[PATCH v3 0/2] Improved Memory Tier Creation for CPUless NUMA Nodes

[PATCH v3 0/2] Improved Memory Tier Creation for CPUless NUMA Nodes

[Ho-Ren (Jack) Chuang]

When a memory device, such as CXL1.1 type3 memory, is emulated as
normal memory (E820_TYPE_RAM), the memory device is indistinguishable
from normal DRAM in terms of memory tiering with the current implementation.
The current memory tiering assigns all detected normal memory nodes
to the same DRAM tier. This results in normal memory devices with
different attributions being unable to be assigned to the correct memory tier,
leading to the inability to migrate pages between different types of memory.
https://lore.kernel.org/linux-mm/PH0PR08MB7955E9F08CCB64F23963B5C3A860A@PH0PR08MB7955.namprd08.prod.outlook.com/T/

This patchset automatically resolves the issues. It delays the initialization
of memory tiers for CPUless NUMA nodes until they obtain HMAT information
and after all devices are initialized at boot time, eliminating the need
for user intervention. If no HMAT is specified, it falls back to
using `default_dram_type`.

Example usecase:
We have CXL memory on the host, and we create VMs with a new system memory
device backed by host CXL memory. We inject CXL memory performance attributes
through QEMU, and the guest now sees memory nodes with performance attributes
in HMAT. With this change, we enable the guest kernel to construct
the correct memory tiering for the memory nodes.

-v3:
 Thanks to Ying's comments,
 * Make the newly added code independent of HMAT
 * Upgrade set_node_memory_tier to support more cases
 * Put all non-driver-initialized memory types into default_memory_types
   instead of using hmat_memory_types
 * find_alloc_memory_type -> mt_find_alloc_memory_type
-v2:
 Thanks to Ying's comments,
 * Rewrite cover letter & patch description
 * Rename functions, don't use _hmat
 * Abstract common functions into find_alloc_memory_type()
 * Use the expected way to use set_node_memory_tier instead of modifying it
 * https://lore.kernel.org/lkml/20240312061729.1997111-1-horenchuang@bytedance.com/T/#u
-v1:
 * https://lore.kernel.org/lkml/20240301082248.3456086-1-horenchuang@bytedance.com/T/#u

Ho-Ren (Jack) Chuang (2):
  memory tier: dax/kmem: create CPUless memory tiers after obtaining
    HMAT info
  memory tier: dax/kmem: abstract memory types put

 drivers/dax/kmem.c           |  20 +------
 include/linux/memory-tiers.h |  13 +++++
 mm/memory-tiers.c            | 106 ++++++++++++++++++++++++++++++++---
 3 files changed, 114 insertions(+), 25 deletions(-)

-- 
Ho-Ren (Jack) Chuang

[PATCH v3 1/2] memory tier: dax/kmem: create CPUless memory tiers after obtaining HMAT info

[Ho-Ren (Jack) Chuang]

The current implementation treats emulated memory devices, such as
CXL1.1 type3 memory, as normal DRAM when they are emulated as normal memory
(E820_TYPE_RAM). However, these emulated devices have different
characteristics than traditional DRAM, making it important to
distinguish them. Thus, we modify the tiered memory initialization process
to introduce a delay specifically for CPUless NUMA nodes. This delay
ensures that the memory tier initialization for these nodes is deferred
until HMAT information is obtained during the boot process. Finally,
demotion tables are recalculated at the end.

More details:
* late_initcall(memory_tier_late_init);
Some device drivers may have initialized memory tiers between
`memory_tier_init()` and `memory_tier_late_init()`, potentially bringing
online memory nodes and configuring memory tiers. They should be excluded
in the late init.

* Abstract common functions into `mt_find_alloc_memory_type()`
Since different memory devices require finding or allocating a memory type,
these common steps are abstracted into a single function,
`mt_find_alloc_memory_type()`, enhancing code scalability and conciseness.

* Handle cases where there is no HMAT when creating memory tiers
There is a scenario where a CPUless node does not provide HMAT information.
If no HMAT is specified, it falls back to using the default DRAM tier.

* Change adist calculation code to use another new lock, `mt_perf_lock`.
In the current implementation, iterating through CPUlist nodes requires
holding the `memory_tier_lock`. However, `mt_calc_adistance()` will end up
trying to acquire the same lock, leading to a potential deadlock.
Therefore, we propose introducing a standalone `mt_perf_lock` to protect
`default_dram_perf`. This approach not only avoids deadlock but also
prevents holding a large lock simultaneously.

* Upgrade `set_node_memory_tier` to support additional cases, including
  default DRAM, late CPUless, and hot-plugged initializations.
To cover hot-plugged memory nodes, `mt_calc_adistance()` and
`mt_find_alloc_memory_type()` are moved into `set_node_memory_tier()` to
handle cases where memtype is not initialized and where HMAT information is
available.

* Introduce `default_memory_types` for those memory types that are not
  initialized by device drivers.
Because late initialized memory and default DRAM memory need to be managed,
a default memory type is created for storing all memory types that are
not initialized by device drivers and as a fallback.

Signed-off-by: Ho-Ren (Jack) Chuang <horenchuang@bytedance.com>
Signed-off-by: Hao Xiang <hao.xiang@bytedance.com>
---
 drivers/dax/kmem.c           | 13 +----
 include/linux/memory-tiers.h |  7 +++
 mm/memory-tiers.c            | 94 +++++++++++++++++++++++++++++++++---
 3 files changed, 95 insertions(+), 19 deletions(-)

diff --git a/drivers/dax/kmem.c b/drivers/dax/kmem.c
index 42ee360cf4e3..de1333aa7b3e 100644
--- a/drivers/dax/kmem.c
+++ b/drivers/dax/kmem.c
@@ -55,21 +55,10 @@ static LIST_HEAD(kmem_memory_types);
 
 static struct memory_dev_type *kmem_find_alloc_memory_type(int adist)
 {
-	bool found = false;
 	struct memory_dev_type *mtype;
 
