| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Message-ID: <BLU436-SMTP81A522DCE2D99125FBFCAC80690@phx.gbl>
Date: Wed, 2 Sep 2015 14:21:00 +0800
From: Wanpeng Li <wanpeng.li@...mail.com>
To: Paolo Bonzini <pbonzini@...hat.com>
CC: David Matlack <dmatlack@...gle.com>, kvm@...r.kernel.org,
linux-kernel@...r.kernel.org, Wanpeng Li <wanpeng.li@...mail.com>
Subject: [PATCH v5 2/3] KVM: dynamic halt_poll_ns adjustment
There is a downside of always-poll since poll is still happened for idle
vCPUs which can waste cpu usage. This patch adds the ability to adjust
halt_poll_ns dynamically, to grow halt_poll_ns when shot halt is detected,
and to shrink halt_poll_ns when long halt is detected.
There are two new kernel parameters for changing the halt_poll_ns:
halt_poll_ns_grow and halt_poll_ns_shrink.
no-poll always-poll dynamic-poll
-----------------------------------------------------------------------
Idle (nohz) vCPU %c0 0.15% 0.3% 0.2%
Idle (250HZ) vCPU %c0 1.1% 4.6%~14% 1.2%
TCP_RR latency 34us 27us 26.7us
"Idle (X) vCPU %c0" is the percent of time the physical cpu spent in
c0 over 60 seconds (each vCPU is pinned to a pCPU). (nohz) means the
guest was tickless. (250HZ) means the guest was ticking at 250HZ.
The big win is with ticking operating systems. Running the linux guest
with nohz=off (and HZ=250), we save 3.4%~12.8% CPUs/second and get close
to no-polling overhead levels by using the dynamic-poll. The savings
should be even higher for higher frequency ticks.
Suggested-by: David Matlack <dmatlack@...gle.com>
Signed-off-by: Wanpeng Li <wanpeng.li@...mail.com>
---
virt/kvm/kvm_main.c | 60 ++++++++++++++++++++++++++++++++++++++++++++++++++---
1 file changed, 57 insertions(+), 3 deletions(-)
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index c06e57c..2206cb0 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -66,9 +66,18 @@
MODULE_AUTHOR("Qumranet");
MODULE_LICENSE("GPL");
-static unsigned int halt_poll_ns;
+/* halt polling only reduces halt latency by 5-7 us, 500us is enough */
+static unsigned int halt_poll_ns = 500000;
module_param(halt_poll_ns, uint, S_IRUGO | S_IWUSR);
+/* Default doubles per-vcpu halt_poll_ns. */
+static unsigned int halt_poll_ns_grow = 2;
+module_param(halt_poll_ns_grow, int, S_IRUGO);
+
+/* Default resets per-vcpu halt_poll_ns . */
+static unsigned int halt_poll_ns_shrink;
+module_param(halt_poll_ns_shrink, int, S_IRUGO);
+
/*
* Ordering of locks:
*
@@ -1907,6 +1916,31 @@ void kvm_vcpu_mark_page_dirty(struct kvm_vcpu *vcpu, gfn_t gfn)
}
EXPORT_SYMBOL_GPL(kvm_vcpu_mark_page_dirty);
+static void grow_halt_poll_ns(struct kvm_vcpu *vcpu)
+{
+ int val = vcpu->halt_poll_ns;
+
+ /* 10us base */
+ if (val == 0 && halt_poll_ns_grow)
+ val = 10000;
+ else
+ val *= halt_poll_ns_grow;
+
+ vcpu->halt_poll_ns = val;
+}
+
+static void shrink_halt_poll_ns(struct kvm_vcpu *vcpu)
+{
+ int val = vcpu->halt_poll_ns;
+
+ if (halt_poll_ns_shrink == 0)
+ val = 0;
+ else
+ val /= halt_poll_ns_shrink;
+
+ vcpu->halt_poll_ns = val;
+}
+
static int kvm_vcpu_check_block(struct kvm_vcpu *vcpu)
{
if (kvm_arch_vcpu_runnable(vcpu)) {
@@ -1929,6 +1963,7 @@ void kvm_vcpu_block(struct kvm_vcpu *vcpu)
ktime_t start, cur;
DEFINE_WAIT(wait);
bool waited = false;
+ u64 poll_ns = 0, wait_ns = 0, block_ns = 0;
start = cur = ktime_get();
if (vcpu->halt_poll_ns) {
@@ -1941,12 +1976,17 @@ void kvm_vcpu_block(struct kvm_vcpu *vcpu)
*/
if (kvm_vcpu_check_block(vcpu) < 0) {
++vcpu->stat.halt_successful_poll;
- goto out;
+ break;
}
cur = ktime_get();
} while (single_task_running() && ktime_before(cur, stop));
}
+ if (ktime_after(cur, start)) {
+ poll_ns = ktime_to_ns(cur) - ktime_to_ns(start);
+ goto out;
+ }
+
for (;;) {
prepare_to_wait(&vcpu->wq, &wait, TASK_INTERRUPTIBLE);
@@ -1959,9 +1999,23 @@ void kvm_vcpu_block(struct kvm_vcpu *vcpu)
finish_wait(&vcpu->wq, &wait);
cur = ktime_get();
+ wait_ns = ktime_to_ns(cur) - ktime_to_ns(start);
out:
- trace_kvm_vcpu_wakeup(ktime_to_ns(cur) - ktime_to_ns(start), waited);
+ block_ns = poll_ns + wait_ns;
+
+ if (halt_poll_ns) {
+ if (block_ns <= vcpu->halt_poll_ns)
+ ;
+ /* we had a long block, shrink polling */
+ else if (vcpu->halt_poll_ns && block_ns > halt_poll_ns)
+ shrink_halt_poll_ns(vcpu);
+ /* we had a short halt and our poll time is too small */
+ else if (vcpu->halt_poll_ns < halt_poll_ns && block_ns < halt_poll_ns)
+ grow_halt_poll_ns(vcpu);
+ }
+
+ trace_kvm_vcpu_wakeup(block_ns, waited);
}
EXPORT_SYMBOL_GPL(kvm_vcpu_block);
--
1.9.1
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists