[PATCH v5 1/4] syscalls: Restore address limit after a syscall

[Thomas Garnier]

This patch ensures a syscall does not return to user-mode with a kernel
address limit. If that happened, a process can corrupt kernel-mode
memory and elevate privileges.

For example, it would mitigation this bug:

- https://bugs.chromium.org/p/project-zero/issues/detail?id=990

The CONFIG_ARCH_NO_SYSCALL_VERIFY_PRE_USERMODE_STATE option is also
added so each architecture can optimize this change.

Signed-off-by: Thomas Garnier <thgarnie at google.com>
---
Based on next-20170322
---
 arch/s390/Kconfig        |  1 +
 include/linux/syscalls.h | 26 +++++++++++++++++++++++++-
 init/Kconfig             |  7 +++++++
 kernel/sys.c             |  7 +++++++
 4 files changed, 40 insertions(+), 1 deletion(-)

diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index a2dcef0aacc7..b73f5b87bc99 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -103,6 +103,7 @@ config S390
 	select ARCH_INLINE_WRITE_UNLOCK_BH
 	select ARCH_INLINE_WRITE_UNLOCK_IRQ
 	select ARCH_INLINE_WRITE_UNLOCK_IRQRESTORE
+	select ARCH_NO_SYSCALL_VERIFY_PRE_USERMODE_STATE
 	select ARCH_SAVE_PAGE_KEYS if HIBERNATION
 	select ARCH_SUPPORTS_ATOMIC_RMW
 	select ARCH_SUPPORTS_NUMA_BALANCING
diff --git a/include/linux/syscalls.h b/include/linux/syscalls.h
index 980c3c9b06f8..f9ff80fa92ff 100644
--- a/include/linux/syscalls.h
+++ b/include/linux/syscalls.h
@@ -191,6 +191,27 @@ extern struct trace_event_functions exit_syscall_print_funcs;
 	SYSCALL_METADATA(sname, x, __VA_ARGS__)			\
 	__SYSCALL_DEFINEx(x, sname, __VA_ARGS__)
 
+
+/*
+ * Called before coming back to user-mode. Returning to user-mode with an
+ * address limit different than USER_DS can allow to overwrite kernel memory.
+ */
+static inline void verify_pre_usermode_state(void) {
+	BUG_ON(!segment_eq(get_fs(), USER_DS));
+}
+
+#ifndef CONFIG_ARCH_NO_SYSCALL_VERIFY_PRE_USERMODE_STATE
+#define __CHECK_USER_CALLER() \
+	bool user_caller = segment_eq(get_fs(), USER_DS)
+#define __VERIFY_PRE_USERMODE_STATE() \
+	if (user_caller) verify_pre_usermode_state()
+#else
+#define __CHECK_USER_CALLER()
+#define __VERIFY_PRE_USERMODE_STATE()
+asmlinkage void asm_verify_pre_usermode_state(void);
+#endif
+
+
 #define __PROTECT(...) asmlinkage_protect(__VA_ARGS__)
 #define __SYSCALL_DEFINEx(x, name, ...)					\
 	asmlinkage long sys##name(__MAP(x,__SC_DECL,__VA_ARGS__))	\
@@ -199,7 +220,10 @@ extern struct trace_event_functions exit_syscall_print_funcs;
 	asmlinkage long SyS##name(__MAP(x,__SC_LONG,__VA_ARGS__));	\
 	asmlinkage long SyS##name(__MAP(x,__SC_LONG,__VA_ARGS__))	\
 	{								\
-		long ret = SYSC##name(__MAP(x,__SC_CAST,__VA_ARGS__));	\
+		long ret;						\
+		__CHECK_USER_CALLER();					\
+		ret = SYSC##name(__MAP(x,__SC_CAST,__VA_ARGS__));	\
+		__VERIFY_PRE_USERMODE_STATE();				\
 		__MAP(x,__SC_TEST,__VA_ARGS__);				\
 		__PROTECT(x, ret,__MAP(x,__SC_ARGS,__VA_ARGS__));	\
 		return ret;						\
diff --git a/init/Kconfig b/init/Kconfig
index c859c993c26f..c4efc3a95e4a 100644
--- a/init/Kconfig
+++ b/init/Kconfig
@@ -1929,6 +1929,13 @@ config PROFILING
 config TRACEPOINTS
 	bool
 
+#
+# Set by each architecture that want to optimize how verify_pre_usermode_state
+# is called.
+#
+config ARCH_NO_SYSCALL_VERIFY_PRE_USERMODE_STATE
+	bool
+
 source "arch/Kconfig"
 
 endmenu		# General setup
diff --git a/kernel/sys.c b/kernel/sys.c
index 196c7134bee6..4ae278fcc290 100644
--- a/kernel/sys.c
+++ b/kernel/sys.c
@@ -2459,3 +2459,10 @@ COMPAT_SYSCALL_DEFINE1(sysinfo, struct compat_sysinfo __user *, info)
 	return 0;
 }
 #endif /* CONFIG_COMPAT */
+
+#ifdef CONFIG_ARCH_NO_SYSCALL_VERIFY_PRE_USERMODE_STATE
+asmlinkage void asm_verify_pre_usermode_state(void)
+{
+	verify_pre_usermode_state();
+}
+#endif
-- 
2.12.1.500.gab5fba24ee-goog