[patch 10/14] x86/irq: Clear move_in_progress before sending cleanup IPI

[Thomas Gleixner <tglx@linutronix.de>]

[-- Attachment #1: x86-irq--Protect-send_cleanup_vector-with-vector_lock.patch --]
[-- Type: text/plain, Size: 1989 bytes --]

send_cleanup_vector() fiddles with the old_domain mask unprotected because it
relies on the protection by the move_in_progress flag. But this is fatal, as
the flag is reset after the IPI has been sent. So a cpu which receives the IPI
can still see the flag set and therefor ignores the cleanup request. If no
other cleanup request happens then the vector stays stale on that cpu and in
case of an irq removal the vector still persists. That can lead to use after
free when the next cleanup IPI happens.

Protect the code with vector_lock and clear move_in_progress before sending
the IPI.

This does not plug the race which Joe reported because:

CPU0                          CPU1                      CPU2
lock_vector()
data->move_in_progress=0
sendIPI()                       
unlock_vector()
                              set_affinity()
                              assign_irq_vector()
                              lock_vector()             handle_IPI
                              move_in_progress = 1      lock_vector()
                              unlock_vector()
                                                        move_in_progress == 1

The full fix comes with a later patch.

Reported-by: Joe Lawrence <joe.lawrence@stratus.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
---
 arch/x86/kernel/apic/vector.c |    4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

--- a/arch/x86/kernel/apic/vector.c
+++ b/arch/x86/kernel/apic/vector.c
@@ -530,6 +530,8 @@ static void __send_cleanup_vector(struct
 {
 	cpumask_var_t cleanup_mask;
 
+	raw_spin_lock(&vector_lock);
+	data->move_in_progress = 0;
 	if (unlikely(!alloc_cpumask_var(&cleanup_mask, GFP_ATOMIC))) {
 		unsigned int i;
 
@@ -541,7 +543,7 @@ static void __send_cleanup_vector(struct
 		apic->send_IPI_mask(cleanup_mask, IRQ_MOVE_CLEANUP_VECTOR);
 		free_cpumask_var(cleanup_mask);
 	}
-	data->move_in_progress = 0;
+	raw_spin_unlock(&vector_lock);
 }
 
 void send_cleanup_vector(struct irq_cfg *cfg)