Re: [PATCH V2 1/1] IMA event log trimming

[Mimi Zohar <zohar@linux.ibm.com>]

Hi Steven,

As I previously said, "The main difference between this patch and Roberto's
version is the length of time needed for locking the measurement list, which
prevents new entries from being appended to the measurement list.  In Roberto's
version, the list head is moved quickly and the lock released.  Measuring the
total amount of time needed to trim the measurement list ignores the benefit of
his version. I plan on reviewing both this version and his (hopefully today)."

The other difference is "when" the IMA measurement list is read and saved,
before the trigger to trim the measurement list or after when the measurement
list is staged.  In this case, the initial trigger trims the measurement list.
In the other case, the measurement list is staged and then deleted.  When
reviewing Roberto's patch, I plan to discuss it.

After trimming the measurement list, existing verifiers, which walk the IMA
measurement list, will obviously fail to match the PCRs.  Breaking existing
userspace applications is a problem and, unfortunately, requires yet another
Kconfig option.  It needs to be at least mentioned here in the patch
description.

There are two places where it says, "the list never shrinks, so we don't need a
lock here".  Either the code, the comment, or both need to be updated.


On Thu, 2025-12-11 at 10:41 -0800, steven chen wrote:
> On 12/10/2025 3:53 PM, steven chen wrote:
> > This patch is for trimming N entries of the IMA event logs. It will also
> > cleaning the hash table if ima_flush_htable is set.

Please refer to "Describe your changes in imperative mood" in the "Describe your
changes" section of Documentation/process/submitting-patches.rst.

> > 
> > It provides a userspace interface ima_trim_log that can be used to input
> > number N to let kernel to trim N entries of IMA event logs. When read

There is only a single kernel measurement list or event log, not plural.  There
are N number of "entries" or "records" in the IMA measurement list.

-> trim N records from the IMA measurement list.

> > this interface, it returns number of entries trimmed last time.

Please provide an example of how to initiate the trim.

After trimming the IMA measurement list, are the other securityfs files correct?
Are they correct after a kexec?  Or are they reset without a way of resurrecting
them without the full measurement list?

> > 
> > Signed-off-by: steven chen <chenste@linux.microsoft.com>

> > ---
> >   .../admin-guide/kernel-parameters.txt         |   4 +
> >   security/integrity/ima/ima.h                  |   2 +
> >   security/integrity/ima/ima_fs.c               | 175 +++++++++++++++++-
> >   security/integrity/ima/ima_queue.c            |  64 +++++++
> >   4 files changed, 241 insertions(+), 4 deletions(-)
> > 
> > diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
> > index e92c0056e4e0..cd1a1d0bf0e2 100644
> > --- a/Documentation/admin-guide/kernel-parameters.txt
> > +++ b/Documentation/admin-guide/kernel-parameters.txt
> > @@ -2197,6 +2197,10 @@
> >   			Use the canonical format for the binary runtime
> >   			measurements, instead of host native format.
> >   
> > +	ima_flush_htable  [IMA]
> > +			Flush the measurement list hash table when trim all
> > +			or a part of it for deletion.
> > +
> >   	ima_hash=	[IMA]
> >   			Format: { md5 | sha1 | rmd160 | sha256 | sha384
> >   				   | sha512 | ... }
> > diff --git a/security/integrity/ima/ima.h b/security/integrity/ima/ima.h
> > index e3d71d8d56e3..ab0e30ee25ea 100644
> > --- a/security/integrity/ima/ima.h
> > +++ b/security/integrity/ima/ima.h
> > @@ -246,8 +246,10 @@ void ima_post_key_create_or_update(struct key *keyring, struct key *key,
> >   
> >   #ifdef CONFIG_IMA_KEXEC
> >   void ima_measure_kexec_event(const char *event_name);
> > +long ima_purge_event_log(long number_logs);
> >   #else
> >   static inline void ima_measure_kexec_event(const char *event_name) {}
> > +static inline long ima_purge_event_log(long number_logs) { return 0; }
> >   #endif
> >   
> >   /*
> > diff --git a/security/integrity/ima/ima_fs.c b/security/integrity/ima/ima_fs.c
> > index 87045b09f120..410f7d03c43f 100644
> > --- a/security/integrity/ima/ima_fs.c
> > +++ b/security/integrity/ima/ima_fs.c
> > @@ -21,6 +21,9 @@
> >   #include <linux/rcupdate.h>
> >   #include <linux/parser.h>
> >   #include <linux/vmalloc.h>
> > +#include <linux/ktime.h>
> > +#include <linux/timekeeping.h>
> > +#include <linux/ima.h>
> >   
> >   #include "ima.h"
> >   
> > @@ -38,6 +41,14 @@ __setup("ima_canonical_fmt", default_canonical_fmt_setup);
> >   
> >   static int valid_policy = 1;
> >   
> > +#define IMA_LOG_TRIM_REQ_LENGTH 11
> > +#define IMA_LOG_TRIM_EVENT_LEN 256
> > +
> > +static long trimcount;
> > +/* mutex protects atomicity of trimming measurement list requests */

