trace: add missing limits.h include

[project/procd.git] / trace / trace.c

diff --git a/trace/trace.c b/trace/trace.c
index 4c25a4f..1c1e757 100644 (file)
--- a/trace/trace.c
+++ b/trace/trace.c
@@ -12,8 +12,10 @@
  */
 
 #define _GNU_SOURCE
  */
 
 #define _GNU_SOURCE

+#include <fcntl.h>

 #include <stddef.h>
 #include <sys/ptrace.h>
 #include <stddef.h>
 #include <sys/ptrace.h>

+#include <sys/stat.h>

 #include <sys/types.h>
 #include <sys/user.h>
 #include <sys/wait.h>
 #include <sys/types.h>
 #include <sys/user.h>
 #include <sys/wait.h>


@@ -23,7 +25,15 @@
 #include <errno.h>
 #include <string.h>
 #include <syslog.h>
 #include <errno.h>
 #include <string.h>
 #include <syslog.h>

+#include <limits.h>

 
 

+#ifndef PTRACE_EVENT_STOP
+/* PTRACE_EVENT_STOP is defined in linux/ptrace.h, but this header
+ * collides with musl's sys/ptrace.h */
+#define PTRACE_EVENT_STOP 128
+#endif
+
+#include <libubox/ulog.h>

 #include <libubox/uloop.h>
 #include <libubox/blobmsg.h>
 #include <libubox/blobmsg_json.h>
 #include <libubox/uloop.h>
 #include <libubox/blobmsg.h>
 #include <libubox/blobmsg_json.h>


@@ -43,19 +53,19 @@
 # endif
 #define reg_syscall_nr (EF_REG2 / 4)
 #elif defined(__arm__)
 # endif
 #define reg_syscall_nr (EF_REG2 / 4)
 #elif defined(__arm__)

+#include <asm/ptrace.h>                /* for PTRACE_SET_SYSCALL */

 #define reg_syscall_nr _offsetof(struct user, regs.uregs[7])
 #define reg_syscall_nr _offsetof(struct user, regs.uregs[7])

+# if defined(__ARM_EABI__)
+# define reg_retval_nr _offsetof(struct user, regs.uregs[0])
+# endif

 #else
 #error tracing is not supported on this architecture
 #endif
 
 #else
 #error tracing is not supported on this architecture
 #endif
 

-#define INFO(fmt, ...) do { \
-       fprintf(stderr, "utrace: "fmt, ## __VA_ARGS__); \
-} while (0)
-
-#define ERROR(fmt, ...) do { \
-       syslog(LOG_ERR, "utrace: "fmt, ## __VA_ARGS__); \
-       fprintf(stderr, "utrace: "fmt, ## __VA_ARGS__); \
-} while (0)
+enum mode {
+       UTRACE,
+       SECCOMP_TRACE,
+} mode = UTRACE;

 
 struct tracee {
        struct uloop_process proc;
 
 struct tracee {
        struct uloop_process proc;


@@ -63,22 +73,24 @@ struct tracee {
 };
 
 static struct tracee tracer;
 };
 
 static struct tracee tracer;

-static int *syscall_count;
+static int syscall_count[SYSCALL_COUNT];
+static int violation_count;

 static struct blob_buf b;
 static struct blob_buf b;

-static int syscall_max;

 static int debug;
 static int debug;

-
-static int max_syscall = ARRAY_SIZE(syscall_names);
+char *json = NULL;
+int ptrace_restart;

 
 static void set_syscall(const char *name, int val)
 {
        int i;
 
 
 static void set_syscall(const char *name, int val)
 {
        int i;
 

-       for (i = 0; i < max_syscall; i++)
-               if (syscall_names[i] && !strcmp(syscall_names[i], name)) {
+       for (i = 0; i < SYSCALL_COUNT; i++) {
+               int sc = syscall_index_to_number(i);
+               if (syscall_name(sc) && !strcmp(syscall_name(sc), name)) {

