Skip to contentSkip to conversation

numa: move numa global variables into MachineState

PATCH Superseded Tao Xu 1 participant 4 messages Apr 23, 2019

v3 of 20 Superseded Latest is v20 →

lore
Share & tools
lore

Tao Xu

author
#0 lore raw

v3 cover numa: move numa global variables into MachineState

The aim of this patch is to add struct NumaState in MachineState and move existing numa global nb_numa_nodes into NumaState. And add variable numa_support into MachineClass to decide which submachines support NUMA.

---
Changes in v3 -> v2:
- rename the "NumaState::nb_numa_nodes" as "NumaState::num_nodes", and rename the "NumaState::numa_info" as "NumaState::nodes" (Eduardo)
- use machine_num_numa_nodes(MachineState *ms) to check if ms->numa_state is NULL before using NumaState::num_nodes (Eduardo)
- check if ms->numa_state == NULL in the set_numa_options to avoid using -numa in a machine-type which don't support numa

Changes in v2:
- fix the mistake in numa_complete_configuration in numa.c
- add MachineState into some functions to avoid using qdev_get_machine
- add some if experssion to avoid the NumaState is null

Tao Xu (3):
numa: move numa global variable nb_numa_nodes into MachineState
numa: move numa global variable have_numa_distance into MachineState
numa: move numa global variable numa_info into MachineState

 exec.c                              |  5 +-
 hw/acpi/aml-build.c                 |  9 ++--
 hw/arm/boot.c                       |  4 +-
 hw/arm/virt-acpi-build.c            | 17 ++++---
 hw/arm/virt.c                       |  8 +++-
 hw/core/machine.c                   | 21 ++++++--
 hw/i386/acpi-build.c                |  4 +-
 hw/i386/pc.c                        | 11 +++--
 hw/mem/pc-dimm.c                    |  2 +
 hw/pci-bridge/pci_expander_bridge.c |  2 +
 hw/ppc/spapr.c                      | 25 ++++++++--
 hw/ppc/spapr_pci.c                  |  2 +
 include/hw/acpi/aml-build.h         |  2 +-
 include/hw/boards.h                 | 22 +++++++++
 include/sysemu/numa.h               | 13 +----
 monitor.c                           |  4 +-
 numa.c                              | 74 +++++++++++++++++------------
 17 files changed, 155 insertions(+), 70 deletions(-)
Signature · 1 line
2.17.1
author
#1 lore raw

PATCH v3 1/3 numa: move numa global variable nb_numa_nodes into MachineState

The aim of this patch is to add struct NumaState in MachineState and move existing numa global nb_numa_nodes(renamed as "num_nodes") into NumaState. And add variable numa_support into MachineClass to decide which submachines support NUMA.

Changes in v3 -> v2:
- rename the "NumaState::nb_numa_nodes" as "NumaState::num_nodes" (Eduardo)
- use machine_num_numa_nodes(MachineState *ms) to check if ms->numa_state is NULL before using NumaState::num_nodes (Eduardo)
- check if ms->numa_state == NULL in the set_numa_options to avoid using -numa in a machine-type which don't support numa

Changes in v2:
- fix the mistake in numa_complete_configuration in numa.c
- add MachineState into some functions to avoid using qdev_get_machine
- add some if experssion to avoid the NumaState is null

---
 exec.c                              |  5 ++-
 hw/acpi/aml-build.c                 |  3 +-
 hw/arm/boot.c                       |  2 ++
 hw/arm/virt-acpi-build.c            |  8 +++--
 hw/arm/virt.c                       |  5 ++-
 hw/core/machine.c                   | 21 ++++++++---
 hw/i386/acpi-build.c                |  2 +-
 hw/i386/pc.c                        |  7 +++-
 hw/mem/pc-dimm.c                    |  2 ++
 hw/pci-bridge/pci_expander_bridge.c |  2 ++
 hw/ppc/spapr.c                      | 17 +++++++--
 include/hw/acpi/aml-build.h         |  2 +-
 include/hw/boards.h                 | 10 ++++++
 include/sysemu/numa.h               |  3 +-
 monitor.c                           |  4 ++-
 numa.c                              | 54 ++++++++++++++++++-----------
 16 files changed, 108 insertions(+), 39 deletions(-)
exec.c +4 -1
index 6ab62f4eee..87f4da207d 100644
@@ -1708,6 +1708,7 @@ long qemu_getrampagesize(void)
     long hpsize = LONG_MAX;
     long mainrampagesize;
     Object *memdev_root;
+    MachineState *ms = MACHINE(qdev_get_machine());
 
     mainrampagesize = qemu_mempath_getpagesize(mem_path);
 
