diff mbox series

[v2,1/1] memory tier: acpi/hmat: create CPUless memory tiers after obtaining HMAT info

Message ID 20240312061729.1997111-2-horenchuang@bytedance.com
State New
Headers show
Series Improved Memory Tier Creation for CPUless NUMA Nodes | expand

Commit Message

Ho-Ren (Jack) Chuang March 12, 2024, 6:17 a.m. UTC
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.

* Abstract common functions into `find_alloc_memory_type()`
Since different memory devices require finding or allocating a memory type,
these common steps are abstracted into a single function,
`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.

Signed-off-by: Ho-Ren (Jack) Chuang <horenchuang@bytedance.com>
Signed-off-by: Hao Xiang <hao.xiang@bytedance.com>
---
 drivers/acpi/numa/hmat.c     | 11 ++++++
 drivers/dax/kmem.c           | 13 +------
 include/linux/acpi.h         |  6 ++++
 include/linux/memory-tiers.h |  8 +++++
 mm/memory-tiers.c            | 70 +++++++++++++++++++++++++++++++++---
 5 files changed, 92 insertions(+), 16 deletions(-)

Comments

Ho-Ren (Jack) Chuang March 13, 2024, 8:16 a.m. UTC | #1
On Tue, Mar 12, 2024 at 2:21 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.
> >
> > * Abstract common functions into `find_alloc_memory_type()`
>
> We should move kmem_put_memory_types() (renamed to
> mt_put_memory_types()?) too.  This can be put in a separate patch.
>

Will do! Thanks,


>
> > Since different memory devices require finding or allocating a memory type,
> > these common steps are abstracted into a single function,
> > `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.
> >
> > Signed-off-by: Ho-Ren (Jack) Chuang <horenchuang@bytedance.com>
> > Signed-off-by: Hao Xiang <hao.xiang@bytedance.com>
> > ---
> >  drivers/acpi/numa/hmat.c     | 11 ++++++
> >  drivers/dax/kmem.c           | 13 +------
> >  include/linux/acpi.h         |  6 ++++
> >  include/linux/memory-tiers.h |  8 +++++
> >  mm/memory-tiers.c            | 70 +++++++++++++++++++++++++++++++++---
> >  5 files changed, 92 insertions(+), 16 deletions(-)
> >
> > diff --git a/drivers/acpi/numa/hmat.c b/drivers/acpi/numa/hmat.c
> > index d6b85f0f6082..28812ec2c793 100644
> > --- a/drivers/acpi/numa/hmat.c
> > +++ b/drivers/acpi/numa/hmat.c
> > @@ -38,6 +38,8 @@ static LIST_HEAD(targets);
> >  static LIST_HEAD(initiators);
> >  static LIST_HEAD(localities);
> >
> > +static LIST_HEAD(hmat_memory_types);
> > +
>
> HMAT isn't a device driver for some memory devices.  So I don't think we
> should manage memory types in HMAT.

I can put it back in memory-tier.c. How about the list? Do we still
need to keep a separate list for storing late_inited memory nodes?
And how about the list name if we need to remove the prefix "hmat_"?


> Instead, if the memory_type of a
> node isn't set by the driver, we should manage it in memory-tier.c as
> fallback.
>

Do you mean some device drivers may init memory tiers between
memory_tier_init() and late_initcall(memory_tier_late_init);?
And this is the reason why you mention to exclude
"node_memory_types[nid].memtype != NULL" in memory_tier_late_init().
Is my understanding correct?

> >  static DEFINE_MUTEX(target_lock);
> >
> >  /*
> > @@ -149,6 +151,12 @@ int acpi_get_genport_coordinates(u32 uid,
> >  }
> >  EXPORT_SYMBOL_NS_GPL(acpi_get_genport_coordinates, CXL);
> >
> > +struct memory_dev_type *hmat_find_alloc_memory_type(int adist)
> > +{
> > +     return find_alloc_memory_type(adist, &hmat_memory_types);
> > +}
> > +EXPORT_SYMBOL_GPL(hmat_find_alloc_memory_type);
> > +
> >  static __init void alloc_memory_initiator(unsigned int cpu_pxm)
> >  {
> >       struct memory_initiator *initiator;
> > @@ -1038,6 +1046,9 @@ static __init int hmat_init(void)
> >       if (!hmat_set_default_dram_perf())
> >               register_mt_adistance_algorithm(&hmat_adist_nb);
> >
> > +     /* Post-create CPUless memory tiers after getting HMAT info */
> > +     memory_tier_late_init();
> > +
>
> This should be called in memory-tier.c via
>
> late_initcall(memory_tier_late_init);
>
> Then, we don't need hmat to call it.
>

Thanks. Learned!


