context_tracking: Restore correct previous context state on exception exit

On exception exit, we restore the previous context tracking state based on
the regs of the interrupted frame. Iff that frame is in user mode as
stated by user_mode() helper, we restore the context tracking user mode.

However there is a tiny chunck of low level arch code after we pass through
user_enter() and until the CPU eventually resumes userspace.
If an exception happens in this tiny area, exception_enter() correctly
exits the context tracking user mode but exception_exit() won't restore
it because of the value returned by user_mode(regs).

As a result we may return to userspace with the wrong context tracking
state.

To fix this, change exception_enter() to return the context tracking state
prior to its call and pass this saved state to exception_exit(). This restores
the real context tracking state of the interrupted frame.

(May be this patch was suggested to me, I don't recall exactly. If so,
sorry for the missing credit).
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Li Zhong <zhong@linux.vnet.ibm.com>
Cc: Kevin Hilman <khilman@linaro.org>
Cc: Mats Liljegren <mats.liljegren@enea.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Namhyung Kim <namhyung.kim@lge.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>

arch/x86/kernel/kvm.c

@@ -254,16 +254,18 @@ EXPORT_SYMBOL_GPL(kvm_read_and_reset_pf_reason);
 dotraplinkage void __kprobes
 do_async_page_fault(struct pt_regs *regs, unsigned long error_code)
 {
+	enum ctx_state prev_state;
+
 	switch (kvm_read_and_reset_pf_reason()) {
 	default:
 		do_page_fault(regs, error_code);
 		break;
 	case KVM_PV_REASON_PAGE_NOT_PRESENT:
 		/* page is swapped out by the host. */
-		exception_enter(regs);
+		prev_state = exception_enter();
 		exit_idle();
 		kvm_async_pf_task_wait((u32)read_cr2());
-		exception_exit(regs);
+		exception_exit(prev_state);
 		break;
 	case KVM_PV_REASON_PAGE_READY:
 		rcu_irq_enter();

arch/x86/kernel/traps.c

@@ -175,34 +175,38 @@ do_trap(int trapnr, int signr, char *str, struct pt_regs *regs,
 #define DO_ERROR(trapnr, signr, str, name)				\
 dotraplinkage void do_##name(struct pt_regs *regs, long error_code)	\
 {									\
-	exception_enter(regs);						\
+	enum ctx_state prev_state;					\
+									\
+	prev_state = exception_enter();					\
 	if (notify_die(DIE_TRAP, str, regs, error_code,			\
 			trapnr, signr) == NOTIFY_STOP) {		\
-		exception_exit(regs);					\
+		exception_exit(prev_state);				\
 		return;							\
 	}								\
 	conditional_sti(regs);						\
 	do_trap(trapnr, signr, str, regs, error_code, NULL);		\
-	exception_exit(regs);						\
+	exception_exit(prev_state);					\
 }
 
 #define DO_ERROR_INFO(trapnr, signr, str, name, sicode, siaddr)		\
 dotraplinkage void do_##name(struct pt_regs *regs, long error_code)	\
 {									\
 	siginfo_t info;							\
+	enum ctx_state prev_state;					\
+									\
 	info.si_signo = signr;						\
 	info.si_errno = 0;						\
 	info.si_code = sicode;						\
 	info.si_addr = (void __user *)siaddr;				\
-	exception_enter(regs);						\
+	prev_state = exception_enter();					\
 	if (notify_die(DIE_TRAP, str, regs, error_code,			\
 			trapnr, signr) == NOTIFY_STOP) {		\
-		exception_exit(regs);					\
+		exception_exit(prev_state);				\
 		return;							\
 	}								\
 	conditional_sti(regs);						\
 	do_trap(trapnr, signr, str, regs, error_code, &info);		\
-	exception_exit(regs);						\
+	exception_exit(prev_state);					\
 }
 
