@@ -237,6 +237,9 @@ typedef struct {
* purposes only.
* Applies only to default memory backend, i.e., explicit memory backend
* wasn't used.
+ * @smbios_memory_device_size:
+ * Default size of memory device,
+ * SMBIOS 3.1.0 "7.18 Memory Device (Type 17)"
*/
struct MachineClass {
/*< private >*/
@@ -304,6 +307,7 @@ struct MachineClass {
const CPUArchIdList *(*possible_cpu_arch_ids)(MachineState *machine);
int64_t (*get_default_cpu_node_id)(const MachineState *ms, int idx);
ram_addr_t (*fixup_ram_size)(ram_addr_t size);
+ uint64_t smbios_memory_device_size;
};
/**
@@ -3308,6 +3308,7 @@ DEFINE_VIRT_MACHINE_AS_LATEST(9, 1)
static void virt_machine_9_0_options(MachineClass *mc)
{
virt_machine_9_1_options(mc);
+ mc->smbios_memory_device_size = 16 * GiB;
compat_props_add(mc->compat_props, hw_compat_9_0, hw_compat_9_0_len);
}
DEFINE_VIRT_MACHINE(9, 0)
@@ -1004,6 +1004,12 @@ static void machine_class_init(ObjectClass *oc, void *data)
/* Default 128 MB as guest ram size */
mc->default_ram_size = 128 * MiB;
mc->rom_file_has_mr = true;
+ /*
+ * SMBIOS 3.1.0 7.18.5 Memory Device — Extended Size
+ * use max possible value that could be encoded into
+ * 'Extended Size' field (2047Tb).
+ */
+ mc->smbios_memory_device_size = 2047 * TiB;
/* numa node memory size aligned on 8MB by default.
* On Linux, each node's border has to be 8MB aligned
@@ -495,6 +495,7 @@ static void pc_i440fx_machine_9_0_options(MachineClass *m)
pc_i440fx_machine_9_1_options(m);
m->alias = NULL;
m->is_default = false;
+ m->smbios_memory_device_size = 16 * GiB;
compat_props_add(m->compat_props, hw_compat_9_0, hw_compat_9_0_len);
compat_props_add(m->compat_props, pc_compat_9_0, pc_compat_9_0_len);
@@ -374,6 +374,7 @@ static void pc_q35_machine_9_0_options(MachineClass *m)
PCMachineClass *pcmc = PC_MACHINE_CLASS(m);
pc_q35_machine_9_1_options(m);
m->alias = NULL;
+ m->smbios_memory_device_size = 16 * GiB;
compat_props_add(m->compat_props, hw_compat_9_0, hw_compat_9_0_len);
compat_props_add(m->compat_props, pc_compat_9_0, pc_compat_9_0_len);
pcmc->isa_bios_alias = false;
@@ -1093,6 +1093,7 @@ static bool smbios_get_tables_ep(MachineState *ms,
Error **errp)
{
unsigned i, dimm_cnt, offset;
+ MachineClass *mc = MACHINE_GET_CLASS(ms);
ERRP_GUARD();
assert(ep_type == SMBIOS_ENTRY_POINT_TYPE_32 ||
@@ -1123,12 +1124,12 @@ static bool smbios_get_tables_ep(MachineState *ms,
smbios_build_type_9_table(errp);
smbios_build_type_11_table();
-#define MAX_DIMM_SZ (16 * GiB)
-#define GET_DIMM_SZ ((i < dimm_cnt - 1) ? MAX_DIMM_SZ \
- : ((current_machine->ram_size - 1) % MAX_DIMM_SZ) + 1)
+#define GET_DIMM_SZ ((i < dimm_cnt - 1) ? mc->smbios_memory_device_size \
+ : ((current_machine->ram_size - 1) % mc->smbios_memory_device_size) + 1)
- dimm_cnt = QEMU_ALIGN_UP(current_machine->ram_size, MAX_DIMM_SZ) /
- MAX_DIMM_SZ;
+ dimm_cnt = QEMU_ALIGN_UP(current_machine->ram_size,
+ mc->smbios_memory_device_size) /
+ mc->smbios_memory_device_size;
/*
* The offset determines if we need to keep additional space between
Currently QEMU describes initial[1] RAM* in SMBIOS as a series of virtual DIMMs (capped at 16Gb max) using type 17 structure entries. Which is fine for the most cases. However when starting guest with terabytes of RAM this leads to too many memory device structures, which eventually upsets linux kernel as it reserves only 64K for these entries and when that border is crossed out it runs out of reserved memory. Instead of partitioning initial RAM on 16Gb DIMMs, use maximum possible chunk size that SMBIOS spec allows[2]. Which lets encode RAM in lower 31 bits of 32bit field (which amounts upto 2047Tb per DIMM). As result initial RAM will generate only one type 17 structure until host/guest reach ability to use more RAM in the future. Compat changes: We can't unconditionally change chunk size as it will break QEMU<->guest ABI (and migration). Thus introduce a new machine class field that would let older versioned machines to use legacy 16Gb chunks, while new(er) machine type[s] use maximum possible chunk size. PS: While it might seem to be risky to rise max entry size this large (much beyond of what current physical RAM modules support), I'd not expect it causing much issues, modulo uncovering bugs in software running within guest. And those should be fixed on guest side to handle SMBIOS spec properly, especially if guest is expected to support so huge RAM configs. In worst case, QEMU can reduce chunk size later if we would care enough about introducing a workaround for some 'unfixable' guest OS, either by fixing up the next machine type or giving users a CLI option to customize it. 1) Initial RAM - is RAM configured with help '-m SIZE' CLI option/ implicitly defined by machine. It doesn't include memory configured with help of '-device' option[s] (pcdimm,nvdimm,...) 2) SMBIOS 3.1.0 7.18.5 Memory Device — Extended Size PS: * tested on 8Tb host with RHEL6 guest, which seems to parse type 17 SMBIOS table entries correctly (according to 'dmidecode'). Signed-off-by: Igor Mammedov <imammedo@redhat.com> --- v2: * add comment in the code describing where 2047Tb comes from (mst) * rephrase commit message a bit and clarify what RAM it applies. --- include/hw/boards.h | 4 ++++ hw/arm/virt.c | 1 + hw/core/machine.c | 6 ++++++ hw/i386/pc_piix.c | 1 + hw/i386/pc_q35.c | 1 + hw/smbios/smbios.c | 11 ++++++----- 6 files changed, 19 insertions(+), 5 deletions(-)