Skip to contentSkip to conversation

kernel: Implement selective syscall userspace redirection

PATCH Superseded Gabriel Krisman Bertazi 3 participants 5 messages Jul 9, 2020 – Jul 10, 2020

Subsystems (7)
x86 VFS MM core Scheduler +3 more subsystems

v2 of 4 Superseded Latest is v4 →

lore
Share & tools
lore
#0 lore raw

PATCH v2 1/1 kernel: Implement selective syscall userspace redirection

Introduce a mechanism to quickly disable/enable syscall handling for a specific process and redirect to userspace via SIGSYS. This is useful for processes with parts that require syscall redirection and parts that don't, but who need to perform this boundary crossing really fast, without paying the cost of a system call to reconfigure syscall handling on each boundary transition. This is particularly important for Windows games running over Wine.

The proposed interface looks like this:

prctl(PR_SET_SYSCALL_USER_DISPATCH, <op>, <dispatcher>, [selector])

Dispatcher is the address of a syscall instruction that is allowed to by-pass the blockage, such that in fast paths you don't need to disable the trap nor check the selector. This is essential to return from SIGSYS to a blocked area without triggering another SIGSYS from the rt_sigreturn.

selector is an optional pointer to an aligned int-sized userspace memory region that has a key switch for the mechanism. This key switch can be PR_SYSCALL_DISPATCH_ENABLE, PR_SYSCALL_DISPATCH_DISABLE.

The feature is meant to be set per-thread and it is disabled on fork/clone/execv.

Internally, this doesn't add overhead to the syscall hot path, and it requires very little per-architecture support. I avoided using seccomp, even though it duplicates some functionality, due to previous feedback that maybe it shouldn't mix with seccomp since it is not a security mechanism. And obviously, this should never be considered a security mechanism, since any part of the program can by-pass it by using the syscall dispatcher.

For the sysinfo benchmark, which measures the overhead added to executing a native syscall that doesn't require interception, the overhead using only dispatcher to issue syscalls is pretty much irrelevant. The overhead of selector goes around 40ns for a native (untrapped) syscall in my system, and it is (as expected) dominated by the supervisor-mode user-address access. In fact, with SMAP off, the overhead is consistently less than 5ns on my test box.

Right now, it is only supported by x86_64 and x86, but it should be easily enabled for other architectures.

An example code using this interface can be found at:
https://gitlab.collabora.com/krisman/syscall-disable-personality

Cc Matthew Wilcox <[email protected]>
Cc Andy Lutomirski <[email protected]>
Cc Paul Gofman <[email protected]>
Cc Kees Cook <[email protected]>

---
Changes since RFC:
(Kees Cook suggestions)
- Don't mention personality while explaining the feature
- Use syscall_get_nr
- Remove header guard on several places
- Convert WARN_ON to WARN_ON_ONCE
- Explicit check for state values
- Rename to syscall user dispatcher

RFC/v1: https://lkml.org/lkml/2020/7/8/96

 arch/Kconfig                          | 20 ++++++
 arch/x86/Kconfig                      |  1 +
 arch/x86/entry/common.c               |  5 ++
 arch/x86/include/asm/thread_info.h    |  4 +-
 fs/exec.c                             |  2 +
 include/linux/sched.h                 |  3 +
 include/linux/syscall_user_dispatch.h | 45 ++++++++++++++
 include/uapi/asm-generic/siginfo.h    |  1 +
 include/uapi/linux/prctl.h            |  4 ++
 kernel/Makefile                       |  1 +
 kernel/fork.c                         |  1 +
 kernel/sys.c                          |  8 +++
 kernel/syscall_user_dispatch.c        | 90 +++++++++++++++++++++++++++
 13 files changed, 184 insertions(+), 1 deletion(-)
 create mode 100644 include/linux/syscall_user_dispatch.h
 create mode 100644 kernel/syscall_user_dispatch.c
arch/Kconfig +20 -0
index 8cc35dc556c7..c182344374a1 100644
@@ -465,6 +465,26 @@ config SECCOMP_FILTER
 
 	  See Documentation/userspace-api/seccomp_filter.rst for details.
 
