2005-04-17 02:20:36 +04:00
|
|
|
/*
|
|
|
|
* ACPI 3.0 based NUMA setup
|
|
|
|
* Copyright 2004 Andi Kleen, SuSE Labs.
|
|
|
|
*
|
|
|
|
* Reads the ACPI SRAT table to figure out what memory belongs to which CPUs.
|
|
|
|
*
|
|
|
|
* Called from acpi_numa_init while reading the SRAT and SLIT tables.
|
|
|
|
* Assumes all memory regions belonging to a single proximity domain
|
|
|
|
* are in one chunk. Holes between them will be included in the node.
|
|
|
|
*/
|
|
|
|
|
|
|
|
#include <linux/kernel.h>
|
|
|
|
#include <linux/acpi.h>
|
|
|
|
#include <linux/mmzone.h>
|
|
|
|
#include <linux/bitmap.h>
|
|
|
|
#include <linux/module.h>
|
|
|
|
#include <linux/topology.h>
|
2006-04-07 21:49:18 +04:00
|
|
|
#include <linux/bootmem.h>
|
2010-08-26 00:39:17 +04:00
|
|
|
#include <linux/memblock.h>
|
2006-04-07 21:49:18 +04:00
|
|
|
#include <linux/mm.h>
|
2005-04-17 02:20:36 +04:00
|
|
|
#include <asm/proto.h>
|
|
|
|
#include <asm/numa.h>
|
2006-01-12 00:44:39 +03:00
|
|
|
#include <asm/e820.h>
|
2009-02-17 15:58:15 +03:00
|
|
|
#include <asm/apic.h>
|
2009-01-21 12:24:27 +03:00
|
|
|
#include <asm/uv/uv.h>
|
2005-04-17 02:20:36 +04:00
|
|
|
|
2006-09-26 12:52:33 +04:00
|
|
|
int acpi_numa __initdata;
|
|
|
|
|
2005-04-17 02:20:36 +04:00
|
|
|
static __init int setup_node(int pxm)
|
|
|
|
{
|
2006-06-23 13:03:19 +04:00
|
|
|
return acpi_map_pxm_to_node(pxm);
|
2005-04-17 02:20:36 +04:00
|
|
|
}
|
|
|
|
|
|
|
|
static __init void bad_srat(void)
|
|
|
|
{
|
|
|
|
printk(KERN_ERR "SRAT: SRAT not used.\n");
|
|
|
|
acpi_numa = -1;
|
|
|
|
}
|
|
|
|
|
|
|
|
static __init inline int srat_disabled(void)
|
|
|
|
{
|
2011-02-16 14:13:06 +03:00
|
|
|
return acpi_numa < 0;
|
2005-04-17 02:20:36 +04:00
|
|
|
}
|
|
|
|
|
|
|
|
/* Callback for SLIT parsing */
|
|
|
|
void __init acpi_numa_slit_init(struct acpi_table_slit *slit)
|
|
|
|
{
|
2011-02-16 19:11:09 +03:00
|
|
|
int i, j;
|
2008-07-11 07:36:37 +04:00
|
|
|
|
2011-02-16 19:11:09 +03:00
|
|
|
for (i = 0; i < slit->locality_count; i++)
|
|
|
|
for (j = 0; j < slit->locality_count; j++)
|
|
|
|
numa_set_distance(pxm_to_node(i), pxm_to_node(j),
|
|
|
|
slit->entry[slit->locality_count * i + j]);
|
2005-04-17 02:20:36 +04:00
|
|
|
}
|
|
|
|
|
2009-03-31 01:55:30 +04:00
|
|
|
/* Callback for Proximity Domain -> x2APIC mapping */
|
|
|
|
void __init
|
|
|
|
acpi_numa_x2apic_affinity_init(struct acpi_srat_x2apic_cpu_affinity *pa)
|
|
|
|
{
|
|
|
|
int pxm, node;
|
|
|
|
int apic_id;
|
|
|
|
|
|
|
|
if (srat_disabled())
|
|
|
|
return;
|
|
|
|
if (pa->header.length < sizeof(struct acpi_srat_x2apic_cpu_affinity)) {
|
|
|
|
bad_srat();
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
if ((pa->flags & ACPI_SRAT_CPU_ENABLED) == 0)
|
|
|
|
return;
|
|
|
|
pxm = pa->proximity_domain;
|
2011-12-22 05:45:16 +04:00
|
|
|
apic_id = pa->apic_id;
|
2012-03-16 23:25:35 +04:00
|
|
|
if (!apic->apic_id_valid(apic_id)) {
|
2011-12-22 05:45:16 +04:00
|
|
|
printk(KERN_INFO "SRAT: PXM %u -> X2APIC 0x%04x ignored\n",
|
|
|
|
pxm, apic_id);
|
|
|
|
return;
|
|
|
|
}
|
2009-03-31 01:55:30 +04:00
|
|
|
node = setup_node(pxm);
|
|
|
|
if (node < 0) {
|
|
|
|
printk(KERN_ERR "SRAT: Too many proximity domains %x\n", pxm);
