qemu/backends/hostmem.c

466 lines
14 KiB
C
Raw Normal View History

/*
* QEMU Host Memory Backend
*
* Copyright (C) 2013-2014 Red Hat Inc
*
* Authors:
* Igor Mammedov <imammedo@redhat.com>
*
* This work is licensed under the terms of the GNU GPL, version 2 or later.
* See the COPYING file in the top-level directory.
*/
#include "qemu/osdep.h"
#include "sysemu/hostmem.h"
#include "hw/boards.h"
2016-03-14 11:01:28 +03:00
#include "qapi/error.h"
#include "qapi/qapi-builtin-visit.h"
#include "qapi/visitor.h"
#include "qemu/config-file.h"
#include "qom/object_interfaces.h"
#include "qemu/mmap-alloc.h"
#ifdef CONFIG_NUMA
#include <numaif.h>
QEMU_BUILD_BUG_ON(HOST_MEM_POLICY_DEFAULT != MPOL_DEFAULT);
QEMU_BUILD_BUG_ON(HOST_MEM_POLICY_PREFERRED != MPOL_PREFERRED);
QEMU_BUILD_BUG_ON(HOST_MEM_POLICY_BIND != MPOL_BIND);
QEMU_BUILD_BUG_ON(HOST_MEM_POLICY_INTERLEAVE != MPOL_INTERLEAVE);
#endif
static void
host_memory_backend_get_size(Object *obj, Visitor *v, const char *name,
void *opaque, Error **errp)
{
HostMemoryBackend *backend = MEMORY_BACKEND(obj);
uint64_t value = backend->size;
qapi: Swap visit_* arguments for consistent 'name' placement JSON uses "name":value, but many of our visitor interfaces were called with visit_type_FOO(v, &value, name, errp). This can be a bit confusing to have to mentally swap the parameter order to match JSON order. It's particularly bad for visit_start_struct(), where the 'name' parameter is smack in the middle of the otherwise-related group of 'obj, kind, size' parameters! It's time to do a global swap of the parameter ordering, so that the 'name' parameter is always immediately after the Visitor argument. Additional reason in favor of the swap: the existing include/qjson.h prefers listing 'name' first in json_prop_*(), and I have plans to unify that file with the qapi visitors; listing 'name' first in qapi will minimize churn to the (admittedly few) qjson.h clients. Later patches will then fix docs, object.h, visitor-impl.h, and those clients to match. Done by first patching scripts/qapi*.py by hand to make generated files do what I want, then by running the following Coccinelle script to affect the rest of the code base: $ spatch --sp-file script `git grep -l '\bvisit_' -- '**/*.[ch]'` I then had to apply some touchups (Coccinelle insisted on TAB indentation in visitor.h, and botched the signature of visit_type_enum() by rewriting 'const char *const strings[]' to the syntactically invalid 'const char*const[] strings'). The movement of parameters is sufficient to provoke compiler errors if any callers were missed. // Part 1: Swap declaration order @@ type TV, TErr, TObj, T1, T2; identifier OBJ, ARG1, ARG2; @@ void visit_start_struct -(TV v, TObj OBJ, T1 ARG1, const char *name, T2 ARG2, TErr errp) +(TV v, const char *name, TObj OBJ, T1 ARG1, T2 ARG2, TErr errp) { ... } @@ type bool, TV, T1; identifier ARG1; @@ bool visit_optional -(TV v, T1 ARG1, const char *name) +(TV v, const char *name, T1 ARG1) { ... } @@ type TV, TErr, TObj, T1; identifier OBJ, ARG1; @@ void visit_get_next_type -(TV v, TObj OBJ, T1 ARG1, const char *name, TErr errp) +(TV v, const char *name, TObj OBJ, T1 ARG1, TErr errp) { ... } @@ type TV, TErr, TObj, T1, T2; identifier OBJ, ARG1, ARG2; @@ void visit_type_enum -(TV v, TObj OBJ, T1 ARG1, T2 ARG2, const char *name, TErr errp) +(TV v, const char *name, TObj OBJ, T1 ARG1, T2 ARG2, TErr errp) { ... } @@ type TV, TErr, TObj; identifier OBJ; identifier VISIT_TYPE =~ "^visit_type_"; @@ void VISIT_TYPE -(TV v, TObj OBJ, const char *name, TErr errp) +(TV v, const char *name, TObj OBJ, TErr errp) { ... } // Part 2: swap caller order @@ expression V, NAME, OBJ, ARG1, ARG2, ERR; identifier VISIT_TYPE =~ "^visit_type_"; @@ ( -visit_start_struct(V, OBJ, ARG1, NAME, ARG2, ERR) +visit_start_struct(V, NAME, OBJ, ARG1, ARG2, ERR) | -visit_optional(V, ARG1, NAME) +visit_optional(V, NAME, ARG1) | -visit_get_next_type(V, OBJ, ARG1, NAME, ERR) +visit_get_next_type(V, NAME, OBJ, ARG1, ERR) | -visit_type_enum(V, OBJ, ARG1, ARG2, NAME, ERR) +visit_type_enum(V, NAME, OBJ, ARG1, ARG2, ERR) | -VISIT_TYPE(V, OBJ, NAME, ERR) +VISIT_TYPE(V, NAME, OBJ, ERR) ) Signed-off-by: Eric Blake <eblake@redhat.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <1454075341-13658-19-git-send-email-eblake@redhat.com> Signed-off-by: Markus Armbruster <armbru@redhat.com>
