diff mbox series

[06/10] x86/entry: Move nmi entry/exit into common code

Message ID 20201022222701.887660-7-ira.weiny@intel.com
State New
Headers show
Series PKS: Add Protection Keys Supervisor (PKS) support | expand

Commit Message

Ira Weiny Oct. 22, 2020, 10:26 p.m. UTC
From: Thomas Gleixner <tglx@linutronix.de>

Lockdep state handling on NMI enter and exit is nothing specific to X86. It's
not any different on other architectures. Also the extra state type is not
necessary, irqentry_state_t can carry the necessary information as well.

Move it to common code and extend irqentry_state_t to carry lockdep state.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ira Weiny <ira.weiny@intel.com>
---
 arch/x86/entry/common.c         | 34 -------------------------------
 arch/x86/include/asm/idtentry.h |  3 ---
 arch/x86/kernel/cpu/mce/core.c  |  6 +++---
 arch/x86/kernel/nmi.c           |  6 +++---
 arch/x86/kernel/traps.c         | 13 ++++++------
 include/linux/entry-common.h    | 24 +++++++++++++++++++++-
 kernel/entry/common.c           | 36 +++++++++++++++++++++++++++++++++
 7 files changed, 72 insertions(+), 50 deletions(-)

Comments

Thomas Gleixner Oct. 23, 2020, 9:50 p.m. UTC | #1
On Thu, Oct 22 2020 at 15:26, ira weiny wrote:

> From: Thomas Gleixner <tglx@linutronix.de>
>
> Lockdep state handling on NMI enter and exit is nothing specific to X86. It's
> not any different on other architectures. Also the extra state type is not
> necessary, irqentry_state_t can carry the necessary information as well.
>
> Move it to common code and extend irqentry_state_t to carry lockdep
> state.

This lacks something like:

 [ Ira: Made the states a union as they are mutually exclusive and added
        the missing kernel doc ]

Hrm.
 
>  #ifndef irqentry_state
>  typedef struct irqentry_state {
> -	bool	exit_rcu;
> +	union {
> +		bool	exit_rcu;
> +		bool	lockdep;
> +	};
>  } irqentry_state_t;
>  #endif

  -E_NO_KERNELDOC

Thanks,

        tglx
Ira Weiny Oct. 27, 2020, 7:07 a.m. UTC | #2
On Fri, Oct 23, 2020 at 11:50:11PM +0200, Thomas Gleixner wrote:
> On Thu, Oct 22 2020 at 15:26, ira weiny wrote:
> 
> > From: Thomas Gleixner <tglx@linutronix.de>
> >
> > Lockdep state handling on NMI enter and exit is nothing specific to X86. It's
> > not any different on other architectures. Also the extra state type is not
> > necessary, irqentry_state_t can carry the necessary information as well.
> >
> > Move it to common code and extend irqentry_state_t to carry lockdep
> > state.
> 
> This lacks something like:
> 
>  [ Ira: Made the states a union as they are mutually exclusive and added
>         the missing kernel doc ]

Fair enough.  done.

> 
> Hrm.
>  
> >  #ifndef irqentry_state
> >  typedef struct irqentry_state {
> > -	bool	exit_rcu;
> > +	union {
> > +		bool	exit_rcu;
> > +		bool	lockdep;
> > +	};
> >  } irqentry_state_t;
> >  #endif
> 
>   -E_NO_KERNELDOC

Adding: Paul McKenney

I'm happy to write something but I'm very unfamiliar with this code.  So I'm
getting confused what exactly exit_rcu is flagging.

I can see that exit_rcu is a bad name for the state used in
irqentry_nmi_[enter|exit]().  Furthermore, I see why 'lockdep' is a better
name.  But similar lockdep handling is used in irqentry_exit() if exit_rcu is
true...