 	mutex_lock(&kmem_memory_type_lock);
-	list_for_each_entry(mtype, &kmem_memory_types, list) {
-		if (mtype->adistance == adist) {
-			found = true;
-			break;
-		}
-	}
-	if (!found) {
-		mtype = alloc_memory_type(adist);
-		if (!IS_ERR(mtype))
-			list_add(&mtype->list, &kmem_memory_types);
-	}
+	mtype = mt_find_alloc_memory_type(adist, &kmem_memory_types);
 	mutex_unlock(&kmem_memory_type_lock);
 
 	return mtype;
diff --git a/include/linux/memory-tiers.h b/include/linux/memory-tiers.h
index 69e781900082..b2135334ac18 100644
--- a/include/linux/memory-tiers.h
+++ b/include/linux/memory-tiers.h
@@ -48,6 +48,8 @@ int mt_calc_adistance(int node, int *adist);
 int mt_set_default_dram_perf(int nid, struct access_coordinate *perf,
 			     const char *source);
 int mt_perf_to_adistance(struct access_coordinate *perf, int *adist);
+struct memory_dev_type *mt_find_alloc_memory_type(int adist,
+							struct list_head *memory_types);
 #ifdef CONFIG_MIGRATION
 int next_demotion_node(int node);
 void node_get_allowed_targets(pg_data_t *pgdat, nodemask_t *targets);
@@ -136,5 +138,10 @@ static inline int mt_perf_to_adistance(struct access_coordinate *perf, int *adis
 {
 	return -EIO;
 }
+
+struct memory_dev_type *mt_find_alloc_memory_type(int adist, struct list_head *memory_types)
+{
+	return NULL;
+}
 #endif	/* CONFIG_NUMA */
 #endif  /* _LINUX_MEMORY_TIERS_H */
diff --git a/mm/memory-tiers.c b/mm/memory-tiers.c
index 0537664620e5..d9b96b21b65a 100644
--- a/mm/memory-tiers.c
+++ b/mm/memory-tiers.c
@@ -6,6 +6,7 @@
 #include <linux/memory.h>
 #include <linux/memory-tiers.h>
 #include <linux/notifier.h>
+#include <linux/acpi.h>
 
 #include "internal.h"
 
@@ -36,6 +37,11 @@ struct node_memory_type_map {
 
 static DEFINE_MUTEX(memory_tier_lock);
 static LIST_HEAD(memory_tiers);
+/*
+ * The list is used to store all memory types that are not created
+ * by a device driver.
+ */
+static LIST_HEAD(default_memory_types);
 static struct node_memory_type_map node_memory_types[MAX_NUMNODES];
 struct memory_dev_type *default_dram_type;
 
@@ -505,7 +511,8 @@ static inline void __init_node_memory_type(int node, struct memory_dev_type *mem
 static struct memory_tier *set_node_memory_tier(int node)
 {
 	struct memory_tier *memtier;
-	struct memory_dev_type *memtype;
+	struct memory_dev_type *memtype, *mtype = NULL;
+	int adist = MEMTIER_ADISTANCE_DRAM;
 	pg_data_t *pgdat = NODE_DATA(node);
 
 
@@ -514,7 +521,18 @@ static struct memory_tier *set_node_memory_tier(int node)
 	if (!node_state(node, N_MEMORY))
 		return ERR_PTR(-EINVAL);
 
-	__init_node_memory_type(node, default_dram_type);
+	mt_calc_adistance(node, &adist);
+	if (adist != MEMTIER_ADISTANCE_DRAM &&
+			node_memory_types[node].memtype == NULL) {
+		mtype = mt_find_alloc_memory_type(adist, &default_memory_types);
+		if (IS_ERR(mtype)) {
+			mtype = default_dram_type;
+			pr_info("Failed to allocate a memory type. Fall back.\n");
+		}
+	} else
+		mtype = default_dram_type;
+
+	__init_node_memory_type(node, mtype);
 
 	memtype = node_memory_types[node].memtype;
 	node_set(node, memtype->nodes);
@@ -623,6 +641,55 @@ void clear_node_memory_type(int node, struct memory_dev_type *memtype)
 }
 EXPORT_SYMBOL_GPL(clear_node_memory_type);
 
+struct memory_dev_type *mt_find_alloc_memory_type(int adist, struct list_head *memory_types)
+{
+	bool found = false;
+	struct memory_dev_type *mtype;
+
+	list_for_each_entry(mtype, memory_types, list) {
+		if (mtype->adistance == adist) {
+			found = true;
+			break;
+		}
+	}
+	if (!found) {
+		mtype = alloc_memory_type(adist);
+		if (!IS_ERR(mtype))
+			list_add(&mtype->list, memory_types);
+	}
+
+	return mtype;
+}
+EXPORT_SYMBOL_GPL(mt_find_alloc_memory_type);
+
+/*
+ * This is invoked via late_initcall() to create
+ * CPUless memory tiers after HMAT info is ready or
+ * when there is no HMAT.
+ */
+static int __init memory_tier_late_init(void)
+{
+	int nid;
+
+	mutex_lock(&memory_tier_lock);
+	for_each_node_state(nid, N_MEMORY)
+		if (!node_state(nid, N_CPU) &&
+			node_memory_types[nid].memtype == NULL)
+			/*
+			 * Some device drivers may have initialized memory tiers
+			 * between `memory_tier_init()` and `memory_tier_late_init()`,
+			 * potentially bringing online memory nodes and
+			 * configuring memory tiers. Exclude them here.
+			 */
+			set_node_memory_tier(nid);
+
+	establish_demotion_targets();
+	mutex_unlock(&memory_tier_lock);
+
+	return 0;
+}
+late_initcall(memory_tier_late_init);
+
 static void dump_hmem_attrs(struct access_coordinate *coord, const char *prefix)
 {
 	pr_info(
@@ -631,12 +698,16 @@ static void dump_hmem_attrs(struct access_coordinate *coord, const char *prefix)
 		coord->read_bandwidth, coord->write_bandwidth);
 }
 