|
|
|
|
bad_srat();
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
2010-12-17 06:09:58 +03:00
|
|
|
if (apic_id >= MAX_LOCAL_APIC) {
|
|
|
|
printk(KERN_INFO "SRAT: PXM %u -> APIC 0x%04x -> Node %u skipped apicid that is too big\n", pxm, apic_id, node);
|
|
|
|
return;
|
|
|
|
}
|
2011-01-23 16:37:39 +03:00
|
|
|
set_apicid_to_node(apic_id, node);
|
2011-02-16 19:11:09 +03:00
|
|
|
node_set(node, numa_nodes_parsed);
|
2009-03-31 01:55:30 +04:00
|
|
|
acpi_numa = 1;
|
2009-11-21 11:23:37 +03:00
|
|
|
printk(KERN_INFO "SRAT: PXM %u -> APIC 0x%04x -> Node %u\n",
|
2009-03-31 01:55:30 +04:00
|
|
|
pxm, apic_id, node);
|
|
|
|
}
|
|
|
|
|
2005-04-17 02:20:36 +04:00
|
|
|
/* Callback for Proximity Domain -> LAPIC mapping */
|
|
|
|
void __init
|
2007-02-02 19:48:22 +03:00
|
|
|
acpi_numa_processor_affinity_init(struct acpi_srat_cpu_affinity *pa)
|
2005-04-17 02:20:36 +04:00
|
|
|
{
|
|
|
|
int pxm, node;
|
2008-01-30 15:33:10 +03:00
|
|
|
int apic_id;
|
|
|
|
|
2006-02-03 23:51:26 +03:00
|
|
|
if (srat_disabled())
|
|
|
|
return;
|
2007-02-02 19:48:22 +03:00
|
|
|
if (pa->header.length != sizeof(struct acpi_srat_cpu_affinity)) {
|
2006-05-15 20:19:44 +04:00
|
|
|
bad_srat();
|
2006-02-03 23:51:26 +03:00
|
|
|
return;
|
|
|
|
}
|
2007-02-02 19:48:22 +03:00
|
|
|
if ((pa->flags & ACPI_SRAT_CPU_ENABLED) == 0)
|
2005-04-17 02:20:36 +04:00
|
|
|
return;
|
2007-02-02 19:48:22 +03:00
|
|
|
pxm = pa->proximity_domain_lo;
|
2012-01-17 13:20:31 +04:00
|
|
|
if (acpi_srat_revision >= 2)
|
|
|
|
pxm |= *((unsigned int*)pa->proximity_domain_hi) << 8;
|
2005-04-17 02:20:36 +04:00
|
|
|
node = setup_node(pxm);
|
|
|
|
if (node < 0) {
|
|
|
|
printk(KERN_ERR "SRAT: Too many proximity domains %x\n", pxm);
|
|
|
|
bad_srat();
|
|
|
|
return;
|
|
|
|
}
|
2008-02-17 10:00:22 +03:00
|
|
|
|
2008-09-24 00:37:13 +04:00
|
|
|
if (get_uv_system_type() >= UV_X2APIC)
|
2008-03-28 22:12:08 +03:00
|
|
|
apic_id = (pa->apic_id << 8) | pa->local_sapic_eid;
|
|
|
|
else
|
|
|
|
apic_id = pa->apic_id;
|
2010-12-17 06:09:58 +03:00
|
|
|
|
|
|
|
if (apic_id >= MAX_LOCAL_APIC) {
|
|
|
|
printk(KERN_INFO "SRAT: PXM %u -> APIC 0x%02x -> Node %u skipped apicid that is too big\n", pxm, apic_id, node);
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
2011-01-23 16:37:39 +03:00
|
|
|
set_apicid_to_node(apic_id, node);
|
2011-02-16 19:11:09 +03:00
|
|
|
node_set(node, numa_nodes_parsed);
|
2005-04-17 02:20:36 +04:00
|
|
|
acpi_numa = 1;
|
2009-11-21 11:23:37 +03:00
|
|
|
printk(KERN_INFO "SRAT: PXM %u -> APIC 0x%02x -> Node %u\n",
|
2008-01-30 15:33:10 +03:00
|
|
|
pxm, apic_id, node);
|
2005-04-17 02:20:36 +04:00
|
|
|
}
|
|
|
|
|
2011-05-02 16:18:51 +04:00
|
|
|
#ifdef CONFIG_MEMORY_HOTPLUG
|
2006-10-01 10:27:05 +04:00
|
|
|
static inline int save_add_info(void) {return 1;}
|
|
|
|
#else
|
|
|
|
static inline int save_add_info(void) {return 0;}
|
|
|
|
#endif
|
2006-04-07 21:49:18 +04:00
|
|
|
|
2013-02-23 04:33:46 +04:00
|
|
|
#ifdef CONFIG_HAVE_MEMBLOCK_NODE_MAP
|
acpi, memory-hotplug: support getting hotplug info from SRAT
We now provide an option for users who don't want to specify physical
memory address in kernel commandline.
