OpenCloudOS-Kernel/arch/x86/mm/numa.c

825 lines
20 KiB
C
Raw Normal View History

/* Common code for 32 and 64-bit NUMA */
#include <linux/kernel.h>
#include <linux/mm.h>
#include <linux/string.h>
#include <linux/init.h>
#include <linux/bootmem.h>
#include <linux/memblock.h>
#include <linux/mmzone.h>
#include <linux/ctype.h>
#include <linux/module.h>
#include <linux/nodemask.h>
#include <linux/sched.h>
#include <linux/topology.h>
#include <asm/e820.h>
#include <asm/proto.h>
#include <asm/dma.h>
#include <asm/acpi.h>
#include <asm/amd_nb.h>
#include "numa_internal.h"
int __initdata numa_off;
nodemask_t numa_nodes_parsed __initdata;
struct pglist_data *node_data[MAX_NUMNODES] __read_mostly;
EXPORT_SYMBOL(node_data);
static struct numa_meminfo numa_meminfo
#ifndef CONFIG_MEMORY_HOTPLUG
__initdata
#endif
;
static int numa_distance_cnt;
static u8 *numa_distance;
static __init int numa_setup(char *opt)
{
if (!opt)
return -EINVAL;
if (!strncmp(opt, "off", 3))
numa_off = 1;
#ifdef CONFIG_NUMA_EMU
if (!strncmp(opt, "fake=", 5))
numa_emu_cmdline(opt + 5);
#endif
#ifdef CONFIG_ACPI_NUMA
if (!strncmp(opt, "noacpi", 6))
acpi_numa = -1;
#endif
return 0;
}
early_param("numa", numa_setup);
/*
* apicid, cpu, node mappings
*/
cpu_hotplug: clear apicid to node when the cpu is hotremoved When a cpu is hotpluged, we call acpi_map_cpu2node() in _acpi_map_lsapic() to store the cpu's node and apicid's node. But we don't clear the cpu's node in acpi_unmap_lsapic() when this cpu is hotremoved. If the node is also hotremoved, we will get the following messages: kernel BUG at include/linux/gfp.h:329! invalid opcode: 0000 [#1] SMP Modules linked in: ebtable_nat ebtables ipt_MASQUERADE iptable_nat nf_nat xt_CHECKSUM iptable_mangle bridge stp llc sunrpc ipt_REJECT nf_conntrack_ipv4 nf_defrag_ipv4 iptable_filter ip_tables ip6t_REJECT nf_conntrack_ipv6 nf_defrag_ipv6 xt_state nf_conntrack ip6table_filter ip6_tables binfmt_misc dm_mirror dm_region_hash dm_log dm_mod vhost_net macvtap macvlan tun uinput iTCO_wdt iTCO_vendor_support coretemp kvm_intel kvm crc32c_intel microcode pcspkr i2c_i801 i2c_core lpc_ich mfd_core ioatdma e1000e i7core_edac edac_core sg acpi_memhotplug igb dca sd_mod crc_t10dif megaraid_sas mptsas mptscsih mptbase scsi_transport_sas scsi_mod Pid: 3126, comm: init Not tainted 3.6.0-rc3-tangchen-hostbridge+ #13 FUJITSU-SV PRIMEQUEST 1800E/SB RIP: 0010:[<ffffffff811bc3fd>] [<ffffffff811bc3fd>] allocate_slab+0x28d/0x300 RSP: 0018:ffff88078a049cf8 EFLAGS: 00010246 RAX: 0000000000000000 RBX: 0000000000000001 RCX: 0000000000000000 RDX: 0000000000000001 RSI: 0000000000000001 RDI: 0000000000000246 RBP: ffff88078a049d38 R08: 00000000000040d0 R09: 0000000000000001 R10: 0000000000000000 R11: 0000000000000b5f R12: 00000000000052d0 R13: ffff8807c1417300 R14: 0000000000030038 R15: 0000000000000003 FS: 00007fa9b1b44700(0000) GS:ffff8807c3800000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b CR2: 00007fa9b09acca0 CR3: 000000078b855000 CR4: 00000000000007e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400 Process init (pid: 3126, threadinfo ffff88078a048000, task ffff8807bb6f2650) Call Trace: new_slab+0x30/0x1b0 __slab_alloc+0x358/0x4c0 kmem_cache_alloc_node_trace+0xb4/0x1e0 alloc_fair_sched_group+0xd0/0x1b0 sched_create_group+0x3e/0x110 sched_autogroup_create_attach+0x4d/0x180 sys_setsid+0xd4/0xf0 system_call_fastpath+0x16/0x1b Code: 89 c4 e9 73 fe ff ff 31 c0 89 de 48 c7 c7 45 de 9e 81 44 89 45 c8 e8 22 05 4b 00 85 db 44 8b 45 c8 0f 89 4f ff ff ff 0f 0b eb fe <0f> 0b 90 eb fd 0f 0b eb fe 89 de 48 c7 c7 45 de 9e 81 31 c0 44 RIP [<ffffffff811bc3fd>] allocate_slab+0x28d/0x300 RSP <ffff88078a049cf8> ---[ end trace adf84c90f3fea3e5 ]--- The reason is that the cpu's node is not NUMA_NO_NODE, we will call alloc_pages_exact_node() to alloc memory on the node, but the node is offlined. If the node is onlined, we still need cpu's node. For example: a task on the cpu is sleeped