> >       return 0;
> >  out_put:
> >       hmat_free_structures();
> > diff --git a/drivers/dax/kmem.c b/drivers/dax/kmem.c
> > index 42ee360cf4e3..aee17ab59f4f 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 = find_alloc_memory_type(adist, &kmem_memory_types);
> >       mutex_unlock(&kmem_memory_type_lock);
> >
> >       return mtype;
> > diff --git a/include/linux/acpi.h b/include/linux/acpi.h
> > index b7165e52b3c6..3f927ff01f02 100644
> > --- a/include/linux/acpi.h
> > +++ b/include/linux/acpi.h
> > @@ -434,12 +434,18 @@ int thermal_acpi_critical_trip_temp(struct acpi_device *adev, int *ret_temp);
> >
> >  #ifdef CONFIG_ACPI_HMAT
> >  int acpi_get_genport_coordinates(u32 uid, struct access_coordinate *coord);
> > +struct memory_dev_type *hmat_find_alloc_memory_type(int adist);
> >  #else
> >  static inline int acpi_get_genport_coordinates(u32 uid,
> >                                              struct access_coordinate *coord)
> >  {
> >       return -EOPNOTSUPP;
> >  }
> > +
> > +static inline struct memory_dev_type *hmat_find_alloc_memory_type(int adist)
> > +{
> > +     return NULL;
> > +}
> >  #endif
> >
> >  #ifdef CONFIG_ACPI_NUMA
> > diff --git a/include/linux/memory-tiers.h b/include/linux/memory-tiers.h
> > index 69e781900082..4bc2596c5774 100644
> > --- a/include/linux/memory-tiers.h
> > +++ b/include/linux/memory-tiers.h
> > @@ -48,6 +48,9 @@ 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 *find_alloc_memory_type(int adist,
> > +                                                     struct list_head *memory_types);
> > +void memory_tier_late_init(void);
> >  #ifdef CONFIG_MIGRATION
> >  int next_demotion_node(int node);
> >  void node_get_allowed_targets(pg_data_t *pgdat, nodemask_t *targets);
> > @@ -136,5 +139,10 @@ static inline int mt_perf_to_adistance(struct access_coordinate *perf, int *adis
> >  {
> >       return -EIO;
> >  }
> > +
> > +static inline void memory_tier_late_init(void)
> > +{
> > +
> > +}
> >  #endif       /* CONFIG_NUMA */
> >  #endif  /* _LINUX_MEMORY_TIERS_H */
> > diff --git a/mm/memory-tiers.c b/mm/memory-tiers.c
> > index 0537664620e5..79f748d60e6f 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"
> >
> > @@ -35,6 +36,7 @@ struct node_memory_type_map {
> >  };
> >
> >  static DEFINE_MUTEX(memory_tier_lock);
> > +static DEFINE_MUTEX(mt_perf_lock);
>
> Please add comments about what it protects.  And put it near the data
> structure it protects.
>

Where is better for me to add comments for this? Code? Patch description?
Will put it closer to the protected data. Thanks.



> >  static LIST_HEAD(memory_tiers);
> >  static struct node_memory_type_map node_memory_types[MAX_NUMNODES];
> >  struct memory_dev_type *default_dram_type;
> > @@ -623,6 +625,58 @@ void clear_node_memory_type(int node, struct memory_dev_type *memtype)
> >  }
> >  EXPORT_SYMBOL_GPL(clear_node_memory_type);
> >
> > +struct memory_dev_type *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(find_alloc_memory_type);
> > +
> > +static void memory_tier_late_create(int node)
> > +{
> > +     struct memory_dev_type *mtype = NULL;
> > +     int adist = MEMTIER_ADISTANCE_DRAM;
> > +
> > +     mt_calc_adistance(node, &adist);
> > +     if (adist != MEMTIER_ADISTANCE_DRAM) {
>
> We can manage default_dram_type() via find_alloc_memory_type()
> too.
>
> And, if "node_memory_types[node].memtype == NULL", we can call
> mt_calc_adistance(node, &adist) and find_alloc_memory_type() in
> set_node_memory_tier().  Then, we can cover hotpluged memory node too.
>
> > +             mtype = hmat_find_alloc_memory_type(adist);
> > +             if (!IS_ERR(mtype))
> > +                     __init_node_memory_type(node, mtype);
> > +             else
> > +                     pr_err("Failed to allocate a memory type at %s()\n", __func__);
> > +     }
> > +
> > +     set_node_memory_tier(node);
> > +}
> > +
> > +void 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))
>
> We should exclude "node_memory_types[nid].memtype != NULL".  Some memory
> nodes may be onlined by some device drivers and setup memory tiers
> already.
>
> > +                     memory_tier_late_create(nid);
> > +
> > +     establish_demotion_targets();
> > +     mutex_unlock(&memory_tier_lock);
> > +}
> > +EXPORT_SYMBOL_GPL(memory_tier_late_init);
> > +
> >  static void dump_hmem_attrs(struct access_coordinate *coord, const char *prefix)
> >  {
> >       pr_info(
> > @@ -636,7 +690,7 @@ int mt_set_default_dram_perf(int nid, struct access_coordinate *perf,
> >  {
> >       int rc = 0;
> >
> > -     mutex_lock(&memory_tier_lock);
> > +     mutex_lock(&mt_perf_lock);
> >       if (default_dram_perf_error) {
> >               rc = -EIO;
> >               goto out;
> > @@ -684,7 +738,7 @@ 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 +754,7 @@ 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 +767,7 @@ 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;
> >  }
> > @@ -836,6 +890,14 @@ 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 when HMAT information is
>
> HMAT is platform specific, we should avoid to mention it in general code
> if possible.
>