                        syscall_count[i] = val;
                        return;
                }
                        syscall_count[i] = val;
                        return;
                }

+       }

 }
 
 struct syscall {
 }
 
 struct syscall {


@@ -96,35 +108,37 @@ static void print_syscalls(int policy, const char *json)
        void *c;
        int i;
 
        void *c;
        int i;
 

-       set_syscall("rt_sigaction", 1);
-       set_syscall("sigreturn", 1);
-       set_syscall("rt_sigreturn", 1);
-       set_syscall("exit_group", 1);
-       set_syscall("exit", 1);
+       if (mode == UTRACE) {
+               set_syscall("rt_sigaction", 1);
+               set_syscall("sigreturn", 1);
+               set_syscall("rt_sigreturn", 1);
+               set_syscall("exit_group", 1);
+               set_syscall("exit", 1);
+       }

 
 

-       struct syscall sorted[ARRAY_SIZE(syscall_names)];
+       struct syscall sorted[SYSCALL_COUNT];

 
 

-       for (i = 0; i < ARRAY_SIZE(syscall_names); i++) {
-               sorted[i].syscall = i;
+       for (i = 0; i < SYSCALL_COUNT; i++) {
+               sorted[i].syscall = syscall_index_to_number(i);

                sorted[i].count = syscall_count[i];
        }
 
                sorted[i].count = syscall_count[i];
        }
 

-       qsort(sorted, ARRAY_SIZE(syscall_names), sizeof(sorted[0]), cmp_count);
+       qsort(sorted, SYSCALL_COUNT, sizeof(sorted[0]), cmp_count);

 
        blob_buf_init(&b, 0);
        c = blobmsg_open_array(&b, "whitelist");
 
 
        blob_buf_init(&b, 0);
        c = blobmsg_open_array(&b, "whitelist");
 

-       for (i = 0; i < ARRAY_SIZE(syscall_names); i++) {
+       for (i = 0; i < SYSCALL_COUNT; i++) {

                int sc = sorted[i].syscall;
                if (!sorted[i].count)
                        break;
                int sc = sorted[i].syscall;
                if (!sorted[i].count)
                        break;

-               if (syscall_names[sc]) {
+               if (syscall_name(sc)) {

                        if (debug)
                                printf("syscall %d (%s) was called %d times\n",
                        if (debug)
                                printf("syscall %d (%s) was called %d times\n",

-                                       sc, syscall_names[sc], sorted[i].count);
-                       blobmsg_add_string(&b, NULL, syscall_names[sc]);
+                                      sc, syscall_name(sc), sorted[i].count);
+                       blobmsg_add_string(&b, NULL, syscall_name(sc));

                } else {
                } else {

-                       ERROR("no name found for syscall(%d)\n", sc);
+                       ULOG_ERR("no name found for syscall(%d)\n", sc);

                }
        }
        blobmsg_close_array(&b, c);
                }
        }
        blobmsg_close_array(&b, c);


@@ -134,9 +148,9 @@ static void print_syscalls(int policy, const char *json)
                if (fp) {
                        fprintf(fp, "%s", blobmsg_format_json_indent(b.head, true, 0));
                        fclose(fp);
                if (fp) {
                        fprintf(fp, "%s", blobmsg_format_json_indent(b.head, true, 0));
                        fclose(fp);

-                       INFO("saving syscall trace to %s\n", json);
+                       ULOG_INFO("saving syscall trace to %s\n", json);

                } else {
                } else {

-                       ERROR("failed to open %s\n", json);
+                       ULOG_ERR("failed to open %s\n", json);

                }
        } else {
                printf("%s\n",
                }
        } else {
                printf("%s\n",


@@ -145,19 +159,48 @@ static void print_syscalls(int policy, const char *json)
 
