xsave_helper: pull xsave and xrstor out of kvm.c into helper function

[qemu.git] / numa.c

diff --git a/numa.c b/numa.c
index 4c4f7f572ef8609d37b52343f48c61173c38d553..b0e75f62685a2984f8e313bbf2bf8e3b02b94cd3 100644 (file)
--- a/numa.c
+++ b/numa.c
@@ -25,13 +25,13 @@
 #include "qemu/osdep.h"
 #include "sysemu/numa.h"
 #include "exec/cpu-common.h"
 #include "qemu/osdep.h"
 #include "sysemu/numa.h"
 #include "exec/cpu-common.h"

+#include "exec/ramlist.h"

 #include "qemu/bitmap.h"
 #include "qom/cpu.h"
 #include "qemu/error-report.h"
 #include "include/exec/cpu-common.h" /* for RAM_ADDR_FMT */
 #include "qapi-visit.h"
 #include "qapi/opts-visitor.h"
 #include "qemu/bitmap.h"
 #include "qom/cpu.h"
 #include "qemu/error-report.h"
 #include "include/exec/cpu-common.h" /* for RAM_ADDR_FMT */
 #include "qapi-visit.h"
 #include "qapi/opts-visitor.h"

-#include "qapi/dealloc-visitor.h"

 #include "hw/boards.h"
 #include "sysemu/hostmem.h"
 #include "qmp-commands.h"
 #include "hw/boards.h"
 #include "sysemu/hostmem.h"
 #include "qmp-commands.h"


@@ -51,6 +51,7 @@ static int max_numa_nodeid; /* Highest specified NUMA node ID, plus one.
                              * For all nodes, nodeid < max_numa_nodeid
                              */
 int nb_numa_nodes;
                              * For all nodes, nodeid < max_numa_nodeid
                              */
 int nb_numa_nodes;

+bool have_numa_distance;

 NodeInfo numa_info[MAX_NODES];
 
 void numa_set_mem_node_id(ram_addr_t addr, uint64_t size, uint32_t node)
 NodeInfo numa_info[MAX_NODES];
 
 void numa_set_mem_node_id(ram_addr_t addr, uint64_t size, uint32_t node)


@@ -140,10 +141,12 @@ uint32_t numa_get_node(ram_addr_t addr, Error **errp)
     return -1;
 }
 
     return -1;
 }
 

-static void numa_node_parse(NumaNodeOptions *node, QemuOpts *opts, Error **errp)
+static void parse_numa_node(MachineState *ms, NumaNodeOptions *node,
+                            QemuOpts *opts, Error **errp)

 {
     uint16_t nodenr;
     uint16List *cpus = NULL;
 {
     uint16_t nodenr;
     uint16List *cpus = NULL;

+    MachineClass *mc = MACHINE_GET_CLASS(ms);

 
     if (node->has_nodeid) {
         nodenr = node->nodeid;
 
     if (node->has_nodeid) {
         nodenr = node->nodeid;


@@ -162,7 +165,12 @@ static void numa_node_parse(NumaNodeOptions *node, QemuOpts *opts, Error **errp)
         return;
     }
 
         return;
     }
 

+    if (!mc->cpu_index_to_instance_props) {
+        error_report("NUMA is not supported by this machine-type");
+        exit(1);
+    }

     for (cpus = node->cpus; cpus; cpus = cpus->next) {
     for (cpus = node->cpus; cpus; cpus = cpus->next) {

+        CpuInstanceProperties props;

         if (cpus->value >= max_cpus) {
             error_setg(errp,
                        "CPU index (%" PRIu16 ")"
         if (cpus->value >= max_cpus) {
             error_setg(errp,
                        "CPU index (%" PRIu16 ")"


@@ -170,11 +178,14 @@ static void numa_node_parse(NumaNodeOptions *node, QemuOpts *opts, Error **errp)
                        cpus->value, max_cpus);
             return;
         }
                        cpus->value, max_cpus);
             return;
         }