+config HAVE_ARCH_SYSCALL_USER_DISPATCH
+	bool
+	help
+	  An arch should select this symbol if it provides all of these things:
+	  - TIF_SYSCALL_USER_DISPATCH
+	  - syscall_get_arch
+	  - syscall_rollback
+	  - syscall_get_nr
+	  - SIGSYS siginfo_t support
+
+config SYSCALL_USER_DISPATCH
+	bool "Support rejecting syscalls not coming from a dispatcher"
+	depends on HAVE_ARCH_SYSCALL_USER_DISPATCH
+	help
+	  Enable tasks to ask the kernel to redirect syscalls not
+	  issued from a predefined dispatcher back to userspace,
+	  depending on a userspace selector.
+
+	  This option is useful to optimize games running over Wine.
+
 config HAVE_ARCH_STACKLEAK
 	bool
 	help
arch/x86/Kconfig +1 -0
index 883da0abf779..466a3a9c0708 100644
@@ -149,6 +149,7 @@ config X86
 	select HAVE_ARCH_COMPAT_MMAP_BASES	if MMU && COMPAT
 	select HAVE_ARCH_PREL32_RELOCATIONS
 	select HAVE_ARCH_SECCOMP_FILTER
+	select HAVE_ARCH_SYSCALL_USER_DISPATCH
 	select HAVE_ARCH_THREAD_STRUCT_WHITELIST
 	select HAVE_ARCH_STACKLEAK
 	select HAVE_ARCH_TRACEHOOK
arch/x86/entry/common.c +5 -0
index bd3f14175193..6c1360a7f260 100644
@@ -138,6 +138,11 @@ static long syscall_trace_enter(struct pt_regs *regs)
 			return -1L;
 	}
 
+	if (work & _TIF_SYSCALL_USER_DISPATCH) {
+		if (do_syscall_user_dispatch(regs))
+			return -1L;
+	}
+
 #ifdef CONFIG_SECCOMP
 	/*
 	 * Do seccomp after ptrace, to catch any tracer changes.
arch/x86/include/asm/thread_info.h +3 -1
index 8de8ceccb8bc..b26a9f2f0491 100644
@@ -93,6 +93,7 @@ struct thread_info {
 #define TIF_NOTSC		16	/* TSC is not accessible in userland */
 #define TIF_IA32		17	/* IA32 compatibility process */
 #define TIF_SLD			18	/* Restore split lock detection on context switch */
+#define TIF_SYSCALL_USER_DISPATCH 19	/* Redirect syscall for userspace handling */
 #define TIF_MEMDIE		20	/* is terminating due to OOM killer */
 #define TIF_POLLING_NRFLAG	21	/* idle is polling for TIF_NEED_RESCHED */
 #define TIF_IO_BITMAP		22	/* uses I/O bitmap */
@@ -123,6 +124,7 @@ struct thread_info {
 #define _TIF_NOTSC		(1 << TIF_NOTSC)
 #define _TIF_IA32		(1 << TIF_IA32)
 #define _TIF_SLD		(1 << TIF_SLD)
+#define _TIF_SYSCALL_USER_DISPATCH (1 << TIF_SYSCALL_USER_DISPATCH)
 #define _TIF_POLLING_NRFLAG	(1 << TIF_POLLING_NRFLAG)
 #define _TIF_IO_BITMAP		(1 << TIF_IO_BITMAP)
 #define _TIF_FORCED_TF		(1 << TIF_FORCED_TF)
@@ -136,7 +138,7 @@ struct thread_info {
 /* Work to do before invoking the actual syscall. */
 #define _TIF_WORK_SYSCALL_ENTRY	\
 	(_TIF_SYSCALL_TRACE | _TIF_SYSCALL_EMU | _TIF_SYSCALL_AUDIT |	\
-	 _TIF_SECCOMP | _TIF_SYSCALL_TRACEPOINT)
+	 _TIF_SECCOMP | _TIF_SYSCALL_TRACEPOINT | _TIF_SYSCALL_USER_DISPATCH)
 
 /* flags to check in __switch_to() */
 #define _TIF_WORK_CTXSW_BASE					\
fs/exec.c +2 -0
index e6e8a9a70327..44f0ce352a0d 100644
@@ -1502,6 +1502,8 @@ void setup_new_exec(struct linux_binprm * bprm)
 
 	arch_setup_new_exec();
 
