kernel-hardening.lists.openwall.com archive mirror
 help / color / mirror / Atom feed
From: John Wood <john.wood@gmx.com>
To: Kees Cook <keescook@chromium.org>, Jann Horn <jannh@google.com>,
	Jonathan Corbet <corbet@lwn.net>,
	James Morris <jmorris@namei.org>,
	"Serge E. Hallyn" <serge@hallyn.com>,
	Shuah Khan <shuah@kernel.org>
Cc: John Wood <john.wood@gmx.com>, Andi Kleen <ak@linux.intel.com>,
	valdis.kletnieks@vt.edu,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	Randy Dunlap <rdunlap@infradead.org>,
	linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-security-module@vger.kernel.org,
	linux-kselftest@vger.kernel.org, linux-hardening@vger.kernel.org,
	kernel-hardening@lists.openwall.com
Subject: [PATCH v7 3/7] security/brute: Detect a brute force attack
Date: Fri, 21 May 2021 19:24:10 +0200	[thread overview]
Message-ID: <20210521172414.69456-4-john.wood@gmx.com> (raw)
In-Reply-To: <20210521172414.69456-1-john.wood@gmx.com>

For a correct management of a fork brute force attack it is necessary to
track all the information related to the application crashes. To do so,
use the extended attributes (xattr) of the executable files and define a
statistical data structure to hold all the necessary information shared
by all the fork hierarchy processes. This info is the number of crashes,
the last crash timestamp and the crash period's moving average.

The same can be achieved using a pointer to the fork hierarchy
statistical data held by the task_struct structure. But this has an
important drawback: a brute force attack that happens through the execve
system call losts the faults info since these statistics are freed when
the fork hierarchy disappears. Using this method makes not possible to
manage this attack type that can be successfully treated using extended
attributes.

Also, to avoid false positives during the attack detection it is
necessary to narrow the possible cases. So, only the following scenarios
are taken into account:

1.- Launching (fork()/exec()) a setuid/setgid process repeatedly until a
    desirable memory layout is got (e.g. Stack Clash).
2.- Connecting to an exec()ing network daemon (e.g. xinetd) repeatedly
    until a desirable memory layout is got (e.g. what CTFs do for simple
    network service).
3.- Launching processes without exec() (e.g. Android Zygote) and
    exposing state to attack a sibling.
4.- Connecting to a fork()ing network daemon (e.g. apache) repeatedly
    until the previously shared memory layout of all the other children
    is exposed (e.g. kind of related to HeartBleed).

In each case, a privilege boundary has been crossed:

Case 1: setuid/setgid process
Case 2: network to local
Case 3: privilege changes
Case 4: network to local

To mark that a privilege boundary has been crossed it is only necessary
to create a new stats for the executable file via the extended attribute
and only if it has no previous statistical data. This is done using four
different LSM hooks, one per privilege boundary:

setuid/setgid process --> bprm_creds_from_file hook (based on secureexec
                          flag).
network to local -------> socket_accept hook (taking into account only
                          external connections).
privilege changes ------> task_fix_setuid and task_fix_setgid hooks.

To detect a brute force attack it is necessary that the executable file
statistics be updated in every fatal crash and the most important data
to update is the application crash period. To do so, use the new
"task_fatal_signal" LSM hook added in a previous step.

The application crash period must be a value that is not prone to change
due to spurious data and follows the real crash period. So, to compute
it, the exponential moving average (EMA) is used.

Based on the updated statistics two different attacks can be handled. A
slow brute force attack that is detected if the maximum number of faults
per fork hierarchy is reached and a fast brute force attack that is
detected if the application crash period falls below a certain
threshold.

Moreover, only the signals delivered by the kernel are taken into
account with the exception of the SIGABRT signal since the latter is
used by glibc for stack canary, malloc, etc failures, which may indicate
that a mitigation has been triggered.

