NUMA series and hostmem improvements

index dd7ac13..ce668b7 100644
@@ -1312,20 +1312,6 @@ static void tcg_exec_all(void)
     exit_request = 0;
 }
 
-void set_numa_modes(void)
-{
-    CPUState *cpu;
-    int i;
-
-    CPU_FOREACH(cpu) {
-        for (i = 0; i < nb_numa_nodes; i++) {
-            if (test_bit(cpu->cpu_index, node_cpumask[i])) {
-                cpu->numa_node = i;
-            }
-        }
-    }
-}
-
 void list_cpus(FILE *f, fprintf_function cpu_fprintf, const char *optarg)
 {
     /* XXX: implement xxx_cpu_list for targets that still miss it */

index e8363d7..ed28f1e 100644
@@ -297,8 +297,6 @@ CPUArchState *cpu_copy(CPUArchState *env);
 
 /* memory API */
 
-extern ram_addr_t ram_size;
-
 /* RAM is pre-allocated and passed into qemu_ram_alloc_from_ptr */
 #define RAM_PREALLOC_MASK   (1 << 0)
 

index a21b65a..e8c7970 100644
@@ -45,6 +45,8 @@ typedef uintptr_t ram_addr_t;
 #  define RAM_ADDR_FMT "%" PRIxPTR
 #endif
 
+extern ram_addr_t ram_size;
+
 /* memory API */
 
 typedef void CPUWriteMemoryFunc(void *opaque, hwaddr addr, uint32_t value);

index 6502488..4f79081 100644
@@ -23,7 +23,6 @@ extern int smp_threads;
 #define smp_threads 1
 #endif
 
-void set_numa_modes(void);
 void list_cpus(FILE *f, fprintf_function cpu_fprintf, const char *optarg);
 
 #endif

index ba5c7f8..565c8f6 100644
@@ -144,6 +144,9 @@ extern QEMUClockType rtc_clock;
 extern int nb_numa_nodes;
 extern uint64_t node_mem[MAX_NODES];
 extern unsigned long *node_cpumask[MAX_NODES];
+void numa_add(const char *optarg);
+void set_numa_nodes(void);
+void set_numa_modes(void);
 
 #define MAX_OPTION_ROMS 16
 typedef struct QEMUOptionRom {

185 lines not on this page · raw message

151 lines not on this page · raw message

index c419deb..b106531 100644
@@ -26,6 +26,8 @@
 #include "exec/cpu-common.h"
 #include "qemu/bitmap.h"
 #include "qom/cpu.h"
+#include "qemu/error-report.h"
+#include "include/exec/cpu-common.h" /* for RAM_ADDR_FMT */
 
 static void numa_node_parse_cpus(int nodenr, const char *cpus)
 {

@@ -126,6 +128,7 @@ void numa_add(const char *optarg)
 void set_numa_nodes(void)
 {
     if (nb_numa_nodes > 0) {
+        uint64_t numa_total;
         int i;
 
         if (nb_numa_nodes > MAX_NODES) {

@@ -153,6 +156,17 @@ void set_numa_nodes(void)
             node_mem[i] = ram_size - usedmem;
         }
 
+        numa_total = 0;
+        for (i = 0; i < nb_numa_nodes; i++) {
+            numa_total += node_mem[i];
+        }
+        if (numa_total != ram_size) {
+            error_report("qemu: total memory size for NUMA nodes (%" PRIu64 ")"
+                         " should equal to RAM size (" RAM_ADDR_FMT ")\n",
+                         numa_total, ram_size);
+            exit(1);
+        }
+
         for (i = 0; i < nb_numa_nodes; i++) {
             if (!bitmap_empty(node_cpumask[i], MAX_CPUMASK_BITS)) {
                 break;

index 7cdba10..2c75ecc 100644
@@ -705,14 +705,14 @@ static FWCfgState *bochs_bios_init(void)
         unsigned int apic_id = x86_cpu_apic_id_from_index(i);
         assert(apic_id < apic_id_limit);
         for (j = 0; j < nb_numa_nodes; j++) {
-            if (test_bit(i, node_cpumask[j])) {
+            if (test_bit(i, numa_info[j].node_cpu)) {
                 numa_fw_cfg[apic_id + 1] = cpu_to_le64(j);
                 break;
             }
         }
     }
     for (i = 0; i < nb_numa_nodes; i++) {
-        numa_fw_cfg[apic_id_limit + 1 + i] = cpu_to_le64(node_mem[i]);
+        numa_fw_cfg[apic_id_limit + 1 + i] = cpu_to_le64(numa_info[i].node_mem);
     }
     fw_cfg_add_bytes(fw_cfg, FW_CFG_NUMA, numa_fw_cfg,
                      (1 + apic_id_limit + nb_numa_nodes) *

@@ -1126,8 +1126,12 @@ PcGuestInfo *pc_guest_info_init(ram_addr_t below_4g_mem_size,
     guest_info->apic_id_limit = pc_apic_id_limit(max_cpus);
     guest_info->apic_xrupt_override = kvm_allows_irq0_override();
     guest_info->numa_nodes = nb_numa_nodes;
-    guest_info->node_mem = g_memdup(node_mem, guest_info->numa_nodes *
+    guest_info->node_mem = g_malloc0(guest_info->numa_nodes *
                                     sizeof *guest_info->node_mem);
+    for (i = 0; i < nb_numa_nodes; i++) {
+        guest_info->node_mem[i] = numa_info[i].node_mem;
+    }
+
     guest_info->node_cpu = g_malloc0(guest_info->apic_id_limit *
                                      sizeof *guest_info->node_cpu);
 

@@ -1135,7 +1139,7 @@ PcGuestInfo *pc_guest_info_init(ram_addr_t below_4g_mem_size,
         unsigned int apic_id = x86_cpu_apic_id_from_index(i);
         assert(apic_id < guest_info->apic_id_limit);
         for (j = 0; j < nb_numa_nodes; j++) {
-            if (test_bit(i, node_cpumask[j])) {
+            if (test_bit(i, numa_info[j].node_cpu)) {
                 guest_info->node_cpu[apic_id] = j;
                 break;
             }

index 57e9578..d252a60 100644
@@ -538,8 +538,8 @@ static int spapr_populate_memory(sPAPREnvironment *spapr, void *fdt)
     int i, off;
 
     /* memory node(s) */
-    if (nb_numa_nodes > 1 && node_mem[0] < ram_size) {
-        node0_size = node_mem[0];
+    if (nb_numa_nodes > 1 && numa_info[0].node_mem < ram_size) {
+        node0_size = numa_info[0].node_mem;
     } else {
         node0_size = ram_size;
     }

@@ -577,7 +577,7 @@ static int spapr_populate_memory(sPAPREnvironment *spapr, void *fdt)
         if (mem_start >= ram_size) {
             node_size = 0;
         } else {
-            node_size = node_mem[i];
+            node_size = numa_info[i].node_mem;
             if (node_size > ram_size - mem_start) {
                 node_size = ram_size - mem_start;
             }

