Nadav Amit
2019-May-25 08:22 UTC
[RFC PATCH 5/6] x86/mm/tlb: Flush remote and local TLBs concurrently
To improve TLB shootdown performance, flush the remote and local TLBs
concurrently. Introduce flush_tlb_multi() that does so. The current
flush_tlb_others() interface is kept, since paravirtual interfaces need
to be adapted first before it can be removed. This is left for future
work. In such PV environments, TLB flushes are not performed, at this
time, concurrently.
Add a static key to tell whether this new interface is supported.
Cc: "K. Y. Srinivasan" <kys at microsoft.com>
Cc: Haiyang Zhang <haiyangz at microsoft.com>
Cc: Stephen Hemminger <sthemmin at microsoft.com>
Cc: Sasha Levin <sashal at kernel.org>
Cc: Thomas Gleixner <tglx at linutronix.de>
Cc: Ingo Molnar <mingo at redhat.com>
Cc: Borislav Petkov <bp at alien8.de>
Cc: x86 at kernel.org
Cc: Juergen Gross <jgross at suse.com>
Cc: Paolo Bonzini <pbonzini at redhat.com>
Cc: Dave Hansen <dave.hansen at linux.intel.com>
Cc: Andy Lutomirski <luto at kernel.org>
Cc: Peter Zijlstra <peterz at infradead.org>
Cc: Boris Ostrovsky <boris.ostrovsky at oracle.com>
Cc: linux-hyperv at vger.kernel.org
Cc: linux-kernel at vger.kernel.org
Cc: virtualization at lists.linux-foundation.org
Cc: kvm at vger.kernel.org
Cc: xen-devel at lists.xenproject.org
Signed-off-by: Nadav Amit <namit at vmware.com>
---
arch/x86/hyperv/mmu.c | 2 +
arch/x86/include/asm/paravirt.h | 8 +++
arch/x86/include/asm/paravirt_types.h | 6 ++
arch/x86/include/asm/tlbflush.h | 6 ++
arch/x86/kernel/kvm.c | 1 +
arch/x86/kernel/paravirt.c | 3 +
arch/x86/mm/tlb.c | 80 +++++++++++++++++++++++----
arch/x86/xen/mmu_pv.c | 2 +
8 files changed, 96 insertions(+), 12 deletions(-)
diff --git a/arch/x86/hyperv/mmu.c b/arch/x86/hyperv/mmu.c
index e65d7fe6489f..ca28b400c87c 100644
--- a/arch/x86/hyperv/mmu.c
+++ b/arch/x86/hyperv/mmu.c
@@ -233,4 +233,6 @@ void hyperv_setup_mmu_ops(void)
pr_info("Using hypercall for remote TLB flush\n");
pv_ops.mmu.flush_tlb_others = hyperv_flush_tlb_others;
pv_ops.mmu.tlb_remove_table = tlb_remove_table;
+
+ static_key_disable(&flush_tlb_multi_enabled.key);
}
diff --git a/arch/x86/include/asm/paravirt.h b/arch/x86/include/asm/paravirt.h
index c25c38a05c1c..192be7254457 100644
--- a/arch/x86/include/asm/paravirt.h
+++ b/arch/x86/include/asm/paravirt.h
@@ -47,6 +47,8 @@ static inline void slow_down_io(void)
#endif
}
+DECLARE_STATIC_KEY_TRUE(flush_tlb_multi_enabled);
+
static inline void __flush_tlb(void)
{
PVOP_VCALL0(mmu.flush_tlb_user);
@@ -62,6 +64,12 @@ static inline void __flush_tlb_one_user(unsigned long addr)
PVOP_VCALL1(mmu.flush_tlb_one_user, addr);
}
+static inline void flush_tlb_multi(const struct cpumask *cpumask,
+ const struct flush_tlb_info *info)
+{
+ PVOP_VCALL2(mmu.flush_tlb_multi, cpumask, info);
+}
+
static inline void flush_tlb_others(const struct cpumask *cpumask,
const struct flush_tlb_info *info)
{
diff --git a/arch/x86/include/asm/paravirt_types.h
b/arch/x86/include/asm/paravirt_types.h
index 946f8f1f1efc..3a156e63c57d 100644
--- a/arch/x86/include/asm/paravirt_types.h
+++ b/arch/x86/include/asm/paravirt_types.h
@@ -211,6 +211,12 @@ struct pv_mmu_ops {
void (*flush_tlb_user)(void);
void (*flush_tlb_kernel)(void);
void (*flush_tlb_one_user)(unsigned long addr);
+ /*
+ * flush_tlb_multi() is the preferred interface. When it is used,
+ * flush_tlb_others() should return false.