Signed-off-by: John Wood <john.wood@gmx.com>
---
 include/uapi/linux/xattr.h |   3 +
 security/brute/brute.c     | 478 +++++++++++++++++++++++++++++++++++++
 2 files changed, 481 insertions(+)

diff --git a/include/uapi/linux/xattr.h b/include/uapi/linux/xattr.h
index 9463db2dfa9d..ce1c8497dceb 100644
--- a/include/uapi/linux/xattr.h
+++ b/include/uapi/linux/xattr.h
@@ -76,6 +76,9 @@
 #define XATTR_CAPS_SUFFIX "capability"
 #define XATTR_NAME_CAPS XATTR_SECURITY_PREFIX XATTR_CAPS_SUFFIX

+#define XATTR_BRUTE_SUFFIX "brute"
+#define XATTR_NAME_BRUTE XATTR_SECURITY_PREFIX XATTR_BRUTE_SUFFIX
+
 #define XATTR_POSIX_ACL_ACCESS  "posix_acl_access"
 #define XATTR_NAME_POSIX_ACL_ACCESS XATTR_SYSTEM_PREFIX XATTR_POSIX_ACL_ACCESS
 #define XATTR_POSIX_ACL_DEFAULT  "posix_acl_default"
diff --git a/security/brute/brute.c b/security/brute/brute.c
index 0edb89a58ab0..66db0dd0664d 100644
--- a/security/brute/brute.c
+++ b/security/brute/brute.c
@@ -2,8 +2,85 @@

 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt

+#include <linux/binfmts.h>
 #include <linux/lsm_hooks.h>
 #include <linux/sysctl.h>
+#include <linux/xattr.h>
+#include <net/ipv6.h>
+#include <net/sock.h>
+
+/**
+ * struct brute_stats - Fork brute force attack statistics.
+ * @faults: Number of crashes.
+ * @nsecs: Last crash timestamp as the number of nanoseconds in the
+ *         International Atomic Time (TAI) reference.
+ * @period: Crash period's moving average.
+ * @flags: Statistics flags as a whole.
+ * @not_allowed: Not allowed executable file flag.
+ * @unused: Remaining unused flags.
+ *
+ * This structure holds the statistical data shared by all the fork hierarchy
+ * processes.
+ */
+struct brute_stats {
+	u32 faults;
+	u64 nsecs;
+	u64 period;
+	union {
+		u8 flags;
+		struct {
+			u8 not_allowed : 1;
+			u8 unused : 7;
+		};
+	};
+};
+
+/**
+ * struct brute_raw_stats - Raw fork brute force attack statistics.
+ * @faults: Number of crashes.
+ * @nsecs: Last crash timestamp as the number of nanoseconds in the
+ *         International Atomic Time (TAI) reference.
+ * @period: Crash period's moving average.
+ * @flags: Statistics flags.
+ *
+ * This structure holds the statistical data on disk as an extended attribute.
+ * Since the filesystems on which extended attributes are stored might also be
+ * used on architectures with a different byte order and machine word size, care
+ * should be taken to store attribute values in an architecture-independent
+ * format.
+ */
+struct brute_raw_stats {
+	__le32 faults;
+	__le64 nsecs;
+	__le64 period;
+	u8 flags;
+} __packed;
+
+/**
+ * brute_get_current_exe_file() - Get the current task's executable file.
+ *
+ * Since all the kernel threads associated with a task share the same executable
+ * file, get the thread group leader's executable file.
+ *
+ * Context: The file must be released via fput().
+ * Return: NULL if the current task has no associated executable file. A pointer
+ *         to the executable file otherwise.
+ */
+static struct file *brute_get_current_exe_file(void)
+{
+	struct task_struct *task = current;
+	struct file *exe_file;
+
+	rcu_read_lock();
+	if (!thread_group_leader(task))
+		task = rcu_dereference(task->group_leader);
+	get_task_struct(task);
+	rcu_read_unlock();
+
+	exe_file = get_task_exe_file(task);
+	put_task_struct(task);
+	return exe_file;
+}

 /**
  * DOC: brute_ema_weight_numerator
@@ -19,6 +96,18 @@ static unsigned int brute_ema_weight_numerator __read_mostly = 7;
  */
 static unsigned int brute_ema_weight_denominator __read_mostly = 10;

+/**
+ * brute_mul_by_ema_weight() - Multiply by EMA weight.
+ * @value: Value to multiply by EMA weight.
+ *
+ * Return: The result of the multiplication operation.
+ */
+static inline u64 brute_mul_by_ema_weight(u64 value)
+{
+	return mul_u64_u32_div(value, brute_ema_weight_numerator,
+			       brute_ema_weight_denominator);
+}
+
 /**
  * DOC: brute_max_faults
  *
@@ -30,6 +119,56 @@ static unsigned int brute_ema_weight_denominator __read_mostly = 10;
  */
 static unsigned int brute_max_faults __read_mostly = 200;