Will fix! Thanks,


> > +                      * available.
> > +                      */
> > +                     continue;
> > +
> >               memtier = set_node_memory_tier(node);
> >               if (IS_ERR(memtier))
> >                       /*
>
> --
> Best Regards,
> Huang, Ying
Huang, Ying March 14, 2024, 7:52 a.m. UTC | #2
"Ho-Ren (Jack) Chuang" <horenchuang@bytedance.com> writes:

> On Tue, Mar 12, 2024 at 2:21 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.
>> >
>> > * Abstract common functions into `find_alloc_memory_type()`
>>
>> We should move kmem_put_memory_types() (renamed to
>> mt_put_memory_types()?) too.  This can be put in a separate patch.
>>
>
> Will do! Thanks,
>
>
>>
>> > Since different memory devices require finding or allocating a memory type,
>> > these common steps are abstracted into a single function,
>> > `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.
>> >
>> > Signed-off-by: Ho-Ren (Jack) Chuang <horenchuang@bytedance.com>
>> > Signed-off-by: Hao Xiang <hao.xiang@bytedance.com>
>> > ---
>> >  drivers/acpi/numa/hmat.c     | 11 ++++++
>> >  drivers/dax/kmem.c           | 13 +------
>> >  include/linux/acpi.h         |  6 ++++
>> >  include/linux/memory-tiers.h |  8 +++++
>> >  mm/memory-tiers.c            | 70 +++++++++++++++++++++++++++++++++---
>> >  5 files changed, 92 insertions(+), 16 deletions(-)
>> >
>> > diff --git a/drivers/acpi/numa/hmat.c b/drivers/acpi/numa/hmat.c
>> > index d6b85f0f6082..28812ec2c793 100644
>> > --- a/drivers/acpi/numa/hmat.c
>> > +++ b/drivers/acpi/numa/hmat.c
>> > @@ -38,6 +38,8 @@ static LIST_HEAD(targets);
>> >  static LIST_HEAD(initiators);
>> >  static LIST_HEAD(localities);
>> >
>> > +static LIST_HEAD(hmat_memory_types);
>> > +
>>
>> HMAT isn't a device driver for some memory devices.  So I don't think we
>> should manage memory types in HMAT.
>
> I can put it back in memory-tier.c. How about the list? Do we still
> need to keep a separate list for storing late_inited memory nodes?
> And how about the list name if we need to remove the prefix "hmat_"?

I don't think we need a separate list for memory-less nodes.  Just
iterate all memory-less nodes.

>
>> Instead, if the memory_type of a
>> node isn't set by the driver, we should manage it in memory-tier.c as
>> fallback.
>>
>
> Do you mean some device drivers may init memory tiers between
> memory_tier_init() and late_initcall(memory_tier_late_init);?
> And this is the reason why you mention to exclude
> "node_memory_types[nid].memtype != NULL" in memory_tier_late_init().
> Is my understanding correct?

Yes.