+ */
+ void (*flush_tlb_multi)(const struct cpumask *cpus,
+ const struct flush_tlb_info *info);
void (*flush_tlb_others)(const struct cpumask *cpus,
const struct flush_tlb_info *info);
diff --git a/arch/x86/include/asm/tlbflush.h b/arch/x86/include/asm/tlbflush.h
index dee375831962..79272938cf79 100644
--- a/arch/x86/include/asm/tlbflush.h
+++ b/arch/x86/include/asm/tlbflush.h
@@ -569,6 +569,9 @@ static inline void flush_tlb_page(struct vm_area_struct
*vma, unsigned long a)
flush_tlb_mm_range(vma->vm_mm, a, a + PAGE_SIZE, PAGE_SHIFT, false);
}
+void native_flush_tlb_multi(const struct cpumask *cpumask,
+ const struct flush_tlb_info *info);
+
void native_flush_tlb_others(const struct cpumask *cpumask,
const struct flush_tlb_info *info);
@@ -593,6 +596,9 @@ static inline void arch_tlbbatch_add_mm(struct
arch_tlbflush_unmap_batch *batch,
extern void arch_tlbbatch_flush(struct arch_tlbflush_unmap_batch *batch);
#ifndef CONFIG_PARAVIRT
+#define flush_tlb_multi(mask, info) \
+ native_flush_tlb_multi(mask, info)
+
#define flush_tlb_others(mask, info) \
native_flush_tlb_others(mask, info)
diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c
index 3f0cc828cc36..c1c2b88ea3f1 100644
--- a/arch/x86/kernel/kvm.c
+++ b/arch/x86/kernel/kvm.c
@@ -643,6 +643,7 @@ static void __init kvm_guest_init(void)
kvm_para_has_feature(KVM_FEATURE_STEAL_TIME)) {
pv_ops.mmu.flush_tlb_others = kvm_flush_tlb_others;
pv_ops.mmu.tlb_remove_table = tlb_remove_table;
+ static_key_disable(&flush_tlb_multi_enabled.key);
}
if (kvm_para_has_feature(KVM_FEATURE_PV_EOI))
diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c
index 5492a669f658..1314f89304a8 100644
--- a/arch/x86/kernel/paravirt.c
+++ b/arch/x86/kernel/paravirt.c
@@ -171,6 +171,8 @@ unsigned paravirt_patch_insns(void *insn_buff, unsigned len,
return insn_len;
}
+DEFINE_STATIC_KEY_TRUE(flush_tlb_multi_enabled);
+
static void native_flush_tlb(void)
{
__native_flush_tlb();
@@ -375,6 +377,7 @@ struct paravirt_patch_template pv_ops = {
.mmu.flush_tlb_user = native_flush_tlb,
.mmu.flush_tlb_kernel = native_flush_tlb_global,
.mmu.flush_tlb_one_user = native_flush_tlb_one_user,
+ .mmu.flush_tlb_multi = native_flush_tlb_multi,
.mmu.flush_tlb_others = native_flush_tlb_others,
.mmu.tlb_remove_table (void (*)(struct mmu_gather *, void
*))tlb_remove_page,
diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c
index afd2859a6966..0ec2bfca7581 100644
--- a/arch/x86/mm/tlb.c
+++ b/arch/x86/mm/tlb.c
@@ -550,7 +550,7 @@ static void flush_tlb_func_common(const struct
flush_tlb_info *f,
* garbage into our TLB. Since switching to init_mm is barely
* slower than a minimal flush, just switch to init_mm.
