sanitize vfsmount refcounting changes
Instead of splitting refcount between (per-cpu) mnt_count
and (SMP-only) mnt_longrefs, make all references contribute
to mnt_count again and keep track of how many are longterm
ones.
Accounting rules for longterm count:
* 1 for each fs_struct.root.mnt
* 1 for each fs_struct.pwd.mnt
* 1 for having non-NULL ->mnt_ns
* decrement to 0 happens only under vfsmount lock exclusive
That allows nice common case for mntput() - since we can't drop the
final reference until after mnt_longterm has reached 0 due to the rules
above, mntput() can grab vfsmount lock shared and check mnt_longterm.
If it turns out to be non-zero (which is the common case), we know
that this is not the final mntput() and can just blindly decrement
percpu mnt_count. Otherwise we grab vfsmount lock exclusive and
do usual decrement-and-check of percpu mnt_count.
For fs_struct.c we have mnt_make_longterm() and mnt_make_shortterm();
namespace.c uses the latter in places where we don't already hold
vfsmount lock exclusive and opencodes a few remaining spots where
we need to manipulate mnt_longterm.
Note that we mostly revert the code outside of fs/namespace.c back
to what we used to have; in particular, normal code doesn't need
to care about two kinds of references, etc. And we get to keep
the optimization Nick's variant had bought us...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
diff --git a/fs/fs_struct.c b/fs/fs_struct.c
index 68ca487..78b519c 100644
--- a/fs/fs_struct.c
+++ b/fs/fs_struct.c
@@ -4,6 +4,19 @@
#include <linux/path.h>
#include <linux/slab.h>
#include <linux/fs_struct.h>
+#include "internal.h"
+
+static inline void path_get_longterm(struct path *path)
+{
+ path_get(path);
+ mnt_make_longterm(path->mnt);
+}
+
+static inline void path_put_longterm(struct path *path)
+{
+ mnt_make_shortterm(path->mnt);
+ path_put(path);
+}
/*
* Replace the fs->{rootmnt,root} with {mnt,dentry}. Put the old values.
@@ -17,11 +30,11 @@
write_seqcount_begin(&fs->seq);
old_root = fs->root;
fs->root = *path;
- path_get_long(path);
+ path_get_longterm(path);
write_seqcount_end(&fs->seq);
spin_unlock(&fs->lock);
if (old_root.dentry)
- path_put_long(&old_root);
+ path_put_longterm(&old_root);
}
/*
@@ -36,12 +49,12 @@
write_seqcount_begin(&fs->seq);
old_pwd = fs->pwd;
fs->pwd = *path;
- path_get_long(path);
+ path_get_longterm(path);
write_seqcount_end(&fs->seq);
spin_unlock(&fs->lock);
if (old_pwd.dentry)
- path_put_long(&old_pwd);
+ path_put_longterm(&old_pwd);
}
void chroot_fs_refs(struct path *old_root, struct path *new_root)
@@ -59,13 +72,13 @@
write_seqcount_begin(&fs->seq);
if (fs->root.dentry == old_root->dentry
&& fs->root.mnt == old_root->mnt) {
- path_get_long(new_root);
+ path_get_longterm(new_root);
fs->root = *new_root;
count++;
}
if (fs->pwd.dentry == old_root->dentry
&& fs->pwd.mnt == old_root->mnt) {
- path_get_long(new_root);
+ path_get_longterm(new_root);
fs->pwd = *new_root;
count++;
}
@@ -76,13 +89,13 @@
} while_each_thread(g, p);
read_unlock(&tasklist_lock);
while (count--)
- path_put_long(old_root);
+ path_put_longterm(old_root);
}
void free_fs_struct(struct fs_struct *fs)
{
- path_put_long(&fs->root);
- path_put_long(&fs->pwd);
+ path_put_longterm(&fs->root);
+ path_put_longterm(&fs->pwd);
kmem_cache_free(fs_cachep, fs);
}
@@ -118,9 +131,9 @@
spin_lock(&old->lock);
fs->root = old->root;
- path_get_long(&fs->root);
+ path_get_longterm(&fs->root);
fs->pwd = old->pwd;
- path_get_long(&fs->pwd);
+ path_get_longterm(&fs->pwd);
spin_unlock(&old->lock);
}
return fs;