+	clear_tsk_syscall_user_dispatch(me);
+
 	/* Set the new mm task size. We have to do that late because it may
 	 * depend on TIF_32BIT which is only updated in flush_thread() on
 	 * some architectures like powerpc
include/linux/sched.h +3 -0
index 692e327d7455..407b868146e1 100644
@@ -32,6 +32,7 @@
 #include <linux/posix-timers.h>
 #include <linux/rseq.h>
 #include <linux/kcsan.h>
+#include <linux/syscall_user_dispatch.h>
 
 /* task_struct member predeclarations (sorted alphabetically): */
 struct audit_context;
@@ -953,6 +954,8 @@ struct task_struct {
 #endif
 	struct seccomp			seccomp;
 
+	struct syscall_user_dispatch	syscall_dispatch;
+
 	/* Thread group tracking: */
 	u64				parent_exec_id;
 	u64				self_exec_id;
include/linux/syscall_user_dispatch.h new file +45 -0

45 lines not on this page · raw message

include/uapi/asm-generic/siginfo.h +1 -0
index cb3d6c267181..16337270467d 100644
@@ -285,6 +285,7 @@ typedef struct siginfo {
  */
 #define SYS_SECCOMP	1	/* seccomp triggered */
 #define NSIGSYS		1
+#define SYS_USER_REDIRECT 2
 
 /*
  * SIGEMT si_codes
include/uapi/linux/prctl.h +4 -0
index 07b4f8131e36..335ea1b5785e 100644
@@ -238,4 +238,8 @@ struct prctl_mm_map {
 #define PR_SET_IO_FLUSHER		57
 #define PR_GET_IO_FLUSHER		58
 
+#define PR_SET_SYSCALL_USER_DISPATCH	59
+# define PR_SYSCALL_DISPATCH_DISABLE	0
+# define PR_SYSCALL_DISPATCH_ENABLE	1
+
 #endif /* _LINUX_PRCTL_H */
kernel/Makefile +1 -0
index f3218bc5ec69..158b8c61592f 100644
@@ -92,6 +92,7 @@ obj-$(CONFIG_DETECT_HUNG_TASK) += hung_task.o
 obj-$(CONFIG_LOCKUP_DETECTOR) += watchdog.o
 obj-$(CONFIG_HARDLOCKUP_DETECTOR_PERF) += watchdog_hld.o
 obj-$(CONFIG_SECCOMP) += seccomp.o
+obj-$(CONFIG_SYSCALL_USER_DISPATCH) += syscall_user_dispatch.o
 obj-$(CONFIG_RELAY) += relay.o
 obj-$(CONFIG_SYSCTL) += utsname_sysctl.o
 obj-$(CONFIG_TASK_DELAY_ACCT) += delayacct.o
kernel/fork.c +1 -0
index 142b23645d82..c6b64a849fec 100644
@@ -921,6 +921,7 @@ static struct task_struct *dup_task_struct(struct task_struct *orig, int node)
 	clear_user_return_notifier(tsk);
 	clear_tsk_need_resched(tsk);
 	set_task_stack_end_magic(tsk);
+	clear_tsk_syscall_user_dispatch(tsk);
 
 #ifdef CONFIG_STACKPROTECTOR
 	tsk->stack_canary = get_random_canary();
kernel/sys.c +8 -0
index 00a96746e28a..549f4dc9340f 100644
@@ -42,6 +42,7 @@
 #include <linux/syscore_ops.h>
 #include <linux/version.h>
 #include <linux/ctype.h>
+#include <linux/syscall_user_dispatch.h>
 
 #include <linux/compat.h>
 #include <linux/syscalls.h>
@@ -2527,6 +2528,13 @@ SYSCALL_DEFINE5(prctl, int, option, unsigned long, arg2, unsigned long, arg3,
 
 		error = (current->flags & PR_IO_FLUSHER) == PR_IO_FLUSHER;
 		break;
+	case PR_SET_SYSCALL_USER_DISPATCH:
+		if (arg5 != 0)
+			return -EINVAL;
+		error = set_syscall_user_dispatch((int) arg2,
+						  (unsigned long __user) arg3,
+						  (int __user *) arg4);
+		break;
 	default:
 		error = -EINVAL;
 		break;
kernel/syscall_user_dispatch.c new file +90 -0

90 lines not on this page · raw message

Signature · 1 line
2.27.0
#1 lore raw
Quoted 9 lines · Gabriel Krisman Bertazi

The proposed interface looks like this:

prctl(PR_SET_SYSCALL_USER_DISPATCH, <op>, <dispatcher>, [selector])

Dispatcher is the address of a syscall instruction that is allowed to by-pass the blockage, such that in fast paths you don't need to disable the trap nor check the selector. This is essential to return from SIGSYS to a blocked area without triggering another SIGSYS from the rt_sigreturn.