>> >  static DEFINE_MUTEX(target_lock);
>> >
>> >  /*
>> > @@ -149,6 +151,12 @@ int acpi_get_genport_coordinates(u32 uid,
>> >  }
>> >  EXPORT_SYMBOL_NS_GPL(acpi_get_genport_coordinates, CXL);
>> >
>> > +struct memory_dev_type *hmat_find_alloc_memory_type(int adist)
>> > +{
>> > +     return find_alloc_memory_type(adist, &hmat_memory_types);
>> > +}
>> > +EXPORT_SYMBOL_GPL(hmat_find_alloc_memory_type);
>> > +
>> >  static __init void alloc_memory_initiator(unsigned int cpu_pxm)
>> >  {
>> >       struct memory_initiator *initiator;
>> > @@ -1038,6 +1046,9 @@ static __init int hmat_init(void)
>> >       if (!hmat_set_default_dram_perf())
>> >               register_mt_adistance_algorithm(&hmat_adist_nb);
>> >
>> > +     /* Post-create CPUless memory tiers after getting HMAT info */
>> > +     memory_tier_late_init();
>> > +
>>
>> This should be called in memory-tier.c via
>>
>> late_initcall(memory_tier_late_init);
>>
>> Then, we don't need hmat to call it.
>>
>
> Thanks. Learned!
>
>
>> >       return 0;
>> >  out_put:
>> >       hmat_free_structures();
>> > diff --git a/drivers/dax/kmem.c b/drivers/dax/kmem.c
>> > index 42ee360cf4e3..aee17ab59f4f 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 = find_alloc_memory_type(adist, &kmem_memory_types);
>> >       mutex_unlock(&kmem_memory_type_lock);
>> >
>> >       return mtype;
>> > diff --git a/include/linux/acpi.h b/include/linux/acpi.h
>> > index b7165e52b3c6..3f927ff01f02 100644
>> > --- a/include/linux/acpi.h
>> > +++ b/include/linux/acpi.h
>> > @@ -434,12 +434,18 @@ int thermal_acpi_critical_trip_temp(struct acpi_device *adev, int *ret_temp);
>> >
>> >  #ifdef CONFIG_ACPI_HMAT
>> >  int acpi_get_genport_coordinates(u32 uid, struct access_coordinate *coord);
>> > +struct memory_dev_type *hmat_find_alloc_memory_type(int adist);
>> >  #else
>> >  static inline int acpi_get_genport_coordinates(u32 uid,
>> >                                              struct access_coordinate *coord)
>> >  {
>> >       return -EOPNOTSUPP;
>> >  }
>> > +
>> > +static inline struct memory_dev_type *hmat_find_alloc_memory_type(int adist)
>> > +{
>> > +     return NULL;
>> > +}
>> >  #endif
>> >
>> >  #ifdef CONFIG_ACPI_NUMA
>> > diff --git a/include/linux/memory-tiers.h b/include/linux/memory-tiers.h
>> > index 69e781900082..4bc2596c5774 100644
>> > --- a/include/linux/memory-tiers.h
>> > +++ b/include/linux/memory-tiers.h
>> > @@ -48,6 +48,9 @@ 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 *find_alloc_memory_type(int adist,
>> > +                                                     struct list_head *memory_types);
>> > +void memory_tier_late_init(void);
>> >  #ifdef CONFIG_MIGRATION
>> >  int next_demotion_node(int node);
>> >  void node_get_allowed_targets(pg_data_t *pgdat, nodemask_t *targets);
>> > @@ -136,5 +139,10 @@ static inline int mt_perf_to_adistance(struct access_coordinate *perf, int *adis
>> >  {
>> >       return -EIO;
>> >  }
>> > +
>> > +static inline void memory_tier_late_init(void)
>> > +{
>> > +
>> > +}
>> >  #endif       /* CONFIG_NUMA */
>> >  #endif  /* _LINUX_MEMORY_TIERS_H */
>> > diff --git a/mm/memory-tiers.c b/mm/memory-tiers.c
>> > index 0537664620e5..79f748d60e6f 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"
>> >
>> > @@ -35,6 +36,7 @@ struct node_memory_type_map {
>> >  };
>> >
>> >  static DEFINE_MUTEX(memory_tier_lock);
>> > +static DEFINE_MUTEX(mt_perf_lock);
>>
>> Please add comments about what it protects.  And put it near the data
>> structure it protects.
>>
>
> Where is better for me to add comments for this? Code? Patch description?
> Will put it closer to the protected data. Thanks.

Just put the comments at the above of the lock in the source code.

