| Message ID | 20171018205626.25391-1-mjg59@google.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
Matthew Garrett <mjg59@google.com> wrote: > commit e462ec50cb5fa ("VFS: Differentiate mount flags (MS_*) from > internal superblock flags") reworked the flags that are passed through > do mount, and in the process dropped MS_I_VERSION. Confusingly, things > continue working if the i_version mount option is used because that's > directly passed to ext4 which interprets it, while iversion is turned > into MS_I_VERSION and dropped on the floor. This breaks IMA and EVM and > my heart. Pass SB_I_VERSION through, and update the consumers to use it > instead of MS_I_VERSION. > > Signed-off-by: Matthew Garrett <mjg59@google.com> > Cc: David Howells <dhowells@redhat.com> > Cc: Alexander Viro <viro@zeniv.linux.org.uk> > Cc: stable@vger.kernel.org Reviewed-by: David Howells <dhowells@redhat.com>
On Wed, Oct 18, 2017 at 01:56:26PM -0700, Matthew Garrett wrote: > commit e462ec50cb5fa ("VFS: Differentiate mount flags (MS_*) from > internal superblock flags") reworked the flags that are passed through > do mount, and in the process dropped MS_I_VERSION. Confusingly, things > continue working if the i_version mount option is used because that's > directly passed to ext4 which interprets it, while iversion is turned > into MS_I_VERSION and dropped on the floor. This breaks IMA and EVM and > my heart. Pass SB_I_VERSION through, and update the consumers to use it > instead of MS_I_VERSION. Remaining bits applied on top of "vfs: fix mounting a filesystem with i_version". And AFAICS Cc:stable is wrong - e462ec50cb5fa is not in 4.13...
On Wed, Oct 18, 2017 at 3:52 PM, Al Viro <viro@zeniv.linux.org.uk> wrote: > On Wed, Oct 18, 2017 at 01:56:26PM -0700, Matthew Garrett wrote: >> commit e462ec50cb5fa ("VFS: Differentiate mount flags (MS_*) from >> internal superblock flags") reworked the flags that are passed through >> do mount, and in the process dropped MS_I_VERSION. Confusingly, things >> continue working if the i_version mount option is used because that's >> directly passed to ext4 which interprets it, while iversion is turned >> into MS_I_VERSION and dropped on the floor. This breaks IMA and EVM and >> my heart. Pass SB_I_VERSION through, and update the consumers to use it >> instead of MS_I_VERSION. > > Remaining bits applied on top of "vfs: fix mounting a filesystem with i_version". > And AFAICS Cc:stable is wrong - e462ec50cb5fa is not in 4.13... Yes, my mistake. Sorry about that.
diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index 35a128acfbd1..161694b66038 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -1135,7 +1135,7 @@ static int btrfs_fill_super(struct super_block *sb, #ifdef CONFIG_BTRFS_FS_POSIX_ACL sb->s_flags |= MS_POSIXACL; #endif - sb->s_flags |= MS_I_VERSION; + sb->s_flags |= SB_I_VERSION; sb->s_iflags |= SB_I_CGROUPWB; err = super_setup_bdi(sb); diff --git a/fs/ext4/super.c b/fs/ext4/super.c index b104096fce9e..b0915b734a38 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -1677,7 +1677,7 @@ static int handle_mount_opt(struct super_block *sb, char *opt, int token, sbi->s_mount_flags |= EXT4_MF_FS_ABORTED; return 1; case Opt_i_version: - sb->s_flags |= MS_I_VERSION; + sb->s_flags |= SB_I_VERSION; return 1; case Opt_lazytime: sb->s_flags |= MS_LAZYTIME; @@ -2060,7 +2060,7 @@ static int _ext4_show_options(struct seq_file *seq, struct super_block *sb, SEQ_OPTS_PRINT("min_batch_time=%u", sbi->s_min_batch_time); if (nodefs || sbi->s_max_batch_time != EXT4_DEF_MAX_BATCH_TIME) SEQ_OPTS_PRINT("max_batch_time=%u", sbi->s_max_batch_time); - if (sb->s_flags & MS_I_VERSION) + if (sb->s_flags & SB_I_VERSION) SEQ_OPTS_PUTS("i_version"); if (nodefs || sbi->s_stripe) SEQ_OPTS_PRINT("stripe=%lu", sbi->s_stripe); diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index 65c88379a3a1..94a745acaef8 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -1059,7 +1059,7 @@ static int fuse_fill_super(struct super_block *sb, void *data, int silent) if (sb->s_flags & MS_MANDLOCK) goto err; - sb->s_flags &= ~(MS_NOSEC | MS_I_VERSION); + sb->s_flags &= ~(MS_NOSEC | SB_I_VERSION); if (!parse_fuse_opt(data, &d, is_bdev)) goto err; diff --git a/fs/namespace.c b/fs/namespace.c index 3b601f115b6c..d18deb4c410b 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -2825,7 +2825,8 @@ long do_mount(const char *dev_name, const char __user *dir_name, SB_MANDLOCK | SB_DIRSYNC | SB_SILENT | - SB_POSIXACL); + SB_POSIXACL | + SB_I_VERSION); if (flags & MS_REMOUNT) retval = do_remount(&path, flags, sb_flags, mnt_flags, diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index 584cf2d573ba..f663022353c0 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -1637,7 +1637,7 @@ xfs_fs_fill_super( /* version 5 superblocks support inode version counters. */ if (XFS_SB_VERSION_NUM(&mp->m_sb) == XFS_SB_VERSION_5) - sb->s_flags |= MS_I_VERSION; + sb->s_flags |= SB_I_VERSION; if (mp->m_flags & XFS_MOUNT_DAX) { xfs_warn(mp,
commit e462ec50cb5fa ("VFS: Differentiate mount flags (MS_*) from internal superblock flags") reworked the flags that are passed through do mount, and in the process dropped MS_I_VERSION. Confusingly, things continue working if the i_version mount option is used because that's directly passed to ext4 which interprets it, while iversion is turned into MS_I_VERSION and dropped on the floor. This breaks IMA and EVM and my heart. Pass SB_I_VERSION through, and update the consumers to use it instead of MS_I_VERSION. Signed-off-by: Matthew Garrett <mjg59@google.com> Cc: David Howells <dhowells@redhat.com> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Cc: stable@vger.kernel.org --- fs/btrfs/super.c | 2 +- fs/ext4/super.c | 4 ++-- fs/fuse/inode.c | 2 +- fs/namespace.c | 3 ++- fs/xfs/xfs_super.c | 2 +- 5 files changed, 7 insertions(+), 6 deletions(-)