@@ -722,7 +722,8 @@ static void spapr_reset_htab(sPAPREnvironment *spapr)
 
     /* Update the RMA size if necessary */
     if (spapr->vrma_adjust) {
-        hwaddr node0_size = (nb_numa_nodes > 1) ? node_mem[0] : ram_size;
+        hwaddr node0_size = (nb_numa_nodes > 1) ?
+            numa_info[0].node_mem : ram_size;
         spapr->rma_size = kvmppc_rma_size(node0_size, spapr->htab_shift);
     }
 }

@@ -1155,7 +1156,7 @@ static void ppc_spapr_init(MachineState *machine)
     MemoryRegion *sysmem = get_system_memory();
     MemoryRegion *ram = g_new(MemoryRegion, 1);
     hwaddr rma_alloc_size;
-    hwaddr node0_size = (nb_numa_nodes > 1) ? node_mem[0] : ram_size;
+    hwaddr node0_size = (nb_numa_nodes > 1) ? numa_info[0].node_mem : ram_size;
     uint32_t initrd_base = 0;
     long kernel_size = 0, initrd_size = 0;
     long load_limit, rtas_limit, fw_size;

index 565c8f6..3a9308b 100644
@@ -9,6 +9,7 @@
 #include "qapi-types.h"
 #include "qemu/notify.h"
 #include "qemu/main-loop.h"
+#include "qemu/bitmap.h"
 
 /* vl.c */
 

@@ -142,8 +143,11 @@ extern QEMUClockType rtc_clock;
 #define MAX_CPUMASK_BITS 255
 
 extern int nb_numa_nodes;
-extern uint64_t node_mem[MAX_NODES];
-extern unsigned long *node_cpumask[MAX_NODES];
+typedef struct node_info {
+    uint64_t node_mem;
+    DECLARE_BITMAP(node_cpu, MAX_CPUMASK_BITS);
+} NodeInfo;
+extern NodeInfo numa_info[MAX_NODES];
 void numa_add(const char *optarg);
 void set_numa_nodes(void);
 void set_numa_modes(void);

index 0565816..a2a4466 100644
@@ -2006,7 +2006,7 @@ static void do_info_numa(Monitor *mon, const QDict *qdict)
         }
         monitor_printf(mon, "\n");
         monitor_printf(mon, "node %d size: %" PRId64 " MB\n", i,
-            node_mem[i] >> 20);
+            numa_info[i].node_mem >> 20);
     }
 }
 

index b106531..fb9bffc 100644
@@ -65,7 +65,7 @@ static void numa_node_parse_cpus(int nodenr, const char *cpus)
         goto error;
     }
 
-    bitmap_set(node_cpumask[nodenr], value, endvalue-value+1);
+    bitmap_set(numa_info[nodenr].node_cpu, value, endvalue-value+1);
     return;
 
 error:

@@ -105,7 +105,7 @@ void numa_add(const char *optarg)
         }
 
         if (get_param_value(option, 128, "mem", optarg) == 0) {
-            node_mem[nodenr] = 0;
+            numa_info[nodenr].node_mem = 0;
         } else {
             int64_t sval;
             sval = strtosz(option, &endptr);

@@ -113,7 +113,7 @@ void numa_add(const char *optarg)
                 fprintf(stderr, "qemu: invalid numa mem size: %s\n", optarg);
                 exit(1);
             }
-            node_mem[nodenr] = sval;
+            numa_info[nodenr].node_mem = sval;
         }
         if (get_param_value(option, 128, "cpus", optarg) != 0) {
             numa_node_parse_cpus(nodenr, option);

@@ -139,7 +139,7 @@ void set_numa_nodes(void)
          * and distribute the available memory equally across all nodes
          */
         for (i = 0; i < nb_numa_nodes; i++) {
-            if (node_mem[i] != 0) {
+            if (numa_info[i].node_mem != 0) {
                 break;
             }
         }

@@ -150,15 +150,16 @@ void set_numa_nodes(void)
              * the final node gets the rest.
              */
             for (i = 0; i < nb_numa_nodes - 1; i++) {
-                node_mem[i] = (ram_size / nb_numa_nodes) & ~((1 << 23UL) - 1);
-                usedmem += node_mem[i];
+                numa_info[i].node_mem = (ram_size / nb_numa_nodes) &
+                                        ~((1 << 23UL) - 1);
+                usedmem += numa_info[i].node_mem;
             }
-            node_mem[i] = ram_size - usedmem;
+            numa_info[i].node_mem = ram_size - usedmem;
         }
 
         numa_total = 0;
         for (i = 0; i < nb_numa_nodes; i++) {
-            numa_total += node_mem[i];
+            numa_total += numa_info[i].node_mem;
         }
         if (numa_total != ram_size) {
             error_report("qemu: total memory size for NUMA nodes (%" PRIu64 ")"

@@ -168,7 +169,7 @@ void set_numa_nodes(void)
         }
 
         for (i = 0; i < nb_numa_nodes; i++) {
-            if (!bitmap_empty(node_cpumask[i], MAX_CPUMASK_BITS)) {
+            if (!bitmap_empty(numa_info[i].node_cpu, MAX_CPUMASK_BITS)) {
                 break;
             }
         }

@@ -178,7 +179,7 @@ void set_numa_nodes(void)
          */
         if (i == nb_numa_nodes) {
             for (i = 0; i < max_cpus; i++) {
-                set_bit(i, node_cpumask[i % nb_numa_nodes]);
+                set_bit(i, numa_info[i % nb_numa_nodes].node_cpu);
             }
         }
     }

@@ -191,7 +192,7 @@ void set_numa_modes(void)
 
     CPU_FOREACH(cpu) {
         for (i = 0; i < nb_numa_nodes; i++) {
-            if (test_bit(cpu->cpu_index, node_cpumask[i])) {
+            if (test_bit(cpu->cpu_index, numa_info[i].node_cpu)) {
                 cpu->numa_node = i;
             }
         }

index eacbcc7..a38e6dd 100644
@@ -195,8 +195,7 @@ static QTAILQ_HEAD(, FWBootEntry) fw_boot_order =
     QTAILQ_HEAD_INITIALIZER(fw_boot_order);
 
 int nb_numa_nodes;
-uint64_t node_mem[MAX_NODES];
-unsigned long *node_cpumask[MAX_NODES];
+NodeInfo numa_info[MAX_NODES];
 
 uint8_t qemu_uuid[16];
 bool qemu_uuid_set;