/*
* For movablemem_map=acpi:
*
* SRAT: |_____| |_____| |_________| |_________| ......
* node id: 0 1 1 2
* hotpluggable: n y y n
* movablemem_map: |_____| |_________|
*
* Using movablemem_map, we can prevent memblock from allocating memory
* on ZONE_MOVABLE at boot time.
*/
So user just specify movablemem_map=acpi, and the kernel will use
hotpluggable info in SRAT to determine which memory ranges should be set
as ZONE_MOVABLE.
If all the memory ranges in SRAT is hotpluggable, then no memory can be
used by kernel. But before parsing SRAT, memblock has already reserve
some memory ranges for other purposes, such as for kernel image, and so
on. We cannot prevent kernel from using these memory. So we need to
exclude these ranges even if these memory is hotpluggable.
Furthermore, there could be several memory ranges in the single node
which the kernel resides in. We may skip one range that have memory
reserved by memblock, but if the rest of memory is too small, then the
kernel will fail to boot. So, make the whole node which the kernel
resides in un-hotpluggable. Then the kernel has enough memory to use.
NOTE: Using this way will cause NUMA performance down because the
whole node will be set as ZONE_MOVABLE, and kernel cannot use memory
on it. If users don't want to lose NUMA performance, just don't use
it.
[akpm@linux-foundation.org: fix warning]
[akpm@linux-foundation.org: use strcmp()]
Signed-off-by: Tang Chen <tangchen@cn.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Jiang Liu <jiang.liu@huawei.com>
Cc: Jianguo Wu <wujianguo@huawei.com>
Cc: Kamezawa Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Wu Jianguo <wujianguo@huawei.com>
Cc: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Len Brown <lenb@kernel.org>
Cc: "Brown, Len" <len.brown@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-23 04:33:49 +04:00
|
|
|
static void __init
|
|
|
|
handle_movablemem(int node, u64 start, u64 end, u32 hotpluggable)
|
2013-02-23 04:33:46 +04:00
|
|
|
{
|
acpi, memory-hotplug: support getting hotplug info from SRAT
We now provide an option for users who don't want to specify physical
memory address in kernel commandline.
/*
* For movablemem_map=acpi:
*
* SRAT: |_____| |_____| |_________| |_________| ......
* node id: 0 1 1 2
* hotpluggable: n y y n
* movablemem_map: |_____| |_________|
*
* Using movablemem_map, we can prevent memblock from allocating memory
* on ZONE_MOVABLE at boot time.
*/
So user just specify movablemem_map=acpi, and the kernel will use
hotpluggable info in SRAT to determine which memory ranges should be set
as ZONE_MOVABLE.
If all the memory ranges in SRAT is hotpluggable, then no memory can be
used by kernel. But before parsing SRAT, memblock has already reserve
some memory ranges for other purposes, such as for kernel image, and so
on. We cannot prevent kernel from using these memory. So we need to
exclude these ranges even if these memory is hotpluggable.
Furthermore, there could be several memory ranges in the single node
which the kernel resides in. We may skip one range that have memory
reserved by memblock, but if the rest of memory is too small, then the
kernel will fail to boot. So, make the whole node which the kernel
resides in un-hotpluggable. Then the kernel has enough memory to use.
NOTE: Using this way will cause NUMA performance down because the
whole node will be set as ZONE_MOVABLE, and kernel cannot use memory
on it. If users don't want to lose NUMA performance, just don't use
it.