when the cpu is hotremoved. We will choose another cpu to run this task when it is waked up. If we know the cpu's node, we will choose the cpu on the same node first. So we should clear cpu-to-node mapping when the node is offlined. This patch only clears apicid-to-node mapping when the cpu is hotremoved. [akpm@linux-foundation.org: fix section error] Signed-off-by: Wen Congyang <wency@cn.fujitsu.com> Signed-off-by: Tang Chen <tangchen@cn.fujitsu.com> Cc: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com> Cc: David Rientjes <rientjes@google.com> Cc: Jiang Liu <liuj97@gmail.com> Cc: Minchan Kim <minchan.kim@gmail.com> Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Cc: Mel Gorman <mel@csn.ul.ie> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Ingo Molnar <mingo@redhat.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Peter Zijlstra <peterz@infradead.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-23 08:33:24 +08:00
s16 __apicid_to_node[MAX_LOCAL_APIC] = {
[0 ... MAX_LOCAL_APIC-1] = NUMA_NO_NODE
};
int __cpuinit numa_cpu_node(int cpu)
{
int apicid = early_per_cpu(x86_cpu_to_apicid, cpu);
if (apicid != BAD_APICID)
return __apicid_to_node[apicid];
return NUMA_NO_NODE;
}
cpumask_var_t node_to_cpumask_map[MAX_NUMNODES];
EXPORT_SYMBOL(node_to_cpumask_map);
/*
* Map cpu index to node index
*/
DEFINE_EARLY_PER_CPU(int, x86_cpu_to_node_map, NUMA_NO_NODE);
EXPORT_EARLY_PER_CPU_SYMBOL(x86_cpu_to_node_map);
void numa_set_node(int cpu, int node)
{
int *cpu_to_node_map = early_per_cpu_ptr(x86_cpu_to_node_map);
/* early setting, no percpu area yet */
if (cpu_to_node_map) {
cpu_to_node_map[cpu] = node;
return;
}
#ifdef CONFIG_DEBUG_PER_CPU_MAPS
if (cpu >= nr_cpu_ids || !cpu_possible(cpu)) {
printk(KERN_ERR "numa_set_node: invalid cpu# (%d)\n", cpu);
dump_stack();
return;
}
#endif
per_cpu(x86_cpu_to_node_map, cpu) = node;
set_cpu_numa_node(cpu, node);
}
void numa_clear_node(int cpu)
{
numa_set_node(cpu, NUMA_NO_NODE);
}
/*
* Allocate node_to_cpumask_map based on number of available nodes
* Requires node_possible_map to be valid.
*
* Note: cpumask_of_node() is not valid until after this is done.
* (Use CONFIG_DEBUG_PER_CPU_MAPS to check this.)
*/
void __init setup_node_to_cpumask_map(void)
{
unsigned int node, num = 0;
/* setup nr_node_ids if not done yet */
if (nr_node_ids == MAX_NUMNODES) {
for_each_node_mask(node, node_possible_map)
num = node;
nr_node_ids = num + 1;
}
/* allocate the map */
for (node = 0; node < nr_node_ids; node++)
alloc_bootmem_cpumask_var(&node_to_cpumask_map[node]);
/* cpumask_of_node() will now work */
pr_debug("Node to cpumask map for %d nodes\n", nr_node_ids);
}
static int __init numa_add_memblk_to(int nid, u64 start, u64 end,
struct numa_meminfo *mi)
{
/* ignore zero length blks */
if (start == end)
return 0;
/* whine about and ignore invalid blks */
if (start > end || nid < 0 || nid >= MAX_NUMNODES) {
pr_warning("NUMA: Warning: invalid memblk node %d [mem %#010Lx-%#010Lx]\n",
nid, start, end - 1);
return 0;
}
if (mi->nr_blks >= NR_NODE_MEMBLKS) {
pr_err("NUMA: too many memblk ranges\n");
return -EINVAL;
}
mi->blk[mi->nr_blks].start = start;
mi->blk[mi->nr_blks].end = end;
mi->blk[mi->nr_blks].nid = nid;
mi->nr_blks++;
return 0;
}
/**
* numa_remove_memblk_from - Remove one numa_memblk from a numa_meminfo
* @idx: Index of memblk to remove
* @mi: numa_meminfo to remove memblk from
*
* Remove @idx'th numa_memblk from @mi by shifting @mi->blk[] and
* decrementing @mi->nr_blks.
*/
void __init numa_remove_memblk_from(int idx, struct numa_meminfo *mi)
{
mi->nr_blks--;
memmove(&mi->blk[idx], &mi->blk[idx + 1],
(mi->nr_blks - idx) * sizeof(mi->blk[0]));
}
/**
* numa_add_memblk - Add one numa_memblk to numa_meminfo
* @nid: NUMA node ID of the new memblk
* @start: Start address of the new memblk
* @end: End address of the new memblk
*
* Add a new memblk to the default numa_meminfo.
*
* RETURNS:
* 0 on success, -errno on failure.