+/*
+ * The lock is used to protect the default_dram_perf.
+ */
+static DEFINE_MUTEX(mt_perf_lock);
 int mt_set_default_dram_perf(int nid, struct access_coordinate *perf,
 			     const char *source)
 {
 	int rc = 0;
 
-	mutex_lock(&memory_tier_lock);
+	mutex_lock(&mt_perf_lock);
 	if (default_dram_perf_error) {
 		rc = -EIO;
 		goto out;
@@ -684,7 +755,7 @@ int mt_set_default_dram_perf(int nid, struct access_coordinate *perf,
 	}
 
 out:
-	mutex_unlock(&memory_tier_lock);
+	mutex_unlock(&mt_perf_lock);
 	return rc;
 }
 
@@ -700,7 +771,7 @@ int mt_perf_to_adistance(struct access_coordinate *perf, int *adist)
 	    perf->read_bandwidth + perf->write_bandwidth == 0)
 		return -EINVAL;
 
-	mutex_lock(&memory_tier_lock);
+	mutex_lock(&mt_perf_lock);
 	/*
 	 * The abstract distance of a memory node is in direct proportion to
 	 * its memory latency (read + write) and inversely proportional to its
@@ -713,7 +784,7 @@ int mt_perf_to_adistance(struct access_coordinate *perf, int *adist)
 		(default_dram_perf.read_latency + default_dram_perf.write_latency) *
 		(default_dram_perf.read_bandwidth + default_dram_perf.write_bandwidth) /
 		(perf->read_bandwidth + perf->write_bandwidth);
-	mutex_unlock(&memory_tier_lock);
+	mutex_unlock(&mt_perf_lock);
 
 	return 0;
 }
@@ -826,7 +897,8 @@ static int __init memory_tier_init(void)
 	 * For now we can have 4 faster memory tiers with smaller adistance
 	 * than default DRAM tier.
 	 */
-	default_dram_type = alloc_memory_type(MEMTIER_ADISTANCE_DRAM);
+	default_dram_type = mt_find_alloc_memory_type(
+					MEMTIER_ADISTANCE_DRAM, &default_memory_types);
 	if (IS_ERR(default_dram_type))
 		panic("%s() failed to allocate default DRAM tier\n", __func__);
 
@@ -836,6 +908,14 @@ static int __init memory_tier_init(void)
 	 * types assigned.
 	 */
 	for_each_node_state(node, N_MEMORY) {
+		if (!node_state(node, N_CPU))
+			/*
+			 * Defer memory tier initialization on CPUless numa nodes.
+			 * These will be initialized after firmware and devices are
+			 * initialized.
+			 */
+			continue;
+
 		memtier = set_node_memory_tier(node);
 		if (IS_ERR(memtier))
 			/*
-- 
Ho-Ren (Jack) Chuang

[PATCH v3 2/2] memory tier: dax/kmem: abstract memory types put

[Ho-Ren (Jack) Chuang]

Abstract `kmem_put_memory_types()` into `mt_put_memory_types()` to
accommodate various memory types and enhance flexibility,
similar to `mt_find_alloc_memory_type()`.

Signed-off-by: Ho-Ren (Jack) Chuang <horenchuang@bytedance.com>
---
 drivers/dax/kmem.c           |  7 +------
 include/linux/memory-tiers.h |  6 ++++++
 mm/memory-tiers.c            | 12 ++++++++++++
 3 files changed, 19 insertions(+), 6 deletions(-)

diff --git a/drivers/dax/kmem.c b/drivers/dax/kmem.c
index de1333aa7b3e..01399e5b53b2 100644
--- a/drivers/dax/kmem.c
+++ b/drivers/dax/kmem.c
@@ -66,13 +66,8 @@ static struct memory_dev_type *kmem_find_alloc_memory_type(int adist)
 
 static void kmem_put_memory_types(void)
 {
-	struct memory_dev_type *mtype, *mtn;
-
 	mutex_lock(&kmem_memory_type_lock);
-	list_for_each_entry_safe(mtype, mtn, &kmem_memory_types, list) {
-		list_del(&mtype->list);
-		put_memory_type(mtype);
-	}
+	mt_put_memory_types(&kmem_memory_types);
 	mutex_unlock(&kmem_memory_type_lock);
 }
 
diff --git a/include/linux/memory-tiers.h b/include/linux/memory-tiers.h
index b2135334ac18..a44c03c2ba3a 100644
--- a/include/linux/memory-tiers.h
+++ b/include/linux/memory-tiers.h
@@ -50,6 +50,7 @@ int mt_set_default_dram_perf(int nid, struct access_coordinate *perf,
 int mt_perf_to_adistance(struct access_coordinate *perf, int *adist);
 struct memory_dev_type *mt_find_alloc_memory_type(int adist,
 							struct list_head *memory_types);
+void mt_put_memory_types(struct list_head *memory_types);
 #ifdef CONFIG_MIGRATION
 int next_demotion_node(int node);
 void node_get_allowed_targets(pg_data_t *pgdat, nodemask_t *targets);
@@ -143,5 +144,10 @@ struct memory_dev_type *mt_find_alloc_memory_type(int adist, struct list_head *m
 {
 	return NULL;
 }
+
+void mt_put_memory_types(struct list_head *memory_types)
+{
+
+}
 #endif	/* CONFIG_NUMA */
 #endif  /* _LINUX_MEMORY_TIERS_H */
diff --git a/mm/memory-tiers.c b/mm/memory-tiers.c
index d9b96b21b65a..6246c28546ba 100644
--- a/mm/memory-tiers.c
+++ b/mm/memory-tiers.c
@@ -662,6 +662,18 @@ struct memory_dev_type *mt_find_alloc_memory_type(int adist, struct list_head *m
 }
 EXPORT_SYMBOL_GPL(mt_find_alloc_memory_type);
 