 }
 
 
 }
 

+static void report_seccomp_vialation(pid_t pid, unsigned syscall)
+{
+       char buf[200];
+       snprintf(buf, sizeof(buf), "/proc/%d/cmdline", pid);
+       int f = open(buf, O_RDONLY);
+       int r = read(f, buf, sizeof(buf) - 1);
+       if (r >= 0)
+               buf[r] = 0;
+       else
+               strcpy(buf, "unknown?");
+       close(f);
+
+       if (violation_count < INT_MAX)
+               violation_count++;
+       int i = syscall_index(syscall);
+       if (i >= 0) {
+               syscall_count[i]++;
+               ULOG_ERR("%s[%u] tried to call non-whitelisted syscall: %s (see %s)\n",
+                        buf, pid,  syscall_name(syscall), json);
+       } else {
+               ULOG_ERR("%s[%u] tried to call non-whitelisted syscall: %d (see %s)\n",
+                        buf, pid,  syscall, json);
+       }
+}
+

 static void tracer_cb(struct uloop_process *c, int ret)
 {
        struct tracee *tracee = container_of(c, struct tracee, proc);
 static void tracer_cb(struct uloop_process *c, int ret)
 {
        struct tracee *tracee = container_of(c, struct tracee, proc);

+       int inject_signal = 0;

 
 

-       if (WIFSTOPPED(ret)) {
+       /* We explicitely check for events in upper 16 bits, because
+        * musl (as opposed to glibc) does not report
+        * PTRACE_EVENT_STOP as WIFSTOPPED */
+       if (WIFSTOPPED(ret) || (ret >> 16)) {

                if (WSTOPSIG(ret) & 0x80) {
                        if (!tracee->in_syscall) {
                                int syscall = ptrace(PTRACE_PEEKUSER, c->pid, reg_syscall_nr);
                if (WSTOPSIG(ret) & 0x80) {
                        if (!tracee->in_syscall) {
                                int syscall = ptrace(PTRACE_PEEKUSER, c->pid, reg_syscall_nr);

-
-                               if (syscall < syscall_max) {
-                                       syscall_count[syscall]++;
+                               int i = syscall_index(syscall);
+                               if (i >= 0) {
+                                       syscall_count[i]++;

                                        if (debug)
                                        if (debug)

-                                               fprintf(stderr, "%s()\n", syscall_names[syscall]);
+                                               fprintf(stderr, "%s()\n", syscall_name(syscall));

                                } else if (debug) {
                                        fprintf(stderr, "syscal(%d)\n", syscall);
                                }
                                } else if (debug) {
                                        fprintf(stderr, "syscal(%d)\n", syscall);
                                }


@@ -170,12 +213,28 @@ static void tracer_cb(struct uloop_process *c, int ret)
 
                        ptrace(PTRACE_GETEVENTMSG, c->pid, 0, &child->proc.pid);
                        child->proc.cb = tracer_cb;
 
                        ptrace(PTRACE_GETEVENTMSG, c->pid, 0, &child->proc.pid);
                        child->proc.cb = tracer_cb;

-                       ptrace(PTRACE_SYSCALL, child->proc.pid, 0, 0);
+                       ptrace(ptrace_restart, child->proc.pid, 0, 0);

                        uloop_process_add(&child->proc);
                        if (debug)
                                fprintf(stderr, "Tracing new child %d\n", child->proc.pid);
                        uloop_process_add(&child->proc);
                        if (debug)
                                fprintf(stderr, "Tracing new child %d\n", child->proc.pid);

+               } else if ((ret >> 16) == PTRACE_EVENT_STOP) {
+                       /* Nothing special to do here */
+               } else if ((ret >> 8) == (SIGTRAP | (PTRACE_EVENT_SECCOMP << 8))) {
+                       int syscall = ptrace(PTRACE_PEEKUSER, c->pid, reg_syscall_nr);
+#if defined(__arm__)
+                       ptrace(PTRACE_SET_SYSCALL, c->pid, 0, -1);
+                       ptrace(PTRACE_POKEUSER, c->pid, reg_retval_nr, -ENOSYS);
+#else
+                       ptrace(PTRACE_POKEUSER, c->pid, reg_syscall_nr, -1);
+#endif
+                       report_seccomp_vialation(c->pid, syscall);
+               } else {
+                       inject_signal = WSTOPSIG(ret);
+                       if (debug)
+                               fprintf(stderr, "Injecting signal %d into pid %d\n",
+                                       inject_signal, tracee->proc.pid);