ima_measure_lock is taken for more than just synchronization of trimming the
measurement list.  Please update comment.

> > +static DEFINE_MUTEX(ima_measure_lock);
> > +static long ima_measure_users;
> > +
> >   static ssize_t ima_show_htable_value(char __user *buf, size_t count,
> >   				     loff_t *ppos, atomic_long_t *val)
> >   {
> > @@ -202,16 +213,65 @@ static const struct seq_operations ima_measurments_seqops = {
> >   	.show = ima_measurements_show
> >   };
> >   

_ima_measurements_open() seems pretty fundamental to the locking scheme. 
Preventing opening the IMA measurement list is new.  There should at least be a
short, regular comment explaining what you're locking and why.

> > +static int _ima_measurements_open(struct inode *inode, struct file *file,
> > +				  const struct seq_operations *seq_ops)
> > +{
> > +	bool write = !!(file->f_mode & FMODE_WRITE);
> > +	int ret;
> > +
> > +	if (write && !capable(CAP_SYS_ADMIN))
> > +		return -EPERM;
> > +
> > +	mutex_lock(&ima_measure_lock);
> > +	if ((write && ima_measure_users != 0) ||
> > +	    (!write && ima_measure_users < 0)) {
> > +		mutex_unlock(&ima_measure_lock);
> > +		return -EBUSY;
> > +	}
> > +
> > +	ret = seq_open(file, seq_ops);
> > +	if (ret < 0) {
> > +		mutex_unlock(&ima_measure_lock);
> > +		return ret;
> > +	}
> > +
> > +	if (write)
> > +		ima_measure_users--;
> > +	else
> > +		ima_measure_users++;
> > +
> > +	mutex_unlock(&ima_measure_lock);
> > +	return ret;
> > +}
> > +
> >   static int ima_measurements_open(struct inode *inode, struct file *file)
> >   {
> > -	return seq_open(file, &ima_measurments_seqops);
> > +	return _ima_measurements_open(inode, file, &ima_measurments_seqops);
> > +}
> > +
> > +static int ima_measurements_release(struct inode *inode, struct file *file)
> > +{
> > +	bool write = !!(file->f_mode & FMODE_WRITE);
> > +	int ret;
> > +
> > +	mutex_lock(&ima_measure_lock);
> > +	ret = seq_release(inode, file);
> > +	if (!ret) {
> > +		if (write)
> > +			ima_measure_users++;
> > +		else
> > +			ima_measure_users--;
> > +	}
> > +
> > +	mutex_unlock(&ima_measure_lock);
> > +	return ret;
> >   }
> >   
> >   static const struct file_operations ima_measurements_ops = {
> >   	.open = ima_measurements_open,
> >   	.read = seq_read,
> >   	.llseek = seq_lseek,
> > -	.release = seq_release,
> > +	.release = ima_measurements_release,
> >   };
> >   
> >   void ima_print_digest(struct seq_file *m, u8 *digest, u32 size)
> > @@ -279,14 +339,111 @@ static const struct seq_operations ima_ascii_measurements_seqops = {
> >   
> >   static int ima_ascii_measurements_open(struct inode *inode, struct file *file)
> >   {
> > -	return seq_open(file, &ima_ascii_measurements_seqops);
> > +	return _ima_measurements_open(inode, file, &ima_ascii_measurements_seqops);
> >   }
> >   
> >   static const struct file_operations ima_ascii_measurements_ops = {
> >   	.open = ima_ascii_measurements_open,
> >   	.read = seq_read,
> >   	.llseek = seq_lseek,
> > -	.release = seq_release,
> > +	.release = ima_measurements_release,
> > +};
> > +
> > +static void ima_measure_trim_event(const long number_logs)
> > +{
> > +	char ima_log_trim_event[IMA_LOG_TRIM_EVENT_LEN];
> > +	struct timespec64 ts;
> > +	u64 time_ns;
> > +	int n;
> > +
> > +	ktime_get_real_ts64(&ts);
> > +	time_ns = (u64)ts.tv_sec * 1000000000ULL + ts.tv_nsec;
> > +	n = scnprintf(ima_log_trim_event, IMA_LOG_TRIM_EVENT_LEN,
> > +		      "time=%llu; log trim this time=%lu;",
> > +		       time_ns, number_logs);
> > +
> > +	ima_measure_critical_data("ima_log_trim", "trim ima event logs", ima_log_trim_event, n, false, NULL, 0);
> > +}