+
+void mt_put_memory_types(struct list_head *memory_types)
+{
+	struct memory_dev_type *mtype, *mtn;
+
+	list_for_each_entry_safe(mtype, mtn, memory_types, list) {
+		list_del(&mtype->list);
+		put_memory_type(mtype);
+	}
+}
+EXPORT_SYMBOL_GPL(mt_put_memory_types);
+
 /*
  * This is invoked via late_initcall() to create
  * CPUless memory tiers after HMAT info is ready or
-- 
Ho-Ren (Jack) Chuang

Re: [PATCH v3 1/2] memory tier: dax/kmem: create CPUless memory tiers after obtaining HMAT info

[Huang, Ying]

"Ho-Ren (Jack) Chuang" <horenchuang@bytedance.com> writes:

> The current implementation treats emulated memory devices, such as
> CXL1.1 type3 memory, as normal DRAM when they are emulated as normal memory
> (E820_TYPE_RAM). However, these emulated devices have different
> characteristics than traditional DRAM, making it important to
> distinguish them. Thus, we modify the tiered memory initialization process
> to introduce a delay specifically for CPUless NUMA nodes. This delay
> ensures that the memory tier initialization for these nodes is deferred
> until HMAT information is obtained during the boot process. Finally,
> demotion tables are recalculated at the end.
>
> More details:

You have done several stuff in one patch.  So you need "more details".
You may separate them into multiple patches.  One for echo "*" below.
But I have no strong opinion on that.

> * late_initcall(memory_tier_late_init);
> Some device drivers may have initialized memory tiers between
> `memory_tier_init()` and `memory_tier_late_init()`, potentially bringing
> online memory nodes and configuring memory tiers. They should be excluded
> in the late init.
>
> * Abstract common functions into `mt_find_alloc_memory_type()`
> Since different memory devices require finding or allocating a memory type,
> these common steps are abstracted into a single function,
> `mt_find_alloc_memory_type()`, enhancing code scalability and conciseness.
>
> * Handle cases where there is no HMAT when creating memory tiers
> There is a scenario where a CPUless node does not provide HMAT information.
> If no HMAT is specified, it falls back to using the default DRAM tier.
>
> * Change adist calculation code to use another new lock, `mt_perf_lock`.
> In the current implementation, iterating through CPUlist nodes requires
> holding the `memory_tier_lock`. However, `mt_calc_adistance()` will end up
> trying to acquire the same lock, leading to a potential deadlock.
> Therefore, we propose introducing a standalone `mt_perf_lock` to protect
> `default_dram_perf`. This approach not only avoids deadlock but also
> prevents holding a large lock simultaneously.
>
> * Upgrade `set_node_memory_tier` to support additional cases, including
>   default DRAM, late CPUless, and hot-plugged initializations.
> To cover hot-plugged memory nodes, `mt_calc_adistance()` and
> `mt_find_alloc_memory_type()` are moved into `set_node_memory_tier()` to
> handle cases where memtype is not initialized and where HMAT information is
> available.
>
> * Introduce `default_memory_types` for those memory types that are not
>   initialized by device drivers.
> Because late initialized memory and default DRAM memory need to be managed,
> a default memory type is created for storing all memory types that are
> not initialized by device drivers and as a fallback.
>
> Signed-off-by: Ho-Ren (Jack) Chuang <horenchuang@bytedance.com>
> Signed-off-by: Hao Xiang <hao.xiang@bytedance.com>
> ---
>  drivers/dax/kmem.c           | 13 +----
>  include/linux/memory-tiers.h |  7 +++
>  mm/memory-tiers.c            | 94 +++++++++++++++++++++++++++++++++---
>  3 files changed, 95 insertions(+), 19 deletions(-)
>
> diff --git a/drivers/dax/kmem.c b/drivers/dax/kmem.c
> index 42ee360cf4e3..de1333aa7b3e 100644
> --- a/drivers/dax/kmem.c
> +++ b/drivers/dax/kmem.c
> @@ -55,21 +55,10 @@ static LIST_HEAD(kmem_memory_types);
>  
>  static struct memory_dev_type *kmem_find_alloc_memory_type(int adist)
>  {
> -	bool found = false;
>  	struct memory_dev_type *mtype;
>  
>  	mutex_lock(&kmem_memory_type_lock);
> -	list_for_each_entry(mtype, &kmem_memory_types, list) {
> -		if (mtype->adistance == adist) {
> -			found = true;
> -			break;
> -		}
> -	}
> -	if (!found) {
> -		mtype = alloc_memory_type(adist);
> -		if (!IS_ERR(mtype))
> -			list_add(&mtype->list, &kmem_memory_types);
> -	}
> +	mtype = mt_find_alloc_memory_type(adist, &kmem_memory_types);
>  	mutex_unlock(&kmem_memory_type_lock);
>  
>  	return mtype;

It seems that there's some miscommunication about my previous comments
about this.  What I suggested is to create one separate patch, which
moves mt_find_alloc_memory_type() and mt_put_memory_types() into
memory-tiers.c.  And make this patch the first one of the series.

> diff --git a/include/linux/memory-tiers.h b/include/linux/memory-tiers.h
> index 69e781900082..b2135334ac18 100644
> --- a/include/linux/memory-tiers.h
> +++ b/include/linux/memory-tiers.h
> @@ -48,6 +48,8 @@ int mt_calc_adistance(int node, int *adist);
>  int mt_set_default_dram_perf(int nid, struct access_coordinate *perf,
>  			     const char *source);
>  int mt_perf_to_adistance(struct access_coordinate *perf, int *adist);
> +struct memory_dev_type *mt_find_alloc_memory_type(int adist,
> +							struct list_head *memory_types);
>  #ifdef CONFIG_MIGRATION
>  int next_demotion_node(int node);
>  void node_get_allowed_targets(pg_data_t *pgdat, nodemask_t *targets);
> @@ -136,5 +138,10 @@ static inline int mt_perf_to_adistance(struct access_coordinate *perf, int *adis
>  {
>  	return -EIO;
>  }
> +
> +struct memory_dev_type *mt_find_alloc_memory_type(int adist, struct list_head *memory_types)
> +{
> +	return NULL;
> +}
>  #endif	/* CONFIG_NUMA */
>  #endif  /* _LINUX_MEMORY_TIERS_H */
> diff --git a/mm/memory-tiers.c b/mm/memory-tiers.c
> index 0537664620e5..d9b96b21b65a 100644
> --- a/mm/memory-tiers.c
> +++ b/mm/memory-tiers.c
> @@ -6,6 +6,7 @@
>  #include <linux/memory.h>
>  #include <linux/memory-tiers.h>
>  #include <linux/notifier.h>
> +#include <linux/acpi.h>

We don't need this anymore.

