1
linux/fs/attr.c
Andi Kleen 69b4573296 Cache xattr security drop check for write v2
Some recent benchmarking on btrfs showed that a major scaling bottleneck
on large systems on btrfs is currently the xattr lookup on every write.

Why xattr lookup on every write I hear you ask?

write wants to drop suid and security related xattrs that could set o
capabilities for executables.  To do that it currently looks up
security.capability on EVERY write (even for non executables) to decide
whether to drop it or not.

In btrfs this causes an additional tree walk, hitting some per file system
locks and quite bad scalability. In a simple read workload on a 8S
system I saw over 90% CPU time in spinlocks related to that.

Chris Mason tells me this is also a problem in ext4, where it hits
the global mbcache lock.

This patch adds a simple per inode to avoid this problem.  We only
do the lookup once per file and then if there is no xattr cache
the decision. All xattr changes clear the flag.

I also used the same flag to avoid the suid check, although
that one is pretty cheap.

A file system can also set this flag when it creates the inode,
if it has a cheap way to do so.  This is done for some common file systems
in followon patches.

With this patch a major part of the lock contention disappears
for btrfs. Some testing on smaller systems didn't show significant
performance changes, but at least it helps the larger systems
and is generally more efficient.

v2: Rename is_sgid. add file system helper.
Cc: chris.mason@oracle.com
Cc: josef@redhat.com
Cc: viro@zeniv.linux.org.uk
Cc: agruen@linbit.com
Cc: Serge E. Hallyn <serue@us.ibm.com>
Signed-off-by: Andi Kleen <ak@linux.intel.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2011-05-28 12:02:09 -04:00