@@ -2959,8 +2958,8 @@ int main(int argc, char **argv, char **envp)
     translation = BIOS_ATA_TRANSLATION_AUTO;
 
     for (i = 0; i < MAX_NODES; i++) {
-        node_mem[i] = 0;
-        node_cpumask[i] = bitmap_new(MAX_CPUMASK_BITS);
+        numa_info[i].node_mem = 0;
+        bitmap_zero(numa_info[i].node_cpu, MAX_CPUMASK_BITS);
     }
 
     nb_numa_nodes = 0;

index 3a9308b..4102be3 100644
@@ -148,9 +148,10 @@ typedef struct node_info {
     DECLARE_BITMAP(node_cpu, MAX_CPUMASK_BITS);
 } NodeInfo;
 extern NodeInfo numa_info[MAX_NODES];
-void numa_add(const char *optarg);
 void set_numa_nodes(void);
 void set_numa_modes(void);
+extern QemuOptsList qemu_numa_opts;
+int numa_init_func(QemuOpts *opts, void *opaque);
 
 #define MAX_OPTION_ROMS 16
 typedef struct QEMUOptionRom {

171 lines not on this page · raw message

index a38e6dd..730bd1e 100644
@@ -2938,6 +2938,7 @@ int main(int argc, char **argv, char **envp)
     qemu_add_opts(&qemu_realtime_opts);
     qemu_add_opts(&qemu_msg_opts);
     qemu_add_opts(&qemu_name_opts);
+    qemu_add_opts(&qemu_numa_opts);
 
     runstate_init();
 

@@ -3133,7 +3134,10 @@ int main(int argc, char **argv, char **envp)
                 }
                 break;
             case QEMU_OPTION_numa:
-                numa_add(optarg);
+                opts = qemu_opts_parse(qemu_find_opts("numa"), optarg, 1);
+                if (!opts) {
+                    exit(1);
+                }
                 break;
             case QEMU_OPTION_display:
                 display_type = select_display(optarg);

@@ -4303,6 +4307,11 @@ int main(int argc, char **argv, char **envp)
     default_drive(default_floppy, snapshot, IF_FLOPPY, 0, FD_OPTS);
     default_drive(default_sdcard, snapshot, IF_SD, 0, SD_OPTS);
 
+    if (qemu_opts_foreach(qemu_find_opts("numa"), numa_init_func,
+                          NULL, 1) != 0) {
+        exit(1);
+    }
+
     set_numa_nodes();
 
     if (qemu_opts_foreach(qemu_find_opts("mon"), mon_init_func, NULL, 1) != 0) {

index 4102be3..423d49e 100644
@@ -133,7 +133,7 @@ extern size_t boot_splash_filedata_size;
 extern uint8_t qemu_extra_params_fw[2];
 extern QEMUClockType rtc_clock;
 
-#define MAX_NODES 64
+#define MAX_NODES 128
 
 /* The following shall be true for all CPUs:
  *   cpu->cpu_index < max_cpus <= MAX_CPUMASK_BITS

index 730bd1e..1617013 100644
@@ -116,7 +116,7 @@ int main(int argc, char **argv)
 
 #include "ui/qemu-spice.h"
 #include "qapi/string-input-visitor.h"
-#include "qom/object_interfaces.h"
+#include "qapi/opts-visitor.h"
 
 #define DEFAULT_RAM_SIZE 128
 

@@ -2822,44 +2822,51 @@ static int object_set_property(const char *name, const char *value, void *opaque
 
 static int object_create(QemuOpts *opts, void *opaque)
 {
-    const char *type = qemu_opt_get(opts, "qom-type");
-    const char *id = qemu_opts_id(opts);
-    Error *local_err = NULL;
-    Object *obj;
-
-    g_assert(type != NULL);
-
-    if (id == NULL) {
-        qerror_report(QERR_MISSING_PARAMETER, "id");
-        return -1;
+    Error *err = NULL;
+    char *type = NULL;
+    char *id = NULL;
+    void *dummy = NULL;
+    OptsVisitor *ov;
+    QDict *pdict;
+
+    ov = opts_visitor_new(opts);
+    pdict = qemu_opts_to_qdict(opts, NULL);
+
+    visit_start_struct(opts_get_visitor(ov), &dummy, NULL, NULL, 0, &err);
+    if (err) {
+        goto out;
     }
 
-    obj = object_new(type);
-    if (qemu_opt_foreach(opts, object_set_property, obj, 1) < 0) {
-        object_unref(obj);
-        return -1;
+    qdict_del(pdict, "qom-type");
+    visit_type_str(opts_get_visitor(ov), &type, "qom-type", &err);
+    if (err) {
+        goto out;
     }
 
-    if (!object_dynamic_cast(obj, TYPE_USER_CREATABLE)) {
-        error_setg(&local_err, "object '%s' isn't supported by -object",
-                   id);
+    qdict_del(pdict, "id");
+    visit_type_str(opts_get_visitor(ov), &id, "id", &err);
+    if (err) {
         goto out;
     }
 
-    object_property_add_child(container_get(object_get_root(), "/objects"),
-                              id, obj, &local_err);
-
-    user_creatable_complete(obj, &local_err);
-    if (local_err) {
-        object_property_del(container_get(object_get_root(), "/objects"),
-                            id, &error_abort);
+    object_add(type, id, pdict, opts_get_visitor(ov), &err);
+    if (err) {
         goto out;
     }
+    visit_end_struct(opts_get_visitor(ov), &err);
+    if (err) {
+        qmp_object_del(id, NULL);
+    }
+
 out:
-    object_unref(obj);
-    if (local_err) {
-        qerror_report_err(local_err);
-        error_free(local_err);
+    opts_visitor_cleanup(ov);
+
+    QDECREF(pdict);
+    g_free(id);
+    g_free(type);
+    g_free(dummy);
+    if (err) {
+        qerror_report_err(err);
         return -1;
     }
     return 0;

index b722dbe..cef60fb 100644
@@ -540,7 +540,8 @@ void object_add(const char *type, const char *id, const QDict *qdict,
 
     klass = object_class_by_name(type);
     if (!klass) {
-        error_setg(errp, "invalid class name");
+        error_set(errp, QERR_INVALID_PARAMETER_VALUE,
+                  "qom-type", "a valid class name");
         return;
     }
 

index 2c75ecc..9860e3f 100644
@@ -1194,10 +1194,8 @@ void pc_acpi_init(const char *default_dsdt)
     }
 }
 
-FWCfgState *pc_memory_init(MemoryRegion *system_memory,
-                           const char *kernel_filename,
-                           const char *kernel_cmdline,
-                           const char *initrd_filename,
+FWCfgState *pc_memory_init(MachineState *machine,
+                           MemoryRegion *system_memory,
                            ram_addr_t below_4g_mem_size,
                            ram_addr_t above_4g_mem_size,
                            MemoryRegion *rom_memory,

