inotify: do not send a block of zeros when no pathname is available
Subsystems (2)
PATCH 1/3 inotify: do not send a block of zeros when no pathname is available
From: Brian Rogers <[email protected]>
When an event has no pathname, there's no need to pad it with a null byte and therefore generate an inotify_event sized block of zeros. This fixes a regression introduced by commit 0db501bd0610ee0c0aca84d927f90bcccd09e2bd where my system wouldn't finish booting because some process was being confused by this.
---
fs/notify/inotify/inotify_user.c | 8 +++++--- 1 files changed, 5 insertions(+), 3 deletions(-)
fs/notify/inotify/inotify_user.c +5 -3
@@ -180,7 +180,7 @@ static ssize_t copy_event_to_user(struct fsnotify_group *group,struct fsnotify_event_private_data *fsn_priv;struct inotify_event_private_data *priv;size_t event_size =sizeof (struct inotify_event ); -size_t name_len; +size_t name_len =0 ;/* we get the inotify watch descriptor from the event private data */ spin_lock (&event->lock);
@@ -196,10 +196,12 @@ static ssize_t copy_event_to_user(struct fsnotify_group *group,inotify_free_event_priv (fsn_priv); } -/* round up event->name_len so it is a multiple of event_size +/* +* round up event->name_len so it is a multiple of event_size * plus an extra byte for the terminating '\0'. */ - name_len =roundup (event->name_len +1 , event_size); +if (event->name_len) + name_len =roundup (event->name_len +1 , event_size); inotify_event.len = name_len; inotify_event.mask =inotify_mask_to_arg (event->mask);
PATCH 2/3 inotify: fix length reporting and size checking
0db501bd0610ee0c0 introduced a regresion in that it now sends a nul terminator but the length accounting when checking for space or reporting to userspace did not take this into account. This corrects all of the rounding logic.
---
fs/notify/inotify/inotify_user.c | 8 +++++--- 1 files changed, 5 insertions(+), 3 deletions(-)
fs/notify/inotify/inotify_user.c +5 -3
@@ -154,7 +154,8 @@ static struct fsnotify_event *get_one_event(struct fsnotify_group *group, event =fsnotify_peek_notify_event (group); - event_size +=roundup (event->name_len, event_size); +if (event->name_len) + event_size +=roundup (event->name_len +1 , event_size);if (event_size > count)return ERR_PTR (-EINVAL );
@@ -327,8 +328,9 @@ static long inotify_ioctl(struct file *file, unsigned int cmd,list_for_each_entry (holder, &group->notification_list, event_list) { event = holder->event; send_len +=sizeof (struct inotify_event ); - send_len +=roundup (event->name_len, -sizeof (struct inotify_event )); +if (event->name_len) + send_len +=roundup (event->name_len +1 , +sizeof (struct inotify_event )); }mutex_unlock (&group->notification_mutex); ret =put_user (send_len, (int __user *) p);
PATCH 3/3 inotify: update the group mask on mark addition
In reply to Eric ParisSeperating the addition and update of marks in inotify resulted in a regression in that inotify never gets events. The inotify group mask is always 0. This mask should be updated any time a new mark is added.
---
fs/notify/inotify/inotify_user.c | 4 ++++ 1 files changed, 4 insertions(+), 0 deletions(-)
fs/notify/inotify/inotify_user.c +4 -0
@@ -591,6 +591,10 @@ retry:/* match the ref from fsnotify_init_markentry() */ fsnotify_put_mark (&tmp_ientry->fsn_entry); +/* if this mark added a new event update the group mask */ +if (mask & ~group->mask) +fsnotify_recalc_group_mask (group); + out_err:if (ret <0 )kmem_cache_free (inotify_inode_mark_cachep, tmp_ientry);