Message-ID: <aWktm4vdzxF0b596@CMGLRV3>
Date: Thu, 15 Jan 2026 12:10:35 -0600
From: Frederick Lawler <fred@...udflare.com>
To: Roberto Sassu <roberto.sassu@...weicloud.com>
Cc: Mimi Zohar <zohar@...ux.ibm.com>,
	Roberto Sassu <roberto.sassu@...wei.com>,
	Dmitry Kasatkin <dmitry.kasatkin@...il.com>,
	Eric Snowberg <eric.snowberg@...cle.com>,
	Paul Moore <paul@...l-moore.com>, James Morris <jmorris@...ei.org>,
	"Serge E. Hallyn" <serge@...lyn.com>,
	"Darrick J. Wong" <djwong@...nel.org>,
	Christian Brauner <brauner@...nel.org>,
	Josef Bacik <josef@...icpanda.com>,
	Jeff Layton <jlayton@...nel.org>, linux-kernel@...r.kernel.org,
	linux-integrity@...r.kernel.org,
	linux-security-module@...r.kernel.org, kernel-team@...udflare.com
Subject: Re: [PATCH] ima: Detect changes to files via kstat changes rather
 than i_version

On Thu, Jan 15, 2026 at 12:46:37PM +0100, Roberto Sassu wrote:
> On Mon, 2026-01-12 at 16:32 -0600, Frederick Lawler wrote:
> > Commit 1cf7e834a6fb ("xfs: switch to multigrain timestamps")
> > introduced a means to track change detection for an inode
> > via ctime updates, opposed to setting kstat.change_cookie when
> > calling into xfs_vn_getattr().
> > 
> > This introduced a regression because IMA caches kstat.change_cookie
> > to compare against an inode's i_version directly in
> > integrity_inode_attrs_changed(), and thus could be out of date
> > depending on how file systems increment i_version.
> > 
> > To address this, require integrity_inode_attrs_changed() to query
> > vfs_getattr_nosec() to compare the cached version against
> > kstat.change_cookie directly. This ensures that when updates occur,
> > we're accessing the same changed inode version on changes, and fallback
> > to compare against an artificial version generated from kstat.ctime
> > via integrity_ctime_guard() when there's no detected change
> > to the kstat.change_cookie.
> > 
> > This ensures that in the absence of i_version support for file systems,
> > and in the absence of a kstat.change_cookie update, we ultimately have a
> > unique-enough version to compare against.
> > 
> > The exact implementation for integrity_ctime_guard() is to ensure that
> > if tv_sec or tv_nsec are zero, there's some value to store back into
> > struct integrity_inode_attributes.version. This also avoids the need to
> > add additional storage and comparisons.
> > 
> > Lastly, because EVM still relies on querying and caching a backing inode's
> > i_version, the integrity_inode_attrs_changed() falls back to the
> > original inode.i_version != cached comparison. This maintains the
> > invariant that a re-evaluation in unknown change detection circumstances
> > is required.
> > 
> > Link: https://lore.kernel.org/all/aTspr4_h9IU4EyrR@CMGLRV3
> > Suggested-by: Jeff Layton <jlayton@...nel.org>
> > Signed-off-by: Frederick Lawler <fred@...udflare.com>
> > ---
> > We uncovered a case in kernels >= 6.13 where XFS is no longer updating
> > struct kstat.change_cookie on i_op getattr() access calls. Instead, XFS is
> > using multigrain ctime (as well as other file systems) for
> > change detection in commit 1cf7e834a6fb ("xfs: switch to
> > multigrain timestamps").
> > 
> > Because file systems may implement i_version as they see fit, IMA
> > caching may be behind as well as file systems that don't support/export
> > i_version. Thus we're proposing to compare against the kstat.change_cookie