@@ -1208,18 +1206,18 @@ FWCfgState *pc_memory_init(MemoryRegion *system_memory,
     MemoryRegion *ram, *option_rom_mr;
     MemoryRegion *ram_below_4g, *ram_above_4g;
     FWCfgState *fw_cfg;
-    ram_addr_t ram_size = below_4g_mem_size + above_4g_mem_size;
-    MachineState *machine = MACHINE(qdev_get_machine());
     PCMachineState *pcms = PC_MACHINE(machine);
 
-    linux_boot = (kernel_filename != NULL);
+    assert(machine->ram_size == below_4g_mem_size + above_4g_mem_size);
+
+    linux_boot = (machine->kernel_filename != NULL);
 
     /* Allocate RAM.  We allocate it as a single memory region and use
      * aliases to address portions of it, mostly for backwards compatibility
      * with older qemus that used qemu_ram_alloc().
      */
     ram = g_malloc(sizeof(*ram));
-    memory_region_init_ram(ram, NULL, "pc.ram", ram_size);
+    memory_region_init_ram(ram, NULL, "pc.ram", machine->ram_size);
     vmstate_register_ram_global(ram);
     *ram_memory = ram;
     ram_below_4g = g_malloc(sizeof(*ram_below_4g));

@@ -1238,7 +1236,7 @@ FWCfgState *pc_memory_init(MemoryRegion *system_memory,
 
     if (!guest_info->has_reserved_memory &&
         (machine->ram_slots ||
-         (machine->maxram_size > ram_size))) {
+         (machine->maxram_size > machine->ram_size))) {
         MachineClass *mc = MACHINE_GET_CLASS(machine);
 
         error_report("\"-memory 'slots|maxmem'\" is not supported by: %s",

@@ -1248,9 +1246,9 @@ FWCfgState *pc_memory_init(MemoryRegion *system_memory,
 
     /* initialize hotplug memory address space */
     if (guest_info->has_reserved_memory &&
-        (ram_size < machine->maxram_size)) {
+        (machine->ram_size < machine->maxram_size)) {
         ram_addr_t hotplug_mem_size =
-            machine->maxram_size - ram_size;
+            machine->maxram_size - machine->ram_size;
 
         if (machine->ram_slots > ACPI_MAX_RAM_SLOTS) {
             error_report("unsupported amount of memory slots: %"PRIu64,

@@ -1295,7 +1293,8 @@ FWCfgState *pc_memory_init(MemoryRegion *system_memory,
     }
 
     if (linux_boot) {
-        load_linux(fw_cfg, kernel_filename, initrd_filename, kernel_cmdline, below_4g_mem_size);
+        load_linux(fw_cfg, machine->kernel_filename, machine->initrd_filename,
+                   machine->kernel_cmdline, below_4g_mem_size);
     }
 
     for (i = 0; i < nb_option_roms; i++) {

index a13e8d6..3e7524b 100644
@@ -156,11 +156,9 @@ static void pc_init1(MachineState *machine,
 
     /* allocate ram and load rom/bios */
     if (!xen_enabled()) {
-        fw_cfg = pc_memory_init(system_memory,
-                       machine->kernel_filename, machine->kernel_cmdline,
-                       machine->initrd_filename,
-                       below_4g_mem_size, above_4g_mem_size,
-                       rom_memory, &ram_memory, guest_info);
+        fw_cfg = pc_memory_init(machine, system_memory,
+                                below_4g_mem_size, above_4g_mem_size,
+                                rom_memory, &ram_memory, guest_info);
     }
 
     gsi_state = g_malloc0(sizeof(*gsi_state));

index 629eb2d..aa71332 100644
@@ -143,9 +143,7 @@ static void pc_q35_init(MachineState *machine)
 
     /* allocate ram and load rom/bios */
     if (!xen_enabled()) {
-        pc_memory_init(get_system_memory(),
-                       machine->kernel_filename, machine->kernel_cmdline,
-                       machine->initrd_filename,
+        pc_memory_init(machine, get_system_memory(),
                        below_4g_mem_size, above_4g_mem_size,
                        rom_memory, &ram_memory, guest_info);
     }

index fe9e18b..f337d54 100644
@@ -3,6 +3,7 @@
 
 #include "qemu-common.h"
 #include "exec/memory.h"
+#include "hw/boards.h"
 #include "hw/isa/isa.h"
 #include "hw/block/fdc.h"
 #include "net/net.h"

@@ -183,10 +184,8 @@ PcGuestInfo *pc_guest_info_init(ram_addr_t below_4g_mem_size,
 void pc_pci_as_mapping_init(Object *owner, MemoryRegion *system_memory,
                             MemoryRegion *pci_address_space);
 
-FWCfgState *pc_memory_init(MemoryRegion *system_memory,
-                           const char *kernel_filename,
-                           const char *kernel_cmdline,
-                           const char *initrd_filename,
+FWCfgState *pc_memory_init(MachineState *machine,
+                           MemoryRegion *system_memory,
                            ram_addr_t below_4g_mem_size,
                            ram_addr_t above_4g_mem_size,
                            MemoryRegion *rom_memory,

index 9860e3f..37344ce 100644
@@ -1217,8 +1217,8 @@ FWCfgState *pc_memory_init(MachineState *machine,
      * with older qemus that used qemu_ram_alloc().
      */
     ram = g_malloc(sizeof(*ram));
-    memory_region_init_ram(ram, NULL, "pc.ram", machine->ram_size);
-    vmstate_register_ram_global(ram);
+    memory_region_allocate_system_memory(ram, NULL, "pc.ram",
+                                         machine->ram_size);
     *ram_memory = ram;
     ram_below_4g = g_malloc(sizeof(*ram_below_4g));
     memory_region_init_alias(ram_below_4g, NULL, "ram-below-4g", ram,

index 429ac43..605a970 100644
@@ -43,9 +43,13 @@ struct QEMUMachine {
     const char *hw_version;
 };
 
-#define TYPE_MACHINE_SUFFIX "-machine"
+void memory_region_allocate_system_memory(MemoryRegion *mr, Object *owner,
+                                          const char *name,
+                                          uint64_t ram_size);
+
 int qemu_register_machine(QEMUMachine *m);
 
+#define TYPE_MACHINE_SUFFIX "-machine"
 #define TYPE_MACHINE "machine"
 #undef MACHINE  /* BSD defines it and QEMU does not use it */
 #define MACHINE(obj) \

index 423d49e..caf88dd 100644
@@ -10,6 +10,7 @@
 #include "qemu/notify.h"
 #include "qemu/main-loop.h"
 #include "qemu/bitmap.h"
+#include "qom/object.h"
 
 /* vl.c */
 

index 43573c5..efdebf4 100644
@@ -32,6 +32,7 @@
 #include "qapi/opts-visitor.h"
 #include "qapi/dealloc-visitor.h"
 #include "qapi/qmp/qerror.h"
+#include "hw/boards.h"
 