Given my limited knowledge; here is my proposed text:

/**
 * struct irqentry_state - Opaque object for exception state storage
 * @exit_rcu: Used exclusively in the irqentry_*() calls; tracks if the
 *            exception hit the idle task which requires special handling,
 *            including calling rcu_irq_exit(), when the exception exits.
 * @lockdep: Used exclusively in the irqentry_nmi_*() calls; ensures lockdep
 *           tracking is maintained if hardirqs were already enabled
 *
 * This opaque object is filled in by the irqentry_*_enter() functions and
 * should be passed back into the corresponding irqentry_*_exit() functions
 * when the exception is complete.
 *
 * Callers of irqentry_*_[enter|exit]() should consider this structure opaque
 * and all members private.  Descriptions of the members are provided to aid in
 * the maintenance of the irqentry_*() functions.
 */


Perhaps Paul can enlighten me on how exit_rcu is used beyond just flagging a
call to rcu_irq_exit()?

Why do we call lockdep_hardirqs_off() only when in the idle task?  That implies
that regs_irqs_disabled() can only be false if we were in the idle task to
match up the lockdep on/off calls.  This does not make sense to me because why
do we need the extra check for exit_rcu?  I'm still trying to understand when
regs_irqs_disabled() is false.


        } else if (!regs_irqs_disabled(regs)) {
...
        } else {
                /*
                 * IRQ flags state is correct already. Just tell RCU if it
                 * was not watching on entry.
                 */
                if (state.exit_rcu)
                        rcu_irq_exit();
        }

Also, the comment in irqentry_enter() refers to irq_enter_from_user_mode() which
does not seem to exist anymore.  So I'm not sure what careful sequence it is
referring to.

        /*
         * If RCU is not watching then the same careful
         * sequence vs. lockdep and tracing is required
         * as in irq_enter_from_user_mode().
         */

?

Ira
Thomas Gleixner Oct. 27, 2020, 2:18 p.m. UTC | #3
On Tue, Oct 27 2020 at 00:07, Ira Weiny wrote:
> On Fri, Oct 23, 2020 at 11:50:11PM +0200, Thomas Gleixner wrote:
>> >  #ifndef irqentry_state
>> >  typedef struct irqentry_state {
>> > -	bool	exit_rcu;
>> > +	union {
>> > +		bool	exit_rcu;
>> > +		bool	lockdep;
>> > +	};
>> >  } irqentry_state_t;
>> >  #endif
>> 
>>   -E_NO_KERNELDOC
>
> Adding: Paul McKenney
>
> I'm happy to write something but I'm very unfamiliar with this code.  So I'm
> getting confused what exactly exit_rcu is flagging.
>
> I can see that exit_rcu is a bad name for the state used in
> irqentry_nmi_[enter|exit]().  Furthermore, I see why 'lockdep' is a better
> name.  But similar lockdep handling is used in irqentry_exit() if exit_rcu is
> true...

No, it's not similar at all. Lockdep state vs. interrupts and regular
exceptions is always consistent.

In the NMI case, that's not guaranteed because of

       local_irq_disable()
         arch_local_irq_disable()
                                        <- NMI race window
         trace_hardirqs_off()

same the other way round

       local_irq_enable()
         trace_hardirqs_on()
                                        <- NMI race window
         arch_local_irq_enable()

IOW, the hardware state and the lockdep state are not consistent.

> /**
>  * struct irqentry_state - Opaque object for exception state storage
>  * @exit_rcu: Used exclusively in the irqentry_*() calls; tracks if the
>  *            exception hit the idle task which requires special handling,
>  *            including calling rcu_irq_exit(), when the exception
>  exits.

calls; signals whether the exit path has to invoke rcu_irq_exit().

>  * @lockdep: Used exclusively in the irqentry_nmi_*() calls; ensures lockdep
>  *           tracking is maintained if hardirqs were already enabled

   ensures that lockdep state is restored correctly on exit from nmi.

>  *
>  * This opaque object is filled in by the irqentry_*_enter() functions and
>  * should be passed back into the corresponding irqentry_*_exit()
>  functions

s/should/must/

>  * when the exception is complete.
>  *
>  * Callers of irqentry_*_[enter|exit]() should consider this structure
>  opaque

s/should/must/

>  * and all members private.  Descriptions of the members are provided to aid in
>  * the maintenance of the irqentry_*() functions.
>  */
>
> Perhaps Paul can enlighten me on how exit_rcu is used beyond just flagging a
> call to rcu_irq_exit()?