> > directly to the cached version, and fall back to a ctime guard when
> > that's not updated.
> > 
> > EVM is largely left alone since there's no trivial way to query a file
> > directly in the LSM call paths to obtain kstat.change_cookie &
> > kstat.ctime to cache. Thus retains accessing i_version directly.
> > 
> > Regression tests will be added to the Linux Test Project instead of
> > selftest to help catch future file system changes that may impact
> > future evaluation of IMA.
> > 
> > I'd like this to be backported to at least 6.18 if possible.
> > 
> > Below is a simplified test that demonstrates the issue:
> > 
> > _fragment.config_
> > CONFIG_XFS_FS=y
> > CONFIG_OVERLAY_FS=y
> > CONFIG_IMA=y
> > CONFIG_IMA_WRITE_POLICY=y
> > CONFIG_IMA_READ_POLICY=y
> > 
> > _./test.sh_
> > 
> > IMA_POLICY="/sys/kernel/security/ima/policy"
> > TEST_BIN="/bin/date"
> > MNT_BASE="/tmp/ima_test_root"
> > 
> > mkdir -p "$MNT_BASE"
> > mount -t tmpfs tmpfs "$MNT_BASE"
> > mkdir -p "$MNT_BASE"/{xfs_disk,upper,work,ovl}
> > 
> > dd if=/dev/zero of="$MNT_BASE/xfs.img" bs=1M count=300
> > mkfs.xfs -q "$MNT_BASE/xfs.img"
> > mount "$MNT_BASE/xfs.img" "$MNT_BASE/xfs_disk"
> > cp "$TEST_BIN" "$MNT_BASE/xfs_disk/test_prog"
> > 
> > mount -t overlay overlay -o \
> > "lowerdir=$MNT_BASE/xfs_disk,upperdir=$MNT_BASE/upper,workdir=$MNT_BASE/work" \
> > "$MNT_BASE/ovl"
> > 
> > echo "audit func=BPRM_CHECK uid=$(id -u nobody)" > "$IMA_POLICY"
> > 
> > target_prog="$MNT_BASE/ovl/test_prog"
> > setpriv --reuid nobody "$target_prog"
> > setpriv --reuid nobody "$target_prog"
> > setpriv --reuid nobody "$target_prog"
> > 
> > audit_count=$(dmesg | grep -c "file=\"$target_prog\"")
> > 
> > if [[ "$audit_count" -eq 1 ]]; then
> >         echo "PASS: Found exactly 1 audit event."
> > else
> >         echo "FAIL: Expected 1 audit event, but found $audit_count."
> >         exit 1
> > fi
> > ---
> > Changes since RFC:
> > - Remove calls to I_IS_VERSION()
> > - Function documentation/comments
> > - Abide IMA/EVM change detection fallback invariants
> > - Combined ctime guard into version for attributes struct
> > - Link to RFC: https://lore.kernel.org/r/20251229-xfs-ima-fixup-v1-1-6a717c939f7c@cloudflare.com
> > ---
> >  include/linux/integrity.h         | 42 +++++++++++++++++++++++++++++++++++----
> >  security/integrity/evm/evm_main.c |  5 ++---
> >  security/integrity/ima/ima_api.c  | 11 +++++++---
> >  security/integrity/ima/ima_main.c | 15 +++++---------
> >  4 files changed, 53 insertions(+), 20 deletions(-)
> > 
> > diff --git a/include/linux/integrity.h b/include/linux/integrity.h
> > index f5842372359be5341b6870a43b92e695e8fc78af..5eca8aa2769f9238c68bb40885ecc46910524f11 100644
> > --- a/include/linux/integrity.h
> > +++ b/include/linux/integrity.h
> > @@ -9,6 +9,7 @@
> >  
> >  #include <linux/fs.h>
> >  #include <linux/iversion.h>
> > +#include <linux/kernel.h>
> >  
> >  enum integrity_status {
> >  	INTEGRITY_PASS = 0,
> > @@ -36,6 +37,14 @@ struct integrity_inode_attributes {
> >  	dev_t dev;
> >  };
> >  
> > +/*
> > + * Wrapper to generate an artificial version for a file.
> > + */
> > +static inline u64 integrity_ctime_guard(struct kstat stat)
> > +{
> > +	return stat.ctime.tv_sec ^ stat.ctime.tv_nsec;
> 
> Unfortunately, we cannot take the risk of a collision. Better use all
> or a packed version.