2016-01-29 16:48:54 +03:00
visit_type_size(v, name, &value, errp);
}
static void
host_memory_backend_set_size(Object *obj, Visitor *v, const char *name,
void *opaque, Error **errp)
{
HostMemoryBackend *backend = MEMORY_BACKEND(obj);
Error *local_err = NULL;
uint64_t value;
if (host_memory_backend_mr_inited(backend)) {
error_setg(&local_err, "cannot change property value");
goto out;
}
qapi: Swap visit_* arguments for consistent 'name' placement JSON uses "name":value, but many of our visitor interfaces were called with visit_type_FOO(v, &value, name, errp). This can be a bit confusing to have to mentally swap the parameter order to match JSON order. It's particularly bad for visit_start_struct(), where the 'name' parameter is smack in the middle of the otherwise-related group of 'obj, kind, size' parameters! It's time to do a global swap of the parameter ordering, so that the 'name' parameter is always immediately after the Visitor argument. Additional reason in favor of the swap: the existing include/qjson.h prefers listing 'name' first in json_prop_*(), and I have plans to unify that file with the qapi visitors; listing 'name' first in qapi will minimize churn to the (admittedly few) qjson.h clients. Later patches will then fix docs, object.h, visitor-impl.h, and those clients to match. Done by first patching scripts/qapi*.py by hand to make generated files do what I want, then by running the following Coccinelle script to affect the rest of the code base: $ spatch --sp-file script `git grep -l '\bvisit_' -- '**/*.[ch]'` I then had to apply some touchups (Coccinelle insisted on TAB indentation in visitor.h, and botched the signature of visit_type_enum() by rewriting 'const char *const strings[]' to the syntactically invalid 'const char*const[] strings'). The movement of parameters is sufficient to provoke compiler errors if any callers were missed. // Part 1: Swap declaration order @@ type TV, TErr, TObj, T1, T2; identifier OBJ, ARG1, ARG2; @@ void visit_start_struct -(TV v, TObj OBJ, T1 ARG1, const char *name, T2 ARG2, TErr errp) +(TV v, const char *name, TObj OBJ, T1 ARG1, T2 ARG2, TErr errp) { ... } @@ type bool, TV, T1; identifier ARG1; @@ bool visit_optional -(TV v, T1 ARG1, const char *name) +(TV v, const char *name, T1 ARG1) { ... } @@ type TV, TErr, TObj, T1; identifier OBJ, ARG1; @@ void visit_get_next_type -(TV v, TObj OBJ, T1 ARG1, const char *name, TErr errp) +(TV v, const char *name, TObj OBJ, T1 ARG1, TErr errp) { ... } @@ type TV, TErr, TObj, T1, T2; identifier OBJ, ARG1, ARG2; @@ void visit_type_enum -(TV v, TObj OBJ, T1 ARG1, T2 ARG2, const char *name, TErr errp) +(TV v, const char *name, TObj OBJ, T1 ARG1, T2 ARG2, TErr errp) { ... } @@ type TV, TErr, TObj; identifier OBJ; identifier VISIT_TYPE =~ "^visit_type_"; @@ void VISIT_TYPE -(TV v, TObj OBJ, const char *name, TErr errp) +(TV v, const char *name, TObj OBJ, TErr errp) { ... } // Part 2: swap caller order @@ expression V, NAME, OBJ, ARG1, ARG2, ERR; identifier VISIT_TYPE =~ "^visit_type_"; @@ ( -visit_start_struct(V, OBJ, ARG1, NAME, ARG2, ERR) +visit_start_struct(V, NAME, OBJ, ARG1, ARG2, ERR) | -visit_optional(V, ARG1, NAME) +visit_optional(V, NAME, ARG1) | -visit_get_next_type(V, OBJ, ARG1, NAME, ERR) +visit_get_next_type(V, NAME, OBJ, ARG1, ERR) | -visit_type_enum(V, OBJ, ARG1, ARG2, NAME, ERR) +visit_type_enum(V, NAME, OBJ, ARG1, ARG2, ERR) | -VISIT_TYPE(V, OBJ, NAME, ERR) +VISIT_TYPE(V, NAME, OBJ, ERR) ) Signed-off-by: Eric Blake <eblake@redhat.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <1454075341-13658-19-git-send-email-eblake@redhat.com> Signed-off-by: Markus Armbruster <armbru@redhat.com>