 QemuOptsList qemu_numa_opts = {
     .name = "numa",

@@ -193,3 +194,11 @@ void set_numa_modes(void)
         }
     }
 }
+
+void memory_region_allocate_system_memory(MemoryRegion *mr, Object *owner,
+                                          const char *name,
+                                          uint64_t ram_size)
+{
+    memory_region_init_ram(mr, owner, name, ram_size);
+    vmstate_register_ram_global(mr);
+}

index bba2ebc..d9a8290 100644
@@ -16,9 +16,8 @@
 
 
 static void
-ram_backend_memory_init(UserCreatable *uc, Error **errp)
+ram_backend_memory_alloc(HostMemoryBackend *backend, Error **errp)
 {
-    HostMemoryBackend *backend = MEMORY_BACKEND(uc);
     char *path;
 
     if (!backend->size) {

@@ -35,9 +34,9 @@ ram_backend_memory_init(UserCreatable *uc, Error **errp)
 static void
 ram_backend_class_init(ObjectClass *oc, void *data)
 {
-    UserCreatableClass *ucc = USER_CREATABLE_CLASS(oc);
+    HostMemoryBackendClass *bc = MEMORY_BACKEND_CLASS(oc);
 
-    ucc->complete = ram_backend_memory_init;
+    bc->alloc = ram_backend_memory_alloc;
 }
 
 static const TypeInfo ram_backend_info = {

index 2f578ac..cc57c13 100644
@@ -17,7 +17,7 @@
 #include "qom/object_interfaces.h"
 
 static void
-hostmemory_backend_get_size(Object *obj, Visitor *v, void *opaque,
+host_memory_backend_get_size(Object *obj, Visitor *v, void *opaque,
                             const char *name, Error **errp)
 {
     HostMemoryBackend *backend = MEMORY_BACKEND(obj);

@@ -27,7 +27,7 @@ hostmemory_backend_get_size(Object *obj, Visitor *v, void *opaque,
 }
 
 static void
-hostmemory_backend_set_size(Object *obj, Visitor *v, void *opaque,
+host_memory_backend_set_size(Object *obj, Visitor *v, void *opaque,
                             const char *name, Error **errp)
 {
     HostMemoryBackend *backend = MEMORY_BACKEND(obj);

@@ -53,14 +53,14 @@ out:
     error_propagate(errp, local_err);
 }
 
-static void hostmemory_backend_init(Object *obj)
+static void host_memory_backend_init(Object *obj)
 {
     object_property_add(obj, "size", "int",
-                        hostmemory_backend_get_size,
-                        hostmemory_backend_set_size, NULL, NULL, NULL);
+                        host_memory_backend_get_size,
+                        host_memory_backend_set_size, NULL, NULL, NULL);
 }
 
-static void hostmemory_backend_finalize(Object *obj)
+static void host_memory_backend_finalize(Object *obj)
 {
     HostMemoryBackend *backend = MEMORY_BACKEND(obj);
 

@@ -75,14 +75,34 @@ host_memory_backend_get_memory(HostMemoryBackend *backend, Error **errp)
     return memory_region_size(&backend->mr) ? &backend->mr : NULL;
 }
 
-static const TypeInfo hostmemory_backend_info = {
+static void
+host_memory_backend_memory_complete(UserCreatable *uc, Error **errp)
+{
+    HostMemoryBackend *backend = MEMORY_BACKEND(uc);
+    HostMemoryBackendClass *bc = MEMORY_BACKEND_GET_CLASS(uc);
+
+    if (bc->alloc) {
+        bc->alloc(backend, errp);
+    }
+}
+
+static void
+host_memory_backend_class_init(ObjectClass *oc, void *data)
+{
+    UserCreatableClass *ucc = USER_CREATABLE_CLASS(oc);
+
+    ucc->complete = host_memory_backend_memory_complete;
+}
+
+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 = hostmemory_backend_init,
-    .instance_finalize = hostmemory_backend_finalize,
+    .instance_init = host_memory_backend_init,
+    .instance_finalize = host_memory_backend_finalize,
     .interfaces = (InterfaceInfo[]) {
         { TYPE_USER_CREATABLE },
         { }

@@ -91,7 +111,7 @@ static const TypeInfo hostmemory_backend_info = {
 
 static void register_types(void)
 {
-    type_register_static(&hostmemory_backend_info);
+    type_register_static(&host_memory_backend_info);
 }
 
 type_init(register_types);

index 4fc081e..923f672 100644
@@ -34,6 +34,8 @@ typedef struct HostMemoryBackendClass HostMemoryBackendClass;
  */
 struct HostMemoryBackendClass {
     ObjectClass parent_class;
+
+    void (*alloc)(HostMemoryBackend *backend, Error **errp);
 };
 
 /**

index caf88dd..1e141e3 100644
@@ -147,6 +147,7 @@ extern int nb_numa_nodes;
 typedef struct node_info {
     uint64_t node_mem;
     DECLARE_BITMAP(node_cpu, MAX_CPUMASK_BITS);
+    struct HostMemoryBackend *node_memdev;
 } NodeInfo;
 extern NodeInfo numa_info[MAX_NODES];
 void set_numa_nodes(void);

94 lines not on this page · raw message

index 4e179a6..525fc04 100644
@@ -1246,56 +1246,30 @@ static int memory_try_enable_merging(void *addr, size_t len)
     return qemu_madvise(addr, len, QEMU_MADV_MERGEABLE);
 }
 
-ram_addr_t qemu_ram_alloc_from_ptr(ram_addr_t size, void *host,
-                                   MemoryRegion *mr)
+static ram_addr_t ram_block_add(RAMBlock *new_block)
 {
-    RAMBlock *block, *new_block;
+    RAMBlock *block;
     ram_addr_t old_ram_size, new_ram_size;
 
     old_ram_size = last_ram_offset() >> TARGET_PAGE_BITS;
 
-    size = TARGET_PAGE_ALIGN(size);
-    new_block = g_malloc0(sizeof(*new_block));
-    new_block->fd = -1;
-
     /* This assumes the iothread lock is taken here too.  */
     qemu_mutex_lock_ramlist();
-    new_block->mr = mr;
-    new_block->offset = find_ram_offset(size);
-    if (host) {
-        new_block->host = host;
-        new_block->flags |= RAM_PREALLOC_MASK;
-    } else if (xen_enabled()) {
-        if (mem_path) {
-            fprintf(stderr, "-mem-path not supported with Xen\n");
-            exit(1);
-        }
-        xen_ram_alloc(new_block->offset, size, mr);
-    } else {
-        if (mem_path) {
-            if (phys_mem_alloc != qemu_anon_ram_alloc) {
-                /*
-                 * file_ram_alloc() needs to allocate just like
-                 * phys_mem_alloc, but we haven't bothered to provide
-                 * a hook there.
-                 */
-                fprintf(stderr,
-                        "-mem-path not supported with this accelerator\n");
-                exit(1);
-            }
-            new_block->host = file_ram_alloc(new_block, size, mem_path);
-        }
-        if (!new_block->host) {
-            new_block->host = phys_mem_alloc(size);
+    new_block->offset = find_ram_offset(new_block->length);
+
+    if (!new_block->host) {
+        if (xen_enabled()) {
+            xen_ram_alloc(new_block->offset, new_block->length, new_block->mr);
+        } else {
+            new_block->host = phys_mem_alloc(new_block->length);
             if (!new_block->host) {
                 fprintf(stderr, "Cannot set up guest memory '%s': %s\n",
                         new_block->mr->name, strerror(errno));
                 exit(1);
             }
-            memory_try_enable_merging(new_block->host, size);
+            memory_try_enable_merging(new_block->host, new_block->length);
         }
     }
-    new_block->length = size;
 
