numa: move numa global variable nb_numa_nodes into MachineState
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. Reviewed-by: Igor Mammedov <imammedo@redhat.com> Suggested-by: Igor Mammedov <imammedo@redhat.com> Suggested-by: Eduardo Habkost <ehabkost@redhat.com> Signed-off-by: Tao Xu <tao3.xu@intel.com> Message-Id: <20190809065731.9097-3-tao3.xu@intel.com> [ehabkost: include hw/boards.h again to fix build failures] Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
This commit is contained in:
parent
2744ece809
commit
aa57020774
5
exec.c
5
exec.c
@ -1750,6 +1750,7 @@ long qemu_minrampagesize(void)
|
|||||||
long hpsize = LONG_MAX;
|
long hpsize = LONG_MAX;
|
||||||
long mainrampagesize;
|
long mainrampagesize;
|
||||||
Object *memdev_root;
|
Object *memdev_root;
|
||||||
|
MachineState *ms = MACHINE(qdev_get_machine());
|
||||||
|
|
||||||
mainrampagesize = qemu_mempath_getpagesize(mem_path);
|
mainrampagesize = qemu_mempath_getpagesize(mem_path);
|
||||||
|
|
||||||
@ -1777,7 +1778,9 @@ long qemu_minrampagesize(void)
|
|||||||
* so if its page size is smaller we have got to report that size instead.
|
* so if its page size is smaller we have got to report that size instead.
|
||||||
*/
|
*/
|
||||||
if (hpsize > mainrampagesize &&
|
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;
|
static bool warned;
|
||||||
if (!warned) {
|
if (!warned) {
|
||||||
error_report("Huge page support disabled (n/a for main memory).");
|
error_report("Huge page support disabled (n/a for main memory).");
|
||||||
|
@ -25,6 +25,7 @@
|
|||||||
#include "qemu/bswap.h"
|
#include "qemu/bswap.h"
|
||||||
#include "qemu/bitops.h"
|
#include "qemu/bitops.h"
|
||||||
#include "sysemu/numa.h"
|
#include "sysemu/numa.h"
|
||||||
|
#include "hw/boards.h"
|
||||||
|
|
||||||
static GArray *build_alloc_array(void)
|
static GArray *build_alloc_array(void)
|
||||||
{
|
{
|
||||||
@ -1726,10 +1727,11 @@ void build_srat_memory(AcpiSratMemoryAffinity *numamem, uint64_t base,
|
|||||||
* ACPI spec 5.2.17 System Locality Distance Information Table
|
* ACPI spec 5.2.17 System Locality Distance Information Table
|
||||||
* (Revision 2.0 or later)
|
* (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;
|
int slit_start, i, j;
|
||||||
slit_start = table_data->len;
|
slit_start = table_data->len;
|
||||||
|
int nb_numa_nodes = ms->numa_state->num_nodes;
|
||||||
|
|
||||||
acpi_data_push(table_data, sizeof(AcpiTableHeader));
|
acpi_data_push(table_data, sizeof(AcpiTableHeader));
|
||||||
|
|
||||||
|
@ -598,9 +598,9 @@ int arm_load_dtb(hwaddr addr, const struct arm_boot_info *binfo,
|
|||||||
}
|
}
|
||||||
g_strfreev(node_path);
|
g_strfreev(node_path);
|
||||||
|
|
||||||
if (nb_numa_nodes > 0) {
|
if (ms->numa_state != NULL && ms->numa_state->num_nodes > 0) {
|
||||||
mem_base = binfo->loader_start;
|
mem_base = binfo->loader_start;
|
||||||
for (i = 0; i < nb_numa_nodes; i++) {
|
for (i = 0; i < ms->numa_state->num_nodes; i++) {
|
||||||
mem_len = numa_info[i].node_mem;
|
mem_len = numa_info[i].node_mem;
|
||||||
rc = fdt_add_memory_node(fdt, acells, mem_base,
|
rc = fdt_add_memory_node(fdt, acells, mem_base,
|
||||||
scells, mem_len, i);
|
scells, mem_len, i);
|
||||||
|
@ -146,6 +146,7 @@ static void create_fdt(SBSAMachineState *sms)
|
|||||||
{
|
{
|
||||||
void *fdt = create_device_tree(&sms->fdt_size);
|
void *fdt = create_device_tree(&sms->fdt_size);
|
||||||
const MachineState *ms = MACHINE(sms);
|
const MachineState *ms = MACHINE(sms);
|
||||||
|
int nb_numa_nodes = ms->numa_state->num_nodes;
|
||||||
int cpu;
|
int cpu;
|
||||||
|
|
||||||
if (!fdt) {
|
if (!fdt) {
|
||||||
@ -762,7 +763,7 @@ sbsa_ref_cpu_index_to_props(MachineState *ms, unsigned cpu_index)
|
|||||||
static int64_t
|
static int64_t
|
||||||
sbsa_ref_get_default_cpu_node_id(const MachineState *ms, int idx)
|
sbsa_ref_get_default_cpu_node_id(const MachineState *ms, int idx)
|
||||||
{
|
{
|
||||||
return idx % nb_numa_nodes;
|
return idx % ms->numa_state->num_nodes;
|
||||||
}
|
}
|
||||||
|
|
||||||
static void sbsa_ref_instance_init(Object *obj)
|
static void sbsa_ref_instance_init(Object *obj)
|
||||||