-        bitmap_set(numa_info[nodenr].node_cpu, cpus->value, 1);
+        props = mc->cpu_index_to_instance_props(ms, cpus->value);
+        props.node_id = nodenr;
+        props.has_node_id = true;
+        machine_set_cpu_numa_node(ms, &props, &error_fatal);

     }
 
     if (node->has_mem && node->has_memdev) {
     }
 
     if (node->has_mem && node->has_memdev) {

-        error_setg(errp, "qemu: cannot specify both mem= and memdev=");
+        error_setg(errp, "cannot specify both mem= and memdev=");

         return;
     }
 
         return;
     }
 


@@ -182,7 +193,7 @@ static void numa_node_parse(NumaNodeOptions *node, QemuOpts *opts, Error **errp)
         have_memdevs = node->has_memdev;
     }
     if (node->has_memdev != have_memdevs) {
         have_memdevs = node->has_memdev;
     }
     if (node->has_memdev != have_memdevs) {

-        error_setg(errp, "qemu: memdev option must be specified for either "
+        error_setg(errp, "memdev option must be specified for either "

                    "all or no nodes");
         return;
     }
                    "all or no nodes");
         return;
     }


@@ -205,105 +216,219 @@ static void numa_node_parse(NumaNodeOptions *node, QemuOpts *opts, Error **errp)
         }
 
         object_ref(o);
         }
 
         object_ref(o);

-        numa_info[nodenr].node_mem = object_property_get_int(o, "size", NULL);
+        numa_info[nodenr].node_mem = object_property_get_uint(o, "size", NULL);

         numa_info[nodenr].node_memdev = MEMORY_BACKEND(o);
     }
     numa_info[nodenr].present = true;
     max_numa_nodeid = MAX(max_numa_nodeid, nodenr + 1);
 }
 
         numa_info[nodenr].node_memdev = MEMORY_BACKEND(o);
     }
     numa_info[nodenr].present = true;
     max_numa_nodeid = MAX(max_numa_nodeid, nodenr + 1);
 }
 

+static void parse_numa_distance(NumaDistOptions *dist, Error **errp)
+{
+    uint16_t src = dist->src;
+    uint16_t dst = dist->dst;
+    uint8_t val = dist->val;
+
+    if (src >= MAX_NODES || dst >= MAX_NODES) {
+        error_setg(errp,
+                   "Invalid node %d, max possible could be %d",
+                   MAX(src, dst), MAX_NODES);
+        return;
+    }
+
+    if (!numa_info[src].present || !numa_info[dst].present) {
+        error_setg(errp, "Source/Destination NUMA node is missing. "
+                   "Please use '-numa node' option to declare it first.");
+        return;
+    }
+
+    if (val < NUMA_DISTANCE_MIN) {
+        error_setg(errp, "NUMA distance (%" PRIu8 ") is invalid, "
+                   "it shouldn't be less than %d.",
+                   val, NUMA_DISTANCE_MIN);
+        return;
+    }
+
+    if (src == dst && val != NUMA_DISTANCE_MIN) {
+        error_setg(errp, "Local distance of node %d should be %d.",
+                   src, NUMA_DISTANCE_MIN);
+        return;
+    }
+
+    numa_info[src].distance[dst] = val;
+    have_numa_distance = true;
+}
+

 static int parse_numa(void *opaque, QemuOpts *opts, Error **errp)
 {
     NumaOptions *object = NULL;
 static int parse_numa(void *opaque, QemuOpts *opts, Error **errp)
 {
     NumaOptions *object = NULL;

+    MachineState *ms = opaque;

     Error *err = NULL;
 
     {
     Error *err = NULL;
 
     {

-        OptsVisitor *ov = opts_visitor_new(opts);
-        visit_type_NumaOptions(opts_get_visitor(ov), NULL, &object, &err);
-        opts_visitor_cleanup(ov);
+        Visitor *v = opts_visitor_new(opts);
+        visit_type_NumaOptions(v, NULL, &object, &err);
+        visit_free(v);