@@ -1735,7 +1736,9 @@ long qemu_getrampagesize(void)
      * so if its page size is smaller we have got to report that size instead.
      */
     if (hpsize > mainrampagesize &&
-        (nb_numa_nodes == 0 || numa_info[0].node_memdev == NULL)) {
+        (ms->numa_state == NULL ||
+         ms->numa_state->num_nodes == 0 ||
+         numa_info[0].node_memdev == NULL)) {
         static bool warned;
         if (!warned) {
             error_report("Huge page support disabled (n/a for main memory).");
hw/acpi/aml-build.c +2 -1
index 555c24f21d..c67f4561a4 100644
@@ -1726,10 +1726,11 @@ void build_srat_memory(AcpiSratMemoryAffinity *numamem, uint64_t base,
  * ACPI spec 5.2.17 System Locality Distance Information Table
  * (Revision 2.0 or later)
  */
-void build_slit(GArray *table_data, BIOSLinker *linker)
+void build_slit(GArray *table_data, BIOSLinker *linker, MachineState *ms)
 {
     int slit_start, i, j;
     slit_start = table_data->len;
+    int nb_numa_nodes = machine_num_numa_nodes(ms);
 
     acpi_data_push(table_data, sizeof(AcpiTableHeader));
 
hw/arm/boot.c +2 -0
index a830655e1a..8ff08814fd 100644
@@ -532,6 +532,8 @@ int arm_load_dtb(hwaddr addr, const struct arm_boot_info *binfo,
     hwaddr mem_base, mem_len;
     char **node_path;
     Error *err = NULL;
+    MachineState *ms = MACHINE(qdev_get_machine());
+    int nb_numa_nodes = machine_num_numa_nodes(ms);
 
     if (binfo->dtb_filename) {
         char *filename;
hw/arm/virt-acpi-build.c +6 -2
index bf9c0bc2f4..6805b4de51 100644
@@ -516,7 +516,9 @@ build_srat(GArray *table_data, BIOSLinker *linker, VirtMachineState *vms)
     int i, srat_start;
     uint64_t mem_base;
     MachineClass *mc = MACHINE_GET_CLASS(vms);
-    const CPUArchIdList *cpu_list = mc->possible_cpu_arch_ids(MACHINE(vms));
+    MachineState *ms = MACHINE(vms);
+    const CPUArchIdList *cpu_list = mc->possible_cpu_arch_ids(ms);
+    int nb_numa_nodes = machine_num_numa_nodes(ms);
 
     srat_start = table_data->len;
     srat = acpi_data_push(table_data, sizeof(*srat));
@@ -780,6 +782,8 @@ void virt_acpi_build(VirtMachineState *vms, AcpiBuildTables *tables)
     GArray *table_offsets;
     unsigned dsdt, xsdt;
     GArray *tables_blob = tables->table_data;
+    MachineState *ms = MACHINE(vms);
+    int nb_numa_nodes = machine_num_numa_nodes(ms);
 
     table_offsets = g_array_new(false, true /* clear */,
                                         sizeof(uint32_t));
@@ -813,7 +817,7 @@ void virt_acpi_build(VirtMachineState *vms, AcpiBuildTables *tables)
         build_srat(tables_blob, tables->linker, vms);
         if (have_numa_distance) {
             acpi_add_table(table_offsets, tables_blob);
-            build_slit(tables_blob, tables->linker);
+            build_slit(tables_blob, tables->linker, ms);
         }
     }
 
hw/arm/virt.c +4 -1
index ce2664a30b..280174b1b7 100644
@@ -195,6 +195,8 @@ static bool cpu_type_valid(const char *cpu)
 
 static void create_fdt(VirtMachineState *vms)
 {
+    MachineState *ms = MACHINE(vms);
+    int nb_numa_nodes = machine_num_numa_nodes(ms);
     void *fdt = create_device_tree(&vms->fdt_size);
 
     if (!fdt) {
@@ -1780,7 +1782,7 @@ virt_cpu_index_to_props(MachineState *ms, unsigned cpu_index)
 
 static int64_t virt_get_default_cpu_node_id(const MachineState *ms, int idx)
 {
-    return idx % nb_numa_nodes;
+    return idx % machine_num_numa_nodes(ms);
 }
 
 static const CPUArchIdList *virt_possible_cpu_arch_ids(MachineState *ms)
@@ -1886,6 +1888,7 @@ static void virt_machine_class_init(ObjectClass *oc, void *data)
     mc->kvm_type = virt_kvm_type;
     assert(!mc->get_hotplug_handler);
     mc->get_hotplug_handler = virt_machine_get_hotplug_handler;
+    mc->numa_supported = true;
     hc->plug = virt_machine_device_plug_cb;
 }
 
hw/core/machine.c +17 -4
index 743fef2898..752a90199d 100644
@@ -854,6 +854,11 @@ static void machine_initfn(Object *obj)
                                         NULL);
     }
 
+    if (mc->numa_supported) {
+        ms->numa_state = g_new0(NumaState, 1);
+    } else {
+        ms->numa_state = NULL;
+    }
 
     /* Register notifier when init is done for sysbus sanity checks */
     ms->sysbus_notifier.notify = machine_init_notify;
@@ -874,6 +879,7 @@ static void machine_finalize(Object *obj)
     g_free(ms->firmware);
     g_free(ms->device_memory);
     g_free(ms->nvdimms_state);
+    g_free(ms->numa_state);
 }
 
 bool machine_usb(MachineState *machine)
@@ -916,6 +922,11 @@ bool machine_mem_merge(MachineState *machine)
     return machine->mem_merge;
 }
 
+int machine_num_numa_nodes(const MachineState *machine)
+{
+    return machine->numa_state ? machine->numa_state->num_nodes : 0;
+}
+
 static char *cpu_slot_to_string(const CPUArchId *cpu)
 {
     GString *s = g_string_new(NULL);
@@ -945,7 +956,7 @@ static void machine_numa_finish_cpu_init(MachineState *machine)
     MachineClass *mc = MACHINE_GET_CLASS(machine);
     const CPUArchIdList *possible_cpus = mc->possible_cpu_arch_ids(machine);
 