Sounds good.

> 
> > +}
> > +
> >  /*
> >   * On stacked filesystems the i_version alone is not enough to detect file data
> >   * or metadata change. Additional metadata is required.
> > @@ -51,14 +60,39 @@ integrity_inode_attrs_store(struct integrity_inode_attributes *attrs,
> >  
> >  /*
> >   * On stacked filesystems detect whether the inode or its content has changed.
> > + *
> > + * Must be called in process context.
> >   */
> >  static inline bool
> >  integrity_inode_attrs_changed(const struct integrity_inode_attributes *attrs,
> > -			      const struct inode *inode)
> > +			      struct file *file, struct inode *inode)
> >  {
> > -	return (inode->i_sb->s_dev != attrs->dev ||
> > -		inode->i_ino != attrs->ino ||
> > -		!inode_eq_iversion(inode, attrs->version));
> > +	struct kstat stat;
> > +
> > +	might_sleep();
> > +
> > +	if (inode->i_sb->s_dev != attrs->dev || inode->i_ino != attrs->ino)
> > +		return true;
> > +
> > +	/*
> > +	 * EVM currently relies on backing inode i_version. While IS_I_VERSION
> > +	 * is not a good indicator of i_version support, this still retains
> > +	 * the logic such that a re-evaluation should still occur for EVM, and
> > +	 * only for IMA if vfs_getattr_nosec() fails.
> > +	 */
> > +	if (!file || vfs_getattr_nosec(&file->f_path, &stat,
> > +				       STATX_CHANGE_COOKIE | STATX_CTIME,
> > +				       AT_STATX_SYNC_AS_STAT))
> > +		return !IS_I_VERSION(inode) ||
> > +			!inode_eq_iversion(inode, attrs->version);
> > +
> > +	if (stat.result_mask & STATX_CHANGE_COOKIE)
> > +		return stat.change_cookie != attrs->version;
> > +
> > +	if (stat.result_mask & STATX_CTIME)
> > +		return integrity_ctime_guard(stat) != attrs->version;
> 
> Yes, switching to the new field I guess it works, but I'm wondering if
> we could have more uniformity across the filesystems, otherwise one has
> to use one source for filesystem X, another source for filesystem Y.

Agreed. But I'm under the impression from casual searching, that most
file systems are likely to support ctime, than setting the change cookie
with an i_version or even having/updating i_version consistently.

Is there someone we could CC in here to get another opinion?