253 lines
6.8 KiB
C

/*
* linux/fs/attr.c
*
* Copyright (C) 1991, 1992 Linus Torvalds
* changes by Thomas Schoebel-Theuer
*/
#include <linux/module.h>
#include <linux/time.h>
#include <linux/mm.h>
#include <linux/string.h>
#include <linux/capability.h>
#include <linux/fsnotify.h>
#include <linux/fcntl.h>
#include <linux/security.h>
/**
* inode_change_ok - check if attribute changes to an inode are allowed
* @inode: inode to check
* @attr: attributes to change
*
* Check if we are allowed to change the attributes contained in @attr
* in the given inode. This includes the normal unix access permission
* checks, as well as checks for rlimits and others.
*
* Should be called as the first thing in ->setattr implementations,
* possibly after taking additional locks.
*/
int inode_change_ok(const struct inode *inode, struct iattr *attr)
{
unsigned int ia_valid = attr->ia_valid;
/*
* First check size constraints. These can't be overriden using
* ATTR_FORCE.
*/
if (ia_valid & ATTR_SIZE) {
int error = inode_newsize_ok(inode, attr->ia_size);
if (error)
return error;
}
/* If force is set do it anyway. */
if (ia_valid & ATTR_FORCE)
return 0;
/* Make sure a caller can chown. */
if ((ia_valid & ATTR_UID) &&
(current_fsuid() != inode->i_uid ||
attr->ia_uid != inode->i_uid) && !capable(CAP_CHOWN))
return -EPERM;
/* Make sure caller can chgrp. */
if ((ia_valid & ATTR_GID) &&
(current_fsuid() != inode->i_uid ||
(!in_group_p(attr->ia_gid) && attr->ia_gid != inode->i_gid)) &&
!capable(CAP_CHOWN))
return -EPERM;
/* Make sure a caller can chmod. */
if (ia_valid & ATTR_MODE) {
if (!inode_owner_or_capable(inode))
return -EPERM;
/* Also check the setgid bit! */
if (!in_group_p((ia_valid & ATTR_GID) ? attr->ia_gid :
inode->i_gid) && !capable(CAP_FSETID))
attr->ia_mode &= ~S_ISGID;
}
/* Check for setting the inode time. */
if (ia_valid & (ATTR_MTIME_SET | ATTR_ATIME_SET | ATTR_TIMES_SET)) {
if (!inode_owner_or_capable(inode))
return -EPERM;
}
return 0;
}
EXPORT_SYMBOL(inode_change_ok);
/**
* inode_newsize_ok - may this inode be truncated to a given size
* @inode: the inode to be truncated
* @offset: the new size to assign to the inode
* @Returns: 0 on success, -ve errno on failure
*
* inode_newsize_ok must be called with i_mutex held.
*
* inode_newsize_ok will check filesystem limits and ulimits to check that the
* new inode size is within limits. inode_newsize_ok will also send SIGXFSZ
* when necessary. Caller must not proceed with inode size change if failure is
* returned. @inode must be a file (not directory), with appropriate
* permissions to allow truncate (inode_newsize_ok does NOT check these
* conditions).
*/
int inode_newsize_ok(const struct inode *inode, loff_t offset)
{
if (inode->i_size < offset) {
unsigned long limit;
limit = rlimit(RLIMIT_FSIZE);
if (limit != RLIM_INFINITY && offset > limit)
goto out_sig;
if (offset > inode->i_sb->s_maxbytes)
goto out_big;
} else {
/*
* truncation of in-use swapfiles is disallowed - it would
* cause subsequent swapout to scribble on the now-freed
* blocks.
*/
if (IS_SWAPFILE(inode))
return -ETXTBSY;
}
return 0;
out_sig:
send_sig(SIGXFSZ, current, 0);
out_big:
return -EFBIG;
}
EXPORT_SYMBOL(inode_newsize_ok);
/**
* setattr_copy - copy simple metadata updates into the generic inode
* @inode: the inode to be updated
* @attr: the new attributes
*
* setattr_copy must be called with i_mutex held.
*
* setattr_copy updates the inode's metadata with that specified
* in attr. Noticeably missing is inode size update, which is more complex
* as it requires pagecache updates.
*
* The inode is not marked as dirty after this operation. The rationale is
* that for "simple" filesystems, the struct inode is the inode storage.
* The caller is free to mark the inode dirty afterwards if needed.
*/
void setattr_copy(struct inode *inode, const struct iattr *attr)
{
unsigned int ia_valid = attr->ia_valid;
if (ia_valid & ATTR_UID)
inode->i_uid = attr->ia_uid;
if (ia_valid & ATTR_GID)
inode->i_gid = attr->ia_gid;
if (ia_valid & ATTR_ATIME)
inode->i_atime = timespec_trunc(attr->ia_atime,
inode->i_sb->s_time_gran);
if (ia_valid & ATTR_MTIME)
inode->i_mtime = timespec_trunc(attr->ia_mtime,
inode->i_sb->s_time_gran);
if (ia_valid & ATTR_CTIME)
inode->i_ctime = timespec_trunc(attr->ia_ctime,
inode->i_sb->s_time_gran);
if (ia_valid & ATTR_MODE) {
umode_t mode = attr->ia_mode;
if (!in_group_p(inode->i_gid) && !capable(CAP_FSETID))
mode &= ~S_ISGID;
inode->i_mode = mode;
}
}
EXPORT_SYMBOL(setattr_copy);
int notify_change(struct dentry * dentry, struct iattr * attr)
{
struct inode *inode = dentry->d_inode;
mode_t mode = inode->i_mode;
int error;
struct timespec now;
unsigned int ia_valid = attr->ia_valid;
if (ia_valid & (ATTR_MODE | ATTR_UID | ATTR_GID | ATTR_TIMES_SET)) {
if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
return -EPERM;
}
if ((ia_valid & ATTR_MODE)) {
mode_t amode = attr->ia_mode;
/* Flag setting protected by i_mutex */
if (is_sxid(amode))
inode->i_flags &= ~S_NOSEC;
}
now = current_fs_time(inode->i_sb);
attr->ia_ctime = now;
if (!(ia_valid & ATTR_ATIME_SET))
attr->ia_atime = now;
if (!(ia_valid & ATTR_MTIME_SET))
attr->ia_mtime = now;
if (ia_valid & ATTR_KILL_PRIV) {
attr->ia_valid &= ~ATTR_KILL_PRIV;
ia_valid &= ~ATTR_KILL_PRIV;
error = security_inode_need_killpriv(dentry);
if (error > 0)
error = security_inode_killpriv(dentry);
if (error)
return error;
}
/*
* We now pass ATTR_KILL_S*ID to the lower level setattr function so
* that the function has the ability to reinterpret a mode change
* that's due to these bits. This adds an implicit restriction that
* no function will ever call notify_change with both ATTR_MODE and
* ATTR_KILL_S*ID set.
*/
if ((ia_valid & (ATTR_KILL_SUID|ATTR_KILL_SGID)) &&
(ia_valid & ATTR_MODE))
BUG();
if (ia_valid & ATTR_KILL_SUID) {
if (mode & S_ISUID) {
ia_valid = attr->ia_valid |= ATTR_MODE;
attr->ia_mode = (inode->i_mode & ~S_ISUID);
}
}
if (ia_valid & ATTR_KILL_SGID) {
if ((mode & (S_ISGID | S_IXGRP)) == (S_ISGID | S_IXGRP)) {
if (!(ia_valid & ATTR_MODE)) {
ia_valid = attr->ia_valid |= ATTR_MODE;
attr->ia_mode = inode->i_mode;
}
attr->ia_mode &= ~S_ISGID;
}
}
if (!(attr->ia_valid & ~(ATTR_KILL_SUID | ATTR_KILL_SGID)))
return 0;
error = security_inode_setattr(dentry, attr);
if (error)
return error;
if (ia_valid & ATTR_SIZE)
down_write(&dentry->d_inode->i_alloc_sem);
if (inode->i_op->setattr)
error = inode->i_op->setattr(dentry, attr);
else
error = simple_setattr(dentry, attr);
if (ia_valid & ATTR_SIZE)
up_write(&dentry->d_inode->i_alloc_sem);
if (!error)
fsnotify_change(dentry, ia_valid);
return error;
}
EXPORT_SYMBOL(notify_change);