@ -789,6 +790,7 @@ static void sbsa_ref_class_init(ObjectClass *oc, void *data)
|
|||||||
mc->possible_cpu_arch_ids = sbsa_ref_possible_cpu_arch_ids;
|
mc->possible_cpu_arch_ids = sbsa_ref_possible_cpu_arch_ids;
|
||||||
mc->cpu_index_to_instance_props = sbsa_ref_cpu_index_to_props;
|
mc->cpu_index_to_instance_props = sbsa_ref_cpu_index_to_props;
|
||||||
mc->get_default_cpu_node_id = sbsa_ref_get_default_cpu_node_id;
|
mc->get_default_cpu_node_id = sbsa_ref_get_default_cpu_node_id;
|
||||||
|
mc->numa_mem_supported = true;
|
||||||
}
|
}
|
||||||
|
|
||||||
static const TypeInfo sbsa_ref_info = {
|
static const TypeInfo sbsa_ref_info = {
|
||||||
|
@ -517,7 +517,8 @@ build_srat(GArray *table_data, BIOSLinker *linker, VirtMachineState *vms)
|
|||||||
int i, srat_start;
|
int i, srat_start;
|
||||||
uint64_t mem_base;
|
uint64_t mem_base;
|
||||||
MachineClass *mc = MACHINE_GET_CLASS(vms);
|
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);
|
||||||
|
|
||||||
srat_start = table_data->len;
|
srat_start = table_data->len;
|
||||||
srat = acpi_data_push(table_data, sizeof(*srat));
|
srat = acpi_data_push(table_data, sizeof(*srat));
|
||||||
@ -533,7 +534,7 @@ build_srat(GArray *table_data, BIOSLinker *linker, VirtMachineState *vms)
|
|||||||
}
|
}
|
||||||
|
|
||||||
mem_base = vms->memmap[VIRT_MEM].base;
|
mem_base = vms->memmap[VIRT_MEM].base;
|
||||||
for (i = 0; i < nb_numa_nodes; ++i) {
|
for (i = 0; i < ms->numa_state->num_nodes; ++i) {
|
||||||
if (numa_info[i].node_mem > 0) {
|
if (numa_info[i].node_mem > 0) {
|
||||||
numamem = acpi_data_push(table_data, sizeof(*numamem));
|
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, numa_info[i].node_mem, i,
|
||||||
@ -759,6 +760,7 @@ void virt_acpi_build(VirtMachineState *vms, AcpiBuildTables *tables)
|
|||||||
GArray *table_offsets;
|
GArray *table_offsets;
|
||||||
unsigned dsdt, xsdt;
|
unsigned dsdt, xsdt;
|
||||||
GArray *tables_blob = tables->table_data;
|
GArray *tables_blob = tables->table_data;
|
||||||
|
MachineState *ms = MACHINE(vms);
|
||||||
|
|
||||||
table_offsets = g_array_new(false, true /* clear */,
|
table_offsets = g_array_new(false, true /* clear */,
|
||||||
sizeof(uint32_t));
|
sizeof(uint32_t));
|
||||||
@ -793,12 +795,12 @@ void virt_acpi_build(VirtMachineState *vms, AcpiBuildTables *tables)
|
|||||||
acpi_add_table(table_offsets, tables_blob);
|
acpi_add_table(table_offsets, tables_blob);
|
||||||
build_spcr(tables_blob, tables->linker, vms);
|
build_spcr(tables_blob, tables->linker, vms);
|
||||||
|
|
||||||
if (nb_numa_nodes > 0) {
|
if (ms->numa_state->num_nodes > 0) {
|
||||||
acpi_add_table(table_offsets, tables_blob);
|
acpi_add_table(table_offsets, tables_blob);
|
||||||
build_srat(tables_blob, tables->linker, vms);
|
build_srat(tables_blob, tables->linker, vms);
|
||||||
if (have_numa_distance) {
|
if (have_numa_distance) {
|
||||||
acpi_add_table(table_offsets, tables_blob);
|
acpi_add_table(table_offsets, tables_blob);
|
||||||
build_slit(tables_blob, tables->linker);
|
build_slit(tables_blob, tables->linker, ms);
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
|
@ -203,6 +203,8 @@ static bool cpu_type_valid(const char *cpu)
|
|||||||
|
|
||||||
static void create_fdt(VirtMachineState *vms)
|
static void create_fdt(VirtMachineState *vms)
|
||||||
{
|
{
|
||||||
|
MachineState *ms = MACHINE(vms);
|
||||||
|
int nb_numa_nodes = ms->numa_state->num_nodes;
|
||||||
void *fdt = create_device_tree(&vms->fdt_size);
|
void *fdt = create_device_tree(&vms->fdt_size);
|
||||||
|
|
||||||
if (!fdt) {
|
if (!fdt) {
|
||||||
@ -1846,7 +1848,7 @@ virt_cpu_index_to_props(MachineState *ms, unsigned cpu_index)
|
|||||||
|
|
||||||
static int64_t virt_get_default_cpu_node_id(const MachineState *ms, int idx)
|
static int64_t virt_get_default_cpu_node_id(const MachineState *ms, int idx)
|
||||||
{
|
{
|
||||||
return idx % nb_numa_nodes;
|
return idx % ms->numa_state->num_nodes;
|
||||||
}
|
}
|
||||||
|
|
||||||
static const CPUArchIdList *virt_possible_cpu_arch_ids(MachineState *ms)
|
static const CPUArchIdList *virt_possible_cpu_arch_ids(MachineState *ms)
|
||||||
|
@ -23,6 +23,7 @@
|
|||||||
