no_fs: Use landlock if possible

This is not 100% the same, but good enough and more importantly,
does not require unsharing user/mount namespace and the chroot call.

README.md

@@ -1,60 +1,168 @@
 # exile.h
-`exile.h` is a header-only library, enabling processes to easily isolate themselves on Linux for exploit mitigation. exile.h wants to make existing technologies, such as Seccomp and Linux Namespaces, easier to use. Those generally
-require knowledge of details and are not trivial for developers to employ, which prevents a more widespread adoption.
+`exile.h` provides an API for processes on Linux to easily isolate themselves in order
+to mitigate the effect of exploited vulnerabilities, i. e. when attacker has achieved
+arbitrary code execution. exile.h makes it simpler for developers to use existing technologies such as Seccomp and Linux Namespaces. Those generally require knowledge of details and are not trivial for developers to employ, which prevents a more widespread adoption.
 
-The following section gives small quick examples. Then the motivation is explained in more detail.
-Proper API documentation will be maintained in other files.
+The following section offers small examples. Then the motivation is explained in more detail. Proper API documentation will be maintained in other files.
 
 ## Quick demo
-TODO This section will demonstrate the simplicity of the API, but only serves as an overview.
+This section quickly demonstrates the simplicity of the API. It serves as an overview to get a first impression.
+
+system() is used to keep the example C code short. It also demonstrates that subprocesses are also subject to restrictions imposed by exile.h.
+
+While the examples show different features separately, it is generally possible to combine those.
 
 ### Filesystem isolation
+```c
+#include "exile.h"
+#include <assert.h>
+int main(void)
+{
+	system("echo test > /home/user/testfile");
+	struct exile_policy *policy = exile_init_policy();
+	exile_append_path_policies(policy, EXILE_FS_ALLOW_ALL_READ, "/home/user");
+	exile_append_path_policies(policy, EXILE_FS_ALLOW_ALL_READ | EXILE_FS_ALLOW_ALL_WRITE, "/tmp");
+	int ret = exile_enable_policy(policy);
+	if(ret != 0)
+	{
+		exit(EXIT_FAILURE);
+	}
+	int fd = open("/home/user/test", O_CREAT | O_WRONLY | O_TRUNC, 0600);
+	assert(fd == -1);
+	fd = open("/home/user/testfile", O_RDONLY);
+	//use fd
+	assert(fd != -1);
+	fd = open("/tmp/testfile", O_CREAT | O_WRONLY | O_TRUNC, 0600);
+	//use fd
+	assert(fd != -1);
+	return 0;
+}
+```
 
+The assert() calls won't be fired, consistent with the policy that allows only reading
+from /home/user. We can write to /tmp/ though as it was specified in the policy.
 
+### vows(): pledge()-like API / System call policies
+exile.h allows specifying which syscalls are permitted or denied. In the following example,
+'ls' is never executed, as the specified "vows" do not allow the execve() system call. The process will be killed.
 
-### System call policies / vows
+```c
+#include "exile.h"
 
+int main(void)
+{
+	struct exile_policy *policy = exile_init_policy();
+	policy->vow_promises = exile_vows_from_str("stdio rpath wpath cpath");
+	exile_enable_policy(policy);
+	printf("Trying to execute...");
+	execlp("/bin/ls", "ls", "/", NULL);
+}
+```
 
-### Isolation of single functions
-exile_launch() demo
+### Isolation from network
+exile offers a quick way to isolate a process from the default network namespace.
+
+```c
+#include "exile.h"
+
+int main(void)
+{
+	struct exile_policy *policy = exile_init_policy();
+	policy->namespace_options |= EXILE_UNSHARE_NETWORK;
+	int ret = exile_enable_policy(policy);
+	if(ret != 0)
+	{
+		exit(EXIT_FAILURE);
+	}
+	system("curl -I https://evil.tld");
+}
+```
+Produces ```curl: (6) Could not resolve host: evil.tld```. For example, this is useful for subprocesses which do not need
+network access, but perform tasks such as parsing user-supplied file formats.
+
+### Isolation of single functions (EXPERIMENTAL)
+Currently, work is being done that hopefully will allow isolation of individual function calls in a mostly pain-free manner.
+
+Consider the following C++ code:
+```cpp
+#include <iostream>
+#include <fstream>
+#include "exile.hpp"
+std::string cat(std::string path)
+{
+	std::fstream f1;
+	f1.open(path.c_str(), std::ios::in);
+	std::string content;
+	std::string line;
+	while(getline(f1, line)) {
+		content += line + "\n";
+	}
+	return content;
+}
+
+int main(void)
+{
+	struct exile_policy *policy = exile_init_policy();
+	policy->vow_promises = exile_vows_from_str("stdio rpath");
+
+	std::string content = exile_launch<std::string>(policy, cat, "/etc/hosts");
+	std::cout << content;
+
+	policy = exile_init_policy();
+	policy->vow_promises = exile_vows_from_str("stdio");
+
+	try
+	{
+	content = exile_launch<std::string>(policy, cat, "/etc/hosts");
+	std::cout << content;
+	}
+	catch(std::exception &e)
+	{
+		std::cout << "launch failure: " << e.what() << std::endl;
+	}
+}
+```
+
+We execute "cat()". The first call succeeds. In the second, we get an exception, because
+the subprocess "cat()" was launched in violated the policy (missing "rpath" vow). 
+
+Naturally, there is a performance overhead. Certain challenges remain, such as the fact
+that being executed in a subproces, we operate on copies, so handling references
+is not something that has been given much thought. There is also the fact
+that clone()ing from threads opens a can of worms, particularly with locks. Hence, exile_launch() is best avoided in multi-threaded contexts.
 