                }
                }

-       } else if (WIFEXITED(ret)) {
+       } else if (WIFEXITED(ret) || (WIFSIGNALED(ret) && WTERMSIG(ret))) {

                if (tracee == &tracer) {
                        uloop_end(); /* Main process exit */
                } else {
                if (tracee == &tracer) {
                        uloop_end(); /* Main process exit */
                } else {


@@ -186,16 +245,29 @@ static void tracer_cb(struct uloop_process *c, int ret)
                return;
        }
 
                return;
        }
 

-       ptrace(PTRACE_SYSCALL, c->pid, 0, 0);
+       ptrace(ptrace_restart, c->pid, 0, inject_signal);

        uloop_process_add(c);
 }
 
        uloop_process_add(c);
 }
 

+static void sigterm_handler(int signum)
+{
+       /* When we receive SIGTERM, we forward it to the tracee. After
+        * the tracee exits, trace_cb() will be called and make us
+        * exit too. */
+       kill(tracer.proc.pid, SIGTERM);
+}
+
+

 int main(int argc, char **argv, char **envp)
 {
 int main(int argc, char **argv, char **envp)
 {

-       char *json = NULL;

        int status, ch, policy = EPERM;
        pid_t child;
 
        int status, ch, policy = EPERM;
        pid_t child;
 

+       /* When invoked via seccomp-trace symlink, work as seccomp
+        * violation logger rather than as syscall tracer */
+       if (strstr(argv[0], "seccomp-trace"))
+               mode = SECCOMP_TRACE;
+

        while ((ch = getopt(argc, argv, "f:p:")) != -1) {
                switch (ch) {
                case 'f':
        while ((ch = getopt(argc, argv, "f:p:")) != -1) {
                switch (ch) {
                case 'f':


@@ -207,6 +279,9 @@ int main(int argc, char **argv, char **envp)
                }
        }
 
                }
        }
 

+       if (!json)
+               json = getenv("SECCOMP_FILE");
+

        argc -= optind;
        argv += optind;
 
        argc -= optind;
        argv += optind;
 


@@ -222,7 +297,9 @@ int main(int argc, char **argv, char **envp)
        if (child == 0) {
                char **_argv = calloc(argc + 1, sizeof(char *));
                char **_envp;
        if (child == 0) {
                char **_argv = calloc(argc + 1, sizeof(char *));
                char **_envp;

-               char *preload = "LD_PRELOAD=/lib/libpreload-trace.so";
+               char *preload = NULL;
+               const char *old_preload = getenv("LD_PRELOAD");
+               int newenv = 0;

                int envc = 0;
                int ret;
 
                int envc = 0;
                int ret;
 


@@ -231,12 +308,26 @@ int main(int argc, char **argv, char **envp)
                while (envp[envc++])
                        ;
 
                while (envp[envc++])
                        ;
 

-               _envp = calloc(envc + 1, sizeof(char *));
-               memcpy(&_envp[1], envp, envc * sizeof(char *));
-               *_envp = preload;
+               _envp = calloc(envc + 2, sizeof(char *));
+               switch (mode) {
+               case UTRACE:
+                       preload = "/lib/libpreload-trace.so";
+                       newenv = 1;
+                       break;
+               case SECCOMP_TRACE:
+                       preload = "/lib/libpreload-seccomp.so";
+                       newenv = 2;
+                       asprintf(&_envp[1], "SECCOMP_FILE=%s", json ? json : "");
+                       kill(getpid(), SIGSTOP);
+                       break;
+               }
+               asprintf(&_envp[0], "LD_PRELOAD=%s%s%s", preload,
+                        old_preload ? ":" : "",
+                        old_preload ? old_preload : "");
+               memcpy(&_envp[newenv], envp, envc * sizeof(char *));

 
                ret = execve(_argv[0], _argv, _envp);
 
                ret = execve(_argv[0], _argv, _envp);

