache */ - page = alloc_huge_page(&pseudo_vma, addr, avoid_reserve); + /* + * Allocate page without setting the avoid_reserve argument. + * There certainly are no reserves associated with the + * pseudo_vma. However, there could be shared mappings with + * reserves for the file at the inode level. If we fallocate + * pages in these areas, we need to consume the reserves + * to keep reservation accounting consistent. + */ + page = alloc_huge_page(&pseudo_vma, addr, 0); hugetlb_drop_vma_policy(&pseudo_vma); if (IS_ERR(page)) { mutex_unlock(&hugetlb_fault_mutex_table[hash]); @@ -735,7 +740,7 @@ static long hugetlbfs_fallocate(struct file *file, int mode, loff_t offset, mutex_unlock(&hugetlb_fault_mutex_table[hash]); - set_page_huge_active(page); + SetHPageMigratable(page); /* * unlock_page because locked by add_to_page_cache() * put_page() due to reference from alloc_huge_page() @@ -752,7 +757,8 @@ out: return error; } -static int hugetlbfs_setattr(struct dentry *dentry, struct iattr *attr) +static int hugetlbfs_setattr(struct user_namespace *mnt_userns, + struct dentry *dentry, struct iattr *attr) { struct inode *inode = d_inode(dentry); struct hstate *h = hstate_inode(inode); @@ -760,9 +766,7 @@ static int hugetlbfs_setattr(struct dentry *dentry, struct iattr *attr) unsigned int ia_valid = attr->ia_valid; struct hugetlbfs_inode_info *info = HUGETLBFS_I(inode); - BUG_ON(!inode); - - error = setattr_prepare(dentry, attr); + error = setattr_prepare(&init_user_ns, dentry, attr); if (error) return error; @@ -772,16 +776,14 @@ static int hugetlbfs_setattr(struct dentry *dentry, struct iattr *attr) if (newsize & ~huge_page_mask(h)) return -EINVAL; - /* protected by i_mutex */ + /* protected by i_rwsem */ if ((newsize < oldsize && (info->seals & F_SEAL_SHRINK)) || (newsize > oldsize && (info->seals & F_SEAL_GROW))) return -EPERM; - error = hugetlb_vmtruncate(inode, newsize); - if (error) - return error; + hugetlb_vmtruncate(inode, newsize); } - setattr_copy(inode, attr); + setattr_copy(&init_user_ns, inode, attr); mark_inode_dirty(inode); return 0; } @@ -837,7 +839,7 @@ static struct inode *hugetlbfs_get_inode(struct super_block *sb, struct hugetlbfs_inode_info *info = HUGETLBFS_I(inode); inode->i_ino = get_next_ino(); - inode_init_owner(inode, dir, mode); + inode_init_owner(&init_user_ns, inode, dir, mode); lockdep_set_class(&inode->i_mapping->i_mmap_rwsem, &hugetlbfs_i_mmap_rwsem_key); inode->i_mapping->a_ops = &hugetlbfs_aops; @@ -899,33 +901,39 @@ static int do_hugetlbfs_mknod(struct inode *dir, return error; } -static int hugetlbfs_mknod(struct inode *dir, - struct dentry *dentry, umode_t mode, dev_t dev) +static int hugetlbfs_mknod(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, dev_t dev) { return do_hugetlbfs_mknod(dir, dentry, mode, dev, false); } -static int hugetlbfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) +static int hugetlbfs_mkdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { - int retval = hugetlbfs_mknod(dir, dentry, mode | S_IFDIR, 0); + int retval = hugetlbfs_mknod(&init_user_ns, dir, dentry, + mode | S_IFDIR, 0); if (!retval) inc_nlink(dir); return retval; } -static int hugetlbfs_create(struct inode *dir, struct dentry *dentry, umode_t mode, bool excl) +static int hugetlbfs_create(struct user_namespace *mnt_userns, + struct inode *dir, struct dentry *dentry, + umode_t mode, bool excl) { - return hugetlbfs_mknod(dir, dentry, mode | S_IFREG, 0); + return hugetlbfs_mknod(&init_user_ns, dir, dentry, mode | S_IFREG, 0); } -static int hugetlbfs_tmpfile(struct inode *dir, - struct dentry *dentry, umode_t mode) +static int hugetlbfs_tmpfile(struct user_namespace *mnt_userns, + struct inode *dir, struct dentry *dentry, + umode_t mode) { return do_hugetlbfs_mknod(dir, dentry, mode | S_IFREG, 0, true); } -static int hugetlbfs_symlink(struct inode *dir, - struct dentry *dentry, const char *symname) +static int hugetlbfs_symlink(struct user_namespace *mnt_userns, + struct inode *dir, struct dentry *dentry, + const char *symname) { struct inode *inode; int error = -ENOSPC; @@ -945,17 +953,6 @@ static int hugetlbfs_symlink(struct inode *dir, return error; } -/* - * mark the head page dirty - */ -static int hugetlbfs_set_page_dirty(struct page *page) -{ - struct page *head = compound_head(page); - - SetPageDirty(head); - return 0; -} - static int hugetlbfs_migrate_page(struct address_space *mapping, struct page *newpage, struct page *page, enum migrate_mode mode) @@ -966,15 +963,9 @@ static int hugetlbfs_migrate_page(struct address_space *mapping, if (rc != MIGRATEPAGE_SUCCESS) return rc; - /* - * page_private is subpool pointer in hugetlb pages. Transfer to - * new page. PagePrivate is not associated with page_private for - * hugetlb pages and can not be set here as only page_huge_active - * pages can be migrated. - */ - if (page_private(page)) { - set_page_private(newpage, page_private(page)); - set_page_private(page, 0); + if (hugetlb_page_subpool(page)) { + hugetlb_set_page_subpool(newpage, hugetlb_page_subpool(page)); + hugetlb_set_page_subpool(page, NULL); } if (mode != MIGRATE_SYNC_NO_COPY) @@ -1149,7 +1140,7 @@ static void hugetlbfs_destroy_inode(struct inode *inode) static const struct address_space_operations hugetlbfs_aops = { .write_begin = hugetlbfs_write_begin, .write_end = hugetlbfs_write_end, - .set_page_dirty = hugetlbfs_set_page_dirty, + .set_page_dirty = __set_page_dirty_no_writeback, .migratepage = hugetlbfs_migrate_page, .error_remove_page = hugetlbfs_error_remove_page, }; @@ -1349,7 +1340,7 @@ hugetlbfs_fill_super(struct super_block *sb, struct fs_context *fc) /* * Allocate and initialize subpool if maximum or minimum size is - * specified. Any needed reservations (for minimim size) are taken + * specified. Any needed reservations (for minimum size) are taken * taken when the subpool is created. */ if (ctx->max_hpages != -1 || ctx->min_hpages != -1) { @@ -1492,7 +1483,7 @@ struct file *hugetlb_file_setup(const char *name, size_t size, inode->i_size = size; clear_nlink(inode); - if (hugetlb_reserve_pages(inode, 0, + if (!hugetlb_reserve_pages(inode, 0, size >> huge_page_shift(hstate_inode(inode)), NULL, acctflag)) file = ERR_PTR(-ENOMEM); @@ -1526,8 +1517,8 @@ static struct vfsmount *__init mount_one_hugetlbfs(struct hstate *h) put_fs_context(fc); } if (IS_ERR(mnt)) - pr_err("Cannot mount internal hugetlbfs for page size %uK", - 1U << (h->order + PAGE_SHIFT - 10)); + pr_err("Cannot mount internal hugetlbfs for page size %luK", + huge_page_size(h) >> 10); return mnt; } @@ -1555,7 +1546,7 @@ static int __init init_hugetlbfs_fs(void) goto out_free; /* default hstate mount is required */ - mnt = mount_one_hugetlbfs(&hstates[default_hstate_idx]); + mnt = mount_one_hugetlbfs(&default_hstate); if (IS_ERR(mnt)) { error = PTR_ERR(mnt); goto out_unreg; diff --git a/fs/init.c b/fs/init.c index e9c320a48cf1..5c36adaa9b44 100644 --- a/fs/init.c +++ b/fs/init.c @@ -49,7 +49,7 @@ int __init init_chdir(const char *filename) error = kern_path(filename, LOOKUP_FOLLOW | LOOKUP_DIRECTORY, &path); if (error) return error; - error = inode_permission(path.dentry->d_inode, MAY_EXEC | MAY_CHDIR); + error = path_permission(&path, MAY_EXEC | MAY_CHDIR); if (!error) set_fs_pwd(current->fs, &path); path_put(&path); @@ -64,7 +64,7 @@ int __init init_chroot(const char *filename) error = kern_path(filename, LOOKUP_FOLLOW | LOOKUP_DIRECTORY, &path); if (error) return error; - error = inode_permission(path.dentry->d_inode, MAY_EXEC | MAY_CHDIR); + error = path_permission(&path, MAY_EXEC | MAY_CHDIR); if (error) goto dput_and_out; error = -EPERM; @@ -118,7 +118,7 @@ int __init init_eaccess(const char *filename) error = kern_path(filename, LOOKUP_FOLLOW, &path); if (error) return error; - error = inode_permission(d_inode(path.dentry), MAY_ACCESS); + error = path_permission(&path, MAY_ACCESS); path_put(&path); return error; } @@ -157,8 +157,8 @@ int __init init_mknod(const char *filename, umode_t mode, unsigned int dev) mode &= ~current_umask(); error = security_path_mknod(&path, dentry, mode, dev); if (!error) - error = vfs_mknod(path.dentry->d_inode, dentry, mode, - new_decode_dev(dev)); + error = vfs_mknod(mnt_user_ns(path.mnt), path.dentry->d_inode, + dentry, mode, new_decode_dev(dev)); done_path_create(&path, dentry); return error; } @@ -167,6 +167,7 @@ int __init init_link(const char *oldname, const char *newname) { struct dentry *new_dentry; struct path old_path, new_path; + struct user_namespace *mnt_userns; int error; error = kern_path(oldname, 0, &old_path); @@ -181,14 +182,15 @@ int __init init_link(const char *oldname, const char *newname) error = -EXDEV; if (old_path.mnt != new_path.mnt) goto out_dput; - error = may_linkat(&old_path); + mnt_userns = mnt_user_ns(new_path.mnt); + error = may_linkat(mnt_userns, &old_path); if (unlikely(error)) goto out_dput; error = security_path_link(old_path.dentry, &new_path, new_dentry); if (error) goto out_dput; - error = vfs_link(old_path.dentry, new_path.dentry->d_inode, new_dentry, - NULL); + error = vfs_link(old_path.dentry, mnt_userns, new_path.dentry->d_inode, + new_dentry, NULL); out_dput: done_path_create(&new_path, new_dentry); out: @@ -207,7 +209,8 @@ int __init init_symlink(const char *oldname, const char *newname) return PTR_ERR(dentry); error = security_path_symlink(&path, dentry, oldname); if (!error) - error = vfs_symlink(path.dentry->d_inode, dentry, oldname); + error = vfs_symlink(mnt_user_ns(path.mnt), path.dentry->d_inode, + dentry, oldname); done_path_create(&path, dentry); return error; } @@ -230,7 +233,8 @@ int __init init_mkdir(const char *pathname, umode_t mode) mode &= ~current_umask(); error = security_path_mkdir(&path, dentry, mode); if (!error) - error = vfs_mkdir(path.dentry->d_inode, dentry, mode); + error = vfs_mkdir(mnt_user_ns(path.mnt), path.dentry->d_inode, + dentry, mode); done_path_create(&path, dentry); return error; } diff --git a/fs/inode.c b/fs/inode.c index 6442d97d9a4a..a047ab306f9a 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -142,6 +142,7 @@ int inode_init_always(struct super_block *sb, struct inode *inode) atomic_set(&inode->i_count, 1); inode->i_op = &empty_iops; inode->i_fop = &no_open_fops; + inode->i_ino = 0; inode->__i_nlink = 1; inode->i_opflags = 0; if (sb->s_xattr) @@ -1493,7 +1494,7 @@ struct inode *find_inode_rcu(struct super_block *sb, unsigned long hashval, EXPORT_SYMBOL(find_inode_rcu); /** - * find_inode_by_rcu - Find an inode in the inode cache + * find_inode_by_ino_rcu - Find an inode in the inode cache * @sb: Super block of file system to search * @ino: The inode number to match * @@ -1743,24 +1744,26 @@ static int relatime_need_update(struct vfsmount *mnt, struct inode *inode, int generic_update_time(struct inode *inode, struct timespec64 *time, int flags) { - int iflags = I_DIRTY_TIME; - bool dirty = false; - - if (flags & S_ATIME) - inode->i_atime = *time; - if (flags & S_VERSION) - dirty = inode_maybe_inc_iversion(inode, false); - if (flags & S_CTIME) - inode->i_ctime = *time; - if (flags & S_MTIME) - inode->i_mtime = *time; - if ((flags & (S_ATIME | S_CTIME | S_MTIME)) && - !(inode->i_sb->s_flags & SB_LAZYTIME)) - dirty = true; - - if (dirty) - iflags |= I_DIRTY_SYNC; - __mark_inode_dirty(inode, iflags); + int dirty_flags = 0; + + if (flags & (S_ATIME | S_CTIME | S_MTIME)) { + if (flags & S_ATIME) + inode->i_atime = *time; + if (flags & S_CTIME) + inode->i_ctime = *time; + if (flags & S_MTIME) + inode->i_mtime = *time; + + if (inode->i_sb->s_flags & SB_LAZYTIME) + dirty_flags |= I_DIRTY_TIME; + else + dirty_flags |= I_DIRTY_SYNC; + } + + if ((flags & S_VERSION) && inode_maybe_inc_iversion(inode, false)) + dirty_flags |= I_DIRTY_SYNC; + + __mark_inode_dirty(inode, dirty_flags); return 0; } EXPORT_SYMBOL(generic_update_time); @@ -1777,7 +1780,7 @@ static int update_time(struct inode *inode, struct timespec64 *time, int flags) } /** - * touch_atime - update the access time + * atime_needs_update - update the access time * @path: the &struct path to update * @inode: inode to update * @@ -1796,7 +1799,7 @@ bool atime_needs_update(const struct path *path, struct inode *inode) /* Atime updates will likely cause i_uid and i_gid to be written * back improprely if their true value is unknown to the vfs. */ - if (HAS_UNMAPPED_ID(inode)) + if (HAS_UNMAPPED_ID(mnt_user_ns(mnt), inode)) return false; if (IS_NOATIME(inode)) @@ -1903,7 +1906,8 @@ int dentry_needs_remove_privs(struct dentry *dentry) return mask; } -static int __remove_privs(struct dentry *dentry, int kill) +static int __remove_privs(struct user_namespace *mnt_userns, + struct dentry *dentry, int kill) { struct iattr newattrs; @@ -1912,7 +1916,7 @@ static int __remove_privs(struct dentry *dentry, int kill) * Note we call this on write, so notify_change will not * encounter any conflicting delegations: */ - return notify_change(dentry, &newattrs, NULL); + return notify_change(mnt_userns, dentry, &newattrs, NULL); } /* @@ -1939,7 +1943,7 @@ int file_remove_privs(struct file *file) if (kill < 0) return kill; if (kill) - error = __remove_privs(dentry, kill); + error = __remove_privs(file_mnt_user_ns(file), dentry, kill); if (!error) inode_has_no_xattr(inode); @@ -2130,14 +2134,21 @@ EXPORT_SYMBOL(init_special_inode); /** * inode_init_owner - Init uid,gid,mode for new inode according to posix standards + * @mnt_userns: User namespace of the mount the inode was created from * @inode: New inode * @dir: Directory inode * @mode: mode of the new inode + * + * If the inode has been created through an idmapped mount the user namespace of + * the vfsmount must be passed through @mnt_userns. This function will then take + * care to map the inode according to @mnt_userns before checking permissions + * and initializing i_uid and i_gid. On non-idmapped mounts or if permission + * checking is to be performed on the raw inode simply passs init_user_ns. */ -void inode_init_owner(struct inode *inode, const struct inode *dir, - umode_t mode) +void inode_init_owner(struct user_namespace *mnt_userns, struct inode *inode, + const struct inode *dir, umode_t mode) { - inode->i_uid = current_fsuid(); + inode->i_uid = fsuid_into_mnt(mnt_userns); if (dir && dir->i_mode & S_ISGID) { inode->i_gid = dir->i_gid; @@ -2145,31 +2156,41 @@ void inode_init_owner(struct inode *inode, const struct inode *dir, if (S_ISDIR(mode)) mode |= S_ISGID; else if ((mode & (S_ISGID | S_IXGRP)) == (S_ISGID | S_IXGRP) && - !in_group_p(inode->i_gid) && - !capable_wrt_inode_uidgid(dir, CAP_FSETID)) + !in_group_p(i_gid_into_mnt(mnt_userns, dir)) && + !capable_wrt_inode_uidgid(mnt_userns, dir, CAP_FSETID)) mode &= ~S_ISGID; } else - inode->i_gid = current_fsgid(); + inode->i_gid = fsgid_into_mnt(mnt_userns); inode->i_mode = mode; } EXPORT_SYMBOL(inode_init_owner); /** * inode_owner_or_capable - check current task permissions to inode + * @mnt_userns: user namespace of the mount the inode was found from * @inode: inode being checked * * Return true if current either has CAP_FOWNER in a namespace with the * inode owner uid mapped, or owns the file. + * + * If the inode has been found through an idmapped mount the user namespace of + * the vfsmount must be passed through @mnt_userns. This function will then take + * care to map the inode according to @mnt_userns before checking permissions. + * On non-idmapped mounts or if permission checking is to be performed on the + * raw inode simply passs init_user_ns. */ -bool inode_owner_or_capable(const struct inode *inode) +bool inode_owner_or_capable(struct user_namespace *mnt_userns, + const struct inode *inode) { + kuid_t i_uid; struct user_namespace *ns; - if (uid_eq(current_fsuid(), inode->i_uid)) + i_uid = i_uid_into_mnt(mnt_userns, inode); + if (uid_eq(current_fsuid(), i_uid)) return true; ns = current_user_ns(); - if (kuid_has_mapping(ns, inode->i_uid) && ns_capable(ns, CAP_FOWNER)) + if (kuid_has_mapping(ns, i_uid) && ns_capable(ns, CAP_FOWNER)) return true; return false; } diff --git a/fs/internal.h b/fs/internal.h index 77c50befbfbe..6aeae7ef3380 100644 --- a/fs/internal.h +++ b/fs/internal.h @@ -15,6 +15,7 @@ struct mount; struct shrink_control; struct fs_context; struct user_namespace; +struct pipe_inode_info; /* * block_dev.c @@ -73,7 +74,7 @@ extern int vfs_path_lookup(struct dentry *, struct vfsmount *, const char *, unsigned int, struct path *); long do_rmdir(int dfd, struct filename *name); long do_unlinkat(int dfd, struct filename *name); -int may_linkat(struct path *link); +int may_linkat(struct user_namespace *mnt_userns, struct path *link); int do_renameat2(int olddfd, struct filename *oldname, int newdfd, struct filename *newname, unsigned int flags); @@ -132,6 +133,7 @@ extern struct file *do_file_open_root(struct dentry *, struct vfsmount *, const char *, const struct open_flags *); extern struct open_how build_open_how(int flags, umode_t mode); extern int build_open_flags(const struct open_how *how, struct open_flags *op); +extern int __close_fd_get_file(unsigned int fd, struct file **res); long do_sys_ftruncate(unsigned int fd, loff_t length, int small); int chmod_common(const struct path *path, umode_t mode); @@ -193,3 +195,11 @@ int sb_init_dio_done_wq(struct super_block *sb); */ int do_statx(int dfd, const char __user *filename, unsigned flags, unsigned int mask, struct statx __user *buffer); + +/* + * fs/splice.c: + */ +long splice_file_to_pipe(struct file *in, + struct pipe_inode_info *opipe, + loff_t *offset, + size_t len, unsigned int flags); diff --git a/fs/io-wq.c b/fs/io-wq.c index a564f36e260c..3dc10bfd8c3b 100644 --- a/fs/io-wq.c +++ b/fs/io-wq.c @@ -13,13 +13,10 @@ #include <linux/sched/mm.h> #include <linux/percpu.h> #include <linux/slab.h> -#include <linux/kthread.h> #include <linux/rculist_nulls.h> -#include <linux/fs_struct.h> -#include <linux/task_work.h> -#include <linux/blk-cgroup.h> -#include <linux/audit.h> #include <linux/cpu.h> +#include <linux/tracehook.h> +#include <linux/freezer.h> #include "../kernel/sched/sched.h" #include "io-wq.h" @@ -36,7 +33,6 @@ enum { enum { IO_WQ_BIT_EXIT = 0, /* wq exiting */ - IO_WQ_BIT_ERROR = 1, /* error on setup */ }; enum { @@ -57,16 +53,9 @@ struct io_worker { struct io_wq_work *cur_work; spinlock_t lock; + struct completion ref_done; + struct rcu_head rcu; - struct mm_struct *mm; -#ifdef CONFIG_BLK_CGROUP - struct cgroup_subsys_state *blkcg_css; -#endif - const struct cred *cur_creds; - const struct cred *saved_creds; - struct files_struct *restore_files; - struct nsproxy *restore_nsproxy; - struct fs_struct *restore_fs; }; #if BITS_PER_LONG == 64 @@ -95,7 +84,6 @@ struct io_wqe { struct { raw_spinlock_t lock; struct io_wq_work_list work_list; - unsigned long hash_map; unsigned flags; } ____cacheline_aligned_in_smp; @@ -105,6 +93,8 @@ struct io_wqe { struct hlist_nulls_head free_list; struct list_head all_list; + struct wait_queue_entry wait; + struct io_wq *wq; struct io_wq_work *hash_tail[IO_WQ_NR_HASH_BUCKETS]; }; @@ -120,17 +110,33 @@ struct io_wq { io_wq_work_fn *do_work; struct task_struct *manager; - struct user_struct *user; + + struct io_wq_hash *hash; + refcount_t refs; - struct completion done; + struct completion exited; + + atomic_t worker_refs; + struct completion worker_done; struct hlist_node cpuhp_node; - refcount_t use_refs; + pid_t task_pid; }; static enum cpuhp_state io_wq_online; +struct io_cb_cancel_data { + work_cancel_fn *fn; + void *data; + int nr_running; + int nr_pending; + bool cancel_all; +}; + +static void io_wqe_cancel_pending_work(struct io_wqe *wqe, + struct io_cb_cancel_data *match); + static bool io_worker_get(struct io_worker *worker) { return refcount_inc_not_zero(&worker->ref); @@ -139,62 +145,7 @@ static bool io_worker_get(struct io_worker *worker) static void io_worker_release(struct io_worker *worker) { if (refcount_dec_and_test(&worker->ref)) - wake_up_process(worker->task); -} - -/* - * Note: drops the wqe->lock if returning true! The caller must re-acquire - * the lock in that case. Some callers need to restart handling if this - * happens, so we can't just re-acquire the lock on behalf of the caller. - */ -static bool __io_worker_unuse(struct io_wqe *wqe, struct io_worker *worker) -{ - bool dropped_lock = false; - - if (worker->saved_creds) { - revert_creds(worker->saved_creds); - worker->cur_creds = worker->saved_creds = NULL; - } - - if (current->files != worker->restore_files) { - __acquire(&wqe->lock); - raw_spin_unlock_irq(&wqe->lock); - dropped_lock = true; - - task_lock(current); - current->files = worker->restore_files; - current->nsproxy = worker->restore_nsproxy; - task_unlock(current); - } - - if (current->fs != worker->restore_fs) - current->fs = worker->restore_fs; - - /* - * If we have an active mm, we need to drop the wq lock before unusing - * it. If we do, return true and let the caller retry the idle loop. - */ - if (worker->mm) { - if (!dropped_lock) { - __acquire(&wqe->lock); - raw_spin_unlock_irq(&wqe->lock); - dropped_lock = true; - } - __set_current_state(TASK_RUNNING); - kthread_unuse_mm(worker->mm); - mmput(worker->mm); - worker->mm = NULL; - } - -#ifdef CONFIG_BLK_CGROUP - if (worker->blkcg_css) { - kthread_associate_blkcg(NULL); - worker->blkcg_css = NULL; - } -#endif - if (current->signal->rlim[RLIMIT_FSIZE].rlim_cur != RLIM_INFINITY) - current->signal->rlim[RLIMIT_FSIZE].rlim_cur = RLIM_INFINITY; - return dropped_lock; + complete(&worker->ref_done); } static inline struct io_wqe_acct *io_work_get_acct(struct io_wqe *wqe, @@ -206,9 +157,10 @@ static inline struct io_wqe_acct *io_work_get_acct(struct io_wqe *wqe, return &wqe->acct[IO_WQ_ACCT_BOUND]; } -static inline struct io_wqe_acct *io_wqe_get_acct(struct io_wqe *wqe, - struct io_worker *worker) +static inline struct io_wqe_acct *io_wqe_get_acct(struct io_worker *worker) { + struct io_wqe *wqe = worker->wqe; + if (worker->flags & IO_WORKER_F_BOUND) return &wqe->acct[IO_WQ_ACCT_BOUND]; @@ -218,39 +170,33 @@ static inline struct io_wqe_acct *io_wqe_get_acct(struct io_wqe *wqe, static void io_worker_exit(struct io_worker *worker) { struct io_wqe *wqe = worker->wqe; - struct io_wqe_acct *acct = io_wqe_get_acct(wqe, worker); + struct io_wqe_acct *acct = io_wqe_get_acct(worker); + unsigned flags; - /* - * If we're not at zero, someone else is holding a brief reference - * to the worker. Wait for that to go away. - */ - set_current_state(TASK_INTERRUPTIBLE); - if (!refcount_dec_and_test(&worker->ref)) - schedule(); - __set_current_state(TASK_RUNNING); + if (refcount_dec_and_test(&worker->ref)) + complete(&worker->ref_done); + wait_for_completion(&worker->ref_done); preempt_disable(); current->flags &= ~PF_IO_WORKER; - if (worker->flags & IO_WORKER_F_RUNNING) + flags = worker->flags; + worker->flags = 0; + if (flags & IO_WORKER_F_RUNNING) atomic_dec(&acct->nr_running); - if (!(worker->flags & IO_WORKER_F_BOUND)) - atomic_dec(&wqe->wq->user->processes); worker->flags = 0; preempt_enable(); raw_spin_lock_irq(&wqe->lock); - hlist_nulls_del_rcu(&worker->nulls_node); + if (flags & IO_WORKER_F_FREE) + hlist_nulls_del_rcu(&worker->nulls_node); list_del_rcu(&worker->all_list); - if (__io_worker_unuse(wqe, worker)) { - __release(&wqe->lock); - raw_spin_lock_irq(&wqe->lock); - } acct->nr_workers--; raw_spin_unlock_irq(&wqe->lock); kfree_rcu(worker, rcu); - if (refcount_dec_and_test(&wqe->wq->refs)) - complete(&wqe->wq->done); + if (atomic_dec_and_test(&wqe->wq->worker_refs)) + complete(&wqe->wq->worker_done); + do_exit(0); } static inline bool io_wqe_run_queue(struct io_wqe *wqe) @@ -308,35 +254,23 @@ static void io_wqe_wake_worker(struct io_wqe *wqe, struct io_wqe_acct *acct) wake_up_process(wqe->wq->manager); } -static void io_wqe_inc_running(struct io_wqe *wqe, struct io_worker *worker) +static void io_wqe_inc_running(struct io_worker *worker) { - struct io_wqe_acct *acct = io_wqe_get_acct(wqe, worker); + struct io_wqe_acct *acct = io_wqe_get_acct(worker); atomic_inc(&acct->nr_running); } -static void io_wqe_dec_running(struct io_wqe *wqe, struct io_worker *worker) +static void io_wqe_dec_running(struct io_worker *worker) __must_hold(wqe->lock) { - struct io_wqe_acct *acct = io_wqe_get_acct(wqe, worker); + struct io_wqe_acct *acct = io_wqe_get_acct(worker); + struct io_wqe *wqe = worker->wqe; if (atomic_dec_and_test(&acct->nr_running) && io_wqe_run_queue(wqe)) io_wqe_wake_worker(wqe, acct); } -static void io_worker_start(struct io_wqe *wqe, struct io_worker *worker) -{ - allow_kernel_signal(SIGINT); - - current->flags |= PF_IO_WORKER; - - worker->flags |= (IO_WORKER_F_UP | IO_WORKER_F_RUNNING); - worker->restore_files = current->files; - worker->restore_nsproxy = current->nsproxy; - worker->restore_fs = current->fs; - io_wqe_inc_running(wqe, worker); -} - /* * Worker will start processing some work. Move it to the busy list, if * it's currently on the freelist @@ -359,19 +293,17 @@ static void __io_worker_busy(struct io_wqe *wqe, struct io_worker *worker, worker_bound = (worker->flags & IO_WORKER_F_BOUND) != 0; work_bound = (work->flags & IO_WQ_WORK_UNBOUND) == 0; if (worker_bound != work_bound) { - io_wqe_dec_running(wqe, worker); + io_wqe_dec_running(worker); if (work_bound) { worker->flags |= IO_WORKER_F_BOUND; wqe->acct[IO_WQ_ACCT_UNBOUND].nr_workers--; wqe->acct[IO_WQ_ACCT_BOUND].nr_workers++; - atomic_dec(&wqe->wq->user->processes); } else { worker->flags &= ~IO_WORKER_F_BOUND; wqe->acct[IO_WQ_ACCT_UNBOUND].nr_workers++; wqe->acct[IO_WQ_ACCT_BOUND].nr_workers--; - atomic_inc(&wqe->wq->user->processes); } - io_wqe_inc_running(wqe, worker); + io_wqe_inc_running(worker); } } @@ -382,15 +314,13 @@ static void __io_worker_busy(struct io_wqe *wqe, struct io_worker *worker, * retry the loop in that case (we changed task state), we don't regrab * the lock if we return success. */ -static bool __io_worker_idle(struct io_wqe *wqe, struct io_worker *worker) +static void __io_worker_idle(struct io_wqe *wqe, struct io_worker *worker) __must_hold(wqe->lock) { if (!(worker->flags & IO_WORKER_F_FREE)) { worker->flags |= IO_WORKER_F_FREE; hlist_nulls_add_head_rcu(&worker->nulls_node, &wqe->free_list); } - - return __io_worker_unuse(wqe, worker); } static inline unsigned int io_get_work_hash(struct io_wq_work *work) @@ -398,14 +328,31 @@ static inline unsigned int io_get_work_hash(struct io_wq_work *work) return work->flags >> IO_WQ_HASH_SHIFT; } +static void io_wait_on_hash(struct io_wqe *wqe, unsigned int hash) +{ + struct io_wq *wq = wqe->wq; + + spin_lock(&wq->hash->wait.lock); + if (list_empty(&wqe->wait.entry)) { + __add_wait_queue(&wq->hash->wait, &wqe->wait); + if (!test_bit(hash, &wq->hash->map)) { + __set_current_state(TASK_RUNNING); + list_del_init(&wqe->wait.entry); + } + } + spin_unlock(&wq->hash->wait.lock); +} + static struct io_wq_work *io_get_next_work(struct io_wqe *wqe) __must_hold(wqe->lock) { struct io_wq_work_node *node, *prev; struct io_wq_work *work, *tail; - unsigned int hash; + unsigned int stall_hash = -1U; wq_list_for_each(node, prev, &wqe->work_list) { + unsigned int hash; + work = container_of(node, struct io_wq_work, list); /* not hashed, can run anytime */ @@ -414,111 +361,51 @@ static struct io_wq_work *io_get_next_work(struct io_wqe *wqe) return work; } - /* hashed, can run if not already running */ hash = io_get_work_hash(work); - if (!(wqe->hash_map & BIT(hash))) { - wqe->hash_map |= BIT(hash); - /* all items with this hash lie in [work, tail] */ - tail = wqe->hash_tail[hash]; + /* all items with this hash lie in [work, tail] */ + tail = wqe->hash_tail[hash]; + + /* hashed, can run if not already running */ + if (!test_and_set_bit(hash, &wqe->wq->hash->map)) { wqe->hash_tail[hash] = NULL; wq_list_cut(&wqe->work_list, &tail->list, prev); return work; } + if (stall_hash == -1U) + stall_hash = hash; + /* fast forward to a next hash, for-each will fix up @prev */ + node = &tail->list; } - return NULL; -} - -static void io_wq_switch_mm(struct io_worker *worker, struct io_wq_work *work) -{ - if (worker->mm) { - kthread_unuse_mm(worker->mm); - mmput(worker->mm); - worker->mm = NULL; - } - - if (mmget_not_zero(work->identity->mm)) { - kthread_use_mm(work->identity->mm); - worker->mm = work->identity->mm; - return; + if (stall_hash != -1U) { + raw_spin_unlock(&wqe->lock); + io_wait_on_hash(wqe, stall_hash); + raw_spin_lock(&wqe->lock); } - /* failed grabbing mm, ensure work gets cancelled */ - work->flags |= IO_WQ_WORK_CANCEL; -} - -static inline void io_wq_switch_blkcg(struct io_worker *worker, - struct io_wq_work *work) -{ -#ifdef CONFIG_BLK_CGROUP - if (!(work->flags & IO_WQ_WORK_BLKCG)) - return; - if (work->identity->blkcg_css != worker->blkcg_css) { - kthread_associate_blkcg(work->identity->blkcg_css); - worker->blkcg_css = work->identity->blkcg_css; - } -#endif -} - -static void io_wq_switch_creds(struct io_worker *worker, - struct io_wq_work *work) -{ - const struct cred *old_creds = override_creds(work->identity->creds); - - worker->cur_creds = work->identity->creds; - if (worker->saved_creds) - put_cred(old_creds); /* creds set by previous switch */ - else - worker->saved_creds = old_creds; + return NULL; } -static void io_impersonate_work(struct io_worker *worker, - struct io_wq_work *work) +static bool io_flush_signals(void) { - if ((work->flags & IO_WQ_WORK_FILES) && - current->files != work->identity->files) { - task_lock(current); - current->files = work->identity->files; - current->nsproxy = work->identity->nsproxy; - task_unlock(current); - if (!work->identity->files) { - /* failed grabbing files, ensure work gets cancelled */ - work->flags |= IO_WQ_WORK_CANCEL; - } + if (unlikely(test_tsk_thread_flag(current, TIF_NOTIFY_SIGNAL))) { + __set_current_state(TASK_RUNNING); + if (current->task_works) + task_work_run(); + clear_tsk_thread_flag(current, TIF_NOTIFY_SIGNAL); + return true; } - if ((work->flags & IO_WQ_WORK_FS) && current->fs != work->identity->fs) - current->fs = work->identity->fs; - if ((work->flags & IO_WQ_WORK_MM) && work->identity->mm != worker->mm) - io_wq_switch_mm(worker, work); - if ((work->flags & IO_WQ_WORK_CREDS) && - worker->cur_creds != work->identity->creds) - io_wq_switch_creds(worker, work); - if (work->flags & IO_WQ_WORK_FSIZE) - current->signal->rlim[RLIMIT_FSIZE].rlim_cur = work->identity->fsize; - else if (current->signal->rlim[RLIMIT_FSIZE].rlim_cur != RLIM_INFINITY) - current->signal->rlim[RLIMIT_FSIZE].rlim_cur = RLIM_INFINITY; - io_wq_switch_blkcg(worker, work); -#ifdef CONFIG_AUDIT - current->loginuid = work->identity->loginuid; - current->sessionid = work->identity->sessionid; -#endif + return false; } static void io_assign_current_work(struct io_worker *worker, struct io_wq_work *work) { if (work) { - /* flush pending signals before assigning new work */ - if (signal_pending(current)) - flush_signals(current); + io_flush_signals(); cond_resched(); } -#ifdef CONFIG_AUDIT - current->loginuid = KUIDT_INIT(AUDIT_UID_UNSET); - current->sessionid = AUDIT_SID_UNSET; -#endif - spin_lock_irq(&worker->lock); worker->cur_work = work; spin_unlock_irq(&worker->lock); @@ -552,32 +439,32 @@ get_next: if (!work) break; io_assign_current_work(worker, work); + __set_current_state(TASK_RUNNING); /* handle a whole dependent link */ do { - struct io_wq_work *old_work, *next_hashed, *linked; + struct io_wq_work *next_hashed, *linked; unsigned int hash = io_get_work_hash(work); next_hashed = wq_next_work(work); - io_impersonate_work(worker, work); - - old_work = work; - linked = wq->do_work(work); + wq->do_work(work); + io_assign_current_work(worker, NULL); + linked = wq->free_work(work); work = next_hashed; if (!work && linked && !io_wq_is_hashed(linked)) { work = linked; linked = NULL; } io_assign_current_work(worker, work); - wq->free_work(old_work); - if (linked) io_wqe_enqueue(wqe, linked); if (hash != -1U && !next_hashed) { + clear_bit(hash, &wq->hash->map); + if (wq_has_sleeper(&wq->hash->wait)) + wake_up(&wq->hash->wait); raw_spin_lock_irq(&wqe->lock); - wqe->hash_map &= ~BIT_ULL(hash); wqe->flags &= ~IO_WQE_FLAG_STALLED; /* skip unnecessary unlock-lock wqe->lock */ if (!work) @@ -595,28 +482,33 @@ static int io_wqe_worker(void *data) struct io_worker *worker = data; struct io_wqe *wqe = worker->wqe; struct io_wq *wq = wqe->wq; + char buf[TASK_COMM_LEN]; - io_worker_start(wqe, worker); + worker->flags |= (IO_WORKER_F_UP | IO_WORKER_F_RUNNING); + io_wqe_inc_running(worker); + + sprintf(buf, "iou-wrk-%d", wq->task_pid); + set_task_comm(current, buf); while (!test_bit(IO_WQ_BIT_EXIT, &wq->state)) { + long ret; + set_current_state(TASK_INTERRUPTIBLE); loop: raw_spin_lock_irq(&wqe->lock); if (io_wqe_run_queue(wqe)) { - __set_current_state(TASK_RUNNING); io_worker_handle_work(worker); goto loop; } - /* drops the lock on success, retry */ - if (__io_worker_idle(wqe, worker)) { - __release(&wqe->lock); - goto loop; - } + __io_worker_idle(wqe, worker); raw_spin_unlock_irq(&wqe->lock); - if (signal_pending(current)) - flush_signals(current); - if (schedule_timeout(WORKER_IDLE_TIMEOUT)) + if (io_flush_signals()) continue; + ret = schedule_timeout(WORKER_IDLE_TIMEOUT); + if (try_to_freeze() || ret) + continue; + if (fatal_signal_pending(current)) + break; /* timed out, exit unless we're the fixed worker */ if (test_bit(IO_WQ_BIT_EXIT, &wq->state) || !(worker->flags & IO_WORKER_F_FIXED)) @@ -640,15 +532,16 @@ loop: */ void io_wq_worker_running(struct task_struct *tsk) { - struct io_worker *worker = kthread_data(tsk); - struct io_wqe *wqe = worker->wqe; + struct io_worker *worker = tsk->pf_io_worker; + if (!worker) + return; if (!(worker->flags & IO_WORKER_F_UP)) return; if (worker->flags & IO_WORKER_F_RUNNING) return; worker->flags |= IO_WORKER_F_RUNNING; - io_wqe_inc_running(wqe, worker); + io_wqe_inc_running(worker); } /* @@ -658,9 +551,10 @@ void io_wq_worker_running(struct task_struct *tsk) */ void io_wq_worker_sleeping(struct task_struct *tsk) { - struct io_worker *worker = kthread_data(tsk); - struct io_wqe *wqe = worker->wqe; + struct io_worker *worker = tsk->pf_io_worker; + if (!worker) + return; if (!(worker->flags & IO_WORKER_F_UP)) return; if (!(worker->flags & IO_WORKER_F_RUNNING)) @@ -668,15 +562,18 @@ void io_wq_worker_sleeping(struct task_struct *tsk) worker->flags &= ~IO_WORKER_F_RUNNING; - raw_spin_lock_irq(&wqe->lock); - io_wqe_dec_running(wqe, worker); - raw_spin_unlock_irq(&wqe->lock); + raw_spin_lock_irq(&worker->wqe->lock); + io_wqe_dec_running(worker); + raw_spin_unlock_irq(&worker->wqe->lock); } static bool create_io_worker(struct io_wq *wq, struct io_wqe *wqe, int index) { struct io_wqe_acct *acct = &wqe->acct[index]; struct io_worker *worker; + struct task_struct *tsk; + + __set_current_state(TASK_RUNNING); worker = kzalloc_node(sizeof(*worker), GFP_KERNEL, wqe->node); if (!worker) @@ -686,14 +583,22 @@ static bool create_io_worker(struct io_wq *wq, struct io_wqe *wqe, int index) worker->nulls_node.pprev = NULL; worker->wqe = wqe; spin_lock_init(&worker->lock); + init_completion(&worker->ref_done); + + atomic_inc(&wq->worker_refs); - worker->task = kthread_create_on_node(io_wqe_worker, worker, wqe->node, - "io_wqe_worker-%d/%d", index, wqe->node); - if (IS_ERR(worker->task)) { + tsk = create_io_thread(io_wqe_worker, worker, wqe->node); + if (IS_ERR(tsk)) { + if (atomic_dec_and_test(&wq->worker_refs)) + complete(&wq->worker_done); kfree(worker); return false; } - kthread_bind_mask(worker->task, cpumask_of_node(wqe->node)); + + tsk->pf_io_worker = worker; + worker->task = tsk; + set_cpus_allowed_ptr(tsk, cpumask_of_node(wqe->node)); + tsk->flags |= PF_NO_SETAFFINITY; raw_spin_lock_irq(&wqe->lock); hlist_nulls_add_head_rcu(&worker->nulls_node, &wqe->free_list); @@ -705,12 +610,7 @@ static bool create_io_worker(struct io_wq *wq, struct io_wqe *wqe, int index) worker->flags |= IO_WORKER_F_FIXED; acct->nr_workers++; raw_spin_unlock_irq(&wqe->lock); - - if (index == IO_WQ_ACCT_UNBOUND) - atomic_inc(&wq->user->processes); - - refcount_inc(&wq->refs); - wake_up_process(worker->task); + wake_up_new_task(tsk); return true; } @@ -719,6 +619,8 @@ static inline bool io_wqe_need_worker(struct io_wqe *wqe, int index) { struct io_wqe_acct *acct = &wqe->acct[index]; + if (acct->nr_workers && test_bit(IO_WQ_BIT_EXIT, &wqe->wq->state)) + return false; /* if we have available workers or no work, no need */ if (!hlist_nulls_empty(&wqe->free_list) || !io_wqe_run_queue(wqe)) return false; @@ -752,97 +654,92 @@ static bool io_wq_for_each_worker(struct io_wqe *wqe, static bool io_wq_worker_wake(struct io_worker *worker, void *data) { + set_notify_signal(worker->task); wake_up_process(worker->task); return false; } -/* - * Manager thread. Tasked with creating new workers, if we need them. - */ -static int io_wq_manager(void *data) +static void io_wq_check_workers(struct io_wq *wq) { - struct io_wq *wq = data; int node; - /* create fixed workers */ - refcount_set(&wq->refs, 1); for_each_node(node) { + struct io_wqe *wqe = wq->wqes[node]; + bool fork_worker[2] = { false, false }; + if (!node_online(node)) continue; - if (create_io_worker(wq, wq->wqes[node], IO_WQ_ACCT_BOUND)) - continue; - set_bit(IO_WQ_BIT_ERROR, &wq->state); - set_bit(IO_WQ_BIT_EXIT, &wq->state); - goto out; + + raw_spin_lock_irq(&wqe->lock); + if (io_wqe_need_worker(wqe, IO_WQ_ACCT_BOUND)) + fork_worker[IO_WQ_ACCT_BOUND] = true; + if (io_wqe_need_worker(wqe, IO_WQ_ACCT_UNBOUND)) + fork_worker[IO_WQ_ACCT_UNBOUND] = true; + raw_spin_unlock_irq(&wqe->lock); + if (fork_worker[IO_WQ_ACCT_BOUND]) + create_io_worker(wq, wqe, IO_WQ_ACCT_BOUND); + if (fork_worker[IO_WQ_ACCT_UNBOUND]) + create_io_worker(wq, wqe, IO_WQ_ACCT_UNBOUND); } +} - complete(&wq->done); +static bool io_wq_work_match_all(struct io_wq_work *work, void *data) +{ + return true; +} - while (!kthread_should_stop()) { - if (current->task_works) - task_work_run(); +static void io_wq_cancel_pending(struct io_wq *wq) +{ + struct io_cb_cancel_data match = { + .fn = io_wq_work_match_all, + .cancel_all = true, + }; + int node; + + for_each_node(node) + io_wqe_cancel_pending_work(wq->wqes[node], &match); +} - for_each_node(node) { - struct io_wqe *wqe = wq->wqes[node]; - bool fork_worker[2] = { false, false }; +/* + * Manager thread. Tasked with creating new workers, if we need them. + */ +static int io_wq_manager(void *data) +{ + struct io_wq *wq = data; + char buf[TASK_COMM_LEN]; + int node; - if (!node_online(node)) - continue; + sprintf(buf, "iou-mgr-%d", wq->task_pid); + set_task_comm(current, buf); - raw_spin_lock_irq(&wqe->lock); - if (io_wqe_need_worker(wqe, IO_WQ_ACCT_BOUND)) - fork_worker[IO_WQ_ACCT_BOUND] = true; - if (io_wqe_need_worker(wqe, IO_WQ_ACCT_UNBOUND)) - fork_worker[IO_WQ_ACCT_UNBOUND] = true; - raw_spin_unlock_irq(&wqe->lock); - if (fork_worker[IO_WQ_ACCT_BOUND]) - create_io_worker(wq, wqe, IO_WQ_ACCT_BOUND); - if (fork_worker[IO_WQ_ACCT_UNBOUND]) - create_io_worker(wq, wqe, IO_WQ_ACCT_UNBOUND); - } + do { set_current_state(TASK_INTERRUPTIBLE); + io_wq_check_workers(wq); schedule_timeout(HZ); - } - - if (current->task_works) - task_work_run(); - -out: - if (refcount_dec_and_test(&wq->refs)) { - complete(&wq->done); - return 0; - } - /* if ERROR is set and we get here, we have workers to wake */ - if (test_bit(IO_WQ_BIT_ERROR, &wq->state)) { - rcu_read_lock(); - for_each_node(node) - io_wq_for_each_worker(wq->wqes[node], io_wq_worker_wake, NULL); - rcu_read_unlock(); - } - return 0; -} - -static bool io_wq_can_queue(struct io_wqe *wqe, struct io_wqe_acct *acct, - struct io_wq_work *work) -{ - bool free_worker; + try_to_freeze(); + if (fatal_signal_pending(current)) + set_bit(IO_WQ_BIT_EXIT, &wq->state); + } while (!test_bit(IO_WQ_BIT_EXIT, &wq->state)); - if (!(work->flags & IO_WQ_WORK_UNBOUND)) - return true; - if (atomic_read(&acct->nr_running)) - return true; + io_wq_check_workers(wq); rcu_read_lock(); - free_worker = !hlist_nulls_empty(&wqe->free_list); + for_each_node(node) + io_wq_for_each_worker(wq->wqes[node], io_wq_worker_wake, NULL); rcu_read_unlock(); - if (free_worker) - return true; - if (atomic_read(&wqe->wq->user->processes) >= acct->max_workers && - !(capable(CAP_SYS_RESOURCE) || capable(CAP_SYS_ADMIN))) - return false; + if (atomic_dec_and_test(&wq->worker_refs)) + complete(&wq->worker_done); + wait_for_completion(&wq->worker_done); - return true; + spin_lock_irq(&wq->hash->wait.lock); + for_each_node(node) + list_del_init(&wq->wqes[node]->wait.entry); + spin_unlock_irq(&wq->hash->wait.lock); + + io_wq_cancel_pending(wq); + complete(&wq->exited); + do_exit(0); } static void io_run_cancel(struct io_wq_work *work, struct io_wqe *wqe) @@ -850,11 +747,9 @@ static void io_run_cancel(struct io_wq_work *work, struct io_wqe *wqe) struct io_wq *wq = wqe->wq; do { - struct io_wq_work *old_work = work; - work->flags |= IO_WQ_WORK_CANCEL; - work = wq->do_work(work); - wq->free_work(old_work); + wq->do_work(work); + work = wq->free_work(work); } while (work); } @@ -878,19 +773,39 @@ append: wq_list_add_after(&work->list, &tail->list, &wqe->work_list); } +static int io_wq_fork_manager(struct io_wq *wq) +{ + struct task_struct *tsk; + + if (wq->manager) + return 0; + + WARN_ON_ONCE(test_bit(IO_WQ_BIT_EXIT, &wq->state)); + + init_completion(&wq->worker_done); + atomic_set(&wq->worker_refs, 1); + tsk = create_io_thread(io_wq_manager, wq, NUMA_NO_NODE); + if (!IS_ERR(tsk)) { + wq->manager = get_task_struct(tsk); + wake_up_new_task(tsk); + return 0; + } + + if (atomic_dec_and_test(&wq->worker_refs)) + complete(&wq->worker_done); + + return PTR_ERR(tsk); +} + static void io_wqe_enqueue(struct io_wqe *wqe, struct io_wq_work *work) { struct io_wqe_acct *acct = io_work_get_acct(wqe, work); int work_flags; unsigned long flags; - /* - * Do early check to see if we need a new unbound worker, and if we do, - * if we're allowed to do so. This isn't 100% accurate as there's a - * gap between this check and incrementing the value, but that's OK. - * It's close enough to not be an issue, fork() has the same delay. - */ - if (unlikely(!io_wq_can_queue(wqe, acct, work))) { + /* Can only happen if manager creation fails after exec */ + if (io_wq_fork_manager(wqe->wq) || + test_bit(IO_WQ_BIT_EXIT, &wqe->wq->state)) { io_run_cancel(work, wqe); return; } @@ -925,14 +840,6 @@ void io_wq_hash_work(struct io_wq_work *work, void *val) work->flags |= (IO_WQ_WORK_HASHED | (bit << IO_WQ_HASH_SHIFT)); } -struct io_cb_cancel_data { - work_cancel_fn *fn; - void *data; - int nr_running; - int nr_pending; - bool cancel_all; -}; - static bool io_wq_worker_cancel(struct io_worker *worker, void *data) { struct io_cb_cancel_data *match = data; @@ -944,9 +851,8 @@ static bool io_wq_worker_cancel(struct io_worker *worker, void *data) */ spin_lock_irqsave(&worker->lock, flags); if (worker->cur_work && - !(worker->cur_work->flags & IO_WQ_WORK_NO_CANCEL) && match->fn(worker->cur_work, match->data)) { - send_sig(SIGINT, worker->task, 1); + set_notify_signal(worker->task); match->nr_running++; } spin_unlock_irqrestore(&worker->lock, flags); @@ -1050,6 +956,24 @@ enum io_wq_cancel io_wq_cancel_cb(struct io_wq *wq, work_cancel_fn *cancel, return IO_WQ_CANCEL_NOTFOUND; } +static int io_wqe_hash_wake(struct wait_queue_entry *wait, unsigned mode, + int sync, void *key) +{ + struct io_wqe *wqe = container_of(wait, struct io_wqe, wait); + int ret; + + list_del_init(&wait->entry); + + rcu_read_lock(); + ret = io_wqe_activate_free_worker(wqe); + rcu_read_unlock(); + + if (!ret) + wake_up_process(wqe->wq->manager); + + return 1; +} + struct io_wq *io_wq_create(unsigned bounded, struct io_wq_data *data) { int ret = -ENOMEM, node; @@ -1070,12 +994,11 @@ struct io_wq *io_wq_create(unsigned bounded, struct io_wq_data *data) if (ret) goto err_wqes; + refcount_inc(&data->hash->refs); + wq->hash = data->hash; wq->free_work = data->free_work; wq->do_work = data->do_work; - /* caller must already hold a reference to this */ - wq->user = data->user; - ret = -ENOMEM; for_each_node(node) { struct io_wqe *wqe; @@ -1090,11 +1013,11 @@ struct io_wq *io_wq_create(unsigned bounded, struct io_wq_data *data) wqe->node = alloc_node; wqe->acct[IO_WQ_ACCT_BOUND].max_workers = bounded; atomic_set(&wqe->acct[IO_WQ_ACCT_BOUND].nr_running, 0); - if (wq->user) { - wqe->acct[IO_WQ_ACCT_UNBOUND].max_workers = + wqe->acct[IO_WQ_ACCT_UNBOUND].max_workers = task_rlimit(current, RLIMIT_NPROC); - } atomic_set(&wqe->acct[IO_WQ_ACCT_UNBOUND].nr_running, 0); + wqe->wait.func = io_wqe_hash_wake; + INIT_LIST_HEAD(&wqe->wait.entry); wqe->wq = wq; raw_spin_lock_init(&wqe->lock); INIT_WQ_LIST(&wqe->work_list); @@ -1102,24 +1025,15 @@ struct io_wq *io_wq_create(unsigned bounded, struct io_wq_data *data) INIT_LIST_HEAD(&wqe->all_list); } - init_completion(&wq->done); + wq->task_pid = current->pid; + init_completion(&wq->exited); + refcount_set(&wq->refs, 1); - wq->manager = kthread_create(io_wq_manager, wq, "io_wq_manager"); - if (!IS_ERR(wq->manager)) { - wake_up_process(wq->manager); - wait_for_completion(&wq->done); - if (test_bit(IO_WQ_BIT_ERROR, &wq->state)) { - ret = -ENOMEM; - goto err; - } - refcount_set(&wq->use_refs, 1); - reinit_completion(&wq->done); + ret = io_wq_fork_manager(wq); + if (!ret) return wq; - } - - ret = PTR_ERR(wq->manager); - complete(&wq->done); err: + io_wq_put_hash(data->hash); cpuhp_state_remove_instance_nocalls(io_wq_online, &wq->cpuhp_node); for_each_node(node) kfree(wq->wqes[node]); @@ -1130,46 +1044,46 @@ err_wq: return ERR_PTR(ret); } -bool io_wq_get(struct io_wq *wq, struct io_wq_data *data) +static void io_wq_destroy_manager(struct io_wq *wq) { - if (data->free_work != wq->free_work || data->do_work != wq->do_work) - return false; - - return refcount_inc_not_zero(&wq->use_refs); + if (wq->manager) { + wake_up_process(wq->manager); + wait_for_completion(&wq->exited); + put_task_struct(wq->manager); + wq->manager = NULL; + } } -static void __io_wq_destroy(struct io_wq *wq) +static void io_wq_destroy(struct io_wq *wq) { int node; cpuhp_state_remove_instance_nocalls(io_wq_online, &wq->cpuhp_node); set_bit(IO_WQ_BIT_EXIT, &wq->state); - if (wq->manager) - kthread_stop(wq->manager); - - rcu_read_lock(); - for_each_node(node) - io_wq_for_each_worker(wq->wqes[node], io_wq_worker_wake, NULL); - rcu_read_unlock(); + io_wq_destroy_manager(wq); - wait_for_completion(&wq->done); - - for_each_node(node) - kfree(wq->wqes[node]); + for_each_node(node) { + struct io_wqe *wqe = wq->wqes[node]; + WARN_ON_ONCE(!wq_list_empty(&wqe->work_list)); + kfree(wqe); + } + io_wq_put_hash(wq->hash); kfree(wq->wqes); kfree(wq); } -void io_wq_destroy(struct io_wq *wq) +void io_wq_put(struct io_wq *wq) { - if (refcount_dec_and_test(&wq->use_refs)) - __io_wq_destroy(wq); + if (refcount_dec_and_test(&wq->refs)) + io_wq_destroy(wq); } -struct task_struct *io_wq_get_task(struct io_wq *wq) +void io_wq_put_and_exit(struct io_wq *wq) { - return wq->manager; + set_bit(IO_WQ_BIT_EXIT, &wq->state); + io_wq_destroy_manager(wq); + io_wq_put(wq); } static bool io_wq_worker_affinity(struct io_worker *worker, void *data) diff --git a/fs/io-wq.h b/fs/io-wq.h index b158f8addcf3..80d590564ff9 100644 --- a/fs/io-wq.h +++ b/fs/io-wq.h @@ -1,7 +1,7 @@ #ifndef INTERNAL_IO_WQ_H #define INTERNAL_IO_WQ_H -#include <linux/io_uring.h> +#include <linux/refcount.h> struct io_wq; @@ -9,16 +9,8 @@ enum { IO_WQ_WORK_CANCEL = 1, IO_WQ_WORK_HASHED = 2, IO_WQ_WORK_UNBOUND = 4, - IO_WQ_WORK_NO_CANCEL = 8, IO_WQ_WORK_CONCURRENT = 16, - IO_WQ_WORK_FILES = 32, - IO_WQ_WORK_FS = 64, - IO_WQ_WORK_MM = 128, - IO_WQ_WORK_CREDS = 256, - IO_WQ_WORK_BLKCG = 512, - IO_WQ_WORK_FSIZE = 1024, - IO_WQ_HASH_SHIFT = 24, /* upper 8 bits are used for hash key */ }; @@ -95,7 +87,7 @@ static inline void wq_list_del(struct io_wq_work_list *list, struct io_wq_work { struct io_wq_work_node list; - struct io_identity *identity; + const struct cred *creds; unsigned flags; }; @@ -107,19 +99,30 @@ static inline struct io_wq_work *wq_next_work(struct io_wq_work *work) return container_of(work->list.next, struct io_wq_work, list); } -typedef void (free_work_fn)(struct io_wq_work *); -typedef struct io_wq_work *(io_wq_work_fn)(struct io_wq_work *); +typedef struct io_wq_work *(free_work_fn)(struct io_wq_work *); +typedef void (io_wq_work_fn)(struct io_wq_work *); -struct io_wq_data { - struct user_struct *user; +struct io_wq_hash { + refcount_t refs; + unsigned long map; + struct wait_queue_head wait; +}; + +static inline void io_wq_put_hash(struct io_wq_hash *hash) +{ + if (refcount_dec_and_test(&hash->refs)) + kfree(hash); +} +struct io_wq_data { + struct io_wq_hash *hash; io_wq_work_fn *do_work; free_work_fn *free_work; }; struct io_wq *io_wq_create(unsigned bounded, struct io_wq_data *data); -bool io_wq_get(struct io_wq *wq, struct io_wq_data *data); -void io_wq_destroy(struct io_wq *wq); +void io_wq_put(struct io_wq *wq); +void io_wq_put_and_exit(struct io_wq *wq); void io_wq_enqueue(struct io_wq *wq, struct io_wq_work *work); void io_wq_hash_work(struct io_wq_work *work, void *val); @@ -134,8 +137,6 @@ typedef bool (work_cancel_fn)(struct io_wq_work *, void *); enum io_wq_cancel io_wq_cancel_cb(struct io_wq *wq, work_cancel_fn *cancel, void *data, bool cancel_all); -struct task_struct *io_wq_get_task(struct io_wq *wq); - #if defined(CONFIG_IO_WQ) extern void io_wq_worker_sleeping(struct task_struct *); extern void io_wq_worker_running(struct task_struct *); @@ -150,6 +151,7 @@ static inline void io_wq_worker_running(struct task_struct *tsk) static inline bool io_wq_current_is_worker(void) { - return in_task() && (current->flags & PF_IO_WORKER); + return in_task() && (current->flags & PF_IO_WORKER) && + current->pf_io_worker; } #endif diff --git a/fs/io_uring.c b/fs/io_uring.c index 931671082e61..543551d70327 100644 --- a/fs/io_uring.c +++ b/fs/io_uring.c @@ -57,7 +57,6 @@ #include <linux/mman.h> #include <linux/percpu.h> #include <linux/slab.h> -#include <linux/kthread.h> #include <linux/blkdev.h> #include <linux/bvec.h> #include <linux/net.h> @@ -75,13 +74,11 @@ #include <linux/fsnotify.h> #include <linux/fadvise.h> #include <linux/eventpoll.h> -#include <linux/fs_struct.h> #include <linux/splice.h> #include <linux/task_work.h> #include <linux/pagemap.h> #include <linux/io_uring.h> -#include <linux/blk-cgroup.h> -#include <linux/audit.h> +#include <linux/freezer.h> #define CREATE_TRACE_POINTS #include <trace/events/io_uring.h> @@ -104,6 +101,10 @@ #define IORING_MAX_RESTRICTIONS (IORING_RESTRICTION_LAST + \ IORING_REGISTER_LAST + IORING_OP_LAST) +#define SQE_VALID_FLAGS (IOSQE_FIXED_FILE|IOSQE_IO_DRAIN|IOSQE_IO_LINK| \ + IOSQE_IO_HARDLINK | IOSQE_ASYNC | \ + IOSQE_BUFFER_SELECT) + struct io_uring { u32 head ____cacheline_aligned_in_smp; u32 tail ____cacheline_aligned_in_smp; @@ -187,6 +188,11 @@ struct io_rings { struct io_uring_cqe cqes[] ____cacheline_aligned_in_smp; }; +enum io_uring_cmd_flags { + IO_URING_F_NONBLOCK = 1, + IO_URING_F_COMPLETE_DEFER = 2, +}; + struct io_mapped_ubuf { u64 ubuf; size_t len; @@ -195,28 +201,39 @@ struct io_mapped_ubuf { unsigned long acct_pages; }; -struct fixed_file_table { +struct io_ring_ctx; + +struct io_rsrc_put { + struct list_head list; + union { + void *rsrc; + struct file *file; + }; +}; + +struct fixed_rsrc_table { struct file **files; }; -struct fixed_file_ref_node { +struct fixed_rsrc_ref_node { struct percpu_ref refs; struct list_head node; - struct list_head file_list; - struct fixed_file_data *file_data; + struct list_head rsrc_list; + struct fixed_rsrc_data *rsrc_data; + void (*rsrc_put)(struct io_ring_ctx *ctx, + struct io_rsrc_put *prsrc); struct llist_node llist; bool done; }; -struct fixed_file_data { - struct fixed_file_table *table; +struct fixed_rsrc_data { + struct fixed_rsrc_table *table; struct io_ring_ctx *ctx; - struct fixed_file_ref_node *node; + struct fixed_rsrc_ref_node *node; struct percpu_ref refs; struct completion done; - struct list_head ref_list; - spinlock_t lock; + bool quiesce; }; struct io_buffer { @@ -234,19 +251,76 @@ struct io_restriction { bool registered; }; +enum { + IO_SQ_THREAD_SHOULD_STOP = 0, + IO_SQ_THREAD_SHOULD_PARK, +}; + struct io_sq_data { refcount_t refs; + atomic_t park_pending; struct mutex lock; /* ctx's that are using this sqd */ struct list_head ctx_list; - struct list_head ctx_new_list; - struct mutex ctx_lock; struct task_struct *thread; struct wait_queue_head wait; unsigned sq_thread_idle; + int sq_cpu; + pid_t task_pid; + pid_t task_tgid; + + unsigned long state; + struct completion exited; + struct callback_head *park_task_work; +}; + +#define IO_IOPOLL_BATCH 8 +#define IO_COMPL_BATCH 32 +#define IO_REQ_CACHE_SIZE 32 +#define IO_REQ_ALLOC_BATCH 8 + +struct io_comp_state { + struct io_kiocb *reqs[IO_COMPL_BATCH]; + unsigned int nr; + unsigned int locked_free_nr; + /* inline/task_work completion list, under ->uring_lock */ + struct list_head free_list; + /* IRQ completion list, under ->completion_lock */ + struct list_head locked_free_list; +}; + +struct io_submit_link { + struct io_kiocb *head; + struct io_kiocb *last; +}; + +struct io_submit_state { + struct blk_plug plug; + struct io_submit_link link; + + /* + * io_kiocb alloc cache + */ + void *reqs[IO_REQ_CACHE_SIZE]; + unsigned int free_reqs; + + bool plug_started; + + /* + * Batch completion logic + */ + struct io_comp_state comp; + + /* + * File reference cache + */ + struct file *file; + unsigned int fd; + unsigned int file_refs; + unsigned int ios_left; }; struct io_ring_ctx { @@ -257,12 +331,10 @@ struct io_ring_ctx { struct { unsigned int flags; unsigned int compat: 1; - unsigned int limit_mem: 1; unsigned int cq_overflow_flushed: 1; unsigned int drain_next: 1; unsigned int eventfd_async: 1; unsigned int restricted: 1; - unsigned int sqo_dead: 1; /* * Ring buffer of indices into array of io_uring_sqe, which is @@ -284,6 +356,9 @@ struct io_ring_ctx { unsigned cached_cq_overflow; unsigned long sq_check_overflow; + /* hashed buffered write serialization */ + struct io_wq_hash *hash_map; + struct list_head defer_list; struct list_head timeout_list; struct list_head cq_overflow_list; @@ -291,24 +366,19 @@ struct io_ring_ctx { struct io_uring_sqe *sq_sqes; } ____cacheline_aligned_in_smp; - struct io_rings *rings; + struct { + struct mutex uring_lock; + wait_queue_head_t wait; + } ____cacheline_aligned_in_smp; - /* IO offload */ - struct io_wq *io_wq; + struct io_submit_state submit_state; - /* - * For SQPOLL usage - we hold a reference to the parent task, so we - * have access to the ->files - */ - struct task_struct *sqo_task; + struct io_rings *rings; /* Only used for accounting purposes */ struct mm_struct *mm_account; -#ifdef CONFIG_BLK_CGROUP - struct cgroup_subsys_state *sqo_blkcg_css; -#endif - + const struct cred *sq_creds; /* cred used for __io_sq_thread() */ struct io_sq_data *sq_data; /* if using sq thread polling */ struct wait_queue_head sqo_sq_wait; @@ -319,7 +389,7 @@ struct io_ring_ctx { * readers must ensure that ->refs is alive as long as the file* is * used. Only updated through io_uring_register(2). */ - struct fixed_file_data *file_data; + struct fixed_rsrc_data *file_data; unsigned nr_user_files; /* if used, fixed mapped user buffers */ @@ -328,26 +398,16 @@ struct io_ring_ctx { struct user_struct *user; - const struct cred *creds; - -#ifdef CONFIG_AUDIT - kuid_t loginuid; - unsigned int sessionid; -#endif - struct completion ref_comp; - struct completion sq_thread_comp; - - /* if all else fails... */ - struct io_kiocb *fallback_req; #if defined(CONFIG_UNIX) struct socket *ring_sock; #endif - struct idr io_buffer_idr; + struct xarray io_buffers; - struct idr personality_idr; + struct xarray personalities; + u32 pers_next; struct { unsigned cached_cq_tail; @@ -362,11 +422,6 @@ struct io_ring_ctx { } ____cacheline_aligned_in_smp; struct { - struct mutex uring_lock; - wait_queue_head_t wait; - } ____cacheline_aligned_in_smp; - - struct { spinlock_t completion_lock; /* @@ -384,11 +439,37 @@ struct io_ring_ctx { struct list_head inflight_list; } ____cacheline_aligned_in_smp; - struct delayed_work file_put_work; - struct llist_head file_put_llist; + struct delayed_work rsrc_put_work; + struct llist_head rsrc_put_llist; + struct list_head rsrc_ref_list; + spinlock_t rsrc_ref_lock; - struct work_struct exit_work; struct io_restriction restrictions; + + /* exit task_work */ + struct callback_head *exit_task_work; + + struct wait_queue_head hash_wait; + + /* Keep this last, we don't need it for the fast path */ + struct work_struct exit_work; + struct list_head tctx_list; +}; + +struct io_uring_task { + /* submission side */ + struct xarray xa; + struct wait_queue_head wait; + const struct io_ring_ctx *last; + struct io_wq *io_wq; + struct percpu_counter inflight; + atomic_t in_idle; + bool sqpoll; + + spinlock_t task_lock; + struct io_wq_work_list task_list; + unsigned long task_state; + struct callback_head task_work; }; /* @@ -411,7 +492,6 @@ struct io_poll_remove { struct io_close { struct file *file; - struct file *put_file; int fd; }; @@ -489,13 +569,12 @@ struct io_sr_msg { struct io_open { struct file *file; int dfd; - bool ignore_nonblock; struct filename *filename; struct open_how how; unsigned long nofile; }; -struct io_files_update { +struct io_rsrc_update { struct file *file; u64 arg; u32 nr_args; @@ -584,7 +663,8 @@ struct io_async_connect { struct io_async_msghdr { struct iovec fast_iov[UIO_FASTIOV]; - struct iovec *iov; + /* points to an allocated iov, if NULL we use fast_iov instead */ + struct iovec *free_iov; struct sockaddr __user *uaddr; struct msghdr msg; struct sockaddr_storage addr; @@ -616,8 +696,8 @@ enum { REQ_F_POLLED_BIT, REQ_F_BUFFER_SELECTED_BIT, REQ_F_NO_FILE_TABLE_BIT, - REQ_F_WORK_INITIALIZED_BIT, REQ_F_LTIMEOUT_ACTIVE_BIT, + REQ_F_COMPLETE_INLINE_BIT, /* not a real bit, just to check we're not overflowing the space */ __REQ_F_LAST_BIT, @@ -639,7 +719,7 @@ enum { /* fail rest of links */ REQ_F_FAIL_LINK = BIT(REQ_F_FAIL_LINK_BIT), - /* on inflight list */ + /* on inflight list, should be cancelled and waited on exit reliably */ REQ_F_INFLIGHT = BIT(REQ_F_INFLIGHT_BIT), /* read/write uses file position */ REQ_F_CUR_POS = BIT(REQ_F_CUR_POS_BIT), @@ -657,10 +737,10 @@ enum { REQ_F_BUFFER_SELECTED = BIT(REQ_F_BUFFER_SELECTED_BIT), /* doesn't need file table for this request */ REQ_F_NO_FILE_TABLE = BIT(REQ_F_NO_FILE_TABLE_BIT), - /* io_wq_work is initialized */ - REQ_F_WORK_INITIALIZED = BIT(REQ_F_WORK_INITIALIZED_BIT), /* linked timeout is active, i.e. prepared by link's head */ REQ_F_LTIMEOUT_ACTIVE = BIT(REQ_F_LTIMEOUT_ACTIVE_BIT), + /* completion is deferred through io_comp_state */ + REQ_F_COMPLETE_INLINE = BIT(REQ_F_COMPLETE_INLINE_BIT), }; struct async_poll { @@ -668,6 +748,11 @@ struct async_poll { struct io_poll_iocb *double_poll; }; +struct io_task_work { + struct io_wq_work_node node; + task_work_func_t func; +}; + /* * NOTE! Each of the iocb union members has the file pointer * as the first entry in their struct definition. So you can @@ -689,7 +774,7 @@ struct io_kiocb { struct io_sr_msg sr_msg; struct io_open open; struct io_close close; - struct io_files_update files_update; + struct io_rsrc_update rsrc_update; struct io_fadvise fadvise; struct io_madvise madvise; struct io_epoll epoll; @@ -719,64 +804,38 @@ struct io_kiocb { u64 user_data; struct io_kiocb *link; - struct percpu_ref *fixed_file_refs; + struct percpu_ref *fixed_rsrc_refs; /* * 1. used with ctx->iopoll_list with reads/writes * 2. to track reqs with ->files (see io_op_def::file_table) */ struct list_head inflight_entry; - struct callback_head task_work; + union { + struct io_task_work io_task_work; + struct callback_head task_work; + }; /* for polled requests, i.e. IORING_OP_POLL_ADD and async armed poll */ struct hlist_node hash_node; struct async_poll *apoll; struct io_wq_work work; }; +struct io_tctx_node { + struct list_head ctx_node; + struct task_struct *task; + struct io_ring_ctx *ctx; +}; + struct io_defer_entry { struct list_head list; struct io_kiocb *req; u32 seq; }; -#define IO_IOPOLL_BATCH 8 - -struct io_comp_state { - unsigned int nr; - struct list_head list; - struct io_ring_ctx *ctx; -}; - -struct io_submit_state { - struct blk_plug plug; - - /* - * io_kiocb alloc cache - */ - void *reqs[IO_IOPOLL_BATCH]; - unsigned int free_reqs; - - bool plug_started; - - /* - * Batch completion logic - */ - struct io_comp_state comp; - - /* - * File reference cache - */ - struct file *file; - unsigned int fd; - unsigned int file_refs; - unsigned int ios_left; -}; - struct io_op_def { /* needs req->file assigned */ unsigned needs_file : 1; - /* don't fail if file grab fails */ - unsigned needs_file_no_error : 1; /* hash wq insertion if file is a regular file */ unsigned hash_reg_file : 1; /* unbound wq insertion if file is a non-regular file */ @@ -794,7 +853,6 @@ struct io_op_def { unsigned plug : 1; /* size of async data needed, if any */ unsigned short async_size; - unsigned work_flags; }; static const struct io_op_def io_op_defs[] = { @@ -807,7 +865,6 @@ static const struct io_op_def io_op_defs[] = { .needs_async_data = 1, .plug = 1, .async_size = sizeof(struct io_async_rw), - .work_flags = IO_WQ_WORK_MM | IO_WQ_WORK_BLKCG, }, [IORING_OP_WRITEV] = { .needs_file = 1, @@ -817,12 +874,9 @@ static const struct io_op_def io_op_defs[] = { .needs_async_data = 1, .plug = 1, .async_size = sizeof(struct io_async_rw), - .work_flags = IO_WQ_WORK_MM | IO_WQ_WORK_BLKCG | - IO_WQ_WORK_FSIZE, }, [IORING_OP_FSYNC] = { .needs_file = 1, - .work_flags = IO_WQ_WORK_BLKCG, }, [IORING_OP_READ_FIXED] = { .needs_file = 1, @@ -830,7 +884,6 @@ static const struct io_op_def io_op_defs[] = { .pollin = 1, .plug = 1, .async_size = sizeof(struct io_async_rw), - .work_flags = IO_WQ_WORK_BLKCG | IO_WQ_WORK_MM, }, [IORING_OP_WRITE_FIXED] = { .needs_file = 1, @@ -839,8 +892,6 @@ static const struct io_op_def io_op_defs[] = { .pollout = 1, .plug = 1, .async_size = sizeof(struct io_async_rw), - .work_flags = IO_WQ_WORK_BLKCG | IO_WQ_WORK_FSIZE | - IO_WQ_WORK_MM, }, [IORING_OP_POLL_ADD] = { .needs_file = 1, @@ -849,7 +900,6 @@ static const struct io_op_def io_op_defs[] = { [IORING_OP_POLL_REMOVE] = {}, [IORING_OP_SYNC_FILE_RANGE] = { .needs_file = 1, - .work_flags = IO_WQ_WORK_BLKCG, }, [IORING_OP_SENDMSG] = { .needs_file = 1, @@ -857,8 +907,6 @@ static const struct io_op_def io_op_defs[] = { .pollout = 1, .needs_async_data = 1, .async_size = sizeof(struct io_async_msghdr), - .work_flags = IO_WQ_WORK_MM | IO_WQ_WORK_BLKCG | - IO_WQ_WORK_FS, }, [IORING_OP_RECVMSG] = { .needs_file = 1, @@ -867,29 +915,23 @@ static const struct io_op_def io_op_defs[] = { .buffer_select = 1, .needs_async_data = 1, .async_size = sizeof(struct io_async_msghdr), - .work_flags = IO_WQ_WORK_MM | IO_WQ_WORK_BLKCG | - IO_WQ_WORK_FS, }, [IORING_OP_TIMEOUT] = { .needs_async_data = 1, .async_size = sizeof(struct io_timeout_data), - .work_flags = IO_WQ_WORK_MM, }, [IORING_OP_TIMEOUT_REMOVE] = { /* used by timeout updates' prep() */ - .work_flags = IO_WQ_WORK_MM, }, [IORING_OP_ACCEPT] = { .needs_file = 1, .unbound_nonreg_file = 1, .pollin = 1, - .work_flags = IO_WQ_WORK_MM | IO_WQ_WORK_FILES, }, [IORING_OP_ASYNC_CANCEL] = {}, [IORING_OP_LINK_TIMEOUT] = { .needs_async_data = 1, .async_size = sizeof(struct io_timeout_data), - .work_flags = IO_WQ_WORK_MM, }, [IORING_OP_CONNECT] = { .needs_file = 1, @@ -897,28 +939,14 @@ static const struct io_op_def io_op_defs[] = { .pollout = 1, .needs_async_data = 1, .async_size = sizeof(struct io_async_connect), - .work_flags = IO_WQ_WORK_MM, }, [IORING_OP_FALLOCATE] = { .needs_file = 1, - .work_flags = IO_WQ_WORK_BLKCG | IO_WQ_WORK_FSIZE, - }, - [IORING_OP_OPENAT] = { - .work_flags = IO_WQ_WORK_FILES | IO_WQ_WORK_BLKCG | - IO_WQ_WORK_FS | IO_WQ_WORK_MM, - }, - [IORING_OP_CLOSE] = { - .needs_file = 1, - .needs_file_no_error = 1, - .work_flags = IO_WQ_WORK_FILES | IO_WQ_WORK_BLKCG, - }, - [IORING_OP_FILES_UPDATE] = { - .work_flags = IO_WQ_WORK_FILES | IO_WQ_WORK_MM, - }, - [IORING_OP_STATX] = { - .work_flags = IO_WQ_WORK_FILES | IO_WQ_WORK_MM | - IO_WQ_WORK_FS | IO_WQ_WORK_BLKCG, }, + [IORING_OP_OPENAT] = {}, + [IORING_OP_CLOSE] = {}, + [IORING_OP_FILES_UPDATE] = {}, + [IORING_OP_STATX] = {}, [IORING_OP_READ] = { .needs_file = 1, .unbound_nonreg_file = 1, @@ -926,7 +954,6 @@ static const struct io_op_def io_op_defs[] = { .buffer_select = 1, .plug = 1, .async_size = sizeof(struct io_async_rw), - .work_flags = IO_WQ_WORK_MM | IO_WQ_WORK_BLKCG, }, [IORING_OP_WRITE] = { .needs_file = 1, @@ -934,42 +961,31 @@ static const struct io_op_def io_op_defs[] = { .pollout = 1, .plug = 1, .async_size = sizeof(struct io_async_rw), - .work_flags = IO_WQ_WORK_MM | IO_WQ_WORK_BLKCG | - IO_WQ_WORK_FSIZE, }, [IORING_OP_FADVISE] = { .needs_file = 1, - .work_flags = IO_WQ_WORK_BLKCG, - }, - [IORING_OP_MADVISE] = { - .work_flags = IO_WQ_WORK_MM | IO_WQ_WORK_BLKCG, }, + [IORING_OP_MADVISE] = {}, [IORING_OP_SEND] = { .needs_file = 1, .unbound_nonreg_file = 1, .pollout = 1, - .work_flags = IO_WQ_WORK_MM | IO_WQ_WORK_BLKCG, }, [IORING_OP_RECV] = { .needs_file = 1, .unbound_nonreg_file = 1, .pollin = 1, .buffer_select = 1, - .work_flags = IO_WQ_WORK_MM | IO_WQ_WORK_BLKCG, }, [IORING_OP_OPENAT2] = { - .work_flags = IO_WQ_WORK_FILES | IO_WQ_WORK_FS | - IO_WQ_WORK_BLKCG | IO_WQ_WORK_MM, }, [IORING_OP_EPOLL_CTL] = { .unbound_nonreg_file = 1, - .work_flags = IO_WQ_WORK_FILES, }, [IORING_OP_SPLICE] = { .needs_file = 1, .hash_reg_file = 1, .unbound_nonreg_file = 1, - .work_flags = IO_WQ_WORK_BLKCG, }, [IORING_OP_PROVIDE_BUFFERS] = {}, [IORING_OP_REMOVE_BUFFERS] = {}, @@ -981,54 +997,49 @@ static const struct io_op_def io_op_defs[] = { [IORING_OP_SHUTDOWN] = { .needs_file = 1, }, - [IORING_OP_RENAMEAT] = { - .work_flags = IO_WQ_WORK_MM | IO_WQ_WORK_FILES | - IO_WQ_WORK_FS | IO_WQ_WORK_BLKCG, - }, - [IORING_OP_UNLINKAT] = { - .work_flags = IO_WQ_WORK_MM | IO_WQ_WORK_FILES | - IO_WQ_WORK_FS | IO_WQ_WORK_BLKCG, - }, -}; - -enum io_mem_account { - ACCT_LOCKED, - ACCT_PINNED, + [IORING_OP_RENAMEAT] = {}, + [IORING_OP_UNLINKAT] = {}, }; -static void __io_uring_cancel_task_requests(struct io_ring_ctx *ctx, - struct task_struct *task); - -static void destroy_fixed_file_ref_node(struct fixed_file_ref_node *ref_node); -static struct fixed_file_ref_node *alloc_fixed_file_ref_node( +static bool io_disarm_next(struct io_kiocb *req); +static void io_uring_del_task_file(unsigned long index); +static void io_uring_try_cancel_requests(struct io_ring_ctx *ctx, + struct task_struct *task, + struct files_struct *files); +static void io_uring_cancel_sqpoll(struct io_ring_ctx *ctx); +static void destroy_fixed_rsrc_ref_node(struct fixed_rsrc_ref_node *ref_node); +static struct fixed_rsrc_ref_node *alloc_fixed_rsrc_ref_node( struct io_ring_ctx *ctx); +static void io_ring_file_put(struct io_ring_ctx *ctx, struct io_rsrc_put *prsrc); -static void __io_complete_rw(struct io_kiocb *req, long res, long res2, - struct io_comp_state *cs); +static bool io_rw_reissue(struct io_kiocb *req); static void io_cqring_fill_event(struct io_kiocb *req, long res); static void io_put_req(struct io_kiocb *req); static void io_put_req_deferred(struct io_kiocb *req, int nr); static void io_double_put_req(struct io_kiocb *req); +static void io_dismantle_req(struct io_kiocb *req); +static void io_put_task(struct task_struct *task, int nr); +static void io_queue_next(struct io_kiocb *req); static struct io_kiocb *io_prep_linked_timeout(struct io_kiocb *req); static void __io_queue_linked_timeout(struct io_kiocb *req); static void io_queue_linked_timeout(struct io_kiocb *req); static int __io_sqe_files_update(struct io_ring_ctx *ctx, - struct io_uring_files_update *ip, + struct io_uring_rsrc_update *ip, unsigned nr_args); static void __io_clean_op(struct io_kiocb *req); static struct file *io_file_get(struct io_submit_state *state, struct io_kiocb *req, int fd, bool fixed); -static void __io_queue_sqe(struct io_kiocb *req, struct io_comp_state *cs); -static void io_file_put_work(struct work_struct *work); +static void __io_queue_sqe(struct io_kiocb *req); +static void io_rsrc_put_work(struct work_struct *work); -static ssize_t io_import_iovec(int rw, struct io_kiocb *req, - struct iovec **iovec, struct iov_iter *iter, - bool needs_lock); +static int io_import_iovec(int rw, struct io_kiocb *req, struct iovec **iovec, + struct iov_iter *iter, bool needs_lock); static int io_setup_async_rw(struct io_kiocb *req, const struct iovec *iovec, const struct iovec *fast_iov, struct iov_iter *iter, bool force); -static void io_req_drop_files(struct io_kiocb *req); static void io_req_task_queue(struct io_kiocb *req); +static void io_submit_flush_completions(struct io_comp_state *cs, + struct io_ring_ctx *ctx); static struct kmem_cache *req_cachep; @@ -1060,9 +1071,9 @@ static inline void io_set_resource_node(struct io_kiocb *req) { struct io_ring_ctx *ctx = req->ctx; - if (!req->fixed_file_refs) { - req->fixed_file_refs = &ctx->file_data->node->refs; - percpu_ref_get(req->fixed_file_refs); + if (!req->fixed_rsrc_refs) { + req->fixed_rsrc_refs = &ctx->file_data->node->refs; + percpu_ref_get(req->fixed_rsrc_refs); } } @@ -1082,196 +1093,20 @@ static bool io_match_task(struct io_kiocb *head, return true; io_for_each_link(req, head) { - if (!(req->flags & REQ_F_WORK_INITIALIZED)) - continue; - if (req->file && req->file->f_op == &io_uring_fops) + if (req->flags & REQ_F_INFLIGHT) return true; - if ((req->work.flags & IO_WQ_WORK_FILES) && - req->work.identity->files == files) + if (req->task->files == files) return true; } return false; } -static void io_sq_thread_drop_mm_files(void) -{ - struct files_struct *files = current->files; - struct mm_struct *mm = current->mm; - - if (mm) { - kthread_unuse_mm(mm); - mmput(mm); - current->mm = NULL; - } - if (files) { - struct nsproxy *nsproxy = current->nsproxy; - - task_lock(current); - current->files = NULL; - current->nsproxy = NULL; - task_unlock(current); - put_files_struct(files); - put_nsproxy(nsproxy); - } -} - -static int __io_sq_thread_acquire_files(struct io_ring_ctx *ctx) -{ - if (current->flags & PF_EXITING) - return -EFAULT; - - if (!current->files) { - struct files_struct *files; - struct nsproxy *nsproxy; - - task_lock(ctx->sqo_task); - files = ctx->sqo_task->files; - if (!files) { - task_unlock(ctx->sqo_task); - return -EOWNERDEAD; - } - atomic_inc(&files->count); - get_nsproxy(ctx->sqo_task->nsproxy); - nsproxy = ctx->sqo_task->nsproxy; - task_unlock(ctx->sqo_task); - - task_lock(current); - current->files = files; - current->nsproxy = nsproxy; - task_unlock(current); - } - return 0; -} - -static int __io_sq_thread_acquire_mm(struct io_ring_ctx *ctx) -{ - struct mm_struct *mm; - - if (current->flags & PF_EXITING) - return -EFAULT; - if (current->mm) - return 0; - - /* Should never happen */ - if (unlikely(!(ctx->flags & IORING_SETUP_SQPOLL))) - return -EFAULT; - - task_lock(ctx->sqo_task); - mm = ctx->sqo_task->mm; - if (unlikely(!mm || !mmget_not_zero(mm))) - mm = NULL; - task_unlock(ctx->sqo_task); - - if (mm) { - kthread_use_mm(mm); - return 0; - } - - return -EFAULT; -} - -static int io_sq_thread_acquire_mm_files(struct io_ring_ctx *ctx, - struct io_kiocb *req) -{ - const struct io_op_def *def = &io_op_defs[req->opcode]; - int ret; - - if (def->work_flags & IO_WQ_WORK_MM) { - ret = __io_sq_thread_acquire_mm(ctx); - if (unlikely(ret)) - return ret; - } - - if (def->needs_file || (def->work_flags & IO_WQ_WORK_FILES)) { - ret = __io_sq_thread_acquire_files(ctx); - if (unlikely(ret)) - return ret; - } - - return 0; -} - -static void io_sq_thread_associate_blkcg(struct io_ring_ctx *ctx, - struct cgroup_subsys_state **cur_css) - -{ -#ifdef CONFIG_BLK_CGROUP - /* puts the old one when swapping */ - if (*cur_css != ctx->sqo_blkcg_css) { - kthread_associate_blkcg(ctx->sqo_blkcg_css); - *cur_css = ctx->sqo_blkcg_css; - } -#endif -} - -static void io_sq_thread_unassociate_blkcg(void) -{ -#ifdef CONFIG_BLK_CGROUP - kthread_associate_blkcg(NULL); -#endif -} - static inline void req_set_fail_links(struct io_kiocb *req) { if ((req->flags & (REQ_F_LINK | REQ_F_HARDLINK)) == REQ_F_LINK) req->flags |= REQ_F_FAIL_LINK; } -/* - * None of these are dereferenced, they are simply used to check if any of - * them have changed. If we're under current and check they are still the - * same, we're fine to grab references to them for actual out-of-line use. - */ -static void io_init_identity(struct io_identity *id) -{ - id->files = current->files; - id->mm = current->mm; -#ifdef CONFIG_BLK_CGROUP - rcu_read_lock(); - id->blkcg_css = blkcg_css(); - rcu_read_unlock(); -#endif - id->creds = current_cred(); - id->nsproxy = current->nsproxy; - id->fs = current->fs; - id->fsize = rlimit(RLIMIT_FSIZE); -#ifdef CONFIG_AUDIT - id->loginuid = current->loginuid; - id->sessionid = current->sessionid; -#endif - refcount_set(&id->count, 1); -} - -static inline void __io_req_init_async(struct io_kiocb *req) -{ - memset(&req->work, 0, sizeof(req->work)); - req->flags |= REQ_F_WORK_INITIALIZED; -} - -/* - * Note: must call io_req_init_async() for the first time you - * touch any members of io_wq_work. - */ -static inline void io_req_init_async(struct io_kiocb *req) -{ - struct io_uring_task *tctx = current->io_uring; - - if (req->flags & REQ_F_WORK_INITIALIZED) - return; - - __io_req_init_async(req); - - /* Grab a ref if this isn't our static identity */ - req->work.identity = tctx->identity; - if (tctx->identity != &tctx->__identity) - refcount_inc(&req->work.identity->count); -} - -static inline bool io_async_submit(struct io_ring_ctx *ctx) -{ - return ctx->flags & IORING_SETUP_SQPOLL; -} - static void io_ring_ctx_ref_free(struct percpu_ref *ref) { struct io_ring_ctx *ctx = container_of(ref, struct io_ring_ctx, refs); @@ -1293,10 +1128,6 @@ static struct io_ring_ctx *io_ring_ctx_alloc(struct io_uring_params *p) if (!ctx) return NULL; - ctx->fallback_req = kmem_cache_alloc(req_cachep, GFP_KERNEL); - if (!ctx->fallback_req) - goto err; - /* * Use 5 bits less than the max cq entries, that should give us around * 32 entries per hash list if totally full and uniformly spread. @@ -1322,9 +1153,8 @@ static struct io_ring_ctx *io_ring_ctx_alloc(struct io_uring_params *p) init_waitqueue_head(&ctx->cq_wait); INIT_LIST_HEAD(&ctx->cq_overflow_list); init_completion(&ctx->ref_comp); - init_completion(&ctx->sq_thread_comp); - idr_init(&ctx->io_buffer_idr); - idr_init(&ctx->personality_idr); + xa_init_flags(&ctx->io_buffers, XA_FLAGS_ALLOC1); + xa_init_flags(&ctx->personalities, XA_FLAGS_ALLOC1); mutex_init(&ctx->uring_lock); init_waitqueue_head(&ctx->wait); spin_lock_init(&ctx->completion_lock); @@ -1333,12 +1163,15 @@ static struct io_ring_ctx *io_ring_ctx_alloc(struct io_uring_params *p) INIT_LIST_HEAD(&ctx->timeout_list); spin_lock_init(&ctx->inflight_lock); INIT_LIST_HEAD(&ctx->inflight_list); - INIT_DELAYED_WORK(&ctx->file_put_work, io_file_put_work); - init_llist_head(&ctx->file_put_llist); + spin_lock_init(&ctx->rsrc_ref_lock); + INIT_LIST_HEAD(&ctx->rsrc_ref_list); + INIT_DELAYED_WORK(&ctx->rsrc_put_work, io_rsrc_put_work); + init_llist_head(&ctx->rsrc_put_llist); + INIT_LIST_HEAD(&ctx->tctx_list); + INIT_LIST_HEAD(&ctx->submit_state.comp.free_list); + INIT_LIST_HEAD(&ctx->submit_state.comp.locked_free_list); return ctx; err: - if (ctx->fallback_req) - kmem_cache_free(req_cachep, ctx->fallback_req); kfree(ctx->cancel_hash); kfree(ctx); return NULL; @@ -1356,184 +1189,17 @@ static bool req_need_defer(struct io_kiocb *req, u32 seq) return false; } -static void __io_commit_cqring(struct io_ring_ctx *ctx) -{ - struct io_rings *rings = ctx->rings; - - /* order cqe stores with ring update */ - smp_store_release(&rings->cq.tail, ctx->cached_cq_tail); -} - -static void io_put_identity(struct io_uring_task *tctx, struct io_kiocb *req) -{ - if (req->work.identity == &tctx->__identity) - return; - if (refcount_dec_and_test(&req->work.identity->count)) - kfree(req->work.identity); -} - -static void io_req_clean_work(struct io_kiocb *req) +static void io_req_track_inflight(struct io_kiocb *req) { - if (!(req->flags & REQ_F_WORK_INITIALIZED)) - return; - - req->flags &= ~REQ_F_WORK_INITIALIZED; - - if (req->work.flags & IO_WQ_WORK_MM) { - mmdrop(req->work.identity->mm); - req->work.flags &= ~IO_WQ_WORK_MM; - } -#ifdef CONFIG_BLK_CGROUP - if (req->work.flags & IO_WQ_WORK_BLKCG) { - css_put(req->work.identity->blkcg_css); - req->work.flags &= ~IO_WQ_WORK_BLKCG; - } -#endif - if (req->work.flags & IO_WQ_WORK_CREDS) { - put_cred(req->work.identity->creds); - req->work.flags &= ~IO_WQ_WORK_CREDS; - } - if (req->work.flags & IO_WQ_WORK_FS) { - struct fs_struct *fs = req->work.identity->fs; - - spin_lock(&req->work.identity->fs->lock); - if (--fs->users) - fs = NULL; - spin_unlock(&req->work.identity->fs->lock); - if (fs) - free_fs_struct(fs); - req->work.flags &= ~IO_WQ_WORK_FS; - } - if (req->flags & REQ_F_INFLIGHT) - io_req_drop_files(req); - - io_put_identity(req->task->io_uring, req); -} - -/* - * Create a private copy of io_identity, since some fields don't match - * the current context. - */ -static bool io_identity_cow(struct io_kiocb *req) -{ - struct io_uring_task *tctx = current->io_uring; - const struct cred *creds = NULL; - struct io_identity *id; - - if (req->work.flags & IO_WQ_WORK_CREDS) - creds = req->work.identity->creds; - - id = kmemdup(req->work.identity, sizeof(*id), GFP_KERNEL); - if (unlikely(!id)) { - req->work.flags |= IO_WQ_WORK_CANCEL; - return false; - } - - /* - * We can safely just re-init the creds we copied Either the field - * matches the current one, or we haven't grabbed it yet. The only - * exception is ->creds, through registered personalities, so handle - * that one separately. - */ - io_init_identity(id); - if (creds) - id->creds = creds; - - /* add one for this request */ - refcount_inc(&id->count); - - /* drop tctx and req identity references, if needed */ - if (tctx->identity != &tctx->__identity && - refcount_dec_and_test(&tctx->identity->count)) - kfree(tctx->identity); - if (req->work.identity != &tctx->__identity && - refcount_dec_and_test(&req->work.identity->count)) - kfree(req->work.identity); - - req->work.identity = id; - tctx->identity = id; - return true; -} - -static bool io_grab_identity(struct io_kiocb *req) -{ - const struct io_op_def *def = &io_op_defs[req->opcode]; - struct io_identity *id = req->work.identity; struct io_ring_ctx *ctx = req->ctx; - if (def->work_flags & IO_WQ_WORK_FSIZE) { - if (id->fsize != rlimit(RLIMIT_FSIZE)) - return false; - req->work.flags |= IO_WQ_WORK_FSIZE; - } -#ifdef CONFIG_BLK_CGROUP - if (!(req->work.flags & IO_WQ_WORK_BLKCG) && - (def->work_flags & IO_WQ_WORK_BLKCG)) { - rcu_read_lock(); - if (id->blkcg_css != blkcg_css()) { - rcu_read_unlock(); - return false; - } - /* - * This should be rare, either the cgroup is dying or the task - * is moving cgroups. Just punt to root for the handful of ios. - */ - if (css_tryget_online(id->blkcg_css)) - req->work.flags |= IO_WQ_WORK_BLKCG; - rcu_read_unlock(); - } -#endif - if (!(req->work.flags & IO_WQ_WORK_CREDS)) { - if (id->creds != current_cred()) - return false; - get_cred(id->creds); - req->work.flags |= IO_WQ_WORK_CREDS; - } -#ifdef CONFIG_AUDIT - if (!uid_eq(current->loginuid, id->loginuid) || - current->sessionid != id->sessionid) - return false; -#endif - if (!(req->work.flags & IO_WQ_WORK_FS) && - (def->work_flags & IO_WQ_WORK_FS)) { - if (current->fs != id->fs) - return false; - spin_lock(&id->fs->lock); - if (!id->fs->in_exec) { - id->fs->users++; - req->work.flags |= IO_WQ_WORK_FS; - } else { - req->work.flags |= IO_WQ_WORK_CANCEL; - } - spin_unlock(¤t->fs->lock); - } - if (!(req->work.flags & IO_WQ_WORK_FILES) && - (def->work_flags & IO_WQ_WORK_FILES) && - !(req->flags & REQ_F_NO_FILE_TABLE)) { - if (id->files != current->files || - id->nsproxy != current->nsproxy) - return false; - atomic_inc(&id->files->count); - get_nsproxy(id->nsproxy); - - if (!(req->flags & REQ_F_INFLIGHT)) { - req->flags |= REQ_F_INFLIGHT; - - spin_lock_irq(&ctx->inflight_lock); - list_add(&req->inflight_entry, &ctx->inflight_list); - spin_unlock_irq(&ctx->inflight_lock); - } - req->work.flags |= IO_WQ_WORK_FILES; - } - if (!(req->work.flags & IO_WQ_WORK_MM) && - (def->work_flags & IO_WQ_WORK_MM)) { - if (id->mm != current->mm) - return false; - mmgrab(id->mm); - req->work.flags |= IO_WQ_WORK_MM; - } + if (!(req->flags & REQ_F_INFLIGHT)) { + req->flags |= REQ_F_INFLIGHT; - return true; + spin_lock_irq(&ctx->inflight_lock); + list_add(&req->inflight_entry, &ctx->inflight_list); + spin_unlock_irq(&ctx->inflight_lock); + } } static void io_prep_async_work(struct io_kiocb *req) @@ -1541,7 +1207,8 @@ static void io_prep_async_work(struct io_kiocb *req) const struct io_op_def *def = &io_op_defs[req->opcode]; struct io_ring_ctx *ctx = req->ctx; - io_req_init_async(req); + if (!req->work.creds) + req->work.creds = get_current_cred(); if (req->flags & REQ_F_FORCE_ASYNC) req->work.flags |= IO_WQ_WORK_CONCURRENT; @@ -1553,17 +1220,6 @@ static void io_prep_async_work(struct io_kiocb *req) if (def->unbound_nonreg_file) req->work.flags |= IO_WQ_WORK_UNBOUND; } - - /* if we fail grabbing identity, we must COW, regrab, and retry */ - if (io_grab_identity(req)) - return; - - if (!io_identity_cow(req)) - return; - - /* can't fail at this point */ - if (!io_grab_identity(req)) - WARN_ON(1); } static void io_prep_async_link(struct io_kiocb *req) @@ -1574,25 +1230,20 @@ static void io_prep_async_link(struct io_kiocb *req) io_prep_async_work(cur); } -static struct io_kiocb *__io_queue_async_work(struct io_kiocb *req) +static void io_queue_async_work(struct io_kiocb *req) { struct io_ring_ctx *ctx = req->ctx; struct io_kiocb *link = io_prep_linked_timeout(req); + struct io_uring_task *tctx = req->task->io_uring; + + BUG_ON(!tctx); + BUG_ON(!tctx->io_wq); trace_io_uring_queue_async_work(ctx, io_wq_is_hashed(&req->work), req, &req->work, req->flags); - io_wq_enqueue(ctx->io_wq, &req->work); - return link; -} - -static void io_queue_async_work(struct io_kiocb *req) -{ - struct io_kiocb *link; - /* init ->work of the whole link before punting */ io_prep_async_link(req); - link = __io_queue_async_work(req); - + io_wq_enqueue(tctx->io_wq, &req->work); if (link) io_queue_linked_timeout(link); } @@ -1685,7 +1336,9 @@ static void io_flush_timeouts(struct io_ring_ctx *ctx) static void io_commit_cqring(struct io_ring_ctx *ctx) { io_flush_timeouts(ctx); - __io_commit_cqring(ctx); + + /* order cqe stores with ring update */ + smp_store_release(&ctx->rings->cq.tail, ctx->cached_cq_tail); if (unlikely(!list_empty(&ctx->defer_list))) __io_queue_deferred(ctx); @@ -1698,21 +1351,25 @@ static inline bool io_sqring_full(struct io_ring_ctx *ctx) return READ_ONCE(r->sq.tail) - ctx->cached_sq_head == r->sq_ring_entries; } +static inline unsigned int __io_cqring_events(struct io_ring_ctx *ctx) +{ + return ctx->cached_cq_tail - READ_ONCE(ctx->rings->cq.head); +} + static struct io_uring_cqe *io_get_cqring(struct io_ring_ctx *ctx) { struct io_rings *rings = ctx->rings; unsigned tail; - tail = ctx->cached_cq_tail; /* * writes to the cq entry need to come after reading head; the * control dependency is enough as we're using WRITE_ONCE to * fill the cq entry */ - if (tail - READ_ONCE(rings->cq.head) == rings->cq_ring_entries) + if (__io_cqring_events(ctx) == rings->cq_ring_entries) return NULL; - ctx->cached_cq_tail++; + tail = ctx->cached_cq_tail++; return &rings->cqes[tail & ctx->cq_mask]; } @@ -1727,11 +1384,6 @@ static inline bool io_should_trigger_evfd(struct io_ring_ctx *ctx) return io_wq_current_is_worker(); } -static inline unsigned __io_cqring_events(struct io_ring_ctx *ctx) -{ - return ctx->cached_cq_tail - READ_ONCE(ctx->rings->cq.head); -} - static void io_cqring_ev_posted(struct io_ring_ctx *ctx) { /* see waitqueue_active() comment */ @@ -1826,18 +1478,22 @@ static bool __io_cqring_overflow_flush(struct io_ring_ctx *ctx, bool force, return all_flushed; } -static void io_cqring_overflow_flush(struct io_ring_ctx *ctx, bool force, +static bool io_cqring_overflow_flush(struct io_ring_ctx *ctx, bool force, struct task_struct *tsk, struct files_struct *files) { + bool ret = true; + if (test_bit(0, &ctx->cq_check_overflow)) { /* iopoll syncs against uring_lock, not completion_lock */ if (ctx->flags & IORING_SETUP_IOPOLL) mutex_lock(&ctx->uring_lock); - __io_cqring_overflow_flush(ctx, force, tsk, files); + ret = __io_cqring_overflow_flush(ctx, force, tsk, files); if (ctx->flags & IORING_SETUP_IOPOLL) mutex_unlock(&ctx->uring_lock); } + + return ret; } static void __io_cqring_fill_event(struct io_kiocb *req, long res, long cflags) @@ -1885,100 +1541,114 @@ static void io_cqring_fill_event(struct io_kiocb *req, long res) __io_cqring_fill_event(req, res, 0); } -static void io_cqring_add_event(struct io_kiocb *req, long res, long cflags) +static void io_req_complete_post(struct io_kiocb *req, long res, + unsigned int cflags) { struct io_ring_ctx *ctx = req->ctx; unsigned long flags; spin_lock_irqsave(&ctx->completion_lock, flags); __io_cqring_fill_event(req, res, cflags); - io_commit_cqring(ctx); - spin_unlock_irqrestore(&ctx->completion_lock, flags); - - io_cqring_ev_posted(ctx); -} - -static void io_submit_flush_completions(struct io_comp_state *cs) -{ - struct io_ring_ctx *ctx = cs->ctx; - - spin_lock_irq(&ctx->completion_lock); - while (!list_empty(&cs->list)) { - struct io_kiocb *req; - - req = list_first_entry(&cs->list, struct io_kiocb, compl.list); - list_del(&req->compl.list); - __io_cqring_fill_event(req, req->result, req->compl.cflags); + /* + * If we're the last reference to this request, add to our locked + * free_list cache. + */ + if (refcount_dec_and_test(&req->refs)) { + struct io_comp_state *cs = &ctx->submit_state.comp; - /* - * io_free_req() doesn't care about completion_lock unless one - * of these flags is set. REQ_F_WORK_INITIALIZED is in the list - * because of a potential deadlock with req->work.fs->lock - */ - if (req->flags & (REQ_F_FAIL_LINK|REQ_F_LINK_TIMEOUT - |REQ_F_WORK_INITIALIZED)) { - spin_unlock_irq(&ctx->completion_lock); - io_put_req(req); - spin_lock_irq(&ctx->completion_lock); - } else { - io_put_req(req); + if (req->flags & (REQ_F_LINK | REQ_F_HARDLINK)) { + if (req->flags & (REQ_F_LINK_TIMEOUT | REQ_F_FAIL_LINK)) + io_disarm_next(req); + if (req->link) { + io_req_task_queue(req->link); + req->link = NULL; + } } + io_dismantle_req(req); + io_put_task(req->task, 1); + list_add(&req->compl.list, &cs->locked_free_list); + cs->locked_free_nr++; + } else { + if (!percpu_ref_tryget(&ctx->refs)) + req = NULL; } io_commit_cqring(ctx); - spin_unlock_irq(&ctx->completion_lock); + spin_unlock_irqrestore(&ctx->completion_lock, flags); - io_cqring_ev_posted(ctx); - cs->nr = 0; + if (req) { + io_cqring_ev_posted(ctx); + percpu_ref_put(&ctx->refs); + } } -static void __io_req_complete(struct io_kiocb *req, long res, unsigned cflags, - struct io_comp_state *cs) +static void io_req_complete_state(struct io_kiocb *req, long res, + unsigned int cflags) { - if (!cs) { - io_cqring_add_event(req, res, cflags); - io_put_req(req); - } else { - io_clean_op(req); - req->result = res; - req->compl.cflags = cflags; - list_add_tail(&req->compl.list, &cs->list); - if (++cs->nr >= 32) - io_submit_flush_completions(cs); - } + io_clean_op(req); + req->result = res; + req->compl.cflags = cflags; + req->flags |= REQ_F_COMPLETE_INLINE; } -static void io_req_complete(struct io_kiocb *req, long res) +static inline void __io_req_complete(struct io_kiocb *req, unsigned issue_flags, + long res, unsigned cflags) { - __io_req_complete(req, res, 0, NULL); + if (issue_flags & IO_URING_F_COMPLETE_DEFER) + io_req_complete_state(req, res, cflags); + else + io_req_complete_post(req, res, cflags); } -static inline bool io_is_fallback_req(struct io_kiocb *req) +static inline void io_req_complete(struct io_kiocb *req, long res) { - return req == (struct io_kiocb *) - ((unsigned long) req->ctx->fallback_req & ~1UL); + __io_req_complete(req, 0, res, 0); } -static struct io_kiocb *io_get_fallback_req(struct io_ring_ctx *ctx) +static bool io_flush_cached_reqs(struct io_ring_ctx *ctx) { - struct io_kiocb *req; + struct io_submit_state *state = &ctx->submit_state; + struct io_comp_state *cs = &state->comp; + struct io_kiocb *req = NULL; - req = ctx->fallback_req; - if (!test_and_set_bit_lock(0, (unsigned long *) &ctx->fallback_req)) - return req; + /* + * If we have more than a batch's worth of requests in our IRQ side + * locked cache, grab the lock and move them over to our submission + * side cache. + */ + if (READ_ONCE(cs->locked_free_nr) > IO_COMPL_BATCH) { + spin_lock_irq(&ctx->completion_lock); + list_splice_init(&cs->locked_free_list, &cs->free_list); + cs->locked_free_nr = 0; + spin_unlock_irq(&ctx->completion_lock); + } - return NULL; + while (!list_empty(&cs->free_list)) { + req = list_first_entry(&cs->free_list, struct io_kiocb, + compl.list); + list_del(&req->compl.list); + state->reqs[state->free_reqs++] = req; + if (state->free_reqs == ARRAY_SIZE(state->reqs)) + break; + } + + return req != NULL; } -static struct io_kiocb *io_alloc_req(struct io_ring_ctx *ctx, - struct io_submit_state *state) +static struct io_kiocb *io_alloc_req(struct io_ring_ctx *ctx) { + struct io_submit_state *state = &ctx->submit_state; + + BUILD_BUG_ON(IO_REQ_ALLOC_BATCH > ARRAY_SIZE(state->reqs)); + if (!state->free_reqs) { gfp_t gfp = GFP_KERNEL | __GFP_NOWARN; - size_t sz; int ret; - sz = min_t(size_t, state->ios_left, ARRAY_SIZE(state->reqs)); - ret = kmem_cache_alloc_bulk(req_cachep, gfp, sz, state->reqs); + if (io_flush_cached_reqs(ctx)) + goto got_req; + + ret = kmem_cache_alloc_bulk(req_cachep, gfp, IO_REQ_ALLOC_BATCH, + state->reqs); /* * Bulk alloc is all-or-nothing. If we fail to get a batch, @@ -1987,16 +1657,14 @@ static struct io_kiocb *io_alloc_req(struct io_ring_ctx *ctx, if (unlikely(ret <= 0)) { state->reqs[0] = kmem_cache_alloc(req_cachep, gfp); if (!state->reqs[0]) - goto fallback; + return NULL; ret = 1; } state->free_reqs = ret; } - +got_req: state->free_reqs--; return state->reqs[state->free_reqs]; -fallback: - return io_get_fallback_req(ctx); } static inline void io_put_file(struct io_kiocb *req, struct file *file, @@ -2014,27 +1682,43 @@ static void io_dismantle_req(struct io_kiocb *req) kfree(req->async_data); if (req->file) io_put_file(req, req->file, (req->flags & REQ_F_FIXED_FILE)); - if (req->fixed_file_refs) - percpu_ref_put(req->fixed_file_refs); - io_req_clean_work(req); + if (req->fixed_rsrc_refs) + percpu_ref_put(req->fixed_rsrc_refs); + if (req->work.creds) { + put_cred(req->work.creds); + req->work.creds = NULL; + } + + if (req->flags & REQ_F_INFLIGHT) { + struct io_ring_ctx *ctx = req->ctx; + unsigned long flags; + + spin_lock_irqsave(&ctx->inflight_lock, flags); + list_del(&req->inflight_entry); + spin_unlock_irqrestore(&ctx->inflight_lock, flags); + req->flags &= ~REQ_F_INFLIGHT; + } +} + +/* must to be called somewhat shortly after putting a request */ +static inline void io_put_task(struct task_struct *task, int nr) +{ + struct io_uring_task *tctx = task->io_uring; + + percpu_counter_sub(&tctx->inflight, nr); + if (unlikely(atomic_read(&tctx->in_idle))) + wake_up(&tctx->wait); + put_task_struct_many(task, nr); } static void __io_free_req(struct io_kiocb *req) { - struct io_uring_task *tctx = req->task->io_uring; struct io_ring_ctx *ctx = req->ctx; io_dismantle_req(req); + io_put_task(req->task, 1); - percpu_counter_dec(&tctx->inflight); - if (atomic_read(&tctx->in_idle)) - wake_up(&tctx->wait); - put_task_struct(req->task); - - if (likely(!io_is_fallback_req(req))) - kmem_cache_free(req_cachep, req); - else - clear_bit_unlock(0, (unsigned long *) &ctx->fallback_req); + kmem_cache_free(req_cachep, req); percpu_ref_put(&ctx->refs); } @@ -2046,15 +1730,11 @@ static inline void io_remove_next_linked(struct io_kiocb *req) nxt->link = NULL; } -static void io_kill_linked_timeout(struct io_kiocb *req) +static bool io_kill_linked_timeout(struct io_kiocb *req) + __must_hold(&req->ctx->completion_lock) { - struct io_ring_ctx *ctx = req->ctx; - struct io_kiocb *link; + struct io_kiocb *link = req->link; bool cancelled = false; - unsigned long flags; - - spin_lock_irqsave(&ctx->completion_lock, flags); - link = req->link; /* * Can happen if a linked timeout fired and link had been like @@ -2069,58 +1749,48 @@ static void io_kill_linked_timeout(struct io_kiocb *req) ret = hrtimer_try_to_cancel(&io->timer); if (ret != -1) { io_cqring_fill_event(link, -ECANCELED); - io_commit_cqring(ctx); + io_put_req_deferred(link, 1); cancelled = true; } } req->flags &= ~REQ_F_LINK_TIMEOUT; - spin_unlock_irqrestore(&ctx->completion_lock, flags); - - if (cancelled) { - io_cqring_ev_posted(ctx); - io_put_req(link); - } + return cancelled; } - static void io_fail_links(struct io_kiocb *req) + __must_hold(&req->ctx->completion_lock) { - struct io_kiocb *link, *nxt; - struct io_ring_ctx *ctx = req->ctx; - unsigned long flags; + struct io_kiocb *nxt, *link = req->link; - spin_lock_irqsave(&ctx->completion_lock, flags); - link = req->link; req->link = NULL; - while (link) { nxt = link->link; link->link = NULL; trace_io_uring_fail_link(req, link); io_cqring_fill_event(link, -ECANCELED); - - /* - * It's ok to free under spinlock as they're not linked anymore, - * but avoid REQ_F_WORK_INITIALIZED because it may deadlock on - * work.fs->lock. - */ - if (link->flags & REQ_F_WORK_INITIALIZED) - io_put_req_deferred(link, 2); - else - io_double_put_req(link); + io_put_req_deferred(link, 2); link = nxt; } - io_commit_cqring(ctx); - spin_unlock_irqrestore(&ctx->completion_lock, flags); +} - io_cqring_ev_posted(ctx); +static bool io_disarm_next(struct io_kiocb *req) + __must_hold(&req->ctx->completion_lock) +{ + bool posted = false; + + if (likely(req->flags & REQ_F_LINK_TIMEOUT)) + posted = io_kill_linked_timeout(req); + if (unlikely(req->flags & REQ_F_FAIL_LINK)) { + posted |= (req->link != NULL); + io_fail_links(req); + } + return posted; } static struct io_kiocb *__io_req_find_next(struct io_kiocb *req) { - if (req->flags & REQ_F_LINK_TIMEOUT) - io_kill_linked_timeout(req); + struct io_kiocb *nxt; /* * If LINK is set, we have dependent requests in this chain. If we @@ -2128,23 +1798,127 @@ static struct io_kiocb *__io_req_find_next(struct io_kiocb *req) * dependencies to the next request. In case of failure, fail the rest * of the chain. */ - if (likely(!(req->flags & REQ_F_FAIL_LINK))) { - struct io_kiocb *nxt = req->link; + if (req->flags & (REQ_F_LINK_TIMEOUT | REQ_F_FAIL_LINK)) { + struct io_ring_ctx *ctx = req->ctx; + unsigned long flags; + bool posted; - req->link = NULL; - return nxt; + spin_lock_irqsave(&ctx->completion_lock, flags); + posted = io_disarm_next(req); + if (posted) + io_commit_cqring(req->ctx); + spin_unlock_irqrestore(&ctx->completion_lock, flags); + if (posted) + io_cqring_ev_posted(ctx); } - io_fail_links(req); - return NULL; + nxt = req->link; + req->link = NULL; + return nxt; } static inline struct io_kiocb *io_req_find_next(struct io_kiocb *req) { - if (likely(!(req->link) && !(req->flags & REQ_F_LINK_TIMEOUT))) + if (likely(!(req->flags & (REQ_F_LINK|REQ_F_HARDLINK)))) return NULL; return __io_req_find_next(req); } +static void ctx_flush_and_put(struct io_ring_ctx *ctx) +{ + if (!ctx) + return; + if (ctx->submit_state.comp.nr) { + mutex_lock(&ctx->uring_lock); + io_submit_flush_completions(&ctx->submit_state.comp, ctx); + mutex_unlock(&ctx->uring_lock); + } + percpu_ref_put(&ctx->refs); +} + +static bool __tctx_task_work(struct io_uring_task *tctx) +{ + struct io_ring_ctx *ctx = NULL; + struct io_wq_work_list list; + struct io_wq_work_node *node; + + if (wq_list_empty(&tctx->task_list)) + return false; + + spin_lock_irq(&tctx->task_lock); + list = tctx->task_list; + INIT_WQ_LIST(&tctx->task_list); + spin_unlock_irq(&tctx->task_lock); + + node = list.first; + while (node) { + struct io_wq_work_node *next = node->next; + struct io_kiocb *req; + + req = container_of(node, struct io_kiocb, io_task_work.node); + if (req->ctx != ctx) { + ctx_flush_and_put(ctx); + ctx = req->ctx; + percpu_ref_get(&ctx->refs); + } + + req->task_work.func(&req->task_work); + node = next; + } + + ctx_flush_and_put(ctx); + return list.first != NULL; +} + +static void tctx_task_work(struct callback_head *cb) +{ + struct io_uring_task *tctx = container_of(cb, struct io_uring_task, task_work); + + clear_bit(0, &tctx->task_state); + + while (__tctx_task_work(tctx)) + cond_resched(); +} + +static int io_task_work_add(struct task_struct *tsk, struct io_kiocb *req, + enum task_work_notify_mode notify) +{ + struct io_uring_task *tctx = tsk->io_uring; + struct io_wq_work_node *node, *prev; + unsigned long flags; + int ret; + + WARN_ON_ONCE(!tctx); + + spin_lock_irqsave(&tctx->task_lock, flags); + wq_list_add_tail(&req->io_task_work.node, &tctx->task_list); + spin_unlock_irqrestore(&tctx->task_lock, flags); + + /* task_work already pending, we're done */ + if (test_bit(0, &tctx->task_state) || + test_and_set_bit(0, &tctx->task_state)) + return 0; + + if (!task_work_add(tsk, &tctx->task_work, notify)) + return 0; + + /* + * Slow path - we failed, find and delete work. if the work is not + * in the list, it got run and we're fine. + */ + ret = 0; + spin_lock_irqsave(&tctx->task_lock, flags); + wq_list_for_each(node, prev, &tctx->task_list) { + if (&req->io_task_work.node == node) { + wq_list_del(&tctx->task_list, node, prev); + ret = 1; + break; + } + } + spin_unlock_irqrestore(&tctx->task_lock, flags); + clear_bit(0, &tctx->task_state); + return ret; +} + static int io_req_task_work_add(struct io_kiocb *req) { struct task_struct *tsk = req->task; @@ -2165,13 +1939,53 @@ static int io_req_task_work_add(struct io_kiocb *req) if (!(ctx->flags & IORING_SETUP_SQPOLL)) notify = TWA_SIGNAL; - ret = task_work_add(tsk, &req->task_work, notify); + ret = io_task_work_add(tsk, req, notify); if (!ret) wake_up_process(tsk); return ret; } +static bool io_run_task_work_head(struct callback_head **work_head) +{ + struct callback_head *work, *next; + bool executed = false; + + do { + work = xchg(work_head, NULL); + if (!work) + break; + + do { + next = work->next; + work->func(work); + work = next; + cond_resched(); + } while (work); + executed = true; + } while (1); + + return executed; +} + +static void io_task_work_add_head(struct callback_head **work_head, + struct callback_head *task_work) +{ + struct callback_head *head; + + do { + head = READ_ONCE(*work_head); + task_work->next = head; + } while (cmpxchg(work_head, head, task_work) != head); +} + +static void io_req_task_work_add_fallback(struct io_kiocb *req, + task_work_func_t cb) +{ + init_task_work(&req->task_work, cb); + io_task_work_add_head(&req->ctx->exit_task_work, &req->task_work); +} + static void __io_req_task_cancel(struct io_kiocb *req, int error) { struct io_ring_ctx *ctx = req->ctx; @@ -2191,7 +2005,9 @@ static void io_req_task_cancel(struct callback_head *cb) struct io_kiocb *req = container_of(cb, struct io_kiocb, task_work); struct io_ring_ctx *ctx = req->ctx; - __io_req_task_cancel(req, -ECANCELED); + mutex_lock(&ctx->uring_lock); + __io_req_task_cancel(req, req->result); + mutex_unlock(&ctx->uring_lock); percpu_ref_put(&ctx->refs); } @@ -2199,46 +2015,45 @@ static void __io_req_task_submit(struct io_kiocb *req) { struct io_ring_ctx *ctx = req->ctx; + /* ctx stays valid until unlock, even if we drop all ours ctx->refs */ mutex_lock(&ctx->uring_lock); - if (!ctx->sqo_dead && - !__io_sq_thread_acquire_mm(ctx) && - !__io_sq_thread_acquire_files(ctx)) - __io_queue_sqe(req, NULL); + if (!(current->flags & PF_EXITING) && !current->in_execve) + __io_queue_sqe(req); else __io_req_task_cancel(req, -EFAULT); mutex_unlock(&ctx->uring_lock); - - if (ctx->flags & IORING_SETUP_SQPOLL) - io_sq_thread_drop_mm_files(); } static void io_req_task_submit(struct callback_head *cb) { struct io_kiocb *req = container_of(cb, struct io_kiocb, task_work); - struct io_ring_ctx *ctx = req->ctx; __io_req_task_submit(req); - percpu_ref_put(&ctx->refs); } static void io_req_task_queue(struct io_kiocb *req) { int ret; - init_task_work(&req->task_work, io_req_task_submit); - percpu_ref_get(&req->ctx->refs); - + req->task_work.func = io_req_task_submit; ret = io_req_task_work_add(req); if (unlikely(ret)) { - struct task_struct *tsk; - - init_task_work(&req->task_work, io_req_task_cancel); - tsk = io_wq_get_task(req->ctx->io_wq); - task_work_add(tsk, &req->task_work, TWA_NONE); - wake_up_process(tsk); + req->result = -ECANCELED; + percpu_ref_get(&req->ctx->refs); + io_req_task_work_add_fallback(req, io_req_task_cancel); } } +static void io_req_task_queue_fail(struct io_kiocb *req, int ret) +{ + percpu_ref_get(&req->ctx->refs); + req->result = ret; + req->task_work.func = io_req_task_cancel; + + if (unlikely(io_req_task_work_add(req))) + io_req_task_work_add_fallback(req, io_req_task_cancel); +} + static inline void io_queue_next(struct io_kiocb *req) { struct io_kiocb *nxt = io_req_find_next(req); @@ -2254,70 +2069,75 @@ static void io_free_req(struct io_kiocb *req) } struct req_batch { - void *reqs[IO_IOPOLL_BATCH]; - int to_free; - struct task_struct *task; int task_refs; + int ctx_refs; }; static inline void io_init_req_batch(struct req_batch *rb) { - rb->to_free = 0; rb->task_refs = 0; + rb->ctx_refs = 0; rb->task = NULL; } -static void __io_req_free_batch_flush(struct io_ring_ctx *ctx, - struct req_batch *rb) -{ - kmem_cache_free_bulk(req_cachep, rb->to_free, rb->reqs); - percpu_ref_put_many(&ctx->refs, rb->to_free); - rb->to_free = 0; -} - static void io_req_free_batch_finish(struct io_ring_ctx *ctx, struct req_batch *rb) { - if (rb->to_free) - __io_req_free_batch_flush(ctx, rb); - if (rb->task) { - struct io_uring_task *tctx = rb->task->io_uring; - - percpu_counter_sub(&tctx->inflight, rb->task_refs); - if (atomic_read(&tctx->in_idle)) - wake_up(&tctx->wait); - put_task_struct_many(rb->task, rb->task_refs); - rb->task = NULL; - } + if (rb->task) + io_put_task(rb->task, rb->task_refs); + if (rb->ctx_refs) + percpu_ref_put_many(&ctx->refs, rb->ctx_refs); } -static void io_req_free_batch(struct req_batch *rb, struct io_kiocb *req) +static void io_req_free_batch(struct req_batch *rb, struct io_kiocb *req, + struct io_submit_state *state) { - if (unlikely(io_is_fallback_req(req))) { - io_free_req(req); - return; - } io_queue_next(req); if (req->task != rb->task) { - if (rb->task) { - struct io_uring_task *tctx = rb->task->io_uring; - - percpu_counter_sub(&tctx->inflight, rb->task_refs); - if (atomic_read(&tctx->in_idle)) - wake_up(&tctx->wait); - put_task_struct_many(rb->task, rb->task_refs); - } + if (rb->task) + io_put_task(rb->task, rb->task_refs); rb->task = req->task; rb->task_refs = 0; } rb->task_refs++; + rb->ctx_refs++; io_dismantle_req(req); - rb->reqs[rb->to_free++] = req; - if (unlikely(rb->to_free == ARRAY_SIZE(rb->reqs))) - __io_req_free_batch_flush(req->ctx, rb); + if (state->free_reqs != ARRAY_SIZE(state->reqs)) + state->reqs[state->free_reqs++] = req; + else + list_add(&req->compl.list, &state->comp.free_list); +} + +static void io_submit_flush_completions(struct io_comp_state *cs, + struct io_ring_ctx *ctx) +{ + int i, nr = cs->nr; + struct io_kiocb *req; + struct req_batch rb; + + io_init_req_batch(&rb); + spin_lock_irq(&ctx->completion_lock); + for (i = 0; i < nr; i++) { + req = cs->reqs[i]; + __io_cqring_fill_event(req, req->result, req->compl.cflags); + } + io_commit_cqring(ctx); + spin_unlock_irq(&ctx->completion_lock); + + io_cqring_ev_posted(ctx); + for (i = 0; i < nr; i++) { + req = cs->reqs[i]; + + /* submission and completion refs */ + if (refcount_sub_and_test(2, &req->refs)) + io_req_free_batch(&rb, req, &ctx->submit_state); + } + + io_req_free_batch_finish(ctx, &rb); + cs->nr = 0; } /* @@ -2352,15 +2172,10 @@ static void io_free_req_deferred(struct io_kiocb *req) { int ret; - init_task_work(&req->task_work, io_put_req_deferred_cb); + req->task_work.func = io_put_req_deferred_cb; ret = io_req_task_work_add(req); - if (unlikely(ret)) { - struct task_struct *tsk; - - tsk = io_wq_get_task(req->ctx->io_wq); - task_work_add(tsk, &req->task_work, TWA_NONE); - wake_up_process(tsk); - } + if (unlikely(ret)) + io_req_task_work_add_fallback(req, io_put_req_deferred_cb); } static inline void io_put_req_deferred(struct io_kiocb *req, int refs) @@ -2369,22 +2184,6 @@ static inline void io_put_req_deferred(struct io_kiocb *req, int refs) io_free_req_deferred(req); } -static struct io_wq_work *io_steal_work(struct io_kiocb *req) -{ - struct io_kiocb *nxt; - - /* - * A ref is owned by io-wq in which context we're. So, if that's the - * last one, it's safe to steal next work. False negatives are Ok, - * it just will be re-punted async in io_put_work() - */ - if (refcount_read(&req->refs) != 1) - return NULL; - - nxt = io_req_find_next(req); - return nxt ? &nxt->work : NULL; -} - static void io_double_put_req(struct io_kiocb *req) { /* drop both submit and complete references */ @@ -2443,17 +2242,6 @@ static inline bool io_run_task_work(void) return false; } -static void io_iopoll_queue(struct list_head *again) -{ - struct io_kiocb *req; - - do { - req = list_first_entry(again, struct io_kiocb, inflight_entry); - list_del(&req->inflight_entry); - __io_complete_rw(req, -EAGAIN, 0, NULL); - } while (!list_empty(again)); -} - /* * Find and free completed poll iocbs */ @@ -2462,7 +2250,6 @@ static void io_iopoll_complete(struct io_ring_ctx *ctx, unsigned int *nr_events, { struct req_batch rb; struct io_kiocb *req; - LIST_HEAD(again); /* order with ->result store in io_complete_rw_iopoll() */ smp_rmb(); @@ -2472,13 +2259,13 @@ static void io_iopoll_complete(struct io_ring_ctx *ctx, unsigned int *nr_events, int cflags = 0; req = list_first_entry(done, struct io_kiocb, inflight_entry); + list_del(&req->inflight_entry); + if (READ_ONCE(req->result) == -EAGAIN) { - req->result = 0; req->iopoll_completed = 0; - list_move_tail(&req->inflight_entry, &again); - continue; + if (io_rw_reissue(req)) + continue; } - list_del(&req->inflight_entry); if (req->flags & REQ_F_BUFFER_SELECTED) cflags = io_put_rw_kbuf(req); @@ -2487,15 +2274,12 @@ static void io_iopoll_complete(struct io_ring_ctx *ctx, unsigned int *nr_events, (*nr_events)++; if (refcount_dec_and_test(&req->refs)) - io_req_free_batch(&rb, req); + io_req_free_batch(&rb, req, &ctx->submit_state); } io_commit_cqring(ctx); io_cqring_ev_posted_iopoll(ctx); io_req_free_batch_finish(ctx, &rb); - - if (!list_empty(&again)) - io_iopoll_queue(&again); } static int io_do_iopoll(struct io_ring_ctx *ctx, unsigned int *nr_events, @@ -2662,34 +2446,16 @@ static void kiocb_end_write(struct io_kiocb *req) file_end_write(req->file); } -static void io_complete_rw_common(struct kiocb *kiocb, long res, - struct io_comp_state *cs) -{ - struct io_kiocb *req = container_of(kiocb, struct io_kiocb, rw.kiocb); - int cflags = 0; - - if (kiocb->ki_flags & IOCB_WRITE) - kiocb_end_write(req); - - if (res != req->result) - req_set_fail_links(req); - if (req->flags & REQ_F_BUFFER_SELECTED) - cflags = io_put_rw_kbuf(req); - __io_req_complete(req, res, cflags, cs); -} - #ifdef CONFIG_BLOCK -static bool io_resubmit_prep(struct io_kiocb *req, int error) +static bool io_resubmit_prep(struct io_kiocb *req) { struct iovec inline_vecs[UIO_FASTIOV], *iovec = inline_vecs; - ssize_t ret = -ECANCELED; + int rw, ret; struct iov_iter iter; - int rw; - if (error) { - ret = error; - goto end_req; - } + /* already prepared */ + if (req->async_data) + return true; switch (req->opcode) { case IORING_OP_READV: @@ -2705,69 +2471,95 @@ static bool io_resubmit_prep(struct io_kiocb *req, int error) default: printk_once(KERN_WARNING "io_uring: bad opcode in resubmit %d\n", req->opcode); - goto end_req; + return false; } - if (!req->async_data) { - ret = io_import_iovec(rw, req, &iovec, &iter, false); - if (ret < 0) - goto end_req; - ret = io_setup_async_rw(req, iovec, inline_vecs, &iter, false); - if (!ret) - return true; - kfree(iovec); - } else { - return true; - } -end_req: - req_set_fail_links(req); - return false; + ret = io_import_iovec(rw, req, &iovec, &iter, false); + if (ret < 0) + return false; + return !io_setup_async_rw(req, iovec, inline_vecs, &iter, false); } -#endif -static bool io_rw_reissue(struct io_kiocb *req, long res) +static bool io_rw_should_reissue(struct io_kiocb *req) { -#ifdef CONFIG_BLOCK umode_t mode = file_inode(req->file)->i_mode; - int ret; + struct io_ring_ctx *ctx = req->ctx; if (!S_ISBLK(mode) && !S_ISREG(mode)) return false; - if ((res != -EAGAIN && res != -EOPNOTSUPP) || io_wq_current_is_worker()) + if ((req->flags & REQ_F_NOWAIT) || (io_wq_current_is_worker() && + !(ctx->flags & IORING_SETUP_IOPOLL))) + return false; + /* + * If ref is dying, we might be running poll reap from the exit work. + * Don't attempt to reissue from that path, just let it fail with + * -EAGAIN. + */ + if (percpu_ref_is_dying(&ctx->refs)) return false; + return true; +} +#endif - lockdep_assert_held(&req->ctx->uring_lock); +static bool io_rw_reissue(struct io_kiocb *req) +{ +#ifdef CONFIG_BLOCK + if (!io_rw_should_reissue(req)) + return false; - ret = io_sq_thread_acquire_mm_files(req->ctx, req); + lockdep_assert_held(&req->ctx->uring_lock); - if (io_resubmit_prep(req, ret)) { + if (io_resubmit_prep(req)) { refcount_inc(&req->refs); io_queue_async_work(req); return true; } - + req_set_fail_links(req); #endif return false; } static void __io_complete_rw(struct io_kiocb *req, long res, long res2, - struct io_comp_state *cs) + unsigned int issue_flags) { - if (!io_rw_reissue(req, res)) - io_complete_rw_common(&req->rw.kiocb, res, cs); + int cflags = 0; + + if ((res == -EAGAIN || res == -EOPNOTSUPP) && io_rw_reissue(req)) + return; + if (res != req->result) + req_set_fail_links(req); + + if (req->rw.kiocb.ki_flags & IOCB_WRITE) + kiocb_end_write(req); + if (req->flags & REQ_F_BUFFER_SELECTED) + cflags = io_put_rw_kbuf(req); + __io_req_complete(req, issue_flags, res, cflags); } static void io_complete_rw(struct kiocb *kiocb, long res, long res2) { struct io_kiocb *req = container_of(kiocb, struct io_kiocb, rw.kiocb); - __io_complete_rw(req, res, res2, NULL); + __io_complete_rw(req, res, res2, 0); } static void io_complete_rw_iopoll(struct kiocb *kiocb, long res, long res2) { struct io_kiocb *req = container_of(kiocb, struct io_kiocb, rw.kiocb); +#ifdef CONFIG_BLOCK + /* Rewind iter, if we have one. iopoll path resubmits as usual */ + if (res == -EAGAIN && io_rw_should_reissue(req)) { + struct io_async_rw *rw = req->async_data; + + if (rw) + iov_iter_revert(&rw->iter, + req->result - iov_iter_count(&rw->iter)); + else if (!io_resubmit_prep(req)) + res = -EIO; + } +#endif + if (kiocb->ki_flags & IOCB_WRITE) kiocb_end_write(req); @@ -2825,16 +2617,12 @@ static void io_iopoll_req_issued(struct io_kiocb *req, bool in_async) wake_up(&ctx->sq_data->wait); } -static inline void __io_state_file_put(struct io_submit_state *state) -{ - fput_many(state->file, state->file_refs); - state->file_refs = 0; -} - static inline void io_state_file_put(struct io_submit_state *state) { - if (state->file_refs) - __io_state_file_put(state); + if (state->file_refs) { + fput_many(state->file, state->file_refs); + state->file_refs = 0; + } } /* @@ -2852,7 +2640,7 @@ static struct file *__io_file_get(struct io_submit_state *state, int fd) state->file_refs--; return state->file; } - __io_state_file_put(state); + io_state_file_put(state); } state->file = fget_many(fd, state->ios_left); if (unlikely(!state->file)) @@ -2910,16 +2698,17 @@ static int io_prep_rw(struct io_kiocb *req, const struct io_uring_sqe *sqe) { struct io_ring_ctx *ctx = req->ctx; struct kiocb *kiocb = &req->rw.kiocb; + struct file *file = req->file; unsigned ioprio; int ret; - if (S_ISREG(file_inode(req->file)->i_mode)) + if (S_ISREG(file_inode(file)->i_mode)) req->flags |= REQ_F_ISREG; kiocb->ki_pos = READ_ONCE(sqe->off); - if (kiocb->ki_pos == -1 && !(req->file->f_mode & FMODE_STREAM)) { + if (kiocb->ki_pos == -1 && !(file->f_mode & FMODE_STREAM)) { req->flags |= REQ_F_CUR_POS; - kiocb->ki_pos = req->file->f_pos; + kiocb->ki_pos = file->f_pos; } kiocb->ki_hint = ki_hint_validate(file_write_hint(kiocb->ki_filp)); kiocb->ki_flags = iocb_flags(kiocb->ki_filp); @@ -2927,6 +2716,10 @@ static int io_prep_rw(struct io_kiocb *req, const struct io_uring_sqe *sqe) if (unlikely(ret)) return ret; + /* don't allow async punt for O_NONBLOCK or RWF_NOWAIT */ + if ((kiocb->ki_flags & IOCB_NOWAIT) || (file->f_flags & O_NONBLOCK)) + req->flags |= REQ_F_NOWAIT; + ioprio = READ_ONCE(sqe->ioprio); if (ioprio) { ret = ioprio_check_cap(ioprio); @@ -2937,10 +2730,6 @@ static int io_prep_rw(struct io_kiocb *req, const struct io_uring_sqe *sqe) } else kiocb->ki_ioprio = get_current_ioprio(); - /* don't allow async punt if RWF_NOWAIT was requested */ - if (kiocb->ki_flags & IOCB_NOWAIT) - req->flags |= REQ_F_NOWAIT; - if (ctx->flags & IORING_SETUP_IOPOLL) { if (!(kiocb->ki_flags & IOCB_DIRECT) || !kiocb->ki_filp->f_op->iopoll) @@ -2983,7 +2772,7 @@ static inline void io_rw_done(struct kiocb *kiocb, ssize_t ret) } static void kiocb_done(struct kiocb *kiocb, ssize_t ret, - struct io_comp_state *cs) + unsigned int issue_flags) { struct io_kiocb *req = container_of(kiocb, struct io_kiocb, rw.kiocb); struct io_async_rw *io = req->async_data; @@ -2999,13 +2788,12 @@ static void kiocb_done(struct kiocb *kiocb, ssize_t ret, if (req->flags & REQ_F_CUR_POS) req->file->f_pos = kiocb->ki_pos; if (ret >= 0 && kiocb->ki_complete == io_complete_rw) - __io_complete_rw(req, ret, 0, cs); + __io_complete_rw(req, ret, 0, issue_flags); else io_rw_done(kiocb, ret); } -static ssize_t io_import_fixed(struct io_kiocb *req, int rw, - struct iov_iter *iter) +static int io_import_fixed(struct io_kiocb *req, int rw, struct iov_iter *iter) { struct io_ring_ctx *ctx = req->ctx; size_t len = req->rw.len; @@ -3069,7 +2857,7 @@ static ssize_t io_import_fixed(struct io_kiocb *req, int rw, } } - return len; + return 0; } static void io_ring_submit_unlock(struct io_ring_ctx *ctx, bool needs_lock) @@ -3103,7 +2891,7 @@ static struct io_buffer *io_buffer_select(struct io_kiocb *req, size_t *len, lockdep_assert_held(&req->ctx->uring_lock); - head = idr_find(&req->ctx->io_buffer_idr, bgid); + head = xa_load(&req->ctx->io_buffers, bgid); if (head) { if (!list_empty(&head->list)) { kbuf = list_last_entry(&head->list, struct io_buffer, @@ -3111,7 +2899,7 @@ static struct io_buffer *io_buffer_select(struct io_kiocb *req, size_t *len, list_del(&kbuf->list); } else { kbuf = head; - idr_remove(&req->ctx->io_buffer_idr, bgid); + xa_erase(&req->ctx->io_buffers, bgid); } if (*len > kbuf->len) *len = kbuf->len; @@ -3210,16 +2998,14 @@ static ssize_t io_iov_buffer_select(struct io_kiocb *req, struct iovec *iov, return __io_iov_buffer_select(req, iov, needs_lock); } -static ssize_t io_import_iovec(int rw, struct io_kiocb *req, - struct iovec **iovec, struct iov_iter *iter, - bool needs_lock) +static int io_import_iovec(int rw, struct io_kiocb *req, struct iovec **iovec, + struct iov_iter *iter, bool needs_lock) { void __user *buf = u64_to_user_ptr(req->rw.addr); size_t sqe_len = req->rw.len; + u8 opcode = req->opcode; ssize_t ret; - u8 opcode; - opcode = req->opcode; if (opcode == IORING_OP_READ_FIXED || opcode == IORING_OP_WRITE_FIXED) { *iovec = NULL; return io_import_fixed(req, rw, iter); @@ -3244,10 +3030,8 @@ static ssize_t io_import_iovec(int rw, struct io_kiocb *req, if (req->flags & REQ_F_BUFFER_SELECT) { ret = io_iov_buffer_select(req, *iovec, needs_lock); - if (!ret) { - ret = (*iovec)->iov_len; - iov_iter_init(iter, rw, *iovec, 1, ret); - } + if (!ret) + iov_iter_init(iter, rw, *iovec, 1, (*iovec)->iov_len); *iovec = NULL; return ret; } @@ -3365,8 +3149,10 @@ static int io_setup_async_rw(struct io_kiocb *req, const struct iovec *iovec, if (!force && !io_op_defs[req->opcode].needs_async_data) return 0; if (!req->async_data) { - if (__io_alloc_async_data(req)) + if (__io_alloc_async_data(req)) { + kfree(iovec); return -ENOMEM; + } io_req_map_rw(req, iovec, fast_iov, iter); } @@ -3377,7 +3163,7 @@ static inline int io_rw_prep_async(struct io_kiocb *req, int rw) { struct io_async_rw *iorw = req->async_data; struct iovec *iov = iorw->fast_iov; - ssize_t ret; + int ret; ret = io_import_iovec(rw, req, &iov, &iorw->iter, false); if (unlikely(ret < 0)) @@ -3392,19 +3178,9 @@ static inline int io_rw_prep_async(struct io_kiocb *req, int rw) static int io_read_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) { - ssize_t ret; - - ret = io_prep_rw(req, sqe); - if (ret) - return ret; - if (unlikely(!