>
>
>> >  static LIST_HEAD(memory_tiers);
>> >  static struct node_memory_type_map node_memory_types[MAX_NUMNODES];
>> >  struct memory_dev_type *default_dram_type;
>> > @@ -623,6 +625,58 @@ void clear_node_memory_type(int node, struct memory_dev_type *memtype)
>> >  }
>> >  EXPORT_SYMBOL_GPL(clear_node_memory_type);
>> >
>> > +struct memory_dev_type *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(find_alloc_memory_type);
>> > +
>> > +static void memory_tier_late_create(int node)
>> > +{
>> > +     struct memory_dev_type *mtype = NULL;
>> > +     int adist = MEMTIER_ADISTANCE_DRAM;
>> > +
>> > +     mt_calc_adistance(node, &adist);
>> > +     if (adist != MEMTIER_ADISTANCE_DRAM) {
>>
>> We can manage default_dram_type() via find_alloc_memory_type()
>> too.
>>
>> And, if "node_memory_types[node].memtype == NULL", we can call
>> mt_calc_adistance(node, &adist) and find_alloc_memory_type() in
>> set_node_memory_tier().  Then, we can cover hotpluged memory node too.
>>
>> > +             mtype = hmat_find_alloc_memory_type(adist);
>> > +             if (!IS_ERR(mtype))
>> > +                     __init_node_memory_type(node, mtype);
>> > +             else
>> > +                     pr_err("Failed to allocate a memory type at %s()\n", __func__);
>> > +     }
>> > +
>> > +     set_node_memory_tier(node);
>> > +}
>> > +
>> > +void 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))
>>
>> We should exclude "node_memory_types[nid].memtype != NULL".  Some memory
>> nodes may be onlined by some device drivers and setup memory tiers
>> already.
>>
>> > +                     memory_tier_late_create(nid);
>> > +
>> > +     establish_demotion_targets();
>> > +     mutex_unlock(&memory_tier_lock);
>> > +}
>> > +EXPORT_SYMBOL_GPL(memory_tier_late_init);
>> > +
>> >  static void dump_hmem_attrs(struct access_coordinate *coord, const char *prefix)
>> >  {
>> >       pr_info(
>> > @@ -636,7 +690,7 @@ int mt_set_default_dram_perf(int nid, struct access_coordinate *perf,
>> >  {
>> >       int rc = 0;
>> >
>> > -     mutex_lock(&memory_tier_lock);
>> > +     mutex_lock(&mt_perf_lock);
>> >       if (default_dram_perf_error) {
>> >               rc = -EIO;
>> >               goto out;
>> > @@ -684,7 +738,7 @@ 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 +754,7 @@ 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 +767,7 @@ 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;
>> >  }
>> > @@ -836,6 +890,14 @@ 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 when HMAT information is
>>
>> HMAT is platform specific, we should avoid to mention it in general code
>> if possible.
>>
>
> Will fix! Thanks,
>
>
>> > +                      * available.
>> > +                      */
>> > +                     continue;
>> > +
>> >               memtier = set_node_memory_tier(node);
>> >               if (IS_ERR(memtier))
>> >                       /*
>>

--
Best Regards,
Huang, Ying
Ho-Ren (Jack) Chuang March 18, 2024, 8:12 p.m. UTC | #3
I'm working on V3. Thanks for Ying's feedback.

cc: sthanneeru@micron.com


On Thu, Mar 14, 2024 at 12:54 AM Huang, Ying <ying.huang@intel.com> wrote:
>
> "Ho-Ren (Jack) Chuang" <horenchuang@bytedance.com> writes:
>
> > On Tue, Mar 12, 2024 at 2:21 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.
> >> >
> >> > * Abstract common functions into `find_alloc_memory_type()`
> >>
> >> We should move kmem_put_memory_types() (renamed to
> >> mt_put_memory_types()?) too.  This can be put in a separate patch.
> >>
> >
> > Will do! Thanks,
> >
> >
> >>
> >> > Since different memory devices require finding or allocating a memory type,
> >> > these common steps are abstracted into a single function,
> >> > `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.
> >> >
> >> > Signed-off-by: Ho-Ren (Jack) Chuang <horenchuang@bytedance.com>
> >> > Signed-off-by: Hao Xiang <hao.xiang@bytedance.com>
> >> > ---
> >> >  drivers/acpi/numa/hmat.c     | 11 ++++++
> >> >  drivers/dax/kmem.c           | 13 +------
> >> >  include/linux/acpi.h         |  6 ++++
> >> >  include/linux/memory-tiers.h |  8 +++++
> >> >  mm/memory-tiers.c            | 70 +++++++++++++++++++++++++++++++++---
> >> >  5 files changed, 92 insertions(+), 16 deletions(-)
> >> >
> >> > diff --git a/drivers/acpi/numa/hmat.c b/drivers/acpi/numa/hmat.c
> >> > index d6b85f0f6082..28812ec2c793 100644
> >> > --- a/drivers/acpi/numa/hmat.c
> >> > +++ b/drivers/acpi/numa/hmat.c
> >> > @@ -38,6 +38,8 @@ static LIST_HEAD(targets);
> >> >  static LIST_HEAD(initiators);
> >> >  static LIST_HEAD(localities);
> >> >
> >> > +static LIST_HEAD(hmat_memory_types);
> >> > +
> >>
> >> HMAT isn't a device driver for some memory devices.  So I don't think we
> >> should manage memory types in HMAT.
> >
> > I can put it back in memory-tier.c. How about the list? Do we still
> > need to keep a separate list for storing late_inited memory nodes?
> > And how about the list name if we need to remove the prefix "hmat_"?
>
> I don't think we need a separate list for memory-less nodes.  Just
> iterate all memory-less nodes.
>

Ok. There is no need to keep a list of memory-less nodes. I will
only keep a memory_type list to manage created memory types.


> >
> >> Instead, if the memory_type of a
> >> node isn't set by the driver, we should manage it in memory-tier.c as
> >> fallback.
> >>
> >
> > Do you mean some device drivers may init memory tiers between
> > memory_tier_init() and late_initcall(memory_tier_late_init);?
> > And this is the reason why you mention to exclude
> > "node_memory_types[nid].memtype != NULL" in memory_tier_late_init().
> > Is my understanding correct?
>
> Yes.
>

Thanks.

