QEMU-Devel Archive on lore.kernel.org
 help / color / Atom feed
From: Peter Xu <peterx@redhat.com>
To: qemu-devel@nongnu.org
Cc: Paolo Bonzini <pbonzini@redhat.com>,
	"Dr . David Alan Gilbert" <dgilbert@redhat.com>,
	peterx@redhat.com
Subject: [PATCH RFC v3 10/11] KVM: Add dirty-gfn-count property
Date: Sat, 23 May 2020 19:20:34 -0400
Message-ID: <20200523232035.1029349-11-peterx@redhat.com> (raw)
In-Reply-To: <20200523232035.1029349-1-peterx@redhat.com>

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>
---
 accel/kvm/kvm-all.c | 72 +++++++++++++++++++++++++++++++++++++++++++++
 qemu-options.hx     |  5 ++++
 2 files changed, 77 insertions(+)

diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c
index b9aaa7912c..dd017d0720 100644
--- a/accel/kvm/kvm-all.c
+++ b/accel/kvm/kvm-all.c
@@ -128,6 +128,9 @@ struct KVMState
         KVMMemoryListener *ml;
         AddressSpace *as;
     } *as;
+    bool kvm_dirty_ring_enabled;    /* Whether KVM dirty ring is enabled */
+    uint64_t kvm_dirty_ring_size;   /* Size of the per-vcpu dirty ring */
+    uint32_t kvm_dirty_gfn_count;   /* Number of dirty GFNs per ring */
 };
 
 KVMState *kvm_state;
@@ -2129,6 +2132,40 @@ static int kvm_init(MachineState *ms)
     s->memory_listener.listener.coalesced_io_add = kvm_coalesce_mmio_region;
     s->memory_listener.listener.coalesced_io_del = kvm_uncoalesce_mmio_region;
 
+    /*
+     * Enable KVM dirty ring if supported, otherwise fall back to
+     * dirty logging mode
+     */
+    if (s->kvm_dirty_gfn_count > 0) {
+        uint64_t ring_size;
+
+        ring_size = s->kvm_dirty_gfn_count * sizeof(struct kvm_dirty_gfn);
+
+        /* Read the max supported pages */
+        ret = kvm_vm_check_extension(kvm_state, KVM_CAP_DIRTY_LOG_RING);
+        if (ret > 0) {
+            if (ring_size > ret) {
+                error_report("KVM dirty GFN count %" PRIu32 " too big "
+                             "(maximum is %ld).  Please use a smaller value.",
+                             s->kvm_dirty_gfn_count,
+                             ret / sizeof(struct kvm_dirty_gfn));
+                ret = -EINVAL;
+                goto err;
+            }
+
+            ret = kvm_vm_enable_cap(s, KVM_CAP_DIRTY_LOG_RING, 0, ring_size);
+            if (ret) {
+                error_report("Enabling of KVM dirty ring failed: %d. "
+                             "Suggested mininum value is 1024. "
+                             "Please also make sure it's a power of two.", ret);
+                goto err;
+            }
+
+            s->kvm_dirty_ring_size = ring_size;
+            s->kvm_dirty_ring_enabled = true;
+        }
+    }
+
     kvm_memory_listener_register(s, &s->memory_listener,
                                  &address_space_memory, 0);
     memory_listener_register(&kvm_io_listener,
@@ -3089,6 +3126,33 @@ bool kvm_kernel_irqchip_split(void)
     return kvm_state->kernel_irqchip_split == ON_OFF_AUTO_ON;
 }
 
+static void kvm_get_dirty_gfn_count(Object *obj, Visitor *v,
+                                    const char *name, void *opaque,
+                                    Error **errp)
+{
+    KVMState *s = KVM_STATE(obj);
+    uint32_t value = s->kvm_dirty_gfn_count;
+
+    visit_type_uint32(v, name, &value, errp);
+}
+
+static void kvm_set_dirty_gfn_count(Object *obj, Visitor *v,
+                                    const char *name, void *opaque,
+                                    Error **errp)
+{
+    KVMState *s = KVM_STATE(obj);
+    Error *error = NULL;
+    uint32_t value;
+
+    visit_type_uint32(v, name, &value, &error);
+    if (error) {
+        error_propagate(errp, error);
+        return;
+    }
+
+    s->kvm_dirty_gfn_count = value;
+}
+
 static void kvm_accel_instance_init(Object *obj)
 {
     KVMState *s = KVM_STATE(obj);
@@ -3096,6 +3160,8 @@ 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_gfn_count = 0;
 }
 