Should <dispatcher> be a single pointer or should the interface specify a range from which syscalls may be made without being redirected? eg, one could specify the whole of libc.

prctl(PR_SET_SYSCALL_USER_DISPATCH, <op>, <start>, <inclusive-end>, [selector])

include/linux/syscall_user_dispatch.h line 11

Quoted 13 lines

+++ b/include/linux/syscall_user_dispatch.h

include/linux/syscall_user_dispatch.h new file +11 -0
@@ -0,0 +1,45 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#ifndef _SYSCALL_USER_DISPATCH_H
+#define _SYSCALL_USER_DISPATCH_H
+
+struct task_struct;
+static void clear_tsk_thread_flag(struct task_struct *tsk, int flag);
+
+#ifdef CONFIG_SYSCALL_USER_DISPATCH
+struct syscall_user_dispatch {
+	int __user *selector;
+	unsigned long __user dispatcher;

The __user annotation is on the pointer, not the value. ie, it's

	unsigned long foo;
	unsigned long __user *p;

	get_user(foo, p)

include/uapi/asm-generic/siginfo.h typedef struct siginfo { · line 288

Quoted 6 lines
include/uapi/asm-generic/siginfo.h +2 -0
+++ b/include/uapi/asm-generic/siginfo.h
@@ -285,6 +285,7 @@ typedef struct siginfo {
  */
 #define SYS_SECCOMP	1	/* seccomp triggered */
 #define NSIGSYS		1
+#define SYS_USER_REDIRECT 2

I'd suggest that SYS_USER_REDIRECT should be moved up by one line.

Quoted 8 lines
kernel/syscall_user_dispatch.c new file +8 -0
+int set_syscall_user_dispatch(int mode, unsigned long __user dispatcher,
+			      int __user *selector)
+{
+	switch (mode) {
+	case PR_SYSCALL_DISPATCH_DISABLE:
+		if (dispatcher || selector)
+			return -EINVAL;
+		break;

kernel/syscall_user_dispatch.c line 72

Quoted 10 lines
kernel/syscall_user_dispatch.c new file +10 -0
+	case PR_SYSCALL_DISPATCH_ENABLE:
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	if (selector) {
+		if (!access_ok(selector, sizeof(int)))
+			return -EFAULT;
+	}

You're not enforcing the alignment requirement here.

kernel/syscall_user_dispatch.c line 79

Quoted 6 lines
kernel/syscall_user_dispatch.c new file +6 -0
+	spin_lock_irq(&current->sighand->siglock);
+
+	current->syscall_dispatch.selector = selector;
+	current->syscall_dispatch.dispatcher = dispatcher;
+
+	/* make sure fastlock is committed before setting the flag. */

fastlock? ;-)
I don't think you actually need this. You're setting per-thread state on yourself, so what's the race that you're concerned about?

Quoted 14 lines
kernel/syscall_user_dispatch.c new file +11 -0
+	smp_mb__before_atomic();
+
+	if (mode == PR_SYSCALL_DISPATCH_ENABLE)
+		set_tsk_thread_flag(current, TIF_SYSCALL_USER_DISPATCH);
+	else
+		clear_tsk_thread_flag(current, TIF_SYSCALL_USER_DISPATCH);
+
+	spin_unlock_irq(&current->sighand->siglock);
+
+	return 0;
+}
Signature · 1 line
2.27.0
#2 lore raw
Quoted 11 lines · Matthew Wilcox
Quoted 9 lines · Gabriel Krisman Bertazi

The proposed interface looks like this:

prctl(PR_SET_SYSCALL_USER_DISPATCH, <op>, <dispatcher>, [selector])

Dispatcher is the address of a syscall instruction that is allowed to by-pass the blockage, such that in fast paths you don't need to disable the trap nor check the selector. This is essential to return from SIGSYS to a blocked area without triggering another SIGSYS from the rt_sigreturn.