#include "qapi/string-output-visitor.h"
|
#include "qapi/string-output-visitor.h"
|
||||||
#include "qemu/error-report.h"
|
#include "qemu/error-report.h"
|
||||||
#include "sysemu/numa.h"
|
#include "sysemu/numa.h"
|
||||||
|
#include "hw/boards.h"
|
||||||
|
|
||||||
void hmp_info_cpus(Monitor *mon, const QDict *qdict)
|
void hmp_info_cpus(Monitor *mon, const QDict *qdict)
|
||||||
{
|
{
|
||||||
@ -139,15 +140,21 @@ void hmp_info_memdev(Monitor *mon, const QDict *qdict)
|
|||||||
|
|
||||||
void hmp_info_numa(Monitor *mon, const QDict *qdict)
|
void hmp_info_numa(Monitor *mon, const QDict *qdict)
|
||||||
{
|
{
|
||||||
int i;
|
int i, nb_numa_nodes;
|
||||||
NumaNodeMem *node_mem;
|
NumaNodeMem *node_mem;
|
||||||
CpuInfoList *cpu_list, *cpu;
|
CpuInfoList *cpu_list, *cpu;
|
||||||
|
MachineState *ms = MACHINE(qdev_get_machine());
|
||||||
|
|
||||||
|
nb_numa_nodes = ms->numa_state ? ms->numa_state->num_nodes : 0;
|
||||||
|
monitor_printf(mon, "%d nodes\n", nb_numa_nodes);
|
||||||
|
if (!nb_numa_nodes) {
|
||||||
|
return;
|
||||||
|
}
|
||||||
|
|
||||||
cpu_list = qmp_query_cpus(&error_abort);
|
cpu_list = qmp_query_cpus(&error_abort);
|
||||||
node_mem = g_new0(NumaNodeMem, nb_numa_nodes);
|
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++) {
|
for (i = 0; i < nb_numa_nodes; i++) {
|
||||||
monitor_printf(mon, "node %d cpus:", i);
|
monitor_printf(mon, "node %d cpus:", i);
|
||||||
for (cpu = cpu_list; cpu; cpu = cpu->next) {
|
for (cpu = cpu_list; cpu; cpu = cpu->next) {
|
||||||
|
@ -956,6 +956,9 @@ static void machine_initfn(Object *obj)
|
|||||||
NULL);
|
NULL);
|
||||||
}
|
}
|
||||||
|
|
||||||
|
if (mc->numa_mem_supported) {
|
||||||
|
ms->numa_state = g_new0(NumaState, 1);
|
||||||
|
}
|
||||||
|
|
||||||
/* Register notifier when init is done for sysbus sanity checks */
|
/* Register notifier when init is done for sysbus sanity checks */
|
||||||
ms->sysbus_notifier.notify = machine_init_notify;
|
ms->sysbus_notifier.notify = machine_init_notify;
|
||||||
@ -976,6 +979,7 @@ static void machine_finalize(Object *obj)
|
|||||||
g_free(ms->firmware);
|
g_free(ms->firmware);
|
||||||
g_free(ms->device_memory);
|
g_free(ms->device_memory);
|
||||||
g_free(ms->nvdimms_state);
|
g_free(ms->nvdimms_state);
|
||||||
|
g_free(ms->numa_state);
|
||||||
}
|
}
|
||||||
|
|
||||||
bool machine_usb(MachineState *machine)
|
bool machine_usb(MachineState *machine)
|
||||||
@ -1050,7 +1054,7 @@ static void machine_numa_finish_cpu_init(MachineState *machine)
|
|||||||
MachineClass *mc = MACHINE_GET_CLASS(machine);
|
MachineClass *mc = MACHINE_GET_CLASS(machine);
|
||||||
const CPUArchIdList *possible_cpus = mc->possible_cpu_arch_ids(machine);
|
const CPUArchIdList *possible_cpus = mc->possible_cpu_arch_ids(machine);
|
||||||
|
|
||||||
assert(nb_numa_nodes);
|
assert(machine->numa_state->num_nodes);
|
||||||
for (i = 0; i < possible_cpus->len; i++) {
|
for (i = 0; i < possible_cpus->len; i++) {
|
||||||
if (possible_cpus->cpus[i].props.has_node_id) {
|
if (possible_cpus->cpus[i].props.has_node_id) {
|
||||||
break;
|
break;
|
||||||
@ -1096,10 +1100,12 @@ void machine_run_board_init(MachineState *machine)
|
|||||||
{
|
{
|
||||||
MachineClass *machine_class = MACHINE_GET_CLASS(machine);
|
MachineClass *machine_class = MACHINE_GET_CLASS(machine);
|
||||||
|
|
||||||
|
if (machine_class->numa_mem_supported) {
|
||||||
numa_complete_configuration(machine);
|
numa_complete_configuration(machine);
|
||||||
if (nb_numa_nodes) {
|
if (machine->numa_state->num_nodes) {
|
||||||
machine_numa_finish_cpu_init(machine);
|
machine_numa_finish_cpu_init(machine);
|
||||||
}
|
}
|
||||||
|
}
|
||||||
|
|
||||||
/* If the machine supports the valid_cpu_types check and the user
|
/* If the machine supports the valid_cpu_types check and the user
|
||||||
* specified a CPU with -cpu check here that the user CPU is supported.
|
* specified a CPU with -cpu check here that the user CPU is supported.
|
||||||
|
@ -55,7 +55,6 @@ static int have_mem;
|
|||||||
static int max_numa_nodeid; /* Highest specified NUMA node ID, plus one.
|
static int max_numa_nodeid; /* Highest specified NUMA node ID, plus one.