*
- * This should be rare, with native_flush_tlb_others skipping
+ * This should be rare, with native_flush_tlb_multi skipping
* IPIs to lazy TLB mode CPUs.
*/
switch_mm_irqs_off(NULL, &init_mm, NULL);
@@ -634,9 +634,12 @@ static void flush_tlb_func_common(const struct
flush_tlb_info *f,
this_cpu_write(cpu_tlbstate.ctxs[loaded_mm_asid].tlb_gen, mm_tlb_gen);
}
-static void flush_tlb_func_local(const void *info, enum tlb_flush_reason
reason)
+static void flush_tlb_func_local(void *info)
{
const struct flush_tlb_info *f = info;
+ enum tlb_flush_reason reason;
+
+ reason = (f->mm == NULL) ? TLB_LOCAL_SHOOTDOWN : TLB_LOCAL_MM_SHOOTDOWN;
flush_tlb_func_common(f, true, reason);
}
@@ -654,14 +657,30 @@ static void flush_tlb_func_remote(void *info)
flush_tlb_func_common(f, false, TLB_REMOTE_SHOOTDOWN);
}
-static bool tlb_is_not_lazy(int cpu, void *data)
+static inline bool tlb_is_not_lazy(int cpu)
{
return !per_cpu(cpu_tlbstate.is_lazy, cpu);
}
-void native_flush_tlb_others(const struct cpumask *cpumask,
- const struct flush_tlb_info *info)
+static DEFINE_PER_CPU(cpumask_t, flush_tlb_mask);
+
+void native_flush_tlb_multi(const struct cpumask *cpumask,
+ const struct flush_tlb_info *info)
{
+ /*
+ * native_flush_tlb_multi() can handle a single CPU, but it is
+ * suboptimal if the local TLB should be flushed, and therefore should
+ * not be used in such case. Check that it is not used in such case,
+ * and use this assumption for tracing and accounting of remote TLB
+ * flushes.
+ */
+ VM_WARN_ON(!cpumask_any_but(cpumask, smp_processor_id()));
+
+ /*
+ * Do accounting and tracing. Note that there are (and have always been)
+ * cases in which a remote TLB flush will be traced, but eventually
+ * would not happen.
+ */
count_vm_tlb_event(NR_TLB_REMOTE_FLUSH);
if (info->end == TLB_FLUSH_ALL)
trace_tlb_flush(TLB_REMOTE_SEND_IPI, TLB_FLUSH_ALL);
@@ -681,10 +700,14 @@ void native_flush_tlb_others(const struct cpumask
*cpumask,
* means that the percpu tlb_gen variables won't be updated
* and we'll do pointless flushes on future context switches.
*
- * Rather than hooking native_flush_tlb_others() here, I think
+ * Rather than hooking native_flush_tlb_multi() here, I think
* that UV should be updated so that smp_call_function_many(),
* etc, are optimal on UV.
*/
+ local_irq_disable();
+ flush_tlb_func_local((__force void *)info);
+ local_irq_enable();
+
cpumask = uv_flush_tlb_others(cpumask, info);
if (cpumask)
smp_call_function_many(cpumask, flush_tlb_func_remote,
@@ -703,11 +726,39 @@ void native_flush_tlb_others(const struct cpumask
*cpumask,
* doing a speculative memory access.
*/
if (info->freed_tables)
- smp_call_function_many(cpumask, flush_tlb_func_remote,
- (void *)info, 1);
- else
- on_each_cpu_cond_mask(tlb_is_not_lazy, flush_tlb_func_remote,
- (void *)info, 1, GFP_ATOMIC, cpumask);
+ __smp_call_function_many(cpumask, flush_tlb_func_remote,
+ flush_tlb_func_local, (void *)info, 1);
+ else {
+ /*
+ * Although we could have used on_each_cpu_cond_mask(),
+ * open-coding it has several performance advantages: (1) we can
+ * use specialized functions for remote and local flushes; (2)
+ * no need for indirect branch to test if TLB is lazy; (3) we
+ * can use a designated cpumask for evaluating the condition
+ * instead of allocating a new one.