*/
int __init numa_add_memblk(int nid, u64 start, u64 end)
{
return numa_add_memblk_to(nid, start, end, &numa_meminfo);
}
/* Initialize NODE_DATA for a node on the local memory */
static void __init setup_node_data(int nid, u64 start, u64 end)
{
const size_t nd_size = roundup(sizeof(pg_data_t), PAGE_SIZE);
u64 nd_pa;
void *nd;
int tnid;
/*
* Don't confuse VM with a node that doesn't have the
* minimum amount of memory:
*/
if (end && (end - start) < NODE_MIN_SIZE)
return;
start = roundup(start, ZONE_ALIGN);
printk(KERN_INFO "Initmem setup node %d [mem %#010Lx-%#010Lx]\n",
nid, start, end - 1);
/*
* Allocate node data. Try node-local memory and then any node.
* Never allocate in DMA zone.
*/
x86: get pg_data_t's memory from other node During the implementation of SRAT support, we met a problem. In setup_arch(), we have the following call series: 1) memblock is ready; 2) some functions use memblock to allocate memory; 3) parse ACPI tables, such as SRAT. Before 3), we don't know which memory is hotpluggable, and as a result, we cannot prevent memblock from allocating hotpluggable memory. So, in 2), there could be some hotpluggable memory allocated by memblock. Now, we are trying to parse SRAT earlier, before memblock is ready. But I think we need more investigation on this topic. So in this v5, I dropped all the SRAT support, and v5 is just the same as v3, and it is based on 3.8-rc3. As we planned, we will support getting info from SRAT without users' participation at last. And we will post another patch-set to do so. And also, I think for now, we can add this boot option as the first step of supporting movable node. Since Linux cannot migrate the direct mapped pages, the only way for now is to limit the whole node containing only movable memory. Using SRAT is one way. But even if we can use SRAT, users still need an interface to enable/disable this functionality if they don't want to loose their NUMA performance. So I think, a user interface is always needed. For now, users can disable this functionality by not specifying the boot option. Later, we will post SRAT support, and add another option value "movablecore_map=acpi" to using SRAT. This patch: If system can create movable node which all memory of the node is allocated as ZONE_MOVABLE, setup_node_data() cannot allocate memory for the node's pg_data_t. So, use memblock_alloc_try_nid() instead of memblock_alloc_nid() to retry when the first allocation fails. Signed-off-by: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com> Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com> Signed-off-by: Tang Chen <tangchen@cn.fujitsu.com> Signed-off-by: Jiang Liu <jiang.liu@huawei.com> Cc: Wu Jianguo <wujianguo@huawei.com> Cc: Wen Congyang <wency@cn.fujitsu.com> Cc: Mel Gorman <mel@csn.ul.ie> Cc: Ingo Molnar <mingo@elte.hu> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-23 08:33:35 +08:00
nd_pa = memblock_alloc_try_nid(nd_size, SMP_CACHE_BYTES, nid);
if (!nd_pa) {
x86: get pg_data_t's memory from other node During the implementation of SRAT support, we met a problem. In setup_arch(), we have the following call series: 1) memblock is ready; 2) some functions use memblock to allocate memory; 3) parse ACPI tables, such as SRAT. Before 3), we don't know which memory is hotpluggable, and as a result, we cannot prevent memblock from allocating hotpluggable memory. So, in 2), there could be some hotpluggable memory allocated by memblock. Now, we are trying to parse SRAT earlier, before memblock is ready. But I think we need more investigation on this topic. So in this v5, I dropped all the SRAT support, and v5 is just the same as v3, and it is based on 3.8-rc3. As we planned, we will support getting info from SRAT without users' participation at last. And we will post another patch-set to do so. And also, I think for now, we can add this boot option as the first step of supporting movable node. Since Linux cannot migrate the direct mapped pages, the only way for now is to limit the whole node containing only movable memory. Using SRAT is one way. But even if we can use SRAT, users still need an interface to enable/disable this functionality if they don't want to loose their NUMA performance. So I think, a user interface is always needed. For now, users can disable this functionality by not specifying the boot option. Later, we will post SRAT support, and add another option value "movablecore_map=acpi" to using SRAT. This patch: If system can create movable node which all memory of the node is allocated as ZONE_MOVABLE, setup_node_data() cannot allocate memory for the node's pg_data_t. So, use memblock_alloc_try_nid() instead of memblock_alloc_nid() to retry when the first allocation fails. Signed-off-by: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com> Signed-off-by: Lai Jiangshan <laijs@cn.fujitsu.com> Signed-off-by: Tang Chen <tangchen@cn.fujitsu.com> Signed-off-by: Jiang Liu <jiang.liu@huawei.com> Cc: Wu Jianguo <wujianguo@huawei.com> Cc: Wen Congyang <wency@cn.fujitsu.com> Cc: Mel Gorman <mel@csn.ul.ie> Cc: Ingo Molnar <mingo@elte.hu> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: Thomas Gleixner <tglx@linutronix.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-23 08:33:35 +08:00
pr_err("Cannot find %zu bytes in any node\n", nd_size);
return;
}
nd = __va(nd_pa);
/* report and initialize */
printk(KERN_INFO " NODE_DATA [mem %#010Lx-%#010Lx]\n",
nd_pa, nd_pa + nd_size - 1);
tnid = early_pfn_to_nid(nd_pa >> PAGE_SHIFT);
if (tnid != nid)
printk(KERN_INFO " NODE_DATA(%d) on node %d\n", nid, tnid);
node_data[nid] = nd;
memset(NODE_DATA(nid), 0, sizeof(pg_data_t));
NODE_DATA(nid)->node_id = nid;
NODE_DATA(nid)->node_start_pfn = start >> PAGE_SHIFT;
NODE_DATA(nid)->node_spanned_pages = (end - start) >> PAGE_SHIFT;
node_set_online(nid);
}
/**
* numa_cleanup_meminfo - Cleanup a numa_meminfo
* @mi: numa_meminfo to clean up
*
* Sanitize @mi by merging and removing unncessary memblks. Also check for
* conflicts and clear unused memblks.