-               ERROR("failed to exec %s: %s\n", _argv[0], strerror(errno));
+               ULOG_ERR("failed to exec %s: %m\n", _argv[0]);

 
                free(_argv);
                free(_envp);
 
                free(_argv);
                free(_envp);


@@ -246,33 +337,55 @@ int main(int argc, char **argv, char **envp)
        if (child < 0)
                return -1;
 
        if (child < 0)
                return -1;
 

-       syscall_max = ARRAY_SIZE(syscall_names);
-       syscall_count = calloc(syscall_max, sizeof(int));
-       waitpid(child, &status, 0);
+       waitpid(child, &status, WUNTRACED);

        if (!WIFSTOPPED(status)) {
        if (!WIFSTOPPED(status)) {

-               ERROR("failed to start %s\n", *argv);
+               ULOG_ERR("failed to start %s\n", *argv);

                return -1;
        }
 
                return -1;
        }
 

-       ptrace(PTRACE_SETOPTIONS, child, 0,
-              PTRACE_O_TRACESYSGOOD |
-              PTRACE_O_TRACEFORK |
-              PTRACE_O_TRACEVFORK |
-              PTRACE_O_TRACECLONE);
-       ptrace(PTRACE_SYSCALL, child, 0, 0);
-
+       /* Initialize uloop to catch all ptrace stops from now on. */

        uloop_init();
        uloop_init();

+
+       int ptrace_options = PTRACE_O_TRACEFORK | PTRACE_O_TRACEVFORK | PTRACE_O_TRACECLONE;
+       switch (mode) {
+       case UTRACE:
+               ptrace_options |= PTRACE_O_TRACESYSGOOD;
+               ptrace_restart = PTRACE_SYSCALL;
+               break;
+       case SECCOMP_TRACE:
+               ptrace_options |= PTRACE_O_TRACESECCOMP;
+               ptrace_restart = PTRACE_CONT;
+               break;
+       }
+       if (ptrace(PTRACE_SEIZE, child, 0, ptrace_options) == -1) {
+               ULOG_ERR("PTRACE_SEIZE: %m\n");
+               return -1;
+       }
+       if (ptrace(ptrace_restart, child, 0, SIGCONT) == -1) {
+               ULOG_ERR("ptrace_restart: %m\n");
+               return -1;
+       }
+

        tracer.proc.pid = child;
        tracer.proc.cb = tracer_cb;
        uloop_process_add(&tracer.proc);
        tracer.proc.pid = child;
        tracer.proc.cb = tracer_cb;
        uloop_process_add(&tracer.proc);

+       signal(SIGTERM, sigterm_handler); /* Override uloop's SIGTERM handler */

        uloop_run();
        uloop_done();
 
        uloop_run();
        uloop_done();
 

-       if (!json)
-               if (asprintf(&json, "/tmp/%s.%u.json", basename(*argv), child) < 0)
-                       ERROR("failed to allocate output path: %s\n", strerror(errno));

 
 

+       switch (mode) {
+       case UTRACE:
+               if (!json)
+                       if (asprintf(&json, "/tmp/%s.%u.json", basename(*argv), child) < 0)
+                               ULOG_ERR("failed to allocate output path: %m\n");
+               break;
+       case SECCOMP_TRACE:
+               if (!violation_count)
+                       return 0;
+               asprintf(&json, "/tmp/%s.%u.violations.json", basename(*argv), child);
+               break;
+       }

        print_syscalls(policy, json);
        print_syscalls(policy, json);

-

        return 0;
 }
        return 0;
 }