> >> >  static DEFINE_MUTEX(target_lock);
> >> >
> >> >  /*
> >> > @@ -149,6 +151,12 @@ int acpi_get_genport_coordinates(u32 uid,
> >> >  }
> >> >  EXPORT_SYMBOL_NS_GPL(acpi_get_genport_coordinates, CXL);
> >> >
> >> > +struct memory_dev_type *hmat_find_alloc_memory_type(int adist)
> >> > +{
> >> > +     return find_alloc_memory_type(adist, &hmat_memory_types);
> >> > +}
> >> > +EXPORT_SYMBOL_GPL(hmat_find_alloc_memory_type);
> >> > +
> >> >  static __init void alloc_memory_initiator(unsigned int cpu_pxm)
> >> >  {
> >> >       struct memory_initiator *initiator;
> >> > @@ -1038,6 +1046,9 @@ static __init int hmat_init(void)
> >> >       if (!hmat_set_default_dram_perf())
> >> >               register_mt_adistance_algorithm(&hmat_adist_nb);
> >> >
> >> > +     /* Post-create CPUless memory tiers after getting HMAT info */
> >> > +     memory_tier_late_init();
> >> > +
> >>
> >> This should be called in memory-tier.c via
> >>
> >> late_initcall(memory_tier_late_init);
> >>
> >> Then, we don't need hmat to call it.
> >>
> >
> > Thanks. Learned!
> >
> >
> >> >       return 0;
> >> >  out_put:
> >> >       hmat_free_structures();
> >> > diff --git a/drivers/dax/kmem.c b/drivers/dax/kmem.c
> >> > index 42ee360cf4e3..aee17ab59f4f 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 = find_alloc_memory_type(adist, &kmem_memory_types);
> >> >       mutex_unlock(&kmem_memory_type_lock);
> >> >
> >> >       return mtype;
> >> > diff --git a/include/linux/acpi.h b/include/linux/acpi.h
> >> > index b7165e52b3c6..3f927ff01f02 100644
> >> > --- a/include/linux/acpi.h
> >> > +++ b/include/linux/acpi.h
> >> > @@ -434,12 +434,18 @@ int thermal_acpi_critical_trip_temp(struct acpi_device *adev, int *ret_temp);
> >> >
> >> >  #ifdef CONFIG_ACPI_HMAT
> >> >  int acpi_get_genport_coordinates(u32 uid, struct access_coordinate *coord);
> >> > +struct memory_dev_type *hmat_find_alloc_memory_type(int adist);
> >> >  #else
> >> >  static inline int acpi_get_genport_coordinates(u32 uid,
> >> >                                              struct access_coordinate *coord)
> >> >  {
> >> >       return -EOPNOTSUPP;
> >> >  }
> >> > +
> >> > +static inline struct memory_dev_type *hmat_find_alloc_memory_type(int adist)
> >> > +{
> >> > +     return NULL;
> >> > +}
> >> >  #endif
> >> >
> >> >  #ifdef CONFIG_ACPI_NUMA
> >> > diff --git a/include/linux/memory-tiers.h b/include/linux/memory-tiers.h
> >> > index 69e781900082..4bc2596c5774 100644
> >> > --- a/include/linux/memory-tiers.h
> >> > +++ b/include/linux/memory-tiers.h
> >> > @@ -48,6 +48,9 @@ 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 *find_alloc_memory_type(int adist,
> >> > +                                                     struct list_head *memory_types);
> >> > +void memory_tier_late_init(void);
> >> >  #ifdef CONFIG_MIGRATION
> >> >  int next_demotion_node(int node);
> >> >  void node_get_allowed_targets(pg_data_t *pgdat, nodemask_t *targets);
> >> > @@ -136,5 +139,10 @@ static inline int mt_perf_to_adistance(struct access_coordinate *perf, int *adis
> >> >  {
> >> >       return -EIO;
> >> >  }
> >> > +
> >> > +static inline void memory_tier_late_init(void)
> >> > +{
> >> > +
> >> > +}
> >> >  #endif       /* CONFIG_NUMA */
> >> >  #endif  /* _LINUX_MEMORY_TIERS_H */
> >> > diff --git a/mm/memory-tiers.c b/mm/memory-tiers.c
> >> > index 0537664620e5..79f748d60e6f 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"
> >> >
> >> > @@ -35,6 +36,7 @@ struct node_memory_type_map {
> >> >  };
> >> >
> >> >  static DEFINE_MUTEX(memory_tier_lock);
> >> > +static DEFINE_MUTEX(mt_perf_lock);
> >>
> >> Please add comments about what it protects.  And put it near the data
> >> structure it protects.
> >>
> >
> > Where is better for me to add comments for this? Code? Patch description?
> > Will put it closer to the protected data. Thanks.
>
> Just put the comments at the above of the lock in the source code.
>

Got it. Thanks!