[akpm@linux-foundation.org: fix warning]
[akpm@linux-foundation.org: use strcmp()]
Signed-off-by: Tang Chen <tangchen@cn.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Jiang Liu <jiang.liu@huawei.com>
Cc: Jianguo Wu <wujianguo@huawei.com>
Cc: Kamezawa Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Wu Jianguo <wujianguo@huawei.com>
Cc: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Len Brown <lenb@kernel.org>
Cc: "Brown, Len" <len.brown@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-23 04:33:49 +04:00
|
|
|
int overlap, i;
|
2013-02-23 04:33:46 +04:00
|
|
|
unsigned long start_pfn, end_pfn;
|
|
|
|
|
|
|
|
start_pfn = PFN_DOWN(start);
|
|
|
|
end_pfn = PFN_UP(end);
|
|
|
|
|
|
|
|
/*
|
acpi, memory-hotplug: support getting hotplug info from SRAT
We now provide an option for users who don't want to specify physical
memory address in kernel commandline.
/*
* For movablemem_map=acpi:
*
* SRAT: |_____| |_____| |_________| |_________| ......
* node id: 0 1 1 2
* hotpluggable: n y y n
* movablemem_map: |_____| |_________|
*
* Using movablemem_map, we can prevent memblock from allocating memory
* on ZONE_MOVABLE at boot time.
*/
So user just specify movablemem_map=acpi, and the kernel will use
hotpluggable info in SRAT to determine which memory ranges should be set
as ZONE_MOVABLE.
If all the memory ranges in SRAT is hotpluggable, then no memory can be
used by kernel. But before parsing SRAT, memblock has already reserve
some memory ranges for other purposes, such as for kernel image, and so
on. We cannot prevent kernel from using these memory. So we need to
exclude these ranges even if these memory is hotpluggable.
Furthermore, there could be several memory ranges in the single node
which the kernel resides in. We may skip one range that have memory
reserved by memblock, but if the rest of memory is too small, then the
kernel will fail to boot. So, make the whole node which the kernel
resides in un-hotpluggable. Then the kernel has enough memory to use.
NOTE: Using this way will cause NUMA performance down because the
whole node will be set as ZONE_MOVABLE, and kernel cannot use memory
on it. If users don't want to lose NUMA performance, just don't use
it.
[akpm@linux-foundation.org: fix warning]
[akpm@linux-foundation.org: use strcmp()]
Signed-off-by: Tang Chen <tangchen@cn.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Jiang Liu <jiang.liu@huawei.com>
Cc: Jianguo Wu <wujianguo@huawei.com>
Cc: Kamezawa Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Wu Jianguo <wujianguo@huawei.com>
Cc: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Len Brown <lenb@kernel.org>
Cc: "Brown, Len" <len.brown@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-23 04:33:49 +04:00
|
|
|
* For movablemem_map=acpi:
|
|
|
|
*
|
|
|
|
* SRAT: |_____| |_____| |_________| |_________| ......
|
|
|
|
* node id: 0 1 1 2
|
|
|
|
* hotpluggable: n y y n
|
|
|
|
* movablemem_map: |_____| |_________|
|
|
|
|
*
|
|
|
|
* Using movablemem_map, we can prevent memblock from allocating memory
|
|
|
|
* on ZONE_MOVABLE at boot time.
|
|
|
|
*
|
|
|
|
* Before parsing SRAT, memblock has already reserve some memory ranges
|
|
|
|
* for other purposes, such as for kernel image. We cannot prevent
|
|
|
|
* kernel from using these memory, so we need to exclude these memory
|
|
|
|
* even if it is hotpluggable.
|
|
|
|
* Furthermore, to ensure the kernel has enough memory to boot, we make
|
|
|
|
* all the memory on the node which the kernel resides in
|
|
|
|
* un-hotpluggable.
|
|
|
|
*/
|
|
|
|
if (hotpluggable && movablemem_map.acpi) {
|
|
|
|
/* Exclude ranges reserved by memblock. */
|
|
|
|
struct memblock_type *rgn = &memblock.reserved;
|
|
|
|
|
|
|
|
for (i = 0; i < rgn->cnt; i++) {
|
|
|
|
if (end <= rgn->regions[i].base ||
|
|
|
|
start >= rgn->regions[i].base +
|
|
|
|
rgn->regions[i].size)
|
|
|
|
continue;
|
|
|
|
|
|
|
|
/*
|
|
|
|
* If the memory range overlaps the memory reserved by
|
|
|
|
* memblock, then the kernel resides in this node.