(req->file->f_mode & FMODE_READ))) return -EBADF; - - /* either don't need iovec imported or already have it */ - if (!req->async_data) - return 0; - return io_rw_prep_async(req, READ); + return io_prep_rw(req, sqe); } /* @@ -3423,7 +3199,6 @@ static int io_async_buf_func(struct wait_queue_entry *wait, unsigned mode, struct wait_page_queue *wpq; struct io_kiocb *req = wait->private; struct wait_page_key *key = arg; - int ret; wpq = container_of(wait, struct wait_page_queue, wait); @@ -3433,21 +3208,9 @@ static int io_async_buf_func(struct wait_queue_entry *wait, unsigned mode, req->rw.kiocb.ki_flags &= ~IOCB_WAITQ; list_del_init(&wait->entry); - init_task_work(&req->task_work, io_req_task_submit); - percpu_ref_get(&req->ctx->refs); - /* submit ref gets dropped, acquire a new one */ refcount_inc(&req->refs); - ret = io_req_task_work_add(req); - if (unlikely(ret)) { - struct task_struct *tsk; - - /* queue just for cancelation */ - init_task_work(&req->task_work, io_req_task_cancel); - tsk = io_wq_get_task(req->ctx->io_wq); - task_work_add(tsk, &req->task_work, TWA_NONE); - wake_up_process(tsk); - } + io_req_task_queue(req); return 1; } @@ -3504,15 +3267,14 @@ static int io_iter_do_read(struct io_kiocb *req, struct iov_iter *iter) return -EINVAL; } -static int io_read(struct io_kiocb *req, bool force_nonblock, - struct io_comp_state *cs) +static int io_read(struct io_kiocb *req, unsigned int issue_flags) { struct iovec inline_vecs[UIO_FASTIOV], *iovec = inline_vecs; struct kiocb *kiocb = &req->rw.kiocb; struct iov_iter __iter, *iter = &__iter; struct io_async_rw *rw = req->async_data; ssize_t io_size, ret, ret2; - bool no_async; + bool force_nonblock = issue_flags & IO_URING_F_NONBLOCK; if (rw) { iter = &rw->iter; @@ -3524,7 +3286,6 @@ static int io_read(struct io_kiocb *req, bool force_nonblock, } io_size = iov_iter_count(iter); req->result = io_size; - ret = 0; /* Ensure we clear previously set non-block flag */ if (!force_nonblock) @@ -3532,115 +3293,94 @@ static int io_read(struct io_kiocb *req, bool force_nonblock, else kiocb->ki_flags |= IOCB_NOWAIT; - /* If the file doesn't support async, just async punt */ - no_async = force_nonblock && !io_file_supports_async(req->file, READ); - if (no_async) - goto copy_iov; + if (force_nonblock && !io_file_supports_async(req->file, READ)) { + ret = io_setup_async_rw(req, iovec, inline_vecs, iter, true); + return ret ?: -EAGAIN; + } ret = rw_verify_area(READ, req->file, io_kiocb_ppos(kiocb), io_size); - if (unlikely(ret)) - goto out_free; + if (unlikely(ret)) { + kfree(iovec); + return ret; + } ret = io_iter_do_read(req, iter); - if (!ret) { - goto done; - } else if (ret == -EIOCBQUEUED) { - ret = 0; + if (ret == -EIOCBQUEUED) { + if (req->async_data) + iov_iter_revert(iter, io_size - iov_iter_count(iter)); goto out_free; } else if (ret == -EAGAIN) { /* IOPOLL retry should happen for io-wq threads */ if (!force_nonblock && !(req->ctx->flags & IORING_SETUP_IOPOLL)) goto done; - /* no retry on NONBLOCK marked file */ - if (req->file->f_flags & O_NONBLOCK) + /* no retry on NONBLOCK nor RWF_NOWAIT */ + if (req->flags & REQ_F_NOWAIT) goto done; /* some cases will consume bytes even on error returns */ iov_iter_revert(iter, io_size - iov_iter_count(iter)); ret = 0; - goto copy_iov; - } else if (ret < 0) { - /* make sure -ERESTARTSYS -> -EINTR is done */ + } else if (ret <= 0 || ret == io_size || !force_nonblock || + (req->flags & REQ_F_NOWAIT) || !(req->flags & REQ_F_ISREG)) { + /* read all, failed, already did sync or don't want to retry */ goto done; } - /* read it all, or we did blocking attempt. no retry. */ - if (!iov_iter_count(iter) || !force_nonblock || - (req->file->f_flags & O_NONBLOCK) || !(req->flags & REQ_F_ISREG)) - goto done; - - io_size -= ret; -copy_iov: ret2 = io_setup_async_rw(req, iovec, inline_vecs, iter, true); - if (ret2) { - ret = ret2; - goto out_free; - } - if (no_async) - return -EAGAIN; - rw = req->async_data; - /* it's copied and will be cleaned with ->io */ + if (ret2) + return ret2; + iovec = NULL; + rw = req->async_data; /* now use our persistent iterator, if we aren't already */ iter = &rw->iter; -retry: - rw->bytes_done += ret; - /* if we can retry, do so with the callbacks armed */ - if (!io_rw_should_retry(req)) { - kiocb->ki_flags &= ~IOCB_WAITQ; - return -EAGAIN; - } - /* - * Now retry read with the IOCB_WAITQ parts set in the iocb. If we - * get -EIOCBQUEUED, then we'll get a notification when the desired - * page gets unlocked. We can also get a partial read here, and if we - * do, then just retry at the new offset. - */ - ret = io_iter_do_read(req, iter); - if (ret == -EIOCBQUEUED) { - ret = 0; - goto out_free; - } else if (ret > 0 && ret < io_size) { + do { + io_size -= ret; + rw->bytes_done += ret; + /* if we can retry, do so with the callbacks armed */ + if (!io_rw_should_retry(req)) { + kiocb->ki_flags &= ~IOCB_WAITQ; + return -EAGAIN; + } + + /* + * Now retry read with the IOCB_WAITQ parts set in the iocb. If + * we get -EIOCBQUEUED, then we'll get a notification when the + * desired page gets unlocked. We can also get a partial read + * here, and if we do, then just retry at the new offset. + */ + ret = io_iter_do_read(req, iter); + if (ret == -EIOCBQUEUED) + return 0; /* we got some bytes, but not all. retry. */ - goto retry; - } + kiocb->ki_flags &= ~IOCB_WAITQ; + } while (ret > 0 && ret < io_size); done: - kiocb_done(kiocb, ret, cs); - ret = 0; + kiocb_done(kiocb, ret, issue_flags); out_free: - /* it's reportedly faster than delegating the null check to kfree() */ + /* it's faster to check here then delegate to kfree */ if (iovec) kfree(iovec); - return ret; + return 0; } static int io_write_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) { - ssize_t ret; - - ret = io_prep_rw(req, sqe); - if (ret) - return ret; - if (unlikely(!(req->file->f_mode & FMODE_WRITE))) return -EBADF; - - /* either don't need iovec imported or already have it */ - if (!req->async_data) - return 0; - return io_rw_prep_async(req, WRITE); + return io_prep_rw(req, sqe); } -static int io_write(struct io_kiocb *req, bool force_nonblock, - struct io_comp_state *cs) +static int io_write(struct io_kiocb *req, unsigned int issue_flags) { struct iovec inline_vecs[UIO_FASTIOV], *iovec = inline_vecs; struct kiocb *kiocb = &req->rw.kiocb; struct iov_iter __iter, *iter = &__iter; struct io_async_rw *rw = req->async_data; ssize_t ret, ret2, io_size; + bool force_nonblock = issue_flags & IO_URING_F_NONBLOCK; if (rw) { iter = &rw->iter; @@ -3699,22 +3439,23 @@ static int io_write(struct io_kiocb *req, bool force_nonblock, */ if (ret2 == -EOPNOTSUPP && (kiocb->ki_flags & IOCB_NOWAIT)) ret2 = -EAGAIN; - /* no retry on NONBLOCK marked file */ - if (ret2 == -EAGAIN && (req->file->f_flags & O_NONBLOCK)) + /* no retry on NONBLOCK nor RWF_NOWAIT */ + if (ret2 == -EAGAIN && (req->flags & REQ_F_NOWAIT)) goto done; + if (ret2 == -EIOCBQUEUED && req->async_data) + iov_iter_revert(iter, io_size - iov_iter_count(iter)); if (!force_nonblock || ret2 != -EAGAIN) { /* IOPOLL retry should happen for io-wq threads */ if ((req->ctx->flags & IORING_SETUP_IOPOLL) && ret2 == -EAGAIN) goto copy_iov; done: - kiocb_done(kiocb, ret2, cs); + kiocb_done(kiocb, ret2, issue_flags); } else { copy_iov: /* some cases will consume bytes even on error returns */ iov_iter_revert(iter, io_size - iov_iter_count(iter)); ret = io_setup_async_rw(req, iovec, inline_vecs, iter, false); - if (!ret) - return -EAGAIN; + return ret ?: -EAGAIN; } out_free: /* it's reportedly faster than delegating the null check to kfree() */ @@ -3752,12 +3493,12 @@ static int io_renameat_prep(struct io_kiocb *req, return 0; } -static int io_renameat(struct io_kiocb *req, bool force_nonblock) +static int io_renameat(struct io_kiocb *req, unsigned int issue_flags) { struct io_rename *ren = &req->rename; int ret; - if (force_nonblock) + if (issue_flags & IO_URING_F_NONBLOCK) return -EAGAIN; ret = do_renameat2(ren->old_dfd, ren->oldpath, ren->new_dfd, @@ -3794,12 +3535,12 @@ static int io_unlinkat_prep(struct io_kiocb *req, return 0; } -static int io_unlinkat(struct io_kiocb *req, bool force_nonblock) +static int io_unlinkat(struct io_kiocb *req, unsigned int issue_flags) { struct io_unlink *un = &req->unlink; int ret; - if (force_nonblock) + if (issue_flags & IO_URING_F_NONBLOCK) return -EAGAIN; if (un->flags & AT_REMOVEDIR) @@ -3831,13 +3572,13 @@ static int io_shutdown_prep(struct io_kiocb *req, #endif } -static int io_shutdown(struct io_kiocb *req, bool force_nonblock) +static int io_shutdown(struct io_kiocb *req, unsigned int issue_flags) { #if defined(CONFIG_NET) struct socket *sock; int ret; - if (force_nonblock) + if (issue_flags & IO_URING_F_NONBLOCK) return -EAGAIN; sock = sock_from_file(req->file); @@ -3881,7 +3622,6 @@ static int __io_splice_prep(struct io_kiocb *req, * Splice operation will be punted aync, and here need to * modify io_wq_work.flags, so initialize io_wq_work firstly. */ - io_req_init_async(req); req->work.flags |= IO_WQ_WORK_UNBOUND; } @@ -3896,7 +3636,7 @@ static int io_tee_prep(struct io_kiocb *req, return __io_splice_prep(req, sqe); } -static int io_tee(struct io_kiocb *req, bool force_nonblock) +static int io_tee(struct io_kiocb *req, unsigned int issue_flags) { struct io_splice *sp = &req->splice; struct file *in = sp->file_in; @@ -3904,7 +3644,7 @@ static int io_tee(struct io_kiocb *req, bool force_nonblock) unsigned int flags = sp->flags & ~SPLICE_F_FD_IN_FIXED; long ret = 0; - if (force_nonblock) + if (issue_flags & IO_URING_F_NONBLOCK) return -EAGAIN; if (sp->len) ret = do_tee(in, out, sp->len, flags); @@ -3927,7 +3667,7 @@ static int io_splice_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) return __io_splice_prep(req, sqe); } -static int io_splice(struct io_kiocb *req, bool force_nonblock) +static int io_splice(struct io_kiocb *req, unsigned int issue_flags) { struct io_splice *sp = &req->splice; struct file *in = sp->file_in; @@ -3936,7 +3676,7 @@ static int io_splice(struct io_kiocb *req, bool force_nonblock) loff_t *poff_in, *poff_out; long ret = 0; - if (force_nonblock) + if (issue_flags & IO_URING_F_NONBLOCK) return -EAGAIN; poff_in = (sp->off_in == -1) ? NULL : &sp->off_in; @@ -3957,18 +3697,18 @@ static int io_splice(struct io_kiocb *req, bool force_nonblock) /* * IORING_OP_NOP just posts a completion event, nothing else. */ -static int io_nop(struct io_kiocb *req, struct io_comp_state *cs) +static int io_nop(struct io_kiocb *req, unsigned int issue_flags) { struct io_ring_ctx *ctx = req->ctx; if (unlikely(ctx->flags & IORING_SETUP_IOPOLL)) return -EINVAL; - __io_req_complete(req, 0, 0, cs); + __io_req_complete(req, issue_flags, 0, 0); return 0; } -static int io_prep_fsync(struct io_kiocb *req, const struct io_uring_sqe *sqe) +static int io_fsync_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) { struct io_ring_ctx *ctx = req->ctx; @@ -3989,13 +3729,13 @@ static int io_prep_fsync(struct io_kiocb *req, const struct io_uring_sqe *sqe) return 0; } -static int io_fsync(struct io_kiocb *req, bool force_nonblock) +static int io_fsync(struct io_kiocb *req, unsigned int issue_flags) { loff_t end = req->sync.off + req->sync.len; int ret; /* fsync always requires a blocking context */ - if (force_nonblock) + if (issue_flags & IO_URING_F_NONBLOCK) return -EAGAIN; ret = vfs_fsync_range(req->file, req->sync.off, @@ -4021,12 +3761,12 @@ static int io_fallocate_prep(struct io_kiocb *req, return 0; } -static int io_fallocate(struct io_kiocb *req, bool force_nonblock) +static int io_fallocate(struct io_kiocb *req, unsigned int issue_flags) { int ret; /* fallocate always requiring blocking context */ - if (force_nonblock) + if (issue_flags & IO_URING_F_NONBLOCK) return -EAGAIN; ret = vfs_fallocate(req->file, req->sync.mode, req->sync.off, req->sync.len); @@ -4059,7 +3799,6 @@ static int __io_openat_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe return ret; } req->open.nofile = rlimit(RLIMIT_NOFILE); - req->open.ignore_nonblock = false; req->flags |= REQ_F_NEED_CLEANUP; return 0; } @@ -4097,43 +3836,53 @@ static int io_openat2_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) return __io_openat_prep(req, sqe); } -static int io_openat2(struct io_kiocb *req, bool force_nonblock) +static int io_openat2(struct io_kiocb *req, unsigned int issue_flags) { struct open_flags op; struct file *file; + bool nonblock_set; + bool resolve_nonblock; int ret; - if (force_nonblock && !req->open.ignore_nonblock) - return -EAGAIN; - ret = build_open_flags(&req->open.how, &op); if (ret) goto err; + nonblock_set = op.open_flag & O_NONBLOCK; + resolve_nonblock = req->open.how.resolve & RESOLVE_CACHED; + if (issue_flags & IO_URING_F_NONBLOCK) { + /* + * Don't bother trying for O_TRUNC, O_CREAT, or O_TMPFILE open, + * it'll always -EAGAIN + */ + if (req->open.how.flags & (O_TRUNC | O_CREAT | O_TMPFILE)) + return -EAGAIN; + op.lookup_flags |= LOOKUP_CACHED; + op.open_flag |= O_NONBLOCK; + } ret = __get_unused_fd_flags(req->open.how.flags, req->open.nofile); if (ret < 0) goto err; file = do_filp_open(req->open.dfd, req->open.filename, &op); + /* only retry if RESOLVE_CACHED wasn't already set by application */ + if ((!resolve_nonblock && (issue_flags & IO_URING_F_NONBLOCK)) && + file == ERR_PTR(-EAGAIN)) { + /* + * We could hang on to this 'fd', but seems like marginal + * gain for something that is now known to be a slower path. + * So just put it, and we'll get a new one when we retry. + */ + put_unused_fd(ret); + return -EAGAIN; + } + if (IS_ERR(file)) { put_unused_fd(ret); ret = PTR_ERR(file); - /* - * A work-around to ensure that /proc/self works that way - * that it should - if we get -EOPNOTSUPP back, then assume - * that proc_self_get_link() failed us because we're in async - * context. We should be safe to retry this from the task - * itself with force_nonblock == false set, as it should not - * block on lookup. Would be nice to know this upfront and - * avoid the async dance, but doesn't seem feasible. - */ - if (ret == -EOPNOTSUPP && io_wq_current_is_worker()) { - req->open.ignore_nonblock = true; - refcount_inc(&req->refs); - io_req_task_queue(req); - return 0; - } } else { + if ((issue_flags & IO_URING_F_NONBLOCK) && !nonblock_set) + file->f_flags &= ~O_NONBLOCK; fsnotify_open(file); fd_install(ret, file); } @@ -4146,9 +3895,9 @@ err: return 0; } -static int io_openat(struct io_kiocb *req, bool force_nonblock) +static int io_openat(struct io_kiocb *req, unsigned int issue_flags) { - return io_openat2(req, force_nonblock); + return io_openat2(req, issue_flags); } static int io_remove_buffers_prep(struct io_kiocb *req, @@ -4191,25 +3940,25 @@ static int __io_remove_buffers(struct io_ring_ctx *ctx, struct io_buffer *buf, } i++; kfree(buf); - idr_remove(&ctx->io_buffer_idr, bgid); + xa_erase(&ctx->io_buffers, bgid); return i; } -static int io_remove_buffers(struct io_kiocb *req, bool force_nonblock, - struct io_comp_state *cs) +static int io_remove_buffers(struct io_kiocb *req, unsigned int issue_flags) { struct io_provide_buf *p = &req->pbuf; struct io_ring_ctx *ctx = req->ctx; struct io_buffer *head; int ret = 0; + bool force_nonblock = issue_flags & IO_URING_F_NONBLOCK; io_ring_submit_lock(ctx, !force_nonblock); lockdep_assert_held(&ctx->uring_lock); ret = -ENOENT; - head = idr_find(&ctx->io_buffer_idr, p->bgid); + head = xa_load(&ctx->io_buffers, p->bgid); if (head) ret = __io_remove_buffers(ctx, head, p->bgid, p->nbufs); if (ret < 0) @@ -4217,11 +3966,11 @@ static int io_remove_buffers(struct io_kiocb *req, bool force_nonblock, /* need to hold the lock to complete IOPOLL requests */ if (ctx->flags & IORING_SETUP_IOPOLL) { - __io_req_complete(req, ret, 0, cs); + __io_req_complete(req, issue_flags, ret, 0); io_ring_submit_unlock(ctx, !force_nonblock); } else { io_ring_submit_unlock(ctx, !force_nonblock); - __io_req_complete(req, ret, 0, cs); + __io_req_complete(req, issue_flags, ret, 0); } return 0; } @@ -4280,43 +4029,36 @@ static int io_add_buffers(struct io_provide_buf *pbuf, struct io_buffer **head) return i ? i : -ENOMEM; } -static int io_provide_buffers(struct io_kiocb *req, bool force_nonblock, - struct io_comp_state *cs) +static int io_provide_buffers(struct io_kiocb *req, unsigned int issue_flags) { struct io_provide_buf *p = &req->pbuf; struct io_ring_ctx *ctx = req->ctx; struct io_buffer *head, *list; int ret = 0; + bool force_nonblock = issue_flags & IO_URING_F_NONBLOCK; io_ring_submit_lock(ctx, !force_nonblock); lockdep_assert_held(&ctx->uring_lock); - list = head = idr_find(&ctx->io_buffer_idr, p->bgid); + list = head = xa_load(&ctx->io_buffers, p->bgid); ret = io_add_buffers(p, &head); - if (ret < 0) - goto out; - - if (!list) { - ret = idr_alloc(&ctx->io_buffer_idr, head, p->bgid, p->bgid + 1, - GFP_KERNEL); - if (ret < 0) { + if (ret >= 0 && !list) { + ret = xa_insert(&ctx->io_buffers, p->bgid, head, GFP_KERNEL); + if (ret < 0) __io_remove_buffers(ctx, head, p->bgid, -1U); - goto out; - } } -out: if (ret < 0) req_set_fail_links(req); /* need to hold the lock to complete IOPOLL requests */ if (ctx->flags & IORING_SETUP_IOPOLL) { - __io_req_complete(req, ret, 0, cs); + __io_req_complete(req, issue_flags, ret, 0); io_ring_submit_unlock(ctx, !force_nonblock); } else { io_ring_submit_unlock(ctx, !force_nonblock); - __io_req_complete(req, ret, 0, cs); + __io_req_complete(req, issue_flags, ret, 0); } return 0; } @@ -4348,12 +4090,12 @@ static int io_epoll_ctl_prep(struct io_kiocb *req, #endif } -static int io_epoll_ctl(struct io_kiocb *req, bool force_nonblock, - struct io_comp_state *cs) +static int io_epoll_ctl(struct io_kiocb *req, unsigned int issue_flags) { #if defined(CONFIG_EPOLL) struct io_epoll *ie = &req->epoll; int ret; + bool force_nonblock = issue_flags & IO_URING_F_NONBLOCK; ret = do_epoll_ctl(ie->epfd, ie->op, ie->fd, &ie->event, force_nonblock); if (force_nonblock && ret == -EAGAIN) @@ -4361,7 +4103,7 @@ static int io_epoll_ctl(struct io_kiocb *req, bool force_nonblock, if (ret < 0) req_set_fail_links(req); - __io_req_complete(req, ret, 0, cs); + __io_req_complete(req, issue_flags, ret, 0); return 0; #else return -EOPNOTSUPP; @@ -4385,13 +4127,13 @@ static int io_madvise_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) #endif } -static int io_madvise(struct io_kiocb *req, bool force_nonblock) +static int io_madvise(struct io_kiocb *req, unsigned int issue_flags) { #if defined(CONFIG_ADVISE_SYSCALLS) && defined(CONFIG_MMU) struct io_madvise *ma = &req->madvise; int ret; - if (force_nonblock) + if (issue_flags & IO_URING_F_NONBLOCK) return -EAGAIN; ret = do_madvise(current->mm, ma->addr, ma->len, ma->advice); @@ -4417,12 +4159,12 @@ static int io_fadvise_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) return 0; } -static int io_fadvise(struct io_kiocb *req, bool force_nonblock) +static int io_fadvise(struct io_kiocb *req, unsigned int issue_flags) { struct io_fadvise *fa = &req->fadvise; int ret; - if (force_nonblock) { + if (issue_flags & IO_URING_F_NONBLOCK) { switch (fa->advice) { case POSIX_FADV_NORMAL: case POSIX_FADV_RANDOM: @@ -4458,12 +4200,12 @@ static int io_statx_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) return 0; } -static int io_statx(struct io_kiocb *req, bool force_nonblock) +static int io_statx(struct io_kiocb *req, unsigned int issue_flags) { struct io_statx *ctx = &req->statx; int ret; - if (force_nonblock) { + if (issue_flags & IO_URING_F_NONBLOCK) { /* only need file table for an actual valid fd */ if (ctx->dfd == -1 || ctx->dfd == AT_FDCWD) req->flags |= REQ_F_NO_FILE_TABLE; @@ -4481,13 +4223,6 @@ static int io_statx(struct io_kiocb *req, bool force_nonblock) static int io_close_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) { - /* - * If we queue this for async, it must not be cancellable. That would - * leave the 'file' in an undeterminate state, and here need to modify - * io_wq_work.flags, so initialize io_wq_work firstly. - */ - io_req_init_async(req); - if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL)) return -EINVAL; if (sqe->ioprio || sqe->off || sqe->addr || sqe->len || @@ -4497,54 +4232,66 @@ static int io_close_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) return -EBADF; req->close.fd = READ_ONCE(sqe->fd); - if ((req->file && req->file->f_op == &io_uring_fops)) - return -EBADF; - - req->close.put_file = NULL; return 0; } -static int io_close(struct io_kiocb *req, bool force_nonblock, - struct io_comp_state *cs) +static int io_close(struct io_kiocb *req, unsigned int issue_flags) { + struct files_struct *files = current->files; struct io_close *close = &req->close; + struct fdtable *fdt; + struct file *file; int ret; - /* might be already done during nonblock submission */ - if (!close->put_file) { - ret = close_fd_get_file(close->fd, &close->put_file); - if (ret < 0) - return (ret == -ENOENT) ? -EBADF : ret; + file = NULL; + ret = -EBADF; + spin_lock(&files->file_lock); + fdt = files_fdtable(files); + if (close->fd >= fdt->max_fds) { + spin_unlock(&files->file_lock); + goto err; + } + file = fdt->fd[close->fd]; + if (!file) { + spin_unlock(&files->file_lock); + goto err; + } + + if (file->f_op == &io_uring_fops) { + spin_unlock(&files->file_lock); + file = NULL; + goto err; } /* if the file has a flush method, be safe and punt to async */ - if (close->put_file->f_op->flush && force_nonblock) { - /* not safe to cancel at this point */ - req->work.flags |= IO_WQ_WORK_NO_CANCEL; - /* was never set, but play safe */ - req->flags &= ~REQ_F_NOWAIT; - /* avoid grabbing files - we don't need the files */ - req->flags |= REQ_F_NO_FILE_TABLE; + if (file->f_op->flush && (issue_flags & IO_URING_F_NONBLOCK)) { + spin_unlock(&files->file_lock); return -EAGAIN; } + ret = __close_fd_get_file(close->fd, &file); + spin_unlock(&files->file_lock); + if (ret < 0) { + if (ret == -ENOENT) + ret = -EBADF; + goto err; + } + /* No ->flush() or already async, safely close from here */ - ret = filp_close(close->put_file, req->work.identity->files); + ret = filp_close(file, current->files); +err: if (ret < 0) req_set_fail_links(req); - fput(close->put_file); - close->put_file = NULL; - __io_req_complete(req, ret, 0, cs); + if (file) + fput(file); + __io_req_complete(req, issue_flags, ret, 0); return 0; } -static int io_prep_sfr(struct io_kiocb *req, const struct io_uring_sqe *sqe) +static int io_sfr_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) { struct io_ring_ctx *ctx = req->ctx; - if (!req->file) - return -EBADF; - if (unlikely(ctx->flags & IORING_SETUP_IOPOLL)) return -EINVAL; if (unlikely(sqe->addr || sqe->ioprio || sqe->buf_index)) @@ -4556,12 +4303,12 @@ static int io_prep_sfr(struct io_kiocb *req, const struct io_uring_sqe *sqe) return 0; } -static int io_sync_file_range(struct io_kiocb *req, bool force_nonblock) +static int io_sync_file_range(struct io_kiocb *req, unsigned int issue_flags) { int ret; /* sync_file_range always requires a blocking context */ - if (force_nonblock) + if (issue_flags & IO_URING_F_NONBLOCK) return -EAGAIN; ret = sync_file_range(req->file, req->sync.off, req->sync.len, @@ -4581,30 +4328,44 @@ static int io_setup_async_msg(struct io_kiocb *req, if (async_msg) return -EAGAIN; if (io_alloc_async_data(req)) { - if (kmsg->iov != kmsg->fast_iov) - kfree(kmsg->iov); + kfree(kmsg->free_iov); return -ENOMEM; } async_msg = req->async_data; req->flags |= REQ_F_NEED_CLEANUP; memcpy(async_msg, kmsg, sizeof(*kmsg)); + async_msg->msg.msg_name = &async_msg->addr; + /* if were using fast_iov, set it to the new one */ + if (!async_msg->free_iov) + async_msg->msg.msg_iter.iov = async_msg->fast_iov; + return -EAGAIN; } static int io_sendmsg_copy_hdr(struct io_kiocb *req, struct io_async_msghdr *iomsg) { - iomsg->iov = iomsg->fast_iov; iomsg->msg.msg_name = &iomsg->addr; + iomsg->free_iov = iomsg->fast_iov; return sendmsg_copy_msghdr(&iomsg->msg, req->sr_msg.umsg, - req->sr_msg.msg_flags, &iomsg->iov); + req->sr_msg.msg_flags, &iomsg->free_iov); +} + +static int io_sendmsg_prep_async(struct io_kiocb *req) +{ + int ret; + + if (!io_op_defs[req->opcode].needs_async_data) + return 0; + ret = io_sendmsg_copy_hdr(req, req->async_data); + if (!ret) + req->flags |= REQ_F_NEED_CLEANUP; + return ret; } static int io_sendmsg_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) { - struct io_async_msghdr *async_msg = req->async_data; struct io_sr_msg *sr = &req->sr_msg; - int ret; if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL)) return -EINVAL; @@ -4617,70 +4378,62 @@ static int io_sendmsg_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) if (req->ctx->compat) sr->msg_flags |= MSG_CMSG_COMPAT; #endif - - if (!async_msg || !io_op_defs[req->opcode].needs_async_data) - return 0; - ret = io_sendmsg_copy_hdr(req, async_msg); - if (!ret) - req->flags |= REQ_F_NEED_CLEANUP; - return ret; + return 0; } -static int io_sendmsg(struct io_kiocb *req, bool force_nonblock, - struct io_comp_state *cs) +static int io_sendmsg(struct io_kiocb *req, unsigned int issue_flags) { struct io_async_msghdr iomsg, *kmsg; struct socket *sock; unsigned flags; + int min_ret = 0; int ret; sock = sock_from_file(req->file); if (unlikely(!sock)) return -ENOTSOCK; - if (req->async_data) { - kmsg = req->async_data; - kmsg->msg.msg_name = &kmsg->addr; - /* if iov is set, it's allocated already */ - if (!kmsg->iov) - kmsg->iov = kmsg->fast_iov; - kmsg->msg.msg_iter.iov = kmsg->iov; - } else { + kmsg = req->async_data; + if (!kmsg) { ret = io_sendmsg_copy_hdr(req, &iomsg); if (ret) return ret; kmsg = &iomsg; } - flags = req->sr_msg.msg_flags; + flags = req->sr_msg.msg_flags | MSG_NOSIGNAL; if (flags & MSG_DONTWAIT) req->flags |= REQ_F_NOWAIT; - else if (force_nonblock) + else if (issue_flags & IO_URING_F_NONBLOCK) flags |= MSG_DONTWAIT; + if (flags & MSG_WAITALL) + min_ret = iov_iter_count(&kmsg->msg.msg_iter); + ret = __sys_sendmsg_sock(sock, &kmsg->msg, flags); - if (force_nonblock && ret == -EAGAIN) + if ((issue_flags & IO_URING_F_NONBLOCK) && ret == -EAGAIN) return io_setup_async_msg(req, kmsg); if (ret == -ERESTARTSYS) ret = -EINTR; - if (kmsg->iov != kmsg->fast_iov) - kfree(kmsg->iov); + /* fast path, check for non-NULL to avoid function call */ + if (kmsg->free_iov) + kfree(kmsg->free_iov); req->flags &= ~REQ_F_NEED_CLEANUP; - if (ret < 0) + if (ret < min_ret) req_set_fail_links(req); - __io_req_complete(req, ret, 0, cs); + __io_req_complete(req, issue_flags, ret, 0); return 0; } -static int io_send(struct io_kiocb *req, bool force_nonblock, - struct io_comp_state *cs) +static int io_send(struct io_kiocb *req, unsigned int issue_flags) { struct io_sr_msg *sr = &req->sr_msg; struct msghdr msg; struct iovec iov; struct socket *sock; unsigned flags; + int min_ret = 0; int ret; sock = sock_from_file(req->file); @@ -4696,22 +4449,25 @@ static int io_send(struct io_kiocb *req, bool force_nonblock, msg.msg_controllen = 0; msg.msg_namelen = 0; - flags = req->sr_msg.msg_flags; + flags = req->sr_msg.msg_flags | MSG_NOSIGNAL; if (flags & MSG_DONTWAIT) req->flags |= REQ_F_NOWAIT; - else if (force_nonblock) + else if (issue_flags & IO_URING_F_NONBLOCK) flags |= MSG_DONTWAIT; + if (flags & MSG_WAITALL) + min_ret = iov_iter_count(&msg.msg_iter); + msg.msg_flags = flags; ret = sock_sendmsg(sock, &msg); - if (force_nonblock && ret == -EAGAIN) + if ((issue_flags & IO_URING_F_NONBLOCK) && ret == -EAGAIN) return -EAGAIN; if (ret == -ERESTARTSYS) ret = -EINTR; - if (ret < 0) + if (ret < min_ret) req_set_fail_links(req); - __io_req_complete(req, ret, 0, cs); + __io_req_complete(req, issue_flags, ret, 0); return 0; } @@ -4731,15 +4487,14 @@ static int __io_recvmsg_copy_hdr(struct io_kiocb *req, if (req->flags & REQ_F_BUFFER_SELECT) { if (iov_len > 1) return -EINVAL; - if (copy_from_user(iomsg->iov, uiov, sizeof(*uiov))) + if (copy_from_user(iomsg->fast_iov, uiov, sizeof(*uiov))) return -EFAULT; - sr->len = iomsg->iov[0].iov_len; - iov_iter_init(&iomsg->msg.msg_iter, READ, iomsg->iov, 1, - sr->len); - iomsg->iov = NULL; + sr->len = iomsg->fast_iov[0].iov_len; + iomsg->free_iov = NULL; } else { + iomsg->free_iov = iomsg->fast_iov; ret = __import_iovec(READ, uiov, iov_len, UIO_FASTIOV, - &iomsg->iov, &iomsg->msg.msg_iter, + &iomsg->free_iov, &iomsg->msg.msg_iter, false); if (ret > 0) ret = 0; @@ -4778,11 +4533,11 @@ static int __io_compat_recvmsg_copy_hdr(struct io_kiocb *req, if (clen < 0) return -EINVAL; sr->len = clen; - iomsg->iov[0].iov_len = clen; - iomsg->iov = NULL; + iomsg->free_iov = NULL; } else { + iomsg->free_iov = iomsg->fast_iov; ret = __import_iovec(READ, (struct iovec __user *)uiov, len, - UIO_FASTIOV, &iomsg->iov, + UIO_FASTIOV, &iomsg->free_iov, &iomsg->msg.msg_iter, true); if (ret < 0) return ret; @@ -4796,7 +4551,6 @@ static int io_recvmsg_copy_hdr(struct io_kiocb *req, struct io_async_msghdr *iomsg) { iomsg->msg.msg_name = &iomsg->addr; - iomsg->iov = iomsg->fast_iov; #ifdef CONFIG_COMPAT if (req->ctx->compat) @@ -4826,13 +4580,22 @@ static inline unsigned int io_put_recv_kbuf(struct io_kiocb *req) return io_put_kbuf(req, req->sr_msg.kbuf); } -static int io_recvmsg_prep(struct io_kiocb *req, - const struct io_uring_sqe *sqe) +static int io_recvmsg_prep_async(struct io_kiocb *req) { - struct io_async_msghdr *async_msg = req->async_data; - struct io_sr_msg *sr = &req->sr_msg; int ret; + if (!io_op_defs[req->opcode].needs_async_data) + return 0; + ret = io_recvmsg_copy_hdr(req, req->async_data); + if (!ret) + req->flags |= REQ_F_NEED_CLEANUP; + return ret; +} + +static int io_recvmsg_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) +{ + struct io_sr_msg *sr = &req->sr_msg; + if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL)) return -EINVAL; @@ -4845,36 +4608,25 @@ static int io_recvmsg_prep(struct io_kiocb *req, if (req->ctx->compat) sr->msg_flags |= MSG_CMSG_COMPAT; #endif - - if (!async_msg || !io_op_defs[req->opcode].needs_async_data) - return 0; - ret = io_recvmsg_copy_hdr(req, async_msg); - if (!ret) - req->flags |= REQ_F_NEED_CLEANUP; - return ret; + return 0; } -static int io_recvmsg(struct io_kiocb *req, bool force_nonblock, - struct io_comp_state *cs) +static int io_recvmsg(struct io_kiocb *req, unsigned int issue_flags) { struct io_async_msghdr iomsg, *kmsg; struct socket *sock; struct io_buffer *kbuf; unsigned flags; + int min_ret = 0; int ret, cflags = 0; + bool force_nonblock = issue_flags & IO_URING_F_NONBLOCK; sock = sock_from_file(req->file); if (unlikely(!sock)) return -ENOTSOCK; - if (req->async_data) { - kmsg = req->async_data; - kmsg->msg.msg_name = &kmsg->addr; - /* if iov is set, it's allocated already */ - if (!kmsg->iov) - kmsg->iov = kmsg->fast_iov; - kmsg->msg.msg_iter.iov = kmsg->iov; - } else { + kmsg = req->async_data; + if (!kmsg) { ret = io_recvmsg_copy_hdr(req, &iomsg); if (ret) return ret; @@ -4886,16 +4638,20 @@ static int io_recvmsg(struct io_kiocb *req, bool force_nonblock, if (IS_ERR(kbuf)) return PTR_ERR(kbuf); kmsg->fast_iov[0].iov_base = u64_to_user_ptr(kbuf->addr); - iov_iter_init(&kmsg->msg.msg_iter, READ, kmsg->iov, + kmsg->fast_iov[0].iov_len = req->sr_msg.len; + iov_iter_init(&kmsg->msg.msg_iter, READ, kmsg->fast_iov, 1, req->sr_msg.len); } - flags = req->sr_msg.msg_flags; + flags = req->sr_msg.msg_flags | MSG_NOSIGNAL; if (flags & MSG_DONTWAIT) req->flags |= REQ_F_NOWAIT; else if (force_nonblock) flags |= MSG_DONTWAIT; + if (flags & MSG_WAITALL) + min_ret = iov_iter_count(&kmsg->msg.msg_iter); + ret = __sys_recvmsg_sock(sock, &kmsg->msg, req->sr_msg.umsg, kmsg->uaddr, flags); if (force_nonblock && ret == -EAGAIN) @@ -4905,17 +4661,17 @@ static int io_recvmsg(struct io_kiocb *req, bool force_nonblock, if (req->flags & REQ_F_BUFFER_SELECTED) cflags = io_put_recv_kbuf(req); - if (kmsg->iov != kmsg->fast_iov) - kfree(kmsg->iov); + /* fast path, check for non-NULL to avoid function call */ + if (kmsg->free_iov) + kfree(kmsg->free_iov); req->flags &= ~REQ_F_NEED_CLEANUP; - if (ret < 0) + if (ret < min_ret || ((flags & MSG_WAITALL) && (kmsg->msg.msg_flags & (MSG_TRUNC | MSG_CTRUNC)))) req_set_fail_links(req); - __io_req_complete(req, ret, cflags, cs); + __io_req_complete(req, issue_flags, ret, cflags); return 0; } -static int io_recv(struct io_kiocb *req, bool force_nonblock, - struct io_comp_state *cs) +static int io_recv(struct io_kiocb *req, unsigned int issue_flags) { struct io_buffer *kbuf; struct io_sr_msg *sr = &req->sr_msg; @@ -4924,7 +4680,9 @@ static int io_recv(struct io_kiocb *req, bool force_nonblock, struct socket *sock; struct iovec iov; unsigned flags; + int min_ret = 0; int ret, cflags = 0; + bool force_nonblock = issue_flags & IO_URING_F_NONBLOCK; sock = sock_from_file(req->file); if (unlikely(!sock)) @@ -4948,12 +4706,15 @@ static int io_recv(struct io_kiocb *req, bool force_nonblock, msg.msg_iocb = NULL; msg.msg_flags = 0; - flags = req->sr_msg.msg_flags; + flags = req->sr_msg.msg_flags | MSG_NOSIGNAL; if (flags & MSG_DONTWAIT) req->flags |= REQ_F_NOWAIT; else if (force_nonblock) flags |= MSG_DONTWAIT; + if (flags & MSG_WAITALL) + min_ret = iov_iter_count(&msg.msg_iter); + ret = sock_recvmsg(sock, &msg, flags); if (force_nonblock && ret == -EAGAIN) return -EAGAIN; @@ -4962,9 +4723,9 @@ static int io_recv(struct io_kiocb *req, bool force_nonblock, out_free: if (req->flags & REQ_F_BUFFER_SELECTED) cflags = io_put_recv_kbuf(req); - if (ret < 0) + if (ret < min_ret || ((flags & MSG_WAITALL) && (msg.msg_flags & (MSG_TRUNC | MSG_CTRUNC)))) req_set_fail_links(req); - __io_req_complete(req, ret, cflags, cs); + __io_req_complete(req, issue_flags, ret, cflags); return 0; } @@ -4984,10 +4745,10 @@ static int io_accept_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) return 0; } -static int io_accept(struct io_kiocb *req, bool force_nonblock, - struct io_comp_state *cs) +static int io_accept(struct io_kiocb *req, unsigned int issue_flags) { struct io_accept *accept = &req->accept; + bool force_nonblock = issue_flags & IO_URING_F_NONBLOCK; unsigned int file_flags = force_nonblock ? O_NONBLOCK : 0; int ret; @@ -5004,14 +4765,21 @@ static int io_accept(struct io_kiocb *req, bool force_nonblock, ret = -EINTR; req_set_fail_links(req); } - __io_req_complete(req, ret, 0, cs); + __io_req_complete(req, issue_flags, ret, 0); return 0; } +static int io_connect_prep_async(struct io_kiocb *req) +{ + struct io_async_connect *io = req->async_data; + struct io_connect *conn = &req->connect; + + return move_addr_to_kernel(conn->addr, conn->addr_len, &io->address); +} + static int io_connect_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) { struct io_connect *conn = &req->connect; - struct io_async_connect *io = req->async_data; if (unlikely(req->ctx->flags & IORING_SETUP_IOPOLL)) return -EINVAL; @@ -5020,20 +4788,15 @@ static int io_connect_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) conn->addr = u64_to_user_ptr(READ_ONCE(sqe->addr)); conn->addr_len = READ_ONCE(sqe->addr2); - - if (!io) - return 0; - - return move_addr_to_kernel(conn->addr, conn->addr_len, - &io->address); + return 0; } -static int io_connect(struct io_kiocb *req, bool force_nonblock, - struct io_comp_state *cs) +static int io_connect(struct io_kiocb *req, unsigned int issue_flags) { struct io_async_connect __io, *io; unsigned file_flags; int ret; + bool force_nonblock = issue_flags & IO_URING_F_NONBLOCK; if (req->async_data) { io = req->async_data; @@ -5066,66 +4829,36 @@ static int io_connect(struct io_kiocb *req, bool force_nonblock, out: if (ret < 0) req_set_fail_links(req); - __io_req_complete(req, ret, 0, cs); + __io_req_complete(req, issue_flags, ret, 0); return 0; } #else /* !CONFIG_NET */ -static int io_sendmsg_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) -{ - return -EOPNOTSUPP; -} - -static int io_sendmsg(struct io_kiocb *req, bool force_nonblock, - struct io_comp_state *cs) -{ - return -EOPNOTSUPP; -} - -static int io_send(struct io_kiocb *req, bool force_nonblock, - struct io_comp_state *cs) -{ - return -EOPNOTSUPP; -} - -static int io_recvmsg_prep(struct io_kiocb *req, - const struct io_uring_sqe *sqe) -{ - return -EOPNOTSUPP; -} - -static int io_recvmsg(struct io_kiocb *req, bool force_nonblock, - struct io_comp_state *cs) -{ - return -EOPNOTSUPP; -} - -static int io_recv(struct io_kiocb *req, bool force_nonblock, - struct io_comp_state *cs) -{ - return -EOPNOTSUPP; -} - -static int io_accept_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) -{ - return -EOPNOTSUPP; -} - -static int io_accept(struct io_kiocb *req, bool force_nonblock, - struct io_comp_state *cs) -{ - return -EOPNOTSUPP; -} - -static int io_connect_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) -{ - return -EOPNOTSUPP; -} - -static int io_connect(struct io_kiocb *req, bool force_nonblock, - struct io_comp_state *cs) -{ - return -EOPNOTSUPP; -} +#define IO_NETOP_FN(op) \ +static int io_##op(struct io_kiocb *req, unsigned int issue_flags) \ +{ \ + return -EOPNOTSUPP; \ +} + +#define IO_NETOP_PREP(op) \ +IO_NETOP_FN(op) \ +static int io_##op##_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) \ +{ \ + return -EOPNOTSUPP; \ +} \ + +#define IO_NETOP_PREP_ASYNC(op) \ +IO_NETOP_PREP(op) \ +static int io_##op##_prep_async(struct io_kiocb *req) \ +{ \ + return -EOPNOTSUPP; \ +} + +IO_NETOP_PREP_ASYNC(sendmsg); +IO_NETOP_PREP_ASYNC(recvmsg); +IO_NETOP_PREP_ASYNC(connect); +IO_NETOP_PREP(accept); +IO_NETOP_FN(send); +IO_NETOP_FN(recv); #endif /* CONFIG_NET */ struct io_poll_table { @@ -5148,7 +4881,7 @@ static int __io_async_wake(struct io_kiocb *req, struct io_poll_iocb *poll, list_del_init(&poll->wait.entry); req->result = mask; - init_task_work(&req->task_work, func); + req->task_work.func = func; percpu_ref_get(&req->ctx->refs); /* @@ -5159,12 +4892,8 @@ static int __io_async_wake(struct io_kiocb *req, struct io_poll_iocb *poll, */ ret = io_req_task_work_add(req); if (unlikely(ret)) { - struct task_struct *tsk; - WRITE_ONCE(poll->canceled, true); - tsk = io_wq_get_task(req->ctx->io_wq); - task_work_add(tsk, &req->task_work, TWA_NONE); - wake_up_process(tsk); + io_req_task_work_add_fallback(req, func); } return 1; } @@ -5316,6 +5045,9 @@ static void __io_queue_proc(struct io_poll_iocb *poll, struct io_poll_table *pt, pt->error = -EINVAL; return; } + /* double add on the same waitqueue head, ignore */ + if (poll->head == head) + return; poll = kmalloc(sizeof(*poll), GFP_ATOMIC); if (!poll) { pt->error = -ENOMEM; @@ -5607,7 +5339,7 @@ static int io_poll_remove_prep(struct io_kiocb *req, * Find a running poll command that matches one specified in sqe->addr, * and remove it if found. */ -static int io_poll_remove(struct io_kiocb *req) +static int io_poll_remove(struct io_kiocb *req, unsigned int issue_flags) { struct io_ring_ctx *ctx = req->ctx; int ret; @@ -5658,7 +5390,7 @@ static int io_poll_add_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe return 0; } -static int io_poll_add(struct io_kiocb *req) +static int io_poll_add(struct io_kiocb *req, unsigned int issue_flags) { struct io_poll_iocb *poll = &req->poll; struct io_ring_ctx *ctx = req->ctx; @@ -5789,24 +5521,27 @@ static int io_timeout_remove_prep(struct io_kiocb *req, return 0; } +static inline enum hrtimer_mode io_translate_timeout_mode(unsigned int flags) +{ + return (flags & IORING_TIMEOUT_ABS) ? HRTIMER_MODE_ABS + : HRTIMER_MODE_REL; +} + /* * Remove or update an existing timeout command */ -static int io_timeout_remove(struct io_kiocb *req) +static int io_timeout_remove(struct io_kiocb *req, unsigned int issue_flags) { struct io_timeout_rem *tr = &req->timeout_rem; struct io_ring_ctx *ctx = req->ctx; int ret; spin_lock_irq(&ctx->completion_lock); - if (req->timeout_rem.flags & IORING_TIMEOUT_UPDATE) { - enum hrtimer_mode mode = (tr->flags & IORING_TIMEOUT_ABS) - ? HRTIMER_MODE_ABS : HRTIMER_MODE_REL; - - ret = io_timeout_update(ctx, tr->addr, &tr->ts, mode); - } else { + if (!(req->timeout_rem.flags & IORING_TIMEOUT_UPDATE)) ret = io_timeout_cancel(ctx, tr->addr); - } + else + ret = io_timeout_update(ctx, tr->addr, &tr->ts, + io_translate_timeout_mode(tr->flags)); io_cqring_fill_event(req, ret); io_commit_cqring(ctx); @@ -5846,16 +5581,13 @@ static int io_timeout_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe, if (get_timespec64(&data->ts, u64_to_user_ptr(sqe->addr))) return -EFAULT; - if (flags & IORING_TIMEOUT_ABS) - data->mode = HRTIMER_MODE_ABS; - else - data->mode = HRTIMER_MODE_REL; - + data->mode = io_translate_timeout_mode(flags); hrtimer_init(&data->timer, CLOCK_MONOTONIC, data->mode); + io_req_track_inflight(req); return 0; } -static int io_timeout(struct io_kiocb *req) +static int io_timeout(struct io_kiocb *req, unsigned int issue_flags) { struct io_ring_ctx *ctx = req->ctx; struct io_timeout_data *data = req->async_data; @@ -5905,19 +5637,30 @@ add: return 0; } +struct io_cancel_data { + struct io_ring_ctx *ctx; + u64 user_data; +}; + static bool io_cancel_cb(struct io_wq_work *work, void *data) { struct io_kiocb *req = container_of(work, struct io_kiocb, work); + struct io_cancel_data *cd = data; - return req->user_data == (unsigned long) data; + return req->ctx == cd->ctx && req->user_data == cd->user_data; } -static int io_async_cancel_one(struct io_ring_ctx *ctx, void *sqe_addr) +static int io_async_cancel_one(struct io_uring_task *tctx, u64 user_data, + struct io_ring_ctx *ctx) { + struct io_cancel_data data = { .ctx = ctx, .user_data = user_data, }; enum io_wq_cancel cancel_ret; int ret = 0; - cancel_ret = io_wq_cancel_cb(ctx->io_wq, io_cancel_cb, sqe_addr, false); + if (!tctx || !tctx->io_wq) + return -ENOENT; + + cancel_ret = io_wq_cancel_cb(tctx->io_wq, io_cancel_cb, &data, false); switch (cancel_ret) { case IO_WQ_CANCEL_OK: ret = 0; @@ -5940,7 +5683,7 @@ static void io_async_find_and_cancel(struct io_ring_ctx *ctx, unsigned long flags; int ret; - ret = io_async_cancel_one(ctx, (void *) (unsigned long) sqe_addr); + ret = io_async_cancel_one(req->task->io_uring, sqe_addr, ctx); if (ret != -ENOENT) { spin_lock_irqsave(&ctx->completion_lock, flags); goto done; @@ -5978,15 +5721,54 @@ static int io_async_cancel_prep(struct io_kiocb *req, return 0; } -static int io_async_cancel(struct io_kiocb *req) +static int io_async_cancel(struct io_kiocb *req, unsigned int issue_flags) { struct io_ring_ctx *ctx = req->ctx; + u64 sqe_addr = req->cancel.addr; + struct io_tctx_node *node; + int ret; + + /* tasks should wait for their io-wq threads, so safe w/o sync */ + ret = io_async_cancel_one(req->task->io_uring, sqe_addr, ctx); + spin_lock_irq(&ctx->completion_lock); + if (ret != -ENOENT) + goto done; + ret = io_timeout_cancel(ctx, sqe_addr); + if (ret != -ENOENT) + goto done; + ret = io_poll_cancel(ctx, sqe_addr); + if (ret != -ENOENT) + goto done; + spin_unlock_irq(&ctx->completion_lock); + + /* slow path, try all io-wq's */ + io_ring_submit_lock(ctx, !(issue_flags & IO_URING_F_NONBLOCK)); + ret = -ENOENT; + list_for_each_entry(node, &ctx->tctx_list, ctx_node) { + struct io_uring_task *tctx = node->task->io_uring; + + if (!tctx || !tctx->io_wq) + continue; + ret = io_async_cancel_one(tctx, req->cancel.addr, ctx); + if (ret != -ENOENT) + break; + } + io_ring_submit_unlock(ctx, !(issue_flags & IO_URING_F_NONBLOCK)); - io_async_find_and_cancel(ctx, req, req->cancel.addr, 0); + spin_lock_irq(&ctx->completion_lock); +done: + io_cqring_fill_event(req, ret); + io_commit_cqring(ctx); + spin_unlock_irq(&ctx->completion_lock); + io_cqring_ev_posted(ctx); + + if (ret < 0) + req_set_fail_links(req); + io_put_req(req); return 0; } -static int io_files_update_prep(struct io_kiocb *req, +static int io_rsrc_update_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) { if (unlikely(req->ctx->flags & IORING_SETUP_SQPOLL)) @@ -5996,34 +5778,33 @@ static int io_files_update_prep(struct io_kiocb *req, if (sqe->ioprio || sqe->rw_flags) return -EINVAL; - req->files_update.offset = READ_ONCE(sqe->off); - req->files_update.nr_args = READ_ONCE(sqe->len); - if (!req->files_update.nr_args) + req->rsrc_update.offset = READ_ONCE(sqe->off); + req->rsrc_update.nr_args = READ_ONCE(sqe->len); + if (!req->rsrc_update.nr_args) return -EINVAL; - req->files_update.arg = READ_ONCE(sqe->addr); + req->rsrc_update.arg = READ_ONCE(sqe->addr); return 0; } -static int io_files_update(struct io_kiocb *req, bool force_nonblock, - struct io_comp_state *cs) +static int io_files_update(struct io_kiocb *req, unsigned int issue_flags) { struct io_ring_ctx *ctx = req->ctx; - struct io_uring_files_update up; + struct io_uring_rsrc_update up; int ret; - if (force_nonblock) + if (issue_flags & IO_URING_F_NONBLOCK) return -EAGAIN; - up.offset = req->files_update.offset; - up.fds = req->files_update.arg; + up.offset = req->rsrc_update.offset; + up.data = req->rsrc_update.arg; mutex_lock(&ctx->uring_lock); - ret = __io_sqe_files_update(ctx, &up, req->files_update.nr_args); + ret = __io_sqe_files_update(ctx, &up, req->rsrc_update.nr_args); mutex_unlock(&ctx->uring_lock); if (ret < 0) req_set_fail_links(req); - __io_req_complete(req, ret, 0, cs); + __io_req_complete(req, issue_flags, ret, 0); return 0; } @@ -6045,9 +5826,9 @@ static int io_req_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) case IORING_OP_POLL_REMOVE: return io_poll_remove_prep(req, sqe); case IORING_OP_FSYNC: - return io_prep_fsync(req, sqe); + return io_fsync_prep(req, sqe); case IORING_OP_SYNC_FILE_RANGE: - return io_prep_sfr(req, sqe); + return io_sfr_prep(req, sqe); case IORING_OP_SENDMSG: case IORING_OP_SEND: return io_sendmsg_prep(req, sqe); @@ -6073,7 +5854,7 @@ static int io_req_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) case IORING_OP_CLOSE: return io_close_prep(req, sqe); case IORING_OP_FILES_UPDATE: - return io_files_update_prep(req, sqe); + return io_rsrc_update_prep(req, sqe); case IORING_OP_STATX: return io_statx_prep(req, sqe); case IORING_OP_FADVISE: @@ -6105,14 +5886,39 @@ static int io_req_prep(struct io_kiocb *req, const struct io_uring_sqe *sqe) return-EINVAL; } -static int io_req_defer_prep(struct io_kiocb *req, - const struct io_uring_sqe *sqe) +static int io_req_prep_async(struct io_kiocb *req) { - if (!sqe) + switch (req->opcode) { + case IORING_OP_READV: + case IORING_OP_READ_FIXED: + case IORING_OP_READ: + return io_rw_prep_async(req, READ); + case IORING_OP_WRITEV: + case IORING_OP_WRITE_FIXED: + case IORING_OP_WRITE: + return io_rw_prep_async(req, WRITE); + case IORING_OP_SENDMSG: + case IORING_OP_SEND: + return io_sendmsg_prep_async(req); + case IORING_OP_RECVMSG: + case IORING_OP_RECV: + return io_recvmsg_prep_async(req); + case IORING_OP_CONNECT: + return io_connect_prep_async(req); + } + return 0; +} + +static int io_req_defer_prep(struct io_kiocb *req) +{ + if (!io_op_defs[req->opcode].needs_async_data) + return 0; + /* some opcodes init it during the inital prep */ + if (req->async_data) return 0; - if (io_alloc_async_data(req)) + if (__io_alloc_async_data(req)) return -EAGAIN; - return io_req_prep(req, sqe); + return io_req_prep_async(req); } static u32 io_get_sequence(struct io_kiocb *req) @@ -6128,7 +5934,7 @@ static u32 io_get_sequence(struct io_kiocb *req) return total_submitted - nr_reqs; } -static int io_req_defer(struct io_kiocb *req, const struct io_uring_sqe *sqe) +static int io_req_defer(struct io_kiocb *req) { struct io_ring_ctx *ctx = req->ctx; struct io_defer_entry *de; @@ -6145,11 +5951,9 @@ static int io_req_defer(struct io_kiocb *req, const struct io_uring_sqe *sqe) if (!req_need_defer(req, seq) && list_empty_careful(&ctx->defer_list)) return 0; - if (!req->async_data) { - ret = io_req_defer_prep(req, sqe); - if (ret) - return ret; - } + ret = io_req_defer_prep(req); + if (ret) + return ret; io_prep_async_link(req); de = kmalloc(sizeof(*de), GFP_KERNEL); if (!de) @@ -6171,25 +5975,6 @@ static int io_req_defer(struct io_kiocb *req, const struct io_uring_sqe *sqe) return -EIOCBQUEUED; } -static void io_req_drop_files(struct io_kiocb *req) -{ - struct io_ring_ctx *ctx = req->ctx; - struct io_uring_task *tctx = req->task->io_uring; - unsigned long flags; - - if (req->work.flags & IO_WQ_WORK_FILES) { - put_files_struct(req->work.identity->files); - put_nsproxy(req->work.identity->nsproxy); - } - spin_lock_irqsave(&ctx->inflight_lock, flags); - list_del(&req->inflight_entry); - spin_unlock_irqrestore(&ctx->inflight_lock, flags); - req->flags &= ~REQ_F_INFLIGHT; - req->work.flags &= ~IO_WQ_WORK_FILES; - if (atomic_read(&tctx->in_idle)) - wake_up(&tctx->wait); -} - static void __io_clean_op(struct io_kiocb *req) { if (req->flags & REQ_F_BUFFER_SELECTED) { @@ -6223,8 +6008,8 @@ static void __io_clean_op(struct io_kiocb *req) case IORING_OP_RECVMSG: case IORING_OP_SENDMSG: { struct io_async_msghdr *io = req->async_data; - if (io->iov != io->fast_iov) - kfree(io->iov); + + kfree(io->free_iov); break; } case IORING_OP_SPLICE: @@ -6249,118 +6034,124 @@ static void __io_clean_op(struct io_kiocb *req) } } -static int io_issue_sqe(struct io_kiocb *req, bool force_nonblock, - struct io_comp_state *cs) +static int io_issue_sqe(struct io_kiocb *req, unsigned int issue_flags) { struct io_ring_ctx *ctx = req->ctx; + const struct cred *creds = NULL; int ret; + if (req->work.creds && req->work.creds != current_cred()) + creds = override_creds(req->work.creds); + switch (req->opcode) { case IORING_OP_NOP: - ret = io_nop(req, cs); + ret = io_nop(req, issue_flags); break; case IORING_OP_READV: case IORING_OP_READ_FIXED: case IORING_OP_READ: - ret = io_read(req, force_nonblock, cs); + ret = io_read(req, issue_flags); break; case IORING_OP_WRITEV: case IORING_OP_WRITE_FIXED: case IORING_OP_WRITE: - ret = io_write(req, force_nonblock, cs); + ret = io_write(req, issue_flags); break; case IORING_OP_FSYNC: - ret = io_fsync(req, force_nonblock); + ret = io_fsync(req, issue_flags); break; case IORING_OP_POLL_ADD: - ret = io_poll_add(req); + ret = io_poll_add(req, issue_flags); break; case IORING_OP_POLL_REMOVE: - ret = io_poll_remove(req); + ret = io_poll_remove(req, issue_flags); break; case IORING_OP_SYNC_FILE_RANGE: - ret = io_sync_file_range(req, force_nonblock); + ret = io_sync_file_range(req, issue_flags); break; case IORING_OP_SENDMSG: - ret = io_sendmsg(req, force_nonblock, cs); + ret = io_sendmsg(req, issue_flags); break; case IORING_OP_SEND: - ret = io_send(req, force_nonblock, cs); + ret = io_send(req, issue_flags); break; case IORING_OP_RECVMSG: - ret = io_recvmsg(req, force_nonblock, cs); + ret = io_recvmsg(req, issue_flags); break; case IORING_OP_RECV: - ret = io_recv(req, force_nonblock, cs); + ret = io_recv(req, issue_flags); break; case IORING_OP_TIMEOUT: - ret = io_timeout(req); + ret = io_timeout(req, issue_flags); break; case IORING_OP_TIMEOUT_REMOVE: - ret = io_timeout_remove(req); + ret = io_timeout_remove(req, issue_flags); break; case IORING_OP_ACCEPT: - ret = io_accept(req, force_nonblock, cs); + ret = io_accept(req, issue_flags); break; case IORING_OP_CONNECT: - ret = io_connect(req, force_nonblock, cs); + ret = io_connect(req, issue_flags); break; case IORING_OP_ASYNC_CANCEL: - ret = io_async_cancel(req); + ret = io_async_cancel(req, issue_flags); break; case IORING_OP_FALLOCATE: - ret = io_fallocate(req, force_nonblock); + ret = io_fallocate(req, issue_flags); break; case IORING_OP_OPENAT: - ret = io_openat(req, force_nonblock); + ret = io_openat(req, issue_flags); break; case IORING_OP_CLOSE: - ret = io_close(req, force_nonblock, cs); + ret = io_close(req, issue_flags); break; case IORING_OP_FILES_UPDATE: - ret = io_files_update(req, force_nonblock, cs); + ret = io_files_update(req, issue_flags); break; case IORING_OP_STATX: - ret = io_statx(req, force_nonblock); + ret = io_statx(req, issue_flags); break; case IORING_OP_FADVISE: - ret = io_fadvise(req, force_nonblock); + ret = io_fadvise(req, issue_flags); break; case IORING_OP_MADVISE: - ret = io_madvise(req, force_nonblock); + ret = io_madvise(req, issue_flags); break; case IORING_OP_OPENAT2: - ret = io_openat2(req, force_nonblock); + ret = io_openat2(req, issue_flags); break; case IORING_OP_EPOLL_CTL: - ret = io_epoll_ctl(req, force_nonblock, cs); + ret = io_epoll_ctl(req, issue_flags); break; case IORING_OP_SPLICE: - ret = io_splice(req, force_nonblock); + ret = io_splice(req, issue_flags); break; case IORING_OP_PROVIDE_BUFFERS: - ret = io_provide_buffers(req, force_nonblock, cs); + ret = io_provide_buffers(req, issue_flags); break; case IORING_OP_REMOVE_BUFFERS: - ret = io_remove_buffers(req, force_nonblock, cs); + ret = io_remove_buffers(req, issue_flags); break; case IORING_OP_TEE: - ret = io_tee(req, force_nonblock); + ret = io_tee(req, issue_flags); break; case IORING_OP_SHUTDOWN: - ret = io_shutdown(req, force_nonblock); + ret = io_shutdown(req, issue_flags); break; case IORING_OP_RENAMEAT: - ret = io_renameat(req, force_nonblock); + ret = io_renameat(req, issue_flags); break; case IORING_OP_UNLINKAT: - ret = io_unlinkat(req, force_nonblock); + ret = io_unlinkat(req, issue_flags); break; default: ret = -EINVAL; break; } + if (creds) + revert_creds(creds); + if (ret) return ret; @@ -6381,7 +6172,7 @@ static int io_issue_sqe(struct io_kiocb *req, bool force_nonblock, return 0; } -static struct io_wq_work *io_wq_submit_work(struct io_wq_work *work) +static void io_wq_submit_work(struct io_wq_work *work) { struct io_kiocb *req = container_of(work, struct io_kiocb, work); struct io_kiocb *timeout; @@ -6391,15 +6182,12 @@ static struct io_wq_work *io_wq_submit_work(struct io_wq_work *work) if (timeout) io_queue_linked_timeout(timeout); - /* if NO_CANCEL is set, we must still run the work */ - if ((work->flags & (IO_WQ_WORK_CANCEL|IO_WQ_WORK_NO_CANCEL)) == - IO_WQ_WORK_CANCEL) { + if (work->flags & IO_WQ_WORK_CANCEL) ret = -ECANCELED; - } if (!ret) { do { - ret = io_issue_sqe(req, false, NULL); + ret = io_issue_sqe(req, 0); /* * We can get EAGAIN for polled IO even though we're * forcing a sync submission from here, since we can't @@ -6411,38 +6199,18 @@ static struct io_wq_work *io_wq_submit_work(struct io_wq_work *work) } while (1); } + /* avoid locking problems by failing it from a clean context */ if (ret) { - struct io_ring_ctx *lock_ctx = NULL; - - if (req->ctx->flags & IORING_SETUP_IOPOLL) - lock_ctx = req->ctx; - - /* - * io_iopoll_complete() does not hold completion_lock to - * complete polled io, so here for polled io, we can not call - * io_req_complete() directly, otherwise there maybe concurrent - * access to cqring, defer_list, etc, which is not safe. Given - * that io_iopoll_complete() is always called under uring_lock, - * so here for polled io, we also get uring_lock to complete - * it. - */ - if (lock_ctx) - mutex_lock(&lock_ctx->uring_lock); - - req_set_fail_links(req); - io_req_complete(req, ret); - - if (lock_ctx) - mutex_unlock(&lock_ctx->uring_lock); + /* io-wq is going to take one down */ + refcount_inc(&req->refs); + io_req_task_queue_fail(req, ret); } - - return io_steal_work(req); } static inline struct file *io_file_from_index(struct io_ring_ctx *ctx, int index) { - struct fixed_file_table *table; + struct fixed_rsrc_table *table; table = &ctx->file_data->table[index >> IORING_FILE_TABLE_SHIFT]; return table->files[index & IORING_FILE_TABLE_MASK]; @@ -6465,16 +6233,8 @@ static struct file *io_file_get(struct io_submit_state *state, file = __io_file_get(state, fd); } - if (file && file->f_op == &io_uring_fops && - !(req->flags & REQ_F_INFLIGHT)) { - io_req_init_async(req); - req->flags |= REQ_F_INFLIGHT; - - spin_lock_irq(&ctx->inflight_lock); - list_add(&req->inflight_entry, &ctx->inflight_list); - spin_unlock_irq(&ctx->inflight_lock); - } - + if (file && unlikely(file->f_op == &io_uring_fops)) + io_req_track_inflight(req); return file; } @@ -6501,11 +6261,11 @@ static enum hrtimer_restart io_link_timeout_fn(struct hrtimer *timer) spin_unlock_irqrestore(&ctx->completion_lock, flags); if (prev) { - req_set_fail_links(prev); io_async_find_and_cancel(ctx, req, prev->user_data, -ETIME); - io_put_req(prev); + io_put_req_deferred(prev, 1); } else { - io_req_complete(req, -ETIME); + io_req_complete_post(req, -ETIME, 0); + io_put_req_deferred(req, 1); } return HRTIMER_NORESTART; } @@ -6551,27 +6311,12 @@ static struct io_kiocb *io_prep_linked_timeout(struct io_kiocb *req) return nxt; } -static void __io_queue_sqe(struct io_kiocb *req, struct io_comp_state *cs) +static void __io_queue_sqe(struct io_kiocb *req) { - struct io_kiocb *linked_timeout; - const struct cred *old_creds = NULL; + struct io_kiocb *linked_timeout = io_prep_linked_timeout(req); int ret; -again: - linked_timeout = io_prep_linked_timeout(req); - - if ((req->flags & REQ_F_WORK_INITIALIZED) && - (req->work.flags & IO_WQ_WORK_CREDS) && - req->work.identity->creds != current_cred()) { - if (old_creds) - revert_creds(old_creds); - if (old_creds == req->work.identity->creds) - old_creds = NULL; /* restored original creds */ - else - old_creds = override_creds(req->work.identity->creds); - } - - ret = io_issue_sqe(req, true, cs); + ret = io_issue_sqe(req, IO_URING_F_NONBLOCK|IO_URING_F_COMPLETE_DEFER); /* * We async punt it if the file wasn't marked NOWAIT, or if the file @@ -6585,38 +6330,32 @@ again: */ io_queue_async_work(req); } - - if (linked_timeout) - io_queue_linked_timeout(linked_timeout); } else if (likely(!ret)) { /* drop submission reference */ - req = io_put_req_find_next(req); - if (linked_timeout) - io_queue_linked_timeout(linked_timeout); + if (req->flags & REQ_F_COMPLETE_INLINE) { + struct io_ring_ctx *ctx = req->ctx; + struct io_comp_state *cs = &ctx->submit_state.comp; - if (req) { - if (!(req->flags & REQ_F_FORCE_ASYNC)) - goto again; - io_queue_async_work(req); + cs->reqs[cs->nr++] = req; + if (cs->nr == ARRAY_SIZE(cs->reqs)) + io_submit_flush_completions(cs, ctx); + } else { + io_put_req(req); } } else { - /* un-prep timeout, so it'll be killed as any other linked */ - req->flags &= ~REQ_F_LINK_TIMEOUT; req_set_fail_links(req); io_put_req(req); io_req_complete(req, ret); } - - if (old_creds) - revert_creds(old_creds); + if (linked_timeout) + io_queue_linked_timeout(linked_timeout); } -static void io_queue_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe, - struct io_comp_state *cs) +static void io_queue_sqe(struct io_kiocb *req) { int ret; - ret = io_req_defer(req, sqe); + ret = io_req_defer(req); if (ret) { if (ret != -EIOCBQUEUED) { fail_req: @@ -6625,43 +6364,140 @@ fail_req: io_req_complete(req, ret); } } else if (req->flags & REQ_F_FORCE_ASYNC) { - if (!req->async_data) { - ret = io_req_defer_prep(req, sqe); - if (unlikely(ret)) - goto fail_req; - } + ret = io_req_defer_prep(req); + if (unlikely(ret)) + goto fail_req; io_queue_async_work(req); } else { - if (sqe) { - ret = io_req_prep(req, sqe); - if (unlikely(ret)) - goto fail_req; - } - __io_queue_sqe(req, cs); + __io_queue_sqe(req); } } -static inline void io_queue_link_head(struct io_kiocb *req, - struct io_comp_state *cs) +/* + * Check SQE restrictions (opcode and flags). + * + * Returns 'true' if SQE is allowed, 'false' otherwise. + */ +static inline bool io_check_restriction(struct io_ring_ctx *ctx, + struct io_kiocb *req, + unsigned int sqe_flags) { - if (unlikely(req->flags & REQ_F_FAIL_LINK)) { - io_put_req(req); - io_req_complete(req, -ECANCELED); - } else - io_queue_sqe(req, NULL, cs); + if (!ctx->restricted) + return true; + + if (!test_bit(req->opcode, ctx->restrictions.sqe_op)) + return false; + + if ((sqe_flags & ctx->restrictions.sqe_flags_required) != + ctx->restrictions.sqe_flags_required) + return false; + + if (sqe_flags & ~(ctx->restrictions.sqe_flags_allowed | + ctx->restrictions.sqe_flags_required)) + return false; + + return true; } -struct io_submit_link { - struct io_kiocb *head; - struct io_kiocb *last; -}; +static int io_init_req(struct io_ring_ctx *ctx, struct io_kiocb *req, + const struct io_uring_sqe *sqe) +{ + struct io_submit_state *state; + unsigned int sqe_flags; + int personality, ret = 0; -static int io_submit_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe, - struct io_submit_link *link, struct io_comp_state *cs) + req->opcode = READ_ONCE(sqe->opcode); + /* same numerical values with corresponding REQ_F_*, safe to copy */ + req->flags = sqe_flags = READ_ONCE(sqe->flags); + req->user_data = READ_ONCE(sqe->user_data); + req->async_data = NULL; + req->file = NULL; + req->ctx = ctx; + req->link = NULL; + req->fixed_rsrc_refs = NULL; + /* one is dropped after submission, the other at completion */ + refcount_set(&req->refs, 2); + req->task = current; + req->result = 0; + req->work.list.next = NULL; + req->work.creds = NULL; + req->work.flags = 0; + + /* enforce forwards compatibility on users */ + if (unlikely(sqe_flags & ~SQE_VALID_FLAGS)) { + req->flags = 0; + return -EINVAL; + } + + if (unlikely(req->opcode >= IORING_OP_LAST)) + return -EINVAL; + + if (unlikely(!io_check_restriction(ctx, req, sqe_flags))) + return -EACCES; + + if ((sqe_flags & IOSQE_BUFFER_SELECT) && + !io_op_defs[req->opcode].buffer_select) + return -EOPNOTSUPP; + + personality = READ_ONCE(sqe->personality); + if (personality) { + req->work.creds = xa_load(&ctx->personalities, personality); + if (!req->work.creds) + return -EINVAL; + get_cred(req->work.creds); + } + state = &ctx->submit_state; + + /* + * Plug now if we have more than 1 IO left after this, and the target + * is potentially a read/write to block based storage. + */ + if (!state->plug_started && state->ios_left > 1 && + io_op_defs[req->opcode].plug) { + blk_start_plug(&state->plug); + state->plug_started = true; + } + + if (io_op_defs[req->opcode].needs_file) { + bool fixed = req->flags & REQ_F_FIXED_FILE; + + req->file = io_file_get(state, req, READ_ONCE(sqe->fd), fixed); + if (unlikely(!req->file)) + ret = -EBADF; + } + + state->ios_left--; + return ret; +} + +static int io_submit_sqe(struct io_ring_ctx *ctx, struct io_kiocb *req, + const struct io_uring_sqe *sqe) { - struct io_ring_ctx *ctx = req->ctx; + struct io_submit_link *link = &ctx->submit_state.link; int ret; + ret = io_init_req(ctx, req, sqe); + if (unlikely(ret)) { +fail_req: + io_put_req(req); + io_req_complete(req, ret); + if (link->head) { + /* fail even hard links since we don't submit */ + link->head->flags |= REQ_F_FAIL_LINK; + io_put_req(link->head); + io_req_complete(link->head, -ECANCELED); + link->head = NULL; + } + return ret; + } + ret = io_req_prep(req, sqe); + if (unlikely(ret)) + goto fail_req; + + /* don't need @sqe from now on */ + trace_io_uring_submit_sqe(ctx, req->opcode, req->user_data, + true, ctx->flags & IORING_SETUP_SQPOLL); + /* * If we already have a head request, queue this one for async * submittal once the head completes. If we don't have a head but @@ -6683,19 +6519,16 @@ static int io_submit_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe, head->flags |= REQ_F_IO_DRAIN; ctx->drain_next = 1; } - ret = io_req_defer_prep(req, sqe); - if (unlikely(ret)) { - /* fail even hard links since we don't submit */ - head->flags |= REQ_F_FAIL_LINK; - return ret; - } + ret = io_req_defer_prep(req); + if (unlikely(ret)) + goto fail_req; trace_io_uring_link(ctx, req, head); link->last->link = req; link->last = req; /* last request of a link, enqueue the link */ if (!(req->flags & (REQ_F_LINK | REQ_F_HARDLINK))) { - io_queue_link_head(head, cs); + io_queue_sqe(head); link->head = NULL; } } else { @@ -6704,13 +6537,10 @@ static int io_submit_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe, ctx->drain_next = 0; } if (req->flags & (REQ_F_LINK | REQ_F_HARDLINK)) { - ret = io_req_defer_prep(req, sqe); - if (unlikely(ret)) - req->flags |= REQ_F_FAIL_LINK; link->head = req; link->last = req; } else { - io_queue_sqe(req, sqe, cs); + io_queue_sqe(req); } } @@ -6720,30 +6550,28 @@ static int io_submit_sqe(struct io_kiocb *req, const struct io_uring_sqe *sqe, /* * Batched submission is done, ensure local IO is flushed out. */ -static void io_submit_state_end(struct io_submit_state *state) +static void io_submit_state_end(struct io_submit_state *state, + struct io_ring_ctx *ctx) { - if (!list_empty(&state->comp.list)) - io_submit_flush_completions(&state->comp); + if (state->link.head) + io_queue_sqe(state->link.head); + if (state->comp.nr) + io_submit_flush_completions(&state->comp, ctx); if (state->plug_started) blk_finish_plug(&state->plug); io_state_file_put(state); - if (state->free_reqs) - kmem_cache_free_bulk(req_cachep, state->free_reqs, state->reqs); } /* * Start submission side cache. */ static void io_submit_state_start(struct io_submit_state *state, - struct io_ring_ctx *ctx, unsigned int max_ios) + unsigned int max_ios) { state->plug_started = false; - state->comp.nr = 0; - INIT_LIST_HEAD(&state->comp.list); - state->comp.ctx = ctx; - state->free_reqs = 0; - state->file_refs = 0; state->ios_left = max_ios; + /* set only head, no need to init link_last in advance */ + state->link.head = NULL; } static void io_commit_sqring(struct io_ring_ctx *ctx) @@ -6779,7 +6607,7 @@ static const struct io_uring_sqe *io_get_sqe(struct io_ring_ctx *ctx) * 2) allows the kernel side to track the head on its own, even * though the application is the one updating it. */ - head = READ_ONCE(sq_array[ctx->cached_sq_head & ctx->sq_mask]); + head = READ_ONCE(sq_array[ctx->cached_sq_head++ & ctx->sq_mask]); if (likely(head < ctx->sq_entries)) return &ctx->sq_sqes[head]; @@ -6789,126 +6617,9 @@ static const struct io_uring_sqe *io_get_sqe(struct io_ring_ctx *ctx) return NULL; } -static inline void io_consume_sqe(struct io_ring_ctx *ctx) -{ - ctx->cached_sq_head++; -} - -/* - * Check SQE restrictions (opcode and flags). - * - * Returns 'true' if SQE is allowed, 'false' otherwise. - */ -static inline bool io_check_restriction(struct io_ring_ctx *ctx, - struct io_kiocb *req, - unsigned int sqe_flags) -{ - if (!ctx->restricted) - return true; - - if (!test_bit(req->opcode, ctx->restrictions.sqe_op)) - return false; - - if ((sqe_flags & ctx->restrictions.sqe_flags_required) != - ctx->restrictions.sqe_flags_required) - return false; - - if (sqe_flags & ~(ctx->restrictions.sqe_flags_allowed | - ctx->restrictions.sqe_flags_required)) - return false; - - return true; -} - -#define SQE_VALID_FLAGS (IOSQE_FIXED_FILE|IOSQE_IO_DRAIN|IOSQE_IO_LINK| \ - IOSQE_IO_HARDLINK | IOSQE_ASYNC | \ - IOSQE_BUFFER_SELECT) - -static int io_init_req(struct io_ring_ctx *ctx, struct io_kiocb *req, - const struct io_uring_sqe *sqe, - struct io_submit_state *state) -{ - unsigned int sqe_flags; - int id, ret; - - req->opcode = READ_ONCE(sqe->opcode); - req->user_data = READ_ONCE(sqe->user_data); - req->async_data = NULL; - req->file = NULL; - req->ctx = ctx; - req->flags = 0; - req->link = NULL; - req->fixed_file_refs = NULL; - /* one is dropped after submission, the other at completion */ - refcount_set(&req->refs, 2); - req->task = current; - req->result = 0; - - if (unlikely(req->opcode >= IORING_OP_LAST)) - return -EINVAL; - - if (unlikely(io_sq_thread_acquire_mm_files(ctx, req))) - return -EFAULT; - - sqe_flags = READ_ONCE(sqe->flags); - /* enforce forwards compatibility on users */ - if (unlikely(sqe_flags & ~SQE_VALID_FLAGS)) - return -EINVAL; - - if (unlikely(!io_check_restriction(ctx, req, sqe_flags))) - return -EACCES; - - if ((sqe_flags & IOSQE_BUFFER_SELECT) && - !io_op_defs[req->opcode].buffer_select) - return -EOPNOTSUPP; - - id = READ_ONCE(sqe->personality); - if (id) { - struct io_identity *iod; - - iod = idr_find(&ctx->personality_idr, id); - if (unlikely(!iod)) - return -EINVAL; - refcount_inc(&iod->count); - - __io_req_init_async(req); - get_cred(iod->creds); - req->work.identity = iod; - req->work.flags |= IO_WQ_WORK_CREDS; - } - - /* same numerical values with corresponding REQ_F_*, safe to copy */ - req->flags |= sqe_flags; - - /* - * Plug now if we have more than 1 IO left after this, and the target - * is potentially a read/write to block based storage. - */ - if (!state->plug_started && state->ios_left > 1 && - io_op_defs[req->opcode].plug) { - blk_start_plug(&state->plug); - state->plug_started = true; - } - - ret = 0; - if (io_op_defs[req->opcode].needs_file) { - bool fixed = req->flags & REQ_F_FIXED_FILE; - - req->file = io_file_get(state, req, READ_ONCE(sqe->fd), fixed); - if (unlikely(!req->file && - !io_op_defs[req->opcode].needs_file_no_error)) - ret = -EBADF; - } - - state->ios_left--; - return ret; -} - static int io_submit_sqes(struct io_ring_ctx *ctx, unsigned int nr) { - struct io_submit_state state; - struct io_submit_link link; - int i, submitted = 0; + int submitted = 0; /* if we have a backlog and couldn't flush it all, return BUSY */ if (test_bit(0, &ctx->sq_check_overflow)) { @@ -6924,43 +6635,27 @@ static int io_submit_sqes(struct io_ring_ctx *ctx, unsigned int nr) percpu_counter_add(¤t->io_uring->inflight, nr); refcount_add(nr, ¤t->usage); + io_submit_state_start(&ctx->submit_state, nr); - io_submit_state_start(&state, ctx, nr); - link.head = NULL; - - for (i = 0; i < nr; i++) { + while (submitted < nr) { const struct io_uring_sqe *sqe; struct io_kiocb *req; - int err; - sqe = io_get_sqe(ctx); - if (unlikely(!sqe)) { - io_consume_sqe(ctx); - break; - } - req = io_alloc_req(ctx, &state); + req = io_alloc_req(ctx); if (unlikely(!req)) { if (!submitted) submitted = -EAGAIN; break; } - io_consume_sqe(ctx); + sqe = io_get_sqe(ctx); + if (unlikely(!sqe)) { + kmem_cache_free(req_cachep, req); + break; + } /* will complete beyond this point, count as submitted */ submitted++; - - err = io_init_req(ctx, req, sqe, &state); - if (unlikely(err)) { -fail_req: - io_put_req(req); - io_req_complete(req, err); + if (io_submit_sqe(ctx, req, sqe)) break; - } - - trace_io_uring_submit_sqe(ctx, req->opcode, req->user_data, - true, io_async_submit(ctx)); - err = io_submit_sqe(req, sqe, &link, &state.comp); - if (err) - goto fail_req; } if (unlikely(submitted != nr)) { @@ -6972,10 +6667,8 @@ fail_req: percpu_counter_sub(&tctx->inflight, unused); put_task_struct_many(current, unused); } - if (link.head) - io_queue_link_head(link.head, &state.comp); - io_submit_state_end(&state); + io_submit_state_end(&ctx->submit_state, ctx); /* Commit SQ ring head once we've consumed and submitted all SQEs */ io_commit_sqring(ctx); @@ -7014,8 +6707,8 @@ static int __io_sq_thread(struct io_ring_ctx *ctx, bool cap_entries) if (!list_empty(&ctx->iopoll_list)) io_do_iopoll(ctx, &nr_events, 0); - if (to_submit && !ctx->sqo_dead && - likely(!percpu_ref_is_dying(&ctx->refs))) + if (to_submit && likely(!percpu_ref_is_dying(&ctx->refs)) && + !(ctx->flags & IORING_SETUP_R_DISABLED)) ret = io_submit_sqes(ctx, to_submit); mutex_unlock(&ctx->uring_lock); } @@ -7039,93 +6732,62 @@ static void io_sqd_update_thread_idle(struct io_sq_data *sqd) sqd->sq_thread_idle = sq_thread_idle; } -static void io_sqd_init_new(struct io_sq_data *sqd) -{ - struct io_ring_ctx *ctx; - - while (!list_empty(&sqd->ctx_new_list)) { - ctx = list_first_entry(&sqd->ctx_new_list, struct io_ring_ctx, sqd_list); - list_move_tail(&ctx->sqd_list, &sqd->ctx_list); - complete(&ctx->sq_thread_comp); - } - - io_sqd_update_thread_idle(sqd); -} - static int io_sq_thread(void *data) { - struct cgroup_subsys_state *cur_css = NULL; - struct files_struct *old_files = current->files; - struct nsproxy *old_nsproxy = current->nsproxy; - const struct cred *old_cred = NULL; struct io_sq_data *sqd = data; struct io_ring_ctx *ctx; unsigned long timeout = 0; + char buf[TASK_COMM_LEN]; DEFINE_WAIT(wait); - task_lock(current); - current->files = NULL; - current->nsproxy = NULL; - task_unlock(current); + sprintf(buf, "iou-sqp-%d", sqd->task_pid); + set_task_comm(current, buf); + current->pf_io_worker = NULL; + + if (sqd->sq_cpu != -1) + set_cpus_allowed_ptr(current, cpumask_of(sqd->sq_cpu)); + else + set_cpus_allowed_ptr(current, cpu_online_mask); + current->flags |= PF_NO_SETAFFINITY; - while (!kthread_should_stop()) { + mutex_lock(&sqd->lock); + while (!test_bit(IO_SQ_THREAD_SHOULD_STOP, &sqd->state)) { int ret; bool cap_entries, sqt_spin, needs_sched; - /* - * Any changes to the sqd lists are synchronized through the - * kthread parking. This synchronizes the thread vs users, - * the users are synchronized on the sqd->ctx_lock. - */ - if (kthread_should_park()) { - kthread_parkme(); - /* - * When sq thread is unparked, in case the previous park operation - * comes from io_put_sq_data(), which means that sq thread is going - * to be stopped, so here needs to have a check. - */ - if (kthread_should_stop()) - break; - } - - if (unlikely(!list_empty(&sqd->ctx_new_list))) { - io_sqd_init_new(sqd); + if (test_bit(IO_SQ_THREAD_SHOULD_PARK, &sqd->state)) { + mutex_unlock(&sqd->lock); + cond_resched(); + mutex_lock(&sqd->lock); + io_run_task_work(); + io_run_task_work_head(&sqd->park_task_work); timeout = jiffies + sqd->sq_thread_idle; + continue; } - + if (fatal_signal_pending(current)) + break; sqt_spin = false; cap_entries = !list_is_singular(&sqd->ctx_list); list_for_each_entry(ctx, &sqd->ctx_list, sqd_list) { - if (current->cred != ctx->creds) { - if (old_cred) - revert_creds(old_cred); - old_cred = override_creds(ctx->creds); - } - io_sq_thread_associate_blkcg(ctx, &cur_css); -#ifdef CONFIG_AUDIT - current->loginuid = ctx->loginuid; - current->sessionid = ctx->sessionid; -#endif + const struct cred *creds = NULL; + if (ctx->sq_creds != current_cred()) + creds = override_creds(ctx->sq_creds); ret = __io_sq_thread(ctx, cap_entries); + if (creds) + revert_creds(creds); if (!sqt_spin && (ret > 0 || !list_empty(&ctx->iopoll_list))) sqt_spin = true; - - io_sq_thread_drop_mm_files(); } if (sqt_spin || !time_after(jiffies, timeout)) { io_run_task_work(); - io_sq_thread_drop_mm_files(); cond_resched(); if (sqt_spin) timeout = jiffies + sqd->sq_thread_idle; continue; } - if (kthread_should_park()) - continue; - needs_sched = true; prepare_to_wait(&sqd->wait, &wait, TASK_INTERRUPTIBLE); list_for_each_entry(ctx, &sqd->ctx_list, sqd_list) { @@ -7140,35 +6802,34 @@ static int io_sq_thread(void *data) } } - if (needs_sched) { + if (needs_sched && !test_bit(IO_SQ_THREAD_SHOULD_PARK, &sqd->state)) { list_for_each_entry(ctx, &sqd->ctx_list, sqd_list) io_ring_set_wakeup_flag(ctx); + mutex_unlock(&sqd->lock); schedule(); + try_to_freeze(); + mutex_lock(&sqd->lock); list_for_each_entry(ctx, &sqd->ctx_list, sqd_list) io_ring_clear_wakeup_flag(ctx); } finish_wait(&sqd->wait, &wait); + io_run_task_work_head(&sqd->park_task_work); timeout = jiffies + sqd->sq_thread_idle; } - io_run_task_work(); - io_sq_thread_drop_mm_files(); - - if (cur_css) - io_sq_thread_unassociate_blkcg(); - if (old_cred) - revert_creds(old_cred); - - task_lock(current); - current->files = old_files; - current->nsproxy = old_nsproxy; - task_unlock(current); - - kthread_parkme(); + list_for_each_entry(ctx, &sqd->ctx_list, sqd_list) + io_uring_cancel_sqpoll(ctx); + sqd->thread = NULL; + list_for_each_entry(ctx, &sqd->ctx_list, sqd_list) + io_ring_set_wakeup_flag(ctx); + mutex_unlock(&sqd->lock); - return 0; + io_run_task_work(); + io_run_task_work_head(&sqd->park_task_work); + complete(&sqd->exited); + do_exit(0); } struct io_wait_queue { @@ -7217,6 +6878,25 @@ static int io_run_task_work_sig(void) return -EINTR; } +/* when returns >0, the caller should retry */ +static inline int io_cqring_wait_schedule(struct io_ring_ctx *ctx, + struct io_wait_queue *iowq, + signed long *timeout) +{ + int ret; + + /* make sure we run task_work before checking for signals */ + ret = io_run_task_work_sig(); + if (ret || io_should_wake(iowq)) + return ret; + /* let the caller flush overflows, retry */ + if (test_bit(0, &ctx->cq_check_overflow)) + return 1; + + *timeout = schedule_timeout(*timeout); + return !*timeout ? -ETIME : 1; +} + /* * Wait until events become available, if we don't already have some. The * application must reap them itself, as they reside on the shared cq ring. @@ -7235,9 +6915,8 @@ static int io_cqring_wait(struct io_ring_ctx *ctx, int min_events, .to_wait = min_events, }; struct io_rings *rings = ctx->rings; - struct timespec64 ts; - signed long timeout = 0; - int ret = 0; + signed long timeout = MAX_SCHEDULE_TIMEOUT; + int ret; do { io_cqring_overflow_flush(ctx, false, NULL, NULL); @@ -7261,6 +6940,8 @@ static int io_cqring_wait(struct io_ring_ctx *ctx, int min_events, } if (uts) { + struct timespec64 ts; + if (get_timespec64(&ts, uts)) return -EFAULT; timeout = timespec64_to_jiffies(&ts); @@ -7269,34 +6950,17 @@ static int io_cqring_wait(struct io_ring_ctx *ctx, int min_events, iowq.nr_timeouts = atomic_read(&ctx->cq_timeouts); trace_io_uring_cqring_wait(ctx, min_events); do { - io_cqring_overflow_flush(ctx, false, NULL, NULL); - prepare_to_wait_exclusive(&ctx->wait, &iowq.wq, - TASK_INTERRUPTIBLE); - /* make sure we run task_work before checking for signals */ - ret = io_run_task_work_sig(); - if (ret > 0) { - finish_wait(&ctx->wait, &iowq.wq); - continue; - } - else if (ret < 0) + /* if we can't even flush overflow, don't wait for more */ + if (!io_cqring_overflow_flush(ctx, false, NULL, NULL)) { + ret = -EBUSY; break; - if (io_should_wake(&iowq)) - break; - if (test_bit(0, &ctx->cq_check_overflow)) { - finish_wait(&ctx->wait, &iowq.wq); - continue; } - if (uts) { - timeout = schedule_timeout(timeout); - if (timeout == 0) { - ret = -ETIME; - break; - } - } else { - schedule(); - } - } while (1); - finish_wait(&ctx->wait, &iowq.wq); + prepare_to_wait_exclusive(&ctx->wait, &iowq.wq, + TASK_INTERRUPTIBLE); + ret = io_cqring_wait_schedule(ctx, &iowq, &timeout); + finish_wait(&ctx->wait, &iowq.wq); + cond_resched(); + } while (ret > 0); restore_saved_sigmask_unless(ret == -EINTR); @@ -7326,90 +6990,208 @@ static void __io_sqe_files_unregister(struct io_ring_ctx *ctx) #endif } -static void io_file_ref_kill(struct percpu_ref *ref) +static void io_rsrc_data_ref_zero(struct percpu_ref *ref) { - struct fixed_file_data *data; + struct fixed_rsrc_data *data; - data = container_of(ref, struct fixed_file_data, refs); + data = container_of(ref, struct fixed_rsrc_data, refs); complete(&data->done); } -static void io_sqe_files_set_node(struct fixed_file_data *file_data, - struct fixed_file_ref_node *ref_node) +static inline void io_rsrc_ref_lock(struct io_ring_ctx *ctx) { - spin_lock_bh(&file_data->lock); - file_data->node = ref_node; - list_add_tail(&ref_node->node, &file_data->ref_list); - spin_unlock_bh(&file_data->lock); - percpu_ref_get(&file_data->refs); + spin_lock_bh(&ctx->rsrc_ref_lock); } -static int io_sqe_files_unregister(struct io_ring_ctx *ctx) +static inline void io_rsrc_ref_unlock(struct io_ring_ctx *ctx) { - struct fixed_file_data *data = ctx->file_data; - struct fixed_file_ref_node *backup_node, *ref_node = NULL; - unsigned nr_tables, i; - int ret; + spin_unlock_bh(&ctx->rsrc_ref_lock); +} - if (!data) - return -ENXIO; - backup_node = alloc_fixed_file_ref_node(ctx); - if (!backup_node) - return -ENOMEM; +static void io_sqe_rsrc_set_node(struct io_ring_ctx *ctx, + struct fixed_rsrc_data *rsrc_data, + struct fixed_rsrc_ref_node *ref_node) +{ + io_rsrc_ref_lock(ctx); + rsrc_data->node = ref_node; + list_add_tail(&ref_node->node, &ctx->rsrc_ref_list); + io_rsrc_ref_unlock(ctx); + percpu_ref_get(&rsrc_data->refs); +} - spin_lock_bh(&data->lock); +static void io_sqe_rsrc_kill_node(struct io_ring_ctx *ctx, struct fixed_rsrc_data *data) +{ + struct fixed_rsrc_ref_node *ref_node = NULL; + + io_rsrc_ref_lock(ctx); ref_node = data->node; - spin_unlock_bh(&data->lock); + data->node = NULL; + io_rsrc_ref_unlock(ctx); if (ref_node) percpu_ref_kill(&ref_node->refs); +} + +static int io_rsrc_ref_quiesce(struct fixed_rsrc_data *data, + struct io_ring_ctx *ctx, + void (*rsrc_put)(struct io_ring_ctx *ctx, + struct io_rsrc_put *prsrc)) +{ + struct fixed_rsrc_ref_node *backup_node; + int ret; - percpu_ref_kill(&data->refs); + if (data->quiesce) + return -ENXIO; - /* wait for all refs nodes to complete */ - flush_delayed_work(&ctx->file_put_work); + data->quiesce = true; do { + ret = -ENOMEM; + backup_node = alloc_fixed_rsrc_ref_node(ctx); + if (!backup_node) + break; + backup_node->rsrc_data = data; + backup_node->rsrc_put = rsrc_put; + + io_sqe_rsrc_kill_node(ctx, data); + percpu_ref_kill(&data->refs); + flush_delayed_work(&ctx->rsrc_put_work); + ret = wait_for_completion_interruptible(&data->done); if (!ret) break; + + percpu_ref_resurrect(&data->refs); + io_sqe_rsrc_set_node(ctx, data, backup_node); + backup_node = NULL; + reinit_completion(&data->done); + mutex_unlock(&ctx->uring_lock); ret = io_run_task_work_sig(); - if (ret < 0) { - percpu_ref_resurrect(&data->refs); - reinit_completion(&data->done); - io_sqe_files_set_node(data, backup_node); - return ret; - } - } while (1); + mutex_lock(&ctx->uring_lock); + } while (ret >= 0); + data->quiesce = false; + + if (backup_node) + destroy_fixed_rsrc_ref_node(backup_node); + return ret; +} + +static struct fixed_rsrc_data *alloc_fixed_rsrc_data(struct io_ring_ctx *ctx) +{ + struct fixed_rsrc_data *data; + + data = kzalloc(sizeof(*data), GFP_KERNEL); + if (!data) + return NULL; + + if (percpu_ref_init(&data->refs, io_rsrc_data_ref_zero, + PERCPU_REF_ALLOW_REINIT, GFP_KERNEL)) { + kfree(data); + return NULL; + } + data->ctx = ctx; + init_completion(&data->done); + return data; +} + +static void free_fixed_rsrc_data(struct fixed_rsrc_data *data) +{ + percpu_ref_exit(&data->refs); + kfree(data->table); + kfree(data); +} + +static int io_sqe_files_unregister(struct io_ring_ctx *ctx) +{ + struct fixed_rsrc_data *data = ctx->file_data; + unsigned nr_tables, i; + int ret; + + /* + * percpu_ref_is_dying() is to stop parallel files unregister + * Since we possibly drop uring lock later in this function to + * run task work. + */ + if (!data || percpu_ref_is_dying(&data->refs)) + return -ENXIO; + ret = io_rsrc_ref_quiesce(data, ctx, io_ring_file_put); + if (ret) + return ret; __io_sqe_files_unregister(ctx); nr_tables = DIV_ROUND_UP(ctx->nr_user_files, IORING_MAX_FILES_TABLE); for (i = 0; i < nr_tables; i++) kfree(data->table[i].files); - kfree(data->table); - percpu_ref_exit(&data->refs); - kfree(data); + free_fixed_rsrc_data(data); ctx->file_data = NULL; ctx->nr_user_files = 0; - destroy_fixed_file_ref_node(backup_node); return 0; } +static void io_sq_thread_unpark(struct io_sq_data *sqd) + __releases(&sqd->lock) +{ + WARN_ON_ONCE(sqd->thread == current); + + /* + * Do the dance but not conditional clear_bit() because it'd race with + * other threads incrementing park_pending and setting the bit. + */ + clear_bit(IO_SQ_THREAD_SHOULD_PARK, &sqd->state); + if (atomic_dec_return(&sqd->park_pending)) + set_bit(IO_SQ_THREAD_SHOULD_PARK, &sqd->state); + mutex_unlock(&sqd->lock); +} + +static void io_sq_thread_park(struct io_sq_data *sqd) + __acquires(&sqd->lock) +{ + WARN_ON_ONCE(sqd->thread == current); + + atomic_inc(&sqd->park_pending); + set_bit(IO_SQ_THREAD_SHOULD_PARK, &sqd->state); + mutex_lock(&sqd->lock); + if (sqd->thread) + wake_up_process(sqd->thread); +} + +static void io_sq_thread_stop(struct io_sq_data *sqd) +{ + WARN_ON_ONCE(sqd->thread == current); + + mutex_lock(&sqd->lock); + set_bit(IO_SQ_THREAD_SHOULD_STOP, &sqd->state); + if (sqd->thread) + wake_up_process(sqd->thread); + mutex_unlock(&sqd->lock); + wait_for_completion(&sqd->exited); +} + static void io_put_sq_data(struct io_sq_data *sqd) { if (refcount_dec_and_test(&sqd->refs)) { - /* - * The park is a bit of a work-around, without it we get - * warning spews on shutdown with SQPOLL set and affinity - * set to a single CPU. - */ - if (sqd->thread) { - kthread_park(sqd->thread); - kthread_stop(sqd->thread); - } + WARN_ON_ONCE(atomic_read(&sqd->park_pending)); + io_sq_thread_stop(sqd); kfree(sqd); } } +static void io_sq_thread_finish(struct io_ring_ctx *ctx) +{ + struct io_sq_data *sqd = ctx->sq_data; + + if (sqd) { + io_sq_thread_park(sqd); + list_del_init(&ctx->sqd_list); + io_sqd_update_thread_idle(sqd); + io_sq_thread_unpark(sqd); + + io_put_sq_data(sqd); + ctx->sq_data = NULL; + if (ctx->sq_creds) + put_cred(ctx->sq_creds); + } +} + static struct io_sq_data *io_attach_sq_data(struct io_uring_params *p) { struct io_ring_ctx *ctx_attach; @@ -7430,91 +7212,46 @@ static struct io_sq_data *io_attach_sq_data(struct io_uring_params *p) fdput(f); return ERR_PTR(-EINVAL); } + if (sqd->task_tgid != current->tgid) { + fdput(f); + return ERR_PTR(-EPERM); + } refcount_inc(&sqd->refs); fdput(f); return sqd; } -static struct io_sq_data *io_get_sq_data(struct io_uring_params *p) +static struct io_sq_data *io_get_sq_data(struct io_uring_params *p, + bool *attached) { struct io_sq_data *sqd; - if (p->flags & IORING_SETUP_ATTACH_WQ) - return io_attach_sq_data(p); + *attached = false; + if (p->flags & IORING_SETUP_ATTACH_WQ) { + sqd = io_attach_sq_data(p); + if (!IS_ERR(sqd)) { + *attached = true; + return sqd; + } + /* fall through for EPERM case, setup new sqd/task */ + if (PTR_ERR(sqd) != -EPERM) + return sqd; + } sqd = kzalloc(sizeof(*sqd), GFP_KERNEL); if (!sqd) return ERR_PTR(-ENOMEM); + atomic_set(&sqd->park_pending, 0); refcount_set(&sqd->refs, 1); INIT_LIST_HEAD(&sqd->ctx_list); - INIT_LIST_HEAD(&sqd->ctx_new_list); - mutex_init(&sqd->ctx_lock); mutex_init(&sqd->lock); init_waitqueue_head(&sqd->wait); + init_completion(&sqd->exited); return sqd; } -static void io_sq_thread_unpark(struct io_sq_data *sqd) - __releases(&sqd->lock) -{ - if (!sqd->thread) - return; - kthread_unpark(sqd->thread); - mutex_unlock(&sqd->lock); -} - -static void io_sq_thread_park(struct io_sq_data *sqd) - __acquires(&sqd->lock) -{ - if (!sqd->thread) - return; - mutex_lock(&sqd->lock); - kthread_park(sqd->thread); -} - -static void io_sq_thread_stop(struct io_ring_ctx *ctx) -{ - struct io_sq_data *sqd = ctx->sq_data; - - if (sqd) { - if (sqd->thread) { - /* - * We may arrive here from the error branch in - * io_sq_offload_create() where the kthread is created - * without being waked up, thus wake it up now to make - * sure the wait will complete. - */ - wake_up_process(sqd->thread); - wait_for_completion(&ctx->sq_thread_comp); - - io_sq_thread_park(sqd); - } - - mutex_lock(&sqd->ctx_lock); - list_del(&ctx->sqd_list); - io_sqd_update_thread_idle(sqd); - mutex_unlock(&sqd->ctx_lock); - - if (sqd->thread) - io_sq_thread_unpark(sqd); - - io_put_sq_data(sqd); - ctx->sq_data = NULL; - } -} - -static void io_finish_async(struct io_ring_ctx *ctx) -{ - io_sq_thread_stop(ctx); - - if (ctx->io_wq) { - io_wq_destroy(ctx->io_wq); - ctx->io_wq = NULL; - } -} - #if defined(CONFIG_UNIX) /* * Ensure the UNIX gc is aware of our file set, so we are certain that @@ -7541,7 +7278,7 @@ static int __io_sqe_files_scm(struct io_ring_ctx *ctx, int nr, int offset) skb->sk = sk; nr_files = 0; - fpl->user = get_uid(ctx->user); + fpl->user = get_uid(current_user()); for (i = 0; i < nr; i++) { struct file *file = io_file_from_index(ctx, i + offset); @@ -7612,13 +7349,13 @@ static int io_sqe_files_scm(struct io_ring_ctx *ctx) } #endif -static int io_sqe_alloc_file_tables(struct fixed_file_data *file_data, +static int io_sqe_alloc_file_tables(struct fixed_rsrc_data *file_data, unsigned nr_tables, unsigned nr_files) { int i; for (i = 0; i < nr_tables; i++) { - struct fixed_file_table *table = &file_data->table[i]; + struct fixed_rsrc_table *table = &file_data->table[i]; unsigned this_files; this_files = min(nr_files, IORING_MAX_FILES_TABLE); @@ -7633,14 +7370,15 @@ static int io_sqe_alloc_file_tables(struct fixed_file_data *file_data, return 0; for (i = 0; i < nr_tables; i++) { - struct fixed_file_table *table = &file_data->table[i]; + struct fixed_rsrc_table *table = &file_data->table[i]; kfree(table->files); } return 1; } -static void io_ring_file_put(struct io_ring_ctx *ctx, struct file *file) +static void io_ring_file_put(struct io_ring_ctx *ctx, struct io_rsrc_put *prsrc) { + struct file *file = prsrc->file; #if defined(CONFIG_UNIX) struct sock *sock = ctx->ring_sock->sk; struct sk_buff_head list, *head = &sock->sk_receive_queue; @@ -7701,108 +7439,119 @@ static void io_ring_file_put(struct io_ring_ctx *ctx, struct file *file) #endif } -struct io_file_put { - struct list_head list; - struct file *file; -}; - -static void __io_file_put_work(struct fixed_file_ref_node *ref_node) +static void __io_rsrc_put_work(struct fixed_rsrc_ref_node *ref_node) { - struct fixed_file_data *file_data = ref_node->file_data; - struct io_ring_ctx *ctx = file_data->ctx; - struct io_file_put *pfile, *tmp; + struct fixed_rsrc_data *rsrc_data = ref_node->rsrc_data; + struct io_ring_ctx *ctx = rsrc_data->ctx; + struct io_rsrc_put *prsrc, *tmp; - list_for_each_entry_safe(pfile, tmp, &ref_node->file_list, list) { - list_del(&pfile->list); - io_ring_file_put(ctx, pfile->file); - kfree(pfile); + list_for_each_entry_safe(prsrc, tmp, &ref_node->rsrc_list, list) { + list_del(&prsrc->list); + ref_node->rsrc_put(ctx, prsrc); + kfree(prsrc); } percpu_ref_exit(&ref_node->refs); kfree(ref_node); - percpu_ref_put(&file_data->refs); + percpu_ref_put(&rsrc_data->refs); } -static void io_file_put_work(struct work_struct *work) +static void io_rsrc_put_work(struct work_struct *work) { struct io_ring_ctx *ctx; struct llist_node *node; - ctx = container_of(work, struct io_ring_ctx, file_put_work.work); - node = llist_del_all(&ctx->file_put_llist); + ctx = container_of(work, struct io_ring_ctx, rsrc_put_work.work); + node = llist_del_all(&ctx->rsrc_put_llist); while (node) { - struct fixed_file_ref_node *ref_node; + struct fixed_rsrc_ref_node *ref_node; struct llist_node *next = node->next; - ref_node = llist_entry(node, struct fixed_file_ref_node, llist); - __io_file_put_work(ref_node); + ref_node = llist_entry(node, struct fixed_rsrc_ref_node, llist); + __io_rsrc_put_work(ref_node); node = next; } } -static void io_file_data_ref_zero(struct percpu_ref *ref) +static struct file **io_fixed_file_slot(struct fixed_rsrc_data *file_data, + unsigned i) +{ + struct fixed_rsrc_table *table; + + table = &file_data->table[i >> IORING_FILE_TABLE_SHIFT]; + return &table->files[i & IORING_FILE_TABLE_MASK]; +} + +static void io_rsrc_node_ref_zero(struct percpu_ref *ref) { - struct fixed_file_ref_node *ref_node; - struct fixed_file_data *data; + struct fixed_rsrc_ref_node *ref_node; + struct fixed_rsrc_data *data; struct io_ring_ctx *ctx; bool first_add = false; int delay = HZ; - ref_node = container_of(ref, struct fixed_file_ref_node, refs); - data = ref_node->file_data; + ref_node = container_of(ref, struct fixed_rsrc_ref_node, refs); + data = ref_node->rsrc_data; ctx = data->ctx; - spin_lock_bh(&data->lock); + io_rsrc_ref_lock(ctx); ref_node->done = true; - while (!list_empty(&data->ref_list)) { - ref_node = list_first_entry(&data->ref_list, - struct fixed_file_ref_node, node); + while (!list_empty(&ctx->rsrc_ref_list)) { + ref_node = list_first_entry(&ctx->rsrc_ref_list, + struct fixed_rsrc_ref_node, node); /* recycle ref nodes in order */ if (!ref_node->done) break; list_del(&ref_node->node); - first_add |= llist_add(&ref_node->llist, &ctx->file_put_llist); + first_add |= llist_add(&ref_node->llist, &ctx->rsrc_put_llist); } - spin_unlock_bh(&data->lock); + io_rsrc_ref_unlock(ctx); if (percpu_ref_is_dying(&data->refs)) delay = 0; if (!delay) - mod_delayed_work(system_wq, &ctx->file_put_work, 0); + mod_delayed_work(system_wq, &ctx->rsrc_put_work, 0); else if (first_add) - queue_delayed_work(system_wq, &ctx->file_put_work, delay); + queue_delayed_work(system_wq, &ctx->rsrc_put_work, delay); } -static struct fixed_file_ref_node *alloc_fixed_file_ref_node( +static struct fixed_rsrc_ref_node *alloc_fixed_rsrc_ref_node( struct io_ring_ctx *ctx) { - struct fixed_file_ref_node *ref_node; + struct fixed_rsrc_ref_node *ref_node; ref_node = kzalloc(sizeof(*ref_node), GFP_KERNEL); if (!ref_node) return NULL; - if (percpu_ref_init(&ref_node->refs, io_file_data_ref_zero, + if (percpu_ref_init(&ref_node->refs, io_rsrc_node_ref_zero, 0, GFP_KERNEL)) { kfree(ref_node); return NULL; } INIT_LIST_HEAD(&ref_node->node); - INIT_LIST_HEAD(&ref_node->file_list); - ref_node->file_data = ctx->file_data; + INIT_LIST_HEAD(&ref_node->rsrc_list); ref_node->done = false; return ref_node; } -static void destroy_fixed_file_ref_node(struct fixed_file_ref_node *ref_node) +static void init_fixed_file_ref_node(struct io_ring_ctx *ctx, + struct fixed_rsrc_ref_node *ref_node) +{ + ref_node->rsrc_data = ctx->file_data; + ref_node->rsrc_put = io_ring_file_put; +} + +static void destroy_fixed_rsrc_ref_node(struct fixed_rsrc_ref_node *ref_node) { percpu_ref_exit(&ref_node->refs); kfree(ref_node); } + static int io_sqe_files_register(struct io_ring_ctx *ctx, void __user *arg, unsigned nr_args) { @@ -7810,8 +7559,8 @@ static int io_sqe_files_register(struct io_ring_ctx *ctx, void __user *arg, unsigned nr_tables, i; struct file *file; int fd, ret = -ENOMEM; - struct fixed_file_ref_node *ref_node; - struct fixed_file_data *file_data; + struct fixed_rsrc_ref_node *ref_node; + struct fixed_rsrc_data *file_data; if (ctx->file_data) return -EBUSY; @@ -7820,13 +7569,10 @@ static int io_sqe_files_register(struct io_ring_ctx *ctx, void __user *arg, if (nr_args > IORING_MAX_FIXED_FILES) return -EMFILE; - file_data = kzalloc(sizeof(*ctx->file_data), GFP_KERNEL); + file_data = alloc_fixed_rsrc_data(ctx); if (!file_data) return -ENOMEM; - file_data->ctx = ctx; - init_completion(&file_data->done); - INIT_LIST_HEAD(&file_data->ref_list); - spin_lock_init(&file_data->lock); + ctx->file_data = file_data; nr_tables = DIV_ROUND_UP(nr_args, IORING_MAX_FILES_TABLE); file_data->table = kcalloc(nr_tables, sizeof(*file_data->table), @@ -7834,18 +7580,10 @@ static int io_sqe_files_register(struct io_ring_ctx *ctx, void __user *arg, if (!file_data->table) goto out_free; - if (percpu_ref_init(&file_data->refs, io_file_ref_kill, - PERCPU_REF_ALLOW_REINIT, GFP_KERNEL)) - goto out_free; - if (io_sqe_alloc_file_tables(file_data, nr_tables, nr_args)) - goto out_ref; - ctx->file_data = file_data; + goto out_free; for (i = 0; i < nr_args; i++, ctx->nr_user_files++) { - struct fixed_file_table *table; - unsigned index; - if (copy_from_user(&fd, &fds[i], sizeof(fd))) { ret = -EFAULT; goto out_fput; @@ -7870,9 +7608,7 @@ static int io_sqe_files_register(struct io_ring_ctx *ctx, void __user *arg, fput(file); goto out_fput; } - table = &file_data->table[i >> IORING_FILE_TABLE_SHIFT]; - index = i & IORING_FILE_TABLE_MASK; - table->files[index] = file; + *io_fixed_file_slot(file_data, i) = file; } ret = io_sqe_files_scm(ctx); @@ -7881,13 +7617,14 @@ static int io_sqe_files_register(struct io_ring_ctx *ctx, void __user *arg, return ret; } - ref_node = alloc_fixed_file_ref_node(ctx); + ref_node = alloc_fixed_rsrc_ref_node(ctx); if (!ref_node) { io_sqe_files_unregister(ctx); return -ENOMEM; } + init_fixed_file_ref_node(ctx, ref_node); - io_sqe_files_set_node(file_data, ref_node); + io_sqe_rsrc_set_node(ctx, file_data, ref_node); return ret; out_fput: for (i = 0; i < ctx->nr_user_files; i++) { @@ -7898,11 +7635,8 @@ out_fput: for (i = 0; i < nr_tables; i++) kfree(file_data->table[i].files); ctx->nr_user_files = 0; -out_ref: - percpu_ref_exit(&file_data->refs); out_free: - kfree(file_data->table); - kfree(file_data); + free_fixed_rsrc_data(ctx->file_data); ctx->file_data = NULL; return ret; } @@ -7950,29 +7684,34 @@ static int io_sqe_file_register(struct io_ring_ctx *ctx, struct file *file, #endif } -static int io_queue_file_removal(struct fixed_file_data *data, - struct file *file) +static int io_queue_rsrc_removal(struct fixed_rsrc_data *data, void *rsrc) { - struct io_file_put *pfile; - struct fixed_file_ref_node *ref_node = data->node; + struct io_rsrc_put *prsrc; + struct fixed_rsrc_ref_node *ref_node = data->node; - pfile = kzalloc(sizeof(*pfile), GFP_KERNEL); - if (!pfile) + prsrc = kzalloc(sizeof(*prsrc), GFP_KERNEL); + if (!prsrc) return -ENOMEM; - pfile->file = file; - list_add(&pfile->list, &ref_node->file_list); + prsrc->rsrc = rsrc; + list_add(&prsrc->list, &ref_node->rsrc_list); return 0; } +static inline int io_queue_file_removal(struct fixed_rsrc_data *data, + struct file *file) +{ + return io_queue_rsrc_removal(data, (void *)file); +} + static int __io_sqe_files_update(struct io_ring_ctx *ctx, - struct io_uring_files_update *up, + struct io_uring_rsrc_update *up, unsigned nr_args) { - struct fixed_file_data *data = ctx->file_data; - struct fixed_file_ref_node *ref_node; - struct file *file; + struct fixed_rsrc_data *data = ctx->file_data; + struct fixed_rsrc_ref_node *ref_node; + struct file *file, **file_slot; __s32 __user *fds; int fd, i, err; __u32 done; @@ -7983,30 +7722,29 @@ static int __io_sqe_files_update(struct io_ring_ctx *ctx, if (done > ctx->nr_user_files) return -EINVAL; - ref_node = alloc_fixed_file_ref_node(ctx); + ref_node = alloc_fixed_rsrc_ref_node(ctx); if (!ref_node) return -ENOMEM; + init_fixed_file_ref_node(ctx, ref_node); - done = 0; - fds = u64_to_user_ptr(up->fds); - while (nr_args) { - struct fixed_file_table *table; - unsigned index; - + fds = u64_to_user_ptr(up->data); + for (done = 0; done < nr_args; done++) { err = 0; if (copy_from_user(&fd, &fds[done], sizeof(fd))) { err = -EFAULT; break; } - i = array_index_nospec(up->offset, ctx->nr_user_files); - table = &ctx->file_data->table[i >> IORING_FILE_TABLE_SHIFT]; - index = i & IORING_FILE_TABLE_MASK; - if (table->files[index]) { - file = table->files[index]; - err = io_queue_file_removal(data, file); + if (fd == IORING_REGISTER_FILES_SKIP) + continue; + + i = array_index_nospec(up->offset + done, ctx->nr_user_files); + file_slot = io_fixed_file_slot(ctx->file_data, i); + + if (*file_slot) { + err = io_queue_file_removal(data, *file_slot); if (err) break; - table->files[index] = NULL; + *file_slot = NULL; needs_switch = true; } if (fd != -1) { @@ -8028,24 +7766,21 @@ static int __io_sqe_files_update(struct io_ring_ctx *ctx, err = -EBADF; break; } - table->files[index] = file; + *file_slot = file; err = io_sqe_file_register(ctx, file, i); if (err) { - table->files[index] = NULL; + *file_slot = NULL; fput(file); break; } } - nr_args--; - done++; - up->offset++; } if (needs_switch) { percpu_ref_kill(&data->node->refs); - io_sqe_files_set_node(data, ref_node); + io_sqe_rsrc_set_node(ctx, data, ref_node); } else - destroy_fixed_file_ref_node(ref_node); + destroy_fixed_rsrc_ref_node(ref_node); return done ? done : err; } @@ -8053,7 +7788,7 @@ static int __io_sqe_files_update(struct io_ring_ctx *ctx, static int io_sqe_files_update(struct io_ring_ctx *ctx, void __user *arg, unsigned nr_args) { - struct io_uring_files_update up; + struct io_uring_rsrc_update up; if (!ctx->file_data) return -ENXIO; @@ -8067,62 +7802,42 @@ static int io_sqe_files_update(struct io_ring_ctx *ctx, void __user *arg, return __io_sqe_files_update(ctx, &up, nr_args); } -static void io_free_work(struct io_wq_work *work) +static struct io_wq_work *io_free_work(struct io_wq_work *work) { struct io_kiocb *req = container_of(work, struct io_kiocb, work); - /* Consider that io_steal_work() relies on this ref */ - io_put_req(req); + req = io_put_req_find_next(req); + return req ? &req->work : NULL; } -static int io_init_wq_offload(struct io_ring_ctx *ctx, - struct io_uring_params *p) +static struct io_wq *io_init_wq_offload(struct io_ring_ctx *ctx) { + struct io_wq_hash *hash; struct io_wq_data data; - struct fd f; - struct io_ring_ctx *ctx_attach; unsigned int concurrency; - int ret = 0; - data.user = ctx->user; - data.free_work = io_free_work; - data.do_work = io_wq_submit_work; - - if (!(p->flags & IORING_SETUP_ATTACH_WQ)) { - /* Do QD, or 4 * CPUS, whatever is smallest */ - concurrency = min(ctx->sq_entries, 4 * num_online_cpus()); - - ctx->io_wq = io_wq_create(concurrency, &data); - if (IS_ERR(ctx->io_wq)) { - ret = PTR_ERR(ctx->io_wq); - ctx->io_wq = NULL; - } - return ret; + hash = ctx->hash_map; + if (!hash) { + hash = kzalloc(sizeof(*hash), GFP_KERNEL); + if (!hash) + return ERR_PTR(-ENOMEM); + refcount_set(&hash->refs, 1); + init_waitqueue_head(&hash->wait); + ctx->hash_map = hash; } - f = fdget(p->wq_fd); - if (!f.file) - return -EBADF; - - if (f.file->f_op != &io_uring_fops) { - ret = -EINVAL; - goto out_fput; - } + data.hash = hash; + data.free_work = io_free_work; + data.do_work = io_wq_submit_work; - ctx_attach = f.file->private_data; - /* @io_wq is protected by holding the fd */ - if (!io_wq_get(ctx_attach->io_wq, &data)) { - ret = -EINVAL; - goto out_fput; - } + /* Do QD, or 4 * CPUS, whatever is smallest */ + concurrency = min(ctx->sq_entries, 4 * num_online_cpus()); - ctx->io_wq = ctx_attach->io_wq; -out_fput: - fdput(f); - return ret; + return io_wq_create(concurrency, &data); } -static int io_uring_alloc_task_context(struct task_struct *task) +static int io_uring_alloc_task_context(struct task_struct *task, + struct io_ring_ctx *ctx) { struct io_uring_task *tctx; int ret; @@ -8137,14 +7852,23 @@ static int io_uring_alloc_task_context(struct task_struct *task) return ret; } + tctx->io_wq = io_init_wq_offload(ctx); + if (IS_ERR(tctx->io_wq)) { + ret = PTR_ERR(tctx->io_wq); + percpu_counter_destroy(&tctx->inflight); + kfree(tctx); + return ret; + } + xa_init(&tctx->xa); init_waitqueue_head(&tctx->wait); tctx->last = NULL; atomic_set(&tctx->in_idle, 0); - tctx->sqpoll = false; - io_init_identity(&tctx->__identity); - tctx->identity = &tctx->__identity; task->io_uring = tctx; + spin_lock_init(&tctx->task_lock); + INIT_WQ_LIST(&tctx->task_list); + tctx->task_state = 0; + init_task_work(&tctx->task_work, tctx_task_work); return 0; } @@ -8153,9 +7877,8 @@ void __io_uring_free(struct task_struct *tsk) struct io_uring_task *tctx = tsk->io_uring; WARN_ON_ONCE(!xa_empty(&tctx->xa)); - WARN_ON_ONCE(refcount_read(&tctx->identity->count) != 1); - if (tctx->identity != &tctx->__identity) - kfree(tctx->identity); + WARN_ON_ONCE(tctx->io_wq); + percpu_counter_destroy(&tctx->inflight); kfree(tctx); tsk->io_uring = NULL; @@ -8166,54 +7889,80 @@ static int io_sq_offload_create(struct io_ring_ctx *ctx, { int ret; + /* Retain compatibility with failing for an invalid attach attempt */ + if ((ctx->flags & (IORING_SETUP_ATTACH_WQ | IORING_SETUP_SQPOLL)) == + IORING_SETUP_ATTACH_WQ) { + struct fd f; + + f = fdget(p->wq_fd); + if (!f.file) + return -ENXIO; + if (f.file->f_op != &io_uring_fops) { + fdput(f); + return -EINVAL; + } + fdput(f); + } if (ctx->flags & IORING_SETUP_SQPOLL) { + struct task_struct *tsk; struct io_sq_data *sqd; + bool attached; ret = -EPERM; if (!capable(CAP_SYS_ADMIN) && !capable(CAP_SYS_NICE)) goto err; - sqd = io_get_sq_data(p); + sqd = io_get_sq_data(p, &attached); if (IS_ERR(sqd)) { ret = PTR_ERR(sqd); goto err; } + ctx->sq_creds = get_current_cred(); ctx->sq_data = sqd; - io_sq_thread_park(sqd); - mutex_lock(&sqd->ctx_lock); - list_add(&ctx->sqd_list, &sqd->ctx_new_list); - mutex_unlock(&sqd->ctx_lock); - io_sq_thread_unpark(sqd); - ctx->sq_thread_idle = msecs_to_jiffies(p->sq_thread_idle); if (!ctx->sq_thread_idle) ctx->sq_thread_idle = HZ; - if (sqd->thread) - goto done; + ret = 0; + io_sq_thread_park(sqd); + list_add(&ctx->sqd_list, &sqd->ctx_list); + io_sqd_update_thread_idle(sqd); + /* don't attach to a dying SQPOLL thread, would be racy */ + if (attached && !sqd->thread) + ret = -ENXIO; + io_sq_thread_unpark(sqd); + + if (ret < 0) + goto err; + if (attached) + return 0; if (p->flags & IORING_SETUP_SQ_AFF) { int cpu = p->sq_thread_cpu; ret = -EINVAL; if (cpu >= nr_cpu_ids) - goto err; + goto err_sqpoll; if (!cpu_online(cpu)) - goto err; + goto err_sqpoll; - sqd->thread = kthread_create_on_cpu(io_sq_thread, sqd, - cpu, "io_uring-sq"); + sqd->sq_cpu = cpu; } else { - sqd->thread = kthread_create(io_sq_thread, sqd, - "io_uring-sq"); + sqd->sq_cpu = -1; } - if (IS_ERR(sqd->thread)) { - ret = PTR_ERR(sqd->thread); - sqd->thread = NULL; - goto err; + + sqd->task_pid = current->pid; + sqd->task_tgid = current->tgid; + tsk = create_io_thread(io_sq_thread, sqd, NUMA_NO_NODE); + if (IS_ERR(tsk)) { + ret = PTR_ERR(tsk); + goto err_sqpoll; } - ret = io_uring_alloc_task_context(sqd->thread); + + sqd->thread = tsk; + ret = io_uring_alloc_task_context(tsk, ctx); + wake_up_new_task(tsk); if (ret) goto err; } else if (p->flags & IORING_SETUP_SQ_AFF) { @@ -8222,23 +7971,13 @@ static int io_sq_offload_create(struct io_ring_ctx *ctx, goto err; } -done: - ret = io_init_wq_offload(ctx, p); - if (ret) - goto err; - return 0; err: - io_finish_async(ctx); + io_sq_thread_finish(ctx); return ret; -} - -static void io_sq_offload_start(struct io_ring_ctx *ctx) -{ - struct io_sq_data *sqd = ctx->sq_data; - - if ((ctx->flags & IORING_SETUP_SQPOLL) && sqd->thread) - wake_up_process(sqd->thread); +err_sqpoll: + complete(&ctx->sq_data->exited); + goto err; } static inline void __io_unaccount_mem(struct user_struct *user, @@ -8266,43 +8005,27 @@ static inline int __io_account_mem(struct user_struct *user, return 0; } -static void io_unaccount_mem(struct io_ring_ctx *ctx, unsigned long nr_pages, - enum io_mem_account acct) +static void io_unaccount_mem(struct io_ring_ctx *ctx, unsigned long nr_pages) { - if (ctx->limit_mem) + if (ctx->user) __io_unaccount_mem(ctx->user, nr_pages); - if (ctx->mm_account) { - if (acct == ACCT_LOCKED) { - mmap_write_lock(ctx->mm_account); - ctx->mm_account->locked_vm -= nr_pages; - mmap_write_unlock(ctx->mm_account); - }else if (acct == ACCT_PINNED) { - atomic64_sub(nr_pages, &ctx->mm_account->pinned_vm); - } - } + if (ctx->mm_account) + atomic64_sub(nr_pages, &ctx->mm_account->pinned_vm); } -static int io_account_mem(struct io_ring_ctx *ctx, unsigned long nr_pages, - enum io_mem_account acct) +static int io_account_mem(struct io_ring_ctx *ctx, unsigned long nr_pages) { int ret; - if (ctx->limit_mem) { + if (ctx->user) { ret = __io_account_mem(ctx->user, nr_pages); if (ret) return ret; } - if (ctx->mm_account) { - if (acct == ACCT_LOCKED) { - mmap_write_lock(ctx->mm_account); - ctx->mm_account->locked_vm += nr_pages; - mmap_write_unlock(ctx->mm_account); - } else if (acct == ACCT_PINNED) { - atomic64_add(nr_pages, &ctx->mm_account->pinned_vm); - } - } + if (ctx->mm_account) + atomic64_add(nr_pages, &ctx->mm_account->pinned_vm); return 0; } @@ -8322,7 +8045,7 @@ static void io_mem_free(void *ptr) static void *io_mem_alloc(size_t size) { gfp_t gfp_flags = GFP_KERNEL | __GFP_ZERO | __GFP_NOWARN | __GFP_COMP | - __GFP_NORETRY; + __GFP_NORETRY | __GFP_ACCOUNT; return (void *) __get_free_pages(gfp_flags, get_order(size)); } @@ -8356,19 +8079,7 @@ static unsigned long rings_size(unsigned sq_entries, unsigned cq_entries, return off; } -static unsigned long ring_pages(unsigned sq_entries, unsigned cq_entries) -{ - size_t pages; - - pages = (size_t)1 << get_order( - rings_size(sq_entries, cq_entries, NULL)); - pages += (size_t)1 << get_order( - array_size(sizeof(struct io_uring_sqe), sq_entries)); - - return pages; -} - -static int io_sqe_buffer_unregister(struct io_ring_ctx *ctx) +static int io_sqe_buffers_unregister(struct io_ring_ctx *ctx) { int i, j; @@ -8382,7 +8093,7 @@ static int io_sqe_buffer_unregister(struct io_ring_ctx *ctx) unpin_user_page(imu->bvec[j].bv_page); if (imu->acct_pages) - io_unaccount_mem(ctx, imu->acct_pages, ACCT_PINNED); + io_unaccount_mem(ctx, imu->acct_pages); kvfree(imu->bvec); imu->nr_bvecs = 0; } @@ -8480,21 +8191,105 @@ static int io_buffer_account_pin(struct io_ring_ctx *ctx, struct page **pages, if (!imu->acct_pages) return 0; - ret = io_account_mem(ctx, imu->acct_pages, ACCT_PINNED); + ret = io_account_mem(ctx, imu->acct_pages); if (ret) imu->acct_pages = 0; return ret; } -static int io_sqe_buffer_register(struct io_ring_ctx *ctx, void __user *arg, - unsigned nr_args) +static int io_sqe_buffer_register(struct io_ring_ctx *ctx, struct iovec *iov, + struct io_mapped_ubuf *imu, + struct page **last_hpage) { struct vm_area_struct **vmas = NULL; struct page **pages = NULL; - struct page *last_hpage = NULL; - int i, j, got_pages = 0; - int ret = -EINVAL; + unsigned long off, start, end, ubuf; + size_t size; + int ret, pret, nr_pages, i; + + ubuf = (unsigned long) iov->iov_base; + end = (ubuf + iov->iov_len + PAGE_SIZE - 1) >> PAGE_SHIFT; + start = ubuf >> PAGE_SHIFT; + nr_pages = end - start; + + ret = -ENOMEM; + + pages = kvmalloc_array(nr_pages, sizeof(struct page *), GFP_KERNEL); + if (!pages) + goto done; + + vmas = kvmalloc_array(nr_pages, sizeof(struct vm_area_struct *), + GFP_KERNEL); + if (!vmas) + goto done; + + imu->bvec = kvmalloc_array(nr_pages, sizeof(struct bio_vec), + GFP_KERNEL); + if (!imu->bvec) + goto done; + + ret = 0; + mmap_read_lock(current->mm); + pret = pin_user_pages(ubuf, nr_pages, FOLL_WRITE | FOLL_LONGTERM, + pages, vmas); + if (pret == nr_pages) { + /* don't support file backed memory */ + for (i = 0; i < nr_pages; i++) { + struct vm_area_struct *vma = vmas[i]; + + if (vma->vm_file && + !is_file_hugepages(vma->vm_file)) { + ret = -EOPNOTSUPP; + break; + } + } + } else { + ret = pret < 0 ? pret : -EFAULT; + } + mmap_read_unlock(current->mm); + if (ret) { + /* + * if we did partial map, or found file backed vmas, + * release any pages we did get + */ + if (pret > 0) + unpin_user_pages(pages, pret); + kvfree(imu->bvec); + goto done; + } + ret = io_buffer_account_pin(ctx, pages, pret, imu, last_hpage); + if (ret) { + unpin_user_pages(pages, pret); + kvfree(imu->bvec); + goto done; + } + + off = ubuf & ~PAGE_MASK; + size = iov->iov_len; + for (i = 0; i < nr_pages; i++) { + size_t vec_len; + + vec_len = min_t(size_t, size, PAGE_SIZE - off); + imu->bvec[i].bv_page = pages[i]; + imu->bvec[i].bv_len = vec_len; + imu->bvec[i].bv_offset = off; + off = 0; + size -= vec_len; + } + /* store original address for later verification */ + imu->ubuf = ubuf; + imu->len = iov->iov_len; + imu->nr_bvecs = nr_pages; + ret = 0; +done: + kvfree(pages); + kvfree(vmas); + return ret; +} + +static int io_buffers_map_alloc(struct io_ring_ctx *ctx, unsigned int nr_args) +{ if (ctx->user_bufs) return -EBUSY; if (!nr_args || nr_args > UIO_MAXIOV) @@ -8505,121 +8300,58 @@ static int io_sqe_buffer_register(struct io_ring_ctx *ctx, void __user *arg, if (!ctx->user_bufs) return -ENOMEM; - for (i = 0; i < nr_args; i++) { - struct io_mapped_ubuf *imu = &ctx->user_bufs[i]; - unsigned long off, start, end, ubuf; - int pret, nr_pages; - struct iovec iov; - size_t size; + return 0; +} - ret = io_copy_iov(ctx, &iov, arg, i); - if (ret) - goto err; +static int io_buffer_validate(struct iovec *iov) +{ + /* + * Don't impose further limits on the size and buffer + * constraints here, we'll -EINVAL later when IO is + * submitted if they are wrong. + */ + if (!iov->iov_base || !iov->iov_len) + return -EFAULT; - /* - * Don't impose further limits on the size and buffer - * constraints here, we'll -EINVAL later when IO is - * submitted if they are wrong. - */ - ret = -EFAULT; - if (!iov.iov_base || !iov.iov_len) - goto err; + /* arbitrary limit, but we need something */ + if (iov->iov_len > SZ_1G) + return -EFAULT; - /* arbitrary limit, but we need something */ - if (iov.iov_len > SZ_1G) - goto err; + return 0; +} - ubuf = (unsigned long) iov.iov_base; - end = (ubuf + iov.iov_len + PAGE_SIZE - 1) >> PAGE_SHIFT; - start = ubuf >> PAGE_SHIFT; - nr_pages = end - start; +static int io_sqe_buffers_register(struct io_ring_ctx *ctx, void __user *arg, + unsigned int nr_args) +{ + int i, ret; + struct iovec iov; + struct page *last_hpage = NULL; - ret = 0; - if (!pages || nr_pages > got_pages) { - kvfree(vmas); - kvfree(pages); - pages = kvmalloc_array(nr_pages, sizeof(struct page *), - GFP_KERNEL); - vmas = kvmalloc_array(nr_pages, - sizeof(struct vm_area_struct *), - GFP_KERNEL); - if (!pages || !vmas) { - ret = -ENOMEM; - goto err; - } - got_pages = nr_pages; - } + ret = io_buffers_map_alloc(ctx, nr_args); + if (ret) + return ret; - imu->bvec = kvmalloc_array(nr_pages, sizeof(struct bio_vec), - GFP_KERNEL); - ret = -ENOMEM; - if (!imu->bvec) - goto err; + for (i = 0; i < nr_args; i++) { + struct io_mapped_ubuf *imu = &ctx->user_bufs[i]; - ret = 0; - mmap_read_lock(current->mm); - pret = pin_user_pages(ubuf, nr_pages, - FOLL_WRITE | FOLL_LONGTERM, - pages, vmas); - if (pret == nr_pages) { - /* don't support file backed memory */ - for (j = 0; j < nr_pages; j++) { - struct vm_area_struct *vma = vmas[j]; - - if (vma->vm_file && - !is_file_hugepages(vma->vm_file)) { - ret = -EOPNOTSUPP; - break; - } - } - } else { - ret = pret < 0 ? pret : -EFAULT; - } - mmap_read_unlock(current->mm); - if (ret) { - /* - * if we did partial map, or found file backed vmas, - * release any pages we did get - */ - if (pret > 0) - unpin_user_pages(pages, pret); - kvfree(imu->bvec); - goto err; - } + ret = io_copy_iov(ctx, &iov, arg, i); + if (ret) + break; - ret = io_buffer_account_pin(ctx, pages, pret, imu, &last_hpage); - if (ret) { - unpin_user_pages(pages, pret); - kvfree(imu->bvec); - goto err; - } + ret = io_buffer_validate(&iov); + if (ret) + break; - off = ubuf & ~PAGE_MASK; - size = iov.iov_len; - for (j = 0; j < nr_pages; j++) { - size_t vec_len; - - vec_len = min_t(size_t, size, PAGE_SIZE - off); - imu->bvec[j].bv_page = pages[j]; - imu->bvec[j].bv_len = vec_len; - imu->bvec[j].bv_offset = off; - off = 0; - size -= vec_len; - } - /* store original address for later verification */ - imu->ubuf = ubuf; - imu->len = iov.iov_len; - imu->nr_bvecs = nr_pages; + ret = io_sqe_buffer_register(ctx, &iov, imu, &last_hpage); + if (ret) + break; ctx->nr_user_bufs++; } - kvfree(pages); - kvfree(vmas); - return 0; -err: - kvfree(pages); - kvfree(vmas); - io_sqe_buffer_unregister(ctx); + + if (ret) + io_sqe_buffers_unregister(ctx); + return ret; } @@ -8655,42 +8387,75 @@ static int io_eventfd_unregister(struct io_ring_ctx *ctx) return -ENXIO; } -static int __io_destroy_buffers(int id, void *p, void *data) +static void io_destroy_buffers(struct io_ring_ctx *ctx) { - struct io_ring_ctx *ctx = data; - struct io_buffer *buf = p; + struct io_buffer *buf; + unsigned long index; - __io_remove_buffers(ctx, buf, id, -1U); - return 0; + xa_for_each(&ctx->io_buffers, index, buf) + __io_remove_buffers(ctx, buf, index, -1U); } -static void io_destroy_buffers(struct io_ring_ctx *ctx) +static void io_req_cache_free(struct list_head *list, struct task_struct *tsk) { - idr_for_each(&ctx->io_buffer_idr, __io_destroy_buffers, ctx); - idr_destroy(&ctx->io_buffer_idr); + struct io_kiocb *req, *nxt; + + list_for_each_entry_safe(req, nxt, list, compl.list) { + if (tsk && req->task != tsk) + continue; + list_del(&req->compl.list); + kmem_cache_free(req_cachep, req); + } +} + +static void io_req_caches_free(struct io_ring_ctx *ctx) +{ + struct io_submit_state *submit_state = &ctx->submit_state; + struct io_comp_state *cs = &ctx->submit_state.comp; + + mutex_lock(&ctx->uring_lock); + + if (submit_state->free_reqs) { + kmem_cache_free_bulk(req_cachep, submit_state->free_reqs, + submit_state->reqs); + submit_state->free_reqs = 0; + } + + spin_lock_irq(&ctx->completion_lock); + list_splice_init(&cs->locked_free_list, &cs->free_list); + cs->locked_free_nr = 0; + spin_unlock_irq(&ctx->completion_lock); + + io_req_cache_free(&cs->free_list, NULL); + + mutex_unlock(&ctx->uring_lock); } static void io_ring_ctx_free(struct io_ring_ctx *ctx) { - io_finish_async(ctx); - io_sqe_buffer_unregister(ctx); + /* + * Some may use context even when all refs and requests have been put, + * and they are free to do so while still holding uring_lock or + * completion_lock, see __io_req_task_submit(). Wait for them to finish. + */ + mutex_lock(&ctx->uring_lock); + mutex_unlock(&ctx->uring_lock); + spin_lock_irq(&ctx->completion_lock); + spin_unlock_irq(&ctx->completion_lock); - if (ctx->sqo_task) { - put_task_struct(ctx->sqo_task); - ctx->sqo_task = NULL; + io_sq_thread_finish(ctx); + io_sqe_buffers_unregister(ctx); + + if (ctx->mm_account) { mmdrop(ctx->mm_account); ctx->mm_account = NULL; } -#ifdef CONFIG_BLK_CGROUP - if (ctx->sqo_blkcg_css) - css_put(ctx->sqo_blkcg_css); -#endif - + mutex_lock(&ctx->uring_lock); io_sqe_files_unregister(ctx); + mutex_unlock(&ctx->uring_lock); io_eventfd_unregister(ctx); io_destroy_buffers(ctx); - idr_destroy(&ctx->personality_idr); #if defined(CONFIG_UNIX) if (ctx->ring_sock) { @@ -8704,9 +8469,10 @@ static void io_ring_ctx_free(struct io_ring_ctx *ctx) percpu_ref_exit(&ctx->refs); free_uid(ctx->user); - put_cred(ctx->creds); + io_req_caches_free(ctx); + if (ctx->hash_map) + io_wq_put_hash(ctx->hash_map); kfree(ctx->cancel_hash); - kmem_cache_free(req_cachep, ctx->fallback_req); kfree(ctx); } @@ -8723,8 +8489,21 @@ static __poll_t io_uring_poll(struct file *file, poll_table *wait) smp_rmb(); if (!io_sqring_full(ctx)) mask |= EPOLLOUT | EPOLLWRNORM; - io_cqring_overflow_flush(ctx, false, NULL, NULL); - if (io_cqring_events(ctx)) + + /* + * Don't flush cqring overflow list here, just do a simple check. + * Otherwise there could possible be ABBA deadlock: + * CPU0 CPU1 + * ---- ---- + * lock(&ctx->uring_lock); + * lock(&ep->mtx); + * lock(&ctx->uring_lock); + * lock(&ep->mtx); + * + * Users may get EPOLLIN meanwhile seeing nothing in cqring, this + * pushs them to do the flush. + */ + if (io_cqring_events(ctx) || test_bit(0, &ctx->cq_check_overflow)) mask |= EPOLLIN | EPOLLRDNORM; return mask; @@ -8737,24 +8516,52 @@ static int io_uring_fasync(int fd, struct file *file, int on) return fasync_helper(fd, file, on, &ctx->cq_fasync); } -static int io_remove_personalities(int id, void *p, void *data) +static int io_unregister_personality(struct io_ring_ctx *ctx, unsigned id) { - struct io_ring_ctx *ctx = data; - struct io_identity *iod; + const struct cred *creds; - iod = idr_remove(&ctx->personality_idr, id); - if (iod) { - put_cred(iod->creds); - if (refcount_dec_and_test(&iod->count)) - kfree(iod); + creds = xa_erase(&ctx->personalities, id); + if (creds) { + put_cred(creds); + return 0; } - return 0; + + return -EINVAL; +} + +static inline bool io_run_ctx_fallback(struct io_ring_ctx *ctx) +{ + return io_run_task_work_head(&ctx->exit_task_work); +} + +struct io_tctx_exit { + struct callback_head task_work; + struct completion completion; + struct io_ring_ctx *ctx; +}; + +static void io_tctx_exit_cb(struct callback_head *cb) +{ + struct io_uring_task *tctx = current->io_uring; + struct io_tctx_exit *work; + + work = container_of(cb, struct io_tctx_exit, task_work); + /* + * When @in_idle, we're in cancellation and it's racy to remove the + * node. It'll be removed by the end of cancellation, just ignore it. + */ + if (!atomic_read(&tctx->in_idle)) + io_uring_del_task_file((unsigned long)work->ctx); + complete(&work->completion); } static void io_ring_exit_work(struct work_struct *work) { - struct io_ring_ctx *ctx = container_of(work, struct io_ring_ctx, - exit_work); + struct io_ring_ctx *ctx = container_of(work, struct io_ring_ctx, exit_work); + unsigned long timeout = jiffies + HZ * 60 * 5; + struct io_tctx_exit exit; + struct io_tctx_node *node; + int ret; /* * If we're doing polled IO and end up having requests being @@ -8763,49 +8570,63 @@ static void io_ring_exit_work(struct work_struct *work) * as nobody else will be looking for them. */ do { - __io_uring_cancel_task_requests(ctx, NULL); + io_uring_try_cancel_requests(ctx, NULL, NULL); + + WARN_ON_ONCE(time_after(jiffies, timeout)); } while (!wait_for_completion_timeout(&ctx->ref_comp, HZ/20)); - io_ring_ctx_free(ctx); -} -static bool io_cancel_ctx_cb(struct io_wq_work *work, void *data) -{ - struct io_kiocb *req = container_of(work, struct io_kiocb, work); + mutex_lock(&ctx->uring_lock); + while (!list_empty(&ctx->tctx_list)) { + WARN_ON_ONCE(time_after(jiffies, timeout)); + + node = list_first_entry(&ctx->tctx_list, struct io_tctx_node, + ctx_node); + exit.ctx = ctx; + init_completion(&exit.completion); + init_task_work(&exit.task_work, io_tctx_exit_cb); + ret = task_work_add(node->task, &exit.task_work, TWA_SIGNAL); + if (WARN_ON_ONCE(ret)) + continue; + wake_up_process(node->task); - return req->ctx == data; + mutex_unlock(&ctx->uring_lock); + wait_for_completion(&exit.completion); + cond_resched(); + mutex_lock(&ctx->uring_lock); + } + mutex_unlock(&ctx->uring_lock); + + io_ring_ctx_free(ctx); } static void io_ring_ctx_wait_and_kill(struct io_ring_ctx *ctx) { + unsigned long index; + struct creds *creds; + mutex_lock(&ctx->uring_lock); percpu_ref_kill(&ctx->refs); - - if (WARN_ON_ONCE((ctx->flags & IORING_SETUP_SQPOLL) && !ctx->sqo_dead)) - ctx->sqo_dead = 1; - /* if force is set, the ring is going away. always drop after that */ ctx->cq_overflow_flushed = 1; if (ctx->rings) __io_cqring_overflow_flush(ctx, true, NULL, NULL); + xa_for_each(&ctx->personalities, index, creds) + io_unregister_personality(ctx, index); mutex_unlock(&ctx->uring_lock); + /* prevent SQPOLL from submitting new requests */ + if (ctx->sq_data) { + io_sq_thread_park(ctx->sq_data); + list_del_init(&ctx->sqd_list); + io_sqd_update_thread_idle(ctx->sq_data); + io_sq_thread_unpark(ctx->sq_data); + } + io_kill_timeouts(ctx, NULL, NULL); io_poll_remove_all(ctx, NULL, NULL); - if (ctx->io_wq) - io_wq_cancel_cb(ctx->io_wq, io_cancel_ctx_cb, ctx, true); - /* if we failed setting up the ctx, we might not have any rings */ io_iopoll_try_reap_events(ctx); - idr_for_each(&ctx->personality_idr, io_remove_personalities, ctx); - - /* - * Do this upfront, so we won't have a grace period where the ring - * is closed but resources aren't reaped yet. This can cause - * spurious failure in setting up a new ring. - */ - io_unaccount_mem(ctx, ring_pages(ctx->sq_entries, ctx->cq_entries), - ACCT_LOCKED); INIT_WORK(&ctx->exit_work, io_ring_exit_work); /* @@ -8851,11 +8672,11 @@ static bool io_cancel_task_cb(struct io_wq_work *work, void *data) return ret; } -static void io_cancel_defer_files(struct io_ring_ctx *ctx, +static bool io_cancel_defer_files(struct io_ring_ctx *ctx, struct task_struct *task, struct files_struct *files) { - struct io_defer_entry *de = NULL; + struct io_defer_entry *de; LIST_HEAD(list); spin_lock_irq(&ctx->completion_lock); @@ -8866,6 +8687,8 @@ static void io_cancel_defer_files(struct io_ring_ctx *ctx, } } spin_unlock_irq(&ctx->completion_lock); + if (list_empty(&list)) + return false; while (!list_empty(&list)) { de = list_first_entry(&list, struct io_defer_entry, list); @@ -8875,226 +8698,292 @@ static void io_cancel_defer_files(struct io_ring_ctx *ctx, io_req_complete(de->req, -ECANCELED); kfree(de); } + return true; } -static int io_uring_count_inflight(struct io_ring_ctx *ctx, - struct task_struct *task, - struct files_struct *files) +static bool io_cancel_ctx_cb(struct io_wq_work *work, void *data) { - struct io_kiocb *req; - int cnt = 0; + struct io_kiocb *req = container_of(work, struct io_kiocb, work); - spin_lock_irq(&ctx->inflight_lock); - list_for_each_entry(req, &ctx->inflight_list, inflight_entry) - cnt += io_match_task(req, task, files); - spin_unlock_irq(&ctx->inflight_lock); - return cnt; + return req->ctx == data; } -static void io_uring_cancel_files(struct io_ring_ctx *ctx, - struct task_struct *task, - struct files_struct *files) +static bool io_uring_try_cancel_iowq(struct io_ring_ctx *ctx) { - while (!list_empty_careful(&ctx->inflight_list)) { - struct io_task_cancel cancel = { .task = task, .files = files }; - DEFINE_WAIT(wait); - int inflight; + struct io_tctx_node *node; + enum io_wq_cancel cret; + bool ret = false; - inflight = io_uring_count_inflight(ctx, task, files); - if (!inflight) - break; - - io_wq_cancel_cb(ctx->io_wq, io_cancel_task_cb, &cancel, true); - io_poll_remove_all(ctx, task, files); - io_kill_timeouts(ctx, task, files); - io_cqring_overflow_flush(ctx, true, task, files); - /* cancellations _may_ trigger task work */ - io_run_task_work(); + mutex_lock(&ctx->uring_lock); + list_for_each_entry(node, &ctx->tctx_list, ctx_node) { + struct io_uring_task *tctx = node->task->io_uring; - prepare_to_wait(&task->io_uring->wait, &wait, - TASK_UNINTERRUPTIBLE); - if (inflight == io_uring_count_inflight(ctx, task, files)) - schedule(); - finish_wait(&task->io_uring->wait, &wait); + /* + * io_wq will stay alive while we hold uring_lock, because it's + * killed after ctx nodes, which requires to take the lock. + */ + if (!tctx || !tctx->io_wq) + continue; + cret = io_wq_cancel_cb(tctx->io_wq, io_cancel_ctx_cb, ctx, true); + ret |= (cret != IO_WQ_CANCEL_NOTFOUND); } + mutex_unlock(&ctx->uring_lock); + + return ret; } -static void __io_uring_cancel_task_requests(struct io_ring_ctx *ctx, - struct task_struct *task) +static void io_uring_try_cancel_requests(struct io_ring_ctx *ctx, + struct task_struct *task, + struct files_struct *files) { + struct io_task_cancel cancel = { .task = task, .files = files, }; + struct io_uring_task *tctx = task ? task->io_uring : NULL; + while (1) { - struct io_task_cancel cancel = { .task = task, .files = NULL, }; enum io_wq_cancel cret; bool ret = false; - if (ctx->io_wq) { - cret = io_wq_cancel_cb(ctx->io_wq, io_cancel_task_cb, + if (!task) { + ret |= io_uring_try_cancel_iowq(ctx); + } else if (tctx && tctx->io_wq) { + /* + * Cancels requests of all rings, not only @ctx, but + * it's fine as the task is in exit/exec. + */ + cret = io_wq_cancel_cb(tctx->io_wq, io_cancel_task_cb, &cancel, true); ret |= (cret != IO_WQ_CANCEL_NOTFOUND); } /* SQPOLL thread does its own polling */ - if (!