+ *
+ * This works under the assumption that there are no nested TLB
+ * flushes, an assumption that is already made in
+ * flush_tlb_mm_range().
+ */
+ struct cpumask *cond_cpumask = this_cpu_ptr(&flush_tlb_mask);
+ int cpu;
+
+ cpumask_clear(cond_cpumask);
+
+ for_each_cpu(cpu, cpumask) {
+ if (tlb_is_not_lazy(cpu))
+ __cpumask_set_cpu(cpu, cond_cpumask);
+ }
+ __smp_call_function_many(cond_cpumask, flush_tlb_func_remote,
+ flush_tlb_func_local, (void *)info, 1);
+ }
+}
+
+void native_flush_tlb_others(const struct cpumask *cpumask,
+ const struct flush_tlb_info *info)
+{
+ native_flush_tlb_multi(cpumask, info);
}
/*
@@ -773,10 +824,15 @@ static void flush_tlb_on_cpus(const cpumask_t *cpumask,
{
int this_cpu = smp_processor_id();
+ if (static_branch_likely(&flush_tlb_multi_enabled)) {
+ flush_tlb_multi(cpumask, info);
+ return;
+ }
+
if (cpumask_test_cpu(this_cpu, cpumask)) {
lockdep_assert_irqs_enabled();
local_irq_disable();
- flush_tlb_func_local(info, TLB_LOCAL_MM_SHOOTDOWN);
+ flush_tlb_func_local((__force void *)info);
local_irq_enable();
}
diff --git a/arch/x86/xen/mmu_pv.c b/arch/x86/xen/mmu_pv.c
index beb44e22afdf..0cb277848cb4 100644
--- a/arch/x86/xen/mmu_pv.c
+++ b/arch/x86/xen/mmu_pv.c
@@ -2474,6 +2474,8 @@ void __init xen_init_mmu_ops(void)
pv_ops.mmu = xen_mmu_ops;
+ static_key_disable(&flush_tlb_multi_enabled.key);
+
memset(dummy_mapping, 0xff, PAGE_SIZE);
}
--
2.20.1
Nadav Amit
2019-May-25 08:38 UTC
[RFC PATCH 5/6] x86/mm/tlb: Flush remote and local TLBs concurrently
> On May 25, 2019, at 1:22 AM, Nadav Amit <namit at vmware.com> wrote: > > To improve TLB shootdown performance, flush the remote and local TLBs > concurrently. Introduce flush_tlb_multi() that does so. The current > flush_tlb_others() interface is kept, since paravirtual interfaces need > to be adapted first before it can be removed. This is left for future > work. In such PV environments, TLB flushes are not performed, at this > time, concurrently. > > +void native_flush_tlb_multi(const struct cpumask *cpumask, > + const struct flush_tlb_info *info) > { > + /* > + * native_flush_tlb_multi() can handle a single CPU, but it is > + * suboptimal if the local TLB should be flushed, and therefore should > + * not be used in such case. Check that it is not used in such case, > + * and use this assumption for tracing and accounting of remote TLB > + * flushes. > + */ > + VM_WARN_ON(!cpumask_any_but(cpumask, smp_processor_id()));This warning might fire off incorrectly and will be removed.