Quoted 6 lines

Should <dispatcher> be a single pointer or should the interface specify a range from which syscalls may be made without being redirected? eg, one could specify the whole of libc.

prctl(PR_SET_SYSCALL_USER_DISPATCH, <op>, <start>, <inclusive-end>, [selector])

I liked this suggestion a lot, since user can just pass a single address to get the original interface, but it still let us not pay the cost of __get_user on more paths. I will add it to v3.

Quoted 52 lines
Quoted 4 lines

+++ b/include/linux/syscall_user_dispatch.h

include/linux/syscall_user_dispatch.h new file +1 -0
@@ -0,0 +1,45 @@
+/* SPDX-License-Identifier: GPL-2.0 */

include/linux/syscall_user_dispatch.h line 11

Quoted 10 lines
include/linux/syscall_user_dispatch.h new file +10 -0
+#ifndef _SYSCALL_USER_DISPATCH_H
+#define _SYSCALL_USER_DISPATCH_H
+
+struct task_struct;
+static void clear_tsk_thread_flag(struct task_struct *tsk, int flag);
+
+#ifdef CONFIG_SYSCALL_USER_DISPATCH
+struct syscall_user_dispatch {
+	int __user *selector;
+	unsigned long __user dispatcher;

The __user annotation is on the pointer, not the value. ie, it's

	unsigned long foo;
	unsigned long __user *p;

	get_user(foo, p)

include/uapi/asm-generic/siginfo.h typedef struct siginfo { · line 288

Quoted 6 lines
include/uapi/asm-generic/siginfo.h +2 -0
+++ b/include/uapi/asm-generic/siginfo.h
@@ -285,6 +285,7 @@ typedef struct siginfo {
  */
 #define SYS_SECCOMP	1	/* seccomp triggered */
 #define NSIGSYS		1
+#define SYS_USER_REDIRECT 2

I'd suggest that SYS_USER_REDIRECT should be moved up by one line.

Quoted 8 lines
kernel/syscall_user_dispatch.c new file +8 -0
+int set_syscall_user_dispatch(int mode, unsigned long __user dispatcher,
+			      int __user *selector)
+{
+	switch (mode) {
+	case PR_SYSCALL_DISPATCH_DISABLE:
+		if (dispatcher || selector)
+			return -EINVAL;
+		break;

kernel/syscall_user_dispatch.c line 72

Quoted 10 lines
kernel/syscall_user_dispatch.c new file +10 -0
+	case PR_SYSCALL_DISPATCH_ENABLE:
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	if (selector) {
+		if (!access_ok(selector, sizeof(int)))
+			return -EFAULT;
+	}

You're not enforcing the alignment requirement here.

Quoted 8 lines

kernel/syscall_user_dispatch.c line 79

Quoted 6 lines
kernel/syscall_user_dispatch.c new file +6 -0
+	spin_lock_irq(&current->sighand->siglock);
+
+	current->syscall_dispatch.selector = selector;
+	current->syscall_dispatch.dispatcher = dispatcher;
+
+	/* make sure fastlock is committed before setting the flag. */

fastlock? ;-)

Gee, keeping variable renames updated on comments is hard, compiler won't catch those. :)

Quoted 2 lines

I don't think you actually need this. You're setting per-thread state on yourself, so what's the race that you're concerned about?

Good point. I was assuming this would be modifiable from under it, but it is not the case.

Quoted 15 lines
Quoted 15 lines
kernel/syscall_user_dispatch.c new file +11 -0
+	smp_mb__before_atomic();
+
+	if (mode == PR_SYSCALL_DISPATCH_ENABLE)
+		set_tsk_thread_flag(current, TIF_SYSCALL_USER_DISPATCH);
+	else
+		clear_tsk_thread_flag(current, TIF_SYSCALL_USER_DISPATCH);
+
+	spin_unlock_irq(&current->sighand->siglock);
+
+	return 0;
+}
Signature · 1 line
2.27.0
Signature · 1 line
Gabriel Krisman Bertazi
#3 lore raw
In reply to Gabriel Krisman Bertazi

arch/Kconfig config SECCOMP_FILTER · line 479

Quoted 3 lines · Gabriel Krisman Bertazi

[...]

arch/Kconfig +2 -0
+config SYSCALL_USER_DISPATCH
+	bool "Support rejecting syscalls not coming from a dispatcher"

bike shed: this doesn't really describe why it's useful. Maybe:

bool "Support syscall redirection to userspace dispatcher"

arch/Kconfig config SECCOMP_FILTER · line 484

Quoted 5 lines
arch/Kconfig +5 -0
+	depends on HAVE_ARCH_SYSCALL_USER_DISPATCH
+	help
+	  Enable tasks to ask the kernel to redirect syscalls not
+	  issued from a predefined dispatcher back to userspace,
+	  depending on a userspace selector.

depending on a userspace memory selector.