     }
 
     if (err) {
     }
 
     if (err) {

-        goto error;
+        goto end;

     }
 
     switch (object->type) {
     }
 
     switch (object->type) {

-    case NUMA_OPTIONS_KIND_NODE:
-        numa_node_parse(object->u.node, opts, &err);
+    case NUMA_OPTIONS_TYPE_NODE:
+        parse_numa_node(ms, &object->u.node, opts, &err);

         if (err) {
         if (err) {

-            goto error;
+            goto end;

         }
         nb_numa_nodes++;
         break;
         }
         nb_numa_nodes++;
         break;

+    case NUMA_OPTIONS_TYPE_DIST:
+        parse_numa_distance(&object->u.dist, &err);
+        if (err) {
+            goto end;
+        }
+        break;
+    case NUMA_OPTIONS_TYPE_CPU:
+        if (!object->u.cpu.has_node_id) {
+            error_setg(&err, "Missing mandatory node-id property");
+            goto end;
+        }
+        if (!numa_info[object->u.cpu.node_id].present) {
+            error_setg(&err, "Invalid node-id=%" PRId64 ", NUMA node must be "
+                "defined with -numa node,nodeid=ID before it's used with "
+                "-numa cpu,node-id=ID", object->u.cpu.node_id);
+            goto end;
+        }
+
+        machine_set_cpu_numa_node(ms, qapi_NumaCpuOptions_base(&object->u.cpu),
+                                  &err);
+        break;

     default:
         abort();
     }
 
     default:
         abort();
     }
 

+end:
+    qapi_free_NumaOptions(object);
+    if (err) {
+        error_report_err(err);
+        return -1;
+    }
+

     return 0;
     return 0;

+}

 
 

-error:
-    error_report_err(err);
+/* If all node pair distances are symmetric, then only distances
+ * in one direction are enough. If there is even one asymmetric
+ * pair, though, then all distances must be provided. The
+ * distance from a node to itself is always NUMA_DISTANCE_MIN,
+ * so providing it is never necessary.
+ */
+static void validate_numa_distance(void)
+{
+    int src, dst;
+    bool is_asymmetrical = false;
+
+    for (src = 0; src < nb_numa_nodes; src++) {
+        for (dst = src; dst < nb_numa_nodes; dst++) {
+            if (numa_info[src].distance[dst] == 0 &&
+                numa_info[dst].distance[src] == 0) {
+                if (src != dst) {
+                    error_report("The distance between node %d and %d is "
+                                 "missing, at least one distance value "
+                                 "between each nodes should be provided.",
+                                 src, dst);
+                    exit(EXIT_FAILURE);
+                }
+            }

 
 

-    if (object) {
-        QapiDeallocVisitor *dv = qapi_dealloc_visitor_new();
-        visit_type_NumaOptions(qapi_dealloc_get_visitor(dv), NULL, &object,
-                               NULL);
-        qapi_dealloc_visitor_cleanup(dv);
+            if (numa_info[src].distance[dst] != 0 &&
+                numa_info[dst].distance[src] != 0 &&
+                numa_info[src].distance[dst] !=
+                numa_info[dst].distance[src]) {
+                is_asymmetrical = true;
+            }
+        }

     }
 
     }
 

-    return -1;
+    if (is_asymmetrical) {
+        for (src = 0; src < nb_numa_nodes; src++) {
+            for (dst = 0; dst < nb_numa_nodes; dst++) {
+                if (src != dst && numa_info[src].distance[dst] == 0) {
+                    error_report("At least one asymmetrical pair of "
+                            "distances is given, please provide distances "
+                            "for both directions of all node pairs.");
+                    exit(EXIT_FAILURE);
+                }
+            }
+        }
+    }