There's no mention of a new critical data record in the patch description.  It
should be a separate patch with a full patch description describing what it is
and how to verify it.

> > +
> > +static int ima_log_trim_open(struct inode *inode, struct file *file)
> > +{
> > +	bool write = !!(file->f_mode & FMODE_WRITE);
> > +
> > +	if (!write && capable(CAP_SYS_ADMIN))
> > +		return 0;
> > +	else if (!capable(CAP_SYS_ADMIN))
> > +		return -EPERM;
> > +
> > +	return _ima_measurements_open(inode, file, &ima_measurments_seqops);
> > +}
> > +
> > +static ssize_t ima_log_trim_read(struct file *file, char __user *buf, size_t size, loff_t *ppos)
> > +{
> > +	char tmpbuf[IMA_LOG_TRIM_REQ_LENGTH];	/* greater than largest 'long' string value */
> > +	ssize_t len;
> > +
> > +	len = scnprintf(tmpbuf, sizeof(tmpbuf), "%li\n", trimcount);
> > +	return simple_read_from_buffer(buf, size, ppos, tmpbuf, len);
> > +}
> > +
> > +static ssize_t ima_log_trim_write(struct file *file,
> > +				  const char __user *buf, size_t datalen, loff_t *ppos)
> > +{
> > +	unsigned char req[IMA_LOG_TRIM_REQ_LENGTH];
> > +	long count, n;
> > +	int ret;
> > +
> > +	if (*ppos > 0 || datalen > IMA_LOG_TRIM_REQ_LENGTH || datalen < 2) {
> > +		ret = -EINVAL;
> > +		goto out;
> > +	}
> > +
> > +	n = (int)datalen;
> > +
> > +	ret = copy_from_user(req, buf, datalen);
> > +	if (ret < 0)
> > +		goto out;
> > +
> > +	count = 0;
> > +	for (int i = 0; i < n; ++i) {
> > +		if (req[i] < '0' || req[i] > '9') {
> > +			ret = -EINVAL;
> > +			goto out;
> > +		}
> > +		count = count * 10 + req[i] - '0';
> > +	}

This code can be simplified by using the kstrto*_from_user() family of
functions.  The patch description should include an example how to trim the
measurement list.

> > +	ret = ima_purge_event_log(count);

The term "purge" is really strong wording.  I hope you're not purging the
measurement list, but simply removing them from kernel memory.