+/**
+ * brute_update_crash_period() - Update the application crash period.
+ * @stats: Statistics that hold the application crash period to update. Cannot
+ *         be NULL.
+ *
+ * The application crash period must be a value that is not prone to change due
+ * to spurious data and follows the real crash period. So, to compute it, the
+ * exponential moving average (EMA) is used.
+ *
+ * This kind of average defines a weight (between 0 and 1) for the new value to
+ * add and applies the remainder of the weight to the current average value.
+ * This way, some spurious data will not excessively modify the average and only
+ * if the new values are persistent, the moving average will tend towards them.
+ *
+ * Mathematically the application crash period's EMA can be expressed as
+ * follows:
+ *
+ * period_ema = period * weight + period_ema * (1 - weight)
+ *
+ * If the operations are applied:
+ *
+ * period_ema = period * weight + period_ema - period_ema * weight
+ *
+ * If the operands are ordered:
+ *
+ * period_ema = period_ema - period_ema * weight + period * weight
+ *
+ * Finally, this formula can be written as follows:
+ *
+ * period_ema -= period_ema * weight;
+ * period_ema += period * weight;
+ */
+static void brute_update_crash_period(struct brute_stats *stats)
+{
+	u64 current_period;
+	u64 now = ktime_get_clocktai_ns();
+
+	if (stats->faults >= (u32)brute_max_faults)
+		return;
+
+	if (stats->nsecs) {
+		current_period = now > stats->nsecs ? now - stats->nsecs : 0;
+		stats->period -= brute_mul_by_ema_weight(stats->period);
+		stats->period += brute_mul_by_ema_weight(current_period);
+	}
+
+	stats->nsecs = now;
+	stats->faults += 1;
+}
+
 /**
  * DOC: brute_min_faults
  *
@@ -51,6 +190,343 @@ static unsigned int brute_min_faults __read_mostly = 5;
  */
 static unsigned int brute_crash_period_threshold __read_mostly = 30;