 ## Status
 No release yet, experimental, API is unstable, builds will break on updates of this library.
 
-Currently, it's mainly evolving from the needs of my other projects.
+Currently, it's mainly evolving from the needs of my other projects which use exile.h.
+
+
+### Real-world usage
+  - looqs: https://github.com/quitesimpleorg/looqs
+  - qswiki: https://gitea.quitesimple.org/crtxcr/qswiki
+
   
 ## Motivation and Background
-exile.h unlocks existing Linux mechanisms to facilite isolation of processes from resources. Limiting the scope of what programs can do helps defending the rest of the system when a process gets under attacker's control (when classic mitigations such as ASLR etc. failed). To this end, OpenBSD has the pledge() and unveil() functions available. Those functions are helpful mitigation mechanisms, but such accessible ways are unfortunately not readily available on Linux. This is where exile.h steps in.
+exile.h unlocks existing Linux mechanisms to facilitate isolation of processes from resources. Limiting the scope of what programs can do helps defending the rest of the system when a process gets under attacker's control (when classic mitigations such as ASLR etc. failed). To this end, OpenBSD has the pledge() and unveil() functions available. Those functions are helpful mitigation mechanisms, but such accessible ways are unfortunately not readily available on Linux. This is where exile.h steps in.
 
-Seccomp allows to restrict system calls available to a process and thus decrease the systems attack surface, but it generally is not easy to use. Requiring BPF filter instructions, you generally just can't make use of it right away. exile.h provides an API inspired by pledge(), building on top of seccomp. It also provides an interface to manually restrict the system calls that can be issued.
+Seccomp allows restricting the system calls available to a process and thus decrease the systems attack surface, but it generally is not easy to use. Requiring BPF filter instructions, you generally just can't make use of it right away without learning
+about BPF. exile.h provides an API inspired by pledge(), building on top of seccomp. It also provides an interface to manually restrict the system calls that can be issued.
 
-Traditional methods employed to restrict file system access, like different uids/gids, chroot, bind-mounts, namespaces etc. may require administrator intervention, are perhaps only suitable
-for daemons and not desktop applications, or are generally rather involved. As a positive example, Landlock since 5.13  is a vast improvement to limit file system access of processes. It also greatly simplifies exile.h' implementation of fs isolation.
+Traditional methods employed to restrict file system access, like different uids/gids, chroot, bind-mounts, namespaces etc. may require administrator intervention, are perhaps only suitable for daemons and not desktop applications, or are generally rather involved. As a positive example, Landlock since 5.13 is a vast improvement to limit file system access of processes. It also greatly simplifies exile.h' implementation of fs isolation.
 
 Abstracting those details may help developers bring sandboxing into their applications.
 
-## Example: Archive extraction
-A programming uncompressing archives does not need network access, but should a bug allow code execution, obviously the payload may also access the network. Once the target path is known, it doesn't need access to the whole file system, only write-permissions to the target directory and read on the archive file(s).
-
-TODO example with exile.h applied on "tar" or "unzip". Link to repo.
-
-## Example: Web apps
-Those generally don't need access to the whole filesystem hierarchy, nor do they necessarily require the ability to execute other processes.
-
-Way more examples can be given, but we can put it in simple words: A general purpose OS allow a process to do more things than it actually needs to do.
-
 ## Features
   - Restricting file system access (using Landlock or Namespaces/chroot as fallback)
-  - Systemcall filtering (using seccomp-bpf). An interface inspired by OpenBSD's pledge() is available, removing the need to specifc rules for syscalls.
+  - Systemcall filtering (using seccomp-bpf). An interface inspired by OpenBSD's pledge() is available
   - Dropping privileges in general, such as capabilities
   - Isolating the application from the network, etc. through Namespaces
   - Helpers to isolate single functions
 
 
 ## What it's not
-A way for end users/administrators to restrict processes. In the future, a wrapper binary may be available to achieve this, but it generally aims for developers to bring sandboxing/isolation into their software, like web browsers do. This allows a more fine-grained approach, as the developers
-is more familiar with the software. Applying restrictions with solutions like AppArmor requires
-them to be present on the system and it's easy to break things this way.
+A way for end users/administrators to restrict processes. In the future, a wrapper binary may be available to achieve this, but it generally aims for developers to bring sandboxing/isolation into their software. This allows a more fine-grained approach, as the developers are more familiar with their software. Applying restrictions with solutions like AppArmor requires
+them to be present and installed on the system and it's easy to break things this way.
 
 Therefore, software should ideally be written with sandboxing in mind from the beginning.
 
@@ -66,40 +174,44 @@ It's recommended to start with [README.usage.md] to get a feeling for exile.h.
 API-Documentation: [README.api.md]
 
 ## Limitations
+Built upon kernel technologies, exile.h naturally inherits their limitations:
+
+  - New syscalls can be introduced by new kernel versions. exile.h must keep in sync, and users must keep the library up to date.
+  - seccomp has no deep argument inspection (yet), particularly new syscalls
+  cannot be reasonably filtered, such as clone3(), or io_uring.
+  - You can't know what syscalls libraries will issue. An update to existing
+  libraries may cause them to use different syscalls not allowed by a policy. However, using vows and keeping up to date with exile.h should cover that. 
+  - Landlock, currently, does not apply to syscalls such as stat().
+  
 TODO: 
- - seccomp must be kept up to date syscalls kernel
  - ioctl does not know the fd, so checking values is kind of strange
  - redundancies: some things are handled by capabilties, other by seccomp or both
- - seccomp no deep argument inspection
- - landlock: stat() does not apply
  - no magic, be reasonable, devs should not  get sloppy, restrict IPC.
 
 ## Requirements
 Kernel >=3.17
 
-While mostly transparent to users of this API, kernel >= 5.13 is required to take advantage of Landlock and furthermore it depends on distro-provided kernels being reasonable and enabling it by default. In practise, this means that Landlock probably won't be used for now, and exile.h will use a combination of namespaces, bind mounts and chroot as fallbacks.
+While mostly transparent to users of this API, kernel >= 5.13 is required to take advantage of Landlock. Furthermore, it depends on distro-provided kernels being reasonable and enabling it by default. In practise, Landlock maybe won't be used in some cases so exile.h will use a combination of namespaces, bind mounts and chroot as fallbacks.
 
 
 ## FAQ
 
 
-### Does the process need to be priviliged to utilize the library?
+### Does the process need to be privileged to utilize the library?
 
 No.
 
-### It doesn't work on Debian!
-
-You can thank a Debian-specific kernel patch for that. In the future,
-the library may check against that. Execute
+### It doesn't work on my Debian version!
+You can thank a Debian-specific kernel patch for that. Execute
 `echo 1 > /proc/sys/kernel/unprivileged_userns_clone` to disable that patch for now.
 