I can do that as well :) The only purpose is to invoke rcu_irq_exit()
conditionally.

> Why do we call lockdep_hardirqs_off() only when in the idle task?  That implies
> that regs_irqs_disabled() can only be false if we were in the idle task to
> match up the lockdep on/off calls.

You're reading the code slightly wrong.

> This does not make sense to me because why do we need the extra check
> for exit_rcu?  I'm still trying to understand when regs_irqs_disabled() is false.

It's false when the interrupted context had interrupts enabled.

So we have the following scenarios:

 Usermode   Idletask   irqs enabled  RCU entry  RCU exit
   Y           N           Y		Y          Y

   N           N           Y            N          N 
   N           N           N            N          N
   N           Y           Y            Y          Y
   N           Y           N            Y          Y                       

Now you might wonder about irqs enabled/disabled. This code is not only
used for interrupts (device, ipi, local timer...) where interrupts are
obviously enabled, it's also used for exception entry/exit. You can have
e.g. pagefaults in interrupt disabled regions.

> Also, the comment in irqentry_enter() refers to irq_enter_from_user_mode() which
> does not seem to exist anymore.  So I'm not sure what careful sequence it is
> referring to.

That was renamed to irqentry_enter_from_user_mode() and the comment was
not updated. Sorry for leaving this hard to solve puzzle around.

Thanks,

        tglx
diff mbox series

Patch

diff --git a/arch/x86/entry/common.c b/arch/x86/entry/common.c
index 870efeec8bda..18d8f17f755c 100644
--- a/arch/x86/entry/common.c
+++ b/arch/x86/entry/common.c
@@ -209,40 +209,6 @@  SYSCALL_DEFINE0(ni_syscall)
 	return -ENOSYS;
 }
 
-noinstr bool idtentry_enter_nmi(struct pt_regs *regs)
-{
-	bool irq_state = lockdep_hardirqs_enabled();
-
-	__nmi_enter();
-	lockdep_hardirqs_off(CALLER_ADDR0);
-	lockdep_hardirq_enter();
-	rcu_nmi_enter();
-
-	instrumentation_begin();
-	trace_hardirqs_off_finish();
-	ftrace_nmi_enter();
-	instrumentation_end();
-
-	return irq_state;
-}
-
-noinstr void idtentry_exit_nmi(struct pt_regs *regs, bool restore)
-{
-	instrumentation_begin();
-	ftrace_nmi_exit();
-	if (restore) {
-		trace_hardirqs_on_prepare();
-		lockdep_hardirqs_on_prepare(CALLER_ADDR0);
-	}
-	instrumentation_end();
-
-	rcu_nmi_exit();
-	lockdep_hardirq_exit();
-	if (restore)
-		lockdep_hardirqs_on(CALLER_ADDR0);
-	__nmi_exit();
-}
-
 #ifdef CONFIG_XEN_PV
 #ifndef CONFIG_PREEMPTION
 /*
diff --git a/arch/x86/include/asm/idtentry.h b/arch/x86/include/asm/idtentry.h
index b2442eb0ac2f..247a60a47331 100644
--- a/arch/x86/include/asm/idtentry.h
+++ b/arch/x86/include/asm/idtentry.h
@@ -11,9 +11,6 @@ 
 
 #include <asm/irq_stack.h>
 
-bool idtentry_enter_nmi(struct pt_regs *regs);
-void idtentry_exit_nmi(struct pt_regs *regs, bool irq_state);
-
 /**
  * DECLARE_IDTENTRY - Declare functions for simple IDT entry points
  *		      No error code pushed by hardware
diff --git a/arch/x86/kernel/cpu/mce/core.c b/arch/x86/kernel/cpu/mce/core.c
index 1c08cb9eb9f6..eb3338c0bbc1 100644
--- a/arch/x86/kernel/cpu/mce/core.c
+++ b/arch/x86/kernel/cpu/mce/core.c
@@ -1983,7 +1983,7 @@  void (*machine_check_vector)(struct pt_regs *) = unexpected_machine_check;
 
 static __always_inline void exc_machine_check_kernel(struct pt_regs *regs)
 {
-	bool irq_state;
+	irqentry_state_t irq_state;
 