|
|
|
|
*/
|
|
|
|
node_set(node, movablemem_map.numa_nodes_kernel);
|
|
|
|
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* If the kernel resides in this node, then the whole node
|
|
|
|
* should not be hotpluggable.
|
|
|
|
*/
|
|
|
|
if (node_isset(node, movablemem_map.numa_nodes_kernel))
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
insert_movablemem_map(start_pfn, end_pfn);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* numa_nodes_hotplug nodemask represents which nodes are put
|
|
|
|
* into movablemem_map.map[].
|
|
|
|
*/
|
|
|
|
node_set(node, movablemem_map.numa_nodes_hotplug);
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
|
|
|
|
/*
|
|
|
|
* For movablemem_map=nn[KMG]@ss[KMG]:
|
2013-02-23 04:33:46 +04:00
|
|
|
*
|
|
|
|
* SRAT: |_____| |_____| |_________| |_________| ......
|
|
|
|
* node id: 0 1 1 2
|
|
|
|
* user specified: |__| |___|
|
|
|
|
* movablemem_map: |___| |_________| |______| ......
|
|
|
|
*
|
|
|
|
* Using movablemem_map, we can prevent memblock from allocating memory
|
|
|
|
* on ZONE_MOVABLE at boot time.
|
acpi, memory-hotplug: support getting hotplug info from SRAT
We now provide an option for users who don't want to specify physical
memory address in kernel commandline.
/*
* For movablemem_map=acpi:
*
* SRAT: |_____| |_____| |_________| |_________| ......
* node id: 0 1 1 2
* hotpluggable: n y y n
* movablemem_map: |_____| |_________|
*
* Using movablemem_map, we can prevent memblock from allocating memory
* on ZONE_MOVABLE at boot time.
*/
So user just specify movablemem_map=acpi, and the kernel will use
hotpluggable info in SRAT to determine which memory ranges should be set
as ZONE_MOVABLE.
If all the memory ranges in SRAT is hotpluggable, then no memory can be
used by kernel. But before parsing SRAT, memblock has already reserve
some memory ranges for other purposes, such as for kernel image, and so
on. We cannot prevent kernel from using these memory. So we need to
exclude these ranges even if these memory is hotpluggable.
Furthermore, there could be several memory ranges in the single node
which the kernel resides in. We may skip one range that have memory
reserved by memblock, but if the rest of memory is too small, then the
kernel will fail to boot. So, make the whole node which the kernel
resides in un-hotpluggable. Then the kernel has enough memory to use.
NOTE: Using this way will cause NUMA performance down because the
whole node will be set as ZONE_MOVABLE, and kernel cannot use memory
on it. If users don't want to lose NUMA performance, just don't use
it.
[akpm@linux-foundation.org: fix warning]
[akpm@linux-foundation.org: use strcmp()]
Signed-off-by: Tang Chen <tangchen@cn.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Jiang Liu <jiang.liu@huawei.com>
Cc: Jianguo Wu <wujianguo@huawei.com>
Cc: Kamezawa Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Wu Jianguo <wujianguo@huawei.com>
Cc: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Len Brown <lenb@kernel.org>
Cc: "Brown, Len" <len.brown@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-23 04:33:49 +04:00
|
|
|
*
|
|
|
|
* NOTE: In this case, SRAT info will be ingored.
|
2013-02-23 04:33:46 +04:00
|
|
|
*/
|
|
|
|
overlap = movablemem_map_overlap(start_pfn, end_pfn);
|
|
|
|
if (overlap >= 0) {
|
|
|
|
/*
|
|
|
|
* If part of this range is in movablemem_map, we need to
|
|
|
|
* add the range after it to extend the range to the end
|
|
|
|
* of the node, because from the min address specified to
|
|
|
|
* the end of the node will be ZONE_MOVABLE.
|
|
|
|
*/
|
|
|
|
start_pfn = max(start_pfn,
|
|
|
|
movablemem_map.map[overlap].start_pfn);
|
|
|
|
insert_movablemem_map(start_pfn, end_pfn);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Set the nodemask, so that if the address range on one node
|
|
|
|
* is not continuse, we can add the subsequent ranges on the
|
|
|
|
* same node into movablemem_map.
|
|
|
|
*/
|
|
|
|
node_set(node, movablemem_map.numa_nodes_hotplug);
|
|
|
|
} else {
|
|
|
|
if (node_isset(node, movablemem_map.numa_nodes_hotplug))
|
|
|
|
/*
|
|
|
|
* Insert the range if we already have movable ranges
|
|
|
|
* on the same node.
|
|
|
|
*/
|
|
|
|
insert_movablemem_map(start_pfn, end_pfn);
|
|
|
|
}
|
acpi, memory-hotplug: support getting hotplug info from SRAT
We now provide an option for users who don't want to specify physical
memory address in kernel commandline.