2016-01-29 16:48:54 +03:00
visit_type_size(v, name, &value, &local_err);
if (local_err) {
goto out;
}
if (!value) {
error_setg(&local_err, "Property '%s.%s' doesn't take value '%"
PRIu64 "'", object_get_typename(obj), name, value);
goto out;
}
backend->size = value;
out:
error_propagate(errp, local_err);
}
static void
host_memory_backend_get_host_nodes(Object *obj, Visitor *v, const char *name,
void *opaque, Error **errp)
{
HostMemoryBackend *backend = MEMORY_BACKEND(obj);
uint16List *host_nodes = NULL;
uint16List **node = &host_nodes;
unsigned long value;
value = find_first_bit(backend->host_nodes, MAX_NODES);
if (value == MAX_NODES) {
Revert "hostmem: fix QEMU crash by 'info memdev'" This reverts commit 1454d33f0507cb54d62ed80f494884157c9e7130. The string input visitor regression fixed in the previous commit made visit_type_uint16List() fail on empty input. query_memdev() calls it via object_property_get_uint16List(). Because it doesn't expect it to fail, it passes &error_abort, and duly crashes. Commit 1454d33 "fixes" this crash by making host_memory_backend_get_host_nodes() return a list containing just MAX_NODES instead of the empty list. Papers over the regression, and leads to bogus "info memdev" output, as shown below; revert. I suspect that if we had bisected the crash back then, we would have found and fixed the actual bug instead of papering over it. To reproduce, run HMP command "info memdev" with $ qemu-system-x86_64 --nodefaults -S -display none -monitor stdio -object memory-backend-ram,id=mem1,size=4k With this commit, "info memdev" prints memory backend: mem1 size: 4096 merge: true dump: true prealloc: false policy: default host nodes: exactly like before commit 74f24cb. Between commit 1454d33 and this commit, it prints memory backend: mem1 size: 4096 merge: true dump: true prealloc: false policy: default host nodes: 128 The last line is bogus. Between commit 74f24cb and 1454d33, it crashes like this: Unexpected error in parse_str() at /work/armbru/tmp/qemu/qapi/string-input-visitor.c:126: Parameter 'null' expects an int64 value or range Aborted (core dumped) Cc: Xiao Guangrong <guangrong.xiao@linux.intel.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Markus Armbruster <armbru@redhat.com> Message-Id: <1490026424-11330-3-git-send-email-armbru@redhat.com> Reviewed-by: Michael Roth <mdroth@linux.vnet.ibm.com> Reviewed-by: Eric Blake <eblake@redhat.com>
2017-03-20 19:13:44 +03:00
return;
}
Revert "hostmem: fix QEMU crash by 'info memdev'" This reverts commit 1454d33f0507cb54d62ed80f494884157c9e7130. The string input visitor regression fixed in the previous commit made visit_type_uint16List() fail on empty input. query_memdev() calls it via object_property_get_uint16List(). Because it doesn't expect it to fail, it passes &error_abort, and duly crashes. Commit 1454d33 "fixes" this crash by making host_memory_backend_get_host_nodes() return a list containing just MAX_NODES instead of the empty list. Papers over the regression, and leads to bogus "info memdev" output, as shown below; revert. I suspect that if we had bisected the crash back then, we would have found and fixed the actual bug instead of papering over it. To reproduce, run HMP command "info memdev" with $ qemu-system-x86_64 --nodefaults -S -display none -monitor stdio -object memory-backend-ram,id=mem1,size=4k With this commit, "info memdev" prints memory backend: mem1 size: 4096 merge: true dump: true prealloc: false policy: default host nodes: exactly like before commit 74f24cb. Between commit 1454d33 and this commit, it prints memory backend: mem1 size: 4096 merge: true dump: true prealloc: false policy: default host nodes: 128 The last line is bogus. Between commit 74f24cb and 1454d33, it crashes like this: Unexpected error in parse_str() at /work/armbru/tmp/qemu/qapi/string-input-visitor.c:126: Parameter 'null' expects an int64 value or range Aborted (core dumped) Cc: Xiao Guangrong <guangrong.xiao@linux.intel.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Markus Armbruster <armbru@redhat.com> Message-Id: <1490026424-11330-3-git-send-email-armbru@redhat.com> Reviewed-by: Michael Roth <mdroth@linux.vnet.ibm.com> Reviewed-by: Eric Blake <eblake@redhat.com>