-    assert(nb_numa_nodes);
+    assert(machine_num_numa_nodes(machine));
     for (i = 0; i < possible_cpus->len; i++) {
         if (possible_cpus->cpus[i].props.has_node_id) {
             break;
@@ -991,9 +1002,11 @@ void machine_run_board_init(MachineState *machine)
 {
     MachineClass *machine_class = MACHINE_GET_CLASS(machine);
 
-    numa_complete_configuration(machine);
-    if (nb_numa_nodes) {
-        machine_numa_finish_cpu_init(machine);
+    if (machine_class->numa_supported) {
+        numa_complete_configuration(machine);
+        if (machine->numa_state->num_nodes) {
+            machine_numa_finish_cpu_init(machine);
+        }
     }
 
     /* If the machine supports the valid_cpu_types check and the user
hw/i386/acpi-build.c +1 -1
index 416da318ae..7d9bc88ac9 100644
@@ -2687,7 +2687,7 @@ void acpi_build(AcpiBuildTables *tables, MachineState *machine)
         build_srat(tables_blob, tables->linker, machine);
         if (have_numa_distance) {
             acpi_add_table(table_offsets, tables_blob);
-            build_slit(tables_blob, tables->linker);
+            build_slit(tables_blob, tables->linker, machine);
         }
     }
     if (acpi_get_mcfg(&mcfg)) {
hw/i386/pc.c +6 -1
index f2c15bf1f2..a752594df2 100644
@@ -996,6 +996,8 @@ static FWCfgState *bochs_bios_init(AddressSpace *as, PCMachineState *pcms)
     int i;
     const CPUArchIdList *cpus;
     MachineClass *mc = MACHINE_GET_CLASS(pcms);
+    MachineState *ms = MACHINE(pcms);
+    int nb_numa_nodes = machine_num_numa_nodes(ms);
 
     fw_cfg = fw_cfg_init_io_dma(FW_CFG_IO_BASE, FW_CFG_IO_BASE + 4, as);
     fw_cfg_add_i16(fw_cfg, FW_CFG_NB_CPUS, pcms->boot_cpus);
@@ -1672,6 +1674,8 @@ void pc_machine_done(Notifier *notifier, void *data)
 void pc_guest_info_init(PCMachineState *pcms)
 {
     int i;
+    MachineState *ms = MACHINE(pcms);
+    int nb_numa_nodes = machine_num_numa_nodes(ms);
 
     pcms->apic_xrupt_override = kvm_allows_irq0_override();
     pcms->numa_nodes = nb_numa_nodes;
@@ -2655,7 +2659,7 @@ static int64_t pc_get_default_cpu_node_id(const MachineState *ms, int idx)
    assert(idx < ms->possible_cpus->len);
    x86_topo_ids_from_apicid(ms->possible_cpus->cpus[idx].arch_id,
                             smp_cores, smp_threads, &topo);
-   return topo.pkg_id % nb_numa_nodes;
+   return topo.pkg_id % machine_num_numa_nodes(ms);
 }
 
 static const CPUArchIdList *pc_possible_cpu_arch_ids(MachineState *ms)
@@ -2749,6 +2753,7 @@ static void pc_machine_class_init(ObjectClass *oc, void *data)
     nc->nmi_monitor_handler = x86_nmi;
     mc->default_cpu_type = TARGET_DEFAULT_CPU_TYPE;
     mc->nvdimm_supported = true;
+    mc->numa_supported = true;
 
     object_class_property_add(oc, PC_MACHINE_DEVMEM_REGION_SIZE, "int",
         pc_machine_get_device_memory_region_size, NULL,
hw/mem/pc-dimm.c +2 -0
index 152400b1fc..48cbd53e6b 100644
@@ -160,6 +160,8 @@ static void pc_dimm_realize(DeviceState *dev, Error **errp)
 {
     PCDIMMDevice *dimm = PC_DIMM(dev);
     PCDIMMDeviceClass *ddc = PC_DIMM_GET_CLASS(dimm);
+    MachineState *ms = MACHINE(qdev_get_machine());
+    int nb_numa_nodes = machine_num_numa_nodes(ms);
 
     if (!dimm->hostmem) {
         error_setg(errp, "'" PC_DIMM_MEMDEV_PROP "' property is not set");
hw/pci-bridge/pci_expander_bridge.c +2 -0
index e62de4218f..d0590c0973 100644
@@ -217,6 +217,8 @@ static void pxb_dev_realize_common(PCIDevice *dev, bool pcie, Error **errp)
     PCIBus *bus;
     const char *dev_name = NULL;
     Error *local_err = NULL;
+    MachineState *ms = MACHINE(qdev_get_machine());
+    int nb_numa_nodes = machine_num_numa_nodes(ms);
 
     if (pxb->numa_node != NUMA_NODE_UNASSIGNED &&
         pxb->numa_node >= nb_numa_nodes) {
hw/ppc/spapr.c +14 -3
index b52b82d298..23789256f7 100644
@@ -290,6 +290,8 @@ static int spapr_fixup_cpu_dt(void *fdt, SpaprMachineState *spapr)
     CPUState *cs;
     char cpu_model[32];
     uint32_t pft_size_prop[] = {0, cpu_to_be32(spapr->htab_shift)};
+    MachineState *ms = MACHINE(spapr);
+    int nb_numa_nodes = machine_num_numa_nodes(ms);
 