/*
* For movablemem_map=acpi:
*
* SRAT: |_____| |_____| |_________| |_________| ......
* node id: 0 1 1 2
* hotpluggable: n y y n
* movablemem_map: |_____| |_________|
*
* Using movablemem_map, we can prevent memblock from allocating memory
* on ZONE_MOVABLE at boot time.
*/
So user just specify movablemem_map=acpi, and the kernel will use
hotpluggable info in SRAT to determine which memory ranges should be set
as ZONE_MOVABLE.
If all the memory ranges in SRAT is hotpluggable, then no memory can be
used by kernel. But before parsing SRAT, memblock has already reserve
some memory ranges for other purposes, such as for kernel image, and so
on. We cannot prevent kernel from using these memory. So we need to
exclude these ranges even if these memory is hotpluggable.
Furthermore, there could be several memory ranges in the single node
which the kernel resides in. We may skip one range that have memory
reserved by memblock, but if the rest of memory is too small, then the
kernel will fail to boot. So, make the whole node which the kernel
resides in un-hotpluggable. Then the kernel has enough memory to use.
NOTE: Using this way will cause NUMA performance down because the
whole node will be set as ZONE_MOVABLE, and kernel cannot use memory
on it. If users don't want to lose NUMA performance, just don't use
it.
[akpm@linux-foundation.org: fix warning]
[akpm@linux-foundation.org: use strcmp()]
Signed-off-by: Tang Chen <tangchen@cn.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Jiang Liu <jiang.liu@huawei.com>
Cc: Jianguo Wu <wujianguo@huawei.com>
Cc: Kamezawa Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Wu Jianguo <wujianguo@huawei.com>
Cc: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Len Brown <lenb@kernel.org>
Cc: "Brown, Len" <len.brown@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-23 04:33:49 +04:00
|
|
|
out:
|
|
|
|
return;
|
2013-02-23 04:33:46 +04:00
|
|
|
}
|
|
|
|
#else /* CONFIG_HAVE_MEMBLOCK_NODE_MAP */
|
acpi, memory-hotplug: support getting hotplug info from SRAT
We now provide an option for users who don't want to specify physical
memory address in kernel commandline.
/*
* For movablemem_map=acpi:
*
* SRAT: |_____| |_____| |_________| |_________| ......
* node id: 0 1 1 2
* hotpluggable: n y y n
* movablemem_map: |_____| |_________|
*
* Using movablemem_map, we can prevent memblock from allocating memory
* on ZONE_MOVABLE at boot time.
*/
So user just specify movablemem_map=acpi, and the kernel will use
hotpluggable info in SRAT to determine which memory ranges should be set
as ZONE_MOVABLE.
If all the memory ranges in SRAT is hotpluggable, then no memory can be
used by kernel. But before parsing SRAT, memblock has already reserve
some memory ranges for other purposes, such as for kernel image, and so
on. We cannot prevent kernel from using these memory. So we need to
exclude these ranges even if these memory is hotpluggable.
Furthermore, there could be several memory ranges in the single node
which the kernel resides in. We may skip one range that have memory
reserved by memblock, but if the rest of memory is too small, then the
kernel will fail to boot. So, make the whole node which the kernel
resides in un-hotpluggable. Then the kernel has enough memory to use.
NOTE: Using this way will cause NUMA performance down because the
whole node will be set as ZONE_MOVABLE, and kernel cannot use memory
on it. If users don't want to lose NUMA performance, just don't use
it.