?

arch/x86/entry/common.c static long syscall_trace_enter(struct pt_regs *regs) · line 145

Quoted 12 lines

[...]

arch/x86/entry/common.c +5 -0
@@ -138,6 +138,11 @@ static long syscall_trace_enter(struct pt_regs *regs)
 			return -1L;
 	}
 
+	if (work & _TIF_SYSCALL_USER_DISPATCH) {
+		if (do_syscall_user_dispatch(regs))
+			return -1L;
+	}
+

Nice; I like this! It's very small, and now I want to refactor seccomp to be so pretty. :)

fs/exec.c void setup_new_exec(struct linux_binprm * bprm) · line 1506

Quoted 11 lines

[...]

fs/exec.c +2 -0
index e6e8a9a70327..44f0ce352a0d 100644
@@ -1502,6 +1502,8 @@ void setup_new_exec(struct linux_binprm * bprm)
 
 	arch_setup_new_exec();
 
+	clear_tsk_syscall_user_dispatch(me);
+

To keep this as arch-agnostic as possible, I actually think this belongs in begin_new_exec() instead, after the personality clearing. If it were to be less arch-agnostic, I'd recommend it live in arch_setup_new_exec() (which most other TIF flags get cleared), but I'd like to have this feature as a good example of an "easy" refactor into arch-agnostic in the future. :P

fs/exec.c typedef struct siginfo { · line 288

Quoted 6 lines

[...]

fs/exec.c +1 -0
@@ -285,6 +285,7 @@ typedef struct siginfo {
  */
 #define SYS_SECCOMP	1	/* seccomp triggered */
 #define NSIGSYS		1
+#define SYS_USER_REDIRECT 2

More than just what willy suggested, but you also need to bump NSIGSYS (I missed that in the RFC). This should read as:

#define SYS_SECCOMP 1 /* seccomp triggered */
#define SYS_USER_REDIRECT 2
#define NSIGSYS 2

(i.e. NSIGSYS is "how many si codes are there for sigsys?")

fs/exec.c

Quoted 7 lines

[...]

fs/exec.c +6 -0
+int do_syscall_user_dispatch(struct pt_regs *regs)
+{
+	int state;
+
+	if (current->syscall_dispatch.dispatcher == instruction_pointer(regs))
+		return 0;

Just to clarify what willy was talking about -- since you're using "dispatcher" as a scalar (and not dereferencing it, etc), it can just stay "unsigned long" without __user.

In the documentation for the future "range inclusive" check, maybe also note that it's the inclusive ranger covering the _starting_ address of the syscall entry, in case anyone thinks they need to decode instruction lengths to get the right range, which they don't and I imagine you don't care about.

Quoted 9 lines
fs/exec.c +9 -0
+
+	if (current->syscall_dispatch.selector) {
+		if (__get_user(state, current->syscall_dispatch.selector))
+			do_exit(SIGSEGV);
+
+		switch (state) {
+		case PR_SYSCALL_DISPATCH_DISABLE:
+			return 0;
+		case PR_SYSCALL_DISPATCH_ENABLE:

fs/exec.c

Quoted 10 lines
fs/exec.c +10 -0
+			break;
+		default:
+			do_exit(SIGSEGV);
+		}
+	}
+
+	syscall_rollback(current, regs);
+	trigger_sigsys(regs);
+
+	return 1;

How should do_syscall_user_dispatch() approach branch hinting? For example, is dispatcher == instruction_pointer going to be the common case for users of this? I would expect not. So maybe, at the top:

	if (unlikely(current->syscall_dispatch.dispatcher == instruction_pointer(regs)))
		return 0;

and what about the selector? I assume it would be common for the selector to be set, and enabled:

	if (likely(current->syscall_dispatch.selector)) {
		if (__get_user(state, current->syscall_dispatch.selector))
			do_exit(SIGSEGV);

		if (unlikely(state != PR_SYSCALL_DISPATCH_ENABLE)) {
			if (likely(state == PR_SYSCALL_DISPATCH_DISABLE))
				return 0;
			do_exit(SIGSEGV);
		}
	}

	syscall_rollback...