*
* RETURNS:
* 0 on success, -errno on failure.
*/
int __init numa_cleanup_meminfo(struct numa_meminfo *mi)
{
const u64 low = 0;
const u64 high = PFN_PHYS(max_pfn);
int i, j, k;
/* first, trim all entries */
for (i = 0; i < mi->nr_blks; i++) {
struct numa_memblk *bi = &mi->blk[i];
/* make sure all blocks are inside the limits */
bi->start = max(bi->start, low);
bi->end = min(bi->end, high);
/* and there's no empty block */
if (bi->start >= bi->end)
numa_remove_memblk_from(i--, mi);
}
/* merge neighboring / overlapping entries */
for (i = 0; i < mi->nr_blks; i++) {
struct numa_memblk *bi = &mi->blk[i];
for (j = i + 1; j < mi->nr_blks; j++) {
struct numa_memblk *bj = &mi->blk[j];
u64 start, end;
/*
* See whether there are overlapping blocks. Whine
* about but allow overlaps of the same nid. They
* will be merged below.
*/
if (bi->end > bj->start && bi->start < bj->end) {
if (bi->nid != bj->nid) {
pr_err("NUMA: node %d [mem %#010Lx-%#010Lx] overlaps with node %d [mem %#010Lx-%#010Lx]\n",
bi->nid, bi->start, bi->end - 1,
bj->nid, bj->start, bj->end - 1);
return -EINVAL;
}
pr_warning("NUMA: Warning: node %d [mem %#010Lx-%#010Lx] overlaps with itself [mem %#010Lx-%#010Lx]\n",
bi->nid, bi->start, bi->end - 1,
bj->start, bj->end - 1);
}
/*
* Join together blocks on the same node, holes
* between which don't overlap with memory on other
* nodes.
*/
if (bi->nid != bj->nid)
continue;
start = min(bi->start, bj->start);
end = max(bi->end, bj->end);
for (k = 0; k < mi->nr_blks; k++) {
struct numa_memblk *bk = &mi->blk[k];
if (bi->nid == bk->nid)
continue;
if (start < bk->end && end > bk->start)
break;
}
if (k < mi->nr_blks)
continue;
printk(KERN_INFO "NUMA: Node %d [mem %#010Lx-%#010Lx] + [mem %#010Lx-%#010Lx] -> [mem %#010Lx-%#010Lx]\n",
bi->nid, bi->start, bi->end - 1, bj->start,
bj->end - 1, start, end - 1);
bi->start = start;
bi->end = end;
numa_remove_memblk_from(j--, mi);
}
}
/* clear unused ones */
for (i = mi->nr_blks; i < ARRAY_SIZE(mi->blk); i++) {
mi->blk[i].start = mi->blk[i].end = 0;
mi->blk[i].nid = NUMA_NO_NODE;
}
return 0;
}
/*
* Set nodes, which have memory in @mi, in *@nodemask.
*/
static void __init numa_nodemask_from_meminfo(nodemask_t *nodemask,
const struct numa_meminfo *mi)
{
int i;
for (i = 0; i < ARRAY_SIZE(mi->blk); i++)
if (mi->blk[i].start != mi->blk[i].end &&
mi->blk[i].nid != NUMA_NO_NODE)
node_set(mi->blk[i].nid, *nodemask);
}
/**
* numa_reset_distance - Reset NUMA distance table
*
* The current table is freed. The next numa_set_distance() call will
* create a new one.
*/
void __init numa_reset_distance(void)
{
size_t size = numa_distance_cnt * numa_distance_cnt * sizeof(numa_distance[0]);
/* numa_distance could be 1LU marking allocation failure, test cnt */
if (numa_distance_cnt)
memblock_free(__pa(numa_distance), size);
numa_distance_cnt = 0;
numa_distance = NULL; /* enable table creation */
}
static int __init numa_alloc_distance(void)
{
nodemask_t nodes_parsed;
size_t size;
int i, j, cnt = 0;
u64 phys;
/* size the new table and allocate it */
nodes_parsed = numa_nodes_parsed;
numa_nodemask_from_meminfo(&nodes_parsed, &numa_meminfo);
for_each_node_mask(i, nodes_parsed)
cnt = i;
cnt++;
size = cnt * cnt * sizeof(numa_distance[0]);
phys = memblock_find_in_range(0, PFN_PHYS(max_pfn_mapped),
size, PAGE_SIZE);
if (!phys) {
pr_warning("NUMA: Warning: can't allocate distance table!\n");
/* don't retry until explicitly reset */
numa_distance = (void *)1LU;
return -ENOMEM;
}
memblock_reserve(phys, size);
numa_distance = __va(phys);
numa_distance_cnt = cnt;
/* fill with the default distances */
for (i = 0; i < cnt; i++)
for (j = 0; j < cnt; j++)
numa_distance[i * cnt + j] = i == j ?