     CPU_FOREACH(cs) {
         PowerPCCPU *cpu = POWERPC_CPU(cs);
@@ -344,6 +346,7 @@ static int spapr_fixup_cpu_dt(void *fdt, SpaprMachineState *spapr)
 
 static hwaddr spapr_node0_size(MachineState *machine)
 {
+    int nb_numa_nodes = machine_num_numa_nodes(machine);
     if (nb_numa_nodes) {
         int i;
         for (i = 0; i < nb_numa_nodes; ++i) {
@@ -390,6 +393,7 @@ static int spapr_populate_memory_node(void *fdt, int nodeid, hwaddr start,
 static int spapr_populate_memory(SpaprMachineState *spapr, void *fdt)
 {
     MachineState *machine = MACHINE(spapr);
+    int nb_numa_nodes = machine_num_numa_nodes(machine);
     hwaddr mem_start, node_size;
     int i, nb_nodes = nb_numa_nodes;
     NodeInfo *nodes = numa_info;
@@ -444,6 +448,8 @@ static void spapr_populate_cpu_dt(CPUState *cs, void *fdt, int offset,
     PowerPCCPU *cpu = POWERPC_CPU(cs);
     CPUPPCState *env = &cpu->env;
     PowerPCCPUClass *pcc = POWERPC_CPU_GET_CLASS(cs);
+    MachineState *ms = MACHINE(spapr);
+    int nb_numa_nodes = machine_num_numa_nodes(ms);
     int index = spapr_get_vcpu_id(cpu);
     uint32_t segs[] = {cpu_to_be32(28), cpu_to_be32(40),
                        0xffffffff, 0xffffffff};
@@ -849,6 +855,7 @@ static int spapr_populate_drmem_v1(SpaprMachineState *spapr, void *fdt,
 static int spapr_populate_drconf_memory(SpaprMachineState *spapr, void *fdt)
 {
     MachineState *machine = MACHINE(spapr);
+    int nb_numa_nodes = machine_num_numa_nodes(machine);
     int ret, i, offset;
     uint64_t lmb_size = SPAPR_MEMORY_BLOCK_SIZE;
     uint32_t prop_lmb_size[] = {0, cpu_to_be32(lmb_size)};
@@ -1023,11 +1030,13 @@ int spapr_h_cas_compose_response(SpaprMachineState *spapr,
 static void spapr_dt_rtas(SpaprMachineState *spapr, void *fdt)
 {
     int rtas;
+    MachineState *ms = MACHINE(spapr);
+    int nb_numa_nodes = machine_num_numa_nodes(ms);
     GString *hypertas = g_string_sized_new(256);
     GString *qemu_hypertas = g_string_sized_new(256);
     uint32_t refpoints[] = { cpu_to_be32(0x4), cpu_to_be32(0x4) };
-    uint64_t max_device_addr = MACHINE(spapr)->device_memory->base +
-        memory_region_size(&MACHINE(spapr)->device_memory->mr);
+    uint64_t max_device_addr = ms->device_memory->base +
+        memory_region_size(&ms->device_memory->mr);
     uint32_t lrdr_capacity[] = {
         cpu_to_be32(max_device_addr >> 32),
         cpu_to_be32(max_device_addr & 0xffffffff),
@@ -2466,6 +2475,7 @@ static void spapr_create_lmb_dr_connectors(SpaprMachineState *spapr)
 static void spapr_validate_node_memory(MachineState *machine, Error **errp)
 {
     int i;
+    int nb_numa_nodes = machine_num_numa_nodes(machine);
 
     if (machine->ram_size % SPAPR_MEMORY_BLOCK_SIZE) {
         error_setg(errp, "Memory size 0x" RAM_ADDR_FMT
@@ -4066,7 +4076,7 @@ spapr_cpu_index_to_props(MachineState *machine, unsigned cpu_index)
 
 static int64_t spapr_get_default_cpu_node_id(const MachineState *ms, int idx)
 {
-    return idx / smp_cores % nb_numa_nodes;
+    return idx / smp_cores % machine_num_numa_nodes(ms);
 }
 
 static const CPUArchIdList *spapr_possible_cpu_arch_ids(MachineState *machine)
@@ -4266,6 +4276,7 @@ static void spapr_machine_class_init(ObjectClass *oc, void *data)
     smc->update_dt_enabled = true;
     mc->default_cpu_type = POWERPC_CPU_TYPE_NAME("power9_v2.0");
     mc->has_hotpluggable_cpus = true;
+    mc->numa_supported = true;
     smc->resize_hpt_default = SPAPR_RESIZE_HPT_ENABLED;
     fwc->get_dev_path = spapr_get_fw_dev_path;
     nc->nmi_monitor_handler = spapr_nmi;
include/hw/acpi/aml-build.h +1 -1
index 1a563ad756..991cf05134 100644
@@ -414,7 +414,7 @@ build_append_gas_from_struct(GArray *table, const struct AcpiGenericAddress *s)
 void build_srat_memory(AcpiSratMemoryAffinity *numamem, uint64_t base,
                        uint64_t len, int node, MemoryAffinityFlags flags);
 
