From 2ea5cb0a47c9dbf78235de30a0d5846a4ab6d995 Mon Sep 17 00:00:00 2001
From: Peter Xu <peterx@redhat.com>
Date: Thu, 6 May 2021 12:05:47 -0400
Subject: [PATCH] KVM: Add dirty-ring-size property

Add a parameter for dirty gfn count for dirty rings.  If zero, dirty ring is
disabled.  Otherwise dirty ring will be enabled with the per-vcpu gfn count as
specified.  If dirty ring cannot be enabled due to unsupported kernel or
illegal parameter, it'll fallback to dirty logging.

By default, dirty ring is not enabled (dirty-gfn-count default to 0).

Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <20210506160549.130416-9-peterx@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 accel/kvm/kvm-all.c | 46 +++++++++++++++++++++++++++++++++++++++++++++
 qemu-options.hx     | 12 ++++++++++++
 2 files changed, 58 insertions(+)

diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c
index df9fbf59a6..5afe15ae66 100644
--- a/accel/kvm/kvm-all.c
+++ b/accel/kvm/kvm-all.c
@@ -126,6 +126,8 @@ struct KVMState
         KVMMemoryListener *ml;
         AddressSpace *as;
     } *as;
+    uint64_t kvm_dirty_ring_bytes;  /* Size of the per-vcpu dirty ring */
+    uint32_t kvm_dirty_ring_size;   /* Number of dirty GFNs per ring */
 };
 
 KVMState *kvm_state;
@@ -3182,6 +3184,42 @@ bool kvm_kernel_irqchip_split(void)
     return kvm_state->kernel_irqchip_split == ON_OFF_AUTO_ON;
 }
 
+static void kvm_get_dirty_ring_size(Object *obj, Visitor *v,
+                                    const char *name, void *opaque,
+                                    Error **errp)
+{
+    KVMState *s = KVM_STATE(obj);
+    uint32_t value = s->kvm_dirty_ring_size;
+
+    visit_type_uint32(v, name, &value, errp);
+}
+
+static void kvm_set_dirty_ring_size(Object *obj, Visitor *v,
+                                    const char *name, void *opaque,
+                                    Error **errp)
+{
+    KVMState *s = KVM_STATE(obj);
+    Error *error = NULL;
+    uint32_t value;
+
+    if (s->fd != -1) {
+        error_setg(errp, "Cannot set properties after the accelerator has been initialized");
+        return;
+    }
+
+    visit_type_uint32(v, name, &value, &error);
+    if (error) {
+        error_propagate(errp, error);
+        return;
+    }
+    if (value & (value - 1)) {
+        error_setg(errp, "dirty-ring-size must be a power of two.");
+        return;
+    }
+
+    s->kvm_dirty_ring_size = value;
+}
+
 static void kvm_accel_instance_init(Object *obj)
 {
     KVMState *s = KVM_STATE(obj);
@@ -3191,6 +3229,8 @@ static void kvm_accel_instance_init(Object *obj)
     s->kvm_shadow_mem = -1;
     s->kernel_irqchip_allowed = true;
     s->kernel_irqchip_split = ON_OFF_AUTO_AUTO;
+    /* KVM dirty ring is by default off */
+    s->kvm_dirty_ring_size = 0;
 }
 
 static void kvm_accel_class_init(ObjectClass *oc, void *data)
@@ -3212,6 +3252,12 @@ static void kvm_accel_class_init(ObjectClass *oc, void *data)
         NULL, NULL);
     object_class_property_set_description(oc, "kvm-shadow-mem",
         "KVM shadow MMU size");
+
+    object_class_property_add(oc, "dirty-ring-size", "uint32",
+        kvm_get_dirty_ring_size, kvm_set_dirty_ring_size,
+        NULL, NULL);
+    object_class_property_set_description(oc, "dirty-ring-size",
+        "Size of KVM dirty page ring buffer (default: 0, i.e. use bitmap)");
 }
 
 static const TypeInfo kvm_accel_type = {
diff --git a/qemu-options.hx b/qemu-options.hx
index e22fb94d99..ecdb064409 100644
--- a/qemu-options.hx
+++ b/qemu-options.hx
@@ -141,6 +141,7 @@ DEF("accel", HAS_ARG, QEMU_OPTION_accel,
     "                kvm-shadow-mem=size of KVM shadow MMU in bytes\n"
     "                split-wx=on|off (enable TCG split w^x mapping)\n"
     "                tb-size=n (TCG translation block cache size)\n"
+    "                dirty-ring-size=n (KVM dirty ring GFN count, default 0)\n"
     "                thread=single|multi (enable multi-threaded TCG)\n", QEMU_ARCH_ALL)
 SRST
 ``-accel name[,prop=value[,...]]``
@@ -181,6 +182,17 @@ SRST
         where both the back-end and front-ends support it and no
         incompatible TCG features have been enabled (e.g.
         icount/replay).
+
+    ``dirty-ring-size=n``
+        When the KVM accelerator is used, it controls the size of the per-vCPU
+        dirty page ring buffer (number of entries for each vCPU). It should
+        be a value that is power of two, and it should be 1024 or bigger (but
+        still less than the maximum value that the kernel supports).  4096
+        could be a good initial value if you have no idea which is the best.
+        Set this value to 0 to disable the feature.  By default, this feature
+        is disabled (dirty-ring-size=0).  When enabled, KVM will instead
+        record dirty pages in a bitmap.
+
 ERST
 
 DEF("smp", HAS_ARG, QEMU_OPTION_smp,