 	WARN_ON_ONCE(user_mode(regs));
 
@@ -1995,7 +1995,7 @@  static __always_inline void exc_machine_check_kernel(struct pt_regs *regs)
 	    mce_check_crashing_cpu())
 		return;
 
-	irq_state = idtentry_enter_nmi(regs);
+	irq_state = irqentry_nmi_enter(regs);
 	/*
 	 * The call targets are marked noinstr, but objtool can't figure
 	 * that out because it's an indirect call. Annotate it.
@@ -2006,7 +2006,7 @@  static __always_inline void exc_machine_check_kernel(struct pt_regs *regs)
 	if (regs->flags & X86_EFLAGS_IF)
 		trace_hardirqs_on_prepare();
 	instrumentation_end();
-	idtentry_exit_nmi(regs, irq_state);
+	irqentry_nmi_exit(regs, irq_state);
 }
 
 static __always_inline void exc_machine_check_user(struct pt_regs *regs)
diff --git a/arch/x86/kernel/nmi.c b/arch/x86/kernel/nmi.c
index 4bc77aaf1303..bf250a339655 100644
--- a/arch/x86/kernel/nmi.c
+++ b/arch/x86/kernel/nmi.c
@@ -475,7 +475,7 @@  static DEFINE_PER_CPU(unsigned long, nmi_dr7);
 
 DEFINE_IDTENTRY_RAW(exc_nmi)
 {
-	bool irq_state;
+	irqentry_state_t irq_state;
 
 	/*
 	 * Re-enable NMIs right here when running as an SEV-ES guest. This might
@@ -502,14 +502,14 @@  DEFINE_IDTENTRY_RAW(exc_nmi)
 
 	this_cpu_write(nmi_dr7, local_db_save());
 
-	irq_state = idtentry_enter_nmi(regs);
+	irq_state = irqentry_nmi_enter(regs);
 
 	inc_irq_stat(__nmi_count);
 
 	if (!ignore_nmis)
 		default_do_nmi(regs);
 
-	idtentry_exit_nmi(regs, irq_state);
+	irqentry_nmi_exit(regs, irq_state);
 
 	local_db_restore(this_cpu_read(nmi_dr7));
 
diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c
index 3c70fb34028b..bffbbe29fc8c 100644
--- a/arch/x86/kernel/traps.c
+++ b/arch/x86/kernel/traps.c
@@ -405,7 +405,7 @@  DEFINE_IDTENTRY_DF(exc_double_fault)
 	}
 #endif
 
-	idtentry_enter_nmi(regs);
+	irqentry_nmi_enter(regs);
 	instrumentation_begin();
 	notify_die(DIE_TRAP, str, regs, error_code, X86_TRAP_DF, SIGSEGV);
 
@@ -651,12 +651,13 @@  DEFINE_IDTENTRY_RAW(exc_int3)
 		instrumentation_end();
 		irqentry_exit_to_user_mode(regs);
 	} else {
-		bool irq_state = idtentry_enter_nmi(regs);
+		irqentry_state_t irq_state = irqentry_nmi_enter(regs);
+
 		instrumentation_begin();
 		if (!do_int3(regs))
 			die("int3", regs, 0);
 		instrumentation_end();
-		idtentry_exit_nmi(regs, irq_state);
+		irqentry_nmi_exit(regs, irq_state);
 	}
 }
 
@@ -864,7 +865,7 @@  static __always_inline void exc_debug_kernel(struct pt_regs *regs,
 	 * includes the entry stack is excluded for everything.
 	 */
 	unsigned long dr7 = local_db_save();
-	bool irq_state = idtentry_enter_nmi(regs);
+	irqentry_state_t irq_state = irqentry_nmi_enter(regs);
 	instrumentation_begin();
 
