summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorDan Williams <dan.j.williams@intel.com>2020-04-02 19:50:31 -0700
committerDan Williams <dan.j.williams@intel.com>2020-04-02 19:50:31 -0700
commitd3b88655c0a157c11370b8faf50e82ecb1c17d54 (patch)
treee3c868d5243a6073625287979bd52cf76f4eaaa8
parent91bf79bcb61df7a89690f894f06b78b0e66fb43c (diff)
parent7b27a8622f802761d5c6abd6c37b22312a35343c (diff)
Merge branch 'for-5.7/numa' into libnvdimm-for-next
- Promote numa_map_to_online_node() to a cross-kernel generic facility. - Save x86 numa information to allow for node-id lookups for reserved memory ranges, deploy that capability for the e820-pmem driver. - Introduce phys_to_target_node() to facilitate drivers that want to know resulting numa node if a given reserved address range was onlined.
-rw-r--r--arch/powerpc/platforms/pseries/papr_scm.c21
-rw-r--r--arch/x86/Kconfig1
-rw-r--r--arch/x86/mm/numa.c67
-rw-r--r--drivers/acpi/numa/srat.c41
-rw-r--r--drivers/nvdimm/e820.c18
-rw-r--r--include/linux/acpi.h23
-rw-r--r--include/linux/numa.h30
-rw-r--r--mm/Kconfig5
-rw-r--r--mm/mempolicy.c26
9 files changed, 140 insertions, 92 deletions
diff --git a/arch/powerpc/platforms/pseries/papr_scm.c b/arch/powerpc/platforms/pseries/papr_scm.c
index 589858cb3203..5499fc06ed5a 100644
--- a/arch/powerpc/platforms/pseries/papr_scm.c
+++ b/arch/powerpc/platforms/pseries/papr_scm.c
@@ -285,25 +285,6 @@ int papr_scm_ndctl(struct nvdimm_bus_descriptor *nd_desc, struct nvdimm *nvdimm,
return 0;
}
-static inline int papr_scm_node(int node)
-{
- int min_dist = INT_MAX, dist;
- int nid, min_node;
-
- if ((node == NUMA_NO_NODE) || node_online(node))
- return node;
-
- min_node = first_online_node;
- for_each_online_node(nid) {
- dist = node_distance(node, nid);
- if (dist < min_dist) {
- min_dist = dist;
- min_node = nid;
- }
- }
- return min_node;
-}
-
static int papr_scm_nvdimm_init(struct papr_scm_priv *p)
{
struct device *dev = &p->pdev->dev;
@@ -349,7 +330,7 @@ static int papr_scm_nvdimm_init(struct papr_scm_priv *p)
memset(&ndr_desc, 0, sizeof(ndr_desc));
target_nid = dev_to_node(&p->pdev->dev);
- online_nid = papr_scm_node(target_nid);
+ online_nid = numa_map_to_online_node(target_nid);
ndr_desc.numa_node = online_nid;
ndr_desc.target_node = target_nid;
ndr_desc.res = &p->res;
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index beea77046f9b..d4e446daf457 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -1664,6 +1664,7 @@ config X86_PMEM_LEGACY
depends on PHYS_ADDR_T_64BIT
depends on BLK_DEV
select X86_PMEM_LEGACY_DEVICE
+ select NUMA_KEEP_MEMINFO if NUMA
select LIBNVDIMM
help
Treat memory marked using the non-standard e820 type of 12 as used
diff --git a/arch/x86/mm/numa.c b/arch/x86/mm/numa.c
index 99f7a68738f0..59ba008504dc 100644
--- a/arch/x86/mm/numa.c
+++ b/arch/x86/mm/numa.c
@@ -25,11 +25,8 @@ 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 struct numa_meminfo numa_meminfo __initdata_or_meminfo;
+static struct numa_meminfo numa_reserved_meminfo __initdata_or_meminfo;
static int numa_distance_cnt;
static u8 *numa_distance;
@@ -169,6 +166,19 @@ void __init numa_remove_memblk_from(int idx, struct numa_meminfo *mi)
}
/**
+ * numa_move_tail_memblk - Move a numa_memblk from one numa_meminfo to another
+ * @dst: numa_meminfo to append block to
+ * @idx: Index of memblk to remove
+ * @src: numa_meminfo to remove memblk from
+ */
+static void __init numa_move_tail_memblk(struct numa_meminfo *dst, int idx,
+ struct numa_meminfo *src)
+{
+ dst->blk[dst->nr_blks++] = src->blk[idx];
+ numa_remove_memblk_from(idx, src);
+}
+
+/**
* 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
@@ -237,14 +247,19 @@ int __init numa_cleanup_meminfo(struct numa_meminfo *mi)
for (i = 0; i < mi->nr_blks; i++) {
struct numa_memblk *bi = &mi->blk[i];
- /* make sure all blocks are inside the limits */
+ /* move / save reserved memory ranges */
+ if (!memblock_overlaps_region(&memblock.memory,
+ bi->start, bi->end - bi->start)) {
+ numa_move_tail_memblk(&numa_reserved_meminfo, i--, mi);
+ continue;
+ }
+
+ /* make sure all non-reserved blocks are inside the limits */
bi->start = max(bi->start, low);
bi->end = min(bi->end, high);
- /* and there's no empty or non-exist block */
- if (bi->start >= bi->end ||
- !memblock_overlaps_region(&memblock.memory,
- bi->start, bi->end - bi->start))
+ /* and there's no empty block */
+ if (bi->start >= bi->end)
numa_remove_memblk_from(i--, mi);
}
@@ -881,16 +896,38 @@ EXPORT_SYMBOL(cpumask_of_node);
#endif /* !CONFIG_DEBUG_PER_CPU_MAPS */
-#ifdef CONFIG_MEMORY_HOTPLUG
-int memory_add_physaddr_to_nid(u64 start)
+#ifdef CONFIG_NUMA_KEEP_MEMINFO
+static int meminfo_to_nid(struct numa_meminfo *mi, 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 mi->blk[i].nid;
+ return NUMA_NO_NODE;
+}
+
+int phys_to_target_node(phys_addr_t start)
+{
+ int nid = meminfo_to_nid(&numa_meminfo, start);
+
+ /*
+ * Prefer online nodes, but if reserved memory might be
+ * hot-added continue the search with reserved ranges.