LOCAL_DISTANCE : REMOTE_DISTANCE;
printk(KERN_DEBUG "NUMA: Initialized distance table, cnt=%d\n", cnt);
return 0;
}
/**
* numa_set_distance - Set NUMA distance from one NUMA to another
* @from: the 'from' node to set distance
* @to: the 'to' node to set distance
* @distance: NUMA distance
*
* Set the distance from node @from to @to to @distance. If distance table
* doesn't exist, one which is large enough to accommodate all the currently
* known nodes will be created.
*
* If such table cannot be allocated, a warning is printed and further
* calls are ignored until the distance table is reset with
* numa_reset_distance().
*
* If @from or @to is higher than the highest known node or lower than zero
* at the time of table creation or @distance doesn't make sense, the call
* is ignored.
* This is to allow simplification of specific NUMA config implementations.
*/
void __init numa_set_distance(int from, int to, int distance)
{
if (!numa_distance && numa_alloc_distance() < 0)
return;
if (from >= numa_distance_cnt || to >= numa_distance_cnt ||
from < 0 || to < 0) {
pr_warn_once("NUMA: Warning: node ids are out of bound, from=%d to=%d distance=%d\n",
from, to, distance);
return;
}
if ((u8)distance != distance ||
(from == to && distance != LOCAL_DISTANCE)) {
pr_warn_once("NUMA: Warning: invalid distance parameter, from=%d to=%d distance=%d\n",
from, to, distance);
return;
}
numa_distance[from * numa_distance_cnt + to] = distance;
}
int __node_distance(int from, int to)
{
if (from >= numa_distance_cnt || to >= numa_distance_cnt)
return from == to ? LOCAL_DISTANCE : REMOTE_DISTANCE;
return numa_distance[from * numa_distance_cnt + to];
}
EXPORT_SYMBOL(__node_distance);
/*
* Sanity check to catch more bad NUMA configurations (they are amazingly
* common). Make sure the nodes cover all memory.
*/
static bool __init numa_meminfo_cover_memory(const struct numa_meminfo *mi)
{
u64 numaram, e820ram;
int i;
numaram = 0;
for (i = 0; i < mi->nr_blks; i++) {
u64 s = mi->blk[i].start >> PAGE_SHIFT;
u64 e = mi->blk[i].end >> PAGE_SHIFT;
numaram += e - s;
numaram -= __absent_pages_in_range(mi->blk[i].nid, s, e);
if ((s64)numaram < 0)
numaram = 0;
}
e820ram = max_pfn - absent_pages_in_range(0, max_pfn);
/* We seem to lose 3 pages somewhere. Allow 1M of slack. */
if ((s64)(e820ram - numaram) >= (1 << (20 - PAGE_SHIFT))) {
printk(KERN_ERR "NUMA: nodes only cover %LuMB of your %LuMB e820 RAM. Not used.\n",
(numaram << PAGE_SHIFT) >> 20,
(e820ram << PAGE_SHIFT) >> 20);
return false;
}
return true;
}
static int __init numa_register_memblks(struct numa_meminfo *mi)
{
x86, numa: Implement pfn -> nid mapping granularity check SPARSEMEM w/o VMEMMAP and DISCONTIGMEM, both used only on 32bit, use sections array to map pfn to nid which is limited in granularity. If NUMA nodes are laid out such that the mapping cannot be accurate, boot will fail triggering BUG_ON() in mminit_verify_page_links(). On 32bit, it's 512MiB w/ PAE and SPARSEMEM. This seems to have been granular enough until commit 2706a0bf7b (x86, NUMA: Enable CONFIG_AMD_NUMA on 32bit too). Apparently, there is a machine which aligns NUMA nodes to 128MiB and has only AMD NUMA but not SRAT. This led to the following BUG_ON(). On node 0 totalpages: 2096615 DMA zone: 32 pages used for memmap DMA zone: 0 pages reserved DMA zone: 3927 pages, LIFO batch:0 Normal zone: 1740 pages used for memmap Normal zone: 220978 pages, LIFO batch:31 HighMem zone: 16405 pages used for memmap HighMem zone: 1853533 pages, LIFO batch:31 BUG: Int 6: CR2 (null) EDI (null) ESI 00000002 EBP 00000002 ESP c1543ecc EBX f2400000 EDX 00000006 ECX (null) EAX 00000001 err (null) EIP c16209aa CS 00000060 flg 00010002 Stack: f2400000 00220000 f7200800 c1620613 00220000 01000000 04400000 00238000 (null) f7200000 00000002 f7200b58 f7200800 c1620929 000375fe (null) f7200b80 c16395f0 00200a02 f7200a80 (null) 000375fe 00000002 (null) Pid: 0, comm: swapper Not tainted 2.6.39-rc5-00181-g2706a0b #17 Call Trace: [<c136b1e5>] ? early_fault+0x2e/0x2e [<c16209aa>] ? mminit_verify_page_links+0x12/0x42 [<c1620613>] ? memmap_init_zone+0xaf/0x10c [<c1620929>] ? free_area_init_node+0x2b9/0x2e3 [<c1607e99>] ? free_area_init_nodes+0x3f2/0x451 [<c1601d80>] ? paging_init+0x112/0x118 [<c15f578d>] ? setup_arch+0x791/0x82f [<c15f43d9>] ? start_kernel+0x6a/0x257 This patch implements node_map_pfn_alignment() which determines maximum internode alignment and update numa_register_memblks() to reject NUMA configuration if alignment exceeds the pfn -> nid mapping granularity of the memory model as determined by PAGES_PER_SECTION. This makes the problematic machine boot w/ flatmem by rejecting the NUMA config and provides protection against crazy NUMA configurations. Signed-off-by: Tejun Heo <tj@kernel.org> Link: http://lkml.kernel.org/r/20110712074534.GB2872@htj.dyndns.org LKML-Reference: <20110628174613.GP478@escobedo.osrc.amd.com> Reported-and-Tested-by: Hans Rosenfeld <hans.rosenfeld@amd.com> Cc: Conny Seidel <conny.seidel@amd.com> Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2011-07-12 15:45:34 +08:00
unsigned long uninitialized_var(pfn_align);
int i, nid;
/* Account for nodes with cpus and no memory */
node_possible_map = numa_nodes_parsed;
numa_nodemask_from_meminfo(&node_possible_map, mi);
if (WARN_ON(nodes_empty(node_possible_map)))
return -EINVAL;
for (i = 0; i < mi->nr_blks; i++) {
struct numa_memblk *mb = &mi->blk[i];
memblock_set_node(mb->start, mb->end - mb->start, mb->nid);
}
x86, numa: Implement pfn -> nid mapping granularity check SPARSEMEM w/o VMEMMAP and DISCONTIGMEM, both used only on 32bit, use sections array to map pfn to nid which is limited in granularity. If NUMA nodes are laid out such that the mapping cannot be accurate, boot will fail triggering BUG_ON() in mminit_verify_page_links(). On 32bit, it's 512MiB w/ PAE and SPARSEMEM. This seems to have been granular enough until commit 2706a0bf7b (x86, NUMA: Enable CONFIG_AMD_NUMA on 32bit too). Apparently, there is a machine which aligns NUMA nodes to 128MiB and has only AMD NUMA but not SRAT. This led to the following BUG_ON(). On node 0 totalpages: 2096615 DMA zone: 32 pages used for memmap DMA zone: 0 pages reserved DMA zone: 3927 pages, LIFO batch:0 Normal zone: 1740 pages used for memmap Normal zone: 220978 pages, LIFO batch:31 HighMem zone: 16405 pages used for memmap HighMem zone: 1853533 pages, LIFO batch:31 BUG: Int 6: CR2 (null) EDI (null) ESI 00000002 EBP 00000002 ESP c1543ecc EBX f2400000 EDX 00000006 ECX (null) EAX 00000001 err (null) EIP c16209aa CS 00000060 flg 00010002 Stack: f2400000 00220000 f7200800 c1620613 00220000 01000000 04400000 00238000 (null) f7200000 00000002 f7200b58 f7200800 c1620929 000375fe (null) f7200b80 c16395f0 00200a02 f7200a80 (null) 000375fe 00000002 (null) Pid: 0, comm: swapper Not tainted 2.6.39-rc5-00181-g2706a0b #17 Call Trace: [<c136b1e5>] ? early_fault+0x2e/0x2e [<c16209aa>] ? mminit_verify_page_links+0x12/0x42 [<c1620613>] ? memmap_init_zone+0xaf/0x10c [<c1620929>] ? free_area_init_node+0x2b9/0x2e3 [<c1607e99>] ? free_area_init_nodes+0x3f2/0x451 [<c1601d80>] ? paging_init+0x112/0x118 [<c15f578d>] ? setup_arch+0x791/0x82f [<c15f43d9>] ? start_kernel+0x6a/0x257 This patch implements node_map_pfn_alignment() which determines maximum internode alignment and update numa_register_memblks() to reject NUMA configuration if alignment exceeds the pfn -> nid mapping granularity of the memory model as determined by PAGES_PER_SECTION. This makes the problematic machine boot w/ flatmem by rejecting the NUMA config and provides protection against crazy NUMA configurations. Signed-off-by: Tejun Heo <tj@kernel.org> Link: http://lkml.kernel.org/r/20110712074534.GB2872@htj.dyndns.org LKML-Reference: <20110628174613.GP478@escobedo.osrc.amd.com> Reported-and-Tested-by: Hans Rosenfeld <hans.rosenfeld@amd.com> Cc: Conny Seidel <conny.seidel@amd.com> Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
2011-07-12 15:45:34 +08:00
/*
* If sections array is gonna be used for pfn -> nid mapping, check
* whether its granularity is fine enough.