 	/*
@@ -907,7 +908,7 @@  static __always_inline void exc_debug_kernel(struct pt_regs *regs,
 		regs->flags &= ~X86_EFLAGS_TF;
 out:
 	instrumentation_end();
-	idtentry_exit_nmi(regs, irq_state);
+	irqentry_nmi_exit(regs, irq_state);
 
 	local_db_restore(dr7);
 }
@@ -925,7 +926,7 @@  static __always_inline void exc_debug_user(struct pt_regs *regs,
 
 	/*
 	 * NB: We can't easily clear DR7 here because
-	 * idtentry_exit_to_usermode() can invoke ptrace, schedule, access
+	 * irqentry_exit_to_usermode() can invoke ptrace, schedule, access
 	 * user memory, etc.  This means that a recursive #DB is possible.  If
 	 * this happens, that #DB will hit exc_debug_kernel() and clear DR7.
 	 * Since we're not on the IST stack right now, everything will be
diff --git a/include/linux/entry-common.h b/include/linux/entry-common.h
index 474f29638d2c..47f9a0658acf 100644
--- a/include/linux/entry-common.h
+++ b/include/linux/entry-common.h
@@ -342,7 +342,10 @@  void irqentry_exit_to_user_mode(struct pt_regs *regs);
 
 #ifndef irqentry_state
 typedef struct irqentry_state {
-	bool	exit_rcu;
+	union {
+		bool	exit_rcu;
+		bool	lockdep;
+	};
 } irqentry_state_t;
 #endif
 
@@ -402,4 +405,23 @@  void irqentry_exit_cond_resched(void);
  */
 void noinstr irqentry_exit(struct pt_regs *regs, irqentry_state_t state);
 
+/**
+ * irqentry_nmi_enter - Handle NMI entry
+ * @regs:	Pointer to currents pt_regs
+ *
+ * Similar to irqentry_enter() but taking care of the NMI constraints.
+ */
+irqentry_state_t noinstr irqentry_nmi_enter(struct pt_regs *regs);
+
+/**
+ * irqentry_nmi_exit - Handle return from NMI handling
+ * @regs:	Pointer to pt_regs (NMI entry regs)
+ * @irq_state:	Return value from matching call to irqentry_nmi_enter()
+ *
+ * Last action before returning to the low level assmenbly code.
+ *
+ * Counterpart to irqentry_nmi_enter().
+ */
+void noinstr irqentry_nmi_exit(struct pt_regs *regs, irqentry_state_t irq_state);
+
 #endif
diff --git a/kernel/entry/common.c b/kernel/entry/common.c
index 0a1e20f8d4e8..5cc2e4174d7c 100644
--- a/kernel/entry/common.c
+++ b/kernel/entry/common.c
@@ -398,3 +398,39 @@  noinstr void irqentry_exit(struct pt_regs *regs, irqentry_state_t state)
 			rcu_irq_exit();
 	}
 }
+
+irqentry_state_t noinstr irqentry_nmi_enter(struct pt_regs *regs)
+{
+	irqentry_state_t irq_state;
+
+	irq_state.lockdep = lockdep_hardirqs_enabled();
+
+	__nmi_enter();
+	lockdep_hardirqs_off(CALLER_ADDR0);
+	lockdep_hardirq_enter();
+	rcu_nmi_enter();
+
+	instrumentation_begin();
+	trace_hardirqs_off_finish();
+	ftrace_nmi_enter();
+	instrumentation_end();
+
+	return irq_state;
+}
+
+void noinstr irqentry_nmi_exit(struct pt_regs *regs, irqentry_state_t irq_state)
+{
+	instrumentation_begin();
+	ftrace_nmi_exit();
+	if (irq_state.lockdep) {
+		trace_hardirqs_on_prepare();
+		lockdep_hardirqs_on_prepare(CALLER_ADDR0);
+	}
+	instrumentation_end();
+
+	rcu_nmi_exit();
+	lockdep_hardirq_exit();
+	if (irq_state.lockdep)
+		lockdep_hardirqs_on(CALLER_ADDR0);
+	__nmi_exit();
+}