-void build_slit(GArray *table_data, BIOSLinker *linker);
+void build_slit(GArray *table_data, BIOSLinker *linker, MachineState *ms);
 
 void build_fadt(GArray *tbl, BIOSLinker *linker, const AcpiFadtData *f,
                 const char *oem_id, const char *oem_table_id);
include/hw/boards.h +10 -0
index e231860666..579d73f6c4 100644
@@ -5,6 +5,7 @@
 
 #include "sysemu/blockdev.h"
 #include "sysemu/accel.h"
+#include "sysemu/sysemu.h"
 #include "hw/qdev.h"
 #include "qom/object.h"
 #include "qom/cpu.h"
@@ -69,6 +70,7 @@ int machine_kvm_shadow_mem(MachineState *machine);
 int machine_phandle_start(MachineState *machine);
 bool machine_dump_guest_core(MachineState *machine);
 bool machine_mem_merge(MachineState *machine);
+int machine_num_numa_nodes(const MachineState *machine);
 HotpluggableCPUList *machine_query_hotpluggable_cpus(MachineState *machine);
 void machine_set_cpu_numa_node(MachineState *machine,
                                const CpuInstanceProperties *props,
@@ -211,6 +213,7 @@ struct MachineClass {
     bool ignore_boot_device_suffixes;
     bool smbus_no_migration_support;
     bool nvdimm_supported;
+    bool numa_supported;
 
     HotplugHandler *(*get_hotplug_handler)(MachineState *machine,
                                            DeviceState *dev);
@@ -231,6 +234,12 @@ typedef struct DeviceMemoryState {
     MemoryRegion mr;
 } DeviceMemoryState;
 
+typedef struct NumaState {
+    /* Number of NUMA nodes */
+    int num_nodes;
+
+} NumaState;
+
 /**
  * MachineState:
  */
@@ -274,6 +283,7 @@ struct MachineState {
     AccelState *accelerator;
     CPUArchIdList *possible_cpus;
     struct NVDIMMState *nvdimms_state;
+    NumaState *numa_state;
 };
 
 #define DEFINE_MACHINE(namestr, machine_initfn) \
include/sysemu/numa.h +1 -2
index b6ac7de43e..a55e2be563 100644
@@ -6,7 +6,6 @@
 #include "sysemu/hostmem.h"
 #include "hw/boards.h"
 
-extern int nb_numa_nodes;   /* Number of NUMA nodes */
 extern bool have_numa_distance;
 
 struct NodeInfo {
@@ -24,7 +23,7 @@ struct NumaNodeMem {
 extern NodeInfo numa_info[MAX_NODES];
 void parse_numa_opts(MachineState *ms);
 void numa_complete_configuration(MachineState *ms);
-void query_numa_node_mem(NumaNodeMem node_mem[]);
+void query_numa_node_mem(NumaNodeMem node_mem[], MachineState *ms);
 extern QemuOptsList qemu_numa_opts;
 void numa_legacy_auto_assign_ram(MachineClass *mc, NodeInfo *nodes,
                                  int nb_nodes, ram_addr_t size);
monitor.c +3 -1
index 4807bbe811..80b6e01c9b 100644
@@ -1908,11 +1908,13 @@ static void hmp_info_numa(Monitor *mon, const QDict *qdict)
     int i;
     NumaNodeMem *node_mem;
     CpuInfoList *cpu_list, *cpu;
+    MachineState *ms = MACHINE(qdev_get_machine());
+    int nb_numa_nodes = machine_num_numa_nodes(ms);
 
     cpu_list = qmp_query_cpus(&error_abort);
     node_mem = g_new0(NumaNodeMem, nb_numa_nodes);
 
-    query_numa_node_mem(node_mem);
+    query_numa_node_mem(node_mem, ms);
     monitor_printf(mon, "%d nodes\n", nb_numa_nodes);
     for (i = 0; i < nb_numa_nodes; i++) {
         monitor_printf(mon, "node %d cpus:", i);
numa.c +33 -21

158 lines not on this page · raw message

Signature · 1 line
2.17.1
author
#2 lore raw

PATCH v3 2/3 numa: move numa global variable have_numa_distance into MachineState

In reply to Tao Xu

The aim of this patch is to move existing numa global have_numa_distance into NumaState.

---
 hw/arm/virt-acpi-build.c | 2 +-
 hw/arm/virt.c            | 2 +-
 hw/i386/acpi-build.c     | 2 +-
 include/hw/boards.h      | 2 ++
 include/sysemu/numa.h    | 2 --
 numa.c                   | 5 ++---
 6 files changed, 7 insertions(+), 8 deletions(-)
hw/arm/virt-acpi-build.c +1 -1
index 6805b4de51..65f070843c 100644
@@ -815,7 +815,7 @@ void virt_acpi_build(VirtMachineState *vms, AcpiBuildTables *tables)
     if (nb_numa_nodes > 0) {
         acpi_add_table(table_offsets, tables_blob);
         build_srat(tables_blob, tables->linker, vms);
-        if (have_numa_distance) {
+        if (ms->numa_state->have_numa_distance) {
             acpi_add_table(table_offsets, tables_blob);
             build_slit(tables_blob, tables->linker, ms);
         }
hw/arm/virt.c +1 -1
index 280174b1b7..6cd5e13eb9 100644
@@ -228,7 +228,7 @@ static void create_fdt(VirtMachineState *vms)
                                 "clk24mhz");
     qemu_fdt_setprop_cell(fdt, "/apb-pclk", "phandle", vms->clock_phandle);
 