     /* Keep the list sorted from biggest to smallest block.  */
     QTAILQ_FOREACH(block, &ram_list.blocks, next) {

@@ -1323,18 +1297,65 @@ ram_addr_t qemu_ram_alloc_from_ptr(ram_addr_t size, void *host,
                                    old_ram_size, new_ram_size);
        }
     }
-    cpu_physical_memory_set_dirty_range(new_block->offset, size);
+    cpu_physical_memory_set_dirty_range(new_block->offset, new_block->length);
 
-    qemu_ram_setup_dump(new_block->host, size);
-    qemu_madvise(new_block->host, size, QEMU_MADV_HUGEPAGE);
-    qemu_madvise(new_block->host, size, QEMU_MADV_DONTFORK);
+    qemu_ram_setup_dump(new_block->host, new_block->length);
+    qemu_madvise(new_block->host, new_block->length, QEMU_MADV_HUGEPAGE);
+    qemu_madvise(new_block->host, new_block->length, QEMU_MADV_DONTFORK);
 
-    if (kvm_enabled())
-        kvm_setup_guest_memory(new_block->host, size);
+    if (kvm_enabled()) {
+        kvm_setup_guest_memory(new_block->host, new_block->length);
+    }
 
     return new_block->offset;
 }
 
+ram_addr_t qemu_ram_alloc_from_file(ram_addr_t size, MemoryRegion *mr,
+                                    const char *mem_path)
+{
+    RAMBlock *new_block;
+
+    if (xen_enabled()) {
+        fprintf(stderr, "-mem-path not supported with Xen\n");
+        exit(1);
+    }
+
+    if (phys_mem_alloc != qemu_anon_ram_alloc) {
+        /*
+         * file_ram_alloc() needs to allocate just like
+         * phys_mem_alloc, but we haven't bothered to provide
+         * a hook there.
+         */
+        fprintf(stderr,
+                "-mem-path not supported with this accelerator\n");
+        exit(1);
+    }
+
+    size = TARGET_PAGE_ALIGN(size);
+    new_block = g_malloc0(sizeof(*new_block));
+    new_block->mr = mr;
+    new_block->length = size;
+    new_block->host = file_ram_alloc(new_block, size, mem_path);
+    return ram_block_add(new_block);
+}
+
+ram_addr_t qemu_ram_alloc_from_ptr(ram_addr_t size, void *host,
+                                   MemoryRegion *mr)
+{
+    RAMBlock *new_block;
+
+    size = TARGET_PAGE_ALIGN(size);
+    new_block = g_malloc0(sizeof(*new_block));
+    new_block->mr = mr;
+    new_block->length = size;
+    new_block->fd = -1;
+    new_block->host = host;
+    if (host) {
+        new_block->flags |= RAM_PREALLOC_MASK;
+    }
+    return ram_block_add(new_block);
+}
+
 ram_addr_t qemu_ram_alloc(ram_addr_t size, MemoryRegion *mr)
 {
     return qemu_ram_alloc_from_ptr(size, NULL, mr);

index ed28f1e..eaddea6 100644
@@ -325,9 +325,6 @@ typedef struct RAMList {
 } RAMList;
 extern RAMList ram_list;
 
-extern const char *mem_path;
-extern int mem_prealloc;
-
 /* Flags stored in the low bits of the TLB virtual address.  These are
    defined so that fast path ram access is all zeros.  */
 /* Zero if TLB entry is valid.  */

index 2edfa96..dedb258 100644
@@ -22,6 +22,8 @@
 #ifndef CONFIG_USER_ONLY
 #include "hw/xen/xen.h"
 
+ram_addr_t qemu_ram_alloc_from_file(ram_addr_t size, MemoryRegion *mr,
+                                    const char *mem_path);
 ram_addr_t qemu_ram_alloc_from_ptr(ram_addr_t size, void *host,
                                    MemoryRegion *mr);
 ram_addr_t qemu_ram_alloc(ram_addr_t size, MemoryRegion *mr);

index 1e141e3..277230d 100644
@@ -133,6 +133,8 @@ extern uint8_t *boot_splash_filedata;
 extern size_t boot_splash_filedata_size;
 extern uint8_t qemu_extra_params_fw[2];
 extern QEMUClockType rtc_clock;
+extern const char *mem_path;
+extern int mem_prealloc;
 
 #define MAX_NODES 128
 

index 93afea7..43b90eb 100644
@@ -23,6 +23,7 @@
 
 #include "exec/memory-internal.h"
 #include "exec/ram_addr.h"
+#include "sysemu/sysemu.h"
 
 //#define DEBUG_UNASSIGNED
 

@@ -1016,7 +1017,11 @@ void memory_region_init_ram(MemoryRegion *mr,
     mr->ram = true;
     mr->terminates = true;
     mr->destructor = memory_region_destructor_ram;
-    mr->ram_addr = qemu_ram_alloc(size, mr);
+    if (mem_path) {
+        mr->ram_addr = qemu_ram_alloc_from_file(size, mr, mem_path);
+    } else {
+        mr->ram_addr = qemu_ram_alloc(size, mr);
+    }
 }
 
 void memory_region_init_ram_ptr(MemoryRegion *mr,

index 525fc04..cbbb2a7 100644
@@ -1129,14 +1129,6 @@ error:
     }
     return NULL;
 }
-#else
-static void *file_ram_alloc(RAMBlock *block,
-                            ram_addr_t memory,
-                            const char *path)
-{
-    fprintf(stderr, "-mem-path not supported on this host\n");
-    exit(1);
-}
 #endif
 
 static ram_addr_t find_ram_offset(ram_addr_t size)

@@ -1310,6 +1302,7 @@ static ram_addr_t ram_block_add(RAMBlock *new_block)
     return new_block->offset;
 }
 
+#ifdef __linux__
 ram_addr_t qemu_ram_alloc_from_file(ram_addr_t size, MemoryRegion *mr,
                                     const char *mem_path)
 {

