s390/mm: use invalid asce for user space when switching to init_mm

Currently only idle_task_exit() explicitly switches (switch_mm) to
init_mm. This causes the kernel asce to be loaded into cr7 and
therefore it would be used for potential user space accesses.

This is currently no problem since idle_task_exit() is nearly the last
thing a CPU executes before it is taken down. However things might
change - and therefore make sure that always the invalid asce is used
for cr7 when active_mm is init_mm.

This makes sure that all potential user space accesses will fail,
instead of accessing kernel address space.

Reviewed-by: Alexander Gordeev <agordeev@linux.ibm.com>
Signed-off-by: Heiko Carstens <hca@linux.ibm.com>

diff --git a/arch/s390/include/asm/mmu_context.h b/arch/s390/include/asm/mmu_context.h
index 87a84fc..5dc49c4 100644 (file)
--- a/arch/s390/include/asm/mmu_context.h
+++ b/arch/s390/include/asm/mmu_context.h
@@ -76,7 +76,10 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
 {
        int cpu = smp_processor_id();
 
-       S390_lowcore.user_asce = next->context.asce;
+       if (next == &init_mm)
+               S390_lowcore.user_asce = s390_invalid_asce;
+       else
+               S390_lowcore.user_asce = next->context.asce;
        cpumask_set_cpu(cpu, &next->context.cpu_attach_mask);
        /* Clear previous user-ASCE from CR7 */
        __ctl_load(s390_invalid_asce, 7, 7);