>
> >
> >
> >> >  static LIST_HEAD(memory_tiers);
> >> >  static struct node_memory_type_map node_memory_types[MAX_NUMNODES];
> >> >  struct memory_dev_type *default_dram_type;
> >> > @@ -623,6 +625,58 @@ void clear_node_memory_type(int node, struct memory_dev_type *memtype)
> >> >  }
> >> >  EXPORT_SYMBOL_GPL(clear_node_memory_type);
> >> >
> >> > +struct memory_dev_type *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(find_alloc_memory_type);
> >> > +
> >> > +static void memory_tier_late_create(int node)
> >> > +{
> >> > +     struct memory_dev_type *mtype = NULL;
> >> > +     int adist = MEMTIER_ADISTANCE_DRAM;
> >> > +
> >> > +     mt_calc_adistance(node, &adist);
> >> > +     if (adist != MEMTIER_ADISTANCE_DRAM) {
> >>
> >> We can manage default_dram_type() via find_alloc_memory_type()
> >> too.
> >>
> >> And, if "node_memory_types[node].memtype == NULL", we can call
> >> mt_calc_adistance(node, &adist) and find_alloc_memory_type() in
> >> set_node_memory_tier().  Then, we can cover hotpluged memory node too.
> >>
> >> > +             mtype = hmat_find_alloc_memory_type(adist);
> >> > +             if (!IS_ERR(mtype))
> >> > +                     __init_node_memory_type(node, mtype);
> >> > +             else
> >> > +                     pr_err("Failed to allocate a memory type at %s()\n", __func__);
> >> > +     }
> >> > +
> >> > +     set_node_memory_tier(node);
> >> > +}
> >> > +
> >> > +void 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))
> >>
> >> We should exclude "node_memory_types[nid].memtype != NULL".  Some memory
> >> nodes may be onlined by some device drivers and setup memory tiers
> >> already.
> >>
> >> > +                     memory_tier_late_create(nid);
> >> > +
> >> > +     establish_demotion_targets();
> >> > +     mutex_unlock(&memory_tier_lock);
> >> > +}
> >> > +EXPORT_SYMBOL_GPL(memory_tier_late_init);
> >> > +
> >> >  static void dump_hmem_attrs(struct access_coordinate *coord, const char *prefix)
> >> >  {
> >> >       pr_info(
> >> > @@ -636,7 +690,7 @@ int mt_set_default_dram_perf(int nid, struct access_coordinate *perf,
> >> >  {
> >> >       int rc = 0;
> >> >
> >> > -     mutex_lock(&memory_tier_lock);
> >> > +     mutex_lock(&mt_perf_lock);
> >> >       if (default_dram_perf_error) {
> >> >               rc = -EIO;
> >> >               goto out;
> >> > @@ -684,7 +738,7 @@ 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 +754,7 @@ 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 +767,7 @@ 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;
> >> >  }
> >> > @@ -836,6 +890,14 @@ 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 when HMAT information is
> >>
> >> HMAT is platform specific, we should avoid to mention it in general code
> >> if possible.
> >>
> >
> > Will fix! Thanks,
> >
> >
> >> > +                      * available.
> >> > +                      */
> >> > +                     continue;
> >> > +
> >> >               memtier = set_node_memory_tier(node);
> >> >               if (IS_ERR(memtier))
> >> >                       /*
> >>
>
> --
> Best Regards,
> Huang, Ying



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

Patch

diff --git a/drivers/acpi/numa/hmat.c b/drivers/acpi/numa/hmat.c
index d6b85f0f6082..28812ec2c793 100644
--- a/drivers/acpi/numa/hmat.c
+++ b/drivers/acpi/numa/hmat.c
@@ -38,6 +38,8 @@  static LIST_HEAD(targets);
 static LIST_HEAD(initiators);
 static LIST_HEAD(localities);
 
+static LIST_HEAD(hmat_memory_types);
+
 static DEFINE_MUTEX(target_lock);
 
 /*
@@ -149,6 +151,12 @@  int acpi_get_genport_coordinates(u32 uid,
 }
 EXPORT_SYMBOL_NS_GPL(acpi_get_genport_coordinates, CXL);
 
+struct memory_dev_type *hmat_find_alloc_memory_type(int adist)
+{
+	return find_alloc_memory_type(adist, &hmat_memory_types);
+}
+EXPORT_SYMBOL_GPL(hmat_find_alloc_memory_type);
+
 static __init void alloc_memory_initiator(unsigned int cpu_pxm)
 {
 	struct memory_initiator *initiator;
@@ -1038,6 +1046,9 @@  static __init int hmat_init(void)
 	if (!hmat_set_default_dram_perf())
 		register_mt_adistance_algorithm(&hmat_adist_nb);
 