 DO_ERROR_INFO(X86_TRAP_DE, SIGFPE, "divide error", divide_error, FPE_INTDIV,
@@ -225,14 +229,16 @@ DO_ERROR_INFO(X86_TRAP_AC, SIGBUS, "alignment check", alignment_check,
 /* Runs on IST stack */
 dotraplinkage void do_stack_segment(struct pt_regs *regs, long error_code)
 {
-	exception_enter(regs);
+	enum ctx_state prev_state;
+
+	prev_state = exception_enter();
 	if (notify_die(DIE_TRAP, "stack segment", regs, error_code,
 		       X86_TRAP_SS, SIGBUS) != NOTIFY_STOP) {
 		preempt_conditional_sti(regs);
 		do_trap(X86_TRAP_SS, SIGBUS, "stack segment", regs, error_code, NULL);
 		preempt_conditional_cli(regs);
 	}
-	exception_exit(regs);
+	exception_exit(prev_state);
 }
 
 dotraplinkage void do_double_fault(struct pt_regs *regs, long error_code)
@@ -240,7 +246,7 @@ dotraplinkage void do_double_fault(struct pt_regs *regs, long error_code)
 	static const char str[] = "double fault";
 	struct task_struct *tsk = current;
 
-	exception_enter(regs);
+	exception_enter();
 	/* Return not checked because double check cannot be ignored */
 	notify_die(DIE_TRAP, str, regs, error_code, X86_TRAP_DF, SIGSEGV);
 
@@ -260,8 +266,9 @@ dotraplinkage void __kprobes
 do_general_protection(struct pt_regs *regs, long error_code)
 {
 	struct task_struct *tsk;
+	enum ctx_state prev_state;
 
-	exception_enter(regs);
+	prev_state = exception_enter();
 	conditional_sti(regs);
 
 #ifdef CONFIG_X86_32
@@ -299,12 +306,14 @@ do_general_protection(struct pt_regs *regs, long error_code)
 
 	force_sig(SIGSEGV, tsk);
 exit:
-	exception_exit(regs);
+	exception_exit(prev_state);
 }
 
 /* May run on IST stack. */
 dotraplinkage void __kprobes notrace do_int3(struct pt_regs *regs, long error_code)
 {
+	enum ctx_state prev_state;
+
 #ifdef CONFIG_DYNAMIC_FTRACE
 	/*
 	 * ftrace must be first, everything else may cause a recursive crash.
@@ -314,7 +323,7 @@ dotraplinkage void __kprobes notrace do_int3(struct pt_regs *regs, long error_co
 	    ftrace_int3_handler(regs))
 		return;
 #endif
-	exception_enter(regs);
+	prev_state = exception_enter();
 #ifdef CONFIG_KGDB_LOW_LEVEL_TRAP
 	if (kgdb_ll_trap(DIE_INT3, "int3", regs, error_code, X86_TRAP_BP,
 				SIGTRAP) == NOTIFY_STOP)
@@ -335,7 +344,7 @@ dotraplinkage void __kprobes notrace do_int3(struct pt_regs *regs, long error_co
 	preempt_conditional_cli(regs);
 	debug_stack_usage_dec();
 exit:
-	exception_exit(regs);
+	exception_exit(prev_state);
 }
 
 #ifdef CONFIG_X86_64
@@ -392,11 +401,12 @@ asmlinkage __kprobes struct pt_regs *sync_regs(struct pt_regs *eregs)
 dotraplinkage void __kprobes do_debug(struct pt_regs *regs, long error_code)
 {
 	struct task_struct *tsk = current;
+	enum ctx_state prev_state;
 	int user_icebp = 0;
 	unsigned long dr6;
 	int si_code;
 
-	exception_enter(regs);
+	prev_state = exception_enter();
 
 	get_debugreg(dr6, 6);
 
@@ -466,7 +476,7 @@ dotraplinkage void __kprobes do_debug(struct pt_regs *regs, long error_code)
 	debug_stack_usage_dec();
 
 exit:
-	exception_exit(regs);
+	exception_exit(prev_state);
 }
 
 /*
@@ -560,17 +570,21 @@ void math_error(struct pt_regs *regs, int error_code, int trapnr)
 
 dotraplinkage void do_coprocessor_error(struct pt_regs *regs, long error_code)
 {
-	exception_enter(regs);
+	enum ctx_state prev_state;
+
+	prev_state = exception_enter();
 	math_error(regs, error_code, X86_TRAP_MF);
-	exception_exit(regs);
+	exception_exit(prev_state);
 }
 
 dotraplinkage void
 do_simd_coprocessor_error(struct pt_regs *regs, long error_code)
 {
-	exception_enter(regs);
+	enum ctx_state prev_state;
+
+	prev_state = exception_enter();
 	math_error(regs, error_code, X86_TRAP_XF);
-	exception_exit(regs);
+	exception_exit(prev_state);
 }
 
 dotraplinkage void
@@ -638,7 +652,9 @@ EXPORT_SYMBOL_GPL(math_state_restore);
 dotraplinkage void __kprobes
 do_device_not_available(struct pt_regs *regs, long error_code)
 {
-	exception_enter(regs);
+	enum ctx_state prev_state;
+
+	prev_state = exception_enter();
 	BUG_ON(use_eager_fpu());
 
