Message ID | 20170816154127.7048-17-jack@suse.cz (mailing list archive) |
---|---|
State | New, archived |
Headers | show |
On Aug 16, 2017, at 9:41 AM, Jan Kara <jack@suse.cz> wrote: > > dqi_flags modifications are protected by dq_data_lock. It would be useful to annotate dqi_flags with a comment that dq_data_lock is protecting it (along with any other fields similarly protected). > However the > modifications in vfs_load_quota_inode() and in mark_info_dirty() were > not which could lead to corruption of dqi_flags. Since modifications to > dqi_flags are rare, this is hard to observe in practice but in theory it > could happen. Fix the problem by always using dq_data_lock for > protection. > What about v1_write_file_info()? It is also modifying dqi_flags, but it doesn't get dq_data_lock either. > Signed-off-by: Jan Kara <jack@suse.cz> > --- > fs/quota/dquot.c | 9 +++++++-- > 1 file changed, 7 insertions(+), 2 deletions(-) > > diff --git a/fs/quota/dquot.c b/fs/quota/dquot.c > index 5e77c4da69a6..e1a155e8db15 100644 > --- a/fs/quota/dquot.c > +++ b/fs/quota/dquot.c > @@ -389,7 +389,9 @@ static inline int clear_dquot_dirty(struct dquot *dquot) > > void mark_info_dirty(struct super_block *sb, int type) > { > - set_bit(DQF_INFO_DIRTY_B, &sb_dqopt(sb)->info[type].dqi_flags); > + spin_lock(&dq_data_lock); > + sb_dqopt(sb)->info[type].dqi_flags |= DQF_INFO_DIRTY; > + spin_unlock(&dq_data_lock); > } > EXPORT_SYMBOL(mark_info_dirty); > > @@ -2316,8 +2318,11 @@ static int vfs_load_quota_inode(struct inode *inode, int type, int format_id, > error = dqopt->ops[type]->read_file_info(sb, type); > if (error < 0) > goto out_file_init; > - if (dqopt->flags & DQUOT_QUOTA_SYS_FILE) > + if (dqopt->flags & DQUOT_QUOTA_SYS_FILE) { > + spin_lock(&dq_data_lock); > dqopt->info[type].dqi_flags |= DQF_SYS_FILE; > + spin_unlock(&dq_data_lock); > + } > spin_lock(&dq_state_lock); > dqopt->flags |= dquot_state_flag(flags, type); > spin_unlock(&dq_state_lock); > -- > 2.12.3 > Cheers, Andreas
diff --git a/fs/quota/dquot.c b/fs/quota/dquot.c index 5e77c4da69a6..e1a155e8db15 100644 --- a/fs/quota/dquot.c +++ b/fs/quota/dquot.c @@ -389,7 +389,9 @@ static inline int clear_dquot_dirty(struct dquot *dquot) void mark_info_dirty(struct super_block *sb, int type) { - set_bit(DQF_INFO_DIRTY_B, &sb_dqopt(sb)->info[type].dqi_flags); + spin_lock(&dq_data_lock); + sb_dqopt(sb)->info[type].dqi_flags |= DQF_INFO_DIRTY; + spin_unlock(&dq_data_lock); } EXPORT_SYMBOL(mark_info_dirty); @@ -2316,8 +2318,11 @@ static int vfs_load_quota_inode(struct inode *inode, int type, int format_id, error = dqopt->ops[type]->read_file_info(sb, type); if (error < 0) goto out_file_init; - if (dqopt->flags & DQUOT_QUOTA_SYS_FILE) + if (dqopt->flags & DQUOT_QUOTA_SYS_FILE) { + spin_lock(&dq_data_lock); dqopt->info[type].dqi_flags |= DQF_SYS_FILE; + spin_unlock(&dq_data_lock); + } spin_lock(&dq_state_lock); dqopt->flags |= dquot_state_flag(flags, type); spin_unlock(&dq_state_lock);
dqi_flags modifications are protected by dq_data_lock. However the modifications in vfs_load_quota_inode() and in mark_info_dirty() were not which could lead to corruption of dqi_flags. Since modifications to dqi_flags are rare, this is hard to observe in practice but in theory it could happen. Fix the problem by always using dq_data_lock for protection. Signed-off-by: Jan Kara <jack@suse.cz> --- fs/quota/dquot.c | 9 +++++++-- 1 file changed, 7 insertions(+), 2 deletions(-)