2017-03-20 19:13:44 +03:00
*node = g_malloc0(sizeof(**node));
(*node)->value = value;
node = &(*node)->next;
do {
value = find_next_bit(backend->host_nodes, MAX_NODES, value + 1);
if (value == MAX_NODES) {
break;
}
Revert "hostmem: fix QEMU crash by 'info memdev'" This reverts commit 1454d33f0507cb54d62ed80f494884157c9e7130. The string input visitor regression fixed in the previous commit made visit_type_uint16List() fail on empty input. query_memdev() calls it via object_property_get_uint16List(). Because it doesn't expect it to fail, it passes &error_abort, and duly crashes. Commit 1454d33 "fixes" this crash by making host_memory_backend_get_host_nodes() return a list containing just MAX_NODES instead of the empty list. Papers over the regression, and leads to bogus "info memdev" output, as shown below; revert. I suspect that if we had bisected the crash back then, we would have found and fixed the actual bug instead of papering over it. To reproduce, run HMP command "info memdev" with $ qemu-system-x86_64 --nodefaults -S -display none -monitor stdio -object memory-backend-ram,id=mem1,size=4k With this commit, "info memdev" prints memory backend: mem1 size: 4096 merge: true dump: true prealloc: false policy: default host nodes: exactly like before commit 74f24cb. Between commit 1454d33 and this commit, it prints memory backend: mem1 size: 4096 merge: true dump: true prealloc: false policy: default host nodes: 128 The last line is bogus. Between commit 74f24cb and 1454d33, it crashes like this: Unexpected error in parse_str() at /work/armbru/tmp/qemu/qapi/string-input-visitor.c:126: Parameter 'null' expects an int64 value or range Aborted (core dumped) Cc: Xiao Guangrong <guangrong.xiao@linux.intel.com> Cc: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Markus Armbruster <armbru@redhat.com> Message-Id: <1490026424-11330-3-git-send-email-armbru@redhat.com> Reviewed-by: Michael Roth <mdroth@linux.vnet.ibm.com> Reviewed-by: Eric Blake <eblake@redhat.com>
2017-03-20 19:13:44 +03:00
*node = g_malloc0(sizeof(**node));
(*node)->value = value;
node = &(*node)->next;
} while (true);
qapi: Swap visit_* arguments for consistent 'name' placement JSON uses "name":value, but many of our visitor interfaces were called with visit_type_FOO(v, &value, name, errp). This can be a bit confusing to have to mentally swap the parameter order to match JSON order. It's particularly bad for visit_start_struct(), where the 'name' parameter is smack in the middle of the otherwise-related group of 'obj, kind, size' parameters! It's time to do a global swap of the parameter ordering, so that the 'name' parameter is always immediately after the Visitor argument. Additional reason in favor of the swap: the existing include/qjson.h prefers listing 'name' first in json_prop_*(), and I have plans to unify that file with the qapi visitors; listing 'name' first in qapi will minimize churn to the (admittedly few) qjson.h clients. Later patches will then fix docs, object.h, visitor-impl.h, and those clients to match. Done by first patching scripts/qapi*.py by hand to make generated files do what I want, then by running the following Coccinelle script to affect the rest of the code base: $ spatch --sp-file script `git grep -l '\bvisit_' -- '**/*.