|
||||||
* For all nodes, nodeid < max_numa_nodeid
|
* For all nodes, nodeid < max_numa_nodeid
|
||||||
*/
|
*/
|
||||||
int nb_numa_nodes;
|
|
||||||
bool have_numa_distance;
|
bool have_numa_distance;
|
||||||
NodeInfo numa_info[MAX_NODES];
|
NodeInfo numa_info[MAX_NODES];
|
||||||
|
|
||||||
@ -72,7 +71,7 @@ static void parse_numa_node(MachineState *ms, NumaNodeOptions *node,
|
|||||||
if (node->has_nodeid) {
|
if (node->has_nodeid) {
|
||||||
nodenr = node->nodeid;
|
nodenr = node->nodeid;
|
||||||
} else {
|
} else {
|
||||||
nodenr = nb_numa_nodes;
|
nodenr = ms->numa_state->num_nodes;
|
||||||
}
|
}
|
||||||
|
|
||||||
if (nodenr >= MAX_NODES) {
|
if (nodenr >= MAX_NODES) {
|
||||||
@ -138,10 +137,11 @@ static void parse_numa_node(MachineState *ms, NumaNodeOptions *node,
|
|||||||
}
|
}
|
||||||
numa_info[nodenr].present = true;
|
numa_info[nodenr].present = true;
|
||||||
max_numa_nodeid = MAX(max_numa_nodeid, nodenr + 1);
|
max_numa_nodeid = MAX(max_numa_nodeid, nodenr + 1);
|
||||||
nb_numa_nodes++;
|
ms->numa_state->num_nodes++;
|
||||||
}
|
}
|
||||||
|
|
||||||
static void parse_numa_distance(NumaDistOptions *dist, Error **errp)
|
static
|
||||||
|
void parse_numa_distance(MachineState *ms, NumaDistOptions *dist, Error **errp)
|
||||||
{
|
{
|
||||||
uint16_t src = dist->src;
|
uint16_t src = dist->src;
|
||||||
uint16_t dst = dist->dst;
|
uint16_t dst = dist->dst;
|
||||||
@ -179,6 +179,12 @@ static void parse_numa_distance(NumaDistOptions *dist, Error **errp)
|
|||||||
void set_numa_options(MachineState *ms, NumaOptions *object, Error **errp)
|
void set_numa_options(MachineState *ms, NumaOptions *object, Error **errp)
|
||||||
{
|
{
|
||||||
Error *err = NULL;
|
Error *err = NULL;
|
||||||
|
MachineClass *mc = MACHINE_GET_CLASS(ms);
|
||||||
|
|
||||||
|
if (!mc->numa_mem_supported) {
|
||||||
|
error_setg(errp, "NUMA is not supported by this machine-type");
|
||||||
|
goto end;
|
||||||
|
}
|
||||||
|
|
||||||
switch (object->type) {
|
switch (object->type) {
|
||||||
case NUMA_OPTIONS_TYPE_NODE:
|
case NUMA_OPTIONS_TYPE_NODE:
|
||||||
@ -188,7 +194,7 @@ void set_numa_options(MachineState *ms, NumaOptions *object, Error **errp)
|
|||||||
}
|
}
|
||||||
break;
|
break;
|
||||||
case NUMA_OPTIONS_TYPE_DIST:
|
case NUMA_OPTIONS_TYPE_DIST:
|
||||||
parse_numa_distance(&object->u.dist, &err);
|
parse_numa_distance(ms, &object->u.dist, &err);
|
||||||
if (err) {
|
if (err) {
|
||||||
goto end;
|
goto end;
|
||||||
}
|
}
|
||||||
@ -253,10 +259,11 @@ end:
|
|||||||
* distance from a node to itself is always NUMA_DISTANCE_MIN,
|
* distance from a node to itself is always NUMA_DISTANCE_MIN,
|
||||||
* so providing it is never necessary.
|
* so providing it is never necessary.
|
||||||
*/
|
*/
|
||||||
static void validate_numa_distance(void)
|
static void validate_numa_distance(MachineState *ms)
|
||||||
{
|
{
|
||||||
int src, dst;
|
int src, dst;
|
||||||
bool is_asymmetrical = false;
|
bool is_asymmetrical = false;
|
||||||
|
int nb_numa_nodes = ms->numa_state->num_nodes;
|
||||||
|
|
||||||
for (src = 0; src < nb_numa_nodes; src++) {
|
for (src = 0; src < nb_numa_nodes; src++) {
|
||||||
for (dst = src; dst < nb_numa_nodes; dst++) {
|
for (dst = src; dst < nb_numa_nodes; dst++) {
|
||||||
@ -294,7 +301,7 @@ static void validate_numa_distance(void)
|
|||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
|
||||||
static void complete_init_numa_distance(void)
|
static void complete_init_numa_distance(MachineState *ms)
|
||||||
{
|
{
|
||||||
int src, dst;
|
int src, dst;
|
||||||
|
|
||||||
@ -303,8 +310,8 @@ static void complete_init_numa_distance(void)
|
|||||||
* there would not be any missing distance except local node, which
|
* there would not be any missing distance except local node, which
|
||||||
* is verified by validate_numa_distance above.
|
* is verified by validate_numa_distance above.
|
||||||
*/
|
*/
|
||||||
for (src = 0; src < nb_numa_nodes; src++) {
|
for (src = 0; src < ms->numa_state->num_nodes; src++) {
|
||||||
for (dst = 0; dst < nb_numa_nodes; dst++) {
|
for (dst = 0; dst < ms->numa_state->num_nodes; dst++) {
|
||||||
if (numa_info[src].distance[dst] == 0) {
|
if (numa_info[src].distance[dst] == 0) {
|
||||||
if (src == dst) {
|
if (src == dst) {
|
||||||
numa_info[src].distance[dst] = NUMA_DISTANCE_MIN;
|
numa_info[src].distance[dst] = NUMA_DISTANCE_MIN;
|
||||||
@ -370,7 +377,7 @@ void numa_complete_configuration(MachineState *ms)
|
|||||||
*
|
*
|
||||||
* Enable NUMA implicitly by adding a new NUMA node automatically.
|
* Enable NUMA implicitly by adding a new NUMA node automatically.