-    if (have_numa_distance) {
+    if (nb_numa_nodes > 0 && ms->numa_state->have_numa_distance) {
         int size = nb_numa_nodes * nb_numa_nodes * 3 * sizeof(uint32_t);
         uint32_t *matrix = g_malloc0(size);
         int idx, i, j;
hw/i386/acpi-build.c +1 -1
index 7d9bc88ac9..43a807c483 100644
@@ -2685,7 +2685,7 @@ void acpi_build(AcpiBuildTables *tables, MachineState *machine)
     if (pcms->numa_nodes) {
         acpi_add_table(table_offsets, tables_blob);
         build_srat(tables_blob, tables->linker, machine);
-        if (have_numa_distance) {
+        if (machine->numa_state->have_numa_distance) {
             acpi_add_table(table_offsets, tables_blob);
             build_slit(tables_blob, tables->linker, machine);
         }
include/hw/boards.h +2 -0
index 579d73f6c4..c3ff2dbd4a 100644
@@ -238,6 +238,8 @@ typedef struct NumaState {
     /* Number of NUMA nodes */
     int num_nodes;
 
+    /* Allow setting NUMA distance for different NUMA nodes */
+    bool have_numa_distance;
 } NumaState;
 
 /**
include/sysemu/numa.h +0 -2
index a55e2be563..1a29408db9 100644
@@ -6,8 +6,6 @@
 #include "sysemu/hostmem.h"
 #include "hw/boards.h"
 
-extern bool have_numa_distance;
-
 struct NodeInfo {
     uint64_t node_mem;
     struct HostMemoryBackend *node_memdev;
numa.c +2 -3
index 343fcaf13f..d4f5ff5193 100644
@@ -52,7 +52,6 @@ static int have_memdevs = -1;
 static int max_numa_nodeid; /* Highest specified NUMA node ID, plus one.
                              * For all nodes, nodeid < max_numa_nodeid
                              */
-bool have_numa_distance;
 NodeInfo numa_info[MAX_NODES];
 
 
@@ -171,7 +170,7 @@ void parse_numa_distance(MachineState *ms, NumaDistOptions *dist, Error **errp)
     }
 
     numa_info[src].distance[dst] = val;
-    have_numa_distance = true;
+    ms->numa_state->have_numa_distance = true;
 }
 
 static
@@ -442,7 +441,7 @@ void numa_complete_configuration(MachineState *ms)
          * asymmetric. In this case, the distances for both directions
          * of all node pairs are required.
          */
-        if (have_numa_distance) {
+        if (ms->numa_state->have_numa_distance) {
             /* Validate enough NUMA distance information was provided. */
             validate_numa_distance(ms);
 
Signature · 1 line
2.17.1
author
#3 lore raw

PATCH v3 3/3 numa: move numa global variable numa_info into MachineState

In reply to Tao Xu

The aim of this patch is to move existing numa global numa_info (renamed as "nodes") into NumaState.

Changes in v3 -> v2:
- rename the "NumaState::numa_info" as "NumaState::nodes" (Eduardo)

---
 exec.c                   |  2 +-
 hw/acpi/aml-build.c      |  6 ++++--
 hw/arm/boot.c            |  2 +-
 hw/arm/virt-acpi-build.c |  7 ++++---
 hw/arm/virt.c            |  1 +
 hw/i386/pc.c             |  4 ++--
 hw/ppc/spapr.c           |  8 +++++++-
 hw/ppc/spapr_pci.c       |  2 ++
 include/hw/boards.h      | 10 ++++++++++
 include/sysemu/numa.h    |  8 --------
 numa.c                   | 15 +++++++++------
 11 files changed, 41 insertions(+), 24 deletions(-)
exec.c +1 -1
index 87f4da207d..6b3bda2766 100644
@@ -1738,7 +1738,7 @@ long qemu_getrampagesize(void)
     if (hpsize > mainrampagesize &&
         (ms->numa_state == NULL ||
          ms->numa_state->num_nodes == 0 ||
-         numa_info[0].node_memdev == NULL)) {
+         ms->numa_state->nodes[0].node_memdev == NULL)) {
         static bool warned;
         if (!warned) {
             error_report("Huge page support disabled (n/a for main memory).");
hw/acpi/aml-build.c +4 -2
index c67f4561a4..b53a55cb56 100644
@@ -1737,8 +1737,10 @@ void build_slit(GArray *table_data, BIOSLinker *linker, MachineState *ms)
     build_append_int_noprefix(table_data, nb_numa_nodes, 8);
     for (i = 0; i < nb_numa_nodes; i++) {
         for (j = 0; j < nb_numa_nodes; j++) {
-            assert(numa_info[i].distance[j]);
-            build_append_int_noprefix(table_data, numa_info[i].distance[j], 1);
+            assert(ms->numa_state->nodes[i].distance[j]);
+            build_append_int_noprefix(table_data,
+                                      ms->numa_state->nodes[i].distance[j],
+                                      1);
         }
     }
 