> > +
> > +	if (ret < 0)
> > +		goto out;
> > +
> > +	trimcount = ret;
> > +
> > +	if (trimcount > 0)
> > +		ima_measure_trim_event(trimcount);
> > +
> > +	ret = datalen;
> > +out:
> > +	return ret;
> > +}
> > +
> > +static int ima_log_trim_release(struct inode *inode, struct file *file)
> > +{
> > +	bool write = !!(file->f_mode & FMODE_WRITE);
> > +	if (!write && capable(CAP_SYS_ADMIN))
> > +		return 0;
> > +	else if (!capable(CAP_SYS_ADMIN))
> > +		return -EPERM;
> > +
> > +	return ima_measurements_release(inode, file);
> > +}
> > +
> > +static const struct file_operations ima_log_trim_ops = {
> > +	.open = ima_log_trim_open,
> > +	.read = ima_log_trim_read,
> > +	.write = ima_log_trim_write,
> > +	.llseek = generic_file_llseek,
> > +	.release = ima_log_trim_release
> >   };
> >   
> >   static ssize_t ima_read_policy(char *path)
> > @@ -528,6 +685,16 @@ int __init ima_fs_init(void)
> >   		goto out;
> >   	}
> >   
> > +	dentry = securityfs_create_file("ima_trim_log",
> > +					S_IRUSR | S_IRGRP | S_IWUSR | S_IWGRP,
> > +					ima_dir, NULL, &ima_log_trim_ops);
> > +	if (IS_ERR(dentry)) {
> > +		ret = PTR_ERR(dentry);
> > +		goto out;
> > +	}
> > +
> > +	trimcount = 0;
> > +
> >   	dentry = securityfs_create_file("runtime_measurements_count",
> >   				   S_IRUSR | S_IRGRP, ima_dir, NULL,
> >   				   &ima_measurements_count_ops);
> > diff --git a/security/integrity/ima/ima_queue.c b/security/integrity/ima/ima_queue.c
> > index 590637e81ad1..77ab52469727 100644
> > --- a/security/integrity/ima/ima_queue.c
> > +++ b/security/integrity/ima/ima_queue.c
> > @@ -22,6 +22,14 @@
> >   
> >   #define AUDIT_CAUSE_LEN_MAX 32
> >   
> > +bool ima_flush_htable;
> > +static int __init ima_flush_htable_setup(char *str)
> > +{
> > +	ima_flush_htable = true;
> > +	return 1;
> > +}
> > +__setup("ima_flush_htable", ima_flush_htable_setup);
> > +
> >   /* pre-allocated array of tpm_digest structures to extend a PCR */
> >   static struct tpm_digest *digests;
> >   
> > @@ -220,6 +228,62 @@ int ima_add_template_entry(struct ima_template_entry *entry, int violation,
> >   	return result;
> >   }
> >   
> > +/* Delete the IMA event logs */

Do you mean delete the IMA event records/entries?

> > +long ima_purge_event_log(long number_logs)

If this function is not defined as static, then it requires a kernel-doc.

> > +{
> > +	struct ima_queue_entry *qe, *qe_tmp;
> > +	LIST_HEAD(ima_measurements_staged);
> > +	unsigned int i;
> > +	long cur = number_logs;

The variable name "number_logs" is confusing.  As I mentioned in the patch
description, there is one measurement list with multiple records.  There aren't
multiple logs in the kernel (other than the staged list).

> > +
> > +	if (number_logs <= 0)
> > +		return number_logs;
> > +
> > +	mutex_lock(&ima_extend_list_mutex);
> > +
> > +
> > +	list_for_each_entry(qe, &ima_measurements, later) {
> > +		if (--number_logs == 0)
> > +			break;
> > +	}
> > +
> > +	if (number_logs > 0) {
> > +		mutex_unlock(&ima_extend_list_mutex);
> > +		return -ENOENT;
> > +	}
> > +
> > +	__list_cut_position(&ima_measurements_staged, &ima_measurements,
> > +				    &qe->later);
> > +	atomic_long_sub(cur, &ima_htable.len);
> > +
> > +	if (!IS_ENABLED(CONFIG_IMA_DISABLE_HTABLE) && ima_flush_htable) {
> > +		list_for_each_entry(qe, &ima_measurements_staged, later)
> > +			/* It can race with ima_lookup_digest_entry(). */
> > +			hlist_del_rcu(&qe->hnext);
> > +	}
> 
> If the h table can be staged during the locking period and deleted after 
> unlocking, the time
> the list is held will be reduced.
> 
> I will work on this, and any suggestions are greatly appreciated.
> 
> Thanks,
> 
> Steven
> 
> > +
> > +	mutex_unlock(&ima_extend_list_mutex);
> > +
> > +
> > +	list_for_each_entry_safe(qe, qe_tmp, &ima_measurements_staged, later) {
> > +		for (i = 0; i < qe->entry->template_desc->num_fields; i++) {
> > +			kfree(qe->entry->template_data[i].data);
> > +			qe->entry->template_data[i].data = NULL;
> > +			qe->entry->template_data[i].len = 0;
> > +		}
> > +
> > +		list_del(&qe->later);
> > +
> > +		if (ima_flush_htable) {
> > +			kfree(qe->entry->digests);
> > +			kfree(qe->entry);
> > +			kfree(qe);
> > +		}
> > +	}

To avoid code duplication, there's a similar function named
ima_free_template_entry().  Any changes needed to the function should be done as
a preparatory patch.

> > +
> > +	return cur;
> > +}
> > +
> >   int ima_restore_measurement_entry(struct ima_template_entry *entry)
> >   {
> >   	int result = 0;


-- 
thanks,

Mimi