+/**
+ * brute_attack_running() - Test if a brute force attack is happening.
+ * @stats: Statistical data shared by all the fork hierarchy processes. Cannot
+ *         be NULL.
+ *
+ * The decision if a brute force attack is running is based on the statistical
+ * data shared by all the fork hierarchy processes.
+ *
+ * There are two types of brute force attacks that can be detected using the
+ * statistical data. The first one is a slow brute force attack that is detected
+ * if the maximum number of faults per fork hierarchy is reached. The second
+ * type is a fast brute force attack that is detected if the application crash
+ * period falls below a certain threshold.
+ *
+ * Moreover, it is important to note that no attacks will be detected until a
+ * minimum number of faults have occurred. This allows to have a trend in the
+ * crash period when the EMA is used.
+ *
+ * Return: True if a brute force attack is happening. False otherwise.
+ */
+static bool brute_attack_running(const struct brute_stats *stats)
+{
+	u64 threshold;
+
+	if (stats->faults < (u32)brute_min_faults)
+		return false;
+
+	if (stats->faults >= (u32)brute_max_faults)
+		return true;
+
+	threshold = (u64)brute_crash_period_threshold * (u64)NSEC_PER_SEC;
+	return stats->period < threshold;
+}
+
+/**
+ * brute_print_attack_running() - Warn about a fork brute force attack.
+ */
+static inline void brute_print_attack_running(void)
+{
+	pr_warn("fork brute force attack detected [pid %d: %s]\n", current->pid,
+		current->comm);
+}
+
+/**
+ * brute_get_xattr_stats() - Get the stats from an extended attribute.
+ * @dentry: The dentry of the file to get the extended attribute.
+ * @inode: The inode of the file to get the extended attribute.
+ * @stats: The stats where to store the info obtained from the extended
+ *         attribute. Cannot be NULL.
+ *
+ * Return: An error code if it is not possible to get the statistical data. Zero
+ *         otherwise.
+ */
+static int brute_get_xattr_stats(struct dentry *dentry, struct inode *inode,
+				 struct brute_stats *stats)
+{
+	int rc;
+	struct brute_raw_stats raw_stats;
+
+	rc = __vfs_getxattr(dentry, inode, XATTR_NAME_BRUTE, &raw_stats,
+			    sizeof(raw_stats));
+	if (rc < 0)
+		return rc;
+
+	stats->faults = le32_to_cpu(raw_stats.faults);
+	stats->nsecs = le64_to_cpu(raw_stats.nsecs);
+	stats->period = le64_to_cpu(raw_stats.period);
+	stats->flags = raw_stats.flags;
+	return 0;
+}
+
+/**
+ * brute_set_xattr_stats() - Set the stats to an extended attribute.
+ * @dentry: The dentry of the file to set the extended attribute.
+ * @inode: The inode of the file to set the extended attribute.
+ * @stats: The stats from where to extract the info to set the extended attribute.
+ *         Cannot be NULL.
+ *
+ * Return: An error code if it is not possible to set the statistical data. Zero
+ *         otherwise.
+ */
+static int brute_set_xattr_stats(struct dentry *dentry, struct inode *inode,
+				 const struct brute_stats *stats)
+{
+	struct brute_raw_stats raw_stats;
+
+	raw_stats.faults = cpu_to_le32(stats->faults);
+	raw_stats.nsecs = cpu_to_le64(stats->nsecs);
+	raw_stats.period = cpu_to_le64(stats->period);
+	raw_stats.flags = stats->flags;
+
+	return __vfs_setxattr(&init_user_ns, dentry, inode, XATTR_NAME_BRUTE,
+			      &raw_stats, sizeof(raw_stats), 0);
+}
+
+/**
+ * brute_update_xattr_stats() - Update the stats of a file.
+ * @file: The file that holds the statistical data to update. Cannot be NULL.
+ *
+ * For a correct management of a fork brute force attack it is only necessary to
+ * update the statistics and test if an attack is happening based on these data.
+ * It is important to note that if the file has no stats nothing is updated nor
+ * created. This way, the scenario where an application has not crossed any
+ * privilege boundary is avoided since the existence of the extended attribute
+ * denotes the crossing of bounds.
+ */
+static void brute_update_xattr_stats(const struct file *file)
+{
+	struct dentry *dentry = file_dentry(file);
+	struct inode *inode = file_inode(file);
+	struct brute_stats stats;
+	int rc;
+
+	inode_lock(inode);
+	rc = brute_get_xattr_stats(dentry, inode, &stats);
+	WARN_ON_ONCE(rc && rc != -ENODATA);
+	if (rc) {
+		inode_unlock(inode);
+		return;
+	}
+
+	brute_update_crash_period(&stats);
+	if (brute_attack_running(&stats)) {
+		brute_print_attack_running();
+		stats.not_allowed = true;
+	}
+
+	rc = brute_set_xattr_stats(dentry, inode, &stats);
+	WARN_ON_ONCE(rc);
+	inode_unlock(inode);
+}
+
+/**
+ * brute_reset_stats() - Reset the statistical data.
+ * @stats: Statistics to be reset. Cannot be NULL.
+ */
+static inline void brute_reset_stats(struct brute_stats *stats)
+{
+	memset(stats, 0, sizeof(*stats));
+}
+
+/**
+ * brute_new_xattr_stats() - New statistics for a file.
+ * @file: The file in which to create the new statistical data. Cannot be NULL.
+ *
+ * Only if the file has no statistical data create it. This function will be
+ * called to mark that a privilege boundary has been crossed so, if new stats
+ * are required, they do not contain any useful data. The existence of the
+ * extended attribute denotes the crossing of privilege bounds.
+ *
+ * Return: An error code if it is not possible to get or set the statistical
+ *         data. Zero otherwise.
+ */
+static int brute_new_xattr_stats(const struct file *file)
+{
+	struct dentry *dentry = file_dentry(file);
+	struct inode *inode = file_inode(file);
+	struct brute_stats stats;
+	int rc;
+
+	inode_lock(inode);
+	rc = brute_get_xattr_stats(dentry, inode, &stats);
+	if (rc && rc != -ENODATA)
+		goto unlock;
+
+	if (rc == -ENODATA) {
+		brute_reset_stats(&stats);
+		rc = brute_set_xattr_stats(dentry, inode, &stats);
+		if (rc)
+			goto unlock;
+	}
+
+unlock:
+	inode_unlock(inode);
+	return rc;
+}
+
+/**