(ctx->flags & IORING_SETUP_SQPOLL)) { + if ((!(ctx->flags & IORING_SETUP_SQPOLL) && !files) || + (ctx->sq_data && ctx->sq_data->thread == current)) { while (!list_empty_careful(&ctx->iopoll_list)) { io_iopoll_try_reap_events(ctx); ret = true; } } - ret |= io_poll_remove_all(ctx, task, NULL); - ret |= io_kill_timeouts(ctx, task, NULL); + ret |= io_cancel_defer_files(ctx, task, files); + ret |= io_poll_remove_all(ctx, task, files); + ret |= io_kill_timeouts(ctx, task, files); ret |= io_run_task_work(); + ret |= io_run_ctx_fallback(ctx); + io_cqring_overflow_flush(ctx, true, task, files); if (!ret) break; cond_resched(); } } -static void io_disable_sqo_submit(struct io_ring_ctx *ctx) +static int io_uring_count_inflight(struct io_ring_ctx *ctx, + struct task_struct *task, + struct files_struct *files) { - mutex_lock(&ctx->uring_lock); - ctx->sqo_dead = 1; - mutex_unlock(&ctx->uring_lock); + struct io_kiocb *req; + int cnt = 0; - /* make sure callers enter the ring to get error */ - if (ctx->rings) - io_ring_set_wakeup_flag(ctx); + spin_lock_irq(&ctx->inflight_lock); + list_for_each_entry(req, &ctx->inflight_list, inflight_entry) + cnt += io_match_task(req, task, files); + spin_unlock_irq(&ctx->inflight_lock); + return cnt; } -/* - * We need to iteratively cancel requests, in case a request has dependent - * hard links. These persist even for failure of cancelations, hence keep - * looping until none are found. - */ -static void io_uring_cancel_task_requests(struct io_ring_ctx *ctx, - struct files_struct *files) +static void io_uring_cancel_files(struct io_ring_ctx *ctx, + struct task_struct *task, + struct files_struct *files) { - struct task_struct *task = current; - - if ((ctx->flags & IORING_SETUP_SQPOLL) && ctx->sq_data) { - io_disable_sqo_submit(ctx); - task = ctx->sq_data->thread; - atomic_inc(&task->io_uring->in_idle); - io_sq_thread_park(ctx->sq_data); - } + while (!list_empty_careful(&ctx->inflight_list)) { + DEFINE_WAIT(wait); + int inflight; - io_cancel_defer_files(ctx, task, files); - io_cqring_overflow_flush(ctx, true, task, files); + inflight = io_uring_count_inflight(ctx, task, files); + if (!inflight) + break; - io_uring_cancel_files(ctx, task, files); - if (!files) - __io_uring_cancel_task_requests(ctx, task); + io_uring_try_cancel_requests(ctx, task, files); - if ((ctx->flags & IORING_SETUP_SQPOLL) && ctx->sq_data) { - atomic_dec(&task->io_uring->in_idle); - io_sq_thread_unpark(ctx->sq_data); + prepare_to_wait(&task->io_uring->wait, &wait, + TASK_UNINTERRUPTIBLE); + if (inflight == io_uring_count_inflight(ctx, task, files)) + schedule(); + finish_wait(&task->io_uring->wait, &wait); } } /* * Note that this task has used io_uring. We use it for cancelation purposes. */ -static int io_uring_add_task_file(struct io_ring_ctx *ctx, struct file *file) +static int io_uring_add_task_file(struct io_ring_ctx *ctx) { struct io_uring_task *tctx = current->io_uring; + struct io_tctx_node *node; int ret; if (unlikely(!tctx)) { - ret = io_uring_alloc_task_context(current); + ret = io_uring_alloc_task_context(current, ctx); if (unlikely(ret)) return ret; tctx = current->io_uring; } - if (tctx->last != file) { - void *old = xa_load(&tctx->xa, (unsigned long)file); + if (tctx->last != ctx) { + void *old = xa_load(&tctx->xa, (unsigned long)ctx); if (!old) { - get_file(file); - ret = xa_err(xa_store(&tctx->xa, (unsigned long)file, - file, GFP_KERNEL)); + node = kmalloc(sizeof(*node), GFP_KERNEL); + if (!node) + return -ENOMEM; + node->ctx = ctx; + node->task = current; + + ret = xa_err(xa_store(&tctx->xa, (unsigned long)ctx, + node, GFP_KERNEL)); if (ret) { - fput(file); + kfree(node); return ret; } + + mutex_lock(&ctx->uring_lock); + list_add(&node->ctx_node, &ctx->tctx_list); + mutex_unlock(&ctx->uring_lock); } - tctx->last = file; + tctx->last = ctx; } - - /* - * This is race safe in that the task itself is doing this, hence it - * cannot be going through the exit/cancel paths at the same time. - * This cannot be modified while exit/cancel is running. - */ - if (!tctx->sqpoll && (ctx->flags & IORING_SETUP_SQPOLL)) - tctx->sqpoll = true; - return 0; } /* * Remove this io_uring_file -> task mapping. */ -static void io_uring_del_task_file(struct file *file) +static void io_uring_del_task_file(unsigned long index) { struct io_uring_task *tctx = current->io_uring; + struct io_tctx_node *node; + + if (!tctx) + return; + node = xa_erase(&tctx->xa, index); + if (!node) + return; + + WARN_ON_ONCE(current != node->task); + WARN_ON_ONCE(list_empty(&node->ctx_node)); - if (tctx->last == file) + mutex_lock(&node->ctx->uring_lock); + list_del(&node->ctx_node); + mutex_unlock(&node->ctx->uring_lock); + + if (tctx->last == node->ctx) tctx->last = NULL; - file = xa_erase(&tctx->xa, (unsigned long)file); - if (file) - fput(file); + kfree(node); } -static void io_uring_remove_task_files(struct io_uring_task *tctx) +static void io_uring_clean_tctx(struct io_uring_task *tctx) { - struct file *file; + struct io_tctx_node *node; unsigned long index; - xa_for_each(&tctx->xa, index, file) - io_uring_del_task_file(file); + xa_for_each(&tctx->xa, index, node) + io_uring_del_task_file(index); + if (tctx->io_wq) { + io_wq_put_and_exit(tctx->io_wq); + tctx->io_wq = NULL; + } +} + +static s64 tctx_inflight(struct io_uring_task *tctx) +{ + return percpu_counter_sum(&tctx->inflight); +} + +static void io_sqpoll_cancel_cb(struct callback_head *cb) +{ + struct io_tctx_exit *work = container_of(cb, struct io_tctx_exit, task_work); + struct io_ring_ctx *ctx = work->ctx; + struct io_sq_data *sqd = ctx->sq_data; + + if (sqd->thread) + io_uring_cancel_sqpoll(ctx); + complete(&work->completion); +} + +static void io_sqpoll_cancel_sync(struct io_ring_ctx *ctx) +{ + struct io_sq_data *sqd = ctx->sq_data; + struct io_tctx_exit work = { .ctx = ctx, }; + struct task_struct *task; + + io_sq_thread_park(sqd); + list_del_init(&ctx->sqd_list); + io_sqd_update_thread_idle(sqd); + task = sqd->thread; + if (task) { + init_completion(&work.completion); + init_task_work(&work.task_work, io_sqpoll_cancel_cb); + io_task_work_add_head(&sqd->park_task_work, &work.task_work); + wake_up_process(task); + } + io_sq_thread_unpark(sqd); + + if (task) + wait_for_completion(&work.completion); } void __io_uring_files_cancel(struct files_struct *files) { struct io_uring_task *tctx = current->io_uring; - struct file *file; + struct io_tctx_node *node; unsigned long index; /* make sure overflow events are dropped */ atomic_inc(&tctx->in_idle); - xa_for_each(&tctx->xa, index, file) - io_uring_cancel_task_requests(file->private_data, files); + xa_for_each(&tctx->xa, index, node) { + struct io_ring_ctx *ctx = node->ctx; + + if (ctx->sq_data) { + io_sqpoll_cancel_sync(ctx); + continue; + } + io_uring_cancel_files(ctx, current, files); + if (!files) + io_uring_try_cancel_requests(ctx, current, NULL); + } atomic_dec(&tctx->in_idle); if (files) - io_uring_remove_task_files(tctx); + io_uring_clean_tctx(tctx); } -static s64 tctx_inflight(struct io_uring_task *tctx) +/* should only be called by SQPOLL task */ +static void io_uring_cancel_sqpoll(struct io_ring_ctx *ctx) { - unsigned long index; - struct file *file; + struct io_sq_data *sqd = ctx->sq_data; + struct io_uring_task *tctx = current->io_uring; s64 inflight; + DEFINE_WAIT(wait); - inflight = percpu_counter_sum(&tctx->inflight); - if (!tctx->sqpoll) - return inflight; - - /* - * If we have SQPOLL rings, then we need to iterate and find them, and - * add the pending count for those. - */ - xa_for_each(&tctx->xa, index, file) { - struct io_ring_ctx *ctx = file->private_data; - - if (ctx->flags & IORING_SETUP_SQPOLL) { - struct io_uring_task *__tctx = ctx->sqo_task->io_uring; + WARN_ON_ONCE(!sqd || ctx->sq_data->thread != current); - inflight += percpu_counter_sum(&__tctx->inflight); - } - } + atomic_inc(&tctx->in_idle); + do { + /* read completions before cancelations */ + inflight = tctx_inflight(tctx); + if (!inflight) + break; + io_uring_try_cancel_requests(ctx, current, NULL); - return inflight; + prepare_to_wait(&tctx->wait, &wait, TASK_UNINTERRUPTIBLE); + /* + * If we've seen completions, retry without waiting. This + * avoids a race where a completion comes in before we did + * prepare_to_wait(). + */ + if (inflight == tctx_inflight(tctx)) + schedule(); + finish_wait(&tctx->wait, &wait); + } while (1); + atomic_dec(&tctx->in_idle); } /* @@ -9109,11 +8998,6 @@ void __io_uring_task_cancel(void) /* make sure overflow events are dropped */ atomic_inc(&tctx->in_idle); - - /* trigger io_disable_sqo_submit() */ - if (tctx->sqpoll) - __io_uring_files_cancel(NULL); - do { /* read completions before cancelations */ inflight = tctx_inflight(tctx); @@ -9135,45 +9019,9 @@ void __io_uring_task_cancel(void) atomic_dec(&tctx->in_idle); - io_uring_remove_task_files(tctx); -} - -static int io_uring_flush(struct file *file, void *data) -{ - struct io_uring_task *tctx = current->io_uring; - struct io_ring_ctx *ctx = file->private_data; - - if (fatal_signal_pending(current) || (current->flags & PF_EXITING)) - io_uring_cancel_task_requests(ctx, NULL); - - if (!tctx) - return 0; - - /* we should have cancelled and erased it before PF_EXITING */ - WARN_ON_ONCE((current->flags & PF_EXITING) && - xa_load(&tctx->xa, (unsigned long)file)); - - /* - * fput() is pending, will be 2 if the only other ref is our potential - * task file note. If the task is exiting, drop regardless of count. - */ - if (atomic_long_read(&file->f_count) != 2) - return 0; - - if (ctx->flags & IORING_SETUP_SQPOLL) { - /* there is only one file note, which is owned by sqo_task */ - WARN_ON_ONCE(ctx->sqo_task != current && - xa_load(&tctx->xa, (unsigned long)file)); - /* sqo_dead check is for when this happens after cancellation */ - WARN_ON_ONCE(ctx->sqo_task == current && !ctx->sqo_dead && - !xa_load(&tctx->xa, (unsigned long)file)); - - io_disable_sqo_submit(ctx); - } - - if (!(ctx->flags & IORING_SETUP_SQPOLL) || ctx->sqo_task == current) - io_uring_del_task_file(file); - return 0; + io_uring_clean_tctx(tctx); + /* all current's requests should be gone, we can kill tctx */ + __io_uring_free(current); } static void *io_uring_validate_mmap_request(struct file *file, @@ -9248,29 +9096,20 @@ static unsigned long io_uring_nommu_get_unmapped_area(struct file *file, static int io_sqpoll_wait_sq(struct io_ring_ctx *ctx) { - int ret = 0; DEFINE_WAIT(wait); do { if (!io_sqring_full(ctx)) break; - prepare_to_wait(&ctx->sqo_sq_wait, &wait, TASK_INTERRUPTIBLE); - if (unlikely(ctx->sqo_dead)) { - ret = -EOWNERDEAD; - goto out; - } - if (!io_sqring_full(ctx)) break; - schedule(); } while (!signal_pending(current)); finish_wait(&ctx->sqo_sq_wait, &wait); -out: - return ret; + return 0; } static int io_get_ext_arg(unsigned flags, const void __user *argp, size_t *argsz, @@ -9345,8 +9184,9 @@ SYSCALL_DEFINE6(io_uring_enter, unsigned int, fd, u32, to_submit, io_cqring_overflow_flush(ctx, false, NULL, NULL); ret = -EOWNERDEAD; - if (unlikely(ctx->sqo_dead)) + if (unlikely(ctx->sq_data->thread == NULL)) { goto out; + } if (flags & IORING_ENTER_SQ_WAKEUP) wake_up(&ctx->sq_data->wait); if (flags & IORING_ENTER_SQ_WAIT) { @@ -9356,7 +9196,7 @@ SYSCALL_DEFINE6(io_uring_enter, unsigned int, fd, u32, to_submit, } submitted = to_submit; } else if (to_submit) { - ret = io_uring_add_task_file(ctx, f.file); + ret = io_uring_add_task_file(ctx); if (unlikely(ret)) goto out; mutex_lock(&ctx->uring_lock); @@ -9398,11 +9238,9 @@ out_fput: } #ifdef CONFIG_PROC_FS -static int io_uring_show_cred(int id, void *p, void *data) +static int io_uring_show_cred(struct seq_file *m, unsigned int id, + const struct cred *cred) { - struct io_identity *iod = p; - const struct cred *cred = iod->creds; - struct seq_file *m = data; struct user_namespace *uns = seq_user_ns(m); struct group_info *gi; kernel_cap_t cap; @@ -9446,18 +9284,18 @@ static void __io_uring_show_fdinfo(struct io_ring_ctx *ctx, struct seq_file *m) */ has_lock = mutex_trylock(&ctx->uring_lock); - if (has_lock && (ctx->flags & IORING_SETUP_SQPOLL)) + if (has_lock && (ctx->flags & IORING_SETUP_SQPOLL)) { sq = ctx->sq_data; + if (!sq->thread) + sq = NULL; + } seq_printf(m, "SqThread:\t%d\n", sq ? task_pid_nr(sq->thread) : -1); seq_printf(m, "SqThreadCpu:\t%d\n", sq ? task_cpu(sq->thread) : -1); seq_printf(m, "UserFiles:\t%u\n", ctx->nr_user_files); for (i = 0; has_lock && i < ctx->nr_user_files; i++) { - struct fixed_file_table *table; - struct file *f; + struct file *f = *io_fixed_file_slot(ctx->file_data, i); - table = &ctx->file_data->table[i >> IORING_FILE_TABLE_SHIFT]; - f = table->files[i & IORING_FILE_TABLE_MASK]; if (f) seq_printf(m, "%5u: %s\n", i, file_dentry(f)->d_iname); else @@ -9470,9 +9308,13 @@ static void __io_uring_show_fdinfo(struct io_ring_ctx *ctx, struct seq_file *m) seq_printf(m, "%5u: 0x%llx/%u\n", i, buf->ubuf, (unsigned int) buf->len); } - if (has_lock && !idr_is_empty(&ctx->personality_idr)) { + if (has_lock && !xa_empty(&ctx->personalities)) { + unsigned long index; + const struct cred *cred; + seq_printf(m, "Personalities:\n"); - idr_for_each(&ctx->personality_idr, io_uring_show_cred, m); + xa_for_each(&ctx->personalities, index, cred) + io_uring_show_cred(m, index, cred); } seq_printf(m, "PollList:\n"); spin_lock_irq(&ctx->completion_lock); @@ -9502,7 +9344,6 @@ static void io_uring_show_fdinfo(struct seq_file *m, struct file *f) static const struct file_operations io_uring_fops = { .release = io_uring_release, - .flush = io_uring_flush, .mmap = io_uring_mmap, #ifndef CONFIG_MMU .get_unmapped_area = io_uring_nommu_get_unmapped_area, @@ -9567,7 +9408,7 @@ static int io_uring_install_fd(struct io_ring_ctx *ctx, struct file *file) if (fd < 0) return fd; - ret = io_uring_add_task_file(ctx, file); + ret = io_uring_add_task_file(ctx); if (ret) { put_unused_fd(fd); return ret; @@ -9610,10 +9451,8 @@ static struct file *io_uring_get_file(struct io_ring_ctx *ctx) static int io_uring_create(unsigned entries, struct io_uring_params *p, struct io_uring_params __user *params) { - struct user_struct *user = NULL; struct io_ring_ctx *ctx; struct file *file; - bool limit_mem; int ret; if (!entries) @@ -9653,34 +9492,12 @@ static int io_uring_create(unsigned entries, struct io_uring_params *p, p->cq_entries = 2 * p->sq_entries; } - user = get_uid(current_user()); - limit_mem = !capable(CAP_IPC_LOCK); - - if (limit_mem) { - ret = __io_account_mem(user, - ring_pages(p->sq_entries, p->cq_entries)); - if (ret) { - free_uid(user); - return ret; - } - } - ctx = io_ring_ctx_alloc(p); - if (!ctx) { - if (limit_mem) - __io_unaccount_mem(user, ring_pages(p->sq_entries, - p->cq_entries)); - free_uid(user); + if (!ctx) return -ENOMEM; - } ctx->compat = in_compat_syscall(); - ctx->user = user; - ctx->creds = get_current_cred(); -#ifdef CONFIG_AUDIT - ctx->loginuid = current->loginuid; - ctx->sessionid = current->sessionid; -#endif - ctx->sqo_task = get_task_struct(current); + if (!capable(CAP_IPC_LOCK)) + ctx->user = get_uid(current_user()); /* * This is just grabbed for accounting purposes. When a process exits, @@ -9691,35 +9508,6 @@ static int io_uring_create(unsigned entries, struct io_uring_params *p, mmgrab(current->mm); ctx->mm_account = current->mm; -#ifdef CONFIG_BLK_CGROUP - /* - * The sq thread will belong to the original cgroup it was inited in. - * If the cgroup goes offline (e.g. disabling the io controller), then - * issued bios will be associated with the closest cgroup later in the - * block layer. - */ - rcu_read_lock(); - ctx->sqo_blkcg_css = blkcg_css(); - ret = css_tryget_online(ctx->sqo_blkcg_css); - rcu_read_unlock(); - if (!ret) { - /* don't init against a dying cgroup, have the user try again */ - ctx->sqo_blkcg_css = NULL; - ret = -ENODEV; - goto err; - } -#endif - - /* - * Account memory _before_ installing the file descriptor. Once - * the descriptor is installed, it can get closed at any time. Also - * do this before hitting the general error path, as ring freeing - * will un-account as well. - */ - io_account_mem(ctx, ring_pages(p->sq_entries, p->cq_entries), - ACCT_LOCKED); - ctx->limit_mem = limit_mem; - ret = io_allocate_scq_urings(ctx, p); if (ret) goto err; @@ -9728,9 +9516,6 @@ static int io_uring_create(unsigned entries, struct io_uring_params *p, if (ret) goto err; - if (!(p->flags & IORING_SETUP_R_DISABLED)) - io_sq_offload_start(ctx); - memset(&p->sq_off, 0, sizeof(p->sq_off)); p->sq_off.head = offsetof(struct io_rings, sq.head); p->sq_off.tail = offsetof(struct io_rings, sq.tail); @@ -9753,7 +9538,7 @@ static int io_uring_create(unsigned entries, struct io_uring_params *p, IORING_FEAT_SUBMIT_STABLE | IORING_FEAT_RW_CUR_POS | IORING_FEAT_CUR_PERSONALITY | IORING_FEAT_FAST_POLL | IORING_FEAT_POLL_32BITS | IORING_FEAT_SQPOLL_NONFIXED | - IORING_FEAT_EXT_ARG; + IORING_FEAT_EXT_ARG | IORING_FEAT_NATIVE_WORKERS; if (copy_to_user(params, p, sizeof(*p))) { ret = -EFAULT; @@ -9772,7 +9557,6 @@ static int io_uring_create(unsigned entries, struct io_uring_params *p, */ ret = io_uring_install_fd(ctx, file); if (ret < 0) { - io_disable_sqo_submit(ctx); /* fput will clean it up */ fput(file); return ret; @@ -9781,7 +9565,6 @@ static int io_uring_create(unsigned entries, struct io_uring_params *p, trace_io_uring_create(ret, ctx, p->sq_entries, p->cq_entries, p->flags); return ret; err: - io_disable_sqo_submit(ctx); io_ring_ctx_wait_and_kill(ctx); return ret; } @@ -9859,39 +9642,20 @@ out: static int io_register_personality(struct io_ring_ctx *ctx) { - struct io_identity *id; + const struct cred *creds; + u32 id; int ret; - id = kmalloc(sizeof(*id), GFP_KERNEL); - if (unlikely(!id)) - return -ENOMEM; - - io_init_identity(id); - id->creds = get_current_cred(); + creds = get_current_cred(); - ret = idr_alloc_cyclic(&ctx->personality_idr, id, 1, USHRT_MAX, GFP_KERNEL); - if (ret < 0) { - put_cred(id->creds); - kfree(id); - } + ret = xa_alloc_cyclic(&ctx->personalities, &id, (void *)creds, + XA_LIMIT(0, USHRT_MAX), &ctx->pers_next, GFP_KERNEL); + if (!ret) + return id; + put_cred(creds); return ret; } -static int io_unregister_personality(struct io_ring_ctx *ctx, unsigned id) -{ - struct io_identity *iod; - - iod = idr_remove(&ctx->personality_idr, id); - if (iod) { - put_cred(iod->creds); - if (refcount_dec_and_test(&iod->count)) - kfree(iod); - return 0; - } - - return -EINVAL; -} - static int io_register_restrictions(struct io_ring_ctx *ctx, void __user *arg, unsigned int nr_args) { @@ -9971,9 +9735,8 @@ static int io_register_enable_rings(struct io_ring_ctx *ctx) ctx->restricted = 1; ctx->flags &= ~IORING_SETUP_R_DISABLED; - - io_sq_offload_start(ctx); - + if (ctx->sq_data && wq_has_sleeper(&ctx->sq_data->wait)) + wake_up(&ctx->sq_data->wait); return 0; } @@ -10049,13 +9812,13 @@ static int __io_uring_register(struct io_ring_ctx *ctx, unsigned opcode, switch (opcode) { case IORING_REGISTER_BUFFERS: - ret = io_sqe_buffer_register(ctx, arg, nr_args); + ret = io_sqe_buffers_register(ctx, arg, nr_args); break; case IORING_UNREGISTER_BUFFERS: ret = -EINVAL; if (arg || nr_args) break; - ret = io_sqe_buffer_unregister(ctx); + ret = io_sqe_buffers_unregister(ctx); break; case IORING_REGISTER_FILES: ret = io_sqe_files_register(ctx, arg, nr_args); @@ -10147,6 +9910,8 @@ SYSCALL_DEFINE4(io_uring_register, unsigned int, fd, unsigned int, opcode, ctx = f.file->private_data; + io_run_task_work(); + mutex_lock(&ctx->uring_lock); ret = __io_uring_register(ctx, opcode, arg, nr_args); mutex_unlock(&ctx->uring_lock); @@ -10198,7 +9963,8 @@ static int __init io_uring_init(void) BUILD_BUG_ON(ARRAY_SIZE(io_op_defs) != IORING_OP_LAST); BUILD_BUG_ON(__REQ_F_LAST_BIT >= 8 * sizeof(int)); - req_cachep = KMEM_CACHE(io_kiocb, SLAB_HWCACHE_ALIGN | SLAB_PANIC); + req_cachep = KMEM_CACHE(io_kiocb, SLAB_HWCACHE_ALIGN | SLAB_PANIC | + SLAB_ACCOUNT); return 0; }; __initcall(io_uring_init); diff --git a/fs/iomap/buffered-io.c b/fs/iomap/buffered-io.c index 16a1e82e3aeb..414769a6ad11 100644 --- a/fs/iomap/buffered-io.c +++ b/fs/iomap/buffered-io.c @@ -278,14 +278,14 @@ iomap_readpage_actor(struct inode *inode, loff_t pos, loff_t length, void *data, if (!is_contig || bio_full(ctx->bio, plen)) { gfp_t gfp = mapping_gfp_constraint(page->mapping, GFP_KERNEL); gfp_t orig_gfp = gfp; - int nr_vecs = (length + PAGE_SIZE - 1) >> PAGE_SHIFT; + unsigned int nr_vecs = DIV_ROUND_UP(length, PAGE_SIZE); if (ctx->bio) submit_bio(ctx->bio); if (ctx->rac) /* same as readahead_gfp_mask */ gfp |= __GFP_NORETRY | __GFP_NOWARN; - ctx->bio = bio_alloc(gfp, min(BIO_MAX_PAGES, nr_vecs)); + ctx->bio = bio_alloc(gfp, bio_max_segs(nr_vecs)); /* * If the bio_alloc fails, try it again for a single page to * avoid having to deal with partial page reads. This emulates @@ -1221,7 +1221,7 @@ iomap_alloc_ioend(struct inode *inode, struct iomap_writepage_ctx *wpc, struct iomap_ioend *ioend; struct bio *bio; - bio = bio_alloc_bioset(GFP_NOFS, BIO_MAX_PAGES, &iomap_ioend_bioset); + bio = bio_alloc_bioset(GFP_NOFS, BIO_MAX_VECS, &iomap_ioend_bioset); bio_set_dev(bio, wpc->iomap.bdev); bio->bi_iter.bi_sector = sector; bio->bi_opf = REQ_OP_WRITE | wbc_to_write_flags(wbc); @@ -1252,7 +1252,7 @@ iomap_chain_bio(struct bio *prev) { struct bio *new; - new = bio_alloc(GFP_NOFS, BIO_MAX_PAGES); + new = bio_alloc(GFP_NOFS, BIO_MAX_VECS); bio_copy_dev(new, prev);/* also copies over blkcg information */ new->bi_iter.bi_sector = bio_end_sector(prev); new->bi_opf = prev->bi_opf; @@ -1459,13 +1459,6 @@ iomap_do_writepage(struct page *page, struct writeback_control *wbc, void *data) goto redirty; /* - * Given that we do not allow direct reclaim to call us, we should - * never be called in a recursive filesystem reclaim context. - */ - if (WARN_ON_ONCE(current->flags & PF_MEMALLOC_NOFS)) - goto redirty; - - /* * Is this page beyond the end of the file? * * The page index is less than the end_index, adjust the end_offset diff --git a/fs/iomap/direct-io.c b/fs/iomap/direct-io.c index 933f234d5bec..bdd0d89bbf0a 100644 --- a/fs/iomap/direct-io.c +++ b/fs/iomap/direct-io.c @@ -201,6 +201,34 @@ iomap_dio_zero(struct iomap_dio *dio, struct iomap *iomap, loff_t pos, iomap_dio_submit_bio(dio, iomap, bio, pos); } +/* + * Figure out the bio's operation flags from the dio request, the + * mapping, and whether or not we want FUA. Note that we can end up + * clearing the WRITE_FUA flag in the dio request. + */ +static inline unsigned int +iomap_dio_bio_opflags(struct iomap_dio *dio, struct iomap *iomap, bool use_fua) +{ + unsigned int opflags = REQ_SYNC | REQ_IDLE; + + if (!(dio->flags & IOMAP_DIO_WRITE)) { + WARN_ON_ONCE(iomap->flags & IOMAP_F_ZONE_APPEND); + return REQ_OP_READ; + } + + if (iomap->flags & IOMAP_F_ZONE_APPEND) + opflags |= REQ_OP_ZONE_APPEND; + else + opflags |= REQ_OP_WRITE; + + if (use_fua) + opflags |= REQ_FUA; + else + dio->flags &= ~IOMAP_DIO_WRITE_FUA; + + return opflags; +} + static loff_t iomap_dio_bio_actor(struct inode *inode, loff_t pos, loff_t length, struct iomap_dio *dio, struct iomap *iomap) @@ -208,6 +236,7 @@ iomap_dio_bio_actor(struct inode *inode, loff_t pos, loff_t length, unsigned int blkbits = blksize_bits(bdev_logical_block_size(iomap->bdev)); unsigned int fs_block_size = i_blocksize(inode), pad; unsigned int align = iov_iter_alignment(dio->submit.iter); + unsigned int bio_opf; struct bio *bio; bool need_zeroout = false; bool use_fua = false; @@ -250,11 +279,8 @@ iomap_dio_bio_actor(struct inode *inode, loff_t pos, loff_t length, orig_count = iov_iter_count(dio->submit.iter); iov_iter_truncate(dio->submit.iter, length); - nr_pages = iov_iter_npages(dio->submit.iter, BIO_MAX_PAGES); - if (nr_pages <= 0) { - ret = nr_pages; + if (!iov_iter_count(dio->submit.iter)) goto out; - } if (need_zeroout) { /* zero out from the start of the block to the write offset */ @@ -263,6 +289,14 @@ iomap_dio_bio_actor(struct inode *inode, loff_t pos, loff_t length, iomap_dio_zero(dio, iomap, pos - pad, pad); } + /* + * Set the operation flags early so that bio_iov_iter_get_pages + * can set up the page vector appropriately for a ZONE_APPEND + * operation. + */ + bio_opf = iomap_dio_bio_opflags(dio, iomap, use_fua); + + nr_pages = bio_iov_vecs_to_alloc(dio->submit.iter, BIO_MAX_VECS); do { size_t n; if (dio->error) { @@ -278,6 +312,7 @@ iomap_dio_bio_actor(struct inode *inode, loff_t pos, loff_t length, bio->bi_ioprio = dio->iocb->ki_ioprio; bio->bi_private = dio; bio->bi_end_io = iomap_dio_bio_end_io; + bio->bi_opf = bio_opf; ret = bio_iov_iter_get_pages(bio, dio->submit.iter); if (unlikely(ret)) { @@ -293,14 +328,8 @@ iomap_dio_bio_actor(struct inode *inode, loff_t pos, loff_t length, n = bio->bi_iter.bi_size; if (dio->flags & IOMAP_DIO_WRITE) { - bio->bi_opf = REQ_OP_WRITE | REQ_SYNC | REQ_IDLE; - if (use_fua) - bio->bi_opf |= REQ_FUA; - else - dio->flags &= ~IOMAP_DIO_WRITE_FUA; task_io_account_write(n); } else { - bio->bi_opf = REQ_OP_READ; if (dio->flags & IOMAP_DIO_DIRTY) bio_set_pages_dirty(bio); } @@ -308,7 +337,8 @@ iomap_dio_bio_actor(struct inode *inode, loff_t pos, loff_t length, dio->size += n; copied += n; - nr_pages = iov_iter_npages(dio->submit.iter, BIO_MAX_PAGES); + nr_pages = bio_iov_vecs_to_alloc(dio->submit.iter, + BIO_MAX_VECS); iomap_dio_submit_bio(dio, iomap, bio, pos); pos += n; } while (nr_pages); @@ -420,23 +450,22 @@ iomap_dio_actor(struct inode *inode, loff_t pos, loff_t length, struct iomap_dio * __iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter, const struct iomap_ops *ops, const struct iomap_dio_ops *dops, - bool wait_for_completion) + unsigned int dio_flags) { struct address_space *mapping = iocb->ki_filp->f_mapping; struct inode *inode = file_inode(iocb->ki_filp); size_t count = iov_iter_count(iter); loff_t pos = iocb->ki_pos; loff_t end = iocb->ki_pos + count - 1, ret = 0; - unsigned int flags = IOMAP_DIRECT; + bool wait_for_completion = + is_sync_kiocb(iocb) || (dio_flags & IOMAP_DIO_FORCE_WAIT); + unsigned int iomap_flags = IOMAP_DIRECT; struct blk_plug plug; struct iomap_dio *dio; if (!count) return NULL; - if (WARN_ON(is_sync_kiocb(iocb) && !wait_for_completion)) - return ERR_PTR(-EIO); - dio = kmalloc(sizeof(*dio), GFP_KERNEL); if (!dio) return ERR_PTR(-ENOMEM); @@ -461,7 +490,7 @@ __iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter, if (iter_is_iovec(iter)) dio->flags |= IOMAP_DIO_DIRTY; } else { - flags |= IOMAP_WRITE; + iomap_flags |= IOMAP_WRITE; dio->flags |= IOMAP_DIO_WRITE; /* for data sync or sync, we need sync completion processing */ @@ -483,7 +512,14 @@ __iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter, ret = -EAGAIN; goto out_free_dio; } - flags |= IOMAP_NOWAIT; + iomap_flags |= IOMAP_NOWAIT; + } + + if (dio_flags & IOMAP_DIO_OVERWRITE_ONLY) { + ret = -EAGAIN; + if (pos >= dio->i_size || pos + count > dio->i_size) + goto out_free_dio; + iomap_flags |= IOMAP_OVERWRITE_ONLY; } ret = filemap_write_and_wait_range(mapping, pos, end); @@ -514,7 +550,7 @@ __iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter, blk_start_plug(&plug); do { - ret = iomap_apply(inode, pos, count, flags, ops, dio, + ret = iomap_apply(inode, pos, count, iomap_flags, ops, dio, iomap_dio_actor); if (ret <= 0) { /* magic error code to fall back to buffered I/O */ @@ -598,11 +634,11 @@ EXPORT_SYMBOL_GPL(__iomap_dio_rw); ssize_t iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter, const struct iomap_ops *ops, const struct iomap_dio_ops *dops, - bool wait_for_completion) + unsigned int dio_flags) { struct iomap_dio *dio; - dio = __iomap_dio_rw(iocb, iter, ops, dops, wait_for_completion); + dio = __iomap_dio_rw(iocb, iter, ops, dops, dio_flags); if (IS_ERR_OR_NULL(dio)) return PTR_ERR_OR_ZERO(dio); return iomap_dio_complete(dio); diff --git a/fs/iomap/seek.c b/fs/iomap/seek.c index 107ee80c3568..dab1b02eba5b 100644 --- a/fs/iomap/seek.c +++ b/fs/iomap/seek.c @@ -10,122 +10,17 @@ #include <linux/pagemap.h> #include <linux/pagevec.h> -/* - * Seek for SEEK_DATA / SEEK_HOLE within @page, starting at @lastoff. - * Returns true if found and updates @lastoff to the offset in file. - */ -static bool -page_seek_hole_data(struct inode *inode, struct page *page, loff_t *lastoff, - int whence) -{ - const struct address_space_operations *ops = inode->i_mapping->a_ops; - unsigned int bsize = i_blocksize(inode), off; - bool seek_data = whence == SEEK_DATA; - loff_t poff = page_offset(page); - - if (WARN_ON_ONCE(*lastoff >= poff + PAGE_SIZE)) - return false; - - if (*lastoff < poff) { - /* - * Last offset smaller than the start of the page means we found - * a hole: - */ - if (whence == SEEK_HOLE) - return true; - *lastoff = poff; - } - - /* - * Just check the page unless we can and should check block ranges: - */ - if (bsize == PAGE_SIZE || !ops->is_partially_uptodate) - return PageUptodate(page) == seek_data; - - lock_page(page); - if (unlikely(page->mapping != inode->i_mapping)) - goto out_unlock_not_found; - - for (off = 0; off < PAGE_SIZE; off += bsize) { - if (offset_in_page(*lastoff) >= off + bsize) - continue; - if (ops->is_partially_uptodate(page, off, bsize) == seek_data) { - unlock_page(page); - return true; - } - *lastoff = poff + off + bsize; - } - -out_unlock_not_found: - unlock_page(page); - return false; -} - -/* - * Seek for SEEK_DATA / SEEK_HOLE in the page cache. - * - * Within unwritten extents, the page cache determines which parts are holes - * and which are data: uptodate buffer heads count as data; everything else - * counts as a hole. - * - * Returns the resulting offset on successs, and -ENOENT otherwise. - */ static loff_t -page_cache_seek_hole_data(struct inode *inode, loff_t offset, loff_t length, - int whence) -{ - pgoff_t index = offset >> PAGE_SHIFT; - pgoff_t end = DIV_ROUND_UP(offset + length, PAGE_SIZE); - loff_t lastoff = offset; - struct pagevec pvec; - - if (length <= 0) - return -ENOENT; - - pagevec_init(&pvec); - - do { - unsigned nr_pages, i; - - nr_pages = pagevec_lookup_range(&pvec, inode->i_mapping, &index, - end - 1); - if (nr_pages == 0) - break; - - for (i = 0; i < nr_pages; i++) { - struct page *page = pvec.pages[i]; - - if (page_seek_hole_data(inode, page, &lastoff, whence)) - goto check_range; - lastoff = page_offset(page) + PAGE_SIZE; - } - pagevec_release(&pvec); - } while (index < end); - - /* When no page at lastoff and we are not done, we found a hole. */ - if (whence != SEEK_HOLE) - goto not_found; - -check_range: - if (lastoff < offset + length) - goto out; -not_found: - lastoff = -ENOENT; -out: - pagevec_release(&pvec); - return lastoff; -} - - -static loff_t -iomap_seek_hole_actor(struct inode *inode, loff_t offset, loff_t length, +iomap_seek_hole_actor(struct inode *inode, loff_t start, loff_t length, void *data, struct iomap *iomap, struct iomap *srcmap) { + loff_t offset = start; + switch (iomap->type) { case IOMAP_UNWRITTEN: - offset = page_cache_seek_hole_data(inode, offset, length, - SEEK_HOLE); - if (offset < 0) + offset = mapping_seek_hole_data(inode->i_mapping, start, + start + length, SEEK_HOLE); + if (offset == start + length) return length; fallthrough; case IOMAP_HOLE: @@ -164,15 +59,17 @@ iomap_seek_hole(struct inode *inode, loff_t offset, const struct iomap_ops *ops) EXPORT_SYMBOL_GPL(iomap_seek_hole); static loff_t -iomap_seek_data_actor(struct inode *inode, loff_t offset, loff_t length, +iomap_seek_data_actor(struct inode *inode, loff_t start, loff_t length, void *data, struct iomap *iomap, struct iomap *srcmap) { + loff_t offset = start; + switch (iomap->type) { case IOMAP_HOLE: return length; case IOMAP_UNWRITTEN: - offset = page_cache_seek_hole_data(inode, offset, length, - SEEK_DATA); + offset = mapping_seek_hole_data(inode->i_mapping, start, + start + length, SEEK_DATA); if (offset < 0) return length; fallthrough; diff --git a/fs/iomap/swapfile.c b/fs/iomap/swapfile.c index a648dbf6991e..a5e478de1417 100644 --- a/fs/iomap/swapfile.c +++ b/fs/iomap/swapfile.c @@ -170,6 +170,16 @@ int iomap_swapfile_activate(struct swap_info_struct *sis, return ret; } + /* + * If this swapfile doesn't contain even a single page-aligned + * contiguous range of blocks, reject this useless swapfile to + * prevent confusion later on. + */ + if (isi.nr_pages == 0) { + pr_warn("swapon: Cannot find a single usable page in file.\n"); + return -EINVAL; + } + *pagespan = 1 + isi.highest_ppage - isi.lowest_ppage; sis->max = isi.nr_pages; sis->pages = isi.nr_pages - 1; diff --git a/fs/isofs/dir.c b/fs/isofs/dir.c index f0fe641893a5..b9e6a7ec78be 100644 --- a/fs/isofs/dir.c +++ b/fs/isofs/dir.c @@ -152,6 +152,7 @@ static int do_isofs_readdir(struct inode *inode, struct file *file, printk(KERN_NOTICE "iso9660: Corrupted directory entry" " in block %lu of inode %lu\n", block, inode->i_ino); + brelse(bh); return -EIO; } diff --git a/fs/isofs/inode.c b/fs/isofs/inode.c index ec90773527ee..21edc423b79f 100644 --- a/fs/isofs/inode.c +++ b/fs/isofs/inode.c @@ -339,6 +339,7 @@ static int parse_options(char *options, struct iso9660_options *popt) { char *p; int option; + unsigned int uv; popt->map = 'n'; popt->rock = 1; @@ -434,17 +435,17 @@ static int parse_options(char *options, struct iso9660_options *popt) case Opt_ignore: break; case Opt_uid: - if (match_int(&args[0], &option)) + if (match_uint(&args[0], &uv)) return 0; - popt->uid = make_kuid(current_user_ns(), option); + popt->uid = make_kuid(current_user_ns(), uv); if (!uid_valid(popt->uid)) return 0; popt->uid_set = 1; break; case Opt_gid: - if (match_int(&args[0], &option)) + if (match_uint(&args[0], &uv)) return 0; - popt->gid = make_kgid(current_user_ns(), option); + popt->gid = make_kgid(current_user_ns(), uv); if (!gid_valid(popt->gid)) return 0; popt->gid_set = 1; diff --git a/fs/isofs/namei.c b/fs/isofs/namei.c index 402769881c32..58f80e1b3ac0 100644 --- a/fs/isofs/namei.c +++ b/fs/isofs/namei.c @@ -102,6 +102,7 @@ isofs_find_entry(struct inode *dir, struct dentry *dentry, printk(KERN_NOTICE "iso9660: Corrupted directory entry" " in block %lu of inode %lu\n", block, dir->i_ino); + brelse(bh); return 0; } diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c index 472932b9e6bc..63b526d44886 100644 --- a/fs/jbd2/checkpoint.c +++ b/fs/jbd2/checkpoint.c @@ -416,7 +416,7 @@ int jbd2_cleanup_journal_tail(journal_t *journal) * jbd2_cleanup_journal_tail() doesn't get called all that often. */ if (journal->j_flags & JBD2_BARRIER) - blkdev_issue_flush(journal->j_fs_dev, GFP_NOFS); + blkdev_issue_flush(journal->j_fs_dev); return __jbd2_update_log_tail(journal, first_tid, blocknr); } diff --git a/fs/jbd2/commit.c b/fs/jbd2/commit.c index b121d7d434c6..3cc4ab2ba7f4 100644 --- a/fs/jbd2/commit.c +++ b/fs/jbd2/commit.c @@ -825,7 +825,7 @@ start_journal_io: if (commit_transaction->t_need_data_flush && (journal->j_fs_dev != journal->j_dev) && (journal->j_flags & JBD2_BARRIER)) - blkdev_issue_flush(journal->j_fs_dev, GFP_NOFS); + blkdev_issue_flush(journal->j_fs_dev); /* Done it all: now write the commit record asynchronously. */ if (jbd2_has_feature_async_commit(journal)) { @@ -932,7 +932,7 @@ start_journal_io: stats.run.rs_blocks_logged++; if (jbd2_has_feature_async_commit(journal) && journal->j_flags & JBD2_BARRIER) { - blkdev_issue_flush(journal->j_dev, GFP_NOFS); + blkdev_issue_flush(journal->j_dev); } if (err) diff --git a/fs/jbd2/recovery.c b/fs/jbd2/recovery.c index dc0694fcfcd1..69f18fe20923 100644 --- a/fs/jbd2/recovery.c +++ b/fs/jbd2/recovery.c @@ -326,7 +326,7 @@ int jbd2_journal_recover(journal_t *journal) err = err2; /* Make sure all replayed data is on permanent storage */ if (journal->j_flags & JBD2_BARRIER) { - err2 = blkdev_issue_flush(journal->j_fs_dev, GFP_KERNEL); + err2 = blkdev_issue_flush(journal->j_fs_dev); if (!err) err = err2; } diff --git a/fs/jffs2/acl.c b/fs/jffs2/acl.c index 093ffbd82395..55a79df70d24 100644 --- a/fs/jffs2/acl.c +++ b/fs/jffs2/acl.c @@ -226,7 +226,8 @@ static int __jffs2_set_acl(struct inode *inode, int xprefix, struct posix_acl *a return rc; } -int jffs2_set_acl(struct inode *inode, struct posix_acl *acl, int type) +int jffs2_set_acl(struct user_namespace *mnt_userns, struct inode *inode, + struct posix_acl *acl, int type) { int rc, xprefix; @@ -236,7 +237,8 @@ int jffs2_set_acl(struct inode *inode, struct posix_acl *acl, int type) if (acl) { umode_t mode; - rc = posix_acl_update_mode(inode, &mode, &acl); + rc = posix_acl_update_mode(&init_user_ns, inode, &mode, + &acl); if (rc) return rc; if (inode->i_mode != mode) { diff --git a/fs/jffs2/acl.h b/fs/jffs2/acl.h index 12d0271bdde3..62c50da9d493 100644 --- a/fs/jffs2/acl.h +++ b/fs/jffs2/acl.h @@ -28,7 +28,8 @@ struct jffs2_acl_header { #ifdef CONFIG_JFFS2_FS_POSIX_ACL struct posix_acl *jffs2_get_acl(struct inode *inode, int type); -int jffs2_set_acl(struct inode *inode, struct posix_acl *acl, int type); +int jffs2_set_acl(struct user_namespace *mnt_userns, struct inode *inode, + struct posix_acl *acl, int type); extern int jffs2_init_acl_pre(struct inode *, struct inode *, umode_t *); extern int jffs2_init_acl_post(struct inode *); diff --git a/fs/jffs2/compr_rtime.c b/fs/jffs2/compr_rtime.c index 406d9cc84ba8..79e771ab624f 100644 --- a/fs/jffs2/compr_rtime.c +++ b/fs/jffs2/compr_rtime.c @@ -37,6 +37,9 @@ static int jffs2_rtime_compress(unsigned char *data_in, int outpos = 0; int pos=0; + if (*dstlen <= 3) + return -1; + memset(positions,0,sizeof(positions)); while (pos < (*sourcelen) && outpos <= (*dstlen)-2) { diff --git a/fs/jffs2/dir.c b/fs/jffs2/dir.c index 776493713153..c0aabbcbfd58 100644 --- a/fs/jffs2/dir.c +++ b/fs/jffs2/dir.c @@ -24,18 +24,21 @@ static int jffs2_readdir (struct file *, struct dir_context *); -static int jffs2_create (struct inode *,struct dentry *,umode_t, - bool); +static int jffs2_create (struct user_namespace *, struct inode *, + struct dentry *, umode_t, bool); static struct dentry *jffs2_lookup (struct inode *,struct dentry *, unsigned int); static int jffs2_link (struct dentry *,struct inode *,struct dentry *); static int jffs2_unlink (struct inode *,struct dentry *); -static int jffs2_symlink (struct inode *,struct dentry *,const char *); -static int jffs2_mkdir (struct inode *,struct dentry *,umode_t); +static int jffs2_symlink (struct user_namespace *, struct inode *, + struct dentry *, const char *); +static int jffs2_mkdir (struct user_namespace *, struct inode *,struct dentry *, + umode_t); static int jffs2_rmdir (struct inode *,struct dentry *); -static int jffs2_mknod (struct inode *,struct dentry *,umode_t,dev_t); -static int jffs2_rename (struct inode *, struct dentry *, - struct inode *, struct dentry *, +static int jffs2_mknod (struct user_namespace *, struct inode *,struct dentry *, + umode_t,dev_t); +static int jffs2_rename (struct user_namespace *, struct inode *, + struct dentry *, struct inode *, struct dentry *, unsigned int); const struct file_operations jffs2_dir_operations = @@ -157,8 +160,8 @@ static int jffs2_readdir(struct file *file, struct dir_context *ctx) /***********************************************************************/ -static int jffs2_create(struct inode *dir_i, struct dentry *dentry, - umode_t mode, bool excl) +static int jffs2_create(struct user_namespace *mnt_userns, struct inode *dir_i, + struct dentry *dentry, umode_t mode, bool excl) { struct jffs2_raw_inode *ri; struct jffs2_inode_info *f, *dir_f; @@ -276,7 +279,8 @@ static int jffs2_link (struct dentry *old_dentry, struct inode *dir_i, struct de /***********************************************************************/ -static int jffs2_symlink (struct inode *dir_i, struct dentry *dentry, const char *target) +static int jffs2_symlink (struct user_namespace *mnt_userns, struct inode *dir_i, + struct dentry *dentry, const char *target) { struct jffs2_inode_info *f, *dir_f; struct jffs2_sb_info *c; @@ -438,7 +442,8 @@ static int jffs2_symlink (struct inode *dir_i, struct dentry *dentry, const char } -static int jffs2_mkdir (struct inode *dir_i, struct dentry *dentry, umode_t mode) +static int jffs2_mkdir (struct user_namespace *mnt_userns, struct inode *dir_i, + struct dentry *dentry, umode_t mode) { struct jffs2_inode_info *f, *dir_f; struct jffs2_sb_info *c; @@ -609,7 +614,8 @@ static int jffs2_rmdir (struct inode *dir_i, struct dentry *dentry) return ret; } -static int jffs2_mknod (struct inode *dir_i, struct dentry *dentry, umode_t mode, dev_t rdev) +static int jffs2_mknod (struct user_namespace *mnt_userns, struct inode *dir_i, + struct dentry *dentry, umode_t mode, dev_t rdev) { struct jffs2_inode_info *f, *dir_f; struct jffs2_sb_info *c; @@ -756,7 +762,8 @@ static int jffs2_mknod (struct inode *dir_i, struct dentry *dentry, umode_t mode return ret; } -static int jffs2_rename (struct inode *old_dir_i, struct dentry *old_dentry, +static int jffs2_rename (struct user_namespace *mnt_userns, + struct inode *old_dir_i, struct dentry *old_dentry, struct inode *new_dir_i, struct dentry *new_dentry, unsigned int flags) { diff --git a/fs/jffs2/fs.c b/fs/jffs2/fs.c index 78858f6e9583..2ac410477c4f 100644 --- a/fs/jffs2/fs.c +++ b/fs/jffs2/fs.c @@ -190,18 +190,19 @@ int jffs2_do_setattr (struct inode *inode, struct iattr *iattr) return 0; } -int jffs2_setattr(struct dentry *dentry, struct iattr *iattr) +int jffs2_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *iattr) { struct inode *inode = d_inode(dentry); int rc; - rc = setattr_prepare(dentry, iattr); + rc = setattr_prepare(&init_user_ns, dentry, iattr); if (rc) return rc; rc = jffs2_do_setattr(inode, iattr); if (!rc && (iattr->ia_valid & ATTR_MODE)) - rc = posix_acl_chmod(inode, inode->i_mode); + rc = posix_acl_chmod(&init_user_ns, inode, inode->i_mode); return rc; } diff --git a/fs/jffs2/os-linux.h b/fs/jffs2/os-linux.h index ef1cfa61549e..173eccac691d 100644 --- a/fs/jffs2/os-linux.h +++ b/fs/jffs2/os-linux.h @@ -164,7 +164,7 @@ long jffs2_ioctl(struct file *, unsigned int, unsigned long); extern const struct inode_operations jffs2_symlink_inode_operations; /* fs.c */ -int jffs2_setattr (struct dentry *, struct iattr *); +int jffs2_setattr (struct user_namespace *, struct dentry *, struct iattr *); int jffs2_do_setattr (struct inode *, struct iattr *); struct inode *jffs2_iget(struct super_block *, unsigned long); void jffs2_evict_inode (struct inode *); diff --git a/fs/jffs2/security.c b/fs/jffs2/security.c index c2332e30f218..aef5522551db 100644 --- a/fs/jffs2/security.c +++ b/fs/jffs2/security.c @@ -57,6 +57,7 @@ static int jffs2_security_getxattr(const struct xattr_handler *handler, } static int jffs2_security_setxattr(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *unused, struct inode *inode, const char *name, const void *buffer, size_t size, int flags) diff --git a/fs/jffs2/summary.c b/fs/jffs2/summary.c index be7c8a6a5748..4fe64519870f 100644 --- a/fs/jffs2/summary.c +++ b/fs/jffs2/summary.c @@ -783,6 +783,8 @@ static int jffs2_sum_write_data(struct jffs2_sb_info *c, struct jffs2_eraseblock dbg_summary("Writing unknown RWCOMPAT_COPY node type %x\n", je16_to_cpu(temp->u.nodetype)); jffs2_sum_disable_collecting(c->summary); + /* The above call removes the list, nothing more to do */ + goto bail_rwcompat; } else { BUG(); /* unknown node in summary information */ } @@ -794,6 +796,7 @@ static int jffs2_sum_write_data(struct jffs2_sb_info *c, struct jffs2_eraseblock c->summary->sum_num--; } + bail_rwcompat: jffs2_sum_reset_collected(c->summary); diff --git a/fs/jffs2/xattr_trusted.c b/fs/jffs2/xattr_trusted.c index 5d6030826c52..cc3f24883e7d 100644 --- a/fs/jffs2/xattr_trusted.c +++ b/fs/jffs2/xattr_trusted.c @@ -25,6 +25,7 @@ static int jffs2_trusted_getxattr(const struct xattr_handler *handler, } static int jffs2_trusted_setxattr(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *unused, struct inode *inode, const char *name, const void *buffer, size_t size, int flags) diff --git a/fs/jffs2/xattr_user.c b/fs/jffs2/xattr_user.c index 9d027b4abcf9..fb945977c013 100644 --- a/fs/jffs2/xattr_user.c +++ b/fs/jffs2/xattr_user.c @@ -25,6 +25,7 @@ static int jffs2_user_getxattr(const struct xattr_handler *handler, } static int jffs2_user_setxattr(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *unused, struct inode *inode, const char *name, const void *buffer, size_t size, int flags) diff --git a/fs/jfs/acl.c b/fs/jfs/acl.c index 92cc0ac2d1fc..43c285c3d2a7 100644 --- a/fs/jfs/acl.c +++ b/fs/jfs/acl.c @@ -91,7 +91,8 @@ out: return rc; } -int jfs_set_acl(struct inode *inode, struct posix_acl *acl, int type) +int jfs_set_acl(struct user_namespace *mnt_userns, struct inode *inode, + struct posix_acl *acl, int type) { int rc; tid_t tid; @@ -101,7 +102,7 @@ int jfs_set_acl(struct inode *inode, struct posix_acl *acl, int type) tid = txBegin(inode->i_sb, 0); mutex_lock(&JFS_IP(inode)->commit_mutex); if (type == ACL_TYPE_ACCESS && acl) { - rc = posix_acl_update_mode(inode, &mode, &acl); + rc = posix_acl_update_mode(&init_user_ns, inode, &mode, &acl); if (rc) goto end_tx; if (mode != inode->i_mode) diff --git a/fs/jfs/file.c b/fs/jfs/file.c index 930d2701f206..28b70e7c7dd4 100644 --- a/fs/jfs/file.c +++ b/fs/jfs/file.c @@ -85,12 +85,13 @@ static int jfs_release(struct inode *inode, struct file *file) return 0; } -int jfs_setattr(struct dentry *dentry, struct iattr *iattr) +int jfs_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *iattr) { struct inode *inode = d_inode(dentry); int rc; - rc = setattr_prepare(dentry, iattr); + rc = setattr_prepare(&init_user_ns, dentry, iattr); if (rc) return rc; @@ -118,11 +119,11 @@ int jfs_setattr(struct dentry *dentry, struct iattr *iattr) jfs_truncate(inode); } - setattr_copy(inode, iattr); + setattr_copy(&init_user_ns, inode, iattr); mark_inode_dirty(inode); if (iattr->ia_valid & ATTR_MODE) - rc = posix_acl_chmod(inode, inode->i_mode); + rc = posix_acl_chmod(&init_user_ns, inode, inode->i_mode); return rc; } diff --git a/fs/jfs/ioctl.c b/fs/jfs/ioctl.c index 10ee0ecca1a8..2581d4db58ff 100644 --- a/fs/jfs/ioctl.c +++ b/fs/jfs/ioctl.c @@ -76,7 +76,7 @@ long jfs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) if (err) return err; - if (!inode_owner_or_capable(inode)) { + if (!inode_owner_or_capable(&init_user_ns, inode)) { err = -EACCES; goto setflags_out; } diff --git a/fs/jfs/jfs_acl.h b/fs/jfs/jfs_acl.h index 9f8f92dd6f84..7ae389a7a366 100644 --- a/fs/jfs/jfs_acl.h +++ b/fs/jfs/jfs_acl.h @@ -8,7 +8,8 @@ #ifdef CONFIG_JFS_POSIX_ACL struct posix_acl *jfs_get_acl(struct inode *inode, int type); -int jfs_set_acl(struct inode *inode, struct posix_acl *acl, int type); +int jfs_set_acl(struct user_namespace *mnt_userns, struct inode *inode, + struct posix_acl *acl, int type); int jfs_init_acl(tid_t, struct inode *, struct inode *); #else diff --git a/fs/jfs/jfs_dmap.c b/fs/jfs/jfs_dmap.c index 94b7c1cb5ceb..7aee15608619 100644 --- a/fs/jfs/jfs_dmap.c +++ b/fs/jfs/jfs_dmap.c @@ -1656,7 +1656,7 @@ s64 dbDiscardAG(struct inode *ip, int agno, s64 minlen) } else if (rc == -ENOSPC) { /* search for next smaller log2 block */ l2nb = BLKSTOL2(nblocks) - 1; - nblocks = 1 << l2nb; + nblocks = 1LL << l2nb; } else { /* Trim any already allocated blocks */ jfs_error(bmp->db_ipbmap->i_sb, "-EIO\n"); diff --git a/fs/jfs/jfs_filsys.h b/fs/jfs/jfs_filsys.h index 1e899298f7f0..b5d702df7111 100644 --- a/fs/jfs/jfs_filsys.h +++ b/fs/jfs/jfs_filsys.h @@ -268,5 +268,6 @@ * fsck() must be run to repair */ #define FM_EXTENDFS 0x00000008 /* file system extendfs() in progress */ +#define FM_STATE_MAX 0x0000000f /* max value of s_state */ #endif /* _H_JFS_FILSYS */ diff --git a/fs/jfs/jfs_inode.c b/fs/jfs/jfs_inode.c index 4cef170630db..59379089e939 100644 --- a/fs/jfs/jfs_inode.c +++ b/fs/jfs/jfs_inode.c @@ -64,7 +64,7 @@ struct inode *ialloc(struct inode *parent, umode_t mode) goto fail_put; } - inode_init_owner(inode, parent, mode); + inode_init_owner(&init_user_ns, inode, parent, mode); /* * New inodes need to save sane values on disk when * uid & gid mount options are used diff --git a/fs/jfs/jfs_inode.h b/fs/jfs/jfs_inode.h index 70a0d12e427e..01daa0cb0ae5 100644 --- a/fs/jfs/jfs_inode.h +++ b/fs/jfs/jfs_inode.h @@ -26,7 +26,7 @@ extern struct dentry *jfs_fh_to_parent(struct super_block *sb, struct fid *fid, int fh_len, int fh_type); extern void jfs_set_inode_flags(struct inode *); extern int jfs_get_block(struct inode *, sector_t, struct buffer_head *, int); -extern int jfs_setattr(struct dentry *, struct iattr *); +extern int jfs_setattr(struct user_namespace *, struct dentry *, struct iattr *); extern const struct address_space_operations jfs_aops; extern const struct inode_operations jfs_dir_inode_operations; diff --git a/fs/jfs/jfs_mount.c b/fs/jfs/jfs_mount.c index 2935d4c776ec..5d7d7170c03c 100644 --- a/fs/jfs/jfs_mount.c +++ b/fs/jfs/jfs_mount.c @@ -37,6 +37,7 @@ #include <linux/fs.h> #include <linux/buffer_head.h> #include <linux/blkdev.h> +#include <linux/log2.h> #include "jfs_incore.h" #include "jfs_filsys.h" @@ -366,6 +367,15 @@ static int chkSuper(struct super_block *sb) sbi->bsize = bsize; sbi->l2bsize = le16_to_cpu(j_sb->s_l2bsize); + /* check some fields for possible corruption */ + if (sbi->l2bsize != ilog2((u32)bsize) || + j_sb->pad != 0 || + le32_to_cpu(j_sb->s_state) > FM_STATE_MAX) { + rc = -EINVAL; + jfs_err("jfs_mount: Mount Failure: superblock is corrupt!"); + goto out; + } + /* * For now, ignore s_pbsize, l2bfactor. All I/O going through buffer * cache. diff --git a/fs/jfs/jfs_txnmgr.c b/fs/jfs/jfs_txnmgr.c index dca8edd2378c..053295cd7bc6 100644 --- a/fs/jfs/jfs_txnmgr.c +++ b/fs/jfs/jfs_txnmgr.c @@ -148,10 +148,10 @@ static struct { /* * forward references */ -static int diLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, - struct tlock * tlck, struct commit * cd); -static int dataLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, - struct tlock * tlck); +static void diLog(struct jfs_log *log, struct tblock *tblk, struct lrd *lrd, + struct tlock *tlck, struct commit *cd); +static void dataLog(struct jfs_log *log, struct tblock *tblk, struct lrd *lrd, + struct tlock *tlck); static void dtLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, struct tlock * tlck); static void mapLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, @@ -159,8 +159,8 @@ static void mapLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, static void txAllocPMap(struct inode *ip, struct maplock * maplock, struct tblock * tblk); static void txForce(struct tblock * tblk); -static int txLog(struct jfs_log * log, struct tblock * tblk, - struct commit * cd); +static void txLog(struct jfs_log *log, struct tblock *tblk, + struct commit *cd); static void txUpdateMap(struct tblock * tblk); static void txRelease(struct tblock * tblk); static void xtLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, @@ -1256,8 +1256,7 @@ int txCommit(tid_t tid, /* transaction identifier */ * * txUpdateMap() resets XAD_NEW in XAD. */ - if ((rc = txLog(log, tblk, &cd))) - goto TheEnd; + txLog(log, tblk, &cd); /* * Ensure that inode isn't reused before @@ -1365,9 +1364,8 @@ int txCommit(tid_t tid, /* transaction identifier */ * * RETURN : */ -static int txLog(struct jfs_log * log, struct tblock * tblk, struct commit * cd) +static void txLog(struct jfs_log *log, struct tblock *tblk, struct commit *cd) { - int rc = 0; struct inode *ip; lid_t lid; struct tlock *tlck; @@ -1414,7 +1412,7 @@ static int txLog(struct jfs_log * log, struct tblock * tblk, struct commit * cd) } } - return rc; + return; } /* @@ -1422,10 +1420,9 @@ static int txLog(struct jfs_log * log, struct tblock * tblk, struct commit * cd) * * function: log inode tlock and format maplock to update bmap; */ -static int diLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, - struct tlock * tlck, struct commit * cd) +static void diLog(struct jfs_log *log, struct tblock *tblk, struct lrd *lrd, + struct tlock *tlck, struct commit *cd) { - int rc = 0; struct metapage *mp; pxd_t *pxd; struct pxd_lock *pxdlock; @@ -1527,7 +1524,7 @@ static int diLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, } #endif /* _JFS_WIP */ - return rc; + return; } /* @@ -1535,8 +1532,8 @@ static int diLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, * * function: log data tlock */ -static int dataLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, - struct tlock * tlck) +static void dataLog(struct jfs_log *log, struct tblock *tblk, struct lrd *lrd, + struct tlock *tlck) { struct metapage *mp; pxd_t *pxd; @@ -1562,7 +1559,7 @@ static int dataLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, metapage_homeok(mp); discard_metapage(mp); tlck->mp = NULL; - return 0; + return; } PXDaddress(pxd, mp->index); @@ -1573,7 +1570,7 @@ static int dataLog(struct jfs_log * log, struct tblock * tblk, struct lrd * lrd, /* mark page as homeward bound */ tlck->flag |= tlckWRITEPAGE; - return 0; + return; } /* diff --git a/fs/jfs/namei.c b/fs/jfs/namei.c index 7a55d14cc1af..9abed0d750e5 100644 --- a/fs/jfs/namei.c +++ b/fs/jfs/namei.c @@ -59,8 +59,8 @@ static inline void free_ea_wmap(struct inode *inode) * RETURN: Errors from subroutines * */ -static int jfs_create(struct inode *dip, struct dentry *dentry, umode_t mode, - bool excl) +static int jfs_create(struct user_namespace *mnt_userns, struct inode *dip, + struct dentry *dentry, umode_t mode, bool excl) { int rc = 0; tid_t tid; /* transaction id */ @@ -192,7 +192,8 @@ static int jfs_create(struct inode *dip, struct dentry *dentry, umode_t mode, * note: * EACCES: user needs search+write permission on the parent directory */ -static int jfs_mkdir(struct inode *dip, struct dentry *dentry, umode_t mode) +static int jfs_mkdir(struct user_namespace *mnt_userns, struct inode *dip, + struct dentry *dentry, umode_t mode) { int rc = 0; tid_t tid; /* transaction id */ @@ -868,8 +869,8 @@ static int jfs_link(struct dentry *old_dentry, * an intermediate result whose length exceeds PATH_MAX [XPG4.2] */ -static int jfs_symlink(struct inode *dip, struct dentry *dentry, - const char *name) +static int jfs_symlink(struct user_namespace *mnt_userns, struct inode *dip, + struct dentry *dentry, const char *name) { int rc; tid_t tid; @@ -1058,9 +1059,9 @@ static int jfs_symlink(struct inode *dip, struct dentry *dentry, * * FUNCTION: rename a file or directory */ -static int jfs_rename(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry, - unsigned int flags) +static int jfs_rename(struct user_namespace *mnt_userns, struct inode *old_dir, + struct dentry *old_dentry, struct inode *new_dir, + struct dentry *new_dentry, unsigned int flags) { struct btstack btstack; ino_t ino; @@ -1344,8 +1345,8 @@ static int jfs_rename(struct inode *old_dir, struct dentry *old_dentry, * * FUNCTION: Create a special file (device) */ -static int jfs_mknod(struct inode *dir, struct dentry *dentry, - umode_t mode, dev_t rdev) +static int jfs_mknod(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, dev_t rdev) { struct jfs_inode_info *jfs_ip; struct btstack btstack; diff --git a/fs/jfs/super.c b/fs/jfs/super.c index b2dc4d1f9dcc..1f0ffabbde56 100644 --- a/fs/jfs/super.c +++ b/fs/jfs/super.c @@ -551,7 +551,6 @@ static int jfs_fill_super(struct super_block *sb, void *data, int silent) ret = -ENOMEM; goto out_unload; } - inode->i_ino = 0; inode->i_size = i_size_read(sb->s_bdev->bd_inode); inode->i_mapping->a_ops = &jfs_metapage_aops; inode_fake_hash(inode); diff --git a/fs/jfs/xattr.c b/fs/jfs/xattr.c index db41e7803163..f9273f6901c8 100644 --- a/fs/jfs/xattr.c +++ b/fs/jfs/xattr.c @@ -932,6 +932,7 @@ static int jfs_xattr_get(const struct xattr_handler *handler, } static int jfs_xattr_set(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *unused, struct inode *inode, const char *name, const void *value, size_t size, int flags) @@ -950,6 +951,7 @@ static int jfs_xattr_get_os2(const struct xattr_handler *handler, } static int jfs_xattr_set_os2(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *unused, struct inode *inode, const char *name, const void *value, size_t size, int flags) diff --git a/fs/kernfs/dir.c b/fs/kernfs/dir.c index 7a53eed69fef..7e0e62deab53 100644 --- a/fs/kernfs/dir.c +++ b/fs/kernfs/dir.c @@ -1110,7 +1110,8 @@ static struct dentry *kernfs_iop_lookup(struct inode *dir, return ret; } -static int kernfs_iop_mkdir(struct inode *dir, struct dentry *dentry, +static int kernfs_iop_mkdir(struct user_namespace *mnt_userns, + struct inode *dir, struct dentry *dentry, umode_t mode) { struct kernfs_node *parent = dir->i_private; @@ -1147,7 +1148,8 @@ static int kernfs_iop_rmdir(struct inode *dir, struct dentry *dentry) return ret; } -static int kernfs_iop_rename(struct inode *old_dir, struct dentry *old_dentry, +static int kernfs_iop_rename(struct user_namespace *mnt_userns, + struct inode *old_dir, struct dentry *old_dentry, struct inode *new_dir, struct dentry *new_dentry, unsigned int flags) { diff --git a/fs/kernfs/inode.c b/fs/kernfs/inode.c index fc2469a20fed..d73950fc3d57 100644 --- a/fs/kernfs/inode.c +++ b/fs/kernfs/inode.c @@ -112,7 +112,8 @@ int kernfs_setattr(struct kernfs_node *kn, const struct iattr *iattr) return ret; } -int kernfs_iop_setattr(struct dentry *dentry, struct iattr *iattr) +int kernfs_iop_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *iattr) { struct inode *inode = d_inode(dentry); struct kernfs_node *kn = inode->i_private; @@ -122,7 +123,7 @@ int kernfs_iop_setattr(struct dentry *dentry, struct iattr *iattr) return -EINVAL; mutex_lock(&kernfs_mutex); - error = setattr_prepare(dentry, iattr); + error = setattr_prepare(&init_user_ns, dentry, iattr); if (error) goto out; @@ -131,7 +132,7 @@ int kernfs_iop_setattr(struct dentry *dentry, struct iattr *iattr) goto out; /* this ignores size changes */ - setattr_copy(inode, iattr); + setattr_copy(&init_user_ns, inode, iattr); out: mutex_unlock(&kernfs_mutex); @@ -183,7 +184,8 @@ static void kernfs_refresh_inode(struct kernfs_node *kn, struct inode *inode) set_nlink(inode, kn->dir.subdirs + 2); } -int kernfs_iop_getattr(const struct path *path, struct kstat *stat, +int kernfs_iop_getattr(struct user_namespace *mnt_userns, + const struct path *path, struct kstat *stat, u32 request_mask, unsigned int query_flags) { struct inode *inode = d_inode(path->dentry); @@ -193,7 +195,7 @@ int kernfs_iop_getattr(const struct path *path, struct kstat *stat, kernfs_refresh_inode(kn, inode); mutex_unlock(&kernfs_mutex); - generic_fillattr(inode, stat); + generic_fillattr(&init_user_ns, inode, stat); return 0; } @@ -272,7 +274,8 @@ void kernfs_evict_inode(struct inode *inode) kernfs_put(kn); } -int kernfs_iop_permission(struct inode *inode, int mask) +int kernfs_iop_permission(struct user_namespace *mnt_userns, + struct inode *inode, int mask) { struct kernfs_node *kn; @@ -285,7 +288,7 @@ int kernfs_iop_permission(struct inode *inode, int mask) kernfs_refresh_inode(kn, inode); mutex_unlock(&kernfs_mutex); - return generic_permission(inode, mask); + return generic_permission(&init_user_ns, inode, mask); } int kernfs_xattr_get(struct kernfs_node *kn, const char *name, @@ -319,6 +322,7 @@ static int kernfs_vfs_xattr_get(const struct xattr_handler *handler, } static int kernfs_vfs_xattr_set(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *unused, struct inode *inode, const char *suffix, const void *value, size_t size, int flags) @@ -385,6 +389,7 @@ static int kernfs_vfs_user_xattr_rm(struct kernfs_node *kn, } static int kernfs_vfs_user_xattr_set(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *unused, struct inode *inode, const char *suffix, const void *value, size_t size, int flags) diff --git a/fs/kernfs/kernfs-internal.h b/fs/kernfs/kernfs-internal.h index 7ee97ef59184..ccc3b44f6306 100644 --- a/fs/kernfs/kernfs-internal.h +++ b/fs/kernfs/kernfs-internal.h @@ -89,9 +89,12 @@ extern struct kmem_cache *kernfs_node_cache, *kernfs_iattrs_cache; */ extern const struct xattr_handler *kernfs_xattr_handlers[]; void kernfs_evict_inode(struct inode *inode); -int kernfs_iop_permission(struct inode *inode, int mask); -int kernfs_iop_setattr(struct dentry *dentry, struct iattr *iattr); -int kernfs_iop_getattr(const struct path *path, struct kstat *stat, +int kernfs_iop_permission(struct user_namespace *mnt_userns, + struct inode *inode, int mask); +int kernfs_iop_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *iattr); +int kernfs_iop_getattr(struct user_namespace *mnt_userns, + const struct path *path, struct kstat *stat, u32 request_mask, unsigned int query_flags); ssize_t kernfs_iop_listxattr(struct dentry *dentry, char *buf, size_t size); int __kernfs_setattr(struct kernfs_node *kn, const struct iattr *iattr); diff --git a/fs/libfs.c b/fs/libfs.c index d1c3bade9f30..e2de5401abca 100644 --- a/fs/libfs.c +++ b/fs/libfs.c @@ -27,11 +27,12 @@ #include "internal.h" -int simple_getattr(const struct path *path, struct kstat *stat, - u32 request_mask, unsigned int query_flags) +int simple_getattr(struct user_namespace *mnt_userns, const struct path *path, + struct kstat *stat, u32 request_mask, + unsigned int query_flags) { struct inode *inode = d_inode(path->dentry); - generic_fillattr(inode, stat); + generic_fillattr(&init_user_ns, inode, stat); stat->blocks = inode->i_mapping->nrpages << (PAGE_SHIFT - 9); return 0; } @@ -447,9 +448,9 @@ int simple_rmdir(struct inode *dir, struct dentry *dentry) } EXPORT_SYMBOL(simple_rmdir); -int simple_rename(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry, - unsigned int flags) +int simple_rename(struct user_namespace *mnt_userns, struct inode *old_dir, + struct dentry *old_dentry, struct inode *new_dir, + struct dentry *new_dentry, unsigned int flags) { struct inode *inode = d_inode(old_dentry); int they_are_dirs = d_is_dir(old_dentry); @@ -492,18 +493,19 @@ EXPORT_SYMBOL(simple_rename); * on simple regular filesystems. Anything that needs to change on-disk * or wire state on size changes needs its own setattr method. */ -int simple_setattr(struct dentry *dentry, struct iattr *iattr) +int simple_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *iattr) { struct inode *inode = d_inode(dentry); int error; - error = setattr_prepare(dentry, iattr); + error = setattr_prepare(mnt_userns, dentry, iattr); if (error) return error; if (iattr->ia_valid & ATTR_SIZE) truncate_setsize(inode, iattr->ia_size); - setattr_copy(inode, iattr); + setattr_copy(mnt_userns, inode, iattr); mark_inode_dirty(inode); return 0; } @@ -1117,7 +1119,7 @@ int generic_file_fsync(struct file *file, loff_t start, loff_t end, err = __generic_file_fsync(file, start, end, datasync); if (err) return err; - return blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL); + return blkdev_issue_flush(inode->i_sb->s_bdev); } EXPORT_SYMBOL(generic_file_fsync); @@ -1214,11 +1216,6 @@ static int anon_set_page_dirty(struct page *page) return 0; }; -/* - * A single inode exists for all anon_inode files. Contrary to pipes, - * anon_inode inodes have no associated per-instance data, so we need - * only allocate one of them. - */ struct inode *alloc_anon_inode(struct super_block *s) { static const struct address_space_operations anon_aops = { @@ -1300,15 +1297,17 @@ static struct dentry *empty_dir_lookup(struct inode *dir, struct dentry *dentry, return ERR_PTR(-ENOENT); } -static int empty_dir_getattr(const struct path *path, struct kstat *stat, +static int empty_dir_getattr(struct user_namespace *mnt_userns, + const struct path *path, struct kstat *stat, u32 request_mask, unsigned int query_flags) { struct inode *inode = d_inode(path->dentry); - generic_fillattr(inode, stat); + generic_fillattr(&init_user_ns, inode, stat); return 0; } -static int empty_dir_setattr(struct dentry *dentry, struct iattr *attr) +static int empty_dir_setattr(struct user_namespace *mnt_userns, + struct dentry *dentry, struct iattr *attr) { return -EPERM; } @@ -1388,8 +1387,8 @@ static bool needs_casefold(const struct inode *dir) * * Return: 0 if names match, 1 if mismatch, or -ERRNO */ -int generic_ci_d_compare(const struct dentry *dentry, unsigned int len, - const char *str, const struct qstr *name) +static int generic_ci_d_compare(const struct dentry *dentry, unsigned int len, + const char *str, const struct qstr *name) { const struct dentry *parent = READ_ONCE(dentry->d_parent); const struct inode *dir = READ_ONCE(parent->d_inode); @@ -1426,7 +1425,6 @@ fallback: return 1; return !!memcmp(str, name->name, len); } -EXPORT_SYMBOL(generic_ci_d_compare); /** * generic_ci_d_hash - generic d_hash implementation for casefolding filesystems @@ -1435,7 +1433,7 @@ EXPORT_SYMBOL(generic_ci_d_compare); * * Return: 0 if hash was successful or unchanged, and -EINVAL on error */ -int generic_ci_d_hash(const struct dentry *dentry, struct qstr *str) +static int generic_ci_d_hash(const struct dentry *dentry, struct qstr *str) { const struct inode *dir = READ_ONCE(dentry->d_inode); struct super_block *sb = dentry->d_sb; @@ -1450,7 +1448,6 @@ int generic_ci_d_hash(const struct dentry *dentry, struct qstr *str) return -EINVAL; return 0; } -EXPORT_SYMBOL(generic_ci_d_hash); static const struct dentry_operations generic_ci_dentry_ops = { .d_hash = generic_ci_d_hash, diff --git a/fs/lockd/svc4proc.c b/fs/lockd/svc4proc.c index fa41dda39925..4c10fb5138f1 100644 --- a/fs/lockd/svc4proc.c +++ b/fs/lockd/svc4proc.c @@ -512,6 +512,7 @@ const struct svc_procedure nlmsvc_procedures4[24] = { .pc_argsize = sizeof(struct nlm_void), .pc_ressize = sizeof(struct nlm_void), .pc_xdrressize = St, + .pc_name = "NULL", }, [NLMPROC_TEST] = { .pc_func = nlm4svc_proc_test, @@ -520,6 +521,7 @@ const struct svc_procedure nlmsvc_procedures4[24] = { .pc_argsize = sizeof(struct nlm_args), .pc_ressize = sizeof(struct nlm_res), .pc_xdrressize = Ck+St+2+No+Rg, + .pc_name = "TEST", }, [NLMPROC_LOCK] = { .pc_func = nlm4svc_proc_lock, @@ -528,6 +530,7 @@ const struct svc_procedure nlmsvc_procedures4[24] = { .pc_argsize = sizeof(struct nlm_args), .pc_ressize = sizeof(struct nlm_res), .pc_xdrressize = Ck+St, + .pc_name = "LOCK", }, [NLMPROC_CANCEL] = { .pc_func = nlm4svc_proc_cancel, @@ -536,6 +539,7 @@ const struct svc_procedure nlmsvc_procedures4[24] = { .pc_argsize = sizeof(struct nlm_args), .pc_ressize = sizeof(struct nlm_res), .pc_xdrressize = Ck+St, + .pc_name = "CANCEL", }, [NLMPROC_UNLOCK] = { .pc_func = nlm4svc_proc_unlock, @@ -544,6 +548,7 @@ const struct svc_procedure nlmsvc_procedures4[24] = { .pc_argsize = sizeof(struct nlm_args), .pc_ressize = sizeof(struct nlm_res), .pc_xdrressize = Ck+St, + .pc_name = "UNLOCK", }, [NLMPROC_GRANTED] = { .pc_func = nlm4svc_proc_granted, @@ -552,6 +557,7 @@ const struct svc_procedure nlmsvc_procedures4[24] = { .pc_argsize = sizeof(struct nlm_args), .pc_ressize = sizeof(struct nlm_res), .pc_xdrressize = Ck+St, + .pc_name = "GRANTED", }, [NLMPROC_TEST_MSG] = { .pc_func = nlm4svc_proc_test_msg, @@ -560,6 +566,7 @@ const struct svc_procedure nlmsvc_procedures4[24] = { .pc_argsize = sizeof(struct nlm_args), .pc_ressize = sizeof(struct nlm_void), .pc_xdrressize = St, + .pc_name = "TEST_MSG", }, [NLMPROC_LOCK_MSG] = { .pc_func = nlm4svc_proc_lock_msg, @@ -568,6 +575,7 @@ const struct svc_procedure nlmsvc_procedures4[24] = { .pc_argsize = sizeof(struct nlm_args), .pc_ressize = sizeof(struct nlm_void), .pc_xdrressize = St, + .pc_name = "LOCK_MSG", }, [NLMPROC_CANCEL_MSG] = { .pc_func = nlm4svc_proc_cancel_msg, @@ -576,6 +584,7 @@ const struct svc_procedure nlmsvc_procedures4[24] = { .pc_argsize = sizeof(struct nlm_args), .pc_ressize = sizeof(struct nlm_void), .pc_xdrressize = St, + .pc_name = "CANCEL_MSG", }, [NLMPROC_UNLOCK_MSG] = { .pc_func = nlm4svc_proc_unlock_msg, @@ -584,6 +593,7 @@ const struct svc_procedure nlmsvc_procedures4[24] = { .pc_argsize = sizeof(struct nlm_args), .pc_ressize = sizeof(struct nlm_void), .pc_xdrressize = St, + .pc_name = "UNLOCK_MSG", }, [NLMPROC_GRANTED_MSG] = { .pc_func = nlm4svc_proc_granted_msg, @@ -592,6 +602,7 @@ const struct svc_procedure nlmsvc_procedures4[24] = { .pc_argsize = sizeof(struct nlm_args), .pc_ressize = sizeof(struct nlm_void), .pc_xdrressize = St, + .pc_name = "GRANTED_MSG", }, [NLMPROC_TEST_RES] = { .pc_func = nlm4svc_proc_null, @@ -600,6 +611,7 @@ const struct svc_procedure nlmsvc_procedures4[24] = { .pc_argsize = sizeof(struct nlm_res), .pc_ressize = sizeof(struct nlm_void), .pc_xdrressize = St, + .pc_name = "TEST_RES", }, [NLMPROC_LOCK_RES] = { .pc_func = nlm4svc_proc_null, @@ -608,6 +620,7 @@ const struct svc_procedure nlmsvc_procedures4[24] = { .pc_argsize = sizeof(struct nlm_res), .pc_ressize = sizeof(struct nlm_void), .pc_xdrressize = St, + .pc_name = "LOCK_RES", }, [NLMPROC_CANCEL_RES] = { .pc_func = nlm4svc_proc_null, @@ -616,6 +629,7 @@ const struct svc_procedure nlmsvc_procedures4[24] = { .pc_argsize = sizeof(struct nlm_res), .pc_ressize = sizeof(struct nlm_void), .pc_xdrressize = St, + .pc_name = "CANCEL_RES", }, [NLMPROC_UNLOCK_RES] = { .pc_func = nlm4svc_proc_null, @@ -624,6 +638,7 @@ const struct svc_procedure nlmsvc_procedures4[24] = { .pc_argsize = sizeof(struct nlm_res), .pc_ressize = sizeof(struct nlm_void), .pc_xdrressize = St, + .pc_name = "UNLOCK_RES", }, [NLMPROC_GRANTED_RES] = { .pc_func = nlm4svc_proc_granted_res, @@ -632,6 +647,7 @@ const struct svc_procedure nlmsvc_procedures4[24] = { .pc_argsize = sizeof(struct nlm_res), .pc_ressize = sizeof(struct nlm_void), .pc_xdrressize = St, + .pc_name = "GRANTED_RES", }, [NLMPROC_NSM_NOTIFY] = { .pc_func = nlm4svc_proc_sm_notify, @@ -640,6 +656,7 @@ const struct svc_procedure nlmsvc_procedures4[24] = { .pc_argsize = sizeof(struct nlm_reboot), .pc_ressize = sizeof(struct nlm_void), .pc_xdrressize = St, + .pc_name = "SM_NOTIFY", }, [17] = { .pc_func = nlm4svc_proc_unused, @@ -648,6 +665,7 @@ const struct svc_procedure nlmsvc_procedures4[24] = { .pc_argsize = sizeof(struct nlm_void), .pc_ressize = sizeof(struct nlm_void), .pc_xdrressize = 0, + .pc_name = "UNUSED", }, [18] = { .pc_func = nlm4svc_proc_unused, @@ -656,6 +674,7 @@ const struct svc_procedure nlmsvc_procedures4[24] = { .pc_argsize = sizeof(struct nlm_void), .pc_ressize = sizeof(struct nlm_void), .pc_xdrressize = 0, + .pc_name = "UNUSED", }, [19] = { .pc_func = nlm4svc_proc_unused, @@ -664,6 +683,7 @@ const struct svc_procedure nlmsvc_procedures4[24] = { .pc_argsize = sizeof(struct nlm_void), .pc_ressize = sizeof(struct nlm_void), .pc_xdrressize = 0, + .pc_name = "UNUSED", }, [NLMPROC_SHARE] = { .pc_func = nlm4svc_proc_share, @@ -672,6 +692,7 @@ const struct svc_procedure nlmsvc_procedures4[24] = { .pc_argsize = sizeof(struct nlm_args), .pc_ressize = sizeof(struct nlm_res), .pc_xdrressize = Ck+St+1, + .pc_name = "SHARE", }, [NLMPROC_UNSHARE] = { .pc_func = nlm4svc_proc_unshare, @@ -680,6 +701,7 @@ const struct svc_procedure nlmsvc_procedures4[24] = { .pc_argsize = sizeof(struct nlm_args), .pc_ressize = sizeof(struct nlm_res), .pc_xdrressize = Ck+St+1, + .pc_name = "UNSHARE", }, [NLMPROC_NM_LOCK] = { .pc_func = nlm4svc_proc_nm_lock, @@ -688,6 +710,7 @@ const struct svc_procedure nlmsvc_procedures4[24] = { .pc_argsize = sizeof(struct nlm_args), .pc_ressize = sizeof(struct nlm_res), .pc_xdrressize = Ck+St, + .pc_name = "NM_LOCK", }, [NLMPROC_FREE_ALL] = { .pc_func = nlm4svc_proc_free_all, @@ -696,5 +719,6 @@ const struct svc_procedure nlmsvc_procedures4[24] = { .pc_argsize = sizeof(struct nlm_args), .pc_ressize = sizeof(struct nlm_void), .pc_xdrressize = St, + .pc_name = "FREE_ALL", }, }; diff --git a/fs/lockd/svcproc.c b/fs/lockd/svcproc.c index 50855f2c1f4b..4ae4b63b5392 100644 --- a/fs/lockd/svcproc.c +++ b/fs/lockd/svcproc.c @@ -554,6 +554,7 @@ const struct svc_procedure nlmsvc_procedures[24] = { .pc_argsize = sizeof(struct nlm_void), .pc_ressize = sizeof(struct nlm_void), .pc_xdrressize = St, + .pc_name = "NULL", }, [NLMPROC_TEST] = { .pc_func = nlmsvc_proc_test, @@ -562,6 +563,7 @@ const struct svc_procedure nlmsvc_procedures[24] = { .pc_argsize = sizeof(struct nlm_args), .pc_ressize = sizeof(struct nlm_res), .pc_xdrressize = Ck+St+2+No+Rg, + .pc_name = "TEST", }, [NLMPROC_LOCK] = { .pc_func = nlmsvc_proc_lock, @@ -570,6 +572,7 @@ const struct svc_procedure nlmsvc_procedures[24] = { .pc_argsize = sizeof(struct nlm_args), .pc_ressize = sizeof(struct nlm_res), .pc_xdrressize = Ck+St, + .pc_name = "LOCK", }, [NLMPROC_CANCEL] = { .pc_func = nlmsvc_proc_cancel, @@ -578,6 +581,7 @@ const struct svc_procedure nlmsvc_procedures[24] = { .pc_argsize = sizeof(struct nlm_args), .pc_ressize = sizeof(struct nlm_res), .pc_xdrressize = Ck+St, + .pc_name = "CANCEL", }, [NLMPROC_UNLOCK] = { .pc_func = nlmsvc_proc_unlock, @@ -586,6 +590,7 @@ const struct svc_procedure nlmsvc_procedures[24] = { .pc_argsize = sizeof(struct nlm_args), .pc_ressize = sizeof(struct nlm_res), .pc_xdrressize = Ck+St, + .pc_name = "UNLOCK", }, [NLMPROC_GRANTED] = { .pc_func = nlmsvc_proc_granted, @@ -594,6 +599,7 @@ const struct svc_procedure nlmsvc_procedures[24] = { .pc_argsize = sizeof(struct nlm_args), .pc_ressize = sizeof(struct nlm_res), .pc_xdrressize = Ck+St, + .pc_name = "GRANTED", }, [NLMPROC_TEST_MSG] = { .pc_func = nlmsvc_proc_test_msg, @@ -602,6 +608,7 @@ const struct svc_procedure nlmsvc_procedures[24] = { .pc_argsize = sizeof(struct nlm_args), .pc_ressize = sizeof(struct nlm_void), .pc_xdrressize = St, + .pc_name = "TEST_MSG", }, [NLMPROC_LOCK_MSG] = { .pc_func = nlmsvc_proc_lock_msg, @@ -610,6 +617,7 @@ const struct svc_procedure nlmsvc_procedures[24] = { .pc_argsize = sizeof(struct nlm_args), .pc_ressize = sizeof(struct nlm_void), .pc_xdrressize = St, + .pc_name = "LOCK_MSG", }, [NLMPROC_CANCEL_MSG] = { .pc_func = nlmsvc_proc_cancel_msg, @@ -618,6 +626,7 @@ const struct svc_procedure nlmsvc_procedures[24] = { .pc_argsize = sizeof(struct nlm_args), .pc_ressize = sizeof(struct nlm_void), .pc_xdrressize = St, + .pc_name = "CANCEL_MSG", }, [NLMPROC_UNLOCK_MSG] = { .pc_func = nlmsvc_proc_unlock_msg, @@ -626,6 +635,7 @@ const struct svc_procedure nlmsvc_procedures[24] = { .pc_argsize = sizeof(struct nlm_args), .pc_ressize = sizeof(struct nlm_void), .pc_xdrressize = St, + .pc_name = "UNLOCK_MSG", }, [NLMPROC_GRANTED_MSG] = { .pc_func = nlmsvc_proc_granted_msg, @@ -634,6 +644,7 @@ const struct svc_procedure nlmsvc_procedures[24] = { .pc_argsize = sizeof(struct nlm_args), .pc_ressize = sizeof(struct nlm_void), .pc_xdrressize = St, + .pc_name = "GRANTED_MSG", }, [NLMPROC_TEST_RES] = { .pc_func = nlmsvc_proc_null, @@ -642,6 +653,7 @@ const struct svc_procedure nlmsvc_procedures[24] = { .pc_argsize = sizeof(struct nlm_res), .pc_ressize = sizeof(struct nlm_void), .pc_xdrressize = St, + .pc_name = "TEST_RES", }, [NLMPROC_LOCK_RES] = { .pc_func = nlmsvc_proc_null, @@ -650,6 +662,7 @@ const struct svc_procedure nlmsvc_procedures[24] = { .pc_argsize = sizeof(struct nlm_res), .pc_ressize = sizeof(struct nlm_void), .pc_xdrressize = St, + .pc_name = "LOCK_RES", }, [NLMPROC_CANCEL_RES] = { .pc_func = nlmsvc_proc_null, @@ -658,6 +671,7 @@ const struct svc_procedure nlmsvc_procedures[24] = { .pc_argsize = sizeof(struct nlm_res), .pc_ressize = sizeof(struct nlm_void), .pc_xdrressize = St, + .pc_name = "CANCEL_RES", }, [NLMPROC_UNLOCK_RES] = { .pc_func = nlmsvc_proc_null, @@ -666,6 +680,7 @@ const struct svc_procedure nlmsvc_procedures[24] = { .pc_argsize = sizeof(struct nlm_res), .pc_ressize = sizeof(struct nlm_void), .pc_xdrressize = St, + .pc_name = "UNLOCK_RES", }, [NLMPROC_GRANTED_RES] = { .pc_func = nlmsvc_proc_granted_res, @@ -674,6 +689,7 @@ const struct svc_procedure nlmsvc_procedures[24] = { .pc_argsize = sizeof(struct nlm_res), .pc_ressize = sizeof(struct nlm_void), .pc_xdrressize = St, + .pc_name = "GRANTED_RES", }, [NLMPROC_NSM_NOTIFY] = { .pc_func = nlmsvc_proc_sm_notify, @@ -682,6 +698,7 @@ const struct svc_procedure nlmsvc_procedures[24] = { .pc_argsize = sizeof(struct nlm_reboot), .pc_ressize = sizeof(struct nlm_void), .pc_xdrressize = St, + .pc_name = "SM_NOTIFY", }, [17] = { .pc_func = nlmsvc_proc_unused, @@ -690,6 +707,7 @@ const struct svc_procedure nlmsvc_procedures[24] = { .pc_argsize = sizeof(struct nlm_void), .pc_ressize = sizeof(struct nlm_void), .pc_xdrressize = St, + .pc_name = "UNUSED", }, [18] = { .pc_func = nlmsvc_proc_unused, @@ -698,6 +716,7 @@ const struct svc_procedure nlmsvc_procedures[24] = { .pc_argsize = sizeof(struct nlm_void), .pc_ressize = sizeof(struct nlm_void), .pc_xdrressize = St, + .pc_name = "UNUSED", }, [19] = { .pc_func = nlmsvc_proc_unused, @@ -706,6 +725,7 @@ const struct svc_procedure nlmsvc_procedures[24] = { .pc_argsize = sizeof(struct nlm_void), .pc_ressize = sizeof(struct nlm_void), .pc_xdrressize = St, + .pc_name = "UNUSED", }, [NLMPROC_SHARE] = { .pc_func = nlmsvc_proc_share, @@ -714,6 +734,7 @@ const struct svc_procedure nlmsvc_procedures[24] = { .pc_argsize = sizeof(struct nlm_args), .pc_ressize = sizeof(struct nlm_res), .pc_xdrressize = Ck+St+1, + .pc_name = "SHARE", }, [NLMPROC_UNSHARE] = { .pc_func = nlmsvc_proc_unshare, @@ -722,6 +743,7 @@ const struct svc_procedure nlmsvc_procedures[24] = { .pc_argsize = sizeof(struct nlm_args), .pc_ressize = sizeof(struct nlm_res), .pc_xdrressize = Ck+St+1, + .pc_name = "UNSHARE", }, [NLMPROC_NM_LOCK] = { .pc_func = nlmsvc_proc_nm_lock, @@ -730,6 +752,7 @@ const struct svc_procedure nlmsvc_procedures[24] = { .pc_argsize = sizeof(struct nlm_args), .pc_ressize = sizeof(struct nlm_res), .pc_xdrressize = Ck+St, + .pc_name = "NM_LOCK", }, [NLMPROC_FREE_ALL] = { .pc_func = nlmsvc_proc_free_all, @@ -738,5 +761,6 @@ const struct svc_procedure nlmsvc_procedures[24] = { .pc_argsize = sizeof(struct nlm_args), .pc_ressize = sizeof(struct nlm_void), .pc_xdrressize = 0, + .pc_name = "FREE_ALL", }, }; diff --git a/fs/locks.c b/fs/locks.c index 99ca97e81b7a..6125d2de39b8 100644 --- a/fs/locks.c +++ b/fs/locks.c @@ -1808,9 +1808,6 @@ check_conflicting_open(struct file *filp, const long arg, int flags) if (flags & FL_LAYOUT) return 0; - if (flags & FL_DELEG) - /* We leave these checks to the caller. */ - return 0; if (arg == F_RDLCK) return inode_is_open_for_write(inode) ? -EAGAIN : 0; diff --git a/fs/minix/bitmap.c b/fs/minix/bitmap.c index f4e5e5181a14..9115948c624e 100644 --- a/fs/minix/bitmap.c +++ b/fs/minix/bitmap.c @@ -252,7 +252,7 @@ struct inode *minix_new_inode(const struct inode *dir, umode_t mode, int *error) iput(inode); return NULL; } - inode_init_owner(inode, dir, mode); + inode_init_owner(&init_user_ns, inode, dir, mode); inode->i_ino = j; inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode); inode->i_blocks = 0; diff --git a/fs/minix/file.c b/fs/minix/file.c index c50b0a20fcd9..6a7bd2d9eec0 100644 --- a/fs/minix/file.c +++ b/fs/minix/file.c @@ -22,12 +22,13 @@ const struct file_operations minix_file_operations = { .splice_read = generic_file_splice_read, }; -static int minix_setattr(struct dentry *dentry, struct iattr *attr) +static int minix_setattr(struct user_namespace *mnt_userns, + struct dentry *dentry, struct iattr *attr) { struct inode *inode = d_inode(dentry); int error; - error = setattr_prepare(dentry, attr); + error = setattr_prepare(&init_user_ns, dentry, attr); if (error) return error; @@ -41,7 +42,7 @@ static int minix_setattr(struct dentry *dentry, struct iattr *attr) minix_truncate(inode); } - setattr_copy(inode, attr); + setattr_copy(&init_user_ns, inode, attr); mark_inode_dirty(inode); return 0; } diff --git a/fs/minix/inode.c b/fs/minix/inode.c index 34f546404aa1..a532a99bbe81 100644 --- a/fs/minix/inode.c +++ b/fs/minix/inode.c @@ -652,13 +652,13 @@ static int minix_write_inode(struct inode *inode, struct writeback_control *wbc) return err; } -int minix_getattr(const struct path *path, struct kstat *stat, - u32 request_mask, unsigned int flags) +int minix_getattr(struct user_namespace *mnt_userns, const struct path *path, + struct kstat *stat, u32 request_mask, unsigned int flags) { struct super_block *sb = path->dentry->d_sb; struct inode *inode = d_inode(path->dentry); - generic_fillattr(inode, stat); + generic_fillattr(&init_user_ns, inode, stat); if (INODE_VERSION(inode) == MINIX_V1) stat->blocks = (BLOCK_SIZE / 512) * V1_minix_blocks(stat->size, sb); else diff --git a/fs/minix/minix.h b/fs/minix/minix.h index 168d45d3de73..202173368025 100644 --- a/fs/minix/minix.h +++ b/fs/minix/minix.h @@ -51,7 +51,8 @@ extern unsigned long minix_count_free_inodes(struct super_block *sb); extern int minix_new_block(struct inode * inode); extern void minix_free_block(struct inode *inode, unsigned long block); extern unsigned long minix_count_free_blocks(struct super_block *sb); -extern int minix_getattr(const struct path *, struct kstat *, u32, unsigned int); +extern int minix_getattr(struct user_namespace *, const struct path *, + struct kstat *, u32, unsigned int); extern int minix_prepare_chunk(struct page *page, loff_t pos, unsigned len); extern void V1_minix_truncate(struct inode *); diff --git a/fs/minix/namei.c b/fs/minix/namei.c index 1a6084d2b02e..937fa5fae2b8 100644 --- a/fs/minix/namei.c +++ b/fs/minix/namei.c @@ -33,7 +33,8 @@ static struct dentry *minix_lookup(struct inode * dir, struct dentry *dentry, un return d_splice_alias(inode, dentry); } -static int minix_mknod(struct inode * dir, struct dentry *dentry, umode_t mode, dev_t rdev) +static int minix_mknod(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, dev_t rdev) { int error; struct inode *inode; @@ -51,7 +52,8 @@ static int minix_mknod(struct inode * dir, struct dentry *dentry, umode_t mode, return error; } -static int minix_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode) +static int minix_tmpfile(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { int error; struct inode *inode = minix_new_inode(dir, mode, &error); @@ -63,14 +65,14 @@ static int minix_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode) return error; } -static int minix_create(struct inode *dir, struct dentry *dentry, umode_t mode, - bool excl) +static int minix_create(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, bool excl) { - return minix_mknod(dir, dentry, mode, 0); + return minix_mknod(mnt_userns, dir, dentry, mode, 0); } -static int minix_symlink(struct inode * dir, struct dentry *dentry, - const char * symname) +static int minix_symlink(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, const char *symname) { int err = -ENAMETOOLONG; int i = strlen(symname)+1; @@ -109,7 +111,8 @@ static int minix_link(struct dentry * old_dentry, struct inode * dir, return add_nondir(dentry, inode); } -static int minix_mkdir(struct inode * dir, struct dentry *dentry, umode_t mode) +static int minix_mkdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { struct inode * inode; int err; @@ -181,8 +184,9 @@ static int minix_rmdir(struct inode * dir, struct dentry *dentry) return err; } -static int minix_rename(struct inode * old_dir, struct dentry *old_dentry, - struct inode * new_dir, struct dentry *new_dentry, +static int minix_rename(struct user_namespace *mnt_userns, + struct inode *old_dir, struct dentry *old_dentry, + struct inode *new_dir, struct dentry *new_dentry, unsigned int flags) { struct inode * old_inode = d_inode(old_dentry); diff --git a/fs/mount.h b/fs/mount.h index ce6c376e0bc2..0b6e08cf8afb 100644 --- a/fs/mount.h +++ b/fs/mount.h @@ -124,16 +124,6 @@ static inline void get_mnt_ns(struct mnt_namespace *ns) extern seqlock_t mount_lock; -static inline void lock_mount_hash(void) -{ - write_seqlock(&mount_lock); -} - -static inline void unlock_mount_hash(void) -{ - write_sequnlock(&mount_lock); -} - struct proc_mounts { struct mnt_namespace *ns; struct path root; diff --git a/fs/mpage.c b/fs/mpage.c index 830e6cc2a9e7..334e7d09aa65 100644 --- a/fs/mpage.c +++ b/fs/mpage.c @@ -304,9 +304,7 @@ alloc_new: goto out; } args->bio = mpage_alloc(bdev, blocks[0] << (blkbits - 9), - min_t(int, args->nr_pages, - BIO_MAX_PAGES), - gfp); + bio_max_segs(args->nr_pages), gfp); if (args->bio == NULL) goto confused; } @@ -618,7 +616,7 @@ alloc_new: goto out; } bio = mpage_alloc(bdev, blocks[0] << (blkbits - 9), - BIO_MAX_PAGES, GFP_NOFS|__GFP_HIGH); + BIO_MAX_VECS, GFP_NOFS|__GFP_HIGH); if (bio == NULL) goto confused; diff --git a/fs/namei.c b/fs/namei.c index 78443a85480a..216f16e74351 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -259,7 +259,24 @@ void putname(struct filename *name) __putname(name); } -static int check_acl(struct inode *inode, int mask) +/** + * check_acl - perform ACL permission checking + * @mnt_userns: user namespace of the mount the inode was found from + * @inode: inode to check permissions on + * @mask: right to check for (%MAY_READ, %MAY_WRITE, %MAY_EXEC ...) + * + * This function performs the ACL permission checking. Since this function + * retrieve POSIX acls it needs to know whether it is called from a blocking or + * non-blocking context and thus cares about the MAY_NOT_BLOCK bit. + * + * If the inode has been found through an idmapped mount the user namespace of + * the vfsmount must be passed through @mnt_userns. This function will then take + * care to map the inode according to @mnt_userns before checking permissions. + * On non-idmapped mounts or if permission checking is to be performed on the + * raw inode simply passs init_user_ns. + */ +static int check_acl(struct user_namespace *mnt_userns, + struct inode *inode, int mask) { #ifdef CONFIG_FS_POSIX_ACL struct posix_acl *acl; @@ -271,14 +288,14 @@ static int check_acl(struct inode *inode, int mask) /* no ->get_acl() calls in RCU mode... */ if (is_uncached_acl(acl)) return -ECHILD; - return posix_acl_permission(inode, acl, mask); + return posix_acl_permission(mnt_userns, inode, acl, mask); } acl = get_acl(inode, ACL_TYPE_ACCESS); if (IS_ERR(acl)) return PTR_ERR(acl); if (acl) { - int error = posix_acl_permission(inode, acl, mask); + int error = posix_acl_permission(mnt_userns, inode, acl, mask); posix_acl_release(acl); return error; } @@ -287,18 +304,31 @@ static int check_acl(struct inode *inode, int mask) return -EAGAIN; } -/* - * This does the basic UNIX permission checking. +/** + * acl_permission_check - perform basic UNIX permission checking + * @mnt_userns: user namespace of the mount the inode was found from + * @inode: inode to check permissions on + * @mask: right to check for (%MAY_READ, %MAY_WRITE, %MAY_EXEC ...) + * + * This function performs the basic UNIX permission checking. Since this + * function may retrieve POSIX acls it needs to know whether it is called from a + * blocking or non-blocking context and thus cares about the MAY_NOT_BLOCK bit. * - * Note that the POSIX ACL check cares about the MAY_NOT_BLOCK bit, - * for RCU walking. + * If the inode has been found through an idmapped mount the user namespace of + * the vfsmount must be passed through @mnt_userns. This function will then take + * care to map the inode according to @mnt_userns before checking permissions. + * On non-idmapped mounts or if permission checking is to be performed on the + * raw inode simply passs init_user_ns. */ -static int acl_permission_check(struct inode *inode, int mask) +static int acl_permission_check(struct user_namespace *mnt_userns, + struct inode *inode, int mask) { unsigned int mode = inode->i_mode; + kuid_t i_uid; /* Are we the owner? If so, ACL's don't matter */ - if (likely(uid_eq(current_fsuid(), inode->i_uid))) { + i_uid = i_uid_into_mnt(mnt_userns, inode); + if (likely(uid_eq(current_fsuid(), i_uid))) { mask &= 7; mode >>= 6; return (mask & ~mode) ? -EACCES : 0; @@ -306,7 +336,7 @@ static int acl_permission_check(struct inode *inode, int mask) /* Do we have ACL's? */ if (IS_POSIXACL(inode) && (mode & S_IRWXG)) { - int error = check_acl(inode, mask); + int error = check_acl(mnt_userns, inode, mask); if (error != -EAGAIN) return error; } @@ -320,7 +350,8 @@ static int acl_permission_check(struct inode *inode, int mask) * about? Need to check group ownership if so. */ if (mask & (mode ^ (mode >> 3))) { - if (in_group_p(inode->i_gid)) + kgid_t kgid = i_gid_into_mnt(mnt_userns, inode); + if (in_group_p(kgid)) mode >>= 3; } @@ -330,6 +361,7 @@ static int acl_permission_check(struct inode *inode, int mask) /** * generic_permission - check for access rights on a Posix-like filesystem + * @mnt_userns: user namespace of the mount the inode was found from * @inode: inode to check access rights for * @mask: right to check for (%MAY_READ, %MAY_WRITE, %MAY_EXEC, * %MAY_NOT_BLOCK ...) @@ -342,25 +374,33 @@ static int acl_permission_check(struct inode *inode, int mask) * generic_permission is rcu-walk aware. It returns -ECHILD in case an rcu-walk * request cannot be satisfied (eg. requires blocking or too much complexity). * It would then be called again in ref-walk mode. + * + * If the inode has been found through an idmapped mount the user namespace of + * the vfsmount must be passed through @mnt_userns. This function will then take + * care to map the inode according to @mnt_userns before checking permissions. + * On non-idmapped mounts or if permission checking is to be performed on the + * raw inode simply passs init_user_ns. */ -int generic_permission(struct inode *inode, int mask) +int generic_permission(struct user_namespace *mnt_userns, struct inode *inode, + int mask) { int ret; /* * Do the basic permission checks. */ - ret = acl_permission_check(inode, mask); + ret = acl_permission_check(mnt_userns, inode, mask); if (ret != -EACCES) return ret; if (S_ISDIR(inode->i_mode)) { /* DACs are overridable for directories */ if (!(mask & MAY_WRITE)) - if (capable_wrt_inode_uidgid(inode, + if (capable_wrt_inode_uidgid(mnt_userns, inode, CAP_DAC_READ_SEARCH)) return 0; - if (capable_wrt_inode_uidgid(inode, CAP_DAC_OVERRIDE)) + if (capable_wrt_inode_uidgid(mnt_userns, inode, + CAP_DAC_OVERRIDE)) return 0; return -EACCES; } @@ -370,7 +410,8 @@ int generic_permission(struct inode *inode, int mask) */ mask &= MAY_READ | MAY_WRITE | MAY_EXEC; if (mask == MAY_READ) - if (capable_wrt_inode_uidgid(inode, CAP_DAC_READ_SEARCH)) + if (capable_wrt_inode_uidgid(mnt_userns, inode, + CAP_DAC_READ_SEARCH)) return 0; /* * Read/write DACs are always overridable. @@ -378,31 +419,38 @@ int generic_permission(struct inode *inode, int mask) * at least one exec bit set. */ if (!(mask & MAY_EXEC) || (inode->i_mode & S_IXUGO)) - if (capable_wrt_inode_uidgid(inode, CAP_DAC_OVERRIDE)) + if (capable_wrt_inode_uidgid(mnt_userns, inode, + CAP_DAC_OVERRIDE)) return 0; return -EACCES; } EXPORT_SYMBOL(generic_permission); -/* +/** + * do_inode_permission - UNIX permission checking + * @mnt_userns: user namespace of the mount the inode was found from + * @inode: inode to check permissions on + * @mask: right to check for (%MAY_READ, %MAY_WRITE, %MAY_EXEC ...) + * * We _really_ want to just do "generic_permission()" without * even looking at the inode->i_op values. So we keep a cache * flag in inode->i_opflags, that says "this has not special * permission function, use the fast case". */ -static inline int do_inode_permission(struct inode *inode, int mask) +static inline int do_inode_permission(struct user_namespace *mnt_userns, + struct inode *inode, int mask) { if (unlikely(!(inode->i_opflags & IOP_FASTPERM))) { if (likely(inode->i_op->permission)) - return inode->i_op->permission(inode, mask); + return inode->i_op->permission(mnt_userns, inode, mask); /* This gets set once for the inode lifetime */ spin_lock(&inode->i_lock); inode->i_opflags |= IOP_FASTPERM; spin_unlock(&inode->i_lock); } - return generic_permission(inode, mask); + return generic_permission(mnt_userns, inode, mask); } /** @@ -427,8 +475,9 @@ static int sb_permission(struct super_block *sb, struct inode *inode, int mask) /** * inode_permission - Check for access rights to a given inode - * @inode: Inode to check permission on - * @mask: Right to check for (%MAY_READ, %MAY_WRITE, %MAY_EXEC) + * @mnt_userns: User namespace of the mount the inode was found from + * @inode: Inode to check permission on + * @mask: Right to check for (%MAY_READ, %MAY_WRITE, %MAY_EXEC) * * Check for read/write/execute permissions on an inode. We use fs[ug]id for * this, letting us set arbitrary permissions for filesystem access without @@ -436,7 +485,8 @@ static int sb_permission(struct super_block *sb, struct inode *inode, int mask) * * When checking for MAY_APPEND, MAY_WRITE must also be set in @mask. */ -int inode_permission(struct inode *inode, int mask) +int inode_permission(struct user_namespace *mnt_userns, + struct inode *inode, int mask) { int retval; @@ -456,11 +506,11 @@ int inode_permission(struct inode *inode, int mask) * written back improperly if their true value is unknown * to the vfs. */ - if (HAS_UNMAPPED_ID(inode)) + if (HAS_UNMAPPED_ID(mnt_userns, inode)) return -EACCES; } - retval = do_inode_permission(inode, mask); + retval = do_inode_permission(mnt_userns, inode, mask); if (retval) return retval; @@ -630,6 +680,11 @@ static inline bool legitimize_path(struct nameidata *nd, static bool legitimize_links(struct nameidata *nd) { int i; + if (unlikely(nd->flags & LOOKUP_CACHED)) { + drop_links(nd); + nd->depth = 0; + return false; + } for (i = 0; i < nd->depth; i++) { struct saved *last = nd->stack + i; if (unlikely(!legitimize_path(nd, &last->link, last->seq))) { @@ -669,17 +724,17 @@ static bool legitimize_root(struct nameidata *nd) */ /** - * unlazy_walk - try to switch to ref-walk mode. + * try_to_unlazy - try to switch to ref-walk mode. * @nd: nameidata pathwalk data - * Returns: 0 on success, -ECHILD on failure + * Returns: true on success, false on failure * - * unlazy_walk attempts to legitimize the current nd->path and nd->root + * try_to_unlazy attempts to legitimize the current nd->path and nd->root * for ref-walk mode. * Must be called from rcu-walk context. - * Nothing should touch nameidata between unlazy_walk() failure and + * Nothing should touch nameidata between try_to_unlazy() failure and * terminate_walk(). */ -static int unlazy_walk(struct nameidata *nd) +static bool try_to_unlazy(struct nameidata *nd) { struct dentry *parent = nd->path.dentry; @@ -694,30 +749,30 @@ static int unlazy_walk(struct nameidata *nd) goto out; rcu_read_unlock(); BUG_ON(nd->inode != parent->d_inode); - return 0; + return true; out1: nd->path.mnt = NULL; nd->path.dentry = NULL; out: rcu_read_unlock(); - return -ECHILD; + return false; } /** - * unlazy_child - try to switch to ref-walk mode. + * try_to_unlazy_next - try to switch to ref-walk mode. * @nd: nameidata pathwalk data - * @dentry: child of nd->path.dentry - * @seq: seq number to check dentry against - * Returns: 0 on success, -ECHILD on failure + * @dentry: next dentry to step into + * @seq: seq number to check @dentry against + * Returns: true on success, false on failure * - * unlazy_child attempts to legitimize the current nd->path, nd->root and dentry - * for ref-walk mode. @dentry must be a path found by a do_lookup call on - * @nd. Must be called from rcu-walk context. - * Nothing should touch nameidata between unlazy_child() failure and + * Similar to to try_to_unlazy(), but here we have the next dentry already + * picked by rcu-walk and want to legitimize that in addition to the current + * nd->path and nd->root for ref-walk mode. Must be called from rcu-walk context. + * Nothing should touch nameidata between try_to_unlazy_next() failure and * terminate_walk(). */ -static int unlazy_child(struct nameidata *nd, struct dentry *dentry, unsigned seq) +static bool try_to_unlazy_next(struct nameidata *nd, struct dentry *dentry, unsigned seq) { BUG_ON(!(nd->flags & LOOKUP_RCU)); @@ -747,7 +802,7 @@ static int unlazy_child(struct nameidata *nd, struct dentry *dentry, unsigned se if (unlikely(!legitimize_root(nd))) goto out_dput; rcu_read_unlock(); - return 0; + return true; out2: nd->path.mnt = NULL; @@ -755,11 +810,11 @@ out1: nd->path.dentry = NULL; out: rcu_read_unlock(); - return -ECHILD; + return false; out_dput: rcu_read_unlock(); dput(dentry); - return -ECHILD; + return false; } static inline int d_revalidate(struct dentry *dentry, unsigned int flags) @@ -792,7 +847,8 @@ static int complete_walk(struct nameidata *nd) */ if (!(nd->flags & (LOOKUP_ROOT | LOOKUP_IS_SCOPED))) nd->root.mnt = NULL; - if (unlikely(unlazy_walk(nd))) + nd->flags &= ~LOOKUP_CACHED; + if (!try_to_unlazy(nd)) return -ECHILD; } @@ -954,11 +1010,16 @@ int sysctl_protected_regular __read_mostly; */ static inline int may_follow_link(struct nameidata *nd, const struct inode *inode) { + struct user_namespace *mnt_userns; + kuid_t i_uid; + if (!sysctl_protected_symlinks) return 0; + mnt_userns = mnt_user_ns(nd->path.mnt); + i_uid = i_uid_into_mnt(mnt_userns, inode); /* Allowed if owner and follower match. */ - if (uid_eq(current_cred()->fsuid, inode->i_uid)) + if (uid_eq(current_cred()->fsuid, i_uid)) return 0; /* Allowed if parent directory not sticky and world-writable. */ @@ -966,7 +1027,7 @@ static inline int may_follow_link(struct nameidata *nd, const struct inode *inod return 0; /* Allowed if parent directory and link owner match. */ - if (uid_valid(nd->dir_uid) && uid_eq(nd->dir_uid, inode->i_uid)) + if (uid_valid(nd->dir_uid) && uid_eq(nd->dir_uid, i_uid)) return 0; if (nd->flags & LOOKUP_RCU) @@ -979,6 +1040,7 @@ static inline int may_follow_link(struct nameidata *nd, const struct inode *inod /** * safe_hardlink_source - Check for safe hardlink conditions + * @mnt_userns: user namespace of the mount the inode was found from * @inode: the source inode to hardlink from * * Return false if at least one of the following conditions: @@ -989,7 +1051,8 @@ static inline int may_follow_link(struct nameidata *nd, const struct inode *inod * * Otherwise returns true. */ -static bool safe_hardlink_source(struct inode *inode) +static bool safe_hardlink_source(struct user_namespace *mnt_userns, + struct inode *inode) { umode_t mode = inode->i_mode; @@ -1006,7 +1069,7 @@ static bool safe_hardlink_source(struct inode *inode) return false; /* Hardlinking to unreadable or unwritable sources is dangerous. */ - if (inode_permission(inode, MAY_READ | MAY_WRITE)) + if (inode_permission(mnt_userns, inode, MAY_READ | MAY_WRITE)) return false; return true; @@ -1014,6 +1077,7 @@ static bool safe_hardlink_source(struct inode *inode) /** * may_linkat - Check permissions for creating a hardlink + * @mnt_userns: user namespace of the mount the inode was found from * @link: the source to hardlink from * * Block hardlink when all of: @@ -1022,14 +1086,21 @@ static bool safe_hardlink_source(struct inode *inode) * - hardlink source is unsafe (see safe_hardlink_source() above) * - not CAP_FOWNER in a namespace with the inode owner uid mapped * + * If the inode has been found through an idmapped mount the user namespace of + * the vfsmount must be passed through @mnt_userns. This function will then take + * care to map the inode according to @mnt_userns before checking permissions. + * On non-idmapped mounts or if permission checking is to be performed on the + * raw inode simply passs init_user_ns. + * * Returns 0 if successful, -ve on error. */ -int may_linkat(struct path *link) +int may_linkat(struct user_namespace *mnt_userns, struct path *link) { struct inode *inode = link->dentry->d_inode; /* Inode writeback is not safe when the uid or gid are invalid. */ - if (!uid_valid(inode->i_uid) || !gid_valid(inode->i_gid)) + if (!uid_valid(i_uid_into_mnt(mnt_userns, inode)) || + !gid_valid(i_gid_into_mnt(mnt_userns, inode))) return -EOVERFLOW; if (!sysctl_protected_hardlinks) @@ -1038,7 +1109,8 @@ int may_linkat(struct path *link) /* Source inode owner (or CAP_FOWNER) can hardlink all they like, * otherwise, it must be a safe source. */ - if (safe_hardlink_source(inode) || inode_owner_or_capable(inode)) + if (safe_hardlink_source(mnt_userns, inode) || + inode_owner_or_capable(mnt_userns, inode)) return 0; audit_log_path_denied(AUDIT_ANOM_LINK, "linkat"); @@ -1049,6 +1121,7 @@ int may_linkat(struct path *link) * may_create_in_sticky - Check whether an O_CREAT open in a sticky directory * should be allowed, or not, on files that already * exist. + * @mnt_userns: user namespace of the mount the inode was found from * @dir_mode: mode bits of directory * @dir_uid: owner of directory * @inode: the inode of the file to open @@ -1064,16 +1137,25 @@ int may_linkat(struct path *link) * the directory doesn't have to be world writable: being group writable will * be enough. * + * If the inode has been found through an idmapped mount the user namespace of + * the vfsmount must be passed through @mnt_userns. This function will then take + * care to map the inode according to @mnt_userns before checking permissions. + * On non-idmapped mounts or if permission checking is to be performed on the + * raw inode simply passs init_user_ns. + * * Returns 0 if the open is allowed, -ve on error. */ -static int may_create_in_sticky(umode_t dir_mode, kuid_t dir_uid, - struct inode * const inode) +static int may_create_in_sticky(struct user_namespace *mnt_userns, + struct nameidata *nd, struct inode *const inode) { + umode_t dir_mode = nd->dir_mode; + kuid_t dir_uid = nd->dir_uid; + if ((!sysctl_protected_fifos && S_ISFIFO(inode->i_mode)) || (!sysctl_protected_regular && S_ISREG(inode->i_mode)) || likely(!(dir_mode & S_ISVTX)) || - uid_eq(inode->i_uid, dir_uid) || - uid_eq(current_fsuid(), inode->i_uid)) + uid_eq(i_uid_into_mnt(mnt_userns, inode), dir_uid) || + uid_eq(current_fsuid(), i_uid_into_mnt(mnt_userns, inode))) return 0; if (likely(dir_mode & 0002) || @@ -1372,7 +1454,7 @@ static inline int handle_mounts(struct nameidata *nd, struct dentry *dentry, return -ENOENT; if (likely(__follow_mount_rcu(nd, path, inode, seqp))) return 0; - if (unlazy_child(nd, dentry, seq)) + if (!try_to_unlazy_next(nd, dentry, seq)) return -ECHILD; // *path might've been clobbered by __follow_mount_rcu() path->mnt = nd->path.mnt; @@ -1466,7 +1548,7 @@ static struct dentry *lookup_fast(struct nameidata *nd, unsigned seq; dentry = __d_lookup_rcu(parent, &nd->last, &seq); if (unlikely(!dentry)) { - if (unlazy_walk(nd)) + if (!try_to_unlazy(nd)) return ERR_PTR(-ECHILD); return NULL; } @@ -1493,9 +1575,9 @@ static struct dentry *lookup_fast(struct nameidata *nd, status = d_revalidate(dentry, nd->flags); if (likely(status > 0)) return dentry; - if (unlazy_child(nd, dentry, seq)) + if (!try_to_unlazy_next(nd, dentry, seq)) return ERR_PTR(-ECHILD); - if (unlikely(status == -ECHILD)) + if (status == -ECHILD) /* we'd been told to redo it in non-rcu mode */ status = d_revalidate(dentry, nd->flags); } else { @@ -1563,16 +1645,15 @@ static struct dentry *lookup_slow(const struct qstr *name, return res; } -static inline int may_lookup(struct nameidata *nd) +static inline int may_lookup(struct user_namespace *mnt_userns, + struct nameidata *nd) { if (nd->flags & LOOKUP_RCU) { - int err = inode_permission(nd->inode, MAY_EXEC|MAY_NOT_BLOCK); - if (err != -ECHILD) + int err = inode_permission(mnt_userns, nd->inode, MAY_EXEC|MAY_NOT_BLOCK); + if (err != -ECHILD || !try_to_unlazy(nd)) return err; - if (unlazy_walk(nd)) - return -ECHILD; } - return inode_permission(nd->inode, MAY_EXEC); + return inode_permission(mnt_userns, nd->inode, MAY_EXEC); } static int reserve_stack(struct nameidata *nd, struct path *link, unsigned seq) @@ -1592,7 +1673,7 @@ static int reserve_stack(struct nameidata *nd, struct path *link, unsigned seq) // unlazy even if we fail to grab the link - cleanup needs it bool grabbed_link = legitimize_path(nd, link, seq); - if (unlazy_walk(nd) != 0 || !grabbed_link) + if (!try_to_unlazy(nd) != 0 || !grabbed_link) return -ECHILD; if (nd_alloc_stack(nd)) @@ -1634,7 +1715,7 @@ static const char *pick_link(struct nameidata *nd, struct path *link, touch_atime(&last->link); cond_resched(); } else if (atime_needs_update(&last->link, inode)) { - if (unlikely(unlazy_walk(nd))) + if (!try_to_unlazy(nd)) return ERR_PTR(-ECHILD); touch_atime(&last->link); } @@ -1651,11 +1732,8 @@ static const char *pick_link(struct nameidata *nd, struct path *link, get = inode->i_op->get_link; if (nd->flags & LOOKUP_RCU) { res = get(NULL, inode, &last->done); - if (res == ERR_PTR(-ECHILD)) { - if (unlikely(unlazy_walk(nd))) - return ERR_PTR(-ECHILD); + if (res == ERR_PTR(-ECHILD) && try_to_unlazy(nd)) res = get(link->dentry, inode, &last->done); - } } else { res = get(link->dentry, inode, &last->done); } @@ -2121,11 +2199,13 @@ static int link_path_walk(const char *name, struct nameidata *nd) /* At this point we know we have a real path component. */ for(;;) { + struct user_namespace *mnt_userns; const char *link; u64 hash_len; int type; - err = may_lookup(nd); + mnt_userns = mnt_user_ns(nd->path.mnt); + err = may_lookup(mnt_userns, nd); if (err) return err; @@ -2173,7 +2253,7 @@ static int link_path_walk(const char *name, struct nameidata *nd) OK: /* pathname or trailing symlink, done */ if (!depth) { - nd->dir_uid = nd->inode->i_uid; + nd->dir_uid = i_uid_into_mnt(mnt_userns, nd->inode); nd->dir_mode = nd->inode->i_mode; nd->flags &= ~LOOKUP_PARENT; return 0; @@ -2195,7 +2275,7 @@ OK: } if (unlikely(!d_can_lookup(nd->path.dentry))) { if (nd->flags & LOOKUP_RCU) { - if (unlazy_walk(nd)) + if (!try_to_unlazy(nd)) return -ECHILD; } return -ENOTDIR; @@ -2209,6 +2289,10 @@ static const char *path_init(struct nameidata *nd, unsigned flags) int error; const char *s = nd->name->name; + /* LOOKUP_CACHED requires RCU, ask caller to retry */ + if ((flags & (LOOKUP_RCU | LOOKUP_CACHED)) == LOOKUP_CACHED) + return ERR_PTR(-EAGAIN); + if (!*s) flags &= ~LOOKUP_RCU; if (flags & LOOKUP_RCU) @@ -2506,7 +2590,7 @@ static int lookup_one_len_common(const char *name, struct dentry *base, return err; } - return inode_permission(base->d_inode, MAY_EXEC); + return inode_permission(&init_user_ns, base->d_inode, MAY_EXEC); } /** @@ -2651,15 +2735,16 @@ int user_path_at_empty(int dfd, const char __user *name, unsigned flags, } EXPORT_SYMBOL(user_path_at_empty); -int __check_sticky(struct inode *dir, struct inode *inode) +int __check_sticky(struct user_namespace *mnt_userns, struct inode *dir, + struct inode *inode) { kuid_t fsuid = current_fsuid(); - if (uid_eq(inode->i_uid, fsuid)) + if (uid_eq(i_uid_into_mnt(mnt_userns, inode), fsuid)) return 0; - if (uid_eq(dir->i_uid, fsuid)) + if (uid_eq(i_uid_into_mnt(mnt_userns, dir), fsuid)) return 0; - return !capable_wrt_inode_uidgid(inode, CAP_FOWNER); + return !capable_wrt_inode_uidgid(mnt_userns, inode, CAP_FOWNER); } EXPORT_SYMBOL(__check_sticky); @@ -2683,7 +2768,8 @@ EXPORT_SYMBOL(__check_sticky); * 11. We don't allow removal of NFS sillyrenamed files; it's handled by * nfs_async_unlink(). */ -static int may_delete(struct inode *dir, struct dentry *victim, bool isdir) +static int may_delete(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *victim, bool isdir) { struct inode *inode = d_backing_inode(victim); int error; @@ -2695,19 +2781,21 @@ static int may_delete(struct inode *dir, struct dentry *victim, bool isdir) BUG_ON(victim->d_parent->d_inode != dir); /* Inode writeback is not safe when the uid or gid are invalid. */ - if (!uid_valid(inode->i_uid) || !gid_valid(inode->i_gid)) + if (!uid_valid(i_uid_into_mnt(mnt_userns, inode)) || + !gid_valid(i_gid_into_mnt(mnt_userns, inode))) return -EOVERFLOW; audit_inode_child(dir, victim, AUDIT_TYPE_CHILD_DELETE); - error = inode_permission(dir, MAY_WRITE | MAY_EXEC); + error = inode_permission(mnt_userns, dir, MAY_WRITE | MAY_EXEC); if (error) return error; if (IS_APPEND(dir)) return -EPERM; - if (check_sticky(dir, inode) || IS_APPEND(inode) || - IS_IMMUTABLE(inode) || IS_SWAPFILE(inode) || HAS_UNMAPPED_ID(inode)) + if (check_sticky(mnt_userns, dir, inode) || IS_APPEND(inode) || + IS_IMMUTABLE(inode) || IS_SWAPFILE(inode) || + HAS_UNMAPPED_ID(mnt_userns, inode)) return -EPERM; if (isdir) { if (!d_is_dir(victim)) @@ -2732,7 +2820,8 @@ static int may_delete(struct inode *dir, struct dentry *victim, bool isdir) * 4. We should have write and exec permissions on dir * 5. We can't do it if dir is immutable (done in permission()) */ -static inline int may_create(struct inode *dir, struct dentry *child) +static inline int may_create(struct user_namespace *mnt_userns, + struct inode *dir, struct dentry *child) { struct user_namespace *s_user_ns; audit_inode_child(dir, child, AUDIT_TYPE_CHILD_CREATE); @@ -2741,10 +2830,10 @@ static inline int may_create(struct inode *dir, struct dentry *child) if (IS_DEADDIR(dir)) return -ENOENT; s_user_ns = dir->i_sb->s_user_ns; - if (!kuid_has_mapping(s_user_ns, current_fsuid()) || - !kgid_has_mapping(s_user_ns, current_fsgid())) + if (!kuid_has_mapping(s_user_ns, fsuid_into_mnt(mnt_userns)) || + !kgid_has_mapping(s_user_ns, fsgid_into_mnt(mnt_userns))) return -EOVERFLOW; - return inode_permission(dir, MAY_WRITE | MAY_EXEC); + return inode_permission(mnt_userns, dir, MAY_WRITE | MAY_EXEC); } /* @@ -2791,10 +2880,26 @@ void unlock_rename(struct dentry *p1, struct dentry *p2) } EXPORT_SYMBOL(unlock_rename); -int vfs_create(struct inode *dir, struct dentry *dentry, umode_t mode, - bool want_excl) +/** + * vfs_create - create new file + * @mnt_userns: user namespace of the mount the inode was found from + * @dir: inode of @dentry + * @dentry: pointer to dentry of the base directory + * @mode: mode of the new file + * @want_excl: whether the file must not yet exist + * + * Create a new file. + * + * If the inode has been found through an idmapped mount the user namespace of + * the vfsmount must be passed through @mnt_userns. This function will then take + * care to map the inode according to @mnt_userns before checking permissions. + * On non-idmapped mounts or if permission checking is to be performed on the + * raw inode simply passs init_user_ns. + */ +int vfs_create(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, bool want_excl) { - int error = may_create(dir, dentry); + int error = may_create(mnt_userns, dir, dentry); if (error) return error; @@ -2805,7 +2910,7 @@ int vfs_create(struct inode *dir, struct dentry *dentry, umode_t mode, error = security_inode_create(dir, dentry, mode); if (error) return error; - error = dir->i_op->create(dir, dentry, mode, want_excl); + error = dir->i_op->create(mnt_userns, dir, dentry, mode, want_excl); if (!error) fsnotify_create(dir, dentry); return error; @@ -2817,7 +2922,7 @@ int vfs_mkobj(struct dentry *dentry, umode_t mode, void *arg) { struct inode *dir = dentry->d_parent->d_inode; - int error = may_create(dir, dentry); + int error = may_create(&init_user_ns, dir, dentry); if (error) return error; @@ -2839,7 +2944,8 @@ bool may_open_dev(const struct path *path) !