|
||||||
*/
|
*/
|
||||||
if (ms->ram_slots > 0 && nb_numa_nodes == 0 &&
|
if (ms->ram_slots > 0 && ms->numa_state->num_nodes == 0 &&
|
||||||
mc->auto_enable_numa_with_memhp) {
|
mc->auto_enable_numa_with_memhp) {
|
||||||
NumaNodeOptions node = { };
|
NumaNodeOptions node = { };
|
||||||
parse_numa_node(ms, &node, &error_abort);
|
parse_numa_node(ms, &node, &error_abort);
|
||||||
@ -388,26 +395,27 @@ void numa_complete_configuration(MachineState *ms)
|
|||||||
}
|
}
|
||||||
|
|
||||||
/* This must be always true if all nodes are present: */
|
/* This must be always true if all nodes are present: */
|
||||||
assert(nb_numa_nodes == max_numa_nodeid);
|
assert(ms->numa_state->num_nodes == max_numa_nodeid);
|
||||||
|
|
||||||
if (nb_numa_nodes > 0) {
|
if (ms->numa_state->num_nodes > 0) {
|
||||||
uint64_t numa_total;
|
uint64_t numa_total;
|
||||||
|
|
||||||
if (nb_numa_nodes > MAX_NODES) {
|
if (ms->numa_state->num_nodes > MAX_NODES) {
|
||||||
nb_numa_nodes = MAX_NODES;
|
ms->numa_state->num_nodes = MAX_NODES;
|
||||||
}
|
}
|
||||||
|
|
||||||
/* If no memory size is given for any node, assume the default case
|
/* If no memory size is given for any node, assume the default case
|
||||||
* and distribute the available memory equally across all nodes
|
* and distribute the available memory equally across all nodes
|
||||||
*/
|
*/
|
||||||
for (i = 0; i < nb_numa_nodes; i++) {
|
for (i = 0; i < ms->numa_state->num_nodes; i++) {
|
||||||
if (numa_info[i].node_mem != 0) {
|
if (numa_info[i].node_mem != 0) {
|
||||||
break;
|
break;
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
if (i == nb_numa_nodes) {
|
if (i == ms->numa_state->num_nodes) {
|
||||||
assert(mc->numa_auto_assign_ram);
|
assert(mc->numa_auto_assign_ram);
|
||||||
mc->numa_auto_assign_ram(mc, numa_info, nb_numa_nodes, ram_size);
|
mc->numa_auto_assign_ram(mc, numa_info,
|
||||||
|
ms->numa_state->num_nodes, ram_size);
|
||||||
if (!qtest_enabled()) {
|
if (!qtest_enabled()) {
|
||||||
warn_report("Default splitting of RAM between nodes is deprecated,"
|
warn_report("Default splitting of RAM between nodes is deprecated,"
|
||||||
" Use '-numa node,memdev' to explictly define RAM"
|
" Use '-numa node,memdev' to explictly define RAM"
|
||||||
@ -416,7 +424,7 @@ void numa_complete_configuration(MachineState *ms)
|
|||||||
}
|
}
|
||||||
|
|
||||||
numa_total = 0;
|
numa_total = 0;
|
||||||
for (i = 0; i < nb_numa_nodes; i++) {
|
for (i = 0; i < ms->numa_state->num_nodes; i++) {
|
||||||
numa_total += numa_info[i].node_mem;
|
numa_total += numa_info[i].node_mem;
|
||||||
}
|
}
|
||||||
if (numa_total != ram_size) {
|
if (numa_total != ram_size) {
|
||||||
@ -440,10 +448,10 @@ void numa_complete_configuration(MachineState *ms)
|
|||||||
*/
|
*/
|
||||||
if (have_numa_distance) {
|
if (have_numa_distance) {
|
||||||
/* Validate enough NUMA distance information was provided. */
|
/* Validate enough NUMA distance information was provided. */
|
||||||
validate_numa_distance();
|
validate_numa_distance(ms);
|
||||||
|
|
||||||
/* Validation succeeded, now fill in any missing distances. */
|
/* Validation succeeded, now fill in any missing distances. */
|
||||||
complete_init_numa_distance();
|
complete_init_numa_distance(ms);
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
@ -510,14 +518,16 @@ void memory_region_allocate_system_memory(MemoryRegion *mr, Object *owner,
|
|||||||
{
|
{
|
||||||
uint64_t addr = 0;
|
uint64_t addr = 0;
|
||||||
int i;
|
int i;
|
||||||
|
MachineState *ms = MACHINE(qdev_get_machine());
|
||||||
|
|
||||||
if (nb_numa_nodes == 0 || !have_memdevs) {
|
if (ms->numa_state == NULL ||
|
||||||
|
ms->numa_state->num_nodes == 0 || !have_memdevs) {
|
||||||
allocate_system_memory_nonnuma(mr, owner, name, ram_size);
|
allocate_system_memory_nonnuma(mr, owner, name, ram_size);
|
||||||
return;
|
return;
|
||||||
}
|
}
|
||||||
|
|
||||||
memory_region_init(mr, owner, name, ram_size);
|
memory_region_init(mr, owner, name, ram_size);
|
||||||
for (i = 0; i < nb_numa_nodes; i++) {
|
for (i = 0; i < ms->numa_state->num_nodes; i++) {
|
||||||
uint64_t size = numa_info[i].node_mem;
|
uint64_t size = numa_info[i].node_mem;
|
||||||
HostMemoryBackend *backend = numa_info[i].node_memdev;
|
HostMemoryBackend *backend = numa_info[i].node_memdev;
|
||||||
if (!backend) {
|
if (!backend) {
|
||||||