-### Examples
-  - looqs: https://gitea.quitesimple.org/crtxcr/looqs
-  - qswiki: https://gitea.quitesimple.org/crtxcr/qswiki
+Note that newer releases should not cause this problem any longer, as [explained](https://www.debian.org/releases/bullseye/amd64/release-notes/ch-information.en.html#linux-user-namespaces) in the Debian release notes.
 
-Outdated:
-  - cgit sandboxed: https://gitea.quitesimple.org/crtxcr/cgitsb
-  - qpdfviewsb sandboxed (quick and dirty): https://gitea.quitesimple.org/crtxcr/qpdfviewsb
+### Why "vows"?
+pledge() cannot be properly implemented using seccomp. The "vow" concept here may look similiar, and it is, but it's not pledge(). 
+
+### Other projects
+ - [sandbox2](https://developers.google.com/code-sandboxing/sandbox2/)
 
 
 ### Contributing

exile.c

@@ -274,11 +274,13 @@ static struct syscall_vow_map exile_vow_map[] =
 	{EXILE_SYS(sched_getattr), EXILE_SYSCALL_VOW_SCHED},
 	{EXILE_SYS(renameat2), EXILE_SYSCALL_VOW_CPATH},
 	{EXILE_SYS(getrandom), EXILE_SYSCALL_VOW_STDIO},
+	{EXILE_SYS(memfd_create), EXILE_SYSCALL_VOW_STDIO},
 	{EXILE_SYS(execveat), EXILE_SYSCALL_VOW_EXEC},
 	{EXILE_SYS(mlock2), EXILE_SYSCALL_VOW_STDIO},
 	{EXILE_SYS(copy_file_range), EXILE_SYSCALL_VOW_STDIO},
 	{EXILE_SYS(statx), EXILE_SYSCALL_VOW_RPATH},
-	{EXILE_SYS(clone3), EXILE_SYSCALL_VOW_CLONE},
+	{EXILE_SYS(rseq), EXILE_SYSCALL_VOW_THREAD},
+	{EXILE_SYS(clone3), EXILE_SYSCALL_VOW_CLONE|EXILE_SYSCALL_VOW_THREAD},
 	{EXILE_SYS(close_range), EXILE_SYSCALL_VOW_STDIO},
 	{EXILE_SYS(openat2), EXILE_SYSCALL_VOW_RPATH|EXILE_SYSCALL_VOW_WPATH},
 	{EXILE_SYS(faccessat2), EXILE_SYSCALL_VOW_RPATH},
@@ -359,10 +361,11 @@ inline int exile_landlock_is_available()
 {
 	#if HAVE_LANDLOCK == 1
 	int ruleset = landlock_create_ruleset(NULL, 0, LANDLOCK_CREATE_RULESET_VERSION);
-	return ruleset == 1;
+	return ruleset > 0;
 	#endif
 	return 0;
 }
+
 int exile_append_syscall_policy(struct exile_policy *exile_policy, long syscall, unsigned int syscall_policy, struct sock_filter *argfilters, size_t n)
 {
 	struct exile_syscall_policy *newpolicy = (struct exile_syscall_policy *) calloc(1, sizeof(struct exile_syscall_policy));
@@ -379,6 +382,7 @@ int exile_append_syscall_policy(struct exile_policy *exile_policy, long syscall,
 	{
 		EXILE_LOG_ERROR("Too many argfilters supplied\n");
 		exile_policy->exile_flags |= EXILE_FLAG_ADD_SYSCALL_POLICY_FAIL;
+		free(newpolicy);
 		return -1;
 	}
 	for(size_t i = 0; i < n; i++)
@@ -430,6 +434,7 @@ int get_vow_argfilter(long syscall, uint64_t vow_promises, struct sock_filter *f
 
 	struct exile_syscall_filter ioctl_filter[] = {
 		EXILE_SYSCALL_FILTER_LOAD_ARG(1),
+		{ EXILE_SYSCALL_VOW_IOCTL, EXILE_BPF_NO_MATCH_SET(TIOCSTI), 1 },
 		{ EXILE_SYSCALL_VOW_IOCTL, EXILE_BPF_RETURN_MATCHING, 1 },
 		{ EXILE_SYSCALL_VOW_STDIO, EXILE_BPF_MATCH(FIONREAD), 1},
 		{ EXILE_SYSCALL_VOW_STDIO, EXILE_BPF_MATCH(FIONBIO), 1},
@@ -518,7 +523,7 @@ int get_vow_argfilter(long syscall, uint64_t vow_promises, struct sock_filter *f
 			current_count = COUNT_EXILE_SYSCALL_FILTER(open_filter);
 			break;
 		case EXILE_SYS(openat2):
-			*policy = EXILE_SYSCALL_DENY_RET_ERROR;
+			*policy = EXILE_SYSCALL_DENY_RET_NOSYS;
 			return 0;
 			break;
 		case EXILE_SYS(socket):
@@ -536,7 +541,7 @@ int get_vow_argfilter(long syscall, uint64_t vow_promises, struct sock_filter *f
 		case EXILE_SYS(clone3):
 			if((vow_promises & EXILE_SYSCALL_VOW_CLONE) == 0)
 			{
-				*policy = EXILE_SYSCALL_DENY_RET_ERROR;
+				*policy = EXILE_SYSCALL_DENY_RET_NOSYS;
 				return 0;
 			}
 			break;
@@ -616,10 +621,12 @@ struct exile_policy *exile_init_policy()
 	{
 		return NULL;
 	}
-	result->drop_caps = 1;
+	result->drop_caps = 0;
 	result->not_dumpable = 1;
 	result->no_new_privs = 1;
-	result->namespace_options = EXILE_UNSHARE_MOUNT | EXILE_UNSHARE_USER;
+	result->namespace_options = EXILE_UNSHARE_AUTOMATIC;
+	result->namespace_uid = 0;
+	result->namespace_gid = 0;
 	return result;
 }
 