 static void kvm_accel_class_init(ObjectClass *oc, void *data)
@@ -3117,6 +3183,12 @@ 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-gfn-count", "uint32",
+        kvm_get_dirty_gfn_count, kvm_set_dirty_gfn_count,
+        NULL, NULL);
+    object_class_property_set_description(oc, "dirty-gfn-count",
+        "KVM dirty GFN count (=0 to disable dirty ring)");
 }
 
 static const TypeInfo kvm_accel_type = {
diff --git a/qemu-options.hx b/qemu-options.hx
index 93bde2bbc8..b59d47473e 100644
--- a/qemu-options.hx
+++ b/qemu-options.hx
@@ -124,6 +124,7 @@ DEF("accel", HAS_ARG, QEMU_OPTION_accel,
     "                kernel-irqchip=on|off|split controls accelerated irqchip support (default=on)\n"
     "                kvm-shadow-mem=size of KVM shadow MMU in bytes\n"
     "                tb-size=n (TCG translation block cache size)\n"
+    "                dirty-gfn-count=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[,...]]``
@@ -158,6 +159,10 @@ SRST
         where both the back-end and front-ends support it and no
         incompatible TCG features have been enabled (e.g.
         icount/replay).
+
+    ``dirty-gfn-count=n``
+        Controls the per-vcpu KVM dirty ring GFN count (=0 to disable).
+
 ERST
 
 DEF("smp", HAS_ARG, QEMU_OPTION_smp,
-- 
2.26.2



  parent reply index

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-05-23 23:20 [PATCH RFC v3 00/11] KVM: Dirty ring support (QEMU part) Peter Xu
2020-05-23 23:20 ` [PATCH RFC v3 01/11] linux-headers: Update Peter Xu
2020-05-24 13:27   ` Peter Maydell
2020-05-24 14:06     ` Peter Xu
2020-05-24 17:50       ` Peter Maydell
2020-05-25 14:29         ` Peter Xu
2020-05-23 23:20 ` [PATCH RFC v3 02/11] memory: Introduce log_sync_global() to memory listener Peter Xu
2020-05-23 23:20 ` [PATCH RFC v3 03/11] KVM: Fixup kvm_log_clear_one_slot() ioctl return check Peter Xu
2020-05-24 16:39   ` Philippe Mathieu-Daudé
2020-05-23 23:20 ` [PATCH RFC v3 04/11] KVM: Use a big lock to replace per-kml slots_lock Peter Xu
2020-05-23 23:20 ` [PATCH RFC v3 05/11] KVM: Create the KVMSlot dirty bitmap on flag changes Peter Xu
2020-05-23 23:20 ` [PATCH RFC v3 06/11] KVM: Provide helper to get kvm dirty log Peter Xu
2020-05-23 23:20 ` [PATCH RFC v3 07/11] KVM: Provide helper to sync dirty bitmap from slot to ramblock Peter Xu
2020-05-23 23:20 ` [PATCH RFC v3 08/11] KVM: Simplify dirty log sync in kvm_set_phys_mem Peter Xu
2020-05-23 23:20 ` [PATCH RFC v3 09/11] KVM: Cache kvm slot dirty bitmap size Peter Xu
2020-05-23 23:20 ` Peter Xu [this message]
2020-05-23 23:20 ` [PATCH RFC v3 11/11] KVM: Dirty ring support Peter Xu
2020-05-24 13:06 ` [PATCH RFC v3 00/11] KVM: Dirty ring support (QEMU part) Peter Xu
2020-05-26 14:17 ` Peter Xu

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20200523232035.1029349-11-peterx@redhat.com \
    --to=peterx@redhat.com \
    --cc=dgilbert@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=qemu-devel@nongnu.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link

QEMU-Devel Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/qemu-devel/0 qemu-devel/git/0.git
	git clone --mirror https://lore.kernel.org/qemu-devel/1 qemu-devel/git/1.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 qemu-devel qemu-devel/ https://lore.kernel.org/qemu-devel \
		qemu-devel@nongnu.org
	public-inbox-index qemu-devel

Example config snippet for mirrors

Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.nongnu.qemu-devel


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git