KVM: do not shrink halt_poll_ns below grow_start

grow_halt_poll_ns() ignores values between 0 and
halt_poll_ns_grow_start (10000 by default). However,
when we shrink halt_poll_ns we may fall way below
halt_poll_ns_grow_start and endup with halt_poll_ns
values that don't make a lot of sense: like 1 or 9,
or 19.

VCPU1 trace (halt_poll_ns_shrink equals 2):

VCPU1 grow 10000
VCPU1 shrink 5000
VCPU1 shrink 2500
VCPU1 shrink 1250
VCPU1 shrink 625
VCPU1 shrink 312
VCPU1 shrink 156
VCPU1 shrink 78
VCPU1 shrink 39
VCPU1 shrink 19
VCPU1 shrink 9
VCPU1 shrink 4

Mirror what grow_halt_poll_ns() does and set halt_poll_ns
to 0 as soon as new shrink-ed halt_poll_ns value falls
below halt_poll_ns_grow_start.

Signed-off-by: Sergey Senozhatsky <senozhatsky@chromium.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20210902031100.252080-1-senozhatsky@chromium.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 439d3b4cd1a941ea8aca486ccc7e6c2b50241e48..8495a01d1e41e469ec3bfe1ca5427f5565828422 100644 (file)
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -3134,15 +3134,19 @@ out:
 
 static void shrink_halt_poll_ns(struct kvm_vcpu *vcpu)
 {
-       unsigned int old, val, shrink;
+       unsigned int old, val, shrink, grow_start;
 
        old = val = vcpu->halt_poll_ns;
        shrink = READ_ONCE(halt_poll_ns_shrink);
+       grow_start = READ_ONCE(halt_poll_ns_grow_start);
        if (shrink == 0)
                val = 0;
        else
                val /= shrink;
 
+       if (val < grow_start)
+               val = 0;
+
        vcpu->halt_poll_ns = val;
        trace_kvm_halt_poll_ns_shrink(vcpu->vcpu_id, val, old);
 }