>  #include "internal.h"
>  
> @@ -36,6 +37,11 @@ struct node_memory_type_map {
>  
>  static DEFINE_MUTEX(memory_tier_lock);
>  static LIST_HEAD(memory_tiers);
> +/*
> + * The list is used to store all memory types that are not created
> + * by a device driver.
> + */
> +static LIST_HEAD(default_memory_types);
>  static struct node_memory_type_map node_memory_types[MAX_NUMNODES];
>  struct memory_dev_type *default_dram_type;
>  
> @@ -505,7 +511,8 @@ static inline void __init_node_memory_type(int node, struct memory_dev_type *mem
>  static struct memory_tier *set_node_memory_tier(int node)
>  {
>  	struct memory_tier *memtier;
> -	struct memory_dev_type *memtype;
> +	struct memory_dev_type *memtype, *mtype = NULL;

It seems unnecessary to introduce another variable, just use memtype?

> +	int adist = MEMTIER_ADISTANCE_DRAM;
>  	pg_data_t *pgdat = NODE_DATA(node);
>  
>  
> @@ -514,7 +521,18 @@ static struct memory_tier *set_node_memory_tier(int node)
>  	if (!node_state(node, N_MEMORY))
>  		return ERR_PTR(-EINVAL);
>  
> -	__init_node_memory_type(node, default_dram_type);
> +	mt_calc_adistance(node, &adist);
> +	if (adist != MEMTIER_ADISTANCE_DRAM &&
> +			node_memory_types[node].memtype == NULL) {
> +		mtype = mt_find_alloc_memory_type(adist, &default_memory_types);
> +		if (IS_ERR(mtype)) {
> +			mtype = default_dram_type;
> +			pr_info("Failed to allocate a memory type. Fall back.\n");
> +		}
> +	} else
> +		mtype = default_dram_type;

This can be simplified to

	mt_calc_adistance(node, &adist);
	if (node_memory_types[node].memtype == NULL) {
		mtype = mt_find_alloc_memory_type(adist, &default_memory_types);
		if (IS_ERR(mtype)) {
			mtype = default_dram_type;
			pr_info("Failed to allocate a memory type. Fall back.\n");
		}
	}

> +	__init_node_memory_type(node, mtype);
>  
>  	memtype = node_memory_types[node].memtype;
>  	node_set(node, memtype->nodes);
> @@ -623,6 +641,55 @@ void clear_node_memory_type(int node, struct memory_dev_type *memtype)
>  }
>  EXPORT_SYMBOL_GPL(clear_node_memory_type);
>  
> +struct memory_dev_type *mt_find_alloc_memory_type(int adist, struct list_head *memory_types)
> +{
> +	bool found = false;
> +	struct memory_dev_type *mtype;
> +
> +	list_for_each_entry(mtype, memory_types, list) {
> +		if (mtype->adistance == adist) {
> +			found = true;
> +			break;
> +		}
> +	}
> +	if (!found) {
> +		mtype = alloc_memory_type(adist);
> +		if (!IS_ERR(mtype))
> +			list_add(&mtype->list, memory_types);
> +	}
> +
> +	return mtype;
> +}
> +EXPORT_SYMBOL_GPL(mt_find_alloc_memory_type);
> +
> +/*
> + * This is invoked via late_initcall() to create
> + * CPUless memory tiers after HMAT info is ready or
> + * when there is no HMAT.

Better to avoid HMAT in general code.  How about something as below?

This is invoked via late_initcall() to initialize memory tiers for
CPU-less memory nodes after drivers initialization.  Which is
expect to provide adistance algorithms.

> + */
> +static int __init memory_tier_late_init(void)
> +{
> +	int nid;
> +
> +	mutex_lock(&memory_tier_lock);
> +	for_each_node_state(nid, N_MEMORY)
> +		if (!node_state(nid, N_CPU) &&
> +			node_memory_types[nid].memtype == NULL)
> +			/*
> +			 * Some device drivers may have initialized memory tiers
> +			 * between `memory_tier_init()` and `memory_tier_late_init()`,
> +			 * potentially bringing online memory nodes and
> +			 * configuring memory tiers. Exclude them here.
> +			 */
> +			set_node_memory_tier(nid);
> +
> +	establish_demotion_targets();
> +	mutex_unlock(&memory_tier_lock);
> +
> +	return 0;
> +}
> +late_initcall(memory_tier_late_init);
> +
>  static void dump_hmem_attrs(struct access_coordinate *coord, const char *prefix)
>  {
>  	pr_info(
> @@ -631,12 +698,16 @@ static void dump_hmem_attrs(struct access_coordinate *coord, const char *prefix)
>  		coord->read_bandwidth, coord->write_bandwidth);
>  }
>  
> +/*
> + * The lock is used to protect the default_dram_perf.
> + */
> +static DEFINE_MUTEX(mt_perf_lock);

Miscommunication here too.  Should be moved to near the
"default_dram_perf" definition.  And it protects not only
default_dram_perf.

