[Devel] [RFC][PATCH 13/16] Remove proc_mnt's use for killing inodes
sukadev at us.ibm.com
sukadev at us.ibm.com
Wed May 23 18:14:17 PDT 2007
Subject: Remove proc_mnt's use for killing inodes
From: Dave Hansen <hansendc at us.ibm.com>
We use proc_mnt as a shortcut to find a superblock on which
to go killing /proc inodes. This will break if we ever have
more than one /proc superblock. So, use the superblock list
to go find each of the /proc sb's and kill inodes on each
superblock.
This does introduce an extra lock grab from what was there
before, but the list should be only 1 long 99% of the time,
and we don't exactly remove proc entries in hot paths. Note
that this *isn't* the path that we use to get rid of the
actual /proc pid entries. Those are a different beast.
Signed-off-by: Dave Hansen <haveblue at us.ibm.com>
---
fs/proc/generic.c | 28 +++++++++++++++++++++++-----
1 file changed, 23 insertions(+), 5 deletions(-)
Index: lx26-21-mm2/fs/proc/generic.c
===================================================================
--- lx26-21-mm2.orig/fs/proc/generic.c 2007-05-22 16:51:19.000000000 -0700
+++ lx26-21-mm2/fs/proc/generic.c 2007-05-22 16:59:55.000000000 -0700
@@ -554,13 +554,10 @@ static int proc_register(struct proc_dir
return 0;
}
-/*
- * Kill an inode that got unregistered..
- */
-static void proc_kill_inodes(struct proc_dir_entry *de)
+static void proc_kill_inodes_sb(struct proc_dir_entry *de,
+ struct super_block *sb)
{
struct list_head *p;
- struct super_block *sb = proc_mnt->mnt_sb;
/*
* Actually it's a partial revoke().
@@ -584,6 +581,27 @@ static void proc_kill_inodes(struct proc
file_list_unlock();
}
+/*
+ * Kill an inode that got unregistered..
+ */
+static void proc_kill_inodes(struct proc_dir_entry *de)
+{
+ struct list_head *l;
+ struct file_system_type *procfs;
+
+ procfs = get_fs_type("proc");
+ if (!procfs)
+ return;
+
+ spin_lock(&sb_lock);
+ list_for_each(l, &procfs->fs_supers) {
+ struct super_block *sb;
+ sb = list_entry(l, struct super_block, s_instances);
+ proc_kill_inodes_sb(de, sb);
+ }
+ spin_unlock(&sb_lock);
+}
+
static struct proc_dir_entry *proc_create(struct proc_dir_entry **parent,
const char *name,
mode_t mode,
_______________________________________________
Containers mailing list
Containers at lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/containers
More information about the Devel
mailing list