> 
> Thanks
> 
> Roberto
> 
> > +
> > +	return true;
> >  }
> >  
> >  
> > diff --git a/security/integrity/evm/evm_main.c b/security/integrity/evm/evm_main.c
> > index 73d500a375cb37a54f295b0e1e93fd6e5d9ecddc..6a4e0e246005246d5700b1db590c1759242b9cb6 100644
> > --- a/security/integrity/evm/evm_main.c
> > +++ b/security/integrity/evm/evm_main.c
> > @@ -752,9 +752,8 @@ bool evm_metadata_changed(struct inode *inode, struct inode *metadata_inode)
> >  	bool ret = false;
> >  
> >  	if (iint) {
> > -		ret = (!IS_I_VERSION(metadata_inode) ||
> > -		       integrity_inode_attrs_changed(&iint->metadata_inode,
> > -						     metadata_inode));
> > +		ret = integrity_inode_attrs_changed(&iint->metadata_inode,
> > +						    NULL, metadata_inode);
> >  		if (ret)
> >  			iint->evm_status = INTEGRITY_UNKNOWN;
> >  	}
> > diff --git a/security/integrity/ima/ima_api.c b/security/integrity/ima/ima_api.c
> > index c35ea613c9f8d404ba4886e3b736c3bab29d1668..8096986f3689781d3cdf6595f330033782f9cc45 100644
> > --- a/security/integrity/ima/ima_api.c
> > +++ b/security/integrity/ima/ima_api.c
> > @@ -272,10 +272,15 @@ int ima_collect_measurement(struct ima_iint_cache *iint, struct file *file,
> >  	 * to an initial measurement/appraisal/audit, but was modified to
> >  	 * assume the file changed.
> >  	 */
> > -	result = vfs_getattr_nosec(&file->f_path, &stat, STATX_CHANGE_COOKIE,
> > +	result = vfs_getattr_nosec(&file->f_path, &stat,
> > +				   STATX_CHANGE_COOKIE | STATX_CTIME,
> >  				   AT_STATX_SYNC_AS_STAT);
> > -	if (!result && (stat.result_mask & STATX_CHANGE_COOKIE))
> > -		i_version = stat.change_cookie;
> > +	if (!result) {
> > +		if (stat.result_mask & STATX_CHANGE_COOKIE)
> > +			i_version = stat.change_cookie;
> > +		else if (stat.result_mask & STATX_CTIME)
> > +			i_version = integrity_ctime_guard(stat);
> > +	}
> >  	hash.hdr.algo = algo;
> >  	hash.hdr.length = hash_digest_size[algo];
> >  
> > diff --git a/security/integrity/ima/ima_main.c b/security/integrity/ima/ima_main.c
> > index 5770cf691912aa912fc65280c59f5baac35dd725..3a4c32e254f925bba85cb91b63744ac142b3b049 100644
> > --- a/security/integrity/ima/ima_main.c
> > +++ b/security/integrity/ima/ima_main.c
> > @@ -22,6 +22,7 @@
> >  #include <linux/mount.h>
> >  #include <linux/mman.h>
> >  #include <linux/slab.h>
> > +#include <linux/stat.h>
> >  #include <linux/xattr.h>
> >  #include <linux/ima.h>
> >  #include <linux/fs.h>
> > @@ -191,18 +192,13 @@ static void ima_check_last_writer(struct ima_iint_cache *iint,
> >  
> >  	mutex_lock(&iint->mutex);
> >  	if (atomic_read(&inode->i_writecount) == 1) {
> > -		struct kstat stat;
> > -
> >  		clear_bit(IMA_EMITTED_OPENWRITERS, &iint->atomic_flags);
> >  
> >  		update = test_and_clear_bit(IMA_UPDATE_XATTR,
> >  					    &iint->atomic_flags);
> >  		if ((iint->flags & IMA_NEW_FILE) ||
> > -		    vfs_getattr_nosec(&file->f_path, &stat,
> > -				      STATX_CHANGE_COOKIE,
> > -				      AT_STATX_SYNC_AS_STAT) ||
> > -		    !(stat.result_mask & STATX_CHANGE_COOKIE) ||
> > -		    stat.change_cookie != iint->real_inode.version) {
> > +		    integrity_inode_attrs_changed(&iint->real_inode, file,
> > +						  inode)) {
> >  			iint->flags &= ~(IMA_DONE_MASK | IMA_NEW_FILE);
> >  			iint->measured_pcrs = 0;
> >  			if (update)
> > @@ -328,9 +324,8 @@ static int process_measurement(struct file *file, const struct cred *cred,
> >  	real_inode = d_real_inode(file_dentry(file));
> >  	if (real_inode != inode &&
> >  	    (action & IMA_DO_MASK) && (iint->flags & IMA_DONE_MASK)) {
> > -		if (!IS_I_VERSION(real_inode) ||
> > -		    integrity_inode_attrs_changed(&iint->real_inode,
> > -						  real_inode)) {
> > +		if (integrity_inode_attrs_changed(&iint->real_inode,
> > +						  file, real_inode)) {
> >  			iint->flags &= ~IMA_DONE_MASK;
> >  			iint->measured_pcrs = 0;
> >  		}
> > 
> > ---
> > base-commit: 8f0b4cce4481fb22653697cced8d0d04027cb1e8
> > change-id: 20251212-xfs-ima-fixup-931780a62c2c
> > 
> > Best regards,
> 

Powered by blists - more mailing lists