>  int mt_set_default_dram_perf(int nid, struct access_coordinate *perf,
>  			     const char *source)
>  {
>  	int rc = 0;
>  
> -	mutex_lock(&memory_tier_lock);
> +	mutex_lock(&mt_perf_lock);
>  	if (default_dram_perf_error) {
>  		rc = -EIO;
>  		goto out;
> @@ -684,7 +755,7 @@ int mt_set_default_dram_perf(int nid, struct access_coordinate *perf,
>  	}
>  
>  out:
> -	mutex_unlock(&memory_tier_lock);
> +	mutex_unlock(&mt_perf_lock);
>  	return rc;
>  }
>  
> @@ -700,7 +771,7 @@ int mt_perf_to_adistance(struct access_coordinate *perf, int *adist)
>  	    perf->read_bandwidth + perf->write_bandwidth == 0)
>  		return -EINVAL;
>  
> -	mutex_lock(&memory_tier_lock);
> +	mutex_lock(&mt_perf_lock);
>  	/*
>  	 * The abstract distance of a memory node is in direct proportion to
>  	 * its memory latency (read + write) and inversely proportional to its
> @@ -713,7 +784,7 @@ int mt_perf_to_adistance(struct access_coordinate *perf, int *adist)
>  		(default_dram_perf.read_latency + default_dram_perf.write_latency) *
>  		(default_dram_perf.read_bandwidth + default_dram_perf.write_bandwidth) /
>  		(perf->read_bandwidth + perf->write_bandwidth);
> -	mutex_unlock(&memory_tier_lock);
> +	mutex_unlock(&mt_perf_lock);
>  
>  	return 0;
>  }
> @@ -826,7 +897,8 @@ static int __init memory_tier_init(void)
>  	 * For now we can have 4 faster memory tiers with smaller adistance
>  	 * than default DRAM tier.
>  	 */
> -	default_dram_type = alloc_memory_type(MEMTIER_ADISTANCE_DRAM);
> +	default_dram_type = mt_find_alloc_memory_type(
> +					MEMTIER_ADISTANCE_DRAM, &default_memory_types);
>  	if (IS_ERR(default_dram_type))
>  		panic("%s() failed to allocate default DRAM tier\n", __func__);
>  
> @@ -836,6 +908,14 @@ static int __init memory_tier_init(void)
>  	 * types assigned.
>  	 */
>  	for_each_node_state(node, N_MEMORY) {
> +		if (!node_state(node, N_CPU))
> +			/*
> +			 * Defer memory tier initialization on CPUless numa nodes.
> +			 * These will be initialized after firmware and devices are
> +			 * initialized.
> +			 */
> +			continue;
> +
>  		memtier = set_node_memory_tier(node);
>  		if (IS_ERR(memtier))
>  			/*

--
Best Regards,
Huang, Ying

Re: [External] Re: [PATCH v3 1/2] memory tier: dax/kmem: create CPUless memory tiers after obtaining HMAT info

[Ho-Ren (Jack) Chuang]

On Wed, Mar 20, 2024 at 12:15 AM Huang, Ying <ying.huang@intel.com> wrote:
>
> "Ho-Ren (Jack) Chuang" <horenchuang@bytedance.com> writes:
>
> > The current implementation treats emulated memory devices, such as
> > CXL1.1 type3 memory, as normal DRAM when they are emulated as normal memory
> > (E820_TYPE_RAM). However, these emulated devices have different
> > characteristics than traditional DRAM, making it important to
> > distinguish them. Thus, we modify the tiered memory initialization process
> > to introduce a delay specifically for CPUless NUMA nodes. This delay
> > ensures that the memory tier initialization for these nodes is deferred
> > until HMAT information is obtained during the boot process. Finally,
> > demotion tables are recalculated at the end.
> >
> > More details:
>
> You have done several stuff in one patch.  So you need "more details".
> You may separate them into multiple patches.  One for echo "*" below.
> But I have no strong opinion on that.
>
> > * late_initcall(memory_tier_late_init);
> > Some device drivers may have initialized memory tiers between
> > `memory_tier_init()` and `memory_tier_late_init()`, potentially bringing
> > online memory nodes and configuring memory tiers. They should be excluded
> > in the late init.
> >
> > * Abstract common functions into `mt_find_alloc_memory_type()`
> > Since different memory devices require finding or allocating a memory type,
> > these common steps are abstracted into a single function,
> > `mt_find_alloc_memory_type()`, enhancing code scalability and conciseness.
> >
> > * Handle cases where there is no HMAT when creating memory tiers
> > There is a scenario where a CPUless node does not provide HMAT information.
> > If no HMAT is specified, it falls back to using the default DRAM tier.
> >
> > * Change adist calculation code to use another new lock, `mt_perf_lock`.
> > In the current implementation, iterating through CPUlist nodes requires
> > holding the `memory_tier_lock`. However, `mt_calc_adistance()` will end up
> > trying to acquire the same lock, leading to a potential deadlock.
> > Therefore, we propose introducing a standalone `mt_perf_lock` to protect
> > `default_dram_perf`. This approach not only avoids deadlock but also
> > prevents holding a large lock simultaneously.
> >
> > * Upgrade `set_node_memory_tier` to support additional cases, including
> >   default DRAM, late CPUless, and hot-plugged initializations.
> > To cover hot-plugged memory nodes, `mt_calc_adistance()` and
> > `mt_find_alloc_memory_type()` are moved into `set_node_memory_tier()` to
> > handle cases where memtype is not initialized and where HMAT information is
> > available.
> >
> > * Introduce `default_memory_types` for those memory types that are not
> >   initialized by device drivers.
> > Because late initialized memory and default DRAM memory need to be managed,
> > a default memory type is created for storing all memory types that are
> > not initialized by device drivers and as a fallback.
> >
> > Signed-off-by: Ho-Ren (Jack) Chuang <horenchuang@bytedance.com>
> > Signed-off-by: Hao Xiang <hao.xiang@bytedance.com>
> > ---
> >  drivers/dax/kmem.c           | 13 +----
> >  include/linux/memory-tiers.h |  7 +++
> >  mm/memory-tiers.c            | 94 +++++++++++++++++++++++++++++++++---
> >  3 files changed, 95 insertions(+), 19 deletions(-)
> >
> > diff --git a/drivers/dax/kmem.c b/drivers/dax/kmem.c
> > index 42ee360cf4e3..de1333aa7b3e 100644
> > --- a/drivers/dax/kmem.c
> > +++ b/drivers/dax/kmem.c
> > @@ -55,21 +55,10 @@ static LIST_HEAD(kmem_memory_types);
> >
> >  static struct memory_dev_type *kmem_find_alloc_memory_type(int adist)
> >  {
> > -     bool found = false;
> >       struct memory_dev_type *mtype;
> >
> >       mutex_lock(&kmem_memory_type_lock);
> > -     list_for_each_entry(mtype, &kmem_memory_types, list) {
> > -             if (mtype->adistance == adist) {
> > -                     found = true;
> > -                     break;
> > -             }
> > -     }
> > -     if (!found) {
> > -             mtype = alloc_memory_type(adist);
> > -             if (!IS_ERR(mtype))
> > -                     list_add(&mtype->list, &kmem_memory_types);
> > -     }
> > +     mtype = mt_find_alloc_memory_type(adist, &kmem_memory_types);
> >       mutex_unlock(&kmem_memory_type_lock);
> >
> >       return mtype;
>
> It seems that there's some miscommunication about my previous comments
> about this.  What I suggested is to create one separate patch, which
> moves mt_find_alloc_memory_type() and mt_put_memory_types() into
> memory-tiers.c.  And make this patch the first one of the series.
>