 }
 
 }
 

-static char *enumerate_cpus(unsigned long *cpus, int max_cpus)
+static void complete_init_numa_distance(void)

 {
 {

-    int cpu;
-    bool first = true;
-    GString *s = g_string_new(NULL);
+    int src, dst;

 
 

-    for (cpu = find_first_bit(cpus, max_cpus);
-        cpu < max_cpus;
-        cpu = find_next_bit(cpus, max_cpus, cpu + 1)) {
-        g_string_append_printf(s, "%s%d", first ? "" : " ", cpu);
-        first = false;
+    /* Fixup NUMA distance by symmetric policy because if it is an
+     * asymmetric distance table, it should be a complete table and
+     * there would not be any missing distance except local node, which
+     * is verified by validate_numa_distance above.
+     */
+    for (src = 0; src < nb_numa_nodes; src++) {
+        for (dst = 0; dst < nb_numa_nodes; dst++) {
+            if (numa_info[src].distance[dst] == 0) {
+                if (src == dst) {
+                    numa_info[src].distance[dst] = NUMA_DISTANCE_MIN;
+                } else {
+                    numa_info[src].distance[dst] = numa_info[dst].distance[src];
+                }
+            }
+        }

     }
     }

-    return g_string_free(s, FALSE);

 }
 
 }
 

-static void validate_numa_cpus(void)
+void numa_legacy_auto_assign_ram(MachineClass *mc, NodeInfo *nodes,
+                                 int nb_nodes, ram_addr_t size)

 {
     int i;
 {
     int i;

-    DECLARE_BITMAP(seen_cpus, MAX_CPUMASK_BITS);
+    uint64_t usedmem = 0;

 
 

-    bitmap_zero(seen_cpus, MAX_CPUMASK_BITS);
-    for (i = 0; i < nb_numa_nodes; i++) {
-        if (bitmap_intersects(seen_cpus, numa_info[i].node_cpu,
-                              MAX_CPUMASK_BITS)) {
-            bitmap_and(seen_cpus, seen_cpus,
-                       numa_info[i].node_cpu, MAX_CPUMASK_BITS);
-            error_report("CPU(s) present in multiple NUMA nodes: %s",
-                         enumerate_cpus(seen_cpus, max_cpus));
-            exit(EXIT_FAILURE);
-        }
-        bitmap_or(seen_cpus, seen_cpus,
-                  numa_info[i].node_cpu, MAX_CPUMASK_BITS);
-    }
+    /* Align each node according to the alignment
+     * requirements of the machine class
+     */

 
 

-    if (!bitmap_full(seen_cpus, max_cpus)) {
-        char *msg;
-        bitmap_complement(seen_cpus, seen_cpus, max_cpus);
-        msg = enumerate_cpus(seen_cpus, max_cpus);
-        error_report("warning: CPU(s) not present in any NUMA nodes: %s", msg);
-        error_report("warning: All CPU(s) up to maxcpus should be described "
-                     "in NUMA config");
-        g_free(msg);
+    for (i = 0; i < nb_nodes - 1; i++) {
+        nodes[i].node_mem = (size / nb_nodes) &
+                            ~((1 << mc->numa_mem_align_shift) - 1);
+        usedmem += nodes[i].node_mem;

     }
     }

+    nodes[i].node_mem = size - usedmem;

 }
 
 }
 

-void parse_numa_opts(MachineClass *mc)
+void numa_default_auto_assign_ram(MachineClass *mc, NodeInfo *nodes,
+                                  int nb_nodes, ram_addr_t size)

 {
     int i;
 {
     int i;

+    uint64_t usedmem = 0, node_mem;
+    uint64_t granularity = size / nb_nodes;
+    uint64_t propagate = 0;
+
+    for (i = 0; i < nb_nodes - 1; i++) {
+        node_mem = (granularity + propagate) &
+                   ~((1 << mc->numa_mem_align_shift) - 1);
+        propagate = granularity + propagate - node_mem;
+        nodes[i].node_mem = node_mem;
+        usedmem += node_mem;
+    }
+    nodes[i].node_mem = size - usedmem;
+}