+ */
+ if (nid != NUMA_NO_NODE)
+ return nid;
+
+ return meminfo_to_nid(&numa_reserved_meminfo, start);
+}
+EXPORT_SYMBOL_GPL(phys_to_target_node);
+
+int memory_add_physaddr_to_nid(u64 start)
+{
+ int nid = meminfo_to_nid(&numa_meminfo, start);
+
+ if (nid == NUMA_NO_NODE)
+ nid = numa_meminfo.blk[0].nid;
return nid;
}
EXPORT_SYMBOL_GPL(memory_add_physaddr_to_nid);
diff --git a/drivers/acpi/numa/srat.c b/drivers/acpi/numa/srat.c
index eadbf90e65d1..47b4969d9b93 100644
--- a/drivers/acpi/numa/srat.c
+++ b/drivers/acpi/numa/srat.c
@@ -72,47 +72,6 @@ int acpi_map_pxm_to_node(int pxm)
}
EXPORT_SYMBOL(acpi_map_pxm_to_node);
-/**
- * acpi_map_pxm_to_online_node - Map proximity ID to online node
- * @pxm: ACPI proximity ID
- *
- * This is similar to acpi_map_pxm_to_node(), but always returns an online
- * node. When the mapped node from a given proximity ID is offline, it
- * looks up the node distance table and returns the nearest online node.
- *
- * ACPI device drivers, which are called after the NUMA initialization has
- * completed in the kernel, can call this interface to obtain their device
- * NUMA topology from ACPI tables. Such drivers do not have to deal with
- * offline nodes. A node may be offline when a device proximity ID is
- * unique, SRAT memory entry does not exist, or NUMA is disabled, ex.
- * "numa=off" on x86.
- */
-int acpi_map_pxm_to_online_node(int pxm)
-{
- int node, min_node;
-
- node = acpi_map_pxm_to_node(pxm);
-
- if (node == NUMA_NO_NODE)
- node = 0;
-
- min_node = node;
- if (!node_online(node)) {
- int min_dist = INT_MAX, dist, n;
-
- for_each_online_node(n) {
- dist = node_distance(node, n);
- if (dist < min_dist) {
- min_dist = dist;
- min_node = n;
- }
- }
- }
-
- return min_node;
-}
-EXPORT_SYMBOL(acpi_map_pxm_to_online_node);
-
static void __init
acpi_table_print_srat_entry(struct acpi_subtable_header *header)
{
diff --git a/drivers/nvdimm/e820.c b/drivers/nvdimm/e820.c
index e02f60ad6c99..4cd18be9d0e9 100644
--- a/drivers/nvdimm/e820.c
+++ b/drivers/nvdimm/e820.c
@@ -7,6 +7,7 @@
#include <linux/memory_hotplug.h>
#include <linux/libnvdimm.h>
#include <linux/module.h>
+#include <linux/numa.h>
static int e820_pmem_remove(struct platform_device *pdev)
{
@@ -16,27 +17,16 @@ static int e820_pmem_remove(struct platform_device *pdev)
return 0;
}
-#ifdef CONFIG_MEMORY_HOTPLUG
-static int e820_range_to_nid(resource_size_t addr)
-{
- return memory_add_physaddr_to_nid(addr);
-}
-#else
-static int e820_range_to_nid(resource_size_t addr)
-{
- return NUMA_NO_NODE;
-}
-#endif
-
static int e820_register_one(struct resource *res, void *data)
{
struct nd_region_desc ndr_desc;
struct nvdimm_bus *nvdimm_bus = data;
+ int nid = phys_to_target_node(res->start);
memset(&ndr_desc, 0, sizeof(ndr_desc));
ndr_desc.res = res;
- ndr_desc.numa_node = e820_range_to_nid(res->start);
- ndr_desc.target_node = ndr_desc.numa_node;
+ ndr_desc.numa_node = numa_map_to_online_node(nid);
+ ndr_desc.target_node = nid;
set_bit(ND_REGION_PAGEMAP, &ndr_desc.flags);
if (!nvdimm_pmem_region_create(nvdimm_bus, &ndr_desc))
return -ENXIO;
diff --git a/include/linux/acpi.h b/include/linux/acpi.h
index 0f24d701fbdc..3839363081f3 100644
--- a/include/linux/acpi.h
+++ b/include/linux/acpi.h
@@ -416,9 +416,30 @@ extern void acpi_osi_setup(char *str);
extern bool acpi_osi_is_win8(void);
#ifdef CONFIG_ACPI_NUMA
-int acpi_map_pxm_to_online_node(int pxm);
int acpi_map_pxm_to_node(int pxm);
int acpi_get_node(acpi_handle handle);
+
+/**
+ * acpi_map_pxm_to_online_node - Map proximity ID to online node
+ * @pxm: ACPI proximity ID
+ *
+ * This is similar to acpi_map_pxm_to_node(), but always returns an online
+ * node. When the mapped node from a given proximity ID is offline, it
+ * looks up the node distance table and returns the nearest online node.