@ -575,16 +585,16 @@ static void numa_stat_memory_devices(NumaNodeMem node_mem[])
|
|||||||
qapi_free_MemoryDeviceInfoList(info_list);
|
qapi_free_MemoryDeviceInfoList(info_list);
|
||||||
}
|
}
|
||||||
|
|
||||||
void query_numa_node_mem(NumaNodeMem node_mem[])
|
void query_numa_node_mem(NumaNodeMem node_mem[], MachineState *ms)
|
||||||
{
|
{
|
||||||
int i;
|
int i;
|
||||||
|
|
||||||
if (nb_numa_nodes <= 0) {
|
if (ms->numa_state == NULL || ms->numa_state->num_nodes <= 0) {
|
||||||
return;
|
return;
|
||||||
}
|
}
|
||||||
|
|
||||||
numa_stat_memory_devices(node_mem);
|
numa_stat_memory_devices(node_mem);
|
||||||
for (i = 0; i < nb_numa_nodes; i++) {
|
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 += numa_info[i].node_mem;
|
||||||
}
|
}
|
||||||
}
|
}
|
||||||
|
@ -2696,7 +2696,7 @@ void acpi_build(AcpiBuildTables *tables, MachineState *machine)
|
|||||||
build_srat(tables_blob, tables->linker, machine);
|
build_srat(tables_blob, tables->linker, machine);
|
||||||
if (have_numa_distance) {
|
if (have_numa_distance) {
|
||||||
acpi_add_table(table_offsets, tables_blob);
|
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)) {
|
if (acpi_get_mcfg(&mcfg)) {
|
||||||
|
@ -1002,6 +1002,8 @@ static FWCfgState *bochs_bios_init(AddressSpace *as, PCMachineState *pcms)
|
|||||||
int i;
|
int i;
|
||||||
const CPUArchIdList *cpus;
|
const CPUArchIdList *cpus;
|
||||||
MachineClass *mc = MACHINE_GET_CLASS(pcms);
|
MachineClass *mc = MACHINE_GET_CLASS(pcms);
|
||||||
|
MachineState *ms = MACHINE(pcms);
|
||||||
|
int nb_numa_nodes = ms->numa_state->num_nodes;
|
||||||
|
|
||||||
fw_cfg = fw_cfg_init_io_dma(FW_CFG_IO_BASE, FW_CFG_IO_BASE + 4, as);
|
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);
|
fw_cfg_add_i16(fw_cfg, FW_CFG_NB_CPUS, pcms->boot_cpus);
|
||||||
@ -1774,12 +1776,13 @@ void pc_machine_done(Notifier *notifier, void *data)
|
|||||||
void pc_guest_info_init(PCMachineState *pcms)
|
void pc_guest_info_init(PCMachineState *pcms)
|
||||||
{
|
{
|
||||||
int i;
|
int i;
|
||||||
|
MachineState *ms = MACHINE(pcms);
|
||||||
|
|
||||||
pcms->apic_xrupt_override = kvm_allows_irq0_override();
|
pcms->apic_xrupt_override = kvm_allows_irq0_override();
|
||||||
pcms->numa_nodes = nb_numa_nodes;
|
pcms->numa_nodes = ms->numa_state->num_nodes;
|
||||||
pcms->node_mem = g_malloc0(pcms->numa_nodes *
|
pcms->node_mem = g_malloc0(pcms->numa_nodes *
|
||||||
sizeof *pcms->node_mem);
|
sizeof *pcms->node_mem);
|
||||||
for (i = 0; i < nb_numa_nodes; i++) {
|
for (i = 0; i < ms->numa_state->num_nodes; i++) {
|
||||||
pcms->node_mem[i] = numa_info[i].node_mem;
|
pcms->node_mem[i] = numa_info[i].node_mem;
|
||||||
}
|
}
|
||||||
|
|
||||||
@ -2869,7 +2872,7 @@ static int64_t pc_get_default_cpu_node_id(const MachineState *ms, int idx)
|
|||||||
x86_topo_ids_from_apicid(ms->possible_cpus->cpus[idx].arch_id,
|
x86_topo_ids_from_apicid(ms->possible_cpus->cpus[idx].arch_id,
|
||||||
pcms->smp_dies, ms->smp.cores,
|
pcms->smp_dies, ms->smp.cores,
|
||||||
ms->smp.threads, &topo);
|
ms->smp.threads, &topo);
|
||||||
return topo.pkg_id % nb_numa_nodes;
|
return topo.pkg_id % ms->numa_state->num_nodes;
|
||||||
}
|
}
|
||||||
|
|
||||||
static const CPUArchIdList *pc_possible_cpu_arch_ids(MachineState *ms)
|
static const CPUArchIdList *pc_possible_cpu_arch_ids(MachineState *ms)
|
||||||
|
@ -172,6 +172,8 @@ static void pc_dimm_realize(DeviceState *dev, Error **errp)
|
|||||||
{
|
{
|
||||||
PCDIMMDevice *dimm = PC_DIMM(dev);
|
PCDIMMDevice *dimm = PC_DIMM(dev);
|
||||||
PCDIMMDeviceClass *ddc = PC_DIMM_GET_CLASS(dimm);
|
PCDIMMDeviceClass *ddc = PC_DIMM_GET_CLASS(dimm);
|
||||||
|
MachineState *ms = MACHINE(qdev_get_machine());
|
||||||
|
int nb_numa_nodes = ms->numa_state->num_nodes;
|
||||||
|
|
||||||
if (!dimm->hostmem) {
|
if (!dimm->hostmem) {
|
||||||
error_setg(errp, "'" PC_DIMM_MEMDEV_PROP "' property is not set");
|
error_setg(errp, "'" PC_DIMM_MEMDEV_PROP "' property is not set");
|
||||||
|
@ -21,6 +21,7 @@
|
|||||||
#include "qemu/error-report.h"
|
#include "qemu/error-report.h"
|
||||||
#include "qemu/module.h"
|
#include "qemu/module.h"
|
||||||
#include "sysemu/numa.h"
|
#include "sysemu/numa.h"
|
||||||
|
#include "hw/boards.h"
|
||||||
|
|
||||||
#define TYPE_PXB_BUS "pxb-bus"
|
#define TYPE_PXB_BUS "pxb-bus"
|
||||||
#define PXB_BUS(obj) OBJECT_CHECK(PXBBus, (obj), TYPE_PXB_BUS)
|
