[PATCH RFC v2] x86,mm,sched: make lazy TLB mode even lazier

From: Rik van Riel
Date: Thu Aug 25 2016 - 16:15:21 EST


Now using cmpxchg and closing a race, thanks to Peter Anvin.

---8<---

Subject: x86,mm,sched: make lazy TLB mode even lazier

Lazy TLB mode can result in an idle CPU being woken up for a TLB
flush, when all it really needed to do was flush %cr3 before the
next context switch.

This is mostly fine on bare metal, though sub-optimal from a power
saving point of view, and deeper C states could make TLB flushes
take a little longer than desired.

On virtual machines, the pain can be much worse, especially if a
currently non-running VCPU is woken up for a TLB invalidation
IPI, on a CPU that is busy running another task. It could take
a while before that IPI is handled, leading to performance issues.

This patch deals with the issue by introducing a third tlb state,
TLBSTATE_FLUSH, which causes %cr3 to be flushed at the next
context switch.

If there was a race transitioning a CPU from TLBSTATE_LAZY to
TLBSTATE_FLUSH, an invalidation IPI will be sent.

Nothing is done for a CPU that is already in TLBSTATE_FLUH mode.

This patch is totally untested, because I am at a conference right
now, and Benjamin has the test case :)

Benjamin, does this help your issue?

Signed-off-by: Rik van Riel <riel@xxxxxxxxxx>
Reported-by: Benjamin Serebrin <serebrin@xxxxxxxxxx>
---
arch/x86/include/asm/tlbflush.h | 1 +
arch/x86/mm/tlb.c | 41 ++++++++++++++++++++++++++++++++++++++---
2 files changed, 39 insertions(+), 3 deletions(-)

diff --git a/arch/x86/include/asm/tlbflush.h b/arch/x86/include/asm/tlbflush.h
index 4e5be94e079a..5ae8e4b174f8 100644
--- a/arch/x86/include/asm/tlbflush.h
+++ b/arch/x86/include/asm/tlbflush.h
@@ -310,6 +310,7 @@ void native_flush_tlb_others(const struct cpumask *cpumask,

#define TLBSTATE_OK 1
#define TLBSTATE_LAZY 2
+#define TLBSTATE_FLUSH 3

static inline void reset_lazy_tlbstate(void)
{
diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c
index 5643fd0b1a7d..8dcc0947681c 100644
--- a/arch/x86/mm/tlb.c
+++ b/arch/x86/mm/tlb.c
@@ -140,10 +140,12 @@ void switch_mm_irqs_off(struct mm_struct *prev, struct mm_struct *next,
}
#ifdef CONFIG_SMP
else {
+ int oldstate = this_cpu_read(cpu_tlbstate.state);
this_cpu_write(cpu_tlbstate.state, TLBSTATE_OK);
BUG_ON(this_cpu_read(cpu_tlbstate.active_mm) != next);

- if (!cpumask_test_cpu(cpu, mm_cpumask(next))) {
+ if (oldstate == TLBSTATE_FLUSH ||
+ !cpumask_test_cpu(cpu, mm_cpumask(next))) {
/*
* On established mms, the mm_cpumask is only changed
* from irq context, from ptep_clear_flush() while in
@@ -242,11 +244,33 @@ static void flush_tlb_func(void *info)

}

+/*
+ * This function moves a CPU from TLBSTATE_LAZY to TLBSTATE_FLUSH, which
+ * will force it to flush %cr3 at the next context switch, effectively
+ * doing a delayed TLB flush for a CPU in lazy TLB mode.
+ * Do nothing if the TLB state is already set to TLBSTATE_FLUSH.
+ */
+static bool set_lazy_tlbstate_flush(int cpu) {
+ int *tlbstate = &per_cpu(cpu_tlbstate.state, cpu);
+ bool skipflush = true;
+ if (*tlbstate == TLBSTATE_LAZY) {
+ /*
+ * If cmpxchg fails, the CPU may have context switched from
+ * TLBSTATE_LAZY to TLBSTATE_OK. Send a TLB flush IPI.
+ */
+ if (cmpxchg(tlbstate, TLBSTATE_LAZY, TLBSTATE_FLUSH) !=
+ TLBSTATE_LAZY)
+ skipflush = false;
+ }
+ return skipflush;
+}
+
void native_flush_tlb_others(const struct cpumask *cpumask,
struct mm_struct *mm, unsigned long start,
unsigned long end)
{
struct flush_tlb_info info;
+ unsigned int cpu;

if (end == 0)
end = start + PAGE_SIZE;
@@ -262,8 +286,6 @@ void native_flush_tlb_others(const struct cpumask *cpumask,
(end - start) >> PAGE_SHIFT);

if (is_uv_system()) {
- unsigned int cpu;
-
cpu = smp_processor_id();
cpumask = uv_flush_tlb_others(cpumask, mm, start, end, cpu);
if (cpumask)
@@ -271,6 +293,19 @@ void native_flush_tlb_others(const struct cpumask *cpumask,
&info, 1);
return;
}
+
+ /*
+ * Instead of sending IPIs to CPUs in lazy TLB mode, move that
+ * CPUs TLB state to TLBSTATE_FLUSH, causing the TLB to be flushed
+ * at the next context switch.
+ */
+ for_each_cpu(cpu, cpumask) {
+ if (per_cpu(cpu_tlbstate.state, cpu) != TLBSTATE_OK) {
+ if (set_lazy_tlbstate_flush(cpu))
+ cpumask_clear_cpu(cpu, (struct cpumask *)cpumask);
+ }
+ }
+
smp_call_function_many(cpumask, flush_tlb_func, &info, 1);
}