@@ -1338,6 +1331,7 @@ ram_addr_t qemu_ram_alloc_from_file(ram_addr_t size, MemoryRegion *mr,
     new_block->host = file_ram_alloc(new_block, size, mem_path);
     return ram_block_add(new_block);
 }
+#endif
 
 ram_addr_t qemu_ram_alloc_from_ptr(ram_addr_t size, void *host,
                                    MemoryRegion *mr)

index ab11c32..58c3fe4 100644
@@ -311,6 +311,24 @@ void memory_region_init_ram(MemoryRegion *mr,
                             const char *name,
                             uint64_t size);
 
+#ifdef __linux__
+/**
+ * memory_region_init_ram_from_file:  Initialize RAM memory region with a
+ *                                    mmap-ed backend.
+ *
+ * @mr: the #MemoryRegion to be initialized.
+ * @owner: the object that tracks the region's reference count
+ * @name: the name of the region.
+ * @size: size of the region.
+ * @path: the path in which to allocate the RAM.
+ */
+void memory_region_init_ram_from_file(MemoryRegion *mr,
+                                      struct Object *owner,
+                                      const char *name,
+                                      uint64_t size,
+                                      const char *path);
+#endif
+
 /**
  * memory_region_init_ram_ptr:  Initialize RAM memory region from a
  *                              user-provided pointer.  Accesses into the

index 43b90eb..6192377 100644
@@ -1017,13 +1017,24 @@ void memory_region_init_ram(MemoryRegion *mr,
     mr->ram = true;
     mr->terminates = true;
     mr->destructor = memory_region_destructor_ram;
-    if (mem_path) {
-        mr->ram_addr = qemu_ram_alloc_from_file(size, mr, mem_path);
-    } else {
-        mr->ram_addr = qemu_ram_alloc(size, mr);
-    }
+    mr->ram_addr = qemu_ram_alloc(size, mr);
 }
 
+#ifdef __linux__
+void memory_region_init_ram_from_file(MemoryRegion *mr,
+                                      struct Object *owner,
+                                      const char *name,
+                                      uint64_t size,
+                                      const char *path)
+{
+    memory_region_init(mr, owner, name, size);
+    mr->ram = true;
+    mr->terminates = true;
+    mr->destructor = memory_region_destructor_ram;
+    mr->ram_addr = qemu_ram_alloc_from_file(size, mr, path);
+}
+#endif
+
 void memory_region_init_ram_ptr(MemoryRegion *mr,
                                 Object *owner,
                                 const char *name,

index ce43e69..7846ba8 100644
@@ -228,7 +228,16 @@ static void allocate_system_memory_nonnuma(MemoryRegion *mr, Object *owner,
                                            const char *name,
                                            uint64_t ram_size)
 {
-    memory_region_init_ram(mr, owner, name, ram_size);
+    if (mem_path) {
+#ifdef __linux__
+        memory_region_init_ram_from_file(mr, owner, name, ram_size, mem_path);
+#else
+        fprintf(stderr, "-mem-path not supported on this host\n");
+        exit(1);
+#endif
+    } else {
+        memory_region_init_ram(mr, owner, name, ram_size);
+    }
     vmstate_register_ram_global(mr);
 }
 

index cbbb2a7..36301e2 100644
@@ -1020,7 +1020,8 @@ static void sigbus_handler(int signal)
 
 static void *file_ram_alloc(RAMBlock *block,
                             ram_addr_t memory,
-                            const char *path)
+                            const char *path,
+                            Error **errp)
 {
     char *filename;
     char *sanitized_name;

@@ -1039,7 +1040,8 @@ static void *file_ram_alloc(RAMBlock *block,
     }
 
     if (kvm_enabled() && !kvm_has_sync_mmu()) {
-        fprintf(stderr, "host lacks kvm mmu notifiers, -mem-path unsupported\n");
+        error_setg(errp,
+                   "host lacks kvm mmu notifiers, -mem-path unsupported\n");
         goto error;
     }
 

@@ -1056,7 +1058,8 @@ static void *file_ram_alloc(RAMBlock *block,
 
     fd = mkstemp(filename);
     if (fd < 0) {
-        perror("unable to create backing store for hugepages");
+        error_setg_errno(errp, errno,
+                         "unable to create backing store for hugepages");
         g_free(filename);
         goto error;
     }

@@ -1071,12 +1074,14 @@ static void *file_ram_alloc(RAMBlock *block,
      * If anything goes wrong with it under other filesystems,
      * mmap will fail.
      */
-    if (ftruncate(fd, memory))
+    if (ftruncate(fd, memory)) {
         perror("ftruncate");
+    }
 
     area = mmap(0, memory, PROT_READ | PROT_WRITE, MAP_PRIVATE, fd, 0);
     if (area == MAP_FAILED) {
-        perror("file_ram_alloc: can't mmap RAM pages");
+        error_setg_errno(errp, errno,
+                         "unable to map backing store for hugepages");
         close(fd);
         goto error;
     }

@@ -1304,13 +1309,14 @@ static ram_addr_t ram_block_add(RAMBlock *new_block)
 
 #ifdef __linux__
 ram_addr_t qemu_ram_alloc_from_file(ram_addr_t size, MemoryRegion *mr,
-                                    const char *mem_path)
+                                    const char *mem_path,
+                                    Error **errp)
 {
     RAMBlock *new_block;
 
     if (xen_enabled()) {
-        fprintf(stderr, "-mem-path not supported with Xen\n");
-        exit(1);
+        error_setg(errp, "-mem-path not supported with Xen\n");
+        return -1;
     }
 
     if (phys_mem_alloc != qemu_anon_ram_alloc) {

@@ -1319,16 +1325,22 @@ ram_addr_t qemu_ram_alloc_from_file(ram_addr_t size, MemoryRegion *mr,
          * phys_mem_alloc, but we haven't bothered to provide
          * a hook there.
          */
-        fprintf(stderr,
-                "-mem-path not supported with this accelerator\n");
-        exit(1);
+        error_setg(errp,
+                   "-mem-path not supported with this accelerator\n");
+        return -1;
     }
 
     size = TARGET_PAGE_ALIGN(size);
     new_block = g_malloc0(sizeof(*new_block));
     new_block->mr = mr;
     new_block->length = size;
-    new_block->host = file_ram_alloc(new_block, size, mem_path);
+    new_block->host = file_ram_alloc(new_block, size,
+                                     mem_path, errp);
+    if (!new_block->host) {
+        g_free(new_block);
+        return -1;
+    }
+
     return ram_block_add(new_block);
 }
 #endif

index 58c3fe4..82d7781 100644
@@ -31,6 +31,7 @@
 #include "qemu/queue.h"
 #include "qemu/int128.h"
 #include "qemu/notify.h"
+#include "qapi/error.h"
 
 #define MAX_PHYS_ADDR_SPACE_BITS 62
 #define MAX_PHYS_ADDR            (((hwaddr)1 << MAX_PHYS_ADDR_SPACE_BITS) - 1)