+ * brute_current_new_xattr_stats() - New stats for the current task's exe file.
+ *
+ * Return: An error code if it is not possible to get or set the statistical
+ *         data. Zero otherwise.
+ */
+static int brute_current_new_xattr_stats(void)
+{
+	struct file *exe_file;
+	int rc;
+
+	exe_file = brute_get_current_exe_file();
+	if (WARN_ON_ONCE(!exe_file))
+		return -ENOENT;
+
+	rc = brute_new_xattr_stats(exe_file);
+	WARN_ON_ONCE(rc);
+	fput(exe_file);
+	return rc;
+}
+
+/**
+ * brute_signal_from_user() - Test if a signal is coming from userspace.
+ * @siginfo: Contains the signal information.
+ *
+ * To avoid false positives during the attack detection it is necessary to
+ * narrow the possible cases. So, only the signals delivered by the kernel are
+ * taken into account with the exception of the SIGABRT signal since the latter
+ * is used by glibc for stack canary, malloc, etc failures, which may indicate
+ * that a mitigation has been triggered.
+ *
+ * Return: True if the signal is coming from usersapce. False otherwise.
+ */
+static inline bool brute_signal_from_user(const kernel_siginfo_t *siginfo)
+{
+	return siginfo->si_signo == SIGKILL && siginfo->si_code != SIGABRT;
+}
+
+/**
+ * brute_task_fatal_signal() - Target for the task_fatal_signal hook.
+ * @siginfo: Contains the signal information.
+ *
+ * To detect a brute force attack it is necessary, as a first step, to test in
+ * every fatal crash if the signal is delibered by the kernel. If so, update the
+ * statistics and act based on these data.
+ */
+static void brute_task_fatal_signal(const kernel_siginfo_t *siginfo)
+{
+	struct file *exe_file;
+
+	if (brute_signal_from_user(siginfo))
+		return;
+
+	exe_file = brute_get_current_exe_file();
+	if (WARN_ON_ONCE(!exe_file))
+		return;
+
+	brute_update_xattr_stats(exe_file);
+	fput(exe_file);
+}
+
+/**
+ * brute_task_execve() - Target for the bprm_creds_from_file hook.
+ * @bprm: Contains the linux_binprm structure.
+ * @file: Binary that will be executed without an interpreter.
+ *
+ * This hook is useful to mark that a privilege boundary (setuid/setgid process)
+ * has been crossed. This is done based on the "secureexec" flag.
+ *
+ * To be defensive return an error code if it is not possible to get or set the
+ * stats using an extended attribute since this blocks the execution of the
+ * file. This scenario is treated as an attack.
+ *
+ * It is important to note that here the brute_new_xattr_stats function could be
+ * used with a previous test of the secureexec flag. However it is better to use
+ * the basic xattr functions since in a future commit a test if the execution is
+ * allowed (via the brute_stats::not_allowed flag) will be necessary. This way,
+ * the stats of the file will be get only once.
+ *
+ * Return: An error code if it is not possible to get or set the statistical
+ *         data. Zero otherwise.
+ */
+static int brute_task_execve(struct linux_binprm *bprm, struct file *file)
+{
+	struct dentry *dentry = file_dentry(bprm->file);
+	struct inode *inode = file_inode(bprm->file);
+	struct brute_stats stats;
+	int rc;
+
+	inode_lock(inode);
+	rc = brute_get_xattr_stats(dentry, inode, &stats);
+	if (WARN_ON_ONCE(rc && rc != -ENODATA))
+		goto unlock;
+
+	if (rc == -ENODATA && bprm->secureexec) {
+		brute_reset_stats(&stats);
+		rc = brute_set_xattr_stats(dentry, inode, &stats);
+		if (WARN_ON_ONCE(rc))
+			goto unlock;
+	}
+
+	rc = 0;
+unlock:
+	inode_unlock(inode);
+	return rc;
+}
+
+/**
+ * brute_task_change_priv() - Target for the task_fix_setid hooks.
+ * @new: The set of credentials that will be installed.
+ * @old: The set of credentials that are being replaced.
+ * @flags: Contains one of the LSM_SETID_* values.
+ *
+ * This hook is useful to mark that a privilege boundary (privilege changes) has
+ * been crossed.
+ *
+ * Return: An error code if it is not possible to get or set the statistical
+ *         data. Zero otherwise.
+ */
+static int brute_task_change_priv(struct cred *new, const struct cred *old, int flags)
+{
+	return brute_current_new_xattr_stats();
+}
+
+/**
+ * brute_socket_accept() - Target for the socket_accept hook.
+ * @sock: Contains the listening socket structure.
+ * @newsock: Contains the newly created server socket for connection.
+ *
+ * This hook is useful to mark that a privilege boundary (network to local) has
+ * been crossed. This is done only if the listening socket accepts external
+ * connections. The sockets for inter-process communication (IPC) and those that
+ * are listening on loopback addresses are not taken into account.
+ *
+ * Return: An error code if it is not possible to get or set the statistical
+ *         data. Zero otherwise.
+ */
+static int brute_socket_accept(struct socket *sock, struct socket *newsock)
+{
+	struct sock *sk = sock->sk;
+
+	if (sk->sk_family == AF_UNIX || sk->sk_family == AF_NETLINK ||
+	    sk->sk_rcv_saddr == htonl(INADDR_LOOPBACK) ||
+	    ipv6_addr_equal(&sk->sk_v6_rcv_saddr, &in6addr_loopback))
+		return 0;
+
+	return brute_current_new_xattr_stats();
+}
+
+/*
+ * brute_hooks - Targets for the LSM's hooks.
+ */
+static struct security_hook_list brute_hooks[] __lsm_ro_after_init = {
+	LSM_HOOK_INIT(task_fatal_signal, brute_task_fatal_signal),
+	LSM_HOOK_INIT(bprm_creds_from_file, brute_task_execve),
+	LSM_HOOK_INIT(task_fix_setuid, brute_task_change_priv),
+	LSM_HOOK_INIT(task_fix_setgid, brute_task_change_priv),
+	LSM_HOOK_INIT(socket_accept, brute_socket_accept),
+};
+
 #ifdef CONFIG_SYSCTL
 static unsigned int uint_max = UINT_MAX;
 #define SYSCTL_UINT_MAX (&uint_max)
@@ -137,6 +613,8 @@ static inline void brute_init_sysctl(void) { }
 static int __init brute_init(void)
 {
 	pr_info("becoming mindful\n");
+	security_add_hooks(brute_hooks, ARRAY_SIZE(brute_hooks),
+			   KBUILD_MODNAME);
 	brute_init_sysctl();
 	return 0;
 }
--
2.25.1