Juergen Gross
2019-May-25 08:54 UTC
[RFC PATCH 5/6] x86/mm/tlb: Flush remote and local TLBs concurrently
On 25/05/2019 10:22, Nadav Amit wrote:> To improve TLB shootdown performance, flush the remote and local TLBs > concurrently. Introduce flush_tlb_multi() that does so. The current > flush_tlb_others() interface is kept, since paravirtual interfaces need > to be adapted first before it can be removed. This is left for future > work. In such PV environments, TLB flushes are not performed, at this > time, concurrently. > > Add a static key to tell whether this new interface is supported. > > Cc: "K. Y. Srinivasan" <kys at microsoft.com> > Cc: Haiyang Zhang <haiyangz at microsoft.com> > Cc: Stephen Hemminger <sthemmin at microsoft.com> > Cc: Sasha Levin <sashal at kernel.org> > Cc: Thomas Gleixner <tglx at linutronix.de> > Cc: Ingo Molnar <mingo at redhat.com> > Cc: Borislav Petkov <bp at alien8.de> > Cc: x86 at kernel.org > Cc: Juergen Gross <jgross at suse.com> > Cc: Paolo Bonzini <pbonzini at redhat.com> > Cc: Dave Hansen <dave.hansen at linux.intel.com> > Cc: Andy Lutomirski <luto at kernel.org> > Cc: Peter Zijlstra <peterz at infradead.org> > Cc: Boris Ostrovsky <boris.ostrovsky at oracle.com> > Cc: linux-hyperv at vger.kernel.org > Cc: linux-kernel at vger.kernel.org > Cc: virtualization at lists.linux-foundation.org > Cc: kvm at vger.kernel.org > Cc: xen-devel at lists.xenproject.org > Signed-off-by: Nadav Amit <namit at vmware.com> > --- > arch/x86/hyperv/mmu.c | 2 + > arch/x86/include/asm/paravirt.h | 8 +++ > arch/x86/include/asm/paravirt_types.h | 6 ++ > arch/x86/include/asm/tlbflush.h | 6 ++ > arch/x86/kernel/kvm.c | 1 + > arch/x86/kernel/paravirt.c | 3 + > arch/x86/mm/tlb.c | 80 +++++++++++++++++++++++---- > arch/x86/xen/mmu_pv.c | 2 + > 8 files changed, 96 insertions(+), 12 deletions(-) > > diff --git a/arch/x86/hyperv/mmu.c b/arch/x86/hyperv/mmu.c > index e65d7fe6489f..ca28b400c87c 100644 > --- a/arch/x86/hyperv/mmu.c > +++ b/arch/x86/hyperv/mmu.c > @@ -233,4 +233,6 @@ void hyperv_setup_mmu_ops(void) > pr_info("Using hypercall for remote TLB flush\n"); > pv_ops.mmu.flush_tlb_others = hyperv_flush_tlb_others; > pv_ops.mmu.tlb_remove_table = tlb_remove_table; > + > + static_key_disable(&flush_tlb_multi_enabled.key); > } > diff --git a/arch/x86/include/asm/paravirt.h b/arch/x86/include/asm/paravirt.h > index c25c38a05c1c..192be7254457 100644 > --- a/arch/x86/include/asm/paravirt.h > +++ b/arch/x86/include/asm/paravirt.h > @@ -47,6 +47,8 @@ static inline void slow_down_io(void) > #endif > } > > +DECLARE_STATIC_KEY_TRUE(flush_tlb_multi_enabled); > + > static inline void __flush_tlb(void) > { > PVOP_VCALL0(mmu.flush_tlb_user); > @@ -62,6 +64,12 @@ static inline void __flush_tlb_one_user(unsigned long addr) > PVOP_VCALL1(mmu.flush_tlb_one_user, addr); > } > > +static inline void flush_tlb_multi(const struct cpumask *cpumask, > + const struct flush_tlb_info *info) > +{ > + PVOP_VCALL2(mmu.flush_tlb_multi, cpumask, info); > +} > + > static inline void flush_tlb_others(const struct cpumask *cpumask, > const struct flush_tlb_info *info) > { > diff --git a/arch/x86/include/asm/paravirt_types.h b/arch/x86/include/asm/paravirt_types.h > index 