(path->mnt->mnt_sb->s_iflags & SB_I_NODEV); } -static int may_open(const struct path *path, int acc_mode, int flag) +static int may_open(struct user_namespace *mnt_userns, const struct path *path, + int acc_mode, int flag) { struct dentry *dentry = path->dentry; struct inode *inode = dentry->d_inode; @@ -2874,7 +2980,7 @@ static int may_open(const struct path *path, int acc_mode, int flag) break; } - error = inode_permission(inode, MAY_OPEN | acc_mode); + error = inode_permission(mnt_userns, inode, MAY_OPEN | acc_mode); if (error) return error; @@ -2889,13 +2995,13 @@ static int may_open(const struct path *path, int acc_mode, int flag) } /* O_NOATIME can only be set by the owner or superuser */ - if (flag & O_NOATIME && !inode_owner_or_capable(inode)) + if (flag & O_NOATIME && !inode_owner_or_capable(mnt_userns, inode)) return -EPERM; return 0; } -static int handle_truncate(struct file *filp) +static int handle_truncate(struct user_namespace *mnt_userns, struct file *filp) { const struct path *path = &filp->f_path; struct inode *inode = path->dentry->d_inode; @@ -2909,7 +3015,7 @@ static int handle_truncate(struct file *filp) if (!error) error = security_path_truncate(path); if (!error) { - error = do_truncate(path->dentry, 0, + error = do_truncate(mnt_userns, path->dentry, 0, ATTR_MTIME|ATTR_CTIME|ATTR_OPEN, filp); } @@ -2924,7 +3030,9 @@ static inline int open_to_namei_flags(int flag) return flag; } -static int may_o_create(const struct path *dir, struct dentry *dentry, umode_t mode) +static int may_o_create(struct user_namespace *mnt_userns, + const struct path *dir, struct dentry *dentry, + umode_t mode) { struct user_namespace *s_user_ns; int error = security_path_mknod(dir, dentry, mode, 0); @@ -2932,11 +3040,12 @@ static int may_o_create(const struct path *dir, struct dentry *dentry, umode_t m return error; s_user_ns = dir->dentry->d_sb->s_user_ns; - if (!kuid_has_mapping(s_user_ns, current_fsuid()) || - !kgid_has_mapping(s_user_ns, current_fsgid())) + if (!kuid_has_mapping(s_user_ns, fsuid_into_mnt(mnt_userns)) || + !kgid_has_mapping(s_user_ns, fsgid_into_mnt(mnt_userns))) return -EOVERFLOW; - error = inode_permission(dir->dentry->d_inode, MAY_WRITE | MAY_EXEC); + error = inode_permission(mnt_userns, dir->dentry->d_inode, + MAY_WRITE | MAY_EXEC); if (error) return error; @@ -3015,6 +3124,7 @@ static struct dentry *lookup_open(struct nameidata *nd, struct file *file, const struct open_flags *op, bool got_write) { + struct user_namespace *mnt_userns; struct dentry *dir = nd->path.dentry; struct inode *dir_inode = dir->d_inode; int open_flag = op->open_flag; @@ -3062,13 +3172,15 @@ static struct dentry *lookup_open(struct nameidata *nd, struct file *file, */ if (unlikely(!got_write)) open_flag &= ~O_TRUNC; + mnt_userns = mnt_user_ns(nd->path.mnt); if (open_flag & O_CREAT) { if (open_flag & O_EXCL) open_flag &= ~O_TRUNC; if (!IS_POSIXACL(dir->d_inode)) mode &= ~current_umask(); if (likely(got_write)) - create_error = may_o_create(&nd->path, dentry, mode); + create_error = may_o_create(mnt_userns, &nd->path, + dentry, mode); else create_error = -EROFS; } @@ -3103,8 +3215,9 @@ static struct dentry *lookup_open(struct nameidata *nd, struct file *file, error = -EACCES; goto out_dput; } - error = dir_inode->i_op->create(dir_inode, dentry, mode, - open_flag & O_EXCL); + + error = dir_inode->i_op->create(mnt_userns, dir_inode, dentry, + mode, open_flag & O_EXCL); if (error) goto out_dput; } @@ -3129,7 +3242,6 @@ static const char *open_last_lookups(struct nameidata *nd, struct inode *inode; struct dentry *dentry; const char *res; - int error; nd->flags |= op->intent; @@ -3153,9 +3265,8 @@ static const char *open_last_lookups(struct nameidata *nd, } else { /* create side of things */ if (nd->flags & LOOKUP_RCU) { - error = unlazy_walk(nd); - if (unlikely(error)) - return ERR_PTR(error); + if (!try_to_unlazy(nd)) + return ERR_PTR(-ECHILD); } audit_inode(nd->name, dir, AUDIT_INODE_PARENT); /* trailing slashes? */ @@ -3164,9 +3275,7 @@ static const char *open_last_lookups(struct nameidata *nd, } if (open_flag & (O_CREAT | O_TRUNC | O_WRONLY | O_RDWR)) { - error = mnt_want_write(nd->path.mnt); - if (!error) - got_write = true; + got_write = !mnt_want_write(nd->path.mnt); /* * do _not_ fail yet - we might not need that or fail with * a different error; let lookup_open() decide; we'll be @@ -3212,6 +3321,7 @@ finish_lookup: static int do_open(struct nameidata *nd, struct file *file, const struct open_flags *op) { + struct user_namespace *mnt_userns; int open_flag = op->open_flag; bool do_truncate; int acc_mode; @@ -3224,12 +3334,13 @@ static int do_open(struct nameidata *nd, } if (!(file->f_mode & FMODE_CREATED)) audit_inode(nd->name, nd->path.dentry, 0); + mnt_userns = mnt_user_ns(nd->path.mnt); if (open_flag & O_CREAT) { if ((open_flag & O_EXCL) && !(file->f_mode & FMODE_CREATED)) return -EEXIST; if (d_is_dir(nd->path.dentry)) return -EISDIR; - error = may_create_in_sticky(nd->dir_mode, nd->dir_uid, + error = may_create_in_sticky(mnt_userns, nd, d_backing_inode(nd->path.dentry)); if (unlikely(error)) return error; @@ -3249,13 +3360,13 @@ static int do_open(struct nameidata *nd, return error; do_truncate = true; } - error = may_open(&nd->path, acc_mode, open_flag); + error = may_open(mnt_userns, &nd->path, acc_mode, open_flag); if (!error && !(file->f_mode & FMODE_OPENED)) error = vfs_open(&nd->path, file); if (!error) error = ima_file_check(file, op->acc_mode); if (!error && do_truncate) - error = handle_truncate(file); + error = handle_truncate(mnt_userns, file); if (unlikely(error > 0)) { WARN_ON(1); error = -EINVAL; @@ -3265,7 +3376,23 @@ static int do_open(struct nameidata *nd, return error; } -struct dentry *vfs_tmpfile(struct dentry *dentry, umode_t mode, int open_flag) +/** + * vfs_tmpfile - create tmpfile + * @mnt_userns: user namespace of the mount the inode was found from + * @dentry: pointer to dentry of the base directory + * @mode: mode of the new tmpfile + * @open_flags: flags + * + * Create a temporary file. + * + * If the inode has been found through an idmapped mount the user namespace of + * the vfsmount must be passed through @mnt_userns. This function will then take + * care to map the inode according to @mnt_userns before checking permissions. + * On non-idmapped mounts or if permission checking is to be performed on the + * raw inode simply passs init_user_ns. + */ +struct dentry *vfs_tmpfile(struct user_namespace *mnt_userns, + struct dentry *dentry, umode_t mode, int open_flag) { struct dentry *child = NULL; struct inode *dir = dentry->d_inode; @@ -3273,7 +3400,7 @@ struct dentry *vfs_tmpfile(struct dentry *dentry, umode_t mode, int open_flag) int error; /* we want directory to be writable */ - error = inode_permission(dir, MAY_WRITE | MAY_EXEC); + error = inode_permission(mnt_userns, dir, MAY_WRITE | MAY_EXEC); if (error) goto out_err; error = -EOPNOTSUPP; @@ -3283,7 +3410,7 @@ struct dentry *vfs_tmpfile(struct dentry *dentry, umode_t mode, int open_flag) child = d_alloc(dentry, &slash_name); if (unlikely(!child)) goto out_err; - error = dir->i_op->tmpfile(dir, child, mode); + error = dir->i_op->tmpfile(mnt_userns, dir, child, mode); if (error) goto out_err; error = -ENOENT; @@ -3295,7 +3422,7 @@ struct dentry *vfs_tmpfile(struct dentry *dentry, umode_t mode, int open_flag) inode->i_state |= I_LINKABLE; spin_unlock(&inode->i_lock); } - ima_post_create_tmpfile(inode); + ima_post_create_tmpfile(mnt_userns, inode); return child; out_err: @@ -3308,6 +3435,7 @@ static int do_tmpfile(struct nameidata *nd, unsigned flags, const struct open_flags *op, struct file *file) { + struct user_namespace *mnt_userns; struct dentry *child; struct path path; int error = path_lookupat(nd, flags | LOOKUP_DIRECTORY, &path); @@ -3316,7 +3444,8 @@ static int do_tmpfile(struct nameidata *nd, unsigned flags, error = mnt_want_write(path.mnt); if (unlikely(error)) goto out; - child = vfs_tmpfile(path.dentry, op->mode, op->open_flag); + mnt_userns = mnt_user_ns(path.mnt); + child = vfs_tmpfile(mnt_userns, path.dentry, op->mode, op->open_flag); error = PTR_ERR(child); if (IS_ERR(child)) goto out2; @@ -3324,11 +3453,9 @@ static int do_tmpfile(struct nameidata *nd, unsigned flags, path.dentry = child; audit_inode(nd->name, child, 0); /* Don't check for other permissions, the inode was just created */ - error = may_open(&path, 0, op->open_flag); - if (error) - goto out2; - file->f_path.mnt = path.mnt; - error = finish_open(file, child, NULL); + error = may_open(mnt_userns, &path, 0, op->open_flag); + if (!error) + error = vfs_open(&path, file); out2: mnt_drop_write(path.mnt); out: @@ -3528,10 +3655,27 @@ inline struct dentry *user_path_create(int dfd, const char __user *pathname, } EXPORT_SYMBOL(user_path_create); -int vfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t dev) +/** + * vfs_mknod - create device node or file + * @mnt_userns: user namespace of the mount the inode was found from + * @dir: inode of @dentry + * @dentry: pointer to dentry of the base directory + * @mode: mode of the new device node or file + * @dev: device number of device to create + * + * Create a device node or file. + * + * If the inode has been found through an idmapped mount the user namespace of + * the vfsmount must be passed through @mnt_userns. This function will then take + * care to map the inode according to @mnt_userns before checking permissions. + * On non-idmapped mounts or if permission checking is to be performed on the + * raw inode simply passs init_user_ns. + */ +int vfs_mknod(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, dev_t dev) { bool is_whiteout = S_ISCHR(mode) && dev == WHITEOUT_DEV; - int error = may_create(dir, dentry); + int error = may_create(mnt_userns, dir, dentry); if (error) return error; @@ -3551,7 +3695,7 @@ int vfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t dev) if (error) return error; - error = dir->i_op->mknod(dir, dentry, mode, dev); + error = dir->i_op->mknod(mnt_userns, dir, dentry, mode, dev); if (!error) fsnotify_create(dir, dentry); return error; @@ -3578,6 +3722,7 @@ static int may_mknod(umode_t mode) static long do_mknodat(int dfd, const char __user *filename, umode_t mode, unsigned int dev) { + struct user_namespace *mnt_userns; struct dentry *dentry; struct path path; int error; @@ -3596,18 +3741,22 @@ retry: error = security_path_mknod(&path, dentry, mode, dev); if (error) goto out; + + mnt_userns = mnt_user_ns(path.mnt); switch (mode & S_IFMT) { case 0: case S_IFREG: - error = vfs_create(path.dentry->d_inode,dentry,mode,true); + error = vfs_create(mnt_userns, path.dentry->d_inode, + dentry, mode, true); if (!error) - ima_post_path_mknod(dentry); + ima_post_path_mknod(mnt_userns, dentry); break; case S_IFCHR: case S_IFBLK: - error = vfs_mknod(path.dentry->d_inode,dentry,mode, - new_decode_dev(dev)); + error = vfs_mknod(mnt_userns, path.dentry->d_inode, + dentry, mode, new_decode_dev(dev)); break; case S_IFIFO: case S_IFSOCK: - error = vfs_mknod(path.dentry->d_inode,dentry,mode,0); + error = vfs_mknod(mnt_userns, path.dentry->d_inode, + dentry, mode, 0); break; } out: @@ -3630,9 +3779,25 @@ SYSCALL_DEFINE3(mknod, const char __user *, filename, umode_t, mode, unsigned, d return do_mknodat(AT_FDCWD, filename, mode, dev); } -int vfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) +/** + * vfs_mkdir - create directory + * @mnt_userns: user namespace of the mount the inode was found from + * @dir: inode of @dentry + * @dentry: pointer to dentry of the base directory + * @mode: mode of the new directory + * + * Create a directory. + * + * If the inode has been found through an idmapped mount the user namespace of + * the vfsmount must be passed through @mnt_userns. This function will then take + * care to map the inode according to @mnt_userns before checking permissions. + * On non-idmapped mounts or if permission checking is to be performed on the + * raw inode simply passs init_user_ns. + */ +int vfs_mkdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { - int error = may_create(dir, dentry); + int error = may_create(mnt_userns, dir, dentry); unsigned max_links = dir->i_sb->s_max_links; if (error) @@ -3649,7 +3814,7 @@ int vfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) if (max_links && dir->i_nlink >= max_links) return -EMLINK; - error = dir->i_op->mkdir(dir, dentry, mode); + error = dir->i_op->mkdir(mnt_userns, dir, dentry, mode); if (!error) fsnotify_mkdir(dir, dentry); return error; @@ -3671,8 +3836,12 @@ retry: if (!IS_POSIXACL(path.dentry->d_inode)) mode &= ~current_umask(); error = security_path_mkdir(&path, dentry, mode); - if (!error) - error = vfs_mkdir(path.dentry->d_inode, dentry, mode); + if (!error) { + struct user_namespace *mnt_userns; + mnt_userns = mnt_user_ns(path.mnt); + error = vfs_mkdir(mnt_userns, path.dentry->d_inode, dentry, + mode); + } done_path_create(&path, dentry); if (retry_estale(error, lookup_flags)) { lookup_flags |= LOOKUP_REVAL; @@ -3691,9 +3860,24 @@ SYSCALL_DEFINE2(mkdir, const char __user *, pathname, umode_t, mode) return do_mkdirat(AT_FDCWD, pathname, mode); } -int vfs_rmdir(struct inode *dir, struct dentry *dentry) +/** + * vfs_rmdir - remove directory + * @mnt_userns: user namespace of the mount the inode was found from + * @dir: inode of @dentry + * @dentry: pointer to dentry of the base directory + * + * Remove a directory. + * + * If the inode has been found through an idmapped mount the user namespace of + * the vfsmount must be passed through @mnt_userns. This function will then take + * care to map the inode according to @mnt_userns before checking permissions. + * On non-idmapped mounts or if permission checking is to be performed on the + * raw inode simply passs init_user_ns. + */ +int vfs_rmdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry) { - int error = may_delete(dir, dentry, 1); + int error = may_delete(mnt_userns, dir, dentry, 1); if (error) return error; @@ -3733,6 +3917,7 @@ EXPORT_SYMBOL(vfs_rmdir); long do_rmdir(int dfd, struct filename *name) { + struct user_namespace *mnt_userns; int error = 0; struct dentry *dentry; struct path path; @@ -3773,7 +3958,8 @@ retry: error = security_path_rmdir(&path, dentry); if (error) goto exit3; - error = vfs_rmdir(path.dentry->d_inode, dentry); + mnt_userns = mnt_user_ns(path.mnt); + error = vfs_rmdir(mnt_userns, path.dentry->d_inode, dentry); exit3: dput(dentry); exit2: @@ -3796,6 +3982,7 @@ SYSCALL_DEFINE1(rmdir, const char __user *, pathname) /** * vfs_unlink - unlink a filesystem object + * @mnt_userns: user namespace of the mount the inode was found from * @dir: parent directory * @dentry: victim * @delegated_inode: returns victim inode, if the inode is delegated. @@ -3811,11 +3998,18 @@ SYSCALL_DEFINE1(rmdir, const char __user *, pathname) * Alternatively, a caller may pass NULL for delegated_inode. This may * be appropriate for callers that expect the underlying filesystem not * to be NFS exported. + * + * If the inode has been found through an idmapped mount the user namespace of + * the vfsmount must be passed through @mnt_userns. This function will then take + * care to map the inode according to @mnt_userns before checking permissions. + * On non-idmapped mounts or if permission checking is to be performed on the + * raw inode simply passs init_user_ns. */ -int vfs_unlink(struct inode *dir, struct dentry *dentry, struct inode **delegated_inode) +int vfs_unlink(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, struct inode **delegated_inode) { struct inode *target = dentry->d_inode; - int error = may_delete(dir, dentry, 0); + int error = may_delete(mnt_userns, dir, dentry, 0); if (error) return error; @@ -3886,6 +4080,8 @@ retry_deleg: dentry = __lookup_hash(&last, path.dentry, lookup_flags); error = PTR_ERR(dentry); if (!IS_ERR(dentry)) { + struct user_namespace *mnt_userns; + /* Why not before? Because we want correct error value */ if (last.name[last.len]) goto slashes; @@ -3896,7 +4092,9 @@ retry_deleg: error = security_path_unlink(&path, dentry); if (error) goto exit2; - error = vfs_unlink(path.dentry->d_inode, dentry, &delegated_inode); + mnt_userns = mnt_user_ns(path.mnt); + error = vfs_unlink(mnt_userns, path.dentry->d_inode, dentry, + &delegated_inode); exit2: dput(dentry); } @@ -3945,9 +4143,25 @@ SYSCALL_DEFINE1(unlink, const char __user *, pathname) return do_unlinkat(AT_FDCWD, getname(pathname)); } -int vfs_symlink(struct inode *dir, struct dentry *dentry, const char *oldname) +/** + * vfs_symlink - create symlink + * @mnt_userns: user namespace of the mount the inode was found from + * @dir: inode of @dentry + * @dentry: pointer to dentry of the base directory + * @oldname: name of the file to link to + * + * Create a symlink. + * + * If the inode has been found through an idmapped mount the user namespace of + * the vfsmount must be passed through @mnt_userns. This function will then take + * care to map the inode according to @mnt_userns before checking permissions. + * On non-idmapped mounts or if permission checking is to be performed on the + * raw inode simply passs init_user_ns. + */ +int vfs_symlink(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, const char *oldname) { - int error = may_create(dir, dentry); + int error = may_create(mnt_userns, dir, dentry); if (error) return error; @@ -3959,7 +4173,7 @@ int vfs_symlink(struct inode *dir, struct dentry *dentry, const char *oldname) if (error) return error; - error = dir->i_op->symlink(dir, dentry, oldname); + error = dir->i_op->symlink(mnt_userns, dir, dentry, oldname); if (!error) fsnotify_create(dir, dentry); return error; @@ -3985,8 +4199,13 @@ retry: goto out_putname; error = security_path_symlink(&path, dentry, from->name); - if (!error) - error = vfs_symlink(path.dentry->d_inode, dentry, from->name); + if (!error) { + struct user_namespace *mnt_userns; + + mnt_userns = mnt_user_ns(path.mnt); + error = vfs_symlink(mnt_userns, path.dentry->d_inode, dentry, + from->name); + } done_path_create(&path, dentry); if (retry_estale(error, lookup_flags)) { lookup_flags |= LOOKUP_REVAL; @@ -4011,6 +4230,7 @@ SYSCALL_DEFINE2(symlink, const char __user *, oldname, const char __user *, newn /** * vfs_link - create a new link * @old_dentry: object to be linked + * @mnt_userns: the user namespace of the mount * @dir: new parent * @new_dentry: where to create the new link * @delegated_inode: returns inode needing a delegation break @@ -4026,8 +4246,16 @@ SYSCALL_DEFINE2(symlink, const char __user *, oldname, const char __user *, newn * Alternatively, a caller may pass NULL for delegated_inode. This may * be appropriate for callers that expect the underlying filesystem not * to be NFS exported. + * + * If the inode has been found through an idmapped mount the user namespace of + * the vfsmount must be passed through @mnt_userns. This function will then take + * care to map the inode according to @mnt_userns before checking permissions. + * On non-idmapped mounts or if permission checking is to be performed on the + * raw inode simply passs init_user_ns. */ -int vfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *new_dentry, struct inode **delegated_inode) +int vfs_link(struct dentry *old_dentry, struct user_namespace *mnt_userns, + struct inode *dir, struct dentry *new_dentry, + struct inode **delegated_inode) { struct inode *inode = old_dentry->d_inode; unsigned max_links = dir->i_sb->s_max_links; @@ -4036,7 +4264,7 @@ int vfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *new_de if (!inode) return -ENOENT; - error = may_create(dir, new_dentry); + error = may_create(mnt_userns, dir, new_dentry); if (error) return error; @@ -4053,7 +4281,7 @@ int vfs_link(struct dentry *old_dentry, struct inode *dir, struct dentry *new_de * be writen back improperly if their true value is unknown to * the vfs. */ - if (HAS_UNMAPPED_ID(inode)) + if (HAS_UNMAPPED_ID(mnt_userns, inode)) return -EPERM; if (!dir->i_op->link) return -EPERM; @@ -4100,6 +4328,7 @@ EXPORT_SYMBOL(vfs_link); static int do_linkat(int olddfd, const char __user *oldname, int newdfd, const char __user *newname, int flags) { + struct user_namespace *mnt_userns; struct dentry *new_dentry; struct path old_path, new_path; struct inode *delegated_inode = NULL; @@ -4135,13 +4364,15 @@ retry: error = -EXDEV; if (old_path.mnt != new_path.mnt) goto out_dput; - error = may_linkat(&old_path); + mnt_userns = mnt_user_ns(new_path.mnt); + error = may_linkat(mnt_userns, &old_path); if (unlikely(error)) goto out_dput; error = security_path_link(old_path.dentry, &new_path, new_dentry); if (error) goto out_dput; - error = vfs_link(old_path.dentry, new_path.dentry->d_inode, new_dentry, &delegated_inode); + error = vfs_link(old_path.dentry, mnt_userns, new_path.dentry->d_inode, + new_dentry, &delegated_inode); out_dput: done_path_create(&new_path, new_dentry); if (delegated_inode) { @@ -4175,12 +4406,14 @@ SYSCALL_DEFINE2(link, const char __user *, oldname, const char __user *, newname /** * vfs_rename - rename a filesystem object - * @old_dir: parent of source - * @old_dentry: source - * @new_dir: parent of destination - * @new_dentry: destination - * @delegated_inode: returns an inode needing a delegation break - * @flags: rename flags + * @old_mnt_userns: old user namespace of the mount the inode was found from + * @old_dir: parent of source + * @old_dentry: source + * @new_mnt_userns: new user namespace of the mount the inode was found from + * @new_dir: parent of destination + * @new_dentry: destination + * @delegated_inode: returns an inode needing a delegation break + * @flags: rename flags * * The caller must hold multiple mutexes--see lock_rename()). * @@ -4223,11 +4456,14 @@ SYSCALL_DEFINE2(link, const char __user *, oldname, const char __user *, newname * ->i_mutex on parents, which works but leads to some truly excessive * locking]. */ -int vfs_rename(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry, - struct inode **delegated_inode, unsigned int flags) +int vfs_rename(struct renamedata *rd) { int error; + struct inode *old_dir = rd->old_dir, *new_dir = rd->new_dir; + struct dentry *old_dentry = rd->old_dentry; + struct dentry *new_dentry = rd->new_dentry; + struct inode **delegated_inode = rd->delegated_inode; + unsigned int flags = rd->flags; bool is_dir = d_is_dir(old_dentry); struct inode *source = old_dentry->d_inode; struct inode *target = new_dentry->d_inode; @@ -4238,19 +4474,21 @@ int vfs_rename(struct inode *old_dir, struct dentry *old_dentry, if (source == target) return 0; - error = may_delete(old_dir, old_dentry, is_dir); + error = may_delete(rd->old_mnt_userns, old_dir, old_dentry, is_dir); if (error) return error; if (!target) { - error = may_create(new_dir, new_dentry); + error = may_create(rd->new_mnt_userns, new_dir, new_dentry); } else { new_is_dir = d_is_dir(new_dentry); if (!(flags & RENAME_EXCHANGE)) - error = may_delete(new_dir, new_dentry, is_dir); + error = may_delete(rd->new_mnt_userns, new_dir, + new_dentry, is_dir); else - error = may_delete(new_dir, new_dentry, new_is_dir); + error = may_delete(rd->new_mnt_userns, new_dir, + new_dentry, new_is_dir); } if (error) return error; @@ -4264,12 +4502,14 @@ int vfs_rename(struct inode *old_dir, struct dentry *old_dentry, */ if (new_dir != old_dir) { if (is_dir) { - error = inode_permission(source, MAY_WRITE); + error = inode_permission(rd->old_mnt_userns, source, + MAY_WRITE); if (error) return error; } if ((flags & RENAME_EXCHANGE) && new_is_dir) { - error = inode_permission(target, MAY_WRITE); + error = inode_permission(rd->new_mnt_userns, target, + MAY_WRITE); if (error) return error; } @@ -4309,8 +4549,8 @@ int vfs_rename(struct inode *old_dir, struct dentry *old_dentry, if (error) goto out; } - error = old_dir->i_op->rename(old_dir, old_dentry, - new_dir, new_dentry, flags); + error = old_dir->i_op->rename(rd->new_mnt_userns, old_dir, old_dentry, + new_dir, new_dentry, flags); if (error) goto out; @@ -4351,6 +4591,7 @@ EXPORT_SYMBOL(vfs_rename); int do_renameat2(int olddfd, struct filename *from, int newdfd, struct filename *to, unsigned int flags) { + struct renamedata rd; struct dentry *old_dentry, *new_dentry; struct dentry *trap; struct path old_path, new_path; @@ -4454,9 +4695,16 @@ retry_deleg: &new_path, new_dentry, flags); if (error) goto exit5; - error = vfs_rename(old_path.dentry->d_inode, old_dentry, - new_path.dentry->d_inode, new_dentry, - &delegated_inode, flags); + + rd.old_dir = old_path.dentry->d_inode; + rd.old_dentry = old_dentry; + rd.old_mnt_userns = mnt_user_ns(old_path.mnt); + rd.new_dir = new_path.dentry->d_inode; + rd.new_dentry = new_dentry; + rd.new_mnt_userns = mnt_user_ns(new_path.mnt); + rd.delegated_inode = &delegated_inode; + rd.flags = flags; + error = vfs_rename(&rd); exit5: dput(new_dentry); exit4: diff --git a/fs/namespace.c b/fs/namespace.c index 9d33909d0f9e..56bb5a5fdc0d 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -25,6 +25,7 @@ #include <linux/proc_ns.h> #include <linux/magic.h> #include <linux/memblock.h> +#include <linux/proc_fs.h> #include <linux/task_work.h> #include <linux/sched/task.h> #include <uapi/linux/mount.h> @@ -73,6 +74,15 @@ static DECLARE_RWSEM(namespace_sem); static HLIST_HEAD(unmounted); /* protected by namespace_sem */ static LIST_HEAD(ex_mountpoints); /* protected by namespace_sem */ +struct mount_kattr { + unsigned int attr_set; + unsigned int attr_clr; + unsigned int propagation; + unsigned int lookup_flags; + bool recurse; + struct user_namespace *mnt_userns; +}; + /* /sys/fs */ struct kobject *fs_kobj; EXPORT_SYMBOL_GPL(fs_kobj); @@ -87,6 +97,16 @@ EXPORT_SYMBOL_GPL(fs_kobj); */ __cacheline_aligned_in_smp DEFINE_SEQLOCK(mount_lock); +static inline void lock_mount_hash(void) +{ + write_seqlock(&mount_lock); +} + +static inline void unlock_mount_hash(void) +{ + write_sequnlock(&mount_lock); +} + static inline struct hlist_head *m_hash(struct vfsmount *mnt, struct dentry *dentry) { unsigned long tmp = ((unsigned long)mnt / L1_CACHE_BYTES); @@ -210,6 +230,7 @@ static struct mount *alloc_vfsmnt(const char *name) INIT_HLIST_NODE(&mnt->mnt_mp_list); INIT_LIST_HEAD(&mnt->mnt_umounting); INIT_HLIST_HEAD(&mnt->mnt_stuck_children); + mnt->mnt.mnt_userns = &init_user_ns; } return mnt; @@ -360,50 +381,36 @@ int mnt_want_write(struct vfsmount *m) EXPORT_SYMBOL_GPL(mnt_want_write); /** - * mnt_clone_write - get write access to a mount - * @mnt: the mount on which to take a write - * - * This is effectively like mnt_want_write, except - * it must only be used to take an extra write reference - * on a mountpoint that we already know has a write reference - * on it. This allows some optimisation. - * - * After finished, mnt_drop_write must be called as usual to - * drop the reference. - */ -int mnt_clone_write(struct vfsmount *mnt) -{ - /* superblock may be r/o */ - if (__mnt_is_readonly(mnt)) - return -EROFS; - preempt_disable(); - mnt_inc_writers(real_mount(mnt)); - preempt_enable(); - return 0; -} -EXPORT_SYMBOL_GPL(mnt_clone_write); - -/** * __mnt_want_write_file - get write access to a file's mount * @file: the file who's mount on which to take a write * - * This is like __mnt_want_write, but it takes a file and can - * do some optimisations if the file is open for write already + * This is like __mnt_want_write, but if the file is already open for writing it + * skips incrementing mnt_writers (since the open file already has a reference) + * and instead only does the check for emergency r/o remounts. This must be + * paired with __mnt_drop_write_file. */ int __mnt_want_write_file(struct file *file) { - if (!(file->f_mode & FMODE_WRITER)) - return __mnt_want_write(file->f_path.mnt); - else - return mnt_clone_write(file->f_path.mnt); + if (file->f_mode & FMODE_WRITER) { + /* + * Superblock may have become readonly while there are still + * writable fd's, e.g. due to a fs error with errors=remount-ro + */ + if (__mnt_is_readonly(file->f_path.mnt)) + return -EROFS; + return 0; + } + return __mnt_want_write(file->f_path.mnt); } /** * mnt_want_write_file - get write access to a file's mount * @file: the file who's mount on which to take a write * - * This is like mnt_want_write, but it takes a file and can - * do some optimisations if the file is open for write already + * This is like mnt_want_write, but if the file is already open for writing it + * skips incrementing mnt_writers (since the open file already has a reference) + * and instead only does the freeze protection and the check for emergency r/o + * remounts. This must be paired with mnt_drop_write_file. */ int mnt_want_write_file(struct file *file) { @@ -449,7 +456,8 @@ EXPORT_SYMBOL_GPL(mnt_drop_write); void __mnt_drop_write_file(struct file *file) { - __mnt_drop_write(file->f_path.mnt); + if (!(file->f_mode & FMODE_WRITER)) + __mnt_drop_write(file->f_path.mnt); } void mnt_drop_write_file(struct file *file) @@ -459,11 +467,8 @@ void mnt_drop_write_file(struct file *file) } EXPORT_SYMBOL(mnt_drop_write_file); -static int mnt_make_readonly(struct mount *mnt) +static inline int mnt_hold_writers(struct mount *mnt) { - int ret = 0; - - lock_mount_hash(); mnt->mnt.mnt_flags |= MNT_WRITE_HOLD; /* * After storing MNT_WRITE_HOLD, we'll read the counters. This store @@ -488,25 +493,30 @@ static int mnt_make_readonly(struct mount *mnt) * we're counting up here. */ if (mnt_get_writers(mnt) > 0) - ret = -EBUSY; - else - mnt->mnt.mnt_flags |= MNT_READONLY; + return -EBUSY; + + return 0; +} + +static inline void mnt_unhold_writers(struct mount *mnt) +{ /* * MNT_READONLY must become visible before ~MNT_WRITE_HOLD, so writers * that become unheld will see MNT_READONLY. */ smp_wmb(); mnt->mnt.mnt_flags &= ~MNT_WRITE_HOLD; - unlock_mount_hash(); - return ret; } -static int __mnt_unmake_readonly(struct mount *mnt) +static int mnt_make_readonly(struct mount *mnt) { - lock_mount_hash(); - mnt->mnt.mnt_flags &= ~MNT_READONLY; - unlock_mount_hash(); - return 0; + int ret; + + ret = mnt_hold_writers(mnt); + if (!ret) + mnt->mnt.mnt_flags |= MNT_READONLY; + mnt_unhold_writers(mnt); + return ret; } int sb_prepare_remount_readonly(struct super_block *sb) @@ -547,6 +557,11 @@ int sb_prepare_remount_readonly(struct super_block *sb) static void free_vfsmnt(struct mount *mnt) { + struct user_namespace *mnt_userns; + + mnt_userns = mnt_user_ns(&mnt->mnt); + if (mnt_userns != &init_user_ns) + put_user_ns(mnt_userns); kfree_const(mnt->mnt_devname); #ifdef CONFIG_SMP free_percpu(mnt->mnt_pcp); @@ -1055,6 +1070,9 @@ static struct mount *clone_mnt(struct mount *old, struct dentry *root, mnt->mnt.mnt_flags &= ~(MNT_WRITE_HOLD|MNT_MARKED|MNT_INTERNAL); atomic_inc(&sb->s_active); + mnt->mnt.mnt_userns = mnt_user_ns(&old->mnt); + if (mnt->mnt.mnt_userns != &init_user_ns) + mnt->mnt.mnt_userns = get_user_ns(mnt->mnt.mnt_userns); mnt->mnt.mnt_sb = sb; mnt->mnt.mnt_root = dget(root); mnt->mnt_mountpoint = mnt->mnt.mnt_root; @@ -2514,20 +2532,15 @@ static int change_mount_ro_state(struct mount *mnt, unsigned int mnt_flags) if (readonly_request) return mnt_make_readonly(mnt); - return __mnt_unmake_readonly(mnt); + mnt->mnt.mnt_flags &= ~MNT_READONLY; + return 0; } -/* - * Update the user-settable attributes on a mount. The caller must hold - * sb->s_umount for writing. - */ static void set_mount_attributes(struct mount *mnt, unsigned int mnt_flags) { - lock_mount_hash(); mnt_flags |= mnt->mnt.mnt_flags & ~MNT_USER_SETTABLE_MASK; mnt->mnt.mnt_flags = mnt_flags; touch_mnt_namespace(mnt->mnt_ns); - unlock_mount_hash(); } static void mnt_warn_timestamp_expiry(struct path *mountpoint, struct vfsmount *mnt) @@ -2572,11 +2585,17 @@ static int do_reconfigure_mnt(struct path *path, unsigned int mnt_flags) if (!can_change_locked_flags(mnt, mnt_flags)) return -EPERM; - down_write(&sb->s_umount); + /* + * We're only checking whether the superblock is read-only not + * changing it, so only take down_read(&sb->s_umount). + */ + down_read(&sb->s_umount); + lock_mount_hash(); ret = change_mount_ro_state(mnt, mnt_flags); if (ret == 0) set_mount_attributes(mnt, mnt_flags); - up_write(&sb->s_umount); + unlock_mount_hash(); + up_read(&sb->s_umount); mnt_warn_timestamp_expiry(path, &mnt->mnt); @@ -2616,8 +2635,11 @@ static int do_remount(struct path *path, int ms_flags, int sb_flags, err = -EPERM; if (ns_capable(sb->s_user_ns, CAP_SYS_ADMIN)) { err = reconfigure_super(fc); - if (!err) + if (!err) { + lock_mount_hash(); set_mount_attributes(mnt, mnt_flags); + unlock_mount_hash(); + } } up_write(&sb->s_umount); } @@ -3440,6 +3462,33 @@ out_type: return ret; } +#define FSMOUNT_VALID_FLAGS \ + (MOUNT_ATTR_RDONLY | MOUNT_ATTR_NOSUID | MOUNT_ATTR_NODEV | \ + MOUNT_ATTR_NOEXEC | MOUNT_ATTR__ATIME | MOUNT_ATTR_NODIRATIME) + +#define MOUNT_SETATTR_VALID_FLAGS (FSMOUNT_VALID_FLAGS | MOUNT_ATTR_IDMAP) + +#define MOUNT_SETATTR_PROPAGATION_FLAGS \ + (MS_UNBINDABLE | MS_PRIVATE | MS_SLAVE | MS_SHARED) + +static unsigned int attr_flags_to_mnt_flags(u64 attr_flags) +{ + unsigned int mnt_flags = 0; + + if (attr_flags & MOUNT_ATTR_RDONLY) + mnt_flags |= MNT_READONLY; + if (attr_flags & MOUNT_ATTR_NOSUID) + mnt_flags |= MNT_NOSUID; + if (attr_flags & MOUNT_ATTR_NODEV) + mnt_flags |= MNT_NODEV; + if (attr_flags & MOUNT_ATTR_NOEXEC) + mnt_flags |= MNT_NOEXEC; + if (attr_flags & MOUNT_ATTR_NODIRATIME) + mnt_flags |= MNT_NODIRATIME; + + return mnt_flags; +} + /* * Create a kernel mount representation for a new, prepared superblock * (specified by fs_fd) and attach to an open_tree-like file descriptor. @@ -3462,24 +3511,10 @@ SYSCALL_DEFINE3(fsmount, int, fs_fd, unsigned int, flags, if ((flags & ~(FSMOUNT_CLOEXEC)) != 0) return -EINVAL; - if (attr_flags & ~(MOUNT_ATTR_RDONLY | - MOUNT_ATTR_NOSUID | - MOUNT_ATTR_NODEV | - MOUNT_ATTR_NOEXEC | - MOUNT_ATTR__ATIME | - MOUNT_ATTR_NODIRATIME)) + if (attr_flags & ~FSMOUNT_VALID_FLAGS) return -EINVAL; - if (attr_flags & MOUNT_ATTR_RDONLY) - mnt_flags |= MNT_READONLY; - if (attr_flags & MOUNT_ATTR_NOSUID) - mnt_flags |= MNT_NOSUID; - if (attr_flags & MOUNT_ATTR_NODEV) - mnt_flags |= MNT_NODEV; - if (attr_flags & MOUNT_ATTR_NOEXEC) - mnt_flags |= MNT_NOEXEC; - if (attr_flags & MOUNT_ATTR_NODIRATIME) - mnt_flags |= MNT_NODIRATIME; + mnt_flags = attr_flags_to_mnt_flags(attr_flags); switch (attr_flags & MOUNT_ATTR__ATIME) { case MOUNT_ATTR_STRICTATIME: @@ -3787,6 +3822,362 @@ out0: return error; } +static unsigned int recalc_flags(struct mount_kattr *kattr, struct mount *mnt) +{ + unsigned int flags = mnt->mnt.mnt_flags; + + /* flags to clear */ + flags &= ~kattr->attr_clr; + /* flags to raise */ + flags |= kattr->attr_set; + + return flags; +} + +static int can_idmap_mount(const struct mount_kattr *kattr, struct mount *mnt) +{ + struct vfsmount *m = &mnt->mnt; + + if (!kattr->mnt_userns) + return 0; + + /* + * Once a mount has been idmapped we don't allow it to change its + * mapping. It makes things simpler and callers can just create + * another bind-mount they can idmap if they want to. + */ + if (mnt_user_ns(m) != &init_user_ns) + return -EPERM; + + /* The underlying filesystem doesn't support idmapped mounts yet. */ + if (!(m->mnt_sb->s_type->fs_flags & FS_ALLOW_IDMAP)) + return -EINVAL; + + /* We're not controlling the superblock. */ + if (!ns_capable(m->mnt_sb->s_user_ns, CAP_SYS_ADMIN)) + return -EPERM; + + /* Mount has already been visible in the filesystem hierarchy. */ + if (!is_anon_ns(mnt->mnt_ns)) + return -EINVAL; + + return 0; +} + +static struct mount *mount_setattr_prepare(struct mount_kattr *kattr, + struct mount *mnt, int *err) +{ + struct mount *m = mnt, *last = NULL; + + if (!is_mounted(&m->mnt)) { + *err = -EINVAL; + goto out; + } + + if (!(mnt_has_parent(m) ? check_mnt(m) : is_anon_ns(m->mnt_ns))) { + *err = -EINVAL; + goto out; + } + + do { + unsigned int flags; + + flags = recalc_flags(kattr, m); + if (!can_change_locked_flags(m, flags)) { + *err = -EPERM; + goto out; + } + + *err = can_idmap_mount(kattr, m); + if (*err) + goto out; + + last = m; + + if ((kattr->attr_set & MNT_READONLY) && + !(m->mnt.mnt_flags & MNT_READONLY)) { + *err = mnt_hold_writers(m); + if (*err) + goto out; + } + } while (kattr->recurse && (m = next_mnt(m, mnt))); + +out: + return last; +} + +static void do_idmap_mount(const struct mount_kattr *kattr, struct mount *mnt) +{ + struct user_namespace *mnt_userns; + + if (!kattr->mnt_userns) + return; + + mnt_userns = get_user_ns(kattr->mnt_userns); + /* Pairs with smp_load_acquire() in mnt_user_ns(). */ + smp_store_release(&mnt->mnt.mnt_userns, mnt_userns); +} + +static void mount_setattr_commit(struct mount_kattr *kattr, + struct mount *mnt, struct mount *last, + int err) +{ + struct mount *m = mnt; + + do { + if (!err) { + unsigned int flags; + + do_idmap_mount(kattr, m); + flags = recalc_flags(kattr, m); + WRITE_ONCE(m->mnt.mnt_flags, flags); + } + + /* + * We either set MNT_READONLY above so make it visible + * before ~MNT_WRITE_HOLD or we failed to recursively + * apply mount options. + */ + if ((kattr->attr_set & MNT_READONLY) && + (m->mnt.mnt_flags & MNT_WRITE_HOLD)) + mnt_unhold_writers(m); + + if (!err && kattr->propagation) + change_mnt_propagation(m, kattr->propagation); + + /* + * On failure, only cleanup until we found the first mount + * we failed to handle. + */ + if (err && m == last) + break; + } while (kattr->recurse && (m = next_mnt(m, mnt))); + + if (!err) + touch_mnt_namespace(mnt->mnt_ns); +} + +static int do_mount_setattr(struct path *path, struct mount_kattr *kattr) +{ + struct mount *mnt = real_mount(path->mnt), *last = NULL; + int err = 0; + + if (path->dentry != mnt->mnt.mnt_root) + return -EINVAL; + + if (kattr->propagation) { + /* + * Only take namespace_lock() if we're actually changing + * propagation. + */ + namespace_lock(); + if (kattr->propagation == MS_SHARED) { + err = invent_group_ids(mnt, kattr->recurse); + if (err) { + namespace_unlock(); + return err; + } + } + } + + lock_mount_hash(); + + /* + * Get the mount tree in a shape where we can change mount + * properties without failure. + */ + last = mount_setattr_prepare(kattr, mnt, &err); + if (last) /* Commit all changes or revert to the old state. */ + mount_setattr_commit(kattr, mnt, last, err); + + unlock_mount_hash(); + + if (kattr->propagation) { + namespace_unlock(); + if (err) + cleanup_group_ids(mnt, NULL); + } + + return err; +} + +static int build_mount_idmapped(const struct mount_attr *attr, size_t usize, + struct mount_kattr *kattr, unsigned int flags) +{ + int err = 0; + struct ns_common *ns; + struct user_namespace *mnt_userns; + struct file *file; + + if (!((attr->attr_set | attr->attr_clr) & MOUNT_ATTR_IDMAP)) + return 0; + + /* + * We currently do not support clearing an idmapped mount. If this ever + * is a use-case we can revisit this but for now let's keep it simple + * and not allow it. + */ + if (attr->attr_clr & MOUNT_ATTR_IDMAP) + return -EINVAL; + + if (attr->userns_fd > INT_MAX) + return -EINVAL; + + file = fget(attr->userns_fd); + if (!file) + return -EBADF; + + if (!proc_ns_file(file)) { + err = -EINVAL; + goto out_fput; + } + + ns = get_proc_ns(file_inode(file)); + if (ns->ops->type != CLONE_NEWUSER) { + err = -EINVAL; + goto out_fput; + } + + /* + * The init_user_ns is used to indicate that a vfsmount is not idmapped. + * This is simpler than just having to treat NULL as unmapped. Users + * wanting to idmap a mount to init_user_ns can just use a namespace + * with an identity mapping. + */ + mnt_userns = container_of(ns, struct user_namespace, ns); + if (mnt_userns == &init_user_ns) { + err = -EPERM; + goto out_fput; + } + kattr->mnt_userns = get_user_ns(mnt_userns); + +out_fput: + fput(file); + return err; +} + +static int build_mount_kattr(const struct mount_attr *attr, size_t usize, + struct mount_kattr *kattr, unsigned int flags) +{ + unsigned int lookup_flags = LOOKUP_AUTOMOUNT | LOOKUP_FOLLOW; + + if (flags & AT_NO_AUTOMOUNT) + lookup_flags &= ~LOOKUP_AUTOMOUNT; + if (flags & AT_SYMLINK_NOFOLLOW) + lookup_flags &= ~LOOKUP_FOLLOW; + if (flags & AT_EMPTY_PATH) + lookup_flags |= LOOKUP_EMPTY; + + *kattr = (struct mount_kattr) { + .lookup_flags = lookup_flags, + .recurse = !!(flags & AT_RECURSIVE), + }; + + if (attr->propagation & ~MOUNT_SETATTR_PROPAGATION_FLAGS) + return -EINVAL; + if (hweight32(attr->propagation & MOUNT_SETATTR_PROPAGATION_FLAGS) > 1) + return -EINVAL; + kattr->propagation = attr->propagation; + + if ((attr->attr_set | attr->attr_clr) & ~MOUNT_SETATTR_VALID_FLAGS) + return -EINVAL; + + kattr->attr_set = attr_flags_to_mnt_flags(attr->attr_set); + kattr->attr_clr = attr_flags_to_mnt_flags(attr->attr_clr); + + /* + * Since the MOUNT_ATTR_<atime> values are an enum, not a bitmap, + * users wanting to transition to a different atime setting cannot + * simply specify the atime setting in @attr_set, but must also + * specify MOUNT_ATTR__ATIME in the @attr_clr field. + * So ensure that MOUNT_ATTR__ATIME can't be partially set in + * @attr_clr and that @attr_set can't have any atime bits set if + * MOUNT_ATTR__ATIME isn't set in @attr_clr. + */ + if (attr->attr_clr & MOUNT_ATTR__ATIME) { + if ((attr->attr_clr & MOUNT_ATTR__ATIME) != MOUNT_ATTR__ATIME) + return -EINVAL; + + /* + * Clear all previous time settings as they are mutually + * exclusive. + */ + kattr->attr_clr |= MNT_RELATIME | MNT_NOATIME; + switch (attr->attr_set & MOUNT_ATTR__ATIME) { + case MOUNT_ATTR_RELATIME: + kattr->attr_set |= MNT_RELATIME; + break; + case MOUNT_ATTR_NOATIME: + kattr->attr_set |= MNT_NOATIME; + break; + case MOUNT_ATTR_STRICTATIME: + break; + default: + return -EINVAL; + } + } else { + if (attr->attr_set & MOUNT_ATTR__ATIME) + return -EINVAL; + } + + return build_mount_idmapped(attr, usize, kattr, flags); +} + +static void finish_mount_kattr(struct mount_kattr *kattr) +{ + put_user_ns(kattr->mnt_userns); + kattr->mnt_userns = NULL; +} + +SYSCALL_DEFINE5(mount_setattr, int, dfd, const char __user *, path, + unsigned int, flags, struct mount_attr __user *, uattr, + size_t, usize) +{ + int err; + struct path target; + struct mount_attr attr; + struct mount_kattr kattr; + + BUILD_BUG_ON(sizeof(struct mount_attr) != MOUNT_ATTR_SIZE_VER0); + + if (flags & ~(AT_EMPTY_PATH | + AT_RECURSIVE | + AT_SYMLINK_NOFOLLOW | + AT_NO_AUTOMOUNT)) + return -EINVAL; + + if (unlikely(usize > PAGE_SIZE)) + return -E2BIG; + if (unlikely(usize < MOUNT_ATTR_SIZE_VER0)) + return -EINVAL; + + if (!may_mount()) + return -EPERM; + + err = copy_struct_from_user(&attr, sizeof(attr), uattr, usize); + if (err) + return err; + + /* Don't bother walking through the mounts if this is a nop. */ + if (attr.attr_set == 0 && + attr.attr_clr == 0 && + attr.propagation == 0) + return 0; + + err = build_mount_kattr(&attr, usize, &kattr, flags); + if (err) + return err; + + err = user_path_at(dfd, path, kattr.lookup_flags, &target); + if (err) + return err; + + err = do_mount_setattr(&target, &kattr); + finish_mount_kattr(&kattr); + path_put(&target); + return err; +} + static void __init init_mount_tree(void) { struct vfsmount *mnt; diff --git a/fs/nfs/Kconfig b/fs/nfs/Kconfig index e2a488d403a6..14a72224b657 100644 --- a/fs/nfs/Kconfig +++ b/fs/nfs/Kconfig @@ -127,7 +127,7 @@ config PNFS_BLOCK config PNFS_FLEXFILE_LAYOUT tristate depends on NFS_V4_1 && NFS_V3 - default m + default NFS_V4 config NFS_V4_1_IMPLEMENTATION_ID_DOMAIN string "NFSv4.1 Implementation ID Domain" diff --git a/fs/nfs/blocklayout/blocklayout.c b/fs/nfs/blocklayout/blocklayout.c index 3be6836074ae..fe860c538747 100644 --- a/fs/nfs/blocklayout/blocklayout.c +++ b/fs/nfs/blocklayout/blocklayout.c @@ -115,19 +115,14 @@ bl_submit_bio(struct bio *bio) return NULL; } -static struct bio * -bl_alloc_init_bio(int npg, struct block_device *bdev, sector_t disk_sector, +static struct bio *bl_alloc_init_bio(unsigned int npg, + struct block_device *bdev, sector_t disk_sector, bio_end_io_t end_io, struct parallel_io *par) { struct bio *bio; - npg = min(npg, BIO_MAX_PAGES); + npg = bio_max_segs(npg); bio = bio_alloc(GFP_NOIO, npg); - if (!bio && (current->flags & PF_MEMALLOC)) { - while (!bio && (npg /= 2)) - bio = bio_alloc(GFP_NOIO, npg); - } - if (bio) { bio->bi_iter.bi_sector = disk_sector; bio_set_dev(bio, bdev); diff --git a/fs/nfs/callback_xdr.c b/fs/nfs/callback_xdr.c index 79ff172eb1c8..c5348ba81129 100644 --- a/fs/nfs/callback_xdr.c +++ b/fs/nfs/callback_xdr.c @@ -1060,6 +1060,7 @@ static const struct svc_procedure nfs4_callback_procedures1[] = { .pc_decode = nfs4_decode_void, .pc_encode = nfs4_encode_void, .pc_xdrressize = 1, + .pc_name = "NULL", }, [CB_COMPOUND] = { .pc_func = nfs4_callback_compound, @@ -1067,6 +1068,7 @@ static const struct svc_procedure nfs4_callback_procedures1[] = { .pc_argsize = 256, .pc_ressize = 256, .pc_xdrressize = NFS4_CALLBACK_BUFSIZE, + .pc_name = "COMPOUND", } }; diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index ef827ae193d2..fc4f490f2d78 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -81,8 +81,9 @@ static struct nfs_open_dir_context *alloc_nfs_open_dir_context(struct inode *dir spin_lock(&dir->i_lock); if (list_empty(&nfsi->open_files) && (nfsi->cache_validity & NFS_INO_DATA_INVAL_DEFER)) - nfsi->cache_validity |= NFS_INO_INVALID_DATA | - NFS_INO_REVAL_FORCED; + nfs_set_cache_invalid(dir, + NFS_INO_INVALID_DATA | + NFS_INO_REVAL_FORCED); list_add(&ctx->list, &nfsi->open_files); spin_unlock(&dir->i_lock); return ctx; @@ -1401,6 +1402,13 @@ out_force: goto out; } +static void nfs_mark_dir_for_revalidate(struct inode *inode) +{ + spin_lock(&inode->i_lock); + nfs_set_cache_invalid(inode, NFS_INO_REVAL_PAGECACHE); + spin_unlock(&inode->i_lock); +} + /* * We judge how long we want to trust negative * dentries by looking at the parent inode mtime. @@ -1435,19 +1443,14 @@ nfs_lookup_revalidate_done(struct inode *dir, struct dentry *dentry, __func__, dentry); return 1; case 0: - nfs_mark_for_revalidate(dir); - if (inode && S_ISDIR(inode->i_mode)) { - /* Purge readdir caches. */ - nfs_zap_caches(inode); - /* - * We can't d_drop the root of a disconnected tree: - * its d_hash is on the s_anon list and d_drop() would hide - * it from shrink_dcache_for_unmount(), leading to busy - * inodes on unmount and further oopses. - */ - if (IS_ROOT(dentry)) - return 1; - } + /* + * We can't d_drop the root of a disconnected tree: + * its d_hash is on the s_anon list and d_drop() would hide + * it from shrink_dcache_for_unmount(), leading to busy + * inodes on unmount and further oopses. + */ + if (inode && IS_ROOT(dentry)) + return 1; dfprintk(LOOKUPCACHE, "NFS: %s(%pd2) is invalid\n", __func__, dentry); return 0; @@ -1525,6 +1528,13 @@ out: nfs_free_fattr(fattr); nfs_free_fhandle(fhandle); nfs4_label_free(label); + + /* + * If the lookup failed despite the dentry change attribute being + * a match, then we should revalidate the directory cache. + */ + if (!ret && nfs_verify_change_attribute(dir, dentry->d_time)) + nfs_mark_dir_for_revalidate(dir); return nfs_lookup_revalidate_done(dir, dentry, inode, ret); } @@ -1567,7 +1577,7 @@ nfs_do_lookup_revalidate(struct inode *dir, struct dentry *dentry, error = nfs_lookup_verify_inode(inode, flags); if (error) { if (error == -ESTALE) - nfs_zap_caches(dir); + nfs_mark_dir_for_revalidate(dir); goto out_bad; } nfs_advise_use_readdirplus(dir); @@ -1691,10 +1701,9 @@ static void nfs_drop_nlink(struct inode *inode) if (inode->i_nlink > 0) drop_nlink(inode); NFS_I(inode)->attr_gencount = nfs_inc_attr_generation_counter(); - NFS_I(inode)->cache_validity |= NFS_INO_INVALID_CHANGE - | NFS_INO_INVALID_CTIME - | NFS_INO_INVALID_OTHER - | NFS_INO_REVAL_FORCED; + nfs_set_cache_invalid( + inode, NFS_INO_INVALID_CHANGE | NFS_INO_INVALID_CTIME | + NFS_INO_INVALID_OTHER | NFS_INO_REVAL_FORCED); spin_unlock(&inode->i_lock); } @@ -1706,7 +1715,7 @@ static void nfs_dentry_iput(struct dentry *dentry, struct inode *inode) { if (S_ISDIR(inode->i_mode)) /* drop any readdir cache as it could easily be old */ - NFS_I(inode)->cache_validity |= NFS_INO_INVALID_DATA; + nfs_set_cache_invalid(inode, NFS_INO_INVALID_DATA); if (dentry->d_flags & DCACHE_NFSFS_RENAMED) { nfs_complete_unlink(dentry, inode); @@ -2064,7 +2073,6 @@ out: dput(parent); return d; out_error: - nfs_mark_for_revalidate(dir); d = ERR_PTR(error); goto out; } @@ -2095,8 +2103,8 @@ EXPORT_SYMBOL_GPL(nfs_instantiate); * that the operation succeeded on the server, but an error in the * reply path made it appear to have failed. */ -int nfs_create(struct inode *dir, struct dentry *dentry, - umode_t mode, bool excl) +int nfs_create(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, bool excl) { struct iattr attr; int open_flags = excl ? O_CREAT | O_EXCL : O_CREAT; @@ -2124,7 +2132,8 @@ EXPORT_SYMBOL_GPL(nfs_create); * See comments for nfs_proc_create regarding failed operations. */ int -nfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t rdev) +nfs_mknod(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, dev_t rdev) { struct iattr attr; int status; @@ -2150,7 +2159,8 @@ EXPORT_SYMBOL_GPL(nfs_mknod); /* * See comments for nfs_proc_create regarding failed operations. */ -int nfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) +int nfs_mkdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { struct iattr attr; int error; @@ -2295,7 +2305,8 @@ EXPORT_SYMBOL_GPL(nfs_unlink); * now have a new file handle and can instantiate an in-core NFS inode * and move the raw page into its mapping. */ -int nfs_symlink(struct inode *dir, struct dentry *dentry, const char *symname) +int nfs_symlink(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, const char *symname) { struct page *page; char *kaddr; @@ -2398,9 +2409,9 @@ EXPORT_SYMBOL_GPL(nfs_link); * If these conditions are met, we can drop the dentries before doing * the rename. */ -int nfs_rename(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry, - unsigned int flags) +int nfs_rename(struct user_namespace *mnt_userns, struct inode *old_dir, + struct dentry *old_dentry, struct inode *new_dir, + struct dentry *new_dentry, unsigned int flags) { struct inode *old_inode = d_inode(old_dentry); struct inode *new_inode = d_inode(new_dentry); @@ -2470,9 +2481,9 @@ int nfs_rename(struct inode *old_dir, struct dentry *old_dentry, if (error == 0) { spin_lock(&old_inode->i_lock); NFS_I(old_inode)->attr_gencount = nfs_inc_attr_generation_counter(); - NFS_I(old_inode)->cache_validity |= NFS_INO_INVALID_CHANGE - | NFS_INO_INVALID_CTIME - | NFS_INO_REVAL_FORCED; + nfs_set_cache_invalid(old_inode, NFS_INO_INVALID_CHANGE | + NFS_INO_INVALID_CTIME | + NFS_INO_REVAL_FORCED); spin_unlock(&old_inode->i_lock); } out: @@ -2939,7 +2950,9 @@ static int nfs_execute_ok(struct inode *inode, int mask) return ret; } -int nfs_permission(struct inode *inode, int mask) +int nfs_permission(struct user_namespace *mnt_userns, + struct inode *inode, + int mask) { const struct cred *cred = current_cred(); int res = 0; @@ -2987,7 +3000,7 @@ out_notsup: res = nfs_revalidate_inode(NFS_SERVER(inode), inode); if (res == 0) - res = generic_permission(inode, mask); + res = generic_permission(&init_user_ns, inode, mask); goto out; } EXPORT_SYMBOL_GPL(nfs_permission); diff --git a/fs/nfs/export.c b/fs/nfs/export.c index 7412bb164fa7..f2b34cfe286c 100644 --- a/fs/nfs/export.c +++ b/fs/nfs/export.c @@ -167,10 +167,28 @@ out: return parent; } +static u64 nfs_fetch_iversion(struct inode *inode) +{ + struct nfs_server *server = NFS_SERVER(inode); + + /* Is this the right call?: */ + nfs_revalidate_inode(server, inode); + /* + * Also, note we're ignoring any returned error. That seems to be + * the practice for cache consistency information elsewhere in + * the server, but I'm not sure why. + */ + if (server->nfs_client->rpc_ops->version >= 4) + return inode_peek_iversion_raw(inode); + else + return time_to_chattr(&inode->i_ctime); +} + const struct export_operations nfs_export_ops = { .encode_fh = nfs_encode_fh, .fh_to_dentry = nfs_fh_to_dentry, .get_parent = nfs_get_parent, + .fetch_iversion = nfs_fetch_iversion, .flags = EXPORT_OP_NOWCC|EXPORT_OP_NOSUBTREECHK| EXPORT_OP_CLOSE_BEFORE_UNLINK|EXPORT_OP_REMOTE_FS| EXPORT_OP_NOATOMIC_ATTR, diff --git a/fs/nfs/file.c b/fs/nfs/file.c index 63940a7a70be..16ad5050e046 100644 --- a/fs/nfs/file.c +++ b/fs/nfs/file.c @@ -89,7 +89,7 @@ nfs_file_release(struct inode *inode, struct file *filp) EXPORT_SYMBOL_GPL(nfs_file_release); /** - * nfs_revalidate_size - Revalidate the file size + * nfs_revalidate_file_size - Revalidate the file size * @inode: pointer to inode struct * @filp: pointer to struct file * @@ -606,8 +606,8 @@ ssize_t nfs_file_write(struct kiocb *iocb, struct iov_iter *from) { struct file *file = iocb->ki_filp; struct inode *inode = file_inode(file); - unsigned long written = 0; - ssize_t result; + unsigned int mntflags = NFS_SERVER(inode)->flags; + ssize_t result, written; errseq_t since; int error; @@ -626,13 +626,13 @@ ssize_t nfs_file_write(struct kiocb *iocb, struct iov_iter *from) /* * O_APPEND implies that we must revalidate the file length. */ - if (iocb->ki_flags & IOCB_APPEND) { + if (iocb->ki_flags & IOCB_APPEND || iocb->ki_pos > i_size_read(inode)) { result = nfs_revalidate_file_size(inode, file); if (result) goto out; } - if (iocb->ki_pos > i_size_read(inode)) - nfs_revalidate_mapping(inode, file->f_mapping); + + nfs_clear_invalid_mapping(file->f_mapping); since = filemap_sample_wb_err(file->f_mapping); nfs_start_io_write(inode); @@ -648,6 +648,21 @@ ssize_t nfs_file_write(struct kiocb *iocb, struct iov_iter *from) written = result; iocb->ki_pos += written; + + if (mntflags & NFS_MOUNT_WRITE_EAGER) { + result = filemap_fdatawrite_range(file->f_mapping, + iocb->ki_pos - written, + iocb->ki_pos - 1); + if (result < 0) + goto out; + } + if (mntflags & NFS_MOUNT_WRITE_WAIT) { + result = filemap_fdatawait_range(file->f_mapping, + iocb->ki_pos - written, + iocb->ki_pos - 1); + if (result < 0) + goto out; + } result = generic_write_sync(iocb, written); if (result < 0) goto out; diff --git a/fs/nfs/fs_context.c b/fs/nfs/fs_context.c index 06894bcdea2d..971a9251c1d9 100644 --- a/fs/nfs/fs_context.c +++ b/fs/nfs/fs_context.c @@ -82,6 +82,7 @@ enum nfs_param { Opt_v, Opt_vers, Opt_wsize, + Opt_write, }; enum { @@ -113,6 +114,19 @@ static const struct constant_table nfs_param_enums_lookupcache[] = { {} }; +enum { + Opt_write_lazy, + Opt_write_eager, + Opt_write_wait, +}; + +static const struct constant_table nfs_param_enums_write[] = { + { "lazy", Opt_write_lazy }, + { "eager", Opt_write_eager }, + { "wait", Opt_write_wait }, + {} +}; + static const struct fs_parameter_spec nfs_fs_parameters[] = { fsparam_flag_no("ac", Opt_ac), fsparam_u32 ("acdirmax", Opt_acdirmax), @@ -171,6 +185,7 @@ static const struct fs_parameter_spec nfs_fs_parameters[] = { fsparam_flag ("v4.1", Opt_v), fsparam_flag ("v4.2", Opt_v), fsparam_string("vers", Opt_vers), + fsparam_enum ("write", Opt_write, nfs_param_enums_write), fsparam_u32 ("wsize", Opt_wsize), {} }; @@ -770,6 +785,24 @@ static int nfs_fs_context_parse_param(struct fs_context *fc, goto out_invalid_value; } break; + case Opt_write: + switch (result.uint_32) { + case Opt_write_lazy: + ctx->flags &= + ~(NFS_MOUNT_WRITE_EAGER | NFS_MOUNT_WRITE_WAIT); + break; + case Opt_write_eager: + ctx->flags |= NFS_MOUNT_WRITE_EAGER; + ctx->flags &= ~NFS_MOUNT_WRITE_WAIT; + break; + case Opt_write_wait: + ctx->flags |= + NFS_MOUNT_WRITE_EAGER | NFS_MOUNT_WRITE_WAIT; + break; + default: + goto out_invalid_value; + } + break; /* * Special options @@ -1479,6 +1512,8 @@ static int nfs_init_fs_context(struct fs_context *fc) ctx->selected_flavor = RPC_AUTH_MAXFLAVOR; ctx->minorversion = 0; ctx->need_mount = true; + + fc->s_iflags |= SB_I_STABLE_WRITES; } fc->fs_private = ctx; fc->ops = &nfs_fs_context_ops; diff --git a/fs/nfs/fscache.c b/fs/nfs/fscache.c index a60df88efc40..c4c021c6ebbd 100644 --- a/fs/nfs/fscache.c +++ b/fs/nfs/fscache.c @@ -390,10 +390,6 @@ static void nfs_readpage_from_fscache_complete(struct page *page, if (!error) { SetPageUptodate(page); unlock_page(page); - } else { - error = nfs_readpage_async(context, page->mapping->host, page); - if (error) - unlock_page(page); } } diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 522aa10a1a3e..a7fb076a5f44 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -195,7 +195,19 @@ bool nfs_check_cache_invalid(struct inode *inode, unsigned long flags) } EXPORT_SYMBOL_GPL(nfs_check_cache_invalid); -static void nfs_set_cache_invalid(struct inode *inode, unsigned long flags) +#ifdef CONFIG_NFS_V4_2 +static bool nfs_has_xattr_cache(const struct nfs_inode *nfsi) +{ + return nfsi->xattr_cache != NULL; +} +#else +static bool nfs_has_xattr_cache(const struct nfs_inode *nfsi) +{ + return false; +} +#endif + +void nfs_set_cache_invalid(struct inode *inode, unsigned long flags) { struct nfs_inode *nfsi = NFS_I(inode); bool have_delegation = NFS_PROTO(inode)->have_delegation(inode, FMODE_READ); @@ -209,12 +221,15 @@ static void nfs_set_cache_invalid(struct inode *inode, unsigned long flags) | NFS_INO_INVALID_XATTR); } + if (!nfs_has_xattr_cache(nfsi)) + flags &= ~NFS_INO_INVALID_XATTR; if (inode->i_mapping->nrpages == 0) flags &= ~(NFS_INO_INVALID_DATA|NFS_INO_DATA_INVAL_DEFER); nfsi->cache_validity |= flags; if (flags & NFS_INO_INVALID_DATA) nfs_fscache_invalidate(inode); } +EXPORT_SYMBOL_GPL(nfs_set_cache_invalid); /* * Invalidate the local caches @@ -594,7 +609,8 @@ EXPORT_SYMBOL_GPL(nfs_fhget); #define NFS_VALID_ATTRS (ATTR_MODE|ATTR_UID|ATTR_GID|ATTR_SIZE|ATTR_ATIME|ATTR_ATIME_SET|ATTR_MTIME|ATTR_MTIME_SET|ATTR_FILE|ATTR_OPEN) int -nfs_setattr(struct dentry *dentry, struct iattr *attr) +nfs_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *attr) { struct inode *inode = d_inode(dentry); struct nfs_fattr *fattr; @@ -787,8 +803,8 @@ static bool nfs_need_revalidate_inode(struct inode *inode) return false; } -int nfs_getattr(const struct path *path, struct kstat *stat, - u32 request_mask, unsigned int query_flags) +int nfs_getattr(struct user_namespace *mnt_userns, const struct path *path, + struct kstat *stat, u32 request_mask, unsigned int query_flags) { struct inode *inode = d_inode(path->dentry); struct nfs_server *server = NFS_SERVER(inode); @@ -857,7 +873,7 @@ out_no_revalidate: /* Only return attributes that were revalidated. */ stat->result_mask &= request_mask; out_no_update: - generic_fillattr(inode, stat); + generic_fillattr(&init_user_ns, inode, stat); stat->ino = nfs_compat_user_ino64(NFS_FILEID(inode)); if (S_ISDIR(inode->i_mode)) stat->blksize = NFS_SERVER(inode)->dtsize; @@ -1052,8 +1068,8 @@ void nfs_inode_attach_open_context(struct nfs_open_context *ctx) spin_lock(&inode->i_lock); if (list_empty(&nfsi->open_files) && (nfsi->cache_validity & NFS_INO_DATA_INVAL_DEFER)) - nfsi->cache_validity |= NFS_INO_INVALID_DATA | - NFS_INO_REVAL_FORCED; + nfs_set_cache_invalid(inode, NFS_INO_INVALID_DATA | + NFS_INO_REVAL_FORCED); list_add_tail_rcu(&ctx->list, &nfsi->open_files); spin_unlock(&inode->i_lock); } @@ -1257,55 +1273,19 @@ static int nfs_invalidate_mapping(struct inode *inode, struct address_space *map return 0; } -bool nfs_mapping_need_revalidate_inode(struct inode *inode) -{ - return nfs_check_cache_invalid(inode, NFS_INO_REVAL_PAGECACHE) || - NFS_STALE(inode); -} - -int nfs_revalidate_mapping_rcu(struct inode *inode) -{ - struct nfs_inode *nfsi = NFS_I(inode); - unsigned long *bitlock = &nfsi->flags; - int ret = 0; - - if (IS_SWAPFILE(inode)) - goto out; - if (nfs_mapping_need_revalidate_inode(inode)) { - ret = -ECHILD; - goto out; - } - spin_lock(&inode->i_lock); - if (test_bit(NFS_INO_INVALIDATING, bitlock) || - (nfsi->cache_validity & NFS_INO_INVALID_DATA)) - ret = -ECHILD; - spin_unlock(&inode->i_lock); -out: - return ret; -} - /** - * nfs_revalidate_mapping - Revalidate the pagecache - * @inode: pointer to host inode + * nfs_clear_invalid_mapping - Conditionally clear a mapping * @mapping: pointer to mapping + * + * If the NFS_INO_INVALID_DATA inode flag is set, clear the mapping. */ -int nfs_revalidate_mapping(struct inode *inode, - struct address_space *mapping) +int nfs_clear_invalid_mapping(struct address_space *mapping) { + struct inode *inode = mapping->host; struct nfs_inode *nfsi = NFS_I(inode); unsigned long *bitlock = &nfsi->flags; int ret = 0; - /* swapfiles are not supposed to be shared. */ - if (IS_SWAPFILE(inode)) - goto out; - - if (nfs_mapping_need_revalidate_inode(inode)) { - ret = __nfs_revalidate_inode(NFS_SERVER(inode), inode); - if (ret < 0) - goto out; - } - /* * We must clear NFS_INO_INVALID_DATA first to ensure that * invalidations that come in while we're shooting down the mappings @@ -1336,8 +1316,8 @@ int nfs_revalidate_mapping(struct inode *inode, set_bit(NFS_INO_INVALIDATING, bitlock); smp_wmb(); - nfsi->cache_validity &= ~(NFS_INO_INVALID_DATA| - NFS_INO_DATA_INVAL_DEFER); + nfsi->cache_validity &= + ~(NFS_INO_INVALID_DATA | NFS_INO_DATA_INVAL_DEFER); spin_unlock(&inode->i_lock); trace_nfs_invalidate_mapping_enter(inode); ret = nfs_invalidate_mapping(inode, mapping); @@ -1350,6 +1330,53 @@ out: return ret; } +bool nfs_mapping_need_revalidate_inode(struct inode *inode) +{ + return nfs_check_cache_invalid(inode, NFS_INO_REVAL_PAGECACHE) || + NFS_STALE(inode); +} + +int nfs_revalidate_mapping_rcu(struct inode *inode) +{ + struct nfs_inode *nfsi = NFS_I(inode); + unsigned long *bitlock = &nfsi->flags; + int ret = 0; + + if (IS_SWAPFILE(inode)) + goto out; + if (nfs_mapping_need_revalidate_inode(inode)) { + ret = -ECHILD; + goto out; + } + spin_lock(&inode->i_lock); + if (test_bit(NFS_INO_INVALIDATING, bitlock) || + (nfsi->cache_validity & NFS_INO_INVALID_DATA)) + ret = -ECHILD; + spin_unlock(&inode->i_lock); +out: + return ret; +} + +/** + * nfs_revalidate_mapping - Revalidate the pagecache + * @inode: pointer to host inode + * @mapping: pointer to mapping + */ +int nfs_revalidate_mapping(struct inode *inode, struct address_space *mapping) +{ + /* swapfiles are not supposed to be shared. */ + if (IS_SWAPFILE(inode)) + return 0; + + if (nfs_mapping_need_revalidate_inode(inode)) { + int ret = __nfs_revalidate_inode(NFS_SERVER(inode), inode); + if (ret < 0) + return ret; + } + + return nfs_clear_invalid_mapping(mapping); +} + static bool nfs_file_has_writers(struct nfs_inode *nfsi) { struct inode *inode = &nfsi->vfs_inode; diff --git a/fs/nfs/internal.h b/fs/nfs/internal.h index 62d3189745cd..7b644d6c09e4 100644 --- a/fs/nfs/internal.h +++ b/fs/nfs/internal.h @@ -378,14 +378,18 @@ extern unsigned long nfs_access_cache_count(struct shrinker *shrink, extern unsigned long nfs_access_cache_scan(struct shrinker *shrink, struct shrink_control *sc); struct dentry *nfs_lookup(struct inode *, struct dentry *, unsigned int); -int nfs_create(struct inode *, struct dentry *, umode_t, bool); -int nfs_mkdir(struct inode *, struct dentry *, umode_t); +int nfs_create(struct user_namespace *, struct inode *, struct dentry *, + umode_t, bool); +int nfs_mkdir(struct user_namespace *, struct inode *, struct dentry *, + umode_t); int nfs_rmdir(struct inode *, struct dentry *); int nfs_unlink(struct inode *, struct dentry *); -int nfs_symlink(struct inode *, struct dentry *, const char *); +int nfs_symlink(struct user_namespace *, struct inode *, struct dentry *, + const char *); int nfs_link(struct dentry *, struct inode *, struct dentry *); -int nfs_mknod(struct inode *, struct dentry *, umode_t, dev_t); -int nfs_rename(struct inode *, struct dentry *, +int nfs_mknod(struct user_namespace *, struct inode *, struct dentry *, umode_t, + dev_t); +int nfs_rename(struct user_namespace *, struct inode *, struct dentry *, struct inode *, struct dentry *, unsigned int); /* file.c */ @@ -407,7 +411,8 @@ extern int nfs_write_inode(struct inode *, struct writeback_control *); extern int nfs_drop_inode(struct inode *); extern void nfs_clear_inode(struct inode *); extern void nfs_evict_inode(struct inode *); -void nfs_zap_acl_cache(struct inode *inode); +extern void nfs_zap_acl_cache(struct inode *inode); +extern void nfs_set_cache_invalid(struct inode *inode, unsigned long flags); extern bool nfs_check_cache_invalid(struct inode *, unsigned long); extern int nfs_wait_bit_killable(struct wait_bit_key *key, int mode); extern int nfs_wait_atomic_killable(atomic_t *p, unsigned int mode); diff --git a/fs/nfs/namespace.c b/fs/nfs/namespace.c index 2bcbe38afe2e..93e60e921f92 100644 --- a/fs/nfs/namespace.c +++ b/fs/nfs/namespace.c @@ -208,20 +208,23 @@ out_fc: } static int -nfs_namespace_getattr(const struct path *path, struct kstat *stat, - u32 request_mask, unsigned int query_flags) +nfs_namespace_getattr(struct user_namespace *mnt_userns, + const struct path *path, struct kstat *stat, + u32 request_mask, unsigned int query_flags) { if (NFS_FH(d_inode(path->dentry))->size != 0) - return nfs_getattr(path, stat, request_mask, query_flags); - generic_fillattr(d_inode(path->dentry), stat); + return nfs_getattr(mnt_userns, path, stat, request_mask, + query_flags); + generic_fillattr(&init_user_ns, d_inode(path->dentry), stat); return 0; } static int -nfs_namespace_setattr(struct dentry *dentry, struct iattr *attr) +nfs_namespace_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *attr) { if (NFS_FH(d_inode(dentry))->size != 0) - return nfs_setattr(dentry, attr); + return nfs_setattr(mnt_userns, dentry, attr); return -EACCES; } diff --git a/fs/nfs/nfs3_fs.h b/fs/nfs/nfs3_fs.h index 1b950b66b3bb..c8a192802dda 100644 --- a/fs/nfs/nfs3_fs.h +++ b/fs/nfs/nfs3_fs.h @@ -12,7 +12,8 @@ */ #ifdef CONFIG_NFS_V3_ACL extern struct posix_acl *nfs3_get_acl(struct inode *inode, int type); -extern int nfs3_set_acl(struct inode *inode, struct posix_acl *acl, int type); +extern int nfs3_set_acl(struct user_namespace *mnt_userns, struct inode *inode, + struct posix_acl *acl, int type); extern int nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl, struct posix_acl *dfacl); extern ssize_t nfs3_listxattr(struct dentry *, char *, size_t); diff --git a/fs/nfs/nfs3acl.c b/fs/nfs/nfs3acl.c index c6c863382f37..bb386a691e69 100644 --- a/fs/nfs/nfs3acl.c +++ b/fs/nfs/nfs3acl.c @@ -111,6 +111,7 @@ struct posix_acl *nfs3_get_acl(struct inode *inode, int type) fallthrough; case -ENOTSUPP: status = -EOPNOTSUPP; + goto getout; default: goto getout; } @@ -251,7 +252,8 @@ int nfs3_proc_setacls(struct inode *inode, struct posix_acl *acl, } -int nfs3_set_acl(struct inode *inode, struct posix_acl *acl, int type) +int nfs3_set_acl(struct user_namespace *mnt_userns, struct inode *inode, + struct posix_acl *acl, int type) { struct posix_acl *orig = acl, *dfacl = NULL, *alloc; int status; diff --git a/fs/nfs/nfs3xdr.c b/fs/nfs/nfs3xdr.c index ca10072644ff..ed1c83738c30 100644 --- a/fs/nfs/nfs3xdr.c +++ b/fs/nfs/nfs3xdr.c @@ -36,6 +36,7 @@ #define NFS3_pagepad_sz (1) /* Page padding */ #define NFS3_fhandle_sz (1+16) #define NFS3_fh_sz (NFS3_fhandle_sz) /* shorthand */ +#define NFS3_post_op_fh_sz (1+NFS3_fh_sz) #define NFS3_sattr_sz (15) #define NFS3_filename_sz (1+(NFS3_MAXNAMLEN>>2)) #define NFS3_path_sz (1+(NFS3_MAXPATHLEN>>2)) @@ -73,7 +74,7 @@ #define NFS3_readlinkres_sz (1+NFS3_post_op_attr_sz+1+NFS3_pagepad_sz) #define NFS3_readres_sz (1+NFS3_post_op_attr_sz+3+NFS3_pagepad_sz) #define NFS3_writeres_sz (1+NFS3_wcc_data_sz+4) -#define NFS3_createres_sz (1+NFS3_fh_sz+NFS3_post_op_attr_sz+NFS3_wcc_data_sz) +#define NFS3_createres_sz (1+NFS3_post_op_fh_sz+NFS3_post_op_attr_sz+NFS3_wcc_data_sz) #define NFS3_renameres_sz (1+(2 * NFS3_wcc_data_sz)) #define NFS3_linkres_sz (1+NFS3_post_op_attr_sz+NFS3_wcc_data_sz) #define NFS3_readdirres_sz (1+NFS3_post_op_attr_sz+2+NFS3_pagepad_sz) diff --git a/fs/nfs/nfs42proc.c b/fs/nfs/nfs42proc.c index f3fd935620fc..094024b0aca1 100644 --- a/fs/nfs/nfs42proc.c +++ b/fs/nfs/nfs42proc.c @@ -357,13 +357,15 @@ static ssize_t _nfs42_proc_copy(struct file *src, truncate_pagecache_range(dst_inode, pos_dst, pos_dst + res->write_res.count); spin_lock(&dst_inode->i_lock); - NFS_I(dst_inode)->cache_validity |= (NFS_INO_REVAL_PAGECACHE | - NFS_INO_REVAL_FORCED | NFS_INO_INVALID_SIZE | - NFS_INO_INVALID_ATTR | NFS_INO_INVALID_DATA); + nfs_set_cache_invalid( + dst_inode, NFS_INO_REVAL_PAGECACHE | NFS_INO_REVAL_FORCED | + NFS_INO_INVALID_SIZE | NFS_INO_INVALID_ATTR | + NFS_INO_INVALID_DATA); spin_unlock(&dst_inode->i_lock); spin_lock(&src_inode->i_lock); - NFS_I(src_inode)->cache_validity |= (NFS_INO_REVAL_PAGECACHE | - NFS_INO_REVAL_FORCED | NFS_INO_INVALID_ATIME); + nfs_set_cache_invalid(src_inode, NFS_INO_REVAL_PAGECACHE | + NFS_INO_REVAL_FORCED | + NFS_INO_INVALID_ATIME); spin_unlock(&src_inode->i_lock); status = res->write_res.count; out: diff --git a/fs/nfs/nfs4client.c b/fs/nfs/nfs4client.c index 86acffe7335c..889a9f4c0310 100644 --- a/fs/nfs/nfs4client.c +++ b/fs/nfs/nfs4client.c @@ -609,6 +609,7 @@ found: * changed. Schedule recovery! */ nfs4_schedule_path_down_recovery(pos); + goto out; default: goto out; } diff --git a/fs/nfs/nfs4file.c b/fs/nfs/nfs4file.c index 57b3821d975a..441a2fa073c8 100644 --- a/fs/nfs/nfs4file.c +++ b/fs/nfs/nfs4file.c @@ -420,7 +420,9 @@ static const struct nfs4_ssc_client_ops nfs4_ssc_clnt_ops_tbl = { */ void nfs42_ssc_register_ops(void) { +#ifdef CONFIG_NFSD_V4 nfs42_ssc_register(&nfs4_ssc_clnt_ops_tbl); +#endif } /** @@ -431,7 +433,9 @@ void nfs42_ssc_register_ops(void) */ void nfs42_ssc_unregister_ops(void) { +#ifdef CONFIG_NFSD_V4 nfs42_ssc_unregister(&nfs4_ssc_clnt_ops_tbl); +#endif } #endif /* CONFIG_NFS_V4_2 */ diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c index 2f4679a62712..c65c4b41e2c1 100644 --- a/fs/nfs/nfs4proc.c +++ b/fs/nfs/nfs4proc.c @@ -71,10 +71,6 @@ #include "nfs4trace.h" -#ifdef CONFIG_NFS_V4_2 -#include "nfs42.h" -#endif /* CONFIG_NFS_V4_2 */ - #define NFSDBG_FACILITY NFSDBG_PROC #define NFS4_BITMASK_SZ 3 @@ -1173,14 +1169,14 @@ int nfs4_call_sync(struct rpc_clnt *clnt, static void nfs4_inc_nlink_locked(struct inode *inode) { - NFS_I(inode)->cache_validity |= NFS_INO_INVALID_OTHER; + nfs_set_cache_invalid(inode, NFS_INO_INVALID_OTHER); inc_nlink(inode); } static void nfs4_dec_nlink_locked(struct inode *inode) { - NFS_I(inode)->cache_validity |= NFS_INO_INVALID_OTHER; + nfs_set_cache_invalid(inode, NFS_INO_INVALID_OTHER); drop_nlink(inode); } @@ -1191,35 +1187,31 @@ nfs4_update_changeattr_locked(struct inode *inode, { struct nfs_inode *nfsi = NFS_I(inode); - nfsi->cache_validity |= NFS_INO_INVALID_CTIME - | NFS_INO_INVALID_MTIME - | cache_validity; + cache_validity |= NFS_INO_INVALID_CTIME | NFS_INO_INVALID_MTIME; if (cinfo->atomic && cinfo->before == inode_peek_iversion_raw(inode)) { nfsi->cache_validity &= ~NFS_INO_REVAL_PAGECACHE; nfsi->attrtimeo_timestamp = jiffies; } else { if (S_ISDIR(inode->i_mode)) { - nfsi->cache_validity |= NFS_INO_INVALID_DATA; + cache_validity |= NFS_INO_INVALID_DATA; nfs_force_lookup_revalidate(inode); } else { if (!NFS_PROTO(inode)->have_delegation(inode, FMODE_READ)) - nfsi->cache_validity |= NFS_INO_REVAL_PAGECACHE; + cache_validity |= NFS_INO_REVAL_PAGECACHE; } if (cinfo->before != inode_peek_iversion_raw(inode)) - nfsi->cache_validity |= NFS_INO_INVALID_ACCESS | - NFS_INO_INVALID_ACL | - NFS_INO_INVALID_XATTR; + cache_validity |= NFS_INO_INVALID_ACCESS | + NFS_INO_INVALID_ACL | + NFS_INO_INVALID_XATTR; } inode_set_iversion_raw(inode, cinfo->after); nfsi->read_cache_jiffies = timestamp; nfsi->attr_gencount = nfs_inc_attr_generation_counter(); + nfs_set_cache_invalid(inode, cache_validity); nfsi->cache_validity &= ~NFS_INO_INVALID_CHANGE; - - if (nfsi->cache_validity & NFS_INO_INVALID_DATA) - nfs_fscache_invalidate(inode); } void @@ -2231,6 +2223,7 @@ static int nfs4_handle_delegation_recall_error(struct nfs_server *server, struct default: printk(KERN_ERR "NFS: %s: unhandled error " "%d.\n", __func__, err); + fallthrough; case 0: case -ENOENT: case -EAGAIN: @@ -5438,15 +5431,16 @@ static void nfs4_bitmask_adjust(__u32 *bitmask, struct inode *inode, if (cache_validity & NFS_INO_INVALID_ATIME) bitmask[1] |= FATTR4_WORD1_TIME_ACCESS; - if (cache_validity & NFS_INO_INVALID_ACCESS) - bitmask[0] |= FATTR4_WORD1_MODE | FATTR4_WORD1_OWNER | - FATTR4_WORD1_OWNER_GROUP; - if (cache_validity & NFS_INO_INVALID_ACL) - bitmask[0] |= FATTR4_WORD0_ACL; - if (cache_validity & NFS_INO_INVALID_LABEL) + if (cache_validity & NFS_INO_INVALID_OTHER) + bitmask[1] |= FATTR4_WORD1_MODE | FATTR4_WORD1_OWNER | + FATTR4_WORD1_OWNER_GROUP | + FATTR4_WORD1_NUMLINKS; + if (label && label->len && cache_validity & NFS_INO_INVALID_LABEL) bitmask[2] |= FATTR4_WORD2_SECURITY_LABEL; - if (cache_validity & NFS_INO_INVALID_CTIME) + if (cache_validity & NFS_INO_INVALID_CHANGE) bitmask[0] |= FATTR4_WORD0_CHANGE; + if (cache_validity & NFS_INO_INVALID_CTIME) + bitmask[1] |= FATTR4_WORD1_TIME_METADATA; if (cache_validity & NFS_INO_INVALID_MTIME) bitmask[1] |= FATTR4_WORD1_TIME_MODIFY; if (cache_validity & NFS_INO_INVALID_SIZE) @@ -5895,6 +5889,9 @@ static int __nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t bufl unsigned int npages = DIV_ROUND_UP(buflen, PAGE_SIZE); int ret, i; + /* You can't remove system.nfs4_acl: */ + if (buflen == 0) + return -EINVAL; if (!nfs4_server_supports_acls(server)) return -EOPNOTSUPP; if (npages > ARRAY_SIZE(pages)) @@ -5917,9 +5914,9 @@ static int __nfs4_proc_set_acl(struct inode *inode, const void *buf, size_t bufl * so mark the attribute cache invalid. */ spin_lock(&inode->i_lock); - NFS_I(inode)->cache_validity |= NFS_INO_INVALID_CHANGE - | NFS_INO_INVALID_CTIME - | NFS_INO_REVAL_FORCED; + nfs_set_cache_invalid(inode, NFS_INO_INVALID_CHANGE | + NFS_INO_INVALID_CTIME | + NFS_INO_REVAL_FORCED); spin_unlock(&inode->i_lock); nfs_access_zap_cache(inode); nfs_zap_acl_cache(inode); @@ -5971,7 +5968,7 @@ static int _nfs4_get_security_label(struct inode *inode, void *buf, return ret; if (!(fattr.valid & NFS_ATTR_FATTR_V4_SECURITY_LABEL)) return -ENOENT; - return 0; + return label.len; } static int nfs4_get_security_label(struct inode *inode, void *buf, @@ -7491,6 +7488,7 @@ nfs4_release_lockowner(struct nfs_server *server, struct nfs4_lock_state *lsp) #define XATTR_NAME_NFSV4_ACL "system.nfs4_acl" static int nfs4_xattr_set_nfs4_acl(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *unused, struct inode *inode, const char *key, const void *buf, size_t buflen, int flags) @@ -7513,6 +7511,7 @@ static bool nfs4_xattr_list_nfs4_acl(struct dentry *dentry) #ifdef CONFIG_NFS_V4_SECURITY_LABEL static int nfs4_xattr_set_nfs4_label(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *unused, struct inode *inode, const char *key, const void *buf, size_t buflen, int flags) @@ -7563,6 +7562,7 @@ nfs4_listxattr_nfs4_label(struct inode *inode, char *list, size_t list_len) #ifdef CONFIG_NFS_V4_2 static int nfs4_xattr_set_nfs4_user(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *unused, struct inode *inode, const char *key, const void *buf, size_t buflen, int flags) @@ -9705,6 +9705,7 @@ nfs4_layoutcommit_done(struct rpc_task *task, void *calldata) case -NFS4ERR_BADLAYOUT: /* no layout */ case -NFS4ERR_GRACE: /* loca_recalim always false */ task->tk_status = 0; + break; case 0: break; default: diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index 4bf10792cb5b..3a51351bdc6a 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -1125,6 +1125,7 @@ static void nfs_increment_seqid(int status, struct nfs_seqid *seqid) " sequence-id error on an" " unconfirmed sequence %p!\n", seqid->sequence); + return; case -NFS4ERR_STALE_CLIENTID: case -NFS4ERR_STALE_STATEID: case -NFS4ERR_BAD_STATEID: diff --git a/fs/nfs/pnfs.c b/fs/nfs/pnfs.c index af64b4e6fd1f..102b66e0bdef 100644 --- a/fs/nfs/pnfs.c +++ b/fs/nfs/pnfs.c @@ -2875,6 +2875,7 @@ pnfs_do_write(struct nfs_pageio_descriptor *desc, switch (trypnfs) { case PNFS_NOT_ATTEMPTED: pnfs_write_through_mds(desc, hdr); + break; case PNFS_ATTEMPTED: break; case PNFS_TRY_AGAIN: @@ -3019,6 +3020,7 @@ pnfs_do_read(struct nfs_pageio_descriptor *desc, struct nfs_pgio_header *hdr) switch (trypnfs) { case PNFS_NOT_ATTEMPTED: pnfs_read_through_mds(desc, hdr); + break; case PNFS_ATTEMPTED: break; case PNFS_TRY_AGAIN: diff --git a/fs/nfs/read.c b/fs/nfs/read.c index eb854f1f86e2..d2b6dce1f99f 100644 --- a/fs/nfs/read.c +++ b/fs/nfs/read.c @@ -74,6 +74,24 @@ void nfs_pageio_init_read(struct nfs_pageio_descriptor *pgio, } EXPORT_SYMBOL_GPL(nfs_pageio_init_read); +static void nfs_pageio_complete_read(struct nfs_pageio_descriptor *pgio, + struct inode *inode) +{ + struct nfs_pgio_mirror *pgm; + unsigned long npages; + + nfs_pageio_complete(pgio); + + /* It doesn't make sense to do mirrored reads! */ + WARN_ON_ONCE(pgio->pg_mirror_count != 1); + + pgm = &pgio->pg_mirrors[0]; + NFS_I(inode)->read_io += pgm->pg_bytes_written; + npages = (pgm->pg_bytes_written + PAGE_SIZE - 1) >> PAGE_SHIFT; + nfs_add_stats(inode, NFSIOS_READPAGES, npages); +} + + void nfs_pageio_reset_read_mds(struct nfs_pageio_descriptor *pgio) { struct nfs_pgio_mirror *mirror; @@ -114,41 +132,10 @@ static void nfs_readpage_release(struct nfs_page *req, int error) nfs_release_request(req); } -int nfs_readpage_async(struct nfs_open_context *ctx, struct inode *inode, - struct page *page) -{ - struct nfs_page *new; - unsigned int len; +struct nfs_readdesc { struct nfs_pageio_descriptor pgio; - struct nfs_pgio_mirror *pgm; - - len = nfs_page_length(page); - if (len == 0) - return nfs_return_empty_page(page); - new = nfs_create_request(ctx, page, 0, len); - if (IS_ERR(new)) { - unlock_page(page); - return PTR_ERR(new); - } - if (len < PAGE_SIZE) - zero_user_segment(page, len, PAGE_SIZE); - - nfs_pageio_init_read(&pgio, inode, false, - &nfs_async_read_completion_ops); - if (!nfs_pageio_add_request(&pgio, new)) { - nfs_list_remove_request(new); - nfs_readpage_release(new, pgio.pg_error); - } - nfs_pageio_complete(&pgio); - - /* It doesn't make sense to do mirrored reads! */ - WARN_ON_ONCE(pgio.pg_mirror_count != 1); - - pgm = &pgio.pg_mirrors[0]; - NFS_I(inode)->read_io += pgm->pg_bytes_written; - - return pgio.pg_error < 0 ? pgio.pg_error : 0; -} + struct nfs_open_context *ctx; +}; static void nfs_page_group_set_uptodate(struct nfs_page *req) { @@ -171,8 +158,7 @@ static void nfs_read_completion(struct nfs_pgio_header *hdr) if (test_bit(NFS_IOHDR_EOF, &hdr->flags)) { /* note: regions of the page not covered by a - * request are zeroed in nfs_readpage_async / - * readpage_async_filler */ + * request are zeroed in readpage_async_filler */ if (bytes > hdr->good_bytes) { /* nothing in this request was good, so zero * the full extent of the request */ @@ -304,6 +290,38 @@ static void nfs_readpage_result(struct rpc_task *task, nfs_readpage_retry(task, hdr); } +static int +readpage_async_filler(void *data, struct page *page) +{ + struct nfs_readdesc *desc = data; + struct nfs_page *new; + unsigned int len; + int error; + + len = nfs_page_length(page); + if (len == 0) + return nfs_return_empty_page(page); + + new = nfs_create_request(desc->ctx, page, 0, len); + if (IS_ERR(new)) + goto out_error; + + if (len < PAGE_SIZE) + zero_user_segment(page, len, PAGE_SIZE); + if (!nfs_pageio_add_request(&desc->pgio, new)) { + nfs_list_remove_request(new); + error = desc->pgio.pg_error; + nfs_readpage_release(new, error); + goto out; + } + return 0; +out_error: + error = PTR_ERR(new); + unlock_page(page); +out: + return error; +} + /* * Read a page over NFS. * We read the page synchronously in the following case: @@ -312,14 +330,13 @@ static void nfs_readpage_result(struct rpc_task *task, */ int nfs_readpage(struct file *file, struct page *page) { - struct nfs_open_context *ctx; + struct nfs_readdesc desc; struct inode *inode = page_file_mapping(page)->host; - int error; + int ret; dprintk("NFS: nfs_readpage (%p %ld@%lu)\n", page, PAGE_SIZE, page_index(page)); nfs_inc_stats(inode, NFSIOS_VFSREADPAGE); - nfs_add_stats(inode, NFSIOS_READPAGES, 1); /* * Try to flush any pending writes to the file.. @@ -328,93 +345,59 @@ int nfs_readpage(struct file *file, struct page *page) * be any new pending writes generated at this point * for this page (other pages can be written to). */ - error = nfs_wb_page(inode, page); - if (error) + ret = nfs_wb_page(inode, page); + if (ret) goto out_unlock; if (PageUptodate(page)) goto out_unlock; - error = -ESTALE; + ret = -ESTALE; if (NFS_STALE(inode)) goto out_unlock; if (file == NULL) { - error = -EBADF; - ctx = nfs_find_open_context(inode, NULL, FMODE_READ); - if (ctx == NULL) + ret = -EBADF; + desc.ctx = nfs_find_open_context(inode, NULL, FMODE_READ); + if (desc.ctx == NULL) goto out_unlock; } else - ctx = get_nfs_open_context(nfs_file_open_context(file)); + desc.ctx = get_nfs_open_context(nfs_file_open_context(file)); if (!IS_SYNC(inode)) { - error = nfs_readpage_from_fscache(ctx, inode, page); - if (error == 0) + ret = nfs_readpage_from_fscache(desc.ctx, inode, page); + if (ret == 0) goto out; } - xchg(&ctx->error, 0); - error = nfs_readpage_async(ctx, inode, page); - if (!error) { - error = wait_on_page_locked_killable(page); - if (!PageUptodate(page) && !error) - error = xchg(&ctx->error, 0); - } -out: - put_nfs_open_context(ctx); - return error; -out_unlock: - unlock_page(page); - return error; -} - -struct nfs_readdesc { - struct nfs_pageio_descriptor *pgio; - struct nfs_open_context *ctx; -}; - -static int -readpage_async_filler(void *data, struct page *page) -{ - struct nfs_readdesc *desc = (struct nfs_readdesc *)data; - struct nfs_page *new; - unsigned int len; - int error; + xchg(&desc.ctx->error, 0); + nfs_pageio_init_read(&desc.pgio, inode, false, + &nfs_async_read_completion_ops); - len = nfs_page_length(page); - if (len == 0) - return nfs_return_empty_page(page); + ret = readpage_async_filler(&desc, page); - new = nfs_create_request(desc->ctx, page, 0, len); - if (IS_ERR(new)) - goto out_error; + if (!ret) + nfs_pageio_complete_read(&desc.pgio, inode); - if (len < PAGE_SIZE) - zero_user_segment(page, len, PAGE_SIZE); - if (!nfs_pageio_add_request(desc->pgio, new)) { - nfs_list_remove_request(new); - error = desc->pgio->pg_error; - nfs_readpage_release(new, error); - goto out; + ret = desc.pgio.pg_error < 0 ? desc.pgio.pg_error : 0; + if (!ret) { + ret = wait_on_page_locked_killable(page); + if (!PageUptodate(page) && !ret) + ret = xchg(&desc.ctx->error, 0); } - return 0; -out_error: - error = PTR_ERR(new); - unlock_page(page); out: - return error; + put_nfs_open_context(desc.ctx); + return ret; +out_unlock: + unlock_page(page); + return ret; } -int nfs_readpages(struct file *filp, struct address_space *mapping, +int nfs_readpages(struct file *file, struct address_space *mapping, struct list_head *pages, unsigned nr_pages) { - struct nfs_pageio_descriptor pgio; - struct nfs_pgio_mirror *pgm; - struct nfs_readdesc desc = { - .pgio = &pgio, - }; + struct nfs_readdesc desc; struct inode *inode = mapping->host; - unsigned long npages; - int ret = -ESTALE; + int ret; dprintk("NFS: nfs_readpages (%s/%Lu %d)\n", inode->i_sb->s_id, @@ -422,15 +405,17 @@ int nfs_readpages(struct file *filp, struct address_space *mapping, nr_pages); nfs_inc_stats(inode, NFSIOS_VFSREADPAGES); + ret = -ESTALE; if (NFS_STALE(inode)) goto out; - if (filp == NULL) { + if (file == NULL) { + ret = -EBADF; desc.ctx = nfs_find_open_context(inode, NULL, FMODE_READ); if (desc.ctx == NULL) - return -EBADF; + goto out; } else - desc.ctx = get_nfs_open_context(nfs_file_open_context(filp)); + desc.ctx = get_nfs_open_context(nfs_file_open_context(file)); /* attempt to read as many of the pages as possible from the cache * - this returns -ENOBUFS immediately if the cookie is negative @@ -440,20 +425,13 @@ int nfs_readpages(struct file *filp, struct address_space *mapping, if (ret == 0) goto read_complete; /* all pages were read */ - nfs_pageio_init_read(&pgio, inode, false, + nfs_pageio_init_read(&desc.pgio, inode, false, &nfs_async_read_completion_ops); ret = read_cache_pages(mapping, pages, readpage_async_filler, &desc); - nfs_pageio_complete(&pgio); - /* It doesn't make sense to do mirrored reads! */ - WARN_ON_ONCE(pgio.pg_mirror_count != 1); + nfs_pageio_complete_read(&desc.pgio, inode); - pgm = &pgio.pg_mirrors[0]; - NFS_I(inode)->read_io += pgm->pg_bytes_written; - npages = (pgm->pg_bytes_written + PAGE_SIZE - 1) >> - PAGE_SHIFT; - nfs_add_stats(inode, NFSIOS_READPAGES, npages); read_complete: put_nfs_open_context(desc.ctx); out: diff --git a/fs/nfs/super.c b/fs/nfs/super.c index 4034102010f0..94885c6f8f54 100644 --- a/fs/nfs/super.c +++ b/fs/nfs/super.c @@ -86,9 +86,11 @@ const struct super_operations nfs_sops = { }; EXPORT_SYMBOL_GPL(nfs_sops); +#ifdef CONFIG_NFS_V4_2 static const struct nfs_ssc_client_ops nfs_ssc_clnt_ops_tbl = { .sco_sb_deactive = nfs_sb_deactive, }; +#endif #if IS_ENABLED(CONFIG_NFS_V4) static int __init register_nfs4_fs(void) @@ -111,15 +113,21 @@ static void unregister_nfs4_fs(void) } #endif +#ifdef CONFIG_NFS_V4_2 static void nfs_ssc_register_ops(void) { +#ifdef CONFIG_NFSD_V4 nfs_ssc_register(&nfs_ssc_clnt_ops_tbl); +#endif } static void nfs_ssc_unregister_ops(void) { +#ifdef CONFIG_NFSD_V4 nfs_ssc_unregister(&nfs_ssc_clnt_ops_tbl); +#endif } +#endif /* CONFIG_NFS_V4_2 */ static struct shrinker acl_shrinker = { .count_objects = nfs_access_cache_count, @@ -148,7 +156,9 @@ int __init register_nfs_fs(void) ret = register_shrinker(&acl_shrinker); if (ret < 0) goto error_3; +#ifdef CONFIG_NFS_V4_2 nfs_ssc_register_ops(); +#endif return 0; error_3: nfs_unregister_sysctl(); @@ -168,7 +178,9 @@ void __exit unregister_nfs_fs(void) unregister_shrinker(&acl_shrinker); nfs_unregister_sysctl(); unregister_nfs4_fs(); +#ifdef CONFIG_NFS_V4_2 nfs_ssc_unregister_ops(); +#endif unregister_filesystem(&nfs_fs_type); } @@ -511,6 +523,13 @@ static void nfs_show_mount_options(struct seq_file *m, struct nfs_server *nfss, seq_puts(m, ",local_lock=flock"); else seq_puts(m, ",local_lock=posix"); + + if (nfss->flags & NFS_MOUNT_WRITE_EAGER) { + if (nfss->flags & NFS_MOUNT_WRITE_WAIT) + seq_puts(m, ",write=wait"); + else + seq_puts(m, ",write=eager"); + } } /* diff --git a/fs/nfs/unlink.c b/fs/nfs/unlink.c index b27ebdccef70..5fa11e1aca4c 100644 --- a/fs/nfs/unlink.c +++ b/fs/nfs/unlink.c @@ -500,9 +500,9 @@ nfs_sillyrename(struct inode *dir, struct dentry *dentry) nfs_set_verifier(dentry, nfs_save_change_attribute(dir)); spin_lock(&inode->i_lock); NFS_I(inode)->attr_gencount = nfs_inc_attr_generation_counter(); - NFS_I(inode)->cache_validity |= NFS_INO_INVALID_CHANGE - | NFS_INO_INVALID_CTIME - | NFS_INO_REVAL_FORCED; + nfs_set_cache_invalid(inode, NFS_INO_INVALID_CHANGE | + NFS_INO_INVALID_CTIME | + NFS_INO_REVAL_FORCED); spin_unlock(&inode->i_lock); d_move(dentry, sdentry); break; diff --git a/fs/nfs/write.c b/fs/nfs/write.c index 639c34fec04a..f05a90338a76 100644 --- a/fs/nfs/write.c +++ b/fs/nfs/write.c @@ -303,9 +303,9 @@ static void nfs_set_pageerror(struct address_space *mapping) nfs_zap_mapping(mapping->host, mapping); /* Force file size revalidation */ spin_lock(&inode->i_lock); - NFS_I(inode)->cache_validity |= NFS_INO_REVAL_FORCED | - NFS_INO_REVAL_PAGECACHE | - NFS_INO_INVALID_SIZE; + nfs_set_cache_invalid(inode, NFS_INO_REVAL_FORCED | + NFS_INO_REVAL_PAGECACHE | + NFS_INO_INVALID_SIZE); spin_unlock(&inode->i_lock); } @@ -712,16 +712,23 @@ int nfs_writepages(struct address_space *mapping, struct writeback_control *wbc) { struct inode *inode = mapping->host; struct nfs_pageio_descriptor pgio; - struct nfs_io_completion *ioc; + struct nfs_io_completion *ioc = NULL; + unsigned int mntflags = NFS_SERVER(inode)->flags; + int priority = 0; int err; nfs_inc_stats(inode, NFSIOS_VFSWRITEPAGES); - ioc = nfs_io_completion_alloc(GFP_KERNEL); - if (ioc) - nfs_io_completion_init(ioc, nfs_io_completion_commit, inode); + if (!(mntflags & NFS_MOUNT_WRITE_EAGER) || wbc->for_kupdate || + wbc->for_background || wbc->for_sync || wbc->for_reclaim) { + ioc = nfs_io_completion_alloc(GFP_KERNEL); + if (ioc) + nfs_io_completion_init(ioc, nfs_io_completion_commit, + inode); + priority = wb_priority(wbc); + } - nfs_pageio_init_write(&pgio, inode, wb_priority(wbc), false, + nfs_pageio_init_write(&pgio, inode, priority, false, &nfs_async_write_completion_ops); pgio.pg_io_completion = ioc; err = write_cache_pages(mapping, wbc, nfs_writepages_callback, &pgio); @@ -1278,19 +1285,21 @@ bool nfs_ctx_key_to_expire(struct nfs_open_context *ctx, struct inode *inode) * the PageUptodate() flag. In this case, we will need to turn off * write optimisations that depend on the page contents being correct. */ -static bool nfs_write_pageuptodate(struct page *page, struct inode *inode) +static bool nfs_write_pageuptodate(struct page *page, struct inode *inode, + unsigned int pagelen) { struct nfs_inode *nfsi = NFS_I(inode); if (nfs_have_delegated_attributes(inode)) goto out; - if (nfsi->cache_validity & NFS_INO_REVAL_PAGECACHE) + if (nfsi->cache_validity & + (NFS_INO_REVAL_PAGECACHE | NFS_INO_INVALID_SIZE)) return false; smp_rmb(); - if (test_bit(NFS_INO_INVALIDATING, &nfsi->flags)) + if (test_bit(NFS_INO_INVALIDATING, &nfsi->flags) && pagelen != 0) return false; out: - if (nfsi->cache_validity & NFS_INO_INVALID_DATA) + if (nfsi->cache_validity & NFS_INO_INVALID_DATA && pagelen != 0) return false; return PageUptodate(page) != 0; } @@ -1310,7 +1319,8 @@ is_whole_file_wrlock(struct file_lock *fl) * If the file is opened for synchronous writes then we can just skip the rest * of the checks. */ -static int nfs_can_extend_write(struct file *file, struct page *page, struct inode *inode) +static int nfs_can_extend_write(struct file *file, struct page *page, + struct inode *inode, unsigned int pagelen) { int ret; struct file_lock_context *flctx = inode->i_flctx; @@ -1318,7 +1328,7 @@ static int nfs_can_extend_write(struct file *file, struct page *page, struct ino if (file->f_flags & O_DSYNC) return 0; - if (!nfs_write_pageuptodate(page, inode)) + if (!nfs_write_pageuptodate(page, inode, pagelen)) return 0; if (NFS_PROTO(inode)->have_delegation(inode, FMODE_WRITE)) return 1; @@ -1356,6 +1366,7 @@ int nfs_updatepage(struct file *file, struct page *page, struct nfs_open_context *ctx = nfs_file_open_context(file); struct address_space *mapping = page_file_mapping(page); struct inode *inode = mapping->host; + unsigned int pagelen = nfs_page_length(page); int status = 0; nfs_inc_stats(inode, NFSIOS_VFSUPDATEPAGE); @@ -1366,8 +1377,8 @@ int nfs_updatepage(struct file *file, struct page *page, if (!count) goto out; - if (nfs_can_extend_write(file, page, inode)) { - count = max(count + offset, nfs_page_length(page)); + if (nfs_can_extend_write(file, page, inode, pagelen)) { + count = max(count + offset, pagelen); offset = 0; } @@ -1593,7 +1604,7 @@ static int nfs_writeback_done(struct rpc_task *task, /* Deal with the suid/sgid bit corner case */ if (nfs_should_remove_suid(inode)) { spin_lock(&inode->i_lock); - NFS_I(inode)->cache_validity |= NFS_INO_INVALID_OTHER; + nfs_set_cache_invalid(inode, NFS_INO_INVALID_OTHER); spin_unlock(&inode->i_lock); } return 0; diff --git a/fs/nfs_common/Makefile b/fs/nfs_common/Makefile index fa82f5aaa6d9..119c75ab9fd0 100644 --- a/fs/nfs_common/Makefile +++ b/fs/nfs_common/Makefile @@ -7,4 +7,4 @@ obj-$(CONFIG_NFS_ACL_SUPPORT) += nfs_acl.o nfs_acl-objs := nfsacl.o obj-$(CONFIG_GRACE_PERIOD) += grace.o -obj-$(CONFIG_GRACE_PERIOD) += nfs_ssc.o +obj-$(CONFIG_NFS_V4_2_SSC_HELPER) += nfs_ssc.o diff --git a/fs/nfs_common/nfs_ssc.c b/fs/nfs_common/nfs_ssc.c index f43bbb373913..7c1509e968c8 100644 --- a/fs/nfs_common/nfs_ssc.c +++ b/fs/nfs_common/nfs_ssc.c @@ -1,7 +1,5 @@ // SPDX-License-Identifier: GPL-2.0-only /* - * fs/nfs_common/nfs_ssc_comm.c - * * Helper for knfsd's SSC to access ops in NFS client modules * * Author: Dai Ngo <dai.ngo@oracle.com> diff --git a/fs/nfs_common/nfsacl.c b/fs/nfs_common/nfsacl.c index d056ad2fdefd..79c563c1a5e8 100644 --- a/fs/nfs_common/nfsacl.c +++ b/fs/nfs_common/nfsacl.c @@ -295,3 +295,55 @@ int nfsacl_decode(struct xdr_buf *buf, unsigned int base, unsigned int *aclcnt, nfsacl_desc.desc.array_len; } EXPORT_SYMBOL_GPL(nfsacl_decode); + +/** + * nfs_stream_decode_acl - Decode an NFSv3 ACL + * + * @xdr: an xdr_stream positioned at an encoded ACL + * @aclcnt: OUT: count of ACEs in decoded posix_acl + * @pacl: OUT: a dynamically-allocated buffer containing the decoded posix_acl + * + * Return values: + * %false: The encoded ACL is not valid + * %true: @pacl contains a decoded ACL, and @xdr is advanced + * + * On a successful return, caller must release *pacl using posix_acl_release(). + */ +bool nfs_stream_decode_acl(struct xdr_stream *xdr, unsigned int *aclcnt, + struct posix_acl **pacl) +{ + const size_t elem_size = XDR_UNIT * 3; + struct nfsacl_decode_desc nfsacl_desc = { + .desc = { + .elem_size = elem_size, + .xcode = pacl ? xdr_nfsace_decode : NULL, + }, + }; + unsigned int base; + u32 entries; + + if (xdr_stream_decode_u32(xdr, &entries) < 0) + return false; + if (entries > NFS_ACL_MAX_ENTRIES) + return false; + + base = xdr_stream_pos(xdr); + if (!xdr_inline_decode(xdr, XDR_UNIT + elem_size * entries)) + return false; + nfsacl_desc.desc.array_maxlen = entries; + if (xdr_decode_array2(xdr->buf, base, &nfsacl_desc.desc)) + return false; + + if (pacl) { + if (entries != nfsacl_desc.desc.array_len || + posix_acl_from_nfsacl(nfsacl_desc.acl) != 0) { + posix_acl_release(nfsacl_desc.acl); + return false; + } + *pacl = nfsacl_desc.acl; + } + if (aclcnt) + *aclcnt = entries; + return true; +} +EXPORT_SYMBOL_GPL(nfs_stream_decode_acl); diff --git a/fs/nfsd/Kconfig b/fs/nfsd/Kconfig index dbbc583d6273..d6cff5fbe705 100644 --- a/fs/nfsd/Kconfig +++ b/fs/nfsd/Kconfig @@ -73,9 +73,11 @@ config NFSD_V4 select NFSD_V3 select FS_POSIX_ACL select SUNRPC_GSS + select CRYPTO select CRYPTO_MD5 select CRYPTO_SHA256 select GRACE_PERIOD + select NFS_V4_2_SSC_HELPER if NFS_V4_2 help This option enables support in your system's NFS server for version 4 of the NFS protocol (RFC 3530). diff --git a/fs/nfsd/blocklayout.c b/fs/nfsd/blocklayout.c index a07c39c94bbd..1058659a8d31 100644 --- a/fs/nfsd/blocklayout.c +++ b/fs/nfsd/blocklayout.c @@ -254,7 +254,7 @@ again: req->cmd[4] = bufflen & 0xff; req->cmd_len = COMMAND_SIZE(INQUIRY); - blk_execute_rq(rq->q, NULL, rq, 1); + blk_execute_rq(NULL, rq, 1); if (req->result) { pr_err("pNFS: INQUIRY 0x83 failed with: %x\n", req->result); diff --git a/fs/nfsd/export.c b/fs/nfsd/export.c index 81e7bb12aca6..9421dae22737 100644 --- a/fs/nfsd/export.c +++ b/fs/nfsd/export.c @@ -331,12 +331,29 @@ static void nfsd4_fslocs_free(struct nfsd4_fs_locations *fsloc) fsloc->locations = NULL; } +static int export_stats_init(struct export_stats *stats) +{ + stats->start_time = ktime_get_seconds(); + return nfsd_percpu_counters_init(stats->counter, EXP_STATS_COUNTERS_NUM); +} + +static void export_stats_reset(struct export_stats *stats) +{ + nfsd_percpu_counters_reset(stats->counter, EXP_STATS_COUNTERS_NUM); +} + +static void export_stats_destroy(struct export_stats *stats) +{ + nfsd_percpu_counters_destroy(stats->counter, EXP_STATS_COUNTERS_NUM); +} + static void svc_export_put(struct kref *ref) { struct svc_export *exp = container_of(ref, struct svc_export, h.ref); path_put(&exp->ex_path); auth_domain_put(exp->ex_client); nfsd4_fslocs_free(&exp->ex_fslocs); + export_stats_destroy(&exp->ex_stats); kfree(exp->ex_uuid); kfree_rcu(exp, ex_rcu); } @@ -369,8 +386,9 @@ static struct svc_export *svc_export_update(struct svc_export *new, struct svc_export *old); static struct svc_export *svc_export_lookup(struct svc_export *); -static int check_export(struct inode *inode, int *flags, unsigned char *uuid) +static int check_export(struct path *path, int *flags, unsigned char *uuid) { + struct inode *inode = d_inode(path->dentry); /* * We currently export only dirs, regular files, and (for v4 @@ -394,6 +412,7 @@ static int check_export(struct inode *inode, int *flags, unsigned char *uuid) * or an FSID number (so NFSEXP_FSID or ->uuid is needed). * 2: We must be able to find an inode from a filehandle. * This means that s_export_op must be set. + * 3: We must not currently be on an idmapped mount. */ if (!(inode->i_sb->s_type->fs_flags & FS_REQUIRES_DEV) && !(*flags & NFSEXP_FSID) && @@ -408,6 +427,11 @@ static int check_export(struct inode *inode, int *flags, unsigned char *uuid) return -EINVAL; } + if (mnt_user_ns(path->mnt) != &init_user_ns) { + dprintk("exp_export: export of idmapped mounts not yet supported.\n"); + return -EINVAL; + } + if (inode->i_sb->s_export_op->flags & EXPORT_OP_NOSUBTREECHK && !(*flags & NFSEXP_NOSUBTREECHECK)) { dprintk("%s: %s does not support subtree checking!\n", @@ -636,8 +660,7 @@ static int svc_export_parse(struct cache_detail *cd, char *mesg, int mlen) goto out4; } - err = check_export(d_inode(exp.ex_path.dentry), &exp.ex_flags, - exp.ex_uuid); + err = check_export(&exp.ex_path, &exp.ex_flags, exp.ex_uuid); if (err) goto out4; /* @@ -692,22 +715,47 @@ static void exp_flags(struct seq_file *m, int flag, int fsid, kuid_t anonu, kgid_t anong, struct nfsd4_fs_locations *fslocs); static void show_secinfo(struct seq_file *m, struct svc_export *exp); +static int is_export_stats_file(struct seq_file *m) +{ + /* + * The export_stats file uses the same ops as the exports file. + * We use the file's name to determine the reported info per export. + * There is no rename in nsfdfs, so d_name.name is stable. + */ + return !strcmp(m->file->f_path.dentry->d_name.name, "export_stats"); +} + static int svc_export_show(struct seq_file *m, struct cache_detail *cd, struct cache_head *h) { - struct svc_export *exp ; + struct svc_export *exp; + bool export_stats = is_export_stats_file(m); - if (h ==NULL) { - seq_puts(m, "#path domain(flags)\n"); + if (h == NULL) { + if (export_stats) + seq_puts(m, "#path domain start-time\n#\tstats\n"); + else + seq_puts(m, "#path domain(flags)\n"); return 0; } exp = container_of(h, struct svc_export, h); seq_path(m, &exp->ex_path, " \t\n\\"); seq_putc(m, '\t'); seq_escape(m, exp->ex_client->name, " \t\n\\"); + if (export_stats) { + seq_printf(m, "\t%lld\n", exp->ex_stats.start_time); + seq_printf(m, "\tfh_stale: %lld\n", + percpu_counter_sum_positive(&exp->ex_stats.counter[EXP_STATS_FH_STALE])); + seq_printf(m, "\tio_read: %lld\n", + percpu_counter_sum_positive(&exp->ex_stats.counter[EXP_STATS_IO_READ])); + seq_printf(m, "\tio_write: %lld\n", + percpu_counter_sum_positive(&exp->ex_stats.counter[EXP_STATS_IO_WRITE])); + seq_putc(m, '\n'); + return 0; + } seq_putc(m, '('); - if (test_bit(CACHE_VALID, &h->flags) && + if (test_bit(CACHE_VALID, &h->flags) && !test_bit(CACHE_NEGATIVE, &h->flags)) { exp_flags(m, exp->ex_flags, exp->ex_fsid, exp->ex_anon_uid, exp->ex_anon_gid, &exp->ex_fslocs); @@ -748,6 +796,7 @@ static void svc_export_init(struct cache_head *cnew, struct cache_head *citem) new->ex_layout_types = 0; new->ex_uuid = NULL; new->cd = item->cd; + export_stats_reset(&new->ex_stats); } static void export_update(struct cache_head *cnew, struct cache_head *citem) @@ -780,10 +829,15 @@ static void export_update(struct cache_head *cnew, struct cache_head *citem) static struct cache_head *svc_export_alloc(void) { struct svc_export *i = kmalloc(sizeof(*i), GFP_KERNEL); - if (i) - return &i->h; - else + if (!i) + return NULL; + + if (export_stats_init(&i->ex_stats)) { + kfree(i); return NULL; + } + + return &i->h; } static const struct cache_detail svc_export_cache_template = { @@ -1245,10 +1299,14 @@ static int e_show(struct seq_file *m, void *p) struct cache_head *cp = p; struct svc_export *exp = container_of(cp, struct svc_export, h); struct cache_detail *cd = m->private; + bool export_stats = is_export_stats_file(m); if (p == SEQ_START_TOKEN) { seq_puts(m, "# Version 1.1\n"); - seq_puts(m, "# Path Client(Flags) # IPs\n"); + if (export_stats) + seq_puts(m, "# Path Client Start-time\n#\tStats\n"); + else + seq_puts(m, "# Path Client(Flags) # IPs\n"); return 0; } diff --git a/fs/nfsd/export.h b/fs/nfsd/export.h index e7daa1f246f0..ee0e3aba4a6e 100644 --- a/fs/nfsd/export.h +++ b/fs/nfsd/export.h @@ -6,6 +6,7 @@ #define NFSD_EXPORT_H #include <linux/sunrpc/cache.h> +#include <linux/percpu_counter.h> #include <uapi/linux/nfsd/export.h> #include <linux/nfs4.h> @@ -46,6 +47,19 @@ struct exp_flavor_info { u32 flags; }; +/* Per-export stats */ +enum { + EXP_STATS_FH_STALE, + EXP_STATS_IO_READ, + EXP_STATS_IO_WRITE, + EXP_STATS_COUNTERS_NUM +}; + +struct export_stats { + time64_t start_time; + struct percpu_counter counter[EXP_STATS_COUNTERS_NUM]; +}; + struct svc_export { struct cache_head h; struct auth_domain * ex_client; @@ -62,6 +76,7 @@ struct svc_export { struct nfsd4_deviceid_map *ex_devid_map; struct cache_detail *cd; struct rcu_head ex_rcu; + struct export_stats ex_stats; }; /* an "export key" (expkey) maps a filehandlefragement to an diff --git a/fs/nfsd/filecache.c b/fs/nfsd/filecache.c index 53fcbf79bdca..7629248fdd53 100644 --- a/fs/nfsd/filecache.c +++ b/fs/nfsd/filecache.c @@ -898,6 +898,8 @@ nfsd_file_find_locked(struct inode *inode, unsigned int may_flags, continue; if (!nfsd_match_cred(nf->nf_cred, current_cred())) continue; + if (!test_bit(NFSD_FILE_HASHED, &nf->nf_flags)) + continue; if (nfsd_file_get(nf) != NULL) return nf; } diff --git a/fs/nfsd/netns.h b/fs/nfsd/netns.h index 7346acda9d76..c330f5bd0cf3 100644 --- a/fs/nfsd/netns.h +++ b/fs/nfsd/netns.h @@ -10,6 +10,7 @@ #include <net/net_namespace.h> #include <net/netns/generic.h> +#include <linux/percpu_counter.h> /* Hash tables for nfs4_clientid state */ #define CLIENT_HASH_BITS 4 @@ -21,6 +22,14 @@ struct cld_net; struct nfsd4_client_tracking_ops; +enum { + /* cache misses due only to checksum comparison failures */ + NFSD_NET_PAYLOAD_MISSES, + /* amount of memory (in bytes) currently consumed by the DRC */ + NFSD_NET_DRC_MEM_USAGE, + NFSD_NET_COUNTERS_NUM +}; + /* * Represents a nfsd "container". With respect to nfsv4 state tracking, the * fields of interest are the *_id_hashtbls and the *_name_tree. These track @@ -149,20 +158,16 @@ struct nfsd_net { /* * Stats and other tracking of on the duplicate reply cache. - * These fields and the "rc" fields in nfsdstats are modified - * with only the per-bucket cache lock, which isn't really safe - * and should be fixed if we want the statistics to be - * completely accurate. + * The longest_chain* fields are modified with only the per-bucket + * cache lock, which isn't really safe and should be fixed if we want + * these statistics to be completely accurate. */ /* total number of entries */ atomic_t num_drc_entries; - /* cache misses due only to checksum comparison failures */ - unsigned int payload_misses; - - /* amount of memory (in bytes) currently consumed by the DRC */ - unsigned int drc_mem_usage; + /* Per-netns stats counters */ + struct percpu_counter counter[NFSD_NET_COUNTERS_NUM]; /* longest hash chain seen */ unsigned int longest_chain; diff --git a/fs/nfsd/nfs2acl.c b/fs/nfsd/nfs2acl.c index b0f66604532a..855e17772eba 100644 --- a/fs/nfsd/nfs2acl.c +++ b/fs/nfsd/nfs2acl.c @@ -113,10 +113,12 @@ static __be32 nfsacld_proc_setacl(struct svc_rqst *rqstp) fh_lock(fh); - error = set_posix_acl(inode, ACL_TYPE_ACCESS, argp->acl_access); + error = set_posix_acl(&init_user_ns, inode, ACL_TYPE_ACCESS, + argp->acl_access); if (error) goto out_drop_lock; - error = set_posix_acl(inode, ACL_TYPE_DEFAULT, argp->acl_default); + error = set_posix_acl(&init_user_ns, inode, ACL_TYPE_DEFAULT, + argp->acl_default); if (error) goto out_drop_lock; @@ -188,63 +190,49 @@ out: static int nfsaclsvc_decode_getaclargs(struct svc_rqst *rqstp, __be32 *p) { + struct xdr_stream *xdr = &rqstp->rq_arg_stream; struct nfsd3_getaclargs *argp = rqstp->rq_argp; - p = nfs2svc_decode_fh(p, &argp->fh); - if (!p) + if (!svcxdr_decode_fhandle(xdr, &argp->fh)) + return 0; + if (xdr_stream_decode_u32(xdr, &argp->mask) < 0) return 0; - argp->mask = ntohl(*p); p++; - return xdr_argsize_check(rqstp, p); + return 1; } - static int nfsaclsvc_decode_setaclargs(struct svc_rqst *rqstp, __be32 *p) { + struct xdr_stream *xdr = &rqstp->rq_arg_stream; struct nfsd3_setaclargs *argp = rqstp->rq_argp; - struct kvec *head = rqstp->rq_arg.head; - unsigned int base; - int n; - p = nfs2svc_decode_fh(p, &argp->fh); - if (!p) + if (!svcxdr_decode_fhandle(xdr, &argp->fh)) return 0; - argp->mask = ntohl(*p++); - if (argp->mask & ~NFS_ACL_MASK || - !xdr_argsize_check(rqstp, p)) + if (xdr_stream_decode_u32(xdr, &argp->mask) < 0) return 0; - - base = (char *)p - (char *)head->iov_base; - n = nfsacl_decode(&rqstp->rq_arg, base, NULL, - (argp->mask & NFS_ACL) ? - &argp->acl_access : NULL); - if (n > 0) - n = nfsacl_decode(&rqstp->rq_arg, base + n, NULL, - (argp->mask & NFS_DFACL) ? - &argp->acl_default : NULL); - return (n > 0); -} - -static int nfsaclsvc_decode_fhandleargs(struct svc_rqst *rqstp, __be32 *p) -{ - struct nfsd_fhandle *argp = rqstp->rq_argp; - - p = nfs2svc_decode_fh(p, &argp->fh); - if (!p) + if (argp->mask & ~NFS_ACL_MASK) + return 0; + if (!nfs_stream_decode_acl(xdr, NULL, (argp->mask & NFS_ACL) ? + &argp->acl_access : NULL)) return 0; - return xdr_argsize_check(rqstp, p); + if (!nfs_stream_decode_acl(xdr, NULL, (argp->mask & NFS_DFACL) ? + &argp->acl_default : NULL)) + return 0; + + return 1; } static int nfsaclsvc_decode_accessargs(struct svc_rqst *rqstp, __be32 *p) { - struct nfsd3_accessargs *argp = rqstp->rq_argp; + struct xdr_stream *xdr = &rqstp->rq_arg_stream; + struct nfsd3_accessargs *args = rqstp->rq_argp; - p = nfs2svc_decode_fh(p, &argp->fh); - if (!p) + if (!svcxdr_decode_fhandle(xdr, &args->fh)) + return 0; + if (xdr_stream_decode_u32(xdr, &args->access) < 0) return 0; - argp->access = ntohl(*p++); - return xdr_argsize_check(rqstp, p); + return 1; } /* @@ -371,6 +359,7 @@ static const struct svc_procedure nfsd_acl_procedures2[5] = { .pc_ressize = sizeof(struct nfsd_voidres), .pc_cachetype = RC_NOCACHE, .pc_xdrressize = ST, + .pc_name = "NULL", }, [ACLPROC2_GETACL] = { .pc_func = nfsacld_proc_getacl, @@ -381,6 +370,7 @@ static const struct svc_procedure nfsd_acl_procedures2[5] = { .pc_ressize = sizeof(struct nfsd3_getaclres), .pc_cachetype = RC_NOCACHE, .pc_xdrressize = ST+1+2*(1+ACL), + .pc_name = "GETACL", }, [ACLPROC2_SETACL] = { .pc_func = nfsacld_proc_setacl, @@ -391,16 +381,18 @@ static const struct svc_procedure nfsd_acl_procedures2[5] = { .pc_ressize = sizeof(struct nfsd_attrstat), .pc_cachetype = RC_NOCACHE, .pc_xdrressize = ST+AT, + .pc_name = "SETACL", }, [ACLPROC2_GETATTR] = { .pc_func = nfsacld_proc_getattr, - .pc_decode = nfsaclsvc_decode_fhandleargs, + .pc_decode = nfssvc_decode_fhandleargs, .pc_encode = nfsaclsvc_encode_attrstatres, .pc_release = nfsaclsvc_release_attrstat, .pc_argsize = sizeof(struct nfsd_fhandle), .pc_ressize = sizeof(struct nfsd_attrstat), .pc_cachetype = RC_NOCACHE, .pc_xdrressize = ST+AT, + .pc_name = "GETATTR", }, [ACLPROC2_ACCESS] = { .pc_func = nfsacld_proc_access, @@ -411,6 +403,7 @@ static const struct svc_procedure nfsd_acl_procedures2[5] = { .pc_ressize = sizeof(struct nfsd3_accessres), .pc_cachetype = RC_NOCACHE, .pc_xdrressize = ST+AT+1, + .pc_name = "SETATTR", }, }; diff --git a/fs/nfsd/nfs3acl.c b/fs/nfsd/nfs3acl.c index 7c30876a31a1..9a6f18d74d14 100644 --- a/fs/nfsd/nfs3acl.c +++ b/fs/nfsd/nfs3acl.c @@ -103,10 +103,12 @@ static __be32 nfsd3_proc_setacl(struct svc_rqst *rqstp) fh_lock(fh); - error = set_posix_acl(inode, ACL_TYPE_ACCESS, argp->acl_access); + error = set_posix_acl(&init_user_ns, inode, ACL_TYPE_ACCESS, + argp->acl_access); if (error) goto out_drop_lock; - error = set_posix_acl(inode, ACL_TYPE_DEFAULT, argp->acl_default); + error = set_posix_acl(&init_user_ns, inode, ACL_TYPE_DEFAULT, + argp->acl_default); out_drop_lock: fh_unlock(fh); @@ -124,43 +126,39 @@ out: /* * XDR decode functions */ + static int nfs3svc_decode_getaclargs(struct svc_rqst *rqstp, __be32 *p) { + struct xdr_stream *xdr = &rqstp->rq_arg_stream; struct nfsd3_getaclargs *args = rqstp->rq_argp; - p = nfs3svc_decode_fh(p, &args->fh); - if (!p) + if (!svcxdr_decode_nfs_fh3(xdr, &args->fh)) + return 0; + if (xdr_stream_decode_u32(xdr, &args->mask) < 0) return 0; - args->mask = ntohl(*p); p++; - return xdr_argsize_check(rqstp, p); + return 1; } - static int nfs3svc_decode_setaclargs(struct svc_rqst *rqstp, __be32 *p) { - struct nfsd3_setaclargs *args = rqstp->rq_argp; - struct kvec *head = rqstp->rq_arg.head; - unsigned int base; - int n; + struct xdr_stream *xdr = &rqstp->rq_arg_stream; + struct nfsd3_setaclargs *argp = rqstp->rq_argp; - p = nfs3svc_decode_fh(p, &args->fh); - if (!p) + if (!svcxdr_decode_nfs_fh3(xdr, &argp->fh)) return 0; - args->mask = ntohl(*p++); - if (args->mask & ~NFS_ACL_MASK || - !xdr_argsize_check(rqstp, p)) + if (xdr_stream_decode_u32(xdr, &argp->mask) < 0) + return 0; + if (argp->mask & ~NFS_ACL_MASK) + return 0; + if (!nfs_stream_decode_acl(xdr, NULL, (argp->mask & NFS_ACL) ? + &argp->acl_access : NULL)) + return 0; + if (!nfs_stream_decode_acl(xdr, NULL, (argp->mask & NFS_DFACL) ? + &argp->acl_default : NULL)) return 0; - base = (char *)p - (char *)head->iov_base; - n = nfsacl_decode(&rqstp->rq_arg, base, NULL, - (args->mask & NFS_ACL) ? - &args->acl_access : NULL); - if (n > 0) - n = nfsacl_decode(&rqstp->rq_arg, base + n, NULL, - (args->mask & NFS_DFACL) ? - &args->acl_default : NULL); - return (n > 0); + return 1; } /* @@ -251,6 +249,7 @@ static const struct svc_procedure nfsd_acl_procedures3[3] = { .pc_ressize = sizeof(struct nfsd_voidres), .pc_cachetype = RC_NOCACHE, .pc_xdrressize = ST, + .pc_name = "NULL", }, [ACLPROC3_GETACL] = { .pc_func = nfsd3_proc_getacl, @@ -261,6 +260,7 @@ static const struct svc_procedure nfsd_acl_procedures3[3] = { .pc_ressize = sizeof(struct nfsd3_getaclres), .pc_cachetype = RC_NOCACHE, .pc_xdrressize = ST+1+2*(1+ACL), + .pc_name = "GETACL", }, [ACLPROC3_SETACL] = { .pc_func = nfsd3_proc_setacl, @@ -271,6 +271,7 @@ static const struct svc_procedure nfsd_acl_procedures3[3] = { .pc_ressize = sizeof(struct nfsd3_attrstat), .pc_cachetype = RC_NOCACHE, .pc_xdrressize = ST+pAT, + .pc_name = "SETACL", }, }; diff --git a/fs/nfsd/nfs3proc.c b/fs/nfsd/nfs3proc.c index 76931f4f57c3..8675851199f8 100644 --- a/fs/nfsd/nfs3proc.c +++ b/fs/nfsd/nfs3proc.c @@ -124,15 +124,16 @@ nfsd3_proc_access(struct svc_rqst *rqstp) static __be32 nfsd3_proc_readlink(struct svc_rqst *rqstp) { - struct nfsd3_readlinkargs *argp = rqstp->rq_argp; + struct nfsd_fhandle *argp = rqstp->rq_argp; struct nfsd3_readlinkres *resp = rqstp->rq_resp; + char *buffer = page_address(*(rqstp->rq_next_page++)); dprintk("nfsd: READLINK(3) %s\n", SVCFH_fmt(&argp->fh)); /* Read the symlink. */ fh_copy(&resp->fh, &argp->fh); resp->len = NFS3_MAXPATHLEN; - resp->status = nfsd_readlink(rqstp, &resp->fh, argp->buffer, &resp->len); + resp->status = nfsd_readlink(rqstp, &resp->fh, buffer, &resp->len); return rpc_success; } @@ -144,25 +145,38 @@ nfsd3_proc_read(struct svc_rqst *rqstp) { struct nfsd3_readargs *argp = rqstp->rq_argp; struct nfsd3_readres *resp = rqstp->rq_resp; - u32 max_blocksize = svc_max_payload(rqstp); - unsigned long cnt = min(argp->count, max_blocksize); + u32 max_blocksize = svc_max_payload(rqstp); + unsigned int len; + int v; + + argp->count = min_t(u32, argp->count, max_blocksize); dprintk("nfsd: READ(3) %s %lu bytes at %Lu\n", SVCFH_fmt(&argp->fh), (unsigned long) argp->count, (unsigned long long) argp->offset); + v = 0; + len = argp->count; + while (len > 0) { + struct page *page = *(rqstp->rq_next_page++); + + rqstp->rq_vec[v].iov_base = page_address(page); + rqstp->rq_vec[v].iov_len = min_t(unsigned int, len, PAGE_SIZE); + len -= rqstp->rq_vec[v].iov_len; + v++; + } + /* Obtain buffer pointer for payload. * 1 (status) + 22 (post_op_attr) + 1 (count) + 1 (eof) * + 1 (xdr opaque byte count) = 26 */ - resp->count = cnt; + resp->count = argp->count; svc_reserve_auth(rqstp, ((1 + NFS3_POST_OP_ATTR_WORDS + 3)<<2) + resp->count +4); fh_copy(&resp->fh, &argp->fh); resp->status = nfsd_read(rqstp, &resp->fh, argp->offset, - rqstp->rq_vec, argp->vlen, &resp->count, - &resp->eof); + rqstp->rq_vec, v, &resp->count, &resp->eof); return rpc_success; } @@ -421,6 +435,23 @@ nfsd3_proc_link(struct svc_rqst *rqstp) return rpc_success; } +static void nfsd3_init_dirlist_pages(struct svc_rqst *rqstp, + struct nfsd3_readdirres *resp, + int count) +{ + count = min_t(u32, count, svc_max_payload(rqstp)); + + /* Convert byte count to number of words (i.e. >> 2), + * and reserve room for the NULL ptr & eof flag (-2 words) */ + resp->buflen = (count >> 2) - 2; + + resp->buffer = page_address(*rqstp->rq_next_page); + while (count > 0) { + rqstp->rq_next_page++; + count -= PAGE_SIZE; + } +} + /* * Read a portion of a directory. */ @@ -430,6 +461,7 @@ nfsd3_proc_readdir(struct svc_rqst *rqstp) struct nfsd3_readdirargs *argp = rqstp->rq_argp; struct nfsd3_readdirres *resp = rqstp->rq_resp; int count = 0; + loff_t offset; struct page **p; caddr_t page_addr = NULL; @@ -437,18 +469,16 @@ nfsd3_proc_readdir(struct svc_rqst *rqstp) SVCFH_fmt(&argp->fh), argp->count, (u32) argp->cookie); - /* Make sure we've room for the NULL ptr & eof flag, and shrink to - * client read size */ - count = (argp->count >> 2) - 2; + nfsd3_init_dirlist_pages(rqstp, resp, argp->count); /* Read directory and encode entries on the fly */ fh_copy(&resp->fh, &argp->fh); - resp->buflen = count; resp->common.err = nfs_ok; - resp->buffer = argp->buffer; resp->rqstp = rqstp; - resp->status = nfsd_readdir(rqstp, &resp->fh, (loff_t *)&argp->cookie, + offset = argp->cookie; + + resp->status = nfsd_readdir(rqstp, &resp->fh, &offset, &resp->common, nfs3svc_encode_entry); memcpy(resp->verf, argp->verf, 8); count = 0; @@ -464,8 +494,6 @@ nfsd3_proc_readdir(struct svc_rqst *rqstp) } resp->count = count >> 2; if (resp->offset) { - loff_t offset = argp->cookie; - if (unlikely(resp->offset1)) { /* we ended up with offset on a page boundary */ *resp->offset = htonl(offset >> 32); @@ -498,16 +526,12 @@ nfsd3_proc_readdirplus(struct svc_rqst *rqstp) SVCFH_fmt(&argp->fh), argp->count, (u32) argp->cookie); - /* Convert byte count to number of words (i.e. >> 2), - * and reserve room for the NULL ptr & eof flag (-2 words) */ - resp->count = (argp->count >> 2) - 2; + nfsd3_init_dirlist_pages(rqstp, resp, argp->count); /* Read directory and encode entries on the fly */ fh_copy(&resp->fh, &argp->fh); resp->common.err = nfs_ok; - resp->buffer = argp->buffer; - resp->buflen = resp->count; resp->rqstp = rqstp; offset = argp->cookie; @@ -683,7 +707,6 @@ out: * NFSv3 Server procedures. * Only the results of non-idempotent operations are cached. */ -#define nfs3svc_decode_fhandleargs nfs3svc_decode_fhandle #define nfs3svc_encode_attrstatres nfs3svc_encode_attrstat #define nfs3svc_encode_wccstatres nfs3svc_encode_wccstat #define nfsd3_mkdirargs nfsd3_createargs @@ -708,16 +731,18 @@ static const struct svc_procedure nfsd_procedures3[22] = { .pc_ressize = sizeof(struct nfsd_voidres), .pc_cachetype = RC_NOCACHE, .pc_xdrressize = ST, + .pc_name = "NULL", }, [NFS3PROC_GETATTR] = { .pc_func = nfsd3_proc_getattr, .pc_decode = nfs3svc_decode_fhandleargs, .pc_encode = nfs3svc_encode_attrstatres, .pc_release = nfs3svc_release_fhandle, - .pc_argsize = sizeof(struct nfsd3_fhandleargs), + .pc_argsize = sizeof(struct nfsd_fhandle), .pc_ressize = sizeof(struct nfsd3_attrstatres), .pc_cachetype = RC_NOCACHE, .pc_xdrressize = ST+AT, + .pc_name = "GETATTR", }, [NFS3PROC_SETATTR] = { .pc_func = nfsd3_proc_setattr, @@ -728,6 +753,7 @@ static const struct svc_procedure nfsd_procedures3[22] = { .pc_ressize = sizeof(struct nfsd3_wccstatres), .pc_cachetype = RC_REPLBUFF, .pc_xdrressize = ST+WC, + .pc_name = "SETATTR", }, [NFS3PROC_LOOKUP] = { .pc_func = nfsd3_proc_lookup, @@ -738,6 +764,7 @@ static const struct svc_procedure nfsd_procedures3[22] = { .pc_ressize = sizeof(struct nfsd3_diropres), .pc_cachetype = RC_NOCACHE, .pc_xdrressize = ST+FH+pAT+pAT, + .pc_name = "LOOKUP", }, [NFS3PROC_ACCESS] = { .pc_func = nfsd3_proc_access, @@ -748,16 +775,18 @@ static const struct svc_procedure nfsd_procedures3[22] = { .pc_ressize = sizeof(struct nfsd3_accessres), .pc_cachetype = RC_NOCACHE, .pc_xdrressize = ST+pAT+1, + .pc_name = "ACCESS", }, [NFS3PROC_READLINK] = { .pc_func = nfsd3_proc_readlink, - .pc_decode = nfs3svc_decode_readlinkargs, + .pc_decode = nfs3svc_decode_fhandleargs, .pc_encode = nfs3svc_encode_readlinkres, .pc_release = nfs3svc_release_fhandle, - .pc_argsize = sizeof(struct nfsd3_readlinkargs), + .pc_argsize = sizeof(struct nfsd_fhandle), .pc_ressize = sizeof(struct nfsd3_readlinkres), .pc_cachetype = RC_NOCACHE, .pc_xdrressize = ST+pAT+1+NFS3_MAXPATHLEN/4, + .pc_name = "READLINK", }, [NFS3PROC_READ] = { .pc_func = nfsd3_proc_read, @@ -768,6 +797,7 @@ static const struct svc_procedure nfsd_procedures3[22] = { .pc_ressize = sizeof(struct nfsd3_readres), .pc_cachetype = RC_NOCACHE, .pc_xdrressize = ST+pAT+4+NFSSVC_MAXBLKSIZE/4, + .pc_name = "READ", }, [NFS3PROC_WRITE] = { .pc_func = nfsd3_proc_write, @@ -778,6 +808,7 @@ static const struct svc_procedure nfsd_procedures3[22] = { .pc_ressize = sizeof(struct nfsd3_writeres), .pc_cachetype = RC_REPLBUFF, .pc_xdrressize = ST+WC+4, + .pc_name = "WRITE", }, [NFS3PROC_CREATE] = { .pc_func = nfsd3_proc_create, @@ -788,6 +819,7 @@ static const struct svc_procedure nfsd_procedures3[22] = { .pc_ressize = sizeof(struct nfsd3_createres), .pc_cachetype = RC_REPLBUFF, .pc_xdrressize = ST+(1+FH+pAT)+WC, + .pc_name = "CREATE", }, [NFS3PROC_MKDIR] = { .pc_func = nfsd3_proc_mkdir, @@ -798,6 +830,7 @@ static const struct svc_procedure nfsd_procedures3[22] = { .pc_ressize = sizeof(struct nfsd3_createres), .pc_cachetype = RC_REPLBUFF, .pc_xdrressize = ST+(1+FH+pAT)+WC, + .pc_name = "MKDIR", }, [NFS3PROC_SYMLINK] = { .pc_func = nfsd3_proc_symlink, @@ -808,6 +841,7 @@ static const struct svc_procedure nfsd_procedures3[22] = { .pc_ressize = sizeof(struct nfsd3_createres), .pc_cachetype = RC_REPLBUFF, .pc_xdrressize = ST+(1+FH+pAT)+WC, + .pc_name = "SYMLINK", }, [NFS3PROC_MKNOD] = { .pc_func = nfsd3_proc_mknod, @@ -818,6 +852,7 @@ static const struct svc_procedure nfsd_procedures3[22] = { .pc_ressize = sizeof(struct nfsd3_createres), .pc_cachetype = RC_REPLBUFF, .pc_xdrressize = ST+(1+FH+pAT)+WC, + .pc_name = "MKNOD", }, [NFS3PROC_REMOVE] = { .pc_func = nfsd3_proc_remove, @@ -828,6 +863,7 @@ static const struct svc_procedure nfsd_procedures3[22] = { .pc_ressize = sizeof(struct nfsd3_wccstatres), .pc_cachetype = RC_REPLBUFF, .pc_xdrressize = ST+WC, + .pc_name = "REMOVE", }, [NFS3PROC_RMDIR] = { .pc_func = nfsd3_proc_rmdir, @@ -838,6 +874,7 @@ static const struct svc_procedure nfsd_procedures3[22] = { .pc_ressize = sizeof(struct nfsd3_wccstatres), .pc_cachetype = RC_REPLBUFF, .pc_xdrressize = ST+WC, + .pc_name = "RMDIR", }, [NFS3PROC_RENAME] = { .pc_func = nfsd3_proc_rename, @@ -848,6 +885,7 @@ static const struct svc_procedure nfsd_procedures3[22] = { .pc_ressize = sizeof(struct nfsd3_renameres), .pc_cachetype = RC_REPLBUFF, .pc_xdrressize = ST+WC+WC, + .pc_name = "RENAME", }, [NFS3PROC_LINK] = { .pc_func = nfsd3_proc_link, @@ -858,6 +896,7 @@ static const struct svc_procedure nfsd_procedures3[22] = { .pc_ressize = sizeof(struct nfsd3_linkres), .pc_cachetype = RC_REPLBUFF, .pc_xdrressize = ST+pAT+WC, + .pc_name = "LINK", }, [NFS3PROC_READDIR] = { .pc_func = nfsd3_proc_readdir, @@ -867,6 +906,7 @@ static const struct svc_procedure nfsd_procedures3[22] = { .pc_argsize = sizeof(struct nfsd3_readdirargs), .pc_ressize = sizeof(struct nfsd3_readdirres), .pc_cachetype = RC_NOCACHE, + .pc_name = "READDIR", }, [NFS3PROC_READDIRPLUS] = { .pc_func = nfsd3_proc_readdirplus, @@ -876,6 +916,7 @@ static const struct svc_procedure nfsd_procedures3[22] = { .pc_argsize = sizeof(struct nfsd3_readdirplusargs), .pc_ressize = sizeof(struct nfsd3_readdirres), .pc_cachetype = RC_NOCACHE, + .pc_name = "READDIRPLUS", }, [NFS3PROC_FSSTAT] = { .pc_func = nfsd3_proc_fsstat, @@ -885,6 +926,7 @@ static const struct svc_procedure nfsd_procedures3[22] = { .pc_ressize = sizeof(struct nfsd3_fsstatres), .pc_cachetype = RC_NOCACHE, .pc_xdrressize = ST+pAT+2*6+1, + .pc_name = "FSSTAT", }, [NFS3PROC_FSINFO] = { .pc_func = nfsd3_proc_fsinfo, @@ -894,6 +936,7 @@ static const struct svc_procedure nfsd_procedures3[22] = { .pc_ressize = sizeof(struct nfsd3_fsinfores), .pc_cachetype = RC_NOCACHE, .pc_xdrressize = ST+pAT+12, + .pc_name = "FSINFO", }, [NFS3PROC_PATHCONF] = { .pc_func = nfsd3_proc_pathconf, @@ -903,6 +946,7 @@ static const struct svc_procedure nfsd_procedures3[22] = { .pc_ressize = sizeof(struct nfsd3_pathconfres), .pc_cachetype = RC_NOCACHE, .pc_xdrressize = ST+pAT+6, + .pc_name = "PATHCONF", }, [NFS3PROC_COMMIT] = { .pc_func = nfsd3_proc_commit, @@ -913,6 +957,7 @@ static const struct svc_procedure nfsd_procedures3[22] = { .pc_ressize = sizeof(struct nfsd3_commitres), .pc_cachetype = RC_NOCACHE, .pc_xdrressize = ST+WC+2, + .pc_name = "COMMIT", }, }; diff --git a/fs/nfsd/nfs3xdr.c b/fs/nfsd/nfs3xdr.c index 34b880211e5e..9d9a01ce0b27 100644 --- a/fs/nfsd/nfs3xdr.c +++ b/fs/nfsd/nfs3xdr.c @@ -29,8 +29,9 @@ static u32 nfs3_ftypes[] = { /* - * XDR functions for basic NFS types + * Basic NFSv3 data types (RFC 1813 Sections 2.5 and 2.6) */ + static __be32 * encode_time3(__be32 *p, struct timespec64 *time) { @@ -38,32 +39,47 @@ encode_time3(__be32 *p, struct timespec64 *time) return p; } -static __be32 * -decode_time3(__be32 *p, struct timespec64 *time) +static bool +svcxdr_decode_nfstime3(struct xdr_stream *xdr, struct timespec64 *timep) { - time->tv_sec = ntohl(*p++); - time->tv_nsec = ntohl(*p++); - return p; + __be32 *p; + + p = xdr_inline_decode(xdr, XDR_UNIT * 2); + if (!p) + return false; + timep->tv_sec = be32_to_cpup(p++); + timep->tv_nsec = be32_to_cpup(p); + + return true; } -static __be32 * -decode_fh(__be32 *p, struct svc_fh *fhp) +/** + * svcxdr_decode_nfs_fh3 - Decode an NFSv3 file handle + * @xdr: XDR stream positioned at an undecoded NFSv3 FH + * @fhp: OUT: filled-in server file handle + * + * Return values: + * %false: The encoded file handle was not valid + * %true: @fhp has been initialized + */ +bool +svcxdr_decode_nfs_fh3(struct xdr_stream *xdr, struct svc_fh *fhp) { - unsigned int size; + __be32 *p; + u32 size; + + if (xdr_stream_decode_u32(xdr, &size) < 0) + return false; + if (size == 0 || size > NFS3_FHSIZE) + return false; + p = xdr_inline_decode(xdr, size); + if (!p) + return false; fh_init(fhp, NFS3_FHSIZE); - size = ntohl(*p++); - if (size > NFS3_FHSIZE) - return NULL; - - memcpy(&fhp->fh_handle.fh_base, p, size); fhp->fh_handle.fh_size = size; - return p + XDR_QUADLEN(size); -} + memcpy(&fhp->fh_handle.fh_base, p, size); -/* Helper function for NFSv3 ACL code */ -__be32 *nfs3svc_decode_fh(__be32 *p, struct svc_fh *fhp) -{ - return decode_fh(p, fhp); + return true; } static __be32 * @@ -76,69 +92,165 @@ encode_fh(__be32 *p, struct svc_fh *fhp) return p + XDR_QUADLEN(size); } -/* - * Decode a file name and make sure that the path contains - * no slashes or null bytes. - */ -static __be32 * -decode_filename(__be32 *p, char **namp, unsigned int *lenp) +static bool +svcxdr_decode_filename3(struct xdr_stream *xdr, char **name, unsigned int *len) { - char *name; - unsigned int i; + u32 size, i; + __be32 *p; + char *c; + + if (xdr_stream_decode_u32(xdr, &size) < 0) + return false; + if (size == 0 || size > NFS3_MAXNAMLEN) + return false; + p = xdr_inline_decode(xdr, size); + if (!p) + return false; - if ((p = xdr_decode_string_inplace(p, namp, lenp, NFS3_MAXNAMLEN)) != NULL) { - for (i = 0, name = *namp; i < *lenp; i++, name++) { - if (*name == '\0' || *name == '/') - return NULL; - } + *len = size; + *name = (char *)p; + for (i = 0, c = *name; i < size; i++, c++) { + if (*c == '\0' || *c == '/') + return false; } - return p; + return true; } -static __be32 * -decode_sattr3(__be32 *p, struct iattr *iap, struct user_namespace *userns) +static bool +svcxdr_decode_diropargs3(struct xdr_stream *xdr, struct svc_fh *fhp, + char **name, unsigned int *len) +{ + return svcxdr_decode_nfs_fh3(xdr, fhp) && + svcxdr_decode_filename3(xdr, name, len); +} + +static bool +svcxdr_decode_sattr3(struct svc_rqst *rqstp, struct xdr_stream *xdr, + struct iattr *iap) { - u32 tmp; + u32 set_it; iap->ia_valid = 0; - if (*p++) { + if (xdr_stream_decode_bool(xdr, &set_it) < 0) + return false; + if (set_it) { + u32 mode; + + if (xdr_stream_decode_u32(xdr, &mode) < 0) + return false; iap->ia_valid |= ATTR_MODE; - iap->ia_mode = ntohl(*p++); + iap->ia_mode = mode; } - if (*p++) { - iap->ia_uid = make_kuid(userns, ntohl(*p++)); + if (xdr_stream_decode_bool(xdr, &set_it) < 0) + return false; + if (set_it) { + u32 uid; + + if (xdr_stream_decode_u32(xdr, &uid) < 0) + return false; + iap->ia_uid = make_kuid(nfsd_user_namespace(rqstp), uid); if (uid_valid(iap->ia_uid)) iap->ia_valid |= ATTR_UID; } - if (*p++) { - iap->ia_gid = make_kgid(userns, ntohl(*p++)); + if (xdr_stream_decode_bool(xdr, &set_it) < 0) + return false; + if (set_it) { + u32 gid; + + if (xdr_stream_decode_u32(xdr, &gid) < 0) + return false; + iap->ia_gid = make_kgid(nfsd_user_namespace(rqstp), gid); if (gid_valid(iap->ia_gid)) iap->ia_valid |= ATTR_GID; } - if (*p++) { - u64 newsize; + if (xdr_stream_decode_bool(xdr, &set_it) < 0) + return false; + if (set_it) { + u64 newsize; + if (xdr_stream_decode_u64(xdr, &newsize) < 0) + return false; iap->ia_valid |= ATTR_SIZE; - p = xdr_decode_hyper(p, &newsize); iap->ia_size = min_t(u64, newsize, NFS_OFFSET_MAX); } - if ((tmp = ntohl(*p++)) == 1) { /* set to server time */ + if (xdr_stream_decode_u32(xdr, &set_it) < 0) + return false; + switch (set_it) { + case DONT_CHANGE: + break; + case SET_TO_SERVER_TIME: iap->ia_valid |= ATTR_ATIME; - } else if (tmp == 2) { /* set to client time */ + break; + case SET_TO_CLIENT_TIME: + if (!svcxdr_decode_nfstime3(xdr, &iap->ia_atime)) + return false; iap->ia_valid |= ATTR_ATIME | ATTR_ATIME_SET; - iap->ia_atime.tv_sec = ntohl(*p++); - iap->ia_atime.tv_nsec = ntohl(*p++); + break; + default: + return false; } - if ((tmp = ntohl(*p++)) == 1) { /* set to server time */ + if (xdr_stream_decode_u32(xdr, &set_it) < 0) + return false; + switch (set_it) { + case DONT_CHANGE: + break; + case SET_TO_SERVER_TIME: iap->ia_valid |= ATTR_MTIME; - } else if (tmp == 2) { /* set to client time */ + break; + case SET_TO_CLIENT_TIME: + if (!svcxdr_decode_nfstime3(xdr, &iap->ia_mtime)) + return false; iap->ia_valid |= ATTR_MTIME | ATTR_MTIME_SET; - iap->ia_mtime.tv_sec = ntohl(*p++); - iap->ia_mtime.tv_nsec = ntohl(*p++); + break; + default: + return false; } - return p; + + return true; +} + +static bool +svcxdr_decode_sattrguard3(struct xdr_stream *xdr, struct nfsd3_sattrargs *args) +{ + __be32 *p; + u32 check; + + if (xdr_stream_decode_bool(xdr, &check) < 0) + return false; + if (check) { + p = xdr_inline_decode(xdr, XDR_UNIT * 2); + if (!p) + return false; + args->check_guard = 1; + args->guardtime = be32_to_cpup(p); + } else + args->check_guard = 0; + + return true; +} + +static bool +svcxdr_decode_specdata3(struct xdr_stream *xdr, struct nfsd3_mknodargs *args) +{ + __be32 *p; + + p = xdr_inline_decode(xdr, XDR_UNIT * 2); + if (!p) + return false; + args->major = be32_to_cpup(p++); + args->minor = be32_to_cpup(p); + + return true; +} + +static bool +svcxdr_decode_devicedata3(struct svc_rqst *rqstp, struct xdr_stream *xdr, + struct nfsd3_mknodargs *args) +{ + return svcxdr_decode_sattr3(rqstp, xdr, &args->attrs) && + svcxdr_decode_specdata3(xdr, args); } static __be32 *encode_fsid(__be32 *p, struct svc_fh *fhp) @@ -252,6 +364,11 @@ encode_wcc_data(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp) return encode_post_op_attr(rqstp, p, fhp); } +static bool fs_supports_change_attribute(struct super_block *sb) +{ + return sb->s_flags & SB_I_VERSION || sb->s_export_op->fetch_iversion; +} + /* * Fill in the pre_op attr for the wcc data */ @@ -260,24 +377,26 @@ void fill_pre_wcc(struct svc_fh *fhp) struct inode *inode; struct kstat stat; bool v4 = (fhp->fh_maxsize == NFS4_FHSIZE); - __be32 err; if (fhp->fh_no_wcc || fhp->fh_pre_saved) return; inode = d_inode(fhp->fh_dentry); - err = fh_getattr(fhp, &stat); - if (err) { - /* Grab the times from inode anyway */ - stat.mtime = inode->i_mtime; - stat.ctime = inode->i_ctime; - stat.size = inode->i_size; + if (fs_supports_change_attribute(inode->i_sb) || !v4) { + __be32 err = fh_getattr(fhp, &stat); + + if (err) { + /* Grab the times from inode anyway */ + stat.mtime = inode->i_mtime; + stat.ctime = inode->i_ctime; + stat.size = inode->i_size; + } + fhp->fh_pre_mtime = stat.mtime; + fhp->fh_pre_ctime = stat.ctime; + fhp->fh_pre_size = stat.size; } if (v4) fhp->fh_pre_change = nfsd4_change_attribute(&stat, inode); - fhp->fh_pre_mtime = stat.mtime; - fhp->fh_pre_ctime = stat.ctime; - fhp->fh_pre_size = stat.size; fhp->fh_pre_saved = true; } @@ -288,7 +407,6 @@ void fill_post_wcc(struct svc_fh *fhp) { bool v4 = (fhp->fh_maxsize == NFS4_FHSIZE); struct inode *inode = d_inode(fhp->fh_dentry); - __be32 err; if (fhp->fh_no_wcc) return; @@ -296,12 +414,16 @@ void fill_post_wcc(struct svc_fh *fhp) if (fhp->fh_post_saved) printk("nfsd: inode locked twice during operation.