*/
#ifdef NODE_NOT_IN_PAGE_FLAGS
pfn_align = node_map_pfn_alignment();
if (pfn_align && pfn_align < PAGES_PER_SECTION) {
printk(KERN_WARNING "Node alignment %LuMB < min %LuMB, rejecting NUMA config\n",
PFN_PHYS(pfn_align) >> 20,
PFN_PHYS(PAGES_PER_SECTION) >> 20);
return -EINVAL;
}
#endif
if (!numa_meminfo_cover_memory(mi))
return -EINVAL;
/* Finally register nodes. */
for_each_node_mask(nid, node_possible_map) {
u64 start = PFN_PHYS(max_pfn);
u64 end = 0;
for (i = 0; i < mi->nr_blks; i++) {
if (nid != mi->blk[i].nid)
continue;
start = min(mi->blk[i].start, start);
end = max(mi->blk[i].end, end);
}
if (start < end)
setup_node_data(nid, start, end);
}
/* Dump memblock with node info and return. */
memblock_dump_all();
return 0;
}
/*
* There are unfortunately some poorly designed mainboards around that
* only connect memory to a single CPU. This breaks the 1:1 cpu->node
* mapping. To avoid this fill in the mapping for all possible CPUs,
* as the number of CPUs is not known yet. We round robin the existing
* nodes.
*/
static void __init numa_init_array(void)
{
int rr, i;
rr = first_node(node_online_map);
for (i = 0; i < nr_cpu_ids; i++) {
if (early_cpu_to_node(i) != NUMA_NO_NODE)
continue;
numa_set_node(i, rr);
rr = next_node(rr, node_online_map);
if (rr == MAX_NUMNODES)
rr = first_node(node_online_map);
}
}
static int __init numa_init(int (*init_func)(void))
{
int i;
int ret;
for (i = 0; i < MAX_LOCAL_APIC; i++)
set_apicid_to_node(i, NUMA_NO_NODE);
acpi, memory-hotplug: parse SRAT before memblock is ready On linux, the pages used by kernel could not be migrated. As a result, if a memory range is used by kernel, it cannot be hot-removed. So if we want to hot-remove memory, we should prevent kernel from using it. The way now used to prevent this is specify a memory range by movablemem_map boot option and set it as ZONE_MOVABLE. But when the system is booting, memblock will allocate memory, and reserve the memory for kernel. And before we parse SRAT, and know the node memory ranges, memblock is working. And it may allocate memory in ranges to be set as ZONE_MOVABLE. This memory can be used by kernel, and never be freed. So, let's parse SRAT before memblock is called first. And it is early enough. The first call of memblock_find_in_range_node() is in: setup_arch() |-->setup_real_mode() so, this patch add a function early_parse_srat() to parse SRAT, and call it before setup_real_mode() is called. NOTE: 1) early_parse_srat() is called before numa_init(), and has initialized numa_meminfo. So DO NOT clear numa_nodes_parsed in numa_init() and DO NOT zero numa_meminfo in numa_init(), otherwise we will lose memory numa info. 2) I don't know why using count of memory affinities parsed from SRAT as a return value in original acpi_numa_init(). So I add a static variable srat_mem_cnt to remember this count and use it as the return value of the new acpi_numa_init() [mhocko@suse.cz: parse SRAT before memblock is ready fix] Signed-off-by: Tang Chen <tangchen@cn.fujitsu.com> Reviewed-by: Wen Congyang <wency@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 08:33:44 +08:00
/*
* Do not clear numa_nodes_parsed or zero numa_meminfo here, because
* SRAT was parsed earlier in early_parse_srat().
*/
nodes_clear(node_possible_map);
nodes_clear(node_online_map);
WARN_ON(memblock_set_node(0, ULLONG_MAX, MAX_NUMNODES));
numa_reset_distance();
ret = init_func();
if (ret < 0)
return ret;
ret = numa_cleanup_meminfo(&numa_meminfo);
if (ret < 0)
return ret;
numa_emulation(&numa_meminfo, numa_distance_cnt);
ret = numa_register_memblks(&numa_meminfo);
if (ret < 0)
return ret;
for (i = 0; i < nr_cpu_ids; i++) {
int nid = early_cpu_to_node(i);
if (nid == NUMA_NO_NODE)
continue;
if (!node_online(nid))
numa_clear_node(i);
}
numa_init_array();
return 0;
}
/**
* dummy_numa_init - Fallback dummy NUMA init
*
* Used if there's no underlying NUMA architecture, NUMA initialization
* fails, or NUMA is disabled on the command line.
*
* Must online at least one node and add memory blocks that cover all
* allowed memory. This function must not fail.
*/
static int __init dummy_numa_init(void)
{
printk(KERN_INFO "%s\n",
numa_off ? "NUMA turned off" : "No NUMA configuration found");
printk(KERN_INFO "Faking a node at [mem %#018Lx-%#018Lx]\n",
0LLU, PFN_PHYS(max_pfn) - 1);
node_set(0, numa_nodes_parsed);
numa_add_memblk(0, 0, PFN_PHYS(max_pfn));
return 0;
}
/**
* x86_numa_init - Initialize NUMA
*
* Try each configured NUMA initialization method until one succeeds. The
* last fallback is dummy single node config encomapssing whole memory and
* never fails.
*/
void __init x86_numa_init(void)
{
if (!numa_off) {
#ifdef CONFIG_X86_NUMAQ
if (!numa_init(numaq_numa_init))
return;
#endif
#ifdef CONFIG_ACPI_NUMA
if (!numa_init(x86_acpi_numa_init))
return;
#endif
#ifdef CONFIG_AMD_NUMA
if (!numa_init(amd_numa_init))
return;
#endif
}
numa_init(dummy_numa_init);
}
static __init int find_near_online_node(int node)
{
int n, val;
int min_val = INT_MAX;
int best_node = -1;
for_each_online_node(n) {
val = node_distance(node, n);
if (val < min_val) {
min_val = val;
best_node = n;
}
}
return best_node;
}
/*
* Setup early cpu_to_node.