  parent reply	other threads:[~2021-05-21 17:55 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-05-21 17:24 [PATCH v7 0/7] Fork brute force attack mitigation John Wood
2021-05-21 17:24 ` [PATCH v7 1/7] security: Add LSM hook at the point where a task gets a fatal signal John Wood
2021-05-21 17:24 ` [PATCH v7 2/7] security/brute: Define a LSM and add sysctl attributes John Wood
2021-05-21 17:24 ` John Wood [this message]
2021-05-21 17:24 ` [PATCH v7 4/7] security/brute: Mitigate a brute force attack John Wood
2021-05-21 18:02 ` [PATCH v7 0/7] Fork brute force attack mitigation Andi Kleen
2021-05-23  7:31   ` John Wood
2021-05-23 14:43     ` Andi Kleen
2021-05-23 15:47       ` John Wood
2021-05-22  6:35 ` [PATCH v7 5/7] selftests/brute: Add tests for the Brute LSM John Wood
2021-05-22  6:39 ` [PATCH v7 6/7] Documentation: Add documentation " John Wood
2021-05-22  6:44 ` [PATCH v7 7/7] MAINTAINERS: Add a new entry " John Wood

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20210521172414.69456-4-john.wood@gmx.com \
    --to=john.wood@gmx.com \
    --cc=ak@linux.intel.com \
    --cc=corbet@lwn.net \
    --cc=gregkh@linuxfoundation.org \
    --cc=jannh@google.com \
    --cc=jmorris@namei.org \
    --cc=keescook@chromium.org \
    --cc=kernel-hardening@lists.openwall.com \
    --cc=linux-doc@vger.kernel.org \
    --cc=linux-hardening@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-kselftest@vger.kernel.org \
    --cc=linux-security-module@vger.kernel.org \
    --cc=rdunlap@infradead.org \
    --cc=serge@hallyn.com \
    --cc=shuah@kernel.org \
    --cc=valdis.kletnieks@vt.edu \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).