hw/arm/boot.c +1 -1
index 8ff08814fd..845b737ab9 100644
@@ -602,7 +602,7 @@ int arm_load_dtb(hwaddr addr, const struct arm_boot_info *binfo,
     if (nb_numa_nodes > 0) {
         mem_base = binfo->loader_start;
         for (i = 0; i < nb_numa_nodes; i++) {
-            mem_len = numa_info[i].node_mem;
+            mem_len = ms->numa_state->nodes[i].node_mem;
             rc = fdt_add_memory_node(fdt, acells, mem_base,
                                      scells, mem_len, i);
             if (rc < 0) {
hw/arm/virt-acpi-build.c +4 -3
index 65f070843c..b22c3d27ad 100644
@@ -535,11 +535,12 @@ build_srat(GArray *table_data, BIOSLinker *linker, VirtMachineState *vms)
 
     mem_base = vms->memmap[VIRT_MEM].base;
     for (i = 0; i < nb_numa_nodes; ++i) {
-        if (numa_info[i].node_mem > 0) {
+        if (ms->numa_state->nodes[i].node_mem > 0) {
             numamem = acpi_data_push(table_data, sizeof(*numamem));
-            build_srat_memory(numamem, mem_base, numa_info[i].node_mem, i,
+            build_srat_memory(numamem, mem_base,
+                              ms->numa_state->nodes[i].node_mem, i,
                               MEM_AFFINITY_ENABLED);
-            mem_base += numa_info[i].node_mem;
+            mem_base += ms->numa_state->nodes[i].node_mem;
         }
     }
 
hw/arm/virt.c +1 -0
index 6cd5e13eb9..8a8fc0d9b4 100644
@@ -232,6 +232,7 @@ static void create_fdt(VirtMachineState *vms)
         int size = nb_numa_nodes * nb_numa_nodes * 3 * sizeof(uint32_t);
         uint32_t *matrix = g_malloc0(size);
         int idx, i, j;
+        NodeInfo *numa_info = ms->numa_state->nodes;
 
         for (i = 0; i < nb_numa_nodes; i++) {
             for (j = 0; j < nb_numa_nodes; j++) {
hw/i386/pc.c +2 -2
index a752594df2..4ace0ad893 100644
@@ -1040,7 +1040,7 @@ static FWCfgState *bochs_bios_init(AddressSpace *as, PCMachineState *pcms)
     }
     for (i = 0; i < nb_numa_nodes; i++) {
         numa_fw_cfg[pcms->apic_id_limit + 1 + i] =
-            cpu_to_le64(numa_info[i].node_mem);
+            cpu_to_le64(ms->numa_state->nodes[i].node_mem);
     }
     fw_cfg_add_bytes(fw_cfg, FW_CFG_NUMA, numa_fw_cfg,
                      (1 + pcms->apic_id_limit + nb_numa_nodes) *
@@ -1682,7 +1682,7 @@ void pc_guest_info_init(PCMachineState *pcms)
     pcms->node_mem = g_malloc0(pcms->numa_nodes *
                                     sizeof *pcms->node_mem);
     for (i = 0; i < nb_numa_nodes; i++) {
-        pcms->node_mem[i] = numa_info[i].node_mem;
+        pcms->node_mem[i] = ms->numa_state->nodes[i].node_mem;
     }
 
     pcms->machine_done.notify = pc_machine_done;
hw/ppc/spapr.c +7 -1
index 23789256f7..7701cb1dac 100644
@@ -349,6 +349,7 @@ static hwaddr spapr_node0_size(MachineState *machine)
     int nb_numa_nodes = machine_num_numa_nodes(machine);
     if (nb_numa_nodes) {
         int i;
+        NodeInfo *numa_info = machine->numa_state->nodes;
         for (i = 0; i < nb_numa_nodes; ++i) {
             if (numa_info[i].node_mem) {
                 return MIN(pow2floor(numa_info[i].node_mem),
@@ -396,7 +397,9 @@ static int spapr_populate_memory(SpaprMachineState *spapr, void *fdt)
     int nb_numa_nodes = machine_num_numa_nodes(machine);
     hwaddr mem_start, node_size;
     int i, nb_nodes = nb_numa_nodes;
-    NodeInfo *nodes = numa_info;
+    NodeInfo *nodes = machine->numa_state ?
+                      machine->numa_state->nodes :
+                      NULL;
     NodeInfo ramnode;
 
     /* No NUMA nodes, assume there is just one node with whole RAM */
@@ -2476,6 +2479,9 @@ static void spapr_validate_node_memory(MachineState *machine, Error **errp)
 {
     int i;
     int nb_numa_nodes = machine_num_numa_nodes(machine);
+    NodeInfo *numa_info = machine->numa_state ?
+                          machine->numa_state->nodes :
+                          NULL;
 
     if (machine->ram_size % SPAPR_MEMORY_BLOCK_SIZE) {
         error_setg(errp, "Memory size 0x" RAM_ADDR_FMT
hw/ppc/spapr_pci.c +2 -0
index f62e6833b8..d4378558d3 100644
@@ -1672,6 +1672,8 @@ static void spapr_phb_realize(DeviceState *dev, Error **errp)
     SysBusDevice *s = SYS_BUS_DEVICE(dev);
     SpaprPhbState *sphb = SPAPR_PCI_HOST_BRIDGE(s);
     PCIHostState *phb = PCI_HOST_BRIDGE(s);
+    MachineState *ms = MACHINE(spapr);
+    NodeInfo *numa_info = ms->numa_state ? ms->numa_state->nodes : NULL;
     char *namebuf;
     int i;
     PCIBus *bus;
include/hw/boards.h +10 -0
index c3ff2dbd4a..e2676009fe 100644
@@ -234,12 +234,22 @@ typedef struct DeviceMemoryState {
     MemoryRegion mr;
 } DeviceMemoryState;
 