@@ -321,12 +322,14 @@ void memory_region_init_ram(MemoryRegion *mr,
  * @name: the name of the region.
  * @size: size of the region.
  * @path: the path in which to allocate the RAM.
+ * @errp: pointer to Error*, to store an error if it happens.
  */
 void memory_region_init_ram_from_file(MemoryRegion *mr,
                                       struct Object *owner,
                                       const char *name,
                                       uint64_t size,
-                                      const char *path);
+                                      const char *path,
+                                      Error **errp);
 #endif
 
 /**

index dedb258..f9518a6 100644
@@ -23,7 +23,7 @@
 #include "hw/xen/xen.h"
 
 ram_addr_t qemu_ram_alloc_from_file(ram_addr_t size, MemoryRegion *mr,
-                                    const char *mem_path);
+                                    const char *mem_path, Error **errp);
 ram_addr_t qemu_ram_alloc_from_ptr(ram_addr_t size, void *host,
                                    MemoryRegion *mr);
 ram_addr_t qemu_ram_alloc(ram_addr_t size, MemoryRegion *mr);

index 6192377..310729a 100644
@@ -1025,13 +1025,14 @@ void memory_region_init_ram_from_file(MemoryRegion *mr,
                                       struct Object *owner,
                                       const char *name,
                                       uint64_t size,
-                                      const char *path)
+                                      const char *path,
+                                      Error **errp)
 {
     memory_region_init(mr, owner, name, size);
     mr->ram = true;
     mr->terminates = true;
     mr->destructor = memory_region_destructor_ram;
-    mr->ram_addr = qemu_ram_alloc_from_file(size, mr, path);
+    mr->ram_addr = qemu_ram_alloc_from_file(size, mr, path, errp);
 }
 #endif
 

index 7846ba8..039d401 100644
@@ -230,7 +230,18 @@ static void allocate_system_memory_nonnuma(MemoryRegion *mr, Object *owner,
 {
     if (mem_path) {
 #ifdef __linux__
-        memory_region_init_ram_from_file(mr, owner, name, ram_size, mem_path);
+        Error *err = NULL;
+        memory_region_init_ram_from_file(mr, owner, name, ram_size,
+                                         mem_path, &err);
+
+        /* Legacy behavior: if allocation failed, fall back to
+         * regular RAM allocation.
+         */
+        if (!memory_region_size(mr)) {
+            qerror_report_err(err);
+            error_free(err);
+            memory_region_init_ram(mr, owner, name, ram_size);
+        }
 #else
         fprintf(stderr, "-mem-path not supported on this host\n");
         exit(1);

index 36301e2..b640425 100644
@@ -1011,13 +1011,6 @@ static long gethugepagesize(const char *path)
     return fs.f_bsize;
 }
 
-static sigjmp_buf sigjump;
-
-static void sigbus_handler(int signal)
-{
-    siglongjmp(sigjump, 1);
-}
-
 static void *file_ram_alloc(RAMBlock *block,
                             ram_addr_t memory,
                             const char *path,

@@ -1087,42 +1080,7 @@ static void *file_ram_alloc(RAMBlock *block,
     }
 
     if (mem_prealloc) {
-        int ret, i;
-        struct sigaction act, oldact;
-        sigset_t set, oldset;
-
-        memset(&act, 0, sizeof(act));
-        act.sa_handler = &sigbus_handler;
-        act.sa_flags = 0;
-
-        ret = sigaction(SIGBUS, &act, &oldact);
-        if (ret) {
-            perror("file_ram_alloc: failed to install signal handler");
-            exit(1);
-        }
-
-        /* unblock SIGBUS */
-        sigemptyset(&set);
-        sigaddset(&set, SIGBUS);
-        pthread_sigmask(SIG_UNBLOCK, &set, &oldset);
-
-        if (sigsetjmp(sigjump, 1)) {
-            fprintf(stderr, "file_ram_alloc: failed to preallocate pages\n");
-            exit(1);
-        }
-
-        /* MAP_POPULATE silently ignores failures */
-        for (i = 0; i < (memory/hpagesize); i++) {
-            memset(area + (hpagesize*i), 0, 1);
-        }
-
-        ret = sigaction(SIGBUS, &oldact, NULL);
-        if (ret) {
-            perror("file_ram_alloc: failed to reinstall signal handler");
-            exit(1);
-        }
-
-        pthread_sigmask(SIG_SETMASK, &oldset, NULL);
+        os_mem_prealloc(fd, area, memory);
     }
 
     block->fd = fd;

index ffb2966..9c1a119 100644
@@ -251,4 +251,6 @@ void qemu_init_auxval(char **envp);
 
 void qemu_set_tty_echo(int fd, bool echo);
 
+void os_mem_prealloc(int fd, char *area, size_t sz);
+
 #endif

91 lines not on this page · raw message

index b640425..739f0cf 100644
@@ -70,6 +70,9 @@ AddressSpace address_space_memory;
 MemoryRegion io_mem_rom, io_mem_notdirty;
 static MemoryRegion io_mem_unassigned;
 
+/* RAM is pre-allocated and passed into qemu_ram_alloc_from_ptr */
+#define RAM_PREALLOC   (1 << 0)
+
 #endif
 
 struct CPUTailQ cpus = QTAILQ_HEAD_INITIALIZER(cpus);

@@ -1315,7 +1318,7 @@ ram_addr_t qemu_ram_alloc_from_ptr(ram_addr_t size, void *host,
     new_block->fd = -1;
     new_block->host = host;
     if (host) {
-        new_block->flags |= RAM_PREALLOC_MASK;
+        new_block->flags |= RAM_PREALLOC;
     }
     return ram_block_add(new_block);
 }

@@ -1354,7 +1357,7 @@ void qemu_ram_free(ram_addr_t addr)
             QTAILQ_REMOVE(&ram_list.blocks, block, next);
             ram_list.mru_block = NULL;
             ram_list.version++;
-            if (block->flags & RAM_PREALLOC_MASK) {
+            if (block->flags & RAM_PREALLOC) {
                 ;
             } else if (xen_enabled()) {
                 xen_invalidate_map_cache_entry(block->host);

@@ -1386,7 +1389,7 @@ void qemu_ram_remap(ram_addr_t addr, ram_addr_t length)
         offset = addr - block->offset;
         if (offset < block->length) {
             vaddr = block->host + offset;
-            if (block->flags & RAM_PREALLOC_MASK) {
+            if (block->flags & RAM_PREALLOC) {
                 ;
             } else if (xen_enabled()) {
                 abort();

index eaddea6..f91581f 100644
@@ -297,9 +297,6 @@ CPUArchState *cpu_copy(CPUArchState *env);
 
 /* memory API */
 
-/* RAM is pre-allocated and passed into qemu_ram_alloc_from_ptr */
-#define RAM_PREALLOC_MASK   (1 << 0)
-
 typedef struct RAMBlock {
     struct MemoryRegion *mr;
     uint8_t *host;

107 lines not on this page · raw message