#define PXB_BUS(obj) OBJECT_CHECK(PXBBus, (obj), TYPE_PXB_BUS)
|
||||||
@ -213,9 +214,15 @@ static void pxb_dev_realize_common(PCIDevice *dev, bool pcie, Error **errp)
|
|||||||
PCIBus *bus;
|
PCIBus *bus;
|
||||||
const char *dev_name = NULL;
|
const char *dev_name = NULL;
|
||||||
Error *local_err = NULL;
|
Error *local_err = NULL;
|
||||||
|
MachineState *ms = MACHINE(qdev_get_machine());
|
||||||
|
|
||||||
|
if (ms->numa_state == NULL) {
|
||||||
|
error_setg(errp, "NUMA is not supported by this machine-type");
|
||||||
|
return;
|
||||||
|
}
|
||||||
|
|
||||||
if (pxb->numa_node != NUMA_NODE_UNASSIGNED &&
|
if (pxb->numa_node != NUMA_NODE_UNASSIGNED &&
|
||||||
pxb->numa_node >= nb_numa_nodes) {
|
pxb->numa_node >= ms->numa_state->num_nodes) {
|
||||||
error_setg(errp, "Illegal numa node %d", pxb->numa_node);
|
error_setg(errp, "Illegal numa node %d", pxb->numa_node);
|
||||||
return;
|
return;
|
||||||
}
|
}
|
||||||
|
@ -336,7 +336,7 @@ static int spapr_fixup_cpu_dt(void *fdt, SpaprMachineState *spapr)
|
|||||||
return ret;
|
return ret;
|
||||||
}
|
}
|
||||||
|
|
||||||
if (nb_numa_nodes > 1) {
|
if (ms->numa_state->num_nodes > 1) {
|
||||||
ret = spapr_fixup_cpu_numa_dt(fdt, offset, cpu);
|
ret = spapr_fixup_cpu_numa_dt(fdt, offset, cpu);
|
||||||
if (ret < 0) {
|
if (ret < 0) {
|
||||||
return ret;
|
return ret;
|
||||||
@ -356,9 +356,9 @@ static int spapr_fixup_cpu_dt(void *fdt, SpaprMachineState *spapr)
|
|||||||
|
|
||||||
static hwaddr spapr_node0_size(MachineState *machine)
|
static hwaddr spapr_node0_size(MachineState *machine)
|
||||||
{
|
{
|
||||||
if (nb_numa_nodes) {
|
if (machine->numa_state->num_nodes) {
|
||||||
int i;
|
int i;
|
||||||
for (i = 0; i < nb_numa_nodes; ++i) {
|
for (i = 0; i < machine->numa_state->num_nodes; ++i) {
|
||||||
if (numa_info[i].node_mem) {
|
if (numa_info[i].node_mem) {
|
||||||
return MIN(pow2floor(numa_info[i].node_mem),
|
return MIN(pow2floor(numa_info[i].node_mem),
|
||||||
machine->ram_size);
|
machine->ram_size);
|
||||||
@ -403,12 +403,12 @@ static int spapr_populate_memory(SpaprMachineState *spapr, void *fdt)
|
|||||||
{
|
{
|
||||||
MachineState *machine = MACHINE(spapr);
|
MachineState *machine = MACHINE(spapr);
|
||||||
hwaddr mem_start, node_size;
|
hwaddr mem_start, node_size;
|
||||||
int i, nb_nodes = nb_numa_nodes;
|
int i, nb_nodes = machine->numa_state->num_nodes;
|
||||||
NodeInfo *nodes = numa_info;
|
NodeInfo *nodes = numa_info;
|
||||||
NodeInfo ramnode;
|
NodeInfo ramnode;
|
||||||
|
|
||||||
/* No NUMA nodes, assume there is just one node with whole RAM */
|
/* No NUMA nodes, assume there is just one node with whole RAM */
|
||||||
if (!nb_numa_nodes) {
|
if (!nb_nodes) {
|
||||||
nb_nodes = 1;
|
nb_nodes = 1;
|
||||||
ramnode.node_mem = machine->ram_size;
|
ramnode.node_mem = machine->ram_size;
|
||||||
nodes = &ramnode;
|
nodes = &ramnode;
|
||||||
@ -559,7 +559,7 @@ static void spapr_populate_cpu_dt(CPUState *cs, void *fdt, int offset,
|
|||||||
_FDT((fdt_setprop(fdt, offset, "ibm,pft-size",
|
_FDT((fdt_setprop(fdt, offset, "ibm,pft-size",
|
||||||
pft_size_prop, sizeof(pft_size_prop))));
|
pft_size_prop, sizeof(pft_size_prop))));
|
||||||
|
|
||||||
if (nb_numa_nodes > 1) {
|
if (ms->numa_state->num_nodes > 1) {
|
||||||
_FDT(spapr_fixup_cpu_numa_dt(fdt, offset, cpu));
|
_FDT(spapr_fixup_cpu_numa_dt(fdt, offset, cpu));
|
||||||
}
|
}
|
||||||
|
|
||||||
@ -866,6 +866,7 @@ static int spapr_populate_drmem_v1(SpaprMachineState *spapr, void *fdt,
|
|||||||
static int spapr_populate_drconf_memory(SpaprMachineState *spapr, void *fdt)
|
static int spapr_populate_drconf_memory(SpaprMachineState *spapr, void *fdt)
|
||||||
{
|
{
|
||||||
MachineState *machine = MACHINE(spapr);
|
MachineState *machine = MACHINE(spapr);
|
||||||
|
int nb_numa_nodes = machine->numa_state->num_nodes;
|
||||||
int ret, i, offset;
|
int ret, i, offset;
|
||||||
uint64_t lmb_size = SPAPR_MEMORY_BLOCK_SIZE;
|
uint64_t lmb_size = SPAPR_MEMORY_BLOCK_SIZE;
|
||||||
uint32_t prop_lmb_size[] = {0, cpu_to_be32(lmb_size)};
|
uint32_t prop_lmb_size[] = {0, cpu_to_be32(lmb_size)};
|
||||||
@ -1741,7 +1742,7 @@ static void spapr_machine_reset(MachineState *machine)
|
|||||||
* The final value of spapr->gpu_numa_id is going to be written to
|
* The final value of spapr->gpu_numa_id is going to be written to
|
||||||
* max-associativity-domains in spapr_build_fdt().