\n"); - err = fh_getattr(fhp, &fhp->fh_post_attr); - if (err) { - fhp->fh_post_saved = false; - fhp->fh_post_attr.ctime = inode->i_ctime; - } else - fhp->fh_post_saved = true; + fhp->fh_post_saved = true; + + if (fs_supports_change_attribute(inode->i_sb) || !v4) { + __be32 err = fh_getattr(fhp, &fhp->fh_post_attr); + + if (err) { + fhp->fh_post_saved = false; + fhp->fh_post_attr.ctime = inode->i_ctime; + } + } if (v4) fhp->fh_post_change = nfsd4_change_attribute(&fhp->fh_post_attr, inode); @@ -312,331 +434,277 @@ void fill_post_wcc(struct svc_fh *fhp) */ int -nfs3svc_decode_fhandle(struct svc_rqst *rqstp, __be32 *p) +nfs3svc_decode_fhandleargs(struct svc_rqst *rqstp, __be32 *p) { + struct xdr_stream *xdr = &rqstp->rq_arg_stream; struct nfsd_fhandle *args = rqstp->rq_argp; - p = decode_fh(p, &args->fh); - if (!p) - return 0; - return xdr_argsize_check(rqstp, p); + return svcxdr_decode_nfs_fh3(xdr, &args->fh); } int nfs3svc_decode_sattrargs(struct svc_rqst *rqstp, __be32 *p) { + struct xdr_stream *xdr = &rqstp->rq_arg_stream; struct nfsd3_sattrargs *args = rqstp->rq_argp; - p = decode_fh(p, &args->fh); - if (!p) - return 0; - p = decode_sattr3(p, &args->attrs, nfsd_user_namespace(rqstp)); - - if ((args->check_guard = ntohl(*p++)) != 0) { - struct timespec64 time; - p = decode_time3(p, &time); - args->guardtime = time.tv_sec; - } - - return xdr_argsize_check(rqstp, p); + return svcxdr_decode_nfs_fh3(xdr, &args->fh) && + svcxdr_decode_sattr3(rqstp, xdr, &args->attrs) && + svcxdr_decode_sattrguard3(xdr, args); } int nfs3svc_decode_diropargs(struct svc_rqst *rqstp, __be32 *p) { + struct xdr_stream *xdr = &rqstp->rq_arg_stream; struct nfsd3_diropargs *args = rqstp->rq_argp; - if (!(p = decode_fh(p, &args->fh)) - || !(p = decode_filename(p, &args->name, &args->len))) - return 0; - - return xdr_argsize_check(rqstp, p); + return svcxdr_decode_diropargs3(xdr, &args->fh, &args->name, &args->len); } int nfs3svc_decode_accessargs(struct svc_rqst *rqstp, __be32 *p) { + struct xdr_stream *xdr = &rqstp->rq_arg_stream; struct nfsd3_accessargs *args = rqstp->rq_argp; - p = decode_fh(p, &args->fh); - if (!p) + if (!svcxdr_decode_nfs_fh3(xdr, &args->fh)) + return 0; + if (xdr_stream_decode_u32(xdr, &args->access) < 0) return 0; - args->access = ntohl(*p++); - return xdr_argsize_check(rqstp, p); + return 1; } int nfs3svc_decode_readargs(struct svc_rqst *rqstp, __be32 *p) { + struct xdr_stream *xdr = &rqstp->rq_arg_stream; struct nfsd3_readargs *args = rqstp->rq_argp; - unsigned int len; - int v; - u32 max_blocksize = svc_max_payload(rqstp); - p = decode_fh(p, &args->fh); - if (!p) + if (!svcxdr_decode_nfs_fh3(xdr, &args->fh)) + return 0; + if (xdr_stream_decode_u64(xdr, &args->offset) < 0) + return 0; + if (xdr_stream_decode_u32(xdr, &args->count) < 0) return 0; - p = xdr_decode_hyper(p, &args->offset); - - args->count = ntohl(*p++); - len = min(args->count, max_blocksize); - - /* set up the kvec */ - v=0; - while (len > 0) { - struct page *p = *(rqstp->rq_next_page++); - rqstp->rq_vec[v].iov_base = page_address(p); - rqstp->rq_vec[v].iov_len = min_t(unsigned int, len, PAGE_SIZE); - len -= rqstp->rq_vec[v].iov_len; - v++; - } - args->vlen = v; - return xdr_argsize_check(rqstp, p); + return 1; } int nfs3svc_decode_writeargs(struct svc_rqst *rqstp, __be32 *p) { + struct xdr_stream *xdr = &rqstp->rq_arg_stream; struct nfsd3_writeargs *args = rqstp->rq_argp; - unsigned int len, hdr, dlen; u32 max_blocksize = svc_max_payload(rqstp); struct kvec *head = rqstp->rq_arg.head; struct kvec *tail = rqstp->rq_arg.tail; + size_t remaining; - p = decode_fh(p, &args->fh); - if (!p) + if (!svcxdr_decode_nfs_fh3(xdr, &args->fh)) return 0; - p = xdr_decode_hyper(p, &args->offset); - - args->count = ntohl(*p++); - args->stable = ntohl(*p++); - len = args->len = ntohl(*p++); - if ((void *)p > head->iov_base + head->iov_len) + if (xdr_stream_decode_u64(xdr, &args->offset) < 0) return 0; - /* - * The count must equal the amount of data passed. - */ - if (args->count != args->len) + if (xdr_stream_decode_u32(xdr, &args->count) < 0) + return 0; + if (xdr_stream_decode_u32(xdr, &args->stable) < 0) return 0; - /* - * Check to make sure that we got the right number of - * bytes. - */ - hdr = (void*)p - head->iov_base; - dlen = head->iov_len + rqstp->rq_arg.page_len + tail->iov_len - hdr; - /* - * Round the length of the data which was specified up to - * the next multiple of XDR units and then compare that - * against the length which was actually received. - * Note that when RPCSEC/GSS (for example) is used, the - * data buffer can be padded so dlen might be larger - * than required. It must never be smaller. - */ - if (dlen < XDR_QUADLEN(len)*4) + /* opaque data */ + if (xdr_stream_decode_u32(xdr, &args->len) < 0) return 0; + /* request sanity */ + if (args->count != args->len) + return 0; + remaining = head->iov_len + rqstp->rq_arg.page_len + tail->iov_len; + remaining -= xdr_stream_pos(xdr); + if (remaining < xdr_align_size(args->len)) + return 0; if (args->count > max_blocksize) { args->count = max_blocksize; - len = args->len = max_blocksize; + args->len = max_blocksize; } - args->first.iov_base = (void *)p; - args->first.iov_len = head->iov_len - hdr; + args->first.iov_base = xdr->p; + args->first.iov_len = head->iov_len - xdr_stream_pos(xdr); + return 1; } int nfs3svc_decode_createargs(struct svc_rqst *rqstp, __be32 *p) { + struct xdr_stream *xdr = &rqstp->rq_arg_stream; struct nfsd3_createargs *args = rqstp->rq_argp; - if (!(p = decode_fh(p, &args->fh)) - || !(p = decode_filename(p, &args->name, &args->len))) + if (!svcxdr_decode_diropargs3(xdr, &args->fh, &args->name, &args->len)) return 0; - - switch (args->createmode = ntohl(*p++)) { + if (xdr_stream_decode_u32(xdr, &args->createmode) < 0) + return 0; + switch (args->createmode) { case NFS3_CREATE_UNCHECKED: case NFS3_CREATE_GUARDED: - p = decode_sattr3(p, &args->attrs, nfsd_user_namespace(rqstp)); - break; + return svcxdr_decode_sattr3(rqstp, xdr, &args->attrs); case NFS3_CREATE_EXCLUSIVE: - args->verf = p; - p += 2; + args->verf = xdr_inline_decode(xdr, NFS3_CREATEVERFSIZE); + if (!args->verf) + return 0; break; default: return 0; } - - return xdr_argsize_check(rqstp, p); + return 1; } int nfs3svc_decode_mkdirargs(struct svc_rqst *rqstp, __be32 *p) { + struct xdr_stream *xdr = &rqstp->rq_arg_stream; struct nfsd3_createargs *args = rqstp->rq_argp; - if (!(p = decode_fh(p, &args->fh)) || - !(p = decode_filename(p, &args->name, &args->len))) - return 0; - p = decode_sattr3(p, &args->attrs, nfsd_user_namespace(rqstp)); - - return xdr_argsize_check(rqstp, p); + return svcxdr_decode_diropargs3(xdr, &args->fh, + &args->name, &args->len) && + svcxdr_decode_sattr3(rqstp, xdr, &args->attrs); } int nfs3svc_decode_symlinkargs(struct svc_rqst *rqstp, __be32 *p) { + struct xdr_stream *xdr = &rqstp->rq_arg_stream; struct nfsd3_symlinkargs *args = rqstp->rq_argp; - char *base = (char *)p; - size_t dlen; + struct kvec *head = rqstp->rq_arg.head; + struct kvec *tail = rqstp->rq_arg.tail; + size_t remaining; - if (!(p = decode_fh(p, &args->ffh)) || - !(p = decode_filename(p, &args->fname, &args->flen))) + if (!svcxdr_decode_diropargs3(xdr, &args->ffh, &args->fname, &args->flen)) + return 0; + if (!svcxdr_decode_sattr3(rqstp, xdr, &args->attrs)) + return 0; + if (xdr_stream_decode_u32(xdr, &args->tlen) < 0) return 0; - p = decode_sattr3(p, &args->attrs, nfsd_user_namespace(rqstp)); - args->tlen = ntohl(*p++); + /* request sanity */ + remaining = head->iov_len + rqstp->rq_arg.page_len + tail->iov_len; + remaining -= xdr_stream_pos(xdr); + if (remaining < xdr_align_size(args->tlen)) + return 0; - args->first.iov_base = p; - args->first.iov_len = rqstp->rq_arg.head[0].iov_len; - args->first.iov_len -= (char *)p - base; + args->first.iov_base = xdr->p; + args->first.iov_len = head->iov_len - xdr_stream_pos(xdr); - dlen = args->first.iov_len + rqstp->rq_arg.page_len + - rqstp->rq_arg.tail[0].iov_len; - if (dlen < XDR_QUADLEN(args->tlen) << 2) - return 0; return 1; } int nfs3svc_decode_mknodargs(struct svc_rqst *rqstp, __be32 *p) { + struct xdr_stream *xdr = &rqstp->rq_arg_stream; struct nfsd3_mknodargs *args = rqstp->rq_argp; - if (!(p = decode_fh(p, &args->fh)) - || !(p = decode_filename(p, &args->name, &args->len))) + if (!svcxdr_decode_diropargs3(xdr, &args->fh, &args->name, &args->len)) + return 0; + if (xdr_stream_decode_u32(xdr, &args->ftype) < 0) + return 0; + switch (args->ftype) { + case NF3CHR: + case NF3BLK: + return svcxdr_decode_devicedata3(rqstp, xdr, args); + case NF3SOCK: + case NF3FIFO: + return svcxdr_decode_sattr3(rqstp, xdr, &args->attrs); + case NF3REG: + case NF3DIR: + case NF3LNK: + /* Valid XDR but illegal file types */ + break; + default: return 0; - - args->ftype = ntohl(*p++); - - if (args->ftype == NF3BLK || args->ftype == NF3CHR - || args->ftype == NF3SOCK || args->ftype == NF3FIFO) - p = decode_sattr3(p, &args->attrs, nfsd_user_namespace(rqstp)); - - if (args->ftype == NF3BLK || args->ftype == NF3CHR) { - args->major = ntohl(*p++); - args->minor = ntohl(*p++); } - return xdr_argsize_check(rqstp, p); + return 1; } int nfs3svc_decode_renameargs(struct svc_rqst *rqstp, __be32 *p) { + struct xdr_stream *xdr = &rqstp->rq_arg_stream; struct nfsd3_renameargs *args = rqstp->rq_argp; - if (!(p = decode_fh(p, &args->ffh)) - || !(p = decode_filename(p, &args->fname, &args->flen)) - || !(p = decode_fh(p, &args->tfh)) - || !(p = decode_filename(p, &args->tname, &args->tlen))) - return 0; - - return xdr_argsize_check(rqstp, p); -} - -int -nfs3svc_decode_readlinkargs(struct svc_rqst *rqstp, __be32 *p) -{ - struct nfsd3_readlinkargs *args = rqstp->rq_argp; - - p = decode_fh(p, &args->fh); - if (!p) - return 0; - args->buffer = page_address(*(rqstp->rq_next_page++)); - - return xdr_argsize_check(rqstp, p); + return svcxdr_decode_diropargs3(xdr, &args->ffh, + &args->fname, &args->flen) && + svcxdr_decode_diropargs3(xdr, &args->tfh, + &args->tname, &args->tlen); } int nfs3svc_decode_linkargs(struct svc_rqst *rqstp, __be32 *p) { + struct xdr_stream *xdr = &rqstp->rq_arg_stream; struct nfsd3_linkargs *args = rqstp->rq_argp; - if (!(p = decode_fh(p, &args->ffh)) - || !(p = decode_fh(p, &args->tfh)) - || !(p = decode_filename(p, &args->tname, &args->tlen))) - return 0; - - return xdr_argsize_check(rqstp, p); + return svcxdr_decode_nfs_fh3(xdr, &args->ffh) && + svcxdr_decode_diropargs3(xdr, &args->tfh, + &args->tname, &args->tlen); } int nfs3svc_decode_readdirargs(struct svc_rqst *rqstp, __be32 *p) { + struct xdr_stream *xdr = &rqstp->rq_arg_stream; struct nfsd3_readdirargs *args = rqstp->rq_argp; - int len; - u32 max_blocksize = svc_max_payload(rqstp); - p = decode_fh(p, &args->fh); - if (!p) + if (!svcxdr_decode_nfs_fh3(xdr, &args->fh)) + return 0; + if (xdr_stream_decode_u64(xdr, &args->cookie) < 0) + return 0; + args->verf = xdr_inline_decode(xdr, NFS3_COOKIEVERFSIZE); + if (!args->verf) + return 0; + if (xdr_stream_decode_u32(xdr, &args->count) < 0) return 0; - p = xdr_decode_hyper(p, &args->cookie); - args->verf = p; p += 2; - args->dircount = ~0; - args->count = ntohl(*p++); - len = args->count = min_t(u32, args->count, max_blocksize); - - while (len > 0) { - struct page *p = *(rqstp->rq_next_page++); - if (!args->buffer) - args->buffer = page_address(p); - len -= PAGE_SIZE; - } - return xdr_argsize_check(rqstp, p); + return 1; } int nfs3svc_decode_readdirplusargs(struct svc_rqst *rqstp, __be32 *p) { + struct xdr_stream *xdr = &rqstp->rq_arg_stream; struct nfsd3_readdirargs *args = rqstp->rq_argp; - int len; - u32 max_blocksize = svc_max_payload(rqstp); + u32 dircount; - p = decode_fh(p, &args->fh); - if (!p) + if (!svcxdr_decode_nfs_fh3(xdr, &args->fh)) + return 0; + if (xdr_stream_decode_u64(xdr, &args->cookie) < 0) + return 0; + args->verf = xdr_inline_decode(xdr, NFS3_COOKIEVERFSIZE); + if (!args->verf) + return 0; + /* dircount is ignored */ + if (xdr_stream_decode_u32(xdr, &dircount) < 0) + return 0; + if (xdr_stream_decode_u32(xdr, &args->count) < 0) return 0; - p = xdr_decode_hyper(p, &args->cookie); - args->verf = p; p += 2; - args->dircount = ntohl(*p++); - args->count = ntohl(*p++); - - len = args->count = min(args->count, max_blocksize); - while (len > 0) { - struct page *p = *(rqstp->rq_next_page++); - if (!args->buffer) - args->buffer = page_address(p); - len -= PAGE_SIZE; - } - return xdr_argsize_check(rqstp, p); + return 1; } int nfs3svc_decode_commitargs(struct svc_rqst *rqstp, __be32 *p) { + struct xdr_stream *xdr = &rqstp->rq_arg_stream; struct nfsd3_commitargs *args = rqstp->rq_argp; - p = decode_fh(p, &args->fh); - if (!p) + + if (!svcxdr_decode_nfs_fh3(xdr, &args->fh)) + return 0; + if (xdr_stream_decode_u64(xdr, &args->offset) < 0) + return 0; + if (xdr_stream_decode_u32(xdr, &args->count) < 0) return 0; - p = xdr_decode_hyper(p, &args->offset); - args->count = ntohl(*p++); - return xdr_argsize_check(rqstp, p); + return 1; } /* diff --git a/fs/nfsd/nfs4acl.c b/fs/nfsd/nfs4acl.c index 71292a0d6f09..eaa3a0cf38f1 100644 --- a/fs/nfsd/nfs4acl.c +++ b/fs/nfsd/nfs4acl.c @@ -781,12 +781,13 @@ nfsd4_set_nfs4_acl(struct svc_rqst *rqstp, struct svc_fh *fhp, fh_lock(fhp); - host_error = set_posix_acl(inode, ACL_TYPE_ACCESS, pacl); + host_error = set_posix_acl(&init_user_ns, inode, ACL_TYPE_ACCESS, pacl); if (host_error < 0) goto out_drop_lock; if (S_ISDIR(inode->i_mode)) { - host_error = set_posix_acl(inode, ACL_TYPE_DEFAULT, dpacl); + host_error = set_posix_acl(&init_user_ns, inode, + ACL_TYPE_DEFAULT, dpacl); } out_drop_lock: diff --git a/fs/nfsd/nfs4callback.c b/fs/nfsd/nfs4callback.c index 052be5bf9ef5..7325592b456e 100644 --- a/fs/nfsd/nfs4callback.c +++ b/fs/nfsd/nfs4callback.c @@ -1189,6 +1189,7 @@ static void nfsd4_cb_done(struct rpc_task *task, void *calldata) switch (task->tk_status) { case -EIO: case -ETIMEDOUT: + case -EACCES: nfsd4_mark_cb_down(clp, task->tk_status); } break; diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index 8d6d2678abad..dd9f38d072dd 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -378,8 +378,7 @@ nfsd4_open(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, * Before RECLAIM_COMPLETE done, server should deny new lock */ if (nfsd4_has_session(cstate) && - !test_bit(NFSD4_CLIENT_RECLAIM_COMPLETE, - &cstate->session->se_client->cl_flags) && + !test_bit(NFSD4_CLIENT_RECLAIM_COMPLETE, &cstate->clp->cl_flags) && open->op_claim_type != NFS4_OPEN_CLAIM_PREVIOUS) return nfserr_grace; @@ -428,8 +427,7 @@ nfsd4_open(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, goto out; break; case NFS4_OPEN_CLAIM_PREVIOUS: - status = nfs4_check_open_reclaim(&open->op_clientid, - cstate, nn); + status = nfs4_check_open_reclaim(cstate->clp); if (status) goto out; open->op_openowner->oo_flags |= NFS4_OO_CONFIRMED; @@ -1304,7 +1302,7 @@ nfsd4_cleanup_inter_ssc(struct vfsmount *ss_mnt, struct nfsd_file *src, struct nfsd_file *dst) { nfs42_ssc_close(src->nf_file); - /* 'src' is freed by nfsd4_do_async_copy */ + fput(src->nf_file); nfsd_file_put(dst); mntput(ss_mnt); } @@ -1888,7 +1886,7 @@ nfsd4_getdeviceinfo(struct svc_rqst *rqstp, nfserr = nfs_ok; if (gdp->gd_maxcount != 0) { nfserr = ops->proc_getdeviceinfo(exp->ex_path.mnt->mnt_sb, - rqstp, cstate->session->se_client, gdp); + rqstp, cstate->clp, gdp); } gdp->gd_notify_types &= ops->notify_types; @@ -2174,7 +2172,7 @@ nfsd4_proc_null(struct svc_rqst *rqstp) static inline void nfsd4_increment_op_stats(u32 opnum) { if (opnum >= FIRST_NFS4_OP && opnum <= LAST_NFS4_OP) - nfsdstats.nfs4_opcount[opnum]++; + percpu_counter_inc(&nfsdstats.counter[NFSD_STATS_NFS4_OP(opnum)]); } static const struct nfsd4_operation nfsd4_ops[]; @@ -3305,6 +3303,7 @@ static const struct svc_procedure nfsd_procedures4[2] = { .pc_ressize = sizeof(struct nfsd_voidres), .pc_cachetype = RC_NOCACHE, .pc_xdrressize = 1, + .pc_name = "NULL", }, [NFSPROC4_COMPOUND] = { .pc_func = nfsd4_proc_compound, @@ -3315,6 +3314,7 @@ static const struct svc_procedure nfsd_procedures4[2] = { .pc_release = nfsd4_release_compoundargs, .pc_cachetype = RC_NOCACHE, .pc_xdrressize = NFSD_BUFSIZE/4, + .pc_name = "COMPOUND", }, }; diff --git a/fs/nfsd/nfs4recover.c b/fs/nfsd/nfs4recover.c index 186fa2c2c6ba..891395c6c7d3 100644 --- a/fs/nfsd/nfs4recover.c +++ b/fs/nfsd/nfs4recover.c @@ -233,7 +233,7 @@ nfsd4_create_clid_dir(struct nfs4_client *clp) * as well be forgiving and just succeed silently. */ goto out_put; - status = vfs_mkdir(d_inode(dir), dentry, S_IRWXU); + status = vfs_mkdir(&init_user_ns, d_inode(dir), dentry, S_IRWXU); out_put: dput(dentry); out_unlock: @@ -353,7 +353,7 @@ nfsd4_unlink_clid_dir(char *name, int namlen, struct nfsd_net *nn) status = -ENOENT; if (d_really_is_negative(dentry)) goto out; - status = vfs_rmdir(d_inode(dir), dentry); + status = vfs_rmdir(&init_user_ns, d_inode(dir), dentry); out: dput(dentry); out_unlock: @@ -443,7 +443,7 @@ purge_old(struct dentry *parent, struct dentry *child, struct nfsd_net *nn) if (nfs4_has_reclaimed_state(name, nn)) goto out_free; - status = vfs_rmdir(d_inode(parent), child); + status = vfs_rmdir(&init_user_ns, d_inode(parent), child); if (status) printk("failed to remove client recovery directory %pd\n", child); diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 1d2cd6a88f61..97447a64bad0 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -3891,6 +3891,7 @@ nfsd4_reclaim_complete(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, union nfsd4_op_u *u) { struct nfsd4_reclaim_complete *rc = &u->reclaim_complete; + struct nfs4_client *clp = cstate->clp; __be32 status = 0; if (rc->rca_one_fs) { @@ -3904,12 +3905,11 @@ nfsd4_reclaim_complete(struct svc_rqst *rqstp, } status = nfserr_complete_already; - if (test_and_set_bit(NFSD4_CLIENT_RECLAIM_COMPLETE, - &cstate->session->se_client->cl_flags)) + if (test_and_set_bit(NFSD4_CLIENT_RECLAIM_COMPLETE, &clp->cl_flags)) goto out; status = nfserr_stale_clientid; - if (is_client_expired(cstate->session->se_client)) + if (is_client_expired(clp)) /* * The following error isn't really legal. * But we only get here if the client just explicitly @@ -3920,8 +3920,8 @@ nfsd4_reclaim_complete(struct svc_rqst *rqstp, goto out; status = nfs_ok; - nfsd4_client_record_create(cstate->session->se_client); - inc_reclaim_complete(cstate->session->se_client); + nfsd4_client_record_create(clp); + inc_reclaim_complete(clp); out: return status; } @@ -4633,40 +4633,37 @@ static __be32 nfsd4_check_seqid(struct nfsd4_compound_state *cstate, struct nfs4 return nfserr_bad_seqid; } -static __be32 lookup_clientid(clientid_t *clid, - struct nfsd4_compound_state *cstate, - struct nfsd_net *nn, - bool sessions) +static struct nfs4_client *lookup_clientid(clientid_t *clid, bool sessions, + struct nfsd_net *nn) { struct nfs4_client *found; + spin_lock(&nn->client_lock); + found = find_confirmed_client(clid, sessions, nn); + if (found) + atomic_inc(&found->cl_rpc_users); + spin_unlock(&nn->client_lock); + return found; +} + +static __be32 set_client(clientid_t *clid, + struct nfsd4_compound_state *cstate, + struct nfsd_net *nn) +{ if (cstate->clp) { - found = cstate->clp; - if (!same_clid(&found->cl_clientid, clid)) + if (!same_clid(&cstate->clp->cl_clientid, clid)) return nfserr_stale_clientid; return nfs_ok; } - if (STALE_CLIENTID(clid, nn)) return nfserr_stale_clientid; - /* - * For v4.1+ we get the client in the SEQUENCE op. If we don't have one - * cached already then we know this is for is for v4.0 and "sessions" - * will be false. + * We're in the 4.0 case (otherwise the SEQUENCE op would have + * set cstate->clp), so session = false: */ - WARN_ON_ONCE(cstate->session); - spin_lock(&nn->client_lock); - found = find_confirmed_client(clid, sessions, nn); - if (!found) { - spin_unlock(&nn->client_lock); + cstate->clp = lookup_clientid(clid, false, nn); + if (!cstate->clp) return nfserr_expired; - } - atomic_inc(&found->cl_rpc_users); - spin_unlock(&nn->client_lock); - - /* Cache the nfs4_client in cstate! */ - cstate->clp = found; return nfs_ok; } @@ -4680,8 +4677,6 @@ nfsd4_process_open1(struct nfsd4_compound_state *cstate, struct nfs4_openowner *oo = NULL; __be32 status; - if (STALE_CLIENTID(&open->op_clientid, nn)) - return nfserr_stale_clientid; /* * In case we need it later, after we've already created the * file and don't want to risk a further failure: @@ -4690,7 +4685,7 @@ nfsd4_process_open1(struct nfsd4_compound_state *cstate, if (open->op_file == NULL) return nfserr_jukebox; - status = lookup_clientid(clientid, cstate, nn, false); + status = set_client(clientid, cstate, nn); if (status) return status; clp = cstate->clp; @@ -4945,31 +4940,6 @@ static struct file_lock *nfs4_alloc_init_lease(struct nfs4_delegation *dp, return fl; } -static int nfsd4_check_conflicting_opens(struct nfs4_client *clp, - struct nfs4_file *fp) -{ - struct nfs4_clnt_odstate *co; - struct file *f = fp->fi_deleg_file->nf_file; - struct inode *ino = locks_inode(f); - int writes = atomic_read(&ino->i_writecount); - - if (fp->fi_fds[O_WRONLY]) - writes--; - if (fp->fi_fds[O_RDWR]) - writes--; - if (writes > 0) - return -EAGAIN; - spin_lock(&fp->fi_lock); - list_for_each_entry(co, &fp->fi_clnt_odstate, co_perfile) { - if (co->co_client != clp) { - spin_unlock(&fp->fi_lock); - return -EAGAIN; - } - } - spin_unlock(&fp->fi_lock); - return 0; -} - static struct nfs4_delegation * nfs4_set_delegation(struct nfs4_client *clp, struct svc_fh *fh, struct nfs4_file *fp, struct nfs4_clnt_odstate *odstate) @@ -4989,12 +4959,9 @@ nfs4_set_delegation(struct nfs4_client *clp, struct svc_fh *fh, nf = find_readable_file(fp); if (!nf) { - /* - * We probably could attempt another open and get a read - * delegation, but for now, don't bother until the - * client actually sends us one. - */ - return ERR_PTR(-EAGAIN); + /* We should always have a readable file here */ + WARN_ON_ONCE(1); + return ERR_PTR(-EBADF); } spin_lock(&state_lock); spin_lock(&fp->fi_lock); @@ -5024,19 +4991,11 @@ nfs4_set_delegation(struct nfs4_client *clp, struct svc_fh *fh, if (!fl) goto out_clnt_odstate; - status = nfsd4_check_conflicting_opens(clp, fp); - if (status) { - locks_free_lock(fl); - goto out_clnt_odstate; - } status = vfs_setlease(fp->fi_deleg_file->nf_file, fl->fl_type, &fl, NULL); if (fl) locks_free_lock(fl); if (status) goto out_clnt_odstate; - status = nfsd4_check_conflicting_opens(clp, fp); - if (status) - goto out_clnt_odstate; spin_lock(&state_lock); spin_lock(&fp->fi_lock); @@ -5118,6 +5077,17 @@ nfs4_open_delegation(struct svc_fh *fh, struct nfsd4_open *open, goto out_no_deleg; if (!cb_up || !(oo->oo_flags & NFS4_OO_CONFIRMED)) goto out_no_deleg; + /* + * Also, if the file was opened for write or + * create, there's a good chance the client's + * about to write to it, resulting in an + * immediate recall (since we don't support + * write delegations): + */ + if (open->op_share_access & NFS4_SHARE_ACCESS_WRITE) + goto out_no_deleg; + if (open->op_create == NFS4_OPEN_CREATE) + goto out_no_deleg; break; default: goto out_no_deleg; @@ -5300,17 +5270,14 @@ nfsd4_renew(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, struct nfsd_net *nn = net_generic(SVC_NET(rqstp), nfsd_net_id); trace_nfsd_clid_renew(clid); - status = lookup_clientid(clid, cstate, nn, false); + status = set_client(clid, cstate, nn); if (status) - goto out; + return status; clp = cstate->clp; - status = nfserr_cb_path_down; if (!list_empty(&clp->cl_delegations) && clp->cl_cb_state != NFSD4_CB_UP) - goto out; - status = nfs_ok; -out: - return status; + return nfserr_cb_path_down; + return nfs_ok; } void @@ -5397,7 +5364,7 @@ nfs4_laundromat(struct nfsd_net *nn) idr_for_each_entry(&nn->s2s_cp_stateids, cps_t, i) { cps = container_of(cps_t, struct nfs4_cpntf_state, cp_stateid); if (cps->cp_stateid.sc_type == NFS4_COPYNOTIFY_STID && - cps->cpntf_time > cutoff) + cps->cpntf_time < cutoff) _free_cpntf_state_locked(nn, cps); } spin_unlock(&nn->s2s_cp_lock); @@ -5686,8 +5653,7 @@ nfsd4_lookup_stateid(struct nfsd4_compound_state *cstate, if (ZERO_STATEID(stateid) || ONE_STATEID(stateid) || CLOSE_STATEID(stateid)) return nfserr_bad_stateid; - status = lookup_clientid(&stateid->si_opaque.so_clid, cstate, nn, - false); + status = set_client(&stateid->si_opaque.so_clid, cstate, nn); if (status == nfserr_stale_clientid) { if (cstate->session) return nfserr_bad_stateid; @@ -5818,21 +5784,27 @@ static __be32 find_cpntf_state(struct nfsd_net *nn, stateid_t *st, { __be32 status; struct nfs4_cpntf_state *cps = NULL; - struct nfsd4_compound_state cstate; + struct nfs4_client *found; status = manage_cpntf_state(nn, st, NULL, &cps); if (status) return status; cps->cpntf_time = ktime_get_boottime_seconds(); - memset(&cstate, 0, sizeof(cstate)); - status = lookup_clientid(&cps->cp_p_clid, &cstate, nn, true); - if (status) + + status = nfserr_expired; + found = lookup_clientid(&cps->cp_p_clid, true, nn); + if (!found) goto out; - status = nfsd4_lookup_stateid(&cstate, &cps->cp_p_stateid, - NFS4_DELEG_STID|NFS4_OPEN_STID|NFS4_LOCK_STID, - stid, nn); - put_client_renew(cstate.clp); + + *stid = find_stateid_by_type(found, &cps->cp_p_stateid, + NFS4_DELEG_STID|NFS4_OPEN_STID|NFS4_LOCK_STID); + if (*stid) + status = nfs_ok; + else + status = nfserr_bad_stateid; + + put_client_renew(found); out: nfs4_put_cpntf_state(nn, cps); return status; @@ -5921,7 +5893,7 @@ nfsd4_test_stateid(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, { struct nfsd4_test_stateid *test_stateid = &u->test_stateid; struct nfsd4_test_stateid_id *stateid; - struct nfs4_client *cl = cstate->session->se_client; + struct nfs4_client *cl = cstate->clp; list_for_each_entry(stateid, &test_stateid->ts_stateid_list, ts_id_list) stateid->ts_id_status = @@ -5967,7 +5939,7 @@ nfsd4_free_stateid(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, stateid_t *stateid = &free_stateid->fr_stateid; struct nfs4_stid *s; struct nfs4_delegation *dp; - struct nfs4_client *cl = cstate->session->se_client; + struct nfs4_client *cl = cstate->clp; __be32 ret = nfserr_bad_stateid; spin_lock(&cl->cl_lock); @@ -6696,13 +6668,9 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, if (nfsd4_has_session(cstate)) /* See rfc 5661 18.10.3: given clientid is ignored: */ memcpy(&lock->lk_new_clientid, - &cstate->session->se_client->cl_clientid, + &cstate->clp->cl_clientid, sizeof(clientid_t)); - status = nfserr_stale_clientid; - if (STALE_CLIENTID(&lock->lk_new_clientid, nn)) - goto out; - /* validate and update open stateid and open seqid */ status = nfs4_preprocess_confirmed_seqid_op(cstate, lock->lk_new_open_seqid, @@ -6909,8 +6877,7 @@ nfsd4_lockt(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, return nfserr_inval; if (!nfsd4_has_session(cstate)) { - status = lookup_clientid(&lockt->lt_clientid, cstate, nn, - false); + status = set_client(&lockt->lt_clientid, cstate, nn); if (status) goto out; } @@ -7094,7 +7061,7 @@ nfsd4_release_lockowner(struct svc_rqst *rqstp, dprintk("nfsd4_release_lockowner clientid: (%08x/%08x):\n", clid->cl_boot, clid->cl_id); - status = lookup_clientid(clid, cstate, nn, false); + status = set_client(clid, cstate, nn); if (status) return status; @@ -7230,25 +7197,13 @@ nfsd4_find_reclaim_client(struct xdr_netobj name, struct nfsd_net *nn) return NULL; } -/* -* Called from OPEN. Look for clientid in reclaim list. -*/ __be32 -nfs4_check_open_reclaim(clientid_t *clid, - struct nfsd4_compound_state *cstate, - struct nfsd_net *nn) +nfs4_check_open_reclaim(struct nfs4_client *clp) { - __be32 status; - - /* find clientid in conf_id_hashtbl */ - status = lookup_clientid(clid, cstate, nn, false); - if (status) - return nfserr_reclaim_bad; - - if (test_bit(NFSD4_CLIENT_RECLAIM_COMPLETE, &cstate->clp->cl_flags)) + if (test_bit(NFSD4_CLIENT_RECLAIM_COMPLETE, &clp->cl_flags)) return nfserr_no_grace; - if (nfsd4_client_record_check(cstate->clp)) + if (nfsd4_client_record_check(clp)) return nfserr_reclaim_bad; return nfs_ok; diff --git a/fs/nfsd/nfscache.c b/fs/nfsd/nfscache.c index 80c90fc231a5..96cdf77925f3 100644 --- a/fs/nfsd/nfscache.c +++ b/fs/nfsd/nfscache.c @@ -121,14 +121,14 @@ nfsd_reply_cache_free_locked(struct nfsd_drc_bucket *b, struct svc_cacherep *rp, struct nfsd_net *nn) { if (rp->c_type == RC_REPLBUFF && rp->c_replvec.iov_base) { - nn->drc_mem_usage -= rp->c_replvec.iov_len; + nfsd_stats_drc_mem_usage_sub(nn, rp->c_replvec.iov_len); kfree(rp->c_replvec.iov_base); } if (rp->c_state != RC_UNUSED) { rb_erase(&rp->c_node, &b->rb_head); list_del(&rp->c_lru); atomic_dec(&nn->num_drc_entries); - nn->drc_mem_usage -= sizeof(*rp); + nfsd_stats_drc_mem_usage_sub(nn, sizeof(*rp)); } kmem_cache_free(drc_slab, rp); } @@ -154,6 +154,16 @@ void nfsd_drc_slab_free(void) kmem_cache_destroy(drc_slab); } +static int nfsd_reply_cache_stats_init(struct nfsd_net *nn) +{ + return nfsd_percpu_counters_init(nn->counter, NFSD_NET_COUNTERS_NUM); +} + +static void nfsd_reply_cache_stats_destroy(struct nfsd_net *nn) +{ + nfsd_percpu_counters_destroy(nn->counter, NFSD_NET_COUNTERS_NUM); +} + int nfsd_reply_cache_init(struct nfsd_net *nn) { unsigned int hashsize; @@ -165,12 +175,16 @@ int nfsd_reply_cache_init(struct nfsd_net *nn) hashsize = nfsd_hashsize(nn->max_drc_entries); nn->maskbits = ilog2(hashsize); + status = nfsd_reply_cache_stats_init(nn); + if (status) + goto out_nomem; + nn->nfsd_reply_cache_shrinker.scan_objects = nfsd_reply_cache_scan; nn->nfsd_reply_cache_shrinker.count_objects = nfsd_reply_cache_count; nn->nfsd_reply_cache_shrinker.seeks = 1; status = register_shrinker(&nn->nfsd_reply_cache_shrinker); if (status) - goto out_nomem; + goto out_stats_destroy; nn->drc_hashtbl = kvzalloc(array_size(hashsize, sizeof(*nn->drc_hashtbl)), GFP_KERNEL); @@ -186,6 +200,8 @@ int nfsd_reply_cache_init(struct nfsd_net *nn) return 0; out_shrinker: unregister_shrinker(&nn->nfsd_reply_cache_shrinker); +out_stats_destroy: + nfsd_reply_cache_stats_destroy(nn); out_nomem: printk(KERN_ERR "nfsd: failed to allocate reply cache\n"); return -ENOMEM; @@ -196,6 +212,7 @@ void nfsd_reply_cache_shutdown(struct nfsd_net *nn) struct svc_cacherep *rp; unsigned int i; + nfsd_reply_cache_stats_destroy(nn); unregister_shrinker(&nn->nfsd_reply_cache_shrinker); for (i = 0; i < nn->drc_hashsize; i++) { @@ -324,7 +341,7 @@ nfsd_cache_key_cmp(const struct svc_cacherep *key, { if (key->c_key.k_xid == rp->c_key.k_xid && key->c_key.k_csum != rp->c_key.k_csum) { - ++nn->payload_misses; + nfsd_stats_payload_misses_inc(nn); trace_nfsd_drc_mismatch(nn, key, rp); } @@ -407,7 +424,7 @@ int nfsd_cache_lookup(struct svc_rqst *rqstp) rqstp->rq_cacherep = NULL; if (type == RC_NOCACHE) { - nfsdstats.rcnocache++; + nfsd_stats_rc_nocache_inc(); goto out; } @@ -429,12 +446,12 @@ int nfsd_cache_lookup(struct svc_rqst *rqstp) goto found_entry; } - nfsdstats.rcmisses++; + nfsd_stats_rc_misses_inc(); rqstp->rq_cacherep = rp; rp->c_state = RC_INPROG; atomic_inc(&nn->num_drc_entries); - nn->drc_mem_usage += sizeof(*rp); + nfsd_stats_drc_mem_usage_add(nn, sizeof(*rp)); /* go ahead and prune the cache */ prune_bucket(b, nn); @@ -446,7 +463,7 @@ out: found_entry: /* We found a matching entry which is either in progress or done. */ - nfsdstats.rchits++; + nfsd_stats_rc_hits_inc(); rtn = RC_DROPIT; /* Request being processed */ @@ -548,7 +565,7 @@ void nfsd_cache_update(struct svc_rqst *rqstp, int cachetype, __be32 *statp) return; } spin_lock(&b->cache_lock); - nn->drc_mem_usage += bufsize; + nfsd_stats_drc_mem_usage_add(nn, bufsize); lru_put_end(b, rp); rp->c_secure = test_bit(RQ_SECURE, &rqstp->rq_flags); rp->c_type = cachetype; @@ -588,13 +605,18 @@ static int nfsd_reply_cache_stats_show(struct seq_file *m, void *v) seq_printf(m, "max entries: %u\n", nn->max_drc_entries); seq_printf(m, "num entries: %u\n", - atomic_read(&nn->num_drc_entries)); + atomic_read(&nn->num_drc_entries)); seq_printf(m, "hash buckets: %u\n", 1 << nn->maskbits); - seq_printf(m, "mem usage: %u\n", nn->drc_mem_usage); - seq_printf(m, "cache hits: %u\n", nfsdstats.rchits); - seq_printf(m, "cache misses: %u\n", nfsdstats.rcmisses); - seq_printf(m, "not cached: %u\n", nfsdstats.rcnocache); - seq_printf(m, "payload misses: %u\n", nn->payload_misses); + seq_printf(m, "mem usage: %lld\n", + percpu_counter_sum_positive(&nn->counter[NFSD_NET_DRC_MEM_USAGE])); + seq_printf(m, "cache hits: %lld\n", + percpu_counter_sum_positive(&nfsdstats.counter[NFSD_STATS_RC_HITS])); + seq_printf(m, "cache misses: %lld\n", + percpu_counter_sum_positive(&nfsdstats.counter[NFSD_STATS_RC_MISSES])); + seq_printf(m, "not cached: %lld\n", + percpu_counter_sum_positive(&nfsdstats.counter[NFSD_STATS_RC_NOCACHE])); + seq_printf(m, "payload misses: %lld\n", + percpu_counter_sum_positive(&nn->counter[NFSD_NET_PAYLOAD_MISSES])); seq_printf(m, "longest chain len: %u\n", nn->longest_chain); seq_printf(m, "cachesize at longest: %u\n", nn->longest_chain_cachesize); return 0; diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c index f6d5d783f4a4..ef86ed23af82 100644 --- a/fs/nfsd/nfsctl.c +++ b/fs/nfsd/nfsctl.c @@ -32,6 +32,7 @@ enum { NFSD_Root = 1, NFSD_List, + NFSD_Export_Stats, NFSD_Export_features, NFSD_Fh, NFSD_FO_UnlockIP, @@ -1348,6 +1349,8 @@ static int nfsd_fill_super(struct super_block *sb, struct fs_context *fc) static const struct tree_descr nfsd_files[] = { [NFSD_List] = {"exports", &exports_nfsd_operations, S_IRUGO}, + /* Per-export io stats use same ops as exports file */ + [NFSD_Export_Stats] = {"export_stats", &exports_nfsd_operations, S_IRUGO}, [NFSD_Export_features] = {"export_features", &export_features_operations, S_IRUGO}, [NFSD_FO_UnlockIP] = {"unlock_ip", @@ -1522,19 +1525,18 @@ static int __init init_nfsd(void) int retval; printk(KERN_INFO "Installing knfsd (copyright (C) 1996 okir@monad.swb.de).\n"); - retval = register_pernet_subsys(&nfsd_net_ops); - if (retval < 0) - return retval; retval = register_cld_notifier(); if (retval) - goto out_unregister_pernet; + return retval; retval = nfsd4_init_slabs(); if (retval) goto out_unregister_notifier; retval = nfsd4_init_pnfs(); if (retval) goto out_free_slabs; - nfsd_stat_init(); /* Statistics */ + retval = nfsd_stat_init(); /* Statistics */ + if (retval) + goto out_free_pnfs; retval = nfsd_drc_slab_create(); if (retval) goto out_free_stat; @@ -1544,9 +1546,14 @@ static int __init init_nfsd(void) goto out_free_lockd; retval = register_filesystem(&nfsd_fs_type); if (retval) + goto out_free_exports; + retval = register_pernet_subsys(&nfsd_net_ops); + if (retval < 0) goto out_free_all; return 0; out_free_all: + unregister_pernet_subsys(&nfsd_net_ops); +out_free_exports: remove_proc_entry("fs/nfs/exports", NULL); remove_proc_entry("fs/nfs", NULL); out_free_lockd: @@ -1554,18 +1561,18 @@ out_free_lockd: nfsd_drc_slab_free(); out_free_stat: nfsd_stat_shutdown(); +out_free_pnfs: nfsd4_exit_pnfs(); out_free_slabs: nfsd4_free_slabs(); out_unregister_notifier: unregister_cld_notifier(); -out_unregister_pernet: - unregister_pernet_subsys(&nfsd_net_ops); return retval; } static void __exit exit_nfsd(void) { + unregister_pernet_subsys(&nfsd_net_ops); nfsd_drc_slab_free(); remove_proc_entry("fs/nfs/exports", NULL); remove_proc_entry("fs/nfs", NULL); @@ -1575,7 +1582,6 @@ static void __exit exit_nfsd(void) nfsd4_exit_pnfs(); unregister_filesystem(&nfsd_fs_type); unregister_cld_notifier(); - unregister_pernet_subsys(&nfsd_net_ops); } MODULE_AUTHOR("Olaf Kirch <okir@monad.swb.de>"); diff --git a/fs/nfsd/nfsd.h b/fs/nfsd/nfsd.h index d63cf8196fed..8bdc37aa2c2e 100644 --- a/fs/nfsd/nfsd.h +++ b/fs/nfsd/nfsd.h @@ -24,8 +24,8 @@ #include <uapi/linux/nfsd/debug.h> #include "netns.h" -#include "stats.h" #include "export.h" +#include "stats.h" #undef ifdebug #ifdef CONFIG_SUNRPC_DEBUG diff --git a/fs/nfsd/nfsfh.c b/fs/nfsd/nfsfh.c index 66f2ef67792a..10b44421eace 100644 --- a/fs/nfsd/nfsfh.c +++ b/fs/nfsd/nfsfh.c @@ -40,7 +40,8 @@ static int nfsd_acceptable(void *expv, struct dentry *dentry) /* make sure parents give x permission to user */ int err; parent = dget_parent(tdentry); - err = inode_permission(d_inode(parent), MAY_EXEC); + err = inode_permission(&init_user_ns, + d_inode(parent), MAY_EXEC); if (err < 0) { dput(parent); break; @@ -349,7 +350,7 @@ out: __be32 fh_verify(struct svc_rqst *rqstp, struct svc_fh *fhp, umode_t type, int access) { - struct svc_export *exp; + struct svc_export *exp = NULL; struct dentry *dentry; __be32 error; @@ -422,7 +423,7 @@ skip_pseudoflavor_check: } out: if (error == nfserr_stale) - nfsdstats.fh_stale++; + nfsd_stats_fh_stale_inc(exp); return error; } diff --git a/fs/nfsd/nfsfh.h b/fs/nfsd/nfsfh.h index cb20c2cd3469..f58933519f38 100644 --- a/fs/nfsd/nfsfh.h +++ b/fs/nfsd/nfsfh.h @@ -12,6 +12,7 @@ #include <linux/sunrpc/svc.h> #include <uapi/linux/nfsd/nfsfh.h> #include <linux/iversion.h> +#include <linux/exportfs.h> static inline __u32 ino_t_to_u32(ino_t ino) { @@ -264,7 +265,9 @@ fh_clear_wcc(struct svc_fh *fhp) static inline u64 nfsd4_change_attribute(struct kstat *stat, struct inode *inode) { - if (IS_I_VERSION(inode)) { + if (inode->i_sb->s_export_op->fetch_iversion) + return inode->i_sb->s_export_op->fetch_iversion(inode); + else if (IS_I_VERSION(inode)) { u64 chattr; chattr = stat->ctime.tv_sec; diff --git a/fs/nfsd/nfsproc.c b/fs/nfsd/nfsproc.c index 9473d048efec..a8d5449dd0e9 100644 --- a/fs/nfsd/nfsproc.c +++ b/fs/nfsd/nfsproc.c @@ -90,7 +90,7 @@ nfsd_proc_setattr(struct svc_rqst *rqstp) if (delta < 0) delta = -delta; if (delta < MAX_TOUCH_TIME_ERROR && - setattr_prepare(fhp->fh_dentry, iap) != 0) { + setattr_prepare(&init_user_ns, fhp->fh_dentry, iap) != 0) { /* * Turn off ATTR_[AM]TIME_SET but leave ATTR_[AM]TIME. * This will cause notify_change to set these times @@ -149,14 +149,15 @@ out: static __be32 nfsd_proc_readlink(struct svc_rqst *rqstp) { - struct nfsd_readlinkargs *argp = rqstp->rq_argp; + struct nfsd_fhandle *argp = rqstp->rq_argp; struct nfsd_readlinkres *resp = rqstp->rq_resp; + char *buffer = page_address(*(rqstp->rq_next_page++)); dprintk("nfsd: READLINK %s\n", SVCFH_fmt(&argp->fh)); /* Read the symlink. */ resp->len = NFS_MAXPATHLEN; - resp->status = nfsd_readlink(rqstp, &argp->fh, argp->buffer, &resp->len); + resp->status = nfsd_readlink(rqstp, &argp->fh, buffer, &resp->len); fh_put(&argp->fh); return rpc_success; @@ -171,32 +172,36 @@ nfsd_proc_read(struct svc_rqst *rqstp) { struct nfsd_readargs *argp = rqstp->rq_argp; struct nfsd_readres *resp = rqstp->rq_resp; + unsigned int len; u32 eof; + int v; dprintk("nfsd: READ %s %d bytes at %d\n", SVCFH_fmt(&argp->fh), argp->count, argp->offset); + argp->count = min_t(u32, argp->count, NFSSVC_MAXBLKSIZE_V2); + + v = 0; + len = argp->count; + while (len > 0) { + struct page *page = *(rqstp->rq_next_page++); + + rqstp->rq_vec[v].iov_base = page_address(page); + rqstp->rq_vec[v].iov_len = min_t(unsigned int, len, PAGE_SIZE); + len -= rqstp->rq_vec[v].iov_len; + v++; + } + /* Obtain buffer pointer for payload. 19 is 1 word for * status, 17 words for fattr, and 1 word for the byte count. */ - - if (NFSSVC_MAXBLKSIZE_V2 < argp->count) { - char buf[RPC_MAX_ADDRBUFLEN]; - printk(KERN_NOTICE - "oversized read request from %s (%d bytes)\n", - svc_print_addr(rqstp, buf, sizeof(buf)), - argp->count); - argp->count = NFSSVC_MAXBLKSIZE_V2; - } svc_reserve_auth(rqstp, (19<<2) + argp->count + 4); resp->count = argp->count; - resp->status = nfsd_read(rqstp, fh_copy(&resp->fh, &argp->fh), - argp->offset, - rqstp->rq_vec, argp->vlen, - &resp->count, - &eof); + fh_copy(&resp->fh, &argp->fh); + resp->status = nfsd_read(rqstp, &resp->fh, argp->offset, + rqstp->rq_vec, v, &resp->count, &eof); if (resp->status == nfs_ok) resp->status = fh_getattr(&resp->fh, &resp->stat); else if (resp->status == nfserr_jukebox) @@ -548,6 +553,20 @@ nfsd_proc_rmdir(struct svc_rqst *rqstp) return rpc_success; } +static void nfsd_init_dirlist_pages(struct svc_rqst *rqstp, + struct nfsd_readdirres *resp, + int count) +{ + count = min_t(u32, count, PAGE_SIZE); + + /* Convert byte count to number of words (i.e. >> 2), + * and reserve room for the NULL ptr & eof flag (-2 words) */ + resp->buflen = (count >> 2) - 2; + + resp->buffer = page_address(*rqstp->rq_next_page); + rqstp->rq_next_page++; +} + /* * Read a portion of a directory. */ @@ -556,31 +575,24 @@ nfsd_proc_readdir(struct svc_rqst *rqstp) { struct nfsd_readdirargs *argp = rqstp->rq_argp; struct nfsd_readdirres *resp = rqstp->rq_resp; - int count; loff_t offset; + __be32 *buffer; dprintk("nfsd: READDIR %s %d bytes at %d\n", SVCFH_fmt(&argp->fh), argp->count, argp->cookie); - /* Shrink to the client read size */ - count = (argp->count >> 2) - 2; - - /* Make sure we've room for the NULL ptr & eof flag */ - count -= 2; - if (count < 0) - count = 0; + nfsd_init_dirlist_pages(rqstp, resp, argp->count); + buffer = resp->buffer; - resp->buffer = argp->buffer; resp->offset = NULL; - resp->buflen = count; resp->common.err = nfs_ok; /* Read directory and encode entries on the fly */ offset = argp->cookie; resp->status = nfsd_readdir(rqstp, &argp->fh, &offset, &resp->common, nfssvc_encode_entry); - resp->count = resp->buffer - argp->buffer; + resp->count = resp->buffer - buffer; if (resp->offset) *resp->offset = htonl(offset); @@ -623,16 +635,18 @@ static const struct svc_procedure nfsd_procedures2[18] = { .pc_ressize = sizeof(struct nfsd_voidres), .pc_cachetype = RC_NOCACHE, .pc_xdrressize = 0, + .pc_name = "NULL", }, [NFSPROC_GETATTR] = { .pc_func = nfsd_proc_getattr, - .pc_decode = nfssvc_decode_fhandle, + .pc_decode = nfssvc_decode_fhandleargs, .pc_encode = nfssvc_encode_attrstat, .pc_release = nfssvc_release_attrstat, .pc_argsize = sizeof(struct nfsd_fhandle), .pc_ressize = sizeof(struct nfsd_attrstat), .pc_cachetype = RC_NOCACHE, .pc_xdrressize = ST+AT, + .pc_name = "GETATTR", }, [NFSPROC_SETATTR] = { .pc_func = nfsd_proc_setattr, @@ -643,6 +657,7 @@ static const struct svc_procedure nfsd_procedures2[18] = { .pc_ressize = sizeof(struct nfsd_attrstat), .pc_cachetype = RC_REPLBUFF, .pc_xdrressize = ST+AT, + .pc_name = "SETATTR", }, [NFSPROC_ROOT] = { .pc_func = nfsd_proc_root, @@ -652,6 +667,7 @@ static const struct svc_procedure nfsd_procedures2[18] = { .pc_ressize = sizeof(struct nfsd_voidres), .pc_cachetype = RC_NOCACHE, .pc_xdrressize = 0, + .pc_name = "ROOT", }, [NFSPROC_LOOKUP] = { .pc_func = nfsd_proc_lookup, @@ -662,15 +678,17 @@ static const struct svc_procedure nfsd_procedures2[18] = { .pc_ressize = sizeof(struct nfsd_diropres), .pc_cachetype = RC_NOCACHE, .pc_xdrressize = ST+FH+AT, + .pc_name = "LOOKUP", }, [NFSPROC_READLINK] = { .pc_func = nfsd_proc_readlink, - .pc_decode = nfssvc_decode_readlinkargs, + .pc_decode = nfssvc_decode_fhandleargs, .pc_encode = nfssvc_encode_readlinkres, - .pc_argsize = sizeof(struct nfsd_readlinkargs), + .pc_argsize = sizeof(struct nfsd_fhandle), .pc_ressize = sizeof(struct nfsd_readlinkres), .pc_cachetype = RC_NOCACHE, .pc_xdrressize = ST+1+NFS_MAXPATHLEN/4, + .pc_name = "READLINK", }, [NFSPROC_READ] = { .pc_func = nfsd_proc_read, @@ -681,6 +699,7 @@ static const struct svc_procedure nfsd_procedures2[18] = { .pc_ressize = sizeof(struct nfsd_readres), .pc_cachetype = RC_NOCACHE, .pc_xdrressize = ST+AT+1+NFSSVC_MAXBLKSIZE_V2/4, + .pc_name = "READ", }, [NFSPROC_WRITECACHE] = { .pc_func = nfsd_proc_writecache, @@ -690,6 +709,7 @@ static const struct svc_procedure nfsd_procedures2[18] = { .pc_ressize = sizeof(struct nfsd_voidres), .pc_cachetype = RC_NOCACHE, .pc_xdrressize = 0, + .pc_name = "WRITECACHE", }, [NFSPROC_WRITE] = { .pc_func = nfsd_proc_write, @@ -700,6 +720,7 @@ static const struct svc_procedure nfsd_procedures2[18] = { .pc_ressize = sizeof(struct nfsd_attrstat), .pc_cachetype = RC_REPLBUFF, .pc_xdrressize = ST+AT, + .pc_name = "WRITE", }, [NFSPROC_CREATE] = { .pc_func = nfsd_proc_create, @@ -710,6 +731,7 @@ static const struct svc_procedure nfsd_procedures2[18] = { .pc_ressize = sizeof(struct nfsd_diropres), .pc_cachetype = RC_REPLBUFF, .pc_xdrressize = ST+FH+AT, + .pc_name = "CREATE", }, [NFSPROC_REMOVE] = { .pc_func = nfsd_proc_remove, @@ -719,6 +741,7 @@ static const struct svc_procedure nfsd_procedures2[18] = { .pc_ressize = sizeof(struct nfsd_stat), .pc_cachetype = RC_REPLSTAT, .pc_xdrressize = ST, + .pc_name = "REMOVE", }, [NFSPROC_RENAME] = { .pc_func = nfsd_proc_rename, @@ -728,6 +751,7 @@ static const struct svc_procedure nfsd_procedures2[18] = { .pc_ressize = sizeof(struct nfsd_stat), .pc_cachetype = RC_REPLSTAT, .pc_xdrressize = ST, + .pc_name = "RENAME", }, [NFSPROC_LINK] = { .pc_func = nfsd_proc_link, @@ -737,6 +761,7 @@ static const struct svc_procedure nfsd_procedures2[18] = { .pc_ressize = sizeof(struct nfsd_stat), .pc_cachetype = RC_REPLSTAT, .pc_xdrressize = ST, + .pc_name = "LINK", }, [NFSPROC_SYMLINK] = { .pc_func = nfsd_proc_symlink, @@ -746,6 +771,7 @@ static const struct svc_procedure nfsd_procedures2[18] = { .pc_ressize = sizeof(struct nfsd_stat), .pc_cachetype = RC_REPLSTAT, .pc_xdrressize = ST, + .pc_name = "SYMLINK", }, [NFSPROC_MKDIR] = { .pc_func = nfsd_proc_mkdir, @@ -756,6 +782,7 @@ static const struct svc_procedure nfsd_procedures2[18] = { .pc_ressize = sizeof(struct nfsd_diropres), .pc_cachetype = RC_REPLBUFF, .pc_xdrressize = ST+FH+AT, + .pc_name = "MKDIR", }, [NFSPROC_RMDIR] = { .pc_func = nfsd_proc_rmdir, @@ -765,6 +792,7 @@ static const struct svc_procedure nfsd_procedures2[18] = { .pc_ressize = sizeof(struct nfsd_stat), .pc_cachetype = RC_REPLSTAT, .pc_xdrressize = ST, + .pc_name = "RMDIR", }, [NFSPROC_READDIR] = { .pc_func = nfsd_proc_readdir, @@ -773,15 +801,17 @@ static const struct svc_procedure nfsd_procedures2[18] = { .pc_argsize = sizeof(struct nfsd_readdirargs), .pc_ressize = sizeof(struct nfsd_readdirres), .pc_cachetype = RC_NOCACHE, + .pc_name = "READDIR", }, [NFSPROC_STATFS] = { .pc_func = nfsd_proc_statfs, - .pc_decode = nfssvc_decode_fhandle, + .pc_decode = nfssvc_decode_fhandleargs, .pc_encode = nfssvc_encode_statfsres, .pc_argsize = sizeof(struct nfsd_fhandle), .pc_ressize = sizeof(struct nfsd_statfsres), .pc_cachetype = RC_NOCACHE, .pc_xdrressize = ST+5, + .pc_name = "STATFS", }, }; diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c index f9c9f4c63cc7..6de406322106 100644 --- a/fs/nfsd/nfssvc.c +++ b/fs/nfsd/nfssvc.c @@ -955,37 +955,6 @@ out: return 0; } -/* - * A write procedure can have a large argument, and a read procedure can - * have a large reply, but no NFSv2 or NFSv3 procedure has argument and - * reply that can both be larger than a page. The xdr code has taken - * advantage of this assumption to be a sloppy about bounds checking in - * some cases. Pending a rewrite of the NFSv2/v3 xdr code to fix that - * problem, we enforce these assumptions here: - */ -static bool nfs_request_too_big(struct svc_rqst *rqstp, - const struct svc_procedure *proc) -{ - /* - * The ACL code has more careful bounds-checking and is not - * susceptible to this problem: - */ - if (rqstp->rq_prog != NFS_PROGRAM) - return false; - /* - * Ditto NFSv4 (which can in theory have argument and reply both - * more than a page): - */ - if (rqstp->rq_vers >= 4) - return false; - /* The reply will be small, we're OK: */ - if (proc->pc_xdrressize > 0 && - proc->pc_xdrressize < XDR_QUADLEN(PAGE_SIZE)) - return false; - - return rqstp->rq_arg.len > PAGE_SIZE; -} - /** * nfsd_dispatch - Process an NFS or NFSACL Request * @rqstp: incoming request @@ -1004,9 +973,6 @@ int nfsd_dispatch(struct svc_rqst *rqstp, __be32 *statp) struct kvec *resv = &rqstp->rq_res.head[0]; __be32 *p; - if (nfs_request_too_big(rqstp, proc)) - goto out_decode_err; - /* * Give the xdr decoder a chance to change this if it wants * (necessary in the NFSv4.0 compound case) diff --git a/fs/nfsd/nfsxdr.c b/fs/nfsd/nfsxdr.c index 7aa6e8aca2c1..5d79ef6a0c7f 100644 --- a/fs/nfsd/nfsxdr.c +++ b/fs/nfsd/nfsxdr.c @@ -23,24 +23,31 @@ static u32 nfs_ftypes[] = { /* - * XDR functions for basic NFS types + * Basic NFSv2 data types (RFC 1094 Section 2.3) */ -static __be32 * -decode_fh(__be32 *p, struct svc_fh *fhp) + +/** + * svcxdr_decode_fhandle - Decode an NFSv2 file handle + * @xdr: XDR stream positioned at an encoded NFSv2 FH + * @fhp: OUT: filled-in server file handle + * + * Return values: + * %false: The encoded file handle was not valid + * %true: @fhp has been initialized + */ +bool +svcxdr_decode_fhandle(struct xdr_stream *xdr, struct svc_fh *fhp) { + __be32 *p; + + p = xdr_inline_decode(xdr, NFS_FHSIZE); + if (!p) + return false; fh_init(fhp, NFS_FHSIZE); memcpy(&fhp->fh_handle.fh_base, p, NFS_FHSIZE); fhp->fh_handle.fh_size = NFS_FHSIZE; - /* FIXME: Look up export pointer here and verify - * Sun Secure RPC if requested */ - return p + (NFS_FHSIZE >> 2); -} - -/* Helper function for NFSv2 ACL code */ -__be32 *nfs2svc_decode_fh(__be32 *p, struct svc_fh *fhp) -{ - return decode_fh(p, fhp); + return true; } static __be32 * @@ -50,66 +57,95 @@ encode_fh(__be32 *p, struct svc_fh *fhp) return p + (NFS_FHSIZE>> 2); } -/* - * Decode a file name and make sure that the path contains - * no slashes or null bytes. - */ -static __be32 * -decode_filename(__be32 *p, char **namp, unsigned int *lenp) +static bool +svcxdr_decode_filename(struct xdr_stream *xdr, char **name, unsigned int *len) { - char *name; - unsigned int i; - - if ((p = xdr_decode_string_inplace(p, namp, lenp, NFS_MAXNAMLEN)) != NULL) { - for (i = 0, name = *namp; i < *lenp; i++, name++) { - if (*name == '\0' || *name == '/') - return NULL; - } - } + u32 size, i; + __be32 *p; + char *c; + + if (xdr_stream_decode_u32(xdr, &size) < 0) + return false; + if (size == 0 || size > NFS_MAXNAMLEN) + return false; + p = xdr_inline_decode(xdr, size); + if (!p) + return false; - return p; + *len = size; + *name = (char *)p; + for (i = 0, c = *name; i < size; i++, c++) + if (*c == '\0' || *c == '/') + return false; + + return true; } -static __be32 * -decode_sattr(__be32 *p, struct iattr *iap, struct user_namespace *userns) +static bool +svcxdr_decode_diropargs(struct xdr_stream *xdr, struct svc_fh *fhp, + char **name, unsigned int *len) { - u32 tmp, tmp1; + return svcxdr_decode_fhandle(xdr, fhp) && + svcxdr_decode_filename(xdr, name, len); +} + +static bool +svcxdr_decode_sattr(struct svc_rqst *rqstp, struct xdr_stream *xdr, + struct iattr *iap) +{ + u32 tmp1, tmp2; + __be32 *p; + + p = xdr_inline_decode(xdr, XDR_UNIT * 8); + if (!p) + return false; iap->ia_valid = 0; - /* Sun client bug compatibility check: some sun clients seem to - * put 0xffff in the mode field when they mean 0xffffffff. - * Quoting the 4.4BSD nfs server code: Nah nah nah nah na nah. + /* + * Some Sun clients put 0xffff in the mode field when they + * mean 0xffffffff. */ - if ((tmp = ntohl(*p++)) != (u32)-1 && tmp != 0xffff) { + tmp1 = be32_to_cpup(p++); + if (tmp1 != (u32)-1 && tmp1 != 0xffff) { iap->ia_valid |= ATTR_MODE; - iap->ia_mode = tmp; + iap->ia_mode = tmp1; } - if ((tmp = ntohl(*p++)) != (u32)-1) { - iap->ia_uid = make_kuid(userns, tmp); + + tmp1 = be32_to_cpup(p++); + if (tmp1 != (u32)-1) { + iap->ia_uid = make_kuid(nfsd_user_namespace(rqstp), tmp1); if (uid_valid(iap->ia_uid)) iap->ia_valid |= ATTR_UID; } - if ((tmp = ntohl(*p++)) != (u32)-1) { - iap->ia_gid = make_kgid(userns, tmp); + + tmp1 = be32_to_cpup(p++); + if (tmp1 != (u32)-1) { + iap->ia_gid = make_kgid(nfsd_user_namespace(rqstp), tmp1); if (gid_valid(iap->ia_gid)) iap->ia_valid |= ATTR_GID; } - if ((tmp = ntohl(*p++)) != (u32)-1) { + + tmp1 = be32_to_cpup(p++); + if (tmp1 != (u32)-1) { iap->ia_valid |= ATTR_SIZE; - iap->ia_size = tmp; + iap->ia_size = tmp1; } - tmp = ntohl(*p++); tmp1 = ntohl(*p++); - if (tmp != (u32)-1 && tmp1 != (u32)-1) { + + tmp1 = be32_to_cpup(p++); + tmp2 = be32_to_cpup(p++); + if (tmp1 != (u32)-1 && tmp2 != (u32)-1) { iap->ia_valid |= ATTR_ATIME | ATTR_ATIME_SET; - iap->ia_atime.tv_sec = tmp; - iap->ia_atime.tv_nsec = tmp1 * 1000; + iap->ia_atime.tv_sec = tmp1; + iap->ia_atime.tv_nsec = tmp2 * NSEC_PER_USEC; } - tmp = ntohl(*p++); tmp1 = ntohl(*p++); - if (tmp != (u32)-1 && tmp1 != (u32)-1) { + + tmp1 = be32_to_cpup(p++); + tmp2 = be32_to_cpup(p++); + if (tmp1 != (u32)-1 && tmp2 != (u32)-1) { iap->ia_valid |= ATTR_MTIME | ATTR_MTIME_SET; - iap->ia_mtime.tv_sec = tmp; - iap->ia_mtime.tv_nsec = tmp1 * 1000; + iap->ia_mtime.tv_sec = tmp1; + iap->ia_mtime.tv_nsec = tmp2 * NSEC_PER_USEC; /* * Passing the invalid value useconds=1000000 for mtime * is a Sun convention for "set both mtime and atime to @@ -119,10 +155,11 @@ decode_sattr(__be32 *p, struct iattr *iap, struct user_namespace *userns) * sattr in section 6.1 of "NFS Illustrated" by * Brent Callaghan, Addison-Wesley, ISBN 0-201-32750-5 */ - if (tmp1 == 1000000) + if (tmp2 == 1000000) iap->ia_valid &= ~(ATTR_ATIME_SET|ATTR_MTIME_SET); } - return p; + + return true; } static __be32 * @@ -194,225 +231,158 @@ __be32 *nfs2svc_encode_fattr(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *f */ int -nfssvc_decode_fhandle(struct svc_rqst *rqstp, __be32 *p) +nfssvc_decode_fhandleargs(struct svc_rqst *rqstp, __be32 *p) { + struct xdr_stream *xdr = &rqstp->rq_arg_stream; struct nfsd_fhandle *args = rqstp->rq_argp; - p = decode_fh(p, &args->fh); - if (!p) - return 0; - return xdr_argsize_check(rqstp, p); + return svcxdr_decode_fhandle(xdr, &args->fh); } int nfssvc_decode_sattrargs(struct svc_rqst *rqstp, __be32 *p) { + struct xdr_stream *xdr = &rqstp->rq_arg_stream; struct nfsd_sattrargs *args = rqstp->rq_argp; - p = decode_fh(p, &args->fh); - if (!p) - return 0; - p = decode_sattr(p, &args->attrs, nfsd_user_namespace(rqstp)); - - return xdr_argsize_check(rqstp, p); + return svcxdr_decode_fhandle(xdr, &args->fh) && + svcxdr_decode_sattr(rqstp, xdr, &args->attrs); } int nfssvc_decode_diropargs(struct svc_rqst *rqstp, __be32 *p) { + struct xdr_stream *xdr = &rqstp->rq_arg_stream; struct nfsd_diropargs *args = rqstp->rq_argp; - if (!(p = decode_fh(p, &args->fh)) - || !(p = decode_filename(p, &args->name, &args->len))) - return 0; - - return xdr_argsize_check(rqstp, p); + return svcxdr_decode_diropargs(xdr, &args->fh, &args->name, &args->len); } int nfssvc_decode_readargs(struct svc_rqst *rqstp, __be32 *p) { + struct xdr_stream *xdr = &rqstp->rq_arg_stream; struct nfsd_readargs *args = rqstp->rq_argp; - unsigned int len; - int v; - p = decode_fh(p, &args->fh); - if (!p) - return 0; - - args->offset = ntohl(*p++); - len = args->count = ntohl(*p++); - p++; /* totalcount - unused */ + u32 totalcount; - len = min_t(unsigned int, len, NFSSVC_MAXBLKSIZE_V2); + if (!svcxdr_decode_fhandle(xdr, &args->fh)) + return 0; + if (xdr_stream_decode_u32(xdr, &args->offset) < 0) + return 0; + if (xdr_stream_decode_u32(xdr, &args->count) < 0) + return 0; + /* totalcount is ignored */ + if (xdr_stream_decode_u32(xdr, &totalcount) < 0) + return 0; - /* set up somewhere to store response. - * We take pages, put them on reslist and include in iovec - */ - v=0; - while (len > 0) { - struct page *p = *(rqstp->rq_next_page++); - - rqstp->rq_vec[v].iov_base = page_address(p); - rqstp->rq_vec[v].iov_len = min_t(unsigned int, len, PAGE_SIZE); - len -= rqstp->rq_vec[v].iov_len; - v++; - } - args->vlen = v; - return xdr_argsize_check(rqstp, p); + return 1; } int nfssvc_decode_writeargs(struct svc_rqst *rqstp, __be32 *p) { + struct xdr_stream *xdr = &rqstp->rq_arg_stream; struct nfsd_writeargs *args = rqstp->rq_argp; - unsigned int len, hdr, dlen; struct kvec *head = rqstp->rq_arg.head; + struct kvec *tail = rqstp->rq_arg.tail; + u32 beginoffset, totalcount; + size_t remaining; - p = decode_fh(p, &args->fh); - if (!p) + if (!svcxdr_decode_fhandle(xdr, &args->fh)) return 0; - - p++; /* beginoffset */ - args->offset = ntohl(*p++); /* offset */ - p++; /* totalcount */ - len = args->len = ntohl(*p++); - /* - * The protocol specifies a maximum of 8192 bytes. - */ - if (len > NFSSVC_MAXBLKSIZE_V2) + /* beginoffset is ignored */ + if (xdr_stream_decode_u32(xdr, &beginoffset) < 0) return 0; - - /* - * Check to make sure that we got the right number of - * bytes. - */ - hdr = (void*)p - head->iov_base; - if (hdr > head->iov_len) + if (xdr_stream_decode_u32(xdr, &args->offset) < 0) + return 0; + /* totalcount is ignored */ + if (xdr_stream_decode_u32(xdr, &totalcount) < 0) return 0; - dlen = head->iov_len + rqstp->rq_arg.page_len - hdr; - /* - * Round the length of the data which was specified up to - * the next multiple of XDR units and then compare that - * against the length which was actually received. - * Note that when RPCSEC/GSS (for example) is used, the - * data buffer can be padded so dlen might be larger - * than required. It must never be smaller. - */ - if (dlen < XDR_QUADLEN(len)*4) + /* opaque data */ + if (xdr_stream_decode_u32(xdr, &args->len) < 0) return 0; + if (args->len > NFSSVC_MAXBLKSIZE_V2) + return 0; + remaining = head->iov_len + rqstp->rq_arg.page_len + tail->iov_len; + remaining -= xdr_stream_pos(xdr); + if (remaining < xdr_align_size(args->len)) + return 0; + args->first.iov_base = xdr->p; + args->first.iov_len = head->iov_len - xdr_stream_pos(xdr); - args->first.iov_base = (void *)p; - args->first.iov_len = head->iov_len - hdr; return 1; } int nfssvc_decode_createargs(struct svc_rqst *rqstp, __be32 *p) { + struct xdr_stream *xdr = &rqstp->rq_arg_stream; struct nfsd_createargs *args = rqstp->rq_argp; - if ( !(p = decode_fh(p, &args->fh)) - || !(p = decode_filename(p, &args->name, &args->len))) - return 0; - p = decode_sattr(p, &args->attrs, nfsd_user_namespace(rqstp)); - - return xdr_argsize_check(rqstp, p); + return svcxdr_decode_diropargs(xdr, &args->fh, + &args->name, &args->len) && + svcxdr_decode_sattr(rqstp, xdr, &args->attrs); } int nfssvc_decode_renameargs(struct svc_rqst *rqstp, __be32 *p) { + struct xdr_stream *xdr = &rqstp->rq_arg_stream; struct nfsd_renameargs *args = rqstp->rq_argp; - if (!(p = decode_fh(p, &args->ffh)) - || !(p = decode_filename(p, &args->fname, &args->flen)) - || !(p = decode_fh(p, &args->tfh)) - || !(p = decode_filename(p, &args->tname, &args->tlen))) - return 0; - - return xdr_argsize_check(rqstp, p); -} - -int -nfssvc_decode_readlinkargs(struct svc_rqst *rqstp, __be32 *p) -{ - struct nfsd_readlinkargs *args = rqstp->rq_argp; - - p = decode_fh(p, &args->fh); - if (!p) - return 0; - args->buffer = page_address(*(rqstp->rq_next_page++)); - - return xdr_argsize_check(rqstp, p); + return svcxdr_decode_diropargs(xdr, &args->ffh, + &args->fname, &args->flen) && + svcxdr_decode_diropargs(xdr, &args->tfh, + &args->tname, &args->tlen); } int nfssvc_decode_linkargs(struct svc_rqst *rqstp, __be32 *p) { + struct xdr_stream *xdr = &rqstp->rq_arg_stream; struct nfsd_linkargs *args = rqstp->rq_argp; - if (!(p = decode_fh(p, &args->ffh)) - || !(p = decode_fh(p, &args->tfh)) - || !(p = decode_filename(p, &args->tname, &args->tlen))) - return 0; - - return xdr_argsize_check(rqstp, p); + return svcxdr_decode_fhandle(xdr, &args->ffh) && + svcxdr_decode_diropargs(xdr, &args->tfh, + &args->tname, &args->tlen); } int nfssvc_decode_symlinkargs(struct svc_rqst *rqstp, __be32 *p) { + struct xdr_stream *xdr = &rqstp->rq_arg_stream; struct nfsd_symlinkargs *args = rqstp->rq_argp; - char *base = (char *)p; - size_t xdrlen; + struct kvec *head = rqstp->rq_arg.head; - if ( !(p = decode_fh(p, &args->ffh)) - || !(p = decode_filename(p, &args->fname, &args->flen))) + if (!svcxdr_decode_diropargs(xdr, &args->ffh, &args->fname, &args->flen)) + return 0; + if (xdr_stream_decode_u32(xdr, &args->tlen) < 0) return 0; - - args->tlen = ntohl(*p++); if (args->tlen == 0) return 0; - args->first.iov_base = p; - args->first.iov_len = rqstp->rq_arg.head[0].iov_len; - args->first.iov_len -= (char *)p - base; - - /* This request is never larger than a page. Therefore, - * transport will deliver either: - * 1. pathname in the pagelist -> sattr is in the tail. - * 2. everything in the head buffer -> sattr is in the head. - */ - if (rqstp->rq_arg.page_len) { - if (args->tlen != rqstp->rq_arg.page_len) - return 0; - p = rqstp->rq_arg.tail[0].iov_base; - } else { - xdrlen = XDR_QUADLEN(args->tlen); - if (xdrlen > args->first.iov_len - (8 * sizeof(__be32))) - return 0; - p += xdrlen; - } - decode_sattr(p, &args->attrs, nfsd_user_namespace(rqstp)); - - return 1; + args->first.iov_len = head->iov_len - xdr_stream_pos(xdr); + args->first.iov_base = xdr_inline_decode(xdr, args->tlen); + if (!args->first.iov_base) + return 0; + return svcxdr_decode_sattr(rqstp, xdr, &args->attrs); } int nfssvc_decode_readdirargs(struct svc_rqst *rqstp, __be32 *p) { + struct xdr_stream *xdr = &rqstp->rq_arg_stream; struct nfsd_readdirargs *args = rqstp->rq_argp; - p = decode_fh(p, &args->fh); - if (!p) + if (!svcxdr_decode_fhandle(xdr, &args->fh)) + return 0; + if (xdr_stream_decode_u32(xdr, &args->cookie) < 0) + return 0; + if (xdr_stream_decode_u32(xdr, &args->count) < 0) return 0; - args->cookie = ntohl(*p++); - args->count = ntohl(*p++); - args->count = min_t(u32, args->count, PAGE_SIZE); - args->buffer = page_address(*(rqstp->rq_next_page++)); - return xdr_argsize_check(rqstp, p); + return 1; } /* diff --git a/fs/nfsd/state.h b/fs/nfsd/state.h index 9eae11a9d21c..73deea353169 100644 --- a/fs/nfsd/state.h +++ b/fs/nfsd/state.h @@ -649,8 +649,7 @@ void nfs4_remove_reclaim_record(struct nfs4_client_reclaim *, struct nfsd_net *) extern void nfs4_release_reclaim(struct nfsd_net *); extern struct nfs4_client_reclaim *nfsd4_find_reclaim_client(struct xdr_netobj name, struct nfsd_net *nn); -extern __be32 nfs4_check_open_reclaim(clientid_t *clid, - struct nfsd4_compound_state *cstate, struct nfsd_net *nn); +extern __be32 nfs4_check_open_reclaim(struct nfs4_client *); extern void nfsd4_probe_callback(struct nfs4_client *clp); extern void nfsd4_probe_callback_sync(struct nfs4_client *clp); extern void nfsd4_change_callback(struct nfs4_client *clp, struct nfs4_cb_conn *); diff --git a/fs/nfsd/stats.c b/fs/nfsd/stats.c index b1bc582b0493..1d3b881e7382 100644 --- a/fs/nfsd/stats.c +++ b/fs/nfsd/stats.c @@ -7,16 +7,14 @@ * Format: * rc <hits> <misses> <nocache> * Statistsics for the reply cache - * fh <stale> <total-lookups> <anonlookups> <dir-not-in-dcache> <nondir-not-in-dcache> + * fh <stale> <deprecated filehandle cache stats> * statistics for filehandle lookup * io <bytes-read> <bytes-written> * statistics for IO throughput - * th <threads> <fullcnt> <10%-20%> <20%-30%> ... <90%-100%> <100%> - * time (seconds) when nfsd thread usage above thresholds - * and number of times that all threads were in use - * ra cache-size <10% <20% <30% ... <100% not-found - * number of times that read-ahead entry was found that deep in - * the cache. + * th <threads> <deprecated thread usage histogram stats> + * number of threads + * ra <deprecated ra-cache stats> + * * plus generic RPC stats (see net/sunrpc/stats.c) * * Copyright (C) 1995, 1996, 1997 Olaf Kirch <okir@monad.swb.de> @@ -38,31 +36,24 @@ static int nfsd_proc_show(struct seq_file *seq, void *v) { int i; - seq_printf(seq, "rc %u %u %u\nfh %u %u %u %u %u\nio %u %u\n", - nfsdstats.rchits, - nfsdstats.rcmisses, - nfsdstats.rcnocache, - nfsdstats.fh_stale, - nfsdstats.fh_lookup, - nfsdstats.fh_anon, - nfsdstats.fh_nocache_dir, - nfsdstats.fh_nocache_nondir, - nfsdstats.io_read, - nfsdstats.io_write); + seq_printf(seq, "rc %lld %lld %lld\nfh %lld 0 0 0 0\nio %lld %lld\n", + percpu_counter_sum_positive(&nfsdstats.counter[NFSD_STATS_RC_HITS]), + percpu_counter_sum_positive(&nfsdstats.counter[NFSD_STATS_RC_MISSES]), + percpu_counter_sum_positive(&nfsdstats.counter[NFSD_STATS_RC_NOCACHE]), + percpu_counter_sum_positive(&nfsdstats.counter[NFSD_STATS_FH_STALE]), + percpu_counter_sum_positive(&nfsdstats.counter[NFSD_STATS_IO_READ]), + percpu_counter_sum_positive(&nfsdstats.counter[NFSD_STATS_IO_WRITE])); + /* thread usage: */ - seq_printf(seq, "th %u %u", nfsdstats.th_cnt, nfsdstats.th_fullcnt); - for (i=0; i<10; i++) { - unsigned int jifs = nfsdstats.th_usage[i]; - unsigned int sec = jifs / HZ, msec = (jifs % HZ)*1000/HZ; - seq_printf(seq, " %u.%03u", sec, msec); - } + seq_printf(seq, "th %u 0", nfsdstats.th_cnt); + + /* deprecated thread usage histogram stats */ + for (i = 0; i < 10; i++) + seq_puts(seq, " 0.000"); + + /* deprecated ra-cache stats */ + seq_puts(seq, "\nra 0 0 0 0 0 0 0 0 0 0 0 0\n"); - /* newline and ra-cache */ - seq_printf(seq, "\nra %u", nfsdstats.ra_size); - for (i=0; i<11; i++) - seq_printf(seq, " %u", nfsdstats.ra_depth[i]); - seq_putc(seq, '\n'); - /* show my rpc info */ svc_seq_show(seq, &nfsd_svcstats); @@ -70,8 +61,10 @@ static int nfsd_proc_show(struct seq_file *seq, void *v) /* Show count for individual nfsv4 operations */ /* Writing operation numbers 0 1 2 also for maintaining uniformity */ seq_printf(seq,"proc4ops %u", LAST_NFS4_OP + 1); - for (i = 0; i <= LAST_NFS4_OP; i++) - seq_printf(seq, " %u", nfsdstats.nfs4_opcount[i]); + for (i = 0; i <= LAST_NFS4_OP; i++) { + seq_printf(seq, " %lld", + percpu_counter_sum_positive(&nfsdstats.counter[NFSD_STATS_NFS4_OP(i)])); + } seq_putc(seq, '\n'); #endif @@ -91,14 +84,63 @@ static const struct proc_ops nfsd_proc_ops = { .proc_release = single_release, }; -void -nfsd_stat_init(void) +int nfsd_percpu_counters_init(struct percpu_counter counters[], int num) +{ + int i, err = 0; + + for (i = 0; !err && i < num; i++) + err = percpu_counter_init(&counters[i], 0, GFP_KERNEL); + + if (!err) + return 0; + + for (; i > 0; i--) + percpu_counter_destroy(&counters[i-1]); + + return err; +} + +void nfsd_percpu_counters_reset(struct percpu_counter counters[], int num) +{ + int i; + + for (i = 0; i < num; i++) + percpu_counter_set(&counters[i], 0); +} + +void nfsd_percpu_counters_destroy(struct percpu_counter counters[], int num) { + int i; + + for (i = 0; i < num; i++) + percpu_counter_destroy(&counters[i]); +} + +static int nfsd_stat_counters_init(void) +{ + return nfsd_percpu_counters_init(nfsdstats.counter, NFSD_STATS_COUNTERS_NUM); +} + +static void nfsd_stat_counters_destroy(void) +{ + nfsd_percpu_counters_destroy(nfsdstats.counter, NFSD_STATS_COUNTERS_NUM); +} + +int nfsd_stat_init(void) +{ + int err; + + err = nfsd_stat_counters_init(); + if (err) + return err; + svc_proc_register(&init_net, &nfsd_svcstats, &nfsd_proc_ops); + + return 0; } -void -nfsd_stat_shutdown(void) +void nfsd_stat_shutdown(void) { + nfsd_stat_counters_destroy(); svc_proc_unregister(&init_net, "nfsd"); } diff --git a/fs/nfsd/stats.h b/fs/nfsd/stats.h index b23fdac69820..51ecda852e23 100644 --- a/fs/nfsd/stats.h +++ b/fs/nfsd/stats.h @@ -8,37 +8,91 @@ #define _NFSD_STATS_H #include <uapi/linux/nfsd/stats.h> +#include <linux/percpu_counter.h> -struct nfsd_stats { - unsigned int rchits; /* repcache hits */ - unsigned int rcmisses; /* repcache hits */ - unsigned int rcnocache; /* uncached reqs */ - unsigned int fh_stale; /* FH stale error */ - unsigned int fh_lookup; /* dentry cached */ - unsigned int fh_anon; /* anon file dentry returned */ - unsigned int fh_nocache_dir; /* filehandle not found in dcache */ - unsigned int fh_nocache_nondir; /* filehandle not found in dcache */ - unsigned int io_read; /* bytes returned to read requests */ - unsigned int io_write; /* bytes passed in write requests */ - unsigned int th_cnt; /* number of available threads */ - unsigned int th_usage[10]; /* number of ticks during which n perdeciles - * of available threads were in use */ - unsigned int th_fullcnt; /* number of times last free thread was used */ - unsigned int ra_size; /* size of ra cache */ - unsigned int ra_depth[11]; /* number of times ra entry was found that deep - * in the cache (10percentiles). [10] = not found */ +enum { + NFSD_STATS_RC_HITS, /* repcache hits */ + NFSD_STATS_RC_MISSES, /* repcache misses */ + NFSD_STATS_RC_NOCACHE, /* uncached reqs */ + NFSD_STATS_FH_STALE, /* FH stale error */ + NFSD_STATS_IO_READ, /* bytes returned to read requests */ + NFSD_STATS_IO_WRITE, /* bytes passed in write requests */ #ifdef CONFIG_NFSD_V4 - unsigned int nfs4_opcount[LAST_NFS4_OP + 1]; /* count of individual nfsv4 operations */ + NFSD_STATS_FIRST_NFS4_OP, /* count of individual nfsv4 operations */ + NFSD_STATS_LAST_NFS4_OP = NFSD_STATS_FIRST_NFS4_OP + LAST_NFS4_OP, +#define NFSD_STATS_NFS4_OP(op) (NFSD_STATS_FIRST_NFS4_OP + (op)) #endif + NFSD_STATS_COUNTERS_NUM +}; + +struct nfsd_stats { + struct percpu_counter counter[NFSD_STATS_COUNTERS_NUM]; + /* Protected by nfsd_mutex */ + unsigned int th_cnt; /* number of available threads */ }; extern struct nfsd_stats nfsdstats; + extern struct svc_stat nfsd_svcstats; -void nfsd_stat_init(void); -void nfsd_stat_shutdown(void); +int nfsd_percpu_counters_init(struct percpu_counter counters[], int num); +void nfsd_percpu_counters_reset(struct percpu_counter counters[], int num); +void nfsd_percpu_counters_destroy(struct percpu_counter counters[], int num); +int nfsd_stat_init(void); +void nfsd_stat_shutdown(void); + +static inline void nfsd_stats_rc_hits_inc(void) +{ + percpu_counter_inc(&nfsdstats.counter[NFSD_STATS_RC_HITS]); +} + +static inline void nfsd_stats_rc_misses_inc(void) +{ + percpu_counter_inc(&nfsdstats.counter[NFSD_STATS_RC_MISSES]); +} + +static inline void nfsd_stats_rc_nocache_inc(void) +{ + percpu_counter_inc(&nfsdstats.counter[NFSD_STATS_RC_NOCACHE]); +} + +static inline void nfsd_stats_fh_stale_inc(struct svc_export *exp) +{ + percpu_counter_inc(&nfsdstats.counter[NFSD_STATS_FH_STALE]); + if (exp) + percpu_counter_inc(&exp->ex_stats.counter[EXP_STATS_FH_STALE]); +} + +static inline void nfsd_stats_io_read_add(struct svc_export *exp, s64 amount) +{ + percpu_counter_add(&nfsdstats.counter[NFSD_STATS_IO_READ], amount); + if (exp) + percpu_counter_add(&exp->ex_stats.counter[EXP_STATS_IO_READ], amount); +} + +static inline void nfsd_stats_io_write_add(struct svc_export *exp, s64 amount) +{ + percpu_counter_add(&nfsdstats.counter[NFSD_STATS_IO_WRITE], amount); + if (exp) + percpu_counter_add(&exp->ex_stats.counter[EXP_STATS_IO_WRITE], amount); +} + +static inline void nfsd_stats_payload_misses_inc(struct nfsd_net *nn) +{ + percpu_counter_inc(&nn->counter[NFSD_NET_PAYLOAD_MISSES]); +} + +static inline void nfsd_stats_drc_mem_usage_add(struct nfsd_net *nn, s64 amount) +{ + percpu_counter_add(&nn->counter[NFSD_NET_DRC_MEM_USAGE], amount); +} + +static inline void nfsd_stats_drc_mem_usage_sub(struct nfsd_net *nn, s64 amount) +{ + percpu_counter_sub(&nn->counter[NFSD_NET_DRC_MEM_USAGE], amount); +} #endif /* _NFSD_STATS_H */ diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index 04937e51de56..fd6be35a1642 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -448,7 +448,7 @@ nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap, .ia_size = iap->ia_size, }; - host_err = notify_change(dentry, &size_attr, NULL); + host_err = notify_change(&init_user_ns, dentry, &size_attr, NULL); if (host_err) goto out_unlock; iap->ia_valid &= ~ATTR_SIZE; @@ -463,7 +463,7 @@ nfsd_setattr(struct svc_rqst *rqstp, struct svc_fh *fhp, struct iattr *iap, } iap->ia_valid |= ATTR_CTIME; - host_err = notify_change(dentry, iap, NULL); + host_err = notify_change(&init_user_ns, dentry, iap, NULL); out_unlock: fh_unlock(fhp); @@ -499,7 +499,8 @@ int nfsd4_is_junction(struct dentry *dentry) return 0; if (!(inode->i_mode & S_ISVTX)) return 0; - if (vfs_getxattr(dentry, NFSD_JUNCTION_XATTR_NAME, NULL, 0) <= 0) + if (vfs_getxattr(&init_user_ns, dentry, NFSD_JUNCTION_XATTR_NAME, + NULL, 0) <= 0) return 0; return 1; } @@ -889,7 +890,7 @@ static __be32 nfsd_finish_read(struct svc_rqst *rqstp, struct svc_fh *fhp, unsigned long *count, u32 *eof, ssize_t host_err) { if (host_err >= 0) { - nfsdstats.io_read += host_err; + nfsd_stats_io_read_add(fhp->fh_export, host_err); *eof = nfsd_eof_on_read(file, offset, host_err, *count); *count = host_err; fsnotify_access(file); @@ -1040,7 +1041,7 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct nfsd_file *nf, goto out_nfserr; } *cnt = host_err; - nfsdstats.io_write += *cnt; + nfsd_stats_io_write_add(exp, *cnt); fsnotify_modify(file); if (stable && use_wgather) { @@ -1254,12 +1255,12 @@ nfsd_create_locked(struct svc_rqst *rqstp, struct svc_fh *fhp, host_err = 0; switch (type) { case S_IFREG: - host_err = vfs_create(dirp, dchild, iap->ia_mode, true); + host_err = vfs_create(&init_user_ns, dirp, dchild, iap->ia_mode, true); if (!host_err) nfsd_check_ignore_resizing(iap); break; case S_IFDIR: - host_err = vfs_mkdir(dirp, dchild, iap->ia_mode); + host_err = vfs_mkdir(&init_user_ns, dirp, dchild, iap->ia_mode); if (!host_err && unlikely(d_unhashed(dchild))) { struct dentry *d; d = lookup_one_len(dchild->d_name.name, @@ -1287,7 +1288,8 @@ nfsd_create_locked(struct svc_rqst *rqstp, struct svc_fh *fhp, case S_IFBLK: case S_IFIFO: case S_IFSOCK: - host_err = vfs_mknod(dirp, dchild, iap->ia_mode, rdev); + host_err = vfs_mknod(&init_user_ns, dirp, dchild, + iap->ia_mode, rdev); break; default: printk(KERN_WARNING "nfsd: bad file type %o in nfsd_create\n", @@ -1485,7 +1487,7 @@ do_nfsd_create(struct svc_rqst *rqstp, struct svc_fh *fhp, if (!IS_POSIXACL(dirp)) iap->ia_mode &= ~current_umask(); - host_err = vfs_create(dirp, dchild, iap->ia_mode, true); + host_err = vfs_create(&init_user_ns, dirp, dchild, iap->ia_mode, true); if (host_err < 0) { fh_drop_write(fhp); goto out_nfserr; @@ -1609,7 +1611,7 @@ nfsd_symlink(struct svc_rqst *rqstp, struct svc_fh *fhp, if (IS_ERR(dnew)) goto out_nfserr; - host_err = vfs_symlink(d_inode(dentry), dnew, path); + host_err = vfs_symlink(&init_user_ns, d_inode(dentry), dnew, path); err = nfserrno(host_err); if (!err) err = nfserrno(commit_metadata(fhp)); @@ -1677,7 +1679,7 @@ nfsd_link(struct svc_rqst *rqstp, struct svc_fh *ffhp, err = nfserr_noent; if (d_really_is_negative(dold)) goto out_dput; - host_err = vfs_link(dold, dirp, dnew, NULL); + host_err = vfs_link(dold, &init_user_ns, dirp, dnew, NULL); if (!host_err) { err = nfserrno(commit_metadata(ffhp)); if (!err) @@ -1797,7 +1799,15 @@ retry: close_cached = true; goto out_dput_old; } else { - host_err = vfs_rename(fdir, odentry, tdir, ndentry, NULL, 0); + struct renamedata rd = { + .old_mnt_userns = &init_user_ns, + .old_dir = fdir, + .old_dentry = odentry, + .new_mnt_userns = &init_user_ns, + .new_dir = tdir, + .new_dentry = ndentry, + }; + host_err = vfs_rename(&rd); if (!host_err) { host_err = commit_metadata(tfhp); if (!host_err) @@ -1884,9 +1894,9 @@ nfsd_unlink(struct svc_rqst *rqstp, struct svc_fh *fhp, int type, if (type != S_IFDIR) { if (rdentry->d_sb->s_export_op->flags & EXPORT_OP_CLOSE_BEFORE_UNLINK) nfsd_close_cached_files(rdentry); - host_err = vfs_unlink(dirp, rdentry, NULL); + host_err = vfs_unlink(&init_user_ns, dirp, rdentry, NULL); } else { - host_err = vfs_rmdir(dirp, rdentry); + host_err = vfs_rmdir(&init_user_ns, dirp, rdentry); } if (!host_err) @@ -2149,7 +2159,7 @@ nfsd_getxattr(struct svc_rqst *rqstp, struct svc_fh *fhp, char *name, inode_lock_shared(inode); - len = vfs_getxattr(dentry, name, NULL, 0); + len = vfs_getxattr(&init_user_ns, dentry, name, NULL, 0); /* * Zero-length attribute, just return. @@ -2176,7 +2186,7 @@ nfsd_getxattr(struct svc_rqst *rqstp, struct svc_fh *fhp, char *name, goto out; } - len = vfs_getxattr(dentry, name, buf, len); + len = vfs_getxattr(&init_user_ns, dentry, name, buf, len); if (len <= 0) { kvfree(buf); buf = NULL; @@ -2283,7 +2293,8 @@ nfsd_removexattr(struct svc_rqst *rqstp, struct svc_fh *fhp, char *name) fh_lock(fhp); - ret = __vfs_removexattr_locked(fhp->fh_dentry, name, NULL); + ret = __vfs_removexattr_locked(&init_user_ns, fhp->fh_dentry, + name, NULL); fh_unlock(fhp); fh_drop_write(fhp); @@ -2307,8 +2318,8 @@ nfsd_setxattr(struct svc_rqst *rqstp, struct svc_fh *fhp, char *name, return nfserrno(ret); fh_lock(fhp); - ret = __vfs_setxattr_locked(fhp->fh_dentry, name, buf, len, flags, - NULL); + ret = __vfs_setxattr_locked(&init_user_ns, fhp->fh_dentry, name, buf, + len, flags, NULL); fh_unlock(fhp); fh_drop_write(fhp); @@ -2391,13 +2402,14 @@ nfsd_permission(struct svc_rqst *rqstp, struct svc_export *exp, return 0; /* This assumes NFSD_MAY_{READ,WRITE,EXEC} == MAY_{READ,WRITE,EXEC} */ - err = inode_permission(inode, acc & (MAY_READ|MAY_WRITE|MAY_EXEC)); + err = inode_permission(&init_user_ns, inode, + acc & (MAY_READ | MAY_WRITE | MAY_EXEC)); /* Allow read access to binaries even when mode 111 */ if (err == -EACCES && S_ISREG(inode->i_mode) && (acc == (NFSD_MAY_READ | NFSD_MAY_OWNER_OVERRIDE) || acc == (NFSD_MAY_READ | NFSD_MAY_READ_IF_EXEC))) - err = inode_permission(inode, MAY_EXEC); + err = inode_permission(&init_user_ns, inode, MAY_EXEC); return err? nfserrno(err) : 0; } diff --git a/fs/nfsd/xdr.h b/fs/nfsd/xdr.h index ad77387734cc..3018b52b6d5e 100644 --- a/fs/nfsd/xdr.h +++ b/fs/nfsd/xdr.h @@ -27,7 +27,6 @@ struct nfsd_readargs { struct svc_fh fh; __u32 offset; __u32 count; - int vlen; }; struct nfsd_writeargs { @@ -53,11 +52,6 @@ struct nfsd_renameargs { unsigned int tlen; }; -struct nfsd_readlinkargs { - struct svc_fh fh; - char * buffer; -}; - struct nfsd_linkargs { struct svc_fh ffh; struct svc_fh tfh; @@ -79,7 +73,6 @@ struct nfsd_readdirargs { struct svc_fh fh; __u32 cookie; __u32 count; - __be32 * buffer; }; struct nfsd_stat { @@ -144,14 +137,13 @@ union nfsd_xdrstore { #define NFS2_SVC_XDRSIZE sizeof(union nfsd_xdrstore) -int nfssvc_decode_fhandle(struct svc_rqst *, __be32 *); +int nfssvc_decode_fhandleargs(struct svc_rqst *, __be32 *); int nfssvc_decode_sattrargs(struct svc_rqst *, __be32 *); int nfssvc_decode_diropargs(struct svc_rqst *, __be32 *); int nfssvc_decode_readargs(struct svc_rqst *, __be32 *); int nfssvc_decode_writeargs(struct svc_rqst *, __be32 *); int nfssvc_decode_createargs(struct svc_rqst *, __be32 *); int nfssvc_decode_renameargs(struct svc_rqst *, __be32 *); -int nfssvc_decode_readlinkargs(struct svc_rqst *, __be32 *); int nfssvc_decode_linkargs(struct svc_rqst *, __be32 *); int nfssvc_decode_symlinkargs(struct svc_rqst *, __be32 *); int nfssvc_decode_readdirargs(struct svc_rqst *, __be32 *); @@ -172,6 +164,6 @@ void nfssvc_release_readres(struct svc_rqst *rqstp); /* Helper functions for NFSv2 ACL code */ __be32 *nfs2svc_encode_fattr(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp, struct kstat *stat); -__be32 *nfs2svc_decode_fh(__be32 *p, struct svc_fh *fhp); +bool svcxdr_decode_fhandle(struct xdr_stream *xdr, struct svc_fh *fhp); #endif /* LINUX_NFSD_H */ diff --git a/fs/nfsd/xdr3.h b/fs/nfsd/xdr3.h index 456fcd7a1038..3e1578953f54 100644 --- a/fs/nfsd/xdr3.h +++ b/fs/nfsd/xdr3.h @@ -25,14 +25,13 @@ struct nfsd3_diropargs { struct nfsd3_accessargs { struct svc_fh fh; - unsigned int access; + __u32 access; }; struct nfsd3_readargs { struct svc_fh fh; __u64 offset; __u32 count; - int vlen; }; struct nfsd3_writeargs { @@ -71,11 +70,6 @@ struct nfsd3_renameargs { unsigned int tlen; }; -struct nfsd3_readlinkargs { - struct svc_fh fh; - char * buffer; -}; - struct nfsd3_linkargs { struct svc_fh ffh; struct svc_fh tfh; @@ -96,10 +90,8 @@ struct nfsd3_symlinkargs { struct nfsd3_readdirargs { struct svc_fh fh; __u64 cookie; - __u32 dircount; __u32 count; __be32 * verf; - __be32 * buffer; }; struct nfsd3_commitargs { @@ -110,13 +102,13 @@ struct nfsd3_commitargs { struct nfsd3_getaclargs { struct svc_fh fh; - int mask; + __u32 mask; }; struct posix_acl; struct nfsd3_setaclargs { struct svc_fh fh; - int mask; + __u32 mask; struct posix_acl *acl_access; struct posix_acl *acl_default; }; @@ -273,7 +265,7 @@ union nfsd3_xdrstore { #define NFS3_SVC_XDRSIZE sizeof(union nfsd3_xdrstore) -int nfs3svc_decode_fhandle(struct svc_rqst *, __be32 *); +int nfs3svc_decode_fhandleargs(struct svc_rqst *, __be32 *); int nfs3svc_decode_sattrargs(struct svc_rqst *, __be32 *); int nfs3svc_decode_diropargs(struct svc_rqst *, __be32 *); int nfs3svc_decode_accessargs(struct svc_rqst *, __be32 *); @@ -283,7 +275,6 @@ int nfs3svc_decode_createargs(struct svc_rqst *, __be32 *); int nfs3svc_decode_mkdirargs(struct svc_rqst *, __be32 *); int nfs3svc_decode_mknodargs(struct svc_rqst *, __be32 *); int nfs3svc_decode_renameargs(struct svc_rqst *, __be32 *); -int nfs3svc_decode_readlinkargs(struct svc_rqst *, __be32 *); int nfs3svc_decode_linkargs(struct svc_rqst *, __be32 *); int nfs3svc_decode_symlinkargs(struct svc_rqst *, __be32 *); int nfs3svc_decode_readdirargs(struct svc_rqst *, __be32 *); @@ -316,7 +307,6 @@ int nfs3svc_encode_entry_plus(void *, const char *name, /* Helper functions for NFSv3 ACL code */ __be32 *nfs3svc_encode_post_op_attr(struct svc_rqst *rqstp, __be32 *p, struct svc_fh *fhp); -__be32 *nfs3svc_decode_fh(__be32 *p, struct svc_fh *fhp); - +bool svcxdr_decode_nfs_fh3(struct xdr_stream *xdr, struct svc_fh *fhp); #endif /* _LINUX_NFSD_XDR3_H */ diff --git a/fs/nilfs2/inode.c b/fs/nilfs2/inode.c index 745d371d6fea..2e8eb263cf0f 100644 --- a/fs/nilfs2/inode.c +++ b/fs/nilfs2/inode.c @@ -348,7 +348,7 @@ struct inode *nilfs_new_inode(struct inode *dir, umode_t mode) /* reference count of i_bh inherits from nilfs_mdt_read_block() */ atomic64_inc(&root->inodes_count); - inode_init_owner(inode, dir, mode); + inode_init_owner(&init_user_ns, inode, dir, mode); inode->i_ino = ino; inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode); @@ -805,14 +805,15 @@ void nilfs_evict_inode(struct inode *inode) */ } -int nilfs_setattr(struct dentry *dentry, struct iattr *iattr) +int nilfs_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *iattr) { struct nilfs_transaction_info ti; struct inode *inode = d_inode(dentry); struct super_block *sb = inode->i_sb; int err; - err = setattr_prepare(dentry, iattr); + err = setattr_prepare(&init_user_ns, dentry, iattr); if (err) return err; @@ -827,7 +828,7 @@ int nilfs_setattr(struct dentry *dentry, struct iattr *iattr) nilfs_truncate(inode); } - setattr_copy(inode, iattr); + setattr_copy(&init_user_ns, inode, iattr); mark_inode_dirty(inode); if (iattr->ia_valid & ATTR_MODE) { @@ -843,7 +844,8 @@ out_err: return err; } -int nilfs_permission(struct inode *inode, int mask) +int nilfs_permission(struct user_namespace *mnt_userns, struct inode *inode, + int mask) { struct nilfs_root *root = NILFS_I(inode)->i_root; @@ -851,7 +853,7 @@ int nilfs_permission(struct inode *inode, int mask) root->cno != NILFS_CPTREE_CURRENT_CNO) return -EROFS; /* snapshot is not writable */ - return generic_permission(inode, mask); + return generic_permission(&init_user_ns, inode, mask); } int nilfs_load_inode_block(struct inode *inode, struct buffer_head **pbh) diff --git a/fs/nilfs2/ioctl.c b/fs/nilfs2/ioctl.c index 07d26f61f22a..b053b40315bf 100644 --- a/fs/nilfs2/ioctl.c +++ b/fs/nilfs2/ioctl.c @@ -132,7 +132,7 @@ static int nilfs_ioctl_setflags(struct inode *inode, struct file *filp, unsigned int flags, oldflags; int ret; - if (!inode_owner_or_capable(inode)) + if (!inode_owner_or_capable(&init_user_ns, inode)) return -EACCES; if (get_user(flags, (int __user *)argp)) diff --git a/fs/nilfs2/namei.c b/fs/nilfs2/namei.c index a6ec7961d4f5..ecace5f96a95 100644 --- a/fs/nilfs2/namei.c +++ b/fs/nilfs2/namei.c @@ -72,8 +72,8 @@ nilfs_lookup(struct inode *dir, struct dentry *dentry, unsigned int flags) * If the create succeeds, we fill in the inode information * with d_instantiate(). */ -static int nilfs_create(struct inode *dir, struct dentry *dentry, umode_t mode, - bool excl) +static int nilfs_create(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, bool excl) { struct inode *inode; struct nilfs_transaction_info ti; @@ -100,7 +100,8 @@ static int nilfs_create(struct inode *dir, struct dentry *dentry, umode_t mode, } static int -nilfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t rdev) +nilfs_mknod(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, dev_t rdev) { struct inode *inode; struct nilfs_transaction_info ti; @@ -124,8 +125,8 @@ nilfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t rdev) return err; } -static int nilfs_symlink(struct inode *dir, struct dentry *dentry, - const char *symname) +static int nilfs_symlink(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, const char *symname) { struct nilfs_transaction_info ti; struct super_block *sb = dir->i_sb; @@ -201,7 +202,8 @@ static int nilfs_link(struct dentry *old_dentry, struct inode *dir, return err; } -static int nilfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) +static int nilfs_mkdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { struct inode *inode; struct nilfs_transaction_info ti; @@ -338,8 +340,9 @@ static int nilfs_rmdir(struct inode *dir, struct dentry *dentry) return err; } -static int nilfs_rename(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry, +static int nilfs_rename(struct user_namespace *mnt_userns, + struct inode *old_dir, struct dentry *old_dentry, + struct inode *new_dir, struct dentry *new_dentry, unsigned int flags) { struct inode *old_inode = d_inode(old_dentry); diff --git a/fs/nilfs2/nilfs.h b/fs/nilfs2/nilfs.h index f8450ee3fd06..c4a45a081ade 100644 --- a/fs/nilfs2/nilfs.h +++ b/fs/nilfs2/nilfs.h @@ -267,9 +267,11 @@ extern struct inode *nilfs_iget_for_gc(struct super_block *sb, extern void nilfs_update_inode(struct inode *, struct buffer_head *, int); extern void nilfs_truncate(struct inode *); extern void nilfs_evict_inode(struct inode *); -extern int nilfs_setattr(struct dentry *, struct iattr *); +extern int nilfs_setattr(struct user_namespace *, struct dentry *, + struct iattr *); extern void nilfs_write_failed(struct address_space *mapping, loff_t to); -int nilfs_permission(struct inode *inode, int mask); +int nilfs_permission(struct user_namespace *mnt_userns, struct inode *inode, + int mask); int nilfs_load_inode_block(struct inode *inode, struct buffer_head **pbh); extern int nilfs_inode_dirty(struct inode *); int nilfs_set_file_dirty(struct inode *inode, unsigned int nr_dirty); diff --git a/fs/nilfs2/segbuf.c b/fs/nilfs2/segbuf.c index 1a8729eded8b..56872e93823d 100644 --- a/fs/nilfs2/segbuf.c +++ b/fs/nilfs2/segbuf.c @@ -386,10 +386,6 @@ static struct bio *nilfs_alloc_seg_bio(struct the_nilfs *nilfs, sector_t start, struct bio *bio; bio = bio_alloc(GFP_NOIO, nr_vecs); - if (bio == NULL) { - while (!bio && (nr_vecs >>= 1)) - bio = bio_alloc(GFP_NOIO, nr_vecs); - } if (likely(bio)) { bio_set_dev(bio, nilfs->ns_bdev); bio->bi_iter.bi_sector = @@ -403,7 +399,7 @@ static void nilfs_segbuf_prepare_write(struct nilfs_segment_buffer *segbuf, { wi->bio = NULL; wi->rest_blocks = segbuf->sb_sum.nblocks; - wi->max_pages = BIO_MAX_PAGES; + wi->max_pages = BIO_MAX_VECS; wi->nr_vecs = min(wi->max_pages, wi->rest_blocks); wi->start = wi->end = 0; wi->blocknr = segbuf->sb_pseg_start; diff --git a/fs/nilfs2/the_nilfs.h b/fs/nilfs2/the_nilfs.h index b55cdeb4d169..987c8ab02aee 100644 --- a/fs/nilfs2/the_nilfs.h +++ b/fs/nilfs2/the_nilfs.h @@ -375,7 +375,7 @@ static inline int nilfs_flush_device(struct the_nilfs *nilfs) */ smp_wmb(); - err = blkdev_issue_flush(nilfs->ns_bdev, GFP_KERNEL); + err = blkdev_issue_flush(nilfs->ns_bdev); if (err != -EIO) err = 0; return err; diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c index dcab112e1f00..9e0c1afac8bd 100644 --- a/fs/notify/fanotify/fanotify_user.c +++ b/fs/notify/fanotify/fanotify_user.c @@ -702,7 +702,7 @@ static int fanotify_find_path(int dfd, const char __user *filename, } /* you can only watch an inode if you have read permissions on it */ - ret = inode_permission(path->dentry->d_inode, MAY_READ); + ret = path_permission(path, MAY_READ); if (ret) { path_put(path); goto out; @@ -976,7 +976,7 @@ SYSCALL_DEFINE2(fanotify_init, unsigned int, flags, unsigned int, event_f_flags) f_flags |= O_NONBLOCK; /* fsnotify_alloc_group takes a ref. Dropped in fanotify_release */ - group = fsnotify_alloc_group(&fanotify_fsnotify_ops); + group = fsnotify_alloc_user_group(&fanotify_fsnotify_ops); if (IS_ERR(group)) { free_uid(user); return PTR_ERR(group); diff --git a/fs/notify/group.c b/fs/notify/group.c index a4a4b1c64d32..ffd723ffe46d 100644 --- a/fs/notify/group.c +++ b/fs/notify/group.c @@ -111,14 +111,12 @@ void fsnotify_put_group(struct fsnotify_group *group) } EXPORT_SYMBOL_GPL(fsnotify_put_group); -/* - * Create a new fsnotify_group and hold a reference for the group returned. - */ -struct fsnotify_group *fsnotify_alloc_group(const struct fsnotify_ops *ops) +static struct fsnotify_group *__fsnotify_alloc_group( + const struct fsnotify_ops *ops, gfp_t gfp) { struct fsnotify_group *group; - group = kzalloc(sizeof(struct fsnotify_group), GFP_KERNEL); + group = kzalloc(sizeof(struct fsnotify_group), gfp); if (!group) return ERR_PTR(-ENOMEM); @@ -139,8 +137,25 @@ struct fsnotify_group *fsnotify_alloc_group(const struct fsnotify_ops *ops) return group; } + +/* + * Create a new fsnotify_group and hold a reference for the group returned. + */ +struct fsnotify_group *fsnotify_alloc_group(const struct fsnotify_ops *ops) +{ + return __fsnotify_alloc_group(ops, GFP_KERNEL); +} EXPORT_SYMBOL_GPL(fsnotify_alloc_group); +/* + * Create a new fsnotify_group and hold a reference for the group returned. + */ +struct fsnotify_group *fsnotify_alloc_user_group(const struct fsnotify_ops *ops) +{ + return __fsnotify_alloc_group(ops, GFP_KERNEL_ACCOUNT); +} +EXPORT_SYMBOL_GPL(fsnotify_alloc_user_group); + int fsnotify_fasync(int fd, struct file *file, int on) { struct fsnotify_group *group = file->private_data; diff --git a/fs/notify/inotify/inotify_user.c b/fs/notify/inotify/inotify_user.c index 59c177011a0f..c71be4fb7dc5 100644 --- a/fs/notify/inotify/inotify_user.c +++ b/fs/notify/inotify/inotify_user.c @@ -352,7 +352,7 @@ static int inotify_find_inode(const char __user *dirname, struct path *path, if (error) return error; /* you can only watch an inode if you have read permissions on it */ - error = inode_permission(path->dentry->d_inode, MAY_READ); + error = path_permission(path, MAY_READ); if (error) { path_put(path); return error; @@ -632,11 +632,11 @@ static struct fsnotify_group *inotify_new_group(unsigned int max_events) struct fsnotify_group *group; struct inotify_event_info *oevent; - group = fsnotify_alloc_group(&inotify_fsnotify_ops); + group = fsnotify_alloc_user_group(&inotify_fsnotify_ops); if (IS_ERR(group)) return group; - oevent = kmalloc(sizeof(struct inotify_event_info), GFP_KERNEL); + oevent = kmalloc(sizeof(struct inotify_event_info), GFP_KERNEL_ACCOUNT); if (unlikely(!oevent)) { fsnotify_destroy_group(group); return ERR_PTR(-ENOMEM); diff --git a/fs/ntfs/inode.c b/fs/ntfs/inode.c index f7e4cbc26eaf..f5c058b3192c 100644 --- a/fs/ntfs/inode.c +++ b/fs/ntfs/inode.c @@ -629,6 +629,12 @@ static int ntfs_read_locked_inode(struct inode *vi) } a = ctx->attr; /* Get the standard information attribute value. */ + if ((u8 *)a + le16_to_cpu(a->data.resident.value_offset) + + le32_to_cpu(a->data.resident.value_length) > + (u8 *)ctx->mrec + vol->mft_record_size) { + ntfs_error(vi->i_sb, "Corrupt standard information attribute in inode."); + goto unm_err_out; + } si = (STANDARD_INFORMATION*)((u8*)a + le16_to_cpu(a->data.resident.value_offset)); @@ -2848,6 +2854,7 @@ void ntfs_truncate_vfs(struct inode *vi) { /** * ntfs_setattr - called from notify_change() when an attribute is being changed + * @mnt_userns: user namespace of the mount the inode was found from * @dentry: dentry whose attributes to change * @attr: structure describing the attributes and the changes * @@ -2860,13 +2867,14 @@ void ntfs_truncate_vfs(struct inode *vi) { * * Called with ->i_mutex held. */ -int ntfs_setattr(struct dentry *dentry, struct iattr *attr) +int ntfs_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *attr) { struct inode *vi = d_inode(dentry); int err; unsigned int ia_valid = attr->ia_valid; - err = setattr_prepare(dentry, attr); + err = setattr_prepare(&init_user_ns, dentry, attr); if (err) goto out; /* We do not support NTFS ACLs yet. */ diff --git a/fs/ntfs/inode.h b/fs/ntfs/inode.h index 363e4e820673..6f78ee00f57f 100644 --- a/fs/ntfs/inode.h +++ b/fs/ntfs/inode.h @@ -289,7 +289,8 @@ extern int ntfs_show_options(struct seq_file *sf, struct dentry *root); extern int ntfs_truncate(struct inode *vi); extern void ntfs_truncate_vfs(struct inode *vi); -extern int ntfs_setattr(struct dentry *dentry, struct iattr *attr); +extern int ntfs_setattr(struct user_namespace *mnt_userns, + struct dentry *dentry, struct iattr *attr); extern int __ntfs_write_inode(struct inode *vi, int sync); diff --git a/fs/ntfs/layout.h b/fs/ntfs/layout.h index 85422761ff43..5d4bf7a3259f 100644 --- a/fs/ntfs/layout.h +++ b/fs/ntfs/layout.h @@ -703,7 +703,7 @@ typedef struct { /* 14*/ le16 instance; /* The instance of this attribute record. This number is unique within this mft record (see MFT_RECORD/next_attribute_instance notes in - in mft.h for more details). */ + mft.h for more details). */ /* 16*/ union { /* Resident attributes. */ struct { @@ -1838,7 +1838,7 @@ typedef struct { * Also, each security descriptor is stored twice in the $SDS stream with a * fixed offset of 0x40000 bytes (256kib, the Windows cache manager's max size) * between them; i.e. if a SDS_ENTRY specifies an offset of 0x51d0, then the - * the first copy of the security descriptor will be at offset 0x51d0 in the + * first copy of the security descriptor will be at offset 0x51d0 in the * $SDS data stream and the second copy will be at offset 0x451d0. */ typedef struct { diff --git a/fs/ocfs2/acl.c b/fs/ocfs2/acl.c index 7b07f5df3a29..5259badabb56 100644 --- a/fs/ocfs2/acl.c +++ b/fs/ocfs2/acl.c @@ -262,7 +262,8 @@ static int ocfs2_set_acl(handle_t *handle, return ret; } -int ocfs2_iop_set_acl(struct inode *inode, struct posix_acl *acl, int type) +int ocfs2_iop_set_acl(struct user_namespace *mnt_userns, struct inode *inode, + struct posix_acl *acl, int type) { struct buffer_head *bh = NULL; int status, had_lock; @@ -274,7 +275,8 @@ int ocfs2_iop_set_acl(struct inode *inode, struct posix_acl *acl, int type) if (type == ACL_TYPE_ACCESS && acl) { umode_t mode; - status = posix_acl_update_mode(inode, &mode, &acl); + status = posix_acl_update_mode(&init_user_ns, inode, &mode, + &acl); if (status) goto unlock; diff --git a/fs/ocfs2/acl.h b/fs/ocfs2/acl.h index 127b13432146..4e86450917b2 100644 --- a/fs/ocfs2/acl.h +++ b/fs/ocfs2/acl.h @@ -19,7 +19,8 @@ struct ocfs2_acl_entry { }; struct posix_acl *ocfs2_iop_get_acl(struct inode *inode, int type); -int ocfs2_iop_set_acl(struct inode *inode, struct posix_acl *acl, int type); +int ocfs2_iop_set_acl(struct user_namespace *mnt_userns, struct inode *inode, + struct posix_acl *acl, int type); extern int ocfs2_acl_chmod(struct inode *, struct buffer_head *); extern int ocfs2_init_acl(handle_t *, struct inode *, struct inode *, struct buffer_head *, struct buffer_head *, diff --git a/fs/ocfs2/cluster/heartbeat.c b/fs/ocfs2/cluster/heartbeat.c index 0179a73a3fa2..12a7590601dd 100644 --- a/fs/ocfs2/cluster/heartbeat.c +++ b/fs/ocfs2/cluster/heartbeat.c @@ -2042,7 +2042,7 @@ static struct config_item *o2hb_heartbeat_group_make_item(struct config_group *g o2hb_nego_timeout_handler, reg, NULL, ®->hr_handler_list); if (ret) - goto free; + goto remove_item; ret = o2net_register_handler(O2HB_NEGO_APPROVE_MSG, reg->hr_key, sizeof(struct o2hb_nego_msg), @@ -2057,6 +2057,12 @@ static struct config_item *o2hb_heartbeat_group_make_item(struct config_group *g unregister_handler: o2net_unregister_handler_list(®->hr_handler_list); +remove_item: + spin_lock(&o2hb_live_lock); + list_del(®->hr_all_item); + if (o2hb_global_heartbeat_active()) + clear_bit(reg->hr_region_num, o2hb_region_bitmap); + spin_unlock(&o2hb_live_lock); free: kfree(reg); return ERR_PTR(ret); diff --git a/fs/ocfs2/dlm/dlmast.c b/fs/ocfs2/dlm/dlmast.c index 6abaded3ff6b..70a10764f249 100644 --- a/fs/ocfs2/dlm/dlmast.c +++ b/fs/ocfs2/dlm/dlmast.c @@ -165,16 +165,6 @@ void __dlm_queue_bast(struct dlm_ctxt *dlm, struct dlm_lock *lock) spin_unlock(&lock->spinlock); } -void dlm_queue_bast(struct dlm_ctxt *dlm, struct dlm_lock *lock) -{ - BUG_ON(!dlm); - BUG_ON(!lock); - - spin_lock(&dlm->ast_lock); - __dlm_queue_bast(dlm, lock); - spin_unlock(&dlm->ast_lock); -} - static void dlm_update_lvb(struct dlm_ctxt *dlm, struct dlm_lock_resource *res, struct dlm_lock *lock) { diff --git a/fs/ocfs2/dlm/dlmcommon.h b/fs/ocfs2/dlm/dlmcommon.h index c8a444622faa..58d57e25d384 100644 --- a/fs/ocfs2/dlm/dlmcommon.h +++ b/fs/ocfs2/dlm/dlmcommon.h @@ -17,10 +17,7 @@ #define DLM_LOCKID_NAME_MAX 32 -#define DLM_DOMAIN_NAME_MAX_LEN 255 #define DLM_LOCK_RES_OWNER_UNKNOWN O2NM_MAX_NODES -#define DLM_THREAD_SHUFFLE_INTERVAL 5 // flush everything every 5 passes -#define DLM_THREAD_MS 200 // flush at least every 200 ms #define DLM_HASH_SIZE_DEFAULT (1 << 17) #if DLM_HASH_SIZE_DEFAULT < PAGE_SIZE @@ -902,7 +899,6 @@ void __dlm_lockres_grab_inflight_worker(struct dlm_ctxt *dlm, struct dlm_lock_resource *res); void dlm_queue_ast(struct dlm_ctxt *dlm, struct dlm_lock *lock); -void dlm_queue_bast(struct dlm_ctxt *dlm, struct dlm_lock *lock); void __dlm_queue_ast(struct dlm_ctxt *dlm, struct dlm_lock *lock); void __dlm_queue_bast(struct dlm_ctxt *dlm, struct dlm_lock *lock); void dlm_do_local_ast(struct dlm_ctxt *dlm, diff --git a/fs/ocfs2/dlmfs/dlmfs.c b/fs/ocfs2/dlmfs/dlmfs.c index 583820ec63e2..b2870f1a31df 100644 --- a/fs/ocfs2/dlmfs/dlmfs.c +++ b/fs/ocfs2/dlmfs/dlmfs.c @@ -190,17 +190,18 @@ static int dlmfs_file_release(struct inode *inode, * We do ->setattr() just to override size changes. Our size is the size * of the LVB and nothing else. */ -static int dlmfs_file_setattr(struct dentry *dentry, struct iattr *attr) +static int dlmfs_file_setattr(struct user_namespace *mnt_userns, + struct dentry *dentry, struct iattr *attr) { int error; struct inode *inode = d_inode(dentry); attr->ia_valid &= ~ATTR_SIZE; - error = setattr_prepare(dentry, attr); + error = setattr_prepare(&init_user_ns, dentry, attr); if (error) return error; - setattr_copy(inode, attr); + setattr_copy(&init_user_ns, inode, attr); mark_inode_dirty(inode); return 0; } @@ -329,7 +330,7 @@ static struct inode *dlmfs_get_root_inode(struct super_block *sb) if (inode) { inode->i_ino = get_next_ino(); - inode_init_owner(inode, NULL, mode); + inode_init_owner(&init_user_ns, inode, NULL, mode); inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode); inc_nlink(inode); @@ -352,7 +353,7 @@ static struct inode *dlmfs_get_inode(struct inode *parent, return NULL; inode->i_ino = get_next_ino(); - inode_init_owner(inode, parent, mode); + inode_init_owner(&init_user_ns, inode, parent, mode); inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode); ip = DLMFS_I(inode); @@ -395,7 +396,8 @@ static struct inode *dlmfs_get_inode(struct inode *parent, * File creation. Allocate an inode, and we're done.. */ /* SMP-safe */ -static int dlmfs_mkdir(struct inode * dir, +static int dlmfs_mkdir(struct user_namespace * mnt_userns, + struct inode * dir, struct dentry * dentry, umode_t mode) { @@ -443,7 +445,8 @@ bail: return status; } -static int dlmfs_create(struct inode *dir, +static int dlmfs_create(struct user_namespace *mnt_userns, + struct inode *dir, struct dentry *dentry, umode_t mode, bool excl) diff --git a/fs/ocfs2/file.c b/fs/ocfs2/file.c index 85979e2214b3..6611c64ca0be 100644 --- a/fs/ocfs2/file.c +++ b/fs/ocfs2/file.c @@ -194,7 +194,7 @@ static int ocfs2_sync_file(struct file *file, loff_t start, loff_t end, needs_barrier = true; err = jbd2_complete_transaction(journal, commit_tid); if (needs_barrier) { - ret = blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL); + ret = blkdev_issue_flush(inode->i_sb->s_bdev); if (!err) err = ret; } @@ -1112,7 +1112,8 @@ out: return ret; } -int ocfs2_setattr(struct dentry *dentry, struct iattr *attr) +int ocfs2_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *attr) { int status = 0, size_change; int inode_locked = 0; @@ -1142,7 +1143,7 @@ int ocfs2_setattr(struct dentry *dentry, struct iattr *attr) if (!