I will make mt_find_alloc/mt_put_memory_type changes as
a separate patch and the first of my patch series. Thanks.


> > diff --git a/include/linux/memory-tiers.h b/include/linux/memory-tiers.h
> > index 69e781900082..b2135334ac18 100644
> > --- a/include/linux/memory-tiers.h
> > +++ b/include/linux/memory-tiers.h
> > @@ -48,6 +48,8 @@ int mt_calc_adistance(int node, int *adist);
> >  int mt_set_default_dram_perf(int nid, struct access_coordinate *perf,
> >                            const char *source);
> >  int mt_perf_to_adistance(struct access_coordinate *perf, int *adist);
> > +struct memory_dev_type *mt_find_alloc_memory_type(int adist,
> > +                                                     struct list_head *memory_types);
> >  #ifdef CONFIG_MIGRATION
> >  int next_demotion_node(int node);
> >  void node_get_allowed_targets(pg_data_t *pgdat, nodemask_t *targets);
> > @@ -136,5 +138,10 @@ static inline int mt_perf_to_adistance(struct access_coordinate *perf, int *adis
> >  {
> >       return -EIO;
> >  }
> > +
> > +struct memory_dev_type *mt_find_alloc_memory_type(int adist, struct list_head *memory_types)
> > +{
> > +     return NULL;
> > +}
> >  #endif       /* CONFIG_NUMA */
> >  #endif  /* _LINUX_MEMORY_TIERS_H */
> > diff --git a/mm/memory-tiers.c b/mm/memory-tiers.c
> > index 0537664620e5..d9b96b21b65a 100644
> > --- a/mm/memory-tiers.c
> > +++ b/mm/memory-tiers.c
> > @@ -6,6 +6,7 @@
> >  #include <linux/memory.h>
> >  #include <linux/memory-tiers.h>
> >  #include <linux/notifier.h>
> > +#include <linux/acpi.h>
>
> We don't need this anymore.
>

Thanks. I will remove it.

> >  #include "internal.h"
> >
> > @@ -36,6 +37,11 @@ struct node_memory_type_map {
> >
> >  static DEFINE_MUTEX(memory_tier_lock);
> >  static LIST_HEAD(memory_tiers);
> > +/*
> > + * The list is used to store all memory types that are not created
> > + * by a device driver.
> > + */
> > +static LIST_HEAD(default_memory_types);
> >  static struct node_memory_type_map node_memory_types[MAX_NUMNODES];
> >  struct memory_dev_type *default_dram_type;
> >
> > @@ -505,7 +511,8 @@ static inline void __init_node_memory_type(int node, struct memory_dev_type *mem
> >  static struct memory_tier *set_node_memory_tier(int node)
> >  {
> >       struct memory_tier *memtier;
> > -     struct memory_dev_type *memtype;
> > +     struct memory_dev_type *memtype, *mtype = NULL;
>
> It seems unnecessary to introduce another variable, just use memtype?
>

Yes, I will consolidate them.


> > +     int adist = MEMTIER_ADISTANCE_DRAM;
> >       pg_data_t *pgdat = NODE_DATA(node);
> >
> >
> > @@ -514,7 +521,18 @@ static struct memory_tier *set_node_memory_tier(int node)
> >       if (!node_state(node, N_MEMORY))
> >               return ERR_PTR(-EINVAL);
> >
> > -     __init_node_memory_type(node, default_dram_type);
> > +     mt_calc_adistance(node, &adist);
> > +     if (adist != MEMTIER_ADISTANCE_DRAM &&
> > +                     node_memory_types[node].memtype == NULL) {
> > +             mtype = mt_find_alloc_memory_type(adist, &default_memory_types);
> > +             if (IS_ERR(mtype)) {
> > +                     mtype = default_dram_type;
> > +                     pr_info("Failed to allocate a memory type. Fall back.\n");
> > +             }
> > +     } else
> > +             mtype = default_dram_type;
>
> This can be simplified to
>
>         mt_calc_adistance(node, &adist);
>         if (node_memory_types[node].memtype == NULL) {
>                 mtype = mt_find_alloc_memory_type(adist, &default_memory_types);
>                 if (IS_ERR(mtype)) {
>                         mtype = default_dram_type;
>                         pr_info("Failed to allocate a memory type. Fall back.\n");
>                 }
>         }
>

Sounds good! I will do.