+	/* Post-create CPUless memory tiers after getting HMAT info */
+	memory_tier_late_init();
+
 	return 0;
 out_put:
 	hmat_free_structures();
diff --git a/drivers/dax/kmem.c b/drivers/dax/kmem.c
index 42ee360cf4e3..aee17ab59f4f 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 = find_alloc_memory_type(adist, &kmem_memory_types);
 	mutex_unlock(&kmem_memory_type_lock);
 
 	return mtype;
diff --git a/include/linux/acpi.h b/include/linux/acpi.h
index b7165e52b3c6..3f927ff01f02 100644
--- a/include/linux/acpi.h
+++ b/include/linux/acpi.h
@@ -434,12 +434,18 @@  int thermal_acpi_critical_trip_temp(struct acpi_device *adev, int *ret_temp);
 
 #ifdef CONFIG_ACPI_HMAT
 int acpi_get_genport_coordinates(u32 uid, struct access_coordinate *coord);
+struct memory_dev_type *hmat_find_alloc_memory_type(int adist);
 #else
 static inline int acpi_get_genport_coordinates(u32 uid,
 					       struct access_coordinate *coord)
 {
 	return -EOPNOTSUPP;
 }
+
+static inline struct memory_dev_type *hmat_find_alloc_memory_type(int adist)
+{
+	return NULL;
+}
 #endif
 
 #ifdef CONFIG_ACPI_NUMA
diff --git a/include/linux/memory-tiers.h b/include/linux/memory-tiers.h
index 69e781900082..4bc2596c5774 100644
--- a/include/linux/memory-tiers.h
+++ b/include/linux/memory-tiers.h
@@ -48,6 +48,9 @@  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 *find_alloc_memory_type(int adist,
+							struct list_head *memory_types);
+void memory_tier_late_init(void);
 #ifdef CONFIG_MIGRATION
 int next_demotion_node(int node);
 void node_get_allowed_targets(pg_data_t *pgdat, nodemask_t *targets);
@@ -136,5 +139,10 @@  static inline int mt_perf_to_adistance(struct access_coordinate *perf, int *adis
 {
 	return -EIO;
 }
+
+static inline void memory_tier_late_init(void)
+{
+
+}
 #endif	/* CONFIG_NUMA */
 #endif  /* _LINUX_MEMORY_TIERS_H */
diff --git a/mm/memory-tiers.c b/mm/memory-tiers.c
index 0537664620e5..79f748d60e6f 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"
 
@@ -35,6 +36,7 @@  struct node_memory_type_map {
 };
 
 static DEFINE_MUTEX(memory_tier_lock);
+static DEFINE_MUTEX(mt_perf_lock);
 static LIST_HEAD(memory_tiers);
 static struct node_memory_type_map node_memory_types[MAX_NUMNODES];
 struct memory_dev_type *default_dram_type;
@@ -623,6 +625,58 @@  void clear_node_memory_type(int node, struct memory_dev_type *memtype)
 }
 EXPORT_SYMBOL_GPL(clear_node_memory_type);
 
+struct memory_dev_type *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(find_alloc_memory_type);
+
+static void memory_tier_late_create(int node)
+{
+	struct memory_dev_type *mtype = NULL;
+	int adist = MEMTIER_ADISTANCE_DRAM;
+
+	mt_calc_adistance(node, &adist);
+	if (adist != MEMTIER_ADISTANCE_DRAM) {
+		mtype = hmat_find_alloc_memory_type(adist);
+		if (!IS_ERR(mtype))
+			__init_node_memory_type(node, mtype);
+		else
+			pr_err("Failed to allocate a memory type at %s()\n", __func__);
+	}
+
+	set_node_memory_tier(node);
+}
+
+void 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))
+			memory_tier_late_create(nid);
+
+	establish_demotion_targets();
+	mutex_unlock(&memory_tier_lock);
+}
+EXPORT_SYMBOL_GPL(memory_tier_late_init);
+
 static void dump_hmem_attrs(struct access_coordinate *coord, const char *prefix)
 {
 	pr_info(
@@ -636,7 +690,7 @@  int mt_set_default_dram_perf(int nid, struct access_coordinate *perf,
 {
 	int rc = 0;
 
-	mutex_lock(&memory_tier_lock);
+	mutex_lock(&mt_perf_lock);
 	if (default_dram_perf_error) {
 		rc = -EIO;
 		goto out;
@@ -684,7 +738,7 @@  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 +754,7 @@  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 +767,7 @@  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;
 }
@@ -836,6 +890,14 @@  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 when HMAT information is
+			 * available.
+			 */
+			continue;
+
 		memtier = set_node_memory_tier(node);
 		if (IS_ERR(memtier))
 			/*