[ch]'` I then had to apply some touchups (Coccinelle insisted on TAB indentation in visitor.h, and botched the signature of visit_type_enum() by rewriting 'const char *const strings[]' to the syntactically invalid 'const char*const[] strings'). The movement of parameters is sufficient to provoke compiler errors if any callers were missed. // Part 1: Swap declaration order @@ type TV, TErr, TObj, T1, T2; identifier OBJ, ARG1, ARG2; @@ void visit_start_struct -(TV v, TObj OBJ, T1 ARG1, const char *name, T2 ARG2, TErr errp) +(TV v, const char *name, TObj OBJ, T1 ARG1, T2 ARG2, TErr errp) { ... } @@ type bool, TV, T1; identifier ARG1; @@ bool visit_optional -(TV v, T1 ARG1, const char *name) +(TV v, const char *name, T1 ARG1) { ... } @@ type TV, TErr, TObj, T1; identifier OBJ, ARG1; @@ void visit_get_next_type -(TV v, TObj OBJ, T1 ARG1, const char *name, TErr errp) +(TV v, const char *name, TObj OBJ, T1 ARG1, TErr errp) { ... } @@ type TV, TErr, TObj, T1, T2; identifier OBJ, ARG1, ARG2; @@ void visit_type_enum -(TV v, TObj OBJ, T1 ARG1, T2 ARG2, const char *name, TErr errp) +(TV v, const char *name, TObj OBJ, T1 ARG1, T2 ARG2, TErr errp) { ... } @@ type TV, TErr, TObj; identifier OBJ; identifier VISIT_TYPE =~ "^visit_type_"; @@ void VISIT_TYPE -(TV v, TObj OBJ, const char *name, TErr errp) +(TV v, const char *name, TObj OBJ, TErr errp) { ... } // Part 2: swap caller order @@ expression V, NAME, OBJ, ARG1, ARG2, ERR; identifier VISIT_TYPE =~ "^visit_type_"; @@ ( -visit_start_struct(V, OBJ, ARG1, NAME, ARG2, ERR) +visit_start_struct(V, NAME, OBJ, ARG1, ARG2, ERR) | -visit_optional(V, ARG1, NAME) +visit_optional(V, NAME, ARG1) | -visit_get_next_type(V, OBJ, ARG1, NAME, ERR) +visit_get_next_type(V, NAME, OBJ, ARG1, ERR) | -visit_type_enum(V, OBJ, ARG1, ARG2, NAME, ERR) +visit_type_enum(V, NAME, OBJ, ARG1, ARG2, ERR) | -VISIT_TYPE(V, OBJ, NAME, ERR) +VISIT_TYPE(V, NAME, OBJ, ERR) ) Signed-off-by: Eric Blake <eblake@redhat.com> Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com> Message-Id: <1454075341-13658-19-git-send-email-eblake@redhat.com> Signed-off-by: Markus Armbruster <armbru@redhat.com>
2016-01-29 16:48:54 +03:00
visit_type_uint16List(v, name, &host_nodes, errp);
}
static void
host_memory_backend_set_host_nodes(Object *obj, Visitor *v, const char *name,
void *opaque, Error **errp)
{
#ifdef CONFIG_NUMA
HostMemoryBackend *backend = MEMORY_BACKEND(obj);
uint16List *l, *host_nodes = NULL;
visit_type_uint16List(v, name, &host_nodes, errp);
for (l = host_nodes; l; l = l->next) {
if (l->value >= MAX_NODES) {
error_setg(errp, "Invalid host-nodes value: %d", l->value);
goto out;
}
}
for (l = host_nodes; l; l = l->next) {
bitmap_set(backend->host_nodes, l->value, 1);
}
out:
qapi_free_uint16List(host_nodes);
#else
error_setg(errp, "NUMA node binding are not supported by this QEMU");
#endif
}
static int
host_memory_backend_get_policy(Object *obj, Error **errp G_GNUC_UNUSED)
{
HostMemoryBackend *backend = MEMORY_BACKEND(obj);
return backend->policy;
}
static void
host_memory_backend_set_policy(Object *obj, int policy, Error **errp)
{
HostMemoryBackend *backend = MEMORY_BACKEND(obj);
backend->policy = policy;
#ifndef CONFIG_NUMA
if (policy != HOST_MEM_POLICY_DEFAULT) {
error_setg(errp, "NUMA policies are not supported by this QEMU");
}
#endif
}
static bool host_memory_backend_get_merge(Object *obj, Error **errp)
{
HostMemoryBackend *backend = MEMORY_BACKEND(obj);
return backend->merge;
}
static void host_memory_backend_set_merge(Object *obj, bool value, Error **errp)
{
HostMemoryBackend *backend = MEMORY_BACKEND(obj);
if (!host_memory_backend_mr_inited(backend)) {
backend->merge = value;
return;
}
if (value != backend->merge) {
void *ptr = memory_region_get_ram_ptr(&backend->mr);
uint64_t sz = memory_region_size(&backend->mr);
qemu_madvise(ptr, sz,
value ? QEMU_MADV_MERGEABLE : QEMU_MADV_UNMERGEABLE);
backend->merge = value;
}
}
static bool host_memory_backend_get_dump(Object *obj, Error **errp)
{
HostMemoryBackend *backend = MEMORY_BACKEND(obj);
return backend->dump;
}
static void host_memory_backend_set_dump(Object *obj, bool value, Error **errp)
{
HostMemoryBackend *backend = MEMORY_BACKEND(obj);
if (!host_memory_backend_mr_inited(backend)) {
backend->dump = value;
return;
}
if (value != backend->dump) {
void *ptr = memory_region_get_ram_ptr(&backend->mr);
uint64_t sz = memory_region_size(&backend->mr);
qemu_madvise(ptr, sz,
value ? QEMU_MADV_DODUMP : QEMU_MADV_DONTDUMP);
backend->dump = value;
}
}
