security: add trace event for cap_capable

In cases where we want a stable way to observe/trace
cap_capable (e.g. protection from inlining and API updates)
add a tracepoint that passes:
- The credentials used
- The user namespace of the resource being accessed
- The user namespace in which the credential provides the
capability to access the targeted resource
- The capability to check for
- The return value of the check

Signed-off-by: Jordan Rome <linux@jordanrome.com>
Acked-by: Andrii Nakryiko <andrii@kernel.org>
Reviewed-by: Paul Moore <paul@paul-moore.com>
Reviewed-by: Serge Hallyn <serge@hallyn.com>
Link: https://lore.kernel.org/r/20241204155911.1817092-1-linux@jordanrome.com
Signed-off-by: Serge Hallyn <sergeh@kernel.org>

MAINTAINERS

@@ -5147,6 +5147,7 @@ M:	Serge Hallyn <serge@hallyn.com>
 L:	linux-security-module@vger.kernel.org
 S:	Supported
 F:	include/linux/capability.h
+F:	include/trace/events/capability.h
 F:	include/uapi/linux/capability.h
 F:	kernel/capability.c
 F:	security/commoncap.c

include/trace/events/capability.h

@@ -0,0 +1,57 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM capability
+
+#if !defined(_TRACE_CAPABILITY_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_CAPABILITY_H
+
+#include <linux/cred.h>
+#include <linux/tracepoint.h>
+#include <linux/user_namespace.h>
+
+/**
+ * cap_capable - called after it's determined if a task has a particular
+ * effective capability
+ *
+ * @cred: The credentials used
+ * @target_ns: The user namespace of the resource being accessed
+ * @capable_ns: The user namespace in which the credential provides the
+ *              capability to access the targeted resource.
+ *              This will be NULL if ret is not 0.
+ * @cap: The capability to check for
+ * @ret: The return value of the check: 0 if it does, -ve if it does not
+ *
+ * Allows to trace calls to cap_capable in commoncap.c
+ */
+TRACE_EVENT(cap_capable,
+
+	TP_PROTO(const struct cred *cred, struct user_namespace *target_ns,
+		const struct user_namespace *capable_ns, int cap, int ret),
+
+	TP_ARGS(cred, target_ns, capable_ns, cap, ret),
+
+	TP_STRUCT__entry(
+		__field(const struct cred *, cred)
+		__field(struct user_namespace *, target_ns)
+		__field(const struct user_namespace *, capable_ns)
+		__field(int, cap)
+		__field(int, ret)
+	),
+
+	TP_fast_assign(
+		__entry->cred       = cred;
+		__entry->target_ns    = target_ns;
+		__entry->capable_ns = ret == 0 ? capable_ns : NULL;
+		__entry->cap        = cap;
+		__entry->ret        = ret;
+	),
+
+	TP_printk("cred %p, target_ns %p, capable_ns %p, cap %d, ret %d",
+		__entry->cred, __entry->target_ns, __entry->capable_ns, __entry->cap,
+		__entry->ret)
+);
+
+#endif /* _TRACE_CAPABILITY_H */
+
+/* This part must be outside protection */
+#include <trace/define_trace.h>

security/commoncap.c

@@ -27,6 +27,9 @@
 #include <linux/mnt_idmapping.h>
 #include <uapi/linux/lsm.h>
 
+#define CREATE_TRACE_POINTS
+#include <trace/events/capability.h>
+
 /*
  * If a non-root user executes a setuid-root binary in
  * !secure(SECURE_NOROOT) mode, then we raise capabilities.
@@ -50,24 +53,24 @@ static void warn_setuid_and_fcaps_mixed(const char *fname)
 }
 
 /**
- * cap_capable - Determine whether a task has a particular effective capability
+ * cap_capable_helper - Determine whether a task has a particular effective
+ * capability.
  * @cred: The credentials to use
- * @targ_ns:  The user namespace in which we need the capability
+ * @target_ns:  The user namespace of the resource being accessed
+ * @cred_ns:  The user namespace of the credentials
  * @cap: The capability to check for
- * @opts: Bitmask of options defined in include/linux/security.h
  *
  * Determine whether the nominated task has the specified capability amongst
  * its effective set, returning 0 if it does, -ve if it does not.
  *
- * NOTE WELL: cap_has_capability() cannot be used like the kernel's capable()
+ * See cap_capable for more details.
- * and has_capability() functions.  That is, it has the reverse semantics:
- * cap_has_capability() returns 0 when a task has a capability, but the
- * kernel's capable() and has_capability() returns 1 for this case.
  */
-int cap_capable(const struct cred *cred, struct user_namespace *targ_ns,
+static inline int cap_capable_helper(const struct cred *cred,
-		int cap, unsigned int opts)
+				     struct user_namespace *target_ns,
+				     const struct user_namespace *cred_ns,
+				     int cap)
 {
-	struct user_namespace *ns = targ_ns;
+	struct user_namespace *ns = target_ns;
 
 	/* See if cred has the capability in the target user namespace
 	 * by examining the target user namespace and all of the target
@@ -75,21 +78,21 @@ int cap_capable(const struct cred *cred, struct user_namespace *targ_ns,
 	 */
 	for (;;) {
 		/* Do we have the necessary capabilities? */
-		if (ns == cred->user_ns)
+		if (likely(ns == cred_ns))
 			return cap_raised(cred->cap_effective, cap) ? 0 : -EPERM;
 
 		/*
 		 * If we're already at a lower level than we're looking for,
 		 * we're done searching.
 		 */
-		if (ns->level <= cred->user_ns->level)
+		if (ns->level <= cred_ns->level)
 			return -EPERM;
 
 		/* 
 		 * The owner of the user namespace in the parent of the
 		 * user namespace has all caps.
 		 */
-		if ((ns->parent == cred->user_ns) && uid_eq(ns->owner, cred->euid))
+		if ((ns->parent == cred_ns) && uid_eq(ns->owner, cred->euid))
 			return 0;
 
 		/*
@@ -102,6 +105,31 @@ int cap_capable(const struct cred *cred, struct user_namespace *targ_ns,
 	/* We never get here */
 }
 
+/**
+ * cap_capable - Determine whether a task has a particular effective capability
+ * @cred: The credentials to use
+ * @target_ns:  The user namespace of the resource being accessed
+ * @cap: The capability to check for
+ * @opts: Bitmask of options defined in include/linux/security.h (unused)
+ *
+ * Determine whether the nominated task has the specified capability amongst
+ * its effective set, returning 0 if it does, -ve if it does not.
+ *
+ * NOTE WELL: cap_has_capability() cannot be used like the kernel's capable()
+ * and has_capability() functions.  That is, it has the reverse semantics:
+ * cap_has_capability() returns 0 when a task has a capability, but the
+ * kernel's capable() and has_capability() returns 1 for this case.
+ */
+int cap_capable(const struct cred *cred, struct user_namespace *target_ns,
+		int cap, unsigned int opts)
+{
+	const struct user_namespace *cred_ns = cred->user_ns;
+	int ret = cap_capable_helper(cred, target_ns, cred_ns, cap);
+
+	trace_cap_capable(cred, target_ns, cred_ns, cap, ret);
+	return ret;
+}
+
 /**
  * cap_settime - Determine whether the current process may set the system clock
  * @ts: The time to set