 
 

-    if (qemu_opts_foreach(qemu_find_opts("numa"), parse_numa, NULL, NULL)) {
+void parse_numa_opts(MachineState *ms)
+{
+    int i;
+    MachineClass *mc = MACHINE_GET_CLASS(ms);
+
+    if (qemu_opts_foreach(qemu_find_opts("numa"), parse_numa, ms, NULL)) {

         exit(1);
     }
 
         exit(1);
     }
 


@@ -337,17 +462,8 @@ void parse_numa_opts(MachineClass *mc)
             }
         }
         if (i == nb_numa_nodes) {
             }
         }
         if (i == nb_numa_nodes) {

-            uint64_t usedmem = 0;
-
-            /* On Linux, each node's border has to be 8MB aligned,
-             * the final node gets the rest.
-             */
-            for (i = 0; i < nb_numa_nodes - 1; i++) {
-                numa_info[i].node_mem = (ram_size / nb_numa_nodes) &
-                                        ~((1 << 23UL) - 1);
-                usedmem += numa_info[i].node_mem;
-            }
-            numa_info[i].node_mem = ram_size - usedmem;
+            assert(mc->numa_auto_assign_ram);
+            mc->numa_auto_assign_ram(mc, numa_info, nb_numa_nodes, ram_size);

         }
 
         numa_total = 0;
         }
 
         numa_total = 0;


@@ -367,46 +483,44 @@ void parse_numa_opts(MachineClass *mc)
 
         numa_set_mem_ranges();
 
 
         numa_set_mem_ranges();
 

-        for (i = 0; i < nb_numa_nodes; i++) {
-            if (!bitmap_empty(numa_info[i].node_cpu, MAX_CPUMASK_BITS)) {
-                break;
-            }
-        }
-        /* Historically VCPUs were assigned in round-robin order to NUMA
-         * nodes. However it causes issues with guest not handling it nice
-         * in case where cores/threads from a multicore CPU appear on
-         * different nodes. So allow boards to override default distribution
-         * rule grouping VCPUs by socket so that VCPUs from the same socket
-         * would be on the same node.
+        /* QEMU needs at least all unique node pair distances to build
+         * the whole NUMA distance table. QEMU treats the distance table
+         * as symmetric by default, i.e. distance A->B == distance B->A.
+         * Thus, QEMU is able to complete the distance table
+         * initialization even though only distance A->B is provided and
+         * distance B->A is not. QEMU knows the distance of a node to
+         * itself is always 10, so A->A distances may be omitted. When
+         * the distances of two nodes of a pair differ, i.e. distance
+         * A->B != distance B->A, then that means the distance table is
+         * asymmetric. In this case, the distances for both directions
+         * of all node pairs are required.

          */
          */

-        if (i == nb_numa_nodes) {
-            for (i = 0; i < max_cpus; i++) {
-                unsigned node_id = i % nb_numa_nodes;
-                if (mc->cpu_index_to_socket_id) {
-                    node_id = mc->cpu_index_to_socket_id(i) % nb_numa_nodes;
-                }
+        if (have_numa_distance) {
+            /* Validate enough NUMA distance information was provided. */
+            validate_numa_distance();

 
 

-                set_bit(i, numa_info[node_id].node_cpu);
-            }
+            /* Validation succeeded, now fill in any missing distances. */
+            complete_init_numa_distance();

         }
         }

-
-        validate_numa_cpus();

     } else {
         numa_set_mem_node_id(0, ram_size, 0);
     }
 }
 
     } else {
         numa_set_mem_node_id(0, ram_size, 0);
     }
 }
 

-void numa_post_machine_init(void)
+void numa_cpu_pre_plug(const CPUArchId *slot, DeviceState *dev, Error **errp)