 #ifdef CONFIG_MATH_EMULATION
@@ -649,7 +665,7 @@ do_device_not_available(struct pt_regs *regs, long error_code)
 
 		info.regs = regs;
 		math_emulate(&info);
-		exception_exit(regs);
+		exception_exit(prev_state);
 		return;
 	}
 #endif
@@ -657,15 +673,16 @@ do_device_not_available(struct pt_regs *regs, long error_code)
 #ifdef CONFIG_X86_32
 	conditional_sti(regs);
 #endif
-	exception_exit(regs);
+	exception_exit(prev_state);
 }
 
 #ifdef CONFIG_X86_32
 dotraplinkage void do_iret_error(struct pt_regs *regs, long error_code)
 {
 	siginfo_t info;
+	enum ctx_state prev_state;
 
-	exception_enter(regs);
+	prev_state = exception_enter();
 	local_irq_enable();
 
 	info.si_signo = SIGILL;
@@ -677,7 +694,7 @@ dotraplinkage void do_iret_error(struct pt_regs *regs, long error_code)
 		do_trap(X86_TRAP_IRET, SIGILL, "iret exception", regs, error_code,
 			&info);
 	}
-	exception_exit(regs);
+	exception_exit(prev_state);
 }
 #endif
 

arch/x86/mm/fault.c

@@ -1222,7 +1222,9 @@ __do_page_fault(struct pt_regs *regs, unsigned long error_code)
 dotraplinkage void __kprobes
 do_page_fault(struct pt_regs *regs, unsigned long error_code)
 {
-	exception_enter(regs);
+	enum ctx_state prev_state;
+
+	prev_state = exception_enter();
 	__do_page_fault(regs, error_code);
-	exception_exit(regs);
+	exception_exit(prev_state);
 }

include/linux/context_tracking.h

@@ -5,7 +5,6 @@
 #include <linux/percpu.h>
 #include <asm/ptrace.h>
 
-#ifdef CONFIG_CONTEXT_TRACKING
 struct context_tracking {
 	/*
 	 * When active is false, probes are unset in order
@@ -14,12 +13,13 @@ struct context_tracking {
 	 * may be further optimized using static keys.
 	 */
 	bool active;
-	enum {
+	enum ctx_state {
 		IN_KERNEL = 0,
 		IN_USER,
 	} state;
 };
 
+#ifdef CONFIG_CONTEXT_TRACKING
 DECLARE_PER_CPU(struct context_tracking, context_tracking);
 
 static inline bool context_tracking_in_user(void)
@@ -35,14 +35,19 @@ static inline bool context_tracking_active(void)
 extern void user_enter(void);
 extern void user_exit(void);
 
-static inline void exception_enter(struct pt_regs *regs)
+static inline enum ctx_state exception_enter(void)
 {
+	enum ctx_state prev_ctx;
+
+	prev_ctx = this_cpu_read(context_tracking.state);
 	user_exit();
+
+	return prev_ctx;
 }
 
-static inline void exception_exit(struct pt_regs *regs)
+static inline void exception_exit(enum ctx_state prev_ctx)
 {
-	if (user_mode(regs))
+	if (prev_ctx == IN_USER)
 		user_enter();
 }
 
@@ -52,8 +57,8 @@ extern void context_tracking_task_switch(struct task_struct *prev,
 static inline bool context_tracking_in_user(void) { return false; }
 static inline void user_enter(void) { }
 static inline void user_exit(void) { }
-static inline void exception_enter(struct pt_regs *regs) { }
-static inline void exception_exit(struct pt_regs *regs) { }
+static inline enum ctx_state exception_enter(void) { return 0; }
+static inline void exception_exit(enum ctx_state prev_ctx) { }
 static inline void context_tracking_task_switch(struct task_struct *prev,
 						struct task_struct *next) { }
 #endif /* !CONFIG_CONTEXT_TRACKING */