+ *
+ * ACPI device drivers, which are called after the NUMA initialization has
+ * completed in the kernel, can call this interface to obtain their device
+ * NUMA topology from ACPI tables. Such drivers do not have to deal with
+ * offline nodes. A node may be offline when a device proximity ID is
+ * unique, SRAT memory entry does not exist, or NUMA is disabled, ex.
+ * "numa=off" on x86.
+ */
+static inline int acpi_map_pxm_to_online_node(int pxm)
+{
+ int node = acpi_map_pxm_to_node(pxm);
+
+ return numa_map_to_online_node(node);
+}
#else
static inline int acpi_map_pxm_to_online_node(int pxm)
{
diff --git a/include/linux/numa.h b/include/linux/numa.h
index 110b0e5d0fb0..a42df804679e 100644
--- a/include/linux/numa.h
+++ b/include/linux/numa.h
@@ -1,7 +1,7 @@
/* SPDX-License-Identifier: GPL-2.0 */
#ifndef _LINUX_NUMA_H
#define _LINUX_NUMA_H
-
+#include <linux/types.h>
#ifdef CONFIG_NODES_SHIFT
#define NODES_SHIFT CONFIG_NODES_SHIFT
@@ -13,4 +13,32 @@
#define NUMA_NO_NODE (-1)
+/* optionally keep NUMA memory info available post init */
+#ifdef CONFIG_NUMA_KEEP_MEMINFO
+#define __initdata_or_meminfo
+#else
+#define __initdata_or_meminfo __initdata
+#endif
+
+#ifdef CONFIG_NUMA
+/* Generic implementation available */
+int numa_map_to_online_node(int node);
+
+/*
+ * Optional architecture specific implementation, users need a "depends
+ * on $ARCH"
+ */
+int phys_to_target_node(phys_addr_t addr);
+#else
+static inline int numa_map_to_online_node(int node)
+{
+ return NUMA_NO_NODE;
+}
+
+static inline int phys_to_target_node(phys_addr_t addr)
+{
+ return NUMA_NO_NODE;
+}
+#endif
+
#endif /* _LINUX_NUMA_H */
diff --git a/mm/Kconfig b/mm/Kconfig
index ab80933be65f..328268473fec 100644
--- a/mm/Kconfig
+++ b/mm/Kconfig
@@ -139,6 +139,10 @@ config HAVE_FAST_GUP
config ARCH_KEEP_MEMBLOCK
bool
+# Keep arch NUMA mapping infrastructure post-init.
+config NUMA_KEEP_MEMINFO
+ bool
+
config MEMORY_ISOLATION
bool
@@ -154,6 +158,7 @@ config MEMORY_HOTPLUG
bool "Allow for memory hot-add"
depends on SPARSEMEM || X86_64_ACPI_NUMA
depends on ARCH_ENABLE_MEMORY_HOTPLUG
+ select NUMA_KEEP_MEMINFO if NUMA
config MEMORY_HOTPLUG_SPARSE
def_bool y
diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index 977c641f78cf..19f7e71945a7 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -127,6 +127,32 @@ static struct mempolicy default_policy = {
static struct mempolicy preferred_node_policy[MAX_NUMNODES];
+/**
+ * numa_map_to_online_node - Find closest online node
+ * @nid: Node id to start the search
+ *
+ * Lookup the next closest node by distance if @nid is not online.
+ */
+int numa_map_to_online_node(int node)
+{
+ int min_dist = INT_MAX, dist, n, min_node;
+
+ if (node == NUMA_NO_NODE || node_online(node))
+ return node;
+
+ min_node = node;
+ for_each_online_node(n) {
+ dist = node_distance(node, n);
+ if (dist < min_dist) {
+ min_dist = dist;
+ min_node = n;
+ }
+ }
+
+ return min_node;
+}
+EXPORT_SYMBOL_GPL(numa_map_to_online_node);
+
struct mempolicy *get_task_policy(struct task_struct *p)
{
struct mempolicy *pol = p->mempolicy;