@@ -643,7 +650,7 @@ int (exile_append_path_policies)(struct exile_policy *exile_policy, unsigned int
 		int fd = open(path, O_PATH);
 		if(fd == -1)
 		{
-			EXILE_LOG_ERROR("Failed to open the specified path: %s\n", strerror(errno));
+			EXILE_LOG_ERROR("Failed to open %s: %s\n", path, strerror(errno));
 			exile_policy->exile_flags |= EXILE_FLAG_ADD_PATH_POLICY_FAIL;
 			return -1;
 		}
@@ -811,11 +818,13 @@ char *concat_path(const char *first, const char *second)
 	if(written < 0)
 	{
 		EXILE_LOG_ERROR("Error during path concatination\n");
+		free(result);
 		return NULL;
 	}
 	if(written >= PATH_MAX)
 	{
 		EXILE_LOG_ERROR("path concatination truncated\n");
+		free(result);
 		return NULL;
 	}
 	return result;
@@ -851,7 +860,7 @@ static int create_chroot_dirs(const char *chroot_target_path, struct exile_path_
 		ret = mkpath(path_inside_chroot, 0700, baseisfile);
 		if(ret < 0)
 		{
-			EXILE_LOG_ERROR("Error creating directory structure while mounting paths to chroot. %s\n", strerror(errno));
+			EXILE_LOG_ERROR("Error creating directory structure %s while mounting paths to chroot: %s\n", path_inside_chroot, strerror(errno));
 			free(path_inside_chroot);
 			return ret;
 		}
@@ -865,6 +874,8 @@ static int create_chroot_dirs(const char *chroot_target_path, struct exile_path_
 static int perform_mounts(const char *chroot_target_path, struct exile_path_policy *path_policy)
 {
 	while(path_policy != NULL)
+	{
+		if(path_policy->policy & EXILE_FS_ALLOW_ALL_READ || path_policy->policy & EXILE_FS_ALLOW_ALL_WRITE)
 		{
 			int mount_flags = get_policy_mount_flags(path_policy);
 
@@ -876,8 +887,6 @@ static int perform_mounts(const char *chroot_target_path, struct exile_path_poli
 			//all we do is bind mounts
 			mount_flags |= MS_BIND;
 
-		if(path_policy->policy & EXILE_FS_ALLOW_ALL_READ || path_policy->policy & EXILE_FS_ALLOW_ALL_WRITE)
-		{
 			int ret = mount(path_policy->path, path_inside_chroot,  NULL, mount_flags, NULL);
 			if(ret < 0 )
 			{
@@ -894,9 +903,10 @@ static int perform_mounts(const char *chroot_target_path, struct exile_path_poli
 				free(path_inside_chroot);
 				return ret;
 			}
-			path_policy = path_policy->next;
+
 			free(path_inside_chroot);
 		}
+		path_policy = path_policy->next;
 	}
 	return 0;
 }
@@ -930,10 +940,15 @@ void exile_free_policy(struct exile_policy *ctxt)
 }
 
 /* Enters the specified namespaces */
-static int enter_namespaces(int namespace_options)
+static int enter_namespaces(int namespace_options, uid_t namespace_uid, gid_t namespace_gid)
 {
 	if(namespace_options & EXILE_UNSHARE_USER)
 	{
+		uid_t current_uid = getuid();
+		gid_t current_gid = getgid();
+
+		char buf[1024] = {0};
+
 		int ret = unshare(CLONE_NEWUSER);
 		if(ret == -1)
 		{
@@ -941,47 +956,51 @@ static int enter_namespaces(int namespace_options)
 			return ret;
 		}
 
-		uid_t current_uid = getuid();
-		gid_t current_gid = getgid();
+		int fd = open("/proc/self/setgroups", O_WRONLY);
+		if(fd == -1)
+		{
+			EXILE_LOG_ERROR("Failed to open /proc/self/setgroups for writing\n");
+			return -1;
+		}
+		int writesize = snprintf(buf, sizeof(buf), "deny");
+		int writeret = write(fd, buf, writesize);
+		if(writeret < 0 || writeret < writesize)
+		{
+			EXILE_LOG_ERROR("Failed to write to /proc/self/setgroups: %i (%s)\n", writeret, strerror(errno));
+			return -1;
+		}
+		close(fd);
 
-		FILE *fp = fopen("/proc/self/setgroups", "w");
-		if(fp == NULL)
+		fd = open("/proc/self/uid_map", O_WRONLY);
+		if(fd == -1)
 		{
-			EXILE_LOG_ERROR("fopen failed while trying to deny setgroups\n");
+			EXILE_LOG_ERROR("Failed to open /proc/self/uid_map for writing\n");
 			return -1;
 		}
-		if(fprintf(fp, "deny") < 0)
+		writesize = snprintf(buf, sizeof(buf), "%u %u 1\n", namespace_uid, current_uid);
+		writeret = write(fd, buf, writesize);
+		if(writeret < 0 || writeret < writesize)
 		{
-			EXILE_LOG_ERROR("fprintf failed while trying to write setgroups\n");
+			EXILE_LOG_ERROR("Failed to write to /proc/self/uid_map: %i (%s)\n", writeret, strerror(errno));
 			return -1;
 		}
-		fclose(fp);
+		close(fd);
 
-		fp = fopen("/proc/self/uid_map", "w");
-		if(fp == NULL)
-		{
-			EXILE_LOG_ERROR("fopen failed while trying to write uid_map\n");
-			return -1;
-		}
-		if(fprintf(fp, "0 %i", current_uid) < 0)
-		{
-			EXILE_LOG_ERROR("fprintf failed while trying to write uid_map\n");
-			return -1;
-		}
-		fclose(fp);
 