(attr->ia_valid & OCFS2_VALID_ATTRS)) return 0; - status = setattr_prepare(dentry, attr); + status = setattr_prepare(&init_user_ns, dentry, attr); if (status) return status; @@ -1263,7 +1264,7 @@ int ocfs2_setattr(struct dentry *dentry, struct iattr *attr) } } - setattr_copy(inode, attr); + setattr_copy(&init_user_ns, inode, attr); mark_inode_dirty(inode); status = ocfs2_mark_inode_dirty(handle, inode, bh); @@ -1298,8 +1299,8 @@ bail: return status; } -int ocfs2_getattr(const struct path *path, struct kstat *stat, - u32 request_mask, unsigned int flags) +int ocfs2_getattr(struct user_namespace *mnt_userns, const struct path *path, + struct kstat *stat, u32 request_mask, unsigned int flags) { struct inode *inode = d_inode(path->dentry); struct super_block *sb = path->dentry->d_sb; @@ -1313,7 +1314,7 @@ int ocfs2_getattr(const struct path *path, struct kstat *stat, goto bail; } - generic_fillattr(inode, stat); + generic_fillattr(&init_user_ns, inode, stat); /* * If there is inline data in the inode, the inode will normally not * have data blocks allocated (it may have an external xattr block). @@ -1330,7 +1331,8 @@ bail: return err; } -int ocfs2_permission(struct inode *inode, int mask) +int ocfs2_permission(struct user_namespace *mnt_userns, struct inode *inode, + int mask) { int ret, had_lock; struct ocfs2_lock_holder oh; @@ -1355,7 +1357,7 @@ int ocfs2_permission(struct inode *inode, int mask) dump_stack(); } - ret = generic_permission(inode, mask); + ret = generic_permission(&init_user_ns, inode, mask); ocfs2_inode_unlock_tracker(inode, 0, &oh, had_lock); out: diff --git a/fs/ocfs2/file.h b/fs/ocfs2/file.h index 4832cbceba5b..8536cec5f122 100644 --- a/fs/ocfs2/file.h +++ b/fs/ocfs2/file.h @@ -51,10 +51,13 @@ int ocfs2_extend_no_holes(struct inode *inode, struct buffer_head *di_bh, u64 new_i_size, u64 zero_to); int ocfs2_zero_extend(struct inode *inode, struct buffer_head *di_bh, loff_t zero_to); -int ocfs2_setattr(struct dentry *dentry, struct iattr *attr); -int ocfs2_getattr(const struct path *path, struct kstat *stat, - u32 request_mask, unsigned int flags); -int ocfs2_permission(struct inode *inode, int mask); +int ocfs2_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *attr); +int ocfs2_getattr(struct user_namespace *mnt_userns, const struct path *path, + struct kstat *stat, u32 request_mask, unsigned int flags); +int ocfs2_permission(struct user_namespace *mnt_userns, + struct inode *inode, + int mask); int ocfs2_should_update_atime(struct inode *inode, struct vfsmount *vfsmnt); diff --git a/fs/ocfs2/ioctl.c b/fs/ocfs2/ioctl.c index 89984172fc4a..50c9b30ee9f6 100644 --- a/fs/ocfs2/ioctl.c +++ b/fs/ocfs2/ioctl.c @@ -96,7 +96,7 @@ static int ocfs2_set_inode_attr(struct inode *inode, unsigned flags, } status = -EACCES; - if (!inode_owner_or_capable(inode)) + if (!inode_owner_or_capable(&init_user_ns, inode)) goto bail_unlock; if (!S_ISDIR(inode->i_mode)) diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c index 2a237ab00453..3abdd36da2e2 100644 --- a/fs/ocfs2/namei.c +++ b/fs/ocfs2/namei.c @@ -198,7 +198,7 @@ static struct inode *ocfs2_get_init_inode(struct inode *dir, umode_t mode) * callers. */ if (S_ISDIR(mode)) set_nlink(inode, 2); - inode_init_owner(inode, dir, mode); + inode_init_owner(&init_user_ns, inode, dir, mode); status = dquot_initialize(inode); if (status) return ERR_PTR(status); @@ -221,7 +221,8 @@ static void ocfs2_cleanup_add_entry_failure(struct ocfs2_super *osb, iput(inode); } -static int ocfs2_mknod(struct inode *dir, +static int ocfs2_mknod(struct user_namespace *mnt_userns, + struct inode *dir, struct dentry *dentry, umode_t mode, dev_t dev) @@ -645,7 +646,8 @@ static int ocfs2_mknod_locked(struct ocfs2_super *osb, return status; } -static int ocfs2_mkdir(struct inode *dir, +static int ocfs2_mkdir(struct user_namespace *mnt_userns, + struct inode *dir, struct dentry *dentry, umode_t mode) { @@ -653,14 +655,15 @@ static int ocfs2_mkdir(struct inode *dir, trace_ocfs2_mkdir(dir, dentry, dentry->d_name.len, dentry->d_name.name, OCFS2_I(dir)->ip_blkno, mode); - ret = ocfs2_mknod(dir, dentry, mode | S_IFDIR, 0); + ret = ocfs2_mknod(&init_user_ns, dir, dentry, mode | S_IFDIR, 0); if (ret) mlog_errno(ret); return ret; } -static int ocfs2_create(struct inode *dir, +static int ocfs2_create(struct user_namespace *mnt_userns, + struct inode *dir, struct dentry *dentry, umode_t mode, bool excl) @@ -669,7 +672,7 @@ static int ocfs2_create(struct inode *dir, trace_ocfs2_create(dir, dentry, dentry->d_name.len, dentry->d_name.name, (unsigned long long)OCFS2_I(dir)->ip_blkno, mode); - ret = ocfs2_mknod(dir, dentry, mode | S_IFREG, 0); + ret = ocfs2_mknod(&init_user_ns, dir, dentry, mode | S_IFREG, 0); if (ret) mlog_errno(ret); @@ -1195,7 +1198,8 @@ static void ocfs2_double_unlock(struct inode *inode1, struct inode *inode2) ocfs2_inode_unlock(inode2, 1); } -static int ocfs2_rename(struct inode *old_dir, +static int ocfs2_rename(struct user_namespace *mnt_userns, + struct inode *old_dir, struct dentry *old_dentry, struct inode *new_dir, struct dentry *new_dentry, @@ -1784,7 +1788,8 @@ bail: return status; } -static int ocfs2_symlink(struct inode *dir, +static int ocfs2_symlink(struct user_namespace *mnt_userns, + struct inode *dir, struct dentry *dentry, const char *symname) { diff --git a/fs/ocfs2/refcounttree.c b/fs/ocfs2/refcounttree.c index 3b397fa9c9e8..c19a463fac55 100644 --- a/fs/ocfs2/refcounttree.c +++ b/fs/ocfs2/refcounttree.c @@ -978,7 +978,7 @@ static int ocfs2_get_refcount_cpos_end(struct ocfs2_caching_info *ci, return 0; } - if (!eb || (eb && !eb->h_next_leaf_blk)) { + if (!eb || !eb->h_next_leaf_blk) { /* * We are the last extent rec, so any high cpos should * be stored in this leaf refcount block. @@ -4346,7 +4346,7 @@ static inline int ocfs2_may_create(struct inode *dir, struct dentry *child) return -EEXIST; if (IS_DEADDIR(dir)) return -ENOENT; - return inode_permission(dir, MAY_WRITE | MAY_EXEC); + return inode_permission(&init_user_ns, dir, MAY_WRITE | MAY_EXEC); } /** @@ -4400,7 +4400,7 @@ static int ocfs2_vfs_reflink(struct dentry *old_dentry, struct inode *dir, * file. */ if (!preserve) { - error = inode_permission(inode, MAY_READ); + error = inode_permission(&init_user_ns, inode, MAY_READ); if (error) return error; } diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index 2febc76e9de7..079f8826993e 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -973,8 +973,6 @@ static void ocfs2_disable_quotas(struct ocfs2_super *osb) * quota files */ dquot_disable(sb, type, DQUOT_USAGE_ENABLED | DQUOT_LIMITS_ENABLED); - if (!inode) - continue; iput(inode); } } diff --git a/fs/ocfs2/xattr.c b/fs/ocfs2/xattr.c index 9ccd19d8f7b1..36ae47a4aef6 100644 --- a/fs/ocfs2/xattr.c +++ b/fs/ocfs2/xattr.c @@ -7249,6 +7249,7 @@ static int ocfs2_xattr_security_get(const struct xattr_handler *handler, } static int ocfs2_xattr_security_set(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *unused, struct inode *inode, const char *name, const void *value, size_t size, int flags) @@ -7321,6 +7322,7 @@ static int ocfs2_xattr_trusted_get(const struct xattr_handler *handler, } static int ocfs2_xattr_trusted_set(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *unused, struct inode *inode, const char *name, const void *value, size_t size, int flags) @@ -7351,6 +7353,7 @@ static int ocfs2_xattr_user_get(const struct xattr_handler *handler, } static int ocfs2_xattr_user_set(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *unused, struct inode *inode, const char *name, const void *value, size_t size, int flags) diff --git a/fs/omfs/dir.c b/fs/omfs/dir.c index a0f45651f3b7..c219f91f44e9 100644 --- a/fs/omfs/dir.c +++ b/fs/omfs/dir.c @@ -279,13 +279,14 @@ out_free_inode: return err; } -static int omfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) +static int omfs_mkdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { return omfs_add_node(dir, dentry, mode | S_IFDIR); } -static int omfs_create(struct inode *dir, struct dentry *dentry, umode_t mode, - bool excl) +static int omfs_create(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, bool excl) { return omfs_add_node(dir, dentry, mode | S_IFREG); } @@ -369,9 +370,9 @@ static bool omfs_fill_chain(struct inode *dir, struct dir_context *ctx, return true; } -static int omfs_rename(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry, - unsigned int flags) +static int omfs_rename(struct user_namespace *mnt_userns, struct inode *old_dir, + struct dentry *old_dentry, struct inode *new_dir, + struct dentry *new_dentry, unsigned int flags) { struct inode *new_inode = d_inode(new_dentry); struct inode *old_inode = d_inode(old_dentry); diff --git a/fs/omfs/file.c b/fs/omfs/file.c index 2c7b70ee1388..11e733aab25d 100644 --- a/fs/omfs/file.c +++ b/fs/omfs/file.c @@ -343,12 +343,13 @@ const struct file_operations omfs_file_operations = { .splice_read = generic_file_splice_read, }; -static int omfs_setattr(struct dentry *dentry, struct iattr *attr) +static int omfs_setattr(struct user_namespace *mnt_userns, + struct dentry *dentry, struct iattr *attr) { struct inode *inode = d_inode(dentry); int error; - error = setattr_prepare(dentry, attr); + error = setattr_prepare(&init_user_ns, dentry, attr); if (error) return error; @@ -361,7 +362,7 @@ static int omfs_setattr(struct dentry *dentry, struct iattr *attr) omfs_truncate(inode); } - setattr_copy(inode, attr); + setattr_copy(&init_user_ns, inode, attr); mark_inode_dirty(inode); return 0; } diff --git a/fs/omfs/inode.c b/fs/omfs/inode.c index ce93ccca8639..2a0e83236c01 100644 --- a/fs/omfs/inode.c +++ b/fs/omfs/inode.c @@ -48,7 +48,7 @@ struct inode *omfs_new_inode(struct inode *dir, umode_t mode) goto fail; inode->i_ino = new_block; - inode_init_owner(inode, NULL, mode); + inode_init_owner(&init_user_ns, inode, NULL, mode); inode->i_mapping->a_ops = &omfs_aops; inode->i_atime = inode->i_mtime = inode->i_ctime = current_time(inode); diff --git a/fs/open.c b/fs/open.c index 1e06e443a565..e53af13b5835 100644 --- a/fs/open.c +++ b/fs/open.c @@ -35,8 +35,8 @@ #include "internal.h" -int do_truncate(struct dentry *dentry, loff_t length, unsigned int time_attrs, - struct file *filp) +int do_truncate(struct user_namespace *mnt_userns, struct dentry *dentry, + loff_t length, unsigned int time_attrs, struct file *filp) { int ret; struct iattr newattrs; @@ -61,13 +61,14 @@ int do_truncate(struct dentry *dentry, loff_t length, unsigned int time_attrs, inode_lock(dentry->d_inode); /* Note any delegations or leases have already been broken: */ - ret = notify_change(dentry, &newattrs, NULL); + ret = notify_change(mnt_userns, dentry, &newattrs, NULL); inode_unlock(dentry->d_inode); return ret; } long vfs_truncate(const struct path *path, loff_t length) { + struct user_namespace *mnt_userns; struct inode *inode; long error; @@ -83,7 +84,8 @@ long vfs_truncate(const struct path *path, loff_t length) if (error) goto out; - error = inode_permission(inode, MAY_WRITE); + mnt_userns = mnt_user_ns(path->mnt); + error = inode_permission(mnt_userns, inode, MAY_WRITE); if (error) goto mnt_drop_write_and_out; @@ -107,7 +109,7 @@ long vfs_truncate(const struct path *path, loff_t length) if (!error) error = security_path_truncate(path); if (!error) - error = do_truncate(path->dentry, length, 0, NULL); + error = do_truncate(mnt_userns, path->dentry, length, 0, NULL); put_write_and_out: put_write_access(inode); @@ -186,13 +188,13 @@ long do_sys_ftruncate(unsigned int fd, loff_t length, int small) /* Check IS_APPEND on real upper inode */ if (IS_APPEND(file_inode(f.file))) goto out_putf; - sb_start_write(inode->i_sb); error = locks_verify_truncate(inode, f.file, length); if (!error) error = security_path_truncate(&f.file->f_path); if (!error) - error = do_truncate(dentry, length, ATTR_MTIME|ATTR_CTIME, f.file); + error = do_truncate(file_mnt_user_ns(f.file), dentry, length, + ATTR_MTIME | ATTR_CTIME, f.file); sb_end_write(inode->i_sb); out_putf: fdput(f); @@ -436,7 +438,7 @@ retry: goto out_path_release; } - res = inode_permission(inode, mode | MAY_ACCESS); + res = inode_permission(mnt_user_ns(path.mnt), inode, mode | MAY_ACCESS); /* SuS v2 requires we report a read only fs too */ if (res || !(mode & S_IWOTH) || special_file(inode->i_mode)) goto out_path_release; @@ -492,7 +494,7 @@ retry: if (error) goto out; - error = inode_permission(path.dentry->d_inode, MAY_EXEC | MAY_CHDIR); + error = path_permission(&path, MAY_EXEC | MAY_CHDIR); if (error) goto dput_and_out; @@ -521,7 +523,7 @@ SYSCALL_DEFINE1(fchdir, unsigned int, fd) if (!d_can_lookup(f.file->f_path.dentry)) goto out_putf; - error = inode_permission(file_inode(f.file), MAY_EXEC | MAY_CHDIR); + error = file_permission(f.file, MAY_EXEC | MAY_CHDIR); if (!error) set_fs_pwd(current->fs, &f.file->f_path); out_putf: @@ -540,7 +542,7 @@ retry: if (error) goto out; - error = inode_permission(path.dentry->d_inode, MAY_EXEC | MAY_CHDIR); + error = path_permission(&path, MAY_EXEC | MAY_CHDIR); if (error) goto dput_and_out; @@ -580,7 +582,8 @@ retry_deleg: goto out_unlock; newattrs.ia_mode = (mode & S_IALLUGO) | (inode->i_mode & ~S_IALLUGO); newattrs.ia_valid = ATTR_MODE | ATTR_CTIME; - error = notify_change(path->dentry, &newattrs, &delegated_inode); + error = notify_change(mnt_user_ns(path->mnt), path->dentry, + &newattrs, &delegated_inode); out_unlock: inode_unlock(inode); if (delegated_inode) { @@ -641,6 +644,7 @@ SYSCALL_DEFINE2(chmod, const char __user *, filename, umode_t, mode) int chown_common(const struct path *path, uid_t user, gid_t group) { + struct user_namespace *mnt_userns; struct inode *inode = path->dentry->d_inode; struct inode *delegated_inode = NULL; int error; @@ -651,6 +655,10 @@ int chown_common(const struct path *path, uid_t user, gid_t group) uid = make_kuid(current_user_ns(), user); gid = make_kgid(current_user_ns(), group); + mnt_userns = mnt_user_ns(path->mnt); + uid = kuid_from_mnt(mnt_userns, uid); + gid = kgid_from_mnt(mnt_userns, gid); + retry_deleg: newattrs.ia_valid = ATTR_CTIME; if (user != (uid_t) -1) { @@ -671,7 +679,8 @@ retry_deleg: inode_lock(inode); error = security_path_chown(path, uid, gid); if (!error) - error = notify_change(path->dentry, &newattrs, &delegated_inode); + error = notify_change(mnt_userns, path->dentry, &newattrs, + &delegated_inode); inode_unlock(inode); if (delegated_inode) { error = break_deleg_wait(&delegated_inode); @@ -1091,6 +1100,12 @@ inline int build_open_flags(const struct open_how *how, struct open_flags *op) lookup_flags |= LOOKUP_BENEATH; if (how->resolve & RESOLVE_IN_ROOT) lookup_flags |= LOOKUP_IN_ROOT; + if (how->resolve & RESOLVE_CACHED) { + /* Don't bother even trying for create/truncate/tmpfile open */ + if (flags & (O_TRUNC | O_CREAT | O_TMPFILE)) + return -EAGAIN; + lookup_flags |= LOOKUP_CACHED; + } op->lookup_flags = lookup_flags; return 0; diff --git a/fs/orangefs/acl.c b/fs/orangefs/acl.c index a25e6c890975..18852b9ed82b 100644 --- a/fs/orangefs/acl.c +++ b/fs/orangefs/acl.c @@ -116,7 +116,8 @@ out: return error; } -int orangefs_set_acl(struct inode *inode, struct posix_acl *acl, int type) +int orangefs_set_acl(struct user_namespace *mnt_userns, struct inode *inode, + struct posix_acl *acl, int type) { int error; struct iattr iattr; @@ -132,7 +133,8 @@ int orangefs_set_acl(struct inode *inode, struct posix_acl *acl, int type) * and "mode" to the new desired value. It is up to * us to propagate the new mode back to the server... */ - error = posix_acl_update_mode(inode, &iattr.ia_mode, &acl); + error = posix_acl_update_mode(&init_user_ns, inode, + &iattr.ia_mode, &acl); if (error) { gossip_err("%s: posix_acl_update_mode err: %d\n", __func__, diff --git a/fs/orangefs/file.c b/fs/orangefs/file.c index ec8ae4257975..9b28a7132466 100644 --- a/fs/orangefs/file.c +++ b/fs/orangefs/file.c @@ -487,10 +487,7 @@ static int orangefs_file_mmap(struct file *file, struct vm_area_struct *vma) return ret; gossip_debug(GOSSIP_FILE_DEBUG, - "orangefs_file_mmap: called on %s\n", - (file ? - (char *)file->f_path.dentry->d_name.name : - (char *)"Unknown")); + "orangefs_file_mmap: called on %pD\n", file); /* set the sequential readahead hint */ vma->vm_flags |= VM_SEQ_READ; diff --git a/fs/orangefs/inode.c b/fs/orangefs/inode.c index 48f0547d4850..5079cfafa8d7 100644 --- a/fs/orangefs/inode.c +++ b/fs/orangefs/inode.c @@ -855,13 +855,13 @@ again: ORANGEFS_I(inode)->attr_uid = current_fsuid(); ORANGEFS_I(inode)->attr_gid = current_fsgid(); } - setattr_copy(inode, iattr); + setattr_copy(&init_user_ns, inode, iattr); spin_unlock(&inode->i_lock); mark_inode_dirty(inode); if (iattr->ia_valid & ATTR_MODE) /* change mod on a file that has ACLs */ - ret = posix_acl_chmod(inode, inode->i_mode); + ret = posix_acl_chmod(&init_user_ns, inode, inode->i_mode); ret = 0; out: @@ -871,12 +871,13 @@ out: /* * Change attributes of an object referenced by dentry. */ -int orangefs_setattr(struct dentry *dentry, struct iattr *iattr) +int orangefs_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *iattr) { int ret; gossip_debug(GOSSIP_INODE_DEBUG, "__orangefs_setattr: called on %pd\n", dentry); - ret = setattr_prepare(dentry, iattr); + ret = setattr_prepare(&init_user_ns, dentry, iattr); if (ret) goto out; ret = __orangefs_setattr(d_inode(dentry), iattr); @@ -890,8 +891,8 @@ out: /* * Obtain attributes of an object given a dentry */ -int orangefs_getattr(const struct path *path, struct kstat *stat, - u32 request_mask, unsigned int flags) +int orangefs_getattr(struct user_namespace *mnt_userns, const struct path *path, + struct kstat *stat, u32 request_mask, unsigned int flags) { int ret; struct inode *inode = path->dentry->d_inode; @@ -903,7 +904,7 @@ int orangefs_getattr(const struct path *path, struct kstat *stat, ret = orangefs_inode_getattr(inode, request_mask & STATX_SIZE ? ORANGEFS_GETATTR_SIZE : 0); if (ret == 0) { - generic_fillattr(inode, stat); + generic_fillattr(&init_user_ns, inode, stat); /* override block size reported to stat */ if (!(request_mask & STATX_SIZE)) @@ -919,7 +920,8 @@ int orangefs_getattr(const struct path *path, struct kstat *stat, return ret; } -int orangefs_permission(struct inode *inode, int mask) +int orangefs_permission(struct user_namespace *mnt_userns, + struct inode *inode, int mask) { int ret; @@ -933,7 +935,7 @@ int orangefs_permission(struct inode *inode, int mask) if (ret < 0) return ret; - return generic_permission(inode, mask); + return generic_permission(&init_user_ns, inode, mask); } int orangefs_update_time(struct inode *inode, struct timespec64 *time, int flags) diff --git a/fs/orangefs/namei.c b/fs/orangefs/namei.c index 3e7cf3d0a494..600e8eee541f 100644 --- a/fs/orangefs/namei.c +++ b/fs/orangefs/namei.c @@ -15,7 +15,8 @@ /* * Get a newly allocated inode to go with a negative dentry. */ -static int orangefs_create(struct inode *dir, +static int orangefs_create(struct user_namespace *mnt_userns, + struct inode *dir, struct dentry *dentry, umode_t mode, bool exclusive) @@ -215,7 +216,8 @@ static int orangefs_unlink(struct inode *dir, struct dentry *dentry) return ret; } -static int orangefs_symlink(struct inode *dir, +static int orangefs_symlink(struct user_namespace *mnt_userns, + struct inode *dir, struct dentry *dentry, const char *symname) { @@ -303,7 +305,8 @@ out: return ret; } -static int orangefs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) +static int orangefs_mkdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { struct orangefs_inode_s *parent = ORANGEFS_I(dir); struct orangefs_kernel_op_s *new_op; @@ -372,7 +375,8 @@ out: return ret; } -static int orangefs_rename(struct inode *old_dir, +static int orangefs_rename(struct user_namespace *mnt_userns, + struct inode *old_dir, struct dentry *old_dentry, struct inode *new_dir, struct dentry *new_dentry, diff --git a/fs/orangefs/orangefs-kernel.h b/fs/orangefs/orangefs-kernel.h index e12aeb9623d6..0e6b97682e41 100644 --- a/fs/orangefs/orangefs-kernel.h +++ b/fs/orangefs/orangefs-kernel.h @@ -107,7 +107,9 @@ extern int orangefs_init_acl(struct inode *inode, struct inode *dir); extern const struct xattr_handler *orangefs_xattr_handlers[]; extern struct posix_acl *orangefs_get_acl(struct inode *inode, int type); -extern int orangefs_set_acl(struct inode *inode, struct posix_acl *acl, int type); +extern int orangefs_set_acl(struct user_namespace *mnt_userns, + struct inode *inode, struct posix_acl *acl, + int type); /* * orangefs data structures @@ -359,12 +361,13 @@ struct inode *orangefs_new_inode(struct super_block *sb, struct orangefs_object_kref *ref); int __orangefs_setattr(struct inode *, struct iattr *); -int orangefs_setattr(struct dentry *, struct iattr *); +int orangefs_setattr(struct user_namespace *, struct dentry *, struct iattr *); -int orangefs_getattr(const struct path *path, struct kstat *stat, - u32 request_mask, unsigned int flags); +int orangefs_getattr(struct user_namespace *mnt_userns, const struct path *path, + struct kstat *stat, u32 request_mask, unsigned int flags); -int orangefs_permission(struct inode *inode, int mask); +int orangefs_permission(struct user_namespace *mnt_userns, + struct inode *inode, int mask); int orangefs_update_time(struct inode *, struct timespec64 *, int); diff --git a/fs/orangefs/xattr.c b/fs/orangefs/xattr.c index bdc285aea360..9a5b757fbd2f 100644 --- a/fs/orangefs/xattr.c +++ b/fs/orangefs/xattr.c @@ -526,6 +526,7 @@ out_unlock: } static int orangefs_xattr_set_default(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *unused, struct inode *inode, const char *name, diff --git a/fs/overlayfs/copy_up.c b/fs/overlayfs/copy_up.c index 0fed532efa68..0b2891c6c71e 100644 --- a/fs/overlayfs/copy_up.c +++ b/fs/overlayfs/copy_up.c @@ -93,9 +93,9 @@ int ovl_copy_xattr(struct super_block *sb, struct dentry *old, continue; /* Discard */ } retry: - size = vfs_getxattr(old, name, value, value_size); + size = vfs_getxattr(&init_user_ns, old, name, value, value_size); if (size == -ERANGE) - size = vfs_getxattr(old, name, NULL, 0); + size = vfs_getxattr(&init_user_ns, old, name, NULL, 0); if (size < 0) { error = size; @@ -115,7 +115,7 @@ retry: goto retry; } - error = vfs_setxattr(new, name, value, size, 0); + error = vfs_setxattr(&init_user_ns, new, name, value, size, 0); if (error) { if (error != -EOPNOTSUPP || ovl_must_copy_xattr(name)) break; @@ -236,7 +236,7 @@ static int ovl_set_size(struct dentry *upperdentry, struct kstat *stat) .ia_size = stat->size, }; - return notify_change(upperdentry, &attr, NULL); + return notify_change(&init_user_ns, upperdentry, &attr, NULL); } static int ovl_set_timestamps(struct dentry *upperdentry, struct kstat *stat) @@ -248,7 +248,7 @@ static int ovl_set_timestamps(struct dentry *upperdentry, struct kstat *stat) .ia_mtime = stat->mtime, }; - return notify_change(upperdentry, &attr, NULL); + return notify_change(&init_user_ns, upperdentry, &attr, NULL); } int ovl_set_attr(struct dentry *upperdentry, struct kstat *stat) @@ -260,7 +260,7 @@ int ovl_set_attr(struct dentry *upperdentry, struct kstat *stat) .ia_valid = ATTR_MODE, .ia_mode = stat->mode, }; - err = notify_change(upperdentry, &attr, NULL); + err = notify_change(&init_user_ns, upperdentry, &attr, NULL); } if (!err) { struct iattr attr = { @@ -268,7 +268,7 @@ int ovl_set_attr(struct dentry *upperdentry, struct kstat *stat) .ia_uid = stat->uid, .ia_gid = stat->gid, }; - err = notify_change(upperdentry, &attr, NULL); + err = notify_change(&init_user_ns, upperdentry, &attr, NULL); } if (!err) ovl_set_timestamps(upperdentry, stat); @@ -796,7 +796,7 @@ static ssize_t ovl_getxattr(struct dentry *dentry, char *name, char **value) ssize_t res; char *buf; - res = vfs_getxattr(dentry, name, NULL, 0); + res = vfs_getxattr(&init_user_ns, dentry, name, NULL, 0); if (res == -ENODATA || res == -EOPNOTSUPP) res = 0; @@ -805,7 +805,7 @@ static ssize_t ovl_getxattr(struct dentry *dentry, char *name, char **value) if (!buf) return -ENOMEM; - res = vfs_getxattr(dentry, name, buf, res); + res = vfs_getxattr(&init_user_ns, dentry, name, buf, res); if (res < 0) kfree(buf); else @@ -847,8 +847,8 @@ static int ovl_copy_up_meta_inode_data(struct ovl_copy_up_ctx *c) * don't want that to happen for normal copy-up operation. */ if (capability) { - err = vfs_setxattr(upperpath.dentry, XATTR_NAME_CAPS, - capability, cap_size, 0); + err = vfs_setxattr(&init_user_ns, upperpath.dentry, + XATTR_NAME_CAPS, capability, cap_size, 0); if (err) goto out_free; } diff --git a/fs/overlayfs/dir.c b/fs/overlayfs/dir.c index d1efa3a5a503..836f14b9d3a6 100644 --- a/fs/overlayfs/dir.c +++ b/fs/overlayfs/dir.c @@ -449,7 +449,7 @@ static int ovl_set_upper_acl(struct dentry *upperdentry, const char *name, if (err < 0) goto out_free; - err = vfs_setxattr(upperdentry, name, buffer, size, XATTR_CREATE); + err = vfs_setxattr(&init_user_ns, upperdentry, name, buffer, size, XATTR_CREATE); out_free: kfree(buffer); return err; @@ -508,7 +508,7 @@ static int ovl_create_over_whiteout(struct dentry *dentry, struct inode *inode, .ia_mode = cattr->mode, }; inode_lock(newdentry->d_inode); - err = notify_change(newdentry, &attr, NULL); + err = notify_change(&init_user_ns, newdentry, &attr, NULL); inode_unlock(newdentry->d_inode); if (err) goto out_cleanup; @@ -636,7 +636,7 @@ static int ovl_create_object(struct dentry *dentry, int mode, dev_t rdev, inode->i_state |= I_CREATING; spin_unlock(&inode->i_lock); - inode_init_owner(inode, dentry->d_parent->d_inode, mode); + inode_init_owner(&init_user_ns, inode, dentry->d_parent->d_inode, mode); attr.mode = inode->i_mode; err = ovl_create_or_link(dentry, inode, &attr, false); @@ -650,19 +650,20 @@ out: return err; } -static int ovl_create(struct inode *dir, struct dentry *dentry, umode_t mode, - bool excl) +static int ovl_create(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, bool excl) { return ovl_create_object(dentry, (mode & 07777) | S_IFREG, 0, NULL); } -static int ovl_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) +static int ovl_mkdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { return ovl_create_object(dentry, (mode & 07777) | S_IFDIR, 0, NULL); } -static int ovl_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, - dev_t rdev) +static int ovl_mknod(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, dev_t rdev) { /* Don't allow creation of "whiteout" on overlay */ if (S_ISCHR(mode) && rdev == WHITEOUT_DEV) @@ -671,8 +672,8 @@ static int ovl_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, return ovl_create_object(dentry, mode, rdev, NULL); } -static int ovl_symlink(struct inode *dir, struct dentry *dentry, - const char *link) +static int ovl_symlink(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, const char *link) { return ovl_create_object(dentry, S_IFLNK, 0, link); } @@ -821,9 +822,9 @@ static int ovl_remove_upper(struct dentry *dentry, bool is_dir, goto out_dput_upper; if (is_dir) - err = vfs_rmdir(dir, upper); + err = vfs_rmdir(&init_user_ns, dir, upper); else - err = vfs_unlink(dir, upper, NULL); + err = vfs_unlink(&init_user_ns, dir, upper, NULL); ovl_dir_modified(dentry->d_parent, ovl_type_origin(dentry)); /* @@ -1069,9 +1070,9 @@ static int ovl_set_redirect(struct dentry *dentry, bool samedir) return err; } -static int ovl_rename(struct inode *olddir, struct dentry *old, - struct inode *newdir, struct dentry *new, - unsigned int flags) +static int ovl_rename(struct user_namespace *mnt_userns, struct inode *olddir, + struct dentry *old, struct inode *newdir, + struct dentry *new, unsigned int flags) { int err; struct dentry *old_upperdir; diff --git a/fs/overlayfs/file.c b/fs/overlayfs/file.c index 077d3ad343f6..dbfb35fb0ff7 100644 --- a/fs/overlayfs/file.c +++ b/fs/overlayfs/file.c @@ -50,11 +50,11 @@ static struct file *ovl_open_realfile(const struct file *file, acc_mode |= MAY_APPEND; old_cred = ovl_override_creds(inode->i_sb); - err = inode_permission(realinode, MAY_OPEN | acc_mode); + err = inode_permission(&init_user_ns, realinode, MAY_OPEN | acc_mode); if (err) { realfile = ERR_PTR(err); } else { - if (!inode_owner_or_capable(realinode)) + if (!inode_owner_or_capable(&init_user_ns, realinode)) flags &= ~O_NOATIME; realfile = open_with_fake_path(&file->f_path, flags, realinode, @@ -521,7 +521,7 @@ static long ovl_ioctl_set_flags(struct file *file, unsigned int cmd, long ret; struct inode *inode = file_inode(file); - if (!inode_owner_or_capable(inode)) + if (!inode_owner_or_capable(&init_user_ns, inode)) return -EACCES; ret = mnt_want_write_file(file); diff --git a/fs/overlayfs/inode.c b/fs/overlayfs/inode.c index cf41bcb664bc..003cf83bf78a 100644 --- a/fs/overlayfs/inode.c +++ b/fs/overlayfs/inode.c @@ -14,14 +14,15 @@ #include "overlayfs.h" -int ovl_setattr(struct dentry *dentry, struct iattr *attr) +int ovl_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *attr) { int err; bool full_copy_up = false; struct dentry *upperdentry; const struct cred *old_cred; - err = setattr_prepare(dentry, attr); + err = setattr_prepare(&init_user_ns, dentry, attr); if (err) return err; @@ -79,7 +80,7 @@ int ovl_setattr(struct dentry *dentry, struct iattr *attr) inode_lock(upperdentry->d_inode); old_cred = ovl_override_creds(dentry->d_sb); - err = notify_change(upperdentry, attr, NULL); + err = notify_change(&init_user_ns, upperdentry, attr, NULL); revert_creds(old_cred); if (!err) ovl_copyattr(upperdentry->d_inode, dentry->d_inode); @@ -154,8 +155,8 @@ static int ovl_map_dev_ino(struct dentry *dentry, struct kstat *stat, int fsid) return 0; } -int ovl_getattr(const struct path *path, struct kstat *stat, - u32 request_mask, unsigned int flags) +int ovl_getattr(struct user_namespace *mnt_userns, const struct path *path, + struct kstat *stat, u32 request_mask, unsigned int flags) { struct dentry *dentry = path->dentry; enum ovl_path_type type; @@ -277,7 +278,8 @@ out: return err; } -int ovl_permission(struct inode *inode, int mask) +int ovl_permission(struct user_namespace *mnt_userns, + struct inode *inode, int mask) { struct inode *upperinode = ovl_inode_upper(inode); struct inode *realinode = upperinode ?: ovl_inode_lower(inode); @@ -294,7 +296,7 @@ int ovl_permission(struct inode *inode, int mask) * Check overlay inode with the creds of task and underlying inode * with creds of mounter */ - err = generic_permission(inode, mask); + err = generic_permission(&init_user_ns, inode, mask); if (err) return err; @@ -305,7 +307,7 @@ int ovl_permission(struct inode *inode, int mask) /* Make sure mounter can read file for copy up later */ mask |= MAY_READ; } - err = inode_permission(realinode, mask); + err = inode_permission(&init_user_ns, realinode, mask); revert_creds(old_cred); return err; @@ -353,7 +355,7 @@ int ovl_xattr_set(struct dentry *dentry, struct inode *inode, const char *name, if (!value && !upperdentry) { old_cred = ovl_override_creds(dentry->d_sb); - err = vfs_getxattr(realdentry, name, NULL, 0); + err = vfs_getxattr(&init_user_ns, realdentry, name, NULL, 0); revert_creds(old_cred); if (err < 0) goto out_drop_write; @@ -369,10 +371,11 @@ int ovl_xattr_set(struct dentry *dentry, struct inode *inode, const char *name, old_cred = ovl_override_creds(dentry->d_sb); if (value) - err = vfs_setxattr(realdentry, name, value, size, flags); + err = vfs_setxattr(&init_user_ns, realdentry, name, value, size, + flags); else { WARN_ON(flags != XATTR_REPLACE); - err = vfs_removexattr(realdentry, name); + err = vfs_removexattr(&init_user_ns, realdentry, name); } revert_creds(old_cred); @@ -394,7 +397,7 @@ int ovl_xattr_get(struct dentry *dentry, struct inode *inode, const char *name, ovl_i_dentry_upper(inode) ?: ovl_dentry_lower(dentry); old_cred = ovl_override_creds(dentry->d_sb); - res = vfs_getxattr(realdentry, name, value, size); + res = vfs_getxattr(&init_user_ns, realdentry, name, value, size); revert_creds(old_cred); return res; } diff --git a/fs/overlayfs/overlayfs.h b/fs/overlayfs/overlayfs.h index cb4e2d60ecf9..95cff83786a5 100644 --- a/fs/overlayfs/overlayfs.h +++ b/fs/overlayfs/overlayfs.h @@ -123,7 +123,7 @@ static inline const char *ovl_xattr(struct ovl_fs *ofs, enum ovl_xattr ox) static inline int ovl_do_rmdir(struct inode *dir, struct dentry *dentry) { - int err = vfs_rmdir(dir, dentry); + int err = vfs_rmdir(&init_user_ns, dir, dentry); pr_debug("rmdir(%pd2) = %i\n", dentry, err); return err; @@ -131,7 +131,7 @@ static inline int ovl_do_rmdir(struct inode *dir, struct dentry *dentry) static inline int ovl_do_unlink(struct inode *dir, struct dentry *dentry) { - int err = vfs_unlink(dir, dentry, NULL); + int err = vfs_unlink(&init_user_ns, dir, dentry, NULL); pr_debug("unlink(%pd2) = %i\n", dentry, err); return err; @@ -140,7 +140,7 @@ static inline int ovl_do_unlink(struct inode *dir, struct dentry *dentry) static inline int ovl_do_link(struct dentry *old_dentry, struct inode *dir, struct dentry *new_dentry) { - int err = vfs_link(old_dentry, dir, new_dentry, NULL); + int err = vfs_link(old_dentry, &init_user_ns, dir, new_dentry, NULL); pr_debug("link(%pd2, %pd2) = %i\n", old_dentry, new_dentry, err); return err; @@ -149,7 +149,7 @@ static inline int ovl_do_link(struct dentry *old_dentry, struct inode *dir, static inline int ovl_do_create(struct inode *dir, struct dentry *dentry, umode_t mode) { - int err = vfs_create(dir, dentry, mode, true); + int err = vfs_create(&init_user_ns, dir, dentry, mode, true); pr_debug("create(%pd2, 0%o) = %i\n", dentry, mode, err); return err; @@ -158,7 +158,7 @@ static inline int ovl_do_create(struct inode *dir, struct dentry *dentry, static inline int ovl_do_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) { - int err = vfs_mkdir(dir, dentry, mode); + int err = vfs_mkdir(&init_user_ns, dir, dentry, mode); pr_debug("mkdir(%pd2, 0%o) = %i\n", dentry, mode, err); return err; } @@ -166,7 +166,7 @@ static inline int ovl_do_mkdir(struct inode *dir, struct dentry *dentry, static inline int ovl_do_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t dev) { - int err = vfs_mknod(dir, dentry, mode, dev); + int err = vfs_mknod(&init_user_ns, dir, dentry, mode, dev); pr_debug("mknod(%pd2, 0%o, 0%o) = %i\n", dentry, mode, dev, err); return err; @@ -175,7 +175,7 @@ static inline int ovl_do_mknod(struct inode *dir, struct dentry *dentry, static inline int ovl_do_symlink(struct inode *dir, struct dentry *dentry, const char *oldname) { - int err = vfs_symlink(dir, dentry, oldname); + int err = vfs_symlink(&init_user_ns, dir, dentry, oldname); pr_debug("symlink(\"%s\", %pd2) = %i\n", oldname, dentry, err); return err; @@ -186,7 +186,7 @@ static inline ssize_t ovl_do_getxattr(struct ovl_fs *ofs, struct dentry *dentry, size_t size) { const char *name = ovl_xattr(ofs, ox); - return vfs_getxattr(dentry, name, value, size); + return vfs_getxattr(&init_user_ns, dentry, name, value, size); } static inline int ovl_do_setxattr(struct ovl_fs *ofs, struct dentry *dentry, @@ -194,7 +194,7 @@ static inline int ovl_do_setxattr(struct ovl_fs *ofs, struct dentry *dentry, size_t size) { const char *name = ovl_xattr(ofs, ox); - int err = vfs_setxattr(dentry, name, value, size, 0); + int err = vfs_setxattr(&init_user_ns, dentry, name, value, size, 0); pr_debug("setxattr(%pd2, \"%s\", \"%*pE\", %zu, 0) = %i\n", dentry, name, min((int)size, 48), value, size, err); return err; @@ -204,7 +204,7 @@ static inline int ovl_do_removexattr(struct ovl_fs *ofs, struct dentry *dentry, enum ovl_xattr ox) { const char *name = ovl_xattr(ofs, ox); - int err = vfs_removexattr(dentry, name); + int err = vfs_removexattr(&init_user_ns, dentry, name); pr_debug("removexattr(%pd2, \"%s\") = %i\n", dentry, name, err); return err; } @@ -214,9 +214,18 @@ static inline int ovl_do_rename(struct inode *olddir, struct dentry *olddentry, unsigned int flags) { int err; + struct renamedata rd = { + .old_mnt_userns = &init_user_ns, + .old_dir = olddir, + .old_dentry = olddentry, + .new_mnt_userns = &init_user_ns, + .new_dir = newdir, + .new_dentry = newdentry, + .flags = flags, + }; pr_debug("rename(%pd2, %pd2, 0x%x)\n", olddentry, newdentry, flags); - err = vfs_rename(olddir, olddentry, newdir, newdentry, NULL, flags); + err = vfs_rename(&rd); if (err) { pr_debug("...rename(%pd2, %pd2, ...) = %i\n", olddentry, newdentry, err); @@ -226,14 +235,14 @@ static inline int ovl_do_rename(struct inode *olddir, struct dentry *olddentry, static inline int ovl_do_whiteout(struct inode *dir, struct dentry *dentry) { - int err = vfs_whiteout(dir, dentry); + int err = vfs_whiteout(&init_user_ns, dir, dentry); pr_debug("whiteout(%pd2) = %i\n", dentry, err); return err; } static inline struct dentry *ovl_do_tmpfile(struct dentry *dentry, umode_t mode) { - struct dentry *ret = vfs_tmpfile(dentry, mode, 0); + struct dentry *ret = vfs_tmpfile(&init_user_ns, dentry, mode, 0); int err = PTR_ERR_OR_ZERO(ret); pr_debug("tmpfile(%pd2, 0%o) = %i\n", dentry, mode, err); @@ -436,10 +445,12 @@ int ovl_set_nlink_lower(struct dentry *dentry); unsigned int ovl_get_nlink(struct ovl_fs *ofs, struct dentry *lowerdentry, struct dentry *upperdentry, unsigned int fallback); -int ovl_setattr(struct dentry *dentry, struct iattr *attr); -int ovl_getattr(const struct path *path, struct kstat *stat, - u32 request_mask, unsigned int flags); -int ovl_permission(struct inode *inode, int mask); +int ovl_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *attr); +int ovl_getattr(struct user_namespace *mnt_userns, const struct path *path, + struct kstat *stat, u32 request_mask, unsigned int flags); +int ovl_permission(struct user_namespace *mnt_userns, struct inode *inode, + int mask); int ovl_xattr_set(struct dentry *dentry, struct inode *inode, const char *name, const void *value, size_t size, int flags); int ovl_xattr_get(struct dentry *dentry, struct inode *inode, const char *name, diff --git a/fs/overlayfs/super.c b/fs/overlayfs/super.c index d58b8f2bf9d0..fdd72f1a9c5e 100644 --- a/fs/overlayfs/super.c +++ b/fs/overlayfs/super.c @@ -803,17 +803,19 @@ retry: * allowed as upper are limited to "normal" ones, where checking * for the above two errors is sufficient. */ - err = vfs_removexattr(work, XATTR_NAME_POSIX_ACL_DEFAULT); + err = vfs_removexattr(&init_user_ns, work, + XATTR_NAME_POSIX_ACL_DEFAULT); if (err && err != -ENODATA && err != -EOPNOTSUPP) goto out_dput; - err = vfs_removexattr(work, XATTR_NAME_POSIX_ACL_ACCESS); + err = vfs_removexattr(&init_user_ns, work, + XATTR_NAME_POSIX_ACL_ACCESS); if (err && err != -ENODATA && err != -EOPNOTSUPP) goto out_dput; /* Clear any inherited mode bits */ inode_lock(work->d_inode); - err = notify_change(work, &attr, NULL); + err = notify_change(&init_user_ns, work, &attr, NULL); inode_unlock(work->d_inode); if (err) goto out_dput; @@ -865,6 +867,10 @@ static int ovl_mount_dir_noesc(const char *name, struct path *path) pr_err("filesystem on '%s' not supported\n", name); goto out_put; } + if (mnt_user_ns(path->mnt) != &init_user_ns) { + pr_err("idmapped layers are currently not supported\n"); + goto out_put; + } if (!d_is_dir(path->dentry)) { pr_err("'%s' not a directory\n", name); goto out_put; @@ -989,6 +995,7 @@ ovl_posix_acl_xattr_get(const struct xattr_handler *handler, static int __maybe_unused ovl_posix_acl_xattr_set(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *dentry, struct inode *inode, const char *name, const void *value, size_t size, int flags) @@ -1014,7 +1021,7 @@ ovl_posix_acl_xattr_set(const struct xattr_handler *handler, goto out_acl_release; } err = -EPERM; - if (!inode_owner_or_capable(inode)) + if (!inode_owner_or_capable(&init_user_ns, inode)) goto out_acl_release; posix_acl_release(acl); @@ -1026,10 +1033,10 @@ ovl_posix_acl_xattr_set(const struct xattr_handler *handler, if (unlikely(inode->i_mode & S_ISGID) && handler->flags == ACL_TYPE_ACCESS && !in_group_p(inode->i_gid) && - !capable_wrt_inode_uidgid(inode, CAP_FSETID)) { + !capable_wrt_inode_uidgid(&init_user_ns, inode, CAP_FSETID)) { struct iattr iattr = { .ia_valid = ATTR_KILL_SGID }; - err = ovl_setattr(dentry, &iattr); + err = ovl_setattr(&init_user_ns, dentry, &iattr); if (err) return err; } @@ -1053,6 +1060,7 @@ static int ovl_own_xattr_get(const struct xattr_handler *handler, } static int ovl_own_xattr_set(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *dentry, struct inode *inode, const char *name, const void *value, size_t size, int flags) @@ -1068,6 +1076,7 @@ static int ovl_other_xattr_get(const struct xattr_handler *handler, } static int ovl_other_xattr_set(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *dentry, struct inode *inode, const char *name, const void *value, size_t size, int flags) diff --git a/fs/overlayfs/util.c b/fs/overlayfs/util.c index 9826b003f1d2..7f5a01a11f97 100644 --- a/fs/overlayfs/util.c +++ b/fs/overlayfs/util.c @@ -479,12 +479,12 @@ struct file *ovl_path_open(struct path *path, int flags) BUG(); } - err = inode_permission(inode, acc_mode | MAY_OPEN); + err = inode_permission(&init_user_ns, inode, acc_mode | MAY_OPEN); if (err) return ERR_PTR(err); /* O_NOATIME is an optimization, don't fail if not permitted */ - if (inode_owner_or_capable(inode)) + if (inode_owner_or_capable(&init_user_ns, inode)) flags |= O_NOATIME; return dentry_open(path, flags, current_cred()); diff --git a/fs/pipe.c b/fs/pipe.c index 39c96845a72f..bfd946a9ad01 100644 --- a/fs/pipe.c +++ b/fs/pipe.c @@ -171,7 +171,7 @@ EXPORT_SYMBOL(generic_pipe_buf_try_steal); * * Description: * This function grabs an extra reference to @buf. It's used in - * in the tee() system call, when we duplicate the buffers in one + * the tee() system call, when we duplicate the buffers in one * pipe into another. */ bool generic_pipe_buf_get(struct pipe_inode_info *pipe, struct pipe_buffer *buf) diff --git a/fs/pnode.h b/fs/pnode.h index 26f74e092bd9..988f1aa9b02a 100644 --- a/fs/pnode.h +++ b/fs/pnode.h @@ -12,7 +12,7 @@ #define IS_MNT_SHARED(m) ((m)->mnt.mnt_flags & MNT_SHARED) #define IS_MNT_SLAVE(m) ((m)->mnt_master) -#define IS_MNT_NEW(m) (!(m)->mnt_ns) +#define IS_MNT_NEW(m) (!(m)->mnt_ns || is_anon_ns((m)->mnt_ns)) #define CLEAR_MNT_SHARED(m) ((m)->mnt.mnt_flags &= ~MNT_SHARED) #define IS_MNT_UNBINDABLE(m) ((m)->mnt.mnt_flags & MNT_UNBINDABLE) #define IS_MNT_MARKED(m) ((m)->mnt.mnt_flags & MNT_MARKED) diff --git a/fs/posix_acl.c b/fs/posix_acl.c index 95882b3f5f62..f3309a7edb49 100644 --- a/fs/posix_acl.c +++ b/fs/posix_acl.c @@ -345,10 +345,13 @@ EXPORT_SYMBOL(posix_acl_from_mode); * by the acl. Returns -E... otherwise. */ int -posix_acl_permission(struct inode *inode, const struct posix_acl *acl, int want) +posix_acl_permission(struct user_namespace *mnt_userns, struct inode *inode, + const struct posix_acl *acl, int want) { const struct posix_acl_entry *pa, *pe, *mask_obj; int found = 0; + kuid_t uid; + kgid_t gid; want &= MAY_READ | MAY_WRITE | MAY_EXEC; @@ -356,22 +359,26 @@ posix_acl_permission(struct inode *inode, const struct posix_acl *acl, int want) switch(pa->e_tag) { case ACL_USER_OBJ: /* (May have been checked already) */ - if (uid_eq(inode->i_uid, current_fsuid())) + uid = i_uid_into_mnt(mnt_userns, inode); + if (uid_eq(uid, current_fsuid())) goto check_perm; break; case ACL_USER: - if (uid_eq(pa->e_uid, current_fsuid())) + uid = kuid_into_mnt(mnt_userns, pa->e_uid); + if (uid_eq(uid, current_fsuid())) goto mask; break; case ACL_GROUP_OBJ: - if (in_group_p(inode->i_gid)) { + gid = i_gid_into_mnt(mnt_userns, inode); + if (in_group_p(gid)) { found = 1; if ((pa->e_perm & want) == want) goto mask; } break; case ACL_GROUP: - if (in_group_p(pa->e_gid)) { + gid = kgid_into_mnt(mnt_userns, pa->e_gid); + if (in_group_p(gid)) { found = 1; if ((pa->e_perm & want) == want) goto mask; @@ -551,8 +558,22 @@ __posix_acl_chmod(struct posix_acl **acl, gfp_t gfp, umode_t mode) } EXPORT_SYMBOL(__posix_acl_chmod); +/** + * posix_acl_chmod - chmod a posix acl + * + * @mnt_userns: user namespace of the mount @inode was found from + * @inode: inode to check permissions on + * @mode: the new mode of @inode + * + * If the inode has been found through an idmapped mount the user namespace of + * the vfsmount must be passed through @mnt_userns. This function will then + * take care to map the inode according to @mnt_userns before checking + * permissions. On non-idmapped mounts or if permission checking is to be + * performed on the raw inode simply passs init_user_ns. + */ int -posix_acl_chmod(struct inode *inode, umode_t mode) + posix_acl_chmod(struct user_namespace *mnt_userns, struct inode *inode, + umode_t mode) { struct posix_acl *acl; int ret = 0; @@ -572,7 +593,7 @@ posix_acl_chmod(struct inode *inode, umode_t mode) ret = __posix_acl_chmod(&acl, GFP_KERNEL, mode); if (ret) return ret; - ret = inode->i_op->set_acl(inode, acl, ACL_TYPE_ACCESS); + ret = inode->i_op->set_acl(mnt_userns, inode, acl, ACL_TYPE_ACCESS); posix_acl_release(acl); return ret; } @@ -631,9 +652,10 @@ EXPORT_SYMBOL_GPL(posix_acl_create); /** * posix_acl_update_mode - update mode in set_acl - * @inode: target inode - * @mode_p: mode (pointer) for update - * @acl: acl pointer + * @mnt_userns: user namespace of the mount @inode was found from + * @inode: target inode + * @mode_p: mode (pointer) for update + * @acl: acl pointer * * Update the file mode when setting an ACL: compute the new file permission * bits based on the ACL. In addition, if the ACL is equivalent to the new @@ -642,9 +664,16 @@ EXPORT_SYMBOL_GPL(posix_acl_create); * As with chmod, clear the setgid bit if the caller is not in the owning group * or capable of CAP_FSETID (see inode_change_ok). * + * If the inode has been found through an idmapped mount the user namespace of + * the vfsmount must be passed through @mnt_userns. This function will then + * take care to map the inode according to @mnt_userns before checking + * permissions. On non-idmapped mounts or if permission checking is to be + * performed on the raw inode simply passs init_user_ns. + * * Called from set_acl inode operations. */ -int posix_acl_update_mode(struct inode *inode, umode_t *mode_p, +int posix_acl_update_mode(struct user_namespace *mnt_userns, + struct inode *inode, umode_t *mode_p, struct posix_acl **acl) { umode_t mode = inode->i_mode; @@ -655,8 +684,8 @@ int posix_acl_update_mode(struct inode *inode, umode_t *mode_p, return error; if (error == 0) *acl = NULL; - if (!in_group_p(inode->i_gid) && - !capable_wrt_inode_uidgid(inode, CAP_FSETID)) + if (!in_group_p(i_gid_into_mnt(mnt_userns, inode)) && + !capable_wrt_inode_uidgid(mnt_userns, inode, CAP_FSETID)) mode &= ~S_ISGID; *mode_p = mode; return 0; @@ -668,7 +697,8 @@ EXPORT_SYMBOL(posix_acl_update_mode); */ static void posix_acl_fix_xattr_userns( struct user_namespace *to, struct user_namespace *from, - void *value, size_t size) + struct user_namespace *mnt_userns, + void *value, size_t size, bool from_user) { struct posix_acl_xattr_header *header = value; struct posix_acl_xattr_entry *entry = (void *)(header + 1), *end; @@ -693,10 +723,18 @@ static void posix_acl_fix_xattr_userns( switch(le16_to_cpu(entry->e_tag)) { case ACL_USER: uid = make_kuid(from, le32_to_cpu(entry->e_id)); + if (from_user) + uid = kuid_from_mnt(mnt_userns, uid); + else + uid = kuid_into_mnt(mnt_userns, uid); entry->e_id = cpu_to_le32(from_kuid(to, uid)); break; case ACL_GROUP: gid = make_kgid(from, le32_to_cpu(entry->e_id)); + if (from_user) + gid = kgid_from_mnt(mnt_userns, gid); + else + gid = kgid_into_mnt(mnt_userns, gid); entry->e_id = cpu_to_le32(from_kgid(to, gid)); break; default: @@ -705,20 +743,24 @@ static void posix_acl_fix_xattr_userns( } } -void posix_acl_fix_xattr_from_user(void *value, size_t size) +void posix_acl_fix_xattr_from_user(struct user_namespace *mnt_userns, + void *value, size_t size) { struct user_namespace *user_ns = current_user_ns(); - if (user_ns == &init_user_ns) + if ((user_ns == &init_user_ns) && (mnt_userns == &init_user_ns)) return; - posix_acl_fix_xattr_userns(&init_user_ns, user_ns, value, size); + posix_acl_fix_xattr_userns(&init_user_ns, user_ns, mnt_userns, value, + size, true); } -void posix_acl_fix_xattr_to_user(void *value, size_t size) +void posix_acl_fix_xattr_to_user(struct user_namespace *mnt_userns, + void *value, size_t size) { struct user_namespace *user_ns = current_user_ns(); - if (user_ns == &init_user_ns) + if ((user_ns == &init_user_ns) && (mnt_userns == &init_user_ns)) return; - posix_acl_fix_xattr_userns(user_ns, &init_user_ns, value, size); + posix_acl_fix_xattr_userns(user_ns, &init_user_ns, mnt_userns, value, + size, false); } /* @@ -858,7 +900,8 @@ posix_acl_xattr_get(const struct xattr_handler *handler, } int -set_posix_acl(struct inode *inode, int type, struct posix_acl *acl) +set_posix_acl(struct user_namespace *mnt_userns, struct inode *inode, + int type, struct posix_acl *acl) { if (!IS_POSIXACL(inode)) return -EOPNOTSUPP; @@ -867,7 +910,7 @@ set_posix_acl(struct inode *inode, int type, struct posix_acl *acl) if (type == ACL_TYPE_DEFAULT && !S_ISDIR(inode->i_mode)) return acl ? -EACCES : 0; - if (!inode_owner_or_capable(inode)) + if (!inode_owner_or_capable(mnt_userns, inode)) return -EPERM; if (acl) { @@ -875,15 +918,16 @@ set_posix_acl(struct inode *inode, int type, struct posix_acl *acl) if (ret) return ret; } - return inode->i_op->set_acl(inode, acl, type); + return inode->i_op->set_acl(mnt_userns, inode, acl, type); } EXPORT_SYMBOL(set_posix_acl); static int posix_acl_xattr_set(const struct xattr_handler *handler, - struct dentry *unused, struct inode *inode, - const char *name, const void *value, - size_t size, int flags) + struct user_namespace *mnt_userns, + struct dentry *unused, struct inode *inode, + const char *name, const void *value, size_t size, + int flags) { struct posix_acl *acl = NULL; int ret; @@ -893,7 +937,7 @@ posix_acl_xattr_set(const struct xattr_handler *handler, if (IS_ERR(acl)) return PTR_ERR(acl); } - ret = set_posix_acl(inode, handler->flags, acl); + ret = set_posix_acl(mnt_userns, inode, handler->flags, acl); posix_acl_release(acl); return ret; } @@ -922,12 +966,13 @@ const struct xattr_handler posix_acl_default_xattr_handler = { }; EXPORT_SYMBOL_GPL(posix_acl_default_xattr_handler); -int simple_set_acl(struct inode *inode, struct posix_acl *acl, int type) +int simple_set_acl(struct user_namespace *mnt_userns, struct inode *inode, + struct posix_acl *acl, int type) { int error; if (type == ACL_TYPE_ACCESS) { - error = posix_acl_update_mode(inode, + error = posix_acl_update_mode(mnt_userns, inode, &inode->i_mode, &acl); if (error) return error; diff --git a/fs/proc/base.c b/fs/proc/base.c index b3422cda2a91..3851bfcdba56 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -67,7 +67,6 @@ #include <linux/mm.h> #include <linux/swap.h> #include <linux/rcupdate.h> -#include <linux/kallsyms.h> #include <linux/stacktrace.h> #include <linux/resource.h> #include <linux/module.h> @@ -386,19 +385,17 @@ static int proc_pid_wchan(struct seq_file *m, struct pid_namespace *ns, struct pid *pid, struct task_struct *task) { unsigned long wchan; - char symname[KSYM_NAME_LEN]; - if (!ptrace_may_access(task, PTRACE_MODE_READ_FSCREDS)) - goto print0; + if (ptrace_may_access(task, PTRACE_MODE_READ_FSCREDS)) + wchan = get_wchan(task); + else + wchan = 0; - wchan = get_wchan(task); - if (wchan && !lookup_symbol_name(wchan, symname)) { - seq_puts(m, symname); - return 0; - } + if (wchan) + seq_printf(m, "%ps", (void *) wchan); + else + seq_putc(m, '0'); -print0: - seq_putc(m, '0'); return 0; } #endif /* CONFIG_KALLSYMS */ @@ -685,7 +682,8 @@ static int proc_fd_access_allowed(struct inode *inode) return allowed; } -int proc_setattr(struct dentry *dentry, struct iattr *attr) +int proc_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *attr) { int error; struct inode *inode = d_inode(dentry); @@ -693,11 +691,11 @@ int proc_setattr(struct dentry *dentry, struct iattr *attr) if (attr->ia_valid & ATTR_MODE) return -EPERM; - error = setattr_prepare(dentry, attr); + error = setattr_prepare(&init_user_ns, dentry, attr); if (error) return error; - setattr_copy(inode, attr); + setattr_copy(&init_user_ns, inode, attr); mark_inode_dirty(inode); return 0; } @@ -726,7 +724,8 @@ static bool has_pid_permissions(struct proc_fs_info *fs_info, } -static int proc_pid_permission(struct inode *inode, int mask) +static int proc_pid_permission(struct user_namespace *mnt_userns, + struct inode *inode, int mask) { struct proc_fs_info *fs_info = proc_sb_info(inode->i_sb); struct task_struct *task; @@ -751,7 +750,7 @@ static int proc_pid_permission(struct inode *inode, int mask) return -EPERM; } - return generic_permission(inode, mask); + return generic_permission(&init_user_ns, inode, mask); } @@ -1927,14 +1926,14 @@ out_unlock: return NULL; } -int pid_getattr(const struct path *path, struct kstat *stat, - u32 request_mask, unsigned int query_flags) +int pid_getattr(struct user_namespace *mnt_userns, const struct path *path, + struct kstat *stat, u32 request_mask, unsigned int query_flags) { struct inode *inode = d_inode(path->dentry); struct proc_fs_info *fs_info = proc_sb_info(inode->i_sb); struct task_struct *task; - generic_fillattr(inode, stat); + generic_fillattr(&init_user_ns, inode, stat); stat->uid = GLOBAL_ROOT_UID; stat->gid = GLOBAL_ROOT_GID; @@ -3473,7 +3472,8 @@ int proc_pid_readdir(struct file *file, struct dir_context *ctx) * This function makes sure that the node is always accessible for members of * same thread group. */ -static int proc_tid_comm_permission(struct inode *inode, int mask) +static int proc_tid_comm_permission(struct user_namespace *mnt_userns, + struct inode *inode, int mask) { bool is_same_tgroup; struct task_struct *task; @@ -3492,7 +3492,7 @@ static int proc_tid_comm_permission(struct inode *inode, int mask) return 0; } - return generic_permission(inode, mask); + return generic_permission(&init_user_ns, inode, mask); } static const struct inode_operations proc_tid_comm_inode_operations = { @@ -3798,12 +3798,13 @@ static int proc_task_readdir(struct file *file, struct dir_context *ctx) return 0; } -static int proc_task_getattr(const struct path *path, struct kstat *stat, +static int proc_task_getattr(struct user_namespace *mnt_userns, + const struct path *path, struct kstat *stat, u32 request_mask, unsigned int query_flags) { struct inode *inode = d_inode(path->dentry); struct task_struct *p = get_proc_task(inode); - generic_fillattr(inode, stat); + generic_fillattr(&init_user_ns, inode, stat); if (p) { stat->nlink += get_nr_threads(p); diff --git a/fs/proc/fd.c b/fs/proc/fd.c index cb51763ed554..07fc4fad2602 100644 --- a/fs/proc/fd.c +++ b/fs/proc/fd.c @@ -276,12 +276,13 @@ static struct dentry *proc_lookupfd(struct inode *dir, struct dentry *dentry, * /proc/pid/fd needs a special permission handler so that a process can still * access /proc/self/fd after it has executed a setuid(). */ -int proc_fd_permission(struct inode *inode, int mask) +int proc_fd_permission(struct user_namespace *mnt_userns, + struct inode *inode, int mask) { struct task_struct *p; int rv; - rv = generic_permission(inode, mask); + rv = generic_permission(&init_user_ns, inode, mask); if (rv == 0) return rv; diff --git a/fs/proc/fd.h b/fs/proc/fd.h index f371a602bf58..c5a921a06a0b 100644 --- a/fs/proc/fd.h +++ b/fs/proc/fd.h @@ -10,7 +10,8 @@ extern const struct inode_operations proc_fd_inode_operations; extern const struct file_operations proc_fdinfo_operations; extern const struct inode_operations proc_fdinfo_inode_operations; -extern int proc_fd_permission(struct inode *inode, int mask); +extern int proc_fd_permission(struct user_namespace *mnt_userns, + struct inode *inode, int mask); static inline unsigned int proc_fd(struct inode *inode) { diff --git a/fs/proc/generic.c b/fs/proc/generic.c index 6c0a05f55d6b..bc86aa87cc41 100644 --- a/fs/proc/generic.c +++ b/fs/proc/generic.c @@ -115,17 +115,18 @@ static bool pde_subdir_insert(struct proc_dir_entry *dir, return true; } -static int proc_notify_change(struct dentry *dentry, struct iattr *iattr) +static int proc_notify_change(struct user_namespace *mnt_userns, + struct dentry *dentry, struct iattr *iattr) { struct inode *inode = d_inode(dentry); struct proc_dir_entry *de = PDE(inode); int error; - error = setattr_prepare(dentry, iattr); + error = setattr_prepare(&init_user_ns, dentry, iattr); if (error) return error; - setattr_copy(inode, iattr); + setattr_copy(&init_user_ns, inode, iattr); mark_inode_dirty(inode); proc_set_user(de, inode->i_uid, inode->i_gid); @@ -133,7 +134,8 @@ static int proc_notify_change(struct dentry *dentry, struct iattr *iattr) return 0; } -static int proc_getattr(const struct path *path, struct kstat *stat, +static int proc_getattr(struct user_namespace *mnt_userns, + const struct path *path, struct kstat *stat, u32 request_mask, unsigned int query_flags) { struct inode *inode = d_inode(path->dentry); @@ -145,7 +147,7 @@ static int proc_getattr(const struct path *path, struct kstat *stat, } } - generic_fillattr(inode, stat); + generic_fillattr(&init_user_ns, inode, stat); return 0; } diff --git a/fs/proc/internal.h b/fs/proc/internal.h index f60b379dcdc7..03415f3fb3a8 100644 --- a/fs/proc/internal.h +++ b/fs/proc/internal.h @@ -162,8 +162,10 @@ extern int proc_pid_statm(struct seq_file *, struct pid_namespace *, * base.c */ extern const struct dentry_operations pid_dentry_operations; -extern int pid_getattr(const struct path *, struct kstat *, u32, unsigned int); -extern int proc_setattr(struct dentry *, struct iattr *); +extern int pid_getattr(struct user_namespace *, const struct path *, + struct kstat *, u32, unsigned int); +extern int proc_setattr(struct user_namespace *, struct dentry *, + struct iattr *); extern void proc_pid_evict_inode(struct proc_inode *); extern struct inode *proc_pid_make_inode(struct super_block *, struct task_struct *, umode_t); extern void pid_update_inode(struct task_struct *, struct inode *); diff --git a/fs/proc/meminfo.c b/fs/proc/meminfo.c index d6fc74619625..6fa761c9cc78 100644 --- a/fs/proc/meminfo.c +++ b/fs/proc/meminfo.c @@ -129,15 +129,15 @@ static int meminfo_proc_show(struct seq_file *m, void *v) #ifdef CONFIG_TRANSPARENT_HUGEPAGE show_val_kb(m, "AnonHugePages: ", - global_node_page_state(NR_ANON_THPS) * HPAGE_PMD_NR); + global_node_page_state(NR_ANON_THPS)); show_val_kb(m, "ShmemHugePages: ", - global_node_page_state(NR_SHMEM_THPS) * HPAGE_PMD_NR); + global_node_page_state(NR_SHMEM_THPS)); show_val_kb(m, "ShmemPmdMapped: ", - global_node_page_state(NR_SHMEM_PMDMAPPED) * HPAGE_PMD_NR); + global_node_page_state(NR_SHMEM_PMDMAPPED)); show_val_kb(m, "FileHugePages: ", - global_node_page_state(NR_FILE_THPS) * HPAGE_PMD_NR); + global_node_page_state(NR_FILE_THPS)); show_val_kb(m, "FilePmdMapped: ", - global_node_page_state(NR_FILE_PMDMAPPED) * HPAGE_PMD_NR); + global_node_page_state(NR_FILE_PMDMAPPED)); #endif #ifdef CONFIG_CMA diff --git a/fs/proc/proc_net.c b/fs/proc/proc_net.c index 18601042af99..15c2e55d2ed2 100644 --- a/fs/proc/proc_net.c +++ b/fs/proc/proc_net.c @@ -289,7 +289,8 @@ static struct dentry *proc_tgid_net_lookup(struct inode *dir, return de; } -static int proc_tgid_net_getattr(const struct path *path, struct kstat *stat, +static int proc_tgid_net_getattr(struct user_namespace *mnt_userns, + const struct path *path, struct kstat *stat, u32 request_mask, unsigned int query_flags) { struct inode *inode = d_inode(path->dentry); @@ -297,7 +298,7 @@ static int proc_tgid_net_getattr(const struct path *path, struct kstat *stat, net = get_proc_task_net(inode); - generic_fillattr(inode, stat); + generic_fillattr(&init_user_ns, inode, stat); if (net != NULL) { stat->nlink = net->proc_net->nlink; diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c index d2018f70d1fa..984e42f8cb11 100644 --- a/fs/proc/proc_sysctl.c +++ b/fs/proc/proc_sysctl.c @@ -571,7 +571,7 @@ static ssize_t proc_sys_call_handler(struct kiocb *iocb, struct iov_iter *iter, error = -ENOMEM; if (count >= KMALLOC_MAX_SIZE) goto out; - kbuf = kzalloc(count + 1, GFP_KERNEL); + kbuf = kvzalloc(count + 1, GFP_KERNEL); if (!kbuf) goto out; @@ -600,7 +600,7 @@ static ssize_t proc_sys_call_handler(struct kiocb *iocb, struct iov_iter *iter, error = count; out_free_buf: - kfree(kbuf); + kvfree(kbuf); out: sysctl_head_finish(head); @@ -785,7 +785,8 @@ out: return 0; } -static int proc_sys_permission(struct inode *inode, int mask) +static int proc_sys_permission(struct user_namespace *mnt_userns, + struct inode *inode, int mask) { /* * sysctl entries that are not writeable, @@ -813,7 +814,8 @@ static int proc_sys_permission(struct inode *inode, int mask) return error; } -static int proc_sys_setattr(struct dentry *dentry, struct iattr *attr) +static int proc_sys_setattr(struct user_namespace *mnt_userns, + struct dentry *dentry, struct iattr *attr) { struct inode *inode = d_inode(dentry); int error; @@ -821,16 +823,17 @@ static int proc_sys_setattr(struct dentry *dentry, struct iattr *attr) if (attr->ia_valid & (ATTR_MODE | ATTR_UID | ATTR_GID)) return -EPERM; - error = setattr_prepare(dentry, attr); + error = setattr_prepare(&init_user_ns, dentry, attr); if (error) return error; - setattr_copy(inode, attr); + setattr_copy(&init_user_ns, inode, attr); mark_inode_dirty(inode); return 0; } -static int proc_sys_getattr(const struct path *path, struct kstat *stat, +static int proc_sys_getattr(struct user_namespace *mnt_userns, + const struct path *path, struct kstat *stat, u32 request_mask, unsigned int query_flags) { struct inode *inode = d_inode(path->dentry); @@ -840,7 +843,7 @@ static int proc_sys_getattr(const struct path *path, struct kstat *stat, if (IS_ERR(head)) return PTR_ERR(head); - generic_fillattr(inode, stat); + generic_fillattr(&init_user_ns, inode, stat); if (table) stat->mode = (stat->mode & S_IFMT) | table->mode; diff --git a/fs/proc/root.c b/fs/proc/root.c index 5e444d4f9717..c7e3b1350ef8 100644 --- a/fs/proc/root.c +++ b/fs/proc/root.c @@ -308,10 +308,11 @@ void __init proc_root_init(void) register_filesystem(&proc_fs_type); } -static int proc_root_getattr(const struct path *path, struct kstat *stat, +static int proc_root_getattr(struct user_namespace *mnt_userns, + const struct path *path, struct kstat *stat, u32 request_mask, unsigned int query_flags) { - generic_fillattr(d_inode(path->dentry), stat); + generic_fillattr(&init_user_ns, d_inode(path->dentry), stat); stat->nlink = proc_root.nlink + nr_processes(); return 0; } diff --git a/fs/proc/self.c b/fs/proc/self.c index cc71ce3466dc..72cd69bcaf4a 100644 --- a/fs/proc/self.c +++ b/fs/proc/self.c @@ -16,13 +16,6 @@ static const char *proc_self_get_link(struct dentry *dentry, pid_t tgid = task_tgid_nr_ns(current, ns); char *name; - /* - * Not currently supported. Once we can inherit all of struct pid, - * we can allow this. - */ - if (current->flags & PF_KTHREAD) - return ERR_PTR(-EOPNOTSUPP); - if (!tgid) return ERR_PTR(-ENOENT); /* max length of unsigned int in decimal + NULL term */ diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index 602e3a52884d..e862cab69583 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -1036,8 +1036,6 @@ struct clear_refs_private { #ifdef CONFIG_MEM_SOFT_DIRTY -#define is_cow_mapping(flags) (((flags) & (VM_SHARED | VM_MAYWRITE)) == VM_MAYWRITE) - static inline bool pte_is_pinned(struct vm_area_struct *vma, unsigned long addr, pte_t pte) { struct page *page; @@ -1210,7 +1208,6 @@ static ssize_t clear_refs_write(struct file *file, const char __user *buf, struct mm_struct *mm; struct vm_area_struct *vma; enum clear_refs_types type; - struct mmu_gather tlb; int itype; int rv; @@ -1249,7 +1246,6 @@ static ssize_t clear_refs_write(struct file *file, const char __user *buf, goto out_unlock; } - tlb_gather_mmu(&tlb, mm, 0, -1); if (type == CLEAR_REFS_SOFT_DIRTY) { for (vma = mm->mmap; vma; vma = vma->vm_next) { if (!(vma->vm_flags & VM_SOFTDIRTY)) @@ -1258,15 +1254,18 @@ static ssize_t clear_refs_write(struct file *file, const char __user *buf, vma_set_page_prot(vma); } + inc_tlb_flush_pending(mm); mmu_notifier_range_init(&range, MMU_NOTIFY_SOFT_DIRTY, 0, NULL, mm, 0, -1UL); mmu_notifier_invalidate_range_start(&range); } walk_page_range(mm, 0, mm->highest_vm_end, &clear_refs_walk_ops, &cp); - if (type == CLEAR_REFS_SOFT_DIRTY) + if (type == CLEAR_REFS_SOFT_DIRTY) { mmu_notifier_invalidate_range_end(&range); - tlb_finish_mmu(&tlb, 0, -1); + flush_tlb_mm(mm); + dec_tlb_flush_pending(mm); + } out_unlock: mmap_write_unlock(mm); out_mm: diff --git a/fs/proc/vmcore.c b/fs/proc/vmcore.c index c3a345c28a93..9a15334da208 100644 --- a/fs/proc/vmcore.c +++ b/fs/proc/vmcore.c @@ -1503,11 +1503,8 @@ int vmcore_add_device_dump(struct vmcoredd_data *data) return 0; out_err: - if (buf) - vfree(buf); - - if (dump) - vfree(dump); + vfree(buf); + vfree(dump); return ret; } diff --git a/fs/proc_namespace.c b/fs/proc_namespace.c index eafb75755fa3..392ef5162655 100644 --- a/fs/proc_namespace.c +++ b/fs/proc_namespace.c @@ -79,6 +79,9 @@ static void show_mnt_opts(struct seq_file *m, struct vfsmount *mnt) if (mnt->mnt_flags & fs_infop->flag) seq_puts(m, fs_infop->str); } + + if (mnt_user_ns(mnt) != &init_user_ns) + seq_puts(m, ",idmapped"); } static inline void mangle(struct seq_file *m, const char *s) diff --git a/fs/pstore/inode.c b/fs/pstore/inode.c index 93a217e4f563..14658b009f1b 100644 --- a/fs/pstore/inode.c +++ b/fs/pstore/inode.c @@ -467,7 +467,7 @@ static struct dentry *pstore_mount(struct file_system_type *fs_type, static void pstore_kill_sb(struct super_block *sb) { mutex_lock(&pstore_sb_lock); - WARN_ON(pstore_sb != sb); + WARN_ON(pstore_sb && pstore_sb != sb); kill_litter_super(sb); pstore_sb = NULL; diff --git a/fs/pstore/platform.c b/fs/pstore/platform.c index 32f64abc277c..d963ae7902f9 100644 --- a/fs/pstore/platform.c +++ b/fs/pstore/platform.c @@ -269,7 +269,7 @@ static int pstore_compress(const void *in, void *out, { int ret; - if (!IS_ENABLED(CONFIG_PSTORE_COMPRESSION)) + if (!IS_ENABLED(CONFIG_PSTORE_COMPRESS)) return -EINVAL; ret = crypto_comp_compress(tfm, in, inlen, out, &outlen); @@ -671,7 +671,7 @@ static void decompress_record(struct pstore_record *record) int unzipped_len; char *unzipped, *workspace; - if (!IS_ENABLED(CONFIG_PSTORE_COMPRESSION) || !record->compressed) + if (!IS_ENABLED(CONFIG_PSTORE_COMPRESS) || !record->compressed) return; /* Only PSTORE_TYPE_DMESG support compression. */ diff --git a/fs/pstore/ram_core.c b/fs/pstore/ram_core.c index aa8e0b65ff1a..fff363bfd484 100644 --- a/fs/pstore/ram_core.c +++ b/fs/pstore/ram_core.c @@ -246,7 +246,7 @@ static int persistent_ram_init_ecc(struct persistent_ram_zone *prz, pr_info("error in header, %d\n", numerr); prz->corrected_bytes += numerr; } else if (numerr < 0) { - pr_info("uncorrectable error in header\n"); + pr_info_ratelimited("uncorrectable error in header\n"); prz->bad_blocks++; } diff --git a/fs/pstore/zone.c b/fs/pstore/zone.c index 5266ccbec007..7c8f8feac6c3 100644 --- a/fs/pstore/zone.c +++ b/fs/pstore/zone.c @@ -23,7 +23,7 @@ #include "internal.h" /** - * struct psz_head - header of zone to flush to storage + * struct psz_buffer - header of zone to flush to storage * * @sig: signature to indicate header (PSZ_SIG xor PSZONE-type value) * @datalen: length of data in @data diff --git a/fs/quota/quota_v2.c b/fs/quota/quota_v2.c index c21106557a37..b1467f3921c2 100644 --- a/fs/quota/quota_v2.c +++ b/fs/quota/quota_v2.c @@ -164,19 +164,24 @@ static int v2_read_file_info(struct super_block *sb, int type) quota_error(sb, "Number of blocks too big for quota file size (%llu > %llu).", (loff_t)qinfo->dqi_blocks << qinfo->dqi_blocksize_bits, i_size_read(sb_dqopt(sb)->files[type])); - goto out; + goto out_free; } if (qinfo->dqi_free_blk >= qinfo->dqi_blocks) { quota_error(sb, "Free block number too big (%u >= %u).", qinfo->dqi_free_blk, qinfo->dqi_blocks); - goto out; + goto out_free; } if (qinfo->dqi_free_entry >= qinfo->dqi_blocks) { quota_error(sb, "Block with free entry too big (%u >= %u).", qinfo->dqi_free_entry, qinfo->dqi_blocks); - goto out; + goto out_free; } ret = 0; +out_free: + if (ret) { + kfree(info->dqi_priv); + info->dqi_priv = NULL; + } out: up_read(&dqopt->dqio_sem); return ret; diff --git a/fs/ramfs/file-nommu.c b/fs/ramfs/file-nommu.c index 355523f4a4bf..ba3525ccc27e 100644 --- a/fs/ramfs/file-nommu.c +++ b/fs/ramfs/file-nommu.c @@ -22,7 +22,7 @@ #include <linux/uaccess.h> #include "internal.h" -static int ramfs_nommu_setattr(struct dentry *, struct iattr *); +static int ramfs_nommu_setattr(struct user_namespace *, struct dentry *, struct iattr *); static unsigned long ramfs_nommu_get_unmapped_area(struct file *file, unsigned long addr, unsigned long len, @@ -158,14 +158,15 @@ static int ramfs_nommu_resize(struct inode *inode, loff_t newsize, loff_t size) * handle a change of attributes * - we're specifically interested in a change of size */ -static int ramfs_nommu_setattr(struct dentry *dentry, struct iattr *ia) +static int ramfs_nommu_setattr(struct user_namespace *mnt_userns, + struct dentry *dentry, struct iattr *ia) { struct inode *inode = d_inode(dentry); unsigned int old_ia_valid = ia->ia_valid; int ret = 0; /* POSIX UID/GID verification for setting inode attributes */ - ret = setattr_prepare(dentry, ia); + ret = setattr_prepare(&init_user_ns, dentry, ia); if (ret) return ret; @@ -185,7 +186,7 @@ static int ramfs_nommu_setattr(struct dentry *dentry, struct iattr *ia) } } - setattr_copy(inode, ia); + setattr_copy(&init_user_ns, inode, ia); out: ia->ia_valid = old_ia_valid; return ret; diff --git a/fs/ramfs/inode.c b/fs/ramfs/inode.c index ee179a81b3da..9ebd17d7befb 100644 --- a/fs/ramfs/inode.c +++ b/fs/ramfs/inode.c @@ -67,7 +67,7 @@ struct inode *ramfs_get_inode(struct super_block *sb, if (inode) { inode->i_ino = get_next_ino(); - inode_init_owner(inode, dir, mode); + inode_init_owner(&init_user_ns, inode, dir, mode); inode->i_mapping->a_ops = &ramfs_aops; mapping_set_gfp_mask(inode->i_mapping, GFP_HIGHUSER); mapping_set_unevictable(inode->i_mapping); @@ -101,7 +101,8 @@ struct inode *ramfs_get_inode(struct super_block *sb, */ /* SMP-safe */ static int -ramfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t dev) +ramfs_mknod(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, dev_t dev) { struct inode * inode = ramfs_get_inode(dir->i_sb, dir, mode, dev); int error = -ENOSPC; @@ -115,20 +116,23 @@ ramfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t dev) return error; } -static int ramfs_mkdir(struct inode * dir, struct dentry * dentry, umode_t mode) +static int ramfs_mkdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { - int retval = ramfs_mknod(dir, dentry, mode | S_IFDIR, 0); + int retval = ramfs_mknod(&init_user_ns, dir, dentry, mode | S_IFDIR, 0); if (!retval) inc_nlink(dir); return retval; } -static int ramfs_create(struct inode *dir, struct dentry *dentry, umode_t mode, bool excl) +static int ramfs_create(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, bool excl) { - return ramfs_mknod(dir, dentry, mode | S_IFREG, 0); + return ramfs_mknod(&init_user_ns, dir, dentry, mode | S_IFREG, 0); } -static int ramfs_symlink(struct inode * dir, struct dentry *dentry, const char * symname) +static int ramfs_symlink(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, const char *symname) { struct inode *inode; int error = -ENOSPC; @@ -147,6 +151,18 @@ static int ramfs_symlink(struct inode * dir, struct dentry *dentry, const char * return error; } +static int ramfs_tmpfile(struct user_namespace *mnt_userns, + struct inode *dir, struct dentry *dentry, umode_t mode) +{ + struct inode *inode; + + inode = ramfs_get_inode(dir->i_sb, dir, mode, 0); + if (!inode) + return -ENOSPC; + d_tmpfile(dentry, inode); + return 0; +} + static const struct inode_operations ramfs_dir_inode_operations = { .create = ramfs_create, .lookup = simple_lookup, @@ -157,6 +173,7 @@ static const struct inode_operations ramfs_dir_inode_operations = { .rmdir = simple_rmdir, .mknod = ramfs_mknod, .rename = simple_rename, + .tmpfile = ramfs_tmpfile, }; /* diff --git a/fs/read_write.c b/fs/read_write.c index 75f764b43418..9db7adf160d2 100644 --- a/fs/read_write.c +++ b/fs/read_write.c @@ -1188,6 +1188,7 @@ static ssize_t do_sendfile(int out_fd, int in_fd, loff_t *ppos, { struct fd in, out; struct inode *in_inode, *out_inode; + struct pipe_inode_info *opipe; loff_t pos; loff_t out_pos; ssize_t retval; @@ -1228,9 +1229,6 @@ static ssize_t do_sendfile(int out_fd, int in_fd, loff_t *ppos, in_inode = file_inode(in.file); out_inode = file_inode(out.file); out_pos = out.file->f_pos; - retval = rw_verify_area(WRITE, out.file, &out_pos, count); - if (retval < 0) - goto fput_out; if (!max) max = min(in_inode->i_sb->s_maxbytes, out_inode->i_sb->s_maxbytes); @@ -1253,9 +1251,18 @@ static ssize_t do_sendfile(int out_fd, int in_fd, loff_t *ppos, if (in.file->f_flags & O_NONBLOCK) fl = SPLICE_F_NONBLOCK; #endif - file_start_write(out.file); - retval = do_splice_direct(in.file, &pos, out.file, &out_pos, count, fl); - file_end_write(out.file); + opipe = get_pipe_info(out.file, true); + if (!opipe) { + retval = rw_verify_area(WRITE, out.file, &out_pos, count); + if (retval < 0) + goto fput_out; + file_start_write(out.file); + retval = do_splice_direct(in.file, &pos, out.file, &out_pos, + count, fl); + file_end_write(out.file); + } else { + retval = splice_file_to_pipe(in.file, opipe, &pos, count, fl); + } if (retval > 0) { add_rchar(current, retval); diff --git a/fs/reiserfs/acl.h b/fs/reiserfs/acl.h index 0c1c847f992f..fd58618da360 100644 --- a/fs/reiserfs/acl.h +++ b/fs/reiserfs/acl.h @@ -49,7 +49,8 @@ static inline int reiserfs_acl_count(size_t size) #ifdef CONFIG_REISERFS_FS_POSIX_ACL struct posix_acl *reiserfs_get_acl(struct inode *inode, int type); -int reiserfs_set_acl(struct inode *inode, struct posix_acl *acl, int type); +int reiserfs_set_acl(struct user_namespace *mnt_userns, struct inode *inode, + struct posix_acl *acl, int type); int reiserfs_acl_chmod(struct inode *inode); int reiserfs_inherit_default_acl(struct reiserfs_transaction_handle *th, struct inode *dir, struct dentry *dentry, diff --git a/fs/reiserfs/file.c b/fs/reiserfs/file.c index 0b641ae694f1..1db0254bc38b 100644 --- a/fs/reiserfs/file.c +++ b/fs/reiserfs/file.c @@ -159,7 +159,7 @@ static int reiserfs_sync_file(struct file *filp, loff_t start, loff_t end, barrier_done = reiserfs_commit_for_inode(inode); reiserfs_write_unlock(inode->i_sb); if (barrier_done != 1 && reiserfs_barrier_flush(inode->i_sb)) - blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL); + blkdev_issue_flush(inode->i_sb->s_bdev); inode_unlock(inode); if (barrier_done < 0) return barrier_done; diff --git a/fs/reiserfs/inode.c b/fs/reiserfs/inode.c index c76d563dec0e..780bb90c1804 100644 --- a/fs/reiserfs/inode.c +++ b/fs/reiserfs/inode.c @@ -3282,13 +3282,14 @@ static ssize_t reiserfs_direct_IO(struct kiocb *iocb, struct iov_iter *iter) return ret; } -int reiserfs_setattr(struct dentry *dentry, struct iattr *attr) +int reiserfs_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *attr) { struct inode *inode = d_inode(dentry); unsigned int ia_valid; int error; - error = setattr_prepare(dentry, attr); + error = setattr_prepare(&init_user_ns, dentry, attr); if (error) return error; @@ -3413,7 +3414,7 @@ int reiserfs_setattr(struct dentry *dentry, struct iattr *attr) } if (!error) { - setattr_copy(inode, attr); + setattr_copy(&init_user_ns, inode, attr); mark_inode_dirty(inode); } diff --git a/fs/reiserfs/ioctl.c b/fs/reiserfs/ioctl.c index adb21bea3d60..4f1cbd930179 100644 --- a/fs/reiserfs/ioctl.c +++ b/fs/reiserfs/ioctl.c @@ -59,7 +59,7 @@ long reiserfs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) if (err) break; - if (!inode_owner_or_capable(inode)) { + if (!inode_owner_or_capable(&init_user_ns, inode)) { err = -EPERM; goto setflags_out; } @@ -101,7 +101,7 @@ setflags_out: err = put_user(inode->i_generation, (int __user *)arg); break; case REISERFS_IOC_SETVERSION: - if (!inode_owner_or_capable(inode)) { + if (!inode_owner_or_capable(&init_user_ns, inode)) { err = -EPERM; break; } diff --git a/fs/reiserfs/namei.c b/fs/reiserfs/namei.c index 1594687582f0..e6eb05e2b2f1 100644 --- a/fs/reiserfs/namei.c +++ b/fs/reiserfs/namei.c @@ -615,12 +615,12 @@ static int new_inode_init(struct inode *inode, struct inode *dir, umode_t mode) * the quota init calls have to know who to charge the quota to, so * we have to set uid and gid here */ - inode_init_owner(inode, dir, mode); + inode_init_owner(&init_user_ns, inode, dir, mode); return dquot_initialize(inode); } -static int reiserfs_create(struct inode *dir, struct dentry *dentry, umode_t mode, - bool excl) +static int reiserfs_create(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, bool excl) { int retval; struct inode *inode; @@ -698,8 +698,8 @@ out_failed: return retval; } -static int reiserfs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, - dev_t rdev) +static int reiserfs_mknod(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, dev_t rdev) { int retval; struct inode *inode; @@ -781,7 +781,8 @@ out_failed: return retval; } -static int reiserfs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) +static int reiserfs_mkdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { int retval; struct inode *inode; @@ -1094,8 +1095,9 @@ out_unlink: return retval; } -static int reiserfs_symlink(struct inode *parent_dir, - struct dentry *dentry, const char *symname) +static int reiserfs_symlink(struct user_namespace *mnt_userns, + struct inode *parent_dir, struct dentry *dentry, + const char *symname) { int retval; struct inode *inode; @@ -1304,7 +1306,8 @@ static void set_ino_in_dir_entry(struct reiserfs_dir_entry *de, * one path. If it holds 2 or more, it can get into endless waiting in * get_empty_nodes or its clones */ -static int reiserfs_rename(struct inode *old_dir, struct dentry *old_dentry, +static int reiserfs_rename(struct user_namespace *mnt_userns, + struct inode *old_dir, struct dentry *old_dentry, struct inode *new_dir, struct dentry *new_dentry, unsigned int flags) { diff --git a/fs/reiserfs/reiserfs.h b/fs/reiserfs/reiserfs.h index f69871516167..0ca2ac62e534 100644 --- a/fs/reiserfs/reiserfs.h +++ b/fs/reiserfs/reiserfs.h @@ -3102,7 +3102,8 @@ static inline void reiserfs_update_sd(struct reiserfs_transaction_handle *th, } void sd_attrs_to_i_attrs(__u16 sd_attrs, struct inode *inode); -int reiserfs_setattr(struct dentry *dentry, struct iattr *attr); +int reiserfs_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *attr); int __reiserfs_write_begin(struct page *page, unsigned from, unsigned len); diff --git a/fs/reiserfs/xattr.c b/fs/reiserfs/xattr.c index fe63a7c3e0da..bd073836e141 100644 --- a/fs/reiserfs/xattr.c +++ b/fs/reiserfs/xattr.c @@ -66,14 +66,14 @@ static int xattr_create(struct inode *dir, struct dentry *dentry, int mode) { BUG_ON(!inode_is_locked(dir)); - return dir->i_op->create(dir, dentry, mode, true); + return dir->i_op->create(&init_user_ns, dir, dentry, mode, true); } #endif static int xattr_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) { BUG_ON(!inode_is_locked(dir)); - return dir->i_op->mkdir(dir, dentry, mode); + return dir->i_op->mkdir(&init_user_ns, dir, dentry, mode); } /* @@ -352,7 +352,7 @@ static int chown_one_xattr(struct dentry *dentry, void *data) * ATTR_MODE is set. */ attrs->ia_valid &= (ATTR_UID|ATTR_GID); - err = reiserfs_setattr(dentry, attrs); + err = reiserfs_setattr(&init_user_ns, dentry, attrs); attrs->ia_valid = ia_valid; return err; @@ -604,7 +604,7 @@ reiserfs_xattr_set_handle(struct reiserfs_transaction_handle *th, inode_lock_nested(d_inode(dentry), I_MUTEX_XATTR); inode_dio_wait(d_inode(dentry)); - err = reiserfs_setattr(dentry, &newattrs); + err = reiserfs_setattr(&init_user_ns, dentry, &newattrs); inode_unlock(d_inode(dentry)); } else update_ctime(inode); @@ -948,7 +948,8 @@ static int xattr_mount_check(struct super_block *s) return 0; } -int reiserfs_permission(struct inode *inode, int mask) +int reiserfs_permission(struct user_namespace *mnt_userns, struct inode *inode, + int mask) { /* * We don't do permission checks on the internal objects. @@ -957,7 +958,7 @@ int reiserfs_permission(struct inode *inode, int mask) if (IS_PRIVATE(inode)) return 0; - return generic_permission(inode, mask); + return generic_permission(&init_user_ns, inode, mask); } static int xattr_hide_revalidate(struct dentry *dentry, unsigned int flags) diff --git a/fs/reiserfs/xattr.h b/fs/reiserfs/xattr.h index c764352447ba..9b3b06da568c 100644 --- a/fs/reiserfs/xattr.h +++ b/fs/reiserfs/xattr.h @@ -16,7 +16,8 @@ int reiserfs_xattr_init(struct super_block *sb, int mount_flags); int reiserfs_lookup_privroot(struct super_block *sb); int reiserfs_delete_xattrs(struct inode *inode); int reiserfs_chown_xattrs(struct inode *inode, struct iattr *attrs); -int reiserfs_permission(struct inode *inode, int mask); +int reiserfs_permission(struct user_namespace *mnt_userns, + struct inode *inode, int mask); #ifdef CONFIG_REISERFS_FS_XATTR #define has_xattr_dir(inode) (REISERFS_I(inode)->i_flags & i_has_xattr_dir) diff --git a/fs/reiserfs/xattr_acl.c b/fs/reiserfs/xattr_acl.c index ccd40df6eb45..a9547144a099 100644 --- a/fs/reiserfs/xattr_acl.c +++ b/fs/reiserfs/xattr_acl.c @@ -18,7 +18,8 @@ static int __reiserfs_set_acl(struct reiserfs_transaction_handle *th, int -reiserfs_set_acl(struct inode *inode, struct posix_acl *acl, int type) +reiserfs_set_acl(struct user_namespace *mnt_userns, struct inode *inode, + struct posix_acl *acl, int type) { int error, error2; struct reiserfs_transaction_handle th; @@ -40,7 +41,8 @@ reiserfs_set_acl(struct inode *inode, struct posix_acl *acl, int type) reiserfs_write_unlock(inode->i_sb); if (error == 0) { if (type == ACL_TYPE_ACCESS && acl) { - error = posix_acl_update_mode(inode, &mode, &acl); + error = posix_acl_update_mode(&init_user_ns, inode, + &mode, &acl); if (error) goto unlock; update_mode = 1; @@ -399,5 +401,5 @@ int reiserfs_acl_chmod(struct inode *inode) !reiserfs_posixacl(inode->i_sb)) return 0; - return posix_acl_chmod(inode, inode->i_mode); + return posix_acl_chmod(&init_user_ns, inode, inode->i_mode); } diff --git a/fs/reiserfs/xattr_security.c b/fs/reiserfs/xattr_security.c index 20be9a0e5870..8965c8e5e172 100644 --- a/fs/reiserfs/xattr_security.c +++ b/fs/reiserfs/xattr_security.c @@ -21,7 +21,8 @@ security_get(const struct xattr_handler *handler, struct dentry *unused, } static int -security_set(const struct xattr_handler *handler, struct dentry *unused, +security_set(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *unused, struct inode *inode, const char *name, const void *buffer, size_t size, int flags) { diff --git a/fs/reiserfs/xattr_trusted.c b/fs/reiserfs/xattr_trusted.c index 5ed48da3d02b..d853cea2afcd 100644 --- a/fs/reiserfs/xattr_trusted.c +++ b/fs/reiserfs/xattr_trusted.c @@ -20,7 +20,8 @@ trusted_get(const struct xattr_handler *handler, struct dentry *unused, } static int -trusted_set(const struct xattr_handler *handler, struct dentry *unused, +trusted_set(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *unused, struct inode *inode, const char *name, const void *buffer, size_t size, int flags) { diff --git a/fs/reiserfs/xattr_user.c b/fs/reiserfs/xattr_user.c index a573ca45bacc..65d9cd10a5ea 100644 --- a/fs/reiserfs/xattr_user.c +++ b/fs/reiserfs/xattr_user.c @@ -18,7 +18,8 @@ user_get(const struct xattr_handler *handler, struct dentry *unused, } static int -user_set(const struct xattr_handler *handler, struct dentry *unused, +user_set(const struct xattr_handler *handler, struct user_namespace *mnt_userns, + struct dentry *unused, struct inode *inode, const char *name, const void *buffer, size_t size, int flags) { diff --git a/fs/remap_range.c b/fs/remap_range.c index 77dba3a49e65..e4a5fdd7ad7b 100644 --- a/fs/remap_range.c +++ b/fs/remap_range.c @@ -432,13 +432,16 @@ EXPORT_SYMBOL(vfs_clone_file_range); /* Check whether we are allowed to dedupe the destination file */ static bool allow_file_dedupe(struct file *file) { + struct user_namespace *mnt_userns = file_mnt_user_ns(file); + struct inode *inode = file_inode(file); + if (capable(CAP_SYS_ADMIN)) return true; if (file->f_mode & FMODE_WRITE) return true; - if (uid_eq(current_fsuid(), file_inode(file)->i_uid)) + if (uid_eq(current_fsuid(), i_uid_into_mnt(mnt_userns, inode))) return true; - if (!inode_permission(file_inode(file), MAY_WRITE)) + if (!inode_permission(mnt_userns, inode, MAY_WRITE)) return true; return false; } diff --git a/fs/select.c b/fs/select.c index 37aaa8317f3a..945896d0ac9e 100644 --- a/fs/select.c +++ b/fs/select.c @@ -1055,10 +1055,9 @@ static long do_restart_poll(struct restart_block *restart_block) ret = do_sys_poll(ufds, nfds, to); - if (ret == -ERESTARTNOHAND) { - restart_block->fn = do_restart_poll; - ret = -ERESTART_RESTARTBLOCK; - } + if (ret == -ERESTARTNOHAND) + ret = set_restart_fn(restart_block, do_restart_poll); + return ret; } @@ -1080,7 +1079,6 @@ SYSCALL_DEFINE3(poll, struct pollfd __user *, ufds, unsigned int, nfds, struct restart_block *restart_block; restart_block = ¤t->restart_block; - restart_block->fn = do_restart_poll; restart_block->poll.ufds = ufds; restart_block->poll.nfds = nfds; @@ -1091,7 +1089,7 @@ SYSCALL_DEFINE3(poll, struct pollfd __user *, ufds, unsigned int, nfds, } else restart_block->poll.has_timeout = 0; - ret = -ERESTART_RESTARTBLOCK; + ret = set_restart_fn(restart_block, do_restart_poll); } return ret; } diff --git a/fs/seq_file.c b/fs/seq_file.c index 03a369ccd28c..cb11a34fb871 100644 --- a/fs/seq_file.c +++ b/fs/seq_file.c @@ -669,7 +669,8 @@ void seq_puts(struct seq_file *m, const char *s) EXPORT_SYMBOL(seq_puts); /** - * A helper routine for putting decimal numbers without rich format of printf(). + * seq_put_decimal_ull_width - A helper routine for putting decimal numbers + * without rich format of printf(). * only 'unsigned long long' is supported. * @m: seq_file identifying the buffer to which data should be written * @delimiter: a string which is printed before the number @@ -1044,7 +1045,7 @@ struct hlist_node *seq_hlist_next_rcu(void *v, EXPORT_SYMBOL(seq_hlist_next_rcu); /** - * seq_hlist_start_precpu - start an iteration of a percpu hlist array + * seq_hlist_start_percpu - start an iteration of a percpu hlist array * @head: pointer to percpu array of struct hlist_heads * @cpu: pointer to cpu "cursor" * @pos: start position of sequence diff --git a/fs/splice.c b/fs/splice.c index 866d5c2367b2..5dbce4dcc1a7 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -662,12 +662,14 @@ iter_file_splice_write(struct pipe_inode_info *pipe, struct file *out, /* build the vector */ left = sd.total_len; - for (n = 0; !pipe_empty(head, tail) && left && n < nbufs; tail++, n++) { + for (n = 0; !pipe_empty(head, tail) && left && n < nbufs; tail++) { struct pipe_buffer *buf = &pipe->bufs[tail & mask]; size_t this_len = buf->len; - if (this_len > left) - this_len = left; + /* zero-length bvecs are not supported, skip them */ + if (!this_len) + continue; + this_len = min(this_len, left); ret = pipe_buf_confirm(pipe, buf); if (unlikely(ret)) { @@ -680,6 +682,7 @@ iter_file_splice_write(struct pipe_inode_info *pipe, struct file *out, array[n].bv_len = this_len; array[n].bv_offset = buf->offset; left -= this_len; + n++; } iov_iter_bvec(&from, WRITE, array, n, sd.total_len - left); @@ -771,11 +774,16 @@ static long do_splice_to(struct file *in, loff_t *ppos, struct pipe_inode_info *pipe, size_t len, unsigned int flags) { + unsigned int p_space; int ret; if (unlikely(!(in->f_mode & FMODE_READ))) return -EBADF; + /* Don't try to read more the pipe has space for. */ + p_space = pipe->max_usage - pipe_occupancy(pipe->head, pipe->tail); + len = min_t(size_t, len, p_space << PAGE_SHIFT); + ret = rw_verify_area(READ, in, ppos, len); if (unlikely(ret < 0)) return ret; @@ -856,15 +864,10 @@ ssize_t splice_direct_to_actor(struct file *in, struct splice_desc *sd, WARN_ON_ONCE(!pipe_empty(pipe->head, pipe->tail)); while (len) { - unsigned int p_space; size_t read_len; loff_t pos = sd->pos, prev_pos = pos; - /* Don't try to read more the pipe has space for. */ - p_space = pipe->max_usage - - pipe_occupancy(pipe->head, pipe->tail); - read_len = min_t(size_t, len, p_space << PAGE_SHIFT); - ret = do_splice_to(in, &pos, pipe, read_len, flags); + ret = do_splice_to(in, &pos, pipe, len, flags); if (unlikely(ret <= 0)) goto out_release; @@ -1002,6 +1005,23 @@ static int splice_pipe_to_pipe(struct pipe_inode_info *ipipe, struct pipe_inode_info *opipe, size_t len, unsigned int flags); +long splice_file_to_pipe(struct file *in, + struct pipe_inode_info *opipe, + loff_t *offset, + size_t len, unsigned int flags) +{ + long ret; + + pipe_lock(opipe); + ret = wait_for_space(opipe, flags); + if (!ret) + ret = do_splice_to(in, offset, opipe, len, flags); + pipe_unlock(opipe); + if (ret > 0) + wakeup_pipe_readers(opipe); + return ret; +} + /* * Determine where to splice to/from. */ @@ -1081,20 +1101,7 @@ long do_splice(struct file *in, loff_t *off_in, struct file *out, if (out->f_flags & O_NONBLOCK) flags |= SPLICE_F_NONBLOCK; - pipe_lock(opipe); - ret = wait_for_space(opipe, flags); - if (!ret) { - unsigned int p_space; - - /* Don't try to read more the pipe has space for. */ - p_space = opipe->max_usage - pipe_occupancy(opipe->head, opipe->tail); - len = min_t(size_t, len, p_space << PAGE_SHIFT); - - ret = do_splice_to(in, &offset, opipe, len, flags); - } - pipe_unlock(opipe); - if (ret > 0) - wakeup_pipe_readers(opipe); + ret = splice_file_to_pipe(in, opipe, &offset, len, flags); if (!off_in) in->f_pos = offset; else diff --git a/fs/squashfs/block.c b/fs/squashfs/block.c index 45f44425d856..b9e87ebb1060 100644 --- a/fs/squashfs/block.c +++ b/fs/squashfs/block.c @@ -87,7 +87,7 @@ static int squashfs_bio_read(struct super_block *sb, u64 index, int length, int error, i; struct bio *bio; - if (page_count <= BIO_MAX_PAGES) + if (page_count <= BIO_MAX_VECS) bio = bio_alloc(GFP_NOIO, page_count); else bio = bio_kmalloc(GFP_NOIO, page_count); diff --git a/fs/stat.c b/fs/stat.c index dacecdda2e79..fbc171d038aa 100644 --- a/fs/stat.c +++ b/fs/stat.c @@ -26,21 +26,29 @@ /** * generic_fillattr - Fill in the basic attributes from the inode struct - * @inode: Inode to use as the source - * @stat: Where to fill in the attributes + * @mnt_userns: user namespace of the mount the inode was found from + * @inode: Inode to use as the source + * @stat: Where to fill in the attributes * * Fill in the basic attributes in the kstat structure from data that's to be * found on the VFS inode structure. This is the default if no getattr inode * operation is supplied. + * + * If the inode has been found through an idmapped mount the user namespace of + * the vfsmount must be passed through @mnt_userns. This function will then + * take care to map the inode according to @mnt_userns before filling in the + * uid and gid filds. On non-idmapped mounts or if permission checking is to be + * performed on the raw inode simply passs init_user_ns. */ -void generic_fillattr(struct inode *inode, struct kstat *stat) +void generic_fillattr(struct user_namespace *mnt_userns, struct inode *inode, + struct kstat *stat) { stat->dev = inode->i_sb->s_dev; stat->ino = inode->i_ino; stat->mode = inode->i_mode; stat->nlink = inode->i_nlink; - stat->uid = inode->i_uid; - stat->gid = inode->i_gid; + stat->uid = i_uid_into_mnt(mnt_userns, inode); + stat->gid = i_gid_into_mnt(mnt_userns, inode); stat->rdev = inode->i_rdev; stat->size = i_size_read(inode); stat->atime = inode->i_atime; @@ -67,6 +75,7 @@ EXPORT_SYMBOL(generic_fillattr); int vfs_getattr_nosec(const struct path *path, struct kstat *stat, u32 request_mask, unsigned int query_flags) { + struct user_namespace *mnt_userns; struct inode *inode = d_backing_inode(path->dentry); memset(stat, 0, sizeof(*stat)); @@ -83,11 +92,12 @@ int vfs_getattr_nosec(const struct path *path, struct kstat *stat, if (IS_DAX(inode)) stat->attributes |= STATX_ATTR_DAX; + mnt_userns = mnt_user_ns(path->mnt); if (inode->i_op->getattr) - return inode->i_op->getattr(path, stat, request_mask, - query_flags); + return inode->i_op->getattr(mnt_userns, path, stat, + request_mask, query_flags); - generic_fillattr(inode, stat); + generic_fillattr(mnt_userns, inode, stat); return 0; } EXPORT_SYMBOL(vfs_getattr_nosec); diff --git a/fs/statfs.c b/fs/statfs.c index 68cb07788750..0ba34c135593 100644 --- a/fs/statfs.c +++ b/fs/statfs.c @@ -255,7 +255,10 @@ SYSCALL_DEFINE2(ustat, unsigned, dev, struct ustat __user *, ubuf) memset(&tmp,0,sizeof(struct ustat)); tmp.f_tfree = sbuf.f_bfree; - tmp.f_tinode = sbuf.f_ffree; + if (IS_ENABLED(CONFIG_ARCH_32BIT_USTAT_F_TINODE)) + tmp.f_tinode = min_t(u64, sbuf.f_ffree, UINT_MAX); + else + tmp.f_tinode = sbuf.f_ffree; return copy_to_user(ubuf, &tmp, sizeof(struct ustat)) ? -EFAULT : 0; } diff --git a/fs/super.c b/fs/super.c index 2c6cdea2ab2d..8c1baca35c16 100644 --- a/fs/super.c +++ b/fs/super.c @@ -865,7 +865,8 @@ int reconfigure_super(struct fs_context *fc) if (fc->sb_flags_mask & SB_RDONLY) { #ifdef CONFIG_BLOCK - if (!(fc->sb_flags & SB_RDONLY) && bdev_read_only(sb->s_bdev)) + if (!(fc->sb_flags & SB_RDONLY) && sb->s_bdev && + bdev_read_only(sb->s_bdev)) return -EACCES; #endif @@ -1718,12 +1719,6 @@ int freeze_super(struct super_block *sb) } EXPORT_SYMBOL(freeze_super); -/** - * thaw_super -- unlock filesystem - * @sb: the super to thaw - * - * Unlocks the filesystem and marks it writeable again after freeze_super(). - */ static int thaw_super_locked(struct super_block *sb) { int error; @@ -1759,6 +1754,12 @@ out: return 0; } +/** + * thaw_super -- unlock filesystem + * @sb: the super to thaw + * + * Unlocks the filesystem and marks it writeable again after freeze_super(). + */ int thaw_super(struct super_block *sb) { down_write(&sb->s_umount); diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c index 96d0da65e088..9aefa7779b29 100644 --- a/fs/sysfs/file.c +++ b/fs/sysfs/file.c @@ -170,6 +170,16 @@ static int sysfs_kf_bin_mmap(struct kernfs_open_file *of, return battr->mmap(of->file, kobj, battr, vma); } +static int sysfs_kf_bin_open(struct kernfs_open_file *of) +{ + struct bin_attribute *battr = of->kn->priv; + + if (battr->mapping) + of->file->f_mapping = battr->mapping; + + return 0; +} + void sysfs_notify(struct kobject *kobj, const char *dir, const char *attr) { struct kernfs_node *kn = kobj->sd, *tmp; @@ -241,6 +251,7 @@ static const struct kernfs_ops sysfs_bin_kfops_mmap = { .read = sysfs_kf_bin_read, .write = sysfs_kf_bin_write, .mmap = sysfs_kf_bin_mmap, + .open = sysfs_kf_bin_open, }; int sysfs_add_file_mode_ns(struct kernfs_node *parent, diff --git a/fs/sysv/file.c b/fs/sysv/file.c index 45fc79a18594..90e00124ea07 100644 --- a/fs/sysv/file.c +++ b/fs/sysv/file.c @@ -29,12 +29,13 @@ const struct file_operations sysv_file_operations = { .splice_read = generic_file_splice_read, }; -static int sysv_setattr(struct dentry *dentry, struct iattr *attr) +static int sysv_setattr(struct user_namespace *mnt_userns, + struct dentry *dentry, struct iattr *attr) { struct inode *inode = d_inode(dentry); int error; - error = setattr_prepare(dentry, attr); + error = setattr_prepare(&init_user_ns, dentry, attr); if (error) return error; @@ -47,7 +48,7 @@ static int sysv_setattr(struct dentry *dentry, struct iattr *attr) sysv_truncate(inode); } - setattr_copy(inode, attr); + setattr_copy(&init_user_ns, inode, attr); mark_inode_dirty(inode); return 0; } diff --git a/fs/sysv/ialloc.c b/fs/sysv/ialloc.c index 6c9801986af6..50df794a3c1f 100644 --- a/fs/sysv/ialloc.c +++ b/fs/sysv/ialloc.c @@ -163,7 +163,7 @@ struct inode * sysv_new_inode(const struct inode * dir, umode_t mode) *sbi->s_sb_fic_count = cpu_to_fs16(sbi, count); fs16_add(sbi, sbi->s_sb_total_free_inodes, -1); dirty_sb(sb); - inode_init_owner(inode, dir, mode); + inode_init_owner(&init_user_ns, inode, dir, mode); inode->i_ino = fs16_to_cpu(sbi, ino); inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode); inode->i_blocks = 0; diff --git a/fs/sysv/itree.c b/fs/sysv/itree.c index bcb67b0cabe7..8b2e99b7bc9f 100644 --- a/fs/sysv/itree.c +++ b/fs/sysv/itree.c @@ -441,11 +441,11 @@ static unsigned sysv_nblocks(struct super_block *s, loff_t size) return blocks; } -int sysv_getattr(const struct path *path, struct kstat *stat, - u32 request_mask, unsigned int flags) +int sysv_getattr(struct user_namespace *mnt_userns, const struct path *path, + struct kstat *stat, u32 request_mask, unsigned int flags) { struct super_block *s = path->dentry->d_sb; - generic_fillattr(d_inode(path->dentry), stat); + generic_fillattr(&init_user_ns, d_inode(path->dentry), stat); stat->blocks = (s->s_blocksize / 512) * sysv_nblocks(s, stat->size); stat->blksize = s->s_blocksize; return 0; diff --git a/fs/sysv/namei.c b/fs/sysv/namei.c index ea2414b385ec..b2e6abc06a2d 100644 --- a/fs/sysv/namei.c +++ b/fs/sysv/namei.c @@ -41,7 +41,8 @@ static struct dentry *sysv_lookup(struct inode * dir, struct dentry * dentry, un return d_splice_alias(inode, dentry); } -static int sysv_mknod(struct inode * dir, struct dentry * dentry, umode_t mode, dev_t rdev) +static int sysv_mknod(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, dev_t rdev) { struct inode * inode; int err; @@ -60,13 +61,14 @@ static int sysv_mknod(struct inode * dir, struct dentry * dentry, umode_t mode, return err; } -static int sysv_create(struct inode * dir, struct dentry * dentry, umode_t mode, bool excl) +static int sysv_create(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, bool excl) { - return sysv_mknod(dir, dentry, mode, 0); + return sysv_mknod(&init_user_ns, dir, dentry, mode, 0); } -static int sysv_symlink(struct inode * dir, struct dentry * dentry, - const char * symname) +static int sysv_symlink(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, const char *symname) { int err = -ENAMETOOLONG; int l = strlen(symname)+1; @@ -108,7 +110,8 @@ static int sysv_link(struct dentry * old_dentry, struct inode * dir, return add_nondir(dentry, inode); } -static int sysv_mkdir(struct inode * dir, struct dentry *dentry, umode_t mode) +static int sysv_mkdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { struct inode * inode; int err; @@ -186,9 +189,9 @@ static int sysv_rmdir(struct inode * dir, struct dentry * dentry) * Anybody can rename anything with this: the permission checks are left to the * higher-level routines. */ -static int sysv_rename(struct inode * old_dir, struct dentry * old_dentry, - struct inode * new_dir, struct dentry * new_dentry, - unsigned int flags) +static int sysv_rename(struct user_namespace *mnt_userns, struct inode *old_dir, + struct dentry *old_dentry, struct inode *new_dir, + struct dentry *new_dentry, unsigned int flags) { struct inode * old_inode = d_inode(old_dentry); struct inode * new_inode = d_inode(new_dentry); diff --git a/fs/sysv/sysv.h b/fs/sysv/sysv.h index 1cff585526b1..99ddf033da4f 100644 --- a/fs/sysv/sysv.h +++ b/fs/sysv/sysv.h @@ -141,7 +141,8 @@ extern struct inode *sysv_iget(struct super_block *, unsigned int); extern int sysv_write_inode(struct inode *, struct writeback_control *wbc); extern int sysv_sync_inode(struct inode *); extern void sysv_set_inode(struct inode *, dev_t); -extern int sysv_getattr(const struct path *, struct kstat *, u32, unsigned int); +extern int sysv_getattr(struct user_namespace *, const struct path *, + struct kstat *, u32, unsigned int); extern int sysv_init_icache(void); extern void sysv_destroy_icache(void); diff --git a/fs/tracefs/inode.c b/fs/tracefs/inode.c index 0ee8c6dfb036..4b83cbded559 100644 --- a/fs/tracefs/inode.c +++ b/fs/tracefs/inode.c @@ -67,7 +67,9 @@ static char *get_dname(struct dentry *dentry) return name; } -static int tracefs_syscall_mkdir(struct inode *inode, struct dentry *dentry, umode_t mode) +static int tracefs_syscall_mkdir(struct user_namespace *mnt_userns, + struct inode *inode, struct dentry *dentry, + umode_t mode) { char *name; int ret; diff --git a/fs/ubifs/auth.c b/fs/ubifs/auth.c index 51a7c8c2c3f0..e564d5ff8781 100644 --- a/fs/ubifs/auth.c +++ b/fs/ubifs/auth.c @@ -327,7 +327,7 @@ int ubifs_init_authentication(struct ubifs_info *c) ubifs_err(c, "hmac %s is bigger than maximum allowed hmac size (%d > %d)", hmac_name, c->hmac_desc_len, UBIFS_HMAC_ARR_SZ); err = -EINVAL; - goto out_free_hash; + goto out_free_hmac; } err = crypto_shash_setkey(c->hmac_tfm, ukp->data, ukp->datalen); diff --git a/fs/ubifs/dir.c b/fs/ubifs/dir.c index 9a6b8660425a..d9d8d7794eff 100644 --- a/fs/ubifs/dir.c +++ b/fs/ubifs/dir.c @@ -94,7 +94,7 @@ struct inode *ubifs_new_inode(struct ubifs_info *c, struct inode *dir, */ inode->i_flags |= S_NOCMTIME; - inode_init_owner(inode, dir, mode); + inode_init_owner(&init_user_ns, inode, dir, mode); inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode); inode->i_mapping->nrpages = 0; @@ -280,8 +280,8 @@ static int ubifs_prepare_create(struct inode *dir, struct dentry *dentry, return fscrypt_setup_filename(dir, &dentry->d_name, 0, nm); } -static int ubifs_create(struct inode *dir, struct dentry *dentry, umode_t mode, - bool excl) +static int ubifs_create(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, bool excl) { struct inode *inode; struct ubifs_info *c = dir->i_sb->s_fs_info; @@ -441,8 +441,8 @@ out_budg: return err; } -static int ubifs_tmpfile(struct inode *dir, struct dentry *dentry, - umode_t mode) +static int ubifs_tmpfile(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { return do_tmpfile(dir, dentry, mode, NULL); } @@ -942,7 +942,8 @@ out_fname: return err; } -static int ubifs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) +static int ubifs_mkdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { struct inode *inode; struct ubifs_inode *dir_ui = ubifs_inode(dir); @@ -1013,8 +1014,8 @@ out_budg: return err; } -static int ubifs_mknod(struct inode *dir, struct dentry *dentry, - umode_t mode, dev_t rdev) +static int ubifs_mknod(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, dev_t rdev) { struct inode *inode; struct ubifs_inode *ui; @@ -1102,8 +1103,8 @@ out_budg: return err; } -static int ubifs_symlink(struct inode *dir, struct dentry *dentry, - const char *symname) +static int ubifs_symlink(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, const char *symname) { struct inode *inode; struct ubifs_inode *ui; @@ -1542,7 +1543,8 @@ static int ubifs_xrename(struct inode *old_dir, struct dentry *old_dentry, return err; } -static int ubifs_rename(struct inode *old_dir, struct dentry *old_dentry, +static int ubifs_rename(struct user_namespace *mnt_userns, + struct inode *old_dir, struct dentry *old_dentry, struct inode *new_dir, struct dentry *new_dentry, unsigned int flags) { @@ -1566,8 +1568,8 @@ static int ubifs_rename(struct inode *old_dir, struct dentry *old_dentry, return do_rename(old_dir, old_dentry, new_dir, new_dentry, flags); } -int ubifs_getattr(const struct path *path, struct kstat *stat, - u32 request_mask, unsigned int flags) +int ubifs_getattr(struct user_namespace *mnt_userns, const struct path *path, + struct kstat *stat, u32 request_mask, unsigned int flags) { loff_t size; struct inode *inode = d_inode(path->dentry); @@ -1589,7 +1591,7 @@ int ubifs_getattr(const struct path *path, struct kstat *stat, STATX_ATTR_ENCRYPTED | STATX_ATTR_IMMUTABLE); - generic_fillattr(inode, stat); + generic_fillattr(&init_user_ns, inode, stat); stat->blksize = UBIFS_BLOCK_SIZE; stat->size = ui->ui_size; diff --git a/fs/ubifs/file.c b/fs/ubifs/file.c index 2bc7780d2963..0e4b4be3aa26 100644 --- a/fs/ubifs/file.c +++ b/fs/ubifs/file.c @@ -1257,7 +1257,8 @@ static int do_setattr(struct ubifs_info *c, struct inode *inode, return err; } -int ubifs_setattr(struct dentry *dentry, struct iattr *attr) +int ubifs_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *attr) { int err; struct inode *inode = d_inode(dentry); @@ -1265,7 +1266,7 @@ int ubifs_setattr(struct dentry *dentry, struct iattr *attr) dbg_gen("ino %lu, mode %#x, ia_valid %#x", inode->i_ino, inode->i_mode, attr->ia_valid); - err = setattr_prepare(dentry, attr); + err = setattr_prepare(&init_user_ns, dentry, attr); if (err) return err; diff --git a/fs/ubifs/ioctl.c b/fs/ubifs/ioctl.c index 4363d85a3fd4..2326d5122beb 100644 --- a/fs/ubifs/ioctl.c +++ b/fs/ubifs/ioctl.c @@ -155,7 +155,7 @@ long ubifs_ioctl(struct file *file, unsigned int cmd, unsigned long arg) if (IS_RDONLY(inode)) return -EROFS; - if (!inode_owner_or_capable(inode)) + if (!inode_owner_or_capable(&init_user_ns, inode)) return -EACCES; if (get_user(flags, (int __user *) arg)) diff --git a/fs/ubifs/journal.c b/fs/ubifs/journal.c index 03410ae0813a..2857e64d673d 100644 --- a/fs/ubifs/journal.c +++ b/fs/ubifs/journal.c @@ -881,7 +881,7 @@ int ubifs_jnl_write_inode(struct ubifs_info *c, const struct inode *inode) struct inode *xino; struct ubifs_dent_node *xent, *pxent = NULL; - if (ui->xattr_cnt >= ubifs_xattr_max_cnt(c)) { + if (ui->xattr_cnt > ubifs_xattr_max_cnt(c)) { ubifs_err(c, "Cannot delete inode, it has too much xattrs!"); goto out_release; } diff --git a/fs/ubifs/replay.c b/fs/ubifs/replay.c index 79801c9a5b87..0f8a6a16421b 100644 --- a/fs/ubifs/replay.c +++ b/fs/ubifs/replay.c @@ -559,7 +559,9 @@ static int is_last_bud(struct ubifs_info *c, struct ubifs_bud *bud) } /* authenticate_sleb_hash is split out for stack usage */ -static int authenticate_sleb_hash(struct ubifs_info *c, struct shash_desc *log_hash, u8 *hash) +static int noinline_for_stack +authenticate_sleb_hash(struct ubifs_info *c, + struct shash_desc *log_hash, u8 *hash) { SHASH_DESC_ON_STACK(hash_desc, c->hash_tfm); diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c index 138b9426c6c1..ddb2ca636c93 100644 --- a/fs/ubifs/super.c +++ b/fs/ubifs/super.c @@ -838,8 +838,10 @@ static int alloc_wbufs(struct ubifs_info *c) c->jheads[i].wbuf.jhead = i; c->jheads[i].grouped = 1; c->jheads[i].log_hash = ubifs_hash_get_desc(c); - if (IS_ERR(c->jheads[i].log_hash)) + if (IS_ERR(c->jheads[i].log_hash)) { + err = PTR_ERR(c->jheads[i].log_hash); goto out; + } } /* diff --git a/fs/ubifs/ubifs.h b/fs/ubifs/ubifs.h index fc2cdde3b549..7fdfdbda4b8a 100644 --- a/fs/ubifs/ubifs.h +++ b/fs/ubifs/ubifs.h @@ -1989,13 +1989,14 @@ int ubifs_calc_dark(const struct ubifs_info *c, int spc); /* file.c */ int ubifs_fsync(struct file *file, loff_t start, loff_t end, int datasync); -int ubifs_setattr(struct dentry *dentry, struct iattr *attr); +int ubifs_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *attr); int ubifs_update_time(struct inode *inode, struct timespec64 *time, int flags); /* dir.c */ struct inode *ubifs_new_inode(struct ubifs_info *c, struct inode *dir, umode_t mode); -int ubifs_getattr(const struct path *path, struct kstat *stat, +int ubifs_getattr(struct user_namespace *mnt_userns, const struct path *path, struct kstat *stat, u32 request_mask, unsigned int flags); int ubifs_check_dir_empty(struct inode *dir); diff --git a/fs/ubifs/xattr.c b/fs/ubifs/xattr.c index a0b9b349efe6..6b1e9830b274 100644 --- a/fs/ubifs/xattr.c +++ b/fs/ubifs/xattr.c @@ -498,7 +498,7 @@ int ubifs_purge_xattrs(struct inode *host) struct fscrypt_name nm = {0}; int err; - if (ubifs_inode(host)->xattr_cnt < ubifs_xattr_max_cnt(c)) + if (ubifs_inode(host)->xattr_cnt <= ubifs_xattr_max_cnt(c)) return 0; ubifs_warn(c, "inode %lu has too many xattrs, doing a non-atomic deletion", @@ -681,6 +681,7 @@ static int xattr_get(const struct xattr_handler *handler, } static int xattr_set(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *dentry, struct inode *inode, const char *name, const void *value, size_t size, int flags) diff --git a/fs/udf/file.c b/fs/udf/file.c index ad8eefad27d7..2846dcd92197 100644 --- a/fs/udf/file.c +++ b/fs/udf/file.c @@ -183,7 +183,7 @@ long udf_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) long old_block, new_block; int result; - if (inode_permission(inode, MAY_READ) != 0) { + if (file_permission(filp, MAY_READ) != 0) { udf_debug("no permission to access inode %lu\n", inode->i_ino); return -EPERM; } @@ -253,13 +253,14 @@ const struct file_operations udf_file_operations = { .llseek = generic_file_llseek, }; -static int udf_setattr(struct dentry *dentry, struct iattr *attr) +static int udf_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *attr) { struct inode *inode = d_inode(dentry); struct super_block *sb = inode->i_sb; int error; - error = setattr_prepare(dentry, attr); + error = setattr_prepare(&init_user_ns, dentry, attr); if (error) return error; @@ -282,7 +283,7 @@ static int udf_setattr(struct dentry *dentry, struct iattr *attr) if (attr->ia_valid & ATTR_MODE) udf_update_extra_perms(inode, attr->ia_mode); - setattr_copy(inode, attr); + setattr_copy(&init_user_ns, inode, attr); mark_inode_dirty(inode); return 0; } diff --git a/fs/udf/ialloc.c b/fs/udf/ialloc.c index 84ed23edebfd..2ecf0e87660e 100644 --- a/fs/udf/ialloc.c +++ b/fs/udf/ialloc.c @@ -103,7 +103,7 @@ struct inode *udf_new_inode(struct inode *dir, umode_t mode) mutex_unlock(&sbi->s_alloc_mutex); } - inode_init_owner(inode, dir, mode); + inode_init_owner(&init_user_ns, inode, dir, mode); if (UDF_QUERY_FLAG(sb, UDF_FLAG_UID_SET)) inode->i_uid = sbi->s_uid; if (UDF_QUERY_FLAG(sb, UDF_FLAG_GID_SET)) diff --git a/fs/udf/inode.c b/fs/udf/inode.c index bb89c3e43212..0dd2f93ac048 100644 --- a/fs/udf/inode.c +++ b/fs/udf/inode.c @@ -544,11 +544,14 @@ static int udf_do_extend_file(struct inode *inode, udf_write_aext(inode, last_pos, &last_ext->extLocation, last_ext->extLength, 1); + /* - * We've rewritten the last extent but there may be empty - * indirect extent after it - enter it. + * We've rewritten the last extent. If we are going to add + * more extents, we may need to enter possible following + * empty indirect extent. */ - udf_next_aext(inode, last_pos, &tmploc, &tmplen, 0); + if (new_block_bytes || prealloc_len) + udf_next_aext(inode, last_pos, &tmploc, &tmplen, 0); } /* Managed to do everything necessary? */ diff --git a/fs/udf/namei.c b/fs/udf/namei.c index e169d8fe35b5..f146b3089f3d 100644 --- a/fs/udf/namei.c +++ b/fs/udf/namei.c @@ -604,8 +604,8 @@ static int udf_add_nondir(struct dentry *dentry, struct inode *inode) return 0; } -static int udf_create(struct inode *dir, struct dentry *dentry, umode_t mode, - bool excl) +static int udf_create(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, bool excl) { struct inode *inode = udf_new_inode(dir, mode); @@ -623,7 +623,8 @@ static int udf_create(struct inode *dir, struct dentry *dentry, umode_t mode, return udf_add_nondir(dentry, inode); } -static int udf_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode) +static int udf_tmpfile(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { struct inode *inode = udf_new_inode(dir, mode); @@ -642,8 +643,8 @@ static int udf_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode) return 0; } -static int udf_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, - dev_t rdev) +static int udf_mknod(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, dev_t rdev) { struct inode *inode; @@ -658,7 +659,8 @@ static int udf_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, return udf_add_nondir(dentry, inode); } -static int udf_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) +static int udf_mkdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { struct inode *inode; struct udf_fileident_bh fibh; @@ -877,8 +879,8 @@ out: return retval; } -static int udf_symlink(struct inode *dir, struct dentry *dentry, - const char *symname) +static int udf_symlink(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, const char *symname) { struct inode *inode = udf_new_inode(dir, S_IFLNK | 0777); struct pathComponent *pc; @@ -1065,9 +1067,9 @@ static int udf_link(struct dentry *old_dentry, struct inode *dir, /* Anybody can rename anything with this: the permission checks are left to the * higher-level routines. */ -static int udf_rename(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry, - unsigned int flags) +static int udf_rename(struct user_namespace *mnt_userns, struct inode *old_dir, + struct dentry *old_dentry, struct inode *new_dir, + struct dentry *new_dentry, unsigned int flags) { struct inode *old_inode = d_inode(old_dentry); struct inode *new_inode = d_inode(new_dentry); diff --git a/fs/udf/super.c b/fs/udf/super.c index d0df217f4712..2f83c1204e20 100644 --- a/fs/udf/super.c +++ b/fs/udf/super.c @@ -459,6 +459,7 @@ static int udf_parse_options(char *options, struct udf_options *uopt, { char *p; int option; + unsigned int uv; uopt->novrs = 0; uopt->session = 0xFFFFFFFF; @@ -508,17 +509,17 @@ static int udf_parse_options(char *options, struct udf_options *uopt, uopt->flags &= ~(1 << UDF_FLAG_USE_SHORT_AD); break; case Opt_gid: - if (match_int(args, &option)) + if (match_uint(args, &uv)) return 0; - uopt->gid = make_kgid(current_user_ns(), option); + uopt->gid = make_kgid(current_user_ns(), uv); if (!gid_valid(uopt->gid)) return 0; uopt->flags |= (1 << UDF_FLAG_GID_SET); break; case Opt_uid: - if (match_int(args, &option)) + if (match_uint(args, &uv)) return 0; - uopt->uid = make_kuid(current_user_ns(), option); + uopt->uid = make_kuid(current_user_ns(), uv); if (!uid_valid(uopt->uid)) return 0; uopt->flags |= (1 << UDF_FLAG_UID_SET); diff --git a/fs/udf/symlink.c b/fs/udf/symlink.c index c973db239604..9b223421a3c5 100644 --- a/fs/udf/symlink.c +++ b/fs/udf/symlink.c @@ -152,14 +152,15 @@ out_unmap: return err; } -static int udf_symlink_getattr(const struct path *path, struct kstat *stat, - u32 request_mask, unsigned int flags) +static int udf_symlink_getattr(struct user_namespace *mnt_userns, + const struct path *path, struct kstat *stat, + u32 request_mask, unsigned int flags) { struct dentry *dentry = path->dentry; struct inode *inode = d_backing_inode(dentry); struct page *page; - generic_fillattr(inode, stat); + generic_fillattr(&init_user_ns, inode, stat); page = read_mapping_page(inode->i_mapping, 0, NULL); if (IS_ERR(page)) return PTR_ERR(page); diff --git a/fs/ufs/ialloc.c b/fs/ufs/ialloc.c index 969fd60436d3..7e3e08c0166f 100644 --- a/fs/ufs/ialloc.c +++ b/fs/ufs/ialloc.c @@ -289,7 +289,7 @@ cg_found: ufs_mark_sb_dirty(sb); inode->i_ino = cg * uspi->s_ipg + bit; - inode_init_owner(inode, dir, mode); + inode_init_owner(&init_user_ns, inode, dir, mode); inode->i_blocks = 0; inode->i_generation = 0; inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode); diff --git a/fs/ufs/inode.c b/fs/ufs/inode.c index c843ec858cf7..debc282c1bb4 100644 --- a/fs/ufs/inode.c +++ b/fs/ufs/inode.c @@ -1211,13 +1211,14 @@ out: return err; } -int ufs_setattr(struct dentry *dentry, struct iattr *attr) +int ufs_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *attr) { struct inode *inode = d_inode(dentry); unsigned int ia_valid = attr->ia_valid; int error; - error = setattr_prepare(dentry, attr); + error = setattr_prepare(&init_user_ns, dentry, attr); if (error) return error; @@ -1227,7 +1228,7 @@ int ufs_setattr(struct dentry *dentry, struct iattr *attr) return error; } - setattr_copy(inode, attr); + setattr_copy(&init_user_ns, inode, attr); mark_inode_dirty(inode); return 0; } diff --git a/fs/ufs/namei.c b/fs/ufs/namei.c index 9ef40f100415..29d5a0e0c8f0 100644 --- a/fs/ufs/namei.c +++ b/fs/ufs/namei.c @@ -69,7 +69,8 @@ static struct dentry *ufs_lookup(struct inode * dir, struct dentry *dentry, unsi * If the create succeeds, we fill in the inode information * with d_instantiate(). */ -static int ufs_create (struct inode * dir, struct dentry * dentry, umode_t mode, +static int ufs_create (struct user_namespace * mnt_userns, + struct inode * dir, struct dentry * dentry, umode_t mode, bool excl) { struct inode *inode; @@ -85,7 +86,8 @@ static int ufs_create (struct inode * dir, struct dentry * dentry, umode_t mode, return ufs_add_nondir(dentry, inode); } -static int ufs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev_t rdev) +static int ufs_mknod(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode, dev_t rdev) { struct inode *inode; int err; @@ -104,8 +106,8 @@ static int ufs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode, dev return err; } -static int ufs_symlink (struct inode * dir, struct dentry * dentry, - const char * symname) +static int ufs_symlink (struct user_namespace * mnt_userns, struct inode * dir, + struct dentry * dentry, const char * symname) { struct super_block * sb = dir->i_sb; int err; @@ -164,7 +166,8 @@ static int ufs_link (struct dentry * old_dentry, struct inode * dir, return error; } -static int ufs_mkdir(struct inode * dir, struct dentry * dentry, umode_t mode) +static int ufs_mkdir(struct user_namespace * mnt_userns, struct inode * dir, + struct dentry * dentry, umode_t mode) { struct inode * inode; int err; @@ -240,9 +243,9 @@ static int ufs_rmdir (struct inode * dir, struct dentry *dentry) return err; } -static int ufs_rename(struct inode *old_dir, struct dentry *old_dentry, - struct inode *new_dir, struct dentry *new_dentry, - unsigned int flags) +static int ufs_rename(struct user_namespace *mnt_userns, struct inode *old_dir, + struct dentry *old_dentry, struct inode *new_dir, + struct dentry *new_dentry, unsigned int flags) { struct inode *old_inode = d_inode(old_dentry); struct inode *new_inode = d_inode(new_dentry); diff --git a/fs/ufs/ufs.h b/fs/ufs/ufs.h index b49e0efdf3d7..550f7c5a3636 100644 --- a/fs/ufs/ufs.h +++ b/fs/ufs/ufs.h @@ -123,7 +123,8 @@ extern struct inode *ufs_iget(struct super_block *, unsigned long); extern int ufs_write_inode (struct inode *, struct writeback_control *); extern int ufs_sync_inode (struct inode *); extern void ufs_evict_inode (struct inode *); -extern int ufs_setattr(struct dentry *dentry, struct iattr *attr); +extern int ufs_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *attr); /* namei.c */ extern const struct file_operations ufs_dir_operations; diff --git a/fs/userfaultfd.c b/fs/userfaultfd.c index 894cc28142e7..0be8cdd4425a 100644 --- a/fs/userfaultfd.c +++ b/fs/userfaultfd.c @@ -979,14 +979,14 @@ static __poll_t userfaultfd_poll(struct file *file, poll_table *wait) static const struct file_operations userfaultfd_fops; -static int resolve_userfault_fork(struct userfaultfd_ctx *ctx, - struct userfaultfd_ctx *new, +static int resolve_userfault_fork(struct userfaultfd_ctx *new, + struct inode *inode, struct uffd_msg *msg) { int fd; - fd = anon_inode_getfd("[userfaultfd]", &userfaultfd_fops, new, - O_RDWR | (new->flags & UFFD_SHARED_FCNTL_FLAGS)); + fd = anon_inode_getfd_secure("[userfaultfd]", &userfaultfd_fops, new, + O_RDWR | (new->flags & UFFD_SHARED_FCNTL_FLAGS), inode); if (fd < 0) return fd; @@ -996,7 +996,7 @@ static int resolve_userfault_fork(struct userfaultfd_ctx *ctx, } static ssize_t userfaultfd_ctx_read(struct userfaultfd_ctx *ctx, int no_wait, - struct uffd_msg *msg) + struct uffd_msg *msg, struct inode *inode) { ssize_t ret; DECLARE_WAITQUEUE(wait, current); @@ -1107,7 +1107,7 @@ static ssize_t userfaultfd_ctx_read(struct userfaultfd_ctx *ctx, int no_wait, spin_unlock_irq(&ctx->fd_wqh.lock); if (!ret && msg->event == UFFD_EVENT_FORK) { - ret = resolve_userfault_fork(ctx, fork_nctx, msg); + ret = resolve_userfault_fork(fork_nctx, inode, msg); spin_lock_irq(&ctx->event_wqh.lock); if (!list_empty(&fork_event)) { /* @@ -1167,6 +1167,7 @@ static ssize_t userfaultfd_read(struct file *file, char __user *buf, ssize_t _ret, ret = 0; struct uffd_msg msg; int no_wait = file->f_flags & O_NONBLOCK; + struct inode *inode = file_inode(file); if (ctx->state == UFFD_STATE_WAIT_API) return -EINVAL; @@ -1174,7 +1175,7 @@ static ssize_t userfaultfd_read(struct file *file, char __user *buf, for (;;) { if (count < sizeof(msg)) return ret ? ret : -EINVAL; - _ret = userfaultfd_ctx_read(ctx, no_wait, &msg); + _ret = userfaultfd_ctx_read(ctx, no_wait, &msg, inode); if (_ret < 0) return ret ? ret : _ret; if (copy_to_user((__u64 __user *) buf, &msg, sizeof(msg))) @@ -1999,8 +2000,8 @@ SYSCALL_DEFINE1(userfaultfd, int, flags) /* prevent the mm struct to be freed */ mmgrab(ctx->mm); - fd = anon_inode_getfd("[userfaultfd]", &userfaultfd_fops, ctx, - O_RDWR | (flags & UFFD_SHARED_FCNTL_FLAGS)); + fd = anon_inode_getfd_secure("[userfaultfd]", &userfaultfd_fops, ctx, + O_RDWR | (flags & UFFD_SHARED_FCNTL_FLAGS), NULL); if (fd < 0) { mmdrop(ctx->mm); kmem_cache_free(userfaultfd_ctx_cachep, ctx); diff --git a/fs/utimes.c b/fs/utimes.c index fd3cc4226224..39f356017635 100644 --- a/fs/utimes.c +++ b/fs/utimes.c @@ -62,7 +62,8 @@ int vfs_utimes(const struct path *path, struct timespec64 *times) } retry_deleg: inode_lock(inode); - error = notify_change(path->dentry, &newattrs, &delegated_inode); + error = notify_change(mnt_user_ns(path->mnt), path->dentry, &newattrs, + &delegated_inode); inode_unlock(inode); if (delegated_inode) { error = break_deleg_wait(&delegated_inode); diff --git a/fs/vboxsf/dir.c b/fs/vboxsf/dir.c index 4d569f14a8d8..7aee0ec63ade 100644 --- a/fs/vboxsf/dir.c +++ b/fs/vboxsf/dir.c @@ -288,13 +288,15 @@ static int vboxsf_dir_create(struct inode *parent, struct dentry *dentry, return 0; } -static int vboxsf_dir_mkfile(struct inode *parent, struct dentry *dentry, +static int vboxsf_dir_mkfile(struct user_namespace *mnt_userns, + struct inode *parent, struct dentry *dentry, umode_t mode, bool excl) { return vboxsf_dir_create(parent, dentry, mode, 0); } -static int vboxsf_dir_mkdir(struct inode *parent, struct dentry *dentry, +static int vboxsf_dir_mkdir(struct user_namespace *mnt_userns, + struct inode *parent, struct dentry *dentry, umode_t mode) { return vboxsf_dir_create(parent, dentry, mode, 1); @@ -332,7 +334,8 @@ static int vboxsf_dir_unlink(struct inode *parent, struct dentry *dentry) return 0; } -static int vboxsf_dir_rename(struct inode *old_parent, +static int vboxsf_dir_rename(struct user_namespace *mnt_userns, + struct inode *old_parent, struct dentry *old_dentry, struct inode *new_parent, struct dentry *new_dentry, @@ -374,7 +377,8 @@ err_put_old_path: return err; } -static int vboxsf_dir_symlink(struct inode *parent, struct dentry *dentry, +static int vboxsf_dir_symlink(struct user_namespace *mnt_userns, + struct inode *parent, struct dentry *dentry, const char *symname) { struct vboxsf_inode *sf_parent_i = VBOXSF_I(parent); diff --git a/fs/vboxsf/utils.c b/fs/vboxsf/utils.c index 018057546067..3b847e3fba24 100644 --- a/fs/vboxsf/utils.c +++ b/fs/vboxsf/utils.c @@ -212,8 +212,8 @@ int vboxsf_inode_revalidate(struct dentry *dentry) return 0; } -int vboxsf_getattr(const struct path *path, struct kstat *kstat, - u32 request_mask, unsigned int flags) +int vboxsf_getattr(struct user_namespace *mnt_userns, const struct path *path, + struct kstat *kstat, u32 request_mask, unsigned int flags) { int err; struct dentry *dentry = path->dentry; @@ -233,11 +233,12 @@ int vboxsf_getattr(const struct path *path, struct kstat *kstat, if (err) return err; - generic_fillattr(d_inode(dentry), kstat); + generic_fillattr(&init_user_ns, d_inode(dentry), kstat); return 0; } -int vboxsf_setattr(struct dentry *dentry, struct iattr *iattr) +int vboxsf_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *iattr) { struct vboxsf_inode *sf_i = VBOXSF_I(d_inode(dentry)); struct vboxsf_sbi *sbi = VBOXSF_SBI(dentry->d_sb); diff --git a/fs/vboxsf/vfsmod.h b/fs/vboxsf/vfsmod.h index 18f95b00fc33..760524e78c88 100644 --- a/fs/vboxsf/vfsmod.h +++ b/fs/vboxsf/vfsmod.h @@ -90,9 +90,11 @@ int vboxsf_stat(struct vboxsf_sbi *sbi, struct shfl_string *path, struct shfl_fsobjinfo *info); int vboxsf_stat_dentry(struct dentry *dentry, struct shfl_fsobjinfo *info); int vboxsf_inode_revalidate(struct dentry *dentry); -int vboxsf_getattr(const struct path *path, struct kstat *kstat, - u32 request_mask, unsigned int query_flags); -int vboxsf_setattr(struct dentry *dentry, struct iattr *iattr); +int vboxsf_getattr(struct user_namespace *mnt_userns, const struct path *path, + struct kstat *kstat, u32 request_mask, + unsigned int query_flags); +int vboxsf_setattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct iattr *iattr); struct shfl_string *vboxsf_path_from_dentry(struct vboxsf_sbi *sbi, struct dentry *dentry); int vboxsf_nlscpy(struct vboxsf_sbi *sbi, char *name, size_t name_bound_len, diff --git a/fs/verity/Makefile b/fs/verity/Makefile index 570e9136334d..435559a4fa9e 100644 --- a/fs/verity/Makefile +++ b/fs/verity/Makefile @@ -5,6 +5,7 @@ obj-$(CONFIG_FS_VERITY) += enable.o \ init.o \ measure.o \ open.o \ + read_metadata.o \ verify.o obj-$(CONFIG_FS_VERITY_BUILTIN_SIGNATURES) += signature.o diff --git a/fs/verity/enable.c b/fs/verity/enable.c index f7e997a01ad0..77e159a0346b 100644 --- a/fs/verity/enable.c +++ b/fs/verity/enable.c @@ -369,7 +369,7 @@ int fsverity_ioctl_enable(struct file *filp, const void __user *uarg) * has verity enabled, and to stabilize the data being hashed. */ - err = inode_permission(inode, MAY_WRITE); + err = file_permission(filp, MAY_WRITE); if (err) return err; diff --git a/fs/verity/fsverity_private.h b/fs/verity/fsverity_private.h index 6413d28664d6..a7920434bae5 100644 --- a/fs/verity/fsverity_private.h +++ b/fs/verity/fsverity_private.h @@ -122,12 +122,17 @@ int fsverity_init_merkle_tree_params(struct merkle_tree_params *params, const u8 *salt, size_t salt_size); struct fsverity_info *fsverity_create_info(const struct inode *inode, - void *desc, size_t desc_size); + struct fsverity_descriptor *desc, + size_t desc_size); void fsverity_set_info(struct inode *inode, struct fsverity_info *vi); void fsverity_free_info(struct fsverity_info *vi); +int fsverity_get_descriptor(struct inode *inode, + struct fsverity_descriptor **desc_ret, + size_t *desc_size_ret); + int __init fsverity_init_info_cache(void); void __init fsverity_exit_info_cache(void); @@ -135,15 +140,13 @@ void __init fsverity_exit_info_cache(void); #ifdef CONFIG_FS_VERITY_BUILTIN_SIGNATURES int fsverity_verify_signature(const struct fsverity_info *vi, - const struct fsverity_descriptor *desc, - size_t desc_size); + const u8 *signature, size_t sig_size); int __init fsverity_init_signature(void); #else /* !CONFIG_FS_VERITY_BUILTIN_SIGNATURES */ static inline int fsverity_verify_signature(const struct fsverity_info *vi, - const struct fsverity_descriptor *desc, - size_t desc_size) + const u8 *signature, size_t sig_size) { return 0; } diff --git a/fs/verity/open.c b/fs/verity/open.c index 228d0eca3e2e..60ff8af7219f 100644 --- a/fs/verity/open.c +++ b/fs/verity/open.c @@ -142,45 +142,17 @@ static int compute_file_digest(struct fsverity_hash_alg *hash_alg, } /* - * Validate the given fsverity_descriptor and create a new fsverity_info from - * it. The signature (if present) is also checked. + * Create a new fsverity_info from the given fsverity_descriptor (with optional + * appended signature), and check the signature if present. The + * fsverity_descriptor must have already undergone basic validation. */ struct fsverity_info *fsverity_create_info(const struct inode *inode, - void *_desc, size_t desc_size) + struct fsverity_descriptor *desc, + size_t desc_size) { - struct fsverity_descriptor *desc = _desc; struct fsverity_info *vi; int err; - if (desc_size < sizeof(*desc)) { - fsverity_err(inode, "Unrecognized descriptor size: %zu bytes", - desc_size); - return ERR_PTR(-EINVAL); - } - - if (desc->version != 1) { - fsverity_err(inode, "Unrecognized descriptor version: %u", - desc->version); - return ERR_PTR(-EINVAL); - } - - if (memchr_inv(desc->__reserved, 0, sizeof(desc->__reserved))) { - fsverity_err(inode, "Reserved bits set in descriptor"); - return ERR_PTR(-EINVAL); - } - - if (desc->salt_size > sizeof(desc->salt)) { - fsverity_err(inode, "Invalid salt_size: %u", desc->salt_size); - return ERR_PTR(-EINVAL); - } - - if (le64_to_cpu(desc->data_size) != inode->i_size) { - fsverity_err(inode, - "Wrong data_size: %llu (desc) != %lld (inode)", - le64_to_cpu(desc->data_size), inode->i_size); - return ERR_PTR(-EINVAL); - } - vi = kmem_cache_zalloc(fsverity_info_cachep, GFP_KERNEL); if (!vi) return ERR_PTR(-ENOMEM); @@ -209,7 +181,8 @@ struct fsverity_info *fsverity_create_info(const struct inode *inode, vi->tree_params.hash_alg->name, vi->tree_params.digest_size, vi->file_digest); - err = fsverity_verify_signature(vi, desc, desc_size); + err = fsverity_verify_signature(vi, desc->signature, + le32_to_cpu(desc->sig_size)); out: if (err) { fsverity_free_info(vi); @@ -245,15 +218,57 @@ void fsverity_free_info(struct fsverity_info *vi) kmem_cache_free(fsverity_info_cachep, vi); } -/* Ensure the inode has an ->i_verity_info */ -static int ensure_verity_info(struct inode *inode) +static bool validate_fsverity_descriptor(struct inode *inode, + const struct fsverity_descriptor *desc, + size_t desc_size) { - struct fsverity_info *vi = fsverity_get_info(inode); - struct fsverity_descriptor *desc; - int res; + if (desc_size < sizeof(*desc)) { + fsverity_err(inode, "Unrecognized descriptor size: %zu bytes", + desc_size); + return false; + } - if (vi) - return 0; + if (desc->version != 1) { + fsverity_err(inode, "Unrecognized descriptor version: %u", + desc->version); + return false; + } + + if (memchr_inv(desc->__reserved, 0, sizeof(desc->__reserved))) { + fsverity_err(inode, "Reserved bits set in descriptor"); + return false; + } + + if (desc->salt_size > sizeof(desc->salt)) { + fsverity_err(inode, "Invalid salt_size: %u", desc->salt_size); + return false; + } + + if (le64_to_cpu(desc->data_size) != inode->i_size) { + fsverity_err(inode, + "Wrong data_size: %llu (desc) != %lld (inode)", + le64_to_cpu(desc->data_size), inode->i_size); + return false; + } + + if (le32_to_cpu(desc->sig_size) > desc_size - sizeof(*desc)) { + fsverity_err(inode, "Signature overflows verity descriptor"); + return false; + } + + return true; +} + +/* + * Read the inode's fsverity_descriptor (with optional appended signature) from + * the filesystem, and do basic validation of it. + */ +int fsverity_get_descriptor(struct inode *inode, + struct fsverity_descriptor **desc_ret, + size_t *desc_size_ret) +{ + int res; + struct fsverity_descriptor *desc; res = inode->i_sb->s_vop->get_verity_descriptor(inode, NULL, 0); if (res < 0) { @@ -272,20 +287,46 @@ static int ensure_verity_info(struct inode *inode) res = inode->i_sb->s_vop->get_verity_descriptor(inode, desc, res); if (res < 0) { fsverity_err(inode, "Error %d reading verity descriptor", res); - goto out_free_desc; + kfree(desc); + return res; + } + + if (!validate_fsverity_descriptor(inode, desc, res)) { + kfree(desc); + return -EINVAL; } - vi = fsverity_create_info(inode, desc, res); + *desc_ret = desc; + *desc_size_ret = res; + return 0; +} + +/* Ensure the inode has an ->i_verity_info */ +static int ensure_verity_info(struct inode *inode) +{ + struct fsverity_info *vi = fsverity_get_info(inode); + struct fsverity_descriptor *desc; + size_t desc_size; + int err; + + if (vi) + return 0; + + err = fsverity_get_descriptor(inode, &desc, &desc_size); + if (err) + return err; + + vi = fsverity_create_info(inode, desc, desc_size); if (IS_ERR(vi)) { - res = PTR_ERR(vi); + err = PTR_ERR(vi); goto out_free_desc; } fsverity_set_info(inode, vi); - res = 0; + err = 0; out_free_desc: kfree(desc); - return res; + return err; } /** diff --git a/fs/verity/read_metadata.c b/fs/verity/read_metadata.c new file mode 100644 index 000000000000..7e2d0c7bdf0d --- /dev/null +++ b/fs/verity/read_metadata.c @@ -0,0 +1,195 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Ioctl to read verity metadata + * + * Copyright 2021 Google LLC + */ + +#include "fsverity_private.h" + +#include <linux/backing-dev.h> +#include <linux/highmem.h> +#include <linux/sched/signal.h> +#include <linux/uaccess.h> + +static int fsverity_read_merkle_tree(struct inode *inode, + const struct fsverity_info *vi, + void __user *buf, u64 offset, int length) +{ + const struct fsverity_operations *vops = inode->i_sb->s_vop; + u64 end_offset; + unsigned int offs_in_page; + pgoff_t index, last_index; + int retval = 0; + int err = 0; + + end_offset = min(offset + length, vi->tree_params.tree_size); + if (offset >= end_offset) + return 0; + offs_in_page = offset_in_page(offset); + last_index = (end_offset - 1) >> PAGE_SHIFT; + + /* + * Iterate through each Merkle tree page in the requested range and copy + * the requested portion to userspace. Note that the Merkle tree block + * size isn't important here, as we are returning a byte stream; i.e., + * we can just work with pages even if the tree block size != PAGE_SIZE. + */ + for (index = offset >> PAGE_SHIFT; index <= last_index; index++) { + unsigned long num_ra_pages = + min_t(unsigned long, last_index - index + 1, + inode->i_sb->s_bdi->io_pages); + unsigned int bytes_to_copy = min_t(u64, end_offset - offset, + PAGE_SIZE - offs_in_page); + struct page *page; + const void *virt; + + page = vops->read_merkle_tree_page(inode, index, num_ra_pages); + if (IS_ERR(page)) { + err = PTR_ERR(page); + fsverity_err(inode, + "Error %d reading Merkle tree page %lu", + err, index); + break; + } + + virt = kmap(page); + if (copy_to_user(buf, virt + offs_in_page, bytes_to_copy)) { + kunmap(page); + put_page(page); + err = -EFAULT; + break; + } + kunmap(page); + put_page(page); + + retval += bytes_to_copy; + buf += bytes_to_copy; + offset += bytes_to_copy; + + if (fatal_signal_pending(current)) { + err = -EINTR; + break; + } + cond_resched(); + offs_in_page = 0; + } + return retval ? retval : err; +} + +/* Copy the requested portion of the buffer to userspace. */ +static int fsverity_read_buffer(void __user *dst, u64 offset, int length, + const void *src, size_t src_length) +{ + if (offset >= src_length) + return 0; + src += offset; + src_length -= offset; + + length = min_t(size_t, length, src_length); + + if (copy_to_user(dst, src, length)) + return -EFAULT; + + return length; +} + +static int fsverity_read_descriptor(struct inode *inode, + void __user *buf, u64 offset, int length) +{ + struct fsverity_descriptor *desc; + size_t desc_size; + int res; + + res = fsverity_get_descriptor(inode, &desc, &desc_size); + if (res) + return res; + + /* don't include the signature */ + desc_size = offsetof(struct fsverity_descriptor, signature); + desc->sig_size = 0; + + res = fsverity_read_buffer(buf, offset, length, desc, desc_size); + + kfree(desc); + return res; +} + +static int fsverity_read_signature(struct inode *inode, + void __user *buf, u64 offset, int length) +{ + struct fsverity_descriptor *desc; + size_t desc_size; + int res; + + res = fsverity_get_descriptor(inode, &desc, &desc_size); + if (res) + return res; + + if (desc->sig_size == 0) { + res = -ENODATA; + goto out; + } + + /* + * Include only the signature. Note that fsverity_get_descriptor() + * already verified that sig_size is in-bounds. + */ + res = fsverity_read_buffer(buf, offset, length, desc->signature, + le32_to_cpu(desc->sig_size)); +out: + kfree(desc); + return res; +} + +/** + * fsverity_ioctl_read_metadata() - read verity metadata from a file + * @filp: file to read the metadata from + * @uarg: user pointer to fsverity_read_metadata_arg + * + * Return: length read on success, 0 on EOF, -errno on failure + */ +int fsverity_ioctl_read_metadata(struct file *filp, const void __user *uarg) +{ + struct inode *inode = file_inode(filp); + const struct fsverity_info *vi; + struct fsverity_read_metadata_arg arg; + int length; + void __user *buf; + + vi = fsverity_get_info(inode); + if (!vi) + return -ENODATA; /* not a verity file */ + /* + * Note that we don't have to explicitly check that the file is open for + * reading, since verity files can only be opened for reading. + */ + + if (copy_from_user(&arg, uarg, sizeof(arg))) + return -EFAULT; + + if (arg.