Or, perhaps micro-optimization doesn't matter at all here give the overhead of signal delivery, and you can just ignore me and leave this as-is, which is fine too. :)

This continues to look really nice. Very simple, very powerful. I think I'd like to see one more thing: a selftest! I think it should be very easy to add; model it after the seccomp selftests:

tools/testing/selftests/seccomp/seccomp_bpf.c

The testing harness there should make it very easy to produce a test, and it's easy to wire up to the Makefiles. I'm happy to help point you in the right directions. If you want, you could even share seccomp's directory, but maybe you want your own. :)

Signature · 1 line
Kees Cook
#4 lore raw
Quoted 64 lines · Kees Cook

arch/Kconfig config SECCOMP_FILTER · line 479

Quoted 3 lines · Gabriel Krisman Bertazi

[...]

arch/Kconfig +2 -0
+config SYSCALL_USER_DISPATCH
+	bool "Support rejecting syscalls not coming from a dispatcher"

bike shed: this doesn't really describe why it's useful. Maybe:

bool "Support syscall redirection to userspace dispatcher"

arch/Kconfig config SECCOMP_FILTER · line 484

Quoted 5 lines
arch/Kconfig +5 -0
+	depends on HAVE_ARCH_SYSCALL_USER_DISPATCH
+	help
+	  Enable tasks to ask the kernel to redirect syscalls not
+	  issued from a predefined dispatcher back to userspace,
+	  depending on a userspace selector.

depending on a userspace memory selector.

?

arch/x86/entry/common.c static long syscall_trace_enter(struct pt_regs *regs) · line 145

Quoted 12 lines

[...]

arch/x86/entry/common.c +5 -0
@@ -138,6 +138,11 @@ static long syscall_trace_enter(struct pt_regs *regs)
 			return -1L;
 	}
 
+	if (work & _TIF_SYSCALL_USER_DISPATCH) {
+		if (do_syscall_user_dispatch(regs))
+			return -1L;
+	}
+

Nice; I like this! It's very small, and now I want to refactor seccomp to be so pretty. :)

fs/exec.c void setup_new_exec(struct linux_binprm * bprm) · line 1506

Quoted 11 lines

[...]

fs/exec.c +2 -0
index e6e8a9a70327..44f0ce352a0d 100644
@@ -1502,6 +1502,8 @@ void setup_new_exec(struct linux_binprm * bprm)
 
 	arch_setup_new_exec();
 
+	clear_tsk_syscall_user_dispatch(me);
+

To keep this as arch-agnostic as possible, I actually think this belongs in begin_new_exec() instead, after the personality clearing. If it were to be less arch-agnostic, I'd recommend it live in arch_setup_new_exec() (which most other TIF flags get cleared), but I'd like to have this feature as a good example of an "easy" refactor into arch-agnostic in the future. :P

fs/exec.c typedef struct siginfo { · line 288

Quoted 6 lines

[...]

fs/exec.c +1 -0
@@ -285,6 +285,7 @@ typedef struct siginfo {
  */
 #define SYS_SECCOMP	1	/* seccomp triggered */
 #define NSIGSYS		1
+#define SYS_USER_REDIRECT 2

More than just what willy suggested, but you also need to bump NSIGSYS (I missed that in the RFC). This should read as:

Quoted 5 lines

#define SYS_SECCOMP 1 /* seccomp triggered */
#define SYS_USER_REDIRECT 2
#define NSIGSYS 2

(i.e. NSIGSYS is "how many si codes are there for sigsys?")

Hey Kees,

Thanks for the very detailed feedback.

I was actually asking myself what NSIGSYS was and, if it meant something else, whether it was a bug to have the same value as SYS_SECCOMP. The name isn't very meaningful to me :). I'll slip in a comment here in my next version.

Quoted 11 lines

fs/exec.c

Quoted 7 lines

[...]

fs/exec.c +6 -0
+int do_syscall_user_dispatch(struct pt_regs *regs)
+{
+	int state;
+
+	if (current->syscall_dispatch.dispatcher == instruction_pointer(regs))
+		return 0;

Just to clarify what willy was talking about -- since you're using "dispatcher" as a scalar (and not dereferencing it, etc), it can just stay "unsigned long" without __user.