|
* max-associativity-domains in spapr_build_fdt().
|
||||||
*/
|
*/
|
||||||
spapr->gpu_numa_id = MAX(1, nb_numa_nodes);
|
spapr->gpu_numa_id = MAX(1, machine->numa_state->num_nodes);
|
||||||
qemu_devices_reset();
|
qemu_devices_reset();
|
||||||
|
|
||||||
/*
|
/*
|
||||||
@ -2539,7 +2540,7 @@ static void spapr_validate_node_memory(MachineState *machine, Error **errp)
|
|||||||
return;
|
return;
|
||||||
}
|
}
|
||||||
|
|
||||||
for (i = 0; i < nb_numa_nodes; i++) {
|
for (i = 0; i < machine->numa_state->num_nodes; i++) {
|
||||||
if (numa_info[i].node_mem % SPAPR_MEMORY_BLOCK_SIZE) {
|
if (numa_info[i].node_mem % SPAPR_MEMORY_BLOCK_SIZE) {
|
||||||
error_setg(errp,
|
error_setg(errp,
|
||||||
"Node %d memory size 0x%" PRIx64
|
"Node %d memory size 0x%" PRIx64
|
||||||
@ -4178,7 +4179,7 @@ spapr_cpu_index_to_props(MachineState *machine, unsigned cpu_index)
|
|||||||
|
|
||||||
static int64_t spapr_get_default_cpu_node_id(const MachineState *ms, int idx)
|
static int64_t spapr_get_default_cpu_node_id(const MachineState *ms, int idx)
|
||||||
{
|
{
|
||||||
return idx / ms->smp.cores % nb_numa_nodes;
|
return idx / ms->smp.cores % ms->numa_state->num_nodes;
|
||||||
}
|
}
|
||||||
|
|
||||||
static const CPUArchIdList *spapr_possible_cpu_arch_ids(MachineState *machine)
|
static const CPUArchIdList *spapr_possible_cpu_arch_ids(MachineState *machine)
|
||||||
|
@ -414,7 +414,7 @@ build_append_gas_from_struct(GArray *table, const struct AcpiGenericAddress *s)
|
|||||||
void build_srat_memory(AcpiSratMemoryAffinity *numamem, uint64_t base,
|
void build_srat_memory(AcpiSratMemoryAffinity *numamem, uint64_t base,
|
||||||
uint64_t len, int node, MemoryAffinityFlags flags);
|
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,
|
void build_fadt(GArray *tbl, BIOSLinker *linker, const AcpiFadtData *f,
|
||||||
const char *oem_id, const char *oem_table_id);
|
const char *oem_id, const char *oem_table_id);
|
||||||
|
@ -299,6 +299,7 @@ struct MachineState {
|
|||||||
CPUArchIdList *possible_cpus;
|
CPUArchIdList *possible_cpus;
|
||||||
CpuTopology smp;
|
CpuTopology smp;
|
||||||
struct NVDIMMState *nvdimms_state;
|
struct NVDIMMState *nvdimms_state;
|
||||||
|
struct NumaState *numa_state;
|
||||||
};
|
};
|
||||||
|
|
||||||
#define DEFINE_MACHINE(namestr, machine_initfn) \
|
#define DEFINE_MACHINE(namestr, machine_initfn) \
|
||||||
|
@ -14,7 +14,6 @@ struct CPUArchId;
|
|||||||
#define NUMA_DISTANCE_MAX 254
|
#define NUMA_DISTANCE_MAX 254
|
||||||
#define NUMA_DISTANCE_UNREACHABLE 255
|
#define NUMA_DISTANCE_UNREACHABLE 255
|
||||||
|
|
||||||
extern int nb_numa_nodes; /* Number of NUMA nodes */
|
|
||||||
extern bool have_numa_distance;
|
extern bool have_numa_distance;
|
||||||
|
|
||||||
struct NodeInfo {
|
struct NodeInfo {
|
||||||
@ -31,10 +30,17 @@ struct NumaNodeMem {
|
|||||||
|
|
||||||
extern NodeInfo numa_info[MAX_NODES];
|
extern NodeInfo numa_info[MAX_NODES];
|
||||||
|
|
||||||
|
struct NumaState {
|
||||||
|
/* Number of NUMA nodes */
|
||||||
|
int num_nodes;
|
||||||
|
|
||||||
|
};
|
||||||
|
typedef struct NumaState NumaState;
|
||||||
|
|
||||||
void set_numa_options(MachineState *ms, NumaOptions *object, Error **errp);
|
void set_numa_options(MachineState *ms, NumaOptions *object, Error **errp);
|
||||||
void parse_numa_opts(MachineState *ms);
|
void parse_numa_opts(MachineState *ms);
|
||||||
void numa_complete_configuration(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;
|
extern QemuOptsList qemu_numa_opts;
|
||||||
void numa_legacy_auto_assign_ram(MachineClass *mc, NodeInfo *nodes,
|
void numa_legacy_auto_assign_ram(MachineClass *mc, NodeInfo *nodes,
|
||||||
int nb_nodes, ram_addr_t size);
|
int nb_nodes, ram_addr_t size);
|
||||||
|
Loading…
Reference in New Issue
Block a user