__reserved) + return -EINVAL; + + /* offset + length must not overflow. */ + if (arg.offset + arg.length < arg.offset) + return -EINVAL; + + /* Ensure that the return value will fit in INT_MAX. */ + length = min_t(u64, arg.length, INT_MAX); + + buf = u64_to_user_ptr(arg.buf_ptr); + + switch (arg.metadata_type) { + case FS_VERITY_METADATA_TYPE_MERKLE_TREE: + return fsverity_read_merkle_tree(inode, vi, buf, arg.offset, + length); + case FS_VERITY_METADATA_TYPE_DESCRIPTOR: + return fsverity_read_descriptor(inode, buf, arg.offset, length); + case FS_VERITY_METADATA_TYPE_SIGNATURE: + return fsverity_read_signature(inode, buf, arg.offset, length); + default: + return -EINVAL; + } +} +EXPORT_SYMBOL_GPL(fsverity_ioctl_read_metadata); diff --git a/fs/verity/signature.c b/fs/verity/signature.c index 012468eda2a7..143a530a8008 100644 --- a/fs/verity/signature.c +++ b/fs/verity/signature.c @@ -29,21 +29,19 @@ static struct key *fsverity_keyring; /** * fsverity_verify_signature() - check a verity file's signature * @vi: the file's fsverity_info - * @desc: the file's fsverity_descriptor - * @desc_size: size of @desc + * @signature: the file's built-in signature + * @sig_size: size of signature in bytes, or 0 if no signature * - * If the file's fs-verity descriptor includes a signature of the file digest, - * verify it against the certificates in the fs-verity keyring. + * If the file includes a signature of its fs-verity file digest, verify it + * against the certificates in the fs-verity keyring. * * Return: 0 on success (signature valid or not required); -errno on failure */ int fsverity_verify_signature(const struct fsverity_info *vi, - const struct fsverity_descriptor *desc, - size_t desc_size) + const u8 *signature, size_t sig_size) { const struct inode *inode = vi->inode; const struct fsverity_hash_alg *hash_alg = vi->tree_params.hash_alg; - const u32 sig_size = le32_to_cpu(desc->sig_size); struct fsverity_formatted_digest *d; int err; @@ -56,11 +54,6 @@ int fsverity_verify_signature(const struct fsverity_info *vi, return 0; } - if (sig_size > desc_size - sizeof(*desc)) { - fsverity_err(inode, "Signature overflows verity descriptor"); - return -EBADMSG; - } - d = kzalloc(sizeof(*d) + hash_alg->digest_size, GFP_KERNEL); if (!d) return -ENOMEM; @@ -70,8 +63,7 @@ int fsverity_verify_signature(const struct fsverity_info *vi, memcpy(d->digest, vi->file_digest, hash_alg->digest_size); err = verify_pkcs7_signature(d, sizeof(*d) + hash_alg->digest_size, - desc->signature, sig_size, - fsverity_keyring, + signature, sig_size, fsverity_keyring, VERIFYING_UNSPECIFIED_SIGNATURE, NULL, NULL); kfree(d); diff --git a/fs/xattr.c b/fs/xattr.c index fd57153b1f61..b3444e06cded 100644 --- a/fs/xattr.c +++ b/fs/xattr.c @@ -83,7 +83,8 @@ xattr_resolve_name(struct inode *inode, const char **name) * because different namespaces have very different rules. */ static int -xattr_permission(struct inode *inode, const char *name, int mask) +xattr_permission(struct user_namespace *mnt_userns, struct inode *inode, + const char *name, int mask) { /* * We can never set or remove an extended attribute on a read-only @@ -97,7 +98,7 @@ xattr_permission(struct inode *inode, const char *name, int mask) * to be writen back improperly if their true value is * unknown to the vfs. */ - if (HAS_UNMAPPED_ID(inode)) + if (HAS_UNMAPPED_ID(mnt_userns, inode)) return -EPERM; } @@ -127,11 +128,12 @@ xattr_permission(struct inode *inode, const char *name, int mask) if (!S_ISREG(inode->i_mode) && !S_ISDIR(inode->i_mode)) return (mask & MAY_WRITE) ? -EPERM : -ENODATA; if (S_ISDIR(inode->i_mode) && (inode->i_mode & S_ISVTX) && - (mask & MAY_WRITE) && !inode_owner_or_capable(inode)) + (mask & MAY_WRITE) && + !inode_owner_or_capable(mnt_userns, inode)) return -EPERM; } - return inode_permission(inode, mask); + return inode_permission(mnt_userns, inode, mask); } /* @@ -162,8 +164,9 @@ xattr_supported_namespace(struct inode *inode, const char *prefix) EXPORT_SYMBOL(xattr_supported_namespace); int -__vfs_setxattr(struct dentry *dentry, struct inode *inode, const char *name, - const void *value, size_t size, int flags) +__vfs_setxattr(struct user_namespace *mnt_userns, struct dentry *dentry, + struct inode *inode, const char *name, const void *value, + size_t size, int flags) { const struct xattr_handler *handler; @@ -174,7 +177,8 @@ __vfs_setxattr(struct dentry *dentry, struct inode *inode, const char *name, return -EOPNOTSUPP; if (size == 0) value = ""; /* empty EA, do not remove */ - return handler->set(handler, dentry, inode, name, value, size, flags); + return handler->set(handler, mnt_userns, dentry, inode, name, value, + size, flags); } EXPORT_SYMBOL(__vfs_setxattr); @@ -182,6 +186,7 @@ EXPORT_SYMBOL(__vfs_setxattr); * __vfs_setxattr_noperm - perform setxattr operation without performing * permission checks. * + * @mnt_userns - user namespace of the mount the inode was found from * @dentry - object to perform setxattr on * @name - xattr name to set * @value - value to set @name to @@ -194,8 +199,9 @@ EXPORT_SYMBOL(__vfs_setxattr); * is executed. It also assumes that the caller will make the appropriate * permission checks. */ -int __vfs_setxattr_noperm(struct dentry *dentry, const char *name, - const void *value, size_t size, int flags) +int __vfs_setxattr_noperm(struct user_namespace *mnt_userns, + struct dentry *dentry, const char *name, + const void *value, size_t size, int flags) { struct inode *inode = dentry->d_inode; int error = -EAGAIN; @@ -205,7 +211,8 @@ int __vfs_setxattr_noperm(struct dentry *dentry, const char *name, if (issec) inode->i_flags &= ~S_NOSEC; if (inode->i_opflags & IOP_XATTR) { - error = __vfs_setxattr(dentry, inode, name, value, size, flags); + error = __vfs_setxattr(mnt_userns, dentry, inode, name, value, + size, flags); if (!error) { fsnotify_xattr(dentry); security_inode_post_setxattr(dentry, name, value, @@ -244,18 +251,19 @@ int __vfs_setxattr_noperm(struct dentry *dentry, const char *name, * a delegation was broken on, NULL if none. */ int -__vfs_setxattr_locked(struct dentry *dentry, const char *name, - const void *value, size_t size, int flags, - struct inode **delegated_inode) +__vfs_setxattr_locked(struct user_namespace *mnt_userns, struct dentry *dentry, + const char *name, const void *value, size_t size, + int flags, struct inode **delegated_inode) { struct inode *inode = dentry->d_inode; int error; - error = xattr_permission(inode, name, MAY_WRITE); + error = xattr_permission(mnt_userns, inode, name, MAY_WRITE); if (error) return error; - error = security_inode_setxattr(dentry, name, value, size, flags); + error = security_inode_setxattr(mnt_userns, dentry, name, value, size, + flags); if (error) goto out; @@ -263,7 +271,8 @@ __vfs_setxattr_locked(struct dentry *dentry, const char *name, if (error) goto out; - error = __vfs_setxattr_noperm(dentry, name, value, size, flags); + error = __vfs_setxattr_noperm(mnt_userns, dentry, name, value, + size, flags); out: return error; @@ -271,8 +280,8 @@ out: EXPORT_SYMBOL_GPL(__vfs_setxattr_locked); int -vfs_setxattr(struct dentry *dentry, const char *name, const void *value, - size_t size, int flags) +vfs_setxattr(struct user_namespace *mnt_userns, struct dentry *dentry, + const char *name, const void *value, size_t size, int flags) { struct inode *inode = dentry->d_inode; struct inode *delegated_inode = NULL; @@ -280,7 +289,7 @@ vfs_setxattr(struct dentry *dentry, const char *name, const void *value, int error; if (size && strcmp(name, XATTR_NAME_CAPS) == 0) { - error = cap_convert_nscap(dentry, &value, size); + error = cap_convert_nscap(mnt_userns, dentry, &value, size); if (error < 0) return error; size = error; @@ -288,8 +297,8 @@ vfs_setxattr(struct dentry *dentry, const char *name, const void *value, retry_deleg: inode_lock(inode); - error = __vfs_setxattr_locked(dentry, name, value, size, flags, - &delegated_inode); + error = __vfs_setxattr_locked(mnt_userns, dentry, name, value, size, + flags, &delegated_inode); inode_unlock(inode); if (delegated_inode) { @@ -305,18 +314,20 @@ retry_deleg: EXPORT_SYMBOL_GPL(vfs_setxattr); static ssize_t -xattr_getsecurity(struct inode *inode, const char *name, void *value, - size_t size) +xattr_getsecurity(struct user_namespace *mnt_userns, struct inode *inode, + const char *name, void *value, size_t size) { void *buffer = NULL; ssize_t len; if (!value || !size) { - len = security_inode_getsecurity(inode, name, &buffer, false); + len = security_inode_getsecurity(mnt_userns, inode, name, + &buffer, false); goto out_noalloc; } - len = security_inode_getsecurity(inode, name, &buffer, true); + len = security_inode_getsecurity(mnt_userns, inode, name, &buffer, + true); if (len < 0) return len; if (size < len) { @@ -339,15 +350,16 @@ out_noalloc: * Returns the result of alloc, if failed, or the getxattr operation. */ ssize_t -vfs_getxattr_alloc(struct dentry *dentry, const char *name, char **xattr_value, - size_t xattr_size, gfp_t flags) +vfs_getxattr_alloc(struct user_namespace *mnt_userns, struct dentry *dentry, + const char *name, char **xattr_value, size_t xattr_size, + gfp_t flags) { const struct xattr_handler *handler; struct inode *inode = dentry->d_inode; char *value = *xattr_value; int error; - error = xattr_permission(inode, name, MAY_READ); + error = xattr_permission(mnt_userns, inode, name, MAY_READ); if (error) return error; @@ -388,12 +400,13 @@ __vfs_getxattr(struct dentry *dentry, struct inode *inode, const char *name, EXPORT_SYMBOL(__vfs_getxattr); ssize_t -vfs_getxattr(struct dentry *dentry, const char *name, void *value, size_t size) +vfs_getxattr(struct user_namespace *mnt_userns, struct dentry *dentry, + const char *name, void *value, size_t size) { struct inode *inode = dentry->d_inode; int error; - error = xattr_permission(inode, name, MAY_READ); + error = xattr_permission(mnt_userns, inode, name, MAY_READ); if (error) return error; @@ -404,7 +417,8 @@ vfs_getxattr(struct dentry *dentry, const char *name, void *value, size_t size) if (!strncmp(name, XATTR_SECURITY_PREFIX, XATTR_SECURITY_PREFIX_LEN)) { const char *suffix = name + XATTR_SECURITY_PREFIX_LEN; - int ret = xattr_getsecurity(inode, suffix, value, size); + int ret = xattr_getsecurity(mnt_userns, inode, suffix, value, + size); /* * Only overwrite the return value if a security module * is actually active. @@ -439,7 +453,8 @@ vfs_listxattr(struct dentry *dentry, char *list, size_t size) EXPORT_SYMBOL_GPL(vfs_listxattr); int -__vfs_removexattr(struct dentry *dentry, const char *name) +__vfs_removexattr(struct user_namespace *mnt_userns, struct dentry *dentry, + const char *name) { struct inode *inode = d_inode(dentry); const struct xattr_handler *handler; @@ -449,7 +464,8 @@ __vfs_removexattr(struct dentry *dentry, const char *name) return PTR_ERR(handler); if (!handler->set) return -EOPNOTSUPP; - return handler->set(handler, dentry, inode, name, NULL, 0, XATTR_REPLACE); + return handler->set(handler, mnt_userns, dentry, inode, name, NULL, 0, + XATTR_REPLACE); } EXPORT_SYMBOL(__vfs_removexattr); @@ -463,17 +479,18 @@ EXPORT_SYMBOL(__vfs_removexattr); * a delegation was broken on, NULL if none. */ int -__vfs_removexattr_locked(struct dentry *dentry, const char *name, - struct inode **delegated_inode) +__vfs_removexattr_locked(struct user_namespace *mnt_userns, + struct dentry *dentry, const char *name, + struct inode **delegated_inode) { struct inode *inode = dentry->d_inode; int error; - error = xattr_permission(inode, name, MAY_WRITE); + error = xattr_permission(mnt_userns, inode, name, MAY_WRITE); if (error) return error; - error = security_inode_removexattr(dentry, name); + error = security_inode_removexattr(mnt_userns, dentry, name); if (error) goto out; @@ -481,7 +498,7 @@ __vfs_removexattr_locked(struct dentry *dentry, const char *name, if (error) goto out; - error = __vfs_removexattr(dentry, name); + error = __vfs_removexattr(mnt_userns, dentry, name); if (!error) { fsnotify_xattr(dentry); @@ -494,7 +511,8 @@ out: EXPORT_SYMBOL_GPL(__vfs_removexattr_locked); int -vfs_removexattr(struct dentry *dentry, const char *name) +vfs_removexattr(struct user_namespace *mnt_userns, struct dentry *dentry, + const char *name) { struct inode *inode = dentry->d_inode; struct inode *delegated_inode = NULL; @@ -502,7 +520,8 @@ vfs_removexattr(struct dentry *dentry, const char *name) retry_deleg: inode_lock(inode); - error = __vfs_removexattr_locked(dentry, name, &delegated_inode); + error = __vfs_removexattr_locked(mnt_userns, dentry, + name, &delegated_inode); inode_unlock(inode); if (delegated_inode) { @@ -519,8 +538,9 @@ EXPORT_SYMBOL_GPL(vfs_removexattr); * Extended attribute SET operations */ static long -setxattr(struct dentry *d, const char __user *name, const void __user *value, - size_t size, int flags) +setxattr(struct user_namespace *mnt_userns, struct dentry *d, + const char __user *name, const void __user *value, size_t size, + int flags) { int error; void *kvalue = NULL; @@ -547,10 +567,10 @@ setxattr(struct dentry *d, const char __user *name, const void __user *value, } if ((strcmp(kname, XATTR_NAME_POSIX_ACL_ACCESS) == 0) || (strcmp(kname, XATTR_NAME_POSIX_ACL_DEFAULT) == 0)) - posix_acl_fix_xattr_from_user(kvalue, size); + posix_acl_fix_xattr_from_user(mnt_userns, kvalue, size); } - error = vfs_setxattr(d, kname, kvalue, size, flags); + error = vfs_setxattr(mnt_userns, d, kname, kvalue, size, flags); out: kvfree(kvalue); @@ -563,13 +583,15 @@ static int path_setxattr(const char __user *pathname, { struct path path; int error; + retry: error = user_path_at(AT_FDCWD, pathname, lookup_flags, &path); if (error) return error; error = mnt_want_write(path.mnt); if (!error) { - error = setxattr(path.dentry, name, value, size, flags); + error = setxattr(mnt_user_ns(path.mnt), path.dentry, name, + value, size, flags); mnt_drop_write(path.mnt); } path_put(&path); @@ -605,7 +627,9 @@ SYSCALL_DEFINE5(fsetxattr, int, fd, const char __user *, name, audit_file(f.file); error = mnt_want_write_file(f.file); if (!error) { - error = setxattr(f.file->f_path.dentry, name, value, size, flags); + error = setxattr(file_mnt_user_ns(f.file), + f.file->f_path.dentry, name, + value, size, flags); mnt_drop_write_file(f.file); } fdput(f); @@ -616,8 +640,8 @@ SYSCALL_DEFINE5(fsetxattr, int, fd, const char __user *, name, * Extended attribute GET operations */ static ssize_t -getxattr(struct dentry *d, const char __user *name, void __user *value, - size_t size) +getxattr(struct user_namespace *mnt_userns, struct dentry *d, + const char __user *name, void __user *value, size_t size) { ssize_t error; void *kvalue = NULL; @@ -637,11 +661,11 @@ getxattr(struct dentry *d, const char __user *name, void __user *value, return -ENOMEM; } - error = vfs_getxattr(d, kname, kvalue, size); + error = vfs_getxattr(mnt_userns, d, kname, kvalue, size); if (error > 0) { if ((strcmp(kname, XATTR_NAME_POSIX_ACL_ACCESS) == 0) || (strcmp(kname, XATTR_NAME_POSIX_ACL_DEFAULT) == 0)) - posix_acl_fix_xattr_to_user(kvalue, error); + posix_acl_fix_xattr_to_user(mnt_userns, kvalue, error); if (size && copy_to_user(value, kvalue, error)) error = -EFAULT; } else if (error == -ERANGE && size >= XATTR_SIZE_MAX) { @@ -665,7 +689,7 @@ retry: error = user_path_at(AT_FDCWD, pathname, lookup_flags, &path); if (error) return error; - error = getxattr(path.dentry, name, value, size); + error = getxattr(mnt_user_ns(path.mnt), path.dentry, name, value, size); path_put(&path); if (retry_estale(error, lookup_flags)) { lookup_flags |= LOOKUP_REVAL; @@ -695,7 +719,8 @@ SYSCALL_DEFINE4(fgetxattr, int, fd, const char __user *, name, if (!f.file) return error; audit_file(f.file); - error = getxattr(f.file->f_path.dentry, name, value, size); + error = getxattr(file_mnt_user_ns(f.file), f.file->f_path.dentry, + name, value, size); fdput(f); return error; } @@ -779,7 +804,8 @@ SYSCALL_DEFINE3(flistxattr, int, fd, char __user *, list, size_t, size) * Extended attribute REMOVE operations */ static long -removexattr(struct dentry *d, const char __user *name) +removexattr(struct user_namespace *mnt_userns, struct dentry *d, + const char __user *name) { int error; char kname[XATTR_NAME_MAX + 1]; @@ -790,7 +816,7 @@ removexattr(struct dentry *d, const char __user *name) if (error < 0) return error; - return vfs_removexattr(d, kname); + return vfs_removexattr(mnt_userns, d, kname); } static int path_removexattr(const char __user *pathname, @@ -804,7 +830,7 @@ retry: return error; error = mnt_want_write(path.mnt); if (!error) { - error = removexattr(path.dentry, name); + error = removexattr(mnt_user_ns(path.mnt), path.dentry, name); mnt_drop_write(path.mnt); } path_put(&path); @@ -837,7 +863,8 @@ SYSCALL_DEFINE2(fremovexattr, int, fd, const char __user *, name) audit_file(f.file); error = mnt_want_write_file(f.file); if (!error) { - error = removexattr(f.file->f_path.dentry, name); + error = removexattr(file_mnt_user_ns(f.file), + f.file->f_path.dentry, name); mnt_drop_write_file(f.file); } fdput(f); diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c index 7cb9f064ac64..0c623d3c1036 100644 --- a/fs/xfs/libxfs/xfs_alloc.c +++ b/fs/xfs/libxfs/xfs_alloc.c @@ -2474,6 +2474,47 @@ xfs_defer_agfl_block( xfs_defer_add(tp, XFS_DEFER_OPS_TYPE_AGFL_FREE, &new->xefi_list); } +#ifdef DEBUG +/* + * Check if an AGF has a free extent record whose length is equal to + * args->minlen. + */ +STATIC int +xfs_exact_minlen_extent_available( + struct xfs_alloc_arg *args, + struct xfs_buf *agbp, + int *stat) +{ + struct xfs_btree_cur *cnt_cur; + xfs_agblock_t fbno; + xfs_extlen_t flen; + int error = 0; + + cnt_cur = xfs_allocbt_init_cursor(args->mp, args->tp, agbp, + args->agno, XFS_BTNUM_CNT); + error = xfs_alloc_lookup_ge(cnt_cur, 0, args->minlen, stat); + if (error) + goto out; + + if (*stat == 0) { + error = -EFSCORRUPTED; + goto out; + } + + error = xfs_alloc_get_rec(cnt_cur, &fbno, &flen, stat); + if (error) + goto out; + + if (*stat == 1 && flen != args->minlen) + *stat = 0; + +out: + xfs_btree_del_cursor(cnt_cur, error); + + return error; +} +#endif + /* * Decide whether to use this allocation group for this allocation. * If so, fix up the btree freelist's size. @@ -2545,6 +2586,15 @@ xfs_alloc_fix_freelist( if (!xfs_alloc_space_available(args, need, flags)) goto out_agbp_relse; +#ifdef DEBUG + if (args->alloc_minlen_only) { + int stat; + + error = xfs_exact_minlen_extent_available(args, agbp, &stat); + if (error || !stat) + goto out_agbp_relse; + } +#endif /* * Make the freelist shorter if it's too long. * diff --git a/fs/xfs/libxfs/xfs_alloc.h b/fs/xfs/libxfs/xfs_alloc.h index 6c22b12176b8..a4427c5775c2 100644 --- a/fs/xfs/libxfs/xfs_alloc.h +++ b/fs/xfs/libxfs/xfs_alloc.h @@ -75,6 +75,9 @@ typedef struct xfs_alloc_arg { char wasfromfl; /* set if allocation is from freelist */ struct xfs_owner_info oinfo; /* owner of blocks being allocated */ enum xfs_ag_resv_type resv; /* block reservation to use */ +#ifdef DEBUG + bool alloc_minlen_only; /* allocate exact minlen extent */ +#endif } xfs_alloc_arg_t; /* diff --git a/fs/xfs/libxfs/xfs_attr.c b/fs/xfs/libxfs/xfs_attr.c index fd8e6418a0d3..472b3039eabb 100644 --- a/fs/xfs/libxfs/xfs_attr.c +++ b/fs/xfs/libxfs/xfs_attr.c @@ -396,6 +396,7 @@ xfs_attr_set( struct xfs_trans_res tres; bool rsvd = (args->attr_filter & XFS_ATTR_ROOT); int error, local; + int rmt_blks = 0; unsigned int total; if (XFS_FORCED_SHUTDOWN(dp->i_mount)) @@ -442,34 +443,33 @@ xfs_attr_set( tres.tr_logcount = XFS_ATTRSET_LOG_COUNT; tres.tr_logflags = XFS_TRANS_PERM_LOG_RES; total = args->total; + + if (!local) + rmt_blks = xfs_attr3_rmt_blocks(mp, args->valuelen); } else { XFS_STATS_INC(mp, xs_attr_remove); tres = M_RES(mp)->tr_attrrm; total = XFS_ATTRRM_SPACE_RES(mp); + rmt_blks = xfs_attr3_rmt_blocks(mp, XFS_XATTR_SIZE_MAX); } /* * Root fork attributes can use reserved data blocks for this * operation if necessary */ - error = xfs_trans_alloc(mp, &tres, total, 0, - rsvd ? XFS_TRANS_RESERVE : 0, &args->trans); + error = xfs_trans_alloc_inode(dp, &tres, total, 0, rsvd, &args->trans); if (error) return error; - xfs_ilock(dp, XFS_ILOCK_EXCL); - xfs_trans_ijoin(args->trans, dp, 0); - if (args->value) { - unsigned int quota_flags = XFS_QMOPT_RES_REGBLKS; - - if (rsvd) - quota_flags |= XFS_QMOPT_FORCE_RES; - error = xfs_trans_reserve_quota_nblks(args->trans, dp, - args->total, 0, quota_flags); + if (args->value || xfs_inode_hasattr(dp)) { + error = xfs_iext_count_may_overflow(dp, XFS_ATTR_FORK, + XFS_IEXT_ATTR_MANIP_CNT(rmt_blks)); if (error) goto out_trans_cancel; + } + if (args->value) { error = xfs_has_attr(args); if (error == -EEXIST && (args->attr_flags & XATTR_CREATE)) goto out_trans_cancel; diff --git a/fs/xfs/libxfs/xfs_bmap.c b/fs/xfs/libxfs/xfs_bmap.c index bc446418e227..e0905ad171f0 100644 --- a/fs/xfs/libxfs/xfs_bmap.c +++ b/fs/xfs/libxfs/xfs_bmap.c @@ -1079,21 +1079,13 @@ xfs_bmap_add_attrfork( blks = XFS_ADDAFORK_SPACE_RES(mp); - error = xfs_trans_alloc(mp, &M_RES(mp)->tr_addafork, blks, 0, - rsvd ? XFS_TRANS_RESERVE : 0, &tp); + error = xfs_trans_alloc_inode(ip, &M_RES(mp)->tr_addafork, blks, 0, + rsvd, &tp); if (error) return error; - - xfs_ilock(ip, XFS_ILOCK_EXCL); - error = xfs_trans_reserve_quota_nblks(tp, ip, blks, 0, rsvd ? - XFS_QMOPT_RES_REGBLKS | XFS_QMOPT_FORCE_RES : - XFS_QMOPT_RES_REGBLKS); - if (error) - goto trans_cancel; if (XFS_IFORK_Q(ip)) goto trans_cancel; - xfs_trans_ijoin(tp, ip, 0); xfs_trans_log_inode(tp, ip, XFS_ILOG_CORE); error = xfs_bmap_set_attrforkoff(ip, size, &version); if (error) @@ -3463,34 +3455,16 @@ xfs_bmap_btalloc_accounting( args->len); } -STATIC int -xfs_bmap_btalloc( - struct xfs_bmalloca *ap) /* bmap alloc argument struct */ +static int +xfs_bmap_compute_alignments( + struct xfs_bmalloca *ap, + struct xfs_alloc_arg *args) { - xfs_mount_t *mp; /* mount point structure */ - xfs_alloctype_t atype = 0; /* type for allocation routines */ - xfs_extlen_t align = 0; /* minimum allocation alignment */ - xfs_agnumber_t fb_agno; /* ag number of ap->firstblock */ - xfs_agnumber_t ag; - xfs_alloc_arg_t args; - xfs_fileoff_t orig_offset; - xfs_extlen_t orig_length; - xfs_extlen_t blen; - xfs_extlen_t nextminlen = 0; - int nullfb; /* true if ap->firstblock isn't set */ - int isaligned; - int tryagain; - int error; - int stripe_align; - - ASSERT(ap->length); - orig_offset = ap->offset; - orig_length = ap->length; - - mp = ap->ip->i_mount; + struct xfs_mount *mp = args->mp; + xfs_extlen_t align = 0; /* minimum allocation alignment */ + int stripe_align = 0; /* stripe alignment for allocation is determined by mount parameters */ - stripe_align = 0; if (mp->m_swidth && (mp->m_flags & XFS_MOUNT_SWALLOC)) stripe_align = mp->m_swidth; else if (mp->m_dalign) @@ -3501,13 +3475,171 @@ xfs_bmap_btalloc( else if (ap->datatype & XFS_ALLOC_USERDATA) align = xfs_get_extsz_hint(ap->ip); if (align) { - error = xfs_bmap_extsize_align(mp, &ap->got, &ap->prev, - align, 0, ap->eof, 0, ap->conv, - &ap->offset, &ap->length); - ASSERT(!error); + if (xfs_bmap_extsize_align(mp, &ap->got, &ap->prev, align, 0, + ap->eof, 0, ap->conv, &ap->offset, + &ap->length)) + ASSERT(0); ASSERT(ap->length); } + /* apply extent size hints if obtained earlier */ + if (align) { + args->prod = align; + div_u64_rem(ap->offset, args->prod, &args->mod); + if (args->mod) + args->mod = args->prod - args->mod; + } else if (mp->m_sb.sb_blocksize >= PAGE_SIZE) { + args->prod = 1; + args->mod = 0; + } else { + args->prod = PAGE_SIZE >> mp->m_sb.sb_blocklog; + div_u64_rem(ap->offset, args->prod, &args->mod); + if (args->mod) + args->mod = args->prod - args->mod; + } + + return stripe_align; +} + +static void +xfs_bmap_process_allocated_extent( + struct xfs_bmalloca *ap, + struct xfs_alloc_arg *args, + xfs_fileoff_t orig_offset, + xfs_extlen_t orig_length) +{ + int nullfb; + + nullfb = ap->tp->t_firstblock == NULLFSBLOCK; + + /* + * check the allocation happened at the same or higher AG than + * the first block that was allocated. + */ + ASSERT(nullfb || + XFS_FSB_TO_AGNO(args->mp, ap->tp->t_firstblock) <= + XFS_FSB_TO_AGNO(args->mp, args->fsbno)); + + ap->blkno = args->fsbno; + if (nullfb) + ap->tp->t_firstblock = args->fsbno; + ap->length = args->len; + /* + * If the extent size hint is active, we tried to round the + * caller's allocation request offset down to extsz and the + * length up to another extsz boundary. If we found a free + * extent we mapped it in starting at this new offset. If the + * newly mapped space isn't long enough to cover any of the + * range of offsets that was originally requested, move the + * mapping up so that we can fill as much of the caller's + * original request as possible. Free space is apparently + * very fragmented so we're unlikely to be able to satisfy the + * hints anyway. + */ + if (ap->length <= orig_length) + ap->offset = orig_offset; + else if (ap->offset + ap->length < orig_offset + orig_length) + ap->offset = orig_offset + orig_length - ap->length; + xfs_bmap_btalloc_accounting(ap, args); +} + +#ifdef DEBUG +static int +xfs_bmap_exact_minlen_extent_alloc( + struct xfs_bmalloca *ap) +{ + struct xfs_mount *mp = ap->ip->i_mount; + struct xfs_alloc_arg args = { .tp = ap->tp, .mp = mp }; + xfs_fileoff_t orig_offset; + xfs_extlen_t orig_length; + int error; + + ASSERT(ap->length); + + if (ap->minlen != 1) { + ap->blkno = NULLFSBLOCK; + ap->length = 0; + return 0; + } + + orig_offset = ap->offset; + orig_length = ap->length; + + args.alloc_minlen_only = 1; + + xfs_bmap_compute_alignments(ap, &args); + + if (ap->tp->t_firstblock == NULLFSBLOCK) { + /* + * Unlike the longest extent available in an AG, we don't track + * the length of an AG's shortest extent. + * XFS_ERRTAG_BMAP_ALLOC_MINLEN_EXTENT is a debug only knob and + * hence we can afford to start traversing from the 0th AG since + * we need not be concerned about a drop in performance in + * "debug only" code paths. + */ + ap->blkno = XFS_AGB_TO_FSB(mp, 0, 0); + } else { + ap->blkno = ap->tp->t_firstblock; + } + + args.fsbno = ap->blkno; + args.oinfo = XFS_RMAP_OINFO_SKIP_UPDATE; + args.type = XFS_ALLOCTYPE_FIRST_AG; + args.total = args.minlen = args.maxlen = ap->minlen; + + args.alignment = 1; + args.minalignslop = 0; + + args.minleft = ap->minleft; + args.wasdel = ap->wasdel; + args.resv = XFS_AG_RESV_NONE; + args.datatype = ap->datatype; + + error = xfs_alloc_vextent(&args); + if (error) + return error; + + if (args.fsbno != NULLFSBLOCK) { + xfs_bmap_process_allocated_extent(ap, &args, orig_offset, + orig_length); + } else { + ap->blkno = NULLFSBLOCK; + ap->length = 0; + } + + return 0; +} +#else + +#define xfs_bmap_exact_minlen_extent_alloc(bma) (-EFSCORRUPTED) + +#endif + +STATIC int +xfs_bmap_btalloc( + struct xfs_bmalloca *ap) +{ + struct xfs_mount *mp = ap->ip->i_mount; + struct xfs_alloc_arg args = { .tp = ap->tp, .mp = mp }; + xfs_alloctype_t atype = 0; + xfs_agnumber_t fb_agno; /* ag number of ap->firstblock */ + xfs_agnumber_t ag; + xfs_fileoff_t orig_offset; + xfs_extlen_t orig_length; + xfs_extlen_t blen; + xfs_extlen_t nextminlen = 0; + int nullfb; /* true if ap->firstblock isn't set */ + int isaligned; + int tryagain; + int error; + int stripe_align; + + ASSERT(ap->length); + orig_offset = ap->offset; + orig_length = ap->length; + + stripe_align = xfs_bmap_compute_alignments(ap, &args); nullfb = ap->tp->t_firstblock == NULLFSBLOCK; fb_agno = nullfb ? NULLAGNUMBER : XFS_FSB_TO_AGNO(mp, @@ -3538,9 +3670,6 @@ xfs_bmap_btalloc( * Normal allocation, done through xfs_alloc_vextent. */ tryagain = isaligned = 0; - memset(&args, 0, sizeof(args)); - args.tp = ap->tp; - args.mp = mp; args.fsbno = ap->blkno; args.oinfo = XFS_RMAP_OINFO_SKIP_UPDATE; @@ -3571,21 +3700,7 @@ xfs_bmap_btalloc( args.total = ap->total; args.minlen = ap->minlen; } - /* apply extent size hints if obtained earlier */ - if (align) { - args.prod = align; - div_u64_rem(ap->offset, args.prod, &args.mod); - if (args.mod) - args.mod = args.prod - args.mod; - } else if (mp->m_sb.sb_blocksize >= PAGE_SIZE) { - args.prod = 1; - args.mod = 0; - } else { - args.prod = PAGE_SIZE >> mp->m_sb.sb_blocklog; - div_u64_rem(ap->offset, args.prod, &args.mod); - if (args.mod) - args.mod = args.prod - args.mod; - } + /* * If we are not low on available data blocks, and the underlying * logical volume manager is a stripe, and the file offset is zero then @@ -3687,37 +3802,10 @@ xfs_bmap_btalloc( return error; ap->tp->t_flags |= XFS_TRANS_LOWMODE; } + if (args.fsbno != NULLFSBLOCK) { - /* - * check the allocation happened at the same or higher AG than - * the first block that was allocated. - */ - ASSERT(ap->tp->t_firstblock == NULLFSBLOCK || - XFS_FSB_TO_AGNO(mp, ap->tp->t_firstblock) <= - XFS_FSB_TO_AGNO(mp, args.fsbno)); - - ap->blkno = args.fsbno; - if (ap->tp->t_firstblock == NULLFSBLOCK) - ap->tp->t_firstblock = args.fsbno; - ASSERT(nullfb || fb_agno <= args.agno); - ap->length = args.len; - /* - * If the extent size hint is active, we tried to round the - * caller's allocation request offset down to extsz and the - * length up to another extsz boundary. If we found a free - * extent we mapped it in starting at this new offset. If the - * newly mapped space isn't long enough to cover any of the - * range of offsets that was originally requested, move the - * mapping up so that we can fill as much of the caller's - * original request as possible. Free space is apparently - * very fragmented so we're unlikely to be able to satisfy the - * hints anyway. - */ - if (ap->length <= orig_length) - ap->offset = orig_offset; - else if (ap->offset + ap->length < orig_offset + orig_length) - ap->offset = orig_offset + orig_length - ap->length; - xfs_bmap_btalloc_accounting(ap, &args); + xfs_bmap_process_allocated_extent(ap, &args, orig_offset, + orig_length); } else { ap->blkno = NULLFSBLOCK; ap->length = 0; @@ -4001,8 +4089,7 @@ xfs_bmapi_reserve_delalloc( * blocks. This number gets adjusted later. We return if we haven't * allocated blocks already inside this loop. */ - error = xfs_trans_reserve_quota_nblks(NULL, ip, (long)alen, 0, - XFS_QMOPT_RES_REGBLKS); + error = xfs_quota_reserve_blkres(ip, alen); if (error) return error; @@ -4048,8 +4135,7 @@ out_unreserve_blocks: xfs_mod_fdblocks(mp, alen, false); out_unreserve_quota: if (XFS_IS_QUOTA_ON(mp)) - xfs_trans_unreserve_quota_nblks(NULL, ip, (long)alen, 0, - XFS_QMOPT_RES_REGBLKS); + xfs_quota_unreserve_blkres(ip, alen); return error; } @@ -4083,6 +4169,10 @@ xfs_bmap_alloc_userdata( return xfs_bmap_rtalloc(bma); } + if (unlikely(XFS_TEST_ERROR(false, mp, + XFS_ERRTAG_BMAP_ALLOC_MINLEN_EXTENT))) + return xfs_bmap_exact_minlen_extent_alloc(bma); + return xfs_bmap_btalloc(bma); } @@ -4119,10 +4209,15 @@ xfs_bmapi_allocate( else bma->minlen = 1; - if (bma->flags & XFS_BMAPI_METADATA) - error = xfs_bmap_btalloc(bma); - else + if (bma->flags & XFS_BMAPI_METADATA) { + if (unlikely(XFS_TEST_ERROR(false, mp, + XFS_ERRTAG_BMAP_ALLOC_MINLEN_EXTENT))) + error = xfs_bmap_exact_minlen_extent_alloc(bma); + else + error = xfs_bmap_btalloc(bma); + } else { error = xfs_bmap_alloc_userdata(bma); + } if (error || bma->blkno == NULLFSBLOCK) return error; @@ -4527,6 +4622,12 @@ xfs_bmapi_convert_delalloc( return error; xfs_ilock(ip, XFS_ILOCK_EXCL); + + error = xfs_iext_count_may_overflow(ip, whichfork, + XFS_IEXT_ADD_NOSPLIT_CNT); + if (error) + goto out_trans_cancel; + xfs_trans_ijoin(tp, ip, 0); if (!xfs_iext_lookup_extent(ip, ifp, offset_fsb, &bma.icur, &bma.got) || @@ -4826,9 +4927,8 @@ xfs_bmap_del_extent_delay( * sb counters as we might have to borrow some blocks for the * indirect block accounting. */ - error = xfs_trans_reserve_quota_nblks(NULL, ip, - -((long)del->br_blockcount), 0, - isrt ? XFS_QMOPT_RES_RTBLKS : XFS_QMOPT_RES_REGBLKS); + ASSERT(!isrt); + error = xfs_quota_unreserve_blkres(ip, del->br_blockcount); if (error) return error; ip->i_delayed_blks -= del->br_blockcount; @@ -5145,6 +5245,27 @@ xfs_bmap_del_extent_real( /* * Deleting the middle of the extent. */ + + /* + * For directories, -ENOSPC is returned since a directory entry + * remove operation must not fail due to low extent count + * availability. -ENOSPC will be handled by higher layers of XFS + * by letting the corresponding empty Data/Free blocks to linger + * until a future remove operation. Dabtree blocks would be + * swapped with the last block in the leaf space and then the + * new last block will be unmapped. + * + * The above logic also applies to the source directory entry of + * a rename operation. + */ + error = xfs_iext_count_may_overflow(ip, whichfork, 1); + if (error) { + ASSERT(S_ISDIR(VFS_I(ip)->i_mode) && + whichfork == XFS_DATA_FORK); + error = -ENOSPC; + goto done; + } + old = got; got.br_blockcount = del->br_startoff - got.br_startoff; diff --git a/fs/xfs/libxfs/xfs_btree.c b/fs/xfs/libxfs/xfs_btree.c index c4d7a9241dc3..5b6fcb9b44e2 100644 --- a/fs/xfs/libxfs/xfs_btree.c +++ b/fs/xfs/libxfs/xfs_btree.c @@ -353,20 +353,17 @@ xfs_btree_free_block( */ void xfs_btree_del_cursor( - xfs_btree_cur_t *cur, /* btree cursor */ - int error) /* del because of error */ + struct xfs_btree_cur *cur, /* btree cursor */ + int error) /* del because of error */ { - int i; /* btree level */ + int i; /* btree level */ /* - * Clear the buffer pointers, and release the buffers. - * If we're doing this in the face of an error, we - * need to make sure to inspect all of the entries - * in the bc_bufs array for buffers to be unlocked. - * This is because some of the btree code works from - * level n down to 0, and if we get an error along - * the way we won't have initialized all the entries - * down to 0. + * Clear the buffer pointers and release the buffers. If we're doing + * this because of an error, inspect all of the entries in the bc_bufs + * array for buffers to be unlocked. This is because some of the btree + * code works from level n down to 0, and if we get an error along the + * way we won't have initialized all the entries down to 0. */ for (i = 0; i < cur->bc_nlevels; i++) { if (cur->bc_bufs[i]) @@ -374,17 +371,11 @@ xfs_btree_del_cursor( else if (!error) break; } - /* - * Can't free a bmap cursor without having dealt with the - * allocated indirect blocks' accounting. - */ - ASSERT(cur->bc_btnum != XFS_BTNUM_BMAP || - cur->bc_ino.allocated == 0); - /* - * Free the cursor. - */ + + ASSERT(cur->bc_btnum != XFS_BTNUM_BMAP || cur->bc_ino.allocated == 0 || + XFS_FORCED_SHUTDOWN(cur->bc_mp)); if (unlikely(cur->bc_flags & XFS_BTREE_STAGING)) - kmem_free((void *)cur->bc_ops); + kmem_free(cur->bc_ops); kmem_cache_free(xfs_btree_cur_zone, cur); } @@ -2814,7 +2805,7 @@ xfs_btree_split_worker( struct xfs_btree_split_args *args = container_of(work, struct xfs_btree_split_args, work); unsigned long pflags; - unsigned long new_pflags = PF_MEMALLOC_NOFS; + unsigned long new_pflags = 0; /* * we are in a transaction context here, but may also be doing work @@ -2826,12 +2817,20 @@ xfs_btree_split_worker( new_pflags |= PF_MEMALLOC | PF_SWAPWRITE | PF_KSWAPD; current_set_flags_nested(&pflags, new_pflags); + xfs_trans_set_context(args->cur->bc_tp); args->result = __xfs_btree_split(args->cur, args->level, args->ptrp, args->key, args->curp, args->stat); - complete(args->done); + xfs_trans_clear_context(args->cur->bc_tp); current_restore_flags_nested(&pflags, new_pflags); + + /* + * Do not access args after complete() has run here. We don't own args + * and the owner may run and free args before we return here. + */ + complete(args->done); + } /* diff --git a/fs/xfs/libxfs/xfs_dir2.h b/fs/xfs/libxfs/xfs_dir2.h index e55378640b05..d03e6098ded9 100644 --- a/fs/xfs/libxfs/xfs_dir2.h +++ b/fs/xfs/libxfs/xfs_dir2.h @@ -47,8 +47,6 @@ extern int xfs_dir_lookup(struct xfs_trans *tp, struct xfs_inode *dp, extern int xfs_dir_removename(struct xfs_trans *tp, struct xfs_inode *dp, struct xfs_name *name, xfs_ino_t ino, xfs_extlen_t tot); -extern bool xfs_dir2_sf_replace_needblock(struct xfs_inode *dp, - xfs_ino_t inum); extern int xfs_dir_replace(struct xfs_trans *tp, struct xfs_inode *dp, struct xfs_name *name, xfs_ino_t inum, xfs_extlen_t tot); diff --git a/fs/xfs/libxfs/xfs_dir2_sf.c b/fs/xfs/libxfs/xfs_dir2_sf.c index 2463b5d73447..8c4f76bba88b 100644 --- a/fs/xfs/libxfs/xfs_dir2_sf.c +++ b/fs/xfs/libxfs/xfs_dir2_sf.c @@ -1018,7 +1018,7 @@ xfs_dir2_sf_removename( /* * Check whether the sf dir replace operation need more blocks. */ -bool +static bool xfs_dir2_sf_replace_needblock( struct xfs_inode *dp, xfs_ino_t inum) diff --git a/fs/xfs/libxfs/xfs_errortag.h b/fs/xfs/libxfs/xfs_errortag.h index 53b305dea381..6ca9084b6934 100644 --- a/fs/xfs/libxfs/xfs_errortag.h +++ b/fs/xfs/libxfs/xfs_errortag.h @@ -56,7 +56,9 @@ #define XFS_ERRTAG_FORCE_SUMMARY_RECALC 33 #define XFS_ERRTAG_IUNLINK_FALLBACK 34 #define XFS_ERRTAG_BUF_IOERROR 35 -#define XFS_ERRTAG_MAX 36 +#define XFS_ERRTAG_REDUCE_MAX_IEXTENTS 36 +#define XFS_ERRTAG_BMAP_ALLOC_MINLEN_EXTENT 37 +#define XFS_ERRTAG_MAX 38 /* * Random factors for above tags, 1 means always, 2 means 1/2 time, etc. @@ -97,5 +99,7 @@ #define XFS_RANDOM_FORCE_SUMMARY_RECALC 1 #define XFS_RANDOM_IUNLINK_FALLBACK (XFS_RANDOM_DEFAULT/10) #define XFS_RANDOM_BUF_IOERROR XFS_RANDOM_DEFAULT +#define XFS_RANDOM_REDUCE_MAX_IEXTENTS 1 +#define XFS_RANDOM_BMAP_ALLOC_MINLEN_EXTENT 1 #endif /* __XFS_ERRORTAG_H_ */ diff --git a/fs/xfs/libxfs/xfs_fs.h b/fs/xfs/libxfs/xfs_fs.h index 2a2e3cfd94f0..6fad140d4c8e 100644 --- a/fs/xfs/libxfs/xfs_fs.h +++ b/fs/xfs/libxfs/xfs_fs.h @@ -250,6 +250,7 @@ typedef struct xfs_fsop_resblks { #define XFS_FSOP_GEOM_FLAGS_RMAPBT (1 << 19) /* reverse mapping btree */ #define XFS_FSOP_GEOM_FLAGS_REFLINK (1 << 20) /* files can share blocks */ #define XFS_FSOP_GEOM_FLAGS_BIGTIME (1 << 21) /* 64-bit nsec timestamps */ +#define XFS_FSOP_GEOM_FLAGS_INOBTCNT (1 << 22) /* inobt btree counter */ /* * Minimum and maximum sizes need for growth checks. diff --git a/fs/xfs/libxfs/xfs_inode_fork.c b/fs/xfs/libxfs/xfs_inode_fork.c index 7575de5cecb1..e080d7e07643 100644 --- a/fs/xfs/libxfs/xfs_inode_fork.c +++ b/fs/xfs/libxfs/xfs_inode_fork.c @@ -23,6 +23,8 @@ #include "xfs_da_btree.h" #include "xfs_dir2_priv.h" #include "xfs_attr_leaf.h" +#include "xfs_types.h" +#include "xfs_errortag.h" kmem_zone_t *xfs_ifork_zone; @@ -728,3 +730,28 @@ xfs_ifork_verify_local_attr( return 0; } + +int +xfs_iext_count_may_overflow( + struct xfs_inode *ip, + int whichfork, + int nr_to_add) +{ + struct xfs_ifork *ifp = XFS_IFORK_PTR(ip, whichfork); + uint64_t max_exts; + uint64_t nr_exts; + + if (whichfork == XFS_COW_FORK) + return 0; + + max_exts = (whichfork == XFS_ATTR_FORK) ? MAXAEXTNUM : MAXEXTNUM; + + if (XFS_TEST_ERROR(false, ip->i_mount, XFS_ERRTAG_REDUCE_MAX_IEXTENTS)) + max_exts = 10; + + nr_exts = ifp->if_nextents + nr_to_add; + if (nr_exts < ifp->if_nextents || nr_exts > max_exts) + return -EFBIG; + + return 0; +} diff --git a/fs/xfs/libxfs/xfs_inode_fork.h b/fs/xfs/libxfs/xfs_inode_fork.h index a4953e95c4f3..9e2137cd7372 100644 --- a/fs/xfs/libxfs/xfs_inode_fork.h +++ b/fs/xfs/libxfs/xfs_inode_fork.h @@ -35,6 +35,67 @@ struct xfs_ifork { #define XFS_IFBROOT 0x04 /* i_broot points to the bmap b-tree root */ /* + * Worst-case increase in the fork extent count when we're adding a single + * extent to a fork and there's no possibility of splitting an existing mapping. + */ +#define XFS_IEXT_ADD_NOSPLIT_CNT (1) + +/* + * Punching out an extent from the middle of an existing extent can cause the + * extent count to increase by 1. + * i.e. | Old extent | Hole | Old extent | + */ +#define XFS_IEXT_PUNCH_HOLE_CNT (1) + +/* + * Directory entry addition can cause the following, + * 1. Data block can be added/removed. + * A new extent can cause extent count to increase by 1. + * 2. Free disk block can be added/removed. + * Same behaviour as described above for Data block. + * 3. Dabtree blocks. + * XFS_DA_NODE_MAXDEPTH blocks can be added. Each of these can be new + * extents. Hence extent count can increase by XFS_DA_NODE_MAXDEPTH. + */ +#define XFS_IEXT_DIR_MANIP_CNT(mp) \ + ((XFS_DA_NODE_MAXDEPTH + 1 + 1) * (mp)->m_dir_geo->fsbcount) + +/* + * Adding/removing an xattr can cause XFS_DA_NODE_MAXDEPTH extents to + * be added. One extra extent for dabtree in case a local attr is + * large enough to cause a double split. It can also cause extent + * count to increase proportional to the size of a remote xattr's + * value. + */ +#define XFS_IEXT_ATTR_MANIP_CNT(rmt_blks) \ + (XFS_DA_NODE_MAXDEPTH + max(1, rmt_blks)) + +/* + * A write to a sub-interval of an existing unwritten extent causes the original + * extent to be split into 3 extents + * i.e. | Unwritten | Real | Unwritten | + * Hence extent count can increase by 2. + */ +#define XFS_IEXT_WRITE_UNWRITTEN_CNT (2) + + +/* + * Moving an extent to data fork can cause a sub-interval of an existing extent + * to be unmapped. This will increase extent count by 1. Mapping in the new + * extent can increase the extent count by 1 again i.e. + * | Old extent | New extent | Old extent | + * Hence number of extents increases by 2. + */ +#define XFS_IEXT_REFLINK_END_COW_CNT (2) + +/* + * Removing an initial range of source/donor file's extent and adding a new + * extent (from donor/source file) in its place will cause extent count to + * increase by 1. + */ +#define XFS_IEXT_SWAP_RMAP_CNT (1) + +/* * Fork handling. */ @@ -172,5 +233,7 @@ extern void xfs_ifork_init_cow(struct xfs_inode *ip); int xfs_ifork_verify_local_data(struct xfs_inode *ip); int xfs_ifork_verify_local_attr(struct xfs_inode *ip); +int xfs_iext_count_may_overflow(struct xfs_inode *ip, int whichfork, + int nr_to_add); #endif /* __XFS_INODE_FORK_H__ */ diff --git a/fs/xfs/libxfs/xfs_sb.c b/fs/xfs/libxfs/xfs_sb.c index bbda117e5d85..60e6d255e5e2 100644 --- a/fs/xfs/libxfs/xfs_sb.c +++ b/fs/xfs/libxfs/xfs_sb.c @@ -1138,6 +1138,8 @@ xfs_fs_geometry( geo->flags |= XFS_FSOP_GEOM_FLAGS_REFLINK; if (xfs_sb_version_hasbigtime(sbp)) geo->flags |= XFS_FSOP_GEOM_FLAGS_BIGTIME; + if (xfs_sb_version_hasinobtcounts(sbp)) + geo->flags |= XFS_FSOP_GEOM_FLAGS_INOBTCNT; if (xfs_sb_version_hassector(sbp)) geo->logsectsize = sbp->sb_logsectsize; else diff --git a/fs/xfs/scrub/common.c b/fs/xfs/scrub/common.c index 8ea6d4aa3f55..53456f3de881 100644 --- a/fs/xfs/scrub/common.c +++ b/fs/xfs/scrub/common.c @@ -888,7 +888,7 @@ xchk_stop_reaping( struct xfs_scrub *sc) { sc->flags |= XCHK_REAPING_DISABLED; - xfs_stop_block_reaping(sc->mp); + xfs_blockgc_stop(sc->mp); } /* Restart background reaping of resources. */ @@ -896,6 +896,6 @@ void xchk_start_reaping( struct xfs_scrub *sc) { - xfs_start_block_reaping(sc->mp); + xfs_blockgc_start(sc->mp); sc->flags &= ~XCHK_REAPING_DISABLED; } diff --git a/fs/xfs/xfs_acl.c b/fs/xfs/xfs_acl.c index 779cb73b3d00..d02bef24b32b 100644 --- a/fs/xfs/xfs_acl.c +++ b/fs/xfs/xfs_acl.c @@ -238,7 +238,8 @@ xfs_acl_set_mode( } int -xfs_set_acl(struct inode *inode, struct posix_acl *acl, int type) +xfs_set_acl(struct user_namespace *mnt_userns, struct inode *inode, + struct posix_acl *acl, int type) { umode_t mode; bool set_mode = false; @@ -252,7 +253,7 @@ xfs_set_acl(struct inode *inode, struct posix_acl *acl, int type) return error; if (type == ACL_TYPE_ACCESS) { - error = posix_acl_update_mode(inode, &mode, &acl); + error = posix_acl_update_mode(mnt_userns, inode, &mode, &acl); if (error) return error; set_mode = true; diff --git a/fs/xfs/xfs_acl.h b/fs/xfs/xfs_acl.h index c042c0868016..7bdb3a4ed798 100644 --- a/fs/xfs/xfs_acl.h +++ b/fs/xfs/xfs_acl.h @@ -11,7 +11,8 @@ struct posix_acl; #ifdef CONFIG_XFS_POSIX_ACL extern struct posix_acl *xfs_get_acl(struct inode *inode, int type); -extern int xfs_set_acl(struct inode *inode, struct posix_acl *acl, int type); +extern int xfs_set_acl(struct user_namespace *mnt_userns, struct inode *inode, + struct posix_acl *acl, int type); extern int __xfs_set_acl(struct inode *inode, struct posix_acl *acl, int type); void xfs_forget_acl(struct inode *inode, const char *name); #else diff --git a/fs/xfs/xfs_aops.c b/fs/xfs/xfs_aops.c index 4304c6416fbb..b4186d666157 100644 --- a/fs/xfs/xfs_aops.c +++ b/fs/xfs/xfs_aops.c @@ -62,7 +62,7 @@ xfs_setfilesize_trans_alloc( * We hand off the transaction to the completion thread now, so * clear the flag here. */ - current_restore_flags_nested(&tp->t_pflags, PF_MEMALLOC_NOFS); + xfs_trans_clear_context(tp); return 0; } @@ -125,7 +125,7 @@ xfs_setfilesize_ioend( * thus we need to mark ourselves as being in a transaction manually. * Similarly for freeze protection. */ - current_set_flags_nested(&tp->t_pflags, PF_MEMALLOC_NOFS); + xfs_trans_set_context(tp); __sb_writers_acquired(VFS_I(ip)->i_sb, SB_FREEZE_FS); /* we abort the update if there was an IO error */ @@ -568,6 +568,12 @@ xfs_vm_writepage( { struct xfs_writepage_ctx wpc = { }; + if (WARN_ON_ONCE(current->journal_info)) { + redirty_page_for_writepage(wbc, page); + unlock_page(page); + return 0; + } + return iomap_writepage(page, wbc, &wpc.ctx, &xfs_writeback_ops); } @@ -578,6 +584,13 @@ xfs_vm_writepages( { struct xfs_writepage_ctx wpc = { }; + /* + * Writing back data in a transaction context can result in recursive + * transactions. This is bad, so issue a warning and get out of here. + */ + if (WARN_ON_ONCE(current->journal_info)) + return 0; + xfs_iflags_clear(XFS_I(mapping->host), XFS_ITRUNCATED); return iomap_writepages(mapping, wbc, &wpc.ctx, &xfs_writeback_ops); } diff --git a/fs/xfs/xfs_bio_io.c b/fs/xfs/xfs_bio_io.c index e2148f2d5d6b..17f36db2f792 100644 --- a/fs/xfs/xfs_bio_io.c +++ b/fs/xfs/xfs_bio_io.c @@ -6,7 +6,7 @@ static inline unsigned int bio_max_vecs(unsigned int count) { - return min_t(unsigned, howmany(count, PAGE_SIZE), BIO_MAX_PAGES); + return bio_max_segs(howmany(count, PAGE_SIZE)); } int diff --git a/fs/xfs/xfs_bmap_item.c b/fs/xfs/xfs_bmap_item.c index 93e4d8ae6e92..2344757ede63 100644 --- a/fs/xfs/xfs_bmap_item.c +++ b/fs/xfs/xfs_bmap_item.c @@ -471,6 +471,7 @@ xfs_bui_item_recover( xfs_exntst_t state; unsigned int bui_type; int whichfork; + int iext_delta; int error = 0; if (!xfs_bui_validate(mp, buip)) { @@ -508,6 +509,15 @@ xfs_bui_item_recover( xfs_ilock(ip, XFS_ILOCK_EXCL); xfs_trans_ijoin(tp, ip, 0); + if (bui_type == XFS_BMAP_MAP) + iext_delta = XFS_IEXT_ADD_NOSPLIT_CNT; + else + iext_delta = XFS_IEXT_PUNCH_HOLE_CNT; + + error = xfs_iext_count_may_overflow(ip, whichfork, iext_delta); + if (error) + goto err_cancel; + count = bmap->me_len; error = xfs_trans_log_finish_bmap_update(tp, budp, bui_type, ip, whichfork, bmap->me_startoff, bmap->me_startblock, diff --git a/fs/xfs/xfs_bmap_util.c b/fs/xfs/xfs_bmap_util.c index 7371a7f7c652..e7d68318e6a5 100644 --- a/fs/xfs/xfs_bmap_util.c +++ b/fs/xfs/xfs_bmap_util.c @@ -727,11 +727,9 @@ xfs_alloc_file_space( xfs_fileoff_t startoffset_fsb; xfs_fileoff_t endoffset_fsb; int nimaps; - int quota_flag; int rt; xfs_trans_t *tp; xfs_bmbt_irec_t imaps[1], *imapp; - uint qblocks, resblks, resrtextents; int error; trace_xfs_alloc_file_space(ip); @@ -761,6 +759,7 @@ xfs_alloc_file_space( */ while (allocatesize_fsb && !error) { xfs_fileoff_t s, e; + unsigned int dblocks, rblocks, resblks; /* * Determine space reservations for data/realtime. @@ -790,45 +789,31 @@ xfs_alloc_file_space( */ resblks = min_t(xfs_fileoff_t, (e - s), (MAXEXTLEN * nimaps)); if (unlikely(rt)) { - resrtextents = qblocks = resblks; - resrtextents /= mp->m_sb.sb_rextsize; - resblks = XFS_DIOSTRAT_SPACE_RES(mp, 0); - quota_flag = XFS_QMOPT_RES_RTBLKS; + dblocks = XFS_DIOSTRAT_SPACE_RES(mp, 0); + rblocks = resblks; } else { - resrtextents = 0; - resblks = qblocks = XFS_DIOSTRAT_SPACE_RES(mp, resblks); - quota_flag = XFS_QMOPT_RES_REGBLKS; + dblocks = XFS_DIOSTRAT_SPACE_RES(mp, resblks); + rblocks = 0; } /* * Allocate and setup the transaction. */ - error = xfs_trans_alloc(mp, &M_RES(mp)->tr_write, resblks, - resrtextents, 0, &tp); - - /* - * Check for running out of space - */ - if (error) { - /* - * Free the transaction structure. - */ - ASSERT(error == -ENOSPC || XFS_FORCED_SHUTDOWN(mp)); - break; - } - xfs_ilock(ip, XFS_ILOCK_EXCL); - error = xfs_trans_reserve_quota_nblks(tp, ip, qblocks, - 0, quota_flag); + error = xfs_trans_alloc_inode(ip, &M_RES(mp)->tr_write, + dblocks, rblocks, false, &tp); if (error) - goto error1; + break; - xfs_trans_ijoin(tp, ip, 0); + error = xfs_iext_count_may_overflow(ip, XFS_DATA_FORK, + XFS_IEXT_ADD_NOSPLIT_CNT); + if (error) + goto error; error = xfs_bmapi_write(tp, ip, startoffset_fsb, allocatesize_fsb, alloc_type, 0, imapp, &nimaps); if (error) - goto error0; + goto error; /* * Complete the transaction @@ -851,10 +836,7 @@ xfs_alloc_file_space( return error; -error0: /* unlock inode, unreserve quota blocks, cancel trans */ - xfs_trans_unreserve_quota_nblks(tp, ip, (long)qblocks, 0, quota_flag); - -error1: /* Just cancel transaction */ +error: xfs_trans_cancel(tp); xfs_iunlock(ip, XFS_ILOCK_EXCL); return error; @@ -872,20 +854,16 @@ xfs_unmap_extent( uint resblks = XFS_DIOSTRAT_SPACE_RES(mp, 0); int error; - error = xfs_trans_alloc(mp, &M_RES(mp)->tr_write, resblks, 0, 0, &tp); - if (error) { - ASSERT(error == -ENOSPC || XFS_FORCED_SHUTDOWN(mp)); + error = xfs_trans_alloc_inode(ip, &M_RES(mp)->tr_write, resblks, 0, + false, &tp); + if (error) return error; - } - xfs_ilock(ip, XFS_ILOCK_EXCL); - error = xfs_trans_reserve_quota(tp, mp, ip->i_udquot, ip->i_gdquot, - ip->i_pdquot, resblks, 0, XFS_QMOPT_RES_REGBLKS); + error = xfs_iext_count_may_overflow(ip, XFS_DATA_FORK, + XFS_IEXT_PUNCH_HOLE_CNT); if (error) goto out_trans_cancel; - xfs_trans_ijoin(tp, ip, 0); - error = xfs_bunmapi(tp, ip, startoffset_fsb, len_fsb, 0, 2, done); if (error) goto out_trans_cancel; @@ -1163,6 +1141,11 @@ xfs_insert_file_space( xfs_ilock(ip, XFS_ILOCK_EXCL); xfs_trans_ijoin(tp, ip, 0); + error = xfs_iext_count_may_overflow(ip, XFS_DATA_FORK, + XFS_IEXT_PUNCH_HOLE_CNT); + if (error) + goto out_trans_cancel; + /* * The extent shifting code works on extent granularity. So, if stop_fsb * is not the starting block of extent, we need to split the extent at @@ -1384,6 +1367,22 @@ xfs_swap_extent_rmap( irec.br_blockcount); trace_xfs_swap_extent_rmap_remap_piece(tip, &uirec); + if (xfs_bmap_is_real_extent(&uirec)) { + error = xfs_iext_count_may_overflow(ip, + XFS_DATA_FORK, + XFS_IEXT_SWAP_RMAP_CNT); + if (error) + goto out; + } + + if (xfs_bmap_is_real_extent(&irec)) { + error = xfs_iext_count_may_overflow(tip, + XFS_DATA_FORK, + XFS_IEXT_SWAP_RMAP_CNT); + if (error) + goto out; + } + /* Remove the mapping from the donor file. */ xfs_bmap_unmap_extent(tp, tip, &uirec); diff --git a/fs/xfs/xfs_buf.c b/fs/xfs/xfs_buf.c index f8400bbd6473..37a1d12762d8 100644 --- a/fs/xfs/xfs_buf.c +++ b/fs/xfs/xfs_buf.c @@ -43,7 +43,7 @@ static kmem_zone_t *xfs_buf_zone; * pag_buf_lock * lru_lock * - * xfs_buftarg_wait_rele + * xfs_buftarg_drain_rele * lru_lock * b_lock (trylock due to inversion) * @@ -88,7 +88,7 @@ xfs_buf_vmap_len( * because the corresponding decrement is deferred to buffer release. Buffers * can undergo I/O multiple times in a hold-release cycle and per buffer I/O * tracking adds unnecessary overhead. This is used for sychronization purposes - * with unmount (see xfs_wait_buftarg()), so all we really need is a count of + * with unmount (see xfs_buftarg_drain()), so all we really need is a count of * in-flight buffers. * * Buffers that are never released (e.g., superblock, iclog buffers) must set @@ -1480,7 +1480,7 @@ xfs_buf_ioapply_map( int op) { int page_index; - int total_nr_pages = bp->b_page_count; + unsigned int total_nr_pages = bp->b_page_count; int nr_pages; struct bio *bio; sector_t sector = bp->b_maps[map].bm_bn; @@ -1505,7 +1505,7 @@ xfs_buf_ioapply_map( next_chunk: atomic_inc(&bp->b_io_remaining); - nr_pages = min(total_nr_pages, BIO_MAX_PAGES); + nr_pages = bio_max_segs(total_nr_pages); bio = bio_alloc(GFP_NOIO, nr_pages); bio_set_dev(bio, bp->b_target->bt_bdev); @@ -1786,7 +1786,7 @@ __xfs_buf_mark_corrupt( * while freeing all the buffers only held by the LRU. */ static enum lru_status -xfs_buftarg_wait_rele( +xfs_buftarg_drain_rele( struct list_head *item, struct list_lru_one *lru, spinlock_t *lru_lock, @@ -1798,7 +1798,7 @@ xfs_buftarg_wait_rele( if (atomic_read(&bp->b_hold) > 1) { /* need to wait, so skip it this pass */ - trace_xfs_buf_wait_buftarg(bp, _RET_IP_); + trace_xfs_buf_drain_buftarg(bp, _RET_IP_); return LRU_SKIP; } if (!spin_trylock(&bp->b_lock)) @@ -1815,14 +1815,13 @@ xfs_buftarg_wait_rele( return LRU_REMOVED; } +/* + * Wait for outstanding I/O on the buftarg to complete. + */ void -xfs_wait_buftarg( +xfs_buftarg_wait( struct xfs_buftarg *btp) { - LIST_HEAD(dispose); - int loop = 0; - bool write_fail = false; - /* * First wait on the buftarg I/O count for all in-flight buffers to be * released. This is critical as new buffers do not make the LRU until @@ -1838,10 +1837,21 @@ xfs_wait_buftarg( while (percpu_counter_sum(&btp->bt_io_count)) delay(100); flush_workqueue(btp->bt_mount->m_buf_workqueue); +} + +void +xfs_buftarg_drain( + struct xfs_buftarg *btp) +{ + LIST_HEAD(dispose); + int loop = 0; + bool write_fail = false; + + xfs_buftarg_wait(btp); /* loop until there is nothing left on the lru list. */ while (list_lru_count(&btp->bt_lru)) { - list_lru_walk(&btp->bt_lru, xfs_buftarg_wait_rele, + list_lru_walk(&btp->bt_lru, xfs_buftarg_drain_rele, &dispose, LONG_MAX); while (!list_empty(&dispose)) { diff --git a/fs/xfs/xfs_buf.h b/fs/xfs/xfs_buf.h index 5d91a31298a4..459ca34f26f5 100644 --- a/fs/xfs/xfs_buf.h +++ b/fs/xfs/xfs_buf.h @@ -152,7 +152,7 @@ struct xfs_buf { struct list_head b_list; struct xfs_perag *b_pag; /* contains rbtree root */ struct xfs_mount *b_mount; - xfs_buftarg_t *b_target; /* buffer target (device) */ + struct xfs_buftarg *b_target; /* buffer target (device) */ void *b_addr; /* virtual address of buffer */ struct work_struct b_ioend_work; struct completion b_iowait; /* queue for I/O waiters */ @@ -344,11 +344,12 @@ xfs_buf_update_cksum(struct xfs_buf *bp, unsigned long cksum_offset) /* * Handling of buftargs. */ -extern xfs_buftarg_t *xfs_alloc_buftarg(struct xfs_mount *, - struct block_device *, struct dax_device *); +extern struct xfs_buftarg *xfs_alloc_buftarg(struct xfs_mount *, + struct block_device *, struct dax_device *); extern void xfs_free_buftarg(struct xfs_buftarg *); -extern void xfs_wait_buftarg(xfs_buftarg_t *); -extern int xfs_setsize_buftarg(xfs_buftarg_t *, unsigned int); +extern void xfs_buftarg_wait(struct xfs_buftarg *); +extern void xfs_buftarg_drain(struct xfs_buftarg *); +extern int xfs_setsize_buftarg(struct xfs_buftarg *, unsigned int); #define xfs_getsize_buftarg(buftarg) block_size((buftarg)->bt_bdev) #define xfs_readonly_buftarg(buftarg) bdev_read_only((buftarg)->bt_bdev) diff --git a/fs/xfs/xfs_dquot.c b/fs/xfs/xfs_dquot.c index 1d95ed387d66..bd8379b98374 100644 --- a/fs/xfs/xfs_dquot.c +++ b/fs/xfs/xfs_dquot.c @@ -314,8 +314,14 @@ xfs_dquot_disk_alloc( return -ESRCH; } - /* Create the block mapping. */ xfs_trans_ijoin(tp, quotip, XFS_ILOCK_EXCL); + + error = xfs_iext_count_may_overflow(quotip, XFS_DATA_FORK, + XFS_IEXT_ADD_NOSPLIT_CNT); + if (error) + return error; + + /* Create the block mapping. */ error = xfs_bmapi_write(tp, quotip, dqp->q_fileoffset, XFS_DQUOT_CLUSTER_SIZE_FSB, XFS_BMAPI_METADATA, 0, &map, &nmaps); @@ -500,6 +506,42 @@ xfs_dquot_alloc( return dqp; } +/* Check the ondisk dquot's id and type match what the incore dquot expects. */ +static bool +xfs_dquot_check_type( + struct xfs_dquot *dqp, + struct xfs_disk_dquot *ddqp) +{ + uint8_t ddqp_type; + uint8_t dqp_type; + + ddqp_type = ddqp->d_type & XFS_DQTYPE_REC_MASK; + dqp_type = xfs_dquot_type(dqp); + + if (be32_to_cpu(ddqp->d_id) != dqp->q_id) + return false; + + /* + * V5 filesystems always expect an exact type match. V4 filesystems + * expect an exact match for user dquots and for non-root group and + * project dquots. + */ + if (xfs_sb_version_hascrc(&dqp->q_mount->m_sb) || + dqp_type == XFS_DQTYPE_USER || dqp->q_id != 0) + return ddqp_type == dqp_type; + + /* + * V4 filesystems support either group or project quotas, but not both + * at the same time. The non-user quota file can be switched between + * group and project quota uses depending on the mount options, which + * means that we can encounter the other type when we try to load quota + * defaults. Quotacheck will soon reset the the entire quota file + * (including the root dquot) anyway, but don't log scary corruption + * reports to dmesg. + */ + return ddqp_type == XFS_DQTYPE_GROUP || ddqp_type == XFS_DQTYPE_PROJ; +} + /* Copy the in-core quota fields in from the on-disk buffer. */ STATIC int xfs_dquot_from_disk( @@ -512,8 +554,7 @@ xfs_dquot_from_disk( * Ensure that we got the type and ID we were looking for. * Everything else was checked by the dquot buffer verifier. */ - if ((ddqp->d_type & XFS_DQTYPE_REC_MASK) != xfs_dquot_type(dqp) || - be32_to_cpu(ddqp->d_id) != dqp->q_id) { + if (!xfs_dquot_check_type(dqp, ddqp)) { xfs_alert_tag(bp->b_mount, XFS_PTAG_VERIFIER_ERROR, "Metadata corruption detected at %pS, quota %u", __this_address, dqp->q_id); diff --git a/fs/xfs/xfs_error.c b/fs/xfs/xfs_error.c index 7f6e20899473..185b4915b7bf 100644 --- a/fs/xfs/xfs_error.c +++ b/fs/xfs/xfs_error.c @@ -54,6 +54,8 @@ static unsigned int xfs_errortag_random_default[] = { XFS_RANDOM_FORCE_SUMMARY_RECALC, XFS_RANDOM_IUNLINK_FALLBACK, XFS_RANDOM_BUF_IOERROR, + XFS_RANDOM_REDUCE_MAX_IEXTENTS, + XFS_RANDOM_BMAP_ALLOC_MINLEN_EXTENT, }; struct xfs_errortag_attr { @@ -164,6 +166,8 @@ XFS_ERRORTAG_ATTR_RW(force_repair, XFS_ERRTAG_FORCE_SCRUB_REPAIR); XFS_ERRORTAG_ATTR_RW(bad_summary, XFS_ERRTAG_FORCE_SUMMARY_RECALC); XFS_ERRORTAG_ATTR_RW(iunlink_fallback, XFS_ERRTAG_IUNLINK_FALLBACK); XFS_ERRORTAG_ATTR_RW(buf_ioerror, XFS_ERRTAG_BUF_IOERROR); +XFS_ERRORTAG_ATTR_RW(reduce_max_iextents, XFS_ERRTAG_REDUCE_MAX_IEXTENTS); +XFS_ERRORTAG_ATTR_RW(bmap_alloc_minlen_extent, XFS_ERRTAG_BMAP_ALLOC_MINLEN_EXTENT); static struct attribute *xfs_errortag_attrs[] = { XFS_ERRORTAG_ATTR_LIST(noerror), @@ -202,6 +206,8 @@ static struct attribute *xfs_errortag_attrs[] = { XFS_ERRORTAG_ATTR_LIST(bad_summary), XFS_ERRORTAG_ATTR_LIST(iunlink_fallback), XFS_ERRORTAG_ATTR_LIST(buf_ioerror), + XFS_ERRORTAG_ATTR_LIST(reduce_max_iextents), + XFS_ERRORTAG_ATTR_LIST(bmap_alloc_minlen_extent), NULL, }; diff --git a/fs/xfs/xfs_extent_busy.c b/fs/xfs/xfs_extent_busy.c index 3991e59cfd18..ef17c1f6db32 100644 --- a/fs/xfs/xfs_extent_busy.c +++ b/fs/xfs/xfs_extent_busy.c @@ -344,7 +344,6 @@ xfs_extent_busy_trim( ASSERT(*len > 0); spin_lock(&args->pag->pagb_lock); -restart: fbno = *bno; flen = *len; rbp = args->pag->pagb_tree.rb_node; @@ -363,19 +362,6 @@ restart: continue; } - /* - * If this is a metadata allocation, try to reuse the busy - * extent instead of trimming the allocation. - */ - if (!(args->datatype & XFS_ALLOC_USERDATA) && - !(busyp->flags & XFS_EXTENT_BUSY_DISCARDED)) { - if (!xfs_extent_busy_update_extent(args->mp, args->pag, - busyp, fbno, flen, - false)) - goto restart; - continue; - } - if (bbno <= fbno) { /* start overlap */ diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index 5b0f93f73837..a007ca0711d9 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -29,6 +29,7 @@ #include <linux/backing-dev.h> #include <linux/mman.h> #include <linux/fadvise.h> +#include <linux/mount.h> static const struct vm_operations_struct xfs_file_vm_ops; @@ -118,6 +119,54 @@ xfs_dir_fsync( return xfs_log_force_inode(ip); } +static xfs_lsn_t +xfs_fsync_lsn( + struct xfs_inode *ip, + bool datasync) +{ + if (!xfs_ipincount(ip)) + return 0; + if (datasync && !(ip->i_itemp->ili_fsync_fields & ~XFS_ILOG_TIMESTAMP)) + return 0; + return ip->i_itemp->ili_last_lsn; +} + +/* + * All metadata updates are logged, which means that we just have to flush the + * log up to the latest LSN that touched the inode. + * + * If we have concurrent fsync/fdatasync() calls, we need them to all block on + * the log force before we clear the ili_fsync_fields field. This ensures that + * we don't get a racing sync operation that does not wait for the metadata to + * hit the journal before returning. If we race with clearing ili_fsync_fields, + * then all that will happen is the log force will do nothing as the lsn will + * already be on disk. We can't race with setting ili_fsync_fields because that + * is done under XFS_ILOCK_EXCL, and that can't happen because we hold the lock + * shared until after the ili_fsync_fields is cleared. + */ +static int +xfs_fsync_flush_log( + struct xfs_inode *ip, + bool datasync, + int *log_flushed) +{ + int error = 0; + xfs_lsn_t lsn; + + xfs_ilock(ip, XFS_ILOCK_SHARED); + lsn = xfs_fsync_lsn(ip, datasync); + if (lsn) { + error = xfs_log_force_lsn(ip->i_mount, lsn, XFS_LOG_SYNC, + log_flushed); + + spin_lock(&ip->i_itemp->ili_lock); + ip->i_itemp->ili_fsync_fields = 0; + spin_unlock(&ip->i_itemp->ili_lock); + } + xfs_iunlock(ip, XFS_ILOCK_SHARED); + return error; +} + STATIC int xfs_file_fsync( struct file *file, @@ -125,13 +174,10 @@ xfs_file_fsync( loff_t end, int datasync) { - struct inode *inode = file->f_mapping->host; - struct xfs_inode *ip = XFS_I(inode); - struct xfs_inode_log_item *iip = ip->i_itemp; + struct xfs_inode *ip = XFS_I(file->f_mapping->host); struct xfs_mount *mp = ip->i_mount; int error = 0; int log_flushed = 0; - xfs_lsn_t lsn = 0; trace_xfs_file_fsync(ip); @@ -156,32 +202,13 @@ xfs_file_fsync( xfs_blkdev_issue_flush(mp->m_ddev_targp); /* - * All metadata updates are logged, which means that we just have to - * flush the log up to the latest LSN that touched the inode. If we have - * concurrent fsync/fdatasync() calls, we need them to all block on the - * log force before we clear the ili_fsync_fields field. This ensures - * that we don't get a racing sync operation that does not wait for the - * metadata to hit the journal before returning. If we race with - * clearing the ili_fsync_fields, then all that will happen is the log - * force will do nothing as the lsn will already be on disk. We can't - * race with setting ili_fsync_fields because that is done under - * XFS_ILOCK_EXCL, and that can't happen because we hold the lock shared - * until after the ili_fsync_fields is cleared. + * Any inode that has dirty modifications in the log is pinned. The + * racy check here for a pinned inode while not catch modifications + * that happen concurrently to the fsync call, but fsync semantics + * only require to sync previously completed I/O. */ - xfs_ilock(ip, XFS_ILOCK_SHARED); - if (xfs_ipincount(ip)) { - if (!datasync || - (iip->ili_fsync_fields & ~XFS_ILOG_TIMESTAMP)) - lsn = iip->ili_last_lsn; - } - - if (lsn) { - error = xfs_log_force_lsn(mp, lsn, XFS_LOG_SYNC, &log_flushed); - spin_lock(&iip->ili_lock); - iip->ili_fsync_fields = 0; - spin_unlock(&iip->ili_lock); - } - xfs_iunlock(ip, XFS_ILOCK_SHARED); + if (xfs_ipincount(ip)) + error = xfs_fsync_flush_log(ip, datasync, &log_flushed); /* * If we only have a single device, and the log force about was @@ -197,30 +224,42 @@ xfs_file_fsync( return error; } +static int +xfs_ilock_iocb( + struct kiocb *iocb, + unsigned int lock_mode) +{ + struct xfs_inode *ip = XFS_I(file_inode(iocb->ki_filp)); + + if (iocb->ki_flags & IOCB_NOWAIT) { + if (!xfs_ilock_nowait(ip, lock_mode)) + return -EAGAIN; + } else { + xfs_ilock(ip, lock_mode); + } + + return 0; +} + STATIC ssize_t -xfs_file_dio_aio_read( +xfs_file_dio_read( struct kiocb *iocb, struct iov_iter *to) { struct xfs_inode *ip = XFS_I(file_inode(iocb->ki_filp)); - size_t count = iov_iter_count(to); ssize_t ret; - trace_xfs_file_direct_read(ip, count, iocb->ki_pos); + trace_xfs_file_direct_read(iocb, to); - if (!count) + if (!iov_iter_count(to)) return 0; /* skip atime */ file_accessed(iocb->ki_filp); - if (iocb->ki_flags & IOCB_NOWAIT) { - if (!xfs_ilock_nowait(ip, XFS_IOLOCK_SHARED)) - return -EAGAIN; - } else { - xfs_ilock(ip, XFS_IOLOCK_SHARED); - } - ret = iomap_dio_rw(iocb, to, &xfs_read_iomap_ops, NULL, - is_sync_kiocb(iocb)); + ret = xfs_ilock_iocb(iocb, XFS_IOLOCK_SHARED); + if (ret) + return ret; + ret = iomap_dio_rw(iocb, to, &xfs_read_iomap_ops, NULL, 0); xfs_iunlock(ip, XFS_IOLOCK_SHARED); return ret; @@ -232,21 +271,16 @@ xfs_file_dax_read( struct iov_iter *to) { struct xfs_inode *ip = XFS_I(iocb->ki_filp->f_mapping->host); - size_t count = iov_iter_count(to); ssize_t ret = 0; - trace_xfs_file_dax_read(ip, count, iocb->ki_pos); + trace_xfs_file_dax_read(iocb, to); - if (!count) + if (!iov_iter_count(to)) return 0; /* skip atime */ - if (iocb->ki_flags & IOCB_NOWAIT) { - if (!xfs_ilock_nowait(ip, XFS_IOLOCK_SHARED)) - return -EAGAIN; - } else { - xfs_ilock(ip, XFS_IOLOCK_SHARED); - } - + ret = xfs_ilock_iocb(iocb, XFS_IOLOCK_SHARED); + if (ret) + return ret; ret = dax_iomap_rw(iocb, to, &xfs_read_iomap_ops); xfs_iunlock(ip, XFS_IOLOCK_SHARED); @@ -255,21 +289,18 @@ xfs_file_dax_read( } STATIC ssize_t -xfs_file_buffered_aio_read( +xfs_file_buffered_read( struct kiocb *iocb, struct iov_iter *to) { struct xfs_inode *ip = XFS_I(file_inode(iocb->ki_filp)); ssize_t ret; - trace_xfs_file_buffered_read(ip, iov_iter_count(to), iocb->ki_pos); + trace_xfs_file_buffered_read(iocb, to); - if (iocb->ki_flags & IOCB_NOWAIT) { - if (!xfs_ilock_nowait(ip, XFS_IOLOCK_SHARED)) - return -EAGAIN; - } else { - xfs_ilock(ip, XFS_IOLOCK_SHARED); - } + ret = xfs_ilock_iocb(iocb, XFS_IOLOCK_SHARED); + if (ret) + return ret; ret = generic_file_read_iter(iocb, to); xfs_iunlock(ip, XFS_IOLOCK_SHARED); @@ -293,9 +324,9 @@ xfs_file_read_iter( if (IS_DAX(inode)) ret = xfs_file_dax_read(iocb, to); else if (iocb->ki_flags & IOCB_DIRECT) - ret = xfs_file_dio_aio_read(iocb, to); + ret = xfs_file_dio_read(iocb, to); else - ret = xfs_file_buffered_aio_read(iocb, to); + ret = xfs_file_buffered_read(iocb, to); if (ret > 0) XFS_STATS_ADD(mp, xs_read_bytes, ret); @@ -310,7 +341,7 @@ xfs_file_read_iter( * if called for a direct write beyond i_size. */ STATIC ssize_t -xfs_file_aio_write_checks( +xfs_file_write_checks( struct kiocb *iocb, struct iov_iter *from, int *iolock) @@ -328,7 +359,14 @@ restart: if (error <= 0) return error; - error = xfs_break_layouts(inode, iolock, BREAK_WRITE); + if (iocb->ki_flags & IOCB_NOWAIT) { + error = break_layout(inode, false); + if (error == -EWOULDBLOCK) + error = -EAGAIN; + } else { + error = xfs_break_layouts(inode, iolock, BREAK_WRITE); + } + if (error) return error; @@ -339,7 +377,11 @@ restart: if (*iolock == XFS_IOLOCK_SHARED && !IS_NOSEC(inode)) { xfs_iunlock(ip, *iolock); *iolock = XFS_IOLOCK_EXCL; - xfs_ilock(ip, *iolock); + error = xfs_ilock_iocb(iocb, *iolock); + if (error) { + *iolock = 0; + return error; + } goto restart; } /* @@ -361,6 +403,10 @@ restart: isize = i_size_read(inode); if (iocb->ki_pos > isize) { spin_unlock(&ip->i_flags_lock); + + if (iocb->ki_flags & IOCB_NOWAIT) + return -EAGAIN; + if (!drained_dio) { if (*iolock == XFS_IOLOCK_SHARED) { xfs_iunlock(ip, *iolock); @@ -389,12 +435,6 @@ restart: } else spin_unlock(&ip->i_flags_lock); - /* - * Updating the timestamps will grab the ilock again from - * xfs_fs_dirty_inode, so we have to call it after dropping the - * lock above. Eventually we should look into a way to avoid - * the pointless lock roundtrip. - */ return file_modified(file); } @@ -480,122 +520,149 @@ static const struct iomap_dio_ops xfs_dio_write_ops = { }; /* - * xfs_file_dio_aio_write - handle direct IO writes - * - * Lock the inode appropriately to prepare for and issue a direct IO write. - * By separating it from the buffered write path we remove all the tricky to - * follow locking changes and looping. - * - * If there are cached pages or we're extending the file, we need IOLOCK_EXCL - * until we're sure the bytes at the new EOF have been zeroed and/or the cached - * pages are flushed out. - * - * In most cases the direct IO writes will be done holding IOLOCK_SHARED - * allowing them to be done in parallel with reads and other direct IO writes. - * However, if the IO is not aligned to filesystem blocks, the direct IO layer - * needs to do sub-block zeroing and that requires serialisation against other - * direct IOs to the same block. In this case we need to serialise the - * submission of the unaligned IOs so that we don't get racing block zeroing in - * the dio layer. To avoid the problem with aio, we also need to wait for - * outstanding IOs to complete so that unwritten extent conversion is completed - * before we try to map the overlapping block. This is currently implemented by - * hitting it with a big hammer (i.e. inode_dio_wait()). - * - * Returns with locks held indicated by @iolock and errors indicated by - * negative return values. + * Handle block aligned direct I/O writes */ -STATIC ssize_t -xfs_file_dio_aio_write( +static noinline ssize_t +xfs_file_dio_write_aligned( + struct xfs_inode *ip, struct kiocb *iocb, struct iov_iter *from) { - struct file *file = iocb->ki_filp; - struct address_space *mapping = file->f_mapping; - struct inode *inode = mapping->host; - struct xfs_inode *ip = XFS_I(inode); - struct xfs_mount *mp = ip->i_mount; - ssize_t ret = 0; - int unaligned_io = 0; - int iolock; - size_t count = iov_iter_count(from); - struct xfs_buftarg *target = xfs_inode_buftarg(ip); + int iolock = XFS_IOLOCK_SHARED; + ssize_t ret; - /* DIO must be aligned to device logical sector size */ - if ((iocb->ki_pos | count) & target->bt_logical_sectormask) - return -EINVAL; + ret = xfs_ilock_iocb(iocb, iolock); + if (ret) + return ret; + ret = xfs_file_write_checks(iocb, from, &iolock); + if (ret) + goto out_unlock; /* - * Don't take the exclusive iolock here unless the I/O is unaligned to - * the file system block size. We don't need to consider the EOF - * extension case here because xfs_file_aio_write_checks() will relock - * the inode as necessary for EOF zeroing cases and fill out the new - * inode size as appropriate. + * We don't need to hold the IOLOCK exclusively across the IO, so demote + * the iolock back to shared if we had to take the exclusive lock in + * xfs_file_write_checks() for other reasons. */ - if ((iocb->ki_pos & mp->m_blockmask) || - ((iocb->ki_pos + count) & mp->m_blockmask)) { - unaligned_io = 1; - - /* - * We can't properly handle unaligned direct I/O to reflink - * files yet, as we can't unshare a partial block. - */ - if (xfs_is_cow_inode(ip)) { - trace_xfs_reflink_bounce_dio_write(ip, iocb->ki_pos, count); - return -ENOTBLK; - } - iolock = XFS_IOLOCK_EXCL; - } else { + if (iolock == XFS_IOLOCK_EXCL) { + xfs_ilock_demote(ip, XFS_IOLOCK_EXCL); iolock = XFS_IOLOCK_SHARED; } + trace_xfs_file_direct_write(iocb, from); + ret = iomap_dio_rw(iocb, from, &xfs_direct_write_iomap_ops, + &xfs_dio_write_ops, 0); +out_unlock: + if (iolock) + xfs_iunlock(ip, iolock); + return ret; +} - if (iocb->ki_flags & IOCB_NOWAIT) { - /* unaligned dio always waits, bail */ - if (unaligned_io) - return -EAGAIN; - if (!xfs_ilock_nowait(ip, iolock)) +/* + * Handle block unaligned direct I/O writes + * + * In most cases direct I/O writes will be done holding IOLOCK_SHARED, allowing + * them to be done in parallel with reads and other direct I/O writes. However, + * if the I/O is not aligned to filesystem blocks, the direct I/O layer may need + * to do sub-block zeroing and that requires serialisation against other direct + * I/O to the same block. In this case we need to serialise the submission of + * the unaligned I/O so that we don't get racing block zeroing in the dio layer. + * In the case where sub-block zeroing is not required, we can do concurrent + * sub-block dios to the same block successfully. + * + * Optimistically submit the I/O using the shared lock first, but use the + * IOMAP_DIO_OVERWRITE_ONLY flag to tell the lower layers to return -EAGAIN + * if block allocation or partial block zeroing would be required. In that case + * we try again with the exclusive lock. + */ +static noinline ssize_t +xfs_file_dio_write_unaligned( + struct xfs_inode *ip, + struct kiocb *iocb, + struct iov_iter *from) +{ + size_t isize = i_size_read(VFS_I(ip)); + size_t count = iov_iter_count(from); + int iolock = XFS_IOLOCK_SHARED; + unsigned int flags = IOMAP_DIO_OVERWRITE_ONLY; + ssize_t ret; + + /* + * Extending writes need exclusivity because of the sub-block zeroing + * that the DIO code always does for partial tail blocks beyond EOF, so + * don't even bother trying the fast path in this case. + */ + if (iocb->ki_pos > isize || iocb->ki_pos + count >= isize) { +retry_exclusive: + if (iocb->ki_flags & IOCB_NOWAIT) return -EAGAIN; - } else { - xfs_ilock(ip, iolock); + iolock = XFS_IOLOCK_EXCL; + flags = IOMAP_DIO_FORCE_WAIT; } - ret = xfs_file_aio_write_checks(iocb, from, &iolock); + ret = xfs_ilock_iocb(iocb, iolock); if (ret) - goto out; - count = iov_iter_count(from); + return ret; /* - * If we are doing unaligned IO, we can't allow any other overlapping IO - * in-flight at the same time or we risk data corruption. Wait for all - * other IO to drain before we submit. If the IO is aligned, demote the - * iolock if we had to take the exclusive lock in - * xfs_file_aio_write_checks() for other reasons. + * We can't properly handle unaligned direct I/O to reflink files yet, + * as we can't unshare a partial block. */ - if (unaligned_io) { - inode_dio_wait(inode); - } else if (iolock == XFS_IOLOCK_EXCL) { - xfs_ilock_demote(ip, XFS_IOLOCK_EXCL); - iolock = XFS_IOLOCK_SHARED; + if (xfs_is_cow_inode(ip)) { + trace_xfs_reflink_bounce_dio_write(iocb, from); + ret = -ENOTBLK; + goto out_unlock; } - trace_xfs_file_direct_write(ip, count, iocb->ki_pos); + ret = xfs_file_write_checks(iocb, from, &iolock); + if (ret) + goto out_unlock; + /* - * If unaligned, this is the only IO in-flight. Wait on it before we - * release the iolock to prevent subsequent overlapping IO. + * If we are doing exclusive unaligned I/O, this must be the only I/O + * in-flight. Otherwise we risk data corruption due to unwritten extent + * conversions from the AIO end_io handler. Wait for all other I/O to + * drain first. */ + if (flags & IOMAP_DIO_FORCE_WAIT) + inode_dio_wait(VFS_I(ip)); + + trace_xfs_file_direct_write(iocb, from); ret = iomap_dio_rw(iocb, from, &xfs_direct_write_iomap_ops, - &xfs_dio_write_ops, - is_sync_kiocb(iocb) || unaligned_io); -out: - xfs_iunlock(ip, iolock); + &xfs_dio_write_ops, flags); /* - * No fallback to buffered IO after short writes for XFS, direct I/O - * will either complete fully or return an error. + * Retry unaligned I/O with exclusive blocking semantics if the DIO + * layer rejected it for mapping or locking reasons. If we are doing + * nonblocking user I/O, propagate the error. */ - ASSERT(ret < 0 || ret == count); + if (ret == -EAGAIN && !(iocb->ki_flags & IOCB_NOWAIT)) { + ASSERT(flags & IOMAP_DIO_OVERWRITE_ONLY); + xfs_iunlock(ip, iolock); + goto retry_exclusive; + } + +out_unlock: + if (iolock) + xfs_iunlock(ip, iolock); return ret; } +static ssize_t +xfs_file_dio_write( + struct kiocb *iocb, + struct iov_iter *from) +{ + struct xfs_inode *ip = XFS_I(file_inode(iocb->ki_filp)); + struct xfs_buftarg *target = xfs_inode_buftarg(ip); + size_t count = iov_iter_count(from); + + /* direct I/O must be aligned to device logical sector size */ + if ((iocb->ki_pos | count) & target->bt_logical_sectormask) + return -EINVAL; + if ((iocb->ki_pos | count) & ip->i_mount->m_blockmask) + return xfs_file_dio_write_unaligned(ip, iocb, from); + return xfs_file_dio_write_aligned(ip, iocb, from); +} + static noinline ssize_t xfs_file_dax_write( struct kiocb *iocb, @@ -605,31 +672,26 @@ xfs_file_dax_write( struct xfs_inode *ip = XFS_I(inode); int iolock = XFS_IOLOCK_EXCL; ssize_t ret, error = 0; - size_t count; loff_t pos; - if (iocb->ki_flags & IOCB_NOWAIT) { - if (!xfs_ilock_nowait(ip, iolock)) - return -EAGAIN; - } else { - xfs_ilock(ip, iolock); - } - - ret = xfs_file_aio_write_checks(iocb, from, &iolock); + ret = xfs_ilock_iocb(iocb, iolock); + if (ret) + return ret; + ret = xfs_file_write_checks(iocb, from, &iolock); if (ret) goto out; pos = iocb->ki_pos; - count = iov_iter_count(from); - trace_xfs_file_dax_write(ip, count, pos); + trace_xfs_file_dax_write(iocb, from); ret = dax_iomap_rw(iocb, from, &xfs_direct_write_iomap_ops); if (ret > 0 && iocb->ki_pos > i_size_read(inode)) { i_size_write(inode, iocb->ki_pos); error = xfs_setfilesize(ip, pos, ret); } out: - xfs_iunlock(ip, iolock); + if (iolock) + xfs_iunlock(ip, iolock); if (error) return error; @@ -643,7 +705,7 @@ out: } STATIC ssize_t -xfs_file_buffered_aio_write( +xfs_file_buffered_write( struct kiocb *iocb, struct iov_iter *from) { @@ -652,7 +714,7 @@ xfs_file_buffered_aio_write( struct inode *inode = mapping->host; struct xfs_inode *ip = XFS_I(inode); ssize_t ret; - int enospc = 0; + bool cleared_space = false; int iolock; if (iocb->ki_flags & IOCB_NOWAIT) @@ -662,14 +724,14 @@ write_retry: iolock = XFS_IOLOCK_EXCL; xfs_ilock(ip, iolock); - ret = xfs_file_aio_write_checks(iocb, from, &iolock); + ret = xfs_file_write_checks(iocb, from, &iolock); if (ret) goto out; /* We can write back this queue in page reclaim */ current->backing_dev_info = inode_to_bdi(inode); - trace_xfs_file_buffered_write(ip, iov_iter_count(from), iocb->ki_pos); + trace_xfs_file_buffered_write(iocb, from); ret = iomap_file_buffered_write(iocb, from, &xfs_buffered_write_iomap_ops); if (likely(ret >= 0)) @@ -682,27 +744,23 @@ write_retry: * metadata space. This reduces the chances that the eofblocks scan * waits on dirty mappings. Since xfs_flush_inodes() is serialized, this * also behaves as a filter to prevent too many eofblocks scans from - * running at the same time. + * running at the same time. Use a synchronous scan to increase the + * effectiveness of the scan. */ - if (ret == -EDQUOT && !enospc) { + if (ret == -EDQUOT && !cleared_space) { xfs_iunlock(ip, iolock); - enospc = xfs_inode_free_quota_eofblocks(ip); - if (enospc) - goto write_retry; - enospc = xfs_inode_free_quota_cowblocks(ip); - if (enospc) - goto write_retry; - iolock = 0; - } else if (ret == -ENOSPC && !enospc) { + xfs_blockgc_free_quota(ip, XFS_EOF_FLAGS_SYNC); + cleared_space = true; + goto write_retry; + } else if (ret == -ENOSPC && !cleared_space) { struct xfs_eofblocks eofb = {0}; - enospc = 1; + cleared_space = true; xfs_flush_inodes(ip->i_mount); xfs_iunlock(ip, iolock); eofb.eof_flags = XFS_EOF_FLAGS_SYNC; - xfs_icache_free_eofblocks(ip->i_mount, &eofb); - xfs_icache_free_cowblocks(ip->i_mount, &eofb); + xfs_blockgc_free_space(ip->i_mount, &eofb); goto write_retry; } @@ -749,12 +807,12 @@ xfs_file_write_iter( * CoW. In all other directio scenarios we do not * allow an operation to fall back to buffered mode. */ - ret = xfs_file_dio_aio_write(iocb, from); + ret = xfs_file_dio_write(iocb, from); if (ret != -ENOTBLK) return ret; } - return xfs_file_buffered_aio_write(iocb, from); + return xfs_file_buffered_write(iocb, from); } static void @@ -994,7 +1052,8 @@ xfs_file_fallocate( iattr.ia_valid = ATTR_SIZE; iattr.ia_size = new_size; - error = xfs_vn_setattr_size(file_dentry(file), &iattr); + error = xfs_vn_setattr_size(file_mnt_user_ns(file), + file_dentry(file), &iattr); if (error) goto out_unlock; } @@ -1319,17 +1378,19 @@ xfs_filemap_pfn_mkwrite( return __xfs_filemap_fault(vmf, PE_SIZE_PTE, true); } -static void +static vm_fault_t xfs_filemap_map_pages( struct vm_fault *vmf, pgoff_t start_pgoff, pgoff_t end_pgoff) { struct inode *inode = file_inode(vmf->vma->vm_file); + vm_fault_t ret; xfs_ilock(XFS_I(inode), XFS_MMAPLOCK_SHARED); - filemap_map_pages(vmf, start_pgoff, end_pgoff); + ret = filemap_map_pages(vmf, start_pgoff, end_pgoff); xfs_iunlock(XFS_I(inode), XFS_MMAPLOCK_SHARED); + return ret; } static const struct vm_operations_struct xfs_file_vm_ops = { diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c index 959ce91a3755..a2a407039227 100644 --- a/fs/xfs/xfs_fsops.c +++ b/fs/xfs/xfs_fsops.c @@ -25,17 +25,17 @@ */ static int xfs_growfs_data_private( - xfs_mount_t *mp, /* mount point for filesystem */ - xfs_growfs_data_t *in) /* growfs data input struct */ + struct xfs_mount *mp, /* mount point for filesystem */ + struct xfs_growfs_data *in) /* growfs data input struct */ { struct xfs_buf *bp; int error; xfs_agnumber_t nagcount; xfs_agnumber_t nagimax = 0; - xfs_rfsblock_t nb, nb_mod; - xfs_rfsblock_t new; + xfs_rfsblock_t nb, nb_div, nb_mod; + xfs_rfsblock_t delta; xfs_agnumber_t oagcount; - xfs_trans_t *tp; + struct xfs_trans *tp; struct aghdr_init_data id = {}; nb = in->newblocks; @@ -50,16 +50,16 @@ xfs_growfs_data_private( return error; xfs_buf_relse(bp); - new = nb; /* use new as a temporary here */ - nb_mod = do_div(new, mp->m_sb.sb_agblocks); - nagcount = new + (nb_mod != 0); + nb_div = nb; + nb_mod = do_div(nb_div, mp->m_sb.sb_agblocks); + nagcount = nb_div + (nb_mod != 0); if (nb_mod && nb_mod < XFS_MIN_AG_BLOCKS) { nagcount--; nb = (xfs_rfsblock_t)nagcount * mp->m_sb.sb_agblocks; if (nb < mp->m_sb.sb_dblocks) return -EINVAL; } - new = nb - mp->m_sb.sb_dblocks; + delta = nb - mp->m_sb.sb_dblocks; oagcount = mp->m_sb.sb_agcount; /* allocate the new per-ag structures */ @@ -89,7 +89,7 @@ xfs_growfs_data_private( INIT_LIST_HEAD(&id.buffer_list); for (id.agno = nagcount - 1; id.agno >= oagcount; - id.agno--, new -= id.agsize) { + id.agno--, delta -= id.agsize) { if (id.agno == nagcount - 1) id.agsize = nb - @@ -110,8 +110,8 @@ xfs_growfs_data_private( xfs_trans_agblocks_delta(tp, id.nfree); /* If there are new blocks in the old last AG, extend it. */ - if (new) { - error = xfs_ag_extend_space(mp, tp, &id, new); + if (delta) { + error = xfs_ag_extend_space(mp, tp, &id, delta); if (error) goto out_trans_cancel; } @@ -143,7 +143,7 @@ xfs_growfs_data_private( * If we expanded the last AG, free the per-AG reservation * so we can reinitialize it with the new size. */ - if (new) { + if (delta) { struct xfs_perag *pag; pag = xfs_perag_get(mp, id.agno); @@ -170,8 +170,8 @@ out_trans_cancel: static int xfs_growfs_log_private( - xfs_mount_t *mp, /* mount point for filesystem */ - xfs_growfs_log_t *in) /* growfs log input struct */ + struct xfs_mount *mp, /* mount point for filesystem */ + struct xfs_growfs_log *in) /* growfs log input struct */ { xfs_extlen_t nb; @@ -268,7 +268,7 @@ out_error: int xfs_growfs_log( xfs_mount_t *mp, - xfs_growfs_log_t *in) + struct xfs_growfs_log *in) { int error; diff --git a/fs/xfs/xfs_fsops.h b/fs/xfs/xfs_fsops.h index 92869f6ec8d3..2cffe51a31e8 100644 --- a/fs/xfs/xfs_fsops.h +++ b/fs/xfs/xfs_fsops.h @@ -6,8 +6,8 @@ #ifndef __XFS_FSOPS_H__ #define __XFS_FSOPS_H__ -extern int xfs_growfs_data(xfs_mount_t *mp, xfs_growfs_data_t *in); -extern int xfs_growfs_log(xfs_mount_t *mp, xfs_growfs_log_t *in); +extern int xfs_growfs_data(struct xfs_mount *mp, struct xfs_growfs_data *in); +extern int xfs_growfs_log(struct xfs_mount *mp, struct xfs_growfs_log *in); extern void xfs_fs_counts(xfs_mount_t *mp, xfs_fsop_counts_t *cnt); extern int xfs_reserve_blocks(xfs_mount_t *mp, uint64_t *inval, xfs_fsop_resblks_t *outval); diff --git a/fs/xfs/xfs_globals.c b/fs/xfs/xfs_globals.c index fa55ab8b8d80..f62fa652c2fd 100644 --- a/fs/xfs/xfs_globals.c +++ b/fs/xfs/xfs_globals.c @@ -8,8 +8,8 @@ /* * Tunable XFS parameters. xfs_params is required even when CONFIG_SYSCTL=n, * other XFS code uses these values. Times are measured in centisecs (i.e. - * 100ths of a second) with the exception of eofb_timer and cowb_timer, which - * are measured in seconds. + * 100ths of a second) with the exception of blockgc_timer, which is measured + * in seconds. */ xfs_param_t xfs_params = { /* MIN DFLT MAX */ @@ -28,8 +28,7 @@ xfs_param_t xfs_params = { .rotorstep = { 1, 1, 255 }, .inherit_nodfrg = { 0, 1, 1 }, .fstrm_timer = { 1, 30*100, 3600*100}, - .eofb_timer = { 1, 300, 3600*24}, - .cowb_timer = { 1, 1800, 3600*24}, + .blockgc_timer = { 1, 300, 3600*24}, }; struct xfs_globals xfs_globals = { diff --git a/fs/xfs/xfs_icache.c b/fs/xfs/xfs_icache.c index deb99300d171..1d7720a0c068 100644 --- a/fs/xfs/xfs_icache.c +++ b/fs/xfs/xfs_icache.c @@ -916,69 +916,6 @@ xfs_inode_walk( } /* - * Background scanning to trim post-EOF preallocated space. This is queued - * based on the 'speculative_prealloc_lifetime' tunable (5m by default). - */ -void -xfs_queue_eofblocks( - struct xfs_mount *mp) -{ - rcu_read_lock(); - if (radix_tree_tagged(&mp->m_perag_tree, XFS_ICI_EOFBLOCKS_TAG)) - queue_delayed_work(mp->m_eofblocks_workqueue, - &mp->m_eofblocks_work, - msecs_to_jiffies(xfs_eofb_secs * 1000)); - rcu_read_unlock(); -} - -void -xfs_eofblocks_worker( - struct work_struct *work) -{ - struct xfs_mount *mp = container_of(to_delayed_work(work), - struct xfs_mount, m_eofblocks_work); - - if (!sb_start_write_trylock(mp->m_super)) - return; - xfs_icache_free_eofblocks(mp, NULL); - sb_end_write(mp->m_super); - - xfs_queue_eofblocks(mp); -} - -/* - * Background scanning to trim preallocated CoW space. This is queued - * based on the 'speculative_cow_prealloc_lifetime' tunable (5m by default). - * (We'll just piggyback on the post-EOF prealloc space workqueue.) - */ -void -xfs_queue_cowblocks( - struct xfs_mount *mp) -{ - rcu_read_lock(); - if (radix_tree_tagged(&mp->m_perag_tree, XFS_ICI_COWBLOCKS_TAG)) - queue_delayed_work(mp->m_eofblocks_workqueue, - &mp->m_cowblocks_work, - msecs_to_jiffies(xfs_cowb_secs * 1000)); - rcu_read_unlock(); -} - -void -xfs_cowblocks_worker( - struct work_struct *work) -{ - struct xfs_mount *mp = container_of(to_delayed_work(work), - struct xfs_mount, m_cowblocks_work); - - if (!sb_start_write_trylock(mp->m_super)) - return; - xfs_icache_free_cowblocks(mp, NULL); - sb_end_write(mp->m_super); - - xfs_queue_cowblocks(mp); -} - -/* * Grab the inode for reclaim exclusively. * * We have found this inode via a lookup under RCU, so the inode may have @@ -1346,14 +1283,17 @@ xfs_reclaim_worker( STATIC int xfs_inode_free_eofblocks( struct xfs_inode *ip, - void *args) + void *args, + unsigned int *lockflags) { struct xfs_eofblocks *eofb = args; bool wait; - int ret; wait = eofb && (eofb->eof_flags & XFS_EOF_FLAGS_SYNC); + if (!xfs_iflags_test(ip, XFS_IEOFBLOCKS)) + return 0; + if (!xfs_can_free_eofblocks(ip, false)) { /* inode could be preallocated or append-only */ trace_xfs_inode_free_eofblocks_invalid(ip); @@ -1380,130 +1320,68 @@ xfs_inode_free_eofblocks( return -EAGAIN; return 0; } + *lockflags |= XFS_IOLOCK_EXCL; - ret = xfs_free_eofblocks(ip); - xfs_iunlock(ip, XFS_IOLOCK_EXCL); - - return ret; -} - -int -xfs_icache_free_eofblocks( - struct xfs_mount *mp, - struct xfs_eofblocks *eofb) -{ - return xfs_inode_walk(mp, 0, xfs_inode_free_eofblocks, eofb, - XFS_ICI_EOFBLOCKS_TAG); + return xfs_free_eofblocks(ip); } /* - * Run eofblocks scans on the quotas applicable to the inode. For inodes with - * multiple quotas, we don't know exactly which quota caused an allocation - * failure. We make a best effort by including each quota under low free space - * conditions (less than 1% free space) in the scan. + * Background scanning to trim preallocated space. This is queued based on the + * 'speculative_prealloc_lifetime' tunable (5m by default). */ -static int -__xfs_inode_free_quota_eofblocks( - struct xfs_inode *ip, - int (*execute)(struct xfs_mount *mp, - struct xfs_eofblocks *eofb)) -{ - int scan = 0; - struct xfs_eofblocks eofb = {0}; - struct xfs_dquot *dq; - - /* - * Run a sync scan to increase effectiveness and use the union filter to - * cover all applicable quotas in a single scan. - */ - eofb.eof_flags = XFS_EOF_FLAGS_UNION|XFS_EOF_FLAGS_SYNC; - - if (XFS_IS_UQUOTA_ENFORCED(ip->i_mount)) { - dq = xfs_inode_dquot(ip, XFS_DQTYPE_USER); - if (dq && xfs_dquot_lowsp(dq)) { - eofb.eof_uid = VFS_I(ip)->i_uid; - eofb.eof_flags |= XFS_EOF_FLAGS_UID; - scan = 1; - } - } - - if (XFS_IS_GQUOTA_ENFORCED(ip->i_mount)) { - dq = xfs_inode_dquot(ip, XFS_DQTYPE_GROUP); - if (dq && xfs_dquot_lowsp(dq)) { - eofb.eof_gid = VFS_I(ip)->i_gid; - eofb.eof_flags |= XFS_EOF_FLAGS_GID; - scan = 1; - } - } - - if (scan) - execute(ip->i_mount, &eofb); - - return scan; -} - -int -xfs_inode_free_quota_eofblocks( - struct xfs_inode *ip) -{ - return __xfs_inode_free_quota_eofblocks(ip, xfs_icache_free_eofblocks); -} - -static inline unsigned long -xfs_iflag_for_tag( - int tag) +static inline void +xfs_blockgc_queue( + struct xfs_perag *pag) { - switch (tag) { - case XFS_ICI_EOFBLOCKS_TAG: - return XFS_IEOFBLOCKS; - case XFS_ICI_COWBLOCKS_TAG: - return XFS_ICOWBLOCKS; - default: - ASSERT(0); - return 0; - } + rcu_read_lock(); + if (radix_tree_tagged(&pag->pag_ici_root, XFS_ICI_BLOCKGC_TAG)) + queue_delayed_work(pag->pag_mount->m_blockgc_workqueue, + &pag->pag_blockgc_work, + msecs_to_jiffies(xfs_blockgc_secs * 1000)); + rcu_read_unlock(); } static void -__xfs_inode_set_blocks_tag( - xfs_inode_t *ip, - void (*execute)(struct xfs_mount *mp), - void (*set_tp)(struct xfs_mount *mp, xfs_agnumber_t agno, - int error, unsigned long caller_ip), - int tag) +xfs_blockgc_set_iflag( + struct xfs_inode *ip, + unsigned long iflag) { - struct xfs_mount *mp = ip->i_mount; - struct xfs_perag *pag; - int tagged; + struct xfs_mount *mp = ip->i_mount; + struct xfs_perag *pag; + int tagged; + + ASSERT((iflag & ~(XFS_IEOFBLOCKS | XFS_ICOWBLOCKS)) == 0); /* * Don't bother locking the AG and looking up in the radix trees * if we already know that we have the tag set. */ - if (ip->i_flags & xfs_iflag_for_tag(tag)) + if (ip->i_flags & iflag) return; spin_lock(&ip->i_flags_lock); - ip->i_flags |= xfs_iflag_for_tag(tag); + ip->i_flags |= iflag; spin_unlock(&ip->i_flags_lock); pag = xfs_perag_get(mp, XFS_INO_TO_AGNO(mp, ip->i_ino)); spin_lock(&pag->pag_ici_lock); - tagged = radix_tree_tagged(&pag->pag_ici_root, tag); + tagged = radix_tree_tagged(&pag->pag_ici_root, XFS_ICI_BLOCKGC_TAG); radix_tree_tag_set(&pag->pag_ici_root, - XFS_INO_TO_AGINO(ip->i_mount, ip->i_ino), tag); + XFS_INO_TO_AGINO(ip->i_mount, ip->i_ino), + XFS_ICI_BLOCKGC_TAG); if (!tagged) { - /* propagate the eofblocks tag up into the perag radix tree */ + /* propagate the blockgc tag up into the perag radix tree */ spin_lock(&ip->i_mount->m_perag_lock); radix_tree_tag_set(&ip->i_mount->m_perag_tree, XFS_INO_TO_AGNO(ip->i_mount, ip->i_ino), - tag); + XFS_ICI_BLOCKGC_TAG); spin_unlock(&ip->i_mount->m_perag_lock); /* kick off background trimming */ - execute(ip->i_mount); + xfs_blockgc_queue(pag); - set_tp(ip->i_mount, pag->pag_agno, -1, _RET_IP_); + trace_xfs_perag_set_blockgc(ip->i_mount, pag->pag_agno, -1, + _RET_IP_); } spin_unlock(&pag->pag_ici_lock); @@ -1515,38 +1393,43 @@ xfs_inode_set_eofblocks_tag( xfs_inode_t *ip) { trace_xfs_inode_set_eofblocks_tag(ip); - return __xfs_inode_set_blocks_tag(ip, xfs_queue_eofblocks, - trace_xfs_perag_set_eofblocks, - XFS_ICI_EOFBLOCKS_TAG); + return xfs_blockgc_set_iflag(ip, XFS_IEOFBLOCKS); } static void -__xfs_inode_clear_blocks_tag( - xfs_inode_t *ip, - void (*clear_tp)(struct xfs_mount *mp, xfs_agnumber_t agno, - int error, unsigned long caller_ip), - int tag) +xfs_blockgc_clear_iflag( + struct xfs_inode *ip, + unsigned long iflag) { - struct xfs_mount *mp = ip->i_mount; - struct xfs_perag *pag; + struct xfs_mount *mp = ip->i_mount; + struct xfs_perag *pag; + bool clear_tag; + + ASSERT((iflag & ~(XFS_IEOFBLOCKS | XFS_ICOWBLOCKS)) == 0); spin_lock(&ip->i_flags_lock); - ip->i_flags &= ~xfs_iflag_for_tag(tag); + ip->i_flags &= ~iflag; + clear_tag = (ip->i_flags & (XFS_IEOFBLOCKS | XFS_ICOWBLOCKS)) == 0; spin_unlock(&ip->i_flags_lock); + if (!clear_tag) + return; + pag = xfs_perag_get(mp, XFS_INO_TO_AGNO(mp, ip->i_ino)); spin_lock(&pag->pag_ici_lock); radix_tree_tag_clear(&pag->pag_ici_root, - XFS_INO_TO_AGINO(ip->i_mount, ip->i_ino), tag); - if (!radix_tree_tagged(&pag->pag_ici_root, tag)) { - /* clear the eofblocks tag from the perag radix tree */ + XFS_INO_TO_AGINO(ip->i_mount, ip->i_ino), + XFS_ICI_BLOCKGC_TAG); + if (!radix_tree_tagged(&pag->pag_ici_root, XFS_ICI_BLOCKGC_TAG)) { + /* clear the blockgc tag from the perag radix tree */ spin_lock(&ip->i_mount->m_perag_lock); radix_tree_tag_clear(&ip->i_mount->m_perag_tree, XFS_INO_TO_AGNO(ip->i_mount, ip->i_ino), - tag); + XFS_ICI_BLOCKGC_TAG); spin_unlock(&ip->i_mount->m_perag_lock); - clear_tp(ip->i_mount, pag->pag_agno, -1, _RET_IP_); + trace_xfs_perag_clear_blockgc(ip->i_mount, pag->pag_agno, -1, + _RET_IP_); } spin_unlock(&pag->pag_ici_lock); @@ -1558,8 +1441,7 @@ xfs_inode_clear_eofblocks_tag( xfs_inode_t *ip) { trace_xfs_inode_clear_eofblocks_tag(ip); - return __xfs_inode_clear_blocks_tag(ip, - trace_xfs_perag_clear_eofblocks, XFS_ICI_EOFBLOCKS_TAG); + return xfs_blockgc_clear_iflag(ip, XFS_IEOFBLOCKS); } /* @@ -1609,20 +1491,42 @@ xfs_prep_free_cowblocks( STATIC int xfs_inode_free_cowblocks( struct xfs_inode *ip, - void *args) + void *args, + unsigned int *lockflags) { struct xfs_eofblocks *eofb = args; + bool wait; int ret = 0; + wait = eofb && (eofb->eof_flags & XFS_EOF_FLAGS_SYNC); + + if (!xfs_iflags_test(ip, XFS_ICOWBLOCKS)) + return 0; + if (!xfs_prep_free_cowblocks(ip)) return 0; if (!xfs_inode_matches_eofb(ip, eofb)) return 0; - /* Free the CoW blocks */ - xfs_ilock(ip, XFS_IOLOCK_EXCL); - xfs_ilock(ip, XFS_MMAPLOCK_EXCL); + /* + * If the caller is waiting, return -EAGAIN to keep the background + * scanner moving and revisit the inode in a subsequent pass. + */ + if (!(*lockflags & XFS_IOLOCK_EXCL) && + !xfs_ilock_nowait(ip, XFS_IOLOCK_EXCL)) { + if (wait) + return -EAGAIN; + return 0; + } + *lockflags |= XFS_IOLOCK_EXCL; + + if (!xfs_ilock_nowait(ip, XFS_MMAPLOCK_EXCL)) { + if (wait) + return -EAGAIN; + return 0; + } + *lockflags |= XFS_MMAPLOCK_EXCL; /* * Check again, nobody else should be able to dirty blocks or change @@ -1630,37 +1534,15 @@ xfs_inode_free_cowblocks( */ if (xfs_prep_free_cowblocks(ip)) ret = xfs_reflink_cancel_cow_range(ip, 0, NULLFILEOFF, false); - - xfs_iunlock(ip, XFS_MMAPLOCK_EXCL); - xfs_iunlock(ip, XFS_IOLOCK_EXCL); - return ret; } -int -xfs_icache_free_cowblocks( - struct xfs_mount *mp, - struct xfs_eofblocks *eofb) -{ - return xfs_inode_walk(mp, 0, xfs_inode_free_cowblocks, eofb, - XFS_ICI_COWBLOCKS_TAG); -} - -int -xfs_inode_free_quota_cowblocks( - struct xfs_inode *ip) -{ - return __xfs_inode_free_quota_eofblocks(ip, xfs_icache_free_cowblocks); -} - void xfs_inode_set_cowblocks_tag( xfs_inode_t *ip) { trace_xfs_inode_set_cowblocks_tag(ip); - return __xfs_inode_set_blocks_tag(ip, xfs_queue_cowblocks, - trace_xfs_perag_set_cowblocks, - XFS_ICI_COWBLOCKS_TAG); + return xfs_blockgc_set_iflag(ip, XFS_ICOWBLOCKS); } void @@ -1668,24 +1550,158 @@ xfs_inode_clear_cowblocks_tag( xfs_inode_t *ip) { trace_xfs_inode_clear_cowblocks_tag(ip); - return __xfs_inode_clear_blocks_tag(ip, - trace_xfs_perag_clear_cowblocks, XFS_ICI_COWBLOCKS_TAG); + return xfs_blockgc_clear_iflag(ip, XFS_ICOWBLOCKS); } +#define for_each_perag_tag(mp, next_agno, pag, tag) \ + for ((next_agno) = 0, (pag) = xfs_perag_get_tag((mp), 0, (tag)); \ + (pag) != NULL; \ + (next_agno) = (pag)->pag_agno + 1, \ + xfs_perag_put(pag), \ + (pag) = xfs_perag_get_tag((mp), (next_agno), (tag))) + + /* Disable post-EOF and CoW block auto-reclamation. */ void -xfs_stop_block_reaping( +xfs_blockgc_stop( struct xfs_mount *mp) { - cancel_delayed_work_sync(&mp->m_eofblocks_work); - cancel_delayed_work_sync(&mp->m_cowblocks_work); + struct xfs_perag *pag; + xfs_agnumber_t agno; + + for_each_perag_tag(mp, agno, pag, XFS_ICI_BLOCKGC_TAG) + cancel_delayed_work_sync(&pag->pag_blockgc_work); } /* Enable post-EOF and CoW block auto-reclamation. */ void -xfs_start_block_reaping( +xfs_blockgc_start( struct xfs_mount *mp) { - xfs_queue_eofblocks(mp); - xfs_queue_cowblocks(mp); + struct xfs_perag *pag; + xfs_agnumber_t agno; + + for_each_perag_tag(mp, agno, pag, XFS_ICI_BLOCKGC_TAG) + xfs_blockgc_queue(pag); +} + +/* Scan one incore inode for block preallocations that we can remove. */ +static int +xfs_blockgc_scan_inode( + struct xfs_inode *ip, + void *args) +{ + unsigned int lockflags = 0; + int error; + + error = xfs_inode_free_eofblocks(ip, args, &lockflags); + if (error) + goto unlock; + + error = xfs_inode_free_cowblocks(ip, args, &lockflags); +unlock: + if (lockflags) + xfs_iunlock(ip, lockflags); + return error; +} + +/* Background worker that trims preallocated space. */ +void +xfs_blockgc_worker( + struct work_struct *work) +{ + struct xfs_perag *pag = container_of(to_delayed_work(work), + struct xfs_perag, pag_blockgc_work); + struct xfs_mount *mp = pag->pag_mount; + int error; + + if (!sb_start_write_trylock(mp->m_super)) + return; + error = xfs_inode_walk_ag(pag, 0, xfs_blockgc_scan_inode, NULL, + XFS_ICI_BLOCKGC_TAG); + if (error) + xfs_info(mp, "AG %u preallocation gc worker failed, err=%d", + pag->pag_agno, error); + sb_end_write(mp->m_super); + xfs_blockgc_queue(pag); +} + +/* + * Try to free space in the filesystem by purging eofblocks and cowblocks. + */ +int +xfs_blockgc_free_space( + struct xfs_mount *mp, + struct xfs_eofblocks *eofb) +{ + trace_xfs_blockgc_free_space(mp, eofb, _RET_IP_); + + return xfs_inode_walk(mp, 0, xfs_blockgc_scan_inode, eofb, + XFS_ICI_BLOCKGC_TAG); +} + +/* + * Run cow/eofblocks scans on the supplied dquots. We don't know exactly which + * quota caused an allocation failure, so we make a best effort by including + * each quota under low free space conditions (less than 1% free space) in the + * scan. + * + * Callers must not hold any inode's ILOCK. If requesting a synchronous scan + * (XFS_EOF_FLAGS_SYNC), the caller also must not hold any inode's IOLOCK or + * MMAPLOCK. + */ +int +xfs_blockgc_free_dquots( + struct xfs_mount *mp, + struct xfs_dquot *udqp, + struct xfs_dquot *gdqp, + struct xfs_dquot *pdqp, + unsigned int eof_flags) +{ + struct xfs_eofblocks eofb = {0}; + bool do_work = false; + + if (!udqp && !gdqp && !pdqp) + return 0; + + /* + * Run a scan to free blocks using the union filter to cover all + * applicable quotas in a single scan. + */ + eofb.eof_flags = XFS_EOF_FLAGS_UNION | eof_flags; + + if (XFS_IS_UQUOTA_ENFORCED(mp) && udqp && xfs_dquot_lowsp(udqp)) { + eofb.eof_uid = make_kuid(mp->m_super->s_user_ns, udqp->q_id); + eofb.eof_flags |= XFS_EOF_FLAGS_UID; + do_work = true; + } + + if (XFS_IS_UQUOTA_ENFORCED(mp) && gdqp && xfs_dquot_lowsp(gdqp)) { + eofb.eof_gid = make_kgid(mp->m_super->s_user_ns, gdqp->q_id); + eofb.eof_flags |= XFS_EOF_FLAGS_GID; + do_work = true; + } + + if (XFS_IS_PQUOTA_ENFORCED(mp) && pdqp && xfs_dquot_lowsp(pdqp)) { + eofb.eof_prid = pdqp->q_id; + eofb.eof_flags |= XFS_EOF_FLAGS_PRID; + do_work = true; + } + + if (!do_work) + return 0; + + return xfs_blockgc_free_space(mp, &eofb); +} + +/* Run cow/eofblocks scans on the quotas attached to the inode. */ +int +xfs_blockgc_free_quota( + struct xfs_inode *ip, + unsigned int eof_flags) +{ + return xfs_blockgc_free_dquots(ip->i_mount, + xfs_inode_dquot(ip, XFS_DQTYPE_USER), + xfs_inode_dquot(ip, XFS_DQTYPE_GROUP), + xfs_inode_dquot(ip, XFS_DQTYPE_PROJ), eof_flags); } diff --git a/fs/xfs/xfs_icache.h b/fs/xfs/xfs_icache.h index 3a4c8b382cd0..d1fddb152420 100644 --- a/fs/xfs/xfs_icache.h +++ b/fs/xfs/xfs_icache.h @@ -23,8 +23,8 @@ struct xfs_eofblocks { #define XFS_ICI_NO_TAG (-1) /* special flag for an untagged lookup in xfs_inode_walk */ #define XFS_ICI_RECLAIM_TAG 0 /* inode is to be reclaimed */ -#define XFS_ICI_EOFBLOCKS_TAG 1 /* inode has blocks beyond EOF */ -#define XFS_ICI_COWBLOCKS_TAG 2 /* inode can have cow blocks to gc */ +/* Inode has speculative preallocations (posteof or cow) to clean. */ +#define XFS_ICI_BLOCKGC_TAG 1 /* * Flags for xfs_iget() @@ -54,19 +54,19 @@ long xfs_reclaim_inodes_nr(struct xfs_mount *mp, int nr_to_scan); void xfs_inode_set_reclaim_tag(struct xfs_inode *ip); +int xfs_blockgc_free_dquots(struct xfs_mount *mp, struct xfs_dquot *udqp, + struct xfs_dquot *gdqp, struct xfs_dquot *pdqp, + unsigned int eof_flags); +int xfs_blockgc_free_quota(struct xfs_inode *ip, unsigned int eof_flags); +int xfs_blockgc_free_space(struct xfs_mount *mp, struct xfs_eofblocks *eofb); + void xfs_inode_set_eofblocks_tag(struct xfs_inode *ip); void xfs_inode_clear_eofblocks_tag(struct xfs_inode *ip); -int xfs_icache_free_eofblocks(struct xfs_mount *, struct xfs_eofblocks *); -int xfs_inode_free_quota_eofblocks(struct xfs_inode *ip); -void xfs_eofblocks_worker(struct work_struct *); -void xfs_queue_eofblocks(struct xfs_mount *); void xfs_inode_set_cowblocks_tag(struct xfs_inode *ip); void xfs_inode_clear_cowblocks_tag(struct xfs_inode *ip); -int xfs_icache_free_cowblocks(struct xfs_mount *, struct xfs_eofblocks *); -int xfs_inode_free_quota_cowblocks(struct xfs_inode *ip); -void xfs_cowblocks_worker(struct work_struct *); -void xfs_queue_cowblocks(struct xfs_mount *); + +void xfs_blockgc_worker(struct work_struct *work); int xfs_inode_walk(struct xfs_mount *mp, int iter_flags, int (*execute)(struct xfs_inode *ip, void *args), @@ -75,7 +75,7 @@ int xfs_inode_walk(struct xfs_mount *mp, int iter_flags, int xfs_icache_inode_is_allocated(struct xfs_mount *mp, struct xfs_trans *tp, xfs_ino_t ino, bool *inuse); -void xfs_stop_block_reaping(struct xfs_mount *mp); -void xfs_start_block_reaping(struct xfs_mount *mp); +void xfs_blockgc_stop(struct xfs_mount *mp); +void xfs_blockgc_start(struct xfs_mount *mp); #endif diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c index b7352bc4c815..f93370bd7b1e 100644 --- a/fs/xfs/xfs_inode.c +++ b/fs/xfs/xfs_inode.c @@ -766,6 +766,7 @@ xfs_inode_inherit_flags2( */ static int xfs_init_new_inode( + struct user_namespace *mnt_userns, struct xfs_trans *tp, struct xfs_inode *pip, xfs_ino_t ino, @@ -775,6 +776,7 @@ xfs_init_new_inode( prid_t prid, struct xfs_inode **ipp) { + struct inode *dir = pip ? VFS_I(pip) : NULL; struct xfs_mount *mp = tp->t_mountp; struct xfs_inode *ip; unsigned int flags; @@ -804,18 +806,17 @@ xfs_init_new_inode( ASSERT(ip != NULL); inode = VFS_I(ip); - inode->i_mode = mode; set_nlink(inode, nlink); - inode->i_uid = current_fsuid(); inode->i_rdev = rdev; ip->i_d.di_projid = prid; - if (pip && XFS_INHERIT_GID(pip)) { - inode->i_gid = VFS_I(pip)->i_gid; - if ((VFS_I(pip)->i_mode & S_ISGID) && S_ISDIR(mode)) - inode->i_mode |= S_ISGID; + if (dir && !