[akpm@linux-foundation.org: fix warning]
[akpm@linux-foundation.org: use strcmp()]
Signed-off-by: Tang Chen <tangchen@cn.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Jiang Liu <jiang.liu@huawei.com>
Cc: Jianguo Wu <wujianguo@huawei.com>
Cc: Kamezawa Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Wu Jianguo <wujianguo@huawei.com>
Cc: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Len Brown <lenb@kernel.org>
Cc: "Brown, Len" <len.brown@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-23 04:33:49 +04:00
|
|
|
static inline void
|
|
|
|
handle_movablemem(int node, u64 start, u64 end, u32 hotpluggable)
|
2013-02-23 04:33:46 +04:00
|
|
|
{
|
|
|
|
}
|
|
|
|
#endif /* CONFIG_HAVE_MEMBLOCK_NODE_MAP */
|
|
|
|
|
2005-04-17 02:20:36 +04:00
|
|
|
/* Callback for parsing of the Proximity Domain <-> Memory Area mappings */
|
2012-07-31 19:41:09 +04:00
|
|
|
int __init
|
2007-02-02 19:48:22 +03:00
|
|
|
acpi_numa_memory_affinity_init(struct acpi_srat_mem_affinity *ma)
|
2005-04-17 02:20:36 +04:00
|
|
|
{
|
2011-05-02 16:18:52 +04:00
|
|
|
u64 start, end;
|
2013-02-23 04:33:46 +04:00
|
|
|
u32 hotpluggable;
|
2005-04-17 02:20:36 +04:00
|
|
|
int node, pxm;
|
|
|
|
|
2006-02-03 23:51:26 +03:00
|
|
|
if (srat_disabled())
|
2013-01-09 04:18:41 +04:00
|
|
|
goto out_err;
|
|
|
|
if (ma->header.length != sizeof(struct acpi_srat_mem_affinity))
|
|
|
|
goto out_err_bad_srat;
|
2007-02-02 19:48:22 +03:00
|
|
|
if ((ma->flags & ACPI_SRAT_MEM_ENABLED) == 0)
|
2013-01-09 04:18:41 +04:00
|
|
|
goto out_err;
|
2013-02-23 04:33:46 +04:00
|
|
|
hotpluggable = ma->flags & ACPI_SRAT_MEM_HOT_PLUGGABLE;
|
|
|
|
if (hotpluggable && !save_add_info())
|
2013-01-09 04:18:41 +04:00
|
|
|
goto out_err;
|
|
|
|
|
2007-02-02 19:48:22 +03:00
|
|
|
start = ma->base_address;
|
|
|
|
end = start + ma->length;
|
2005-04-17 02:20:36 +04:00
|
|
|
pxm = ma->proximity_domain;
|
2012-01-17 13:20:31 +04:00
|
|
|
if (acpi_srat_revision <= 1)
|
|
|
|
pxm &= 0xff;
|
2013-01-09 04:18:41 +04:00
|
|
|
|
2005-04-17 02:20:36 +04:00
|
|
|
node = setup_node(pxm);
|
|
|
|
if (node < 0) {
|
|
|
|
printk(KERN_ERR "SRAT: Too many proximity domains.\n");
|
2013-01-09 04:18:41 +04:00
|
|
|
goto out_err_bad_srat;
|
2005-04-17 02:20:36 +04:00
|
|
|
}
|
2011-02-16 19:11:07 +03:00
|
|
|
|
2013-01-09 04:18:41 +04:00
|
|
|
if (numa_add_memblk(node, start, end) < 0)
|
|
|
|
goto out_err_bad_srat;
|
2006-04-07 21:49:18 +04:00
|
|
|
|
x86/numa: Set numa_nodes_parsed at acpi_numa_memory_affinity_init()
When hot-adding a CPU, the system outputs following messages
since node_to_cpumask_map[2] was not allocated memory.
Booting Node 2 Processor 32 APIC 0xc0
node_to_cpumask_map[2] NULL
Pid: 0, comm: swapper/32 Tainted: G A 3.3.5-acd #21
Call Trace:
[<ffffffff81048845>] debug_cpumask_set_cpu+0x155/0x160
[<ffffffff8105e28a>] ? add_timer_on+0xaa/0x120
[<ffffffff8150665f>] numa_add_cpu+0x1e/0x22
[<ffffffff815020bb>] identify_cpu+0x1df/0x1e4
[<ffffffff815020d6>] identify_econdary_cpu+0x16/0x1d
[<ffffffff81504614>] smp_store_cpu_info+0x3c/0x3e
[<ffffffff81505263>] smp_callin+0x139/0x1be
[<ffffffff815052fb>] start_secondary+0x13/0xeb
The reason is that the bit of node 2 was not set at
numa_nodes_parsed. numa_nodes_parsed is set by only
acpi_numa_processor_affinity_init /
acpi_numa_x2apic_affinity_init. Thus even if hot-added memory
which is same PXM as hot-added CPU is written in ACPI SRAT
Table, if the hot-added CPU is not written in ACPI SRAT table,
numa_nodes_parsed is not set.
But according to ACPI Spec Rev 5.0, it says about ACPI SRAT
table as follows: This optional table provides information that
allows OSPM to associate processors and memory ranges, including
ranges of memory provided by hot-added memory devices, with
system localities / proximity domains and clock domains.
It means that ACPI SRAT table only provides information for CPUs
present at boot time and for memory including hot-added memory.