-		fp = fopen("/proc/self/gid_map", "w");
-		if(fp == NULL)
+		fd = open("/proc/self/gid_map", O_WRONLY);
+		if(fd == -1)
 		{
-			EXILE_LOG_ERROR("fopen failed while trying to write gid_map\n");
+			EXILE_LOG_ERROR("Failed to open /proc/self/gid_map for writing\n");
 			return -1;
 		}
-		if(fprintf(fp, "0 %i", current_gid) < 0)
+		writesize = snprintf(buf, sizeof(buf), "%u %u 1\n", namespace_gid, current_gid);
+		writeret = write(fd, buf, writesize);
+		if(writeret < 0 || writeret < writesize)
 		{
-			EXILE_LOG_ERROR("fprintf failed while trying to write gid_map\n");
+			EXILE_LOG_ERROR("Failed to write to /proc/self/gid_map: %i (%s)\n", writeret, strerror(errno));
 			return -1;
 		}
-		fclose(fp);
+		close(fd);
 	}
 
 	if(namespace_options & EXILE_UNSHARE_MOUNT)
@@ -1072,6 +1091,10 @@ static struct sock_filter *append_syscall_to_bpf(struct exile_syscall_policy *sy
 	{
 		action = SECCOMP_RET_ERRNO|EACCES;
 	}
+	if(action == EXILE_SYSCALL_DENY_RET_NOSYS)
+	{
+		action = SECCOMP_RET_ERRNO|ENOSYS;
+	}
 	long syscall = syscallpolicy->syscall;
 
 	struct sock_filter syscall_load = BPF_STMT(BPF_LD+BPF_W+BPF_ABS, offsetof(struct seccomp_data, nr));
@@ -1138,7 +1161,7 @@ static struct sock_filter *append_syscall_to_bpf(struct exile_syscall_policy *sy
 
 static int is_valid_syscall_policy(unsigned int policy)
 {
-	return policy == EXILE_SYSCALL_ALLOW || policy == EXILE_SYSCALL_DENY_RET_ERROR || policy == EXILE_SYSCALL_DENY_KILL_PROCESS;
+	return policy == EXILE_SYSCALL_ALLOW || policy == EXILE_SYSCALL_DENY_RET_ERROR || policy == EXILE_SYSCALL_DENY_KILL_PROCESS || policy == EXILE_SYSCALL_DENY_RET_NOSYS;
 }
 
 /*
@@ -1206,12 +1229,21 @@ static unsigned int exile_flags_to_landlock(unsigned int flags, int statmode)
 	if(flags & EXILE_FS_ALLOW_ALL_WRITE)
 	{
 		result |= LANDLOCK_ACCESS_FS_WRITE_FILE;
+#ifdef LANDLOCK_ACCESS_FS_TRUNCATE
+		result |= LANDLOCK_ACCESS_FS_TRUNCATE;
+#endif
 		if(S_ISDIR(statmode))
 		{
-			result |= LANDLOCK_ACCESS_FS_REMOVE_FILE;
-			result |= LANDLOCK_ACCESS_FS_MAKE_REG;
 			result |= LANDLOCK_ACCESS_FS_REMOVE_DIR;
+			result |= LANDLOCK_ACCESS_FS_REMOVE_FILE;
+			result |= LANDLOCK_ACCESS_FS_MAKE_DIR;
+			result |= LANDLOCK_ACCESS_FS_MAKE_FIFO;
+			result |= LANDLOCK_ACCESS_FS_MAKE_REG;
+			result |= LANDLOCK_ACCESS_FS_MAKE_SOCK;
 			result |= LANDLOCK_ACCESS_FS_MAKE_SYM;
+#ifdef LANDLOCK_ACCESS_FS_REFER
+			result |= LANDLOCK_ACCESS_FS_REFER;
+#endif
 		}
 	}
 	if(flags & EXILE_FS_ALLOW_EXEC)
@@ -1278,15 +1310,42 @@ static unsigned int exile_flags_to_landlock(unsigned int flags, int statmode)
 	return result;
 }
 
+/* Sets maximum values for the handled access fs... */
+static int landlock_set_max_handled_access(struct landlock_ruleset_attr *ruleset)
+{
+	int abi = landlock_create_ruleset(NULL, 0,
+                                  LANDLOCK_CREATE_RULESET_VERSION);
+	if(abi < 0)
+	{
+		EXILE_LOG_ERROR("Can't determine landlock ABI version\n");
+		return -1;
+	}
+	ruleset->handled_access_net = 0;
+	if(abi == 1)
+	{
+		ruleset->handled_access_fs = ((LANDLOCK_ACCESS_FS_MAKE_SYM << 1) - 1);
+	}
+	if(abi == 2)
+	{
+		ruleset->handled_access_fs = ((LANDLOCK_ACCESS_FS_REFER << 1) - 1);
+	}
+	if(abi >= 3)
+	{
+		ruleset->handled_access_fs = ((LANDLOCK_ACCESS_FS_TRUNCATE << 1) - 1);
+		/* TODO: think about net */
+	}
+	return 0;
+
+}
+
 static int landlock_prepare_ruleset(struct exile_path_policy *policies)
 {
 	int ruleset_fd = -1;
-	struct landlock_ruleset_attr ruleset_attr;
-	/* We here want the maximum possible ruleset, so set the var to the max possible bitmask.
-	   Stolen/Adapted from: [linux src]/security/landlock/limits.h
-	*/
-	ruleset_attr.handled_access_fs = ((LANDLOCK_ACCESS_FS_MAKE_SYM << 1) - 1);
-
+	struct landlock_ruleset_attr ruleset_attr = {0};
+	if(landlock_set_max_handled_access(&ruleset_attr) != 0)
+	{
+		return -1;
+	}
 	ruleset_fd = landlock_create_ruleset(&ruleset_attr, sizeof(ruleset_attr), 0);
 	if (ruleset_fd < 0)
 	{
@@ -1296,7 +1355,7 @@ static int landlock_prepare_ruleset(struct exile_path_policy *policies)
 	struct exile_path_policy *policy = policies;
 	while(policy != NULL)
 	{
-		struct landlock_path_beneath_attr path_beneath;
+		struct landlock_path_beneath_attr path_beneath = {0};
 		path_beneath.parent_fd = open(policy->path, O_PATH | O_CLOEXEC);
 		if(path_beneath.parent_fd < 0)
 		{
@@ -1313,6 +1372,13 @@ static int landlock_prepare_ruleset(struct exile_path_policy *policies)
 			return ret;
 		}
 		path_beneath.allowed_access = exile_flags_to_landlock(policy->policy, sb.st_mode);
+
+		/* Required, so the .allowed_access fits .handled_access_fs of the ruleset.
+		 * Needed for backwards compatibility, e. g. new binary compiled with new headers,
+		 executed on a kernel with an older ABI version which does not have some constant defined...
+		 */
+		path_beneath.allowed_access &= ruleset_attr.handled_access_fs;
+
 		ret = landlock_add_rule(ruleset_fd, LANDLOCK_RULE_PATH_BENEATH, &path_beneath, 0);
 		if(ret)
 		{
@@ -1347,7 +1413,7 @@ static int check_policy_sanity(struct exile_policy *policy)
 		{
 			if(path_policy_needs_landlock(path_policy))
 			{
-				EXILE_LOG_ERROR("A path policy needs landlock, but landlock is not available. Fallback not possible\n");
+				EXILE_LOG_ERROR("A path policy (%s) needs landlock, but landlock is not available. Fallback not possible\n", path_policy->path);
 				return -1;
 			}
 			path_policy = path_policy->next;
@@ -1400,6 +1466,11 @@ static int check_policy_sanity(struct exile_policy *policy)
 		{
 			if(syscall_policy->syscall == EXILE_SYSCALL_MATCH_ALL)
 			{
+				if(policy->vow_promises != 0)
+				{
+					EXILE_LOG_ERROR("It's not possible to specify a default, all matching syscall policy while also using vows\n");
+					return -1;
+				}
 				last_match_all = i;
 				match_all_policy = syscall_policy->policy;
 			}
@@ -1410,7 +1481,7 @@ static int check_policy_sanity(struct exile_policy *policy)
 			syscall_policy = syscall_policy->next;
 			++i;
 		}
-		if(last_match_all == -1 || i - last_match_all != 1)
+		if(policy->vow_promises == 0 && (last_match_all == -1 || i - last_match_all != 1))
 		{
 			EXILE_LOG_ERROR("The last entry in the syscall policy list must match all syscalls (default rule)\n");
 			return -1;
@@ -1431,7 +1502,20 @@ static void close_file_fds()
 	long max_files = sysconf(_SC_OPEN_MAX);
 	for(long i = 3; i <= max_files; i++)
 	{
-		close((int)i);
+		struct stat statbuf;
+		int fd = (int) max_files;
+		int result = fstat(i, &statbuf);
+		if(result == -1 && errno != EBADF && errno != EACCES)
+		{
+			EXILE_LOG_ERROR("Could not fstat %i: %s\n", fd, strerror(errno));
+			abort();
+		}
+		int type = statbuf.st_mode & S_IFMT;
+		if(type != S_IFIFO && type != S_IFSOCK)
+		{
+			/* No error check, retrying not recommended */
+			close(fd);
+		}
 	}
 }
 