(dir->i_mode & S_ISGID) && + (mp->m_flags & XFS_MOUNT_GRPID)) { + inode->i_uid = fsuid_into_mnt(mnt_userns); + inode->i_gid = dir->i_gid; + inode->i_mode = mode; } else { - inode->i_gid = current_fsgid(); + inode_init_owner(mnt_userns, inode, dir, mode); } /* @@ -824,7 +825,8 @@ xfs_init_new_inode( * (and only if the irix_sgid_inherit compatibility variable is set). */ if (irix_sgid_inherit && - (inode->i_mode & S_ISGID) && !in_group_p(inode->i_gid)) + (inode->i_mode & S_ISGID) && + !in_group_p(i_gid_into_mnt(mnt_userns, inode))) inode->i_mode &= ~S_ISGID; ip->i_d.di_size = 0; @@ -901,6 +903,7 @@ xfs_init_new_inode( */ int xfs_dir_ialloc( + struct user_namespace *mnt_userns, struct xfs_trans **tpp, struct xfs_inode *dp, umode_t mode, @@ -933,7 +936,8 @@ xfs_dir_ialloc( return error; ASSERT(ino != NULLFSINO); - return xfs_init_new_inode(*tpp, dp, ino, mode, nlink, rdev, prid, ipp); + return xfs_init_new_inode(mnt_userns, *tpp, dp, ino, mode, nlink, rdev, + prid, ipp); } /* @@ -973,6 +977,7 @@ xfs_bumplink( int xfs_create( + struct user_namespace *mnt_userns, xfs_inode_t *dp, struct xfs_name *name, umode_t mode, @@ -1002,9 +1007,10 @@ xfs_create( /* * Make sure that we have allocated dquot(s) on disk. */ - error = xfs_qm_vop_dqalloc(dp, current_fsuid(), current_fsgid(), prid, - XFS_QMOPT_QUOTALL | XFS_QMOPT_INHERIT, - &udqp, &gdqp, &pdqp); + error = xfs_qm_vop_dqalloc(dp, fsuid_into_mnt(mnt_userns), + fsgid_into_mnt(mnt_userns), prid, + XFS_QMOPT_QUOTALL | XFS_QMOPT_INHERIT, + &udqp, &gdqp, &pdqp); if (error) return error; @@ -1022,23 +1028,22 @@ xfs_create( * the case we'll drop the one we have and get a more * appropriate transaction later. */ - error = xfs_trans_alloc(mp, tres, resblks, 0, 0, &tp); + error = xfs_trans_alloc_icreate(mp, tres, udqp, gdqp, pdqp, resblks, + &tp); if (error == -ENOSPC) { /* flush outstanding delalloc blocks and retry */ xfs_flush_inodes(mp); - error = xfs_trans_alloc(mp, tres, resblks, 0, 0, &tp); + error = xfs_trans_alloc_icreate(mp, tres, udqp, gdqp, pdqp, + resblks, &tp); } if (error) - goto out_release_inode; + goto out_release_dquots; xfs_ilock(dp, XFS_ILOCK_EXCL | XFS_ILOCK_PARENT); unlock_dp_on_error = true; - /* - * Reserve disk quota and the inode. - */ - error = xfs_trans_reserve_quota(tp, mp, udqp, gdqp, - pdqp, resblks, 1, 0); + error = xfs_iext_count_may_overflow(dp, XFS_DATA_FORK, + XFS_IEXT_DIR_MANIP_CNT(mp)); if (error) goto out_trans_cancel; @@ -1047,7 +1052,8 @@ xfs_create( * entry pointing to them, but a directory also the "." entry * pointing to itself. */ - error = xfs_dir_ialloc(&tp, dp, mode, is_dir ? 2 : 1, rdev, prid, &ip); + error = xfs_dir_ialloc(mnt_userns, &tp, dp, mode, is_dir ? 2 : 1, rdev, + prid, &ip); if (error) goto out_trans_cancel; @@ -1116,7 +1122,7 @@ xfs_create( xfs_finish_inode_setup(ip); xfs_irele(ip); } - + out_release_dquots: xfs_qm_dqrele(udqp); xfs_qm_dqrele(gdqp); xfs_qm_dqrele(pdqp); @@ -1128,6 +1134,7 @@ xfs_create( int xfs_create_tmpfile( + struct user_namespace *mnt_userns, struct xfs_inode *dp, umode_t mode, struct xfs_inode **ipp) @@ -1151,25 +1158,22 @@ xfs_create_tmpfile( /* * Make sure that we have allocated dquot(s) on disk. */ - error = xfs_qm_vop_dqalloc(dp, current_fsuid(), current_fsgid(), prid, - XFS_QMOPT_QUOTALL | XFS_QMOPT_INHERIT, - &udqp, &gdqp, &pdqp); + error = xfs_qm_vop_dqalloc(dp, fsuid_into_mnt(mnt_userns), + fsgid_into_mnt(mnt_userns), prid, + XFS_QMOPT_QUOTALL | XFS_QMOPT_INHERIT, + &udqp, &gdqp, &pdqp); if (error) return error; resblks = XFS_IALLOC_SPACE_RES(mp); tres = &M_RES(mp)->tr_create_tmpfile; - error = xfs_trans_alloc(mp, tres, resblks, 0, 0, &tp); - if (error) - goto out_release_inode; - - error = xfs_trans_reserve_quota(tp, mp, udqp, gdqp, - pdqp, resblks, 1, 0); + error = xfs_trans_alloc_icreate(mp, tres, udqp, gdqp, pdqp, resblks, + &tp); if (error) - goto out_trans_cancel; + goto out_release_dquots; - error = xfs_dir_ialloc(&tp, dp, mode, 0, 0, prid, &ip); + error = xfs_dir_ialloc(mnt_userns, &tp, dp, mode, 0, 0, prid, &ip); if (error) goto out_trans_cancel; @@ -1210,7 +1214,7 @@ xfs_create_tmpfile( xfs_finish_inode_setup(ip); xfs_irele(ip); } - + out_release_dquots: xfs_qm_dqrele(udqp); xfs_qm_dqrele(gdqp); xfs_qm_dqrele(pdqp); @@ -1258,6 +1262,11 @@ xfs_link( xfs_trans_ijoin(tp, sip, XFS_ILOCK_EXCL); xfs_trans_ijoin(tp, tdp, XFS_ILOCK_EXCL); + error = xfs_iext_count_may_overflow(tdp, XFS_DATA_FORK, + XFS_IEXT_DIR_MANIP_CNT(mp)); + if (error) + goto error_return; + /* * If we are using project inheritance, we only allow hard link * creation in our tree when the project IDs are the same; else @@ -2977,13 +2986,15 @@ out_trans_abort: */ static int xfs_rename_alloc_whiteout( + struct user_namespace *mnt_userns, struct xfs_inode *dp, struct xfs_inode **wip) { struct xfs_inode *tmpfile; int error; - error = xfs_create_tmpfile(dp, S_IFCHR | WHITEOUT_MODE, &tmpfile); + error = xfs_create_tmpfile(mnt_userns, dp, S_IFCHR | WHITEOUT_MODE, + &tmpfile); if (error) return error; @@ -3005,6 +3016,7 @@ xfs_rename_alloc_whiteout( */ int xfs_rename( + struct user_namespace *mnt_userns, struct xfs_inode *src_dp, struct xfs_name *src_name, struct xfs_inode *src_ip, @@ -3017,7 +3029,7 @@ xfs_rename( struct xfs_trans *tp; struct xfs_inode *wip = NULL; /* whiteout inode */ struct xfs_inode *inodes[__XFS_SORT_INODES]; - struct xfs_buf *agibp; + int i; int num_inodes = __XFS_SORT_INODES; bool new_parent = (src_dp != target_dp); bool src_is_directory = S_ISDIR(VFS_I(src_ip)->i_mode); @@ -3036,7 +3048,7 @@ xfs_rename( */ if (flags & RENAME_WHITEOUT) { ASSERT(!(flags & (RENAME_NOREPLACE | RENAME_EXCHANGE))); - error = xfs_rename_alloc_whiteout(target_dp, &wip); + error = xfs_rename_alloc_whiteout(mnt_userns, target_dp, &wip); if (error) return error; @@ -3106,6 +3118,35 @@ xfs_rename( /* * Check for expected errors before we dirty the transaction * so we can return an error without a transaction abort. + * + * Extent count overflow check: + * + * From the perspective of src_dp, a rename operation is essentially a + * directory entry remove operation. Hence the only place where we check + * for extent count overflow for src_dp is in + * xfs_bmap_del_extent_real(). xfs_bmap_del_extent_real() returns + * -ENOSPC when it detects a possible extent count overflow and in + * response, the higher layers of directory handling code do the + * following: + * 1. Data/Free blocks: XFS lets these blocks linger until a + * future remove operation removes them. + * 2. Dabtree blocks: XFS swaps the blocks with the last block in the + * Leaf space and unmaps the last block. + * + * For target_dp, there are two cases depending on whether the + * destination directory entry exists or not. + * + * When destination directory entry does not exist (i.e. target_ip == + * NULL), extent count overflow check is performed only when transaction + * has a non-zero sized space reservation associated with it. With a + * zero-sized space reservation, XFS allows a rename operation to + * continue only when the directory has sufficient free space in its + * data/leaf/free space blocks to hold the new entry. + * + * When destination directory entry exists (i.e. target_ip != NULL), all + * we need to do is change the inode number associated with the already + * existing entry. Hence there is no need to perform an extent count + * overflow check. */ if (target_ip == NULL) { /* @@ -3116,6 +3157,12 @@ xfs_rename( error = xfs_dir_canenter(tp, target_dp, target_name); if (error) goto out_trans_cancel; + } else { + error = xfs_iext_count_may_overflow(target_dp, + XFS_DATA_FORK, + XFS_IEXT_DIR_MANIP_CNT(mp)); + if (error) + goto out_trans_cancel; } } else { /* @@ -3131,6 +3178,30 @@ xfs_rename( } /* + * Lock the AGI buffers we need to handle bumping the nlink of the + * whiteout inode off the unlinked list and to handle dropping the + * nlink of the target inode. Per locking order rules, do this in + * increasing AG order and before directory block allocation tries to + * grab AGFs because we grab AGIs before AGFs. + * + * The (vfs) caller must ensure that if src is a directory then + * target_ip is either null or an empty directory. + */ + for (i = 0; i < num_inodes && inodes[i] != NULL; i++) { + if (inodes[i] == wip || + (inodes[i] == target_ip && + (VFS_I(target_ip)->i_nlink == 1 || src_is_directory))) { + struct xfs_buf *bp; + xfs_agnumber_t agno; + + agno = XFS_INO_TO_AGNO(mp, inodes[i]->i_ino); + error = xfs_read_agi(mp, tp, agno, &bp); + if (error) + goto out_trans_cancel; + } + } + + /* * Directory entry creation below may acquire the AGF. Remove * the whiteout from the unlinked list first to preserve correct * AGI/AGF locking order. This dirties the transaction so failures @@ -3182,22 +3253,6 @@ xfs_rename( * In case there is already an entry with the same * name at the destination directory, remove it first. */ - - /* - * Check whether the replace operation will need to allocate - * blocks. This happens when the shortform directory lacks - * space and we have to convert it to a block format directory. - * When more blocks are necessary, we must lock the AGI first - * to preserve locking order (AGI -> AGF). - */ - if (xfs_dir2_sf_replace_needblock(target_dp, src_ip->i_ino)) { - error = xfs_read_agi(mp, tp, - XFS_INO_TO_AGNO(mp, target_ip->i_ino), - &agibp); - if (error) - goto out_trans_cancel; - } - error = xfs_dir_replace(tp, target_dp, target_name, src_ip->i_ino, spaceres); if (error) @@ -3273,9 +3328,16 @@ xfs_rename( if (wip) { error = xfs_dir_replace(tp, src_dp, src_name, wip->i_ino, spaceres); - } else + } else { + /* + * NOTE: We don't need to check for extent count overflow here + * because the dir remove name code will leave the dir block in + * place if the extent count would overflow. + */ error = xfs_dir_removename(tp, src_dp, src_name, src_ip->i_ino, spaceres); + } + if (error) goto out_trans_cancel; diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h index eca333f5f715..88ee4c3930ae 100644 --- a/fs/xfs/xfs_inode.h +++ b/fs/xfs/xfs_inode.h @@ -369,15 +369,18 @@ int xfs_release(struct xfs_inode *ip); void xfs_inactive(struct xfs_inode *ip); int xfs_lookup(struct xfs_inode *dp, struct xfs_name *name, struct xfs_inode **ipp, struct xfs_name *ci_name); -int xfs_create(struct xfs_inode *dp, struct xfs_name *name, +int xfs_create(struct user_namespace *mnt_userns, + struct xfs_inode *dp, struct xfs_name *name, umode_t mode, dev_t rdev, struct xfs_inode **ipp); -int xfs_create_tmpfile(struct xfs_inode *dp, umode_t mode, +int xfs_create_tmpfile(struct user_namespace *mnt_userns, + struct xfs_inode *dp, umode_t mode, struct xfs_inode **ipp); int xfs_remove(struct xfs_inode *dp, struct xfs_name *name, struct xfs_inode *ip); int xfs_link(struct xfs_inode *tdp, struct xfs_inode *sip, struct xfs_name *target_name); -int xfs_rename(struct xfs_inode *src_dp, struct xfs_name *src_name, +int xfs_rename(struct user_namespace *mnt_userns, + struct xfs_inode *src_dp, struct xfs_name *src_name, struct xfs_inode *src_ip, struct xfs_inode *target_dp, struct xfs_name *target_name, struct xfs_inode *target_ip, unsigned int flags); @@ -407,9 +410,10 @@ void xfs_lock_two_inodes(struct xfs_inode *ip0, uint ip0_mode, xfs_extlen_t xfs_get_extsz_hint(struct xfs_inode *ip); xfs_extlen_t xfs_get_cowextsz_hint(struct xfs_inode *ip); -int xfs_dir_ialloc(struct xfs_trans **tpp, struct xfs_inode *dp, umode_t mode, - xfs_nlink_t nlink, dev_t dev, prid_t prid, - struct xfs_inode **ipp); +int xfs_dir_ialloc(struct user_namespace *mnt_userns, + struct xfs_trans **tpp, struct xfs_inode *dp, + umode_t mode, xfs_nlink_t nlink, dev_t dev, + prid_t prid, struct xfs_inode **ipp); static inline int xfs_itruncate_extents( diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c index 3fbd98f61ea5..99dfe89a8d08 100644 --- a/fs/xfs/xfs_ioctl.c +++ b/fs/xfs/xfs_ioctl.c @@ -693,7 +693,8 @@ xfs_ioc_space( iattr.ia_valid = ATTR_SIZE; iattr.ia_size = bf->l_start; - error = xfs_vn_setattr_size(file_dentry(filp), &iattr); + error = xfs_vn_setattr_size(file_mnt_user_ns(filp), file_dentry(filp), + &iattr); if (error) goto out_unlock; @@ -734,13 +735,15 @@ xfs_fsinumbers_fmt( STATIC int xfs_ioc_fsbulkstat( - xfs_mount_t *mp, + struct file *file, unsigned int cmd, void __user *arg) { + struct xfs_mount *mp = XFS_I(file_inode(file))->i_mount; struct xfs_fsop_bulkreq bulkreq; struct xfs_ibulk breq = { .mp = mp, + .mnt_userns = file_mnt_user_ns(file), .ocount = 0, }; xfs_ino_t lastino; @@ -908,13 +911,15 @@ xfs_bulk_ireq_teardown( /* Handle the v5 bulkstat ioctl. */ STATIC int xfs_ioc_bulkstat( - struct xfs_mount *mp, + struct file *file, unsigned int cmd, struct xfs_bulkstat_req __user *arg) { + struct xfs_mount *mp = XFS_I(file_inode(file))->i_mount; struct xfs_bulk_ireq hdr; struct xfs_ibulk breq = { .mp = mp, + .mnt_userns = file_mnt_user_ns(file), }; int error; @@ -1275,24 +1280,24 @@ xfs_ioctl_setattr_prepare_dax( */ static struct xfs_trans * xfs_ioctl_setattr_get_trans( - struct xfs_inode *ip) + struct file *file, + struct xfs_dquot *pdqp) { + struct xfs_inode *ip = XFS_I(file_inode(file)); struct xfs_mount *mp = ip->i_mount; struct xfs_trans *tp; int error = -EROFS; if (mp->m_flags & XFS_MOUNT_RDONLY) - goto out_unlock; + goto out_error; error = -EIO; if (XFS_FORCED_SHUTDOWN(mp)) - goto out_unlock; + goto out_error; - error = xfs_trans_alloc(mp, &M_RES(mp)->tr_ichange, 0, 0, 0, &tp); + error = xfs_trans_alloc_ichange(ip, NULL, NULL, pdqp, + capable(CAP_FOWNER), &tp); if (error) - goto out_unlock; - - xfs_ilock(ip, XFS_ILOCK_EXCL); - xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL); + goto out_error; /* * CAP_FOWNER overrides the following restrictions: @@ -1300,7 +1305,7 @@ xfs_ioctl_setattr_get_trans( * The user ID of the calling process must be equal to the file owner * ID, except in cases where the CAP_FSETID capability is applicable. */ - if (!inode_owner_or_capable(VFS_I(ip))) { + if (!inode_owner_or_capable(file_mnt_user_ns(file), VFS_I(ip))) { error = -EPERM; goto out_cancel; } @@ -1312,7 +1317,7 @@ xfs_ioctl_setattr_get_trans( out_cancel: xfs_trans_cancel(tp); -out_unlock: +out_error: return ERR_PTR(error); } @@ -1428,21 +1433,23 @@ xfs_ioctl_setattr_check_projid( STATIC int xfs_ioctl_setattr( - xfs_inode_t *ip, + struct file *file, struct fsxattr *fa) { + struct user_namespace *mnt_userns = file_mnt_user_ns(file); + struct xfs_inode *ip = XFS_I(file_inode(file)); struct fsxattr old_fa; struct xfs_mount *mp = ip->i_mount; struct xfs_trans *tp; struct xfs_dquot *pdqp = NULL; struct xfs_dquot *olddquot = NULL; - int code; + int error; trace_xfs_ioctl_setattr(ip); - code = xfs_ioctl_setattr_check_projid(ip, fa); - if (code) - return code; + error = xfs_ioctl_setattr_check_projid(ip, fa); + if (error) + return error; /* * If disk quotas is on, we make sure that the dquots do exist on disk, @@ -1453,44 +1460,36 @@ xfs_ioctl_setattr( * because the i_*dquot fields will get updated anyway. */ if (XFS_IS_QUOTA_ON(mp)) { - code = xfs_qm_vop_dqalloc(ip, VFS_I(ip)->i_uid, + error = xfs_qm_vop_dqalloc(ip, VFS_I(ip)->i_uid, VFS_I(ip)->i_gid, fa->fsx_projid, XFS_QMOPT_PQUOTA, NULL, NULL, &pdqp); - if (code) - return code; + if (error) + return error; } xfs_ioctl_setattr_prepare_dax(ip, fa); - tp = xfs_ioctl_setattr_get_trans(ip); + tp = xfs_ioctl_setattr_get_trans(file, pdqp); if (IS_ERR(tp)) { - code = PTR_ERR(tp); + error = PTR_ERR(tp); goto error_free_dquots; } - if (XFS_IS_QUOTA_RUNNING(mp) && XFS_IS_PQUOTA_ON(mp) && - ip->i_d.di_projid != fa->fsx_projid) { - code = xfs_qm_vop_chown_reserve(tp, ip, NULL, NULL, pdqp, - capable(CAP_FOWNER) ? XFS_QMOPT_FORCE_RES : 0); - if (code) /* out of quota */ - goto error_trans_cancel; - } - xfs_fill_fsxattr(ip, false, &old_fa); - code = vfs_ioc_fssetxattr_check(VFS_I(ip), &old_fa, fa); - if (code) + error = vfs_ioc_fssetxattr_check(VFS_I(ip), &old_fa, fa); + if (error) goto error_trans_cancel; - code = xfs_ioctl_setattr_check_extsize(ip, fa); - if (code) + error = xfs_ioctl_setattr_check_extsize(ip, fa); + if (error) goto error_trans_cancel; - code = xfs_ioctl_setattr_check_cowextsize(ip, fa); - if (code) + error = xfs_ioctl_setattr_check_cowextsize(ip, fa); + if (error) goto error_trans_cancel; - code = xfs_ioctl_setattr_xflags(tp, ip, fa); - if (code) + error = xfs_ioctl_setattr_xflags(tp, ip, fa); + if (error) goto error_trans_cancel; /* @@ -1502,7 +1501,7 @@ xfs_ioctl_setattr( */ if ((VFS_I(ip)->i_mode & (S_ISUID|S_ISGID)) && - !capable_wrt_inode_uidgid(VFS_I(ip), CAP_FSETID)) + !capable_wrt_inode_uidgid(mnt_userns, VFS_I(ip), CAP_FSETID)) VFS_I(ip)->i_mode &= ~(S_ISUID|S_ISGID); /* Change the ownerships and register project quota modifications */ @@ -1530,7 +1529,7 @@ xfs_ioctl_setattr( else ip->i_d.di_cowextsize = 0; - code = xfs_trans_commit(tp); + error = xfs_trans_commit(tp); /* * Release any dquot(s) the inode had kept before chown. @@ -1538,18 +1537,17 @@ xfs_ioctl_setattr( xfs_qm_dqrele(olddquot); xfs_qm_dqrele(pdqp); - return code; + return error; error_trans_cancel: xfs_trans_cancel(tp); error_free_dquots: xfs_qm_dqrele(pdqp); - return code; + return error; } STATIC int xfs_ioc_fssetxattr( - xfs_inode_t *ip, struct file *filp, void __user *arg) { @@ -1562,7 +1560,7 @@ xfs_ioc_fssetxattr( error = mnt_want_write_file(filp); if (error) return error; - error = xfs_ioctl_setattr(ip, &fa); + error = xfs_ioctl_setattr(filp, &fa); mnt_drop_write_file(filp); return error; } @@ -1608,7 +1606,7 @@ xfs_ioc_setxflags( xfs_ioctl_setattr_prepare_dax(ip, &fa); - tp = xfs_ioctl_setattr_get_trans(ip); + tp = xfs_ioctl_setattr_get_trans(filp, NULL); if (IS_ERR(tp)) { error = PTR_ERR(tp); goto out_drop_write; @@ -2119,10 +2117,10 @@ xfs_file_ioctl( case XFS_IOC_FSBULKSTAT_SINGLE: case XFS_IOC_FSBULKSTAT: case XFS_IOC_FSINUMBERS: - return xfs_ioc_fsbulkstat(mp, cmd, arg); + return xfs_ioc_fsbulkstat(filp, cmd, arg); case XFS_IOC_BULKSTAT: - return xfs_ioc_bulkstat(mp, cmd, arg); + return xfs_ioc_bulkstat(filp, cmd, arg); case XFS_IOC_INUMBERS: return xfs_ioc_inumbers(mp, cmd, arg); @@ -2144,7 +2142,7 @@ xfs_file_ioctl( case XFS_IOC_FSGETXATTRA: return xfs_ioc_fsgetxattr(ip, 1, arg); case XFS_IOC_FSSETXATTR: - return xfs_ioc_fssetxattr(ip, filp, arg); + return xfs_ioc_fssetxattr(filp, arg); case XFS_IOC_GETXFLAGS: return xfs_ioc_getxflags(ip, arg); case XFS_IOC_SETXFLAGS: @@ -2260,7 +2258,7 @@ xfs_file_ioctl( } case XFS_IOC_FSGROWFSDATA: { - xfs_growfs_data_t in; + struct xfs_growfs_data in; if (copy_from_user(&in, arg, sizeof(in))) return -EFAULT; @@ -2274,7 +2272,7 @@ xfs_file_ioctl( } case XFS_IOC_FSGROWFSLOG: { - xfs_growfs_log_t in; + struct xfs_growfs_log in; if (copy_from_user(&in, arg, sizeof(in))) return -EFAULT; @@ -2348,8 +2346,10 @@ xfs_file_ioctl( if (error) return error; + trace_xfs_ioc_free_eofblocks(mp, &keofb, _RET_IP_); + sb_start_write(mp->m_super); - error = xfs_icache_free_eofblocks(mp, &keofb); + error = xfs_blockgc_free_space(mp, &keofb); sb_end_write(mp->m_super); return error; } diff --git a/fs/xfs/xfs_ioctl32.c b/fs/xfs/xfs_ioctl32.c index c1771e728117..33c09ec8e6c0 100644 --- a/fs/xfs/xfs_ioctl32.c +++ b/fs/xfs/xfs_ioctl32.c @@ -209,14 +209,16 @@ xfs_fsbulkstat_one_fmt_compat( /* copied from xfs_ioctl.c */ STATIC int xfs_compat_ioc_fsbulkstat( - xfs_mount_t *mp, + struct file *file, unsigned int cmd, struct compat_xfs_fsop_bulkreq __user *p32) { + struct xfs_mount *mp = XFS_I(file_inode(file))->i_mount; u32 addr; struct xfs_fsop_bulkreq bulkreq; struct xfs_ibulk breq = { .mp = mp, + .mnt_userns = file_mnt_user_ns(file), .ocount = 0, }; xfs_ino_t lastino; @@ -436,7 +438,6 @@ xfs_file_compat_ioctl( { struct inode *inode = file_inode(filp); struct xfs_inode *ip = XFS_I(inode); - struct xfs_mount *mp = ip->i_mount; void __user *arg = compat_ptr(p); int error; @@ -456,7 +457,7 @@ xfs_file_compat_ioctl( return xfs_ioc_space(filp, &bf); } case XFS_IOC_FSGEOMETRY_V1_32: - return xfs_compat_ioc_fsgeometry_v1(mp, arg); + return xfs_compat_ioc_fsgeometry_v1(ip->i_mount, arg); case XFS_IOC_FSGROWFSDATA_32: { struct xfs_growfs_data in; @@ -465,7 +466,7 @@ xfs_file_compat_ioctl( error = mnt_want_write_file(filp); if (error) return error; - error = xfs_growfs_data(mp, &in); + error = xfs_growfs_data(ip->i_mount, &in); mnt_drop_write_file(filp); return error; } @@ -477,7 +478,7 @@ xfs_file_compat_ioctl( error = mnt_want_write_file(filp); if (error) return error; - error = xfs_growfs_rt(mp, &in); + error = xfs_growfs_rt(ip->i_mount, &in); mnt_drop_write_file(filp); return error; } @@ -507,7 +508,7 @@ xfs_file_compat_ioctl( case XFS_IOC_FSBULKSTAT_32: case XFS_IOC_FSBULKSTAT_SINGLE_32: case XFS_IOC_FSINUMBERS_32: - return xfs_compat_ioc_fsbulkstat(mp, cmd, arg); + return xfs_compat_ioc_fsbulkstat(filp, cmd, arg); case XFS_IOC_FD_TO_HANDLE_32: case XFS_IOC_PATH_TO_HANDLE_32: case XFS_IOC_PATH_TO_FSHANDLE_32: { diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c index 7b9ff824e82d..e17ab7f42928 100644 --- a/fs/xfs/xfs_iomap.c +++ b/fs/xfs/xfs_iomap.c @@ -194,25 +194,21 @@ xfs_iomap_write_direct( struct xfs_trans *tp; xfs_filblks_t resaligned; int nimaps; - int quota_flag; - uint qblocks, resblks; - unsigned int resrtextents = 0; + unsigned int dblocks, rblocks; + bool force = false; int error; int bmapi_flags = XFS_BMAPI_PREALLOC; - uint tflags = 0; ASSERT(count_fsb > 0); resaligned = xfs_aligned_fsb_count(offset_fsb, count_fsb, xfs_get_extsz_hint(ip)); if (unlikely(XFS_IS_REALTIME_INODE(ip))) { - resrtextents = qblocks = resaligned; - resrtextents /= mp->m_sb.sb_rextsize; - resblks = XFS_DIOSTRAT_SPACE_RES(mp, 0); - quota_flag = XFS_QMOPT_RES_RTBLKS; + dblocks = XFS_DIOSTRAT_SPACE_RES(mp, 0); + rblocks = resaligned; } else { - resblks = qblocks = XFS_DIOSTRAT_SPACE_RES(mp, resaligned); - quota_flag = XFS_QMOPT_RES_REGBLKS; + dblocks = XFS_DIOSTRAT_SPACE_RES(mp, resaligned); + rblocks = 0; } error = xfs_qm_dqattach(ip); @@ -235,23 +231,21 @@ xfs_iomap_write_direct( if (IS_DAX(VFS_I(ip))) { bmapi_flags = XFS_BMAPI_CONVERT | XFS_BMAPI_ZERO; if (imap->br_state == XFS_EXT_UNWRITTEN) { - tflags |= XFS_TRANS_RESERVE; - resblks = XFS_DIOSTRAT_SPACE_RES(mp, 0) << 1; + force = true; + dblocks = XFS_DIOSTRAT_SPACE_RES(mp, 0) << 1; } } - error = xfs_trans_alloc(mp, &M_RES(mp)->tr_write, resblks, resrtextents, - tflags, &tp); + + error = xfs_trans_alloc_inode(ip, &M_RES(mp)->tr_write, dblocks, + rblocks, force, &tp); if (error) return error; - xfs_ilock(ip, XFS_ILOCK_EXCL); - - error = xfs_trans_reserve_quota_nblks(tp, ip, qblocks, 0, quota_flag); + error = xfs_iext_count_may_overflow(ip, XFS_DATA_FORK, + XFS_IEXT_ADD_NOSPLIT_CNT); if (error) goto out_trans_cancel; - xfs_trans_ijoin(tp, ip, 0); - /* * From this point onwards we overwrite the imap pointer that the * caller gave to us. @@ -260,7 +254,7 @@ xfs_iomap_write_direct( error = xfs_bmapi_write(tp, ip, offset_fsb, count_fsb, bmapi_flags, 0, imap, &nimaps); if (error) - goto out_res_cancel; + goto out_trans_cancel; /* * Complete the transaction @@ -284,8 +278,6 @@ out_unlock: xfs_iunlock(ip, XFS_ILOCK_EXCL); return error; -out_res_cancel: - xfs_trans_unreserve_quota_nblks(tp, ip, (long)qblocks, 0, quota_flag); out_trans_cancel: xfs_trans_cancel(tp); goto out_unlock; @@ -548,16 +540,13 @@ xfs_iomap_write_unwritten( * here as we might be asked to write out the same inode that we * complete here and might deadlock on the iolock. */ - error = xfs_trans_alloc(mp, &M_RES(mp)->tr_write, resblks, 0, - XFS_TRANS_RESERVE, &tp); + error = xfs_trans_alloc_inode(ip, &M_RES(mp)->tr_write, resblks, + 0, true, &tp); if (error) return error; - xfs_ilock(ip, XFS_ILOCK_EXCL); - xfs_trans_ijoin(tp, ip, 0); - - error = xfs_trans_reserve_quota_nblks(tp, ip, resblks, 0, - XFS_QMOPT_RES_REGBLKS | XFS_QMOPT_FORCE_RES); + error = xfs_iext_count_may_overflow(ip, XFS_DATA_FORK, + XFS_IEXT_WRITE_UNWRITTEN_CNT); if (error) goto error_on_bmapi_transaction; @@ -784,15 +773,28 @@ xfs_direct_write_iomap_begin( goto allocate_blocks; /* - * NOWAIT IO needs to span the entire requested IO with a single map so - * that we avoid partial IO failures due to the rest of the IO range not - * covered by this map triggering an EAGAIN condition when it is - * subsequently mapped and aborting the IO. + * NOWAIT and OVERWRITE I/O needs to span the entire requested I/O with + * a single map so that we avoid partial IO failures due to the rest of + * the I/O range not covered by this map triggering an EAGAIN condition + * when it is subsequently mapped and aborting the I/O. */ - if ((flags & IOMAP_NOWAIT) && - !imap_spans_range(&imap, offset_fsb, end_fsb)) { + if (flags & (IOMAP_NOWAIT | IOMAP_OVERWRITE_ONLY)) { error = -EAGAIN; - goto out_unlock; + if (!imap_spans_range(&imap, offset_fsb, end_fsb)) + goto out_unlock; + } + + /* + * For overwrite only I/O, we cannot convert unwritten extents without + * requiring sub-block zeroing. This can only be done under an + * exclusive IOLOCK, hence return -EAGAIN if this is not a written + * extent to tell the caller to try again. + */ + if (flags & IOMAP_OVERWRITE_ONLY) { + error = -EAGAIN; + if (imap.br_state != XFS_EXT_NORM && + ((offset | length) & mp->m_blockmask)) + goto out_unlock; } xfs_iunlock(ip, lockmode); @@ -801,7 +803,7 @@ xfs_direct_write_iomap_begin( allocate_blocks: error = -EAGAIN; - if (flags & IOMAP_NOWAIT) + if (flags & (IOMAP_NOWAIT | IOMAP_OVERWRITE_ONLY)) goto out_unlock; /* @@ -842,7 +844,8 @@ out_found_cow: return xfs_bmbt_to_iomap(ip, iomap, &cmap, IOMAP_F_SHARED); out_unlock: - xfs_iunlock(ip, lockmode); + if (lockmode) + xfs_iunlock(ip, lockmode); return error; } @@ -870,6 +873,9 @@ xfs_buffered_write_iomap_begin( int allocfork = XFS_DATA_FORK; int error = 0; + if (XFS_FORCED_SHUTDOWN(mp)) + return -EIO; + /* we can't use delayed allocations when using extent size hints */ if (xfs_get_extsz_hint(ip)) return xfs_direct_write_iomap_begin(inode, offset, count, diff --git a/fs/xfs/xfs_iops.c b/fs/xfs/xfs_iops.c index 67c8dc9de8aa..66ebccb5a6ff 100644 --- a/fs/xfs/xfs_iops.c +++ b/fs/xfs/xfs_iops.c @@ -128,6 +128,7 @@ xfs_cleanup_inode( STATIC int xfs_generic_create( + struct user_namespace *mnt_userns, struct inode *dir, struct dentry *dentry, umode_t mode, @@ -161,9 +162,10 @@ xfs_generic_create( goto out_free_acl; if (!tmpfile) { - error = xfs_create(XFS_I(dir), &name, mode, rdev, &ip); + error = xfs_create(mnt_userns, XFS_I(dir), &name, mode, rdev, + &ip); } else { - error = xfs_create_tmpfile(XFS_I(dir), mode, &ip); + error = xfs_create_tmpfile(mnt_userns, XFS_I(dir), mode, &ip); } if (unlikely(error)) goto out_free_acl; @@ -220,31 +222,35 @@ xfs_generic_create( STATIC int xfs_vn_mknod( - struct inode *dir, - struct dentry *dentry, - umode_t mode, - dev_t rdev) + struct user_namespace *mnt_userns, + struct inode *dir, + struct dentry *dentry, + umode_t mode, + dev_t rdev) { - return xfs_generic_create(dir, dentry, mode, rdev, false); + return xfs_generic_create(mnt_userns, dir, dentry, mode, rdev, false); } STATIC int xfs_vn_create( - struct inode *dir, - struct dentry *dentry, - umode_t mode, - bool flags) + struct user_namespace *mnt_userns, + struct inode *dir, + struct dentry *dentry, + umode_t mode, + bool flags) { - return xfs_generic_create(dir, dentry, mode, 0, false); + return xfs_generic_create(mnt_userns, dir, dentry, mode, 0, false); } STATIC int xfs_vn_mkdir( - struct inode *dir, - struct dentry *dentry, - umode_t mode) + struct user_namespace *mnt_userns, + struct inode *dir, + struct dentry *dentry, + umode_t mode) { - return xfs_generic_create(dir, dentry, mode | S_IFDIR, 0, false); + return xfs_generic_create(mnt_userns, dir, dentry, mode | S_IFDIR, 0, + false); } STATIC struct dentry * @@ -361,9 +367,10 @@ xfs_vn_unlink( STATIC int xfs_vn_symlink( - struct inode *dir, - struct dentry *dentry, - const char *symname) + struct user_namespace *mnt_userns, + struct inode *dir, + struct dentry *dentry, + const char *symname) { struct inode *inode; struct xfs_inode *cip = NULL; @@ -377,7 +384,7 @@ xfs_vn_symlink( if (unlikely(error)) goto out; - error = xfs_symlink(XFS_I(dir), &name, symname, mode, &cip); + error = xfs_symlink(mnt_userns, XFS_I(dir), &name, symname, mode, &cip); if (unlikely(error)) goto out; @@ -403,11 +410,12 @@ xfs_vn_symlink( STATIC int xfs_vn_rename( - struct inode *odir, - struct dentry *odentry, - struct inode *ndir, - struct dentry *ndentry, - unsigned int flags) + struct user_namespace *mnt_userns, + struct inode *odir, + struct dentry *odentry, + struct inode *ndir, + struct dentry *ndentry, + unsigned int flags) { struct inode *new_inode = d_inode(ndentry); int omode = 0; @@ -431,8 +439,8 @@ xfs_vn_rename( if (unlikely(error)) return error; - return xfs_rename(XFS_I(odir), &oname, XFS_I(d_inode(odentry)), - XFS_I(ndir), &nname, + return xfs_rename(mnt_userns, XFS_I(odir), &oname, + XFS_I(d_inode(odentry)), XFS_I(ndir), &nname, new_inode ? XFS_I(new_inode) : NULL, flags); } @@ -529,6 +537,7 @@ xfs_stat_blksize( STATIC int xfs_vn_getattr( + struct user_namespace *mnt_userns, const struct path *path, struct kstat *stat, u32 request_mask, @@ -547,8 +556,8 @@ xfs_vn_getattr( stat->dev = inode->i_sb->s_dev; stat->mode = inode->i_mode; stat->nlink = inode->i_nlink; - stat->uid = inode->i_uid; - stat->gid = inode->i_gid; + stat->uid = i_uid_into_mnt(mnt_userns, inode); + stat->gid = i_gid_into_mnt(mnt_userns, inode); stat->ino = ip->i_ino; stat->atime = inode->i_atime; stat->mtime = inode->i_mtime; @@ -626,8 +635,9 @@ xfs_setattr_time( static int xfs_vn_change_ok( - struct dentry *dentry, - struct iattr *iattr) + struct user_namespace *mnt_userns, + struct dentry *dentry, + struct iattr *iattr) { struct xfs_mount *mp = XFS_I(d_inode(dentry))->i_mount; @@ -637,7 +647,7 @@ xfs_vn_change_ok( if (XFS_FORCED_SHUTDOWN(mp)) return -EIO; - return setattr_prepare(dentry, iattr); + return setattr_prepare(mnt_userns, dentry, iattr); } /* @@ -648,6 +658,7 @@ xfs_vn_change_ok( */ static int xfs_setattr_nonsize( + struct user_namespace *mnt_userns, struct xfs_inode *ip, struct iattr *iattr) { @@ -700,13 +711,11 @@ xfs_setattr_nonsize( return error; } - error = xfs_trans_alloc(mp, &M_RES(mp)->tr_ichange, 0, 0, 0, &tp); + error = xfs_trans_alloc_ichange(ip, udqp, gdqp, NULL, + capable(CAP_FOWNER), &tp); if (error) goto out_dqrele; - xfs_ilock(ip, XFS_ILOCK_EXCL); - xfs_trans_ijoin(tp, ip, 0); - /* * Change file ownership. Must be the owner or privileged. */ @@ -723,21 +732,6 @@ xfs_setattr_nonsize( uid = (mask & ATTR_UID) ? iattr->ia_uid : iuid; /* - * Do a quota reservation only if uid/gid is actually - * going to change. - */ - if (XFS_IS_QUOTA_RUNNING(mp) && - ((XFS_IS_UQUOTA_ON(mp) && !uid_eq(iuid, uid)) || - (XFS_IS_GQUOTA_ON(mp) && !gid_eq(igid, gid)))) { - ASSERT(tp); - error = xfs_qm_vop_chown_reserve(tp, ip, udqp, gdqp, - NULL, capable(CAP_FOWNER) ? - XFS_QMOPT_FORCE_RES : 0); - if (error) /* out of quota */ - goto out_cancel; - } - - /* * CAP_FSETID overrides the following restrictions: * * The set-user-ID and set-group-ID bits of a file will be @@ -786,8 +780,6 @@ xfs_setattr_nonsize( xfs_trans_set_sync(tp); error = xfs_trans_commit(tp); - xfs_iunlock(ip, XFS_ILOCK_EXCL); - /* * Release any dquot(s) the inode had kept before chown. */ @@ -807,16 +799,13 @@ xfs_setattr_nonsize( * Posix ACL code seems to care about this issue either. */ if (mask & ATTR_MODE) { - error = posix_acl_chmod(inode, inode->i_mode); + error = posix_acl_chmod(mnt_userns, inode, inode->i_mode); if (error) return error; } return 0; -out_cancel: - xfs_trans_cancel(tp); - xfs_iunlock(ip, XFS_ILOCK_EXCL); out_dqrele: xfs_qm_dqrele(udqp); xfs_qm_dqrele(gdqp); @@ -831,6 +820,7 @@ out_dqrele: */ STATIC int xfs_setattr_size( + struct user_namespace *mnt_userns, struct xfs_inode *ip, struct iattr *iattr) { @@ -846,7 +836,7 @@ xfs_setattr_size( ASSERT(xfs_isilocked(ip, XFS_MMAPLOCK_EXCL)); ASSERT(S_ISREG(inode->i_mode)); ASSERT((iattr->ia_valid & (ATTR_UID|ATTR_GID|ATTR_ATIME|ATTR_ATIME_SET| - ATTR_MTIME_SET|ATTR_KILL_PRIV|ATTR_TIMES_SET)) == 0); + ATTR_MTIME_SET|ATTR_TIMES_SET)) == 0); oldsize = inode->i_size; newsize = iattr->ia_size; @@ -862,7 +852,7 @@ xfs_setattr_size( * Use the regular setattr path to update the timestamps. */ iattr->ia_valid &= ~ATTR_SIZE; - return xfs_setattr_nonsize(ip, iattr); + return xfs_setattr_nonsize(mnt_userns, ip, iattr); } /* @@ -1031,6 +1021,7 @@ out_trans_cancel: int xfs_vn_setattr_size( + struct user_namespace *mnt_userns, struct dentry *dentry, struct iattr *iattr) { @@ -1039,14 +1030,15 @@ xfs_vn_setattr_size( trace_xfs_setattr(ip); - error = xfs_vn_change_ok(dentry, iattr); + error = xfs_vn_change_ok(mnt_userns, dentry, iattr); if (error) return error; - return xfs_setattr_size(ip, iattr); + return xfs_setattr_size(mnt_userns, ip, iattr); } STATIC int xfs_vn_setattr( + struct user_namespace *mnt_userns, struct dentry *dentry, struct iattr *iattr) { @@ -1066,14 +1058,14 @@ xfs_vn_setattr( return error; } - error = xfs_vn_setattr_size(dentry, iattr); + error = xfs_vn_setattr_size(mnt_userns, dentry, iattr); xfs_iunlock(ip, XFS_MMAPLOCK_EXCL); } else { trace_xfs_setattr(ip); - error = xfs_vn_change_ok(dentry, iattr); + error = xfs_vn_change_ok(mnt_userns, dentry, iattr); if (!error) - error = xfs_setattr_nonsize(ip, iattr); + error = xfs_setattr_nonsize(mnt_userns, ip, iattr); } return error; @@ -1144,11 +1136,12 @@ xfs_vn_fiemap( STATIC int xfs_vn_tmpfile( - struct inode *dir, - struct dentry *dentry, - umode_t mode) + struct user_namespace *mnt_userns, + struct inode *dir, + struct dentry *dentry, + umode_t mode) { - return xfs_generic_create(dir, dentry, mode, 0, true); + return xfs_generic_create(mnt_userns, dir, dentry, mode, 0, true); } static const struct inode_operations xfs_inode_operations = { diff --git a/fs/xfs/xfs_iops.h b/fs/xfs/xfs_iops.h index 99ca745c1071..278949056048 100644 --- a/fs/xfs/xfs_iops.h +++ b/fs/xfs/xfs_iops.h @@ -14,6 +14,7 @@ extern const struct file_operations xfs_dir_file_operations; extern ssize_t xfs_vn_listxattr(struct dentry *, char *data, size_t size); extern void xfs_setattr_time(struct xfs_inode *ip, struct iattr *iattr); -extern int xfs_vn_setattr_size(struct dentry *dentry, struct iattr *vap); +int xfs_vn_setattr_size(struct user_namespace *mnt_userns, + struct dentry *dentry, struct iattr *vap); #endif /* __XFS_IOPS_H__ */ diff --git a/fs/xfs/xfs_itable.c b/fs/xfs/xfs_itable.c index 16ca97a7ff00..3498b97fb06d 100644 --- a/fs/xfs/xfs_itable.c +++ b/fs/xfs/xfs_itable.c @@ -54,10 +54,12 @@ struct xfs_bstat_chunk { STATIC int xfs_bulkstat_one_int( struct xfs_mount *mp, + struct user_namespace *mnt_userns, struct xfs_trans *tp, xfs_ino_t ino, struct xfs_bstat_chunk *bc) { + struct user_namespace *sb_userns = mp->m_super->s_user_ns; struct xfs_icdinode *dic; /* dinode core info pointer */ struct xfs_inode *ip; /* incore inode pointer */ struct inode *inode; @@ -86,8 +88,8 @@ xfs_bulkstat_one_int( */ buf->bs_projectid = ip->i_d.di_projid; buf->bs_ino = ino; - buf->bs_uid = i_uid_read(inode); - buf->bs_gid = i_gid_read(inode); + buf->bs_uid = from_kuid(sb_userns, i_uid_into_mnt(mnt_userns, inode)); + buf->bs_gid = from_kgid(sb_userns, i_gid_into_mnt(mnt_userns, inode)); buf->bs_size = dic->di_size; buf->bs_nlink = inode->i_nlink; @@ -166,6 +168,12 @@ xfs_bulkstat_one( }; int error; + if (breq->mnt_userns != &init_user_ns) { + xfs_warn_ratelimited(breq->mp, + "bulkstat not supported inside of idmapped mounts."); + return -EINVAL; + } + ASSERT(breq->icount == 1); bc.buf = kmem_zalloc(sizeof(struct xfs_bulkstat), @@ -173,7 +181,8 @@ xfs_bulkstat_one( if (!bc.buf) return -ENOMEM; - error = xfs_bulkstat_one_int(breq->mp, NULL, breq->startino, &bc); + error = xfs_bulkstat_one_int(breq->mp, breq->mnt_userns, NULL, + breq->startino, &bc); kmem_free(bc.buf); @@ -194,9 +203,10 @@ xfs_bulkstat_iwalk( xfs_ino_t ino, void *data) { + struct xfs_bstat_chunk *bc = data; int error; - error = xfs_bulkstat_one_int(mp, tp, ino, data); + error = xfs_bulkstat_one_int(mp, bc->breq->mnt_userns, tp, ino, data); /* bulkstat just skips over missing inodes */ if (error == -ENOENT || error == -EINVAL) return 0; @@ -239,6 +249,11 @@ xfs_bulkstat( }; int error; + if (breq->mnt_userns != &init_user_ns) { + xfs_warn_ratelimited(breq->mp, + "bulkstat not supported inside of idmapped mounts."); + return -EINVAL; + } if (xfs_bulkstat_already_done(breq->mp, breq->startino)) return 0; diff --git a/fs/xfs/xfs_itable.h b/fs/xfs/xfs_itable.h index 96a1e2a9be3f..7078d10c9b12 100644 --- a/fs/xfs/xfs_itable.h +++ b/fs/xfs/xfs_itable.h @@ -8,6 +8,7 @@ /* In-memory representation of a userspace request for batch inode data. */ struct xfs_ibulk { struct xfs_mount *mp; + struct user_namespace *mnt_userns; void __user *ubuffer; /* user output buffer */ xfs_ino_t startino; /* start with this inode */ unsigned int icount; /* number of elements in ubuffer */ diff --git a/fs/xfs/xfs_iwalk.c b/fs/xfs/xfs_iwalk.c index eae3aff9bc97..c4a340f1f1e1 100644 --- a/fs/xfs/xfs_iwalk.c +++ b/fs/xfs/xfs_iwalk.c @@ -618,15 +618,12 @@ xfs_iwalk_threaded( { struct xfs_pwork_ctl pctl; xfs_agnumber_t agno = XFS_INO_TO_AGNO(mp, startino); - unsigned int nr_threads; int error; ASSERT(agno < mp->m_sb.sb_agcount); ASSERT(!(flags & ~XFS_IWALK_FLAGS_ALL)); - nr_threads = xfs_pwork_guess_datadev_parallelism(mp); - error = xfs_pwork_init(mp, &pctl, xfs_iwalk_ag_work, "xfs_iwalk", - nr_threads); + error = xfs_pwork_init(mp, &pctl, xfs_iwalk_ag_work, "xfs_iwalk"); if (error) return error; diff --git a/fs/xfs/xfs_linux.h b/fs/xfs/xfs_linux.h index 5b7a1e201559..af6be9b9ccdf 100644 --- a/fs/xfs/xfs_linux.h +++ b/fs/xfs/xfs_linux.h @@ -98,8 +98,7 @@ typedef __u32 xfs_nlink_t; #define xfs_rotorstep xfs_params.rotorstep.val #define xfs_inherit_nodefrag xfs_params.inherit_nodfrg.val #define xfs_fstrm_centisecs xfs_params.fstrm_timer.val -#define xfs_eofb_secs xfs_params.eofb_timer.val -#define xfs_cowb_secs xfs_params.cowb_timer.val +#define xfs_blockgc_secs xfs_params.blockgc_timer.val #define current_cpu() (raw_smp_processor_id()) #define current_set_flags_nested(sp, f) \ diff --git a/fs/xfs/xfs_log.c b/fs/xfs/xfs_log.c index fa2d05e65ff1..06041834daa3 100644 --- a/fs/xfs/xfs_log.c +++ b/fs/xfs/xfs_log.c @@ -91,6 +91,9 @@ STATIC int xlog_iclogs_empty( struct xlog *log); +static int +xfs_log_cover(struct xfs_mount *); + static void xlog_grant_sub_space( struct xlog *log, @@ -347,6 +350,25 @@ xlog_tic_add_region(xlog_ticket_t *tic, uint len, uint type) tic->t_res_num++; } +bool +xfs_log_writable( + struct xfs_mount *mp) +{ + /* + * Never write to the log on norecovery mounts, if the block device is + * read-only, or if the filesystem is shutdown. Read-only mounts still + * allow internal writes for log recovery and unmount purposes, so don't + * restrict that case here. + */ + if (mp->m_flags & XFS_MOUNT_NORECOVERY) + return false; + if (xfs_readonly_buftarg(mp->m_log->l_targ)) + return false; + if (XFS_FORCED_SHUTDOWN(mp)) + return false; + return true; +} + /* * Replenish the byte reservation required by moving the grant write head. */ @@ -741,7 +763,7 @@ xfs_log_mount_finish( xfs_log_force(mp, XFS_LOG_SYNC); xfs_ail_push_all_sync(mp->m_ail); } - xfs_wait_buftarg(mp->m_ddev_targp); + xfs_buftarg_drain(mp->m_ddev_targp); if (readonly) mp->m_flags |= XFS_MOUNT_RDONLY; @@ -886,15 +908,8 @@ xfs_log_unmount_write( { struct xlog *log = mp->m_log; - /* - * Don't write out unmount record on norecovery mounts or ro devices. - * Or, if we are doing a forced umount (typically because of IO errors). - */ - if (mp->m_flags & XFS_MOUNT_NORECOVERY || - xfs_readonly_buftarg(log->l_targ)) { - ASSERT(mp->m_flags & XFS_MOUNT_RDONLY); + if (!xfs_log_writable(mp)) return; - } xfs_log_force(mp, XFS_LOG_SYNC); @@ -924,10 +939,9 @@ xfs_log_unmount_write( * To do this, we first need to shut down the background log work so it is not * trying to cover the log as we clean up. We then need to unpin all objects in * the log so we can then flush them out. Once they have completed their IO and - * run the callbacks removing themselves from the AIL, we can write the unmount - * record. + * run the callbacks removing themselves from the AIL, we can cover the log. */ -void +int xfs_log_quiesce( struct xfs_mount *mp) { @@ -936,16 +950,24 @@ xfs_log_quiesce( /* * The superblock buffer is uncached and while xfs_ail_push_all_sync() - * will push it, xfs_wait_buftarg() will not wait for it. Further, + * will push it, xfs_buftarg_wait() will not wait for it. Further, * xfs_buf_iowait() cannot be used because it was pushed with the * XBF_ASYNC flag set, so we need to use a lock/unlock pair to wait for * the IO to complete. */ xfs_ail_push_all_sync(mp->m_ail); - xfs_wait_buftarg(mp->m_ddev_targp); + xfs_buftarg_wait(mp->m_ddev_targp); xfs_buf_lock(mp->m_sb_bp); xfs_buf_unlock(mp->m_sb_bp); + return xfs_log_cover(mp); +} + +void +xfs_log_clean( + struct xfs_mount *mp) +{ + xfs_log_quiesce(mp); xfs_log_unmount_write(mp); } @@ -960,7 +982,9 @@ void xfs_log_unmount( struct xfs_mount *mp) { - xfs_log_quiesce(mp); + xfs_log_clean(mp); + + xfs_buftarg_drain(mp->m_ddev_targp); xfs_trans_ail_destroy(mp); @@ -1037,17 +1061,15 @@ xfs_log_space_wake( * there's no point in running a dummy transaction at this point because we * can't start trying to idle the log until both the CIL and AIL are empty. */ -static int -xfs_log_need_covered(xfs_mount_t *mp) +static bool +xfs_log_need_covered( + struct xfs_mount *mp) { - struct xlog *log = mp->m_log; - int needed = 0; - - if (!xfs_fs_writable(mp, SB_FREEZE_WRITE)) - return 0; + struct xlog *log = mp->m_log; + bool needed = false; if (!xlog_cil_empty(log)) - return 0; + return false; spin_lock(&log->l_icloglock); switch (log->l_covered_state) { @@ -1062,14 +1084,14 @@ xfs_log_need_covered(xfs_mount_t *mp) if (!xlog_iclogs_empty(log)) break; - needed = 1; + needed = true; if (log->l_covered_state == XLOG_STATE_COVER_NEED) log->l_covered_state = XLOG_STATE_COVER_DONE; else log->l_covered_state = XLOG_STATE_COVER_DONE2; break; default: - needed = 1; + needed = true; break; } spin_unlock(&log->l_icloglock); @@ -1077,6 +1099,60 @@ xfs_log_need_covered(xfs_mount_t *mp) } /* + * Explicitly cover the log. This is similar to background log covering but + * intended for usage in quiesce codepaths. The caller is responsible to ensure + * the log is idle and suitable for covering. The CIL, iclog buffers and AIL + * must all be empty. + */ +static int +xfs_log_cover( + struct xfs_mount *mp) +{ + int error = 0; + bool need_covered; + + ASSERT((xlog_cil_empty(mp->m_log) && xlog_iclogs_empty(mp->m_log) && + !xfs_ail_min_lsn(mp->m_log->l_ailp)) || + XFS_FORCED_SHUTDOWN(mp)); + + if (!xfs_log_writable(mp)) + return 0; + + /* + * xfs_log_need_covered() is not idempotent because it progresses the + * state machine if the log requires covering. Therefore, we must call + * this function once and use the result until we've issued an sb sync. + * Do so first to make that abundantly clear. + * + * Fall into the covering sequence if the log needs covering or the + * mount has lazy superblock accounting to sync to disk. The sb sync + * used for covering accumulates the in-core counters, so covering + * handles this for us. + */ + need_covered = xfs_log_need_covered(mp); + if (!need_covered && !xfs_sb_version_haslazysbcount(&mp->m_sb)) + return 0; + + /* + * To cover the log, commit the superblock twice (at most) in + * independent checkpoints. The first serves as a reference for the + * tail pointer. The sync transaction and AIL push empties the AIL and + * updates the in-core tail to the LSN of the first checkpoint. The + * second commit updates the on-disk tail with the in-core LSN, + * covering the log. Push the AIL one more time to leave it empty, as + * we found it. + */ + do { + error = xfs_sync_sb(mp, true); + if (error) + break; + xfs_ail_push_all_sync(mp->m_ail); + } while (xfs_log_need_covered(mp)); + + return error; +} + +/* * We may be holding the log iclog lock upon entering this routine. */ xfs_lsn_t @@ -1259,7 +1335,7 @@ xfs_log_worker( struct xfs_mount *mp = log->l_mp; /* dgc: errors ignored - not fatal and nowhere to report them */ - if (xfs_log_need_covered(mp)) { + if (xfs_fs_writable(mp, SB_FREEZE_WRITE) && xfs_log_need_covered(mp)) { /* * Dump a transaction into the log that contains no real change. * This is needed to stamp the current tail LSN into the log @@ -1416,8 +1492,9 @@ xlog_alloc_log( log->l_iclog->ic_prev = prev_iclog; /* re-write 1st prev ptr */ log->l_ioend_workqueue = alloc_workqueue("xfs-log/%s", - WQ_MEM_RECLAIM | WQ_FREEZABLE | WQ_HIGHPRI, 0, - mp->m_super->s_id); + XFS_WQFLAGS(WQ_FREEZABLE | WQ_MEM_RECLAIM | + WQ_HIGHPRI), + 0, mp->m_super->s_id); if (!log->l_ioend_workqueue) goto out_free_iclog; @@ -2538,12 +2615,15 @@ xlog_covered_state( int iclogs_changed) { /* - * We usually go to NEED. But we go to NEED2 if the changed indicates we - * are done writing the dummy record. If we are done with the second - * dummy recored (DONE2), then we go to IDLE. + * We go to NEED for any non-covering writes. We go to NEED2 if we just + * wrote the first covering record (DONE). We go to IDLE if we just + * wrote the second covering record (DONE2) and remain in IDLE until a + * non-covering write occurs. */ switch (prev_state) { case XLOG_STATE_COVER_IDLE: + if (iclogs_changed == 1) + return XLOG_STATE_COVER_IDLE; case XLOG_STATE_COVER_NEED: case XLOG_STATE_COVER_NEED2: break; diff --git a/fs/xfs/xfs_log.h b/fs/xfs/xfs_log.h index 58c3fcbec94a..044e02cb8921 100644 --- a/fs/xfs/xfs_log.h +++ b/fs/xfs/xfs_log.h @@ -127,6 +127,7 @@ int xfs_log_reserve(struct xfs_mount *mp, int xfs_log_regrant(struct xfs_mount *mp, struct xlog_ticket *tic); void xfs_log_unmount(struct xfs_mount *mp); int xfs_log_force_umount(struct xfs_mount *mp, int logerror); +bool xfs_log_writable(struct xfs_mount *mp); struct xlog_ticket *xfs_log_ticket_get(struct xlog_ticket *ticket); void xfs_log_ticket_put(struct xlog_ticket *ticket); @@ -137,7 +138,8 @@ void xlog_cil_process_committed(struct list_head *list); bool xfs_log_item_in_current_chkpt(struct xfs_log_item *lip); void xfs_log_work_queue(struct xfs_mount *mp); -void xfs_log_quiesce(struct xfs_mount *mp); +int xfs_log_quiesce(struct xfs_mount *mp); +void xfs_log_clean(struct xfs_mount *mp); bool xfs_log_check_lsn(struct xfs_mount *, xfs_lsn_t); bool xfs_log_in_recovery(struct xfs_mount *); diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index 7110507a2b6b..1c97b155a8ee 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -126,6 +126,7 @@ __xfs_free_perag( { struct xfs_perag *pag = container_of(head, struct xfs_perag, rcu_head); + ASSERT(!delayed_work_pending(&pag->pag_blockgc_work)); ASSERT(atomic_read(&pag->pag_ref) == 0); kmem_free(pag); } @@ -146,6 +147,7 @@ xfs_free_perag( spin_unlock(&mp->m_perag_lock); ASSERT(pag); ASSERT(atomic_read(&pag->pag_ref) == 0); + cancel_delayed_work_sync(&pag->pag_blockgc_work); xfs_iunlink_destroy(pag); xfs_buf_hash_destroy(pag); call_rcu(&pag->rcu_head, __xfs_free_perag); @@ -201,6 +203,7 @@ xfs_initialize_perag( pag->pag_agno = index; pag->pag_mount = mp; spin_lock_init(&pag->pag_ici_lock); + INIT_DELAYED_WORK(&pag->pag_blockgc_work, xfs_blockgc_worker); INIT_RADIX_TREE(&pag->pag_ici_root, GFP_ATOMIC); error = xfs_buf_hash_init(pag); @@ -632,6 +635,47 @@ xfs_check_summary_counts( } /* + * Flush and reclaim dirty inodes in preparation for unmount. Inodes and + * internal inode structures can be sitting in the CIL and AIL at this point, + * so we need to unpin them, write them back and/or reclaim them before unmount + * can proceed. + * + * An inode cluster that has been freed can have its buffer still pinned in + * memory because the transaction is still sitting in a iclog. The stale inodes + * on that buffer will be pinned to the buffer until the transaction hits the + * disk and the callbacks run. Pushing the AIL will skip the stale inodes and + * may never see the pinned buffer, so nothing will push out the iclog and + * unpin the buffer. + * + * Hence we need to force the log to unpin everything first. However, log + * forces don't wait for the discards they issue to complete, so we have to + * explicitly wait for them to complete here as well. + * + * Then we can tell the world we are unmounting so that error handling knows + * that the filesystem is going away and we should error out anything that we + * have been retrying in the background. This will prevent never-ending + * retries in AIL pushing from hanging the unmount. + * + * Finally, we can push the AIL to clean all the remaining dirty objects, then + * reclaim the remaining inodes that are still in memory at this point in time. + */ +static void +xfs_unmount_flush_inodes( + struct xfs_mount *mp) +{ + xfs_log_force(mp, XFS_LOG_SYNC); + xfs_extent_busy_wait_all(mp); + flush_workqueue(xfs_discard_wq); + + mp->m_flags |= XFS_MOUNT_UNMOUNTING; + + xfs_ail_push_all_sync(mp->m_ail); + cancel_delayed_work_sync(&mp->m_reclaim_work); + xfs_reclaim_inodes(mp); + xfs_health_unmount(mp); +} + +/* * This function does the following on an initial mount of a file system: * - reads the superblock from disk and init the mount struct * - if we're a 32-bit kernel, do a size check on the superblock @@ -946,7 +990,7 @@ xfs_mountfs( */ if ((mp->m_flags & (XFS_MOUNT_RDONLY|XFS_MOUNT_NORECOVERY)) == XFS_MOUNT_RDONLY) { - xfs_quiesce_attr(mp); + xfs_log_clean(mp); } /* @@ -1005,7 +1049,7 @@ xfs_mountfs( /* Clean out dquots that might be in memory after quotacheck. */ xfs_qm_unmount(mp); /* - * Cancel all delayed reclaim work and reclaim the inodes directly. + * Flush all inode reclamation work and flush the log. * We have to do this /after/ rtunmount and qm_unmount because those * two will have scheduled delayed reclaim for the rt/quota inodes. * @@ -1015,16 +1059,13 @@ xfs_mountfs( * qm_unmount_quotas and therefore rely on qm_unmount to release the * quota inodes. */ - cancel_delayed_work_sync(&mp->m_reclaim_work); - xfs_reclaim_inodes(mp); - xfs_health_unmount(mp); + xfs_unmount_flush_inodes(mp); out_log_dealloc: - mp->m_flags |= XFS_MOUNT_UNMOUNTING; xfs_log_mount_cancel(mp); out_fail_wait: if (mp->m_logdev_targp && mp->m_logdev_targp != mp->m_ddev_targp) - xfs_wait_buftarg(mp->m_logdev_targp); - xfs_wait_buftarg(mp->m_ddev_targp); + xfs_buftarg_drain(mp->m_logdev_targp); + xfs_buftarg_drain(mp->m_ddev_targp); out_free_perag: xfs_free_perag(mp); out_free_dir: @@ -1054,53 +1095,13 @@ xfs_unmountfs( uint64_t resblks; int error; - xfs_stop_block_reaping(mp); + xfs_blockgc_stop(mp); xfs_fs_unreserve_ag_blocks(mp); xfs_qm_unmount_quotas(mp); xfs_rtunmount_inodes(mp); xfs_irele(mp->m_rootip); - /* - * We can potentially deadlock here if we have an inode cluster - * that has been freed has its buffer still pinned in memory because - * the transaction is still sitting in a iclog. The stale inodes - * on that buffer will be pinned to the buffer until the - * transaction hits the disk and the callbacks run. Pushing the AIL will - * skip the stale inodes and may never see the pinned buffer, so - * nothing will push out the iclog and unpin the buffer. Hence we - * need to force the log here to ensure all items are flushed into the - * AIL before we go any further. - */ - xfs_log_force(mp, XFS_LOG_SYNC); - - /* - * Wait for all busy extents to be freed, including completion of - * any discard operation. - */ - xfs_extent_busy_wait_all(mp); - flush_workqueue(xfs_discard_wq); - - /* - * We now need to tell the world we are unmounting. This will allow - * us to detect that the filesystem is going away and we should error - * out anything that we have been retrying in the background. This will - * prevent neverending retries in AIL pushing from hanging the unmount. - */ - mp->m_flags |= XFS_MOUNT_UNMOUNTING; - - /* - * Flush all pending changes from the AIL. - */ - xfs_ail_push_all_sync(mp->m_ail); - - /* - * Reclaim all inodes. At this point there should be no dirty inodes and - * none should be pinned or locked. Stop background inode reclaim here - * if it is still running. - */ - cancel_delayed_work_sync(&mp->m_reclaim_work); - xfs_reclaim_inodes(mp); - xfs_health_unmount(mp); + xfs_unmount_flush_inodes(mp); xfs_qm_unmount(mp); @@ -1124,12 +1125,6 @@ xfs_unmountfs( xfs_warn(mp, "Unable to free reserved block pool. " "Freespace may not be correct on next mount."); - error = xfs_log_sbcount(mp); - if (error) - xfs_warn(mp, "Unable to update superblock counters. " - "Freespace may not be correct on next mount."); - - xfs_log_unmount(mp); xfs_da_unmount(mp); xfs_uuid_unmount(mp); @@ -1165,32 +1160,6 @@ xfs_fs_writable( } /* - * xfs_log_sbcount - * - * Sync the superblock counters to disk. - * - * Note this code can be called during the process of freezing, so we use the - * transaction allocator that does not block when the transaction subsystem is - * in its frozen state. - */ -int -xfs_log_sbcount(xfs_mount_t *mp) -{ - /* allow this to proceed during the freeze sequence... */ - if (!xfs_fs_writable(mp, SB_FREEZE_COMPLETE)) - return 0; - - /* - * we don't need to do this if we are updating the superblock - * counters on every modification. - */ - if (!xfs_sb_version_haslazysbcount(&mp->m_sb)) - return 0; - - return xfs_sync_sb(mp, true); -} - -/* * Deltas for the block count can vary from 1 to very large, but lock contention * only occurs on frequent small block count updates such as in the delayed * allocation path for buffered writes (page a time updates). Hence we set diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h index dfa429b77ee2..659ad95fe3e0 100644 --- a/fs/xfs/xfs_mount.h +++ b/fs/xfs/xfs_mount.h @@ -93,7 +93,7 @@ typedef struct xfs_mount { struct workqueue_struct *m_unwritten_workqueue; struct workqueue_struct *m_cil_workqueue; struct workqueue_struct *m_reclaim_workqueue; - struct workqueue_struct *m_eofblocks_workqueue; + struct workqueue_struct *m_blockgc_workqueue; struct workqueue_struct *m_sync_workqueue; int m_bsize; /* fs logical block size */ @@ -177,10 +177,6 @@ typedef struct xfs_mount { uint64_t m_resblks_avail;/* available reserved blocks */ uint64_t m_resblks_save; /* reserved blks @ remount,ro */ struct delayed_work m_reclaim_work; /* background inode reclaim */ - struct delayed_work m_eofblocks_work; /* background eof blocks - trimming */ - struct delayed_work m_cowblocks_work; /* background cow blocks - trimming */ struct xfs_kobj m_kobj; struct xfs_kobj m_error_kobj; struct xfs_kobj m_error_meta_kobj; @@ -369,6 +365,9 @@ typedef struct xfs_perag { /* Blocks reserved for the reverse mapping btree. */ struct xfs_ag_resv pag_rmapbt_resv; + /* background prealloc block trimming */ + struct delayed_work pag_blockgc_work; + /* reference count */ uint8_t pagf_refcount_level; @@ -399,7 +398,6 @@ int xfs_buf_hash_init(xfs_perag_t *pag); void xfs_buf_hash_destroy(xfs_perag_t *pag); extern void xfs_uuid_table_free(void); -extern int xfs_log_sbcount(xfs_mount_t *); extern uint64_t xfs_default_resblks(xfs_mount_t *mp); extern int xfs_mountfs(xfs_mount_t *mp); extern int xfs_initialize_perag(xfs_mount_t *mp, xfs_agnumber_t agcount, diff --git a/fs/xfs/xfs_mru_cache.c b/fs/xfs/xfs_mru_cache.c index a06661dac5be..34c3b16f834f 100644 --- a/fs/xfs/xfs_mru_cache.c +++ b/fs/xfs/xfs_mru_cache.c @@ -294,7 +294,7 @@ int xfs_mru_cache_init(void) { xfs_mru_reap_wq = alloc_workqueue("xfs_mru_cache", - WQ_MEM_RECLAIM|WQ_FREEZABLE, 1); + XFS_WQFLAGS(WQ_MEM_RECLAIM | WQ_FREEZABLE), 1); if (!xfs_mru_reap_wq) return -ENOMEM; return 0; diff --git a/fs/xfs/xfs_pwork.c b/fs/xfs/xfs_pwork.c index b03333f1c84a..c283b801cc5d 100644 --- a/fs/xfs/xfs_pwork.c +++ b/fs/xfs/xfs_pwork.c @@ -61,16 +61,18 @@ xfs_pwork_init( struct xfs_mount *mp, struct xfs_pwork_ctl *pctl, xfs_pwork_work_fn work_fn, - const char *tag, - unsigned int nr_threads) + const char *tag) { + unsigned int nr_threads = 0; + #ifdef DEBUG if (xfs_globals.pwork_threads >= 0) nr_threads = xfs_globals.pwork_threads; #endif trace_xfs_pwork_init(mp, nr_threads, current->pid); - pctl->wq = alloc_workqueue("%s-%d", WQ_FREEZABLE, nr_threads, tag, + pctl->wq = alloc_workqueue("%s-%d", + WQ_UNBOUND | WQ_SYSFS | WQ_FREEZABLE, nr_threads, tag, current->pid); if (!pctl->wq) return -ENOMEM; @@ -117,20 +119,3 @@ xfs_pwork_poll( atomic_read(&pctl->nr_work) == 0, HZ) == 0) touch_softlockup_watchdog(); } - -/* - * Return the amount of parallelism that the data device can handle, or 0 for - * no limit. - */ -unsigned int -xfs_pwork_guess_datadev_parallelism( - struct xfs_mount *mp) -{ - struct xfs_buftarg *btp = mp->m_ddev_targp; - - /* - * For now we'll go with the most conservative setting possible, - * which is two threads for an SSD and 1 thread everywhere else. - */ - return blk_queue_nonrot(btp->bt_bdev->bd_disk->queue) ? 2 : 1; -} diff --git a/fs/xfs/xfs_pwork.h b/fs/xfs/xfs_pwork.h index 8133124cf3bb..c0ef81fc85dd 100644 --- a/fs/xfs/xfs_pwork.h +++ b/fs/xfs/xfs_pwork.h @@ -51,11 +51,9 @@ xfs_pwork_want_abort( } int xfs_pwork_init(struct xfs_mount *mp, struct xfs_pwork_ctl *pctl, - xfs_pwork_work_fn work_fn, const char *tag, - unsigned int nr_threads); + xfs_pwork_work_fn work_fn, const char *tag); void xfs_pwork_queue(struct xfs_pwork_ctl *pctl, struct xfs_pwork *pwork); int xfs_pwork_destroy(struct xfs_pwork_ctl *pctl); void xfs_pwork_poll(struct xfs_pwork_ctl *pctl); -unsigned int xfs_pwork_guess_datadev_parallelism(struct xfs_mount *mp); #endif /* __XFS_PWORK_H__ */ diff --git a/fs/xfs/xfs_qm.c b/fs/xfs/xfs_qm.c index c134eb4aeaa8..bfa4164990b1 100644 --- a/fs/xfs/xfs_qm.c +++ b/fs/xfs/xfs_qm.c @@ -787,7 +787,8 @@ xfs_qm_qino_alloc( return error; if (need_alloc) { - error = xfs_dir_ialloc(&tp, NULL, S_IFREG, 1, 0, 0, ipp); + error = xfs_dir_ialloc(&init_user_ns, &tp, NULL, S_IFREG, 1, 0, + 0, ipp); if (error) { xfs_trans_cancel(tp); return error; @@ -1786,105 +1787,35 @@ xfs_qm_vop_chown( xfs_trans_mod_dquot(tp, newdq, XFS_TRANS_DQ_ICOUNT, 1); /* - * Take an extra reference, because the inode is going to keep - * this dquot pointer even after the trans_commit. + * Back when we made quota reservations for the chown, we reserved the + * ondisk blocks + delalloc blocks with the new dquot. Now that we've + * switched the dquots, decrease the new dquot's block reservation + * (having already bumped up the real counter) so that we don't have + * any reservation to give back when we commit. */ - *IO_olddq = xfs_qm_dqhold(newdq); + xfs_trans_mod_dquot(tp, newdq, XFS_TRANS_DQ_RES_BLKS, + -ip->i_delayed_blks); - return prevdq; -} - -/* - * Quota reservations for setattr(AT_UID|AT_GID|AT_PROJID). - */ -int -xfs_qm_vop_chown_reserve( - struct xfs_trans *tp, - struct xfs_inode *ip, - struct xfs_dquot *udqp, - struct xfs_dquot *gdqp, - struct xfs_dquot *pdqp, - uint flags) -{ - struct xfs_mount *mp = ip->i_mount; - uint64_t delblks; - unsigned int blkflags; - struct xfs_dquot *udq_unres = NULL; - struct xfs_dquot *gdq_unres = NULL; - struct xfs_dquot *pdq_unres = NULL; - struct xfs_dquot *udq_delblks = NULL; - struct xfs_dquot *gdq_delblks = NULL; - struct xfs_dquot *pdq_delblks = NULL; - int error; - - - ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL|XFS_ILOCK_SHARED)); - ASSERT(XFS_IS_QUOTA_RUNNING(mp)); - - delblks = ip->i_delayed_blks; - blkflags = XFS_IS_REALTIME_INODE(ip) ? - XFS_QMOPT_RES_RTBLKS : XFS_QMOPT_RES_REGBLKS; - - if (XFS_IS_UQUOTA_ON(mp) && udqp && - i_uid_read(VFS_I(ip)) != udqp->q_id) { - udq_delblks = udqp; - /* - * If there are delayed allocation blocks, then we have to - * unreserve those from the old dquot, and add them to the - * new dquot. - */ - if (delblks) { - ASSERT(ip->i_udquot); - udq_unres = ip->i_udquot; - } - } - if (XFS_IS_GQUOTA_ON(ip->i_mount) && gdqp && - i_gid_read(VFS_I(ip)) != gdqp->q_id) { - gdq_delblks = gdqp; - if (delblks) { - ASSERT(ip->i_gdquot); - gdq_unres = ip->i_gdquot; - } - } - - if (XFS_IS_PQUOTA_ON(ip->i_mount) && pdqp && - ip->i_d.di_projid != pdqp->q_id) { - pdq_delblks = pdqp; - if (delblks) { - ASSERT(ip->i_pdquot); - pdq_unres = ip->i_pdquot; - } - } - - error = xfs_trans_reserve_quota_bydquots(tp, ip->i_mount, - udq_delblks, gdq_delblks, pdq_delblks, - ip->i_d.di_nblocks, 1, flags | blkflags); - if (error) - return error; + /* + * Give the incore reservation for delalloc blocks back to the old + * dquot. We don't normally handle delalloc quota reservations + * transactionally, so just lock the dquot and subtract from the + * reservation. Dirty the transaction because it's too late to turn + * back now. + */ + tp->t_flags |= XFS_TRANS_DIRTY; + xfs_dqlock(prevdq); + ASSERT(prevdq->q_blk.reserved >= ip->i_delayed_blks); + prevdq->q_blk.reserved -= ip->i_delayed_blks; + xfs_dqunlock(prevdq); /* - * Do the delayed blks reservations/unreservations now. Since, these - * are done without the help of a transaction, if a reservation fails - * its previous reservations won't be automatically undone by trans - * code. So, we have to do it manually here. + * Take an extra reference, because the inode is going to keep + * this dquot pointer even after the trans_commit. */ - if (delblks) { - /* - * Do the reservations first. Unreservation can't fail. - */ - ASSERT(udq_delblks || gdq_delblks || pdq_delblks); - ASSERT(udq_unres || gdq_unres || pdq_unres); - error = xfs_trans_reserve_quota_bydquots(NULL, ip->i_mount, - udq_delblks, gdq_delblks, pdq_delblks, - (xfs_qcnt_t)delblks, 0, flags | blkflags); - if (error) - return error; - xfs_trans_reserve_quota_bydquots(NULL, ip->i_mount, - udq_unres, gdq_unres, pdq_unres, - -((xfs_qcnt_t)delblks), 0, blkflags); - } + *IO_olddq = xfs_qm_dqhold(newdq); - return 0; + return prevdq; } int diff --git a/fs/xfs/xfs_quota.h b/fs/xfs/xfs_quota.h index 5a62398940d0..d00d01302545 100644 --- a/fs/xfs/xfs_quota.h +++ b/fs/xfs/xfs_quota.h @@ -81,11 +81,14 @@ extern void xfs_trans_mod_dquot_byino(struct xfs_trans *, struct xfs_inode *, uint, int64_t); extern void xfs_trans_apply_dquot_deltas(struct xfs_trans *); extern void xfs_trans_unreserve_and_mod_dquots(struct xfs_trans *); -extern int xfs_trans_reserve_quota_nblks(struct xfs_trans *, - struct xfs_inode *, int64_t, long, uint); +int xfs_trans_reserve_quota_nblks(struct xfs_trans *tp, struct xfs_inode *ip, + int64_t dblocks, int64_t rblocks, bool force); extern int xfs_trans_reserve_quota_bydquots(struct xfs_trans *, struct xfs_mount *, struct xfs_dquot *, struct xfs_dquot *, struct xfs_dquot *, int64_t, long, uint); +int xfs_trans_reserve_quota_icreate(struct xfs_trans *tp, + struct xfs_dquot *udqp, struct xfs_dquot *gdqp, + struct xfs_dquot *pdqp, int64_t dblocks); extern int xfs_qm_vop_dqalloc(struct xfs_inode *, kuid_t, kgid_t, prid_t, uint, struct xfs_dquot **, struct xfs_dquot **, @@ -95,9 +98,6 @@ extern void xfs_qm_vop_create_dqattach(struct xfs_trans *, struct xfs_inode *, extern int xfs_qm_vop_rename_dqattach(struct xfs_inode **); extern struct xfs_dquot *xfs_qm_vop_chown(struct xfs_trans *, struct xfs_inode *, struct xfs_dquot **, struct xfs_dquot *); -extern int xfs_qm_vop_chown_reserve(struct xfs_trans *, struct xfs_inode *, - struct xfs_dquot *, struct xfs_dquot *, - struct xfs_dquot *, uint); extern int xfs_qm_dqattach(struct xfs_inode *); extern int xfs_qm_dqattach_locked(struct xfs_inode *ip, bool doalloc); extern void xfs_qm_dqdetach(struct xfs_inode *); @@ -108,6 +108,11 @@ extern void xfs_qm_mount_quotas(struct xfs_mount *); extern void xfs_qm_unmount(struct xfs_mount *); extern void xfs_qm_unmount_quotas(struct xfs_mount *); +static inline int +xfs_quota_reserve_blkres(struct xfs_inode *ip, int64_t blocks) +{ + return xfs_trans_reserve_quota_nblks(NULL, ip, blocks, 0, false); +} #else static inline int xfs_qm_vop_dqalloc(struct xfs_inode *ip, kuid_t kuid, kgid_t kgid, @@ -121,11 +126,12 @@ xfs_qm_vop_dqalloc(struct xfs_inode *ip, kuid_t kuid, kgid_t kgid, } #define xfs_trans_dup_dqinfo(tp, tp2) #define xfs_trans_free_dqinfo(tp) -#define xfs_trans_mod_dquot_byino(tp, ip, fields, delta) +#define xfs_trans_mod_dquot_byino(tp, ip, fields, delta) do { } while (0) #define xfs_trans_apply_dquot_deltas(tp) #define xfs_trans_unreserve_and_mod_dquots(tp) static inline int xfs_trans_reserve_quota_nblks(struct xfs_trans *tp, - struct xfs_inode *ip, int64_t nblks, long ninos, uint flags) + struct xfs_inode *ip, int64_t dblocks, int64_t rblocks, + bool force) { return 0; } @@ -136,26 +142,39 @@ static inline int xfs_trans_reserve_quota_bydquots(struct xfs_trans *tp, { return 0; } + +static inline int +xfs_quota_reserve_blkres(struct xfs_inode *ip, int64_t blocks) +{ + return 0; +} + +static inline int +xfs_trans_reserve_quota_icreate(struct xfs_trans *tp, struct xfs_dquot *udqp, + struct xfs_dquot *gdqp, struct xfs_dquot *pdqp, int64_t dblocks) +{ + return 0; +} + #define xfs_qm_vop_create_dqattach(tp, ip, u, g, p) #define xfs_qm_vop_rename_dqattach(it) (0) #define xfs_qm_vop_chown(tp, ip, old, new) (NULL) -#define xfs_qm_vop_chown_reserve(tp, ip, u, g, p, fl) (0) #define xfs_qm_dqattach(ip) (0) #define xfs_qm_dqattach_locked(ip, fl) (0) #define xfs_qm_dqdetach(ip) -#define xfs_qm_dqrele(d) -#define xfs_qm_statvfs(ip, s) +#define xfs_qm_dqrele(d) do { (d) = (d); } while(0) +#define xfs_qm_statvfs(ip, s) do { } while(0) #define xfs_qm_newmount(mp, a, b) (0) #define xfs_qm_mount_quotas(mp) #define xfs_qm_unmount(mp) #define xfs_qm_unmount_quotas(mp) #endif /* CONFIG_XFS_QUOTA */ -#define xfs_trans_unreserve_quota_nblks(tp, ip, nblks, ninos, flags) \ - xfs_trans_reserve_quota_nblks(tp, ip, -(nblks), -(ninos), flags) -#define xfs_trans_reserve_quota(tp, mp, ud, gd, pd, nb, ni, f) \ - xfs_trans_reserve_quota_bydquots(tp, mp, ud, gd, pd, nb, ni, \ - f | XFS_QMOPT_RES_REGBLKS) +static inline int +xfs_quota_unreserve_blkres(struct xfs_inode *ip, int64_t blocks) +{ + return xfs_quota_reserve_blkres(ip, -blocks); +} extern int xfs_mount_reset_sbqflags(struct xfs_mount *); diff --git a/fs/xfs/xfs_reflink.c b/fs/xfs/xfs_reflink.c index 6fa05fb78189..725c7d8e4438 100644 --- a/fs/xfs/xfs_reflink.c +++ b/fs/xfs/xfs_reflink.c @@ -376,16 +376,14 @@ xfs_reflink_allocate_cow( resblks = XFS_DIOSTRAT_SPACE_RES(mp, resaligned); xfs_iunlock(ip, *lockmode); - error = xfs_trans_alloc(mp, &M_RES(mp)->tr_write, resblks, 0, 0, &tp); - *lockmode = XFS_ILOCK_EXCL; - xfs_ilock(ip, *lockmode); + *lockmode = 0; + error = xfs_trans_alloc_inode(ip, &M_RES(mp)->tr_write, resblks, 0, + false, &tp); if (error) return error; - error = xfs_qm_dqattach_locked(ip, false); - if (error) - goto out_trans_cancel; + *lockmode = XFS_ILOCK_EXCL; /* * Check for an overlapping extent again now that we dropped the ilock. @@ -398,20 +396,13 @@ xfs_reflink_allocate_cow( goto convert; } - error = xfs_trans_reserve_quota_nblks(tp, ip, resblks, 0, - XFS_QMOPT_RES_REGBLKS); - if (error) - goto out_trans_cancel; - - xfs_trans_ijoin(tp, ip, 0); - /* Allocate the entire reservation as unwritten blocks. */ nimaps = 1; error = xfs_bmapi_write(tp, ip, imap->br_startoff, imap->br_blockcount, XFS_BMAPI_COWFORK | XFS_BMAPI_PREALLOC, 0, cmap, &nimaps); if (error) - goto out_unreserve; + goto out_trans_cancel; xfs_inode_set_cowblocks_tag(ip); error = xfs_trans_commit(tp); @@ -436,9 +427,6 @@ convert: trace_xfs_reflink_convert_cow(ip, cmap); return xfs_reflink_convert_cow_locked(ip, offset_fsb, count_fsb); -out_unreserve: - xfs_trans_unreserve_quota_nblks(tp, ip, (long)resblks, 0, - XFS_QMOPT_RES_REGBLKS); out_trans_cancel: xfs_trans_cancel(tp); return error; @@ -508,9 +496,8 @@ xfs_reflink_cancel_cow_blocks( xfs_bmap_del_extent_cow(ip, &icur, &got, &del); /* Remove the quota reservation */ - error = xfs_trans_reserve_quota_nblks(NULL, ip, - -(long)del.br_blockcount, 0, - XFS_QMOPT_RES_REGBLKS); + error = xfs_quota_unreserve_blkres(ip, + del.br_blockcount); if (error) break; } else { @@ -628,6 +615,11 @@ xfs_reflink_end_cow_extent( xfs_ilock(ip, XFS_ILOCK_EXCL); xfs_trans_ijoin(tp, ip, 0); + error = xfs_iext_count_may_overflow(ip, XFS_DATA_FORK, + XFS_IEXT_REFLINK_END_COW_CNT); + if (error) + goto out_cancel; + /* * In case of racing, overlapping AIO writes no COW extents might be * left by the time I/O completes for the loser of the race. In that @@ -997,22 +989,47 @@ xfs_reflink_remap_extent( struct xfs_mount *mp = ip->i_mount; struct xfs_trans *tp; xfs_off_t newlen; - int64_t qres, qdelta; + int64_t qdelta = 0; unsigned int resblks; + bool quota_reserved = true; bool smap_real; bool dmap_written = xfs_bmap_is_written_extent(dmap); + int iext_delta = 0; int nimaps; int error; - /* Start a rolling transaction to switch the mappings */ + /* + * Start a rolling transaction to switch the mappings. + * + * Adding a written extent to the extent map can cause a bmbt split, + * and removing a mapped extent from the extent can cause a bmbt split. + * The two operations cannot both cause a split since they operate on + * the same index in the bmap btree, so we only need a reservation for + * one bmbt split if either thing is happening. However, we haven't + * locked the inode yet, so we reserve assuming this is the case. + * + * The first allocation call tries to reserve enough space to handle + * mapping dmap into a sparse part of the file plus the bmbt split. We + * haven't locked the inode or read the existing mapping yet, so we do + * not know for sure that we need the space. This should succeed most + * of the time. + * + * If the first attempt fails, try again but reserving only enough + * space to handle a bmbt split. This is the hard minimum requirement, + * and we revisit quota reservations later when we know more about what + * we're remapping. + */ resblks = XFS_EXTENTADD_SPACE_RES(mp, XFS_DATA_FORK); - error = xfs_trans_alloc(mp, &M_RES(mp)->tr_write, resblks, 0, 0, &tp); + error = xfs_trans_alloc_inode(ip, &M_RES(mp)->tr_write, + resblks + dmap->br_blockcount, 0, false, &tp); + if (error == -EDQUOT || error == -ENOSPC) { + quota_reserved = false; + error = xfs_trans_alloc_inode(ip, &M_RES(mp)->tr_write, + resblks, 0, false, &tp); + } if (error) goto out; - xfs_ilock(ip, XFS_ILOCK_EXCL); - xfs_trans_ijoin(tp, ip, 0); - /* * Read what's currently mapped in the destination file into smap. * If smap isn't a hole, we will have to remove it before we can add @@ -1060,15 +1077,9 @@ xfs_reflink_remap_extent( } /* - * Compute quota reservation if we think the quota block counter for + * Increase quota reservation if we think the quota block counter for * this file could increase. * - * Adding a written extent to the extent map can cause a bmbt split, - * and removing a mapped extent from the extent can cause a bmbt split. - * The two operations cannot both cause a split since they operate on - * the same index in the bmap btree, so we only need a reservation for - * one bmbt split if either thing is happening. - * * If we are mapping a written extent into the file, we need to have * enough quota block count reservation to handle the blocks in that * extent. We log only the delta to the quota block counts, so if the @@ -1081,19 +1092,29 @@ xfs_reflink_remap_extent( * count. This is suboptimal, but the VFS flushed the dest range * before we started. That should have removed all the delalloc * reservations, but we code defensively. + * + * xfs_trans_alloc_inode above already tried to grab an even larger + * quota reservation, and kicked off a blockgc scan if it couldn't. + * If we can't get a potentially smaller quota reservation now, we're + * done. */ - qres = qdelta = 0; - if (smap_real || dmap_written) - qres = XFS_EXTENTADD_SPACE_RES(mp, XFS_DATA_FORK); - if (!smap_real && dmap_written) - qres += dmap->br_blockcount; - if (qres > 0) { - error = xfs_trans_reserve_quota_nblks(tp, ip, qres, 0, - XFS_QMOPT_RES_REGBLKS); + if (!quota_reserved && !smap_real && dmap_written) { + error = xfs_trans_reserve_quota_nblks(tp, ip, + dmap->br_blockcount, 0, false); if (error) goto out_cancel; } + if (smap_real) + ++iext_delta; + + if (dmap_written) + ++iext_delta; + + error = xfs_iext_count_may_overflow(ip, XFS_DATA_FORK, iext_delta); + if (error) + goto out_cancel; + if (smap_real) { /* * If the extent we're unmapping is backed by storage (written diff --git a/fs/xfs/xfs_rtalloc.c b/fs/xfs/xfs_rtalloc.c index b4999fb01ff7..161b0e8992ba 100644 --- a/fs/xfs/xfs_rtalloc.c +++ b/fs/xfs/xfs_rtalloc.c @@ -804,6 +804,11 @@ xfs_growfs_rt_alloc( xfs_ilock(ip, XFS_ILOCK_EXCL); xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL); + error = xfs_iext_count_may_overflow(ip, XFS_DATA_FORK, + XFS_IEXT_ADD_NOSPLIT_CNT); + if (error) + goto out_trans_cancel; + /* * Allocate blocks to the bitmap file. */ diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index 813be879a5e5..e5e0713bebcd 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -35,6 +35,7 @@ #include "xfs_refcount_item.h" #include "xfs_bmap_item.h" #include "xfs_reflink.h" +#include "xfs_pwork.h" #include <linux/magic.h> #include <linux/fs_context.h> @@ -342,7 +343,7 @@ void xfs_blkdev_issue_flush( xfs_buftarg_t *buftarg) { - blkdev_issue_flush(buftarg->bt_bdev, GFP_NOFS); + blkdev_issue_flush(buftarg->bt_bdev); } STATIC void @@ -495,40 +496,44 @@ xfs_init_mount_workqueues( struct xfs_mount *mp) { mp->m_buf_workqueue = alloc_workqueue("xfs-buf/%s", - WQ_MEM_RECLAIM|WQ_FREEZABLE, 1, mp->m_super->s_id); + XFS_WQFLAGS(WQ_FREEZABLE | WQ_MEM_RECLAIM), + 1, mp->m_super->s_id); if (!mp->m_buf_workqueue) goto out; mp->m_unwritten_workqueue = alloc_workqueue("xfs-conv/%s", - WQ_MEM_RECLAIM|WQ_FREEZABLE, 0, mp->m_super->s_id); + XFS_WQFLAGS(WQ_FREEZABLE | WQ_MEM_RECLAIM), + 0, mp->m_super->s_id); if (!mp->m_unwritten_workqueue) goto out_destroy_buf; mp->m_cil_workqueue = alloc_workqueue("xfs-cil/%s", - WQ_MEM_RECLAIM | WQ_FREEZABLE | WQ_UNBOUND, + XFS_WQFLAGS(WQ_FREEZABLE | WQ_MEM_RECLAIM | WQ_UNBOUND), 0, mp->m_super->s_id); if (!mp->m_cil_workqueue) goto out_destroy_unwritten; mp->m_reclaim_workqueue = alloc_workqueue("xfs-reclaim/%s", - WQ_MEM_RECLAIM|WQ_FREEZABLE, 0, mp->m_super->s_id); + XFS_WQFLAGS(WQ_FREEZABLE | WQ_MEM_RECLAIM), + 0, mp->m_super->s_id); if (!mp->m_reclaim_workqueue) goto out_destroy_cil; - mp->m_eofblocks_workqueue = alloc_workqueue("xfs-eofblocks/%s", - WQ_MEM_RECLAIM|WQ_FREEZABLE, 0, mp->m_super->s_id); - if (!mp->m_eofblocks_workqueue) + mp->m_blockgc_workqueue = alloc_workqueue("xfs-blockgc/%s", + WQ_SYSFS | WQ_UNBOUND | WQ_FREEZABLE | WQ_MEM_RECLAIM, + 0, mp->m_super->s_id); + if (!mp->m_blockgc_workqueue) goto out_destroy_reclaim; - mp->m_sync_workqueue = alloc_workqueue("xfs-sync/%s", WQ_FREEZABLE, 0, - mp->m_super->s_id); + mp->m_sync_workqueue = alloc_workqueue("xfs-sync/%s", + XFS_WQFLAGS(WQ_FREEZABLE), 0, mp->m_super->s_id); if (!mp->m_sync_workqueue) goto out_destroy_eofb; return 0; out_destroy_eofb: - destroy_workqueue(mp->m_eofblocks_workqueue); + destroy_workqueue(mp->m_blockgc_workqueue); out_destroy_reclaim: destroy_workqueue(mp->m_reclaim_workqueue); out_destroy_cil: @@ -546,7 +551,7 @@ xfs_destroy_mount_workqueues( struct xfs_mount *mp) { destroy_workqueue(mp->m_sync_workqueue); - destroy_workqueue(mp->m_eofblocks_workqueue); + destroy_workqueue(mp->m_blockgc_workqueue); destroy_workqueue(mp->m_reclaim_workqueue); destroy_workqueue(mp->m_cil_workqueue); destroy_workqueue(mp->m_unwritten_workqueue); @@ -868,39 +873,6 @@ xfs_restore_resvblks(struct xfs_mount *mp) } /* - * Trigger writeback of all the dirty metadata in the file system. - * - * This ensures that the metadata is written to their location on disk rather - * than just existing in transactions in the log. This means after a quiesce - * there is no log replay required to write the inodes to disk - this is the - * primary difference between a sync and a quiesce. - * - * We cancel log work early here to ensure all transactions the log worker may - * run have finished before we clean up and log the superblock and write an - * unmount record. The unfreeze process is responsible for restarting the log - * worker correctly. - */ -void -xfs_quiesce_attr( - struct xfs_mount *mp) -{ - int error = 0; - - cancel_delayed_work_sync(&mp->m_log->l_work); - - /* force the log to unpin objects from the now complete transactions */ - xfs_log_force(mp, XFS_LOG_SYNC); - - - /* Push the superblock and write an unmount record */ - error = xfs_log_sbcount(mp); - if (error) - xfs_warn(mp, "xfs_attr_quiesce: failed to log sb changes. " - "Frozen image may not be consistent."); - xfs_log_quiesce(mp); -} - -/* * Second stage of a freeze. The data is already frozen so we only * need to take care of the metadata. Once that's done sync the superblock * to the log to dirty it in case of a crash while frozen. This ensures that we @@ -920,10 +892,9 @@ xfs_fs_freeze( * set a GFP_NOFS context here to avoid recursion deadlocks. */ flags = memalloc_nofs_save(); - xfs_stop_block_reaping(mp); + xfs_blockgc_stop(mp); xfs_save_resvblks(mp); - xfs_quiesce_attr(mp); - ret = xfs_sync_sb(mp, true); + ret = xfs_log_quiesce(mp); memalloc_nofs_restore(flags); return ret; } @@ -936,7 +907,7 @@ xfs_fs_unfreeze( xfs_restore_resvblks(mp); xfs_log_work_queue(mp); - xfs_start_block_reaping(mp); + xfs_blockgc_start(mp); return 0; } @@ -1720,7 +1691,7 @@ xfs_remount_rw( xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE); return error; } - xfs_start_block_reaping(mp); + xfs_blockgc_start(mp); /* Create the per-AG metadata reservation pool .*/ error = xfs_fs_reserve_ag_blocks(mp); @@ -1740,10 +1711,10 @@ xfs_remount_ro( * Cancel background eofb scanning so it cannot race with the final * log force+buftarg wait and deadlock the remount. */ - xfs_stop_block_reaping(mp); + xfs_blockgc_stop(mp); /* Get rid of any leftover CoW reservations... */ - error = xfs_icache_free_cowblocks(mp, NULL); + error = xfs_blockgc_free_space(mp, NULL); if (error) { xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE); return error; @@ -1765,7 +1736,7 @@ xfs_remount_ro( */ xfs_save_resvblks(mp); - xfs_quiesce_attr(mp); + xfs_log_clean(mp); mp->m_flags |= XFS_MOUNT_RDONLY; return 0; @@ -1872,8 +1843,6 @@ static int xfs_init_fs_context( mutex_init(&mp->m_growlock); INIT_WORK(&mp->m_flush_inodes_work, xfs_flush_inodes_worker); INIT_DELAYED_WORK(&mp->m_reclaim_work, xfs_reclaim_worker); - INIT_DELAYED_WORK(&mp->m_eofblocks_work, xfs_eofblocks_worker); - INIT_DELAYED_WORK(&mp->m_cowblocks_work, xfs_cowblocks_worker); mp->m_kobj.kobject.kset = xfs_kset; /* * We don't create the finobt per-ag space reservation until after log @@ -1912,7 +1881,7 @@ static struct file_system_type xfs_fs_type = { .init_fs_context = xfs_init_fs_context, .parameters = xfs_fs_parameters, .kill_sb = kill_block_super, - .fs_flags = FS_REQUIRES_DEV, + .fs_flags = FS_REQUIRES_DEV | FS_ALLOW_IDMAP, }; MODULE_ALIAS_FS("xfs"); @@ -2119,11 +2088,12 @@ xfs_init_workqueues(void) * max_active value for this workqueue. */ xfs_alloc_wq = alloc_workqueue("xfsalloc", - WQ_MEM_RECLAIM|WQ_FREEZABLE, 0); + XFS_WQFLAGS(WQ_MEM_RECLAIM | WQ_FREEZABLE), 0); if (!xfs_alloc_wq) return -ENOMEM; - xfs_discard_wq = alloc_workqueue("xfsdiscard", WQ_UNBOUND, 0); + xfs_discard_wq = alloc_workqueue("xfsdiscard", XFS_WQFLAGS(WQ_UNBOUND), + 0); if (!xfs_discard_wq) goto out_free_alloc_wq; diff --git a/fs/xfs/xfs_super.h b/fs/xfs/xfs_super.h index b552cf6d3379..1ca484b8357f 100644 --- a/fs/xfs/xfs_super.h +++ b/fs/xfs/xfs_super.h @@ -75,6 +75,12 @@ extern void xfs_qm_exit(void); XFS_ASSERT_FATAL_STRING \ XFS_DBG_STRING /* DBG must be last */ +#ifdef DEBUG +# define XFS_WQFLAGS(wqflags) (WQ_SYSFS | (wqflags)) +#else +# define XFS_WQFLAGS(wqflags) (wqflags) +#endif + struct xfs_inode; struct xfs_mount; struct xfs_buftarg; diff --git a/fs/xfs/xfs_symlink.c b/fs/xfs/xfs_symlink.c index 1f43fd7f3209..7f368b10ded1 100644 --- a/fs/xfs/xfs_symlink.c +++ b/fs/xfs/xfs_symlink.c @@ -134,6 +134,7 @@ xfs_readlink( int xfs_symlink( + struct user_namespace *mnt_userns, struct xfs_inode *dp, struct xfs_name *link_name, const char *target_path, @@ -181,7 +182,8 @@ xfs_symlink( /* * Make sure that we have allocated dquot(s) on disk. */ - error = xfs_qm_vop_dqalloc(dp, current_fsuid(), current_fsgid(), prid, + error = xfs_qm_vop_dqalloc(dp, fsuid_into_mnt(mnt_userns), + fsgid_into_mnt(mnt_userns), prid, XFS_QMOPT_QUOTALL | XFS_QMOPT_INHERIT, &udqp, &gdqp, &pdqp); if (error) @@ -197,9 +199,10 @@ xfs_symlink( fs_blocks = xfs_symlink_blocks(mp, pathlen); resblks = XFS_SYMLINK_SPACE_RES(mp, link_name->len, fs_blocks); - error = xfs_trans_alloc(mp, &M_RES(mp)->tr_symlink, resblks, 0, 0, &tp); + error = xfs_trans_alloc_icreate(mp, &M_RES(mp)->tr_symlink, udqp, gdqp, + pdqp, resblks, &tp); if (error) - goto out_release_inode; + goto out_release_dquots; xfs_ilock(dp, XFS_ILOCK_EXCL | XFS_ILOCK_PARENT); unlock_dp_on_error = true; @@ -212,19 +215,16 @@ xfs_symlink( goto out_trans_cancel; } - /* - * Reserve disk quota : blocks and inode. - */ - error = xfs_trans_reserve_quota(tp, mp, udqp, gdqp, - pdqp, resblks, 1, 0); + error = xfs_iext_count_may_overflow(dp, XFS_DATA_FORK, + XFS_IEXT_DIR_MANIP_CNT(mp)); if (error) goto out_trans_cancel; /* * Allocate an inode for the symlink. */ - error = xfs_dir_ialloc(&tp, dp, S_IFLNK | (mode & ~S_IFMT), 1, 0, - prid, &ip); + error = xfs_dir_ialloc(mnt_userns, &tp, dp, S_IFLNK | (mode & ~S_IFMT), + 1, 0, prid, &ip); if (error) goto out_trans_cancel; @@ -300,6 +300,7 @@ xfs_symlink( } ASSERT(pathlen == 0); } + i_size_write(VFS_I(ip), ip->i_d.di_size); /* * Create the directory entry for the symlink. @@ -342,7 +343,7 @@ out_release_inode: xfs_finish_inode_setup(ip); xfs_irele(ip); } - +out_release_dquots: xfs_qm_dqrele(udqp); xfs_qm_dqrele(gdqp); xfs_qm_dqrele(pdqp); diff --git a/fs/xfs/xfs_symlink.h b/fs/xfs/xfs_symlink.h index b1fa091427e6..2586b7e393f3 100644 --- a/fs/xfs/xfs_symlink.h +++ b/fs/xfs/xfs_symlink.h @@ -7,8 +7,9 @@ /* Kernel only symlink definitions */ -int xfs_symlink(struct xfs_inode *dp, struct xfs_name *link_name, - const char *target_path, umode_t mode, struct xfs_inode **ipp); +int xfs_symlink(struct user_namespace *mnt_userns, struct xfs_inode *dp, + struct xfs_name *link_name, const char *target_path, + umode_t mode, struct xfs_inode **ipp); int xfs_readlink_bmap_ilocked(struct xfs_inode *ip, char *link); int xfs_readlink(struct xfs_inode *ip, char *link); int xfs_inactive_symlink(struct xfs_inode *ip); diff --git a/fs/xfs/xfs_sysctl.c b/fs/xfs/xfs_sysctl.c index fac9de7ee6d0..546a6cd96729 100644 --- a/fs/xfs/xfs_sysctl.c +++ b/fs/xfs/xfs_sysctl.c @@ -51,7 +51,7 @@ xfs_panic_mask_proc_handler( #endif /* CONFIG_PROC_FS */ STATIC int -xfs_deprecate_irix_sgid_inherit_proc_handler( +xfs_deprecated_dointvec_minmax( struct ctl_table *ctl, int write, void *buffer, @@ -59,24 +59,8 @@ xfs_deprecate_irix_sgid_inherit_proc_handler( loff_t *ppos) { if (write) { - printk_once(KERN_WARNING - "XFS: " "%s sysctl option is deprecated.\n", - ctl->procname); - } - return proc_dointvec_minmax(ctl, write, buffer, lenp, ppos); -} - -STATIC int -xfs_deprecate_irix_symlink_mode_proc_handler( - struct ctl_table *ctl, - int write, - void *buffer, - size_t *lenp, - loff_t *ppos) -{ - if (write) { - printk_once(KERN_WARNING - "XFS: " "%s sysctl option is deprecated.\n", + printk_ratelimited(KERN_WARNING + "XFS: %s sysctl option is deprecated.\n", ctl->procname); } return proc_dointvec_minmax(ctl, write, buffer, lenp, ppos); @@ -88,7 +72,7 @@ static struct ctl_table xfs_table[] = { .data = &xfs_params.sgid_inherit.val, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = xfs_deprecate_irix_sgid_inherit_proc_handler, + .proc_handler = xfs_deprecated_dointvec_minmax, .extra1 = &xfs_params.sgid_inherit.min, .extra2 = &xfs_params.sgid_inherit.max }, @@ -97,7 +81,7 @@ static struct ctl_table xfs_table[] = { .data = &xfs_params.symlink_mode.val, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = xfs_deprecate_irix_symlink_mode_proc_handler, + .proc_handler = xfs_deprecated_dointvec_minmax, .extra1 = &xfs_params.symlink_mode.min, .extra2 = &xfs_params.symlink_mode.max }, @@ -194,21 +178,21 @@ static struct ctl_table xfs_table[] = { }, { .procname = "speculative_prealloc_lifetime", - .data = &xfs_params.eofb_timer.val, + .data = &xfs_params.blockgc_timer.val, .maxlen = sizeof(int), .mode = 0644, .proc_handler = proc_dointvec_minmax, - .extra1 = &xfs_params.eofb_timer.min, - .extra2 = &xfs_params.eofb_timer.max, + .extra1 = &xfs_params.blockgc_timer.min, + .extra2 = &xfs_params.blockgc_timer.max, }, { .procname = "speculative_cow_prealloc_lifetime", - .data = &xfs_params.cowb_timer.val, + .data = &xfs_params.blockgc_timer.val, .maxlen = sizeof(int), .mode = 0644, - .proc_handler = proc_dointvec_minmax, - .extra1 = &xfs_params.cowb_timer.min, - .extra2 = &xfs_params.cowb_timer.max, + .proc_handler = xfs_deprecated_dointvec_minmax, + .extra1 = &xfs_params.blockgc_timer.min, + .extra2 = &xfs_params.blockgc_timer.max, }, /* please keep this the last entry */ #ifdef CONFIG_PROC_FS diff --git a/fs/xfs/xfs_sysctl.h b/fs/xfs/xfs_sysctl.h index 8abf4640f1d5..7692e76ead33 100644 --- a/fs/xfs/xfs_sysctl.h +++ b/fs/xfs/xfs_sysctl.h @@ -35,8 +35,7 @@ typedef struct xfs_param { xfs_sysctl_val_t rotorstep; /* inode32 AG rotoring control knob */ xfs_sysctl_val_t inherit_nodfrg;/* Inherit the "nodefrag" inode flag. */ xfs_sysctl_val_t fstrm_timer; /* Filestream dir-AG assoc'n timeout. */ - xfs_sysctl_val_t eofb_timer; /* Interval between eofb scan wakeups */ - xfs_sysctl_val_t cowb_timer; /* Interval between cowb scan wakeups */ + xfs_sysctl_val_t blockgc_timer; /* Interval between blockgc scans */ } xfs_param_t; /* diff --git a/fs/xfs/xfs_trace.c b/fs/xfs/xfs_trace.c index 120398a37c2a..9b8d703dc9fd 100644 --- a/fs/xfs/xfs_trace.c +++ b/fs/xfs/xfs_trace.c @@ -29,6 +29,7 @@ #include "xfs_filestream.h" #include "xfs_fsmap.h" #include "xfs_btree_staging.h" +#include "xfs_icache.h" /* * We include this last to have the helpers above available for the trace diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h index 5a263ae3d4f0..e74bbb648f83 100644 --- a/fs/xfs/xfs_trace.h +++ b/fs/xfs/xfs_trace.h @@ -37,6 +37,7 @@ struct xfs_trans_res; struct xfs_inobt_rec_incore; union xfs_btree_ptr; struct xfs_dqtrx; +struct xfs_eofblocks; #define XFS_ATTR_FILTER_FLAGS \ { XFS_ATTR_ROOT, "ROOT" }, \ @@ -154,10 +155,8 @@ DEFINE_PERAG_REF_EVENT(xfs_perag_get_tag); DEFINE_PERAG_REF_EVENT(xfs_perag_put); DEFINE_PERAG_REF_EVENT(xfs_perag_set_reclaim); DEFINE_PERAG_REF_EVENT(xfs_perag_clear_reclaim); -DEFINE_PERAG_REF_EVENT(xfs_perag_set_eofblocks); -DEFINE_PERAG_REF_EVENT(xfs_perag_clear_eofblocks); -DEFINE_PERAG_REF_EVENT(xfs_perag_set_cowblocks); -DEFINE_PERAG_REF_EVENT(xfs_perag_clear_cowblocks); +DEFINE_PERAG_REF_EVENT(xfs_perag_set_blockgc); +DEFINE_PERAG_REF_EVENT(xfs_perag_clear_blockgc); DECLARE_EVENT_CLASS(xfs_ag_class, TP_PROTO(struct xfs_mount *mp, xfs_agnumber_t agno), @@ -358,7 +357,7 @@ DEFINE_BUF_EVENT(xfs_buf_get_uncached); DEFINE_BUF_EVENT(xfs_buf_item_relse); DEFINE_BUF_EVENT(xfs_buf_iodone_async); DEFINE_BUF_EVENT(xfs_buf_error_relse); -DEFINE_BUF_EVENT(xfs_buf_wait_buftarg); +DEFINE_BUF_EVENT(xfs_buf_drain_buftarg); DEFINE_BUF_EVENT(xfs_trans_read_buf_shut); /* not really buffer traces, but the buf provides useful information */ @@ -1287,8 +1286,8 @@ TRACE_EVENT(xfs_log_assign_tail_lsn, ) DECLARE_EVENT_CLASS(xfs_file_class, - TP_PROTO(struct xfs_inode *ip, size_t count, loff_t offset), - TP_ARGS(ip, count, offset), + TP_PROTO(struct kiocb *iocb, struct iov_iter *iter), + TP_ARGS(iocb, iter), TP_STRUCT__entry( __field(dev_t, dev) __field(xfs_ino_t, ino) @@ -1297,11 +1296,11 @@ DECLARE_EVENT_CLASS(xfs_file_class, __field(size_t, count) ), TP_fast_assign( - __entry->dev = VFS_I(ip)->i_sb->s_dev; - __entry->ino = ip->i_ino; - __entry->size = ip->i_d.di_size; - __entry->offset = offset; - __entry->count = count; + __entry->dev = file_inode(iocb->ki_filp)->i_sb->s_dev; + __entry->ino = XFS_I(file_inode(iocb->ki_filp))->i_ino; + __entry->size = XFS_I(file_inode(iocb->ki_filp))->i_d.di_size; + __entry->offset = iocb->ki_pos; + __entry->count = iov_iter_count(iter); ), TP_printk("dev %d:%d ino 0x%llx size 0x%llx offset 0x%llx count 0x%zx", MAJOR(__entry->dev), MINOR(__entry->dev), @@ -1313,14 +1312,16 @@ DECLARE_EVENT_CLASS(xfs_file_class, #define DEFINE_RW_EVENT(name) \ DEFINE_EVENT(xfs_file_class, name, \ - TP_PROTO(struct xfs_inode *ip, size_t count, loff_t offset), \ - TP_ARGS(ip, count, offset)) + TP_PROTO(struct kiocb *iocb, struct iov_iter *iter), \ + TP_ARGS(iocb, iter)) DEFINE_RW_EVENT(xfs_file_buffered_read); DEFINE_RW_EVENT(xfs_file_direct_read); DEFINE_RW_EVENT(xfs_file_dax_read); DEFINE_RW_EVENT(xfs_file_buffered_write); DEFINE_RW_EVENT(xfs_file_direct_write); DEFINE_RW_EVENT(xfs_file_dax_write); +DEFINE_RW_EVENT(xfs_reflink_bounce_dio_write); + DECLARE_EVENT_CLASS(xfs_imap_class, TP_PROTO(struct xfs_inode *ip, xfs_off_t offset, ssize_t count, @@ -3294,8 +3295,6 @@ DEFINE_INODE_IREC_EVENT(xfs_reflink_cow_found); DEFINE_INODE_IREC_EVENT(xfs_reflink_cow_enospc); DEFINE_INODE_IREC_EVENT(xfs_reflink_convert_cow); -DEFINE_SIMPLE_IO_EVENT(xfs_reflink_bounce_dio_write); - DEFINE_SIMPLE_IO_EVENT(xfs_reflink_cancel_cow_range); DEFINE_SIMPLE_IO_EVENT(xfs_reflink_end_cow); DEFINE_INODE_IREC_EVENT(xfs_reflink_cow_remap); @@ -3888,6 +3887,47 @@ DEFINE_EVENT(xfs_timestamp_range_class, name, \ DEFINE_TIMESTAMP_RANGE_EVENT(xfs_inode_timestamp_range); DEFINE_TIMESTAMP_RANGE_EVENT(xfs_quota_expiry_range); +DECLARE_EVENT_CLASS(xfs_eofblocks_class, + TP_PROTO(struct xfs_mount *mp, struct xfs_eofblocks *eofb, + unsigned long caller_ip), + TP_ARGS(mp, eofb, caller_ip), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(__u32, flags) + __field(uint32_t, uid) + __field(uint32_t, gid) + __field(prid_t, prid) + __field(__u64, min_file_size) + __field(unsigned long, caller_ip) + ), + TP_fast_assign( + __entry->dev = mp->m_super->s_dev; + __entry->flags = eofb ? eofb->eof_flags : 0; + __entry->uid = eofb ? from_kuid(mp->m_super->s_user_ns, + eofb->eof_uid) : 0; + __entry->gid = eofb ? from_kgid(mp->m_super->s_user_ns, + eofb->eof_gid) : 0; + __entry->prid = eofb ? eofb->eof_prid : 0; + __entry->min_file_size = eofb ? eofb->eof_min_file_size : 0; + __entry->caller_ip = caller_ip; + ), + TP_printk("dev %d:%d flags 0x%x uid %u gid %u prid %u minsize %llu caller %pS", + MAJOR(__entry->dev), MINOR(__entry->dev), + __entry->flags, + __entry->uid, + __entry->gid, + __entry->prid, + __entry->min_file_size, + (char *)__entry->caller_ip) +); +#define DEFINE_EOFBLOCKS_EVENT(name) \ +DEFINE_EVENT(xfs_eofblocks_class, name, \ + TP_PROTO(struct xfs_mount *mp, struct xfs_eofblocks *eofb, \ + unsigned long caller_ip), \ + TP_ARGS(mp, eofb, caller_ip)) +DEFINE_EOFBLOCKS_EVENT(xfs_ioc_free_eofblocks); +DEFINE_EOFBLOCKS_EVENT(xfs_blockgc_free_space); + #endif /* _TRACE_XFS_H */ #undef TRACE_INCLUDE_PATH diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c index e72730f85af1..b22a09e9daee 100644 --- a/fs/xfs/xfs_trans.c +++ b/fs/xfs/xfs_trans.c @@ -20,6 +20,10 @@ #include "xfs_trace.h" #include "xfs_error.h" #include "xfs_defer.h" +#include "xfs_inode.h" +#include "xfs_dquot_item.h" +#include "xfs_dquot.h" +#include "xfs_icache.h" kmem_zone_t *xfs_trans_zone; @@ -68,6 +72,7 @@ xfs_trans_free( xfs_extent_busy_clear(tp->t_mountp, &tp->t_busy, false); trace_xfs_trans_free(tp, _RET_IP_); + xfs_trans_clear_context(tp); if (!(tp->t_flags & XFS_TRANS_NO_WRITECOUNT)) sb_end_intwrite(tp->t_mountp->m_super); xfs_trans_free_dqinfo(tp); @@ -119,7 +124,8 @@ xfs_trans_dup( ntp->t_rtx_res = tp->t_rtx_res - tp->t_rtx_res_used; tp->t_rtx_res = tp->t_rtx_res_used; - ntp->t_pflags = tp->t_pflags; + + xfs_trans_switch_context(tp, ntp); /* move deferred ops over to the new tp */ xfs_defer_move(ntp, tp); @@ -153,9 +159,6 @@ xfs_trans_reserve( int error = 0; bool rsvd = (tp->t_flags & XFS_TRANS_RESERVE) != 0; - /* Mark this thread as being in a transaction */ - current_set_flags_nested(&tp->t_pflags, PF_MEMALLOC_NOFS); - /* * Attempt to reserve the needed disk blocks by decrementing * the number needed from the number available. This will @@ -163,10 +166,8 @@ xfs_trans_reserve( */ if (blocks > 0) { error = xfs_mod_fdblocks(mp, -((int64_t)blocks), rsvd); - if (error != 0) { - current_restore_flags_nested(&tp->t_pflags, PF_MEMALLOC_NOFS); + if (error != 0) return -ENOSPC; - } tp->t_blk_res += blocks; } @@ -240,9 +241,6 @@ undo_blocks: xfs_mod_fdblocks(mp, (int64_t)blocks, rsvd); tp->t_blk_res = 0; } - - current_restore_flags_nested(&tp->t_pflags, PF_MEMALLOC_NOFS); - return error; } @@ -256,6 +254,7 @@ xfs_trans_alloc( struct xfs_trans **tpp) { struct xfs_trans *tp; + bool want_retry = true; int error; /* @@ -263,9 +262,11 @@ xfs_trans_alloc( * GFP_NOFS allocation context so that we avoid lockdep false positives * by doing GFP_KERNEL allocations inside sb_start_intwrite(). */ +retry: tp = kmem_cache_zalloc(xfs_trans_zone, GFP_KERNEL | __GFP_NOFAIL); if (!(flags & XFS_TRANS_NO_WRITECOUNT)) sb_start_intwrite(mp->m_super); + xfs_trans_set_context(tp); /* * Zero-reservation ("empty") transactions can't modify anything, so @@ -285,6 +286,22 @@ xfs_trans_alloc( tp->t_firstblock = NULLFSBLOCK; error = xfs_trans_reserve(tp, resp, blocks, rtextents); + if (error == -ENOSPC && want_retry) { + xfs_trans_cancel(tp); + + /* + * We weren't able to reserve enough space for the transaction. + * Flush the other speculative space allocations to free space. + * Do not perform a synchronous scan because callers can hold + * other locks. + */ + error = xfs_blockgc_free_space(mp, NULL); + if (error) + return error; + + want_retry = false; + goto retry; + } if (error) { xfs_trans_cancel(tp); return error; @@ -878,7 +895,6 @@ __xfs_trans_commit( xfs_log_commit_cil(mp, tp, &commit_lsn, regrant); - current_restore_flags_nested(&tp->t_pflags, PF_MEMALLOC_NOFS); xfs_trans_free(tp); /* @@ -910,7 +926,6 @@ out_unreserve: xfs_log_ticket_ungrant(mp->m_log, tp->t_ticket); tp->t_ticket = NULL; } - current_restore_flags_nested(&tp->t_pflags, PF_MEMALLOC_NOFS); xfs_trans_free_items(tp, !!error); xfs_trans_free(tp); @@ -970,9 +985,6 @@ xfs_trans_cancel( tp->t_ticket = NULL; } - /* mark this thread as no longer being in a transaction */ - current_restore_flags_nested(&tp->t_pflags, PF_MEMALLOC_NOFS); - xfs_trans_free_items(tp, dirty); xfs_trans_free(tp); } @@ -1024,3 +1036,183 @@ xfs_trans_roll( tres.tr_logflags = XFS_TRANS_PERM_LOG_RES; return xfs_trans_reserve(*tpp, &tres, 0, 0); } + +/* + * Allocate an transaction, lock and join the inode to it, and reserve quota. + * + * The caller must ensure that the on-disk dquots attached to this inode have + * already been allocated and initialized. The caller is responsible for + * releasing ILOCK_EXCL if a new transaction is returned. + */ +int +xfs_trans_alloc_inode( + struct xfs_inode *ip, + struct xfs_trans_res *resv, + unsigned int dblocks, + unsigned int rblocks, + bool force, + struct xfs_trans **tpp) +{ + struct xfs_trans *tp; + struct xfs_mount *mp = ip->i_mount; + bool retried = false; + int error; + +retry: + error = xfs_trans_alloc(mp, resv, dblocks, + rblocks / mp->m_sb.sb_rextsize, + force ? XFS_TRANS_RESERVE : 0, &tp); + if (error) + return error; + + xfs_ilock(ip, XFS_ILOCK_EXCL); + xfs_trans_ijoin(tp, ip, 0); + + error = xfs_qm_dqattach_locked(ip, false); + if (error) { + /* Caller should have allocated the dquots! */ + ASSERT(error != -ENOENT); + goto out_cancel; + } + + error = xfs_trans_reserve_quota_nblks(tp, ip, dblocks, rblocks, force); + if ((error == -EDQUOT || error == -ENOSPC) && !retried) { + xfs_trans_cancel(tp); + xfs_iunlock(ip, XFS_ILOCK_EXCL); + xfs_blockgc_free_quota(ip, 0); + retried = true; + goto retry; + } + if (error) + goto out_cancel; + + *tpp = tp; + return 0; + +out_cancel: + xfs_trans_cancel(tp); + xfs_iunlock(ip, XFS_ILOCK_EXCL); + return error; +} + +/* + * Allocate an transaction in preparation for inode creation by reserving quota + * against the given dquots. Callers are not required to hold any inode locks. + */ +int +xfs_trans_alloc_icreate( + struct xfs_mount *mp, + struct xfs_trans_res *resv, + struct xfs_dquot *udqp, + struct xfs_dquot *gdqp, + struct xfs_dquot *pdqp, + unsigned int dblocks, + struct xfs_trans **tpp) +{ + struct xfs_trans *tp; + bool retried = false; + int error; + +retry: + error = xfs_trans_alloc(mp, resv, dblocks, 0, 0, &tp); + if (error) + return error; + + error = xfs_trans_reserve_quota_icreate(tp, udqp, gdqp, pdqp, dblocks); + if ((error == -EDQUOT || error == -ENOSPC) && !retried) { + xfs_trans_cancel(tp); + xfs_blockgc_free_dquots(mp, udqp, gdqp, pdqp, 0); + retried = true; + goto retry; + } + if (error) { + xfs_trans_cancel(tp); + return error; + } + + *tpp = tp; + return 0; +} + +/* + * Allocate an transaction, lock and join the inode to it, and reserve quota + * in preparation for inode attribute changes that include uid, gid, or prid + * changes. + * + * The caller must ensure that the on-disk dquots attached to this inode have + * already been allocated and initialized. The ILOCK will be dropped when the + * transaction is committed or cancelled. + */ +int +xfs_trans_alloc_ichange( + struct xfs_inode *ip, + struct xfs_dquot *new_udqp, + struct xfs_dquot *new_gdqp, + struct xfs_dquot *new_pdqp, + bool force, + struct xfs_trans **tpp) +{ + struct xfs_trans *tp; + struct xfs_mount *mp = ip->i_mount; + struct xfs_dquot *udqp; + struct xfs_dquot *gdqp; + struct xfs_dquot *pdqp; + bool retried = false; + int error; + +retry: + error = xfs_trans_alloc(mp, &M_RES(mp)->tr_ichange, 0, 0, 0, &tp); + if (error) + return error; + + xfs_ilock(ip, XFS_ILOCK_EXCL); + xfs_trans_ijoin(tp, ip, XFS_ILOCK_EXCL); + + error = xfs_qm_dqattach_locked(ip, false); + if (error) { + /* Caller should have allocated the dquots! */ + ASSERT(error != -ENOENT); + goto out_cancel; + } + + /* + * For each quota type, skip quota reservations if the inode's dquots + * now match the ones that came from the caller, or the caller didn't + * pass one in. The inode's dquots can change if we drop the ILOCK to + * perform a blockgc scan, so we must preserve the caller's arguments. + */ + udqp = (new_udqp != ip->i_udquot) ? new_udqp : NULL; + gdqp = (new_gdqp != ip->i_gdquot) ? new_gdqp : NULL; + pdqp = (new_pdqp != ip->i_pdquot) ? new_pdqp : NULL; + if (udqp || gdqp || pdqp) { + unsigned int qflags = XFS_QMOPT_RES_REGBLKS; + + if (force) + qflags |= XFS_QMOPT_FORCE_RES; + + /* + * Reserve enough quota to handle blocks on disk and reserved + * for a delayed allocation. We'll actually transfer the + * delalloc reservation between dquots at chown time, even + * though that part is only semi-transactional. + */ + error = xfs_trans_reserve_quota_bydquots(tp, mp, udqp, gdqp, + pdqp, ip->i_d.di_nblocks + ip->i_delayed_blks, + 1, qflags); + if ((error == -EDQUOT || error == -ENOSPC) && !retried) { + xfs_trans_cancel(tp); + xfs_blockgc_free_dquots(mp, udqp, gdqp, pdqp, 0); + retried = true; + goto retry; + } + if (error) + goto out_cancel; + } + + *tpp = tp; + return 0; + +out_cancel: + xfs_trans_cancel(tp); + return error; +} diff --git a/fs/xfs/xfs_trans.h b/fs/xfs/xfs_trans.h index 084658946cc8..9dd745cf77c9 100644 --- a/fs/xfs/xfs_trans.h +++ b/fs/xfs/xfs_trans.h @@ -268,4 +268,47 @@ xfs_trans_item_relog( return lip->li_ops->iop_relog(lip, tp); } +struct xfs_dquot; + +int xfs_trans_alloc_inode(struct xfs_inode *ip, struct xfs_trans_res *resv, + unsigned int dblocks, unsigned int rblocks, bool force, + struct xfs_trans **tpp); +int xfs_trans_alloc_icreate(struct xfs_mount *mp, struct xfs_trans_res *resv, + struct xfs_dquot *udqp, struct xfs_dquot *gdqp, + struct xfs_dquot *pdqp, unsigned int dblocks, + struct xfs_trans **tpp); +int xfs_trans_alloc_ichange(struct xfs_inode *ip, struct xfs_dquot *udqp, + struct xfs_dquot *gdqp, struct xfs_dquot *pdqp, bool force, + struct xfs_trans **tpp); + +static inline void +xfs_trans_set_context( + struct xfs_trans *tp) +{ + ASSERT(current->journal_info == NULL); + tp->t_pflags = memalloc_nofs_save(); + current->journal_info = tp; +} + +static inline void +xfs_trans_clear_context( + struct xfs_trans *tp) +{ + if (current->journal_info == tp) { + memalloc_nofs_restore(tp->t_pflags); + current->journal_info = NULL; + } +} + +static inline void +xfs_trans_switch_context( + struct xfs_trans *old_tp, + struct xfs_trans *new_tp) +{ + ASSERT(current->journal_info == old_tp); + new_tp->t_pflags = old_tp->t_pflags; + old_tp->t_pflags = 0; + current->journal_info = new_tp; +} + #endif /* __XFS_TRANS_H__ */ diff --git a/fs/xfs/xfs_trans_dquot.c b/fs/xfs/xfs_trans_dquot.c index 28b8ac701919..48e09ea30ee5 100644 --- a/fs/xfs/xfs_trans_dquot.c +++ b/fs/xfs/xfs_trans_dquot.c @@ -16,6 +16,7 @@ #include "xfs_quota.h" #include "xfs_qm.h" #include "xfs_trace.h" +#include "xfs_error.h" STATIC void xfs_trans_alloc_dqinfo(xfs_trans_t *); @@ -691,9 +692,11 @@ xfs_trans_dqresv( nblks); xfs_trans_mod_dquot(tp, dqp, XFS_TRANS_DQ_RES_INOS, ninos); } - ASSERT(dqp->q_blk.reserved >= dqp->q_blk.count); - ASSERT(dqp->q_rtb.reserved >= dqp->q_rtb.count); - ASSERT(dqp->q_ino.reserved >= dqp->q_ino.count); + + if (XFS_IS_CORRUPT(mp, dqp->q_blk.reserved < dqp->q_blk.count) || + XFS_IS_CORRUPT(mp, dqp->q_rtb.reserved < dqp->q_rtb.count) || + XFS_IS_CORRUPT(mp, dqp->q_ino.reserved < dqp->q_ino.count)) + goto error_corrupt; xfs_dqunlock(dqp); return 0; @@ -703,6 +706,10 @@ error_return: if (xfs_dquot_type(dqp) == XFS_DQTYPE_PROJ) return -ENOSPC; return -EDQUOT; +error_corrupt: + xfs_dqunlock(dqp); + xfs_force_shutdown(mp, SHUTDOWN_CORRUPT_INCORE); + return -EFSCORRUPTED; } @@ -780,28 +787,60 @@ int xfs_trans_reserve_quota_nblks( struct xfs_trans *tp, struct xfs_inode *ip, - int64_t nblks, - long ninos, - uint flags) + int64_t dblocks, + int64_t rblocks, + bool force) { struct xfs_mount *mp = ip->i_mount; + unsigned int qflags = 0; + int error; if (!XFS_IS_QUOTA_RUNNING(mp) || !XFS_IS_QUOTA_ON(mp)) return 0; ASSERT(!xfs_is_quota_inode(&mp->m_sb, ip->i_ino)); - ASSERT(xfs_isilocked(ip, XFS_ILOCK_EXCL)); - ASSERT((flags & ~(XFS_QMOPT_FORCE_RES)) == XFS_TRANS_DQ_RES_RTBLKS || - (flags & ~(XFS_QMOPT_FORCE_RES)) == XFS_TRANS_DQ_RES_BLKS); - /* - * Reserve nblks against these dquots, with trans as the mediator. - */ - return xfs_trans_reserve_quota_bydquots(tp, mp, - ip->i_udquot, ip->i_gdquot, - ip->i_pdquot, - nblks, ninos, flags); + if (force) + qflags |= XFS_QMOPT_FORCE_RES; + + /* Reserve data device quota against the inode's dquots. */ + error = xfs_trans_reserve_quota_bydquots(tp, mp, ip->i_udquot, + ip->i_gdquot, ip->i_pdquot, dblocks, 0, + XFS_QMOPT_RES_REGBLKS | qflags); + if (error) + return error; + + /* Do the same but for realtime blocks. */ + error = xfs_trans_reserve_quota_bydquots(tp, mp, ip->i_udquot, + ip->i_gdquot, ip->i_pdquot, rblocks, 0, + XFS_QMOPT_RES_RTBLKS | qflags); + if (error) { + xfs_trans_reserve_quota_bydquots(tp, mp, ip->i_udquot, + ip->i_gdquot, ip->i_pdquot, -dblocks, 0, + XFS_QMOPT_RES_REGBLKS); + return error; + } + + return 0; +} + +/* Change the quota reservations for an inode creation activity. */ +int +xfs_trans_reserve_quota_icreate( + struct xfs_trans *tp, + struct xfs_dquot *udqp, + struct xfs_dquot *gdqp, + struct xfs_dquot *pdqp, + int64_t dblocks) +{ + struct xfs_mount *mp = tp->t_mountp; + + if (!XFS_IS_QUOTA_RUNNING(mp) || !XFS_IS_QUOTA_ON(mp)) + return 0; + + return xfs_trans_reserve_quota_bydquots(tp, mp, udqp, gdqp, pdqp, + dblocks, 1, XFS_QMOPT_RES_REGBLKS); } /* diff --git a/fs/xfs/xfs_xattr.c b/fs/xfs/xfs_xattr.c index bca48b308c02..12be32f66dc1 100644 --- a/fs/xfs/xfs_xattr.c +++ b/fs/xfs/xfs_xattr.c @@ -38,9 +38,10 @@ xfs_xattr_get(const struct xattr_handler *handler, struct dentry *unused, } static int -xfs_xattr_set(const struct xattr_handler *handler, struct dentry *unused, - struct inode *inode, const char *name, const void *value, - size_t size, int flags) +xfs_xattr_set(const struct xattr_handler *handler, + struct user_namespace *mnt_userns, struct dentry *unused, + struct inode *inode, const char *name, const void *value, + size_t size, int flags) { struct xfs_da_args args = { .dp = XFS_I(inode), diff --git a/fs/zonefs/Makefile b/fs/zonefs/Makefile index 75a380aa1ae1..33c1a4f1132e 100644 --- a/fs/zonefs/Makefile +++ b/fs/zonefs/Makefile @@ -1,4 +1,6 @@ # SPDX-License-Identifier: GPL-2.0 +ccflags-y += -I$(src) + obj-$(CONFIG_ZONEFS_FS) += zonefs.o zonefs-y := super.o diff --git a/fs/zonefs/super.c b/fs/zonefs/super.c index bec47f2d074b..049e36c69ed7 100644 --- a/fs/zonefs/super.c +++ b/fs/zonefs/super.c @@ -24,6 +24,9 @@ #include "zonefs.h" +#define CREATE_TRACE_POINTS +#include "trace.h" + static inline int zonefs_zone_mgmt(struct inode *inode, enum req_opf op) { @@ -32,6 +35,7 @@ static inline int zonefs_zone_mgmt(struct inode *inode, lockdep_assert_held(&zi->i_truncate_mutex); + trace_zonefs_zone_mgmt(inode, op); ret = blkdev_zone_mgmt(inode->i_sb->s_bdev, op, zi->i_zsector, zi->i_zone_size >> SECTOR_SHIFT, GFP_NOFS); if (ret) { @@ -100,6 +104,8 @@ static int zonefs_iomap_begin(struct inode *inode, loff_t offset, loff_t length, iomap->bdev = inode->i_sb->s_bdev; iomap->addr = (zi->i_zsector << SECTOR_SHIFT) + iomap->offset; + trace_zonefs_iomap_begin(inode, iomap); + return 0; } @@ -159,6 +165,21 @@ static int zonefs_writepages(struct address_space *mapping, return iomap_writepages(mapping, wbc, &wpc, &zonefs_writeback_ops); } +static int zonefs_swap_activate(struct swap_info_struct *sis, + struct file *swap_file, sector_t *span) +{ + struct inode *inode = file_inode(swap_file); + struct zonefs_inode_info *zi = ZONEFS_I(inode); + + if (zi->i_ztype != ZONEFS_ZTYPE_CNV) { + zonefs_err(inode->i_sb, + "swap file: not a conventional zone file\n"); + return -EINVAL; + } + + return iomap_swapfile_activate(sis, swap_file, span, &zonefs_iomap_ops); +} + static const struct address_space_operations zonefs_file_aops = { .readpage = zonefs_readpage, .readahead = zonefs_readahead, @@ -171,6 +192,7 @@ static const struct address_space_operations zonefs_file_aops = { .is_partially_uptodate = iomap_is_partially_uptodate, .error_remove_page = generic_error_remove_page, .direct_IO = noop_direct_IO, + .swap_activate = zonefs_swap_activate, }; static void zonefs_update_stats(struct inode *inode, loff_t new_isize) @@ -250,6 +272,9 @@ static loff_t zonefs_check_zone_condition(struct inode *inode, } inode->i_mode &= ~0222; return i_size_read(inode); + case BLK_ZONE_COND_FULL: + /* The write pointer of full zones is invalid. */ + return zi->i_max_size; default: if (zi->i_ztype == ZONEFS_ZTYPE_CNV) return zi->i_max_size; @@ -480,7 +505,8 @@ unlock: return ret; } -static int zonefs_inode_setattr(struct dentry *dentry, struct iattr *iattr) +static int zonefs_inode_setattr(struct user_namespace *mnt_userns, + struct dentry *dentry, struct iattr *iattr) { struct inode *inode = d_inode(dentry); int ret; @@ -488,7 +514,7 @@ static int zonefs_inode_setattr(struct dentry *dentry, struct iattr *iattr) if (unlikely(IS_IMMUTABLE(inode))) return -EPERM; - ret = setattr_prepare(dentry, iattr); + ret = setattr_prepare(&init_user_ns, dentry, iattr); if (ret) return ret; @@ -516,7 +542,7 @@ static int zonefs_inode_setattr(struct dentry *dentry, struct iattr *iattr) return ret; } - setattr_copy(inode, iattr); + setattr_copy(&init_user_ns, inode, iattr); return 0; } @@ -541,7 +567,7 @@ static int zonefs_file_fsync(struct file *file, loff_t start, loff_t end, if (ZONEFS_I(inode)->i_ztype == ZONEFS_ZTYPE_CNV) ret = file_write_and_wait_range(file, start, end); if (!ret) - ret = blkdev_issue_flush(inode->i_sb->s_bdev, GFP_KERNEL); + ret = blkdev_issue_flush(inode->i_sb->s_bdev); if (ret) zonefs_io_error(inode, true); @@ -674,11 +700,11 @@ static ssize_t zonefs_file_dio_append(struct kiocb *iocb, struct iov_iter *from) max = ALIGN_DOWN(max << SECTOR_SHIFT, inode->i_sb->s_blocksize); iov_iter_truncate(from, max); - nr_pages = iov_iter_npages(from, BIO_MAX_PAGES); + nr_pages = iov_iter_npages(from, BIO_MAX_VECS); if (!nr_pages) return 0; - bio = bio_alloc_bioset(GFP_NOFS, nr_pages, &fs_bio_set); + bio = bio_alloc(GFP_NOFS, nr_pages); if (!bio) return -ENOMEM; @@ -703,6 +729,7 @@ static ssize_t zonefs_file_dio_append(struct kiocb *iocb, struct iov_iter *from) ret = submit_bio_wait(bio); zonefs_file_write_dio_end_io(iocb, size, ret, 0); + trace_zonefs_file_dio_append(inode, size, ret); out_release: bio_release_pages(bio, false); @@ -717,6 +744,68 @@ out_release: } /* + * Do not exceed the LFS limits nor the file zone size. If pos is under the + * limit it becomes a short access. If it exceeds the limit, return -EFBIG. + */ +static loff_t zonefs_write_check_limits(struct file *file, loff_t pos, + loff_t count) +{ + struct inode *inode = file_inode(file); + struct zonefs_inode_info *zi = ZONEFS_I(inode); + loff_t limit = rlimit(RLIMIT_FSIZE); + loff_t max_size = zi->i_max_size; + + if (limit != RLIM_INFINITY) { + if (pos >= limit) { + send_sig(SIGXFSZ, current, 0); + return -EFBIG; + } + count = min(count, limit - pos); + } + + if (!(file->f_flags & O_LARGEFILE)) + max_size = min_t(loff_t, MAX_NON_LFS, max_size); + + if (unlikely(pos >= max_size)) + return -EFBIG; + + return min(count, max_size - pos); +} + +static ssize_t zonefs_write_checks(struct kiocb *iocb, struct iov_iter *from) +{ + struct file *file = iocb->ki_filp; + struct inode *inode = file_inode(file); + struct zonefs_inode_info *zi = ZONEFS_I(inode); + loff_t count; + + if (IS_SWAPFILE(inode)) + return -ETXTBSY; + + if (!iov_iter_count(from)) + return 0; + + if ((iocb->ki_flags & IOCB_NOWAIT) && !(iocb->ki_flags & IOCB_DIRECT)) + return -EINVAL; + + if (iocb->ki_flags & IOCB_APPEND) { + if (zi->i_ztype != ZONEFS_ZTYPE_SEQ) + return -EINVAL; + mutex_lock(&zi->i_truncate_mutex); + iocb->ki_pos = zi->i_wpoffset; + mutex_unlock(&zi->i_truncate_mutex); + } + + count = zonefs_write_check_limits(file, iocb->ki_pos, + iov_iter_count(from)); + if (count < 0) + return count; + + iov_iter_truncate(from, count); + return iov_iter_count(from); +} + +/* * Handle direct writes. For sequential zone files, this is the only possible * write path. For these files, check that the user is issuing writes * sequentially from the end of the file. This code assumes that the block layer @@ -733,8 +822,7 @@ static ssize_t zonefs_file_dio_write(struct kiocb *iocb, struct iov_iter *from) struct super_block *sb = inode->i_sb; bool sync = is_sync_kiocb(iocb); bool append = false; - size_t count; - ssize_t ret; + ssize_t ret, count; /* * For async direct IOs to sequential zone files, refuse IOCB_NOWAIT @@ -752,12 +840,11 @@ static ssize_t zonefs_file_dio_write(struct kiocb *iocb, struct iov_iter *from) inode_lock(inode); } - ret = generic_write_checks(iocb, from); - if (ret <= 0) + count = zonefs_write_checks(iocb, from); + if (count <= 0) { + ret = count; goto inode_unlock; - - iov_iter_truncate(from, zi->i_max_size - iocb->ki_pos); - count = iov_iter_count(from); + } if ((iocb->ki_pos | count) & (sb->s_blocksize - 1)) { ret = -EINVAL; @@ -780,7 +867,7 @@ static ssize_t zonefs_file_dio_write(struct kiocb *iocb, struct iov_iter *from) ret = zonefs_file_dio_append(iocb, from); else ret = iomap_dio_rw(iocb, from, &zonefs_iomap_ops, - &zonefs_write_dio_ops, sync); + &zonefs_write_dio_ops, 0); if (zi->i_ztype == ZONEFS_ZTYPE_SEQ && (ret > 0 || ret == -EIOCBQUEUED)) { if (ret > 0) @@ -817,12 +904,10 @@ static ssize_t zonefs_file_buffered_write(struct kiocb *iocb, inode_lock(inode); } - ret = generic_write_checks(iocb, from); + ret = zonefs_write_checks(iocb, from); if (ret <= 0) goto inode_unlock; - iov_iter_truncate(from, zi->i_max_size - iocb->ki_pos); - ret = iomap_file_buffered_write(iocb, from, &zonefs_iomap_ops); if (ret > 0) iocb->ki_pos += ret; @@ -917,7 +1002,7 @@ static ssize_t zonefs_file_read_iter(struct kiocb *iocb, struct iov_iter *to) } file_accessed(iocb->ki_filp); ret = iomap_dio_rw(iocb, to, &zonefs_iomap_ops, - &zonefs_read_dio_ops, is_sync_kiocb(iocb)); + &zonefs_read_dio_ops, 0); } else { ret = generic_file_read_iter(iocb, to); if (ret == -EIO) @@ -955,9 +1040,7 @@ static int zonefs_open_zone(struct inode *inode) mutex_lock(&zi->i_truncate_mutex); - zi->i_wr_refcnt++; - if (zi->i_wr_refcnt == 1) { - + if (!zi->i_wr_refcnt) { if (atomic_inc_return(&sbi->s_open_zones) > sbi->s_max_open_zones) { atomic_dec(&sbi->s_open_zones); ret = -EBUSY; @@ -967,7 +1050,6 @@ static int zonefs_open_zone(struct inode *inode) if (i_size_read(inode) < zi->i_max_size) { ret = zonefs_zone_mgmt(inode, REQ_OP_ZONE_OPEN); if (ret) { - zi->i_wr_refcnt--; atomic_dec(&sbi->s_open_zones); goto unlock; } @@ -975,6 +1057,8 @@ static int zonefs_open_zone(struct inode *inode) } } + zi->i_wr_refcnt++; + unlock: mutex_unlock(&zi->i_truncate_mutex); @@ -1223,7 +1307,7 @@ static void zonefs_init_dir_inode(struct inode *parent, struct inode *inode, struct super_block *sb = parent->i_sb; inode->i_ino = blkdev_nr_zones(sb->s_bdev->bd_disk) + type + 1; - inode_init_owner(inode, parent, S_IFDIR | 0555); + inode_init_owner(&init_user_ns, inode, parent, S_IFDIR | 0555); inode->i_op = &zonefs_dir_inode_operations; inode->i_fop = &simple_dir_operations; set_nlink(inode, 2); @@ -1581,12 +1665,11 @@ static int zonefs_fill_super(struct super_block *sb, void *data, int silent) sb->s_time_gran = 1; /* - * The block size is set to the device physical sector size to ensure - * that write operations on 512e devices (512B logical block and 4KB - * physical block) are always aligned to the device physical blocks, - * as mandated by the ZBC/ZAC specifications. + * The block size is set to the device zone write granularity to ensure + * that write operations are always aligned according to the device + * interface constraints. */ - sb_set_blocksize(sb, bdev_physical_block_size(sb->s_bdev)); + sb_set_blocksize(sb, bdev_zone_write_granularity(sb->s_bdev)); sbi->s_zone_sectors_shift = ilog2(bdev_zone_sectors(sb->s_bdev)); sbi->s_uid = GLOBAL_ROOT_UID; sbi->s_gid = GLOBAL_ROOT_GID; diff --git a/fs/zonefs/trace.h b/fs/zonefs/trace.h new file mode 100644 index 000000000000..f369d7d50303 --- /dev/null +++ b/fs/zonefs/trace.h @@ -0,0 +1,104 @@ +/* SPDX-License-Identifier: GPL-2.0 */ +/* + * zonefs filesystem driver tracepoints. + * + * Copyright (C) 2021 Western Digital Corporation or its affiliates. + */ + +#undef TRACE_SYSTEM +#define TRACE_SYSTEM zonefs + +#if !defined(_TRACE_ZONEFS_H) || defined(TRACE_HEADER_MULTI_READ) +#define _TRACE_ZONEFS_H + +#include <linux/tracepoint.h> +#include <linux/trace_seq.h> +#include <linux/blkdev.h> + +#include "zonefs.h" + +#define show_dev(dev) MAJOR(dev), MINOR(dev) + +TRACE_EVENT(zonefs_zone_mgmt, + TP_PROTO(struct inode *inode, enum req_opf op), + TP_ARGS(inode, op), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(ino_t, ino) + __field(int, op) + __field(sector_t, sector) + __field(sector_t, nr_sectors) + ), + TP_fast_assign( + __entry->dev = inode->i_sb->s_dev; + __entry->ino = inode->i_ino; + __entry->op = op; + __entry->sector = ZONEFS_I(inode)->i_zsector; + __entry->nr_sectors = + ZONEFS_I(inode)->i_zone_size >> SECTOR_SHIFT; + ), + TP_printk("bdev=(%d,%d), ino=%lu op=%s, sector=%llu, nr_sectors=%llu", + show_dev(__entry->dev), (unsigned long)__entry->ino, + blk_op_str(__entry->op), __entry->sector, + __entry->nr_sectors + ) +); + +TRACE_EVENT(zonefs_file_dio_append, + TP_PROTO(struct inode *inode, ssize_t size, ssize_t ret), + TP_ARGS(inode, size, ret), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(ino_t, ino) + __field(sector_t, sector) + __field(ssize_t, size) + __field(loff_t, wpoffset) + __field(ssize_t, ret) + ), + TP_fast_assign( + __entry->dev = inode->i_sb->s_dev; + __entry->ino = inode->i_ino; + __entry->sector = ZONEFS_I(inode)->i_zsector; + __entry->size = size; + __entry->wpoffset = ZONEFS_I(inode)->i_wpoffset; + __entry->ret = ret; + ), + TP_printk("bdev=(%d, %d), ino=%lu, sector=%llu, size=%zu, wpoffset=%llu, ret=%zu", + show_dev(__entry->dev), (unsigned long)__entry->ino, + __entry->sector, __entry->size, __entry->wpoffset, + __entry->ret + ) +); + +TRACE_EVENT(zonefs_iomap_begin, + TP_PROTO(struct inode *inode, struct iomap *iomap), + TP_ARGS(inode, iomap), + TP_STRUCT__entry( + __field(dev_t, dev) + __field(ino_t, ino) + __field(u64, addr) + __field(loff_t, offset) + __field(u64, length) + ), + TP_fast_assign( + __entry->dev = inode->i_sb->s_dev; + __entry->ino = inode->i_ino; + __entry->addr = iomap->addr; + __entry->offset = iomap->offset; + __entry->length = iomap->length; + ), + TP_printk("bdev=(%d,%d), ino=%lu, addr=%llu, offset=%llu, length=%llu", + show_dev(__entry->dev), (unsigned long)__entry->ino, + __entry->addr, __entry->offset, __entry->length + ) +); + +#endif /* _TRACE_ZONEFS_H */ + +#undef TRACE_INCLUDE_PATH +#define TRACE_INCLUDE_PATH . +#undef TRACE_INCLUDE_FILE +#define TRACE_INCLUDE_FILE trace + +/* This part must be outside protection */ +#include <trace/define_trace.h> |