Thanks, i think I got the __user flag correctly now.

I was actually wondering more if this should be (void*), (uintptr) or some other type, and if ulong was the right call, but looking at other code in the kernel ulong seems fine.

Quoted 5 lines

In the documentation for the future "range inclusive" check, maybe also note that it's the inclusive ranger covering the _starting_ address of the syscall entry, in case anyone thinks they need to decode instruction lengths to get the right range, which they don't and I imagine you don't care about.

This is something I was wondering. For the end of the range, shouldn't we document it must include exactly the __end__ of the syscall instruction, i.e., the start of the next instruction, instead of the start? If syscall is in the last address of the range, we get IP for whatever is next and in that case, a syscall issued from a "valid" address is rejected. This is a corner case, for sure, but for simplicity I think we could just let the kernel reject the syscall, otherwise we need to find the actual syscall address when checking the range. Is that fine?

Quoted 31 lines
Quoted 10 lines
fs/exec.c +9 -0
+
+	if (current->syscall_dispatch.selector) {
+		if (__get_user(state, current->syscall_dispatch.selector))
+			do_exit(SIGSEGV);
+
+		switch (state) {
+		case PR_SYSCALL_DISPATCH_DISABLE:
+			return 0;
+		case PR_SYSCALL_DISPATCH_ENABLE:

fs/exec.c

Quoted 10 lines
fs/exec.c +10 -0
+			break;
+		default:
+			do_exit(SIGSEGV);
+		}
+	}
+
+	syscall_rollback(current, regs);
+	trigger_sigsys(regs);
+
+	return 1;

How should do_syscall_user_dispatch() approach branch hinting? For example, is dispatcher == instruction_pointer going to be the common case for users of this? I would expect not. So maybe, at the top:

	if (unlikely(current->syscall_dispatch.dispatcher == instruction_pointer(regs)))
		return 0;

and what about the selector? I assume it would be common for the selector to be set, and enabled:

Quoted 12 lines
	if (likely(current->syscall_dispatch.selector)) {
		if (__get_user(state, current->syscall_dispatch.selector))
			do_exit(SIGSEGV);

		if (unlikely(state != PR_SYSCALL_DISPATCH_ENABLE)) {
			if (likely(state == PR_SYSCALL_DISPATCH_DISABLE))
				return 0;
			do_exit(SIGSEGV);
		}
	}

	syscall_rollback...

My first version had all the unlikely/likely added, but I didn't observe high branch misses when profiling so I dropped before sending the RFC. The fast path, considering the new range scheme (specially when you allow the glibc range), is entering the first 'if' condition, so that should be made likely. Once we are reading selector then __get_user has such a high cost with SMAP that I didn't notice any difference in the wall-clock time.

Also, I think placing the DISABLE check before ENABLE might be better, since we won't pay the high cost of SIGSYS in that case. Once we know that we will either SIGSEGV or SIGSYS, a branch miss won't do much difference.. Does the following sound good?

if (likely(current->syscall_dispatch.dispatcher == instruction_pointer(regs)))

	return 0;
if (likely(current->syscall_dispatch.selector)) {
	if (__get_user(state, current->syscall_dispatch.selector))
		do_exit(SIGSEGV);

	if (likely(state == PR_SYSCALL_DISPATCH_DISABLE)
		return 0;

	if (state != PR_SYSCALL_DISPATCH_ENABLE)
		do_exit(SIGSEGV);
}

// syscall_rollback

Quoted 9 lines

Or, perhaps micro-optimization doesn't matter at all here give the overhead of signal delivery, and you can just ignore me and leave this as-is, which is fine too. :)

This continues to look really nice. Very simple, very powerful. I think I'd like to see one more thing: a selftest! I think it should be very easy to add; model it after the seccomp selftests:

Quoted 6 lines

tools/testing/selftests/seccomp/seccomp_bpf.c

The testing harness there should make it very easy to produce a test, and it's easy to wire up to the Makefiles. I'm happy to help point you in the right directions. If you want, you could even share seccomp's directory, but maybe you want your own. :)

Definitely. Added to my todo list for the next iteration.

Thanks!

Signature · 1 line
Gabriel Krisman Bertazi