946f8f1f1efc..3a156e63c57d 100644 > --- a/arch/x86/include/asm/paravirt_types.h > +++ b/arch/x86/include/asm/paravirt_types.h > @@ -211,6 +211,12 @@ struct pv_mmu_ops { > void (*flush_tlb_user)(void); > void (*flush_tlb_kernel)(void); > void (*flush_tlb_one_user)(unsigned long addr); > + /* > + * flush_tlb_multi() is the preferred interface. When it is used, > + * flush_tlb_others() should return false.This comment does not make sense. flush_tlb_others() return type is void. Juergen
Peter Zijlstra
2019-May-27 09:47 UTC
[RFC PATCH 5/6] x86/mm/tlb: Flush remote and local TLBs concurrently
On Sat, May 25, 2019 at 10:54:50AM +0200, Juergen Gross wrote:> On 25/05/2019 10:22, Nadav Amit wrote:> > diff --git a/arch/x86/include/asm/paravirt_types.h b/arch/x86/include/asm/paravirt_types.h > > index 946f8f1f1efc..3a156e63c57d 100644 > > --- a/arch/x86/include/asm/paravirt_types.h > > +++ b/arch/x86/include/asm/paravirt_types.h > > @@ -211,6 +211,12 @@ struct pv_mmu_ops { > > void (*flush_tlb_user)(void); > > void (*flush_tlb_kernel)(void); > > void (*flush_tlb_one_user)(unsigned long addr); > > + /* > > + * flush_tlb_multi() is the preferred interface. When it is used, > > + * flush_tlb_others() should return false. > > This comment does not make sense. flush_tlb_others() return type is > void.I suspect that is an artifact from before the static_key; an attempt to make the pv interface less awkward. Something like the below would work for KVM I suspect, the others (Hyper-V and Xen are more 'interesting'). --- --- a/arch/x86/kernel/kvm.c +++ b/arch/x86/kernel/kvm.c @@ -580,7 +580,7 @@ static void __init kvm_apf_trap_init(voi static DEFINE_PER_CPU(cpumask_var_t, __pv_tlb_mask); -static void kvm_flush_tlb_others(const struct cpumask *cpumask, +static void kvm_flush_tlb_multi(const struct cpumask *cpumask, const struct flush_tlb_info *info) { u8 state; @@ -594,6 +594,9 @@ static void kvm_flush_tlb_others(const s * queue flush_on_enter for pre-empted vCPUs */ for_each_cpu(cpu, flushmask) { + if (cpu == smp_processor_id()) + continue; + src = &per_cpu(steal_time, cpu); state = READ_ONCE(src->preempted); if ((state & KVM_VCPU_PREEMPTED)) { @@ -603,7 +606,7 @@ static void kvm_flush_tlb_others(const s } } - native_flush_tlb_others(flushmask, info); + native_flush_tlb_multi(flushmask, info); } static void __init kvm_guest_init(void) @@ -628,9 +631,8 @@ static void __init kvm_guest_init(void) if (kvm_para_has_feature(KVM_FEATURE_PV_TLB_FLUSH) && !kvm_para_has_hint(KVM_HINTS_REALTIME) && kvm_para_has_feature(KVM_FEATURE_STEAL_TIME)) { - pv_ops.mmu.flush_tlb_others = kvm_flush_tlb_others; + pv_ops.mmu.flush_tlb_multi = kvm_flush_tlb_multi; pv_ops.mmu.tlb_remove_table = tlb_remove_table; - static_key_disable(&flush_tlb_multi_enabled.key); } if (kvm_para_has_feature(KVM_FEATURE_PV_EOI))
Possibly Parallel Threads
- [RFC PATCH 5/6] x86/mm/tlb: Flush remote and local TLBs concurrently
- [RFC PATCH 5/6] x86/mm/tlb: Flush remote and local TLBs concurrently
- [RFC PATCH 5/6] x86/mm/tlb: Flush remote and local TLBs concurrently
- [RFC PATCH 5/6] x86/mm/tlb: Flush remote and local TLBs concurrently
- [RFC PATCH 5/6] x86/mm/tlb: Flush remote and local TLBs concurrently