@@ -1448,6 +1532,30 @@ static int enable_no_fs(struct exile_policy *policy)
 {
 		close_file_fds();
 
+		if(exile_landlock_is_available())
+		{
+			struct landlock_ruleset_attr ruleset_attr = {0};
+			if(landlock_set_max_handled_access(&ruleset_attr) != 0)
+			{
+				return -1;
+			}
+			int ruleset_fd = landlock_create_ruleset(&ruleset_attr, sizeof(ruleset_attr), 0);
+			if (ruleset_fd < 0)
+			{
+				EXILE_LOG_ERROR("Failed to create landlock ruleset\n");
+				return -1;
+			}
+			int ret = landlock_restrict_self(ruleset_fd, 0);
+			if(ret != 0)
+			{
+				EXILE_LOG_ERROR("Failed to enable no_fs with landlock: %s\n", strerror(errno));
+				close(ruleset_fd);
+				return -1;
+			}
+			close(ret);
+			return 0;
+		}
+
 		if(chdir("/proc/self/fdinfo") != 0)
 		{
 			EXILE_LOG_ERROR("Failed to change to safe directory: %s\n", strerror(errno));
@@ -1494,7 +1602,12 @@ int exile_enable_policy(struct exile_policy *policy)
 		return -EINVAL;
 	}
 
-	if(enter_namespaces(policy->namespace_options) < 0)
+	if(policy->keep_fds_open != 1)
+	{
+		close_file_fds();
+	}
+
+	if(enter_namespaces(policy->namespace_options, policy->namespace_uid, policy->namespace_gid) < 0)
 	{
 		EXILE_LOG_ERROR("Error while trying to enter namespaces\n");
 		return -1;
@@ -1577,14 +1690,6 @@ int exile_enable_policy(struct exile_policy *policy)
 	}
 #endif
 
-	if(policy->no_fs)
-	{
-		if(enable_no_fs(policy) != 0)
-		{
-			EXILE_LOG_ERROR("Failed to take away filesystem access of process\n");
-			return -1;
-		}
-	}
 
 	if(policy->no_new_fds)
 	{
@@ -1596,15 +1701,6 @@ int exile_enable_policy(struct exile_policy *policy)
 		}
 	}
 
-	if(policy->drop_caps)
-	{
-		if(drop_caps() < 0)
-		{
-			EXILE_LOG_ERROR("failed to drop capabilities\n");
-			return -1;
-		}
-	}
-
 	if(policy->not_dumpable)
 	{
 		if(prctl(PR_SET_DUMPABLE, 0) == -1)
@@ -1623,6 +1719,15 @@ int exile_enable_policy(struct exile_policy *policy)
 		}
 	}
 
+	if(policy->no_fs)
+	{
+		if(enable_no_fs(policy) != 0)
+		{
+			EXILE_LOG_ERROR("Failed to take away filesystem access of process\n");
+			return -1;
+		}
+	}
+
 #if HAVE_LANDLOCK == 1
 	if (can_use_landlock && policy->path_policies != NULL && landlock_restrict_self(landlock_ruleset_fd, 0) != 0)
 	{
@@ -1643,12 +1748,19 @@ int exile_enable_policy(struct exile_policy *policy)
 		}
 	}
 
+	if(policy->drop_caps)
+	{
+		if(drop_caps() < 0)
+		{
+			EXILE_LOG_ERROR("failed to drop capabilities\n");
+			return -1;
+		}
+	}
+
 	if(policy->syscall_policies != NULL)
 	{
 		return exile_enable_syscall_policy(policy);
 	}
-
-
 	return 0;
 }
 
