s390/mm: simplify the TLB flushing code

ptep_flush_lazy and pmdp_flush_lazy use mm->context.attach_count to
decide between a lazy TLB flush vs an immediate TLB flush. The field
contains two 16-bit counters, the number of CPUs that have the mm
attached and can create TLB entries for it and the number of CPUs in
the middle of a page table update.

The __tlb_flush_asce, ptep_flush_direct and pmdp_flush_direct functions
use the attach counter and a mask check with mm_cpumask(mm) to decide
between a local flush local of the current CPU and a global flush.

For all these functions the decision between lazy vs immediate and
local vs global TLB flush can be based on CPU masks. There are two
masks:  the mm->context.cpu_attach_mask with the CPUs that are actively
using the mm, and the mm_cpumask(mm) with the CPUs that have used the
mm since the last full flush. The decision between lazy vs immediate
flush is based on the mm->context.cpu_attach_mask, to decide between
local vs global flush the mm_cpumask(mm) is used.

With this patch all checks will use the CPU masks, the old counter
mm->context.attach_count with its two 16-bit values is turned into a
single counter mm->context.flush_count that keeps track of the number
of CPUs with incomplete page table updates. The sole user of this
counter is finish_arch_post_lock_switch() which waits for the end of
all page table updates.

Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>

3 files changed, 11 insertions, 19 deletions

diff --git a/arch/s390/include/asm/mmu.h b/arch/s390/include/asm/mmu.h
index 081b2ad99d73..18226437a832 100644
--- a/arch/s390/include/asm/mmu.h
+++ b/arch/s390/include/asm/mmu.h
@@ -6,7 +6,7 @@
 
 typedef struct {
 	cpumask_t cpu_attach_mask;
-	atomic_t attach_count;
+	atomic_t flush_count;
 	unsigned int flush_mm;
 	spinlock_t list_lock;
 	struct list_head pgtable_list;
diff --git a/arch/s390/include/asm/mmu_context.h b/arch/s390/include/asm/mmu_context.h
index c837b79b455d..f77c638bf397 100644
--- a/arch/s390/include/asm/mmu_context.h
+++ b/arch/s390/include/asm/mmu_context.h
@@ -19,7 +19,7 @@ static inline int init_new_context(struct task_struct *tsk,
 	INIT_LIST_HEAD(&mm->context.pgtable_list);
 	INIT_LIST_HEAD(&mm->context.gmap_list);
 	cpumask_clear(&mm->context.cpu_attach_mask);
-	atomic_set(&mm->context.attach_count, 0);
+	atomic_set(&mm->context.flush_count, 0);
 	mm->context.flush_mm = 0;
 #ifdef CONFIG_PGSTE
 	mm->context.alloc_pgste = page_table_allocate_pgste;
@@ -90,15 +90,12 @@ static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
 	S390_lowcore.user_asce = next->context.asce;
 	if (prev == next)
 		return;
-	if (MACHINE_HAS_TLB_LC)
-		cpumask_set_cpu(cpu, &next->context.cpu_attach_mask);
+	cpumask_set_cpu(cpu, &next->context.cpu_attach_mask);
+	cpumask_set_cpu(cpu, mm_cpumask(next));
 	/* Clear old ASCE by loading the kernel ASCE. */
 	__ctl_load(S390_lowcore.kernel_asce, 1, 1);
 	__ctl_load(S390_lowcore.kernel_asce, 7, 7);
-	atomic_inc(&next->context.attach_count);
-	atomic_dec(&prev->context.attach_count);
-	if (MACHINE_HAS_TLB_LC)
-		cpumask_clear_cpu(cpu, &prev->context.cpu_attach_mask);
+	cpumask_clear_cpu(cpu, &prev->context.cpu_attach_mask);
 }
 
 #define finish_arch_post_lock_switch finish_arch_post_lock_switch
@@ -110,10 +107,9 @@ static inline void finish_arch_post_lock_switch(void)
 	load_kernel_asce();
 	if (mm) {
 		preempt_disable();
-		while (atomic_read(&mm->context.attach_count) >> 16)
+		while (atomic_read(&mm->context.flush_count))
 			cpu_relax();
 
-		cpumask_set_cpu(smp_processor_id(), mm_cpumask(mm));
 		if (mm->context.flush_mm)
 			__tlb_flush_mm(mm);
 		preempt_enable();
@@ -128,7 +124,6 @@ static inline void activate_mm(struct mm_struct *prev,
                                struct mm_struct *next)
 {
 	switch_mm(prev, next, current);
-	cpumask_set_cpu(smp_processor_id(), mm_cpumask(next));
 	set_user_asce(next);
 }
 
diff --git a/arch/s390/include/asm/tlbflush.h b/arch/s390/include/asm/tlbflush.h
index ac02a6c37a3e..e72cea7a4bfe 100644
--- a/arch/s390/include/asm/tlbflush.h
+++ b/arch/s390/include/asm/tlbflush.h
@@ -57,7 +57,7 @@ static inline void __tlb_flush_global(void)
 static inline void __tlb_flush_full(struct mm_struct *mm)
 {
 	preempt_disable();
-	atomic_add(0x10000, &mm->context.attach_count);
+	atomic_inc(&mm->context.flush_count);
 	if (cpumask_equal(mm_cpumask(mm), cpumask_of(smp_processor_id()))) {
 		/* Local TLB flush */
 		__tlb_flush_local();
@@ -69,7 +69,7 @@ static inline void __tlb_flush_full(struct mm_struct *mm)
 			cpumask_copy(mm_cpumask(mm),
 				     &mm->context.cpu_attach_mask);
 	}
-	atomic_sub(0x10000, &mm->context.attach_count);
+	atomic_dec(&mm->context.flush_count);
 	preempt_enable();
 }
 
@@ -78,12 +78,9 @@ static inline void __tlb_flush_full(struct mm_struct *mm)
  */
 static inline void __tlb_flush_asce(struct mm_struct *mm, unsigned long asce)
 {
-	int active, count;
-
 	preempt_disable();
-	active = (mm == current->active_mm) ? 1 : 0;
-	count = atomic_add_return(0x10000, &mm->context.attach_count);
-	if (MACHINE_HAS_TLB_LC && (count & 0xffff) <= active &&
+	atomic_inc(&mm->context.flush_count);
+	if (MACHINE_HAS_TLB_LC &&
 	    cpumask_equal(mm_cpumask(mm), cpumask_of(smp_processor_id()))) {
 		__tlb_flush_idte_local(asce);
 	} else {
@@ -96,7 +93,7 @@ static inline void __tlb_flush_asce(struct mm_struct *mm, unsigned long asce)
 			cpumask_copy(mm_cpumask(mm),
 				     &mm->context.cpu_attach_mask);
 	}
-	atomic_sub(0x10000, &mm->context.attach_count);
+	atomic_dec(&mm->context.flush_count);
 	preempt_enable();
 }