*
* Populate cpu_to_node[] only if x86_cpu_to_apicid[],
* and apicid_to_node[] tables have valid entries for a CPU.
* This means we skip cpu_to_node[] initialisation for NUMA
* emulation and faking node case (when running a kernel compiled
* for NUMA on a non NUMA box), which is OK as cpu_to_node[]
* is already initialized in a round robin manner at numa_init_array,
* prior to this call, and this initialization is good enough
* for the fake NUMA cases.
*
* Called before the per_cpu areas are setup.
*/
void __init init_cpu_to_node(void)
{
int cpu;
u16 *cpu_to_apicid = early_per_cpu_ptr(x86_cpu_to_apicid);
BUG_ON(cpu_to_apicid == NULL);
for_each_possible_cpu(cpu) {
int node = numa_cpu_node(cpu);
if (node == NUMA_NO_NODE)
continue;
if (!node_online(node))
node = find_near_online_node(node);
numa_set_node(cpu, node);
}
}
#ifndef CONFIG_DEBUG_PER_CPU_MAPS
# ifndef CONFIG_NUMA_EMU
void __cpuinit numa_add_cpu(int cpu)
{
cpumask_set_cpu(cpu, node_to_cpumask_map[early_cpu_to_node(cpu)]);
}
void __cpuinit numa_remove_cpu(int cpu)
{
cpumask_clear_cpu(cpu, node_to_cpumask_map[early_cpu_to_node(cpu)]);
}
# endif /* !CONFIG_NUMA_EMU */
#else /* !CONFIG_DEBUG_PER_CPU_MAPS */
int __cpu_to_node(int cpu)
{
if (early_per_cpu_ptr(x86_cpu_to_node_map)) {
printk(KERN_WARNING
"cpu_to_node(%d): usage too early!\n", cpu);
dump_stack();
return early_per_cpu_ptr(x86_cpu_to_node_map)[cpu];
}
return per_cpu(x86_cpu_to_node_map, cpu);
}
EXPORT_SYMBOL(__cpu_to_node);
/*
* Same function as cpu_to_node() but used if called before the
* per_cpu areas are setup.
*/
int early_cpu_to_node(int cpu)
{
if (early_per_cpu_ptr(x86_cpu_to_node_map))
return early_per_cpu_ptr(x86_cpu_to_node_map)[cpu];
if (!cpu_possible(cpu)) {
printk(KERN_WARNING
"early_cpu_to_node(%d): no per_cpu area!\n", cpu);
dump_stack();
return NUMA_NO_NODE;
}
return per_cpu(x86_cpu_to_node_map, cpu);
}
void debug_cpumask_set_cpu(int cpu, int node, bool enable)
{
struct cpumask *mask;
char buf[64];
if (node == NUMA_NO_NODE) {
/* early_cpu_to_node() already emits a warning and trace */
return;
}
mask = node_to_cpumask_map[node];
if (!mask) {
pr_err("node_to_cpumask_map[%i] NULL\n", node);
dump_stack();
return;
}
if (enable)
cpumask_set_cpu(cpu, mask);
else
cpumask_clear_cpu(cpu, mask);
cpulist_scnprintf(buf, sizeof(buf), mask);
printk(KERN_DEBUG "%s cpu %d node %d: mask now %s\n",
enable ? "numa_add_cpu" : "numa_remove_cpu",
cpu, node, buf);
return;
}
# ifndef CONFIG_NUMA_EMU
static void __cpuinit numa_set_cpumask(int cpu, bool enable)
{
debug_cpumask_set_cpu(cpu, early_cpu_to_node(cpu), enable);
}
void __cpuinit numa_add_cpu(int cpu)
{
numa_set_cpumask(cpu, true);
}
void __cpuinit numa_remove_cpu(int cpu)
{
numa_set_cpumask(cpu, false);
}
# endif /* !CONFIG_NUMA_EMU */
/*
* Returns a pointer to the bitmask of CPUs on Node 'node'.
*/
const struct cpumask *cpumask_of_node(int node)
{
if (node >= nr_node_ids) {
printk(KERN_WARNING
"cpumask_of_node(%d): node > nr_node_ids(%d)\n",
node, nr_node_ids);
dump_stack();
return cpu_none_mask;
}
if (node_to_cpumask_map[node] == NULL) {
printk(KERN_WARNING
"cpumask_of_node(%d): no node_to_cpumask_map!\n",
node);
dump_stack();
return cpu_online_mask;
}
return node_to_cpumask_map[node];
}
EXPORT_SYMBOL(cpumask_of_node);
#endif /* !CONFIG_DEBUG_PER_CPU_MAPS */
#ifdef CONFIG_MEMORY_HOTPLUG
int memory_add_physaddr_to_nid(u64 start)
{
struct numa_meminfo *mi = &numa_meminfo;
int nid = mi->blk[0].nid;
int i;
for (i = 0; i < mi->nr_blks; i++)
if (mi->blk[i].start <= start && mi->blk[i].end > start)
nid = mi->blk[i].nid;
return nid;
}
EXPORT_SYMBOL_GPL(memory_add_physaddr_to_nid);
#endif