@@ -1841,13 +1953,6 @@ char *exile_launch_get(struct exile_launch_params *launch_params, size_t *n)
 		}
 	}
 	fclose(stream);
-	int seek = fseek(stream, 0, SEEK_SET);
-	if(seek == -1)
-	{
-		EXILE_LOG_ERROR("fseek failed\n");
-		close(launch_result.read_fd);
-		return NULL;
-	}
 	close(launch_result.read_fd);
 	*n = size;
 	return result;

exile.h

@@ -75,6 +75,7 @@
 #define EXILE_UNSHARE_NETWORK 1<<1
 #define EXILE_UNSHARE_USER 1<<2
 #define EXILE_UNSHARE_MOUNT 1<<3
+#define EXILE_UNSHARE_AUTOMATIC 1<<4
 
 #ifndef EXILE_LOG_ERROR
 #define EXILE_LOG_ERROR(...) do { fprintf(stderr, "exile.h: %s(): Error: ", __func__); fprintf(stderr, __VA_ARGS__); } while(0)
@@ -273,6 +274,7 @@ struct exile_path_policy
 #define EXILE_SYSCALL_ALLOW 1
 #define EXILE_SYSCALL_DENY_KILL_PROCESS 2
 #define EXILE_SYSCALL_DENY_RET_ERROR 3
+#define EXILE_SYSCALL_DENY_RET_NOSYS 4
 
 #define EXILE_BPF_NOP \
 BPF_STMT(BPF_JMP+BPF_JA,0)
@@ -362,6 +364,7 @@ struct exile_policy
 	int no_new_privs;
 	int no_fs;
 	int no_new_fds;
+	int keep_fds_open;
 	int namespace_options;
 	int disable_syscall_filter;
 	/* Bind mounts all paths in path_policies into the chroot and applies
@@ -372,6 +375,9 @@ struct exile_policy
 
 	uint64_t vow_promises;
 
+	uid_t namespace_uid;
+	gid_t namespace_gid;
+
 	/* Do not manually add policies here, use exile_append_path_policies() */
 	struct exile_path_policy *path_policies;
 	struct exile_path_policy **path_policies_tail;

exile.hpp

@@ -100,6 +100,7 @@ inline int do_clone(int (*clonefn)(void *), void *launcharg)
 	}
 	size_t size = rlimit.rlim_cur;
 	char *stack = (char *) calloc(1, size);
+	char *stackbegin = stack;
 	if(stack == NULL)
 	{
 		EXILE_LOG_ERROR("Failed to allocate stack memory for child\n");
@@ -110,6 +111,7 @@ inline int do_clone(int (*clonefn)(void *), void *launcharg)
 	ret = clone(clonefn, stack, 17 /* SIGCHLD */, launcharg);
 	int status = 0;
 	waitpid(ret, &status, __WALL);
+	free(stackbegin);
 	if(WIFEXITED(status))
 	{
 		return WEXITSTATUS(status);
@@ -119,7 +121,7 @@ inline int do_clone(int (*clonefn)(void *), void *launcharg)
 }
 
 template<typename T, typename U, typename ... Args>
-typename std::enable_if_t<std::is_trivially_copyable_v<T>, T> exile_launch(struct exile_policy *policy, U fn, Args && ... args)
+typename std::enable_if_t<std::is_trivially_copyable_v<T> && !std::is_pointer_v<T>, T> exile_launch(struct exile_policy *policy, U fn, Args && ... args)
 {
 	size_t mapsize = sizeof(T);
 	T * sharedbuf =  (T *) mmap(NULL, mapsize , PROT_READ | PROT_WRITE, MAP_SHARED | MAP_ANON, -1, 0);
@@ -145,7 +147,7 @@ typename std::enable_if_t<std::is_trivially_copyable_v<T>, T> exile_launch(struc
 
 
 template<typename T, typename U, typename ... Args>
-typename std::enable_if_t<!std::is_trivially_copyable_v<T> && std::is_copy_constructible_v<T>, T>
+typename std::enable_if_t<std::is_pointer_v<T> || (!std::is_trivially_copyable_v<T> && std::is_copy_constructible_v<T>), T>
 	exile_launch(struct exile_policy *policy, const std::function<size_t (const T &, char *, size_t)> &serializer, const std::function<T(const char *, size_t)> &deserializer, U fn, Args && ... args)
 {
 	size_t mapsize = EXILE_MMAP_SIZE;

test.c

@@ -618,9 +618,9 @@ int test_launch_get()
 	size_t n = 0;
 	char *content = exile_launch_get(&params, &n);
 	unsigned int len = strlen(LAUNCH_GET_TEST_STR);
-	if(n != strlen(LAUNCH_GET_TEST_STR))
+	if(n != len)
 	{
-		LOG("Lenght does does not match: %lu vs %u\n", n, len);
+		LOG("Lenght does not match: %lu vs %u\n", n, len);
 		return 1;
 	}
 	if(strcmp(content, LAUNCH_GET_TEST_STR) != 0)
@@ -643,6 +643,164 @@ int test_vows_from_str()
 	return 0;
 }
 