static bool host_memory_backend_get_prealloc(Object *obj, Error **errp)
{
HostMemoryBackend *backend = MEMORY_BACKEND(obj);
return backend->prealloc || backend->force_prealloc;
}
static void host_memory_backend_set_prealloc(Object *obj, bool value,
Error **errp)
{
Error *local_err = NULL;
HostMemoryBackend *backend = MEMORY_BACKEND(obj);
if (backend->force_prealloc) {
if (value) {
error_setg(errp,
"remove -mem-prealloc to use the prealloc property");
return;
}
}
if (!host_memory_backend_mr_inited(backend)) {
backend->prealloc = value;
return;
}
if (value && !backend->prealloc) {
int fd = memory_region_get_fd(&backend->mr);
void *ptr = memory_region_get_ram_ptr(&backend->mr);
uint64_t sz = memory_region_size(&backend->mr);
mem-prealloc: reduce large guest start-up and migration time. Using "-mem-prealloc" option for a large guest leads to higher guest start-up and migration time. This is because with "-mem-prealloc" option qemu tries to map every guest page (create address translations), and make sure the pages are available during runtime. virsh/libvirt by default, seems to use "-mem-prealloc" option in case the guest is configured to use huge pages. The patch tries to map all guest pages simultaneously by spawning multiple threads. Currently limiting the change to QEMU library functions on POSIX compliant host only, as we are not sure if the problem exists on win32. Below are some stats with "-mem-prealloc" option for guest configured to use huge pages. ------------------------------------------------------------------------ Idle Guest | Start-up time | Migration time ------------------------------------------------------------------------ Guest stats with 2M HugePage usage - single threaded (existing code) ------------------------------------------------------------------------ 64 Core - 4TB | 54m11.796s | 75m43.843s 64 Core - 1TB | 8m56.576s | 14m29.049s 64 Core - 256GB | 2m11.245s | 3m26.598s ------------------------------------------------------------------------ Guest stats with 2M HugePage usage - map guest pages using 8 threads ------------------------------------------------------------------------ 64 Core - 4TB | 5m1.027s | 34m10.565s 64 Core - 1TB | 1m10.366s | 8m28.188s 64 Core - 256GB | 0m19.040s | 2m10.148s ----------------------------------------------------------------------- Guest stats with 2M HugePage usage - map guest pages using 16 threads ----------------------------------------------------------------------- 64 Core - 4TB | 1m58.970s | 31m43.400s 64 Core - 1TB | 0m39.885s | 7m55.289s 64 Core - 256GB | 0m11.960s | 2m0.135s ----------------------------------------------------------------------- Changed in v2: - modify number of memset threads spawned to min(smp_cpus, 16). - removed 64GB memory restriction for spawning memset threads. Changed in v3: - limit number of threads spawned based on min(sysconf(_SC_NPROCESSORS_ONLN), 16, smp_cpus) - implement memset thread specific siglongjmp in SIGBUS signal_handler. Changed in v4 - remove sigsetjmp/siglongjmp and SIGBUS unblock/block for main thread as main thread no longer touches any pages. - simplify code my returning memset_thread_failed status from touch_all_pages. Signed-off-by: Jitendra Kolhe <jitendra.kolhe@hpe.com> Message-Id: <1487907103-32350-1-git-send-email-jitendra.kolhe@hpe.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2017-02-24 06:31:43 +03:00
os_mem_prealloc(fd, ptr, sz, smp_cpus, &local_err);
if (local_err) {
error_propagate(errp, local_err);
return;
}
backend->prealloc = true;
}
}
static void host_memory_backend_init(Object *obj)
{
HostMemoryBackend *backend = MEMORY_BACKEND(obj);
MachineState *machine = MACHINE(qdev_get_machine());
backend->merge = machine_mem_merge(machine);
backend->dump = machine_dump_guest_core(machine);
backend->prealloc = mem_prealloc;
}
bool host_memory_backend_mr_inited(HostMemoryBackend *backend)
{
/*
* NOTE: We forbid zero-length memory backend, so here zero means
* "we haven't inited the backend memory region yet".