 {
 {

-    CPUState *cpu;
-    int i;
-
-    CPU_FOREACH(cpu) {
-        for (i = 0; i < nb_numa_nodes; i++) {
-            if (test_bit(cpu->cpu_index, numa_info[i].node_cpu)) {
-                cpu->numa_node = i;
-            }
+    int node_id = object_property_get_int(OBJECT(dev), "node-id", &error_abort);
+
+    if (node_id == CPU_UNSET_NUMA_NODE_ID) {
+        /* due to bug in libvirt, it doesn't pass node-id from props on
+         * device_add as expected, so we have to fix it up here */
+        if (slot->props.has_node_id) {
+            object_property_set_int(OBJECT(dev), slot->props.node_id,
+                                    "node-id", errp);

         }
         }

+    } else if (node_id != slot->props.node_id) {
+        error_setg(errp, "node-id=%d must match numa node specified "
+                   "with -numa option", node_id);

     }
 }
 
     }
 }
 


@@ -470,6 +584,7 @@ void memory_region_allocate_system_memory(MemoryRegion *mr, Object *owner,
             exit(1);
         }
 
             exit(1);
         }
 

+        host_memory_backend_set_mapped(backend, true);

         memory_region_add_subregion(mr, addr, seg);
         vmstate_register_ram_global(seg);
         addr += size;
         memory_region_add_subregion(mr, addr, seg);
         vmstate_register_ram_global(seg);
         addr += size;


@@ -489,7 +604,7 @@ static void numa_stat_memory_devices(uint64_t node_mem[])
         if (value) {
             switch (value->type) {
             case MEMORY_DEVICE_INFO_KIND_DIMM:
         if (value) {
             switch (value->type) {
             case MEMORY_DEVICE_INFO_KIND_DIMM:

-                node_mem[value->u.dimm->node] += value->u.dimm->size;
+                node_mem[value->u.dimm.data->node] += value->u.dimm.data->size;

                 break;
             default:
                 break;
                 break;
             default:
                 break;


@@ -523,8 +638,11 @@ static int query_memdev(Object *obj, void *opaque)
 
         m->value = g_malloc0(sizeof(*m->value));
 
 
         m->value = g_malloc0(sizeof(*m->value));
 

-        m->value->size = object_property_get_int(obj, "size",
-                                                 &error_abort);
+        m->value->id = object_property_get_str(obj, "id", NULL);
+        m->value->has_id = !!m->value->id;
+
+        m->value->size = object_property_get_uint(obj, "size",
+                                                  &error_abort);

         m->value->merge = object_property_get_bool(obj, "merge",
                                                    &error_abort);
         m->value->dump = object_property_get_bool(obj, "dump",
         m->value->merge = object_property_get_bool(obj, "merge",
                                                    &error_abort);
         m->value->dump = object_property_get_bool(obj, "dump",


@@ -555,3 +673,31 @@ MemdevList *qmp_query_memdev(Error **errp)
     object_child_foreach(obj, query_memdev, &list);
     return list;
 }
     object_child_foreach(obj, query_memdev, &list);
     return list;
 }

+
+void ram_block_notifier_add(RAMBlockNotifier *n)
+{
+    QLIST_INSERT_HEAD(&ram_list.ramblock_notifiers, n, next);
+}
+
+void ram_block_notifier_remove(RAMBlockNotifier *n)
+{
+    QLIST_REMOVE(n, next);
+}
+
+void ram_block_notify_add(void *host, size_t size)
+{
+    RAMBlockNotifier *notifier;
+
+    QLIST_FOREACH(notifier, &ram_list.ramblock_notifiers, next) {
+        notifier->ram_block_added(notifier, host, size);
+    }
+}
+
+void ram_block_notify_remove(void *host, size_t size)
+{
+    RAMBlockNotifier *notifier;
+
+    QLIST_FOREACH(notifier, &ram_list.ramblock_notifiers, next) {
+        notifier->ram_block_removed(notifier, host, size);
+    }
+}