+struct NodeInfo {
+    uint64_t node_mem;
+    struct HostMemoryBackend *node_memdev;
+    bool present;
+    uint8_t distance[MAX_NODES];
+};
+
 typedef struct NumaState {
     /* Number of NUMA nodes */
     int num_nodes;
 
     /* Allow setting NUMA distance for different NUMA nodes */
     bool have_numa_distance;
+
+    /* NUMA nodes information */
+    NodeInfo nodes[MAX_NODES];
 } NumaState;
 
 /**
include/sysemu/numa.h +0 -8
index 1a29408db9..7b8011f9ea 100644
@@ -6,19 +6,11 @@
 #include "sysemu/hostmem.h"
 #include "hw/boards.h"
 
-struct NodeInfo {
-    uint64_t node_mem;
-    struct HostMemoryBackend *node_memdev;
-    bool present;
-    uint8_t distance[MAX_NODES];
-};
-
 struct NumaNodeMem {
     uint64_t node_mem;
     uint64_t node_plugged_mem;
 };
 
-extern NodeInfo numa_info[MAX_NODES];
 void parse_numa_opts(MachineState *ms);
 void numa_complete_configuration(MachineState *ms);
 void query_numa_node_mem(NumaNodeMem node_mem[], MachineState *ms);
numa.c +9 -6
index d4f5ff5193..ddea376d72 100644
@@ -52,8 +52,6 @@ static int have_memdevs = -1;
 static int max_numa_nodeid; /* Highest specified NUMA node ID, plus one.
                              * For all nodes, nodeid < max_numa_nodeid
                              */
-NodeInfo numa_info[MAX_NODES];
-
 
 static void parse_numa_node(MachineState *ms, NumaNodeOptions *node,
                             Error **errp)
@@ -62,6 +60,7 @@ static void parse_numa_node(MachineState *ms, NumaNodeOptions *node,
     uint16_t nodenr;
     uint16List *cpus = NULL;
     MachineClass *mc = MACHINE_GET_CLASS(ms);
+    NodeInfo *numa_info = ms->numa_state->nodes;
 
     if (node->has_nodeid) {
         nodenr = node->nodeid;
@@ -143,6 +142,7 @@ void parse_numa_distance(MachineState *ms, NumaDistOptions *dist, Error **errp)
     uint16_t src = dist->src;
     uint16_t dst = dist->dst;
     uint8_t val = dist->val;
+    NodeInfo *numa_info = ms->numa_state->nodes;
 
     if (src >= MAX_NODES || dst >= MAX_NODES) {
         error_setg(errp, "Parameter '%s' expects an integer between 0 and %d",
@@ -201,7 +201,7 @@ void set_numa_options(MachineState *ms, NumaOptions *object, Error **errp)
             error_setg(&err, "Missing mandatory node-id property");
             goto end;
         }
-        if (!numa_info[object->u.cpu.node_id].present) {
+        if (!ms->numa_state->nodes[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);
@@ -261,6 +261,7 @@ static void validate_numa_distance(MachineState *ms)
     int src, dst;
     bool is_asymmetrical = false;
     int nb_numa_nodes = machine_num_numa_nodes(ms);
+    NodeInfo *numa_info = ms->numa_state->nodes;
 
     for (src = 0; src < nb_numa_nodes; src++) {
         for (dst = src; dst < nb_numa_nodes; dst++) {
@@ -302,6 +303,7 @@ static void complete_init_numa_distance(MachineState *ms)
 {
     int src, dst;
     int nb_numa_nodes = machine_num_numa_nodes(ms);
+    NodeInfo *numa_info = ms->numa_state->nodes;
 
     /* Fixup NUMA distance by symmetric policy because if it is an
      * asymmetric distance table, it should be a complete table and
@@ -361,6 +363,7 @@ void numa_complete_configuration(MachineState *ms)
 {
     int i;
     MachineClass *mc = MACHINE_GET_CLASS(ms);
+    NodeInfo *numa_info = ms->numa_state->nodes;
 
     /*
      * If memory hotplug is enabled (slots > 0) but without '-numa'
@@ -532,8 +535,8 @@ void memory_region_allocate_system_memory(MemoryRegion *mr, Object *owner,
 
     memory_region_init(mr, owner, name, ram_size);
     for (i = 0; i < nb_numa_nodes; i++) {
-        uint64_t size = numa_info[i].node_mem;
-        HostMemoryBackend *backend = numa_info[i].node_memdev;
+        uint64_t size = ms->numa_state->nodes[i].node_mem;
+        HostMemoryBackend *backend = ms->numa_state->nodes[i].node_memdev;
         if (!backend) {
             continue;
         }
@@ -599,7 +602,7 @@ void query_numa_node_mem(NumaNodeMem node_mem[], MachineState *ms)
 
     numa_stat_memory_devices(node_mem);
     for (i = 0; i < ms->numa_state->num_nodes; i++) {
-        node_mem[i].node_mem += numa_info[i].node_mem;
+        node_mem[i].node_mem += ms->numa_state->nodes[i].node_mem;
     }
 }
 
Signature · 1 line
2.17.1