*/
return memory_region_size(&backend->mr) != 0;
}
MemoryRegion *host_memory_backend_get_memory(HostMemoryBackend *backend)
{
return host_memory_backend_mr_inited(backend) ? &backend->mr : NULL;
}
void host_memory_backend_set_mapped(HostMemoryBackend *backend, bool mapped)
{
backend->is_mapped = mapped;
}
bool host_memory_backend_is_mapped(HostMemoryBackend *backend)
{
return backend->is_mapped;
}
#ifdef __linux__
size_t host_memory_backend_pagesize(HostMemoryBackend *memdev)
{
Object *obj = OBJECT(memdev);
char *path = object_property_get_str(obj, "mem-path", NULL);
size_t pagesize = qemu_mempath_getpagesize(path);
g_free(path);
return pagesize;
}
#else
size_t host_memory_backend_pagesize(HostMemoryBackend *memdev)
{
return getpagesize();
}
#endif
static void
host_memory_backend_memory_complete(UserCreatable *uc, Error **errp)
{
HostMemoryBackend *backend = MEMORY_BACKEND(uc);
HostMemoryBackendClass *bc = MEMORY_BACKEND_GET_CLASS(uc);
Error *local_err = NULL;
void *ptr;
uint64_t sz;
if (bc->alloc) {
bc->alloc(backend, &local_err);
if (local_err) {
goto out;
}
ptr = memory_region_get_ram_ptr(&backend->mr);
sz = memory_region_size(&backend->mr);
if (backend->merge) {
qemu_madvise(ptr, sz, QEMU_MADV_MERGEABLE);
}
if (!backend->dump) {
qemu_madvise(ptr, sz, QEMU_MADV_DONTDUMP);
}
#ifdef CONFIG_NUMA
unsigned long lastbit = find_last_bit(backend->host_nodes, MAX_NODES);
/* lastbit == MAX_NODES means maxnode = 0 */
unsigned long maxnode = (lastbit + 1) % (MAX_NODES + 1);
/* ensure policy won't be ignored in case memory is preallocated
* before mbind(). note: MPOL_MF_STRICT is ignored on hugepages so
* this doesn't catch hugepage case. */
unsigned flags = MPOL_MF_STRICT | MPOL_MF_MOVE;
/* check for invalid host-nodes and policies and give more verbose
* error messages than mbind(). */
if (maxnode && backend->policy == MPOL_DEFAULT) {
error_setg(errp, "host-nodes must be empty for policy default,"
" or you should explicitly specify a policy other"
" than default");
return;
} else if (maxnode == 0 && backend->policy != MPOL_DEFAULT) {
error_setg(errp, "host-nodes must be set for policy %s",
HostMemPolicy_str(backend->policy));
return;
}
/* We can have up to MAX_NODES nodes, but we need to pass maxnode+1
* as argument to mbind() due to an old Linux bug (feature?) which
* cuts off the last specified node. This means backend->host_nodes
* must have MAX_NODES+1 bits available.
*/
assert(sizeof(backend->host_nodes) >=
BITS_TO_LONGS(MAX_NODES + 1) * sizeof(unsigned long));
assert(maxnode <= MAX_NODES);
if (mbind(ptr, sz, backend->policy,
maxnode ? backend->host_nodes : NULL, maxnode + 1, flags)) {
if (backend->policy != MPOL_DEFAULT || errno != ENOSYS) {
error_setg_errno(errp, errno,
"cannot bind memory to host NUMA nodes");
return;
}
}
#endif
/* Preallocate memory after the NUMA policy has been instantiated.
* This is necessary to guarantee memory is allocated with
* specified NUMA policy in place.
*/
if (backend->prealloc) {
os_mem_prealloc(memory_region_get_fd(&backend->mr), ptr, sz,
mem-prealloc: reduce large guest start-up and migration time. Using "-mem-prealloc" option for a large guest leads to higher guest start-up and migration time. This is because with "-mem-prealloc" option qemu tries to map every guest page (create address translations), and make sure the pages are available during runtime. virsh/libvirt by default, seems to use "-mem-prealloc" option in case the guest is configured to use huge pages. The patch tries to map all guest pages simultaneously by spawning multiple threads. Currently limiting the change to QEMU library functions on POSIX compliant host only, as we are not sure if the problem exists on win32. Below are some stats with "-mem-prealloc" option for guest configured to use huge pages. ------------------------------------------------------------------------ Idle Guest | Start-up time | Migration time ------------------------------------------------------------------------ Guest stats with 2M HugePage usage - single threaded (existing code) ------------------------------------------------------------------------ 64 Core - 4TB | 54m11.796s | 75m43.843s 64 Core - 1TB | 8m56.576s | 14m29.049s 64 Core - 256GB | 2m11.245s | 3m26.598s ------------------------------------------------------------------------ Guest stats with 2M HugePage usage - map guest pages using 8 threads ------------------------------------------------------------------------ 64 Core - 4TB | 5m1.027s | 34m10.565s 64 Core - 1TB | 1m10.366s | 8m28.188s 64 Core - 256GB | 0m19.040s | 2m10.148s ----------------------------------------------------------------------- Guest stats with 2M HugePage usage - map guest pages using 16 threads ----------------------------------------------------------------------- 64 Core - 4TB | 1m58.970s | 31m43.400s 64 Core - 1TB | 0m39.885s | 7m55.289s 64 Core - 256GB | 0m11.960s | 2m0.135s ----------------------------------------------------------------------- Changed in v2: - modify number of memset threads spawned to min(smp_cpus, 16). - removed 64GB memory restriction for spawning memset threads. Changed in v3: - limit number of threads spawned based on min(sysconf(_SC_NPROCESSORS_ONLN), 16, smp_cpus) - implement memset thread specific siglongjmp in SIGBUS signal_handler. Changed in v4 - remove sigsetjmp/siglongjmp and SIGBUS unblock/block for main thread as main thread no longer touches any pages. - simplify code my returning memset_thread_failed status from touch_all_pages. Signed-off-by: Jitendra Kolhe <jitendra.kolhe@hpe.com> Message-Id: <1487907103-32350-1-git-send-email-jitendra.kolhe@hpe.com> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2017-02-24 06:31:43 +03:00
smp_cpus, &local_err);
if (local_err) {
goto out;
}
}
}
out:
error_propagate(errp, local_err);
}
static bool
host_memory_backend_can_be_deleted(UserCreatable *uc)
{
if (host_memory_backend_is_mapped(MEMORY_BACKEND(uc))) {
return false;
} else {
return true;
}
}
static bool host_memory_backend_get_share(Object *o, Error **errp)
{
HostMemoryBackend *backend = MEMORY_BACKEND(o);
return backend->share;
}
static void host_memory_backend_set_share(Object *o, bool value, Error **errp)
{
HostMemoryBackend *backend = MEMORY_BACKEND(o);
if (host_memory_backend_mr_inited(backend)) {
error_setg(errp, "cannot change property value");
return;
}
backend->share = value;
}
static void
host_memory_backend_class_init(ObjectClass *oc, void *data)
{
UserCreatableClass *ucc = USER_CREATABLE_CLASS(oc);
ucc->complete = host_memory_backend_memory_complete;
ucc->can_be_deleted = host_memory_backend_can_be_deleted;
object_class_property_add_bool(oc, "merge",
host_memory_backend_get_merge,
host_memory_backend_set_merge, &error_abort);
object_class_property_set_description(oc, "merge",
"Mark memory as mergeable", &error_abort);
object_class_property_add_bool(oc, "dump",
host_memory_backend_get_dump,
host_memory_backend_set_dump, &error_abort);
object_class_property_set_description(oc, "dump",
"Set to 'off' to exclude from core dump", &error_abort);
object_class_property_add_bool(oc, "prealloc",
host_memory_backend_get_prealloc,
host_memory_backend_set_prealloc, &error_abort);
object_class_property_set_description(oc, "prealloc",
"Preallocate memory", &error_abort);
object_class_property_add(oc, "size", "int",
host_memory_backend_get_size,
host_memory_backend_set_size,
NULL, NULL, &error_abort);
object_class_property_set_description(oc, "size",
"Size of the memory region (ex: 500M)", &error_abort);
object_class_property_add(oc, "host-nodes", "int",
host_memory_backend_get_host_nodes,
host_memory_backend_set_host_nodes,
NULL, NULL, &error_abort);
object_class_property_set_description(oc, "host-nodes",
"Binds memory to the list of NUMA host nodes", &error_abort);
object_class_property_add_enum(oc, "policy", "HostMemPolicy",
&HostMemPolicy_lookup,
host_memory_backend_get_policy,
host_memory_backend_set_policy, &error_abort);
object_class_property_set_description(oc, "policy",
"Set the NUMA policy", &error_abort);
object_class_property_add_bool(oc, "share",
host_memory_backend_get_share, host_memory_backend_set_share,
&error_abort);
object_class_property_set_description(oc, "share",
"Mark the memory as private to QEMU or shared", &error_abort);
}
static const TypeInfo host_memory_backend_info = {
.name = TYPE_MEMORY_BACKEND,
.parent = TYPE_OBJECT,
.abstract = true,
.class_size = sizeof(HostMemoryBackendClass),
.class_init = host_memory_backend_class_init,
.instance_size = sizeof(HostMemoryBackend),
.instance_init = host_memory_backend_init,
.interfaces = (InterfaceInfo[]) {
{ TYPE_USER_CREATABLE },
{ }
}
};
static void register_types(void)
{
type_register_static(&host_memory_backend_info);
}
type_init(register_types);