mirrors/qemu
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
3b64349539
qemu/include/sysemu/numa.h

26 lines
625 B
C
Raw Normal View History

numa: Move NUMA declarations from sysemu.h to numa.h Not all sysemu.h users need the NUMA declarations, and keeping them in a separate file makes it easier to see what are the interfaces provided by numa.c. Reviewed-by: Paolo Bonzini <pbonzini@redhat.com> Acked-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
2015-02-08 21:51:16 +03:00
#ifndef SYSEMU_NUMA_H
#define SYSEMU_NUMA_H
#include <stdint.h>
#include "qemu/bitmap.h"
#include "qemu/option.h"
#include "sysemu/sysemu.h"
#include "sysemu/hostmem.h"
numa: introduce machine callback for VCPU to node mapping Current default round-robin way of distributing VCPUs among NUMA nodes might be wrong in case on multi-core/threads CPUs. Making guests confused wrt topology where cores from the same socket are on different nodes. Allow a machine to override default mapping by providing MachineClass::cpu_index_to_socket_id() callback which would allow it group VCPUs from a socket on the same NUMA node. Signed-off-by: Igor Mammedov <imammedo@redhat.com> Reviewed-by: Andreas Färber <afaerber@suse.de> Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
2015-03-19 20:09:21 +03:00
#include "hw/boards.h"
numa: Move NUMA declarations from sysemu.h to numa.h Not all sysemu.h users need the NUMA declarations, and keeping them in a separate file makes it easier to see what are the interfaces provided by numa.c. Reviewed-by: Paolo Bonzini <pbonzini@redhat.com> Acked-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
2015-02-08 21:51:16 +03:00
extern int nb_numa_nodes; /* Number of NUMA nodes */
typedef struct node_info {
uint64_t node_mem;
DECLARE_BITMAP(node_cpu, MAX_CPUMASK_BITS);
struct HostMemoryBackend *node_memdev;
bool present;
} NodeInfo;
extern NodeInfo numa_info[MAX_NODES];
numa: introduce machine callback for VCPU to node mapping Current default round-robin way of distributing VCPUs among NUMA nodes might be wrong in case on multi-core/threads CPUs. Making guests confused wrt topology where cores from the same socket are on different nodes. Allow a machine to override default mapping by providing MachineClass::cpu_index_to_socket_id() callback which would allow it group VCPUs from a socket on the same NUMA node. Signed-off-by: Igor Mammedov <imammedo@redhat.com> Reviewed-by: Andreas Färber <afaerber@suse.de> Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
2015-03-19 20:09:21 +03:00
void parse_numa_opts(MachineClass *mc);
numa: Rename set_numa_modes() to numa_post_machine_init() This function does some initialization that needs to be done after machine init. The function may be eventually removed if we move the CPUState.numa_node initialization to the CPU init code, but while the function exists, lets give it a name that makes sense. Reviewed-by: Paolo Bonzini <pbonzini@redhat.com> Acked-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
2015-02-08 21:51:22 +03:00
void numa_post_machine_init(void);
numa: Move NUMA declarations from sysemu.h to numa.h Not all sysemu.h users need the NUMA declarations, and keeping them in a separate file makes it easier to see what are the interfaces provided by numa.c. Reviewed-by: Paolo Bonzini <pbonzini@redhat.com> Acked-by: Michael S. Tsirkin <mst@redhat.com> Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
2015-02-08 21:51:16 +03:00
void query_numa_node_mem(uint64_t node_mem[]);
extern QemuOptsList qemu_numa_opts;
#endif
Reference in New Issue Copy Permalink