tools: bpftool: Allow unprivileged users to probe features

There is demand for a way to identify what BPF helper functions are
available to unprivileged users. To do so, allow unprivileged users to
run "bpftool feature probe" to list BPF-related features. This will only
show features accessible to those users, and may not reflect the full
list of features available (to administrators) on the system.

To avoid the case where bpftool is inadvertently run as non-root and
would list only a subset of the features supported by the system when it
would be expected to list all of them, running as unprivileged is gated
behind the "unprivileged" keyword passed to the command line. When used
by a privileged user, this keyword allows to drop the CAP_SYS_ADMIN and
to list the features available to unprivileged users. Note that this
addsd a dependency on libpcap for compiling bpftool.

Note that there is no particular reason why the probes were restricted
to root, other than the fact I did not need them for unprivileged and
did not bother with the additional checks at the time probes were added.

Signed-off-by: Quentin Monnet <quentin@isovalent.com>
Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
Acked-by: John Fastabend <john.fastabend@gmail.com>
Link: https://lore.kernel.org/bpf/20200429144506.8999-3-quentin@isovalent.com

1 files changed, 89 insertions, 13 deletions

diff --git a/tools/bpf/bpftool/feature.c b/tools/bpf/bpftool/feature.c
index 59e4cb44efbc..952f4b1987c0 100644
--- a/tools/bpf/bpftool/feature.c
+++ b/tools/bpf/bpftool/feature.c
@@ -6,6 +6,7 @@
 #include <string.h>
 #include <unistd.h>
 #include <net/if.h>
+#include <sys/capability.h>
 #include <sys/utsname.h>
 #include <sys/vfs.h>
 
@@ -36,6 +37,7 @@ static const char * const helper_name[] = {
 #undef BPF_HELPER_MAKE_ENTRY
 
 static bool full_mode;
+static bool run_as_unprivileged;
 
 /* Miscellaneous utility functions */
 
@@ -473,6 +475,11 @@ probe_prog_type(enum bpf_prog_type prog_type, bool *supported_types,
 		}
 
 	res = bpf_probe_prog_type(prog_type, ifindex);
+	/* Probe may succeed even if program load fails, for unprivileged users
+	 * check that we did not fail because of insufficient permissions
+	 */
+	if (run_as_unprivileged && errno == EPERM)
+		res = false;
 
 	supported_types[prog_type] |= res;
 
@@ -501,6 +508,10 @@ probe_map_type(enum bpf_map_type map_type, const char *define_prefix,
 
 	res = bpf_probe_map_type(map_type, ifindex);
 
+	/* Probe result depends on the success of map creation, no additional
+	 * check required for unprivileged users
+	 */
+
 	maxlen = sizeof(plain_desc) - strlen(plain_comment) - 1;
 	if (strlen(map_type_name[map_type]) > maxlen) {
 		p_info("map type name too long");
@@ -520,12 +531,17 @@ probe_helper_for_progtype(enum bpf_prog_type prog_type, bool supported_type,
 			  const char *define_prefix, unsigned int id,
 			  const char *ptype_name, __u32 ifindex)
 {
-	bool res;
+	bool res = false;
 
-	if (!supported_type)
-		res = false;
-	else
+	if (supported_type) {
 		res = bpf_probe_helper(id, prog_type, ifindex);
+		/* Probe may succeed even if program load fails, for
+		 * unprivileged users check that we did not fail because of
+		 * insufficient permissions
+		 */
+		if (run_as_unprivileged && errno == EPERM)
+			res = false;
+	}
 
 	if (json_output) {
 		if (res)
@@ -720,6 +736,65 @@ static void section_misc(const char *define_prefix, __u32 ifindex)
 	print_end_section();
 }
 
+static int handle_perms(void)
+{
+	cap_value_t cap_list[1] = { CAP_SYS_ADMIN };
+	bool has_sys_admin_cap = false;
+	cap_flag_value_t val;
+	int res = -1;
+	cap_t caps;
+
+	caps = cap_get_proc();
+	if (!caps) {
+		p_err("failed to get capabilities for process: %s",
+		      strerror(errno));
+		return -1;
+	}
+
+	if (cap_get_flag(caps, CAP_SYS_ADMIN, CAP_EFFECTIVE, &val)) {
+		p_err("bug: failed to retrieve CAP_SYS_ADMIN status");
+		goto exit_free;
+	}
+	if (val == CAP_SET)
+		has_sys_admin_cap = true;
+
+	if (!run_as_unprivileged && !has_sys_admin_cap) {
+		p_err("full feature probing requires CAP_SYS_ADMIN, run as root or use 'unprivileged'");
+		goto exit_free;
+	}
+
+	if ((run_as_unprivileged && !has_sys_admin_cap) ||
+	    (!run_as_unprivileged && has_sys_admin_cap)) {
+		/* We are all good, exit now */
+		res = 0;
+		goto exit_free;
+	}
+
+	/* if (run_as_unprivileged && has_sys_admin_cap), drop CAP_SYS_ADMIN */
+
+	if (cap_set_flag(caps, CAP_EFFECTIVE, ARRAY_SIZE(cap_list), cap_list,
+			 CAP_CLEAR)) {
+		p_err("bug: failed to clear CAP_SYS_ADMIN from capabilities");
+		goto exit_free;
+	}
+
+	if (cap_set_proc(caps)) {
+		p_err("failed to drop CAP_SYS_ADMIN: %s", strerror(errno));
+		goto exit_free;
+	}
+
+	res = 0;
+
+exit_free:
+	if (cap_free(caps) && !res) {
+		p_err("failed to clear storage object for capabilities: %s",
+		      strerror(errno));
+		res = -1;
+	}
+
+	return res;
+}
+
 static int do_probe(int argc, char **argv)
 {
 	enum probe_component target = COMPONENT_UNSPEC;
@@ -728,14 +803,6 @@ static int do_probe(int argc, char **argv)
 	__u32 ifindex = 0;
 	char *ifname;
 
-	/* Detection assumes user has sufficient privileges (CAP_SYS_ADMIN).
-	 * Let's approximate, and restrict usage to root user only.
-	 */
-	if (geteuid()) {
-		p_err("please run this command as root user");
-		return -1;
-	}
-
 	set_max_rlimit();
 
 	while (argc) {
@@ -784,6 +851,9 @@ static int do_probe(int argc, char **argv)
 			if (!REQ_ARGS(1))
 				return -1;
 			define_prefix = GET_ARG();
+		} else if (is_prefix(*argv, "unprivileged")) {
+			run_as_unprivileged = true;
+			NEXT_ARG();
 		} else {
 			p_err("expected no more arguments, 'kernel', 'dev', 'macros' or 'prefix', got: '%s'?",
 			      *argv);
@@ -791,6 +861,12 @@ static int do_probe(int argc, char **argv)
 		}
 	}
 
+	/* Full feature detection requires CAP_SYS_ADMIN privilege.
+	 * Let's approximate, and warn if user is not root.
+	 */
+	if (handle_perms())
+		return -1;
+
 	if (json_output) {
 		define_prefix = NULL;
 		jsonw_start_object(json_wtr);
@@ -821,7 +897,7 @@ static int do_help(int argc, char **argv)
 	}
 
 	fprintf(stderr,
-		"Usage: %s %s probe [COMPONENT] [full] [macros [prefix PREFIX]]\n"
+		"Usage: %s %s probe [COMPONENT] [full] [unprivileged] [macros [prefix PREFIX]]\n"
 		"       %s %s help\n"
 		"\n"
 		"       COMPONENT := { kernel | dev NAME }\n"