> > +     __init_node_memory_type(node, mtype);
> >
> >       memtype = node_memory_types[node].memtype;
> >       node_set(node, memtype->nodes);
> > @@ -623,6 +641,55 @@ void clear_node_memory_type(int node, struct memory_dev_type *memtype)
> >  }
> >  EXPORT_SYMBOL_GPL(clear_node_memory_type);
> >
> > +struct memory_dev_type *mt_find_alloc_memory_type(int adist, struct list_head *memory_types)
> > +{
> > +     bool found = false;
> > +     struct memory_dev_type *mtype;
> > +
> > +     list_for_each_entry(mtype, memory_types, list) {
> > +             if (mtype->adistance == adist) {
> > +                     found = true;
> > +                     break;
> > +             }
> > +     }
> > +     if (!found) {
> > +             mtype = alloc_memory_type(adist);
> > +             if (!IS_ERR(mtype))
> > +                     list_add(&mtype->list, memory_types);
> > +     }
> > +
> > +     return mtype;
> > +}
> > +EXPORT_SYMBOL_GPL(mt_find_alloc_memory_type);
> > +
> > +/*
> > + * This is invoked via late_initcall() to create
> > + * CPUless memory tiers after HMAT info is ready or
> > + * when there is no HMAT.
>
> Better to avoid HMAT in general code.  How about something as below?
>
> This is invoked via late_initcall() to initialize memory tiers for
> CPU-less memory nodes after drivers initialization.  Which is
> expect to provide adistance algorithms.
>

Got it. Thanks.
"
This is invoked via `late_initcall()` to initialize memory tiers for
CPU-less memory nodes after driver initialization, which is
expected to provide `adistance` algorithms.
"


> > + */
> > +static int __init memory_tier_late_init(void)
> > +{
> > +     int nid;
> > +
> > +     mutex_lock(&memory_tier_lock);
> > +     for_each_node_state(nid, N_MEMORY)
> > +             if (!node_state(nid, N_CPU) &&
> > +                     node_memory_types[nid].memtype == NULL)
> > +                     /*
> > +                      * Some device drivers may have initialized memory tiers
> > +                      * between `memory_tier_init()` and `memory_tier_late_init()`,
> > +                      * potentially bringing online memory nodes and
> > +                      * configuring memory tiers. Exclude them here.
> > +                      */
> > +                     set_node_memory_tier(nid);
> > +
> > +     establish_demotion_targets();
> > +     mutex_unlock(&memory_tier_lock);
> > +
> > +     return 0;
> > +}
> > +late_initcall(memory_tier_late_init);
> > +
> >  static void dump_hmem_attrs(struct access_coordinate *coord, const char *prefix)
> >  {
> >       pr_info(
> > @@ -631,12 +698,16 @@ static void dump_hmem_attrs(struct access_coordinate *coord, const char *prefix)
> >               coord->read_bandwidth, coord->write_bandwidth);
> >  }
> >
> > +/*
> > + * The lock is used to protect the default_dram_perf.
> > + */
> > +static DEFINE_MUTEX(mt_perf_lock);
>
> Miscommunication here too.  Should be moved to near the
> "default_dram_perf" definition.  And it protects not only
> default_dram_perf.
>

I will move it closer to default_dram_perf*.
And change it to:
+/*
+ * The lock is used to protect default_dram_perf*.
+ */


> >  int mt_set_default_dram_perf(int nid, struct access_coordinate *perf,
> >                            const char *source)
> >  {
> >       int rc = 0;
> >
> > -     mutex_lock(&memory_tier_lock);
> > +     mutex_lock(&mt_perf_lock);
> >       if (default_dram_perf_error) {
> >               rc = -EIO;
> >               goto out;
> > @@ -684,7 +755,7 @@ int mt_set_default_dram_perf(int nid, struct access_coordinate *perf,
> >       }
> >
> >  out:
> > -     mutex_unlock(&memory_tier_lock);
> > +     mutex_unlock(&mt_perf_lock);
> >       return rc;
> >  }
> >
> > @@ -700,7 +771,7 @@ int mt_perf_to_adistance(struct access_coordinate *perf, int *adist)
> >           perf->read_bandwidth + perf->write_bandwidth == 0)
> >               return -EINVAL;
> >
> > -     mutex_lock(&memory_tier_lock);
> > +     mutex_lock(&mt_perf_lock);
> >       /*
> >        * The abstract distance of a memory node is in direct proportion to
> >        * its memory latency (read + write) and inversely proportional to its
> > @@ -713,7 +784,7 @@ int mt_perf_to_adistance(struct access_coordinate *perf, int *adist)
> >               (default_dram_perf.read_latency + default_dram_perf.write_latency) *
> >               (default_dram_perf.read_bandwidth + default_dram_perf.write_bandwidth) /
> >               (perf->read_bandwidth + perf->write_bandwidth);
> > -     mutex_unlock(&memory_tier_lock);
> > +     mutex_unlock(&mt_perf_lock);
> >
> >       return 0;
> >  }
> > @@ -826,7 +897,8 @@ static int __init memory_tier_init(void)
> >        * For now we can have 4 faster memory tiers with smaller adistance
> >        * than default DRAM tier.
> >        */
> > -     default_dram_type = alloc_memory_type(MEMTIER_ADISTANCE_DRAM);
> > +     default_dram_type = mt_find_alloc_memory_type(
> > +                                     MEMTIER_ADISTANCE_DRAM, &default_memory_types);
> >       if (IS_ERR(default_dram_type))
> >               panic("%s() failed to allocate default DRAM tier\n", __func__);
> >
> > @@ -836,6 +908,14 @@ static int __init memory_tier_init(void)
> >        * types assigned.
> >        */
> >       for_each_node_state(node, N_MEMORY) {
> > +             if (!node_state(node, N_CPU))
> > +                     /*
> > +                      * Defer memory tier initialization on CPUless numa nodes.
> > +                      * These will be initialized after firmware and devices are
> > +                      * initialized.
> > +                      */
> > +                     continue;
> > +
> >               memtier = set_node_memory_tier(node);
> >               if (IS_ERR(memtier))
> >                       /*
>
> --
> Best Regards,
> Huang, Ying



-- 
Best regards,
Ho-Ren (Jack) Chuang
莊賀任