So hot-added memory is written in ACPI SRAT table, but hot-added
CPU is not written in it. Thus numa_nodes_parsed should be set
by not only acpi_numa_processor_affinity_init /
acpi_numa_x2apic_affinity_init but also
acpi_numa_memory_affinity_init for the case.
Additionally, if system has cpuless memory node,
acpi_numa_processor_affinity_init /
acpi_numa_x2apic_affinity_init cannot set numa_nodes_parseds
since these functions cannot find cpu description for the node.
In this case, numa_nodes_parsed needs to be set by
acpi_numa_memory_affinity_init.
Signed-off-by: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
Acked-by: David Rientjes <rientjes@google.com>
Acked-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: liuj97@gmail.com
Cc: kosaki.motohiro@gmail.com
Link: http://lkml.kernel.org/r/4FCC2098.4030007@jp.fujitsu.com
[ merged it ]
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2012-06-04 06:42:32 +04:00
|
|
|
node_set(node, numa_nodes_parsed);
|
|
|
|
|
2013-02-23 04:33:46 +04:00
|
|
|
printk(KERN_INFO "SRAT: Node %u PXM %u [mem %#010Lx-%#010Lx] %s\n",
|
2012-05-30 02:06:29 +04:00
|
|
|
node, pxm,
|
2013-02-23 04:33:46 +04:00
|
|
|
(unsigned long long) start, (unsigned long long) end - 1,
|
|
|
|
hotpluggable ? "Hot Pluggable": "");
|
|
|
|
|
acpi, memory-hotplug: support getting hotplug info from SRAT
We now provide an option for users who don't want to specify physical
memory address in kernel commandline.
/*
* For movablemem_map=acpi:
*
* SRAT: |_____| |_____| |_________| |_________| ......
* node id: 0 1 1 2
* hotpluggable: n y y n
* movablemem_map: |_____| |_________|
*
* Using movablemem_map, we can prevent memblock from allocating memory
* on ZONE_MOVABLE at boot time.
*/
So user just specify movablemem_map=acpi, and the kernel will use
hotpluggable info in SRAT to determine which memory ranges should be set
as ZONE_MOVABLE.
If all the memory ranges in SRAT is hotpluggable, then no memory can be
used by kernel. But before parsing SRAT, memblock has already reserve
some memory ranges for other purposes, such as for kernel image, and so
on. We cannot prevent kernel from using these memory. So we need to
exclude these ranges even if these memory is hotpluggable.
Furthermore, there could be several memory ranges in the single node
which the kernel resides in. We may skip one range that have memory
reserved by memblock, but if the rest of memory is too small, then the
kernel will fail to boot. So, make the whole node which the kernel
resides in un-hotpluggable. Then the kernel has enough memory to use.
NOTE: Using this way will cause NUMA performance down because the
whole node will be set as ZONE_MOVABLE, and kernel cannot use memory
on it. If users don't want to lose NUMA performance, just don't use
it.
[akpm@linux-foundation.org: fix warning]
[akpm@linux-foundation.org: use strcmp()]
Signed-off-by: Tang Chen <tangchen@cn.fujitsu.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Jiang Liu <jiang.liu@huawei.com>
Cc: Jianguo Wu <wujianguo@huawei.com>
Cc: Kamezawa Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Lai Jiangshan <laijs@cn.fujitsu.com>
Cc: Wu Jianguo <wujianguo@huawei.com>
Cc: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Len Brown <lenb@kernel.org>
Cc: "Brown, Len" <len.brown@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-23 04:33:49 +04:00
|
|
|
handle_movablemem(node, start, end, hotpluggable);
|
2013-01-09 04:18:41 +04:00
|
|
|
|
2012-07-31 19:41:09 +04:00
|
|
|
return 0;
|
2013-01-09 04:18:41 +04:00
|
|
|
out_err_bad_srat:
|
|
|
|
bad_srat();
|
|
|
|
out_err:
|
|
|
|
return -1;
|
2005-04-17 02:20:36 +04:00
|
|
|
}
|
|
|
|
|
|
|
|
void __init acpi_numa_arch_fixup(void) {}
|
|
|
|
|
2011-02-16 14:13:06 +03:00
|
|
|
int __init x86_acpi_numa_init(void)
|
|
|
|
{
|
|
|
|
int ret;
|
|
|
|
|
|
|
|
ret = acpi_numa_init();
|
|
|
|
if (ret < 0)
|
|
|
|
return ret;
|
|
|
|
return srat_disabled() ? -EINVAL : 0;
|
|
|
|
}
|