+int test_clone3_nosys()
+{
+	struct exile_policy *policy = exile_init_policy();
+	policy->vow_promises = exile_vows_from_str("stdio rpath wpath cpath thread error");
+
+	exile_enable_policy(policy);
+	/* While args are invalid, it should never reach clone3 syscall handler, so it's irrelevant for
+	 our test*/
+	long ret =  syscall(__NR_clone3, NULL, 0);
+
+	if(ret == -1 && errno != ENOSYS)
+	{
+		LOG("clone3() was not allowed but did not return ENOSYS. It returned: %li, errno: %i\n", ret, errno);
+		return 1;
+	}
+	return 0;
+}
+
+int do_test_nsuidmap(const char *path,  const char *firstfield, const char *secondfield, const char *thirdfield)
+{
+	char *line = NULL;
+	size_t n = 0;
+	FILE *fp = fopen(path, "r");
+
+	int ret = getdelim(&line, &n, ' ', fp);
+	while(ret != -1 && strlen(line) == 1 && *line == ' ')
+		ret = getdelim(&line, &n, ' ', fp);
+	if(ret == -1)
+	{
+		LOG("getdelim() failed to read a line from %s\n", path);
+		return 1;
+	}
+	line[ret-1] = '\0';
+	if(strcmp(line, firstfield) != 0)
+	{
+		LOG("Invalid value for first entry in map: Expected: %s, was: %s\n", firstfield, line);
+		return 1;
+	}
+
+	ret = getdelim(&line, &n, ' ', fp);
+	while(ret != -1 && strlen(line) == 1 && *line == ' ')
+		ret = getdelim(&line, &n, ' ', fp);
+	if(ret == -1)
+	{
+		LOG("getdelim() failed to read a line from map\n");
+		return 1;
+	}
+	line[ret-1] = '\0';
+
+	if(strcmp(line, secondfield) != 0)
+	{
+		LOG("Invalid value for second entry in map: Expected: %s, was: %s\n", secondfield, line);
+		return 1;
+	}
+
+
+	ret = getdelim(&line, &n, ' ', fp);
+	while(ret != -1 && strlen(line) == 1 && *line == ' ')
+		ret = getdelim(&line, &n, ' ', fp);
+	if(ret == -1)
+	{
+		LOG("getdelim() failed to read a line from uid_map\n");
+		return 1;
+	}
+	line[ret-1] = '\0';
+	if(strcmp(line, thirdfield) != 0)
+	{
+		LOG("Invalid value for second entry in map: Expected: %s, was: %s\n", thirdfield, line);
+		return 1;
+	}
+
+	fclose(fp);
+	return 0;
+}
+
+int test_unshare_user()
+{
+	char uidstr[64];
+	snprintf(uidstr, sizeof(uidstr), "%u", getuid());
+
+	char gidstr[64];
+	snprintf(gidstr, sizeof(gidstr), "%u", getgid());
+
+	struct exile_policy *policy = exile_init_policy();
+	policy->namespace_options = EXILE_UNSHARE_USER;
+	xexile_enable_policy(policy);
+
+	if(do_test_nsuidmap("/proc/self/uid_map", "0", uidstr, "1") != 0)
+	{
+		LOG("/proc/self/uid_map failed\n");
+		return 1;
+	}
+
+	if(do_test_nsuidmap("/proc/self/gid_map", "0", gidstr, "1") != 0)
+	{
+		LOG("/proc/self/gid_map failed\n");
+		return 1;
+	}
+
+	FILE *fp = fopen("/proc/self/setgroups", "r");
+
+	char buffer[4096] = { 0 };
+	fread(buffer, sizeof(buffer), 1, fp);
+	fclose(fp);
+
+	if(strcmp(buffer, "deny\n") != 0)
+	{
+		LOG("/proc/self/setgroups does not contain 'deny'\n");
+		return 1;
+	}
+
+	return 0;
+}
+
+int test_unshare_user_own_uid()
+{
+	uid_t uid = getuid();
+	gid_t gid = getgid();
+
+	char uidstr[64];
+	snprintf(uidstr, sizeof(uidstr), "%u", uid);
+
+	char gidstr[64];
+	snprintf(gidstr, sizeof(gidstr), "%u", gid);
+
+	struct exile_policy *policy = exile_init_policy();
+	policy->namespace_options = EXILE_UNSHARE_USER;
+	policy->namespace_gid = gid;
+	policy->namespace_uid = uid;
+	xexile_enable_policy(policy);
+
+	if(do_test_nsuidmap("/proc/self/uid_map", uidstr, uidstr, "1") != 0)
+	{
+		LOG("/proc/self/uid_map failed\n");
+		return 1;
+	}
+
+	if(do_test_nsuidmap("/proc/self/gid_map", gidstr, gidstr, "1") != 0)
+	{
+		LOG("/proc/self/gid_map failed\n");
+		return 1;
+	}
+
+	FILE *fp = fopen("/proc/self/setgroups", "r");
+
+	char buffer[4096] = { 0 };
+	fread(buffer, sizeof(buffer), 1, fp);
+	fclose(fp);
+
+	if(strcmp(buffer, "deny\n") != 0)
+	{
+		LOG("/proc/self/setgroups does not contain 'deny'\n");
+		return 1;
+	}
+
+	return 0;
+}
+
 struct dispatcher
 {
 	char *name;
@@ -670,6 +828,11 @@ struct dispatcher dispatchers[] = {
 	{ "launch", &test_launch},
 	{ "launch-get", &test_launch_get},
 	{ "vow_from_str", &test_vows_from_str},
+	{ "clone3_nosys", &test_clone3_nosys},
+	{ "unshare-user", &test_unshare_user},
+	{ "unshare-user-own-uid", &test_unshare_user_own_uid},
+
+
 };
 
 int main(int argc, char *argv[])

test.sh

@@ -8,41 +8,41 @@ COUNT_SUCCEEDED=0
 COUNT_FAILED=0
 COUNT_SKIPPED=0
 
-function print_fail()
+print_fail()
 {
-	echo -e "${RED}$@${NC}" 1>&2
+	printf "${RED}$@${NC}\n" 1>&2
 }
 
-function print_success()
+print_success()
 {
-	echo -e "${GREEN}$@${NC}"
+	printf "${GREEN}$@${NC}\n"
 }
 
-function print_skipped()
+print_skipped()
 {
-	echo -e "${YELLOW}$@${NC}"
+	printf "${YELLOW}$@${NC}\n"
 }
 
-function runtest_fail()
+runtest_fail()
 {
 	print_fail "failed"
 	COUNT_FAILED=$(($COUNT_FAILED+1))
 }
 
-function runtest_success()
+runtest_success()
 {
 	print_success "ok"
 	COUNT_SUCCEEDED=$((COUNT_SUCCEEDED+1))
 }
 
-function runtest_skipped()
+runtest_skipped()
 {
 	print_skipped "skipped"
 	COUNT_SKIPPED=$((COUNT_SKIPPED+1))
 }
 
 
-function runtest()
+runtest()
 {
 	testbin="$1"
 	testname="$2"
@@ -52,7 +52,8 @@ function runtest()
 
 	echo -n "Running $testname... "
 	#exit $? to suppress shell message like "./test.sh: line 18: pid Bad system call"
-	(./$testbin "$testname" || exit $?) &>> "${test_log_file}"
+	(./$testbin "$testname" || exit $?) >> "${test_log_file}" 2>&1
+
 	ret=$?
 	SUCCESS="no"
 	if [ $ret -eq 0 ] ; then